Add an unpatched version of xz, XZ Utils, to /trunk/deps/third_party.
README.chromium will be placed in the Chromium tree in the directory that
receives xz. Here are its contents, which provide information about xz:
--
Name: XZ Utils
URL: http://tukaani.org/xz/
Source URL: git://ctrl.tukaani.org/xz.git
Version: git commit c15c42abb3c8c6e77c778ef06c97a4a10b8b5d00
(2010-06-15 14:06:29 +0300), newer than 4.999.9beta
License: Public domain/license grant, LGPL 2.1, GPL 2, GPL 3
License File: COPYING
Description:
XZ Utils is general-purpose data compression software implementing the
modified Lempel-Ziv-Markov chain algorithm (LZMA2).
Local Modifications:
- None.
--
BUG=47199
TEST=none
Review URL: http://codereview.chromium.org/2869016
git-svn-id: http://src.chromium.org/svn/trunk/deps/third_party/xz@50504 4ff67af0-8c30-449e-8e8b-ad334ec8d88c
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..0b5bb53
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,56 @@
+*~
+*.bak
+*.bak[0-9]
+
+.deps
+.libs
+*.la
+*.lo
+*.o
+Makefile.in
+
+/ABOUT-NLS
+/autom4te.cache
+/Doxyfile
+/aclocal.m4
+/build-aux
+/config.h
+/config.h.in
+/config.log
+/config.status
+/configure
+/libtool
+/stamp-h1
+
+/src/liblzma/liblzma.pc
+/src/lzmainfo/lzmainfo
+/src/xz/xz
+/src/xzdec/lzmadec
+/src/xzdec/xzdec
+
+/src/scripts/xzdiff
+/src/scripts/xzgrep
+/src/scripts/xzless
+/src/scripts/xzmore
+
+/tests/compress_generated_abc
+/tests/compress_generated_random
+/tests/compress_generated_text
+/tests/create_compress_files
+/tests/test_block_header
+/tests/test_check
+/tests/test_filter_flags
+/tests/test_index
+/tests/test_stream_flags
+
+/lib/Makefile
+/tests/Makefile
+/Makefile
+/debug/Makefile
+/src/scripts/Makefile
+/src/xz/Makefile
+/src/Makefile
+/src/liblzma/Makefile
+/src/liblzma/api/Makefile
+/src/lzmainfo/Makefile
+/src/xzdec/Makefile
diff --git a/AUTHORS b/AUTHORS
new file mode 100644
index 0000000..63a9815
--- /dev/null
+++ b/AUTHORS
@@ -0,0 +1,27 @@
+
+Authors of XZ Utils
+===================
+
+ XZ Utils is developed and maintained by Lasse Collin
+ <lasse.collin@tukaani.org>.
+
+ Major parts of liblzma are based on code written by Igor Pavlov,
+ specifically the LZMA SDK <http://7-zip.org/sdk.html>. Without
+ this code, XZ Utils wouldn't exist.
+
+ The SHA-256 implementation in liblzma is based on the code found from
+ 7-Zip <http://7-zip.org/>, which has a modified version of the SHA-256
+ code found from Crypto++ <http://www.cryptopp.com/>. The SHA-256 code
+ in Crypto++ was written by Kevin Springle and Wei Dai.
+
+ Some scripts have been adapted from gzip. The original versions
+ were written by Jean-loup Gailly, Charles Levert, and Paul Eggert.
+ Andrew Dudman helped adapting the script and their man pages for
+ XZ Utils.
+
+ The GNU Autotools based build system contains files from many authors,
+ which I'm not trying list here.
+
+ Several people have contributed fixes or reported bugs. Most of them
+ are mentioned in the file THANKS.
+
diff --git a/COPYING b/COPYING
new file mode 100644
index 0000000..43c90d0
--- /dev/null
+++ b/COPYING
@@ -0,0 +1,65 @@
+
+XZ Utils Licensing
+==================
+
+ Different licenses apply to different files in this package. Here
+ is a rough summary of which licenses apply to which parts of this
+ package (but check the individual files to be sure!):
+
+ - liblzma is in the public domain.
+
+ - xz, xzdec, and lzmadec command line tools are in the public
+ domain unless GNU getopt_long had to be compiled and linked
+ in from the lib directory. The getopt_long code is under
+ GNU LGPLv2.1+.
+
+ - The scripts to grep, diff, and view compressed files have been
+ adapted from gzip. These scripts and their documentation are
+ under GNU GPLv2+.
+
+ - All the documentation in the doc directory and most of the
+ XZ Utils specific documentation files in other directories
+ are in the public domain.
+
+ - Translated messages are in the public domain.
+
+ - The build system contains public domain files, and files that
+ are under GNU GPLv2+ or GNU GPLv3+. None of these files end up
+ in the binaries being built.
+
+ - Test files and test code in the tests directory, and debugging
+ utilities in the debug directory are in the public domain.
+
+ - The extra directory may contain public domain files, and files
+ that are under various free software licenses.
+
+ You can do whatever you want with the files that have been put into
+ the public domain. If you find public domain legally problematic,
+ take the previous sentence as a license grant. If you still find
+ the lack of copyright legally problematic, you have too many
+ lawyers.
+
+ As usual, this software is provided "as is", without any warranty.
+
+ If you copy significant amounts of public domain code from XZ Utils
+ into your project, acknowledging this somewhere in your software is
+ polite (especially if it is proprietary, non-free software), but
+ naturally it is not legally required. Here is an example of a good
+ notice to put into "about box" or into documentation:
+
+ This software includes code from XZ Utils <http://tukaani.org/xz/>.
+
+ The following license texts are included in the following files:
+ - COPYING.LGPLv2.1: GNU Lesser General Public License version 2.1
+ - COPYING.GPLv2: GNU General Public License version 2
+ - COPYING.GPLv3: GNU General Public License version 3
+
+ Note that the toolchain (compiler, linker etc.) may add some code
+ pieces that are copyrighted. Thus, it is possible that e.g. liblzma
+ binary wouldn't actually be in the public domain in its entirety
+ even though it contains no copyrighted code from the XZ Utils source
+ package.
+
+ If you have questions, don't hesitate to ask the author(s) for more
+ information.
+
diff --git a/COPYING.GPLv2 b/COPYING.GPLv2
new file mode 100644
index 0000000..d511905
--- /dev/null
+++ b/COPYING.GPLv2
@@ -0,0 +1,339 @@
+ GNU GENERAL PUBLIC LICENSE
+ Version 2, June 1991
+
+ Copyright (C) 1989, 1991 Free Software Foundation, Inc.,
+ 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+ Preamble
+
+ The licenses for most software are designed to take away your
+freedom to share and change it. By contrast, the GNU General Public
+License is intended to guarantee your freedom to share and change free
+software--to make sure the software is free for all its users. This
+General Public License applies to most of the Free Software
+Foundation's software and to any other program whose authors commit to
+using it. (Some other Free Software Foundation software is covered by
+the GNU Lesser General Public License instead.) You can apply it to
+your programs, too.
+
+ When we speak of free software, we are referring to freedom, not
+price. Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+this service if you wish), that you receive source code or can get it
+if you want it, that you can change the software or use pieces of it
+in new free programs; and that you know you can do these things.
+
+ To protect your rights, we need to make restrictions that forbid
+anyone to deny you these rights or to ask you to surrender the rights.
+These restrictions translate to certain responsibilities for you if you
+distribute copies of the software, or if you modify it.
+
+ For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must give the recipients all the rights that
+you have. You must make sure that they, too, receive or can get the
+source code. And you must show them these terms so they know their
+rights.
+
+ We protect your rights with two steps: (1) copyright the software, and
+(2) offer you this license which gives you legal permission to copy,
+distribute and/or modify the software.
+
+ Also, for each author's protection and ours, we want to make certain
+that everyone understands that there is no warranty for this free
+software. If the software is modified by someone else and passed on, we
+want its recipients to know that what they have is not the original, so
+that any problems introduced by others will not reflect on the original
+authors' reputations.
+
+ Finally, any free program is threatened constantly by software
+patents. We wish to avoid the danger that redistributors of a free
+program will individually obtain patent licenses, in effect making the
+program proprietary. To prevent this, we have made it clear that any
+patent must be licensed for everyone's free use or not licensed at all.
+
+ The precise terms and conditions for copying, distribution and
+modification follow.
+
+ GNU GENERAL PUBLIC LICENSE
+ TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+
+ 0. This License applies to any program or other work which contains
+a notice placed by the copyright holder saying it may be distributed
+under the terms of this General Public License. The "Program", below,
+refers to any such program or work, and a "work based on the Program"
+means either the Program or any derivative work under copyright law:
+that is to say, a work containing the Program or a portion of it,
+either verbatim or with modifications and/or translated into another
+language. (Hereinafter, translation is included without limitation in
+the term "modification".) Each licensee is addressed as "you".
+
+Activities other than copying, distribution and modification are not
+covered by this License; they are outside its scope. The act of
+running the Program is not restricted, and the output from the Program
+is covered only if its contents constitute a work based on the
+Program (independent of having been made by running the Program).
+Whether that is true depends on what the Program does.
+
+ 1. You may copy and distribute verbatim copies of the Program's
+source code as you receive it, in any medium, provided that you
+conspicuously and appropriately publish on each copy an appropriate
+copyright notice and disclaimer of warranty; keep intact all the
+notices that refer to this License and to the absence of any warranty;
+and give any other recipients of the Program a copy of this License
+along with the Program.
+
+You may charge a fee for the physical act of transferring a copy, and
+you may at your option offer warranty protection in exchange for a fee.
+
+ 2. You may modify your copy or copies of the Program or any portion
+of it, thus forming a work based on the Program, and copy and
+distribute such modifications or work under the terms of Section 1
+above, provided that you also meet all of these conditions:
+
+ a) You must cause the modified files to carry prominent notices
+ stating that you changed the files and the date of any change.
+
+ b) You must cause any work that you distribute or publish, that in
+ whole or in part contains or is derived from the Program or any
+ part thereof, to be licensed as a whole at no charge to all third
+ parties under the terms of this License.
+
+ c) If the modified program normally reads commands interactively
+ when run, you must cause it, when started running for such
+ interactive use in the most ordinary way, to print or display an
+ announcement including an appropriate copyright notice and a
+ notice that there is no warranty (or else, saying that you provide
+ a warranty) and that users may redistribute the program under
+ these conditions, and telling the user how to view a copy of this
+ License. (Exception: if the Program itself is interactive but
+ does not normally print such an announcement, your work based on
+ the Program is not required to print an announcement.)
+
+These requirements apply to the modified work as a whole. If
+identifiable sections of that work are not derived from the Program,
+and can be reasonably considered independent and separate works in
+themselves, then this License, and its terms, do not apply to those
+sections when you distribute them as separate works. But when you
+distribute the same sections as part of a whole which is a work based
+on the Program, the distribution of the whole must be on the terms of
+this License, whose permissions for other licensees extend to the
+entire whole, and thus to each and every part regardless of who wrote it.
+
+Thus, it is not the intent of this section to claim rights or contest
+your rights to work written entirely by you; rather, the intent is to
+exercise the right to control the distribution of derivative or
+collective works based on the Program.
+
+In addition, mere aggregation of another work not based on the Program
+with the Program (or with a work based on the Program) on a volume of
+a storage or distribution medium does not bring the other work under
+the scope of this License.
+
+ 3. You may copy and distribute the Program (or a work based on it,
+under Section 2) in object code or executable form under the terms of
+Sections 1 and 2 above provided that you also do one of the following:
+
+ a) Accompany it with the complete corresponding machine-readable
+ source code, which must be distributed under the terms of Sections
+ 1 and 2 above on a medium customarily used for software interchange; or,
+
+ b) Accompany it with a written offer, valid for at least three
+ years, to give any third party, for a charge no more than your
+ cost of physically performing source distribution, a complete
+ machine-readable copy of the corresponding source code, to be
+ distributed under the terms of Sections 1 and 2 above on a medium
+ customarily used for software interchange; or,
+
+ c) Accompany it with the information you received as to the offer
+ to distribute corresponding source code. (This alternative is
+ allowed only for noncommercial distribution and only if you
+ received the program in object code or executable form with such
+ an offer, in accord with Subsection b above.)
+
+The source code for a work means the preferred form of the work for
+making modifications to it. For an executable work, complete source
+code means all the source code for all modules it contains, plus any
+associated interface definition files, plus the scripts used to
+control compilation and installation of the executable. However, as a
+special exception, the source code distributed need not include
+anything that is normally distributed (in either source or binary
+form) with the major components (compiler, kernel, and so on) of the
+operating system on which the executable runs, unless that component
+itself accompanies the executable.
+
+If distribution of executable or object code is made by offering
+access to copy from a designated place, then offering equivalent
+access to copy the source code from the same place counts as
+distribution of the source code, even though third parties are not
+compelled to copy the source along with the object code.
+
+ 4. You may not copy, modify, sublicense, or distribute the Program
+except as expressly provided under this License. Any attempt
+otherwise to copy, modify, sublicense or distribute the Program is
+void, and will automatically terminate your rights under this License.
+However, parties who have received copies, or rights, from you under
+this License will not have their licenses terminated so long as such
+parties remain in full compliance.
+
+ 5. You are not required to accept this License, since you have not
+signed it. However, nothing else grants you permission to modify or
+distribute the Program or its derivative works. These actions are
+prohibited by law if you do not accept this License. Therefore, by
+modifying or distributing the Program (or any work based on the
+Program), you indicate your acceptance of this License to do so, and
+all its terms and conditions for copying, distributing or modifying
+the Program or works based on it.
+
+ 6. Each time you redistribute the Program (or any work based on the
+Program), the recipient automatically receives a license from the
+original licensor to copy, distribute or modify the Program subject to
+these terms and conditions. You may not impose any further
+restrictions on the recipients' exercise of the rights granted herein.
+You are not responsible for enforcing compliance by third parties to
+this License.
+
+ 7. If, as a consequence of a court judgment or allegation of patent
+infringement or for any other reason (not limited to patent issues),
+conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot
+distribute so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you
+may not distribute the Program at all. For example, if a patent
+license would not permit royalty-free redistribution of the Program by
+all those who receive copies directly or indirectly through you, then
+the only way you could satisfy both it and this License would be to
+refrain entirely from distribution of the Program.
+
+If any portion of this section is held invalid or unenforceable under
+any particular circumstance, the balance of the section is intended to
+apply and the section as a whole is intended to apply in other
+circumstances.
+
+It is not the purpose of this section to induce you to infringe any
+patents or other property right claims or to contest validity of any
+such claims; this section has the sole purpose of protecting the
+integrity of the free software distribution system, which is
+implemented by public license practices. Many people have made
+generous contributions to the wide range of software distributed
+through that system in reliance on consistent application of that
+system; it is up to the author/donor to decide if he or she is willing
+to distribute software through any other system and a licensee cannot
+impose that choice.
+
+This section is intended to make thoroughly clear what is believed to
+be a consequence of the rest of this License.
+
+ 8. If the distribution and/or use of the Program is restricted in
+certain countries either by patents or by copyrighted interfaces, the
+original copyright holder who places the Program under this License
+may add an explicit geographical distribution limitation excluding
+those countries, so that distribution is permitted only in or among
+countries not thus excluded. In such case, this License incorporates
+the limitation as if written in the body of this License.
+
+ 9. The Free Software Foundation may publish revised and/or new versions
+of the General Public License from time to time. Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+Each version is given a distinguishing version number. If the Program
+specifies a version number of this License which applies to it and "any
+later version", you have the option of following the terms and conditions
+either of that version or of any later version published by the Free
+Software Foundation. If the Program does not specify a version number of
+this License, you may choose any version ever published by the Free Software
+Foundation.
+
+ 10. If you wish to incorporate parts of the Program into other free
+programs whose distribution conditions are different, write to the author
+to ask for permission. For software which is copyrighted by the Free
+Software Foundation, write to the Free Software Foundation; we sometimes
+make exceptions for this. Our decision will be guided by the two goals
+of preserving the free status of all derivatives of our free software and
+of promoting the sharing and reuse of software generally.
+
+ NO WARRANTY
+
+ 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
+FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
+OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
+PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
+OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
+TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
+PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
+REPAIR OR CORRECTION.
+
+ 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
+REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
+INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
+OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
+TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
+YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
+PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGES.
+
+ END OF TERMS AND CONDITIONS
+
+ How to Apply These Terms to Your New Programs
+
+ If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+ To do so, attach the following notices to the program. It is safest
+to attach them to the start of each source file to most effectively
+convey the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+ <one line to give the program's name and a brief idea of what it does.>
+ Copyright (C) <year> <name of author>
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License along
+ with this program; if not, write to the Free Software Foundation, Inc.,
+ 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+
+Also add information on how to contact you by electronic and paper mail.
+
+If the program is interactive, make it output a short notice like this
+when it starts in an interactive mode:
+
+ Gnomovision version 69, Copyright (C) year name of author
+ Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+ This is free software, and you are welcome to redistribute it
+ under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License. Of course, the commands you use may
+be called something other than `show w' and `show c'; they could even be
+mouse-clicks or menu items--whatever suits your program.
+
+You should also get your employer (if you work as a programmer) or your
+school, if any, to sign a "copyright disclaimer" for the program, if
+necessary. Here is a sample; alter the names:
+
+ Yoyodyne, Inc., hereby disclaims all copyright interest in the program
+ `Gnomovision' (which makes passes at compilers) written by James Hacker.
+
+ <signature of Ty Coon>, 1 April 1989
+ Ty Coon, President of Vice
+
+This General Public License does not permit incorporating your program into
+proprietary programs. If your program is a subroutine library, you may
+consider it more useful to permit linking proprietary applications with the
+library. If this is what you want to do, use the GNU Lesser General
+Public License instead of this License.
diff --git a/COPYING.GPLv3 b/COPYING.GPLv3
new file mode 100644
index 0000000..94a9ed0
--- /dev/null
+++ b/COPYING.GPLv3
@@ -0,0 +1,674 @@
+ GNU GENERAL PUBLIC LICENSE
+ Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+ Preamble
+
+ The GNU General Public License is a free, copyleft license for
+software and other kinds of works.
+
+ The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works. By contrast,
+the GNU General Public License is intended to guarantee your freedom to
+share and change all versions of a program--to make sure it remains free
+software for all its users. We, the Free Software Foundation, use the
+GNU General Public License for most of our software; it applies also to
+any other work released this way by its authors. You can apply it to
+your programs, too.
+
+ When we speak of free software, we are referring to freedom, not
+price. Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+them if you wish), that you receive source code or can get it if you
+want it, that you can change the software or use pieces of it in new
+free programs, and that you know you can do these things.
+
+ To protect your rights, we need to prevent others from denying you
+these rights or asking you to surrender the rights. Therefore, you have
+certain responsibilities if you distribute copies of the software, or if
+you modify it: responsibilities to respect the freedom of others.
+
+ For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must pass on to the recipients the same
+freedoms that you received. You must make sure that they, too, receive
+or can get the source code. And you must show them these terms so they
+know their rights.
+
+ Developers that use the GNU GPL protect your rights with two steps:
+(1) assert copyright on the software, and (2) offer you this License
+giving you legal permission to copy, distribute and/or modify it.
+
+ For the developers' and authors' protection, the GPL clearly explains
+that there is no warranty for this free software. For both users' and
+authors' sake, the GPL requires that modified versions be marked as
+changed, so that their problems will not be attributed erroneously to
+authors of previous versions.
+
+ Some devices are designed to deny users access to install or run
+modified versions of the software inside them, although the manufacturer
+can do so. This is fundamentally incompatible with the aim of
+protecting users' freedom to change the software. The systematic
+pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable. Therefore, we
+have designed this version of the GPL to prohibit the practice for those
+products. If such problems arise substantially in other domains, we
+stand ready to extend this provision to those domains in future versions
+of the GPL, as needed to protect the freedom of users.
+
+ Finally, every program is threatened constantly by software patents.
+States should not allow patents to restrict development and use of
+software on general-purpose computers, but in those that do, we wish to
+avoid the special danger that patents applied to a free program could
+make it effectively proprietary. To prevent this, the GPL assures that
+patents cannot be used to render the program non-free.
+
+ The precise terms and conditions for copying, distribution and
+modification follow.
+
+ TERMS AND CONDITIONS
+
+ 0. Definitions.
+
+ "This License" refers to version 3 of the GNU General Public License.
+
+ "Copyright" also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+ "The Program" refers to any copyrightable work licensed under this
+License. Each licensee is addressed as "you". "Licensees" and
+"recipients" may be individuals or organizations.
+
+ To "modify" a work means to copy from or adapt all or part of the work
+in a fashion requiring copyright permission, other than the making of an
+exact copy. The resulting work is called a "modified version" of the
+earlier work or a work "based on" the earlier work.
+
+ A "covered work" means either the unmodified Program or a work based
+on the Program.
+
+ To "propagate" a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for
+infringement under applicable copyright law, except executing it on a
+computer or modifying a private copy. Propagation includes copying,
+distribution (with or without modification), making available to the
+public, and in some countries other activities as well.
+
+ To "convey" a work means any kind of propagation that enables other
+parties to make or receive copies. Mere interaction with a user through
+a computer network, with no transfer of a copy, is not conveying.
+
+ An interactive user interface displays "Appropriate Legal Notices"
+to the extent that it includes a convenient and prominently visible
+feature that (1) displays an appropriate copyright notice, and (2)
+tells the user that there is no warranty for the work (except to the
+extent that warranties are provided), that licensees may convey the
+work under this License, and how to view a copy of this License. If
+the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+ 1. Source Code.
+
+ The "source code" for a work means the preferred form of the work
+for making modifications to it. "Object code" means any non-source
+form of a work.
+
+ A "Standard Interface" means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of
+interfaces specified for a particular programming language, one that
+is widely used among developers working in that language.
+
+ The "System Libraries" of an executable work include anything, other
+than the work as a whole, that (a) is included in the normal form of
+packaging a Major Component, but which is not part of that Major
+Component, and (b) serves only to enable use of the work with that
+Major Component, or to implement a Standard Interface for which an
+implementation is available to the public in source code form. A
+"Major Component", in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system
+(if any) on which the executable work runs, or a compiler used to
+produce the work, or an object code interpreter used to run it.
+
+ The "Corresponding Source" for a work in object code form means all
+the source code needed to generate, install, and (for an executable
+work) run the object code and to modify the work, including scripts to
+control those activities. However, it does not include the work's
+System Libraries, or general-purpose tools or generally available free
+programs which are used unmodified in performing those activities but
+which are not part of the work. For example, Corresponding Source
+includes interface definition files associated with source files for
+the work, and the source code for shared libraries and dynamically
+linked subprograms that the work is specifically designed to require,
+such as by intimate data communication or control flow between those
+subprograms and other parts of the work.
+
+ The Corresponding Source need not include anything that users
+can regenerate automatically from other parts of the Corresponding
+Source.
+
+ The Corresponding Source for a work in source code form is that
+same work.
+
+ 2. Basic Permissions.
+
+ All rights granted under this License are granted for the term of
+copyright on the Program, and are irrevocable provided the stated
+conditions are met. This License explicitly affirms your unlimited
+permission to run the unmodified Program. The output from running a
+covered work is covered by this License only if the output, given its
+content, constitutes a covered work. This License acknowledges your
+rights of fair use or other equivalent, as provided by copyright law.
+
+ You may make, run and propagate covered works that you do not
+convey, without conditions so long as your license otherwise remains
+in force. You may convey covered works to others for the sole purpose
+of having them make modifications exclusively for you, or provide you
+with facilities for running those works, provided that you comply with
+the terms of this License in conveying all material for which you do
+not control copyright. Those thus making or running the covered works
+for you must do so exclusively on your behalf, under your direction
+and control, on terms that prohibit them from making any copies of
+your copyrighted material outside their relationship with you.
+
+ Conveying under any other circumstances is permitted solely under
+the conditions stated below. Sublicensing is not allowed; section 10
+makes it unnecessary.
+
+ 3. Protecting Users' Legal Rights From Anti-Circumvention Law.
+
+ No covered work shall be deemed part of an effective technological
+measure under any applicable law fulfilling obligations under article
+11 of the WIPO copyright treaty adopted on 20 December 1996, or
+similar laws prohibiting or restricting circumvention of such
+measures.
+
+ When you convey a covered work, you waive any legal power to forbid
+circumvention of technological measures to the extent such circumvention
+is effected by exercising rights under this License with respect to
+the covered work, and you disclaim any intention to limit operation or
+modification of the work as a means of enforcing, against the work's
+users, your or third parties' legal rights to forbid circumvention of
+technological measures.
+
+ 4. Conveying Verbatim Copies.
+
+ You may convey verbatim copies of the Program's source code as you
+receive it, in any medium, provided that you conspicuously and
+appropriately publish on each copy an appropriate copyright notice;
+keep intact all notices stating that this License and any
+non-permissive terms added in accord with section 7 apply to the code;
+keep intact all notices of the absence of any warranty; and give all
+recipients a copy of this License along with the Program.
+
+ You may charge any price or no price for each copy that you convey,
+and you may offer support or warranty protection for a fee.
+
+ 5. Conveying Modified Source Versions.
+
+ You may convey a work based on the Program, or the modifications to
+produce it from the Program, in the form of source code under the
+terms of section 4, provided that you also meet all of these conditions:
+
+ a) The work must carry prominent notices stating that you modified
+ it, and giving a relevant date.
+
+ b) The work must carry prominent notices stating that it is
+ released under this License and any conditions added under section
+ 7. This requirement modifies the requirement in section 4 to
+ "keep intact all notices".
+
+ c) You must license the entire work, as a whole, under this
+ License to anyone who comes into possession of a copy. This
+ License will therefore apply, along with any applicable section 7
+ additional terms, to the whole of the work, and all its parts,
+ regardless of how they are packaged. This License gives no
+ permission to license the work in any other way, but it does not
+ invalidate such permission if you have separately received it.
+
+ d) If the work has interactive user interfaces, each must display
+ Appropriate Legal Notices; however, if the Program has interactive
+ interfaces that do not display Appropriate Legal Notices, your
+ work need not make them do so.
+
+ A compilation of a covered work with other separate and independent
+works, which are not by their nature extensions of the covered work,
+and which are not combined with it such as to form a larger program,
+in or on a volume of a storage or distribution medium, is called an
+"aggregate" if the compilation and its resulting copyright are not
+used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit. Inclusion of a covered work
+in an aggregate does not cause this License to apply to the other
+parts of the aggregate.
+
+ 6. Conveying Non-Source Forms.
+
+ You may convey a covered work in object code form under the terms
+of sections 4 and 5, provided that you also convey the
+machine-readable Corresponding Source under the terms of this License,
+in one of these ways:
+
+ a) Convey the object code in, or embodied in, a physical product
+ (including a physical distribution medium), accompanied by the
+ Corresponding Source fixed on a durable physical medium
+ customarily used for software interchange.
+
+ b) Convey the object code in, or embodied in, a physical product
+ (including a physical distribution medium), accompanied by a
+ written offer, valid for at least three years and valid for as
+ long as you offer spare parts or customer support for that product
+ model, to give anyone who possesses the object code either (1) a
+ copy of the Corresponding Source for all the software in the
+ product that is covered by this License, on a durable physical
+ medium customarily used for software interchange, for a price no
+ more than your reasonable cost of physically performing this
+ conveying of source, or (2) access to copy the
+ Corresponding Source from a network server at no charge.
+
+ c) Convey individual copies of the object code with a copy of the
+ written offer to provide the Corresponding Source. This
+ alternative is allowed only occasionally and noncommercially, and
+ only if you received the object code with such an offer, in accord
+ with subsection 6b.
+
+ d) Convey the object code by offering access from a designated
+ place (gratis or for a charge), and offer equivalent access to the
+ Corresponding Source in the same way through the same place at no
+ further charge. You need not require recipients to copy the
+ Corresponding Source along with the object code. If the place to
+ copy the object code is a network server, the Corresponding Source
+ may be on a different server (operated by you or a third party)
+ that supports equivalent copying facilities, provided you maintain
+ clear directions next to the object code saying where to find the
+ Corresponding Source. Regardless of what server hosts the
+ Corresponding Source, you remain obligated to ensure that it is
+ available for as long as needed to satisfy these requirements.
+
+ e) Convey the object code using peer-to-peer transmission, provided
+ you inform other peers where the object code and Corresponding
+ Source of the work are being offered to the general public at no
+ charge under subsection 6d.
+
+ A separable portion of the object code, whose source code is excluded
+from the Corresponding Source as a System Library, need not be
+included in conveying the object code work.
+
+ A "User Product" is either (1) a "consumer product", which means any
+tangible personal property which is normally used for personal, family,
+or household purposes, or (2) anything designed or sold for incorporation
+into a dwelling. In determining whether a product is a consumer product,
+doubtful cases shall be resolved in favor of coverage. For a particular
+product received by a particular user, "normally used" refers to a
+typical or common use of that class of product, regardless of the status
+of the particular user or of the way in which the particular user
+actually uses, or expects or is expected to use, the product. A product
+is a consumer product regardless of whether the product has substantial
+commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+ "Installation Information" for a User Product means any methods,
+procedures, authorization keys, or other information required to install
+and execute modified versions of a covered work in that User Product from
+a modified version of its Corresponding Source. The information must
+suffice to ensure that the continued functioning of the modified object
+code is in no case prevented or interfered with solely because
+modification has been made.
+
+ If you convey an object code work under this section in, or with, or
+specifically for use in, a User Product, and the conveying occurs as
+part of a transaction in which the right of possession and use of the
+User Product is transferred to the recipient in perpetuity or for a
+fixed term (regardless of how the transaction is characterized), the
+Corresponding Source conveyed under this section must be accompanied
+by the Installation Information. But this requirement does not apply
+if neither you nor any third party retains the ability to install
+modified object code on the User Product (for example, the work has
+been installed in ROM).
+
+ The requirement to provide Installation Information does not include a
+requirement to continue to provide support service, warranty, or updates
+for a work that has been modified or installed by the recipient, or for
+the User Product in which it has been modified or installed. Access to a
+network may be denied when the modification itself materially and
+adversely affects the operation of the network or violates the rules and
+protocols for communication across the network.
+
+ Corresponding Source conveyed, and Installation Information provided,
+in accord with this section must be in a format that is publicly
+documented (and with an implementation available to the public in
+source code form), and must require no special password or key for
+unpacking, reading or copying.
+
+ 7. Additional Terms.
+
+ "Additional permissions" are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions.
+Additional permissions that are applicable to the entire Program shall
+be treated as though they were included in this License, to the extent
+that they are valid under applicable law. If additional permissions
+apply only to part of the Program, that part may be used separately
+under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+ When you convey a copy of a covered work, you may at your option
+remove any additional permissions from that copy, or from any part of
+it. (Additional permissions may be written to require their own
+removal in certain cases when you modify the work.) You may place
+additional permissions on material, added by you to a covered work,
+for which you have or can give appropriate copyright permission.
+
+ Notwithstanding any other provision of this License, for material you
+add to a covered work, you may (if authorized by the copyright holders of
+that material) supplement the terms of this License with terms:
+
+ a) Disclaiming warranty or limiting liability differently from the
+ terms of sections 15 and 16 of this License; or
+
+ b) Requiring preservation of specified reasonable legal notices or
+ author attributions in that material or in the Appropriate Legal
+ Notices displayed by works containing it; or
+
+ c) Prohibiting misrepresentation of the origin of that material, or
+ requiring that modified versions of such material be marked in
+ reasonable ways as different from the original version; or
+
+ d) Limiting the use for publicity purposes of names of licensors or
+ authors of the material; or
+
+ e) Declining to grant rights under trademark law for use of some
+ trade names, trademarks, or service marks; or
+
+ f) Requiring indemnification of licensors and authors of that
+ material by anyone who conveys the material (or modified versions of
+ it) with contractual assumptions of liability to the recipient, for
+ any liability that these contractual assumptions directly impose on
+ those licensors and authors.
+
+ All other non-permissive additional terms are considered "further
+restrictions" within the meaning of section 10. If the Program as you
+received it, or any part of it, contains a notice stating that it is
+governed by this License along with a term that is a further
+restriction, you may remove that term. If a license document contains
+a further restriction but permits relicensing or conveying under this
+License, you may add to a covered work material governed by the terms
+of that license document, provided that the further restriction does
+not survive such relicensing or conveying.
+
+ If you add terms to a covered work in accord with this section, you
+must place, in the relevant source files, a statement of the
+additional terms that apply to those files, or a notice indicating
+where to find the applicable terms.
+
+ Additional terms, permissive or non-permissive, may be stated in the
+form of a separately written license, or stated as exceptions;
+the above requirements apply either way.
+
+ 8. Termination.
+
+ You may not propagate or modify a covered work except as expressly
+provided under this License. Any attempt otherwise to propagate or
+modify it is void, and will automatically terminate your rights under
+this License (including any patent licenses granted under the third
+paragraph of section 11).
+
+ However, if you cease all violation of this License, then your
+license from a particular copyright holder is reinstated (a)
+provisionally, unless and until the copyright holder explicitly and
+finally terminates your license, and (b) permanently, if the copyright
+holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+ Moreover, your license from a particular copyright holder is
+reinstated permanently if the copyright holder notifies you of the
+violation by some reasonable means, this is the first time you have
+received notice of violation of this License (for any work) from that
+copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+ Termination of your rights under this section does not terminate the
+licenses of parties who have received copies or rights from you under
+this License. If your rights have been terminated and not permanently
+reinstated, you do not qualify to receive new licenses for the same
+material under section 10.
+
+ 9. Acceptance Not Required for Having Copies.
+
+ You are not required to accept this License in order to receive or
+run a copy of the Program. Ancillary propagation of a covered work
+occurring solely as a consequence of using peer-to-peer transmission
+to receive a copy likewise does not require acceptance. However,
+nothing other than this License grants you permission to propagate or
+modify any covered work. These actions infringe copyright if you do
+not accept this License. Therefore, by modifying or propagating a
+covered work, you indicate your acceptance of this License to do so.
+
+ 10. Automatic Licensing of Downstream Recipients.
+
+ Each time you convey a covered work, the recipient automatically
+receives a license from the original licensors, to run, modify and
+propagate that work, subject to this License. You are not responsible
+for enforcing compliance by third parties with this License.
+
+ An "entity transaction" is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an
+organization, or merging organizations. If propagation of a covered
+work results from an entity transaction, each party to that
+transaction who receives a copy of the work also receives whatever
+licenses to the work the party's predecessor in interest had or could
+give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if
+the predecessor has it or can get it with reasonable efforts.
+
+ You may not impose any further restrictions on the exercise of the
+rights granted or affirmed under this License. For example, you may
+not impose a license fee, royalty, or other charge for exercise of
+rights granted under this License, and you may not initiate litigation
+(including a cross-claim or counterclaim in a lawsuit) alleging that
+any patent claim is infringed by making, using, selling, offering for
+sale, or importing the Program or any portion of it.
+
+ 11. Patents.
+
+ A "contributor" is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based. The
+work thus licensed is called the contributor's "contributor version".
+
+ A contributor's "essential patent claims" are all patent claims
+owned or controlled by the contributor, whether already acquired or
+hereafter acquired, that would be infringed by some manner, permitted
+by this License, of making, using, or selling its contributor version,
+but do not include claims that would be infringed only as a
+consequence of further modification of the contributor version. For
+purposes of this definition, "control" includes the right to grant
+patent sublicenses in a manner consistent with the requirements of
+this License.
+
+ Each contributor grants you a non-exclusive, worldwide, royalty-free
+patent license under the contributor's essential patent claims, to
+make, use, sell, offer for sale, import and otherwise run, modify and
+propagate the contents of its contributor version.
+
+ In the following three paragraphs, a "patent license" is any express
+agreement or commitment, however denominated, not to enforce a patent
+(such as an express permission to practice a patent or covenant not to
+sue for patent infringement). To "grant" such a patent license to a
+party means to make such an agreement or commitment not to enforce a
+patent against the party.
+
+ If you convey a covered work, knowingly relying on a patent license,
+and the Corresponding Source of the work is not available for anyone
+to copy, free of charge and under the terms of this License, through a
+publicly available network server or other readily accessible means,
+then you must either (1) cause the Corresponding Source to be so
+available, or (2) arrange to deprive yourself of the benefit of the
+patent license for this particular work, or (3) arrange, in a manner
+consistent with the requirements of this License, to extend the patent
+license to downstream recipients. "Knowingly relying" means you have
+actual knowledge that, but for the patent license, your conveying the
+covered work in a country, or your recipient's use of the covered work
+in a country, would infringe one or more identifiable patents in that
+country that you have reason to believe are valid.
+
+ If, pursuant to or in connection with a single transaction or
+arrangement, you convey, or propagate by procuring conveyance of, a
+covered work, and grant a patent license to some of the parties
+receiving the covered work authorizing them to use, propagate, modify
+or convey a specific copy of the covered work, then the patent license
+you grant is automatically extended to all recipients of the covered
+work and works based on it.
+
+ A patent license is "discriminatory" if it does not include within
+the scope of its coverage, prohibits the exercise of, or is
+conditioned on the non-exercise of one or more of the rights that are
+specifically granted under this License. You may not convey a covered
+work if you are a party to an arrangement with a third party that is
+in the business of distributing software, under which you make payment
+to the third party based on the extent of your activity of conveying
+the work, and under which the third party grants, to any of the
+parties who would receive the covered work from you, a discriminatory
+patent license (a) in connection with copies of the covered work
+conveyed by you (or copies made from those copies), or (b) primarily
+for and in connection with specific products or compilations that
+contain the covered work, unless you entered into that arrangement,
+or that patent license was granted, prior to 28 March 2007.
+
+ Nothing in this License shall be construed as excluding or limiting
+any implied license or other defenses to infringement that may
+otherwise be available to you under applicable patent law.
+
+ 12. No Surrender of Others' Freedom.
+
+ If conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot convey a
+covered work so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you may
+not convey it at all. For example, if you agree to terms that obligate you
+to collect a royalty for further conveying from those to whom you convey
+the Program, the only way you could satisfy both those terms and this
+License would be to refrain entirely from conveying the Program.
+
+ 13. Use with the GNU Affero General Public License.
+
+ Notwithstanding any other provision of this License, you have
+permission to link or combine any covered work with a work licensed
+under version 3 of the GNU Affero General Public License into a single
+combined work, and to convey the resulting work. The terms of this
+License will continue to apply to the part which is the covered work,
+but the special requirements of the GNU Affero General Public License,
+section 13, concerning interaction through a network will apply to the
+combination as such.
+
+ 14. Revised Versions of this License.
+
+ The Free Software Foundation may publish revised and/or new versions of
+the GNU General Public License from time to time. Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+ Each version is given a distinguishing version number. If the
+Program specifies that a certain numbered version of the GNU General
+Public License "or any later version" applies to it, you have the
+option of following the terms and conditions either of that numbered
+version or of any later version published by the Free Software
+Foundation. If the Program does not specify a version number of the
+GNU General Public License, you may choose any version ever published
+by the Free Software Foundation.
+
+ If the Program specifies that a proxy can decide which future
+versions of the GNU General Public License can be used, that proxy's
+public statement of acceptance of a version permanently authorizes you
+to choose that version for the Program.
+
+ Later license versions may give you additional or different
+permissions. However, no additional obligations are imposed on any
+author or copyright holder as a result of your choosing to follow a
+later version.
+
+ 15. Disclaimer of Warranty.
+
+ THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
+APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
+HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
+OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
+THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
+IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
+ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+ 16. Limitation of Liability.
+
+ IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
+THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
+GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
+USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
+DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
+PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
+EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGES.
+
+ 17. Interpretation of Sections 15 and 16.
+
+ If the disclaimer of warranty and limitation of liability provided
+above cannot be given local legal effect according to their terms,
+reviewing courts shall apply local law that most closely approximates
+an absolute waiver of all civil liability in connection with the
+Program, unless a warranty or assumption of liability accompanies a
+copy of the Program in return for a fee.
+
+ END OF TERMS AND CONDITIONS
+
+ How to Apply These Terms to Your New Programs
+
+ If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+ To do so, attach the following notices to the program. It is safest
+to attach them to the start of each source file to most effectively
+state the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+ <one line to give the program's name and a brief idea of what it does.>
+ Copyright (C) <year> <name of author>
+
+ This program is free software: you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation, either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+ If the program does terminal interaction, make it output a short
+notice like this when it starts in an interactive mode:
+
+ <program> Copyright (C) <year> <name of author>
+ This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+ This is free software, and you are welcome to redistribute it
+ under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License. Of course, your program's commands
+might be different; for a GUI interface, you would use an "about box".
+
+ You should also get your employer (if you work as a programmer) or school,
+if any, to sign a "copyright disclaimer" for the program, if necessary.
+For more information on this, and how to apply and follow the GNU GPL, see
+<http://www.gnu.org/licenses/>.
+
+ The GNU General Public License does not permit incorporating your program
+into proprietary programs. If your program is a subroutine library, you
+may consider it more useful to permit linking proprietary applications with
+the library. If this is what you want to do, use the GNU Lesser General
+Public License instead of this License. But first, please read
+<http://www.gnu.org/philosophy/why-not-lgpl.html>.
diff --git a/COPYING.LGPLv2.1 b/COPYING.LGPLv2.1
new file mode 100644
index 0000000..5ab7695
--- /dev/null
+++ b/COPYING.LGPLv2.1
@@ -0,0 +1,504 @@
+ GNU LESSER GENERAL PUBLIC LICENSE
+ Version 2.1, February 1999
+
+ Copyright (C) 1991, 1999 Free Software Foundation, Inc.
+ 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+[This is the first released version of the Lesser GPL. It also counts
+ as the successor of the GNU Library Public License, version 2, hence
+ the version number 2.1.]
+
+ Preamble
+
+ The licenses for most software are designed to take away your
+freedom to share and change it. By contrast, the GNU General Public
+Licenses are intended to guarantee your freedom to share and change
+free software--to make sure the software is free for all its users.
+
+ This license, the Lesser General Public License, applies to some
+specially designated software packages--typically libraries--of the
+Free Software Foundation and other authors who decide to use it. You
+can use it too, but we suggest you first think carefully about whether
+this license or the ordinary General Public License is the better
+strategy to use in any particular case, based on the explanations below.
+
+ When we speak of free software, we are referring to freedom of use,
+not price. Our General Public Licenses are designed to make sure that
+you have the freedom to distribute copies of free software (and charge
+for this service if you wish); that you receive source code or can get
+it if you want it; that you can change the software and use pieces of
+it in new free programs; and that you are informed that you can do
+these things.
+
+ To protect your rights, we need to make restrictions that forbid
+distributors to deny you these rights or to ask you to surrender these
+rights. These restrictions translate to certain responsibilities for
+you if you distribute copies of the library or if you modify it.
+
+ For example, if you distribute copies of the library, whether gratis
+or for a fee, you must give the recipients all the rights that we gave
+you. You must make sure that they, too, receive or can get the source
+code. If you link other code with the library, you must provide
+complete object files to the recipients, so that they can relink them
+with the library after making changes to the library and recompiling
+it. And you must show them these terms so they know their rights.
+
+ We protect your rights with a two-step method: (1) we copyright the
+library, and (2) we offer you this license, which gives you legal
+permission to copy, distribute and/or modify the library.
+
+ To protect each distributor, we want to make it very clear that
+there is no warranty for the free library. Also, if the library is
+modified by someone else and passed on, the recipients should know
+that what they have is not the original version, so that the original
+author's reputation will not be affected by problems that might be
+introduced by others.
+
+ Finally, software patents pose a constant threat to the existence of
+any free program. We wish to make sure that a company cannot
+effectively restrict the users of a free program by obtaining a
+restrictive license from a patent holder. Therefore, we insist that
+any patent license obtained for a version of the library must be
+consistent with the full freedom of use specified in this license.
+
+ Most GNU software, including some libraries, is covered by the
+ordinary GNU General Public License. This license, the GNU Lesser
+General Public License, applies to certain designated libraries, and
+is quite different from the ordinary General Public License. We use
+this license for certain libraries in order to permit linking those
+libraries into non-free programs.
+
+ When a program is linked with a library, whether statically or using
+a shared library, the combination of the two is legally speaking a
+combined work, a derivative of the original library. The ordinary
+General Public License therefore permits such linking only if the
+entire combination fits its criteria of freedom. The Lesser General
+Public License permits more lax criteria for linking other code with
+the library.
+
+ We call this license the "Lesser" General Public License because it
+does Less to protect the user's freedom than the ordinary General
+Public License. It also provides other free software developers Less
+of an advantage over competing non-free programs. These disadvantages
+are the reason we use the ordinary General Public License for many
+libraries. However, the Lesser license provides advantages in certain
+special circumstances.
+
+ For example, on rare occasions, there may be a special need to
+encourage the widest possible use of a certain library, so that it becomes
+a de-facto standard. To achieve this, non-free programs must be
+allowed to use the library. A more frequent case is that a free
+library does the same job as widely used non-free libraries. In this
+case, there is little to gain by limiting the free library to free
+software only, so we use the Lesser General Public License.
+
+ In other cases, permission to use a particular library in non-free
+programs enables a greater number of people to use a large body of
+free software. For example, permission to use the GNU C Library in
+non-free programs enables many more people to use the whole GNU
+operating system, as well as its variant, the GNU/Linux operating
+system.
+
+ Although the Lesser General Public License is Less protective of the
+users' freedom, it does ensure that the user of a program that is
+linked with the Library has the freedom and the wherewithal to run
+that program using a modified version of the Library.
+
+ The precise terms and conditions for copying, distribution and
+modification follow. Pay close attention to the difference between a
+"work based on the library" and a "work that uses the library". The
+former contains code derived from the library, whereas the latter must
+be combined with the library in order to run.
+
+ GNU LESSER GENERAL PUBLIC LICENSE
+ TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+
+ 0. This License Agreement applies to any software library or other
+program which contains a notice placed by the copyright holder or
+other authorized party saying it may be distributed under the terms of
+this Lesser General Public License (also called "this License").
+Each licensee is addressed as "you".
+
+ A "library" means a collection of software functions and/or data
+prepared so as to be conveniently linked with application programs
+(which use some of those functions and data) to form executables.
+
+ The "Library", below, refers to any such software library or work
+which has been distributed under these terms. A "work based on the
+Library" means either the Library or any derivative work under
+copyright law: that is to say, a work containing the Library or a
+portion of it, either verbatim or with modifications and/or translated
+straightforwardly into another language. (Hereinafter, translation is
+included without limitation in the term "modification".)
+
+ "Source code" for a work means the preferred form of the work for
+making modifications to it. For a library, complete source code means
+all the source code for all modules it contains, plus any associated
+interface definition files, plus the scripts used to control compilation
+and installation of the library.
+
+ Activities other than copying, distribution and modification are not
+covered by this License; they are outside its scope. The act of
+running a program using the Library is not restricted, and output from
+such a program is covered only if its contents constitute a work based
+on the Library (independent of the use of the Library in a tool for
+writing it). Whether that is true depends on what the Library does
+and what the program that uses the Library does.
+
+ 1. You may copy and distribute verbatim copies of the Library's
+complete source code as you receive it, in any medium, provided that
+you conspicuously and appropriately publish on each copy an
+appropriate copyright notice and disclaimer of warranty; keep intact
+all the notices that refer to this License and to the absence of any
+warranty; and distribute a copy of this License along with the
+Library.
+
+ You may charge a fee for the physical act of transferring a copy,
+and you may at your option offer warranty protection in exchange for a
+fee.
+
+ 2. You may modify your copy or copies of the Library or any portion
+of it, thus forming a work based on the Library, and copy and
+distribute such modifications or work under the terms of Section 1
+above, provided that you also meet all of these conditions:
+
+ a) The modified work must itself be a software library.
+
+ b) You must cause the files modified to carry prominent notices
+ stating that you changed the files and the date of any change.
+
+ c) You must cause the whole of the work to be licensed at no
+ charge to all third parties under the terms of this License.
+
+ d) If a facility in the modified Library refers to a function or a
+ table of data to be supplied by an application program that uses
+ the facility, other than as an argument passed when the facility
+ is invoked, then you must make a good faith effort to ensure that,
+ in the event an application does not supply such function or
+ table, the facility still operates, and performs whatever part of
+ its purpose remains meaningful.
+
+ (For example, a function in a library to compute square roots has
+ a purpose that is entirely well-defined independent of the
+ application. Therefore, Subsection 2d requires that any
+ application-supplied function or table used by this function must
+ be optional: if the application does not supply it, the square
+ root function must still compute square roots.)
+
+These requirements apply to the modified work as a whole. If
+identifiable sections of that work are not derived from the Library,
+and can be reasonably considered independent and separate works in
+themselves, then this License, and its terms, do not apply to those
+sections when you distribute them as separate works. But when you
+distribute the same sections as part of a whole which is a work based
+on the Library, the distribution of the whole must be on the terms of
+this License, whose permissions for other licensees extend to the
+entire whole, and thus to each and every part regardless of who wrote
+it.
+
+Thus, it is not the intent of this section to claim rights or contest
+your rights to work written entirely by you; rather, the intent is to
+exercise the right to control the distribution of derivative or
+collective works based on the Library.
+
+In addition, mere aggregation of another work not based on the Library
+with the Library (or with a work based on the Library) on a volume of
+a storage or distribution medium does not bring the other work under
+the scope of this License.
+
+ 3. You may opt to apply the terms of the ordinary GNU General Public
+License instead of this License to a given copy of the Library. To do
+this, you must alter all the notices that refer to this License, so
+that they refer to the ordinary GNU General Public License, version 2,
+instead of to this License. (If a newer version than version 2 of the
+ordinary GNU General Public License has appeared, then you can specify
+that version instead if you wish.) Do not make any other change in
+these notices.
+
+ Once this change is made in a given copy, it is irreversible for
+that copy, so the ordinary GNU General Public License applies to all
+subsequent copies and derivative works made from that copy.
+
+ This option is useful when you wish to copy part of the code of
+the Library into a program that is not a library.
+
+ 4. You may copy and distribute the Library (or a portion or
+derivative of it, under Section 2) in object code or executable form
+under the terms of Sections 1 and 2 above provided that you accompany
+it with the complete corresponding machine-readable source code, which
+must be distributed under the terms of Sections 1 and 2 above on a
+medium customarily used for software interchange.
+
+ If distribution of object code is made by offering access to copy
+from a designated place, then offering equivalent access to copy the
+source code from the same place satisfies the requirement to
+distribute the source code, even though third parties are not
+compelled to copy the source along with the object code.
+
+ 5. A program that contains no derivative of any portion of the
+Library, but is designed to work with the Library by being compiled or
+linked with it, is called a "work that uses the Library". Such a
+work, in isolation, is not a derivative work of the Library, and
+therefore falls outside the scope of this License.
+
+ However, linking a "work that uses the Library" with the Library
+creates an executable that is a derivative of the Library (because it
+contains portions of the Library), rather than a "work that uses the
+library". The executable is therefore covered by this License.
+Section 6 states terms for distribution of such executables.
+
+ When a "work that uses the Library" uses material from a header file
+that is part of the Library, the object code for the work may be a
+derivative work of the Library even though the source code is not.
+Whether this is true is especially significant if the work can be
+linked without the Library, or if the work is itself a library. The
+threshold for this to be true is not precisely defined by law.
+
+ If such an object file uses only numerical parameters, data
+structure layouts and accessors, and small macros and small inline
+functions (ten lines or less in length), then the use of the object
+file is unrestricted, regardless of whether it is legally a derivative
+work. (Executables containing this object code plus portions of the
+Library will still fall under Section 6.)
+
+ Otherwise, if the work is a derivative of the Library, you may
+distribute the object code for the work under the terms of Section 6.
+Any executables containing that work also fall under Section 6,
+whether or not they are linked directly with the Library itself.
+
+ 6. As an exception to the Sections above, you may also combine or
+link a "work that uses the Library" with the Library to produce a
+work containing portions of the Library, and distribute that work
+under terms of your choice, provided that the terms permit
+modification of the work for the customer's own use and reverse
+engineering for debugging such modifications.
+
+ You must give prominent notice with each copy of the work that the
+Library is used in it and that the Library and its use are covered by
+this License. You must supply a copy of this License. If the work
+during execution displays copyright notices, you must include the
+copyright notice for the Library among them, as well as a reference
+directing the user to the copy of this License. Also, you must do one
+of these things:
+
+ a) Accompany the work with the complete corresponding
+ machine-readable source code for the Library including whatever
+ changes were used in the work (which must be distributed under
+ Sections 1 and 2 above); and, if the work is an executable linked
+ with the Library, with the complete machine-readable "work that
+ uses the Library", as object code and/or source code, so that the
+ user can modify the Library and then relink to produce a modified
+ executable containing the modified Library. (It is understood
+ that the user who changes the contents of definitions files in the
+ Library will not necessarily be able to recompile the application
+ to use the modified definitions.)
+
+ b) Use a suitable shared library mechanism for linking with the
+ Library. A suitable mechanism is one that (1) uses at run time a
+ copy of the library already present on the user's computer system,
+ rather than copying library functions into the executable, and (2)
+ will operate properly with a modified version of the library, if
+ the user installs one, as long as the modified version is
+ interface-compatible with the version that the work was made with.
+
+ c) Accompany the work with a written offer, valid for at
+ least three years, to give the same user the materials
+ specified in Subsection 6a, above, for a charge no more
+ than the cost of performing this distribution.
+
+ d) If distribution of the work is made by offering access to copy
+ from a designated place, offer equivalent access to copy the above
+ specified materials from the same place.
+
+ e) Verify that the user has already received a copy of these
+ materials or that you have already sent this user a copy.
+
+ For an executable, the required form of the "work that uses the
+Library" must include any data and utility programs needed for
+reproducing the executable from it. However, as a special exception,
+the materials to be distributed need not include anything that is
+normally distributed (in either source or binary form) with the major
+components (compiler, kernel, and so on) of the operating system on
+which the executable runs, unless that component itself accompanies
+the executable.
+
+ It may happen that this requirement contradicts the license
+restrictions of other proprietary libraries that do not normally
+accompany the operating system. Such a contradiction means you cannot
+use both them and the Library together in an executable that you
+distribute.
+
+ 7. You may place library facilities that are a work based on the
+Library side-by-side in a single library together with other library
+facilities not covered by this License, and distribute such a combined
+library, provided that the separate distribution of the work based on
+the Library and of the other library facilities is otherwise
+permitted, and provided that you do these two things:
+
+ a) Accompany the combined library with a copy of the same work
+ based on the Library, uncombined with any other library
+ facilities. This must be distributed under the terms of the
+ Sections above.
+
+ b) Give prominent notice with the combined library of the fact
+ that part of it is a work based on the Library, and explaining
+ where to find the accompanying uncombined form of the same work.
+
+ 8. You may not copy, modify, sublicense, link with, or distribute
+the Library except as expressly provided under this License. Any
+attempt otherwise to copy, modify, sublicense, link with, or
+distribute the Library is void, and will automatically terminate your
+rights under this License. However, parties who have received copies,
+or rights, from you under this License will not have their licenses
+terminated so long as such parties remain in full compliance.
+
+ 9. You are not required to accept this License, since you have not
+signed it. However, nothing else grants you permission to modify or
+distribute the Library or its derivative works. These actions are
+prohibited by law if you do not accept this License. Therefore, by
+modifying or distributing the Library (or any work based on the
+Library), you indicate your acceptance of this License to do so, and
+all its terms and conditions for copying, distributing or modifying
+the Library or works based on it.
+
+ 10. Each time you redistribute the Library (or any work based on the
+Library), the recipient automatically receives a license from the
+original licensor to copy, distribute, link with or modify the Library
+subject to these terms and conditions. You may not impose any further
+restrictions on the recipients' exercise of the rights granted herein.
+You are not responsible for enforcing compliance by third parties with
+this License.
+
+ 11. If, as a consequence of a court judgment or allegation of patent
+infringement or for any other reason (not limited to patent issues),
+conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot
+distribute so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you
+may not distribute the Library at all. For example, if a patent
+license would not permit royalty-free redistribution of the Library by
+all those who receive copies directly or indirectly through you, then
+the only way you could satisfy both it and this License would be to
+refrain entirely from distribution of the Library.
+
+If any portion of this section is held invalid or unenforceable under any
+particular circumstance, the balance of the section is intended to apply,
+and the section as a whole is intended to apply in other circumstances.
+
+It is not the purpose of this section to induce you to infringe any
+patents or other property right claims or to contest validity of any
+such claims; this section has the sole purpose of protecting the
+integrity of the free software distribution system which is
+implemented by public license practices. Many people have made
+generous contributions to the wide range of software distributed
+through that system in reliance on consistent application of that
+system; it is up to the author/donor to decide if he or she is willing
+to distribute software through any other system and a licensee cannot
+impose that choice.
+
+This section is intended to make thoroughly clear what is believed to
+be a consequence of the rest of this License.
+
+ 12. If the distribution and/or use of the Library is restricted in
+certain countries either by patents or by copyrighted interfaces, the
+original copyright holder who places the Library under this License may add
+an explicit geographical distribution limitation excluding those countries,
+so that distribution is permitted only in or among countries not thus
+excluded. In such case, this License incorporates the limitation as if
+written in the body of this License.
+
+ 13. The Free Software Foundation may publish revised and/or new
+versions of the Lesser General Public License from time to time.
+Such new versions will be similar in spirit to the present version,
+but may differ in detail to address new problems or concerns.
+
+Each version is given a distinguishing version number. If the Library
+specifies a version number of this License which applies to it and
+"any later version", you have the option of following the terms and
+conditions either of that version or of any later version published by
+the Free Software Foundation. If the Library does not specify a
+license version number, you may choose any version ever published by
+the Free Software Foundation.
+
+ 14. If you wish to incorporate parts of the Library into other free
+programs whose distribution conditions are incompatible with these,
+write to the author to ask for permission. For software which is
+copyrighted by the Free Software Foundation, write to the Free
+Software Foundation; we sometimes make exceptions for this. Our
+decision will be guided by the two goals of preserving the free status
+of all derivatives of our free software and of promoting the sharing
+and reuse of software generally.
+
+ NO WARRANTY
+
+ 15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
+WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
+EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
+OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY
+KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE
+LIBRARY IS WITH YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME
+THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+ 16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
+WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY
+AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU
+FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR
+CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
+LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING
+RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A
+FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
+SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
+DAMAGES.
+
+ END OF TERMS AND CONDITIONS
+
+ How to Apply These Terms to Your New Libraries
+
+ If you develop a new library, and you want it to be of the greatest
+possible use to the public, we recommend making it free software that
+everyone can redistribute and change. You can do so by permitting
+redistribution under these terms (or, alternatively, under the terms of the
+ordinary General Public License).
+
+ To apply these terms, attach the following notices to the library. It is
+safest to attach them to the start of each source file to most effectively
+convey the exclusion of warranty; and each file should have at least the
+"copyright" line and a pointer to where the full notice is found.
+
+ <one line to give the library's name and a brief idea of what it does.>
+ Copyright (C) <year> <name of author>
+
+ This library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ This library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with this library; if not, write to the Free Software
+ Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+
+Also add information on how to contact you by electronic and paper mail.
+
+You should also get your employer (if you work as a programmer) or your
+school, if any, to sign a "copyright disclaimer" for the library, if
+necessary. Here is a sample; alter the names:
+
+ Yoyodyne, Inc., hereby disclaims all copyright interest in the
+ library `Frob' (a library for tweaking knobs) written by James Random Hacker.
+
+ <signature of Ty Coon>, 1 April 1990
+ Ty Coon, President of Vice
+
+That's all there is to it!
+
+
diff --git a/ChangeLog b/ChangeLog
new file mode 100644
index 0000000..ff22a97
--- /dev/null
+++ b/ChangeLog
@@ -0,0 +1,7 @@
+See the commit log in the git repository:
+
+ git://ctrl.tukaani.org/xz.git
+
+Note that "make dist" doesn't put this tiny file into the package.
+Instead, the git commit log is used as ChangeLog. See dist-hook in
+Makefile.am for details.
diff --git a/Doxyfile.in b/Doxyfile.in
new file mode 100644
index 0000000..fa3c1e1
--- /dev/null
+++ b/Doxyfile.in
@@ -0,0 +1,1234 @@
+# Doxyfile 1.4.7
+
+# Copyright (C) 1997-2007 by Dimitri van Heesch
+# License: GNU GPLv2+
+
+# This file describes the settings to be used by the documentation system
+# doxygen (www.doxygen.org) for a project
+#
+# All text after a hash (#) is considered a comment and will be ignored
+# The format is:
+# TAG = value [value, ...]
+# For lists items can also be appended using:
+# TAG += value [value, ...]
+# Values that contain spaces should be placed between quotes (" ")
+
+#---------------------------------------------------------------------------
+# Project related configuration options
+#---------------------------------------------------------------------------
+
+# The PROJECT_NAME tag is a single word (or a sequence of words surrounded
+# by quotes) that should identify the project.
+
+PROJECT_NAME = "@PACKAGE_NAME@"
+
+# The PROJECT_NUMBER tag can be used to enter a project or revision number.
+# This could be handy for archiving the generated documentation or
+# if some version control system is used.
+
+PROJECT_NUMBER = "@PACKAGE_VERSION@"
+
+# The OUTPUT_DIRECTORY tag is used to specify the (relative or absolute)
+# base path where the generated documentation will be put.
+# If a relative path is entered, it will be relative to the location
+# where doxygen was started. If left blank the current directory will be used.
+
+OUTPUT_DIRECTORY = doc
+
+# If the CREATE_SUBDIRS tag is set to YES, then doxygen will create
+# 4096 sub-directories (in 2 levels) under the output directory of each output
+# format and will distribute the generated files over these directories.
+# Enabling this option can be useful when feeding doxygen a huge amount of
+# source files, where putting all generated files in the same directory would
+# otherwise cause performance problems for the file system.
+
+CREATE_SUBDIRS = NO
+
+# The OUTPUT_LANGUAGE tag is used to specify the language in which all
+# documentation generated by doxygen is written. Doxygen will use this
+# information to generate all constant output in the proper language.
+# The default language is English, other supported languages are:
+# Brazilian, Catalan, Chinese, Chinese-Traditional, Croatian, Czech, Danish,
+# Dutch, Finnish, French, German, Greek, Hungarian, Italian, Japanese,
+# Japanese-en (Japanese with English messages), Korean, Korean-en, Norwegian,
+# Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Slovene, Spanish,
+# Swedish, and Ukrainian.
+
+OUTPUT_LANGUAGE = English
+
+# This tag can be used to specify the encoding used in the generated output.
+# The encoding is not always determined by the language that is chosen,
+# but also whether or not the output is meant for Windows or non-Windows users.
+# In case there is a difference, setting the USE_WINDOWS_ENCODING tag to YES
+# forces the Windows encoding (this is the default for the Windows binary),
+# whereas setting the tag to NO uses a Unix-style encoding (the default for
+# all platforms other than Windows).
+
+USE_WINDOWS_ENCODING = NO
+
+# If the BRIEF_MEMBER_DESC tag is set to YES (the default) Doxygen will
+# include brief member descriptions after the members that are listed in
+# the file and class documentation (similar to JavaDoc).
+# Set to NO to disable this.
+
+BRIEF_MEMBER_DESC = YES
+
+# If the REPEAT_BRIEF tag is set to YES (the default) Doxygen will prepend
+# the brief description of a member or function before the detailed description.
+# Note: if both HIDE_UNDOC_MEMBERS and BRIEF_MEMBER_DESC are set to NO, the
+# brief descriptions will be completely suppressed.
+
+REPEAT_BRIEF = YES
+
+# This tag implements a quasi-intelligent brief description abbreviator
+# that is used to form the text in various listings. Each string
+# in this list, if found as the leading text of the brief description, will be
+# stripped from the text and the result after processing the whole list, is
+# used as the annotated text. Otherwise, the brief description is used as-is.
+# If left blank, the following values are used ("$name" is automatically
+# replaced with the name of the entity): "The $name class" "The $name widget"
+# "The $name file" "is" "provides" "specifies" "contains"
+# "represents" "a" "an" "the"
+
+ABBREVIATE_BRIEF =
+
+# If the ALWAYS_DETAILED_SEC and REPEAT_BRIEF tags are both set to YES then
+# Doxygen will generate a detailed section even if there is only a brief
+# description.
+
+ALWAYS_DETAILED_SEC = YES
+
+# If the INLINE_INHERITED_MEMB tag is set to YES, doxygen will show all
+# inherited members of a class in the documentation of that class as if those
+# members were ordinary class members. Constructors, destructors and assignment
+# operators of the base classes will not be shown.
+
+INLINE_INHERITED_MEMB = NO
+
+# If the FULL_PATH_NAMES tag is set to YES then Doxygen will prepend the full
+# path before files name in the file list and in the header files. If set
+# to NO the shortest path that makes the file name unique will be used.
+
+FULL_PATH_NAMES = YES
+
+# If the FULL_PATH_NAMES tag is set to YES then the STRIP_FROM_PATH tag
+# can be used to strip a user-defined part of the path. Stripping is
+# only done if one of the specified strings matches the left-hand part of
+# the path. The tag can be used to show relative paths in the file list.
+# If left blank the directory from which doxygen is run is used as the
+# path to strip.
+
+STRIP_FROM_PATH =
+
+# The STRIP_FROM_INC_PATH tag can be used to strip a user-defined part of
+# the path mentioned in the documentation of a class, which tells
+# the reader which header file to include in order to use a class.
+# If left blank only the name of the header file containing the class
+# definition is used. Otherwise one should specify the include paths that
+# are normally passed to the compiler using the -I flag.
+
+STRIP_FROM_INC_PATH =
+
+# If the SHORT_NAMES tag is set to YES, doxygen will generate much shorter
+# (but less readable) file names. This can be useful is your file systems
+# doesn't support long names like on DOS, Mac, or CD-ROM.
+
+SHORT_NAMES = NO
+
+# If the JAVADOC_AUTOBRIEF tag is set to YES then Doxygen
+# will interpret the first line (until the first dot) of a JavaDoc-style
+# comment as the brief description. If set to NO, the JavaDoc
+# comments will behave just like the Qt-style comments (thus requiring an
+# explicit @brief command for a brief description.
+
+JAVADOC_AUTOBRIEF = NO
+
+# The MULTILINE_CPP_IS_BRIEF tag can be set to YES to make Doxygen
+# treat a multi-line C++ special comment block (i.e. a block of //! or ///
+# comments) as a brief description. This used to be the default behaviour.
+# The new default is to treat a multi-line C++ comment block as a detailed
+# description. Set this tag to YES if you prefer the old behaviour instead.
+
+MULTILINE_CPP_IS_BRIEF = NO
+
+# If the DETAILS_AT_TOP tag is set to YES then Doxygen
+# will output the detailed description near the top, like JavaDoc.
+# If set to NO, the detailed description appears after the member
+# documentation.
+
+DETAILS_AT_TOP = NO
+
+# If the INHERIT_DOCS tag is set to YES (the default) then an undocumented
+# member inherits the documentation from any documented member that it
+# re-implements.
+
+INHERIT_DOCS = YES
+
+# If the SEPARATE_MEMBER_PAGES tag is set to YES, then doxygen will produce
+# a new page for each member. If set to NO, the documentation of a member will
+# be part of the file/class/namespace that contains it.
+
+SEPARATE_MEMBER_PAGES = NO
+
+# The TAB_SIZE tag can be used to set the number of spaces in a tab.
+# Doxygen uses this value to replace tabs by spaces in code fragments.
+
+TAB_SIZE = 8
+
+# This tag can be used to specify a number of aliases that acts
+# as commands in the documentation. An alias has the form "name=value".
+# For example adding "sideeffect=\par Side Effects:\n" will allow you to
+# put the command \sideeffect (or @sideeffect) in the documentation, which
+# will result in a user-defined paragraph with heading "Side Effects:".
+# You can put \n's in the value part of an alias to insert newlines.
+
+ALIASES =
+
+# Set the OPTIMIZE_OUTPUT_FOR_C tag to YES if your project consists of C
+# sources only. Doxygen will then generate output that is more tailored for C.
+# For instance, some of the names that are used will be different. The list
+# of all members will be omitted, etc.
+
+OPTIMIZE_OUTPUT_FOR_C = YES
+
+# If member grouping is used in the documentation and the DISTRIBUTE_GROUP_DOC
+# tag is set to YES, then doxygen will reuse the documentation of the first
+# member in the group (if any) for the other members of the group. By default
+# all members of a group must be documented explicitly.
+
+DISTRIBUTE_GROUP_DOC = NO
+
+# Set the SUBGROUPING tag to YES (the default) to allow class member groups of
+# the same type (for instance a group of public functions) to be put as a
+# subgroup of that type (e.g. under the Public Functions section). Set it to
+# NO to prevent subgrouping. Alternatively, this can be done per class using
+# the \nosubgrouping command.
+
+SUBGROUPING = YES
+
+#---------------------------------------------------------------------------
+# Build related configuration options
+#---------------------------------------------------------------------------
+
+# If the EXTRACT_ALL tag is set to YES doxygen will assume all entities in
+# documentation are documented, even if no documentation was available.
+# Private class members and static file members will be hidden unless
+# the EXTRACT_PRIVATE and EXTRACT_STATIC tags are set to YES
+
+EXTRACT_ALL = NO
+
+# If the EXTRACT_PRIVATE tag is set to YES all private members of a class
+# will be included in the documentation.
+
+EXTRACT_PRIVATE = YES
+
+# If the EXTRACT_STATIC tag is set to YES all static members of a file
+# will be included in the documentation.
+
+EXTRACT_STATIC = YES
+
+# If the EXTRACT_LOCAL_CLASSES tag is set to YES classes (and structs)
+# defined locally in source files will be included in the documentation.
+# If set to NO only classes defined in header files are included.
+
+EXTRACT_LOCAL_CLASSES = YES
+
+# If the HIDE_UNDOC_MEMBERS tag is set to YES, Doxygen will hide all
+# undocumented members of documented classes, files or namespaces.
+# If set to NO (the default) these members will be included in the
+# various overviews, but no documentation section is generated.
+# This option has no effect if EXTRACT_ALL is enabled.
+
+HIDE_UNDOC_MEMBERS = NO
+
+# If the HIDE_UNDOC_CLASSES tag is set to YES, Doxygen will hide all
+# undocumented classes that are normally visible in the class hierarchy.
+# If set to NO (the default) these classes will be included in the various
+# overviews. This option has no effect if EXTRACT_ALL is enabled.
+
+HIDE_UNDOC_CLASSES = NO
+
+# If the HIDE_FRIEND_COMPOUNDS tag is set to YES, Doxygen will hide all
+# friend (class|struct|union) declarations.
+# If set to NO (the default) these declarations will be included in the
+# documentation.
+
+HIDE_FRIEND_COMPOUNDS = NO
+
+# If the HIDE_IN_BODY_DOCS tag is set to YES, Doxygen will hide any
+# documentation blocks found inside the body of a function.
+# If set to NO (the default) these blocks will be appended to the
+# function's detailed documentation block.
+
+HIDE_IN_BODY_DOCS = NO
+
+# The INTERNAL_DOCS tag determines if documentation
+# that is typed after a \internal command is included. If the tag is set
+# to NO (the default) then the documentation will be excluded.
+# Set it to YES to include the internal documentation.
+
+INTERNAL_DOCS = NO
+
+# If the CASE_SENSE_NAMES tag is set to NO then Doxygen will only generate
+# file names in lower-case letters. If set to YES upper-case letters are also
+# allowed. This is useful if you have classes or files whose names only differ
+# in case and if your file system supports case sensitive file names. Windows
+# and Mac users are advised to set this option to NO.
+
+CASE_SENSE_NAMES = YES
+
+# If the HIDE_SCOPE_NAMES tag is set to NO (the default) then Doxygen
+# will show members with their full class and namespace scopes in the
+# documentation. If set to YES the scope will be hidden.
+
+HIDE_SCOPE_NAMES = NO
+
+# If the SHOW_INCLUDE_FILES tag is set to YES (the default) then Doxygen
+# will put a list of the files that are included by a file in the documentation
+# of that file.
+
+SHOW_INCLUDE_FILES = YES
+
+# If the INLINE_INFO tag is set to YES (the default) then a tag [inline]
+# is inserted in the documentation for inline members.
+
+INLINE_INFO = YES
+
+# If the SORT_MEMBER_DOCS tag is set to YES (the default) then doxygen
+# will sort the (detailed) documentation of file and class members
+# alphabetically by member name. If set to NO the members will appear in
+# declaration order.
+
+SORT_MEMBER_DOCS = NO
+
+# If the SORT_BRIEF_DOCS tag is set to YES then doxygen will sort the
+# brief documentation of file, namespace and class members alphabetically
+# by member name. If set to NO (the default) the members will appear in
+# declaration order.
+
+SORT_BRIEF_DOCS = NO
+
+# If the SORT_BY_SCOPE_NAME tag is set to YES, the class list will be
+# sorted by fully-qualified names, including namespaces. If set to
+# NO (the default), the class list will be sorted only by class name,
+# not including the namespace part.
+# Note: This option is not very useful if HIDE_SCOPE_NAMES is set to YES.
+# Note: This option applies only to the class list, not to the
+# alphabetical list.
+
+SORT_BY_SCOPE_NAME = NO
+
+# The GENERATE_TODOLIST tag can be used to enable (YES) or
+# disable (NO) the todo list. This list is created by putting \todo
+# commands in the documentation.
+
+GENERATE_TODOLIST = YES
+
+# The GENERATE_TESTLIST tag can be used to enable (YES) or
+# disable (NO) the test list. This list is created by putting \test
+# commands in the documentation.
+
+GENERATE_TESTLIST = YES
+
+# The GENERATE_BUGLIST tag can be used to enable (YES) or
+# disable (NO) the bug list. This list is created by putting \bug
+# commands in the documentation.
+
+GENERATE_BUGLIST = YES
+
+# The GENERATE_DEPRECATEDLIST tag can be used to enable (YES) or
+# disable (NO) the deprecated list. This list is created by putting
+# \deprecated commands in the documentation.
+
+GENERATE_DEPRECATEDLIST= YES
+
+# The ENABLED_SECTIONS tag can be used to enable conditional
+# documentation sections, marked by \if sectionname ... \endif.
+
+ENABLED_SECTIONS =
+
+# The MAX_INITIALIZER_LINES tag determines the maximum number of lines
+# the initial value of a variable or define consists of for it to appear in
+# the documentation. If the initializer consists of more lines than specified
+# here it will be hidden. Use a value of 0 to hide initializers completely.
+# The appearance of the initializer of individual variables and defines in the
+# documentation can be controlled using \showinitializer or \hideinitializer
+# command in the documentation regardless of this setting.
+
+MAX_INITIALIZER_LINES = 30
+
+# Set the SHOW_USED_FILES tag to NO to disable the list of files generated
+# at the bottom of the documentation of classes and structs. If set to YES the
+# list will mention the files that were used to generate the documentation.
+
+SHOW_USED_FILES = YES
+
+# If the sources in your project are distributed over multiple directories
+# then setting the SHOW_DIRECTORIES tag to YES will show the directory hierarchy
+# in the documentation. The default is NO.
+
+SHOW_DIRECTORIES = NO
+
+# The FILE_VERSION_FILTER tag can be used to specify a program or script that
+# doxygen should invoke to get the current version for each file (typically from the
+# version control system). Doxygen will invoke the program by executing (via
+# popen()) the command <command> <input-file>, where <command> is the value of
+# the FILE_VERSION_FILTER tag, and <input-file> is the name of an input file
+# provided by doxygen. Whatever the program writes to standard output
+# is used as the file version. See the manual for examples.
+
+FILE_VERSION_FILTER =
+
+#---------------------------------------------------------------------------
+# configuration options related to warning and progress messages
+#---------------------------------------------------------------------------
+
+# The QUIET tag can be used to turn on/off the messages that are generated
+# by doxygen. Possible values are YES and NO. If left blank NO is used.
+
+QUIET = NO
+
+# The WARNINGS tag can be used to turn on/off the warning messages that are
+# generated by doxygen. Possible values are YES and NO. If left blank
+# NO is used.
+
+WARNINGS = YES
+
+# If WARN_IF_UNDOCUMENTED is set to YES, then doxygen will generate warnings
+# for undocumented members. If EXTRACT_ALL is set to YES then this flag will
+# automatically be disabled.
+
+WARN_IF_UNDOCUMENTED = YES
+
+# If WARN_IF_DOC_ERROR is set to YES, doxygen will generate warnings for
+# potential errors in the documentation, such as not documenting some
+# parameters in a documented function, or documenting parameters that
+# don't exist or using markup commands wrongly.
+
+WARN_IF_DOC_ERROR = YES
+
+# This WARN_NO_PARAMDOC option can be abled to get warnings for
+# functions that are documented, but have no documentation for their parameters
+# or return value. If set to NO (the default) doxygen will only warn about
+# wrong or incomplete parameter documentation, but not about the absence of
+# documentation.
+
+WARN_NO_PARAMDOC = NO
+
+# The WARN_FORMAT tag determines the format of the warning messages that
+# doxygen can produce. The string should contain the $file, $line, and $text
+# tags, which will be replaced by the file and line number from which the
+# warning originated and the warning text. Optionally the format may contain
+# $version, which will be replaced by the version of the file (if it could
+# be obtained via FILE_VERSION_FILTER)
+
+WARN_FORMAT = "$file:$line: $text"
+
+# The WARN_LOGFILE tag can be used to specify a file to which warning
+# and error messages should be written. If left blank the output is written
+# to stderr.
+
+WARN_LOGFILE =
+
+#---------------------------------------------------------------------------
+# configuration options related to the input files
+#---------------------------------------------------------------------------
+
+# The INPUT tag can be used to specify the files and/or directories that contain
+# documented source files. You may enter file names like "myfile.cpp" or
+# directories like "/usr/src/myproject". Separate the files or directories
+# with spaces.
+
+INPUT = @top_srcdir@/src
+
+# If the value of the INPUT tag contains directories, you can use the
+# FILE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp
+# and *.h) to filter out the source-files in the directories. If left
+# blank the following patterns are tested:
+# *.c *.cc *.cxx *.cpp *.c++ *.java *.ii *.ixx *.ipp *.i++ *.inl *.h *.hh *.hxx
+# *.hpp *.h++ *.idl *.odl *.cs *.php *.php3 *.inc *.m *.mm *.py
+
+FILE_PATTERNS = *.h *.c
+
+# The RECURSIVE tag can be used to turn specify whether or not subdirectories
+# should be searched for input files as well. Possible values are YES and NO.
+# If left blank NO is used.
+
+RECURSIVE = YES
+
+# The EXCLUDE tag can be used to specify files and/or directories that should
+# excluded from the INPUT source files. This way you can easily exclude a
+# subdirectory from a directory tree whose root is specified with the INPUT tag.
+
+EXCLUDE =
+
+# The EXCLUDE_SYMLINKS tag can be used select whether or not files or
+# directories that are symbolic links (a Unix filesystem feature) are excluded
+# from the input.
+
+EXCLUDE_SYMLINKS = YES
+
+# If the value of the INPUT tag contains directories, you can use the
+# EXCLUDE_PATTERNS tag to specify one or more wildcard patterns to exclude
+# certain files from those directories. Note that the wildcards are matched
+# against the file with absolute path, so to exclude all test directories
+# for example use the pattern */test/*
+
+EXCLUDE_PATTERNS =
+
+# The EXAMPLE_PATH tag can be used to specify one or more files or
+# directories that contain example code fragments that are included (see
+# the \include command).
+
+EXAMPLE_PATH =
+
+# If the value of the EXAMPLE_PATH tag contains directories, you can use the
+# EXAMPLE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp
+# and *.h) to filter out the source-files in the directories. If left
+# blank all files are included.
+
+EXAMPLE_PATTERNS =
+
+# If the EXAMPLE_RECURSIVE tag is set to YES then subdirectories will be
+# searched for input files to be used with the \include or \dontinclude
+# commands irrespective of the value of the RECURSIVE tag.
+# Possible values are YES and NO. If left blank NO is used.
+
+EXAMPLE_RECURSIVE = NO
+
+# The IMAGE_PATH tag can be used to specify one or more files or
+# directories that contain image that are included in the documentation (see
+# the \image command).
+
+IMAGE_PATH =
+
+# The INPUT_FILTER tag can be used to specify a program that doxygen should
+# invoke to filter for each input file. Doxygen will invoke the filter program
+# by executing (via popen()) the command <filter> <input-file>, where <filter>
+# is the value of the INPUT_FILTER tag, and <input-file> is the name of an
+# input file. Doxygen will then use the output that the filter program writes
+# to standard output. If FILTER_PATTERNS is specified, this tag will be
+# ignored.
+
+INPUT_FILTER =
+
+# The FILTER_PATTERNS tag can be used to specify filters on a per file pattern
+# basis. Doxygen will compare the file name with each pattern and apply the
+# filter if there is a match. The filters are a list of the form:
+# pattern=filter (like *.cpp=my_cpp_filter). See INPUT_FILTER for further
+# info on how filters are used. If FILTER_PATTERNS is empty, INPUT_FILTER
+# is applied to all files.
+
+FILTER_PATTERNS =
+
+# If the FILTER_SOURCE_FILES tag is set to YES, the input filter (if set using
+# INPUT_FILTER) will be used to filter the input files when producing source
+# files to browse (i.e. when SOURCE_BROWSER is set to YES).
+
+FILTER_SOURCE_FILES = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to source browsing
+#---------------------------------------------------------------------------
+
+# If the SOURCE_BROWSER tag is set to YES then a list of source files will
+# be generated. Documented entities will be cross-referenced with these sources.
+# Note: To get rid of all source code in the generated output, make sure also
+# VERBATIM_HEADERS is set to NO.
+
+SOURCE_BROWSER = NO
+
+# Setting the INLINE_SOURCES tag to YES will include the body
+# of functions and classes directly in the documentation.
+
+INLINE_SOURCES = NO
+
+# Setting the STRIP_CODE_COMMENTS tag to YES (the default) will instruct
+# doxygen to hide any special comment blocks from generated source code
+# fragments. Normal C and C++ comments will always remain visible.
+
+STRIP_CODE_COMMENTS = YES
+
+# If the REFERENCED_BY_RELATION tag is set to YES (the default)
+# then for each documented function all documented
+# functions referencing it will be listed.
+
+REFERENCED_BY_RELATION = YES
+
+# If the REFERENCES_RELATION tag is set to YES (the default)
+# then for each documented function all documented entities
+# called/used by that function will be listed.
+
+REFERENCES_RELATION = YES
+
+# If the REFERENCES_LINK_SOURCE tag is set to YES (the default)
+# and SOURCE_BROWSER tag is set to YES, then the hyperlinks from
+# functions in REFERENCES_RELATION and REFERENCED_BY_RELATION lists will
+# link to the source code. Otherwise they will link to the documentstion.
+
+REFERENCES_LINK_SOURCE = YES
+
+# If the USE_HTAGS tag is set to YES then the references to source code
+# will point to the HTML generated by the htags(1) tool instead of doxygen
+# built-in source browser. The htags tool is part of GNU's global source
+# tagging system (see http://www.gnu.org/software/global/global.html). You
+# will need version 4.8.6 or higher.
+
+USE_HTAGS = NO
+
+# If the VERBATIM_HEADERS tag is set to YES (the default) then Doxygen
+# will generate a verbatim copy of the header file for each class for
+# which an include is specified. Set to NO to disable this.
+
+VERBATIM_HEADERS = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the alphabetical class index
+#---------------------------------------------------------------------------
+
+# If the ALPHABETICAL_INDEX tag is set to YES, an alphabetical index
+# of all compounds will be generated. Enable this if the project
+# contains a lot of classes, structs, unions or interfaces.
+
+ALPHABETICAL_INDEX = NO
+
+# If the alphabetical index is enabled (see ALPHABETICAL_INDEX) then
+# the COLS_IN_ALPHA_INDEX tag can be used to specify the number of columns
+# in which this list will be split (can be a number in the range [1..20])
+
+COLS_IN_ALPHA_INDEX = 5
+
+# In case all classes in a project start with a common prefix, all
+# classes will be put under the same header in the alphabetical index.
+# The IGNORE_PREFIX tag can be used to specify one or more prefixes that
+# should be ignored while generating the index headers.
+
+IGNORE_PREFIX =
+
+#---------------------------------------------------------------------------
+# configuration options related to the HTML output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_HTML tag is set to YES (the default) Doxygen will
+# generate HTML output.
+
+GENERATE_HTML = YES
+
+# The HTML_OUTPUT tag is used to specify where the HTML docs will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `html' will be used as the default path.
+
+HTML_OUTPUT = html
+
+# The HTML_FILE_EXTENSION tag can be used to specify the file extension for
+# each generated HTML page (for example: .htm,.php,.asp). If it is left blank
+# doxygen will generate files with .html extension.
+
+HTML_FILE_EXTENSION = .html
+
+# The HTML_HEADER tag can be used to specify a personal HTML header for
+# each generated HTML page. If it is left blank doxygen will generate a
+# standard header.
+
+HTML_HEADER =
+
+# The HTML_FOOTER tag can be used to specify a personal HTML footer for
+# each generated HTML page. If it is left blank doxygen will generate a
+# standard footer.
+
+HTML_FOOTER =
+
+# The HTML_STYLESHEET tag can be used to specify a user-defined cascading
+# style sheet that is used by each HTML page. It can be used to
+# fine-tune the look of the HTML output. If the tag is left blank doxygen
+# will generate a default style sheet. Note that doxygen will try to copy
+# the style sheet file to the HTML output directory, so don't put your own
+# stylesheet in the HTML output directory as well, or it will be erased!
+
+HTML_STYLESHEET =
+
+# If the HTML_ALIGN_MEMBERS tag is set to YES, the members of classes,
+# files or namespaces will be aligned in HTML using tables. If set to
+# NO a bullet list will be used.
+
+HTML_ALIGN_MEMBERS = YES
+
+# If the GENERATE_HTMLHELP tag is set to YES, additional index files
+# will be generated that can be used as input for tools like the
+# Microsoft HTML help workshop to generate a compressed HTML help file (.chm)
+# of the generated HTML documentation.
+
+GENERATE_HTMLHELP = NO
+
+# If the GENERATE_HTMLHELP tag is set to YES, the CHM_FILE tag can
+# be used to specify the file name of the resulting .chm file. You
+# can add a path in front of the file if the result should not be
+# written to the html output directory.
+
+CHM_FILE =
+
+# If the GENERATE_HTMLHELP tag is set to YES, the HHC_LOCATION tag can
+# be used to specify the location (absolute path including file name) of
+# the HTML help compiler (hhc.exe). If non-empty doxygen will try to run
+# the HTML help compiler on the generated index.hhp.
+
+HHC_LOCATION =
+
+# If the GENERATE_HTMLHELP tag is set to YES, the GENERATE_CHI flag
+# controls if a separate .chi index file is generated (YES) or that
+# it should be included in the master .chm file (NO).
+
+GENERATE_CHI = NO
+
+# If the GENERATE_HTMLHELP tag is set to YES, the BINARY_TOC flag
+# controls whether a binary table of contents is generated (YES) or a
+# normal table of contents (NO) in the .chm file.
+
+BINARY_TOC = NO
+
+# The TOC_EXPAND flag can be set to YES to add extra items for group members
+# to the contents of the HTML help documentation and to the tree view.
+
+TOC_EXPAND = NO
+
+# The DISABLE_INDEX tag can be used to turn on/off the condensed index at
+# top of each HTML page. The value NO (the default) enables the index and
+# the value YES disables it.
+
+DISABLE_INDEX = NO
+
+# This tag can be used to set the number of enum values (range [1..20])
+# that doxygen will group on one line in the generated HTML documentation.
+
+ENUM_VALUES_PER_LINE = 4
+
+# If the GENERATE_TREEVIEW tag is set to YES, a side panel will be
+# generated containing a tree-like index structure (just like the one that
+# is generated for HTML Help). For this to work a browser that supports
+# JavaScript, DHTML, CSS and frames is required (for instance Mozilla 1.0+,
+# Netscape 6.0+, Internet explorer 5.0+, or Konqueror). Windows users are
+# probably better off using the HTML help feature.
+
+GENERATE_TREEVIEW = NO
+
+# If the treeview is enabled (see GENERATE_TREEVIEW) then this tag can be
+# used to set the initial width (in pixels) of the frame in which the tree
+# is shown.
+
+TREEVIEW_WIDTH = 250
+
+#---------------------------------------------------------------------------
+# configuration options related to the LaTeX output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_LATEX tag is set to YES (the default) Doxygen will
+# generate Latex output.
+
+GENERATE_LATEX = YES
+
+# The LATEX_OUTPUT tag is used to specify where the LaTeX docs will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `latex' will be used as the default path.
+
+LATEX_OUTPUT = latex
+
+# The LATEX_CMD_NAME tag can be used to specify the LaTeX command name to be
+# invoked. If left blank `latex' will be used as the default command name.
+
+LATEX_CMD_NAME = latex
+
+# The MAKEINDEX_CMD_NAME tag can be used to specify the command name to
+# generate index for LaTeX. If left blank `makeindex' will be used as the
+# default command name.
+
+MAKEINDEX_CMD_NAME = makeindex
+
+# If the COMPACT_LATEX tag is set to YES Doxygen generates more compact
+# LaTeX documents. This may be useful for small projects and may help to
+# save some trees in general.
+
+COMPACT_LATEX = NO
+
+# The PAPER_TYPE tag can be used to set the paper type that is used
+# by the printer. Possible values are: a4, a4wide, letter, legal and
+# executive. If left blank a4wide will be used.
+
+PAPER_TYPE = a4wide
+
+# The EXTRA_PACKAGES tag can be to specify one or more names of LaTeX
+# packages that should be included in the LaTeX output.
+
+EXTRA_PACKAGES =
+
+# The LATEX_HEADER tag can be used to specify a personal LaTeX header for
+# the generated latex document. The header should contain everything until
+# the first chapter. If it is left blank doxygen will generate a
+# standard header. Notice: only use this tag if you know what you are doing!
+
+LATEX_HEADER =
+
+# If the PDF_HYPERLINKS tag is set to YES, the LaTeX that is generated
+# is prepared for conversion to pdf (using ps2pdf). The pdf file will
+# contain links (just like the HTML output) instead of page references
+# This makes the output suitable for online browsing using a pdf viewer.
+
+PDF_HYPERLINKS = NO
+
+# If the USE_PDFLATEX tag is set to YES, pdflatex will be used instead of
+# plain latex in the generated Makefile. Set this option to YES to get a
+# higher quality PDF documentation.
+
+USE_PDFLATEX = YES
+
+# If the LATEX_BATCHMODE tag is set to YES, doxygen will add the \\batchmode.
+# command to the generated LaTeX files. This will instruct LaTeX to keep
+# running if errors occur, instead of asking the user for help.
+# This option is also used when generating formulas in HTML.
+
+LATEX_BATCHMODE = NO
+
+# If LATEX_HIDE_INDICES is set to YES then doxygen will not
+# include the index chapters (such as File Index, Compound Index, etc.)
+# in the output.
+
+LATEX_HIDE_INDICES = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the RTF output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_RTF tag is set to YES Doxygen will generate RTF output
+# The RTF output is optimized for Word 97 and may not look very pretty with
+# other RTF readers or editors.
+
+GENERATE_RTF = NO
+
+# The RTF_OUTPUT tag is used to specify where the RTF docs will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `rtf' will be used as the default path.
+
+RTF_OUTPUT = rtf
+
+# If the COMPACT_RTF tag is set to YES Doxygen generates more compact
+# RTF documents. This may be useful for small projects and may help to
+# save some trees in general.
+
+COMPACT_RTF = NO
+
+# If the RTF_HYPERLINKS tag is set to YES, the RTF that is generated
+# will contain hyperlink fields. The RTF file will
+# contain links (just like the HTML output) instead of page references.
+# This makes the output suitable for online browsing using WORD or other
+# programs which support those fields.
+# Note: wordpad (write) and others do not support links.
+
+RTF_HYPERLINKS = NO
+
+# Load stylesheet definitions from file. Syntax is similar to doxygen's
+# config file, i.e. a series of assignments. You only have to provide
+# replacements, missing definitions are set to their default value.
+
+RTF_STYLESHEET_FILE =
+
+# Set optional variables used in the generation of an rtf document.
+# Syntax is similar to doxygen's config file.
+
+RTF_EXTENSIONS_FILE =
+
+#---------------------------------------------------------------------------
+# configuration options related to the man page output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_MAN tag is set to YES (the default) Doxygen will
+# generate man pages
+
+GENERATE_MAN = NO
+
+# The MAN_OUTPUT tag is used to specify where the man pages will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `man' will be used as the default path.
+
+MAN_OUTPUT = man
+
+# The MAN_EXTENSION tag determines the extension that is added to
+# the generated man pages (default is the subroutine's section .3)
+
+MAN_EXTENSION = .3
+
+# If the MAN_LINKS tag is set to YES and Doxygen generates man output,
+# then it will generate one additional man file for each entity
+# documented in the real man page(s). These additional files
+# only source the real man page, but without them the man command
+# would be unable to find the correct page. The default is NO.
+
+MAN_LINKS = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the XML output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_XML tag is set to YES Doxygen will
+# generate an XML file that captures the structure of
+# the code including all documentation.
+
+GENERATE_XML = NO
+
+# The XML_OUTPUT tag is used to specify where the XML pages will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `xml' will be used as the default path.
+
+XML_OUTPUT = xml
+
+# The XML_SCHEMA tag can be used to specify an XML schema,
+# which can be used by a validating XML parser to check the
+# syntax of the XML files.
+
+XML_SCHEMA =
+
+# The XML_DTD tag can be used to specify an XML DTD,
+# which can be used by a validating XML parser to check the
+# syntax of the XML files.
+
+XML_DTD =
+
+# If the XML_PROGRAMLISTING tag is set to YES Doxygen will
+# dump the program listings (including syntax highlighting
+# and cross-referencing information) to the XML output. Note that
+# enabling this will significantly increase the size of the XML output.
+
+XML_PROGRAMLISTING = YES
+
+#---------------------------------------------------------------------------
+# configuration options for the AutoGen Definitions output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_AUTOGEN_DEF tag is set to YES Doxygen will
+# generate an AutoGen Definitions (see autogen.sf.net) file
+# that captures the structure of the code including all
+# documentation. Note that this feature is still experimental
+# and incomplete at the moment.
+
+GENERATE_AUTOGEN_DEF = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the Perl module output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_PERLMOD tag is set to YES Doxygen will
+# generate a Perl module file that captures the structure of
+# the code including all documentation. Note that this
+# feature is still experimental and incomplete at the
+# moment.
+
+GENERATE_PERLMOD = NO
+
+# If the PERLMOD_LATEX tag is set to YES Doxygen will generate
+# the necessary Makefile rules, Perl scripts and LaTeX code to be able
+# to generate PDF and DVI output from the Perl module output.
+
+PERLMOD_LATEX = NO
+
+# If the PERLMOD_PRETTY tag is set to YES the Perl module output will be
+# nicely formatted so it can be parsed by a human reader. This is useful
+# if you want to understand what is going on. On the other hand, if this
+# tag is set to NO the size of the Perl module output will be much smaller
+# and Perl will parse it just the same.
+
+PERLMOD_PRETTY = YES
+
+# The names of the make variables in the generated doxyrules.make file
+# are prefixed with the string contained in PERLMOD_MAKEVAR_PREFIX.
+# This is useful so different doxyrules.make files included by the same
+# Makefile don't overwrite each other's variables.
+
+PERLMOD_MAKEVAR_PREFIX =
+
+#---------------------------------------------------------------------------
+# Configuration options related to the preprocessor
+#---------------------------------------------------------------------------
+
+# If the ENABLE_PREPROCESSING tag is set to YES (the default) Doxygen will
+# evaluate all C-preprocessor directives found in the sources and include
+# files.
+
+ENABLE_PREPROCESSING = YES
+
+# If the MACRO_EXPANSION tag is set to YES Doxygen will expand all macro
+# names in the source code. If set to NO (the default) only conditional
+# compilation will be performed. Macro expansion can be done in a controlled
+# way by setting EXPAND_ONLY_PREDEF to YES.
+
+MACRO_EXPANSION = YES
+
+# If the EXPAND_ONLY_PREDEF and MACRO_EXPANSION tags are both set to YES
+# then the macro expansion is limited to the macros specified with the
+# PREDEFINED and EXPAND_AS_DEFINED tags.
+
+EXPAND_ONLY_PREDEF = YES
+
+# If the SEARCH_INCLUDES tag is set to YES (the default) the includes files
+# in the INCLUDE_PATH (see below) will be search if a #include is found.
+
+SEARCH_INCLUDES = YES
+
+# The INCLUDE_PATH tag can be used to specify one or more directories that
+# contain include files that are not input files but should be processed by
+# the preprocessor.
+
+INCLUDE_PATH =
+
+# You can use the INCLUDE_FILE_PATTERNS tag to specify one or more wildcard
+# patterns (like *.h and *.hpp) to filter out the header-files in the
+# directories. If left blank, the patterns specified with FILE_PATTERNS will
+# be used.
+
+INCLUDE_FILE_PATTERNS =
+
+# The PREDEFINED tag can be used to specify one or more macro names that
+# are defined before the preprocessor is started (similar to the -D option of
+# gcc). The argument of the tag is a list of macros of the form: name
+# or name=definition (no spaces). If the definition and the = are
+# omitted =1 is assumed. To prevent a macro definition from being
+# undefined via #undef or recursively expanded use the := operator
+# instead of the = operator.
+
+PREDEFINED = LZMA_API(type)=type \
+ LZMA_API_IMPORT \
+ LZMA_API_CALL=
+
+# If the MACRO_EXPANSION and EXPAND_ONLY_PREDEF tags are set to YES then
+# this tag can be used to specify a list of macro names that should be expanded.
+# The macro definition that is found in the sources will be used.
+# Use the PREDEFINED tag if you want to use a different macro definition.
+
+EXPAND_AS_DEFINED =
+
+# If the SKIP_FUNCTION_MACROS tag is set to YES (the default) then
+# doxygen's preprocessor will remove all function-like macros that are alone
+# on a line, have an all uppercase name, and do not end with a semicolon. Such
+# function macros are typically used for boiler-plate code, and will confuse
+# the parser if not removed.
+
+SKIP_FUNCTION_MACROS = YES
+
+#---------------------------------------------------------------------------
+# Configuration::additions related to external references
+#---------------------------------------------------------------------------
+
+# The TAGFILES option can be used to specify one or more tagfiles.
+# Optionally an initial location of the external documentation
+# can be added for each tagfile. The format of a tag file without
+# this location is as follows:
+# TAGFILES = file1 file2 ...
+# Adding location for the tag files is done as follows:
+# TAGFILES = file1=loc1 "file2 = loc2" ...
+# where "loc1" and "loc2" can be relative or absolute paths or
+# URLs. If a location is present for each tag, the installdox tool
+# does not have to be run to correct the links.
+# Note that each tag file must have a unique name
+# (where the name does NOT include the path)
+# If a tag file is not located in the directory in which doxygen
+# is run, you must also specify the path to the tagfile here.
+
+TAGFILES =
+
+# When a file name is specified after GENERATE_TAGFILE, doxygen will create
+# a tag file that is based on the input files it reads.
+
+GENERATE_TAGFILE =
+
+# If the ALLEXTERNALS tag is set to YES all external classes will be listed
+# in the class index. If set to NO only the inherited external classes
+# will be listed.
+
+ALLEXTERNALS = NO
+
+# If the EXTERNAL_GROUPS tag is set to YES all external groups will be listed
+# in the modules index. If set to NO, only the current project's groups will
+# be listed.
+
+EXTERNAL_GROUPS = YES
+
+# The PERL_PATH should be the absolute path and name of the perl script
+# interpreter (i.e. the result of `which perl').
+
+PERL_PATH = /usr/bin/perl
+
+#---------------------------------------------------------------------------
+# Configuration options related to the dot tool
+#---------------------------------------------------------------------------
+
+# If the CLASS_DIAGRAMS tag is set to YES (the default) Doxygen will
+# generate a inheritance diagram (in HTML, RTF and LaTeX) for classes with base
+# or super classes. Setting the tag to NO turns the diagrams off. Note that
+# this option is superseded by the HAVE_DOT option below. This is only a
+# fallback. It is recommended to install and use dot, since it yields more
+# powerful graphs.
+
+CLASS_DIAGRAMS = YES
+
+# If set to YES, the inheritance and collaboration graphs will hide
+# inheritance and usage relations if the target is undocumented
+# or is not a class.
+
+HIDE_UNDOC_RELATIONS = YES
+
+# If you set the HAVE_DOT tag to YES then doxygen will assume the dot tool is
+# available from the path. This tool is part of Graphviz, a graph visualization
+# toolkit from AT&T and Lucent Bell Labs. The other options in this section
+# have no effect if this option is set to NO (the default)
+
+HAVE_DOT = NO
+
+# If the CLASS_GRAPH and HAVE_DOT tags are set to YES then doxygen
+# will generate a graph for each documented class showing the direct and
+# indirect inheritance relations. Setting this tag to YES will force the
+# the CLASS_DIAGRAMS tag to NO.
+
+CLASS_GRAPH = YES
+
+# If the COLLABORATION_GRAPH and HAVE_DOT tags are set to YES then doxygen
+# will generate a graph for each documented class showing the direct and
+# indirect implementation dependencies (inheritance, containment, and
+# class references variables) of the class with other documented classes.
+
+COLLABORATION_GRAPH = YES
+
+# If the GROUP_GRAPHS and HAVE_DOT tags are set to YES then doxygen
+# will generate a graph for groups, showing the direct groups dependencies
+
+GROUP_GRAPHS = YES
+
+# If the UML_LOOK tag is set to YES doxygen will generate inheritance and
+# collaboration diagrams in a style similar to the OMG's Unified Modeling
+# Language.
+
+UML_LOOK = NO
+
+# If set to YES, the inheritance and collaboration graphs will show the
+# relations between templates and their instances.
+
+TEMPLATE_RELATIONS = NO
+
+# If the ENABLE_PREPROCESSING, SEARCH_INCLUDES, INCLUDE_GRAPH, and HAVE_DOT
+# tags are set to YES then doxygen will generate a graph for each documented
+# file showing the direct and indirect include dependencies of the file with
+# other documented files.
+
+INCLUDE_GRAPH = YES
+
+# If the ENABLE_PREPROCESSING, SEARCH_INCLUDES, INCLUDED_BY_GRAPH, and
+# HAVE_DOT tags are set to YES then doxygen will generate a graph for each
+# documented header file showing the documented files that directly or
+# indirectly include this file.
+
+INCLUDED_BY_GRAPH = YES
+
+# If the CALL_GRAPH and HAVE_DOT tags are set to YES then doxygen will
+# generate a call dependency graph for every global function or class method.
+# Note that enabling this option will significantly increase the time of a run.
+# So in most cases it will be better to enable call graphs for selected
+# functions only using the \callgraph command.
+
+CALL_GRAPH = NO
+
+# If the CALLER_GRAPH and HAVE_DOT tags are set to YES then doxygen will
+# generate a caller dependency graph for every global function or class method.
+# Note that enabling this option will significantly increase the time of a run.
+# So in most cases it will be better to enable caller graphs for selected
+# functions only using the \callergraph command.
+
+CALLER_GRAPH = NO
+
+# If the GRAPHICAL_HIERARCHY and HAVE_DOT tags are set to YES then doxygen
+# will graphical hierarchy of all classes instead of a textual one.
+
+GRAPHICAL_HIERARCHY = YES
+
+# If the DIRECTORY_GRAPH, SHOW_DIRECTORIES and HAVE_DOT tags are set to YES
+# then doxygen will show the dependencies a directory has on other directories
+# in a graphical way. The dependency relations are determined by the #include
+# relations between the files in the directories.
+
+DIRECTORY_GRAPH = YES
+
+# The DOT_IMAGE_FORMAT tag can be used to set the image format of the images
+# generated by dot. Possible values are png, jpg, or gif
+# If left blank png will be used.
+
+DOT_IMAGE_FORMAT = png
+
+# The tag DOT_PATH can be used to specify the path where the dot tool can be
+# found. If left blank, it is assumed the dot tool can be found in the path.
+
+DOT_PATH =
+
+# The DOTFILE_DIRS tag can be used to specify one or more directories that
+# contain dot files that are included in the documentation (see the
+# \dotfile command).
+
+DOTFILE_DIRS =
+
+# The MAX_DOT_GRAPH_WIDTH tag can be used to set the maximum allowed width
+# (in pixels) of the graphs generated by dot. If a graph becomes larger than
+# this value, doxygen will try to truncate the graph, so that it fits within
+# the specified constraint. Beware that most browsers cannot cope with very
+# large images.
+
+MAX_DOT_GRAPH_WIDTH = 1024
+
+# The MAX_DOT_GRAPH_HEIGHT tag can be used to set the maximum allows height
+# (in pixels) of the graphs generated by dot. If a graph becomes larger than
+# this value, doxygen will try to truncate the graph, so that it fits within
+# the specified constraint. Beware that most browsers cannot cope with very
+# large images.
+
+MAX_DOT_GRAPH_HEIGHT = 1024
+
+# The MAX_DOT_GRAPH_DEPTH tag can be used to set the maximum depth of the
+# graphs generated by dot. A depth value of 3 means that only nodes reachable
+# from the root by following a path via at most 3 edges will be shown. Nodes
+# that lay further from the root node will be omitted. Note that setting this
+# option to 1 or 2 may greatly reduce the computation time needed for large
+# code bases. Also note that a graph may be further truncated if the graph's
+# image dimensions are not sufficient to fit the graph (see MAX_DOT_GRAPH_WIDTH
+# and MAX_DOT_GRAPH_HEIGHT). If 0 is used for the depth value (the default),
+# the graph is not depth-constrained.
+
+MAX_DOT_GRAPH_DEPTH = 0
+
+# Set the DOT_TRANSPARENT tag to YES to generate images with a transparent
+# background. This is disabled by default, which results in a white background.
+# Warning: Depending on the platform used, enabling this option may lead to
+# badly anti-aliased labels on the edges of a graph (i.e. they become hard to
+# read).
+
+DOT_TRANSPARENT = NO
+
+# Set the DOT_MULTI_TARGETS tag to YES allow dot to generate multiple output
+# files in one run (i.e. multiple -o and -T options on the command line). This
+# makes dot run faster, but since only newer versions of dot (>1.8.10)
+# support this, this feature is disabled by default.
+
+DOT_MULTI_TARGETS = NO
+
+# If the GENERATE_LEGEND tag is set to YES (the default) Doxygen will
+# generate a legend page explaining the meaning of the various boxes and
+# arrows in the dot generated graphs.
+
+GENERATE_LEGEND = YES
+
+# If the DOT_CLEANUP tag is set to YES (the default) Doxygen will
+# remove the intermediate dot files that are used to generate
+# the various graphs.
+
+DOT_CLEANUP = YES
+
+#---------------------------------------------------------------------------
+# Configuration::additions related to the search engine
+#---------------------------------------------------------------------------
+
+# The SEARCHENGINE tag specifies whether or not a search engine should be
+# used. If set to NO the values of all tags below this one will be ignored.
+
+SEARCHENGINE = NO
diff --git a/INSTALL b/INSTALL
new file mode 100644
index 0000000..5f38085
--- /dev/null
+++ b/INSTALL
@@ -0,0 +1,369 @@
+
+XZ Utils Installation
+=====================
+
+ 0. Preface
+ 1. Supported platforms
+ 1.1. Compilers
+ 1.2. Platform-specific notes
+ 1.2.1. IRIX
+ 1.2.2. Tru64
+ 1.2.3. Windows
+ 1.2.4. DOS
+ 1.2.5. OS/2
+ 1.2.6. OpenVMS
+ 1.3. Adding support for new platforms
+ 2. configure options
+ 3. xzgrep and other scripts
+ 3.1. Dependencies
+ 3.2. PATH
+ 4. Troubleshooting
+ 4.1. "No C99 compiler was found."
+ 4.1. "No POSIX conforming shell (sh) was found."
+ 4.2. configure works but build fails at crc32_x86.S
+
+
+0. Preface
+----------
+
+ If you aren't familiar with building packages that use GNU Autotools,
+ see the file INSTALL.generic for generic instructions before reading
+ further.
+
+ If you are going to build a package for distribution, see also the
+ file PACKAGERS. It contains information that should help making the
+ binary packages as good as possible, but the information isn't very
+ interesting to those making local builds for private use or for use
+ in special situations like embedded systems.
+
+
+1. Supported platforms
+----------------------
+
+ XZ Utils are developed on GNU/Linux, but they should work on many
+ POSIX-like operating systems like *BSDs and Solaris, and even on
+ a few non-POSIX operating systems.
+
+
+1.1. Compilers
+
+ A C99 compiler is required to compile XZ Utils. If you use GCC, you
+ need at least version 3.x.x. GCC version 2.xx.x doesn't support some
+ C99 features used in XZ Utils source code, thus GCC 2 won't compile
+ XZ Utils.
+
+ XZ Utils takes advantage of some GNU C extensions when building
+ with GCC. Because these extensions are used only when building
+ with GCC, it should be possible to use any C99 compiler.
+
+
+1.2. Platform-specific notes
+
+1.2.1. IRIX
+
+ MIPSpro 7.4.4m has been reported to produce broken code if using
+ the -O2 optimization flag ("make check" fails). Using -O1 should
+ work.
+
+
+1.2.2. Tru64
+
+ If you try to use the native C compiler on Tru64 (passing CC=cc to
+ configure), it is possible that the configure script will complain
+ that no C99 compiler was found even when the native compiler supports
+ C99. You can safely override the test for C99 compiler by passing
+ ac_cv_prog_cc_c99= as the argument to the configure script.
+
+
+1.2.3. Windows
+
+ Building XZ Utils on Windows is supported under MinGW + MSYS and
+ Cygwin. There is windows/build.sh to ease packaging XZ Utils with
+ MinGW + MSYS into a redistributable .zip or .7z file. See
+ windows/INSTALL-Windows.txt for more information.
+
+ It might be possible to build liblzma with a non-GNU toolchain too,
+ but that will probably require writing a separate makefile. Building
+ the command line tools with non-GNU toolchains will be harder than
+ building only liblzma.
+
+ Even if liblzma is built with MinGW, the resulting DLL or static
+ library can be used by other compilers and linkers, including MSVC.
+ Thus, it shouldn't be a problem to use MinGW to build liblzma even
+ if you cannot use MinGW to build the rest of your project. See
+ windows/README-Windows.txt for details.
+
+
+1.2.4. DOS
+
+ There is an experimental Makefile in the "dos" directory to build
+ XZ Utils on DOS using DJGPP. Support for long file names (LFN) is
+ needed. See dos/README for more information.
+
+ GNU Autotools based build hasn't been tried on DOS. If you try, I
+ would like to hear if it worked.
+
+
+1.2.5. OS/2
+
+ To omit large number of harmless warnings about visibility support,
+ pass gl_cv_cc_visibility=no as an argument to the configure script.
+ This isn't mandatory since it should have no effect on the resulting
+ binaries.
+
+
+1.2.6. OpenVMS
+
+ XZ Utils can be built for OpenVMS, but the build system files are
+ currently not included in the XZ Utils source package. The required
+ OpenVMS-specific files are maintained by Jouk Jansen and can be
+ downloaded here:
+
+ http://nchrem.tnw.tudelft.nl/openvms/software2.html#xzutils
+
+
+1.3. Adding support for new platforms
+
+ If you have written patches to make XZ Utils to work on previously
+ unsupported platform, please send the patches to me! I will consider
+ including them to the official version. It's nice to minimize the
+ need of third-party patching.
+
+ One exception: Don't request or send patches to change the whole
+ source package to C89. I find C99 substantially nicer to write and
+ maintain. However, the public library headers must be in C89 to
+ avoid frustrating those who maintain programs, which are strictly
+ in C89 or C++.
+
+
+2. configure options
+--------------------
+
+ In most cases, the defaults are what you want. Most of the options
+ below are useful only when building a size-optimized version of
+ liblzma or command line tools.
+
+ --enable-encoders=LIST
+ --disable-encoders
+ Specify a comma-separated LIST of filter encoders to
+ build. See "./configure --help" for exact list of
+ available filter encoders. The default is to build all
+ supported encoders.
+
+ If LIST is empty or --disable-encoders is used, no filter
+ encoders will be built and also the code shared between
+ encoders will be omitted.
+
+ Disabling encoders will remove some symbols from the
+ liblzma ABI, so this option should be used only when it
+ is known to not cause problems.
+
+ --enable-decoders=LIST
+ --disable-decoders
+ This is like --enable-encoders but for decoders. The
+ default is to build all supported decoders.
+
+ --enable-match-finders=LIST
+ liblzma includes two categories of match finders:
+ hash chains and binary trees. Hash chains (hc3 and hc4)
+ are quite fast but they don't provide the best compression
+ ratio. Binary trees (bt2, bt3 and bt4) give excellent
+ compression ratio, but they are slower and need more
+ memory than hash chains.
+
+ You need to enable at least one match finder to build the
+ LZMA1 or LZMA2 filter encoders. Usually hash chains are
+ used only in the fast mode, while binary trees are used to
+ when the best compression ratio is wanted.
+
+ The default is to build all the match finders if LZMA1
+ or LZMA2 filter encoders are being built.
+
+ --enable-checks=LIST
+ liblzma support multiple integrity checks. CRC32 is
+ mandatory, and cannot be omitted. See "./configure --help"
+ for exact list of available integrity check types.
+
+ liblzma and the command line tools can decompress files
+ which use unsupported integrity check type, but naturally
+ the file integrity cannot be verified in that case.
+
+ Disabling integrity checks may remove some symbols from
+ the liblzma ABI, so this option should be used only when
+ it is known to not cause problems.
+
+ --disable-assembler
+ liblzma includes some assembler optimizations. Currently
+ there is only assembler code for CRC32 and CRC64 for
+ 32-bit x86.
+
+ All the assembler code in liblzma is position-independent
+ code, which is suitable for use in shared libraries and
+ position-independent executables. So far only i386
+ instructions are used, but the code is optimized for i686
+ class CPUs. If you are compiling liblzma exclusively for
+ pre-i686 systems, you may want to disable the assembler
+ code.
+
+ --enable-unaligned-access
+ Allow liblzma to use unaligned memory access for 16-bit
+ and 32-bit loads and stores. This should be enabled only
+ when the hardware supports this, i.e. when unaligned
+ access is fast. Some operating system kernels emulate
+ unaligned access, which is extremely slow. This option
+ shouldn't be used on systems that rely on such emulation.
+
+ Unaligned access is enabled by default on x86, x86-64,
+ and big endian PowerPC.
+
+ --enable-small
+ Reduce the size of liblzma by selecting smaller but
+ semantically equivalent version of some functions, and
+ omit precomputed lookup tables. This option tends to
+ make liblzma slightly slower.
+
+ Note that while omitting the precomputed tables makes
+ liblzma smaller on disk, the tables are still needed at
+ run time, and need to be computed at startup. This also
+ means that the RAM holding the tables won't be shared
+ between applications linked against shared liblzma.
+
+ This option doesn't modify CFLAGS to tell the compiler
+ to optimize for size. You need to add -Os or equivalent
+ flag(s) to CFLAGS manually.
+
+ --enable-assume-ram=SIZE
+ On the most common operating systems, XZ Utils is able to
+ detect the amount of physical memory on the system. This
+ information is used to set the default memory usage limit.
+
+ On some systems, there is no code to detect the amount of
+ RAM though. Using --enable-assume-ram one can set how much
+ memory to assume on these systems. SIZE is given as MiB.
+ The default is 128 MiB, which allows decompressing files
+ created with "xz -9".
+
+ Feel free to send patches to add support for detecting
+ the amount of RAM on the operating system you use. See
+ src/common/tuklib_physmem.c for details.
+
+ --disable-threads
+ Disable threading support. This makes some things
+ thread-unsafe, meaning that if multithreaded application
+ calls liblzma functions from more than one thread,
+ something bad may happen.
+
+ Use this option if threading support causes you trouble,
+ or if you know that you will use liblzma only from
+ single-threaded applications and want to avoid dependency
+ on libpthread.
+
+ --enable-dynamic=TYPE
+ Specify how command line tools should be linked against
+ liblzma. Possible TYPES:
+
+ yes All command line tools are linked against
+ shared liblzma (if shared liblzma was built).
+ This is equivalent to --enable-dynamic (i.e.
+ no =TYPE).
+
+ mixed Some tools are linked against static liblzma
+ and some against shared liblzma. This is the
+ default and recommended way.
+
+ no All command line tools are linked against
+ static liblzma (if static liblzma was built).
+ This is equivalent to --disable-dynamic.
+
+ This option is mostly useful for packagers, if distro
+ policy requires linking against shared libaries. See the
+ file PACKAGERS for more information about pros and cons
+ of this option.
+
+ --enable-debug
+ This enables the assert() macro and possibly some other
+ run-time consistency checks. It makes the code slower, so
+ you normally don't want to have this enabled.
+
+ --enable-werror
+ If building with GCC, make all compiler warnings an error,
+ that abort the compilation. This may help catching bugs,
+ and should work on most systems. This has no effect on the
+ resulting binaries.
+
+
+3. xzgrep and other scripts
+---------------------------
+
+3.1. Dependencies
+
+ POSIX shell (sh) and bunch of other standard POSIX tools are required
+ to run the scripts. The configure script tries to find a POSIX
+ compliant sh, but if it fails, you can force the shell by passing
+ gl_cv_posix_shell=/path/to/posix-sh as an argument to the configure
+ script.
+
+ Some of the scripts require also mktemp. The original mktemp can be
+ found from <http://www.mktemp.org/>. On GNU, most will use the mktemp
+ program from GNU coreutils instead of the original implementation.
+ Both mktemp versions are fine for XZ Utils (and practically for
+ everything else too).
+
+
+3.2. PATH
+
+ The scripts assume that the required tools (standard POSIX utilities,
+ mktemp, and xz) are in PATH; the scripts don't set the PATH themselves.
+ Some people like this while some think this is a bug. Those in the
+ latter group can easily patch the scripts before running the configure
+ script by taking advantage of a placeholder line in the scripts.
+
+ For example, to make the scripts prefix /usr/bin:/bin to PATH:
+
+ perl -pi -e 's|^#SET_PATH.*$|PATH=/usr/bin:/bin:\$PATH|' \
+ src/scripts/xz*.in
+
+
+4. Troubleshooting
+------------------
+
+4.1. "No C99 compiler was found."
+
+ You need a C99 compiler to build XZ Utils. If the configure script
+ cannot find a C99 compiler and you think you have such a compiler
+ installed, set the compiler command by passing CC=/path/to/c99 as
+ an argument to the configure script.
+
+ If you get this error even when you think your compiler supports C99,
+ you can override the test by passing ac_cv_prog_cc_c99= as an argument
+ to the configure script. The test for C99 compiler is not perfect (and
+ it is not as easy to make it perfect as it sounds), so sometimes this
+ may be needed. You will get a compile error if your compiler doesn't
+ support enough C99.
+
+
+4.1. "No POSIX conforming shell (sh) was found."
+
+ xzgrep and other scripts need a shell that (roughly) conforms
+ to POSIX. The configure script tries to find such a shell. If
+ it fails, you can force the shell to be used by passing
+ gl_cv_posix_shell=/path/to/posix-sh as an argument to the configure
+ script.
+
+
+4.2. configure works but build fails at crc32_x86.S
+
+ The easy fix is to pass --disable-assembler to the configure script.
+
+ The configure script determines if assembler code can be used by
+ looking at the configure triplet; there is currently no check if
+ the assembler code can actually actually be built. The x86 assembler
+ code should work on x86 GNU/Linux, *BSDs, Solaris, Darwin, MinGW,
+ Cygwin, and DJGPP. On other x86 systems, there may be problems and
+ the assembler code may need to be disabled with the configure option.
+
+ If you get this error when building for x86-64, you have specified or
+ the configure script has misguessed your architecture. Pass the
+ correct configure triplet using the --build=CPU-COMPANY-SYSTEM option
+ (see INSTALL.generic).
+
diff --git a/INSTALL.generic b/INSTALL.generic
new file mode 100644
index 0000000..2550dab
--- /dev/null
+++ b/INSTALL.generic
@@ -0,0 +1,302 @@
+Installation Instructions
+*************************
+
+Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004, 2005,
+2006, 2007, 2008, 2009 Free Software Foundation, Inc.
+
+ This file is free documentation; the Free Software Foundation gives
+unlimited permission to copy, distribute and modify it.
+
+Basic Installation
+==================
+
+ Briefly, the shell commands `./configure; make; make install' should
+configure, build, and install this package. The following
+more-detailed instructions are generic; see the `README' file for
+instructions specific to this package.
+
+ The `configure' shell script attempts to guess correct values for
+various system-dependent variables used during compilation. It uses
+those values to create a `Makefile' in each directory of the package.
+It may also create one or more `.h' files containing system-dependent
+definitions. Finally, it creates a shell script `config.status' that
+you can run in the future to recreate the current configuration, and a
+file `config.log' containing compiler output (useful mainly for
+debugging `configure').
+
+ It can also use an optional file (typically called `config.cache'
+and enabled with `--cache-file=config.cache' or simply `-C') that saves
+the results of its tests to speed up reconfiguring. Caching is
+disabled by default to prevent problems with accidental use of stale
+cache files.
+
+ If you need to do unusual things to compile the package, please try
+to figure out how `configure' could check whether to do them, and mail
+diffs or instructions to the address given in the `README' so they can
+be considered for the next release. If you are using the cache, and at
+some point `config.cache' contains results you don't want to keep, you
+may remove or edit it.
+
+ The file `configure.ac' (or `configure.in') is used to create
+`configure' by a program called `autoconf'. You need `configure.ac' if
+you want to change it or regenerate `configure' using a newer version
+of `autoconf'.
+
+The simplest way to compile this package is:
+
+ 1. `cd' to the directory containing the package's source code and type
+ `./configure' to configure the package for your system.
+
+ Running `configure' might take a while. While running, it prints
+ some messages telling which features it is checking for.
+
+ 2. Type `make' to compile the package.
+
+ 3. Optionally, type `make check' to run any self-tests that come with
+ the package.
+
+ 4. Type `make install' to install the programs and any data files and
+ documentation.
+
+ 5. You can remove the program binaries and object files from the
+ source code directory by typing `make clean'. To also remove the
+ files that `configure' created (so you can compile the package for
+ a different kind of computer), type `make distclean'. There is
+ also a `make maintainer-clean' target, but that is intended mainly
+ for the package's developers. If you use it, you may have to get
+ all sorts of other programs in order to regenerate files that came
+ with the distribution.
+
+ 6. Often, you can also type `make uninstall' to remove the installed
+ files again.
+
+Compilers and Options
+=====================
+
+ Some systems require unusual options for compilation or linking that
+the `configure' script does not know about. Run `./configure --help'
+for details on some of the pertinent environment variables.
+
+ You can give `configure' initial values for configuration parameters
+by setting variables in the command line or in the environment. Here
+is an example:
+
+ ./configure CC=c99 CFLAGS=-g LIBS=-lposix
+
+ *Note Defining Variables::, for more details.
+
+Compiling For Multiple Architectures
+====================================
+
+ You can compile the package for more than one kind of computer at the
+same time, by placing the object files for each architecture in their
+own directory. To do this, you can use GNU `make'. `cd' to the
+directory where you want the object files and executables to go and run
+the `configure' script. `configure' automatically checks for the
+source code in the directory that `configure' is in and in `..'.
+
+ With a non-GNU `make', it is safer to compile the package for one
+architecture at a time in the source code directory. After you have
+installed the package for one architecture, use `make distclean' before
+reconfiguring for another architecture.
+
+ On MacOS X 10.5 and later systems, you can create libraries and
+executables that work on multiple system types--known as "fat" or
+"universal" binaries--by specifying multiple `-arch' options to the
+compiler but only a single `-arch' option to the preprocessor. Like
+this:
+
+ ./configure CC="gcc -arch i386 -arch x86_64 -arch ppc -arch ppc64" \
+ CXX="g++ -arch i386 -arch x86_64 -arch ppc -arch ppc64" \
+ CPP="gcc -E" CXXCPP="g++ -E"
+
+ This is not guaranteed to produce working output in all cases, you
+may have to build one architecture at a time and combine the results
+using the `lipo' tool if you have problems.
+
+Installation Names
+==================
+
+ By default, `make install' installs the package's commands under
+`/usr/local/bin', include files under `/usr/local/include', etc. You
+can specify an installation prefix other than `/usr/local' by giving
+`configure' the option `--prefix=PREFIX'.
+
+ You can specify separate installation prefixes for
+architecture-specific files and architecture-independent files. If you
+pass the option `--exec-prefix=PREFIX' to `configure', the package uses
+PREFIX as the prefix for installing programs and libraries.
+Documentation and other data files still use the regular prefix.
+
+ In addition, if you use an unusual directory layout you can give
+options like `--bindir=DIR' to specify different values for particular
+kinds of files. Run `configure --help' for a list of the directories
+you can set and what kinds of files go in them.
+
+ If the package supports it, you can cause programs to be installed
+with an extra prefix or suffix on their names by giving `configure' the
+option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'.
+
+Optional Features
+=================
+
+ Some packages pay attention to `--enable-FEATURE' options to
+`configure', where FEATURE indicates an optional part of the package.
+They may also pay attention to `--with-PACKAGE' options, where PACKAGE
+is something like `gnu-as' or `x' (for the X Window System). The
+`README' should mention any `--enable-' and `--with-' options that the
+package recognizes.
+
+ For packages that use the X Window System, `configure' can usually
+find the X include and library files automatically, but if it doesn't,
+you can use the `configure' options `--x-includes=DIR' and
+`--x-libraries=DIR' to specify their locations.
+
+Particular systems
+==================
+
+ On HP-UX, the default C compiler is not ANSI C compatible. If GNU
+CC is not installed, it is recommended to use the following options in
+order to use an ANSI C compiler:
+
+ ./configure CC="cc -Ae -D_XOPEN_SOURCE=500"
+
+and if that doesn't work, install pre-built binaries of GCC for HP-UX.
+
+ On OSF/1 a.k.a. Tru64, some versions of the default C compiler cannot
+parse its `<wchar.h>' header file. The option `-nodtk' can be used as
+a workaround. If GNU CC is not installed, it is therefore recommended
+to try
+
+ ./configure CC="cc"
+
+and if that doesn't work, try
+
+ ./configure CC="cc -nodtk"
+
+ On Solaris, don't put `/usr/ucb' early in your `PATH'. This
+directory contains several dysfunctional programs; working variants of
+these programs are available in `/usr/bin'. So, if you need `/usr/ucb'
+in your `PATH', put it _after_ `/usr/bin'.
+
+ On Haiku, software installed for all users goes in `/boot/common',
+not `/usr/local'. It is recommended to use the following options:
+
+ ./configure --prefix=/boot/common
+
+Specifying the System Type
+==========================
+
+ There may be some features `configure' cannot figure out
+automatically, but needs to determine by the type of machine the package
+will run on. Usually, assuming the package is built to be run on the
+_same_ architectures, `configure' can figure that out, but if it prints
+a message saying it cannot guess the machine type, give it the
+`--build=TYPE' option. TYPE can either be a short name for the system
+type, such as `sun4', or a canonical name which has the form:
+
+ CPU-COMPANY-SYSTEM
+
+where SYSTEM can have one of these forms:
+
+ OS
+ KERNEL-OS
+
+ See the file `config.sub' for the possible values of each field. If
+`config.sub' isn't included in this package, then this package doesn't
+need to know the machine type.
+
+ If you are _building_ compiler tools for cross-compiling, you should
+use the option `--target=TYPE' to select the type of system they will
+produce code for.
+
+ If you want to _use_ a cross compiler, that generates code for a
+platform different from the build platform, you should specify the
+"host" platform (i.e., that on which the generated programs will
+eventually be run) with `--host=TYPE'.
+
+Sharing Defaults
+================
+
+ If you want to set default values for `configure' scripts to share,
+you can create a site shell script called `config.site' that gives
+default values for variables like `CC', `cache_file', and `prefix'.
+`configure' looks for `PREFIX/share/config.site' if it exists, then
+`PREFIX/etc/config.site' if it exists. Or, you can set the
+`CONFIG_SITE' environment variable to the location of the site script.
+A warning: not all `configure' scripts look for a site script.
+
+Defining Variables
+==================
+
+ Variables not defined in a site shell script can be set in the
+environment passed to `configure'. However, some packages may run
+configure again during the build, and the customized values of these
+variables may be lost. In order to avoid this problem, you should set
+them in the `configure' command line, using `VAR=value'. For example:
+
+ ./configure CC=/usr/local2/bin/gcc
+
+causes the specified `gcc' to be used as the C compiler (unless it is
+overridden in the site shell script).
+
+Unfortunately, this technique does not work for `CONFIG_SHELL' due to
+an Autoconf bug. Until the bug is fixed you can use this workaround:
+
+ CONFIG_SHELL=/bin/bash /bin/bash ./configure CONFIG_SHELL=/bin/bash
+
+`configure' Invocation
+======================
+
+ `configure' recognizes the following options to control how it
+operates.
+
+`--help'
+`-h'
+ Print a summary of all of the options to `configure', and exit.
+
+`--help=short'
+`--help=recursive'
+ Print a summary of the options unique to this package's
+ `configure', and exit. The `short' variant lists options used
+ only in the top level, while the `recursive' variant lists options
+ also present in any nested packages.
+
+`--version'
+`-V'
+ Print the version of Autoconf used to generate the `configure'
+ script, and exit.
+
+`--cache-file=FILE'
+ Enable the cache: use and save the results of the tests in FILE,
+ traditionally `config.cache'. FILE defaults to `/dev/null' to
+ disable caching.
+
+`--config-cache'
+`-C'
+ Alias for `--cache-file=config.cache'.
+
+`--quiet'
+`--silent'
+`-q'
+ Do not print messages saying which checks are being made. To
+ suppress all normal output, redirect it to `/dev/null' (any error
+ messages will still be shown).
+
+`--srcdir=DIR'
+ Look for the package's source code in directory DIR. Usually
+ `configure' can determine that directory automatically.
+
+`--prefix=DIR'
+ Use DIR as the installation prefix. *Note Installation Names::
+ for more details, including other options available for fine-tuning
+ the installation locations.
+
+`--no-create'
+`-n'
+ Run the configure checks, but stop before creating any output
+ files.
+
+`configure' also accepts some other, not widely useful, options. Run
+`configure --help' for more details.
+
diff --git a/Makefile.am b/Makefile.am
new file mode 100644
index 0000000..9d6cf1e
--- /dev/null
+++ b/Makefile.am
@@ -0,0 +1,85 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+DIST_SUBDIRS = lib src po tests debug
+SUBDIRS =
+
+if COND_GNULIB
+SUBDIRS += lib
+endif
+
+SUBDIRS += src po tests
+
+dist_doc_DATA = \
+ AUTHORS \
+ COPYING \
+ COPYING.GPLv2 \
+ NEWS \
+ README \
+ THANKS \
+ TODO \
+ doc/faq.txt \
+ doc/history.txt \
+ doc/xz-file-format.txt \
+ doc/lzma-file-format.txt
+
+EXTRA_DIST = \
+ extra \
+ dos \
+ windows \
+ autogen.sh \
+ version.sh \
+ Doxyfile.in \
+ COPYING.GPLv2 \
+ COPYING.GPLv3 \
+ COPYING.LGPLv2.1 \
+ INSTALL.generic \
+ PACKAGERS
+
+ACLOCAL_AMFLAGS = -I m4
+
+# List of man pages to conver to PDF and plain text in the dist-hook target.
+manfiles = \
+ src/xz/xz.1 \
+ src/xzdec/xzdec.1 \
+ src/lzmainfo/lzmainfo.1 \
+ src/scripts/xzdiff.1 \
+ src/scripts/xzgrep.1 \
+ src/scripts/xzless.1 \
+ src/scripts/xzmore.1
+
+# Create ChangeLog from output of "git log --date=iso --stat".
+# Convert the man pages to PDF and plain text (ASCII only) formats.
+dist-hook:
+ if test -d "$(srcdir)/.git" && type git > /dev/null 2>&1; then \
+ ( cd "$(srcdir)" && git log --date=iso --stat ) \
+ > "$(distdir)/ChangeLog"; \
+ fi
+ if type groff > /dev/null 2>&1 && type ps2pdf > /dev/null 2>&1; then \
+ dest="$(distdir)/doc/man" && \
+ $(MKDIR_P) "$$dest/pdf-a4" "$$dest/pdf-letter" "$$dest/txt" && \
+ for FILE in $(manfiles); do \
+ BASE=`basename $$FILE .1` && \
+ groff -man -t -Tps -P-pa4 < "$(srcdir)/$$FILE" \
+ | ps2pdf - - \
+ > "$$dest/pdf-a4/$$BASE-a4.pdf" && \
+ groff -man -t -Tps -P-pletter < "$(srcdir)/$$FILE" \
+ | ps2pdf - - \
+ > "$$dest/pdf-letter/$$BASE-letter.pdf" && \
+ groff -man -t -Tascii < "$(srcdir)/$$FILE" \
+ | col -bx > "$$dest/txt/$$BASE.txt"; \
+ done; \
+ fi
+
+# This works with GNU tar and gives cleaner package than normal 'make dist'.
+mydist:
+ if test -d "$(srcdir)/.git" && type git > /dev/null 2>&1; then \
+ SNAPSHOT=`cd "$(srcdir)" && git describe --abbrev=4 | cut -b2-`; \
+ test -n "$$SNAPSHOT" && VERSION=$$SNAPSHOT; \
+ fi; \
+ TAR_OPTIONS='--owner=0 --group=0 --numeric-owner --mode=u+rw,go+r-w' \
+ $(MAKE) VERSION="$$VERSION" dist-gzip
diff --git a/NEWS b/NEWS
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/NEWS
diff --git a/PACKAGERS b/PACKAGERS
new file mode 100644
index 0000000..c5d375a
--- /dev/null
+++ b/PACKAGERS
@@ -0,0 +1,279 @@
+
+Information to packagers of XZ Utils
+====================================
+
+ 0. Preface
+ 1. Package naming
+ 2. Package description
+ 3. License
+ 4. configure options
+ 4.1. Static vs. dynamic linking of liblzma
+ 4.2. Optimizing xzdec and lzmadec
+ 5. Additional documentation
+ 6. Extra files
+ 7. Installing XZ Utils and LZMA Utils in parallel
+ 8. Example
+
+
+0. Preface
+----------
+
+ This document is meant for people who create and maintain XZ Utils
+ packages for operating system distributions. The focus is on GNU/Linux
+ systems, but most things apply to other systems too.
+
+ While the standard "configure && make DESTDIR=$PKG install" should
+ give a pretty good package, there are some details which packagers
+ may want to tweak.
+
+ Packagers should also read the INSTALL file.
+
+
+1. Package naming
+-----------------
+
+ The preferred name for the XZ Utils package is "xz", because that's
+ the name of the upstream tarball. Naturally you may have good reasons
+ to use some other name; I won't get angry about it. ;-) It's just nice
+ to be able to point people to the correct package name without asking
+ what distro they have.
+
+ If your distro policy is to split things into small pieces, here is
+ one suggestion:
+
+ xz xz, xzdec, scripts (xzdiff, xzgrep, etc.), docs
+ xz-lzma lzma, unlzma, lzcat, lzgrep etc. symlinks and
+ lzmadec binary for compatibility with LZMA Utils
+ liblzma liblzma.so.*
+ liblzma-devel liblzma.so, liblzma.a, API headers
+
+
+2. Package description
+----------------------
+
+ Here is a suggestion which you may use as the package description.
+ If you can use only one-line description, pick only the first line.
+ Naturally, feel free to use some other description if you find it
+ better, and maybe send it to me too.
+
+ Library and command line tools for XZ and LZMA compressed files
+
+ XZ Utils provide a general purpose data compression library
+ and command line tools. The native file format is the .xz
+ format, but also the legacy .lzma format is supported. The .xz
+ format supports multiple compression algorithms, of which LZMA2
+ is currently the primary algorithm. With typical files, XZ Utils
+ create about 30 % smaller files than gzip.
+
+ If you are splitting XZ Utils into multiple packages, here are some
+ suggestions for package descriptions:
+
+ xz:
+
+ Command line tools for XZ and LZMA compressed files
+
+ This package includes the xz compression tool and other command
+ line tools from XZ Utils. xz has command line syntax similar to
+ that of gzip. The native file format is the .xz format, but also
+ the legacy .lzma format is supported. The .xz format supports
+ multiple compression algorithms, of which LZMA2 is currently the
+ primary algorithm. With typical files, XZ Utils create about 30 %
+ smaller files than gzip.
+
+ Note that this package doesn't include the files needed for
+ LZMA Utils 4.32.x compatibility. Install also the xz-lzma
+ package to make XZ Utils emulate LZMA Utils 4.32.x.
+
+ xz-lzma:
+
+ LZMA Utils emulation with XZ Utils
+
+ This package includes executables and symlinks to make
+ XZ Utils emulate lzma, unlzma, lzcat, and other command
+ line tools found from the legacy LZMA Utils 4.32.x package.
+
+ liblzma:
+
+ Library for XZ and LZMA compressed files
+
+ liblzma is a general purpose data compression library with
+ an API similar to that of zlib. liblzma supports multiple
+ algorithms, of which LZMA2 is currently the primary algorithm.
+ The native file format is .xz, but also the legacy .lzma
+ format and raw streams (no headers at all) are supported.
+
+ This package includes the shared library.
+
+ liblzma-devel:
+
+ Library for XZ and LZMA compressed files
+
+ This package includes the API headers, static library, and
+ other development files related to liblzma.
+
+
+3. License
+----------
+
+ If the package manager supports a license field, you probably should
+ put GPLv2+ there (GNU GPL v2 or later). The interesting parts of
+ XZ Utils are in the public domain, but some less important files
+ ending up into the binary package are under GPLv2+. So it is simplest
+ to just say GPLv2+ if you cannot specify "public domain and GPLv2+".
+
+ If you split XZ Utils into multiple packages as described earlier
+ in this file, liblzma and liblzma-dev packages will contain only
+ public domain code (from XZ Utils at least; compiler or linker may
+ add some third-party code, which may be copyrighted).
+
+
+4. configure options
+--------------------
+
+ Unless you are building a package for a distribution that is meant
+ only for embedded systems, don't use the following configure options:
+
+ --enable-debug
+ --enable-encoders (*)
+ --enable-decoders
+ --enable-match-finders
+ --enable-checks
+ --enable-small (*)
+ --disable-threads (*)
+
+ (*) These are OK when building xzdec and lzmadec as explained later.
+
+ You may use --enable-werror but be careful with it since it may break
+ the build due to some useless warning when the build environment
+ changes (like CPU architecture or compiler version).
+
+
+4.1. Static vs. dynamic linking of liblzma
+
+ The default is to link the most important command line tools against
+ static liblzma, and the less important tools against shared liblzma.
+ This can be changed by passing --enable-dynamic to configure, or by
+ not building static libraries at all by passing --disable-static
+ to configure. It is mildly recommended that you use the default, but
+ the configure options make it easy to do otherwise if the distro policy
+ so requires.
+
+ On 32-bit x86, linking against static liblzma can give a minor
+ speed improvement. Static libraries on x86 are usually compiled as
+ position-dependent code (non-PIC) and shared libraries are built as
+ position-independent code (PIC). PIC wastes one register, which can
+ make the code slightly slower compared to a non-PIC version. (Note
+ that this doesn't apply to x86-64.)
+
+ Linking against static liblzma avoids a dependency on liblzma shared
+ library, and makes it slightly easier to copy the command line tools
+ between systems (e.g. quick 'n' dirty emergency recovery of some
+ files). It also allows putting the command line tools to /bin while
+ leaving liblzma to /usr/lib (assuming that your distribution uses
+ such a file system hierarchy), if no other file in /bin would require
+ liblzma.
+
+ If you don't want to distribute static libraries but you still
+ want to link the command line tools against static liblzma, it is
+ probably easiest to build both static and shared liblzma, but after
+ "make DESTDIR=$PKG install" remove liblzma.a and modify liblzma.la
+ to not contain a reference to liblzma.a.
+
+
+4.2. Optimizing xzdec and lzmadec
+
+ xzdec and lzmadec are intended to be relatively small instead of
+ optimizing for the best speed. Thus, it is a good idea to build
+ xzdec and lzmadec separately:
+
+ - Only decoder code is needed, so you can speed up the build
+ slightly by passing --disable-encoders to configure. This
+ shouldn't affect the final size of the executables though,
+ because the linker is able to omit the encoder code anyway.
+
+ - xzdec and lzmadec will never use multithreading capabilities of
+ liblzma. You can avoid dependency on libpthread by passing
+ --disable-threads to configure.
+
+ - There are and will be no translated messages for xzdec and
+ lzmadec, so it is fine to pass also --disable-nls to configure.
+
+ - To select somewhat size-optimized variant of some things in
+ liblzma, pass --enable-small to configure.
+
+ - Tell the compiler to optimize for size instead of speed.
+ E.g. with GCC, put -Os into CFLAGS.
+
+
+5. Additional documentation
+---------------------------
+
+ "make install" copies some additional documentation to $docdir
+ (--docdir in configure). These a copy of the GNU GPL v2, which can
+ be replaced with a symlink if your distro ships with shared copies
+ of the common license texts.
+
+
+6. Extra files
+--------------
+
+ The "extra" directory contains some small extra tools or other files.
+ The exact set of extra files can vary between XZ Utils releases. The
+ extra files have only limited use or they are too dangerous to be
+ put directly to $bindir (7z2lzma.sh is a good example, since it can
+ silently create corrupt output if certain conditions are not met).
+
+ If you feel like it, you may copy the extra directory under the doc
+ directory (e.g. /usr/share/doc/xz/extra). Maybe some people will find
+ them useful. However, most people needing these tools probably are
+ able to find them from the source package too.
+
+ The "debug" directory contains some tools that are useful only when
+ hacking on XZ Utils. Don't package these tools.
+
+
+7. Installing XZ Utils and LZMA Utils in parallel
+-------------------------------------------------
+
+ XZ Utils and LZMA Utils 4.32.x can be installed in parallel by
+ omitting the compatibility symlinks (lzma, unlzma, lzcat, lzgrep etc.)
+ from the XZ Utils package. It's probably a good idea to still package
+ the symlinks into a separate package so that users may choose if they
+ want to use XZ Utils or LZMA Utils for handling .lzma files.
+
+
+8. Example
+----------
+
+ Here is an example for i686 GNU/Linux that
+ - links xz against static liblzma;
+ - includes only shared liblzma in the final package;
+ - links xzdec and lzmadec against static liblzma while
+ avoiding libpthread dependency.
+
+ PKG=/tmp/xz-pkg
+ tar xf xz-x.y.z.tar.gz
+ cd xz-x.y.z
+ ./configure \
+ --prefix=/usr \
+ --sysconfdir=/etc \
+ CFLAGS='-march=i686 -O2'
+ make
+ make DESTDIR=$PKG install-strip
+ rm -f $PKG/usr/lib/lib*.a
+ sed -i "s/^old_library=.*$/old_library=''/" $PKG/usr/lib/lib*.la
+ make clean
+ ./configure \
+ --prefix=/usr \
+ --sysconfdir=/etc \
+ --disable-shared \
+ --disable-nls \
+ --disable-encoders \
+ --enable-small \
+ --disable-threads \
+ CFLAGS='-march=i686 -Os'
+ make -C src/liblzma
+ make -C src/xzdec
+ make -C src/xzdec DESTDIR=$PKG install-strip
+ cp -a extra $PKG/usr/share/doc/xz
+
diff --git a/README b/README
new file mode 100644
index 0000000..6b695d5
--- /dev/null
+++ b/README
@@ -0,0 +1,218 @@
+
+XZ Utils
+========
+
+ 0. Overview
+ 1. Documentation
+ 1.1. Overall documentation
+ 1.2. Documentation for command line tools
+ 1.3. Documentation for liblzma
+ 2. Version numbering
+ 3. Reporting bugs
+ 4. Other implementations of the .xz format
+ 5. Contact information
+
+
+0. Overview
+-----------
+
+ XZ Utils provide a general-purpose data compression library and
+ command line tools. The native file format is the .xz format, but
+ also the legacy .lzma format is supported. The .xz format supports
+ multiple compression algorithms, which are called "filters" in
+ context of XZ Utils. The primary filter is currently LZMA2. With
+ typical files, XZ Utils create about 30 % smaller files than gzip.
+
+ To ease adapting support for the .xz format into existing applications
+ and scripts, the API of liblzma is somewhat similar to the API of the
+ popular zlib library. For the same reason, the command line tool xz
+ has similar command line syntax than that of gzip.
+
+ When aiming for the highest compression ratio, LZMA2 encoder uses
+ a lot of CPU time and may use, depending on the settings, even
+ hundreds of megabytes of RAM. However, in fast modes, LZMA2 encoder
+ competes with bzip2 in compression speed, RAM usage, and compression
+ ratio.
+
+ LZMA2 is reasonably fast to decompress. It is a little slower than
+ gzip, but a lot faster than bzip2. Being fast to decompress means
+ that the .xz format is especially nice when the same file will be
+ decompressed very many times (usually on different computers), which
+ is the case e.g. when distributing software packages. In such
+ situations, it's not too bad if the compression takes some time,
+ since that needs to be done only once to benefit many people.
+
+ With some file types, combining (or "chaining") LZMA2 with an
+ additional filter can improve compression ratio. A filter chain may
+ contain up to four filters, although usually only one two is used.
+ For example, putting a BCJ (Branch/Call/Jump) filter before LZMA2
+ in the filter chain can improve compression ratio of executable files.
+
+ Since the .xz format allows adding new filter IDs, it is possible that
+ some day there will be a filter that is, for example, much faster to
+ compress than LZMA2 (but probably with worse compression ratio).
+ Similarly, it is possible that some day there is a filter that will
+ compress better than LZMA2.
+
+ XZ Utils doesn't support multithreaded compression or decompression
+ yet. It has been planned though and taken into account when designing
+ the .xz file format.
+
+
+1. Documentation
+----------------
+
+1.1. Overall documentation
+
+ README This file
+
+ INSTALL.generic Generic install instructions for those not familiar
+ with packages using GNU Autotools
+ INSTALL Installation instructions specific to XZ Utils
+ PACKAGERS Information to packagers of XZ Utils
+
+ COPYING XZ Utils copyright and license information
+ COPYING.GPLv2 GNU General Public License version 2
+ COPYING.GPLv3 GNU General Public License version 3
+ COPYING.LGPLv2.1 GNU Lesser General Public License version 2.1
+
+ AUTHORS The main authors of XZ Utils
+ THANKS Incomplete list of people who have helped making
+ this software
+ NEWS User-visible changes between XZ Utils releases
+ ChangeLog Detailed list of changes (commit log)
+ TODO Known bugs and some sort of to-do list
+
+ Note that only some of the above files are included in binary
+ packages.
+
+
+1.2. Documentation for command line tools
+
+ The command line tools are documented as man pages. In source code
+ releases (and possibly also in some binary packages), the man pages
+ are also provided in plain text (ASCII only) and PDF formats in the
+ directory "doc/man" to make the man pages more accessible to those
+ whose operating system doesn't provide an easy way to view man pages.
+
+
+1.3. Documentation for liblzma
+
+ The liblzma API headers include short docs about each function
+ and data type as Doxygen tags. These docs should be quite OK as
+ a quick reference.
+
+ I have planned to write a bunch of very well documented example
+ programs, which (due to comments) should work as a tutorial to
+ various features of liblzma. No such example programs have been
+ written yet.
+
+ For now, if you have never used liblzma, libbzip2, or zlib, I
+ recommend learning *basics* of zlib API. Once you know that, it
+ should be easier to learn liblzma.
+
+ http://zlib.net/manual.html
+ http://zlib.net/zlib_how.html
+
+
+2. Version numbering
+--------------------
+
+ The version number format of XZ Utils is X.Y.ZS:
+
+ - X is the major version. When this is incremented, the library
+ API and ABI break.
+
+ - Y is the minor version. It is incremented when new features are
+ added without breaking existing API or ABI. Even Y indicates
+ stable release and odd Y indicates unstable (alpha or beta
+ version).
+
+ - Z is the revision. This has different meaning for stable and
+ unstable releases:
+ * Stable: Z is incremented when bugs get fixed without adding
+ any new features.
+ * Unstable: Z is just a counter. API or ABI of features added
+ in earlier unstable releases having the same X.Y may break.
+
+ - S indicates stability of the release. It is missing from the
+ stable releases where Y is an even number. When Y is odd, S
+ is either "alpha" or "beta" to make it very clear that such
+ versions are not stable releases. The same X.Y.Z combination is
+ not used for more than one stability level i.e. after X.Y.Zalpha,
+ the next version can be X.Y.(Z+1)beta but not X.Y.Zbeta.
+
+
+3. Reporting bugs
+-----------------
+
+ Naturally it is easiest for me if you already know what causes the
+ unexpected behavior. Even better if you have a patch to propose.
+ However, quite often the reason for unexpected behavior is unknown,
+ so here are a few things to do before sending a bug report:
+
+ 1. Try to create a small example how to reproduce the issue.
+
+ 2. Compile XZ Utils with debugging code using configure switches
+ --enable-debug and, if possible, --disable-shared. If you are
+ using GCC, use CFLAGS='-O0 -ggdb3'. Don't strip the resulting
+ binaries.
+
+ 3. Turn on core dumps. The exact command depends on your shell;
+ for example in GNU bash it is done with "ulimit -c unlimited",
+ and in tcsh with "limit coredumpsize unlimited".
+
+ 4. Try to reproduce the suspected bug. If you get "assertion failed"
+ message, be sure to include the complete message in your bug
+ report. If the application leaves a coredump, get a backtrace
+ using gdb:
+ $ gdb /path/to/app-binary # Load the app to the debugger.
+ (gdb) core core # Open the coredump.
+ (gdb) bt # Print the backtrace. Copy & paste to bug report.
+ (gdb) quit # Quit gdb.
+
+ Report your bug via email or IRC (see Contact information below).
+ Don't send core dump files or any executables. If you have a small
+ example file(s) (total size less than 256 KiB), please include
+ it/them as an attachment. If you have bigger test files, put them
+ online somewhere and include an URL to the file(s) in the bug report.
+
+ Always include the exact version number of XZ Utils in the bug report.
+ If you are using a snapshot from the git repository, use "git describe"
+ to get the exact snapshot version. If you are using XZ Utils shipped
+ in an operating system distribution, mention the distribution name,
+ distribution version, and exact xz package version; if you cannot
+ repeat the bug with the code compiled from unpatched source code,
+ you probably need to report a bug to your distribution's bug tracking
+ system.
+
+
+4. Other implementations of the .xz format
+------------------------------------------
+
+ 7-Zip and the p7zip port of 7-Zip support the .xz format starting
+ from the version 9.00alpha.
+
+ http://7-zip.org/
+ http://p7zip.sourceforge.net/
+
+ XZ Embedded is a limited implementation written for use in the Linux
+ kernel, but it is also suitable for other embedded use.
+
+ http://tukaani.org/xz/embedded.html
+
+
+5. Contact information
+----------------------
+
+ If you have questions, bug reports, patches etc. related to XZ Utils,
+ contact Lasse Collin <lasse.collin@tukaani.org> (in Finnish or English).
+ tukaani.org uses greylisting to reduce spam, thus when you send your
+ first email, it may get delayed by a few hours. In addition to that,
+ I'm sometimes slow at replying. If you haven't got a reply within two
+ weeks, assume that your email has got lost and resend it or use IRC.
+
+ You can find me also from #tukaani on Freenode; my nick is Larhzu.
+ The channel tends to be pretty quiet, so just ask your question and
+ someone may wake up.
+
diff --git a/THANKS b/THANKS
new file mode 100644
index 0000000..4f79e47
--- /dev/null
+++ b/THANKS
@@ -0,0 +1,69 @@
+
+Thanks
+======
+
+Some people have helped more, some less, but nevertheless everyone's help
+has been important. :-) In alphabetical order:
+ - Mark Adler
+ - H. Peter Anvin
+ - Nelson H. F. Beebe
+ - Karl Berry
+ - Anders F. Björklund
+ - Emmanuel Blot
+ - Trent W. Buck
+ - David Burklund
+ - Marek Černocký
+ - Andrew Dudman
+ - Markus Duft
+ - İsmail Dönmez
+ - Robert Elz
+ - Gilles Espinasse
+ - Denis Excoffier
+ - Mike Frysinger
+ - Joachim Henke
+ - Peter Ivanov
+ - Jouk Jansen
+ - Per Øyvind Karlsen
+ - Thomas Klausner
+ - Richard Koch
+ - Ville Koskinen
+ - Stephan Kulow
+ - Peter Lawler
+ - Hin-Tak Leung
+ - Andraž 'ruskie' Levstik
+ - Jim Meyering
+ - Rafał Mużyło
+ - Adrien Nader
+ - Hongbo Ni
+ - Jonathan Nieder
+ - Peter O'Gorman
+ - Igor Pavlov
+ - Elbert Pol
+ - Mikko Pouru
+ - Robert Readman
+ - Bernhard Reutner-Fischer
+ - Christian von Roques
+ - Jukka Salmi
+ - Alexandre Sauvé
+ - Andreas Schwab
+ - Dan Shechter
+ - Stuart Shelton
+ - Jonathan Stott
+ - Paul Townsend
+ - Mohammed Adnène Trojette
+ - Alexey Tourbin
+ - Patrick J. Volkerding
+ - Christian Weisgerber
+ - Bert Wesarg
+ - Ralf Wildenhues
+ - Charles Wilson
+ - Lars Wirzenius
+ - Pilorz Wojciech
+ - Ryan Young
+ - Andreas Zieringer
+
+Also thanks to all the people who have participated in the Tukaani project.
+
+I have probably forgot to add some names to the above list. Sorry about
+that and thanks for your help.
+
diff --git a/TODO b/TODO
new file mode 100644
index 0000000..9fac1b3
--- /dev/null
+++ b/TODO
@@ -0,0 +1,60 @@
+
+XZ Utils To-Do List
+===================
+
+Known bugs
+----------
+
+ The test suite is too incomplete.
+
+ If the memory usage limit is less than about 13 MiB, xz is unable to
+ automatically scale down the compression settings enough even though
+ it would be possible by switching from BT2/BT3/BT4 match finder to
+ HC3/HC4.
+
+ The code to detect number of CPU cores doesn't count hyperthreading
+ as multiple cores. In context of xz, it probably should.
+ Hyperthreading is good at least with p7zip.
+
+ XZ Utils compress some files significantly worse than LZMA Utils.
+ This is due to faster compression presets used by XZ Utils, and
+ can be worked around by using "xz --extreme". However, the presets
+ need some tweaking and maybe this issue can be minimized without
+ making the typical case too much slower.
+
+ xz doesn't quote unprintable characters when it displays file names
+ given on the command line.
+
+ tuklib_exit() doesn't block signals => EINTR is possible.
+
+
+Missing features
+----------------
+
+ xz doesn't support copying extended attributes, access control
+ lists etc. from source to target file.
+
+ Multithreaded compression
+
+ Multithreaded decompression
+
+ Buffer-to-buffer coding could use less RAM (especially when
+ decompressing LZMA1 or LZMA2).
+
+ I/O library is not implemented. It will possibly be named libzzf.
+
+ lzma_strerror() to convert lzma_ret to human readable form?
+ This is tricky, because the same error codes are used with
+ slightly different meanings.
+
+
+Documentation
+-------------
+
+ Some tutorial is needed for liblzma. I have planned to write some
+ extremely well commented example programs, which would work as
+ a tutorial. I suppose the Doxygen tags are quite OK as a quick
+ reference once one is familiar with the liblzma API.
+
+ Document the LZMA1 and LZMA2 algorithms.
+
diff --git a/autogen.sh b/autogen.sh
new file mode 100755
index 0000000..f0195ec
--- /dev/null
+++ b/autogen.sh
@@ -0,0 +1,22 @@
+#!/bin/sh
+
+###############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+# The result of using "autoreconf -fi" should be identical to using this
+# script. I'm leaving this script here just in case someone finds it useful.
+
+set -e -x
+
+${AUTOPOINT:-autopoint} -f
+${LIBTOOLIZE:-libtoolize} -c -f || glibtoolize -c -f
+${ACLOCAL:-aclocal} -I m4
+${AUTOCONF:-autoconf}
+${AUTOHEADER:-autoheader}
+${AUTOMAKE:-automake} -acf --foreign
diff --git a/configure.ac b/configure.ac
new file mode 100644
index 0000000..e93e806
--- /dev/null
+++ b/configure.ac
@@ -0,0 +1,693 @@
+# -*- Autoconf -*-
+# Process this file with autoconf to produce a configure script.
+
+###############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+# NOTE: Don't add useless checks. autoscan detects this and that, but don't
+# let it confuse you. For example, we don't care about checking for behavior
+# of malloc(), stat(), or lstat(), since we don't use those functions in
+# a way that would cause the problems the autoconf macros check.
+
+AC_PREREQ([2.64])
+
+AC_INIT([XZ Utils], m4_esyscmd([/bin/sh version.sh]),
+ [lasse.collin@tukaani.org], [xz], [http://tukaani.org/xz/])
+AC_CONFIG_SRCDIR([src/liblzma/common/common.h])
+AC_CONFIG_AUX_DIR([build-aux])
+AC_CONFIG_MACRO_DIR([m4])
+AC_CONFIG_HEADER([config.h])
+
+echo
+echo "$PACKAGE_STRING"
+
+echo
+echo "System type:"
+# This is needed to know if assembler optimizations can be used.
+AC_CANONICAL_HOST
+
+# We do some special things on Windows (32-bit or 64-bit) builds.
+case $host_os in
+ mingw* | cygwin*) is_w32=yes ;;
+ *) is_w32=no ;;
+esac
+AM_CONDITIONAL([COND_W32], [test "$is_w32" = yes])
+
+# We need to use $EXEEXT with $(LN_S) when creating symlinks to
+# executables. Cygwin is an exception to this, since it is recommended
+# that symlinks don't have the .exe suffix. To make this work, we
+# define LN_EXEEXT.
+case $host_os in
+ cygwin) LN_EXEEXT= ;;
+ *) LN_EXEEXT='$(EXEEXT)' ;;
+esac
+AC_SUBST([LN_EXEEXT])
+
+echo
+echo "Configure options:"
+AM_CFLAGS=
+
+
+#############
+# Debugging #
+#############
+
+AC_MSG_CHECKING([if debugging code should be compiled])
+AC_ARG_ENABLE([debug], AC_HELP_STRING([--enable-debug], [Enable debugging code.]),
+ [], enable_debug=no)
+if test "x$enable_debug" = xyes; then
+ AC_MSG_RESULT([yes])
+else
+ AC_DEFINE([NDEBUG], [1], [Define to 1 to disable debugging code.])
+ AC_MSG_RESULT([no])
+fi
+
+
+###########
+# Filters #
+###########
+
+m4_define([SUPPORTED_FILTERS], [lzma1,lzma2,delta,x86,powerpc,ia64,arm,armthumb,sparc])dnl
+m4_define([SIMPLE_FILTERS], [x86,powerpc,ia64,arm,armthumb,sparc])
+m4_define([LZ_FILTERS], [lzma1,lzma2])
+
+m4_foreach([NAME], [SUPPORTED_FILTERS],
+[enable_filter_[]NAME=no
+enable_encoder_[]NAME=no
+enable_decoder_[]NAME=no
+])dnl
+
+AC_MSG_CHECKING([which encoders to build])
+AC_ARG_ENABLE([encoders], AC_HELP_STRING([--enable-encoders=LIST],
+ [Comma-separated list of encoders to build. Default=all.
+ Available encoders:]
+ m4_translit(m4_defn([SUPPORTED_FILTERS]), [,], [ ])),
+ [], [enable_encoders=SUPPORTED_FILTERS])
+enable_encoders=`echo "$enable_encoders" | sed 's/,/ /g'`
+if test "x$enable_encoders" = xno || test "x$enable_encoders" = x; then
+ AC_MSG_RESULT([(none)])
+else
+ AC_DEFINE([HAVE_ENCODER], [1],
+ [Define to 1 if encoder components are enabled.])
+ for arg in $enable_encoders
+ do
+ case $arg in m4_foreach([NAME], [SUPPORTED_FILTERS], [
+ NAME)
+ enable_filter_[]NAME=yes
+ enable_encoder_[]NAME=yes
+ AC_DEFINE(HAVE_ENCODER_[]m4_toupper(NAME), [1],
+ [Define to 1 if] NAME [encoder is enabled.])
+ ;;])
+ *)
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([unknown filter: $arg])
+ ;;
+ esac
+ done
+ AC_MSG_RESULT([$enable_encoders])
+fi
+
+AC_MSG_CHECKING([which decoders to build])
+AC_ARG_ENABLE([decoders], AC_HELP_STRING([--enable-decoders=LIST],
+ [Comma-separated list of decoders to build. Default=all.
+ Available decoders are the same as available encoders.]),
+ [], [enable_decoders=SUPPORTED_FILTERS])
+enable_decoders=`echo "$enable_decoders" | sed 's/,/ /g'`
+if test "x$enable_decoders" = xno || test "x$enable_decoders" = x; then
+ AC_MSG_RESULT([(none)])
+else
+ AC_DEFINE([HAVE_DECODER], [1],
+ [Define to 1 if decoder components are enabled.])
+ for arg in $enable_decoders
+ do
+ case $arg in m4_foreach([NAME], [SUPPORTED_FILTERS], [
+ NAME)
+ enable_filter_[]NAME=yes
+ enable_decoder_[]NAME=yes
+ AC_DEFINE(HAVE_DECODER_[]m4_toupper(NAME), [1],
+ [Define to 1 if] NAME [decoder is enabled.])
+ ;;])
+ *)
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([unknown filter: $arg])
+ ;;
+ esac
+ done
+
+ # LZMA2 requires that LZMA1 is enabled.
+ test "x$enable_encoder_lzma2" = xyes && enable_encoder_lzma1=yes
+ test "x$enable_decoder_lzma2" = xyes && enable_decoder_lzma1=yes
+
+ AC_MSG_RESULT([$enable_decoders])
+fi
+
+if test "x$enable_encoder_lzma2$enable_encoder_lzma1" = xyesno \
+ || test "x$enable_decoder_lzma2$enable_decoder_lzma1" = xyesno; then
+ AC_MSG_ERROR([LZMA2 requires that LZMA1 is also enabled.])
+fi
+
+AM_CONDITIONAL(COND_MAIN_ENCODER, test "x$enable_encoders" != xno && test "x$enable_encoders" != x)
+AM_CONDITIONAL(COND_MAIN_DECODER, test "x$enable_decoders" != xno && test "x$enable_decoders" != x)
+
+m4_foreach([NAME], [SUPPORTED_FILTERS],
+[AM_CONDITIONAL(COND_FILTER_[]m4_toupper(NAME), test "x$enable_filter_[]NAME" = xyes)
+AM_CONDITIONAL(COND_ENCODER_[]m4_toupper(NAME), test "x$enable_encoder_[]NAME" = xyes)
+AM_CONDITIONAL(COND_DECODER_[]m4_toupper(NAME), test "x$enable_decoder_[]NAME" = xyes)
+])dnl
+
+# The so called "simple filters" share common code.
+enable_filter_simple=no
+enable_encoder_simple=no
+enable_decoder_simple=no
+m4_foreach([NAME], [SIMPLE_FILTERS],
+[test "x$enable_filter_[]NAME" = xyes && enable_filter_simple=yes
+test "x$enable_encoder_[]NAME" = xyes && enable_encoder_simple=yes
+test "x$enable_decoder_[]NAME" = xyes && enable_decoder_simple=yes
+])dnl
+AM_CONDITIONAL(COND_FILTER_SIMPLE, test "x$enable_filter_simple" = xyes)
+AM_CONDITIONAL(COND_ENCODER_SIMPLE, test "x$enable_encoder_simple" = xyes)
+AM_CONDITIONAL(COND_DECODER_SIMPLE, test "x$enable_decoder_simple" = xyes)
+
+# LZ-based filters share common code.
+enable_filter_lz=no
+enable_encoder_lz=no
+enable_decoder_lz=no
+m4_foreach([NAME], [LZ_FILTERS],
+[test "x$enable_filter_[]NAME" = xyes && enable_filter_lz=yes
+test "x$enable_encoder_[]NAME" = xyes && enable_encoder_lz=yes
+test "x$enable_decoder_[]NAME" = xyes && enable_decoder_lz=yes
+])dnl
+AM_CONDITIONAL(COND_FILTER_LZ, test "x$enable_filter_lz" = xyes)
+AM_CONDITIONAL(COND_ENCODER_LZ, test "x$enable_encoder_lz" = xyes)
+AM_CONDITIONAL(COND_DECODER_LZ, test "x$enable_decoder_lz" = xyes)
+
+
+#################
+# Match finders #
+#################
+
+m4_define([SUPPORTED_MATCH_FINDERS], [hc3,hc4,bt2,bt3,bt4])
+
+m4_foreach([NAME], [SUPPORTED_MATCH_FINDERS],
+[enable_match_finder_[]NAME=no
+])
+
+AC_MSG_CHECKING([which match finders to build])
+AC_ARG_ENABLE([match-finders], AC_HELP_STRING([--enable-match-finders=LIST],
+ [Comma-separated list of match finders to build. Default=all.
+ At least one match finder is required for encoding with
+ the LZMA1 and LZMA2 filters. Available match finders:]
+ m4_translit(m4_defn([SUPPORTED_MATCH_FINDERS]), [,], [ ])), [],
+ [enable_match_finders=SUPPORTED_MATCH_FINDERS])
+enable_match_finders=`echo "$enable_match_finders" | sed 's/,/ /g'`
+if test "x$enable_encoder_lz" = xyes ; then
+ for arg in $enable_match_finders
+ do
+ case $arg in m4_foreach([NAME], [SUPPORTED_MATCH_FINDERS], [
+ NAME)
+ enable_match_finder_[]NAME=yes
+ AC_DEFINE(HAVE_MF_[]m4_toupper(NAME), [1],
+ [Define to 1 to enable] NAME [match finder.])
+ ;;])
+ *)
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([unknown match finder: $arg])
+ ;;
+ esac
+ done
+ AC_MSG_RESULT([$enable_match_finders])
+else
+ AC_MSG_RESULT([(none because not building any LZ-based encoder)])
+fi
+
+
+####################
+# Integrity checks #
+####################
+
+m4_define([SUPPORTED_CHECKS], [crc32,crc64,sha256])
+
+m4_foreach([NAME], [SUPPORTED_FILTERS],
+[enable_check_[]NAME=no
+])dnl
+
+AC_MSG_CHECKING([which integrity checks to build])
+AC_ARG_ENABLE([checks], AC_HELP_STRING([--enable-checks=LIST],
+ [Comma-separated list of integrity checks to build.
+ Default=all. Available integrity checks:]
+ m4_translit(m4_defn([SUPPORTED_CHECKS]), [,], [ ])),
+ [], [enable_checks=SUPPORTED_CHECKS])
+enable_checks=`echo "$enable_checks" | sed 's/,/ /g'`
+if test "x$enable_checks" = xno || test "x$enable_checks" = x; then
+ AC_MSG_RESULT([(none)])
+else
+ for arg in $enable_checks
+ do
+ case $arg in m4_foreach([NAME], [SUPPORTED_CHECKS], [
+ NAME)
+ enable_check_[]NAME=yes
+ AC_DEFINE(HAVE_CHECK_[]m4_toupper(NAME), [1],
+ [Define to 1 if] NAME
+ [integrity check is enabled.])
+ ;;])
+ *)
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([unknown integrity check: $arg])
+ ;;
+ esac
+ done
+ AC_MSG_RESULT([$enable_checks])
+fi
+if test "x$enable_checks_crc32" = xno ; then
+ AC_MSG_ERROR([For now, the CRC32 check must always be enabled.])
+fi
+
+m4_foreach([NAME], [SUPPORTED_CHECKS],
+[AM_CONDITIONAL(COND_CHECK_[]m4_toupper(NAME), test "x$enable_check_[]NAME" = xyes)
+])dnl
+
+
+###########################
+# Assembler optimizations #
+###########################
+
+AC_MSG_CHECKING([if assembler optimizations should be used])
+AC_ARG_ENABLE([assembler], AC_HELP_STRING([--disable-assembler],
+ [Do not use assembler optimizations even if such exist
+ for the architecture.]),
+ [], [enable_assembler=yes])
+if test "x$enable_assembler" = xyes; then
+ enable_assembler=no
+ case $host_os in
+ # Darwin should work too but only if not creating universal
+ # binaries. Solaris x86 could work too but I cannot test.
+ linux* | *bsd* | mingw* | cygwin*)
+ case $host_cpu in
+ i?86) enable_assembler=x86 ;;
+ x86_64) enable_assembler=x86_64 ;;
+ esac
+ ;;
+ esac
+fi
+case $enable_assembler in
+ x86 | x86_64 | no)
+ AC_MSG_RESULT([$enable_assembler])
+ ;;
+ *)
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([--enable-assembler accepts only \`yes', \`no', \`x86', or \`x86_64'.])
+ ;;
+esac
+AM_CONDITIONAL(COND_ASM_X86, test "x$enable_assembler" = xx86)
+AM_CONDITIONAL(COND_ASM_X86_64, test "x$enable_assembler" = xx86_64)
+
+
+#####################
+# Size optimization #
+#####################
+
+AC_MSG_CHECKING([if small size is preferred over speed])
+AC_ARG_ENABLE([small], AC_HELP_STRING([--enable-small],
+ [Make liblzma smaller and a little slower.
+ This is disabled by default to optimize for speed.]),
+ [], [enable_small=no])
+if test "x$enable_small" = xyes; then
+ AC_DEFINE([HAVE_SMALL], [1], [Define to 1 if optimizing for size.])
+elif test "x$enable_small" != xno; then
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([--enable-small accepts only \`yes' or \`no'])
+fi
+AC_MSG_RESULT([$enable_small])
+AM_CONDITIONAL(COND_SMALL, test "x$enable_small" = xyes)
+
+
+#############
+# Threading #
+#############
+
+AC_MSG_CHECKING([if threading support is wanted])
+AC_ARG_ENABLE([threads], AC_HELP_STRING([--disable-threads],
+ [Disable threading support.
+ This makes some things thread-unsafe.]),
+ [], [enable_threads=yes])
+if test "x$enable_threads" != xyes && test "x$enable_threads" != xno; then
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([--enable-threads accepts only \`yes' or \`no'])
+fi
+AC_MSG_RESULT([$enable_threads])
+# We use the actual result a little later.
+
+
+#########################
+# Assumed amount of RAM #
+#########################
+
+# We use 128 MiB as default, because it will allow decompressing files
+# created with "xz -9". It would be slightly safer to guess a lower value,
+# but most systems, on which we don't have any way to determine the amount
+# of RAM, will probably have at least 128 MiB of RAM.
+AC_MSG_CHECKING([how much RAM to assume if the real amount is unknown])
+AC_ARG_ENABLE([assume-ram], AC_HELP_STRING([--enable-assume-ram=SIZE],
+ [If and only if the real amount of RAM cannot be determined,
+ assume SIZE MiB. The default is 128 MiB. This affects the
+ default memory usage limit.]),
+ [], [enable_assume_ram=128])
+assume_ram_check=`echo "$enable_assume_ram" | tr -d 0123456789`
+if test -z "$enable_assume_ram" || test -n "$assume_ram_check"; then
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([--enable-assume-ram accepts only an integer argument])
+fi
+AC_MSG_RESULT([$enable_assume_ram MiB])
+AC_DEFINE_UNQUOTED([ASSUME_RAM], [$enable_assume_ram],
+ [How many MiB of RAM to assume if the real amount cannot
+ be determined.])
+
+
+############################################
+# xz/xzdec/lzmadec linkage against liblzma #
+############################################
+
+# Link the xz, xzdec, and lzmadec command line tools against static liblzma
+# unless using --enable-dynamic. Using static liblzma gives a little bit
+# faster executable on x86, because no register is wasted for PIC. We also
+# have one dependency less, which allows users to more freely copy the xz
+# binary to other boxes. However, I wouldn't be surprised if distro
+# maintainers still prefer dynamic linking, so let's make it easy for them.
+
+AC_MSG_CHECKING([how programs should be linked against liblzma])
+AC_ARG_ENABLE([dynamic], [AC_HELP_STRING([--enable-dynamic=TYPE],
+ [Set how command line tools are linked against liblzma.
+ TYPE can be mixed, yes, or no. The default is mixed.])],
+ [], [enable_dynamic=mixed])
+case $enable_dynamic in
+ mixed)
+ AC_MSG_RESULT([mixed (some dynamically, some statically)])
+ ;;
+ yes)
+ AC_MSG_RESULT([dynamically])
+ ;;
+ no)
+ AC_MSG_RESULT([statically])
+ ;;
+ *)
+ AC_MSG_RESULT([])
+ AC_MSG_ERROR([--enable-dynamic accepts only \`mixed', \`yes', or \`no'])
+ ;;
+esac
+# We use the actual results later, because we don't know yet
+# if --disable-shared or --disable-static was used.
+
+
+###############################################################################
+# Checks for programs.
+###############################################################################
+
+echo
+gl_POSIX_SHELL
+if test -z "$POSIX_SHELL" ; then
+ AC_MSG_ERROR([No POSIX conforming shell (sh) was found.])
+fi
+
+echo
+echo "Initializing Automake:"
+
+AM_INIT_AUTOMAKE([1.10 foreign tar-v7 filename-length-max=99])
+AC_PROG_LN_S
+
+AC_PROG_CC_C99
+if test x$ac_cv_prog_cc_c99 = xno ; then
+ AC_MSG_ERROR([No C99 compiler was found.])
+fi
+
+AM_PROG_CC_C_O
+AM_PROG_AS
+AC_USE_SYSTEM_EXTENSIONS
+
+if test "x$enable_threads" = xyes; then
+ echo
+ echo "Threading support:"
+ ACX_PTHREAD
+ LIBS="$LIBS $PTHREAD_LIBS"
+ AM_CFLAGS="$AM_CFLAGS $PTHREAD_CFLAGS"
+ CC="$PTHREAD_CC"
+fi
+
+echo
+echo "Initializing Libtool:"
+LT_PREREQ([2.2])
+LT_INIT([win32-dll])
+LT_LANG([Windows Resource])
+
+# This is a bit wrong since it is possible to request that only some libs
+# are built as shared. Using that feature isn't so common though, and this
+# breaks only on Windows (at least for now) if the user enables only some
+# libs as shared.
+AM_CONDITIONAL([COND_SHARED], [test "x$enable_shared" != xno])
+
+
+###############################################################################
+# Checks for libraries.
+###############################################################################
+
+echo
+echo "Initializing gettext:"
+AM_GNU_GETTEXT_VERSION([0.16.1])
+AM_GNU_GETTEXT([external])
+
+###############################################################################
+# Checks for header files.
+###############################################################################
+
+echo
+echo "System headers and functions:"
+
+# There is currently no workarounds in this package if some of
+# these headers are missing.
+AC_CHECK_HEADERS([fcntl.h limits.h sys/time.h],
+ [],
+ [AC_MSG_ERROR([Required header file(s) are missing.])])
+
+
+###############################################################################
+# Checks for typedefs, structures, and compiler characteristics.
+###############################################################################
+
+dnl We don't need these as long as we need a C99 compiler anyway.
+dnl AC_C_INLINE
+dnl AC_C_RESTRICT
+
+AC_HEADER_STDBOOL
+
+AC_TYPE_UINT8_T
+AC_TYPE_UINT16_T
+AC_TYPE_INT32_T
+AC_TYPE_UINT32_T
+AC_TYPE_INT64_T
+AC_TYPE_UINT64_T
+AC_TYPE_UINTPTR_T
+
+AC_CHECK_SIZEOF([size_t])
+
+# The command line tool can copy high resolution timestamps if such
+# information is availabe in struct stat. Otherwise one second accuracy
+# is used.
+AC_CHECK_MEMBERS([
+ struct stat.st_atim.tv_nsec,
+ struct stat.st_atimespec.tv_nsec,
+ struct stat.st_atimensec,
+ struct stat.st_uatime,
+ struct stat.st_atim.st__tim.tv_nsec])
+
+AC_SYS_LARGEFILE
+AC_C_BIGENDIAN
+
+
+###############################################################################
+# Checks for library functions.
+###############################################################################
+
+# Gnulib replacements as needed
+gl_GETOPT
+
+# Find the best function to set timestamps.
+AC_CHECK_FUNCS([futimens futimes futimesat utimes utime], [break])
+
+TUKLIB_PROGNAME
+TUKLIB_INTEGER
+TUKLIB_PHYSMEM
+TUKLIB_CPUCORES
+
+
+###############################################################################
+# If using GCC, set some additional AM_CFLAGS:
+###############################################################################
+
+if test "$GCC" = yes ; then
+ echo
+ echo "GCC extensions:"
+fi
+
+# Always do the visibility check but don't set AM_CFLAGS on Windows.
+# This way things get set properly even on Windows.
+gl_VISIBILITY
+if test -n "$CFLAG_VISIBILITY" && test "$is_w32" = no; then
+ AM_CFLAGS="$AM_CFLAGS $CFLAG_VISIBILITY"
+fi
+
+if test "$GCC" = yes ; then
+ # Enable as much warnings as possible. These commented warnings won't
+ # work for this package though:
+ # * -Wunreachable-code breaks several assert(0) cases, which are
+ # backed up with "return LZMA_PROG_ERROR".
+ # * -Wcast-qual would break various things where we need a non-const
+ # pointer although we don't modify anything through it.
+ # * -Wcast-align breaks optimized CRC32 and CRC64 implementation
+ # on some architectures (not on x86), where this warning is bogus,
+ # because we take care of correct alignment.
+ # * -Winline, -Wdisabled-optimization, -Wunsafe-loop-optimizations
+ # don't seem so useful here; at least the last one gives some
+ # warnings which are not bugs.
+ for NEW_FLAG in \
+ -Wall \
+ -Wextra \
+ -Wformat=2 \
+ -Winit-self \
+ -Wmissing-include-dirs \
+ -Wstrict-aliasing \
+ -Wfloat-equal \
+ -Wundef \
+ -Wshadow \
+ -Wpointer-arith \
+ -Wbad-function-cast \
+ -Wwrite-strings \
+ -Wlogical-op \
+ -Waggregate-return \
+ -Wstrict-prototypes \
+ -Wold-style-definition \
+ -Wmissing-prototypes \
+ -Wmissing-declarations \
+ -Wmissing-noreturn \
+ -Wredundant-decls
+ do
+ AC_MSG_CHECKING([if $CC accepts $NEW_FLAG])
+ OLD_CFLAGS="$CFLAGS"
+ CFLAGS="$CFLAGS $NEW_FLAG"
+ AC_COMPILE_IFELSE([void foo(void) { }], [
+ AM_CFLAGS="$AM_CFLAGS $NEW_FLAG"
+ AC_MSG_RESULT([yes])
+ ], [
+ AC_MSG_RESULT([no])
+ ])
+ CFLAGS="$OLD_CFLAGS"
+ done
+
+ AC_ARG_ENABLE([werror],
+ AC_HELP_STRING([--enable-werror], [Enable -Werror to abort
+ compilation on all compiler warnings.]),
+ [], [enable_werror=no])
+ if test "x$enable_werror" = "xyes"; then
+ AM_CFLAGS="$AM_CFLAGS -Werror"
+ fi
+fi
+
+
+###############################################################################
+# Create the makefiles and config.h
+###############################################################################
+
+echo
+
+# Don't build the lib directory at all if we don't need any replacement
+# functions.
+AM_CONDITIONAL([COND_GNULIB], test -n "$LIBOBJS")
+
+# Add default AM_CFLAGS.
+AC_SUBST([AM_CFLAGS])
+
+# Set additional flags for static/dynamic linking. The idea is that every
+# program (not library) being built will use either STATIC_{CPPFLAGS,LDFLAGS}
+# or DYNAMIC_{CPPFLAGS,LDFLAGS} depending on which type of linkage is
+# preferred. These preferences get overridden by use of --disable-static,
+# --disable-shared, or --enable-dynamic.
+#
+# This is quite messy, because we want to use LZMA_API_STATIC when linking
+# against static liblzma. It's needed on Windows.
+if test "x$enable_static" = xno; then
+ enable_dynamic=yes
+fi
+if test "x$enable_shared" = xno; then
+ enable_dynamic=no
+fi
+case $enable_dynamic in
+ yes)
+ STATIC_CPPFLAGS=
+ STATIC_LDFLAGS=
+ DYNAMIC_CPPFLAGS=
+ DYNAMIC_LDFLAGS=
+ ;;
+ mixed)
+ STATIC_CPPFLAGS="-DLZMA_API_STATIC"
+ STATIC_LDFLAGS="-static"
+ DYNAMIC_CPPFLAGS=
+ DYNAMIC_LDFLAGS=
+ ;;
+ no)
+ STATIC_CPPFLAGS="-DLZMA_API_STATIC"
+ STATIC_LDFLAGS="-static"
+ DYNAMIC_CPPFLAGS="-DLZMA_API_STATIC"
+ DYNAMIC_LDFLAGS="-static"
+ ;;
+esac
+AC_SUBST([STATIC_CPPFLAGS])
+AC_SUBST([STATIC_LDFLAGS])
+AC_SUBST([DYNAMIC_CPPFLAGS])
+AC_SUBST([DYNAMIC_LDFLAGS])
+
+# This is needed for src/scripts.
+xz=`echo xz | sed "$program_transform_name"`
+AC_SUBST([xz])
+
+AC_CONFIG_FILES([
+ Doxyfile
+ Makefile
+ po/Makefile.in
+ lib/Makefile
+ src/Makefile
+ src/liblzma/liblzma.pc
+ src/liblzma/Makefile
+ src/liblzma/api/Makefile
+ src/xz/Makefile
+ src/xzdec/Makefile
+ src/lzmainfo/Makefile
+ src/scripts/Makefile
+ src/scripts/xzdiff
+ src/scripts/xzgrep
+ src/scripts/xzmore
+ src/scripts/xzless
+ tests/Makefile
+ debug/Makefile
+])
+
+AC_OUTPUT
+
+# Some warnings
+if test x$tuklib_cv_physmem_method = xunknown; then
+ echo
+ echo "WARNING:"
+ echo "No supported method to detect the amount of RAM."
+ echo "Consider using --enable-assume-ram (if you didn't already)"
+ echo "or make a patch to add support for this operating system."
+fi
+
+# Not threading yet so don't warn.
+#if test x$tuklib_cv_cpucores_method = xunknown; then
+# echo
+# echo "WARNING:"
+# echo "No supported method to detect the number of CPU cores."
+#fi
diff --git a/debug/Makefile.am b/debug/Makefile.am
new file mode 100644
index 0000000..23834f4
--- /dev/null
+++ b/debug/Makefile.am
@@ -0,0 +1,30 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+noinst_PROGRAMS = \
+ repeat \
+ sync_flush \
+ full_flush \
+ memusage \
+ crc32 \
+ known_sizes \
+ hex2bin
+
+AM_CPPFLAGS = \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ $(STATIC_CPPFLAGS)
+
+AM_LDFLAGS = $(STATIC_LDFLAGS)
+
+LDADD = $(top_builddir)/src/liblzma/liblzma.la
+
+if COND_GNULIB
+LDADD += $(top_builddir)/lib/libgnu.a
+endif
+
+LDADD += $(LTLIBINTL)
diff --git a/debug/README b/debug/README
new file mode 100644
index 0000000..749610d
--- /dev/null
+++ b/debug/README
@@ -0,0 +1,17 @@
+
+Debug tools
+-----------
+
+ This directory contains a few tiny programs that may be helpful when
+ debugging LZMA Utils.
+
+ These tools are not meant to be installed. Often one needs to edit
+ the source code a little to make the programs do the wanted things.
+ If you don't know how these programs could help you, it is likely
+ that they really are useless to you.
+
+ These aren't intended to be used as example programs. They take some
+ shortcuts here and there, which correct programs should not do. Many
+ possible errors (especially I/O errors) are ignored. Don't report
+ bugs or send patches to fix this kind of bugs.
+
diff --git a/debug/crc32.c b/debug/crc32.c
new file mode 100644
index 0000000..e545a3c
--- /dev/null
+++ b/debug/crc32.c
@@ -0,0 +1,39 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc32.c
+/// \brief Primitive CRC32 calculation tool
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "lzma.h"
+#include <stdio.h>
+
+
+int
+main(void)
+{
+ uint32_t crc = 0;
+
+ do {
+ uint8_t buf[BUFSIZ];
+ const size_t size = fread(buf, 1, sizeof(buf), stdin);
+ crc = lzma_crc32(buf, size, crc);
+ } while (!ferror(stdin) && !feof(stdin));
+
+ //printf("%08" PRIX32 "\n", crc);
+
+ // I want it little endian so it's easy to work with hex editor.
+ printf("%02" PRIX32 " ", crc & 0xFF);
+ printf("%02" PRIX32 " ", (crc >> 8) & 0xFF);
+ printf("%02" PRIX32 " ", (crc >> 16) & 0xFF);
+ printf("%02" PRIX32 " ", crc >> 24);
+ printf("\n");
+
+ return 0;
+}
diff --git a/debug/full_flush.c b/debug/full_flush.c
new file mode 100644
index 0000000..67442b1
--- /dev/null
+++ b/debug/full_flush.c
@@ -0,0 +1,102 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file full_flush.c
+/// \brief Encode files using LZMA_FULL_FLUSH
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "lzma.h"
+#include <stdio.h>
+
+
+static lzma_stream strm = LZMA_STREAM_INIT;
+static FILE *file_in;
+
+
+static void
+encode(size_t size, lzma_action action)
+{
+ static const size_t CHUNK = 64;
+ uint8_t in[CHUNK];
+ uint8_t out[CHUNK];
+ lzma_ret ret;
+
+ do {
+ if (strm.avail_in == 0 && size > 0) {
+ const size_t amount = my_min(size, CHUNK);
+ strm.avail_in = fread(in, 1, amount, file_in);
+ strm.next_in = in;
+ size -= amount; // Intentionally not using avail_in.
+ }
+
+ strm.next_out = out;
+ strm.avail_out = CHUNK;
+
+ ret = lzma_code(&strm, size == 0 ? action : LZMA_RUN);
+
+ if (ret != LZMA_OK && ret != LZMA_STREAM_END) {
+ fprintf(stderr, "%s:%u: %s: ret == %d\n",
+ __FILE__, __LINE__, __func__, ret);
+ exit(1);
+ }
+
+ fwrite(out, 1, CHUNK - strm.avail_out, stdout);
+
+ } while (size > 0 || strm.avail_out == 0);
+
+ if ((action == LZMA_RUN && ret != LZMA_OK)
+ || (action != LZMA_RUN && ret != LZMA_STREAM_END)) {
+ fprintf(stderr, "%s:%u: %s: ret == %d\n",
+ __FILE__, __LINE__, __func__, ret);
+ exit(1);
+ }
+}
+
+
+int
+main(int argc, char **argv)
+{
+ file_in = argc > 1 ? fopen(argv[1], "rb") : stdin;
+
+
+ // Config
+ lzma_options_lzma opt_lzma;
+ if (lzma_lzma_preset(&opt_lzma, 1)) {
+ fprintf(stderr, "preset failed\n");
+ exit(1);
+ }
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+ filters[0].id = LZMA_FILTER_LZMA2;
+ filters[0].options = &opt_lzma;
+ filters[1].id = LZMA_VLI_UNKNOWN;
+
+ // Init
+ if (lzma_stream_encoder(&strm, filters, LZMA_CHECK_CRC32) != LZMA_OK) {
+ fprintf(stderr, "init failed\n");
+ exit(1);
+ }
+
+// if (lzma_easy_encoder(&strm, 1)) {
+// fprintf(stderr, "init failed\n");
+// exit(1);
+// }
+
+ // Encoding
+ encode(0, LZMA_FULL_FLUSH);
+ encode(6, LZMA_FULL_FLUSH);
+ encode(0, LZMA_FULL_FLUSH);
+ encode(7, LZMA_FULL_FLUSH);
+ encode(0, LZMA_FULL_FLUSH);
+ encode(0, LZMA_FINISH);
+
+ // Clean up
+ lzma_end(&strm);
+
+ return 0;
+}
diff --git a/debug/hex2bin.c b/debug/hex2bin.c
new file mode 100644
index 0000000..7324624
--- /dev/null
+++ b/debug/hex2bin.c
@@ -0,0 +1,53 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file hex2bin.c
+/// \brief Converts hexadecimal input strings to binary
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include <stdio.h>
+#include <ctype.h>
+
+
+static int
+getbin(int x)
+{
+ if (x >= '0' && x <= '9')
+ return x - '0';
+
+ if (x >= 'A' && x <= 'F')
+ return x - 'A' + 10;
+
+ return x - 'a' + 10;
+}
+
+
+int
+main(void)
+{
+ while (true) {
+ int byte = getchar();
+ if (byte == EOF)
+ return 0;
+ if (!isxdigit(byte))
+ continue;
+
+ const int digit = getchar();
+ if (digit == EOF || !isxdigit(digit)) {
+ fprintf(stderr, "Invalid input\n");
+ return 1;
+ }
+
+ byte = (getbin(byte) << 4) | getbin(digit);
+ if (putchar(byte) == EOF) {
+ perror(NULL);
+ return 1;
+ }
+ }
+}
diff --git a/debug/known_sizes.c b/debug/known_sizes.c
new file mode 100644
index 0000000..6c3641f
--- /dev/null
+++ b/debug/known_sizes.c
@@ -0,0 +1,129 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file known_sizes.c
+/// \brief Encodes .lzma Stream with sizes known in Block Header
+///
+/// The input file is encoded in RAM, and the known Compressed Size
+/// and/or Uncompressed Size values are stored in the Block Header.
+/// As of writing there's no such Stream encoder in liblzma.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "lzma.h"
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/unistd.h>
+#include <stdio.h>
+
+
+// Support file sizes up to 1 MiB. We use this for output space too, so files
+// close to 1 MiB had better compress at least a little or we have a buffer
+// overflow.
+#define BUFFER_SIZE (1U << 20)
+
+
+int
+main(void)
+{
+ // Allocate the buffers.
+ uint8_t *in = malloc(BUFFER_SIZE);
+ uint8_t *out = malloc(BUFFER_SIZE);
+ if (in == NULL || out == NULL)
+ return 1;
+
+ // Fill the input buffer.
+ const size_t in_size = fread(in, 1, BUFFER_SIZE, stdin);
+
+ // Filter setup
+ lzma_options_lzma opt_lzma;
+ if (lzma_lzma_preset(&opt_lzma, 1))
+ return 1;
+
+ lzma_filter filters[] = {
+ {
+ .id = LZMA_FILTER_LZMA2,
+ .options = &opt_lzma
+ },
+ {
+ .id = LZMA_VLI_UNKNOWN
+ }
+ };
+
+ lzma_block block = {
+ .check = LZMA_CHECK_CRC32,
+ .compressed_size = BUFFER_SIZE, // Worst case reserve
+ .uncompressed_size = in_size,
+ .filters = filters,
+ };
+
+ lzma_stream strm = LZMA_STREAM_INIT;
+ if (lzma_block_encoder(&strm, &block) != LZMA_OK)
+ return 1;
+
+ // Reserve space for Stream Header and Block Header. We need to
+ // calculate the size of the Block Header first.
+ if (lzma_block_header_size(&block) != LZMA_OK)
+ return 1;
+
+ size_t out_size = LZMA_STREAM_HEADER_SIZE + block.header_size;
+
+ strm.next_in = in;
+ strm.avail_in = in_size;
+ strm.next_out = out + out_size;
+ strm.avail_out = BUFFER_SIZE - out_size;
+
+ if (lzma_code(&strm, LZMA_FINISH) != LZMA_STREAM_END)
+ return 1;
+
+ out_size += strm.total_out;
+
+ if (lzma_block_header_encode(&block, out + LZMA_STREAM_HEADER_SIZE)
+ != LZMA_OK)
+ return 1;
+
+ lzma_index *idx = lzma_index_init(NULL);
+ if (idx == NULL)
+ return 1;
+
+ if (lzma_index_append(idx, NULL, block.header_size + strm.total_out,
+ strm.total_in) != LZMA_OK)
+ return 1;
+
+ if (lzma_index_encoder(&strm, idx) != LZMA_OK)
+ return 1;
+
+ if (lzma_code(&strm, LZMA_RUN) != LZMA_STREAM_END)
+ return 1;
+
+ out_size += strm.total_out;
+
+ lzma_end(&strm);
+
+ lzma_index_end(idx, NULL);
+
+ // Encode the Stream Header and Stream Footer. backwards_size is
+ // needed only for the Stream Footer.
+ lzma_stream_flags sf = {
+ .backward_size = strm.total_out,
+ .check = block.check,
+ };
+
+ if (lzma_stream_header_encode(&sf, out) != LZMA_OK)
+ return 1;
+
+ if (lzma_stream_footer_encode(&sf, out + out_size) != LZMA_OK)
+ return 1;
+
+ out_size += LZMA_STREAM_HEADER_SIZE;
+
+ // Write out the file.
+ fwrite(out, 1, out_size, stdout);
+
+ return 0;
+}
diff --git a/debug/memusage.c b/debug/memusage.c
new file mode 100644
index 0000000..bd72b5a
--- /dev/null
+++ b/debug/memusage.c
@@ -0,0 +1,51 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file memusage.c
+/// \brief Calculates memory usage using lzma_memory_usage()
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "lzma.h"
+#include <stdio.h>
+
+int
+main(void)
+{
+ lzma_options_lzma lzma = {
+ .dict_size = (1U << 30) + (1U << 29),
+ .lc = 3,
+ .lp = 0,
+ .pb = 2,
+ .preset_dict = NULL,
+ .preset_dict_size = 0,
+ .mode = LZMA_MODE_NORMAL,
+ .nice_len = 48,
+ .mf = LZMA_MF_BT4,
+ .depth = 0,
+ };
+
+/*
+ lzma_options_filter filters[] = {
+ { LZMA_FILTER_LZMA1,
+ (lzma_options_lzma *)&lzma_preset_lzma[6 - 1] },
+ { UINT64_MAX, NULL }
+ };
+*/
+ lzma_filter filters[] = {
+ { LZMA_FILTER_LZMA1, &lzma },
+ { UINT64_MAX, NULL }
+ };
+
+ printf("Encoder: %10" PRIu64 " B\n",
+ lzma_raw_encoder_memusage(filters));
+ printf("Decoder: %10" PRIu64 " B\n",
+ lzma_raw_decoder_memusage(filters));
+
+ return 0;
+}
diff --git a/debug/repeat.c b/debug/repeat.c
new file mode 100644
index 0000000..a00bde2
--- /dev/null
+++ b/debug/repeat.c
@@ -0,0 +1,36 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file repeat.c
+/// \brief Repeats given string given times
+///
+/// This program can be useful when debugging run-length encoder in
+/// the Subblock filter, especially the condition when repeat count
+/// doesn't fit into 28-bit integer.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include <stdio.h>
+
+
+int
+main(int argc, char **argv)
+{
+ if (argc != 3) {
+ fprintf(stderr, "Usage: %s COUNT STRING\n", argv[0]);
+ exit(1);
+ }
+
+ unsigned long long count = strtoull(argv[1], NULL, 10);
+ const size_t size = strlen(argv[2]);
+
+ while (count-- != 0)
+ fwrite(argv[2], 1, size, stdout);
+
+ return !!(ferror(stdout) || fclose(stdout));
+}
diff --git a/debug/sync_flush.c b/debug/sync_flush.c
new file mode 100644
index 0000000..fc2ebcb
--- /dev/null
+++ b/debug/sync_flush.c
@@ -0,0 +1,124 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file sync_flush.c
+/// \brief Encode files using LZMA_SYNC_FLUSH
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "lzma.h"
+#include <stdio.h>
+
+
+static lzma_stream strm = LZMA_STREAM_INIT;
+static FILE *file_in;
+
+
+static void
+encode(size_t size, lzma_action action)
+{
+ static const size_t CHUNK = 64;
+ uint8_t in[CHUNK];
+ uint8_t out[CHUNK];
+ lzma_ret ret;
+
+ do {
+ if (strm.avail_in == 0 && size > 0) {
+ const size_t amount = my_min(size, CHUNK);
+ strm.avail_in = fread(in, 1, amount, file_in);
+ strm.next_in = in;
+ size -= amount; // Intentionally not using avail_in.
+ }
+
+ strm.next_out = out;
+ strm.avail_out = CHUNK;
+
+ ret = lzma_code(&strm, size == 0 ? action : LZMA_RUN);
+
+ if (ret != LZMA_OK && ret != LZMA_STREAM_END) {
+ fprintf(stderr, "%s:%u: %s: ret == %d\n",
+ __FILE__, __LINE__, __func__, ret);
+ exit(1);
+ }
+
+ fwrite(out, 1, CHUNK - strm.avail_out, stdout);
+
+ } while (size > 0 || strm.avail_out == 0);
+
+ if ((action == LZMA_RUN && ret != LZMA_OK)
+ || (action != LZMA_RUN && ret != LZMA_STREAM_END)) {
+ fprintf(stderr, "%s:%u: %s: ret == %d\n",
+ __FILE__, __LINE__, __func__, ret);
+ exit(1);
+ }
+}
+
+
+int
+main(int argc, char **argv)
+{
+ file_in = argc > 1 ? fopen(argv[1], "rb") : stdin;
+
+ // Config
+ lzma_options_lzma opt_lzma = {
+ .dict_size = 1U << 16,
+ .lc = LZMA_LC_DEFAULT,
+ .lp = LZMA_LP_DEFAULT,
+ .pb = LZMA_PB_DEFAULT,
+ .preset_dict = NULL,
+ .mode = LZMA_MODE_NORMAL,
+ .nice_len = 32,
+ .mf = LZMA_MF_HC3,
+ .depth = 0,
+ };
+
+ lzma_options_delta opt_delta = {
+ .dist = 16
+ };
+
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+ filters[0].id = LZMA_FILTER_LZMA2;
+ filters[0].options = &opt_lzma;
+ filters[1].id = LZMA_VLI_UNKNOWN;
+
+ // Init
+ if (lzma_stream_encoder(&strm, filters, LZMA_CHECK_CRC32) != LZMA_OK) {
+ fprintf(stderr, "init failed\n");
+ exit(1);
+ }
+
+ // Encoding
+ encode(0, LZMA_SYNC_FLUSH);
+ encode(6, LZMA_SYNC_FLUSH);
+ encode(0, LZMA_SYNC_FLUSH);
+ encode(7, LZMA_SYNC_FLUSH);
+ encode(0, LZMA_SYNC_FLUSH);
+ encode(0, LZMA_FINISH);
+
+/*
+ encode(53, LZMA_SYNC_FLUSH);
+ opt_lzma.lc = 2;
+ opt_lzma.lp = 1;
+ opt_lzma.pb = 0;
+ if (lzma_filters_update(&strm, filters) != LZMA_OK) {
+ fprintf(stderr, "update failed\n");
+ exit(1);
+ }
+ encode(404, LZMA_FINISH);
+*/
+
+ // Clean up
+ lzma_end(&strm);
+
+ return 0;
+
+ // Prevent useless warnings so we don't need to have special CFLAGS
+ // to disable -Werror.
+ (void)opt_lzma;
+ (void)opt_delta;
+}
diff --git a/doc/faq.txt b/doc/faq.txt
new file mode 100644
index 0000000..8e87457
--- /dev/null
+++ b/doc/faq.txt
@@ -0,0 +1,128 @@
+
+XZ Utils FAQ
+============
+
+Q: What do the letters XZ mean?
+
+A: Nothing. They are just two letters, which come from the file format
+ suffix .xz. The .xz suffix was selected, because it seemed to be
+ pretty much unused. It is no deeper meaning.
+
+
+Q: What are LZMA and LZMA2?
+
+A: LZMA stands for Lempel-Ziv-Markov chain-Algorithm. It is the name
+ of the compression algorithm designed by Igor Pavlov for 7-Zip.
+ LZMA is based on LZ77 and range encoding.
+
+ LZMA2 is an updated version of the original LZMA to fix a couple of
+ practical issues. In context of XZ Utils, LZMA is called LZMA1 to
+ emphasize that LZMA is not the same thing as LZMA2. LZMA2 is the
+ primary compression algorithm in the .xz file format.
+
+
+Q: There are many LZMA related projects. How does XZ Utils relate to them?
+
+A: 7-Zip and LZMA SDK are the original projects. LZMA SDK is roughly
+ a subset of the 7-Zip source tree.
+
+ p7zip is 7-Zip's command line tools ported to POSIX-like systems.
+
+ LZMA Utils provide a gzip-like lzma tool for POSIX-like systems.
+ LZMA Utils are based on LZMA SDK. XZ Utils are the successor to
+ LZMA Utils.
+
+ There are several other projects using LZMA. Most are more or less
+ based on LZMA SDK.
+
+
+Q: Do XZ Utils support the .7z format?
+
+A: No. Use 7-Zip (Windows) or p7zip (POSIX-like systems) to handle .7z
+ files.
+
+
+Q: I have many .tar.7z files. Can I convert them to .tar.xz without
+ spending hours recompressing the data?
+
+A: In the "extra" directory, there is a script named 7z2lzma.bash which
+ is able to convert some .7z files to the .lzma format (not .xz). It
+ needs the 7za (or 7z) command from p7zip. The script may silently
+ produce corrupt output if certain assumptions are not met, so
+ decompress the resulting .lzma file and compare it against the
+ original before deleting the original file!
+
+
+Q: I have many .lzma files. Can I quickly convert them to the .xz format?
+
+A: For now, no. Since XZ Utils supports the .lzma format, it's usually
+ not too bad to keep the old files in the old format. If you want to
+ do the conversion anyway, you need to decompress the .lzma files and
+ then recompress to the .xz format.
+
+ Technically, there is a way to make the conversion relatively fast
+ (roughly twice the time that normal decompression takes). Writing
+ such a tool would take quite a bit time though, and would probably
+ be useful to only a few people. If you really want such a conversion
+ tool, contact Lasse Collin and offer some money.
+
+
+Q: I have installed xz, but my tar doesn't recognize .tar.xz files.
+ How can I extract .tar.xz files?
+
+A: xz -dc foo.tar.xz | tar xf -
+
+
+Q: Can I recover parts of a broken .xz file (e.g. corrupted CD-R)?
+
+A: It may be possible if the file consists of multiple blocks, which
+ typically is not the case if the file was created in single-threaded
+ mode. There is no recovery program yet.
+
+
+Q: Is (some part of) XZ Utils patented?
+
+A: Lasse Collin is not aware of any patents that could affect XZ Utils.
+ However, due to nature of software patents, it's not possible to
+ guarantee that XZ Utils isn't affected by any third party patent(s).
+
+
+Q: Where can I find documentation about the file format and algorithms?
+
+A: The .xz format is documented in xz-file-format.txt. It is a container
+ format only, and doesn't include descriptions of any non-trivial
+ filters.
+
+ Documenting LZMA and LZMA2 is planned, but for now, there is no other
+ documentation that the source code. Before you begin, you should know
+ the basics of LZ77 and range coding algorithms. LZMA is based on LZ77,
+ but LZMA is *a lot* more complex. Range coding is used to compress
+ the final bitstream like Huffman coding is used in Deflate.
+
+
+Q: I cannot find BCJ and BCJ2 filters. Don't they exist in liblzma?
+
+A: BCJ filter is called "x86" in liblzma. BCJ2 is not included,
+ because it requires using more than one encoded output stream.
+
+
+Q: How do I build a program that needs liblzmadec (lzmadec.h)?
+
+A: liblzmadec is part of LZMA Utils. XZ Utils has liblzma, but no
+ liblzmadec. The code using liblzmadec should be ported to use
+ liblzma instead. If you cannot or don't want to do that, download
+ LZMA Utils from <http://tukaani.org/lzma/>.
+
+
+Q: The default build of liblzma is too big. How can I make it smaller?
+
+A: Give --enable-small to the configure script. Use also appropriate
+ --enable or --disable options to include only those filter encoders
+ and decoders and integrity checks that you actually need. Use
+ CFLAGS=-Os (with GCC) or equivalent to tell your compiler to optimize
+ for size. See INSTALL for information about configure options.
+
+ If the result is still too big, take a look at XZ Embedded. It is
+ a separate project, which provides a limited but significantly
+ smaller XZ decoder implementation than XZ Utils.
+
diff --git a/doc/history.txt b/doc/history.txt
new file mode 100644
index 0000000..c97492e
--- /dev/null
+++ b/doc/history.txt
@@ -0,0 +1,149 @@
+
+History of LZMA Utils and XZ Utils
+==================================
+
+Tukaani distribution
+
+ In 2005, there was a small group working on Tukaani distribution, which
+ was a Slackware fork. One of the project goals was to fit the distro on
+ a single 700 MiB ISO-9660 image. Using LZMA instead of gzip helped a
+ lot. Roughly speaking, one could fit data that took 1000 MiB in gzipped
+ form into 700 MiB with LZMA. Naturally compression ratio varied across
+ packages, but this was what we got on average.
+
+ Slackware packages have traditionally had .tgz as the filename suffix,
+ which is an abbreviation of .tar.gz. A logical naming for LZMA
+ compressed packages was .tlz, being an abbreviation of .tar.lzma.
+
+ At the end of the year 2007, there was no distribution under the
+ Tukaani project anymore, but development of LZMA Utils was kept going.
+ Still, there were .tlz packages around, because at least Vector Linux
+ (a Slackware based distribution) used LZMA for its packages.
+
+ First versions of the modified pkgtools used the LZMA_Alone tool from
+ Igor Pavlov's LZMA SDK as is. It was fine, because users wouldn't need
+ to interact with LZMA_Alone directly. But people soon wanted to use
+ LZMA for other files too, and the interface of LZMA_Alone wasn't
+ comfortable for those used to gzip and bzip2.
+
+
+First steps of LZMA Utils
+
+ The first version of LZMA Utils (4.22.0) included a shell script called
+ lzmash. It was wrapper that had gzip-like command line interface. It
+ used the LZMA_Alone tool from LZMA SDK to do all the real work. zgrep,
+ zdiff, and related scripts from gzip were adapted work with LZMA and
+ were part of the first LZMA Utils release too.
+
+ LZMA Utils 4.22.0 included also lzmadec, which was a small (less than
+ 10 KiB) decoder-only command line tool. It was written on top of the
+ decoder-only C code found from the LZMA SDK. lzmadec was convenient in
+ situations where LZMA_Alone (a few hundred KiB) would be too big.
+
+ lzmash and lzmadec were written by Lasse Collin.
+
+
+Second generation
+
+ The lzmash script was an ugly and not very secure hack. The last
+ version of LZMA Utils to use lzmash was 4.27.1.
+
+ LZMA Utils 4.32.0beta1 introduced a new lzma command line tool written
+ by Ville Koskinen. It was written in C++, and used the encoder and
+ decoder from C++ LZMA SDK with little modifications. This tool replaced
+ both the lzmash script and the LZMA_Alone command line tool in LZMA
+ Utils.
+
+ Introducing this new tool caused some temporary incompatibilities,
+ because LZMA_Alone executable was simply named lzma like the new
+ command line tool, but they had completely different command line
+ interface. The file format was still the same.
+
+ Lasse wrote liblzmadec, which was a small decoder-only library based
+ on the C code found from LZMA SDK. liblzmadec had API similar to zlib,
+ although there were some significant differences, which made it
+ non-trivial to use it in some applications designed for zlib and
+ libbzip2.
+
+ The lzmadec command line tool was converted to use liblzmadec.
+
+ Alexandre Sauvé helped converting build system to use GNU Autotools.
+ This made is easier to test for certain less portable features needed
+ by the new command line tool.
+
+ Since the new command line tool never got completely finished (for
+ example, it didn't support LZMA_OPT environment variable), the intent
+ was to not call 4.32.x stable. Similarly, liblzmadec wasn't polished,
+ but appeared to work well enough, so some people started using it too.
+
+ Because the development of the third generation of LZMA Utils was
+ delayed considerably (3-4 years), the 4.32.x branch had to be kept
+ maintained. It got some bug fixes now and then, and finally it was
+ decided to call it stable, although most of the missing features were
+ never added.
+
+
+File format problems
+
+ The file format used by LZMA_Alone was primitive. It was designed for
+ embedded systems in mind, and thus provided only minimal set of
+ features. The two biggest problems for non-embedded use were lack of
+ magic bytes and integrity check.
+
+ Igor and Lasse started developing a new file format with some help
+ from Ville Koskinen. Also Mark Adler, Mikko Pouru, H. Peter Anvin,
+ and Lars Wirzenius helped with some minor things at some point of the
+ development. Designing the new format took quite a long time (actually,
+ too long time would be more appropriate expression). It was mostly
+ because Lasse was quite slow at getting things done due to personal
+ reasons.
+
+ Originally the new format was supposed to use the same .lzma suffix
+ that was already used by the old file format. Switching to the new
+ format wouldn't have caused much trouble when the old format wasn't
+ used by many people. But since the development of the new format took
+ so long time, the old format got quite popular, and it was decided
+ that the new file format must use a different suffix.
+
+ It was decided to use .xz as the suffix of the new file format. The
+ first stable .xz file format specification was finally released in
+ December 2008. In addition to fixing the most obvious problems of
+ the old .lzma format, the .xz format added some new features like
+ support for multiple filters (compression algorithms), filter chaining
+ (like piping on the command line), and limited random-access reading.
+
+ Currently the primary compression algorithm used in .xz is LZMA2.
+ It is an extension on top of the original LZMA to fix some practical
+ problems: LZMA2 adds support for flushing the encoder, uncompressed
+ chunks, eases stateful decoder implementations, and improves support
+ for multithreading. Since LZMA2 is better than the original LZMA, the
+ original LZMA is not supported in .xz.
+
+
+Transition to XZ Utils
+
+ The early versions of XZ Utils were called LZMA Utils. The first
+ releases were 4.42.0alphas. They dropped the rest of the C++ LZMA SDK.
+ The code was still directly based on LZMA SDK but ported to C and
+ converted from callback API to stateful API. Later, Igor Pavlov made
+ C version of the LZMA encoder too; these ports from C++ to C were
+ independent in LZMA SDK and LZMA Utils.
+
+ The core of the new LZMA Utils was liblzma, a compression library with
+ zlib-like API. liblzma supported both the old and new file format. The
+ gzip-like lzma command line tool was rewritten to use liblzma.
+
+ The new LZMA Utils code base was renamed to XZ Utils when the name
+ of the new file format had been decided. The liblzma compression
+ library retained its name though, because changing it would have
+ caused unnecessary breakage in applications already using the early
+ liblzma snapshots.
+
+ The xz command line tool can emulate the gzip-like lzma tool by
+ creating appropriate symlinks (e.g. lzma -> xz). Thus, practically
+ all scripts using the lzma tool from LZMA Utils will work as is with
+ XZ Utils (and will keep using the old .lzma format). Still, the .lzma
+ format is more or less deprecated. XZ Utils will keep supporting it,
+ but new applications should use the .xz format, and migrating old
+ applications to .xz is often a good idea too.
+
diff --git a/doc/lzma-file-format.txt b/doc/lzma-file-format.txt
new file mode 100644
index 0000000..21fcb19
--- /dev/null
+++ b/doc/lzma-file-format.txt
@@ -0,0 +1,166 @@
+
+The .lzma File Format
+=====================
+
+ 0. Preface
+ 0.1. Notices and Acknowledgements
+ 0.2. Changes
+ 1. File Format
+ 1.1. Header
+ 1.1.1. Properties
+ 1.1.2. Dictionary Size
+ 1.1.3. Uncompressed Size
+ 1.2. LZMA Compressed Data
+ 2. References
+
+
+0. Preface
+
+ This document describes the .lzma file format, which is
+ sometimes also called LZMA_Alone format. It is a legacy file
+ format, which is being or has been replaced by the .xz format.
+ The MIME type of the .lzma format is `application/x-lzma'.
+
+ The most commonly used software to handle .lzma files are
+ LZMA SDK, LZMA Utils, 7-Zip, and XZ Utils. This document
+ describes some of the differences between these implementations
+ and gives hints what subset of the .lzma format is the most
+ portable.
+
+
+0.1. Notices and Acknowledgements
+
+ This file format was designed by Igor Pavlov for use in
+ LZMA SDK. This document was written by Lasse Collin
+ <lasse.collin@tukaani.org> using the documentation found
+ from the LZMA SDK.
+
+ This document has been put into the public domain.
+
+
+0.2. Changes
+
+ Last modified: 2009-05-01 11:15+0300
+
+
+1. File Format
+
+ +-+-+-+-+-+-+-+-+-+-+-+-+-+==========================+
+ | Header | LZMA Compressed Data |
+ +-+-+-+-+-+-+-+-+-+-+-+-+-+==========================+
+
+ The .lzma format file consist of 13-byte Header followed by
+ the LZMA Compressed Data.
+
+ Unlike the .gz, .bz2, and .xz formats, it is not possible to
+ concatenate multiple .lzma files as is and expect the
+ decompression tool to decode the resulting file as if it were
+ a single .lzma file.
+
+ For example, the command line tools from LZMA Utils and
+ LZMA SDK silently ignore all the data after the first .lzma
+ stream. In contrast, the command line tool from XZ Utils
+ considers the .lzma file to be corrupt if there is data after
+ the first .lzma stream.
+
+
+1.1. Header
+
+ +------------+----+----+----+----+--+--+--+--+--+--+--+--+
+ | Properties | Dictionary Size | Uncompressed Size |
+ +------------+----+----+----+----+--+--+--+--+--+--+--+--+
+
+
+1.1.1. Properties
+
+ The Properties field contains three properties. An abbreviation
+ is given in parentheses, followed by the value range of the
+ property. The field consists of
+
+ 1) the number of literal context bits (lc, [0, 8]);
+ 2) the number of literal position bits (lp, [0, 4]); and
+ 3) the number of position bits (pb, [0, 4]).
+
+ The properties are encoded using the following formula:
+
+ Properties = (pb * 5 + lp) * 9 + lc
+
+ The following C code illustrates a straightforward way to
+ decode the Properties field:
+
+ uint8_t lc, lp, pb;
+ uint8_t prop = get_lzma_properties();
+ if (prop > (4 * 5 + 4) * 9 + 8)
+ return LZMA_PROPERTIES_ERROR;
+
+ pb = prop / (9 * 5);
+ prop -= pb * 9 * 5;
+ lp = prop / 9;
+ lc = prop - lp * 9;
+
+ XZ Utils has an additional requirement: lc + lp <= 4. Files
+ which don't follow this requirement cannot be decompressed
+ with XZ Utils. Usually this isn't a problem since the most
+ common lc/lp/pb values are 3/0/2. It is the only lc/lp/pb
+ combination that the files created by LZMA Utils can have,
+ but LZMA Utils can decompress files with any lc/lp/pb.
+
+
+1.1.2. Dictionary Size
+
+ Dictionary Size is stored as an unsigned 32-bit little endian
+ integer. Any 32-bit value is possible, but for maximum
+ portability, only sizes of 2^n and 2^n + 2^(n-1) should be
+ used.
+
+ LZMA Utils creates only files with dictionary size 2^n,
+ 16 <= n <= 25. LZMA Utils can decompress files with any
+ dictionary size.
+
+ XZ Utils creates and decompresses .lzma files only with
+ dictionary sizes 2^n and 2^n + 2^(n-1). If some other
+ dictionary size is specified when compressing, the value
+ stored in the Dictionary Size field is a rounded up, but the
+ specified value is still used in the actual compression code.
+
+
+1.1.3. Uncompressed Size
+
+ Uncompressed Size is stored as unsigned 64-bit little endian
+ integer. A special value of 0xFFFF_FFFF_FFFF_FFFF indicates
+ that Uncompressed Size is unknown. End of Payload Marker (*)
+ is used if and only if Uncompressed Size is unknown.
+
+ XZ Utils rejects files whose Uncompressed Size field specifies
+ a known size that is 256 GiB or more. This is to reject false
+ positives when trying to guess if the input file is in the
+ .lzma format. When Uncompressed Size is unknown, there is no
+ limit for the uncompressed size of the file.
+
+ (*) Some tools use the term End of Stream (EOS) marker
+ instead of End of Payload Marker.
+
+
+1.2. LZMA Compressed Data
+
+ Detailed description of the format of this field is out of
+ scope of this document.
+
+
+2. References
+
+ LZMA SDK - The original LZMA implementation
+ http://7-zip.org/sdk.html
+
+ 7-Zip
+ http://7-zip.org/
+
+ LZMA Utils - LZMA adapted to POSIX-like systems
+ http://tukaani.org/lzma/
+
+ XZ Utils - The next generation of LZMA Utils
+ http://tukaani.org/xz/
+
+ The .xz file format - The successor of the the .lzma format
+ http://tukaani.org/xz/xz-file-format.txt
+
diff --git a/doc/xz-file-format.txt b/doc/xz-file-format.txt
new file mode 100644
index 0000000..4ed6650
--- /dev/null
+++ b/doc/xz-file-format.txt
@@ -0,0 +1,1150 @@
+
+The .xz File Format
+===================
+
+Version 1.0.4 (2009-08-27)
+
+
+ 0. Preface
+ 0.1. Notices and Acknowledgements
+ 0.2. Getting the Latest Version
+ 0.3. Version History
+ 1. Conventions
+ 1.1. Byte and Its Representation
+ 1.2. Multibyte Integers
+ 2. Overall Structure of .xz File
+ 2.1. Stream
+ 2.1.1. Stream Header
+ 2.1.1.1. Header Magic Bytes
+ 2.1.1.2. Stream Flags
+ 2.1.1.3. CRC32
+ 2.1.2. Stream Footer
+ 2.1.2.1. CRC32
+ 2.1.2.2. Backward Size
+ 2.1.2.3. Stream Flags
+ 2.1.2.4. Footer Magic Bytes
+ 2.2. Stream Padding
+ 3. Block
+ 3.1. Block Header
+ 3.1.1. Block Header Size
+ 3.1.2. Block Flags
+ 3.1.3. Compressed Size
+ 3.1.4. Uncompressed Size
+ 3.1.5. List of Filter Flags
+ 3.1.6. Header Padding
+ 3.1.7. CRC32
+ 3.2. Compressed Data
+ 3.3. Block Padding
+ 3.4. Check
+ 4. Index
+ 4.1. Index Indicator
+ 4.2. Number of Records
+ 4.3. List of Records
+ 4.3.1. Unpadded Size
+ 4.3.2. Uncompressed Size
+ 4.4. Index Padding
+ 4.5. CRC32
+ 5. Filter Chains
+ 5.1. Alignment
+ 5.2. Security
+ 5.3. Filters
+ 5.3.1. LZMA2
+ 5.3.2. Branch/Call/Jump Filters for Executables
+ 5.3.3. Delta
+ 5.3.3.1. Format of the Encoded Output
+ 5.4. Custom Filter IDs
+ 5.4.1. Reserved Custom Filter ID Ranges
+ 6. Cyclic Redundancy Checks
+ 7. References
+
+
+0. Preface
+
+ This document describes the .xz file format (filename suffix
+ ".xz", MIME type "application/x-xz"). It is intended that this
+ this format replace the old .lzma format used by LZMA SDK and
+ LZMA Utils.
+
+
+0.1. Notices and Acknowledgements
+
+ This file format was designed by Lasse Collin
+ <lasse.collin@tukaani.org> and Igor Pavlov.
+
+ Special thanks for helping with this document goes to
+ Ville Koskinen. Thanks for helping with this document goes to
+ Mark Adler, H. Peter Anvin, Mikko Pouru, and Lars Wirzenius.
+
+ This document has been put into the public domain.
+
+
+0.2. Getting the Latest Version
+
+ The latest official version of this document can be downloaded
+ from <http://tukaani.org/xz/xz-file-format.txt>.
+
+ Specific versions of this document have a filename
+ xz-file-format-X.Y.Z.txt where X.Y.Z is the version number.
+ For example, the version 1.0.0 of this document is available
+ at <http://tukaani.org/xz/xz-file-format-1.0.0.txt>.
+
+
+0.3. Version History
+
+ Version Date Description
+
+ 1.0.4 2009-08-27 Language improvements in Sections 1.2,
+ 2.1.1.2, 3.1.1, 3.1.2, and 5.3.1
+
+ 1.0.3 2009-06-05 Spelling fixes in Sections 5.1 and 5.4
+
+ 1.0.2 2009-06-04 Typo fixes in Sections 4 and 5.3.1
+
+ 1.0.1 2009-06-01 Typo fix in Section 0.3 and minor
+ clarifications to Sections 2, 2.2,
+ 3.3, 4.4, and 5.3.2
+
+ 1.0.0 2009-01-14 The first official version
+
+
+1. Conventions
+
+ The key words "MUST", "MUST NOT", "REQUIRED", "SHOULD",
+ "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
+ document are to be interpreted as described in [RFC-2119].
+
+ Indicating a warning means displaying a message, returning
+ appropriate exit status, or doing something else to let the
+ user know that something worth warning occurred. The operation
+ SHOULD still finish if a warning is indicated.
+
+ Indicating an error means displaying a message, returning
+ appropriate exit status, or doing something else to let the
+ user know that something prevented successfully finishing the
+ operation. The operation MUST be aborted once an error has
+ been indicated.
+
+
+1.1. Byte and Its Representation
+
+ In this document, byte is always 8 bits.
+
+ A "null byte" has all bits unset. That is, the value of a null
+ byte is 0x00.
+
+ To represent byte blocks, this document uses notation that
+ is similar to the notation used in [RFC-1952]:
+
+ +-------+
+ | Foo | One byte.
+ +-------+
+
+ +---+---+
+ | Foo | Two bytes; that is, some of the vertical bars
+ +---+---+ can be missing.
+
+ +=======+
+ | Foo | Zero or more bytes.
+ +=======+
+
+ In this document, a boxed byte or a byte sequence declared
+ using this notation is called "a field". The example field
+ above would be called "the Foo field" or plain "Foo".
+
+ If there are many fields, they may be split to multiple lines.
+ This is indicated with an arrow ("--->"):
+
+ +=====+
+ | Foo |
+ +=====+
+
+ +=====+
+ ---> | Bar |
+ +=====+
+
+ The above is equivalent to this:
+
+ +=====+=====+
+ | Foo | Bar |
+ +=====+=====+
+
+
+1.2. Multibyte Integers
+
+ Multibyte integers of static length, such as CRC values,
+ are stored in little endian byte order (least significant
+ byte first).
+
+ When smaller values are more likely than bigger values (for
+ example file sizes), multibyte integers are encoded in a
+ variable-length representation:
+ - Numbers in the range [0, 127] are copied as is, and take
+ one byte of space.
+ - Bigger numbers will occupy two or more bytes. All but the
+ last byte of the multibyte representation have the highest
+ (eighth) bit set.
+
+ For now, the value of the variable-length integers is limited
+ to 63 bits, which limits the encoded size of the integer to
+ nine bytes. These limits may be increased in the future if
+ needed.
+
+ The following C code illustrates encoding and decoding of
+ variable-length integers. The functions return the number of
+ bytes occupied by the integer (1-9), or zero on error.
+
+ #include <stddef.h>
+ #include <inttypes.h>
+
+ size_t
+ encode(uint8_t buf[static 9], uint64_t num)
+ {
+ if (num > UINT64_MAX / 2)
+ return 0;
+
+ size_t i = 0;
+
+ while (num >= 0x80) {
+ buf[i++] = (uint8_t)(num) | 0x80;
+ num >>= 7;
+ }
+
+ buf[i++] = (uint8_t)(num);
+
+ return i;
+ }
+
+ size_t
+ decode(const uint8_t buf[], size_t size_max, uint64_t *num)
+ {
+ if (size_max == 0)
+ return 0;
+
+ if (size_max > 9)
+ size_max = 9;
+
+ *num = buf[0] & 0x7F;
+ size_t i = 0;
+
+ while (buf[i++] & 0x80) {
+ if (i >= size_max || buf[i] == 0x00)
+ return 0;
+
+ *num |= (uint64_t)(buf[i] & 0x7F) << (i * 7);
+ }
+
+ return i;
+ }
+
+
+2. Overall Structure of .xz File
+
+ A standalone .xz files consist of one or more Streams which may
+ have Stream Padding between or after them:
+
+ +========+================+========+================+
+ | Stream | Stream Padding | Stream | Stream Padding | ...
+ +========+================+========+================+
+
+ The sizes of Stream and Stream Padding are always multiples
+ of four bytes, thus the size of every valid .xz file MUST be
+ a multiple of four bytes.
+
+ While a typical file contains only one Stream and no Stream
+ Padding, a decoder handling standalone .xz files SHOULD support
+ files that have more than one Stream or Stream Padding.
+
+ In contrast to standalone .xz files, when the .xz file format
+ is used as an internal part of some other file format or
+ communication protocol, it usually is expected that the decoder
+ stops after the first Stream, and doesn't look for Stream
+ Padding or possibly other Streams.
+
+
+2.1. Stream
+
+ +-+-+-+-+-+-+-+-+-+-+-+-+=======+=======+ +=======+
+ | Stream Header | Block | Block | ... | Block |
+ +-+-+-+-+-+-+-+-+-+-+-+-+=======+=======+ +=======+
+
+ +=======+-+-+-+-+-+-+-+-+-+-+-+-+
+ ---> | Index | Stream Footer |
+ +=======+-+-+-+-+-+-+-+-+-+-+-+-+
+
+ All the above fields have a size that is a multiple of four. If
+ Stream is used as an internal part of another file format, it
+ is RECOMMENDED to make the Stream start at an offset that is
+ a multiple of four bytes.
+
+ Stream Header, Index, and Stream Footer are always present in
+ a Stream. The maximum size of the Index field is 16 GiB (2^34).
+
+ There are zero or more Blocks. The maximum number of Blocks is
+ limited only by the maximum size of the Index field.
+
+ Total size of a Stream MUST be less than 8 EiB (2^63 bytes).
+ The same limit applies to the total amount of uncompressed
+ data stored in a Stream.
+
+ If an implementation supports handling .xz files with multiple
+ concatenated Streams, it MAY apply the above limits to the file
+ as a whole instead of limiting per Stream basis.
+
+
+2.1.1. Stream Header
+
+ +---+---+---+---+---+---+-------+------+--+--+--+--+
+ | Header Magic Bytes | Stream Flags | CRC32 |
+ +---+---+---+---+---+---+-------+------+--+--+--+--+
+
+
+2.1.1.1. Header Magic Bytes
+
+ The first six (6) bytes of the Stream are so called Header
+ Magic Bytes. They can be used to identify the file type.
+
+ Using a C array and ASCII:
+ const uint8_t HEADER_MAGIC[6]
+ = { 0xFD, '7', 'z', 'X', 'Z', 0x00 };
+
+ In plain hexadecimal:
+ FD 37 7A 58 5A 00
+
+ Notes:
+ - The first byte (0xFD) was chosen so that the files cannot
+ be erroneously detected as being in .lzma format, in which
+ the first byte is in the range [0x00, 0xE0].
+ - The sixth byte (0x00) was chosen to prevent applications
+ from misdetecting the file as a text file.
+
+ If the Header Magic Bytes don't match, the decoder MUST
+ indicate an error.
+
+
+2.1.1.2. Stream Flags
+
+ The first byte of Stream Flags is always a null byte. In the
+ future, this byte may be used to indicate a new Stream version
+ or other Stream properties.
+
+ The second byte of Stream Flags is a bit field:
+
+ Bit(s) Mask Description
+ 0-3 0x0F Type of Check (see Section 3.4):
+ ID Size Check name
+ 0x00 0 bytes None
+ 0x01 4 bytes CRC32
+ 0x02 4 bytes (Reserved)
+ 0x03 4 bytes (Reserved)
+ 0x04 8 bytes CRC64
+ 0x05 8 bytes (Reserved)
+ 0x06 8 bytes (Reserved)
+ 0x07 16 bytes (Reserved)
+ 0x08 16 bytes (Reserved)
+ 0x09 16 bytes (Reserved)
+ 0x0A 32 bytes SHA-256
+ 0x0B 32 bytes (Reserved)
+ 0x0C 32 bytes (Reserved)
+ 0x0D 64 bytes (Reserved)
+ 0x0E 64 bytes (Reserved)
+ 0x0F 64 bytes (Reserved)
+ 4-7 0xF0 Reserved for future use; MUST be zero for now.
+
+ Implementations SHOULD support at least the Check IDs 0x00
+ (None) and 0x01 (CRC32). Supporting other Check IDs is
+ OPTIONAL. If an unsupported Check is used, the decoder SHOULD
+ indicate a warning or error.
+
+ If any reserved bit is set, the decoder MUST indicate an error.
+ It is possible that there is a new field present which the
+ decoder is not aware of, and can thus parse the Stream Header
+ incorrectly.
+
+
+2.1.1.3. CRC32
+
+ The CRC32 is calculated from the Stream Flags field. It is
+ stored as an unsigned 32-bit little endian integer. If the
+ calculated value does not match the stored one, the decoder
+ MUST indicate an error.
+
+ The idea is that Stream Flags would always be two bytes, even
+ if new features are needed. This way old decoders will be able
+ to verify the CRC32 calculated from Stream Flags, and thus
+ distinguish between corrupt files (CRC32 doesn't match) and
+ files that the decoder doesn't support (CRC32 matches but
+ Stream Flags has reserved bits set).
+
+
+2.1.2. Stream Footer
+
+ +-+-+-+-+---+---+---+---+-------+------+----------+---------+
+ | CRC32 | Backward Size | Stream Flags | Footer Magic Bytes |
+ +-+-+-+-+---+---+---+---+-------+------+----------+---------+
+
+
+2.1.2.1. CRC32
+
+ The CRC32 is calculated from the Backward Size and Stream Flags
+ fields. It is stored as an unsigned 32-bit little endian
+ integer. If the calculated value does not match the stored one,
+ the decoder MUST indicate an error.
+
+ The reason to have the CRC32 field before the Backward Size and
+ Stream Flags fields is to keep the four-byte fields aligned to
+ a multiple of four bytes.
+
+
+2.1.2.2. Backward Size
+
+ Backward Size is stored as a 32-bit little endian integer,
+ which indicates the size of the Index field as multiple of
+ four bytes, minimum value being four bytes:
+
+ real_backward_size = (stored_backward_size + 1) * 4;
+
+ If the stored value does not match the real size of the Index
+ field, the decoder MUST indicate an error.
+
+ Using a fixed-size integer to store Backward Size makes
+ it slightly simpler to parse the Stream Footer when the
+ application needs to parse the Stream backwards.
+
+
+2.1.2.3. Stream Flags
+
+ This is a copy of the Stream Flags field from the Stream
+ Header. The information stored to Stream Flags is needed
+ when parsing the Stream backwards. The decoder MUST compare
+ the Stream Flags fields in both Stream Header and Stream
+ Footer, and indicate an error if they are not identical.
+
+
+2.1.2.4. Footer Magic Bytes
+
+ As the last step of the decoding process, the decoder MUST
+ verify the existence of Footer Magic Bytes. If they don't
+ match, an error MUST be indicated.
+
+ Using a C array and ASCII:
+ const uint8_t FOOTER_MAGIC[2] = { 'Y', 'Z' };
+
+ In hexadecimal:
+ 59 5A
+
+ The primary reason to have Footer Magic Bytes is to make
+ it easier to detect incomplete files quickly, without
+ uncompressing. If the file does not end with Footer Magic Bytes
+ (excluding Stream Padding described in Section 2.2), it cannot
+ be undamaged, unless someone has intentionally appended garbage
+ after the end of the Stream.
+
+
+2.2. Stream Padding
+
+ Only the decoders that support decoding of concatenated Streams
+ MUST support Stream Padding.
+
+ Stream Padding MUST contain only null bytes. To preserve the
+ four-byte alignment of consecutive Streams, the size of Stream
+ Padding MUST be a multiple of four bytes. Empty Stream Padding
+ is allowed. If these requirements are not met, the decoder MUST
+ indicate an error.
+
+ Note that non-empty Stream Padding is allowed at the end of the
+ file; there doesn't need to be a new Stream after non-empty
+ Stream Padding. This can be convenient in certain situations
+ [GNU-tar].
+
+ The possibility of Stream Padding MUST be taken into account
+ when designing an application that parses Streams backwards,
+ and the application supports concatenated Streams.
+
+
+3. Block
+
+ +==============+=================+===============+=======+
+ | Block Header | Compressed Data | Block Padding | Check |
+ +==============+=================+===============+=======+
+
+
+3.1. Block Header
+
+ +-------------------+-------------+=================+
+ | Block Header Size | Block Flags | Compressed Size |
+ +-------------------+-------------+=================+
+
+ +===================+======================+
+ ---> | Uncompressed Size | List of Filter Flags |
+ +===================+======================+
+
+ +================+--+--+--+--+
+ ---> | Header Padding | CRC32 |
+ +================+--+--+--+--+
+
+
+3.1.1. Block Header Size
+
+ This field overlaps with the Index Indicator field (see
+ Section 4.1).
+
+ This field contains the size of the Block Header field,
+ including the Block Header Size field itself. Valid values are
+ in the range [0x01, 0xFF], which indicate the size of the Block
+ Header as multiples of four bytes, minimum size being eight
+ bytes:
+
+ real_header_size = (encoded_header_size + 1) * 4;
+
+ If a Block Header bigger than 1024 bytes is needed in the
+ future, a new field can be added between the Block Header and
+ Compressed Data fields. The presence of this new field would
+ be indicated in the Block Header field.
+
+
+3.1.2. Block Flags
+
+ The Block Flags field is a bit field:
+
+ Bit(s) Mask Description
+ 0-1 0x03 Number of filters (1-4)
+ 2-5 0x3C Reserved for future use; MUST be zero for now.
+ 6 0x40 The Compressed Size field is present.
+ 7 0x80 The Uncompressed Size field is present.
+
+ If any reserved bit is set, the decoder MUST indicate an error.
+ It is possible that there is a new field present which the
+ decoder is not aware of, and can thus parse the Block Header
+ incorrectly.
+
+
+3.1.3. Compressed Size
+
+ This field is present only if the appropriate bit is set in
+ the Block Flags field (see Section 3.1.2).
+
+ The Compressed Size field contains the size of the Compressed
+ Data field, which MUST be non-zero. Compressed Size is stored
+ using the encoding described in Section 1.2. If the Compressed
+ Size doesn't match the size of the Compressed Data field, the
+ decoder MUST indicate an error.
+
+
+3.1.4. Uncompressed Size
+
+ This field is present only if the appropriate bit is set in
+ the Block Flags field (see Section 3.1.2).
+
+ The Uncompressed Size field contains the size of the Block
+ after uncompressing. Uncompressed Size is stored using the
+ encoding described in Section 1.2. If the Uncompressed Size
+ does not match the real uncompressed size, the decoder MUST
+ indicate an error.
+
+ Storing the Compressed Size and Uncompressed Size fields serves
+ several purposes:
+ - The decoder knows how much memory it needs to allocate
+ for a temporary buffer in multithreaded mode.
+ - Simple error detection: wrong size indicates a broken file.
+ - Seeking forwards to a specific location in streamed mode.
+
+ It should be noted that the only reliable way to determine
+ the real uncompressed size is to uncompress the Block,
+ because the Block Header and Index fields may contain
+ (intentionally or unintentionally) invalid information.
+
+
+3.1.5. List of Filter Flags
+
+ +================+================+ +================+
+ | Filter 0 Flags | Filter 1 Flags | ... | Filter n Flags |
+ +================+================+ +================+
+
+ The number of Filter Flags fields is stored in the Block Flags
+ field (see Section 3.1.2).
+
+ The format of each Filter Flags field is as follows:
+
+ +===========+====================+===================+
+ | Filter ID | Size of Properties | Filter Properties |
+ +===========+====================+===================+
+
+ Both Filter ID and Size of Properties are stored using the
+ encoding described in Section 1.2. Size of Properties indicates
+ the size of the Filter Properties field as bytes. The list of
+ officially defined Filter IDs and the formats of their Filter
+ Properties are described in Section 5.3.
+
+ Filter IDs greater than or equal to 0x4000_0000_0000_0000
+ (2^62) are reserved for implementation-specific internal use.
+ These Filter IDs MUST never be used in List of Filter Flags.
+
+
+3.1.6. Header Padding
+
+ This field contains as many null byte as it is needed to make
+ the Block Header have the size specified in Block Header Size.
+ If any of the bytes are not null bytes, the decoder MUST
+ indicate an error. It is possible that there is a new field
+ present which the decoder is not aware of, and can thus parse
+ the Block Header incorrectly.
+
+
+3.1.7. CRC32
+
+ The CRC32 is calculated over everything in the Block Header
+ field except the CRC32 field itself. It is stored as an
+ unsigned 32-bit little endian integer. If the calculated
+ value does not match the stored one, the decoder MUST indicate
+ an error.
+
+ By verifying the CRC32 of the Block Header before parsing the
+ actual contents allows the decoder to distinguish between
+ corrupt and unsupported files.
+
+
+3.2. Compressed Data
+
+ The format of Compressed Data depends on Block Flags and List
+ of Filter Flags. Excluding the descriptions of the simplest
+ filters in Section 5.3, the format of the filter-specific
+ encoded data is out of scope of this document.
+
+
+3.3. Block Padding
+
+ Block Padding MUST contain 0-3 null bytes to make the size of
+ the Block a multiple of four bytes. This can be needed when
+ the size of Compressed Data is not a multiple of four. If any
+ of the bytes in Block Padding are not null bytes, the decoder
+ MUST indicate an error.
+
+
+3.4. Check
+
+ The type and size of the Check field depends on which bits
+ are set in the Stream Flags field (see Section 2.1.1.2).
+
+ The Check, when used, is calculated from the original
+ uncompressed data. If the calculated Check does not match the
+ stored one, the decoder MUST indicate an error. If the selected
+ type of Check is not supported by the decoder, it SHOULD
+ indicate a warning or error.
+
+
+4. Index
+
+ +-----------------+===================+
+ | Index Indicator | Number of Records |
+ +-----------------+===================+
+
+ +=================+===============+-+-+-+-+
+ ---> | List of Records | Index Padding | CRC32 |
+ +=================+===============+-+-+-+-+
+
+ Index serves several purposes. Using it, one can
+ - verify that all Blocks in a Stream have been processed;
+ - find out the uncompressed size of a Stream; and
+ - quickly access the beginning of any Block (random access).
+
+
+4.1. Index Indicator
+
+ This field overlaps with the Block Header Size field (see
+ Section 3.1.1). The value of Index Indicator is always 0x00.
+
+
+4.2. Number of Records
+
+ This field indicates how many Records there are in the List
+ of Records field, and thus how many Blocks there are in the
+ Stream. The value is stored using the encoding described in
+ Section 1.2. If the decoder has decoded all the Blocks of the
+ Stream, and then notices that the Number of Records doesn't
+ match the real number of Blocks, the decoder MUST indicate an
+ error.
+
+
+4.3. List of Records
+
+ List of Records consists of as many Records as indicated by the
+ Number of Records field:
+
+ +========+========+
+ | Record | Record | ...
+ +========+========+
+
+ Each Record contains information about one Block:
+
+ +===============+===================+
+ | Unpadded Size | Uncompressed Size |
+ +===============+===================+
+
+ If the decoder has decoded all the Blocks of the Stream, it
+ MUST verify that the contents of the Records match the real
+ Unpadded Size and Uncompressed Size of the respective Blocks.
+
+ Implementation hint: It is possible to verify the Index with
+ constant memory usage by calculating for example SHA-256 of
+ both the real size values and the List of Records, then
+ comparing the hash values. Implementing this using
+ non-cryptographic hash like CRC32 SHOULD be avoided unless
+ small code size is important.
+
+ If the decoder supports random-access reading, it MUST verify
+ that Unpadded Size and Uncompressed Size of every completely
+ decoded Block match the sizes stored in the Index. If only
+ partial Block is decoded, the decoder MUST verify that the
+ processed sizes don't exceed the sizes stored in the Index.
+
+
+4.3.1. Unpadded Size
+
+ This field indicates the size of the Block excluding the Block
+ Padding field. That is, Unpadded Size is the size of the Block
+ Header, Compressed Data, and Check fields. Unpadded Size is
+ stored using the encoding described in Section 1.2. The value
+ MUST never be zero; with the current structure of Blocks, the
+ actual minimum value for Unpadded Size is five.
+
+ Implementation note: Because the size of the Block Padding
+ field is not included in Unpadded Size, calculating the total
+ size of a Stream or doing random-access reading requires
+ calculating the actual size of the Blocks by rounding Unpadded
+ Sizes up to the next multiple of four.
+
+ The reason to exclude Block Padding from Unpadded Size is to
+ ease making a raw copy of Compressed Data without Block
+ Padding. This can be useful, for example, if someone wants
+ to convert Streams to some other file format quickly.
+
+
+4.3.2. Uncompressed Size
+
+ This field indicates the Uncompressed Size of the respective
+ Block as bytes. The value is stored using the encoding
+ described in Section 1.2.
+
+
+4.4. Index Padding
+
+ This field MUST contain 0-3 null bytes to pad the Index to
+ a multiple of four bytes. If any of the bytes are not null
+ bytes, the decoder MUST indicate an error.
+
+
+4.5. CRC32
+
+ The CRC32 is calculated over everything in the Index field
+ except the CRC32 field itself. The CRC32 is stored as an
+ unsigned 32-bit little endian integer. If the calculated
+ value does not match the stored one, the decoder MUST indicate
+ an error.
+
+
+5. Filter Chains
+
+ The Block Flags field defines how many filters are used. When
+ more than one filter is used, the filters are chained; that is,
+ the output of one filter is the input of another filter. The
+ following figure illustrates the direction of data flow.
+
+ v Uncompressed Data ^
+ | Filter 0 |
+ Encoder | Filter 1 | Decoder
+ | Filter n |
+ v Compressed Data ^
+
+
+5.1. Alignment
+
+ Alignment of uncompressed input data is usually the job of
+ the application producing the data. For example, to get the
+ best results, an archiver tool should make sure that all
+ PowerPC executable files in the archive stream start at
+ offsets that are multiples of four bytes.
+
+ Some filters, for example LZMA2, can be configured to take
+ advantage of specified alignment of input data. Note that
+ taking advantage of aligned input can be beneficial also when
+ a filter is not the first filter in the chain. For example,
+ if you compress PowerPC executables, you may want to use the
+ PowerPC filter and chain that with the LZMA2 filter. Because
+ not only the input but also the output alignment of the PowerPC
+ filter is four bytes, it is now beneficial to set LZMA2
+ settings so that the LZMA2 encoder can take advantage of its
+ four-byte-aligned input data.
+
+ The output of the last filter in the chain is stored to the
+ Compressed Data field, which is is guaranteed to be aligned
+ to a multiple of four bytes relative to the beginning of the
+ Stream. This can increase
+ - speed, if the filtered data is handled multiple bytes at
+ a time by the filter-specific encoder and decoder,
+ because accessing aligned data in computer memory is
+ usually faster; and
+ - compression ratio, if the output data is later compressed
+ with an external compression tool.
+
+
+5.2. Security
+
+ If filters would be allowed to be chained freely, it would be
+ possible to create malicious files, that would be very slow to
+ decode. Such files could be used to create denial of service
+ attacks.
+
+ Slow files could occur when multiple filters are chained:
+
+ v Compressed input data
+ | Filter 1 decoder (last filter)
+ | Filter 0 decoder (non-last filter)
+ v Uncompressed output data
+
+ The decoder of the last filter in the chain produces a lot of
+ output from little input. Another filter in the chain takes the
+ output of the last filter, and produces very little output
+ while consuming a lot of input. As a result, a lot of data is
+ moved inside the filter chain, but the filter chain as a whole
+ gets very little work done.
+
+ To prevent this kind of slow files, there are restrictions on
+ how the filters can be chained. These restrictions MUST be
+ taken into account when designing new filters.
+
+ The maximum number of filters in the chain has been limited to
+ four, thus there can be at maximum of three non-last filters.
+ Of these three non-last filters, only two are allowed to change
+ the size of the data.
+
+ The non-last filters, that change the size of the data, MUST
+ have a limit how much the decoder can compress the data: the
+ decoder SHOULD produce at least n bytes of output when the
+ filter is given 2n bytes of input. This limit is not
+ absolute, but significant deviations MUST be avoided.
+
+ The above limitations guarantee that if the last filter in the
+ chain produces 4n bytes of output, the chain as a whole will
+ produce at least n bytes of output.
+
+
+5.3. Filters
+
+5.3.1. LZMA2
+
+ LZMA (Lempel-Ziv-Markov chain-Algorithm) is a general-purpose
+ compression algorithm with high compression ratio and fast
+ decompression. LZMA is based on LZ77 and range coding
+ algorithms.
+
+ LZMA2 is an extension on top of the original LZMA. LZMA2 uses
+ LZMA internally, but adds support for flushing the encoder,
+ uncompressed chunks, eases stateful decoder implementations,
+ and improves support for multithreading. Thus, the plain LZMA
+ will not be supported in this file format.
+
+ Filter ID: 0x21
+ Size of Filter Properties: 1 byte
+ Changes size of data: Yes
+ Allow as a non-last filter: No
+ Allow as the last filter: Yes
+
+ Preferred alignment:
+ Input data: Adjustable to 1/2/4/8/16 byte(s)
+ Output data: 1 byte
+
+ The format of the one-byte Filter Properties field is as
+ follows:
+
+ Bits Mask Description
+ 0-5 0x3F Dictionary Size
+ 6-7 0xC0 Reserved for future use; MUST be zero for now.
+
+ Dictionary Size is encoded with one-bit mantissa and five-bit
+ exponent. The smallest dictionary size is 4 KiB and the biggest
+ is 4 GiB.
+
+ Raw value Mantissa Exponent Dictionary size
+ 0 2 11 4 KiB
+ 1 3 11 6 KiB
+ 2 2 12 8 KiB
+ 3 3 12 12 KiB
+ 4 2 13 16 KiB
+ 5 3 13 24 KiB
+ 6 2 14 32 KiB
+ ... ... ... ...
+ 35 3 27 768 MiB
+ 36 2 28 1024 MiB
+ 37 3 29 1536 MiB
+ 38 2 30 2048 MiB
+ 39 3 30 3072 MiB
+ 40 2 31 4096 MiB - 1 B
+
+ Instead of having a table in the decoder, the dictionary size
+ can be decoded using the following C code:
+
+ const uint8_t bits = get_dictionary_flags() & 0x3F;
+ if (bits > 40)
+ return DICTIONARY_TOO_BIG; // Bigger than 4 GiB
+
+ uint32_t dictionary_size;
+ if (bits == 40) {
+ dictionary_size = UINT32_MAX;
+ } else {
+ dictionary_size = 2 | (bits & 1);
+ dictionary_size <<= bits / 2 + 11;
+ }
+
+
+5.3.2. Branch/Call/Jump Filters for Executables
+
+ These filters convert relative branch, call, and jump
+ instructions to their absolute counterparts in executable
+ files. This conversion increases redundancy and thus
+ compression ratio.
+
+ Size of Filter Properties: 0 or 4 bytes
+ Changes size of data: No
+ Allow as a non-last filter: Yes
+ Allow as the last filter: No
+
+ Below is the list of filters in this category. The alignment
+ is the same for both input and output data.
+
+ Filter ID Alignment Description
+ 0x04 1 byte x86 filter (BCJ)
+ 0x05 4 bytes PowerPC (big endian) filter
+ 0x06 16 bytes IA64 filter
+ 0x07 4 bytes ARM (little endian) filter
+ 0x08 2 bytes ARM Thumb (little endian) filter
+ 0x09 4 bytes SPARC filter
+
+ If the size of Filter Properties is four bytes, the Filter
+ Properties field contains the start offset used for address
+ conversions. It is stored as an unsigned 32-bit little endian
+ integer. The start offset MUST be a multiple of the alignment
+ of the filter as listed in the table above; if it isn't, the
+ decoder MUST indicate an error. If the size of Filter
+ Properties is zero, the start offset is zero.
+
+ Setting the start offset may be useful if an executable has
+ multiple sections, and there are many cross-section calls.
+ Taking advantage of this feature usually requires usage of
+ the Subblock filter, whose design is not complete yet.
+
+
+5.3.3. Delta
+
+ The Delta filter may increase compression ratio when the value
+ of the next byte correlates with the value of an earlier byte
+ at specified distance.
+
+ Filter ID: 0x03
+ Size of Filter Properties: 1 byte
+ Changes size of data: No
+ Allow as a non-last filter: Yes
+ Allow as the last filter: No
+
+ Preferred alignment:
+ Input data: 1 byte
+ Output data: Same as the original input data
+
+ The Properties byte indicates the delta distance, which can be
+ 1-256 bytes backwards from the current byte: 0x00 indicates
+ distance of 1 byte and 0xFF distance of 256 bytes.
+
+
+5.3.3.1. Format of the Encoded Output
+
+ The code below illustrates both encoding and decoding with
+ the Delta filter.
+
+ // Distance is in the range [1, 256].
+ const unsigned int distance = get_properties_byte() + 1;
+ uint8_t pos = 0;
+ uint8_t delta[256];
+
+ memset(delta, 0, sizeof(delta));
+
+ while (1) {
+ const int byte = read_byte();
+ if (byte == EOF)
+ break;
+
+ uint8_t tmp = delta[(uint8_t)(distance + pos)];
+ if (is_encoder) {
+ tmp = (uint8_t)(byte) - tmp;
+ delta[pos] = (uint8_t)(byte);
+ } else {
+ tmp = (uint8_t)(byte) + tmp;
+ delta[pos] = tmp;
+ }
+
+ write_byte(tmp);
+ --pos;
+ }
+
+
+5.4. Custom Filter IDs
+
+ If a developer wants to use custom Filter IDs, he has two
+ choices. The first choice is to contact Lasse Collin and ask
+ him to allocate a range of IDs for the developer.
+
+ The second choice is to generate a 40-bit random integer,
+ which the developer can use as his personal Developer ID.
+ To minimize the risk of collisions, Developer ID has to be
+ a randomly generated integer, not manually selected "hex word".
+ The following command, which works on many free operating
+ systems, can be used to generate Developer ID:
+
+ dd if=/dev/urandom bs=5 count=1 | hexdump
+
+ The developer can then use his Developer ID to create unique
+ (well, hopefully unique) Filter IDs.
+
+ Bits Mask Description
+ 0-15 0x0000_0000_0000_FFFF Filter ID
+ 16-55 0x00FF_FFFF_FFFF_0000 Developer ID
+ 56-62 0x3F00_0000_0000_0000 Static prefix: 0x3F
+
+ The resulting 63-bit integer will use 9 bytes of space when
+ stored using the encoding described in Section 1.2. To get
+ a shorter ID, see the beginning of this Section how to
+ request a custom ID range.
+
+
+5.4.1. Reserved Custom Filter ID Ranges
+
+ Range Description
+ 0x0000_0300 - 0x0000_04FF Reserved to ease .7z compatibility
+ 0x0002_0000 - 0x0007_FFFF Reserved to ease .7z compatibility
+ 0x0200_0000 - 0x07FF_FFFF Reserved to ease .7z compatibility
+
+
+6. Cyclic Redundancy Checks
+
+ There are several incompatible variations to calculate CRC32
+ and CRC64. For simplicity and clarity, complete examples are
+ provided to calculate the checks as they are used in this file
+ format. Implementations MAY use different code as long as it
+ gives identical results.
+
+ The program below reads data from standard input, calculates
+ the CRC32 and CRC64 values, and prints the calculated values
+ as big endian hexadecimal strings to standard output.
+
+ #include <stddef.h>
+ #include <inttypes.h>
+ #include <stdio.h>
+
+ uint32_t crc32_table[256];
+ uint64_t crc64_table[256];
+
+ void
+ init(void)
+ {
+ static const uint32_t poly32 = UINT32_C(0xEDB88320);
+ static const uint64_t poly64
+ = UINT64_C(0xC96C5795D7870F42);
+
+ for (size_t i = 0; i < 256; ++i) {
+ uint32_t crc32 = i;
+ uint64_t crc64 = i;
+
+ for (size_t j = 0; j < 8; ++j) {
+ if (crc32 & 1)
+ crc32 = (crc32 >> 1) ^ poly32;
+ else
+ crc32 >>= 1;
+
+ if (crc64 & 1)
+ crc64 = (crc64 >> 1) ^ poly64;
+ else
+ crc64 >>= 1;
+ }
+
+ crc32_table[i] = crc32;
+ crc64_table[i] = crc64;
+ }
+ }
+
+ uint32_t
+ crc32(const uint8_t *buf, size_t size, uint32_t crc)
+ {
+ crc = ~crc;
+ for (size_t i = 0; i < size; ++i)
+ crc = crc32_table[buf[i] ^ (crc & 0xFF)]
+ ^ (crc >> 8);
+ return ~crc;
+ }
+
+ uint64_t
+ crc64(const uint8_t *buf, size_t size, uint64_t crc)
+ {
+ crc = ~crc;
+ for (size_t i = 0; i < size; ++i)
+ crc = crc64_table[buf[i] ^ (crc & 0xFF)]
+ ^ (crc >> 8);
+ return ~crc;
+ }
+
+ int
+ main()
+ {
+ init();
+
+ uint32_t value32 = 0;
+ uint64_t value64 = 0;
+ uint64_t total_size = 0;
+ uint8_t buf[8192];
+
+ while (1) {
+ const size_t buf_size
+ = fread(buf, 1, sizeof(buf), stdin);
+ if (buf_size == 0)
+ break;
+
+ total_size += buf_size;
+ value32 = crc32(buf, buf_size, value32);
+ value64 = crc64(buf, buf_size, value64);
+ }
+
+ printf("Bytes: %" PRIu64 "\n", total_size);
+ printf("CRC-32: 0x%08" PRIX32 "\n", value32);
+ printf("CRC-64: 0x%016" PRIX64 "\n", value64);
+
+ return 0;
+ }
+
+
+7. References
+
+ LZMA SDK - The original LZMA implementation
+ http://7-zip.org/sdk.html
+
+ LZMA Utils - LZMA adapted to POSIX-like systems
+ http://tukaani.org/lzma/
+
+ XZ Utils - The next generation of LZMA Utils
+ http://tukaani.org/xz/
+
+ [RFC-1952]
+ GZIP file format specification version 4.3
+ http://www.ietf.org/rfc/rfc1952.txt
+ - Notation of byte boxes in section "2.1. Overall conventions"
+
+ [RFC-2119]
+ Key words for use in RFCs to Indicate Requirement Levels
+ http://www.ietf.org/rfc/rfc2119.txt
+
+ [GNU-tar]
+ GNU tar 1.21 manual
+ http://www.gnu.org/software/tar/manual/html_node/Blocking-Factor.html
+ - Node 9.4.2 "Blocking Factor", paragraph that begins
+ "gzip will complain about trailing garbage"
+ - Note that this URL points to the latest version of the
+ manual, and may some day not contain the note which is in
+ 1.21. For the exact version of the manual, download GNU
+ tar 1.21: ftp://ftp.gnu.org/pub/gnu/tar/tar-1.21.tar.gz
+
diff --git a/dos/Makefile b/dos/Makefile
new file mode 100644
index 0000000..0b8a088
--- /dev/null
+++ b/dos/Makefile
@@ -0,0 +1,255 @@
+###############################################################################
+#
+# Makefile to build XZ Utils using DJGPP
+#
+# Make flags to alter compilation:
+#
+# DEBUG=1 Enable assertions. Don't use this for production builds!
+# You may also want to set CFLAGS="-g -O0" to disable
+# optimizations.
+#
+# The usual CPPFLAGS and CFLAGS are supported too.
+#
+###############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+CC = gcc
+AR = ar
+STRIP = strip
+SED = sed
+RM = rm -f
+
+CFLAGS = -g -Wextra -Wfatal-errors -Wall -march=i386 -mtune=i686 -O2
+
+# NOTE: -fgnu89-inline is needed on DJGPP 2.04 beta and GCC 4.3.2
+# because time.h uses GNU-style "extern inline".
+ALL_CFLAGS = -std=gnu99 -fgnu89-inline
+
+ALL_CPPFLAGS = \
+ -I. \
+ -I../lib \
+ -I../src/common \
+ -I../src/liblzma/api \
+ -I../src/liblzma/common \
+ -I../src/liblzma/check \
+ -I../src/liblzma/rangecoder \
+ -I../src/liblzma/lz \
+ -I../src/liblzma/lzma \
+ -I../src/liblzma/delta \
+ -I../src/liblzma/simple \
+ -I../src/liblzma/subblock
+
+ALL_CPPFLAGS += -DHAVE_CONFIG_H
+
+ifdef DEBUG
+STRIP := rem Skipping strip
+else
+ALL_CPPFLAGS += -DNDEBUG
+endif
+
+ALL_CPPFLAGS += $(CPPFLAGS)
+ALL_CFLAGS += $(CFLAGS)
+
+
+################
+# Common rules #
+################
+
+.PHONY: all clean
+all: liblzma.a getopt.a xzdec.exe lzmadec.exe xz.exe
+clean: liblzma-clean getopt-clean xzdec-clean xz-clean
+
+
+#############
+# liblzma.a #
+#############
+
+LIBLZMA_SRCS_C = \
+ ../src/liblzma/common/alone_decoder.c \
+ ../src/liblzma/common/alone_encoder.c \
+ ../src/liblzma/common/auto_decoder.c \
+ ../src/liblzma/common/block_buffer_decoder.c \
+ ../src/liblzma/common/block_buffer_encoder.c \
+ ../src/liblzma/common/block_decoder.c \
+ ../src/liblzma/common/block_encoder.c \
+ ../src/liblzma/common/block_header_decoder.c \
+ ../src/liblzma/common/block_header_encoder.c \
+ ../src/liblzma/common/block_util.c \
+ ../src/liblzma/common/common.c \
+ ../src/liblzma/common/easy_buffer_encoder.c \
+ ../src/liblzma/common/easy_decoder_memusage.c \
+ ../src/liblzma/common/easy_encoder.c \
+ ../src/liblzma/common/easy_encoder_memusage.c \
+ ../src/liblzma/common/easy_preset.c \
+ ../src/liblzma/common/filter_common.c \
+ ../src/liblzma/common/filter_decoder.c \
+ ../src/liblzma/common/filter_encoder.c \
+ ../src/liblzma/common/filter_flags_decoder.c \
+ ../src/liblzma/common/filter_flags_encoder.c \
+ ../src/liblzma/common/index.c \
+ ../src/liblzma/common/index_decoder.c \
+ ../src/liblzma/common/index_encoder.c \
+ ../src/liblzma/common/index_hash.c \
+ ../src/liblzma/common/stream_buffer_decoder.c \
+ ../src/liblzma/common/stream_buffer_encoder.c \
+ ../src/liblzma/common/stream_decoder.c \
+ ../src/liblzma/common/stream_encoder.c \
+ ../src/liblzma/common/stream_flags_common.c \
+ ../src/liblzma/common/stream_flags_decoder.c \
+ ../src/liblzma/common/stream_flags_encoder.c \
+ ../src/liblzma/common/vli_decoder.c \
+ ../src/liblzma/common/vli_encoder.c \
+ ../src/liblzma/common/vli_size.c \
+ ../src/liblzma/check/check.c \
+ ../src/liblzma/check/crc32_table.c \
+ ../src/liblzma/check/crc64_table.c \
+ ../src/liblzma/check/sha256.c \
+ ../src/liblzma/rangecoder/price_table.c \
+ ../src/liblzma/lz/lz_decoder.c \
+ ../src/liblzma/lz/lz_encoder.c \
+ ../src/liblzma/lz/lz_encoder_mf.c \
+ ../src/liblzma/lzma/fastpos_table.c \
+ ../src/liblzma/lzma/lzma2_decoder.c \
+ ../src/liblzma/lzma/lzma2_encoder.c \
+ ../src/liblzma/lzma/lzma_decoder.c \
+ ../src/liblzma/lzma/lzma_encoder.c \
+ ../src/liblzma/lzma/lzma_encoder_optimum_fast.c \
+ ../src/liblzma/lzma/lzma_encoder_optimum_normal.c \
+ ../src/liblzma/lzma/lzma_encoder_presets.c \
+ ../src/liblzma/delta/delta_common.c \
+ ../src/liblzma/delta/delta_decoder.c \
+ ../src/liblzma/delta/delta_encoder.c \
+ ../src/liblzma/simple/arm.c \
+ ../src/liblzma/simple/armthumb.c \
+ ../src/liblzma/simple/ia64.c \
+ ../src/liblzma/simple/powerpc.c \
+ ../src/liblzma/simple/simple_coder.c \
+ ../src/liblzma/simple/simple_decoder.c \
+ ../src/liblzma/simple/simple_encoder.c \
+ ../src/liblzma/simple/sparc.c \
+ ../src/liblzma/simple/x86.c
+
+LIBLZMA_SRCS_ASM = \
+ ../src/liblzma/check/crc32_x86.S \
+ ../src/liblzma/check/crc64_x86.S
+
+LIBLZMA_OBJS_C = $(LIBLZMA_SRCS_C:.c=.o)
+LIBLZMA_OBJS_ASM = $(LIBLZMA_SRCS_ASM:.S=.o)
+LIBLZMA_OBJS = $(LIBLZMA_OBJS_C) $(LIBLZMA_OBJS_ASM)
+
+$(LIBLZMA_OBJS_C): %.o: %.c
+ $(CC) $(ALL_CPPFLAGS) $(ALL_CFLAGS) -c -o $@ $<
+
+$(LIBLZMA_OBJS_ASM): %.o: %.S
+ $(CC) $(ALL_CPPFLAGS) $(ALL_CFLAGS) -c -o $@ $<
+
+liblzma.a: $(LIBLZMA_OBJS)
+ $(RM) $@
+ $(AR) rcs $@ $(LIBLZMA_OBJS)
+ $(STRIP) --strip-unneeded $@
+
+# Avoid too long command lines.
+.PHONY: liblzma-clean $(LIBLZMA_OBJS:.o=-clean)
+liblzma-clean: $(LIBLZMA_OBJS:.o=-clean)
+ -$(RM) liblzma.a
+
+$(LIBLZMA_OBJS:.o=-clean):
+ -$(RM) $(@:-clean=.o)
+
+
+###############
+# getopt_long #
+###############
+
+GETOPT_SRCS = \
+ ../lib/getopt.c \
+ ../lib/getopt1.c
+
+GETOPT_OBJS = $(GETOPT_SRCS:.c=.o)
+
+GETOPT_H = ../lib/getopt.h
+
+$(GETOPT_H): %.h: %.in.h
+ $(SED) "" $< > $@
+
+$(GETOPT_OBJS): %.o: %.c $(GETOPT_H)
+ $(CC) $(ALL_CPPFLAGS) $(ALL_CFLAGS) -c -o $@ $<
+
+getopt.a: $(GETOPT_OBJS)
+ $(RM) $@
+ $(AR) rcs $@ $(GETOPT_OBJS)
+ $(STRIP) --strip-unneeded $@
+
+getopt-clean:
+ $(RM) $(GETOPT_H) $(GETOPT_OBJS) getopt.a
+
+
+###########################
+# xzdec.exe & lzmadec.exe #
+###########################
+
+XZDEC_SRCS = ../src/xzdec/xzdec.c
+
+xzdec.exe: getopt.a liblzma.a $(XZDEC_SRCS)
+ $(CC) $(ALL_CPPFLAGS) $(ALL_CFLAGS) $(XZDEC_SRCS) -o $@ getopt.a liblzma.a
+ $(STRIP) --strip-all $@
+ exe2coff $@
+ $(RM) $@
+ copy /b $(DJGPP:DJGPP.ENV=BIN\CWSDSTUB.EXE) + $(@:.exe=) $@
+ $(RM) $(@:.exe=)
+
+lzmadec.exe: getopt.a liblzma.a $(XZDEC_SRCS)
+ $(CC) $(ALL_CPPFLAGS) -DLZMADEC $(ALL_CFLAGS) $(XZDEC_SRCS) -o $@ getopt.a liblzma.a
+ $(STRIP) --strip-all $@
+ exe2coff $@
+ $(RM) $@
+ copy /b $(DJGPP:DJGPP.ENV=BIN\CWSDSTUB.EXE) + $(@:.exe=) $@
+ $(RM) $(@:.exe=)
+
+.PHONY: xzdec-clean
+xzdec-clean:
+ -$(RM) xzdec.exe lzmadec.exe xzdec lzmadec
+
+
+##########
+# xz.exe #
+##########
+
+XZ_SRCS = \
+ ../src/xz/args.c \
+ ../src/xz/coder.c \
+ ../src/xz/file_io.c \
+ ../src/xz/hardware.c \
+ ../src/xz/main.c \
+ ../src/xz/message.c \
+ ../src/xz/options.c \
+ ../src/xz/signals.c \
+ ../src/xz/suffix.c \
+ ../src/xz/util.c
+
+XZ_OBJS = $(XZ_SRCS:.c=.o)
+
+$(XZ_OBJS): %.o: %.c
+ $(CC) $(ALL_CPPFLAGS) $(ALL_CFLAGS) -c -o $@ $<
+
+xz.exe: getopt.a liblzma.a $(XZ_OBJS)
+ $(CC) $(ALL_CFLAGS) $(XZ_OBJS) -o $@ getopt.a liblzma.a
+ $(STRIP) --strip-all $@
+ exe2coff $@
+ $(RM) $@
+ copy /b $(DJGPP:DJGPP.ENV=BIN\CWSDSTUB.EXE) + $(@:.exe=) $@
+ $(RM) $(@:.exe=)
+
+# Avoid too long command lines.
+.PHONY: xz-clean $(XZ_OBJS:.o=-clean)
+xz-clean: $(XZ_OBJS:.o=-clean)
+ -$(RM) xz.exe xz
+
+$(XZ_OBJS:.o=-clean):
+ -$(RM) $(@:-clean=.o)
diff --git a/dos/README b/dos/README
new file mode 100644
index 0000000..649c58c
--- /dev/null
+++ b/dos/README
@@ -0,0 +1,113 @@
+
+XZ Utils on DOS
+===============
+
+Introduction
+
+ This document explains how to build XZ Utils for DOS using DJGPP.
+ The resulting binaries should run at least on various DOS versions
+ and under Windows 95/98/98SE/ME, which cannot run the Windows version
+ of XZ Utils.
+
+ This is currently experimental and has got very little testing.
+
+
+Getting and Installing DJGPP
+
+ You may use <http://www.delorie.com/djgpp/zip-picker.html> to help
+ deciding what to download, but as of writing (2009-02-13) that may
+ not be the most convenient way taking into account what components
+ are actually required to build XZ Utils. However, using the
+ zip-picker can still be worth doing to get nice short summary of
+ installation instructions (they can be found from readme.1st too).
+
+ For more manual method, first select a mirror from
+ <http://www.delorie.com/djgpp/getting.html>. You need
+ the following files:
+
+ unzip32.exe
+ beta/v2/djdev204.zip
+ v2gnu/bnu219b.zip
+ v2gnu/gcc432b.zip
+ v2gnu/mak3791b.zip
+ v2gnu/sed415b.zip
+ v2misc/csdpmi5b.zip
+
+ If newer versions are available, probably you should try them first.
+ Note that djdev203.zip is too old to build XZ Utils; you need at
+ least djdev204.zip. Also note that you want csdpmi5b.zip even if you
+ run under Windows or DOSEMU, because the XZ Utils Makefile will embed
+ cwsdstub.exe to the resulting binaries.
+
+ See the instructions in readme.1st found from djdev204.zip. Here's
+ a short summary, but you should still read readme.1st.
+
+ C:\> mkdir DJGPP
+ C:\> cd DJGPP
+ C:\DJGPP> c:\download\unzip32 c:\download\djdev204.zip
+ C:\DJGPP> c:\download\unzip32 c:\download\bnu219b.zip
+ C:\DJGPP> c:\download\unzip32 c:\download\gcc432b.zip
+ C:\DJGPP> c:\download\unzip32 c:\download\mak3791b.zip
+ C:\DJGPP> c:\download\unzip32 c:\download\sed415b.zip
+ C:\DJGPP> c:\download\unzip32 c:\download\csdpmi5b.zip
+
+ C:\DJGPP> set PATH=C:\DJGPP\BIN;%PATH%
+ C:\DJGPP> set DJGPP=C:\DJGPP\DJGPP.ENV
+
+ You may want to add the last two lines into AUTOEXEC.BAT or have,
+ for example, DJGPP.BAT which you can run before using DJGPP.
+
+ Make sure you use completely upper case path in the DJGPP environment
+ variable. This is not required by DJGPP, but the XZ Utils Makefile is
+ a bit stupid and expects that everything in DJGPP environment variable
+ is uppercase.
+
+
+Building
+
+ Just run "make" in this directory (the directory containing this
+ README). You should get liblzma.a, xz.exe, xzdec.exe, and
+ lzmadec.exe. Of these, probably xz.exe is the only interesting one.
+
+ Note: You need to have an environment that supports long filenames.
+ Once you have built XZ Utils, the resulting binaries can be run
+ without long filename support.
+
+
+Additional Make Flags and Targets
+
+ You may want to try some additional optimizations, which may or
+ may not make the code faster (and may or may not hit possible
+ compiler bugs more easily):
+
+ make CFLAGS="-O3 -fomit-frame-pointer -funroll-loops"
+
+ If you want to enable assertions (the assert() macro), use DEBUG=1.
+ You may want to disable optimizations too if you plan to actually
+ debug the code. Never use DEBUG=1 for production builds!
+
+ make DEBUG=1 CFLAGS="-g -O0"
+
+
+Bugs
+
+ "make clean" may remove src/xz/hardware.c when it tries to remove
+ src/xz/hardware-fixed.c. This is probably a bug somewhere in the
+ DOS environment I use. Maybe it tries truncated 8.3 name first and
+ since that gives a name of an existing file, it doesn't look for
+ long filename.
+
+ "xz -fc /dev/tty" hangs at least in DOSEMU and cannot be interrupted
+ by pressing C-c. Maybe xz should never accept non-regular files on
+ DOS even when --force is used.
+
+ Using different memory usage limit for encoding and decoding doesn't
+ make sense under pure DOS. Maybe it is still OK when running under
+ Windows.
+
+ The progress indicator of "xz -v" doesn't get updated when running
+ under Dosbox, but it works in DOSEMU. I currently (2009-02-13) don't
+ know if it works in other environments.
+
+ Report bugs to <lasse.collin@tukaani.org> (in English or Finnish).
+
diff --git a/dos/config.h b/dos/config.h
new file mode 100644
index 0000000..dfd6f79
--- /dev/null
+++ b/dos/config.h
@@ -0,0 +1,153 @@
+/* Define to 1 if using x86 assembler optimizations. */
+#define HAVE_ASM_X86 1
+
+/* Define to 1 if crc32 integrity check is enabled. */
+#define HAVE_CHECK_CRC32 1
+
+/* Define to 1 if crc64 integrity check is enabled. */
+#define HAVE_CHECK_CRC64 1
+
+/* Define to 1 if sha256 integrity check is enabled. */
+#define HAVE_CHECK_SHA256 1
+
+/* Define to 1 if decoder components are enabled. */
+#define HAVE_DECODER 1
+
+/* Define to 1 if arm decoder is enabled. */
+#define HAVE_DECODER_ARM 1
+
+/* Define to 1 if armthumb decoder is enabled. */
+#define HAVE_DECODER_ARMTHUMB 1
+
+/* Define to 1 if delta decoder is enabled. */
+#define HAVE_DECODER_DELTA 1
+
+/* Define to 1 if ia64 decoder is enabled. */
+#define HAVE_DECODER_IA64 1
+
+/* Define to 1 if lzma1 decoder is enabled. */
+#define HAVE_DECODER_LZMA1 1
+
+/* Define to 1 if lzma2 decoder is enabled. */
+#define HAVE_DECODER_LZMA2 1
+
+/* Define to 1 if powerpc decoder is enabled. */
+#define HAVE_DECODER_POWERPC 1
+
+/* Define to 1 if sparc decoder is enabled. */
+#define HAVE_DECODER_SPARC 1
+
+/* Define to 1 if subblock decoder is enabled. */
+/* #undef HAVE_DECODER_SUBBLOCK */
+
+/* Define to 1 if x86 decoder is enabled. */
+#define HAVE_DECODER_X86 1
+
+/* Define to 1 if encoder components are enabled. */
+#define HAVE_ENCODER 1
+
+/* Define to 1 if arm encoder is enabled. */
+#define HAVE_ENCODER_ARM 1
+
+/* Define to 1 if armthumb encoder is enabled. */
+#define HAVE_ENCODER_ARMTHUMB 1
+
+/* Define to 1 if delta encoder is enabled. */
+#define HAVE_ENCODER_DELTA 1
+
+/* Define to 1 if ia64 encoder is enabled. */
+#define HAVE_ENCODER_IA64 1
+
+/* Define to 1 if lzma1 encoder is enabled. */
+#define HAVE_ENCODER_LZMA1 1
+
+/* Define to 1 if lzma2 encoder is enabled. */
+#define HAVE_ENCODER_LZMA2 1
+
+/* Define to 1 if powerpc encoder is enabled. */
+#define HAVE_ENCODER_POWERPC 1
+
+/* Define to 1 if sparc encoder is enabled. */
+#define HAVE_ENCODER_SPARC 1
+
+/* Define to 1 if subblock encoder is enabled. */
+/* #undef HAVE_ENCODER_SUBBLOCK */
+
+/* Define to 1 if x86 encoder is enabled. */
+#define HAVE_ENCODER_X86 1
+
+/* Define to 1 if the system supports fast unaligned memory access. */
+#define HAVE_FAST_UNALIGNED_ACCESS 1
+
+/* Define to 1 if you have the <inttypes.h> header file. */
+#define HAVE_INTTYPES_H 1
+
+/* Define to 1 if you have the <limits.h> header file. */
+#define HAVE_LIMITS_H 1
+
+/* Define to 1 if you have the <memory.h> header file. */
+#define HAVE_MEMORY_H 1
+
+/* Define to 1 to enable bt2 match finder. */
+#define HAVE_MF_BT2 1
+
+/* Define to 1 to enable bt3 match finder. */
+#define HAVE_MF_BT3 1
+
+/* Define to 1 to enable bt4 match finder. */
+#define HAVE_MF_BT4 1
+
+/* Define to 1 to enable hc3 match finder. */
+#define HAVE_MF_HC3 1
+
+/* Define to 1 to enable hc4 match finder. */
+#define HAVE_MF_HC4 1
+
+/* Define to 1 if optimizing for size. */
+/* #undef HAVE_SMALL */
+
+/* Define to 1 if stdbool.h conforms to C99. */
+#define HAVE_STDBOOL_H 1
+
+/* Define to 1 if you have the <stdint.h> header file. */
+#define HAVE_STDINT_H 1
+
+/* Define to 1 if you have the <stdlib.h> header file. */
+#define HAVE_STDLIB_H 1
+
+/* Define to 1 if you have the <strings.h> header file. */
+#define HAVE_STRINGS_H 1
+
+/* Define to 1 if you have the <string.h> header file. */
+#define HAVE_STRING_H 1
+
+/* Define to 1 if the system has the type `uintptr_t'. */
+#define HAVE_UINTPTR_T 1
+
+/* Define to 1 if you have the <sys/time.h> header file. */
+#define HAVE_SYS_TIME_H 1
+
+/* Define to 1 if you have the `utimes' function. */
+#define HAVE_UTIMES 1
+
+/* Define to 1 or 0, depending whether the compiler supports simple visibility
+ declarations. */
+#define HAVE_VISIBILITY 0
+
+/* Define to 1 if the system has the type `_Bool'. */
+#define HAVE__BOOL 1
+
+/* Define to the address where bug reports for this package should be sent. */
+#define PACKAGE_BUGREPORT "lasse.collin@tukaani.org"
+
+/* Define to the URL of the home page of this package. */
+#define PACKAGE_HOMEPAGE "http://tukaani.org/xz/"
+
+/* Define to the full name of this package. */
+#define PACKAGE_NAME "XZ Utils"
+
+/* The size of `size_t', as computed by sizeof. */
+#define SIZEOF_SIZE_T 4
+
+/* Define to 1 if you have the ANSI C header files. */
+#define STDC_HEADERS 1
diff --git a/extra/7z2lzma/7z2lzma.bash b/extra/7z2lzma/7z2lzma.bash
new file mode 100755
index 0000000..1777c78
--- /dev/null
+++ b/extra/7z2lzma/7z2lzma.bash
@@ -0,0 +1,115 @@
+#!/bin/bash
+#
+#############################################################################
+#
+# 7z2lzma.bash is very primitive .7z to .lzma converter. The input file must
+# have exactly one LZMA compressed stream, which has been created with the
+# default lc, lp, and pb values. The CRC32 in the .7z archive is not checked,
+# and the script may seem to succeed while it actually created a corrupt .lzma
+# file. You should always try uncompressing both the original .7z and the
+# created .lzma and compare that the output is identical.
+#
+# This script requires basic GNU tools and 7z or 7za tool from p7zip.
+#
+# Last modified: 2009-01-15 14:25+0200
+#
+#############################################################################
+#
+# Author: Lasse Collin <lasse.collin@tukaani.org>
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+#############################################################################
+
+# You can use 7z or 7za, both will work.
+SEVENZIP=7za
+
+if [ $# != 2 -o -z "$1" -o -z "$2" ]; then
+ echo "Usage: $0 input.7z output.lzma"
+ exit 1
+fi
+
+# Converts an integer variable to little endian binary integer.
+int2bin()
+{
+ local LEN=$1
+ local NUM=$2
+ local HEX=(0 1 2 3 4 5 6 7 8 9 A B C D E F)
+ local I
+ for ((I=0; I < "$LEN"; ++I)); do
+ printf "\\x${HEX[(NUM >> 4) & 0x0F]}${HEX[NUM & 0x0F]}"
+ NUM=$((NUM >> 8))
+ done
+}
+
+# Make sure we get possible errors from pipes.
+set -o pipefail
+
+# Get information about the input file. At least older 7z and 7za versions
+# may return with zero exit status even when an error occurred, so check
+# if the output has any lines beginning with "Error".
+INFO=$("$SEVENZIP" l -slt "$1")
+if [ $? != 0 ] || printf '%s\n' "$INFO" | grep -q ^Error; then
+ printf '%s\n' "$INFO"
+ exit 1
+fi
+
+# Check if the input file has more than one compressed block.
+if printf '%s\n' "$INFO" | grep -q '^Block = 1'; then
+ echo "Cannot convert, because the input file has more than"
+ echo "one compressed block."
+ exit 1
+fi
+
+# Get compressed, uncompressed, and dictionary size.
+CSIZE=$(printf '%s\n' "$INFO" | sed -rn 's|^Packed Size = ([0-9]+$)|\1|p')
+USIZE=$(printf '%s\n' "$INFO" | sed -rn 's|^Size = ([0-9]+$)|\1|p')
+DICT=$(printf '%s\n' "$INFO" | sed -rn 's|^Method = LZMA:([0-9]+[bkm]?)$|\1|p')
+
+if [ -z "$CSIZE" -o -z "$USIZE" -o -z "$DICT" ]; then
+ echo "Parsing output of $SEVENZIP failed. Maybe the file uses some"
+ echo "other compression method than plain LZMA."
+ exit 1
+fi
+
+# The following assumes that the default lc, lp, and pb settings were used.
+# Otherwise the output will be corrupt.
+printf '\x5D' > "$2"
+
+# Dictionary size can be either was power of two, bytes, kibibytes, or
+# mebibytes. We need to convert it to bytes.
+case $DICT in
+ *b)
+ DICT=${DICT%b}
+ ;;
+ *k)
+ DICT=${DICT%k}
+ DICT=$((DICT << 10))
+ ;;
+ *m)
+ DICT=${DICT%m}
+ DICT=$((DICT << 20))
+ ;;
+ *)
+ DICT=$((1 << DICT))
+ ;;
+esac
+int2bin 4 "$DICT" >> "$2"
+
+# Uncompressed size
+int2bin 8 "$USIZE" >> "$2"
+
+# Copy the actual compressed data. Using multiple dd commands to avoid
+# copying large amount of data with one-byte block size, which would be
+# annoyingly slow.
+BS=8192
+BIGSIZE=$((CSIZE / BS))
+CSIZE=$((CSIZE % BS))
+{
+ dd of=/dev/null bs=32 count=1 \
+ && dd bs="$BS" count="$BIGSIZE" \
+ && dd bs=1 count="$CSIZE"
+} < "$1" >> "$2"
+
+exit $?
diff --git a/extra/scanlzma/scanlzma.c b/extra/scanlzma/scanlzma.c
new file mode 100644
index 0000000..5c3b25b
--- /dev/null
+++ b/extra/scanlzma/scanlzma.c
@@ -0,0 +1,84 @@
+/*
+ scanlzma, scan for lzma compressed data in stdin and echo it to stdout.
+ Copyright (C) 2006 Timo Lindfors
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+*/
+
+/* Usage example:
+
+ $ wget http://www.wifi-shop.cz/Files/produkty/wa2204/wa2204av1.4.1.zip
+ $ unzip wa2204av1.4.1.zip
+ $ gcc scanlzma.c -o scanlzma -Wall
+ $ ./scanlzma 0 < WA2204-FW1.4.1/linux-1.4.bin | lzma -c -d | strings | grep -i "copyright"
+ UpdateDD version 2.5, Copyright (C) 2005 Philipp Benner.
+ Copyright (C) 2005 Philipp Benner.
+ Copyright (C) 2005 Philipp Benner.
+ mawk 1.3%s%s %s, Copyright (C) Michael D. Brennan
+ # Copyright (C) 1998, 1999, 2001 Henry Spencer.
+ ...
+
+*/
+
+
+/* LZMA compressed file format */
+/* --------------------------- */
+/* Offset Size Description */
+/* 0 1 Special LZMA properties for compressed data */
+/* 1 4 Dictionary size (little endian) */
+/* 5 8 Uncompressed size (little endian). -1 means unknown size */
+/* 13 Compressed data */
+
+#define BUFSIZE 4096
+
+int find_lzma_header(unsigned char *buf) {
+ return (buf[0] < 0xE1
+ && buf[0] == 0x5d
+ && buf[4] < 0x20
+ && (memcmp (buf + 10 , "\x00\x00\x00", 3) == 0
+ || (memcmp (buf + 5, "\xFF\xFF\xFF\xFF\xFF\xFF\xFF\xFF", 8) == 0)));
+}
+
+int main(int argc, char *argv[]) {
+ char buf[BUFSIZE];
+ int ret, i, numlzma, blocks=0;
+
+ if (argc != 2) {
+ printf("usage: %s numlzma < infile | lzma -c -d > outfile\n"
+ "where numlzma is index of lzma file to extract, starting from zero.\n",
+ argv[0]);
+ exit(1);
+ }
+ numlzma = atoi(argv[1]);
+
+ for (;;) {
+ /* Read data. */
+ ret = fread(buf, BUFSIZE, 1, stdin);
+ if (ret != 1)
+ break;
+
+ /* Scan for signature. */
+ for (i = 0; i<BUFSIZE-23; i++) {
+ if (find_lzma_header(buf+i) && numlzma-- <= 0) {
+ fwrite(buf+i, (BUFSIZE-i), 1, stdout);
+ for (;;) {
+ int ch;
+ ch = getchar();
+ if (ch == EOF)
+ exit(0);
+ putchar(ch);
+ }
+ }
+ }
+ blocks++;
+ }
+ return 1;
+}
diff --git a/lib/Makefile.am b/lib/Makefile.am
new file mode 100644
index 0000000..0dbd9c4
--- /dev/null
+++ b/lib/Makefile.am
@@ -0,0 +1,32 @@
+##
+## Copyright (C) 2004-2007 Free Software Foundation, Inc.
+##
+## This program is free software; you can redistribute it and/or modify
+## it under the terms of the GNU General Public License as published by
+## the Free Software Foundation; either version 2 of the License, or
+## (at your option) any later version.
+##
+## This program is distributed in the hope that it will be useful,
+## but WITHOUT ANY WARRANTY; without even the implied warranty of
+## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+## GNU General Public License for more details.
+##
+
+## Not using gnulib-tool, at least for now. It is likely that we won't
+## need anything else from Gnulib than getopt_long().
+
+noinst_LIBRARIES = libgnu.a
+
+libgnu_a_SOURCES =
+libgnu_a_DEPENDENCIES = $(LIBOBJS)
+libgnu_a_LIBADD = $(LIBOBJS)
+
+EXTRA_DIST = getopt.in.h getopt.c getopt1.c getopt_int.h
+BUILT_SOURCES = $(GETOPT_H)
+MOSTLYCLEANFILES = getopt.h getopt.h-t
+
+getopt.h: getopt.in.h
+ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \
+ cat $(srcdir)/getopt.in.h; \
+ } > $@-t
+ mv -f $@-t $@
diff --git a/lib/getopt.c b/lib/getopt.c
new file mode 100644
index 0000000..1d14b24
--- /dev/null
+++ b/lib/getopt.c
@@ -0,0 +1,1197 @@
+/* Getopt for GNU.
+ NOTE: getopt is now part of the C library, so if you don't know what
+ "Keep this file name-space clean" means, talk to drepper@gnu.org
+ before changing it!
+ Copyright (C) 1987,88,89,90,91,92,93,94,95,96,98,99,2000,2001,2002,2003,2004,2006
+ Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License along
+ with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#ifndef _LIBC
+# include <config.h>
+#endif
+
+#include "getopt.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#ifdef __VMS
+# include <unixlib.h>
+#endif
+
+/* Completely disable NLS for getopt. We won't include translations for it
+ anyway. If the system lacks getopt_long, missing translations probably
+ aren't a problem. */
+/*
+#ifdef _LIBC
+# include <libintl.h>
+#else
+# include "gettext.h"
+# define _(msgid) gettext (msgid)
+#endif
+*/
+#define _(msgid) (msgid)
+
+#if defined _LIBC && defined USE_IN_LIBIO
+# include <wchar.h>
+#endif
+
+#ifndef attribute_hidden
+# define attribute_hidden
+#endif
+
+/* Unlike standard Unix `getopt', functions like `getopt_long'
+ let the user intersperse the options with the other arguments.
+
+ As `getopt_long' works, it permutes the elements of ARGV so that,
+ when it is done, all the options precede everything else. Thus
+ all application programs are extended to handle flexible argument order.
+
+ Using `getopt' or setting the environment variable POSIXLY_CORRECT
+ disables permutation.
+ Then the application's behavior is completely standard.
+
+ GNU application programs can use a third alternative mode in which
+ they can distinguish the relative order of options and other arguments. */
+
+#include "getopt_int.h"
+
+/* For communication from `getopt' to the caller.
+ When `getopt' finds an option that takes an argument,
+ the argument value is returned here.
+ Also, when `ordering' is RETURN_IN_ORDER,
+ each non-option ARGV-element is returned here. */
+
+char *optarg;
+
+/* Index in ARGV of the next element to be scanned.
+ This is used for communication to and from the caller
+ and for communication between successive calls to `getopt'.
+
+ On entry to `getopt', zero means this is the first call; initialize.
+
+ When `getopt' returns -1, this is the index of the first of the
+ non-option elements that the caller should itself scan.
+
+ Otherwise, `optind' communicates from one call to the next
+ how much of ARGV has been scanned so far. */
+
+/* 1003.2 says this must be 1 before any call. */
+int optind = 1;
+
+/* Callers store zero here to inhibit the error message
+ for unrecognized options. */
+
+int opterr = 1;
+
+/* Set to an option character which was unrecognized.
+ This must be initialized on some systems to avoid linking in the
+ system's own getopt implementation. */
+
+int optopt = '?';
+
+/* Keep a global copy of all internal members of getopt_data. */
+
+static struct _getopt_data getopt_data;
+
+
+#if defined HAVE_DECL_GETENV && !HAVE_DECL_GETENV
+extern char *getenv ();
+#endif
+
+#ifdef _LIBC
+/* Stored original parameters.
+ XXX This is no good solution. We should rather copy the args so
+ that we can compare them later. But we must not use malloc(3). */
+extern int __libc_argc;
+extern char **__libc_argv;
+
+/* Bash 2.0 gives us an environment variable containing flags
+ indicating ARGV elements that should not be considered arguments. */
+
+# ifdef USE_NONOPTION_FLAGS
+/* Defined in getopt_init.c */
+extern char *__getopt_nonoption_flags;
+# endif
+
+# ifdef USE_NONOPTION_FLAGS
+# define SWAP_FLAGS(ch1, ch2) \
+ if (d->__nonoption_flags_len > 0) \
+ { \
+ char __tmp = __getopt_nonoption_flags[ch1]; \
+ __getopt_nonoption_flags[ch1] = __getopt_nonoption_flags[ch2]; \
+ __getopt_nonoption_flags[ch2] = __tmp; \
+ }
+# else
+# define SWAP_FLAGS(ch1, ch2)
+# endif
+#else /* !_LIBC */
+# define SWAP_FLAGS(ch1, ch2)
+#endif /* _LIBC */
+
+/* Exchange two adjacent subsequences of ARGV.
+ One subsequence is elements [first_nonopt,last_nonopt)
+ which contains all the non-options that have been skipped so far.
+ The other is elements [last_nonopt,optind), which contains all
+ the options processed since those non-options were skipped.
+
+ `first_nonopt' and `last_nonopt' are relocated so that they describe
+ the new indices of the non-options in ARGV after they are moved. */
+
+static void
+exchange (char **argv, struct _getopt_data *d)
+{
+ int bottom = d->__first_nonopt;
+ int middle = d->__last_nonopt;
+ int top = d->optind;
+ char *tem;
+
+ /* Exchange the shorter segment with the far end of the longer segment.
+ That puts the shorter segment into the right place.
+ It leaves the longer segment in the right place overall,
+ but it consists of two parts that need to be swapped next. */
+
+#if defined _LIBC && defined USE_NONOPTION_FLAGS
+ /* First make sure the handling of the `__getopt_nonoption_flags'
+ string can work normally. Our top argument must be in the range
+ of the string. */
+ if (d->__nonoption_flags_len > 0 && top >= d->__nonoption_flags_max_len)
+ {
+ /* We must extend the array. The user plays games with us and
+ presents new arguments. */
+ char *new_str = malloc (top + 1);
+ if (new_str == NULL)
+ d->__nonoption_flags_len = d->__nonoption_flags_max_len = 0;
+ else
+ {
+ memset (__mempcpy (new_str, __getopt_nonoption_flags,
+ d->__nonoption_flags_max_len),
+ '\0', top + 1 - d->__nonoption_flags_max_len);
+ d->__nonoption_flags_max_len = top + 1;
+ __getopt_nonoption_flags = new_str;
+ }
+ }
+#endif
+
+ while (top > middle && middle > bottom)
+ {
+ if (top - middle > middle - bottom)
+ {
+ /* Bottom segment is the short one. */
+ int len = middle - bottom;
+ register int i;
+
+ /* Swap it with the top part of the top segment. */
+ for (i = 0; i < len; i++)
+ {
+ tem = argv[bottom + i];
+ argv[bottom + i] = argv[top - (middle - bottom) + i];
+ argv[top - (middle - bottom) + i] = tem;
+ SWAP_FLAGS (bottom + i, top - (middle - bottom) + i);
+ }
+ /* Exclude the moved bottom segment from further swapping. */
+ top -= len;
+ }
+ else
+ {
+ /* Top segment is the short one. */
+ int len = top - middle;
+ register int i;
+
+ /* Swap it with the bottom part of the bottom segment. */
+ for (i = 0; i < len; i++)
+ {
+ tem = argv[bottom + i];
+ argv[bottom + i] = argv[middle + i];
+ argv[middle + i] = tem;
+ SWAP_FLAGS (bottom + i, middle + i);
+ }
+ /* Exclude the moved top segment from further swapping. */
+ bottom += len;
+ }
+ }
+
+ /* Update records for the slots the non-options now occupy. */
+
+ d->__first_nonopt += (d->optind - d->__last_nonopt);
+ d->__last_nonopt = d->optind;
+}
+
+/* Initialize the internal data when the first call is made. */
+
+static const char *
+_getopt_initialize (int argc, char **argv, const char *optstring,
+ int posixly_correct, struct _getopt_data *d)
+{
+ /* Start processing options with ARGV-element 1 (since ARGV-element 0
+ is the program name); the sequence of previously skipped
+ non-option ARGV-elements is empty. */
+
+ d->__first_nonopt = d->__last_nonopt = d->optind;
+
+ d->__nextchar = NULL;
+
+ d->__posixly_correct = posixly_correct || !!getenv ("POSIXLY_CORRECT");
+
+ /* Determine how to handle the ordering of options and nonoptions. */
+
+ if (optstring[0] == '-')
+ {
+ d->__ordering = RETURN_IN_ORDER;
+ ++optstring;
+ }
+ else if (optstring[0] == '+')
+ {
+ d->__ordering = REQUIRE_ORDER;
+ ++optstring;
+ }
+ else if (d->__posixly_correct)
+ d->__ordering = REQUIRE_ORDER;
+ else
+ d->__ordering = PERMUTE;
+
+#if defined _LIBC && defined USE_NONOPTION_FLAGS
+ if (!d->__posixly_correct
+ && argc == __libc_argc && argv == __libc_argv)
+ {
+ if (d->__nonoption_flags_max_len == 0)
+ {
+ if (__getopt_nonoption_flags == NULL
+ || __getopt_nonoption_flags[0] == '\0')
+ d->__nonoption_flags_max_len = -1;
+ else
+ {
+ const char *orig_str = __getopt_nonoption_flags;
+ int len = d->__nonoption_flags_max_len = strlen (orig_str);
+ if (d->__nonoption_flags_max_len < argc)
+ d->__nonoption_flags_max_len = argc;
+ __getopt_nonoption_flags =
+ (char *) malloc (d->__nonoption_flags_max_len);
+ if (__getopt_nonoption_flags == NULL)
+ d->__nonoption_flags_max_len = -1;
+ else
+ memset (__mempcpy (__getopt_nonoption_flags, orig_str, len),
+ '\0', d->__nonoption_flags_max_len - len);
+ }
+ }
+ d->__nonoption_flags_len = d->__nonoption_flags_max_len;
+ }
+ else
+ d->__nonoption_flags_len = 0;
+#endif
+
+ return optstring;
+}
+
+/* Scan elements of ARGV (whose length is ARGC) for option characters
+ given in OPTSTRING.
+
+ If an element of ARGV starts with '-', and is not exactly "-" or "--",
+ then it is an option element. The characters of this element
+ (aside from the initial '-') are option characters. If `getopt'
+ is called repeatedly, it returns successively each of the option characters
+ from each of the option elements.
+
+ If `getopt' finds another option character, it returns that character,
+ updating `optind' and `nextchar' so that the next call to `getopt' can
+ resume the scan with the following option character or ARGV-element.
+
+ If there are no more option characters, `getopt' returns -1.
+ Then `optind' is the index in ARGV of the first ARGV-element
+ that is not an option. (The ARGV-elements have been permuted
+ so that those that are not options now come last.)
+
+ OPTSTRING is a string containing the legitimate option characters.
+ If an option character is seen that is not listed in OPTSTRING,
+ return '?' after printing an error message. If you set `opterr' to
+ zero, the error message is suppressed but we still return '?'.
+
+ If a char in OPTSTRING is followed by a colon, that means it wants an arg,
+ so the following text in the same ARGV-element, or the text of the following
+ ARGV-element, is returned in `optarg'. Two colons mean an option that
+ wants an optional arg; if there is text in the current ARGV-element,
+ it is returned in `optarg', otherwise `optarg' is set to zero.
+
+ If OPTSTRING starts with `-' or `+', it requests different methods of
+ handling the non-option ARGV-elements.
+ See the comments about RETURN_IN_ORDER and REQUIRE_ORDER, above.
+
+ Long-named options begin with `--' instead of `-'.
+ Their names may be abbreviated as long as the abbreviation is unique
+ or is an exact match for some defined option. If they have an
+ argument, it follows the option name in the same ARGV-element, separated
+ from the option name by a `=', or else the in next ARGV-element.
+ When `getopt' finds a long-named option, it returns 0 if that option's
+ `flag' field is nonzero, the value of the option's `val' field
+ if the `flag' field is zero.
+
+ LONGOPTS is a vector of `struct option' terminated by an
+ element containing a name which is zero.
+
+ LONGIND returns the index in LONGOPT of the long-named option found.
+ It is only valid when a long-named option has been found by the most
+ recent call.
+
+ If LONG_ONLY is nonzero, '-' as well as '--' can introduce
+ long-named options.
+
+ If POSIXLY_CORRECT is nonzero, behave as if the POSIXLY_CORRECT
+ environment variable were set. */
+
+int
+_getopt_internal_r (int argc, char **argv, const char *optstring,
+ const struct option *longopts, int *longind,
+ int long_only, int posixly_correct, struct _getopt_data *d)
+{
+ int print_errors = d->opterr;
+ if (optstring[0] == ':')
+ print_errors = 0;
+
+ if (argc < 1)
+ return -1;
+
+ d->optarg = NULL;
+
+ if (d->optind == 0 || !d->__initialized)
+ {
+ if (d->optind == 0)
+ d->optind = 1; /* Don't scan ARGV[0], the program name. */
+ optstring = _getopt_initialize (argc, argv, optstring,
+ posixly_correct, d);
+ d->__initialized = 1;
+ }
+
+ /* Test whether ARGV[optind] points to a non-option argument.
+ Either it does not have option syntax, or there is an environment flag
+ from the shell indicating it is not an option. The later information
+ is only used when the used in the GNU libc. */
+#if defined _LIBC && defined USE_NONOPTION_FLAGS
+# define NONOPTION_P (argv[d->optind][0] != '-' || argv[d->optind][1] == '\0' \
+ || (d->optind < d->__nonoption_flags_len \
+ && __getopt_nonoption_flags[d->optind] == '1'))
+#else
+# define NONOPTION_P (argv[d->optind][0] != '-' || argv[d->optind][1] == '\0')
+#endif
+
+ if (d->__nextchar == NULL || *d->__nextchar == '\0')
+ {
+ /* Advance to the next ARGV-element. */
+
+ /* Give FIRST_NONOPT & LAST_NONOPT rational values if OPTIND has been
+ moved back by the user (who may also have changed the arguments). */
+ if (d->__last_nonopt > d->optind)
+ d->__last_nonopt = d->optind;
+ if (d->__first_nonopt > d->optind)
+ d->__first_nonopt = d->optind;
+
+ if (d->__ordering == PERMUTE)
+ {
+ /* If we have just processed some options following some non-options,
+ exchange them so that the options come first. */
+
+ if (d->__first_nonopt != d->__last_nonopt
+ && d->__last_nonopt != d->optind)
+ exchange ((char **) argv, d);
+ else if (d->__last_nonopt != d->optind)
+ d->__first_nonopt = d->optind;
+
+ /* Skip any additional non-options
+ and extend the range of non-options previously skipped. */
+
+ while (d->optind < argc && NONOPTION_P)
+ d->optind++;
+ d->__last_nonopt = d->optind;
+ }
+
+ /* The special ARGV-element `--' means premature end of options.
+ Skip it like a null option,
+ then exchange with previous non-options as if it were an option,
+ then skip everything else like a non-option. */
+
+ if (d->optind != argc && !strcmp (argv[d->optind], "--"))
+ {
+ d->optind++;
+
+ if (d->__first_nonopt != d->__last_nonopt
+ && d->__last_nonopt != d->optind)
+ exchange ((char **) argv, d);
+ else if (d->__first_nonopt == d->__last_nonopt)
+ d->__first_nonopt = d->optind;
+ d->__last_nonopt = argc;
+
+ d->optind = argc;
+ }
+
+ /* If we have done all the ARGV-elements, stop the scan
+ and back over any non-options that we skipped and permuted. */
+
+ if (d->optind == argc)
+ {
+ /* Set the next-arg-index to point at the non-options
+ that we previously skipped, so the caller will digest them. */
+ if (d->__first_nonopt != d->__last_nonopt)
+ d->optind = d->__first_nonopt;
+ return -1;
+ }
+
+ /* If we have come to a non-option and did not permute it,
+ either stop the scan or describe it to the caller and pass it by. */
+
+ if (NONOPTION_P)
+ {
+ if (d->__ordering == REQUIRE_ORDER)
+ return -1;
+ d->optarg = argv[d->optind++];
+ return 1;
+ }
+
+ /* We have found another option-ARGV-element.
+ Skip the initial punctuation. */
+
+ d->__nextchar = (argv[d->optind] + 1
+ + (longopts != NULL && argv[d->optind][1] == '-'));
+ }
+
+ /* Decode the current option-ARGV-element. */
+
+ /* Check whether the ARGV-element is a long option.
+
+ If long_only and the ARGV-element has the form "-f", where f is
+ a valid short option, don't consider it an abbreviated form of
+ a long option that starts with f. Otherwise there would be no
+ way to give the -f short option.
+
+ On the other hand, if there's a long option "fubar" and
+ the ARGV-element is "-fu", do consider that an abbreviation of
+ the long option, just like "--fu", and not "-f" with arg "u".
+
+ This distinction seems to be the most useful approach. */
+
+ if (longopts != NULL
+ && (argv[d->optind][1] == '-'
+ || (long_only && (argv[d->optind][2]
+ || !strchr (optstring, argv[d->optind][1])))))
+ {
+ char *nameend;
+ const struct option *p;
+ const struct option *pfound = NULL;
+ int exact = 0;
+ int ambig = 0;
+ int indfound = -1;
+ int option_index;
+
+ for (nameend = d->__nextchar; *nameend && *nameend != '='; nameend++)
+ /* Do nothing. */ ;
+
+ /* Test all long options for either exact match
+ or abbreviated matches. */
+ for (p = longopts, option_index = 0; p->name; p++, option_index++)
+ if (!strncmp (p->name, d->__nextchar, nameend - d->__nextchar))
+ {
+ if ((unsigned int) (nameend - d->__nextchar)
+ == (unsigned int) strlen (p->name))
+ {
+ /* Exact match found. */
+ pfound = p;
+ indfound = option_index;
+ exact = 1;
+ break;
+ }
+ else if (pfound == NULL)
+ {
+ /* First nonexact match found. */
+ pfound = p;
+ indfound = option_index;
+ }
+ else if (long_only
+ || pfound->has_arg != p->has_arg
+ || pfound->flag != p->flag
+ || pfound->val != p->val)
+ /* Second or later nonexact match found. */
+ ambig = 1;
+ }
+
+ if (ambig && !exact)
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf, _("%s: option `%s' is ambiguous\n"),
+ argv[0], argv[d->optind]) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr, _("%s: option `%s' is ambiguous\n"),
+ argv[0], argv[d->optind]);
+#endif
+ }
+ d->__nextchar += strlen (d->__nextchar);
+ d->optind++;
+ d->optopt = 0;
+ return '?';
+ }
+
+ if (pfound != NULL)
+ {
+ option_index = indfound;
+ d->optind++;
+ if (*nameend)
+ {
+ /* Don't test has_arg with >, because some C compilers don't
+ allow it to be used on enums. */
+ if (pfound->has_arg)
+ d->optarg = nameend + 1;
+ else
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+ int n;
+#endif
+
+ if (argv[d->optind - 1][1] == '-')
+ {
+ /* --option */
+#if defined _LIBC && defined USE_IN_LIBIO
+ n = __asprintf (&buf, _("\
+%s: option `--%s' doesn't allow an argument\n"),
+ argv[0], pfound->name);
+#else
+ fprintf (stderr, _("\
+%s: option `--%s' doesn't allow an argument\n"),
+ argv[0], pfound->name);
+#endif
+ }
+ else
+ {
+ /* +option or -option */
+#if defined _LIBC && defined USE_IN_LIBIO
+ n = __asprintf (&buf, _("\
+%s: option `%c%s' doesn't allow an argument\n"),
+ argv[0], argv[d->optind - 1][0],
+ pfound->name);
+#else
+ fprintf (stderr, _("\
+%s: option `%c%s' doesn't allow an argument\n"),
+ argv[0], argv[d->optind - 1][0],
+ pfound->name);
+#endif
+ }
+
+#if defined _LIBC && defined USE_IN_LIBIO
+ if (n >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2
+ |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#endif
+ }
+
+ d->__nextchar += strlen (d->__nextchar);
+
+ d->optopt = pfound->val;
+ return '?';
+ }
+ }
+ else if (pfound->has_arg == 1)
+ {
+ if (d->optind < argc)
+ d->optarg = argv[d->optind++];
+ else
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf, _("\
+%s: option `%s' requires an argument\n"),
+ argv[0], argv[d->optind - 1]) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2
+ |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr,
+ _("%s: option `%s' requires an argument\n"),
+ argv[0], argv[d->optind - 1]);
+#endif
+ }
+ d->__nextchar += strlen (d->__nextchar);
+ d->optopt = pfound->val;
+ return optstring[0] == ':' ? ':' : '?';
+ }
+ }
+ d->__nextchar += strlen (d->__nextchar);
+ if (longind != NULL)
+ *longind = option_index;
+ if (pfound->flag)
+ {
+ *(pfound->flag) = pfound->val;
+ return 0;
+ }
+ return pfound->val;
+ }
+
+ /* Can't find it as a long option. If this is not getopt_long_only,
+ or the option starts with '--' or is not a valid short
+ option, then it's an error.
+ Otherwise interpret it as a short option. */
+ if (!long_only || argv[d->optind][1] == '-'
+ || strchr (optstring, *d->__nextchar) == NULL)
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+ int n;
+#endif
+
+ if (argv[d->optind][1] == '-')
+ {
+ /* --option */
+#if defined _LIBC && defined USE_IN_LIBIO
+ n = __asprintf (&buf, _("%s: unrecognized option `--%s'\n"),
+ argv[0], d->__nextchar);
+#else
+ fprintf (stderr, _("%s: unrecognized option `--%s'\n"),
+ argv[0], d->__nextchar);
+#endif
+ }
+ else
+ {
+ /* +option or -option */
+#if defined _LIBC && defined USE_IN_LIBIO
+ n = __asprintf (&buf, _("%s: unrecognized option `%c%s'\n"),
+ argv[0], argv[d->optind][0], d->__nextchar);
+#else
+ fprintf (stderr, _("%s: unrecognized option `%c%s'\n"),
+ argv[0], argv[d->optind][0], d->__nextchar);
+#endif
+ }
+
+#if defined _LIBC && defined USE_IN_LIBIO
+ if (n >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#endif
+ }
+ d->__nextchar = (char *) "";
+ d->optind++;
+ d->optopt = 0;
+ return '?';
+ }
+ }
+
+ /* Look at and handle the next short option-character. */
+
+ {
+ char c = *d->__nextchar++;
+ char *temp = strchr (optstring, c);
+
+ /* Increment `optind' when we start to process its last character. */
+ if (*d->__nextchar == '\0')
+ ++d->optind;
+
+ if (temp == NULL || c == ':')
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+ int n;
+#endif
+
+ if (d->__posixly_correct)
+ {
+ /* 1003.2 specifies the format of this message. */
+#if defined _LIBC && defined USE_IN_LIBIO
+ n = __asprintf (&buf, _("%s: illegal option -- %c\n"),
+ argv[0], c);
+#else
+ fprintf (stderr, _("%s: illegal option -- %c\n"), argv[0], c);
+#endif
+ }
+ else
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ n = __asprintf (&buf, _("%s: invalid option -- %c\n"),
+ argv[0], c);
+#else
+ fprintf (stderr, _("%s: invalid option -- %c\n"), argv[0], c);
+#endif
+ }
+
+#if defined _LIBC && defined USE_IN_LIBIO
+ if (n >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#endif
+ }
+ d->optopt = c;
+ return '?';
+ }
+ /* Convenience. Treat POSIX -W foo same as long option --foo */
+ if (temp[0] == 'W' && temp[1] == ';')
+ {
+ char *nameend;
+ const struct option *p;
+ const struct option *pfound = NULL;
+ int exact = 0;
+ int ambig = 0;
+ int indfound = 0;
+ int option_index;
+
+ /* This is an option that requires an argument. */
+ if (*d->__nextchar != '\0')
+ {
+ d->optarg = d->__nextchar;
+ /* If we end this ARGV-element by taking the rest as an arg,
+ we must advance to the next element now. */
+ d->optind++;
+ }
+ else if (d->optind == argc)
+ {
+ if (print_errors)
+ {
+ /* 1003.2 specifies the format of this message. */
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf,
+ _("%s: option requires an argument -- %c\n"),
+ argv[0], c) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr, _("%s: option requires an argument -- %c\n"),
+ argv[0], c);
+#endif
+ }
+ d->optopt = c;
+ if (optstring[0] == ':')
+ c = ':';
+ else
+ c = '?';
+ return c;
+ }
+ else
+ /* We already incremented `d->optind' once;
+ increment it again when taking next ARGV-elt as argument. */
+ d->optarg = argv[d->optind++];
+
+ /* optarg is now the argument, see if it's in the
+ table of longopts. */
+
+ for (d->__nextchar = nameend = d->optarg; *nameend && *nameend != '=';
+ nameend++)
+ /* Do nothing. */ ;
+
+ /* Test all long options for either exact match
+ or abbreviated matches. */
+ for (p = longopts, option_index = 0; p->name; p++, option_index++)
+ if (!strncmp (p->name, d->__nextchar, nameend - d->__nextchar))
+ {
+ if ((unsigned int) (nameend - d->__nextchar) == strlen (p->name))
+ {
+ /* Exact match found. */
+ pfound = p;
+ indfound = option_index;
+ exact = 1;
+ break;
+ }
+ else if (pfound == NULL)
+ {
+ /* First nonexact match found. */
+ pfound = p;
+ indfound = option_index;
+ }
+ else
+ /* Second or later nonexact match found. */
+ ambig = 1;
+ }
+ if (ambig && !exact)
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf, _("%s: option `-W %s' is ambiguous\n"),
+ argv[0], argv[d->optind]) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr, _("%s: option `-W %s' is ambiguous\n"),
+ argv[0], argv[d->optind]);
+#endif
+ }
+ d->__nextchar += strlen (d->__nextchar);
+ d->optind++;
+ return '?';
+ }
+ if (pfound != NULL)
+ {
+ option_index = indfound;
+ if (*nameend)
+ {
+ /* Don't test has_arg with >, because some C compilers don't
+ allow it to be used on enums. */
+ if (pfound->has_arg)
+ d->optarg = nameend + 1;
+ else
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf, _("\
+%s: option `-W %s' doesn't allow an argument\n"),
+ argv[0], pfound->name) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2
+ |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr, _("\
+%s: option `-W %s' doesn't allow an argument\n"),
+ argv[0], pfound->name);
+#endif
+ }
+
+ d->__nextchar += strlen (d->__nextchar);
+ return '?';
+ }
+ }
+ else if (pfound->has_arg == 1)
+ {
+ if (d->optind < argc)
+ d->optarg = argv[d->optind++];
+ else
+ {
+ if (print_errors)
+ {
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf, _("\
+%s: option `%s' requires an argument\n"),
+ argv[0], argv[d->optind - 1]) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2
+ |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr,
+ _("%s: option `%s' requires an argument\n"),
+ argv[0], argv[d->optind - 1]);
+#endif
+ }
+ d->__nextchar += strlen (d->__nextchar);
+ return optstring[0] == ':' ? ':' : '?';
+ }
+ }
+ d->__nextchar += strlen (d->__nextchar);
+ if (longind != NULL)
+ *longind = option_index;
+ if (pfound->flag)
+ {
+ *(pfound->flag) = pfound->val;
+ return 0;
+ }
+ return pfound->val;
+ }
+ d->__nextchar = NULL;
+ return 'W'; /* Let the application handle it. */
+ }
+ if (temp[1] == ':')
+ {
+ if (temp[2] == ':')
+ {
+ /* This is an option that accepts an argument optionally. */
+ if (*d->__nextchar != '\0')
+ {
+ d->optarg = d->__nextchar;
+ d->optind++;
+ }
+ else
+ d->optarg = NULL;
+ d->__nextchar = NULL;
+ }
+ else
+ {
+ /* This is an option that requires an argument. */
+ if (*d->__nextchar != '\0')
+ {
+ d->optarg = d->__nextchar;
+ /* If we end this ARGV-element by taking the rest as an arg,
+ we must advance to the next element now. */
+ d->optind++;
+ }
+ else if (d->optind == argc)
+ {
+ if (print_errors)
+ {
+ /* 1003.2 specifies the format of this message. */
+#if defined _LIBC && defined USE_IN_LIBIO
+ char *buf;
+
+ if (__asprintf (&buf, _("\
+%s: option requires an argument -- %c\n"),
+ argv[0], c) >= 0)
+ {
+ _IO_flockfile (stderr);
+
+ int old_flags2 = ((_IO_FILE *) stderr)->_flags2;
+ ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL;
+
+ __fxprintf (NULL, "%s", buf);
+
+ ((_IO_FILE *) stderr)->_flags2 = old_flags2;
+ _IO_funlockfile (stderr);
+
+ free (buf);
+ }
+#else
+ fprintf (stderr,
+ _("%s: option requires an argument -- %c\n"),
+ argv[0], c);
+#endif
+ }
+ d->optopt = c;
+ if (optstring[0] == ':')
+ c = ':';
+ else
+ c = '?';
+ }
+ else
+ /* We already incremented `optind' once;
+ increment it again when taking next ARGV-elt as argument. */
+ d->optarg = argv[d->optind++];
+ d->__nextchar = NULL;
+ }
+ }
+ return c;
+ }
+}
+
+int
+_getopt_internal (int argc, char **argv, const char *optstring,
+ const struct option *longopts, int *longind,
+ int long_only, int posixly_correct)
+{
+ int result;
+
+ getopt_data.optind = optind;
+ getopt_data.opterr = opterr;
+
+ result = _getopt_internal_r (argc, argv, optstring, longopts, longind,
+ long_only, posixly_correct, &getopt_data);
+
+ optind = getopt_data.optind;
+ optarg = getopt_data.optarg;
+ optopt = getopt_data.optopt;
+
+ return result;
+}
+
+/* glibc gets a LSB-compliant getopt.
+ Standalone applications get a POSIX-compliant getopt. */
+#if _LIBC
+enum { POSIXLY_CORRECT = 0 };
+#else
+enum { POSIXLY_CORRECT = 1 };
+#endif
+
+int
+getopt (int argc, char *const *argv, const char *optstring)
+{
+ return _getopt_internal (argc, (char **) argv, optstring, NULL, NULL, 0,
+ POSIXLY_CORRECT);
+}
+
+
+#ifdef TEST
+
+/* Compile with -DTEST to make an executable for use in testing
+ the above definition of `getopt'. */
+
+int
+main (int argc, char **argv)
+{
+ int c;
+ int digit_optind = 0;
+
+ while (1)
+ {
+ int this_option_optind = optind ? optind : 1;
+
+ c = getopt (argc, argv, "abc:d:0123456789");
+ if (c == -1)
+ break;
+
+ switch (c)
+ {
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ if (digit_optind != 0 && digit_optind != this_option_optind)
+ printf ("digits occur in two different argv-elements.\n");
+ digit_optind = this_option_optind;
+ printf ("option %c\n", c);
+ break;
+
+ case 'a':
+ printf ("option a\n");
+ break;
+
+ case 'b':
+ printf ("option b\n");
+ break;
+
+ case 'c':
+ printf ("option c with value `%s'\n", optarg);
+ break;
+
+ case '?':
+ break;
+
+ default:
+ printf ("?? getopt returned character code 0%o ??\n", c);
+ }
+ }
+
+ if (optind < argc)
+ {
+ printf ("non-option ARGV-elements: ");
+ while (optind < argc)
+ printf ("%s ", argv[optind++]);
+ printf ("\n");
+ }
+
+ exit (0);
+}
+
+#endif /* TEST */
diff --git a/lib/getopt.in.h b/lib/getopt.in.h
new file mode 100644
index 0000000..ea77e3d
--- /dev/null
+++ b/lib/getopt.in.h
@@ -0,0 +1,226 @@
+/* Declarations for getopt.
+ Copyright (C) 1989-1994,1996-1999,2001,2003,2004,2005,2006,2007
+ Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License along
+ with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#ifndef _GETOPT_H
+
+#ifndef __need_getopt
+# define _GETOPT_H 1
+#endif
+
+/* Standalone applications should #define __GETOPT_PREFIX to an
+ identifier that prefixes the external functions and variables
+ defined in this header. When this happens, include the
+ headers that might declare getopt so that they will not cause
+ confusion if included after this file. Then systematically rename
+ identifiers so that they do not collide with the system functions
+ and variables. Renaming avoids problems with some compilers and
+ linkers. */
+#if defined __GETOPT_PREFIX && !defined __need_getopt
+# include <stdlib.h>
+# include <stdio.h>
+# include <unistd.h>
+# undef __need_getopt
+# undef getopt
+# undef getopt_long
+# undef getopt_long_only
+# undef optarg
+# undef opterr
+# undef optind
+# undef optopt
+# define __GETOPT_CONCAT(x, y) x ## y
+# define __GETOPT_XCONCAT(x, y) __GETOPT_CONCAT (x, y)
+# define __GETOPT_ID(y) __GETOPT_XCONCAT (__GETOPT_PREFIX, y)
+# define getopt __GETOPT_ID (getopt)
+# define getopt_long __GETOPT_ID (getopt_long)
+# define getopt_long_only __GETOPT_ID (getopt_long_only)
+# define optarg __GETOPT_ID (optarg)
+# define opterr __GETOPT_ID (opterr)
+# define optind __GETOPT_ID (optind)
+# define optopt __GETOPT_ID (optopt)
+#endif
+
+/* Standalone applications get correct prototypes for getopt_long and
+ getopt_long_only; they declare "char **argv". libc uses prototypes
+ with "char *const *argv" that are incorrect because getopt_long and
+ getopt_long_only can permute argv; this is required for backward
+ compatibility (e.g., for LSB 2.0.1).
+
+ This used to be `#if defined __GETOPT_PREFIX && !defined __need_getopt',
+ but it caused redefinition warnings if both unistd.h and getopt.h were
+ included, since unistd.h includes getopt.h having previously defined
+ __need_getopt.
+
+ The only place where __getopt_argv_const is used is in definitions
+ of getopt_long and getopt_long_only below, but these are visible
+ only if __need_getopt is not defined, so it is quite safe to rewrite
+ the conditional as follows:
+*/
+#if !defined __need_getopt
+# if defined __GETOPT_PREFIX
+# define __getopt_argv_const /* empty */
+# else
+# define __getopt_argv_const const
+# endif
+#endif
+
+/* If __GNU_LIBRARY__ is not already defined, either we are being used
+ standalone, or this is the first header included in the source file.
+ If we are being used with glibc, we need to include <features.h>, but
+ that does not exist if we are standalone. So: if __GNU_LIBRARY__ is
+ not defined, include <ctype.h>, which will pull in <features.h> for us
+ if it's from glibc. (Why ctype.h? It's guaranteed to exist and it
+ doesn't flood the namespace with stuff the way some other headers do.) */
+#if !defined __GNU_LIBRARY__
+# include <ctype.h>
+#endif
+
+#ifndef __THROW
+# ifndef __GNUC_PREREQ
+# define __GNUC_PREREQ(maj, min) (0)
+# endif
+# if defined __cplusplus && __GNUC_PREREQ (2,8)
+# define __THROW throw ()
+# else
+# define __THROW
+# endif
+#endif
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* For communication from `getopt' to the caller.
+ When `getopt' finds an option that takes an argument,
+ the argument value is returned here.
+ Also, when `ordering' is RETURN_IN_ORDER,
+ each non-option ARGV-element is returned here. */
+
+extern char *optarg;
+
+/* Index in ARGV of the next element to be scanned.
+ This is used for communication to and from the caller
+ and for communication between successive calls to `getopt'.
+
+ On entry to `getopt', zero means this is the first call; initialize.
+
+ When `getopt' returns -1, this is the index of the first of the
+ non-option elements that the caller should itself scan.
+
+ Otherwise, `optind' communicates from one call to the next
+ how much of ARGV has been scanned so far. */
+
+extern int optind;
+
+/* Callers store zero here to inhibit the error message `getopt' prints
+ for unrecognized options. */
+
+extern int opterr;
+
+/* Set to an option character which was unrecognized. */
+
+extern int optopt;
+
+#ifndef __need_getopt
+/* Describe the long-named options requested by the application.
+ The LONG_OPTIONS argument to getopt_long or getopt_long_only is a vector
+ of `struct option' terminated by an element containing a name which is
+ zero.
+
+ The field `has_arg' is:
+ no_argument (or 0) if the option does not take an argument,
+ required_argument (or 1) if the option requires an argument,
+ optional_argument (or 2) if the option takes an optional argument.
+
+ If the field `flag' is not NULL, it points to a variable that is set
+ to the value given in the field `val' when the option is found, but
+ left unchanged if the option is not found.
+
+ To have a long-named option do something other than set an `int' to
+ a compiled-in constant, such as set a value from `optarg', set the
+ option's `flag' field to zero and its `val' field to a nonzero
+ value (the equivalent single-letter option character, if there is
+ one). For long options that have a zero `flag' field, `getopt'
+ returns the contents of the `val' field. */
+
+struct option
+{
+ const char *name;
+ /* has_arg can't be an enum because some compilers complain about
+ type mismatches in all the code that assumes it is an int. */
+ int has_arg;
+ int *flag;
+ int val;
+};
+
+/* Names for the values of the `has_arg' field of `struct option'. */
+
+# define no_argument 0
+# define required_argument 1
+# define optional_argument 2
+#endif /* need getopt */
+
+
+/* Get definitions and prototypes for functions to process the
+ arguments in ARGV (ARGC of them, minus the program name) for
+ options given in OPTS.
+
+ Return the option character from OPTS just read. Return -1 when
+ there are no more options. For unrecognized options, or options
+ missing arguments, `optopt' is set to the option letter, and '?' is
+ returned.
+
+ The OPTS string is a list of characters which are recognized option
+ letters, optionally followed by colons, specifying that that letter
+ takes an argument, to be placed in `optarg'.
+
+ If a letter in OPTS is followed by two colons, its argument is
+ optional. This behavior is specific to the GNU `getopt'.
+
+ The argument `--' causes premature termination of argument
+ scanning, explicitly telling `getopt' that there are no more
+ options.
+
+ If OPTS begins with `-', then non-option arguments are treated as
+ arguments to the option '\1'. This behavior is specific to the GNU
+ `getopt'. If OPTS begins with `+', or POSIXLY_CORRECT is set in
+ the environment, then do not permute arguments. */
+
+extern int getopt (int ___argc, char *const *___argv, const char *__shortopts)
+ __THROW;
+
+#ifndef __need_getopt
+extern int getopt_long (int ___argc, char *__getopt_argv_const *___argv,
+ const char *__shortopts,
+ const struct option *__longopts, int *__longind)
+ __THROW;
+extern int getopt_long_only (int ___argc, char *__getopt_argv_const *___argv,
+ const char *__shortopts,
+ const struct option *__longopts, int *__longind)
+ __THROW;
+
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+/* Make sure we later can get all the definitions and declarations. */
+#undef __need_getopt
+
+#endif /* getopt.h */
diff --git a/lib/getopt1.c b/lib/getopt1.c
new file mode 100644
index 0000000..da5d533
--- /dev/null
+++ b/lib/getopt1.c
@@ -0,0 +1,171 @@
+/* getopt_long and getopt_long_only entry points for GNU getopt.
+ Copyright (C) 1987,88,89,90,91,92,93,94,96,97,98,2004,2006
+ Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License along
+ with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#ifdef _LIBC
+# include <getopt.h>
+#else
+# include <config.h>
+# include "getopt.h"
+#endif
+#include "getopt_int.h"
+
+#include <stdio.h>
+
+/* This needs to come after some library #include
+ to get __GNU_LIBRARY__ defined. */
+#ifdef __GNU_LIBRARY__
+#include <stdlib.h>
+#endif
+
+#ifndef NULL
+#define NULL 0
+#endif
+
+int
+getopt_long (int argc, char *__getopt_argv_const *argv, const char *options,
+ const struct option *long_options, int *opt_index)
+{
+ return _getopt_internal (argc, (char **) argv, options, long_options,
+ opt_index, 0, 0);
+}
+
+int
+_getopt_long_r (int argc, char **argv, const char *options,
+ const struct option *long_options, int *opt_index,
+ struct _getopt_data *d)
+{
+ return _getopt_internal_r (argc, argv, options, long_options, opt_index,
+ 0, 0, d);
+}
+
+/* Like getopt_long, but '-' as well as '--' can indicate a long option.
+ If an option that starts with '-' (not '--') doesn't match a long option,
+ but does match a short option, it is parsed as a short option
+ instead. */
+
+int
+getopt_long_only (int argc, char *__getopt_argv_const *argv,
+ const char *options,
+ const struct option *long_options, int *opt_index)
+{
+ return _getopt_internal (argc, (char **) argv, options, long_options,
+ opt_index, 1, 0);
+}
+
+int
+_getopt_long_only_r (int argc, char **argv, const char *options,
+ const struct option *long_options, int *opt_index,
+ struct _getopt_data *d)
+{
+ return _getopt_internal_r (argc, argv, options, long_options, opt_index,
+ 1, 0, d);
+}
+
+
+#ifdef TEST
+
+#include <stdio.h>
+
+int
+main (int argc, char **argv)
+{
+ int c;
+ int digit_optind = 0;
+
+ while (1)
+ {
+ int this_option_optind = optind ? optind : 1;
+ int option_index = 0;
+ static struct option long_options[] =
+ {
+ {"add", 1, 0, 0},
+ {"append", 0, 0, 0},
+ {"delete", 1, 0, 0},
+ {"verbose", 0, 0, 0},
+ {"create", 0, 0, 0},
+ {"file", 1, 0, 0},
+ {0, 0, 0, 0}
+ };
+
+ c = getopt_long (argc, argv, "abc:d:0123456789",
+ long_options, &option_index);
+ if (c == -1)
+ break;
+
+ switch (c)
+ {
+ case 0:
+ printf ("option %s", long_options[option_index].name);
+ if (optarg)
+ printf (" with arg %s", optarg);
+ printf ("\n");
+ break;
+
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ if (digit_optind != 0 && digit_optind != this_option_optind)
+ printf ("digits occur in two different argv-elements.\n");
+ digit_optind = this_option_optind;
+ printf ("option %c\n", c);
+ break;
+
+ case 'a':
+ printf ("option a\n");
+ break;
+
+ case 'b':
+ printf ("option b\n");
+ break;
+
+ case 'c':
+ printf ("option c with value `%s'\n", optarg);
+ break;
+
+ case 'd':
+ printf ("option d with value `%s'\n", optarg);
+ break;
+
+ case '?':
+ break;
+
+ default:
+ printf ("?? getopt returned character code 0%o ??\n", c);
+ }
+ }
+
+ if (optind < argc)
+ {
+ printf ("non-option ARGV-elements: ");
+ while (optind < argc)
+ printf ("%s ", argv[optind++]);
+ printf ("\n");
+ }
+
+ exit (0);
+}
+
+#endif /* TEST */
diff --git a/lib/getopt_int.h b/lib/getopt_int.h
new file mode 100644
index 0000000..401579f
--- /dev/null
+++ b/lib/getopt_int.h
@@ -0,0 +1,131 @@
+/* Internal declarations for getopt.
+ Copyright (C) 1989-1994,1996-1999,2001,2003,2004
+ Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License along
+ with this program; if not, write to the Free Software Foundation,
+ Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
+
+#ifndef _GETOPT_INT_H
+#define _GETOPT_INT_H 1
+
+extern int _getopt_internal (int ___argc, char **___argv,
+ const char *__shortopts,
+ const struct option *__longopts, int *__longind,
+ int __long_only, int __posixly_correct);
+
+
+/* Reentrant versions which can handle parsing multiple argument
+ vectors at the same time. */
+
+/* Data type for reentrant functions. */
+struct _getopt_data
+{
+ /* These have exactly the same meaning as the corresponding global
+ variables, except that they are used for the reentrant
+ versions of getopt. */
+ int optind;
+ int opterr;
+ int optopt;
+ char *optarg;
+
+ /* Internal members. */
+
+ /* True if the internal members have been initialized. */
+ int __initialized;
+
+ /* The next char to be scanned in the option-element
+ in which the last option character we returned was found.
+ This allows us to pick up the scan where we left off.
+
+ If this is zero, or a null string, it means resume the scan
+ by advancing to the next ARGV-element. */
+ char *__nextchar;
+
+ /* Describe how to deal with options that follow non-option ARGV-elements.
+
+ If the caller did not specify anything,
+ the default is REQUIRE_ORDER if the environment variable
+ POSIXLY_CORRECT is defined, PERMUTE otherwise.
+
+ REQUIRE_ORDER means don't recognize them as options;
+ stop option processing when the first non-option is seen.
+ This is what Unix does.
+ This mode of operation is selected by either setting the environment
+ variable POSIXLY_CORRECT, or using `+' as the first character
+ of the list of option characters, or by calling getopt.
+
+ PERMUTE is the default. We permute the contents of ARGV as we
+ scan, so that eventually all the non-options are at the end.
+ This allows options to be given in any order, even with programs
+ that were not written to expect this.
+
+ RETURN_IN_ORDER is an option available to programs that were
+ written to expect options and other ARGV-elements in any order
+ and that care about the ordering of the two. We describe each
+ non-option ARGV-element as if it were the argument of an option
+ with character code 1. Using `-' as the first character of the
+ list of option characters selects this mode of operation.
+
+ The special argument `--' forces an end of option-scanning regardless
+ of the value of `ordering'. In the case of RETURN_IN_ORDER, only
+ `--' can cause `getopt' to return -1 with `optind' != ARGC. */
+
+ enum
+ {
+ REQUIRE_ORDER, PERMUTE, RETURN_IN_ORDER
+ } __ordering;
+
+ /* If the POSIXLY_CORRECT environment variable is set
+ or getopt was called. */
+ int __posixly_correct;
+
+
+ /* Handle permutation of arguments. */
+
+ /* Describe the part of ARGV that contains non-options that have
+ been skipped. `first_nonopt' is the index in ARGV of the first
+ of them; `last_nonopt' is the index after the last of them. */
+
+ int __first_nonopt;
+ int __last_nonopt;
+
+#if defined _LIBC && defined USE_NONOPTION_FLAGS
+ int __nonoption_flags_max_len;
+ int __nonoption_flags_len;
+# endif
+};
+
+/* The initializer is necessary to set OPTIND and OPTERR to their
+ default values and to clear the initialization flag. */
+#define _GETOPT_DATA_INITIALIZER { 1, 1 }
+
+extern int _getopt_internal_r (int ___argc, char **___argv,
+ const char *__shortopts,
+ const struct option *__longopts, int *__longind,
+ int __long_only, int __posixly_correct,
+ struct _getopt_data *__data);
+
+extern int _getopt_long_r (int ___argc, char **___argv,
+ const char *__shortopts,
+ const struct option *__longopts, int *__longind,
+ struct _getopt_data *__data);
+
+extern int _getopt_long_only_r (int ___argc, char **___argv,
+ const char *__shortopts,
+ const struct option *__longopts,
+ int *__longind,
+ struct _getopt_data *__data);
+
+#endif /* getopt_int.h */
diff --git a/m4/.gitignore b/m4/.gitignore
new file mode 100644
index 0000000..6e6c301
--- /dev/null
+++ b/m4/.gitignore
@@ -0,0 +1,35 @@
+codeset.m4
+gettext.m4
+glibc2.m4
+glibc21.m4
+iconv.m4
+intdiv0.m4
+intl.m4
+intldir.m4
+intmax.m4
+inttypes-pri.m4
+inttypes_h.m4
+lcmessage.m4
+lib-ld.m4
+lib-link.m4
+lib-prefix.m4
+libtool.m4
+lock.m4
+longdouble.m4
+longlong.m4
+ltoptions.m4
+ltsugar.m4
+ltversion.m4
+lt~obsolete.m4
+nls.m4
+po.m4
+printf-posix.m4
+progtest.m4
+size_max.m4
+stdint_h.m4
+uintmax_t.m4
+ulonglong.m4
+visibility.m4
+wchar_t.m4
+wint_t.m4
+xsize.m4
diff --git a/m4/acx_pthread.m4 b/m4/acx_pthread.m4
new file mode 100644
index 0000000..d2b1169
--- /dev/null
+++ b/m4/acx_pthread.m4
@@ -0,0 +1,279 @@
+##### http://autoconf-archive.cryp.to/acx_pthread.html
+#
+# SYNOPSIS
+#
+# ACX_PTHREAD([ACTION-IF-FOUND[, ACTION-IF-NOT-FOUND]])
+#
+# DESCRIPTION
+#
+# This macro figures out how to build C programs using POSIX threads.
+# It sets the PTHREAD_LIBS output variable to the threads library and
+# linker flags, and the PTHREAD_CFLAGS output variable to any special
+# C compiler flags that are needed. (The user can also force certain
+# compiler flags/libs to be tested by setting these environment
+# variables.)
+#
+# Also sets PTHREAD_CC to any special C compiler that is needed for
+# multi-threaded programs (defaults to the value of CC otherwise).
+# (This is necessary on AIX to use the special cc_r compiler alias.)
+#
+# NOTE: You are assumed to not only compile your program with these
+# flags, but also link it with them as well. e.g. you should link
+# with $PTHREAD_CC $CFLAGS $PTHREAD_CFLAGS $LDFLAGS ... $PTHREAD_LIBS
+# $LIBS
+#
+# If you are only building threads programs, you may wish to use
+# these variables in your default LIBS, CFLAGS, and CC:
+#
+# LIBS="$PTHREAD_LIBS $LIBS"
+# CFLAGS="$CFLAGS $PTHREAD_CFLAGS"
+# CC="$PTHREAD_CC"
+#
+# In addition, if the PTHREAD_CREATE_JOINABLE thread-attribute
+# constant has a nonstandard name, defines PTHREAD_CREATE_JOINABLE to
+# that name (e.g. PTHREAD_CREATE_UNDETACHED on AIX).
+#
+# ACTION-IF-FOUND is a list of shell commands to run if a threads
+# library is found, and ACTION-IF-NOT-FOUND is a list of commands to
+# run it if it is not found. If ACTION-IF-FOUND is not specified, the
+# default action will define HAVE_PTHREAD.
+#
+# Please let the authors know if this macro fails on any platform, or
+# if you have any other suggestions or comments. This macro was based
+# on work by SGJ on autoconf scripts for FFTW (http://www.fftw.org/)
+# (with help from M. Frigo), as well as ac_pthread and hb_pthread
+# macros posted by Alejandro Forero Cuervo to the autoconf macro
+# repository. We are also grateful for the helpful feedback of
+# numerous users.
+#
+# LAST MODIFICATION
+#
+# 2007-07-29
+#
+# COPYLEFT
+#
+# Copyright (c) 2007 Steven G. Johnson <stevenj@alum.mit.edu>
+#
+# This program is free software: you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful, but
+# WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see
+# <http://www.gnu.org/licenses/>.
+#
+# As a special exception, the respective Autoconf Macro's copyright
+# owner gives unlimited permission to copy, distribute and modify the
+# configure scripts that are the output of Autoconf when processing
+# the Macro. You need not follow the terms of the GNU General Public
+# License when using or distributing such scripts, even though
+# portions of the text of the Macro appear in them. The GNU General
+# Public License (GPL) does govern all other use of the material that
+# constitutes the Autoconf Macro.
+#
+# This special exception to the GPL applies to versions of the
+# Autoconf Macro released by the Autoconf Macro Archive. When you
+# make and distribute a modified version of the Autoconf Macro, you
+# may extend this special exception to the GPL to apply to your
+# modified version as well.
+
+AC_DEFUN([ACX_PTHREAD], [
+AC_REQUIRE([AC_CANONICAL_HOST])
+AC_LANG_SAVE
+AC_LANG_C
+acx_pthread_ok=no
+
+# We used to check for pthread.h first, but this fails if pthread.h
+# requires special compiler flags (e.g. on True64 or Sequent).
+# It gets checked for in the link test anyway.
+
+# First of all, check if the user has set any of the PTHREAD_LIBS,
+# etcetera environment variables, and if threads linking works using
+# them:
+if test x"$PTHREAD_LIBS$PTHREAD_CFLAGS" != x; then
+ save_CFLAGS="$CFLAGS"
+ CFLAGS="$CFLAGS $PTHREAD_CFLAGS"
+ save_LIBS="$LIBS"
+ LIBS="$PTHREAD_LIBS $LIBS"
+ AC_MSG_CHECKING([for pthread_join in LIBS=$PTHREAD_LIBS with CFLAGS=$PTHREAD_CFLAGS])
+ AC_TRY_LINK_FUNC(pthread_join, acx_pthread_ok=yes)
+ AC_MSG_RESULT($acx_pthread_ok)
+ if test x"$acx_pthread_ok" = xno; then
+ PTHREAD_LIBS=""
+ PTHREAD_CFLAGS=""
+ fi
+ LIBS="$save_LIBS"
+ CFLAGS="$save_CFLAGS"
+fi
+
+# We must check for the threads library under a number of different
+# names; the ordering is very important because some systems
+# (e.g. DEC) have both -lpthread and -lpthreads, where one of the
+# libraries is broken (non-POSIX).
+
+# Create a list of thread flags to try. Items starting with a "-" are
+# C compiler flags, and other items are library names, except for "none"
+# which indicates that we try without any flags at all, and "pthread-config"
+# which is a program returning the flags for the Pth emulation library.
+
+acx_pthread_flags="pthreads none -Kthread -kthread lthread -pthread -pthreads -mthreads pthread --thread-safe -mt pthread-config"
+
+# The ordering *is* (sometimes) important. Some notes on the
+# individual items follow:
+
+# pthreads: AIX (must check this before -lpthread)
+# none: in case threads are in libc; should be tried before -Kthread and
+# other compiler flags to prevent continual compiler warnings
+# -Kthread: Sequent (threads in libc, but -Kthread needed for pthread.h)
+# -kthread: FreeBSD kernel threads (preferred to -pthread since SMP-able)
+# lthread: LinuxThreads port on FreeBSD (also preferred to -pthread)
+# -pthread: Linux/gcc (kernel threads), BSD/gcc (userland threads)
+# -pthreads: Solaris/gcc
+# -mthreads: Mingw32/gcc, Lynx/gcc
+# -mt: Sun Workshop C (may only link SunOS threads [-lthread], but it
+# doesn't hurt to check since this sometimes defines pthreads too;
+# also defines -D_REENTRANT)
+# ... -mt is also the pthreads flag for HP/aCC
+# pthread: Linux, etcetera
+# --thread-safe: KAI C++
+# pthread-config: use pthread-config program (for GNU Pth library)
+
+case "${host_cpu}-${host_os}" in
+ *solaris*)
+
+ # On Solaris (at least, for some versions), libc contains stubbed
+ # (non-functional) versions of the pthreads routines, so link-based
+ # tests will erroneously succeed. (We need to link with -pthreads/-mt/
+ # -lpthread.) (The stubs are missing pthread_cleanup_push, or rather
+ # a function called by this macro, so we could check for that, but
+ # who knows whether they'll stub that too in a future libc.) So,
+ # we'll just look for -pthreads and -lpthread first:
+
+ acx_pthread_flags="-pthreads pthread -mt -pthread $acx_pthread_flags"
+ ;;
+esac
+
+if test x"$acx_pthread_ok" = xno; then
+for flag in $acx_pthread_flags; do
+
+ case $flag in
+ none)
+ AC_MSG_CHECKING([whether pthreads work without any flags])
+ ;;
+
+ -*)
+ AC_MSG_CHECKING([whether pthreads work with $flag])
+ PTHREAD_CFLAGS="$flag"
+ ;;
+
+ pthread-config)
+ AC_CHECK_PROG(acx_pthread_config, pthread-config, yes, no)
+ if test x"$acx_pthread_config" = xno; then continue; fi
+ PTHREAD_CFLAGS="`pthread-config --cflags`"
+ PTHREAD_LIBS="`pthread-config --ldflags` `pthread-config --libs`"
+ ;;
+
+ *)
+ AC_MSG_CHECKING([for the pthreads library -l$flag])
+ PTHREAD_LIBS="-l$flag"
+ ;;
+ esac
+
+ save_LIBS="$LIBS"
+ save_CFLAGS="$CFLAGS"
+ LIBS="$PTHREAD_LIBS $LIBS"
+ CFLAGS="$CFLAGS $PTHREAD_CFLAGS"
+
+ # Check for various functions. We must include pthread.h,
+ # since some functions may be macros. (On the Sequent, we
+ # need a special flag -Kthread to make this header compile.)
+ # We check for pthread_join because it is in -lpthread on IRIX
+ # while pthread_create is in libc. We check for pthread_attr_init
+ # due to DEC craziness with -lpthreads. We check for
+ # pthread_cleanup_push because it is one of the few pthread
+ # functions on Solaris that doesn't have a non-functional libc stub.
+ # We try pthread_create on general principles.
+ AC_TRY_LINK([#include <pthread.h>],
+ [pthread_t th; pthread_join(th, 0);
+ pthread_attr_init(0); pthread_cleanup_push(0, 0);
+ pthread_create(0,0,0,0); pthread_cleanup_pop(0); ],
+ [acx_pthread_ok=yes])
+
+ LIBS="$save_LIBS"
+ CFLAGS="$save_CFLAGS"
+
+ AC_MSG_RESULT($acx_pthread_ok)
+ if test "x$acx_pthread_ok" = xyes; then
+ break;
+ fi
+
+ PTHREAD_LIBS=""
+ PTHREAD_CFLAGS=""
+done
+fi
+
+# Various other checks:
+if test "x$acx_pthread_ok" = xyes; then
+ save_LIBS="$LIBS"
+ LIBS="$PTHREAD_LIBS $LIBS"
+ save_CFLAGS="$CFLAGS"
+ CFLAGS="$CFLAGS $PTHREAD_CFLAGS"
+
+ # Detect AIX lossage: JOINABLE attribute is called UNDETACHED.
+ AC_MSG_CHECKING([for joinable pthread attribute])
+ attr_name=unknown
+ for attr in PTHREAD_CREATE_JOINABLE PTHREAD_CREATE_UNDETACHED; do
+ AC_TRY_LINK([#include <pthread.h>], [int attr=$attr; return attr;],
+ [attr_name=$attr; break])
+ done
+ AC_MSG_RESULT($attr_name)
+ if test "$attr_name" != PTHREAD_CREATE_JOINABLE; then
+ AC_DEFINE_UNQUOTED(PTHREAD_CREATE_JOINABLE, $attr_name,
+ [Define to necessary symbol if this constant
+ uses a non-standard name on your system.])
+ fi
+
+ AC_MSG_CHECKING([if more special flags are required for pthreads])
+ flag=no
+ case "${host_cpu}-${host_os}" in
+ *-aix* | *-freebsd* | *-darwin*) flag="-D_THREAD_SAFE";;
+ *solaris* | *-osf* | *-hpux*) flag="-D_REENTRANT";;
+ esac
+ AC_MSG_RESULT(${flag})
+ if test "x$flag" != xno; then
+ PTHREAD_CFLAGS="$flag $PTHREAD_CFLAGS"
+ fi
+
+ LIBS="$save_LIBS"
+ CFLAGS="$save_CFLAGS"
+
+ # More AIX lossage: must compile with xlc_r or cc_r
+ if test x"$GCC" != xyes; then
+ AC_CHECK_PROGS(PTHREAD_CC, xlc_r cc_r, ${CC})
+ else
+ PTHREAD_CC=$CC
+ fi
+else
+ PTHREAD_CC="$CC"
+fi
+
+AC_SUBST(PTHREAD_LIBS)
+AC_SUBST(PTHREAD_CFLAGS)
+AC_SUBST(PTHREAD_CC)
+
+# Finally, execute ACTION-IF-FOUND/ACTION-IF-NOT-FOUND:
+if test x"$acx_pthread_ok" = xyes; then
+ ifelse([$1],,AC_DEFINE(HAVE_PTHREAD,1,[Define if you have POSIX threads libraries and header files.]),[$1])
+ :
+else
+ acx_pthread_ok=no
+ $2
+fi
+AC_LANG_RESTORE
+])dnl ACX_PTHREAD
diff --git a/m4/getopt.m4 b/m4/getopt.m4
new file mode 100644
index 0000000..cfbe40f
--- /dev/null
+++ b/m4/getopt.m4
@@ -0,0 +1,71 @@
+# getopt.m4 serial 14 (modified version)
+dnl Copyright (C) 2002-2006, 2008 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+# The getopt module assume you want GNU getopt, with getopt_long etc,
+# rather than vanilla POSIX getopt. This means your code should
+# always include <getopt.h> for the getopt prototypes.
+
+AC_DEFUN([gl_GETOPT_SUBSTITUTE],
+[
+ AC_LIBOBJ([getopt])
+ AC_LIBOBJ([getopt1])
+ gl_GETOPT_SUBSTITUTE_HEADER
+])
+
+AC_DEFUN([gl_GETOPT_SUBSTITUTE_HEADER],
+[
+ GETOPT_H=getopt.h
+ AC_DEFINE([__GETOPT_PREFIX], [[rpl_]],
+ [Define to rpl_ if the getopt replacement functions and variables
+ should be used.])
+ AC_SUBST([GETOPT_H])
+])
+
+AC_DEFUN([gl_GETOPT_CHECK_HEADERS],
+[
+ if test -z "$GETOPT_H"; then
+ AC_CHECK_HEADERS([getopt.h], [], [GETOPT_H=getopt.h])
+ fi
+
+ if test -z "$GETOPT_H"; then
+ AC_CHECK_FUNCS([getopt_long], [], [GETOPT_H=getopt.h])
+ fi
+
+ dnl BSD getopt_long uses a way to reset option processing, that is different
+ dnl from GNU and Solaris (which copied the GNU behavior). We support both
+ dnl GNU and BSD style resetting of getopt_long(), so there's no need to use
+ dnl GNU getopt_long() on BSD due to different resetting style.
+ dnl
+ dnl With getopt_long(), some BSD versions have a bug in handling optional
+ dnl arguments. This bug appears only if the environment variable
+ dnl POSIXLY_CORRECT has been set, so it shouldn't be too bad in most
+ dnl cases; probably most don't have that variable set. But if we actually
+ dnl hit this bug, it is a real problem due to our heavy use of optional
+ dnl arguments.
+ dnl
+ dnl According to CVS logs, the bug was introduced in OpenBSD in 2003-09-22
+ dnl and copied to FreeBSD in 2004-02-24. It was fixed in both in 2006-09-22,
+ dnl so the affected versions shouldn't be popular anymore anyway. NetBSD
+ dnl never had this bug. TODO: What about Darwin and others?
+ if test -z "$GETOPT_H"; then
+ AC_CHECK_DECL([optreset],
+ [AC_DEFINE([HAVE_OPTRESET], 1,
+ [Define to 1 if getopt.h declares extern int optreset.])],
+ [], [#include <getopt.h>])
+ fi
+
+ dnl Solaris 10 getopt doesn't handle `+' as a leading character in an
+ dnl option string (as of 2005-05-05). We don't use that feature, so this
+ dnl is not a problem for us. Thus, the respective test was removed here.
+])
+
+AC_DEFUN([gl_GETOPT_IFELSE],
+[
+ AC_REQUIRE([gl_GETOPT_CHECK_HEADERS])
+ AS_IF([test -n "$GETOPT_H"], [$1], [$2])
+])
+
+AC_DEFUN([gl_GETOPT], [gl_GETOPT_IFELSE([gl_GETOPT_SUBSTITUTE])])
diff --git a/m4/posix-shell.m4 b/m4/posix-shell.m4
new file mode 100644
index 0000000..4c56193
--- /dev/null
+++ b/m4/posix-shell.m4
@@ -0,0 +1,63 @@
+# Find a POSIX-conforming shell.
+
+# Copyright (C) 2007-2008 Free Software Foundation, Inc.
+
+# This file is free software; the Free Software Foundation
+# gives unlimited permission to copy and/or distribute it,
+# with or without modifications, as long as this notice is preserved.
+
+# Written by Paul Eggert.
+
+# If a POSIX-conforming shell can be found, set POSIX_SHELL and
+# PREFERABLY_POSIX_SHELL to it. If not, set POSIX_SHELL to the
+# empty string and PREFERABLY_POSIX_SHELL to '/bin/sh'.
+
+AC_DEFUN([gl_POSIX_SHELL],
+[
+ AC_CACHE_CHECK([for a shell that conforms to POSIX], [gl_cv_posix_shell],
+ [gl_test_posix_shell_script='
+ func_return () {
+ (exit [$]1)
+ }
+ func_success () {
+ func_return 0
+ }
+ func_failure () {
+ func_return 1
+ }
+ func_ret_success () {
+ return 0
+ }
+ func_ret_failure () {
+ return 1
+ }
+ subshell_umask_sanity () {
+ (umask 22; (umask 0); test $(umask) -eq 22)
+ }
+ test "[$](echo foo)" = foo &&
+ func_success &&
+ ! func_failure &&
+ func_ret_success &&
+ ! func_ret_failure &&
+ (set x && func_ret_success y && test x = "[$]1") &&
+ subshell_umask_sanity
+ '
+ for gl_cv_posix_shell in \
+ "$CONFIG_SHELL" "$SHELL" /bin/sh /bin/bash /bin/ksh /bin/sh5 no; do
+ case $gl_cv_posix_shell in
+ /*)
+ "$gl_cv_posix_shell" -c "$gl_test_posix_shell_script" 2>/dev/null \
+ && break;;
+ esac
+ done])
+
+ if test "$gl_cv_posix_shell" != no; then
+ POSIX_SHELL=$gl_cv_posix_shell
+ PREFERABLY_POSIX_SHELL=$POSIX_SHELL
+ else
+ POSIX_SHELL=
+ PREFERABLY_POSIX_SHELL=/bin/sh
+ fi
+ AC_SUBST([POSIX_SHELL])
+ AC_SUBST([PREFERABLY_POSIX_SHELL])
+])
diff --git a/m4/tuklib_common.m4 b/m4/tuklib_common.m4
new file mode 100644
index 0000000..d942a25
--- /dev/null
+++ b/m4/tuklib_common.m4
@@ -0,0 +1,22 @@
+#
+# SYNOPSIS
+#
+# TUKLIB_COMMON
+#
+# DESCRIPTION
+#
+# Common checks for tuklib.
+#
+# COPYING
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+AC_DEFUN_ONCE([TUKLIB_COMMON], [
+AC_REQUIRE([AC_CANONICAL_HOST])
+AC_REQUIRE([AC_PROG_CC_C99])
+AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS])
+])dnl
diff --git a/m4/tuklib_cpucores.m4 b/m4/tuklib_cpucores.m4
new file mode 100644
index 0000000..9e295c8
--- /dev/null
+++ b/m4/tuklib_cpucores.m4
@@ -0,0 +1,105 @@
+#
+# SYNOPSIS
+#
+# TUKLIB_CPUCORES
+#
+# DESCRIPTION
+#
+# Check how to find out the number of available CPU cores in the system.
+# This information is used by tuklib_cpucores.c.
+#
+# Supported methods:
+# - sysctl(): BSDs, OS/2
+# - sysconf(): GNU/Linux, Solaris, Tru64, IRIX, AIX, Cygwin
+# - pstat_getdynamic(): HP-UX
+#
+# COPYING
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+AC_DEFUN_ONCE([TUKLIB_CPUCORES], [
+AC_REQUIRE([TUKLIB_COMMON])
+
+# sys/param.h might be needed by sys/sysctl.h.
+AC_CHECK_HEADERS([sys/param.h])
+
+AC_CACHE_CHECK([how to detect the number of available CPU cores],
+ [tuklib_cv_cpucores_method], [
+
+# Look for sysctl() solution first, because on OS/2, both sysconf()
+# and sysctl() pass the tests in this file, but only sysctl()
+# actually works.
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/types.h>
+#ifdef HAVE_SYS_PARAM_H
+# include <sys/param.h>
+#endif
+#include <sys/sysctl.h>
+int
+main(void)
+{
+ int name[2] = { CTL_HW, HW_NCPU };
+ int cpus;
+ size_t cpus_size = sizeof(cpus);
+ sysctl(name, 2, &cpus, &cpus_size, NULL, 0);
+ return 0;
+}
+]])], [tuklib_cv_cpucores_method=sysctl], [
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <unistd.h>
+int
+main(void)
+{
+ long i;
+#ifdef _SC_NPROCESSORS_ONLN
+ /* Many systems using sysconf() */
+ i = sysconf(_SC_NPROCESSORS_ONLN);
+#else
+ /* IRIX */
+ i = sysconf(_SC_NPROC_ONLN);
+#endif
+ return 0;
+}
+]])], [tuklib_cv_cpucores_method=sysconf], [
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/param.h>
+#include <sys/pstat.h>
+
+int
+main(void)
+{
+ struct pst_dynamic pst;
+ pstat_getdynamic(&pst, sizeof(pst), 1, 0);
+ (void)pst.psd_proc_cnt;
+ return 0;
+}
+]])], [tuklib_cv_cpucores_method=pstat_getdynamic], [
+
+ tuklib_cv_cpucores_method=unknown
+])])])])
+
+case $tuklib_cv_cpucores_method in
+ sysctl)
+ AC_DEFINE([TUKLIB_CPUCORES_SYSCTL], [1],
+ [Define to 1 if the number of available CPU cores
+ can be detected with sysctl().])
+ ;;
+ sysconf)
+ AC_DEFINE([TUKLIB_CPUCORES_SYSCONF], [1],
+ [Define to 1 if the number of available CPU cores
+ can be detected with sysconf(_SC_NPROCESSORS_ONLN)
+ or sysconf(_SC_NPROC_ONLN).])
+ ;;
+ pstat_getdynamic)
+ AC_DEFINE([TUKLIB_CPUCORES_PSTAT_GETDYNAMIC], [1],
+ [Define to 1 if the number of available CPU cores
+ can be detected with pstat_getdynamic().])
+ ;;
+esac
+])dnl
diff --git a/m4/tuklib_integer.m4 b/m4/tuklib_integer.m4
new file mode 100644
index 0000000..5fe66ee
--- /dev/null
+++ b/m4/tuklib_integer.m4
@@ -0,0 +1,74 @@
+#
+# SYNOPSIS
+#
+# TUKLIB_INTEGER
+#
+# DESCRIPTION
+#
+# Checks for tuklib_integer.h:
+# - Endianness
+# - Does operating system provide byte swapping macros
+# - Does the hardware support fast unaligned access to 16-bit
+# and 32-bit integers
+#
+# COPYING
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+AC_DEFUN_ONCE([TUKLIB_INTEGER], [
+AC_REQUIRE([TUKLIB_COMMON])
+AC_REQUIRE([AC_C_BIGENDIAN])
+AC_CHECK_HEADERS([byteswap.h sys/endian.h sys/byteorder.h], [break])
+
+# Even if we have byteswap.h, we may lack the specific macros/functions.
+if test x$ac_cv_header_byteswap_h = xyes ; then
+ m4_foreach([FUNC], [bswap_16,bswap_32,bswap_64], [
+ AC_MSG_CHECKING([if FUNC is available])
+ AC_LINK_IFELSE([AC_LANG_SOURCE([
+#include <byteswap.h>
+int
+main(void)
+{
+ FUNC[](42);
+ return 0;
+}
+ ])], [
+ AC_DEFINE(HAVE_[]m4_toupper(FUNC), [1],
+ [Define to 1 if] FUNC [is available.])
+ AC_MSG_RESULT([yes])
+ ], [AC_MSG_RESULT([no])])
+
+ ])dnl
+fi
+
+AC_MSG_CHECKING([if unaligned memory access should be used])
+AC_ARG_ENABLE([unaligned-access], AC_HELP_STRING([--enable-unaligned-access],
+ [Enable if the system supports *fast* unaligned memory access
+ with 16-bit and 32-bit integers. By default, this is enabled
+ only on x86, x86_64, and big endian PowerPC.]),
+ [], [enable_unaligned_access=auto])
+if test "x$enable_unaligned_access" = xauto ; then
+ # TODO: There may be other architectures, on which unaligned access
+ # is OK.
+ case $host_cpu in
+ i?86|x86_64|powerpc|powerpc64)
+ enable_unaligned_access=yes
+ ;;
+ *)
+ enable_unaligned_access=no
+ ;;
+ esac
+fi
+if test "x$enable_unaligned_access" = xyes ; then
+ AC_DEFINE([TUKLIB_FAST_UNALIGNED_ACCESS], [1], [Define to 1 if
+ the system supports fast unaligned access to 16-bit and
+ 32-bit integers.])
+ AC_MSG_RESULT([yes])
+else
+ AC_MSG_RESULT([no])
+fi
+])dnl
diff --git a/m4/tuklib_physmem.m4 b/m4/tuklib_physmem.m4
new file mode 100644
index 0000000..1249925
--- /dev/null
+++ b/m4/tuklib_physmem.m4
@@ -0,0 +1,211 @@
+#
+# SYNOPSIS
+#
+# TUKLIB_PHYSMEM
+#
+# DESCRIPTION
+#
+# Check how to get the amount of physical memory.
+# This information is used in tuklib_physmem.c.
+#
+# Supported methods:
+#
+# - Windows (including Cygwin), OS/2, DJGPP (DOS), and OpenVMS have
+# operating-system specific functions.
+#
+# - AIX has _system_configuration.physmem.
+#
+# - sysconf() works on GNU/Linux and Solaris, and possibly on
+# some BSDs.
+#
+# - BSDs use sysctl().
+#
+# - Tru64 uses getsysinfo().
+#
+# - HP-UX uses pstat_getstatic().
+#
+# - IRIX has setinvent_r(), getinvent_r(), and endinvent_r().
+#
+# - sysinfo() works on Linux/dietlibc and probably on other Linux
+# systems whose libc may lack sysconf().
+#
+# COPYING
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+AC_DEFUN_ONCE([TUKLIB_PHYSMEM], [
+AC_REQUIRE([TUKLIB_COMMON])
+
+# sys/param.h might be needed by sys/sysctl.h.
+AC_CHECK_HEADERS([sys/param.h])
+
+AC_CACHE_CHECK([how to detect the amount of physical memory],
+ [tuklib_cv_physmem_method], [
+
+# Maybe checking $host_os would be enough but this matches what
+# tuklib_physmem.c does.
+#
+# NOTE: IRIX has a compiler that doesn't error out with #error, so use
+# a non-compilable text instead of #error to generate an error.
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#if defined(_WIN32) || defined(__CYGWIN__) || defined(__OS2__) \
+ || defined(__DJGPP__) || defined(__VMS)
+int main(void) { return 0; }
+#else
+compile error
+#endif
+]])], [tuklib_cv_physmem_method=special], [
+
+# Look for AIX-specific solution before sysconf(), because the test
+# for sysconf() will pass on AIX but won't actually work
+# (sysconf(_SC_PHYS_PAGES) compiles but always returns -1 on AIX).
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/systemcfg.h>
+
+int
+main(void)
+{
+ (void)_system_configuration.physmem;
+ return 0;
+}
+]])], [tuklib_cv_physmem_method=aix], [
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <unistd.h>
+int
+main(void)
+{
+ long i;
+ i = sysconf(_SC_PAGESIZE);
+ i = sysconf(_SC_PHYS_PAGES);
+ return 0;
+}
+]])], [tuklib_cv_physmem_method=sysconf], [
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/types.h>
+#ifdef HAVE_SYS_PARAM_H
+# include <sys/param.h>
+#endif
+#include <sys/sysctl.h>
+int
+main(void)
+{
+ int name[2] = { CTL_HW, HW_PHYSMEM };
+ unsigned long mem;
+ size_t mem_ptr_size = sizeof(mem);
+ sysctl(name, 2, &mem, &mem_ptr_size, NULL, 0);
+ return 0;
+}
+]])], [tuklib_cv_physmem_method=sysctl], [
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/sysinfo.h>
+#include <machine/hal_sysinfo.h>
+
+int
+main(void)
+{
+ int memkb;
+ int start = 0;
+ getsysinfo(GSI_PHYSMEM, (caddr_t)&memkb, sizeof(memkb), &start);
+ return 0;
+}
+]])], [tuklib_cv_physmem_method=getsysinfo],[
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/param.h>
+#include <sys/pstat.h>
+
+int
+main(void)
+{
+ struct pst_static pst;
+ pstat_getstatic(&pst, sizeof(pst), 1, 0);
+ (void)pst.physical_memory;
+ (void)pst.page_size;
+ return 0;
+}
+]])], [tuklib_cv_physmem_method=pstat_getstatic],[
+
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <invent.h>
+int
+main(void)
+{
+ inv_state_t *st = NULL;
+ setinvent_r(&st);
+ getinvent_r(st);
+ endinvent_r(st);
+ return 0;
+}
+]])], [tuklib_cv_physmem_method=getinvent_r], [
+
+# This version of sysinfo() is Linux-specific. Some non-Linux systems have
+# different sysinfo() so we must check $host_os.
+case $host_os in
+ linux*)
+ AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+#include <sys/sysinfo.h>
+int
+main(void)
+{
+ struct sysinfo si;
+ sysinfo(&si);
+ return 0;
+}
+ ]])], [
+ tuklib_cv_physmem_method=sysinfo
+ ], [
+ tuklib_cv_physmem_method=unknown
+ ])
+ ;;
+ *)
+ tuklib_cv_physmem_method=unknown
+ ;;
+esac
+])])])])])])])])
+
+case $tuklib_cv_physmem_method in
+ aix)
+ AC_DEFINE([TUKLIB_PHYSMEM_AIX], [1],
+ [Define to 1 if the amount of physical memory
+ can be detected with _system_configuration.physmem.])
+ ;;
+ sysconf)
+ AC_DEFINE([TUKLIB_PHYSMEM_SYSCONF], [1],
+ [Define to 1 if the amount of physical memory can
+ be detected with sysconf(_SC_PAGESIZE) and
+ sysconf(_SC_PHYS_PAGES).])
+ ;;
+ sysctl)
+ AC_DEFINE([TUKLIB_PHYSMEM_SYSCTL], [1],
+ [Define to 1 if the amount of physical memory can
+ be detected with sysctl().])
+ ;;
+ getsysinfo)
+ AC_DEFINE([TUKLIB_PHYSMEM_GETSYSINFO], [1],
+ [Define to 1 if the amount of physical memory can
+ be detected with getsysinfo().])
+ ;;
+ pstat_getstatic)
+ AC_DEFINE([TUKLIB_PHYSMEM_PSTAT_GETSTATIC], [1],
+ [Define to 1 if the amount of physical memory can
+ be detected with pstat_getstatic().])
+ ;;
+ getinvent_r)
+ AC_DEFINE([TUKLIB_PHYSMEM_GETINVENT_R], [1],
+ [Define to 1 if the amount of physical memory
+ can be detected with getinvent_r().])
+ ;;
+ sysinfo)
+ AC_DEFINE([TUKLIB_PHYSMEM_SYSINFO], [1],
+ [Define to 1 if the amount of physical memory
+ can be detected with Linux sysinfo().])
+ ;;
+esac
+])dnl
diff --git a/m4/tuklib_progname.m4 b/m4/tuklib_progname.m4
new file mode 100644
index 0000000..f3e158b
--- /dev/null
+++ b/m4/tuklib_progname.m4
@@ -0,0 +1,25 @@
+#
+# SYNOPSIS
+#
+# TUKLIB_PROGNAME
+#
+# DESCRIPTION
+#
+# Put argv[0] into a global variable progname. On DOS-like systems,
+# modify it so that it looks nice (no full path or .exe suffix).
+#
+# This .m4 file is needed allow this module to use glibc's
+# program_invocation_name.
+#
+# COPYING
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+AC_DEFUN_ONCE([TUKLIB_PROGNAME], [
+AC_REQUIRE([TUKLIB_COMMON])
+AC_CHECK_DECLS([program_invocation_name], [], [], [#include <errno.h>])
+])dnl
diff --git a/po/.gitignore b/po/.gitignore
new file mode 100644
index 0000000..d201729
--- /dev/null
+++ b/po/.gitignore
@@ -0,0 +1,31 @@
+# autopoint
+Makefile.in.in
+Makevars.template
+Rules-quot
+boldquot.sed
+en@boldquot.header
+en@quot.header
+insert-header.sin
+quot.sed
+remove-potcdate.sin
+
+# configure
+Makefile.in
+Makefile
+POTFILES
+
+# intermediate files (make)
+stamp-poT
+xz.po
+xz.1po
+xz.2po
+*.new.po
+
+# make
+remove-potcdate.sed
+xz.mo
+stamp-po
+*.gmo
+
+# cached templates (make)
+xz.pot
diff --git a/po/LINGUAS b/po/LINGUAS
new file mode 100644
index 0000000..841618a
--- /dev/null
+++ b/po/LINGUAS
@@ -0,0 +1 @@
+cs
diff --git a/po/Makevars b/po/Makevars
new file mode 100644
index 0000000..dc19bc9
--- /dev/null
+++ b/po/Makevars
@@ -0,0 +1,46 @@
+# Makefile variables for PO directory in any package using GNU gettext.
+
+# Usually the message domain is the same as the package name.
+DOMAIN = $(PACKAGE)
+
+# These two variables depend on the location of this directory.
+subdir = po
+top_builddir = ..
+
+# These options get passed to xgettext.
+XGETTEXT_OPTIONS = --keyword=_ --keyword=N_
+
+# This is the copyright holder that gets inserted into the header of the
+# $(DOMAIN).pot file. Set this to the copyright holder of the surrounding
+# package. (Note that the msgstr strings, extracted from the package's
+# sources, belong to the copyright holder of the package.) Translators are
+# expected to transfer the copyright for their translations to this person
+# or entity, or to disclaim their copyright. The empty string stands for
+# the public domain; in this case the translators are expected to disclaim
+# their copyright.
+COPYRIGHT_HOLDER =
+
+# This is the email address or URL to which the translators shall report
+# bugs in the untranslated strings:
+# - Strings which are not entire sentences, see the maintainer guidelines
+# in the GNU gettext documentation, section 'Preparing Strings'.
+# - Strings which use unclear terms or require additional context to be
+# understood.
+# - Strings which make invalid assumptions about notation of date, time or
+# money.
+# - Pluralisation problems.
+# - Incorrect English spelling.
+# - Incorrect formatting.
+# It can be your email address, or a mailing list address where translators
+# can write to without being subscribed, or the URL of a web page through
+# which the translators can contact you.
+MSGID_BUGS_ADDRESS =
+
+# This is the list of locale categories, beyond LC_MESSAGES, for which the
+# message catalogs shall be used. It is usually empty.
+EXTRA_LOCALE_CATEGORIES =
+
+# Although you may need slightly wider terminal than 80 chars, it is
+# much nicer to edit the output of --help when this is set.
+XGETTEXT_OPTIONS += --no-wrap
+MSGMERGE += --no-wrap
diff --git a/po/POTFILES.in b/po/POTFILES.in
new file mode 100644
index 0000000..ee430c5
--- /dev/null
+++ b/po/POTFILES.in
@@ -0,0 +1,10 @@
+# List of source files which contain translatable strings.
+src/xz/args.c
+src/xz/coder.c
+src/xz/file_io.c
+src/xz/hardware.c
+src/xz/main.c
+src/xz/message.c
+src/xz/options.c
+src/xz/suffix.c
+src/xz/util.c
diff --git a/po/cs.po b/po/cs.po
new file mode 100644
index 0000000..a953fd3
--- /dev/null
+++ b/po/cs.po
@@ -0,0 +1,636 @@
+msgid ""
+msgstr ""
+"Project-Id-Version: xz-utils\n"
+"Report-Msgid-Bugs-To: lasse.collin@tukaani.org\n"
+"POT-Creation-Date: 2009-12-04 16:04+0100\n"
+"PO-Revision-Date: 2009-12-05 13:51+0100\n"
+"Last-Translator: Marek Černocký <marek@manet.cz>\n"
+"Language-Team: Czech <diskuze@lists.l10n.cz>\n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=UTF-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+"Plural-Forms: nplurals=3; plural=(n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2\n"
+"X-Poedit-Language: Czech\n"
+"X-Poedit-SourceCharset: utf-8\n"
+
+#: src/xz/args.c:306
+#, c-format
+msgid "%s: Unknown file format type"
+msgstr "%s: Neznámý typ formátu souboru"
+
+#: src/xz/args.c:329 src/xz/args.c:337
+#, c-format
+msgid "%s: Unsupported integrity check type"
+msgstr "%s: Neznámý typ kontroly integrity"
+
+#: src/xz/args.c:355
+msgid "Only one file can be specified with `--files'or `--files0'."
+msgstr ""
+"Můžete zadat pouze jeden soubor spolu s přepínači „--files“ nebo „--files0“"
+
+#: src/xz/args.c:413
+msgid "The environment variable XZ_OPT contains too many arguments"
+msgstr "Proměnná prostředí XZ_OPT obsahuje příliš mnoho argumentů"
+
+#: src/xz/coder.c:95
+msgid "Maximum number of filters is four"
+msgstr "Maximální počet filtrů je čtyři"
+
+#: src/xz/coder.c:108
+#, c-format
+msgid ""
+"Memory usage limit (%<PRIu64> MiB) is too small for the given filter setup (%"
+"<PRIu64> MiB)"
+msgstr ""
+"Omezení použitelné paměti (%<PRIu64> MiB) je příliš malé pro dané nastavení "
+"filtru (%<PRIu64> MiB)"
+
+#: src/xz/coder.c:128
+msgid "Using a preset in raw mode is discouraged."
+msgstr "Použití přednastavení v režimu raw je nevhodné."
+
+#: src/xz/coder.c:130
+msgid "The exact options of the presets may vary between software versions."
+msgstr ""
+"Přesné volby u přednastavení se mohou lišit mezi různými verzemi softwaru."
+
+#: src/xz/coder.c:158
+msgid "The .lzma format supports only the LZMA1 filter"
+msgstr "Formát .lzma podporuje pouze filtr LZMA1"
+
+#: src/xz/coder.c:166
+msgid "LZMA1 cannot be used with the .xz format"
+msgstr "LZMA1 nelze použít s formátem .xz"
+
+#: src/xz/coder.c:186
+#, c-format
+msgid "%s MiB (%s B) of memory is required per thread, limit is %s MiB (%s B)"
+msgstr ""
+"Je vyžadováno %s MiB (%s B) paměti pro každé vlákno, limit je %s MiB (%s B)"
+
+#: src/xz/coder.c:554
+#, c-format
+msgid "Limit was %s MiB, but %s MiB would have been needed"
+msgstr "Limit byl %s MiB, ale bylo by zapotřebí %s MiB"
+
+#: src/xz/file_io.c:113
+#, c-format
+msgid "%s: File seems to be moved, not removing"
+msgstr "%s: Vypadá to, že soubor byl přesunut, neodstraní se"
+
+#: src/xz/file_io.c:120 src/xz/file_io.c:527
+#, c-format
+msgid "%s: Cannot remove: %s"
+msgstr "%s: Nelze odstranit: %s"
+
+#: src/xz/file_io.c:145
+#, c-format
+msgid "%s: Cannot set the file owner: %s"
+msgstr "%s: Nelze nastavit vlastníka souboru: %s"
+
+#: src/xz/file_io.c:151
+#, c-format
+msgid "%s: Cannot set the file group: %s"
+msgstr "%s: Nelze nastavit skupinu souboru: %s"
+
+#: src/xz/file_io.c:170
+#, c-format
+msgid "%s: Cannot set the file permissions: %s"
+msgstr "%s: Nelze nastavit oprávnění souboru: %s"
+
+#: src/xz/file_io.c:308 src/xz/file_io.c:387
+#, c-format
+msgid "%s: Is a symbolic link, skipping"
+msgstr "%s: Jedná se o symbolický odkaz, vynechává se"
+
+#: src/xz/file_io.c:422
+#, c-format
+msgid "%s: Is a directory, skipping"
+msgstr "%s: Jedná se o složku, vynechává se"
+
+#: src/xz/file_io.c:429
+#, c-format
+msgid "%s: Not a regular file, skipping"
+msgstr "%s: Nejedná se o běžný soubor, vynechává se"
+
+#: src/xz/file_io.c:446
+#, c-format
+msgid "%s: File has setuid or setgid bit set, skipping"
+msgstr "%s: Soubor má nastavený bit setuid nebo setgid, vynechává se"
+
+#: src/xz/file_io.c:453
+#, c-format
+msgid "%s: File has sticky bit set, skipping"
+msgstr "%s: Soubor má nastavený bit sticky, vynechává se"
+
+#: src/xz/file_io.c:460
+#, c-format
+msgid "%s: Input file has more than one hard link, skipping"
+msgstr "%s: Vstupní soubor má více než jeden pevný odkaz, vynechává se"
+
+#: src/xz/file_io.c:644
+#, c-format
+msgid "Error restoring the O_APPEND flag to standard output: %s"
+msgstr "Chyba při obnovení příznaku O_APPEND na standardní výstup: %s"
+
+#: src/xz/file_io.c:655
+#, c-format
+msgid "%s: Closing the file failed: %s"
+msgstr "%s: Selhalo zavření souboru: %s"
+
+#: src/xz/file_io.c:732 src/xz/file_io.c:889
+#, c-format
+msgid "%s: Seeking failed when trying to create a sparse file: %s"
+msgstr ""
+"%s: Selhalo nastavení pozice při pokusu o vytvoření záložního souboru: %s"
+
+#: src/xz/file_io.c:789
+#, c-format
+msgid "%s: Read error: %s"
+msgstr "%s: Chyba čtení: %s"
+
+#: src/xz/file_io.c:847
+#, c-format
+msgid "%s: Write error: %s"
+msgstr "%s: Chyba zápisu: %s"
+
+#: src/xz/main.c:76
+#, c-format
+msgid "%s: Error reading filenames: %s"
+msgstr "%s: Chyba při čtení názvů souborů: %s"
+
+#: src/xz/main.c:83
+#, c-format
+msgid "%s: Unexpected end of input when reading filenames"
+msgstr "%s: Neočekávaný konec vstupu při čtení názvů souborů"
+
+#: src/xz/main.c:107
+#, c-format
+msgid ""
+"%s: Null character found when reading filenames; maybe you meant to use `--"
+"files0' instead of `--files'?"
+msgstr ""
+"%s: Byl nalezen nulový znak při čtení názvů souborů; nechtěli jste náhodou "
+"použít „--files0“ místo „--files“?"
+
+#: src/xz/main.c:160
+msgid "Compression and decompression with --robot are not supported yet."
+msgstr "Komprimace a dekomprimace s přepínačem --robot není zatím podporovaná."
+
+#: src/xz/main.c:207
+msgid ""
+"Cannot read data from standard input when reading filenames from standard "
+"input"
+msgstr ""
+"Ze standardního vstupu nelze číst data, když se ze standardního vstupu "
+"načítají názvy souborů"
+
+#: src/xz/message.c:841 src/xz/message.c:885
+msgid "Internal error (bug)"
+msgstr "Interní chyba"
+
+#: src/xz/message.c:848
+msgid "Cannot establish signal handlers"
+msgstr "Nelze ustanovit ovladač signálu"
+
+#: src/xz/message.c:857
+msgid "No integrity check; not verifying file integrity"
+msgstr "Žádná kontrola integrity; integrita souboru se nebude ověřovat"
+
+#: src/xz/message.c:860
+msgid "Unsupported type of integrity check; not verifying file integrity"
+msgstr ""
+"Nepodporovaný typ kontroly integrity; integrita souboru se nebude ověřovat"
+
+#: src/xz/message.c:867
+msgid "Memory usage limit reached"
+msgstr "Dosaženo omezení použitelné paměti"
+
+#: src/xz/message.c:870
+msgid "File format not recognized"
+msgstr "Formát souboru nebyl rozpoznán"
+
+#: src/xz/message.c:873
+msgid "Unsupported options"
+msgstr "Nepodporovaná volba"
+
+#: src/xz/message.c:876
+msgid "Compressed data is corrupt"
+msgstr "Komprimovaná data jsou poškozená"
+
+#: src/xz/message.c:879
+msgid "Unexpected end of input"
+msgstr "Neočekávaný konec vstupu"
+
+#: src/xz/message.c:898
+#, c-format
+msgid "%s: Filter chain:"
+msgstr "%s: Omezující filtr:"
+
+#: src/xz/message.c:1009
+#, c-format
+msgid "Try `%s --help' for more information."
+msgstr "Zkuste „%s --help“ pro více informací"
+
+#: src/xz/message.c:1021
+#, c-format
+msgid "%s MiB (%s bytes)\n"
+msgstr "%s MiB (%s bajtů)\n"
+
+#: src/xz/message.c:1049
+#, c-format
+msgid ""
+"Usage: %s [OPTION]... [FILE]...\n"
+"Compress or decompress FILEs in the .xz format.\n"
+"\n"
+msgstr ""
+"Použití: %s [PŘEPÍNAČ]... [SOUBOR]...\n"
+"Komprimuje nebo dekomprimuje SOUBORy ve formátu xz.\n"
+"\n"
+
+#: src/xz/message.c:1053
+msgid ""
+"Mandatory arguments to long options are mandatory for short options too.\n"
+msgstr ""
+"Povinné argumenty pro dlouhé přepínače jsou povinné rovněž pro krátké "
+"přepínače.\n"
+
+#: src/xz/message.c:1057
+msgid " Operation mode:\n"
+msgstr "Operační režim:\n"
+
+#: src/xz/message.c:1060
+msgid ""
+" -z, --compress force compression\n"
+" -d, --decompress force decompression\n"
+" -t, --test test compressed file integrity\n"
+" -l, --list list information about files"
+msgstr ""
+" -z, --compress provést komprimaci\n"
+" -d, --decompress provést dekomprimaci\n"
+" -t, --test testovat integritu komprimovaného souboru\n"
+" -l, --list vypsat informace o souborech"
+
+#: src/xz/message.c:1066
+msgid ""
+"\n"
+" Operation modifiers:\n"
+msgstr ""
+"\n"
+"Modifikátory operací:\n"
+
+#: src/xz/message.c:1069
+msgid ""
+" -k, --keep keep (don't delete) input files\n"
+" -f, --force force overwrite of output file and (de)compress links\n"
+" -c, --stdout write to standard output and don't delete input files"
+msgstr ""
+" -k, --keep zachovat (nemazat) vstupní soubory\n"
+" -f, --force vynutit přepis výstupního souboru a de/komprimovat "
+"odkazy\n"
+" -c, --stdout zapisovat na standardní výstup a nemazat vstupní "
+"soubory"
+
+#: src/xz/message.c:1075
+msgid ""
+" --no-sparse do not create sparse files when decompressing\n"
+" -S, --suffix=.SUF use the suffix `.SUF' on compressed files\n"
+" --files=[FILE] read filenames to process from FILE; if FILE is\n"
+" omitted, filenames are read from the standard input;\n"
+" filenames must be terminated with the newline "
+"character\n"
+" --files0=[FILE] like --files but use the null character as terminator"
+msgstr ""
+" --no-sparse nevytvářet při dekomprimaci záložní soubory\n"
+" -S, --suffix=.SUF použít u komprimovaných souborů příponu „.SUF“\n"
+" --files=[SOUBOR] číst názvy souborů, které se mají zpracovat, ze "
+"SOUBORu;\n"
+" pokud není SOUBOR zadán, čte se ze standardního "
+"vstupu;\n"
+" názvy souborů musí být zakončeny znakem nového řádku\n"
+" --files0=[SOUBOR] stejné jako --files, ale použít k zakončování nulový "
+"znak"
+
+#: src/xz/message.c:1083
+msgid ""
+"\n"
+" Basic file format and compression options:\n"
+msgstr ""
+"\n"
+"Základní přepínače pro formát souboru a komprimaci:\n"
+
+#: src/xz/message.c:1085
+msgid ""
+" -F, --format=FMT file format to encode or decode; possible values are\n"
+" `auto' (default), `xz', `lzma', and `raw'\n"
+" -C, --check=CHECK integrity check type: `crc32', `crc64' (default),\n"
+" `sha256', or `none' (use with caution)"
+msgstr ""
+" -F, --format=FORMÁT formát souboru k zakódování nebo dekódování; možné\n"
+" hodnoty jsou „auto“ (výchozí), „xz“, „lzma“ a „raw“\n"
+" -C, --check=KONTROLA typ kontroly integrity: „crc32“, „crc64“ (výchozí),\n"
+" „sha256“ nebo „none“ (používejte s rozmyslem)"
+
+#: src/xz/message.c:1092
+msgid ""
+" -0 .. -9 compression preset; 0-2 fast compression, 3-5 good\n"
+" compression, 6-9 excellent compression; default is 6"
+msgstr ""
+" -0 .. -9 přednastavení komprimace; 0-2 rychlá komprimace, 3-5\n"
+" dobrá komprimace, 6-9 skvělá komprimace; výchozí je 6"
+
+#: src/xz/message.c:1096
+msgid ""
+" -e, --extreme use more CPU time when encoding to increase "
+"compression\n"
+" ratio without increasing memory usage of the decoder"
+msgstr ""
+" -e, --extreme využít více procesorového času pro kódování, čímž se\n"
+" zvýší kompresní poměr bez zvýšení paměti použité "
+"kodérem"
+
+#: src/xz/message.c:1101
+msgid ""
+" -M, --memory=NUM use roughly NUM bytes of memory at maximum; 0 "
+"indicates\n"
+" the default setting, which is 40 % of total RAM"
+msgstr ""
+" -M, --memory=POČ použít zhruba POČ bajtů paměti jako maximum; 0 "
+"znamená\n"
+" výchozí nastavení, což je 40% celkového množství "
+"paměti"
+
+#: src/xz/message.c:1106
+msgid ""
+"\n"
+" Custom filter chain for compression (alternative for using presets):"
+msgstr ""
+"\n"
+"Vlastní omezující filtr pro komprimaci (alternativa k použití "
+"přednastavených):"
+
+#: src/xz/message.c:1111
+msgid ""
+"\n"
+" --lzma1[=OPTS] LZMA1 or LZMA2; OPTS is a comma-separated list of zero "
+"or\n"
+" --lzma2[=OPTS] more of the following options (valid values; "
+"default):\n"
+" preset=NUM reset options to preset number NUM (0-9)\n"
+" dict=NUM dictionary size (4KiB - 1536MiB; 8MiB)\n"
+" lc=NUM number of literal context bits (0-4; 3)\n"
+" lp=NUM number of literal position bits (0-4; 0)\n"
+" pb=NUM number of position bits (0-4; 2)\n"
+" mode=MODE compression mode (fast, normal; normal)\n"
+" nice=NUM nice length of a match (2-273; 64)\n"
+" mf=NAME match finder (hc3, hc4, bt2, bt3, bt4; "
+"bt4)\n"
+" depth=NUM maximum search depth; 0=automatic "
+"(default)"
+msgstr ""
+"\n"
+" --lzma1[=VOLBY] LZMA1 nebo LZMA2; VOLBY je čárkou oddělovaný seznam "
+"žádné\n"
+" --lzma2[=VOLBY] nebo více následujících voleb (platné hodnoty; "
+"výchozí):\n"
+" preset=POČ obnovení voleb na přednastavený POČet (0-"
+"9)\n"
+" dict=POČ velikost slovníku (4 KiB - 1536 MiB; 8 "
+"MiB)\n"
+" lc=POČ počet kontextových bitů literálu (0-4; "
+"3)\n"
+" lp=POČ počet pozičních bitů literálu (0-4; 0)\n"
+" pb=POČ počet pozičních bitů (0-4; 2)\n"
+" mode=REŽIM režim komprimace (fast, normal; normal)\n"
+" nice=POČ příznivá délka shody (2-273; 64)\n"
+" mf=NÁZEV vyhledávač shod (hc3, hc4, bt2, bt3, "
+"bt4;\n"
+" bt4)\n"
+" depth=POČ maximální hloubka prohledávání;\n"
+" 0=automaticky (výchozí)"
+
+#: src/xz/message.c:1126
+msgid ""
+"\n"
+" --x86[=OPTS] x86 BCJ filter\n"
+" --powerpc[=OPTS] PowerPC BCJ filter (big endian only)\n"
+" --ia64[=OPTS] IA64 (Itanium) BCJ filter\n"
+" --arm[=OPTS] ARM BCJ filter (little endian only)\n"
+" --armthumb[=OPTS] ARM-Thumb BCJ filter (little endian only)\n"
+" --sparc[=OPTS] SPARC BCJ filter\n"
+" Valid OPTS for all BCJ filters:\n"
+" start=NUM start offset for conversions (default=0)"
+msgstr ""
+"\n"
+" --x86[=VOLBY] Filtr x86 BCJ\n"
+" --powerpc[=VOLBY] Filtr PowerPC BCJ (pouze big endian)\n"
+" --ia64[=VOLBY] Filtr IA64 (Itanium) BCJ\n"
+" --arm[=VOLBY] Filtr ARM BCJ (pouze little endian)\n"
+" --armthumb[=VOLBY] Filtr ARM-Thumb BCJ (pouze little endian)\n"
+" --sparc[=VOLBY] Filtr SPARC BCJ\n"
+" Platné volby pro všechny filtry BCJ:\n"
+" start=POČ počáteční posun pro převody (výchozí=0)"
+
+#: src/xz/message.c:1138
+msgid ""
+"\n"
+" --delta[=OPTS] Delta filter; valid OPTS (valid values; default):\n"
+" dist=NUM distance between bytes being subtracted\n"
+" from each other (1-256; 1)"
+msgstr ""
+"\n"
+" --delta[=VOLBY] Filtr Delta; platné VOLBY (platné hodnoty; výchozí):\n"
+" dist=POČ vzdálenost mezi bajty, které jsou "
+"odečítány\n"
+" jeden od druhého (1-256; 1)"
+
+#: src/xz/message.c:1146
+msgid ""
+"\n"
+" --subblock[=OPTS] Subblock filter; valid OPTS (valid values; default):\n"
+" size=NUM number of bytes of data per subblock\n"
+" (1 - 256Mi; 4Ki)\n"
+" rle=NUM run-length encoder chunk size (0-256; 0)"
+msgstr ""
+"\n"
+" --subblock[=VOLBY] Subblokový filtr; platné VOLBY (platné hodnoty; "
+"výchozí):\n"
+" size=POČ počet bajtů dat na subblok\n"
+" (1 - 256 Mi; 4 Ki)\n"
+" rle=POČ velikost dávky pro kodér run-length (0-"
+"256; 0)"
+
+#: src/xz/message.c:1155
+msgid ""
+"\n"
+" Other options:\n"
+msgstr ""
+"\n"
+" Ostatní přepínače:\n"
+
+#: src/xz/message.c:1158
+msgid ""
+" -q, --quiet suppress warnings; specify twice to suppress errors "
+"too\n"
+" -v, --verbose be verbose; specify twice for even more verbose"
+msgstr ""
+" -q, --quiet potlačit varování; zadáním dvakrát, potlačíte i "
+"chyby\n"
+" -v, --verbose podrobnější zprávy; zadáním dvakrát, budou ještě\n"
+" podrobnější"
+
+#: src/xz/message.c:1163
+msgid " -Q, --no-warn make warnings not affect the exit status"
+msgstr " -Q, --no-warn způsobí, že varování neovlivní stav ukončení"
+
+#: src/xz/message.c:1165
+msgid ""
+" --robot use machine-parsable messages (useful for scripts)"
+msgstr ""
+" --robot použít strojově analyzovatelné zprávy (užitečné pro "
+"skripty)"
+
+#: src/xz/message.c:1168
+msgid " --info-memory display the memory usage limit and exit"
+msgstr " --info-memory zobrazit omezení využití paměti a skončit"
+
+#: src/xz/message.c:1170
+msgid ""
+" -h, --help display the short help (lists only the basic options)\n"
+" -H, --long-help display this long help and exit"
+msgstr ""
+" -h, --help zobrazit krátkou nápovědu (vypíše jen základní "
+"přepínače)\n"
+" -H, --long-help zobrazit tuto úplnou nápovědu a skončit"
+
+#: src/xz/message.c:1174
+msgid ""
+" -h, --help display this short help and exit\n"
+" -H, --long-help display the long help (lists also the advanced options)"
+msgstr ""
+" -h, --help zobrazit tuto zkrácenou nápovědu a skončit\n"
+" -H, --long-help zobrazit úplnou nápovědu (vypíše i pokročilé "
+"přepínače)"
+
+#: src/xz/message.c:1179
+msgid " -V, --version display the version number and exit"
+msgstr " -V, --version zobrazit číslo verze a skončit"
+
+#: src/xz/message.c:1181
+msgid ""
+"\n"
+"With no FILE, or when FILE is -, read standard input.\n"
+msgstr ""
+"\n"
+"Pokud SOUBOR není zadán nebo pokud je -, bude se číst ze standardního "
+"vstupu.\n"
+
+#: src/xz/message.c:1185
+#, c-format
+msgid ""
+"On this system and configuration, this program will use a maximum of "
+"roughly\n"
+"%s MiB RAM and "
+msgstr ""
+"Na tomto systému a s tímto nastavením použije tento program maximum ze "
+"zhruba\n"
+"%s MiB RAM a "
+
+#: src/xz/message.c:1187
+msgid ""
+"one thread.\n"
+"\n"
+msgstr ""
+"jedno vlákno.\n"
+"\n"
+
+#: src/xz/message.c:1192
+#, c-format
+msgid "Report bugs to <%s> (in English or Finnish).\n"
+msgstr "Chyby hlaste na <%s> (v angličtině nebo finštině).\n"
+
+#: src/xz/message.c:1194
+#, c-format
+msgid "%s home page: <%s>\n"
+msgstr "Domovská stránka %s: <%s>\n"
+
+#: src/xz/options.c:86
+#, c-format
+msgid "%s: Options must be `name=value' pairs separated with commas"
+msgstr "%s: Volby musí být páry „název=hodnota“ oddělené čárkami"
+
+#: src/xz/options.c:93
+#, c-format
+msgid "%s: Invalid option name"
+msgstr "%s: Neplatný název volby"
+
+#: src/xz/options.c:113
+#, c-format
+msgid "%s: Invalid option value"
+msgstr "%s: Neplatná hodnota volby"
+
+#: src/xz/options.c:308
+#, c-format
+msgid "Unsupported LZMA1/LZMA2 preset: %s"
+msgstr "Nepodporované přednastavení LZMA1/LZMA2: %s"
+
+#: src/xz/options.c:426
+msgid "The sum of lc and lp must be at maximum of 4"
+msgstr "Součet lc a lp musí být maximálně 4"
+
+#: src/xz/options.c:431
+#, c-format
+msgid "The selected match finder requires at least nice=%<PRIu32>"
+msgstr "Vybraný vyhledávač shod vyžaduje minimálně nice=%<PRIu32>"
+
+#: src/xz/suffix.c:79 src/xz/suffix.c:164
+#, c-format
+msgid ""
+"%s: With --format=raw, --suffix=.SUF is required unless writing to stdout"
+msgstr ""
+"%s: S --format=raw je vyžadováno --sufix=.SUF, vyjma zápisu do standardního "
+"výstupu"
+
+#: src/xz/suffix.c:99
+#, c-format
+msgid "%s: Filename has an unknown suffix, skipping"
+msgstr "%s: Název souboru má neznámou příponu, vynechává se"
+
+#: src/xz/suffix.c:154
+#, c-format
+msgid "%s: File already has `%s' suffix, skipping"
+msgstr "%s: Soubor již má příponu „%s“, vynechává se"
+
+#: src/xz/suffix.c:205
+#, c-format
+msgid "%s: Invalid filename suffix"
+msgstr "%s: Neplatná přípona názvu souboru"
+
+#: src/xz/util.c:53
+#, c-format
+msgid "%s: Value is not a non-negative decimal integer"
+msgstr "%s: Hodnota není nezáporné desítkové číslo"
+
+#: src/xz/util.c:95
+#, c-format
+msgid "%s: Invalid multiplier suffix. Valid suffixes:"
+msgstr "%s: Neplatná přípona. Platné přípony jsou:"
+
+#: src/xz/util.c:115
+#, c-format
+msgid "Value of the option `%s' must be in the range [%<PRIu64>, %<PRIu64>]"
+msgstr "Hodnota volby „%s“ musí být v rozsahu [%<PRIu64>, %<PRIu64>]"
+
+#: src/xz/util.c:212
+msgid "Empty filename, skipping"
+msgstr "Prázdný název souboru, vynechává se"
+
+#: src/xz/util.c:226
+msgid "Compressed data not read from a terminal unless `--force' is used."
+msgstr ""
+"Komprimovaná data se nečtou z terminálu, vyjma použití volby „--force“."
+
+#: src/xz/util.c:239
+msgid "Compressed data not written to a terminal unless `--force' is used."
+msgstr ""
+"Komprimovaná data se nezapisují do terminálu, vyjma použití volby „--force“."
diff --git a/src/Makefile.am b/src/Makefile.am
new file mode 100644
index 0000000..f03f5a3
--- /dev/null
+++ b/src/Makefile.am
@@ -0,0 +1,9 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+SUBDIRS = liblzma xz xzdec lzmainfo scripts
+EXTRA_DIST = common
diff --git a/src/common/common_w32res.rc b/src/common/common_w32res.rc
new file mode 100644
index 0000000..ad9e165
--- /dev/null
+++ b/src/common/common_w32res.rc
@@ -0,0 +1,53 @@
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#include <winresrc.h>
+#include "config.h"
+#define LZMA_H_INTERNAL
+#define LZMA_H_INTERNAL_RC
+#include "lzma/version.h"
+
+#ifndef MY_BUILD
+# define MY_BUILD 0
+#endif
+#define MY_VERSION LZMA_VERSION_MAJOR,LZMA_VERSION_MINOR,LZMA_VERSION_PATCH,MY_BUILD
+
+#define MY_FILENAME MY_NAME MY_SUFFIX
+#define MY_COMPANY "Lasse Collin and Igor Pavlov"
+#define MY_COPYRIGHT "Public Domain by " MY_COMPANY
+#define MY_COMMENTS "http://tukaani.org/xz/"
+
+LANGUAGE LANG_ENGLISH, SUBLANG_ENGLISH_US
+VS_VERSION_INFO VERSIONINFO
+ FILEVERSION MY_VERSION
+ PRODUCTVERSION MY_VERSION
+ FILEFLAGSMASK VS_FFI_FILEFLAGSMASK
+ FILEFLAGS 0
+ FILEOS VOS_NT_WINDOWS32
+ FILETYPE MY_TYPE
+ FILESUBTYPE 0x0L
+BEGIN
+ BLOCK "StringFileInfo"
+ BEGIN
+ BLOCK "040904b0"
+ BEGIN
+ VALUE "Comments", MY_COMMENTS
+ VALUE "CompanyName", MY_COMPANY
+ VALUE "FileDescription", MY_DESC
+ VALUE "FileVersion", LZMA_VERSION_STRING
+ VALUE "InternalName", MY_NAME
+ VALUE "LegalCopyright", MY_COPYRIGHT
+ VALUE "OriginalFilename", MY_FILENAME
+ VALUE "ProductName", PACKAGE_NAME
+ VALUE "ProductVersion", LZMA_VERSION_STRING
+ END
+ END
+ BLOCK "VarFileInfo"
+ BEGIN
+ VALUE "Translation", 0x409, 1200
+ END
+END
diff --git a/src/common/mythread.h b/src/common/mythread.h
new file mode 100644
index 0000000..476c2fc
--- /dev/null
+++ b/src/common/mythread.h
@@ -0,0 +1,42 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file mythread.h
+/// \brief Wrappers for threads
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+
+
+#ifdef HAVE_PTHREAD
+# include <pthread.h>
+
+# define mythread_once(func) \
+ do { \
+ static pthread_once_t once_ = PTHREAD_ONCE_INIT; \
+ pthread_once(&once_, &func); \
+ } while (0)
+
+# define mythread_sigmask(how, set, oset) \
+ pthread_sigmask(how, set, oset)
+
+#else
+
+# define mythread_once(func) \
+ do { \
+ static bool once_ = false; \
+ if (!once_) { \
+ func(); \
+ once_ = true; \
+ } \
+ } while (0)
+
+# define mythread_sigmask(how, set, oset) \
+ sigprocmask(how, set, oset)
+
+#endif
diff --git a/src/common/sysdefs.h b/src/common/sysdefs.h
new file mode 100644
index 0000000..1e138b1
--- /dev/null
+++ b/src/common/sysdefs.h
@@ -0,0 +1,169 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file sysdefs.h
+/// \brief Common includes, definitions, system-specific things etc.
+///
+/// This file is used also by the lzma command line tool, that's why this
+/// file is separate from common.h.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_SYSDEFS_H
+#define LZMA_SYSDEFS_H
+
+//////////////
+// Includes //
+//////////////
+
+#ifdef HAVE_CONFIG_H
+# include <config.h>
+#endif
+
+// size_t and NULL
+#include <stddef.h>
+
+#ifdef HAVE_INTTYPES_H
+# include <inttypes.h>
+#endif
+
+// C99 says that inttypes.h always includes stdint.h, but some systems
+// don't do that, and require including stdint.h separately.
+#ifdef HAVE_STDINT_H
+# include <stdint.h>
+#endif
+
+// Some pre-C99 systems have SIZE_MAX in limits.h instead of stdint.h. The
+// limits are also used to figure out some macros missing from pre-C99 systems.
+#ifdef HAVE_LIMITS_H
+# include <limits.h>
+#endif
+
+// Be more compatible with systems that have non-conforming inttypes.h.
+// We assume that int is 32-bit and that long is either 32-bit or 64-bit.
+// Full Autoconf test could be more correct, but this should work well enough.
+// Note that this duplicates some code from lzma.h, but this is better since
+// we can work without inttypes.h thanks to Autoconf tests.
+#ifndef UINT32_C
+# if UINT_MAX != 4294967295U
+# error UINT32_C is not defined and unsigned int is not 32-bit.
+# endif
+# define UINT32_C(n) n ## U
+#endif
+#ifndef UINT32_MAX
+# define UINT32_MAX UINT32_C(4294967295)
+#endif
+#ifndef PRIu32
+# define PRIu32 "u"
+#endif
+#ifndef PRIX32
+# define PRIX32 "X"
+#endif
+
+#if ULONG_MAX == 4294967295UL
+# ifndef UINT64_C
+# define UINT64_C(n) n ## ULL
+# endif
+# ifndef PRIu64
+# define PRIu64 "llu"
+# endif
+# ifndef PRIX64
+# define PRIX64 "llX"
+# endif
+#else
+# ifndef UINT64_C
+# define UINT64_C(n) n ## UL
+# endif
+# ifndef PRIu64
+# define PRIu64 "lu"
+# endif
+# ifndef PRIX64
+# define PRIX64 "lX"
+# endif
+#endif
+#ifndef UINT64_MAX
+# define UINT64_MAX UINT64_C(18446744073709551615)
+#endif
+
+// Interix has broken header files, which typedef size_t to unsigned long,
+// but a few lines later define SIZE_MAX to INT32_MAX.
+#ifdef __INTERIX
+# undef SIZE_MAX
+#endif
+
+// The code currently assumes that size_t is either 32-bit or 64-bit.
+#ifndef SIZE_MAX
+# if SIZEOF_SIZE_T == 4
+# define SIZE_MAX UINT32_MAX
+# elif SIZEOF_SIZE_T == 8
+# define SIZE_MAX UINT64_MAX
+# else
+# error size_t is not 32-bit or 64-bit
+# endif
+#endif
+#if SIZE_MAX != UINT32_MAX && SIZE_MAX != UINT64_MAX
+# error size_t is not 32-bit or 64-bit
+#endif
+
+#include <stdlib.h>
+#include <assert.h>
+
+// Pre-C99 systems lack stdbool.h. All the code in LZMA Utils must be written
+// so that it works with fake bool type, for example:
+//
+// bool foo = (flags & 0x100) != 0;
+// bool bar = !!(flags & 0x100);
+//
+// This works with the real C99 bool but breaks with fake bool:
+//
+// bool baz = (flags & 0x100);
+//
+#ifdef HAVE_STDBOOL_H
+# include <stdbool.h>
+#else
+# if ! HAVE__BOOL
+typedef unsigned char _Bool;
+# endif
+# define bool _Bool
+# define false 0
+# define true 1
+# define __bool_true_false_are_defined 1
+#endif
+
+// string.h should be enough but let's include strings.h and memory.h too if
+// they exists, since that shouldn't do any harm, but may improve portability.
+#ifdef HAVE_STRING_H
+# include <string.h>
+#endif
+
+#ifdef HAVE_STRINGS_H
+# include <strings.h>
+#endif
+
+#ifdef HAVE_MEMORY_H
+# include <memory.h>
+#endif
+
+
+////////////
+// Macros //
+////////////
+
+#undef memzero
+#define memzero(s, n) memset(s, 0, n)
+
+// NOTE: Avoid using MIN() and MAX(), because even conditionally defining
+// those macros can cause some portability trouble, since on some systems
+// the system headers insist defining their own versions.
+#define my_min(x, y) ((x) < (y) ? (x) : (y))
+#define my_max(x, y) ((x) > (y) ? (x) : (y))
+
+#ifndef ARRAY_SIZE
+# define ARRAY_SIZE(array) (sizeof(array) / sizeof((array)[0]))
+#endif
+
+#endif
diff --git a/src/common/tuklib_common.h b/src/common/tuklib_common.h
new file mode 100644
index 0000000..31fbab5
--- /dev/null
+++ b/src/common/tuklib_common.h
@@ -0,0 +1,71 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_common.h
+/// \brief Common definitions for tuklib modules
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_COMMON_H
+#define TUKLIB_COMMON_H
+
+// The config file may be replaced by a package-specific file.
+// It should include at least stddef.h, inttypes.h, and limits.h.
+#include "tuklib_config.h"
+
+// TUKLIB_SYMBOL_PREFIX is prefixed to all symbols exported by
+// the tuklib modules. If you use a tuklib module in a library,
+// you should use TUKLIB_SYMBOL_PREFIX to make sure that there
+// are no symbol conflicts in case someone links your library
+// into application that also uses the same tuklib module.
+#ifndef TUKLIB_SYMBOL_PREFIX
+# define TUKLIB_SYMBOL_PREFIX
+#endif
+
+#define TUKLIB_CAT_X(a, b) a ## b
+#define TUKLIB_CAT(a, b) TUKLIB_CAT_X(a, b)
+
+#ifndef TUKLIB_SYMBOL
+# define TUKLIB_SYMBOL(sym) TUKLIB_CAT(TUKLIB_SYMBOL_PREFIX, sym)
+#endif
+
+#ifndef TUKLIB_DECLS_BEGIN
+# ifdef __cplusplus
+# define TUKLIB_DECLS_BEGIN extern "C" {
+# else
+# define TUKLIB_DECLS_BEGIN
+# endif
+#endif
+
+#ifndef TUKLIB_DECLS_END
+# ifdef __cplusplus
+# define TUKLIB_DECLS_END }
+# else
+# define TUKLIB_DECLS_END
+# endif
+#endif
+
+#if defined(__GNUC__) && defined(__GNUC_MINOR__)
+# define TUKLIB_GNUC_REQ(major, minor) \
+ ((__GNUC__ == (major) && __GNUC_MINOR__ >= (minor)) \
+ || __GNUC__ > (major))
+#else
+# define TUKLIB_GNUC_REQ(major, minor) 0
+#endif
+
+#if TUKLIB_GNUC_REQ(2, 5)
+# define tuklib_attr_noreturn __attribute__((__noreturn__))
+#else
+# define tuklib_attr_noreturn
+#endif
+
+#if (defined(_WIN32) && !defined(__CYGWIN__)) \
+ || defined(__OS2__) || defined(__MSDOS__)
+# define TUKLIB_DOSLIKE 1
+#endif
+
+#endif
diff --git a/src/common/tuklib_config.h b/src/common/tuklib_config.h
new file mode 100644
index 0000000..549cb24
--- /dev/null
+++ b/src/common/tuklib_config.h
@@ -0,0 +1,7 @@
+#ifdef HAVE_CONFIG_H
+# include "sysdefs.h"
+#else
+# include <stddef.h>
+# include <inttypes.h>
+# include <limits.h>
+#endif
diff --git a/src/common/tuklib_cpucores.c b/src/common/tuklib_cpucores.c
new file mode 100644
index 0000000..1da13df
--- /dev/null
+++ b/src/common/tuklib_cpucores.c
@@ -0,0 +1,62 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_cpucores.c
+/// \brief Get the number of CPU cores online
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tuklib_cpucores.h"
+
+#if defined(TUKLIB_CPUCORES_SYSCTL)
+# ifdef HAVE_SYS_PARAM_H
+# include <sys/param.h>
+# endif
+# include <sys/sysctl.h>
+
+#elif defined(TUKLIB_CPUCORES_SYSCONF)
+# include <unistd.h>
+
+// HP-UX
+#elif defined(TUKLIB_CPUCORES_PSTAT_GETDYNAMIC)
+# include <sys/param.h>
+# include <sys/pstat.h>
+#endif
+
+
+extern uint32_t
+tuklib_cpucores(void)
+{
+ uint32_t ret = 0;
+
+#if defined(TUKLIB_CPUCORES_SYSCTL)
+ int name[2] = { CTL_HW, HW_NCPU };
+ int cpus;
+ size_t cpus_size = sizeof(cpus);
+ if (sysctl(name, 2, &cpus, &cpus_size, NULL, 0) != -1
+ && cpus_size == sizeof(cpus) && cpus > 0)
+ ret = cpus;
+
+#elif defined(TUKLIB_CPUCORES_SYSCONF)
+# ifdef _SC_NPROCESSORS_ONLN
+ // Most systems
+ const long cpus = sysconf(_SC_NPROCESSORS_ONLN);
+# else
+ // IRIX
+ const long cpus = sysconf(_SC_NPROC_ONLN);
+# endif
+ if (cpus > 0)
+ ret = cpus;
+
+#elif defined(TUKLIB_CPUCORES_PSTAT_GETDYNAMIC)
+ struct pst_dynamic pst;
+ if (pstat_getdynamic(&pst, sizeof(pst), 1, 0) != -1)
+ ret = pst.psd_proc_cnt;
+#endif
+
+ return ret;
+}
diff --git a/src/common/tuklib_cpucores.h b/src/common/tuklib_cpucores.h
new file mode 100644
index 0000000..be1ce1c
--- /dev/null
+++ b/src/common/tuklib_cpucores.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_cpucores.h
+/// \brief Get the number of CPU cores online
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_CPUCORES_H
+#define TUKLIB_CPUCORES_H
+
+#include "tuklib_common.h"
+TUKLIB_DECLS_BEGIN
+
+#define tuklib_cpucores TUKLIB_SYMBOL(tuklib_cpucores)
+extern uint32_t tuklib_cpucores(void);
+
+TUKLIB_DECLS_END
+#endif
diff --git a/src/common/tuklib_exit.c b/src/common/tuklib_exit.c
new file mode 100644
index 0000000..c393be6
--- /dev/null
+++ b/src/common/tuklib_exit.c
@@ -0,0 +1,57 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_exit.c
+/// \brief Close stdout and stderr, and exit
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tuklib_common.h"
+
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "tuklib_gettext.h"
+#include "tuklib_progname.h"
+#include "tuklib_exit.h"
+
+
+extern void
+tuklib_exit(int status, int err_status, int show_error)
+{
+ if (status != err_status) {
+ // Close stdout. If something goes wrong,
+ // print an error message to stderr.
+ const int ferror_err = ferror(stdout);
+ const int fclose_err = fclose(stdout);
+ if (ferror_err || fclose_err) {
+ status = err_status;
+
+ // If it was fclose() that failed, we have the reason
+ // in errno. If only ferror() indicated an error,
+ // we have no idea what the reason was.
+ if (show_error)
+ fprintf(stderr, "%s: %s: %s\n", progname,
+ _("Writing to standard "
+ "output failed"),
+ fclose_err ? strerror(errno)
+ : _("Unknown error"));
+ }
+ }
+
+ if (status != err_status) {
+ // Close stderr. If something goes wrong, there's
+ // nothing where we could print an error message.
+ // Just set the exit status.
+ const int ferror_err = ferror(stderr);
+ const int fclose_err = fclose(stderr);
+ if (fclose_err || ferror_err)
+ status = err_status;
+ }
+
+ exit(status);
+}
diff --git a/src/common/tuklib_exit.h b/src/common/tuklib_exit.h
new file mode 100644
index 0000000..b11776f
--- /dev/null
+++ b/src/common/tuklib_exit.h
@@ -0,0 +1,25 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_exit.h
+/// \brief Close stdout and stderr, and exit
+/// \note Requires tuklib_progname and tuklib_gettext modules
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_EXIT_H
+#define TUKLIB_EXIT_H
+
+#include "tuklib_common.h"
+TUKLIB_DECLS_BEGIN
+
+#define tuklib_exit TUKLIB_SYMBOL(tuklib_exit)
+extern void tuklib_exit(int status, int err_status, int show_error)
+ tuklib_attr_noreturn;
+
+TUKLIB_DECLS_END
+#endif
diff --git a/src/common/tuklib_gettext.h b/src/common/tuklib_gettext.h
new file mode 100644
index 0000000..2485213
--- /dev/null
+++ b/src/common/tuklib_gettext.h
@@ -0,0 +1,44 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_gettext.h
+/// \brief Wrapper for gettext and friends
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_GETTEXT_H
+#define TUKLIB_GETTEXT_H
+
+#include "tuklib_common.h"
+#include <locale.h>
+
+#ifndef TUKLIB_GETTEXT
+# ifdef ENABLE_NLS
+# define TUKLIB_GETTEXT 1
+# else
+# define TUKLIB_GETTEXT 0
+# endif
+#endif
+
+#if TUKLIB_GETTEXT
+# include <libintl.h>
+# define tuklib_gettext_init(package, localedir) \
+ do { \
+ setlocale(LC_ALL, ""); \
+ bindtextdomain(package, localedir); \
+ textdomain(package); \
+ } while (0)
+# define _(msgid) gettext(msgid)
+# define N_(msgid1, msgid2, n) ngettext(msgid1, msgid2, n)
+#else
+# define tuklib_gettext_init(package, localedir) \
+ setlocale(LC_ALL, "")
+# define _(msgid) (msgid)
+# define N_(msgid1, msgid2, n) ((n) == 1 ? (msgid1) : (msgid2))
+#endif
+
+#endif
diff --git a/src/common/tuklib_integer.h b/src/common/tuklib_integer.h
new file mode 100644
index 0000000..e6daa77
--- /dev/null
+++ b/src/common/tuklib_integer.h
@@ -0,0 +1,523 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_integer.h
+/// \brief Various integer and bit operations
+///
+/// This file provides macros or functions to do some basic integer and bit
+/// operations.
+///
+/// Endianness related integer operations (XX = 16, 32, or 64; Y = b or l):
+/// - Byte swapping: bswapXX(num)
+/// - Byte order conversions to/from native: convXXYe(num)
+/// - Aligned reads: readXXYe(ptr)
+/// - Aligned writes: writeXXYe(ptr, num)
+/// - Unaligned reads (16/32-bit only): unaligned_readXXYe(ptr)
+/// - Unaligned writes (16/32-bit only): unaligned_writeXXYe(ptr, num)
+///
+/// Since they can macros, the arguments should have no side effects since
+/// they may be evaluated more than once.
+///
+/// \todo PowerPC and possibly some other architectures support
+/// byte swapping load and store instructions. This file
+/// doesn't take advantage of those instructions.
+///
+/// Bit scan operations for non-zero 32-bit integers:
+/// - Bit scan reverse (find highest non-zero bit): bsr32(num)
+/// - Count leading zeros: clz32(num)
+/// - Count trailing zeros: ctz32(num)
+/// - Bit scan forward (simply an alias for ctz32()): bsf32(num)
+///
+/// The above bit scan operations return 0-31. If num is zero,
+/// the result is undefined.
+//
+// Authors: Lasse Collin
+// Joachim Henke
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_INTEGER_H
+#define TUKLIB_INTEGER_H
+
+#include "tuklib_common.h"
+
+
+////////////////////////////////////////
+// Operating system specific features //
+////////////////////////////////////////
+
+#if defined(HAVE_BYTESWAP_H)
+ // glibc, uClibc, dietlibc
+# include <byteswap.h>
+# ifdef HAVE_BSWAP_16
+# define bswap16(num) bswap_16(num)
+# endif
+# ifdef HAVE_BSWAP_32
+# define bswap32(num) bswap_32(num)
+# endif
+# ifdef HAVE_BSWAP_64
+# define bswap64(num) bswap_64(num)
+# endif
+
+#elif defined(HAVE_SYS_ENDIAN_H)
+ // *BSDs and Darwin
+# include <sys/endian.h>
+
+#elif defined(HAVE_SYS_BYTEORDER_H)
+ // Solaris
+# include <sys/byteorder.h>
+# ifdef BSWAP_16
+# define bswap16(num) BSWAP_16(num)
+# endif
+# ifdef BSWAP_32
+# define bswap32(num) BSWAP_32(num)
+# endif
+# ifdef BSWAP_64
+# define bswap64(num) BSWAP_64(num)
+# endif
+# ifdef BE_16
+# define conv16be(num) BE_16(num)
+# endif
+# ifdef BE_32
+# define conv32be(num) BE_32(num)
+# endif
+# ifdef BE_64
+# define conv64be(num) BE_64(num)
+# endif
+# ifdef LE_16
+# define conv16le(num) LE_16(num)
+# endif
+# ifdef LE_32
+# define conv32le(num) LE_32(num)
+# endif
+# ifdef LE_64
+# define conv64le(num) LE_64(num)
+# endif
+#endif
+
+
+///////////////////
+// Byte swapping //
+///////////////////
+
+#ifndef bswap16
+# define bswap16(num) \
+ (((uint16_t)(num) << 8) | ((uint16_t)(num) >> 8))
+#endif
+
+#ifndef bswap32
+# define bswap32(num) \
+ ( (((uint32_t)(num) << 24) ) \
+ | (((uint32_t)(num) << 8) & UINT32_C(0x00FF0000)) \
+ | (((uint32_t)(num) >> 8) & UINT32_C(0x0000FF00)) \
+ | (((uint32_t)(num) >> 24) ) )
+#endif
+
+#ifndef bswap64
+# define bswap64(num) \
+ ( (((uint64_t)(num) << 56) ) \
+ | (((uint64_t)(num) << 40) & UINT64_C(0x00FF000000000000)) \
+ | (((uint64_t)(num) << 24) & UINT64_C(0x0000FF0000000000)) \
+ | (((uint64_t)(num) << 8) & UINT64_C(0x000000FF00000000)) \
+ | (((uint64_t)(num) >> 8) & UINT64_C(0x00000000FF000000)) \
+ | (((uint64_t)(num) >> 24) & UINT64_C(0x0000000000FF0000)) \
+ | (((uint64_t)(num) >> 40) & UINT64_C(0x000000000000FF00)) \
+ | (((uint64_t)(num) >> 56) ) )
+#endif
+
+// Define conversion macros using the basic byte swapping macros.
+#ifdef WORDS_BIGENDIAN
+# ifndef conv16be
+# define conv16be(num) ((uint16_t)(num))
+# endif
+# ifndef conv32be
+# define conv32be(num) ((uint32_t)(num))
+# endif
+# ifndef conv64be
+# define conv64be(num) ((uint64_t)(num))
+# endif
+# ifndef conv16le
+# define conv16le(num) bswap16(num)
+# endif
+# ifndef conv32le
+# define conv32le(num) bswap32(num)
+# endif
+# ifndef conv64le
+# define conv64le(num) bswap64(num)
+# endif
+#else
+# ifndef conv16be
+# define conv16be(num) bswap16(num)
+# endif
+# ifndef conv32be
+# define conv32be(num) bswap32(num)
+# endif
+# ifndef conv64be
+# define conv64be(num) bswap64(num)
+# endif
+# ifndef conv16le
+# define conv16le(num) ((uint16_t)(num))
+# endif
+# ifndef conv32le
+# define conv32le(num) ((uint32_t)(num))
+# endif
+# ifndef conv64le
+# define conv64le(num) ((uint64_t)(num))
+# endif
+#endif
+
+
+//////////////////////////////
+// Aligned reads and writes //
+//////////////////////////////
+
+static inline uint16_t
+read16be(const uint8_t *buf)
+{
+ uint16_t num = *(const uint16_t *)buf;
+ return conv16be(num);
+}
+
+
+static inline uint16_t
+read16le(const uint8_t *buf)
+{
+ uint16_t num = *(const uint16_t *)buf;
+ return conv16le(num);
+}
+
+
+static inline uint32_t
+read32be(const uint8_t *buf)
+{
+ uint32_t num = *(const uint32_t *)buf;
+ return conv32be(num);
+}
+
+
+static inline uint32_t
+read32le(const uint8_t *buf)
+{
+ uint32_t num = *(const uint32_t *)buf;
+ return conv32le(num);
+}
+
+
+static inline uint64_t
+read64be(const uint8_t *buf)
+{
+ uint64_t num = *(const uint64_t *)buf;
+ return conv64be(num);
+}
+
+
+static inline uint64_t
+read64le(const uint8_t *buf)
+{
+ uint64_t num = *(const uint64_t *)buf;
+ return conv64le(num);
+}
+
+
+// NOTE: Possible byte swapping must be done in a macro to allow GCC
+// to optimize byte swapping of constants when using glibc's or *BSD's
+// byte swapping macros. The actual write is done in an inline function
+// to make type checking of the buf pointer possible similarly to readXXYe()
+// functions.
+
+#define write16be(buf, num) write16ne((buf), conv16be(num))
+#define write16le(buf, num) write16ne((buf), conv16le(num))
+#define write32be(buf, num) write32ne((buf), conv32be(num))
+#define write32le(buf, num) write32ne((buf), conv32le(num))
+#define write64be(buf, num) write64ne((buf), conv64be(num))
+#define write64le(buf, num) write64ne((buf), conv64le(num))
+
+
+static inline void
+write16ne(uint8_t *buf, uint16_t num)
+{
+ *(uint16_t *)buf = num;
+ return;
+}
+
+
+static inline void
+write32ne(uint8_t *buf, uint32_t num)
+{
+ *(uint32_t *)buf = num;
+ return;
+}
+
+
+static inline void
+write64ne(uint8_t *buf, uint64_t num)
+{
+ *(uint64_t *)buf = num;
+ return;
+}
+
+
+////////////////////////////////
+// Unaligned reads and writes //
+////////////////////////////////
+
+// NOTE: TUKLIB_FAST_UNALIGNED_ACCESS indicates only support for 16-bit and
+// 32-bit unaligned integer loads and stores. It's possible that 64-bit
+// unaligned access doesn't work or is slower than byte-by-byte access.
+// Since unaligned 64-bit is probably not needed as often as 16-bit or
+// 32-bit, we simply don't support 64-bit unaligned access for now.
+#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
+# define unaligned_read16be read16be
+# define unaligned_read16le read16le
+# define unaligned_read32be read32be
+# define unaligned_read32le read32le
+# define unaligned_write16be write16be
+# define unaligned_write16le write16le
+# define unaligned_write32be write32be
+# define unaligned_write32le write32le
+
+#else
+
+static inline uint16_t
+unaligned_read16be(const uint8_t *buf)
+{
+ uint16_t num = ((uint16_t)buf[0] << 8) | (uint16_t)buf[1];
+ return num;
+}
+
+
+static inline uint16_t
+unaligned_read16le(const uint8_t *buf)
+{
+ uint16_t num = ((uint16_t)buf[0]) | ((uint16_t)buf[1] << 8);
+ return num;
+}
+
+
+static inline uint32_t
+unaligned_read32be(const uint8_t *buf)
+{
+ uint32_t num = (uint32_t)buf[0] << 24;
+ num |= (uint32_t)buf[1] << 16;
+ num |= (uint32_t)buf[2] << 8;
+ num |= (uint32_t)buf[3];
+ return num;
+}
+
+
+static inline uint32_t
+unaligned_read32le(const uint8_t *buf)
+{
+ uint32_t num = (uint32_t)buf[0];
+ num |= (uint32_t)buf[1] << 8;
+ num |= (uint32_t)buf[2] << 16;
+ num |= (uint32_t)buf[3] << 24;
+ return num;
+}
+
+
+static inline void
+unaligned_write16be(uint8_t *buf, uint16_t num)
+{
+ buf[0] = num >> 8;
+ buf[1] = num;
+ return;
+}
+
+
+static inline void
+unaligned_write16le(uint8_t *buf, uint16_t num)
+{
+ buf[0] = num;
+ buf[1] = num >> 8;
+ return;
+}
+
+
+static inline void
+unaligned_write32be(uint8_t *buf, uint32_t num)
+{
+ buf[0] = num >> 24;
+ buf[1] = num >> 16;
+ buf[2] = num >> 8;
+ buf[3] = num;
+ return;
+}
+
+
+static inline void
+unaligned_write32le(uint8_t *buf, uint32_t num)
+{
+ buf[0] = num;
+ buf[1] = num >> 8;
+ buf[2] = num >> 16;
+ buf[3] = num >> 24;
+ return;
+}
+
+#endif
+
+
+static inline uint32_t
+bsr32(uint32_t n)
+{
+ // Check for ICC first, since it tends to define __GNUC__ too.
+#if defined(__INTEL_COMPILER)
+ return _bit_scan_reverse(n);
+
+#elif TUKLIB_GNUC_REQ(3, 4) && UINT_MAX == UINT32_MAX
+ // GCC >= 3.4 has __builtin_clz(), which gives good results on
+ // multiple architectures. On x86, __builtin_clz() ^ 31U becomes
+ // either plain BSR (so the XOR gets optimized away) or LZCNT and
+ // XOR (if -march indicates that SSE4a instructions are supported).
+ return __builtin_clz(n) ^ 31U;
+
+#elif defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__))
+ uint32_t i;
+ __asm__("bsrl %1, %0" : "=r" (i) : "rm" (n));
+ return i;
+
+#elif defined(_MSC_VER) && _MSC_VER >= 1400
+ // MSVC isn't supported by tuklib, but since this code exists,
+ // it doesn't hurt to have it here anyway.
+ uint32_t i;
+ _BitScanReverse((DWORD *)&i, n);
+ return i;
+
+#else
+ uint32_t i = 31;
+
+ if ((n & UINT32_C(0xFFFF0000)) == 0) {
+ n <<= 16;
+ i = 15;
+ }
+
+ if ((n & UINT32_C(0xFF000000)) == 0) {
+ n <<= 8;
+ i -= 8;
+ }
+
+ if ((n & UINT32_C(0xF0000000)) == 0) {
+ n <<= 4;
+ i -= 4;
+ }
+
+ if ((n & UINT32_C(0xC0000000)) == 0) {
+ n <<= 2;
+ i -= 2;
+ }
+
+ if ((n & UINT32_C(0x80000000)) == 0)
+ --i;
+
+ return i;
+#endif
+}
+
+
+static inline uint32_t
+clz32(uint32_t n)
+{
+#if defined(__INTEL_COMPILER)
+ return _bit_scan_reverse(n) ^ 31U;
+
+#elif TUKLIB_GNUC_REQ(3, 4) && UINT_MAX == UINT32_MAX
+ return __builtin_clz(n);
+
+#elif defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__))
+ uint32_t i;
+ __asm__("bsrl %1, %0\n\t"
+ "xorl $31, %0"
+ : "=r" (i) : "rm" (n));
+ return i;
+
+#elif defined(_MSC_VER) && _MSC_VER >= 1400
+ uint32_t i;
+ _BitScanReverse((DWORD *)&i, n);
+ return i ^ 31U;
+
+#else
+ uint32_t i = 0;
+
+ if ((n & UINT32_C(0xFFFF0000)) == 0) {
+ n <<= 16;
+ i = 16;
+ }
+
+ if ((n & UINT32_C(0xFF000000)) == 0) {
+ n <<= 8;
+ i += 8;
+ }
+
+ if ((n & UINT32_C(0xF0000000)) == 0) {
+ n <<= 4;
+ i += 4;
+ }
+
+ if ((n & UINT32_C(0xC0000000)) == 0) {
+ n <<= 2;
+ i += 2;
+ }
+
+ if ((n & UINT32_C(0x80000000)) == 0)
+ ++i;
+
+ return i;
+#endif
+}
+
+
+static inline uint32_t
+ctz32(uint32_t n)
+{
+#if defined(__INTEL_COMPILER)
+ return _bit_scan_forward(n);
+
+#elif TUKLIB_GNUC_REQ(3, 4) && UINT_MAX >= UINT32_MAX
+ return __builtin_ctz(n);
+
+#elif defined(__GNUC__) && (defined(__i386__) || defined(__x86_64__))
+ uint32_t i;
+ __asm__("bsfl %1, %0" : "=r" (i) : "rm" (n));
+ return i;
+
+#elif defined(_MSC_VER) && _MSC_VER >= 1400
+ uint32_t i;
+ _BitScanForward((DWORD *)&i, n);
+ return i;
+
+#else
+ uint32_t i = 0;
+
+ if ((n & UINT32_C(0x0000FFFF)) == 0) {
+ n >>= 16;
+ i = 16;
+ }
+
+ if ((n & UINT32_C(0x000000FF)) == 0) {
+ n >>= 8;
+ i += 8;
+ }
+
+ if ((n & UINT32_C(0x0000000F)) == 0) {
+ n >>= 4;
+ i += 4;
+ }
+
+ if ((n & UINT32_C(0x00000003)) == 0) {
+ n >>= 2;
+ i += 2;
+ }
+
+ if ((n & UINT32_C(0x00000001)) == 0)
+ ++i;
+
+ return i;
+#endif
+}
+
+#define bsf32 ctz32
+
+#endif
diff --git a/src/common/tuklib_open_stdxxx.c b/src/common/tuklib_open_stdxxx.c
new file mode 100644
index 0000000..08bc60d
--- /dev/null
+++ b/src/common/tuklib_open_stdxxx.c
@@ -0,0 +1,55 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_open_stdxxx.c
+/// \brief Make sure that file descriptors 0, 1, and 2 are open
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tuklib_open_stdxxx.h"
+
+#ifndef TUKLIB_DOSLIKE
+# include <stdlib.h>
+# include <errno.h>
+# include <fcntl.h>
+# include <unistd.h>
+#endif
+
+
+extern void
+tuklib_open_stdxxx(int err_status)
+{
+#ifdef TUKLIB_DOSLIKE
+ // Do nothing, just silence warnings.
+ (void)err_status;
+
+#else
+ for (int i = 0; i <= 2; ++i) {
+ // We use fcntl() to check if the file descriptor is open.
+ if (fcntl(i, F_GETFD) == -1 && errno == EBADF) {
+ // With stdin, we could use /dev/full so that
+ // writing to stdin would fail. However, /dev/full
+ // is Linux specific, and if the program tries to
+ // write to stdin, there's already a problem anyway.
+ const int fd = open("/dev/null", O_NOCTTY
+ | (i == 0 ? O_WRONLY : O_RDONLY));
+
+ if (fd != i) {
+ // Something went wrong. Exit with the
+ // exit status we were given. Don't try
+ // to print an error message, since stderr
+ // may very well be non-existent. This
+ // error should be extremely rare.
+ (void)close(fd);
+ exit(err_status);
+ }
+ }
+ }
+#endif
+
+ return;
+}
diff --git a/src/common/tuklib_open_stdxxx.h b/src/common/tuklib_open_stdxxx.h
new file mode 100644
index 0000000..b911616
--- /dev/null
+++ b/src/common/tuklib_open_stdxxx.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_open_stdxxx.h
+/// \brief Make sure that file descriptors 0, 1, and 2 are open
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_OPEN_STDXXX_H
+#define TUKLIB_OPEN_STDXXX_H
+
+#include "tuklib_common.h"
+TUKLIB_DECLS_BEGIN
+
+#define tuklib_open_stdxx TUKLIB_SYMBOL(tuklib_open_stdxxx)
+extern void tuklib_open_stdxxx(int err_status);
+
+TUKLIB_DECLS_END
+#endif
diff --git a/src/common/tuklib_physmem.c b/src/common/tuklib_physmem.c
new file mode 100644
index 0000000..623b6e7
--- /dev/null
+++ b/src/common/tuklib_physmem.c
@@ -0,0 +1,196 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_physmem.c
+/// \brief Get the amount of physical memory
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tuklib_physmem.h"
+
+// We want to use Windows-specific code on Cygwin, which also has memory
+// information available via sysconf(), but on Cygwin 1.5 and older it
+// gives wrong results (from our point of view).
+#if defined(_WIN32) || defined(__CYGWIN__)
+# ifndef _WIN32_WINNT
+# define _WIN32_WINNT 0x0500
+# endif
+# include <windows.h>
+
+#elif defined(__OS2__)
+# define INCL_DOSMISC
+# include <os2.h>
+
+#elif defined(__DJGPP__)
+# include <dpmi.h>
+
+#elif defined(__VMS)
+# include <lib$routines.h>
+# include <syidef.h>
+# include <ssdef.h>
+
+// AIX
+#elif defined(TUKLIB_PHYSMEM_AIX)
+# include <sys/systemcfg.h>
+
+#elif defined(TUKLIB_PHYSMEM_SYSCONF)
+# include <unistd.h>
+
+#elif defined(TUKLIB_PHYSMEM_SYSCTL)
+# ifdef HAVE_SYS_PARAM_H
+# include <sys/param.h>
+# endif
+# include <sys/sysctl.h>
+
+// Tru64
+#elif defined(TUKLIB_PHYSMEM_GETSYSINFO)
+# include <sys/sysinfo.h>
+# include <machine/hal_sysinfo.h>
+
+// HP-UX
+#elif defined(TUKLIB_PHYSMEM_PSTAT_GETSTATIC)
+# include <sys/param.h>
+# include <sys/pstat.h>
+
+// IRIX
+#elif defined(TUKLIB_PHYSMEM_GETINVENT_R)
+# include <invent.h>
+
+// This sysinfo() is Linux-specific.
+#elif defined(TUKLIB_PHYSMEM_SYSINFO)
+# include <sys/sysinfo.h>
+#endif
+
+
+extern uint64_t
+tuklib_physmem(void)
+{
+ uint64_t ret = 0;
+
+#if defined(_WIN32) || defined(__CYGWIN__)
+ if ((GetVersion() & 0xFF) >= 5) {
+ // Windows 2000 and later have GlobalMemoryStatusEx() which
+ // supports reporting values greater than 4 GiB. To keep the
+ // code working also on older Windows versions, use
+ // GlobalMemoryStatusEx() conditionally.
+ HMODULE kernel32 = GetModuleHandle("kernel32.dll");
+ if (kernel32 != NULL) {
+ BOOL (WINAPI *gmse)(LPMEMORYSTATUSEX) = GetProcAddress(
+ kernel32, "GlobalMemoryStatusEx");
+ if (gmse != NULL) {
+ MEMORYSTATUSEX meminfo;
+ meminfo.dwLength = sizeof(meminfo);
+ if (gmse(&meminfo))
+ ret = meminfo.ullTotalPhys;
+ }
+ }
+ }
+
+ if (ret == 0) {
+ // GlobalMemoryStatus() is supported by Windows 95 and later,
+ // so it is fine to link against it unconditionally. Note that
+ // GlobalMemoryStatus() has no return value.
+ MEMORYSTATUS meminfo;
+ meminfo.dwLength = sizeof(meminfo);
+ GlobalMemoryStatus(&meminfo);
+ ret = meminfo.dwTotalPhys;
+ }
+
+#elif defined(__OS2__)
+ unsigned long mem;
+ if (DosQuerySysInfo(QSV_TOTPHYSMEM, QSV_TOTPHYSMEM,
+ &mem, sizeof(mem)) == 0)
+ ret = mem;
+
+#elif defined(__DJGPP__)
+ __dpmi_free_mem_info meminfo;
+ if (__dpmi_get_free_memory_information(&meminfo) == 0
+ && meminfo.total_number_of_physical_pages
+ != (unsigned long)-1)
+ ret = (uint64_t)meminfo.total_number_of_physical_pages * 4096;
+
+#elif defined(__VMS)
+ int vms_mem;
+ int val = SYI$_MEMSIZE;
+ if (LIB$GETSYI(&val, &vms_mem, 0, 0, 0, 0) == SS$_NORMAL)
+ ret = (uint64_t)vms_mem * 8192;
+
+#elif defined(TUKLIB_PHYSMEM_AIX)
+ ret = _system_configuration.physmem;
+
+#elif defined(TUKLIB_PHYSMEM_SYSCONF)
+ const long pagesize = sysconf(_SC_PAGESIZE);
+ const long pages = sysconf(_SC_PHYS_PAGES);
+ if (pagesize != -1 && pages != -1)
+ // According to docs, pagesize * pages can overflow.
+ // Simple case is 32-bit box with 4 GiB or more RAM,
+ // which may report exactly 4 GiB of RAM, and "long"
+ // being 32-bit will overflow. Casting to uint64_t
+ // hopefully avoids overflows in the near future.
+ ret = (uint64_t)pagesize * (uint64_t)pages;
+
+#elif defined(TUKLIB_PHYSMEM_SYSCTL)
+ int name[2] = {
+ CTL_HW,
+#ifdef HW_PHYSMEM64
+ HW_PHYSMEM64
+#else
+ HW_PHYSMEM
+#endif
+ };
+ union {
+ uint32_t u32;
+ uint64_t u64;
+ } mem;
+ size_t mem_ptr_size = sizeof(mem.u64);
+ if (sysctl(name, 2, &mem.u64, &mem_ptr_size, NULL, 0) != -1) {
+ // IIRC, 64-bit "return value" is possible on some 64-bit
+ // BSD systems even with HW_PHYSMEM (instead of HW_PHYSMEM64),
+ // so support both.
+ if (mem_ptr_size == sizeof(mem.u64))
+ ret = mem.u64;
+ else if (mem_ptr_size == sizeof(mem.u32))
+ ret = mem.u32;
+ }
+
+#elif defined(TUKLIB_PHYSMEM_GETSYSINFO)
+ // Docs are unclear if "start" is needed, but it doesn't hurt
+ // much to have it.
+ int memkb;
+ int start = 0;
+ if (getsysinfo(GSI_PHYSMEM, (caddr_t)&memkb, sizeof(memkb), &start)
+ != -1)
+ ret = (uint64_t)memkb * 1024;
+
+#elif defined(TUKLIB_PHYSMEM_PSTAT_GETSTATIC)
+ struct pst_static pst;
+ if (pstat_getstatic(&pst, sizeof(pst), 1, 0) != -1)
+ ret = (uint64_t)pst.physical_memory * (uint64_t)pst.page_size;
+
+#elif defined(TUKLIB_PHYSMEM_GETINVENT_R)
+ inv_state_t *st = NULL;
+ if (setinvent_r(&st) != -1) {
+ inventory_t *i;
+ while ((i = getinvent_r(st)) != NULL) {
+ if (i->inv_class == INV_MEMORY
+ && i->inv_type == INV_MAIN_MB) {
+ ret = (uint64_t)i->inv_state << 20;
+ break;
+ }
+ }
+
+ endinvent_r(st);
+ }
+
+#elif defined(TUKLIB_PHYSMEM_SYSINFO)
+ struct sysinfo si;
+ if (sysinfo(&si) == 0)
+ ret = (uint64_t)si.totalram * si.mem_unit;
+#endif
+
+ return ret;
+}
diff --git a/src/common/tuklib_physmem.h b/src/common/tuklib_physmem.h
new file mode 100644
index 0000000..09e2a51
--- /dev/null
+++ b/src/common/tuklib_physmem.h
@@ -0,0 +1,28 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_physmem.h
+/// \brief Get the amount of physical memory
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_PHYSMEM_H
+#define TUKLIB_PHYSMEM_H
+
+#include "tuklib_common.h"
+TUKLIB_DECLS_BEGIN
+
+#define tuklib_physmem TUKLIB_SYMBOL(tuklib_physmem)
+extern uint64_t tuklib_physmem(void);
+///<
+/// \brief Get the amount of physical memory in bytes
+///
+/// \return Amount of physical memory in bytes. On error, zero is
+/// returned.
+
+TUKLIB_DECLS_END
+#endif
diff --git a/src/common/tuklib_progname.c b/src/common/tuklib_progname.c
new file mode 100644
index 0000000..7cb7e20
--- /dev/null
+++ b/src/common/tuklib_progname.c
@@ -0,0 +1,50 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_progname.c
+/// \brief Program name to be displayed in messages
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tuklib_progname.h"
+#include <string.h>
+
+
+#if !HAVE_DECL_PROGRAM_INVOCATION_NAME
+char *progname = NULL;
+#endif
+
+
+extern void
+tuklib_progname_init(char **argv)
+{
+#ifdef TUKLIB_DOSLIKE
+ // On these systems, argv[0] always has the full path and .exe
+ // suffix even if the user just types the plain program name.
+ // We modify argv[0] to make it nicer to read.
+
+ // Strip the leading path.
+ char *p = argv[0] + strlen(argv[0]);
+ while (argv[0] < p && p[-1] != '/' && p[-1] != '\\')
+ --p;
+
+ argv[0] = p;
+
+ // Strip the .exe suffix.
+ p = strrchr(p, '.');
+ if (p != NULL)
+ *p = '\0';
+
+ // Make it lowercase.
+ for (p = argv[0]; *p != '\0'; ++p)
+ if (*p >= 'A' && *p <= 'Z')
+ *p = *p - 'A' + 'a';
+#endif
+
+ progname = argv[0];
+ return;
+}
diff --git a/src/common/tuklib_progname.h b/src/common/tuklib_progname.h
new file mode 100644
index 0000000..791b125
--- /dev/null
+++ b/src/common/tuklib_progname.h
@@ -0,0 +1,32 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tuklib_progname.h
+/// \brief Program name to be displayed in messages
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef TUKLIB_PROGNAME_H
+#define TUKLIB_PROGNAME_H
+
+#include "tuklib_common.h"
+#include <errno.h>
+
+TUKLIB_DECLS_BEGIN
+
+#if HAVE_DECL_PROGRAM_INVOCATION_NAME
+# define progname program_invocation_name
+#else
+# define progname TUKLIB_SYMBOL(tuklib_progname)
+ extern char *progname;
+#endif
+
+#define tuklib_progname_init TUKLIB_SYMBOL(tuklib_progname_init)
+extern void tuklib_progname_init(char **argv);
+
+TUKLIB_DECLS_END
+#endif
diff --git a/src/liblzma/Makefile.am b/src/liblzma/Makefile.am
new file mode 100644
index 0000000..f6c572b
--- /dev/null
+++ b/src/liblzma/Makefile.am
@@ -0,0 +1,90 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+SUBDIRS = api
+
+EXTRA_DIST =
+CLEANFILES =
+doc_DATA =
+
+lib_LTLIBRARIES = liblzma.la
+liblzma_la_SOURCES = $(top_srcdir)/src/common/tuklib_physmem.c
+liblzma_la_CPPFLAGS = \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_srcdir)/src/liblzma/common \
+ -I$(top_srcdir)/src/liblzma/check \
+ -I$(top_srcdir)/src/liblzma/lz \
+ -I$(top_srcdir)/src/liblzma/rangecoder \
+ -I$(top_srcdir)/src/liblzma/lzma \
+ -I$(top_srcdir)/src/liblzma/delta \
+ -I$(top_srcdir)/src/liblzma/simple \
+ -I$(top_srcdir)/src/common \
+ -DTUKLIB_SYMBOL_PREFIX=lzma_
+liblzma_la_LDFLAGS = -no-undefined -version-info 0:0:0
+
+include $(srcdir)/common/Makefile.inc
+include $(srcdir)/check/Makefile.inc
+
+if COND_FILTER_LZ
+include $(srcdir)/lz/Makefile.inc
+endif
+
+if COND_FILTER_LZMA1
+include $(srcdir)/lzma/Makefile.inc
+include $(srcdir)/rangecoder/Makefile.inc
+endif
+
+if COND_FILTER_DELTA
+include $(srcdir)/delta/Makefile.inc
+endif
+
+if COND_FILTER_SIMPLE
+include $(srcdir)/simple/Makefile.inc
+endif
+
+
+## Windows-specific stuff
+
+# Windows resource compiler support. libtool knows what to do with .rc
+# files, but Automake (<= 1.11 at least) doesn't know.
+#
+# We want the resource file only in shared liblzma. To avoid linking it into
+# static liblzma, we overwrite the static object file with an object file
+# compiled from empty input. Note that GNU-specific features are OK here,
+# because on Windows we are compiled with the GNU toolchain.
+.rc.lo:
+ $(LIBTOOL) --mode=compile $(RC) $(DEFS) $(DEFAULT_INCLUDES) \
+ $(INCLUDES) $(liblzma_la_CPPFLAGS) $(CPPFLAGS) $(RCFLAGS) \
+ -i $< -o $@
+ echo > empty.c
+ $(COMPILE) -c empty.c -o $(*D)/$(*F).o
+
+# Remove ordinals from the generated .def file. People must link by name,
+# not by ordinal, because no one is going to track the ordinal numbers.
+liblzma.def: liblzma.la liblzma.def.in
+ sed 's/ \+@ *[0-9]\+//' liblzma.def.in > liblzma.def
+
+# Creating liblzma.def.in is a side effect of linking the library.
+liblzma.def.in: liblzma.la
+
+if COND_W32
+CLEANFILES += liblzma.def liblzma.def.in empty.c
+liblzma_la_SOURCES += liblzma_w32res.rc
+liblzma_la_LDFLAGS += -Xlinker --output-def -Xlinker liblzma.def.in
+
+## liblzma.def.in is created only when building shared liblzma, so don't
+## try to create liblzma.def when not building shared liblzma.
+if COND_SHARED
+doc_DATA += liblzma.def
+endif
+endif
+
+
+## pkg-config
+pkgconfigdir = $(libdir)/pkgconfig
+pkgconfig_DATA = liblzma.pc
+EXTRA_DIST += liblzma.pc.in
diff --git a/src/liblzma/api/Makefile.am b/src/liblzma/api/Makefile.am
new file mode 100644
index 0000000..e016de5
--- /dev/null
+++ b/src/liblzma/api/Makefile.am
@@ -0,0 +1,23 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+nobase_include_HEADERS = \
+ lzma.h \
+ lzma/base.h \
+ lzma/bcj.h \
+ lzma/block.h \
+ lzma/check.h \
+ lzma/container.h \
+ lzma/delta.h \
+ lzma/filter.h \
+ lzma/hardware.h \
+ lzma/index.h \
+ lzma/index_hash.h \
+ lzma/lzma.h \
+ lzma/stream_flags.h \
+ lzma/version.h \
+ lzma/vli.h
diff --git a/src/liblzma/api/lzma.h b/src/liblzma/api/lzma.h
new file mode 100644
index 0000000..fb593a3
--- /dev/null
+++ b/src/liblzma/api/lzma.h
@@ -0,0 +1,313 @@
+/**
+ * \file api/lzma.h
+ * \brief The public API of liblzma data compression library
+ *
+ * liblzma is a public domain general-purpose data compression library with
+ * a zlib-like API. The native file format is .xz, but also the old .lzma
+ * format and raw (no headers) streams are supported. Multiple compression
+ * algorithms (filters) are supported. Currently LZMA2 is the primary filter.
+ *
+ * liblzma is part of XZ Utils <http://tukaani.org/xz/>. XZ Utils includes
+ * a gzip-like command line tool named xz and some other tools. XZ Utils
+ * is developed and maintained by Lasse Collin.
+ *
+ * Major parts of liblzma are based on Igor Pavlov's public domain LZMA SDK
+ * <http://7-zip.org/sdk.html>.
+ *
+ * The SHA-256 implementation is based on the public domain code found from
+ * 7-Zip <http://7-zip.org/>, which has a modified version of the public
+ * domain SHA-256 code found from Crypto++ <http://www.cryptopp.com/>.
+ * The SHA-256 code in Crypto++ was written by Kevin Springle and Wei Dai.
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#ifndef LZMA_H
+#define LZMA_H
+
+/*****************************
+ * Required standard headers *
+ *****************************/
+
+/*
+ * liblzma API headers need some standard types and macros. To allow
+ * including lzma.h without requiring the application to include other
+ * headers first, lzma.h includes the required standard headers unless
+ * they already seem to be included already or if LZMA_MANUAL_HEADERS
+ * has been defined.
+ *
+ * Here's what types and macros are needed and from which headers:
+ * - stddef.h: size_t, NULL
+ * - stdint.h: uint8_t, uint32_t, uint64_t, UINT32_C(n), uint64_C(n),
+ * UINT32_MAX, UINT64_MAX
+ *
+ * However, inttypes.h is a little more portable than stdint.h, although
+ * inttypes.h declares some unneeded things compared to plain stdint.h.
+ *
+ * The hacks below aren't perfect, specifically they assume that inttypes.h
+ * exists and that it typedefs at least uint8_t, uint32_t, and uint64_t,
+ * and that, in case of incomplete inttypes.h, unsigned int is 32-bit.
+ * If the application already takes care of setting up all the types and
+ * macros properly (for example by using gnulib's stdint.h or inttypes.h),
+ * we try to detect that the macros are already defined and don't include
+ * inttypes.h here again. However, you may define LZMA_MANUAL_HEADERS to
+ * force this file to never include any system headers.
+ *
+ * Some could argue that liblzma API should provide all the required types,
+ * for example lzma_uint64, LZMA_UINT64_C(n), and LZMA_UINT64_MAX. This was
+ * seen unnecessary mess, since most systems already provide all the necessary
+ * types and macros in the standard headers.
+ *
+ * Note that liblzma API still has lzma_bool, because using stdbool.h would
+ * break C89 and C++ programs on many systems. sizeof(bool) in C99 isn't
+ * necessarily the same as sizeof(bool) in C++.
+ */
+
+#ifndef LZMA_MANUAL_HEADERS
+ /*
+ * I suppose this works portably also in C++. Note that in C++,
+ * we need to get size_t into the global namespace.
+ */
+# include <stddef.h>
+
+ /*
+ * Skip inttypes.h if we already have all the required macros. If we
+ * have the macros, we assume that we have the matching typedefs too.
+ */
+# if !defined(UINT32_C) || !defined(UINT64_C) \
+ || !defined(UINT32_MAX) || !defined(UINT64_MAX)
+ /*
+ * MSVC has no C99 support, and thus it cannot be used to
+ * compile liblzma. The liblzma API has to still be usable
+ * from MSVC, so we need to define the required standard
+ * integer types here.
+ */
+# if defined(_WIN32) && defined(_MSC_VER)
+ typedef unsigned __int8 uint8_t;
+ typedef unsigned __int32 uint32_t;
+ typedef unsigned __int64 uint64_t;
+# else
+ /* Use the standard inttypes.h. */
+# ifdef __cplusplus
+ /*
+ * C99 sections 7.18.2 and 7.18.4 specify
+ * that C++ implementations define the limit
+ * and constant macros only if specifically
+ * requested. Note that if you want the
+ * format macros (PRIu64 etc.) too, you need
+ * to define __STDC_FORMAT_MACROS before
+ * including lzma.h, since re-including
+ * inttypes.h with __STDC_FORMAT_MACROS
+ * defined doesn't necessarily work.
+ */
+# ifndef __STDC_LIMIT_MACROS
+# define __STDC_LIMIT_MACROS 1
+# endif
+# ifndef __STDC_CONSTANT_MACROS
+# define __STDC_CONSTANT_MACROS 1
+# endif
+# endif
+
+# include <inttypes.h>
+# endif
+
+ /*
+ * Some old systems have only the typedefs in inttypes.h, and
+ * lack all the macros. For those systems, we need a few more
+ * hacks. We assume that unsigned int is 32-bit and unsigned
+ * long is either 32-bit or 64-bit. If these hacks aren't
+ * enough, the application has to setup the types manually
+ * before including lzma.h.
+ */
+# ifndef UINT32_C
+# if defined(_WIN32) && defined(_MSC_VER)
+# define UINT32_C(n) n ## UI32
+# else
+# define UINT32_C(n) n ## U
+# endif
+# endif
+
+# ifndef UINT64_C
+# if defined(_WIN32) && defined(_MSC_VER)
+# define UINT64_C(n) n ## UI64
+# else
+ /* Get ULONG_MAX. */
+# include <limits.h>
+# if ULONG_MAX == 4294967295UL
+# define UINT64_C(n) n ## ULL
+# else
+# define UINT64_C(n) n ## UL
+# endif
+# endif
+# endif
+
+# ifndef UINT32_MAX
+# define UINT32_MAX (UINT32_C(4294967295))
+# endif
+
+# ifndef UINT64_MAX
+# define UINT64_MAX (UINT64_C(18446744073709551615))
+# endif
+# endif
+#endif /* ifdef LZMA_MANUAL_HEADERS */
+
+
+/******************
+ * LZMA_API macro *
+ ******************/
+
+/*
+ * Some systems require that the functions and function pointers are
+ * declared specially in the headers. LZMA_API_IMPORT is for importing
+ * symbols and LZMA_API_CALL is to specify the calling convention.
+ *
+ * By default it is assumed that the application will link dynamically
+ * against liblzma. #define LZMA_API_STATIC in your application if you
+ * want to link against static liblzma. If you don't care about portability
+ * to operating systems like Windows, or at least don't care about linking
+ * against static liblzma on them, don't worry about LZMA_API_STATIC. That
+ * is, most developers will never need to use LZMA_API_STATIC.
+ *
+ * The GCC variants are a special case on Windows (Cygwin and MinGW).
+ * We rely on GCC doing the right thing with its auto-import feature,
+ * and thus don't use __declspec(dllimport). This way developers don't
+ * need to worry about LZMA_API_STATIC. Also the calling convention is
+ * omitted on Cygwin but not on MinGW.
+ */
+#ifndef LZMA_API_IMPORT
+# if !defined(LZMA_API_STATIC) && defined(_WIN32) && !defined(__GNUC__)
+# define LZMA_API_IMPORT __declspec(dllimport)
+# else
+# define LZMA_API_IMPORT
+# endif
+#endif
+
+#ifndef LZMA_API_CALL
+# if defined(_WIN32) && !defined(__CYGWIN__)
+# define LZMA_API_CALL __cdecl
+# else
+# define LZMA_API_CALL
+# endif
+#endif
+
+#ifndef LZMA_API
+# define LZMA_API(type) LZMA_API_IMPORT type LZMA_API_CALL
+#endif
+
+
+/***********
+ * nothrow *
+ ***********/
+
+/*
+ * None of the functions in liblzma may throw an exception. Even
+ * the functions that use callback functions won't throw exceptions,
+ * because liblzma would break if a callback function threw an exception.
+ */
+#ifndef lzma_nothrow
+# if defined(__cplusplus)
+# define lzma_nothrow throw()
+# elif __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 3)
+# define lzma_nothrow __attribute__((__nothrow__))
+# else
+# define lzma_nothrow
+# endif
+#endif
+
+
+/********************
+ * GNU C extensions *
+ ********************/
+
+/*
+ * GNU C extensions are used conditionally in the public API. It doesn't
+ * break anything if these are sometimes enabled and sometimes not, only
+ * affects warnings and optimizations.
+ */
+#if __GNUC__ >= 3
+# ifndef lzma_attribute
+# define lzma_attribute(attr) __attribute__(attr)
+# endif
+
+ /* warn_unused_result was added in GCC 3.4. */
+# ifndef lzma_attr_warn_unused_result
+# if __GNUC__ == 3 && __GNUC_MINOR__ < 4
+# define lzma_attr_warn_unused_result
+# endif
+# endif
+
+#else
+# ifndef lzma_attribute
+# define lzma_attribute(attr)
+# endif
+#endif
+
+
+#ifndef lzma_attr_pure
+# define lzma_attr_pure lzma_attribute((__pure__))
+#endif
+
+#ifndef lzma_attr_const
+# define lzma_attr_const lzma_attribute((__const__))
+#endif
+
+#ifndef lzma_attr_warn_unused_result
+# define lzma_attr_warn_unused_result \
+ lzma_attribute((__warn_unused_result__))
+#endif
+
+
+/**************
+ * Subheaders *
+ **************/
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/*
+ * Subheaders check that this is defined. It is to prevent including
+ * them directly from applications.
+ */
+#define LZMA_H_INTERNAL 1
+
+/* Basic features */
+#include "lzma/version.h"
+#include "lzma/base.h"
+#include "lzma/vli.h"
+#include "lzma/check.h"
+
+/* Filters */
+#include "lzma/filter.h"
+#include "lzma/bcj.h"
+#include "lzma/delta.h"
+#include "lzma/lzma.h"
+
+/* Container formats */
+#include "lzma/container.h"
+
+/* Advanced features */
+#include "lzma/stream_flags.h"
+#include "lzma/block.h"
+#include "lzma/index.h"
+#include "lzma/index_hash.h"
+
+/* Hardware information */
+#include "lzma/hardware.h"
+
+/*
+ * All subheaders included. Undefine LZMA_H_INTERNAL to prevent applications
+ * re-including the subheaders.
+ */
+#undef LZMA_H_INTERNAL
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* ifndef LZMA_H */
diff --git a/src/liblzma/api/lzma/base.h b/src/liblzma/api/lzma/base.h
new file mode 100644
index 0000000..993626a
--- /dev/null
+++ b/src/liblzma/api/lzma/base.h
@@ -0,0 +1,596 @@
+/**
+ * \file lzma/base.h
+ * \brief Data types and functions used in many places in liblzma API
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Boolean
+ *
+ * This is here because C89 doesn't have stdbool.h. To set a value for
+ * variables having type lzma_bool, you can use
+ * - C99's `true' and `false' from stdbool.h;
+ * - C++'s internal `true' and `false'; or
+ * - integers one (true) and zero (false).
+ */
+typedef unsigned char lzma_bool;
+
+
+/**
+ * \brief Type of reserved enumeration variable in structures
+ *
+ * To avoid breaking library ABI when new features are added, several
+ * structures contain extra variables that may be used in future. Since
+ * sizeof(enum) can be different than sizeof(int), and sizeof(enum) may
+ * even vary depending on the range of enumeration constants, we specify
+ * a separate type to be used for reserved enumeration variables. All
+ * enumeration constants in liblzma API will be non-negative and less
+ * than 128, which should guarantee that the ABI won't break even when
+ * new constants are added to existing enumerations.
+ */
+typedef enum {
+ LZMA_RESERVED_ENUM = 0
+} lzma_reserved_enum;
+
+
+/**
+ * \brief Return values used by several functions in liblzma
+ *
+ * Check the descriptions of specific functions to find out which return
+ * values they can return. With some functions the return values may have
+ * more specific meanings than described here; those differences are
+ * described per-function basis.
+ */
+typedef enum {
+ LZMA_OK = 0,
+ /**<
+ * \brief Operation completed successfully
+ */
+
+ LZMA_STREAM_END = 1,
+ /**<
+ * \brief End of stream was reached
+ *
+ * In encoder, LZMA_SYNC_FLUSH, LZMA_FULL_FLUSH, or
+ * LZMA_FINISH was finished. In decoder, this indicates
+ * that all the data was successfully decoded.
+ *
+ * In all cases, when LZMA_STREAM_END is returned, the last
+ * output bytes should be picked from strm->next_out.
+ */
+
+ LZMA_NO_CHECK = 2,
+ /**<
+ * \brief Input stream has no integrity check
+ *
+ * This return value can be returned only if the
+ * LZMA_TELL_NO_CHECK flag was used when initializing
+ * the decoder. LZMA_NO_CHECK is just a warning, and
+ * the decoding can be continued normally.
+ *
+ * It is possible to call lzma_get_check() immediately after
+ * lzma_code has returned LZMA_NO_CHECK. The result will
+ * naturally be LZMA_CHECK_NONE, but the possibility to call
+ * lzma_get_check() may be convenient in some applications.
+ */
+
+ LZMA_UNSUPPORTED_CHECK = 3,
+ /**<
+ * \brief Cannot calculate the integrity check
+ *
+ * The usage of this return value is different in encoders
+ * and decoders.
+ *
+ * Encoders can return this value only from the initialization
+ * function. If initialization fails with this value, the
+ * encoding cannot be done, because there's no way to produce
+ * output with the correct integrity check.
+ *
+ * Decoders can return this value only from lzma_code() and
+ * only if the LZMA_TELL_UNSUPPORTED_CHECK flag was used when
+ * initializing the decoder. The decoding can still be
+ * continued normally even if the check type is unsupported,
+ * but naturally the check will not be validated, and possible
+ * errors may go undetected.
+ *
+ * With decoder, it is possible to call lzma_get_check()
+ * immediately after lzma_code() has returned
+ * LZMA_UNSUPPORTED_CHECK. This way it is possible to find
+ * out what the unsupported Check ID was.
+ */
+
+ LZMA_GET_CHECK = 4,
+ /**<
+ * \brief Integrity check type is now available
+ *
+ * This value can be returned only by the lzma_code() function
+ * and only if the decoder was initialized with the
+ * LZMA_TELL_ANY_CHECK flag. LZMA_GET_CHECK tells the
+ * application that it may now call lzma_get_check() to find
+ * out the Check ID. This can be used, for example, to
+ * implement a decoder that accepts only files that have
+ * strong enough integrity check.
+ */
+
+ LZMA_MEM_ERROR = 5,
+ /**<
+ * \brief Cannot allocate memory
+ *
+ * Memory allocation failed, or the size of the allocation
+ * would be greater than SIZE_MAX.
+ *
+ * Due to internal implementation reasons, the coding cannot
+ * be continued even if more memory were made available after
+ * LZMA_MEM_ERROR.
+ */
+
+ LZMA_MEMLIMIT_ERROR = 6,
+ /**
+ * \brief Memory usage limit was reached
+ *
+ * Decoder would need more memory than allowed by the
+ * specified memory usage limit. To continue decoding,
+ * the memory usage limit has to be increased with
+ * lzma_memlimit_set().
+ */
+
+ LZMA_FORMAT_ERROR = 7,
+ /**<
+ * \brief File format not recognized
+ *
+ * The decoder did not recognize the input as supported file
+ * format. This error can occur, for example, when trying to
+ * decode .lzma format file with lzma_stream_decoder,
+ * because lzma_stream_decoder accepts only the .xz format.
+ */
+
+ LZMA_OPTIONS_ERROR = 8,
+ /**<
+ * \brief Invalid or unsupported options
+ *
+ * Invalid or unsupported options, for example
+ * - unsupported filter(s) or filter options; or
+ * - reserved bits set in headers (decoder only).
+ *
+ * Rebuilding liblzma with more features enabled, or
+ * upgrading to a newer version of liblzma may help.
+ */
+
+ LZMA_DATA_ERROR = 9,
+ /**<
+ * \brief Data is corrupt
+ *
+ * The usage of this return value is different in encoders
+ * and decoders. In both encoder and decoder, the coding
+ * cannot continue after this error.
+ *
+ * Encoders return this if size limits of the target file
+ * format would be exceeded. These limits are huge, thus
+ * getting this error from an encoder is mostly theoretical.
+ * For example, the maximum compressed and uncompressed
+ * size of a .xz Stream is roughly 8 EiB (2^63 bytes).
+ *
+ * Decoders return this error if the input data is corrupt.
+ * This can mean, for example, invalid CRC32 in headers
+ * or invalid check of uncompressed data.
+ */
+
+ LZMA_BUF_ERROR = 10,
+ /**<
+ * \brief No progress is possible
+ *
+ * This error code is returned when the coder cannot consume
+ * any new input and produce any new output. The most common
+ * reason for this error is that the input stream being
+ * decoded is truncated or corrupt.
+ *
+ * This error is not fatal. Coding can be continued normally
+ * by providing more input and/or more output space, if
+ * possible.
+ *
+ * Typically the first call to lzma_code() that can do no
+ * progress returns LZMA_OK instead of LZMA_BUF_ERROR. Only
+ * the second consecutive call doing no progress will return
+ * LZMA_BUF_ERROR. This is intentional.
+ *
+ * With zlib, Z_BUF_ERROR may be returned even if the
+ * application is doing nothing wrong, so apps will need
+ * to handle Z_BUF_ERROR specially. The above hack
+ * guarantees that liblzma never returns LZMA_BUF_ERROR
+ * to properly written applications unless the input file
+ * is truncated or corrupt. This should simplify the
+ * applications a little.
+ */
+
+ LZMA_PROG_ERROR = 11,
+ /**<
+ * \brief Programming error
+ *
+ * This indicates that the arguments given to the function are
+ * invalid or the internal state of the decoder is corrupt.
+ * - Function arguments are invalid or the structures
+ * pointed by the argument pointers are invalid
+ * e.g. if strm->next_out has been set to NULL and
+ * strm->avail_out > 0 when calling lzma_code().
+ * - lzma_* functions have been called in wrong order
+ * e.g. lzma_code() was called right after lzma_end().
+ * - If errors occur randomly, the reason might be flaky
+ * hardware.
+ *
+ * If you think that your code is correct, this error code
+ * can be a sign of a bug in liblzma. See the documentation
+ * how to report bugs.
+ */
+} lzma_ret;
+
+
+/**
+ * \brief The `action' argument for lzma_code()
+ *
+ * After the first use of LZMA_SYNC_FLUSH, LZMA_FULL_FLUSH, or LZMA_FINISH,
+ * the same `action' must is used until lzma_code() returns LZMA_STREAM_END.
+ * Also, the amount of input (that is, strm->avail_in) must not be modified
+ * by the application until lzma_code() returns LZMA_STREAM_END. Changing the
+ * `action' or modifying the amount of input will make lzma_code() return
+ * LZMA_PROG_ERROR.
+ */
+typedef enum {
+ LZMA_RUN = 0,
+ /**<
+ * \brief Continue coding
+ *
+ * Encoder: Encode as much input as possible. Some internal
+ * buffering will probably be done (depends on the filter
+ * chain in use), which causes latency: the input used won't
+ * usually be decodeable from the output of the same
+ * lzma_code() call.
+ *
+ * Decoder: Decode as much input as possible and produce as
+ * much output as possible.
+ */
+
+ LZMA_SYNC_FLUSH = 1,
+ /**<
+ * \brief Make all the input available at output
+ *
+ * Normally the encoder introduces some latency.
+ * LZMA_SYNC_FLUSH forces all the buffered data to be
+ * available at output without resetting the internal
+ * state of the encoder. This way it is possible to use
+ * compressed stream for example for communication over
+ * network.
+ *
+ * Only some filters support LZMA_SYNC_FLUSH. Trying to use
+ * LZMA_SYNC_FLUSH with filters that don't support it will
+ * make lzma_code() return LZMA_OPTIONS_ERROR. For example,
+ * LZMA1 doesn't support LZMA_SYNC_FLUSH but LZMA2 does.
+ *
+ * Using LZMA_SYNC_FLUSH very often can dramatically reduce
+ * the compression ratio. With some filters (for example,
+ * LZMA2), fine-tuning the compression options may help
+ * mitigate this problem significantly.
+ *
+ * Decoders don't support LZMA_SYNC_FLUSH.
+ */
+
+ LZMA_FULL_FLUSH = 2,
+ /**<
+ * \brief Make all the input available at output
+ *
+ * Finish encoding of the current Block. All the input
+ * data going to the current Block must have been given
+ * to the encoder (the last bytes can still be pending in
+ * next_in). Call lzma_code() with LZMA_FULL_FLUSH until
+ * it returns LZMA_STREAM_END. Then continue normally with
+ * LZMA_RUN or finish the Stream with LZMA_FINISH.
+ *
+ * This action is currently supported only by Stream encoder
+ * and easy encoder (which uses Stream encoder). If there is
+ * no unfinished Block, no empty Block is created.
+ */
+
+ LZMA_FINISH = 3
+ /**<
+ * \brief Finish the coding operation
+ *
+ * Finishes the coding operation. All the input data must
+ * have been given to the encoder (the last bytes can still
+ * be pending in next_in). Call lzma_code() with LZMA_FINISH
+ * until it returns LZMA_STREAM_END. Once LZMA_FINISH has
+ * been used, the amount of input must no longer be changed
+ * by the application.
+ *
+ * When decoding, using LZMA_FINISH is optional unless the
+ * LZMA_CONCATENATED flag was used when the decoder was
+ * initialized. When LZMA_CONCATENATED was not used, the only
+ * effect of LZMA_FINISH is that the amount of input must not
+ * be changed just like in the encoder.
+ */
+} lzma_action;
+
+
+/**
+ * \brief Custom functions for memory handling
+ *
+ * A pointer to lzma_allocator may be passed via lzma_stream structure
+ * to liblzma, and some advanced functions take a pointer to lzma_allocator
+ * as a separate function argument. The library will use the functions
+ * specified in lzma_allocator for memory handling instead of the default
+ * malloc() and free(). C++ users should note that the custom memory
+ * handling functions must not throw exceptions.
+ *
+ * liblzma doesn't make an internal copy of lzma_allocator. Thus, it is
+ * OK to change these function pointers in the middle of the coding
+ * process, but obviously it must be done carefully to make sure that the
+ * replacement `free' can deallocate memory allocated by the earlier
+ * `alloc' function(s).
+ */
+typedef struct {
+ /**
+ * \brief Pointer to a custom memory allocation function
+ *
+ * If you don't want a custom allocator, but still want
+ * custom free(), set this to NULL and liblzma will use
+ * the standard malloc().
+ *
+ * \param opaque lzma_allocator.opaque (see below)
+ * \param nmemb Number of elements like in calloc(). liblzma
+ * will always set nmemb to 1, so it is safe to
+ * ignore nmemb in a custom allocator if you like.
+ * The nmemb argument exists only for
+ * compatibility with zlib and libbzip2.
+ * \param size Size of an element in bytes.
+ * liblzma never sets this to zero.
+ *
+ * \return Pointer to the beginning of a memory block of
+ * `size' bytes, or NULL if allocation fails
+ * for some reason. When allocation fails, functions
+ * of liblzma return LZMA_MEM_ERROR.
+ *
+ * The allocator should not waste time zeroing the allocated buffers.
+ * This is not only about speed, but also memory usage, since the
+ * operating system kernel doesn't necessarily allocate the requested
+ * memory in physical memory until it is actually used. With small
+ * input files, liblzma may actually need only a fraction of the
+ * memory that it requested for allocation.
+ *
+ * \note LZMA_MEM_ERROR is also used when the size of the
+ * allocation would be greater than SIZE_MAX. Thus,
+ * don't assume that the custom allocator must have
+ * returned NULL if some function from liblzma
+ * returns LZMA_MEM_ERROR.
+ */
+ void *(LZMA_API_CALL *alloc)(void *opaque, size_t nmemb, size_t size);
+
+ /**
+ * \brief Pointer to a custom memory freeing function
+ *
+ * If you don't want a custom freeing function, but still
+ * want a custom allocator, set this to NULL and liblzma
+ * will use the standard free().
+ *
+ * \param opaque lzma_allocator.opaque (see below)
+ * \param ptr Pointer returned by lzma_allocator.alloc(),
+ * or when it is set to NULL, a pointer returned
+ * by the standard malloc().
+ */
+ void (LZMA_API_CALL *free)(void *opaque, void *ptr);
+
+ /**
+ * \brief Pointer passed to .alloc() and .free()
+ *
+ * opaque is passed as the first argument to lzma_allocator.alloc()
+ * and lzma_allocator.free(). This intended to ease implementing
+ * custom memory allocation functions for use with liblzma.
+ *
+ * If you don't need this, you should set this to NULL.
+ */
+ void *opaque;
+
+} lzma_allocator;
+
+
+/**
+ * \brief Internal data structure
+ *
+ * The contents of this structure is not visible outside the library.
+ */
+typedef struct lzma_internal_s lzma_internal;
+
+
+/**
+ * \brief Passing data to and from liblzma
+ *
+ * The lzma_stream structure is used for
+ * - passing pointers to input and output buffers to liblzma;
+ * - defining custom memory hander functions; and
+ * - holding a pointer to coder-specific internal data structures.
+ *
+ * Typical usage:
+ *
+ * - After allocating lzma_stream (on stack or with malloc()), it must be
+ * initialized to LZMA_STREAM_INIT (see LZMA_STREAM_INIT for details).
+ *
+ * - Initialize a coder to the lzma_stream, for example by using
+ * lzma_easy_encoder() or lzma_auto_decoder(). Some notes:
+ * - In contrast to zlib, strm->next_in and strm->next_out are
+ * ignored by all initialization functions, thus it is safe
+ * to not initialize them yet.
+ * - The initialization functions always set strm->total_in and
+ * strm->total_out to zero.
+ * - If the initialization function fails, no memory is left allocated
+ * that would require freeing with lzma_end() even if some memory was
+ * associated with the lzma_stream structure when the initialization
+ * function was called.
+ *
+ * - Use lzma_code() to do the actual work.
+ *
+ * - Once the coding has been finished, the existing lzma_stream can be
+ * reused. It is OK to reuse lzma_stream with different initialization
+ * function without calling lzma_end() first. Old allocations are
+ * automatically freed.
+ *
+ * - Finally, use lzma_end() to free the allocated memory. lzma_end() never
+ * frees the lzma_stream structure itself.
+ *
+ * Application may modify the values of total_in and total_out as it wants.
+ * They are updated by liblzma to match the amount of data read and
+ * written, but aren't used for anything else.
+ */
+typedef struct {
+ const uint8_t *next_in; /**< Pointer to the next input byte. */
+ size_t avail_in; /**< Number of available input bytes in next_in. */
+ uint64_t total_in; /**< Total number of bytes read by liblzma. */
+
+ uint8_t *next_out; /**< Pointer to the next output position. */
+ size_t avail_out; /**< Amount of free space in next_out. */
+ uint64_t total_out; /**< Total number of bytes written by liblzma. */
+
+ /**
+ * \brief Custom memory allocation functions
+ *
+ * In most cases this is NULL which makes liblzma use
+ * the standard malloc() and free().
+ */
+ lzma_allocator *allocator;
+
+ /** Internal state is not visible to applications. */
+ lzma_internal *internal;
+
+ /*
+ * Reserved space to allow possible future extensions without
+ * breaking the ABI. Excluding the initialization of this structure,
+ * you should not touch these, because the names of these variables
+ * may change.
+ */
+ void *reserved_ptr1;
+ void *reserved_ptr2;
+ uint64_t reserved_int1;
+ uint64_t reserved_int2;
+ lzma_reserved_enum reserved_enum1;
+ lzma_reserved_enum reserved_enum2;
+
+} lzma_stream;
+
+
+/**
+ * \brief Initialization for lzma_stream
+ *
+ * When you declare an instance of lzma_stream, you can immediately
+ * initialize it so that initialization functions know that no memory
+ * has been allocated yet:
+ *
+ * lzma_stream strm = LZMA_STREAM_INIT;
+ *
+ * If you need to initialize a dynamically allocated lzma_stream, you can use
+ * memset(strm_pointer, 0, sizeof(lzma_stream)). Strictly speaking, this
+ * violates the C standard since NULL may have different internal
+ * representation than zero, but it should be portable enough in practice.
+ * Anyway, for maximum portability, you can use something like this:
+ *
+ * lzma_stream tmp = LZMA_STREAM_INIT;
+ * *strm = tmp;
+ */
+#define LZMA_STREAM_INIT \
+ { NULL, 0, 0, NULL, 0, 0, NULL, NULL, \
+ NULL, NULL, 0, 0, LZMA_RESERVED_ENUM, LZMA_RESERVED_ENUM }
+
+
+/**
+ * \brief Encode or decode data
+ *
+ * Once the lzma_stream has been successfully initialized (e.g. with
+ * lzma_stream_encoder()), the actual encoding or decoding is done
+ * using this function. The application has to update strm->next_in,
+ * strm->avail_in, strm->next_out, and strm->avail_out to pass input
+ * to and get output from liblzma.
+ *
+ * See the description of the coder-specific initialization function to find
+ * out what `action' values are supported by the coder.
+ */
+extern LZMA_API(lzma_ret) lzma_code(lzma_stream *strm, lzma_action action)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Free memory allocated for the coder data structures
+ *
+ * \param strm Pointer to lzma_stream that is at least initialized
+ * with LZMA_STREAM_INIT.
+ *
+ * After lzma_end(strm), strm->internal is guaranteed to be NULL. No other
+ * members of the lzma_stream structure are touched.
+ *
+ * \note zlib indicates an error if application end()s unfinished
+ * stream structure. liblzma doesn't do this, and assumes that
+ * application knows what it is doing.
+ */
+extern LZMA_API(void) lzma_end(lzma_stream *strm) lzma_nothrow;
+
+
+/**
+ * \brief Get the memory usage of decoder filter chain
+ *
+ * This function is currently supported only when *strm has been initialized
+ * with a function that takes a memlimit argument. With other functions, you
+ * should use e.g. lzma_raw_encoder_memusage() or lzma_raw_decoder_memusage()
+ * to estimate the memory requirements.
+ *
+ * This function is useful e.g. after LZMA_MEMLIMIT_ERROR to find out how big
+ * the memory usage limit should have been to decode the input. Note that
+ * this may give misleading information if decoding .xz Streams that have
+ * multiple Blocks, because each Block can have different memory requirements.
+ *
+ * \return Rough estimate of how much memory is currently allocated
+ * for the filter decoders. If no filter chain is currently
+ * allocated, some non-zero value is still returned, which is
+ * less than or equal to what any filter chain would indicate
+ * as its memory requirement.
+ *
+ * If this function isn't supported by *strm or some other error
+ * occurs, zero is returned.
+ */
+extern LZMA_API(uint64_t) lzma_memusage(const lzma_stream *strm)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the current memory usage limit
+ *
+ * This function is supported only when *strm has been initialized with
+ * a function that takes a memlimit argument.
+ *
+ * \return On success, the current memory usage limit is returned
+ * (always non-zero). On error, zero is returned.
+ */
+extern LZMA_API(uint64_t) lzma_memlimit_get(const lzma_stream *strm)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Set the memory usage limit
+ *
+ * This function is supported only when *strm has been initialized with
+ * a function that takes a memlimit argument.
+ *
+ * \return - LZMA_OK: New memory usage limit successfully set.
+ * - LZMA_MEMLIMIT_ERROR: The new limit is too small.
+ * The limit was not changed.
+ * - LZMA_PROG_ERROR: Invalid arguments, e.g. *strm doesn't
+ * support memory usage limit or memlimit was zero.
+ */
+extern LZMA_API(lzma_ret) lzma_memlimit_set(
+ lzma_stream *strm, uint64_t memlimit) lzma_nothrow;
diff --git a/src/liblzma/api/lzma/bcj.h b/src/liblzma/api/lzma/bcj.h
new file mode 100644
index 0000000..82e4a44
--- /dev/null
+++ b/src/liblzma/api/lzma/bcj.h
@@ -0,0 +1,90 @@
+/**
+ * \file lzma/bcj.h
+ * \brief Branch/Call/Jump conversion filters
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/* Filter IDs for lzma_filter.id */
+
+#define LZMA_FILTER_X86 LZMA_VLI_C(0x04)
+ /**<
+ * Filter for x86 binaries
+ */
+
+#define LZMA_FILTER_POWERPC LZMA_VLI_C(0x05)
+ /**<
+ * Filter for Big endian PowerPC binaries
+ */
+
+#define LZMA_FILTER_IA64 LZMA_VLI_C(0x06)
+ /**<
+ * Filter for IA64 (Itanium) binaries.
+ */
+
+#define LZMA_FILTER_ARM LZMA_VLI_C(0x07)
+ /**<
+ * Filter for ARM binaries.
+ */
+
+#define LZMA_FILTER_ARMTHUMB LZMA_VLI_C(0x08)
+ /**<
+ * Filter for ARMThumb binaries.
+ */
+
+#define LZMA_FILTER_SPARC LZMA_VLI_C(0x09)
+ /**<
+ * Filter for SPARC binaries.
+ */
+
+
+/**
+ * \brief Options for BCJ filters
+ *
+ * The BCJ filters never change the size of the data. Specifying options
+ * for them is optional: if pointer to options is NULL, default value is
+ * used. You probably never need to specify options to BCJ filters, so just
+ * set the options pointer to NULL and be happy.
+ *
+ * If options with non-default values have been specified when encoding,
+ * the same options must also be specified when decoding.
+ *
+ * \note At the moment, none of the BCJ filters support
+ * LZMA_SYNC_FLUSH. If LZMA_SYNC_FLUSH is specified,
+ * LZMA_OPTIONS_ERROR will be returned. If there is need,
+ * partial support for LZMA_SYNC_FLUSH can be added in future.
+ * Partial means that flushing would be possible only at
+ * offsets that are multiple of 2, 4, or 16 depending on
+ * the filter, except x86 which cannot be made to support
+ * LZMA_SYNC_FLUSH predictably.
+ */
+typedef struct {
+ /**
+ * \brief Start offset for conversions
+ *
+ * This setting is useful only when the same filter is used
+ * _separately_ for multiple sections of the same executable file,
+ * and the sections contain cross-section branch/call/jump
+ * instructions. In that case it is beneficial to set the start
+ * offset of the non-first sections so that the relative addresses
+ * of the cross-section branch/call/jump instructions will use the
+ * same absolute addresses as in the first section.
+ *
+ * When the pointer to options is NULL, the default value (zero)
+ * is used.
+ */
+ uint32_t start_offset;
+
+} lzma_options_bcj;
diff --git a/src/liblzma/api/lzma/block.h b/src/liblzma/api/lzma/block.h
new file mode 100644
index 0000000..8e681ed
--- /dev/null
+++ b/src/liblzma/api/lzma/block.h
@@ -0,0 +1,529 @@
+/**
+ * \file lzma/block.h
+ * \brief .xz Block handling
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Options for the Block and Block Header encoders and decoders
+ *
+ * Different Block handling functions use different parts of this structure.
+ * Some read some members, other functions write, and some do both. Only the
+ * members listed for reading need to be initialized when the specified
+ * functions are called. The members marked for writing will be assigned
+ * new values at some point either by calling the given function or by
+ * later calls to lzma_code().
+ */
+typedef struct {
+ /**
+ * \brief Block format version
+ *
+ * To prevent API and ABI breakages if new features are needed in
+ * the Block field, a version number is used to indicate which
+ * fields in this structure are in use. For now, version must always
+ * be zero. With non-zero version, most Block related functions will
+ * return LZMA_OPTIONS_ERROR.
+ *
+ * Read by:
+ * - All functions that take pointer to lzma_block as argument,
+ * including lzma_block_header_decode().
+ *
+ * Written by:
+ * - lzma_block_header_decode()
+ */
+ uint32_t version;
+
+ /**
+ * \brief Size of the Block Header field
+ *
+ * This is always a multiple of four.
+ *
+ * Read by:
+ * - lzma_block_header_encode()
+ * - lzma_block_header_decode()
+ * - lzma_block_compressed_size()
+ * - lzma_block_unpadded_size()
+ * - lzma_block_total_size()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_decode()
+ *
+ * Written by:
+ * - lzma_block_header_size()
+ * - lzma_block_buffer_encode()
+ */
+ uint32_t header_size;
+# define LZMA_BLOCK_HEADER_SIZE_MIN 8
+# define LZMA_BLOCK_HEADER_SIZE_MAX 1024
+
+ /**
+ * \brief Type of integrity Check
+ *
+ * The Check ID is not stored into the Block Header, thus its value
+ * must be provided also when decoding.
+ *
+ * Read by:
+ * - lzma_block_header_encode()
+ * - lzma_block_header_decode()
+ * - lzma_block_compressed_size()
+ * - lzma_block_unpadded_size()
+ * - lzma_block_total_size()
+ * - lzma_block_encoder()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_encode()
+ * - lzma_block_buffer_decode()
+ */
+ lzma_check check;
+
+ /**
+ * \brief Size of the Compressed Data in bytes
+ *
+ * Encoding: If this is not LZMA_VLI_UNKNOWN, Block Header encoder
+ * will store this value to the Block Header. Block encoder doesn't
+ * care about this value, but will set it once the encoding has been
+ * finished.
+ *
+ * Decoding: If this is not LZMA_VLI_UNKNOWN, Block decoder will
+ * verify that the size of the Compressed Data field matches
+ * compressed_size.
+ *
+ * Usually you don't know this value when encoding in streamed mode,
+ * and thus cannot write this field into the Block Header.
+ *
+ * In non-streamed mode you can reserve space for this field before
+ * encoding the actual Block. After encoding the data, finish the
+ * Block by encoding the Block Header. Steps in detail:
+ *
+ * - Set compressed_size to some big enough value. If you don't know
+ * better, use LZMA_VLI_MAX, but remember that bigger values take
+ * more space in Block Header.
+ *
+ * - Call lzma_block_header_size() to see how much space you need to
+ * reserve for the Block Header.
+ *
+ * - Encode the Block using lzma_block_encoder() and lzma_code().
+ * It sets compressed_size to the correct value.
+ *
+ * - Use lzma_block_header_encode() to encode the Block Header.
+ * Because space was reserved in the first step, you don't need
+ * to call lzma_block_header_size() anymore, because due to
+ * reserving, header_size has to be big enough. If it is "too big",
+ * lzma_block_header_encode() will add enough Header Padding to
+ * make Block Header to match the size specified by header_size.
+ *
+ * Read by:
+ * - lzma_block_header_size()
+ * - lzma_block_header_encode()
+ * - lzma_block_compressed_size()
+ * - lzma_block_unpadded_size()
+ * - lzma_block_total_size()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_decode()
+ *
+ * Written by:
+ * - lzma_block_header_decode()
+ * - lzma_block_compressed_size()
+ * - lzma_block_encoder()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_encode()
+ * - lzma_block_buffer_decode()
+ */
+ lzma_vli compressed_size;
+
+ /**
+ * \brief Uncompressed Size in bytes
+ *
+ * This is handled very similarly to compressed_size above.
+ *
+ * uncompressed_size is needed by fewer functions than
+ * compressed_size. This is because uncompressed_size isn't
+ * needed to validate that Block stays within proper limits.
+ *
+ * Read by:
+ * - lzma_block_header_size()
+ * - lzma_block_header_encode()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_decode()
+ *
+ * Written by:
+ * - lzma_block_header_decode()
+ * - lzma_block_encoder()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_encode()
+ * - lzma_block_buffer_decode()
+ */
+ lzma_vli uncompressed_size;
+
+ /**
+ * \brief Array of filters
+ *
+ * There can be 1-4 filters. The end of the array is marked with
+ * .id = LZMA_VLI_UNKNOWN.
+ *
+ * Read by:
+ * - lzma_block_header_size()
+ * - lzma_block_header_encode()
+ * - lzma_block_encoder()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_encode()
+ * - lzma_block_buffer_decode()
+ *
+ * Written by:
+ * - lzma_block_header_decode(): Note that this does NOT free()
+ * the old filter options structures. All unused filters[] will
+ * have .id == LZMA_VLI_UNKNOWN and .options == NULL. If
+ * decoding fails, all filters[] are guaranteed to be
+ * LZMA_VLI_UNKNOWN and NULL.
+ *
+ * \note Because of the array is terminated with
+ * .id = LZMA_VLI_UNKNOWN, the actual array must
+ * have LZMA_FILTERS_MAX + 1 members or the Block
+ * Header decoder will overflow the buffer.
+ */
+ lzma_filter *filters;
+
+ /**
+ * \brief Raw value stored in the Check field
+ *
+ * After successful coding, the first lzma_check_size(check) bytes
+ * of this array contain the raw value stored in the Check field.
+ *
+ * Note that CRC32 and CRC64 are stored in little endian byte order.
+ * Take it into account if you display the Check values to the user.
+ *
+ * Written by:
+ * - lzma_block_encoder()
+ * - lzma_block_decoder()
+ * - lzma_block_buffer_encode()
+ * - lzma_block_buffer_decode()
+ */
+ uint8_t raw_check[LZMA_CHECK_SIZE_MAX];
+
+ /*
+ * Reserved space to allow possible future extensions without
+ * breaking the ABI. You should not touch these, because the names
+ * of these variables may change. These are and will never be used
+ * with the currently supported options, so it is safe to leave these
+ * uninitialized.
+ */
+ void *reserved_ptr1;
+ void *reserved_ptr2;
+ void *reserved_ptr3;
+ uint32_t reserved_int1;
+ uint32_t reserved_int2;
+ lzma_vli reserved_int3;
+ lzma_vli reserved_int4;
+ lzma_vli reserved_int5;
+ lzma_vli reserved_int6;
+ lzma_vli reserved_int7;
+ lzma_vli reserved_int8;
+ lzma_reserved_enum reserved_enum1;
+ lzma_reserved_enum reserved_enum2;
+ lzma_reserved_enum reserved_enum3;
+ lzma_reserved_enum reserved_enum4;
+ lzma_bool reserved_bool1;
+ lzma_bool reserved_bool2;
+ lzma_bool reserved_bool3;
+ lzma_bool reserved_bool4;
+ lzma_bool reserved_bool5;
+ lzma_bool reserved_bool6;
+ lzma_bool reserved_bool7;
+ lzma_bool reserved_bool8;
+
+} lzma_block;
+
+
+/**
+ * \brief Decode the Block Header Size field
+ *
+ * To decode Block Header using lzma_block_header_decode(), the size of the
+ * Block Header has to be known and stored into lzma_block.header_size.
+ * The size can be calculated from the first byte of a Block using this macro.
+ * Note that if the first byte is 0x00, it indicates beginning of Index; use
+ * this macro only when the byte is not 0x00.
+ *
+ * There is no encoding macro, because Block Header encoder is enough for that.
+ */
+#define lzma_block_header_size_decode(b) (((uint32_t)(b) + 1) * 4)
+
+
+/**
+ * \brief Calculate Block Header Size
+ *
+ * Calculate the minimum size needed for the Block Header field using the
+ * settings specified in the lzma_block structure. Note that it is OK to
+ * increase the calculated header_size value as long as it is a multiple of
+ * four and doesn't exceed LZMA_BLOCK_HEADER_SIZE_MAX. Increasing header_size
+ * just means that lzma_block_header_encode() will add Header Padding.
+ *
+ * \return - LZMA_OK: Size calculated successfully and stored to
+ * block->header_size.
+ * - LZMA_OPTIONS_ERROR: Unsupported version, filters or
+ * filter options.
+ * - LZMA_PROG_ERROR: Invalid values like compressed_size == 0.
+ *
+ * \note This doesn't check that all the options are valid i.e. this
+ * may return LZMA_OK even if lzma_block_header_encode() or
+ * lzma_block_encoder() would fail. If you want to validate the
+ * filter chain, consider using lzma_memlimit_encoder() which as
+ * a side-effect validates the filter chain.
+ */
+extern LZMA_API(lzma_ret) lzma_block_header_size(lzma_block *block)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Encode Block Header
+ *
+ * The caller must have calculated the size of the Block Header already with
+ * lzma_block_header_size(). If a value larger than the one calculated by
+ * lzma_block_header_size() is used, the Block Header will be padded to the
+ * specified size.
+ *
+ * \param out Beginning of the output buffer. This must be
+ * at least block->header_size bytes.
+ * \param block Block options to be encoded.
+ *
+ * \return - LZMA_OK: Encoding was successful. block->header_size
+ * bytes were written to output buffer.
+ * - LZMA_OPTIONS_ERROR: Invalid or unsupported options.
+ * - LZMA_PROG_ERROR: Invalid arguments, for example
+ * block->header_size is invalid or block->filters is NULL.
+ */
+extern LZMA_API(lzma_ret) lzma_block_header_encode(
+ const lzma_block *block, uint8_t *out)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Decode Block Header
+ *
+ * block->version should be set to the highest value supported by the
+ * application; currently the only possible version is zero. This function
+ * will set version to the lowest value that still supports all the features
+ * required by the Block Header.
+ *
+ * The size of the Block Header must have already been decoded with
+ * lzma_block_header_size_decode() macro and stored to block->header_size.
+ *
+ * block->filters must have been allocated, but not necessarily initialized.
+ * Possible existing filter options are _not_ freed.
+ *
+ * \param block Destination for Block options.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() (and also free()
+ * if an error occurs).
+ * \param in Beginning of the input buffer. This must be
+ * at least block->header_size bytes.
+ *
+ * \return - LZMA_OK: Decoding was successful. block->header_size
+ * bytes were read from the input buffer.
+ * - LZMA_OPTIONS_ERROR: The Block Header specifies some
+ * unsupported options such as unsupported filters. This can
+ * happen also if block->version was set to a too low value
+ * compared to what would be required to properly represent
+ * the information stored in the Block Header.
+ * - LZMA_DATA_ERROR: Block Header is corrupt, for example,
+ * the CRC32 doesn't match.
+ * - LZMA_PROG_ERROR: Invalid arguments, for example
+ * block->header_size is invalid or block->filters is NULL.
+ */
+extern LZMA_API(lzma_ret) lzma_block_header_decode(lzma_block *block,
+ lzma_allocator *allocator, const uint8_t *in)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Validate and set Compressed Size according to Unpadded Size
+ *
+ * Block Header stores Compressed Size, but Index has Unpadded Size. If the
+ * application has already parsed the Index and is now decoding Blocks,
+ * it can calculate Compressed Size from Unpadded Size. This function does
+ * exactly that with error checking:
+ *
+ * - Compressed Size calculated from Unpadded Size must be positive integer,
+ * that is, Unpadded Size must be big enough that after Block Header and
+ * Check fields there's still at least one byte for Compressed Size.
+ *
+ * - If Compressed Size was present in Block Header, the new value
+ * calculated from Unpadded Size is compared against the value
+ * from Block Header.
+ *
+ * \note This function must be called _after_ decoding the Block Header
+ * field so that it can properly validate Compressed Size if it
+ * was present in Block Header.
+ *
+ * \return - LZMA_OK: block->compressed_size was set successfully.
+ * - LZMA_DATA_ERROR: unpadded_size is too small compared to
+ * block->header_size and lzma_check_size(block->check).
+ * - LZMA_PROG_ERROR: Some values are invalid. For example,
+ * block->header_size must be a multiple of four and
+ * between 8 and 1024 inclusive.
+ */
+extern LZMA_API(lzma_ret) lzma_block_compressed_size(
+ lzma_block *block, lzma_vli unpadded_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Calculate Unpadded Size
+ *
+ * The Index field stores Unpadded Size and Uncompressed Size. The latter
+ * can be taken directly from the lzma_block structure after coding a Block,
+ * but Unpadded Size needs to be calculated from Block Header Size,
+ * Compressed Size, and size of the Check field. This is where this function
+ * is needed.
+ *
+ * \return Unpadded Size on success, or zero on error.
+ */
+extern LZMA_API(lzma_vli) lzma_block_unpadded_size(const lzma_block *block)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Calculate the total encoded size of a Block
+ *
+ * This is equivalent to lzma_block_unpadded_size() except that the returned
+ * value includes the size of the Block Padding field.
+ *
+ * \return On success, total encoded size of the Block. On error,
+ * zero is returned.
+ */
+extern LZMA_API(lzma_vli) lzma_block_total_size(const lzma_block *block)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Initialize .xz Block encoder
+ *
+ * Valid actions for lzma_code() are LZMA_RUN, LZMA_SYNC_FLUSH (only if the
+ * filter chain supports it), and LZMA_FINISH.
+ *
+ * \return - LZMA_OK: All good, continue with lzma_code().
+ * - LZMA_MEM_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_UNSUPPORTED_CHECK: block->check specifies a Check ID
+ * that is not supported by this buid of liblzma. Initializing
+ * the encoder failed.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_block_encoder(
+ lzma_stream *strm, lzma_block *block)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Initialize .xz Block decoder
+ *
+ * Valid actions for lzma_code() are LZMA_RUN and LZMA_FINISH. Using
+ * LZMA_FINISH is not required. It is supported only for convenience.
+ *
+ * \return - LZMA_OK: All good, continue with lzma_code().
+ * - LZMA_UNSUPPORTED_CHECK: Initialization was successful, but
+ * the given Check ID is not supported, thus Check will be
+ * ignored.
+ * - LZMA_PROG_ERROR
+ * - LZMA_MEM_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_block_decoder(
+ lzma_stream *strm, lzma_block *block)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Calculate maximum output size for single-call Block encoding
+ *
+ * This is equivalent to lzma_stream_buffer_bound() but for .xz Blocks.
+ * See the documentation of lzma_stream_buffer_bound().
+ */
+extern LZMA_API(size_t) lzma_block_buffer_bound(size_t uncompressed_size)
+ lzma_nothrow;
+
+
+/**
+ * \brief Single-call .xz Block encoder
+ *
+ * In contrast to the multi-call encoder initialized with
+ * lzma_block_encoder(), this function encodes also the Block Header. This
+ * is required to make it possible to write appropriate Block Header also
+ * in case the data isn't compressible, and different filter chain has to be
+ * used to encode the data in uncompressed form using uncompressed chunks
+ * of the LZMA2 filter.
+ *
+ * When the data isn't compressible, header_size, compressed_size, and
+ * uncompressed_size are set just like when the data was compressible, but
+ * it is possible that header_size is too small to hold the filter chain
+ * specified in block->filters, because that isn't necessarily the filter
+ * chain that was actually used to encode the data. lzma_block_unpadded_size()
+ * still works normally, because it doesn't read the filters array.
+ *
+ * \param block Block options: block->version, block->check,
+ * and block->filters must have been initialized.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_size Size of the input buffer
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_BUF_ERROR: Not enough output buffer space.
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_MEM_ERROR
+ * - LZMA_DATA_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_block_buffer_encode(
+ lzma_block *block, lzma_allocator *allocator,
+ const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Single-call .xz Block decoder
+ *
+ * This is single-call equivalent of lzma_block_decoder(), and requires that
+ * the caller has already decoded Block Header and checked its memory usage.
+ *
+ * \param block Block options just like with lzma_block_decoder().
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_pos The next byte will be read from in[*in_pos].
+ * *in_pos is updated only if decoding succeeds.
+ * \param in_size Size of the input buffer; the first byte that
+ * won't be read is in[in_size].
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Decoding was successful.
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_DATA_ERROR
+ * - LZMA_MEM_ERROR
+ * - LZMA_BUF_ERROR: Output buffer was too small.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_block_buffer_decode(
+ lzma_block *block, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+ lzma_nothrow;
diff --git a/src/liblzma/api/lzma/check.h b/src/liblzma/api/lzma/check.h
new file mode 100644
index 0000000..6a243db
--- /dev/null
+++ b/src/liblzma/api/lzma/check.h
@@ -0,0 +1,150 @@
+/**
+ * \file lzma/check.h
+ * \brief Integrity checks
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Type of the integrity check (Check ID)
+ *
+ * The .xz format supports multiple types of checks that are calculated
+ * from the uncompressed data. They vary in both speed and ability to
+ * detect errors.
+ */
+typedef enum {
+ LZMA_CHECK_NONE = 0,
+ /**<
+ * No Check is calculated.
+ *
+ * Size of the Check field: 0 bytes
+ */
+
+ LZMA_CHECK_CRC32 = 1,
+ /**<
+ * CRC32 using the polynomial from the IEEE 802.3 standard
+ *
+ * Size of the Check field: 4 bytes
+ */
+
+ LZMA_CHECK_CRC64 = 4,
+ /**<
+ * CRC64 using the polynomial from the ECMA-182 standard
+ *
+ * Size of the Check field: 8 bytes
+ */
+
+ LZMA_CHECK_SHA256 = 10
+ /**<
+ * SHA-256
+ *
+ * Size of the Check field: 32 bytes
+ */
+} lzma_check;
+
+
+/**
+ * \brief Maximum valid Check ID
+ *
+ * The .xz file format specification specifies 16 Check IDs (0-15). Some
+ * of them are only reserved, that is, no actual Check algorithm has been
+ * assigned. When decoding, liblzma still accepts unknown Check IDs for
+ * future compatibility. If a valid but unsupported Check ID is detected,
+ * liblzma can indicate a warning; see the flags LZMA_TELL_NO_CHECK,
+ * LZMA_TELL_UNSUPPORTED_CHECK, and LZMA_TELL_ANY_CHECK in container.h.
+ */
+#define LZMA_CHECK_ID_MAX 15
+
+
+/**
+ * \brief Test if the given Check ID is supported
+ *
+ * Return true if the given Check ID is supported by this liblzma build.
+ * Otherwise false is returned. It is safe to call this with a value that
+ * is not in the range [0, 15]; in that case the return value is always false.
+ *
+ * You can assume that LZMA_CHECK_NONE and LZMA_CHECK_CRC32 are always
+ * supported (even if liblzma is built with limited features).
+ */
+extern LZMA_API(lzma_bool) lzma_check_is_supported(lzma_check check)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Get the size of the Check field with the given Check ID
+ *
+ * Although not all Check IDs have a check algorithm associated, the size of
+ * every Check is already frozen. This function returns the size (in bytes) of
+ * the Check field with the specified Check ID. The values are:
+ * { 0, 4, 4, 4, 8, 8, 8, 16, 16, 16, 32, 32, 32, 64, 64, 64 }
+ *
+ * If the argument is not in the range [0, 15], UINT32_MAX is returned.
+ */
+extern LZMA_API(uint32_t) lzma_check_size(lzma_check check)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Maximum size of a Check field
+ */
+#define LZMA_CHECK_SIZE_MAX 64
+
+
+/**
+ * \brief Calculate CRC32
+ *
+ * Calculate CRC32 using the polynomial from the IEEE 802.3 standard.
+ *
+ * \param buf Pointer to the input buffer
+ * \param size Size of the input buffer
+ * \param crc Previously returned CRC value. This is used to
+ * calculate the CRC of a big buffer in smaller chunks.
+ * Set to zero when starting a new calculation.
+ *
+ * \return Updated CRC value, which can be passed to this function
+ * again to continue CRC calculation.
+ */
+extern LZMA_API(uint32_t) lzma_crc32(
+ const uint8_t *buf, size_t size, uint32_t crc)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Calculate CRC64
+ *
+ * Calculate CRC64 using the polynomial from the ECMA-182 standard.
+ *
+ * This function is used similarly to lzma_crc32(). See its documentation.
+ */
+extern LZMA_API(uint64_t) lzma_crc64(
+ const uint8_t *buf, size_t size, uint64_t crc)
+ lzma_nothrow lzma_attr_pure;
+
+
+/*
+ * SHA-256 functions are currently not exported to public API.
+ * Contact Lasse Collin if you think it should be.
+ */
+
+
+/**
+ * \brief Get the type of the integrity check
+ *
+ * This function can be called only immediately after lzma_code() has
+ * returned LZMA_NO_CHECK, LZMA_UNSUPPORTED_CHECK, or LZMA_GET_CHECK.
+ * Calling this function in any other situation has undefined behavior.
+ */
+extern LZMA_API(lzma_check) lzma_get_check(const lzma_stream *strm)
+ lzma_nothrow;
diff --git a/src/liblzma/api/lzma/container.h b/src/liblzma/api/lzma/container.h
new file mode 100644
index 0000000..0d90765
--- /dev/null
+++ b/src/liblzma/api/lzma/container.h
@@ -0,0 +1,404 @@
+/**
+ * \file lzma/container.h
+ * \brief File formats
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/************
+ * Encoding *
+ ************/
+
+/**
+ * \brief Default compression preset
+ *
+ * It's not straightforward to recommend a default preset, because in some
+ * cases keeping the resource usage relatively low is more important that
+ * getting the maximum compression ratio.
+ */
+#define LZMA_PRESET_DEFAULT UINT32_C(6)
+
+
+/**
+ * \brief Mask for preset level
+ *
+ * This is useful only if you need to extract the level from the preset
+ * variable. That should be rare.
+ */
+#define LZMA_PRESET_LEVEL_MASK UINT32_C(0x1F)
+
+
+/*
+ * Preset flags
+ *
+ * Currently only one flag is defined.
+ */
+
+/**
+ * \brief Extreme compression preset
+ *
+ * This flag modifies the preset to make the encoding significantly slower
+ * while improving the compression ratio only marginally. This is useful
+ * when you don't mind wasting time to get as small result as possible.
+ *
+ * This flag doesn't affect the memory usage requirements of the decoder (at
+ * least not significantly). The memory usage of the encoder may be increased
+ * a little but only at the lowest preset levels (0-2).
+ */
+#define LZMA_PRESET_EXTREME (UINT32_C(1) << 31)
+
+
+/**
+ * \brief Calculate rough memory usage of easy encoder
+ *
+ * This function is a wrapper for lzma_raw_encoder_memusage().
+ *
+ * \param preset Compression preset (level and possible flags)
+ */
+extern LZMA_API(uint64_t) lzma_easy_encoder_memusage(uint32_t preset)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Calculate rough decoder memory usage of a preset
+ *
+ * This function is a wrapper for lzma_raw_decoder_memusage().
+ *
+ * \param preset Compression preset (level and possible flags)
+ */
+extern LZMA_API(uint64_t) lzma_easy_decoder_memusage(uint32_t preset)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Initialize .xz Stream encoder using a preset number
+ *
+ * This function is intended for those who just want to use the basic features
+ * if liblzma (that is, most developers out there).
+ *
+ * \param strm Pointer to lzma_stream that is at least initialized
+ * with LZMA_STREAM_INIT.
+ * \param preset Compression preset to use. A preset consist of level
+ * number and zero or more flags. Usually flags aren't
+ * used, so preset is simply a number [0, 9] which match
+ * the options -0 .. -9 of the xz command line tool.
+ * Additional flags can be be set using bitwise-or with
+ * the preset level number, e.g. 6 | LZMA_PRESET_EXTREME.
+ * \param check Integrity check type to use. See check.h for available
+ * checks. If you are unsure, use LZMA_CHECK_CRC32.
+ *
+ * \return - LZMA_OK: Initialization succeeded. Use lzma_code() to
+ * encode your data.
+ * - LZMA_MEM_ERROR: Memory allocation failed.
+ * - LZMA_OPTIONS_ERROR: The given compression level is not
+ * supported by this build of liblzma.
+ * - LZMA_UNSUPPORTED_CHECK: The given check type is not
+ * supported by this liblzma build.
+ * - LZMA_PROG_ERROR: One or more of the parameters have values
+ * that will never be valid. For example, strm == NULL.
+ *
+ * If initialization fails (return value is not LZMA_OK), all the memory
+ * allocated for *strm by liblzma is always freed. Thus, there is no need
+ * to call lzma_end() after failed initialization.
+ *
+ * If initialization succeeds, use lzma_code() to do the actual encoding.
+ * Valid values for `action' (the second argument of lzma_code()) are
+ * LZMA_RUN, LZMA_SYNC_FLUSH, LZMA_FULL_FLUSH, and LZMA_FINISH. In future,
+ * there may be compression levels or flags that don't support LZMA_SYNC_FLUSH.
+ */
+extern LZMA_API(lzma_ret) lzma_easy_encoder(
+ lzma_stream *strm, uint32_t preset, lzma_check check)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Single-call .xz Stream encoding using a preset number
+ *
+ * The maximum required output buffer size can be calculated with
+ * lzma_stream_buffer_bound().
+ *
+ * \param preset Compression preset to use. See the description
+ * in lzma_easy_encoder().
+ * \param check Type of the integrity check to calculate from
+ * uncompressed data.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_size Size of the input buffer
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_BUF_ERROR: Not enough output buffer space.
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_MEM_ERROR
+ * - LZMA_DATA_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_easy_buffer_encode(
+ uint32_t preset, lzma_check check,
+ lzma_allocator *allocator, const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size) lzma_nothrow;
+
+
+/**
+ * \brief Initialize .xz Stream encoder using a custom filter chain
+ *
+ * \param strm Pointer to properly prepared lzma_stream
+ * \param filters Array of filters. This must be terminated with
+ * filters[n].id = LZMA_VLI_UNKNOWN. See filter.h for
+ * more information.
+ * \param check Type of the integrity check to calculate from
+ * uncompressed data.
+ *
+ * \return - LZMA_OK: Initialization was successful.
+ * - LZMA_MEM_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_stream_encoder(lzma_stream *strm,
+ const lzma_filter *filters, lzma_check check)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Initialize .lzma encoder (legacy file format)
+ *
+ * The .lzma format is sometimes called the LZMA_Alone format, which is the
+ * reason for the name of this function. The .lzma format supports only the
+ * LZMA1 filter. There is no support for integrity checks like CRC32.
+ *
+ * Use this function if and only if you need to create files readable by
+ * legacy LZMA tools such as LZMA Utils 4.32.x. Moving to the .xz format
+ * is strongly recommended.
+ *
+ * The valid action values for lzma_code() are LZMA_RUN and LZMA_FINISH.
+ * No kind of flushing is supported, because the file format doesn't make
+ * it possible.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_alone_encoder(
+ lzma_stream *strm, const lzma_options_lzma *options)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Calculate output buffer size for single-call Stream encoder
+ *
+ * When trying to compress uncompressible data, the encoded size will be
+ * slightly bigger than the input data. This function calculates how much
+ * output buffer space is required to be sure that lzma_stream_buffer_encode()
+ * doesn't return LZMA_BUF_ERROR.
+ *
+ * The calculated value is not exact, but it is guaranteed to be big enough.
+ * The actual maximum output space required may be slightly smaller (up to
+ * about 100 bytes). This should not be a problem in practice.
+ *
+ * If the calculated maximum size doesn't fit into size_t or would make the
+ * Stream grow past LZMA_VLI_MAX (which should never happen in practice),
+ * zero is returned to indicate the error.
+ *
+ * \note The limit calculated by this function applies only to
+ * single-call encoding. Multi-call encoding may (and probably
+ * will) have larger maximum expansion when encoding
+ * uncompressible data. Currently there is no function to
+ * calculate the maximum expansion of multi-call encoding.
+ */
+extern LZMA_API(size_t) lzma_stream_buffer_bound(size_t uncompressed_size)
+ lzma_nothrow;
+
+
+/**
+ * \brief Single-call .xz Stream encoder
+ *
+ * \param filters Array of filters. This must be terminated with
+ * filters[n].id = LZMA_VLI_UNKNOWN. See filter.h
+ * for more information.
+ * \param check Type of the integrity check to calculate from
+ * uncompressed data.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_size Size of the input buffer
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_BUF_ERROR: Not enough output buffer space.
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_MEM_ERROR
+ * - LZMA_DATA_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_stream_buffer_encode(
+ lzma_filter *filters, lzma_check check,
+ lzma_allocator *allocator, const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/************
+ * Decoding *
+ ************/
+
+/**
+ * This flag makes lzma_code() return LZMA_NO_CHECK if the input stream
+ * being decoded has no integrity check. Note that when used with
+ * lzma_auto_decoder(), all .lzma files will trigger LZMA_NO_CHECK
+ * if LZMA_TELL_NO_CHECK is used.
+ */
+#define LZMA_TELL_NO_CHECK UINT32_C(0x01)
+
+
+/**
+ * This flag makes lzma_code() return LZMA_UNSUPPORTED_CHECK if the input
+ * stream has an integrity check, but the type of the integrity check is not
+ * supported by this liblzma version or build. Such files can still be
+ * decoded, but the integrity check cannot be verified.
+ */
+#define LZMA_TELL_UNSUPPORTED_CHECK UINT32_C(0x02)
+
+
+/**
+ * This flag makes lzma_code() return LZMA_GET_CHECK as soon as the type
+ * of the integrity check is known. The type can then be got with
+ * lzma_get_check().
+ */
+#define LZMA_TELL_ANY_CHECK UINT32_C(0x04)
+
+
+/**
+ * This flag enables decoding of concatenated files with file formats that
+ * allow concatenating compressed files as is. From the formats currently
+ * supported by liblzma, only the .xz format allows concatenated files.
+ * Concatenated files are not allowed with the legacy .lzma format.
+ *
+ * This flag also affects the usage of the `action' argument for lzma_code().
+ * When LZMA_CONCATENATED is used, lzma_code() won't return LZMA_STREAM_END
+ * unless LZMA_FINISH is used as `action'. Thus, the application has to set
+ * LZMA_FINISH in the same way as it does when encoding.
+ *
+ * If LZMA_CONCATENATED is not used, the decoders still accept LZMA_FINISH
+ * as `action' for lzma_code(), but the usage of LZMA_FINISH isn't required.
+ */
+#define LZMA_CONCATENATED UINT32_C(0x08)
+
+
+/**
+ * \brief Initialize .xz Stream decoder
+ *
+ * \param strm Pointer to properly prepared lzma_stream
+ * \param memlimit Rough memory usage limit as bytes
+ * \param flags Bitwise-or of zero or more of the decoder flags:
+ * LZMA_TELL_NO_CHECK, LZMA_TELL_UNSUPPORTED_CHECK,
+ * LZMA_TELL_ANY_CHECK, LZMA_CONCATENATED
+ *
+ * \return - LZMA_OK: Initialization was successful.
+ * - LZMA_MEM_ERROR: Cannot allocate memory.
+ * - LZMA_OPTIONS_ERROR: Unsupported flags
+ */
+extern LZMA_API(lzma_ret) lzma_stream_decoder(
+ lzma_stream *strm, uint64_t memlimit, uint32_t flags)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Decode .xz Streams and .lzma files with autodetection
+ *
+ * This decoder autodetects between the .xz and .lzma file formats, and
+ * calls lzma_stream_decoder() or lzma_alone_decoder() once the type
+ * of the input file has been detected.
+ *
+ * \param strm Pointer to properly prepared lzma_stream
+ * \param memlimit Rough memory usage limit as bytes
+ * \param flags Bitwise-or of flags, or zero for no flags.
+ *
+ * \return - LZMA_OK: Initialization was successful.
+ * - LZMA_MEM_ERROR: Cannot allocate memory.
+ * - LZMA_OPTIONS_ERROR: Unsupported flags
+ */
+extern LZMA_API(lzma_ret) lzma_auto_decoder(
+ lzma_stream *strm, uint64_t memlimit, uint32_t flags)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Initialize .lzma decoder (legacy file format)
+ *
+ * Valid `action' arguments to lzma_code() are LZMA_RUN and LZMA_FINISH.
+ * There is no need to use LZMA_FINISH, but allowing it may simplify
+ * certain types of applications.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_alone_decoder(
+ lzma_stream *strm, uint64_t memlimit)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Single-call .xz Stream decoder
+ *
+ * \param memlimit Pointer to how much memory the decoder is allowed
+ * to allocate. The value pointed by this pointer is
+ * modified if and only if LZMA_MEMLIMIT_ERROR is
+ * returned.
+ * \param flags Bitwise-or of zero or more of the decoder flags:
+ * LZMA_TELL_NO_CHECK, LZMA_TELL_UNSUPPORTED_CHECK,
+ * LZMA_CONCATENATED. Note that LZMA_TELL_ANY_CHECK
+ * is not allowed and will return LZMA_PROG_ERROR.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_pos The next byte will be read from in[*in_pos].
+ * *in_pos is updated only if decoding succeeds.
+ * \param in_size Size of the input buffer; the first byte that
+ * won't be read is in[in_size].
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Decoding was successful.
+ * - LZMA_FORMAT_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_DATA_ERROR
+ * - LZMA_NO_CHECK: This can be returned only if using
+ * the LZMA_TELL_NO_CHECK flag.
+ * - LZMA_UNSUPPORTED_CHECK: This can be returned only if using
+ * the LZMA_TELL_UNSUPPORTED_CHECK flag.
+ * - LZMA_MEM_ERROR
+ * - LZMA_MEMLIMIT_ERROR: Memory usage limit was reached.
+ * The minimum required memlimit value was stored to *memlimit.
+ * - LZMA_BUF_ERROR: Output buffer was too small.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_stream_buffer_decode(
+ uint64_t *memlimit, uint32_t flags, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
diff --git a/src/liblzma/api/lzma/delta.h b/src/liblzma/api/lzma/delta.h
new file mode 100644
index 0000000..592fc4f
--- /dev/null
+++ b/src/liblzma/api/lzma/delta.h
@@ -0,0 +1,77 @@
+/**
+ * \file lzma/delta.h
+ * \brief Delta filter
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Filter ID
+ *
+ * Filter ID of the Delta filter. This is used as lzma_filter.id.
+ */
+#define LZMA_FILTER_DELTA LZMA_VLI_C(0x03)
+
+
+/**
+ * \brief Type of the delta calculation
+ *
+ * Currently only byte-wise delta is supported. Other possible types could
+ * be, for example, delta of 16/32/64-bit little/big endian integers, but
+ * these are not currently planned since byte-wise delta is almost as good.
+ */
+typedef enum {
+ LZMA_DELTA_TYPE_BYTE
+} lzma_delta_type;
+
+
+/**
+ * \brief Options for the Delta filter
+ *
+ * These options are needed by both encoder and decoder.
+ */
+typedef struct {
+ /** For now, this must always be LZMA_DELTA_TYPE_BYTE. */
+ lzma_delta_type type;
+
+ /**
+ * \brief Delta distance
+ *
+ * With the only currently supported type, LZMA_DELTA_TYPE_BYTE,
+ * the distance is as bytes.
+ *
+ * Examples:
+ * - 16-bit stereo audio: distance = 4 bytes
+ * - 24-bit RGB image data: distance = 3 bytes
+ */
+ uint32_t dist;
+# define LZMA_DELTA_DIST_MIN 1
+# define LZMA_DELTA_DIST_MAX 256
+
+ /*
+ * Reserved space to allow possible future extensions without
+ * breaking the ABI. You should not touch these, because the names
+ * of these variables may change. These are and will never be used
+ * when type is LZMA_DELTA_TYPE_BYTE, so it is safe to leave these
+ * uninitialized.
+ */
+ uint32_t reserved_int1;
+ uint32_t reserved_int2;
+ uint32_t reserved_int3;
+ uint32_t reserved_int4;
+ void *reserved_ptr1;
+ void *reserved_ptr2;
+
+} lzma_options_delta;
diff --git a/src/liblzma/api/lzma/filter.h b/src/liblzma/api/lzma/filter.h
new file mode 100644
index 0000000..038a933
--- /dev/null
+++ b/src/liblzma/api/lzma/filter.h
@@ -0,0 +1,421 @@
+/**
+ * \file lzma/filter.h
+ * \brief Common filter related types
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Maximum number of filters in a chain
+ *
+ * A filter chain can have 1-4 filters, of which three are allowed to change
+ * the size of the data. Usually only one or two filters are needed.
+ */
+#define LZMA_FILTERS_MAX 4
+
+
+/**
+ * \brief Filter options
+ *
+ * This structure is used to pass Filter ID and a pointer filter's
+ * options to liblzma. A few functions work with a single lzma_filter
+ * structure, while most functions expect a filter chain.
+ *
+ * A filter chain is indicated with an array of lzma_filter structures.
+ * The array is terminated with .id = LZMA_VLI_UNKNOWN. Thus, the filter
+ * array must have LZMA_FILTERS_MAX + 1 elements (that is, five) to
+ * be able to hold any arbitrary filter chain. This is important when
+ * using lzma_block_header_decode() from block.h, because too small
+ * array would make liblzma write past the end of the filters array.
+ */
+typedef struct {
+ /**
+ * \brief Filter ID
+ *
+ * Use constants whose name begin with `LZMA_FILTER_' to specify
+ * different filters. In an array of lzma_filter structures, use
+ * LZMA_VLI_UNKNOWN to indicate end of filters.
+ *
+ * \note This is not an enum, because on some systems enums
+ * cannot be 64-bit.
+ */
+ lzma_vli id;
+
+ /**
+ * \brief Pointer to filter-specific options structure
+ *
+ * If the filter doesn't need options, set this to NULL. If id is
+ * set to LZMA_VLI_UNKNOWN, options is ignored, and thus
+ * doesn't need be initialized.
+ *
+ * Some filters support changing the options in the middle of
+ * the encoding process. These filters store the pointer of the
+ * options structure and communicate with the application via
+ * modifications of the options structure.
+ */
+ void *options;
+
+} lzma_filter;
+
+
+/**
+ * \brief Test if the given Filter ID is supported for encoding
+ *
+ * Return true if the give Filter ID is supported for encoding by this
+ * liblzma build. Otherwise false is returned.
+ *
+ * There is no way to list which filters are available in this particular
+ * liblzma version and build. It would be useless, because the application
+ * couldn't know what kind of options the filter would need.
+ */
+extern LZMA_API(lzma_bool) lzma_filter_encoder_is_supported(lzma_vli id)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Test if the given Filter ID is supported for decoding
+ *
+ * Return true if the give Filter ID is supported for decoding by this
+ * liblzma build. Otherwise false is returned.
+ */
+extern LZMA_API(lzma_bool) lzma_filter_decoder_is_supported(lzma_vli id)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Copy the filters array
+ *
+ * Copy the Filter IDs and filter-specific options from src to dest.
+ * Up to LZMA_FILTERS_MAX filters are copied, plus the terminating
+ * .id == LZMA_VLI_UNKNOWN. Thus, dest should have at least
+ * LZMA_FILTERS_MAX + 1 elements space unless the caller knows that
+ * src is smaller than that.
+ *
+ * Unless the filter-specific options is NULL, the Filter ID has to be
+ * supported by liblzma, because liblzma needs to know the size of every
+ * filter-specific options structure. The filter-specific options are not
+ * validated. If options is NULL, any unsupported Filter IDs are copied
+ * without returning an error.
+ *
+ * Old filter-specific options in dest are not freed, so dest doesn't
+ * need to be initialized by the caller in any way.
+ *
+ * If an error occurs, memory possibly already allocated by this function
+ * is always freed.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ * - LZMA_OPTIONS_ERROR: Unsupported Filter ID and its options
+ * is not NULL.
+ * - LZMA_PROG_ERROR: src or dest is NULL.
+ */
+extern LZMA_API(lzma_ret) lzma_filters_copy(const lzma_filter *src,
+ lzma_filter *dest, lzma_allocator *allocator) lzma_nothrow;
+
+
+/**
+ * \brief Calculate rough memory requirements for raw encoder
+ *
+ * Because the calculation is rough, this function can be used to calculate
+ * the memory requirements for Block and Stream encoders too.
+ *
+ * \param filters Array of filters terminated with
+ * .id == LZMA_VLI_UNKNOWN.
+ *
+ * \return Rough number of bytes of memory required for the given
+ * filter chain when encoding.
+ */
+extern LZMA_API(uint64_t) lzma_raw_encoder_memusage(const lzma_filter *filters)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Calculate rough memory requirements for raw decoder
+ *
+ * Because the calculation is rough, this function can be used to calculate
+ * the memory requirements for Block and Stream decoders too.
+ *
+ * \param filters Array of filters terminated with
+ * .id == LZMA_VLI_UNKNOWN.
+ *
+ * \return Rough number of bytes of memory required for the given
+ * filter chain when decoding.
+ */
+extern LZMA_API(uint64_t) lzma_raw_decoder_memusage(const lzma_filter *filters)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Initialize raw encoder
+ *
+ * This function may be useful when implementing custom file formats.
+ *
+ * \param strm Pointer to properly prepared lzma_stream
+ * \param filters Array of lzma_filter structures. The end of the
+ * array must be marked with .id = LZMA_VLI_UNKNOWN.
+ *
+ * The `action' with lzma_code() can be LZMA_RUN, LZMA_SYNC_FLUSH (if the
+ * filter chain supports it), or LZMA_FINISH.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_raw_encoder(
+ lzma_stream *strm, const lzma_filter *filters)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Initialize raw decoder
+ *
+ * The initialization of raw decoder goes similarly to raw encoder.
+ *
+ * The `action' with lzma_code() can be LZMA_RUN or LZMA_FINISH. Using
+ * LZMA_FINISH is not required, it is supported just for convenience.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_raw_decoder(
+ lzma_stream *strm, const lzma_filter *filters)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Update the filter chain in the encoder
+ *
+ * This function is for advanced users only. This function has two slightly
+ * different purposes:
+ *
+ * - After LZMA_FULL_FLUSH when using Stream encoder: Set a new filter
+ * chain, which will be used starting from the next Block.
+ *
+ * - After LZMA_SYNC_FLUSH using Raw, Block, or Stream encoder: Change
+ * the filter-specific options in the middle of encoding. The actual
+ * filters in the chain (Filter IDs) cannot be changed. In the future,
+ * it might become possible to change the filter options without
+ * using LZMA_SYNC_FLUSH.
+ *
+ * While rarely useful, this function may be called also when no data has
+ * been compressed yet. In that case, this function will behave as if
+ * LZMA_FULL_FLUSH (Stream encoder) or LZMA_SYNC_FLUSH (Raw or Block
+ * encoder) had been used right before calling this function.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ * - LZMA_MEMLIMIT_ERROR
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_filters_update(
+ lzma_stream *strm, const lzma_filter *filters) lzma_nothrow;
+
+
+/**
+ * \brief Single-call raw encoder
+ *
+ * \param filters Array of lzma_filter structures. The end of the
+ * array must be marked with .id = LZMA_VLI_UNKNOWN.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_size Size of the input buffer
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_BUF_ERROR: Not enough output buffer space.
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_MEM_ERROR
+ * - LZMA_DATA_ERROR
+ * - LZMA_PROG_ERROR
+ *
+ * \note There is no function to calculate how big output buffer
+ * would surely be big enough. (lzma_stream_buffer_bound()
+ * works only for lzma_stream_buffer_encode().)
+ */
+extern LZMA_API(lzma_ret) lzma_raw_buffer_encode(
+ const lzma_filter *filters, lzma_allocator *allocator,
+ const uint8_t *in, size_t in_size, uint8_t *out,
+ size_t *out_pos, size_t out_size) lzma_nothrow;
+
+
+/**
+ * \brief Single-call raw decoder
+ *
+ * \param filters Array of lzma_filter structures. The end of the
+ * array must be marked with .id = LZMA_VLI_UNKNOWN.
+ * \param allocator lzma_allocator for custom allocator functions.
+ * Set to NULL to use malloc() and free().
+ * \param in Beginning of the input buffer
+ * \param in_pos The next byte will be read from in[*in_pos].
+ * *in_pos is updated only if decoding succeeds.
+ * \param in_size Size of the input buffer; the first byte that
+ * won't be read is in[in_size].
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ */
+extern LZMA_API(lzma_ret) lzma_raw_buffer_decode(
+ const lzma_filter *filters, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size) lzma_nothrow;
+
+
+/**
+ * \brief Get the size of the Filter Properties field
+ *
+ * This function may be useful when implementing custom file formats
+ * using the raw encoder and decoder.
+ *
+ * \param size Pointer to uint32_t to hold the size of the properties
+ * \param filter Filter ID and options (the size of the properties may
+ * vary depending on the options)
+ *
+ * \return - LZMA_OK
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ *
+ * \note This function validates the Filter ID, but does not
+ * necessarily validate the options. Thus, it is possible
+ * that this returns LZMA_OK while the following call to
+ * lzma_properties_encode() returns LZMA_OPTIONS_ERROR.
+ */
+extern LZMA_API(lzma_ret) lzma_properties_size(
+ uint32_t *size, const lzma_filter *filter) lzma_nothrow;
+
+
+/**
+ * \brief Encode the Filter Properties field
+ *
+ * \param filter Filter ID and options
+ * \param props Buffer to hold the encoded options. The size of
+ * buffer must have been already determined with
+ * lzma_properties_size().
+ *
+ * \return - LZMA_OK
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_PROG_ERROR
+ *
+ * \note Even this function won't validate more options than actually
+ * necessary. Thus, it is possible that encoding the properties
+ * succeeds but using the same options to initialize the encoder
+ * will fail.
+ *
+ * \note It is OK to skip calling this function if
+ * lzma_properties_size() indicated that the size
+ * of the Filter Properties field is zero.
+ */
+extern LZMA_API(lzma_ret) lzma_properties_encode(
+ const lzma_filter *filter, uint8_t *props) lzma_nothrow;
+
+
+/**
+ * \brief Decode the Filter Properties field
+ *
+ * \param filter filter->id must have been set to the correct
+ * Filter ID. filter->options doesn't need to be
+ * initialized (it's not freed by this function). The
+ * decoded options will be stored to filter->options.
+ * filter->options is set to NULL if there are no
+ * properties or if an error occurs.
+ * \param allocator Custom memory allocator used to allocate the
+ * options. Set to NULL to use the default malloc(),
+ * and in case of an error, also free().
+ * \param props Input buffer containing the properties.
+ * \param props_size Size of the properties. This must be the exact
+ * size; giving too much or too little input will
+ * return LZMA_OPTIONS_ERROR.
+ *
+ * \return - LZMA_OK
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_MEM_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_properties_decode(
+ lzma_filter *filter, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size) lzma_nothrow;
+
+
+/**
+ * \brief Calculate encoded size of a Filter Flags field
+ *
+ * Knowing the size of Filter Flags is useful to know when allocating
+ * memory to hold the encoded Filter Flags.
+ *
+ * \param size Pointer to integer to hold the calculated size
+ * \param filters Filter ID and associated options whose encoded
+ * size is to be calculated
+ *
+ * \return - LZMA_OK: *size set successfully. Note that this doesn't
+ * guarantee that filters->options is valid, thus
+ * lzma_filter_flags_encode() may still fail.
+ * - LZMA_OPTIONS_ERROR: Unknown Filter ID or unsupported options.
+ * - LZMA_PROG_ERROR: Invalid options
+ *
+ * \note If you need to calculate size of List of Filter Flags,
+ * you need to loop over every lzma_filter entry.
+ */
+extern LZMA_API(lzma_ret) lzma_filter_flags_size(
+ uint32_t *size, const lzma_filter *filters)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Encode Filter Flags into given buffer
+ *
+ * In contrast to some functions, this doesn't allocate the needed buffer.
+ * This is due to how this function is used internally by liblzma.
+ *
+ * \param filters Filter ID and options to be encoded
+ * \param out Beginning of the output buffer
+ * \param out_pos out[*out_pos] is the next write position. This
+ * is updated by the encoder.
+ * \param out_size out[out_size] is the first byte to not write.
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_OPTIONS_ERROR: Invalid or unsupported options.
+ * - LZMA_PROG_ERROR: Invalid options or not enough output
+ * buffer space (you should have checked it with
+ * lzma_filter_flags_size()).
+ */
+extern LZMA_API(lzma_ret) lzma_filter_flags_encode(const lzma_filter *filters,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Decode Filter Flags from given buffer
+ *
+ * The decoded result is stored into *filters. filters->options is
+ * initialized but the old value is NOT free()d.
+ *
+ * \return - LZMA_OK
+ * - LZMA_OPTIONS_ERROR
+ * - LZMA_MEM_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_filter_flags_decode(
+ lzma_filter *filters, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
diff --git a/src/liblzma/api/lzma/hardware.h b/src/liblzma/api/lzma/hardware.h
new file mode 100644
index 0000000..f44cb60
--- /dev/null
+++ b/src/liblzma/api/lzma/hardware.h
@@ -0,0 +1,51 @@
+/**
+ * \file lzma/hardware.h
+ * \brief Hardware information
+ *
+ * Since liblzma can consume a lot of system resources, it also provides
+ * ways to limit the resource usage. Applications linking against liblzma
+ * need to do the actual decisions how much resources to let liblzma to use.
+ * To ease making these decisions, liblzma provides functions to find out
+ * the relevant capabilities of the underlaying hardware. Currently there
+ * is only a function to find out the amount of RAM, but in the future there
+ * will be also a function to detect how many concurrent threads the system
+ * can run.
+ *
+ * \note On some operating systems, these function may temporarily
+ * load a shared library or open file descriptor(s) to find out
+ * the requested hardware information. Unless the application
+ * assumes that specific file descriptors are not touched by
+ * other threads, this should have no effect on thread safety.
+ * Possible operations involving file descriptors will restart
+ * the syscalls if they return EINTR.
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Get the total amount of physical memory (RAM) in bytes
+ *
+ * This function may be useful when determining a reasonable memory
+ * usage limit for decompressing or how much memory it is OK to use
+ * for compressing. For example, the default limit used by the xz
+ * command line tool is 40 % of RAM.
+ *
+ * \return On success, the total amount of physical memory in bytes
+ * is returned. If the amount of RAM cannot be determined,
+ * zero is returned. This can happen if an error occurs
+ * or if there is no code in liblzma to detect the amount
+ * of RAM on the specific operating system.
+ */
+extern LZMA_API(uint64_t) lzma_physmem(void) lzma_nothrow;
diff --git a/src/liblzma/api/lzma/index.h b/src/liblzma/api/lzma/index.h
new file mode 100644
index 0000000..5ea12e3
--- /dev/null
+++ b/src/liblzma/api/lzma/index.h
@@ -0,0 +1,676 @@
+/**
+ * \file lzma/index.h
+ * \brief Handling of .xz Index and related information
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Opaque data type to hold the Index(es) and other information
+ *
+ * lzma_index often holds just one .xz Index and possibly the Stream Flags
+ * of the same Stream and size of the Stream Padding field. However,
+ * multiple lzma_indexes can be concatenated with lzma_index_cat() and then
+ * there may be information about multiple Streams in the same lzma_index.
+ *
+ * Notes about thread safety: Only one thread may modify lzma_index at
+ * a time. All functions that take non-const pointer to lzma_index
+ * modify it. As long as no thread is modifying the lzma_index, getting
+ * information from the same lzma_index can be done from multiple threads
+ * at the same time with functions that take a const pointer to
+ * lzma_index or use lzma_index_iter. The same iterator must be used
+ * only by one thread at a time, of course, but there can be as many
+ * iterators for the same lzma_index as needed.
+ */
+typedef struct lzma_index_s lzma_index;
+
+
+/**
+ * \brief Iterator to get information about Blocks and Streams
+ */
+typedef struct {
+ struct {
+ /**
+ * \brief Pointer to Stream Flags
+ *
+ * This is NULL if Stream Flags have not been set for
+ * this Stream with lzma_index_stream_flags().
+ */
+ const lzma_stream_flags *flags;
+
+ const void *reserved_ptr1;
+ const void *reserved_ptr2;
+ const void *reserved_ptr3;
+
+ /**
+ * \brief Stream number in the lzma_index
+ *
+ * The first Stream is 1.
+ */
+ lzma_vli number;
+
+ /**
+ * \brief Number of Blocks in the Stream
+ *
+ * If this is zero, the block structure below has
+ * undefined values.
+ */
+ lzma_vli block_count;
+
+ /**
+ * \brief Compressed start offset of this Stream
+ *
+ * The offset is relative to the beginning of the lzma_index
+ * (i.e. usually the beginning of the .xz file).
+ */
+ lzma_vli compressed_offset;
+
+ /**
+ * \brief Uncompressed start offset of this Stream
+ *
+ * The offset is relative to the beginning of the lzma_index
+ * (i.e. usually the beginning of the .xz file).
+ */
+ lzma_vli uncompressed_offset;
+
+ /**
+ * \brief Compressed size of this Stream
+ *
+ * This includes all headers except the possible
+ * Stream Padding after this Stream.
+ */
+ lzma_vli compressed_size;
+
+ /**
+ * \brief Uncompressed size of this Stream
+ */
+ lzma_vli uncompressed_size;
+
+ /**
+ * \brief Size of Stream Padding after this Stream
+ *
+ * If it hasn't been set with lzma_index_stream_padding(),
+ * this defaults to zero. Stream Padding is always
+ * a multiple of four bytes.
+ */
+ lzma_vli padding;
+
+ lzma_vli reserved_vli1;
+ lzma_vli reserved_vli2;
+ lzma_vli reserved_vli3;
+ lzma_vli reserved_vli4;
+ } stream;
+
+ struct {
+ /**
+ * \brief Block number in the file
+ *
+ * The first Block is 1.
+ */
+ lzma_vli number_in_file;
+
+ /**
+ * \brief Compressed start offset of this Block
+ *
+ * This offset is relative to the beginning of the
+ * lzma_index (i.e. usually the beginning of the .xz file).
+ * Normally this is where you should seek in the .xz file
+ * to start decompressing this Block.
+ */
+ lzma_vli compressed_file_offset;
+
+ /**
+ * \brief Uncompressed start offset of this Block
+ *
+ * This offset is relative to the beginning of the lzma_index
+ * (i.e. usually the beginning of the .xz file).
+ */
+ lzma_vli uncompressed_file_offset;
+
+ /**
+ * \brief Block number in this Stream
+ *
+ * The first Block is 1.
+ */
+ lzma_vli number_in_stream;
+
+ /**
+ * \brief Compressed start offset of this Block
+ *
+ * This offset is relative to the beginning of the Stream
+ * containing this Block.
+ */
+ lzma_vli compressed_stream_offset;
+
+ /**
+ * \brief Uncompressed start offset of this Block
+ *
+ * This offset is relative to the beginning of the Stream
+ * containing this Block.
+ */
+ lzma_vli uncompressed_stream_offset;
+
+ /**
+ * \brief Uncompressed size of this Block
+ *
+ * You should pass this to the Block decoder if you will
+ * decode this Block.
+ *
+ * When doing random-access reading, it is possible that
+ * the target offset is not exactly at Block boundary. One
+ * will need to compare the target offset against
+ * uncompressed_file_offset or uncompressed_stream_offset,
+ * and possibly decode and throw away some amount of data
+ * before reaching the target offset.
+ */
+ lzma_vli uncompressed_size;
+
+ /**
+ * \brief Unpadded size of this Block
+ *
+ * You should pass this to the Block decoder if you will
+ * decode this Block.
+ */
+ lzma_vli unpadded_size;
+
+ /**
+ * \brief Total compressed size
+ *
+ * This includes all headers and padding in this Block.
+ * This is useful if you need to know how many bytes
+ * the Block decoder will actually read.
+ */
+ lzma_vli total_size;
+
+ lzma_vli reserved_vli1;
+ lzma_vli reserved_vli2;
+ lzma_vli reserved_vli3;
+ lzma_vli reserved_vli4;
+
+ const void *reserved_ptr1;
+ const void *reserved_ptr2;
+ const void *reserved_ptr3;
+ const void *reserved_ptr4;
+ } block;
+
+ /*
+ * Internal data which is used to store the state of the iterator.
+ * The exact format may vary between liblzma versions, so don't
+ * touch these in any way.
+ */
+ union {
+ const void *p;
+ size_t s;
+ lzma_vli v;
+ } internal[6];
+} lzma_index_iter;
+
+
+/**
+ * \brief Operation mode for lzma_index_iter_next()
+ */
+typedef enum {
+ LZMA_INDEX_ITER_ANY = 0,
+ /**<
+ * \brief Get the next Block or Stream
+ *
+ * Go to the next Block if the current Stream has at least
+ * one Block left. Otherwise go to the next Stream even if
+ * it has no Blocks. If the Stream has no Blocks
+ * (lzma_index_iter.stream.block_count == 0),
+ * lzma_index_iter.block will have undefined values.
+ */
+
+ LZMA_INDEX_ITER_STREAM = 1,
+ /**<
+ * \brief Get the next Stream
+ *
+ * Go to the next Stream even if the current Stream has
+ * unread Blocks left. If the next Stream has at least one
+ * Block, the iterator will point to the first Block.
+ * If there are no Blocks, lzma_index_iter.block will have
+ * undefined values.
+ */
+
+ LZMA_INDEX_ITER_BLOCK = 2,
+ /**<
+ * \brief Get the next Block
+ *
+ * Go to the next Block if the current Stream has at least
+ * one Block left. If the current Stream has no Blocks left,
+ * the next Stream with at least one Block is located and
+ * the iterator will be made to point to the first Block of
+ * that Stream.
+ */
+
+ LZMA_INDEX_ITER_NONEMPTY_BLOCK = 3
+ /**<
+ * \brief Get the next non-empty Block
+ *
+ * This is like LZMA_INDEX_ITER_BLOCK except that it will
+ * skip Blocks whose Uncompressed Size is zero.
+ */
+
+} lzma_index_iter_mode;
+
+
+/**
+ * \brief Calculate memory usage of lzma_index
+ *
+ * On disk, the size of the Index field depends on both the number of Records
+ * stored and how big values the Records store (due to variable-length integer
+ * encoding). When the Index is kept in lzma_index structure, the memory usage
+ * depends only on the number of Records/Blocks stored in the Index(es), and
+ * in case of concatenated lzma_indexes, the number of Streams. The size in
+ * RAM is almost always significantly bigger than in the encoded form on disk.
+ *
+ * This function calculates an approximate amount of memory needed hold
+ * the given number of Streams and Blocks in lzma_index structure. This
+ * value may vary between CPU architectures and also between liblzma versions
+ * if the internal implementation is modified.
+ */
+extern LZMA_API(uint64_t) lzma_index_memusage(
+ lzma_vli streams, lzma_vli blocks) lzma_nothrow;
+
+
+/**
+ * \brief Calculate the memory usage of an existing lzma_index
+ *
+ * This is a shorthand for lzma_index_memusage(lzma_index_stream_count(i),
+ * lzma_index_block_count(i)).
+ */
+extern LZMA_API(uint64_t) lzma_index_memused(const lzma_index *i)
+ lzma_nothrow;
+
+
+/**
+ * \brief Allocate and initialize a new lzma_index structure
+ *
+ * \return On success, a pointer to an empty initialized lzma_index is
+ * returned. If allocation fails, NULL is returned.
+ */
+extern LZMA_API(lzma_index *) lzma_index_init(lzma_allocator *allocator)
+ lzma_nothrow;
+
+
+/**
+ * \brief Deallocate lzma_index
+ *
+ * If i is NULL, this does nothing.
+ */
+extern LZMA_API(void) lzma_index_end(lzma_index *i, lzma_allocator *allocator)
+ lzma_nothrow;
+
+
+/**
+ * \brief Add a new Block to lzma_index
+ *
+ * \param i Pointer to a lzma_index structure
+ * \param allocator Pointer to lzma_allocator, or NULL to
+ * use malloc()
+ * \param unpadded_size Unpadded Size of a Block. This can be
+ * calculated with lzma_block_unpadded_size()
+ * after encoding or decoding the Block.
+ * \param uncompressed_size Uncompressed Size of a Block. This can be
+ * taken directly from lzma_block structure
+ * after encoding or decoding the Block.
+ *
+ * Appending a new Block does not invalidate iterators. For example,
+ * if an iterator was pointing to the end of the lzma_index, after
+ * lzma_index_append() it is possible to read the next Block with
+ * an existing iterator.
+ *
+ * \return - LZMA_OK
+ * - LZMA_MEM_ERROR
+ * - LZMA_DATA_ERROR: Compressed or uncompressed size of the
+ * Stream or size of the Index field would grow too big.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_append(
+ lzma_index *i, lzma_allocator *allocator,
+ lzma_vli unpadded_size, lzma_vli uncompressed_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Set the Stream Flags
+ *
+ * Set the Stream Flags of the last (and typically the only) Stream
+ * in lzma_index. This can be useful when reading information from the
+ * lzma_index, because to decode Blocks, knowing the integrity check type
+ * is needed.
+ *
+ * The given Stream Flags are copied into internal preallocated structure
+ * in the lzma_index, thus the caller doesn't need to keep the *stream_flags
+ * available after calling this function.
+ *
+ * \return - LZMA_OK
+ * - LZMA_OPTIONS_ERROR: Unsupported stream_flags->version.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_stream_flags(
+ lzma_index *i, const lzma_stream_flags *stream_flags)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Get the types of integrity Checks
+ *
+ * If lzma_index_stream_flags() is used to set the Stream Flags for
+ * every Stream, lzma_index_checks() can be used to get a bitmask to
+ * indicate which Check types have been used. It can be useful e.g. if
+ * showing the Check types to the user.
+ *
+ * The bitmask is 1 << check_id, e.g. CRC32 is 1 << 1 and SHA-256 is 1 << 10.
+ */
+extern LZMA_API(uint32_t) lzma_index_checks(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Set the amount of Stream Padding
+ *
+ * Set the amount of Stream Padding of the last (and typically the only)
+ * Stream in the lzma_index. This is needed when planning to do random-access
+ * reading within multiple concatenated Streams.
+ *
+ * By default, the amount of Stream Padding is assumed to be zero bytes.
+ *
+ * \return - LZMA_OK
+ * - LZMA_DATA_ERROR: The file size would grow too big.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_stream_padding(
+ lzma_index *i, lzma_vli stream_padding)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Get the number of Streams
+ */
+extern LZMA_API(lzma_vli) lzma_index_stream_count(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the number of Blocks
+ *
+ * This returns the total number of Blocks in lzma_index. To get number
+ * of Blocks in individual Streams, use lzma_index_iter.
+ */
+extern LZMA_API(lzma_vli) lzma_index_block_count(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the size of the Index field as bytes
+ *
+ * This is needed to verify the Backward Size field in the Stream Footer.
+ */
+extern LZMA_API(lzma_vli) lzma_index_size(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the total size of the Stream
+ *
+ * If multiple lzma_indexes have been combined, this works as if the Blocks
+ * were in a single Stream. This is useful if you are going to combine
+ * Blocks from multiple Streams into a single new Stream.
+ */
+extern LZMA_API(lzma_vli) lzma_index_stream_size(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the total size of the Blocks
+ *
+ * This doesn't include the Stream Header, Stream Footer, Stream Padding,
+ * or Index fields.
+ */
+extern LZMA_API(lzma_vli) lzma_index_total_size(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the total size of the file
+ *
+ * When no lzma_indexes have been combined with lzma_index_cat() and there is
+ * no Stream Padding, this function is identical to lzma_index_stream_size().
+ * If multiple lzma_indexes have been combined, this includes also the headers
+ * of each separate Stream and the possible Stream Padding fields.
+ */
+extern LZMA_API(lzma_vli) lzma_index_file_size(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Get the uncompressed size of the file
+ */
+extern LZMA_API(lzma_vli) lzma_index_uncompressed_size(const lzma_index *i)
+ lzma_nothrow lzma_attr_pure;
+
+
+/**
+ * \brief Initialize an iterator
+ *
+ * \param iter Pointer to a lzma_index_iter structure
+ * \param i lzma_index to which the iterator will be associated
+ *
+ * This function associates the iterator with the given lzma_index, and calls
+ * lzma_index_iter_rewind() on the iterator.
+ *
+ * This function doesn't allocate any memory, thus there is no
+ * lzma_index_iter_end(). The iterator is valid as long as the
+ * associated lzma_index is valid, that is, until lzma_index_end() or
+ * using it as source in lzma_index_cat(). Specifically, lzma_index doesn't
+ * become invalid if new Blocks are added to it with lzma_index_append() or
+ * if it is used as the destination in lzma_index_cat().
+ *
+ * It is safe to make copies of an initialized lzma_index_iter, for example,
+ * to easily restart reading at some particular position.
+ */
+extern LZMA_API(void) lzma_index_iter_init(
+ lzma_index_iter *iter, const lzma_index *i) lzma_nothrow;
+
+
+/**
+ * \brief Rewind the iterator
+ *
+ * Rewind the iterator so that next call to lzma_index_iter_next() will
+ * return the first Block or Stream.
+ */
+extern LZMA_API(void) lzma_index_iter_rewind(lzma_index_iter *iter)
+ lzma_nothrow;
+
+
+/**
+ * \brief Get the next Block or Stream
+ *
+ * \param iter Iterator initialized with lzma_index_iter_init()
+ * \param mode Specify what kind of information the caller wants
+ * to get. See lzma_index_iter_mode for details.
+ *
+ * \return If next Block or Stream matching the mode was found, *iter
+ * is updated and this function returns false. If no Block or
+ * Stream matching the mode is found, *iter is not modified
+ * and this function returns true. If mode is set to an unknown
+ * value, *iter is not modified and this function returns true.
+ */
+extern LZMA_API(lzma_bool) lzma_index_iter_next(
+ lzma_index_iter *iter, lzma_index_iter_mode mode)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Locate a Block
+ *
+ * If it is possible to seek in the .xz file, it is possible to parse
+ * the Index field(s) and use lzma_index_iter_locate() to do random-access
+ * reading with granularity of Block size.
+ *
+ * \param iter Iterator that was earlier initialized with
+ * lzma_index_iter_init().
+ * \param target Uncompressed target offset which the caller would
+ * like to locate from the Stream
+ *
+ * If the target is smaller than the uncompressed size of the Stream (can be
+ * checked with lzma_index_uncompressed_size()):
+ * - Information about the Stream and Block containing the requested
+ * uncompressed offset is stored into *iter.
+ * - Internal state of the iterator is adjusted so that
+ * lzma_index_iter_next() can be used to read subsequent Blocks or Streams.
+ * - This function returns false.
+ *
+ * If target is greater than the uncompressed size of the Stream, *iter
+ * is not modified, and this function returns true.
+ */
+extern LZMA_API(lzma_bool) lzma_index_iter_locate(
+ lzma_index_iter *iter, lzma_vli target) lzma_nothrow;
+
+
+/**
+ * \brief Concatenate lzma_indexes
+ *
+ * Concatenating lzma_indexes is useful when doing random-access reading in
+ * multi-Stream .xz file, or when combining multiple Streams into single
+ * Stream.
+ *
+ * \param dest lzma_index after which src is appended
+ * \param src lzma_index to be appended after dest. If this
+ * function succeeds, the memory allocated for src
+ * is freed or moved to be part of dest, and all
+ * iterators pointing to src will become invalid.
+ * \param allocator Custom memory allocator; can be NULL to use
+ * malloc() and free().
+ *
+ * \return - LZMA_OK: lzma_indexes were concatenated successfully.
+ * src is now a dangling pointer.
+ * - LZMA_DATA_ERROR: *dest would grow too big.
+ * - LZMA_MEM_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_cat(
+ lzma_index *dest, lzma_index *src, lzma_allocator *allocator)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Duplicate lzma_index
+ *
+ * \return A copy of the lzma_index, or NULL if memory allocation failed.
+ */
+extern LZMA_API(lzma_index *) lzma_index_dup(
+ const lzma_index *i, lzma_allocator *allocator)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Initialize .xz Index encoder
+ *
+ * \param strm Pointer to properly prepared lzma_stream
+ * \param i Pointer to lzma_index which should be encoded.
+ *
+ * The only valid action value for lzma_code() is LZMA_RUN.
+ *
+ * \return - LZMA_OK: Initialization succeeded, continue with lzma_code().
+ * - LZMA_MEM_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_encoder(
+ lzma_stream *strm, const lzma_index *i)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Initialize .xz Index decoder
+ *
+ * \param strm Pointer to properly prepared lzma_stream
+ * \param i The decoded Index will be made available via
+ * this pointer. Initially this function will
+ * set *i to NULL (the old value is ignored). If
+ * decoding succeeds (lzma_code() returns
+ * LZMA_STREAM_END), *i will be set to point
+ * to a new lzma_index, which the application
+ * has to later free with lzma_index_end().
+ * \param memlimit How much memory the resulting lzma_index is
+ * allowed to require.
+ *
+ * The only valid action value for lzma_code() is LZMA_RUN.
+ *
+ * \return - LZMA_OK: Initialization succeeded, continue with lzma_code().
+ * - LZMA_MEM_ERROR
+ * - LZMA_MEMLIMIT_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_decoder(
+ lzma_stream *strm, lzma_index **i, uint64_t memlimit)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Single-call .xz Index encoder
+ *
+ * \param i lzma_index to be encoded
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * *out_pos is updated only if encoding succeeds.
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_BUF_ERROR: Output buffer is too small. Use
+ * lzma_index_size() to find out how much output
+ * space is needed.
+ * - LZMA_PROG_ERROR
+ *
+ * \note This function doesn't take allocator argument since all
+ * the internal data is allocated on stack.
+ */
+extern LZMA_API(lzma_ret) lzma_index_buffer_encode(const lzma_index *i,
+ uint8_t *out, size_t *out_pos, size_t out_size) lzma_nothrow;
+
+
+/**
+ * \brief Single-call .xz Index decoder
+ *
+ * \param i If decoding succeeds, *i will point to a new
+ * lzma_index, which the application has to
+ * later free with lzma_index_end(). If an error
+ * occurs, *i will be NULL. The old value of *i
+ * is always ignored and thus doesn't need to be
+ * initialized by the caller.
+ * \param memlimit Pointer to how much memory the resulting
+ * lzma_index is allowed to require. The value
+ * pointed by this pointer is modified if and only
+ * if LZMA_MEMLIMIT_ERROR is returned.
+ * \param allocator Pointer to lzma_allocator, or NULL to use malloc()
+ * \param in Beginning of the input buffer
+ * \param in_pos The next byte will be read from in[*in_pos].
+ * *in_pos is updated only if decoding succeeds.
+ * \param in_size Size of the input buffer; the first byte that
+ * won't be read is in[in_size].
+ *
+ * \return - LZMA_OK: Decoding was successful.
+ * - LZMA_MEM_ERROR
+ * - LZMA_MEMLIMIT_ERROR: Memory usage limit was reached.
+ * The minimum required memlimit value was stored to *memlimit.
+ * - LZMA_DATA_ERROR
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_buffer_decode(lzma_index **i,
+ uint64_t *memlimit, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size)
+ lzma_nothrow;
diff --git a/src/liblzma/api/lzma/index_hash.h b/src/liblzma/api/lzma/index_hash.h
new file mode 100644
index 0000000..94726e7
--- /dev/null
+++ b/src/liblzma/api/lzma/index_hash.h
@@ -0,0 +1,107 @@
+/**
+ * \file lzma/index_hash.h
+ * \brief Validates Index by using a hash function
+ *
+ * Hashing makes it possible to use constant amount of memory to validate
+ * Index of arbitrary size.
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+/**
+ * \brief Opaque data type to hold the Index hash
+ */
+typedef struct lzma_index_hash_s lzma_index_hash;
+
+
+/**
+ * \brief Allocate and initialize a new lzma_index_hash structure
+ *
+ * If index_hash is NULL, a new lzma_index_hash structure is allocated,
+ * initialized, and a pointer to it returned. If allocation fails, NULL
+ * is returned.
+ *
+ * If index_hash is non-NULL, it is reinitialized and the same pointer
+ * returned. In this case, return value cannot be NULL or a different
+ * pointer than the index_hash that was given as an argument.
+ */
+extern LZMA_API(lzma_index_hash *) lzma_index_hash_init(
+ lzma_index_hash *index_hash, lzma_allocator *allocator)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Deallocate lzma_index_hash structure
+ */
+extern LZMA_API(void) lzma_index_hash_end(
+ lzma_index_hash *index_hash, lzma_allocator *allocator)
+ lzma_nothrow;
+
+
+/**
+ * \brief Add a new Record to an Index hash
+ *
+ * \param index Pointer to a lzma_index_hash structure
+ * \param unpadded_size Unpadded Size of a Block
+ * \param uncompressed_size Uncompressed Size of a Block
+ *
+ * \return - LZMA_OK
+ * - LZMA_DATA_ERROR: Compressed or uncompressed size of the
+ * Stream or size of the Index field would grow too big.
+ * - LZMA_PROG_ERROR: Invalid arguments or this function is being
+ * used when lzma_index_hash_decode() has already been used.
+ */
+extern LZMA_API(lzma_ret) lzma_index_hash_append(lzma_index_hash *index_hash,
+ lzma_vli unpadded_size, lzma_vli uncompressed_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Decode and validate the Index field
+ *
+ * After telling the sizes of all Blocks with lzma_index_hash_append(),
+ * the actual Index field is decoded with this function. Specifically,
+ * once decoding of the Index field has been started, no more Records
+ * can be added using lzma_index_hash_append().
+ *
+ * This function doesn't use lzma_stream structure to pass the input data.
+ * Instead, the input buffer is specified using three arguments. This is
+ * because it matches better the internal APIs of liblzma.
+ *
+ * \param index_hash Pointer to a lzma_index_hash structure
+ * \param in Pointer to the beginning of the input buffer
+ * \param in_pos in[*in_pos] is the next byte to process
+ * \param in_size in[in_size] is the first byte not to process
+ *
+ * \return - LZMA_OK: So far good, but more input is needed.
+ * - LZMA_STREAM_END: Index decoded successfully and it matches
+ * the Records given with lzma_index_hash_append().
+ * - LZMA_DATA_ERROR: Index is corrupt or doesn't match the
+ * information given with lzma_index_hash_append().
+ * - LZMA_BUF_ERROR: Cannot progress because *in_pos >= in_size.
+ * - LZMA_PROG_ERROR
+ */
+extern LZMA_API(lzma_ret) lzma_index_hash_decode(lzma_index_hash *index_hash,
+ const uint8_t *in, size_t *in_pos, size_t in_size)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Get the size of the Index field as bytes
+ *
+ * This is needed to verify the Backward Size field in the Stream Footer.
+ */
+extern LZMA_API(lzma_vli) lzma_index_hash_size(
+ const lzma_index_hash *index_hash)
+ lzma_nothrow lzma_attr_pure;
diff --git a/src/liblzma/api/lzma/lzma.h b/src/liblzma/api/lzma/lzma.h
new file mode 100644
index 0000000..c17736d
--- /dev/null
+++ b/src/liblzma/api/lzma/lzma.h
@@ -0,0 +1,397 @@
+/**
+ * \file lzma/lzma.h
+ * \brief LZMA1 and LZMA2 filters
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief LZMA1 Filter ID
+ *
+ * LZMA1 is the very same thing as what was called just LZMA in LZMA Utils,
+ * 7-Zip, and LZMA SDK. It's called LZMA1 here to prevent developers from
+ * accidentally using LZMA when they actually want LZMA2.
+ *
+ * LZMA1 shouldn't be used for new applications unless you _really_ know
+ * what you are doing. LZMA2 is almost always a better choice.
+ */
+#define LZMA_FILTER_LZMA1 LZMA_VLI_C(0x4000000000000001)
+
+/**
+ * \brief LZMA2 Filter ID
+ *
+ * Usually you want this instead of LZMA1. Compared to LZMA1, LZMA2 adds
+ * support for LZMA_SYNC_FLUSH, uncompressed chunks (smaller expansion
+ * when trying to compress uncompressible data), possibility to change
+ * lc/lp/pb in the middle of encoding, and some other internal improvements.
+ */
+#define LZMA_FILTER_LZMA2 LZMA_VLI_C(0x21)
+
+
+/**
+ * \brief Match finders
+ *
+ * Match finder has major effect on both speed and compression ratio.
+ * Usually hash chains are faster than binary trees.
+ *
+ * The memory usage formulas are only rough estimates, which are closest to
+ * reality when dict_size is a power of two. The formulas are more complex
+ * in reality, and can also change a little between liblzma versions. Use
+ * lzma_memusage_encoder() to get more accurate estimate of memory usage.
+ */
+typedef enum {
+ LZMA_MF_HC3 = 0x03,
+ /**<
+ * \brief Hash Chain with 2- and 3-byte hashing
+ *
+ * Minimum nice_len: 3
+ *
+ * Memory usage:
+ * - dict_size <= 16 MiB: dict_size * 7.5
+ * - dict_size > 16 MiB: dict_size * 5.5 + 64 MiB
+ */
+
+ LZMA_MF_HC4 = 0x04,
+ /**<
+ * \brief Hash Chain with 2-, 3-, and 4-byte hashing
+ *
+ * Minimum nice_len: 4
+ *
+ * Memory usage: dict_size * 7.5
+ */
+
+ LZMA_MF_BT2 = 0x12,
+ /**<
+ * \brief Binary Tree with 2-byte hashing
+ *
+ * Minimum nice_len: 2
+ *
+ * Memory usage: dict_size * 9.5
+ */
+
+ LZMA_MF_BT3 = 0x13,
+ /**<
+ * \brief Binary Tree with 2- and 3-byte hashing
+ *
+ * Minimum nice_len: 3
+ *
+ * Memory usage:
+ * - dict_size <= 16 MiB: dict_size * 11.5
+ * - dict_size > 16 MiB: dict_size * 9.5 + 64 MiB
+ */
+
+ LZMA_MF_BT4 = 0x14
+ /**<
+ * \brief Binary Tree with 2-, 3-, and 4-byte hashing
+ *
+ * Minimum nice_len: 4
+ *
+ * Memory usage: dict_size * 11.5
+ */
+} lzma_match_finder;
+
+
+/**
+ * \brief Test if given match finder is supported
+ *
+ * Return true if the given match finder is supported by this liblzma build.
+ * Otherwise false is returned. It is safe to call this with a value that
+ * isn't listed in lzma_match_finder enumeration; the return value will be
+ * false.
+ *
+ * There is no way to list which match finders are available in this
+ * particular liblzma version and build. It would be useless, because
+ * a new match finder, which the application developer wasn't aware,
+ * could require giving additional options to the encoder that the older
+ * match finders don't need.
+ */
+extern LZMA_API(lzma_bool) lzma_mf_is_supported(lzma_match_finder match_finder)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Compression modes
+ *
+ * This selects the function used to analyze the data produced by the match
+ * finder.
+ */
+typedef enum {
+ LZMA_MODE_FAST = 1,
+ /**<
+ * \brief Fast compression
+ *
+ * Fast mode is usually at its best when combined with
+ * a hash chain match finder.
+ */
+
+ LZMA_MODE_NORMAL = 2
+ /**<
+ * \brief Normal compression
+ *
+ * This is usually notably slower than fast mode. Use this
+ * together with binary tree match finders to expose the
+ * full potential of the LZMA1 or LZMA2 encoder.
+ */
+} lzma_mode;
+
+
+/**
+ * \brief Test if given compression mode is supported
+ *
+ * Return true if the given compression mode is supported by this liblzma
+ * build. Otherwise false is returned. It is safe to call this with a value
+ * that isn't listed in lzma_mode enumeration; the return value will be false.
+ *
+ * There is no way to list which modes are available in this particular
+ * liblzma version and build. It would be useless, because a new compression
+ * mode, which the application developer wasn't aware, could require giving
+ * additional options to the encoder that the older modes don't need.
+ */
+extern LZMA_API(lzma_bool) lzma_mode_is_supported(lzma_mode mode)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Options specific to the LZMA1 and LZMA2 filters
+ *
+ * Since LZMA1 and LZMA2 share most of the code, it's simplest to share
+ * the options structure too. For encoding, all but the reserved variables
+ * need to be initialized unless specifically mentioned otherwise.
+ *
+ * For raw decoding, both LZMA1 and LZMA2 need dict_size, preset_dict, and
+ * preset_dict_size (if preset_dict != NULL). LZMA1 needs also lc, lp, and pb.
+ */
+typedef struct {
+ /**
+ * \brief Dictionary size in bytes
+ *
+ * Dictionary size indicates how many bytes of the recently processed
+ * uncompressed data is kept in memory. One method to reduce size of
+ * the uncompressed data is to store distance-length pairs, which
+ * indicate what data to repeat from the dictionary buffer. Thus,
+ * the bigger the dictionary, the better the compression ratio
+ * usually is.
+ *
+ * Maximum size of the dictionary depends on multiple things:
+ * - Memory usage limit
+ * - Available address space (not a problem on 64-bit systems)
+ * - Selected match finder (encoder only)
+ *
+ * Currently the maximum dictionary size for encoding is 1.5 GiB
+ * (i.e. (UINT32_C(1) << 30) + (UINT32_C(1) << 29)) even on 64-bit
+ * systems for certain match finder implementation reasons. In the
+ * future, there may be match finders that support bigger
+ * dictionaries.
+ *
+ * Decoder already supports dictionaries up to 4 GiB - 1 B (i.e.
+ * UINT32_MAX), so increasing the maximum dictionary size of the
+ * encoder won't cause problems for old decoders.
+ *
+ * Because extremely small dictionaries sizes would have unneeded
+ * overhead in the decoder, the minimum dictionary size is 4096 bytes.
+ *
+ * \note When decoding, too big dictionary does no other harm
+ * than wasting memory.
+ */
+ uint32_t dict_size;
+# define LZMA_DICT_SIZE_MIN UINT32_C(4096)
+# define LZMA_DICT_SIZE_DEFAULT (UINT32_C(1) << 23)
+
+ /**
+ * \brief Pointer to an initial dictionary
+ *
+ * It is possible to initialize the LZ77 history window using
+ * a preset dictionary. It is useful when compressing many
+ * similar, relatively small chunks of data independently from
+ * each other. The preset dictionary should contain typical
+ * strings that occur in the files being compressed. The most
+ * probable strings should be near the end of the preset dictionary.
+ *
+ * This feature should be used only in special situations. For
+ * now, it works correctly only with raw encoding and decoding.
+ * Currently none of the container formats supported by
+ * liblzma allow preset dictionary when decoding, thus if
+ * you create a .xz or .lzma file with preset dictionary, it
+ * cannot be decoded with the regular decoder functions. In the
+ * future, the .xz format will likely get support for preset
+ * dictionary though.
+ */
+ const uint8_t *preset_dict;
+
+ /**
+ * \brief Size of the preset dictionary
+ *
+ * Specifies the size of the preset dictionary. If the size is
+ * bigger than dict_size, only the last dict_size bytes are
+ * processed.
+ *
+ * This variable is read only when preset_dict is not NULL.
+ * If preset_dict is not NULL but preset_dict_size is zero,
+ * no preset dictionary is used (identical to only setting
+ * preset_dict to NULL).
+ */
+ uint32_t preset_dict_size;
+
+ /**
+ * \brief Number of literal context bits
+ *
+ * How many of the highest bits of the previous uncompressed
+ * eight-bit byte (also known as `literal') are taken into
+ * account when predicting the bits of the next literal.
+ *
+ * \todo Example
+ *
+ * There is a limit that applies to literal context bits and literal
+ * position bits together: lc + lp <= 4. Without this limit the
+ * decoding could become very slow, which could have security related
+ * results in some cases like email servers doing virus scanning.
+ * This limit also simplifies the internal implementation in liblzma.
+ *
+ * There may be LZMA1 streams that have lc + lp > 4 (maximum possible
+ * lc would be 8). It is not possible to decode such streams with
+ * liblzma.
+ */
+ uint32_t lc;
+# define LZMA_LCLP_MIN 0
+# define LZMA_LCLP_MAX 4
+# define LZMA_LC_DEFAULT 3
+
+ /**
+ * \brief Number of literal position bits
+ *
+ * How many of the lowest bits of the current position (number
+ * of bytes from the beginning of the uncompressed data) in the
+ * uncompressed data is taken into account when predicting the
+ * bits of the next literal (a single eight-bit byte).
+ *
+ * \todo Example
+ */
+ uint32_t lp;
+# define LZMA_LP_DEFAULT 0
+
+ /**
+ * \brief Number of position bits
+ *
+ * How many of the lowest bits of the current position in the
+ * uncompressed data is taken into account when estimating
+ * probabilities of matches. A match is a sequence of bytes for
+ * which a matching sequence is found from the dictionary and
+ * thus can be stored as distance-length pair.
+ *
+ * Example: If most of the matches occur at byte positions of
+ * 8 * n + 3, that is, 3, 11, 19, ... set pb to 3, because 2**3 == 8.
+ */
+ uint32_t pb;
+# define LZMA_PB_MIN 0
+# define LZMA_PB_MAX 4
+# define LZMA_PB_DEFAULT 2
+
+ /** Compression mode */
+ lzma_mode mode;
+
+ /**
+ * \brief Nice length of a match
+ *
+ * This determines how many bytes the encoder compares from the match
+ * candidates when looking for the best match. Once a match of at
+ * least nice_len bytes long is found, the encoder stops looking for
+ * better candidates and encodes the match. (Naturally, if the found
+ * match is actually longer than nice_len, the actual length is
+ * encoded; it's not truncated to nice_len.)
+ *
+ * Bigger values usually increase the compression ratio and
+ * compression time. For most files, 32 to 128 is a good value,
+ * which gives very good compression ratio at good speed.
+ *
+ * The exact minimum value depends on the match finder. The maximum
+ * is 273, which is the maximum length of a match that LZMA1 and
+ * LZMA2 can encode.
+ */
+ uint32_t nice_len;
+
+ /** Match finder ID */
+ lzma_match_finder mf;
+
+ /**
+ * \brief Maximum search depth in the match finder
+ *
+ * For every input byte, match finder searches through the hash chain
+ * or binary tree in a loop, each iteration going one step deeper in
+ * the chain or tree. The searching stops if
+ * - a match of at least nice_len bytes long is found;
+ * - all match candidates from the hash chain or binary tree have
+ * been checked; or
+ * - maximum search depth is reached.
+ *
+ * Maximum search depth is needed to prevent the match finder from
+ * wasting too much time in case there are lots of short match
+ * candidates. On the other hand, stopping the search before all
+ * candidates have been checked can reduce compression ratio.
+ *
+ * Setting depth to zero tells liblzma to use an automatic default
+ * value, that depends on the selected match finder and nice_len.
+ * The default is in the range [10, 200] or so (it may vary between
+ * liblzma versions).
+ *
+ * Using a bigger depth value than the default can increase
+ * compression ratio in some cases. There is no strict maximum value,
+ * but high values (thousands or millions) should be used with care:
+ * the encoder could remain fast enough with typical input, but
+ * malicious input could cause the match finder to slow down
+ * dramatically, possibly creating a denial of service attack.
+ */
+ uint32_t depth;
+
+ /*
+ * Reserved space to allow possible future extensions without
+ * breaking the ABI. You should not touch these, because the names
+ * of these variables may change. These are and will never be used
+ * with the currently supported options, so it is safe to leave these
+ * uninitialized.
+ */
+ void *reserved_ptr1;
+ void *reserved_ptr2;
+ uint32_t reserved_int1;
+ uint32_t reserved_int2;
+ uint32_t reserved_int3;
+ uint32_t reserved_int4;
+ uint32_t reserved_int5;
+ uint32_t reserved_int6;
+ uint32_t reserved_int7;
+ uint32_t reserved_int8;
+ lzma_reserved_enum reserved_enum1;
+ lzma_reserved_enum reserved_enum2;
+ lzma_reserved_enum reserved_enum3;
+ lzma_reserved_enum reserved_enum4;
+
+} lzma_options_lzma;
+
+
+/**
+ * \brief Set a compression preset to lzma_options_lzma structure
+ *
+ * 0 is the fastest and 9 is the slowest. These match the switches -0 .. -9
+ * of the xz command line tool. In addition, it is possible to bitwise-or
+ * flags to the preset. Currently only LZMA_PRESET_EXTREME is supported.
+ * The flags are defined in container.h, because the flags are used also
+ * with lzma_easy_encoder().
+ *
+ * The preset values are subject to changes between liblzma versions.
+ *
+ * This function is available only if LZMA1 or LZMA2 encoder has been enabled
+ * when building liblzma.
+ */
+extern LZMA_API(lzma_bool) lzma_lzma_preset(
+ lzma_options_lzma *options, uint32_t preset) lzma_nothrow;
diff --git a/src/liblzma/api/lzma/stream_flags.h b/src/liblzma/api/lzma/stream_flags.h
new file mode 100644
index 0000000..d255bdd
--- /dev/null
+++ b/src/liblzma/api/lzma/stream_flags.h
@@ -0,0 +1,227 @@
+/**
+ * \file lzma/stream_flags.h
+ * \brief .xz Stream Header and Stream Footer encoder and decoder
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Size of Stream Header and Stream Footer
+ *
+ * Stream Header and Stream Footer have the same size and they are not
+ * going to change even if a newer version of the .xz file format is
+ * developed in future.
+ */
+#define LZMA_STREAM_HEADER_SIZE 12
+
+
+/**
+ * \brief Options for encoding/decoding Stream Header and Stream Footer
+ */
+typedef struct {
+ /**
+ * \brief Stream Flags format version
+ *
+ * To prevent API and ABI breakages if new features are needed in
+ * Stream Header or Stream Footer, a version number is used to
+ * indicate which fields in this structure are in use. For now,
+ * version must always be zero. With non-zero version, the
+ * lzma_stream_header_encode() and lzma_stream_footer_encode()
+ * will return LZMA_OPTIONS_ERROR.
+ *
+ * lzma_stream_header_decode() and lzma_stream_footer_decode()
+ * will always set this to the lowest value that supports all the
+ * features indicated by the Stream Flags field. The application
+ * must check that the version number set by the decoding functions
+ * is supported by the application. Otherwise it is possible that
+ * the application will decode the Stream incorrectly.
+ */
+ uint32_t version;
+
+ /**
+ * \brief Backward Size
+ *
+ * Backward Size must be a multiple of four bytes. In this Stream
+ * format version, Backward Size is the size of the Index field.
+ *
+ * Backward Size isn't actually part of the Stream Flags field, but
+ * it is convenient to include in this structure anyway. Backward
+ * Size is present only in the Stream Footer. There is no need to
+ * initialize backward_size when encoding Stream Header.
+ *
+ * lzma_stream_header_decode() always sets backward_size to
+ * LZMA_VLI_UNKNOWN so that it is convenient to use
+ * lzma_stream_flags_compare() when both Stream Header and Stream
+ * Footer have been decoded.
+ */
+ lzma_vli backward_size;
+# define LZMA_BACKWARD_SIZE_MIN 4
+# define LZMA_BACKWARD_SIZE_MAX (LZMA_VLI_C(1) << 34)
+
+ /**
+ * \brief Check ID
+ *
+ * This indicates the type of the integrity check calculated from
+ * uncompressed data.
+ */
+ lzma_check check;
+
+ /*
+ * Reserved space to allow possible future extensions without
+ * breaking the ABI. You should not touch these, because the
+ * names of these variables may change.
+ *
+ * (We will never be able to use all of these since Stream Flags
+ * is just two bytes plus Backward Size of four bytes. But it's
+ * nice to have the proper types when they are needed.)
+ */
+ lzma_reserved_enum reserved_enum1;
+ lzma_reserved_enum reserved_enum2;
+ lzma_reserved_enum reserved_enum3;
+ lzma_reserved_enum reserved_enum4;
+ lzma_reserved_enum reserved_enum5;
+ lzma_reserved_enum reserved_enum6;
+ lzma_bool reserved_bool1;
+ lzma_bool reserved_bool2;
+ lzma_bool reserved_bool3;
+ lzma_bool reserved_bool4;
+ lzma_bool reserved_bool5;
+ lzma_bool reserved_bool6;
+ lzma_bool reserved_bool7;
+ lzma_bool reserved_bool8;
+ uint32_t reserved_int1;
+ uint32_t reserved_int2;
+ uint32_t reserved_int3;
+ uint32_t reserved_int4;
+
+} lzma_stream_flags;
+
+
+/**
+ * \brief Encode Stream Header
+ *
+ * \param options Stream Header options to be encoded.
+ * options->backward_size is ignored and doesn't
+ * need to be initialized.
+ * \param out Beginning of the output buffer of
+ * LZMA_STREAM_HEADER_SIZE bytes.
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_OPTIONS_ERROR: options->version is not supported by
+ * this liblzma version.
+ * - LZMA_PROG_ERROR: Invalid options.
+ */
+extern LZMA_API(lzma_ret) lzma_stream_header_encode(
+ const lzma_stream_flags *options, uint8_t *out)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Encode Stream Footer
+ *
+ * \param options Stream Footer options to be encoded.
+ * \param out Beginning of the output buffer of
+ * LZMA_STREAM_HEADER_SIZE bytes.
+ *
+ * \return - LZMA_OK: Encoding was successful.
+ * - LZMA_OPTIONS_ERROR: options->version is not supported by
+ * this liblzma version.
+ * - LZMA_PROG_ERROR: Invalid options.
+ */
+extern LZMA_API(lzma_ret) lzma_stream_footer_encode(
+ const lzma_stream_flags *options, uint8_t *out)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Decode Stream Header
+ *
+ * \param options Stream Header options to be encoded.
+ * \param in Beginning of the input buffer of
+ * LZMA_STREAM_HEADER_SIZE bytes.
+ *
+ * options->backward_size is always set to LZMA_VLI_UNKNOWN. This is to
+ * help comparing Stream Flags from Stream Header and Stream Footer with
+ * lzma_stream_flags_compare().
+ *
+ * \return - LZMA_OK: Decoding was successful.
+ * - LZMA_FORMAT_ERROR: Magic bytes don't match, thus the given
+ * buffer cannot be Stream Header.
+ * - LZMA_DATA_ERROR: CRC32 doesn't match, thus the header
+ * is corrupt.
+ * - LZMA_OPTIONS_ERROR: Unsupported options are present
+ * in the header.
+ *
+ * \note When decoding .xz files that contain multiple Streams, it may
+ * make sense to print "file format not recognized" only if
+ * decoding of the Stream Header of the _first_ Stream gives
+ * LZMA_FORMAT_ERROR. If non-first Stream Header gives
+ * LZMA_FORMAT_ERROR, the message used for LZMA_DATA_ERROR is
+ * probably more appropriate.
+ *
+ * For example, Stream decoder in liblzma uses LZMA_DATA_ERROR if
+ * LZMA_FORMAT_ERROR is returned by lzma_stream_header_decode()
+ * when decoding non-first Stream.
+ */
+extern LZMA_API(lzma_ret) lzma_stream_header_decode(
+ lzma_stream_flags *options, const uint8_t *in)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Decode Stream Footer
+ *
+ * \param options Stream Header options to be encoded.
+ * \param in Beginning of the input buffer of
+ * LZMA_STREAM_HEADER_SIZE bytes.
+ *
+ * \return - LZMA_OK: Decoding was successful.
+ * - LZMA_FORMAT_ERROR: Magic bytes don't match, thus the given
+ * buffer cannot be Stream Footer.
+ * - LZMA_DATA_ERROR: CRC32 doesn't match, thus the Stream Footer
+ * is corrupt.
+ * - LZMA_OPTIONS_ERROR: Unsupported options are present
+ * in Stream Footer.
+ *
+ * \note If Stream Header was already decoded successfully, but
+ * decoding Stream Footer returns LZMA_FORMAT_ERROR, the
+ * application should probably report some other error message
+ * than "file format not recognized", since the file more likely
+ * is corrupt (possibly truncated). Stream decoder in liblzma
+ * uses LZMA_DATA_ERROR in this situation.
+ */
+extern LZMA_API(lzma_ret) lzma_stream_footer_decode(
+ lzma_stream_flags *options, const uint8_t *in)
+ lzma_nothrow lzma_attr_warn_unused_result;
+
+
+/**
+ * \brief Compare two lzma_stream_flags structures
+ *
+ * backward_size values are compared only if both are not
+ * LZMA_VLI_UNKNOWN.
+ *
+ * \return - LZMA_OK: Both are equal. If either had backward_size set
+ * to LZMA_VLI_UNKNOWN, backward_size values were not
+ * compared or validated.
+ * - LZMA_DATA_ERROR: The structures differ.
+ * - LZMA_OPTIONS_ERROR: version in either structure is greater
+ * than the maximum supported version (currently zero).
+ * - LZMA_PROG_ERROR: Invalid value, e.g. invalid check or
+ * backward_size.
+ */
+extern LZMA_API(lzma_ret) lzma_stream_flags_compare(
+ const lzma_stream_flags *a, const lzma_stream_flags *b)
+ lzma_nothrow lzma_attr_pure;
diff --git a/src/liblzma/api/lzma/version.h b/src/liblzma/api/lzma/version.h
new file mode 100644
index 0000000..0f7c2d3
--- /dev/null
+++ b/src/liblzma/api/lzma/version.h
@@ -0,0 +1,121 @@
+/**
+ * \file lzma/version.h
+ * \brief Version number
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/*
+ * Version number split into components
+ */
+#define LZMA_VERSION_MAJOR 4
+#define LZMA_VERSION_MINOR 999
+#define LZMA_VERSION_PATCH 9
+#define LZMA_VERSION_STABILITY LZMA_VERSION_STABILITY_BETA
+
+#ifndef LZMA_VERSION_COMMIT
+# define LZMA_VERSION_COMMIT ""
+#endif
+
+
+/*
+ * Map symbolic stability levels to integers.
+ */
+#define LZMA_VERSION_STABILITY_ALPHA 0
+#define LZMA_VERSION_STABILITY_BETA 1
+#define LZMA_VERSION_STABILITY_STABLE 2
+
+
+/**
+ * \brief Compile-time version number
+ *
+ * The version number is of format xyyyzzzs where
+ * - x = major
+ * - yyy = minor
+ * - zzz = revision
+ * - s indicates stability: 0 = alpha, 1 = beta, 2 = stable
+ *
+ * The same xyyyzzz triplet is never reused with different stability levels.
+ * For example, if 5.1.0alpha has been released, there will never be 5.1.0beta
+ * or 5.1.0 stable.
+ *
+ * \note The version number of liblzma has nothing to with
+ * the version number of Igor Pavlov's LZMA SDK.
+ */
+#define LZMA_VERSION (LZMA_VERSION_MAJOR * UINT32_C(10000000) \
+ + LZMA_VERSION_MINOR * UINT32_C(10000) \
+ + LZMA_VERSION_PATCH * UINT32_C(10) \
+ + LZMA_VERSION_STABILITY)
+
+
+/*
+ * Macros to construct the compile-time version string
+ */
+#if LZMA_VERSION_STABILITY == LZMA_VERSION_STABILITY_ALPHA
+# define LZMA_VERSION_STABILITY_STRING "alpha"
+#elif LZMA_VERSION_STABILITY == LZMA_VERSION_STABILITY_BETA
+# define LZMA_VERSION_STABILITY_STRING "beta"
+#elif LZMA_VERSION_STABILITY == LZMA_VERSION_STABILITY_STABLE
+# define LZMA_VERSION_STABILITY_STRING ""
+#else
+# error Incorrect LZMA_VERSION_STABILITY
+#endif
+
+#define LZMA_VERSION_STRING_C_(major, minor, patch, stability, commit) \
+ #major "." #minor "." #patch stability commit
+
+#define LZMA_VERSION_STRING_C(major, minor, patch, stability, commit) \
+ LZMA_VERSION_STRING_C_(major, minor, patch, stability, commit)
+
+
+/**
+ * \brief Compile-time version as a string
+ *
+ * This can be for example "4.999.5alpha", "4.999.8beta", or "5.0.0" (stable
+ * versions don't have any "stable" suffix). In future, a snapshot built
+ * from source code repository may include an additional suffix, for example
+ * "4.999.8beta-21-g1d92". The commit ID won't be available in numeric form
+ * in LZMA_VERSION macro.
+ */
+#define LZMA_VERSION_STRING LZMA_VERSION_STRING_C( \
+ LZMA_VERSION_MAJOR, LZMA_VERSION_MINOR, \
+ LZMA_VERSION_PATCH, LZMA_VERSION_STABILITY_STRING, \
+ LZMA_VERSION_COMMIT)
+
+
+/* #ifndef is needed for use with windres (MinGW or Cygwin). */
+#ifndef LZMA_H_INTERNAL_RC
+
+/**
+ * \brief Run-time version number as an integer
+ *
+ * Return the value of LZMA_VERSION macro at the compile time of liblzma.
+ * This allows the application to compare if it was built against the same,
+ * older, or newer version of liblzma that is currently running.
+ */
+extern LZMA_API(uint32_t) lzma_version_number(void)
+ lzma_nothrow lzma_attr_const;
+
+
+/**
+ * \brief Run-time version as a string
+ *
+ * This function may be useful if you want to display which version of
+ * liblzma your application is currently using.
+ */
+extern LZMA_API(const char *) lzma_version_string(void)
+ lzma_nothrow lzma_attr_const;
+
+#endif
diff --git a/src/liblzma/api/lzma/vli.h b/src/liblzma/api/lzma/vli.h
new file mode 100644
index 0000000..57e3d38
--- /dev/null
+++ b/src/liblzma/api/lzma/vli.h
@@ -0,0 +1,167 @@
+/**
+ * \file lzma/vli.h
+ * \brief Variable-length integer handling
+ *
+ * In the .xz format, most integers are encoded in a variable-length
+ * representation, which is sometimes called little endian base-128 encoding.
+ * This saves space when smaller values are more likely than bigger values.
+ *
+ * The encoding scheme encodes seven bits to every byte, using minimum
+ * number of bytes required to represent the given value. Encodings that use
+ * non-minimum number of bytes are invalid, thus every integer has exactly
+ * one encoded representation. The maximum number of bits in a VLI is 63,
+ * thus the vli argument must be less than or equal to UINT64_MAX / 2. You
+ * should use LZMA_VLI_MAX for clarity.
+ */
+
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * See ../lzma.h for information about liblzma as a whole.
+ */
+
+#ifndef LZMA_H_INTERNAL
+# error Never include this file directly. Use <lzma.h> instead.
+#endif
+
+
+/**
+ * \brief Maximum supported value of variable-length integer
+ */
+#define LZMA_VLI_MAX (UINT64_MAX / 2)
+
+/**
+ * \brief VLI value to denote that the value is unknown
+ */
+#define LZMA_VLI_UNKNOWN UINT64_MAX
+
+/**
+ * \brief Maximum supported length of variable length integers
+ */
+#define LZMA_VLI_BYTES_MAX 9
+
+
+/**
+ * \brief VLI constant suffix
+ */
+#define LZMA_VLI_C(n) UINT64_C(n)
+
+
+/**
+ * \brief Variable-length integer type
+ *
+ * This will always be unsigned integer. Valid VLI values are in the range
+ * [0, LZMA_VLI_MAX]. Unknown value is indicated with LZMA_VLI_UNKNOWN,
+ * which is the maximum value of the underlaying integer type.
+ *
+ * In future, even if lzma_vli is defined to be something other than uint64_t,
+ * it is guaranteed that 2 * LZMA_VLI_MAX will not overflow lzma_vli.
+ * This simplifies integer overflow detection.
+ */
+typedef uint64_t lzma_vli;
+
+
+/**
+ * \brief Simple macro to validate variable-length integer
+ *
+ * This is useful to test that application has given acceptable values
+ * for example in the uncompressed_size and compressed_size variables.
+ *
+ * \return True if the integer is representable as VLI or if it
+ * indicates unknown value.
+ */
+#define lzma_vli_is_valid(vli) \
+ ((vli) <= LZMA_VLI_MAX || (vli) == LZMA_VLI_UNKNOWN)
+
+
+/**
+ * \brief Encode a variable-length integer
+ *
+ * This function has two modes: single-call and multi-call. Single-call mode
+ * encodes the whole integer at once; it is an error if the output buffer is
+ * too small. Multi-call mode saves the position in *vli_pos, and thus it is
+ * possible to continue encoding if the buffer becomes full before the whole
+ * integer has been encoded.
+ *
+ * \param vli Integer to be encoded
+ * \param vli_pos How many VLI-encoded bytes have already been written
+ * out. When starting to encode a new integer, *vli_pos
+ * must be set to zero. To use single-call encoding,
+ * set vli_pos to NULL.
+ * \param out Beginning of the output buffer
+ * \param out_pos The next byte will be written to out[*out_pos].
+ * \param out_size Size of the out buffer; the first byte into
+ * which no data is written to is out[out_size].
+ *
+ * \return Slightly different return values are used in multi-call and
+ * single-call modes.
+ *
+ * Single-call (vli_pos == NULL):
+ * - LZMA_OK: Integer successfully encoded.
+ * - LZMA_PROG_ERROR: Arguments are not sane. This can be due
+ * to too little output space; single-call mode doesn't use
+ * LZMA_BUF_ERROR, since the application should have checked
+ * the encoded size with lzma_vli_size().
+ *
+ * Multi-call (vli_pos != NULL):
+ * - LZMA_OK: So far all OK, but the integer is not
+ * completely written out yet.
+ * - LZMA_STREAM_END: Integer successfully encoded.
+ * - LZMA_BUF_ERROR: No output space was provided.
+ * - LZMA_PROG_ERROR: Arguments are not sane.
+ */
+extern LZMA_API(lzma_ret) lzma_vli_encode(lzma_vli vli, size_t *vli_pos,
+ uint8_t *out, size_t *out_pos, size_t out_size) lzma_nothrow;
+
+
+/**
+ * \brief Decode a variable-length integer
+ *
+ * Like lzma_vli_encode(), this function has single-call and multi-call modes.
+ *
+ * \param vli Pointer to decoded integer. The decoder will
+ * initialize it to zero when *vli_pos == 0, so
+ * application isn't required to initialize *vli.
+ * \param vli_pos How many bytes have already been decoded. When
+ * starting to decode a new integer, *vli_pos must
+ * be initialized to zero. To use single-call decoding,
+ * set this to NULL.
+ * \param in Beginning of the input buffer
+ * \param in_pos The next byte will be read from in[*in_pos].
+ * \param in_size Size of the input buffer; the first byte that
+ * won't be read is in[in_size].
+ *
+ * \return Slightly different return values are used in multi-call and
+ * single-call modes.
+ *
+ * Single-call (vli_pos == NULL):
+ * - LZMA_OK: Integer successfully decoded.
+ * - LZMA_DATA_ERROR: Integer is corrupt. This includes hitting
+ * the end of the input buffer before the whole integer was
+ * decoded; providing no input at all will use LZMA_DATA_ERROR.
+ * - LZMA_PROG_ERROR: Arguments are not sane.
+ *
+ * Multi-call (vli_pos != NULL):
+ * - LZMA_OK: So far all OK, but the integer is not
+ * completely decoded yet.
+ * - LZMA_STREAM_END: Integer successfully decoded.
+ * - LZMA_DATA_ERROR: Integer is corrupt.
+ * - LZMA_BUF_ERROR: No input was provided.
+ * - LZMA_PROG_ERROR: Arguments are not sane.
+ */
+extern LZMA_API(lzma_ret) lzma_vli_decode(lzma_vli *vli, size_t *vli_pos,
+ const uint8_t *in, size_t *in_pos, size_t in_size)
+ lzma_nothrow;
+
+
+/**
+ * \brief Get the number of bytes required to encode a VLI
+ *
+ * \return Number of bytes on success (1-9). If vli isn't valid,
+ * zero is returned.
+ */
+extern LZMA_API(uint32_t) lzma_vli_size(lzma_vli vli)
+ lzma_nothrow lzma_attr_pure;
diff --git a/src/liblzma/check/Makefile.inc b/src/liblzma/check/Makefile.inc
new file mode 100644
index 0000000..e4067a9
--- /dev/null
+++ b/src/liblzma/check/Makefile.inc
@@ -0,0 +1,51 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+EXTRA_DIST += \
+ check/crc32_tablegen.c \
+ check/crc64_tablegen.c
+
+liblzma_la_SOURCES += \
+ check/check.c \
+ check/check.h \
+ check/crc_macros.h
+
+if COND_CHECK_CRC32
+if COND_SMALL
+liblzma_la_SOURCES += check/crc32_small.c
+else
+liblzma_la_SOURCES += \
+ check/crc32_table.c \
+ check/crc32_table_le.h \
+ check/crc32_table_be.h
+if COND_ASM_X86
+liblzma_la_SOURCES += check/crc32_x86.S
+else
+liblzma_la_SOURCES += check/crc32_fast.c
+endif
+endif
+endif
+
+if COND_CHECK_CRC64
+if COND_SMALL
+liblzma_la_SOURCES += check/crc64_small.c
+else
+liblzma_la_SOURCES += \
+ check/crc64_table.c \
+ check/crc64_table_le.h \
+ check/crc64_table_be.h
+if COND_ASM_X86
+liblzma_la_SOURCES += check/crc64_x86.S
+else
+liblzma_la_SOURCES += check/crc64_fast.c
+endif
+endif
+endif
+
+if COND_CHECK_SHA256
+liblzma_la_SOURCES += check/sha256.c
+endif
diff --git a/src/liblzma/check/check.c b/src/liblzma/check/check.c
new file mode 100644
index 0000000..428ddae
--- /dev/null
+++ b/src/liblzma/check/check.c
@@ -0,0 +1,174 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file check.c
+/// \brief Single API to access different integrity checks
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "check.h"
+
+
+extern LZMA_API(lzma_bool)
+lzma_check_is_supported(lzma_check type)
+{
+ if ((unsigned int)(type) > LZMA_CHECK_ID_MAX)
+ return false;
+
+ static const lzma_bool available_checks[LZMA_CHECK_ID_MAX + 1] = {
+ true, // LZMA_CHECK_NONE
+
+#ifdef HAVE_CHECK_CRC32
+ true,
+#else
+ false,
+#endif
+
+ false, // Reserved
+ false, // Reserved
+
+#ifdef HAVE_CHECK_CRC64
+ true,
+#else
+ false,
+#endif
+
+ false, // Reserved
+ false, // Reserved
+ false, // Reserved
+ false, // Reserved
+ false, // Reserved
+
+#ifdef HAVE_CHECK_SHA256
+ true,
+#else
+ false,
+#endif
+
+ false, // Reserved
+ false, // Reserved
+ false, // Reserved
+ false, // Reserved
+ false, // Reserved
+ };
+
+ return available_checks[(unsigned int)(type)];
+}
+
+
+extern LZMA_API(uint32_t)
+lzma_check_size(lzma_check type)
+{
+ if ((unsigned int)(type) > LZMA_CHECK_ID_MAX)
+ return UINT32_MAX;
+
+ // See file-format.txt section 2.1.1.2.
+ static const uint8_t check_sizes[LZMA_CHECK_ID_MAX + 1] = {
+ 0,
+ 4, 4, 4,
+ 8, 8, 8,
+ 16, 16, 16,
+ 32, 32, 32,
+ 64, 64, 64
+ };
+
+ return check_sizes[(unsigned int)(type)];
+}
+
+
+extern void
+lzma_check_init(lzma_check_state *check, lzma_check type)
+{
+ switch (type) {
+ case LZMA_CHECK_NONE:
+ break;
+
+#ifdef HAVE_CHECK_CRC32
+ case LZMA_CHECK_CRC32:
+ check->state.crc32 = 0;
+ break;
+#endif
+
+#ifdef HAVE_CHECK_CRC64
+ case LZMA_CHECK_CRC64:
+ check->state.crc64 = 0;
+ break;
+#endif
+
+#ifdef HAVE_CHECK_SHA256
+ case LZMA_CHECK_SHA256:
+ lzma_sha256_init(check);
+ break;
+#endif
+
+ default:
+ break;
+ }
+
+ return;
+}
+
+
+extern void
+lzma_check_update(lzma_check_state *check, lzma_check type,
+ const uint8_t *buf, size_t size)
+{
+ switch (type) {
+#ifdef HAVE_CHECK_CRC32
+ case LZMA_CHECK_CRC32:
+ check->state.crc32 = lzma_crc32(buf, size, check->state.crc32);
+ break;
+#endif
+
+#ifdef HAVE_CHECK_CRC64
+ case LZMA_CHECK_CRC64:
+ check->state.crc64 = lzma_crc64(buf, size, check->state.crc64);
+ break;
+#endif
+
+#ifdef HAVE_CHECK_SHA256
+ case LZMA_CHECK_SHA256:
+ lzma_sha256_update(buf, size, check);
+ break;
+#endif
+
+ default:
+ break;
+ }
+
+ return;
+}
+
+
+extern void
+lzma_check_finish(lzma_check_state *check, lzma_check type)
+{
+ switch (type) {
+#ifdef HAVE_CHECK_CRC32
+ case LZMA_CHECK_CRC32:
+ check->buffer.u32[0] = conv32le(check->state.crc32);
+ break;
+#endif
+
+#ifdef HAVE_CHECK_CRC64
+ case LZMA_CHECK_CRC64:
+ check->buffer.u64[0] = conv64le(check->state.crc64);
+ break;
+#endif
+
+#ifdef HAVE_CHECK_SHA256
+ case LZMA_CHECK_SHA256:
+ lzma_sha256_finish(check);
+ break;
+#endif
+
+ default:
+ break;
+ }
+
+ return;
+}
diff --git a/src/liblzma/check/check.h b/src/liblzma/check/check.h
new file mode 100644
index 0000000..e100d2b
--- /dev/null
+++ b/src/liblzma/check/check.h
@@ -0,0 +1,95 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file check.h
+/// \brief Internal API to different integrity check functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_CHECK_H
+#define LZMA_CHECK_H
+
+#include "common.h"
+
+
+// Index hashing needs the best possible hash function (preferably
+// a cryptographic hash) for maximum reliability.
+#if defined(HAVE_CHECK_SHA256)
+# define LZMA_CHECK_BEST LZMA_CHECK_SHA256
+#elif defined(HAVE_CHECK_CRC64)
+# define LZMA_CHECK_BEST LZMA_CHECK_CRC64
+#else
+# define LZMA_CHECK_BEST LZMA_CHECK_CRC32
+#endif
+
+
+/// \brief Structure to hold internal state of the check being calculated
+///
+/// \note This is not in the public API because this structure may
+/// change in future if new integrity check algorithms are added.
+typedef struct {
+ /// Buffer to hold the final result and a temporary buffer for SHA256.
+ union {
+ uint8_t u8[64];
+ uint32_t u32[16];
+ uint64_t u64[8];
+ } buffer;
+
+ /// Check-specific data
+ union {
+ uint32_t crc32;
+ uint64_t crc64;
+
+ struct {
+ /// Internal state
+ uint32_t state[8];
+
+ /// Size of the message excluding padding
+ uint64_t size;
+ } sha256;
+ } state;
+
+} lzma_check_state;
+
+
+/// lzma_crc32_table[0] is needed by LZ encoder so we need to keep
+/// the array two-dimensional.
+#ifdef HAVE_SMALL
+extern uint32_t lzma_crc32_table[1][256];
+extern void lzma_crc32_init(void);
+#else
+extern const uint32_t lzma_crc32_table[8][256];
+extern const uint64_t lzma_crc64_table[4][256];
+#endif
+
+
+/// \brief Initialize *check depending on type
+///
+/// \return LZMA_OK on success. LZMA_UNSUPPORTED_CHECK if the type is not
+/// supported by the current version or build of liblzma.
+/// LZMA_PROG_ERROR if type > LZMA_CHECK_ID_MAX.
+extern void lzma_check_init(lzma_check_state *check, lzma_check type);
+
+/// Update the check state
+extern void lzma_check_update(lzma_check_state *check, lzma_check type,
+ const uint8_t *buf, size_t size);
+
+/// Finish the check calculation and store the result to check->buffer.u8.
+extern void lzma_check_finish(lzma_check_state *check, lzma_check type);
+
+
+/// Prepare SHA-256 state for new input.
+extern void lzma_sha256_init(lzma_check_state *check);
+
+/// Update the SHA-256 hash state
+extern void lzma_sha256_update(
+ const uint8_t *buf, size_t size, lzma_check_state *check);
+
+/// Finish the SHA-256 calculation and store the result to check->buffer.u8.
+extern void lzma_sha256_finish(lzma_check_state *check);
+
+#endif
diff --git a/src/liblzma/check/crc32_fast.c b/src/liblzma/check/crc32_fast.c
new file mode 100644
index 0000000..94da855
--- /dev/null
+++ b/src/liblzma/check/crc32_fast.c
@@ -0,0 +1,82 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc32.c
+/// \brief CRC32 calculation
+///
+/// Calculate the CRC32 using the slice-by-eight algorithm.
+/// It is explained in this document:
+/// http://www.intel.com/technology/comms/perfnet/download/CRC_generators.pdf
+/// The code in this file is not the same as in Intel's paper, but
+/// the basic principle is identical.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "check.h"
+#include "crc_macros.h"
+
+
+// If you make any changes, do some bench marking! Seemingly unrelated
+// changes can very easily ruin the performance (and very probably is
+// very compiler dependent).
+extern LZMA_API(uint32_t)
+lzma_crc32(const uint8_t *buf, size_t size, uint32_t crc)
+{
+ crc = ~crc;
+
+#ifdef WORDS_BIGENDIAN
+ crc = bswap32(crc);
+#endif
+
+ if (size > 8) {
+ // Fix the alignment, if needed. The if statement above
+ // ensures that this won't read past the end of buf[].
+ while ((uintptr_t)(buf) & 7) {
+ crc = lzma_crc32_table[0][*buf++ ^ A(crc)] ^ S8(crc);
+ --size;
+ }
+
+ // Calculate the position where to stop.
+ const uint8_t *const limit = buf + (size & ~(size_t)(7));
+
+ // Calculate how many bytes must be calculated separately
+ // before returning the result.
+ size &= (size_t)(7);
+
+ // Calculate the CRC32 using the slice-by-eight algorithm.
+ while (buf < limit) {
+ crc ^= *(const uint32_t *)(buf);
+ buf += 4;
+
+ crc = lzma_crc32_table[7][A(crc)]
+ ^ lzma_crc32_table[6][B(crc)]
+ ^ lzma_crc32_table[5][C(crc)]
+ ^ lzma_crc32_table[4][D(crc)];
+
+ const uint32_t tmp = *(const uint32_t *)(buf);
+ buf += 4;
+
+ // At least with some compilers, it is critical for
+ // performance, that the crc variable is XORed
+ // between the two table-lookup pairs.
+ crc = lzma_crc32_table[3][A(tmp)]
+ ^ lzma_crc32_table[2][B(tmp)]
+ ^ crc
+ ^ lzma_crc32_table[1][C(tmp)]
+ ^ lzma_crc32_table[0][D(tmp)];
+ }
+ }
+
+ while (size-- != 0)
+ crc = lzma_crc32_table[0][*buf++ ^ A(crc)] ^ S8(crc);
+
+#ifdef WORDS_BIGENDIAN
+ crc = bswap32(crc);
+#endif
+
+ return ~crc;
+}
diff --git a/src/liblzma/check/crc32_small.c b/src/liblzma/check/crc32_small.c
new file mode 100644
index 0000000..5f8a328
--- /dev/null
+++ b/src/liblzma/check/crc32_small.c
@@ -0,0 +1,61 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc32_small.c
+/// \brief CRC32 calculation (size-optimized)
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "check.h"
+
+
+uint32_t lzma_crc32_table[1][256];
+
+
+static void
+crc32_init(void)
+{
+ static const uint32_t poly32 = UINT32_C(0xEDB88320);
+
+ for (size_t b = 0; b < 256; ++b) {
+ uint32_t r = b;
+ for (size_t i = 0; i < 8; ++i) {
+ if (r & 1)
+ r = (r >> 1) ^ poly32;
+ else
+ r >>= 1;
+ }
+
+ lzma_crc32_table[0][b] = r;
+ }
+
+ return;
+}
+
+
+extern void
+lzma_crc32_init(void)
+{
+ mythread_once(crc32_init);
+ return;
+}
+
+
+extern LZMA_API(uint32_t)
+lzma_crc32(const uint8_t *buf, size_t size, uint32_t crc)
+{
+ lzma_crc32_init();
+
+ crc = ~crc;
+
+ while (size != 0) {
+ crc = lzma_crc32_table[0][*buf++ ^ (crc & 0xFF)] ^ (crc >> 8);
+ --size;
+ }
+
+ return ~crc;
+}
diff --git a/src/liblzma/check/crc32_table.c b/src/liblzma/check/crc32_table.c
new file mode 100644
index 0000000..368874e
--- /dev/null
+++ b/src/liblzma/check/crc32_table.c
@@ -0,0 +1,19 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc32_table.c
+/// \brief Precalculated CRC32 table with correct endianness
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+#ifdef WORDS_BIGENDIAN
+# include "crc32_table_be.h"
+#else
+# include "crc32_table_le.h"
+#endif
diff --git a/src/liblzma/check/crc32_table_be.h b/src/liblzma/check/crc32_table_be.h
new file mode 100644
index 0000000..c483cb6
--- /dev/null
+++ b/src/liblzma/check/crc32_table_be.h
@@ -0,0 +1,525 @@
+/* This file has been automatically generated by crc32_tablegen.c. */
+
+const uint32_t lzma_crc32_table[8][256] = {
+ {
+ 0x00000000, 0x96300777, 0x2C610EEE, 0xBA510999,
+ 0x19C46D07, 0x8FF46A70, 0x35A563E9, 0xA395649E,
+ 0x3288DB0E, 0xA4B8DC79, 0x1EE9D5E0, 0x88D9D297,
+ 0x2B4CB609, 0xBD7CB17E, 0x072DB8E7, 0x911DBF90,
+ 0x6410B71D, 0xF220B06A, 0x4871B9F3, 0xDE41BE84,
+ 0x7DD4DA1A, 0xEBE4DD6D, 0x51B5D4F4, 0xC785D383,
+ 0x56986C13, 0xC0A86B64, 0x7AF962FD, 0xECC9658A,
+ 0x4F5C0114, 0xD96C0663, 0x633D0FFA, 0xF50D088D,
+ 0xC8206E3B, 0x5E10694C, 0xE44160D5, 0x727167A2,
+ 0xD1E4033C, 0x47D4044B, 0xFD850DD2, 0x6BB50AA5,
+ 0xFAA8B535, 0x6C98B242, 0xD6C9BBDB, 0x40F9BCAC,
+ 0xE36CD832, 0x755CDF45, 0xCF0DD6DC, 0x593DD1AB,
+ 0xAC30D926, 0x3A00DE51, 0x8051D7C8, 0x1661D0BF,
+ 0xB5F4B421, 0x23C4B356, 0x9995BACF, 0x0FA5BDB8,
+ 0x9EB80228, 0x0888055F, 0xB2D90CC6, 0x24E90BB1,
+ 0x877C6F2F, 0x114C6858, 0xAB1D61C1, 0x3D2D66B6,
+ 0x9041DC76, 0x0671DB01, 0xBC20D298, 0x2A10D5EF,
+ 0x8985B171, 0x1FB5B606, 0xA5E4BF9F, 0x33D4B8E8,
+ 0xA2C90778, 0x34F9000F, 0x8EA80996, 0x18980EE1,
+ 0xBB0D6A7F, 0x2D3D6D08, 0x976C6491, 0x015C63E6,
+ 0xF4516B6B, 0x62616C1C, 0xD8306585, 0x4E0062F2,
+ 0xED95066C, 0x7BA5011B, 0xC1F40882, 0x57C40FF5,
+ 0xC6D9B065, 0x50E9B712, 0xEAB8BE8B, 0x7C88B9FC,
+ 0xDF1DDD62, 0x492DDA15, 0xF37CD38C, 0x654CD4FB,
+ 0x5861B24D, 0xCE51B53A, 0x7400BCA3, 0xE230BBD4,
+ 0x41A5DF4A, 0xD795D83D, 0x6DC4D1A4, 0xFBF4D6D3,
+ 0x6AE96943, 0xFCD96E34, 0x468867AD, 0xD0B860DA,
+ 0x732D0444, 0xE51D0333, 0x5F4C0AAA, 0xC97C0DDD,
+ 0x3C710550, 0xAA410227, 0x10100BBE, 0x86200CC9,
+ 0x25B56857, 0xB3856F20, 0x09D466B9, 0x9FE461CE,
+ 0x0EF9DE5E, 0x98C9D929, 0x2298D0B0, 0xB4A8D7C7,
+ 0x173DB359, 0x810DB42E, 0x3B5CBDB7, 0xAD6CBAC0,
+ 0x2083B8ED, 0xB6B3BF9A, 0x0CE2B603, 0x9AD2B174,
+ 0x3947D5EA, 0xAF77D29D, 0x1526DB04, 0x8316DC73,
+ 0x120B63E3, 0x843B6494, 0x3E6A6D0D, 0xA85A6A7A,
+ 0x0BCF0EE4, 0x9DFF0993, 0x27AE000A, 0xB19E077D,
+ 0x44930FF0, 0xD2A30887, 0x68F2011E, 0xFEC20669,
+ 0x5D5762F7, 0xCB676580, 0x71366C19, 0xE7066B6E,
+ 0x761BD4FE, 0xE02BD389, 0x5A7ADA10, 0xCC4ADD67,
+ 0x6FDFB9F9, 0xF9EFBE8E, 0x43BEB717, 0xD58EB060,
+ 0xE8A3D6D6, 0x7E93D1A1, 0xC4C2D838, 0x52F2DF4F,
+ 0xF167BBD1, 0x6757BCA6, 0xDD06B53F, 0x4B36B248,
+ 0xDA2B0DD8, 0x4C1B0AAF, 0xF64A0336, 0x607A0441,
+ 0xC3EF60DF, 0x55DF67A8, 0xEF8E6E31, 0x79BE6946,
+ 0x8CB361CB, 0x1A8366BC, 0xA0D26F25, 0x36E26852,
+ 0x95770CCC, 0x03470BBB, 0xB9160222, 0x2F260555,
+ 0xBE3BBAC5, 0x280BBDB2, 0x925AB42B, 0x046AB35C,
+ 0xA7FFD7C2, 0x31CFD0B5, 0x8B9ED92C, 0x1DAEDE5B,
+ 0xB0C2649B, 0x26F263EC, 0x9CA36A75, 0x0A936D02,
+ 0xA906099C, 0x3F360EEB, 0x85670772, 0x13570005,
+ 0x824ABF95, 0x147AB8E2, 0xAE2BB17B, 0x381BB60C,
+ 0x9B8ED292, 0x0DBED5E5, 0xB7EFDC7C, 0x21DFDB0B,
+ 0xD4D2D386, 0x42E2D4F1, 0xF8B3DD68, 0x6E83DA1F,
+ 0xCD16BE81, 0x5B26B9F6, 0xE177B06F, 0x7747B718,
+ 0xE65A0888, 0x706A0FFF, 0xCA3B0666, 0x5C0B0111,
+ 0xFF9E658F, 0x69AE62F8, 0xD3FF6B61, 0x45CF6C16,
+ 0x78E20AA0, 0xEED20DD7, 0x5483044E, 0xC2B30339,
+ 0x612667A7, 0xF71660D0, 0x4D476949, 0xDB776E3E,
+ 0x4A6AD1AE, 0xDC5AD6D9, 0x660BDF40, 0xF03BD837,
+ 0x53AEBCA9, 0xC59EBBDE, 0x7FCFB247, 0xE9FFB530,
+ 0x1CF2BDBD, 0x8AC2BACA, 0x3093B353, 0xA6A3B424,
+ 0x0536D0BA, 0x9306D7CD, 0x2957DE54, 0xBF67D923,
+ 0x2E7A66B3, 0xB84A61C4, 0x021B685D, 0x942B6F2A,
+ 0x37BE0BB4, 0xA18E0CC3, 0x1BDF055A, 0x8DEF022D
+ }, {
+ 0x00000000, 0x41311B19, 0x82623632, 0xC3532D2B,
+ 0x04C56C64, 0x45F4777D, 0x86A75A56, 0xC796414F,
+ 0x088AD9C8, 0x49BBC2D1, 0x8AE8EFFA, 0xCBD9F4E3,
+ 0x0C4FB5AC, 0x4D7EAEB5, 0x8E2D839E, 0xCF1C9887,
+ 0x5112C24A, 0x1023D953, 0xD370F478, 0x9241EF61,
+ 0x55D7AE2E, 0x14E6B537, 0xD7B5981C, 0x96848305,
+ 0x59981B82, 0x18A9009B, 0xDBFA2DB0, 0x9ACB36A9,
+ 0x5D5D77E6, 0x1C6C6CFF, 0xDF3F41D4, 0x9E0E5ACD,
+ 0xA2248495, 0xE3159F8C, 0x2046B2A7, 0x6177A9BE,
+ 0xA6E1E8F1, 0xE7D0F3E8, 0x2483DEC3, 0x65B2C5DA,
+ 0xAAAE5D5D, 0xEB9F4644, 0x28CC6B6F, 0x69FD7076,
+ 0xAE6B3139, 0xEF5A2A20, 0x2C09070B, 0x6D381C12,
+ 0xF33646DF, 0xB2075DC6, 0x715470ED, 0x30656BF4,
+ 0xF7F32ABB, 0xB6C231A2, 0x75911C89, 0x34A00790,
+ 0xFBBC9F17, 0xBA8D840E, 0x79DEA925, 0x38EFB23C,
+ 0xFF79F373, 0xBE48E86A, 0x7D1BC541, 0x3C2ADE58,
+ 0x054F79F0, 0x447E62E9, 0x872D4FC2, 0xC61C54DB,
+ 0x018A1594, 0x40BB0E8D, 0x83E823A6, 0xC2D938BF,
+ 0x0DC5A038, 0x4CF4BB21, 0x8FA7960A, 0xCE968D13,
+ 0x0900CC5C, 0x4831D745, 0x8B62FA6E, 0xCA53E177,
+ 0x545DBBBA, 0x156CA0A3, 0xD63F8D88, 0x970E9691,
+ 0x5098D7DE, 0x11A9CCC7, 0xD2FAE1EC, 0x93CBFAF5,
+ 0x5CD76272, 0x1DE6796B, 0xDEB55440, 0x9F844F59,
+ 0x58120E16, 0x1923150F, 0xDA703824, 0x9B41233D,
+ 0xA76BFD65, 0xE65AE67C, 0x2509CB57, 0x6438D04E,
+ 0xA3AE9101, 0xE29F8A18, 0x21CCA733, 0x60FDBC2A,
+ 0xAFE124AD, 0xEED03FB4, 0x2D83129F, 0x6CB20986,
+ 0xAB2448C9, 0xEA1553D0, 0x29467EFB, 0x687765E2,
+ 0xF6793F2F, 0xB7482436, 0x741B091D, 0x352A1204,
+ 0xF2BC534B, 0xB38D4852, 0x70DE6579, 0x31EF7E60,
+ 0xFEF3E6E7, 0xBFC2FDFE, 0x7C91D0D5, 0x3DA0CBCC,
+ 0xFA368A83, 0xBB07919A, 0x7854BCB1, 0x3965A7A8,
+ 0x4B98833B, 0x0AA99822, 0xC9FAB509, 0x88CBAE10,
+ 0x4F5DEF5F, 0x0E6CF446, 0xCD3FD96D, 0x8C0EC274,
+ 0x43125AF3, 0x022341EA, 0xC1706CC1, 0x804177D8,
+ 0x47D73697, 0x06E62D8E, 0xC5B500A5, 0x84841BBC,
+ 0x1A8A4171, 0x5BBB5A68, 0x98E87743, 0xD9D96C5A,
+ 0x1E4F2D15, 0x5F7E360C, 0x9C2D1B27, 0xDD1C003E,
+ 0x120098B9, 0x533183A0, 0x9062AE8B, 0xD153B592,
+ 0x16C5F4DD, 0x57F4EFC4, 0x94A7C2EF, 0xD596D9F6,
+ 0xE9BC07AE, 0xA88D1CB7, 0x6BDE319C, 0x2AEF2A85,
+ 0xED796BCA, 0xAC4870D3, 0x6F1B5DF8, 0x2E2A46E1,
+ 0xE136DE66, 0xA007C57F, 0x6354E854, 0x2265F34D,
+ 0xE5F3B202, 0xA4C2A91B, 0x67918430, 0x26A09F29,
+ 0xB8AEC5E4, 0xF99FDEFD, 0x3ACCF3D6, 0x7BFDE8CF,
+ 0xBC6BA980, 0xFD5AB299, 0x3E099FB2, 0x7F3884AB,
+ 0xB0241C2C, 0xF1150735, 0x32462A1E, 0x73773107,
+ 0xB4E17048, 0xF5D06B51, 0x3683467A, 0x77B25D63,
+ 0x4ED7FACB, 0x0FE6E1D2, 0xCCB5CCF9, 0x8D84D7E0,
+ 0x4A1296AF, 0x0B238DB6, 0xC870A09D, 0x8941BB84,
+ 0x465D2303, 0x076C381A, 0xC43F1531, 0x850E0E28,
+ 0x42984F67, 0x03A9547E, 0xC0FA7955, 0x81CB624C,
+ 0x1FC53881, 0x5EF42398, 0x9DA70EB3, 0xDC9615AA,
+ 0x1B0054E5, 0x5A314FFC, 0x996262D7, 0xD85379CE,
+ 0x174FE149, 0x567EFA50, 0x952DD77B, 0xD41CCC62,
+ 0x138A8D2D, 0x52BB9634, 0x91E8BB1F, 0xD0D9A006,
+ 0xECF37E5E, 0xADC26547, 0x6E91486C, 0x2FA05375,
+ 0xE836123A, 0xA9070923, 0x6A542408, 0x2B653F11,
+ 0xE479A796, 0xA548BC8F, 0x661B91A4, 0x272A8ABD,
+ 0xE0BCCBF2, 0xA18DD0EB, 0x62DEFDC0, 0x23EFE6D9,
+ 0xBDE1BC14, 0xFCD0A70D, 0x3F838A26, 0x7EB2913F,
+ 0xB924D070, 0xF815CB69, 0x3B46E642, 0x7A77FD5B,
+ 0xB56B65DC, 0xF45A7EC5, 0x370953EE, 0x763848F7,
+ 0xB1AE09B8, 0xF09F12A1, 0x33CC3F8A, 0x72FD2493
+ }, {
+ 0x00000000, 0x376AC201, 0x6ED48403, 0x59BE4602,
+ 0xDCA80907, 0xEBC2CB06, 0xB27C8D04, 0x85164F05,
+ 0xB851130E, 0x8F3BD10F, 0xD685970D, 0xE1EF550C,
+ 0x64F91A09, 0x5393D808, 0x0A2D9E0A, 0x3D475C0B,
+ 0x70A3261C, 0x47C9E41D, 0x1E77A21F, 0x291D601E,
+ 0xAC0B2F1B, 0x9B61ED1A, 0xC2DFAB18, 0xF5B56919,
+ 0xC8F23512, 0xFF98F713, 0xA626B111, 0x914C7310,
+ 0x145A3C15, 0x2330FE14, 0x7A8EB816, 0x4DE47A17,
+ 0xE0464D38, 0xD72C8F39, 0x8E92C93B, 0xB9F80B3A,
+ 0x3CEE443F, 0x0B84863E, 0x523AC03C, 0x6550023D,
+ 0x58175E36, 0x6F7D9C37, 0x36C3DA35, 0x01A91834,
+ 0x84BF5731, 0xB3D59530, 0xEA6BD332, 0xDD011133,
+ 0x90E56B24, 0xA78FA925, 0xFE31EF27, 0xC95B2D26,
+ 0x4C4D6223, 0x7B27A022, 0x2299E620, 0x15F32421,
+ 0x28B4782A, 0x1FDEBA2B, 0x4660FC29, 0x710A3E28,
+ 0xF41C712D, 0xC376B32C, 0x9AC8F52E, 0xADA2372F,
+ 0xC08D9A70, 0xF7E75871, 0xAE591E73, 0x9933DC72,
+ 0x1C259377, 0x2B4F5176, 0x72F11774, 0x459BD575,
+ 0x78DC897E, 0x4FB64B7F, 0x16080D7D, 0x2162CF7C,
+ 0xA4748079, 0x931E4278, 0xCAA0047A, 0xFDCAC67B,
+ 0xB02EBC6C, 0x87447E6D, 0xDEFA386F, 0xE990FA6E,
+ 0x6C86B56B, 0x5BEC776A, 0x02523168, 0x3538F369,
+ 0x087FAF62, 0x3F156D63, 0x66AB2B61, 0x51C1E960,
+ 0xD4D7A665, 0xE3BD6464, 0xBA032266, 0x8D69E067,
+ 0x20CBD748, 0x17A11549, 0x4E1F534B, 0x7975914A,
+ 0xFC63DE4F, 0xCB091C4E, 0x92B75A4C, 0xA5DD984D,
+ 0x989AC446, 0xAFF00647, 0xF64E4045, 0xC1248244,
+ 0x4432CD41, 0x73580F40, 0x2AE64942, 0x1D8C8B43,
+ 0x5068F154, 0x67023355, 0x3EBC7557, 0x09D6B756,
+ 0x8CC0F853, 0xBBAA3A52, 0xE2147C50, 0xD57EBE51,
+ 0xE839E25A, 0xDF53205B, 0x86ED6659, 0xB187A458,
+ 0x3491EB5D, 0x03FB295C, 0x5A456F5E, 0x6D2FAD5F,
+ 0x801B35E1, 0xB771F7E0, 0xEECFB1E2, 0xD9A573E3,
+ 0x5CB33CE6, 0x6BD9FEE7, 0x3267B8E5, 0x050D7AE4,
+ 0x384A26EF, 0x0F20E4EE, 0x569EA2EC, 0x61F460ED,
+ 0xE4E22FE8, 0xD388EDE9, 0x8A36ABEB, 0xBD5C69EA,
+ 0xF0B813FD, 0xC7D2D1FC, 0x9E6C97FE, 0xA90655FF,
+ 0x2C101AFA, 0x1B7AD8FB, 0x42C49EF9, 0x75AE5CF8,
+ 0x48E900F3, 0x7F83C2F2, 0x263D84F0, 0x115746F1,
+ 0x944109F4, 0xA32BCBF5, 0xFA958DF7, 0xCDFF4FF6,
+ 0x605D78D9, 0x5737BAD8, 0x0E89FCDA, 0x39E33EDB,
+ 0xBCF571DE, 0x8B9FB3DF, 0xD221F5DD, 0xE54B37DC,
+ 0xD80C6BD7, 0xEF66A9D6, 0xB6D8EFD4, 0x81B22DD5,
+ 0x04A462D0, 0x33CEA0D1, 0x6A70E6D3, 0x5D1A24D2,
+ 0x10FE5EC5, 0x27949CC4, 0x7E2ADAC6, 0x494018C7,
+ 0xCC5657C2, 0xFB3C95C3, 0xA282D3C1, 0x95E811C0,
+ 0xA8AF4DCB, 0x9FC58FCA, 0xC67BC9C8, 0xF1110BC9,
+ 0x740744CC, 0x436D86CD, 0x1AD3C0CF, 0x2DB902CE,
+ 0x4096AF91, 0x77FC6D90, 0x2E422B92, 0x1928E993,
+ 0x9C3EA696, 0xAB546497, 0xF2EA2295, 0xC580E094,
+ 0xF8C7BC9F, 0xCFAD7E9E, 0x9613389C, 0xA179FA9D,
+ 0x246FB598, 0x13057799, 0x4ABB319B, 0x7DD1F39A,
+ 0x3035898D, 0x075F4B8C, 0x5EE10D8E, 0x698BCF8F,
+ 0xEC9D808A, 0xDBF7428B, 0x82490489, 0xB523C688,
+ 0x88649A83, 0xBF0E5882, 0xE6B01E80, 0xD1DADC81,
+ 0x54CC9384, 0x63A65185, 0x3A181787, 0x0D72D586,
+ 0xA0D0E2A9, 0x97BA20A8, 0xCE0466AA, 0xF96EA4AB,
+ 0x7C78EBAE, 0x4B1229AF, 0x12AC6FAD, 0x25C6ADAC,
+ 0x1881F1A7, 0x2FEB33A6, 0x765575A4, 0x413FB7A5,
+ 0xC429F8A0, 0xF3433AA1, 0xAAFD7CA3, 0x9D97BEA2,
+ 0xD073C4B5, 0xE71906B4, 0xBEA740B6, 0x89CD82B7,
+ 0x0CDBCDB2, 0x3BB10FB3, 0x620F49B1, 0x55658BB0,
+ 0x6822D7BB, 0x5F4815BA, 0x06F653B8, 0x319C91B9,
+ 0xB48ADEBC, 0x83E01CBD, 0xDA5E5ABF, 0xED3498BE
+ }, {
+ 0x00000000, 0x6567BCB8, 0x8BC809AA, 0xEEAFB512,
+ 0x5797628F, 0x32F0DE37, 0xDC5F6B25, 0xB938D79D,
+ 0xEF28B4C5, 0x8A4F087D, 0x64E0BD6F, 0x018701D7,
+ 0xB8BFD64A, 0xDDD86AF2, 0x3377DFE0, 0x56106358,
+ 0x9F571950, 0xFA30A5E8, 0x149F10FA, 0x71F8AC42,
+ 0xC8C07BDF, 0xADA7C767, 0x43087275, 0x266FCECD,
+ 0x707FAD95, 0x1518112D, 0xFBB7A43F, 0x9ED01887,
+ 0x27E8CF1A, 0x428F73A2, 0xAC20C6B0, 0xC9477A08,
+ 0x3EAF32A0, 0x5BC88E18, 0xB5673B0A, 0xD00087B2,
+ 0x6938502F, 0x0C5FEC97, 0xE2F05985, 0x8797E53D,
+ 0xD1878665, 0xB4E03ADD, 0x5A4F8FCF, 0x3F283377,
+ 0x8610E4EA, 0xE3775852, 0x0DD8ED40, 0x68BF51F8,
+ 0xA1F82BF0, 0xC49F9748, 0x2A30225A, 0x4F579EE2,
+ 0xF66F497F, 0x9308F5C7, 0x7DA740D5, 0x18C0FC6D,
+ 0x4ED09F35, 0x2BB7238D, 0xC518969F, 0xA07F2A27,
+ 0x1947FDBA, 0x7C204102, 0x928FF410, 0xF7E848A8,
+ 0x3D58149B, 0x583FA823, 0xB6901D31, 0xD3F7A189,
+ 0x6ACF7614, 0x0FA8CAAC, 0xE1077FBE, 0x8460C306,
+ 0xD270A05E, 0xB7171CE6, 0x59B8A9F4, 0x3CDF154C,
+ 0x85E7C2D1, 0xE0807E69, 0x0E2FCB7B, 0x6B4877C3,
+ 0xA20F0DCB, 0xC768B173, 0x29C70461, 0x4CA0B8D9,
+ 0xF5986F44, 0x90FFD3FC, 0x7E5066EE, 0x1B37DA56,
+ 0x4D27B90E, 0x284005B6, 0xC6EFB0A4, 0xA3880C1C,
+ 0x1AB0DB81, 0x7FD76739, 0x9178D22B, 0xF41F6E93,
+ 0x03F7263B, 0x66909A83, 0x883F2F91, 0xED589329,
+ 0x546044B4, 0x3107F80C, 0xDFA84D1E, 0xBACFF1A6,
+ 0xECDF92FE, 0x89B82E46, 0x67179B54, 0x027027EC,
+ 0xBB48F071, 0xDE2F4CC9, 0x3080F9DB, 0x55E74563,
+ 0x9CA03F6B, 0xF9C783D3, 0x176836C1, 0x720F8A79,
+ 0xCB375DE4, 0xAE50E15C, 0x40FF544E, 0x2598E8F6,
+ 0x73888BAE, 0x16EF3716, 0xF8408204, 0x9D273EBC,
+ 0x241FE921, 0x41785599, 0xAFD7E08B, 0xCAB05C33,
+ 0x3BB659ED, 0x5ED1E555, 0xB07E5047, 0xD519ECFF,
+ 0x6C213B62, 0x094687DA, 0xE7E932C8, 0x828E8E70,
+ 0xD49EED28, 0xB1F95190, 0x5F56E482, 0x3A31583A,
+ 0x83098FA7, 0xE66E331F, 0x08C1860D, 0x6DA63AB5,
+ 0xA4E140BD, 0xC186FC05, 0x2F294917, 0x4A4EF5AF,
+ 0xF3762232, 0x96119E8A, 0x78BE2B98, 0x1DD99720,
+ 0x4BC9F478, 0x2EAE48C0, 0xC001FDD2, 0xA566416A,
+ 0x1C5E96F7, 0x79392A4F, 0x97969F5D, 0xF2F123E5,
+ 0x05196B4D, 0x607ED7F5, 0x8ED162E7, 0xEBB6DE5F,
+ 0x528E09C2, 0x37E9B57A, 0xD9460068, 0xBC21BCD0,
+ 0xEA31DF88, 0x8F566330, 0x61F9D622, 0x049E6A9A,
+ 0xBDA6BD07, 0xD8C101BF, 0x366EB4AD, 0x53090815,
+ 0x9A4E721D, 0xFF29CEA5, 0x11867BB7, 0x74E1C70F,
+ 0xCDD91092, 0xA8BEAC2A, 0x46111938, 0x2376A580,
+ 0x7566C6D8, 0x10017A60, 0xFEAECF72, 0x9BC973CA,
+ 0x22F1A457, 0x479618EF, 0xA939ADFD, 0xCC5E1145,
+ 0x06EE4D76, 0x6389F1CE, 0x8D2644DC, 0xE841F864,
+ 0x51792FF9, 0x341E9341, 0xDAB12653, 0xBFD69AEB,
+ 0xE9C6F9B3, 0x8CA1450B, 0x620EF019, 0x07694CA1,
+ 0xBE519B3C, 0xDB362784, 0x35999296, 0x50FE2E2E,
+ 0x99B95426, 0xFCDEE89E, 0x12715D8C, 0x7716E134,
+ 0xCE2E36A9, 0xAB498A11, 0x45E63F03, 0x208183BB,
+ 0x7691E0E3, 0x13F65C5B, 0xFD59E949, 0x983E55F1,
+ 0x2106826C, 0x44613ED4, 0xAACE8BC6, 0xCFA9377E,
+ 0x38417FD6, 0x5D26C36E, 0xB389767C, 0xD6EECAC4,
+ 0x6FD61D59, 0x0AB1A1E1, 0xE41E14F3, 0x8179A84B,
+ 0xD769CB13, 0xB20E77AB, 0x5CA1C2B9, 0x39C67E01,
+ 0x80FEA99C, 0xE5991524, 0x0B36A036, 0x6E511C8E,
+ 0xA7166686, 0xC271DA3E, 0x2CDE6F2C, 0x49B9D394,
+ 0xF0810409, 0x95E6B8B1, 0x7B490DA3, 0x1E2EB11B,
+ 0x483ED243, 0x2D596EFB, 0xC3F6DBE9, 0xA6916751,
+ 0x1FA9B0CC, 0x7ACE0C74, 0x9461B966, 0xF10605DE
+ }, {
+ 0x00000000, 0xB029603D, 0x6053C07A, 0xD07AA047,
+ 0xC0A680F5, 0x708FE0C8, 0xA0F5408F, 0x10DC20B2,
+ 0xC14B7030, 0x7162100D, 0xA118B04A, 0x1131D077,
+ 0x01EDF0C5, 0xB1C490F8, 0x61BE30BF, 0xD1975082,
+ 0x8297E060, 0x32BE805D, 0xE2C4201A, 0x52ED4027,
+ 0x42316095, 0xF21800A8, 0x2262A0EF, 0x924BC0D2,
+ 0x43DC9050, 0xF3F5F06D, 0x238F502A, 0x93A63017,
+ 0x837A10A5, 0x33537098, 0xE329D0DF, 0x5300B0E2,
+ 0x042FC1C1, 0xB406A1FC, 0x647C01BB, 0xD4556186,
+ 0xC4894134, 0x74A02109, 0xA4DA814E, 0x14F3E173,
+ 0xC564B1F1, 0x754DD1CC, 0xA537718B, 0x151E11B6,
+ 0x05C23104, 0xB5EB5139, 0x6591F17E, 0xD5B89143,
+ 0x86B821A1, 0x3691419C, 0xE6EBE1DB, 0x56C281E6,
+ 0x461EA154, 0xF637C169, 0x264D612E, 0x96640113,
+ 0x47F35191, 0xF7DA31AC, 0x27A091EB, 0x9789F1D6,
+ 0x8755D164, 0x377CB159, 0xE706111E, 0x572F7123,
+ 0x4958F358, 0xF9719365, 0x290B3322, 0x9922531F,
+ 0x89FE73AD, 0x39D71390, 0xE9ADB3D7, 0x5984D3EA,
+ 0x88138368, 0x383AE355, 0xE8404312, 0x5869232F,
+ 0x48B5039D, 0xF89C63A0, 0x28E6C3E7, 0x98CFA3DA,
+ 0xCBCF1338, 0x7BE67305, 0xAB9CD342, 0x1BB5B37F,
+ 0x0B6993CD, 0xBB40F3F0, 0x6B3A53B7, 0xDB13338A,
+ 0x0A846308, 0xBAAD0335, 0x6AD7A372, 0xDAFEC34F,
+ 0xCA22E3FD, 0x7A0B83C0, 0xAA712387, 0x1A5843BA,
+ 0x4D773299, 0xFD5E52A4, 0x2D24F2E3, 0x9D0D92DE,
+ 0x8DD1B26C, 0x3DF8D251, 0xED827216, 0x5DAB122B,
+ 0x8C3C42A9, 0x3C152294, 0xEC6F82D3, 0x5C46E2EE,
+ 0x4C9AC25C, 0xFCB3A261, 0x2CC90226, 0x9CE0621B,
+ 0xCFE0D2F9, 0x7FC9B2C4, 0xAFB31283, 0x1F9A72BE,
+ 0x0F46520C, 0xBF6F3231, 0x6F159276, 0xDF3CF24B,
+ 0x0EABA2C9, 0xBE82C2F4, 0x6EF862B3, 0xDED1028E,
+ 0xCE0D223C, 0x7E244201, 0xAE5EE246, 0x1E77827B,
+ 0x92B0E6B1, 0x2299868C, 0xF2E326CB, 0x42CA46F6,
+ 0x52166644, 0xE23F0679, 0x3245A63E, 0x826CC603,
+ 0x53FB9681, 0xE3D2F6BC, 0x33A856FB, 0x838136C6,
+ 0x935D1674, 0x23747649, 0xF30ED60E, 0x4327B633,
+ 0x102706D1, 0xA00E66EC, 0x7074C6AB, 0xC05DA696,
+ 0xD0818624, 0x60A8E619, 0xB0D2465E, 0x00FB2663,
+ 0xD16C76E1, 0x614516DC, 0xB13FB69B, 0x0116D6A6,
+ 0x11CAF614, 0xA1E39629, 0x7199366E, 0xC1B05653,
+ 0x969F2770, 0x26B6474D, 0xF6CCE70A, 0x46E58737,
+ 0x5639A785, 0xE610C7B8, 0x366A67FF, 0x864307C2,
+ 0x57D45740, 0xE7FD377D, 0x3787973A, 0x87AEF707,
+ 0x9772D7B5, 0x275BB788, 0xF72117CF, 0x470877F2,
+ 0x1408C710, 0xA421A72D, 0x745B076A, 0xC4726757,
+ 0xD4AE47E5, 0x648727D8, 0xB4FD879F, 0x04D4E7A2,
+ 0xD543B720, 0x656AD71D, 0xB510775A, 0x05391767,
+ 0x15E537D5, 0xA5CC57E8, 0x75B6F7AF, 0xC59F9792,
+ 0xDBE815E9, 0x6BC175D4, 0xBBBBD593, 0x0B92B5AE,
+ 0x1B4E951C, 0xAB67F521, 0x7B1D5566, 0xCB34355B,
+ 0x1AA365D9, 0xAA8A05E4, 0x7AF0A5A3, 0xCAD9C59E,
+ 0xDA05E52C, 0x6A2C8511, 0xBA562556, 0x0A7F456B,
+ 0x597FF589, 0xE95695B4, 0x392C35F3, 0x890555CE,
+ 0x99D9757C, 0x29F01541, 0xF98AB506, 0x49A3D53B,
+ 0x983485B9, 0x281DE584, 0xF86745C3, 0x484E25FE,
+ 0x5892054C, 0xE8BB6571, 0x38C1C536, 0x88E8A50B,
+ 0xDFC7D428, 0x6FEEB415, 0xBF941452, 0x0FBD746F,
+ 0x1F6154DD, 0xAF4834E0, 0x7F3294A7, 0xCF1BF49A,
+ 0x1E8CA418, 0xAEA5C425, 0x7EDF6462, 0xCEF6045F,
+ 0xDE2A24ED, 0x6E0344D0, 0xBE79E497, 0x0E5084AA,
+ 0x5D503448, 0xED795475, 0x3D03F432, 0x8D2A940F,
+ 0x9DF6B4BD, 0x2DDFD480, 0xFDA574C7, 0x4D8C14FA,
+ 0x9C1B4478, 0x2C322445, 0xFC488402, 0x4C61E43F,
+ 0x5CBDC48D, 0xEC94A4B0, 0x3CEE04F7, 0x8CC764CA
+ }, {
+ 0x00000000, 0xA5D35CCB, 0x0BA1C84D, 0xAE729486,
+ 0x1642919B, 0xB391CD50, 0x1DE359D6, 0xB830051D,
+ 0x6D8253EC, 0xC8510F27, 0x66239BA1, 0xC3F0C76A,
+ 0x7BC0C277, 0xDE139EBC, 0x70610A3A, 0xD5B256F1,
+ 0x9B02D603, 0x3ED18AC8, 0x90A31E4E, 0x35704285,
+ 0x8D404798, 0x28931B53, 0x86E18FD5, 0x2332D31E,
+ 0xF68085EF, 0x5353D924, 0xFD214DA2, 0x58F21169,
+ 0xE0C21474, 0x451148BF, 0xEB63DC39, 0x4EB080F2,
+ 0x3605AC07, 0x93D6F0CC, 0x3DA4644A, 0x98773881,
+ 0x20473D9C, 0x85946157, 0x2BE6F5D1, 0x8E35A91A,
+ 0x5B87FFEB, 0xFE54A320, 0x502637A6, 0xF5F56B6D,
+ 0x4DC56E70, 0xE81632BB, 0x4664A63D, 0xE3B7FAF6,
+ 0xAD077A04, 0x08D426CF, 0xA6A6B249, 0x0375EE82,
+ 0xBB45EB9F, 0x1E96B754, 0xB0E423D2, 0x15377F19,
+ 0xC08529E8, 0x65567523, 0xCB24E1A5, 0x6EF7BD6E,
+ 0xD6C7B873, 0x7314E4B8, 0xDD66703E, 0x78B52CF5,
+ 0x6C0A580F, 0xC9D904C4, 0x67AB9042, 0xC278CC89,
+ 0x7A48C994, 0xDF9B955F, 0x71E901D9, 0xD43A5D12,
+ 0x01880BE3, 0xA45B5728, 0x0A29C3AE, 0xAFFA9F65,
+ 0x17CA9A78, 0xB219C6B3, 0x1C6B5235, 0xB9B80EFE,
+ 0xF7088E0C, 0x52DBD2C7, 0xFCA94641, 0x597A1A8A,
+ 0xE14A1F97, 0x4499435C, 0xEAEBD7DA, 0x4F388B11,
+ 0x9A8ADDE0, 0x3F59812B, 0x912B15AD, 0x34F84966,
+ 0x8CC84C7B, 0x291B10B0, 0x87698436, 0x22BAD8FD,
+ 0x5A0FF408, 0xFFDCA8C3, 0x51AE3C45, 0xF47D608E,
+ 0x4C4D6593, 0xE99E3958, 0x47ECADDE, 0xE23FF115,
+ 0x378DA7E4, 0x925EFB2F, 0x3C2C6FA9, 0x99FF3362,
+ 0x21CF367F, 0x841C6AB4, 0x2A6EFE32, 0x8FBDA2F9,
+ 0xC10D220B, 0x64DE7EC0, 0xCAACEA46, 0x6F7FB68D,
+ 0xD74FB390, 0x729CEF5B, 0xDCEE7BDD, 0x793D2716,
+ 0xAC8F71E7, 0x095C2D2C, 0xA72EB9AA, 0x02FDE561,
+ 0xBACDE07C, 0x1F1EBCB7, 0xB16C2831, 0x14BF74FA,
+ 0xD814B01E, 0x7DC7ECD5, 0xD3B57853, 0x76662498,
+ 0xCE562185, 0x6B857D4E, 0xC5F7E9C8, 0x6024B503,
+ 0xB596E3F2, 0x1045BF39, 0xBE372BBF, 0x1BE47774,
+ 0xA3D47269, 0x06072EA2, 0xA875BA24, 0x0DA6E6EF,
+ 0x4316661D, 0xE6C53AD6, 0x48B7AE50, 0xED64F29B,
+ 0x5554F786, 0xF087AB4D, 0x5EF53FCB, 0xFB266300,
+ 0x2E9435F1, 0x8B47693A, 0x2535FDBC, 0x80E6A177,
+ 0x38D6A46A, 0x9D05F8A1, 0x33776C27, 0x96A430EC,
+ 0xEE111C19, 0x4BC240D2, 0xE5B0D454, 0x4063889F,
+ 0xF8538D82, 0x5D80D149, 0xF3F245CF, 0x56211904,
+ 0x83934FF5, 0x2640133E, 0x883287B8, 0x2DE1DB73,
+ 0x95D1DE6E, 0x300282A5, 0x9E701623, 0x3BA34AE8,
+ 0x7513CA1A, 0xD0C096D1, 0x7EB20257, 0xDB615E9C,
+ 0x63515B81, 0xC682074A, 0x68F093CC, 0xCD23CF07,
+ 0x189199F6, 0xBD42C53D, 0x133051BB, 0xB6E30D70,
+ 0x0ED3086D, 0xAB0054A6, 0x0572C020, 0xA0A19CEB,
+ 0xB41EE811, 0x11CDB4DA, 0xBFBF205C, 0x1A6C7C97,
+ 0xA25C798A, 0x078F2541, 0xA9FDB1C7, 0x0C2EED0C,
+ 0xD99CBBFD, 0x7C4FE736, 0xD23D73B0, 0x77EE2F7B,
+ 0xCFDE2A66, 0x6A0D76AD, 0xC47FE22B, 0x61ACBEE0,
+ 0x2F1C3E12, 0x8ACF62D9, 0x24BDF65F, 0x816EAA94,
+ 0x395EAF89, 0x9C8DF342, 0x32FF67C4, 0x972C3B0F,
+ 0x429E6DFE, 0xE74D3135, 0x493FA5B3, 0xECECF978,
+ 0x54DCFC65, 0xF10FA0AE, 0x5F7D3428, 0xFAAE68E3,
+ 0x821B4416, 0x27C818DD, 0x89BA8C5B, 0x2C69D090,
+ 0x9459D58D, 0x318A8946, 0x9FF81DC0, 0x3A2B410B,
+ 0xEF9917FA, 0x4A4A4B31, 0xE438DFB7, 0x41EB837C,
+ 0xF9DB8661, 0x5C08DAAA, 0xF27A4E2C, 0x57A912E7,
+ 0x19199215, 0xBCCACEDE, 0x12B85A58, 0xB76B0693,
+ 0x0F5B038E, 0xAA885F45, 0x04FACBC3, 0xA1299708,
+ 0x749BC1F9, 0xD1489D32, 0x7F3A09B4, 0xDAE9557F,
+ 0x62D95062, 0xC70A0CA9, 0x6978982F, 0xCCABC4E4
+ }, {
+ 0x00000000, 0xB40B77A6, 0x29119F97, 0x9D1AE831,
+ 0x13244FF4, 0xA72F3852, 0x3A35D063, 0x8E3EA7C5,
+ 0x674EEF33, 0xD3459895, 0x4E5F70A4, 0xFA540702,
+ 0x746AA0C7, 0xC061D761, 0x5D7B3F50, 0xE97048F6,
+ 0xCE9CDE67, 0x7A97A9C1, 0xE78D41F0, 0x53863656,
+ 0xDDB89193, 0x69B3E635, 0xF4A90E04, 0x40A279A2,
+ 0xA9D23154, 0x1DD946F2, 0x80C3AEC3, 0x34C8D965,
+ 0xBAF67EA0, 0x0EFD0906, 0x93E7E137, 0x27EC9691,
+ 0x9C39BDCF, 0x2832CA69, 0xB5282258, 0x012355FE,
+ 0x8F1DF23B, 0x3B16859D, 0xA60C6DAC, 0x12071A0A,
+ 0xFB7752FC, 0x4F7C255A, 0xD266CD6B, 0x666DBACD,
+ 0xE8531D08, 0x5C586AAE, 0xC142829F, 0x7549F539,
+ 0x52A563A8, 0xE6AE140E, 0x7BB4FC3F, 0xCFBF8B99,
+ 0x41812C5C, 0xF58A5BFA, 0x6890B3CB, 0xDC9BC46D,
+ 0x35EB8C9B, 0x81E0FB3D, 0x1CFA130C, 0xA8F164AA,
+ 0x26CFC36F, 0x92C4B4C9, 0x0FDE5CF8, 0xBBD52B5E,
+ 0x79750B44, 0xCD7E7CE2, 0x506494D3, 0xE46FE375,
+ 0x6A5144B0, 0xDE5A3316, 0x4340DB27, 0xF74BAC81,
+ 0x1E3BE477, 0xAA3093D1, 0x372A7BE0, 0x83210C46,
+ 0x0D1FAB83, 0xB914DC25, 0x240E3414, 0x900543B2,
+ 0xB7E9D523, 0x03E2A285, 0x9EF84AB4, 0x2AF33D12,
+ 0xA4CD9AD7, 0x10C6ED71, 0x8DDC0540, 0x39D772E6,
+ 0xD0A73A10, 0x64AC4DB6, 0xF9B6A587, 0x4DBDD221,
+ 0xC38375E4, 0x77880242, 0xEA92EA73, 0x5E999DD5,
+ 0xE54CB68B, 0x5147C12D, 0xCC5D291C, 0x78565EBA,
+ 0xF668F97F, 0x42638ED9, 0xDF7966E8, 0x6B72114E,
+ 0x820259B8, 0x36092E1E, 0xAB13C62F, 0x1F18B189,
+ 0x9126164C, 0x252D61EA, 0xB83789DB, 0x0C3CFE7D,
+ 0x2BD068EC, 0x9FDB1F4A, 0x02C1F77B, 0xB6CA80DD,
+ 0x38F42718, 0x8CFF50BE, 0x11E5B88F, 0xA5EECF29,
+ 0x4C9E87DF, 0xF895F079, 0x658F1848, 0xD1846FEE,
+ 0x5FBAC82B, 0xEBB1BF8D, 0x76AB57BC, 0xC2A0201A,
+ 0xF2EA1688, 0x46E1612E, 0xDBFB891F, 0x6FF0FEB9,
+ 0xE1CE597C, 0x55C52EDA, 0xC8DFC6EB, 0x7CD4B14D,
+ 0x95A4F9BB, 0x21AF8E1D, 0xBCB5662C, 0x08BE118A,
+ 0x8680B64F, 0x328BC1E9, 0xAF9129D8, 0x1B9A5E7E,
+ 0x3C76C8EF, 0x887DBF49, 0x15675778, 0xA16C20DE,
+ 0x2F52871B, 0x9B59F0BD, 0x0643188C, 0xB2486F2A,
+ 0x5B3827DC, 0xEF33507A, 0x7229B84B, 0xC622CFED,
+ 0x481C6828, 0xFC171F8E, 0x610DF7BF, 0xD5068019,
+ 0x6ED3AB47, 0xDAD8DCE1, 0x47C234D0, 0xF3C94376,
+ 0x7DF7E4B3, 0xC9FC9315, 0x54E67B24, 0xE0ED0C82,
+ 0x099D4474, 0xBD9633D2, 0x208CDBE3, 0x9487AC45,
+ 0x1AB90B80, 0xAEB27C26, 0x33A89417, 0x87A3E3B1,
+ 0xA04F7520, 0x14440286, 0x895EEAB7, 0x3D559D11,
+ 0xB36B3AD4, 0x07604D72, 0x9A7AA543, 0x2E71D2E5,
+ 0xC7019A13, 0x730AEDB5, 0xEE100584, 0x5A1B7222,
+ 0xD425D5E7, 0x602EA241, 0xFD344A70, 0x493F3DD6,
+ 0x8B9F1DCC, 0x3F946A6A, 0xA28E825B, 0x1685F5FD,
+ 0x98BB5238, 0x2CB0259E, 0xB1AACDAF, 0x05A1BA09,
+ 0xECD1F2FF, 0x58DA8559, 0xC5C06D68, 0x71CB1ACE,
+ 0xFFF5BD0B, 0x4BFECAAD, 0xD6E4229C, 0x62EF553A,
+ 0x4503C3AB, 0xF108B40D, 0x6C125C3C, 0xD8192B9A,
+ 0x56278C5F, 0xE22CFBF9, 0x7F3613C8, 0xCB3D646E,
+ 0x224D2C98, 0x96465B3E, 0x0B5CB30F, 0xBF57C4A9,
+ 0x3169636C, 0x856214CA, 0x1878FCFB, 0xAC738B5D,
+ 0x17A6A003, 0xA3ADD7A5, 0x3EB73F94, 0x8ABC4832,
+ 0x0482EFF7, 0xB0899851, 0x2D937060, 0x999807C6,
+ 0x70E84F30, 0xC4E33896, 0x59F9D0A7, 0xEDF2A701,
+ 0x63CC00C4, 0xD7C77762, 0x4ADD9F53, 0xFED6E8F5,
+ 0xD93A7E64, 0x6D3109C2, 0xF02BE1F3, 0x44209655,
+ 0xCA1E3190, 0x7E154636, 0xE30FAE07, 0x5704D9A1,
+ 0xBE749157, 0x0A7FE6F1, 0x97650EC0, 0x236E7966,
+ 0xAD50DEA3, 0x195BA905, 0x84414134, 0x304A3692
+ }, {
+ 0x00000000, 0x9E00AACC, 0x7D072542, 0xE3078F8E,
+ 0xFA0E4A84, 0x640EE048, 0x87096FC6, 0x1909C50A,
+ 0xB51BE5D3, 0x2B1B4F1F, 0xC81CC091, 0x561C6A5D,
+ 0x4F15AF57, 0xD115059B, 0x32128A15, 0xAC1220D9,
+ 0x2B31BB7C, 0xB53111B0, 0x56369E3E, 0xC83634F2,
+ 0xD13FF1F8, 0x4F3F5B34, 0xAC38D4BA, 0x32387E76,
+ 0x9E2A5EAF, 0x002AF463, 0xE32D7BED, 0x7D2DD121,
+ 0x6424142B, 0xFA24BEE7, 0x19233169, 0x87239BA5,
+ 0x566276F9, 0xC862DC35, 0x2B6553BB, 0xB565F977,
+ 0xAC6C3C7D, 0x326C96B1, 0xD16B193F, 0x4F6BB3F3,
+ 0xE379932A, 0x7D7939E6, 0x9E7EB668, 0x007E1CA4,
+ 0x1977D9AE, 0x87777362, 0x6470FCEC, 0xFA705620,
+ 0x7D53CD85, 0xE3536749, 0x0054E8C7, 0x9E54420B,
+ 0x875D8701, 0x195D2DCD, 0xFA5AA243, 0x645A088F,
+ 0xC8482856, 0x5648829A, 0xB54F0D14, 0x2B4FA7D8,
+ 0x324662D2, 0xAC46C81E, 0x4F414790, 0xD141ED5C,
+ 0xEDC29D29, 0x73C237E5, 0x90C5B86B, 0x0EC512A7,
+ 0x17CCD7AD, 0x89CC7D61, 0x6ACBF2EF, 0xF4CB5823,
+ 0x58D978FA, 0xC6D9D236, 0x25DE5DB8, 0xBBDEF774,
+ 0xA2D7327E, 0x3CD798B2, 0xDFD0173C, 0x41D0BDF0,
+ 0xC6F32655, 0x58F38C99, 0xBBF40317, 0x25F4A9DB,
+ 0x3CFD6CD1, 0xA2FDC61D, 0x41FA4993, 0xDFFAE35F,
+ 0x73E8C386, 0xEDE8694A, 0x0EEFE6C4, 0x90EF4C08,
+ 0x89E68902, 0x17E623CE, 0xF4E1AC40, 0x6AE1068C,
+ 0xBBA0EBD0, 0x25A0411C, 0xC6A7CE92, 0x58A7645E,
+ 0x41AEA154, 0xDFAE0B98, 0x3CA98416, 0xA2A92EDA,
+ 0x0EBB0E03, 0x90BBA4CF, 0x73BC2B41, 0xEDBC818D,
+ 0xF4B54487, 0x6AB5EE4B, 0x89B261C5, 0x17B2CB09,
+ 0x909150AC, 0x0E91FA60, 0xED9675EE, 0x7396DF22,
+ 0x6A9F1A28, 0xF49FB0E4, 0x17983F6A, 0x899895A6,
+ 0x258AB57F, 0xBB8A1FB3, 0x588D903D, 0xC68D3AF1,
+ 0xDF84FFFB, 0x41845537, 0xA283DAB9, 0x3C837075,
+ 0xDA853B53, 0x4485919F, 0xA7821E11, 0x3982B4DD,
+ 0x208B71D7, 0xBE8BDB1B, 0x5D8C5495, 0xC38CFE59,
+ 0x6F9EDE80, 0xF19E744C, 0x1299FBC2, 0x8C99510E,
+ 0x95909404, 0x0B903EC8, 0xE897B146, 0x76971B8A,
+ 0xF1B4802F, 0x6FB42AE3, 0x8CB3A56D, 0x12B30FA1,
+ 0x0BBACAAB, 0x95BA6067, 0x76BDEFE9, 0xE8BD4525,
+ 0x44AF65FC, 0xDAAFCF30, 0x39A840BE, 0xA7A8EA72,
+ 0xBEA12F78, 0x20A185B4, 0xC3A60A3A, 0x5DA6A0F6,
+ 0x8CE74DAA, 0x12E7E766, 0xF1E068E8, 0x6FE0C224,
+ 0x76E9072E, 0xE8E9ADE2, 0x0BEE226C, 0x95EE88A0,
+ 0x39FCA879, 0xA7FC02B5, 0x44FB8D3B, 0xDAFB27F7,
+ 0xC3F2E2FD, 0x5DF24831, 0xBEF5C7BF, 0x20F56D73,
+ 0xA7D6F6D6, 0x39D65C1A, 0xDAD1D394, 0x44D17958,
+ 0x5DD8BC52, 0xC3D8169E, 0x20DF9910, 0xBEDF33DC,
+ 0x12CD1305, 0x8CCDB9C9, 0x6FCA3647, 0xF1CA9C8B,
+ 0xE8C35981, 0x76C3F34D, 0x95C47CC3, 0x0BC4D60F,
+ 0x3747A67A, 0xA9470CB6, 0x4A408338, 0xD44029F4,
+ 0xCD49ECFE, 0x53494632, 0xB04EC9BC, 0x2E4E6370,
+ 0x825C43A9, 0x1C5CE965, 0xFF5B66EB, 0x615BCC27,
+ 0x7852092D, 0xE652A3E1, 0x05552C6F, 0x9B5586A3,
+ 0x1C761D06, 0x8276B7CA, 0x61713844, 0xFF719288,
+ 0xE6785782, 0x7878FD4E, 0x9B7F72C0, 0x057FD80C,
+ 0xA96DF8D5, 0x376D5219, 0xD46ADD97, 0x4A6A775B,
+ 0x5363B251, 0xCD63189D, 0x2E649713, 0xB0643DDF,
+ 0x6125D083, 0xFF257A4F, 0x1C22F5C1, 0x82225F0D,
+ 0x9B2B9A07, 0x052B30CB, 0xE62CBF45, 0x782C1589,
+ 0xD43E3550, 0x4A3E9F9C, 0xA9391012, 0x3739BADE,
+ 0x2E307FD4, 0xB030D518, 0x53375A96, 0xCD37F05A,
+ 0x4A146BFF, 0xD414C133, 0x37134EBD, 0xA913E471,
+ 0xB01A217B, 0x2E1A8BB7, 0xCD1D0439, 0x531DAEF5,
+ 0xFF0F8E2C, 0x610F24E0, 0x8208AB6E, 0x1C0801A2,
+ 0x0501C4A8, 0x9B016E64, 0x7806E1EA, 0xE6064B26
+ }
+};
diff --git a/src/liblzma/check/crc32_table_le.h b/src/liblzma/check/crc32_table_le.h
new file mode 100644
index 0000000..25f4fc4
--- /dev/null
+++ b/src/liblzma/check/crc32_table_le.h
@@ -0,0 +1,525 @@
+/* This file has been automatically generated by crc32_tablegen.c. */
+
+const uint32_t lzma_crc32_table[8][256] = {
+ {
+ 0x00000000, 0x77073096, 0xEE0E612C, 0x990951BA,
+ 0x076DC419, 0x706AF48F, 0xE963A535, 0x9E6495A3,
+ 0x0EDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988,
+ 0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91,
+ 0x1DB71064, 0x6AB020F2, 0xF3B97148, 0x84BE41DE,
+ 0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7,
+ 0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC,
+ 0x14015C4F, 0x63066CD9, 0xFA0F3D63, 0x8D080DF5,
+ 0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172,
+ 0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B,
+ 0x35B5A8FA, 0x42B2986C, 0xDBBBC9D6, 0xACBCF940,
+ 0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59,
+ 0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116,
+ 0x21B4F4B5, 0x56B3C423, 0xCFBA9599, 0xB8BDA50F,
+ 0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924,
+ 0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D,
+ 0x76DC4190, 0x01DB7106, 0x98D220BC, 0xEFD5102A,
+ 0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433,
+ 0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818,
+ 0x7F6A0DBB, 0x086D3D2D, 0x91646C97, 0xE6635C01,
+ 0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E,
+ 0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457,
+ 0x65B0D9C6, 0x12B7E950, 0x8BBEB8EA, 0xFCB9887C,
+ 0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65,
+ 0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2,
+ 0x4ADFA541, 0x3DD895D7, 0xA4D1C46D, 0xD3D6F4FB,
+ 0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0,
+ 0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7CC9,
+ 0x5005713C, 0x270241AA, 0xBE0B1010, 0xC90C2086,
+ 0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F,
+ 0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4,
+ 0x59B33D17, 0x2EB40D81, 0xB7BD5C3B, 0xC0BA6CAD,
+ 0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A,
+ 0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683,
+ 0xE3630B12, 0x94643B84, 0x0D6D6A3E, 0x7A6A5AA8,
+ 0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1,
+ 0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE,
+ 0xF762575D, 0x806567CB, 0x196C3671, 0x6E6B06E7,
+ 0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC,
+ 0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5,
+ 0xD6D6A3E8, 0xA1D1937E, 0x38D8C2C4, 0x4FDFF252,
+ 0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B,
+ 0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60,
+ 0xDF60EFC3, 0xA867DF55, 0x316E8EEF, 0x4669BE79,
+ 0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236,
+ 0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F,
+ 0xC5BA3BBE, 0xB2BD0B28, 0x2BB45A92, 0x5CB36A04,
+ 0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D,
+ 0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A,
+ 0x9C0906A9, 0xEB0E363F, 0x72076785, 0x05005713,
+ 0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38,
+ 0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21,
+ 0x86D3D2D4, 0xF1D4E242, 0x68DDB3F8, 0x1FDA836E,
+ 0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777,
+ 0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C,
+ 0x8F659EFF, 0xF862AE69, 0x616BFFD3, 0x166CCF45,
+ 0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2,
+ 0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB,
+ 0xAED16A4A, 0xD9D65ADC, 0x40DF0B66, 0x37D83BF0,
+ 0xA9BCAE53, 0xDEBB9EC5, 0x47B2CF7F, 0x30B5FFE9,
+ 0xBDBDF21C, 0xCABAC28A, 0x53B39330, 0x24B4A3A6,
+ 0xBAD03605, 0xCDD70693, 0x54DE5729, 0x23D967BF,
+ 0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94,
+ 0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D
+ }, {
+ 0x00000000, 0x191B3141, 0x32366282, 0x2B2D53C3,
+ 0x646CC504, 0x7D77F445, 0x565AA786, 0x4F4196C7,
+ 0xC8D98A08, 0xD1C2BB49, 0xFAEFE88A, 0xE3F4D9CB,
+ 0xACB54F0C, 0xB5AE7E4D, 0x9E832D8E, 0x87981CCF,
+ 0x4AC21251, 0x53D92310, 0x78F470D3, 0x61EF4192,
+ 0x2EAED755, 0x37B5E614, 0x1C98B5D7, 0x05838496,
+ 0x821B9859, 0x9B00A918, 0xB02DFADB, 0xA936CB9A,
+ 0xE6775D5D, 0xFF6C6C1C, 0xD4413FDF, 0xCD5A0E9E,
+ 0x958424A2, 0x8C9F15E3, 0xA7B24620, 0xBEA97761,
+ 0xF1E8E1A6, 0xE8F3D0E7, 0xC3DE8324, 0xDAC5B265,
+ 0x5D5DAEAA, 0x44469FEB, 0x6F6BCC28, 0x7670FD69,
+ 0x39316BAE, 0x202A5AEF, 0x0B07092C, 0x121C386D,
+ 0xDF4636F3, 0xC65D07B2, 0xED705471, 0xF46B6530,
+ 0xBB2AF3F7, 0xA231C2B6, 0x891C9175, 0x9007A034,
+ 0x179FBCFB, 0x0E848DBA, 0x25A9DE79, 0x3CB2EF38,
+ 0x73F379FF, 0x6AE848BE, 0x41C51B7D, 0x58DE2A3C,
+ 0xF0794F05, 0xE9627E44, 0xC24F2D87, 0xDB541CC6,
+ 0x94158A01, 0x8D0EBB40, 0xA623E883, 0xBF38D9C2,
+ 0x38A0C50D, 0x21BBF44C, 0x0A96A78F, 0x138D96CE,
+ 0x5CCC0009, 0x45D73148, 0x6EFA628B, 0x77E153CA,
+ 0xBABB5D54, 0xA3A06C15, 0x888D3FD6, 0x91960E97,
+ 0xDED79850, 0xC7CCA911, 0xECE1FAD2, 0xF5FACB93,
+ 0x7262D75C, 0x6B79E61D, 0x4054B5DE, 0x594F849F,
+ 0x160E1258, 0x0F152319, 0x243870DA, 0x3D23419B,
+ 0x65FD6BA7, 0x7CE65AE6, 0x57CB0925, 0x4ED03864,
+ 0x0191AEA3, 0x188A9FE2, 0x33A7CC21, 0x2ABCFD60,
+ 0xAD24E1AF, 0xB43FD0EE, 0x9F12832D, 0x8609B26C,
+ 0xC94824AB, 0xD05315EA, 0xFB7E4629, 0xE2657768,
+ 0x2F3F79F6, 0x362448B7, 0x1D091B74, 0x04122A35,
+ 0x4B53BCF2, 0x52488DB3, 0x7965DE70, 0x607EEF31,
+ 0xE7E6F3FE, 0xFEFDC2BF, 0xD5D0917C, 0xCCCBA03D,
+ 0x838A36FA, 0x9A9107BB, 0xB1BC5478, 0xA8A76539,
+ 0x3B83984B, 0x2298A90A, 0x09B5FAC9, 0x10AECB88,
+ 0x5FEF5D4F, 0x46F46C0E, 0x6DD93FCD, 0x74C20E8C,
+ 0xF35A1243, 0xEA412302, 0xC16C70C1, 0xD8774180,
+ 0x9736D747, 0x8E2DE606, 0xA500B5C5, 0xBC1B8484,
+ 0x71418A1A, 0x685ABB5B, 0x4377E898, 0x5A6CD9D9,
+ 0x152D4F1E, 0x0C367E5F, 0x271B2D9C, 0x3E001CDD,
+ 0xB9980012, 0xA0833153, 0x8BAE6290, 0x92B553D1,
+ 0xDDF4C516, 0xC4EFF457, 0xEFC2A794, 0xF6D996D5,
+ 0xAE07BCE9, 0xB71C8DA8, 0x9C31DE6B, 0x852AEF2A,
+ 0xCA6B79ED, 0xD37048AC, 0xF85D1B6F, 0xE1462A2E,
+ 0x66DE36E1, 0x7FC507A0, 0x54E85463, 0x4DF36522,
+ 0x02B2F3E5, 0x1BA9C2A4, 0x30849167, 0x299FA026,
+ 0xE4C5AEB8, 0xFDDE9FF9, 0xD6F3CC3A, 0xCFE8FD7B,
+ 0x80A96BBC, 0x99B25AFD, 0xB29F093E, 0xAB84387F,
+ 0x2C1C24B0, 0x350715F1, 0x1E2A4632, 0x07317773,
+ 0x4870E1B4, 0x516BD0F5, 0x7A468336, 0x635DB277,
+ 0xCBFAD74E, 0xD2E1E60F, 0xF9CCB5CC, 0xE0D7848D,
+ 0xAF96124A, 0xB68D230B, 0x9DA070C8, 0x84BB4189,
+ 0x03235D46, 0x1A386C07, 0x31153FC4, 0x280E0E85,
+ 0x674F9842, 0x7E54A903, 0x5579FAC0, 0x4C62CB81,
+ 0x8138C51F, 0x9823F45E, 0xB30EA79D, 0xAA1596DC,
+ 0xE554001B, 0xFC4F315A, 0xD7626299, 0xCE7953D8,
+ 0x49E14F17, 0x50FA7E56, 0x7BD72D95, 0x62CC1CD4,
+ 0x2D8D8A13, 0x3496BB52, 0x1FBBE891, 0x06A0D9D0,
+ 0x5E7EF3EC, 0x4765C2AD, 0x6C48916E, 0x7553A02F,
+ 0x3A1236E8, 0x230907A9, 0x0824546A, 0x113F652B,
+ 0x96A779E4, 0x8FBC48A5, 0xA4911B66, 0xBD8A2A27,
+ 0xF2CBBCE0, 0xEBD08DA1, 0xC0FDDE62, 0xD9E6EF23,
+ 0x14BCE1BD, 0x0DA7D0FC, 0x268A833F, 0x3F91B27E,
+ 0x70D024B9, 0x69CB15F8, 0x42E6463B, 0x5BFD777A,
+ 0xDC656BB5, 0xC57E5AF4, 0xEE530937, 0xF7483876,
+ 0xB809AEB1, 0xA1129FF0, 0x8A3FCC33, 0x9324FD72
+ }, {
+ 0x00000000, 0x01C26A37, 0x0384D46E, 0x0246BE59,
+ 0x0709A8DC, 0x06CBC2EB, 0x048D7CB2, 0x054F1685,
+ 0x0E1351B8, 0x0FD13B8F, 0x0D9785D6, 0x0C55EFE1,
+ 0x091AF964, 0x08D89353, 0x0A9E2D0A, 0x0B5C473D,
+ 0x1C26A370, 0x1DE4C947, 0x1FA2771E, 0x1E601D29,
+ 0x1B2F0BAC, 0x1AED619B, 0x18ABDFC2, 0x1969B5F5,
+ 0x1235F2C8, 0x13F798FF, 0x11B126A6, 0x10734C91,
+ 0x153C5A14, 0x14FE3023, 0x16B88E7A, 0x177AE44D,
+ 0x384D46E0, 0x398F2CD7, 0x3BC9928E, 0x3A0BF8B9,
+ 0x3F44EE3C, 0x3E86840B, 0x3CC03A52, 0x3D025065,
+ 0x365E1758, 0x379C7D6F, 0x35DAC336, 0x3418A901,
+ 0x3157BF84, 0x3095D5B3, 0x32D36BEA, 0x331101DD,
+ 0x246BE590, 0x25A98FA7, 0x27EF31FE, 0x262D5BC9,
+ 0x23624D4C, 0x22A0277B, 0x20E69922, 0x2124F315,
+ 0x2A78B428, 0x2BBADE1F, 0x29FC6046, 0x283E0A71,
+ 0x2D711CF4, 0x2CB376C3, 0x2EF5C89A, 0x2F37A2AD,
+ 0x709A8DC0, 0x7158E7F7, 0x731E59AE, 0x72DC3399,
+ 0x7793251C, 0x76514F2B, 0x7417F172, 0x75D59B45,
+ 0x7E89DC78, 0x7F4BB64F, 0x7D0D0816, 0x7CCF6221,
+ 0x798074A4, 0x78421E93, 0x7A04A0CA, 0x7BC6CAFD,
+ 0x6CBC2EB0, 0x6D7E4487, 0x6F38FADE, 0x6EFA90E9,
+ 0x6BB5866C, 0x6A77EC5B, 0x68315202, 0x69F33835,
+ 0x62AF7F08, 0x636D153F, 0x612BAB66, 0x60E9C151,
+ 0x65A6D7D4, 0x6464BDE3, 0x662203BA, 0x67E0698D,
+ 0x48D7CB20, 0x4915A117, 0x4B531F4E, 0x4A917579,
+ 0x4FDE63FC, 0x4E1C09CB, 0x4C5AB792, 0x4D98DDA5,
+ 0x46C49A98, 0x4706F0AF, 0x45404EF6, 0x448224C1,
+ 0x41CD3244, 0x400F5873, 0x4249E62A, 0x438B8C1D,
+ 0x54F16850, 0x55330267, 0x5775BC3E, 0x56B7D609,
+ 0x53F8C08C, 0x523AAABB, 0x507C14E2, 0x51BE7ED5,
+ 0x5AE239E8, 0x5B2053DF, 0x5966ED86, 0x58A487B1,
+ 0x5DEB9134, 0x5C29FB03, 0x5E6F455A, 0x5FAD2F6D,
+ 0xE1351B80, 0xE0F771B7, 0xE2B1CFEE, 0xE373A5D9,
+ 0xE63CB35C, 0xE7FED96B, 0xE5B86732, 0xE47A0D05,
+ 0xEF264A38, 0xEEE4200F, 0xECA29E56, 0xED60F461,
+ 0xE82FE2E4, 0xE9ED88D3, 0xEBAB368A, 0xEA695CBD,
+ 0xFD13B8F0, 0xFCD1D2C7, 0xFE976C9E, 0xFF5506A9,
+ 0xFA1A102C, 0xFBD87A1B, 0xF99EC442, 0xF85CAE75,
+ 0xF300E948, 0xF2C2837F, 0xF0843D26, 0xF1465711,
+ 0xF4094194, 0xF5CB2BA3, 0xF78D95FA, 0xF64FFFCD,
+ 0xD9785D60, 0xD8BA3757, 0xDAFC890E, 0xDB3EE339,
+ 0xDE71F5BC, 0xDFB39F8B, 0xDDF521D2, 0xDC374BE5,
+ 0xD76B0CD8, 0xD6A966EF, 0xD4EFD8B6, 0xD52DB281,
+ 0xD062A404, 0xD1A0CE33, 0xD3E6706A, 0xD2241A5D,
+ 0xC55EFE10, 0xC49C9427, 0xC6DA2A7E, 0xC7184049,
+ 0xC25756CC, 0xC3953CFB, 0xC1D382A2, 0xC011E895,
+ 0xCB4DAFA8, 0xCA8FC59F, 0xC8C97BC6, 0xC90B11F1,
+ 0xCC440774, 0xCD866D43, 0xCFC0D31A, 0xCE02B92D,
+ 0x91AF9640, 0x906DFC77, 0x922B422E, 0x93E92819,
+ 0x96A63E9C, 0x976454AB, 0x9522EAF2, 0x94E080C5,
+ 0x9FBCC7F8, 0x9E7EADCF, 0x9C381396, 0x9DFA79A1,
+ 0x98B56F24, 0x99770513, 0x9B31BB4A, 0x9AF3D17D,
+ 0x8D893530, 0x8C4B5F07, 0x8E0DE15E, 0x8FCF8B69,
+ 0x8A809DEC, 0x8B42F7DB, 0x89044982, 0x88C623B5,
+ 0x839A6488, 0x82580EBF, 0x801EB0E6, 0x81DCDAD1,
+ 0x8493CC54, 0x8551A663, 0x8717183A, 0x86D5720D,
+ 0xA9E2D0A0, 0xA820BA97, 0xAA6604CE, 0xABA46EF9,
+ 0xAEEB787C, 0xAF29124B, 0xAD6FAC12, 0xACADC625,
+ 0xA7F18118, 0xA633EB2F, 0xA4755576, 0xA5B73F41,
+ 0xA0F829C4, 0xA13A43F3, 0xA37CFDAA, 0xA2BE979D,
+ 0xB5C473D0, 0xB40619E7, 0xB640A7BE, 0xB782CD89,
+ 0xB2CDDB0C, 0xB30FB13B, 0xB1490F62, 0xB08B6555,
+ 0xBBD72268, 0xBA15485F, 0xB853F606, 0xB9919C31,
+ 0xBCDE8AB4, 0xBD1CE083, 0xBF5A5EDA, 0xBE9834ED
+ }, {
+ 0x00000000, 0xB8BC6765, 0xAA09C88B, 0x12B5AFEE,
+ 0x8F629757, 0x37DEF032, 0x256B5FDC, 0x9DD738B9,
+ 0xC5B428EF, 0x7D084F8A, 0x6FBDE064, 0xD7018701,
+ 0x4AD6BFB8, 0xF26AD8DD, 0xE0DF7733, 0x58631056,
+ 0x5019579F, 0xE8A530FA, 0xFA109F14, 0x42ACF871,
+ 0xDF7BC0C8, 0x67C7A7AD, 0x75720843, 0xCDCE6F26,
+ 0x95AD7F70, 0x2D111815, 0x3FA4B7FB, 0x8718D09E,
+ 0x1ACFE827, 0xA2738F42, 0xB0C620AC, 0x087A47C9,
+ 0xA032AF3E, 0x188EC85B, 0x0A3B67B5, 0xB28700D0,
+ 0x2F503869, 0x97EC5F0C, 0x8559F0E2, 0x3DE59787,
+ 0x658687D1, 0xDD3AE0B4, 0xCF8F4F5A, 0x7733283F,
+ 0xEAE41086, 0x525877E3, 0x40EDD80D, 0xF851BF68,
+ 0xF02BF8A1, 0x48979FC4, 0x5A22302A, 0xE29E574F,
+ 0x7F496FF6, 0xC7F50893, 0xD540A77D, 0x6DFCC018,
+ 0x359FD04E, 0x8D23B72B, 0x9F9618C5, 0x272A7FA0,
+ 0xBAFD4719, 0x0241207C, 0x10F48F92, 0xA848E8F7,
+ 0x9B14583D, 0x23A83F58, 0x311D90B6, 0x89A1F7D3,
+ 0x1476CF6A, 0xACCAA80F, 0xBE7F07E1, 0x06C36084,
+ 0x5EA070D2, 0xE61C17B7, 0xF4A9B859, 0x4C15DF3C,
+ 0xD1C2E785, 0x697E80E0, 0x7BCB2F0E, 0xC377486B,
+ 0xCB0D0FA2, 0x73B168C7, 0x6104C729, 0xD9B8A04C,
+ 0x446F98F5, 0xFCD3FF90, 0xEE66507E, 0x56DA371B,
+ 0x0EB9274D, 0xB6054028, 0xA4B0EFC6, 0x1C0C88A3,
+ 0x81DBB01A, 0x3967D77F, 0x2BD27891, 0x936E1FF4,
+ 0x3B26F703, 0x839A9066, 0x912F3F88, 0x299358ED,
+ 0xB4446054, 0x0CF80731, 0x1E4DA8DF, 0xA6F1CFBA,
+ 0xFE92DFEC, 0x462EB889, 0x549B1767, 0xEC277002,
+ 0x71F048BB, 0xC94C2FDE, 0xDBF98030, 0x6345E755,
+ 0x6B3FA09C, 0xD383C7F9, 0xC1366817, 0x798A0F72,
+ 0xE45D37CB, 0x5CE150AE, 0x4E54FF40, 0xF6E89825,
+ 0xAE8B8873, 0x1637EF16, 0x048240F8, 0xBC3E279D,
+ 0x21E91F24, 0x99557841, 0x8BE0D7AF, 0x335CB0CA,
+ 0xED59B63B, 0x55E5D15E, 0x47507EB0, 0xFFEC19D5,
+ 0x623B216C, 0xDA874609, 0xC832E9E7, 0x708E8E82,
+ 0x28ED9ED4, 0x9051F9B1, 0x82E4565F, 0x3A58313A,
+ 0xA78F0983, 0x1F336EE6, 0x0D86C108, 0xB53AA66D,
+ 0xBD40E1A4, 0x05FC86C1, 0x1749292F, 0xAFF54E4A,
+ 0x322276F3, 0x8A9E1196, 0x982BBE78, 0x2097D91D,
+ 0x78F4C94B, 0xC048AE2E, 0xD2FD01C0, 0x6A4166A5,
+ 0xF7965E1C, 0x4F2A3979, 0x5D9F9697, 0xE523F1F2,
+ 0x4D6B1905, 0xF5D77E60, 0xE762D18E, 0x5FDEB6EB,
+ 0xC2098E52, 0x7AB5E937, 0x680046D9, 0xD0BC21BC,
+ 0x88DF31EA, 0x3063568F, 0x22D6F961, 0x9A6A9E04,
+ 0x07BDA6BD, 0xBF01C1D8, 0xADB46E36, 0x15080953,
+ 0x1D724E9A, 0xA5CE29FF, 0xB77B8611, 0x0FC7E174,
+ 0x9210D9CD, 0x2AACBEA8, 0x38191146, 0x80A57623,
+ 0xD8C66675, 0x607A0110, 0x72CFAEFE, 0xCA73C99B,
+ 0x57A4F122, 0xEF189647, 0xFDAD39A9, 0x45115ECC,
+ 0x764DEE06, 0xCEF18963, 0xDC44268D, 0x64F841E8,
+ 0xF92F7951, 0x41931E34, 0x5326B1DA, 0xEB9AD6BF,
+ 0xB3F9C6E9, 0x0B45A18C, 0x19F00E62, 0xA14C6907,
+ 0x3C9B51BE, 0x842736DB, 0x96929935, 0x2E2EFE50,
+ 0x2654B999, 0x9EE8DEFC, 0x8C5D7112, 0x34E11677,
+ 0xA9362ECE, 0x118A49AB, 0x033FE645, 0xBB838120,
+ 0xE3E09176, 0x5B5CF613, 0x49E959FD, 0xF1553E98,
+ 0x6C820621, 0xD43E6144, 0xC68BCEAA, 0x7E37A9CF,
+ 0xD67F4138, 0x6EC3265D, 0x7C7689B3, 0xC4CAEED6,
+ 0x591DD66F, 0xE1A1B10A, 0xF3141EE4, 0x4BA87981,
+ 0x13CB69D7, 0xAB770EB2, 0xB9C2A15C, 0x017EC639,
+ 0x9CA9FE80, 0x241599E5, 0x36A0360B, 0x8E1C516E,
+ 0x866616A7, 0x3EDA71C2, 0x2C6FDE2C, 0x94D3B949,
+ 0x090481F0, 0xB1B8E695, 0xA30D497B, 0x1BB12E1E,
+ 0x43D23E48, 0xFB6E592D, 0xE9DBF6C3, 0x516791A6,
+ 0xCCB0A91F, 0x740CCE7A, 0x66B96194, 0xDE0506F1
+ }, {
+ 0x00000000, 0x3D6029B0, 0x7AC05360, 0x47A07AD0,
+ 0xF580A6C0, 0xC8E08F70, 0x8F40F5A0, 0xB220DC10,
+ 0x30704BC1, 0x0D106271, 0x4AB018A1, 0x77D03111,
+ 0xC5F0ED01, 0xF890C4B1, 0xBF30BE61, 0x825097D1,
+ 0x60E09782, 0x5D80BE32, 0x1A20C4E2, 0x2740ED52,
+ 0x95603142, 0xA80018F2, 0xEFA06222, 0xD2C04B92,
+ 0x5090DC43, 0x6DF0F5F3, 0x2A508F23, 0x1730A693,
+ 0xA5107A83, 0x98705333, 0xDFD029E3, 0xE2B00053,
+ 0xC1C12F04, 0xFCA106B4, 0xBB017C64, 0x866155D4,
+ 0x344189C4, 0x0921A074, 0x4E81DAA4, 0x73E1F314,
+ 0xF1B164C5, 0xCCD14D75, 0x8B7137A5, 0xB6111E15,
+ 0x0431C205, 0x3951EBB5, 0x7EF19165, 0x4391B8D5,
+ 0xA121B886, 0x9C419136, 0xDBE1EBE6, 0xE681C256,
+ 0x54A11E46, 0x69C137F6, 0x2E614D26, 0x13016496,
+ 0x9151F347, 0xAC31DAF7, 0xEB91A027, 0xD6F18997,
+ 0x64D15587, 0x59B17C37, 0x1E1106E7, 0x23712F57,
+ 0x58F35849, 0x659371F9, 0x22330B29, 0x1F532299,
+ 0xAD73FE89, 0x9013D739, 0xD7B3ADE9, 0xEAD38459,
+ 0x68831388, 0x55E33A38, 0x124340E8, 0x2F236958,
+ 0x9D03B548, 0xA0639CF8, 0xE7C3E628, 0xDAA3CF98,
+ 0x3813CFCB, 0x0573E67B, 0x42D39CAB, 0x7FB3B51B,
+ 0xCD93690B, 0xF0F340BB, 0xB7533A6B, 0x8A3313DB,
+ 0x0863840A, 0x3503ADBA, 0x72A3D76A, 0x4FC3FEDA,
+ 0xFDE322CA, 0xC0830B7A, 0x872371AA, 0xBA43581A,
+ 0x9932774D, 0xA4525EFD, 0xE3F2242D, 0xDE920D9D,
+ 0x6CB2D18D, 0x51D2F83D, 0x167282ED, 0x2B12AB5D,
+ 0xA9423C8C, 0x9422153C, 0xD3826FEC, 0xEEE2465C,
+ 0x5CC29A4C, 0x61A2B3FC, 0x2602C92C, 0x1B62E09C,
+ 0xF9D2E0CF, 0xC4B2C97F, 0x8312B3AF, 0xBE729A1F,
+ 0x0C52460F, 0x31326FBF, 0x7692156F, 0x4BF23CDF,
+ 0xC9A2AB0E, 0xF4C282BE, 0xB362F86E, 0x8E02D1DE,
+ 0x3C220DCE, 0x0142247E, 0x46E25EAE, 0x7B82771E,
+ 0xB1E6B092, 0x8C869922, 0xCB26E3F2, 0xF646CA42,
+ 0x44661652, 0x79063FE2, 0x3EA64532, 0x03C66C82,
+ 0x8196FB53, 0xBCF6D2E3, 0xFB56A833, 0xC6368183,
+ 0x74165D93, 0x49767423, 0x0ED60EF3, 0x33B62743,
+ 0xD1062710, 0xEC660EA0, 0xABC67470, 0x96A65DC0,
+ 0x248681D0, 0x19E6A860, 0x5E46D2B0, 0x6326FB00,
+ 0xE1766CD1, 0xDC164561, 0x9BB63FB1, 0xA6D61601,
+ 0x14F6CA11, 0x2996E3A1, 0x6E369971, 0x5356B0C1,
+ 0x70279F96, 0x4D47B626, 0x0AE7CCF6, 0x3787E546,
+ 0x85A73956, 0xB8C710E6, 0xFF676A36, 0xC2074386,
+ 0x4057D457, 0x7D37FDE7, 0x3A978737, 0x07F7AE87,
+ 0xB5D77297, 0x88B75B27, 0xCF1721F7, 0xF2770847,
+ 0x10C70814, 0x2DA721A4, 0x6A075B74, 0x576772C4,
+ 0xE547AED4, 0xD8278764, 0x9F87FDB4, 0xA2E7D404,
+ 0x20B743D5, 0x1DD76A65, 0x5A7710B5, 0x67173905,
+ 0xD537E515, 0xE857CCA5, 0xAFF7B675, 0x92979FC5,
+ 0xE915E8DB, 0xD475C16B, 0x93D5BBBB, 0xAEB5920B,
+ 0x1C954E1B, 0x21F567AB, 0x66551D7B, 0x5B3534CB,
+ 0xD965A31A, 0xE4058AAA, 0xA3A5F07A, 0x9EC5D9CA,
+ 0x2CE505DA, 0x11852C6A, 0x562556BA, 0x6B457F0A,
+ 0x89F57F59, 0xB49556E9, 0xF3352C39, 0xCE550589,
+ 0x7C75D999, 0x4115F029, 0x06B58AF9, 0x3BD5A349,
+ 0xB9853498, 0x84E51D28, 0xC34567F8, 0xFE254E48,
+ 0x4C059258, 0x7165BBE8, 0x36C5C138, 0x0BA5E888,
+ 0x28D4C7DF, 0x15B4EE6F, 0x521494BF, 0x6F74BD0F,
+ 0xDD54611F, 0xE03448AF, 0xA794327F, 0x9AF41BCF,
+ 0x18A48C1E, 0x25C4A5AE, 0x6264DF7E, 0x5F04F6CE,
+ 0xED242ADE, 0xD044036E, 0x97E479BE, 0xAA84500E,
+ 0x4834505D, 0x755479ED, 0x32F4033D, 0x0F942A8D,
+ 0xBDB4F69D, 0x80D4DF2D, 0xC774A5FD, 0xFA148C4D,
+ 0x78441B9C, 0x4524322C, 0x028448FC, 0x3FE4614C,
+ 0x8DC4BD5C, 0xB0A494EC, 0xF704EE3C, 0xCA64C78C
+ }, {
+ 0x00000000, 0xCB5CD3A5, 0x4DC8A10B, 0x869472AE,
+ 0x9B914216, 0x50CD91B3, 0xD659E31D, 0x1D0530B8,
+ 0xEC53826D, 0x270F51C8, 0xA19B2366, 0x6AC7F0C3,
+ 0x77C2C07B, 0xBC9E13DE, 0x3A0A6170, 0xF156B2D5,
+ 0x03D6029B, 0xC88AD13E, 0x4E1EA390, 0x85427035,
+ 0x9847408D, 0x531B9328, 0xD58FE186, 0x1ED33223,
+ 0xEF8580F6, 0x24D95353, 0xA24D21FD, 0x6911F258,
+ 0x7414C2E0, 0xBF481145, 0x39DC63EB, 0xF280B04E,
+ 0x07AC0536, 0xCCF0D693, 0x4A64A43D, 0x81387798,
+ 0x9C3D4720, 0x57619485, 0xD1F5E62B, 0x1AA9358E,
+ 0xEBFF875B, 0x20A354FE, 0xA6372650, 0x6D6BF5F5,
+ 0x706EC54D, 0xBB3216E8, 0x3DA66446, 0xF6FAB7E3,
+ 0x047A07AD, 0xCF26D408, 0x49B2A6A6, 0x82EE7503,
+ 0x9FEB45BB, 0x54B7961E, 0xD223E4B0, 0x197F3715,
+ 0xE82985C0, 0x23755665, 0xA5E124CB, 0x6EBDF76E,
+ 0x73B8C7D6, 0xB8E41473, 0x3E7066DD, 0xF52CB578,
+ 0x0F580A6C, 0xC404D9C9, 0x4290AB67, 0x89CC78C2,
+ 0x94C9487A, 0x5F959BDF, 0xD901E971, 0x125D3AD4,
+ 0xE30B8801, 0x28575BA4, 0xAEC3290A, 0x659FFAAF,
+ 0x789ACA17, 0xB3C619B2, 0x35526B1C, 0xFE0EB8B9,
+ 0x0C8E08F7, 0xC7D2DB52, 0x4146A9FC, 0x8A1A7A59,
+ 0x971F4AE1, 0x5C439944, 0xDAD7EBEA, 0x118B384F,
+ 0xE0DD8A9A, 0x2B81593F, 0xAD152B91, 0x6649F834,
+ 0x7B4CC88C, 0xB0101B29, 0x36846987, 0xFDD8BA22,
+ 0x08F40F5A, 0xC3A8DCFF, 0x453CAE51, 0x8E607DF4,
+ 0x93654D4C, 0x58399EE9, 0xDEADEC47, 0x15F13FE2,
+ 0xE4A78D37, 0x2FFB5E92, 0xA96F2C3C, 0x6233FF99,
+ 0x7F36CF21, 0xB46A1C84, 0x32FE6E2A, 0xF9A2BD8F,
+ 0x0B220DC1, 0xC07EDE64, 0x46EAACCA, 0x8DB67F6F,
+ 0x90B34FD7, 0x5BEF9C72, 0xDD7BEEDC, 0x16273D79,
+ 0xE7718FAC, 0x2C2D5C09, 0xAAB92EA7, 0x61E5FD02,
+ 0x7CE0CDBA, 0xB7BC1E1F, 0x31286CB1, 0xFA74BF14,
+ 0x1EB014D8, 0xD5ECC77D, 0x5378B5D3, 0x98246676,
+ 0x852156CE, 0x4E7D856B, 0xC8E9F7C5, 0x03B52460,
+ 0xF2E396B5, 0x39BF4510, 0xBF2B37BE, 0x7477E41B,
+ 0x6972D4A3, 0xA22E0706, 0x24BA75A8, 0xEFE6A60D,
+ 0x1D661643, 0xD63AC5E6, 0x50AEB748, 0x9BF264ED,
+ 0x86F75455, 0x4DAB87F0, 0xCB3FF55E, 0x006326FB,
+ 0xF135942E, 0x3A69478B, 0xBCFD3525, 0x77A1E680,
+ 0x6AA4D638, 0xA1F8059D, 0x276C7733, 0xEC30A496,
+ 0x191C11EE, 0xD240C24B, 0x54D4B0E5, 0x9F886340,
+ 0x828D53F8, 0x49D1805D, 0xCF45F2F3, 0x04192156,
+ 0xF54F9383, 0x3E134026, 0xB8873288, 0x73DBE12D,
+ 0x6EDED195, 0xA5820230, 0x2316709E, 0xE84AA33B,
+ 0x1ACA1375, 0xD196C0D0, 0x5702B27E, 0x9C5E61DB,
+ 0x815B5163, 0x4A0782C6, 0xCC93F068, 0x07CF23CD,
+ 0xF6999118, 0x3DC542BD, 0xBB513013, 0x700DE3B6,
+ 0x6D08D30E, 0xA65400AB, 0x20C07205, 0xEB9CA1A0,
+ 0x11E81EB4, 0xDAB4CD11, 0x5C20BFBF, 0x977C6C1A,
+ 0x8A795CA2, 0x41258F07, 0xC7B1FDA9, 0x0CED2E0C,
+ 0xFDBB9CD9, 0x36E74F7C, 0xB0733DD2, 0x7B2FEE77,
+ 0x662ADECF, 0xAD760D6A, 0x2BE27FC4, 0xE0BEAC61,
+ 0x123E1C2F, 0xD962CF8A, 0x5FF6BD24, 0x94AA6E81,
+ 0x89AF5E39, 0x42F38D9C, 0xC467FF32, 0x0F3B2C97,
+ 0xFE6D9E42, 0x35314DE7, 0xB3A53F49, 0x78F9ECEC,
+ 0x65FCDC54, 0xAEA00FF1, 0x28347D5F, 0xE368AEFA,
+ 0x16441B82, 0xDD18C827, 0x5B8CBA89, 0x90D0692C,
+ 0x8DD55994, 0x46898A31, 0xC01DF89F, 0x0B412B3A,
+ 0xFA1799EF, 0x314B4A4A, 0xB7DF38E4, 0x7C83EB41,
+ 0x6186DBF9, 0xAADA085C, 0x2C4E7AF2, 0xE712A957,
+ 0x15921919, 0xDECECABC, 0x585AB812, 0x93066BB7,
+ 0x8E035B0F, 0x455F88AA, 0xC3CBFA04, 0x089729A1,
+ 0xF9C19B74, 0x329D48D1, 0xB4093A7F, 0x7F55E9DA,
+ 0x6250D962, 0xA90C0AC7, 0x2F987869, 0xE4C4ABCC
+ }, {
+ 0x00000000, 0xA6770BB4, 0x979F1129, 0x31E81A9D,
+ 0xF44F2413, 0x52382FA7, 0x63D0353A, 0xC5A73E8E,
+ 0x33EF4E67, 0x959845D3, 0xA4705F4E, 0x020754FA,
+ 0xC7A06A74, 0x61D761C0, 0x503F7B5D, 0xF64870E9,
+ 0x67DE9CCE, 0xC1A9977A, 0xF0418DE7, 0x56368653,
+ 0x9391B8DD, 0x35E6B369, 0x040EA9F4, 0xA279A240,
+ 0x5431D2A9, 0xF246D91D, 0xC3AEC380, 0x65D9C834,
+ 0xA07EF6BA, 0x0609FD0E, 0x37E1E793, 0x9196EC27,
+ 0xCFBD399C, 0x69CA3228, 0x582228B5, 0xFE552301,
+ 0x3BF21D8F, 0x9D85163B, 0xAC6D0CA6, 0x0A1A0712,
+ 0xFC5277FB, 0x5A257C4F, 0x6BCD66D2, 0xCDBA6D66,
+ 0x081D53E8, 0xAE6A585C, 0x9F8242C1, 0x39F54975,
+ 0xA863A552, 0x0E14AEE6, 0x3FFCB47B, 0x998BBFCF,
+ 0x5C2C8141, 0xFA5B8AF5, 0xCBB39068, 0x6DC49BDC,
+ 0x9B8CEB35, 0x3DFBE081, 0x0C13FA1C, 0xAA64F1A8,
+ 0x6FC3CF26, 0xC9B4C492, 0xF85CDE0F, 0x5E2BD5BB,
+ 0x440B7579, 0xE27C7ECD, 0xD3946450, 0x75E36FE4,
+ 0xB044516A, 0x16335ADE, 0x27DB4043, 0x81AC4BF7,
+ 0x77E43B1E, 0xD19330AA, 0xE07B2A37, 0x460C2183,
+ 0x83AB1F0D, 0x25DC14B9, 0x14340E24, 0xB2430590,
+ 0x23D5E9B7, 0x85A2E203, 0xB44AF89E, 0x123DF32A,
+ 0xD79ACDA4, 0x71EDC610, 0x4005DC8D, 0xE672D739,
+ 0x103AA7D0, 0xB64DAC64, 0x87A5B6F9, 0x21D2BD4D,
+ 0xE47583C3, 0x42028877, 0x73EA92EA, 0xD59D995E,
+ 0x8BB64CE5, 0x2DC14751, 0x1C295DCC, 0xBA5E5678,
+ 0x7FF968F6, 0xD98E6342, 0xE86679DF, 0x4E11726B,
+ 0xB8590282, 0x1E2E0936, 0x2FC613AB, 0x89B1181F,
+ 0x4C162691, 0xEA612D25, 0xDB8937B8, 0x7DFE3C0C,
+ 0xEC68D02B, 0x4A1FDB9F, 0x7BF7C102, 0xDD80CAB6,
+ 0x1827F438, 0xBE50FF8C, 0x8FB8E511, 0x29CFEEA5,
+ 0xDF879E4C, 0x79F095F8, 0x48188F65, 0xEE6F84D1,
+ 0x2BC8BA5F, 0x8DBFB1EB, 0xBC57AB76, 0x1A20A0C2,
+ 0x8816EAF2, 0x2E61E146, 0x1F89FBDB, 0xB9FEF06F,
+ 0x7C59CEE1, 0xDA2EC555, 0xEBC6DFC8, 0x4DB1D47C,
+ 0xBBF9A495, 0x1D8EAF21, 0x2C66B5BC, 0x8A11BE08,
+ 0x4FB68086, 0xE9C18B32, 0xD82991AF, 0x7E5E9A1B,
+ 0xEFC8763C, 0x49BF7D88, 0x78576715, 0xDE206CA1,
+ 0x1B87522F, 0xBDF0599B, 0x8C184306, 0x2A6F48B2,
+ 0xDC27385B, 0x7A5033EF, 0x4BB82972, 0xEDCF22C6,
+ 0x28681C48, 0x8E1F17FC, 0xBFF70D61, 0x198006D5,
+ 0x47ABD36E, 0xE1DCD8DA, 0xD034C247, 0x7643C9F3,
+ 0xB3E4F77D, 0x1593FCC9, 0x247BE654, 0x820CEDE0,
+ 0x74449D09, 0xD23396BD, 0xE3DB8C20, 0x45AC8794,
+ 0x800BB91A, 0x267CB2AE, 0x1794A833, 0xB1E3A387,
+ 0x20754FA0, 0x86024414, 0xB7EA5E89, 0x119D553D,
+ 0xD43A6BB3, 0x724D6007, 0x43A57A9A, 0xE5D2712E,
+ 0x139A01C7, 0xB5ED0A73, 0x840510EE, 0x22721B5A,
+ 0xE7D525D4, 0x41A22E60, 0x704A34FD, 0xD63D3F49,
+ 0xCC1D9F8B, 0x6A6A943F, 0x5B828EA2, 0xFDF58516,
+ 0x3852BB98, 0x9E25B02C, 0xAFCDAAB1, 0x09BAA105,
+ 0xFFF2D1EC, 0x5985DA58, 0x686DC0C5, 0xCE1ACB71,
+ 0x0BBDF5FF, 0xADCAFE4B, 0x9C22E4D6, 0x3A55EF62,
+ 0xABC30345, 0x0DB408F1, 0x3C5C126C, 0x9A2B19D8,
+ 0x5F8C2756, 0xF9FB2CE2, 0xC813367F, 0x6E643DCB,
+ 0x982C4D22, 0x3E5B4696, 0x0FB35C0B, 0xA9C457BF,
+ 0x6C636931, 0xCA146285, 0xFBFC7818, 0x5D8B73AC,
+ 0x03A0A617, 0xA5D7ADA3, 0x943FB73E, 0x3248BC8A,
+ 0xF7EF8204, 0x519889B0, 0x6070932D, 0xC6079899,
+ 0x304FE870, 0x9638E3C4, 0xA7D0F959, 0x01A7F2ED,
+ 0xC400CC63, 0x6277C7D7, 0x539FDD4A, 0xF5E8D6FE,
+ 0x647E3AD9, 0xC209316D, 0xF3E12BF0, 0x55962044,
+ 0x90311ECA, 0x3646157E, 0x07AE0FE3, 0xA1D90457,
+ 0x579174BE, 0xF1E67F0A, 0xC00E6597, 0x66796E23,
+ 0xA3DE50AD, 0x05A95B19, 0x34414184, 0x92364A30
+ }, {
+ 0x00000000, 0xCCAA009E, 0x4225077D, 0x8E8F07E3,
+ 0x844A0EFA, 0x48E00E64, 0xC66F0987, 0x0AC50919,
+ 0xD3E51BB5, 0x1F4F1B2B, 0x91C01CC8, 0x5D6A1C56,
+ 0x57AF154F, 0x9B0515D1, 0x158A1232, 0xD92012AC,
+ 0x7CBB312B, 0xB01131B5, 0x3E9E3656, 0xF23436C8,
+ 0xF8F13FD1, 0x345B3F4F, 0xBAD438AC, 0x767E3832,
+ 0xAF5E2A9E, 0x63F42A00, 0xED7B2DE3, 0x21D12D7D,
+ 0x2B142464, 0xE7BE24FA, 0x69312319, 0xA59B2387,
+ 0xF9766256, 0x35DC62C8, 0xBB53652B, 0x77F965B5,
+ 0x7D3C6CAC, 0xB1966C32, 0x3F196BD1, 0xF3B36B4F,
+ 0x2A9379E3, 0xE639797D, 0x68B67E9E, 0xA41C7E00,
+ 0xAED97719, 0x62737787, 0xECFC7064, 0x205670FA,
+ 0x85CD537D, 0x496753E3, 0xC7E85400, 0x0B42549E,
+ 0x01875D87, 0xCD2D5D19, 0x43A25AFA, 0x8F085A64,
+ 0x562848C8, 0x9A824856, 0x140D4FB5, 0xD8A74F2B,
+ 0xD2624632, 0x1EC846AC, 0x9047414F, 0x5CED41D1,
+ 0x299DC2ED, 0xE537C273, 0x6BB8C590, 0xA712C50E,
+ 0xADD7CC17, 0x617DCC89, 0xEFF2CB6A, 0x2358CBF4,
+ 0xFA78D958, 0x36D2D9C6, 0xB85DDE25, 0x74F7DEBB,
+ 0x7E32D7A2, 0xB298D73C, 0x3C17D0DF, 0xF0BDD041,
+ 0x5526F3C6, 0x998CF358, 0x1703F4BB, 0xDBA9F425,
+ 0xD16CFD3C, 0x1DC6FDA2, 0x9349FA41, 0x5FE3FADF,
+ 0x86C3E873, 0x4A69E8ED, 0xC4E6EF0E, 0x084CEF90,
+ 0x0289E689, 0xCE23E617, 0x40ACE1F4, 0x8C06E16A,
+ 0xD0EBA0BB, 0x1C41A025, 0x92CEA7C6, 0x5E64A758,
+ 0x54A1AE41, 0x980BAEDF, 0x1684A93C, 0xDA2EA9A2,
+ 0x030EBB0E, 0xCFA4BB90, 0x412BBC73, 0x8D81BCED,
+ 0x8744B5F4, 0x4BEEB56A, 0xC561B289, 0x09CBB217,
+ 0xAC509190, 0x60FA910E, 0xEE7596ED, 0x22DF9673,
+ 0x281A9F6A, 0xE4B09FF4, 0x6A3F9817, 0xA6959889,
+ 0x7FB58A25, 0xB31F8ABB, 0x3D908D58, 0xF13A8DC6,
+ 0xFBFF84DF, 0x37558441, 0xB9DA83A2, 0x7570833C,
+ 0x533B85DA, 0x9F918544, 0x111E82A7, 0xDDB48239,
+ 0xD7718B20, 0x1BDB8BBE, 0x95548C5D, 0x59FE8CC3,
+ 0x80DE9E6F, 0x4C749EF1, 0xC2FB9912, 0x0E51998C,
+ 0x04949095, 0xC83E900B, 0x46B197E8, 0x8A1B9776,
+ 0x2F80B4F1, 0xE32AB46F, 0x6DA5B38C, 0xA10FB312,
+ 0xABCABA0B, 0x6760BA95, 0xE9EFBD76, 0x2545BDE8,
+ 0xFC65AF44, 0x30CFAFDA, 0xBE40A839, 0x72EAA8A7,
+ 0x782FA1BE, 0xB485A120, 0x3A0AA6C3, 0xF6A0A65D,
+ 0xAA4DE78C, 0x66E7E712, 0xE868E0F1, 0x24C2E06F,
+ 0x2E07E976, 0xE2ADE9E8, 0x6C22EE0B, 0xA088EE95,
+ 0x79A8FC39, 0xB502FCA7, 0x3B8DFB44, 0xF727FBDA,
+ 0xFDE2F2C3, 0x3148F25D, 0xBFC7F5BE, 0x736DF520,
+ 0xD6F6D6A7, 0x1A5CD639, 0x94D3D1DA, 0x5879D144,
+ 0x52BCD85D, 0x9E16D8C3, 0x1099DF20, 0xDC33DFBE,
+ 0x0513CD12, 0xC9B9CD8C, 0x4736CA6F, 0x8B9CCAF1,
+ 0x8159C3E8, 0x4DF3C376, 0xC37CC495, 0x0FD6C40B,
+ 0x7AA64737, 0xB60C47A9, 0x3883404A, 0xF42940D4,
+ 0xFEEC49CD, 0x32464953, 0xBCC94EB0, 0x70634E2E,
+ 0xA9435C82, 0x65E95C1C, 0xEB665BFF, 0x27CC5B61,
+ 0x2D095278, 0xE1A352E6, 0x6F2C5505, 0xA386559B,
+ 0x061D761C, 0xCAB77682, 0x44387161, 0x889271FF,
+ 0x825778E6, 0x4EFD7878, 0xC0727F9B, 0x0CD87F05,
+ 0xD5F86DA9, 0x19526D37, 0x97DD6AD4, 0x5B776A4A,
+ 0x51B26353, 0x9D1863CD, 0x1397642E, 0xDF3D64B0,
+ 0x83D02561, 0x4F7A25FF, 0xC1F5221C, 0x0D5F2282,
+ 0x079A2B9B, 0xCB302B05, 0x45BF2CE6, 0x89152C78,
+ 0x50353ED4, 0x9C9F3E4A, 0x121039A9, 0xDEBA3937,
+ 0xD47F302E, 0x18D530B0, 0x965A3753, 0x5AF037CD,
+ 0xFF6B144A, 0x33C114D4, 0xBD4E1337, 0x71E413A9,
+ 0x7B211AB0, 0xB78B1A2E, 0x39041DCD, 0xF5AE1D53,
+ 0x2C8E0FFF, 0xE0240F61, 0x6EAB0882, 0xA201081C,
+ 0xA8C40105, 0x646E019B, 0xEAE10678, 0x264B06E6
+ }
+};
diff --git a/src/liblzma/check/crc32_tablegen.c b/src/liblzma/check/crc32_tablegen.c
new file mode 100644
index 0000000..31a4d27
--- /dev/null
+++ b/src/liblzma/check/crc32_tablegen.c
@@ -0,0 +1,117 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc32_tablegen.c
+/// \brief Generate crc32_table_le.h and crc32_table_be.h
+///
+/// Compiling: gcc -std=c99 -o crc32_tablegen crc32_tablegen.c
+/// Add -DWORDS_BIGENDIAN to generate big endian table.
+/// Add -DLZ_HASH_TABLE to generate lz_encoder_hash_table.h (little endian).
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include <stdio.h>
+#include "../../common/tuklib_integer.h"
+
+
+static uint32_t crc32_table[8][256];
+
+
+static void
+init_crc32_table(void)
+{
+ static const uint32_t poly32 = UINT32_C(0xEDB88320);
+
+ for (size_t s = 0; s < 8; ++s) {
+ for (size_t b = 0; b < 256; ++b) {
+ uint32_t r = s == 0 ? b : crc32_table[s - 1][b];
+
+ for (size_t i = 0; i < 8; ++i) {
+ if (r & 1)
+ r = (r >> 1) ^ poly32;
+ else
+ r >>= 1;
+ }
+
+ crc32_table[s][b] = r;
+ }
+ }
+
+#ifdef WORDS_BIGENDIAN
+ for (size_t s = 0; s < 8; ++s)
+ for (size_t b = 0; b < 256; ++b)
+ crc32_table[s][b] = bswap32(crc32_table[s][b]);
+#endif
+
+ return;
+}
+
+
+static void
+print_crc32_table(void)
+{
+ printf("/* This file has been automatically generated by "
+ "crc32_tablegen.c. */\n\n"
+ "const uint32_t lzma_crc32_table[8][256] = {\n\t{");
+
+ for (size_t s = 0; s < 8; ++s) {
+ for (size_t b = 0; b < 256; ++b) {
+ if ((b % 4) == 0)
+ printf("\n\t\t");
+
+ printf("0x%08" PRIX32, crc32_table[s][b]);
+
+ if (b != 255)
+ printf(",%s", (b+1) % 4 == 0 ? "" : " ");
+ }
+
+ if (s == 7)
+ printf("\n\t}\n};\n");
+ else
+ printf("\n\t}, {");
+ }
+
+ return;
+}
+
+
+static void
+print_lz_table(void)
+{
+ printf("/* This file has been automatically generated by "
+ "crc32_tablegen.c. */\n\n"
+ "const uint32_t lzma_lz_hash_table[256] = {");
+
+ for (size_t b = 0; b < 256; ++b) {
+ if ((b % 4) == 0)
+ printf("\n\t");
+
+ printf("0x%08" PRIX32, crc32_table[0][b]);
+
+ if (b != 255)
+ printf(",%s", (b+1) % 4 == 0 ? "" : " ");
+ }
+
+ printf("\n};\n");
+
+ return;
+}
+
+
+int
+main(void)
+{
+ init_crc32_table();
+
+#ifdef LZ_HASH_TABLE
+ print_lz_table();
+#else
+ print_crc32_table();
+#endif
+
+ return 0;
+}
diff --git a/src/liblzma/check/crc32_x86.S b/src/liblzma/check/crc32_x86.S
new file mode 100644
index 0000000..67f68a4
--- /dev/null
+++ b/src/liblzma/check/crc32_x86.S
@@ -0,0 +1,304 @@
+/*
+ * Speed-optimized CRC32 using slicing-by-eight algorithm
+ *
+ * This uses only i386 instructions, but it is optimized for i686 and later
+ * (including e.g. Pentium II/III/IV, Athlon XP, and Core 2). For i586
+ * (e.g. Pentium), slicing-by-four would be better, and even the C version
+ * of slicing-by-eight built with gcc -march=i586 tends to be a little bit
+ * better than this. Very few probably run this code on i586 or older x86
+ * so this shouldn't be a problem in practice.
+ *
+ * Authors: Igor Pavlov (original version)
+ * Lasse Collin (AT&T syntax, PIC support, better portability)
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * This code needs lzma_crc32_table, which can be created using the
+ * following C code:
+
+uint32_t lzma_crc32_table[8][256];
+
+void
+init_table(void)
+{
+ // IEEE-802.3
+ static const uint32_t poly32 = UINT32_C(0xEDB88320);
+
+ // Castagnoli
+ // static const uint32_t poly32 = UINT32_C(0x82F63B78);
+
+ // Koopman
+ // static const uint32_t poly32 = UINT32_C(0xEB31D82E);
+
+ for (size_t s = 0; s < 8; ++s) {
+ for (size_t b = 0; b < 256; ++b) {
+ uint32_t r = s == 0 ? b : lzma_crc32_table[s - 1][b];
+
+ for (size_t i = 0; i < 8; ++i) {
+ if (r & 1)
+ r = (r >> 1) ^ poly32;
+ else
+ r >>= 1;
+ }
+
+ lzma_crc32_table[s][b] = r;
+ }
+ }
+}
+
+ * The prototype of the CRC32 function:
+ * extern uint32_t lzma_crc32(const uint8_t *buf, size_t size, uint32_t crc);
+ */
+
+/*
+ * On some systems, the functions need to be prefixed. The prefix is
+ * usually an underscore.
+ */
+#ifndef __USER_LABEL_PREFIX__
+# define __USER_LABEL_PREFIX__
+#endif
+#define MAKE_SYM_CAT(prefix, sym) prefix ## sym
+#define MAKE_SYM(prefix, sym) MAKE_SYM_CAT(prefix, sym)
+#define LZMA_CRC32 MAKE_SYM(__USER_LABEL_PREFIX__, lzma_crc32)
+#define LZMA_CRC32_TABLE MAKE_SYM(__USER_LABEL_PREFIX__, lzma_crc32_table)
+
+/*
+ * Solaris assembler doesn't have .p2align, and Darwin uses .align
+ * differently than GNU/Linux and Solaris.
+ */
+#if defined(__APPLE__) || defined(__MSDOS__)
+# define ALIGN(pow2, abs) .align pow2
+#else
+# define ALIGN(pow2, abs) .align abs
+#endif
+
+ .text
+ .globl LZMA_CRC32
+
+#if !defined(__APPLE__) && !defined(_WIN32) && !defined(__CYGWIN__) \
+ && !defined(__MSDOS__)
+ .type LZMA_CRC32, @function
+#endif
+
+ ALIGN(4, 16)
+LZMA_CRC32:
+ /*
+ * Register usage:
+ * %eax crc
+ * %esi buf
+ * %edi size or buf + size
+ * %ebx lzma_crc32_table
+ * %ebp Table index
+ * %ecx Temporary
+ * %edx Temporary
+ */
+ pushl %ebx
+ pushl %esi
+ pushl %edi
+ pushl %ebp
+ movl 0x14(%esp), %esi /* buf */
+ movl 0x18(%esp), %edi /* size */
+ movl 0x1C(%esp), %eax /* crc */
+
+ /*
+ * Store the address of lzma_crc32_table to %ebx. This is needed to
+ * get position-independent code (PIC).
+ *
+ * The PIC macro is defined by libtool, while __PIC__ is defined
+ * by GCC but only on some systems. Testing for both makes it simpler
+ * to test this code without libtool, and keeps the code working also
+ * when built with libtool but using something else than GCC.
+ *
+ * I understood that libtool may define PIC on Windows even though
+ * the code in Windows DLLs is not PIC in sense that it is in ELF
+ * binaries, so we need a separate check to always use the non-PIC
+ * code on Windows.
+ */
+#if (!defined(PIC) && !defined(__PIC__)) \
+ || (defined(_WIN32) || defined(__CYGWIN__))
+ /* Not PIC */
+ movl $ LZMA_CRC32_TABLE, %ebx
+#elif defined(__APPLE__)
+ /* Mach-O */
+ call .L_get_pc
+.L_pic:
+ leal .L_lzma_crc32_table$non_lazy_ptr-.L_pic(%ebx), %ebx
+ movl (%ebx), %ebx
+#else
+ /* ELF */
+ call .L_get_pc
+ addl $_GLOBAL_OFFSET_TABLE_, %ebx
+ movl LZMA_CRC32_TABLE@GOT(%ebx), %ebx
+#endif
+
+ /* Complement the initial value. */
+ notl %eax
+
+ ALIGN(4, 16)
+.L_align:
+ /*
+ * Check if there is enough input to use slicing-by-eight.
+ * We need 16 bytes, because the loop pre-reads eight bytes.
+ */
+ cmpl $16, %edi
+ jb .L_rest
+
+ /* Check if we have reached alignment of eight bytes. */
+ testl $7, %esi
+ jz .L_slice
+
+ /* Calculate CRC of the next input byte. */
+ movzbl (%esi), %ebp
+ incl %esi
+ movzbl %al, %ecx
+ xorl %ecx, %ebp
+ shrl $8, %eax
+ xorl (%ebx, %ebp, 4), %eax
+ decl %edi
+ jmp .L_align
+
+ ALIGN(2, 4)
+.L_slice:
+ /*
+ * If we get here, there's at least 16 bytes of aligned input
+ * available. Make %edi multiple of eight bytes. Store the possible
+ * remainder over the "size" variable in the argument stack.
+ */
+ movl %edi, 0x18(%esp)
+ andl $-8, %edi
+ subl %edi, 0x18(%esp)
+
+ /*
+ * Let %edi be buf + size - 8 while running the main loop. This way
+ * we can compare for equality to determine when exit the loop.
+ */
+ addl %esi, %edi
+ subl $8, %edi
+
+ /* Read in the first eight aligned bytes. */
+ xorl (%esi), %eax
+ movl 4(%esi), %ecx
+ movzbl %cl, %ebp
+
+.L_loop:
+ movl 0x0C00(%ebx, %ebp, 4), %edx
+ movzbl %ch, %ebp
+ xorl 0x0800(%ebx, %ebp, 4), %edx
+ shrl $16, %ecx
+ xorl 8(%esi), %edx
+ movzbl %cl, %ebp
+ xorl 0x0400(%ebx, %ebp, 4), %edx
+ movzbl %ch, %ebp
+ xorl (%ebx, %ebp, 4), %edx
+ movzbl %al, %ebp
+
+ /*
+ * Read the next four bytes, for which the CRC is calculated
+ * on the next interation of the loop.
+ */
+ movl 12(%esi), %ecx
+
+ xorl 0x1C00(%ebx, %ebp, 4), %edx
+ movzbl %ah, %ebp
+ shrl $16, %eax
+ xorl 0x1800(%ebx, %ebp, 4), %edx
+ movzbl %ah, %ebp
+ movzbl %al, %eax
+ movl 0x1400(%ebx, %eax, 4), %eax
+ addl $8, %esi
+ xorl %edx, %eax
+ xorl 0x1000(%ebx, %ebp, 4), %eax
+
+ /* Check for end of aligned input. */
+ cmpl %edi, %esi
+ movzbl %cl, %ebp
+ jne .L_loop
+
+ /*
+ * Process the remaining eight bytes, which we have already
+ * copied to %ecx and %edx.
+ */
+ movl 0x0C00(%ebx, %ebp, 4), %edx
+ movzbl %ch, %ebp
+ xorl 0x0800(%ebx, %ebp, 4), %edx
+ shrl $16, %ecx
+ movzbl %cl, %ebp
+ xorl 0x0400(%ebx, %ebp, 4), %edx
+ movzbl %ch, %ebp
+ xorl (%ebx, %ebp, 4), %edx
+ movzbl %al, %ebp
+
+ xorl 0x1C00(%ebx, %ebp, 4), %edx
+ movzbl %ah, %ebp
+ shrl $16, %eax
+ xorl 0x1800(%ebx, %ebp, 4), %edx
+ movzbl %ah, %ebp
+ movzbl %al, %eax
+ movl 0x1400(%ebx, %eax, 4), %eax
+ addl $8, %esi
+ xorl %edx, %eax
+ xorl 0x1000(%ebx, %ebp, 4), %eax
+
+ /* Copy the number of remaining bytes to %edi. */
+ movl 0x18(%esp), %edi
+
+.L_rest:
+ /* Check for end of input. */
+ testl %edi, %edi
+ jz .L_return
+
+ /* Calculate CRC of the next input byte. */
+ movzbl (%esi), %ebp
+ incl %esi
+ movzbl %al, %ecx
+ xorl %ecx, %ebp
+ shrl $8, %eax
+ xorl (%ebx, %ebp, 4), %eax
+ decl %edi
+ jmp .L_rest
+
+.L_return:
+ /* Complement the final value. */
+ notl %eax
+
+ popl %ebp
+ popl %edi
+ popl %esi
+ popl %ebx
+ ret
+
+#if defined(PIC) || defined(__PIC__)
+ ALIGN(4, 16)
+.L_get_pc:
+ movl (%esp), %ebx
+ ret
+#endif
+
+#if defined(__APPLE__) && (defined(PIC) || defined(__PIC__))
+ /* Mach-O PIC */
+ .section __IMPORT,__pointers,non_lazy_symbol_pointers
+.L_lzma_crc32_table$non_lazy_ptr:
+ .indirect_symbol LZMA_CRC32_TABLE
+ .long 0
+
+#elif defined(_WIN32) || defined(__CYGWIN__)
+# ifdef DLL_EXPORT
+ /* This is equivalent of __declspec(dllexport). */
+ .section .drectve
+ .ascii " -export:lzma_crc32"
+# endif
+
+#elif !defined(__MSDOS__)
+ /* ELF */
+ .size LZMA_CRC32, .-LZMA_CRC32
+#endif
+
+/*
+ * This is needed to support non-executable stack. It's ugly to
+ * use __linux__ here, but I don't know a way to detect when
+ * we are using GNU assembler.
+ */
+#if defined(__ELF__) && defined(__linux__)
+ .section .note.GNU-stack,"",@progbits
+#endif
diff --git a/src/liblzma/check/crc64_fast.c b/src/liblzma/check/crc64_fast.c
new file mode 100644
index 0000000..52af29e
--- /dev/null
+++ b/src/liblzma/check/crc64_fast.c
@@ -0,0 +1,72 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc64.c
+/// \brief CRC64 calculation
+///
+/// Calculate the CRC64 using the slice-by-four algorithm. This is the same
+/// idea that is used in crc32_fast.c, but for CRC64 we use only four tables
+/// instead of eight to avoid increasing CPU cache usage.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "check.h"
+#include "crc_macros.h"
+
+
+#ifdef WORDS_BIGENDIAN
+# define A1(x) ((x) >> 56)
+#else
+# define A1 A
+#endif
+
+
+// See the comments in crc32_fast.c. They aren't duplicated here.
+extern LZMA_API(uint64_t)
+lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc)
+{
+ crc = ~crc;
+
+#ifdef WORDS_BIGENDIAN
+ crc = bswap64(crc);
+#endif
+
+ if (size > 4) {
+ while ((uintptr_t)(buf) & 3) {
+ crc = lzma_crc64_table[0][*buf++ ^ A1(crc)] ^ S8(crc);
+ --size;
+ }
+
+ const uint8_t *const limit = buf + (size & ~(size_t)(3));
+ size &= (size_t)(3);
+
+ while (buf < limit) {
+#ifdef WORDS_BIGENDIAN
+ const uint32_t tmp = (crc >> 32)
+ ^ *(const uint32_t *)(buf);
+#else
+ const uint32_t tmp = crc ^ *(const uint32_t *)(buf);
+#endif
+ buf += 4;
+
+ crc = lzma_crc64_table[3][A(tmp)]
+ ^ lzma_crc64_table[2][B(tmp)]
+ ^ S32(crc)
+ ^ lzma_crc64_table[1][C(tmp)]
+ ^ lzma_crc64_table[0][D(tmp)];
+ }
+ }
+
+ while (size-- != 0)
+ crc = lzma_crc64_table[0][*buf++ ^ A1(crc)] ^ S8(crc);
+
+#ifdef WORDS_BIGENDIAN
+ crc = bswap64(crc);
+#endif
+
+ return ~crc;
+}
diff --git a/src/liblzma/check/crc64_small.c b/src/liblzma/check/crc64_small.c
new file mode 100644
index 0000000..55d7231
--- /dev/null
+++ b/src/liblzma/check/crc64_small.c
@@ -0,0 +1,53 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc64_small.c
+/// \brief CRC64 calculation (size-optimized)
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "check.h"
+
+
+static uint64_t crc64_table[256];
+
+
+static void
+crc64_init(void)
+{
+ static const uint64_t poly64 = UINT64_C(0xC96C5795D7870F42);
+
+ for (size_t b = 0; b < 256; ++b) {
+ uint64_t r = b;
+ for (size_t i = 0; i < 8; ++i) {
+ if (r & 1)
+ r = (r >> 1) ^ poly64;
+ else
+ r >>= 1;
+ }
+
+ crc64_table[b] = r;
+ }
+
+ return;
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc)
+{
+ mythread_once(crc64_init);
+
+ crc = ~crc;
+
+ while (size != 0) {
+ crc = crc64_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8);
+ --size;
+ }
+
+ return ~crc;
+}
diff --git a/src/liblzma/check/crc64_table.c b/src/liblzma/check/crc64_table.c
new file mode 100644
index 0000000..1fbcd94
--- /dev/null
+++ b/src/liblzma/check/crc64_table.c
@@ -0,0 +1,19 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc64_table.c
+/// \brief Precalculated CRC64 table with correct endianness
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+#ifdef WORDS_BIGENDIAN
+# include "crc64_table_be.h"
+#else
+# include "crc64_table_le.h"
+#endif
diff --git a/src/liblzma/check/crc64_table_be.h b/src/liblzma/check/crc64_table_be.h
new file mode 100644
index 0000000..ea074f3
--- /dev/null
+++ b/src/liblzma/check/crc64_table_be.h
@@ -0,0 +1,521 @@
+/* This file has been automatically generated by crc64_tablegen.c. */
+
+const uint64_t lzma_crc64_table[4][256] = {
+ {
+ UINT64_C(0x0000000000000000), UINT64_C(0x6F5FA703BE4C2EB3),
+ UINT64_C(0x5BA040A8573684F4), UINT64_C(0x34FFE7ABE97AAA47),
+ UINT64_C(0x335E8FFF84C3D07B), UINT64_C(0x5C0128FC3A8FFEC8),
+ UINT64_C(0x68FECF57D3F5548F), UINT64_C(0x07A168546DB97A3C),
+ UINT64_C(0x66BC1EFF0987A1F7), UINT64_C(0x09E3B9FCB7CB8F44),
+ UINT64_C(0x3D1C5E575EB12503), UINT64_C(0x5243F954E0FD0BB0),
+ UINT64_C(0x55E291008D44718C), UINT64_C(0x3ABD360333085F3F),
+ UINT64_C(0x0E42D1A8DA72F578), UINT64_C(0x611D76AB643EDBCB),
+ UINT64_C(0x4966335138A19B7D), UINT64_C(0x2639945286EDB5CE),
+ UINT64_C(0x12C673F96F971F89), UINT64_C(0x7D99D4FAD1DB313A),
+ UINT64_C(0x7A38BCAEBC624B06), UINT64_C(0x15671BAD022E65B5),
+ UINT64_C(0x2198FC06EB54CFF2), UINT64_C(0x4EC75B055518E141),
+ UINT64_C(0x2FDA2DAE31263A8A), UINT64_C(0x40858AAD8F6A1439),
+ UINT64_C(0x747A6D066610BE7E), UINT64_C(0x1B25CA05D85C90CD),
+ UINT64_C(0x1C84A251B5E5EAF1), UINT64_C(0x73DB05520BA9C442),
+ UINT64_C(0x4724E2F9E2D36E05), UINT64_C(0x287B45FA5C9F40B6),
+ UINT64_C(0x92CC66A2704237FB), UINT64_C(0xFD93C1A1CE0E1948),
+ UINT64_C(0xC96C260A2774B30F), UINT64_C(0xA633810999389DBC),
+ UINT64_C(0xA192E95DF481E780), UINT64_C(0xCECD4E5E4ACDC933),
+ UINT64_C(0xFA32A9F5A3B76374), UINT64_C(0x956D0EF61DFB4DC7),
+ UINT64_C(0xF470785D79C5960C), UINT64_C(0x9B2FDF5EC789B8BF),
+ UINT64_C(0xAFD038F52EF312F8), UINT64_C(0xC08F9FF690BF3C4B),
+ UINT64_C(0xC72EF7A2FD064677), UINT64_C(0xA87150A1434A68C4),
+ UINT64_C(0x9C8EB70AAA30C283), UINT64_C(0xF3D11009147CEC30),
+ UINT64_C(0xDBAA55F348E3AC86), UINT64_C(0xB4F5F2F0F6AF8235),
+ UINT64_C(0x800A155B1FD52872), UINT64_C(0xEF55B258A19906C1),
+ UINT64_C(0xE8F4DA0CCC207CFD), UINT64_C(0x87AB7D0F726C524E),
+ UINT64_C(0xB3549AA49B16F809), UINT64_C(0xDC0B3DA7255AD6BA),
+ UINT64_C(0xBD164B0C41640D71), UINT64_C(0xD249EC0FFF2823C2),
+ UINT64_C(0xE6B60BA416528985), UINT64_C(0x89E9ACA7A81EA736),
+ UINT64_C(0x8E48C4F3C5A7DD0A), UINT64_C(0xE11763F07BEBF3B9),
+ UINT64_C(0xD5E8845B929159FE), UINT64_C(0xBAB723582CDD774D),
+ UINT64_C(0xA187C3EBCA2BB664), UINT64_C(0xCED864E8746798D7),
+ UINT64_C(0xFA2783439D1D3290), UINT64_C(0x9578244023511C23),
+ UINT64_C(0x92D94C144EE8661F), UINT64_C(0xFD86EB17F0A448AC),
+ UINT64_C(0xC9790CBC19DEE2EB), UINT64_C(0xA626ABBFA792CC58),
+ UINT64_C(0xC73BDD14C3AC1793), UINT64_C(0xA8647A177DE03920),
+ UINT64_C(0x9C9B9DBC949A9367), UINT64_C(0xF3C43ABF2AD6BDD4),
+ UINT64_C(0xF46552EB476FC7E8), UINT64_C(0x9B3AF5E8F923E95B),
+ UINT64_C(0xAFC512431059431C), UINT64_C(0xC09AB540AE156DAF),
+ UINT64_C(0xE8E1F0BAF28A2D19), UINT64_C(0x87BE57B94CC603AA),
+ UINT64_C(0xB341B012A5BCA9ED), UINT64_C(0xDC1E17111BF0875E),
+ UINT64_C(0xDBBF7F457649FD62), UINT64_C(0xB4E0D846C805D3D1),
+ UINT64_C(0x801F3FED217F7996), UINT64_C(0xEF4098EE9F335725),
+ UINT64_C(0x8E5DEE45FB0D8CEE), UINT64_C(0xE10249464541A25D),
+ UINT64_C(0xD5FDAEEDAC3B081A), UINT64_C(0xBAA209EE127726A9),
+ UINT64_C(0xBD0361BA7FCE5C95), UINT64_C(0xD25CC6B9C1827226),
+ UINT64_C(0xE6A3211228F8D861), UINT64_C(0x89FC861196B4F6D2),
+ UINT64_C(0x334BA549BA69819F), UINT64_C(0x5C14024A0425AF2C),
+ UINT64_C(0x68EBE5E1ED5F056B), UINT64_C(0x07B442E253132BD8),
+ UINT64_C(0x00152AB63EAA51E4), UINT64_C(0x6F4A8DB580E67F57),
+ UINT64_C(0x5BB56A1E699CD510), UINT64_C(0x34EACD1DD7D0FBA3),
+ UINT64_C(0x55F7BBB6B3EE2068), UINT64_C(0x3AA81CB50DA20EDB),
+ UINT64_C(0x0E57FB1EE4D8A49C), UINT64_C(0x61085C1D5A948A2F),
+ UINT64_C(0x66A93449372DF013), UINT64_C(0x09F6934A8961DEA0),
+ UINT64_C(0x3D0974E1601B74E7), UINT64_C(0x5256D3E2DE575A54),
+ UINT64_C(0x7A2D961882C81AE2), UINT64_C(0x1572311B3C843451),
+ UINT64_C(0x218DD6B0D5FE9E16), UINT64_C(0x4ED271B36BB2B0A5),
+ UINT64_C(0x497319E7060BCA99), UINT64_C(0x262CBEE4B847E42A),
+ UINT64_C(0x12D3594F513D4E6D), UINT64_C(0x7D8CFE4CEF7160DE),
+ UINT64_C(0x1C9188E78B4FBB15), UINT64_C(0x73CE2FE4350395A6),
+ UINT64_C(0x4731C84FDC793FE1), UINT64_C(0x286E6F4C62351152),
+ UINT64_C(0x2FCF07180F8C6B6E), UINT64_C(0x4090A01BB1C045DD),
+ UINT64_C(0x746F47B058BAEF9A), UINT64_C(0x1B30E0B3E6F6C129),
+ UINT64_C(0x420F87D795576CC9), UINT64_C(0x2D5020D42B1B427A),
+ UINT64_C(0x19AFC77FC261E83D), UINT64_C(0x76F0607C7C2DC68E),
+ UINT64_C(0x715108281194BCB2), UINT64_C(0x1E0EAF2BAFD89201),
+ UINT64_C(0x2AF1488046A23846), UINT64_C(0x45AEEF83F8EE16F5),
+ UINT64_C(0x24B399289CD0CD3E), UINT64_C(0x4BEC3E2B229CE38D),
+ UINT64_C(0x7F13D980CBE649CA), UINT64_C(0x104C7E8375AA6779),
+ UINT64_C(0x17ED16D718131D45), UINT64_C(0x78B2B1D4A65F33F6),
+ UINT64_C(0x4C4D567F4F2599B1), UINT64_C(0x2312F17CF169B702),
+ UINT64_C(0x0B69B486ADF6F7B4), UINT64_C(0x6436138513BAD907),
+ UINT64_C(0x50C9F42EFAC07340), UINT64_C(0x3F96532D448C5DF3),
+ UINT64_C(0x38373B79293527CF), UINT64_C(0x57689C7A9779097C),
+ UINT64_C(0x63977BD17E03A33B), UINT64_C(0x0CC8DCD2C04F8D88),
+ UINT64_C(0x6DD5AA79A4715643), UINT64_C(0x028A0D7A1A3D78F0),
+ UINT64_C(0x3675EAD1F347D2B7), UINT64_C(0x592A4DD24D0BFC04),
+ UINT64_C(0x5E8B258620B28638), UINT64_C(0x31D482859EFEA88B),
+ UINT64_C(0x052B652E778402CC), UINT64_C(0x6A74C22DC9C82C7F),
+ UINT64_C(0xD0C3E175E5155B32), UINT64_C(0xBF9C46765B597581),
+ UINT64_C(0x8B63A1DDB223DFC6), UINT64_C(0xE43C06DE0C6FF175),
+ UINT64_C(0xE39D6E8A61D68B49), UINT64_C(0x8CC2C989DF9AA5FA),
+ UINT64_C(0xB83D2E2236E00FBD), UINT64_C(0xD762892188AC210E),
+ UINT64_C(0xB67FFF8AEC92FAC5), UINT64_C(0xD920588952DED476),
+ UINT64_C(0xEDDFBF22BBA47E31), UINT64_C(0x8280182105E85082),
+ UINT64_C(0x8521707568512ABE), UINT64_C(0xEA7ED776D61D040D),
+ UINT64_C(0xDE8130DD3F67AE4A), UINT64_C(0xB1DE97DE812B80F9),
+ UINT64_C(0x99A5D224DDB4C04F), UINT64_C(0xF6FA752763F8EEFC),
+ UINT64_C(0xC205928C8A8244BB), UINT64_C(0xAD5A358F34CE6A08),
+ UINT64_C(0xAAFB5DDB59771034), UINT64_C(0xC5A4FAD8E73B3E87),
+ UINT64_C(0xF15B1D730E4194C0), UINT64_C(0x9E04BA70B00DBA73),
+ UINT64_C(0xFF19CCDBD43361B8), UINT64_C(0x90466BD86A7F4F0B),
+ UINT64_C(0xA4B98C738305E54C), UINT64_C(0xCBE62B703D49CBFF),
+ UINT64_C(0xCC47432450F0B1C3), UINT64_C(0xA318E427EEBC9F70),
+ UINT64_C(0x97E7038C07C63537), UINT64_C(0xF8B8A48FB98A1B84),
+ UINT64_C(0xE388443C5F7CDAAD), UINT64_C(0x8CD7E33FE130F41E),
+ UINT64_C(0xB8280494084A5E59), UINT64_C(0xD777A397B60670EA),
+ UINT64_C(0xD0D6CBC3DBBF0AD6), UINT64_C(0xBF896CC065F32465),
+ UINT64_C(0x8B768B6B8C898E22), UINT64_C(0xE4292C6832C5A091),
+ UINT64_C(0x85345AC356FB7B5A), UINT64_C(0xEA6BFDC0E8B755E9),
+ UINT64_C(0xDE941A6B01CDFFAE), UINT64_C(0xB1CBBD68BF81D11D),
+ UINT64_C(0xB66AD53CD238AB21), UINT64_C(0xD935723F6C748592),
+ UINT64_C(0xEDCA9594850E2FD5), UINT64_C(0x829532973B420166),
+ UINT64_C(0xAAEE776D67DD41D0), UINT64_C(0xC5B1D06ED9916F63),
+ UINT64_C(0xF14E37C530EBC524), UINT64_C(0x9E1190C68EA7EB97),
+ UINT64_C(0x99B0F892E31E91AB), UINT64_C(0xF6EF5F915D52BF18),
+ UINT64_C(0xC210B83AB428155F), UINT64_C(0xAD4F1F390A643BEC),
+ UINT64_C(0xCC5269926E5AE027), UINT64_C(0xA30DCE91D016CE94),
+ UINT64_C(0x97F2293A396C64D3), UINT64_C(0xF8AD8E3987204A60),
+ UINT64_C(0xFF0CE66DEA99305C), UINT64_C(0x9053416E54D51EEF),
+ UINT64_C(0xA4ACA6C5BDAFB4A8), UINT64_C(0xCBF301C603E39A1B),
+ UINT64_C(0x7144229E2F3EED56), UINT64_C(0x1E1B859D9172C3E5),
+ UINT64_C(0x2AE46236780869A2), UINT64_C(0x45BBC535C6444711),
+ UINT64_C(0x421AAD61ABFD3D2D), UINT64_C(0x2D450A6215B1139E),
+ UINT64_C(0x19BAEDC9FCCBB9D9), UINT64_C(0x76E54ACA4287976A),
+ UINT64_C(0x17F83C6126B94CA1), UINT64_C(0x78A79B6298F56212),
+ UINT64_C(0x4C587CC9718FC855), UINT64_C(0x2307DBCACFC3E6E6),
+ UINT64_C(0x24A6B39EA27A9CDA), UINT64_C(0x4BF9149D1C36B269),
+ UINT64_C(0x7F06F336F54C182E), UINT64_C(0x105954354B00369D),
+ UINT64_C(0x382211CF179F762B), UINT64_C(0x577DB6CCA9D35898),
+ UINT64_C(0x6382516740A9F2DF), UINT64_C(0x0CDDF664FEE5DC6C),
+ UINT64_C(0x0B7C9E30935CA650), UINT64_C(0x642339332D1088E3),
+ UINT64_C(0x50DCDE98C46A22A4), UINT64_C(0x3F83799B7A260C17),
+ UINT64_C(0x5E9E0F301E18D7DC), UINT64_C(0x31C1A833A054F96F),
+ UINT64_C(0x053E4F98492E5328), UINT64_C(0x6A61E89BF7627D9B),
+ UINT64_C(0x6DC080CF9ADB07A7), UINT64_C(0x029F27CC24972914),
+ UINT64_C(0x3660C067CDED8353), UINT64_C(0x593F676473A1ADE0)
+ }, {
+ UINT64_C(0x0000000000000000), UINT64_C(0x0DF1D05C9279E954),
+ UINT64_C(0x1AE2A1B924F3D2A9), UINT64_C(0x171371E5B68A3BFD),
+ UINT64_C(0xB1DA4DDC62497DC1), UINT64_C(0xBC2B9D80F0309495),
+ UINT64_C(0xAB38EC6546BAAF68), UINT64_C(0xA6C93C39D4C3463C),
+ UINT64_C(0xE7AB9517EE3D2210), UINT64_C(0xEA5A454B7C44CB44),
+ UINT64_C(0xFD4934AECACEF0B9), UINT64_C(0xF0B8E4F258B719ED),
+ UINT64_C(0x5671D8CB8C745FD1), UINT64_C(0x5B8008971E0DB685),
+ UINT64_C(0x4C937972A8878D78), UINT64_C(0x4162A92E3AFE642C),
+ UINT64_C(0xCE572B2FDC7B4420), UINT64_C(0xC3A6FB734E02AD74),
+ UINT64_C(0xD4B58A96F8889689), UINT64_C(0xD9445ACA6AF17FDD),
+ UINT64_C(0x7F8D66F3BE3239E1), UINT64_C(0x727CB6AF2C4BD0B5),
+ UINT64_C(0x656FC74A9AC1EB48), UINT64_C(0x689E171608B8021C),
+ UINT64_C(0x29FCBE3832466630), UINT64_C(0x240D6E64A03F8F64),
+ UINT64_C(0x331E1F8116B5B499), UINT64_C(0x3EEFCFDD84CC5DCD),
+ UINT64_C(0x9826F3E4500F1BF1), UINT64_C(0x95D723B8C276F2A5),
+ UINT64_C(0x82C4525D74FCC958), UINT64_C(0x8F358201E685200C),
+ UINT64_C(0x9CAF565EB8F78840), UINT64_C(0x915E86022A8E6114),
+ UINT64_C(0x864DF7E79C045AE9), UINT64_C(0x8BBC27BB0E7DB3BD),
+ UINT64_C(0x2D751B82DABEF581), UINT64_C(0x2084CBDE48C71CD5),
+ UINT64_C(0x3797BA3BFE4D2728), UINT64_C(0x3A666A676C34CE7C),
+ UINT64_C(0x7B04C34956CAAA50), UINT64_C(0x76F51315C4B34304),
+ UINT64_C(0x61E662F0723978F9), UINT64_C(0x6C17B2ACE04091AD),
+ UINT64_C(0xCADE8E953483D791), UINT64_C(0xC72F5EC9A6FA3EC5),
+ UINT64_C(0xD03C2F2C10700538), UINT64_C(0xDDCDFF708209EC6C),
+ UINT64_C(0x52F87D71648CCC60), UINT64_C(0x5F09AD2DF6F52534),
+ UINT64_C(0x481ADCC8407F1EC9), UINT64_C(0x45EB0C94D206F79D),
+ UINT64_C(0xE32230AD06C5B1A1), UINT64_C(0xEED3E0F194BC58F5),
+ UINT64_C(0xF9C0911422366308), UINT64_C(0xF4314148B04F8A5C),
+ UINT64_C(0xB553E8668AB1EE70), UINT64_C(0xB8A2383A18C80724),
+ UINT64_C(0xAFB149DFAE423CD9), UINT64_C(0xA24099833C3BD58D),
+ UINT64_C(0x0489A5BAE8F893B1), UINT64_C(0x097875E67A817AE5),
+ UINT64_C(0x1E6B0403CC0B4118), UINT64_C(0x139AD45F5E72A84C),
+ UINT64_C(0x385FADBC70EF1181), UINT64_C(0x35AE7DE0E296F8D5),
+ UINT64_C(0x22BD0C05541CC328), UINT64_C(0x2F4CDC59C6652A7C),
+ UINT64_C(0x8985E06012A66C40), UINT64_C(0x8474303C80DF8514),
+ UINT64_C(0x936741D93655BEE9), UINT64_C(0x9E969185A42C57BD),
+ UINT64_C(0xDFF438AB9ED23391), UINT64_C(0xD205E8F70CABDAC5),
+ UINT64_C(0xC5169912BA21E138), UINT64_C(0xC8E7494E2858086C),
+ UINT64_C(0x6E2E7577FC9B4E50), UINT64_C(0x63DFA52B6EE2A704),
+ UINT64_C(0x74CCD4CED8689CF9), UINT64_C(0x793D04924A1175AD),
+ UINT64_C(0xF6088693AC9455A1), UINT64_C(0xFBF956CF3EEDBCF5),
+ UINT64_C(0xECEA272A88678708), UINT64_C(0xE11BF7761A1E6E5C),
+ UINT64_C(0x47D2CB4FCEDD2860), UINT64_C(0x4A231B135CA4C134),
+ UINT64_C(0x5D306AF6EA2EFAC9), UINT64_C(0x50C1BAAA7857139D),
+ UINT64_C(0x11A3138442A977B1), UINT64_C(0x1C52C3D8D0D09EE5),
+ UINT64_C(0x0B41B23D665AA518), UINT64_C(0x06B06261F4234C4C),
+ UINT64_C(0xA0795E5820E00A70), UINT64_C(0xAD888E04B299E324),
+ UINT64_C(0xBA9BFFE10413D8D9), UINT64_C(0xB76A2FBD966A318D),
+ UINT64_C(0xA4F0FBE2C81899C1), UINT64_C(0xA9012BBE5A617095),
+ UINT64_C(0xBE125A5BECEB4B68), UINT64_C(0xB3E38A077E92A23C),
+ UINT64_C(0x152AB63EAA51E400), UINT64_C(0x18DB666238280D54),
+ UINT64_C(0x0FC817878EA236A9), UINT64_C(0x0239C7DB1CDBDFFD),
+ UINT64_C(0x435B6EF52625BBD1), UINT64_C(0x4EAABEA9B45C5285),
+ UINT64_C(0x59B9CF4C02D66978), UINT64_C(0x54481F1090AF802C),
+ UINT64_C(0xF2812329446CC610), UINT64_C(0xFF70F375D6152F44),
+ UINT64_C(0xE8638290609F14B9), UINT64_C(0xE59252CCF2E6FDED),
+ UINT64_C(0x6AA7D0CD1463DDE1), UINT64_C(0x67560091861A34B5),
+ UINT64_C(0x7045717430900F48), UINT64_C(0x7DB4A128A2E9E61C),
+ UINT64_C(0xDB7D9D11762AA020), UINT64_C(0xD68C4D4DE4534974),
+ UINT64_C(0xC19F3CA852D97289), UINT64_C(0xCC6EECF4C0A09BDD),
+ UINT64_C(0x8D0C45DAFA5EFFF1), UINT64_C(0x80FD9586682716A5),
+ UINT64_C(0x97EEE463DEAD2D58), UINT64_C(0x9A1F343F4CD4C40C),
+ UINT64_C(0x3CD6080698178230), UINT64_C(0x3127D85A0A6E6B64),
+ UINT64_C(0x2634A9BFBCE45099), UINT64_C(0x2BC579E32E9DB9CD),
+ UINT64_C(0xF5A054D6CA71FB90), UINT64_C(0xF851848A580812C4),
+ UINT64_C(0xEF42F56FEE822939), UINT64_C(0xE2B325337CFBC06D),
+ UINT64_C(0x447A190AA8388651), UINT64_C(0x498BC9563A416F05),
+ UINT64_C(0x5E98B8B38CCB54F8), UINT64_C(0x536968EF1EB2BDAC),
+ UINT64_C(0x120BC1C1244CD980), UINT64_C(0x1FFA119DB63530D4),
+ UINT64_C(0x08E9607800BF0B29), UINT64_C(0x0518B02492C6E27D),
+ UINT64_C(0xA3D18C1D4605A441), UINT64_C(0xAE205C41D47C4D15),
+ UINT64_C(0xB9332DA462F676E8), UINT64_C(0xB4C2FDF8F08F9FBC),
+ UINT64_C(0x3BF77FF9160ABFB0), UINT64_C(0x3606AFA5847356E4),
+ UINT64_C(0x2115DE4032F96D19), UINT64_C(0x2CE40E1CA080844D),
+ UINT64_C(0x8A2D32257443C271), UINT64_C(0x87DCE279E63A2B25),
+ UINT64_C(0x90CF939C50B010D8), UINT64_C(0x9D3E43C0C2C9F98C),
+ UINT64_C(0xDC5CEAEEF8379DA0), UINT64_C(0xD1AD3AB26A4E74F4),
+ UINT64_C(0xC6BE4B57DCC44F09), UINT64_C(0xCB4F9B0B4EBDA65D),
+ UINT64_C(0x6D86A7329A7EE061), UINT64_C(0x6077776E08070935),
+ UINT64_C(0x7764068BBE8D32C8), UINT64_C(0x7A95D6D72CF4DB9C),
+ UINT64_C(0x690F0288728673D0), UINT64_C(0x64FED2D4E0FF9A84),
+ UINT64_C(0x73EDA3315675A179), UINT64_C(0x7E1C736DC40C482D),
+ UINT64_C(0xD8D54F5410CF0E11), UINT64_C(0xD5249F0882B6E745),
+ UINT64_C(0xC237EEED343CDCB8), UINT64_C(0xCFC63EB1A64535EC),
+ UINT64_C(0x8EA4979F9CBB51C0), UINT64_C(0x835547C30EC2B894),
+ UINT64_C(0x94463626B8488369), UINT64_C(0x99B7E67A2A316A3D),
+ UINT64_C(0x3F7EDA43FEF22C01), UINT64_C(0x328F0A1F6C8BC555),
+ UINT64_C(0x259C7BFADA01FEA8), UINT64_C(0x286DABA6487817FC),
+ UINT64_C(0xA75829A7AEFD37F0), UINT64_C(0xAAA9F9FB3C84DEA4),
+ UINT64_C(0xBDBA881E8A0EE559), UINT64_C(0xB04B584218770C0D),
+ UINT64_C(0x1682647BCCB44A31), UINT64_C(0x1B73B4275ECDA365),
+ UINT64_C(0x0C60C5C2E8479898), UINT64_C(0x0191159E7A3E71CC),
+ UINT64_C(0x40F3BCB040C015E0), UINT64_C(0x4D026CECD2B9FCB4),
+ UINT64_C(0x5A111D096433C749), UINT64_C(0x57E0CD55F64A2E1D),
+ UINT64_C(0xF129F16C22896821), UINT64_C(0xFCD82130B0F08175),
+ UINT64_C(0xEBCB50D5067ABA88), UINT64_C(0xE63A8089940353DC),
+ UINT64_C(0xCDFFF96ABA9EEA11), UINT64_C(0xC00E293628E70345),
+ UINT64_C(0xD71D58D39E6D38B8), UINT64_C(0xDAEC888F0C14D1EC),
+ UINT64_C(0x7C25B4B6D8D797D0), UINT64_C(0x71D464EA4AAE7E84),
+ UINT64_C(0x66C7150FFC244579), UINT64_C(0x6B36C5536E5DAC2D),
+ UINT64_C(0x2A546C7D54A3C801), UINT64_C(0x27A5BC21C6DA2155),
+ UINT64_C(0x30B6CDC470501AA8), UINT64_C(0x3D471D98E229F3FC),
+ UINT64_C(0x9B8E21A136EAB5C0), UINT64_C(0x967FF1FDA4935C94),
+ UINT64_C(0x816C801812196769), UINT64_C(0x8C9D504480608E3D),
+ UINT64_C(0x03A8D24566E5AE31), UINT64_C(0x0E590219F49C4765),
+ UINT64_C(0x194A73FC42167C98), UINT64_C(0x14BBA3A0D06F95CC),
+ UINT64_C(0xB2729F9904ACD3F0), UINT64_C(0xBF834FC596D53AA4),
+ UINT64_C(0xA8903E20205F0159), UINT64_C(0xA561EE7CB226E80D),
+ UINT64_C(0xE403475288D88C21), UINT64_C(0xE9F2970E1AA16575),
+ UINT64_C(0xFEE1E6EBAC2B5E88), UINT64_C(0xF31036B73E52B7DC),
+ UINT64_C(0x55D90A8EEA91F1E0), UINT64_C(0x5828DAD278E818B4),
+ UINT64_C(0x4F3BAB37CE622349), UINT64_C(0x42CA7B6B5C1BCA1D),
+ UINT64_C(0x5150AF3402696251), UINT64_C(0x5CA17F6890108B05),
+ UINT64_C(0x4BB20E8D269AB0F8), UINT64_C(0x4643DED1B4E359AC),
+ UINT64_C(0xE08AE2E860201F90), UINT64_C(0xED7B32B4F259F6C4),
+ UINT64_C(0xFA68435144D3CD39), UINT64_C(0xF799930DD6AA246D),
+ UINT64_C(0xB6FB3A23EC544041), UINT64_C(0xBB0AEA7F7E2DA915),
+ UINT64_C(0xAC199B9AC8A792E8), UINT64_C(0xA1E84BC65ADE7BBC),
+ UINT64_C(0x072177FF8E1D3D80), UINT64_C(0x0AD0A7A31C64D4D4),
+ UINT64_C(0x1DC3D646AAEEEF29), UINT64_C(0x1032061A3897067D),
+ UINT64_C(0x9F07841BDE122671), UINT64_C(0x92F654474C6BCF25),
+ UINT64_C(0x85E525A2FAE1F4D8), UINT64_C(0x8814F5FE68981D8C),
+ UINT64_C(0x2EDDC9C7BC5B5BB0), UINT64_C(0x232C199B2E22B2E4),
+ UINT64_C(0x343F687E98A88919), UINT64_C(0x39CEB8220AD1604D),
+ UINT64_C(0x78AC110C302F0461), UINT64_C(0x755DC150A256ED35),
+ UINT64_C(0x624EB0B514DCD6C8), UINT64_C(0x6FBF60E986A53F9C),
+ UINT64_C(0xC9765CD0526679A0), UINT64_C(0xC4878C8CC01F90F4),
+ UINT64_C(0xD394FD697695AB09), UINT64_C(0xDE652D35E4EC425D)
+ }, {
+ UINT64_C(0x0000000000000000), UINT64_C(0xCB6D6A914AE10B3F),
+ UINT64_C(0x96DBD42295C2177E), UINT64_C(0x5DB6BEB3DF231C41),
+ UINT64_C(0x2CB7A9452A852FFC), UINT64_C(0xE7DAC3D4606424C3),
+ UINT64_C(0xBA6C7D67BF473882), UINT64_C(0x710117F6F5A633BD),
+ UINT64_C(0xDD705D247FA5876A), UINT64_C(0x161D37B535448C55),
+ UINT64_C(0x4BAB8906EA679014), UINT64_C(0x80C6E397A0869B2B),
+ UINT64_C(0xF1C7F4615520A896), UINT64_C(0x3AAA9EF01FC1A3A9),
+ UINT64_C(0x671C2043C0E2BFE8), UINT64_C(0xAC714AD28A03B4D7),
+ UINT64_C(0xBAE1BA48FE4A0FD5), UINT64_C(0x718CD0D9B4AB04EA),
+ UINT64_C(0x2C3A6E6A6B8818AB), UINT64_C(0xE75704FB21691394),
+ UINT64_C(0x9656130DD4CF2029), UINT64_C(0x5D3B799C9E2E2B16),
+ UINT64_C(0x008DC72F410D3757), UINT64_C(0xCBE0ADBE0BEC3C68),
+ UINT64_C(0x6791E76C81EF88BF), UINT64_C(0xACFC8DFDCB0E8380),
+ UINT64_C(0xF14A334E142D9FC1), UINT64_C(0x3A2759DF5ECC94FE),
+ UINT64_C(0x4B264E29AB6AA743), UINT64_C(0x804B24B8E18BAC7C),
+ UINT64_C(0xDDFD9A0B3EA8B03D), UINT64_C(0x1690F09A7449BB02),
+ UINT64_C(0xF1DD7B3ED73AC638), UINT64_C(0x3AB011AF9DDBCD07),
+ UINT64_C(0x6706AF1C42F8D146), UINT64_C(0xAC6BC58D0819DA79),
+ UINT64_C(0xDD6AD27BFDBFE9C4), UINT64_C(0x1607B8EAB75EE2FB),
+ UINT64_C(0x4BB10659687DFEBA), UINT64_C(0x80DC6CC8229CF585),
+ UINT64_C(0x2CAD261AA89F4152), UINT64_C(0xE7C04C8BE27E4A6D),
+ UINT64_C(0xBA76F2383D5D562C), UINT64_C(0x711B98A977BC5D13),
+ UINT64_C(0x001A8F5F821A6EAE), UINT64_C(0xCB77E5CEC8FB6591),
+ UINT64_C(0x96C15B7D17D879D0), UINT64_C(0x5DAC31EC5D3972EF),
+ UINT64_C(0x4B3CC1762970C9ED), UINT64_C(0x8051ABE76391C2D2),
+ UINT64_C(0xDDE71554BCB2DE93), UINT64_C(0x168A7FC5F653D5AC),
+ UINT64_C(0x678B683303F5E611), UINT64_C(0xACE602A24914ED2E),
+ UINT64_C(0xF150BC119637F16F), UINT64_C(0x3A3DD680DCD6FA50),
+ UINT64_C(0x964C9C5256D54E87), UINT64_C(0x5D21F6C31C3445B8),
+ UINT64_C(0x00974870C31759F9), UINT64_C(0xCBFA22E189F652C6),
+ UINT64_C(0xBAFB35177C50617B), UINT64_C(0x71965F8636B16A44),
+ UINT64_C(0x2C20E135E9927605), UINT64_C(0xE74D8BA4A3737D3A),
+ UINT64_C(0xE2BBF77CAE758C71), UINT64_C(0x29D69DEDE494874E),
+ UINT64_C(0x7460235E3BB79B0F), UINT64_C(0xBF0D49CF71569030),
+ UINT64_C(0xCE0C5E3984F0A38D), UINT64_C(0x056134A8CE11A8B2),
+ UINT64_C(0x58D78A1B1132B4F3), UINT64_C(0x93BAE08A5BD3BFCC),
+ UINT64_C(0x3FCBAA58D1D00B1B), UINT64_C(0xF4A6C0C99B310024),
+ UINT64_C(0xA9107E7A44121C65), UINT64_C(0x627D14EB0EF3175A),
+ UINT64_C(0x137C031DFB5524E7), UINT64_C(0xD811698CB1B42FD8),
+ UINT64_C(0x85A7D73F6E973399), UINT64_C(0x4ECABDAE247638A6),
+ UINT64_C(0x585A4D34503F83A4), UINT64_C(0x933727A51ADE889B),
+ UINT64_C(0xCE819916C5FD94DA), UINT64_C(0x05ECF3878F1C9FE5),
+ UINT64_C(0x74EDE4717ABAAC58), UINT64_C(0xBF808EE0305BA767),
+ UINT64_C(0xE2363053EF78BB26), UINT64_C(0x295B5AC2A599B019),
+ UINT64_C(0x852A10102F9A04CE), UINT64_C(0x4E477A81657B0FF1),
+ UINT64_C(0x13F1C432BA5813B0), UINT64_C(0xD89CAEA3F0B9188F),
+ UINT64_C(0xA99DB955051F2B32), UINT64_C(0x62F0D3C44FFE200D),
+ UINT64_C(0x3F466D7790DD3C4C), UINT64_C(0xF42B07E6DA3C3773),
+ UINT64_C(0x13668C42794F4A49), UINT64_C(0xD80BE6D333AE4176),
+ UINT64_C(0x85BD5860EC8D5D37), UINT64_C(0x4ED032F1A66C5608),
+ UINT64_C(0x3FD1250753CA65B5), UINT64_C(0xF4BC4F96192B6E8A),
+ UINT64_C(0xA90AF125C60872CB), UINT64_C(0x62679BB48CE979F4),
+ UINT64_C(0xCE16D16606EACD23), UINT64_C(0x057BBBF74C0BC61C),
+ UINT64_C(0x58CD05449328DA5D), UINT64_C(0x93A06FD5D9C9D162),
+ UINT64_C(0xE2A178232C6FE2DF), UINT64_C(0x29CC12B2668EE9E0),
+ UINT64_C(0x747AAC01B9ADF5A1), UINT64_C(0xBF17C690F34CFE9E),
+ UINT64_C(0xA987360A8705459C), UINT64_C(0x62EA5C9BCDE44EA3),
+ UINT64_C(0x3F5CE22812C752E2), UINT64_C(0xF43188B9582659DD),
+ UINT64_C(0x85309F4FAD806A60), UINT64_C(0x4E5DF5DEE761615F),
+ UINT64_C(0x13EB4B6D38427D1E), UINT64_C(0xD88621FC72A37621),
+ UINT64_C(0x74F76B2EF8A0C2F6), UINT64_C(0xBF9A01BFB241C9C9),
+ UINT64_C(0xE22CBF0C6D62D588), UINT64_C(0x2941D59D2783DEB7),
+ UINT64_C(0x5840C26BD225ED0A), UINT64_C(0x932DA8FA98C4E635),
+ UINT64_C(0xCE9B164947E7FA74), UINT64_C(0x05F67CD80D06F14B),
+ UINT64_C(0xC477EFF95CEB18E3), UINT64_C(0x0F1A8568160A13DC),
+ UINT64_C(0x52AC3BDBC9290F9D), UINT64_C(0x99C1514A83C804A2),
+ UINT64_C(0xE8C046BC766E371F), UINT64_C(0x23AD2C2D3C8F3C20),
+ UINT64_C(0x7E1B929EE3AC2061), UINT64_C(0xB576F80FA94D2B5E),
+ UINT64_C(0x1907B2DD234E9F89), UINT64_C(0xD26AD84C69AF94B6),
+ UINT64_C(0x8FDC66FFB68C88F7), UINT64_C(0x44B10C6EFC6D83C8),
+ UINT64_C(0x35B01B9809CBB075), UINT64_C(0xFEDD7109432ABB4A),
+ UINT64_C(0xA36BCFBA9C09A70B), UINT64_C(0x6806A52BD6E8AC34),
+ UINT64_C(0x7E9655B1A2A11736), UINT64_C(0xB5FB3F20E8401C09),
+ UINT64_C(0xE84D819337630048), UINT64_C(0x2320EB027D820B77),
+ UINT64_C(0x5221FCF4882438CA), UINT64_C(0x994C9665C2C533F5),
+ UINT64_C(0xC4FA28D61DE62FB4), UINT64_C(0x0F9742475707248B),
+ UINT64_C(0xA3E60895DD04905C), UINT64_C(0x688B620497E59B63),
+ UINT64_C(0x353DDCB748C68722), UINT64_C(0xFE50B62602278C1D),
+ UINT64_C(0x8F51A1D0F781BFA0), UINT64_C(0x443CCB41BD60B49F),
+ UINT64_C(0x198A75F26243A8DE), UINT64_C(0xD2E71F6328A2A3E1),
+ UINT64_C(0x35AA94C78BD1DEDB), UINT64_C(0xFEC7FE56C130D5E4),
+ UINT64_C(0xA37140E51E13C9A5), UINT64_C(0x681C2A7454F2C29A),
+ UINT64_C(0x191D3D82A154F127), UINT64_C(0xD2705713EBB5FA18),
+ UINT64_C(0x8FC6E9A03496E659), UINT64_C(0x44AB83317E77ED66),
+ UINT64_C(0xE8DAC9E3F47459B1), UINT64_C(0x23B7A372BE95528E),
+ UINT64_C(0x7E011DC161B64ECF), UINT64_C(0xB56C77502B5745F0),
+ UINT64_C(0xC46D60A6DEF1764D), UINT64_C(0x0F000A3794107D72),
+ UINT64_C(0x52B6B4844B336133), UINT64_C(0x99DBDE1501D26A0C),
+ UINT64_C(0x8F4B2E8F759BD10E), UINT64_C(0x4426441E3F7ADA31),
+ UINT64_C(0x1990FAADE059C670), UINT64_C(0xD2FD903CAAB8CD4F),
+ UINT64_C(0xA3FC87CA5F1EFEF2), UINT64_C(0x6891ED5B15FFF5CD),
+ UINT64_C(0x352753E8CADCE98C), UINT64_C(0xFE4A3979803DE2B3),
+ UINT64_C(0x523B73AB0A3E5664), UINT64_C(0x9956193A40DF5D5B),
+ UINT64_C(0xC4E0A7899FFC411A), UINT64_C(0x0F8DCD18D51D4A25),
+ UINT64_C(0x7E8CDAEE20BB7998), UINT64_C(0xB5E1B07F6A5A72A7),
+ UINT64_C(0xE8570ECCB5796EE6), UINT64_C(0x233A645DFF9865D9),
+ UINT64_C(0x26CC1885F29E9492), UINT64_C(0xEDA17214B87F9FAD),
+ UINT64_C(0xB017CCA7675C83EC), UINT64_C(0x7B7AA6362DBD88D3),
+ UINT64_C(0x0A7BB1C0D81BBB6E), UINT64_C(0xC116DB5192FAB051),
+ UINT64_C(0x9CA065E24DD9AC10), UINT64_C(0x57CD0F730738A72F),
+ UINT64_C(0xFBBC45A18D3B13F8), UINT64_C(0x30D12F30C7DA18C7),
+ UINT64_C(0x6D67918318F90486), UINT64_C(0xA60AFB1252180FB9),
+ UINT64_C(0xD70BECE4A7BE3C04), UINT64_C(0x1C668675ED5F373B),
+ UINT64_C(0x41D038C6327C2B7A), UINT64_C(0x8ABD5257789D2045),
+ UINT64_C(0x9C2DA2CD0CD49B47), UINT64_C(0x5740C85C46359078),
+ UINT64_C(0x0AF676EF99168C39), UINT64_C(0xC19B1C7ED3F78706),
+ UINT64_C(0xB09A0B882651B4BB), UINT64_C(0x7BF761196CB0BF84),
+ UINT64_C(0x2641DFAAB393A3C5), UINT64_C(0xED2CB53BF972A8FA),
+ UINT64_C(0x415DFFE973711C2D), UINT64_C(0x8A30957839901712),
+ UINT64_C(0xD7862BCBE6B30B53), UINT64_C(0x1CEB415AAC52006C),
+ UINT64_C(0x6DEA56AC59F433D1), UINT64_C(0xA6873C3D131538EE),
+ UINT64_C(0xFB31828ECC3624AF), UINT64_C(0x305CE81F86D72F90),
+ UINT64_C(0xD71163BB25A452AA), UINT64_C(0x1C7C092A6F455995),
+ UINT64_C(0x41CAB799B06645D4), UINT64_C(0x8AA7DD08FA874EEB),
+ UINT64_C(0xFBA6CAFE0F217D56), UINT64_C(0x30CBA06F45C07669),
+ UINT64_C(0x6D7D1EDC9AE36A28), UINT64_C(0xA610744DD0026117),
+ UINT64_C(0x0A613E9F5A01D5C0), UINT64_C(0xC10C540E10E0DEFF),
+ UINT64_C(0x9CBAEABDCFC3C2BE), UINT64_C(0x57D7802C8522C981),
+ UINT64_C(0x26D697DA7084FA3C), UINT64_C(0xEDBBFD4B3A65F103),
+ UINT64_C(0xB00D43F8E546ED42), UINT64_C(0x7B602969AFA7E67D),
+ UINT64_C(0x6DF0D9F3DBEE5D7F), UINT64_C(0xA69DB362910F5640),
+ UINT64_C(0xFB2B0DD14E2C4A01), UINT64_C(0x3046674004CD413E),
+ UINT64_C(0x414770B6F16B7283), UINT64_C(0x8A2A1A27BB8A79BC),
+ UINT64_C(0xD79CA49464A965FD), UINT64_C(0x1CF1CE052E486EC2),
+ UINT64_C(0xB08084D7A44BDA15), UINT64_C(0x7BEDEE46EEAAD12A),
+ UINT64_C(0x265B50F53189CD6B), UINT64_C(0xED363A647B68C654),
+ UINT64_C(0x9C372D928ECEF5E9), UINT64_C(0x575A4703C42FFED6),
+ UINT64_C(0x0AECF9B01B0CE297), UINT64_C(0xC181932151EDE9A8)
+ }, {
+ UINT64_C(0x0000000000000000), UINT64_C(0xDCA12C225E8AEE1D),
+ UINT64_C(0xB8435944BC14DD3B), UINT64_C(0x64E27566E29E3326),
+ UINT64_C(0x7087B2887829BA77), UINT64_C(0xAC269EAA26A3546A),
+ UINT64_C(0xC8C4EBCCC43D674C), UINT64_C(0x1465C7EE9AB78951),
+ UINT64_C(0xE00E6511F15274EF), UINT64_C(0x3CAF4933AFD89AF2),
+ UINT64_C(0x584D3C554D46A9D4), UINT64_C(0x84EC107713CC47C9),
+ UINT64_C(0x9089D799897BCE98), UINT64_C(0x4C28FBBBD7F12085),
+ UINT64_C(0x28CA8EDD356F13A3), UINT64_C(0xF46BA2FF6BE5FDBE),
+ UINT64_C(0x4503C48DC90A304C), UINT64_C(0x99A2E8AF9780DE51),
+ UINT64_C(0xFD409DC9751EED77), UINT64_C(0x21E1B1EB2B94036A),
+ UINT64_C(0x35847605B1238A3B), UINT64_C(0xE9255A27EFA96426),
+ UINT64_C(0x8DC72F410D375700), UINT64_C(0x5166036353BDB91D),
+ UINT64_C(0xA50DA19C385844A3), UINT64_C(0x79AC8DBE66D2AABE),
+ UINT64_C(0x1D4EF8D8844C9998), UINT64_C(0xC1EFD4FADAC67785),
+ UINT64_C(0xD58A13144071FED4), UINT64_C(0x092B3F361EFB10C9),
+ UINT64_C(0x6DC94A50FC6523EF), UINT64_C(0xB1686672A2EFCDF2),
+ UINT64_C(0x8A06881B93156098), UINT64_C(0x56A7A439CD9F8E85),
+ UINT64_C(0x3245D15F2F01BDA3), UINT64_C(0xEEE4FD7D718B53BE),
+ UINT64_C(0xFA813A93EB3CDAEF), UINT64_C(0x262016B1B5B634F2),
+ UINT64_C(0x42C263D7572807D4), UINT64_C(0x9E634FF509A2E9C9),
+ UINT64_C(0x6A08ED0A62471477), UINT64_C(0xB6A9C1283CCDFA6A),
+ UINT64_C(0xD24BB44EDE53C94C), UINT64_C(0x0EEA986C80D92751),
+ UINT64_C(0x1A8F5F821A6EAE00), UINT64_C(0xC62E73A044E4401D),
+ UINT64_C(0xA2CC06C6A67A733B), UINT64_C(0x7E6D2AE4F8F09D26),
+ UINT64_C(0xCF054C965A1F50D4), UINT64_C(0x13A460B40495BEC9),
+ UINT64_C(0x774615D2E60B8DEF), UINT64_C(0xABE739F0B88163F2),
+ UINT64_C(0xBF82FE1E2236EAA3), UINT64_C(0x6323D23C7CBC04BE),
+ UINT64_C(0x07C1A75A9E223798), UINT64_C(0xDB608B78C0A8D985),
+ UINT64_C(0x2F0B2987AB4D243B), UINT64_C(0xF3AA05A5F5C7CA26),
+ UINT64_C(0x974870C31759F900), UINT64_C(0x4BE95CE149D3171D),
+ UINT64_C(0x5F8C9B0FD3649E4C), UINT64_C(0x832DB72D8DEE7051),
+ UINT64_C(0xE7CFC24B6F704377), UINT64_C(0x3B6EEE6931FAAD6A),
+ UINT64_C(0x91131E980D8418A2), UINT64_C(0x4DB232BA530EF6BF),
+ UINT64_C(0x295047DCB190C599), UINT64_C(0xF5F16BFEEF1A2B84),
+ UINT64_C(0xE194AC1075ADA2D5), UINT64_C(0x3D3580322B274CC8),
+ UINT64_C(0x59D7F554C9B97FEE), UINT64_C(0x8576D976973391F3),
+ UINT64_C(0x711D7B89FCD66C4D), UINT64_C(0xADBC57ABA25C8250),
+ UINT64_C(0xC95E22CD40C2B176), UINT64_C(0x15FF0EEF1E485F6B),
+ UINT64_C(0x019AC90184FFD63A), UINT64_C(0xDD3BE523DA753827),
+ UINT64_C(0xB9D9904538EB0B01), UINT64_C(0x6578BC676661E51C),
+ UINT64_C(0xD410DA15C48E28EE), UINT64_C(0x08B1F6379A04C6F3),
+ UINT64_C(0x6C538351789AF5D5), UINT64_C(0xB0F2AF7326101BC8),
+ UINT64_C(0xA497689DBCA79299), UINT64_C(0x783644BFE22D7C84),
+ UINT64_C(0x1CD431D900B34FA2), UINT64_C(0xC0751DFB5E39A1BF),
+ UINT64_C(0x341EBF0435DC5C01), UINT64_C(0xE8BF93266B56B21C),
+ UINT64_C(0x8C5DE64089C8813A), UINT64_C(0x50FCCA62D7426F27),
+ UINT64_C(0x44990D8C4DF5E676), UINT64_C(0x983821AE137F086B),
+ UINT64_C(0xFCDA54C8F1E13B4D), UINT64_C(0x207B78EAAF6BD550),
+ UINT64_C(0x1B1596839E91783A), UINT64_C(0xC7B4BAA1C01B9627),
+ UINT64_C(0xA356CFC72285A501), UINT64_C(0x7FF7E3E57C0F4B1C),
+ UINT64_C(0x6B92240BE6B8C24D), UINT64_C(0xB7330829B8322C50),
+ UINT64_C(0xD3D17D4F5AAC1F76), UINT64_C(0x0F70516D0426F16B),
+ UINT64_C(0xFB1BF3926FC30CD5), UINT64_C(0x27BADFB03149E2C8),
+ UINT64_C(0x4358AAD6D3D7D1EE), UINT64_C(0x9FF986F48D5D3FF3),
+ UINT64_C(0x8B9C411A17EAB6A2), UINT64_C(0x573D6D38496058BF),
+ UINT64_C(0x33DF185EABFE6B99), UINT64_C(0xEF7E347CF5748584),
+ UINT64_C(0x5E16520E579B4876), UINT64_C(0x82B77E2C0911A66B),
+ UINT64_C(0xE6550B4AEB8F954D), UINT64_C(0x3AF42768B5057B50),
+ UINT64_C(0x2E91E0862FB2F201), UINT64_C(0xF230CCA471381C1C),
+ UINT64_C(0x96D2B9C293A62F3A), UINT64_C(0x4A7395E0CD2CC127),
+ UINT64_C(0xBE18371FA6C93C99), UINT64_C(0x62B91B3DF843D284),
+ UINT64_C(0x065B6E5B1ADDE1A2), UINT64_C(0xDAFA427944570FBF),
+ UINT64_C(0xCE9F8597DEE086EE), UINT64_C(0x123EA9B5806A68F3),
+ UINT64_C(0x76DCDCD362F45BD5), UINT64_C(0xAA7DF0F13C7EB5C8),
+ UINT64_C(0xA739329F30A7E9D6), UINT64_C(0x7B981EBD6E2D07CB),
+ UINT64_C(0x1F7A6BDB8CB334ED), UINT64_C(0xC3DB47F9D239DAF0),
+ UINT64_C(0xD7BE8017488E53A1), UINT64_C(0x0B1FAC351604BDBC),
+ UINT64_C(0x6FFDD953F49A8E9A), UINT64_C(0xB35CF571AA106087),
+ UINT64_C(0x4737578EC1F59D39), UINT64_C(0x9B967BAC9F7F7324),
+ UINT64_C(0xFF740ECA7DE14002), UINT64_C(0x23D522E8236BAE1F),
+ UINT64_C(0x37B0E506B9DC274E), UINT64_C(0xEB11C924E756C953),
+ UINT64_C(0x8FF3BC4205C8FA75), UINT64_C(0x535290605B421468),
+ UINT64_C(0xE23AF612F9ADD99A), UINT64_C(0x3E9BDA30A7273787),
+ UINT64_C(0x5A79AF5645B904A1), UINT64_C(0x86D883741B33EABC),
+ UINT64_C(0x92BD449A818463ED), UINT64_C(0x4E1C68B8DF0E8DF0),
+ UINT64_C(0x2AFE1DDE3D90BED6), UINT64_C(0xF65F31FC631A50CB),
+ UINT64_C(0x0234930308FFAD75), UINT64_C(0xDE95BF2156754368),
+ UINT64_C(0xBA77CA47B4EB704E), UINT64_C(0x66D6E665EA619E53),
+ UINT64_C(0x72B3218B70D61702), UINT64_C(0xAE120DA92E5CF91F),
+ UINT64_C(0xCAF078CFCCC2CA39), UINT64_C(0x165154ED92482424),
+ UINT64_C(0x2D3FBA84A3B2894E), UINT64_C(0xF19E96A6FD386753),
+ UINT64_C(0x957CE3C01FA65475), UINT64_C(0x49DDCFE2412CBA68),
+ UINT64_C(0x5DB8080CDB9B3339), UINT64_C(0x8119242E8511DD24),
+ UINT64_C(0xE5FB5148678FEE02), UINT64_C(0x395A7D6A3905001F),
+ UINT64_C(0xCD31DF9552E0FDA1), UINT64_C(0x1190F3B70C6A13BC),
+ UINT64_C(0x757286D1EEF4209A), UINT64_C(0xA9D3AAF3B07ECE87),
+ UINT64_C(0xBDB66D1D2AC947D6), UINT64_C(0x6117413F7443A9CB),
+ UINT64_C(0x05F5345996DD9AED), UINT64_C(0xD954187BC85774F0),
+ UINT64_C(0x683C7E096AB8B902), UINT64_C(0xB49D522B3432571F),
+ UINT64_C(0xD07F274DD6AC6439), UINT64_C(0x0CDE0B6F88268A24),
+ UINT64_C(0x18BBCC8112910375), UINT64_C(0xC41AE0A34C1BED68),
+ UINT64_C(0xA0F895C5AE85DE4E), UINT64_C(0x7C59B9E7F00F3053),
+ UINT64_C(0x88321B189BEACDED), UINT64_C(0x5493373AC56023F0),
+ UINT64_C(0x3071425C27FE10D6), UINT64_C(0xECD06E7E7974FECB),
+ UINT64_C(0xF8B5A990E3C3779A), UINT64_C(0x241485B2BD499987),
+ UINT64_C(0x40F6F0D45FD7AAA1), UINT64_C(0x9C57DCF6015D44BC),
+ UINT64_C(0x362A2C073D23F174), UINT64_C(0xEA8B002563A91F69),
+ UINT64_C(0x8E69754381372C4F), UINT64_C(0x52C85961DFBDC252),
+ UINT64_C(0x46AD9E8F450A4B03), UINT64_C(0x9A0CB2AD1B80A51E),
+ UINT64_C(0xFEEEC7CBF91E9638), UINT64_C(0x224FEBE9A7947825),
+ UINT64_C(0xD6244916CC71859B), UINT64_C(0x0A85653492FB6B86),
+ UINT64_C(0x6E671052706558A0), UINT64_C(0xB2C63C702EEFB6BD),
+ UINT64_C(0xA6A3FB9EB4583FEC), UINT64_C(0x7A02D7BCEAD2D1F1),
+ UINT64_C(0x1EE0A2DA084CE2D7), UINT64_C(0xC2418EF856C60CCA),
+ UINT64_C(0x7329E88AF429C138), UINT64_C(0xAF88C4A8AAA32F25),
+ UINT64_C(0xCB6AB1CE483D1C03), UINT64_C(0x17CB9DEC16B7F21E),
+ UINT64_C(0x03AE5A028C007B4F), UINT64_C(0xDF0F7620D28A9552),
+ UINT64_C(0xBBED03463014A674), UINT64_C(0x674C2F646E9E4869),
+ UINT64_C(0x93278D9B057BB5D7), UINT64_C(0x4F86A1B95BF15BCA),
+ UINT64_C(0x2B64D4DFB96F68EC), UINT64_C(0xF7C5F8FDE7E586F1),
+ UINT64_C(0xE3A03F137D520FA0), UINT64_C(0x3F01133123D8E1BD),
+ UINT64_C(0x5BE36657C146D29B), UINT64_C(0x87424A759FCC3C86),
+ UINT64_C(0xBC2CA41CAE3691EC), UINT64_C(0x608D883EF0BC7FF1),
+ UINT64_C(0x046FFD5812224CD7), UINT64_C(0xD8CED17A4CA8A2CA),
+ UINT64_C(0xCCAB1694D61F2B9B), UINT64_C(0x100A3AB68895C586),
+ UINT64_C(0x74E84FD06A0BF6A0), UINT64_C(0xA84963F2348118BD),
+ UINT64_C(0x5C22C10D5F64E503), UINT64_C(0x8083ED2F01EE0B1E),
+ UINT64_C(0xE4619849E3703838), UINT64_C(0x38C0B46BBDFAD625),
+ UINT64_C(0x2CA57385274D5F74), UINT64_C(0xF0045FA779C7B169),
+ UINT64_C(0x94E62AC19B59824F), UINT64_C(0x484706E3C5D36C52),
+ UINT64_C(0xF92F6091673CA1A0), UINT64_C(0x258E4CB339B64FBD),
+ UINT64_C(0x416C39D5DB287C9B), UINT64_C(0x9DCD15F785A29286),
+ UINT64_C(0x89A8D2191F151BD7), UINT64_C(0x5509FE3B419FF5CA),
+ UINT64_C(0x31EB8B5DA301C6EC), UINT64_C(0xED4AA77FFD8B28F1),
+ UINT64_C(0x19210580966ED54F), UINT64_C(0xC58029A2C8E43B52),
+ UINT64_C(0xA1625CC42A7A0874), UINT64_C(0x7DC370E674F0E669),
+ UINT64_C(0x69A6B708EE476F38), UINT64_C(0xB5079B2AB0CD8125),
+ UINT64_C(0xD1E5EE4C5253B203), UINT64_C(0x0D44C26E0CD95C1E)
+ }
+};
diff --git a/src/liblzma/check/crc64_table_le.h b/src/liblzma/check/crc64_table_le.h
new file mode 100644
index 0000000..1196b31
--- /dev/null
+++ b/src/liblzma/check/crc64_table_le.h
@@ -0,0 +1,521 @@
+/* This file has been automatically generated by crc64_tablegen.c. */
+
+const uint64_t lzma_crc64_table[4][256] = {
+ {
+ UINT64_C(0x0000000000000000), UINT64_C(0xB32E4CBE03A75F6F),
+ UINT64_C(0xF4843657A840A05B), UINT64_C(0x47AA7AE9ABE7FF34),
+ UINT64_C(0x7BD0C384FF8F5E33), UINT64_C(0xC8FE8F3AFC28015C),
+ UINT64_C(0x8F54F5D357CFFE68), UINT64_C(0x3C7AB96D5468A107),
+ UINT64_C(0xF7A18709FF1EBC66), UINT64_C(0x448FCBB7FCB9E309),
+ UINT64_C(0x0325B15E575E1C3D), UINT64_C(0xB00BFDE054F94352),
+ UINT64_C(0x8C71448D0091E255), UINT64_C(0x3F5F08330336BD3A),
+ UINT64_C(0x78F572DAA8D1420E), UINT64_C(0xCBDB3E64AB761D61),
+ UINT64_C(0x7D9BA13851336649), UINT64_C(0xCEB5ED8652943926),
+ UINT64_C(0x891F976FF973C612), UINT64_C(0x3A31DBD1FAD4997D),
+ UINT64_C(0x064B62BCAEBC387A), UINT64_C(0xB5652E02AD1B6715),
+ UINT64_C(0xF2CF54EB06FC9821), UINT64_C(0x41E11855055BC74E),
+ UINT64_C(0x8A3A2631AE2DDA2F), UINT64_C(0x39146A8FAD8A8540),
+ UINT64_C(0x7EBE1066066D7A74), UINT64_C(0xCD905CD805CA251B),
+ UINT64_C(0xF1EAE5B551A2841C), UINT64_C(0x42C4A90B5205DB73),
+ UINT64_C(0x056ED3E2F9E22447), UINT64_C(0xB6409F5CFA457B28),
+ UINT64_C(0xFB374270A266CC92), UINT64_C(0x48190ECEA1C193FD),
+ UINT64_C(0x0FB374270A266CC9), UINT64_C(0xBC9D3899098133A6),
+ UINT64_C(0x80E781F45DE992A1), UINT64_C(0x33C9CD4A5E4ECDCE),
+ UINT64_C(0x7463B7A3F5A932FA), UINT64_C(0xC74DFB1DF60E6D95),
+ UINT64_C(0x0C96C5795D7870F4), UINT64_C(0xBFB889C75EDF2F9B),
+ UINT64_C(0xF812F32EF538D0AF), UINT64_C(0x4B3CBF90F69F8FC0),
+ UINT64_C(0x774606FDA2F72EC7), UINT64_C(0xC4684A43A15071A8),
+ UINT64_C(0x83C230AA0AB78E9C), UINT64_C(0x30EC7C140910D1F3),
+ UINT64_C(0x86ACE348F355AADB), UINT64_C(0x3582AFF6F0F2F5B4),
+ UINT64_C(0x7228D51F5B150A80), UINT64_C(0xC10699A158B255EF),
+ UINT64_C(0xFD7C20CC0CDAF4E8), UINT64_C(0x4E526C720F7DAB87),
+ UINT64_C(0x09F8169BA49A54B3), UINT64_C(0xBAD65A25A73D0BDC),
+ UINT64_C(0x710D64410C4B16BD), UINT64_C(0xC22328FF0FEC49D2),
+ UINT64_C(0x85895216A40BB6E6), UINT64_C(0x36A71EA8A7ACE989),
+ UINT64_C(0x0ADDA7C5F3C4488E), UINT64_C(0xB9F3EB7BF06317E1),
+ UINT64_C(0xFE5991925B84E8D5), UINT64_C(0x4D77DD2C5823B7BA),
+ UINT64_C(0x64B62BCAEBC387A1), UINT64_C(0xD7986774E864D8CE),
+ UINT64_C(0x90321D9D438327FA), UINT64_C(0x231C512340247895),
+ UINT64_C(0x1F66E84E144CD992), UINT64_C(0xAC48A4F017EB86FD),
+ UINT64_C(0xEBE2DE19BC0C79C9), UINT64_C(0x58CC92A7BFAB26A6),
+ UINT64_C(0x9317ACC314DD3BC7), UINT64_C(0x2039E07D177A64A8),
+ UINT64_C(0x67939A94BC9D9B9C), UINT64_C(0xD4BDD62ABF3AC4F3),
+ UINT64_C(0xE8C76F47EB5265F4), UINT64_C(0x5BE923F9E8F53A9B),
+ UINT64_C(0x1C4359104312C5AF), UINT64_C(0xAF6D15AE40B59AC0),
+ UINT64_C(0x192D8AF2BAF0E1E8), UINT64_C(0xAA03C64CB957BE87),
+ UINT64_C(0xEDA9BCA512B041B3), UINT64_C(0x5E87F01B11171EDC),
+ UINT64_C(0x62FD4976457FBFDB), UINT64_C(0xD1D305C846D8E0B4),
+ UINT64_C(0x96797F21ED3F1F80), UINT64_C(0x2557339FEE9840EF),
+ UINT64_C(0xEE8C0DFB45EE5D8E), UINT64_C(0x5DA24145464902E1),
+ UINT64_C(0x1A083BACEDAEFDD5), UINT64_C(0xA9267712EE09A2BA),
+ UINT64_C(0x955CCE7FBA6103BD), UINT64_C(0x267282C1B9C65CD2),
+ UINT64_C(0x61D8F8281221A3E6), UINT64_C(0xD2F6B4961186FC89),
+ UINT64_C(0x9F8169BA49A54B33), UINT64_C(0x2CAF25044A02145C),
+ UINT64_C(0x6B055FEDE1E5EB68), UINT64_C(0xD82B1353E242B407),
+ UINT64_C(0xE451AA3EB62A1500), UINT64_C(0x577FE680B58D4A6F),
+ UINT64_C(0x10D59C691E6AB55B), UINT64_C(0xA3FBD0D71DCDEA34),
+ UINT64_C(0x6820EEB3B6BBF755), UINT64_C(0xDB0EA20DB51CA83A),
+ UINT64_C(0x9CA4D8E41EFB570E), UINT64_C(0x2F8A945A1D5C0861),
+ UINT64_C(0x13F02D374934A966), UINT64_C(0xA0DE61894A93F609),
+ UINT64_C(0xE7741B60E174093D), UINT64_C(0x545A57DEE2D35652),
+ UINT64_C(0xE21AC88218962D7A), UINT64_C(0x5134843C1B317215),
+ UINT64_C(0x169EFED5B0D68D21), UINT64_C(0xA5B0B26BB371D24E),
+ UINT64_C(0x99CA0B06E7197349), UINT64_C(0x2AE447B8E4BE2C26),
+ UINT64_C(0x6D4E3D514F59D312), UINT64_C(0xDE6071EF4CFE8C7D),
+ UINT64_C(0x15BB4F8BE788911C), UINT64_C(0xA6950335E42FCE73),
+ UINT64_C(0xE13F79DC4FC83147), UINT64_C(0x521135624C6F6E28),
+ UINT64_C(0x6E6B8C0F1807CF2F), UINT64_C(0xDD45C0B11BA09040),
+ UINT64_C(0x9AEFBA58B0476F74), UINT64_C(0x29C1F6E6B3E0301B),
+ UINT64_C(0xC96C5795D7870F42), UINT64_C(0x7A421B2BD420502D),
+ UINT64_C(0x3DE861C27FC7AF19), UINT64_C(0x8EC62D7C7C60F076),
+ UINT64_C(0xB2BC941128085171), UINT64_C(0x0192D8AF2BAF0E1E),
+ UINT64_C(0x4638A2468048F12A), UINT64_C(0xF516EEF883EFAE45),
+ UINT64_C(0x3ECDD09C2899B324), UINT64_C(0x8DE39C222B3EEC4B),
+ UINT64_C(0xCA49E6CB80D9137F), UINT64_C(0x7967AA75837E4C10),
+ UINT64_C(0x451D1318D716ED17), UINT64_C(0xF6335FA6D4B1B278),
+ UINT64_C(0xB199254F7F564D4C), UINT64_C(0x02B769F17CF11223),
+ UINT64_C(0xB4F7F6AD86B4690B), UINT64_C(0x07D9BA1385133664),
+ UINT64_C(0x4073C0FA2EF4C950), UINT64_C(0xF35D8C442D53963F),
+ UINT64_C(0xCF273529793B3738), UINT64_C(0x7C0979977A9C6857),
+ UINT64_C(0x3BA3037ED17B9763), UINT64_C(0x888D4FC0D2DCC80C),
+ UINT64_C(0x435671A479AAD56D), UINT64_C(0xF0783D1A7A0D8A02),
+ UINT64_C(0xB7D247F3D1EA7536), UINT64_C(0x04FC0B4DD24D2A59),
+ UINT64_C(0x3886B22086258B5E), UINT64_C(0x8BA8FE9E8582D431),
+ UINT64_C(0xCC0284772E652B05), UINT64_C(0x7F2CC8C92DC2746A),
+ UINT64_C(0x325B15E575E1C3D0), UINT64_C(0x8175595B76469CBF),
+ UINT64_C(0xC6DF23B2DDA1638B), UINT64_C(0x75F16F0CDE063CE4),
+ UINT64_C(0x498BD6618A6E9DE3), UINT64_C(0xFAA59ADF89C9C28C),
+ UINT64_C(0xBD0FE036222E3DB8), UINT64_C(0x0E21AC88218962D7),
+ UINT64_C(0xC5FA92EC8AFF7FB6), UINT64_C(0x76D4DE52895820D9),
+ UINT64_C(0x317EA4BB22BFDFED), UINT64_C(0x8250E80521188082),
+ UINT64_C(0xBE2A516875702185), UINT64_C(0x0D041DD676D77EEA),
+ UINT64_C(0x4AAE673FDD3081DE), UINT64_C(0xF9802B81DE97DEB1),
+ UINT64_C(0x4FC0B4DD24D2A599), UINT64_C(0xFCEEF8632775FAF6),
+ UINT64_C(0xBB44828A8C9205C2), UINT64_C(0x086ACE348F355AAD),
+ UINT64_C(0x34107759DB5DFBAA), UINT64_C(0x873E3BE7D8FAA4C5),
+ UINT64_C(0xC094410E731D5BF1), UINT64_C(0x73BA0DB070BA049E),
+ UINT64_C(0xB86133D4DBCC19FF), UINT64_C(0x0B4F7F6AD86B4690),
+ UINT64_C(0x4CE50583738CB9A4), UINT64_C(0xFFCB493D702BE6CB),
+ UINT64_C(0xC3B1F050244347CC), UINT64_C(0x709FBCEE27E418A3),
+ UINT64_C(0x3735C6078C03E797), UINT64_C(0x841B8AB98FA4B8F8),
+ UINT64_C(0xADDA7C5F3C4488E3), UINT64_C(0x1EF430E13FE3D78C),
+ UINT64_C(0x595E4A08940428B8), UINT64_C(0xEA7006B697A377D7),
+ UINT64_C(0xD60ABFDBC3CBD6D0), UINT64_C(0x6524F365C06C89BF),
+ UINT64_C(0x228E898C6B8B768B), UINT64_C(0x91A0C532682C29E4),
+ UINT64_C(0x5A7BFB56C35A3485), UINT64_C(0xE955B7E8C0FD6BEA),
+ UINT64_C(0xAEFFCD016B1A94DE), UINT64_C(0x1DD181BF68BDCBB1),
+ UINT64_C(0x21AB38D23CD56AB6), UINT64_C(0x9285746C3F7235D9),
+ UINT64_C(0xD52F0E859495CAED), UINT64_C(0x6601423B97329582),
+ UINT64_C(0xD041DD676D77EEAA), UINT64_C(0x636F91D96ED0B1C5),
+ UINT64_C(0x24C5EB30C5374EF1), UINT64_C(0x97EBA78EC690119E),
+ UINT64_C(0xAB911EE392F8B099), UINT64_C(0x18BF525D915FEFF6),
+ UINT64_C(0x5F1528B43AB810C2), UINT64_C(0xEC3B640A391F4FAD),
+ UINT64_C(0x27E05A6E926952CC), UINT64_C(0x94CE16D091CE0DA3),
+ UINT64_C(0xD3646C393A29F297), UINT64_C(0x604A2087398EADF8),
+ UINT64_C(0x5C3099EA6DE60CFF), UINT64_C(0xEF1ED5546E415390),
+ UINT64_C(0xA8B4AFBDC5A6ACA4), UINT64_C(0x1B9AE303C601F3CB),
+ UINT64_C(0x56ED3E2F9E224471), UINT64_C(0xE5C372919D851B1E),
+ UINT64_C(0xA26908783662E42A), UINT64_C(0x114744C635C5BB45),
+ UINT64_C(0x2D3DFDAB61AD1A42), UINT64_C(0x9E13B115620A452D),
+ UINT64_C(0xD9B9CBFCC9EDBA19), UINT64_C(0x6A978742CA4AE576),
+ UINT64_C(0xA14CB926613CF817), UINT64_C(0x1262F598629BA778),
+ UINT64_C(0x55C88F71C97C584C), UINT64_C(0xE6E6C3CFCADB0723),
+ UINT64_C(0xDA9C7AA29EB3A624), UINT64_C(0x69B2361C9D14F94B),
+ UINT64_C(0x2E184CF536F3067F), UINT64_C(0x9D36004B35545910),
+ UINT64_C(0x2B769F17CF112238), UINT64_C(0x9858D3A9CCB67D57),
+ UINT64_C(0xDFF2A94067518263), UINT64_C(0x6CDCE5FE64F6DD0C),
+ UINT64_C(0x50A65C93309E7C0B), UINT64_C(0xE388102D33392364),
+ UINT64_C(0xA4226AC498DEDC50), UINT64_C(0x170C267A9B79833F),
+ UINT64_C(0xDCD7181E300F9E5E), UINT64_C(0x6FF954A033A8C131),
+ UINT64_C(0x28532E49984F3E05), UINT64_C(0x9B7D62F79BE8616A),
+ UINT64_C(0xA707DB9ACF80C06D), UINT64_C(0x14299724CC279F02),
+ UINT64_C(0x5383EDCD67C06036), UINT64_C(0xE0ADA17364673F59)
+ }, {
+ UINT64_C(0x0000000000000000), UINT64_C(0x54E979925CD0F10D),
+ UINT64_C(0xA9D2F324B9A1E21A), UINT64_C(0xFD3B8AB6E5711317),
+ UINT64_C(0xC17D4962DC4DDAB1), UINT64_C(0x959430F0809D2BBC),
+ UINT64_C(0x68AFBA4665EC38AB), UINT64_C(0x3C46C3D4393CC9A6),
+ UINT64_C(0x10223DEE1795ABE7), UINT64_C(0x44CB447C4B455AEA),
+ UINT64_C(0xB9F0CECAAE3449FD), UINT64_C(0xED19B758F2E4B8F0),
+ UINT64_C(0xD15F748CCBD87156), UINT64_C(0x85B60D1E9708805B),
+ UINT64_C(0x788D87A87279934C), UINT64_C(0x2C64FE3A2EA96241),
+ UINT64_C(0x20447BDC2F2B57CE), UINT64_C(0x74AD024E73FBA6C3),
+ UINT64_C(0x899688F8968AB5D4), UINT64_C(0xDD7FF16ACA5A44D9),
+ UINT64_C(0xE13932BEF3668D7F), UINT64_C(0xB5D04B2CAFB67C72),
+ UINT64_C(0x48EBC19A4AC76F65), UINT64_C(0x1C02B80816179E68),
+ UINT64_C(0x3066463238BEFC29), UINT64_C(0x648F3FA0646E0D24),
+ UINT64_C(0x99B4B516811F1E33), UINT64_C(0xCD5DCC84DDCFEF3E),
+ UINT64_C(0xF11B0F50E4F32698), UINT64_C(0xA5F276C2B823D795),
+ UINT64_C(0x58C9FC745D52C482), UINT64_C(0x0C2085E60182358F),
+ UINT64_C(0x4088F7B85E56AF9C), UINT64_C(0x14618E2A02865E91),
+ UINT64_C(0xE95A049CE7F74D86), UINT64_C(0xBDB37D0EBB27BC8B),
+ UINT64_C(0x81F5BEDA821B752D), UINT64_C(0xD51CC748DECB8420),
+ UINT64_C(0x28274DFE3BBA9737), UINT64_C(0x7CCE346C676A663A),
+ UINT64_C(0x50AACA5649C3047B), UINT64_C(0x0443B3C41513F576),
+ UINT64_C(0xF9783972F062E661), UINT64_C(0xAD9140E0ACB2176C),
+ UINT64_C(0x91D78334958EDECA), UINT64_C(0xC53EFAA6C95E2FC7),
+ UINT64_C(0x380570102C2F3CD0), UINT64_C(0x6CEC098270FFCDDD),
+ UINT64_C(0x60CC8C64717DF852), UINT64_C(0x3425F5F62DAD095F),
+ UINT64_C(0xC91E7F40C8DC1A48), UINT64_C(0x9DF706D2940CEB45),
+ UINT64_C(0xA1B1C506AD3022E3), UINT64_C(0xF558BC94F1E0D3EE),
+ UINT64_C(0x086336221491C0F9), UINT64_C(0x5C8A4FB0484131F4),
+ UINT64_C(0x70EEB18A66E853B5), UINT64_C(0x2407C8183A38A2B8),
+ UINT64_C(0xD93C42AEDF49B1AF), UINT64_C(0x8DD53B3C839940A2),
+ UINT64_C(0xB193F8E8BAA58904), UINT64_C(0xE57A817AE6757809),
+ UINT64_C(0x18410BCC03046B1E), UINT64_C(0x4CA8725E5FD49A13),
+ UINT64_C(0x8111EF70BCAD5F38), UINT64_C(0xD5F896E2E07DAE35),
+ UINT64_C(0x28C31C54050CBD22), UINT64_C(0x7C2A65C659DC4C2F),
+ UINT64_C(0x406CA61260E08589), UINT64_C(0x1485DF803C307484),
+ UINT64_C(0xE9BE5536D9416793), UINT64_C(0xBD572CA48591969E),
+ UINT64_C(0x9133D29EAB38F4DF), UINT64_C(0xC5DAAB0CF7E805D2),
+ UINT64_C(0x38E121BA129916C5), UINT64_C(0x6C0858284E49E7C8),
+ UINT64_C(0x504E9BFC77752E6E), UINT64_C(0x04A7E26E2BA5DF63),
+ UINT64_C(0xF99C68D8CED4CC74), UINT64_C(0xAD75114A92043D79),
+ UINT64_C(0xA15594AC938608F6), UINT64_C(0xF5BCED3ECF56F9FB),
+ UINT64_C(0x088767882A27EAEC), UINT64_C(0x5C6E1E1A76F71BE1),
+ UINT64_C(0x6028DDCE4FCBD247), UINT64_C(0x34C1A45C131B234A),
+ UINT64_C(0xC9FA2EEAF66A305D), UINT64_C(0x9D135778AABAC150),
+ UINT64_C(0xB177A9428413A311), UINT64_C(0xE59ED0D0D8C3521C),
+ UINT64_C(0x18A55A663DB2410B), UINT64_C(0x4C4C23F46162B006),
+ UINT64_C(0x700AE020585E79A0), UINT64_C(0x24E399B2048E88AD),
+ UINT64_C(0xD9D81304E1FF9BBA), UINT64_C(0x8D316A96BD2F6AB7),
+ UINT64_C(0xC19918C8E2FBF0A4), UINT64_C(0x9570615ABE2B01A9),
+ UINT64_C(0x684BEBEC5B5A12BE), UINT64_C(0x3CA2927E078AE3B3),
+ UINT64_C(0x00E451AA3EB62A15), UINT64_C(0x540D28386266DB18),
+ UINT64_C(0xA936A28E8717C80F), UINT64_C(0xFDDFDB1CDBC73902),
+ UINT64_C(0xD1BB2526F56E5B43), UINT64_C(0x85525CB4A9BEAA4E),
+ UINT64_C(0x7869D6024CCFB959), UINT64_C(0x2C80AF90101F4854),
+ UINT64_C(0x10C66C44292381F2), UINT64_C(0x442F15D675F370FF),
+ UINT64_C(0xB9149F60908263E8), UINT64_C(0xEDFDE6F2CC5292E5),
+ UINT64_C(0xE1DD6314CDD0A76A), UINT64_C(0xB5341A8691005667),
+ UINT64_C(0x480F903074714570), UINT64_C(0x1CE6E9A228A1B47D),
+ UINT64_C(0x20A02A76119D7DDB), UINT64_C(0x744953E44D4D8CD6),
+ UINT64_C(0x8972D952A83C9FC1), UINT64_C(0xDD9BA0C0F4EC6ECC),
+ UINT64_C(0xF1FF5EFADA450C8D), UINT64_C(0xA51627688695FD80),
+ UINT64_C(0x582DADDE63E4EE97), UINT64_C(0x0CC4D44C3F341F9A),
+ UINT64_C(0x308217980608D63C), UINT64_C(0x646B6E0A5AD82731),
+ UINT64_C(0x9950E4BCBFA93426), UINT64_C(0xCDB99D2EE379C52B),
+ UINT64_C(0x90FB71CAD654A0F5), UINT64_C(0xC41208588A8451F8),
+ UINT64_C(0x392982EE6FF542EF), UINT64_C(0x6DC0FB7C3325B3E2),
+ UINT64_C(0x518638A80A197A44), UINT64_C(0x056F413A56C98B49),
+ UINT64_C(0xF854CB8CB3B8985E), UINT64_C(0xACBDB21EEF686953),
+ UINT64_C(0x80D94C24C1C10B12), UINT64_C(0xD43035B69D11FA1F),
+ UINT64_C(0x290BBF007860E908), UINT64_C(0x7DE2C69224B01805),
+ UINT64_C(0x41A405461D8CD1A3), UINT64_C(0x154D7CD4415C20AE),
+ UINT64_C(0xE876F662A42D33B9), UINT64_C(0xBC9F8FF0F8FDC2B4),
+ UINT64_C(0xB0BF0A16F97FF73B), UINT64_C(0xE4567384A5AF0636),
+ UINT64_C(0x196DF93240DE1521), UINT64_C(0x4D8480A01C0EE42C),
+ UINT64_C(0x71C2437425322D8A), UINT64_C(0x252B3AE679E2DC87),
+ UINT64_C(0xD810B0509C93CF90), UINT64_C(0x8CF9C9C2C0433E9D),
+ UINT64_C(0xA09D37F8EEEA5CDC), UINT64_C(0xF4744E6AB23AADD1),
+ UINT64_C(0x094FC4DC574BBEC6), UINT64_C(0x5DA6BD4E0B9B4FCB),
+ UINT64_C(0x61E07E9A32A7866D), UINT64_C(0x350907086E777760),
+ UINT64_C(0xC8328DBE8B066477), UINT64_C(0x9CDBF42CD7D6957A),
+ UINT64_C(0xD073867288020F69), UINT64_C(0x849AFFE0D4D2FE64),
+ UINT64_C(0x79A1755631A3ED73), UINT64_C(0x2D480CC46D731C7E),
+ UINT64_C(0x110ECF10544FD5D8), UINT64_C(0x45E7B682089F24D5),
+ UINT64_C(0xB8DC3C34EDEE37C2), UINT64_C(0xEC3545A6B13EC6CF),
+ UINT64_C(0xC051BB9C9F97A48E), UINT64_C(0x94B8C20EC3475583),
+ UINT64_C(0x698348B826364694), UINT64_C(0x3D6A312A7AE6B799),
+ UINT64_C(0x012CF2FE43DA7E3F), UINT64_C(0x55C58B6C1F0A8F32),
+ UINT64_C(0xA8FE01DAFA7B9C25), UINT64_C(0xFC177848A6AB6D28),
+ UINT64_C(0xF037FDAEA72958A7), UINT64_C(0xA4DE843CFBF9A9AA),
+ UINT64_C(0x59E50E8A1E88BABD), UINT64_C(0x0D0C771842584BB0),
+ UINT64_C(0x314AB4CC7B648216), UINT64_C(0x65A3CD5E27B4731B),
+ UINT64_C(0x989847E8C2C5600C), UINT64_C(0xCC713E7A9E159101),
+ UINT64_C(0xE015C040B0BCF340), UINT64_C(0xB4FCB9D2EC6C024D),
+ UINT64_C(0x49C73364091D115A), UINT64_C(0x1D2E4AF655CDE057),
+ UINT64_C(0x216889226CF129F1), UINT64_C(0x7581F0B03021D8FC),
+ UINT64_C(0x88BA7A06D550CBEB), UINT64_C(0xDC53039489803AE6),
+ UINT64_C(0x11EA9EBA6AF9FFCD), UINT64_C(0x4503E72836290EC0),
+ UINT64_C(0xB8386D9ED3581DD7), UINT64_C(0xECD1140C8F88ECDA),
+ UINT64_C(0xD097D7D8B6B4257C), UINT64_C(0x847EAE4AEA64D471),
+ UINT64_C(0x794524FC0F15C766), UINT64_C(0x2DAC5D6E53C5366B),
+ UINT64_C(0x01C8A3547D6C542A), UINT64_C(0x5521DAC621BCA527),
+ UINT64_C(0xA81A5070C4CDB630), UINT64_C(0xFCF329E2981D473D),
+ UINT64_C(0xC0B5EA36A1218E9B), UINT64_C(0x945C93A4FDF17F96),
+ UINT64_C(0x6967191218806C81), UINT64_C(0x3D8E608044509D8C),
+ UINT64_C(0x31AEE56645D2A803), UINT64_C(0x65479CF41902590E),
+ UINT64_C(0x987C1642FC734A19), UINT64_C(0xCC956FD0A0A3BB14),
+ UINT64_C(0xF0D3AC04999F72B2), UINT64_C(0xA43AD596C54F83BF),
+ UINT64_C(0x59015F20203E90A8), UINT64_C(0x0DE826B27CEE61A5),
+ UINT64_C(0x218CD888524703E4), UINT64_C(0x7565A11A0E97F2E9),
+ UINT64_C(0x885E2BACEBE6E1FE), UINT64_C(0xDCB7523EB73610F3),
+ UINT64_C(0xE0F191EA8E0AD955), UINT64_C(0xB418E878D2DA2858),
+ UINT64_C(0x492362CE37AB3B4F), UINT64_C(0x1DCA1B5C6B7BCA42),
+ UINT64_C(0x5162690234AF5051), UINT64_C(0x058B1090687FA15C),
+ UINT64_C(0xF8B09A268D0EB24B), UINT64_C(0xAC59E3B4D1DE4346),
+ UINT64_C(0x901F2060E8E28AE0), UINT64_C(0xC4F659F2B4327BED),
+ UINT64_C(0x39CDD344514368FA), UINT64_C(0x6D24AAD60D9399F7),
+ UINT64_C(0x414054EC233AFBB6), UINT64_C(0x15A92D7E7FEA0ABB),
+ UINT64_C(0xE892A7C89A9B19AC), UINT64_C(0xBC7BDE5AC64BE8A1),
+ UINT64_C(0x803D1D8EFF772107), UINT64_C(0xD4D4641CA3A7D00A),
+ UINT64_C(0x29EFEEAA46D6C31D), UINT64_C(0x7D0697381A063210),
+ UINT64_C(0x712612DE1B84079F), UINT64_C(0x25CF6B4C4754F692),
+ UINT64_C(0xD8F4E1FAA225E585), UINT64_C(0x8C1D9868FEF51488),
+ UINT64_C(0xB05B5BBCC7C9DD2E), UINT64_C(0xE4B2222E9B192C23),
+ UINT64_C(0x1989A8987E683F34), UINT64_C(0x4D60D10A22B8CE39),
+ UINT64_C(0x61042F300C11AC78), UINT64_C(0x35ED56A250C15D75),
+ UINT64_C(0xC8D6DC14B5B04E62), UINT64_C(0x9C3FA586E960BF6F),
+ UINT64_C(0xA0796652D05C76C9), UINT64_C(0xF4901FC08C8C87C4),
+ UINT64_C(0x09AB957669FD94D3), UINT64_C(0x5D42ECE4352D65DE)
+ }, {
+ UINT64_C(0x0000000000000000), UINT64_C(0x3F0BE14A916A6DCB),
+ UINT64_C(0x7E17C29522D4DB96), UINT64_C(0x411C23DFB3BEB65D),
+ UINT64_C(0xFC2F852A45A9B72C), UINT64_C(0xC3246460D4C3DAE7),
+ UINT64_C(0x823847BF677D6CBA), UINT64_C(0xBD33A6F5F6170171),
+ UINT64_C(0x6A87A57F245D70DD), UINT64_C(0x558C4435B5371D16),
+ UINT64_C(0x149067EA0689AB4B), UINT64_C(0x2B9B86A097E3C680),
+ UINT64_C(0x96A8205561F4C7F1), UINT64_C(0xA9A3C11FF09EAA3A),
+ UINT64_C(0xE8BFE2C043201C67), UINT64_C(0xD7B4038AD24A71AC),
+ UINT64_C(0xD50F4AFE48BAE1BA), UINT64_C(0xEA04ABB4D9D08C71),
+ UINT64_C(0xAB18886B6A6E3A2C), UINT64_C(0x94136921FB0457E7),
+ UINT64_C(0x2920CFD40D135696), UINT64_C(0x162B2E9E9C793B5D),
+ UINT64_C(0x57370D412FC78D00), UINT64_C(0x683CEC0BBEADE0CB),
+ UINT64_C(0xBF88EF816CE79167), UINT64_C(0x80830ECBFD8DFCAC),
+ UINT64_C(0xC19F2D144E334AF1), UINT64_C(0xFE94CC5EDF59273A),
+ UINT64_C(0x43A76AAB294E264B), UINT64_C(0x7CAC8BE1B8244B80),
+ UINT64_C(0x3DB0A83E0B9AFDDD), UINT64_C(0x02BB49749AF09016),
+ UINT64_C(0x38C63AD73E7BDDF1), UINT64_C(0x07CDDB9DAF11B03A),
+ UINT64_C(0x46D1F8421CAF0667), UINT64_C(0x79DA19088DC56BAC),
+ UINT64_C(0xC4E9BFFD7BD26ADD), UINT64_C(0xFBE25EB7EAB80716),
+ UINT64_C(0xBAFE7D685906B14B), UINT64_C(0x85F59C22C86CDC80),
+ UINT64_C(0x52419FA81A26AD2C), UINT64_C(0x6D4A7EE28B4CC0E7),
+ UINT64_C(0x2C565D3D38F276BA), UINT64_C(0x135DBC77A9981B71),
+ UINT64_C(0xAE6E1A825F8F1A00), UINT64_C(0x9165FBC8CEE577CB),
+ UINT64_C(0xD079D8177D5BC196), UINT64_C(0xEF72395DEC31AC5D),
+ UINT64_C(0xEDC9702976C13C4B), UINT64_C(0xD2C29163E7AB5180),
+ UINT64_C(0x93DEB2BC5415E7DD), UINT64_C(0xACD553F6C57F8A16),
+ UINT64_C(0x11E6F50333688B67), UINT64_C(0x2EED1449A202E6AC),
+ UINT64_C(0x6FF1379611BC50F1), UINT64_C(0x50FAD6DC80D63D3A),
+ UINT64_C(0x874ED556529C4C96), UINT64_C(0xB845341CC3F6215D),
+ UINT64_C(0xF95917C370489700), UINT64_C(0xC652F689E122FACB),
+ UINT64_C(0x7B61507C1735FBBA), UINT64_C(0x446AB136865F9671),
+ UINT64_C(0x057692E935E1202C), UINT64_C(0x3A7D73A3A48B4DE7),
+ UINT64_C(0x718C75AE7CF7BBE2), UINT64_C(0x4E8794E4ED9DD629),
+ UINT64_C(0x0F9BB73B5E236074), UINT64_C(0x30905671CF490DBF),
+ UINT64_C(0x8DA3F084395E0CCE), UINT64_C(0xB2A811CEA8346105),
+ UINT64_C(0xF3B432111B8AD758), UINT64_C(0xCCBFD35B8AE0BA93),
+ UINT64_C(0x1B0BD0D158AACB3F), UINT64_C(0x2400319BC9C0A6F4),
+ UINT64_C(0x651C12447A7E10A9), UINT64_C(0x5A17F30EEB147D62),
+ UINT64_C(0xE72455FB1D037C13), UINT64_C(0xD82FB4B18C6911D8),
+ UINT64_C(0x9933976E3FD7A785), UINT64_C(0xA6387624AEBDCA4E),
+ UINT64_C(0xA4833F50344D5A58), UINT64_C(0x9B88DE1AA5273793),
+ UINT64_C(0xDA94FDC5169981CE), UINT64_C(0xE59F1C8F87F3EC05),
+ UINT64_C(0x58ACBA7A71E4ED74), UINT64_C(0x67A75B30E08E80BF),
+ UINT64_C(0x26BB78EF533036E2), UINT64_C(0x19B099A5C25A5B29),
+ UINT64_C(0xCE049A2F10102A85), UINT64_C(0xF10F7B65817A474E),
+ UINT64_C(0xB01358BA32C4F113), UINT64_C(0x8F18B9F0A3AE9CD8),
+ UINT64_C(0x322B1F0555B99DA9), UINT64_C(0x0D20FE4FC4D3F062),
+ UINT64_C(0x4C3CDD90776D463F), UINT64_C(0x73373CDAE6072BF4),
+ UINT64_C(0x494A4F79428C6613), UINT64_C(0x7641AE33D3E60BD8),
+ UINT64_C(0x375D8DEC6058BD85), UINT64_C(0x08566CA6F132D04E),
+ UINT64_C(0xB565CA530725D13F), UINT64_C(0x8A6E2B19964FBCF4),
+ UINT64_C(0xCB7208C625F10AA9), UINT64_C(0xF479E98CB49B6762),
+ UINT64_C(0x23CDEA0666D116CE), UINT64_C(0x1CC60B4CF7BB7B05),
+ UINT64_C(0x5DDA28934405CD58), UINT64_C(0x62D1C9D9D56FA093),
+ UINT64_C(0xDFE26F2C2378A1E2), UINT64_C(0xE0E98E66B212CC29),
+ UINT64_C(0xA1F5ADB901AC7A74), UINT64_C(0x9EFE4CF390C617BF),
+ UINT64_C(0x9C4505870A3687A9), UINT64_C(0xA34EE4CD9B5CEA62),
+ UINT64_C(0xE252C71228E25C3F), UINT64_C(0xDD592658B98831F4),
+ UINT64_C(0x606A80AD4F9F3085), UINT64_C(0x5F6161E7DEF55D4E),
+ UINT64_C(0x1E7D42386D4BEB13), UINT64_C(0x2176A372FC2186D8),
+ UINT64_C(0xF6C2A0F82E6BF774), UINT64_C(0xC9C941B2BF019ABF),
+ UINT64_C(0x88D5626D0CBF2CE2), UINT64_C(0xB7DE83279DD54129),
+ UINT64_C(0x0AED25D26BC24058), UINT64_C(0x35E6C498FAA82D93),
+ UINT64_C(0x74FAE74749169BCE), UINT64_C(0x4BF1060DD87CF605),
+ UINT64_C(0xE318EB5CF9EF77C4), UINT64_C(0xDC130A1668851A0F),
+ UINT64_C(0x9D0F29C9DB3BAC52), UINT64_C(0xA204C8834A51C199),
+ UINT64_C(0x1F376E76BC46C0E8), UINT64_C(0x203C8F3C2D2CAD23),
+ UINT64_C(0x6120ACE39E921B7E), UINT64_C(0x5E2B4DA90FF876B5),
+ UINT64_C(0x899F4E23DDB20719), UINT64_C(0xB694AF694CD86AD2),
+ UINT64_C(0xF7888CB6FF66DC8F), UINT64_C(0xC8836DFC6E0CB144),
+ UINT64_C(0x75B0CB09981BB035), UINT64_C(0x4ABB2A430971DDFE),
+ UINT64_C(0x0BA7099CBACF6BA3), UINT64_C(0x34ACE8D62BA50668),
+ UINT64_C(0x3617A1A2B155967E), UINT64_C(0x091C40E8203FFBB5),
+ UINT64_C(0x4800633793814DE8), UINT64_C(0x770B827D02EB2023),
+ UINT64_C(0xCA382488F4FC2152), UINT64_C(0xF533C5C265964C99),
+ UINT64_C(0xB42FE61DD628FAC4), UINT64_C(0x8B2407574742970F),
+ UINT64_C(0x5C9004DD9508E6A3), UINT64_C(0x639BE59704628B68),
+ UINT64_C(0x2287C648B7DC3D35), UINT64_C(0x1D8C270226B650FE),
+ UINT64_C(0xA0BF81F7D0A1518F), UINT64_C(0x9FB460BD41CB3C44),
+ UINT64_C(0xDEA84362F2758A19), UINT64_C(0xE1A3A228631FE7D2),
+ UINT64_C(0xDBDED18BC794AA35), UINT64_C(0xE4D530C156FEC7FE),
+ UINT64_C(0xA5C9131EE54071A3), UINT64_C(0x9AC2F254742A1C68),
+ UINT64_C(0x27F154A1823D1D19), UINT64_C(0x18FAB5EB135770D2),
+ UINT64_C(0x59E69634A0E9C68F), UINT64_C(0x66ED777E3183AB44),
+ UINT64_C(0xB15974F4E3C9DAE8), UINT64_C(0x8E5295BE72A3B723),
+ UINT64_C(0xCF4EB661C11D017E), UINT64_C(0xF045572B50776CB5),
+ UINT64_C(0x4D76F1DEA6606DC4), UINT64_C(0x727D1094370A000F),
+ UINT64_C(0x3361334B84B4B652), UINT64_C(0x0C6AD20115DEDB99),
+ UINT64_C(0x0ED19B758F2E4B8F), UINT64_C(0x31DA7A3F1E442644),
+ UINT64_C(0x70C659E0ADFA9019), UINT64_C(0x4FCDB8AA3C90FDD2),
+ UINT64_C(0xF2FE1E5FCA87FCA3), UINT64_C(0xCDF5FF155BED9168),
+ UINT64_C(0x8CE9DCCAE8532735), UINT64_C(0xB3E23D8079394AFE),
+ UINT64_C(0x64563E0AAB733B52), UINT64_C(0x5B5DDF403A195699),
+ UINT64_C(0x1A41FC9F89A7E0C4), UINT64_C(0x254A1DD518CD8D0F),
+ UINT64_C(0x9879BB20EEDA8C7E), UINT64_C(0xA7725A6A7FB0E1B5),
+ UINT64_C(0xE66E79B5CC0E57E8), UINT64_C(0xD96598FF5D643A23),
+ UINT64_C(0x92949EF28518CC26), UINT64_C(0xAD9F7FB81472A1ED),
+ UINT64_C(0xEC835C67A7CC17B0), UINT64_C(0xD388BD2D36A67A7B),
+ UINT64_C(0x6EBB1BD8C0B17B0A), UINT64_C(0x51B0FA9251DB16C1),
+ UINT64_C(0x10ACD94DE265A09C), UINT64_C(0x2FA73807730FCD57),
+ UINT64_C(0xF8133B8DA145BCFB), UINT64_C(0xC718DAC7302FD130),
+ UINT64_C(0x8604F9188391676D), UINT64_C(0xB90F185212FB0AA6),
+ UINT64_C(0x043CBEA7E4EC0BD7), UINT64_C(0x3B375FED7586661C),
+ UINT64_C(0x7A2B7C32C638D041), UINT64_C(0x45209D785752BD8A),
+ UINT64_C(0x479BD40CCDA22D9C), UINT64_C(0x789035465CC84057),
+ UINT64_C(0x398C1699EF76F60A), UINT64_C(0x0687F7D37E1C9BC1),
+ UINT64_C(0xBBB45126880B9AB0), UINT64_C(0x84BFB06C1961F77B),
+ UINT64_C(0xC5A393B3AADF4126), UINT64_C(0xFAA872F93BB52CED),
+ UINT64_C(0x2D1C7173E9FF5D41), UINT64_C(0x121790397895308A),
+ UINT64_C(0x530BB3E6CB2B86D7), UINT64_C(0x6C0052AC5A41EB1C),
+ UINT64_C(0xD133F459AC56EA6D), UINT64_C(0xEE3815133D3C87A6),
+ UINT64_C(0xAF2436CC8E8231FB), UINT64_C(0x902FD7861FE85C30),
+ UINT64_C(0xAA52A425BB6311D7), UINT64_C(0x9559456F2A097C1C),
+ UINT64_C(0xD44566B099B7CA41), UINT64_C(0xEB4E87FA08DDA78A),
+ UINT64_C(0x567D210FFECAA6FB), UINT64_C(0x6976C0456FA0CB30),
+ UINT64_C(0x286AE39ADC1E7D6D), UINT64_C(0x176102D04D7410A6),
+ UINT64_C(0xC0D5015A9F3E610A), UINT64_C(0xFFDEE0100E540CC1),
+ UINT64_C(0xBEC2C3CFBDEABA9C), UINT64_C(0x81C922852C80D757),
+ UINT64_C(0x3CFA8470DA97D626), UINT64_C(0x03F1653A4BFDBBED),
+ UINT64_C(0x42ED46E5F8430DB0), UINT64_C(0x7DE6A7AF6929607B),
+ UINT64_C(0x7F5DEEDBF3D9F06D), UINT64_C(0x40560F9162B39DA6),
+ UINT64_C(0x014A2C4ED10D2BFB), UINT64_C(0x3E41CD0440674630),
+ UINT64_C(0x83726BF1B6704741), UINT64_C(0xBC798ABB271A2A8A),
+ UINT64_C(0xFD65A96494A49CD7), UINT64_C(0xC26E482E05CEF11C),
+ UINT64_C(0x15DA4BA4D78480B0), UINT64_C(0x2AD1AAEE46EEED7B),
+ UINT64_C(0x6BCD8931F5505B26), UINT64_C(0x54C6687B643A36ED),
+ UINT64_C(0xE9F5CE8E922D379C), UINT64_C(0xD6FE2FC403475A57),
+ UINT64_C(0x97E20C1BB0F9EC0A), UINT64_C(0xA8E9ED51219381C1)
+ }, {
+ UINT64_C(0x0000000000000000), UINT64_C(0x1DEE8A5E222CA1DC),
+ UINT64_C(0x3BDD14BC445943B8), UINT64_C(0x26339EE26675E264),
+ UINT64_C(0x77BA297888B28770), UINT64_C(0x6A54A326AA9E26AC),
+ UINT64_C(0x4C673DC4CCEBC4C8), UINT64_C(0x5189B79AEEC76514),
+ UINT64_C(0xEF7452F111650EE0), UINT64_C(0xF29AD8AF3349AF3C),
+ UINT64_C(0xD4A9464D553C4D58), UINT64_C(0xC947CC137710EC84),
+ UINT64_C(0x98CE7B8999D78990), UINT64_C(0x8520F1D7BBFB284C),
+ UINT64_C(0xA3136F35DD8ECA28), UINT64_C(0xBEFDE56BFFA26BF4),
+ UINT64_C(0x4C300AC98DC40345), UINT64_C(0x51DE8097AFE8A299),
+ UINT64_C(0x77ED1E75C99D40FD), UINT64_C(0x6A03942BEBB1E121),
+ UINT64_C(0x3B8A23B105768435), UINT64_C(0x2664A9EF275A25E9),
+ UINT64_C(0x0057370D412FC78D), UINT64_C(0x1DB9BD5363036651),
+ UINT64_C(0xA34458389CA10DA5), UINT64_C(0xBEAAD266BE8DAC79),
+ UINT64_C(0x98994C84D8F84E1D), UINT64_C(0x8577C6DAFAD4EFC1),
+ UINT64_C(0xD4FE714014138AD5), UINT64_C(0xC910FB1E363F2B09),
+ UINT64_C(0xEF2365FC504AC96D), UINT64_C(0xF2CDEFA2726668B1),
+ UINT64_C(0x986015931B88068A), UINT64_C(0x858E9FCD39A4A756),
+ UINT64_C(0xA3BD012F5FD14532), UINT64_C(0xBE538B717DFDE4EE),
+ UINT64_C(0xEFDA3CEB933A81FA), UINT64_C(0xF234B6B5B1162026),
+ UINT64_C(0xD4072857D763C242), UINT64_C(0xC9E9A209F54F639E),
+ UINT64_C(0x771447620AED086A), UINT64_C(0x6AFACD3C28C1A9B6),
+ UINT64_C(0x4CC953DE4EB44BD2), UINT64_C(0x5127D9806C98EA0E),
+ UINT64_C(0x00AE6E1A825F8F1A), UINT64_C(0x1D40E444A0732EC6),
+ UINT64_C(0x3B737AA6C606CCA2), UINT64_C(0x269DF0F8E42A6D7E),
+ UINT64_C(0xD4501F5A964C05CF), UINT64_C(0xC9BE9504B460A413),
+ UINT64_C(0xEF8D0BE6D2154677), UINT64_C(0xF26381B8F039E7AB),
+ UINT64_C(0xA3EA36221EFE82BF), UINT64_C(0xBE04BC7C3CD22363),
+ UINT64_C(0x9837229E5AA7C107), UINT64_C(0x85D9A8C0788B60DB),
+ UINT64_C(0x3B244DAB87290B2F), UINT64_C(0x26CAC7F5A505AAF3),
+ UINT64_C(0x00F95917C3704897), UINT64_C(0x1D17D349E15CE94B),
+ UINT64_C(0x4C9E64D30F9B8C5F), UINT64_C(0x5170EE8D2DB72D83),
+ UINT64_C(0x7743706F4BC2CFE7), UINT64_C(0x6AADFA3169EE6E3B),
+ UINT64_C(0xA218840D981E1391), UINT64_C(0xBFF60E53BA32B24D),
+ UINT64_C(0x99C590B1DC475029), UINT64_C(0x842B1AEFFE6BF1F5),
+ UINT64_C(0xD5A2AD7510AC94E1), UINT64_C(0xC84C272B3280353D),
+ UINT64_C(0xEE7FB9C954F5D759), UINT64_C(0xF391339776D97685),
+ UINT64_C(0x4D6CD6FC897B1D71), UINT64_C(0x50825CA2AB57BCAD),
+ UINT64_C(0x76B1C240CD225EC9), UINT64_C(0x6B5F481EEF0EFF15),
+ UINT64_C(0x3AD6FF8401C99A01), UINT64_C(0x273875DA23E53BDD),
+ UINT64_C(0x010BEB384590D9B9), UINT64_C(0x1CE5616667BC7865),
+ UINT64_C(0xEE288EC415DA10D4), UINT64_C(0xF3C6049A37F6B108),
+ UINT64_C(0xD5F59A785183536C), UINT64_C(0xC81B102673AFF2B0),
+ UINT64_C(0x9992A7BC9D6897A4), UINT64_C(0x847C2DE2BF443678),
+ UINT64_C(0xA24FB300D931D41C), UINT64_C(0xBFA1395EFB1D75C0),
+ UINT64_C(0x015CDC3504BF1E34), UINT64_C(0x1CB2566B2693BFE8),
+ UINT64_C(0x3A81C88940E65D8C), UINT64_C(0x276F42D762CAFC50),
+ UINT64_C(0x76E6F54D8C0D9944), UINT64_C(0x6B087F13AE213898),
+ UINT64_C(0x4D3BE1F1C854DAFC), UINT64_C(0x50D56BAFEA787B20),
+ UINT64_C(0x3A78919E8396151B), UINT64_C(0x27961BC0A1BAB4C7),
+ UINT64_C(0x01A58522C7CF56A3), UINT64_C(0x1C4B0F7CE5E3F77F),
+ UINT64_C(0x4DC2B8E60B24926B), UINT64_C(0x502C32B8290833B7),
+ UINT64_C(0x761FAC5A4F7DD1D3), UINT64_C(0x6BF126046D51700F),
+ UINT64_C(0xD50CC36F92F31BFB), UINT64_C(0xC8E24931B0DFBA27),
+ UINT64_C(0xEED1D7D3D6AA5843), UINT64_C(0xF33F5D8DF486F99F),
+ UINT64_C(0xA2B6EA171A419C8B), UINT64_C(0xBF586049386D3D57),
+ UINT64_C(0x996BFEAB5E18DF33), UINT64_C(0x848574F57C347EEF),
+ UINT64_C(0x76489B570E52165E), UINT64_C(0x6BA611092C7EB782),
+ UINT64_C(0x4D958FEB4A0B55E6), UINT64_C(0x507B05B56827F43A),
+ UINT64_C(0x01F2B22F86E0912E), UINT64_C(0x1C1C3871A4CC30F2),
+ UINT64_C(0x3A2FA693C2B9D296), UINT64_C(0x27C12CCDE095734A),
+ UINT64_C(0x993CC9A61F3718BE), UINT64_C(0x84D243F83D1BB962),
+ UINT64_C(0xA2E1DD1A5B6E5B06), UINT64_C(0xBF0F57447942FADA),
+ UINT64_C(0xEE86E0DE97859FCE), UINT64_C(0xF3686A80B5A93E12),
+ UINT64_C(0xD55BF462D3DCDC76), UINT64_C(0xC8B57E3CF1F07DAA),
+ UINT64_C(0xD6E9A7309F3239A7), UINT64_C(0xCB072D6EBD1E987B),
+ UINT64_C(0xED34B38CDB6B7A1F), UINT64_C(0xF0DA39D2F947DBC3),
+ UINT64_C(0xA1538E481780BED7), UINT64_C(0xBCBD041635AC1F0B),
+ UINT64_C(0x9A8E9AF453D9FD6F), UINT64_C(0x876010AA71F55CB3),
+ UINT64_C(0x399DF5C18E573747), UINT64_C(0x24737F9FAC7B969B),
+ UINT64_C(0x0240E17DCA0E74FF), UINT64_C(0x1FAE6B23E822D523),
+ UINT64_C(0x4E27DCB906E5B037), UINT64_C(0x53C956E724C911EB),
+ UINT64_C(0x75FAC80542BCF38F), UINT64_C(0x6814425B60905253),
+ UINT64_C(0x9AD9ADF912F63AE2), UINT64_C(0x873727A730DA9B3E),
+ UINT64_C(0xA104B94556AF795A), UINT64_C(0xBCEA331B7483D886),
+ UINT64_C(0xED6384819A44BD92), UINT64_C(0xF08D0EDFB8681C4E),
+ UINT64_C(0xD6BE903DDE1DFE2A), UINT64_C(0xCB501A63FC315FF6),
+ UINT64_C(0x75ADFF0803933402), UINT64_C(0x6843755621BF95DE),
+ UINT64_C(0x4E70EBB447CA77BA), UINT64_C(0x539E61EA65E6D666),
+ UINT64_C(0x0217D6708B21B372), UINT64_C(0x1FF95C2EA90D12AE),
+ UINT64_C(0x39CAC2CCCF78F0CA), UINT64_C(0x24244892ED545116),
+ UINT64_C(0x4E89B2A384BA3F2D), UINT64_C(0x536738FDA6969EF1),
+ UINT64_C(0x7554A61FC0E37C95), UINT64_C(0x68BA2C41E2CFDD49),
+ UINT64_C(0x39339BDB0C08B85D), UINT64_C(0x24DD11852E241981),
+ UINT64_C(0x02EE8F674851FBE5), UINT64_C(0x1F0005396A7D5A39),
+ UINT64_C(0xA1FDE05295DF31CD), UINT64_C(0xBC136A0CB7F39011),
+ UINT64_C(0x9A20F4EED1867275), UINT64_C(0x87CE7EB0F3AAD3A9),
+ UINT64_C(0xD647C92A1D6DB6BD), UINT64_C(0xCBA943743F411761),
+ UINT64_C(0xED9ADD965934F505), UINT64_C(0xF07457C87B1854D9),
+ UINT64_C(0x02B9B86A097E3C68), UINT64_C(0x1F5732342B529DB4),
+ UINT64_C(0x3964ACD64D277FD0), UINT64_C(0x248A26886F0BDE0C),
+ UINT64_C(0x7503911281CCBB18), UINT64_C(0x68ED1B4CA3E01AC4),
+ UINT64_C(0x4EDE85AEC595F8A0), UINT64_C(0x53300FF0E7B9597C),
+ UINT64_C(0xEDCDEA9B181B3288), UINT64_C(0xF02360C53A379354),
+ UINT64_C(0xD610FE275C427130), UINT64_C(0xCBFE74797E6ED0EC),
+ UINT64_C(0x9A77C3E390A9B5F8), UINT64_C(0x879949BDB2851424),
+ UINT64_C(0xA1AAD75FD4F0F640), UINT64_C(0xBC445D01F6DC579C),
+ UINT64_C(0x74F1233D072C2A36), UINT64_C(0x691FA96325008BEA),
+ UINT64_C(0x4F2C37814375698E), UINT64_C(0x52C2BDDF6159C852),
+ UINT64_C(0x034B0A458F9EAD46), UINT64_C(0x1EA5801BADB20C9A),
+ UINT64_C(0x38961EF9CBC7EEFE), UINT64_C(0x257894A7E9EB4F22),
+ UINT64_C(0x9B8571CC164924D6), UINT64_C(0x866BFB923465850A),
+ UINT64_C(0xA05865705210676E), UINT64_C(0xBDB6EF2E703CC6B2),
+ UINT64_C(0xEC3F58B49EFBA3A6), UINT64_C(0xF1D1D2EABCD7027A),
+ UINT64_C(0xD7E24C08DAA2E01E), UINT64_C(0xCA0CC656F88E41C2),
+ UINT64_C(0x38C129F48AE82973), UINT64_C(0x252FA3AAA8C488AF),
+ UINT64_C(0x031C3D48CEB16ACB), UINT64_C(0x1EF2B716EC9DCB17),
+ UINT64_C(0x4F7B008C025AAE03), UINT64_C(0x52958AD220760FDF),
+ UINT64_C(0x74A614304603EDBB), UINT64_C(0x69489E6E642F4C67),
+ UINT64_C(0xD7B57B059B8D2793), UINT64_C(0xCA5BF15BB9A1864F),
+ UINT64_C(0xEC686FB9DFD4642B), UINT64_C(0xF186E5E7FDF8C5F7),
+ UINT64_C(0xA00F527D133FA0E3), UINT64_C(0xBDE1D8233113013F),
+ UINT64_C(0x9BD246C15766E35B), UINT64_C(0x863CCC9F754A4287),
+ UINT64_C(0xEC9136AE1CA42CBC), UINT64_C(0xF17FBCF03E888D60),
+ UINT64_C(0xD74C221258FD6F04), UINT64_C(0xCAA2A84C7AD1CED8),
+ UINT64_C(0x9B2B1FD69416ABCC), UINT64_C(0x86C59588B63A0A10),
+ UINT64_C(0xA0F60B6AD04FE874), UINT64_C(0xBD188134F26349A8),
+ UINT64_C(0x03E5645F0DC1225C), UINT64_C(0x1E0BEE012FED8380),
+ UINT64_C(0x383870E3499861E4), UINT64_C(0x25D6FABD6BB4C038),
+ UINT64_C(0x745F4D278573A52C), UINT64_C(0x69B1C779A75F04F0),
+ UINT64_C(0x4F82599BC12AE694), UINT64_C(0x526CD3C5E3064748),
+ UINT64_C(0xA0A13C6791602FF9), UINT64_C(0xBD4FB639B34C8E25),
+ UINT64_C(0x9B7C28DBD5396C41), UINT64_C(0x8692A285F715CD9D),
+ UINT64_C(0xD71B151F19D2A889), UINT64_C(0xCAF59F413BFE0955),
+ UINT64_C(0xECC601A35D8BEB31), UINT64_C(0xF1288BFD7FA74AED),
+ UINT64_C(0x4FD56E9680052119), UINT64_C(0x523BE4C8A22980C5),
+ UINT64_C(0x74087A2AC45C62A1), UINT64_C(0x69E6F074E670C37D),
+ UINT64_C(0x386F47EE08B7A669), UINT64_C(0x2581CDB02A9B07B5),
+ UINT64_C(0x03B253524CEEE5D1), UINT64_C(0x1E5CD90C6EC2440D)
+ }
+};
diff --git a/src/liblzma/check/crc64_tablegen.c b/src/liblzma/check/crc64_tablegen.c
new file mode 100644
index 0000000..fddaa7e
--- /dev/null
+++ b/src/liblzma/check/crc64_tablegen.c
@@ -0,0 +1,88 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc64_tablegen.c
+/// \brief Generate crc64_table_le.h and crc64_table_be.h
+///
+/// Compiling: gcc -std=c99 -o crc64_tablegen crc64_tablegen.c
+/// Add -DWORDS_BIGENDIAN to generate big endian table.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include <stdio.h>
+#include "../../common/tuklib_integer.h"
+
+
+static uint64_t crc64_table[4][256];
+
+
+extern void
+init_crc64_table(void)
+{
+ static const uint64_t poly64 = UINT64_C(0xC96C5795D7870F42);
+
+ for (size_t s = 0; s < 4; ++s) {
+ for (size_t b = 0; b < 256; ++b) {
+ uint64_t r = s == 0 ? b : crc64_table[s - 1][b];
+
+ for (size_t i = 0; i < 8; ++i) {
+ if (r & 1)
+ r = (r >> 1) ^ poly64;
+ else
+ r >>= 1;
+ }
+
+ crc64_table[s][b] = r;
+ }
+ }
+
+#ifdef WORDS_BIGENDIAN
+ for (size_t s = 0; s < 4; ++s)
+ for (size_t b = 0; b < 256; ++b)
+ crc64_table[s][b] = bswap64(crc64_table[s][b]);
+#endif
+
+ return;
+}
+
+
+static void
+print_crc64_table(void)
+{
+ printf("/* This file has been automatically generated by "
+ "crc64_tablegen.c. */\n\n"
+ "const uint64_t lzma_crc64_table[4][256] = {\n\t{");
+
+ for (size_t s = 0; s < 4; ++s) {
+ for (size_t b = 0; b < 256; ++b) {
+ if ((b % 2) == 0)
+ printf("\n\t\t");
+
+ printf("UINT64_C(0x%016" PRIX64 ")",
+ crc64_table[s][b]);
+
+ if (b != 255)
+ printf(",%s", (b+1) % 2 == 0 ? "" : " ");
+ }
+
+ if (s == 3)
+ printf("\n\t}\n};\n");
+ else
+ printf("\n\t}, {");
+ }
+
+ return;
+}
+
+
+int
+main(void)
+{
+ init_crc64_table();
+ print_crc64_table();
+ return 0;
+}
diff --git a/src/liblzma/check/crc64_x86.S b/src/liblzma/check/crc64_x86.S
new file mode 100644
index 0000000..f5bb84b
--- /dev/null
+++ b/src/liblzma/check/crc64_x86.S
@@ -0,0 +1,287 @@
+/*
+ * Speed-optimized CRC64 using slicing-by-four algorithm
+ *
+ * This uses only i386 instructions, but it is optimized for i686 and later
+ * (including e.g. Pentium II/III/IV, Athlon XP, and Core 2).
+ *
+ * Authors: Igor Pavlov (original CRC32 assembly code)
+ * Lasse Collin (CRC64 adaptation of the modified CRC32 code)
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ *
+ * This code needs lzma_crc64_table, which can be created using the
+ * following C code:
+
+uint64_t lzma_crc64_table[4][256];
+
+void
+init_table(void)
+{
+ // ECMA-182
+ static const uint64_t poly64 = UINT64_C(0xC96C5795D7870F42);
+
+ for (size_t s = 0; s < 4; ++s) {
+ for (size_t b = 0; b < 256; ++b) {
+ uint64_t r = s == 0 ? b : lzma_crc64_table[s - 1][b];
+
+ for (size_t i = 0; i < 8; ++i) {
+ if (r & 1)
+ r = (r >> 1) ^ poly64;
+ else
+ r >>= 1;
+ }
+
+ lzma_crc64_table[s][b] = r;
+ }
+ }
+}
+
+ * The prototype of the CRC64 function:
+ * extern uint64_t lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc);
+ */
+
+/*
+ * On some systems, the functions need to be prefixed. The prefix is
+ * usually an underscore.
+ */
+#ifndef __USER_LABEL_PREFIX__
+# define __USER_LABEL_PREFIX__
+#endif
+#define MAKE_SYM_CAT(prefix, sym) prefix ## sym
+#define MAKE_SYM(prefix, sym) MAKE_SYM_CAT(prefix, sym)
+#define LZMA_CRC64 MAKE_SYM(__USER_LABEL_PREFIX__, lzma_crc64)
+#define LZMA_CRC64_TABLE MAKE_SYM(__USER_LABEL_PREFIX__, lzma_crc64_table)
+
+/*
+ * Solaris assembler doesn't have .p2align, and Darwin uses .align
+ * differently than GNU/Linux and Solaris.
+ */
+#if defined(__APPLE__) || defined(__MSDOS__)
+# define ALIGN(pow2, abs) .align pow2
+#else
+# define ALIGN(pow2, abs) .align abs
+#endif
+
+ .text
+ .globl LZMA_CRC64
+
+#if !defined(__APPLE__) && !defined(_WIN32) && !defined(__CYGWIN__) \
+ && !defined(__MSDOS__)
+ .type LZMA_CRC64, @function
+#endif
+
+ ALIGN(4, 16)
+LZMA_CRC64:
+ /*
+ * Register usage:
+ * %eax crc LSB
+ * %edx crc MSB
+ * %esi buf
+ * %edi size or buf + size
+ * %ebx lzma_crc64_table
+ * %ebp Table index
+ * %ecx Temporary
+ */
+ pushl %ebx
+ pushl %esi
+ pushl %edi
+ pushl %ebp
+ movl 0x14(%esp), %esi /* buf */
+ movl 0x18(%esp), %edi /* size */
+ movl 0x1C(%esp), %eax /* crc LSB */
+ movl 0x20(%esp), %edx /* crc MSB */
+
+ /*
+ * Store the address of lzma_crc64_table to %ebx. This is needed to
+ * get position-independent code (PIC).
+ *
+ * The PIC macro is defined by libtool, while __PIC__ is defined
+ * by GCC but only on some systems. Testing for both makes it simpler
+ * to test this code without libtool, and keeps the code working also
+ * when built with libtool but using something else than GCC.
+ *
+ * I understood that libtool may define PIC on Windows even though
+ * the code in Windows DLLs is not PIC in sense that it is in ELF
+ * binaries, so we need a separate check to always use the non-PIC
+ * code on Windows.
+ */
+#if (!defined(PIC) && !defined(__PIC__)) \
+ || (defined(_WIN32) || defined(__CYGWIN__))
+ /* Not PIC */
+ movl $ LZMA_CRC64_TABLE, %ebx
+#elif defined(__APPLE__)
+ /* Mach-O */
+ call .L_get_pc
+.L_pic:
+ leal .L_lzma_crc64_table$non_lazy_ptr-.L_pic(%ebx), %ebx
+ movl (%ebx), %ebx
+#else
+ /* ELF */
+ call .L_get_pc
+ addl $_GLOBAL_OFFSET_TABLE_, %ebx
+ movl LZMA_CRC64_TABLE@GOT(%ebx), %ebx
+#endif
+
+ /* Complement the initial value. */
+ notl %eax
+ notl %edx
+
+.L_align:
+ /*
+ * Check if there is enough input to use slicing-by-four.
+ * We need eight bytes, because the loop pre-reads four bytes.
+ */
+ cmpl $8, %edi
+ jb .L_rest
+
+ /* Check if we have reached alignment of four bytes. */
+ testl $3, %esi
+ jz .L_slice
+
+ /* Calculate CRC of the next input byte. */
+ movzbl (%esi), %ebp
+ incl %esi
+ movzbl %al, %ecx
+ xorl %ecx, %ebp
+ shrdl $8, %edx, %eax
+ xorl (%ebx, %ebp, 8), %eax
+ shrl $8, %edx
+ xorl 4(%ebx, %ebp, 8), %edx
+ decl %edi
+ jmp .L_align
+
+.L_slice:
+ /*
+ * If we get here, there's at least eight bytes of aligned input
+ * available. Make %edi multiple of four bytes. Store the possible
+ * remainder over the "size" variable in the argument stack.
+ */
+ movl %edi, 0x18(%esp)
+ andl $-4, %edi
+ subl %edi, 0x18(%esp)
+
+ /*
+ * Let %edi be buf + size - 4 while running the main loop. This way
+ * we can compare for equality to determine when exit the loop.
+ */
+ addl %esi, %edi
+ subl $4, %edi
+
+ /* Read in the first four aligned bytes. */
+ movl (%esi), %ecx
+
+.L_loop:
+ xorl %eax, %ecx
+ movzbl %cl, %ebp
+ movl 0x1800(%ebx, %ebp, 8), %eax
+ xorl %edx, %eax
+ movl 0x1804(%ebx, %ebp, 8), %edx
+ movzbl %ch, %ebp
+ xorl 0x1000(%ebx, %ebp, 8), %eax
+ xorl 0x1004(%ebx, %ebp, 8), %edx
+ shrl $16, %ecx
+ movzbl %cl, %ebp
+ xorl 0x0800(%ebx, %ebp, 8), %eax
+ xorl 0x0804(%ebx, %ebp, 8), %edx
+ movzbl %ch, %ebp
+ addl $4, %esi
+ xorl (%ebx, %ebp, 8), %eax
+ xorl 4(%ebx, %ebp, 8), %edx
+
+ /* Check for end of aligned input. */
+ cmpl %edi, %esi
+
+ /*
+ * Copy the next input byte to %ecx. It is slightly faster to
+ * read it here than at the top of the loop.
+ */
+ movl (%esi), %ecx
+ jb .L_loop
+
+ /*
+ * Process the remaining four bytes, which we have already
+ * copied to %ecx.
+ */
+ xorl %eax, %ecx
+ movzbl %cl, %ebp
+ movl 0x1800(%ebx, %ebp, 8), %eax
+ xorl %edx, %eax
+ movl 0x1804(%ebx, %ebp, 8), %edx
+ movzbl %ch, %ebp
+ xorl 0x1000(%ebx, %ebp, 8), %eax
+ xorl 0x1004(%ebx, %ebp, 8), %edx
+ shrl $16, %ecx
+ movzbl %cl, %ebp
+ xorl 0x0800(%ebx, %ebp, 8), %eax
+ xorl 0x0804(%ebx, %ebp, 8), %edx
+ movzbl %ch, %ebp
+ addl $4, %esi
+ xorl (%ebx, %ebp, 8), %eax
+ xorl 4(%ebx, %ebp, 8), %edx
+
+ /* Copy the number of remaining bytes to %edi. */
+ movl 0x18(%esp), %edi
+
+.L_rest:
+ /* Check for end of input. */
+ testl %edi, %edi
+ jz .L_return
+
+ /* Calculate CRC of the next input byte. */
+ movzbl (%esi), %ebp
+ incl %esi
+ movzbl %al, %ecx
+ xorl %ecx, %ebp
+ shrdl $8, %edx, %eax
+ xorl (%ebx, %ebp, 8), %eax
+ shrl $8, %edx
+ xorl 4(%ebx, %ebp, 8), %edx
+ decl %edi
+ jmp .L_rest
+
+.L_return:
+ /* Complement the final value. */
+ notl %eax
+ notl %edx
+
+ popl %ebp
+ popl %edi
+ popl %esi
+ popl %ebx
+ ret
+
+#if defined(PIC) || defined(__PIC__)
+ ALIGN(4, 16)
+.L_get_pc:
+ movl (%esp), %ebx
+ ret
+#endif
+
+#if defined(__APPLE__) && (defined(PIC) || defined(__PIC__))
+ /* Mach-O PIC */
+ .section __IMPORT,__pointers,non_lazy_symbol_pointers
+.L_lzma_crc64_table$non_lazy_ptr:
+ .indirect_symbol LZMA_CRC64_TABLE
+ .long 0
+
+#elif defined(_WIN32) || defined(__CYGWIN__)
+# ifdef DLL_EXPORT
+ /* This is equivalent of __declspec(dllexport). */
+ .section .drectve
+ .ascii " -export:lzma_crc64"
+# endif
+
+#elif !defined(__MSDOS__)
+ /* ELF */
+ .size LZMA_CRC64, .-LZMA_CRC64
+#endif
+
+/*
+ * This is needed to support non-executable stack. It's ugly to
+ * use __linux__ here, but I don't know a way to detect when
+ * we are using GNU assembler.
+ */
+#if defined(__ELF__) && defined(__linux__)
+ .section .note.GNU-stack,"",@progbits
+#endif
diff --git a/src/liblzma/check/crc_macros.h b/src/liblzma/check/crc_macros.h
new file mode 100644
index 0000000..a7c21b7
--- /dev/null
+++ b/src/liblzma/check/crc_macros.h
@@ -0,0 +1,30 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file crc_macros.h
+/// \brief Some endian-dependent macros for CRC32 and CRC64
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifdef WORDS_BIGENDIAN
+# define A(x) ((x) >> 24)
+# define B(x) (((x) >> 16) & 0xFF)
+# define C(x) (((x) >> 8) & 0xFF)
+# define D(x) ((x) & 0xFF)
+
+# define S8(x) ((x) << 8)
+# define S32(x) ((x) << 32)
+
+#else
+# define A(x) ((x) & 0xFF)
+# define B(x) (((x) >> 8) & 0xFF)
+# define C(x) (((x) >> 16) & 0xFF)
+# define D(x) ((x) >> 24)
+
+# define S8(x) ((x) >> 8)
+# define S32(x) ((x) >> 32)
+#endif
diff --git a/src/liblzma/check/sha256.c b/src/liblzma/check/sha256.c
new file mode 100644
index 0000000..04231db
--- /dev/null
+++ b/src/liblzma/check/sha256.c
@@ -0,0 +1,201 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file sha256.c
+/// \brief SHA-256
+///
+/// \todo Crypto++ has x86 ASM optimizations. They use SSE so if they
+/// are imported to liblzma, SSE instructions need to be used
+/// conditionally to keep the code working on older boxes.
+/// We could also support using some external libary for SHA-256.
+//
+// This code is based on the code found from 7-Zip, which has a modified
+// version of the SHA-256 found from Crypto++ <http://www.cryptopp.com/>.
+// The code was modified a little to fit into liblzma.
+//
+// Authors: Kevin Springle
+// Wei Dai
+// Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+// Avoid bogus warnings in transform().
+#if (__GNUC__ == 4 && __GNUC_MINOR__ >= 2) || __GNUC__ > 4
+# pragma GCC diagnostic ignored "-Wuninitialized"
+#endif
+
+#include "check.h"
+
+// At least on x86, GCC is able to optimize this to a rotate instruction.
+#define rotr_32(num, amount) ((num) >> (amount) | (num) << (32 - (amount)))
+
+#define blk0(i) (W[i] = data[i])
+#define blk2(i) (W[i & 15] += s1(W[(i - 2) & 15]) + W[(i - 7) & 15] \
+ + s0(W[(i - 15) & 15]))
+
+#define Ch(x, y, z) (z ^ (x & (y ^ z)))
+#define Maj(x, y, z) ((x & y) | (z & (x | y)))
+
+#define a(i) T[(0 - i) & 7]
+#define b(i) T[(1 - i) & 7]
+#define c(i) T[(2 - i) & 7]
+#define d(i) T[(3 - i) & 7]
+#define e(i) T[(4 - i) & 7]
+#define f(i) T[(5 - i) & 7]
+#define g(i) T[(6 - i) & 7]
+#define h(i) T[(7 - i) & 7]
+
+#define R(i) \
+ h(i) += S1(e(i)) + Ch(e(i), f(i), g(i)) + SHA256_K[i + j] \
+ + (j ? blk2(i) : blk0(i)); \
+ d(i) += h(i); \
+ h(i) += S0(a(i)) + Maj(a(i), b(i), c(i))
+
+#define S0(x) (rotr_32(x, 2) ^ rotr_32(x, 13) ^ rotr_32(x, 22))
+#define S1(x) (rotr_32(x, 6) ^ rotr_32(x, 11) ^ rotr_32(x, 25))
+#define s0(x) (rotr_32(x, 7) ^ rotr_32(x, 18) ^ (x >> 3))
+#define s1(x) (rotr_32(x, 17) ^ rotr_32(x, 19) ^ (x >> 10))
+
+
+static const uint32_t SHA256_K[64] = {
+ 0x428A2F98, 0x71374491, 0xB5C0FBCF, 0xE9B5DBA5,
+ 0x3956C25B, 0x59F111F1, 0x923F82A4, 0xAB1C5ED5,
+ 0xD807AA98, 0x12835B01, 0x243185BE, 0x550C7DC3,
+ 0x72BE5D74, 0x80DEB1FE, 0x9BDC06A7, 0xC19BF174,
+ 0xE49B69C1, 0xEFBE4786, 0x0FC19DC6, 0x240CA1CC,
+ 0x2DE92C6F, 0x4A7484AA, 0x5CB0A9DC, 0x76F988DA,
+ 0x983E5152, 0xA831C66D, 0xB00327C8, 0xBF597FC7,
+ 0xC6E00BF3, 0xD5A79147, 0x06CA6351, 0x14292967,
+ 0x27B70A85, 0x2E1B2138, 0x4D2C6DFC, 0x53380D13,
+ 0x650A7354, 0x766A0ABB, 0x81C2C92E, 0x92722C85,
+ 0xA2BFE8A1, 0xA81A664B, 0xC24B8B70, 0xC76C51A3,
+ 0xD192E819, 0xD6990624, 0xF40E3585, 0x106AA070,
+ 0x19A4C116, 0x1E376C08, 0x2748774C, 0x34B0BCB5,
+ 0x391C0CB3, 0x4ED8AA4A, 0x5B9CCA4F, 0x682E6FF3,
+ 0x748F82EE, 0x78A5636F, 0x84C87814, 0x8CC70208,
+ 0x90BEFFFA, 0xA4506CEB, 0xBEF9A3F7, 0xC67178F2,
+};
+
+
+static void
+transform(uint32_t state[static 8], const uint32_t data[static 16])
+{
+ uint32_t W[16];
+ uint32_t T[8];
+
+ // Copy state[] to working vars.
+ memcpy(T, state, sizeof(T));
+
+ // 64 operations, partially loop unrolled
+ for (unsigned int j = 0; j < 64; j += 16) {
+ R( 0); R( 1); R( 2); R( 3);
+ R( 4); R( 5); R( 6); R( 7);
+ R( 8); R( 9); R(10); R(11);
+ R(12); R(13); R(14); R(15);
+ }
+
+ // Add the working vars back into state[].
+ state[0] += a(0);
+ state[1] += b(0);
+ state[2] += c(0);
+ state[3] += d(0);
+ state[4] += e(0);
+ state[5] += f(0);
+ state[6] += g(0);
+ state[7] += h(0);
+}
+
+
+static void
+process(lzma_check_state *check)
+{
+#ifdef WORDS_BIGENDIAN
+ transform(check->state.sha256.state, check->buffer.u32);
+
+#else
+ uint32_t data[16];
+
+ for (size_t i = 0; i < 16; ++i)
+ data[i] = bswap32(check->buffer.u32[i]);
+
+ transform(check->state.sha256.state, data);
+#endif
+
+ return;
+}
+
+
+extern void
+lzma_sha256_init(lzma_check_state *check)
+{
+ static const uint32_t s[8] = {
+ 0x6A09E667, 0xBB67AE85, 0x3C6EF372, 0xA54FF53A,
+ 0x510E527F, 0x9B05688C, 0x1F83D9AB, 0x5BE0CD19,
+ };
+
+ memcpy(check->state.sha256.state, s, sizeof(s));
+ check->state.sha256.size = 0;
+
+ return;
+}
+
+
+extern void
+lzma_sha256_update(const uint8_t *buf, size_t size, lzma_check_state *check)
+{
+ // Copy the input data into a properly aligned temporary buffer.
+ // This way we can be called with arbitrarily sized buffers
+ // (no need to be multiple of 64 bytes), and the code works also
+ // on architectures that don't allow unaligned memory access.
+ while (size > 0) {
+ const size_t copy_start = check->state.sha256.size & 0x3F;
+ size_t copy_size = 64 - copy_start;
+ if (copy_size > size)
+ copy_size = size;
+
+ memcpy(check->buffer.u8 + copy_start, buf, copy_size);
+
+ buf += copy_size;
+ size -= copy_size;
+ check->state.sha256.size += copy_size;
+
+ if ((check->state.sha256.size & 0x3F) == 0)
+ process(check);
+ }
+
+ return;
+}
+
+
+extern void
+lzma_sha256_finish(lzma_check_state *check)
+{
+ // Add padding as described in RFC 3174 (it describes SHA-1 but
+ // the same padding style is used for SHA-256 too).
+ size_t pos = check->state.sha256.size & 0x3F;
+ check->buffer.u8[pos++] = 0x80;
+
+ while (pos != 64 - 8) {
+ if (pos == 64) {
+ process(check);
+ pos = 0;
+ }
+
+ check->buffer.u8[pos++] = 0x00;
+ }
+
+ // Convert the message size from bytes to bits.
+ check->state.sha256.size *= 8;
+
+ check->buffer.u64[(64 - 8) / 8] = conv64be(check->state.sha256.size);
+
+ process(check);
+
+ for (size_t i = 0; i < 8; ++i)
+ check->buffer.u32[i] = conv32be(check->state.sha256.state[i]);
+
+ return;
+}
diff --git a/src/liblzma/common/Makefile.inc b/src/liblzma/common/Makefile.inc
new file mode 100644
index 0000000..f154f82
--- /dev/null
+++ b/src/liblzma/common/Makefile.inc
@@ -0,0 +1,67 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+liblzma_la_SOURCES += \
+ common/common.c \
+ common/common.h \
+ common/block_util.c \
+ common/easy_preset.c \
+ common/easy_preset.h \
+ common/filter_common.c \
+ common/filter_common.h \
+ common/hardware_physmem.c \
+ common/index.c \
+ common/index.h \
+ common/stream_flags_common.c \
+ common/stream_flags_common.h \
+ common/vli_size.c
+
+if COND_MAIN_ENCODER
+liblzma_la_SOURCES += \
+ common/alone_encoder.c \
+ common/block_buffer_encoder.c \
+ common/block_encoder.c \
+ common/block_encoder.h \
+ common/block_header_encoder.c \
+ common/easy_buffer_encoder.c \
+ common/easy_encoder.c \
+ common/easy_encoder_memusage.c \
+ common/filter_buffer_encoder.c \
+ common/filter_encoder.c \
+ common/filter_encoder.h \
+ common/filter_flags_encoder.c \
+ common/index_encoder.c \
+ common/index_encoder.h \
+ common/stream_buffer_encoder.c \
+ common/stream_encoder.c \
+ common/stream_encoder.h \
+ common/stream_flags_encoder.c \
+ common/vli_encoder.c
+endif
+
+if COND_MAIN_DECODER
+liblzma_la_SOURCES += \
+ common/alone_decoder.c \
+ common/alone_decoder.h \
+ common/auto_decoder.c \
+ common/block_buffer_decoder.c \
+ common/block_decoder.c \
+ common/block_decoder.h \
+ common/block_header_decoder.c \
+ common/easy_decoder_memusage.c \
+ common/filter_buffer_decoder.c \
+ common/filter_decoder.c \
+ common/filter_decoder.h \
+ common/filter_flags_decoder.c \
+ common/index_decoder.c \
+ common/index_hash.c \
+ common/stream_buffer_decoder.c \
+ common/stream_decoder.c \
+ common/stream_decoder.h \
+ common/stream_flags_decoder.c \
+ common/vli_decoder.c
+endif
diff --git a/src/liblzma/common/alone_decoder.c b/src/liblzma/common/alone_decoder.c
new file mode 100644
index 0000000..039b428
--- /dev/null
+++ b/src/liblzma/common/alone_decoder.c
@@ -0,0 +1,232 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file alone_decoder.c
+/// \brief Decoder for LZMA_Alone files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "alone_decoder.h"
+#include "lzma_decoder.h"
+#include "lz_decoder.h"
+
+
+struct lzma_coder_s {
+ lzma_next_coder next;
+
+ enum {
+ SEQ_PROPERTIES,
+ SEQ_DICTIONARY_SIZE,
+ SEQ_UNCOMPRESSED_SIZE,
+ SEQ_CODER_INIT,
+ SEQ_CODE,
+ } sequence;
+
+ /// Position in the header fields
+ size_t pos;
+
+ /// Uncompressed size decoded from the header
+ lzma_vli uncompressed_size;
+
+ /// Memory usage limit
+ uint64_t memlimit;
+
+ /// Amount of memory actually needed (only an estimate)
+ uint64_t memusage;
+
+ /// Options decoded from the header needed to initialize
+ /// the LZMA decoder
+ lzma_options_lzma options;
+};
+
+
+static lzma_ret
+alone_decode(lzma_coder *coder,
+ lzma_allocator *allocator lzma_attribute((unused)),
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size,
+ lzma_action action)
+{
+ while (*out_pos < out_size
+ && (coder->sequence == SEQ_CODE || *in_pos < in_size))
+ switch (coder->sequence) {
+ case SEQ_PROPERTIES:
+ if (lzma_lzma_lclppb_decode(&coder->options, in[*in_pos]))
+ return LZMA_FORMAT_ERROR;
+
+ coder->sequence = SEQ_DICTIONARY_SIZE;
+ ++*in_pos;
+ break;
+
+ case SEQ_DICTIONARY_SIZE:
+ coder->options.dict_size
+ |= (size_t)(in[*in_pos]) << (coder->pos * 8);
+
+ if (++coder->pos == 4) {
+ if (coder->options.dict_size != UINT32_MAX) {
+ // A hack to ditch tons of false positives:
+ // We allow only dictionary sizes that are
+ // 2^n or 2^n + 2^(n-1). LZMA_Alone created
+ // only files with 2^n, but accepts any
+ // dictionary size. If someone complains, this
+ // will be reconsidered.
+ uint32_t d = coder->options.dict_size - 1;
+ d |= d >> 2;
+ d |= d >> 3;
+ d |= d >> 4;
+ d |= d >> 8;
+ d |= d >> 16;
+ ++d;
+
+ if (d != coder->options.dict_size)
+ return LZMA_FORMAT_ERROR;
+ }
+
+ coder->pos = 0;
+ coder->sequence = SEQ_UNCOMPRESSED_SIZE;
+ }
+
+ ++*in_pos;
+ break;
+
+ case SEQ_UNCOMPRESSED_SIZE:
+ coder->uncompressed_size
+ |= (lzma_vli)(in[*in_pos]) << (coder->pos * 8);
+ ++*in_pos;
+ if (++coder->pos < 8)
+ break;
+
+ // Another hack to ditch false positives: Assume that
+ // if the uncompressed size is known, it must be less
+ // than 256 GiB. Again, if someone complains, this
+ // will be reconsidered.
+ if (coder->uncompressed_size != LZMA_VLI_UNKNOWN
+ && coder->uncompressed_size
+ >= (LZMA_VLI_C(1) << 38))
+ return LZMA_FORMAT_ERROR;
+
+ // Calculate the memory usage so that it is ready
+ // for SEQ_CODER_INIT.
+ coder->memusage = lzma_lzma_decoder_memusage(&coder->options)
+ + LZMA_MEMUSAGE_BASE;
+
+ coder->pos = 0;
+ coder->sequence = SEQ_CODER_INIT;
+
+ // Fall through
+
+ case SEQ_CODER_INIT: {
+ if (coder->memusage > coder->memlimit)
+ return LZMA_MEMLIMIT_ERROR;
+
+ lzma_filter_info filters[2] = {
+ {
+ .init = &lzma_lzma_decoder_init,
+ .options = &coder->options,
+ }, {
+ .init = NULL,
+ }
+ };
+
+ const lzma_ret ret = lzma_next_filter_init(&coder->next,
+ allocator, filters);
+ if (ret != LZMA_OK)
+ return ret;
+
+ // Use a hack to set the uncompressed size.
+ lzma_lz_decoder_uncompressed(coder->next.coder,
+ coder->uncompressed_size);
+
+ coder->sequence = SEQ_CODE;
+ break;
+ }
+
+ case SEQ_CODE: {
+ return coder->next.code(coder->next.coder,
+ allocator, in, in_pos, in_size,
+ out, out_pos, out_size, action);
+ }
+
+ default:
+ return LZMA_PROG_ERROR;
+ }
+
+ return LZMA_OK;
+}
+
+
+static void
+alone_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+alone_decoder_memconfig(lzma_coder *coder, uint64_t *memusage,
+ uint64_t *old_memlimit, uint64_t new_memlimit)
+{
+ *memusage = coder->memusage;
+ *old_memlimit = coder->memlimit;
+
+ if (new_memlimit != 0) {
+ if (new_memlimit < coder->memusage)
+ return LZMA_MEMLIMIT_ERROR;
+
+ coder->memlimit = new_memlimit;
+ }
+
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_alone_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ uint64_t memlimit)
+{
+ lzma_next_coder_init(&lzma_alone_decoder_init, next, allocator);
+
+ if (memlimit == 0)
+ return LZMA_PROG_ERROR;
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &alone_decode;
+ next->end = &alone_decoder_end;
+ next->memconfig = &alone_decoder_memconfig;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ next->coder->sequence = SEQ_PROPERTIES;
+ next->coder->pos = 0;
+ next->coder->options.dict_size = 0;
+ next->coder->options.preset_dict = NULL;
+ next->coder->options.preset_dict_size = 0;
+ next->coder->uncompressed_size = 0;
+ next->coder->memlimit = memlimit;
+ next->coder->memusage = LZMA_MEMUSAGE_BASE;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_alone_decoder(lzma_stream *strm, uint64_t memlimit)
+{
+ lzma_next_strm_init(lzma_alone_decoder_init, strm, memlimit);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/alone_decoder.h b/src/liblzma/common/alone_decoder.h
new file mode 100644
index 0000000..70d0d2a
--- /dev/null
+++ b/src/liblzma/common/alone_decoder.h
@@ -0,0 +1,22 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file alone_decoder.h
+/// \brief Decoder for LZMA_Alone files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_ALONE_DECODER_H
+#define LZMA_ALONE_DECODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_alone_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, uint64_t memlimit);
+
+#endif
diff --git a/src/liblzma/common/alone_encoder.c b/src/liblzma/common/alone_encoder.c
new file mode 100644
index 0000000..d8c0170
--- /dev/null
+++ b/src/liblzma/common/alone_encoder.c
@@ -0,0 +1,157 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file alone_decoder.c
+/// \brief Decoder for LZMA_Alone files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+#include "lzma_encoder.h"
+
+
+#define ALONE_HEADER_SIZE (1 + 4 + 8)
+
+
+struct lzma_coder_s {
+ lzma_next_coder next;
+
+ enum {
+ SEQ_HEADER,
+ SEQ_CODE,
+ } sequence;
+
+ size_t header_pos;
+ uint8_t header[ALONE_HEADER_SIZE];
+};
+
+
+static lzma_ret
+alone_encode(lzma_coder *coder,
+ lzma_allocator *allocator lzma_attribute((unused)),
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size,
+ lzma_action action)
+{
+ while (*out_pos < out_size)
+ switch (coder->sequence) {
+ case SEQ_HEADER:
+ lzma_bufcpy(coder->header, &coder->header_pos,
+ ALONE_HEADER_SIZE,
+ out, out_pos, out_size);
+ if (coder->header_pos < ALONE_HEADER_SIZE)
+ return LZMA_OK;
+
+ coder->sequence = SEQ_CODE;
+ break;
+
+ case SEQ_CODE:
+ return coder->next.code(coder->next.coder,
+ allocator, in, in_pos, in_size,
+ out, out_pos, out_size, action);
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+ return LZMA_OK;
+}
+
+
+static void
+alone_encoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+// At least for now, this is not used by any internal function.
+static lzma_ret
+alone_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_options_lzma *options)
+{
+ lzma_next_coder_init(&alone_encoder_init, next, allocator);
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &alone_encode;
+ next->end = &alone_encoder_end;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ // Basic initializations
+ next->coder->sequence = SEQ_HEADER;
+ next->coder->header_pos = 0;
+
+ // Encode the header:
+ // - Properties (1 byte)
+ if (lzma_lzma_lclppb_encode(options, next->coder->header))
+ return LZMA_OPTIONS_ERROR;
+
+ // - Dictionary size (4 bytes)
+ if (options->dict_size < LZMA_DICT_SIZE_MIN)
+ return LZMA_OPTIONS_ERROR;
+
+ // Round up to to the next 2^n or 2^n + 2^(n - 1) depending on which
+ // one is the next unless it is UINT32_MAX. While the header would
+ // allow any 32-bit integer, we do this to keep the decoder of liblzma
+ // accepting the resulting files.
+ uint32_t d = options->dict_size - 1;
+ d |= d >> 2;
+ d |= d >> 3;
+ d |= d >> 4;
+ d |= d >> 8;
+ d |= d >> 16;
+ if (d != UINT32_MAX)
+ ++d;
+
+ unaligned_write32le(next->coder->header + 1, d);
+
+ // - Uncompressed size (always unknown and using EOPM)
+ memset(next->coder->header + 1 + 4, 0xFF, 8);
+
+ // Initialize the LZMA encoder.
+ const lzma_filter_info filters[2] = {
+ {
+ .init = &lzma_lzma_encoder_init,
+ .options = (void *)(options),
+ }, {
+ .init = NULL,
+ }
+ };
+
+ return lzma_next_filter_init(&next->coder->next, allocator, filters);
+}
+
+
+/*
+extern lzma_ret
+lzma_alone_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_options_alone *options)
+{
+ lzma_next_coder_init(&alone_encoder_init, next, allocator, options);
+}
+*/
+
+
+extern LZMA_API(lzma_ret)
+lzma_alone_encoder(lzma_stream *strm, const lzma_options_lzma *options)
+{
+ lzma_next_strm_init(alone_encoder_init, strm, options);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/auto_decoder.c b/src/liblzma/common/auto_decoder.c
new file mode 100644
index 0000000..ae6c3e7
--- /dev/null
+++ b/src/liblzma/common/auto_decoder.c
@@ -0,0 +1,186 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file auto_decoder.c
+/// \brief Autodetect between .xz Stream and .lzma (LZMA_Alone) formats
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_decoder.h"
+#include "alone_decoder.h"
+
+
+struct lzma_coder_s {
+ /// Stream decoder or LZMA_Alone decoder
+ lzma_next_coder next;
+
+ uint64_t memlimit;
+ uint32_t flags;
+
+ enum {
+ SEQ_INIT,
+ SEQ_CODE,
+ SEQ_FINISH,
+ } sequence;
+};
+
+
+static lzma_ret
+auto_decode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ switch (coder->sequence) {
+ case SEQ_INIT:
+ if (*in_pos >= in_size)
+ return LZMA_OK;
+
+ // Update the sequence now, because we want to continue from
+ // SEQ_CODE even if we return some LZMA_*_CHECK.
+ coder->sequence = SEQ_CODE;
+
+ // Detect the file format. For now this is simple, since if
+ // it doesn't start with 0xFD (the first magic byte of the
+ // new format), it has to be LZMA_Alone, or something that
+ // we don't support at all.
+ if (in[*in_pos] == 0xFD) {
+ return_if_error(lzma_stream_decoder_init(
+ &coder->next, allocator,
+ coder->memlimit, coder->flags));
+ } else {
+ return_if_error(lzma_alone_decoder_init(&coder->next,
+ allocator, coder->memlimit));
+
+ // If the application wants to know about missing
+ // integrity check or about the check in general, we
+ // need to handle it here, because LZMA_Alone decoder
+ // doesn't accept any flags.
+ if (coder->flags & LZMA_TELL_NO_CHECK)
+ return LZMA_NO_CHECK;
+
+ if (coder->flags & LZMA_TELL_ANY_CHECK)
+ return LZMA_GET_CHECK;
+ }
+
+ // Fall through
+
+ case SEQ_CODE: {
+ const lzma_ret ret = coder->next.code(
+ coder->next.coder, allocator,
+ in, in_pos, in_size,
+ out, out_pos, out_size, action);
+ if (ret != LZMA_STREAM_END
+ || (coder->flags & LZMA_CONCATENATED) == 0)
+ return ret;
+
+ coder->sequence = SEQ_FINISH;
+ }
+
+ // Fall through
+
+ case SEQ_FINISH:
+ // When LZMA_DECODE_CONCATENATED was used and we were decoding
+ // LZMA_Alone file, we need to check check that there is no
+ // trailing garbage and wait for LZMA_FINISH.
+ if (*in_pos < in_size)
+ return LZMA_DATA_ERROR;
+
+ return action == LZMA_FINISH ? LZMA_STREAM_END : LZMA_OK;
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+}
+
+
+static void
+auto_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_check
+auto_decoder_get_check(const lzma_coder *coder)
+{
+ // It is LZMA_Alone if get_check is NULL.
+ return coder->next.get_check == NULL ? LZMA_CHECK_NONE
+ : coder->next.get_check(coder->next.coder);
+}
+
+
+static lzma_ret
+auto_decoder_memconfig(lzma_coder *coder, uint64_t *memusage,
+ uint64_t *old_memlimit, uint64_t new_memlimit)
+{
+ lzma_ret ret;
+
+ if (coder->next.memconfig != NULL) {
+ ret = coder->next.memconfig(coder->next.coder,
+ memusage, old_memlimit, new_memlimit);
+ assert(*old_memlimit == coder->memlimit);
+ } else {
+ // No coder is configured yet. Use the base value as
+ // the current memory usage.
+ *memusage = LZMA_MEMUSAGE_BASE;
+ *old_memlimit = coder->memlimit;
+ ret = LZMA_OK;
+ }
+
+ if (ret == LZMA_OK && new_memlimit != 0)
+ coder->memlimit = new_memlimit;
+
+ return ret;
+}
+
+
+static lzma_ret
+auto_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ uint64_t memlimit, uint32_t flags)
+{
+ lzma_next_coder_init(&auto_decoder_init, next, allocator);
+
+ if (memlimit == 0)
+ return LZMA_PROG_ERROR;
+
+ if (flags & ~LZMA_SUPPORTED_FLAGS)
+ return LZMA_OPTIONS_ERROR;
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &auto_decode;
+ next->end = &auto_decoder_end;
+ next->get_check = &auto_decoder_get_check;
+ next->memconfig = &auto_decoder_memconfig;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ next->coder->memlimit = memlimit;
+ next->coder->flags = flags;
+ next->coder->sequence = SEQ_INIT;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_auto_decoder(lzma_stream *strm, uint64_t memlimit, uint32_t flags)
+{
+ lzma_next_strm_init(auto_decoder_init, strm, memlimit, flags);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/block_buffer_decoder.c b/src/liblzma/common/block_buffer_decoder.c
new file mode 100644
index 0000000..ff27a11
--- /dev/null
+++ b/src/liblzma/common/block_buffer_decoder.c
@@ -0,0 +1,80 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_buffer_decoder.c
+/// \brief Single-call .xz Block decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "block_decoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_buffer_decode(lzma_block *block, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ if (in_pos == NULL || (in == NULL && *in_pos != in_size)
+ || *in_pos > in_size || out_pos == NULL
+ || (out == NULL && *out_pos != out_size)
+ || *out_pos > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Initialize the Block decoder.
+ lzma_next_coder block_decoder = LZMA_NEXT_CODER_INIT;
+ lzma_ret ret = lzma_block_decoder_init(
+ &block_decoder, allocator, block);
+
+ if (ret == LZMA_OK) {
+ // Save the positions so that we can restore them in case
+ // an error occurs.
+ const size_t in_start = *in_pos;
+ const size_t out_start = *out_pos;
+
+ // Do the actual decoding.
+ ret = block_decoder.code(block_decoder.coder, allocator,
+ in, in_pos, in_size, out, out_pos, out_size,
+ LZMA_FINISH);
+
+ if (ret == LZMA_STREAM_END) {
+ ret = LZMA_OK;
+ } else {
+ if (ret == LZMA_OK) {
+ // Either the input was truncated or the
+ // output buffer was too small.
+ assert(*in_pos == in_size
+ || *out_pos == out_size);
+
+ // If all the input was consumed, then the
+ // input is truncated, even if the output
+ // buffer is also full. This is because
+ // processing the last byte of the Block
+ // never produces output.
+ //
+ // NOTE: This assumption may break when new
+ // filters are added, if the end marker of
+ // the filter doesn't consume at least one
+ // complete byte.
+ if (*in_pos == in_size)
+ ret = LZMA_DATA_ERROR;
+ else
+ ret = LZMA_BUF_ERROR;
+ }
+
+ // Restore the positions.
+ *in_pos = in_start;
+ *out_pos = out_start;
+ }
+ }
+
+ // Free the decoder memory. This needs to be done even if
+ // initialization fails, because the internal API doesn't
+ // require the initialization function to free its memory on error.
+ lzma_next_end(&block_decoder, allocator);
+
+ return ret;
+}
diff --git a/src/liblzma/common/block_buffer_encoder.c b/src/liblzma/common/block_buffer_encoder.c
new file mode 100644
index 0000000..a8f71c2
--- /dev/null
+++ b/src/liblzma/common/block_buffer_encoder.c
@@ -0,0 +1,299 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_buffer_encoder.c
+/// \brief Single-call .xz Block encoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "block_encoder.h"
+#include "filter_encoder.h"
+#include "lzma2_encoder.h"
+#include "check.h"
+
+
+/// Estimate the maximum size of the Block Header and Check fields for
+/// a Block that uses LZMA2 uncompressed chunks. We could use
+/// lzma_block_header_size() but this is simpler.
+///
+/// Block Header Size + Block Flags + Compressed Size
+/// + Uncompressed Size + Filter Flags for LZMA2 + CRC32 + Check
+/// and round up to the next multiple of four to take Header Padding
+/// into account.
+#define HEADERS_BOUND ((1 + 1 + 2 * LZMA_VLI_BYTES_MAX + 3 + 4 \
+ + LZMA_CHECK_SIZE_MAX + 3) & ~3)
+
+
+static lzma_vli
+lzma2_bound(lzma_vli uncompressed_size)
+{
+ // Prevent integer overflow in overhead calculation.
+ if (uncompressed_size > COMPRESSED_SIZE_MAX)
+ return 0;
+
+ // Calculate the exact overhead of the LZMA2 headers: Round
+ // uncompressed_size up to the next multiple of LZMA2_CHUNK_MAX,
+ // multiply by the size of per-chunk header, and add one byte for
+ // the end marker.
+ const lzma_vli overhead = ((uncompressed_size + LZMA2_CHUNK_MAX - 1)
+ / LZMA2_CHUNK_MAX)
+ * LZMA2_HEADER_UNCOMPRESSED + 1;
+
+ // Catch the possible integer overflow.
+ if (COMPRESSED_SIZE_MAX - overhead < uncompressed_size)
+ return 0;
+
+ return uncompressed_size + overhead;
+}
+
+
+extern LZMA_API(size_t)
+lzma_block_buffer_bound(size_t uncompressed_size)
+{
+ // For now, if the data doesn't compress, we always use uncompressed
+ // chunks of LZMA2. In future we may use Subblock filter too, but
+ // but for simplicity we probably will still use the same bound
+ // calculation even though Subblock filter would have slightly less
+ // overhead.
+ lzma_vli lzma2_size = lzma2_bound(uncompressed_size);
+ if (lzma2_size == 0)
+ return 0;
+
+ // Take Block Padding into account.
+ lzma2_size = (lzma2_size + 3) & ~LZMA_VLI_C(3);
+
+#if SIZE_MAX < LZMA_VLI_MAX
+ // Catch the possible integer overflow on 32-bit systems. There's no
+ // overflow on 64-bit systems, because lzma2_bound() already takes
+ // into account the size of the headers in the Block.
+ if (SIZE_MAX - HEADERS_BOUND < lzma2_size)
+ return 0;
+#endif
+
+ return HEADERS_BOUND + lzma2_size;
+}
+
+
+static lzma_ret
+block_encode_uncompressed(lzma_block *block, const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // TODO: Figure out if the last filter is LZMA2 or Subblock and use
+ // that filter to encode the uncompressed chunks.
+
+ // Use LZMA2 uncompressed chunks. We wouldn't need a dictionary at
+ // all, but LZMA2 always requires a dictionary, so use the minimum
+ // value to minimize memory usage of the decoder.
+ lzma_options_lzma lzma2 = {
+ .dict_size = LZMA_DICT_SIZE_MIN,
+ };
+
+ lzma_filter filters[2];
+ filters[0].id = LZMA_FILTER_LZMA2;
+ filters[0].options = &lzma2;
+ filters[1].id = LZMA_VLI_UNKNOWN;
+
+ // Set the above filter options to *block temporarily so that we can
+ // encode the Block Header.
+ lzma_filter *filters_orig = block->filters;
+ block->filters = filters;
+
+ if (lzma_block_header_size(block) != LZMA_OK) {
+ block->filters = filters_orig;
+ return LZMA_PROG_ERROR;
+ }
+
+ // Check that there's enough output space. The caller has already
+ // set block->compressed_size to what lzma2_bound() has returned,
+ // so we can reuse that value. We know that compressed_size is a
+ // known valid VLI and header_size is a small value so their sum
+ // will never overflow.
+ assert(block->compressed_size == lzma2_bound(in_size));
+ if (out_size - *out_pos
+ < block->header_size + block->compressed_size) {
+ block->filters = filters_orig;
+ return LZMA_BUF_ERROR;
+ }
+
+ if (lzma_block_header_encode(block, out + *out_pos) != LZMA_OK) {
+ block->filters = filters_orig;
+ return LZMA_PROG_ERROR;
+ }
+
+ block->filters = filters_orig;
+ *out_pos += block->header_size;
+
+ // Encode the data using LZMA2 uncompressed chunks.
+ size_t in_pos = 0;
+ uint8_t control = 0x01; // Dictionary reset
+
+ while (in_pos < in_size) {
+ // Control byte: Indicate uncompressed chunk, of which
+ // the first resets the dictionary.
+ out[(*out_pos)++] = control;
+ control = 0x02; // No dictionary reset
+
+ // Size of the uncompressed chunk
+ const size_t copy_size
+ = my_min(in_size - in_pos, LZMA2_CHUNK_MAX);
+ out[(*out_pos)++] = (copy_size - 1) >> 8;
+ out[(*out_pos)++] = (copy_size - 1) & 0xFF;
+
+ // The actual data
+ assert(*out_pos + copy_size <= out_size);
+ memcpy(out + *out_pos, in + in_pos, copy_size);
+
+ in_pos += copy_size;
+ *out_pos += copy_size;
+ }
+
+ // End marker
+ out[(*out_pos)++] = 0x00;
+ assert(*out_pos <= out_size);
+
+ return LZMA_OK;
+}
+
+
+static lzma_ret
+block_encode_normal(lzma_block *block, lzma_allocator *allocator,
+ const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // Find out the size of the Block Header.
+ block->compressed_size = lzma2_bound(in_size);
+ if (block->compressed_size == 0)
+ return LZMA_DATA_ERROR;
+
+ block->uncompressed_size = in_size;
+ return_if_error(lzma_block_header_size(block));
+
+ // Reserve space for the Block Header and skip it for now.
+ if (out_size - *out_pos <= block->header_size)
+ return LZMA_BUF_ERROR;
+
+ const size_t out_start = *out_pos;
+ *out_pos += block->header_size;
+
+ // Limit out_size so that we stop encoding if the output would grow
+ // bigger than what uncompressed Block would be.
+ if (out_size - *out_pos > block->compressed_size)
+ out_size = *out_pos + block->compressed_size;
+
+ // TODO: In many common cases this could be optimized to use
+ // significantly less memory.
+ lzma_next_coder raw_encoder = LZMA_NEXT_CODER_INIT;
+ lzma_ret ret = lzma_raw_encoder_init(
+ &raw_encoder, allocator, block->filters);
+
+ if (ret == LZMA_OK) {
+ size_t in_pos = 0;
+ ret = raw_encoder.code(raw_encoder.coder, allocator,
+ in, &in_pos, in_size, out, out_pos, out_size,
+ LZMA_FINISH);
+ }
+
+ // NOTE: This needs to be run even if lzma_raw_encoder_init() failed.
+ lzma_next_end(&raw_encoder, allocator);
+
+ if (ret == LZMA_STREAM_END) {
+ // Compression was successful. Write the Block Header.
+ block->compressed_size
+ = *out_pos - (out_start + block->header_size);
+ ret = lzma_block_header_encode(block, out + out_start);
+ if (ret != LZMA_OK)
+ ret = LZMA_PROG_ERROR;
+
+ } else if (ret == LZMA_OK) {
+ // Output buffer became full.
+ ret = LZMA_BUF_ERROR;
+ }
+
+ // Reset *out_pos if something went wrong.
+ if (ret != LZMA_OK)
+ *out_pos = out_start;
+
+ return ret;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_buffer_encode(lzma_block *block, lzma_allocator *allocator,
+ const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // Sanity checks
+ if (block == NULL || block->filters == NULL
+ || (in == NULL && in_size != 0) || out == NULL
+ || out_pos == NULL || *out_pos > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Check the version field.
+ if (block->version != 0)
+ return LZMA_OPTIONS_ERROR;
+
+ // Size of a Block has to be a multiple of four, so limit the size
+ // here already. This way we don't need to check it again when adding
+ // Block Padding.
+ out_size -= (out_size - *out_pos) & 3;
+
+ // Get the size of the Check field.
+ const size_t check_size = lzma_check_size(block->check);
+ if (check_size == UINT32_MAX)
+ return LZMA_PROG_ERROR;
+
+ // Reserve space for the Check field.
+ if (out_size - *out_pos <= check_size)
+ return LZMA_BUF_ERROR;
+
+ out_size -= check_size;
+
+ // Do the actual compression.
+ const lzma_ret ret = block_encode_normal(block, allocator,
+ in, in_size, out, out_pos, out_size);
+ if (ret != LZMA_OK) {
+ // If the error was something else than output buffer
+ // becoming full, return the error now.
+ if (ret != LZMA_BUF_ERROR)
+ return ret;
+
+ // The data was uncompressible (at least with the options
+ // given to us) or the output buffer was too small. Use the
+ // uncompressed chunks of LZMA2 to wrap the data into a valid
+ // Block. If we haven't been given enough output space, even
+ // this may fail.
+ return_if_error(block_encode_uncompressed(block, in, in_size,
+ out, out_pos, out_size));
+ }
+
+ assert(*out_pos <= out_size);
+
+ // Block Padding. No buffer overflow here, because we already adjusted
+ // out_size so that (out_size - out_start) is a multiple of four.
+ // Thus, if the buffer is full, the loop body can never run.
+ for (size_t i = (size_t)(block->compressed_size); i & 3; ++i) {
+ assert(*out_pos < out_size);
+ out[(*out_pos)++] = 0x00;
+ }
+
+ // If there's no Check field, we are done now.
+ if (check_size > 0) {
+ // Calculate the integrity check. We reserved space for
+ // the Check field earlier so we don't need to check for
+ // available output space here.
+ lzma_check_state check;
+ lzma_check_init(&check, block->check);
+ lzma_check_update(&check, block->check, in, in_size);
+ lzma_check_finish(&check, block->check);
+
+ memcpy(block->raw_check, check.buffer.u8, check_size);
+ memcpy(out + *out_pos, check.buffer.u8, check_size);
+ *out_pos += check_size;
+ }
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/block_decoder.c b/src/liblzma/common/block_decoder.c
new file mode 100644
index 0000000..a3ce6f4
--- /dev/null
+++ b/src/liblzma/common/block_decoder.c
@@ -0,0 +1,242 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_decoder.c
+/// \brief Decodes .xz Blocks
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "block_decoder.h"
+#include "filter_decoder.h"
+#include "check.h"
+
+
+struct lzma_coder_s {
+ enum {
+ SEQ_CODE,
+ SEQ_PADDING,
+ SEQ_CHECK,
+ } sequence;
+
+ /// The filters in the chain; initialized with lzma_raw_decoder_init().
+ lzma_next_coder next;
+
+ /// Decoding options; we also write Compressed Size and Uncompressed
+ /// Size back to this structure when the decoding has been finished.
+ lzma_block *block;
+
+ /// Compressed Size calculated while decoding
+ lzma_vli compressed_size;
+
+ /// Uncompressed Size calculated while decoding
+ lzma_vli uncompressed_size;
+
+ /// Maximum allowed Compressed Size; this takes into account the
+ /// size of the Block Header and Check fields when Compressed Size
+ /// is unknown.
+ lzma_vli compressed_limit;
+
+ /// Position when reading the Check field
+ size_t check_pos;
+
+ /// Check of the uncompressed data
+ lzma_check_state check;
+};
+
+
+static inline bool
+update_size(lzma_vli *size, lzma_vli add, lzma_vli limit)
+{
+ if (limit > LZMA_VLI_MAX)
+ limit = LZMA_VLI_MAX;
+
+ if (limit < *size || limit - *size < add)
+ return true;
+
+ *size += add;
+
+ return false;
+}
+
+
+static inline bool
+is_size_valid(lzma_vli size, lzma_vli reference)
+{
+ return reference == LZMA_VLI_UNKNOWN || reference == size;
+}
+
+
+static lzma_ret
+block_decode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ switch (coder->sequence) {
+ case SEQ_CODE: {
+ const size_t in_start = *in_pos;
+ const size_t out_start = *out_pos;
+
+ const lzma_ret ret = coder->next.code(coder->next.coder,
+ allocator, in, in_pos, in_size,
+ out, out_pos, out_size, action);
+
+ const size_t in_used = *in_pos - in_start;
+ const size_t out_used = *out_pos - out_start;
+
+ // NOTE: We compare to compressed_limit here, which prevents
+ // the total size of the Block growing past LZMA_VLI_MAX.
+ if (update_size(&coder->compressed_size, in_used,
+ coder->compressed_limit)
+ || update_size(&coder->uncompressed_size,
+ out_used,
+ coder->block->uncompressed_size))
+ return LZMA_DATA_ERROR;
+
+ lzma_check_update(&coder->check, coder->block->check,
+ out + out_start, out_used);
+
+ if (ret != LZMA_STREAM_END)
+ return ret;
+
+ // Compressed and Uncompressed Sizes are now at their final
+ // values. Verify that they match the values given to us.
+ if (!is_size_valid(coder->compressed_size,
+ coder->block->compressed_size)
+ || !is_size_valid(coder->uncompressed_size,
+ coder->block->uncompressed_size))
+ return LZMA_DATA_ERROR;
+
+ // Copy the values into coder->block. The caller
+ // may use this information to construct Index.
+ coder->block->compressed_size = coder->compressed_size;
+ coder->block->uncompressed_size = coder->uncompressed_size;
+
+ coder->sequence = SEQ_PADDING;
+ }
+
+ // Fall through
+
+ case SEQ_PADDING:
+ // Compressed Data is padded to a multiple of four bytes.
+ while (coder->compressed_size & 3) {
+ if (*in_pos >= in_size)
+ return LZMA_OK;
+
+ // We use compressed_size here just get the Padding
+ // right. The actual Compressed Size was stored to
+ // coder->block already, and won't be modified by
+ // us anymore.
+ ++coder->compressed_size;
+
+ if (in[(*in_pos)++] != 0x00)
+ return LZMA_DATA_ERROR;
+ }
+
+ if (coder->block->check == LZMA_CHECK_NONE)
+ return LZMA_STREAM_END;
+
+ lzma_check_finish(&coder->check, coder->block->check);
+ coder->sequence = SEQ_CHECK;
+
+ // Fall through
+
+ case SEQ_CHECK: {
+ const size_t check_size = lzma_check_size(coder->block->check);
+ lzma_bufcpy(in, in_pos, in_size, coder->block->raw_check,
+ &coder->check_pos, check_size);
+ if (coder->check_pos < check_size)
+ return LZMA_OK;
+
+ // Validate the Check only if we support it.
+ // coder->check.buffer may be uninitialized
+ // when the Check ID is not supported.
+ if (lzma_check_is_supported(coder->block->check)
+ && memcmp(coder->block->raw_check,
+ coder->check.buffer.u8,
+ check_size) != 0)
+ return LZMA_DATA_ERROR;
+
+ return LZMA_STREAM_END;
+ }
+ }
+
+ return LZMA_PROG_ERROR;
+}
+
+
+static void
+block_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+extern lzma_ret
+lzma_block_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ lzma_block *block)
+{
+ lzma_next_coder_init(&lzma_block_decoder_init, next, allocator);
+
+ // Validate the options. lzma_block_unpadded_size() does that for us
+ // except for Uncompressed Size and filters. Filters are validated
+ // by the raw decoder.
+ if (lzma_block_unpadded_size(block) == 0
+ || !lzma_vli_is_valid(block->uncompressed_size))
+ return LZMA_PROG_ERROR;
+
+ // Allocate and initialize *next->coder if needed.
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &block_decode;
+ next->end = &block_decoder_end;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ // Basic initializations
+ next->coder->sequence = SEQ_CODE;
+ next->coder->block = block;
+ next->coder->compressed_size = 0;
+ next->coder->uncompressed_size = 0;
+
+ // If Compressed Size is not known, we calculate the maximum allowed
+ // value so that encoded size of the Block (including Block Padding)
+ // is still a valid VLI and a multiple of four.
+ next->coder->compressed_limit
+ = block->compressed_size == LZMA_VLI_UNKNOWN
+ ? (LZMA_VLI_MAX & ~LZMA_VLI_C(3))
+ - block->header_size
+ - lzma_check_size(block->check)
+ : block->compressed_size;
+
+ // Initialize the check. It's caller's problem if the Check ID is not
+ // supported, and the Block decoder cannot verify the Check field.
+ // Caller can test lzma_check_is_supported(block->check).
+ next->coder->check_pos = 0;
+ lzma_check_init(&next->coder->check, block->check);
+
+ // Initialize the filter chain.
+ return lzma_raw_decoder_init(&next->coder->next, allocator,
+ block->filters);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_decoder(lzma_stream *strm, lzma_block *block)
+{
+ lzma_next_strm_init(lzma_block_decoder_init, strm, block);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/block_decoder.h b/src/liblzma/common/block_decoder.h
new file mode 100644
index 0000000..7da9df6
--- /dev/null
+++ b/src/liblzma/common/block_decoder.h
@@ -0,0 +1,22 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_decoder.h
+/// \brief Decodes .xz Blocks
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_BLOCK_DECODER_H
+#define LZMA_BLOCK_DECODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_block_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, lzma_block *block);
+
+#endif
diff --git a/src/liblzma/common/block_encoder.c b/src/liblzma/common/block_encoder.c
new file mode 100644
index 0000000..ca51523
--- /dev/null
+++ b/src/liblzma/common/block_encoder.c
@@ -0,0 +1,212 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_encoder.c
+/// \brief Encodes .xz Blocks
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "block_encoder.h"
+#include "filter_encoder.h"
+#include "check.h"
+
+
+struct lzma_coder_s {
+ /// The filters in the chain; initialized with lzma_raw_decoder_init().
+ lzma_next_coder next;
+
+ /// Encoding options; we also write Unpadded Size, Compressed Size,
+ /// and Uncompressed Size back to this structure when the encoding
+ /// has been finished.
+ lzma_block *block;
+
+ enum {
+ SEQ_CODE,
+ SEQ_PADDING,
+ SEQ_CHECK,
+ } sequence;
+
+ /// Compressed Size calculated while encoding
+ lzma_vli compressed_size;
+
+ /// Uncompressed Size calculated while encoding
+ lzma_vli uncompressed_size;
+
+ /// Position in the Check field
+ size_t pos;
+
+ /// Check of the uncompressed data
+ lzma_check_state check;
+};
+
+
+static lzma_ret
+block_encode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ // Check that our amount of input stays in proper limits.
+ if (LZMA_VLI_MAX - coder->uncompressed_size < in_size - *in_pos)
+ return LZMA_DATA_ERROR;
+
+ switch (coder->sequence) {
+ case SEQ_CODE: {
+ const size_t in_start = *in_pos;
+ const size_t out_start = *out_pos;
+
+ const lzma_ret ret = coder->next.code(coder->next.coder,
+ allocator, in, in_pos, in_size,
+ out, out_pos, out_size, action);
+
+ const size_t in_used = *in_pos - in_start;
+ const size_t out_used = *out_pos - out_start;
+
+ if (COMPRESSED_SIZE_MAX - coder->compressed_size < out_used)
+ return LZMA_DATA_ERROR;
+
+ coder->compressed_size += out_used;
+
+ // No need to check for overflow because we have already
+ // checked it at the beginning of this function.
+ coder->uncompressed_size += in_used;
+
+ lzma_check_update(&coder->check, coder->block->check,
+ in + in_start, in_used);
+
+ if (ret != LZMA_STREAM_END || action == LZMA_SYNC_FLUSH)
+ return ret;
+
+ assert(*in_pos == in_size);
+ assert(action == LZMA_FINISH);
+
+ // Copy the values into coder->block. The caller
+ // may use this information to construct Index.
+ coder->block->compressed_size = coder->compressed_size;
+ coder->block->uncompressed_size = coder->uncompressed_size;
+
+ coder->sequence = SEQ_PADDING;
+ }
+
+ // Fall through
+
+ case SEQ_PADDING:
+ // Pad Compressed Data to a multiple of four bytes. We can
+ // use coder->compressed_size for this since we don't need
+ // it for anything else anymore.
+ while (coder->compressed_size & 3) {
+ if (*out_pos >= out_size)
+ return LZMA_OK;
+
+ out[*out_pos] = 0x00;
+ ++*out_pos;
+ ++coder->compressed_size;
+ }
+
+ if (coder->block->check == LZMA_CHECK_NONE)
+ return LZMA_STREAM_END;
+
+ lzma_check_finish(&coder->check, coder->block->check);
+
+ coder->sequence = SEQ_CHECK;
+
+ // Fall through
+
+ case SEQ_CHECK: {
+ const size_t check_size = lzma_check_size(coder->block->check);
+ lzma_bufcpy(coder->check.buffer.u8, &coder->pos, check_size,
+ out, out_pos, out_size);
+ if (coder->pos < check_size)
+ return LZMA_OK;
+
+ memcpy(coder->block->raw_check, coder->check.buffer.u8,
+ check_size);
+ return LZMA_STREAM_END;
+ }
+ }
+
+ return LZMA_PROG_ERROR;
+}
+
+
+static void
+block_encoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+block_encoder_update(lzma_coder *coder, lzma_allocator *allocator,
+ const lzma_filter *filters lzma_attribute((unused)),
+ const lzma_filter *reversed_filters)
+{
+ if (coder->sequence != SEQ_CODE)
+ return LZMA_PROG_ERROR;
+
+ return lzma_next_filter_update(
+ &coder->next, allocator, reversed_filters);
+}
+
+
+extern lzma_ret
+lzma_block_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ lzma_block *block)
+{
+ lzma_next_coder_init(&lzma_block_encoder_init, next, allocator);
+
+ if (block->version != 0)
+ return LZMA_OPTIONS_ERROR;
+
+ // If the Check ID is not supported, we cannot calculate the check and
+ // thus not create a proper Block.
+ if ((unsigned int)(block->check) > LZMA_CHECK_ID_MAX)
+ return LZMA_PROG_ERROR;
+
+ if (!lzma_check_is_supported(block->check))
+ return LZMA_UNSUPPORTED_CHECK;
+
+ // Allocate and initialize *next->coder if needed.
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &block_encode;
+ next->end = &block_encoder_end;
+ next->update = &block_encoder_update;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ // Basic initializations
+ next->coder->sequence = SEQ_CODE;
+ next->coder->block = block;
+ next->coder->compressed_size = 0;
+ next->coder->uncompressed_size = 0;
+ next->coder->pos = 0;
+
+ // Initialize the check
+ lzma_check_init(&next->coder->check, block->check);
+
+ // Initialize the requested filters.
+ return lzma_raw_encoder_init(&next->coder->next, allocator,
+ block->filters);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_encoder(lzma_stream *strm, lzma_block *block)
+{
+ lzma_next_strm_init(lzma_block_encoder_init, strm, block);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/block_encoder.h b/src/liblzma/common/block_encoder.h
new file mode 100644
index 0000000..b9eff0b
--- /dev/null
+++ b/src/liblzma/common/block_encoder.h
@@ -0,0 +1,47 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_encoder.h
+/// \brief Encodes .xz Blocks
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_BLOCK_ENCODER_H
+#define LZMA_BLOCK_ENCODER_H
+
+#include "common.h"
+
+
+/// \brief Biggest Compressed Size value that the Block encoder supports
+///
+/// The maximum size of a single Block is limited by the maximum size of
+/// a Stream, which in theory is 2^63 - 3 bytes (i.e. LZMA_VLI_MAX - 3).
+/// While the size is really big and no one should hit it in practice, we
+/// take it into account in some places anyway to catch some errors e.g. if
+/// application passes insanely big value to some function.
+///
+/// We could take into account the headers etc. to determine the exact
+/// maximum size of the Compressed Data field, but the complexity would give
+/// us nothing useful. Instead, limit the size of Compressed Data so that
+/// even with biggest possible Block Header and Check fields the total
+/// encoded size of the Block stays as a valid VLI. This doesn't guarantee
+/// that the size of the Stream doesn't grow too big, but that problem is
+/// taken care outside the Block handling code.
+///
+/// ~LZMA_VLI_C(3) is to guarantee that if we need padding at the end of
+/// the Compressed Data field, it will still stay in the proper limit.
+///
+/// This constant is in this file because it is needed in both
+/// block_encoder.c and block_buffer_encoder.c.
+#define COMPRESSED_SIZE_MAX ((LZMA_VLI_MAX - LZMA_BLOCK_HEADER_SIZE_MAX \
+ - LZMA_CHECK_SIZE_MAX) & ~LZMA_VLI_C(3))
+
+
+extern lzma_ret lzma_block_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, lzma_block *block);
+
+#endif
diff --git a/src/liblzma/common/block_header_decoder.c b/src/liblzma/common/block_header_decoder.c
new file mode 100644
index 0000000..2c9573e
--- /dev/null
+++ b/src/liblzma/common/block_header_decoder.c
@@ -0,0 +1,116 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_header_decoder.c
+/// \brief Decodes Block Header from .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+#include "check.h"
+
+
+static void
+free_properties(lzma_block *block, lzma_allocator *allocator)
+{
+ // Free allocated filter options. The last array member is not
+ // touched after the initialization in the beginning of
+ // lzma_block_header_decode(), so we don't need to touch that here.
+ for (size_t i = 0; i < LZMA_FILTERS_MAX; ++i) {
+ lzma_free(block->filters[i].options, allocator);
+ block->filters[i].id = LZMA_VLI_UNKNOWN;
+ block->filters[i].options = NULL;
+ }
+
+ return;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_header_decode(lzma_block *block,
+ lzma_allocator *allocator, const uint8_t *in)
+{
+ // NOTE: We consider the header to be corrupt not only when the
+ // CRC32 doesn't match, but also when variable-length integers
+ // are invalid or over 63 bits, or if the header is too small
+ // to contain the claimed information.
+
+ // Initialize the filter options array. This way the caller can
+ // safely free() the options even if an error occurs in this function.
+ for (size_t i = 0; i <= LZMA_FILTERS_MAX; ++i) {
+ block->filters[i].id = LZMA_VLI_UNKNOWN;
+ block->filters[i].options = NULL;
+ }
+
+ // Always zero for now.
+ block->version = 0;
+
+ // Validate Block Header Size and Check type. The caller must have
+ // already set these, so it is a programming error if this test fails.
+ if (lzma_block_header_size_decode(in[0]) != block->header_size
+ || (unsigned int)(block->check) > LZMA_CHECK_ID_MAX)
+ return LZMA_PROG_ERROR;
+
+ // Exclude the CRC32 field.
+ const size_t in_size = block->header_size - 4;
+
+ // Verify CRC32
+ if (lzma_crc32(in, in_size, 0) != unaligned_read32le(in + in_size))
+ return LZMA_DATA_ERROR;
+
+ // Check for unsupported flags.
+ if (in[1] & 0x3C)
+ return LZMA_OPTIONS_ERROR;
+
+ // Start after the Block Header Size and Block Flags fields.
+ size_t in_pos = 2;
+
+ // Compressed Size
+ if (in[1] & 0x40) {
+ return_if_error(lzma_vli_decode(&block->compressed_size,
+ NULL, in, &in_pos, in_size));
+
+ // Validate Compressed Size. This checks that it isn't zero
+ // and that the total size of the Block is a valid VLI.
+ if (lzma_block_unpadded_size(block) == 0)
+ return LZMA_DATA_ERROR;
+ } else {
+ block->compressed_size = LZMA_VLI_UNKNOWN;
+ }
+
+ // Uncompressed Size
+ if (in[1] & 0x80)
+ return_if_error(lzma_vli_decode(&block->uncompressed_size,
+ NULL, in, &in_pos, in_size));
+ else
+ block->uncompressed_size = LZMA_VLI_UNKNOWN;
+
+ // Filter Flags
+ const size_t filter_count = (in[1] & 3) + 1;
+ for (size_t i = 0; i < filter_count; ++i) {
+ const lzma_ret ret = lzma_filter_flags_decode(
+ &block->filters[i], allocator,
+ in, &in_pos, in_size);
+ if (ret != LZMA_OK) {
+ free_properties(block, allocator);
+ return ret;
+ }
+ }
+
+ // Padding
+ while (in_pos < in_size) {
+ if (in[in_pos++] != 0x00) {
+ free_properties(block, allocator);
+
+ // Possibly some new field present so use
+ // LZMA_OPTIONS_ERROR instead of LZMA_DATA_ERROR.
+ return LZMA_OPTIONS_ERROR;
+ }
+ }
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/block_header_encoder.c b/src/liblzma/common/block_header_encoder.c
new file mode 100644
index 0000000..707dd0c
--- /dev/null
+++ b/src/liblzma/common/block_header_encoder.c
@@ -0,0 +1,132 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_header_encoder.c
+/// \brief Encodes Block Header for .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+#include "check.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_header_size(lzma_block *block)
+{
+ if (block->version != 0)
+ return LZMA_OPTIONS_ERROR;
+
+ // Block Header Size + Block Flags + CRC32.
+ uint32_t size = 1 + 1 + 4;
+
+ // Compressed Size
+ if (block->compressed_size != LZMA_VLI_UNKNOWN) {
+ const uint32_t add = lzma_vli_size(block->compressed_size);
+ if (add == 0 || block->compressed_size == 0)
+ return LZMA_PROG_ERROR;
+
+ size += add;
+ }
+
+ // Uncompressed Size
+ if (block->uncompressed_size != LZMA_VLI_UNKNOWN) {
+ const uint32_t add = lzma_vli_size(block->uncompressed_size);
+ if (add == 0)
+ return LZMA_PROG_ERROR;
+
+ size += add;
+ }
+
+ // List of Filter Flags
+ if (block->filters == NULL || block->filters[0].id == LZMA_VLI_UNKNOWN)
+ return LZMA_PROG_ERROR;
+
+ for (size_t i = 0; block->filters[i].id != LZMA_VLI_UNKNOWN; ++i) {
+ // Don't allow too many filters.
+ if (i == LZMA_FILTERS_MAX)
+ return LZMA_PROG_ERROR;
+
+ uint32_t add;
+ return_if_error(lzma_filter_flags_size(&add,
+ block->filters + i));
+
+ size += add;
+ }
+
+ // Pad to a multiple of four bytes.
+ block->header_size = (size + 3) & ~UINT32_C(3);
+
+ // NOTE: We don't verify that the encoded size of the Block stays
+ // within limits. This is because it is possible that we are called
+ // with exaggerated Compressed Size (e.g. LZMA_VLI_MAX) to reserve
+ // space for Block Header, and later called again with lower,
+ // real values.
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_header_encode(const lzma_block *block, uint8_t *out)
+{
+ // Validate everything but filters.
+ if (lzma_block_unpadded_size(block) == 0
+ || !lzma_vli_is_valid(block->uncompressed_size))
+ return LZMA_PROG_ERROR;
+
+ // Indicate the size of the buffer _excluding_ the CRC32 field.
+ const size_t out_size = block->header_size - 4;
+
+ // Store the Block Header Size.
+ out[0] = out_size / 4;
+
+ // We write Block Flags in pieces.
+ out[1] = 0x00;
+ size_t out_pos = 2;
+
+ // Compressed Size
+ if (block->compressed_size != LZMA_VLI_UNKNOWN) {
+ return_if_error(lzma_vli_encode(block->compressed_size, NULL,
+ out, &out_pos, out_size));
+
+ out[1] |= 0x40;
+ }
+
+ // Uncompressed Size
+ if (block->uncompressed_size != LZMA_VLI_UNKNOWN) {
+ return_if_error(lzma_vli_encode(block->uncompressed_size, NULL,
+ out, &out_pos, out_size));
+
+ out[1] |= 0x80;
+ }
+
+ // Filter Flags
+ if (block->filters == NULL || block->filters[0].id == LZMA_VLI_UNKNOWN)
+ return LZMA_PROG_ERROR;
+
+ size_t filter_count = 0;
+ do {
+ // There can be a maximum of four filters.
+ if (filter_count == LZMA_FILTERS_MAX)
+ return LZMA_PROG_ERROR;
+
+ return_if_error(lzma_filter_flags_encode(
+ block->filters + filter_count,
+ out, &out_pos, out_size));
+
+ } while (block->filters[++filter_count].id != LZMA_VLI_UNKNOWN);
+
+ out[1] |= filter_count - 1;
+
+ // Padding
+ memzero(out + out_pos, out_size - out_pos);
+
+ // CRC32
+ unaligned_write32le(out + out_size, lzma_crc32(out, out_size, 0));
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/block_util.c b/src/liblzma/common/block_util.c
new file mode 100644
index 0000000..62c9345
--- /dev/null
+++ b/src/liblzma/common/block_util.c
@@ -0,0 +1,90 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file block_header.c
+/// \brief Utility functions to handle lzma_block
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+#include "index.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_block_compressed_size(lzma_block *block, lzma_vli unpadded_size)
+{
+ // Validate everything but Uncompressed Size and filters.
+ if (lzma_block_unpadded_size(block) == 0)
+ return LZMA_PROG_ERROR;
+
+ const uint32_t container_size = block->header_size
+ + lzma_check_size(block->check);
+
+ // Validate that Compressed Size will be greater than zero.
+ if (unpadded_size <= container_size)
+ return LZMA_DATA_ERROR;
+
+ // Calculate what Compressed Size is supposed to be.
+ // If Compressed Size was present in Block Header,
+ // compare that the new value matches it.
+ const lzma_vli compressed_size = unpadded_size - container_size;
+ if (block->compressed_size != LZMA_VLI_UNKNOWN
+ && block->compressed_size != compressed_size)
+ return LZMA_DATA_ERROR;
+
+ block->compressed_size = compressed_size;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_block_unpadded_size(const lzma_block *block)
+{
+ // Validate the values that we are interested in i.e. all but
+ // Uncompressed Size and the filters.
+ //
+ // NOTE: This function is used for validation too, so it is
+ // essential that these checks are always done even if
+ // Compressed Size is unknown.
+ if (block == NULL || block->version != 0
+ || block->header_size < LZMA_BLOCK_HEADER_SIZE_MIN
+ || block->header_size > LZMA_BLOCK_HEADER_SIZE_MAX
+ || (block->header_size & 3)
+ || !lzma_vli_is_valid(block->compressed_size)
+ || block->compressed_size == 0
+ || (unsigned int)(block->check) > LZMA_CHECK_ID_MAX)
+ return 0;
+
+ // If Compressed Size is unknown, return that we cannot know
+ // size of the Block either.
+ if (block->compressed_size == LZMA_VLI_UNKNOWN)
+ return LZMA_VLI_UNKNOWN;
+
+ // Calculate Unpadded Size and validate it.
+ const lzma_vli unpadded_size = block->compressed_size
+ + block->header_size
+ + lzma_check_size(block->check);
+
+ assert(unpadded_size >= UNPADDED_SIZE_MIN);
+ if (unpadded_size > UNPADDED_SIZE_MAX)
+ return 0;
+
+ return unpadded_size;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_block_total_size(const lzma_block *block)
+{
+ lzma_vli unpadded_size = lzma_block_unpadded_size(block);
+
+ if (unpadded_size != LZMA_VLI_UNKNOWN)
+ unpadded_size = vli_ceil4(unpadded_size);
+
+ return unpadded_size;
+}
diff --git a/src/liblzma/common/common.c b/src/liblzma/common/common.c
new file mode 100644
index 0000000..07b1d47
--- /dev/null
+++ b/src/liblzma/common/common.c
@@ -0,0 +1,374 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file common.h
+/// \brief Common functions needed in many places in liblzma
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+
+/////////////
+// Version //
+/////////////
+
+extern LZMA_API(uint32_t)
+lzma_version_number(void)
+{
+ return LZMA_VERSION;
+}
+
+
+extern LZMA_API(const char *)
+lzma_version_string(void)
+{
+ return LZMA_VERSION_STRING;
+}
+
+
+///////////////////////
+// Memory allocation //
+///////////////////////
+
+extern void * lzma_attribute((malloc))
+lzma_alloc(size_t size, lzma_allocator *allocator)
+{
+ // Some malloc() variants return NULL if called with size == 0.
+ if (size == 0)
+ size = 1;
+
+ void *ptr;
+
+ if (allocator != NULL && allocator->alloc != NULL)
+ ptr = allocator->alloc(allocator->opaque, 1, size);
+ else
+ ptr = malloc(size);
+
+ return ptr;
+}
+
+
+extern void
+lzma_free(void *ptr, lzma_allocator *allocator)
+{
+ if (allocator != NULL && allocator->free != NULL)
+ allocator->free(allocator->opaque, ptr);
+ else
+ free(ptr);
+
+ return;
+}
+
+
+//////////
+// Misc //
+//////////
+
+extern size_t
+lzma_bufcpy(const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size)
+{
+ const size_t in_avail = in_size - *in_pos;
+ const size_t out_avail = out_size - *out_pos;
+ const size_t copy_size = my_min(in_avail, out_avail);
+
+ memcpy(out + *out_pos, in + *in_pos, copy_size);
+
+ *in_pos += copy_size;
+ *out_pos += copy_size;
+
+ return copy_size;
+}
+
+
+extern lzma_ret
+lzma_next_filter_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ lzma_next_coder_init(filters[0].init, next, allocator);
+ next->id = filters[0].id;
+ return filters[0].init == NULL
+ ? LZMA_OK : filters[0].init(next, allocator, filters);
+}
+
+
+extern lzma_ret
+lzma_next_filter_update(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *reversed_filters)
+{
+ // Check that the application isn't trying to change the Filter ID.
+ // End of filters is indicated with LZMA_VLI_UNKNOWN in both
+ // reversed_filters[0].id and next->id.
+ if (reversed_filters[0].id != next->id)
+ return LZMA_PROG_ERROR;
+
+ if (reversed_filters[0].id == LZMA_VLI_UNKNOWN)
+ return LZMA_OK;
+
+ assert(next->update != NULL);
+ return next->update(next->coder, allocator, NULL, reversed_filters);
+}
+
+
+extern void
+lzma_next_end(lzma_next_coder *next, lzma_allocator *allocator)
+{
+ if (next->init != (uintptr_t)(NULL)) {
+ // To avoid tiny end functions that simply call
+ // lzma_free(coder, allocator), we allow leaving next->end
+ // NULL and call lzma_free() here.
+ if (next->end != NULL)
+ next->end(next->coder, allocator);
+ else
+ lzma_free(next->coder, allocator);
+
+ // Reset the variables so the we don't accidentally think
+ // that it is an already initialized coder.
+ *next = LZMA_NEXT_CODER_INIT;
+ }
+
+ return;
+}
+
+
+//////////////////////////////////////
+// External to internal API wrapper //
+//////////////////////////////////////
+
+extern lzma_ret
+lzma_strm_init(lzma_stream *strm)
+{
+ if (strm == NULL)
+ return LZMA_PROG_ERROR;
+
+ if (strm->internal == NULL) {
+ strm->internal = lzma_alloc(sizeof(lzma_internal),
+ strm->allocator);
+ if (strm->internal == NULL)
+ return LZMA_MEM_ERROR;
+
+ strm->internal->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ strm->internal->supported_actions[LZMA_RUN] = false;
+ strm->internal->supported_actions[LZMA_SYNC_FLUSH] = false;
+ strm->internal->supported_actions[LZMA_FULL_FLUSH] = false;
+ strm->internal->supported_actions[LZMA_FINISH] = false;
+ strm->internal->sequence = ISEQ_RUN;
+ strm->internal->allow_buf_error = false;
+
+ strm->total_in = 0;
+ strm->total_out = 0;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_code(lzma_stream *strm, lzma_action action)
+{
+ // Sanity checks
+ if ((strm->next_in == NULL && strm->avail_in != 0)
+ || (strm->next_out == NULL && strm->avail_out != 0)
+ || strm->internal == NULL
+ || strm->internal->next.code == NULL
+ || (unsigned int)(action) > LZMA_FINISH
+ || !strm->internal->supported_actions[action])
+ return LZMA_PROG_ERROR;
+
+ switch (strm->internal->sequence) {
+ case ISEQ_RUN:
+ switch (action) {
+ case LZMA_RUN:
+ break;
+
+ case LZMA_SYNC_FLUSH:
+ strm->internal->sequence = ISEQ_SYNC_FLUSH;
+ break;
+
+ case LZMA_FULL_FLUSH:
+ strm->internal->sequence = ISEQ_FULL_FLUSH;
+ break;
+
+ case LZMA_FINISH:
+ strm->internal->sequence = ISEQ_FINISH;
+ break;
+ }
+
+ break;
+
+ case ISEQ_SYNC_FLUSH:
+ // The same action must be used until we return
+ // LZMA_STREAM_END, and the amount of input must not change.
+ if (action != LZMA_SYNC_FLUSH
+ || strm->internal->avail_in != strm->avail_in)
+ return LZMA_PROG_ERROR;
+
+ break;
+
+ case ISEQ_FULL_FLUSH:
+ if (action != LZMA_FULL_FLUSH
+ || strm->internal->avail_in != strm->avail_in)
+ return LZMA_PROG_ERROR;
+
+ break;
+
+ case ISEQ_FINISH:
+ if (action != LZMA_FINISH
+ || strm->internal->avail_in != strm->avail_in)
+ return LZMA_PROG_ERROR;
+
+ break;
+
+ case ISEQ_END:
+ return LZMA_STREAM_END;
+
+ case ISEQ_ERROR:
+ default:
+ return LZMA_PROG_ERROR;
+ }
+
+ size_t in_pos = 0;
+ size_t out_pos = 0;
+ lzma_ret ret = strm->internal->next.code(
+ strm->internal->next.coder, strm->allocator,
+ strm->next_in, &in_pos, strm->avail_in,
+ strm->next_out, &out_pos, strm->avail_out, action);
+
+ strm->next_in += in_pos;
+ strm->avail_in -= in_pos;
+ strm->total_in += in_pos;
+
+ strm->next_out += out_pos;
+ strm->avail_out -= out_pos;
+ strm->total_out += out_pos;
+
+ strm->internal->avail_in = strm->avail_in;
+
+ switch (ret) {
+ case LZMA_OK:
+ // Don't return LZMA_BUF_ERROR when it happens the first time.
+ // This is to avoid returning LZMA_BUF_ERROR when avail_out
+ // was zero but still there was no more data left to written
+ // to next_out.
+ if (out_pos == 0 && in_pos == 0) {
+ if (strm->internal->allow_buf_error)
+ ret = LZMA_BUF_ERROR;
+ else
+ strm->internal->allow_buf_error = true;
+ } else {
+ strm->internal->allow_buf_error = false;
+ }
+ break;
+
+ case LZMA_STREAM_END:
+ if (strm->internal->sequence == ISEQ_SYNC_FLUSH
+ || strm->internal->sequence == ISEQ_FULL_FLUSH)
+ strm->internal->sequence = ISEQ_RUN;
+ else
+ strm->internal->sequence = ISEQ_END;
+
+ // Fall through
+
+ case LZMA_NO_CHECK:
+ case LZMA_UNSUPPORTED_CHECK:
+ case LZMA_GET_CHECK:
+ case LZMA_MEMLIMIT_ERROR:
+ // Something else than LZMA_OK, but not a fatal error,
+ // that is, coding may be continued (except if ISEQ_END).
+ strm->internal->allow_buf_error = false;
+ break;
+
+ default:
+ // All the other errors are fatal; coding cannot be continued.
+ assert(ret != LZMA_BUF_ERROR);
+ strm->internal->sequence = ISEQ_ERROR;
+ break;
+ }
+
+ return ret;
+}
+
+
+extern LZMA_API(void)
+lzma_end(lzma_stream *strm)
+{
+ if (strm != NULL && strm->internal != NULL) {
+ lzma_next_end(&strm->internal->next, strm->allocator);
+ lzma_free(strm->internal, strm->allocator);
+ strm->internal = NULL;
+ }
+
+ return;
+}
+
+
+extern LZMA_API(lzma_check)
+lzma_get_check(const lzma_stream *strm)
+{
+ // Return LZMA_CHECK_NONE if we cannot know the check type.
+ // It's a bug in the application if this happens.
+ if (strm->internal->next.get_check == NULL)
+ return LZMA_CHECK_NONE;
+
+ return strm->internal->next.get_check(strm->internal->next.coder);
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_memusage(const lzma_stream *strm)
+{
+ uint64_t memusage;
+ uint64_t old_memlimit;
+
+ if (strm == NULL || strm->internal == NULL
+ || strm->internal->next.memconfig == NULL
+ || strm->internal->next.memconfig(
+ strm->internal->next.coder,
+ &memusage, &old_memlimit, 0) != LZMA_OK)
+ return 0;
+
+ return memusage;
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_memlimit_get(const lzma_stream *strm)
+{
+ uint64_t old_memlimit;
+ uint64_t memusage;
+
+ if (strm == NULL || strm->internal == NULL
+ || strm->internal->next.memconfig == NULL
+ || strm->internal->next.memconfig(
+ strm->internal->next.coder,
+ &memusage, &old_memlimit, 0) != LZMA_OK)
+ return 0;
+
+ return old_memlimit;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_memlimit_set(lzma_stream *strm, uint64_t new_memlimit)
+{
+ // Dummy variables to simplify memconfig functions
+ uint64_t old_memlimit;
+ uint64_t memusage;
+
+ if (strm == NULL || strm->internal == NULL
+ || strm->internal->next.memconfig == NULL)
+ return LZMA_PROG_ERROR;
+
+ if (new_memlimit != 0 && new_memlimit < LZMA_MEMUSAGE_BASE)
+ return LZMA_MEMLIMIT_ERROR;
+
+ return strm->internal->next.memconfig(strm->internal->next.coder,
+ &memusage, &old_memlimit, new_memlimit);
+}
diff --git a/src/liblzma/common/common.h b/src/liblzma/common/common.h
new file mode 100644
index 0000000..3a85168
--- /dev/null
+++ b/src/liblzma/common/common.h
@@ -0,0 +1,284 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file common.h
+/// \brief Definitions common to the whole liblzma library
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_COMMON_H
+#define LZMA_COMMON_H
+
+#include "sysdefs.h"
+#include "mythread.h"
+#include "tuklib_integer.h"
+
+#if defined(_WIN32) || defined(__CYGWIN__)
+# ifdef DLL_EXPORT
+# define LZMA_API_EXPORT __declspec(dllexport)
+# else
+# define LZMA_API_EXPORT
+# endif
+// Don't use ifdef or defined() below.
+#elif HAVE_VISIBILITY
+# define LZMA_API_EXPORT __attribute__((__visibility__("default")))
+#else
+# define LZMA_API_EXPORT
+#endif
+
+#define LZMA_API(type) LZMA_API_EXPORT type LZMA_API_CALL
+
+#include "lzma.h"
+
+// These allow helping the compiler in some often-executed branches, whose
+// result is almost always the same.
+#ifdef __GNUC__
+# define likely(expr) __builtin_expect(expr, true)
+# define unlikely(expr) __builtin_expect(expr, false)
+#else
+# define likely(expr) (expr)
+# define unlikely(expr) (expr)
+#endif
+
+
+/// Size of temporary buffers needed in some filters
+#define LZMA_BUFFER_SIZE 4096
+
+
+/// Starting value for memory usage estimates. Instead of calculating size
+/// of _every_ structure and taking into account malloc() overhead etc., we
+/// add a base size to all memory usage estimates. It's not very accurate
+/// but should be easily good enough.
+#define LZMA_MEMUSAGE_BASE (UINT64_C(1) << 15)
+
+/// Start of internal Filter ID space. These IDs must never be used
+/// in Streams.
+#define LZMA_FILTER_RESERVED_START (LZMA_VLI_C(1) << 62)
+
+
+/// Supported flags that can be passed to lzma_stream_decoder()
+/// or lzma_auto_decoder().
+#define LZMA_SUPPORTED_FLAGS \
+ ( LZMA_TELL_NO_CHECK \
+ | LZMA_TELL_UNSUPPORTED_CHECK \
+ | LZMA_TELL_ANY_CHECK \
+ | LZMA_CONCATENATED )
+
+
+/// Type of encoder/decoder specific data; the actual structure is defined
+/// differently in different coders.
+typedef struct lzma_coder_s lzma_coder;
+
+typedef struct lzma_next_coder_s lzma_next_coder;
+
+typedef struct lzma_filter_info_s lzma_filter_info;
+
+
+/// Type of a function used to initialize a filter encoder or decoder
+typedef lzma_ret (*lzma_init_function)(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters);
+
+/// Type of a function to do some kind of coding work (filters, Stream,
+/// Block encoders/decoders etc.). Some special coders use don't use both
+/// input and output buffers, but for simplicity they still use this same
+/// function prototype.
+typedef lzma_ret (*lzma_code_function)(
+ lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size,
+ lzma_action action);
+
+/// Type of a function to free the memory allocated for the coder
+typedef void (*lzma_end_function)(
+ lzma_coder *coder, lzma_allocator *allocator);
+
+
+/// Raw coder validates and converts an array of lzma_filter structures to
+/// an array of lzma_filter_info structures. This array is used with
+/// lzma_next_filter_init to initialize the filter chain.
+struct lzma_filter_info_s {
+ /// Filter ID. This is used only by the encoder
+ /// with lzma_filters_update().
+ lzma_vli id;
+
+ /// Pointer to function used to initialize the filter.
+ /// This is NULL to indicate end of array.
+ lzma_init_function init;
+
+ /// Pointer to filter's options structure
+ void *options;
+};
+
+
+/// Hold data and function pointers of the next filter in the chain.
+struct lzma_next_coder_s {
+ /// Pointer to coder-specific data
+ lzma_coder *coder;
+
+ /// Filter ID. This is LZMA_VLI_UNKNOWN when this structure doesn't
+ /// point to a filter coder.
+ lzma_vli id;
+
+ /// "Pointer" to init function. This is never called here.
+ /// We need only to detect if we are initializing a coder
+ /// that was allocated earlier. See lzma_next_coder_init and
+ /// lzma_next_strm_init macros in this file.
+ uintptr_t init;
+
+ /// Pointer to function to do the actual coding
+ lzma_code_function code;
+
+ /// Pointer to function to free lzma_next_coder.coder. This can
+ /// be NULL; in that case, lzma_free is called to free
+ /// lzma_next_coder.coder.
+ lzma_end_function end;
+
+ /// Pointer to function to return the type of the integrity check.
+ /// Most coders won't support this.
+ lzma_check (*get_check)(const lzma_coder *coder);
+
+ /// Pointer to function to get and/or change the memory usage limit.
+ /// If new_memlimit == 0, the limit is not changed.
+ lzma_ret (*memconfig)(lzma_coder *coder, uint64_t *memusage,
+ uint64_t *old_memlimit, uint64_t new_memlimit);
+
+ /// Update the filter-specific options or the whole filter chain
+ /// in the encoder.
+ lzma_ret (*update)(lzma_coder *coder, lzma_allocator *allocator,
+ const lzma_filter *filters,
+ const lzma_filter *reversed_filters);
+};
+
+
+/// Macro to initialize lzma_next_coder structure
+#define LZMA_NEXT_CODER_INIT \
+ (lzma_next_coder){ \
+ .coder = NULL, \
+ .init = (uintptr_t)(NULL), \
+ .id = LZMA_VLI_UNKNOWN, \
+ .code = NULL, \
+ .end = NULL, \
+ .get_check = NULL, \
+ .memconfig = NULL, \
+ .update = NULL, \
+ }
+
+
+/// Internal data for lzma_strm_init, lzma_code, and lzma_end. A pointer to
+/// this is stored in lzma_stream.
+struct lzma_internal_s {
+ /// The actual coder that should do something useful
+ lzma_next_coder next;
+
+ /// Track the state of the coder. This is used to validate arguments
+ /// so that the actual coders can rely on e.g. that LZMA_SYNC_FLUSH
+ /// is used on every call to lzma_code until next.code has returned
+ /// LZMA_STREAM_END.
+ enum {
+ ISEQ_RUN,
+ ISEQ_SYNC_FLUSH,
+ ISEQ_FULL_FLUSH,
+ ISEQ_FINISH,
+ ISEQ_END,
+ ISEQ_ERROR,
+ } sequence;
+
+ /// A copy of lzma_stream avail_in. This is used to verify that the
+ /// amount of input doesn't change once e.g. LZMA_FINISH has been
+ /// used.
+ size_t avail_in;
+
+ /// Indicates which lzma_action values are allowed by next.code.
+ bool supported_actions[4];
+
+ /// If true, lzma_code will return LZMA_BUF_ERROR if no progress was
+ /// made (no input consumed and no output produced by next.code).
+ bool allow_buf_error;
+};
+
+
+/// Allocates memory
+extern void *lzma_alloc(size_t size, lzma_allocator *allocator)
+ lzma_attribute((malloc));
+
+/// Frees memory
+extern void lzma_free(void *ptr, lzma_allocator *allocator);
+
+
+/// Allocates strm->internal if it is NULL, and initializes *strm and
+/// strm->internal. This function is only called via lzma_next_strm_init macro.
+extern lzma_ret lzma_strm_init(lzma_stream *strm);
+
+/// Initializes the next filter in the chain, if any. This takes care of
+/// freeing the memory of previously initialized filter if it is different
+/// than the filter being initialized now. This way the actual filter
+/// initialization functions don't need to use lzma_next_coder_init macro.
+extern lzma_ret lzma_next_filter_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+/// Update the next filter in the chain, if any. This checks that
+/// the application is not trying to change the Filter IDs.
+extern lzma_ret lzma_next_filter_update(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *reversed_filters);
+
+/// Frees the memory allocated for next->coder either using next->end or,
+/// if next->end is NULL, using lzma_free.
+extern void lzma_next_end(lzma_next_coder *next, lzma_allocator *allocator);
+
+
+/// Copy as much data as possible from in[] to out[] and update *in_pos
+/// and *out_pos accordingly. Returns the number of bytes copied.
+extern size_t lzma_bufcpy(const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size);
+
+
+/// \brief Return if expression doesn't evaluate to LZMA_OK
+///
+/// There are several situations where we want to return immediately
+/// with the value of expr if it isn't LZMA_OK. This macro shortens
+/// the code a little.
+#define return_if_error(expr) \
+do { \
+ const lzma_ret ret_ = (expr); \
+ if (ret_ != LZMA_OK) \
+ return ret_; \
+} while (0)
+
+
+/// If next isn't already initialized, free the previous coder. Then mark
+/// that next is _possibly_ initialized for the coder using this macro.
+/// "Possibly" means that if e.g. allocation of next->coder fails, the
+/// structure isn't actually initialized for this coder, but leaving
+/// next->init to func is still OK.
+#define lzma_next_coder_init(func, next, allocator) \
+do { \
+ if ((uintptr_t)(func) != (next)->init) \
+ lzma_next_end(next, allocator); \
+ (next)->init = (uintptr_t)(func); \
+} while (0)
+
+
+/// Initializes lzma_strm and calls func() to initialize strm->internal->next.
+/// (The function being called will use lzma_next_coder_init()). If
+/// initialization fails, memory that wasn't freed by func() is freed
+/// along strm->internal.
+#define lzma_next_strm_init(func, strm, ...) \
+do { \
+ return_if_error(lzma_strm_init(strm)); \
+ const lzma_ret ret_ = func(&(strm)->internal->next, \
+ (strm)->allocator, __VA_ARGS__); \
+ if (ret_ != LZMA_OK) { \
+ lzma_end(strm); \
+ return ret_; \
+ } \
+} while (0)
+
+#endif
diff --git a/src/liblzma/common/easy_buffer_encoder.c b/src/liblzma/common/easy_buffer_encoder.c
new file mode 100644
index 0000000..c4be34c
--- /dev/null
+++ b/src/liblzma/common/easy_buffer_encoder.c
@@ -0,0 +1,27 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file easy_buffer_encoder.c
+/// \brief Easy single-call .xz Stream encoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "easy_preset.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_easy_buffer_encode(uint32_t preset, lzma_check check,
+ lzma_allocator *allocator, const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ lzma_options_easy opt_easy;
+ if (lzma_easy_preset(&opt_easy, preset))
+ return LZMA_OPTIONS_ERROR;
+
+ return lzma_stream_buffer_encode(opt_easy.filters, check,
+ allocator, in, in_size, out, out_pos, out_size);
+}
diff --git a/src/liblzma/common/easy_decoder_memusage.c b/src/liblzma/common/easy_decoder_memusage.c
new file mode 100644
index 0000000..20bcd5b
--- /dev/null
+++ b/src/liblzma/common/easy_decoder_memusage.c
@@ -0,0 +1,24 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file easy_decoder_memusage.c
+/// \brief Decoder memory usage calculation to match easy encoder presets
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "easy_preset.h"
+
+
+extern LZMA_API(uint64_t)
+lzma_easy_decoder_memusage(uint32_t preset)
+{
+ lzma_options_easy opt_easy;
+ if (lzma_easy_preset(&opt_easy, preset))
+ return UINT32_MAX;
+
+ return lzma_raw_decoder_memusage(opt_easy.filters);
+}
diff --git a/src/liblzma/common/easy_encoder.c b/src/liblzma/common/easy_encoder.c
new file mode 100644
index 0000000..d13ccd7
--- /dev/null
+++ b/src/liblzma/common/easy_encoder.c
@@ -0,0 +1,25 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file easy_encoder.c
+/// \brief Easy .xz Stream encoder initialization
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "easy_preset.h"
+#include "stream_encoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_easy_encoder(lzma_stream *strm, uint32_t preset, lzma_check check)
+{
+ lzma_options_easy opt_easy;
+ if (lzma_easy_preset(&opt_easy, preset))
+ return LZMA_OPTIONS_ERROR;
+
+ return lzma_stream_encoder(strm, opt_easy.filters, check);
+}
diff --git a/src/liblzma/common/easy_encoder_memusage.c b/src/liblzma/common/easy_encoder_memusage.c
new file mode 100644
index 0000000..e910575
--- /dev/null
+++ b/src/liblzma/common/easy_encoder_memusage.c
@@ -0,0 +1,24 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file easy_encoder_memusage.c
+/// \brief Easy .xz Stream encoder memory usage calculation
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "easy_preset.h"
+
+
+extern LZMA_API(uint64_t)
+lzma_easy_encoder_memusage(uint32_t preset)
+{
+ lzma_options_easy opt_easy;
+ if (lzma_easy_preset(&opt_easy, preset))
+ return UINT32_MAX;
+
+ return lzma_raw_encoder_memusage(opt_easy.filters);
+}
diff --git a/src/liblzma/common/easy_preset.c b/src/liblzma/common/easy_preset.c
new file mode 100644
index 0000000..2f98598
--- /dev/null
+++ b/src/liblzma/common/easy_preset.c
@@ -0,0 +1,27 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file easy_preset.c
+/// \brief Preset handling for easy encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "easy_preset.h"
+
+
+extern bool
+lzma_easy_preset(lzma_options_easy *opt_easy, uint32_t preset)
+{
+ if (lzma_lzma_preset(&opt_easy->opt_lzma, preset))
+ return true;
+
+ opt_easy->filters[0].id = LZMA_FILTER_LZMA2;
+ opt_easy->filters[0].options = &opt_easy->opt_lzma;
+ opt_easy->filters[1].id = LZMA_VLI_UNKNOWN;
+
+ return false;
+}
diff --git a/src/liblzma/common/easy_preset.h b/src/liblzma/common/easy_preset.h
new file mode 100644
index 0000000..382ade8
--- /dev/null
+++ b/src/liblzma/common/easy_preset.h
@@ -0,0 +1,32 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file easy_preset.h
+/// \brief Preset handling for easy encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+
+typedef struct {
+ /// We need to keep the filters array available in case
+ /// LZMA_FULL_FLUSH is used.
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+
+ /// Options for LZMA2
+ lzma_options_lzma opt_lzma;
+
+ // Options for more filters can be added later, so this struct
+ // is not ready to be put into the public API.
+
+} lzma_options_easy;
+
+
+/// Set *easy to the settings given by the preset. Returns true on error,
+/// false on success.
+extern bool lzma_easy_preset(lzma_options_easy *easy, uint32_t preset);
diff --git a/src/liblzma/common/filter_buffer_decoder.c b/src/liblzma/common/filter_buffer_decoder.c
new file mode 100644
index 0000000..2d35ef8
--- /dev/null
+++ b/src/liblzma/common/filter_buffer_decoder.c
@@ -0,0 +1,87 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_buffer_decoder.c
+/// \brief Single-call raw decoding
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_decoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_raw_buffer_decode(const lzma_filter *filters, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // Validate what isn't validated later in filter_common.c.
+ if (in == NULL || in_pos == NULL || *in_pos > in_size || out == NULL
+ || out_pos == NULL || *out_pos > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Initialize the decoer.
+ lzma_next_coder next = LZMA_NEXT_CODER_INIT;
+ return_if_error(lzma_raw_decoder_init(&next, allocator, filters));
+
+ // Store the positions so that we can restore them if something
+ // goes wrong.
+ const size_t in_start = *in_pos;
+ const size_t out_start = *out_pos;
+
+ // Do the actual decoding and free decoder's memory.
+ lzma_ret ret = next.code(next.coder, allocator, in, in_pos, in_size,
+ out, out_pos, out_size, LZMA_FINISH);
+
+ if (ret == LZMA_STREAM_END) {
+ ret = LZMA_OK;
+ } else {
+ if (ret == LZMA_OK) {
+ // Either the input was truncated or the
+ // output buffer was too small.
+ assert(*in_pos == in_size || *out_pos == out_size);
+
+ if (*in_pos != in_size) {
+ // Since input wasn't consumed completely,
+ // the output buffer became full and is
+ // too small.
+ ret = LZMA_BUF_ERROR;
+
+ } else if (*out_pos != out_size) {
+ // Since output didn't became full, the input
+ // has to be truncated.
+ ret = LZMA_DATA_ERROR;
+
+ } else {
+ // All the input was consumed and output
+ // buffer is full. Now we don't immediately
+ // know the reason for the error. Try
+ // decoding one more byte. If it succeeds,
+ // then the output buffer was too small. If
+ // we cannot get a new output byte, the input
+ // is truncated.
+ uint8_t tmp[1];
+ size_t tmp_pos = 0;
+ (void)next.code(next.coder, allocator,
+ in, in_pos, in_size,
+ tmp, &tmp_pos, 1, LZMA_FINISH);
+
+ if (tmp_pos == 1)
+ ret = LZMA_BUF_ERROR;
+ else
+ ret = LZMA_DATA_ERROR;
+ }
+ }
+
+ // Restore the positions.
+ *in_pos = in_start;
+ *out_pos = out_start;
+ }
+
+ lzma_next_end(&next, allocator);
+
+ return ret;
+}
diff --git a/src/liblzma/common/filter_buffer_encoder.c b/src/liblzma/common/filter_buffer_encoder.c
new file mode 100644
index 0000000..646e1b3
--- /dev/null
+++ b/src/liblzma/common/filter_buffer_encoder.c
@@ -0,0 +1,54 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_buffer_encoder.c
+/// \brief Single-call raw encoding
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_encoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_raw_buffer_encode(const lzma_filter *filters, lzma_allocator *allocator,
+ const uint8_t *in, size_t in_size, uint8_t *out,
+ size_t *out_pos, size_t out_size)
+{
+ // Validate what isn't validated later in filter_common.c.
+ if ((in == NULL && in_size != 0) || out == NULL
+ || out_pos == NULL || *out_pos > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Initialize the encoder
+ lzma_next_coder next = LZMA_NEXT_CODER_INIT;
+ return_if_error(lzma_raw_encoder_init(&next, allocator, filters));
+
+ // Store the output position so that we can restore it if
+ // something goes wrong.
+ const size_t out_start = *out_pos;
+
+ // Do the actual encoding and free coder's memory.
+ size_t in_pos = 0;
+ lzma_ret ret = next.code(next.coder, allocator, in, &in_pos, in_size,
+ out, out_pos, out_size, LZMA_FINISH);
+ lzma_next_end(&next, allocator);
+
+ if (ret == LZMA_STREAM_END) {
+ ret = LZMA_OK;
+ } else {
+ if (ret == LZMA_OK) {
+ // Output buffer was too small.
+ assert(*out_pos == out_size);
+ ret = LZMA_BUF_ERROR;
+ }
+
+ // Restore the output position.
+ *out_pos = out_start;
+ }
+
+ return ret;
+}
diff --git a/src/liblzma/common/filter_common.c b/src/liblzma/common/filter_common.c
new file mode 100644
index 0000000..b157c62
--- /dev/null
+++ b/src/liblzma/common/filter_common.c
@@ -0,0 +1,337 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_common.c
+/// \brief Filter-specific stuff common for both encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_common.h"
+
+
+static const struct {
+ /// Filter ID
+ lzma_vli id;
+
+ /// Size of the filter-specific options structure
+ size_t options_size;
+
+ /// True if it is OK to use this filter as non-last filter in
+ /// the chain.
+ bool non_last_ok;
+
+ /// True if it is OK to use this filter as the last filter in
+ /// the chain.
+ bool last_ok;
+
+ /// True if the filter may change the size of the data (that is, the
+ /// amount of encoded output can be different than the amount of
+ /// uncompressed input).
+ bool changes_size;
+
+} features[] = {
+#if defined (HAVE_ENCODER_LZMA1) || defined(HAVE_DECODER_LZMA1)
+ {
+ .id = LZMA_FILTER_LZMA1,
+ .options_size = sizeof(lzma_options_lzma),
+ .non_last_ok = false,
+ .last_ok = true,
+ .changes_size = true,
+ },
+#endif
+#ifdef HAVE_DECODER_LZMA2
+ {
+ .id = LZMA_FILTER_LZMA2,
+ .options_size = sizeof(lzma_options_lzma),
+ .non_last_ok = false,
+ .last_ok = true,
+ .changes_size = true,
+ },
+#endif
+#ifdef HAVE_DECODER_X86
+ {
+ .id = LZMA_FILTER_X86,
+ .options_size = sizeof(lzma_options_bcj),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+#if defined(HAVE_ENCODER_POWERPC) || defined(HAVE_DECODER_POWERPC)
+ {
+ .id = LZMA_FILTER_POWERPC,
+ .options_size = sizeof(lzma_options_bcj),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+#ifdef HAVE_DECODER_IA64
+ {
+ .id = LZMA_FILTER_IA64,
+ .options_size = sizeof(lzma_options_bcj),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+#if defined(HAVE_ENCODER_ARM) || defined(HAVE_DECODER_ARM)
+ {
+ .id = LZMA_FILTER_ARM,
+ .options_size = sizeof(lzma_options_bcj),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+#if defined(HAVE_ENCODER_ARMTHUMB) || defined(HAVE_DECODER_ARMTHUMB)
+ {
+ .id = LZMA_FILTER_ARMTHUMB,
+ .options_size = sizeof(lzma_options_bcj),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+#if defined(HAVE_ENCODER_SPARC) || defined(HAVE_DECODER_SPARC)
+ {
+ .id = LZMA_FILTER_SPARC,
+ .options_size = sizeof(lzma_options_bcj),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+#if defined(HAVE_ENCODER_DELTA) || defined(HAVE_DECODER_DELTA)
+ {
+ .id = LZMA_FILTER_DELTA,
+ .options_size = sizeof(lzma_options_delta),
+ .non_last_ok = true,
+ .last_ok = false,
+ .changes_size = false,
+ },
+#endif
+ {
+ .id = LZMA_VLI_UNKNOWN
+ }
+};
+
+
+extern LZMA_API(lzma_ret)
+lzma_filters_copy(const lzma_filter *src, lzma_filter *dest,
+ lzma_allocator *allocator)
+{
+ if (src == NULL || dest == NULL)
+ return LZMA_PROG_ERROR;
+
+ lzma_ret ret;
+ size_t i;
+ for (i = 0; src[i].id != LZMA_VLI_UNKNOWN; ++i) {
+ // There must be a maximum of four filters plus
+ // the array terminator.
+ if (i == LZMA_FILTERS_MAX) {
+ ret = LZMA_OPTIONS_ERROR;
+ goto error;
+ }
+
+ dest[i].id = src[i].id;
+
+ if (src[i].options == NULL) {
+ dest[i].options = NULL;
+ } else {
+ // See if the filter is supported only when the
+ // options is not NULL. This might be convenient
+ // sometimes if the app is actually copying only
+ // a partial filter chain with a place holder ID.
+ //
+ // When options is not NULL, the Filter ID must be
+ // supported by us, because otherwise we don't know
+ // how big the options are.
+ size_t j;
+ for (j = 0; src[i].id != features[j].id; ++j) {
+ if (features[j].id == LZMA_VLI_UNKNOWN) {
+ ret = LZMA_OPTIONS_ERROR;
+ goto error;
+ }
+ }
+
+ // Allocate and copy the options.
+ dest[i].options = lzma_alloc(features[j].options_size,
+ allocator);
+ if (dest[i].options == NULL) {
+ ret = LZMA_MEM_ERROR;
+ goto error;
+ }
+
+ memcpy(dest[i].options, src[i].options,
+ features[j].options_size);
+ }
+ }
+
+ // Terminate the filter array.
+ assert(i <= LZMA_FILTERS_MAX + 1);
+ dest[i].id = LZMA_VLI_UNKNOWN;
+ dest[i].options = NULL;
+
+ return LZMA_OK;
+
+error:
+ // Free the options which we have already allocated.
+ while (i-- > 0) {
+ lzma_free(dest[i].options, allocator);
+ dest[i].options = NULL;
+ }
+
+ return ret;
+}
+
+
+static lzma_ret
+validate_chain(const lzma_filter *filters, size_t *count)
+{
+ // There must be at least one filter.
+ if (filters == NULL || filters[0].id == LZMA_VLI_UNKNOWN)
+ return LZMA_PROG_ERROR;
+
+ // Number of non-last filters that may change the size of the data
+ // significantly (that is, more than 1-2 % or so).
+ size_t changes_size_count = 0;
+
+ // True if it is OK to add a new filter after the current filter.
+ bool non_last_ok = true;
+
+ // True if the last filter in the given chain is actually usable as
+ // the last filter. Only filters that support embedding End of Payload
+ // Marker can be used as the last filter in the chain.
+ bool last_ok = false;
+
+ size_t i = 0;
+ do {
+ size_t j;
+ for (j = 0; filters[i].id != features[j].id; ++j)
+ if (features[j].id == LZMA_VLI_UNKNOWN)
+ return LZMA_OPTIONS_ERROR;
+
+ // If the previous filter in the chain cannot be a non-last
+ // filter, the chain is invalid.
+ if (!non_last_ok)
+ return LZMA_OPTIONS_ERROR;
+
+ non_last_ok = features[j].non_last_ok;
+ last_ok = features[j].last_ok;
+ changes_size_count += features[j].changes_size;
+
+ } while (filters[++i].id != LZMA_VLI_UNKNOWN);
+
+ // There must be 1-4 filters. The last filter must be usable as
+ // the last filter in the chain. A maximum of three filters are
+ // allowed to change the size of the data.
+ if (i > LZMA_FILTERS_MAX || !last_ok || changes_size_count > 3)
+ return LZMA_OPTIONS_ERROR;
+
+ *count = i;
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_raw_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *options,
+ lzma_filter_find coder_find, bool is_encoder)
+{
+ // Do some basic validation and get the number of filters.
+ size_t count;
+ return_if_error(validate_chain(options, &count));
+
+ // Set the filter functions and copy the options pointer.
+ lzma_filter_info filters[LZMA_FILTERS_MAX + 1];
+ if (is_encoder) {
+ for (size_t i = 0; i < count; ++i) {
+ // The order of the filters is reversed in the
+ // encoder. It allows more efficient handling
+ // of the uncompressed data.
+ const size_t j = count - i - 1;
+
+ const lzma_filter_coder *const fc
+ = coder_find(options[i].id);
+ if (fc == NULL || fc->init == NULL)
+ return LZMA_OPTIONS_ERROR;
+
+ filters[j].id = options[i].id;
+ filters[j].init = fc->init;
+ filters[j].options = options[i].options;
+ }
+ } else {
+ for (size_t i = 0; i < count; ++i) {
+ const lzma_filter_coder *const fc
+ = coder_find(options[i].id);
+ if (fc == NULL || fc->init == NULL)
+ return LZMA_OPTIONS_ERROR;
+
+ filters[i].id = options[i].id;
+ filters[i].init = fc->init;
+ filters[i].options = options[i].options;
+ }
+ }
+
+ // Terminate the array.
+ filters[count].id = LZMA_VLI_UNKNOWN;
+ filters[count].init = NULL;
+
+ // Initialize the filters.
+ const lzma_ret ret = lzma_next_filter_init(next, allocator, filters);
+ if (ret != LZMA_OK)
+ lzma_next_end(next, allocator);
+
+ return ret;
+}
+
+
+extern uint64_t
+lzma_raw_coder_memusage(lzma_filter_find coder_find,
+ const lzma_filter *filters)
+{
+ // The chain has to have at least one filter.
+ {
+ size_t tmp;
+ if (validate_chain(filters, &tmp) != LZMA_OK)
+ return UINT64_MAX;
+ }
+
+ uint64_t total = 0;
+ size_t i = 0;
+
+ do {
+ const lzma_filter_coder *const fc
+ = coder_find(filters[i].id);
+ if (fc == NULL)
+ return UINT64_MAX; // Unsupported Filter ID
+
+ if (fc->memusage == NULL) {
+ // This filter doesn't have a function to calculate
+ // the memory usage and validate the options. Such
+ // filters need only little memory, so we use 1 KiB
+ // as a good estimate. They also accept all possible
+ // options, so there's no need to worry about lack
+ // of validation.
+ total += 1024;
+ } else {
+ // Call the filter-specific memory usage calculation
+ // function.
+ const uint64_t usage
+ = fc->memusage(filters[i].options);
+ if (usage == UINT64_MAX)
+ return UINT64_MAX; // Invalid options
+
+ total += usage;
+ }
+ } while (filters[++i].id != LZMA_VLI_UNKNOWN);
+
+ // Add some fixed amount of extra. It's to compensate memory usage
+ // of Stream, Block etc. coders, malloc() overhead, stack etc.
+ return total + LZMA_MEMUSAGE_BASE;
+}
diff --git a/src/liblzma/common/filter_common.h b/src/liblzma/common/filter_common.h
new file mode 100644
index 0000000..cd61fc0
--- /dev/null
+++ b/src/liblzma/common/filter_common.h
@@ -0,0 +1,48 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_common.c
+/// \brief Filter-specific stuff common for both encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_FILTER_COMMON_H
+#define LZMA_FILTER_COMMON_H
+
+#include "common.h"
+
+
+/// Both lzma_filter_encoder and lzma_filter_decoder begin with these members.
+typedef struct {
+ /// Filter ID
+ lzma_vli id;
+
+ /// Initializes the filter encoder and calls lzma_next_filter_init()
+ /// for filters + 1.
+ lzma_init_function init;
+
+ /// Calculates memory usage of the encoder. If the options are
+ /// invalid, UINT64_MAX is returned.
+ uint64_t (*memusage)(const void *options);
+
+} lzma_filter_coder;
+
+
+typedef const lzma_filter_coder *(*lzma_filter_find)(lzma_vli id);
+
+
+extern lzma_ret lzma_raw_coder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *filters,
+ lzma_filter_find coder_find, bool is_encoder);
+
+
+extern uint64_t lzma_raw_coder_memusage(lzma_filter_find coder_find,
+ const lzma_filter *filters);
+
+
+#endif
diff --git a/src/liblzma/common/filter_decoder.c b/src/liblzma/common/filter_decoder.c
new file mode 100644
index 0000000..1ebbe2a
--- /dev/null
+++ b/src/liblzma/common/filter_decoder.c
@@ -0,0 +1,183 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_decoder.c
+/// \brief Filter ID mapping to filter-specific functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_decoder.h"
+#include "filter_common.h"
+#include "lzma_decoder.h"
+#include "lzma2_decoder.h"
+#include "simple_decoder.h"
+#include "delta_decoder.h"
+
+
+typedef struct {
+ /// Filter ID
+ lzma_vli id;
+
+ /// Initializes the filter encoder and calls lzma_next_filter_init()
+ /// for filters + 1.
+ lzma_init_function init;
+
+ /// Calculates memory usage of the encoder. If the options are
+ /// invalid, UINT64_MAX is returned.
+ uint64_t (*memusage)(const void *options);
+
+ /// Decodes Filter Properties.
+ ///
+ /// \return - LZMA_OK: Properties decoded successfully.
+ /// - LZMA_OPTIONS_ERROR: Unsupported properties
+ /// - LZMA_MEM_ERROR: Memory allocation failed.
+ lzma_ret (*props_decode)(void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size);
+
+} lzma_filter_decoder;
+
+
+static const lzma_filter_decoder decoders[] = {
+#ifdef HAVE_DECODER_LZMA1
+ {
+ .id = LZMA_FILTER_LZMA1,
+ .init = &lzma_lzma_decoder_init,
+ .memusage = &lzma_lzma_decoder_memusage,
+ .props_decode = &lzma_lzma_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_LZMA2
+ {
+ .id = LZMA_FILTER_LZMA2,
+ .init = &lzma_lzma2_decoder_init,
+ .memusage = &lzma_lzma2_decoder_memusage,
+ .props_decode = &lzma_lzma2_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_X86
+ {
+ .id = LZMA_FILTER_X86,
+ .init = &lzma_simple_x86_decoder_init,
+ .memusage = NULL,
+ .props_decode = &lzma_simple_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_POWERPC
+ {
+ .id = LZMA_FILTER_POWERPC,
+ .init = &lzma_simple_powerpc_decoder_init,
+ .memusage = NULL,
+ .props_decode = &lzma_simple_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_IA64
+ {
+ .id = LZMA_FILTER_IA64,
+ .init = &lzma_simple_ia64_decoder_init,
+ .memusage = NULL,
+ .props_decode = &lzma_simple_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_ARM
+ {
+ .id = LZMA_FILTER_ARM,
+ .init = &lzma_simple_arm_decoder_init,
+ .memusage = NULL,
+ .props_decode = &lzma_simple_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_ARMTHUMB
+ {
+ .id = LZMA_FILTER_ARMTHUMB,
+ .init = &lzma_simple_armthumb_decoder_init,
+ .memusage = NULL,
+ .props_decode = &lzma_simple_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_SPARC
+ {
+ .id = LZMA_FILTER_SPARC,
+ .init = &lzma_simple_sparc_decoder_init,
+ .memusage = NULL,
+ .props_decode = &lzma_simple_props_decode,
+ },
+#endif
+#ifdef HAVE_DECODER_DELTA
+ {
+ .id = LZMA_FILTER_DELTA,
+ .init = &lzma_delta_decoder_init,
+ .memusage = &lzma_delta_coder_memusage,
+ .props_decode = &lzma_delta_props_decode,
+ },
+#endif
+};
+
+
+static const lzma_filter_decoder *
+decoder_find(lzma_vli id)
+{
+ for (size_t i = 0; i < ARRAY_SIZE(decoders); ++i)
+ if (decoders[i].id == id)
+ return decoders + i;
+
+ return NULL;
+}
+
+
+extern LZMA_API(lzma_bool)
+lzma_filter_decoder_is_supported(lzma_vli id)
+{
+ return decoder_find(id) != NULL;
+}
+
+
+extern lzma_ret
+lzma_raw_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *options)
+{
+ return lzma_raw_coder_init(next, allocator,
+ options, (lzma_filter_find)(&decoder_find), false);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_raw_decoder(lzma_stream *strm, const lzma_filter *options)
+{
+ lzma_next_strm_init(lzma_raw_decoder_init, strm, options);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_raw_decoder_memusage(const lzma_filter *filters)
+{
+ return lzma_raw_coder_memusage(
+ (lzma_filter_find)(&decoder_find), filters);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_properties_decode(lzma_filter *filter, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size)
+{
+ // Make it always NULL so that the caller can always safely free() it.
+ filter->options = NULL;
+
+ const lzma_filter_decoder *const fd = decoder_find(filter->id);
+ if (fd == NULL)
+ return LZMA_OPTIONS_ERROR;
+
+ if (fd->props_decode == NULL)
+ return props_size == 0 ? LZMA_OK : LZMA_OPTIONS_ERROR;
+
+ return fd->props_decode(
+ &filter->options, allocator, props, props_size);
+}
diff --git a/src/liblzma/common/filter_decoder.h b/src/liblzma/common/filter_decoder.h
new file mode 100644
index 0000000..d5c68bd
--- /dev/null
+++ b/src/liblzma/common/filter_decoder.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_decoder.c
+/// \brief Filter ID mapping to filter-specific functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_FILTER_DECODER_H
+#define LZMA_FILTER_DECODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_raw_decoder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *options);
+
+#endif
diff --git a/src/liblzma/common/filter_encoder.c b/src/liblzma/common/filter_encoder.c
new file mode 100644
index 0000000..436d2cc
--- /dev/null
+++ b/src/liblzma/common/filter_encoder.c
@@ -0,0 +1,286 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_decoder.c
+/// \brief Filter ID mapping to filter-specific functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_encoder.h"
+#include "filter_common.h"
+#include "lzma_encoder.h"
+#include "lzma2_encoder.h"
+#include "simple_encoder.h"
+#include "delta_encoder.h"
+
+
+typedef struct {
+ /// Filter ID
+ lzma_vli id;
+
+ /// Initializes the filter encoder and calls lzma_next_filter_init()
+ /// for filters + 1.
+ lzma_init_function init;
+
+ /// Calculates memory usage of the encoder. If the options are
+ /// invalid, UINT64_MAX is returned.
+ uint64_t (*memusage)(const void *options);
+
+ /// Calculates the minimum sane size for Blocks (or other types of
+ /// chunks) to which the input data can be split to make
+ /// multithreaded encoding possible. If this is NULL, it is assumed
+ /// that the encoder is fast enough with single thread.
+ lzma_vli (*chunk_size)(const void *options);
+
+ /// Tells the size of the Filter Properties field. If options are
+ /// invalid, UINT32_MAX is returned. If this is NULL, props_size_fixed
+ /// is used.
+ lzma_ret (*props_size_get)(uint32_t *size, const void *options);
+ uint32_t props_size_fixed;
+
+ /// Encodes Filter Properties.
+ ///
+ /// \return - LZMA_OK: Properties encoded successfully.
+ /// - LZMA_OPTIONS_ERROR: Unsupported options
+ /// - LZMA_PROG_ERROR: Invalid options or not enough
+ /// output space
+ lzma_ret (*props_encode)(const void *options, uint8_t *out);
+
+} lzma_filter_encoder;
+
+
+static const lzma_filter_encoder encoders[] = {
+#ifdef HAVE_ENCODER_LZMA1
+ {
+ .id = LZMA_FILTER_LZMA1,
+ .init = &lzma_lzma_encoder_init,
+ .memusage = &lzma_lzma_encoder_memusage,
+ .chunk_size = NULL, // FIXME
+ .props_size_get = NULL,
+ .props_size_fixed = 5,
+ .props_encode = &lzma_lzma_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_LZMA2
+ {
+ .id = LZMA_FILTER_LZMA2,
+ .init = &lzma_lzma2_encoder_init,
+ .memusage = &lzma_lzma2_encoder_memusage,
+ .chunk_size = NULL, // FIXME
+ .props_size_get = NULL,
+ .props_size_fixed = 1,
+ .props_encode = &lzma_lzma2_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_X86
+ {
+ .id = LZMA_FILTER_X86,
+ .init = &lzma_simple_x86_encoder_init,
+ .memusage = NULL,
+ .chunk_size = NULL,
+ .props_size_get = &lzma_simple_props_size,
+ .props_encode = &lzma_simple_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_POWERPC
+ {
+ .id = LZMA_FILTER_POWERPC,
+ .init = &lzma_simple_powerpc_encoder_init,
+ .memusage = NULL,
+ .chunk_size = NULL,
+ .props_size_get = &lzma_simple_props_size,
+ .props_encode = &lzma_simple_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_IA64
+ {
+ .id = LZMA_FILTER_IA64,
+ .init = &lzma_simple_ia64_encoder_init,
+ .memusage = NULL,
+ .chunk_size = NULL,
+ .props_size_get = &lzma_simple_props_size,
+ .props_encode = &lzma_simple_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_ARM
+ {
+ .id = LZMA_FILTER_ARM,
+ .init = &lzma_simple_arm_encoder_init,
+ .memusage = NULL,
+ .chunk_size = NULL,
+ .props_size_get = &lzma_simple_props_size,
+ .props_encode = &lzma_simple_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_ARMTHUMB
+ {
+ .id = LZMA_FILTER_ARMTHUMB,
+ .init = &lzma_simple_armthumb_encoder_init,
+ .memusage = NULL,
+ .chunk_size = NULL,
+ .props_size_get = &lzma_simple_props_size,
+ .props_encode = &lzma_simple_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_SPARC
+ {
+ .id = LZMA_FILTER_SPARC,
+ .init = &lzma_simple_sparc_encoder_init,
+ .memusage = NULL,
+ .chunk_size = NULL,
+ .props_size_get = &lzma_simple_props_size,
+ .props_encode = &lzma_simple_props_encode,
+ },
+#endif
+#ifdef HAVE_ENCODER_DELTA
+ {
+ .id = LZMA_FILTER_DELTA,
+ .init = &lzma_delta_encoder_init,
+ .memusage = &lzma_delta_coder_memusage,
+ .chunk_size = NULL,
+ .props_size_get = NULL,
+ .props_size_fixed = 1,
+ .props_encode = &lzma_delta_props_encode,
+ },
+#endif
+};
+
+
+static const lzma_filter_encoder *
+encoder_find(lzma_vli id)
+{
+ for (size_t i = 0; i < ARRAY_SIZE(encoders); ++i)
+ if (encoders[i].id == id)
+ return encoders + i;
+
+ return NULL;
+}
+
+
+extern LZMA_API(lzma_bool)
+lzma_filter_encoder_is_supported(lzma_vli id)
+{
+ return encoder_find(id) != NULL;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_filters_update(lzma_stream *strm, const lzma_filter *filters)
+{
+ if (strm->internal->next.update == NULL)
+ return LZMA_PROG_ERROR;
+
+ // Validate the filter chain.
+ if (lzma_raw_encoder_memusage(filters) == UINT64_MAX)
+ return LZMA_OPTIONS_ERROR;
+
+ // The actual filter chain in the encoder is reversed. Some things
+ // still want the normal order chain, so we provide both.
+ size_t count = 1;
+ while (filters[count].id != LZMA_VLI_UNKNOWN)
+ ++count;
+
+ lzma_filter reversed_filters[LZMA_FILTERS_MAX + 1];
+ for (size_t i = 0; i < count; ++i)
+ reversed_filters[count - i - 1] = filters[i];
+
+ reversed_filters[count].id = LZMA_VLI_UNKNOWN;
+
+ return strm->internal->next.update(strm->internal->next.coder,
+ strm->allocator, filters, reversed_filters);
+}
+
+
+extern lzma_ret
+lzma_raw_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *options)
+{
+ return lzma_raw_coder_init(next, allocator,
+ options, (lzma_filter_find)(&encoder_find), true);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_raw_encoder(lzma_stream *strm, const lzma_filter *options)
+{
+ lzma_next_strm_init(lzma_raw_coder_init, strm, options,
+ (lzma_filter_find)(&encoder_find), true);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_SYNC_FLUSH] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_raw_encoder_memusage(const lzma_filter *filters)
+{
+ return lzma_raw_coder_memusage(
+ (lzma_filter_find)(&encoder_find), filters);
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_chunk_size(const lzma_filter *filters)
+{
+ lzma_vli max = 0;
+
+ for (size_t i = 0; filters[i].id != LZMA_VLI_UNKNOWN; ++i) {
+ const lzma_filter_encoder *const fe
+ = encoder_find(filters[i].id);
+ if (fe->chunk_size != NULL) {
+ const lzma_vli size
+ = fe->chunk_size(filters[i].options);
+ if (size == LZMA_VLI_UNKNOWN)
+ return LZMA_VLI_UNKNOWN;
+
+ if (size > max)
+ max = size;
+ }
+ }
+
+ return max;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_properties_size(uint32_t *size, const lzma_filter *filter)
+{
+ const lzma_filter_encoder *const fe = encoder_find(filter->id);
+ if (fe == NULL) {
+ // Unknown filter - if the Filter ID is a proper VLI,
+ // return LZMA_OPTIONS_ERROR instead of LZMA_PROG_ERROR,
+ // because it's possible that we just don't have support
+ // compiled in for the requested filter.
+ return filter->id <= LZMA_VLI_MAX
+ ? LZMA_OPTIONS_ERROR : LZMA_PROG_ERROR;
+ }
+
+ if (fe->props_size_get == NULL) {
+ // No props_size_get() function, use props_size_fixed.
+ *size = fe->props_size_fixed;
+ return LZMA_OK;
+ }
+
+ return fe->props_size_get(size, filter->options);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_properties_encode(const lzma_filter *filter, uint8_t *props)
+{
+ const lzma_filter_encoder *const fe = encoder_find(filter->id);
+ if (fe == NULL)
+ return LZMA_PROG_ERROR;
+
+ if (fe->props_encode == NULL)
+ return LZMA_OK;
+
+ return fe->props_encode(filter->options, props);
+}
diff --git a/src/liblzma/common/filter_encoder.h b/src/liblzma/common/filter_encoder.h
new file mode 100644
index 0000000..a978932
--- /dev/null
+++ b/src/liblzma/common/filter_encoder.h
@@ -0,0 +1,27 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_encoder.c
+/// \brief Filter ID mapping to filter-specific functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_FILTER_ENCODER_H
+#define LZMA_FILTER_ENCODER_H
+
+#include "common.h"
+
+
+// FIXME !!! Public API
+extern lzma_vli lzma_chunk_size(const lzma_filter *filters);
+
+
+extern lzma_ret lzma_raw_encoder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *filters);
+
+#endif
diff --git a/src/liblzma/common/filter_flags_decoder.c b/src/liblzma/common/filter_flags_decoder.c
new file mode 100644
index 0000000..caae10c
--- /dev/null
+++ b/src/liblzma/common/filter_flags_decoder.c
@@ -0,0 +1,46 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_flags_decoder.c
+/// \brief Decodes a Filter Flags field
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_decoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_filter_flags_decode(
+ lzma_filter *filter, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size)
+{
+ // Set the pointer to NULL so the caller can always safely free it.
+ filter->options = NULL;
+
+ // Filter ID
+ return_if_error(lzma_vli_decode(&filter->id, NULL,
+ in, in_pos, in_size));
+
+ if (filter->id >= LZMA_FILTER_RESERVED_START)
+ return LZMA_DATA_ERROR;
+
+ // Size of Properties
+ lzma_vli props_size;
+ return_if_error(lzma_vli_decode(&props_size, NULL,
+ in, in_pos, in_size));
+
+ // Filter Properties
+ if (in_size - *in_pos < props_size)
+ return LZMA_DATA_ERROR;
+
+ const lzma_ret ret = lzma_properties_decode(
+ filter, allocator, in + *in_pos, props_size);
+
+ *in_pos += props_size;
+
+ return ret;
+}
diff --git a/src/liblzma/common/filter_flags_encoder.c b/src/liblzma/common/filter_flags_encoder.c
new file mode 100644
index 0000000..d110566
--- /dev/null
+++ b/src/liblzma/common/filter_flags_encoder.c
@@ -0,0 +1,56 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file filter_flags_encoder.c
+/// \brief Decodes a Filter Flags field
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "filter_encoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_filter_flags_size(uint32_t *size, const lzma_filter *filter)
+{
+ if (filter->id >= LZMA_FILTER_RESERVED_START)
+ return LZMA_PROG_ERROR;
+
+ return_if_error(lzma_properties_size(size, filter));
+
+ *size += lzma_vli_size(filter->id) + lzma_vli_size(*size);
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_filter_flags_encode(const lzma_filter *filter,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // Filter ID
+ if (filter->id >= LZMA_FILTER_RESERVED_START)
+ return LZMA_PROG_ERROR;
+
+ return_if_error(lzma_vli_encode(filter->id, NULL,
+ out, out_pos, out_size));
+
+ // Size of Properties
+ uint32_t props_size;
+ return_if_error(lzma_properties_size(&props_size, filter));
+ return_if_error(lzma_vli_encode(props_size, NULL,
+ out, out_pos, out_size));
+
+ // Filter Properties
+ if (out_size - *out_pos < props_size)
+ return LZMA_PROG_ERROR;
+
+ return_if_error(lzma_properties_encode(filter, out + *out_pos));
+
+ *out_pos += props_size;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/hardware_physmem.c b/src/liblzma/common/hardware_physmem.c
new file mode 100644
index 0000000..7405b65
--- /dev/null
+++ b/src/liblzma/common/hardware_physmem.c
@@ -0,0 +1,25 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file hardware_physmem.c
+/// \brief Get the total amount of physical memory (RAM)
+//
+// Author: Jonathan Nieder
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+#include "tuklib_physmem.h"
+
+
+extern LZMA_API(uint64_t)
+lzma_physmem(void)
+{
+ // It is simpler to make lzma_physmem() a wrapper for
+ // tuklib_physmem() than to hack appropriate symbol visiblity
+ // support for the tuklib modules.
+ return tuklib_physmem();
+}
diff --git a/src/liblzma/common/index.c b/src/liblzma/common/index.c
new file mode 100644
index 0000000..3941e28
--- /dev/null
+++ b/src/liblzma/common/index.c
@@ -0,0 +1,1241 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file index.c
+/// \brief Handling of .xz Indexes and some other Stream information
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "index.h"
+#include "stream_flags_common.h"
+
+
+/// \brief How many Records to allocate at once
+///
+/// This should be big enough to avoid making lots of tiny allocations
+/// but small enough to avoid too much unused memory at once.
+#define INDEX_GROUP_SIZE 500
+
+
+/// \brief How many Records can be allocated at once at maximum
+#define PREALLOC_MAX ((SIZE_MAX - sizeof(index_group)) / sizeof(index_record))
+
+
+/// \brief Base structure for index_stream and index_group structures
+typedef struct index_tree_node_s index_tree_node;
+struct index_tree_node_s {
+ /// Uncompressed start offset of this Stream (relative to the
+ /// beginning of the file) or Block (relative to the beginning
+ /// of the Stream)
+ lzma_vli uncompressed_base;
+
+ /// Compressed start offset of this Stream or Block
+ lzma_vli compressed_base;
+
+ index_tree_node *parent;
+ index_tree_node *left;
+ index_tree_node *right;
+};
+
+
+/// \brief AVL tree to hold index_stream or index_group structures
+typedef struct {
+ /// Root node
+ index_tree_node *root;
+
+ /// Leftmost node. Since the tree will be filled sequentially,
+ /// this won't change after the first node has been added to
+ /// the tree.
+ index_tree_node *leftmost;
+
+ /// The rightmost node in the tree. Since the tree is filled
+ /// sequentially, this is always the node where to add the new data.
+ index_tree_node *rightmost;
+
+ /// Number of nodes in the tree
+ uint32_t count;
+
+} index_tree;
+
+
+typedef struct {
+ lzma_vli uncompressed_sum;
+ lzma_vli unpadded_sum;
+} index_record;
+
+
+typedef struct {
+ /// Every Record group is part of index_stream.groups tree.
+ index_tree_node node;
+
+ /// Number of Blocks in this Stream before this group.
+ lzma_vli number_base;
+
+ /// Number of Records that can be put in records[].
+ size_t allocated;
+
+ /// Index of the last Record in use.
+ size_t last;
+
+ /// The sizes in this array are stored as cumulative sums relative
+ /// to the beginning of the Stream. This makes it possible to
+ /// use binary search in lzma_index_locate().
+ ///
+ /// Note that the cumulative summing is done specially for
+ /// unpadded_sum: The previous value is rounded up to the next
+ /// multiple of four before adding the Unpadded Size of the new
+ /// Block. The total encoded size of the Blocks in the Stream
+ /// is records[last].unpadded_sum in the last Record group of
+ /// the Stream.
+ ///
+ /// For example, if the Unpadded Sizes are 39, 57, and 81, the
+ /// stored values are 39, 97 (40 + 57), and 181 (100 + 181).
+ /// The total encoded size of these Blocks is 184.
+ ///
+ /// This is a flexible array, because it makes easy to optimize
+ /// memory usage in case someone concatenates many Streams that
+ /// have only one or few Blocks.
+ index_record records[];
+
+} index_group;
+
+
+typedef struct {
+ /// Every index_stream is a node in the tree of Sreams.
+ index_tree_node node;
+
+ /// Number of this Stream (first one is 1)
+ uint32_t number;
+
+ /// Total number of Blocks before this Stream
+ lzma_vli block_number_base;
+
+ /// Record groups of this Stream are stored in a tree.
+ /// It's a T-tree with AVL-tree balancing. There are
+ /// INDEX_GROUP_SIZE Records per node by default.
+ /// This keeps the number of memory allocations reasonable
+ /// and finding a Record is fast.
+ index_tree groups;
+
+ /// Number of Records in this Stream
+ lzma_vli record_count;
+
+ /// Size of the List of Records field in this Stream. This is used
+ /// together with record_count to calculate the size of the Index
+ /// field and thus the total size of the Stream.
+ lzma_vli index_list_size;
+
+ /// Stream Flags of this Stream. This is meaningful only if
+ /// the Stream Flags have been told us with lzma_index_stream_flags().
+ /// Initially stream_flags.version is set to UINT32_MAX to indicate
+ /// that the Stream Flags are unknown.
+ lzma_stream_flags stream_flags;
+
+ /// Amount of Stream Padding after this Stream. This defaults to
+ /// zero and can be set with lzma_index_stream_padding().
+ lzma_vli stream_padding;
+
+} index_stream;
+
+
+struct lzma_index_s {
+ /// AVL-tree containing the Stream(s). Often there is just one
+ /// Stream, but using a tree keeps lookups fast even when there
+ /// are many concatenated Streams.
+ index_tree streams;
+
+ /// Uncompressed size of all the Blocks in the Stream(s)
+ lzma_vli uncompressed_size;
+
+ /// Total size of all the Blocks in the Stream(s)
+ lzma_vli total_size;
+
+ /// Total number of Records in all Streams in this lzma_index
+ lzma_vli record_count;
+
+ /// Size of the List of Records field if all the Streams in this
+ /// lzma_index were packed into a single Stream (makes it simpler to
+ /// take many .xz files and combine them into a single Stream).
+ ///
+ /// This value together with record_count is needed to calculate
+ /// Backward Size that is stored into Stream Footer.
+ lzma_vli index_list_size;
+
+ /// How many Records to allocate at once in lzma_index_append().
+ /// This defaults to INDEX_GROUP_SIZE but can be overriden with
+ /// lzma_index_prealloc().
+ size_t prealloc;
+
+ /// Bitmask indicating what integrity check types have been used
+ /// as set by lzma_index_stream_flags(). The bit of the last Stream
+ /// is not included here, since it is possible to change it by
+ /// calling lzma_index_stream_flags() again.
+ uint32_t checks;
+};
+
+
+static void
+index_tree_init(index_tree *tree)
+{
+ tree->root = NULL;
+ tree->leftmost = NULL;
+ tree->rightmost = NULL;
+ tree->count = 0;
+ return;
+}
+
+
+/// Helper for index_tree_end()
+static void
+index_tree_node_end(index_tree_node *node, lzma_allocator *allocator,
+ void (*free_func)(void *node, lzma_allocator *allocator))
+{
+ // The tree won't ever be very huge, so recursion should be fine.
+ // 20 levels in the tree is likely quite a lot already in practice.
+ if (node->left != NULL)
+ index_tree_node_end(node->left, allocator, free_func);
+
+ if (node->right != NULL)
+ index_tree_node_end(node->right, allocator, free_func);
+
+ if (free_func != NULL)
+ free_func(node, allocator);
+
+ lzma_free(node, allocator);
+ return;
+}
+
+
+/// Free the meory allocated for a tree. If free_func is not NULL,
+/// it is called on each node before freeing the node. This is used
+/// to free the Record groups from each index_stream before freeing
+/// the index_stream itself.
+static void
+index_tree_end(index_tree *tree, lzma_allocator *allocator,
+ void (*free_func)(void *node, lzma_allocator *allocator))
+{
+ if (tree->root != NULL)
+ index_tree_node_end(tree->root, allocator, free_func);
+
+ return;
+}
+
+
+/// Add a new node to the tree. node->uncompressed_base and
+/// node->compressed_base must have been set by the caller already.
+static void
+index_tree_append(index_tree *tree, index_tree_node *node)
+{
+ node->parent = tree->rightmost;
+ node->left = NULL;
+ node->right = NULL;
+
+ ++tree->count;
+
+ // Handle the special case of adding the first node.
+ if (tree->root == NULL) {
+ tree->root = node;
+ tree->leftmost = node;
+ tree->rightmost = node;
+ return;
+ }
+
+ // The tree is always filled sequentially.
+ assert(tree->rightmost->uncompressed_base <= node->uncompressed_base);
+ assert(tree->rightmost->compressed_base < node->compressed_base);
+
+ // Add the new node after the rightmost node. It's the correct
+ // place due to the reason above.
+ tree->rightmost->right = node;
+ tree->rightmost = node;
+
+ // Balance the AVL-tree if needed. We don't need to keep the balance
+ // factors in nodes, because we always fill the tree sequentially,
+ // and thus know the state of the tree just by looking at the node
+ // count. From the node count we can calculate how many steps to go
+ // up in the tree to find the rotation root.
+ uint32_t up = tree->count ^ (UINT32_C(1) << bsr32(tree->count));
+ if (up != 0) {
+ // Locate the root node for the rotation.
+ up = ctz32(tree->count) + 2;
+ do {
+ node = node->parent;
+ } while (--up > 0);
+
+ // Rotate left using node as the rotation root.
+ index_tree_node *pivot = node->right;
+
+ if (node->parent == NULL) {
+ tree->root = pivot;
+ } else {
+ assert(node->parent->right == node);
+ node->parent->right = pivot;
+ }
+
+ pivot->parent = node->parent;
+
+ node->right = pivot->left;
+ if (node->right != NULL)
+ node->right->parent = node;
+
+ pivot->left = node;
+ node->parent = pivot;
+ }
+
+ return;
+}
+
+
+/// Get the next node in the tree. Return NULL if there are no more nodes.
+static void *
+index_tree_next(const index_tree_node *node)
+{
+ if (node->right != NULL) {
+ node = node->right;
+ while (node->left != NULL)
+ node = node->left;
+
+ return (void *)(node);
+ }
+
+ while (node->parent != NULL && node->parent->right == node)
+ node = node->parent;
+
+ return (void *)(node->parent);
+}
+
+
+/// Locate a node that contains the given uncompressed offset. It is
+/// caller's job to check that target is not bigger than the uncompressed
+/// size of the tree (the last node would be returned in that case still).
+static void *
+index_tree_locate(const index_tree *tree, lzma_vli target)
+{
+ const index_tree_node *result = NULL;
+ const index_tree_node *node = tree->root;
+
+ assert(tree->leftmost == NULL
+ || tree->leftmost->uncompressed_base == 0);
+
+ // Consecutive nodes may have the same uncompressed_base.
+ // We must pick the rightmost one.
+ while (node != NULL) {
+ if (node->uncompressed_base > target) {
+ node = node->left;
+ } else {
+ result = node;
+ node = node->right;
+ }
+ }
+
+ return (void *)(result);
+}
+
+
+/// Allocate and initialize a new Stream using the given base offsets.
+static index_stream *
+index_stream_init(lzma_vli compressed_base, lzma_vli uncompressed_base,
+ lzma_vli stream_number, lzma_vli block_number_base,
+ lzma_allocator *allocator)
+{
+ index_stream *s = lzma_alloc(sizeof(index_stream), allocator);
+ if (s == NULL)
+ return NULL;
+
+ s->node.uncompressed_base = uncompressed_base;
+ s->node.compressed_base = compressed_base;
+ s->node.parent = NULL;
+ s->node.left = NULL;
+ s->node.right = NULL;
+
+ s->number = stream_number;
+ s->block_number_base = block_number_base;
+
+ index_tree_init(&s->groups);
+
+ s->record_count = 0;
+ s->index_list_size = 0;
+ s->stream_flags.version = UINT32_MAX;
+ s->stream_padding = 0;
+
+ return s;
+}
+
+
+/// Free the memory allocated for a Stream and its Record groups.
+static void
+index_stream_end(void *node, lzma_allocator *allocator)
+{
+ index_stream *s = node;
+ index_tree_end(&s->groups, allocator, NULL);
+ return;
+}
+
+
+static lzma_index *
+index_init_plain(lzma_allocator *allocator)
+{
+ lzma_index *i = lzma_alloc(sizeof(lzma_index), allocator);
+ if (i != NULL) {
+ index_tree_init(&i->streams);
+ i->uncompressed_size = 0;
+ i->total_size = 0;
+ i->record_count = 0;
+ i->index_list_size = 0;
+ i->prealloc = INDEX_GROUP_SIZE;
+ i->checks = 0;
+ }
+
+ return i;
+}
+
+
+extern LZMA_API(lzma_index *)
+lzma_index_init(lzma_allocator *allocator)
+{
+ lzma_index *i = index_init_plain(allocator);
+ index_stream *s = index_stream_init(0, 0, 1, 0, allocator);
+ if (i == NULL || s == NULL) {
+ index_stream_end(s, allocator);
+ lzma_free(i, allocator);
+ }
+
+ index_tree_append(&i->streams, &s->node);
+
+ return i;
+}
+
+
+extern LZMA_API(void)
+lzma_index_end(lzma_index *i, lzma_allocator *allocator)
+{
+ // NOTE: If you modify this function, check also the bottom
+ // of lzma_index_cat().
+ if (i != NULL) {
+ index_tree_end(&i->streams, allocator, &index_stream_end);
+ lzma_free(i, allocator);
+ }
+
+ return;
+}
+
+
+extern void
+lzma_index_prealloc(lzma_index *i, lzma_vli records)
+{
+ if (records > PREALLOC_MAX)
+ records = PREALLOC_MAX;
+
+ i->prealloc = (size_t)(records);
+ return;
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_index_memusage(lzma_vli streams, lzma_vli blocks)
+{
+ // This calculates an upper bound that is only a little bit
+ // bigger than the exact maximum memory usage with the given
+ // parameters.
+
+ // Typical malloc() overhead is 2 * sizeof(void *) but we take
+ // a little bit extra just in case. Using LZMA_MEMUSAGE_BASE
+ // instead would give too inaccurate estimate.
+ const size_t alloc_overhead = 4 * sizeof(void *);
+
+ // Amount of memory needed for each Stream base structures.
+ // We assume that every Stream has at least one Block and
+ // thus at least one group.
+ const size_t stream_base = sizeof(index_stream)
+ + sizeof(index_group) + 2 * alloc_overhead;
+
+ // Amount of memory needed per group.
+ const size_t group_base = sizeof(index_group)
+ + INDEX_GROUP_SIZE * sizeof(index_record)
+ + alloc_overhead;
+
+ // Number of groups. There may actually be more, but that overhead
+ // has been taken into account in stream_base already.
+ const lzma_vli groups
+ = (blocks + INDEX_GROUP_SIZE - 1) / INDEX_GROUP_SIZE;
+
+ // Memory used by index_stream and index_group structures.
+ const uint64_t streams_mem = streams * stream_base;
+ const uint64_t groups_mem = groups * group_base;
+
+ // Memory used by the base structure.
+ const uint64_t index_base = sizeof(lzma_index) + alloc_overhead;
+
+ // Validate the arguments and catch integer overflows.
+ // Maximum number of Streams is "only" UINT32_MAX, because
+ // that limit is used by the tree containing the Streams.
+ const uint64_t limit = UINT64_MAX - index_base;
+ if (streams == 0 || streams > UINT32_MAX || blocks > LZMA_VLI_MAX
+ || streams > limit / stream_base
+ || groups > limit / group_base
+ || limit - streams_mem < groups_mem)
+ return UINT64_MAX;
+
+ return index_base + streams_mem + groups_mem;
+}
+
+
+extern LZMA_API(uint64_t)
+lzma_index_memused(const lzma_index *i)
+{
+ return lzma_index_memusage(i->streams.count, i->record_count);
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_block_count(const lzma_index *i)
+{
+ return i->record_count;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_stream_count(const lzma_index *i)
+{
+ return i->streams.count;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_size(const lzma_index *i)
+{
+ return index_size(i->record_count, i->index_list_size);
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_total_size(const lzma_index *i)
+{
+ return i->total_size;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_stream_size(const lzma_index *i)
+{
+ // Stream Header + Blocks + Index + Stream Footer
+ return LZMA_STREAM_HEADER_SIZE + i->total_size
+ + index_size(i->record_count, i->index_list_size)
+ + LZMA_STREAM_HEADER_SIZE;
+}
+
+
+static lzma_vli
+index_file_size(lzma_vli compressed_base, lzma_vli unpadded_sum,
+ lzma_vli record_count, lzma_vli index_list_size,
+ lzma_vli stream_padding)
+{
+ // Earlier Streams and Stream Paddings + Stream Header
+ // + Blocks + Index + Stream Footer + Stream Padding
+ //
+ // This might go over LZMA_VLI_MAX due to too big unpadded_sum
+ // when this function is used in lzma_index_append().
+ lzma_vli file_size = compressed_base + 2 * LZMA_STREAM_HEADER_SIZE
+ + stream_padding + vli_ceil4(unpadded_sum);
+ if (file_size > LZMA_VLI_MAX)
+ return LZMA_VLI_UNKNOWN;
+
+ // The same applies here.
+ file_size += index_size(record_count, index_list_size);
+ if (file_size > LZMA_VLI_MAX)
+ return LZMA_VLI_UNKNOWN;
+
+ return file_size;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_file_size(const lzma_index *i)
+{
+ const index_stream *s = (const index_stream *)(i->streams.rightmost);
+ const index_group *g = (const index_group *)(s->groups.rightmost);
+ return index_file_size(s->node.compressed_base,
+ g == NULL ? 0 : g->records[g->last].unpadded_sum,
+ s->record_count, s->index_list_size,
+ s->stream_padding);
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_uncompressed_size(const lzma_index *i)
+{
+ return i->uncompressed_size;
+}
+
+
+extern LZMA_API(uint32_t)
+lzma_index_checks(const lzma_index *i)
+{
+ uint32_t checks = i->checks;
+
+ // Get the type of the Check of the last Stream too.
+ const index_stream *s = (const index_stream *)(i->streams.rightmost);
+ if (s->stream_flags.version != UINT32_MAX)
+ checks |= UINT32_C(1) << s->stream_flags.check;
+
+ return checks;
+}
+
+
+extern uint32_t
+lzma_index_padding_size(const lzma_index *i)
+{
+ return (LZMA_VLI_C(4) - index_size_unpadded(
+ i->record_count, i->index_list_size)) & 3;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_stream_flags(lzma_index *i, const lzma_stream_flags *stream_flags)
+{
+ if (i == NULL || stream_flags == NULL)
+ return LZMA_PROG_ERROR;
+
+ // Validate the Stream Flags.
+ return_if_error(lzma_stream_flags_compare(
+ stream_flags, stream_flags));
+
+ index_stream *s = (index_stream *)(i->streams.rightmost);
+ s->stream_flags = *stream_flags;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_stream_padding(lzma_index *i, lzma_vli stream_padding)
+{
+ if (i == NULL || stream_padding > LZMA_VLI_MAX
+ || (stream_padding & 3) != 0)
+ return LZMA_PROG_ERROR;
+
+ index_stream *s = (index_stream *)(i->streams.rightmost);
+
+ // Check that the new value won't make the file grow too big.
+ const lzma_vli old_stream_padding = s->stream_padding;
+ s->stream_padding = 0;
+ if (lzma_index_file_size(i) + stream_padding > LZMA_VLI_MAX) {
+ s->stream_padding = old_stream_padding;
+ return LZMA_DATA_ERROR;
+ }
+
+ s->stream_padding = stream_padding;
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_append(lzma_index *i, lzma_allocator *allocator,
+ lzma_vli unpadded_size, lzma_vli uncompressed_size)
+{
+ // Validate.
+ if (i == NULL || unpadded_size < UNPADDED_SIZE_MIN
+ || unpadded_size > UNPADDED_SIZE_MAX
+ || uncompressed_size > LZMA_VLI_MAX)
+ return LZMA_PROG_ERROR;
+
+ index_stream *s = (index_stream *)(i->streams.rightmost);
+ index_group *g = (index_group *)(s->groups.rightmost);
+
+ const lzma_vli compressed_base = g == NULL ? 0
+ : vli_ceil4(g->records[g->last].unpadded_sum);
+ const lzma_vli uncompressed_base = g == NULL ? 0
+ : g->records[g->last].uncompressed_sum;
+ const uint32_t index_list_size_add = lzma_vli_size(unpadded_size)
+ + lzma_vli_size(uncompressed_size);
+
+ // Check that the file size will stay within limits.
+ if (index_file_size(s->node.compressed_base,
+ compressed_base + unpadded_size, s->record_count + 1,
+ s->index_list_size + index_list_size_add,
+ s->stream_padding) == LZMA_VLI_UNKNOWN)
+ return LZMA_DATA_ERROR;
+
+ // The size of the Index field must not exceed the maximum value
+ // that can be stored in the Backward Size field.
+ if (index_size(i->record_count + 1,
+ i->index_list_size + index_list_size_add)
+ > LZMA_BACKWARD_SIZE_MAX)
+ return LZMA_DATA_ERROR;
+
+ if (g != NULL && g->last + 1 < g->allocated) {
+ // There is space in the last group at least for one Record.
+ ++g->last;
+ } else {
+ // We need to allocate a new group.
+ g = lzma_alloc(sizeof(index_group)
+ + i->prealloc * sizeof(index_record),
+ allocator);
+ if (g == NULL)
+ return LZMA_MEM_ERROR;
+
+ g->last = 0;
+ g->allocated = i->prealloc;
+
+ // Reset prealloc so that if the application happens to
+ // add new Records, the allocation size will be sane.
+ i->prealloc = INDEX_GROUP_SIZE;
+
+ // Set the start offsets of this group.
+ g->node.uncompressed_base = uncompressed_base;
+ g->node.compressed_base = compressed_base;
+ g->number_base = s->record_count + 1;
+
+ // Add the new group to the Stream.
+ index_tree_append(&s->groups, &g->node);
+ }
+
+ // Add the new Record to the group.
+ g->records[g->last].uncompressed_sum
+ = uncompressed_base + uncompressed_size;
+ g->records[g->last].unpadded_sum
+ = compressed_base + unpadded_size;
+
+ // Update the totals.
+ ++s->record_count;
+ s->index_list_size += index_list_size_add;
+
+ i->total_size += vli_ceil4(unpadded_size);
+ i->uncompressed_size += uncompressed_size;
+ ++i->record_count;
+ i->index_list_size += index_list_size_add;
+
+ return LZMA_OK;
+}
+
+
+/// Structure to pass info to index_cat_helper()
+typedef struct {
+ /// Uncompressed size of the destination
+ lzma_vli uncompressed_size;
+
+ /// Compressed file size of the destination
+ lzma_vli file_size;
+
+ /// Same as above but for Block numbers
+ lzma_vli block_number_add;
+
+ /// Number of Streams that were in the destination index before we
+ /// started appending new Streams from the source index. This is
+ /// used to fix the Stream numbering.
+ uint32_t stream_number_add;
+
+ /// Destination index' Stream tree
+ index_tree *streams;
+
+} index_cat_info;
+
+
+/// Add the Stream nodes from the source index to dest using recursion.
+/// Simplest iterative traversal of the source tree wouldn't work, because
+/// we update the pointers in nodes when moving them to the destination tree.
+static void
+index_cat_helper(const index_cat_info *info, index_stream *this)
+{
+ index_stream *left = (index_stream *)(this->node.left);
+ index_stream *right = (index_stream *)(this->node.right);
+
+ if (left != NULL)
+ index_cat_helper(info, left);
+
+ this->node.uncompressed_base += info->uncompressed_size;
+ this->node.compressed_base += info->file_size;
+ this->number += info->stream_number_add;
+ this->block_number_base += info->block_number_add;
+ index_tree_append(info->streams, &this->node);
+
+ if (right != NULL)
+ index_cat_helper(info, right);
+
+ return;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_cat(lzma_index *restrict dest, lzma_index *restrict src,
+ lzma_allocator *allocator)
+{
+ const lzma_vli dest_file_size = lzma_index_file_size(dest);
+
+ // Check that we don't exceed the file size limits.
+ if (dest_file_size + lzma_index_file_size(src) > LZMA_VLI_MAX
+ || dest->uncompressed_size + src->uncompressed_size
+ > LZMA_VLI_MAX)
+ return LZMA_DATA_ERROR;
+
+ // Check that the encoded size of the combined lzma_indexes stays
+ // within limits. In theory, this should be done only if we know
+ // that the user plans to actually combine the Streams and thus
+ // construct a single Index (probably rare). However, exceeding
+ // this limit is quite theoretical, so we do this check always
+ // to simplify things elsewhere.
+ {
+ const lzma_vli dest_size = index_size_unpadded(
+ dest->record_count, dest->index_list_size);
+ const lzma_vli src_size = index_size_unpadded(
+ src->record_count, src->index_list_size);
+ if (vli_ceil4(dest_size + src_size) > LZMA_BACKWARD_SIZE_MAX)
+ return LZMA_DATA_ERROR;
+ }
+
+ // Optimize the last group to minimize memory usage. Allocation has
+ // to be done before modifying dest or src.
+ {
+ index_stream *s = (index_stream *)(dest->streams.rightmost);
+ index_group *g = (index_group *)(s->groups.rightmost);
+ if (g != NULL && g->last + 1 < g->allocated) {
+ assert(g->node.left == NULL);
+ assert(g->node.right == NULL);
+
+ index_group *newg = lzma_alloc(sizeof(index_group)
+ + (g->last + 1)
+ * sizeof(index_record),
+ allocator);
+ if (newg == NULL)
+ return LZMA_MEM_ERROR;
+
+ newg->node = g->node;
+ newg->allocated = g->last + 1;
+ newg->last = g->last;
+ newg->number_base = g->number_base;
+
+ memcpy(newg->records, g->records, newg->allocated
+ * sizeof(index_record));
+
+ if (g->node.parent != NULL) {
+ assert(g->node.parent->right == &g->node);
+ g->node.parent->right = &newg->node;
+ }
+
+ if (s->groups.leftmost == &g->node) {
+ assert(s->groups.root == &g->node);
+ s->groups.leftmost = &newg->node;
+ s->groups.root = &newg->node;
+ }
+
+ if (s->groups.rightmost == &g->node)
+ s->groups.rightmost = &newg->node;
+
+ lzma_free(g, allocator);
+ }
+ }
+
+ // Add all the Streams from src to dest. Update the base offsets
+ // of each Stream from src.
+ const index_cat_info info = {
+ .uncompressed_size = dest->uncompressed_size,
+ .file_size = dest_file_size,
+ .stream_number_add = dest->streams.count,
+ .block_number_add = dest->record_count,
+ .streams = &dest->streams,
+ };
+ index_cat_helper(&info, (index_stream *)(src->streams.root));
+
+ // Update info about all the combined Streams.
+ dest->uncompressed_size += src->uncompressed_size;
+ dest->total_size += src->total_size;
+ dest->record_count += src->record_count;
+ dest->index_list_size += src->index_list_size;
+ dest->checks = lzma_index_checks(dest) | src->checks;
+
+ // There's nothing else left in src than the base structure.
+ lzma_free(src, allocator);
+
+ return LZMA_OK;
+}
+
+
+/// Duplicate an index_stream.
+static index_stream *
+index_dup_stream(const index_stream *src, lzma_allocator *allocator)
+{
+ // Catch a somewhat theoretical integer overflow.
+ if (src->record_count > PREALLOC_MAX)
+ return NULL;
+
+ // Allocate and initialize a new Stream.
+ index_stream *dest = index_stream_init(src->node.compressed_base,
+ src->node.uncompressed_base, src->number,
+ src->block_number_base, allocator);
+
+ // Return immediately if allocation failed or if there are
+ // no groups to duplicate.
+ if (dest == NULL || src->groups.leftmost == NULL)
+ return dest;
+
+ // Copy the overall information.
+ dest->record_count = src->record_count;
+ dest->index_list_size = src->index_list_size;
+ dest->stream_flags = src->stream_flags;
+ dest->stream_padding = src->stream_padding;
+
+ // Allocate memory for the Records. We put all the Records into
+ // a single group. It's simplest and also tends to make
+ // lzma_index_locate() a little bit faster with very big Indexes.
+ index_group *destg = lzma_alloc(sizeof(index_group)
+ + src->record_count * sizeof(index_record),
+ allocator);
+ if (destg == NULL) {
+ index_stream_end(dest, allocator);
+ return NULL;
+ }
+
+ // Initialize destg.
+ destg->node.uncompressed_base = 0;
+ destg->node.compressed_base = 0;
+ destg->number_base = 1;
+ destg->allocated = src->record_count;
+ destg->last = src->record_count - 1;
+
+ // Go through all the groups in src and copy the Records into destg.
+ const index_group *srcg = (const index_group *)(src->groups.leftmost);
+ size_t i = 0;
+ do {
+ memcpy(destg->records + i, srcg->records,
+ (srcg->last + 1) * sizeof(index_record));
+ i += srcg->last + 1;
+ srcg = index_tree_next(&srcg->node);
+ } while (srcg != NULL);
+
+ assert(i == destg->allocated);
+
+ // Add the group to the new Stream.
+ index_tree_append(&dest->groups, &destg->node);
+
+ return dest;
+}
+
+
+extern LZMA_API(lzma_index *)
+lzma_index_dup(const lzma_index *src, lzma_allocator *allocator)
+{
+ // Allocate the base structure (no initial Stream).
+ lzma_index *dest = index_init_plain(allocator);
+ if (dest == NULL)
+ return NULL;
+
+ // Copy the totals.
+ dest->uncompressed_size = src->uncompressed_size;
+ dest->total_size = src->total_size;
+ dest->record_count = src->record_count;
+ dest->index_list_size = src->index_list_size;
+
+ // Copy the Streams and the groups in them.
+ const index_stream *srcstream
+ = (const index_stream *)(src->streams.leftmost);
+ do {
+ index_stream *deststream = index_dup_stream(
+ srcstream, allocator);
+ if (deststream == NULL) {
+ lzma_index_end(dest, allocator);
+ return NULL;
+ }
+
+ index_tree_append(&dest->streams, &deststream->node);
+
+ srcstream = index_tree_next(&srcstream->node);
+ } while (srcstream != NULL);
+
+ return dest;
+}
+
+
+/// Indexing for lzma_index_iter.internal[]
+enum {
+ ITER_INDEX,
+ ITER_STREAM,
+ ITER_GROUP,
+ ITER_RECORD,
+ ITER_METHOD,
+};
+
+
+/// Values for lzma_index_iter.internal[ITER_METHOD].s
+enum {
+ ITER_METHOD_NORMAL,
+ ITER_METHOD_NEXT,
+ ITER_METHOD_LEFTMOST,
+};
+
+
+static void
+iter_set_info(lzma_index_iter *iter)
+{
+ const lzma_index *i = iter->internal[ITER_INDEX].p;
+ const index_stream *stream = iter->internal[ITER_STREAM].p;
+ const index_group *group = iter->internal[ITER_GROUP].p;
+ const size_t record = iter->internal[ITER_RECORD].s;
+
+ // lzma_index_iter.internal must not contain a pointer to the last
+ // group in the index, because that may be reallocated by
+ // lzma_index_cat().
+ if (group == NULL) {
+ // There are no groups.
+ assert(stream->groups.root == NULL);
+ iter->internal[ITER_METHOD].s = ITER_METHOD_LEFTMOST;
+
+ } else if (i->streams.rightmost != &stream->node
+ || stream->groups.rightmost != &group->node) {
+ // The group is not not the last group in the index.
+ iter->internal[ITER_METHOD].s = ITER_METHOD_NORMAL;
+
+ } else if (stream->groups.leftmost != &group->node) {
+ // The group isn't the only group in the Stream, thus we
+ // know that it must have a parent group i.e. it's not
+ // the root node.
+ assert(stream->groups.root != &group->node);
+ assert(group->node.parent->right == &group->node);
+ iter->internal[ITER_METHOD].s = ITER_METHOD_NEXT;
+ iter->internal[ITER_GROUP].p = group->node.parent;
+
+ } else {
+ // The Stream has only one group.
+ assert(stream->groups.root == &group->node);
+ assert(group->node.parent == NULL);
+ iter->internal[ITER_METHOD].s = ITER_METHOD_LEFTMOST;
+ iter->internal[ITER_GROUP].p = NULL;
+ }
+
+ iter->stream.number = stream->number;
+ iter->stream.block_count = stream->record_count;
+ iter->stream.compressed_offset = stream->node.compressed_base;
+ iter->stream.uncompressed_offset = stream->node.uncompressed_base;
+
+ // iter->stream.flags will be NULL if the Stream Flags haven't been
+ // set with lzma_index_stream_flags().
+ iter->stream.flags = stream->stream_flags.version == UINT32_MAX
+ ? NULL : &stream->stream_flags;
+ iter->stream.padding = stream->stream_padding;
+
+ if (stream->groups.rightmost == NULL) {
+ // Stream has no Blocks.
+ iter->stream.compressed_size = index_size(0, 0)
+ + 2 * LZMA_STREAM_HEADER_SIZE;
+ iter->stream.uncompressed_size = 0;
+ } else {
+ const index_group *g = (const index_group *)(
+ stream->groups.rightmost);
+
+ // Stream Header + Stream Footer + Index + Blocks
+ iter->stream.compressed_size = 2 * LZMA_STREAM_HEADER_SIZE
+ + index_size(stream->record_count,
+ stream->index_list_size)
+ + vli_ceil4(g->records[g->last].unpadded_sum);
+ iter->stream.uncompressed_size
+ = g->records[g->last].uncompressed_sum;
+ }
+
+ if (group != NULL) {
+ iter->block.number_in_stream = group->number_base + record;
+ iter->block.number_in_file = iter->block.number_in_stream
+ + stream->block_number_base;
+
+ iter->block.compressed_stream_offset
+ = record == 0 ? group->node.compressed_base
+ : vli_ceil4(group->records[
+ record - 1].unpadded_sum);
+ iter->block.uncompressed_stream_offset
+ = record == 0 ? group->node.uncompressed_base
+ : group->records[record - 1].uncompressed_sum;
+
+ iter->block.uncompressed_size
+ = group->records[record].uncompressed_sum
+ - iter->block.uncompressed_stream_offset;
+ iter->block.unpadded_size
+ = group->records[record].unpadded_sum
+ - iter->block.compressed_stream_offset;
+ iter->block.total_size = vli_ceil4(iter->block.unpadded_size);
+
+ iter->block.compressed_stream_offset
+ += LZMA_STREAM_HEADER_SIZE;
+
+ iter->block.compressed_file_offset
+ = iter->block.compressed_stream_offset
+ + iter->stream.compressed_offset;
+ iter->block.uncompressed_file_offset
+ = iter->block.uncompressed_stream_offset
+ + iter->stream.uncompressed_offset;
+ }
+
+ return;
+}
+
+
+extern LZMA_API(void)
+lzma_index_iter_init(lzma_index_iter *iter, const lzma_index *i)
+{
+ iter->internal[ITER_INDEX].p = i;
+ lzma_index_iter_rewind(iter);
+ return;
+}
+
+
+extern LZMA_API(void)
+lzma_index_iter_rewind(lzma_index_iter *iter)
+{
+ iter->internal[ITER_STREAM].p = NULL;
+ iter->internal[ITER_GROUP].p = NULL;
+ iter->internal[ITER_RECORD].s = 0;
+ iter->internal[ITER_METHOD].s = ITER_METHOD_NORMAL;
+ return;
+}
+
+
+extern LZMA_API(lzma_bool)
+lzma_index_iter_next(lzma_index_iter *iter, lzma_index_iter_mode mode)
+{
+ // Catch unsupported mode values.
+ if ((unsigned int)(mode) > LZMA_INDEX_ITER_NONEMPTY_BLOCK)
+ return true;
+
+ const lzma_index *i = iter->internal[ITER_INDEX].p;
+ const index_stream *stream = iter->internal[ITER_STREAM].p;
+ const index_group *group = NULL;
+ size_t record = iter->internal[ITER_RECORD].s;
+
+ // If we are being asked for the next Stream, leave group to NULL
+ // so that the rest of the this function thinks that this Stream
+ // has no groups and will thus go to the next Stream.
+ if (mode != LZMA_INDEX_ITER_STREAM) {
+ // Get the pointer to the current group. See iter_set_inf()
+ // for explanation.
+ switch (iter->internal[ITER_METHOD].s) {
+ case ITER_METHOD_NORMAL:
+ group = iter->internal[ITER_GROUP].p;
+ break;
+
+ case ITER_METHOD_NEXT:
+ group = index_tree_next(iter->internal[ITER_GROUP].p);
+ break;
+
+ case ITER_METHOD_LEFTMOST:
+ group = (const index_group *)(
+ stream->groups.leftmost);
+ break;
+ }
+ }
+
+again:
+ if (stream == NULL) {
+ // We at the beginning of the lzma_index.
+ // Locate the first Stream.
+ stream = (const index_stream *)(i->streams.leftmost);
+ if (mode >= LZMA_INDEX_ITER_BLOCK) {
+ // Since we are being asked to return information
+ // about the first a Block, skip Streams that have
+ // no Blocks.
+ while (stream->groups.leftmost == NULL) {
+ stream = index_tree_next(&stream->node);
+ if (stream == NULL)
+ return true;
+ }
+ }
+
+ // Start from the first Record in the Stream.
+ group = (const index_group *)(stream->groups.leftmost);
+ record = 0;
+
+ } else if (group != NULL && record < group->last) {
+ // The next Record is in the same group.
+ ++record;
+
+ } else {
+ // This group has no more Records or this Stream has
+ // no Blocks at all.
+ record = 0;
+
+ // If group is not NULL, this Stream has at least one Block
+ // and thus at least one group. Find the next group.
+ if (group != NULL)
+ group = index_tree_next(&group->node);
+
+ if (group == NULL) {
+ // This Stream has no more Records. Find the next
+ // Stream. If we are being asked to return information
+ // about a Block, we skip empty Streams.
+ do {
+ stream = index_tree_next(&stream->node);
+ if (stream == NULL)
+ return true;
+ } while (mode >= LZMA_INDEX_ITER_BLOCK
+ && stream->groups.leftmost == NULL);
+
+ group = (const index_group *)(
+ stream->groups.leftmost);
+ }
+ }
+
+ if (mode == LZMA_INDEX_ITER_NONEMPTY_BLOCK) {
+ // We need to look for the next Block again if this Block
+ // is empty.
+ if (record == 0) {
+ if (group->node.uncompressed_base
+ == group->records[0].uncompressed_sum)
+ goto again;
+ } else if (group->records[record - 1].uncompressed_sum
+ == group->records[record].uncompressed_sum) {
+ goto again;
+ }
+ }
+
+ iter->internal[ITER_STREAM].p = stream;
+ iter->internal[ITER_GROUP].p = group;
+ iter->internal[ITER_RECORD].s = record;
+
+ iter_set_info(iter);
+
+ return false;
+}
+
+
+extern LZMA_API(lzma_bool)
+lzma_index_iter_locate(lzma_index_iter *iter, lzma_vli target)
+{
+ const lzma_index *i = iter->internal[ITER_INDEX].p;
+
+ // If the target is past the end of the file, return immediately.
+ if (i->uncompressed_size <= target)
+ return true;
+
+ // Locate the Stream containing the target offset.
+ const index_stream *stream = index_tree_locate(&i->streams, target);
+ assert(stream != NULL);
+ target -= stream->node.uncompressed_base;
+
+ // Locate the group containing the target offset.
+ const index_group *group = index_tree_locate(&stream->groups, target);
+ assert(group != NULL);
+
+ // Use binary search to locate the exact Record. It is the first
+ // Record whose uncompressed_sum is greater than target.
+ // This is because we want the rightmost Record that fullfills the
+ // search criterion. It is possible that there are empty Blocks;
+ // we don't want to return them.
+ size_t left = 0;
+ size_t right = group->last;
+
+ while (left < right) {
+ const size_t pos = left + (right - left) / 2;
+ if (group->records[pos].uncompressed_sum <= target)
+ left = pos + 1;
+ else
+ right = pos;
+ }
+
+ iter->internal[ITER_STREAM].p = stream;
+ iter->internal[ITER_GROUP].p = group;
+ iter->internal[ITER_RECORD].s = left;
+
+ iter_set_info(iter);
+
+ return false;
+}
diff --git a/src/liblzma/common/index.h b/src/liblzma/common/index.h
new file mode 100644
index 0000000..64e9724
--- /dev/null
+++ b/src/liblzma/common/index.h
@@ -0,0 +1,73 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file index.h
+/// \brief Handling of Index
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_INDEX_H
+#define LZMA_INDEX_H
+
+#include "common.h"
+
+
+/// Minimum Unpadded Size
+#define UNPADDED_SIZE_MIN LZMA_VLI_C(5)
+
+/// Maximum Unpadded Size
+#define UNPADDED_SIZE_MAX (LZMA_VLI_MAX & ~LZMA_VLI_C(3))
+
+
+/// Get the size of the Index Padding field. This is needed by Index encoder
+/// and decoder, but applications should have no use for this.
+extern uint32_t lzma_index_padding_size(const lzma_index *i);
+
+
+/// Set for how many Records to allocate memory the next time
+/// lzma_index_append() needs to allocate space for a new Record.
+/// This is used only by the Index decoder.
+extern void lzma_index_prealloc(lzma_index *i, lzma_vli records);
+
+
+/// Round the variable-length integer to the next multiple of four.
+static inline lzma_vli
+vli_ceil4(lzma_vli vli)
+{
+ assert(vli <= LZMA_VLI_MAX);
+ return (vli + 3) & ~LZMA_VLI_C(3);
+}
+
+
+/// Calculate the size of the Index field excluding Index Padding
+static inline lzma_vli
+index_size_unpadded(lzma_vli count, lzma_vli index_list_size)
+{
+ // Index Indicator + Number of Records + List of Records + CRC32
+ return 1 + lzma_vli_size(count) + index_list_size + 4;
+}
+
+
+/// Calculate the size of the Index field including Index Padding
+static inline lzma_vli
+index_size(lzma_vli count, lzma_vli index_list_size)
+{
+ return vli_ceil4(index_size_unpadded(count, index_list_size));
+}
+
+
+/// Calculate the total size of the Stream
+static inline lzma_vli
+index_stream_size(lzma_vli blocks_size,
+ lzma_vli count, lzma_vli index_list_size)
+{
+ return LZMA_STREAM_HEADER_SIZE + blocks_size
+ + index_size(count, index_list_size)
+ + LZMA_STREAM_HEADER_SIZE;
+}
+
+#endif
diff --git a/src/liblzma/common/index_decoder.c b/src/liblzma/common/index_decoder.c
new file mode 100644
index 0000000..86a2297
--- /dev/null
+++ b/src/liblzma/common/index_decoder.c
@@ -0,0 +1,343 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file index_decoder.c
+/// \brief Decodes the Index field
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "index.h"
+#include "check.h"
+
+
+struct lzma_coder_s {
+ enum {
+ SEQ_INDICATOR,
+ SEQ_COUNT,
+ SEQ_MEMUSAGE,
+ SEQ_UNPADDED,
+ SEQ_UNCOMPRESSED,
+ SEQ_PADDING_INIT,
+ SEQ_PADDING,
+ SEQ_CRC32,
+ } sequence;
+
+ /// Memory usage limit
+ uint64_t memlimit;
+
+ /// Target Index
+ lzma_index *index;
+
+ /// Pointer give by the application, which is set after
+ /// successful decoding.
+ lzma_index **index_ptr;
+
+ /// Number of Records left to decode.
+ lzma_vli count;
+
+ /// The most recent Unpadded Size field
+ lzma_vli unpadded_size;
+
+ /// The most recent Uncompressed Size field
+ lzma_vli uncompressed_size;
+
+ /// Position in integers
+ size_t pos;
+
+ /// CRC32 of the List of Records field
+ uint32_t crc32;
+};
+
+
+static lzma_ret
+index_decode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out lzma_attribute((unused)),
+ size_t *restrict out_pos lzma_attribute((unused)),
+ size_t out_size lzma_attribute((unused)),
+ lzma_action action lzma_attribute((unused)))
+{
+ // Similar optimization as in index_encoder.c
+ const size_t in_start = *in_pos;
+ lzma_ret ret = LZMA_OK;
+
+ while (*in_pos < in_size)
+ switch (coder->sequence) {
+ case SEQ_INDICATOR:
+ // Return LZMA_DATA_ERROR instead of e.g. LZMA_PROG_ERROR or
+ // LZMA_FORMAT_ERROR, because a typical usage case for Index
+ // decoder is when parsing the Stream backwards. If seeking
+ // backward from the Stream Footer gives us something that
+ // doesn't begin with Index Indicator, the file is considered
+ // corrupt, not "programming error" or "unrecognized file
+ // format". One could argue that the application should
+ // verify the Index Indicator before trying to decode the
+ // Index, but well, I suppose it is simpler this way.
+ if (in[(*in_pos)++] != 0x00)
+ return LZMA_DATA_ERROR;
+
+ coder->sequence = SEQ_COUNT;
+ break;
+
+ case SEQ_COUNT:
+ ret = lzma_vli_decode(&coder->count, &coder->pos,
+ in, in_pos, in_size);
+ if (ret != LZMA_STREAM_END)
+ goto out;
+
+ coder->pos = 0;
+ coder->sequence = SEQ_MEMUSAGE;
+
+ // Fall through
+
+ case SEQ_MEMUSAGE:
+ if (lzma_index_memusage(1, coder->count) > coder->memlimit) {
+ ret = LZMA_MEMLIMIT_ERROR;
+ goto out;
+ }
+
+ // Tell the Index handling code how many Records this
+ // Index has to allow it to allocate memory more efficiently.
+ lzma_index_prealloc(coder->index, coder->count);
+
+ ret = LZMA_OK;
+ coder->sequence = coder->count == 0
+ ? SEQ_PADDING_INIT : SEQ_UNPADDED;
+ break;
+
+ case SEQ_UNPADDED:
+ case SEQ_UNCOMPRESSED: {
+ lzma_vli *size = coder->sequence == SEQ_UNPADDED
+ ? &coder->unpadded_size
+ : &coder->uncompressed_size;
+
+ ret = lzma_vli_decode(size, &coder->pos,
+ in, in_pos, in_size);
+ if (ret != LZMA_STREAM_END)
+ goto out;
+
+ ret = LZMA_OK;
+ coder->pos = 0;
+
+ if (coder->sequence == SEQ_UNPADDED) {
+ // Validate that encoded Unpadded Size isn't too small
+ // or too big.
+ if (coder->unpadded_size < UNPADDED_SIZE_MIN
+ || coder->unpadded_size
+ > UNPADDED_SIZE_MAX)
+ return LZMA_DATA_ERROR;
+
+ coder->sequence = SEQ_UNCOMPRESSED;
+ } else {
+ // Add the decoded Record to the Index.
+ return_if_error(lzma_index_append(
+ coder->index, allocator,
+ coder->unpadded_size,
+ coder->uncompressed_size));
+
+ // Check if this was the last Record.
+ coder->sequence = --coder->count == 0
+ ? SEQ_PADDING_INIT
+ : SEQ_UNPADDED;
+ }
+
+ break;
+ }
+
+ case SEQ_PADDING_INIT:
+ coder->pos = lzma_index_padding_size(coder->index);
+ coder->sequence = SEQ_PADDING;
+
+ // Fall through
+
+ case SEQ_PADDING:
+ if (coder->pos > 0) {
+ --coder->pos;
+ if (in[(*in_pos)++] != 0x00)
+ return LZMA_DATA_ERROR;
+
+ break;
+ }
+
+ // Finish the CRC32 calculation.
+ coder->crc32 = lzma_crc32(in + in_start,
+ *in_pos - in_start, coder->crc32);
+
+ coder->sequence = SEQ_CRC32;
+
+ // Fall through
+
+ case SEQ_CRC32:
+ do {
+ if (*in_pos == in_size)
+ return LZMA_OK;
+
+ if (((coder->crc32 >> (coder->pos * 8)) & 0xFF)
+ != in[(*in_pos)++])
+ return LZMA_DATA_ERROR;
+
+ } while (++coder->pos < 4);
+
+ // Decoding was successful, now we can let the application
+ // see the decoded Index.
+ *coder->index_ptr = coder->index;
+
+ // Make index NULL so we don't free it unintentionally.
+ coder->index = NULL;
+
+ return LZMA_STREAM_END;
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+out:
+ // Update the CRC32,
+ coder->crc32 = lzma_crc32(in + in_start,
+ *in_pos - in_start, coder->crc32);
+
+ return ret;
+}
+
+
+static void
+index_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_index_end(coder->index, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+index_decoder_memconfig(lzma_coder *coder, uint64_t *memusage,
+ uint64_t *old_memlimit, uint64_t new_memlimit)
+{
+ *memusage = lzma_index_memusage(1, coder->count);
+ *old_memlimit = coder->memlimit;
+
+ if (new_memlimit != 0) {
+ if (new_memlimit < *memusage)
+ return LZMA_MEMLIMIT_ERROR;
+
+ coder->memlimit = new_memlimit;
+ }
+
+ return LZMA_OK;
+}
+
+
+static lzma_ret
+index_decoder_reset(lzma_coder *coder, lzma_allocator *allocator,
+ lzma_index **i, uint64_t memlimit)
+{
+ // Remember the pointer given by the application. We will set it
+ // to point to the decoded Index only if decoding is successful.
+ // Before that, keep it NULL so that applications can always safely
+ // pass it to lzma_index_end() no matter did decoding succeed or not.
+ coder->index_ptr = i;
+ *i = NULL;
+
+ // We always allocate a new lzma_index.
+ coder->index = lzma_index_init(allocator);
+ if (coder->index == NULL)
+ return LZMA_MEM_ERROR;
+
+ // Initialize the rest.
+ coder->sequence = SEQ_INDICATOR;
+ coder->memlimit = memlimit;
+ coder->count = 0; // Needs to be initialized due to _memconfig().
+ coder->pos = 0;
+ coder->crc32 = 0;
+
+ return LZMA_OK;
+}
+
+
+static lzma_ret
+index_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ lzma_index **i, uint64_t memlimit)
+{
+ lzma_next_coder_init(&index_decoder_init, next, allocator);
+
+ if (i == NULL || memlimit == 0)
+ return LZMA_PROG_ERROR;
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &index_decode;
+ next->end = &index_decoder_end;
+ next->memconfig = &index_decoder_memconfig;
+ next->coder->index = NULL;
+ } else {
+ lzma_index_end(next->coder->index, allocator);
+ }
+
+ return index_decoder_reset(next->coder, allocator, i, memlimit);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_decoder(lzma_stream *strm, lzma_index **i, uint64_t memlimit)
+{
+ lzma_next_strm_init(index_decoder_init, strm, i, memlimit);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_buffer_decode(
+ lzma_index **i, uint64_t *memlimit, lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size)
+{
+ // Sanity checks
+ if (i == NULL || memlimit == NULL
+ || in == NULL || in_pos == NULL || *in_pos > in_size)
+ return LZMA_PROG_ERROR;
+
+ // Initialize the decoder.
+ lzma_coder coder;
+ return_if_error(index_decoder_reset(&coder, allocator, i, *memlimit));
+
+ // Store the input start position so that we can restore it in case
+ // of an error.
+ const size_t in_start = *in_pos;
+
+ // Do the actual decoding.
+ lzma_ret ret = index_decode(&coder, allocator, in, in_pos, in_size,
+ NULL, NULL, 0, LZMA_RUN);
+
+ if (ret == LZMA_STREAM_END) {
+ ret = LZMA_OK;
+ } else {
+ // Something went wrong, free the Index structure and restore
+ // the input position.
+ lzma_index_end(coder.index, allocator);
+ *in_pos = in_start;
+
+ if (ret == LZMA_OK) {
+ // The input is truncated or otherwise corrupt.
+ // Use LZMA_DATA_ERROR instead of LZMA_BUF_ERROR
+ // like lzma_vli_decode() does in single-call mode.
+ ret = LZMA_DATA_ERROR;
+
+ } else if (ret == LZMA_MEMLIMIT_ERROR) {
+ // Tell the caller how much memory would have
+ // been needed.
+ *memlimit = lzma_index_memusage(1, coder.count);
+ }
+ }
+
+ return ret;
+}
diff --git a/src/liblzma/common/index_encoder.c b/src/liblzma/common/index_encoder.c
new file mode 100644
index 0000000..706f1fd
--- /dev/null
+++ b/src/liblzma/common/index_encoder.c
@@ -0,0 +1,252 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file index_encoder.c
+/// \brief Encodes the Index field
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "index_encoder.h"
+#include "index.h"
+#include "check.h"
+
+
+struct lzma_coder_s {
+ enum {
+ SEQ_INDICATOR,
+ SEQ_COUNT,
+ SEQ_UNPADDED,
+ SEQ_UNCOMPRESSED,
+ SEQ_NEXT,
+ SEQ_PADDING,
+ SEQ_CRC32,
+ } sequence;
+
+ /// Index being encoded
+ const lzma_index *index;
+
+ /// Iterator for the Index being encoded
+ lzma_index_iter iter;
+
+ /// Position in integers
+ size_t pos;
+
+ /// CRC32 of the List of Records field
+ uint32_t crc32;
+};
+
+
+static lzma_ret
+index_encode(lzma_coder *coder,
+ lzma_allocator *allocator lzma_attribute((unused)),
+ const uint8_t *restrict in lzma_attribute((unused)),
+ size_t *restrict in_pos lzma_attribute((unused)),
+ size_t in_size lzma_attribute((unused)),
+ uint8_t *restrict out, size_t *restrict out_pos,
+ size_t out_size, lzma_action action lzma_attribute((unused)))
+{
+ // Position where to start calculating CRC32. The idea is that we
+ // need to call lzma_crc32() only once per call to index_encode().
+ const size_t out_start = *out_pos;
+
+ // Return value to use if we return at the end of this function.
+ // We use "goto out" to jump out of the while-switch construct
+ // instead of returning directly, because that way we don't need
+ // to copypaste the lzma_crc32() call to many places.
+ lzma_ret ret = LZMA_OK;
+
+ while (*out_pos < out_size)
+ switch (coder->sequence) {
+ case SEQ_INDICATOR:
+ out[*out_pos] = 0x00;
+ ++*out_pos;
+ coder->sequence = SEQ_COUNT;
+ break;
+
+ case SEQ_COUNT: {
+ const lzma_vli count = lzma_index_block_count(coder->index);
+ ret = lzma_vli_encode(count, &coder->pos,
+ out, out_pos, out_size);
+ if (ret != LZMA_STREAM_END)
+ goto out;
+
+ ret = LZMA_OK;
+ coder->pos = 0;
+ coder->sequence = SEQ_NEXT;
+ break;
+ }
+
+ case SEQ_NEXT:
+ if (lzma_index_iter_next(
+ &coder->iter, LZMA_INDEX_ITER_BLOCK)) {
+ // Get the size of the Index Padding field.
+ coder->pos = lzma_index_padding_size(coder->index);
+ assert(coder->pos <= 3);
+ coder->sequence = SEQ_PADDING;
+ break;
+ }
+
+ coder->sequence = SEQ_UNPADDED;
+
+ // Fall through
+
+ case SEQ_UNPADDED:
+ case SEQ_UNCOMPRESSED: {
+ const lzma_vli size = coder->sequence == SEQ_UNPADDED
+ ? coder->iter.block.unpadded_size
+ : coder->iter.block.uncompressed_size;
+
+ ret = lzma_vli_encode(size, &coder->pos,
+ out, out_pos, out_size);
+ if (ret != LZMA_STREAM_END)
+ goto out;
+
+ ret = LZMA_OK;
+ coder->pos = 0;
+
+ // Advance to SEQ_UNCOMPRESSED or SEQ_NEXT.
+ ++coder->sequence;
+ break;
+ }
+
+ case SEQ_PADDING:
+ if (coder->pos > 0) {
+ --coder->pos;
+ out[(*out_pos)++] = 0x00;
+ break;
+ }
+
+ // Finish the CRC32 calculation.
+ coder->crc32 = lzma_crc32(out + out_start,
+ *out_pos - out_start, coder->crc32);
+
+ coder->sequence = SEQ_CRC32;
+
+ // Fall through
+
+ case SEQ_CRC32:
+ // We don't use the main loop, because we don't want
+ // coder->crc32 to be touched anymore.
+ do {
+ if (*out_pos == out_size)
+ return LZMA_OK;
+
+ out[*out_pos] = (coder->crc32 >> (coder->pos * 8))
+ & 0xFF;
+ ++*out_pos;
+
+ } while (++coder->pos < 4);
+
+ return LZMA_STREAM_END;
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+out:
+ // Update the CRC32.
+ coder->crc32 = lzma_crc32(out + out_start,
+ *out_pos - out_start, coder->crc32);
+
+ return ret;
+}
+
+
+static void
+index_encoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static void
+index_encoder_reset(lzma_coder *coder, const lzma_index *i)
+{
+ lzma_index_iter_init(&coder->iter, i);
+
+ coder->sequence = SEQ_INDICATOR;
+ coder->index = i;
+ coder->pos = 0;
+ coder->crc32 = 0;
+
+ return;
+}
+
+
+extern lzma_ret
+lzma_index_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_index *i)
+{
+ lzma_next_coder_init(&lzma_index_encoder_init, next, allocator);
+
+ if (i == NULL)
+ return LZMA_PROG_ERROR;
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &index_encode;
+ next->end = &index_encoder_end;
+ }
+
+ index_encoder_reset(next->coder, i);
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_encoder(lzma_stream *strm, const lzma_index *i)
+{
+ lzma_next_strm_init(lzma_index_encoder_init, strm, i);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_buffer_encode(const lzma_index *i,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // Validate the arguments.
+ if (i == NULL || out == NULL || out_pos == NULL || *out_pos > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Don't try to encode if there's not enough output space.
+ if (out_size - *out_pos < lzma_index_size(i))
+ return LZMA_BUF_ERROR;
+
+ // The Index encoder needs just one small data structure so we can
+ // allocate it on stack.
+ lzma_coder coder;
+ index_encoder_reset(&coder, i);
+
+ // Do the actual encoding. This should never fail, but store
+ // the original *out_pos just in case.
+ const size_t out_start = *out_pos;
+ lzma_ret ret = index_encode(&coder, NULL, NULL, NULL, 0,
+ out, out_pos, out_size, LZMA_RUN);
+
+ if (ret == LZMA_STREAM_END) {
+ ret = LZMA_OK;
+ } else {
+ // We should never get here, but just in case, restore the
+ // output position and set the error accordingly if something
+ // goes wrong and debugging isn't enabled.
+ assert(0);
+ *out_pos = out_start;
+ ret = LZMA_PROG_ERROR;
+ }
+
+ return ret;
+}
diff --git a/src/liblzma/common/index_encoder.h b/src/liblzma/common/index_encoder.h
new file mode 100644
index 0000000..a13c94d
--- /dev/null
+++ b/src/liblzma/common/index_encoder.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file index_encoder.h
+/// \brief Encodes the Index field
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_INDEX_ENCODER_H
+#define LZMA_INDEX_ENCODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_index_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_index *i);
+
+
+#endif
diff --git a/src/liblzma/common/index_hash.c b/src/liblzma/common/index_hash.c
new file mode 100644
index 0000000..e3e9386
--- /dev/null
+++ b/src/liblzma/common/index_hash.c
@@ -0,0 +1,332 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file index_hash.c
+/// \brief Validates Index by using a hash function
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+#include "index.h"
+#include "check.h"
+
+
+typedef struct {
+ /// Sum of the Block sizes (including Block Padding)
+ lzma_vli blocks_size;
+
+ /// Sum of the Uncompressed Size fields
+ lzma_vli uncompressed_size;
+
+ /// Number of Records
+ lzma_vli count;
+
+ /// Size of the List of Index Records as bytes
+ lzma_vli index_list_size;
+
+ /// Check calculated from Unpadded Sizes and Uncompressed Sizes.
+ lzma_check_state check;
+
+} lzma_index_hash_info;
+
+
+struct lzma_index_hash_s {
+ enum {
+ SEQ_BLOCK,
+ SEQ_COUNT,
+ SEQ_UNPADDED,
+ SEQ_UNCOMPRESSED,
+ SEQ_PADDING_INIT,
+ SEQ_PADDING,
+ SEQ_CRC32,
+ } sequence;
+
+ /// Information collected while decoding the actual Blocks.
+ lzma_index_hash_info blocks;
+
+ /// Information collected from the Index field.
+ lzma_index_hash_info records;
+
+ /// Number of Records not fully decoded
+ lzma_vli remaining;
+
+ /// Unpadded Size currently being read from an Index Record.
+ lzma_vli unpadded_size;
+
+ /// Uncompressed Size currently being read from an Index Record.
+ lzma_vli uncompressed_size;
+
+ /// Position in variable-length integers when decoding them from
+ /// the List of Records.
+ size_t pos;
+
+ /// CRC32 of the Index
+ uint32_t crc32;
+};
+
+
+extern LZMA_API(lzma_index_hash *)
+lzma_index_hash_init(lzma_index_hash *index_hash, lzma_allocator *allocator)
+{
+ if (index_hash == NULL) {
+ index_hash = lzma_alloc(sizeof(lzma_index_hash), allocator);
+ if (index_hash == NULL)
+ return NULL;
+ }
+
+ index_hash->sequence = SEQ_BLOCK;
+ index_hash->blocks.blocks_size = 0;
+ index_hash->blocks.uncompressed_size = 0;
+ index_hash->blocks.count = 0;
+ index_hash->blocks.index_list_size = 0;
+ index_hash->records.blocks_size = 0;
+ index_hash->records.uncompressed_size = 0;
+ index_hash->records.count = 0;
+ index_hash->records.index_list_size = 0;
+ index_hash->unpadded_size = 0;
+ index_hash->uncompressed_size = 0;
+ index_hash->pos = 0;
+ index_hash->crc32 = 0;
+
+ // These cannot fail because LZMA_CHECK_BEST is known to be supported.
+ (void)lzma_check_init(&index_hash->blocks.check, LZMA_CHECK_BEST);
+ (void)lzma_check_init(&index_hash->records.check, LZMA_CHECK_BEST);
+
+ return index_hash;
+}
+
+
+extern LZMA_API(void)
+lzma_index_hash_end(lzma_index_hash *index_hash, lzma_allocator *allocator)
+{
+ lzma_free(index_hash, allocator);
+ return;
+}
+
+
+extern LZMA_API(lzma_vli)
+lzma_index_hash_size(const lzma_index_hash *index_hash)
+{
+ // Get the size of the Index from ->blocks instead of ->records for
+ // cases where application wants to know the Index Size before
+ // decoding the Index.
+ return index_size(index_hash->blocks.count,
+ index_hash->blocks.index_list_size);
+}
+
+
+/// Updates the sizes and the hash without any validation.
+static lzma_ret
+hash_append(lzma_index_hash_info *info, lzma_vli unpadded_size,
+ lzma_vli uncompressed_size)
+{
+ info->blocks_size += vli_ceil4(unpadded_size);
+ info->uncompressed_size += uncompressed_size;
+ info->index_list_size += lzma_vli_size(unpadded_size)
+ + lzma_vli_size(uncompressed_size);
+ ++info->count;
+
+ const lzma_vli sizes[2] = { unpadded_size, uncompressed_size };
+ lzma_check_update(&info->check, LZMA_CHECK_BEST,
+ (const uint8_t *)(sizes), sizeof(sizes));
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_hash_append(lzma_index_hash *index_hash, lzma_vli unpadded_size,
+ lzma_vli uncompressed_size)
+{
+ // Validate the arguments.
+ if (index_hash->sequence != SEQ_BLOCK
+ || unpadded_size < UNPADDED_SIZE_MIN
+ || unpadded_size > UNPADDED_SIZE_MAX
+ || uncompressed_size > LZMA_VLI_MAX)
+ return LZMA_PROG_ERROR;
+
+ // Update the hash.
+ return_if_error(hash_append(&index_hash->blocks,
+ unpadded_size, uncompressed_size));
+
+ // Validate the properties of *info are still in allowed limits.
+ if (index_hash->blocks.blocks_size > LZMA_VLI_MAX
+ || index_hash->blocks.uncompressed_size > LZMA_VLI_MAX
+ || index_size(index_hash->blocks.count,
+ index_hash->blocks.index_list_size)
+ > LZMA_BACKWARD_SIZE_MAX
+ || index_stream_size(index_hash->blocks.blocks_size,
+ index_hash->blocks.count,
+ index_hash->blocks.index_list_size)
+ > LZMA_VLI_MAX)
+ return LZMA_DATA_ERROR;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_index_hash_decode(lzma_index_hash *index_hash, const uint8_t *in,
+ size_t *in_pos, size_t in_size)
+{
+ // Catch zero input buffer here, because in contrast to Index encoder
+ // and decoder functions, applications call this function directly
+ // instead of via lzma_code(), which does the buffer checking.
+ if (*in_pos >= in_size)
+ return LZMA_BUF_ERROR;
+
+ // NOTE: This function has many similarities to index_encode() and
+ // index_decode() functions found from index_encoder.c and
+ // index_decoder.c. See the comments especially in index_encoder.c.
+ const size_t in_start = *in_pos;
+ lzma_ret ret = LZMA_OK;
+
+ while (*in_pos < in_size)
+ switch (index_hash->sequence) {
+ case SEQ_BLOCK:
+ // Check the Index Indicator is present.
+ if (in[(*in_pos)++] != 0x00)
+ return LZMA_DATA_ERROR;
+
+ index_hash->sequence = SEQ_COUNT;
+ break;
+
+ case SEQ_COUNT: {
+ ret = lzma_vli_decode(&index_hash->remaining,
+ &index_hash->pos, in, in_pos, in_size);
+ if (ret != LZMA_STREAM_END)
+ goto out;
+
+ // The count must match the count of the Blocks decoded.
+ if (index_hash->remaining != index_hash->blocks.count)
+ return LZMA_DATA_ERROR;
+
+ ret = LZMA_OK;
+ index_hash->pos = 0;
+
+ // Handle the special case when there are no Blocks.
+ index_hash->sequence = index_hash->remaining == 0
+ ? SEQ_PADDING_INIT : SEQ_UNPADDED;
+ break;
+ }
+
+ case SEQ_UNPADDED:
+ case SEQ_UNCOMPRESSED: {
+ lzma_vli *size = index_hash->sequence == SEQ_UNPADDED
+ ? &index_hash->unpadded_size
+ : &index_hash->uncompressed_size;
+
+ ret = lzma_vli_decode(size, &index_hash->pos,
+ in, in_pos, in_size);
+ if (ret != LZMA_STREAM_END)
+ goto out;
+
+ ret = LZMA_OK;
+ index_hash->pos = 0;
+
+ if (index_hash->sequence == SEQ_UNPADDED) {
+ if (index_hash->unpadded_size < UNPADDED_SIZE_MIN
+ || index_hash->unpadded_size
+ > UNPADDED_SIZE_MAX)
+ return LZMA_DATA_ERROR;
+
+ index_hash->sequence = SEQ_UNCOMPRESSED;
+ } else {
+ // Update the hash.
+ return_if_error(hash_append(&index_hash->records,
+ index_hash->unpadded_size,
+ index_hash->uncompressed_size));
+
+ // Verify that we don't go over the known sizes. Note
+ // that this validation is simpler than the one used
+ // in lzma_index_hash_append(), because here we know
+ // that values in index_hash->blocks are already
+ // validated and we are fine as long as we don't
+ // exceed them in index_hash->records.
+ if (index_hash->blocks.blocks_size
+ < index_hash->records.blocks_size
+ || index_hash->blocks.uncompressed_size
+ < index_hash->records.uncompressed_size
+ || index_hash->blocks.index_list_size
+ < index_hash->records.index_list_size)
+ return LZMA_DATA_ERROR;
+
+ // Check if this was the last Record.
+ index_hash->sequence = --index_hash->remaining == 0
+ ? SEQ_PADDING_INIT : SEQ_UNPADDED;
+ }
+
+ break;
+ }
+
+ case SEQ_PADDING_INIT:
+ index_hash->pos = (LZMA_VLI_C(4) - index_size_unpadded(
+ index_hash->records.count,
+ index_hash->records.index_list_size)) & 3;
+ index_hash->sequence = SEQ_PADDING;
+
+ // Fall through
+
+ case SEQ_PADDING:
+ if (index_hash->pos > 0) {
+ --index_hash->pos;
+ if (in[(*in_pos)++] != 0x00)
+ return LZMA_DATA_ERROR;
+
+ break;
+ }
+
+ // Compare the sizes.
+ if (index_hash->blocks.blocks_size
+ != index_hash->records.blocks_size
+ || index_hash->blocks.uncompressed_size
+ != index_hash->records.uncompressed_size
+ || index_hash->blocks.index_list_size
+ != index_hash->records.index_list_size)
+ return LZMA_DATA_ERROR;
+
+ // Finish the hashes and compare them.
+ lzma_check_finish(&index_hash->blocks.check, LZMA_CHECK_BEST);
+ lzma_check_finish(&index_hash->records.check, LZMA_CHECK_BEST);
+ if (memcmp(index_hash->blocks.check.buffer.u8,
+ index_hash->records.check.buffer.u8,
+ lzma_check_size(LZMA_CHECK_BEST)) != 0)
+ return LZMA_DATA_ERROR;
+
+ // Finish the CRC32 calculation.
+ index_hash->crc32 = lzma_crc32(in + in_start,
+ *in_pos - in_start, index_hash->crc32);
+
+ index_hash->sequence = SEQ_CRC32;
+
+ // Fall through
+
+ case SEQ_CRC32:
+ do {
+ if (*in_pos == in_size)
+ return LZMA_OK;
+
+ if (((index_hash->crc32 >> (index_hash->pos * 8))
+ & 0xFF) != in[(*in_pos)++])
+ return LZMA_DATA_ERROR;
+
+ } while (++index_hash->pos < 4);
+
+ return LZMA_STREAM_END;
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+out:
+ // Update the CRC32,
+ index_hash->crc32 = lzma_crc32(in + in_start,
+ *in_pos - in_start, index_hash->crc32);
+
+ return ret;
+}
diff --git a/src/liblzma/common/stream_buffer_decoder.c b/src/liblzma/common/stream_buffer_decoder.c
new file mode 100644
index 0000000..ae75315
--- /dev/null
+++ b/src/liblzma/common/stream_buffer_decoder.c
@@ -0,0 +1,91 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_buffer_decoder.c
+/// \brief Single-call .xz Stream decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_decoder.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_buffer_decode(uint64_t *memlimit, uint32_t flags,
+ lzma_allocator *allocator,
+ const uint8_t *in, size_t *in_pos, size_t in_size,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ // Sanity checks
+ if (in_pos == NULL || (in == NULL && *in_pos != in_size)
+ || *in_pos > in_size || out_pos == NULL
+ || (out == NULL && *out_pos != out_size)
+ || *out_pos > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Catch flags that are not allowed in buffer-to-buffer decoding.
+ if (flags & LZMA_TELL_ANY_CHECK)
+ return LZMA_PROG_ERROR;
+
+ // Initialize the Stream decoder.
+ // TODO: We need something to tell the decoder that it can use the
+ // output buffer as workspace, and thus save significant amount of RAM.
+ lzma_next_coder stream_decoder = LZMA_NEXT_CODER_INIT;
+ lzma_ret ret = lzma_stream_decoder_init(
+ &stream_decoder, allocator, *memlimit, flags);
+
+ if (ret == LZMA_OK) {
+ // Save the positions so that we can restore them in case
+ // an error occurs.
+ const size_t in_start = *in_pos;
+ const size_t out_start = *out_pos;
+
+ // Do the actual decoding.
+ ret = stream_decoder.code(stream_decoder.coder, allocator,
+ in, in_pos, in_size, out, out_pos, out_size,
+ LZMA_FINISH);
+
+ if (ret == LZMA_STREAM_END) {
+ ret = LZMA_OK;
+ } else {
+ // Something went wrong, restore the positions.
+ *in_pos = in_start;
+ *out_pos = out_start;
+
+ if (ret == LZMA_OK) {
+ // Either the input was truncated or the
+ // output buffer was too small.
+ assert(*in_pos == in_size
+ || *out_pos == out_size);
+
+ // If all the input was consumed, then the
+ // input is truncated, even if the output
+ // buffer is also full. This is because
+ // processing the last byte of the Stream
+ // never produces output.
+ if (*in_pos == in_size)
+ ret = LZMA_DATA_ERROR;
+ else
+ ret = LZMA_BUF_ERROR;
+
+ } else if (ret == LZMA_MEMLIMIT_ERROR) {
+ // Let the caller know how much memory would
+ // have been needed.
+ uint64_t memusage;
+ (void)stream_decoder.memconfig(
+ stream_decoder.coder,
+ memlimit, &memusage, 0);
+ }
+ }
+ }
+
+ // Free the decoder memory. This needs to be done even if
+ // initialization fails, because the internal API doesn't
+ // require the initialization function to free its memory on error.
+ lzma_next_end(&stream_decoder, allocator);
+
+ return ret;
+}
diff --git a/src/liblzma/common/stream_buffer_encoder.c b/src/liblzma/common/stream_buffer_encoder.c
new file mode 100644
index 0000000..f727d85
--- /dev/null
+++ b/src/liblzma/common/stream_buffer_encoder.c
@@ -0,0 +1,131 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_buffer_encoder.c
+/// \brief Single-call .xz Stream encoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "index.h"
+
+
+/// Maximum size of Index that has exactly one Record.
+/// Index Indicator + Number of Records + Record + CRC32 rounded up to
+/// the next multiple of four.
+#define INDEX_BOUND ((1 + 1 + 2 * LZMA_VLI_BYTES_MAX + 4 + 3) & ~3)
+
+/// Stream Header, Stream Footer, and Index
+#define HEADERS_BOUND (2 * LZMA_STREAM_HEADER_SIZE + INDEX_BOUND)
+
+
+extern LZMA_API(size_t)
+lzma_stream_buffer_bound(size_t uncompressed_size)
+{
+ // Get the maximum possible size of a Block.
+ const size_t block_bound = lzma_block_buffer_bound(uncompressed_size);
+ if (block_bound == 0)
+ return 0;
+
+ // Catch the possible integer overflow and also prevent the size of
+ // the Stream exceeding LZMA_VLI_MAX (theoretically possible on
+ // 64-bit systems).
+ if (my_min(SIZE_MAX, LZMA_VLI_MAX) - block_bound < HEADERS_BOUND)
+ return 0;
+
+ return block_bound + HEADERS_BOUND;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_buffer_encode(lzma_filter *filters, lzma_check check,
+ lzma_allocator *allocator, const uint8_t *in, size_t in_size,
+ uint8_t *out, size_t *out_pos_ptr, size_t out_size)
+{
+ // Sanity checks
+ if (filters == NULL || (unsigned int)(check) > LZMA_CHECK_ID_MAX
+ || (in == NULL && in_size != 0) || out == NULL
+ || out_pos_ptr == NULL || *out_pos_ptr > out_size)
+ return LZMA_PROG_ERROR;
+
+ // Note for the paranoids: Index encoder prevents the Stream from
+ // getting too big and still being accepted with LZMA_OK, and Block
+ // encoder catches if the input is too big. So we don't need to
+ // separately check if the buffers are too big.
+
+ // Use a local copy. We update *out_pos_ptr only if everything
+ // succeeds.
+ size_t out_pos = *out_pos_ptr;
+
+ // Check that there's enough space for both Stream Header and
+ // Stream Footer.
+ if (out_size - out_pos <= 2 * LZMA_STREAM_HEADER_SIZE)
+ return LZMA_BUF_ERROR;
+
+ // Reserve space for Stream Footer so we don't need to check for
+ // available space again before encoding Stream Footer.
+ out_size -= LZMA_STREAM_HEADER_SIZE;
+
+ // Encode the Stream Header.
+ lzma_stream_flags stream_flags = {
+ .version = 0,
+ .check = check,
+ };
+
+ if (lzma_stream_header_encode(&stream_flags, out + out_pos)
+ != LZMA_OK)
+ return LZMA_PROG_ERROR;
+
+ out_pos += LZMA_STREAM_HEADER_SIZE;
+
+ // Block
+ lzma_block block = {
+ .version = 0,
+ .check = check,
+ .filters = filters,
+ };
+
+ return_if_error(lzma_block_buffer_encode(&block, allocator,
+ in, in_size, out, &out_pos, out_size));
+
+ // Index
+ {
+ // Create an Index with one Record.
+ lzma_index *i = lzma_index_init(allocator);
+ if (i == NULL)
+ return LZMA_MEM_ERROR;
+
+ lzma_ret ret = lzma_index_append(i, allocator,
+ lzma_block_unpadded_size(&block),
+ block.uncompressed_size);
+
+ // If adding the Record was successful, encode the Index
+ // and get its size which will be stored into Stream Footer.
+ if (ret == LZMA_OK) {
+ ret = lzma_index_buffer_encode(
+ i, out, &out_pos, out_size);
+
+ stream_flags.backward_size = lzma_index_size(i);
+ }
+
+ lzma_index_end(i, allocator);
+
+ if (ret != LZMA_OK)
+ return ret;
+ }
+
+ // Stream Footer. We have already reserved space for this.
+ if (lzma_stream_footer_encode(&stream_flags, out + out_pos)
+ != LZMA_OK)
+ return LZMA_PROG_ERROR;
+
+ out_pos += LZMA_STREAM_HEADER_SIZE;
+
+ // Everything went fine, make the new output position available
+ // to the application.
+ *out_pos_ptr = out_pos;
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/stream_decoder.c b/src/liblzma/common/stream_decoder.c
new file mode 100644
index 0000000..37ea71e
--- /dev/null
+++ b/src/liblzma/common/stream_decoder.c
@@ -0,0 +1,451 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_decoder.c
+/// \brief Decodes .xz Streams
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_decoder.h"
+#include "block_decoder.h"
+
+
+struct lzma_coder_s {
+ enum {
+ SEQ_STREAM_HEADER,
+ SEQ_BLOCK_HEADER,
+ SEQ_BLOCK,
+ SEQ_INDEX,
+ SEQ_STREAM_FOOTER,
+ SEQ_STREAM_PADDING,
+ } sequence;
+
+ /// Block or Metadata decoder. This takes little memory and the same
+ /// data structure can be used to decode every Block Header, so it's
+ /// a good idea to have a separate lzma_next_coder structure for it.
+ lzma_next_coder block_decoder;
+
+ /// Block options decoded by the Block Header decoder and used by
+ /// the Block decoder.
+ lzma_block block_options;
+
+ /// Stream Flags from Stream Header
+ lzma_stream_flags stream_flags;
+
+ /// Index is hashed so that it can be compared to the sizes of Blocks
+ /// with O(1) memory usage.
+ lzma_index_hash *index_hash;
+
+ /// Memory usage limit
+ uint64_t memlimit;
+
+ /// Amount of memory actually needed (only an estimate)
+ uint64_t memusage;
+
+ /// If true, LZMA_NO_CHECK is returned if the Stream has
+ /// no integrity check.
+ bool tell_no_check;
+
+ /// If true, LZMA_UNSUPPORTED_CHECK is returned if the Stream has
+ /// an integrity check that isn't supported by this liblzma build.
+ bool tell_unsupported_check;
+
+ /// If true, LZMA_GET_CHECK is returned after decoding Stream Header.
+ bool tell_any_check;
+
+ /// If true, we will decode concatenated Streams that possibly have
+ /// Stream Padding between or after them. LZMA_STREAM_END is returned
+ /// once the application isn't giving us any new input, and we aren't
+ /// in the middle of a Stream, and possible Stream Padding is a
+ /// multiple of four bytes.
+ bool concatenated;
+
+ /// When decoding concatenated Streams, this is true as long as we
+ /// are decoding the first Stream. This is needed to avoid misleading
+ /// LZMA_FORMAT_ERROR in case the later Streams don't have valid magic
+ /// bytes.
+ bool first_stream;
+
+ /// Write position in buffer[] and position in Stream Padding
+ size_t pos;
+
+ /// Buffer to hold Stream Header, Block Header, and Stream Footer.
+ /// Block Header has biggest maximum size.
+ uint8_t buffer[LZMA_BLOCK_HEADER_SIZE_MAX];
+};
+
+
+static lzma_ret
+stream_decoder_reset(lzma_coder *coder, lzma_allocator *allocator)
+{
+ // Initialize the Index hash used to verify the Index.
+ coder->index_hash = lzma_index_hash_init(coder->index_hash, allocator);
+ if (coder->index_hash == NULL)
+ return LZMA_MEM_ERROR;
+
+ // Reset the rest of the variables.
+ coder->sequence = SEQ_STREAM_HEADER;
+ coder->pos = 0;
+
+ return LZMA_OK;
+}
+
+
+static lzma_ret
+stream_decode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ // When decoding the actual Block, it may be able to produce more
+ // output even if we don't give it any new input.
+ while (true)
+ switch (coder->sequence) {
+ case SEQ_STREAM_HEADER: {
+ // Copy the Stream Header to the internal buffer.
+ lzma_bufcpy(in, in_pos, in_size, coder->buffer, &coder->pos,
+ LZMA_STREAM_HEADER_SIZE);
+
+ // Return if we didn't get the whole Stream Header yet.
+ if (coder->pos < LZMA_STREAM_HEADER_SIZE)
+ return LZMA_OK;
+
+ coder->pos = 0;
+
+ // Decode the Stream Header.
+ const lzma_ret ret = lzma_stream_header_decode(
+ &coder->stream_flags, coder->buffer);
+ if (ret != LZMA_OK)
+ return ret == LZMA_FORMAT_ERROR && !coder->first_stream
+ ? LZMA_DATA_ERROR : ret;
+
+ // If we are decoding concatenated Streams, and the later
+ // Streams have invalid Header Magic Bytes, we give
+ // LZMA_DATA_ERROR instead of LZMA_FORMAT_ERROR.
+ coder->first_stream = false;
+
+ // Copy the type of the Check so that Block Header and Block
+ // decoders see it.
+ coder->block_options.check = coder->stream_flags.check;
+
+ // Even if we return LZMA_*_CHECK below, we want
+ // to continue from Block Header decoding.
+ coder->sequence = SEQ_BLOCK_HEADER;
+
+ // Detect if there's no integrity check or if it is
+ // unsupported if those were requested by the application.
+ if (coder->tell_no_check && coder->stream_flags.check
+ == LZMA_CHECK_NONE)
+ return LZMA_NO_CHECK;
+
+ if (coder->tell_unsupported_check
+ && !lzma_check_is_supported(
+ coder->stream_flags.check))
+ return LZMA_UNSUPPORTED_CHECK;
+
+ if (coder->tell_any_check)
+ return LZMA_GET_CHECK;
+ }
+
+ // Fall through
+
+ case SEQ_BLOCK_HEADER: {
+ if (*in_pos >= in_size)
+ return LZMA_OK;
+
+ if (coder->pos == 0) {
+ // Detect if it's Index.
+ if (in[*in_pos] == 0x00) {
+ coder->sequence = SEQ_INDEX;
+ break;
+ }
+
+ // Calculate the size of the Block Header. Note that
+ // Block Header decoder wants to see this byte too
+ // so don't advance *in_pos.
+ coder->block_options.header_size
+ = lzma_block_header_size_decode(
+ in[*in_pos]);
+ }
+
+ // Copy the Block Header to the internal buffer.
+ lzma_bufcpy(in, in_pos, in_size, coder->buffer, &coder->pos,
+ coder->block_options.header_size);
+
+ // Return if we didn't get the whole Block Header yet.
+ if (coder->pos < coder->block_options.header_size)
+ return LZMA_OK;
+
+ coder->pos = 0;
+
+ // Version 0 is currently the only possible version.
+ coder->block_options.version = 0;
+
+ // Set up a buffer to hold the filter chain. Block Header
+ // decoder will initialize all members of this array so
+ // we don't need to do it here.
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+ coder->block_options.filters = filters;
+
+ // Decode the Block Header.
+ return_if_error(lzma_block_header_decode(&coder->block_options,
+ allocator, coder->buffer));
+
+ // Check the memory usage limit.
+ const uint64_t memusage = lzma_raw_decoder_memusage(filters);
+ lzma_ret ret;
+
+ if (memusage == UINT64_MAX) {
+ // One or more unknown Filter IDs.
+ ret = LZMA_OPTIONS_ERROR;
+ } else {
+ // Now we can set coder->memusage since we know that
+ // the filter chain is valid. We don't want
+ // lzma_memusage() to return UINT64_MAX in case of
+ // invalid filter chain.
+ coder->memusage = memusage;
+
+ if (memusage > coder->memlimit) {
+ // The chain would need too much memory.
+ ret = LZMA_MEMLIMIT_ERROR;
+ } else {
+ // Memory usage is OK.
+ // Initialize the Block decoder.
+ ret = lzma_block_decoder_init(
+ &coder->block_decoder,
+ allocator,
+ &coder->block_options);
+ }
+ }
+
+ // Free the allocated filter options since they are needed
+ // only to initialize the Block decoder.
+ for (size_t i = 0; i < LZMA_FILTERS_MAX; ++i)
+ lzma_free(filters[i].options, allocator);
+
+ coder->block_options.filters = NULL;
+
+ // Check if memory usage calculation and Block enocoder
+ // initialization succeeded.
+ if (ret != LZMA_OK)
+ return ret;
+
+ coder->sequence = SEQ_BLOCK;
+ }
+
+ // Fall through
+
+ case SEQ_BLOCK: {
+ const lzma_ret ret = coder->block_decoder.code(
+ coder->block_decoder.coder, allocator,
+ in, in_pos, in_size, out, out_pos, out_size,
+ action);
+
+ if (ret != LZMA_STREAM_END)
+ return ret;
+
+ // Block decoded successfully. Add the new size pair to
+ // the Index hash.
+ return_if_error(lzma_index_hash_append(coder->index_hash,
+ lzma_block_unpadded_size(
+ &coder->block_options),
+ coder->block_options.uncompressed_size));
+
+ coder->sequence = SEQ_BLOCK_HEADER;
+ break;
+ }
+
+ case SEQ_INDEX: {
+ // If we don't have any input, don't call
+ // lzma_index_hash_decode() since it would return
+ // LZMA_BUF_ERROR, which we must not do here.
+ if (*in_pos >= in_size)
+ return LZMA_OK;
+
+ // Decode the Index and compare it to the hash calculated
+ // from the sizes of the Blocks (if any).
+ const lzma_ret ret = lzma_index_hash_decode(coder->index_hash,
+ in, in_pos, in_size);
+ if (ret != LZMA_STREAM_END)
+ return ret;
+
+ coder->sequence = SEQ_STREAM_FOOTER;
+ }
+
+ // Fall through
+
+ case SEQ_STREAM_FOOTER: {
+ // Copy the Stream Footer to the internal buffer.
+ lzma_bufcpy(in, in_pos, in_size, coder->buffer, &coder->pos,
+ LZMA_STREAM_HEADER_SIZE);
+
+ // Return if we didn't get the whole Stream Footer yet.
+ if (coder->pos < LZMA_STREAM_HEADER_SIZE)
+ return LZMA_OK;
+
+ coder->pos = 0;
+
+ // Decode the Stream Footer. The decoder gives
+ // LZMA_FORMAT_ERROR if the magic bytes don't match,
+ // so convert that return code to LZMA_DATA_ERROR.
+ lzma_stream_flags footer_flags;
+ const lzma_ret ret = lzma_stream_footer_decode(
+ &footer_flags, coder->buffer);
+ if (ret != LZMA_OK)
+ return ret == LZMA_FORMAT_ERROR
+ ? LZMA_DATA_ERROR : ret;
+
+ // Check that Index Size stored in the Stream Footer matches
+ // the real size of the Index field.
+ if (lzma_index_hash_size(coder->index_hash)
+ != footer_flags.backward_size)
+ return LZMA_DATA_ERROR;
+
+ // Compare that the Stream Flags fields are identical in
+ // both Stream Header and Stream Footer.
+ return_if_error(lzma_stream_flags_compare(
+ &coder->stream_flags, &footer_flags));
+
+ if (!coder->concatenated)
+ return LZMA_STREAM_END;
+
+ coder->sequence = SEQ_STREAM_PADDING;
+ }
+
+ // Fall through
+
+ case SEQ_STREAM_PADDING:
+ assert(coder->concatenated);
+
+ // Skip over possible Stream Padding.
+ while (true) {
+ if (*in_pos >= in_size) {
+ // Unless LZMA_FINISH was used, we cannot
+ // know if there's more input coming later.
+ if (action != LZMA_FINISH)
+ return LZMA_OK;
+
+ // Stream Padding must be a multiple of
+ // four bytes.
+ return coder->pos == 0
+ ? LZMA_STREAM_END
+ : LZMA_DATA_ERROR;
+ }
+
+ // If the byte is not zero, it probably indicates
+ // beginning of a new Stream (or the file is corrupt).
+ if (in[*in_pos] != 0x00)
+ break;
+
+ ++*in_pos;
+ coder->pos = (coder->pos + 1) & 3;
+ }
+
+ // Stream Padding must be a multiple of four bytes (empty
+ // Stream Padding is OK).
+ if (coder->pos != 0) {
+ ++*in_pos;
+ return LZMA_DATA_ERROR;
+ }
+
+ // Prepare to decode the next Stream.
+ return_if_error(stream_decoder_reset(coder, allocator));
+ break;
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+ // Never reached
+}
+
+
+static void
+stream_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->block_decoder, allocator);
+ lzma_index_hash_end(coder->index_hash, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_check
+stream_decoder_get_check(const lzma_coder *coder)
+{
+ return coder->stream_flags.check;
+}
+
+
+static lzma_ret
+stream_decoder_memconfig(lzma_coder *coder, uint64_t *memusage,
+ uint64_t *old_memlimit, uint64_t new_memlimit)
+{
+ *memusage = coder->memusage;
+ *old_memlimit = coder->memlimit;
+
+ if (new_memlimit != 0) {
+ if (new_memlimit < coder->memusage)
+ return LZMA_MEMLIMIT_ERROR;
+
+ coder->memlimit = new_memlimit;
+ }
+
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_stream_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ uint64_t memlimit, uint32_t flags)
+{
+ lzma_next_coder_init(&lzma_stream_decoder_init, next, allocator);
+
+ if (memlimit == 0)
+ return LZMA_PROG_ERROR;
+
+ if (flags & ~LZMA_SUPPORTED_FLAGS)
+ return LZMA_OPTIONS_ERROR;
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &stream_decode;
+ next->end = &stream_decoder_end;
+ next->get_check = &stream_decoder_get_check;
+ next->memconfig = &stream_decoder_memconfig;
+
+ next->coder->block_decoder = LZMA_NEXT_CODER_INIT;
+ next->coder->index_hash = NULL;
+ }
+
+ next->coder->memlimit = memlimit;
+ next->coder->memusage = LZMA_MEMUSAGE_BASE;
+ next->coder->tell_no_check = (flags & LZMA_TELL_NO_CHECK) != 0;
+ next->coder->tell_unsupported_check
+ = (flags & LZMA_TELL_UNSUPPORTED_CHECK) != 0;
+ next->coder->tell_any_check = (flags & LZMA_TELL_ANY_CHECK) != 0;
+ next->coder->concatenated = (flags & LZMA_CONCATENATED) != 0;
+ next->coder->first_stream = true;
+
+ return stream_decoder_reset(next->coder, allocator);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_decoder(lzma_stream *strm, uint64_t memlimit, uint32_t flags)
+{
+ lzma_next_strm_init(lzma_stream_decoder_init, strm, memlimit, flags);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/stream_decoder.h b/src/liblzma/common/stream_decoder.h
new file mode 100644
index 0000000..e54ac28
--- /dev/null
+++ b/src/liblzma/common/stream_decoder.h
@@ -0,0 +1,21 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_decoder.h
+/// \brief Decodes .xz Streams
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_STREAM_DECODER_H
+#define LZMA_STREAM_DECODER_H
+
+#include "common.h"
+
+extern lzma_ret lzma_stream_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, uint64_t memlimit, uint32_t flags);
+
+#endif
diff --git a/src/liblzma/common/stream_encoder.c b/src/liblzma/common/stream_encoder.c
new file mode 100644
index 0000000..48d91da
--- /dev/null
+++ b/src/liblzma/common/stream_encoder.c
@@ -0,0 +1,331 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_encoder.c
+/// \brief Encodes .xz Streams
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_encoder.h"
+#include "block_encoder.h"
+#include "index_encoder.h"
+
+
+struct lzma_coder_s {
+ enum {
+ SEQ_STREAM_HEADER,
+ SEQ_BLOCK_INIT,
+ SEQ_BLOCK_HEADER,
+ SEQ_BLOCK_ENCODE,
+ SEQ_INDEX_ENCODE,
+ SEQ_STREAM_FOOTER,
+ } sequence;
+
+ /// True if Block encoder has been initialized by
+ /// lzma_stream_encoder_init() or stream_encoder_update()
+ /// and thus doesn't need to be initialized in stream_encode().
+ bool block_encoder_is_initialized;
+
+ /// Block
+ lzma_next_coder block_encoder;
+
+ /// Options for the Block encoder
+ lzma_block block_options;
+
+ /// The filter chain currently in use
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+
+ /// Index encoder. This is separate from Block encoder, because this
+ /// doesn't take much memory, and when encoding multiple Streams
+ /// with the same encoding options we avoid reallocating memory.
+ lzma_next_coder index_encoder;
+
+ /// Index to hold sizes of the Blocks
+ lzma_index *index;
+
+ /// Read position in buffer[]
+ size_t buffer_pos;
+
+ /// Total number of bytes in buffer[]
+ size_t buffer_size;
+
+ /// Buffer to hold Stream Header, Block Header, and Stream Footer.
+ /// Block Header has biggest maximum size.
+ uint8_t buffer[LZMA_BLOCK_HEADER_SIZE_MAX];
+};
+
+
+static lzma_ret
+block_encoder_init(lzma_coder *coder, lzma_allocator *allocator)
+{
+ // Prepare the Block options. Even though Block encoder doesn't need
+ // compressed_size, uncompressed_size, and header_size to be
+ // initialized, it is a good idea to do it here, because this way
+ // we catch if someone gave us Filter ID that cannot be used in
+ // Blocks/Streams.
+ coder->block_options.compressed_size = LZMA_VLI_UNKNOWN;
+ coder->block_options.uncompressed_size = LZMA_VLI_UNKNOWN;
+
+ return_if_error(lzma_block_header_size(&coder->block_options));
+
+ // Initialize the actual Block encoder.
+ return lzma_block_encoder_init(&coder->block_encoder, allocator,
+ &coder->block_options);
+}
+
+
+static lzma_ret
+stream_encode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ // Main loop
+ while (*out_pos < out_size)
+ switch (coder->sequence) {
+ case SEQ_STREAM_HEADER:
+ case SEQ_BLOCK_HEADER:
+ case SEQ_STREAM_FOOTER:
+ lzma_bufcpy(coder->buffer, &coder->buffer_pos,
+ coder->buffer_size, out, out_pos, out_size);
+ if (coder->buffer_pos < coder->buffer_size)
+ return LZMA_OK;
+
+ if (coder->sequence == SEQ_STREAM_FOOTER)
+ return LZMA_STREAM_END;
+
+ coder->buffer_pos = 0;
+ ++coder->sequence;
+ break;
+
+ case SEQ_BLOCK_INIT: {
+ if (*in_pos == in_size) {
+ // If we are requested to flush or finish the current
+ // Block, return LZMA_STREAM_END immediately since
+ // there's nothing to do.
+ if (action != LZMA_FINISH)
+ return action == LZMA_RUN
+ ? LZMA_OK : LZMA_STREAM_END;
+
+ // The application had used LZMA_FULL_FLUSH to finish
+ // the previous Block, but now wants to finish without
+ // encoding new data, or it is simply creating an
+ // empty Stream with no Blocks.
+ //
+ // Initialize the Index encoder, and continue to
+ // actually encoding the Index.
+ return_if_error(lzma_index_encoder_init(
+ &coder->index_encoder, allocator,
+ coder->index));
+ coder->sequence = SEQ_INDEX_ENCODE;
+ break;
+ }
+
+ // Initialize the Block encoder unless it was already
+ // initialized by lzma_stream_encoder_init() or
+ // stream_encoder_update().
+ if (!coder->block_encoder_is_initialized)
+ return_if_error(block_encoder_init(coder, allocator));
+
+ // Make it false so that we don't skip the initialization
+ // with the next Block.
+ coder->block_encoder_is_initialized = false;
+
+ // Encode the Block Header. This shouldn't fail since we have
+ // already initialized the Block encoder.
+ if (lzma_block_header_encode(&coder->block_options,
+ coder->buffer) != LZMA_OK)
+ return LZMA_PROG_ERROR;
+
+ coder->buffer_size = coder->block_options.header_size;
+ coder->sequence = SEQ_BLOCK_HEADER;
+ break;
+ }
+
+ case SEQ_BLOCK_ENCODE: {
+ static const lzma_action convert[4] = {
+ LZMA_RUN,
+ LZMA_SYNC_FLUSH,
+ LZMA_FINISH,
+ LZMA_FINISH,
+ };
+
+ const lzma_ret ret = coder->block_encoder.code(
+ coder->block_encoder.coder, allocator,
+ in, in_pos, in_size,
+ out, out_pos, out_size, convert[action]);
+ if (ret != LZMA_STREAM_END || action == LZMA_SYNC_FLUSH)
+ return ret;
+
+ // Add a new Index Record.
+ const lzma_vli unpadded_size = lzma_block_unpadded_size(
+ &coder->block_options);
+ assert(unpadded_size != 0);
+ return_if_error(lzma_index_append(coder->index, allocator,
+ unpadded_size,
+ coder->block_options.uncompressed_size));
+
+ coder->sequence = SEQ_BLOCK_INIT;
+ break;
+ }
+
+ case SEQ_INDEX_ENCODE: {
+ // Call the Index encoder. It doesn't take any input, so
+ // those pointers can be NULL.
+ const lzma_ret ret = coder->index_encoder.code(
+ coder->index_encoder.coder, allocator,
+ NULL, NULL, 0,
+ out, out_pos, out_size, LZMA_RUN);
+ if (ret != LZMA_STREAM_END)
+ return ret;
+
+ // Encode the Stream Footer into coder->buffer.
+ const lzma_stream_flags stream_flags = {
+ .version = 0,
+ .backward_size = lzma_index_size(coder->index),
+ .check = coder->block_options.check,
+ };
+
+ if (lzma_stream_footer_encode(&stream_flags, coder->buffer)
+ != LZMA_OK)
+ return LZMA_PROG_ERROR;
+
+ coder->buffer_size = LZMA_STREAM_HEADER_SIZE;
+ coder->sequence = SEQ_STREAM_FOOTER;
+ break;
+ }
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+ return LZMA_OK;
+}
+
+
+static void
+stream_encoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->block_encoder, allocator);
+ lzma_next_end(&coder->index_encoder, allocator);
+ lzma_index_end(coder->index, allocator);
+
+ for (size_t i = 0; coder->filters[i].id != LZMA_VLI_UNKNOWN; ++i)
+ lzma_free(coder->filters[i].options, allocator);
+
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+stream_encoder_update(lzma_coder *coder, lzma_allocator *allocator,
+ const lzma_filter *filters,
+ const lzma_filter *reversed_filters)
+{
+ if (coder->sequence <= SEQ_BLOCK_INIT) {
+ // There is no incomplete Block waiting to be finished,
+ // thus we can change the whole filter chain. Start by
+ // trying to initialize the Block encoder with the new
+ // chain. This way we detect if the chain is valid.
+ coder->block_encoder_is_initialized = false;
+ coder->block_options.filters = (lzma_filter *)(filters);
+ const lzma_ret ret = block_encoder_init(coder, allocator);
+ coder->block_options.filters = coder->filters;
+ if (ret != LZMA_OK)
+ return ret;
+
+ coder->block_encoder_is_initialized = true;
+
+ } else if (coder->sequence <= SEQ_BLOCK_ENCODE) {
+ // We are in the middle of a Block. Try to update only
+ // the filter-specific options.
+ return_if_error(coder->block_encoder.update(
+ coder->block_encoder.coder, allocator,
+ filters, reversed_filters));
+ } else {
+ // Trying to update the filter chain when we are already
+ // encoding Index or Stream Footer.
+ return LZMA_PROG_ERROR;
+ }
+
+ // Free the copy of the old chain and make a copy of the new chain.
+ for (size_t i = 0; coder->filters[i].id != LZMA_VLI_UNKNOWN; ++i)
+ lzma_free(coder->filters[i].options, allocator);
+
+ return lzma_filters_copy(filters, coder->filters, allocator);
+}
+
+
+extern lzma_ret
+lzma_stream_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *filters, lzma_check check)
+{
+ lzma_next_coder_init(&lzma_stream_encoder_init, next, allocator);
+
+ if (filters == NULL)
+ return LZMA_PROG_ERROR;
+
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &stream_encode;
+ next->end = &stream_encoder_end;
+ next->update = &stream_encoder_update;
+
+ next->coder->block_encoder = LZMA_NEXT_CODER_INIT;
+ next->coder->index_encoder = LZMA_NEXT_CODER_INIT;
+ next->coder->index = NULL;
+ }
+
+ // Basic initializations
+ next->coder->sequence = SEQ_STREAM_HEADER;
+ next->coder->block_options.version = 0;
+ next->coder->block_options.check = check;
+ next->coder->filters[0].id = LZMA_VLI_UNKNOWN;
+
+ // Initialize the Index
+ lzma_index_end(next->coder->index, allocator);
+ next->coder->index = lzma_index_init(allocator);
+ if (next->coder->index == NULL)
+ return LZMA_MEM_ERROR;
+
+ // Encode the Stream Header
+ lzma_stream_flags stream_flags = {
+ .version = 0,
+ .check = check,
+ };
+ return_if_error(lzma_stream_header_encode(
+ &stream_flags, next->coder->buffer));
+
+ next->coder->buffer_pos = 0;
+ next->coder->buffer_size = LZMA_STREAM_HEADER_SIZE;
+
+ // Initialize the Block encoder. This way we detect unsupported
+ // filter chains when initializing the Stream encoder instead of
+ // giving an error after Stream Header has already written out.
+ return stream_encoder_update(
+ next->coder, allocator, filters, NULL);
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_encoder(lzma_stream *strm,
+ const lzma_filter *filters, lzma_check check)
+{
+ lzma_next_strm_init(lzma_stream_encoder_init, strm, filters, check);
+
+ strm->internal->supported_actions[LZMA_RUN] = true;
+ strm->internal->supported_actions[LZMA_SYNC_FLUSH] = true;
+ strm->internal->supported_actions[LZMA_FULL_FLUSH] = true;
+ strm->internal->supported_actions[LZMA_FINISH] = true;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/stream_encoder.h b/src/liblzma/common/stream_encoder.h
new file mode 100644
index 0000000..46a7aed
--- /dev/null
+++ b/src/liblzma/common/stream_encoder.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_encoder.h
+/// \brief Encodes .xz Streams
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_STREAM_ENCODER_H
+#define LZMA_STREAM_ENCODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_stream_encoder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter *filters, lzma_check check);
+
+#endif
diff --git a/src/liblzma/common/stream_flags_common.c b/src/liblzma/common/stream_flags_common.c
new file mode 100644
index 0000000..fbe8eb8
--- /dev/null
+++ b/src/liblzma/common/stream_flags_common.c
@@ -0,0 +1,47 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_flags_common.c
+/// \brief Common stuff for Stream flags coders
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_flags_common.h"
+
+
+const uint8_t lzma_header_magic[6] = { 0xFD, 0x37, 0x7A, 0x58, 0x5A, 0x00 };
+const uint8_t lzma_footer_magic[2] = { 0x59, 0x5A };
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_flags_compare(
+ const lzma_stream_flags *a, const lzma_stream_flags *b)
+{
+ // We can compare only version 0 structures.
+ if (a->version != 0 || b->version != 0)
+ return LZMA_OPTIONS_ERROR;
+
+ // Check type
+ if ((unsigned int)(a->check) > LZMA_CHECK_ID_MAX
+ || (unsigned int)(b->check) > LZMA_CHECK_ID_MAX)
+ return LZMA_PROG_ERROR;
+
+ if (a->check != b->check)
+ return LZMA_DATA_ERROR;
+
+ // Backward Sizes are compared only if they are known in both.
+ if (a->backward_size != LZMA_VLI_UNKNOWN
+ && b->backward_size != LZMA_VLI_UNKNOWN) {
+ if (!is_backward_size_valid(a) || !is_backward_size_valid(b))
+ return LZMA_PROG_ERROR;
+
+ if (a->backward_size != b->backward_size)
+ return LZMA_DATA_ERROR;
+ }
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/stream_flags_common.h b/src/liblzma/common/stream_flags_common.h
new file mode 100644
index 0000000..9f3122a
--- /dev/null
+++ b/src/liblzma/common/stream_flags_common.h
@@ -0,0 +1,33 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_flags_common.h
+/// \brief Common stuff for Stream flags coders
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_STREAM_FLAGS_COMMON_H
+#define LZMA_STREAM_FLAGS_COMMON_H
+
+#include "common.h"
+
+/// Size of the Stream Flags field
+#define LZMA_STREAM_FLAGS_SIZE 2
+
+extern const uint8_t lzma_header_magic[6];
+extern const uint8_t lzma_footer_magic[2];
+
+
+static inline bool
+is_backward_size_valid(const lzma_stream_flags *options)
+{
+ return options->backward_size >= LZMA_BACKWARD_SIZE_MIN
+ && options->backward_size <= LZMA_BACKWARD_SIZE_MAX
+ && (options->backward_size & 3) == 0;
+}
+
+#endif
diff --git a/src/liblzma/common/stream_flags_decoder.c b/src/liblzma/common/stream_flags_decoder.c
new file mode 100644
index 0000000..1bc2f97
--- /dev/null
+++ b/src/liblzma/common/stream_flags_decoder.c
@@ -0,0 +1,82 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_flags_decoder.c
+/// \brief Decodes Stream Header and Stream Footer from .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_flags_common.h"
+
+
+static bool
+stream_flags_decode(lzma_stream_flags *options, const uint8_t *in)
+{
+ // Reserved bits must be unset.
+ if (in[0] != 0x00 || (in[1] & 0xF0))
+ return true;
+
+ options->version = 0;
+ options->check = in[1] & 0x0F;
+
+ return false;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_header_decode(lzma_stream_flags *options, const uint8_t *in)
+{
+ // Magic
+ if (memcmp(in, lzma_header_magic, sizeof(lzma_header_magic)) != 0)
+ return LZMA_FORMAT_ERROR;
+
+ // Verify the CRC32 so we can distinguish between corrupt
+ // and unsupported files.
+ const uint32_t crc = lzma_crc32(in + sizeof(lzma_header_magic),
+ LZMA_STREAM_FLAGS_SIZE, 0);
+ if (crc != unaligned_read32le(in + sizeof(lzma_header_magic)
+ + LZMA_STREAM_FLAGS_SIZE))
+ return LZMA_DATA_ERROR;
+
+ // Stream Flags
+ if (stream_flags_decode(options, in + sizeof(lzma_header_magic)))
+ return LZMA_OPTIONS_ERROR;
+
+ // Set Backward Size to indicate unknown value. That way
+ // lzma_stream_flags_compare() can be used to compare Stream Header
+ // and Stream Footer while keeping it useful also for comparing
+ // two Stream Footers.
+ options->backward_size = LZMA_VLI_UNKNOWN;
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_footer_decode(lzma_stream_flags *options, const uint8_t *in)
+{
+ // Magic
+ if (memcmp(in + sizeof(uint32_t) * 2 + LZMA_STREAM_FLAGS_SIZE,
+ lzma_footer_magic, sizeof(lzma_footer_magic)) != 0)
+ return LZMA_FORMAT_ERROR;
+
+ // CRC32
+ const uint32_t crc = lzma_crc32(in + sizeof(uint32_t),
+ sizeof(uint32_t) + LZMA_STREAM_FLAGS_SIZE, 0);
+ if (crc != unaligned_read32le(in))
+ return LZMA_DATA_ERROR;
+
+ // Stream Flags
+ if (stream_flags_decode(options, in + sizeof(uint32_t) * 2))
+ return LZMA_OPTIONS_ERROR;
+
+ // Backward Size
+ options->backward_size = unaligned_read32le(in + sizeof(uint32_t));
+ options->backward_size = (options->backward_size + 1) * 4;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/stream_flags_encoder.c b/src/liblzma/common/stream_flags_encoder.c
new file mode 100644
index 0000000..4e71715
--- /dev/null
+++ b/src/liblzma/common/stream_flags_encoder.c
@@ -0,0 +1,86 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file stream_flags_encoder.c
+/// \brief Encodes Stream Header and Stream Footer for .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "stream_flags_common.h"
+
+
+static bool
+stream_flags_encode(const lzma_stream_flags *options, uint8_t *out)
+{
+ if ((unsigned int)(options->check) > LZMA_CHECK_ID_MAX)
+ return true;
+
+ out[0] = 0x00;
+ out[1] = options->check;
+
+ return false;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_header_encode(const lzma_stream_flags *options, uint8_t *out)
+{
+ assert(sizeof(lzma_header_magic) + LZMA_STREAM_FLAGS_SIZE
+ + 4 == LZMA_STREAM_HEADER_SIZE);
+
+ if (options->version != 0)
+ return LZMA_OPTIONS_ERROR;
+
+ // Magic
+ memcpy(out, lzma_header_magic, sizeof(lzma_header_magic));
+
+ // Stream Flags
+ if (stream_flags_encode(options, out + sizeof(lzma_header_magic)))
+ return LZMA_PROG_ERROR;
+
+ // CRC32 of the Stream Header
+ const uint32_t crc = lzma_crc32(out + sizeof(lzma_header_magic),
+ LZMA_STREAM_FLAGS_SIZE, 0);
+
+ unaligned_write32le(out + sizeof(lzma_header_magic)
+ + LZMA_STREAM_FLAGS_SIZE, crc);
+
+ return LZMA_OK;
+}
+
+
+extern LZMA_API(lzma_ret)
+lzma_stream_footer_encode(const lzma_stream_flags *options, uint8_t *out)
+{
+ assert(2 * 4 + LZMA_STREAM_FLAGS_SIZE + sizeof(lzma_footer_magic)
+ == LZMA_STREAM_HEADER_SIZE);
+
+ if (options->version != 0)
+ return LZMA_OPTIONS_ERROR;
+
+ // Backward Size
+ if (!is_backward_size_valid(options))
+ return LZMA_PROG_ERROR;
+
+ unaligned_write32le(out + 4, options->backward_size / 4 - 1);
+
+ // Stream Flags
+ if (stream_flags_encode(options, out + 2 * 4))
+ return LZMA_PROG_ERROR;
+
+ // CRC32
+ const uint32_t crc = lzma_crc32(
+ out + 4, 4 + LZMA_STREAM_FLAGS_SIZE, 0);
+
+ unaligned_write32le(out, crc);
+
+ // Magic
+ memcpy(out + 2 * 4 + LZMA_STREAM_FLAGS_SIZE,
+ lzma_footer_magic, sizeof(lzma_footer_magic));
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/common/vli_decoder.c b/src/liblzma/common/vli_decoder.c
new file mode 100644
index 0000000..c181828
--- /dev/null
+++ b/src/liblzma/common/vli_decoder.c
@@ -0,0 +1,86 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file vli_decoder.c
+/// \brief Decodes variable-length integers
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_vli_decode(lzma_vli *restrict vli, size_t *vli_pos,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size)
+{
+ // If we haven't been given vli_pos, work in single-call mode.
+ size_t vli_pos_internal = 0;
+ if (vli_pos == NULL) {
+ vli_pos = &vli_pos_internal;
+ *vli = 0;
+
+ // If there's no input, use LZMA_DATA_ERROR. This way it is
+ // easy to decode VLIs from buffers that have known size,
+ // and get the correct error code in case the buffer is
+ // too short.
+ if (*in_pos >= in_size)
+ return LZMA_DATA_ERROR;
+
+ } else {
+ // Initialize *vli when starting to decode a new integer.
+ if (*vli_pos == 0)
+ *vli = 0;
+
+ // Validate the arguments.
+ if (*vli_pos >= LZMA_VLI_BYTES_MAX
+ || (*vli >> (*vli_pos * 7)) != 0)
+ return LZMA_PROG_ERROR;;
+
+ if (*in_pos >= in_size)
+ return LZMA_BUF_ERROR;
+ }
+
+ do {
+ // Read the next byte. Use a temporary variable so that we
+ // can update *in_pos immediately.
+ const uint8_t byte = in[*in_pos];
+ ++*in_pos;
+
+ // Add the newly read byte to *vli.
+ *vli += (lzma_vli)(byte & 0x7F) << (*vli_pos * 7);
+ ++*vli_pos;
+
+ // Check if this is the last byte of a multibyte integer.
+ if ((byte & 0x80) == 0) {
+ // We don't allow using variable-length integers as
+ // padding i.e. the encoding must use the most the
+ // compact form.
+ if (byte == 0x00 && *vli_pos > 1)
+ return LZMA_DATA_ERROR;
+
+ return vli_pos == &vli_pos_internal
+ ? LZMA_OK : LZMA_STREAM_END;
+ }
+
+ // There is at least one more byte coming. If we have already
+ // read maximum number of bytes, the integer is considered
+ // corrupt.
+ //
+ // If we need bigger integers in future, old versions liblzma
+ // will confusingly indicate the file being corrupt istead of
+ // unsupported. I suppose it's still better this way, because
+ // in the foreseeable future (writing this in 2008) the only
+ // reason why files would appear having over 63-bit integers
+ // is that the files are simply corrupt.
+ if (*vli_pos == LZMA_VLI_BYTES_MAX)
+ return LZMA_DATA_ERROR;
+
+ } while (*in_pos < in_size);
+
+ return vli_pos == &vli_pos_internal ? LZMA_DATA_ERROR : LZMA_OK;
+}
diff --git a/src/liblzma/common/vli_encoder.c b/src/liblzma/common/vli_encoder.c
new file mode 100644
index 0000000..f864269
--- /dev/null
+++ b/src/liblzma/common/vli_encoder.c
@@ -0,0 +1,69 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file vli_encoder.c
+/// \brief Encodes variable-length integers
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+
+extern LZMA_API(lzma_ret)
+lzma_vli_encode(lzma_vli vli, size_t *vli_pos,
+ uint8_t *restrict out, size_t *restrict out_pos,
+ size_t out_size)
+{
+ // If we haven't been given vli_pos, work in single-call mode.
+ size_t vli_pos_internal = 0;
+ if (vli_pos == NULL) {
+ vli_pos = &vli_pos_internal;
+
+ // In single-call mode, we expect that the caller has
+ // reserved enough output space.
+ if (*out_pos >= out_size)
+ return LZMA_PROG_ERROR;
+ } else {
+ // This never happens when we are called by liblzma, but
+ // may happen if called directly from an application.
+ if (*out_pos >= out_size)
+ return LZMA_BUF_ERROR;
+ }
+
+ // Validate the arguments.
+ if (*vli_pos >= LZMA_VLI_BYTES_MAX || vli > LZMA_VLI_MAX)
+ return LZMA_PROG_ERROR;
+
+ // Shift vli so that the next bits to encode are the lowest. In
+ // single-call mode this never changes vli since *vli_pos is zero.
+ vli >>= *vli_pos * 7;
+
+ // Write the non-last bytes in a loop.
+ while (vli >= 0x80) {
+ // We don't need *vli_pos during this function call anymore,
+ // but update it here so that it is ready if we need to
+ // return before the whole integer has been decoded.
+ ++*vli_pos;
+ assert(*vli_pos < LZMA_VLI_BYTES_MAX);
+
+ // Write the next byte.
+ out[*out_pos] = (uint8_t)(vli) | 0x80;
+ vli >>= 7;
+
+ if (++*out_pos == out_size)
+ return vli_pos == &vli_pos_internal
+ ? LZMA_PROG_ERROR : LZMA_OK;
+ }
+
+ // Write the last byte.
+ out[*out_pos] = (uint8_t)(vli);
+ ++*out_pos;
+ ++*vli_pos;
+
+ return vli_pos == &vli_pos_internal ? LZMA_OK : LZMA_STREAM_END;
+
+}
diff --git a/src/liblzma/common/vli_size.c b/src/liblzma/common/vli_size.c
new file mode 100644
index 0000000..ec1b4fa
--- /dev/null
+++ b/src/liblzma/common/vli_size.c
@@ -0,0 +1,30 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file vli_size.c
+/// \brief Calculates the encoded size of a variable-length integer
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+
+extern LZMA_API(uint32_t)
+lzma_vli_size(lzma_vli vli)
+{
+ if (vli > LZMA_VLI_MAX)
+ return 0;
+
+ uint32_t i = 0;
+ do {
+ vli >>= 7;
+ ++i;
+ } while (vli != 0);
+
+ assert(i <= LZMA_VLI_BYTES_MAX);
+ return i;
+}
diff --git a/src/liblzma/delta/Makefile.inc b/src/liblzma/delta/Makefile.inc
new file mode 100644
index 0000000..c7739b4
--- /dev/null
+++ b/src/liblzma/delta/Makefile.inc
@@ -0,0 +1,23 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+liblzma_la_SOURCES += \
+ delta/delta_common.c \
+ delta/delta_common.h \
+ delta/delta_private.h
+
+if COND_ENCODER_DELTA
+liblzma_la_SOURCES += \
+ delta/delta_encoder.c \
+ delta/delta_encoder.h
+endif
+
+if COND_DECODER_DELTA
+liblzma_la_SOURCES += \
+ delta/delta_decoder.c \
+ delta/delta_decoder.h
+endif
diff --git a/src/liblzma/delta/delta_common.c b/src/liblzma/delta/delta_common.c
new file mode 100644
index 0000000..930ad21
--- /dev/null
+++ b/src/liblzma/delta/delta_common.c
@@ -0,0 +1,70 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_common.c
+/// \brief Common stuff for Delta encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "delta_common.h"
+#include "delta_private.h"
+
+
+static void
+delta_coder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+extern lzma_ret
+lzma_delta_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ // Allocate memory for the decoder if needed.
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ // End function is the same for encoder and decoder.
+ next->end = &delta_coder_end;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ // Validate the options.
+ if (lzma_delta_coder_memusage(filters[0].options) == UINT64_MAX)
+ return LZMA_OPTIONS_ERROR;
+
+ // Set the delta distance.
+ const lzma_options_delta *opt = filters[0].options;
+ next->coder->distance = opt->dist;
+
+ // Initialize the rest of the variables.
+ next->coder->pos = 0;
+ memzero(next->coder->history, LZMA_DELTA_DIST_MAX);
+
+ // Initialize the next decoder in the chain, if any.
+ return lzma_next_filter_init(&next->coder->next,
+ allocator, filters + 1);
+}
+
+
+extern uint64_t
+lzma_delta_coder_memusage(const void *options)
+{
+ const lzma_options_delta *opt = options;
+
+ if (opt == NULL || opt->type != LZMA_DELTA_TYPE_BYTE
+ || opt->dist < LZMA_DELTA_DIST_MIN
+ || opt->dist > LZMA_DELTA_DIST_MAX)
+ return UINT64_MAX;
+
+ return sizeof(lzma_coder);
+}
diff --git a/src/liblzma/delta/delta_common.h b/src/liblzma/delta/delta_common.h
new file mode 100644
index 0000000..7e7e1ba
--- /dev/null
+++ b/src/liblzma/delta/delta_common.h
@@ -0,0 +1,20 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_common.h
+/// \brief Common stuff for Delta encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_DELTA_COMMON_H
+#define LZMA_DELTA_COMMON_H
+
+#include "common.h"
+
+extern uint64_t lzma_delta_coder_memusage(const void *options);
+
+#endif
diff --git a/src/liblzma/delta/delta_decoder.c b/src/liblzma/delta/delta_decoder.c
new file mode 100644
index 0000000..2cf60d5
--- /dev/null
+++ b/src/liblzma/delta/delta_decoder.c
@@ -0,0 +1,76 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_decoder.c
+/// \brief Delta filter decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "delta_decoder.h"
+#include "delta_private.h"
+
+
+static void
+decode_buffer(lzma_coder *coder, uint8_t *buffer, size_t size)
+{
+ const size_t distance = coder->distance;
+
+ for (size_t i = 0; i < size; ++i) {
+ buffer[i] += coder->history[(distance + coder->pos) & 0xFF];
+ coder->history[coder->pos-- & 0xFF] = buffer[i];
+ }
+}
+
+
+static lzma_ret
+delta_decode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ assert(coder->next.code != NULL);
+
+ const size_t out_start = *out_pos;
+
+ const lzma_ret ret = coder->next.code(coder->next.coder, allocator,
+ in, in_pos, in_size, out, out_pos, out_size,
+ action);
+
+ decode_buffer(coder, out + out_start, *out_pos - out_start);
+
+ return ret;
+}
+
+
+extern lzma_ret
+lzma_delta_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ next->code = &delta_decode;
+ return lzma_delta_coder_init(next, allocator, filters);
+}
+
+
+extern lzma_ret
+lzma_delta_props_decode(void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size)
+{
+ if (props_size != 1)
+ return LZMA_OPTIONS_ERROR;
+
+ lzma_options_delta *opt
+ = lzma_alloc(sizeof(lzma_options_delta), allocator);
+ if (opt == NULL)
+ return LZMA_MEM_ERROR;
+
+ opt->type = LZMA_DELTA_TYPE_BYTE;
+ opt->dist = props[0] + 1;
+
+ *options = opt;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/delta/delta_decoder.h b/src/liblzma/delta/delta_decoder.h
new file mode 100644
index 0000000..ae89acc
--- /dev/null
+++ b/src/liblzma/delta/delta_decoder.h
@@ -0,0 +1,25 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_decoder.h
+/// \brief Delta filter decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_DELTA_DECODER_H
+#define LZMA_DELTA_DECODER_H
+
+#include "delta_common.h"
+
+extern lzma_ret lzma_delta_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_delta_props_decode(
+ void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size);
+
+#endif
diff --git a/src/liblzma/delta/delta_encoder.c b/src/liblzma/delta/delta_encoder.c
new file mode 100644
index 0000000..ea1cc2c
--- /dev/null
+++ b/src/liblzma/delta/delta_encoder.c
@@ -0,0 +1,121 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_encoder.c
+/// \brief Delta filter encoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "delta_encoder.h"
+#include "delta_private.h"
+
+
+/// Copies and encodes the data at the same time. This is used when Delta
+/// is the first filter in the chain (and thus the last filter in the
+/// encoder's filter stack).
+static void
+copy_and_encode(lzma_coder *coder,
+ const uint8_t *restrict in, uint8_t *restrict out, size_t size)
+{
+ const size_t distance = coder->distance;
+
+ for (size_t i = 0; i < size; ++i) {
+ const uint8_t tmp = coder->history[
+ (distance + coder->pos) & 0xFF];
+ coder->history[coder->pos-- & 0xFF] = in[i];
+ out[i] = in[i] - tmp;
+ }
+}
+
+
+/// Encodes the data in place. This is used when we are the last filter
+/// in the chain (and thus non-last filter in the encoder's filter stack).
+static void
+encode_in_place(lzma_coder *coder, uint8_t *buffer, size_t size)
+{
+ const size_t distance = coder->distance;
+
+ for (size_t i = 0; i < size; ++i) {
+ const uint8_t tmp = coder->history[
+ (distance + coder->pos) & 0xFF];
+ coder->history[coder->pos-- & 0xFF] = buffer[i];
+ buffer[i] -= tmp;
+ }
+}
+
+
+static lzma_ret
+delta_encode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ lzma_ret ret;
+
+ if (coder->next.code == NULL) {
+ const size_t in_avail = in_size - *in_pos;
+ const size_t out_avail = out_size - *out_pos;
+ const size_t size = my_min(in_avail, out_avail);
+
+ copy_and_encode(coder, in + *in_pos, out + *out_pos, size);
+
+ *in_pos += size;
+ *out_pos += size;
+
+ ret = action != LZMA_RUN && *in_pos == in_size
+ ? LZMA_STREAM_END : LZMA_OK;
+
+ } else {
+ const size_t out_start = *out_pos;
+
+ ret = coder->next.code(coder->next.coder, allocator,
+ in, in_pos, in_size, out, out_pos, out_size,
+ action);
+
+ encode_in_place(coder, out + out_start, *out_pos - out_start);
+ }
+
+ return ret;
+}
+
+
+static lzma_ret
+delta_encoder_update(lzma_coder *coder, lzma_allocator *allocator,
+ const lzma_filter *filters_null lzma_attribute((unused)),
+ const lzma_filter *reversed_filters)
+{
+ // Delta doesn't and will never support changing the options in
+ // the middle of encoding. If the app tries to change them, we
+ // simply ignore them.
+ return lzma_next_filter_update(
+ &coder->next, allocator, reversed_filters + 1);
+}
+
+
+extern lzma_ret
+lzma_delta_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ next->code = &delta_encode;
+ next->update = &delta_encoder_update;
+ return lzma_delta_coder_init(next, allocator, filters);
+}
+
+
+extern lzma_ret
+lzma_delta_props_encode(const void *options, uint8_t *out)
+{
+ // The caller must have already validated the options, so it's
+ // LZMA_PROG_ERROR if they are invalid.
+ if (lzma_delta_coder_memusage(options) == UINT64_MAX)
+ return LZMA_PROG_ERROR;
+
+ const lzma_options_delta *opt = options;
+ out[0] = opt->dist - LZMA_DELTA_DIST_MIN;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/delta/delta_encoder.h b/src/liblzma/delta/delta_encoder.h
new file mode 100644
index 0000000..a447862
--- /dev/null
+++ b/src/liblzma/delta/delta_encoder.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_encoder.h
+/// \brief Delta filter encoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_DELTA_ENCODER_H
+#define LZMA_DELTA_ENCODER_H
+
+#include "delta_common.h"
+
+extern lzma_ret lzma_delta_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_delta_props_encode(const void *options, uint8_t *out);
+
+#endif
diff --git a/src/liblzma/delta/delta_private.h b/src/liblzma/delta/delta_private.h
new file mode 100644
index 0000000..62b7fed
--- /dev/null
+++ b/src/liblzma/delta/delta_private.h
@@ -0,0 +1,37 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file delta_private.h
+/// \brief Private common stuff for Delta encoder and decoder
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_DELTA_PRIVATE_H
+#define LZMA_DELTA_PRIVATE_H
+
+#include "delta_common.h"
+
+struct lzma_coder_s {
+ /// Next coder in the chain
+ lzma_next_coder next;
+
+ /// Delta distance
+ size_t distance;
+
+ /// Position in history[]
+ uint8_t pos;
+
+ /// Buffer to hold history of the original data
+ uint8_t history[LZMA_DELTA_DIST_MAX];
+};
+
+
+extern lzma_ret lzma_delta_coder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters);
+
+#endif
diff --git a/src/liblzma/liblzma.pc.in b/src/liblzma/liblzma.pc.in
new file mode 100644
index 0000000..7f11f1a
--- /dev/null
+++ b/src/liblzma/liblzma.pc.in
@@ -0,0 +1,19 @@
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+
+prefix=@prefix@
+exec_prefix=@exec_prefix@
+libdir=@libdir@
+includedir=@includedir@
+
+Name: liblzma
+Description: General purpose data compression library
+URL: @PACKAGE_URL@
+Version: @PACKAGE_VERSION@
+Cflags: -I${includedir}
+Libs: -L${libdir} -llzma
+Libs.private: @PTHREAD_CFLAGS@ @PTHREAD_LIBS@
diff --git a/src/liblzma/liblzma_w32res.rc b/src/liblzma/liblzma_w32res.rc
new file mode 100644
index 0000000..d4d8159
--- /dev/null
+++ b/src/liblzma/liblzma_w32res.rc
@@ -0,0 +1,12 @@
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#define MY_TYPE VFT_DLL
+#define MY_NAME "liblzma"
+#define MY_SUFFIX ".dll"
+#define MY_DESC "liblzma data compression library"
+#include "common_w32res.rc"
diff --git a/src/liblzma/lz/Makefile.inc b/src/liblzma/lz/Makefile.inc
new file mode 100644
index 0000000..75742a8
--- /dev/null
+++ b/src/liblzma/lz/Makefile.inc
@@ -0,0 +1,22 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+if COND_ENCODER_LZ
+liblzma_la_SOURCES += \
+ lz/lz_encoder.c \
+ lz/lz_encoder.h \
+ lz/lz_encoder_hash.h \
+ lz/lz_encoder_hash_table.h \
+ lz/lz_encoder_mf.c
+endif
+
+
+if COND_DECODER_LZ
+liblzma_la_SOURCES += \
+ lz/lz_decoder.c \
+ lz/lz_decoder.h
+endif
diff --git a/src/liblzma/lz/lz_decoder.c b/src/liblzma/lz/lz_decoder.c
new file mode 100644
index 0000000..2c57355
--- /dev/null
+++ b/src/liblzma/lz/lz_decoder.c
@@ -0,0 +1,300 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lz_decoder.c
+/// \brief LZ out window
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+// liblzma supports multiple LZ77-based filters. The LZ part is shared
+// between these filters. The LZ code takes care of dictionary handling
+// and passing the data between filters in the chain. The filter-specific
+// part decodes from the input buffer to the dictionary.
+
+
+#include "lz_decoder.h"
+
+
+struct lzma_coder_s {
+ /// Dictionary (history buffer)
+ lzma_dict dict;
+
+ /// The actual LZ-based decoder e.g. LZMA
+ lzma_lz_decoder lz;
+
+ /// Next filter in the chain, if any. Note that LZMA and LZMA2 are
+ /// only allowed as the last filter, but the long-range filter in
+ /// future can be in the middle of the chain.
+ lzma_next_coder next;
+
+ /// True if the next filter in the chain has returned LZMA_STREAM_END.
+ bool next_finished;
+
+ /// True if the LZ decoder (e.g. LZMA) has detected end of payload
+ /// marker. This may become true before next_finished becomes true.
+ bool this_finished;
+
+ /// Temporary buffer needed when the LZ-based filter is not the last
+ /// filter in the chain. The output of the next filter is first
+ /// decoded into buffer[], which is then used as input for the actual
+ /// LZ-based decoder.
+ struct {
+ size_t pos;
+ size_t size;
+ uint8_t buffer[LZMA_BUFFER_SIZE];
+ } temp;
+};
+
+
+static void
+lz_decoder_reset(lzma_coder *coder)
+{
+ coder->dict.pos = 0;
+ coder->dict.full = 0;
+ coder->dict.buf[coder->dict.size - 1] = '\0';
+ coder->dict.need_reset = false;
+ return;
+}
+
+
+static lzma_ret
+decode_buffer(lzma_coder *coder,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size)
+{
+ while (true) {
+ // Wrap the dictionary if needed.
+ if (coder->dict.pos == coder->dict.size)
+ coder->dict.pos = 0;
+
+ // Store the current dictionary position. It is needed to know
+ // where to start copying to the out[] buffer.
+ const size_t dict_start = coder->dict.pos;
+
+ // Calculate how much we allow coder->lz.code() to decode.
+ // It must not decode past the end of the dictionary
+ // buffer, and we don't want it to decode more than is
+ // actually needed to fill the out[] buffer.
+ coder->dict.limit = coder->dict.pos
+ + my_min(out_size - *out_pos,
+ coder->dict.size - coder->dict.pos);
+
+ // Call the coder->lz.code() to do the actual decoding.
+ const lzma_ret ret = coder->lz.code(
+ coder->lz.coder, &coder->dict,
+ in, in_pos, in_size);
+
+ // Copy the decoded data from the dictionary to the out[]
+ // buffer.
+ const size_t copy_size = coder->dict.pos - dict_start;
+ assert(copy_size <= out_size - *out_pos);
+ memcpy(out + *out_pos, coder->dict.buf + dict_start,
+ copy_size);
+ *out_pos += copy_size;
+
+ // Reset the dictionary if so requested by coder->lz.code().
+ if (coder->dict.need_reset) {
+ lz_decoder_reset(coder);
+
+ // Since we reset dictionary, we don't check if
+ // dictionary became full.
+ if (ret != LZMA_OK || *out_pos == out_size)
+ return ret;
+ } else {
+ // Return if everything got decoded or an error
+ // occurred, or if there's no more data to decode.
+ //
+ // Note that detecting if there's something to decode
+ // is done by looking if dictionary become full
+ // instead of looking if *in_pos == in_size. This
+ // is because it is possible that all the input was
+ // consumed already but some data is pending to be
+ // written to the dictionary.
+ if (ret != LZMA_OK || *out_pos == out_size
+ || coder->dict.pos < coder->dict.size)
+ return ret;
+ }
+ }
+}
+
+
+static lzma_ret
+lz_decode(lzma_coder *coder,
+ lzma_allocator *allocator lzma_attribute((unused)),
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size,
+ lzma_action action)
+{
+ if (coder->next.code == NULL)
+ return decode_buffer(coder, in, in_pos, in_size,
+ out, out_pos, out_size);
+
+ // We aren't the last coder in the chain, we need to decode
+ // our input to a temporary buffer.
+ while (*out_pos < out_size) {
+ // Fill the temporary buffer if it is empty.
+ if (!coder->next_finished
+ && coder->temp.pos == coder->temp.size) {
+ coder->temp.pos = 0;
+ coder->temp.size = 0;
+
+ const lzma_ret ret = coder->next.code(
+ coder->next.coder,
+ allocator, in, in_pos, in_size,
+ coder->temp.buffer, &coder->temp.size,
+ LZMA_BUFFER_SIZE, action);
+
+ if (ret == LZMA_STREAM_END)
+ coder->next_finished = true;
+ else if (ret != LZMA_OK || coder->temp.size == 0)
+ return ret;
+ }
+
+ if (coder->this_finished) {
+ if (coder->temp.size != 0)
+ return LZMA_DATA_ERROR;
+
+ if (coder->next_finished)
+ return LZMA_STREAM_END;
+
+ return LZMA_OK;
+ }
+
+ const lzma_ret ret = decode_buffer(coder, coder->temp.buffer,
+ &coder->temp.pos, coder->temp.size,
+ out, out_pos, out_size);
+
+ if (ret == LZMA_STREAM_END)
+ coder->this_finished = true;
+ else if (ret != LZMA_OK)
+ return ret;
+ else if (coder->next_finished && *out_pos < out_size)
+ return LZMA_DATA_ERROR;
+ }
+
+ return LZMA_OK;
+}
+
+
+static void
+lz_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder->dict.buf, allocator);
+
+ if (coder->lz.end != NULL)
+ coder->lz.end(coder->lz.coder, allocator);
+ else
+ lzma_free(coder->lz.coder, allocator);
+
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+extern lzma_ret
+lzma_lz_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters,
+ lzma_ret (*lz_init)(lzma_lz_decoder *lz,
+ lzma_allocator *allocator, const void *options,
+ lzma_lz_options *lz_options))
+{
+ // Allocate the base structure if it isn't already allocated.
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &lz_decode;
+ next->end = &lz_decoder_end;
+
+ next->coder->dict.buf = NULL;
+ next->coder->dict.size = 0;
+ next->coder->lz = LZMA_LZ_DECODER_INIT;
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ // Allocate and initialize the LZ-based decoder. It will also give
+ // us the dictionary size.
+ lzma_lz_options lz_options;
+ return_if_error(lz_init(&next->coder->lz, allocator,
+ filters[0].options, &lz_options));
+
+ // If the dictionary size is very small, increase it to 4096 bytes.
+ // This is to prevent constant wrapping of the dictionary, which
+ // would slow things down. The downside is that since we don't check
+ // separately for the real dictionary size, we may happily accept
+ // corrupt files.
+ if (lz_options.dict_size < 4096)
+ lz_options.dict_size = 4096;
+
+ // Make dictionary size a multipe of 16. Some LZ-based decoders like
+ // LZMA use the lowest bits lzma_dict.pos to know the alignment of the
+ // data. Aligned buffer is also good when memcpying from the
+ // dictionary to the output buffer, since applications are
+ // recommended to give aligned buffers to liblzma.
+ //
+ // Avoid integer overflow.
+ if (lz_options.dict_size > SIZE_MAX - 15)
+ return LZMA_MEM_ERROR;
+
+ lz_options.dict_size = (lz_options.dict_size + 15) & ~((size_t)(15));
+
+ // Allocate and initialize the dictionary.
+ if (next->coder->dict.size != lz_options.dict_size) {
+ lzma_free(next->coder->dict.buf, allocator);
+ next->coder->dict.buf
+ = lzma_alloc(lz_options.dict_size, allocator);
+ if (next->coder->dict.buf == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->coder->dict.size = lz_options.dict_size;
+ }
+
+ lz_decoder_reset(next->coder);
+
+ // Use the preset dictionary if it was given to us.
+ if (lz_options.preset_dict != NULL
+ && lz_options.preset_dict_size > 0) {
+ // If the preset dictionary is bigger than the actual
+ // dictionary, copy only the tail.
+ const size_t copy_size = my_min(lz_options.preset_dict_size,
+ lz_options.dict_size);
+ const size_t offset = lz_options.preset_dict_size - copy_size;
+ memcpy(next->coder->dict.buf, lz_options.preset_dict + offset,
+ copy_size);
+ next->coder->dict.pos = copy_size;
+ next->coder->dict.full = copy_size;
+ }
+
+ // Miscellaneous initializations
+ next->coder->next_finished = false;
+ next->coder->this_finished = false;
+ next->coder->temp.pos = 0;
+ next->coder->temp.size = 0;
+
+ // Initialize the next filter in the chain, if any.
+ return lzma_next_filter_init(&next->coder->next, allocator,
+ filters + 1);
+}
+
+
+extern uint64_t
+lzma_lz_decoder_memusage(size_t dictionary_size)
+{
+ return sizeof(lzma_coder) + (uint64_t)(dictionary_size);
+}
+
+
+extern void
+lzma_lz_decoder_uncompressed(lzma_coder *coder, lzma_vli uncompressed_size)
+{
+ coder->lz.set_uncompressed(coder->lz.coder, uncompressed_size);
+}
diff --git a/src/liblzma/lz/lz_decoder.h b/src/liblzma/lz/lz_decoder.h
new file mode 100644
index 0000000..7266e80
--- /dev/null
+++ b/src/liblzma/lz/lz_decoder.h
@@ -0,0 +1,234 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lz_decoder.h
+/// \brief LZ out window
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZ_DECODER_H
+#define LZMA_LZ_DECODER_H
+
+#include "common.h"
+
+
+typedef struct {
+ /// Pointer to the dictionary buffer. It can be an allocated buffer
+ /// internal to liblzma, or it can a be a buffer given by the
+ /// application when in single-call mode (not implemented yet).
+ uint8_t *buf;
+
+ /// Write position in dictionary. The next byte will be written to
+ /// buf[pos].
+ size_t pos;
+
+ /// Indicates how full the dictionary is. This is used by
+ /// dict_is_distance_valid() to detect corrupt files that would
+ /// read beyond the beginning of the dictionary.
+ size_t full;
+
+ /// Write limit
+ size_t limit;
+
+ /// Size of the dictionary
+ size_t size;
+
+ /// True when dictionary should be reset before decoding more data.
+ bool need_reset;
+
+} lzma_dict;
+
+
+typedef struct {
+ size_t dict_size;
+ const uint8_t *preset_dict;
+ size_t preset_dict_size;
+} lzma_lz_options;
+
+
+typedef struct {
+ /// Data specific to the LZ-based decoder
+ lzma_coder *coder;
+
+ /// Function to decode from in[] to *dict
+ lzma_ret (*code)(lzma_coder *restrict coder,
+ lzma_dict *restrict dict, const uint8_t *restrict in,
+ size_t *restrict in_pos, size_t in_size);
+
+ void (*reset)(lzma_coder *coder, const void *options);
+
+ /// Set the uncompressed size
+ void (*set_uncompressed)(lzma_coder *coder,
+ lzma_vli uncompressed_size);
+
+ /// Free allocated resources
+ void (*end)(lzma_coder *coder, lzma_allocator *allocator);
+
+} lzma_lz_decoder;
+
+
+#define LZMA_LZ_DECODER_INIT \
+ (lzma_lz_decoder){ \
+ .coder = NULL, \
+ .code = NULL, \
+ .reset = NULL, \
+ .set_uncompressed = NULL, \
+ .end = NULL, \
+ }
+
+
+extern lzma_ret lzma_lz_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters,
+ lzma_ret (*lz_init)(lzma_lz_decoder *lz,
+ lzma_allocator *allocator, const void *options,
+ lzma_lz_options *lz_options));
+
+extern uint64_t lzma_lz_decoder_memusage(size_t dictionary_size);
+
+extern void lzma_lz_decoder_uncompressed(
+ lzma_coder *coder, lzma_vli uncompressed_size);
+
+
+//////////////////////
+// Inline functions //
+//////////////////////
+
+/// Get a byte from the history buffer.
+static inline uint8_t
+dict_get(const lzma_dict *const dict, const uint32_t distance)
+{
+ return dict->buf[dict->pos - distance - 1
+ + (distance < dict->pos ? 0 : dict->size)];
+}
+
+
+/// Test if dictionary is empty.
+static inline bool
+dict_is_empty(const lzma_dict *const dict)
+{
+ return dict->full == 0;
+}
+
+
+/// Validate the match distance
+static inline bool
+dict_is_distance_valid(const lzma_dict *const dict, const size_t distance)
+{
+ return dict->full > distance;
+}
+
+
+/// Repeat *len bytes at distance.
+static inline bool
+dict_repeat(lzma_dict *dict, uint32_t distance, uint32_t *len)
+{
+ // Don't write past the end of the dictionary.
+ const size_t dict_avail = dict->limit - dict->pos;
+ uint32_t left = my_min(dict_avail, *len);
+ *len -= left;
+
+ // Repeat a block of data from the history. Because memcpy() is faster
+ // than copying byte by byte in a loop, the copying process gets split
+ // into three cases.
+ if (distance < left) {
+ // Source and target areas overlap, thus we can't use
+ // memcpy() nor even memmove() safely.
+ do {
+ dict->buf[dict->pos] = dict_get(dict, distance);
+ ++dict->pos;
+ } while (--left > 0);
+
+ } else if (distance < dict->pos) {
+ // The easiest and fastest case
+ memcpy(dict->buf + dict->pos,
+ dict->buf + dict->pos - distance - 1,
+ left);
+ dict->pos += left;
+
+ } else {
+ // The bigger the dictionary, the more rare this
+ // case occurs. We need to "wrap" the dict, thus
+ // we might need two memcpy() to copy all the data.
+ assert(dict->full == dict->size);
+ const uint32_t copy_pos
+ = dict->pos - distance - 1 + dict->size;
+ uint32_t copy_size = dict->size - copy_pos;
+
+ if (copy_size < left) {
+ memmove(dict->buf + dict->pos, dict->buf + copy_pos,
+ copy_size);
+ dict->pos += copy_size;
+ copy_size = left - copy_size;
+ memcpy(dict->buf + dict->pos, dict->buf, copy_size);
+ dict->pos += copy_size;
+ } else {
+ memmove(dict->buf + dict->pos, dict->buf + copy_pos,
+ left);
+ dict->pos += left;
+ }
+ }
+
+ // Update how full the dictionary is.
+ if (dict->full < dict->pos)
+ dict->full = dict->pos;
+
+ return unlikely(*len != 0);
+}
+
+
+/// Puts one byte into the dictionary. Returns true if the dictionary was
+/// already full and the byte couldn't be added.
+static inline bool
+dict_put(lzma_dict *dict, uint8_t byte)
+{
+ if (unlikely(dict->pos == dict->limit))
+ return true;
+
+ dict->buf[dict->pos++] = byte;
+
+ if (dict->pos > dict->full)
+ dict->full = dict->pos;
+
+ return false;
+}
+
+
+/// Copies arbitrary amount of data into the dictionary.
+static inline void
+dict_write(lzma_dict *restrict dict, const uint8_t *restrict in,
+ size_t *restrict in_pos, size_t in_size,
+ size_t *restrict left)
+{
+ // NOTE: If we are being given more data than the size of the
+ // dictionary, it could be possible to optimize the LZ decoder
+ // so that not everything needs to go through the dictionary.
+ // This shouldn't be very common thing in practice though, and
+ // the slowdown of one extra memcpy() isn't bad compared to how
+ // much time it would have taken if the data were compressed.
+
+ if (in_size - *in_pos > *left)
+ in_size = *in_pos + *left;
+
+ *left -= lzma_bufcpy(in, in_pos, in_size,
+ dict->buf, &dict->pos, dict->limit);
+
+ if (dict->pos > dict->full)
+ dict->full = dict->pos;
+
+ return;
+}
+
+
+static inline void
+dict_reset(lzma_dict *dict)
+{
+ dict->need_reset = true;
+ return;
+}
+
+#endif
diff --git a/src/liblzma/lz/lz_encoder.c b/src/liblzma/lz/lz_encoder.c
new file mode 100644
index 0000000..9e980a2
--- /dev/null
+++ b/src/liblzma/lz/lz_encoder.c
@@ -0,0 +1,582 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lz_encoder.c
+/// \brief LZ in window
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lz_encoder.h"
+#include "lz_encoder_hash.h"
+
+// See lz_encoder_hash.h. This is a bit hackish but avoids making
+// endianness a conditional in makefiles.
+#if defined(WORDS_BIGENDIAN) && !defined(HAVE_SMALL)
+# include "lz_encoder_hash_table.h"
+#endif
+
+
+struct lzma_coder_s {
+ /// LZ-based encoder e.g. LZMA
+ lzma_lz_encoder lz;
+
+ /// History buffer and match finder
+ lzma_mf mf;
+
+ /// Next coder in the chain
+ lzma_next_coder next;
+};
+
+
+/// \brief Moves the data in the input window to free space for new data
+///
+/// mf->buffer is a sliding input window, which keeps mf->keep_size_before
+/// bytes of input history available all the time. Now and then we need to
+/// "slide" the buffer to make space for the new data to the end of the
+/// buffer. At the same time, data older than keep_size_before is dropped.
+///
+static void
+move_window(lzma_mf *mf)
+{
+ // Align the move to a multiple of 16 bytes. Some LZ-based encoders
+ // like LZMA use the lowest bits of mf->read_pos to know the
+ // alignment of the uncompressed data. We also get better speed
+ // for memmove() with aligned buffers.
+ assert(mf->read_pos > mf->keep_size_before);
+ const uint32_t move_offset
+ = (mf->read_pos - mf->keep_size_before) & ~UINT32_C(15);
+
+ assert(mf->write_pos > move_offset);
+ const size_t move_size = mf->write_pos - move_offset;
+
+ assert(move_offset + move_size <= mf->size);
+
+ memmove(mf->buffer, mf->buffer + move_offset, move_size);
+
+ mf->offset += move_offset;
+ mf->read_pos -= move_offset;
+ mf->read_limit -= move_offset;
+ mf->write_pos -= move_offset;
+
+ return;
+}
+
+
+/// \brief Tries to fill the input window (mf->buffer)
+///
+/// If we are the last encoder in the chain, our input data is in in[].
+/// Otherwise we call the next filter in the chain to process in[] and
+/// write its output to mf->buffer.
+///
+/// This function must not be called once it has returned LZMA_STREAM_END.
+///
+static lzma_ret
+fill_window(lzma_coder *coder, lzma_allocator *allocator, const uint8_t *in,
+ size_t *in_pos, size_t in_size, lzma_action action)
+{
+ assert(coder->mf.read_pos <= coder->mf.write_pos);
+
+ // Move the sliding window if needed.
+ if (coder->mf.read_pos >= coder->mf.size - coder->mf.keep_size_after)
+ move_window(&coder->mf);
+
+ // Maybe this is ugly, but lzma_mf uses uint32_t for most things
+ // (which I find cleanest), but we need size_t here when filling
+ // the history window.
+ size_t write_pos = coder->mf.write_pos;
+ lzma_ret ret;
+ if (coder->next.code == NULL) {
+ // Not using a filter, simply memcpy() as much as possible.
+ lzma_bufcpy(in, in_pos, in_size, coder->mf.buffer,
+ &write_pos, coder->mf.size);
+
+ ret = action != LZMA_RUN && *in_pos == in_size
+ ? LZMA_STREAM_END : LZMA_OK;
+
+ } else {
+ ret = coder->next.code(coder->next.coder, allocator,
+ in, in_pos, in_size,
+ coder->mf.buffer, &write_pos,
+ coder->mf.size, action);
+ }
+
+ coder->mf.write_pos = write_pos;
+
+ // If end of stream has been reached or flushing completed, we allow
+ // the encoder to process all the input (that is, read_pos is allowed
+ // to reach write_pos). Otherwise we keep keep_size_after bytes
+ // available as prebuffer.
+ if (ret == LZMA_STREAM_END) {
+ assert(*in_pos == in_size);
+ ret = LZMA_OK;
+ coder->mf.action = action;
+ coder->mf.read_limit = coder->mf.write_pos;
+
+ } else if (coder->mf.write_pos > coder->mf.keep_size_after) {
+ // This needs to be done conditionally, because if we got
+ // only little new input, there may be too little input
+ // to do any encoding yet.
+ coder->mf.read_limit = coder->mf.write_pos
+ - coder->mf.keep_size_after;
+ }
+
+ // Restart the match finder after finished LZMA_SYNC_FLUSH.
+ if (coder->mf.pending > 0
+ && coder->mf.read_pos < coder->mf.read_limit) {
+ // Match finder may update coder->pending and expects it to
+ // start from zero, so use a temporary variable.
+ const size_t pending = coder->mf.pending;
+ coder->mf.pending = 0;
+
+ // Rewind read_pos so that the match finder can hash
+ // the pending bytes.
+ assert(coder->mf.read_pos >= pending);
+ coder->mf.read_pos -= pending;
+
+ // Call the skip function directly instead of using
+ // mf_skip(), since we don't want to touch mf->read_ahead.
+ coder->mf.skip(&coder->mf, pending);
+ }
+
+ return ret;
+}
+
+
+static lzma_ret
+lz_encode(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size,
+ uint8_t *restrict out, size_t *restrict out_pos,
+ size_t out_size, lzma_action action)
+{
+ while (*out_pos < out_size
+ && (*in_pos < in_size || action != LZMA_RUN)) {
+ // Read more data to coder->mf.buffer if needed.
+ if (coder->mf.action == LZMA_RUN && coder->mf.read_pos
+ >= coder->mf.read_limit)
+ return_if_error(fill_window(coder, allocator,
+ in, in_pos, in_size, action));
+
+ // Encode
+ const lzma_ret ret = coder->lz.code(coder->lz.coder,
+ &coder->mf, out, out_pos, out_size);
+ if (ret != LZMA_OK) {
+ // Setting this to LZMA_RUN for cases when we are
+ // flushing. It doesn't matter when finishing or if
+ // an error occurred.
+ coder->mf.action = LZMA_RUN;
+ return ret;
+ }
+ }
+
+ return LZMA_OK;
+}
+
+
+static bool
+lz_encoder_prepare(lzma_mf *mf, lzma_allocator *allocator,
+ const lzma_lz_options *lz_options)
+{
+ // For now, the dictionary size is limited to 1.5 GiB. This may grow
+ // in the future if needed, but it needs a little more work than just
+ // changing this check.
+ if (lz_options->dict_size < LZMA_DICT_SIZE_MIN
+ || lz_options->dict_size
+ > (UINT32_C(1) << 30) + (UINT32_C(1) << 29)
+ || lz_options->nice_len > lz_options->match_len_max)
+ return true;
+
+ mf->keep_size_before = lz_options->before_size + lz_options->dict_size;
+
+ mf->keep_size_after = lz_options->after_size
+ + lz_options->match_len_max;
+
+ // To avoid constant memmove()s, allocate some extra space. Since
+ // memmove()s become more expensive when the size of the buffer
+ // increases, we reserve more space when a large dictionary is
+ // used to make the memmove() calls rarer.
+ //
+ // This works with dictionaries up to about 3 GiB. If bigger
+ // dictionary is wanted, some extra work is needed:
+ // - Several variables in lzma_mf have to be changed from uint32_t
+ // to size_t.
+ // - Memory usage calculation needs something too, e.g. use uint64_t
+ // for mf->size.
+ uint32_t reserve = lz_options->dict_size / 2;
+ if (reserve > (UINT32_C(1) << 30))
+ reserve /= 2;
+
+ reserve += (lz_options->before_size + lz_options->match_len_max
+ + lz_options->after_size) / 2 + (UINT32_C(1) << 19);
+
+ const uint32_t old_size = mf->size;
+ mf->size = mf->keep_size_before + reserve + mf->keep_size_after;
+
+ // Deallocate the old history buffer if it exists but has different
+ // size than what is needed now.
+ if (mf->buffer != NULL && old_size != mf->size) {
+ lzma_free(mf->buffer, allocator);
+ mf->buffer = NULL;
+ }
+
+ // Match finder options
+ mf->match_len_max = lz_options->match_len_max;
+ mf->nice_len = lz_options->nice_len;
+
+ // cyclic_size has to stay smaller than 2 Gi. Note that this doesn't
+ // mean limiting dictionary size to less than 2 GiB. With a match
+ // finder that uses multibyte resolution (hashes start at e.g. every
+ // fourth byte), cyclic_size would stay below 2 Gi even when
+ // dictionary size is greater than 2 GiB.
+ //
+ // It would be possible to allow cyclic_size >= 2 Gi, but then we
+ // would need to be careful to use 64-bit types in various places
+ // (size_t could do since we would need bigger than 32-bit address
+ // space anyway). It would also require either zeroing a multigigabyte
+ // buffer at initialization (waste of time and RAM) or allow
+ // normalization in lz_encoder_mf.c to access uninitialized
+ // memory to keep the code simpler. The current way is simple and
+ // still allows pretty big dictionaries, so I don't expect these
+ // limits to change.
+ mf->cyclic_size = lz_options->dict_size + 1;
+
+ // Validate the match finder ID and setup the function pointers.
+ switch (lz_options->match_finder) {
+#ifdef HAVE_MF_HC3
+ case LZMA_MF_HC3:
+ mf->find = &lzma_mf_hc3_find;
+ mf->skip = &lzma_mf_hc3_skip;
+ break;
+#endif
+#ifdef HAVE_MF_HC4
+ case LZMA_MF_HC4:
+ mf->find = &lzma_mf_hc4_find;
+ mf->skip = &lzma_mf_hc4_skip;
+ break;
+#endif
+#ifdef HAVE_MF_BT2
+ case LZMA_MF_BT2:
+ mf->find = &lzma_mf_bt2_find;
+ mf->skip = &lzma_mf_bt2_skip;
+ break;
+#endif
+#ifdef HAVE_MF_BT3
+ case LZMA_MF_BT3:
+ mf->find = &lzma_mf_bt3_find;
+ mf->skip = &lzma_mf_bt3_skip;
+ break;
+#endif
+#ifdef HAVE_MF_BT4
+ case LZMA_MF_BT4:
+ mf->find = &lzma_mf_bt4_find;
+ mf->skip = &lzma_mf_bt4_skip;
+ break;
+#endif
+
+ default:
+ return true;
+ }
+
+ // Calculate the sizes of mf->hash and mf->son and check that
+ // nice_len is big enough for the selected match finder.
+ const uint32_t hash_bytes = lz_options->match_finder & 0x0F;
+ if (hash_bytes > mf->nice_len)
+ return true;
+
+ const bool is_bt = (lz_options->match_finder & 0x10) != 0;
+ uint32_t hs;
+
+ if (hash_bytes == 2) {
+ hs = 0xFFFF;
+ } else {
+ // Round dictionary size up to the next 2^n - 1 so it can
+ // be used as a hash mask.
+ hs = lz_options->dict_size - 1;
+ hs |= hs >> 1;
+ hs |= hs >> 2;
+ hs |= hs >> 4;
+ hs |= hs >> 8;
+ hs >>= 1;
+ hs |= 0xFFFF;
+
+ if (hs > (UINT32_C(1) << 24)) {
+ if (hash_bytes == 3)
+ hs = (UINT32_C(1) << 24) - 1;
+ else
+ hs >>= 1;
+ }
+ }
+
+ mf->hash_mask = hs;
+
+ ++hs;
+ if (hash_bytes > 2)
+ hs += HASH_2_SIZE;
+ if (hash_bytes > 3)
+ hs += HASH_3_SIZE;
+/*
+ No match finder uses this at the moment.
+ if (mf->hash_bytes > 4)
+ hs += HASH_4_SIZE;
+*/
+
+ // If the above code calculating hs is modified, make sure that
+ // this assertion stays valid (UINT32_MAX / 5 is not strictly the
+ // exact limit). If it doesn't, you need to calculate that
+ // hash_size_sum + sons_count cannot overflow.
+ assert(hs < UINT32_MAX / 5);
+
+ const uint32_t old_count = mf->hash_size_sum + mf->sons_count;
+ mf->hash_size_sum = hs;
+ mf->sons_count = mf->cyclic_size;
+ if (is_bt)
+ mf->sons_count *= 2;
+
+ const uint32_t new_count = mf->hash_size_sum + mf->sons_count;
+
+ // Deallocate the old hash array if it exists and has different size
+ // than what is needed now.
+ if (old_count != new_count) {
+ lzma_free(mf->hash, allocator);
+ mf->hash = NULL;
+ }
+
+ // Maximum number of match finder cycles
+ mf->depth = lz_options->depth;
+ if (mf->depth == 0) {
+ mf->depth = 16 + (mf->nice_len / 2);
+ if (!is_bt)
+ mf->depth /= 2;
+ }
+
+ return false;
+}
+
+
+static bool
+lz_encoder_init(lzma_mf *mf, lzma_allocator *allocator,
+ const lzma_lz_options *lz_options)
+{
+ // Allocate the history buffer.
+ if (mf->buffer == NULL) {
+ mf->buffer = lzma_alloc(mf->size, allocator);
+ if (mf->buffer == NULL)
+ return true;
+ }
+
+ // Use cyclic_size as initial mf->offset. This allows
+ // avoiding a few branches in the match finders. The downside is
+ // that match finder needs to be normalized more often, which may
+ // hurt performance with huge dictionaries.
+ mf->offset = mf->cyclic_size;
+ mf->read_pos = 0;
+ mf->read_ahead = 0;
+ mf->read_limit = 0;
+ mf->write_pos = 0;
+ mf->pending = 0;
+
+ // Allocate match finder's hash array.
+ const size_t alloc_count = mf->hash_size_sum + mf->sons_count;
+
+#if UINT32_MAX >= SIZE_MAX / 4
+ // Check for integer overflow. (Huge dictionaries are not
+ // possible on 32-bit CPU.)
+ if (alloc_count > SIZE_MAX / sizeof(uint32_t))
+ return true;
+#endif
+
+ if (mf->hash == NULL) {
+ mf->hash = lzma_alloc(alloc_count * sizeof(uint32_t),
+ allocator);
+ if (mf->hash == NULL)
+ return true;
+ }
+
+ mf->son = mf->hash + mf->hash_size_sum;
+ mf->cyclic_pos = 0;
+
+ // Initialize the hash table. Since EMPTY_HASH_VALUE is zero, we
+ // can use memset().
+/*
+ for (uint32_t i = 0; i < hash_size_sum; ++i)
+ mf->hash[i] = EMPTY_HASH_VALUE;
+*/
+ memzero(mf->hash, (size_t)(mf->hash_size_sum) * sizeof(uint32_t));
+
+ // We don't need to initialize mf->son, but not doing that will
+ // make Valgrind complain in normalization (see normalize() in
+ // lz_encoder_mf.c).
+ //
+ // Skipping this initialization is *very* good when big dictionary is
+ // used but only small amount of data gets actually compressed: most
+ // of the mf->hash won't get actually allocated by the kernel, so
+ // we avoid wasting RAM and improve initialization speed a lot.
+ //memzero(mf->son, (size_t)(mf->sons_count) * sizeof(uint32_t));
+
+ // Handle preset dictionary.
+ if (lz_options->preset_dict != NULL
+ && lz_options->preset_dict_size > 0) {
+ // If the preset dictionary is bigger than the actual
+ // dictionary, use only the tail.
+ mf->write_pos = my_min(lz_options->preset_dict_size, mf->size);
+ memcpy(mf->buffer, lz_options->preset_dict
+ + lz_options->preset_dict_size - mf->write_pos,
+ mf->write_pos);
+ mf->action = LZMA_SYNC_FLUSH;
+ mf->skip(mf, mf->write_pos);
+ }
+
+ mf->action = LZMA_RUN;
+
+ return false;
+}
+
+
+extern uint64_t
+lzma_lz_encoder_memusage(const lzma_lz_options *lz_options)
+{
+ // Old buffers must not exist when calling lz_encoder_prepare().
+ lzma_mf mf = {
+ .buffer = NULL,
+ .hash = NULL,
+ .hash_size_sum = 0,
+ .sons_count = 0,
+ };
+
+ // Setup the size information into mf.
+ if (lz_encoder_prepare(&mf, NULL, lz_options))
+ return UINT64_MAX;
+
+ // Calculate the memory usage.
+ return (uint64_t)(mf.hash_size_sum + mf.sons_count)
+ * sizeof(uint32_t)
+ + (uint64_t)(mf.size) + sizeof(lzma_coder);
+}
+
+
+static void
+lz_encoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+
+ lzma_free(coder->mf.hash, allocator);
+ lzma_free(coder->mf.buffer, allocator);
+
+ if (coder->lz.end != NULL)
+ coder->lz.end(coder->lz.coder, allocator);
+ else
+ lzma_free(coder->lz.coder, allocator);
+
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+lz_encoder_update(lzma_coder *coder, lzma_allocator *allocator,
+ const lzma_filter *filters_null lzma_attribute((unused)),
+ const lzma_filter *reversed_filters)
+{
+ if (coder->lz.options_update == NULL)
+ return LZMA_PROG_ERROR;
+
+ return_if_error(coder->lz.options_update(
+ coder->lz.coder, reversed_filters));
+
+ return lzma_next_filter_update(
+ &coder->next, allocator, reversed_filters + 1);
+}
+
+
+extern lzma_ret
+lzma_lz_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters,
+ lzma_ret (*lz_init)(lzma_lz_encoder *lz,
+ lzma_allocator *allocator, const void *options,
+ lzma_lz_options *lz_options))
+{
+#ifdef HAVE_SMALL
+ // We need that the CRC32 table has been initialized.
+ lzma_crc32_init();
+#endif
+
+ // Allocate and initialize the base data structure.
+ if (next->coder == NULL) {
+ next->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &lz_encode;
+ next->end = &lz_encoder_end;
+ next->update = &lz_encoder_update;
+
+ next->coder->lz.coder = NULL;
+ next->coder->lz.code = NULL;
+ next->coder->lz.end = NULL;
+
+ next->coder->mf.buffer = NULL;
+ next->coder->mf.hash = NULL;
+ next->coder->mf.hash_size_sum = 0;
+ next->coder->mf.sons_count = 0;
+
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ }
+
+ // Initialize the LZ-based encoder.
+ lzma_lz_options lz_options;
+ return_if_error(lz_init(&next->coder->lz, allocator,
+ filters[0].options, &lz_options));
+
+ // Setup the size information into next->coder->mf and deallocate
+ // old buffers if they have wrong size.
+ if (lz_encoder_prepare(&next->coder->mf, allocator, &lz_options))
+ return LZMA_OPTIONS_ERROR;
+
+ // Allocate new buffers if needed, and do the rest of
+ // the initialization.
+ if (lz_encoder_init(&next->coder->mf, allocator, &lz_options))
+ return LZMA_MEM_ERROR;
+
+ // Initialize the next filter in the chain, if any.
+ return lzma_next_filter_init(&next->coder->next, allocator,
+ filters + 1);
+}
+
+
+extern LZMA_API(lzma_bool)
+lzma_mf_is_supported(lzma_match_finder mf)
+{
+ bool ret = false;
+
+#ifdef HAVE_MF_HC3
+ if (mf == LZMA_MF_HC3)
+ ret = true;
+#endif
+
+#ifdef HAVE_MF_HC4
+ if (mf == LZMA_MF_HC4)
+ ret = true;
+#endif
+
+#ifdef HAVE_MF_BT2
+ if (mf == LZMA_MF_BT2)
+ ret = true;
+#endif
+
+#ifdef HAVE_MF_BT3
+ if (mf == LZMA_MF_BT3)
+ ret = true;
+#endif
+
+#ifdef HAVE_MF_BT4
+ if (mf == LZMA_MF_BT4)
+ ret = true;
+#endif
+
+ return ret;
+}
diff --git a/src/liblzma/lz/lz_encoder.h b/src/liblzma/lz/lz_encoder.h
new file mode 100644
index 0000000..741c453
--- /dev/null
+++ b/src/liblzma/lz/lz_encoder.h
@@ -0,0 +1,328 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lz_encoder.h
+/// \brief LZ in window and match finder API
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZ_ENCODER_H
+#define LZMA_LZ_ENCODER_H
+
+#include "common.h"
+
+
+/// A table of these is used by the LZ-based encoder to hold
+/// the length-distance pairs found by the match finder.
+typedef struct {
+ uint32_t len;
+ uint32_t dist;
+} lzma_match;
+
+
+typedef struct lzma_mf_s lzma_mf;
+struct lzma_mf_s {
+ ///////////////
+ // In Window //
+ ///////////////
+
+ /// Pointer to buffer with data to be compressed
+ uint8_t *buffer;
+
+ /// Total size of the allocated buffer (that is, including all
+ /// the extra space)
+ uint32_t size;
+
+ /// Number of bytes that must be kept available in our input history.
+ /// That is, once keep_size_before bytes have been processed,
+ /// buffer[read_pos - keep_size_before] is the oldest byte that
+ /// must be available for reading.
+ uint32_t keep_size_before;
+
+ /// Number of bytes that must be kept in buffer after read_pos.
+ /// That is, read_pos <= write_pos - keep_size_after as long as
+ /// action is LZMA_RUN; when action != LZMA_RUN, read_pos is allowed
+ /// to reach write_pos so that the last bytes get encoded too.
+ uint32_t keep_size_after;
+
+ /// Match finders store locations of matches using 32-bit integers.
+ /// To avoid adjusting several megabytes of integers every time the
+ /// input window is moved with move_window, we only adjust the
+ /// offset of the buffer. Thus, buffer[value_in_hash_table - offset]
+ /// is the byte pointed by value_in_hash_table.
+ uint32_t offset;
+
+ /// buffer[read_pos] is the next byte to run through the match
+ /// finder. This is incremented in the match finder once the byte
+ /// has been processed.
+ uint32_t read_pos;
+
+ /// Number of bytes that have been ran through the match finder, but
+ /// which haven't been encoded by the LZ-based encoder yet.
+ uint32_t read_ahead;
+
+ /// As long as read_pos is less than read_limit, there is enough
+ /// input available in buffer for at least one encoding loop.
+ ///
+ /// Because of the stateful API, read_limit may and will get greater
+ /// than read_pos quite often. This is taken into account when
+ /// calculating the value for keep_size_after.
+ uint32_t read_limit;
+
+ /// buffer[write_pos] is the first byte that doesn't contain valid
+ /// uncompressed data; that is, the next input byte will be copied
+ /// to buffer[write_pos].
+ uint32_t write_pos;
+
+ /// Number of bytes not hashed before read_pos. This is needed to
+ /// restart the match finder after LZMA_SYNC_FLUSH.
+ uint32_t pending;
+
+ //////////////////
+ // Match Finder //
+ //////////////////
+
+ /// Find matches. Returns the number of distance-length pairs written
+ /// to the matches array. This is called only via lzma_mf_find().
+ uint32_t (*find)(lzma_mf *mf, lzma_match *matches);
+
+ /// Skips num bytes. This is like find() but doesn't make the
+ /// distance-length pairs available, thus being a little faster.
+ /// This is called only via mf_skip().
+ void (*skip)(lzma_mf *mf, uint32_t num);
+
+ uint32_t *hash;
+ uint32_t *son;
+ uint32_t cyclic_pos;
+ uint32_t cyclic_size; // Must be dictionary size + 1.
+ uint32_t hash_mask;
+
+ /// Maximum number of loops in the match finder
+ uint32_t depth;
+
+ /// Maximum length of a match that the match finder will try to find.
+ uint32_t nice_len;
+
+ /// Maximum length of a match supported by the LZ-based encoder.
+ /// If the longest match found by the match finder is nice_len,
+ /// mf_find() tries to expand it up to match_len_max bytes.
+ uint32_t match_len_max;
+
+ /// When running out of input, binary tree match finders need to know
+ /// if it is due to flushing or finishing. The action is used also
+ /// by the LZ-based encoders themselves.
+ lzma_action action;
+
+ /// Number of elements in hash[]
+ uint32_t hash_size_sum;
+
+ /// Number of elements in son[]
+ uint32_t sons_count;
+};
+
+
+typedef struct {
+ /// Extra amount of data to keep available before the "actual"
+ /// dictionary.
+ size_t before_size;
+
+ /// Size of the history buffer
+ size_t dict_size;
+
+ /// Extra amount of data to keep available after the "actual"
+ /// dictionary.
+ size_t after_size;
+
+ /// Maximum length of a match that the LZ-based encoder can accept.
+ /// This is used to extend matches of length nice_len to the
+ /// maximum possible length.
+ size_t match_len_max;
+
+ /// Match finder will search matches up to this length.
+ /// This must be less than or equal to match_len_max.
+ size_t nice_len;
+
+ /// Type of the match finder to use
+ lzma_match_finder match_finder;
+
+ /// Maximum search depth
+ uint32_t depth;
+
+ /// TODO: Comment
+ const uint8_t *preset_dict;
+
+ uint32_t preset_dict_size;
+
+} lzma_lz_options;
+
+
+// The total usable buffer space at any moment outside the match finder:
+// before_size + dict_size + after_size + match_len_max
+//
+// In reality, there's some extra space allocated to prevent the number of
+// memmove() calls reasonable. The bigger the dict_size is, the bigger
+// this extra buffer will be since with bigger dictionaries memmove() would
+// also take longer.
+//
+// A single encoder loop in the LZ-based encoder may call the match finder
+// (mf_find() or mf_skip()) at most after_size times. In other words,
+// a single encoder loop may increment lzma_mf.read_pos at most after_size
+// times. Since matches are looked up to
+// lzma_mf.buffer[lzma_mf.read_pos + match_len_max - 1], the total
+// amount of extra buffer needed after dict_size becomes
+// after_size + match_len_max.
+//
+// before_size has two uses. The first one is to keep literals available
+// in cases when the LZ-based encoder has made some read ahead.
+// TODO: Maybe this could be changed by making the LZ-based encoders to
+// store the actual literals as they do with length-distance pairs.
+//
+// Algorithms such as LZMA2 first try to compress a chunk, and then check
+// if the encoded result is smaller than the uncompressed one. If the chunk
+// was uncompressible, it is better to store it in uncompressed form in
+// the output stream. To do this, the whole uncompressed chunk has to be
+// still available in the history buffer. before_size achieves that.
+
+
+typedef struct {
+ /// Data specific to the LZ-based encoder
+ lzma_coder *coder;
+
+ /// Function to encode from *dict to out[]
+ lzma_ret (*code)(lzma_coder *restrict coder,
+ lzma_mf *restrict mf, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size);
+
+ /// Free allocated resources
+ void (*end)(lzma_coder *coder, lzma_allocator *allocator);
+
+ /// Update the options in the middle of the encoding.
+ lzma_ret (*options_update)(lzma_coder *coder,
+ const lzma_filter *filter);
+
+} lzma_lz_encoder;
+
+
+// Basic steps:
+// 1. Input gets copied into the dictionary.
+// 2. Data in dictionary gets run through the match finder byte by byte.
+// 3. The literals and matches are encoded using e.g. LZMA.
+//
+// The bytes that have been ran through the match finder, but not encoded yet,
+// are called `read ahead'.
+
+
+/// Get pointer to the first byte not ran through the match finder
+static inline const uint8_t *
+mf_ptr(const lzma_mf *mf)
+{
+ return mf->buffer + mf->read_pos;
+}
+
+
+/// Get the number of bytes that haven't been ran through the match finder yet.
+static inline uint32_t
+mf_avail(const lzma_mf *mf)
+{
+ return mf->write_pos - mf->read_pos;
+}
+
+
+/// Get the number of bytes that haven't been encoded yet (some of these
+/// bytes may have been ran through the match finder though).
+static inline uint32_t
+mf_unencoded(const lzma_mf *mf)
+{
+ return mf->write_pos - mf->read_pos + mf->read_ahead;
+}
+
+
+/// Calculate the absolute offset from the beginning of the most recent
+/// dictionary reset. Only the lowest four bits are important, so there's no
+/// problem that we don't know the 64-bit size of the data encoded so far.
+///
+/// NOTE: When moving the input window, we need to do it so that the lowest
+/// bits of dict->read_pos are not modified to keep this macro working
+/// as intended.
+static inline uint32_t
+mf_position(const lzma_mf *mf)
+{
+ return mf->read_pos - mf->read_ahead;
+}
+
+
+/// Since everything else begins with mf_, use it also for lzma_mf_find().
+#define mf_find lzma_mf_find
+
+
+/// Skip the given number of bytes. This is used when a good match was found.
+/// For example, if mf_find() finds a match of 200 bytes long, the first byte
+/// of that match was already consumed by mf_find(), and the rest 199 bytes
+/// have to be skipped with mf_skip(mf, 199).
+static inline void
+mf_skip(lzma_mf *mf, uint32_t amount)
+{
+ if (amount != 0) {
+ mf->skip(mf, amount);
+ mf->read_ahead += amount;
+ }
+}
+
+
+/// Copies at most *left number of bytes from the history buffer
+/// to out[]. This is needed by LZMA2 to encode uncompressed chunks.
+static inline void
+mf_read(lzma_mf *mf, uint8_t *out, size_t *out_pos, size_t out_size,
+ size_t *left)
+{
+ const size_t out_avail = out_size - *out_pos;
+ const size_t copy_size = my_min(out_avail, *left);
+
+ assert(mf->read_ahead == 0);
+ assert(mf->read_pos >= *left);
+
+ memcpy(out + *out_pos, mf->buffer + mf->read_pos - *left,
+ copy_size);
+
+ *out_pos += copy_size;
+ *left -= copy_size;
+ return;
+}
+
+
+extern lzma_ret lzma_lz_encoder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters,
+ lzma_ret (*lz_init)(lzma_lz_encoder *lz,
+ lzma_allocator *allocator, const void *options,
+ lzma_lz_options *lz_options));
+
+
+extern uint64_t lzma_lz_encoder_memusage(const lzma_lz_options *lz_options);
+
+
+// These are only for LZ encoder's internal use.
+extern uint32_t lzma_mf_find(
+ lzma_mf *mf, uint32_t *count, lzma_match *matches);
+
+extern uint32_t lzma_mf_hc3_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_hc3_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_hc4_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_hc4_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_bt2_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_bt2_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_bt3_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_bt3_skip(lzma_mf *dict, uint32_t amount);
+
+extern uint32_t lzma_mf_bt4_find(lzma_mf *dict, lzma_match *matches);
+extern void lzma_mf_bt4_skip(lzma_mf *dict, uint32_t amount);
+
+#endif
diff --git a/src/liblzma/lz/lz_encoder_hash.h b/src/liblzma/lz/lz_encoder_hash.h
new file mode 100644
index 0000000..c398d7d
--- /dev/null
+++ b/src/liblzma/lz/lz_encoder_hash.h
@@ -0,0 +1,108 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lz_encoder_hash.h
+/// \brief Hash macros for match finders
+//
+// Author: Igor Pavlov
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZ_ENCODER_HASH_H
+#define LZMA_LZ_ENCODER_HASH_H
+
+#if defined(WORDS_BIGENDIAN) && !defined(HAVE_SMALL)
+ // This is to make liblzma produce the same output on big endian
+ // systems that it does on little endian systems. lz_encoder.c
+ // takes care of including the actual table.
+ extern const uint32_t lzma_lz_hash_table[256];
+# define hash_table lzma_lz_hash_table
+#else
+# include "check.h"
+# define hash_table lzma_crc32_table[0]
+#endif
+
+#define HASH_2_SIZE (UINT32_C(1) << 10)
+#define HASH_3_SIZE (UINT32_C(1) << 16)
+#define HASH_4_SIZE (UINT32_C(1) << 20)
+
+#define HASH_2_MASK (HASH_2_SIZE - 1)
+#define HASH_3_MASK (HASH_3_SIZE - 1)
+#define HASH_4_MASK (HASH_4_SIZE - 1)
+
+#define FIX_3_HASH_SIZE (HASH_2_SIZE)
+#define FIX_4_HASH_SIZE (HASH_2_SIZE + HASH_3_SIZE)
+#define FIX_5_HASH_SIZE (HASH_2_SIZE + HASH_3_SIZE + HASH_4_SIZE)
+
+// Endianness doesn't matter in hash_2_calc() (no effect on the output).
+#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
+# define hash_2_calc() \
+ const uint32_t hash_value = *(const uint16_t *)(cur);
+#else
+# define hash_2_calc() \
+ const uint32_t hash_value \
+ = (uint32_t)(cur[0]) | ((uint32_t)(cur[1]) << 8)
+#endif
+
+#define hash_3_calc() \
+ const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+ const uint32_t hash_2_value = temp & HASH_2_MASK; \
+ const uint32_t hash_value \
+ = (temp ^ ((uint32_t)(cur[2]) << 8)) & mf->hash_mask
+
+#define hash_4_calc() \
+ const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+ const uint32_t hash_2_value = temp & HASH_2_MASK; \
+ const uint32_t hash_3_value \
+ = (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK; \
+ const uint32_t hash_value = (temp ^ ((uint32_t)(cur[2]) << 8) \
+ ^ (hash_table[cur[3]] << 5)) & mf->hash_mask
+
+
+// The following are not currently used.
+
+#define hash_5_calc() \
+ const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+ const uint32_t hash_2_value = temp & HASH_2_MASK; \
+ const uint32_t hash_3_value \
+ = (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK; \
+ uint32_t hash_4_value = (temp ^ ((uint32_t)(cur[2]) << 8) ^ \
+ ^ hash_table[cur[3]] << 5); \
+ const uint32_t hash_value \
+ = (hash_4_value ^ (hash_table[cur[4]] << 3)) \
+ & mf->hash_mask; \
+ hash_4_value &= HASH_4_MASK
+
+/*
+#define hash_zip_calc() \
+ const uint32_t hash_value \
+ = (((uint32_t)(cur[0]) | ((uint32_t)(cur[1]) << 8)) \
+ ^ hash_table[cur[2]]) & 0xFFFF
+*/
+
+#define hash_zip_calc() \
+ const uint32_t hash_value \
+ = (((uint32_t)(cur[2]) | ((uint32_t)(cur[0]) << 8)) \
+ ^ hash_table[cur[1]]) & 0xFFFF
+
+#define mt_hash_2_calc() \
+ const uint32_t hash_2_value \
+ = (hash_table[cur[0]] ^ cur[1]) & HASH_2_MASK
+
+#define mt_hash_3_calc() \
+ const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+ const uint32_t hash_2_value = temp & HASH_2_MASK; \
+ const uint32_t hash_3_value \
+ = (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK
+
+#define mt_hash_4_calc() \
+ const uint32_t temp = hash_table[cur[0]] ^ cur[1]; \
+ const uint32_t hash_2_value = temp & HASH_2_MASK; \
+ const uint32_t hash_3_value \
+ = (temp ^ ((uint32_t)(cur[2]) << 8)) & HASH_3_MASK; \
+ const uint32_t hash_4_value = (temp ^ ((uint32_t)(cur[2]) << 8) ^ \
+ (hash_table[cur[3]] << 5)) & HASH_4_MASK
+
+#endif
diff --git a/src/liblzma/lz/lz_encoder_hash_table.h b/src/liblzma/lz/lz_encoder_hash_table.h
new file mode 100644
index 0000000..8c51717
--- /dev/null
+++ b/src/liblzma/lz/lz_encoder_hash_table.h
@@ -0,0 +1,68 @@
+/* This file has been automatically generated by crc32_tablegen.c. */
+
+const uint32_t lzma_lz_hash_table[256] = {
+ 0x00000000, 0x77073096, 0xEE0E612C, 0x990951BA,
+ 0x076DC419, 0x706AF48F, 0xE963A535, 0x9E6495A3,
+ 0x0EDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988,
+ 0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91,
+ 0x1DB71064, 0x6AB020F2, 0xF3B97148, 0x84BE41DE,
+ 0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7,
+ 0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC,
+ 0x14015C4F, 0x63066CD9, 0xFA0F3D63, 0x8D080DF5,
+ 0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172,
+ 0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B,
+ 0x35B5A8FA, 0x42B2986C, 0xDBBBC9D6, 0xACBCF940,
+ 0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59,
+ 0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116,
+ 0x21B4F4B5, 0x56B3C423, 0xCFBA9599, 0xB8BDA50F,
+ 0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924,
+ 0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D,
+ 0x76DC4190, 0x01DB7106, 0x98D220BC, 0xEFD5102A,
+ 0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433,
+ 0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818,
+ 0x7F6A0DBB, 0x086D3D2D, 0x91646C97, 0xE6635C01,
+ 0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E,
+ 0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457,
+ 0x65B0D9C6, 0x12B7E950, 0x8BBEB8EA, 0xFCB9887C,
+ 0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65,
+ 0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2,
+ 0x4ADFA541, 0x3DD895D7, 0xA4D1C46D, 0xD3D6F4FB,
+ 0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0,
+ 0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7CC9,
+ 0x5005713C, 0x270241AA, 0xBE0B1010, 0xC90C2086,
+ 0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F,
+ 0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4,
+ 0x59B33D17, 0x2EB40D81, 0xB7BD5C3B, 0xC0BA6CAD,
+ 0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A,
+ 0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683,
+ 0xE3630B12, 0x94643B84, 0x0D6D6A3E, 0x7A6A5AA8,
+ 0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1,
+ 0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE,
+ 0xF762575D, 0x806567CB, 0x196C3671, 0x6E6B06E7,
+ 0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC,
+ 0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5,
+ 0xD6D6A3E8, 0xA1D1937E, 0x38D8C2C4, 0x4FDFF252,
+ 0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B,
+ 0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60,
+ 0xDF60EFC3, 0xA867DF55, 0x316E8EEF, 0x4669BE79,
+ 0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236,
+ 0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F,
+ 0xC5BA3BBE, 0xB2BD0B28, 0x2BB45A92, 0x5CB36A04,
+ 0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D,
+ 0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A,
+ 0x9C0906A9, 0xEB0E363F, 0x72076785, 0x05005713,
+ 0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38,
+ 0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21,
+ 0x86D3D2D4, 0xF1D4E242, 0x68DDB3F8, 0x1FDA836E,
+ 0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777,
+ 0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C,
+ 0x8F659EFF, 0xF862AE69, 0x616BFFD3, 0x166CCF45,
+ 0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2,
+ 0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB,
+ 0xAED16A4A, 0xD9D65ADC, 0x40DF0B66, 0x37D83BF0,
+ 0xA9BCAE53, 0xDEBB9EC5, 0x47B2CF7F, 0x30B5FFE9,
+ 0xBDBDF21C, 0xCABAC28A, 0x53B39330, 0x24B4A3A6,
+ 0xBAD03605, 0xCDD70693, 0x54DE5729, 0x23D967BF,
+ 0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94,
+ 0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D
+};
diff --git a/src/liblzma/lz/lz_encoder_mf.c b/src/liblzma/lz/lz_encoder_mf.c
new file mode 100644
index 0000000..f82a1c1
--- /dev/null
+++ b/src/liblzma/lz/lz_encoder_mf.c
@@ -0,0 +1,753 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lz_encoder_mf.c
+/// \brief Match finders
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lz_encoder.h"
+#include "lz_encoder_hash.h"
+
+
+/// \brief Find matches starting from the current byte
+///
+/// \return The length of the longest match found
+extern uint32_t
+lzma_mf_find(lzma_mf *mf, uint32_t *count_ptr, lzma_match *matches)
+{
+ // Call the match finder. It returns the number of length-distance
+ // pairs found.
+ // FIXME: Minimum count is zero, what _exactly_ is the maximum?
+ const uint32_t count = mf->find(mf, matches);
+
+ // Length of the longest match; assume that no matches were found
+ // and thus the maximum length is zero.
+ uint32_t len_best = 0;
+
+ if (count > 0) {
+#ifndef NDEBUG
+ // Validate the matches.
+ for (uint32_t i = 0; i < count; ++i) {
+ assert(matches[i].len <= mf->nice_len);
+ assert(matches[i].dist < mf->read_pos);
+ assert(memcmp(mf_ptr(mf) - 1,
+ mf_ptr(mf) - matches[i].dist - 2,
+ matches[i].len) == 0);
+ }
+#endif
+
+ // The last used element in the array contains
+ // the longest match.
+ len_best = matches[count - 1].len;
+
+ // If a match of maximum search length was found, try to
+ // extend the match to maximum possible length.
+ if (len_best == mf->nice_len) {
+ // The limit for the match length is either the
+ // maximum match length supported by the LZ-based
+ // encoder or the number of bytes left in the
+ // dictionary, whichever is smaller.
+ uint32_t limit = mf_avail(mf) + 1;
+ if (limit > mf->match_len_max)
+ limit = mf->match_len_max;
+
+ // Pointer to the byte we just ran through
+ // the match finder.
+ const uint8_t *p1 = mf_ptr(mf) - 1;
+
+ // Pointer to the beginning of the match. We need -1
+ // here because the match distances are zero based.
+ const uint8_t *p2 = p1 - matches[count - 1].dist - 1;
+
+ while (len_best < limit
+ && p1[len_best] == p2[len_best])
+ ++len_best;
+ }
+ }
+
+ *count_ptr = count;
+
+ // Finally update the read position to indicate that match finder was
+ // run for this dictionary offset.
+ ++mf->read_ahead;
+
+ return len_best;
+}
+
+
+/// Hash value to indicate unused element in the hash. Since we start the
+/// positions from dict_size + 1, zero is always too far to qualify
+/// as usable match position.
+#define EMPTY_HASH_VALUE 0
+
+
+/// Normalization must be done when lzma_mf.offset + lzma_mf.read_pos
+/// reaches MUST_NORMALIZE_POS.
+#define MUST_NORMALIZE_POS UINT32_MAX
+
+
+/// \brief Normalizes hash values
+///
+/// The hash arrays store positions of match candidates. The positions are
+/// relative to an arbitrary offset that is not the same as the absolute
+/// offset in the input stream. The relative position of the current byte
+/// is lzma_mf.offset + lzma_mf.read_pos. The distances of the matches are
+/// the differences of the current read position and the position found from
+/// the hash.
+///
+/// To prevent integer overflows of the offsets stored in the hash arrays,
+/// we need to "normalize" the stored values now and then. During the
+/// normalization, we drop values that indicate distance greater than the
+/// dictionary size, thus making space for new values.
+static void
+normalize(lzma_mf *mf)
+{
+ assert(mf->read_pos + mf->offset == MUST_NORMALIZE_POS);
+
+ // In future we may not want to touch the lowest bits, because there
+ // may be match finders that use larger resolution than one byte.
+ const uint32_t subvalue
+ = (MUST_NORMALIZE_POS - mf->cyclic_size);
+ // & (~(UINT32_C(1) << 10) - 1);
+
+ const uint32_t count = mf->hash_size_sum + mf->sons_count;
+ uint32_t *hash = mf->hash;
+
+ for (uint32_t i = 0; i < count; ++i) {
+ // If the distance is greater than the dictionary size,
+ // we can simply mark the hash element as empty.
+ //
+ // NOTE: Only the first mf->hash_size_sum elements are
+ // initialized for sure. There may be uninitialized elements
+ // in mf->son. Since we go through both mf->hash and
+ // mf->son here in normalization, Valgrind may complain
+ // that the "if" below depends on uninitialized value. In
+ // this case it is safe to ignore the warning. See also the
+ // comments in lz_encoder_init() in lz_encoder.c.
+ if (hash[i] <= subvalue)
+ hash[i] = EMPTY_HASH_VALUE;
+ else
+ hash[i] -= subvalue;
+ }
+
+ // Update offset to match the new locations.
+ mf->offset -= subvalue;
+
+ return;
+}
+
+
+/// Mark the current byte as processed from point of view of the match finder.
+static void
+move_pos(lzma_mf *mf)
+{
+ if (++mf->cyclic_pos == mf->cyclic_size)
+ mf->cyclic_pos = 0;
+
+ ++mf->read_pos;
+ assert(mf->read_pos <= mf->write_pos);
+
+ if (unlikely(mf->read_pos + mf->offset == UINT32_MAX))
+ normalize(mf);
+}
+
+
+/// When flushing, we cannot run the match finder unless there is nice_len
+/// bytes available in the dictionary. Instead, we skip running the match
+/// finder (indicating that no match was found), and count how many bytes we
+/// have ignored this way.
+///
+/// When new data is given after the flushing was completed, the match finder
+/// is restarted by rewinding mf->read_pos backwards by mf->pending. Then
+/// the missed bytes are added to the hash using the match finder's skip
+/// function (with small amount of input, it may start using mf->pending
+/// again if flushing).
+///
+/// Due to this rewinding, we don't touch cyclic_pos or test for
+/// normalization. It will be done when the match finder's skip function
+/// catches up after a flush.
+static void
+move_pending(lzma_mf *mf)
+{
+ ++mf->read_pos;
+ assert(mf->read_pos <= mf->write_pos);
+ ++mf->pending;
+}
+
+
+/// Calculate len_limit and determine if there is enough input to run
+/// the actual match finder code. Sets up "cur" and "pos". This macro
+/// is used by all find functions and binary tree skip functions. Hash
+/// chain skip function doesn't need len_limit so a simpler code is used
+/// in them.
+#define header(is_bt, len_min, ret_op) \
+ uint32_t len_limit = mf_avail(mf); \
+ if (mf->nice_len <= len_limit) { \
+ len_limit = mf->nice_len; \
+ } else if (len_limit < (len_min) \
+ || (is_bt && mf->action == LZMA_SYNC_FLUSH)) { \
+ assert(mf->action != LZMA_RUN); \
+ move_pending(mf); \
+ ret_op; \
+ } \
+ const uint8_t *cur = mf_ptr(mf); \
+ const uint32_t pos = mf->read_pos + mf->offset
+
+
+/// Header for find functions. "return 0" indicates that zero matches
+/// were found.
+#define header_find(is_bt, len_min) \
+ header(is_bt, len_min, return 0); \
+ uint32_t matches_count = 0
+
+
+/// Header for a loop in a skip function. "continue" tells to skip the rest
+/// of the code in the loop.
+#define header_skip(is_bt, len_min) \
+ header(is_bt, len_min, continue)
+
+
+/// Calls hc_find_func() or bt_find_func() and calculates the total number
+/// of matches found. Updates the dictionary position and returns the number
+/// of matches found.
+#define call_find(func, len_best) \
+do { \
+ matches_count = func(len_limit, pos, cur, cur_match, mf->depth, \
+ mf->son, mf->cyclic_pos, mf->cyclic_size, \
+ matches + matches_count, len_best) \
+ - matches; \
+ move_pos(mf); \
+ return matches_count; \
+} while (0)
+
+
+////////////////
+// Hash Chain //
+////////////////
+
+#if defined(HAVE_MF_HC3) || defined(HAVE_MF_HC4)
+///
+///
+/// \param len_limit Don't look for matches longer than len_limit.
+/// \param pos lzma_mf.read_pos + lzma_mf.offset
+/// \param cur Pointer to current byte (mf_ptr(mf))
+/// \param cur_match Start position of the current match candidate
+/// \param depth Maximum length of the hash chain
+/// \param son lzma_mf.son (contains the hash chain)
+/// \param cyclic_pos
+/// \param cyclic_size
+/// \param matches Array to hold the matches.
+/// \param len_best The length of the longest match found so far.
+static lzma_match *
+hc_find_func(
+ const uint32_t len_limit,
+ const uint32_t pos,
+ const uint8_t *const cur,
+ uint32_t cur_match,
+ uint32_t depth,
+ uint32_t *const son,
+ const uint32_t cyclic_pos,
+ const uint32_t cyclic_size,
+ lzma_match *matches,
+ uint32_t len_best)
+{
+ son[cyclic_pos] = cur_match;
+
+ while (true) {
+ const uint32_t delta = pos - cur_match;
+ if (depth-- == 0 || delta >= cyclic_size)
+ return matches;
+
+ const uint8_t *const pb = cur - delta;
+ cur_match = son[cyclic_pos - delta
+ + (delta > cyclic_pos ? cyclic_size : 0)];
+
+ if (pb[len_best] == cur[len_best] && pb[0] == cur[0]) {
+ uint32_t len = 0;
+ while (++len != len_limit)
+ if (pb[len] != cur[len])
+ break;
+
+ if (len_best < len) {
+ len_best = len;
+ matches->len = len;
+ matches->dist = delta - 1;
+ ++matches;
+
+ if (len == len_limit)
+ return matches;
+ }
+ }
+ }
+}
+
+
+#define hc_find(len_best) \
+ call_find(hc_find_func, len_best)
+
+
+#define hc_skip() \
+do { \
+ mf->son[mf->cyclic_pos] = cur_match; \
+ move_pos(mf); \
+} while (0)
+
+#endif
+
+
+#ifdef HAVE_MF_HC3
+extern uint32_t
+lzma_mf_hc3_find(lzma_mf *mf, lzma_match *matches)
+{
+ header_find(false, 3);
+
+ hash_3_calc();
+
+ const uint32_t delta2 = pos - mf->hash[hash_2_value];
+ const uint32_t cur_match = mf->hash[FIX_3_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;
+
+ uint32_t len_best = 2;
+
+ if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
+ for ( ; len_best != len_limit; ++len_best)
+ if (*(cur + len_best - delta2) != cur[len_best])
+ break;
+
+ matches[0].len = len_best;
+ matches[0].dist = delta2 - 1;
+ matches_count = 1;
+
+ if (len_best == len_limit) {
+ hc_skip();
+ return 1; // matches_count
+ }
+ }
+
+ hc_find(len_best);
+}
+
+
+extern void
+lzma_mf_hc3_skip(lzma_mf *mf, uint32_t amount)
+{
+ do {
+ if (mf_avail(mf) < 3) {
+ move_pending(mf);
+ continue;
+ }
+
+ const uint8_t *cur = mf_ptr(mf);
+ const uint32_t pos = mf->read_pos + mf->offset;
+
+ hash_3_calc();
+
+ const uint32_t cur_match
+ = mf->hash[FIX_3_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;
+
+ hc_skip();
+
+ } while (--amount != 0);
+}
+#endif
+
+
+#ifdef HAVE_MF_HC4
+extern uint32_t
+lzma_mf_hc4_find(lzma_mf *mf, lzma_match *matches)
+{
+ header_find(false, 4);
+
+ hash_4_calc();
+
+ uint32_t delta2 = pos - mf->hash[hash_2_value];
+ const uint32_t delta3
+ = pos - mf->hash[FIX_3_HASH_SIZE + hash_3_value];
+ const uint32_t cur_match = mf->hash[FIX_4_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value ] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
+ mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;
+
+ uint32_t len_best = 1;
+
+ if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
+ len_best = 2;
+ matches[0].len = 2;
+ matches[0].dist = delta2 - 1;
+ matches_count = 1;
+ }
+
+ if (delta2 != delta3 && delta3 < mf->cyclic_size
+ && *(cur - delta3) == *cur) {
+ len_best = 3;
+ matches[matches_count++].dist = delta3 - 1;
+ delta2 = delta3;
+ }
+
+ if (matches_count != 0) {
+ for ( ; len_best != len_limit; ++len_best)
+ if (*(cur + len_best - delta2) != cur[len_best])
+ break;
+
+ matches[matches_count - 1].len = len_best;
+
+ if (len_best == len_limit) {
+ hc_skip();
+ return matches_count;
+ }
+ }
+
+ if (len_best < 3)
+ len_best = 3;
+
+ hc_find(len_best);
+}
+
+
+extern void
+lzma_mf_hc4_skip(lzma_mf *mf, uint32_t amount)
+{
+ do {
+ if (mf_avail(mf) < 4) {
+ move_pending(mf);
+ continue;
+ }
+
+ const uint8_t *cur = mf_ptr(mf);
+ const uint32_t pos = mf->read_pos + mf->offset;
+
+ hash_4_calc();
+
+ const uint32_t cur_match
+ = mf->hash[FIX_4_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
+ mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;
+
+ hc_skip();
+
+ } while (--amount != 0);
+}
+#endif
+
+
+/////////////////
+// Binary Tree //
+/////////////////
+
+#if defined(HAVE_MF_BT2) || defined(HAVE_MF_BT3) || defined(HAVE_MF_BT4)
+static lzma_match *
+bt_find_func(
+ const uint32_t len_limit,
+ const uint32_t pos,
+ const uint8_t *const cur,
+ uint32_t cur_match,
+ uint32_t depth,
+ uint32_t *const son,
+ const uint32_t cyclic_pos,
+ const uint32_t cyclic_size,
+ lzma_match *matches,
+ uint32_t len_best)
+{
+ uint32_t *ptr0 = son + (cyclic_pos << 1) + 1;
+ uint32_t *ptr1 = son + (cyclic_pos << 1);
+
+ uint32_t len0 = 0;
+ uint32_t len1 = 0;
+
+ while (true) {
+ const uint32_t delta = pos - cur_match;
+ if (depth-- == 0 || delta >= cyclic_size) {
+ *ptr0 = EMPTY_HASH_VALUE;
+ *ptr1 = EMPTY_HASH_VALUE;
+ return matches;
+ }
+
+ uint32_t *const pair = son + ((cyclic_pos - delta
+ + (delta > cyclic_pos ? cyclic_size : 0))
+ << 1);
+
+ const uint8_t *const pb = cur - delta;
+ uint32_t len = my_min(len0, len1);
+
+ if (pb[len] == cur[len]) {
+ while (++len != len_limit)
+ if (pb[len] != cur[len])
+ break;
+
+ if (len_best < len) {
+ len_best = len;
+ matches->len = len;
+ matches->dist = delta - 1;
+ ++matches;
+
+ if (len == len_limit) {
+ *ptr1 = pair[0];
+ *ptr0 = pair[1];
+ return matches;
+ }
+ }
+ }
+
+ if (pb[len] < cur[len]) {
+ *ptr1 = cur_match;
+ ptr1 = pair + 1;
+ cur_match = *ptr1;
+ len1 = len;
+ } else {
+ *ptr0 = cur_match;
+ ptr0 = pair;
+ cur_match = *ptr0;
+ len0 = len;
+ }
+ }
+}
+
+
+static void
+bt_skip_func(
+ const uint32_t len_limit,
+ const uint32_t pos,
+ const uint8_t *const cur,
+ uint32_t cur_match,
+ uint32_t depth,
+ uint32_t *const son,
+ const uint32_t cyclic_pos,
+ const uint32_t cyclic_size)
+{
+ uint32_t *ptr0 = son + (cyclic_pos << 1) + 1;
+ uint32_t *ptr1 = son + (cyclic_pos << 1);
+
+ uint32_t len0 = 0;
+ uint32_t len1 = 0;
+
+ while (true) {
+ const uint32_t delta = pos - cur_match;
+ if (depth-- == 0 || delta >= cyclic_size) {
+ *ptr0 = EMPTY_HASH_VALUE;
+ *ptr1 = EMPTY_HASH_VALUE;
+ return;
+ }
+
+ uint32_t *pair = son + ((cyclic_pos - delta
+ + (delta > cyclic_pos ? cyclic_size : 0))
+ << 1);
+ const uint8_t *pb = cur - delta;
+ uint32_t len = my_min(len0, len1);
+
+ if (pb[len] == cur[len]) {
+ while (++len != len_limit)
+ if (pb[len] != cur[len])
+ break;
+
+ if (len == len_limit) {
+ *ptr1 = pair[0];
+ *ptr0 = pair[1];
+ return;
+ }
+ }
+
+ if (pb[len] < cur[len]) {
+ *ptr1 = cur_match;
+ ptr1 = pair + 1;
+ cur_match = *ptr1;
+ len1 = len;
+ } else {
+ *ptr0 = cur_match;
+ ptr0 = pair;
+ cur_match = *ptr0;
+ len0 = len;
+ }
+ }
+}
+
+
+#define bt_find(len_best) \
+ call_find(bt_find_func, len_best)
+
+#define bt_skip() \
+do { \
+ bt_skip_func(len_limit, pos, cur, cur_match, mf->depth, \
+ mf->son, mf->cyclic_pos, \
+ mf->cyclic_size); \
+ move_pos(mf); \
+} while (0)
+
+#endif
+
+
+#ifdef HAVE_MF_BT2
+extern uint32_t
+lzma_mf_bt2_find(lzma_mf *mf, lzma_match *matches)
+{
+ header_find(true, 2);
+
+ hash_2_calc();
+
+ const uint32_t cur_match = mf->hash[hash_value];
+ mf->hash[hash_value] = pos;
+
+ bt_find(1);
+}
+
+
+extern void
+lzma_mf_bt2_skip(lzma_mf *mf, uint32_t amount)
+{
+ do {
+ header_skip(true, 2);
+
+ hash_2_calc();
+
+ const uint32_t cur_match = mf->hash[hash_value];
+ mf->hash[hash_value] = pos;
+
+ bt_skip();
+
+ } while (--amount != 0);
+}
+#endif
+
+
+#ifdef HAVE_MF_BT3
+extern uint32_t
+lzma_mf_bt3_find(lzma_mf *mf, lzma_match *matches)
+{
+ header_find(true, 3);
+
+ hash_3_calc();
+
+ const uint32_t delta2 = pos - mf->hash[hash_2_value];
+ const uint32_t cur_match = mf->hash[FIX_3_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;
+
+ uint32_t len_best = 2;
+
+ if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
+ for ( ; len_best != len_limit; ++len_best)
+ if (*(cur + len_best - delta2) != cur[len_best])
+ break;
+
+ matches[0].len = len_best;
+ matches[0].dist = delta2 - 1;
+ matches_count = 1;
+
+ if (len_best == len_limit) {
+ bt_skip();
+ return 1; // matches_count
+ }
+ }
+
+ bt_find(len_best);
+}
+
+
+extern void
+lzma_mf_bt3_skip(lzma_mf *mf, uint32_t amount)
+{
+ do {
+ header_skip(true, 3);
+
+ hash_3_calc();
+
+ const uint32_t cur_match
+ = mf->hash[FIX_3_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;
+
+ bt_skip();
+
+ } while (--amount != 0);
+}
+#endif
+
+
+#ifdef HAVE_MF_BT4
+extern uint32_t
+lzma_mf_bt4_find(lzma_mf *mf, lzma_match *matches)
+{
+ header_find(true, 4);
+
+ hash_4_calc();
+
+ uint32_t delta2 = pos - mf->hash[hash_2_value];
+ const uint32_t delta3
+ = pos - mf->hash[FIX_3_HASH_SIZE + hash_3_value];
+ const uint32_t cur_match = mf->hash[FIX_4_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
+ mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;
+
+ uint32_t len_best = 1;
+
+ if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
+ len_best = 2;
+ matches[0].len = 2;
+ matches[0].dist = delta2 - 1;
+ matches_count = 1;
+ }
+
+ if (delta2 != delta3 && delta3 < mf->cyclic_size
+ && *(cur - delta3) == *cur) {
+ len_best = 3;
+ matches[matches_count++].dist = delta3 - 1;
+ delta2 = delta3;
+ }
+
+ if (matches_count != 0) {
+ for ( ; len_best != len_limit; ++len_best)
+ if (*(cur + len_best - delta2) != cur[len_best])
+ break;
+
+ matches[matches_count - 1].len = len_best;
+
+ if (len_best == len_limit) {
+ bt_skip();
+ return matches_count;
+ }
+ }
+
+ if (len_best < 3)
+ len_best = 3;
+
+ bt_find(len_best);
+}
+
+
+extern void
+lzma_mf_bt4_skip(lzma_mf *mf, uint32_t amount)
+{
+ do {
+ header_skip(true, 4);
+
+ hash_4_calc();
+
+ const uint32_t cur_match
+ = mf->hash[FIX_4_HASH_SIZE + hash_value];
+
+ mf->hash[hash_2_value] = pos;
+ mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
+ mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;
+
+ bt_skip();
+
+ } while (--amount != 0);
+}
+#endif
diff --git a/src/liblzma/lzma/Makefile.inc b/src/liblzma/lzma/Makefile.inc
new file mode 100644
index 0000000..7fc4d17
--- /dev/null
+++ b/src/liblzma/lzma/Makefile.inc
@@ -0,0 +1,43 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+EXTRA_DIST += lzma/fastpos_tablegen.c
+
+liblzma_la_SOURCES += lzma/lzma_common.h
+
+if COND_ENCODER_LZMA1
+liblzma_la_SOURCES += \
+ lzma/fastpos.h \
+ lzma/lzma_encoder.h \
+ lzma/lzma_encoder.c \
+ lzma/lzma_encoder_presets.c \
+ lzma/lzma_encoder_private.h \
+ lzma/lzma_encoder_optimum_fast.c \
+ lzma/lzma_encoder_optimum_normal.c
+
+if !COND_SMALL
+liblzma_la_SOURCES += lzma/fastpos_table.c
+endif
+endif
+
+if COND_DECODER_LZMA1
+liblzma_la_SOURCES += \
+ lzma/lzma_decoder.c \
+ lzma/lzma_decoder.h
+endif
+
+if COND_ENCODER_LZMA2
+liblzma_la_SOURCES += \
+ lzma/lzma2_encoder.c \
+ lzma/lzma2_encoder.h
+endif
+
+if COND_DECODER_LZMA2
+liblzma_la_SOURCES += \
+ lzma/lzma2_decoder.c \
+ lzma/lzma2_decoder.h
+endif
diff --git a/src/liblzma/lzma/fastpos.h b/src/liblzma/lzma/fastpos.h
new file mode 100644
index 0000000..4aea231
--- /dev/null
+++ b/src/liblzma/lzma/fastpos.h
@@ -0,0 +1,140 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file fastpos.h
+/// \brief Kind of two-bit version of bit scan reverse
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_FASTPOS_H
+#define LZMA_FASTPOS_H
+
+// LZMA encodes match distances (positions) by storing the highest two
+// bits using a six-bit value [0, 63], and then the missing lower bits.
+// Dictionary size is also stored using this encoding in the new .lzma
+// file format header.
+//
+// fastpos.h provides a way to quickly find out the correct six-bit
+// values. The following table gives some examples of this encoding:
+//
+// pos return
+// 0 0
+// 1 1
+// 2 2
+// 3 3
+// 4 4
+// 5 4
+// 6 5
+// 7 5
+// 8 6
+// 11 6
+// 12 7
+// ... ...
+// 15 7
+// 16 8
+// 17 8
+// ... ...
+// 23 8
+// 24 9
+// 25 9
+// ... ...
+//
+//
+// Provided functions or macros
+// ----------------------------
+//
+// get_pos_slot(pos) is the basic version. get_pos_slot_2(pos)
+// assumes that pos >= FULL_DISTANCES, thus the result is at least
+// FULL_DISTANCES_BITS * 2. Using get_pos_slot(pos) instead of
+// get_pos_slot_2(pos) would give the same result, but get_pos_slot_2(pos)
+// should be tiny bit faster due to the assumption being made.
+//
+//
+// Size vs. speed
+// --------------
+//
+// With some CPUs that have fast BSR (bit scan reverse) instruction, the
+// size optimized version is slightly faster than the bigger table based
+// approach. Such CPUs include Intel Pentium Pro, Pentium II, Pentium III
+// and Core 2 (possibly others). AMD K7 seems to have slower BSR, but that
+// would still have speed roughly comparable to the table version. Older
+// x86 CPUs like the original Pentium have very slow BSR; on those systems
+// the table version is a lot faster.
+//
+// On some CPUs, the table version is a lot faster when using position
+// dependent code, but with position independent code the size optimized
+// version is slightly faster. This occurs at least on 32-bit SPARC (no
+// ASM optimizations).
+//
+// I'm making the table version the default, because that has good speed
+// on all systems I have tried. The size optimized version is sometimes
+// slightly faster, but sometimes it is a lot slower.
+
+#ifdef HAVE_SMALL
+# define get_pos_slot(pos) ((pos) <= 4 ? (pos) : get_pos_slot_2(pos))
+
+static inline uint32_t
+get_pos_slot_2(uint32_t pos)
+{
+ const uint32_t i = bsr32(pos);
+ return (i + i) + ((pos >> (i - 1)) & 1);
+}
+
+
+#else
+
+#define FASTPOS_BITS 13
+
+extern const uint8_t lzma_fastpos[1 << FASTPOS_BITS];
+
+
+#define fastpos_shift(extra, n) \
+ ((extra) + (n) * (FASTPOS_BITS - 1))
+
+#define fastpos_limit(extra, n) \
+ (UINT32_C(1) << (FASTPOS_BITS + fastpos_shift(extra, n)))
+
+#define fastpos_result(pos, extra, n) \
+ lzma_fastpos[(pos) >> fastpos_shift(extra, n)] \
+ + 2 * fastpos_shift(extra, n)
+
+
+static inline uint32_t
+get_pos_slot(uint32_t pos)
+{
+ // If it is small enough, we can pick the result directly from
+ // the precalculated table.
+ if (pos < fastpos_limit(0, 0))
+ return lzma_fastpos[pos];
+
+ if (pos < fastpos_limit(0, 1))
+ return fastpos_result(pos, 0, 1);
+
+ return fastpos_result(pos, 0, 2);
+}
+
+
+#ifdef FULL_DISTANCES_BITS
+static inline uint32_t
+get_pos_slot_2(uint32_t pos)
+{
+ assert(pos >= FULL_DISTANCES);
+
+ if (pos < fastpos_limit(FULL_DISTANCES_BITS - 1, 0))
+ return fastpos_result(pos, FULL_DISTANCES_BITS - 1, 0);
+
+ if (pos < fastpos_limit(FULL_DISTANCES_BITS - 1, 1))
+ return fastpos_result(pos, FULL_DISTANCES_BITS - 1, 1);
+
+ return fastpos_result(pos, FULL_DISTANCES_BITS - 1, 2);
+}
+#endif
+
+#endif
+
+#endif
diff --git a/src/liblzma/lzma/fastpos_table.c b/src/liblzma/lzma/fastpos_table.c
new file mode 100644
index 0000000..6a3ceac
--- /dev/null
+++ b/src/liblzma/lzma/fastpos_table.c
@@ -0,0 +1,519 @@
+/* This file has been automatically generated by fastpos_tablegen.c. */
+
+#include "common.h"
+#include "fastpos.h"
+
+const uint8_t lzma_fastpos[1 << FASTPOS_BITS] = {
+ 0, 1, 2, 3, 4, 4, 5, 5, 6, 6, 6, 6, 7, 7, 7, 7,
+ 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9,
+ 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10,
+ 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11,
+ 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12,
+ 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12,
+ 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13,
+ 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13,
+ 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
+ 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
+ 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
+ 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
+ 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,
+ 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,
+ 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,
+ 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+ 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25
+};
diff --git a/src/liblzma/lzma/fastpos_tablegen.c b/src/liblzma/lzma/fastpos_tablegen.c
new file mode 100644
index 0000000..c97e6f4
--- /dev/null
+++ b/src/liblzma/lzma/fastpos_tablegen.c
@@ -0,0 +1,56 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file fastpos_tablegen.c
+/// \brief Generates the lzma_fastpos[] lookup table
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include <sys/types.h>
+#include <inttypes.h>
+#include <stdio.h>
+#include "fastpos.h"
+
+
+int
+main(void)
+{
+ uint8_t fastpos[1 << FASTPOS_BITS];
+
+ const uint8_t fast_slots = 2 * FASTPOS_BITS;
+ uint32_t c = 2;
+
+ fastpos[0] = 0;
+ fastpos[1] = 1;
+
+ for (uint8_t slot_fast = 2; slot_fast < fast_slots; ++slot_fast) {
+ const uint32_t k = 1 << ((slot_fast >> 1) - 1);
+ for (uint32_t j = 0; j < k; ++j, ++c)
+ fastpos[c] = slot_fast;
+ }
+
+ printf("/* This file has been automatically generated "
+ "by fastpos_tablegen.c. */\n\n"
+ "#include \"common.h\"\n"
+ "#include \"fastpos.h\"\n\n"
+ "const uint8_t lzma_fastpos[1 << FASTPOS_BITS] = {");
+
+ for (size_t i = 0; i < (1 << FASTPOS_BITS); ++i) {
+ if (i % 16 == 0)
+ printf("\n\t");
+
+ printf("%3u", (unsigned int)(fastpos[i]));
+
+ if (i != (1 << FASTPOS_BITS) - 1)
+ printf(",");
+ }
+
+ printf("\n};\n");
+
+ return 0;
+}
diff --git a/src/liblzma/lzma/lzma2_decoder.c b/src/liblzma/lzma/lzma2_decoder.c
new file mode 100644
index 0000000..b4c2f2d
--- /dev/null
+++ b/src/liblzma/lzma/lzma2_decoder.c
@@ -0,0 +1,305 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma2_decoder.c
+/// \brief LZMA2 decoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lzma2_decoder.h"
+#include "lz_decoder.h"
+#include "lzma_decoder.h"
+
+
+struct lzma_coder_s {
+ enum sequence {
+ SEQ_CONTROL,
+ SEQ_UNCOMPRESSED_1,
+ SEQ_UNCOMPRESSED_2,
+ SEQ_COMPRESSED_0,
+ SEQ_COMPRESSED_1,
+ SEQ_PROPERTIES,
+ SEQ_LZMA,
+ SEQ_COPY,
+ } sequence;
+
+ /// Sequence after the size fields have been decoded.
+ enum sequence next_sequence;
+
+ /// LZMA decoder
+ lzma_lz_decoder lzma;
+
+ /// Uncompressed size of LZMA chunk
+ size_t uncompressed_size;
+
+ /// Compressed size of the chunk (naturally equals to uncompressed
+ /// size of uncompressed chunk)
+ size_t compressed_size;
+
+ /// True if properties are needed. This is false before the
+ /// first LZMA chunk.
+ bool need_properties;
+
+ /// True if dictionary reset is needed. This is false before the
+ /// first chunk (LZMA or uncompressed).
+ bool need_dictionary_reset;
+
+ lzma_options_lzma options;
+};
+
+
+static lzma_ret
+lzma2_decode(lzma_coder *restrict coder, lzma_dict *restrict dict,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size)
+{
+ // With SEQ_LZMA it is possible that no new input is needed to do
+ // some progress. The rest of the sequences assume that there is
+ // at least one byte of input.
+ while (*in_pos < in_size || coder->sequence == SEQ_LZMA)
+ switch (coder->sequence) {
+ case SEQ_CONTROL: {
+ const uint32_t control = in[*in_pos];
+ ++*in_pos;
+
+ if (control >= 0xE0 || control == 1) {
+ // Dictionary reset implies that next LZMA chunk has
+ // to set new properties.
+ coder->need_properties = true;
+ coder->need_dictionary_reset = true;
+ } else if (coder->need_dictionary_reset) {
+ return LZMA_DATA_ERROR;
+ }
+
+ if (control >= 0x80) {
+ // LZMA chunk. The highest five bits of the
+ // uncompressed size are taken from the control byte.
+ coder->uncompressed_size = (control & 0x1F) << 16;
+ coder->sequence = SEQ_UNCOMPRESSED_1;
+
+ // See if there are new properties or if we need to
+ // reset the state.
+ if (control >= 0xC0) {
+ // When there are new properties, state reset
+ // is done at SEQ_PROPERTIES.
+ coder->need_properties = false;
+ coder->next_sequence = SEQ_PROPERTIES;
+
+ } else if (coder->need_properties) {
+ return LZMA_DATA_ERROR;
+
+ } else {
+ coder->next_sequence = SEQ_LZMA;
+
+ // If only state reset is wanted with old
+ // properties, do the resetting here for
+ // simplicity.
+ if (control >= 0xA0)
+ coder->lzma.reset(coder->lzma.coder,
+ &coder->options);
+ }
+ } else {
+ // End marker
+ if (control == 0x00)
+ return LZMA_STREAM_END;
+
+ // Invalid control values
+ if (control > 2)
+ return LZMA_DATA_ERROR;
+
+ // It's uncompressed chunk
+ coder->sequence = SEQ_COMPRESSED_0;
+ coder->next_sequence = SEQ_COPY;
+ }
+
+ if (coder->need_dictionary_reset) {
+ // Finish the dictionary reset and let the caller
+ // flush the dictionary to the actual output buffer.
+ coder->need_dictionary_reset = false;
+ dict_reset(dict);
+ return LZMA_OK;
+ }
+
+ break;
+ }
+
+ case SEQ_UNCOMPRESSED_1:
+ coder->uncompressed_size += (uint32_t)(in[(*in_pos)++]) << 8;
+ coder->sequence = SEQ_UNCOMPRESSED_2;
+ break;
+
+ case SEQ_UNCOMPRESSED_2:
+ coder->uncompressed_size += in[(*in_pos)++] + 1;
+ coder->sequence = SEQ_COMPRESSED_0;
+ coder->lzma.set_uncompressed(coder->lzma.coder,
+ coder->uncompressed_size);
+ break;
+
+ case SEQ_COMPRESSED_0:
+ coder->compressed_size = (uint32_t)(in[(*in_pos)++]) << 8;
+ coder->sequence = SEQ_COMPRESSED_1;
+ break;
+
+ case SEQ_COMPRESSED_1:
+ coder->compressed_size += in[(*in_pos)++] + 1;
+ coder->sequence = coder->next_sequence;
+ break;
+
+ case SEQ_PROPERTIES:
+ if (lzma_lzma_lclppb_decode(&coder->options, in[(*in_pos)++]))
+ return LZMA_DATA_ERROR;
+
+ coder->lzma.reset(coder->lzma.coder, &coder->options);
+
+ coder->sequence = SEQ_LZMA;
+ break;
+
+ case SEQ_LZMA: {
+ // Store the start offset so that we can update
+ // coder->compressed_size later.
+ const size_t in_start = *in_pos;
+
+ // Decode from in[] to *dict.
+ const lzma_ret ret = coder->lzma.code(coder->lzma.coder,
+ dict, in, in_pos, in_size);
+
+ // Validate and update coder->compressed_size.
+ const size_t in_used = *in_pos - in_start;
+ if (in_used > coder->compressed_size)
+ return LZMA_DATA_ERROR;
+
+ coder->compressed_size -= in_used;
+
+ // Return if we didn't finish the chunk, or an error occurred.
+ if (ret != LZMA_STREAM_END)
+ return ret;
+
+ // The LZMA decoder must have consumed the whole chunk now.
+ // We don't need to worry about uncompressed size since it
+ // is checked by the LZMA decoder.
+ if (coder->compressed_size != 0)
+ return LZMA_DATA_ERROR;
+
+ coder->sequence = SEQ_CONTROL;
+ break;
+ }
+
+ case SEQ_COPY: {
+ // Copy from input to the dictionary as is.
+ // FIXME Can copy too much?
+ dict_write(dict, in, in_pos, in_size, &coder->compressed_size);
+ if (coder->compressed_size != 0)
+ return LZMA_OK;
+
+ coder->sequence = SEQ_CONTROL;
+ break;
+ }
+
+ default:
+ assert(0);
+ return LZMA_PROG_ERROR;
+ }
+
+ return LZMA_OK;
+}
+
+
+static void
+lzma2_decoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ assert(coder->lzma.end == NULL);
+ lzma_free(coder->lzma.coder, allocator);
+
+ lzma_free(coder, allocator);
+
+ return;
+}
+
+
+static lzma_ret
+lzma2_decoder_init(lzma_lz_decoder *lz, lzma_allocator *allocator,
+ const void *opt, lzma_lz_options *lz_options)
+{
+ if (lz->coder == NULL) {
+ lz->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (lz->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ lz->code = &lzma2_decode;
+ lz->end = &lzma2_decoder_end;
+
+ lz->coder->lzma = LZMA_LZ_DECODER_INIT;
+ }
+
+ const lzma_options_lzma *options = opt;
+
+ lz->coder->sequence = SEQ_CONTROL;
+ lz->coder->need_properties = true;
+ lz->coder->need_dictionary_reset = options->preset_dict == NULL
+ || options->preset_dict_size == 0;
+
+ return lzma_lzma_decoder_create(&lz->coder->lzma,
+ allocator, options, lz_options);
+}
+
+
+extern lzma_ret
+lzma_lzma2_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ // LZMA2 can only be the last filter in the chain. This is enforced
+ // by the raw_decoder initialization.
+ assert(filters[1].init == NULL);
+
+ return lzma_lz_decoder_init(next, allocator, filters,
+ &lzma2_decoder_init);
+}
+
+
+extern uint64_t
+lzma_lzma2_decoder_memusage(const void *options)
+{
+ return sizeof(lzma_coder)
+ + lzma_lzma_decoder_memusage_nocheck(options);
+}
+
+
+extern lzma_ret
+lzma_lzma2_props_decode(void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size)
+{
+ if (props_size != 1)
+ return LZMA_OPTIONS_ERROR;
+
+ // Check that reserved bits are unset.
+ if (props[0] & 0xC0)
+ return LZMA_OPTIONS_ERROR;
+
+ // Decode the dictionary size.
+ if (props[0] > 40)
+ return LZMA_OPTIONS_ERROR;
+
+ lzma_options_lzma *opt = lzma_alloc(
+ sizeof(lzma_options_lzma), allocator);
+ if (opt == NULL)
+ return LZMA_MEM_ERROR;
+
+ if (props[0] == 40) {
+ opt->dict_size = UINT32_MAX;
+ } else {
+ opt->dict_size = 2 | (props[0] & 1);
+ opt->dict_size <<= props[0] / 2 + 11;
+ }
+
+ opt->preset_dict = NULL;
+ opt->preset_dict_size = 0;
+
+ *options = opt;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/lzma/lzma2_decoder.h b/src/liblzma/lzma/lzma2_decoder.h
new file mode 100644
index 0000000..fac4ac4
--- /dev/null
+++ b/src/liblzma/lzma/lzma2_decoder.h
@@ -0,0 +1,28 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma2_decoder.h
+/// \brief LZMA2 decoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZMA2_DECODER_H
+#define LZMA_LZMA2_DECODER_H
+
+#include "common.h"
+
+extern lzma_ret lzma_lzma2_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern uint64_t lzma_lzma2_decoder_memusage(const void *options);
+
+extern lzma_ret lzma_lzma2_props_decode(
+ void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size);
+
+#endif
diff --git a/src/liblzma/lzma/lzma2_encoder.c b/src/liblzma/lzma/lzma2_encoder.c
new file mode 100644
index 0000000..b48e0d6
--- /dev/null
+++ b/src/liblzma/lzma/lzma2_encoder.c
@@ -0,0 +1,393 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma2_encoder.c
+/// \brief LZMA2 encoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lz_encoder.h"
+#include "lzma_encoder.h"
+#include "fastpos.h"
+#include "lzma2_encoder.h"
+
+
+struct lzma_coder_s {
+ enum {
+ SEQ_INIT,
+ SEQ_LZMA_ENCODE,
+ SEQ_LZMA_COPY,
+ SEQ_UNCOMPRESSED_HEADER,
+ SEQ_UNCOMPRESSED_COPY,
+ } sequence;
+
+ /// LZMA encoder
+ lzma_coder *lzma;
+
+ /// LZMA options currently in use.
+ lzma_options_lzma opt_cur;
+
+ bool need_properties;
+ bool need_state_reset;
+ bool need_dictionary_reset;
+
+ /// Uncompressed size of a chunk
+ size_t uncompressed_size;
+
+ /// Compressed size of a chunk (excluding headers); this is also used
+ /// to indicate the end of buf[] in SEQ_LZMA_COPY.
+ size_t compressed_size;
+
+ /// Read position in buf[]
+ size_t buf_pos;
+
+ /// Buffer to hold the chunk header and LZMA compressed data
+ uint8_t buf[LZMA2_HEADER_MAX + LZMA2_CHUNK_MAX];
+};
+
+
+static void
+lzma2_header_lzma(lzma_coder *coder)
+{
+ assert(coder->uncompressed_size > 0);
+ assert(coder->uncompressed_size <= LZMA2_UNCOMPRESSED_MAX);
+ assert(coder->compressed_size > 0);
+ assert(coder->compressed_size <= LZMA2_CHUNK_MAX);
+
+ size_t pos;
+
+ if (coder->need_properties) {
+ pos = 0;
+
+ if (coder->need_dictionary_reset)
+ coder->buf[pos] = 0x80 + (3 << 5);
+ else
+ coder->buf[pos] = 0x80 + (2 << 5);
+ } else {
+ pos = 1;
+
+ if (coder->need_state_reset)
+ coder->buf[pos] = 0x80 + (1 << 5);
+ else
+ coder->buf[pos] = 0x80;
+ }
+
+ // Set the start position for copying.
+ coder->buf_pos = pos;
+
+ // Uncompressed size
+ size_t size = coder->uncompressed_size - 1;
+ coder->buf[pos++] += size >> 16;
+ coder->buf[pos++] = (size >> 8) & 0xFF;
+ coder->buf[pos++] = size & 0xFF;
+
+ // Compressed size
+ size = coder->compressed_size - 1;
+ coder->buf[pos++] = size >> 8;
+ coder->buf[pos++] = size & 0xFF;
+
+ // Properties, if needed
+ if (coder->need_properties)
+ lzma_lzma_lclppb_encode(&coder->opt_cur, coder->buf + pos);
+
+ coder->need_properties = false;
+ coder->need_state_reset = false;
+ coder->need_dictionary_reset = false;
+
+ // The copying code uses coder->compressed_size to indicate the end
+ // of coder->buf[], so we need add the maximum size of the header here.
+ coder->compressed_size += LZMA2_HEADER_MAX;
+
+ return;
+}
+
+
+static void
+lzma2_header_uncompressed(lzma_coder *coder)
+{
+ assert(coder->uncompressed_size > 0);
+ assert(coder->uncompressed_size <= LZMA2_CHUNK_MAX);
+
+ // If this is the first chunk, we need to include dictionary
+ // reset indicator.
+ if (coder->need_dictionary_reset)
+ coder->buf[0] = 1;
+ else
+ coder->buf[0] = 2;
+
+ coder->need_dictionary_reset = false;
+
+ // "Compressed" size
+ coder->buf[1] = (coder->uncompressed_size - 1) >> 8;
+ coder->buf[2] = (coder->uncompressed_size - 1) & 0xFF;
+
+ // Set the start position for copying.
+ coder->buf_pos = 0;
+ return;
+}
+
+
+static lzma_ret
+lzma2_encode(lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint8_t *restrict out, size_t *restrict out_pos,
+ size_t out_size)
+{
+ while (*out_pos < out_size)
+ switch (coder->sequence) {
+ case SEQ_INIT:
+ // If there's no input left and we are flushing or finishing,
+ // don't start a new chunk.
+ if (mf_unencoded(mf) == 0) {
+ // Write end of payload marker if finishing.
+ if (mf->action == LZMA_FINISH)
+ out[(*out_pos)++] = 0;
+
+ return mf->action == LZMA_RUN
+ ? LZMA_OK : LZMA_STREAM_END;
+ }
+
+ if (coder->need_state_reset)
+ return_if_error(lzma_lzma_encoder_reset(
+ coder->lzma, &coder->opt_cur));
+
+ coder->uncompressed_size = 0;
+ coder->compressed_size = 0;
+ coder->sequence = SEQ_LZMA_ENCODE;
+
+ // Fall through
+
+ case SEQ_LZMA_ENCODE: {
+ // Calculate how much more uncompressed data this chunk
+ // could accept.
+ const uint32_t left = LZMA2_UNCOMPRESSED_MAX
+ - coder->uncompressed_size;
+ uint32_t limit;
+
+ if (left < mf->match_len_max) {
+ // Must flush immediately since the next LZMA symbol
+ // could make the uncompressed size of the chunk too
+ // big.
+ limit = 0;
+ } else {
+ // Calculate maximum read_limit that is OK from point
+ // of view of LZMA2 chunk size.
+ limit = mf->read_pos - mf->read_ahead
+ + left - mf->match_len_max;
+ }
+
+ // Save the start position so that we can update
+ // coder->uncompressed_size.
+ const uint32_t read_start = mf->read_pos - mf->read_ahead;
+
+ // Call the LZMA encoder until the chunk is finished.
+ const lzma_ret ret = lzma_lzma_encode(coder->lzma, mf,
+ coder->buf + LZMA2_HEADER_MAX,
+ &coder->compressed_size,
+ LZMA2_CHUNK_MAX, limit);
+
+ coder->uncompressed_size += mf->read_pos - mf->read_ahead
+ - read_start;
+
+ assert(coder->compressed_size <= LZMA2_CHUNK_MAX);
+ assert(coder->uncompressed_size <= LZMA2_UNCOMPRESSED_MAX);
+
+ if (ret != LZMA_STREAM_END)
+ return LZMA_OK;
+
+ // See if the chunk compressed. If it didn't, we encode it
+ // as uncompressed chunk. This saves a few bytes of space
+ // and makes decoding faster.
+ if (coder->compressed_size >= coder->uncompressed_size) {
+ coder->uncompressed_size += mf->read_ahead;
+ assert(coder->uncompressed_size
+ <= LZMA2_UNCOMPRESSED_MAX);
+ mf->read_ahead = 0;
+ lzma2_header_uncompressed(coder);
+ coder->need_state_reset = true;
+ coder->sequence = SEQ_UNCOMPRESSED_HEADER;
+ break;
+ }
+
+ // The chunk did compress at least by one byte, so we store
+ // the chunk as LZMA.
+ lzma2_header_lzma(coder);
+
+ coder->sequence = SEQ_LZMA_COPY;
+ }
+
+ // Fall through
+
+ case SEQ_LZMA_COPY:
+ // Copy the compressed chunk along its headers to the
+ // output buffer.
+ lzma_bufcpy(coder->buf, &coder->buf_pos,
+ coder->compressed_size,
+ out, out_pos, out_size);
+ if (coder->buf_pos != coder->compressed_size)
+ return LZMA_OK;
+
+ coder->sequence = SEQ_INIT;
+ break;
+
+ case SEQ_UNCOMPRESSED_HEADER:
+ // Copy the three-byte header to indicate uncompressed chunk.
+ lzma_bufcpy(coder->buf, &coder->buf_pos,
+ LZMA2_HEADER_UNCOMPRESSED,
+ out, out_pos, out_size);
+ if (coder->buf_pos != LZMA2_HEADER_UNCOMPRESSED)
+ return LZMA_OK;
+
+ coder->sequence = SEQ_UNCOMPRESSED_COPY;
+
+ // Fall through
+
+ case SEQ_UNCOMPRESSED_COPY:
+ // Copy the uncompressed data as is from the dictionary
+ // to the output buffer.
+ mf_read(mf, out, out_pos, out_size, &coder->uncompressed_size);
+ if (coder->uncompressed_size != 0)
+ return LZMA_OK;
+
+ coder->sequence = SEQ_INIT;
+ break;
+ }
+
+ return LZMA_OK;
+}
+
+
+static void
+lzma2_encoder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_free(coder->lzma, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+lzma2_encoder_options_update(lzma_coder *coder, const lzma_filter *filter)
+{
+ // New options can be set only when there is no incomplete chunk.
+ // This is the case at the beginning of the raw stream and right
+ // after LZMA_SYNC_FLUSH.
+ if (filter->options == NULL || coder->sequence != SEQ_INIT)
+ return LZMA_PROG_ERROR;
+
+ // Look if there are new options. At least for now,
+ // only lc/lp/pb can be changed.
+ const lzma_options_lzma *opt = filter->options;
+ if (coder->opt_cur.lc != opt->lc || coder->opt_cur.lp != opt->lp
+ || coder->opt_cur.pb != opt->pb) {
+ // Validate the options.
+ if (opt->lc > LZMA_LCLP_MAX || opt->lp > LZMA_LCLP_MAX
+ || opt->lc + opt->lp > LZMA_LCLP_MAX
+ || opt->pb > LZMA_PB_MAX)
+ return LZMA_OPTIONS_ERROR;
+
+ // The new options will be used when the encoder starts
+ // a new LZMA2 chunk.
+ coder->opt_cur.lc = opt->lc;
+ coder->opt_cur.lp = opt->lp;
+ coder->opt_cur.pb = opt->pb;
+ coder->need_properties = true;
+ coder->need_state_reset = true;
+ }
+
+ return LZMA_OK;
+}
+
+
+static lzma_ret
+lzma2_encoder_init(lzma_lz_encoder *lz, lzma_allocator *allocator,
+ const void *options, lzma_lz_options *lz_options)
+{
+ if (options == NULL)
+ return LZMA_PROG_ERROR;
+
+ if (lz->coder == NULL) {
+ lz->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (lz->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ lz->code = &lzma2_encode;
+ lz->end = &lzma2_encoder_end;
+ lz->options_update = &lzma2_encoder_options_update;
+
+ lz->coder->lzma = NULL;
+ }
+
+ lz->coder->opt_cur = *(const lzma_options_lzma *)(options);
+
+ lz->coder->sequence = SEQ_INIT;
+ lz->coder->need_properties = true;
+ lz->coder->need_state_reset = false;
+ lz->coder->need_dictionary_reset
+ = lz->coder->opt_cur.preset_dict == NULL
+ || lz->coder->opt_cur.preset_dict_size == 0;
+
+ // Initialize LZMA encoder
+ return_if_error(lzma_lzma_encoder_create(&lz->coder->lzma, allocator,
+ &lz->coder->opt_cur, lz_options));
+
+ // Make sure that we will always have enough history available in
+ // case we need to use uncompressed chunks. They are used when the
+ // compressed size of a chunk is not smaller than the uncompressed
+ // size, so we need to have at least LZMA2_COMPRESSED_MAX bytes
+ // history available.
+ if (lz_options->before_size + lz_options->dict_size < LZMA2_CHUNK_MAX)
+ lz_options->before_size
+ = LZMA2_CHUNK_MAX - lz_options->dict_size;
+
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_lzma2_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ return lzma_lz_encoder_init(
+ next, allocator, filters, &lzma2_encoder_init);
+}
+
+
+extern uint64_t
+lzma_lzma2_encoder_memusage(const void *options)
+{
+ const uint64_t lzma_mem = lzma_lzma_encoder_memusage(options);
+ if (lzma_mem == UINT64_MAX)
+ return UINT64_MAX;
+
+ return sizeof(lzma_coder) + lzma_mem;
+}
+
+
+extern lzma_ret
+lzma_lzma2_props_encode(const void *options, uint8_t *out)
+{
+ const lzma_options_lzma *const opt = options;
+ uint32_t d = my_max(opt->dict_size, LZMA_DICT_SIZE_MIN);
+
+ // Round up to to the next 2^n - 1 or 2^n + 2^(n - 1) - 1 depending
+ // on which one is the next:
+ --d;
+ d |= d >> 2;
+ d |= d >> 3;
+ d |= d >> 4;
+ d |= d >> 8;
+ d |= d >> 16;
+
+ // Get the highest two bits using the proper encoding:
+ if (d == UINT32_MAX)
+ out[0] = 40;
+ else
+ out[0] = get_pos_slot(d + 1) - 24;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/lzma/lzma2_encoder.h b/src/liblzma/lzma/lzma2_encoder.h
new file mode 100644
index 0000000..ca19ef4
--- /dev/null
+++ b/src/liblzma/lzma/lzma2_encoder.h
@@ -0,0 +1,41 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma2_encoder.h
+/// \brief LZMA2 encoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZMA2_ENCODER_H
+#define LZMA_LZMA2_ENCODER_H
+
+#include "common.h"
+
+
+/// Maximum number of bytes of actual data per chunk (no headers)
+#define LZMA2_CHUNK_MAX (UINT32_C(1) << 16)
+
+/// Maximum uncompressed size of LZMA chunk (no headers)
+#define LZMA2_UNCOMPRESSED_MAX (UINT32_C(1) << 21)
+
+/// Maximum size of LZMA2 headers
+#define LZMA2_HEADER_MAX 6
+
+/// Size of a header for uncompressed chunk
+#define LZMA2_HEADER_UNCOMPRESSED 3
+
+
+extern lzma_ret lzma_lzma2_encoder_init(
+ lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters);
+
+extern uint64_t lzma_lzma2_encoder_memusage(const void *options);
+
+extern lzma_ret lzma_lzma2_props_encode(const void *options, uint8_t *out);
+
+#endif
diff --git a/src/liblzma/lzma/lzma_common.h b/src/liblzma/lzma/lzma_common.h
new file mode 100644
index 0000000..e31e285
--- /dev/null
+++ b/src/liblzma/lzma/lzma_common.h
@@ -0,0 +1,223 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_common.h
+/// \brief Private definitions common to LZMA encoder and decoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZMA_COMMON_H
+#define LZMA_LZMA_COMMON_H
+
+#include "common.h"
+#include "range_common.h"
+
+
+///////////////////
+// Miscellaneous //
+///////////////////
+
+/// Maximum number of position states. A position state is the lowest pos bits
+/// number of bits of the current uncompressed offset. In some places there
+/// are different sets of probabilities for different pos states.
+#define POS_STATES_MAX (1 << LZMA_PB_MAX)
+
+
+/// Validates lc, lp, and pb.
+static inline bool
+is_lclppb_valid(const lzma_options_lzma *options)
+{
+ return options->lc <= LZMA_LCLP_MAX && options->lp <= LZMA_LCLP_MAX
+ && options->lc + options->lp <= LZMA_LCLP_MAX
+ && options->pb <= LZMA_PB_MAX;
+}
+
+
+///////////
+// State //
+///////////
+
+/// This enum is used to track which events have occurred most recently and
+/// in which order. This information is used to predict the next event.
+///
+/// Events:
+/// - Literal: One 8-bit byte
+/// - Match: Repeat a chunk of data at some distance
+/// - Long repeat: Multi-byte match at a recently seen distance
+/// - Short repeat: One-byte repeat at a recently seen distance
+///
+/// The event names are in from STATE_oldest_older_previous. REP means
+/// either short or long repeated match, and NONLIT means any non-literal.
+typedef enum {
+ STATE_LIT_LIT,
+ STATE_MATCH_LIT_LIT,
+ STATE_REP_LIT_LIT,
+ STATE_SHORTREP_LIT_LIT,
+ STATE_MATCH_LIT,
+ STATE_REP_LIT,
+ STATE_SHORTREP_LIT,
+ STATE_LIT_MATCH,
+ STATE_LIT_LONGREP,
+ STATE_LIT_SHORTREP,
+ STATE_NONLIT_MATCH,
+ STATE_NONLIT_REP,
+} lzma_lzma_state;
+
+
+/// Total number of states
+#define STATES 12
+
+/// The lowest 7 states indicate that the previous state was a literal.
+#define LIT_STATES 7
+
+
+/// Indicate that the latest state was a literal.
+#define update_literal(state) \
+ state = ((state) <= STATE_SHORTREP_LIT_LIT \
+ ? STATE_LIT_LIT \
+ : ((state) <= STATE_LIT_SHORTREP \
+ ? (state) - 3 \
+ : (state) - 6))
+
+/// Indicate that the latest state was a match.
+#define update_match(state) \
+ state = ((state) < LIT_STATES ? STATE_LIT_MATCH : STATE_NONLIT_MATCH)
+
+/// Indicate that the latest state was a long repeated match.
+#define update_long_rep(state) \
+ state = ((state) < LIT_STATES ? STATE_LIT_LONGREP : STATE_NONLIT_REP)
+
+/// Indicate that the latest state was a short match.
+#define update_short_rep(state) \
+ state = ((state) < LIT_STATES ? STATE_LIT_SHORTREP : STATE_NONLIT_REP)
+
+/// Test if the previous state was a literal.
+#define is_literal_state(state) \
+ ((state) < LIT_STATES)
+
+
+/////////////
+// Literal //
+/////////////
+
+/// Each literal coder is divided in three sections:
+/// - 0x001-0x0FF: Without match byte
+/// - 0x101-0x1FF: With match byte; match bit is 0
+/// - 0x201-0x2FF: With match byte; match bit is 1
+///
+/// Match byte is used when the previous LZMA symbol was something else than
+/// a literal (that is, it was some kind of match).
+#define LITERAL_CODER_SIZE 0x300
+
+/// Maximum number of literal coders
+#define LITERAL_CODERS_MAX (1 << LZMA_LCLP_MAX)
+
+/// Locate the literal coder for the next literal byte. The choice depends on
+/// - the lowest literal_pos_bits bits of the position of the current
+/// byte; and
+/// - the highest literal_context_bits bits of the previous byte.
+#define literal_subcoder(probs, lc, lp_mask, pos, prev_byte) \
+ ((probs)[(((pos) & lp_mask) << lc) + ((prev_byte) >> (8 - lc))])
+
+
+static inline void
+literal_init(probability (*probs)[LITERAL_CODER_SIZE],
+ uint32_t lc, uint32_t lp)
+{
+ assert(lc + lp <= LZMA_LCLP_MAX);
+
+ const uint32_t coders = 1U << (lc + lp);
+
+ for (uint32_t i = 0; i < coders; ++i)
+ for (uint32_t j = 0; j < LITERAL_CODER_SIZE; ++j)
+ bit_reset(probs[i][j]);
+
+ return;
+}
+
+
+//////////////////
+// Match length //
+//////////////////
+
+// Minimum length of a match is two bytes.
+#define MATCH_LEN_MIN 2
+
+// Match length is encoded with 4, 5, or 10 bits.
+//
+// Length Bits
+// 2-9 4 = Choice=0 + 3 bits
+// 10-17 5 = Choice=1 + Choice2=0 + 3 bits
+// 18-273 10 = Choice=1 + Choice2=1 + 8 bits
+#define LEN_LOW_BITS 3
+#define LEN_LOW_SYMBOLS (1 << LEN_LOW_BITS)
+#define LEN_MID_BITS 3
+#define LEN_MID_SYMBOLS (1 << LEN_MID_BITS)
+#define LEN_HIGH_BITS 8
+#define LEN_HIGH_SYMBOLS (1 << LEN_HIGH_BITS)
+#define LEN_SYMBOLS (LEN_LOW_SYMBOLS + LEN_MID_SYMBOLS + LEN_HIGH_SYMBOLS)
+
+// Maximum length of a match is 273 which is a result of the encoding
+// described above.
+#define MATCH_LEN_MAX (MATCH_LEN_MIN + LEN_SYMBOLS - 1)
+
+
+////////////////////
+// Match distance //
+////////////////////
+
+// Different set of probabilities is used for match distances that have very
+// short match length: Lengths of 2, 3, and 4 bytes have a separate set of
+// probabilities for each length. The matches with longer length use a shared
+// set of probabilities.
+#define LEN_TO_POS_STATES 4
+
+// Macro to get the index of the appropriate probability array.
+#define get_len_to_pos_state(len) \
+ ((len) < LEN_TO_POS_STATES + MATCH_LEN_MIN \
+ ? (len) - MATCH_LEN_MIN \
+ : LEN_TO_POS_STATES - 1)
+
+// The highest two bits of a match distance (pos slot) are encoded using six
+// bits. See fastpos.h for more explanation.
+#define POS_SLOT_BITS 6
+#define POS_SLOTS (1 << POS_SLOT_BITS)
+
+// Match distances up to 127 are fully encoded using probabilities. Since
+// the highest two bits (pos slot) are always encoded using six bits, the
+// distances 0-3 don't need any additional bits to encode, since the pos
+// slot itself is the same as the actual distance. START_POS_MODEL_INDEX
+// indicates the first pos slot where at least one additional bit is needed.
+#define START_POS_MODEL_INDEX 4
+
+// Match distances greater than 127 are encoded in three pieces:
+// - pos slot: the highest two bits
+// - direct bits: 2-26 bits below the highest two bits
+// - alignment bits: four lowest bits
+//
+// Direct bits don't use any probabilities.
+//
+// The pos slot value of 14 is for distances 128-191 (see the table in
+// fastpos.h to understand why).
+#define END_POS_MODEL_INDEX 14
+
+// Pos slots that indicate a distance <= 127.
+#define FULL_DISTANCES_BITS (END_POS_MODEL_INDEX / 2)
+#define FULL_DISTANCES (1 << FULL_DISTANCES_BITS)
+
+// For match distances greater than 127, only the highest two bits and the
+// lowest four bits (alignment) is encoded using probabilities.
+#define ALIGN_BITS 4
+#define ALIGN_TABLE_SIZE (1 << ALIGN_BITS)
+#define ALIGN_MASK (ALIGN_TABLE_SIZE - 1)
+
+// LZMA remembers the four most recent match distances. Reusing these distances
+// tends to take less space than re-encoding the actual distance value.
+#define REP_DISTANCES 4
+
+#endif
diff --git a/src/liblzma/lzma/lzma_decoder.c b/src/liblzma/lzma/lzma_decoder.c
new file mode 100644
index 0000000..4329e01
--- /dev/null
+++ b/src/liblzma/lzma/lzma_decoder.c
@@ -0,0 +1,1057 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_decoder.c
+/// \brief LZMA decoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lz_decoder.h"
+#include "lzma_common.h"
+#include "lzma_decoder.h"
+#include "range_decoder.h"
+
+
+#ifdef HAVE_SMALL
+
+// Macros for (somewhat) size-optimized code.
+#define seq_4(seq) seq
+
+#define seq_6(seq) seq
+
+#define seq_8(seq) seq
+
+#define seq_len(seq) \
+ seq ## _CHOICE, \
+ seq ## _CHOICE2, \
+ seq ## _BITTREE
+
+#define len_decode(target, ld, pos_state, seq) \
+do { \
+case seq ## _CHOICE: \
+ rc_if_0(ld.choice, seq ## _CHOICE) { \
+ rc_update_0(ld.choice); \
+ probs = ld.low[pos_state];\
+ limit = LEN_LOW_SYMBOLS; \
+ target = MATCH_LEN_MIN; \
+ } else { \
+ rc_update_1(ld.choice); \
+case seq ## _CHOICE2: \
+ rc_if_0(ld.choice2, seq ## _CHOICE2) { \
+ rc_update_0(ld.choice2); \
+ probs = ld.mid[pos_state]; \
+ limit = LEN_MID_SYMBOLS; \
+ target = MATCH_LEN_MIN + LEN_LOW_SYMBOLS; \
+ } else { \
+ rc_update_1(ld.choice2); \
+ probs = ld.high; \
+ limit = LEN_HIGH_SYMBOLS; \
+ target = MATCH_LEN_MIN + LEN_LOW_SYMBOLS \
+ + LEN_MID_SYMBOLS; \
+ } \
+ } \
+ symbol = 1; \
+case seq ## _BITTREE: \
+ do { \
+ rc_bit(probs[symbol], , , seq ## _BITTREE); \
+ } while (symbol < limit); \
+ target += symbol - limit; \
+} while (0)
+
+#else // HAVE_SMALL
+
+// Unrolled versions
+#define seq_4(seq) \
+ seq ## 0, \
+ seq ## 1, \
+ seq ## 2, \
+ seq ## 3
+
+#define seq_6(seq) \
+ seq ## 0, \
+ seq ## 1, \
+ seq ## 2, \
+ seq ## 3, \
+ seq ## 4, \
+ seq ## 5
+
+#define seq_8(seq) \
+ seq ## 0, \
+ seq ## 1, \
+ seq ## 2, \
+ seq ## 3, \
+ seq ## 4, \
+ seq ## 5, \
+ seq ## 6, \
+ seq ## 7
+
+#define seq_len(seq) \
+ seq ## _CHOICE, \
+ seq ## _LOW0, \
+ seq ## _LOW1, \
+ seq ## _LOW2, \
+ seq ## _CHOICE2, \
+ seq ## _MID0, \
+ seq ## _MID1, \
+ seq ## _MID2, \
+ seq ## _HIGH0, \
+ seq ## _HIGH1, \
+ seq ## _HIGH2, \
+ seq ## _HIGH3, \
+ seq ## _HIGH4, \
+ seq ## _HIGH5, \
+ seq ## _HIGH6, \
+ seq ## _HIGH7
+
+#define len_decode(target, ld, pos_state, seq) \
+do { \
+ symbol = 1; \
+case seq ## _CHOICE: \
+ rc_if_0(ld.choice, seq ## _CHOICE) { \
+ rc_update_0(ld.choice); \
+ rc_bit_case(ld.low[pos_state][symbol], , , seq ## _LOW0); \
+ rc_bit_case(ld.low[pos_state][symbol], , , seq ## _LOW1); \
+ rc_bit_case(ld.low[pos_state][symbol], , , seq ## _LOW2); \
+ target = symbol - LEN_LOW_SYMBOLS + MATCH_LEN_MIN; \
+ } else { \
+ rc_update_1(ld.choice); \
+case seq ## _CHOICE2: \
+ rc_if_0(ld.choice2, seq ## _CHOICE2) { \
+ rc_update_0(ld.choice2); \
+ rc_bit_case(ld.mid[pos_state][symbol], , , \
+ seq ## _MID0); \
+ rc_bit_case(ld.mid[pos_state][symbol], , , \
+ seq ## _MID1); \
+ rc_bit_case(ld.mid[pos_state][symbol], , , \
+ seq ## _MID2); \
+ target = symbol - LEN_MID_SYMBOLS \
+ + MATCH_LEN_MIN + LEN_LOW_SYMBOLS; \
+ } else { \
+ rc_update_1(ld.choice2); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH0); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH1); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH2); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH3); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH4); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH5); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH6); \
+ rc_bit_case(ld.high[symbol], , , seq ## _HIGH7); \
+ target = symbol - LEN_HIGH_SYMBOLS \
+ + MATCH_LEN_MIN \
+ + LEN_LOW_SYMBOLS + LEN_MID_SYMBOLS; \
+ } \
+ } \
+} while (0)
+
+#endif // HAVE_SMALL
+
+
+/// Length decoder probabilities; see comments in lzma_common.h.
+typedef struct {
+ probability choice;
+ probability choice2;
+ probability low[POS_STATES_MAX][LEN_LOW_SYMBOLS];
+ probability mid[POS_STATES_MAX][LEN_MID_SYMBOLS];
+ probability high[LEN_HIGH_SYMBOLS];
+} lzma_length_decoder;
+
+
+struct lzma_coder_s {
+ ///////////////////
+ // Probabilities //
+ ///////////////////
+
+ /// Literals; see comments in lzma_common.h.
+ probability literal[LITERAL_CODERS_MAX][LITERAL_CODER_SIZE];
+
+ /// If 1, it's a match. Otherwise it's a single 8-bit literal.
+ probability is_match[STATES][POS_STATES_MAX];
+
+ /// If 1, it's a repeated match. The distance is one of rep0 .. rep3.
+ probability is_rep[STATES];
+
+ /// If 0, distance of a repeated match is rep0.
+ /// Otherwise check is_rep1.
+ probability is_rep0[STATES];
+
+ /// If 0, distance of a repeated match is rep1.
+ /// Otherwise check is_rep2.
+ probability is_rep1[STATES];
+
+ /// If 0, distance of a repeated match is rep2. Otherwise it is rep3.
+ probability is_rep2[STATES];
+
+ /// If 1, the repeated match has length of one byte. Otherwise
+ /// the length is decoded from rep_len_decoder.
+ probability is_rep0_long[STATES][POS_STATES_MAX];
+
+ /// Probability tree for the highest two bits of the match distance.
+ /// There is a separate probability tree for match lengths of
+ /// 2 (i.e. MATCH_LEN_MIN), 3, 4, and [5, 273].
+ probability pos_slot[LEN_TO_POS_STATES][POS_SLOTS];
+
+ /// Probability trees for additional bits for match distance when the
+ /// distance is in the range [4, 127].
+ probability pos_special[FULL_DISTANCES - END_POS_MODEL_INDEX];
+
+ /// Probability tree for the lowest four bits of a match distance
+ /// that is equal to or greater than 128.
+ probability pos_align[ALIGN_TABLE_SIZE];
+
+ /// Length of a normal match
+ lzma_length_decoder match_len_decoder;
+
+ /// Length of a repeated match
+ lzma_length_decoder rep_len_decoder;
+
+ ///////////////////
+ // Decoder state //
+ ///////////////////
+
+ // Range coder
+ lzma_range_decoder rc;
+
+ // Types of the most recently seen LZMA symbols
+ lzma_lzma_state state;
+
+ uint32_t rep0; ///< Distance of the latest match
+ uint32_t rep1; ///< Distance of second latest match
+ uint32_t rep2; ///< Distance of third latest match
+ uint32_t rep3; ///< Distance of fourth latest match
+
+ uint32_t pos_mask; // (1U << pb) - 1
+ uint32_t literal_context_bits;
+ uint32_t literal_pos_mask;
+
+ /// Uncompressed size as bytes, or LZMA_VLI_UNKNOWN if end of
+ /// payload marker is expected.
+ lzma_vli uncompressed_size;
+
+ ////////////////////////////////
+ // State of incomplete symbol //
+ ////////////////////////////////
+
+ /// Position where to continue the decoder loop
+ enum {
+ SEQ_NORMALIZE,
+ SEQ_IS_MATCH,
+ seq_8(SEQ_LITERAL),
+ seq_8(SEQ_LITERAL_MATCHED),
+ SEQ_LITERAL_WRITE,
+ SEQ_IS_REP,
+ seq_len(SEQ_MATCH_LEN),
+ seq_6(SEQ_POS_SLOT),
+ SEQ_POS_MODEL,
+ SEQ_DIRECT,
+ seq_4(SEQ_ALIGN),
+ SEQ_EOPM,
+ SEQ_IS_REP0,
+ SEQ_SHORTREP,
+ SEQ_IS_REP0_LONG,
+ SEQ_IS_REP1,
+ SEQ_IS_REP2,
+ seq_len(SEQ_REP_LEN),
+ SEQ_COPY,
+ } sequence;
+
+ /// Base of the current probability tree
+ probability *probs;
+
+ /// Symbol being decoded. This is also used as an index variable in
+ /// bittree decoders: probs[symbol]
+ uint32_t symbol;
+
+ /// Used as a loop termination condition on bittree decoders and
+ /// direct bits decoder.
+ uint32_t limit;
+
+ /// Matched literal decoder: 0x100 or 0 to help avoiding branches.
+ /// Bittree reverse decoders: Offset of the next bit: 1 << offset
+ uint32_t offset;
+
+ /// If decoding a literal: match byte.
+ /// If decoding a match: length of the match.
+ uint32_t len;
+};
+
+
+static lzma_ret
+lzma_decode(lzma_coder *restrict coder, lzma_dict *restrict dictptr,
+ const uint8_t *restrict in,
+ size_t *restrict in_pos, size_t in_size)
+{
+ ////////////////////
+ // Initialization //
+ ////////////////////
+
+ if (!rc_read_init(&coder->rc, in, in_pos, in_size))
+ return LZMA_OK;
+
+ ///////////////
+ // Variables //
+ ///////////////
+
+ // Making local copies of often-used variables improves both
+ // speed and readability.
+
+ lzma_dict dict = *dictptr;
+
+ const size_t dict_start = dict.pos;
+
+ // Range decoder
+ rc_to_local(coder->rc, *in_pos);
+
+ // State
+ uint32_t state = coder->state;
+ uint32_t rep0 = coder->rep0;
+ uint32_t rep1 = coder->rep1;
+ uint32_t rep2 = coder->rep2;
+ uint32_t rep3 = coder->rep3;
+
+ const uint32_t pos_mask = coder->pos_mask;
+
+ // These variables are actually needed only if we last time ran
+ // out of input in the middle of the decoder loop.
+ probability *probs = coder->probs;
+ uint32_t symbol = coder->symbol;
+ uint32_t limit = coder->limit;
+ uint32_t offset = coder->offset;
+ uint32_t len = coder->len;
+
+ const uint32_t literal_pos_mask = coder->literal_pos_mask;
+ const uint32_t literal_context_bits = coder->literal_context_bits;
+
+ // Temporary variables
+ uint32_t pos_state = dict.pos & pos_mask;
+
+ lzma_ret ret = LZMA_OK;
+
+ // If uncompressed size is known, there must be no end of payload
+ // marker.
+ const bool no_eopm = coder->uncompressed_size
+ != LZMA_VLI_UNKNOWN;
+ if (no_eopm && coder->uncompressed_size < dict.limit - dict.pos)
+ dict.limit = dict.pos + (size_t)(coder->uncompressed_size);
+
+ // The main decoder loop. The "switch" is used to restart the decoder at
+ // correct location. Once restarted, the "switch" is no longer used.
+ switch (coder->sequence)
+ while (true) {
+ // Calculate new pos_state. This is skipped on the first loop
+ // since we already calculated it when setting up the local
+ // variables.
+ pos_state = dict.pos & pos_mask;
+
+ case SEQ_NORMALIZE:
+ case SEQ_IS_MATCH:
+ if (unlikely(no_eopm && dict.pos == dict.limit))
+ break;
+
+ rc_if_0(coder->is_match[state][pos_state], SEQ_IS_MATCH) {
+ rc_update_0(coder->is_match[state][pos_state]);
+
+ // It's a literal i.e. a single 8-bit byte.
+
+ probs = literal_subcoder(coder->literal,
+ literal_context_bits, literal_pos_mask,
+ dict.pos, dict_get(&dict, 0));
+ symbol = 1;
+
+ if (is_literal_state(state)) {
+ // Decode literal without match byte.
+#ifdef HAVE_SMALL
+ case SEQ_LITERAL:
+ do {
+ rc_bit(probs[symbol], , , SEQ_LITERAL);
+ } while (symbol < (1 << 8));
+#else
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL0);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL1);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL2);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL3);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL4);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL5);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL6);
+ rc_bit_case(probs[symbol], , , SEQ_LITERAL7);
+#endif
+ } else {
+ // Decode literal with match byte.
+ //
+ // We store the byte we compare against
+ // ("match byte") to "len" to minimize the
+ // number of variables we need to store
+ // between decoder calls.
+ len = dict_get(&dict, rep0) << 1;
+
+ // The usage of "offset" allows omitting some
+ // branches, which should give tiny speed
+ // improvement on some CPUs. "offset" gets
+ // set to zero if match_bit didn't match.
+ offset = 0x100;
+
+#ifdef HAVE_SMALL
+ case SEQ_LITERAL_MATCHED:
+ do {
+ const uint32_t match_bit
+ = len & offset;
+ const uint32_t subcoder_index
+ = offset + match_bit
+ + symbol;
+
+ rc_bit(probs[subcoder_index],
+ offset &= ~match_bit,
+ offset &= match_bit,
+ SEQ_LITERAL_MATCHED);
+
+ // It seems to be faster to do this
+ // here instead of putting it to the
+ // beginning of the loop and then
+ // putting the "case" in the middle
+ // of the loop.
+ len <<= 1;
+
+ } while (symbol < (1 << 8));
+#else
+ // Unroll the loop.
+ uint32_t match_bit;
+ uint32_t subcoder_index;
+
+# define d(seq) \
+ case seq: \
+ match_bit = len & offset; \
+ subcoder_index = offset + match_bit + symbol; \
+ rc_bit(probs[subcoder_index], \
+ offset &= ~match_bit, \
+ offset &= match_bit, \
+ seq)
+
+ d(SEQ_LITERAL_MATCHED0);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED1);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED2);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED3);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED4);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED5);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED6);
+ len <<= 1;
+ d(SEQ_LITERAL_MATCHED7);
+# undef d
+#endif
+ }
+
+ //update_literal(state);
+ // Use a lookup table to update to literal state,
+ // since compared to other state updates, this would
+ // need two branches.
+ static const lzma_lzma_state next_state[] = {
+ STATE_LIT_LIT,
+ STATE_LIT_LIT,
+ STATE_LIT_LIT,
+ STATE_LIT_LIT,
+ STATE_MATCH_LIT_LIT,
+ STATE_REP_LIT_LIT,
+ STATE_SHORTREP_LIT_LIT,
+ STATE_MATCH_LIT,
+ STATE_REP_LIT,
+ STATE_SHORTREP_LIT,
+ STATE_MATCH_LIT,
+ STATE_REP_LIT
+ };
+ state = next_state[state];
+
+ case SEQ_LITERAL_WRITE:
+ if (unlikely(dict_put(&dict, symbol))) {
+ coder->sequence = SEQ_LITERAL_WRITE;
+ goto out;
+ }
+
+ continue;
+ }
+
+ // Instead of a new byte we are going to get a byte range
+ // (distance and length) which will be repeated from our
+ // output history.
+
+ rc_update_1(coder->is_match[state][pos_state]);
+
+ case SEQ_IS_REP:
+ rc_if_0(coder->is_rep[state], SEQ_IS_REP) {
+ // Not a repeated match
+ rc_update_0(coder->is_rep[state]);
+ update_match(state);
+
+ // The latest three match distances are kept in
+ // memory in case there are repeated matches.
+ rep3 = rep2;
+ rep2 = rep1;
+ rep1 = rep0;
+
+ // Decode the length of the match.
+ len_decode(len, coder->match_len_decoder,
+ pos_state, SEQ_MATCH_LEN);
+
+ // Prepare to decode the highest two bits of the
+ // match distance.
+ probs = coder->pos_slot[get_len_to_pos_state(len)];
+ symbol = 1;
+
+#ifdef HAVE_SMALL
+ case SEQ_POS_SLOT:
+ do {
+ rc_bit(probs[symbol], , , SEQ_POS_SLOT);
+ } while (symbol < POS_SLOTS);
+#else
+ rc_bit_case(probs[symbol], , , SEQ_POS_SLOT0);
+ rc_bit_case(probs[symbol], , , SEQ_POS_SLOT1);
+ rc_bit_case(probs[symbol], , , SEQ_POS_SLOT2);
+ rc_bit_case(probs[symbol], , , SEQ_POS_SLOT3);
+ rc_bit_case(probs[symbol], , , SEQ_POS_SLOT4);
+ rc_bit_case(probs[symbol], , , SEQ_POS_SLOT5);
+#endif
+ // Get rid of the highest bit that was needed for
+ // indexing of the probability array.
+ symbol -= POS_SLOTS;
+ assert(symbol <= 63);
+
+ if (symbol < START_POS_MODEL_INDEX) {
+ // Match distances [0, 3] have only two bits.
+ rep0 = symbol;
+ } else {
+ // Decode the lowest [1, 29] bits of
+ // the match distance.
+ limit = (symbol >> 1) - 1;
+ assert(limit >= 1 && limit <= 30);
+ rep0 = 2 + (symbol & 1);
+
+ if (symbol < END_POS_MODEL_INDEX) {
+ // Prepare to decode the low bits for
+ // a distance of [4, 127].
+ assert(limit <= 5);
+ rep0 <<= limit;
+ assert(rep0 <= 96);
+ // -1 is fine, because we start
+ // decoding at probs[1], not probs[0].
+ // NOTE: This violates the C standard,
+ // since we are doing pointer
+ // arithmetic past the beginning of
+ // the array.
+ assert((int32_t)(rep0 - symbol - 1)
+ >= -1);
+ assert((int32_t)(rep0 - symbol - 1)
+ <= 82);
+ probs = coder->pos_special + rep0
+ - symbol - 1;
+ symbol = 1;
+ offset = 0;
+ case SEQ_POS_MODEL:
+#ifdef HAVE_SMALL
+ do {
+ rc_bit(probs[symbol], ,
+ rep0 += 1 << offset,
+ SEQ_POS_MODEL);
+ } while (++offset < limit);
+#else
+ switch (limit) {
+ case 5:
+ assert(offset == 0);
+ rc_bit(probs[symbol], ,
+ rep0 += 1,
+ SEQ_POS_MODEL);
+ ++offset;
+ --limit;
+ case 4:
+ rc_bit(probs[symbol], ,
+ rep0 += 1 << offset,
+ SEQ_POS_MODEL);
+ ++offset;
+ --limit;
+ case 3:
+ rc_bit(probs[symbol], ,
+ rep0 += 1 << offset,
+ SEQ_POS_MODEL);
+ ++offset;
+ --limit;
+ case 2:
+ rc_bit(probs[symbol], ,
+ rep0 += 1 << offset,
+ SEQ_POS_MODEL);
+ ++offset;
+ --limit;
+ case 1:
+ // We need "symbol" only for
+ // indexing the probability
+ // array, thus we can use
+ // rc_bit_last() here to omit
+ // the unneeded updating of
+ // "symbol".
+ rc_bit_last(probs[symbol], ,
+ rep0 += 1 << offset,
+ SEQ_POS_MODEL);
+ }
+#endif
+ } else {
+ // The distance is >= 128. Decode the
+ // lower bits without probabilities
+ // except the lowest four bits.
+ assert(symbol >= 14);
+ assert(limit >= 6);
+ limit -= ALIGN_BITS;
+ assert(limit >= 2);
+ case SEQ_DIRECT:
+ // Not worth manual unrolling
+ do {
+ rc_direct(rep0, SEQ_DIRECT);
+ } while (--limit > 0);
+
+ // Decode the lowest four bits using
+ // probabilities.
+ rep0 <<= ALIGN_BITS;
+ symbol = 1;
+#ifdef HAVE_SMALL
+ offset = 0;
+ case SEQ_ALIGN:
+ do {
+ rc_bit(coder->pos_align[
+ symbol], ,
+ rep0 += 1 << offset,
+ SEQ_ALIGN);
+ } while (++offset < ALIGN_BITS);
+#else
+ case SEQ_ALIGN0:
+ rc_bit(coder->pos_align[symbol], ,
+ rep0 += 1, SEQ_ALIGN0);
+ case SEQ_ALIGN1:
+ rc_bit(coder->pos_align[symbol], ,
+ rep0 += 2, SEQ_ALIGN1);
+ case SEQ_ALIGN2:
+ rc_bit(coder->pos_align[symbol], ,
+ rep0 += 4, SEQ_ALIGN2);
+ case SEQ_ALIGN3:
+ // Like in SEQ_POS_MODEL, we don't
+ // need "symbol" for anything else
+ // than indexing the probability array.
+ rc_bit_last(coder->pos_align[symbol], ,
+ rep0 += 8, SEQ_ALIGN3);
+#endif
+
+ if (rep0 == UINT32_MAX) {
+ // End of payload marker was
+ // found. It must not be
+ // present if uncompressed
+ // size is known.
+ if (coder->uncompressed_size
+ != LZMA_VLI_UNKNOWN) {
+ ret = LZMA_DATA_ERROR;
+ goto out;
+ }
+
+ case SEQ_EOPM:
+ // TODO Comment
+ rc_normalize(SEQ_EOPM);
+ ret = LZMA_STREAM_END;
+ goto out;
+ }
+ }
+ }
+
+ // Validate the distance we just decoded.
+ if (unlikely(!dict_is_distance_valid(&dict, rep0))) {
+ ret = LZMA_DATA_ERROR;
+ goto out;
+ }
+
+ } else {
+ rc_update_1(coder->is_rep[state]);
+
+ // Repeated match
+ //
+ // The match distance is a value that we have had
+ // earlier. The latest four match distances are
+ // available as rep0, rep1, rep2 and rep3. We will
+ // now decode which of them is the new distance.
+ //
+ // There cannot be a match if we haven't produced
+ // any output, so check that first.
+ if (unlikely(!dict_is_distance_valid(&dict, 0))) {
+ ret = LZMA_DATA_ERROR;
+ goto out;
+ }
+
+ case SEQ_IS_REP0:
+ rc_if_0(coder->is_rep0[state], SEQ_IS_REP0) {
+ rc_update_0(coder->is_rep0[state]);
+ // The distance is rep0.
+
+ case SEQ_IS_REP0_LONG:
+ rc_if_0(coder->is_rep0_long[state][pos_state],
+ SEQ_IS_REP0_LONG) {
+ rc_update_0(coder->is_rep0_long[
+ state][pos_state]);
+
+ update_short_rep(state);
+
+ case SEQ_SHORTREP:
+ if (unlikely(dict_put(&dict, dict_get(
+ &dict, rep0)))) {
+ coder->sequence = SEQ_SHORTREP;
+ goto out;
+ }
+
+ continue;
+ }
+
+ // Repeating more than one byte at
+ // distance of rep0.
+ rc_update_1(coder->is_rep0_long[
+ state][pos_state]);
+
+ } else {
+ rc_update_1(coder->is_rep0[state]);
+
+ case SEQ_IS_REP1:
+ // The distance is rep1, rep2 or rep3. Once
+ // we find out which one of these three, it
+ // is stored to rep0 and rep1, rep2 and rep3
+ // are updated accordingly.
+ rc_if_0(coder->is_rep1[state], SEQ_IS_REP1) {
+ rc_update_0(coder->is_rep1[state]);
+
+ const uint32_t distance = rep1;
+ rep1 = rep0;
+ rep0 = distance;
+
+ } else {
+ rc_update_1(coder->is_rep1[state]);
+ case SEQ_IS_REP2:
+ rc_if_0(coder->is_rep2[state],
+ SEQ_IS_REP2) {
+ rc_update_0(coder->is_rep2[
+ state]);
+
+ const uint32_t distance = rep2;
+ rep2 = rep1;
+ rep1 = rep0;
+ rep0 = distance;
+
+ } else {
+ rc_update_1(coder->is_rep2[
+ state]);
+
+ const uint32_t distance = rep3;
+ rep3 = rep2;
+ rep2 = rep1;
+ rep1 = rep0;
+ rep0 = distance;
+ }
+ }
+ }
+
+ update_long_rep(state);
+
+ // Decode the length of the repeated match.
+ len_decode(len, coder->rep_len_decoder,
+ pos_state, SEQ_REP_LEN);
+ }
+
+ /////////////////////////////////
+ // Repeat from history buffer. //
+ /////////////////////////////////
+
+ // The length is always between these limits. There is no way
+ // to trigger the algorithm to set len outside this range.
+ assert(len >= MATCH_LEN_MIN);
+ assert(len <= MATCH_LEN_MAX);
+
+ case SEQ_COPY:
+ // Repeat len bytes from distance of rep0.
+ if (unlikely(dict_repeat(&dict, rep0, &len))) {
+ coder->sequence = SEQ_COPY;
+ goto out;
+ }
+ }
+
+ rc_normalize(SEQ_NORMALIZE);
+ coder->sequence = SEQ_IS_MATCH;
+
+out:
+ // Save state
+
+ // NOTE: Must not copy dict.limit.
+ dictptr->pos = dict.pos;
+ dictptr->full = dict.full;
+
+ rc_from_local(coder->rc, *in_pos);
+
+ coder->state = state;
+ coder->rep0 = rep0;
+ coder->rep1 = rep1;
+ coder->rep2 = rep2;
+ coder->rep3 = rep3;
+
+ coder->probs = probs;
+ coder->symbol = symbol;
+ coder->limit = limit;
+ coder->offset = offset;
+ coder->len = len;
+
+ // Update the remaining amount of uncompressed data if uncompressed
+ // size was known.
+ if (coder->uncompressed_size != LZMA_VLI_UNKNOWN) {
+ coder->uncompressed_size -= dict.pos - dict_start;
+
+ // Since there cannot be end of payload marker if the
+ // uncompressed size was known, we check here if we
+ // finished decoding.
+ if (coder->uncompressed_size == 0 && ret == LZMA_OK
+ && coder->sequence != SEQ_NORMALIZE)
+ ret = coder->sequence == SEQ_IS_MATCH
+ ? LZMA_STREAM_END : LZMA_DATA_ERROR;
+ }
+
+ // We can do an additional check in the range decoder to catch some
+ // corrupted files.
+ if (ret == LZMA_STREAM_END) {
+ if (!rc_is_finished(coder->rc))
+ ret = LZMA_DATA_ERROR;
+
+ // Reset the range decoder so that it is ready to reinitialize
+ // for a new LZMA2 chunk.
+ rc_reset(coder->rc);
+ }
+
+ return ret;
+}
+
+
+
+static void
+lzma_decoder_uncompressed(lzma_coder *coder, lzma_vli uncompressed_size)
+{
+ coder->uncompressed_size = uncompressed_size;
+}
+
+/*
+extern void
+lzma_lzma_decoder_uncompressed(void *coder_ptr, lzma_vli uncompressed_size)
+{
+ // This is hack.
+ (*(lzma_coder **)(coder))->uncompressed_size = uncompressed_size;
+}
+*/
+
+static void
+lzma_decoder_reset(lzma_coder *coder, const void *opt)
+{
+ const lzma_options_lzma *options = opt;
+
+ // NOTE: We assume that lc/lp/pb are valid since they were
+ // successfully decoded with lzma_lzma_decode_properties().
+ // FIXME?
+
+ // Calculate pos_mask. We don't need pos_bits as is for anything.
+ coder->pos_mask = (1U << options->pb) - 1;
+
+ // Initialize the literal decoder.
+ literal_init(coder->literal, options->lc, options->lp);
+
+ coder->literal_context_bits = options->lc;
+ coder->literal_pos_mask = (1U << options->lp) - 1;
+
+ // State
+ coder->state = STATE_LIT_LIT;
+ coder->rep0 = 0;
+ coder->rep1 = 0;
+ coder->rep2 = 0;
+ coder->rep3 = 0;
+ coder->pos_mask = (1U << options->pb) - 1;
+
+ // Range decoder
+ rc_reset(coder->rc);
+
+ // Bit and bittree decoders
+ for (uint32_t i = 0; i < STATES; ++i) {
+ for (uint32_t j = 0; j <= coder->pos_mask; ++j) {
+ bit_reset(coder->is_match[i][j]);
+ bit_reset(coder->is_rep0_long[i][j]);
+ }
+
+ bit_reset(coder->is_rep[i]);
+ bit_reset(coder->is_rep0[i]);
+ bit_reset(coder->is_rep1[i]);
+ bit_reset(coder->is_rep2[i]);
+ }
+
+ for (uint32_t i = 0; i < LEN_TO_POS_STATES; ++i)
+ bittree_reset(coder->pos_slot[i], POS_SLOT_BITS);
+
+ for (uint32_t i = 0; i < FULL_DISTANCES - END_POS_MODEL_INDEX; ++i)
+ bit_reset(coder->pos_special[i]);
+
+ bittree_reset(coder->pos_align, ALIGN_BITS);
+
+ // Len decoders (also bit/bittree)
+ const uint32_t num_pos_states = 1U << options->pb;
+ bit_reset(coder->match_len_decoder.choice);
+ bit_reset(coder->match_len_decoder.choice2);
+ bit_reset(coder->rep_len_decoder.choice);
+ bit_reset(coder->rep_len_decoder.choice2);
+
+ for (uint32_t pos_state = 0; pos_state < num_pos_states; ++pos_state) {
+ bittree_reset(coder->match_len_decoder.low[pos_state],
+ LEN_LOW_BITS);
+ bittree_reset(coder->match_len_decoder.mid[pos_state],
+ LEN_MID_BITS);
+
+ bittree_reset(coder->rep_len_decoder.low[pos_state],
+ LEN_LOW_BITS);
+ bittree_reset(coder->rep_len_decoder.mid[pos_state],
+ LEN_MID_BITS);
+ }
+
+ bittree_reset(coder->match_len_decoder.high, LEN_HIGH_BITS);
+ bittree_reset(coder->rep_len_decoder.high, LEN_HIGH_BITS);
+
+ coder->sequence = SEQ_IS_MATCH;
+ coder->probs = NULL;
+ coder->symbol = 0;
+ coder->limit = 0;
+ coder->offset = 0;
+ coder->len = 0;
+
+ return;
+}
+
+
+extern lzma_ret
+lzma_lzma_decoder_create(lzma_lz_decoder *lz, lzma_allocator *allocator,
+ const void *opt, lzma_lz_options *lz_options)
+{
+ if (lz->coder == NULL) {
+ lz->coder = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (lz->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ lz->code = &lzma_decode;
+ lz->reset = &lzma_decoder_reset;
+ lz->set_uncompressed = &lzma_decoder_uncompressed;
+ }
+
+ // All dictionary sizes are OK here. LZ decoder will take care of
+ // the special cases.
+ const lzma_options_lzma *options = opt;
+ lz_options->dict_size = options->dict_size;
+ lz_options->preset_dict = options->preset_dict;
+ lz_options->preset_dict_size = options->preset_dict_size;
+
+ return LZMA_OK;
+}
+
+
+/// Allocate and initialize LZMA decoder. This is used only via LZ
+/// initialization (lzma_lzma_decoder_init() passes function pointer to
+/// the LZ initialization).
+static lzma_ret
+lzma_decoder_init(lzma_lz_decoder *lz, lzma_allocator *allocator,
+ const void *options, lzma_lz_options *lz_options)
+{
+ if (!is_lclppb_valid(options))
+ return LZMA_PROG_ERROR;
+
+ return_if_error(lzma_lzma_decoder_create(
+ lz, allocator, options, lz_options));
+
+ lzma_decoder_reset(lz->coder, options);
+ lzma_decoder_uncompressed(lz->coder, LZMA_VLI_UNKNOWN);
+
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_lzma_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ // LZMA can only be the last filter in the chain. This is enforced
+ // by the raw_decoder initialization.
+ assert(filters[1].init == NULL);
+
+ return lzma_lz_decoder_init(next, allocator, filters,
+ &lzma_decoder_init);
+}
+
+
+extern bool
+lzma_lzma_lclppb_decode(lzma_options_lzma *options, uint8_t byte)
+{
+ if (byte > (4 * 5 + 4) * 9 + 8)
+ return true;
+
+ // See the file format specification to understand this.
+ options->pb = byte / (9 * 5);
+ byte -= options->pb * 9 * 5;
+ options->lp = byte / 9;
+ options->lc = byte - options->lp * 9;
+
+ return options->lc + options->lp > LZMA_LCLP_MAX;
+}
+
+
+extern uint64_t
+lzma_lzma_decoder_memusage_nocheck(const void *options)
+{
+ const lzma_options_lzma *const opt = options;
+ return sizeof(lzma_coder) + lzma_lz_decoder_memusage(opt->dict_size);
+}
+
+
+extern uint64_t
+lzma_lzma_decoder_memusage(const void *options)
+{
+ if (!is_lclppb_valid(options))
+ return UINT64_MAX;
+
+ return lzma_lzma_decoder_memusage_nocheck(options);
+}
+
+
+extern lzma_ret
+lzma_lzma_props_decode(void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size)
+{
+ if (props_size != 5)
+ return LZMA_OPTIONS_ERROR;
+
+ lzma_options_lzma *opt
+ = lzma_alloc(sizeof(lzma_options_lzma), allocator);
+ if (opt == NULL)
+ return LZMA_MEM_ERROR;
+
+ if (lzma_lzma_lclppb_decode(opt, props[0]))
+ goto error;
+
+ // All dictionary sizes are accepted, including zero. LZ decoder
+ // will automatically use a dictionary at least a few KiB even if
+ // a smaller dictionary is requested.
+ opt->dict_size = unaligned_read32le(props + 1);
+
+ opt->preset_dict = NULL;
+ opt->preset_dict_size = 0;
+
+ *options = opt;
+
+ return LZMA_OK;
+
+error:
+ lzma_free(opt, allocator);
+ return LZMA_OPTIONS_ERROR;
+}
diff --git a/src/liblzma/lzma/lzma_decoder.h b/src/liblzma/lzma/lzma_decoder.h
new file mode 100644
index 0000000..a463a76
--- /dev/null
+++ b/src/liblzma/lzma/lzma_decoder.h
@@ -0,0 +1,52 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_decoder.h
+/// \brief LZMA decoder API
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZMA_DECODER_H
+#define LZMA_LZMA_DECODER_H
+
+#include "common.h"
+
+
+/// Allocates and initializes LZMA decoder
+extern lzma_ret lzma_lzma_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern uint64_t lzma_lzma_decoder_memusage(const void *options);
+
+extern lzma_ret lzma_lzma_props_decode(
+ void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size);
+
+
+/// \brief Decodes the LZMA Properties byte (lc/lp/pb)
+///
+/// \return true if error occurred, false on success
+///
+extern bool lzma_lzma_lclppb_decode(
+ lzma_options_lzma *options, uint8_t byte);
+
+
+#ifdef LZMA_LZ_DECODER_H
+/// Allocate and setup function pointers only. This is used by LZMA1 and
+/// LZMA2 decoders.
+extern lzma_ret lzma_lzma_decoder_create(
+ lzma_lz_decoder *lz, lzma_allocator *allocator,
+ const void *opt, lzma_lz_options *lz_options);
+
+/// Gets memory usage without validating lc/lp/pb. This is used by LZMA2
+/// decoder, because raw LZMA2 decoding doesn't need lc/lp/pb.
+extern uint64_t lzma_lzma_decoder_memusage_nocheck(const void *options);
+
+#endif
+
+#endif
diff --git a/src/liblzma/lzma/lzma_encoder.c b/src/liblzma/lzma/lzma_encoder.c
new file mode 100644
index 0000000..0fe992d
--- /dev/null
+++ b/src/liblzma/lzma/lzma_encoder.c
@@ -0,0 +1,675 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_encoder.c
+/// \brief LZMA encoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lzma2_encoder.h"
+#include "lzma_encoder_private.h"
+#include "fastpos.h"
+
+
+/////////////
+// Literal //
+/////////////
+
+static inline void
+literal_matched(lzma_range_encoder *rc, probability *subcoder,
+ uint32_t match_byte, uint32_t symbol)
+{
+ uint32_t offset = 0x100;
+ symbol += UINT32_C(1) << 8;
+
+ do {
+ match_byte <<= 1;
+ const uint32_t match_bit = match_byte & offset;
+ const uint32_t subcoder_index
+ = offset + match_bit + (symbol >> 8);
+ const uint32_t bit = (symbol >> 7) & 1;
+ rc_bit(rc, &subcoder[subcoder_index], bit);
+
+ symbol <<= 1;
+ offset &= ~(match_byte ^ symbol);
+
+ } while (symbol < (UINT32_C(1) << 16));
+}
+
+
+static inline void
+literal(lzma_coder *coder, lzma_mf *mf, uint32_t position)
+{
+ // Locate the literal byte to be encoded and the subcoder.
+ const uint8_t cur_byte = mf->buffer[
+ mf->read_pos - mf->read_ahead];
+ probability *subcoder = literal_subcoder(coder->literal,
+ coder->literal_context_bits, coder->literal_pos_mask,
+ position, mf->buffer[mf->read_pos - mf->read_ahead - 1]);
+
+ if (is_literal_state(coder->state)) {
+ // Previous LZMA-symbol was a literal. Encode a normal
+ // literal without a match byte.
+ rc_bittree(&coder->rc, subcoder, 8, cur_byte);
+ } else {
+ // Previous LZMA-symbol was a match. Use the last byte of
+ // the match as a "match byte". That is, compare the bits
+ // of the current literal and the match byte.
+ const uint8_t match_byte = mf->buffer[
+ mf->read_pos - coder->reps[0] - 1
+ - mf->read_ahead];
+ literal_matched(&coder->rc, subcoder, match_byte, cur_byte);
+ }
+
+ update_literal(coder->state);
+}
+
+
+//////////////////
+// Match length //
+//////////////////
+
+static void
+length_update_prices(lzma_length_encoder *lc, const uint32_t pos_state)
+{
+ const uint32_t table_size = lc->table_size;
+ lc->counters[pos_state] = table_size;
+
+ const uint32_t a0 = rc_bit_0_price(lc->choice);
+ const uint32_t a1 = rc_bit_1_price(lc->choice);
+ const uint32_t b0 = a1 + rc_bit_0_price(lc->choice2);
+ const uint32_t b1 = a1 + rc_bit_1_price(lc->choice2);
+ uint32_t *const prices = lc->prices[pos_state];
+
+ uint32_t i;
+ for (i = 0; i < table_size && i < LEN_LOW_SYMBOLS; ++i)
+ prices[i] = a0 + rc_bittree_price(lc->low[pos_state],
+ LEN_LOW_BITS, i);
+
+ for (; i < table_size && i < LEN_LOW_SYMBOLS + LEN_MID_SYMBOLS; ++i)
+ prices[i] = b0 + rc_bittree_price(lc->mid[pos_state],
+ LEN_MID_BITS, i - LEN_LOW_SYMBOLS);
+
+ for (; i < table_size; ++i)
+ prices[i] = b1 + rc_bittree_price(lc->high, LEN_HIGH_BITS,
+ i - LEN_LOW_SYMBOLS - LEN_MID_SYMBOLS);
+
+ return;
+}
+
+
+static inline void
+length(lzma_range_encoder *rc, lzma_length_encoder *lc,
+ const uint32_t pos_state, uint32_t len, const bool fast_mode)
+{
+ assert(len <= MATCH_LEN_MAX);
+ len -= MATCH_LEN_MIN;
+
+ if (len < LEN_LOW_SYMBOLS) {
+ rc_bit(rc, &lc->choice, 0);
+ rc_bittree(rc, lc->low[pos_state], LEN_LOW_BITS, len);
+ } else {
+ rc_bit(rc, &lc->choice, 1);
+ len -= LEN_LOW_SYMBOLS;
+
+ if (len < LEN_MID_SYMBOLS) {
+ rc_bit(rc, &lc->choice2, 0);
+ rc_bittree(rc, lc->mid[pos_state], LEN_MID_BITS, len);
+ } else {
+ rc_bit(rc, &lc->choice2, 1);
+ len -= LEN_MID_SYMBOLS;
+ rc_bittree(rc, lc->high, LEN_HIGH_BITS, len);
+ }
+ }
+
+ // Only getoptimum uses the prices so don't update the table when
+ // in fast mode.
+ if (!fast_mode)
+ if (--lc->counters[pos_state] == 0)
+ length_update_prices(lc, pos_state);
+}
+
+
+///////////
+// Match //
+///////////
+
+static inline void
+match(lzma_coder *coder, const uint32_t pos_state,
+ const uint32_t distance, const uint32_t len)
+{
+ update_match(coder->state);
+
+ length(&coder->rc, &coder->match_len_encoder, pos_state, len,
+ coder->fast_mode);
+
+ const uint32_t pos_slot = get_pos_slot(distance);
+ const uint32_t len_to_pos_state = get_len_to_pos_state(len);
+ rc_bittree(&coder->rc, coder->pos_slot[len_to_pos_state],
+ POS_SLOT_BITS, pos_slot);
+
+ if (pos_slot >= START_POS_MODEL_INDEX) {
+ const uint32_t footer_bits = (pos_slot >> 1) - 1;
+ const uint32_t base = (2 | (pos_slot & 1)) << footer_bits;
+ const uint32_t pos_reduced = distance - base;
+
+ if (pos_slot < END_POS_MODEL_INDEX) {
+ // Careful here: base - pos_slot - 1 can be -1, but
+ // rc_bittree_reverse starts at probs[1], not probs[0].
+ rc_bittree_reverse(&coder->rc,
+ coder->pos_special + base - pos_slot - 1,
+ footer_bits, pos_reduced);
+ } else {
+ rc_direct(&coder->rc, pos_reduced >> ALIGN_BITS,
+ footer_bits - ALIGN_BITS);
+ rc_bittree_reverse(
+ &coder->rc, coder->pos_align,
+ ALIGN_BITS, pos_reduced & ALIGN_MASK);
+ ++coder->align_price_count;
+ }
+ }
+
+ coder->reps[3] = coder->reps[2];
+ coder->reps[2] = coder->reps[1];
+ coder->reps[1] = coder->reps[0];
+ coder->reps[0] = distance;
+ ++coder->match_price_count;
+}
+
+
+////////////////////
+// Repeated match //
+////////////////////
+
+static inline void
+rep_match(lzma_coder *coder, const uint32_t pos_state,
+ const uint32_t rep, const uint32_t len)
+{
+ if (rep == 0) {
+ rc_bit(&coder->rc, &coder->is_rep0[coder->state], 0);
+ rc_bit(&coder->rc,
+ &coder->is_rep0_long[coder->state][pos_state],
+ len != 1);
+ } else {
+ const uint32_t distance = coder->reps[rep];
+ rc_bit(&coder->rc, &coder->is_rep0[coder->state], 1);
+
+ if (rep == 1) {
+ rc_bit(&coder->rc, &coder->is_rep1[coder->state], 0);
+ } else {
+ rc_bit(&coder->rc, &coder->is_rep1[coder->state], 1);
+ rc_bit(&coder->rc, &coder->is_rep2[coder->state],
+ rep - 2);
+
+ if (rep == 3)
+ coder->reps[3] = coder->reps[2];
+
+ coder->reps[2] = coder->reps[1];
+ }
+
+ coder->reps[1] = coder->reps[0];
+ coder->reps[0] = distance;
+ }
+
+ if (len == 1) {
+ update_short_rep(coder->state);
+ } else {
+ length(&coder->rc, &coder->rep_len_encoder, pos_state, len,
+ coder->fast_mode);
+ update_long_rep(coder->state);
+ }
+}
+
+
+//////////
+// Main //
+//////////
+
+static void
+encode_symbol(lzma_coder *coder, lzma_mf *mf,
+ uint32_t back, uint32_t len, uint32_t position)
+{
+ const uint32_t pos_state = position & coder->pos_mask;
+
+ if (back == UINT32_MAX) {
+ // Literal i.e. eight-bit byte
+ assert(len == 1);
+ rc_bit(&coder->rc,
+ &coder->is_match[coder->state][pos_state], 0);
+ literal(coder, mf, position);
+ } else {
+ // Some type of match
+ rc_bit(&coder->rc,
+ &coder->is_match[coder->state][pos_state], 1);
+
+ if (back < REP_DISTANCES) {
+ // It's a repeated match i.e. the same distance
+ // has been used earlier.
+ rc_bit(&coder->rc, &coder->is_rep[coder->state], 1);
+ rep_match(coder, pos_state, back, len);
+ } else {
+ // Normal match
+ rc_bit(&coder->rc, &coder->is_rep[coder->state], 0);
+ match(coder, pos_state, back - REP_DISTANCES, len);
+ }
+ }
+
+ assert(mf->read_ahead >= len);
+ mf->read_ahead -= len;
+}
+
+
+static bool
+encode_init(lzma_coder *coder, lzma_mf *mf)
+{
+ assert(mf_position(mf) == 0);
+
+ if (mf->read_pos == mf->read_limit) {
+ if (mf->action == LZMA_RUN)
+ return false; // We cannot do anything.
+
+ // We are finishing (we cannot get here when flushing).
+ assert(mf->write_pos == mf->read_pos);
+ assert(mf->action == LZMA_FINISH);
+ } else {
+ // Do the actual initialization. The first LZMA symbol must
+ // always be a literal.
+ mf_skip(mf, 1);
+ mf->read_ahead = 0;
+ rc_bit(&coder->rc, &coder->is_match[0][0], 0);
+ rc_bittree(&coder->rc, coder->literal[0], 8, mf->buffer[0]);
+ }
+
+ // Initialization is done (except if empty file).
+ coder->is_initialized = true;
+
+ return true;
+}
+
+
+static void
+encode_eopm(lzma_coder *coder, uint32_t position)
+{
+ const uint32_t pos_state = position & coder->pos_mask;
+ rc_bit(&coder->rc, &coder->is_match[coder->state][pos_state], 1);
+ rc_bit(&coder->rc, &coder->is_rep[coder->state], 0);
+ match(coder, pos_state, UINT32_MAX, MATCH_LEN_MIN);
+}
+
+
+/// Number of bytes that a single encoding loop in lzma_lzma_encode() can
+/// consume from the dictionary. This limit comes from lzma_lzma_optimum()
+/// and may need to be updated if that function is significantly modified.
+#define LOOP_INPUT_MAX (OPTS + 1)
+
+
+extern lzma_ret
+lzma_lzma_encode(lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint8_t *restrict out, size_t *restrict out_pos,
+ size_t out_size, uint32_t limit)
+{
+ // Initialize the stream if no data has been encoded yet.
+ if (!coder->is_initialized && !encode_init(coder, mf))
+ return LZMA_OK;
+
+ // Get the lowest bits of the uncompressed offset from the LZ layer.
+ uint32_t position = mf_position(mf);
+
+ while (true) {
+ // Encode pending bits, if any. Calling this before encoding
+ // the next symbol is needed only with plain LZMA, since
+ // LZMA2 always provides big enough buffer to flush
+ // everything out from the range encoder. For the same reason,
+ // rc_encode() never returns true when this function is used
+ // as part of LZMA2 encoder.
+ if (rc_encode(&coder->rc, out, out_pos, out_size)) {
+ assert(limit == UINT32_MAX);
+ return LZMA_OK;
+ }
+
+ // With LZMA2 we need to take care that compressed size of
+ // a chunk doesn't get too big.
+ // TODO
+ if (limit != UINT32_MAX
+ && (mf->read_pos - mf->read_ahead >= limit
+ || *out_pos + rc_pending(&coder->rc)
+ >= LZMA2_CHUNK_MAX
+ - LOOP_INPUT_MAX))
+ break;
+
+ // Check that there is some input to process.
+ if (mf->read_pos >= mf->read_limit) {
+ if (mf->action == LZMA_RUN)
+ return LZMA_OK;
+
+ if (mf->read_ahead == 0)
+ break;
+ }
+
+ // Get optimal match (repeat position and length).
+ // Value ranges for pos:
+ // - [0, REP_DISTANCES): repeated match
+ // - [REP_DISTANCES, UINT32_MAX):
+ // match at (pos - REP_DISTANCES)
+ // - UINT32_MAX: not a match but a literal
+ // Value ranges for len:
+ // - [MATCH_LEN_MIN, MATCH_LEN_MAX]
+ uint32_t len;
+ uint32_t back;
+
+ if (coder->fast_mode)
+ lzma_lzma_optimum_fast(coder, mf, &back, &len);
+ else
+ lzma_lzma_optimum_normal(
+ coder, mf, &back, &len, position);
+
+ encode_symbol(coder, mf, back, len, position);
+
+ position += len;
+ }
+
+ if (!coder->is_flushed) {
+ coder->is_flushed = true;
+
+ // We don't support encoding plain LZMA streams without EOPM,
+ // and LZMA2 doesn't use EOPM at LZMA level.
+ if (limit == UINT32_MAX)
+ encode_eopm(coder, position);
+
+ // Flush the remaining bytes from the range encoder.
+ rc_flush(&coder->rc);
+
+ // Copy the remaining bytes to the output buffer. If there
+ // isn't enough output space, we will copy out the remaining
+ // bytes on the next call to this function by using
+ // the rc_encode() call in the encoding loop above.
+ if (rc_encode(&coder->rc, out, out_pos, out_size)) {
+ assert(limit == UINT32_MAX);
+ return LZMA_OK;
+ }
+ }
+
+ // Make it ready for the next LZMA2 chunk.
+ coder->is_flushed = false;
+
+ return LZMA_STREAM_END;
+}
+
+
+static lzma_ret
+lzma_encode(lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint8_t *restrict out, size_t *restrict out_pos,
+ size_t out_size)
+{
+ // Plain LZMA has no support for sync-flushing.
+ if (unlikely(mf->action == LZMA_SYNC_FLUSH))
+ return LZMA_OPTIONS_ERROR;
+
+ return lzma_lzma_encode(coder, mf, out, out_pos, out_size, UINT32_MAX);
+}
+
+
+////////////////////
+// Initialization //
+////////////////////
+
+static bool
+is_options_valid(const lzma_options_lzma *options)
+{
+ // Validate some of the options. LZ encoder validates nice_len too
+ // but we need a valid value here earlier.
+ return is_lclppb_valid(options)
+ && options->nice_len >= MATCH_LEN_MIN
+ && options->nice_len <= MATCH_LEN_MAX
+ && (options->mode == LZMA_MODE_FAST
+ || options->mode == LZMA_MODE_NORMAL);
+}
+
+
+static void
+set_lz_options(lzma_lz_options *lz_options, const lzma_options_lzma *options)
+{
+ // LZ encoder initialization does the validation for these so we
+ // don't need to validate here.
+ lz_options->before_size = OPTS;
+ lz_options->dict_size = options->dict_size;
+ lz_options->after_size = LOOP_INPUT_MAX;
+ lz_options->match_len_max = MATCH_LEN_MAX;
+ lz_options->nice_len = options->nice_len;
+ lz_options->match_finder = options->mf;
+ lz_options->depth = options->depth;
+ lz_options->preset_dict = options->preset_dict;
+ lz_options->preset_dict_size = options->preset_dict_size;
+ return;
+}
+
+
+static void
+length_encoder_reset(lzma_length_encoder *lencoder,
+ const uint32_t num_pos_states, const bool fast_mode)
+{
+ bit_reset(lencoder->choice);
+ bit_reset(lencoder->choice2);
+
+ for (size_t pos_state = 0; pos_state < num_pos_states; ++pos_state) {
+ bittree_reset(lencoder->low[pos_state], LEN_LOW_BITS);
+ bittree_reset(lencoder->mid[pos_state], LEN_MID_BITS);
+ }
+
+ bittree_reset(lencoder->high, LEN_HIGH_BITS);
+
+ if (!fast_mode)
+ for (size_t pos_state = 0; pos_state < num_pos_states;
+ ++pos_state)
+ length_update_prices(lencoder, pos_state);
+
+ return;
+}
+
+
+extern lzma_ret
+lzma_lzma_encoder_reset(lzma_coder *coder, const lzma_options_lzma *options)
+{
+ if (!is_options_valid(options))
+ return LZMA_OPTIONS_ERROR;
+
+ coder->pos_mask = (1U << options->pb) - 1;
+ coder->literal_context_bits = options->lc;
+ coder->literal_pos_mask = (1U << options->lp) - 1;
+
+ // Range coder
+ rc_reset(&coder->rc);
+
+ // State
+ coder->state = STATE_LIT_LIT;
+ for (size_t i = 0; i < REP_DISTANCES; ++i)
+ coder->reps[i] = 0;
+
+ literal_init(coder->literal, options->lc, options->lp);
+
+ // Bit encoders
+ for (size_t i = 0; i < STATES; ++i) {
+ for (size_t j = 0; j <= coder->pos_mask; ++j) {
+ bit_reset(coder->is_match[i][j]);
+ bit_reset(coder->is_rep0_long[i][j]);
+ }
+
+ bit_reset(coder->is_rep[i]);
+ bit_reset(coder->is_rep0[i]);
+ bit_reset(coder->is_rep1[i]);
+ bit_reset(coder->is_rep2[i]);
+ }
+
+ for (size_t i = 0; i < FULL_DISTANCES - END_POS_MODEL_INDEX; ++i)
+ bit_reset(coder->pos_special[i]);
+
+ // Bit tree encoders
+ for (size_t i = 0; i < LEN_TO_POS_STATES; ++i)
+ bittree_reset(coder->pos_slot[i], POS_SLOT_BITS);
+
+ bittree_reset(coder->pos_align, ALIGN_BITS);
+
+ // Length encoders
+ length_encoder_reset(&coder->match_len_encoder,
+ 1U << options->pb, coder->fast_mode);
+
+ length_encoder_reset(&coder->rep_len_encoder,
+ 1U << options->pb, coder->fast_mode);
+
+ // Price counts are incremented every time appropriate probabilities
+ // are changed. price counts are set to zero when the price tables
+ // are updated, which is done when the appropriate price counts have
+ // big enough value, and lzma_mf.read_ahead == 0 which happens at
+ // least every OPTS (a few thousand) possible price count increments.
+ //
+ // By resetting price counts to UINT32_MAX / 2, we make sure that the
+ // price tables will be initialized before they will be used (since
+ // the value is definitely big enough), and that it is OK to increment
+ // price counts without risk of integer overflow (since UINT32_MAX / 2
+ // is small enough). The current code doesn't increment price counts
+ // before initializing price tables, but it maybe done in future if
+ // we add support for saving the state between LZMA2 chunks.
+ coder->match_price_count = UINT32_MAX / 2;
+ coder->align_price_count = UINT32_MAX / 2;
+
+ coder->opts_end_index = 0;
+ coder->opts_current_index = 0;
+
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_lzma_encoder_create(lzma_coder **coder_ptr, lzma_allocator *allocator,
+ const lzma_options_lzma *options, lzma_lz_options *lz_options)
+{
+ // Allocate lzma_coder if it wasn't already allocated.
+ if (*coder_ptr == NULL) {
+ *coder_ptr = lzma_alloc(sizeof(lzma_coder), allocator);
+ if (*coder_ptr == NULL)
+ return LZMA_MEM_ERROR;
+ }
+
+ lzma_coder *coder = *coder_ptr;
+
+ // Set compression mode. We haven't validates the options yet,
+ // but it's OK here, since nothing bad happens with invalid
+ // options in the code below, and they will get rejected by
+ // lzma_lzma_encoder_reset() call at the end of this function.
+ switch (options->mode) {
+ case LZMA_MODE_FAST:
+ coder->fast_mode = true;
+ break;
+
+ case LZMA_MODE_NORMAL: {
+ coder->fast_mode = false;
+
+ // Set dist_table_size.
+ // Round the dictionary size up to next 2^n.
+ uint32_t log_size = 0;
+ while ((UINT32_C(1) << log_size) < options->dict_size)
+ ++log_size;
+
+ coder->dist_table_size = log_size * 2;
+
+ // Length encoders' price table size
+ coder->match_len_encoder.table_size
+ = options->nice_len + 1 - MATCH_LEN_MIN;
+ coder->rep_len_encoder.table_size
+ = options->nice_len + 1 - MATCH_LEN_MIN;
+ break;
+ }
+
+ default:
+ return LZMA_OPTIONS_ERROR;
+ }
+
+ // We don't need to write the first byte as literal if there is
+ // a non-empty preset dictionary. encode_init() wouldn't even work
+ // if there is a non-empty preset dictionary, because encode_init()
+ // assumes that position is zero and previous byte is also zero.
+ coder->is_initialized = options->preset_dict != NULL
+ && options->preset_dict_size > 0;
+ coder->is_flushed = false;
+
+ set_lz_options(lz_options, options);
+
+ return lzma_lzma_encoder_reset(coder, options);
+}
+
+
+static lzma_ret
+lzma_encoder_init(lzma_lz_encoder *lz, lzma_allocator *allocator,
+ const void *options, lzma_lz_options *lz_options)
+{
+ lz->code = &lzma_encode;
+ return lzma_lzma_encoder_create(
+ &lz->coder, allocator, options, lz_options);
+}
+
+
+extern lzma_ret
+lzma_lzma_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ return lzma_lz_encoder_init(
+ next, allocator, filters, &lzma_encoder_init);
+}
+
+
+extern uint64_t
+lzma_lzma_encoder_memusage(const void *options)
+{
+ if (!is_options_valid(options))
+ return UINT64_MAX;
+
+ lzma_lz_options lz_options;
+ set_lz_options(&lz_options, options);
+
+ const uint64_t lz_memusage = lzma_lz_encoder_memusage(&lz_options);
+ if (lz_memusage == UINT64_MAX)
+ return UINT64_MAX;
+
+ return (uint64_t)(sizeof(lzma_coder)) + lz_memusage;
+}
+
+
+extern bool
+lzma_lzma_lclppb_encode(const lzma_options_lzma *options, uint8_t *byte)
+{
+ if (!is_lclppb_valid(options))
+ return true;
+
+ *byte = (options->pb * 5 + options->lp) * 9 + options->lc;
+ assert(*byte <= (4 * 5 + 4) * 9 + 8);
+
+ return false;
+}
+
+
+#ifdef HAVE_ENCODER_LZMA1
+extern lzma_ret
+lzma_lzma_props_encode(const void *options, uint8_t *out)
+{
+ const lzma_options_lzma *const opt = options;
+
+ if (lzma_lzma_lclppb_encode(opt, out))
+ return LZMA_PROG_ERROR;
+
+ unaligned_write32le(out + 1, opt->dict_size);
+
+ return LZMA_OK;
+}
+#endif
+
+
+extern LZMA_API(lzma_bool)
+lzma_mode_is_supported(lzma_mode mode)
+{
+ return mode == LZMA_MODE_FAST || mode == LZMA_MODE_NORMAL;
+}
diff --git a/src/liblzma/lzma/lzma_encoder.h b/src/liblzma/lzma/lzma_encoder.h
new file mode 100644
index 0000000..835e1f5
--- /dev/null
+++ b/src/liblzma/lzma/lzma_encoder.h
@@ -0,0 +1,54 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_encoder.h
+/// \brief LZMA encoder API
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZMA_ENCODER_H
+#define LZMA_LZMA_ENCODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_lzma_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+
+extern uint64_t lzma_lzma_encoder_memusage(const void *options);
+
+extern lzma_ret lzma_lzma_props_encode(const void *options, uint8_t *out);
+
+
+/// Encodes lc/lp/pb into one byte. Returns false on success and true on error.
+extern bool lzma_lzma_lclppb_encode(
+ const lzma_options_lzma *options, uint8_t *byte);
+
+
+#ifdef LZMA_LZ_ENCODER_H
+
+/// Initializes raw LZMA encoder; this is used by LZMA2.
+extern lzma_ret lzma_lzma_encoder_create(
+ lzma_coder **coder_ptr, lzma_allocator *allocator,
+ const lzma_options_lzma *options, lzma_lz_options *lz_options);
+
+
+/// Resets an already initialized LZMA encoder; this is used by LZMA2.
+extern lzma_ret lzma_lzma_encoder_reset(
+ lzma_coder *coder, const lzma_options_lzma *options);
+
+
+extern lzma_ret lzma_lzma_encode(lzma_coder *restrict coder,
+ lzma_mf *restrict mf, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size,
+ uint32_t read_limit);
+
+#endif
+
+#endif
diff --git a/src/liblzma/lzma/lzma_encoder_optimum_fast.c b/src/liblzma/lzma/lzma_encoder_optimum_fast.c
new file mode 100644
index 0000000..f835f69
--- /dev/null
+++ b/src/liblzma/lzma/lzma_encoder_optimum_fast.c
@@ -0,0 +1,179 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_encoder_optimum_fast.c
+//
+// Author: Igor Pavlov
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lzma_encoder_private.h"
+
+
+#define change_pair(small_dist, big_dist) \
+ (((big_dist) >> 7) > (small_dist))
+
+
+extern void
+lzma_lzma_optimum_fast(lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint32_t *restrict back_res, uint32_t *restrict len_res)
+{
+ const uint32_t nice_len = mf->nice_len;
+
+ uint32_t len_main;
+ uint32_t matches_count;
+ if (mf->read_ahead == 0) {
+ len_main = mf_find(mf, &matches_count, coder->matches);
+ } else {
+ assert(mf->read_ahead == 1);
+ len_main = coder->longest_match_length;
+ matches_count = coder->matches_count;
+ }
+
+ const uint8_t *buf = mf_ptr(mf) - 1;
+ const uint32_t buf_avail = my_min(mf_avail(mf) + 1, MATCH_LEN_MAX);
+
+ if (buf_avail < 2) {
+ // There's not enough input left to encode a match.
+ *back_res = UINT32_MAX;
+ *len_res = 1;
+ return;
+ }
+
+ // Look for repeated matches; scan the previous four match distances
+ uint32_t rep_len = 0;
+ uint32_t rep_index = 0;
+
+ for (uint32_t i = 0; i < REP_DISTANCES; ++i) {
+ // Pointer to the beginning of the match candidate
+ const uint8_t *const buf_back = buf - coder->reps[i] - 1;
+
+ // If the first two bytes (2 == MATCH_LEN_MIN) do not match,
+ // this rep is not useful.
+ if (not_equal_16(buf, buf_back))
+ continue;
+
+ // The first two bytes matched.
+ // Calculate the length of the match.
+ uint32_t len;
+ for (len = 2; len < buf_avail
+ && buf[len] == buf_back[len]; ++len) ;
+
+ // If we have found a repeated match that is at least
+ // nice_len long, return it immediately.
+ if (len >= nice_len) {
+ *back_res = i;
+ *len_res = len;
+ mf_skip(mf, len - 1);
+ return;
+ }
+
+ if (len > rep_len) {
+ rep_index = i;
+ rep_len = len;
+ }
+ }
+
+ // We didn't find a long enough repeated match. Encode it as a normal
+ // match if the match length is at least nice_len.
+ if (len_main >= nice_len) {
+ *back_res = coder->matches[matches_count - 1].dist
+ + REP_DISTANCES;
+ *len_res = len_main;
+ mf_skip(mf, len_main - 1);
+ return;
+ }
+
+ uint32_t back_main = 0;
+ if (len_main >= 2) {
+ back_main = coder->matches[matches_count - 1].dist;
+
+ while (matches_count > 1 && len_main ==
+ coder->matches[matches_count - 2].len + 1) {
+ if (!change_pair(coder->matches[
+ matches_count - 2].dist,
+ back_main))
+ break;
+
+ --matches_count;
+ len_main = coder->matches[matches_count - 1].len;
+ back_main = coder->matches[matches_count - 1].dist;
+ }
+
+ if (len_main == 2 && back_main >= 0x80)
+ len_main = 1;
+ }
+
+ if (rep_len >= 2) {
+ if (rep_len + 1 >= len_main
+ || (rep_len + 2 >= len_main
+ && back_main > (UINT32_C(1) << 9))
+ || (rep_len + 3 >= len_main
+ && back_main > (UINT32_C(1) << 15))) {
+ *back_res = rep_index;
+ *len_res = rep_len;
+ mf_skip(mf, rep_len - 1);
+ return;
+ }
+ }
+
+ if (len_main < 2 || buf_avail <= 2) {
+ *back_res = UINT32_MAX;
+ *len_res = 1;
+ return;
+ }
+
+ // Get the matches for the next byte. If we find a better match,
+ // the current byte is encoded as a literal.
+ coder->longest_match_length = mf_find(mf,
+ &coder->matches_count, coder->matches);
+
+ if (coder->longest_match_length >= 2) {
+ const uint32_t new_dist = coder->matches[
+ coder->matches_count - 1].dist;
+
+ if ((coder->longest_match_length >= len_main
+ && new_dist < back_main)
+ || (coder->longest_match_length == len_main + 1
+ && !change_pair(back_main, new_dist))
+ || (coder->longest_match_length > len_main + 1)
+ || (coder->longest_match_length + 1 >= len_main
+ && len_main >= 3
+ && change_pair(new_dist, back_main))) {
+ *back_res = UINT32_MAX;
+ *len_res = 1;
+ return;
+ }
+ }
+
+ // In contrast to LZMA SDK, dictionary could not have been moved
+ // between mf_find() calls, thus it is safe to just increment
+ // the old buf pointer instead of recalculating it with mf_ptr().
+ ++buf;
+
+ const uint32_t limit = len_main - 1;
+
+ for (uint32_t i = 0; i < REP_DISTANCES; ++i) {
+ const uint8_t *const buf_back = buf - coder->reps[i] - 1;
+
+ if (not_equal_16(buf, buf_back))
+ continue;
+
+ uint32_t len;
+ for (len = 2; len < limit
+ && buf[len] == buf_back[len]; ++len) ;
+
+ if (len >= limit) {
+ *back_res = UINT32_MAX;
+ *len_res = 1;
+ return;
+ }
+ }
+
+ *back_res = back_main + REP_DISTANCES;
+ *len_res = len_main;
+ mf_skip(mf, len_main - 2);
+ return;
+}
diff --git a/src/liblzma/lzma/lzma_encoder_optimum_normal.c b/src/liblzma/lzma/lzma_encoder_optimum_normal.c
new file mode 100644
index 0000000..7e85649
--- /dev/null
+++ b/src/liblzma/lzma/lzma_encoder_optimum_normal.c
@@ -0,0 +1,868 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_encoder_optimum_normal.c
+//
+// Author: Igor Pavlov
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "lzma_encoder_private.h"
+#include "fastpos.h"
+
+
+////////////
+// Prices //
+////////////
+
+static uint32_t
+get_literal_price(const lzma_coder *const coder, const uint32_t pos,
+ const uint32_t prev_byte, const bool match_mode,
+ uint32_t match_byte, uint32_t symbol)
+{
+ const probability *const subcoder = literal_subcoder(coder->literal,
+ coder->literal_context_bits, coder->literal_pos_mask,
+ pos, prev_byte);
+
+ uint32_t price = 0;
+
+ if (!match_mode) {
+ price = rc_bittree_price(subcoder, 8, symbol);
+ } else {
+ uint32_t offset = 0x100;
+ symbol += UINT32_C(1) << 8;
+
+ do {
+ match_byte <<= 1;
+
+ const uint32_t match_bit = match_byte & offset;
+ const uint32_t subcoder_index
+ = offset + match_bit + (symbol >> 8);
+ const uint32_t bit = (symbol >> 7) & 1;
+ price += rc_bit_price(subcoder[subcoder_index], bit);
+
+ symbol <<= 1;
+ offset &= ~(match_byte ^ symbol);
+
+ } while (symbol < (UINT32_C(1) << 16));
+ }
+
+ return price;
+}
+
+
+static inline uint32_t
+get_len_price(const lzma_length_encoder *const lencoder,
+ const uint32_t len, const uint32_t pos_state)
+{
+ // NOTE: Unlike the other price tables, length prices are updated
+ // in lzma_encoder.c
+ return lencoder->prices[pos_state][len - MATCH_LEN_MIN];
+}
+
+
+static inline uint32_t
+get_short_rep_price(const lzma_coder *const coder,
+ const lzma_lzma_state state, const uint32_t pos_state)
+{
+ return rc_bit_0_price(coder->is_rep0[state])
+ + rc_bit_0_price(coder->is_rep0_long[state][pos_state]);
+}
+
+
+static inline uint32_t
+get_pure_rep_price(const lzma_coder *const coder, const uint32_t rep_index,
+ const lzma_lzma_state state, uint32_t pos_state)
+{
+ uint32_t price;
+
+ if (rep_index == 0) {
+ price = rc_bit_0_price(coder->is_rep0[state]);
+ price += rc_bit_1_price(coder->is_rep0_long[state][pos_state]);
+ } else {
+ price = rc_bit_1_price(coder->is_rep0[state]);
+
+ if (rep_index == 1) {
+ price += rc_bit_0_price(coder->is_rep1[state]);
+ } else {
+ price += rc_bit_1_price(coder->is_rep1[state]);
+ price += rc_bit_price(coder->is_rep2[state],
+ rep_index - 2);
+ }
+ }
+
+ return price;
+}
+
+
+static inline uint32_t
+get_rep_price(const lzma_coder *const coder, const uint32_t rep_index,
+ const uint32_t len, const lzma_lzma_state state,
+ const uint32_t pos_state)
+{
+ return get_len_price(&coder->rep_len_encoder, len, pos_state)
+ + get_pure_rep_price(coder, rep_index, state, pos_state);
+}
+
+
+static inline uint32_t
+get_pos_len_price(const lzma_coder *const coder, const uint32_t pos,
+ const uint32_t len, const uint32_t pos_state)
+{
+ const uint32_t len_to_pos_state = get_len_to_pos_state(len);
+ uint32_t price;
+
+ if (pos < FULL_DISTANCES) {
+ price = coder->distances_prices[len_to_pos_state][pos];
+ } else {
+ const uint32_t pos_slot = get_pos_slot_2(pos);
+ price = coder->pos_slot_prices[len_to_pos_state][pos_slot]
+ + coder->align_prices[pos & ALIGN_MASK];
+ }
+
+ price += get_len_price(&coder->match_len_encoder, len, pos_state);
+
+ return price;
+}
+
+
+static void
+fill_distances_prices(lzma_coder *coder)
+{
+ for (uint32_t len_to_pos_state = 0;
+ len_to_pos_state < LEN_TO_POS_STATES;
+ ++len_to_pos_state) {
+
+ uint32_t *const pos_slot_prices
+ = coder->pos_slot_prices[len_to_pos_state];
+
+ // Price to encode the pos_slot.
+ for (uint32_t pos_slot = 0;
+ pos_slot < coder->dist_table_size; ++pos_slot)
+ pos_slot_prices[pos_slot] = rc_bittree_price(
+ coder->pos_slot[len_to_pos_state],
+ POS_SLOT_BITS, pos_slot);
+
+ // For matches with distance >= FULL_DISTANCES, add the price
+ // of the direct bits part of the match distance. (Align bits
+ // are handled by fill_align_prices()).
+ for (uint32_t pos_slot = END_POS_MODEL_INDEX;
+ pos_slot < coder->dist_table_size; ++pos_slot)
+ pos_slot_prices[pos_slot] += rc_direct_price(
+ ((pos_slot >> 1) - 1) - ALIGN_BITS);
+
+ // Distances in the range [0, 3] are fully encoded with
+ // pos_slot, so they are used for coder->distances_prices
+ // as is.
+ for (uint32_t i = 0; i < START_POS_MODEL_INDEX; ++i)
+ coder->distances_prices[len_to_pos_state][i]
+ = pos_slot_prices[i];
+ }
+
+ // Distances in the range [4, 127] depend on pos_slot and pos_special.
+ // We do this in a loop separate from the above loop to avoid
+ // redundant calls to get_pos_slot().
+ for (uint32_t i = START_POS_MODEL_INDEX; i < FULL_DISTANCES; ++i) {
+ const uint32_t pos_slot = get_pos_slot(i);
+ const uint32_t footer_bits = ((pos_slot >> 1) - 1);
+ const uint32_t base = (2 | (pos_slot & 1)) << footer_bits;
+ const uint32_t price = rc_bittree_reverse_price(
+ coder->pos_special + base - pos_slot - 1,
+ footer_bits, i - base);
+
+ for (uint32_t len_to_pos_state = 0;
+ len_to_pos_state < LEN_TO_POS_STATES;
+ ++len_to_pos_state)
+ coder->distances_prices[len_to_pos_state][i]
+ = price + coder->pos_slot_prices[
+ len_to_pos_state][pos_slot];
+ }
+
+ coder->match_price_count = 0;
+ return;
+}
+
+
+static void
+fill_align_prices(lzma_coder *coder)
+{
+ for (uint32_t i = 0; i < ALIGN_TABLE_SIZE; ++i)
+ coder->align_prices[i] = rc_bittree_reverse_price(
+ coder->pos_align, ALIGN_BITS, i);
+
+ coder->align_price_count = 0;
+ return;
+}
+
+
+/////////////
+// Optimal //
+/////////////
+
+static inline void
+make_literal(lzma_optimal *optimal)
+{
+ optimal->back_prev = UINT32_MAX;
+ optimal->prev_1_is_literal = false;
+}
+
+
+static inline void
+make_short_rep(lzma_optimal *optimal)
+{
+ optimal->back_prev = 0;
+ optimal->prev_1_is_literal = false;
+}
+
+
+#define is_short_rep(optimal) \
+ ((optimal).back_prev == 0)
+
+
+static void
+backward(lzma_coder *restrict coder, uint32_t *restrict len_res,
+ uint32_t *restrict back_res, uint32_t cur)
+{
+ coder->opts_end_index = cur;
+
+ uint32_t pos_mem = coder->opts[cur].pos_prev;
+ uint32_t back_mem = coder->opts[cur].back_prev;
+
+ do {
+ if (coder->opts[cur].prev_1_is_literal) {
+ make_literal(&coder->opts[pos_mem]);
+ coder->opts[pos_mem].pos_prev = pos_mem - 1;
+
+ if (coder->opts[cur].prev_2) {
+ coder->opts[pos_mem - 1].prev_1_is_literal
+ = false;
+ coder->opts[pos_mem - 1].pos_prev
+ = coder->opts[cur].pos_prev_2;
+ coder->opts[pos_mem - 1].back_prev
+ = coder->opts[cur].back_prev_2;
+ }
+ }
+
+ const uint32_t pos_prev = pos_mem;
+ const uint32_t back_cur = back_mem;
+
+ back_mem = coder->opts[pos_prev].back_prev;
+ pos_mem = coder->opts[pos_prev].pos_prev;
+
+ coder->opts[pos_prev].back_prev = back_cur;
+ coder->opts[pos_prev].pos_prev = cur;
+ cur = pos_prev;
+
+ } while (cur != 0);
+
+ coder->opts_current_index = coder->opts[0].pos_prev;
+ *len_res = coder->opts[0].pos_prev;
+ *back_res = coder->opts[0].back_prev;
+
+ return;
+}
+
+
+//////////
+// Main //
+//////////
+
+static inline uint32_t
+helper1(lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint32_t *restrict back_res, uint32_t *restrict len_res,
+ uint32_t position)
+{
+ const uint32_t nice_len = mf->nice_len;
+
+ uint32_t len_main;
+ uint32_t matches_count;
+
+ if (mf->read_ahead == 0) {
+ len_main = mf_find(mf, &matches_count, coder->matches);
+ } else {
+ assert(mf->read_ahead == 1);
+ len_main = coder->longest_match_length;
+ matches_count = coder->matches_count;
+ }
+
+ const uint32_t buf_avail = my_min(mf_avail(mf) + 1, MATCH_LEN_MAX);
+ if (buf_avail < 2) {
+ *back_res = UINT32_MAX;
+ *len_res = 1;
+ return UINT32_MAX;
+ }
+
+ const uint8_t *const buf = mf_ptr(mf) - 1;
+
+ uint32_t rep_lens[REP_DISTANCES];
+ uint32_t rep_max_index = 0;
+
+ for (uint32_t i = 0; i < REP_DISTANCES; ++i) {
+ const uint8_t *const buf_back = buf - coder->reps[i] - 1;
+
+ if (not_equal_16(buf, buf_back)) {
+ rep_lens[i] = 0;
+ continue;
+ }
+
+ uint32_t len_test;
+ for (len_test = 2; len_test < buf_avail
+ && buf[len_test] == buf_back[len_test];
+ ++len_test) ;
+
+ rep_lens[i] = len_test;
+ if (len_test > rep_lens[rep_max_index])
+ rep_max_index = i;
+ }
+
+ if (rep_lens[rep_max_index] >= nice_len) {
+ *back_res = rep_max_index;
+ *len_res = rep_lens[rep_max_index];
+ mf_skip(mf, *len_res - 1);
+ return UINT32_MAX;
+ }
+
+
+ if (len_main >= nice_len) {
+ *back_res = coder->matches[matches_count - 1].dist
+ + REP_DISTANCES;
+ *len_res = len_main;
+ mf_skip(mf, len_main - 1);
+ return UINT32_MAX;
+ }
+
+ const uint8_t current_byte = *buf;
+ const uint8_t match_byte = *(buf - coder->reps[0] - 1);
+
+ if (len_main < 2 && current_byte != match_byte
+ && rep_lens[rep_max_index] < 2) {
+ *back_res = UINT32_MAX;
+ *len_res = 1;
+ return UINT32_MAX;
+ }
+
+ coder->opts[0].state = coder->state;
+
+ const uint32_t pos_state = position & coder->pos_mask;
+
+ coder->opts[1].price = rc_bit_0_price(
+ coder->is_match[coder->state][pos_state])
+ + get_literal_price(coder, position, buf[-1],
+ !is_literal_state(coder->state),
+ match_byte, current_byte);
+
+ make_literal(&coder->opts[1]);
+
+ const uint32_t match_price = rc_bit_1_price(
+ coder->is_match[coder->state][pos_state]);
+ const uint32_t rep_match_price = match_price
+ + rc_bit_1_price(coder->is_rep[coder->state]);
+
+ if (match_byte == current_byte) {
+ const uint32_t short_rep_price = rep_match_price
+ + get_short_rep_price(
+ coder, coder->state, pos_state);
+
+ if (short_rep_price < coder->opts[1].price) {
+ coder->opts[1].price = short_rep_price;
+ make_short_rep(&coder->opts[1]);
+ }
+ }
+
+ const uint32_t len_end = my_max(len_main, rep_lens[rep_max_index]);
+
+ if (len_end < 2) {
+ *back_res = coder->opts[1].back_prev;
+ *len_res = 1;
+ return UINT32_MAX;
+ }
+
+ coder->opts[1].pos_prev = 0;
+
+ for (uint32_t i = 0; i < REP_DISTANCES; ++i)
+ coder->opts[0].backs[i] = coder->reps[i];
+
+ uint32_t len = len_end;
+ do {
+ coder->opts[len].price = RC_INFINITY_PRICE;
+ } while (--len >= 2);
+
+
+ for (uint32_t i = 0; i < REP_DISTANCES; ++i) {
+ uint32_t rep_len = rep_lens[i];
+ if (rep_len < 2)
+ continue;
+
+ const uint32_t price = rep_match_price + get_pure_rep_price(
+ coder, i, coder->state, pos_state);
+
+ do {
+ const uint32_t cur_and_len_price = price
+ + get_len_price(
+ &coder->rep_len_encoder,
+ rep_len, pos_state);
+
+ if (cur_and_len_price < coder->opts[rep_len].price) {
+ coder->opts[rep_len].price = cur_and_len_price;
+ coder->opts[rep_len].pos_prev = 0;
+ coder->opts[rep_len].back_prev = i;
+ coder->opts[rep_len].prev_1_is_literal = false;
+ }
+ } while (--rep_len >= 2);
+ }
+
+
+ const uint32_t normal_match_price = match_price
+ + rc_bit_0_price(coder->is_rep[coder->state]);
+
+ len = rep_lens[0] >= 2 ? rep_lens[0] + 1 : 2;
+ if (len <= len_main) {
+ uint32_t i = 0;
+ while (len > coder->matches[i].len)
+ ++i;
+
+ for(; ; ++len) {
+ const uint32_t dist = coder->matches[i].dist;
+ const uint32_t cur_and_len_price = normal_match_price
+ + get_pos_len_price(coder,
+ dist, len, pos_state);
+
+ if (cur_and_len_price < coder->opts[len].price) {
+ coder->opts[len].price = cur_and_len_price;
+ coder->opts[len].pos_prev = 0;
+ coder->opts[len].back_prev
+ = dist + REP_DISTANCES;
+ coder->opts[len].prev_1_is_literal = false;
+ }
+
+ if (len == coder->matches[i].len)
+ if (++i == matches_count)
+ break;
+ }
+ }
+
+ return len_end;
+}
+
+
+static inline uint32_t
+helper2(lzma_coder *coder, uint32_t *reps, const uint8_t *buf,
+ uint32_t len_end, uint32_t position, const uint32_t cur,
+ const uint32_t nice_len, const uint32_t buf_avail_full)
+{
+ uint32_t matches_count = coder->matches_count;
+ uint32_t new_len = coder->longest_match_length;
+ uint32_t pos_prev = coder->opts[cur].pos_prev;
+ lzma_lzma_state state;
+
+ if (coder->opts[cur].prev_1_is_literal) {
+ --pos_prev;
+
+ if (coder->opts[cur].prev_2) {
+ state = coder->opts[coder->opts[cur].pos_prev_2].state;
+
+ if (coder->opts[cur].back_prev_2 < REP_DISTANCES)
+ update_long_rep(state);
+ else
+ update_match(state);
+
+ } else {
+ state = coder->opts[pos_prev].state;
+ }
+
+ update_literal(state);
+
+ } else {
+ state = coder->opts[pos_prev].state;
+ }
+
+ if (pos_prev == cur - 1) {
+ if (is_short_rep(coder->opts[cur]))
+ update_short_rep(state);
+ else
+ update_literal(state);
+ } else {
+ uint32_t pos;
+ if (coder->opts[cur].prev_1_is_literal
+ && coder->opts[cur].prev_2) {
+ pos_prev = coder->opts[cur].pos_prev_2;
+ pos = coder->opts[cur].back_prev_2;
+ update_long_rep(state);
+ } else {
+ pos = coder->opts[cur].back_prev;
+ if (pos < REP_DISTANCES)
+ update_long_rep(state);
+ else
+ update_match(state);
+ }
+
+ if (pos < REP_DISTANCES) {
+ reps[0] = coder->opts[pos_prev].backs[pos];
+
+ uint32_t i;
+ for (i = 1; i <= pos; ++i)
+ reps[i] = coder->opts[pos_prev].backs[i - 1];
+
+ for (; i < REP_DISTANCES; ++i)
+ reps[i] = coder->opts[pos_prev].backs[i];
+
+ } else {
+ reps[0] = pos - REP_DISTANCES;
+
+ for (uint32_t i = 1; i < REP_DISTANCES; ++i)
+ reps[i] = coder->opts[pos_prev].backs[i - 1];
+ }
+ }
+
+ coder->opts[cur].state = state;
+
+ for (uint32_t i = 0; i < REP_DISTANCES; ++i)
+ coder->opts[cur].backs[i] = reps[i];
+
+ const uint32_t cur_price = coder->opts[cur].price;
+
+ const uint8_t current_byte = *buf;
+ const uint8_t match_byte = *(buf - reps[0] - 1);
+
+ const uint32_t pos_state = position & coder->pos_mask;
+
+ const uint32_t cur_and_1_price = cur_price
+ + rc_bit_0_price(coder->is_match[state][pos_state])
+ + get_literal_price(coder, position, buf[-1],
+ !is_literal_state(state), match_byte, current_byte);
+
+ bool next_is_literal = false;
+
+ if (cur_and_1_price < coder->opts[cur + 1].price) {
+ coder->opts[cur + 1].price = cur_and_1_price;
+ coder->opts[cur + 1].pos_prev = cur;
+ make_literal(&coder->opts[cur + 1]);
+ next_is_literal = true;
+ }
+
+ const uint32_t match_price = cur_price
+ + rc_bit_1_price(coder->is_match[state][pos_state]);
+ const uint32_t rep_match_price = match_price
+ + rc_bit_1_price(coder->is_rep[state]);
+
+ if (match_byte == current_byte
+ && !(coder->opts[cur + 1].pos_prev < cur
+ && coder->opts[cur + 1].back_prev == 0)) {
+
+ const uint32_t short_rep_price = rep_match_price
+ + get_short_rep_price(coder, state, pos_state);
+
+ if (short_rep_price <= coder->opts[cur + 1].price) {
+ coder->opts[cur + 1].price = short_rep_price;
+ coder->opts[cur + 1].pos_prev = cur;
+ make_short_rep(&coder->opts[cur + 1]);
+ next_is_literal = true;
+ }
+ }
+
+ if (buf_avail_full < 2)
+ return len_end;
+
+ const uint32_t buf_avail = my_min(buf_avail_full, nice_len);
+
+ if (!next_is_literal && match_byte != current_byte) { // speed optimization
+ // try literal + rep0
+ const uint8_t *const buf_back = buf - reps[0] - 1;
+ const uint32_t limit = my_min(buf_avail_full, nice_len + 1);
+
+ uint32_t len_test = 1;
+ while (len_test < limit && buf[len_test] == buf_back[len_test])
+ ++len_test;
+
+ --len_test;
+
+ if (len_test >= 2) {
+ lzma_lzma_state state_2 = state;
+ update_literal(state_2);
+
+ const uint32_t pos_state_next = (position + 1) & coder->pos_mask;
+ const uint32_t next_rep_match_price = cur_and_1_price
+ + rc_bit_1_price(coder->is_match[state_2][pos_state_next])
+ + rc_bit_1_price(coder->is_rep[state_2]);
+
+ //for (; len_test >= 2; --len_test) {
+ const uint32_t offset = cur + 1 + len_test;
+
+ while (len_end < offset)
+ coder->opts[++len_end].price = RC_INFINITY_PRICE;
+
+ const uint32_t cur_and_len_price = next_rep_match_price
+ + get_rep_price(coder, 0, len_test,
+ state_2, pos_state_next);
+
+ if (cur_and_len_price < coder->opts[offset].price) {
+ coder->opts[offset].price = cur_and_len_price;
+ coder->opts[offset].pos_prev = cur + 1;
+ coder->opts[offset].back_prev = 0;
+ coder->opts[offset].prev_1_is_literal = true;
+ coder->opts[offset].prev_2 = false;
+ }
+ //}
+ }
+ }
+
+
+ uint32_t start_len = 2; // speed optimization
+
+ for (uint32_t rep_index = 0; rep_index < REP_DISTANCES; ++rep_index) {
+ const uint8_t *const buf_back = buf - reps[rep_index] - 1;
+ if (not_equal_16(buf, buf_back))
+ continue;
+
+ uint32_t len_test;
+ for (len_test = 2; len_test < buf_avail
+ && buf[len_test] == buf_back[len_test];
+ ++len_test) ;
+
+ while (len_end < cur + len_test)
+ coder->opts[++len_end].price = RC_INFINITY_PRICE;
+
+ const uint32_t len_test_temp = len_test;
+ const uint32_t price = rep_match_price + get_pure_rep_price(
+ coder, rep_index, state, pos_state);
+
+ do {
+ const uint32_t cur_and_len_price = price
+ + get_len_price(&coder->rep_len_encoder,
+ len_test, pos_state);
+
+ if (cur_and_len_price < coder->opts[cur + len_test].price) {
+ coder->opts[cur + len_test].price = cur_and_len_price;
+ coder->opts[cur + len_test].pos_prev = cur;
+ coder->opts[cur + len_test].back_prev = rep_index;
+ coder->opts[cur + len_test].prev_1_is_literal = false;
+ }
+ } while (--len_test >= 2);
+
+ len_test = len_test_temp;
+
+ if (rep_index == 0)
+ start_len = len_test + 1;
+
+
+ uint32_t len_test_2 = len_test + 1;
+ const uint32_t limit = my_min(buf_avail_full,
+ len_test_2 + nice_len);
+ for (; len_test_2 < limit
+ && buf[len_test_2] == buf_back[len_test_2];
+ ++len_test_2) ;
+
+ len_test_2 -= len_test + 1;
+
+ if (len_test_2 >= 2) {
+ lzma_lzma_state state_2 = state;
+ update_long_rep(state_2);
+
+ uint32_t pos_state_next = (position + len_test) & coder->pos_mask;
+
+ const uint32_t cur_and_len_literal_price = price
+ + get_len_price(&coder->rep_len_encoder,
+ len_test, pos_state)
+ + rc_bit_0_price(coder->is_match[state_2][pos_state_next])
+ + get_literal_price(coder, position + len_test,
+ buf[len_test - 1], true,
+ buf_back[len_test], buf[len_test]);
+
+ update_literal(state_2);
+
+ pos_state_next = (position + len_test + 1) & coder->pos_mask;
+
+ const uint32_t next_rep_match_price = cur_and_len_literal_price
+ + rc_bit_1_price(coder->is_match[state_2][pos_state_next])
+ + rc_bit_1_price(coder->is_rep[state_2]);
+
+ //for(; len_test_2 >= 2; len_test_2--) {
+ const uint32_t offset = cur + len_test + 1 + len_test_2;
+
+ while (len_end < offset)
+ coder->opts[++len_end].price = RC_INFINITY_PRICE;
+
+ const uint32_t cur_and_len_price = next_rep_match_price
+ + get_rep_price(coder, 0, len_test_2,
+ state_2, pos_state_next);
+
+ if (cur_and_len_price < coder->opts[offset].price) {
+ coder->opts[offset].price = cur_and_len_price;
+ coder->opts[offset].pos_prev = cur + len_test + 1;
+ coder->opts[offset].back_prev = 0;
+ coder->opts[offset].prev_1_is_literal = true;
+ coder->opts[offset].prev_2 = true;
+ coder->opts[offset].pos_prev_2 = cur;
+ coder->opts[offset].back_prev_2 = rep_index;
+ }
+ //}
+ }
+ }
+
+
+ //for (uint32_t len_test = 2; len_test <= new_len; ++len_test)
+ if (new_len > buf_avail) {
+ new_len = buf_avail;
+
+ matches_count = 0;
+ while (new_len > coder->matches[matches_count].len)
+ ++matches_count;
+
+ coder->matches[matches_count++].len = new_len;
+ }
+
+
+ if (new_len >= start_len) {
+ const uint32_t normal_match_price = match_price
+ + rc_bit_0_price(coder->is_rep[state]);
+
+ while (len_end < cur + new_len)
+ coder->opts[++len_end].price = RC_INFINITY_PRICE;
+
+ uint32_t i = 0;
+ while (start_len > coder->matches[i].len)
+ ++i;
+
+ for (uint32_t len_test = start_len; ; ++len_test) {
+ const uint32_t cur_back = coder->matches[i].dist;
+ uint32_t cur_and_len_price = normal_match_price
+ + get_pos_len_price(coder,
+ cur_back, len_test, pos_state);
+
+ if (cur_and_len_price < coder->opts[cur + len_test].price) {
+ coder->opts[cur + len_test].price = cur_and_len_price;
+ coder->opts[cur + len_test].pos_prev = cur;
+ coder->opts[cur + len_test].back_prev
+ = cur_back + REP_DISTANCES;
+ coder->opts[cur + len_test].prev_1_is_literal = false;
+ }
+
+ if (len_test == coder->matches[i].len) {
+ // Try Match + Literal + Rep0
+ const uint8_t *const buf_back = buf - cur_back - 1;
+ uint32_t len_test_2 = len_test + 1;
+ const uint32_t limit = my_min(buf_avail_full,
+ len_test_2 + nice_len);
+
+ for (; len_test_2 < limit &&
+ buf[len_test_2] == buf_back[len_test_2];
+ ++len_test_2) ;
+
+ len_test_2 -= len_test + 1;
+
+ if (len_test_2 >= 2) {
+ lzma_lzma_state state_2 = state;
+ update_match(state_2);
+ uint32_t pos_state_next
+ = (position + len_test) & coder->pos_mask;
+
+ const uint32_t cur_and_len_literal_price = cur_and_len_price
+ + rc_bit_0_price(
+ coder->is_match[state_2][pos_state_next])
+ + get_literal_price(coder,
+ position + len_test,
+ buf[len_test - 1],
+ true,
+ buf_back[len_test],
+ buf[len_test]);
+
+ update_literal(state_2);
+ pos_state_next = (pos_state_next + 1) & coder->pos_mask;
+
+ const uint32_t next_rep_match_price
+ = cur_and_len_literal_price
+ + rc_bit_1_price(
+ coder->is_match[state_2][pos_state_next])
+ + rc_bit_1_price(coder->is_rep[state_2]);
+
+ // for(; len_test_2 >= 2; --len_test_2) {
+ const uint32_t offset = cur + len_test + 1 + len_test_2;
+
+ while (len_end < offset)
+ coder->opts[++len_end].price = RC_INFINITY_PRICE;
+
+ cur_and_len_price = next_rep_match_price
+ + get_rep_price(coder, 0, len_test_2,
+ state_2, pos_state_next);
+
+ if (cur_and_len_price < coder->opts[offset].price) {
+ coder->opts[offset].price = cur_and_len_price;
+ coder->opts[offset].pos_prev = cur + len_test + 1;
+ coder->opts[offset].back_prev = 0;
+ coder->opts[offset].prev_1_is_literal = true;
+ coder->opts[offset].prev_2 = true;
+ coder->opts[offset].pos_prev_2 = cur;
+ coder->opts[offset].back_prev_2
+ = cur_back + REP_DISTANCES;
+ }
+ //}
+ }
+
+ if (++i == matches_count)
+ break;
+ }
+ }
+ }
+
+ return len_end;
+}
+
+
+extern void
+lzma_lzma_optimum_normal(lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint32_t *restrict back_res, uint32_t *restrict len_res,
+ uint32_t position)
+{
+ // If we have symbols pending, return the next pending symbol.
+ if (coder->opts_end_index != coder->opts_current_index) {
+ assert(mf->read_ahead > 0);
+ *len_res = coder->opts[coder->opts_current_index].pos_prev
+ - coder->opts_current_index;
+ *back_res = coder->opts[coder->opts_current_index].back_prev;
+ coder->opts_current_index = coder->opts[
+ coder->opts_current_index].pos_prev;
+ return;
+ }
+
+ // Update the price tables. In LZMA SDK <= 4.60 (and possibly later)
+ // this was done in both initialization function and in the main loop.
+ // In liblzma they were moved into this single place.
+ if (mf->read_ahead == 0) {
+ if (coder->match_price_count >= (1 << 7))
+ fill_distances_prices(coder);
+
+ if (coder->align_price_count >= ALIGN_TABLE_SIZE)
+ fill_align_prices(coder);
+ }
+
+ // TODO: This needs quite a bit of cleaning still. But splitting
+ // the original function into two pieces makes it at least a little
+ // more readable, since those two parts don't share many variables.
+
+ uint32_t len_end = helper1(coder, mf, back_res, len_res, position);
+ if (len_end == UINT32_MAX)
+ return;
+
+ uint32_t reps[REP_DISTANCES];
+ memcpy(reps, coder->reps, sizeof(reps));
+
+ uint32_t cur;
+ for (cur = 1; cur < len_end; ++cur) {
+ assert(cur < OPTS);
+
+ coder->longest_match_length = mf_find(
+ mf, &coder->matches_count, coder->matches);
+
+ if (coder->longest_match_length >= mf->nice_len)
+ break;
+
+ len_end = helper2(coder, reps, mf_ptr(mf) - 1, len_end,
+ position + cur, cur, mf->nice_len,
+ my_min(mf_avail(mf) + 1, OPTS - 1 - cur));
+ }
+
+ backward(coder, len_res, back_res, cur);
+ return;
+}
diff --git a/src/liblzma/lzma/lzma_encoder_presets.c b/src/liblzma/lzma/lzma_encoder_presets.c
new file mode 100644
index 0000000..c4c9c14
--- /dev/null
+++ b/src/liblzma/lzma/lzma_encoder_presets.c
@@ -0,0 +1,52 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_encoder_presets.c
+/// \brief Encoder presets
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "common.h"
+
+
+extern LZMA_API(lzma_bool)
+lzma_lzma_preset(lzma_options_lzma *options, uint32_t preset)
+{
+ const uint32_t level = preset & LZMA_PRESET_LEVEL_MASK;
+ const uint32_t flags = preset & ~LZMA_PRESET_LEVEL_MASK;
+ const uint32_t supported_flags = LZMA_PRESET_EXTREME;
+
+ if (level > 9 || (flags & ~supported_flags))
+ return true;
+
+ const uint32_t dict_shift = level <= 1 ? 16 : level + 17;
+ options->dict_size = UINT32_C(1) << dict_shift;
+
+ options->preset_dict = NULL;
+ options->preset_dict_size = 0;
+
+ options->lc = LZMA_LC_DEFAULT;
+ options->lp = LZMA_LP_DEFAULT;
+ options->pb = LZMA_PB_DEFAULT;
+
+ options->mode = level <= 2 ? LZMA_MODE_FAST : LZMA_MODE_NORMAL;
+
+ options->nice_len = level == 0 ? 8 : level <= 5 ? 32 : 64;
+ options->mf = level <= 1 ? LZMA_MF_HC3 : level <= 2 ? LZMA_MF_HC4
+ : LZMA_MF_BT4;
+ options->depth = 0;
+
+ if (flags & LZMA_PRESET_EXTREME) {
+ options->lc = 4; // FIXME?
+ options->mode = LZMA_MODE_NORMAL;
+ options->mf = LZMA_MF_BT4;
+ options->nice_len = 273;
+ options->depth = 512;
+ }
+
+ return false;
+}
diff --git a/src/liblzma/lzma/lzma_encoder_private.h b/src/liblzma/lzma/lzma_encoder_private.h
new file mode 100644
index 0000000..6847452
--- /dev/null
+++ b/src/liblzma/lzma/lzma_encoder_private.h
@@ -0,0 +1,148 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzma_encoder_private.h
+/// \brief Private definitions for LZMA encoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_LZMA_ENCODER_PRIVATE_H
+#define LZMA_LZMA_ENCODER_PRIVATE_H
+
+#include "lz_encoder.h"
+#include "range_encoder.h"
+#include "lzma_common.h"
+#include "lzma_encoder.h"
+
+
+// Macro to compare if the first two bytes in two buffers differ. This is
+// needed in lzma_lzma_optimum_*() to test if the match is at least
+// MATCH_LEN_MIN bytes. Unaligned access gives tiny gain so there's no
+// reason to not use it when it is supported.
+#ifdef TUKLIB_FAST_UNALIGNED_ACCESS
+# define not_equal_16(a, b) \
+ (*(const uint16_t *)(a) != *(const uint16_t *)(b))
+#else
+# define not_equal_16(a, b) \
+ ((a)[0] != (b)[0] || (a)[1] != (b)[1])
+#endif
+
+
+// Optimal - Number of entries in the optimum array.
+#define OPTS (1 << 12)
+
+
+typedef struct {
+ probability choice;
+ probability choice2;
+ probability low[POS_STATES_MAX][LEN_LOW_SYMBOLS];
+ probability mid[POS_STATES_MAX][LEN_MID_SYMBOLS];
+ probability high[LEN_HIGH_SYMBOLS];
+
+ uint32_t prices[POS_STATES_MAX][LEN_SYMBOLS];
+ uint32_t table_size;
+ uint32_t counters[POS_STATES_MAX];
+
+} lzma_length_encoder;
+
+
+typedef struct {
+ lzma_lzma_state state;
+
+ bool prev_1_is_literal;
+ bool prev_2;
+
+ uint32_t pos_prev_2;
+ uint32_t back_prev_2;
+
+ uint32_t price;
+ uint32_t pos_prev; // pos_next;
+ uint32_t back_prev;
+
+ uint32_t backs[REP_DISTANCES];
+
+} lzma_optimal;
+
+
+struct lzma_coder_s {
+ /// Range encoder
+ lzma_range_encoder rc;
+
+ /// State
+ lzma_lzma_state state;
+
+ /// The four most recent match distances
+ uint32_t reps[REP_DISTANCES];
+
+ /// Array of match candidates
+ lzma_match matches[MATCH_LEN_MAX + 1];
+
+ /// Number of match candidates in matches[]
+ uint32_t matches_count;
+
+ /// Variable to hold the length of the longest match between calls
+ /// to lzma_lzma_optimum_*().
+ uint32_t longest_match_length;
+
+ /// True if using getoptimumfast
+ bool fast_mode;
+
+ /// True if the encoder has been initialized by encoding the first
+ /// byte as a literal.
+ bool is_initialized;
+
+ /// True if the range encoder has been flushed, but not all bytes
+ /// have been written to the output buffer yet.
+ bool is_flushed;
+
+ uint32_t pos_mask; ///< (1 << pos_bits) - 1
+ uint32_t literal_context_bits;
+ uint32_t literal_pos_mask;
+
+ // These are the same as in lzma_decoder.c. See comments there.
+ probability literal[LITERAL_CODERS_MAX][LITERAL_CODER_SIZE];
+ probability is_match[STATES][POS_STATES_MAX];
+ probability is_rep[STATES];
+ probability is_rep0[STATES];
+ probability is_rep1[STATES];
+ probability is_rep2[STATES];
+ probability is_rep0_long[STATES][POS_STATES_MAX];
+ probability pos_slot[LEN_TO_POS_STATES][POS_SLOTS];
+ probability pos_special[FULL_DISTANCES - END_POS_MODEL_INDEX];
+ probability pos_align[ALIGN_TABLE_SIZE];
+
+ // These are the same as in lzma_decoder.c except that the encoders
+ // include also price tables.
+ lzma_length_encoder match_len_encoder;
+ lzma_length_encoder rep_len_encoder;
+
+ // Price tables
+ uint32_t pos_slot_prices[LEN_TO_POS_STATES][POS_SLOTS];
+ uint32_t distances_prices[LEN_TO_POS_STATES][FULL_DISTANCES];
+ uint32_t dist_table_size;
+ uint32_t match_price_count;
+
+ uint32_t align_prices[ALIGN_TABLE_SIZE];
+ uint32_t align_price_count;
+
+ // Optimal
+ uint32_t opts_end_index;
+ uint32_t opts_current_index;
+ lzma_optimal opts[OPTS];
+};
+
+
+extern void lzma_lzma_optimum_fast(
+ lzma_coder *restrict coder, lzma_mf *restrict mf,
+ uint32_t *restrict back_res, uint32_t *restrict len_res);
+
+extern void lzma_lzma_optimum_normal(lzma_coder *restrict coder,
+ lzma_mf *restrict mf, uint32_t *restrict back_res,
+ uint32_t *restrict len_res, uint32_t position);
+
+#endif
diff --git a/src/liblzma/rangecoder/Makefile.inc b/src/liblzma/rangecoder/Makefile.inc
new file mode 100644
index 0000000..d8a597a
--- /dev/null
+++ b/src/liblzma/rangecoder/Makefile.inc
@@ -0,0 +1,21 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+EXTRA_DIST += rangecoder/price_tablegen.c
+
+liblzma_la_SOURCES += rangecoder/range_common.h
+
+if COND_ENCODER_LZMA1
+liblzma_la_SOURCES += \
+ rangecoder/range_encoder.h \
+ rangecoder/price.h \
+ rangecoder/price_table.c
+endif
+
+if COND_DECODER_LZMA1
+liblzma_la_SOURCES += rangecoder/range_decoder.h
+endif
diff --git a/src/liblzma/rangecoder/price.h b/src/liblzma/rangecoder/price.h
new file mode 100644
index 0000000..8ae02ca
--- /dev/null
+++ b/src/liblzma/rangecoder/price.h
@@ -0,0 +1,92 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file price.h
+/// \brief Probability price calculation
+//
+// Author: Igor Pavlov
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_PRICE_H
+#define LZMA_PRICE_H
+
+
+#define RC_MOVE_REDUCING_BITS 4
+#define RC_BIT_PRICE_SHIFT_BITS 4
+#define RC_PRICE_TABLE_SIZE (RC_BIT_MODEL_TOTAL >> RC_MOVE_REDUCING_BITS)
+
+#define RC_INFINITY_PRICE (UINT32_C(1) << 30)
+
+
+/// Lookup table for the inline functions defined in this file.
+extern const uint8_t lzma_rc_prices[RC_PRICE_TABLE_SIZE];
+
+
+static inline uint32_t
+rc_bit_price(const probability prob, const uint32_t bit)
+{
+ return lzma_rc_prices[(prob ^ ((UINT32_C(0) - bit)
+ & (RC_BIT_MODEL_TOTAL - 1))) >> RC_MOVE_REDUCING_BITS];
+}
+
+
+static inline uint32_t
+rc_bit_0_price(const probability prob)
+{
+ return lzma_rc_prices[prob >> RC_MOVE_REDUCING_BITS];
+}
+
+
+static inline uint32_t
+rc_bit_1_price(const probability prob)
+{
+ return lzma_rc_prices[(prob ^ (RC_BIT_MODEL_TOTAL - 1))
+ >> RC_MOVE_REDUCING_BITS];
+}
+
+
+static inline uint32_t
+rc_bittree_price(const probability *const probs,
+ const uint32_t bit_levels, uint32_t symbol)
+{
+ uint32_t price = 0;
+ symbol += UINT32_C(1) << bit_levels;
+
+ do {
+ const uint32_t bit = symbol & 1;
+ symbol >>= 1;
+ price += rc_bit_price(probs[symbol], bit);
+ } while (symbol != 1);
+
+ return price;
+}
+
+
+static inline uint32_t
+rc_bittree_reverse_price(const probability *const probs,
+ uint32_t bit_levels, uint32_t symbol)
+{
+ uint32_t price = 0;
+ uint32_t model_index = 1;
+
+ do {
+ const uint32_t bit = symbol & 1;
+ symbol >>= 1;
+ price += rc_bit_price(probs[model_index], bit);
+ model_index = (model_index << 1) + bit;
+ } while (--bit_levels != 0);
+
+ return price;
+}
+
+
+static inline uint32_t
+rc_direct_price(const uint32_t bits)
+{
+ return bits << RC_BIT_PRICE_SHIFT_BITS;
+}
+
+#endif
diff --git a/src/liblzma/rangecoder/price_table.c b/src/liblzma/rangecoder/price_table.c
new file mode 100644
index 0000000..ac64bf6
--- /dev/null
+++ b/src/liblzma/rangecoder/price_table.c
@@ -0,0 +1,22 @@
+/* This file has been automatically generated by price_tablegen.c. */
+
+#include "range_encoder.h"
+
+const uint8_t lzma_rc_prices[RC_PRICE_TABLE_SIZE] = {
+ 128, 103, 91, 84, 78, 73, 69, 66,
+ 63, 61, 58, 56, 54, 52, 51, 49,
+ 48, 46, 45, 44, 43, 42, 41, 40,
+ 39, 38, 37, 36, 35, 34, 34, 33,
+ 32, 31, 31, 30, 29, 29, 28, 28,
+ 27, 26, 26, 25, 25, 24, 24, 23,
+ 23, 22, 22, 22, 21, 21, 20, 20,
+ 19, 19, 19, 18, 18, 17, 17, 17,
+ 16, 16, 16, 15, 15, 15, 14, 14,
+ 14, 13, 13, 13, 12, 12, 12, 11,
+ 11, 11, 11, 10, 10, 10, 10, 9,
+ 9, 9, 9, 8, 8, 8, 8, 7,
+ 7, 7, 7, 6, 6, 6, 6, 5,
+ 5, 5, 5, 5, 4, 4, 4, 4,
+ 3, 3, 3, 3, 3, 2, 2, 2,
+ 2, 2, 2, 1, 1, 1, 1, 1
+};
diff --git a/src/liblzma/rangecoder/price_tablegen.c b/src/liblzma/rangecoder/price_tablegen.c
new file mode 100644
index 0000000..bf08ce3
--- /dev/null
+++ b/src/liblzma/rangecoder/price_tablegen.c
@@ -0,0 +1,87 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file price_tablegen.c
+/// \brief Probability price table generator
+///
+/// Compiling: gcc -std=c99 -o price_tablegen price_tablegen.c
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include <inttypes.h>
+#include <stdio.h>
+#include "range_common.h"
+#include "price.h"
+
+
+static uint32_t rc_prices[RC_PRICE_TABLE_SIZE];
+
+
+static void
+init_price_table(void)
+{
+ for (uint32_t i = (UINT32_C(1) << RC_MOVE_REDUCING_BITS) / 2;
+ i < RC_BIT_MODEL_TOTAL;
+ i += (UINT32_C(1) << RC_MOVE_REDUCING_BITS)) {
+ const uint32_t cycles_bits = RC_BIT_PRICE_SHIFT_BITS;
+ uint32_t w = i;
+ uint32_t bit_count = 0;
+
+ for (uint32_t j = 0; j < cycles_bits; ++j) {
+ w *= w;
+ bit_count <<= 1;
+
+ while (w >= (UINT32_C(1) << 16)) {
+ w >>= 1;
+ ++bit_count;
+ }
+ }
+
+ rc_prices[i >> RC_MOVE_REDUCING_BITS]
+ = (RC_BIT_MODEL_TOTAL_BITS << cycles_bits)
+ - 15 - bit_count;
+ }
+
+ return;
+}
+
+
+static void
+print_price_table(void)
+{
+ printf("/* This file has been automatically generated by "
+ "price_tablegen.c. */\n\n"
+ "#include \"range_encoder.h\"\n\n"
+ "const uint8_t lzma_rc_prices["
+ "RC_PRICE_TABLE_SIZE] = {");
+
+ const size_t array_size = sizeof(lzma_rc_prices)
+ / sizeof(lzma_rc_prices[0]);
+ for (size_t i = 0; i < array_size; ++i) {
+ if (i % 8 == 0)
+ printf("\n\t");
+
+ printf("%4" PRIu32, rc_prices[i]);
+
+ if (i != array_size - 1)
+ printf(",");
+ }
+
+ printf("\n};\n");
+
+ return;
+}
+
+
+int
+main(void)
+{
+ init_price_table();
+ print_price_table();
+ return 0;
+}
diff --git a/src/liblzma/rangecoder/range_common.h b/src/liblzma/rangecoder/range_common.h
new file mode 100644
index 0000000..0e64241
--- /dev/null
+++ b/src/liblzma/rangecoder/range_common.h
@@ -0,0 +1,73 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file range_common.h
+/// \brief Common things for range encoder and decoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_RANGE_COMMON_H
+#define LZMA_RANGE_COMMON_H
+
+#ifdef HAVE_CONFIG_H
+# include "common.h"
+#endif
+
+
+///////////////
+// Constants //
+///////////////
+
+#define RC_SHIFT_BITS 8
+#define RC_TOP_BITS 24
+#define RC_TOP_VALUE (UINT32_C(1) << RC_TOP_BITS)
+#define RC_BIT_MODEL_TOTAL_BITS 11
+#define RC_BIT_MODEL_TOTAL (UINT32_C(1) << RC_BIT_MODEL_TOTAL_BITS)
+#define RC_MOVE_BITS 5
+
+
+////////////
+// Macros //
+////////////
+
+// Resets the probability so that both 0 and 1 have probability of 50 %
+#define bit_reset(prob) \
+ prob = RC_BIT_MODEL_TOTAL >> 1
+
+// This does the same for a complete bit tree.
+// (A tree represented as an array.)
+#define bittree_reset(probs, bit_levels) \
+ for (uint32_t bt_i = 0; bt_i < (1 << (bit_levels)); ++bt_i) \
+ bit_reset((probs)[bt_i])
+
+
+//////////////////////
+// Type definitions //
+//////////////////////
+
+/// \brief Type of probabilities used with range coder
+///
+/// This needs to be at least 12-bit integer, so uint16_t is a logical choice.
+/// However, on some architecture and compiler combinations, a bigger type
+/// may give better speed, because the probability variables are accessed
+/// a lot. On the other hand, bigger probability type increases cache
+/// footprint, since there are 2 to 14 thousand probability variables in
+/// LZMA (assuming the limit of lc + lp <= 4; with lc + lp <= 12 there
+/// would be about 1.5 million variables).
+///
+/// With malicious files, the initialization speed of the LZMA decoder can
+/// become important. In that case, smaller probability variables mean that
+/// there is less bytes to write to RAM, which makes initialization faster.
+/// With big probability type, the initialization can become so slow that it
+/// can be a problem e.g. for email servers doing virus scanning.
+///
+/// I will be sticking to uint16_t unless some specific architectures
+/// are *much* faster (20-50 %) with uint32_t.
+typedef uint16_t probability;
+
+#endif
diff --git a/src/liblzma/rangecoder/range_decoder.h b/src/liblzma/rangecoder/range_decoder.h
new file mode 100644
index 0000000..fb96180
--- /dev/null
+++ b/src/liblzma/rangecoder/range_decoder.h
@@ -0,0 +1,179 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file range_decoder.h
+/// \brief Range Decoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_RANGE_DECODER_H
+#define LZMA_RANGE_DECODER_H
+
+#include "range_common.h"
+
+
+typedef struct {
+ uint32_t range;
+ uint32_t code;
+ uint32_t init_bytes_left;
+} lzma_range_decoder;
+
+
+/// Reads the first five bytes to initialize the range decoder.
+static inline bool
+rc_read_init(lzma_range_decoder *rc, const uint8_t *restrict in,
+ size_t *restrict in_pos, size_t in_size)
+{
+ while (rc->init_bytes_left > 0) {
+ if (*in_pos == in_size)
+ return false;
+
+ rc->code = (rc->code << 8) | in[*in_pos];
+ ++*in_pos;
+ --rc->init_bytes_left;
+ }
+
+ return true;
+}
+
+
+/// Makes local copies of range decoder and *in_pos variables. Doing this
+/// improves speed significantly. The range decoder macros expect also
+/// variables `in' and `in_size' to be defined.
+#define rc_to_local(range_decoder, in_pos) \
+ lzma_range_decoder rc = range_decoder; \
+ size_t rc_in_pos = (in_pos); \
+ uint32_t rc_bound
+
+
+/// Stores the local copes back to the range decoder structure.
+#define rc_from_local(range_decoder, in_pos) \
+do { \
+ range_decoder = rc; \
+ in_pos = rc_in_pos; \
+} while (0)
+
+
+/// Resets the range decoder structure.
+#define rc_reset(range_decoder) \
+do { \
+ (range_decoder).range = UINT32_MAX; \
+ (range_decoder).code = 0; \
+ (range_decoder).init_bytes_left = 5; \
+} while (0)
+
+
+/// When decoding has been properly finished, rc.code is always zero unless
+/// the input stream is corrupt. So checking this can catch some corrupt
+/// files especially if they don't have any other integrity check.
+#define rc_is_finished(range_decoder) \
+ ((range_decoder).code == 0)
+
+
+/// Read the next input byte if needed. If more input is needed but there is
+/// no more input available, "goto out" is used to jump out of the main
+/// decoder loop.
+#define rc_normalize(seq) \
+do { \
+ if (rc.range < RC_TOP_VALUE) { \
+ if (unlikely(rc_in_pos == in_size)) { \
+ coder->sequence = seq; \
+ goto out; \
+ } \
+ rc.range <<= RC_SHIFT_BITS; \
+ rc.code = (rc.code << RC_SHIFT_BITS) | in[rc_in_pos++]; \
+ } \
+} while (0)
+
+
+/// Start decoding a bit. This must be used together with rc_update_0()
+/// and rc_update_1():
+///
+/// rc_if_0(prob, seq) {
+/// rc_update_0(prob);
+/// // Do something
+/// } else {
+/// rc_update_1(prob);
+/// // Do something else
+/// }
+///
+#define rc_if_0(prob, seq) \
+ rc_normalize(seq); \
+ rc_bound = (rc.range >> RC_BIT_MODEL_TOTAL_BITS) * (prob); \
+ if (rc.code < rc_bound)
+
+
+/// Update the range decoder state and the used probability variable to
+/// match a decoded bit of 0.
+#define rc_update_0(prob) \
+do { \
+ rc.range = rc_bound; \
+ prob += (RC_BIT_MODEL_TOTAL - (prob)) >> RC_MOVE_BITS; \
+} while (0)
+
+
+/// Update the range decoder state and the used probability variable to
+/// match a decoded bit of 1.
+#define rc_update_1(prob) \
+do { \
+ rc.range -= rc_bound; \
+ rc.code -= rc_bound; \
+ prob -= (prob) >> RC_MOVE_BITS; \
+} while (0)
+
+
+/// Decodes one bit and runs action0 or action1 depending on the decoded bit.
+/// This macro is used as the last step in bittree reverse decoders since
+/// those don't use "symbol" for anything else than indexing the probability
+/// arrays.
+#define rc_bit_last(prob, action0, action1, seq) \
+do { \
+ rc_if_0(prob, seq) { \
+ rc_update_0(prob); \
+ action0; \
+ } else { \
+ rc_update_1(prob); \
+ action1; \
+ } \
+} while (0)
+
+
+/// Decodes one bit, updates "symbol", and runs action0 or action1 depending
+/// on the decoded bit.
+#define rc_bit(prob, action0, action1, seq) \
+ rc_bit_last(prob, \
+ symbol <<= 1; action0, \
+ symbol = (symbol << 1) + 1; action1, \
+ seq);
+
+
+/// Like rc_bit() but add "case seq:" as a prefix. This makes the unrolled
+/// loops more readable because the code isn't littered with "case"
+/// statements. On the other hand this also makes it less readable, since
+/// spotting the places where the decoder loop may be restarted is less
+/// obvious.
+#define rc_bit_case(prob, action0, action1, seq) \
+ case seq: rc_bit(prob, action0, action1, seq)
+
+
+/// Decode a bit without using a probability.
+#define rc_direct(dest, seq) \
+do { \
+ rc_normalize(seq); \
+ rc.range >>= 1; \
+ rc.code -= rc.range; \
+ rc_bound = UINT32_C(0) - (rc.code >> 31); \
+ rc.code += rc.range & rc_bound; \
+ dest = (dest << 1) + (rc_bound + 1); \
+} while (0)
+
+
+// NOTE: No macros are provided for bittree decoding. It seems to be simpler
+// to just write them open in the code.
+
+#endif
diff --git a/src/liblzma/rangecoder/range_encoder.h b/src/liblzma/rangecoder/range_encoder.h
new file mode 100644
index 0000000..1e1c369
--- /dev/null
+++ b/src/liblzma/rangecoder/range_encoder.h
@@ -0,0 +1,231 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file range_encoder.h
+/// \brief Range Encoder
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_RANGE_ENCODER_H
+#define LZMA_RANGE_ENCODER_H
+
+#include "range_common.h"
+#include "price.h"
+
+
+/// Maximum number of symbols that can be put pending into lzma_range_encoder
+/// structure between calls to lzma_rc_encode(). For LZMA, 52+5 is enough
+/// (match with big distance and length followed by range encoder flush).
+#define RC_SYMBOLS_MAX 58
+
+
+typedef struct {
+ uint64_t low;
+ uint64_t cache_size;
+ uint32_t range;
+ uint8_t cache;
+
+ /// Number of symbols in the tables
+ size_t count;
+
+ /// rc_encode()'s position in the tables
+ size_t pos;
+
+ /// Symbols to encode
+ enum {
+ RC_BIT_0,
+ RC_BIT_1,
+ RC_DIRECT_0,
+ RC_DIRECT_1,
+ RC_FLUSH,
+ } symbols[RC_SYMBOLS_MAX];
+
+ /// Probabilities associated with RC_BIT_0 or RC_BIT_1
+ probability *probs[RC_SYMBOLS_MAX];
+
+} lzma_range_encoder;
+
+
+static inline void
+rc_reset(lzma_range_encoder *rc)
+{
+ rc->low = 0;
+ rc->cache_size = 1;
+ rc->range = UINT32_MAX;
+ rc->cache = 0;
+ rc->count = 0;
+ rc->pos = 0;
+}
+
+
+static inline void
+rc_bit(lzma_range_encoder *rc, probability *prob, uint32_t bit)
+{
+ rc->symbols[rc->count] = bit;
+ rc->probs[rc->count] = prob;
+ ++rc->count;
+}
+
+
+static inline void
+rc_bittree(lzma_range_encoder *rc, probability *probs,
+ uint32_t bit_count, uint32_t symbol)
+{
+ uint32_t model_index = 1;
+
+ do {
+ const uint32_t bit = (symbol >> --bit_count) & 1;
+ rc_bit(rc, &probs[model_index], bit);
+ model_index = (model_index << 1) + bit;
+ } while (bit_count != 0);
+}
+
+
+static inline void
+rc_bittree_reverse(lzma_range_encoder *rc, probability *probs,
+ uint32_t bit_count, uint32_t symbol)
+{
+ uint32_t model_index = 1;
+
+ do {
+ const uint32_t bit = symbol & 1;
+ symbol >>= 1;
+ rc_bit(rc, &probs[model_index], bit);
+ model_index = (model_index << 1) + bit;
+ } while (--bit_count != 0);
+}
+
+
+static inline void
+rc_direct(lzma_range_encoder *rc,
+ uint32_t value, uint32_t bit_count)
+{
+ do {
+ rc->symbols[rc->count++]
+ = RC_DIRECT_0 + ((value >> --bit_count) & 1);
+ } while (bit_count != 0);
+}
+
+
+static inline void
+rc_flush(lzma_range_encoder *rc)
+{
+ for (size_t i = 0; i < 5; ++i)
+ rc->symbols[rc->count++] = RC_FLUSH;
+}
+
+
+static inline bool
+rc_shift_low(lzma_range_encoder *rc,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ if ((uint32_t)(rc->low) < (uint32_t)(0xFF000000)
+ || (uint32_t)(rc->low >> 32) != 0) {
+ do {
+ if (*out_pos == out_size)
+ return true;
+
+ out[*out_pos] = rc->cache + (uint8_t)(rc->low >> 32);
+ ++*out_pos;
+ rc->cache = 0xFF;
+
+ } while (--rc->cache_size != 0);
+
+ rc->cache = (rc->low >> 24) & 0xFF;
+ }
+
+ ++rc->cache_size;
+ rc->low = (rc->low & 0x00FFFFFF) << RC_SHIFT_BITS;
+
+ return false;
+}
+
+
+static inline bool
+rc_encode(lzma_range_encoder *rc,
+ uint8_t *out, size_t *out_pos, size_t out_size)
+{
+ assert(rc->count <= RC_SYMBOLS_MAX);
+
+ while (rc->pos < rc->count) {
+ // Normalize
+ if (rc->range < RC_TOP_VALUE) {
+ if (rc_shift_low(rc, out, out_pos, out_size))
+ return true;
+
+ rc->range <<= RC_SHIFT_BITS;
+ }
+
+ // Encode a bit
+ switch (rc->symbols[rc->pos]) {
+ case RC_BIT_0: {
+ probability prob = *rc->probs[rc->pos];
+ rc->range = (rc->range >> RC_BIT_MODEL_TOTAL_BITS)
+ * prob;
+ prob += (RC_BIT_MODEL_TOTAL - prob) >> RC_MOVE_BITS;
+ *rc->probs[rc->pos] = prob;
+ break;
+ }
+
+ case RC_BIT_1: {
+ probability prob = *rc->probs[rc->pos];
+ const uint32_t bound = prob * (rc->range
+ >> RC_BIT_MODEL_TOTAL_BITS);
+ rc->low += bound;
+ rc->range -= bound;
+ prob -= prob >> RC_MOVE_BITS;
+ *rc->probs[rc->pos] = prob;
+ break;
+ }
+
+ case RC_DIRECT_0:
+ rc->range >>= 1;
+ break;
+
+ case RC_DIRECT_1:
+ rc->range >>= 1;
+ rc->low += rc->range;
+ break;
+
+ case RC_FLUSH:
+ // Prevent further normalizations.
+ rc->range = UINT32_MAX;
+
+ // Flush the last five bytes (see rc_flush()).
+ do {
+ if (rc_shift_low(rc, out, out_pos, out_size))
+ return true;
+ } while (++rc->pos < rc->count);
+
+ // Reset the range encoder so we are ready to continue
+ // encoding if we weren't finishing the stream.
+ rc_reset(rc);
+ return false;
+
+ default:
+ assert(0);
+ break;
+ }
+
+ ++rc->pos;
+ }
+
+ rc->count = 0;
+ rc->pos = 0;
+
+ return false;
+}
+
+
+static inline uint64_t
+rc_pending(const lzma_range_encoder *rc)
+{
+ return rc->cache_size + 5 - 1;
+}
+
+#endif
diff --git a/src/liblzma/simple/Makefile.inc b/src/liblzma/simple/Makefile.inc
new file mode 100644
index 0000000..8a5e2d7
--- /dev/null
+++ b/src/liblzma/simple/Makefile.inc
@@ -0,0 +1,47 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+liblzma_la_SOURCES += \
+ simple/simple_coder.c \
+ simple/simple_coder.h \
+ simple/simple_private.h
+
+if COND_ENCODER_SIMPLE
+liblzma_la_SOURCES += \
+ simple/simple_encoder.c \
+ simple/simple_encoder.h
+endif
+
+if COND_DECODER_SIMPLE
+liblzma_la_SOURCES += \
+ simple/simple_decoder.c \
+ simple/simple_decoder.h
+endif
+
+if COND_FILTER_X86
+liblzma_la_SOURCES += simple/x86.c
+endif
+
+if COND_FILTER_POWERPC
+liblzma_la_SOURCES += simple/powerpc.c
+endif
+
+if COND_FILTER_IA64
+liblzma_la_SOURCES += simple/ia64.c
+endif
+
+if COND_FILTER_ARM
+liblzma_la_SOURCES += simple/arm.c
+endif
+
+if COND_FILTER_ARMTHUMB
+liblzma_la_SOURCES += simple/armthumb.c
+endif
+
+if COND_FILTER_SPARC
+liblzma_la_SOURCES += simple/sparc.c
+endif
diff --git a/src/liblzma/simple/arm.c b/src/liblzma/simple/arm.c
new file mode 100644
index 0000000..8fcf643
--- /dev/null
+++ b/src/liblzma/simple/arm.c
@@ -0,0 +1,69 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file arm.c
+/// \brief Filter for ARM binaries
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+static size_t
+arm_code(lzma_simple *simple lzma_attribute((unused)),
+ uint32_t now_pos, bool is_encoder,
+ uint8_t *buffer, size_t size)
+{
+ size_t i;
+ for (i = 0; i + 4 <= size; i += 4) {
+ if (buffer[i + 3] == 0xEB) {
+ uint32_t src = (buffer[i + 2] << 16)
+ | (buffer[i + 1] << 8)
+ | (buffer[i + 0]);
+ src <<= 2;
+
+ uint32_t dest;
+ if (is_encoder)
+ dest = now_pos + (uint32_t)(i) + 8 + src;
+ else
+ dest = src - (now_pos + (uint32_t)(i) + 8);
+
+ dest >>= 2;
+ buffer[i + 2] = (dest >> 16);
+ buffer[i + 1] = (dest >> 8);
+ buffer[i + 0] = dest;
+ }
+ }
+
+ return i;
+}
+
+
+static lzma_ret
+arm_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters, bool is_encoder)
+{
+ return lzma_simple_coder_init(next, allocator, filters,
+ &arm_code, 0, 4, 4, is_encoder);
+}
+
+
+extern lzma_ret
+lzma_simple_arm_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ return arm_coder_init(next, allocator, filters, true);
+}
+
+
+extern lzma_ret
+lzma_simple_arm_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ return arm_coder_init(next, allocator, filters, false);
+}
diff --git a/src/liblzma/simple/armthumb.c b/src/liblzma/simple/armthumb.c
new file mode 100644
index 0000000..eb6a69d
--- /dev/null
+++ b/src/liblzma/simple/armthumb.c
@@ -0,0 +1,74 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file armthumb.c
+/// \brief Filter for ARM-Thumb binaries
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+static size_t
+armthumb_code(lzma_simple *simple lzma_attribute((unused)),
+ uint32_t now_pos, bool is_encoder,
+ uint8_t *buffer, size_t size)
+{
+ size_t i;
+ for (i = 0; i + 4 <= size; i += 2) {
+ if ((buffer[i + 1] & 0xF8) == 0xF0
+ && (buffer[i + 3] & 0xF8) == 0xF8) {
+ uint32_t src = ((buffer[i + 1] & 0x7) << 19)
+ | (buffer[i + 0] << 11)
+ | ((buffer[i + 3] & 0x7) << 8)
+ | (buffer[i + 2]);
+
+ src <<= 1;
+
+ uint32_t dest;
+ if (is_encoder)
+ dest = now_pos + (uint32_t)(i) + 4 + src;
+ else
+ dest = src - (now_pos + (uint32_t)(i) + 4);
+
+ dest >>= 1;
+ buffer[i + 1] = 0xF0 | ((dest >> 19) & 0x7);
+ buffer[i + 0] = (dest >> 11);
+ buffer[i + 3] = 0xF8 | ((dest >> 8) & 0x7);
+ buffer[i + 2] = (dest);
+ i += 2;
+ }
+ }
+
+ return i;
+}
+
+
+static lzma_ret
+armthumb_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters, bool is_encoder)
+{
+ return lzma_simple_coder_init(next, allocator, filters,
+ &armthumb_code, 0, 4, 2, is_encoder);
+}
+
+
+extern lzma_ret
+lzma_simple_armthumb_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return armthumb_coder_init(next, allocator, filters, true);
+}
+
+
+extern lzma_ret
+lzma_simple_armthumb_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return armthumb_coder_init(next, allocator, filters, false);
+}
diff --git a/src/liblzma/simple/ia64.c b/src/liblzma/simple/ia64.c
new file mode 100644
index 0000000..fd263d4
--- /dev/null
+++ b/src/liblzma/simple/ia64.c
@@ -0,0 +1,110 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file ia64.c
+/// \brief Filter for IA64 (Itanium) binaries
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+static size_t
+ia64_code(lzma_simple *simple lzma_attribute((unused)),
+ uint32_t now_pos, bool is_encoder,
+ uint8_t *buffer, size_t size)
+{
+ static const uint32_t BRANCH_TABLE[32] = {
+ 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0,
+ 4, 4, 6, 6, 0, 0, 7, 7,
+ 4, 4, 0, 0, 4, 4, 0, 0
+ };
+
+ size_t i;
+ for (i = 0; i + 16 <= size; i += 16) {
+ const uint32_t instr_template = buffer[i] & 0x1F;
+ const uint32_t mask = BRANCH_TABLE[instr_template];
+ uint32_t bit_pos = 5;
+
+ for (size_t slot = 0; slot < 3; ++slot, bit_pos += 41) {
+ if (((mask >> slot) & 1) == 0)
+ continue;
+
+ const size_t byte_pos = (bit_pos >> 3);
+ const uint32_t bit_res = bit_pos & 0x7;
+ uint64_t instruction = 0;
+
+ for (size_t j = 0; j < 6; ++j)
+ instruction += (uint64_t)(
+ buffer[i + j + byte_pos])
+ << (8 * j);
+
+ uint64_t inst_norm = instruction >> bit_res;
+
+ if (((inst_norm >> 37) & 0xF) == 0x5
+ && ((inst_norm >> 9) & 0x7) == 0
+ /* && (inst_norm & 0x3F)== 0 */
+ ) {
+ uint32_t src = (uint32_t)(
+ (inst_norm >> 13) & 0xFFFFF);
+ src |= ((inst_norm >> 36) & 1) << 20;
+
+ src <<= 4;
+
+ uint32_t dest;
+ if (is_encoder)
+ dest = now_pos + (uint32_t)(i) + src;
+ else
+ dest = src - (now_pos + (uint32_t)(i));
+
+ dest >>= 4;
+
+ inst_norm &= ~((uint64_t)(0x8FFFFF) << 13);
+ inst_norm |= (uint64_t)(dest & 0xFFFFF) << 13;
+ inst_norm |= (uint64_t)(dest & 0x100000)
+ << (36 - 20);
+
+ instruction &= (1 << bit_res) - 1;
+ instruction |= (inst_norm << bit_res);
+
+ for (size_t j = 0; j < 6; j++)
+ buffer[i + j + byte_pos] = (uint8_t)(
+ instruction
+ >> (8 * j));
+ }
+ }
+ }
+
+ return i;
+}
+
+
+static lzma_ret
+ia64_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters, bool is_encoder)
+{
+ return lzma_simple_coder_init(next, allocator, filters,
+ &ia64_code, 0, 16, 16, is_encoder);
+}
+
+
+extern lzma_ret
+lzma_simple_ia64_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return ia64_coder_init(next, allocator, filters, true);
+}
+
+
+extern lzma_ret
+lzma_simple_ia64_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return ia64_coder_init(next, allocator, filters, false);
+}
diff --git a/src/liblzma/simple/powerpc.c b/src/liblzma/simple/powerpc.c
new file mode 100644
index 0000000..aaa14f2
--- /dev/null
+++ b/src/liblzma/simple/powerpc.c
@@ -0,0 +1,73 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file powerpc.c
+/// \brief Filter for PowerPC (big endian) binaries
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+static size_t
+powerpc_code(lzma_simple *simple lzma_attribute((unused)),
+ uint32_t now_pos, bool is_encoder,
+ uint8_t *buffer, size_t size)
+{
+ size_t i;
+ for (i = 0; i + 4 <= size; i += 4) {
+ // PowerPC branch 6(48) 24(Offset) 1(Abs) 1(Link)
+ if ((buffer[i] >> 2) == 0x12
+ && ((buffer[i + 3] & 3) == 1)) {
+
+ const uint32_t src = ((buffer[i + 0] & 3) << 24)
+ | (buffer[i + 1] << 16)
+ | (buffer[i + 2] << 8)
+ | (buffer[i + 3] & (~3));
+
+ uint32_t dest;
+ if (is_encoder)
+ dest = now_pos + (uint32_t)(i) + src;
+ else
+ dest = src - (now_pos + (uint32_t)(i));
+
+ buffer[i + 0] = 0x48 | ((dest >> 24) & 0x03);
+ buffer[i + 1] = (dest >> 16);
+ buffer[i + 2] = (dest >> 8);
+ buffer[i + 3] &= 0x03;
+ buffer[i + 3] |= dest;
+ }
+ }
+
+ return i;
+}
+
+
+static lzma_ret
+powerpc_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters, bool is_encoder)
+{
+ return lzma_simple_coder_init(next, allocator, filters,
+ &powerpc_code, 0, 4, 4, is_encoder);
+}
+
+
+extern lzma_ret
+lzma_simple_powerpc_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return powerpc_coder_init(next, allocator, filters, true);
+}
+
+
+extern lzma_ret
+lzma_simple_powerpc_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return powerpc_coder_init(next, allocator, filters, false);
+}
diff --git a/src/liblzma/simple/simple_coder.c b/src/liblzma/simple/simple_coder.c
new file mode 100644
index 0000000..06db86e
--- /dev/null
+++ b/src/liblzma/simple/simple_coder.c
@@ -0,0 +1,280 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_coder.c
+/// \brief Wrapper for simple filters
+///
+/// Simple filters don't change the size of the data i.e. number of bytes
+/// in equals the number of bytes out.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+/// Copied or encodes/decodes more data to out[].
+static lzma_ret
+copy_or_code(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ assert(!coder->end_was_reached);
+
+ if (coder->next.code == NULL) {
+ lzma_bufcpy(in, in_pos, in_size, out, out_pos, out_size);
+
+ // Check if end of stream was reached.
+ if (coder->is_encoder && action == LZMA_FINISH
+ && *in_pos == in_size)
+ coder->end_was_reached = true;
+
+ } else {
+ // Call the next coder in the chain to provide us some data.
+ // We don't care about uncompressed_size here, because
+ // the next filter in the chain will do it for us (since
+ // we don't change the size of the data).
+ const lzma_ret ret = coder->next.code(
+ coder->next.coder, allocator,
+ in, in_pos, in_size,
+ out, out_pos, out_size, action);
+
+ if (ret == LZMA_STREAM_END) {
+ assert(!coder->is_encoder
+ || action == LZMA_FINISH);
+ coder->end_was_reached = true;
+
+ } else if (ret != LZMA_OK) {
+ return ret;
+ }
+ }
+
+ return LZMA_OK;
+}
+
+
+static size_t
+call_filter(lzma_coder *coder, uint8_t *buffer, size_t size)
+{
+ const size_t filtered = coder->filter(coder->simple,
+ coder->now_pos, coder->is_encoder,
+ buffer, size);
+ coder->now_pos += filtered;
+ return filtered;
+}
+
+
+static lzma_ret
+simple_code(lzma_coder *coder, lzma_allocator *allocator,
+ const uint8_t *restrict in, size_t *restrict in_pos,
+ size_t in_size, uint8_t *restrict out,
+ size_t *restrict out_pos, size_t out_size, lzma_action action)
+{
+ // TODO: Add partial support for LZMA_SYNC_FLUSH. We can support it
+ // in cases when the filter is able to filter everything. With most
+ // simple filters it can be done at offset that is a multiple of 2,
+ // 4, or 16. With x86 filter, it needs good luck, and thus cannot
+ // be made to work predictably.
+ if (action == LZMA_SYNC_FLUSH)
+ return LZMA_OPTIONS_ERROR;
+
+ // Flush already filtered data from coder->buffer[] to out[].
+ if (coder->pos < coder->filtered) {
+ lzma_bufcpy(coder->buffer, &coder->pos, coder->filtered,
+ out, out_pos, out_size);
+
+ // If we couldn't flush all the filtered data, return to
+ // application immediately.
+ if (coder->pos < coder->filtered)
+ return LZMA_OK;
+
+ if (coder->end_was_reached) {
+ assert(coder->filtered == coder->size);
+ return LZMA_STREAM_END;
+ }
+ }
+
+ // If we get here, there is no filtered data left in the buffer.
+ coder->filtered = 0;
+
+ assert(!coder->end_was_reached);
+
+ // If there is more output space left than there is unfiltered data
+ // in coder->buffer[], flush coder->buffer[] to out[], and copy/code
+ // more data to out[] hopefully filling it completely. Then filter
+ // the data in out[]. This step is where most of the data gets
+ // filtered if the buffer sizes used by the application are reasonable.
+ const size_t out_avail = out_size - *out_pos;
+ const size_t buf_avail = coder->size - coder->pos;
+ if (out_avail > buf_avail) {
+ // Store the old position so that we know from which byte
+ // to start filtering.
+ const size_t out_start = *out_pos;
+
+ // Flush data from coder->buffer[] to out[], but don't reset
+ // coder->pos and coder->size yet. This way the coder can be
+ // restarted if the next filter in the chain returns e.g.
+ // LZMA_MEM_ERROR.
+ memcpy(out + *out_pos, coder->buffer + coder->pos, buf_avail);
+ *out_pos += buf_avail;
+
+ // Copy/Encode/Decode more data to out[].
+ {
+ const lzma_ret ret = copy_or_code(coder, allocator,
+ in, in_pos, in_size,
+ out, out_pos, out_size, action);
+ assert(ret != LZMA_STREAM_END);
+ if (ret != LZMA_OK)
+ return ret;
+ }
+
+ // Filter out[].
+ const size_t size = *out_pos - out_start;
+ const size_t filtered = call_filter(
+ coder, out + out_start, size);
+
+ const size_t unfiltered = size - filtered;
+ assert(unfiltered <= coder->allocated / 2);
+
+ // Now we can update coder->pos and coder->size, because
+ // the next coder in the chain (if any) was successful.
+ coder->pos = 0;
+ coder->size = unfiltered;
+
+ if (coder->end_was_reached) {
+ // The last byte has been copied to out[] already.
+ // They are left as is.
+ coder->size = 0;
+
+ } else if (unfiltered > 0) {
+ // There is unfiltered data left in out[]. Copy it to
+ // coder->buffer[] and rewind *out_pos appropriately.
+ *out_pos -= unfiltered;
+ memcpy(coder->buffer, out + *out_pos, unfiltered);
+ }
+ } else if (coder->pos > 0) {
+ memmove(coder->buffer, coder->buffer + coder->pos, buf_avail);
+ coder->size -= coder->pos;
+ coder->pos = 0;
+ }
+
+ assert(coder->pos == 0);
+
+ // If coder->buffer[] isn't empty, try to fill it by copying/decoding
+ // more data. Then filter coder->buffer[] and copy the successfully
+ // filtered data to out[]. It is probable, that some filtered and
+ // unfiltered data will be left to coder->buffer[].
+ if (coder->size > 0) {
+ {
+ const lzma_ret ret = copy_or_code(coder, allocator,
+ in, in_pos, in_size,
+ coder->buffer, &coder->size,
+ coder->allocated, action);
+ assert(ret != LZMA_STREAM_END);
+ if (ret != LZMA_OK)
+ return ret;
+ }
+
+ coder->filtered = call_filter(
+ coder, coder->buffer, coder->size);
+
+ // Everything is considered to be filtered if coder->buffer[]
+ // contains the last bytes of the data.
+ if (coder->end_was_reached)
+ coder->filtered = coder->size;
+
+ // Flush as much as possible.
+ lzma_bufcpy(coder->buffer, &coder->pos, coder->filtered,
+ out, out_pos, out_size);
+ }
+
+ // Check if we got everything done.
+ if (coder->end_was_reached && coder->pos == coder->size)
+ return LZMA_STREAM_END;
+
+ return LZMA_OK;
+}
+
+
+static void
+simple_coder_end(lzma_coder *coder, lzma_allocator *allocator)
+{
+ lzma_next_end(&coder->next, allocator);
+ lzma_free(coder->simple, allocator);
+ lzma_free(coder, allocator);
+ return;
+}
+
+
+static lzma_ret
+simple_coder_update(lzma_coder *coder, lzma_allocator *allocator,
+ const lzma_filter *filters_null lzma_attribute((unused)),
+ const lzma_filter *reversed_filters)
+{
+ // No update support, just call the next filter in the chain.
+ return lzma_next_filter_update(
+ &coder->next, allocator, reversed_filters + 1);
+}
+
+
+extern lzma_ret
+lzma_simple_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters,
+ size_t (*filter)(lzma_simple *simple, uint32_t now_pos,
+ bool is_encoder, uint8_t *buffer, size_t size),
+ size_t simple_size, size_t unfiltered_max,
+ uint32_t alignment, bool is_encoder)
+{
+ // Allocate memory for the lzma_coder structure if needed.
+ if (next->coder == NULL) {
+ // Here we allocate space also for the temporary buffer. We
+ // need twice the size of unfiltered_max, because then it
+ // is always possible to filter at least unfiltered_max bytes
+ // more data in coder->buffer[] if it can be filled completely.
+ next->coder = lzma_alloc(sizeof(lzma_coder)
+ + 2 * unfiltered_max, allocator);
+ if (next->coder == NULL)
+ return LZMA_MEM_ERROR;
+
+ next->code = &simple_code;
+ next->end = &simple_coder_end;
+ next->update = &simple_coder_update;
+
+ next->coder->next = LZMA_NEXT_CODER_INIT;
+ next->coder->filter = filter;
+ next->coder->allocated = 2 * unfiltered_max;
+
+ // Allocate memory for filter-specific data structure.
+ if (simple_size > 0) {
+ next->coder->simple = lzma_alloc(
+ simple_size, allocator);
+ if (next->coder->simple == NULL)
+ return LZMA_MEM_ERROR;
+ } else {
+ next->coder->simple = NULL;
+ }
+ }
+
+ if (filters[0].options != NULL) {
+ const lzma_options_bcj *simple = filters[0].options;
+ next->coder->now_pos = simple->start_offset;
+ if (next->coder->now_pos & (alignment - 1))
+ return LZMA_OPTIONS_ERROR;
+ } else {
+ next->coder->now_pos = 0;
+ }
+
+ // Reset variables.
+ next->coder->is_encoder = is_encoder;
+ next->coder->end_was_reached = false;
+ next->coder->pos = 0;
+ next->coder->filtered = 0;
+ next->coder->size = 0;
+
+ return lzma_next_filter_init(
+ &next->coder->next, allocator, filters + 1);
+}
diff --git a/src/liblzma/simple/simple_coder.h b/src/liblzma/simple/simple_coder.h
new file mode 100644
index 0000000..0952fad
--- /dev/null
+++ b/src/liblzma/simple/simple_coder.h
@@ -0,0 +1,60 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_coder.h
+/// \brief Wrapper for simple filters
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_SIMPLE_CODER_H
+#define LZMA_SIMPLE_CODER_H
+
+#include "common.h"
+
+
+extern lzma_ret lzma_simple_x86_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_simple_x86_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+
+extern lzma_ret lzma_simple_powerpc_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_simple_powerpc_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+
+extern lzma_ret lzma_simple_ia64_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_simple_ia64_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+
+extern lzma_ret lzma_simple_arm_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_simple_arm_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+
+extern lzma_ret lzma_simple_armthumb_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_simple_armthumb_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+
+extern lzma_ret lzma_simple_sparc_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+extern lzma_ret lzma_simple_sparc_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters);
+
+#endif
diff --git a/src/liblzma/simple/simple_decoder.c b/src/liblzma/simple/simple_decoder.c
new file mode 100644
index 0000000..0beccd3
--- /dev/null
+++ b/src/liblzma/simple/simple_decoder.c
@@ -0,0 +1,40 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_decoder.c
+/// \brief Properties decoder for simple filters
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_decoder.h"
+
+
+extern lzma_ret
+lzma_simple_props_decode(void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size)
+{
+ if (props_size == 0)
+ return LZMA_OK;
+
+ if (props_size != 4)
+ return LZMA_OPTIONS_ERROR;
+
+ lzma_options_bcj *opt = lzma_alloc(
+ sizeof(lzma_options_bcj), allocator);
+ if (opt == NULL)
+ return LZMA_MEM_ERROR;
+
+ opt->start_offset = unaligned_read32le(props);
+
+ // Don't leave an options structure allocated if start_offset is zero.
+ if (opt->start_offset == 0)
+ lzma_free(opt, allocator);
+ else
+ *options = opt;
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/simple/simple_decoder.h b/src/liblzma/simple/simple_decoder.h
new file mode 100644
index 0000000..b8bf590
--- /dev/null
+++ b/src/liblzma/simple/simple_decoder.h
@@ -0,0 +1,22 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_decoder.h
+/// \brief Properties decoder for simple filters
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_SIMPLE_DECODER_H
+#define LZMA_SIMPLE_DECODER_H
+
+#include "simple_coder.h"
+
+extern lzma_ret lzma_simple_props_decode(
+ void **options, lzma_allocator *allocator,
+ const uint8_t *props, size_t props_size);
+
+#endif
diff --git a/src/liblzma/simple/simple_encoder.c b/src/liblzma/simple/simple_encoder.c
new file mode 100644
index 0000000..8aa463b
--- /dev/null
+++ b/src/liblzma/simple/simple_encoder.c
@@ -0,0 +1,38 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_encoder.c
+/// \brief Properties encoder for simple filters
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_encoder.h"
+
+
+extern lzma_ret
+lzma_simple_props_size(uint32_t *size, const void *options)
+{
+ const lzma_options_bcj *const opt = options;
+ *size = (opt == NULL || opt->start_offset == 0) ? 0 : 4;
+ return LZMA_OK;
+}
+
+
+extern lzma_ret
+lzma_simple_props_encode(const void *options, uint8_t *out)
+{
+ const lzma_options_bcj *const opt = options;
+
+ // The default start offset is zero, so we don't need to store any
+ // options unless the start offset is non-zero.
+ if (opt == NULL || opt->start_offset == 0)
+ return LZMA_OK;
+
+ unaligned_write32le(out, opt->start_offset);
+
+ return LZMA_OK;
+}
diff --git a/src/liblzma/simple/simple_encoder.h b/src/liblzma/simple/simple_encoder.h
new file mode 100644
index 0000000..1cee482
--- /dev/null
+++ b/src/liblzma/simple/simple_encoder.h
@@ -0,0 +1,23 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_encoder.c
+/// \brief Properties encoder for simple filters
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_SIMPLE_ENCODER_H
+#define LZMA_SIMPLE_ENCODER_H
+
+#include "simple_coder.h"
+
+
+extern lzma_ret lzma_simple_props_size(uint32_t *size, const void *options);
+
+extern lzma_ret lzma_simple_props_encode(const void *options, uint8_t *out);
+
+#endif
diff --git a/src/liblzma/simple/simple_private.h b/src/liblzma/simple/simple_private.h
new file mode 100644
index 0000000..a69f827
--- /dev/null
+++ b/src/liblzma/simple/simple_private.h
@@ -0,0 +1,76 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file simple_private.h
+/// \brief Private definitions for so called simple filters
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_SIMPLE_PRIVATE_H
+#define LZMA_SIMPLE_PRIVATE_H
+
+#include "simple_coder.h"
+
+
+typedef struct lzma_simple_s lzma_simple;
+
+struct lzma_coder_s {
+ /// Next filter in the chain
+ lzma_next_coder next;
+
+ /// True if the next coder in the chain has returned LZMA_STREAM_END
+ /// or if we have processed uncompressed_size bytes.
+ bool end_was_reached;
+
+ /// True if filter() should encode the data; false to decode.
+ /// Currently all simple filters use the same function for encoding
+ /// and decoding, because the difference between encoders and decoders
+ /// is very small.
+ bool is_encoder;
+
+ /// Pointer to filter-specific function, which does
+ /// the actual filtering.
+ size_t (*filter)(lzma_simple *simple, uint32_t now_pos,
+ bool is_encoder, uint8_t *buffer, size_t size);
+
+ /// Pointer to filter-specific data, or NULL if filter doesn't need
+ /// any extra data.
+ lzma_simple *simple;
+
+ /// The lowest 32 bits of the current position in the data. Most
+ /// filters need this to do conversions between absolute and relative
+ /// addresses.
+ uint32_t now_pos;
+
+ /// Size of the memory allocated for the buffer.
+ size_t allocated;
+
+ /// Flushing position in the temporary buffer. buffer[pos] is the
+ /// next byte to be copied to out[].
+ size_t pos;
+
+ /// buffer[filtered] is the first unfiltered byte. When pos is smaller
+ /// than filtered, there is unflushed filtered data in the buffer.
+ size_t filtered;
+
+ /// Total number of bytes (both filtered and unfiltered) currently
+ /// in the temporary buffer.
+ size_t size;
+
+ /// Temporary buffer
+ uint8_t buffer[];
+};
+
+
+extern lzma_ret lzma_simple_coder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters,
+ size_t (*filter)(lzma_simple *simple, uint32_t now_pos,
+ bool is_encoder, uint8_t *buffer, size_t size),
+ size_t simple_size, size_t unfiltered_max,
+ uint32_t alignment, bool is_encoder);
+
+#endif
diff --git a/src/liblzma/simple/sparc.c b/src/liblzma/simple/sparc.c
new file mode 100644
index 0000000..808a59a
--- /dev/null
+++ b/src/liblzma/simple/sparc.c
@@ -0,0 +1,81 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file sparc.c
+/// \brief Filter for SPARC binaries
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+static size_t
+sparc_code(lzma_simple *simple lzma_attribute((unused)),
+ uint32_t now_pos, bool is_encoder,
+ uint8_t *buffer, size_t size)
+{
+ size_t i;
+ for (i = 0; i + 4 <= size; i += 4) {
+
+ if ((buffer[i] == 0x40 && (buffer[i + 1] & 0xC0) == 0x00)
+ || (buffer[i] == 0x7F
+ && (buffer[i + 1] & 0xC0) == 0xC0)) {
+
+ uint32_t src = ((uint32_t)buffer[i + 0] << 24)
+ | ((uint32_t)buffer[i + 1] << 16)
+ | ((uint32_t)buffer[i + 2] << 8)
+ | ((uint32_t)buffer[i + 3]);
+
+ src <<= 2;
+
+ uint32_t dest;
+ if (is_encoder)
+ dest = now_pos + (uint32_t)(i) + src;
+ else
+ dest = src - (now_pos + (uint32_t)(i));
+
+ dest >>= 2;
+
+ dest = (((0 - ((dest >> 22) & 1)) << 22) & 0x3FFFFFFF)
+ | (dest & 0x3FFFFF)
+ | 0x40000000;
+
+ buffer[i + 0] = (uint8_t)(dest >> 24);
+ buffer[i + 1] = (uint8_t)(dest >> 16);
+ buffer[i + 2] = (uint8_t)(dest >> 8);
+ buffer[i + 3] = (uint8_t)(dest);
+ }
+ }
+
+ return i;
+}
+
+
+static lzma_ret
+sparc_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters, bool is_encoder)
+{
+ return lzma_simple_coder_init(next, allocator, filters,
+ &sparc_code, 0, 4, 4, is_encoder);
+}
+
+
+extern lzma_ret
+lzma_simple_sparc_encoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return sparc_coder_init(next, allocator, filters, true);
+}
+
+
+extern lzma_ret
+lzma_simple_sparc_decoder_init(lzma_next_coder *next,
+ lzma_allocator *allocator, const lzma_filter_info *filters)
+{
+ return sparc_coder_init(next, allocator, filters, false);
+}
diff --git a/src/liblzma/simple/x86.c b/src/liblzma/simple/x86.c
new file mode 100644
index 0000000..5d1509b
--- /dev/null
+++ b/src/liblzma/simple/x86.c
@@ -0,0 +1,154 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file x86.c
+/// \brief Filter for x86 binaries (BCJ filter)
+///
+// Authors: Igor Pavlov
+// Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "simple_private.h"
+
+
+#define Test86MSByte(b) ((b) == 0 || (b) == 0xFF)
+
+
+struct lzma_simple_s {
+ uint32_t prev_mask;
+ uint32_t prev_pos;
+};
+
+
+static size_t
+x86_code(lzma_simple *simple, uint32_t now_pos, bool is_encoder,
+ uint8_t *buffer, size_t size)
+{
+ static const bool MASK_TO_ALLOWED_STATUS[8]
+ = { true, true, true, false, true, false, false, false };
+
+ static const uint32_t MASK_TO_BIT_NUMBER[8]
+ = { 0, 1, 2, 2, 3, 3, 3, 3 };
+
+ uint32_t prev_mask = simple->prev_mask;
+ uint32_t prev_pos = simple->prev_pos;
+
+ if (size < 5)
+ return 0;
+
+ if (now_pos - prev_pos > 5)
+ prev_pos = now_pos - 5;
+
+ const size_t limit = size - 5;
+ size_t buffer_pos = 0;
+
+ while (buffer_pos <= limit) {
+ uint8_t b = buffer[buffer_pos];
+ if (b != 0xE8 && b != 0xE9) {
+ ++buffer_pos;
+ continue;
+ }
+
+ const uint32_t offset = now_pos + (uint32_t)(buffer_pos)
+ - prev_pos;
+ prev_pos = now_pos + (uint32_t)(buffer_pos);
+
+ if (offset > 5) {
+ prev_mask = 0;
+ } else {
+ for (uint32_t i = 0; i < offset; ++i) {
+ prev_mask &= 0x77;
+ prev_mask <<= 1;
+ }
+ }
+
+ b = buffer[buffer_pos + 4];
+
+ if (Test86MSByte(b)
+ && MASK_TO_ALLOWED_STATUS[(prev_mask >> 1) & 0x7]
+ && (prev_mask >> 1) < 0x10) {
+
+ uint32_t src = ((uint32_t)(b) << 24)
+ | ((uint32_t)(buffer[buffer_pos + 3]) << 16)
+ | ((uint32_t)(buffer[buffer_pos + 2]) << 8)
+ | (buffer[buffer_pos + 1]);
+
+ uint32_t dest;
+ while (true) {
+ if (is_encoder)
+ dest = src + (now_pos + (uint32_t)(
+ buffer_pos) + 5);
+ else
+ dest = src - (now_pos + (uint32_t)(
+ buffer_pos) + 5);
+
+ if (prev_mask == 0)
+ break;
+
+ const uint32_t i = MASK_TO_BIT_NUMBER[
+ prev_mask >> 1];
+
+ b = (uint8_t)(dest >> (24 - i * 8));
+
+ if (!Test86MSByte(b))
+ break;
+
+ src = dest ^ ((1 << (32 - i * 8)) - 1);
+ }
+
+ buffer[buffer_pos + 4]
+ = (uint8_t)(~(((dest >> 24) & 1) - 1));
+ buffer[buffer_pos + 3] = (uint8_t)(dest >> 16);
+ buffer[buffer_pos + 2] = (uint8_t)(dest >> 8);
+ buffer[buffer_pos + 1] = (uint8_t)(dest);
+ buffer_pos += 5;
+ prev_mask = 0;
+
+ } else {
+ ++buffer_pos;
+ prev_mask |= 1;
+ if (Test86MSByte(b))
+ prev_mask |= 0x10;
+ }
+ }
+
+ simple->prev_mask = prev_mask;
+ simple->prev_pos = prev_pos;
+
+ return buffer_pos;
+}
+
+
+static lzma_ret
+x86_coder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters, bool is_encoder)
+{
+ const lzma_ret ret = lzma_simple_coder_init(next, allocator, filters,
+ &x86_code, sizeof(lzma_simple), 5, 1, is_encoder);
+
+ if (ret == LZMA_OK) {
+ next->coder->simple->prev_mask = 0;
+ next->coder->simple->prev_pos = (uint32_t)(-5);
+ }
+
+ return ret;
+}
+
+
+extern lzma_ret
+lzma_simple_x86_encoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ return x86_coder_init(next, allocator, filters, true);
+}
+
+
+extern lzma_ret
+lzma_simple_x86_decoder_init(lzma_next_coder *next, lzma_allocator *allocator,
+ const lzma_filter_info *filters)
+{
+ return x86_coder_init(next, allocator, filters, false);
+}
diff --git a/src/lzmainfo/Makefile.am b/src/lzmainfo/Makefile.am
new file mode 100644
index 0000000..cbc7e5c
--- /dev/null
+++ b/src/lzmainfo/Makefile.am
@@ -0,0 +1,32 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+bin_PROGRAMS = lzmainfo
+
+lzmainfo_SOURCES = \
+ lzmainfo.c \
+ $(top_srcdir)/src/common/tuklib_progname.c \
+ $(top_srcdir)/src/common/tuklib_exit.c
+
+lzmainfo_CPPFLAGS = \
+ -DLOCALEDIR=\"$(localedir)\" \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_builddir)/lib \
+ $(DYNAMIC_CPPFLAGS)
+
+lzmainfo_LDFLAGS = $(DYNAMIC_LDFLAGS)
+lzmainfo_LDADD = $(top_builddir)/src/liblzma/liblzma.la
+
+if COND_GNULIB
+lzmainfo_LDADD += $(top_builddir)/lib/libgnu.a
+endif
+
+lzmainfo_LDADD += $(LTLIBINTL)
+
+
+dist_man_MANS = lzmainfo.1
diff --git a/src/lzmainfo/lzmainfo.1 b/src/lzmainfo/lzmainfo.1
new file mode 100644
index 0000000..ef736a6
--- /dev/null
+++ b/src/lzmainfo/lzmainfo.1
@@ -0,0 +1,55 @@
+.\"
+.\" Author: Lasse Collin
+.\"
+.\" This file has been put into the public domain.
+.\" You can do whatever you want with this file.
+.\"
+.TH LZMAINFO 1 "2009-08-13" "Tukaani" "XZ Utils"
+.SH NAME
+lzmainfo \- show infomation stored in the .lzma file header
+.SH SYNOPSIS
+.B lzmainfo
+.RB [ \-\-help ]
+.RB [ \-\-version ]
+.RI [ file ]...
+.SH DESCRIPTION
+.B lzmainfo
+shows information stored in the
+.B .lzma
+file header. It reads the first 13 bytes from the specified
+.IR file ,
+decodes the header, and prints it to standard output in human
+readable format. If no
+.I files
+are given or
+.I file
+is
+.BR \- ,
+standard input is read.
+.PP
+Usually the most interesting information is the uncompressed size and
+the dictionary size. Uncompressed size can be shown only if the file is
+in the non-streamed
+.B .lzma
+format variant. The amount of memory required to decompress the file is
+a few dozen kilobytes plus the dictionary size.
+.PP
+.B lzmainfo
+is included in XZ Utils primarily for backward compatibility with LZMA Utils.
+.SH EXIT STATUS
+.TP
+.B 0
+All is good.
+.TP
+.B 1
+An error occurred.
+.SH BUGS
+.B lzmainfo
+uses
+.B MB
+while the correct suffix would be
+.B MiB
+(2^20 bytes).
+This is to keep the output compatible with LZMA Utils.
+.SH SEE ALSO
+.BR xz (1)
diff --git a/src/lzmainfo/lzmainfo.c b/src/lzmainfo/lzmainfo.c
new file mode 100644
index 0000000..af8e66c
--- /dev/null
+++ b/src/lzmainfo/lzmainfo.c
@@ -0,0 +1,210 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file lzmainfo.c
+/// \brief lzmainfo tool for compatibility with LZMA Utils
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include <stdio.h>
+#include <errno.h>
+
+#include "lzma.h"
+#include "getopt.h"
+#include "tuklib_gettext.h"
+#include "tuklib_progname.h"
+#include "tuklib_exit.h"
+
+
+static void lzma_attribute((noreturn))
+help(void)
+{
+ printf(
+_("Usage: %s [--help] [--version] [FILE]...\n"
+"Show information stored in the .lzma file header"), progname);
+
+ printf(_(
+"\nWith no FILE, or when FILE is -, read standard input.\n"));
+ printf("\n");
+
+ printf(_("Report bugs to <%s> (in English or Finnish).\n"),
+ PACKAGE_BUGREPORT);
+ printf(_("%s home page: <%s>\n"), PACKAGE_NAME, PACKAGE_URL);
+
+ tuklib_exit(EXIT_SUCCESS, EXIT_FAILURE, true);
+}
+
+
+static void lzma_attribute((noreturn))
+version(void)
+{
+ puts("lzmainfo (" PACKAGE_NAME ") " PACKAGE_VERSION);
+ tuklib_exit(EXIT_SUCCESS, EXIT_FAILURE, true);
+}
+
+
+/// Parse command line options.
+static void
+parse_args(int argc, char **argv)
+{
+ enum {
+ OPT_HELP,
+ OPT_VERSION,
+ };
+
+ static const struct option long_opts[] = {
+ { "help", no_argument, NULL, OPT_HELP },
+ { "version", no_argument, NULL, OPT_VERSION },
+ { NULL, 0, NULL, 0 }
+ };
+
+ int c;
+ while ((c = getopt_long(argc, argv, "", long_opts, NULL)) != -1) {
+ switch (c) {
+ case OPT_HELP:
+ help();
+
+ case OPT_VERSION:
+ version();
+
+ default:
+ exit(EXIT_FAILURE);
+ }
+ }
+
+ return;
+}
+
+
+/// Primitive base-2 logarithm for integers
+static uint32_t
+my_log2(uint32_t n)
+{
+ uint32_t e;
+ for (e = 0; n > 1; ++e, n /= 2) ;
+ return e;
+}
+
+
+/// Parse the .lzma header and display information about it.
+static bool
+lzmainfo(const char *name, FILE *f)
+{
+ uint8_t buf[13];
+ const size_t size = fread(buf, 1, sizeof(buf), f);
+ if (size != 13) {
+ fprintf(stderr, "%s: %s: %s\n", progname, name,
+ ferror(f) ? strerror(errno)
+ : _("File is too small to be a .lzma file"));
+ return true;
+ }
+
+ lzma_filter filter = { .id = LZMA_FILTER_LZMA1 };
+
+ // Parse the first five bytes.
+ switch (lzma_properties_decode(&filter, NULL, buf, 5)) {
+ case LZMA_OK:
+ break;
+
+ case LZMA_OPTIONS_ERROR:
+ fprintf(stderr, "%s: %s: %s\n", progname, name,
+ _("Not a .lzma file"));
+ return true;
+
+ case LZMA_MEM_ERROR:
+ fprintf(stderr, "%s: %s\n", progname, strerror(ENOMEM));
+ exit(EXIT_FAILURE);
+
+ default:
+ fprintf(stderr, "%s: %s\n", progname,
+ _("Internal error (bug)"));
+ exit(EXIT_FAILURE);
+ }
+
+ // Uncompressed size
+ uint64_t uncompressed_size = 0;
+ for (size_t i = 0; i < 8; ++i)
+ uncompressed_size |= (uint64_t)(buf[5 + i]) << (i * 8);
+
+ // Display the results. We don't want to translate these and also
+ // will use MB instead of MiB, because someone could be parsing
+ // this output and we don't want to break that when people move
+ // from LZMA Utils to XZ Utils.
+ if (f != stdin)
+ printf("%s\n", name);
+
+ printf("Uncompressed size: ");
+ if (uncompressed_size == UINT64_MAX)
+ printf("Unknown");
+ else
+ printf("%" PRIu64 " MB (%" PRIu64 " bytes)",
+ (uncompressed_size + 512 * 1024)
+ / (1024 * 1024),
+ uncompressed_size);
+
+ lzma_options_lzma *opt = filter.options;
+
+ printf("\nDictionary size: "
+ "%u MB (2^%u bytes)\n"
+ "Literal context bits (lc): %" PRIu32 "\n"
+ "Literal pos bits (lp): %" PRIu32 "\n"
+ "Number of pos bits (pb): %" PRIu32 "\n",
+ (opt->dict_size + 512 * 1024) / (1024 * 1024),
+ my_log2(opt->dict_size), opt->lc, opt->lp, opt->pb);
+
+ free(opt);
+
+ return false;
+}
+
+
+extern int
+main(int argc, char **argv)
+{
+ tuklib_progname_init(argv);
+ tuklib_gettext_init(PACKAGE, LOCALEDIR);
+
+ parse_args(argc, argv);
+
+ int ret = EXIT_SUCCESS;
+
+ // We print empty lines around the output only when reading from
+ // files specified on the command line. This is due to how
+ // LZMA Utils did it.
+ if (optind == argc) {
+ if (lzmainfo("(stdin)", stdin))
+ ret = EXIT_FAILURE;
+ } else {
+ printf("\n");
+
+ do {
+ if (strcmp(argv[optind], "-") == 0) {
+ if (lzmainfo("(stdin)", stdin))
+ ret = EXIT_FAILURE;
+ } else {
+ FILE *f = fopen(argv[optind], "r");
+ if (f == NULL) {
+ ret = EXIT_FAILURE;
+ fprintf(stderr, "%s: %s: %s\n",
+ progname,
+ argv[optind],
+ strerror(errno));
+ continue;
+ }
+
+ if (lzmainfo(argv[optind], f))
+ ret = EXIT_FAILURE;
+
+ printf("\n");
+ fclose(f);
+ }
+ } while (++optind < argc);
+ }
+
+ tuklib_exit(ret, EXIT_FAILURE, true);
+}
diff --git a/src/scripts/Makefile.am b/src/scripts/Makefile.am
new file mode 100644
index 0000000..86f2738
--- /dev/null
+++ b/src/scripts/Makefile.am
@@ -0,0 +1,65 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+nodist_bin_SCRIPTS = xzdiff xzgrep xzmore xzless
+dist_man_MANS = xzdiff.1 xzgrep.1 xzmore.1 xzless.1
+
+install-exec-hook:
+ cd $(DESTDIR)$(bindir) && \
+ target=`echo xzdiff | sed '$(transform)'` && \
+ for name in xzcmp lzdiff lzcmp; do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link && \
+ $(LN_S) $$target $$link; \
+ done && \
+ target=`echo xzgrep | sed '$(transform)'` && \
+ for name in xzegrep xzfgrep lzgrep lzegrep lzfgrep; do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link && \
+ $(LN_S) $$target $$link; \
+ done && \
+ for name in xzmore xzless; do \
+ target=`echo $$name | sed '$(transform)'` && \
+ link=`echo $$name | sed 's/xz/lz/;$(transform)'` && \
+ rm -f $$link && \
+ $(LN_S) $$target $$link; \
+ done
+
+install-data-hook:
+ cd $(DESTDIR)$(mandir)/man1 && \
+ target=`echo xzdiff | sed '$(transform)'` && \
+ for name in xzcmp lzdiff lzcmp; do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link.1 && \
+ $(LN_S) $$target.1 $$link.1; \
+ done && \
+ target=`echo xzgrep | sed '$(transform)'` && \
+ for name in xzegrep xzfgrep lzgrep lzegrep lzfgrep; do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link.1 && \
+ $(LN_S) $$target.1 $$link.1; \
+ done && \
+ for name in xzmore xzless; do \
+ target=`echo $$name | sed '$(transform)'` && \
+ link=`echo $$name | sed 's/xz/lz/;$(transform)'` && \
+ rm -f $$link.1 && \
+ $(LN_S) $$target.1 $$link.1; \
+ done
+
+uninstall-hook:
+ cd $(DESTDIR)$(bindir) && \
+ for name in xzcmp lzdiff lzcmp xzegrep xzfgrep \
+ lzgrep lzegrep lzfgrep lzmore lzless; do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link; \
+ done
+ cd $(DESTDIR)$(mandir)/man1 && \
+ for name in xzcmp lzdiff lzcmp xzegrep xzfgrep \
+ lzgrep lzegrep lzfgrep lzmore lzless; do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link.1; \
+ done
diff --git a/src/scripts/xzdiff.1 b/src/scripts/xzdiff.1
new file mode 100644
index 0000000..318d06f
--- /dev/null
+++ b/src/scripts/xzdiff.1
@@ -0,0 +1,74 @@
+.\"
+.\" Original zdiff.1 for gzip: Jean-loup Gailly
+.\"
+.\" Modifications for XZ Utils: Lasse Collin
+.\" Andrew Dudman
+.\"
+.\" License: GNU GPLv2+
+.\"
+.TH XZDIFF 1 "2009-07-05" "Tukaani" "XZ Utils"
+.SH NAME
+xzcmp, xzdiff, lzcmp, lzdiff \- compare compressed files
+.SH SYNOPSIS
+.B xzcmp
+.RI [ cmp_options "] " file1 " [" file2 ]
+.br
+.B xzdiff
+.RI [ diff_options "] " file1 " [" file2 ]
+.br
+.B lzcmp
+.RI [ cmp_options "] " file1 " [" file2 ]
+.br
+.B lzdiff
+.RI [ diff_options "] " file1 " [" file2 ]
+.SH DESCRIPTION
+.B xzcmp
+and
+.B xzdiff
+invoke
+.BR cmp (1)
+or
+.BR diff (1)
+on files compressed with
+.BR xz (1),
+.BR lzma (1),
+.BR gzip (1),
+or
+.BR bzip2 (1).
+All options specified are passed directly to
+.B cmp
+or
+.BR diff .
+If only one file is specified, then the files compared are
+.I file1
+(which must have a suffix of a supported compression format) and
+.I file1
+from which the compression format suffix has been stripped.
+If two files are specified, then they are uncompressed if necessary and fed to
+.BR cmp (1)
+or
+.BR diff (1).
+The exit status from
+.B cmp
+or
+.B diff
+is preserved.
+.PP
+The names
+.B lzcmp
+and
+.B lzdiff
+are provided for backward compatibility with LZMA Utils.
+.SH "SEE ALSO"
+.BR cmp (1),
+.BR diff (1),
+.BR xz (1),
+.BR gzip (1),
+.BR bzip2 (1),
+.BR zdiff (1)
+.SH BUGS
+Messages from the
+.BR cmp (1)
+or
+.BR diff (1)
+programs refer to temporary filenames instead of those specified.
diff --git a/src/scripts/xzdiff.in b/src/scripts/xzdiff.in
new file mode 100644
index 0000000..2d6e5da
--- /dev/null
+++ b/src/scripts/xzdiff.in
@@ -0,0 +1,172 @@
+#!@POSIX_SHELL@
+
+# Copyright (C) 1998, 2002, 2006, 2007 Free Software Foundation
+# Copyright (C) 1993 Jean-loup Gailly
+
+# Modified for XZ Utils by Andrew Dudman and Lasse Collin.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+
+#SET_PATH - This line is a placeholder to ease patching this script.
+
+# Instead of unsetting XZ_OPT, just make sure that xz will use file format
+# autodetection. This way memory usage limit and thread limit can be
+# specified via XZ_OPT. With gzip and bzip2 it's OK to just unset the
+# environment variables.
+xz='@xz@ --format=auto'
+unset GZIP BZIP BZIP2
+
+case ${0##*/} in
+ *cmp*) prog=xzcmp; cmp=${CMP:-cmp};;
+ *) prog=xzdiff; cmp=${DIFF:-diff};;
+esac
+
+version="$prog (@PACKAGE_NAME@) @VERSION@"
+
+usage="Usage: ${0##*/} [OPTION]... FILE1 [FILE2]
+Compare FILE1 to FILE2, using their uncompressed contents if they are
+compressed. If FILE2 is omitted, then the files compared are FILE1 and
+FILE1 from which the compression format suffix has been stripped.
+
+Do comparisons like '$cmp' does. OPTIONs are the same as for '$cmp'.
+
+Report bugs to <@PACKAGE_BUGREPORT@>."
+
+# sed script to escape all ' for the shell, and then (to handle trailing
+# newlines correctly) turn trailing X on last line into '.
+escape='
+ s/'\''/'\''\\'\'''\''/g
+ $s/X$/'\''/
+'
+
+while :; do
+ case $1 in
+ --h*) printf '%s\n' "$usage" || exit 2; exit;;
+ --v*) echo "$version" || exit 2; exit;;
+ --) shift; break;;
+ -*\'*) cmp="$cmp '"`printf '%sX\n' "$1" | sed "$escape"`;;
+ -?*) cmp="$cmp '$1'";;
+ *) break;;
+ esac
+ shift
+done
+cmp="$cmp --"
+
+for file; do
+ test "X$file" = X- || <"$file" || exit 2
+done
+
+xz1=$xz
+xz2=$xz
+xz_status=0
+exec 3>&1
+
+if test $# -eq 1; then
+ case $1 in
+ *[-.]xz | *[-.]lzma | *.t[lx]z)
+ ;;
+ *[-.]bz2 | *.tbz | *.tbz2)
+ xz1=$bzip2;;
+ *[-.][zZ] | *_z | *[-.]gz | *.t[ag]z)
+ xz1=$gzip;;
+ *)
+ echo >&2 "$0: $1: Unknown compressed file name suffix"
+ exit 2;;
+ esac
+ case $1 in
+ *[-.][zZ] | *_z | *[-.][gx]z | *[-.]bz2 | *[-.]lzma)
+ FILE=`expr "X$1" : 'X\(.*\)[-.][abglmxzZ2]*$'`;;
+ *.t[abglx]z)
+ FILE=`expr "X$1" : 'X\(.*[-.]t\)[abglx]z$'`ar;;
+ *.tbz2)
+ FILE=`expr "X$1" : 'X\(.*[-.]t\)bz2$'`ar;;
+ esac
+ xz_status=$(
+ exec 4>&1
+ ($xz1 -cd -- "$1" 4>&-; echo $? >&4) 3>&- | eval "$cmp" - '"$FILE"' >&3
+ )
+elif test $# -eq 2; then
+ case $1 in
+ *[-.]bz2 | *.tbz | *.tbz2) xz1=$bzip2;;
+ *[-.][zZ] | *_z | *[-.]gz | *.t[ag]z) xz1=$gzip;;
+ esac
+ case $2 in
+ *[-.]bz2 | *.tbz | *.tbz2) xz2=$bzip2;;
+ *[-.][zZ] | *_z | *[-.]gz | *.t[ag]z) xz2=$gzip;;
+ esac
+ case $1 in
+ *[-.][zZ] | *_z | *[-.][gx]z | *[-.]bz2 | *[-.]lzma | *.t[abglx]z | *.tbz2 | -)
+ case "$2" in
+ *[-.][zZ] | *_z | *[-.][gx]z | *[-.]bz2 | *[-.]lzma | *.t[abglx]z | *.tbz2 | -)
+ if test "$1$2" = --; then
+ xz_status=$(
+ exec 4>&1
+ ($xz1 -cdfq - 4>&-; echo $? >&4) 3>&- |
+ eval "$cmp" - - >&3
+ )
+ elif # Reject Solaris 8's buggy /bin/bash 2.03.
+ echo X | (echo X | eval "$cmp" /dev/fd/5 - >/dev/null 2>&1) 5<&0; then
+ xz_status=$(
+ exec 4>&1
+ ($xz1 -cdfq -- "$1" 4>&-; echo $? >&4) 3>&- |
+ ( ($xz2 -cdfq -- "$2" 4>&-; echo $? >&4) 3>&- 5<&- </dev/null |
+ eval "$cmp" /dev/fd/5 - >&3) 5<&0
+ )
+ case $xz_status in
+ *[1-9]*) xz_status=1;;
+ *) xz_status=0;;
+ esac
+ else
+ F=`expr "/$2" : '.*/\(.*\)[-.][ablmtxz2]*$'` || F=$prog
+ tmp=
+ trap '
+ test -n "$tmp" && rm -f "$tmp"
+ (exit 2); exit 2
+ ' HUP INT PIPE TERM 0
+ tmp=`mktemp -t -- "$F.XXXXXX"` || exit 2
+ $xz2 -cdfq -- "$2" > "$tmp" || exit 2
+ xz_status=$(
+ exec 4>&1
+ ($xz1 -cdfq -- "$1" 4>&-; echo $? >&4) 3>&- |
+ eval "$cmp" - '"$tmp"' >&3
+ )
+ cmp_status=$?
+ rm -f "$tmp" || xz_status=$?
+ trap - HUP INT PIPE TERM 0
+ (exit $cmp_status)
+ fi;;
+ *)
+ xz_status=$(
+ exec 4>&1
+ ($xz1 -cdfq -- "$1" 4>&-; echo $? >&4) 3>&- |
+ eval "$cmp" - '"$2"' >&3
+ );;
+ esac;;
+ *)
+ case "$2" in
+ *[-.][zZ] | *_z | *[-.][gx]z | *[-.]bz2 | *[-.]lzma | *.t[abglx]z | *.tbz2 | -)
+ xz_status=$(
+ exec 4>&1
+ ($xz2 -cdfq -- "$2" 4>&-; echo $? >&4) 3>&- |
+ eval "$cmp" '"$1"' - >&3
+ );;
+ *)
+ eval "$cmp" '"$1"' '"$2"';;
+ esac;;
+ esac
+else
+ echo >&2 "$0: Invalid number of operands; try \`${0##*/} --help' for help"
+ exit 2
+fi
+
+cmp_status=$?
+test "$xz_status" -eq 0 || exit 2
+exit $cmp_status
diff --git a/src/scripts/xzgrep.1 b/src/scripts/xzgrep.1
new file mode 100644
index 0000000..996d64a
--- /dev/null
+++ b/src/scripts/xzgrep.1
@@ -0,0 +1,94 @@
+.\"
+.\" Original zgrep.1 for gzip: Jean-loup Gailly
+.\" Charles Levert <charles@comm.polymtl.ca>
+.\"
+.\" Modifications for XZ Utils: Lasse Collin
+.\"
+.\" License: GNU GPLv2+
+.\"
+.TH XZGREP 1 "2009-07-05" "Tukaani" "XZ Utils"
+.SH NAME
+xzgrep \- search compressed files for a regular expression
+.SH SYNOPSIS
+.B xzgrep
+.RI [ grep_options ]
+.RB [ \-e ]
+.I pattern
+.IR file ".\|.\|."
+.br
+.B xzegrep
+.RB ...
+.br
+.B xzfgrep
+.RB ...
+.br
+.B lzgrep
+.RB ...
+.br
+.B lzegrep
+.RB ...
+.br
+.B lzfgrep
+.RB ...
+.SH DESCRIPTION
+.B xzgrep
+invokes
+.BR grep (1)
+on
+.I files
+which may be either uncompressed or compressed with
+.BR xz (1),
+.BR lzma (1),
+.BR gzip (1),
+or
+.BR bzip2 (1).
+All options specified are passed directly to
+.BR grep (1).
+.PP
+If no
+.I file
+is specified, then the standard input is decompressed if necessary and fed to
+.BR grep (1).
+When reading from standard input,
+.BR gzip (1)
+and
+.BR bzip2 (1)
+compressed files are not supported.
+.PP
+If
+.B xzgrep
+is invoked as
+.B xzegrep
+or
+.B xzfgrep
+then
+.BR egrep (1)
+or
+.BR fgrep (1)
+is used instead of
+.BR grep (1).
+The same applies to names
+.BR lzgrep ,
+.BR lzegrep ,
+and
+.BR lzfgrep ,
+which are provided for backward compatibility with LZMA Utils.
+.PP
+.SH ENVIRONMENT
+.TP
+.B GREP
+If the
+.B GREP
+environment variable is set,
+.B xzgrep
+uses it instead of
+.BR grep (1),
+.BR egrep (1),
+or
+.BR fgrep (1).
+.SH "SEE ALSO"
+.BR grep (1),
+.BR xz (1),
+.BR gzip (1),
+.BR bzip2 (1),
+.BR zgrep (1)
diff --git a/src/scripts/xzgrep.in b/src/scripts/xzgrep.in
new file mode 100644
index 0000000..bfa9fb5
--- /dev/null
+++ b/src/scripts/xzgrep.in
@@ -0,0 +1,196 @@
+#!@POSIX_SHELL@
+
+# xzgrep -- a wrapper around a grep program that decompresses files as needed
+# Adapted from a version sent by Charles Levert <charles@comm.polymtl.ca>
+
+# Copyright (C) 1998, 2001, 2002, 2006, 2007 Free Software Foundation
+# Copyright (C) 1993 Jean-loup Gailly
+
+# Modified for XZ Utils by Andrew Dudman and Lasse Collin.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+
+#SET_PATH - This line is a placeholder to ease patching this script.
+
+# Instead of unsetting XZ_OPT, just make sure that xz will use file format
+# autodetection. This way memory usage limit and thread limit can be
+# specified via XZ_OPT. With gzip and bzip2 it's OK to just unset the
+# environment variables.
+xz='@xz@ --format=auto'
+unset GZIP BZIP BZIP2
+
+case ${0##/*} in
+ *egrep*) prog=xzegrep; grep=${GREP:-egrep};;
+ *fgrep*) prog=xzfgrep; grep=${GREP:-fgrep};;
+ *) prog=xzgrep; grep=${GREP:-grep};;
+esac
+
+version="$prog (@PACKAGE_NAME@) @VERSION@"
+
+usage="Usage: ${0##/*} [OPTION]... [-e] PATTERN [FILE]...
+Look for instances of PATTERN in the input FILEs, using their
+uncompressed contents if they are compressed.
+
+OPTIONs are the same as for '$grep'.
+
+Report bugs to <@PACKAGE_BUGREPORT@>."
+
+# sed script to escape all ' for the shell, and then (to handle trailing
+# newlines correctly) turn trailing X on last line into '.
+escape='
+ s/'\''/'\''\\'\'''\''/g
+ $s/X$/'\''/
+'
+operands=
+have_pat=0
+files_with_matches=0
+files_without_matches=0
+no_filename=0
+with_filename=0
+
+while test $# -ne 0; do
+ option=$1
+ shift
+ optarg=
+
+ case $option in
+ (-[0123456789abcdhHiIKLlnoqrRsTuUvVwxyzZ]?*)
+ arg2=-\'$(expr "X${option}X" : 'X-.[0-9]*\(.*\)' | sed "$escape")
+ eval "set -- $arg2 "'${1+"$@"}'
+ option=$(expr "X$option" : 'X\(-.[0-9]*\)');;
+ (--binary-*=* | --[lm]a*=* | --reg*=*)
+ ;;
+ (-[ABCDefm] | --binary-* | --file | --[lm]a* | --reg*)
+ case ${1?"$option option requires an argument"} in
+ (*\'*)
+ optarg=" '"$(printf '%sX\n' "$1" | sed "$escape");;
+ (*)
+ optarg=" '$1'";;
+ esac
+ shift;;
+ (--)
+ break;;
+ (-?*)
+ ;;
+ (*)
+ case $option in
+ (*\'*)
+ operands="$operands '"$(printf '%sX\n' "$option" | sed "$escape");;
+ (*)
+ operands="$operands '$option'";;
+ esac
+ ${POSIXLY_CORRECT+break}
+ continue;;
+ esac
+
+ case $option in
+ (-[drRzZ] | --di* | --exc* | --inc* | --rec* | --nu*)
+ printf >&2 '%s: %s: Option not supported\n' "$0" "$option"
+ exit 2;;
+ (-[ef]* | --file | --file=* | --reg*)
+ have_pat=1;;
+ (--h | --he | --hel | --help)
+ echo "$usage" || exit 2
+ exit;;
+ (-H | --wi | --wit | --with | --with- | --with-f | --with-fi \
+ | --with-fil | --with-file | --with-filen | --with-filena | --with-filenam \
+ | --with-filename)
+ with_filename=1
+ continue;;
+ (-l | --files-with-*)
+ files_with_matches=1;;
+ (-L | --files-witho*)
+ files_without_matches=1;;
+ (--no-f*)
+ no_filename=1;;
+ (-V | --v | --ve | --ver | --vers | --versi | --versio | --version)
+ echo "$version" || exit 2
+ exit;;
+ esac
+
+ case $option in
+ (*\'?*)
+ option=\'$(expr "X${option}X" : 'X\(.*\)' | sed "$escape");;
+ (*)
+ option="'$option'";;
+ esac
+
+ grep="$grep $option$optarg"
+done
+
+eval "set -- $operands "'${1+"$@"}'
+
+if test $have_pat -eq 0; then
+ case ${1?"Missing pattern; try \`${0##*/} --help' for help"} in
+ (*\'*)
+ grep="$grep -- '"$(printf '%sX\n' "$1" | sed "$escape");;
+ (*)
+ grep="$grep -- '$1'";;
+ esac
+ shift
+fi
+
+if test $# -eq 0; then
+ set -- -
+fi
+
+exec 3>&1
+res=0
+
+for i; do
+ case $i in
+ *[-.][zZ] | *_z | *[-.]gz | *.t[ag]z) uncompress="gzip -cdfq";;
+ *[-.]bz2 | *[-.]tbz | *.tbz2) uncompress="bzip2 -cdfq";;
+ *) uncompress="$xz -cdfq";;
+ esac
+ # Fail if xz or grep (or sed) fails.
+ xz_status=$(
+ exec 5>&1
+ ($uncompress -- "$i" 5>&-; echo $? >&5) 3>&- |
+ if test $files_with_matches -eq 1; then
+ eval "$grep" -q && { printf '%s\n' "$i" || exit 2; }
+ elif test $files_without_matches -eq 1; then
+ eval "$grep" -q || {
+ r=$?
+ if test $r -eq 1; then
+ printf '%s\n' "$i" || r=2
+ fi
+ exit $r
+ }
+ elif test $with_filename -eq 0 &&
+ { test $# -eq 1 || test $no_filename -eq 1; }; then
+ eval "$grep"
+ else
+ case $i in
+ (*'
+'* | *'&'* | *'\'* | *'|'*)
+ i=$(printf '%s\n' "$i" |
+ sed '
+ $!N
+ $s/[&\|]/\\&/g
+ $s/\n/\\n/g
+ ');;
+ esac
+ sed_script="s|^|$i:|"
+
+ # Fail if grep or sed fails.
+ r=$(
+ exec 4>&1
+ (eval "$grep" 4>&-; echo $? >&4) 3>&- | sed "$sed_script" >&3 4>&-
+ ) || r=2
+ exit $r
+ fi >&3 5>&-
+ )
+ r=$?
+ test "$xz_status" -eq 0 || test "$xz_status" -eq 2 || r=2
+ test $res -lt $r && res=$r
+done
+exit $res
diff --git a/src/scripts/xzless.1 b/src/scripts/xzless.1
new file mode 100644
index 0000000..299806f
--- /dev/null
+++ b/src/scripts/xzless.1
@@ -0,0 +1,66 @@
+.\"
+.\" Authors: Andrew Dudman
+.\" Lasse Collin
+.\"
+.\" This file has been put into the public domain.
+.\" You can do whatever you want with this file.
+.\"
+.\" (Note that this file is not based on gzip's zless.1.)
+.\"
+.TH XZLESS 1 "2009-07-05" "Tukaani" "XZ Utils"
+.SH NAME
+xzless, lzless \- view xz or lzma compressed (text) files
+.SH SYNOPSIS
+.B xzless
+.RI [ file ...]
+.br
+.B lzless
+.RI [ file ...]
+.SH DESCRIPTION
+.B xzless
+is a filter that displays text from compressed files to a terminal.
+It works on files compressed with
+.BR xz (1)
+or
+.BR lzma (1).
+If no
+.I files
+are given,
+.B xzless
+reads from standard input.
+.PP
+.B xzless
+uses
+.BR less (1)
+to present its output. Unlike
+.BR xzmore ,
+its choice of pager cannot be altered by setting an environment variable.
+Commands are based on both
+.BR more (1)
+and
+.BR vi (1)
+and allow back and forth movement and searching.
+See the
+.BR less (1)
+manual for more information.
+.PP
+The command named
+.B lzless
+is provided for backward compatibility with LZMA Utils.
+.SH ENVIRONMENT
+.TP
+.B LESSMETACHARS
+A list of characters special to the shell. Set by
+.B xzless
+unless it is already set in the environment.
+.TP
+.B LESSOPEN
+Set to a command line to invoke the
+.BR xz (1)
+decompressor for preprocessing the input files to
+.BR less (1).
+.SH "SEE ALSO"
+.BR less (1),
+.BR xz (1),
+.BR xzmore (1),
+.BR zless (1)
diff --git a/src/scripts/xzless.in b/src/scripts/xzless.in
new file mode 100644
index 0000000..a3da697
--- /dev/null
+++ b/src/scripts/xzless.in
@@ -0,0 +1,58 @@
+#!@POSIX_SHELL@
+
+# Copyright (C) 1998, 2002, 2006, 2007 Free Software Foundation
+
+# The original version for gzip was written by Paul Eggert.
+# Modified for XZ Utils by Andrew Dudman and Lasse Collin.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+
+#SET_PATH - This line is a placeholder to ease patching this script.
+
+# Instead of unsetting XZ_OPT, just make sure that xz will use file format
+# autodetection. This way memory usage limit and thread limit can be
+# specified via XZ_OPT.
+xz='@xz@ --format=auto'
+
+version='xzless (@PACKAGE_NAME@) @VERSION@'
+
+usage="Usage: ${0##*/} [OPTION]... [FILE]...
+Like 'less', but operate on the uncompressed contents of xz compressed FILEs.
+
+Options are the same as for 'less'.
+
+Report bugs to <@PACKAGE_BUGREPORT@>."
+
+case $1 in
+ --help) echo "$usage" || exit 2; exit;;
+ --version) echo "$version" || exit 2; exit;;
+esac
+
+if test "${LESSMETACHARS+set}" != set; then
+ # Work around a bug in less 394 and earlier;
+ # it mishandles the metacharacters '$%=~'.
+ space=' '
+ tab=' '
+ nl='
+'
+ LESSMETACHARS="$space$tab$nl'"';*?"()<>[|&^`#\$%=~'
+fi
+
+if test "$(less -V | { read ver && echo ${ver#less }; })" -ge 429; then
+ # less 429 or later: LESSOPEN pipe will be used on
+ # standard input if $LESSOPEN begins with |-.
+ LESSOPEN="|-$xz -cdfq -- %s"
+else
+ LESSOPEN="|$xz -cdfq -- %s"
+fi
+export LESSMETACHARS LESSOPEN
+
+exec less "$@"
diff --git a/src/scripts/xzmore.1 b/src/scripts/xzmore.1
new file mode 100644
index 0000000..42542bf
--- /dev/null
+++ b/src/scripts/xzmore.1
@@ -0,0 +1,54 @@
+.\"
+.\" Original zdiff.1 for gzip: Jean-loup Gailly
+.\" Modifications for XZ Utils: Lasse Collin
+.\"
+.\" License: GNU GPLv2+
+.\"
+.TH XZMORE 1 "2009-07-05" "Tukaani" "XZ Utils"
+.SH NAME
+xzmore, lzmore \- view xz or lzma compressed (text) files
+.SH SYNOPSIS
+.B xzmore
+.RI [ "filename ..." ]
+.br
+.B lzmore
+.RI [ "filename ..." ]
+.SH DESCRIPTION
+.B xzmore
+is a filter which allows examination of
+.BR xz (1)
+or
+.BR lzma (1)
+compressed text files one screenful at a time on a soft-copy terminal.
+.PP
+To use a pager other than the default
+.B more,
+set environment variable
+.B PAGER
+to the name of the desired program.
+The name
+.B lzmore
+is provided for backward compatibility with LZMA Utils.
+.TP
+.BR e " or " q
+When the prompt \-\-More\-\-(Next file:
+.IR file )
+is printed, this command causes
+.B xzmore
+to exit.
+.TP
+.B s
+When the prompt \-\-More\-\-(Next file:
+.IR file )
+is printed, this command causes
+.B xzmore
+to skip the next file and continue.
+.PP
+For list of keyboard commands supported while actually viewing the
+content of a file, refer to manual of the pager you use, usually
+.BR more (1).
+.SH "SEE ALSO"
+.BR more (1),
+.BR xz (1),
+.BR xzless (1),
+.BR zmore (1)
diff --git a/src/scripts/xzmore.in b/src/scripts/xzmore.in
new file mode 100644
index 0000000..940d661
--- /dev/null
+++ b/src/scripts/xzmore.in
@@ -0,0 +1,78 @@
+#!@POSIX_SHELL@
+
+# Copyright (C) 2001, 2002, 2007 Free Software Foundation
+# Copyright (C) 1992, 1993 Jean-loup Gailly
+
+# Modified for XZ Utils by Andrew Dudman and Lasse Collin.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+
+#SET_PATH - This line is a placeholder to ease patching this script.
+
+# Instead of unsetting XZ_OPT, just make sure that xz will use file format
+# autodetection. This way memory usage limit and thread limit can be
+# specified via XZ_OPT.
+xz='@xz@ --format=auto'
+
+version='xzmore (@PACKAGE_NAME@) @VERSION@'
+
+usage="Usage: ${0##*/} [OPTION]... [FILE]...
+Like 'more', but operate on the uncompressed contents of xz compressed FILEs.
+
+Report bugs to <@PACKAGE_BUGREPORT@>."
+
+case $1 in
+ --help) echo "$usage" || exit 2; exit;;
+ --version) echo "$version" || exit 2; exit;;
+esac
+
+oldtty=`stty -g 2>/dev/null`
+if stty -cbreak 2>/dev/null; then
+ cb='cbreak'; ncb='-cbreak'
+else
+ # 'stty min 1' resets eof to ^a on both SunOS and SysV!
+ cb='min 1 -icanon'; ncb='icanon eof ^d'
+fi
+if test $? -eq 0 && test -n "$oldtty"; then
+ trap 'stty $oldtty 2>/dev/null; exit' 0 2 3 5 10 13 15
+else
+ trap 'stty $ncb echo 2>/dev/null; exit' 0 2 3 5 10 13 15
+fi
+
+if test $# = 0; then
+ if test -t 0; then
+ echo "$usage"; exit 1
+ else
+ $xz -cdfq | eval "${PAGER:-more}"
+ fi
+else
+ FIRST=1
+ for FILE; do
+ < "$FILE" || continue
+ if test $FIRST -eq 0; then
+ printf "%s--More--(Next file: %s)" "" "$FILE"
+ stty $cb -echo 2>/dev/null
+ ANS=`dd bs=1 count=1 2>/dev/null`
+ stty $ncb echo 2>/dev/null
+ echo " "
+ case "$ANS" in
+ [eq]) exit;;
+ esac
+ fi
+ if test "$ANS" != 's'; then
+ echo "------> $FILE <------"
+ $xz -cdfq -- "$FILE" | eval "${PAGER:-more}"
+ fi
+ if test -t 1; then
+ FIRST=0
+ fi
+ done
+fi
diff --git a/src/xz/Makefile.am b/src/xz/Makefile.am
new file mode 100644
index 0000000..4dbe0f2
--- /dev/null
+++ b/src/xz/Makefile.am
@@ -0,0 +1,102 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+bin_PROGRAMS = xz
+
+xz_SOURCES = \
+ args.c \
+ args.h \
+ coder.c \
+ coder.h \
+ file_io.c \
+ file_io.h \
+ hardware.c \
+ hardware.h \
+ list.c \
+ list.h \
+ main.c \
+ main.h \
+ message.c \
+ message.h \
+ options.c \
+ options.h \
+ private.h \
+ signals.c \
+ signals.h \
+ suffix.c \
+ suffix.h \
+ util.c \
+ util.h \
+ $(top_srcdir)/src/common/tuklib_open_stdxxx.c \
+ $(top_srcdir)/src/common/tuklib_progname.c \
+ $(top_srcdir)/src/common/tuklib_exit.c \
+ $(top_srcdir)/src/common/tuklib_cpucores.c
+
+if COND_W32
+xz_SOURCES += xz_w32res.rc
+endif
+
+xz_CPPFLAGS = \
+ -DLOCALEDIR=\"$(localedir)\" \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_builddir)/lib \
+ $(STATIC_CPPFLAGS)
+
+xz_LDFLAGS = $(STATIC_LDFLAGS)
+xz_LDADD = $(top_builddir)/src/liblzma/liblzma.la
+
+if COND_GNULIB
+xz_LDADD += $(top_builddir)/lib/libgnu.a
+endif
+
+# libgnu.a may need these libs, so this must be after libgnu.a.
+xz_LDADD += $(LTLIBINTL)
+
+
+# Windows resource compiler support
+.rc.o:
+ $(RC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \
+ $(xz_CPPFLAGS) $(CPPFLAGS) $(RCFLAGS) -i $< -o $@
+
+
+dist_man_MANS = xz.1
+
+
+## Create symlinks for unxz and xzcat for convenience. Create symlinks also
+## for lzma, unlzma, and lzcat for compatibility with LZMA Utils 4.32.x.
+xzlinks = unxz xzcat lzma unlzma lzcat
+
+install-exec-hook:
+ cd $(DESTDIR)$(bindir) && \
+ target=`echo xz | sed '$(transform)'`$(EXEEXT) && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'`$(LN_EXEEXT) && \
+ rm -f $$link && \
+ $(LN_S) $$target $$link; \
+ done
+
+install-data-hook:
+ cd $(DESTDIR)$(mandir)/man1 && \
+ target=`echo xz | sed '$(transform)'` && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link.1 && \
+ $(LN_S) $$target.1 $$link.1; \
+ done
+
+uninstall-hook:
+ cd $(DESTDIR)$(bindir) && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'`$(LN_EXEEXT) && \
+ rm -f $$link; \
+ done
+ cd $(DESTDIR)$(mandir)/man1 && \
+ for name in $(xzlinks); do \
+ link=`echo $$name | sed '$(transform)'` && \
+ rm -f $$link.1; \
+ done
diff --git a/src/xz/args.c b/src/xz/args.c
new file mode 100644
index 0000000..be26bf3
--- /dev/null
+++ b/src/xz/args.c
@@ -0,0 +1,549 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file args.c
+/// \brief Argument parsing
+///
+/// \note Filter-specific options parsing is in options.c.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#include "getopt.h"
+#include <ctype.h>
+
+
+bool opt_stdout = false;
+bool opt_force = false;
+bool opt_keep_original = false;
+bool opt_robot = false;
+
+// We don't modify or free() this, but we need to assign it in some
+// non-const pointers.
+const char *stdin_filename = "(stdin)";
+
+
+static void
+parse_real(args_info *args, int argc, char **argv)
+{
+ enum {
+ OPT_X86 = INT_MIN,
+ OPT_POWERPC,
+ OPT_IA64,
+ OPT_ARM,
+ OPT_ARMTHUMB,
+ OPT_SPARC,
+ OPT_DELTA,
+ OPT_LZMA1,
+ OPT_LZMA2,
+
+ OPT_NO_SPARSE,
+ OPT_FILES,
+ OPT_FILES0,
+ OPT_NO_ADJUST,
+ OPT_INFO_MEMORY,
+ OPT_ROBOT,
+ };
+
+ static const char short_opts[]
+ = "cC:defF:hHlkM:qQrS:tT:vVz0123456789";
+
+ static const struct option long_opts[] = {
+ // Operation mode
+ { "compress", no_argument, NULL, 'z' },
+ { "decompress", no_argument, NULL, 'd' },
+ { "uncompress", no_argument, NULL, 'd' },
+ { "test", no_argument, NULL, 't' },
+ { "list", no_argument, NULL, 'l' },
+
+ // Operation modifiers
+ { "keep", no_argument, NULL, 'k' },
+ { "force", no_argument, NULL, 'f' },
+ { "stdout", no_argument, NULL, 'c' },
+ { "to-stdout", no_argument, NULL, 'c' },
+ { "no-sparse", no_argument, NULL, OPT_NO_SPARSE },
+ { "suffix", required_argument, NULL, 'S' },
+ // { "recursive", no_argument, NULL, 'r' }, // TODO
+ { "files", optional_argument, NULL, OPT_FILES },
+ { "files0", optional_argument, NULL, OPT_FILES0 },
+
+ // Basic compression settings
+ { "format", required_argument, NULL, 'F' },
+ { "check", required_argument, NULL, 'C' },
+ { "no-adjust", no_argument, NULL, OPT_NO_ADJUST },
+ { "memory", required_argument, NULL, 'M' },
+ { "threads", required_argument, NULL, 'T' },
+
+ { "extreme", no_argument, NULL, 'e' },
+ { "fast", no_argument, NULL, '0' },
+ { "best", no_argument, NULL, '9' },
+
+ // Filters
+ { "lzma1", optional_argument, NULL, OPT_LZMA1 },
+ { "lzma2", optional_argument, NULL, OPT_LZMA2 },
+ { "x86", optional_argument, NULL, OPT_X86 },
+ { "powerpc", optional_argument, NULL, OPT_POWERPC },
+ { "ia64", optional_argument, NULL, OPT_IA64 },
+ { "arm", optional_argument, NULL, OPT_ARM },
+ { "armthumb", optional_argument, NULL, OPT_ARMTHUMB },
+ { "sparc", optional_argument, NULL, OPT_SPARC },
+ { "delta", optional_argument, NULL, OPT_DELTA },
+
+ // Other options
+ { "quiet", no_argument, NULL, 'q' },
+ { "verbose", no_argument, NULL, 'v' },
+ { "no-warn", no_argument, NULL, 'Q' },
+ { "robot", no_argument, NULL, OPT_ROBOT },
+ { "info-memory", no_argument, NULL, OPT_INFO_MEMORY },
+ { "help", no_argument, NULL, 'h' },
+ { "long-help", no_argument, NULL, 'H' },
+ { "version", no_argument, NULL, 'V' },
+
+ { NULL, 0, NULL, 0 }
+ };
+
+ int c;
+
+ while ((c = getopt_long(argc, argv, short_opts, long_opts, NULL))
+ != -1) {
+ switch (c) {
+ // Compression preset (also for decompression if --format=raw)
+ case '0': case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9':
+ coder_set_preset(c - '0');
+ break;
+
+ // --memory
+ case 'M': {
+ // Support specifying the limit as a percentage of
+ // installed physical RAM.
+ size_t len = strlen(optarg);
+ if (len > 0 && optarg[len - 1] == '%') {
+ optarg[len - 1] = '\0';
+ hardware_memlimit_set_percentage(
+ str_to_uint64(
+ "memory%", optarg, 1, 100));
+ } else {
+ // On 32-bit systems, SIZE_MAX would make more
+ // sense than UINT64_MAX. But use UINT64_MAX
+ // still so that scripts that assume > 4 GiB
+ // values don't break.
+ hardware_memlimit_set(str_to_uint64(
+ "memory", optarg,
+ 0, UINT64_MAX));
+ }
+
+ break;
+ }
+
+ // --suffix
+ case 'S':
+ suffix_set(optarg);
+ break;
+
+ case 'T':
+ hardware_threadlimit_set(str_to_uint64(
+ "threads", optarg, 0, UINT32_MAX));
+ break;
+
+ // --version
+ case 'V':
+ // This doesn't return.
+ message_version();
+
+ // --stdout
+ case 'c':
+ opt_stdout = true;
+ break;
+
+ // --decompress
+ case 'd':
+ opt_mode = MODE_DECOMPRESS;
+ break;
+
+ // --extreme
+ case 'e':
+ coder_set_extreme();
+ break;
+
+ // --force
+ case 'f':
+ opt_force = true;
+ break;
+
+ // --info-memory
+ case OPT_INFO_MEMORY:
+ // This doesn't return.
+ message_memlimit();
+
+ // --help
+ case 'h':
+ // This doesn't return.
+ message_help(false);
+
+ // --long-help
+ case 'H':
+ // This doesn't return.
+ message_help(true);
+
+ // --list
+ case 'l':
+ opt_mode = MODE_LIST;
+ break;
+
+ // --keep
+ case 'k':
+ opt_keep_original = true;
+ break;
+
+ // --quiet
+ case 'q':
+ message_verbosity_decrease();
+ break;
+
+ case 'Q':
+ set_exit_no_warn();
+ break;
+
+ case 't':
+ opt_mode = MODE_TEST;
+ break;
+
+ // --verbose
+ case 'v':
+ message_verbosity_increase();
+ break;
+
+ // --robot
+ case OPT_ROBOT:
+ opt_robot = true;
+
+ // This is to make sure that floating point numbers
+ // always have a dot as decimal separator.
+ setlocale(LC_NUMERIC, "C");
+ break;
+
+ case 'z':
+ opt_mode = MODE_COMPRESS;
+ break;
+
+ // Filter setup
+
+ case OPT_X86:
+ coder_add_filter(LZMA_FILTER_X86,
+ options_bcj(optarg));
+ break;
+
+ case OPT_POWERPC:
+ coder_add_filter(LZMA_FILTER_POWERPC,
+ options_bcj(optarg));
+ break;
+
+ case OPT_IA64:
+ coder_add_filter(LZMA_FILTER_IA64,
+ options_bcj(optarg));
+ break;
+
+ case OPT_ARM:
+ coder_add_filter(LZMA_FILTER_ARM,
+ options_bcj(optarg));
+ break;
+
+ case OPT_ARMTHUMB:
+ coder_add_filter(LZMA_FILTER_ARMTHUMB,
+ options_bcj(optarg));
+ break;
+
+ case OPT_SPARC:
+ coder_add_filter(LZMA_FILTER_SPARC,
+ options_bcj(optarg));
+ break;
+
+ case OPT_DELTA:
+ coder_add_filter(LZMA_FILTER_DELTA,
+ options_delta(optarg));
+ break;
+
+ case OPT_LZMA1:
+ coder_add_filter(LZMA_FILTER_LZMA1,
+ options_lzma(optarg));
+ break;
+
+ case OPT_LZMA2:
+ coder_add_filter(LZMA_FILTER_LZMA2,
+ options_lzma(optarg));
+ break;
+
+ // Other
+
+ // --format
+ case 'F': {
+ // Just in case, support both "lzma" and "alone" since
+ // the latter was used for forward compatibility in
+ // LZMA Utils 4.32.x.
+ static const struct {
+ char str[8];
+ enum format_type format;
+ } types[] = {
+ { "auto", FORMAT_AUTO },
+ { "xz", FORMAT_XZ },
+ { "lzma", FORMAT_LZMA },
+ { "alone", FORMAT_LZMA },
+ // { "gzip", FORMAT_GZIP },
+ // { "gz", FORMAT_GZIP },
+ { "raw", FORMAT_RAW },
+ };
+
+ size_t i = 0;
+ while (strcmp(types[i].str, optarg) != 0)
+ if (++i == ARRAY_SIZE(types))
+ message_fatal(_("%s: Unknown file "
+ "format type"),
+ optarg);
+
+ opt_format = types[i].format;
+ break;
+ }
+
+ // --check
+ case 'C': {
+ static const struct {
+ char str[8];
+ lzma_check check;
+ } types[] = {
+ { "none", LZMA_CHECK_NONE },
+ { "crc32", LZMA_CHECK_CRC32 },
+ { "crc64", LZMA_CHECK_CRC64 },
+ { "sha256", LZMA_CHECK_SHA256 },
+ };
+
+ size_t i = 0;
+ while (strcmp(types[i].str, optarg) != 0) {
+ if (++i == ARRAY_SIZE(types))
+ message_fatal(_("%s: Unsupported "
+ "integrity "
+ "check type"), optarg);
+ }
+
+ // Use a separate check in case we are using different
+ // liblzma than what was used to compile us.
+ if (!lzma_check_is_supported(types[i].check))
+ message_fatal(_("%s: Unsupported integrity "
+ "check type"), optarg);
+
+ coder_set_check(types[i].check);
+ break;
+ }
+
+ case OPT_NO_SPARSE:
+ io_no_sparse();
+ break;
+
+ case OPT_FILES:
+ args->files_delim = '\n';
+
+ // Fall through
+
+ case OPT_FILES0:
+ if (args->files_name != NULL)
+ message_fatal(_("Only one file can be "
+ "specified with `--files' "
+ "or `--files0'."));
+
+ if (optarg == NULL) {
+ args->files_name = (char *)stdin_filename;
+ args->files_file = stdin;
+ } else {
+ args->files_name = optarg;
+ args->files_file = fopen(optarg,
+ c == OPT_FILES ? "r" : "rb");
+ if (args->files_file == NULL)
+ message_fatal("%s: %s", optarg,
+ strerror(errno));
+ }
+
+ break;
+
+ case OPT_NO_ADJUST:
+ opt_auto_adjust = false;
+ break;
+
+ default:
+ message_try_help();
+ tuklib_exit(E_ERROR, E_ERROR, false);
+ }
+ }
+
+ return;
+}
+
+
+static void
+parse_environment(args_info *args, char *argv0)
+{
+ char *env = getenv("XZ_OPT");
+ if (env == NULL)
+ return;
+
+ // We modify the string, so make a copy of it.
+ env = xstrdup(env);
+
+ // Calculate the number of arguments in env. argc stats at one
+ // to include space for the program name.
+ int argc = 1;
+ bool prev_was_space = true;
+ for (size_t i = 0; env[i] != '\0'; ++i) {
+ // NOTE: Cast to unsigned char is needed so that correct
+ // value gets passed to isspace(), which expects
+ // unsigned char cast to int. Casting to int is done
+ // automatically due to integer promotion, but we need to
+ // force char to unsigned char manually. Otherwise 8-bit
+ // characters would get promoted to wrong value if
+ // char is signed.
+ if (isspace((unsigned char)env[i])) {
+ prev_was_space = true;
+ } else if (prev_was_space) {
+ prev_was_space = false;
+
+ // Keep argc small enough to fit into a singed int
+ // and to keep it usable for memory allocation.
+ if (++argc == my_min(
+ INT_MAX, SIZE_MAX / sizeof(char *)))
+ message_fatal(_("The environment variable "
+ "XZ_OPT contains too many "
+ "arguments"));
+ }
+ }
+
+ // Allocate memory to hold pointers to the arguments. Add one to get
+ // space for the terminating NULL (if some systems happen to need it).
+ char **argv = xmalloc(((size_t)(argc) + 1) * sizeof(char *));
+ argv[0] = argv0;
+ argv[argc] = NULL;
+
+ // Go through the string again. Split the arguments using '\0'
+ // characters and add pointers to the resulting strings to argv.
+ argc = 1;
+ prev_was_space = true;
+ for (size_t i = 0; env[i] != '\0'; ++i) {
+ if (isspace((unsigned char)env[i])) {
+ prev_was_space = true;
+ env[i] = '\0';
+ } else if (prev_was_space) {
+ prev_was_space = false;
+ argv[argc++] = env + i;
+ }
+ }
+
+ // Parse the argument list we got from the environment. All non-option
+ // arguments i.e. filenames are ignored.
+ parse_real(args, argc, argv);
+
+ // Reset the state of the getopt_long() so that we can parse the
+ // command line options too. There are two incompatible ways to
+ // do it.
+#ifdef HAVE_OPTRESET
+ // BSD
+ optind = 1;
+ optreset = 1;
+#else
+ // GNU, Solaris
+ optind = 0;
+#endif
+
+ // We don't need the argument list from environment anymore.
+ free(argv);
+ free(env);
+
+ return;
+}
+
+
+extern void
+args_parse(args_info *args, int argc, char **argv)
+{
+ // Initialize those parts of *args that we need later.
+ args->files_name = NULL;
+ args->files_file = NULL;
+ args->files_delim = '\0';
+
+ // Check how we were called.
+ {
+ // Remove the leading path name, if any.
+ const char *name = strrchr(argv[0], '/');
+ if (name == NULL)
+ name = argv[0];
+ else
+ ++name;
+
+ // NOTE: It's possible that name[0] is now '\0' if argv[0]
+ // is weird, but it doesn't matter here.
+
+ // Look for full command names instead of substrings like
+ // "un", "cat", and "lz" to reduce possibility of false
+ // positives when the programs have been renamed.
+ if (strstr(name, "xzcat") != NULL) {
+ opt_mode = MODE_DECOMPRESS;
+ opt_stdout = true;
+ } else if (strstr(name, "unxz") != NULL) {
+ opt_mode = MODE_DECOMPRESS;
+ } else if (strstr(name, "lzcat") != NULL) {
+ opt_format = FORMAT_LZMA;
+ opt_mode = MODE_DECOMPRESS;
+ opt_stdout = true;
+ } else if (strstr(name, "unlzma") != NULL) {
+ opt_format = FORMAT_LZMA;
+ opt_mode = MODE_DECOMPRESS;
+ } else if (strstr(name, "lzma") != NULL) {
+ opt_format = FORMAT_LZMA;
+ }
+ }
+
+ // First the flags from environment
+ parse_environment(args, argv[0]);
+
+ // Then from the command line
+ parse_real(args, argc, argv);
+
+ // Never remove the source file when the destination is not on disk.
+ // In test mode the data is written nowhere, but setting opt_stdout
+ // will make the rest of the code behave well.
+ if (opt_stdout || opt_mode == MODE_TEST) {
+ opt_keep_original = true;
+ opt_stdout = true;
+ }
+
+ // When compressing, if no --format flag was used, or it
+ // was --format=auto, we compress to the .xz format.
+ if (opt_mode == MODE_COMPRESS && opt_format == FORMAT_AUTO)
+ opt_format = FORMAT_XZ;
+
+ // Compression settings need to be validated (options themselves and
+ // their memory usage) when compressing to any file format. It has to
+ // be done also when uncompressing raw data, since for raw decoding
+ // the options given on the command line are used to know what kind
+ // of raw data we are supposed to decode.
+ if (opt_mode == MODE_COMPRESS || opt_format == FORMAT_RAW)
+ coder_set_compression_settings();
+
+ // If no filenames are given, use stdin.
+ if (argv[optind] == NULL && args->files_name == NULL) {
+ // We don't modify or free() the "-" constant. The caller
+ // modifies this so don't make the struct itself const.
+ static char *names_stdin[2] = { (char *)"-", NULL };
+ args->arg_names = names_stdin;
+ args->arg_count = 1;
+ } else {
+ // We got at least one filename from the command line, or
+ // --files or --files0 was specified.
+ args->arg_names = argv + optind;
+ args->arg_count = argc - optind;
+ }
+
+ return;
+}
diff --git a/src/xz/args.h b/src/xz/args.h
new file mode 100644
index 0000000..ac5959b
--- /dev/null
+++ b/src/xz/args.h
@@ -0,0 +1,42 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file args.h
+/// \brief Argument parsing
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+typedef struct {
+ /// Filenames from command line
+ char **arg_names;
+
+ /// Number of filenames from command line
+ size_t arg_count;
+
+ /// Name of the file from which to read filenames. This is NULL
+ /// if --files or --files0 was not used.
+ char *files_name;
+
+ /// File opened for reading from which filenames are read. This is
+ /// non-NULL only if files_name is non-NULL.
+ FILE *files_file;
+
+ /// Delimiter for filenames read from files_file
+ char files_delim;
+
+} args_info;
+
+
+extern bool opt_stdout;
+extern bool opt_force;
+extern bool opt_keep_original;
+// extern bool opt_recursive;
+extern bool opt_robot;
+
+extern const char *stdin_filename;
+
+extern void args_parse(args_info *args, int argc, char **argv);
diff --git a/src/xz/coder.c b/src/xz/coder.c
new file mode 100644
index 0000000..ff50d63
--- /dev/null
+++ b/src/xz/coder.c
@@ -0,0 +1,655 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file coder.c
+/// \brief Compresses or uncompresses a file
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+/// Return value type for coder_init().
+enum coder_init_ret {
+ CODER_INIT_NORMAL,
+ CODER_INIT_PASSTHRU,
+ CODER_INIT_ERROR,
+};
+
+
+enum operation_mode opt_mode = MODE_COMPRESS;
+enum format_type opt_format = FORMAT_AUTO;
+bool opt_auto_adjust = true;
+
+
+/// Stream used to communicate with liblzma
+static lzma_stream strm = LZMA_STREAM_INIT;
+
+/// Filters needed for all encoding all formats, and also decoding in raw data
+static lzma_filter filters[LZMA_FILTERS_MAX + 1];
+
+/// Input and output buffers
+static io_buf in_buf;
+static io_buf out_buf;
+
+/// Number of filters. Zero indicates that we are using a preset.
+static size_t filters_count = 0;
+
+/// Number of the preset (0-9)
+static size_t preset_number = 6;
+
+/// Indicate if no preset has been explicitly given. In that case, if we need
+/// to auto-adjust for lower memory usage, we won't print a warning.
+static bool preset_default = true;
+
+/// If a preset is used (no custom filter chain) and preset_extreme is true,
+/// a significantly slower compression is used to achieve slightly better
+/// compression ratio.
+static bool preset_extreme = false;
+
+/// Integrity check type
+static lzma_check check;
+
+/// This becomes false if the --check=CHECK option is used.
+static bool check_default = true;
+
+
+extern void
+coder_set_check(lzma_check new_check)
+{
+ check = new_check;
+ check_default = false;
+ return;
+}
+
+
+extern void
+coder_set_preset(size_t new_preset)
+{
+ preset_number = new_preset;
+ preset_default = false;
+ return;
+}
+
+
+extern void
+coder_set_extreme(void)
+{
+ preset_extreme = true;
+ return;
+}
+
+
+extern void
+coder_add_filter(lzma_vli id, void *options)
+{
+ if (filters_count == LZMA_FILTERS_MAX)
+ message_fatal(_("Maximum number of filters is four"));
+
+ filters[filters_count].id = id;
+ filters[filters_count].options = options;
+ ++filters_count;
+
+ return;
+}
+
+
+static void lzma_attribute((noreturn))
+memlimit_too_small(uint64_t memory_usage)
+{
+ message(V_ERROR, _("Memory usage limit is too low for the given "
+ "filter setup."));
+ message_mem_needed(V_ERROR, memory_usage);
+ tuklib_exit(E_ERROR, E_ERROR, false);
+}
+
+
+extern void
+coder_set_compression_settings(void)
+{
+ // Options for LZMA1 or LZMA2 in case we are using a preset.
+ static lzma_options_lzma opt_lzma;
+
+ if (filters_count == 0) {
+ // We are using a preset. This is not a good idea in raw mode
+ // except when playing around with things. Different versions
+ // of this software may use different options in presets, and
+ // thus make uncompressing the raw data difficult.
+ if (opt_format == FORMAT_RAW) {
+ // The message is shown only if warnings are allowed
+ // but the exit status isn't changed.
+ message(V_WARNING, _("Using a preset in raw mode "
+ "is discouraged."));
+ message(V_WARNING, _("The exact options of the "
+ "presets may vary between software "
+ "versions."));
+ }
+
+ // Get the preset for LZMA1 or LZMA2.
+ if (preset_extreme)
+ preset_number |= LZMA_PRESET_EXTREME;
+
+ if (lzma_lzma_preset(&opt_lzma, preset_number))
+ message_bug();
+
+ // Use LZMA2 except with --format=lzma we use LZMA1.
+ filters[0].id = opt_format == FORMAT_LZMA
+ ? LZMA_FILTER_LZMA1 : LZMA_FILTER_LZMA2;
+ filters[0].options = &opt_lzma;
+ filters_count = 1;
+ } else {
+ preset_default = false;
+ }
+
+ // Terminate the filter options array.
+ filters[filters_count].id = LZMA_VLI_UNKNOWN;
+
+ // If we are using the .lzma format, allow exactly one filter
+ // which has to be LZMA1.
+ if (opt_format == FORMAT_LZMA && (filters_count != 1
+ || filters[0].id != LZMA_FILTER_LZMA1))
+ message_fatal(_("The .lzma format supports only "
+ "the LZMA1 filter"));
+
+ // If we are using the .xz format, make sure that there is no LZMA1
+ // filter to prevent LZMA_PROG_ERROR.
+ if (opt_format == FORMAT_XZ)
+ for (size_t i = 0; i < filters_count; ++i)
+ if (filters[i].id == LZMA_FILTER_LZMA1)
+ message_fatal(_("LZMA1 cannot be used "
+ "with the .xz format"));
+
+ // Print the selected filter chain.
+ message_filters_show(V_DEBUG, filters);
+
+ // If using --format=raw, we can be decoding. The memusage function
+ // also validates the filter chain and the options used for the
+ // filters.
+ const uint64_t memory_limit = hardware_memlimit_get();
+ uint64_t memory_usage;
+ if (opt_mode == MODE_COMPRESS)
+ memory_usage = lzma_raw_encoder_memusage(filters);
+ else
+ memory_usage = lzma_raw_decoder_memusage(filters);
+
+ if (memory_usage == UINT64_MAX)
+ message_fatal(_("Unsupported filter chain or filter options"));
+
+ // Print memory usage info before possible dictionary
+ // size auto-adjusting.
+ message_mem_needed(V_DEBUG, memory_usage);
+
+ if (memory_usage > memory_limit) {
+ // If --no-auto-adjust was used or we didn't find LZMA1 or
+ // LZMA2 as the last filter, give an error immediately.
+ // --format=raw implies --no-auto-adjust.
+ if (!opt_auto_adjust || opt_format == FORMAT_RAW)
+ memlimit_too_small(memory_usage);
+
+ assert(opt_mode == MODE_COMPRESS);
+
+ // Look for the last filter if it is LZMA2 or LZMA1, so
+ // we can make it use less RAM. With other filters we don't
+ // know what to do.
+ size_t i = 0;
+ while (filters[i].id != LZMA_FILTER_LZMA2
+ && filters[i].id != LZMA_FILTER_LZMA1) {
+ if (filters[i].id == LZMA_VLI_UNKNOWN)
+ memlimit_too_small(memory_usage);
+
+ ++i;
+ }
+
+ // Decrease the dictionary size until we meet the memory
+ // usage limit. First round down to full mebibytes.
+ lzma_options_lzma *opt = filters[i].options;
+ const uint32_t orig_dict_size = opt->dict_size;
+ opt->dict_size &= ~((UINT32_C(1) << 20) - 1);
+ while (true) {
+ // If it is below 1 MiB, auto-adjusting failed. We
+ // could be more sophisticated and scale it down even
+ // more, but let's see if many complain about this
+ // version.
+ //
+ // FIXME: Displays the scaled memory usage instead
+ // of the original.
+ if (opt->dict_size < (UINT32_C(1) << 20))
+ memlimit_too_small(memory_usage);
+
+ memory_usage = lzma_raw_encoder_memusage(filters);
+ if (memory_usage == UINT64_MAX)
+ message_bug();
+
+ // Accept it if it is low enough.
+ if (memory_usage <= memory_limit)
+ break;
+
+ // Otherwise 1 MiB down and try again. I hope this
+ // isn't too slow method for cases where the original
+ // dict_size is very big.
+ opt->dict_size -= UINT32_C(1) << 20;
+ }
+
+ // Tell the user that we decreased the dictionary size.
+ // However, omit the message if no preset or custom chain
+ // was given. FIXME: Always warn?
+ if (!preset_default)
+ message(V_WARNING, _("Adjusted LZMA%c dictionary size "
+ "from %s MiB to %s MiB to not exceed "
+ "the memory usage limit of %s MiB"),
+ filters[i].id == LZMA_FILTER_LZMA2
+ ? '2' : '1',
+ uint64_to_str(orig_dict_size >> 20, 0),
+ uint64_to_str(opt->dict_size >> 20, 1),
+ uint64_to_str(round_up_to_mib(
+ memory_limit), 2));
+ }
+
+/*
+ // Limit the number of worker threads so that memory usage
+ // limit isn't exceeded.
+ assert(memory_usage > 0);
+ size_t thread_limit = memory_limit / memory_usage;
+ if (thread_limit == 0)
+ thread_limit = 1;
+
+ if (opt_threads > thread_limit)
+ opt_threads = thread_limit;
+*/
+
+ if (check_default) {
+ // The default check type is CRC64, but fallback to CRC32
+ // if CRC64 isn't supported by the copy of liblzma we are
+ // using. CRC32 is always supported.
+ check = LZMA_CHECK_CRC64;
+ if (!lzma_check_is_supported(check))
+ check = LZMA_CHECK_CRC32;
+ }
+
+ return;
+}
+
+
+/// Return true if the data in in_buf seems to be in the .xz format.
+static bool
+is_format_xz(void)
+{
+ return strm.avail_in >= 6 && memcmp(in_buf.u8, "\3757zXZ", 6) == 0;
+}
+
+
+/// Return true if the data in in_buf seems to be in the .lzma format.
+static bool
+is_format_lzma(void)
+{
+ // The .lzma header is 13 bytes.
+ if (strm.avail_in < 13)
+ return false;
+
+ // Decode the LZMA1 properties.
+ lzma_filter filter = { .id = LZMA_FILTER_LZMA1 };
+ if (lzma_properties_decode(&filter, NULL, in_buf.u8, 5) != LZMA_OK)
+ return false;
+
+ // A hack to ditch tons of false positives: We allow only dictionary
+ // sizes that are 2^n or 2^n + 2^(n-1) or UINT32_MAX. LZMA_Alone
+ // created only files with 2^n, but accepts any dictionary size.
+ // If someone complains, this will be reconsidered.
+ lzma_options_lzma *opt = filter.options;
+ const uint32_t dict_size = opt->dict_size;
+ free(opt);
+
+ if (dict_size != UINT32_MAX) {
+ uint32_t d = dict_size - 1;
+ d |= d >> 2;
+ d |= d >> 3;
+ d |= d >> 4;
+ d |= d >> 8;
+ d |= d >> 16;
+ ++d;
+ if (d != dict_size || dict_size == 0)
+ return false;
+ }
+
+ // Another hack to ditch false positives: Assume that if the
+ // uncompressed size is known, it must be less than 256 GiB.
+ // Again, if someone complains, this will be reconsidered.
+ uint64_t uncompressed_size = 0;
+ for (size_t i = 0; i < 8; ++i)
+ uncompressed_size |= (uint64_t)(in_buf.u8[5 + i]) << (i * 8);
+
+ if (uncompressed_size != UINT64_MAX
+ && uncompressed_size > (UINT64_C(1) << 38))
+ return false;
+
+ return true;
+}
+
+
+/// Detect the input file type (for now, this done only when decompressing),
+/// and initialize an appropriate coder. Return value indicates if a normal
+/// liblzma-based coder was initialized (CODER_INIT_NORMAL), if passthru
+/// mode should be used (CODER_INIT_PASSTHRU), or if an error occurred
+/// (CODER_INIT_ERROR).
+static enum coder_init_ret
+coder_init(file_pair *pair)
+{
+ lzma_ret ret = LZMA_PROG_ERROR;
+
+ if (opt_mode == MODE_COMPRESS) {
+ switch (opt_format) {
+ case FORMAT_AUTO:
+ // args.c ensures this.
+ assert(0);
+ break;
+
+ case FORMAT_XZ:
+ ret = lzma_stream_encoder(&strm, filters, check);
+ break;
+
+ case FORMAT_LZMA:
+ ret = lzma_alone_encoder(&strm, filters[0].options);
+ break;
+
+ case FORMAT_RAW:
+ ret = lzma_raw_encoder(&strm, filters);
+ break;
+ }
+ } else {
+ const uint32_t flags = LZMA_TELL_UNSUPPORTED_CHECK
+ | LZMA_CONCATENATED;
+
+ // We abuse FORMAT_AUTO to indicate unknown file format,
+ // for which we may consider passthru mode.
+ enum format_type init_format = FORMAT_AUTO;
+
+ switch (opt_format) {
+ case FORMAT_AUTO:
+ if (is_format_xz())
+ init_format = FORMAT_XZ;
+ else if (is_format_lzma())
+ init_format = FORMAT_LZMA;
+ break;
+
+ case FORMAT_XZ:
+ if (is_format_xz())
+ init_format = FORMAT_XZ;
+ break;
+
+ case FORMAT_LZMA:
+ if (is_format_lzma())
+ init_format = FORMAT_LZMA;
+ break;
+
+ case FORMAT_RAW:
+ init_format = FORMAT_RAW;
+ break;
+ }
+
+ switch (init_format) {
+ case FORMAT_AUTO:
+ // Uknown file format. If --decompress --stdout
+ // --force have been given, then we copy the input
+ // as is to stdout. Checking for MODE_DECOMPRESS
+ // is needed, because we don't want to do use
+ // passthru mode with --test.
+ if (opt_mode == MODE_DECOMPRESS
+ && opt_stdout && opt_force)
+ return CODER_INIT_PASSTHRU;
+
+ ret = LZMA_FORMAT_ERROR;
+ break;
+
+ case FORMAT_XZ:
+ ret = lzma_stream_decoder(&strm,
+ hardware_memlimit_get(), flags);
+ break;
+
+ case FORMAT_LZMA:
+ ret = lzma_alone_decoder(&strm,
+ hardware_memlimit_get());
+ break;
+
+ case FORMAT_RAW:
+ // Memory usage has already been checked in
+ // coder_set_compression_settings().
+ ret = lzma_raw_decoder(&strm, filters);
+ break;
+ }
+
+ // Try to decode the headers. This will catch too low
+ // memory usage limit in case it happens in the first
+ // Block of the first Stream, which is where it very
+ // probably will happen if it is going to happen.
+ if (ret == LZMA_OK && init_format != FORMAT_RAW) {
+ strm.next_out = NULL;
+ strm.avail_out = 0;
+ ret = lzma_code(&strm, LZMA_RUN);
+ }
+ }
+
+ if (ret != LZMA_OK) {
+ message_error("%s: %s", pair->src_name, message_strm(ret));
+ if (ret == LZMA_MEMLIMIT_ERROR)
+ message_mem_needed(V_ERROR, lzma_memusage(&strm));
+
+ return CODER_INIT_ERROR;
+ }
+
+ return CODER_INIT_NORMAL;
+}
+
+
+/// Compress or decompress using liblzma.
+static bool
+coder_normal(file_pair *pair)
+{
+ // Encoder needs to know when we have given all the input to it.
+ // The decoders need to know it too when we are using
+ // LZMA_CONCATENATED. We need to check for src_eof here, because
+ // the first input chunk has been already read, and that may
+ // have been the only chunk we will read.
+ lzma_action action = pair->src_eof ? LZMA_FINISH : LZMA_RUN;
+
+ lzma_ret ret;
+
+ // Assume that something goes wrong.
+ bool success = false;
+
+ strm.next_out = out_buf.u8;
+ strm.avail_out = IO_BUFFER_SIZE;
+
+ while (!user_abort) {
+ // Fill the input buffer if it is empty and we haven't reached
+ // end of file yet.
+ if (strm.avail_in == 0 && !pair->src_eof) {
+ strm.next_in = in_buf.u8;
+ strm.avail_in = io_read(
+ pair, &in_buf, IO_BUFFER_SIZE);
+
+ if (strm.avail_in == SIZE_MAX)
+ break;
+
+ if (pair->src_eof)
+ action = LZMA_FINISH;
+ }
+
+ // Let liblzma do the actual work.
+ ret = lzma_code(&strm, action);
+
+ // Write out if the output buffer became full.
+ if (strm.avail_out == 0) {
+ if (opt_mode != MODE_TEST && io_write(pair, &out_buf,
+ IO_BUFFER_SIZE - strm.avail_out))
+ break;
+
+ strm.next_out = out_buf.u8;
+ strm.avail_out = IO_BUFFER_SIZE;
+ }
+
+ if (ret != LZMA_OK) {
+ // Determine if the return value indicates that we
+ // won't continue coding.
+ const bool stop = ret != LZMA_NO_CHECK
+ && ret != LZMA_UNSUPPORTED_CHECK;
+
+ if (stop) {
+ // Write the remaining bytes even if something
+ // went wrong, because that way the user gets
+ // as much data as possible, which can be good
+ // when trying to get at least some useful
+ // data out of damaged files.
+ if (opt_mode != MODE_TEST && io_write(pair,
+ &out_buf, IO_BUFFER_SIZE
+ - strm.avail_out))
+ break;
+ }
+
+ if (ret == LZMA_STREAM_END) {
+ // Check that there is no trailing garbage.
+ // This is needed for LZMA_Alone and raw
+ // streams.
+ if (strm.avail_in == 0 && !pair->src_eof) {
+ // Try reading one more byte.
+ // Hopefully we don't get any more
+ // input, and thus pair->src_eof
+ // becomes true.
+ strm.avail_in = io_read(
+ pair, &in_buf, 1);
+ if (strm.avail_in == SIZE_MAX)
+ break;
+
+ assert(strm.avail_in == 0
+ || strm.avail_in == 1);
+ }
+
+ if (strm.avail_in == 0) {
+ assert(pair->src_eof);
+ success = true;
+ break;
+ }
+
+ // We hadn't reached the end of the file.
+ ret = LZMA_DATA_ERROR;
+ assert(stop);
+ }
+
+ // If we get here and stop is true, something went
+ // wrong and we print an error. Otherwise it's just
+ // a warning and coding can continue.
+ if (stop) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+ } else {
+ message_warning("%s: %s", pair->src_name,
+ message_strm(ret));
+
+ // When compressing, all possible errors set
+ // stop to true.
+ assert(opt_mode != MODE_COMPRESS);
+ }
+
+ if (ret == LZMA_MEMLIMIT_ERROR) {
+ // Display how much memory it would have
+ // actually needed.
+ message_mem_needed(V_ERROR,
+ lzma_memusage(&strm));
+ }
+
+ if (stop)
+ break;
+ }
+
+ // Show progress information under certain conditions.
+ message_progress_update();
+ }
+
+ return success;
+}
+
+
+/// Copy from input file to output file without processing the data in any
+/// way. This is used only when trying to decompress unrecognized files
+/// with --decompress --stdout --force, so the output is always stdout.
+static bool
+coder_passthru(file_pair *pair)
+{
+ while (strm.avail_in != 0) {
+ if (user_abort)
+ return false;
+
+ if (io_write(pair, &in_buf, strm.avail_in))
+ return false;
+
+ strm.total_in += strm.avail_in;
+ strm.total_out = strm.total_in;
+ message_progress_update();
+
+ strm.avail_in = io_read(pair, &in_buf, IO_BUFFER_SIZE);
+ if (strm.avail_in == SIZE_MAX)
+ return false;
+ }
+
+ return true;
+}
+
+
+extern void
+coder_run(const char *filename)
+{
+ // Set and possibly print the filename for the progress message.
+ message_filename(filename);
+
+ // Try to open the input file.
+ file_pair *pair = io_open_src(filename);
+ if (pair == NULL)
+ return;
+
+ // Assume that something goes wrong.
+ bool success = false;
+
+ // Read the first chunk of input data. This is needed to detect
+ // the input file type (for now, only for decompression).
+ strm.next_in = in_buf.u8;
+ strm.avail_in = io_read(pair, &in_buf, IO_BUFFER_SIZE);
+
+ if (strm.avail_in != SIZE_MAX) {
+ // Initialize the coder. This will detect the file format
+ // and, in decompression or testing mode, check the memory
+ // usage of the first Block too. This way we don't try to
+ // open the destination file if we see that coding wouldn't
+ // work at all anyway. This also avoids deleting the old
+ // "target" file if --force was used.
+ const enum coder_init_ret init_ret = coder_init(pair);
+
+ if (init_ret != CODER_INIT_ERROR && !user_abort) {
+ // Don't open the destination file when --test
+ // is used.
+ if (opt_mode == MODE_TEST || !io_open_dest(pair)) {
+ // Initialize the progress indicator.
+ const uint64_t in_size
+ = pair->src_st.st_size <= 0
+ ? 0 : pair->src_st.st_size;
+ message_progress_start(&strm, in_size);
+
+ // Do the actual coding or passthru.
+ if (init_ret == CODER_INIT_NORMAL)
+ success = coder_normal(pair);
+ else
+ success = coder_passthru(pair);
+
+ message_progress_end(success);
+ }
+ }
+ }
+
+ // Close the file pair. It needs to know if coding was successful to
+ // know if the source or target file should be unlinked.
+ io_close(pair, success);
+
+ return;
+}
diff --git a/src/xz/coder.h b/src/xz/coder.h
new file mode 100644
index 0000000..4626466
--- /dev/null
+++ b/src/xz/coder.h
@@ -0,0 +1,61 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file coder.h
+/// \brief Compresses or uncompresses a file
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+enum operation_mode {
+ MODE_COMPRESS,
+ MODE_DECOMPRESS,
+ MODE_TEST,
+ MODE_LIST,
+};
+
+
+// NOTE: The order of these is significant in suffix.c.
+enum format_type {
+ FORMAT_AUTO,
+ FORMAT_XZ,
+ FORMAT_LZMA,
+ // HEADER_GZIP,
+ FORMAT_RAW,
+};
+
+
+/// Operation mode of the command line tool. This is set in args.c and read
+/// in several files.
+extern enum operation_mode opt_mode;
+
+/// File format to use when encoding or what format(s) to accept when
+/// decoding. This is a global because it's needed also in suffix.c.
+/// This is set in args.c.
+extern enum format_type opt_format;
+
+/// If true, the compression settings are automatically adjusted down if
+/// they exceed the memory usage limit.
+extern bool opt_auto_adjust;
+
+
+/// Set the integrity check type used when compressing
+extern void coder_set_check(lzma_check check);
+
+/// Set preset number
+extern void coder_set_preset(size_t new_preset);
+
+/// Enable extreme mode
+extern void coder_set_extreme(void);
+
+/// Add a filter to the custom filter chain
+extern void coder_add_filter(lzma_vli id, void *options);
+
+///
+extern void coder_set_compression_settings(void);
+
+/// Compress or decompress the given file
+extern void coder_run(const char *filename);
diff --git a/src/xz/file_io.c b/src/xz/file_io.c
new file mode 100644
index 0000000..a78002e
--- /dev/null
+++ b/src/xz/file_io.c
@@ -0,0 +1,957 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file file_io.c
+/// \brief File opening, unlinking, and closing
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#include <fcntl.h>
+
+#ifdef TUKLIB_DOSLIKE
+# include <io.h>
+#else
+static bool warn_fchown;
+#endif
+
+#if defined(HAVE_FUTIMES) || defined(HAVE_FUTIMESAT) || defined(HAVE_UTIMES)
+# include <sys/time.h>
+#elif defined(HAVE_UTIME)
+# include <utime.h>
+#endif
+
+#include "tuklib_open_stdxxx.h"
+
+#ifndef O_BINARY
+# define O_BINARY 0
+#endif
+
+#ifndef O_NOCTTY
+# define O_NOCTTY 0
+#endif
+
+
+/// If true, try to create sparse files when decompressing.
+static bool try_sparse = true;
+
+#ifndef TUKLIB_DOSLIKE
+/// File status flags of standard output. This is used by io_open_dest()
+/// and io_close_dest().
+static int stdout_flags = 0;
+#endif
+
+
+static bool io_write_buf(file_pair *pair, const uint8_t *buf, size_t size);
+
+
+extern void
+io_init(void)
+{
+ // Make sure that stdin, stdout, and and stderr are connected to
+ // a valid file descriptor. Exit immediately with exit code ERROR
+ // if we cannot make the file descriptors valid. Maybe we should
+ // print an error message, but our stderr could be screwed anyway.
+ tuklib_open_stdxxx(E_ERROR);
+
+#ifndef TUKLIB_DOSLIKE
+ // If fchown() fails setting the owner, we warn about it only if
+ // we are root.
+ warn_fchown = geteuid() == 0;
+#endif
+
+#ifdef __DJGPP__
+ // Avoid doing useless things when statting files.
+ // This isn't important but doesn't hurt.
+ _djstat_flags = _STAT_INODE | _STAT_EXEC_EXT
+ | _STAT_EXEC_MAGIC | _STAT_DIRSIZE;
+#endif
+
+ return;
+}
+
+
+extern void
+io_no_sparse(void)
+{
+ try_sparse = false;
+ return;
+}
+
+
+/// \brief Unlink a file
+///
+/// This tries to verify that the file being unlinked really is the file that
+/// we want to unlink by verifying device and inode numbers. There's still
+/// a small unavoidable race, but this is much better than nothing (the file
+/// could have been moved/replaced even hours earlier).
+static void
+io_unlink(const char *name, const struct stat *known_st)
+{
+#if defined(TUKLIB_DOSLIKE)
+ // On DOS-like systems, st_ino is meaningless, so don't bother
+ // testing it. Just silence a compiler warning.
+ (void)known_st;
+#else
+ struct stat new_st;
+
+ // If --force was used, use stat() instead of lstat(). This way
+ // (de)compressing symlinks works correctly. However, it also means
+ // that xz cannot detect if a regular file foo is renamed to bar
+ // and then a symlink foo -> bar is created. Because of stat()
+ // instead of lstat(), xz will think that foo hasn't been replaced
+ // with another file. Thus, xz will remove foo even though it no
+ // longer is the same file that xz used when it started compressing.
+ // Probably it's not too bad though, so this doesn't need a more
+ // complex fix.
+ const int stat_ret = opt_force
+ ? stat(name, &new_st) : lstat(name, &new_st);
+
+ if (stat_ret
+# ifdef __VMS
+ // st_ino is an array, and we don't want to
+ // compare st_dev at all.
+ || memcmp(&new_st.st_ino, &known_st->st_ino,
+ sizeof(new_st.st_ino)) != 0
+# else
+ // Typical POSIX-like system
+ || new_st.st_dev != known_st->st_dev
+ || new_st.st_ino != known_st->st_ino
+# endif
+ )
+ // TRANSLATORS: When compression or decompression finishes,
+ // and xz is going to remove the source file, xz first checks
+ // if the source file still exists, and if it does, does its
+ // device and inode numbers match what xz saw when it opened
+ // the source file. If these checks fail, this message is
+ // shown, %s being the filename, and the file is not deleted.
+ // The check for device and inode numbers is there, because
+ // it is possible that the user has put a new file in place
+ // of the original file, and in that case it obviously
+ // shouldn't be removed.
+ message_error(_("%s: File seems to have been moved, "
+ "not removing"), name);
+ else
+#endif
+ // There's a race condition between lstat() and unlink()
+ // but at least we have tried to avoid removing wrong file.
+ if (unlink(name))
+ message_error(_("%s: Cannot remove: %s"),
+ name, strerror(errno));
+
+ return;
+}
+
+
+/// \brief Copies owner/group and permissions
+///
+/// \todo ACL and EA support
+///
+static void
+io_copy_attrs(const file_pair *pair)
+{
+ // Skip chown and chmod on Windows.
+#ifndef TUKLIB_DOSLIKE
+ // This function is more tricky than you may think at first.
+ // Blindly copying permissions may permit users to access the
+ // destination file who didn't have permission to access the
+ // source file.
+
+ // Try changing the owner of the file. If we aren't root or the owner
+ // isn't already us, fchown() probably doesn't succeed. We warn
+ // about failing fchown() only if we are root.
+ if (fchown(pair->dest_fd, pair->src_st.st_uid, -1) && warn_fchown)
+ message_warning(_("%s: Cannot set the file owner: %s"),
+ pair->dest_name, strerror(errno));
+
+ mode_t mode;
+
+ if (fchown(pair->dest_fd, -1, pair->src_st.st_gid)) {
+ message_warning(_("%s: Cannot set the file group: %s"),
+ pair->dest_name, strerror(errno));
+ // We can still safely copy some additional permissions:
+ // `group' must be at least as strict as `other' and
+ // also vice versa.
+ //
+ // NOTE: After this, the owner of the source file may
+ // get additional permissions. This shouldn't be too bad,
+ // because the owner would have had permission to chmod
+ // the original file anyway.
+ mode = ((pair->src_st.st_mode & 0070) >> 3)
+ & (pair->src_st.st_mode & 0007);
+ mode = (pair->src_st.st_mode & 0700) | (mode << 3) | mode;
+ } else {
+ // Drop the setuid, setgid, and sticky bits.
+ mode = pair->src_st.st_mode & 0777;
+ }
+
+ if (fchmod(pair->dest_fd, mode))
+ message_warning(_("%s: Cannot set the file permissions: %s"),
+ pair->dest_name, strerror(errno));
+#endif
+
+ // Copy the timestamps. We have several possible ways to do this, of
+ // which some are better in both security and precision.
+ //
+ // First, get the nanosecond part of the timestamps. As of writing,
+ // it's not standardized by POSIX, and there are several names for
+ // the same thing in struct stat.
+ long atime_nsec;
+ long mtime_nsec;
+
+# if defined(HAVE_STRUCT_STAT_ST_ATIM_TV_NSEC)
+ // GNU and Solaris
+ atime_nsec = pair->src_st.st_atim.tv_nsec;
+ mtime_nsec = pair->src_st.st_mtim.tv_nsec;
+
+# elif defined(HAVE_STRUCT_STAT_ST_ATIMESPEC_TV_NSEC)
+ // BSD
+ atime_nsec = pair->src_st.st_atimespec.tv_nsec;
+ mtime_nsec = pair->src_st.st_mtimespec.tv_nsec;
+
+# elif defined(HAVE_STRUCT_STAT_ST_ATIMENSEC)
+ // GNU and BSD without extensions
+ atime_nsec = pair->src_st.st_atimensec;
+ mtime_nsec = pair->src_st.st_mtimensec;
+
+# elif defined(HAVE_STRUCT_STAT_ST_UATIME)
+ // Tru64
+ atime_nsec = pair->src_st.st_uatime * 1000;
+ mtime_nsec = pair->src_st.st_umtime * 1000;
+
+# elif defined(HAVE_STRUCT_STAT_ST_ATIM_ST__TIM_TV_NSEC)
+ // UnixWare
+ atime_nsec = pair->src_st.st_atim.st__tim.tv_nsec;
+ mtime_nsec = pair->src_st.st_mtim.st__tim.tv_nsec;
+
+# else
+ // Safe fallback
+ atime_nsec = 0;
+ mtime_nsec = 0;
+# endif
+
+ // Construct a structure to hold the timestamps and call appropriate
+ // function to set the timestamps.
+#if defined(HAVE_FUTIMENS)
+ // Use nanosecond precision.
+ struct timespec tv[2];
+ tv[0].tv_sec = pair->src_st.st_atime;
+ tv[0].tv_nsec = atime_nsec;
+ tv[1].tv_sec = pair->src_st.st_mtime;
+ tv[1].tv_nsec = mtime_nsec;
+
+ (void)futimens(pair->dest_fd, tv);
+
+#elif defined(HAVE_FUTIMES) || defined(HAVE_FUTIMESAT) || defined(HAVE_UTIMES)
+ // Use microsecond precision.
+ struct timeval tv[2];
+ tv[0].tv_sec = pair->src_st.st_atime;
+ tv[0].tv_usec = atime_nsec / 1000;
+ tv[1].tv_sec = pair->src_st.st_mtime;
+ tv[1].tv_usec = mtime_nsec / 1000;
+
+# if defined(HAVE_FUTIMES)
+ (void)futimes(pair->dest_fd, tv);
+# elif defined(HAVE_FUTIMESAT)
+ (void)futimesat(pair->dest_fd, NULL, tv);
+# else
+ // Argh, no function to use a file descriptor to set the timestamp.
+ (void)utimes(pair->dest_name, tv);
+# endif
+
+#elif defined(HAVE_UTIME)
+ // Use one-second precision. utime() doesn't support using file
+ // descriptor either. Some systems have broken utime() prototype
+ // so don't make this const.
+ struct utimbuf buf = {
+ .actime = pair->src_st.st_atime,
+ .modtime = pair->src_st.st_mtime,
+ };
+
+ // Avoid warnings.
+ (void)atime_nsec;
+ (void)mtime_nsec;
+
+ (void)utime(pair->dest_name, &buf);
+#endif
+
+ return;
+}
+
+
+/// Opens the source file. Returns false on success, true on error.
+static bool
+io_open_src_real(file_pair *pair)
+{
+ // There's nothing to open when reading from stdin.
+ if (pair->src_name == stdin_filename) {
+ pair->src_fd = STDIN_FILENO;
+#ifdef TUKLIB_DOSLIKE
+ setmode(STDIN_FILENO, O_BINARY);
+#endif
+ return false;
+ }
+
+ // Symlinks are not followed unless writing to stdout or --force
+ // was used.
+ const bool follow_symlinks = opt_stdout || opt_force;
+
+ // We accept only regular files if we are writing the output
+ // to disk too. bzip2 allows overriding this with --force but
+ // gzip and xz don't.
+ const bool reg_files_only = !opt_stdout;
+
+ // Flags for open()
+ int flags = O_RDONLY | O_BINARY | O_NOCTTY;
+
+#ifndef TUKLIB_DOSLIKE
+ // If we accept only regular files, we need to be careful to avoid
+ // problems with special files like devices and FIFOs. O_NONBLOCK
+ // prevents blocking when opening such files. When we want to accept
+ // special files, we must not use O_NONBLOCK, or otherwise we won't
+ // block waiting e.g. FIFOs to become readable.
+ if (reg_files_only)
+ flags |= O_NONBLOCK;
+#endif
+
+#if defined(O_NOFOLLOW)
+ if (!follow_symlinks)
+ flags |= O_NOFOLLOW;
+#elif !defined(TUKLIB_DOSLIKE)
+ // Some POSIX-like systems lack O_NOFOLLOW (it's not required
+ // by POSIX). Check for symlinks with a separate lstat() on
+ // these systems.
+ if (!follow_symlinks) {
+ struct stat st;
+ if (lstat(pair->src_name, &st)) {
+ message_error("%s: %s", pair->src_name,
+ strerror(errno));
+ return true;
+
+ } else if (S_ISLNK(st.st_mode)) {
+ message_warning(_("%s: Is a symbolic link, "
+ "skipping"), pair->src_name);
+ return true;
+ }
+ }
+#else
+ // Avoid warnings.
+ (void)follow_symlinks;
+#endif
+
+ // Try to open the file. If we are accepting non-regular files,
+ // unblock the caught signals so that open() can be interrupted
+ // if it blocks e.g. due to a FIFO file.
+ if (!reg_files_only)
+ signals_unblock();
+
+ // Maybe this wouldn't need a loop, since all the signal handlers for
+ // which we don't use SA_RESTART set user_abort to true. But it
+ // doesn't hurt to have it just in case.
+ do {
+ pair->src_fd = open(pair->src_name, flags);
+ } while (pair->src_fd == -1 && errno == EINTR && !user_abort);
+
+ if (!reg_files_only)
+ signals_block();
+
+ if (pair->src_fd == -1) {
+ // If we were interrupted, don't display any error message.
+ if (errno == EINTR) {
+ // All the signals that don't have SA_RESTART
+ // set user_abort.
+ assert(user_abort);
+ return true;
+ }
+
+#ifdef O_NOFOLLOW
+ // Give an understandable error message in if reason
+ // for failing was that the file was a symbolic link.
+ //
+ // Note that at least Linux, OpenBSD, Solaris, and Darwin
+ // use ELOOP to indicate if O_NOFOLLOW was the reason
+ // that open() failed. Because there may be
+ // directories in the pathname, ELOOP may occur also
+ // because of a symlink loop in the directory part.
+ // So ELOOP doesn't tell us what actually went wrong.
+ //
+ // FreeBSD associates EMLINK with O_NOFOLLOW and
+ // Tru64 uses ENOTSUP. We use these directly here
+ // and skip the lstat() call and the associated race.
+ // I want to hear if there are other kernels that
+ // fail with something else than ELOOP with O_NOFOLLOW.
+ bool was_symlink = false;
+
+# if defined(__FreeBSD__) || defined(__DragonFly__)
+ if (errno == EMLINK)
+ was_symlink = true;
+
+# elif defined(__digital__) && defined(__unix__)
+ if (errno == ENOTSUP)
+ was_symlink = true;
+
+# elif defined(__NetBSD__)
+ // FIXME? As of 2008-11-20, NetBSD doesn't document what
+ // errno is used with O_NOFOLLOW. It seems to be EFTYPE,
+ // but since it isn't documented, it may be wrong to rely
+ // on it here.
+ if (errno == EFTYPE)
+ was_symlink = true;
+
+# else
+ if (errno == ELOOP && !follow_symlinks) {
+ const int saved_errno = errno;
+ struct stat st;
+ if (lstat(pair->src_name, &st) == 0
+ && S_ISLNK(st.st_mode))
+ was_symlink = true;
+
+ errno = saved_errno;
+ }
+# endif
+
+ if (was_symlink)
+ message_warning(_("%s: Is a symbolic link, "
+ "skipping"), pair->src_name);
+ else
+#endif
+ // Something else than O_NOFOLLOW failing
+ // (assuming that the race conditions didn't
+ // confuse us).
+ message_error("%s: %s", pair->src_name,
+ strerror(errno));
+
+ return true;
+ }
+
+#ifndef TUKLIB_DOSLIKE
+ // Drop O_NONBLOCK, which is used only when we are accepting only
+ // regular files. After the open() call, we want things to block
+ // instead of giving EAGAIN.
+ if (reg_files_only) {
+ flags = fcntl(pair->src_fd, F_GETFL);
+ if (flags == -1)
+ goto error_msg;
+
+ flags &= ~O_NONBLOCK;
+
+ if (fcntl(pair->src_fd, F_SETFL, flags))
+ goto error_msg;
+ }
+#endif
+
+ // Stat the source file. We need the result also when we copy
+ // the permissions, and when unlinking.
+ if (fstat(pair->src_fd, &pair->src_st))
+ goto error_msg;
+
+ if (S_ISDIR(pair->src_st.st_mode)) {
+ message_warning(_("%s: Is a directory, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+ if (reg_files_only) {
+ if (!S_ISREG(pair->src_st.st_mode)) {
+ message_warning(_("%s: Not a regular file, "
+ "skipping"), pair->src_name);
+ goto error;
+ }
+
+ // These are meaningless on Windows.
+#ifndef TUKLIB_DOSLIKE
+ if (pair->src_st.st_mode & (S_ISUID | S_ISGID)) {
+ // gzip rejects setuid and setgid files even
+ // when --force was used. bzip2 doesn't check
+ // for them, but calls fchown() after fchmod(),
+ // and many systems automatically drop setuid
+ // and setgid bits there.
+ //
+ // We accept setuid and setgid files if
+ // --force was used. We drop these bits
+ // explicitly in io_copy_attr().
+ message_warning(_("%s: File has setuid or "
+ "setgid bit set, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+ if (pair->src_st.st_mode & S_ISVTX) {
+ message_warning(_("%s: File has sticky bit "
+ "set, skipping"),
+ pair->src_name);
+ goto error;
+ }
+
+ if (pair->src_st.st_nlink > 1) {
+ message_warning(_("%s: Input file has more "
+ "than one hard link, "
+ "skipping"), pair->src_name);
+ goto error;
+ }
+#endif
+ }
+
+ return false;
+
+error_msg:
+ message_error("%s: %s", pair->src_name, strerror(errno));
+error:
+ (void)close(pair->src_fd);
+ return true;
+}
+
+
+extern file_pair *
+io_open_src(const char *src_name)
+{
+ if (is_empty_filename(src_name))
+ return NULL;
+
+ // Since we have only one file open at a time, we can use
+ // a statically allocated structure.
+ static file_pair pair;
+
+ pair = (file_pair){
+ .src_name = src_name,
+ .dest_name = NULL,
+ .src_fd = -1,
+ .dest_fd = -1,
+ .src_eof = false,
+ .dest_try_sparse = false,
+ .dest_pending_sparse = 0,
+ };
+
+ // Block the signals, for which we have a custom signal handler, so
+ // that we don't need to worry about EINTR.
+ signals_block();
+ const bool error = io_open_src_real(&pair);
+ signals_unblock();
+
+ return error ? NULL : &pair;
+}
+
+
+/// \brief Closes source file of the file_pair structure
+///
+/// \param pair File whose src_fd should be closed
+/// \param success If true, the file will be removed from the disk if
+/// closing succeeds and --keep hasn't been used.
+static void
+io_close_src(file_pair *pair, bool success)
+{
+ if (pair->src_fd != STDIN_FILENO && pair->src_fd != -1) {
+#ifdef TUKLIB_DOSLIKE
+ (void)close(pair->src_fd);
+#endif
+
+ // If we are going to unlink(), do it before closing the file.
+ // This way there's no risk that someone replaces the file and
+ // happens to get same inode number, which would make us
+ // unlink() wrong file.
+ //
+ // NOTE: DOS-like systems are an exception to this, because
+ // they don't allow unlinking files that are open. *sigh*
+ if (success && !opt_keep_original)
+ io_unlink(pair->src_name, &pair->src_st);
+
+#ifndef TUKLIB_DOSLIKE
+ (void)close(pair->src_fd);
+#endif
+ }
+
+ return;
+}
+
+
+static bool
+io_open_dest_real(file_pair *pair)
+{
+ if (opt_stdout || pair->src_fd == STDIN_FILENO) {
+ // We don't modify or free() this.
+ pair->dest_name = (char *)"(stdout)";
+ pair->dest_fd = STDOUT_FILENO;
+#ifdef TUKLIB_DOSLIKE
+ setmode(STDOUT_FILENO, O_BINARY);
+#endif
+ } else {
+ pair->dest_name = suffix_get_dest_name(pair->src_name);
+ if (pair->dest_name == NULL)
+ return true;
+
+ // If --force was used, unlink the target file first.
+ if (opt_force && unlink(pair->dest_name) && errno != ENOENT) {
+ message_error(_("%s: Cannot remove: %s"),
+ pair->dest_name, strerror(errno));
+ free(pair->dest_name);
+ return true;
+ }
+
+ // Open the file.
+ const int flags = O_WRONLY | O_BINARY | O_NOCTTY
+ | O_CREAT | O_EXCL;
+ const mode_t mode = S_IRUSR | S_IWUSR;
+ pair->dest_fd = open(pair->dest_name, flags, mode);
+
+ if (pair->dest_fd == -1) {
+ message_error("%s: %s", pair->dest_name,
+ strerror(errno));
+ free(pair->dest_name);
+ return true;
+ }
+ }
+
+ // If this really fails... well, we have a safe fallback.
+ if (fstat(pair->dest_fd, &pair->dest_st)) {
+#if defined(__VMS)
+ pair->dest_st.st_ino[0] = 0;
+ pair->dest_st.st_ino[1] = 0;
+ pair->dest_st.st_ino[2] = 0;
+#elif !defined(TUKLIB_DOSLIKE)
+ pair->dest_st.st_dev = 0;
+ pair->dest_st.st_ino = 0;
+#endif
+#ifndef TUKLIB_DOSLIKE
+ } else if (try_sparse && opt_mode == MODE_DECOMPRESS) {
+ // When writing to standard output, we need to be extra
+ // careful:
+ // - It may be connected to something else than
+ // a regular file.
+ // - We aren't necessarily writing to a new empty file
+ // or to the end of an existing file.
+ // - O_APPEND may be active.
+ //
+ // TODO: I'm keeping this disabled for DOS-like systems
+ // for now. FAT doesn't support sparse files, but NTFS
+ // does, so maybe this should be enabled on Windows after
+ // some testing.
+ if (pair->dest_fd == STDOUT_FILENO) {
+ if (!S_ISREG(pair->dest_st.st_mode))
+ return false;
+
+ const int flags = fcntl(STDOUT_FILENO, F_GETFL);
+ if (flags == -1)
+ return false;
+
+ if (flags & O_APPEND) {
+ // Creating a sparse file is not possible
+ // when O_APPEND is active (it's used by
+ // shell's >> redirection). As I understand
+ // it, it is safe to temporarily disable
+ // O_APPEND in xz, because if someone
+ // happened to write to the same file at the
+ // same time, results would be bad anyway
+ // (users shouldn't assume that xz uses any
+ // specific block size when writing data).
+ //
+ // The write position may be something else
+ // than the end of the file, so we must fix
+ // it to start writing at the end of the file
+ // to imitate O_APPEND.
+ if (lseek(STDOUT_FILENO, 0, SEEK_END) == -1)
+ return false;
+
+ if (fcntl(STDOUT_FILENO, F_SETFL,
+ stdout_flags & ~O_APPEND))
+ return false;
+
+ // Remember the flags so that io_close_dest()
+ // can restore them.
+ stdout_flags = flags;
+
+ } else if (lseek(STDOUT_FILENO, 0, SEEK_CUR)
+ != pair->dest_st.st_size) {
+ // Writing won't start exactly at the end
+ // of the file. We cannot use sparse output,
+ // because it would probably corrupt the file.
+ return false;
+ }
+ }
+
+ pair->dest_try_sparse = true;
+#endif
+ }
+
+ return false;
+}
+
+
+extern bool
+io_open_dest(file_pair *pair)
+{
+ signals_block();
+ const bool ret = io_open_dest_real(pair);
+ signals_unblock();
+ return ret;
+}
+
+
+/// \brief Closes destination file of the file_pair structure
+///
+/// \param pair File whose dest_fd should be closed
+/// \param success If false, the file will be removed from the disk.
+///
+/// \return Zero if closing succeeds. On error, -1 is returned and
+/// error message printed.
+static bool
+io_close_dest(file_pair *pair, bool success)
+{
+#ifndef TUKLIB_DOSLIKE
+ // If io_open_dest() has disabled O_APPEND, restore it here.
+ if (stdout_flags != 0) {
+ assert(pair->dest_fd == STDOUT_FILENO);
+
+ const int fail = fcntl(STDOUT_FILENO, F_SETFL, stdout_flags);
+ stdout_flags = 0;
+
+ if (fail) {
+ message_error(_("Error restoring the O_APPEND flag "
+ "to standard output: %s"),
+ strerror(errno));
+ return true;
+ }
+ }
+#endif
+
+ if (pair->dest_fd == -1 || pair->dest_fd == STDOUT_FILENO)
+ return false;
+
+ if (close(pair->dest_fd)) {
+ message_error(_("%s: Closing the file failed: %s"),
+ pair->dest_name, strerror(errno));
+
+ // Closing destination file failed, so we cannot trust its
+ // contents. Get rid of junk:
+ io_unlink(pair->dest_name, &pair->dest_st);
+ free(pair->dest_name);
+ return true;
+ }
+
+ // If the operation using this file wasn't successful, we git rid
+ // of the junk file.
+ if (!success)
+ io_unlink(pair->dest_name, &pair->dest_st);
+
+ free(pair->dest_name);
+
+ return false;
+}
+
+
+extern void
+io_close(file_pair *pair, bool success)
+{
+ // Take care of sparseness at the end of the output file.
+ if (success && pair->dest_try_sparse
+ && pair->dest_pending_sparse > 0) {
+ // Seek forward one byte less than the size of the pending
+ // hole, then write one zero-byte. This way the file grows
+ // to its correct size. An alternative would be to use
+ // ftruncate() but that isn't portable enough (e.g. it
+ // doesn't work with FAT on Linux; FAT isn't that important
+ // since it doesn't support sparse files anyway, but we don't
+ // want to create corrupt files on it).
+ if (lseek(pair->dest_fd, pair->dest_pending_sparse - 1,
+ SEEK_CUR) == -1) {
+ message_error(_("%s: Seeking failed when trying "
+ "to create a sparse file: %s"),
+ pair->dest_name, strerror(errno));
+ success = false;
+ } else {
+ const uint8_t zero[1] = { '\0' };
+ if (io_write_buf(pair, zero, 1))
+ success = false;
+ }
+ }
+
+ signals_block();
+
+ // Copy the file attributes. We need to skip this if destination
+ // file isn't open or it is standard output.
+ if (success && pair->dest_fd != -1 && pair->dest_fd != STDOUT_FILENO)
+ io_copy_attrs(pair);
+
+ // Close the destination first. If it fails, we must not remove
+ // the source file!
+ if (io_close_dest(pair, success))
+ success = false;
+
+ // Close the source file, and unlink it if the operation using this
+ // file pair was successful and we haven't requested to keep the
+ // source file.
+ io_close_src(pair, success);
+
+ signals_unblock();
+
+ return;
+}
+
+
+extern size_t
+io_read(file_pair *pair, io_buf *buf_union, size_t size)
+{
+ // We use small buffers here.
+ assert(size < SSIZE_MAX);
+
+ uint8_t *buf = buf_union->u8;
+ size_t left = size;
+
+ while (left > 0) {
+ const ssize_t amount = read(pair->src_fd, buf, left);
+
+ if (amount == 0) {
+ pair->src_eof = true;
+ break;
+ }
+
+ if (amount == -1) {
+ if (errno == EINTR) {
+ if (user_abort)
+ return SIZE_MAX;
+
+ continue;
+ }
+
+ message_error(_("%s: Read error: %s"),
+ pair->src_name, strerror(errno));
+
+ // FIXME Is this needed?
+ pair->src_eof = true;
+
+ return SIZE_MAX;
+ }
+
+ buf += (size_t)(amount);
+ left -= (size_t)(amount);
+ }
+
+ return size - left;
+}
+
+
+extern bool
+io_pread(file_pair *pair, io_buf *buf, size_t size, off_t pos)
+{
+ // Using lseek() and read() is more portable than pread() and
+ // for us it is as good as real pread().
+ if (lseek(pair->src_fd, pos, SEEK_SET) != pos) {
+ message_error(_("%s: Error seeking the file: %s"),
+ pair->src_name, strerror(errno));
+ return true;
+ }
+
+ const size_t amount = io_read(pair, buf, size);
+ if (amount == SIZE_MAX)
+ return true;
+
+ if (amount != size) {
+ message_error(_("%s: Unexpected end of file"),
+ pair->src_name);
+ return true;
+ }
+
+ return false;
+}
+
+
+static bool
+is_sparse(const io_buf *buf)
+{
+ assert(IO_BUFFER_SIZE % sizeof(uint64_t) == 0);
+
+ for (size_t i = 0; i < ARRAY_SIZE(buf->u64); ++i)
+ if (buf->u64[i] != 0)
+ return false;
+
+ return true;
+}
+
+
+static bool
+io_write_buf(file_pair *pair, const uint8_t *buf, size_t size)
+{
+ assert(size < SSIZE_MAX);
+
+ while (size > 0) {
+ const ssize_t amount = write(pair->dest_fd, buf, size);
+ if (amount == -1) {
+ if (errno == EINTR) {
+ if (user_abort)
+ return -1;
+
+ continue;
+ }
+
+ // Handle broken pipe specially. gzip and bzip2
+ // don't print anything on SIGPIPE. In addition,
+ // gzip --quiet uses exit status 2 (warning) on
+ // broken pipe instead of whatever raise(SIGPIPE)
+ // would make it return. It is there to hide "Broken
+ // pipe" message on some old shells (probably old
+ // GNU bash).
+ //
+ // We don't do anything special with --quiet, which
+ // is what bzip2 does too. If we get SIGPIPE, we
+ // will handle it like other signals by setting
+ // user_abort, and get EPIPE here.
+ if (errno != EPIPE)
+ message_error(_("%s: Write error: %s"),
+ pair->dest_name, strerror(errno));
+
+ return true;
+ }
+
+ buf += (size_t)(amount);
+ size -= (size_t)(amount);
+ }
+
+ return false;
+}
+
+
+extern bool
+io_write(file_pair *pair, const io_buf *buf, size_t size)
+{
+ assert(size <= IO_BUFFER_SIZE);
+
+ if (pair->dest_try_sparse) {
+ // Check if the block is sparse (contains only zeros). If it
+ // sparse, we just store the amount and return. We will take
+ // care of actually skipping over the hole when we hit the
+ // next data block or close the file.
+ //
+ // Since io_close() requires that dest_pending_sparse > 0
+ // if the file ends with sparse block, we must also return
+ // if size == 0 to avoid doing the lseek().
+ if (size == IO_BUFFER_SIZE) {
+ if (is_sparse(buf)) {
+ pair->dest_pending_sparse += size;
+ return false;
+ }
+ } else if (size == 0) {
+ return false;
+ }
+
+ // This is not a sparse block. If we have a pending hole,
+ // skip it now.
+ if (pair->dest_pending_sparse > 0) {
+ if (lseek(pair->dest_fd, pair->dest_pending_sparse,
+ SEEK_CUR) == -1) {
+ message_error(_("%s: Seeking failed when "
+ "trying to create a sparse "
+ "file: %s"), pair->dest_name,
+ strerror(errno));
+ return true;
+ }
+
+ pair->dest_pending_sparse = 0;
+ }
+ }
+
+ return io_write_buf(pair, buf->u8, size);
+}
diff --git a/src/xz/file_io.h b/src/xz/file_io.h
new file mode 100644
index 0000000..967da86
--- /dev/null
+++ b/src/xz/file_io.h
@@ -0,0 +1,129 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file file_io.h
+/// \brief I/O types and functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+// Some systems have suboptimal BUFSIZ. Use a bit bigger value on them.
+// We also need that IO_BUFFER_SIZE is a multiple of 8 (sizeof(uint64_t))
+#if BUFSIZ <= 1024
+# define IO_BUFFER_SIZE 8192
+#else
+# define IO_BUFFER_SIZE (BUFSIZ & ~7U)
+#endif
+
+
+/// is_sparse() accesses the buffer as uint64_t for maximum speed.
+/// Use an union to make sure that the buffer is properly aligned.
+typedef union {
+ uint8_t u8[IO_BUFFER_SIZE];
+ uint32_t u32[IO_BUFFER_SIZE / sizeof(uint32_t)];
+ uint64_t u64[IO_BUFFER_SIZE / sizeof(uint64_t)];
+} io_buf;
+
+
+typedef struct {
+ /// Name of the source filename (as given on the command line) or
+ /// pointer to static "(stdin)" when reading from standard input.
+ const char *src_name;
+
+ /// Destination filename converted from src_name or pointer to static
+ /// "(stdout)" when writing to standard output.
+ char *dest_name;
+
+ /// File descriptor of the source file
+ int src_fd;
+
+ /// File descriptor of the target file
+ int dest_fd;
+
+ /// True once end of the source file has been detected.
+ bool src_eof;
+
+ /// If true, we look for long chunks of zeros and try to create
+ /// a sparse file.
+ bool dest_try_sparse;
+
+ /// This is used only if dest_try_sparse is true. This holds the
+ /// number of zero bytes we haven't written out, because we plan
+ /// to make that byte range a sparse chunk.
+ off_t dest_pending_sparse;
+
+ /// Stat of the source file.
+ struct stat src_st;
+
+ /// Stat of the destination file.
+ struct stat dest_st;
+
+} file_pair;
+
+
+/// \brief Initialize the I/O module
+extern void io_init(void);
+
+
+/// \brief Disable creation of sparse files when decompressing
+extern void io_no_sparse(void);
+
+
+/// \brief Open the source file
+extern file_pair *io_open_src(const char *src_name);
+
+
+/// \brief Open the destination file
+extern bool io_open_dest(file_pair *pair);
+
+
+/// \brief Closes the file descriptors and frees possible allocated memory
+///
+/// The success argument determines if source or destination file gets
+/// unlinked:
+/// - false: The destination file is unlinked.
+/// - true: The source file is unlinked unless writing to stdout or --keep
+/// was used.
+extern void io_close(file_pair *pair, bool success);
+
+
+/// \brief Reads from the source file to a buffer
+///
+/// \param pair File pair having the source file open for reading
+/// \param buf Destination buffer to hold the read data
+/// \param size Size of the buffer; assumed be smaller than SSIZE_MAX
+///
+/// \return On success, number of bytes read is returned. On end of
+/// file zero is returned and pair->src_eof set to true.
+/// On error, SIZE_MAX is returned and error message printed.
+extern size_t io_read(file_pair *pair, io_buf *buf, size_t size);
+
+
+/// \brief Read from source file from given offset to a buffer
+///
+/// This is remotely similar to standard pread(). This uses lseek() though,
+/// so the read offset is changed on each call.
+///
+/// \param pair Seekable source file
+/// \param buf Destination buffer
+/// \param size Amount of data to read
+/// \param pos Offset relative to the beginning of the file,
+/// from which the data should be read.
+///
+/// \return On success, false is returned. On error, error message
+/// is printed and true is returned.
+extern bool io_pread(file_pair *pair, io_buf *buf, size_t size, off_t pos);
+
+
+/// \brief Writes a buffer to the destination file
+///
+/// \param pair File pair having the destination file open for writing
+/// \param buf Buffer containing the data to be written
+/// \param size Size of the buffer; assumed be smaller than SSIZE_MAX
+///
+/// \return On success, zero is returned. On error, -1 is returned
+/// and error message printed.
+extern bool io_write(file_pair *pair, const io_buf *buf, size_t size);
diff --git a/src/xz/hardware.c b/src/xz/hardware.c
new file mode 100644
index 0000000..74742fc
--- /dev/null
+++ b/src/xz/hardware.c
@@ -0,0 +1,112 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file hardware.c
+/// \brief Detection of available hardware resources
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include "tuklib_cpucores.h"
+
+
+/// Maximum number of free *coder* threads. This can be set with
+/// the --threads=NUM command line option.
+static uint32_t threadlimit;
+
+/// Memory usage limit
+static uint64_t memlimit;
+
+/// Total amount of physical RAM
+static uint64_t total_ram;
+
+
+extern void
+hardware_threadlimit_set(uint32_t new_threadlimit)
+{
+ if (new_threadlimit == 0) {
+ // The default is the number of available CPU cores.
+ threadlimit = tuklib_cpucores();
+ if (threadlimit == 0)
+ threadlimit = 1;
+ } else {
+ threadlimit = new_threadlimit;
+ }
+
+ return;
+}
+
+
+extern uint32_t
+hardware_threadlimit_get(void)
+{
+ return threadlimit;
+}
+
+
+extern void
+hardware_memlimit_set(uint64_t new_memlimit)
+{
+ if (new_memlimit != 0) {
+ memlimit = new_memlimit;
+ } else {
+ // The default depends on the amount of RAM but so that
+ // on "low-memory" systems the relative limit is higher
+ // to make it more likely that files created with "xz -9"
+ // will still decompress without overriding the limit
+ // manually.
+ //
+ // If 40 % of RAM is 80 MiB or more, use 40 % of RAM as
+ // the limit.
+ memlimit = 40 * total_ram / 100;
+ if (memlimit < UINT64_C(80) * 1024 * 1024) {
+ // If 80 % of RAM is less than 80 MiB,
+ // use 80 % of RAM as the limit.
+ memlimit = 80 * total_ram / 100;
+ if (memlimit > UINT64_C(80) * 1024 * 1024) {
+ // Otherwise use 80 MiB as the limit.
+ memlimit = UINT64_C(80) * 1024 * 1024;
+ }
+ }
+ }
+
+ return;
+}
+
+
+extern void
+hardware_memlimit_set_percentage(uint32_t percentage)
+{
+ assert(percentage > 0);
+ assert(percentage <= 100);
+
+ memlimit = percentage * total_ram / 100;
+ return;
+}
+
+
+extern uint64_t
+hardware_memlimit_get(void)
+{
+ return memlimit;
+}
+
+
+extern void
+hardware_init(void)
+{
+ // Get the amount of RAM. If we cannot determine it,
+ // use the assumption defined by the configure script.
+ total_ram = lzma_physmem();
+ if (total_ram == 0)
+ total_ram = (uint64_t)(ASSUME_RAM) * 1024 * 1024;
+
+ // Set the defaults.
+ hardware_memlimit_set(0);
+ hardware_threadlimit_set(0);
+ return;
+}
diff --git a/src/xz/hardware.h b/src/xz/hardware.h
new file mode 100644
index 0000000..b2cf34c
--- /dev/null
+++ b/src/xz/hardware.h
@@ -0,0 +1,35 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file hardware.h
+/// \brief Detection of available hardware resources
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// Initialize some hardware-specific variables, which are needed by other
+/// hardware_* functions.
+extern void hardware_init(void);
+
+
+/// Set custom value for maximum number of coder threads.
+extern void hardware_threadlimit_set(uint32_t threadlimit);
+
+/// Get the maximum number of coder threads. Some additional helper threads
+/// are allowed on top of this).
+extern uint32_t hardware_threadlimit_get(void);
+
+
+/// Set custom memory usage limit. This is used for both encoding and
+/// decoding. Zero indicates resetting the limit back to defaults.
+extern void hardware_memlimit_set(uint64_t memlimit);
+
+/// Set custom memory usage limit as a percentage of installed RAM.
+/// The percentage must be in the range [1, 100].
+extern void hardware_memlimit_set_percentage(uint32_t percentage);
+
+/// Get the current memory usage limit.
+extern uint64_t hardware_memlimit_get(void);
diff --git a/src/xz/list.c b/src/xz/list.c
new file mode 100644
index 0000000..dda7c9b
--- /dev/null
+++ b/src/xz/list.c
@@ -0,0 +1,1032 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file list.c
+/// \brief Listing information about .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include "tuklib_integer.h"
+
+
+/// Information about a .xz file
+typedef struct {
+ /// Combined Index of all Streams in the file
+ lzma_index *idx;
+
+ /// Total amount of Stream Padding
+ uint64_t stream_padding;
+
+ /// Highest memory usage so far
+ uint64_t memusage_max;
+
+ /// True if all Blocks so far have Compressed Size and
+ /// Uncompressed Size fields
+ bool all_have_sizes;
+
+} xz_file_info;
+
+#define XZ_FILE_INFO_INIT { NULL, 0, 0, true }
+
+
+/// Information about a .xz Block
+typedef struct {
+ /// Size of the Block Header
+ uint32_t header_size;
+
+ /// A few of the Block Flags as a string
+ char flags[3];
+
+ /// Size of the Compressed Data field in the Block
+ lzma_vli compressed_size;
+
+ /// Decoder memory usage for this Block
+ uint64_t memusage;
+
+ /// The filter chain of this Block in human-readable form
+ const char *filter_chain;
+
+} block_header_info;
+
+
+/// Check ID to string mapping
+static const char check_names[LZMA_CHECK_ID_MAX + 1][12] = {
+ "None",
+ "CRC32",
+ "Unknown-2",
+ "Unknown-3",
+ "CRC64",
+ "Unknown-5",
+ "Unknown-6",
+ "Unknown-7",
+ "Unknown-8",
+ "Unknown-9",
+ "SHA-256",
+ "Unknown-11",
+ "Unknown-12",
+ "Unknown-13",
+ "Unknown-14",
+ "Unknown-15",
+};
+
+
+/// Value of the Check field as hexadecimal string.
+/// This is set by parse_check_value().
+static char check_value[2 * LZMA_CHECK_SIZE_MAX + 1];
+
+
+/// Totals that are displayed if there was more than one file.
+/// The "files" counter is also used in print_info_adv() to show
+/// the file number.
+static struct {
+ uint64_t files;
+ uint64_t streams;
+ uint64_t blocks;
+ uint64_t compressed_size;
+ uint64_t uncompressed_size;
+ uint64_t stream_padding;
+ uint64_t memusage_max;
+ uint32_t checks;
+ bool all_have_sizes;
+} totals = { 0, 0, 0, 0, 0, 0, 0, 0, true };
+
+
+/// \brief Parse the Index(es) from the given .xz file
+///
+/// \param xfi Pointer to structure where the decoded information
+/// is stored.
+/// \param pair Input file
+///
+/// \return On success, false is returned. On error, true is returned.
+///
+// TODO: This function is pretty big. liblzma should have a function that
+// takes a callback function to parse the Index(es) from a .xz file to make
+// it easy for applications.
+static bool
+parse_indexes(xz_file_info *xfi, file_pair *pair)
+{
+ if (pair->src_st.st_size <= 0) {
+ message_error(_("%s: File is empty"), pair->src_name);
+ return true;
+ }
+
+ if (pair->src_st.st_size < 2 * LZMA_STREAM_HEADER_SIZE) {
+ message_error(_("%s: Too small to be a valid .xz file"),
+ pair->src_name);
+ return true;
+ }
+
+ io_buf buf;
+ lzma_stream_flags header_flags;
+ lzma_stream_flags footer_flags;
+ lzma_ret ret;
+
+ // lzma_stream for the Index decoder
+ lzma_stream strm = LZMA_STREAM_INIT;
+
+ // All Indexes decoded so far
+ lzma_index *combined_index = NULL;
+
+ // The Index currently being decoded
+ lzma_index *this_index = NULL;
+
+ // Current position in the file. We parse the file backwards so
+ // initialize it to point to the end of the file.
+ off_t pos = pair->src_st.st_size;
+
+ // Each loop iteration decodes one Index.
+ do {
+ // Check that there is enough data left to contain at least
+ // the Stream Header and Stream Footer. This check cannot
+ // fail in the first pass of this loop.
+ if (pos < 2 * LZMA_STREAM_HEADER_SIZE) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(LZMA_DATA_ERROR));
+ goto error;
+ }
+
+ pos -= LZMA_STREAM_HEADER_SIZE;
+ lzma_vli stream_padding = 0;
+
+ // Locate the Stream Footer. There may be Stream Padding which
+ // we must skip when reading backwards.
+ while (true) {
+ if (pos < LZMA_STREAM_HEADER_SIZE) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(
+ LZMA_DATA_ERROR));
+ goto error;
+ }
+
+ if (io_pread(pair, &buf,
+ LZMA_STREAM_HEADER_SIZE, pos))
+ goto error;
+
+ // Stream Padding is always a multiple of four bytes.
+ int i = 2;
+ if (buf.u32[i] != 0)
+ break;
+
+ // To avoid calling io_pread() for every four bytes
+ // of Stream Padding, take advantage that we read
+ // 12 bytes (LZMA_STREAM_HEADER_SIZE) already and
+ // check them too before calling io_pread() again.
+ do {
+ stream_padding += 4;
+ pos -= 4;
+ --i;
+ } while (i >= 0 && buf.u32[i] == 0);
+ }
+
+ // Decode the Stream Footer.
+ ret = lzma_stream_footer_decode(&footer_flags, buf.u8);
+ if (ret != LZMA_OK) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+ goto error;
+ }
+
+ // Check that the size of the Index field looks sane.
+ lzma_vli index_size = footer_flags.backward_size;
+ if ((lzma_vli)(pos) < index_size + LZMA_STREAM_HEADER_SIZE) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(LZMA_DATA_ERROR));
+ goto error;
+ }
+
+ // Set pos to the beginning of the Index.
+ pos -= index_size;
+
+ // See how much memory we can use for decoding this Index.
+ uint64_t memlimit = hardware_memlimit_get();
+ uint64_t memused = 0;
+ if (combined_index != NULL) {
+ memused = lzma_index_memused(combined_index);
+ if (memused > memlimit)
+ message_bug();
+
+ memlimit -= memused;
+ }
+
+ // Decode the Index.
+ ret = lzma_index_decoder(&strm, &this_index, memlimit);
+ if (ret != LZMA_OK) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+ goto error;
+ }
+
+ do {
+ // Don't give the decoder more input than the
+ // Index size.
+ strm.avail_in = my_min(IO_BUFFER_SIZE, index_size);
+ if (io_pread(pair, &buf, strm.avail_in, pos))
+ goto error;
+
+ pos += strm.avail_in;
+ index_size -= strm.avail_in;
+
+ strm.next_in = buf.u8;
+ ret = lzma_code(&strm, LZMA_RUN);
+
+ } while (ret == LZMA_OK);
+
+ // If the decoding seems to be successful, check also that
+ // the Index decoder consumed as much input as indicated
+ // by the Backward Size field.
+ if (ret == LZMA_STREAM_END)
+ if (index_size != 0 || strm.avail_in != 0)
+ ret = LZMA_DATA_ERROR;
+
+ if (ret != LZMA_STREAM_END) {
+ // LZMA_BUFFER_ERROR means that the Index decoder
+ // would have liked more input than what the Index
+ // size should be according to Stream Footer.
+ // The message for LZMA_DATA_ERROR makes more
+ // sense in that case.
+ if (ret == LZMA_BUF_ERROR)
+ ret = LZMA_DATA_ERROR;
+
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+
+ // If the error was too low memory usage limit,
+ // show also how much memory would have been needed.
+ if (ret == LZMA_MEMLIMIT_ERROR) {
+ uint64_t needed = lzma_memusage(&strm);
+ if (UINT64_MAX - needed < memused)
+ needed = UINT64_MAX;
+ else
+ needed += memused;
+
+ message_mem_needed(V_ERROR, needed);
+ }
+
+ goto error;
+ }
+
+ // Decode the Stream Header and check that its Stream Flags
+ // match the Stream Footer.
+ pos -= footer_flags.backward_size + LZMA_STREAM_HEADER_SIZE;
+ if ((lzma_vli)(pos) < lzma_index_total_size(this_index)) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(LZMA_DATA_ERROR));
+ goto error;
+ }
+
+ pos -= lzma_index_total_size(this_index);
+ if (io_pread(pair, &buf, LZMA_STREAM_HEADER_SIZE, pos))
+ goto error;
+
+ ret = lzma_stream_header_decode(&header_flags, buf.u8);
+ if (ret != LZMA_OK) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+ goto error;
+ }
+
+ ret = lzma_stream_flags_compare(&header_flags, &footer_flags);
+ if (ret != LZMA_OK) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+ goto error;
+ }
+
+ // Store the decoded Stream Flags into this_index. This is
+ // needed so that we can print which Check is used in each
+ // Stream.
+ ret = lzma_index_stream_flags(this_index, &footer_flags);
+ if (ret != LZMA_OK)
+ message_bug();
+
+ // Store also the size of the Stream Padding field. It is
+ // needed to show the offsets of the Streams correctly.
+ ret = lzma_index_stream_padding(this_index, stream_padding);
+ if (ret != LZMA_OK)
+ message_bug();
+
+ if (combined_index != NULL) {
+ // Append the earlier decoded Indexes
+ // after this_index.
+ ret = lzma_index_cat(
+ this_index, combined_index, NULL);
+ if (ret != LZMA_OK) {
+ message_error("%s: %s", pair->src_name,
+ message_strm(ret));
+ goto error;
+ }
+ }
+
+ combined_index = this_index;
+ this_index = NULL;
+
+ xfi->stream_padding += stream_padding;
+
+ } while (pos > 0);
+
+ lzma_end(&strm);
+
+ // All OK. Make combined_index available to the caller.
+ xfi->idx = combined_index;
+ return false;
+
+error:
+ // Something went wrong, free the allocated memory.
+ lzma_end(&strm);
+ lzma_index_end(combined_index, NULL);
+ lzma_index_end(this_index, NULL);
+ return true;
+}
+
+
+/// \brief Parse the Block Header
+///
+/// The result is stored into *bhi. The caller takes care of initializing it.
+///
+/// \return False on success, true on error.
+static bool
+parse_block_header(file_pair *pair, const lzma_index_iter *iter,
+ block_header_info *bhi, xz_file_info *xfi)
+{
+#if IO_BUFFER_SIZE < LZMA_BLOCK_HEADER_SIZE_MAX
+# error IO_BUFFER_SIZE < LZMA_BLOCK_HEADER_SIZE_MAX
+#endif
+
+ // Get the whole Block Header with one read, but don't read past
+ // the end of the Block (or even its Check field).
+ const uint32_t size = my_min(iter->block.total_size
+ - lzma_check_size(iter->stream.flags->check),
+ LZMA_BLOCK_HEADER_SIZE_MAX);
+ io_buf buf;
+ if (io_pread(pair, &buf, size, iter->block.compressed_file_offset))
+ return true;
+
+ // Zero would mean Index Indicator and thus not a valid Block.
+ if (buf.u8[0] == 0)
+ goto data_error;
+
+ lzma_block block;
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+
+ // Initialize the pointers so that they can be passed to free().
+ for (size_t i = 0; i < ARRAY_SIZE(filters); ++i)
+ filters[i].options = NULL;
+
+ // Initialize the block structure and decode Block Header Size.
+ block.version = 0;
+ block.check = iter->stream.flags->check;
+ block.filters = filters;
+
+ block.header_size = lzma_block_header_size_decode(buf.u8[0]);
+ if (block.header_size > size)
+ goto data_error;
+
+ // Decode the Block Header.
+ switch (lzma_block_header_decode(&block, NULL, buf.u8)) {
+ case LZMA_OK:
+ break;
+
+ case LZMA_OPTIONS_ERROR:
+ message_error("%s: %s", pair->src_name,
+ message_strm(LZMA_OPTIONS_ERROR));
+ return true;
+
+ case LZMA_DATA_ERROR:
+ goto data_error;
+
+ default:
+ message_bug();
+ }
+
+ // Check the Block Flags. These must be done before calling
+ // lzma_block_compressed_size(), because it overwrites
+ // block.compressed_size.
+ bhi->flags[0] = block.compressed_size != LZMA_VLI_UNKNOWN
+ ? 'c' : '-';
+ bhi->flags[1] = block.uncompressed_size != LZMA_VLI_UNKNOWN
+ ? 'u' : '-';
+ bhi->flags[2] = '\0';
+
+ // Collect information if all Blocks have both Compressed Size
+ // and Uncompressed Size fields. They can be useful e.g. for
+ // multi-threaded decompression so it can be useful to know it.
+ xfi->all_have_sizes &= block.compressed_size != LZMA_VLI_UNKNOWN
+ && block.uncompressed_size != LZMA_VLI_UNKNOWN;
+
+ // Validate or set block.compressed_size.
+ switch (lzma_block_compressed_size(&block,
+ iter->block.unpadded_size)) {
+ case LZMA_OK:
+ break;
+
+ case LZMA_DATA_ERROR:
+ goto data_error;
+
+ default:
+ message_bug();
+ }
+
+ // Copy the known sizes.
+ bhi->header_size = block.header_size;
+ bhi->compressed_size = block.compressed_size;
+
+ // Calculate the decoder memory usage and update the maximum
+ // memory usage of this Block.
+ bhi->memusage = lzma_raw_decoder_memusage(filters);
+ if (xfi->memusage_max < bhi->memusage)
+ xfi->memusage_max = bhi->memusage;
+
+ // Convert the filter chain to human readable form.
+ bhi->filter_chain = message_filters_to_str(filters, false);
+
+ // Free the memory allocated by lzma_block_header_decode().
+ for (size_t i = 0; filters[i].id != LZMA_VLI_UNKNOWN; ++i)
+ free(filters[i].options);
+
+ return false;
+
+data_error:
+ // Show the error message.
+ message_error("%s: %s", pair->src_name,
+ message_strm(LZMA_DATA_ERROR));
+
+ // Free the memory allocated by lzma_block_header_decode().
+ // This is truly needed only if we get here after a succcessful
+ // call to lzma_block_header_decode() but it doesn't hurt to
+ // always do it.
+ for (size_t i = 0; filters[i].id != LZMA_VLI_UNKNOWN; ++i)
+ free(filters[i].options);
+
+ return true;
+}
+
+
+/// \brief Parse the Check field and put it into check_value[]
+///
+/// \return False on success, true on error.
+static bool
+parse_check_value(file_pair *pair, const lzma_index_iter *iter)
+{
+ // Don't read anything from the file if there is no integrity Check.
+ if (iter->stream.flags->check == LZMA_CHECK_NONE) {
+ snprintf(check_value, sizeof(check_value), "---");
+ return false;
+ }
+
+ // Locate and read the Check field.
+ const uint32_t size = lzma_check_size(iter->stream.flags->check);
+ const off_t offset = iter->block.compressed_file_offset
+ + iter->block.total_size - size;
+ io_buf buf;
+ if (io_pread(pair, &buf, size, offset))
+ return true;
+
+ // CRC32 and CRC64 are in little endian. Guess that all the future
+ // 32-bit and 64-bit Check values are little endian too. It shouldn't
+ // be a too big problem if this guess is wrong.
+ if (size == 4)
+ snprintf(check_value, sizeof(check_value),
+ "%08" PRIx32, conv32le(buf.u32[0]));
+ else if (size == 8)
+ snprintf(check_value, sizeof(check_value),
+ "%016" PRIx64, conv64le(buf.u64[0]));
+ else
+ for (size_t i = 0; i < size; ++i)
+ snprintf(check_value + i * 2, 3, "%02x", buf.u8[i]);
+
+ return false;
+}
+
+
+/// \brief Parse detailed information about a Block
+///
+/// Since this requires seek(s), listing information about all Blocks can
+/// be slow.
+///
+/// \param pair Input file
+/// \param iter Location of the Block whose Check value should
+/// be printed.
+/// \param bhi Pointer to structure where to store the information
+/// about the Block Header field.
+///
+/// \return False on success, true on error. If an error occurs,
+/// the error message is printed too so the caller doesn't
+/// need to worry about that.
+static bool
+parse_details(file_pair *pair, const lzma_index_iter *iter,
+ block_header_info *bhi, xz_file_info *xfi)
+{
+ if (parse_block_header(pair, iter, bhi, xfi))
+ return true;
+
+ if (parse_check_value(pair, iter))
+ return true;
+
+ return false;
+}
+
+
+/// \brief Get the compression ratio
+///
+/// This has slightly different format than that is used by in message.c.
+static const char *
+get_ratio(uint64_t compressed_size, uint64_t uncompressed_size)
+{
+ if (uncompressed_size == 0)
+ return "---";
+
+ const double ratio = (double)(compressed_size)
+ / (double)(uncompressed_size);
+ if (ratio > 9.999)
+ return "---";
+
+ static char buf[6];
+ snprintf(buf, sizeof(buf), "%.3f", ratio);
+ return buf;
+}
+
+
+/// \brief Get a comma-separated list of Check names
+///
+/// \param checks Bit mask of Checks to print
+/// \param space_after_comma
+/// It's better to not use spaces in table-like listings,
+/// but in more verbose formats a space after a comma
+/// is good for readability.
+static const char *
+get_check_names(uint32_t checks, bool space_after_comma)
+{
+ assert(checks != 0);
+
+ static char buf[sizeof(check_names)];
+ char *pos = buf;
+ size_t left = sizeof(buf);
+
+ const char *sep = space_after_comma ? ", " : ",";
+ bool comma = false;
+
+ for (size_t i = 0; i <= LZMA_CHECK_ID_MAX; ++i) {
+ if (checks & (UINT32_C(1) << i)) {
+ my_snprintf(&pos, &left, "%s%s",
+ comma ? sep : "", check_names[i]);
+ comma = true;
+ }
+ }
+
+ return buf;
+}
+
+
+static bool
+print_info_basic(const xz_file_info *xfi, file_pair *pair)
+{
+ static bool headings_displayed = false;
+ if (!headings_displayed) {
+ headings_displayed = true;
+ // TRANSLATORS: These are column titles. From Strms (Streams)
+ // to Ratio, the columns are right aligned. Check and Filename
+ // are left aligned. If you need longer words, it's OK to
+ // use two lines here. Test with xz --list.
+ puts(_("Strms Blocks Compressed Uncompressed Ratio "
+ "Check Filename"));
+ }
+
+ printf("%5s %7s %11s %11s %5s %-7s %s\n",
+ uint64_to_str(lzma_index_stream_count(xfi->idx), 0),
+ uint64_to_str(lzma_index_block_count(xfi->idx), 1),
+ uint64_to_nicestr(lzma_index_file_size(xfi->idx),
+ NICESTR_B, NICESTR_TIB, false, 2),
+ uint64_to_nicestr(
+ lzma_index_uncompressed_size(xfi->idx),
+ NICESTR_B, NICESTR_TIB, false, 3),
+ get_ratio(lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx)),
+ get_check_names(lzma_index_checks(xfi->idx), false),
+ pair->src_name);
+
+ return false;
+}
+
+
+static void
+print_adv_helper(uint64_t stream_count, uint64_t block_count,
+ uint64_t compressed_size, uint64_t uncompressed_size,
+ uint32_t checks, uint64_t stream_padding)
+{
+ printf(_(" Streams: %s\n"),
+ uint64_to_str(stream_count, 0));
+ printf(_(" Blocks: %s\n"),
+ uint64_to_str(block_count, 0));
+ printf(_(" Compressed size: %s\n"),
+ uint64_to_nicestr(compressed_size,
+ NICESTR_B, NICESTR_TIB, true, 0));
+ printf(_(" Uncompressed size: %s\n"),
+ uint64_to_nicestr(uncompressed_size,
+ NICESTR_B, NICESTR_TIB, true, 0));
+ printf(_(" Ratio: %s\n"),
+ get_ratio(compressed_size, uncompressed_size));
+ printf(_(" Check: %s\n"),
+ get_check_names(checks, true));
+ printf(_(" Stream padding: %s\n"),
+ uint64_to_nicestr(stream_padding,
+ NICESTR_B, NICESTR_TIB, true, 0));
+ return;
+}
+
+
+static bool
+print_info_adv(xz_file_info *xfi, file_pair *pair)
+{
+ // Print the overall information.
+ print_adv_helper(lzma_index_stream_count(xfi->idx),
+ lzma_index_block_count(xfi->idx),
+ lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx),
+ lzma_index_checks(xfi->idx),
+ xfi->stream_padding);
+
+ // Size of the biggest Check. This is used to calculate the width
+ // of the CheckVal field. The table would get insanely wide if
+ // we always reserved space for 64-byte Check (128 chars as hex).
+ uint32_t check_max = 0;
+
+ // Print information about the Streams.
+ puts(_(" Streams:\n Stream Blocks"
+ " CompOffset UncompOffset"
+ " CompSize UncompSize Ratio"
+ " Check Padding"));
+
+ lzma_index_iter iter;
+ lzma_index_iter_init(&iter, xfi->idx);
+
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_STREAM)) {
+ printf(" %6s %9s %15s %15s ",
+ uint64_to_str(iter.stream.number, 0),
+ uint64_to_str(iter.stream.block_count, 1),
+ uint64_to_str(
+ iter.stream.compressed_offset, 2),
+ uint64_to_str(
+ iter.stream.uncompressed_offset, 3));
+ printf("%15s %15s %5s %-10s %7s\n",
+ uint64_to_str(iter.stream.compressed_size, 0),
+ uint64_to_str(
+ iter.stream.uncompressed_size, 1),
+ get_ratio(iter.stream.compressed_size,
+ iter.stream.uncompressed_size),
+ check_names[iter.stream.flags->check],
+ uint64_to_str(iter.stream.padding, 2));
+
+ // Update the maximum Check size.
+ if (lzma_check_size(iter.stream.flags->check) > check_max)
+ check_max = lzma_check_size(iter.stream.flags->check);
+ }
+
+ // Cache the verbosity level to a local variable.
+ const bool detailed = message_verbosity_get() >= V_DEBUG;
+
+ // Information collected from Block Headers
+ block_header_info bhi;
+
+ // Print information about the Blocks but only if there is
+ // at least one Block.
+ if (lzma_index_block_count(xfi->idx) > 0) {
+ // Calculate the width of the CheckVal field.
+ const int checkval_width = my_max(8, 2 * check_max);
+
+ // Print the headings.
+ printf(_(" Blocks:\n Stream Block"
+ " CompOffset UncompOffset"
+ " TotalSize UncompSize Ratio Check"));
+
+ if (detailed)
+ printf(_(" %-*s Header Flags CompSize"
+ " MemUsage Filters"),
+ checkval_width, _("CheckVal"));
+
+ putchar('\n');
+
+ lzma_index_iter_init(&iter, xfi->idx);
+
+ // Iterate over the Blocks.
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_BLOCK)) {
+ if (detailed && parse_details(pair, &iter, &bhi, xfi))
+ return true;
+
+ printf(" %6s %9s %15s %15s ",
+ uint64_to_str(iter.stream.number, 0),
+ uint64_to_str(
+ iter.block.number_in_stream, 1),
+ uint64_to_str(
+ iter.block.compressed_file_offset, 2),
+ uint64_to_str(
+ iter.block.uncompressed_file_offset,
+ 3));
+ printf("%15s %15s %5s %-*s",
+ uint64_to_str(iter.block.total_size, 0),
+ uint64_to_str(iter.block.uncompressed_size,
+ 1),
+ get_ratio(iter.block.total_size,
+ iter.block.uncompressed_size),
+ detailed ? 11 : 1,
+ check_names[iter.stream.flags->check]);
+
+ if (detailed) {
+ // Show MiB for memory usage, because it
+ // is the only size which is not in bytes.
+ const lzma_vli compressed_size
+ = iter.block.unpadded_size
+ - bhi.header_size
+ - lzma_check_size(
+ iter.stream.flags->check);
+ printf("%-*s %6s %-5s %15s %7s MiB %s",
+ checkval_width, check_value,
+ uint64_to_str(bhi.header_size, 0),
+ bhi.flags,
+ uint64_to_str(compressed_size, 1),
+ uint64_to_str(
+ round_up_to_mib(bhi.memusage),
+ 2),
+ bhi.filter_chain);
+ }
+
+ putchar('\n');
+ }
+ }
+
+ if (detailed) {
+ printf(_(" Memory needed: %s MiB\n"), uint64_to_str(
+ round_up_to_mib(xfi->memusage_max), 0));
+ printf(_(" Sizes in headers: %s\n"),
+ xfi->all_have_sizes ? _("Yes") : _("No"));
+ }
+
+ return false;
+}
+
+
+static bool
+print_info_robot(xz_file_info *xfi, file_pair *pair)
+{
+ printf("name\t%s\n", pair->src_name);
+
+ printf("file\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%s\t%s\t%" PRIu64 "\n",
+ lzma_index_stream_count(xfi->idx),
+ lzma_index_block_count(xfi->idx),
+ lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx),
+ get_ratio(lzma_index_file_size(xfi->idx),
+ lzma_index_uncompressed_size(xfi->idx)),
+ get_check_names(lzma_index_checks(xfi->idx), false),
+ xfi->stream_padding);
+
+ if (message_verbosity_get() >= V_VERBOSE) {
+ lzma_index_iter iter;
+ lzma_index_iter_init(&iter, xfi->idx);
+
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_STREAM))
+ printf("stream\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%s\t%s\t%" PRIu64 "\n",
+ iter.stream.number,
+ iter.stream.block_count,
+ iter.stream.compressed_offset,
+ iter.stream.uncompressed_offset,
+ iter.stream.compressed_size,
+ iter.stream.uncompressed_size,
+ get_ratio(iter.stream.compressed_size,
+ iter.stream.uncompressed_size),
+ check_names[iter.stream.flags->check],
+ iter.stream.padding);
+
+ lzma_index_iter_rewind(&iter);
+ block_header_info bhi;
+
+ while (!lzma_index_iter_next(&iter, LZMA_INDEX_ITER_BLOCK)) {
+ if (message_verbosity_get() >= V_DEBUG
+ && parse_details(
+ pair, &iter, &bhi, xfi))
+ return true;
+
+ printf("block\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu64
+ "\t%" PRIu64 "\t%" PRIu64 "\t%s\t%s",
+ iter.stream.number,
+ iter.block.number_in_stream,
+ iter.block.number_in_file,
+ iter.block.compressed_file_offset,
+ iter.block.uncompressed_file_offset,
+ iter.block.total_size,
+ iter.block.uncompressed_size,
+ get_ratio(iter.block.total_size,
+ iter.block.uncompressed_size),
+ check_names[iter.stream.flags->check]);
+
+ if (message_verbosity_get() >= V_DEBUG)
+ printf("\t%s\t%" PRIu32 "\t%s\t%" PRIu64
+ "\t%" PRIu64 "\t%s",
+ check_value,
+ bhi.header_size,
+ bhi.flags,
+ bhi.compressed_size,
+ bhi.memusage,
+ bhi.filter_chain);
+
+ putchar('\n');
+ }
+ }
+
+ if (message_verbosity_get() >= V_DEBUG)
+ printf("summary\t%" PRIu64 "\t%s\n",
+ xfi->memusage_max,
+ xfi->all_have_sizes ? "yes" : "no");
+
+ return false;
+}
+
+
+static void
+update_totals(const xz_file_info *xfi)
+{
+ // TODO: Integer overflow checks
+ ++totals.files;
+ totals.streams += lzma_index_stream_count(xfi->idx);
+ totals.blocks += lzma_index_block_count(xfi->idx);
+ totals.compressed_size += lzma_index_file_size(xfi->idx);
+ totals.uncompressed_size += lzma_index_uncompressed_size(xfi->idx);
+ totals.stream_padding += xfi->stream_padding;
+ totals.checks |= lzma_index_checks(xfi->idx);
+
+ if (totals.memusage_max < xfi->memusage_max)
+ totals.memusage_max = xfi->memusage_max;
+
+ totals.all_have_sizes &= xfi->all_have_sizes;
+
+ return;
+}
+
+
+static void
+print_totals_basic(void)
+{
+ // Print a separator line.
+ char line[80];
+ memset(line, '-', sizeof(line));
+ line[sizeof(line) - 1] = '\0';
+ puts(line);
+
+ // Print the totals except the file count, which needs
+ // special handling.
+ printf("%5s %7s %11s %11s %5s %-7s ",
+ uint64_to_str(totals.streams, 0),
+ uint64_to_str(totals.blocks, 1),
+ uint64_to_nicestr(totals.compressed_size,
+ NICESTR_B, NICESTR_TIB, false, 2),
+ uint64_to_nicestr(totals.uncompressed_size,
+ NICESTR_B, NICESTR_TIB, false, 3),
+ get_ratio(totals.compressed_size,
+ totals.uncompressed_size),
+ get_check_names(totals.checks, false));
+
+ // Since we print totals only when there are at least two files,
+ // the English message will always use "%s files". But some other
+ // languages need different forms for different plurals so we
+ // have to translate this string still.
+ //
+ // TRANSLATORS: This simply indicates the number of files shown
+ // by --list even though the format string uses %s.
+ printf(N_("%s file", "%s files\n",
+ totals.files <= ULONG_MAX ? totals.files
+ : (totals.files % 1000000) + 1000000),
+ uint64_to_str(totals.files, 0));
+
+ return;
+}
+
+
+static void
+print_totals_adv(void)
+{
+ putchar('\n');
+ puts(_("Totals:"));
+ printf(_(" Number of files: %s\n"),
+ uint64_to_str(totals.files, 0));
+ print_adv_helper(totals.streams, totals.blocks,
+ totals.compressed_size, totals.uncompressed_size,
+ totals.checks, totals.stream_padding);
+
+ if (message_verbosity_get() >= V_DEBUG) {
+ printf(_(" Memory needed: %s MiB\n"), uint64_to_str(
+ round_up_to_mib(totals.memusage_max), 0));
+ printf(_(" Sizes in headers: %s\n"),
+ totals.all_have_sizes ? _("Yes") : _("No"));
+ }
+
+ return;
+}
+
+
+static void
+print_totals_robot(void)
+{
+ printf("totals\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64 "\t%" PRIu64
+ "\t%s\t%s\t%" PRIu64 "\t%" PRIu64,
+ totals.streams,
+ totals.blocks,
+ totals.compressed_size,
+ totals.uncompressed_size,
+ get_ratio(totals.compressed_size,
+ totals.uncompressed_size),
+ get_check_names(totals.checks, false),
+ totals.stream_padding,
+ totals.files);
+
+ if (message_verbosity_get() >= V_DEBUG)
+ printf("\t%" PRIu64 "\t%s",
+ totals.memusage_max,
+ totals.all_have_sizes ? "yes" : "no");
+
+ putchar('\n');
+
+ return;
+}
+
+
+extern void
+list_totals(void)
+{
+ if (opt_robot) {
+ // Always print totals in --robot mode. It can be convenient
+ // in some cases and doesn't complicate usage of the
+ // single-file case much.
+ print_totals_robot();
+
+ } else if (totals.files > 1) {
+ // For non-robot mode, totals are printed only if there
+ // is more than one file.
+ if (message_verbosity_get() <= V_WARNING)
+ print_totals_basic();
+ else
+ print_totals_adv();
+ }
+
+ return;
+}
+
+
+extern void
+list_file(const char *filename)
+{
+ if (opt_format != FORMAT_XZ && opt_format != FORMAT_AUTO)
+ message_fatal(_("--list works only on .xz files "
+ "(--format=xz or --format=auto)"));
+
+ message_filename(filename);
+
+ if (filename == stdin_filename) {
+ message_error(_("--list does not support reading from "
+ "standard input"));
+ return;
+ }
+
+ // Unset opt_stdout so that io_open_src() won't accept special files.
+ // Set opt_force so that io_open_src() will follow symlinks.
+ opt_stdout = false;
+ opt_force = true;
+ file_pair *pair = io_open_src(filename);
+ if (pair == NULL)
+ return;
+
+ xz_file_info xfi = XZ_FILE_INFO_INIT;
+ if (!parse_indexes(&xfi, pair)) {
+ bool fail;
+
+ // We have three main modes:
+ // - --robot, which has submodes if --verbose is specified
+ // once or twice
+ // - Normal --list without --verbose
+ // - --list with one or two --verbose
+ if (opt_robot)
+ fail = print_info_robot(&xfi, pair);
+ else if (message_verbosity_get() <= V_WARNING)
+ fail = print_info_basic(&xfi, pair);
+ else
+ fail = print_info_adv(&xfi, pair);
+
+ // Update the totals that are displayed after all
+ // the individual files have been listed. Don't count
+ // broken files.
+ if (!fail)
+ update_totals(&xfi);
+
+ lzma_index_end(xfi.idx, NULL);
+ }
+
+ io_close(pair, false);
+ return;
+}
diff --git a/src/xz/list.h b/src/xz/list.h
new file mode 100644
index 0000000..a4c6ec7
--- /dev/null
+++ b/src/xz/list.h
@@ -0,0 +1,18 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file list.h
+/// \brief List information about .xz files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief List information about the given .xz file
+extern void list_file(const char *filename);
+
+
+/// \brief Show the totals after all files have been listed
+extern void list_totals(void);
diff --git a/src/xz/main.c b/src/xz/main.c
new file mode 100644
index 0000000..8196c6e
--- /dev/null
+++ b/src/xz/main.c
@@ -0,0 +1,302 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file main.c
+/// \brief main()
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include <ctype.h>
+
+/// Exit status to use. This can be changed with set_exit_status().
+static enum exit_status_type exit_status = E_SUCCESS;
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+/// exit_status has to be protected with a critical section due to
+/// how "signal handling" is done on Windows. See signals.c for details.
+static CRITICAL_SECTION exit_status_cs;
+#endif
+
+/// True if --no-warn is specified. When this is true, we don't set
+/// the exit status to E_WARNING when something worth a warning happens.
+static bool no_warn = false;
+
+
+extern void
+set_exit_status(enum exit_status_type new_status)
+{
+ assert(new_status == E_WARNING || new_status == E_ERROR);
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ EnterCriticalSection(&exit_status_cs);
+#endif
+
+ if (exit_status != E_ERROR)
+ exit_status = new_status;
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ LeaveCriticalSection(&exit_status_cs);
+#endif
+
+ return;
+}
+
+
+extern void
+set_exit_no_warn(void)
+{
+ no_warn = true;
+ return;
+}
+
+
+static const char *
+read_name(const args_info *args)
+{
+ // FIXME: Maybe we should have some kind of memory usage limit here
+ // like the tool has for the actual compression and decompression.
+ // Giving some huge text file with --files0 makes us to read the
+ // whole file in RAM.
+ static char *name = NULL;
+ static size_t size = 256;
+
+ // Allocate the initial buffer. This is never freed, since after it
+ // is no longer needed, the program exits very soon. It is safe to
+ // use xmalloc() and xrealloc() in this function, because while
+ // executing this function, no files are open for writing, and thus
+ // there's no need to cleanup anything before exiting.
+ if (name == NULL)
+ name = xmalloc(size);
+
+ // Write position in name
+ size_t pos = 0;
+
+ // Read one character at a time into name.
+ while (!user_abort) {
+ const int c = fgetc(args->files_file);
+
+ if (ferror(args->files_file)) {
+ // Take care of EINTR since we have established
+ // the signal handlers already.
+ if (errno == EINTR)
+ continue;
+
+ message_error(_("%s: Error reading filenames: %s"),
+ args->files_name, strerror(errno));
+ return NULL;
+ }
+
+ if (feof(args->files_file)) {
+ if (pos != 0)
+ message_error(_("%s: Unexpected end of input "
+ "when reading filenames"),
+ args->files_name);
+
+ return NULL;
+ }
+
+ if (c == args->files_delim) {
+ // We allow consecutive newline (--files) or '\0'
+ // characters (--files0), and ignore such empty
+ // filenames.
+ if (pos == 0)
+ continue;
+
+ // A non-empty name was read. Terminate it with '\0'
+ // and return it.
+ name[pos] = '\0';
+ return name;
+ }
+
+ if (c == '\0') {
+ // A null character was found when using --files,
+ // which expects plain text input separated with
+ // newlines.
+ message_error(_("%s: Null character found when "
+ "reading filenames; maybe you meant "
+ "to use `--files0' instead "
+ "of `--files'?"), args->files_name);
+ return NULL;
+ }
+
+ name[pos++] = c;
+
+ // Allocate more memory if needed. There must always be space
+ // at least for one character to allow terminating the string
+ // with '\0'.
+ if (pos == size) {
+ size *= 2;
+ name = xrealloc(name, size);
+ }
+ }
+
+ return NULL;
+}
+
+
+int
+main(int argc, char **argv)
+{
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ InitializeCriticalSection(&exit_status_cs);
+#endif
+
+ // Set up the progname variable.
+ tuklib_progname_init(argv);
+
+ // Initialize the file I/O. This makes sure that
+ // stdin, stdout, and stderr are something valid.
+ io_init();
+
+ // Set up the locale and message translations.
+ tuklib_gettext_init(PACKAGE, LOCALEDIR);
+
+ // Initialize handling of error/warning/other messages.
+ message_init();
+
+ // Set hardware-dependent default values. These can be overriden
+ // on the command line, thus this must be done before args_parse().
+ hardware_init();
+
+ // Parse the command line arguments and get an array of filenames.
+ // This doesn't return if something is wrong with the command line
+ // arguments. If there are no arguments, one filename ("-") is still
+ // returned to indicate stdin.
+ args_info args;
+ args_parse(&args, argc, argv);
+
+ if (opt_mode != MODE_LIST && opt_robot)
+ message_fatal(_("Compression and decompression with --robot "
+ "are not supported yet."));
+
+ // Tell the message handling code how many input files there are if
+ // we know it. This way the progress indicator can show it.
+ if (args.files_name != NULL)
+ message_set_files(0);
+ else
+ message_set_files(args.arg_count);
+
+ // Refuse to write compressed data to standard output if it is
+ // a terminal.
+ if (opt_mode == MODE_COMPRESS) {
+ if (opt_stdout || (args.arg_count == 1
+ && strcmp(args.arg_names[0], "-") == 0)) {
+ if (is_tty_stdout()) {
+ message_try_help();
+ tuklib_exit(E_ERROR, E_ERROR, false);
+ }
+ }
+ }
+
+ // Set up the signal handlers. We don't need these before we
+ // start the actual action and not in --list mode, so this is
+ // done after parsing the command line arguments.
+ //
+ // It's good to keep signal handlers in normal compression and
+ // decompression modes even when only writing to stdout, because
+ // we might need to restore O_APPEND flag on stdout before exiting.
+ // In --test mode, signal handlers aren't really needed, but let's
+ // keep them there for consistency with normal decompression.
+ if (opt_mode != MODE_LIST)
+ signals_init();
+
+ // coder_run() handles compression, decompression, and testing.
+ // list_file() is for --list.
+ void (*run)(const char *filename) = opt_mode == MODE_LIST
+ ? &list_file : &coder_run;
+
+ // Process the files given on the command line. Note that if no names
+ // were given, args_parse() gave us a fake "-" filename.
+ for (size_t i = 0; i < args.arg_count && !user_abort; ++i) {
+ if (strcmp("-", args.arg_names[i]) == 0) {
+ // Processing from stdin to stdout. Check that we
+ // aren't writing compressed data to a terminal or
+ // reading it from a terminal.
+ if (opt_mode == MODE_COMPRESS) {
+ if (is_tty_stdout())
+ continue;
+ } else if (is_tty_stdin()) {
+ continue;
+ }
+
+ // It doesn't make sense to compress data from stdin
+ // if we are supposed to read filenames from stdin
+ // too (enabled with --files or --files0).
+ if (args.files_name == stdin_filename) {
+ message_error(_("Cannot read data from "
+ "standard input when "
+ "reading filenames "
+ "from standard input"));
+ continue;
+ }
+
+ // Replace the "-" with a special pointer, which is
+ // recognized by coder_run() and other things.
+ // This way error messages get a proper filename
+ // string and the code still knows that it is
+ // handling the special case of stdin.
+ args.arg_names[i] = (char *)stdin_filename;
+ }
+
+ // Do the actual compression or decompression.
+ run(args.arg_names[i]);
+ }
+
+ // If --files or --files0 was used, process the filenames from the
+ // given file or stdin. Note that here we don't consider "-" to
+ // indicate stdin like we do with the command line arguments.
+ if (args.files_name != NULL) {
+ // read_name() checks for user_abort so we don't need to
+ // check it as loop termination condition.
+ while (true) {
+ const char *name = read_name(&args);
+ if (name == NULL)
+ break;
+
+ // read_name() doesn't return empty names.
+ assert(name[0] != '\0');
+ run(name);
+ }
+
+ if (args.files_name != stdin_filename)
+ (void)fclose(args.files_file);
+ }
+
+ // All files have now been handled. If in --list mode, display
+ // the totals before exiting. We don't have signal handlers
+ // enabled in --list mode, so we don't need to check user_abort.
+ if (opt_mode == MODE_LIST) {
+ assert(!user_abort);
+ list_totals();
+ }
+
+ // If we have got a signal, raise it to kill the program instead
+ // of calling tuklib_exit().
+ signals_exit();
+
+ // Make a local copy of exit_status to keep the Windows code
+ // thread safe. At this point it is fine if we miss the user
+ // pressing C-c and don't set the exit_status to E_ERROR on
+ // Windows.
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ EnterCriticalSection(&exit_status_cs);
+#endif
+
+ enum exit_status_type es = exit_status;
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+ LeaveCriticalSection(&exit_status_cs);
+#endif
+
+ // Suppress the exit status indicating a warning if --no-warn
+ // was specified.
+ if (es == E_WARNING && no_warn)
+ es = E_SUCCESS;
+
+ tuklib_exit(es, E_ERROR, message_verbosity_get() != V_SILENT);
+}
diff --git a/src/xz/main.h b/src/xz/main.h
new file mode 100644
index 0000000..323f2f7
--- /dev/null
+++ b/src/xz/main.h
@@ -0,0 +1,30 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file main.h
+/// \brief Miscellaneous declarations
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// Possible exit status values. These are the same as used by gzip and bzip2.
+enum exit_status_type {
+ E_SUCCESS = 0,
+ E_ERROR = 1,
+ E_WARNING = 2,
+};
+
+
+/// Sets the exit status after a warning or error has occurred. If new_status
+/// is E_WARNING and the old exit status was already E_ERROR, the exit
+/// status is not changed.
+extern void set_exit_status(enum exit_status_type new_status);
+
+
+/// Use E_SUCCESS instead of E_WARNING if something worth a warning occurs
+/// but nothing worth an error has occurred. This is called when --no-warn
+/// is specified.
+extern void set_exit_no_warn(void);
diff --git a/src/xz/message.c b/src/xz/message.c
new file mode 100644
index 0000000..5044ea2
--- /dev/null
+++ b/src/xz/message.c
@@ -0,0 +1,1237 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file message.c
+/// \brief Printing messages
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+#ifdef HAVE_SYS_TIME_H
+# include <sys/time.h>
+#endif
+
+#include <stdarg.h>
+
+
+/// Number of the current file
+static unsigned int files_pos = 0;
+
+/// Total number of input files; zero if unknown.
+static unsigned int files_total;
+
+/// Verbosity level
+static enum message_verbosity verbosity = V_WARNING;
+
+/// Filename which we will print with the verbose messages
+static const char *filename;
+
+/// True once the a filename has been printed to stderr as part of progress
+/// message. If automatic progress updating isn't enabled, this becomes true
+/// after the first progress message has been printed due to user sending
+/// SIGINFO, SIGUSR1, or SIGALRM. Once this variable is true, we will print
+/// an empty line before the next filename to make the output more readable.
+static bool first_filename_printed = false;
+
+/// This is set to true when we have printed the current filename to stderr
+/// as part of a progress message. This variable is useful only if not
+/// updating progress automatically: if user sends many SIGINFO, SIGUSR1, or
+/// SIGALRM signals, we won't print the name of the same file multiple times.
+static bool current_filename_printed = false;
+
+/// True if we should print progress indicator and update it automatically
+/// if also verbose >= V_VERBOSE.
+static bool progress_automatic;
+
+/// True if message_progress_start() has been called but
+/// message_progress_end() hasn't been called yet.
+static bool progress_started = false;
+
+/// This is true when a progress message was printed and the cursor is still
+/// on the same line with the progress message. In that case, a newline has
+/// to be printed before any error messages.
+static bool progress_active = false;
+
+/// Pointer to lzma_stream used to do the encoding or decoding.
+static lzma_stream *progress_strm;
+
+/// Expected size of the input stream is needed to show completion percentage
+/// and estimate remaining time.
+static uint64_t expected_in_size;
+
+/// Time when we started processing the file
+static uint64_t start_time;
+
+
+// Use alarm() and SIGALRM when they are supported. This has two minor
+// advantages over the alternative of polling gettimeofday():
+// - It is possible for the user to send SIGINFO, SIGUSR1, or SIGALRM to
+// get intermediate progress information even when --verbose wasn't used
+// or stderr is not a terminal.
+// - alarm() + SIGALRM seems to have slightly less overhead than polling
+// gettimeofday().
+#ifdef SIGALRM
+
+/// The signal handler for SIGALRM sets this to true. It is set back to false
+/// once the progress message has been updated.
+static volatile sig_atomic_t progress_needs_updating = false;
+
+/// Signal handler for SIGALRM
+static void
+progress_signal_handler(int sig lzma_attribute((unused)))
+{
+ progress_needs_updating = true;
+ return;
+}
+
+#else
+
+/// This is true when progress message printing is wanted. Using the same
+/// variable name as above to avoid some ifdefs.
+static bool progress_needs_updating = false;
+
+/// Elapsed time when the next progress message update should be done.
+static uint64_t progress_next_update;
+
+#endif
+
+
+/// Get the current time as microseconds since epoch
+static uint64_t
+my_time(void)
+{
+ struct timeval tv;
+ gettimeofday(&tv, NULL);
+ return (uint64_t)(tv.tv_sec) * UINT64_C(1000000) + tv.tv_usec;
+}
+
+
+extern void
+message_init(void)
+{
+ // If --verbose is used, we use a progress indicator if and only
+ // if stderr is a terminal. If stderr is not a terminal, we print
+ // verbose information only after finishing the file. As a special
+ // exception, even if --verbose was not used, user can send SIGALRM
+ // to make us print progress information once without automatic
+ // updating.
+ progress_automatic = isatty(STDERR_FILENO);
+
+ // Commented out because COLUMNS is rarely exported to environment.
+ // Most users have at least 80 columns anyway, let's think something
+ // fancy here if enough people complain.
+/*
+ if (progress_automatic) {
+ // stderr is a terminal. Check the COLUMNS environment
+ // variable to see if the terminal is wide enough. If COLUMNS
+ // doesn't exist or it has some unparsable value, we assume
+ // that the terminal is wide enough.
+ const char *columns_str = getenv("COLUMNS");
+ if (columns_str != NULL) {
+ char *endptr;
+ const long columns = strtol(columns_str, &endptr, 10);
+ if (*endptr != '\0' || columns < 80)
+ progress_automatic = false;
+ }
+ }
+*/
+
+#ifdef SIGALRM
+ // DJGPP lacks SA_RESTART, but it shouldn't give EINTR
+ // in most places either.
+# if defined(__DJGPP__) && !defined(SA_RESTART)
+# define SA_RESTART 0
+# endif
+
+ // Establish the signal handlers which set a flag to tell us that
+ // progress info should be updated. Since these signals don't
+ // require any quick action, we set SA_RESTART. That way we don't
+ // need to block them either in signals_block() to keep stdio
+ // functions from getting EINTR.
+ static const int sigs[] = {
+ SIGALRM,
+#ifdef SIGINFO
+ SIGINFO,
+#endif
+#ifdef SIGUSR1
+ SIGUSR1,
+#endif
+ };
+
+ struct sigaction sa;
+ sigemptyset(&sa.sa_mask);
+ sa.sa_flags = SA_RESTART;
+ sa.sa_handler = &progress_signal_handler;
+
+ for (size_t i = 0; i < ARRAY_SIZE(sigs); ++i)
+ if (sigaction(sigs[i], &sa, NULL))
+ message_signal_handler();
+#endif
+
+ return;
+}
+
+
+extern void
+message_verbosity_increase(void)
+{
+ if (verbosity < V_DEBUG)
+ ++verbosity;
+
+ return;
+}
+
+
+extern void
+message_verbosity_decrease(void)
+{
+ if (verbosity > V_SILENT)
+ --verbosity;
+
+ return;
+}
+
+
+extern enum message_verbosity
+message_verbosity_get(void)
+{
+ return verbosity;
+}
+
+
+extern void
+message_set_files(unsigned int files)
+{
+ files_total = files;
+ return;
+}
+
+
+/// Prints the name of the current file if it hasn't been printed already,
+/// except if we are processing exactly one stream from stdin to stdout.
+/// I think it looks nicer to not print "(stdin)" when --verbose is used
+/// in a pipe and no other files are processed.
+static void
+print_filename(void)
+{
+ if (files_total != 1 || filename != stdin_filename) {
+ signals_block();
+
+ FILE *file = opt_mode == MODE_LIST ? stdout : stderr;
+
+ // If a file was already processed, put an empty line
+ // before the next filename to improve readability.
+ if (first_filename_printed)
+ fputc('\n', file);
+
+ first_filename_printed = true;
+ current_filename_printed = true;
+
+ // If we don't know how many files there will be due
+ // to usage of --files or --files0.
+ if (files_total == 0)
+ fprintf(file, "%s (%u)\n", filename,
+ files_pos);
+ else
+ fprintf(file, "%s (%u/%u)\n", filename,
+ files_pos, files_total);
+
+ signals_unblock();
+ }
+
+ return;
+}
+
+
+extern void
+message_filename(const char *src_name)
+{
+ // Start numbering the files starting from one.
+ ++files_pos;
+ filename = src_name;
+
+ if (verbosity >= V_VERBOSE
+ && (progress_automatic || opt_mode == MODE_LIST))
+ print_filename();
+ else
+ current_filename_printed = false;
+
+ return;
+}
+
+
+extern void
+message_progress_start(lzma_stream *strm, uint64_t in_size)
+{
+ // Store the pointer to the lzma_stream used to do the coding.
+ // It is needed to find out the position in the stream.
+ progress_strm = strm;
+
+ // Store the processing start time of the file and its expected size.
+ // If we aren't printing any statistics, then these are unused. But
+ // since it is possible that the user sends us a signal to show
+ // statistics, we need to have these available anyway.
+ start_time = my_time();
+ expected_in_size = in_size;
+
+ // Indicate that progress info may need to be printed before
+ // printing error messages.
+ progress_started = true;
+
+ // If progress indicator is wanted, print the filename and possibly
+ // the file count now.
+ if (verbosity >= V_VERBOSE && progress_automatic) {
+ // Start the timer to display the first progress message
+ // after one second. An alternative would be to show the
+ // first message almost immediately, but delaying by one
+ // second looks better to me, since extremely early
+ // progress info is pretty much useless.
+#ifdef SIGALRM
+ // First disable a possibly existing alarm.
+ alarm(0);
+ progress_needs_updating = false;
+ alarm(1);
+#else
+ progress_needs_updating = true;
+ progress_next_update = 1000000;
+#endif
+ }
+
+ return;
+}
+
+
+/// Make the string indicating completion percentage.
+static const char *
+progress_percentage(uint64_t in_pos)
+{
+ // If the size of the input file is unknown or the size told us is
+ // clearly wrong since we have processed more data than the alleged
+ // size of the file, show a static string indicating that we have
+ // no idea of the completion percentage.
+ if (expected_in_size == 0 || in_pos > expected_in_size)
+ return "--- %";
+
+ // Never show 100.0 % before we actually are finished.
+ double percentage = (double)(in_pos) / (double)(expected_in_size)
+ * 99.9;
+
+ static char buf[sizeof("99.9 %")];
+ snprintf(buf, sizeof(buf), "%.1f %%", percentage);
+
+ return buf;
+}
+
+
+/// Make the string containing the amount of input processed, amount of
+/// output produced, and the compression ratio.
+static const char *
+progress_sizes(uint64_t compressed_pos, uint64_t uncompressed_pos, bool final)
+{
+ // This is enough to hold sizes up to about 99 TiB if thousand
+ // separator is used, or about 1 PiB without thousand separator.
+ // After that the progress indicator will look a bit silly, since
+ // the compression ratio no longer fits with three decimal places.
+ static char buf[36];
+
+ char *pos = buf;
+ size_t left = sizeof(buf);
+
+ // Print the sizes. If this the final message, use more reasonable
+ // units than MiB if the file was small.
+ const enum nicestr_unit unit_min = final ? NICESTR_B : NICESTR_MIB;
+ my_snprintf(&pos, &left, "%s / %s",
+ uint64_to_nicestr(compressed_pos,
+ unit_min, NICESTR_TIB, false, 0),
+ uint64_to_nicestr(uncompressed_pos,
+ unit_min, NICESTR_TIB, false, 1));
+
+ // Avoid division by zero. If we cannot calculate the ratio, set
+ // it to some nice number greater than 10.0 so that it gets caught
+ // in the next if-clause.
+ const double ratio = uncompressed_pos > 0
+ ? (double)(compressed_pos) / (double)(uncompressed_pos)
+ : 16.0;
+
+ // If the ratio is very bad, just indicate that it is greater than
+ // 9.999. This way the length of the ratio field stays fixed.
+ if (ratio > 9.999)
+ snprintf(pos, left, " > %.3f", 9.999);
+ else
+ snprintf(pos, left, " = %.3f", ratio);
+
+ return buf;
+}
+
+
+/// Make the string containing the processing speed of uncompressed data.
+static const char *
+progress_speed(uint64_t uncompressed_pos, uint64_t elapsed)
+{
+ // Don't print the speed immediately, since the early values look
+ // somewhat random.
+ if (elapsed < 3000000)
+ return "";
+
+ static const char unit[][8] = {
+ "KiB/s",
+ "MiB/s",
+ "GiB/s",
+ };
+
+ size_t unit_index = 0;
+
+ // Calculate the speed as KiB/s.
+ double speed = (double)(uncompressed_pos)
+ / ((double)(elapsed) * (1024.0 / 1e6));
+
+ // Adjust the unit of the speed if needed.
+ while (speed > 999.0) {
+ speed /= 1024.0;
+ if (++unit_index == ARRAY_SIZE(unit))
+ return ""; // Way too fast ;-)
+ }
+
+ // Use decimal point only if the number is small. Examples:
+ // - 0.1 KiB/s
+ // - 9.9 KiB/s
+ // - 99 KiB/s
+ // - 999 KiB/s
+ static char buf[sizeof("999 GiB/s")];
+ snprintf(buf, sizeof(buf), "%.*f %s",
+ speed > 9.9 ? 0 : 1, speed, unit[unit_index]);
+ return buf;
+}
+
+
+/// Make a string indicating elapsed or remaining time. The format is either
+/// M:SS or H:MM:SS depending on if the time is an hour or more.
+static const char *
+progress_time(uint64_t useconds)
+{
+ // 9999 hours = 416 days
+ static char buf[sizeof("9999:59:59")];
+
+ uint32_t seconds = useconds / 1000000;
+
+ // Don't show anything if the time is zero or ridiculously big.
+ if (seconds == 0 || seconds > ((9999 * 60) + 59) * 60 + 59)
+ return "";
+
+ uint32_t minutes = seconds / 60;
+ seconds %= 60;
+
+ if (minutes >= 60) {
+ const uint32_t hours = minutes / 60;
+ minutes %= 60;
+ snprintf(buf, sizeof(buf),
+ "%" PRIu32 ":%02" PRIu32 ":%02" PRIu32,
+ hours, minutes, seconds);
+ } else {
+ snprintf(buf, sizeof(buf), "%" PRIu32 ":%02" PRIu32,
+ minutes, seconds);
+ }
+
+ return buf;
+}
+
+
+/// Return a string containing estimated remaining time when
+/// reasonably possible.
+static const char *
+progress_remaining(uint64_t in_pos, uint64_t elapsed)
+{
+ // Don't show the estimated remaining time when it wouldn't
+ // make sense:
+ // - Input size is unknown.
+ // - Input has grown bigger since we started (de)compressing.
+ // - We haven't processed much data yet, so estimate would be
+ // too inaccurate.
+ // - Only a few seconds has passed since we started (de)compressing,
+ // so estimate would be too inaccurate.
+ if (expected_in_size == 0 || in_pos > expected_in_size
+ || in_pos < (UINT64_C(1) << 19) || elapsed < 8000000)
+ return "";
+
+ // Calculate the estimate. Don't give an estimate of zero seconds,
+ // since it is possible that all the input has been already passed
+ // to the library, but there is still quite a bit of output pending.
+ uint32_t remaining = (double)(expected_in_size - in_pos)
+ * ((double)(elapsed) / 1e6) / (double)(in_pos);
+ if (remaining < 1)
+ remaining = 1;
+
+ static char buf[sizeof("9 h 55 min")];
+
+ // Select appropriate precision for the estimated remaining time.
+ if (remaining <= 10) {
+ // A maximum of 10 seconds remaining.
+ // Show the number of seconds as is.
+ snprintf(buf, sizeof(buf), "%" PRIu32 " s", remaining);
+
+ } else if (remaining <= 50) {
+ // A maximum of 50 seconds remaining.
+ // Round up to the next multiple of five seconds.
+ remaining = (remaining + 4) / 5 * 5;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " s", remaining);
+
+ } else if (remaining <= 590) {
+ // A maximum of 9 minutes and 50 seconds remaining.
+ // Round up to the next multiple of ten seconds.
+ remaining = (remaining + 9) / 10 * 10;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " min %" PRIu32 " s",
+ remaining / 60, remaining % 60);
+
+ } else if (remaining <= 59 * 60) {
+ // A maximum of 59 minutes remaining.
+ // Round up to the next multiple of a minute.
+ remaining = (remaining + 59) / 60;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " min", remaining);
+
+ } else if (remaining <= 9 * 3600 + 50 * 60) {
+ // A maximum of 9 hours and 50 minutes left.
+ // Round up to the next multiple of ten minutes.
+ remaining = (remaining + 599) / 600 * 10;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " h %" PRIu32 " min",
+ remaining / 60, remaining % 60);
+
+ } else if (remaining <= 23 * 3600) {
+ // A maximum of 23 hours remaining.
+ // Round up to the next multiple of an hour.
+ remaining = (remaining + 3599) / 3600;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " h", remaining);
+
+ } else if (remaining <= 9 * 24 * 3600 + 23 * 3600) {
+ // A maximum of 9 days and 23 hours remaining.
+ // Round up to the next multiple of an hour.
+ remaining = (remaining + 3599) / 3600;
+ snprintf(buf, sizeof(buf), "%" PRIu32 " d %" PRIu32 " h",
+ remaining / 24, remaining % 24);
+
+ } else if (remaining <= 999 * 24 * 3600) {
+ // A maximum of 999 days remaining. ;-)
+ // Round up to the next multiple of a day.
+ remaining = (remaining + 24 * 3600 - 1) / (24 * 3600);
+ snprintf(buf, sizeof(buf), "%" PRIu32 " d", remaining);
+
+ } else {
+ // The estimated remaining time is too big. Don't show it.
+ return "";
+ }
+
+ return buf;
+}
+
+
+/// Calculate the elapsed time as microseconds.
+static uint64_t
+progress_elapsed(void)
+{
+ return my_time() - start_time;
+}
+
+
+/// Get information about position in the stream. This is currently simple,
+/// but it will become more complicated once we have multithreading support.
+static void
+progress_pos(uint64_t *in_pos,
+ uint64_t *compressed_pos, uint64_t *uncompressed_pos)
+{
+ *in_pos = progress_strm->total_in;
+
+ if (opt_mode == MODE_COMPRESS) {
+ *compressed_pos = progress_strm->total_out;
+ *uncompressed_pos = progress_strm->total_in;
+ } else {
+ *compressed_pos = progress_strm->total_in;
+ *uncompressed_pos = progress_strm->total_out;
+ }
+
+ return;
+}
+
+
+extern void
+message_progress_update(void)
+{
+ if (!progress_needs_updating)
+ return;
+
+ // Calculate how long we have been processing this file.
+ const uint64_t elapsed = progress_elapsed();
+
+#ifndef SIGALRM
+ if (progress_next_update > elapsed)
+ return;
+
+ progress_next_update = elapsed + 1000000;
+#endif
+
+ // Get our current position in the stream.
+ uint64_t in_pos;
+ uint64_t compressed_pos;
+ uint64_t uncompressed_pos;
+ progress_pos(&in_pos, &compressed_pos, &uncompressed_pos);
+
+ // Block signals so that fprintf() doesn't get interrupted.
+ signals_block();
+
+ // Print the filename if it hasn't been printed yet.
+ if (!current_filename_printed)
+ print_filename();
+
+ // Print the actual progress message. The idea is that there is at
+ // least three spaces between the fields in typical situations, but
+ // even in rare situations there is at least one space.
+ fprintf(stderr, "\r %6s %35s %9s %10s %10s\r",
+ progress_percentage(in_pos),
+ progress_sizes(compressed_pos, uncompressed_pos, false),
+ progress_speed(uncompressed_pos, elapsed),
+ progress_time(elapsed),
+ progress_remaining(in_pos, elapsed));
+
+#ifdef SIGALRM
+ // Updating the progress info was finished. Reset
+ // progress_needs_updating to wait for the next SIGALRM.
+ //
+ // NOTE: This has to be done before alarm(1) or with (very) bad
+ // luck we could be setting this to false after the alarm has already
+ // been triggered.
+ progress_needs_updating = false;
+
+ if (verbosity >= V_VERBOSE && progress_automatic) {
+ // Mark that the progress indicator is active, so if an error
+ // occurs, the error message gets printed cleanly.
+ progress_active = true;
+
+ // Restart the timer so that progress_needs_updating gets
+ // set to true after about one second.
+ alarm(1);
+ } else {
+ // The progress message was printed because user had sent us
+ // SIGALRM. In this case, each progress message is printed
+ // on its own line.
+ fputc('\n', stderr);
+ }
+#else
+ // When SIGALRM isn't supported and we get here, it's always due to
+ // automatic progress update. We set progress_active here too like
+ // described above.
+ assert(verbosity >= V_VERBOSE);
+ assert(progress_automatic);
+ progress_active = true;
+#endif
+
+ signals_unblock();
+
+ return;
+}
+
+
+static void
+progress_flush(bool finished)
+{
+ if (!progress_started || verbosity < V_VERBOSE)
+ return;
+
+ uint64_t in_pos;
+ uint64_t compressed_pos;
+ uint64_t uncompressed_pos;
+ progress_pos(&in_pos, &compressed_pos, &uncompressed_pos);
+
+ // Avoid printing intermediate progress info if some error occurs
+ // in the beginning of the stream. (If something goes wrong later in
+ // the stream, it is sometimes useful to tell the user where the
+ // error approximately occurred, especially if the error occurs
+ // after a time-consuming operation.)
+ if (!finished && !progress_active
+ && (compressed_pos == 0 || uncompressed_pos == 0))
+ return;
+
+ progress_active = false;
+
+ const uint64_t elapsed = progress_elapsed();
+
+ signals_block();
+
+ // When using the auto-updating progress indicator, the final
+ // statistics are printed in the same format as the progress
+ // indicator itself.
+ if (progress_automatic) {
+ fprintf(stderr, "\r %6s %35s %9s %10s %10s\n",
+ finished ? "100 %" : progress_percentage(in_pos),
+ progress_sizes(compressed_pos, uncompressed_pos, true),
+ progress_speed(uncompressed_pos, elapsed),
+ progress_time(elapsed),
+ finished ? "" : progress_remaining(in_pos, elapsed));
+ } else {
+ // The filename is always printed.
+ fprintf(stderr, "%s: ", filename);
+
+ // Percentage is printed only if we didn't finish yet.
+ if (!finished) {
+ // Don't print the percentage when it isn't known
+ // (starts with a dash).
+ const char *percentage = progress_percentage(in_pos);
+ if (percentage[0] != '-')
+ fprintf(stderr, "%s, ", percentage);
+ }
+
+ // Size information is always printed.
+ fprintf(stderr, "%s", progress_sizes(
+ compressed_pos, uncompressed_pos, true));
+
+ // The speed and elapsed time aren't always shown.
+ const char *speed = progress_speed(uncompressed_pos, elapsed);
+ if (speed[0] != '\0')
+ fprintf(stderr, ", %s", speed);
+
+ const char *elapsed_str = progress_time(elapsed);
+ if (elapsed_str[0] != '\0')
+ fprintf(stderr, ", %s", elapsed_str);
+
+ fputc('\n', stderr);
+ }
+
+ signals_unblock();
+
+ return;
+}
+
+
+extern void
+message_progress_end(bool success)
+{
+ assert(progress_started);
+ progress_flush(success);
+ progress_started = false;
+ return;
+}
+
+
+static void
+vmessage(enum message_verbosity v, const char *fmt, va_list ap)
+{
+ if (v <= verbosity) {
+ signals_block();
+
+ progress_flush(false);
+
+ fprintf(stderr, "%s: ", progname);
+ vfprintf(stderr, fmt, ap);
+ fputc('\n', stderr);
+
+ signals_unblock();
+ }
+
+ return;
+}
+
+
+extern void
+message(enum message_verbosity v, const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(v, fmt, ap);
+ va_end(ap);
+ return;
+}
+
+
+extern void
+message_warning(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(V_WARNING, fmt, ap);
+ va_end(ap);
+
+ set_exit_status(E_WARNING);
+ return;
+}
+
+
+extern void
+message_error(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(V_ERROR, fmt, ap);
+ va_end(ap);
+
+ set_exit_status(E_ERROR);
+ return;
+}
+
+
+extern void
+message_fatal(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ vmessage(V_ERROR, fmt, ap);
+ va_end(ap);
+
+ tuklib_exit(E_ERROR, E_ERROR, false);
+}
+
+
+extern void
+message_bug(void)
+{
+ message_fatal(_("Internal error (bug)"));
+}
+
+
+extern void
+message_signal_handler(void)
+{
+ message_fatal(_("Cannot establish signal handlers"));
+}
+
+
+extern const char *
+message_strm(lzma_ret code)
+{
+ switch (code) {
+ case LZMA_NO_CHECK:
+ return _("No integrity check; not verifying file integrity");
+
+ case LZMA_UNSUPPORTED_CHECK:
+ return _("Unsupported type of integrity check; "
+ "not verifying file integrity");
+
+ case LZMA_MEM_ERROR:
+ return strerror(ENOMEM);
+
+ case LZMA_MEMLIMIT_ERROR:
+ return _("Memory usage limit reached");
+
+ case LZMA_FORMAT_ERROR:
+ return _("File format not recognized");
+
+ case LZMA_OPTIONS_ERROR:
+ return _("Unsupported options");
+
+ case LZMA_DATA_ERROR:
+ return _("Compressed data is corrupt");
+
+ case LZMA_BUF_ERROR:
+ return _("Unexpected end of input");
+
+ case LZMA_OK:
+ case LZMA_STREAM_END:
+ case LZMA_GET_CHECK:
+ case LZMA_PROG_ERROR:
+ return _("Internal error (bug)");
+ }
+
+ return NULL;
+}
+
+
+extern void
+message_mem_needed(enum message_verbosity v, uint64_t memusage)
+{
+ if (v > verbosity)
+ return;
+
+ // Convert memusage to MiB, rounding up to the next full MiB.
+ // This way the user can always use the displayed usage as
+ // the new memory usage limit. (If we rounded to the nearest,
+ // the user might need to +1 MiB to get high enough limit.)
+ memusage = round_up_to_mib(memusage);
+
+ // 2^64 with thousand separators + " MiB" suffix + '\0' = 26 + 4 + 1
+ char memlimitstr[32];
+
+ // Show the memory usage limit as MiB unless it is less than 1 MiB.
+ // This way it's easy to notice errors where one has typed
+ // --memory=123 instead of --memory=123MiB.
+ uint64_t memlimit = hardware_memlimit_get();
+ if (memlimit < (UINT32_C(1) << 20)) {
+ snprintf(memlimitstr, sizeof(memlimitstr), "%s B",
+ uint64_to_str(memlimit, 1));
+ } else {
+ // Round up just like with memusage. If this function is
+ // called for informational purposes (to just show the
+ // current usage and limit), we should never show that
+ // the usage is higher than the limit, which would give
+ // a false impression that the memory usage limit isn't
+ // properly enforced.
+ snprintf(memlimitstr, sizeof(memlimitstr), "%s MiB",
+ uint64_to_str(round_up_to_mib(memlimit), 1));
+ }
+
+ message(v, _("%s MiB of memory is required. The limit is %s."),
+ uint64_to_str(memusage, 0), memlimitstr);
+
+ return;
+}
+
+
+/// \brief Convert uint32_t to a nice string for --lzma[12]=dict=SIZE
+///
+/// The idea is to use KiB or MiB suffix when possible.
+static const char *
+uint32_to_optstr(uint32_t num)
+{
+ static char buf[16];
+
+ if ((num & ((UINT32_C(1) << 20) - 1)) == 0)
+ snprintf(buf, sizeof(buf), "%" PRIu32 "MiB", num >> 20);
+ else if ((num & ((UINT32_C(1) << 10) - 1)) == 0)
+ snprintf(buf, sizeof(buf), "%" PRIu32 "KiB", num >> 10);
+ else
+ snprintf(buf, sizeof(buf), "%" PRIu32, num);
+
+ return buf;
+}
+
+
+extern const char *
+message_filters_to_str(const lzma_filter *filters, bool all_known)
+{
+ static char buf[512];
+
+ char *pos = buf;
+ size_t left = sizeof(buf);
+
+ for (size_t i = 0; filters[i].id != LZMA_VLI_UNKNOWN; ++i) {
+ // Add the dashes for the filter option. A space is
+ // needed after the first and later filters.
+ my_snprintf(&pos, &left, "%s", i == 0 ? "--" : " --");
+
+ switch (filters[i].id) {
+ case LZMA_FILTER_LZMA1:
+ case LZMA_FILTER_LZMA2: {
+ const lzma_options_lzma *opt = filters[i].options;
+ const char *mode = NULL;
+ const char *mf = NULL;
+
+ if (all_known) {
+ switch (opt->mode) {
+ case LZMA_MODE_FAST:
+ mode = "fast";
+ break;
+
+ case LZMA_MODE_NORMAL:
+ mode = "normal";
+ break;
+
+ default:
+ mode = "UNKNOWN";
+ break;
+ }
+
+ switch (opt->mf) {
+ case LZMA_MF_HC3:
+ mf = "hc3";
+ break;
+
+ case LZMA_MF_HC4:
+ mf = "hc4";
+ break;
+
+ case LZMA_MF_BT2:
+ mf = "bt2";
+ break;
+
+ case LZMA_MF_BT3:
+ mf = "bt3";
+ break;
+
+ case LZMA_MF_BT4:
+ mf = "bt4";
+ break;
+
+ default:
+ mf = "UNKNOWN";
+ break;
+ }
+ }
+
+ // Add the filter name and dictionary size, which
+ // is always known.
+ my_snprintf(&pos, &left, "lzma%c=dict=%s",
+ filters[i].id == LZMA_FILTER_LZMA2
+ ? '2' : '1',
+ uint32_to_optstr(opt->dict_size));
+
+ // With LZMA1 also lc/lp/pb are known when
+ // decompressing, but this function is never
+ // used to print information about .lzma headers.
+ assert(filters[i].id == LZMA_FILTER_LZMA2
+ || all_known);
+
+ // Print the rest of the options, which are known
+ // only when compressing.
+ if (all_known)
+ my_snprintf(&pos, &left,
+ ",lc=%" PRIu32 ",lp=%" PRIu32
+ ",pb=%" PRIu32
+ ",mode=%s,nice=%" PRIu32 ",mf=%s"
+ ",depth=%" PRIu32,
+ opt->lc, opt->lp, opt->pb,
+ mode, opt->nice_len, mf, opt->depth);
+ break;
+ }
+
+ case LZMA_FILTER_X86:
+ case LZMA_FILTER_POWERPC:
+ case LZMA_FILTER_IA64:
+ case LZMA_FILTER_ARM:
+ case LZMA_FILTER_ARMTHUMB:
+ case LZMA_FILTER_SPARC: {
+ static const char bcj_names[][9] = {
+ "x86",
+ "powerpc",
+ "ia64",
+ "arm",
+ "armthumb",
+ "sparc",
+ };
+
+ const lzma_options_bcj *opt = filters[i].options;
+ my_snprintf(&pos, &left, "%s", bcj_names[filters[i].id
+ - LZMA_FILTER_X86]);
+
+ // Show the start offset only when really needed.
+ if (opt != NULL && opt->start_offset != 0)
+ my_snprintf(&pos, &left, "=start=%" PRIu32,
+ opt->start_offset);
+
+ break;
+ }
+
+ case LZMA_FILTER_DELTA: {
+ const lzma_options_delta *opt = filters[i].options;
+ my_snprintf(&pos, &left, "delta=dist=%" PRIu32,
+ opt->dist);
+ break;
+ }
+
+ default:
+ // This should be possible only if liblzma is
+ // newer than the xz tool.
+ my_snprintf(&pos, &left, "UNKNOWN");
+ break;
+ }
+ }
+
+ return buf;
+}
+
+
+extern void
+message_filters_show(enum message_verbosity v, const lzma_filter *filters)
+{
+ if (v > verbosity)
+ return;
+
+ fprintf(stderr, _("%s: Filter chain: %s\n"), progname,
+ message_filters_to_str(filters, true));
+ return;
+}
+
+
+extern void
+message_try_help(void)
+{
+ // Print this with V_WARNING instead of V_ERROR to prevent it from
+ // showing up when --quiet has been specified.
+ message(V_WARNING, _("Try `%s --help' for more information."),
+ progname);
+ return;
+}
+
+
+extern void
+message_memlimit(void)
+{
+ if (opt_robot)
+ printf("%" PRIu64 "\n", hardware_memlimit_get());
+ else
+ printf(_("%s MiB (%s bytes)\n"),
+ uint64_to_str(
+ round_up_to_mib(hardware_memlimit_get()), 0),
+ uint64_to_str(hardware_memlimit_get(), 1));
+
+ tuklib_exit(E_SUCCESS, E_ERROR, verbosity != V_SILENT);
+}
+
+
+extern void
+message_version(void)
+{
+ // It is possible that liblzma version is different than the command
+ // line tool version, so print both.
+ if (opt_robot) {
+ printf("XZ_VERSION=%d\nLIBLZMA_VERSION=%d\n",
+ LZMA_VERSION, lzma_version_number());
+ } else {
+ printf("xz (" PACKAGE_NAME ") " LZMA_VERSION_STRING "\n");
+ printf("liblzma %s\n", lzma_version_string());
+ }
+
+ tuklib_exit(E_SUCCESS, E_ERROR, verbosity != V_SILENT);
+}
+
+
+extern void
+message_help(bool long_help)
+{
+ printf(_("Usage: %s [OPTION]... [FILE]...\n"
+ "Compress or decompress FILEs in the .xz format.\n\n"),
+ progname);
+
+ puts(_("Mandatory arguments to long options are mandatory for "
+ "short options too.\n"));
+
+ if (long_help)
+ puts(_(" Operation mode:\n"));
+
+ puts(_(
+" -z, --compress force compression\n"
+" -d, --decompress force decompression\n"
+" -t, --test test compressed file integrity\n"
+" -l, --list list information about files"));
+
+ if (long_help)
+ puts(_("\n Operation modifiers:\n"));
+
+ puts(_(
+" -k, --keep keep (don't delete) input files\n"
+" -f, --force force overwrite of output file and (de)compress links\n"
+" -c, --stdout write to standard output and don't delete input files"));
+
+ if (long_help)
+ puts(_(
+" --no-sparse do not create sparse files when decompressing\n"
+" -S, --suffix=.SUF use the suffix `.SUF' on compressed files\n"
+" --files=[FILE] read filenames to process from FILE; if FILE is\n"
+" omitted, filenames are read from the standard input;\n"
+" filenames must be terminated with the newline character\n"
+" --files0=[FILE] like --files but use the null character as terminator"));
+
+ if (long_help) {
+ puts(_("\n Basic file format and compression options:\n"));
+ puts(_(
+" -F, --format=FMT file format to encode or decode; possible values are\n"
+" `auto' (default), `xz', `lzma', and `raw'\n"
+" -C, --check=CHECK integrity check type: `crc32', `crc64' (default),\n"
+" `sha256', or `none' (use with caution)"));
+ }
+
+ puts(_(
+" -0 .. -9 compression preset; 0-2 fast compression, 3-5 good\n"
+" compression, 6-9 excellent compression; default is 6"));
+
+ puts(_(
+" -e, --extreme use more CPU time when encoding to increase compression\n"
+" ratio without increasing memory usage of the decoder"));
+
+ if (long_help) {
+ puts(_(
+" --no-adjust if compression settings exceed the memory usage limit,\n"
+" give an error instead of adjusting the settings downwards"));
+ puts(_( // xgettext:no-c-format
+" -M, --memory=NUM use roughly NUM bytes of memory at maximum; 0 indicates\n"
+" the default setting, which is 40 % of total RAM"));
+ }
+
+ if (long_help) {
+ puts(_(
+"\n Custom filter chain for compression (alternative for using presets):"));
+
+#if defined(HAVE_ENCODER_LZMA1) || defined(HAVE_DECODER_LZMA1) \
+ || defined(HAVE_ENCODER_LZMA2) || defined(HAVE_DECODER_LZMA2)
+ puts(_(
+"\n"
+" --lzma1[=OPTS] LZMA1 or LZMA2; OPTS is a comma-separated list of zero or\n"
+" --lzma2[=OPTS] more of the following options (valid values; default):\n"
+" preset=NUM reset options to preset number NUM (0-9)\n"
+" dict=NUM dictionary size (4KiB - 1536MiB; 8MiB)\n"
+" lc=NUM number of literal context bits (0-4; 3)\n"
+" lp=NUM number of literal position bits (0-4; 0)\n"
+" pb=NUM number of position bits (0-4; 2)\n"
+" mode=MODE compression mode (fast, normal; normal)\n"
+" nice=NUM nice length of a match (2-273; 64)\n"
+" mf=NAME match finder (hc3, hc4, bt2, bt3, bt4; bt4)\n"
+" depth=NUM maximum search depth; 0=automatic (default)"));
+#endif
+
+ puts(_(
+"\n"
+" --x86[=OPTS] x86 BCJ filter\n"
+" --powerpc[=OPTS] PowerPC BCJ filter (big endian only)\n"
+" --ia64[=OPTS] IA64 (Itanium) BCJ filter\n"
+" --arm[=OPTS] ARM BCJ filter (little endian only)\n"
+" --armthumb[=OPTS] ARM-Thumb BCJ filter (little endian only)\n"
+" --sparc[=OPTS] SPARC BCJ filter\n"
+" Valid OPTS for all BCJ filters:\n"
+" start=NUM start offset for conversions (default=0)"));
+
+#if defined(HAVE_ENCODER_DELTA) || defined(HAVE_DECODER_DELTA)
+ puts(_(
+"\n"
+" --delta[=OPTS] Delta filter; valid OPTS (valid values; default):\n"
+" dist=NUM distance between bytes being subtracted\n"
+" from each other (1-256; 1)"));
+#endif
+ }
+
+ if (long_help)
+ puts(_("\n Other options:\n"));
+
+ puts(_(
+" -q, --quiet suppress warnings; specify twice to suppress errors too\n"
+" -v, --verbose be verbose; specify twice for even more verbose"));
+
+ if (long_help) {
+ puts(_(
+" -Q, --no-warn make warnings not affect the exit status"));
+ puts(_(
+" --robot use machine-parsable messages (useful for scripts)"));
+ puts("");
+ puts(_(
+" --info-memory display the memory usage limit and exit"));
+ puts(_(
+" -h, --help display the short help (lists only the basic options)\n"
+" -H, --long-help display this long help and exit"));
+ } else {
+ puts(_(
+" -h, --help display this short help and exit\n"
+" -H, --long-help display the long help (lists also the advanced options)"));
+ }
+
+ puts(_(
+" -V, --version display the version number and exit"));
+
+ puts(_("\nWith no FILE, or when FILE is -, read standard input.\n"));
+
+ if (long_help) {
+ printf(_(
+"On this system and configuration, this program will use a maximum of roughly\n"
+"%s MiB RAM and "), uint64_to_str(round_up_to_mib(hardware_memlimit_get()), 0));
+ printf(N_("one thread.\n\n", "%s threads.\n\n",
+ hardware_threadlimit_get()),
+ uint64_to_str(hardware_threadlimit_get(), 0));
+ }
+
+ // TRANSLATORS: This message indicates the bug reporting address
+ // for this package. Please add _another line_ saying
+ // "Report translation bugs to <...>\n" with the email or WWW
+ // address for translation bugs. Thanks.
+ printf(_("Report bugs to <%s> (in English or Finnish).\n"),
+ PACKAGE_BUGREPORT);
+ printf(_("%s home page: <%s>\n"), PACKAGE_NAME, PACKAGE_URL);
+
+ tuklib_exit(E_SUCCESS, E_ERROR, verbosity != V_SILENT);
+}
diff --git a/src/xz/message.h b/src/xz/message.h
new file mode 100644
index 0000000..aea4fdf
--- /dev/null
+++ b/src/xz/message.h
@@ -0,0 +1,163 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file message.h
+/// \brief Printing messages to stderr
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// Verbosity levels
+enum message_verbosity {
+ V_SILENT, ///< No messages
+ V_ERROR, ///< Only error messages
+ V_WARNING, ///< Errors and warnings
+ V_VERBOSE, ///< Errors, warnings, and verbose statistics
+ V_DEBUG, ///< Debugging, FIXME remove?
+};
+
+
+/// \brief Initializes the message functions
+///
+/// If an error occurs, this function doesn't return.
+///
+extern void message_init(void);
+
+
+/// Increase verbosity level by one step unless it was at maximum.
+extern void message_verbosity_increase(void);
+
+/// Decrease verbosity level by one step unless it was at minimum.
+extern void message_verbosity_decrease(void);
+
+/// Get the current verbosity level.
+extern enum message_verbosity message_verbosity_get(void);
+
+
+/// \brief Print a message if verbosity level is at least "verbosity"
+///
+/// This doesn't touch the exit status.
+extern void message(enum message_verbosity verbosity, const char *fmt, ...)
+ lzma_attribute((format(printf, 2, 3)));
+
+
+/// \brief Prints a warning and possibly sets exit status
+///
+/// The message is printed only if verbosity level is at least V_WARNING.
+/// The exit status is set to WARNING unless it was already at ERROR.
+extern void message_warning(const char *fmt, ...)
+ lzma_attribute((format(printf, 1, 2)));
+
+
+/// \brief Prints an error message and sets exit status
+///
+/// The message is printed only if verbosity level is at least V_ERROR.
+/// The exit status is set to ERROR.
+extern void message_error(const char *fmt, ...)
+ lzma_attribute((format(printf, 1, 2)));
+
+
+/// \brief Prints an error message and exits with EXIT_ERROR
+///
+/// The message is printed only if verbosity level is at least V_ERROR.
+extern void message_fatal(const char *fmt, ...)
+ lzma_attribute((format(printf, 1, 2)))
+ lzma_attribute((noreturn));
+
+
+/// Print an error message that an internal error occurred and exit with
+/// EXIT_ERROR.
+extern void message_bug(void) lzma_attribute((noreturn));
+
+
+/// Print a message that establishing signal handlers failed, and exit with
+/// exit status ERROR.
+extern void message_signal_handler(void) lzma_attribute((noreturn));
+
+
+/// Convert lzma_ret to a string.
+extern const char *message_strm(lzma_ret code);
+
+
+/// Display how much memory was needed and how much the limit was.
+extern void message_mem_needed(enum message_verbosity v, uint64_t memusage);
+
+
+/// \brief Get the filter chain as a string
+///
+/// \param filters Pointer to the filter chain
+/// \param all_known If true, all filter options are printed.
+/// If false, only the options that get stored
+/// into .xz headers are printed.
+///
+/// \return Pointer to a statically allocated buffer.
+extern const char *message_filters_to_str(
+ const lzma_filter *filters, bool all_known);
+
+
+/// Print the filter chain.
+extern void message_filters_show(
+ enum message_verbosity v, const lzma_filter *filters);
+
+
+/// Print a message that user should try --help.
+extern void message_try_help(void);
+
+
+/// Print the memory usage limit and exit.
+extern void message_memlimit(void) lzma_attribute((noreturn));
+
+
+/// Prints the version number to stdout and exits with exit status SUCCESS.
+extern void message_version(void) lzma_attribute((noreturn));
+
+
+/// Print the help message.
+extern void message_help(bool long_help) lzma_attribute((noreturn));
+
+
+/// \brief Set the total number of files to be processed
+///
+/// Standard input is counted as a file here. This is used when printing
+/// the filename via message_filename().
+extern void message_set_files(unsigned int files);
+
+
+/// \brief Set the name of the current file and possibly print it too
+///
+/// The name is printed immediately if --list was used or if --verbose
+/// was used and stderr is a terminal. Even when the filename isn't printed,
+/// it is stored so that it can be printed later if needed for progress
+/// messages.
+extern void message_filename(const char *src_name);
+
+
+/// \brief Start progress info handling
+///
+/// message_filename() must be called before this function to set
+/// the filename.
+///
+/// This must be paired with a call to message_progress_end() before the
+/// given *strm becomes invalid.
+///
+/// \param strm Pointer to lzma_stream used for the coding.
+/// \param in_size Size of the input file, or zero if unknown.
+///
+extern void message_progress_start(lzma_stream *strm, uint64_t in_size);
+
+
+/// Update the progress info if in verbose mode and enough time has passed
+/// since the previous update. This can be called only when
+/// message_progress_start() has already been used.
+extern void message_progress_update(void);
+
+
+/// \brief Finishes the progress message if we were in verbose mode
+///
+/// \param finished True if the whole stream was successfully coded
+/// and output written to the output stream.
+///
+extern void message_progress_end(bool finished);
diff --git a/src/xz/options.c b/src/xz/options.c
new file mode 100644
index 0000000..7186f24
--- /dev/null
+++ b/src/xz/options.c
@@ -0,0 +1,374 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file options.c
+/// \brief Parser for filter-specific options
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+///////////////////
+// Generic stuff //
+///////////////////
+
+typedef struct {
+ const char *name;
+ uint64_t id;
+} name_id_map;
+
+
+typedef struct {
+ const char *name;
+ const name_id_map *map;
+ uint64_t min;
+ uint64_t max;
+} option_map;
+
+
+/// Parses option=value pairs that are separated with colons, semicolons,
+/// or commas: opt=val:opt=val;opt=val,opt=val
+///
+/// Each option is a string, that is converted to an integer using the
+/// index where the option string is in the array.
+///
+/// Value can be
+/// - a string-id map mapping a list of possible string values to integers
+/// (opts[i].map != NULL, opts[i].min and opts[i].max are ignored);
+/// - a number with minimum and maximum value limit
+/// (opts[i].map == NULL && opts[i].min != UINT64_MAX);
+/// - a string that will be parsed by the filter-specific code
+/// (opts[i].map == NULL && opts[i].min == UINT64_MAX, opts[i].max ignored)
+///
+/// When parsing both option and value succeed, a filter-specific function
+/// is called, which should update the given value to filter-specific
+/// options structure.
+///
+/// \param str String containing the options from the command line
+/// \param opts Filter-specific option map
+/// \param set Filter-specific function to update filter_options
+/// \param filter_options Pointer to filter-specific options structure
+///
+/// \return Returns only if no errors occur.
+///
+static void
+parse_options(const char *str, const option_map *opts,
+ void (*set)(void *filter_options,
+ uint32_t key, uint64_t value, const char *valuestr),
+ void *filter_options)
+{
+ if (str == NULL || str[0] == '\0')
+ return;
+
+ char *s = xstrdup(str);
+ char *name = s;
+
+ while (*name != '\0') {
+ if (*name == ',') {
+ ++name;
+ continue;
+ }
+
+ char *split = strchr(name, ',');
+ if (split != NULL)
+ *split = '\0';
+
+ char *value = strchr(name, '=');
+ if (value != NULL)
+ *value++ = '\0';
+
+ if (value == NULL || value[0] == '\0')
+ message_fatal(_("%s: Options must be `name=value' "
+ "pairs separated with commas"), str);
+
+ // Look for the option name from the option map.
+ size_t i = 0;
+ while (true) {
+ if (opts[i].name == NULL)
+ message_fatal(_("%s: Invalid option name"),
+ name);
+
+ if (strcmp(name, opts[i].name) == 0)
+ break;
+
+ ++i;
+ }
+
+ // Option was found from the map. See how we should handle it.
+ if (opts[i].map != NULL) {
+ // value is a string which we should map
+ // to an integer.
+ size_t j;
+ for (j = 0; opts[i].map[j].name != NULL; ++j) {
+ if (strcmp(opts[i].map[j].name, value) == 0)
+ break;
+ }
+
+ if (opts[i].map[j].name == NULL)
+ message_fatal(_("%s: Invalid option value"),
+ value);
+
+ set(filter_options, i, opts[i].map[j].id, value);
+
+ } else if (opts[i].min == UINT64_MAX) {
+ // value is a special string that will be
+ // parsed by set().
+ set(filter_options, i, 0, value);
+
+ } else {
+ // value is an integer.
+ const uint64_t v = str_to_uint64(name, value,
+ opts[i].min, opts[i].max);
+ set(filter_options, i, v, value);
+ }
+
+ // Check if it was the last option.
+ if (split == NULL)
+ break;
+
+ name = split + 1;
+ }
+
+ free(s);
+ return;
+}
+
+
+///////////
+// Delta //
+///////////
+
+enum {
+ OPT_DIST,
+};
+
+
+static void
+set_delta(void *options, uint32_t key, uint64_t value,
+ const char *valuestr lzma_attribute((unused)))
+{
+ lzma_options_delta *opt = options;
+ switch (key) {
+ case OPT_DIST:
+ opt->dist = value;
+ break;
+ }
+}
+
+
+extern lzma_options_delta *
+options_delta(const char *str)
+{
+ static const option_map opts[] = {
+ { "dist", NULL, LZMA_DELTA_DIST_MIN,
+ LZMA_DELTA_DIST_MAX },
+ { NULL, NULL, 0, 0 }
+ };
+
+ lzma_options_delta *options = xmalloc(sizeof(lzma_options_delta));
+ *options = (lzma_options_delta){
+ // It's hard to give a useful default for this.
+ .type = LZMA_DELTA_TYPE_BYTE,
+ .dist = LZMA_DELTA_DIST_MIN,
+ };
+
+ parse_options(str, opts, &set_delta, options);
+
+ return options;
+}
+
+
+/////////
+// BCJ //
+/////////
+
+enum {
+ OPT_START_OFFSET,
+};
+
+
+static void
+set_bcj(void *options, uint32_t key, uint64_t value,
+ const char *valuestr lzma_attribute((unused)))
+{
+ lzma_options_bcj *opt = options;
+ switch (key) {
+ case OPT_START_OFFSET:
+ opt->start_offset = value;
+ break;
+ }
+}
+
+
+extern lzma_options_bcj *
+options_bcj(const char *str)
+{
+ static const option_map opts[] = {
+ { "start", NULL, 0, UINT32_MAX },
+ { NULL, NULL, 0, 0 }
+ };
+
+ lzma_options_bcj *options = xmalloc(sizeof(lzma_options_bcj));
+ *options = (lzma_options_bcj){
+ .start_offset = 0,
+ };
+
+ parse_options(str, opts, &set_bcj, options);
+
+ return options;
+}
+
+
+//////////
+// LZMA //
+//////////
+
+enum {
+ OPT_PRESET,
+ OPT_DICT,
+ OPT_LC,
+ OPT_LP,
+ OPT_PB,
+ OPT_MODE,
+ OPT_NICE,
+ OPT_MF,
+ OPT_DEPTH,
+};
+
+
+static void lzma_attribute((noreturn))
+error_lzma_preset(const char *valuestr)
+{
+ message_fatal(_("Unsupported LZMA1/LZMA2 preset: %s"), valuestr);
+}
+
+
+static void
+set_lzma(void *options, uint32_t key, uint64_t value, const char *valuestr)
+{
+ lzma_options_lzma *opt = options;
+
+ switch (key) {
+ case OPT_PRESET: {
+ if (valuestr[0] < '0' || valuestr[0] > '9')
+ error_lzma_preset(valuestr);
+
+ uint32_t preset = valuestr[0] - '0';
+
+ // Currently only "e" is supported as a modifier,
+ // so keep this simple for now.
+ if (valuestr[1] != '\0') {
+ if (valuestr[1] == 'e')
+ preset |= LZMA_PRESET_EXTREME;
+ else
+ error_lzma_preset(valuestr);
+
+ if (valuestr[2] != '\0')
+ error_lzma_preset(valuestr);
+ }
+
+ if (lzma_lzma_preset(options, preset))
+ error_lzma_preset(valuestr);
+
+ break;
+ }
+
+ case OPT_DICT:
+ opt->dict_size = value;
+ break;
+
+ case OPT_LC:
+ opt->lc = value;
+ break;
+
+ case OPT_LP:
+ opt->lp = value;
+ break;
+
+ case OPT_PB:
+ opt->pb = value;
+ break;
+
+ case OPT_MODE:
+ opt->mode = value;
+ break;
+
+ case OPT_NICE:
+ opt->nice_len = value;
+ break;
+
+ case OPT_MF:
+ opt->mf = value;
+ break;
+
+ case OPT_DEPTH:
+ opt->depth = value;
+ break;
+ }
+}
+
+
+extern lzma_options_lzma *
+options_lzma(const char *str)
+{
+ static const name_id_map modes[] = {
+ { "fast", LZMA_MODE_FAST },
+ { "normal", LZMA_MODE_NORMAL },
+ { NULL, 0 }
+ };
+
+ static const name_id_map mfs[] = {
+ { "hc3", LZMA_MF_HC3 },
+ { "hc4", LZMA_MF_HC4 },
+ { "bt2", LZMA_MF_BT2 },
+ { "bt3", LZMA_MF_BT3 },
+ { "bt4", LZMA_MF_BT4 },
+ { NULL, 0 }
+ };
+
+ static const option_map opts[] = {
+ { "preset", NULL, UINT64_MAX, 0 },
+ { "dict", NULL, LZMA_DICT_SIZE_MIN,
+ (UINT32_C(1) << 30) + (UINT32_C(1) << 29) },
+ { "lc", NULL, LZMA_LCLP_MIN, LZMA_LCLP_MAX },
+ { "lp", NULL, LZMA_LCLP_MIN, LZMA_LCLP_MAX },
+ { "pb", NULL, LZMA_PB_MIN, LZMA_PB_MAX },
+ { "mode", modes, 0, 0 },
+ { "nice", NULL, 2, 273 },
+ { "mf", mfs, 0, 0 },
+ { "depth", NULL, 0, UINT32_MAX },
+ { NULL, NULL, 0, 0 }
+ };
+
+ lzma_options_lzma *options = xmalloc(sizeof(lzma_options_lzma));
+ *options = (lzma_options_lzma){
+ .dict_size = LZMA_DICT_SIZE_DEFAULT,
+ .preset_dict = NULL,
+ .preset_dict_size = 0,
+ .lc = LZMA_LC_DEFAULT,
+ .lp = LZMA_LP_DEFAULT,
+ .pb = LZMA_PB_DEFAULT,
+ .mode = LZMA_MODE_NORMAL,
+ .nice_len = 64,
+ .mf = LZMA_MF_BT4,
+ .depth = 0,
+ };
+
+ parse_options(str, opts, &set_lzma, options);
+
+ if (options->lc + options->lp > LZMA_LCLP_MAX)
+ message_fatal(_("The sum of lc and lp must be at "
+ "maximum of 4"));
+
+ const uint32_t nice_len_min = options->mf & 0x0F;
+ if (options->nice_len < nice_len_min)
+ message_fatal(_("The selected match finder requires at "
+ "least nice=%" PRIu32), nice_len_min);
+
+ return options;
+}
diff --git a/src/xz/options.h b/src/xz/options.h
new file mode 100644
index 0000000..61ec8d5
--- /dev/null
+++ b/src/xz/options.h
@@ -0,0 +1,31 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file options.h
+/// \brief Parser for filter-specific options
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief Parser for Delta options
+///
+/// \return Pointer to allocated options structure.
+/// Doesn't return on error.
+extern lzma_options_delta *options_delta(const char *str);
+
+
+/// \brief Parser for BCJ options
+///
+/// \return Pointer to allocated options structure.
+/// Doesn't return on error.
+extern lzma_options_bcj *options_bcj(const char *str);
+
+
+/// \brief Parser for LZMA options
+///
+/// \return Pointer to allocated options structure.
+/// Doesn't return on error.
+extern lzma_options_lzma *options_lzma(const char *str);
diff --git a/src/xz/private.h b/src/xz/private.h
new file mode 100644
index 0000000..15136bf
--- /dev/null
+++ b/src/xz/private.h
@@ -0,0 +1,56 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file private.h
+/// \brief Common includes, definions, and prototypes
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "mythread.h"
+#include "lzma.h"
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <errno.h>
+#include <signal.h>
+#include <locale.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include "tuklib_gettext.h"
+#include "tuklib_progname.h"
+#include "tuklib_exit.h"
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+# define WIN32_LEAN_AND_MEAN
+# include <windows.h>
+#endif
+
+#ifndef STDIN_FILENO
+# define STDIN_FILENO (fileno(stdin))
+#endif
+
+#ifndef STDOUT_FILENO
+# define STDOUT_FILENO (fileno(stdout))
+#endif
+
+#ifndef STDERR_FILENO
+# define STDERR_FILENO (fileno(stderr))
+#endif
+
+#include "main.h"
+#include "coder.h"
+#include "message.h"
+#include "args.h"
+#include "hardware.h"
+#include "file_io.h"
+#include "options.h"
+#include "signals.h"
+#include "suffix.h"
+#include "util.h"
+#include "list.h"
diff --git a/src/xz/signals.c b/src/xz/signals.c
new file mode 100644
index 0000000..66d6537
--- /dev/null
+++ b/src/xz/signals.c
@@ -0,0 +1,189 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file signals.c
+/// \brief Handling signals to abort operation
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+
+volatile sig_atomic_t user_abort = false;
+
+
+#if !(defined(_WIN32) && !defined(__CYGWIN__))
+
+/// If we were interrupted by a signal, we store the signal number so that
+/// we can raise that signal to kill the program when all cleanups have
+/// been done.
+static volatile sig_atomic_t exit_signal = 0;
+
+/// Mask of signals for which have have established a signal handler to set
+/// user_abort to true.
+static sigset_t hooked_signals;
+
+/// True once signals_init() has finished. This is used to skip blocking
+/// signals (with uninitialized hooked_signals) if signals_block() and
+/// signals_unblock() are called before signals_init() has been called.
+static bool signals_are_initialized = false;
+
+/// signals_block() and signals_unblock() can be called recursively.
+static size_t signals_block_count = 0;
+
+
+static void
+signal_handler(int sig)
+{
+ exit_signal = sig;
+ user_abort = true;
+ return;
+}
+
+
+extern void
+signals_init(void)
+{
+ // List of signals for which we establish the signal handler.
+ static const int sigs[] = {
+ SIGINT,
+ SIGTERM,
+#ifdef SIGHUP
+ SIGHUP,
+#endif
+#ifdef SIGPIPE
+ SIGPIPE,
+#endif
+#ifdef SIGXCPU
+ SIGXCPU,
+#endif
+#ifdef SIGXFSZ
+ SIGXFSZ,
+#endif
+ };
+
+ // Mask of the signals for which we have established a signal handler.
+ sigemptyset(&hooked_signals);
+ for (size_t i = 0; i < ARRAY_SIZE(sigs); ++i)
+ sigaddset(&hooked_signals, sigs[i]);
+
+ struct sigaction sa;
+
+ // All the signals that we handle we also blocked while the signal
+ // handler runs.
+ sa.sa_mask = hooked_signals;
+
+ // Don't set SA_RESTART, because we want EINTR so that we can check
+ // for user_abort and cleanup before exiting. We block the signals
+ // for which we have established a handler when we don't want EINTR.
+ sa.sa_flags = 0;
+ sa.sa_handler = &signal_handler;
+
+ for (size_t i = 0; i < ARRAY_SIZE(sigs); ++i) {
+ // If the parent process has left some signals ignored,
+ // we don't unignore them.
+ struct sigaction old;
+ if (sigaction(sigs[i], NULL, &old) == 0
+ && old.sa_handler == SIG_IGN)
+ continue;
+
+ // Establish the signal handler.
+ if (sigaction(sigs[i], &sa, NULL))
+ message_signal_handler();
+ }
+
+ signals_are_initialized = true;
+
+ return;
+}
+
+
+#ifndef __VMS
+extern void
+signals_block(void)
+{
+ if (signals_are_initialized) {
+ if (signals_block_count++ == 0) {
+ const int saved_errno = errno;
+ mythread_sigmask(SIG_BLOCK, &hooked_signals, NULL);
+ errno = saved_errno;
+ }
+ }
+
+ return;
+}
+
+
+extern void
+signals_unblock(void)
+{
+ if (signals_are_initialized) {
+ assert(signals_block_count > 0);
+
+ if (--signals_block_count == 0) {
+ const int saved_errno = errno;
+ mythread_sigmask(SIG_UNBLOCK, &hooked_signals, NULL);
+ errno = saved_errno;
+ }
+ }
+
+ return;
+}
+#endif
+
+
+extern void
+signals_exit(void)
+{
+ const int sig = exit_signal;
+
+ if (sig != 0) {
+ struct sigaction sa;
+ sa.sa_handler = SIG_DFL;
+ sigfillset(&sa.sa_mask);
+ sa.sa_flags = 0;
+ sigaction(sig, &sa, NULL);
+ raise(exit_signal);
+ }
+
+ return;
+}
+
+#else
+
+// While Windows has some very basic signal handling functions as required
+// by C89, they are not really used, and e.g. SIGINT doesn't work exactly
+// the way it does on POSIX (Windows creates a new thread for the signal
+// handler). Instead, we use SetConsoleCtrlHandler() to catch user
+// pressing C-c, because that seems to be the recommended way to do it.
+//
+// NOTE: This doesn't work under MSYS. Trying with SIGINT doesn't work
+// either even if it appeared to work at first. So test using Windows
+// console window.
+
+static BOOL WINAPI
+signal_handler(DWORD type lzma_attribute((unused)))
+{
+ // Since we don't get a signal number which we could raise() at
+ // signals_exit() like on POSIX, just set the exit status to
+ // indicate an error, so that we cannot return with zero exit status.
+ set_exit_status(E_ERROR);
+ user_abort = true;
+ return TRUE;
+}
+
+
+extern void
+signals_init(void)
+{
+ if (!SetConsoleCtrlHandler(&signal_handler, TRUE))
+ message_signal_handler();
+
+ return;
+}
+
+#endif
diff --git a/src/xz/signals.h b/src/xz/signals.h
new file mode 100644
index 0000000..5b125e0
--- /dev/null
+++ b/src/xz/signals.h
@@ -0,0 +1,43 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file signals.h
+/// \brief Handling signals to abort operation
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// If this is true, we will clean up the possibly incomplete output file,
+/// return to main() as soon as practical. That is, the code needs to poll
+/// this variable in various places.
+extern volatile sig_atomic_t user_abort;
+
+
+/// Initialize the signal handler, which will set user_abort to true when
+/// user e.g. presses C-c.
+extern void signals_init(void);
+
+
+#if (defined(_WIN32) && !defined(__CYGWIN__)) || defined(__VMS)
+# define signals_block() do { } while (0)
+# define signals_unblock() do { } while (0)
+#else
+/// Block the signals which don't have SA_RESTART and which would just set
+/// user_abort to true. This is handy when we don't want to handle EINTR
+/// and don't want SA_RESTART either.
+extern void signals_block(void);
+
+/// Unblock the signals blocked by signals_block().
+extern void signals_unblock(void);
+#endif
+
+#if defined(_WIN32) && !defined(__CYGWIN__)
+# define signals_exit() do { } while (0)
+#else
+/// If user has sent us a signal earlier to terminate the process,
+/// re-raise that signal to actually terminate the process.
+extern void signals_exit(void);
+#endif
diff --git a/src/xz/suffix.c b/src/xz/suffix.c
new file mode 100644
index 0000000..f2a2da2
--- /dev/null
+++ b/src/xz/suffix.c
@@ -0,0 +1,211 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file suffix.c
+/// \brief Checks filename suffix and creates the destination filename
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+
+// For case-insensitive filename suffix on case-insensitive systems
+#if defined(TUKLIB_DOSLIKE) || defined(__VMS)
+# define strcmp strcasecmp
+#endif
+
+
+static char *custom_suffix = NULL;
+
+
+struct suffix_pair {
+ const char *compressed;
+ const char *uncompressed;
+};
+
+
+/// \brief Checks if src_name has given compressed_suffix
+///
+/// \param suffix Filename suffix to look for
+/// \param src_name Input filename
+/// \param src_len strlen(src_name)
+///
+/// \return If src_name has the suffix, src_len - strlen(suffix) is
+/// returned. It's always a positive integer. Otherwise zero
+/// is returned.
+static size_t
+test_suffix(const char *suffix, const char *src_name, size_t src_len)
+{
+ const size_t suffix_len = strlen(suffix);
+
+ // The filename must have at least one character in addition to
+ // the suffix. src_name may contain path to the filename, so we
+ // need to check for directory separator too.
+ if (src_len <= suffix_len || src_name[src_len - suffix_len - 1] == '/')
+ return 0;
+
+ if (strcmp(suffix, src_name + src_len - suffix_len) == 0)
+ return src_len - suffix_len;
+
+ return 0;
+}
+
+
+/// \brief Removes the filename suffix of the compressed file
+///
+/// \return Name of the uncompressed file, or NULL if file has unknown
+/// suffix.
+static char *
+uncompressed_name(const char *src_name, const size_t src_len)
+{
+ static const struct suffix_pair suffixes[] = {
+ { ".xz", "" },
+ { ".txz", ".tar" }, // .txz abbreviation for .txt.gz is rare.
+ { ".lzma", "" },
+ { ".tlz", ".tar" },
+ // { ".gz", "" },
+ // { ".tgz", ".tar" },
+ };
+
+ const char *new_suffix = "";
+ size_t new_len = 0;
+
+ if (opt_format == FORMAT_RAW) {
+ // Don't check for known suffixes when --format=raw was used.
+ if (custom_suffix == NULL) {
+ message_error(_("%s: With --format=raw, "
+ "--suffix=.SUF is required unless "
+ "writing to stdout"), src_name);
+ return NULL;
+ }
+ } else {
+ for (size_t i = 0; i < ARRAY_SIZE(suffixes); ++i) {
+ new_len = test_suffix(suffixes[i].compressed,
+ src_name, src_len);
+ if (new_len != 0) {
+ new_suffix = suffixes[i].uncompressed;
+ break;
+ }
+ }
+ }
+
+ if (new_len == 0 && custom_suffix != NULL)
+ new_len = test_suffix(custom_suffix, src_name, src_len);
+
+ if (new_len == 0) {
+ message_warning(_("%s: Filename has an unknown suffix, "
+ "skipping"), src_name);
+ return NULL;
+ }
+
+ const size_t new_suffix_len = strlen(new_suffix);
+ char *dest_name = xmalloc(new_len + new_suffix_len + 1);
+
+ memcpy(dest_name, src_name, new_len);
+ memcpy(dest_name + new_len, new_suffix, new_suffix_len);
+ dest_name[new_len + new_suffix_len] = '\0';
+
+ return dest_name;
+}
+
+
+/// \brief Appends suffix to src_name
+///
+/// In contrast to uncompressed_name(), we check only suffixes that are valid
+/// for the specified file format.
+static char *
+compressed_name(const char *src_name, const size_t src_len)
+{
+ // The order of these must match the order in args.h.
+ static const struct suffix_pair all_suffixes[][3] = {
+ {
+ { ".xz", "" },
+ { ".txz", ".tar" },
+ { NULL, NULL }
+ }, {
+ { ".lzma", "" },
+ { ".tlz", ".tar" },
+ { NULL, NULL }
+/*
+ }, {
+ { ".gz", "" },
+ { ".tgz", ".tar" },
+ { NULL, NULL }
+*/
+ }, {
+ // --format=raw requires specifying the suffix
+ // manually or using stdout.
+ { NULL, NULL }
+ }
+ };
+
+ // args.c ensures this.
+ assert(opt_format != FORMAT_AUTO);
+
+ const size_t format = opt_format - 1;
+ const struct suffix_pair *const suffixes = all_suffixes[format];
+
+ for (size_t i = 0; suffixes[i].compressed != NULL; ++i) {
+ if (test_suffix(suffixes[i].compressed, src_name, src_len)
+ != 0) {
+ message_warning(_("%s: File already has `%s' "
+ "suffix, skipping"), src_name,
+ suffixes[i].compressed);
+ return NULL;
+ }
+ }
+
+ // TODO: Hmm, maybe it would be better to validate this in args.c,
+ // since the suffix handling when decoding is weird now.
+ if (opt_format == FORMAT_RAW && custom_suffix == NULL) {
+ message_error(_("%s: With --format=raw, "
+ "--suffix=.SUF is required unless "
+ "writing to stdout"), src_name);
+ return NULL;
+ }
+
+ const char *suffix = custom_suffix != NULL
+ ? custom_suffix : suffixes[0].compressed;
+ const size_t suffix_len = strlen(suffix);
+
+ char *dest_name = xmalloc(src_len + suffix_len + 1);
+
+ memcpy(dest_name, src_name, src_len);
+ memcpy(dest_name + src_len, suffix, suffix_len);
+ dest_name[src_len + suffix_len] = '\0';
+
+ return dest_name;
+}
+
+
+extern char *
+suffix_get_dest_name(const char *src_name)
+{
+ assert(src_name != NULL);
+
+ // Length of the name is needed in all cases to locate the end of
+ // the string to compare the suffix, so calculate the length here.
+ const size_t src_len = strlen(src_name);
+
+ return opt_mode == MODE_COMPRESS
+ ? compressed_name(src_name, src_len)
+ : uncompressed_name(src_name, src_len);
+}
+
+
+extern void
+suffix_set(const char *suffix)
+{
+ // Empty suffix and suffixes having a slash are rejected. Such
+ // suffixes would break things later.
+ if (suffix[0] == '\0' || strchr(suffix, '/') != NULL)
+ message_fatal(_("%s: Invalid filename suffix"), optarg);
+
+ // Replace the old custom_suffix (if any) with the new suffix.
+ free(custom_suffix);
+ custom_suffix = xstrdup(suffix);
+ return;
+}
diff --git a/src/xz/suffix.h b/src/xz/suffix.h
new file mode 100644
index 0000000..5537d73
--- /dev/null
+++ b/src/xz/suffix.h
@@ -0,0 +1,28 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file suffix.h
+/// \brief Checks filename suffix and creates the destination filename
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief Get the name of the destination file
+///
+/// Depending on the global variable opt_mode, this tries to find a matching
+/// counterpart for src_name. If the name can be constructed, it is allocated
+/// and returned (caller must free it). On error, a message is printed and
+/// NULL is returned.
+extern char *suffix_get_dest_name(const char *src_name);
+
+
+/// \brief Set a custom filename suffix
+///
+/// This function calls xstrdup() for the given suffix, thus the caller
+/// doesn't need to keep the memory allocated. There can be only one custom
+/// suffix, thus if this is called multiple times, the old suffixes are freed
+/// and forgotten.
+extern void suffix_set(const char *suffix);
diff --git a/src/xz/util.c b/src/xz/util.c
new file mode 100644
index 0000000..19f5eee
--- /dev/null
+++ b/src/xz/util.c
@@ -0,0 +1,320 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file util.c
+/// \brief Miscellaneous utility functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "private.h"
+#include <stdarg.h>
+
+
+extern void *
+xrealloc(void *ptr, size_t size)
+{
+ assert(size > 0);
+
+ ptr = realloc(ptr, size);
+ if (ptr == NULL)
+ message_fatal("%s", strerror(errno));
+
+ return ptr;
+}
+
+
+extern char *
+xstrdup(const char *src)
+{
+ assert(src != NULL);
+ const size_t size = strlen(src) + 1;
+ char *dest = xmalloc(size);
+ return memcpy(dest, src, size);
+}
+
+
+extern uint64_t
+str_to_uint64(const char *name, const char *value, uint64_t min, uint64_t max)
+{
+ uint64_t result = 0;
+
+ // Skip blanks.
+ while (*value == ' ' || *value == '\t')
+ ++value;
+
+ // Accept special value "max". Supporting "min" doesn't seem useful.
+ if (strcmp(value, "max") == 0)
+ return max;
+
+ if (*value < '0' || *value > '9')
+ message_fatal(_("%s: Value is not a non-negative "
+ "decimal integer"), value);
+
+ do {
+ // Don't overflow.
+ if (result > UINT64_MAX / 10)
+ goto error;
+
+ result *= 10;
+
+ // Another overflow check
+ const uint32_t add = *value - '0';
+ if (UINT64_MAX - add < result)
+ goto error;
+
+ result += add;
+ ++value;
+ } while (*value >= '0' && *value <= '9');
+
+ if (*value != '\0') {
+ // Look for suffix. Originally this supported both base-2
+ // and base-10, but since there seems to be little need
+ // for base-10 in this program, treat everything as base-2
+ // and also be more relaxed about the case of the first
+ // letter of the suffix.
+ uint64_t multiplier = 0;
+ if (*value == 'k' || *value == 'K')
+ multiplier = UINT64_C(1) << 10;
+ else if (*value == 'm' || *value == 'M')
+ multiplier = UINT64_C(1) << 20;
+ else if (*value == 'g' || *value == 'G')
+ multiplier = UINT64_C(1) << 30;
+
+ ++value;
+
+ // Allow also e.g. Ki, KiB, and KB.
+ if (*value != '\0' && strcmp(value, "i") != 0
+ && strcmp(value, "iB") != 0
+ && strcmp(value, "B") != 0)
+ multiplier = 0;
+
+ if (multiplier == 0) {
+ message(V_ERROR, _("%s: Invalid multiplier suffix"),
+ value - 1);
+ message_fatal(_("Valid suffixes are `KiB' (2^10), "
+ "`MiB' (2^20), and `GiB' (2^30)."));
+ }
+
+ // Don't overflow here either.
+ if (result > UINT64_MAX / multiplier)
+ goto error;
+
+ result *= multiplier;
+ }
+
+ if (result < min || result > max)
+ goto error;
+
+ return result;
+
+error:
+ message_fatal(_("Value of the option `%s' must be in the range "
+ "[%" PRIu64 ", %" PRIu64 "]"),
+ name, min, max);
+}
+
+
+extern uint64_t
+round_up_to_mib(uint64_t n)
+{
+ return (n >> 20) + ((n & ((UINT32_C(1) << 20) - 1)) != 0);
+}
+
+
+extern const char *
+uint64_to_str(uint64_t value, uint32_t slot)
+{
+ // 2^64 with thousand separators is 26 bytes plus trailing '\0'.
+ static char bufs[4][32];
+
+ assert(slot < ARRAY_SIZE(bufs));
+
+ static enum { UNKNOWN, WORKS, BROKEN } thousand = UNKNOWN;
+ if (thousand == UNKNOWN) {
+ bufs[slot][0] = '\0';
+ snprintf(bufs[slot], sizeof(bufs[slot]), "%'" PRIu64,
+ UINT64_C(1));
+ thousand = bufs[slot][0] == '1' ? WORKS : BROKEN;
+ }
+
+ if (thousand == WORKS)
+ snprintf(bufs[slot], sizeof(bufs[slot]), "%'" PRIu64, value);
+ else
+ snprintf(bufs[slot], sizeof(bufs[slot]), "%" PRIu64, value);
+
+ return bufs[slot];
+}
+
+
+extern const char *
+uint64_to_nicestr(uint64_t value, enum nicestr_unit unit_min,
+ enum nicestr_unit unit_max, bool always_also_bytes,
+ uint32_t slot)
+{
+ assert(unit_min <= unit_max);
+ assert(unit_max <= NICESTR_TIB);
+
+ enum nicestr_unit unit = NICESTR_B;
+ const char *str;
+
+ if ((unit_min == NICESTR_B && value < 10000)
+ || unit_max == NICESTR_B) {
+ // The value is shown as bytes.
+ str = uint64_to_str(value, slot);
+ } else {
+ // Scale the value to a nicer unit. Unless unit_min and
+ // unit_max limit us, we will show at most five significant
+ // digits with one decimal place.
+ double d = (double)(value);
+ do {
+ d /= 1024.0;
+ ++unit;
+ } while (unit < unit_min || (d > 9999.9 && unit < unit_max));
+
+ str = double_to_str(d);
+ }
+
+ static const char suffix[5][4] = { "B", "KiB", "MiB", "GiB", "TiB" };
+
+ // Minimum buffer size:
+ // 26 2^64 with thousand separators
+ // 4 " KiB"
+ // 2 " ("
+ // 26 2^64 with thousand separators
+ // 3 " B)"
+ // 1 '\0'
+ // 62 Total
+ static char buf[4][64];
+ char *pos = buf[slot];
+ size_t left = sizeof(buf[slot]);
+ my_snprintf(&pos, &left, "%s %s", str, suffix[unit]);
+
+ if (always_also_bytes && value >= 10000)
+ snprintf(pos, left, " (%s B)", uint64_to_str(value, slot));
+
+ return buf[slot];
+}
+
+
+extern const char *
+double_to_str(double value)
+{
+ static char buf[64];
+
+ static enum { UNKNOWN, WORKS, BROKEN } thousand = UNKNOWN;
+ if (thousand == UNKNOWN) {
+ buf[0] = '\0';
+ snprintf(buf, sizeof(buf), "%'.1f", 2.0);
+ thousand = buf[0] == '2' ? WORKS : BROKEN;
+ }
+
+ if (thousand == WORKS)
+ snprintf(buf, sizeof(buf), "%'.1f", value);
+ else
+ snprintf(buf, sizeof(buf), "%.1f", value);
+
+ return buf;
+}
+
+
+extern void
+my_snprintf(char **pos, size_t *left, const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+ const int len = vsnprintf(*pos, *left, fmt, ap);
+ va_end(ap);
+
+ // If an error occurred, we want the caller to think that the whole
+ // buffer was used. This way no more data will be written to the
+ // buffer. We don't need better error handling here.
+ if (len < 0 || (size_t)(len) >= *left) {
+ *left = 0;
+ } else {
+ *pos += len;
+ *left -= len;
+ }
+
+ return;
+}
+
+
+/*
+/// \brief Simple quoting to get rid of ASCII control characters
+///
+/// This is not so cool and locale-dependent, but should be good enough
+/// At least we don't print any control characters on the terminal.
+///
+extern char *
+str_quote(const char *str)
+{
+ size_t dest_len = 0;
+ bool has_ctrl = false;
+
+ while (str[dest_len] != '\0')
+ if (*(unsigned char *)(str + dest_len++) < 0x20)
+ has_ctrl = true;
+
+ char *dest = malloc(dest_len + 1);
+ if (dest != NULL) {
+ if (has_ctrl) {
+ for (size_t i = 0; i < dest_len; ++i)
+ if (*(unsigned char *)(str + i) < 0x20)
+ dest[i] = '?';
+ else
+ dest[i] = str[i];
+
+ dest[dest_len] = '\0';
+
+ } else {
+ // Usually there are no control characters,
+ // so we can optimize.
+ memcpy(dest, str, dest_len + 1);
+ }
+ }
+
+ return dest;
+}
+*/
+
+
+extern bool
+is_empty_filename(const char *filename)
+{
+ if (filename[0] == '\0') {
+ message_error(_("Empty filename, skipping"));
+ return true;
+ }
+
+ return false;
+}
+
+
+extern bool
+is_tty_stdin(void)
+{
+ const bool ret = isatty(STDIN_FILENO);
+
+ if (ret)
+ message_error(_("Compressed data cannot be read from "
+ "a terminal"));
+
+ return ret;
+}
+
+
+extern bool
+is_tty_stdout(void)
+{
+ const bool ret = isatty(STDOUT_FILENO);
+
+ if (ret)
+ message_error(_("Compressed data cannot be written to "
+ "a terminal"));
+
+ return ret;
+}
diff --git a/src/xz/util.h b/src/xz/util.h
new file mode 100644
index 0000000..2e08b4a
--- /dev/null
+++ b/src/xz/util.h
@@ -0,0 +1,129 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file util.h
+/// \brief Miscellaneous utility functions
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+/// \brief Safe malloc() that never returns NULL
+///
+/// \note xmalloc(), xrealloc(), and xstrdup() must not be used when
+/// there are files open for writing, that should be cleaned up
+/// before exiting.
+#define xmalloc(size) xrealloc(NULL, size)
+
+
+/// \brief Safe realloc() that never returns NULL
+extern void *xrealloc(void *ptr, size_t size);
+
+
+/// \brief Safe strdup() that never returns NULL
+extern char *xstrdup(const char *src);
+
+
+/// \brief Fancy version of strtoull()
+///
+/// \param name Name of the option to show in case of an error
+/// \param value String containing the number to be parsed; may
+/// contain suffixes "k", "M", "G", "Ki", "Mi", or "Gi"
+/// \param min Minimum valid value
+/// \param max Maximum valid value
+///
+/// \return Parsed value that is in the range [min, max]. Does not return
+/// if an error occurs.
+///
+extern uint64_t str_to_uint64(const char *name, const char *value,
+ uint64_t min, uint64_t max);
+
+
+/// \brief Round an integer up to the next full MiB and convert to MiB
+///
+/// This is used when printing memory usage and limit.
+extern uint64_t round_up_to_mib(uint64_t n);
+
+
+/// \brief Convert uint64_t to a string
+///
+/// Convert the given value to a string with locale-specific thousand
+/// separators, if supported by the snprintf() implementation. The string
+/// is stored into an internal static buffer indicated by the slot argument.
+/// A pointer to the selected buffer is returned.
+///
+/// This function exists, because non-POSIX systems don't support thousand
+/// separator in format strings. Solving the problem in a simple way doesn't
+/// work, because it breaks gettext (specifically, the xgettext tool).
+extern const char *uint64_to_str(uint64_t value, uint32_t slot);
+
+
+enum nicestr_unit {
+ NICESTR_B,
+ NICESTR_KIB,
+ NICESTR_MIB,
+ NICESTR_GIB,
+ NICESTR_TIB,
+};
+
+
+/// \brief Convert uint64_t to a nice human readable string
+///
+/// This is like uint64_to_str() but uses B, KiB, MiB, GiB, or TiB suffix
+/// and optionally includes the exact size in parenthesis.
+///
+/// \param value Value to be printed
+/// \param unit_min Smallest unit to use. This and unit_max are used
+/// e.g. when showing the progress indicator to force
+/// the unit to MiB.
+/// \param unit_max Biggest unit to use. assert(unit_min <= unit_max).
+/// \param always_also_bytes
+/// Show also the exact byte value in parenthesis
+/// if the nicely formatted string uses bigger unit
+/// than bytes.
+/// \param slot Which static buffer to use to hold the string.
+/// This is shared with uint64_to_str().
+///
+/// \return Pointer to statically allocated buffer containing the string.
+///
+/// \note This uses double_to_str() internally so the static buffer
+/// in double_to_str() will be overwritten.
+///
+extern const char *uint64_to_nicestr(uint64_t value,
+ enum nicestr_unit unit_min, enum nicestr_unit unit_max,
+ bool always_also_bytes, uint32_t slot);
+
+
+/// \brief Convert double to a string with one decimal place
+///
+/// This is like uint64_to_str() except that this converts a double and
+/// uses exactly one decimal place.
+extern const char *double_to_str(double value);
+
+
+/// \brief Wrapper for snprintf() to help constructing a string in pieces
+///
+/// A maximum of *left bytes is written starting from *pos. *pos and *left
+/// are updated accordingly.
+extern void my_snprintf(char **pos, size_t *left, const char *fmt, ...)
+ lzma_attribute((format(printf, 3, 4)));
+
+
+/// \brief Check if filename is empty and print an error message
+extern bool is_empty_filename(const char *filename);
+
+
+/// \brief Test if stdin is a terminal
+///
+/// If stdin is a terminal, an error message is printed and exit status set
+/// to EXIT_ERROR.
+extern bool is_tty_stdin(void);
+
+
+/// \brief Test if stdout is a terminal
+///
+/// If stdout is a terminal, an error message is printed and exit status set
+/// to EXIT_ERROR.
+extern bool is_tty_stdout(void);
diff --git a/src/xz/xz.1 b/src/xz/xz.1
new file mode 100644
index 0000000..2c24170
--- /dev/null
+++ b/src/xz/xz.1
@@ -0,0 +1,1707 @@
+'\" t
+.\"
+.\" Author: Lasse Collin
+.\"
+.\" This file has been put into the public domain.
+.\" You can do whatever you want with this file.
+.\"
+.TH XZ 1 "2010-06-15" "Tukaani" "XZ Utils"
+.SH NAME
+xz, unxz, xzcat, lzma, unlzma, lzcat \- Compress or decompress .xz and .lzma files
+.SH SYNOPSIS
+.B xz
+.RI [ option ]...
+.RI [ file ]...
+.PP
+.B unxz
+is equivalent to
+.BR "xz \-\-decompress" .
+.br
+.B xzcat
+is equivalent to
+.BR "xz \-\-decompress \-\-stdout" .
+.br
+.B lzma
+is equivalent to
+.BR "xz \-\-format=lzma" .
+.br
+.B unlzma
+is equivalent to
+.BR "xz \-\-format=lzma \-\-decompress" .
+.br
+.B lzcat
+is equivalent to
+.BR "xz \-\-format=lzma \-\-decompress \-\-stdout" .
+.PP
+When writing scripts that need to decompress files, it is recommended to
+always use the name
+.B xz
+with appropriate arguments
+.RB ( "xz \-d"
+or
+.BR "xz \-dc" )
+instead of the names
+.B unxz
+and
+.BR xzcat.
+.SH DESCRIPTION
+.B xz
+is a general-purpose data compression tool with command line syntax similar to
+.BR gzip (1)
+and
+.BR bzip2 (1).
+The native file format is the
+.B .xz
+format, but also the legacy
+.B .lzma
+format and raw compressed streams with no container format headers
+are supported.
+.PP
+.B xz
+compresses or decompresses each
+.I file
+according to the selected operation mode.
+If no
+.I files
+are given or
+.I file
+is
+.BR \- ,
+.B xz
+reads from standard input and writes the processed data to standard output.
+.B xz
+will refuse (display an error and skip the
+.IR file )
+to write compressed data to standard output if it is a terminal. Similarly,
+.B xz
+will refuse to read compressed data from standard input if it is a terminal.
+.PP
+Unless
+.B \-\-stdout
+is specified,
+.I files
+other than
+.B \-
+are written to a new file whose name is derived from the source
+.I file
+name:
+.IP \(bu 3
+When compressing, the suffix of the target file format
+.RB ( .xz
+or
+.BR .lzma )
+is appended to the source filename to get the target filename.
+.IP \(bu 3
+When decompressing, the
+.B .xz
+or
+.B .lzma
+suffix is removed from the filename to get the target filename.
+.B xz
+also recognizes the suffixes
+.B .txz
+and
+.BR .tlz ,
+and replaces them with the
+.B .tar
+suffix.
+.PP
+If the target file already exists, an error is displayed and the
+.I file
+is skipped.
+.PP
+Unless writing to standard output,
+.B xz
+will display a warning and skip the
+.I file
+if any of the following applies:
+.IP \(bu 3
+.I File
+is not a regular file. Symbolic links are not followed, thus they
+are not considered to be regular files.
+.IP \(bu 3
+.I File
+has more than one hard link.
+.IP \(bu 3
+.I File
+has setuid, setgid, or sticky bit set.
+.IP \(bu 3
+The operation mode is set to compress, and the
+.I file
+already has a suffix of the target file format
+.RB ( .xz
+or
+.B .txz
+when compressing to the
+.B .xz
+format, and
+.B .lzma
+or
+.B .tlz
+when compressing to the
+.B .lzma
+format).
+.IP \(bu 3
+The operation mode is set to decompress, and the
+.I file
+doesn't have a suffix of any of the supported file formats
+.RB ( .xz ,
+.BR .txz ,
+.BR .lzma ,
+or
+.BR .tlz ).
+.PP
+After successfully compressing or decompressing the
+.IR file ,
+.B xz
+copies the owner, group, permissions, access time, and modification time
+from the source
+.I file
+to the target file. If copying the group fails, the permissions are modified
+so that the target file doesn't become accessible to users who didn't have
+permission to access the source
+.IR file .
+.B xz
+doesn't support copying other metadata like access control lists
+or extended attributes yet.
+.PP
+Once the target file has been successfully closed, the source
+.I file
+is removed unless
+.B \-\-keep
+was specified. The source
+.I file
+is never removed if the output is written to standard output.
+.PP
+Sending
+.B SIGINFO
+or
+.B SIGUSR1
+to the
+.B xz
+process makes it print progress information to standard error.
+This has only limited use since when standard error is a terminal, using
+.B \-\-verbose
+will display an automatically updating progress indicator.
+.SS "Memory usage"
+The memory usage of
+.B xz
+varies from a few hundred kilobytes to several gigabytes depending on
+the compression settings. The settings used when compressing a file
+affect also the memory usage of the decompressor. Typically the decompressor
+needs only 5\ % to 20\ % of the amount of RAM that the compressor needed when
+creating the file. Still, the worst-case memory usage of the decompressor
+is several gigabytes.
+.PP
+To prevent uncomfortable surprises caused by huge memory usage,
+.B xz
+has a built-in memory usage limiter. While some operating systems provide
+ways to limit the memory usage of processes, relying on it wasn't deemed
+to be flexible enough. The default limit depends on the total amount of
+physical RAM:
+.IP \(bu 3
+If 40\ % of RAM is at least 80 MiB, 40\ % of RAM is used as the limit.
+.IP \(bu 3
+If 80\ % of RAM is less than 80 MiB, 80\ % of RAM is used as the limit.
+.IP \(bu 3
+Otherwise 80 MiB is used as the limit.
+.PP
+When compressing, if the selected compression settings exceed the memory
+usage limit, the settings are automatically adjusted downwards and a notice
+about this is displayed. As an exception, if the memory usage limit is
+exceeded when compressing with
+.B \-\-format=raw
+or
+.BR \-\-no\-adjust ,
+an error is displayed and
+.B xz
+will exit with exit status
+.BR 1 .
+.PP
+If source
+.I file
+cannot be decompressed without exceeding the memory usage limit, an error
+message is displayed and the file is skipped. Note that compressed files
+may contain many blocks, which may have been compressed with different
+settings. Typically all blocks will have roughly the same memory requirements,
+but it is possible that a block later in the file will exceed the memory usage
+limit, and an error about too low memory usage limit gets displayed after some
+data has already been decompressed.
+.PP
+The absolute value of the active memory usage limit can be seen with
+.B \-\-info-memory
+or near the bottom of the output of
+.BR \-\-long\-help .
+The default limit can be overridden with
+\fB\-\-memory=\fIlimit\fR.
+.SS Concatenation and padding with .xz files
+It is possible to concatenate
+.B .xz
+files as is.
+.B xz
+will decompress such files as if they were a single
+.B .xz
+file.
+.PP
+It is possible to insert padding between the concenated parts
+or after the last part. The padding must be null bytes and the size
+of the padding must be a multiple of four bytes. This can be useful
+if the .xz file is stored on a medium that stores file sizes
+e.g. as 512-byte blocks.
+.PP
+Concatenation and padding are not allowed with
+.B .lzma
+files or raw streams.
+.SH OPTIONS
+.SS "Integer suffixes and special values"
+In most places where an integer argument is expected, an optional suffix
+is supported to easily indicate large integers. There must be no space
+between the integer and the suffix.
+.TP
+.B KiB
+The integer is multiplied by 1,024 (2^10). Also
+.BR Ki ,
+.BR k ,
+.BR kB ,
+.BR K ,
+and
+.B KB
+are accepted as synonyms for
+.BR KiB .
+.TP
+.B MiB
+The integer is multiplied by 1,048,576 (2^20). Also
+.BR Mi ,
+.BR m ,
+.BR M ,
+and
+.B MB
+are accepted as synonyms for
+.BR MiB .
+.TP
+.B GiB
+The integer is multiplied by 1,073,741,824 (2^30). Also
+.BR Gi ,
+.BR g ,
+.BR G ,
+and
+.B GB
+are accepted as synonyms for
+.BR GiB .
+.PP
+A special value
+.B max
+can be used to indicate the maximum integer value supported by the option.
+.SS "Operation mode"
+If multiple operation mode options are given, the last one takes effect.
+.TP
+.BR \-z ", " \-\-compress
+Compress. This is the default operation mode when no operation mode option
+is specified, and no other operation mode is implied from the command name
+(for example,
+.B unxz
+implies
+.BR \-\-decompress ).
+.TP
+.BR \-d ", " \-\-decompress ", " \-\-uncompress
+Decompress.
+.TP
+.BR \-t ", " \-\-test
+Test the integrity of compressed
+.IR files .
+No files are created or removed. This option is equivalent to
+.B "\-\-decompress \-\-stdout"
+except that the decompressed data is discarded instead of being
+written to standard output.
+.TP
+.BR \-l ", " \-\-list
+List information about compressed
+.IR files .
+No uncompressed output is produced, and no files are created or removed.
+In list mode, the program cannot read the compressed data from standard
+input or from other unseekable sources.
+.IP
+The default listing shows basic information about
+.IR files ,
+one file per line. To get more detailed information, use also the
+.B \-\-verbose
+option. For even more information, use
+.B \-\-verbose
+twice, but note that it may be slow, because getting all the extra
+information requires many seeks. The width of verbose output exceeds
+80 characters, so piping the output to e.g.
+.B "less\ \-S"
+may be convenient if the terminal isn't wide enough.
+.IP
+The exact output may vary between
+.B xz
+versions and different locales. To get machine-readable output,
+.B \-\-robot \-\-list
+should be used.
+.SS "Operation modifiers"
+.TP
+.BR \-k ", " \-\-keep
+Keep (don't delete) the input files.
+.TP
+.BR \-f ", " \-\-force
+This option has several effects:
+.RS
+.IP \(bu 3
+If the target file already exists, delete it before compressing or
+decompressing.
+.IP \(bu 3
+Compress or decompress even if the input is a symbolic link to a regular file,
+has more than one hard link, or has setuid, setgid, or sticky bit set.
+The setuid, setgid, and sticky bits are not copied to the target file.
+.IP \(bu 3
+If combined with
+.B \-\-decompress
+.BR \-\-stdout
+and
+.B xz
+doesn't recognize the type of the source file,
+.B xz
+will copy the source file as is to standard output. This allows using
+.B xzcat
+.B \--force
+like
+.BR cat (1)
+for files that have not been compressed with
+.BR xz .
+Note that in future,
+.B xz
+might support new compressed file formats, which may make
+.B xz
+decompress more types of files instead of copying them as is to
+standard output.
+.BI \-\-format= format
+can be used to restrict
+.B xz
+to decompress only a single file format.
+.RE
+.TP
+.BR \-c ", " \-\-stdout ", " \-\-to-stdout
+Write the compressed or decompressed data to standard output instead of
+a file. This implies
+.BR \-\-keep .
+.TP
+.B \-\-no\-sparse
+Disable creation of sparse files. By default, if decompressing into
+a regular file,
+.B xz
+tries to make the file sparse if the decompressed data contains long
+sequences of binary zeros. It works also when writing to standard output
+as long as standard output is connected to a regular file, and certain
+additional conditions are met to make it safe. Creating sparse files may
+save disk space and speed up the decompression by reducing the amount of
+disk I/O.
+.TP
+\fB\-S\fR \fI.suf\fR, \fB\-\-suffix=\fI.suf
+When compressing, use
+.I .suf
+as the suffix for the target file instead of
+.B .xz
+or
+.BR .lzma .
+If not writing to standard output and the source file already has the suffix
+.IR .suf ,
+a warning is displayed and the file is skipped.
+.IP
+When decompressing, recognize also files with the suffix
+.I .suf
+in addition to files with the
+.BR .xz ,
+.BR .txz ,
+.BR .lzma ,
+or
+.B .tlz
+suffix. If the source file has the suffix
+.IR .suf ,
+the suffix is removed to get the target filename.
+.IP
+When compressing or decompressing raw streams
+.RB ( \-\-format=raw ),
+the suffix must always be specified unless writing to standard output,
+because there is no default suffix for raw streams.
+.TP
+\fB\-\-files\fR[\fB=\fIfile\fR]
+Read the filenames to process from
+.IR file ;
+if
+.I file
+is omitted, filenames are read from standard input. Filenames must be
+terminated with the newline character. A dash
+.RB ( \- )
+is taken as a regular filename; it doesn't mean standard input.
+If filenames are given also as command line arguments, they are
+processed before the filenames read from
+.IR file .
+.TP
+\fB\-\-files0\fR[\fB=\fIfile\fR]
+This is identical to \fB\-\-files\fR[\fB=\fIfile\fR] except that the
+filenames must be terminated with the null character.
+.SS "Basic file format and compression options"
+.TP
+\fB\-F\fR \fIformat\fR, \fB\-\-format=\fIformat
+Specify the file format to compress or decompress:
+.RS
+.IP \(bu 3
+.BR auto :
+This is the default. When compressing,
+.B auto
+is equivalent to
+.BR xz .
+When decompressing, the format of the input file is automatically detected.
+Note that raw streams (created with
+.BR \-\-format=raw )
+cannot be auto-detected.
+.IP \(bu 3
+.BR xz :
+Compress to the
+.B .xz
+file format, or accept only
+.B .xz
+files when decompressing.
+.IP \(bu 3
+.B lzma
+or
+.BR alone :
+Compress to the legacy
+.B .lzma
+file format, or accept only
+.B .lzma
+files when decompressing. The alternative name
+.B alone
+is provided for backwards compatibility with LZMA Utils.
+.IP \(bu 3
+.BR raw :
+Compress or uncompress a raw stream (no headers). This is meant for advanced
+users only. To decode raw streams, you need to set not only
+.B \-\-format=raw
+but also specify the filter chain, which would normally be stored in the
+container format headers.
+.RE
+.TP
+\fB\-C\fR \fIcheck\fR, \fB\-\-check=\fIcheck
+Specify the type of the integrity check, which is calculated from the
+uncompressed data. This option has an effect only when compressing into the
+.B .xz
+format; the
+.B .lzma
+format doesn't support integrity checks.
+The integrity check (if any) is verified when the
+.B .xz
+file is decompressed.
+.IP
+Supported
+.I check
+types:
+.RS
+.IP \(bu 3
+.BR none :
+Don't calculate an integrity check at all. This is usually a bad idea. This
+can be useful when integrity of the data is verified by other means anyway.
+.IP \(bu 3
+.BR crc32 :
+Calculate CRC32 using the polynomial from IEEE-802.3 (Ethernet).
+.IP \(bu 3
+.BR crc64 :
+Calculate CRC64 using the polynomial from ECMA-182. This is the default, since
+it is slightly better than CRC32 at detecting damaged files and the speed
+difference is negligible.
+.IP \(bu 3
+.BR sha256 :
+Calculate SHA-256. This is somewhat slower than CRC32 and CRC64.
+.RE
+.IP
+Integrity of the
+.B .xz
+headers is always verified with CRC32. It is not possible to change or
+disable it.
+.TP
+.BR \-0 " ... " \-9
+Select compression preset. If a preset level is specified multiple times,
+the last one takes effect.
+.IP
+The compression preset levels can be categorised roughly into three
+categories:
+.RS
+.IP "\fB\-0\fR ... \fB\-2"
+Fast presets with relatively low memory usage.
+.B \-1
+and
+.B \-2
+should give compression speed and ratios comparable to
+.B "bzip2 \-1"
+and
+.BR "bzip2 \-9" ,
+respectively.
+Currently
+.B \-0
+is not very good (not much faster than
+.B \-1
+but much worse compression). In future,
+.B \-0
+may be indicate some fast algorithm instead of LZMA2.
+.IP "\fB\-3\fR ... \fB\-5"
+Good compression ratio with low to medium memory usage.
+These are significantly slower than levels 0\-2.
+.IP "\fB\-6\fR ... \fB\-9"
+Excellent compression with medium to high memory usage. These are also
+slower than the lower preset levels. The default is
+.BR \-6 .
+Unless you want to maximize the compression ratio, you probably don't want
+a higher preset level than
+.B \-7
+due to speed and memory usage.
+.RE
+.IP
+The exact compression settings (filter chain) used by each preset may
+vary between
+.B xz
+versions. The settings may also vary between files being compressed, if
+.B xz
+determines that modified settings will probably give better compression
+ratio without significantly affecting compression time or memory usage.
+.IP
+Because the settings may vary, the memory usage may vary too. The following
+table lists the maximum memory usage of each preset level, which won't be
+exceeded even in future versions of
+.BR xz .
+.IP
+.B "FIXME: The table below is just a rough idea."
+.RS
+.RS
+.TS
+tab(;);
+c c c
+n n n.
+Preset;Compression;Decompression
+\-0;6 MiB;1 MiB
+\-1;6 MiB;1 MiB
+\-2;10 MiB;1 MiB
+\-3;20 MiB;2 MiB
+\-4;30 MiB;3 MiB
+\-5;60 MiB;6 MiB
+\-6;100 MiB;10 MiB
+\-7;200 MiB;20 MiB
+\-8;400 MiB;40 MiB
+\-9;800 MiB;80 MiB
+.TE
+.RE
+.RE
+.IP
+When compressing,
+.B xz
+automatically adjusts the compression settings downwards if
+the memory usage limit would be exceeded, so it is safe to specify
+a high preset level even on systems that don't have lots of RAM.
+.TP
+.BR \-\-fast " and " \-\-best
+These are somewhat misleading aliases for
+.B \-0
+and
+.BR \-9 ,
+respectively.
+These are provided only for backwards compatibility with LZMA Utils.
+Avoid using these options.
+.IP
+Especially the name of
+.B \-\-best
+is misleading, because the definition of best depends on the input data,
+and that usually people don't want the very best compression ratio anyway,
+because it would be very slow.
+.TP
+.BR \-e ", " \-\-extreme
+Modify the compression preset (\fB\-0\fR ... \fB\-9\fR) so that a little bit
+better compression ratio can be achieved without increasing memory usage
+of the compressor or decompressor (exception: compressor memory usage may
+increase a little with presets \fB\-0\fR ... \fB\-2\fR). The downside is that
+the compression time will increase dramatically (it can easily double).
+.TP
+.B \-\-no\-adjust
+Display an error and exit if the compression settings exceed the
+the memory usage limit. The default is to adjust the settings downwards so
+that the memory usage limit is not exceeded. Automatic adjusting is
+always disabled when creating raw streams
+.RB ( \-\-format=raw ).
+.TP
+\fB\-M\fR \fIlimit\fR, \fB\-\-memory=\fIlimit
+Set the memory usage limit. If this option is specified multiple times,
+the last one takes effect. The
+.I limit
+can be specified in multiple ways:
+.RS
+.IP \(bu 3
+The
+.I limit
+can be an absolute value in bytes. Using an integer suffix like
+.B MiB
+can be useful. Example:
+.B "\-\-memory=80MiB"
+.IP \(bu 3
+The
+.I limit
+can be specified as a percentage of physical RAM. Example:
+.B "\-\-memory=70%"
+.IP \(bu 3
+The
+.I limit
+can be reset back to its default value by setting it to
+.BR 0 .
+See the section
+.B "Memory usage"
+for how the default limit is defined.
+.IP \(bu 3
+The memory usage limiting can be effectively disabled by setting
+.I limit
+to
+.BR max .
+This isn't recommended. It's usually better to use, for example,
+.BR \-\-memory=90% .
+.RE
+.IP
+The current
+.I limit
+can be seen near the bottom of the output of the
+.B \-\-long-help
+option.
+.TP
+\fB\-T\fR \fIthreads\fR, \fB\-\-threads=\fIthreads
+Specify the maximum number of worker threads to use. The default is
+the number of available CPU cores. You can see the current value of
+.I threads
+near the end of the output of the
+.B \-\-long\-help
+option.
+.IP
+The actual number of worker threads can be less than
+.I threads
+if using more threads would exceed the memory usage limit.
+In addition to CPU-intensive worker threads,
+.B xz
+may use a few auxiliary threads, which don't use a lot of CPU time.
+.IP
+.B "Multithreaded compression and decompression are not implemented yet,"
+.B "so this option has no effect for now."
+.SS Custom compressor filter chains
+A custom filter chain allows specifying the compression settings in detail
+instead of relying on the settings associated to the preset levels.
+When a custom filter chain is specified, the compression preset level options
+(\fB\-0\fR ... \fB\-9\fR and \fB\-\-extreme\fR) are silently ignored.
+.PP
+A filter chain is comparable to piping on the UN*X command line.
+When compressing, the uncompressed input goes to the first filter, whose
+output goes to the next filter (if any). The output of the last filter
+gets written to the compressed file. The maximum number of filters in
+the chain is four, but typically a filter chain has only one or two filters.
+.PP
+Many filters have limitations where they can be in the filter chain:
+some filters can work only as the last filter in the chain, some only
+as a non-last filter, and some work in any position in the chain. Depending
+on the filter, this limitation is either inherent to the filter design or
+exists to prevent security issues.
+.PP
+A custom filter chain is specified by using one or more filter options in
+the order they are wanted in the filter chain. That is, the order of filter
+options is significant! When decoding raw streams
+.RB ( \-\-format=raw ),
+the filter chain is specified in the same order as it was specified when
+compressing.
+.PP
+Filters take filter-specific
+.I options
+as a comma-separated list. Extra commas in
+.I options
+are ignored. Every option has a default value, so you need to
+specify only those you want to change.
+.TP
+\fB\-\-lzma1\fR[\fB=\fIoptions\fR], \fB\-\-lzma2\fR[\fB=\fIoptions\fR]
+Add LZMA1 or LZMA2 filter to the filter chain. These filter can be used
+only as the last filter in the chain.
+.IP
+LZMA1 is a legacy filter, which is supported almost solely due to the legacy
+.B .lzma
+file format, which supports only LZMA1. LZMA2 is an updated
+version of LZMA1 to fix some practical issues of LZMA1. The
+.B .xz
+format uses LZMA2, and doesn't support LZMA1 at all. Compression speed and
+ratios of LZMA1 and LZMA2 are practically the same.
+.IP
+LZMA1 and LZMA2 share the same set of
+.IR options :
+.RS
+.TP
+.BI preset= preset
+Reset all LZMA1 or LZMA2
+.I options
+to
+.IR preset .
+.I Preset
+consist of an integer, which may be followed by single-letter preset
+modifiers. The integer can be from
+.B 0
+to
+.BR 9 ,
+matching the command line options \fB\-0\fR ... \fB\-9\fR.
+The only supported modifier is currently
+.BR e ,
+which matches
+.BR \-\-extreme .
+.IP
+The default
+.I preset
+is
+.BR 6 ,
+from which the default values for the rest of the LZMA1 or LZMA2
+.I options
+are taken.
+.TP
+.BI dict= size
+Dictionary (history buffer) size indicates how many bytes of the recently
+processed uncompressed data is kept in memory. One method to reduce size of
+the uncompressed data is to store distance-length pairs, which
+indicate what data to repeat from the dictionary buffer. The bigger
+the dictionary, the better the compression ratio usually is,
+but dictionaries bigger than the uncompressed data are waste of RAM.
+.IP
+Typical dictionary size is from 64 KiB to 64 MiB. The minimum is 4 KiB.
+The maximum for compression is currently 1.5 GiB. The decompressor already
+supports dictionaries up to one byte less than 4 GiB, which is the
+maximum for LZMA1 and LZMA2 stream formats.
+.IP
+Dictionary size has the biggest effect on compression ratio.
+Dictionary size and match finder together determine the memory usage of
+the LZMA1 or LZMA2 encoder. The same dictionary size is required
+for decompressing that was used when compressing, thus the memory usage of
+the decoder is determined by the dictionary size used when compressing.
+.TP
+.BI lc= lc
+Specify the number of literal context bits. The minimum is
+.B 0
+and the maximum is
+.BR 4 ;
+the default is
+.BR 3 .
+In addition, the sum of
+.I lc
+and
+.I lp
+must not exceed
+.BR 4 .
+.TP
+.BI lp= lp
+Specify the number of literal position bits. The minimum is
+.B 0
+and the maximum is
+.BR 4 ;
+the default is
+.BR 0 .
+.TP
+.BI pb= pb
+Specify the number of position bits. The minimum is
+.B 0
+and the maximum is
+.BR 4 ;
+the default is
+.BR 2 .
+.TP
+.BI mode= mode
+Compression
+.I mode
+specifies the function used to analyze the data produced by the match finder.
+Supported
+.I modes
+are
+.B fast
+and
+.BR normal .
+The default is
+.B fast
+for
+.I presets
+.BR 0 \- 2
+and
+.B normal
+for
+.I presets
+.BR 3 \- 9 .
+.TP
+.BI mf= mf
+Match finder has a major effect on encoder speed, memory usage, and
+compression ratio. Usually Hash Chain match finders are faster than
+Binary Tree match finders. Hash Chains are usually used together with
+.B mode=fast
+and Binary Trees with
+.BR mode=normal .
+The memory usage formulas are only rough estimates,
+which are closest to reality when
+.I dict
+is a power of two.
+.RS
+.TP
+.B hc3
+Hash Chain with 2- and 3-byte hashing
+.br
+Minimum value for
+.IR nice :
+3
+.br
+Memory usage:
+.I dict
+* 7.5 (if
+.I dict
+<= 16 MiB);
+.br
+.I dict
+* 5.5 + 64 MiB (if
+.I dict
+> 16 MiB)
+.TP
+.B hc4
+Hash Chain with 2-, 3-, and 4-byte hashing
+.br
+Minimum value for
+.IR nice :
+4
+.br
+Memory usage:
+.I dict
+* 7.5
+.TP
+.B bt2
+Binary Tree with 2-byte hashing
+.br
+Minimum value for
+.IR nice :
+2
+.br
+Memory usage:
+.I dict
+* 9.5
+.TP
+.B bt3
+Binary Tree with 2- and 3-byte hashing
+.br
+Minimum value for
+.IR nice :
+3
+.br
+Memory usage:
+.I dict
+* 11.5 (if
+.I dict
+<= 16 MiB);
+.br
+.I dict
+* 9.5 + 64 MiB (if
+.I dict
+> 16 MiB)
+.TP
+.B bt4
+Binary Tree with 2-, 3-, and 4-byte hashing
+.br
+Minimum value for
+.IR nice :
+4
+.br
+Memory usage:
+.I dict
+* 11.5
+.RE
+.TP
+.BI nice= nice
+Specify what is considered to be a nice length for a match. Once a match
+of at least
+.I nice
+bytes is found, the algorithm stops looking for possibly better matches.
+.IP
+.I nice
+can be 2\-273 bytes. Higher values tend to give better compression ratio
+at expense of speed. The default depends on the
+.I preset
+level.
+.TP
+.BI depth= depth
+Specify the maximum search depth in the match finder. The default is the
+special value
+.BR 0 ,
+which makes the compressor determine a reasonable
+.I depth
+from
+.I mf
+and
+.IR nice .
+.IP
+Using very high values for
+.I depth
+can make the encoder extremely slow with carefully crafted files.
+Avoid setting the
+.I depth
+over 1000 unless you are prepared to interrupt the compression in case it
+is taking too long.
+.RE
+.IP
+When decoding raw streams
+.RB ( \-\-format=raw ),
+LZMA2 needs only the value of
+.BR dict .
+LZMA1 needs also
+.BR lc ,
+.BR lp ,
+and
+.BR pb.
+.TP
+\fB\-\-x86\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-powerpc\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-ia64\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-arm\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-armthumb\fR[\fB=\fIoptions\fR]
+.TP
+\fB\-\-sparc\fR[\fB=\fIoptions\fR]
+Add a branch/call/jump (BCJ) filter to the filter chain. These filters
+can be used only as non-last filter in the filter chain.
+.IP
+A BCJ filter converts relative addresses in the machine code to their
+absolute counterparts. This doesn't change the size of the data, but
+it increases redundancy, which allows e.g. LZMA2 to get better
+compression ratio.
+.IP
+The BCJ filters are always reversible, so using a BCJ filter for wrong
+type of data doesn't cause any data loss. However, applying a BCJ filter
+for wrong type of data is a bad idea, because it tends to make the
+compression ratio worse.
+.IP
+Different instruction sets have have different alignment:
+.RS
+.RS
+.TS
+tab(;);
+l n l
+l n l.
+Filter;Alignment;Notes
+x86;1;32-bit and 64-bit x86
+PowerPC;4;Big endian only
+ARM;4;Little endian only
+ARM-Thumb;2;Little endian only
+IA-64;16;Big or little endian
+SPARC;4;Big or little endian
+.TE
+.RE
+.RE
+.IP
+Since the BCJ-filtered data is usually compressed with LZMA2, the compression
+ratio may be improved slightly if the LZMA2 options are set to match the
+alignment of the selected BCJ filter. For example, with the IA-64 filter,
+it's good to set
+.B pb=4
+with LZMA2 (2^4=16). The x86 filter is an exception; it's usually good to
+stick to LZMA2's default four-byte alignment when compressing x86 executables.
+.IP
+All BCJ filters support the same
+.IR options :
+.RS
+.TP
+.BI start= offset
+Specify the start
+.I offset
+that is used when converting between relative and absolute addresses.
+The
+.I offset
+must be a multiple of the alignment of the filter (see the table above).
+The default is zero. In practice, the default is good; specifying
+a custom
+.I offset
+is almost never useful.
+.IP
+Specifying a non-zero start
+.I offset
+is probably useful only if the executable has multiple sections, and there
+are many cross-section jumps or calls. Applying a BCJ filter separately for
+each section with proper start offset and then compressing the result as
+a single chunk may give some improvement in compression ratio compared
+to applying the BCJ filter with the default
+.I offset
+for the whole executable.
+.RE
+.TP
+\fB\-\-delta\fR[\fB=\fIoptions\fR]
+Add Delta filter to the filter chain. The Delta filter
+can be used only as non-last filter in the filter chain.
+.IP
+Currently only simple byte-wise delta calculation is supported. It can
+be useful when compressing e.g. uncompressed bitmap images or uncompressed
+PCM audio. However, special purpose algorithms may give significantly better
+results than Delta + LZMA2. This is true especially with audio, which
+compresses faster and better e.g. with FLAC.
+.IP
+Supported
+.IR options :
+.RS
+.TP
+.BI dist= distance
+Specify the
+.I distance
+of the delta calculation as bytes.
+.I distance
+must be 1\-256. The default is 1.
+.IP
+For example, with
+.B dist=2
+and eight-byte input A1 B1 A2 B3 A3 B5 A4 B7, the output will be
+A1 B1 01 02 01 02 01 02.
+.RE
+.SS "Other options"
+.TP
+.BR \-q ", " \-\-quiet
+Suppress warnings and notices. Specify this twice to suppress errors too.
+This option has no effect on the exit status. That is, even if a warning
+was suppressed, the exit status to indicate a warning is still used.
+.TP
+.BR \-v ", " \-\-verbose
+Be verbose. If standard error is connected to a terminal,
+.B xz
+will display a progress indicator.
+Specifying
+.B \-\-verbose
+twice will give even more verbose output (useful mostly for debugging).
+.IP
+The progress indicator shows the following information:
+.RS
+.IP \(bu 3
+Completion percentage is shown if the size of the input file is known.
+That is, percentage cannot be shown in pipes.
+.IP \(bu 3
+Amount of compressed data produced (compressing) or consumed (decompressing).
+.IP \(bu 3
+Amount of uncompressed data consumed (compressing) or produced
+(decompressing).
+.IP \(bu 3
+Compression ratio, which is calculated by dividing the amount of
+compressed data processed so far by the amount of uncompressed data
+processed so far.
+.IP \(bu 3
+Compression or decompression speed. This is measured as the amount of
+uncompressed data consumed (compression) or produced (decompression)
+per second. It is shown once a few seconds have passed since
+.B xz
+started processing the file.
+.IP \(bu 3
+Elapsed time or estimated time remaining.
+Elapsed time is displayed in the format M:SS or H:MM:SS.
+The estimated remaining time is displayed in a less precise format
+which never has colons, for example, 2 min 30 s. The estimate can
+be shown only when the size of the input file is known and a couple of
+seconds have already passed since
+.B xz
+started processing the file.
+.RE
+.IP
+When standard error is not a terminal,
+.B \-\-verbose
+will make
+.B xz
+print the filename, compressed size, uncompressed size, compression ratio,
+speed, and elapsed time on a single line to standard error after
+compressing or decompressing the file. If operating took at least a few
+seconds, also the speed and elapsed time are printed. If the operation
+didn't finish, for example due to user interruption, also the completion
+percentage is printed if the size of the input file is known.
+.TP
+.BR \-Q ", " \-\-no\-warn
+Don't set the exit status to
+.B 2
+even if a condition worth a warning was detected. This option doesn't affect
+the verbosity level, thus both
+.B \-\-quiet
+and
+.B \-\-no\-warn
+have to be used to not display warnings and to not alter the exit status.
+.TP
+.B \-\-robot
+Print messages in a machine-parsable format. This is intended to ease
+writing frontends that want to use
+.B xz
+instead of liblzma, which may be the case with various scripts. The output
+with this option enabled is meant to be stable across
+.B xz
+releases. See the section
+.B "ROBOT MODE"
+for details.
+.TP
+.BR \-\-info-memory
+Display the current memory usage limit in human-readable format on
+a single line, and exit successfully. To see how much RAM
+.B xz
+thinks your system has, use
+.BR "\-\-memory=100% \-\-info\-memory" .
+.TP
+.BR \-h ", " \-\-help
+Display a help message describing the most commonly used options,
+and exit successfully.
+.TP
+.BR \-H ", " \-\-long\-help
+Display a help message describing all features of
+.BR xz ,
+and exit successfully
+.TP
+.BR \-V ", " \-\-version
+Display the version number of
+.B xz
+and liblzma in human readable format. To get machine-parsable output, specify
+.B \-\-robot
+before
+.BR \-\-version .
+.SH ROBOT MODE
+The robot mode is activated with the
+.B \-\-robot
+option. It makes the output of
+.B xz
+easier to parse by other programs. Currently
+.B \-\-robot
+is supported only together with
+.BR \-\-version ,
+.BR \-\-info-memory ,
+and
+.BR \-\-list .
+It will be supported for normal compression and decompression in the future.
+.PP
+.SS Version
+.B "xz \-\-robot \-\-version"
+will print the version number of
+.B xz
+and liblzma in the following format:
+.PP
+.BI XZ_VERSION= XYYYZZZS
+.br
+.BI LIBLZMA_VERSION= XYYYZZZS
+.TP
+.I X
+Major version.
+.TP
+.I YYY
+Minor version. Even numbers are stable.
+Odd numbers are alpha or beta versions.
+.TP
+.I ZZZ
+Patch level for stable releases or just a counter for development releases.
+.TP
+.I S
+Stability.
+.B 0
+is alpha,
+.B 1
+is beta, and
+.B 2
+is stable.
+.I S
+should be always
+.B 2
+when
+.I YYY
+is even.
+.PP
+.I XYYYZZZS
+are the same on both lines if
+.B xz
+and liblzma are from the same XZ Utils release.
+.PP
+Examples: 4.999.9beta is
+.B 49990091
+and
+5.0.0 is
+.BR 50000002 .
+.SS Memory limit information
+.B "xz \-\-robot \-\-info-memory"
+prints the current memory usage limit as bytes on a single line.
+To get the total amount of installed RAM, use
+.BR "xz \-\-robot \-\-memory=100% \-\-info-memory" .
+.SS List mode
+.B "xz \-\-robot \-\-list"
+uses tab-separated output. The first column of every line has a string
+that indicates the type of the information found on that line:
+.TP
+.B name
+This is always the first line when starting to list a file. The second
+column on the line is the filename.
+.TP
+.B file
+This line contains overall information about the
+.B .xz
+file. This line is always printed after the
+.B name
+line.
+.TP
+.B stream
+This line type is used only when
+.B \-\-verbose
+was specified. There are as many
+.B stream
+lines as there are streams in the
+.B .xz
+file.
+.TP
+.B block
+This line type is used only when
+.B \-\-verbose
+was specified. There are as many
+.B block
+lines as there are blocks in the
+.B .xz
+file. The
+.B block
+lines are shown after all the
+.B stream
+lines; different line types are not interleaved.
+.TP
+.B summary
+This line type is used only when
+.B \-\-verbose
+was specified twice. This line is printed after all
+.B block
+lines. Like the
+.B file
+line, the
+.B summary
+line contains overall information about the
+.B .xz
+file.
+.TP
+.B totals
+This line is always the very last line of the list output. It shows
+the total counts and sizes.
+.PP
+The columns of the
+.B file
+lines:
+.RS
+.IP 2. 4
+Number of streams in the file
+.IP 3. 4
+Total number of blocks in the stream(s)
+.IP 4. 4
+Compressed size of the file
+.IP 5. 4
+Uncompressed size of the file
+.IP 6. 4
+Compression ratio, for example
+.BR 0.123.
+If ratio is over 9.999, three dashes
+.RB ( \-\-\- )
+are displayed instead of the ratio.
+.IP 7. 4
+Comma-separated list of integrity check names. The following strings are
+used for the known check types:
+.BR None ,
+.BR CRC32 ,
+.BR CRC64 ,
+and
+.BR SHA\-256 .
+For unknown check types,
+.BI Unknown\- N
+is used, where
+.I N
+is the Check ID as a decimal number (one or two digits).
+.IP 8. 4
+Total size of stream padding in the file
+.RE
+.PP
+The columns of the
+.B stream
+lines:
+.RS
+.IP 2. 4
+Stream number (the first stream is 1)
+.IP 3. 4
+Number of blocks in the stream
+.IP 4. 4
+Compressed start offset
+.IP 5. 4
+Uncompressed start offset
+.IP 6. 4
+Compressed size (does not include stream padding)
+.IP 7. 4
+Uncompressed size
+.IP 8. 4
+Compression ratio
+.IP 9. 4
+Name of the integrity check
+.IP 10. 4
+Size of stream padding
+.RE
+.PP
+The columns of the
+.B block
+lines:
+.RS
+.IP 2. 4
+Number of the stream containing this block
+.IP 3. 4
+Block number relative to the beginning of the stream (the first block is 1)
+.IP 4. 4
+Block number relative to the beginning of the file
+.IP 5. 4
+Compressed start offset relative to the beginning of the file
+.IP 6. 4
+Uncompressed start offset relative to the beginning of the file
+.IP 7. 4
+Total compressed size of the block (includes headers)
+.IP 8. 4
+Uncompressed size
+.IP 9. 4
+Compression ratio
+.IP 10. 4
+Name of the integrity check
+.RE
+.PP
+If
+.B \-\-verbose
+was specified twice, additional columns are included on the
+.B block
+lines. These are not displayed with a single
+.BR \-\-verbose ,
+because getting this information requires many seeks and can thus be slow:
+.RS
+.IP 11. 4
+Value of the integrity check in hexadecimal
+.IP 12. 4
+Block header size
+.IP 13. 4
+Block flags:
+.B c
+indicates that compressed size is present, and
+.B u
+indicates that uncompressed size is present.
+If the flag is not set, a dash
+.RB ( \- )
+is shown instead to keep the string length fixed. New flags may be added
+to the end of the string in the future.
+.IP 14. 4
+Size of the actual compressed data in the block (this excludes
+the block header, block padding, and check fields)
+.IP 15. 4
+Amount of memory (as bytes) required to decompress this block with this
+.B xz
+version
+.IP 16. 4
+Filter chain. Note that most of the options used at compression time cannot
+be known, because only the options that are needed for decompression are
+stored in the
+.B .xz
+headers.
+.RE
+.PP
+The columns of the
+.B totals
+line:
+.RS
+.IP 2. 4
+Number of streams
+.IP 3. 4
+Number of blocks
+.IP 4. 4
+Compressed size
+.IP 5. 4
+Uncompressed size
+.IP 6. 4
+Average compression ratio
+.IP 7. 4
+Comma-separated list of integrity check names that were present in the files
+.IP 8. 4
+Stream padding size
+.IP 9. 4
+Number of files. This is here to keep the order of the earlier columns
+the same as on
+.B file
+lines.
+.RE
+.PP
+If
+.B \-\-verbose
+was specified twice, additional columns are included on the
+.B totals
+line:
+.RS
+.IP 10. 4
+Maximum amount of memory (as bytes) required to decompress the files
+with this
+.B xz
+version
+.IP 11. 4
+.B yes
+or
+.B no
+indicating if all block headers have both compressed size and
+uncompressed size stored in them
+.RE
+.PP
+Future versions may add new line types and new columns can be added to
+the existing line types, but the existing columns won't be changed.
+.SH "EXIT STATUS"
+.TP
+.B 0
+All is good.
+.TP
+.B 1
+An error occurred.
+.TP
+.B 2
+Something worth a warning occurred, but no actual errors occurred.
+.PP
+Notices (not warnings or errors) printed on standard error don't affect
+the exit status.
+.SH ENVIRONMENT
+.TP
+.B XZ_OPT
+A space-separated list of options is parsed from
+.B XZ_OPT
+before parsing the options given on the command line. Note that only
+options are parsed from
+.BR XZ_OPT ;
+all non-options are silently ignored. Parsing is done with
+.BR getopt_long (3)
+which is used also for the command line arguments.
+.SH "LZMA UTILS COMPATIBILITY"
+The command line syntax of
+.B xz
+is practically a superset of
+.BR lzma ,
+.BR unlzma ,
+and
+.BR lzcat
+as found from LZMA Utils 4.32.x. In most cases, it is possible to replace
+LZMA Utils with XZ Utils without breaking existing scripts. There are some
+incompatibilities though, which may sometimes cause problems.
+.SS "Compression preset levels"
+The numbering of the compression level presets is not identical in
+.B xz
+and LZMA Utils.
+The most important difference is how dictionary sizes are mapped to different
+presets. Dictionary size is roughly equal to the decompressor memory usage.
+.RS
+.TS
+tab(;);
+c c c
+c n n.
+Level;xz;LZMA Utils
+\-1;64 KiB;64 KiB
+\-2;512 KiB;1 MiB
+\-3;1 MiB;512 KiB
+\-4;2 MiB;1 MiB
+\-5;4 MiB;2 MiB
+\-6;8 MiB;4 MiB
+\-7;16 MiB;8 MiB
+\-8;32 MiB;16 MiB
+\-9;64 MiB;32 MiB
+.TE
+.RE
+.PP
+The dictionary size differences affect the compressor memory usage too,
+but there are some other differences between LZMA Utils and XZ Utils, which
+make the difference even bigger:
+.RS
+.TS
+tab(;);
+c c c
+c n n.
+Level;xz;LZMA Utils 4.32.x
+\-1;2 MiB;2 MiB
+\-2;5 MiB;12 MiB
+\-3;13 MiB;12 MiB
+\-4;25 MiB;16 MiB
+\-5;48 MiB;26 MiB
+\-6;94 MiB;45 MiB
+\-7;186 MiB;83 MiB
+\-8;370 MiB;159 MiB
+\-9;674 MiB;311 MiB
+.TE
+.RE
+.PP
+The default preset level in LZMA Utils is
+.B \-7
+while in XZ Utils it is
+.BR \-6 ,
+so both use 8 MiB dictionary by default.
+.SS "Streamed vs. non-streamed .lzma files"
+Uncompressed size of the file can be stored in the
+.B .lzma
+header. LZMA Utils does that when compressing regular files.
+The alternative is to mark that uncompressed size is unknown and
+use end of payload marker to indicate where the decompressor should stop.
+LZMA Utils uses this method when uncompressed size isn't known, which is
+the case for example in pipes.
+.PP
+.B xz
+supports decompressing
+.B .lzma
+files with or without end of payload marker, but all
+.B .lzma
+files created by
+.B xz
+will use end of payload marker and have uncompressed size marked as unknown
+in the
+.B .lzma
+header. This may be a problem in some (uncommon) situations. For example, a
+.B .lzma
+decompressor in an embedded device might work only with files that have known
+uncompressed size. If you hit this problem, you need to use LZMA Utils or
+LZMA SDK to create
+.B .lzma
+files with known uncompressed size.
+.SS "Unsupported .lzma files"
+The
+.B .lzma
+format allows
+.I lc
+values up to 8, and
+.I lp
+values up to 4. LZMA Utils can decompress files with any
+.I lc
+and
+.IR lp ,
+but always creates files with
+.B lc=3
+and
+.BR lp=0 .
+Creating files with other
+.I lc
+and
+.I lp
+is possible with
+.B xz
+and with LZMA SDK.
+.PP
+The implementation of the LZMA1 filter in liblzma requires
+that the sum of
+.I lc
+and
+.I lp
+must not exceed 4. Thus,
+.B .lzma
+files which exceed this limitation, cannot be decompressed with
+.BR xz .
+.PP
+LZMA Utils creates only
+.B .lzma
+files which have dictionary size of
+.RI "2^" n
+(a power of 2), but accepts files with any dictionary size.
+liblzma accepts only
+.B .lzma
+files which have dictionary size of
+.RI "2^" n
+or
+.RI "2^" n " + 2^(" n "\-1)."
+This is to decrease false positives when detecting
+.B .lzma
+files.
+.PP
+These limitations shouldn't be a problem in practice, since practically all
+.B .lzma
+files have been compressed with settings that liblzma will accept.
+.SS "Trailing garbage"
+When decompressing, LZMA Utils silently ignore everything after the first
+.B .lzma
+stream. In most situations, this is a bug. This also means that LZMA Utils
+don't support decompressing concatenated
+.B .lzma
+files.
+.PP
+If there is data left after the first
+.B .lzma
+stream,
+.B xz
+considers the file to be corrupt. This may break obscure scripts which have
+assumed that trailing garbage is ignored.
+.SH NOTES
+.SS Compressed output may vary
+The exact compressed output produced from the same uncompressed input file
+may vary between XZ Utils versions even if compression options are identical.
+This is because the encoder can be improved (faster or better compression)
+without affecting the file format. The output can vary even between different
+builds of the same XZ Utils version, if different build options are used.
+.PP
+The above means that implementing
+.B \-\-rsyncable
+to create rsyncable
+.B .xz
+files is not going to happen without freezing a part of the encoder
+implementation, which can then be used with
+.BR \-\-rsyncable .
+.SS Embedded .xz decompressors
+Embedded
+.B .xz
+decompressor implementations like XZ Embedded don't necessarily support files
+created with
+.I check
+types other than
+.B none
+and
+.BR crc32 .
+Since the default is \fB\-\-check=\fIcrc64\fR, you must use
+.B \-\-check=none
+or
+.B \-\-check=crc32
+when creating files for embedded systems.
+.PP
+Outside embedded systems, all
+.B .xz
+format decompressors support all the
+.I check
+types, or at least are able to decompress the file without verifying the
+integrity check if the particular
+.I check
+is not supported.
+.PP
+XZ Embedded supports BCJ filters, but only with the default start offset.
+.SH EXAMPLES
+.SS Basics
+A mix of compressed and uncompressed files can be decompressed
+to standard output with a single command:
+.IP
+.B "xz -dcf a.txt b.txt.xz c.txt d.txt.xz > abcd.txt"
+.SS Parallel compression of many files
+On GNU and *BSD,
+.BR find (1)
+and
+.BR xargs (1)
+can be used to parallellize compression of many files:
+.PP
+.IP
+.B "find . \-type f \e! \-name '*.xz' \-print0 | xargs \-0r \-P4 \-n16 xz"
+.PP
+The
+.B \-P
+option sets the number of parallel
+.B xz
+processes. The best value for the
+.B \-n
+option depends on how many files there are to be compressed.
+If there are only a couple of files, the value should probably be
+.BR 1 ;
+with tens of thousands of files,
+.B 100
+or even more may be appropriate to reduce the number of
+.B xz
+processes that
+.BR xargs (1)
+will eventually create.
+.SS Robot mode examples
+Calculating how many bytes have been saved in total after compressing
+multiple files:
+.IP
+.B "xz --robot --list *.xz | awk '/^totals/{print $5\-$4}'"
+.SH "SEE ALSO"
+.BR xzdec (1),
+.BR gzip (1),
+.BR bzip2 (1)
+.PP
+XZ Utils: <http://tukaani.org/xz/>
+.br
+XZ Embedded: <http://tukaani.org/xz/embedded.html>
+.br
+LZMA SDK: <http://7-zip.org/sdk.html>
diff --git a/src/xz/xz_w32res.rc b/src/xz/xz_w32res.rc
new file mode 100644
index 0000000..bad3020
--- /dev/null
+++ b/src/xz/xz_w32res.rc
@@ -0,0 +1,12 @@
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#define MY_TYPE VFT_APP
+#define MY_NAME "xz"
+#define MY_SUFFIX ".exe"
+#define MY_DESC "xz data compression tool for .xz and .lzma files"
+#include "common_w32res.rc"
diff --git a/src/xzdec/Makefile.am b/src/xzdec/Makefile.am
new file mode 100644
index 0000000..ad48772
--- /dev/null
+++ b/src/xzdec/Makefile.am
@@ -0,0 +1,68 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+# Windows resource compiler support. It's fine to use xz_CPPFLAGS
+# also for lzmadec.
+.rc.o:
+ $(RC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) \
+ $(xzdec_CPPFLAGS) $(CPPFLAGS) $(RCFLAGS) -i $< -o $@
+
+
+bin_PROGRAMS = xzdec lzmadec
+
+xzdec_SOURCES = \
+ xzdec.c \
+ $(top_srcdir)/src/common/tuklib_progname.c \
+ $(top_srcdir)/src/common/tuklib_exit.c
+
+if COND_W32
+xzdec_SOURCES += xzdec_w32res.rc
+endif
+
+xzdec_CPPFLAGS = \
+ -DTUKLIB_GETTEXT=0 \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_builddir)/lib \
+ $(STATIC_CPPFLAGS)
+xzdec_LDFLAGS = $(STATIC_LDFLAGS)
+xzdec_LDADD = $(top_builddir)/src/liblzma/liblzma.la
+
+if COND_GNULIB
+xzdec_LDADD += $(top_builddir)/lib/libgnu.a
+endif
+
+xzdec_LDADD += $(LTLIBINTL)
+
+
+lzmadec_SOURCES = \
+ xzdec.c \
+ $(top_srcdir)/src/common/tuklib_progname.c \
+ $(top_srcdir)/src/common/tuklib_exit.c
+
+if COND_W32
+lzmadec_SOURCES += lzmadec_w32res.rc
+endif
+
+lzmadec_CPPFLAGS = $(xzdec_CPPFLAGS) -DLZMADEC
+lzmadec_LDFLAGS = $(xzdec_LDFLAGS)
+lzmadec_LDADD = $(xzdec_LDADD)
+
+
+dist_man_MANS = xzdec.1
+
+install-data-hook:
+ cd $(DESTDIR)$(mandir)/man1 && \
+ target=`echo xzdec | sed '$(transform)'` && \
+ link=`echo lzmadec | sed '$(transform)'` && \
+ rm -f $$link.1 && \
+ $(LN_S) $$target.1 $$link.1
+
+uninstall-hook:
+ cd $(DESTDIR)$(mandir)/man1 && \
+ link=`echo lzmadec | sed '$(transform)'` && \
+ rm -f $$link.1
diff --git a/src/xzdec/lzmadec_w32res.rc b/src/xzdec/lzmadec_w32res.rc
new file mode 100644
index 0000000..2c71a51
--- /dev/null
+++ b/src/xzdec/lzmadec_w32res.rc
@@ -0,0 +1,5 @@
+#define MY_TYPE VFT_APP
+#define MY_NAME "lzmadec"
+#define MY_SUFFIX ".exe"
+#define MY_DESC "lzmadec decompression tool for .lzma files"
+#include "common_w32res.rc"
diff --git a/src/xzdec/xzdec.1 b/src/xzdec/xzdec.1
new file mode 100644
index 0000000..3057c58
--- /dev/null
+++ b/src/xzdec/xzdec.1
@@ -0,0 +1,168 @@
+.\"
+.\" Author: Lasse Collin
+.\"
+.\" This file has been put into the public domain.
+.\" You can do whatever you want with this file.
+.\"
+.TH XZDEC 1 "2010-03-07" "Tukaani" "XZ Utils"
+.SH NAME
+xzdec, lzmadec \- Small .xz and .lzma decompressors
+.SH SYNOPSIS
+.B xzdec
+.RI [ option ]...
+.RI [ file ]...
+.br
+.B lzmadec
+.RI [ option ]...
+.RI [ file ]...
+.SH DESCRIPTION
+.B xzdec
+is a liblzma-based decompression-only tool for
+.B .xz
+(and only
+.BR .xz )
+files.
+.B xzdec
+is intended to work as a drop-in replacement for
+.BR xz (1)
+in the most common situations where a script has been written to use
+.B "xz \-\-decompress \-\-stdout"
+(and possibly a few other commonly used options) to decompress
+.B .xz
+files.
+.B lzmadec
+is identical to
+.B xzdec
+except that
+.B lzmadec
+supports
+.B .lzma
+files instead of
+.B .xz
+files.
+.PP
+To reduce the size of the executable,
+.B xzdec
+doesn't support multithreading or localization, and doesn't read options from
+.B XZ_OPT
+environment variable.
+.B xzdec
+doesn't support displaying intermediate progress information: sending
+.B SIGINFO
+to
+.B xzdec
+does nothing, but sending
+.B SIGUSR1
+terminates the process instead of displaying progress information.
+.SH OPTIONS
+.TP
+.BR \-d ", " \-\-decompress ", " \-\-uncompress
+Ignored for
+.BR xz (1)
+compatibility.
+.B xzdec
+supports only decompression.
+.TP
+.BR \-k ", " \-\-keep
+Ignored for
+.BR xz (1)
+compatibility.
+.B xzdec
+never creates or removes any files.
+.TP
+.BR \-c ", " \-\-stdout ", " \-\-to-stdout
+Ignored for
+.BR xz (1)
+compatibility.
+.B xzdec
+always writes the decompressed data to standard output.
+.TP
+\fB\-M\fR \fIlimit\fR, \fB\-\-memory=\fIlimit
+Set the memory usage
+.IR limit .
+If this option is specified multiple times, the last one takes effect. The
+.I limit
+can be specified in multiple ways:
+.RS
+.IP \(bu 3
+The
+.I limit
+can be an absolute value in bytes. Using an integer suffix like
+.B MiB
+can be useful. Example:
+.B "\-\-memory=80MiB"
+.IP \(bu 3
+The
+.I limit
+can be specified as a percentage of physical RAM. Example:
+.B "\-\-memory=70%"
+.IP \(bu 3
+The
+.I limit
+can be reset back to its default value by setting it to
+.BR 0 .
+.IP \(bu 3
+The memory usage limiting can be effectively disabled by setting
+.I limit
+to
+.BR max .
+This isn't recommended. It's usually better to use, for example,
+.BR \-\-memory=90% .
+.RE
+.IP
+The current
+.I limit
+can be seen near the bottom of the output of the
+.B \-\-help
+option.
+.TP
+.BR \-q ", " \-\-quiet
+Specifying this once does nothing since
+.B xzdec
+never displays any warnings or notices.
+Specify this twice to suppress errors.
+.TP
+.BR \-Q ", " \-\-no-warn
+Ignored for
+.BR xz (1)
+compatibility.
+.B xzdec
+never uses the exit status
+.BR "2" .
+.TP
+.BR \-h ", " \-\-help
+Display a help message and exit successfully.
+.TP
+.BR \-V ", " \-\-version
+Display the version number of
+.B xzdec
+and liblzma.
+.SH "EXIT STATUS"
+.TP
+.B 0
+All was good.
+.TP
+.B 1
+An error occurred.
+.PP
+.B xzdec
+doesn't have any warning messages like
+.BR xz (1)
+has, thus the exit status
+.B 2
+is not used by
+.BR xzdec .
+.SH NOTES
+.B xzdec
+and
+.B lzmadec
+are not really that small. The size can be reduced further by dropping
+features from liblzma at compile time, but that shouldn't usually be done
+for executables distributed in typical non-embedded operating system
+distributions. If you need a truly small
+.B .xz
+decompressor, consider using XZ Embedded.
+.SH "SEE ALSO"
+.BR xz (1)
+.PP
+XZ Embedded: <http://tukaani.org/xz/embedded.html>
diff --git a/src/xzdec/xzdec.c b/src/xzdec/xzdec.c
new file mode 100644
index 0000000..7f2e0fd
--- /dev/null
+++ b/src/xzdec/xzdec.c
@@ -0,0 +1,491 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file xzdec.c
+/// \brief Simple single-threaded tool to uncompress .xz or .lzma files
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include "lzma.h"
+
+#include <stdarg.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include "getopt.h"
+#include "tuklib_progname.h"
+#include "tuklib_exit.h"
+
+#ifdef TUKLIB_DOSLIKE
+# include <fcntl.h>
+# include <io.h>
+#endif
+
+
+#ifdef LZMADEC
+# define TOOL_FORMAT "lzma"
+#else
+# define TOOL_FORMAT "xz"
+#endif
+
+
+/// Number of bytes to use memory at maximum
+static uint64_t memlimit;
+
+/// Total amount of physical RAM
+static uint64_t total_ram;
+
+/// Error messages are suppressed if this is zero, which is the case when
+/// --quiet has been given at least twice.
+static unsigned int display_errors = 2;
+
+
+static void lzma_attribute((format(printf, 1, 2)))
+my_errorf(const char *fmt, ...)
+{
+ va_list ap;
+ va_start(ap, fmt);
+
+ if (display_errors) {
+ fprintf(stderr, "%s: ", progname);
+ vfprintf(stderr, fmt, ap);
+ fprintf(stderr, "\n");
+ }
+
+ va_end(ap);
+ return;
+}
+
+
+static void lzma_attribute((noreturn))
+help(void)
+{
+ // Round up to the next MiB and do it correctly also with UINT64_MAX.
+ const uint64_t mem_mib = (memlimit >> 20)
+ + ((memlimit & ((UINT32_C(1) << 20) - 1)) != 0);
+
+ printf(
+"Usage: %s [OPTION]... [FILE]...\n"
+"Uncompress files in the ." TOOL_FORMAT " format to the standard output.\n"
+"\n"
+" -c, --stdout (ignored)\n"
+" -d, --decompress (ignored)\n"
+" -k, --keep (ignored)\n"
+" -M, --memory=NUM use NUM bytes of memory at maximum (0 means default)\n"
+" -q, --quiet specify *twice* to suppress errors\n"
+" -Q, --no-warn (ignored)\n"
+" -h, --help display this help and exit\n"
+" -V, --version display the version number and exit\n"
+"\n"
+"With no FILE, or when FILE is -, read standard input.\n"
+"\n"
+"On this system and configuration, this program will use a maximum of roughly\n"
+"%" PRIu64 " MiB RAM.\n"
+"\n"
+"Report bugs to <" PACKAGE_BUGREPORT "> (in English or Finnish).\n"
+PACKAGE_NAME " home page: <" PACKAGE_URL ">\n", progname, mem_mib);
+ tuklib_exit(EXIT_SUCCESS, EXIT_FAILURE, display_errors);
+}
+
+
+static void lzma_attribute((noreturn))
+version(void)
+{
+ printf(TOOL_FORMAT "dec (" PACKAGE_NAME ") " LZMA_VERSION_STRING "\n"
+ "liblzma %s\n", lzma_version_string());
+
+ tuklib_exit(EXIT_SUCCESS, EXIT_FAILURE, display_errors);
+}
+
+
+/// Find out the amount of physical memory (RAM) in the system, and set
+/// the memory usage limit to the given percentage of RAM.
+static void
+memlimit_set_percentage(uint32_t percentage)
+{
+ memlimit = percentage * total_ram / 100;
+ return;
+}
+
+
+/// Set the memory usage limit to give number of bytes. Zero is a special
+/// value to indicate the default limit.
+static void
+memlimit_set(uint64_t new_memlimit)
+{
+ if (new_memlimit != 0) {
+ memlimit = new_memlimit;
+ } else {
+ memlimit = 40 * total_ram / 100;
+ if (memlimit < UINT64_C(80) * 1024 * 1024) {
+ memlimit = 80 * total_ram / 100;
+ if (memlimit > UINT64_C(80) * 1024 * 1024)
+ memlimit = UINT64_C(80) * 1024 * 1024;
+ }
+ }
+
+ return;
+}
+
+
+/// Get the total amount of physical RAM and set the memory usage limit
+/// to the default value.
+static void
+memlimit_init(void)
+{
+ // If we cannot determine the amount of RAM, use the assumption
+ // defined by the configure script.
+ total_ram = lzma_physmem();
+ if (total_ram == 0)
+ total_ram = (uint64_t)(ASSUME_RAM) * 1024 * 1024;
+
+ memlimit_set(0);
+ return;
+}
+
+
+/// \brief Convert a string to uint64_t
+///
+/// This is rudely copied from src/xz/util.c and modified a little. :-(
+/// Since this function is used only for parsing the memory usage limit,
+/// this cheats a little and saturates too big values to UINT64_MAX instead
+/// of giving an error.
+///
+/// \param max Return value when the string "max" was specified.
+///
+static uint64_t
+str_to_uint64(const char *value, uint64_t max)
+{
+ uint64_t result = 0;
+
+ // Accept special value "max".
+ if (strcmp(value, "max") == 0)
+ return max;
+
+ if (*value < '0' || *value > '9') {
+ my_errorf("%s: Value is not a non-negative decimal integer",
+ value);
+ exit(EXIT_FAILURE);
+ }
+
+ do {
+ // Don't overflow.
+ if (result > UINT64_MAX / 10)
+ return UINT64_MAX;
+
+ result *= 10;
+
+ // Another overflow check
+ const uint32_t add = *value - '0';
+ if (UINT64_MAX - add < result)
+ return UINT64_MAX;
+
+ result += add;
+ ++value;
+ } while (*value >= '0' && *value <= '9');
+
+ if (*value != '\0') {
+ // Look for suffix.
+ uint64_t multiplier = 0;
+ if (*value == 'k' || *value == 'K')
+ multiplier = UINT64_C(1) << 10;
+ else if (*value == 'm' || *value == 'M')
+ multiplier = UINT64_C(1) << 20;
+ else if (*value == 'g' || *value == 'G')
+ multiplier = UINT64_C(1) << 30;
+
+ ++value;
+
+ // Allow also e.g. Ki, KiB, and KB.
+ if (*value != '\0' && strcmp(value, "i") != 0
+ && strcmp(value, "iB") != 0
+ && strcmp(value, "B") != 0)
+ multiplier = 0;
+
+ if (multiplier == 0) {
+ my_errorf("%s: Invalid suffix", value - 1);
+ exit(EXIT_FAILURE);
+ }
+
+ // Don't overflow here either.
+ if (result > UINT64_MAX / multiplier)
+ result = UINT64_MAX;
+ else
+ result *= multiplier;
+ }
+
+ return result;
+}
+
+
+/// Parses command line options.
+static void
+parse_options(int argc, char **argv)
+{
+ static const char short_opts[] = "cdkM:hqQV";
+ static const struct option long_opts[] = {
+ { "stdout", no_argument, NULL, 'c' },
+ { "to-stdout", no_argument, NULL, 'c' },
+ { "decompress", no_argument, NULL, 'd' },
+ { "uncompress", no_argument, NULL, 'd' },
+ { "keep", no_argument, NULL, 'k' },
+ { "memory", required_argument, NULL, 'M' },
+ { "quiet", no_argument, NULL, 'q' },
+ { "no-warn", no_argument, NULL, 'Q' },
+ { "help", no_argument, NULL, 'h' },
+ { "version", no_argument, NULL, 'V' },
+ { NULL, 0, NULL, 0 }
+ };
+
+ int c;
+
+ while ((c = getopt_long(argc, argv, short_opts, long_opts, NULL))
+ != -1) {
+ switch (c) {
+ case 'c':
+ case 'd':
+ case 'k':
+ case 'Q':
+ break;
+
+ case 'M': {
+ // Support specifying the limit as a percentage of
+ // installed physical RAM.
+ const size_t len = strlen(optarg);
+ if (len > 0 && optarg[len - 1] == '%') {
+ // Memory limit is a percentage of total
+ // installed RAM.
+ optarg[len - 1] = '\0';
+ const uint64_t percentage
+ = str_to_uint64(optarg, 100);
+ if (percentage < 1 || percentage > 100) {
+ my_errorf("Percentage must be in "
+ "the range [1, 100]");
+ exit(EXIT_FAILURE);
+ }
+
+ memlimit_set_percentage(percentage);
+ } else {
+ memlimit_set(str_to_uint64(
+ optarg, UINT64_MAX));
+ }
+
+ break;
+ }
+
+ case 'q':
+ if (display_errors > 0)
+ --display_errors;
+
+ break;
+
+ case 'h':
+ help();
+
+ case 'V':
+ version();
+
+ default:
+ exit(EXIT_FAILURE);
+ }
+ }
+
+ return;
+}
+
+
+static void
+uncompress(lzma_stream *strm, FILE *file, const char *filename)
+{
+ lzma_ret ret;
+
+ // Initialize the decoder
+#ifdef LZMADEC
+ ret = lzma_alone_decoder(strm, memlimit);
+#else
+ ret = lzma_stream_decoder(strm, memlimit, LZMA_CONCATENATED);
+#endif
+
+ // The only reasonable error here is LZMA_MEM_ERROR.
+ // FIXME: Maybe also LZMA_MEMLIMIT_ERROR in future?
+ if (ret != LZMA_OK) {
+ my_errorf("%s", ret == LZMA_MEM_ERROR ? strerror(ENOMEM)
+ : "Internal error (bug)");
+ exit(EXIT_FAILURE);
+ }
+
+ // Input and output buffers
+ uint8_t in_buf[BUFSIZ];
+ uint8_t out_buf[BUFSIZ];
+
+ strm->avail_in = 0;
+ strm->next_out = out_buf;
+ strm->avail_out = BUFSIZ;
+
+ lzma_action action = LZMA_RUN;
+
+ while (true) {
+ if (strm->avail_in == 0) {
+ strm->next_in = in_buf;
+ strm->avail_in = fread(in_buf, 1, BUFSIZ, file);
+
+ if (ferror(file)) {
+ // POSIX says that fread() sets errno if
+ // an error occurred. ferror() doesn't
+ // touch errno.
+ my_errorf("%s: Error reading input file: %s",
+ filename, strerror(errno));
+ exit(EXIT_FAILURE);
+ }
+
+#ifndef LZMADEC
+ // When using LZMA_CONCATENATED, we need to tell
+ // liblzma when it has got all the input.
+ if (feof(file))
+ action = LZMA_FINISH;
+#endif
+ }
+
+ ret = lzma_code(strm, action);
+
+ // Write and check write error before checking decoder error.
+ // This way as much data as possible gets written to output
+ // even if decoder detected an error.
+ if (strm->avail_out == 0 || ret != LZMA_OK) {
+ const size_t write_size = BUFSIZ - strm->avail_out;
+
+ if (fwrite(out_buf, 1, write_size, stdout)
+ != write_size) {
+ // Wouldn't be a surprise if writing to stderr
+ // would fail too but at least try to show an
+ // error message.
+ my_errorf("Cannot write to standard output: "
+ "%s", strerror(errno));
+ exit(EXIT_FAILURE);
+ }
+
+ strm->next_out = out_buf;
+ strm->avail_out = BUFSIZ;
+ }
+
+ if (ret != LZMA_OK) {
+ if (ret == LZMA_STREAM_END) {
+#ifdef LZMADEC
+ // Check that there's no trailing garbage.
+ if (strm->avail_in != 0
+ || fread(in_buf, 1, 1, file)
+ != 0
+ || !feof(file))
+ ret = LZMA_DATA_ERROR;
+ else
+ return;
+#else
+ // lzma_stream_decoder() already guarantees
+ // that there's no trailing garbage.
+ assert(strm->avail_in == 0);
+ assert(action == LZMA_FINISH);
+ assert(feof(file));
+ return;
+#endif
+ }
+
+ const char *msg;
+ switch (ret) {
+ case LZMA_MEM_ERROR:
+ msg = strerror(ENOMEM);
+ break;
+
+ case LZMA_MEMLIMIT_ERROR:
+ msg = "Memory usage limit reached";
+ break;
+
+ case LZMA_FORMAT_ERROR:
+ msg = "File format not recognized";
+ break;
+
+ case LZMA_OPTIONS_ERROR:
+ // FIXME: Better message?
+ msg = "Unsupported compression options";
+ break;
+
+ case LZMA_DATA_ERROR:
+ msg = "File is corrupt";
+ break;
+
+ case LZMA_BUF_ERROR:
+ msg = "Unexpected end of input";
+ break;
+
+ default:
+ msg = "Internal error (bug)";
+ break;
+ }
+
+ my_errorf("%s: %s", filename, msg);
+ exit(EXIT_FAILURE);
+ }
+ }
+}
+
+
+int
+main(int argc, char **argv)
+{
+ // Initialize progname which we will be used in error messages.
+ tuklib_progname_init(argv);
+
+ // Set the default memory usage limit. This is needed before parsing
+ // the command line arguments.
+ memlimit_init();
+
+ // Parse the command line options.
+ parse_options(argc, argv);
+
+ // The same lzma_stream is used for all files that we decode. This way
+ // we don't need to reallocate memory for every file if they use same
+ // compression settings.
+ lzma_stream strm = LZMA_STREAM_INIT;
+
+ // Some systems require setting stdin and stdout to binary mode.
+#ifdef TUKLIB_DOSLIKE
+ setmode(fileno(stdin), O_BINARY);
+ setmode(fileno(stdout), O_BINARY);
+#endif
+
+ if (optind == argc) {
+ // No filenames given, decode from stdin.
+ uncompress(&strm, stdin, "(stdin)");
+ } else {
+ // Loop through the filenames given on the command line.
+ do {
+ // "-" indicates stdin.
+ if (strcmp(argv[optind], "-") == 0) {
+ uncompress(&strm, stdin, "(stdin)");
+ } else {
+ FILE *file = fopen(argv[optind], "rb");
+ if (file == NULL) {
+ my_errorf("%s: %s", argv[optind],
+ strerror(errno));
+ exit(EXIT_FAILURE);
+ }
+
+ uncompress(&strm, file, argv[optind]);
+ fclose(file);
+ }
+ } while (++optind < argc);
+ }
+
+#ifndef NDEBUG
+ // Free the memory only when debugging. Freeing wastes some time,
+ // but allows detecting possible memory leaks with Valgrind.
+ lzma_end(&strm);
+#endif
+
+ tuklib_exit(EXIT_SUCCESS, EXIT_FAILURE, display_errors);
+}
diff --git a/src/xzdec/xzdec_w32res.rc b/src/xzdec/xzdec_w32res.rc
new file mode 100644
index 0000000..0e26a22
--- /dev/null
+++ b/src/xzdec/xzdec_w32res.rc
@@ -0,0 +1,12 @@
+/*
+ * Author: Lasse Collin
+ *
+ * This file has been put into the public domain.
+ * You can do whatever you want with this file.
+ */
+
+#define MY_TYPE VFT_APP
+#define MY_NAME "xzdec"
+#define MY_SUFFIX ".exe"
+#define MY_DESC "xzdec decompression tool for .xz files"
+#include "common_w32res.rc"
diff --git a/tests/Makefile.am b/tests/Makefile.am
new file mode 100644
index 0000000..d01787a
--- /dev/null
+++ b/tests/Makefile.am
@@ -0,0 +1,51 @@
+##
+## Author: Lasse Collin
+##
+## This file has been put into the public domain.
+## You can do whatever you want with this file.
+##
+
+EXTRA_DIST = \
+ files \
+ tests.h \
+ test_files.sh \
+ test_compress.sh \
+ bcj_test.c \
+ compress_prepared_bcj_sparc \
+ compress_prepared_bcj_x86
+
+AM_CPPFLAGS = \
+ -I$(top_srcdir)/src/common \
+ -I$(top_srcdir)/src/liblzma/api \
+ -I$(top_builddir)/lib \
+ $(STATIC_CPPFLAGS)
+
+AM_LDFLAGS = $(STATIC_LDFLAGS)
+
+LDADD = $(top_builddir)/src/liblzma/liblzma.la
+
+if COND_GNULIB
+LDADD += $(top_builddir)/lib/libgnu.a
+endif
+
+LDADD += $(LTLIBINTL)
+
+check_PROGRAMS = \
+ create_compress_files \
+ test_check \
+ test_stream_flags \
+ test_filter_flags \
+ test_block_header \
+ test_index
+
+TESTS = \
+ test_check \
+ test_stream_flags \
+ test_filter_flags \
+ test_block_header \
+ test_index \
+ test_files.sh \
+ test_compress.sh
+
+clean-local:
+ -rm -f compress_generated_*
diff --git a/tests/bcj_test.c b/tests/bcj_test.c
new file mode 100644
index 0000000..05de38a
--- /dev/null
+++ b/tests/bcj_test.c
@@ -0,0 +1,65 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file bcj_test.c
+/// \brief Source code of compress_prepared_bcj_*
+///
+/// This is a simple program that should make the compiler to generate
+/// PC-relative branches, jumps, and calls. The compiled files can then
+/// be used to test the branch conversion filters. Note that this program
+/// itself does nothing useful.
+///
+/// Compiling: gcc -std=c99 -fPIC -c bcj_test.c
+/// Don't optimize or strip.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+extern int jump(int a, int b);
+
+
+extern int
+call(int a, int b)
+{
+ if (a < b)
+ a = jump(a, b);
+
+ return a;
+}
+
+
+extern int
+jump(int a, int b)
+{
+ // The loop generates conditional jump backwards.
+ while (1) {
+ if (a < b) {
+ a *= 2;
+ a += 3 * b;
+ break;
+ } else {
+ // Put enough code here to prevent JMP SHORT on x86.
+ a += b;
+ a /= 2;
+ b += b % 5;
+ a -= b / 3;
+ b = 2 * b + a - 1;
+ a *= b + a + 1;
+ b += a - 1;
+ a += b * 2 - a / 5;
+ }
+ }
+
+ return a;
+}
+
+
+int
+main(int argc, char **argv)
+{
+ int a = call(argc, argc + 1);
+ return a == 0;
+}
diff --git a/tests/compress_prepared_bcj_sparc b/tests/compress_prepared_bcj_sparc
new file mode 100644
index 0000000..86ea7dd
--- /dev/null
+++ b/tests/compress_prepared_bcj_sparc
Binary files differ
diff --git a/tests/compress_prepared_bcj_x86 b/tests/compress_prepared_bcj_x86
new file mode 100644
index 0000000..bcc546f
--- /dev/null
+++ b/tests/compress_prepared_bcj_x86
Binary files differ
diff --git a/tests/create_compress_files.c b/tests/create_compress_files.c
new file mode 100644
index 0000000..44367d8
--- /dev/null
+++ b/tests/create_compress_files.c
@@ -0,0 +1,157 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file create_compress_files.c
+/// \brief Creates bunch of test files to be compressed
+///
+/// Using a test file generator program saves space in the source code
+/// package considerably.
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "sysdefs.h"
+#include <stdio.h>
+
+
+// Avoid re-creating the test files every time the tests are run.
+#define create_test(name) \
+do { \
+ if (!file_exists("compress_generated_" #name)) { \
+ FILE *file = file_create("compress_generated_" #name); \
+ write_ ## name(file); \
+ file_finish(file, "compress_generated_" #name); \
+ } \
+} while (0)
+
+
+static bool
+file_exists(const char *filename)
+{
+ // Trying to be somewhat portable by avoiding stat().
+ FILE *file = fopen(filename, "rb");
+ bool ret;
+
+ if (file != NULL) {
+ fclose(file);
+ ret = true;
+ } else {
+ ret = false;
+ }
+
+ return ret;
+}
+
+
+static FILE *
+file_create(const char *filename)
+{
+ FILE *file = fopen(filename, "wb");
+
+ if (file == NULL) {
+ perror(filename);
+ exit(1);
+ }
+
+ return file;
+}
+
+
+static void
+file_finish(FILE *file, const char *filename)
+{
+ const bool ferror_fail = ferror(file);
+ const bool fclose_fail = fclose(file);
+
+ if (ferror_fail || fclose_fail) {
+ perror(filename);
+ exit(1);
+ }
+}
+
+
+// File that repeats "abc\n" a few thousand times. This is targeted
+// especially at Subblock filter's run-length encoder.
+static void
+write_abc(FILE *file)
+{
+ for (size_t i = 0; i < 12345; ++i)
+ fwrite("abc\n", 4, 1, file);
+}
+
+
+// File that doesn't compress. We always use the same random seed to
+// generate identical files on all systems.
+static void
+write_random(FILE *file)
+{
+ uint32_t n = 5;
+
+ for (size_t i = 0; i < 123456; ++i) {
+ n = 101771 * n + 71777;
+
+ putc(n & 0xFF, file);
+ putc((n >> 8) & 0xFF, file);
+ putc((n >> 16) & 0xFF, file);
+ putc(n >> 24, file);
+ }
+}
+
+
+// Text file
+static void
+write_text(FILE *file)
+{
+ static const char *lorem[] = {
+ "Lorem", "ipsum", "dolor", "sit", "amet,", "consectetur",
+ "adipisicing", "elit,", "sed", "do", "eiusmod", "tempor",
+ "incididunt", "ut", "labore", "et", "dolore", "magna",
+ "aliqua.", "Ut", "enim", "ad", "minim", "veniam,", "quis",
+ "nostrud", "exercitation", "ullamco", "laboris", "nisi",
+ "ut", "aliquip", "ex", "ea", "commodo", "consequat.",
+ "Duis", "aute", "irure", "dolor", "in", "reprehenderit",
+ "in", "voluptate", "velit", "esse", "cillum", "dolore",
+ "eu", "fugiat", "nulla", "pariatur.", "Excepteur", "sint",
+ "occaecat", "cupidatat", "non", "proident,", "sunt", "in",
+ "culpa", "qui", "officia", "deserunt", "mollit", "anim",
+ "id", "est", "laborum."
+ };
+
+ // Let the first paragraph be the original text.
+ for (size_t w = 0; w < ARRAY_SIZE(lorem); ++w) {
+ fprintf(file, "%s ", lorem[w]);
+
+ if (w % 7 == 6)
+ fprintf(file, "\n");
+ }
+
+ // The rest shall be (hopefully) meaningless combinations of
+ // the same words.
+ uint32_t n = 29;
+
+ for (size_t p = 0; p < 500; ++p) {
+ fprintf(file, "\n\n");
+
+ for (size_t w = 0; w < ARRAY_SIZE(lorem); ++w) {
+ n = 101771 * n + 71777;
+
+ fprintf(file, "%s ", lorem[n % ARRAY_SIZE(lorem)]);
+
+ if (w % 7 == 6)
+ fprintf(file, "\n");
+ }
+ }
+}
+
+
+int
+main(void)
+{
+ create_test(abc);
+ create_test(random);
+ create_test(text);
+ return 0;
+}
diff --git a/tests/files/README b/tests/files/README
new file mode 100644
index 0000000..392ff76
--- /dev/null
+++ b/tests/files/README
@@ -0,0 +1,232 @@
+
+.xz Test Files
+----------------
+
+0. Introduction
+
+ This directory contains bunch of files to test handling of .xz files
+ in .xz decoder implementations. Many of the files have been created
+ by hand with a hex editor, thus there is no better "source code" than
+ the files themselves. All the test files (*.xz) and this README have
+ been put into the public domain.
+
+
+1. File Types
+
+ Good files (good-*.xz) must decode successfully without requiring
+ a lot of CPU time or RAM.
+
+ Unsupported files (unsupported-*.xz) are good files, but headers
+ indicate features not supported by the current file format
+ specification.
+
+ Bad files (bad-*.xz) must cause the decoder to give an error. Like
+ with the good files, these files must not require a lot of CPU time
+ or RAM before they get detected to be broken.
+
+
+2. Descriptions of Individual Files
+
+2.1. Good Files
+
+ good-0-empty.xz has one Stream with no Blocks.
+
+ good-0pad-empty.xz has one Stream with no Blocks followed by
+ four-byte Stream Padding.
+
+ good-0cat-empty.xz has two zero-Block Streams concatenated without
+ Stream Padding.
+
+ good-0catpad-empty.xz has two zero-Block Streams concatenated with
+ four-byte Stream Padding between the Streams.
+
+ good-1-check-none.xz has one Stream with one Block with two
+ uncompressed LZMA2 chunks and no integrity check.
+
+ good-1-check-crc32.xz has one Stream with one Block with two
+ uncompressed LZMA2 chunks and CRC32 check.
+
+ good-1-check-crc64.xz is like good-1-check-crc32.xz but with CRC64.
+
+ good-1-check-sha256.xz is like good-1-check-crc32.xz but with
+ SHA256.
+
+ good-2-lzma2.xz has one Stream with two Blocks with one uncompressed
+ LZMA2 chunk in each Block.
+
+ good-1-block_header-1.xz has both Compressed Size and Uncompressed
+ Size in the Block Header. This has also four extra bytes of Header
+ Padding.
+
+ good-1-block_header-2.xz has known Compressed Size.
+
+ good-1-block_header-3.xz has known Uncompressed Size.
+
+ good-1-delta-lzma2.tiff.xz is an image file that compresses
+ better with Delta+LZMA2 than with plain LZMA2.
+
+ good-1-x86-lzma2.xz uses the x86 filter (BCJ) and LZMA2. The
+ uncompressed file is compress_prepared_bcj_x86 found from the tests
+ directory.
+
+ good-1-sparc-lzma2.xz uses the SPARC filter and LZMA. The
+ uncompressed file is compress_prepared_bcj_sparc found from the tests
+ directory.
+
+ good-1-lzma2-1.xz has two LZMA2 chunks, of which the second sets
+ new properties.
+
+ good-1-lzma2-2.xz has two LZMA2 chunks, of which the second resets
+ the state without specifying new properties.
+
+ good-1-lzma2-3.xz has two LZMA2 chunks, of which the first is
+ uncompressed and the second is LZMA. The first chunk resets dictionary
+ and the second sets new properties.
+
+ good-1-lzma2-4.xz has three LZMA2 chunks: First is LZMA, second is
+ uncompressed with dictionary reset, and third is LZMA with new
+ properties but without dictionary reset.
+
+ good-1-3delta-lzma2.xz has three Delta filters and LZMA2.
+
+
+2.2. Unsupported Files
+
+ unsupported-check.xz uses Check ID 0x02 which isn't supported by
+ the current version of the file format. It is implementation-defined
+ how this file handled (it may reject it, or decode it possibly with
+ a warning).
+
+ unsupported-block_header.xz has a non-null byte in Header Padding,
+ which may indicate presence of a new unsupported field.
+
+ unsupported-filter_flags-1.xz has unsupported Filter ID 0x7F.
+
+ unsupported-filter_flags-2.xz specifies only Delta filter in the
+ List of Filter Flags, but Delta isn't allowed as the last filter in
+ the chain. It could be a little more correct to detect this file as
+ corrupt instead of unsupported, but saying it is unsupported is
+ simpler in case of liblzma.
+
+ unsupported-filter_flags-3.xz specifies two LZMA2 filters in the
+ List of Filter Flags. LZMA2 is allowed only as the last filter in the
+ chain. It could be a little more correct to detect this file as
+ corrupt instead of unsupported, but saying it is unsupported is
+ simpler in case of liblzma.
+
+
+2.3. Bad Files
+
+ bad-0pad-empty.xz has one Stream with no Blocks followed by
+ five-byte Stream Padding. Stream Padding must be a multiple of four
+ bytes, thus this file is corrupt.
+
+ bad-0catpad-empty.xz has two zero-Block Streams concatenated with
+ five-byte Stream Padding between the Streams.
+
+ bad-0cat-alone.xz is good-0-empty.xz concatenated with an empty
+ LZMA_Alone file.
+
+ bad-0cat-header_magic.xz is good-0cat-empty.xz but with one byte
+ wrong in the Header Magic Bytes field of the second Stream. liblzma
+ gives LZMA_DATA_ERROR for this. (LZMA_FORMAT_ERROR is used only if
+ the first Stream of a file has invalid Header Magic Bytes.)
+
+ bad-0-header_magic.xz is good-0-empty.xz but with one byte wrong
+ in the Header Magic Bytes field. liblzma gives LZMA_FORMAT_ERROR for
+ this.
+
+ bad-0-footer_magic.xz is good-0-empty.xz but with one byte wrong
+ in the Footer Magic Bytes field. liblzma gives LZMA_DATA_ERROR for
+ this.
+
+ bad-0-empty-truncated.xz is good-0-empty.xz without the last byte
+ of the file.
+
+ bad-0-nonempty_index.xz has no Blocks but Index claims that there is
+ one Block.
+
+ bad-0-backward_size.xz has wrong Backward Size in Stream Footer.
+
+ bad-1-stream_flags-1.xz has different Stream Flags in Stream Header
+ and Stream Footer.
+
+ bad-1-stream_flags-2.xz has wrong CRC32 in Stream Header.
+
+ bad-1-stream_flags-3.xz has wrong CRC32 in Stream Footer.
+
+ bad-1-vli-1.xz has two-byte variable-length integer in the
+ Uncompressed Size field in Block Header while one-byte would be enough
+ for that value. It's important that the file gets rejected due to too
+ big integer encoding instead of due to Uncompressed Size not matching
+ the value stored in the Block Header. That is, the decoder must not
+ try to decode the Compressed Data field.
+
+ bad-1-vli-2.xz has ten-byte variable-length integer as Uncompressed
+ Size in Block Header. It's important that the file gets rejected due
+ to too big integer encoding instead of due to Uncompressed Size not
+ matching the value stored in the Block Header. That is, the decoder
+ must not try to decode the Compressed Data field.
+
+ bad-1-block_header-1.xz has Block Header that ends in the middle of
+ the Filter Flags field.
+
+ bad-1-block_header-2.xz has Block Header that has Compressed Size and
+ Uncompressed Size but no List of Filter Flags field.
+
+ bad-1-block_header-3.xz has wrong CRC32 in Block Header.
+
+ bad-1-block_header-4.xz has too big Compressed Size in Block Header
+ (2^63 - 1 bytes while maximum is a little less, because the whole
+ Block must stay smaller than 2^63). It's important that the file
+ gets rejected due to invalid Compressed Size value; the decoder
+ must not try decoding the Compressed Data field.
+
+ bad-1-block_header-5.xz has zero as Compressed Size in Block Header.
+
+ bad-2-index-1.xz has wrong Unpadded Sizes in Index.
+
+ bad-2-index-2.xz has wrong Uncompressed Sizes in Index.
+
+ bad-2-index-3.xz has non-null byte in Index Padding.
+
+ bad-2-index-4.xz wrong CRC32 in Index.
+
+ bad-2-index-5.xz has zero as Unpadded Size. It is important that the
+ file gets rejected specifically due to Unpadded Size having an invalid
+ value.
+
+ bad-2-compressed_data_padding.xz has non-null byte in the padding of
+ the Compressed Data field of the first Block.
+
+ bad-1-check-crc32.xz has wrong Check (CRC32).
+
+ bad-1-check-crc64.xz has wrong Check (CRC64).
+
+ bad-1-check-sha256.xz has wrong Check (SHA-256).
+
+ bad-1-lzma2-1.xz has LZMA2 stream whose first chunk (uncompressed)
+ doesn't reset the dictionary.
+
+ bad-1-lzma2-2.xz has two LZMA2 chunks, of which the second chunk
+ indicates dictionary reset, but the LZMA compressed data tries to
+ repeat data from the previous chunk.
+
+ bad-1-lzma2-3.xz sets new invalid properties (lc=8, lp=0, pb=0) in
+ the middle of Block.
+
+ bad-1-lzma2-4.xz has two LZMA2 chunks, of which the first is
+ uncompressed and the second is LZMA. The first chunk resets dictionary
+ as it should, but the second chunk tries to reset state without
+ specifying properties for LZMA.
+
+ bad-1-lzma2-5.xz is like bad-1-lzma2-4.xz but doesn't try to reset
+ anything in the header of the second chunk.
+
+ bad-1-lzma2-6.xz has reserved LZMA2 control byte value (0x03).
+
+ bad-1-lzma2-7.xz has EOPM at LZMA level.
+
+ bad-1-lzma2-8.xz is like good-1-lzma2-4.xz but doesn't set new
+ properties in the third LZMA2 chunk.
+
diff --git a/tests/files/bad-0-backward_size.xz b/tests/files/bad-0-backward_size.xz
new file mode 100644
index 0000000..2b46fa9
--- /dev/null
+++ b/tests/files/bad-0-backward_size.xz
Binary files differ
diff --git a/tests/files/bad-0-empty-truncated.xz b/tests/files/bad-0-empty-truncated.xz
new file mode 100644
index 0000000..f879af8
--- /dev/null
+++ b/tests/files/bad-0-empty-truncated.xz
Binary files differ
diff --git a/tests/files/bad-0-footer_magic.xz b/tests/files/bad-0-footer_magic.xz
new file mode 100644
index 0000000..5d9e389
--- /dev/null
+++ b/tests/files/bad-0-footer_magic.xz
Binary files differ
diff --git a/tests/files/bad-0-header_magic.xz b/tests/files/bad-0-header_magic.xz
new file mode 100644
index 0000000..5984a45
--- /dev/null
+++ b/tests/files/bad-0-header_magic.xz
Binary files differ
diff --git a/tests/files/bad-0-nonempty_index.xz b/tests/files/bad-0-nonempty_index.xz
new file mode 100644
index 0000000..ed6e81f
--- /dev/null
+++ b/tests/files/bad-0-nonempty_index.xz
Binary files differ
diff --git a/tests/files/bad-0cat-alone.xz b/tests/files/bad-0cat-alone.xz
new file mode 100644
index 0000000..a915a3a
--- /dev/null
+++ b/tests/files/bad-0cat-alone.xz
Binary files differ
diff --git a/tests/files/bad-0cat-header_magic.xz b/tests/files/bad-0cat-header_magic.xz
new file mode 100644
index 0000000..426bf2d
--- /dev/null
+++ b/tests/files/bad-0cat-header_magic.xz
Binary files differ
diff --git a/tests/files/bad-0catpad-empty.xz b/tests/files/bad-0catpad-empty.xz
new file mode 100644
index 0000000..97c1330
--- /dev/null
+++ b/tests/files/bad-0catpad-empty.xz
Binary files differ
diff --git a/tests/files/bad-0pad-empty.xz b/tests/files/bad-0pad-empty.xz
new file mode 100644
index 0000000..45e00b7
--- /dev/null
+++ b/tests/files/bad-0pad-empty.xz
Binary files differ
diff --git a/tests/files/bad-1-block_header-1.xz b/tests/files/bad-1-block_header-1.xz
new file mode 100644
index 0000000..d991536
--- /dev/null
+++ b/tests/files/bad-1-block_header-1.xz
Binary files differ
diff --git a/tests/files/bad-1-block_header-2.xz b/tests/files/bad-1-block_header-2.xz
new file mode 100644
index 0000000..ae42ecf
--- /dev/null
+++ b/tests/files/bad-1-block_header-2.xz
Binary files differ
diff --git a/tests/files/bad-1-block_header-3.xz b/tests/files/bad-1-block_header-3.xz
new file mode 100644
index 0000000..606cbd2
--- /dev/null
+++ b/tests/files/bad-1-block_header-3.xz
Binary files differ
diff --git a/tests/files/bad-1-block_header-4.xz b/tests/files/bad-1-block_header-4.xz
new file mode 100644
index 0000000..e72dfbf
--- /dev/null
+++ b/tests/files/bad-1-block_header-4.xz
Binary files differ
diff --git a/tests/files/bad-1-block_header-5.xz b/tests/files/bad-1-block_header-5.xz
new file mode 100644
index 0000000..9652112
--- /dev/null
+++ b/tests/files/bad-1-block_header-5.xz
Binary files differ
diff --git a/tests/files/bad-1-check-crc32.xz b/tests/files/bad-1-check-crc32.xz
new file mode 100644
index 0000000..1ebe131
--- /dev/null
+++ b/tests/files/bad-1-check-crc32.xz
Binary files differ
diff --git a/tests/files/bad-1-check-crc64.xz b/tests/files/bad-1-check-crc64.xz
new file mode 100644
index 0000000..cdb7709
--- /dev/null
+++ b/tests/files/bad-1-check-crc64.xz
Binary files differ
diff --git a/tests/files/bad-1-check-sha256.xz b/tests/files/bad-1-check-sha256.xz
new file mode 100644
index 0000000..def7bff
--- /dev/null
+++ b/tests/files/bad-1-check-sha256.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-1.xz b/tests/files/bad-1-lzma2-1.xz
new file mode 100644
index 0000000..640f592
--- /dev/null
+++ b/tests/files/bad-1-lzma2-1.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-2.xz b/tests/files/bad-1-lzma2-2.xz
new file mode 100644
index 0000000..69ab07d
--- /dev/null
+++ b/tests/files/bad-1-lzma2-2.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-3.xz b/tests/files/bad-1-lzma2-3.xz
new file mode 100644
index 0000000..66f48c5
--- /dev/null
+++ b/tests/files/bad-1-lzma2-3.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-4.xz b/tests/files/bad-1-lzma2-4.xz
new file mode 100644
index 0000000..ac97041
--- /dev/null
+++ b/tests/files/bad-1-lzma2-4.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-5.xz b/tests/files/bad-1-lzma2-5.xz
new file mode 100644
index 0000000..700464d
--- /dev/null
+++ b/tests/files/bad-1-lzma2-5.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-6.xz b/tests/files/bad-1-lzma2-6.xz
new file mode 100644
index 0000000..2bda0c4
--- /dev/null
+++ b/tests/files/bad-1-lzma2-6.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-7.xz b/tests/files/bad-1-lzma2-7.xz
new file mode 100644
index 0000000..8cc711c
--- /dev/null
+++ b/tests/files/bad-1-lzma2-7.xz
Binary files differ
diff --git a/tests/files/bad-1-lzma2-8.xz b/tests/files/bad-1-lzma2-8.xz
new file mode 100644
index 0000000..f21a71b
--- /dev/null
+++ b/tests/files/bad-1-lzma2-8.xz
Binary files differ
diff --git a/tests/files/bad-1-stream_flags-1.xz b/tests/files/bad-1-stream_flags-1.xz
new file mode 100644
index 0000000..6511773
--- /dev/null
+++ b/tests/files/bad-1-stream_flags-1.xz
Binary files differ
diff --git a/tests/files/bad-1-stream_flags-2.xz b/tests/files/bad-1-stream_flags-2.xz
new file mode 100644
index 0000000..0c66b36
--- /dev/null
+++ b/tests/files/bad-1-stream_flags-2.xz
Binary files differ
diff --git a/tests/files/bad-1-stream_flags-3.xz b/tests/files/bad-1-stream_flags-3.xz
new file mode 100644
index 0000000..a9b1f98
--- /dev/null
+++ b/tests/files/bad-1-stream_flags-3.xz
Binary files differ
diff --git a/tests/files/bad-1-vli-1.xz b/tests/files/bad-1-vli-1.xz
new file mode 100644
index 0000000..6514ab1
--- /dev/null
+++ b/tests/files/bad-1-vli-1.xz
Binary files differ
diff --git a/tests/files/bad-1-vli-2.xz b/tests/files/bad-1-vli-2.xz
new file mode 100644
index 0000000..c16941b
--- /dev/null
+++ b/tests/files/bad-1-vli-2.xz
Binary files differ
diff --git a/tests/files/bad-2-compressed_data_padding.xz b/tests/files/bad-2-compressed_data_padding.xz
new file mode 100644
index 0000000..382d047
--- /dev/null
+++ b/tests/files/bad-2-compressed_data_padding.xz
Binary files differ
diff --git a/tests/files/bad-2-index-1.xz b/tests/files/bad-2-index-1.xz
new file mode 100644
index 0000000..f51ed21
--- /dev/null
+++ b/tests/files/bad-2-index-1.xz
Binary files differ
diff --git a/tests/files/bad-2-index-2.xz b/tests/files/bad-2-index-2.xz
new file mode 100644
index 0000000..d7d00ff
--- /dev/null
+++ b/tests/files/bad-2-index-2.xz
Binary files differ
diff --git a/tests/files/bad-2-index-3.xz b/tests/files/bad-2-index-3.xz
new file mode 100644
index 0000000..62428b8
--- /dev/null
+++ b/tests/files/bad-2-index-3.xz
Binary files differ
diff --git a/tests/files/bad-2-index-4.xz b/tests/files/bad-2-index-4.xz
new file mode 100644
index 0000000..9cf2df6
--- /dev/null
+++ b/tests/files/bad-2-index-4.xz
Binary files differ
diff --git a/tests/files/bad-2-index-5.xz b/tests/files/bad-2-index-5.xz
new file mode 100644
index 0000000..0a79270
--- /dev/null
+++ b/tests/files/bad-2-index-5.xz
Binary files differ
diff --git a/tests/files/good-0-empty.xz b/tests/files/good-0-empty.xz
new file mode 100644
index 0000000..83b95e0
--- /dev/null
+++ b/tests/files/good-0-empty.xz
Binary files differ
diff --git a/tests/files/good-0cat-empty.xz b/tests/files/good-0cat-empty.xz
new file mode 100644
index 0000000..e6fc314
--- /dev/null
+++ b/tests/files/good-0cat-empty.xz
Binary files differ
diff --git a/tests/files/good-0catpad-empty.xz b/tests/files/good-0catpad-empty.xz
new file mode 100644
index 0000000..4f86b7d
--- /dev/null
+++ b/tests/files/good-0catpad-empty.xz
Binary files differ
diff --git a/tests/files/good-0pad-empty.xz b/tests/files/good-0pad-empty.xz
new file mode 100644
index 0000000..c51e3a6
--- /dev/null
+++ b/tests/files/good-0pad-empty.xz
Binary files differ
diff --git a/tests/files/good-1-3delta-lzma2.xz b/tests/files/good-1-3delta-lzma2.xz
new file mode 100644
index 0000000..a0be1d0
--- /dev/null
+++ b/tests/files/good-1-3delta-lzma2.xz
Binary files differ
diff --git a/tests/files/good-1-block_header-1.xz b/tests/files/good-1-block_header-1.xz
new file mode 100644
index 0000000..fea5ad2
--- /dev/null
+++ b/tests/files/good-1-block_header-1.xz
Binary files differ
diff --git a/tests/files/good-1-block_header-2.xz b/tests/files/good-1-block_header-2.xz
new file mode 100644
index 0000000..6b5dcb3
--- /dev/null
+++ b/tests/files/good-1-block_header-2.xz
Binary files differ
diff --git a/tests/files/good-1-block_header-3.xz b/tests/files/good-1-block_header-3.xz
new file mode 100644
index 0000000..1565312
--- /dev/null
+++ b/tests/files/good-1-block_header-3.xz
Binary files differ
diff --git a/tests/files/good-1-check-crc32.xz b/tests/files/good-1-check-crc32.xz
new file mode 100644
index 0000000..6c89593
--- /dev/null
+++ b/tests/files/good-1-check-crc32.xz
Binary files differ
diff --git a/tests/files/good-1-check-crc64.xz b/tests/files/good-1-check-crc64.xz
new file mode 100644
index 0000000..5a9915d
--- /dev/null
+++ b/tests/files/good-1-check-crc64.xz
Binary files differ
diff --git a/tests/files/good-1-check-none.xz b/tests/files/good-1-check-none.xz
new file mode 100644
index 0000000..1e85faf
--- /dev/null
+++ b/tests/files/good-1-check-none.xz
Binary files differ
diff --git a/tests/files/good-1-check-sha256.xz b/tests/files/good-1-check-sha256.xz
new file mode 100644
index 0000000..fdc556b
--- /dev/null
+++ b/tests/files/good-1-check-sha256.xz
Binary files differ
diff --git a/tests/files/good-1-delta-lzma2.tiff.xz b/tests/files/good-1-delta-lzma2.tiff.xz
new file mode 100644
index 0000000..1f033bc
--- /dev/null
+++ b/tests/files/good-1-delta-lzma2.tiff.xz
Binary files differ
diff --git a/tests/files/good-1-lzma2-1.xz b/tests/files/good-1-lzma2-1.xz
new file mode 100644
index 0000000..d8d6489
--- /dev/null
+++ b/tests/files/good-1-lzma2-1.xz
Binary files differ
diff --git a/tests/files/good-1-lzma2-2.xz b/tests/files/good-1-lzma2-2.xz
new file mode 100644
index 0000000..7e8cdf1
--- /dev/null
+++ b/tests/files/good-1-lzma2-2.xz
Binary files differ
diff --git a/tests/files/good-1-lzma2-3.xz b/tests/files/good-1-lzma2-3.xz
new file mode 100644
index 0000000..c4c72be
--- /dev/null
+++ b/tests/files/good-1-lzma2-3.xz
Binary files differ
diff --git a/tests/files/good-1-lzma2-4.xz b/tests/files/good-1-lzma2-4.xz
new file mode 100644
index 0000000..e0d623a
--- /dev/null
+++ b/tests/files/good-1-lzma2-4.xz
Binary files differ
diff --git a/tests/files/good-1-sparc-lzma2.xz b/tests/files/good-1-sparc-lzma2.xz
new file mode 100644
index 0000000..4532bc6
--- /dev/null
+++ b/tests/files/good-1-sparc-lzma2.xz
Binary files differ
diff --git a/tests/files/good-1-x86-lzma2.xz b/tests/files/good-1-x86-lzma2.xz
new file mode 100644
index 0000000..8053917
--- /dev/null
+++ b/tests/files/good-1-x86-lzma2.xz
Binary files differ
diff --git a/tests/files/good-2-lzma2.xz b/tests/files/good-2-lzma2.xz
new file mode 100644
index 0000000..bed5085
--- /dev/null
+++ b/tests/files/good-2-lzma2.xz
Binary files differ
diff --git a/tests/files/unsupported-block_header.xz b/tests/files/unsupported-block_header.xz
new file mode 100644
index 0000000..3830442
--- /dev/null
+++ b/tests/files/unsupported-block_header.xz
Binary files differ
diff --git a/tests/files/unsupported-check.xz b/tests/files/unsupported-check.xz
new file mode 100644
index 0000000..c28355e
--- /dev/null
+++ b/tests/files/unsupported-check.xz
Binary files differ
diff --git a/tests/files/unsupported-filter_flags-1.xz b/tests/files/unsupported-filter_flags-1.xz
new file mode 100644
index 0000000..48b9373
--- /dev/null
+++ b/tests/files/unsupported-filter_flags-1.xz
Binary files differ
diff --git a/tests/files/unsupported-filter_flags-2.xz b/tests/files/unsupported-filter_flags-2.xz
new file mode 100644
index 0000000..c283359
--- /dev/null
+++ b/tests/files/unsupported-filter_flags-2.xz
Binary files differ
diff --git a/tests/files/unsupported-filter_flags-3.xz b/tests/files/unsupported-filter_flags-3.xz
new file mode 100644
index 0000000..2608498
--- /dev/null
+++ b/tests/files/unsupported-filter_flags-3.xz
Binary files differ
diff --git a/tests/test_block.c b/tests/test_block.c
new file mode 100644
index 0000000..0352dce
--- /dev/null
+++ b/tests/test_block.c
@@ -0,0 +1,52 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file test_block.c
+/// \brief Tests Block coders
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+
+static uint8_t text[] = "Hello world!";
+static uint8_t buffer[4096];
+static lzma_options_block block_options;
+static lzma_stream strm = LZMA_STREAM_INIT;
+
+
+static void
+test1(void)
+{
+
+}
+
+
+int
+main()
+{
+ lzma_init();
+
+ block_options = (lzma_options_block){
+ .check_type = LZMA_CHECK_NONE,
+ .has_eopm = true,
+ .has_uncompressed_size_in_footer = false,
+ .has_backward_size = false,
+ .handle_padding = false,
+ .total_size = LZMA_VLI_UNKNOWN,
+ .compressed_size = LZMA_VLI_UNKNOWN,
+ .uncompressed_size = LZMA_VLI_UNKNOWN,
+ .header_size = 5,
+ };
+ block_options.filters[0].id = LZMA_VLI_UNKNOWN;
+ block_options.filters[0].options = NULL;
+
+
+ lzma_end(&strm);
+
+ return 0;
+}
diff --git a/tests/test_block_header.c b/tests/test_block_header.c
new file mode 100644
index 0000000..3d9b5d9
--- /dev/null
+++ b/tests/test_block_header.c
@@ -0,0 +1,240 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file test_block_header.c
+/// \brief Tests Block Header coders
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+
+static uint8_t buf[LZMA_BLOCK_HEADER_SIZE_MAX];
+static lzma_block known_options;
+static lzma_block decoded_options;
+
+static lzma_options_lzma opt_lzma;
+
+static lzma_filter filters_none[1] = {
+ {
+ .id = LZMA_VLI_UNKNOWN,
+ },
+};
+
+
+static lzma_filter filters_one[2] = {
+ {
+ .id = LZMA_FILTER_LZMA2,
+ .options = &opt_lzma,
+ }, {
+ .id = LZMA_VLI_UNKNOWN,
+ }
+};
+
+
+static lzma_filter filters_four[5] = {
+ {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_LZMA2,
+ .options = &opt_lzma,
+ }, {
+ .id = LZMA_VLI_UNKNOWN,
+ }
+};
+
+
+static lzma_filter filters_five[6] = {
+ {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_X86,
+ .options = NULL,
+ }, {
+ .id = LZMA_FILTER_LZMA2,
+ .options = &opt_lzma,
+ }, {
+ .id = LZMA_VLI_UNKNOWN,
+ }
+};
+
+
+static void
+code(void)
+{
+ expect(lzma_block_header_encode(&known_options, buf) == LZMA_OK);
+
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+ memcrap(filters, sizeof(filters));
+ memcrap(&decoded_options, sizeof(decoded_options));
+
+ decoded_options.header_size = known_options.header_size;
+ decoded_options.check = known_options.check;
+ decoded_options.filters = filters;
+ expect(lzma_block_header_decode(&decoded_options, NULL, buf)
+ == LZMA_OK);
+
+ expect(known_options.compressed_size
+ == decoded_options.compressed_size);
+ expect(known_options.uncompressed_size
+ == decoded_options.uncompressed_size);
+
+ for (size_t i = 0; known_options.filters[i].id
+ != LZMA_VLI_UNKNOWN; ++i)
+ expect(known_options.filters[i].id == filters[i].id);
+
+ for (size_t i = 0; i < LZMA_FILTERS_MAX; ++i)
+ free(decoded_options.filters[i].options);
+}
+
+
+static void
+test1(void)
+{
+ known_options = (lzma_block){
+ .check = LZMA_CHECK_NONE,
+ .compressed_size = LZMA_VLI_UNKNOWN,
+ .uncompressed_size = LZMA_VLI_UNKNOWN,
+ .filters = NULL,
+ };
+
+ expect(lzma_block_header_size(&known_options) == LZMA_PROG_ERROR);
+
+ known_options.filters = filters_none;
+ expect(lzma_block_header_size(&known_options) == LZMA_PROG_ERROR);
+
+ known_options.filters = filters_five;
+ expect(lzma_block_header_size(&known_options) == LZMA_PROG_ERROR);
+
+ known_options.filters = filters_one;
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+
+ known_options.check = 999; // Some invalid value, which gets ignored.
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+
+ known_options.compressed_size = 5;
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+
+ known_options.compressed_size = 0; // Cannot be zero.
+ expect(lzma_block_header_size(&known_options) == LZMA_PROG_ERROR);
+
+ // LZMA_VLI_MAX is too big to keep the total size of the Block
+ // a valid VLI, but lzma_block_header_size() is not meant
+ // to validate it. (lzma_block_header_encode() must validate it.)
+ known_options.compressed_size = LZMA_VLI_MAX;
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+
+ known_options.compressed_size = LZMA_VLI_UNKNOWN;
+ known_options.uncompressed_size = 0;
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+
+ known_options.uncompressed_size = LZMA_VLI_MAX + 1;
+ expect(lzma_block_header_size(&known_options) == LZMA_PROG_ERROR);
+}
+
+
+static void
+test2(void)
+{
+ known_options = (lzma_block){
+ .check = LZMA_CHECK_CRC32,
+ .compressed_size = LZMA_VLI_UNKNOWN,
+ .uncompressed_size = LZMA_VLI_UNKNOWN,
+ .filters = filters_four,
+ };
+
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+ code();
+
+ known_options.compressed_size = 123456;
+ known_options.uncompressed_size = 234567;
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+ code();
+
+ // We can make the sizes smaller while keeping the header size
+ // the same.
+ known_options.compressed_size = 12;
+ known_options.uncompressed_size = 23;
+ code();
+}
+
+
+static void
+test3(void)
+{
+ known_options = (lzma_block){
+ .check = LZMA_CHECK_CRC32,
+ .compressed_size = LZMA_VLI_UNKNOWN,
+ .uncompressed_size = LZMA_VLI_UNKNOWN,
+ .filters = filters_one,
+ };
+
+ expect(lzma_block_header_size(&known_options) == LZMA_OK);
+ known_options.header_size += 4;
+ expect(lzma_block_header_encode(&known_options, buf) == LZMA_OK);
+
+ lzma_filter filters[LZMA_FILTERS_MAX + 1];
+ decoded_options.header_size = known_options.header_size;
+ decoded_options.check = known_options.check;
+ decoded_options.filters = filters;
+
+ // Wrong size
+ ++buf[0];
+ expect(lzma_block_header_decode(&decoded_options, NULL, buf)
+ == LZMA_PROG_ERROR);
+ --buf[0];
+
+ // Wrong CRC32
+ buf[known_options.header_size - 1] ^= 1;
+ expect(lzma_block_header_decode(&decoded_options, NULL, buf)
+ == LZMA_DATA_ERROR);
+ buf[known_options.header_size - 1] ^= 1;
+
+ // Unsupported filter
+ // NOTE: This may need updating when new IDs become supported.
+ buf[2] ^= 0x1F;
+ unaligned_write32le(buf + known_options.header_size - 4,
+ lzma_crc32(buf, known_options.header_size - 4, 0));
+ expect(lzma_block_header_decode(&decoded_options, NULL, buf)
+ == LZMA_OPTIONS_ERROR);
+ buf[2] ^= 0x1F;
+
+ // Non-nul Padding
+ buf[known_options.header_size - 4 - 1] ^= 1;
+ unaligned_write32le(buf + known_options.header_size - 4,
+ lzma_crc32(buf, known_options.header_size - 4, 0));
+ expect(lzma_block_header_decode(&decoded_options, NULL, buf)
+ == LZMA_OPTIONS_ERROR);
+ buf[known_options.header_size - 4 - 1] ^= 1;
+}
+
+
+int
+main(void)
+{
+ succeed(lzma_lzma_preset(&opt_lzma, 1));
+
+ test1();
+ test2();
+ test3();
+
+ return 0;
+}
diff --git a/tests/test_check.c b/tests/test_check.c
new file mode 100644
index 0000000..7d4a360
--- /dev/null
+++ b/tests/test_check.c
@@ -0,0 +1,83 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file test_check.c
+/// \brief Tests integrity checks
+///
+/// \todo Add SHA256
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+
+static const uint8_t test_string[9] = "123456789";
+static const uint8_t test_unaligned[12] = "xxx123456789";
+
+
+static bool
+test_crc32(void)
+{
+ static const uint32_t test_vector = 0xCBF43926;
+
+ // Test 1
+ uint32_t crc = lzma_crc32(test_string, sizeof(test_string), 0);
+ if (crc != test_vector)
+ return true;
+
+ // Test 2
+ crc = lzma_crc32(test_unaligned + 3, sizeof(test_string), 0);
+ if (crc != test_vector)
+ return true;
+
+ // Test 3
+ crc = 0;
+ for (size_t i = 0; i < sizeof(test_string); ++i)
+ crc = lzma_crc32(test_string + i, 1, crc);
+ if (crc != test_vector)
+ return true;
+
+ return false;
+}
+
+
+static bool
+test_crc64(void)
+{
+ static const uint64_t test_vector = 0x995DC9BBDF1939FA;
+
+ // Test 1
+ uint64_t crc = lzma_crc64(test_string, sizeof(test_string), 0);
+ if (crc != test_vector)
+ return true;
+
+ // Test 2
+ crc = lzma_crc64(test_unaligned + 3, sizeof(test_string), 0);
+ if (crc != test_vector)
+ return true;
+
+ // Test 3
+ crc = 0;
+ for (size_t i = 0; i < sizeof(test_string); ++i)
+ crc = lzma_crc64(test_string + i, 1, crc);
+ if (crc != test_vector)
+ return true;
+
+ return false;
+}
+
+
+int
+main(void)
+{
+ bool error = false;
+
+ error |= test_crc32();
+ error |= test_crc64();
+
+ return error ? 1 : 0;
+}
diff --git a/tests/test_compress.sh b/tests/test_compress.sh
new file mode 100755
index 0000000..ff0cb30
--- /dev/null
+++ b/tests/test_compress.sh
@@ -0,0 +1,129 @@
+#!/bin/sh
+
+###############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+# Find out if our shell supports functions.
+eval 'unset foo ; foo() { return 42; } ; foo'
+if test $? != 42 ; then
+ echo "/bin/sh doesn't support functions, skipping this test."
+ (exit 77)
+ exit 77
+fi
+
+test_xz() {
+ if $XZ -c "$@" "$FILE" > tmp_compressed; then
+ :
+ else
+ echo "Compressing failed: $* $FILE"
+ (exit 1)
+ exit 1
+ fi
+
+ if $XZ -cd tmp_compressed > tmp_uncompressed ; then
+ :
+ else
+ echo "Decoding failed: $* $FILE"
+ (exit 1)
+ exit 1
+ fi
+
+ if cmp tmp_uncompressed "$FILE" ; then
+ :
+ else
+ echo "Decoded file does not match the original: $* $FILE"
+ (exit 1)
+ exit 1
+ fi
+
+ if $XZDEC tmp_compressed > tmp_uncompressed ; then
+ :
+ else
+ echo "Decoding failed: $* $FILE"
+ (exit 1)
+ exit 1
+ fi
+
+ if cmp tmp_uncompressed "$FILE" ; then
+ :
+ else
+ echo "Decoded file does not match the original: $* $FILE"
+ (exit 1)
+ exit 1
+ fi
+
+ # Show progress:
+ echo . | tr -d '\n\r'
+}
+
+XZ="../src/xz/xz --memory=28MiB --threads=1"
+XZDEC="../src/xzdec/xzdec --memory=4MiB"
+unset XZ_OPT
+
+# Create the required input files.
+if ./create_compress_files ; then
+ :
+else
+ rm -f compress_*
+ echo "Failed to create files to test compression."
+ (exit 1)
+ exit 1
+fi
+
+# Remove temporary now (in case they are something weird), and on exit.
+rm -f tmp_compressed tmp_uncompressed
+trap 'rm -f tmp_compressed tmp_uncompressed' 0
+
+# Encode and decode each file with various filter configurations.
+# This takes quite a bit of time.
+echo "test_compress.sh:"
+for FILE in compress_generated_* "$srcdir"/compress_prepared_*
+do
+ MSG=`echo "x$FILE" | sed 's,^x,,; s,^.*/,,; s,^compress_,,'`
+ echo " $MSG" | tr -d '\n\r'
+
+ # Don't test with empty arguments; it breaks some ancient
+ # proprietary /bin/sh versions due to $@ used in test_xz().
+ test_xz -1
+ test_xz -2
+ test_xz -3
+ test_xz -4
+
+ # Disabled until Subblock format is stable.
+# --subblock \
+# --subblock=size=1 \
+# --subblock=size=1,rle=1 \
+# --subblock=size=1,rle=4 \
+# --subblock=size=4,rle=4 \
+# --subblock=size=8,rle=4 \
+# --subblock=size=8,rle=8 \
+# --subblock=size=4096,rle=12 \
+#
+ for ARGS in \
+ --delta=dist=1 \
+ --delta=dist=4 \
+ --delta=dist=256 \
+ --x86 \
+ --powerpc \
+ --ia64 \
+ --arm \
+ --armthumb \
+ --sparc
+ do
+ test_xz $ARGS --lzma2=dict=64KiB,nice=32,mode=fast
+
+ # Disabled until Subblock format is stable.
+ # test_xz --subblock $ARGS --lzma2=dict=64KiB,nice=32,mode=fast
+ done
+
+ echo
+done
+
+(exit 0)
+exit 0
diff --git a/tests/test_files.sh b/tests/test_files.sh
new file mode 100755
index 0000000..7dd9a39
--- /dev/null
+++ b/tests/test_files.sh
@@ -0,0 +1,33 @@
+#/bin/sh
+
+###############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+for I in "$srcdir"/files/good-*.xz
+do
+ if ../src/xzdec/xzdec "$I" > /dev/null 2> /dev/null ; then
+ :
+ else
+ echo "Good file failed: $I"
+ (exit 1)
+ exit 1
+ fi
+done
+
+for I in "$srcdir"/files/bad-*.xz
+do
+ if ../src/xzdec/xzdec "$I" > /dev/null 2> /dev/null ; then
+ echo "Bad file succeeded: $I"
+ (exit 1)
+ exit 1
+ fi
+done
+
+(exit 0)
+exit 0
diff --git a/tests/test_filter_flags.c b/tests/test_filter_flags.c
new file mode 100644
index 0000000..ccd9ae9
--- /dev/null
+++ b/tests/test_filter_flags.c
@@ -0,0 +1,258 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file test_filter_flags.c
+/// \brief Tests Filter Flags coders
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+
+static uint8_t buffer[4096];
+static lzma_filter known_flags;
+static lzma_filter decoded_flags;
+static lzma_stream strm = LZMA_STREAM_INIT;
+
+
+static bool
+encode(uint32_t known_size)
+{
+ memcrap(buffer, sizeof(buffer));
+
+ uint32_t tmp;
+ if (lzma_filter_flags_size(&tmp, &known_flags) != LZMA_OK)
+ return true;
+
+ if (tmp != known_size)
+ return true;
+
+ size_t out_pos = 0;
+ if (lzma_filter_flags_encode(&known_flags,
+ buffer, &out_pos, known_size) != LZMA_OK)
+ return true;
+
+ if (out_pos != known_size)
+ return true;
+
+ return false;
+}
+
+
+static bool
+decode_ret(uint32_t known_size, lzma_ret expected_ret)
+{
+ memcrap(&decoded_flags, sizeof(decoded_flags));
+
+ size_t pos = 0;
+ if (lzma_filter_flags_decode(&decoded_flags, NULL,
+ buffer, &pos, known_size) != expected_ret
+ || pos != known_size)
+ return true;
+
+ return false;
+}
+
+
+static bool
+decode(uint32_t known_size)
+{
+ if (decode_ret(known_size, LZMA_OK))
+ return true;
+
+ if (known_flags.id != decoded_flags.id)
+ return true;
+
+ return false;
+}
+
+
+#if defined(HAVE_ENCODER_X86) && defined(HAVE_DECODER_X86)
+static void
+test_bcj(void)
+{
+ // Test 1
+ known_flags.id = LZMA_FILTER_X86;
+ known_flags.options = NULL;
+
+ expect(!encode(2));
+ expect(!decode(2));
+ expect(decoded_flags.options == NULL);
+
+ // Test 2
+ lzma_options_bcj options;
+ options.start_offset = 0;
+ known_flags.options = &options;
+ expect(!encode(2));
+ expect(!decode(2));
+ expect(decoded_flags.options == NULL);
+
+ // Test 3
+ options.start_offset = 123456;
+ known_flags.options = &options;
+ expect(!encode(6));
+ expect(!decode(6));
+ expect(decoded_flags.options != NULL);
+
+ lzma_options_bcj *decoded = decoded_flags.options;
+ expect(decoded->start_offset == options.start_offset);
+
+ free(decoded);
+}
+#endif
+
+
+#if defined(HAVE_ENCODER_DELTA) && defined(HAVE_DECODER_DELTA)
+static void
+test_delta(void)
+{
+ // Test 1
+ known_flags.id = LZMA_FILTER_DELTA;
+ known_flags.options = NULL;
+ expect(encode(99));
+
+ // Test 2
+ lzma_options_delta options = {
+ .type = LZMA_DELTA_TYPE_BYTE,
+ .dist = 0
+ };
+ known_flags.options = &options;
+ expect(encode(99));
+
+ // Test 3
+ options.dist = LZMA_DELTA_DIST_MIN;
+ expect(!encode(3));
+ expect(!decode(3));
+ expect(((lzma_options_delta *)(decoded_flags.options))->dist
+ == options.dist);
+
+ free(decoded_flags.options);
+
+ // Test 4
+ options.dist = LZMA_DELTA_DIST_MAX;
+ expect(!encode(3));
+ expect(!decode(3));
+ expect(((lzma_options_delta *)(decoded_flags.options))->dist
+ == options.dist);
+
+ free(decoded_flags.options);
+
+ // Test 5
+ options.dist = LZMA_DELTA_DIST_MAX + 1;
+ expect(encode(99));
+}
+#endif
+
+/*
+#ifdef HAVE_FILTER_LZMA
+static void
+validate_lzma(void)
+{
+ const lzma_options_lzma *known = known_flags.options;
+ const lzma_options_lzma *decoded = decoded_flags.options;
+
+ expect(known->dictionary_size <= decoded->dictionary_size);
+
+ if (known->dictionary_size == 1)
+ expect(decoded->dictionary_size == 1);
+ else
+ expect(known->dictionary_size + known->dictionary_size / 2
+ > decoded->dictionary_size);
+
+ expect(known->literal_context_bits == decoded->literal_context_bits);
+ expect(known->literal_pos_bits == decoded->literal_pos_bits);
+ expect(known->pos_bits == decoded->pos_bits);
+}
+
+
+static void
+test_lzma(void)
+{
+ // Test 1
+ known_flags.id = LZMA_FILTER_LZMA1;
+ known_flags.options = NULL;
+ expect(encode(99));
+
+ // Test 2
+ lzma_options_lzma options = {
+ .dictionary_size = 0,
+ .literal_context_bits = 0,
+ .literal_pos_bits = 0,
+ .pos_bits = 0,
+ .preset_dictionary = NULL,
+ .preset_dictionary_size = 0,
+ .mode = LZMA_MODE_INVALID,
+ .fast_bytes = 0,
+ .match_finder = LZMA_MF_INVALID,
+ .match_finder_cycles = 0,
+ };
+
+ // Test 3 (empty dictionary not allowed)
+ known_flags.options = &options;
+ expect(encode(99));
+
+ // Test 4 (brute-force test some valid dictionary sizes)
+ options.dictionary_size = LZMA_DICTIONARY_SIZE_MIN;
+ while (options.dictionary_size != LZMA_DICTIONARY_SIZE_MAX) {
+ if (++options.dictionary_size == 5000)
+ options.dictionary_size = LZMA_DICTIONARY_SIZE_MAX - 5;
+
+ expect(!encode(4));
+ expect(!decode(4));
+ validate_lzma();
+
+ free(decoded_flags.options);
+ }
+
+ // Test 5 (too big dictionary size)
+ options.dictionary_size = LZMA_DICTIONARY_SIZE_MAX + 1;
+ expect(encode(99));
+
+ // Test 6 (brute-force test lc/lp/pb)
+ options.dictionary_size = LZMA_DICTIONARY_SIZE_MIN;
+ for (uint32_t lc = LZMA_LITERAL_CONTEXT_BITS_MIN;
+ lc <= LZMA_LITERAL_CONTEXT_BITS_MAX; ++lc) {
+ for (uint32_t lp = LZMA_LITERAL_POS_BITS_MIN;
+ lp <= LZMA_LITERAL_POS_BITS_MAX; ++lp) {
+ for (uint32_t pb = LZMA_POS_BITS_MIN;
+ pb <= LZMA_POS_BITS_MAX; ++pb) {
+ if (lc + lp > LZMA_LITERAL_BITS_MAX)
+ continue;
+
+ options.literal_context_bits = lc;
+ options.literal_pos_bits = lp;
+ options.pos_bits = pb;
+
+ expect(!encode(4));
+ expect(!decode(4));
+ validate_lzma();
+
+ free(decoded_flags.options);
+ }
+ }
+ }
+}
+#endif
+*/
+
+int
+main(void)
+{
+#if defined(HAVE_ENCODER_X86) && defined(HAVE_DECODER_X86)
+ test_bcj();
+#endif
+#if defined(HAVE_ENCODER_DELTA) && defined(HAVE_DECODER_DELTA)
+ test_delta();
+#endif
+// #ifdef HAVE_FILTER_LZMA
+// test_lzma();
+// #endif
+
+ lzma_end(&strm);
+
+ return 0;
+}
diff --git a/tests/test_index.c b/tests/test_index.c
new file mode 100644
index 0000000..06b4d6b
--- /dev/null
+++ b/tests/test_index.c
@@ -0,0 +1,659 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file test_index.c
+/// \brief Tests functions handling the lzma_index structure
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+#define MEMLIMIT (LZMA_VLI_C(1) << 20)
+
+#define SMALL_COUNT 3
+#define BIG_COUNT 5555
+
+
+static lzma_index *
+create_empty(void)
+{
+ lzma_index *i = lzma_index_init(NULL);
+ expect(i != NULL);
+ return i;
+}
+
+
+static lzma_index *
+create_small(void)
+{
+ lzma_index *i = lzma_index_init(NULL);
+ expect(i != NULL);
+ expect(lzma_index_append(i, NULL, 101, 555) == LZMA_OK);
+ expect(lzma_index_append(i, NULL, 602, 777) == LZMA_OK);
+ expect(lzma_index_append(i, NULL, 804, 999) == LZMA_OK);
+ return i;
+}
+
+
+static lzma_index *
+create_big(void)
+{
+ lzma_index *i = lzma_index_init(NULL);
+ expect(i != NULL);
+
+ lzma_vli total_size = 0;
+ lzma_vli uncompressed_size = 0;
+
+ // Add pseudo-random sizes (but always the same size values).
+ uint32_t n = 11;
+ for (size_t j = 0; j < BIG_COUNT; ++j) {
+ n = 7019 * n + 7607;
+ const uint32_t t = n * 3011;
+ expect(lzma_index_append(i, NULL, t, n) == LZMA_OK);
+ total_size += (t + 3) & ~LZMA_VLI_C(3);
+ uncompressed_size += n;
+ }
+
+ expect(lzma_index_block_count(i) == BIG_COUNT);
+ expect(lzma_index_total_size(i) == total_size);
+ expect(lzma_index_uncompressed_size(i) == uncompressed_size);
+ expect(lzma_index_total_size(i) + lzma_index_size(i)
+ + 2 * LZMA_STREAM_HEADER_SIZE
+ == lzma_index_stream_size(i));
+
+ return i;
+}
+
+
+static bool
+is_equal(const lzma_index *a, const lzma_index *b)
+{
+ // Compare only the Stream and Block sizes and offsets.
+ lzma_index_iter ra, rb;
+ lzma_index_iter_init(&ra, a);
+ lzma_index_iter_init(&rb, b);
+
+ while (true) {
+ bool reta = lzma_index_iter_next(&ra, LZMA_INDEX_ITER_ANY);
+ bool retb = lzma_index_iter_next(&rb, LZMA_INDEX_ITER_ANY);
+ if (reta)
+ return !(reta ^ retb);
+
+ if (ra.stream.number != rb.stream.number
+ || ra.stream.block_count
+ != rb.stream.block_count
+ || ra.stream.compressed_offset
+ != rb.stream.compressed_offset
+ || ra.stream.uncompressed_offset
+ != rb.stream.uncompressed_offset
+ || ra.stream.compressed_size
+ != rb.stream.compressed_size
+ || ra.stream.uncompressed_size
+ != rb.stream.uncompressed_size
+ || ra.stream.padding
+ != rb.stream.padding)
+ return false;
+
+ if (ra.stream.block_count == 0)
+ continue;
+
+ if (ra.block.number_in_file != rb.block.number_in_file
+ || ra.block.compressed_file_offset
+ != rb.block.compressed_file_offset
+ || ra.block.uncompressed_file_offset
+ != rb.block.uncompressed_file_offset
+ || ra.block.number_in_stream
+ != rb.block.number_in_stream
+ || ra.block.compressed_stream_offset
+ != rb.block.compressed_stream_offset
+ || ra.block.uncompressed_stream_offset
+ != rb.block.uncompressed_stream_offset
+ || ra.block.uncompressed_size
+ != rb.block.uncompressed_size
+ || ra.block.unpadded_size
+ != rb.block.unpadded_size
+ || ra.block.total_size
+ != rb.block.total_size)
+ return false;
+ }
+}
+
+
+static void
+test_equal(void)
+{
+ lzma_index *a = create_empty();
+ lzma_index *b = create_small();
+ lzma_index *c = create_big();
+ expect(a && b && c);
+
+ expect(is_equal(a, a));
+ expect(is_equal(b, b));
+ expect(is_equal(c, c));
+
+ expect(!is_equal(a, b));
+ expect(!is_equal(a, c));
+ expect(!is_equal(b, c));
+
+ lzma_index_end(a, NULL);
+ lzma_index_end(b, NULL);
+ lzma_index_end(c, NULL);
+}
+
+
+static void
+test_overflow(void)
+{
+ // Integer overflow tests
+ lzma_index *i = create_empty();
+
+ expect(lzma_index_append(i, NULL, LZMA_VLI_MAX - 5, 1234)
+ == LZMA_DATA_ERROR);
+
+ // TODO
+
+ lzma_index_end(i, NULL);
+}
+
+
+static void
+test_copy(const lzma_index *i)
+{
+ lzma_index *d = lzma_index_dup(i, NULL);
+ expect(d != NULL);
+ expect(is_equal(i, d));
+ lzma_index_end(d, NULL);
+}
+
+
+static void
+test_read(lzma_index *i)
+{
+ lzma_index_iter r;
+ lzma_index_iter_init(&r, i);
+
+ // Try twice so we see that rewinding works.
+ for (size_t j = 0; j < 2; ++j) {
+ lzma_vli total_size = 0;
+ lzma_vli uncompressed_size = 0;
+ lzma_vli stream_offset = LZMA_STREAM_HEADER_SIZE;
+ lzma_vli uncompressed_offset = 0;
+ uint32_t count = 0;
+
+ while (!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)) {
+ ++count;
+
+ total_size += r.block.total_size;
+ uncompressed_size += r.block.uncompressed_size;
+
+ expect(r.block.compressed_file_offset
+ == stream_offset);
+ expect(r.block.uncompressed_file_offset
+ == uncompressed_offset);
+
+ stream_offset += r.block.total_size;
+ uncompressed_offset += r.block.uncompressed_size;
+ }
+
+ expect(lzma_index_total_size(i) == total_size);
+ expect(lzma_index_uncompressed_size(i) == uncompressed_size);
+ expect(lzma_index_block_count(i) == count);
+
+ lzma_index_iter_rewind(&r);
+ }
+}
+
+
+static void
+test_code(lzma_index *i)
+{
+ const size_t alloc_size = 128 * 1024;
+ uint8_t *buf = malloc(alloc_size);
+ expect(buf != NULL);
+
+ // Encode
+ lzma_stream strm = LZMA_STREAM_INIT;
+ expect(lzma_index_encoder(&strm, i) == LZMA_OK);
+ const lzma_vli index_size = lzma_index_size(i);
+ succeed(coder_loop(&strm, NULL, 0, buf, index_size,
+ LZMA_STREAM_END, LZMA_RUN));
+
+ // Decode
+ lzma_index *d;
+ expect(lzma_index_decoder(&strm, &d, MEMLIMIT) == LZMA_OK);
+ expect(d == NULL);
+ succeed(decoder_loop(&strm, buf, index_size));
+
+ expect(is_equal(i, d));
+
+ lzma_index_end(d, NULL);
+ lzma_end(&strm);
+
+ // Decode with hashing
+ lzma_index_hash *h = lzma_index_hash_init(NULL, NULL);
+ expect(h != NULL);
+ lzma_index_iter r;
+ lzma_index_iter_init(&r, i);
+ while (!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK))
+ expect(lzma_index_hash_append(h, r.block.unpadded_size,
+ r.block.uncompressed_size) == LZMA_OK);
+ size_t pos = 0;
+ while (pos < index_size - 1)
+ expect(lzma_index_hash_decode(h, buf, &pos, pos + 1)
+ == LZMA_OK);
+ expect(lzma_index_hash_decode(h, buf, &pos, pos + 1)
+ == LZMA_STREAM_END);
+
+ lzma_index_hash_end(h, NULL);
+
+ // Encode buffer
+ size_t buf_pos = 1;
+ expect(lzma_index_buffer_encode(i, buf, &buf_pos, index_size)
+ == LZMA_BUF_ERROR);
+ expect(buf_pos == 1);
+
+ succeed(lzma_index_buffer_encode(i, buf, &buf_pos, index_size + 1));
+ expect(buf_pos == index_size + 1);
+
+ // Decode buffer
+ buf_pos = 1;
+ uint64_t memlimit = MEMLIMIT;
+ expect(lzma_index_buffer_decode(&d, &memlimit, NULL, buf, &buf_pos,
+ index_size) == LZMA_DATA_ERROR);
+ expect(buf_pos == 1);
+ expect(d == NULL);
+
+ succeed(lzma_index_buffer_decode(&d, &memlimit, NULL, buf, &buf_pos,
+ index_size + 1));
+ expect(buf_pos == index_size + 1);
+ expect(is_equal(i, d));
+
+ lzma_index_end(d, NULL);
+
+ free(buf);
+}
+
+
+static void
+test_many(lzma_index *i)
+{
+ test_copy(i);
+ test_read(i);
+ test_code(i);
+}
+
+
+static void
+test_cat(void)
+{
+ lzma_index *a, *b, *c;
+ lzma_index_iter r;
+
+ // Empty Indexes
+ a = create_empty();
+ b = create_empty();
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_block_count(a) == 0);
+ expect(lzma_index_stream_size(a) == 2 * LZMA_STREAM_HEADER_SIZE + 8);
+ expect(lzma_index_file_size(a)
+ == 2 * (2 * LZMA_STREAM_HEADER_SIZE + 8));
+ lzma_index_iter_init(&r, a);
+ expect(lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK));
+
+ b = create_empty();
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_block_count(a) == 0);
+ expect(lzma_index_stream_size(a) == 2 * LZMA_STREAM_HEADER_SIZE + 8);
+ expect(lzma_index_file_size(a)
+ == 3 * (2 * LZMA_STREAM_HEADER_SIZE + 8));
+
+ b = create_empty();
+ c = create_empty();
+ expect(lzma_index_stream_padding(b, 4) == LZMA_OK);
+ expect(lzma_index_cat(b, c, NULL) == LZMA_OK);
+ expect(lzma_index_block_count(b) == 0);
+ expect(lzma_index_stream_size(b) == 2 * LZMA_STREAM_HEADER_SIZE + 8);
+ expect(lzma_index_file_size(b)
+ == 2 * (2 * LZMA_STREAM_HEADER_SIZE + 8) + 4);
+
+ expect(lzma_index_stream_padding(a, 8) == LZMA_OK);
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_block_count(a) == 0);
+ expect(lzma_index_stream_size(a) == 2 * LZMA_STREAM_HEADER_SIZE + 8);
+ expect(lzma_index_file_size(a)
+ == 5 * (2 * LZMA_STREAM_HEADER_SIZE + 8) + 4 + 8);
+
+ expect(lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK));
+ lzma_index_iter_rewind(&r);
+ expect(lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK));
+ lzma_index_end(a, NULL);
+
+ // Small Indexes
+ a = create_small();
+ lzma_vli stream_size = lzma_index_stream_size(a);
+ lzma_index_iter_init(&r, a);
+ for (int i = SMALL_COUNT; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ b = create_small();
+ expect(lzma_index_stream_padding(a, 4) == LZMA_OK);
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_file_size(a) == stream_size * 2 + 4);
+ expect(lzma_index_stream_size(a) > stream_size);
+ expect(lzma_index_stream_size(a) < stream_size * 2);
+ for (int i = SMALL_COUNT; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ lzma_index_iter_rewind(&r);
+ for (int i = SMALL_COUNT * 2; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ b = create_small();
+ c = create_small();
+ expect(lzma_index_stream_padding(b, 8) == LZMA_OK);
+ expect(lzma_index_cat(b, c, NULL) == LZMA_OK);
+ expect(lzma_index_stream_padding(a, 12) == LZMA_OK);
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_file_size(a) == stream_size * 4 + 4 + 8 + 12);
+
+ expect(lzma_index_block_count(a) == SMALL_COUNT * 4);
+ for (int i = SMALL_COUNT * 2; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ lzma_index_iter_rewind(&r);
+ for (int i = SMALL_COUNT * 4; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ lzma_index_end(a, NULL);
+
+ // Mix of empty and small
+ a = create_empty();
+ b = create_small();
+ expect(lzma_index_stream_padding(a, 4) == LZMA_OK);
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ lzma_index_iter_init(&r, a);
+ for (int i = SMALL_COUNT; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ lzma_index_end(a, NULL);
+
+ // Big Indexes
+ a = create_big();
+ stream_size = lzma_index_stream_size(a);
+ b = create_big();
+ expect(lzma_index_stream_padding(a, 4) == LZMA_OK);
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_file_size(a) == stream_size * 2 + 4);
+ expect(lzma_index_stream_size(a) > stream_size);
+ expect(lzma_index_stream_size(a) < stream_size * 2);
+
+ b = create_big();
+ c = create_big();
+ expect(lzma_index_stream_padding(b, 8) == LZMA_OK);
+ expect(lzma_index_cat(b, c, NULL) == LZMA_OK);
+ expect(lzma_index_stream_padding(a, 12) == LZMA_OK);
+ expect(lzma_index_cat(a, b, NULL) == LZMA_OK);
+ expect(lzma_index_file_size(a) == stream_size * 4 + 4 + 8 + 12);
+
+ lzma_index_iter_init(&r, a);
+ for (int i = BIG_COUNT * 4; i >= 0; --i)
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK)
+ ^ (i == 0));
+
+ lzma_index_end(a, NULL);
+}
+
+
+static void
+test_locate(void)
+{
+ lzma_index *i = lzma_index_init(NULL);
+ expect(i != NULL);
+ lzma_index_iter r;
+ lzma_index_iter_init(&r, i);
+
+ // Cannot locate anything from an empty Index.
+ expect(lzma_index_iter_locate(&r, 0));
+ expect(lzma_index_iter_locate(&r, 555));
+
+ // One empty Record: nothing is found since there's no uncompressed
+ // data.
+ expect(lzma_index_append(i, NULL, 16, 0) == LZMA_OK);
+ expect(lzma_index_iter_locate(&r, 0));
+
+ // Non-empty Record and we can find something.
+ expect(lzma_index_append(i, NULL, 32, 5) == LZMA_OK);
+ expect(!lzma_index_iter_locate(&r, 0));
+ expect(r.block.total_size == 32);
+ expect(r.block.uncompressed_size == 5);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 16);
+ expect(r.block.uncompressed_file_offset == 0);
+
+ // Still cannot find anything past the end.
+ expect(lzma_index_iter_locate(&r, 5));
+
+ // Add the third Record.
+ expect(lzma_index_append(i, NULL, 40, 11) == LZMA_OK);
+
+ expect(!lzma_index_iter_locate(&r, 0));
+ expect(r.block.total_size == 32);
+ expect(r.block.uncompressed_size == 5);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 16);
+ expect(r.block.uncompressed_file_offset == 0);
+
+ expect(!lzma_index_iter_next(&r, LZMA_INDEX_ITER_BLOCK));
+ expect(r.block.total_size == 40);
+ expect(r.block.uncompressed_size == 11);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 16 + 32);
+ expect(r.block.uncompressed_file_offset == 5);
+
+ expect(!lzma_index_iter_locate(&r, 2));
+ expect(r.block.total_size == 32);
+ expect(r.block.uncompressed_size == 5);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 16);
+ expect(r.block.uncompressed_file_offset == 0);
+
+ expect(!lzma_index_iter_locate(&r, 5));
+ expect(r.block.total_size == 40);
+ expect(r.block.uncompressed_size == 11);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 16 + 32);
+ expect(r.block.uncompressed_file_offset == 5);
+
+ expect(!lzma_index_iter_locate(&r, 5 + 11 - 1));
+ expect(r.block.total_size == 40);
+ expect(r.block.uncompressed_size == 11);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 16 + 32);
+ expect(r.block.uncompressed_file_offset == 5);
+
+ expect(lzma_index_iter_locate(&r, 5 + 11));
+ expect(lzma_index_iter_locate(&r, 5 + 15));
+
+ // Large Index
+ lzma_index_end(i, NULL);
+ i = lzma_index_init(NULL);
+ expect(i != NULL);
+ lzma_index_iter_init(&r, i);
+
+ for (size_t n = 4; n <= 4 * 5555; n += 4)
+ expect(lzma_index_append(i, NULL, n + 8, n) == LZMA_OK);
+
+ expect(lzma_index_block_count(i) == 5555);
+
+ // First Record
+ expect(!lzma_index_iter_locate(&r, 0));
+ expect(r.block.total_size == 4 + 8);
+ expect(r.block.uncompressed_size == 4);
+ expect(r.block.compressed_file_offset == LZMA_STREAM_HEADER_SIZE);
+ expect(r.block.uncompressed_file_offset == 0);
+
+ expect(!lzma_index_iter_locate(&r, 3));
+ expect(r.block.total_size == 4 + 8);
+ expect(r.block.uncompressed_size == 4);
+ expect(r.block.compressed_file_offset == LZMA_STREAM_HEADER_SIZE);
+ expect(r.block.uncompressed_file_offset == 0);
+
+ // Second Record
+ expect(!lzma_index_iter_locate(&r, 4));
+ expect(r.block.total_size == 2 * 4 + 8);
+ expect(r.block.uncompressed_size == 2 * 4);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + 4 + 8);
+ expect(r.block.uncompressed_file_offset == 4);
+
+ // Last Record
+ expect(!lzma_index_iter_locate(
+ &r, lzma_index_uncompressed_size(i) - 1));
+ expect(r.block.total_size == 4 * 5555 + 8);
+ expect(r.block.uncompressed_size == 4 * 5555);
+ expect(r.block.compressed_file_offset == lzma_index_total_size(i)
+ + LZMA_STREAM_HEADER_SIZE - 4 * 5555 - 8);
+ expect(r.block.uncompressed_file_offset
+ == lzma_index_uncompressed_size(i) - 4 * 5555);
+
+ // Allocation chunk boundaries. See INDEX_GROUP_SIZE in
+ // liblzma/common/index.c.
+ const size_t group_multiple = 256 * 4;
+ const size_t radius = 8;
+ const size_t start = group_multiple - radius;
+ lzma_vli ubase = 0;
+ lzma_vli tbase = 0;
+ size_t n;
+ for (n = 1; n < start; ++n) {
+ ubase += n * 4;
+ tbase += n * 4 + 8;
+ }
+
+ while (n < start + 2 * radius) {
+ expect(!lzma_index_iter_locate(&r, ubase + n * 4));
+
+ expect(r.block.compressed_file_offset == tbase + n * 4 + 8
+ + LZMA_STREAM_HEADER_SIZE);
+ expect(r.block.uncompressed_file_offset == ubase + n * 4);
+
+ tbase += n * 4 + 8;
+ ubase += n * 4;
+ ++n;
+
+ expect(r.block.total_size == n * 4 + 8);
+ expect(r.block.uncompressed_size == n * 4);
+ }
+
+ // Do it also backwards.
+ while (n > start) {
+ expect(!lzma_index_iter_locate(&r, ubase + (n - 1) * 4));
+
+ expect(r.block.total_size == n * 4 + 8);
+ expect(r.block.uncompressed_size == n * 4);
+
+ --n;
+ tbase -= n * 4 + 8;
+ ubase -= n * 4;
+
+ expect(r.block.compressed_file_offset == tbase + n * 4 + 8
+ + LZMA_STREAM_HEADER_SIZE);
+ expect(r.block.uncompressed_file_offset == ubase + n * 4);
+ }
+
+ // Test locating in concatenated Index.
+ lzma_index_end(i, NULL);
+ i = lzma_index_init(NULL);
+ expect(i != NULL);
+ lzma_index_iter_init(&r, i);
+ for (n = 0; n < group_multiple; ++n)
+ expect(lzma_index_append(i, NULL, 8, 0) == LZMA_OK);
+ expect(lzma_index_append(i, NULL, 16, 1) == LZMA_OK);
+ expect(!lzma_index_iter_locate(&r, 0));
+ expect(r.block.total_size == 16);
+ expect(r.block.uncompressed_size == 1);
+ expect(r.block.compressed_file_offset
+ == LZMA_STREAM_HEADER_SIZE + group_multiple * 8);
+ expect(r.block.uncompressed_file_offset == 0);
+
+ lzma_index_end(i, NULL);
+}
+
+
+static void
+test_corrupt(void)
+{
+ const size_t alloc_size = 128 * 1024;
+ uint8_t *buf = malloc(alloc_size);
+ expect(buf != NULL);
+ lzma_stream strm = LZMA_STREAM_INIT;
+
+ lzma_index *i = create_empty();
+ expect(lzma_index_append(i, NULL, 0, 1) == LZMA_PROG_ERROR);
+ lzma_index_end(i, NULL);
+
+ // Create a valid Index and corrupt it in different ways.
+ i = create_small();
+ expect(lzma_index_encoder(&strm, i) == LZMA_OK);
+ succeed(coder_loop(&strm, NULL, 0, buf, 20,
+ LZMA_STREAM_END, LZMA_RUN));
+ lzma_index_end(i, NULL);
+
+ // Wrong Index Indicator
+ buf[0] ^= 1;
+ expect(lzma_index_decoder(&strm, &i, MEMLIMIT) == LZMA_OK);
+ succeed(decoder_loop_ret(&strm, buf, 1, LZMA_DATA_ERROR));
+ buf[0] ^= 1;
+
+ // Wrong Number of Records and thus CRC32 fails.
+ --buf[1];
+ expect(lzma_index_decoder(&strm, &i, MEMLIMIT) == LZMA_OK);
+ succeed(decoder_loop_ret(&strm, buf, 10, LZMA_DATA_ERROR));
+ ++buf[1];
+
+ // Padding not NULs
+ buf[15] ^= 1;
+ expect(lzma_index_decoder(&strm, &i, MEMLIMIT) == LZMA_OK);
+ succeed(decoder_loop_ret(&strm, buf, 16, LZMA_DATA_ERROR));
+
+ lzma_end(&strm);
+ free(buf);
+}
+
+
+int
+main(void)
+{
+ test_equal();
+
+ test_overflow();
+
+ lzma_index *i = create_empty();
+ test_many(i);
+ lzma_index_end(i, NULL);
+
+ i = create_small();
+ test_many(i);
+ lzma_index_end(i, NULL);
+
+ i = create_big();
+ test_many(i);
+ lzma_index_end(i, NULL);
+
+ test_cat();
+
+ test_locate();
+
+ test_corrupt();
+
+ return 0;
+}
diff --git a/tests/test_stream_flags.c b/tests/test_stream_flags.c
new file mode 100644
index 0000000..9611459
--- /dev/null
+++ b/tests/test_stream_flags.c
@@ -0,0 +1,180 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file test_stream_flags.c
+/// \brief Tests Stream Header and Stream Footer coders
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+
+static lzma_stream_flags known_flags;
+static lzma_stream_flags decoded_flags;
+static uint8_t buffer[LZMA_STREAM_HEADER_SIZE];
+
+
+static bool
+validate(void)
+{
+ // TODO: This could require the specific error type as an argument.
+ // We could also test that lzma_stream_flags_compare() gives
+ // the correct return values in different situations.
+ return lzma_stream_flags_compare(&known_flags, &decoded_flags)
+ != LZMA_OK;
+}
+
+
+static bool
+test_header_decoder(lzma_ret expected_ret)
+{
+ memcrap(&decoded_flags, sizeof(decoded_flags));
+
+ if (lzma_stream_header_decode(&decoded_flags, buffer) != expected_ret)
+ return true;
+
+ if (expected_ret != LZMA_OK)
+ return false;
+
+ // Header doesn't have Backward Size, so make
+ // lzma_stream_flags_compare() ignore it.
+ decoded_flags.backward_size = LZMA_VLI_UNKNOWN;
+ return validate();
+}
+
+
+static void
+test_header(void)
+{
+ memcrap(buffer, sizeof(buffer));
+ expect(lzma_stream_header_encode(&known_flags, buffer) == LZMA_OK);
+ succeed(test_header_decoder(LZMA_OK));
+}
+
+
+static bool
+test_footer_decoder(lzma_ret expected_ret)
+{
+ memcrap(&decoded_flags, sizeof(decoded_flags));
+
+ if (lzma_stream_footer_decode(&decoded_flags, buffer) != expected_ret)
+ return true;
+
+ if (expected_ret != LZMA_OK)
+ return false;
+
+ return validate();
+}
+
+
+static void
+test_footer(void)
+{
+ memcrap(buffer, sizeof(buffer));
+ expect(lzma_stream_footer_encode(&known_flags, buffer) == LZMA_OK);
+ succeed(test_footer_decoder(LZMA_OK));
+}
+
+
+static void
+test_encode_invalid(void)
+{
+ known_flags.check = LZMA_CHECK_ID_MAX + 1;
+ known_flags.backward_size = 1024;
+
+ expect(lzma_stream_header_encode(&known_flags, buffer)
+ == LZMA_PROG_ERROR);
+
+ expect(lzma_stream_footer_encode(&known_flags, buffer)
+ == LZMA_PROG_ERROR);
+
+ known_flags.check = (lzma_check)(-1);
+
+ expect(lzma_stream_header_encode(&known_flags, buffer)
+ == LZMA_PROG_ERROR);
+
+ expect(lzma_stream_footer_encode(&known_flags, buffer)
+ == LZMA_PROG_ERROR);
+
+ known_flags.check = LZMA_CHECK_NONE;
+ known_flags.backward_size = 0;
+
+ // Header encoder ignores backward_size.
+ expect(lzma_stream_header_encode(&known_flags, buffer) == LZMA_OK);
+
+ expect(lzma_stream_footer_encode(&known_flags, buffer)
+ == LZMA_PROG_ERROR);
+
+ known_flags.backward_size = LZMA_VLI_MAX;
+
+ expect(lzma_stream_header_encode(&known_flags, buffer) == LZMA_OK);
+
+ expect(lzma_stream_footer_encode(&known_flags, buffer)
+ == LZMA_PROG_ERROR);
+}
+
+
+static void
+test_decode_invalid(void)
+{
+ known_flags.check = LZMA_CHECK_NONE;
+ known_flags.backward_size = 1024;
+
+ expect(lzma_stream_header_encode(&known_flags, buffer) == LZMA_OK);
+
+ // Test 1 (invalid Magic Bytes)
+ buffer[5] ^= 1;
+ succeed(test_header_decoder(LZMA_FORMAT_ERROR));
+ buffer[5] ^= 1;
+
+ // Test 2a (valid CRC32)
+ uint32_t crc = lzma_crc32(buffer + 6, 2, 0);
+ unaligned_write32le(buffer + 8, crc);
+ succeed(test_header_decoder(LZMA_OK));
+
+ // Test 2b (invalid Stream Flags with valid CRC32)
+ buffer[6] ^= 0x20;
+ crc = lzma_crc32(buffer + 6, 2, 0);
+ unaligned_write32le(buffer + 8, crc);
+ succeed(test_header_decoder(LZMA_OPTIONS_ERROR));
+
+ // Test 3 (invalid CRC32)
+ expect(lzma_stream_header_encode(&known_flags, buffer) == LZMA_OK);
+ buffer[9] ^= 1;
+ succeed(test_header_decoder(LZMA_DATA_ERROR));
+
+ // Test 4 (invalid Stream Flags with valid CRC32)
+ expect(lzma_stream_footer_encode(&known_flags, buffer) == LZMA_OK);
+ buffer[9] ^= 0x40;
+ crc = lzma_crc32(buffer + 4, 6, 0);
+ unaligned_write32le(buffer, crc);
+ succeed(test_footer_decoder(LZMA_OPTIONS_ERROR));
+
+ // Test 5 (invalid Magic Bytes)
+ expect(lzma_stream_footer_encode(&known_flags, buffer) == LZMA_OK);
+ buffer[11] ^= 1;
+ succeed(test_footer_decoder(LZMA_FORMAT_ERROR));
+}
+
+
+int
+main(void)
+{
+ // Valid headers
+ known_flags.backward_size = 1024;
+ for (lzma_check check = LZMA_CHECK_NONE;
+ check <= LZMA_CHECK_ID_MAX; ++check) {
+ test_header();
+ test_footer();
+ }
+
+ // Invalid headers
+ test_encode_invalid();
+ test_decode_invalid();
+
+ return 0;
+}
diff --git a/tests/tests.h b/tests/tests.h
new file mode 100644
index 0000000..8f3c745
--- /dev/null
+++ b/tests/tests.h
@@ -0,0 +1,124 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file tests.h
+/// \brief Common definitions for test applications
+//
+// Author: Lasse Collin
+//
+// This file has been put into the public domain.
+// You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#ifndef LZMA_TESTS_H
+#define LZMA_TESTS_H
+
+#include "sysdefs.h"
+#include "tuklib_integer.h"
+#include "lzma.h"
+
+#include <stdio.h>
+
+#define memcrap(buf, size) memset(buf, 0xFD, size)
+
+#define expect(test) ((test) ? 0 : (fprintf(stderr, "%s:%d: %s\n", \
+ __FILE__, __LINE__, #test), abort(), 0))
+
+#define succeed(test) expect(!(test))
+
+#define fail(test) expect(test)
+
+
+static inline const char *
+lzma_ret_sym(lzma_ret ret)
+{
+ if ((unsigned int)(ret) > LZMA_PROG_ERROR)
+ return "UNKNOWN_ERROR";
+
+ static const char *msgs[] = {
+ "LZMA_OK",
+ "LZMA_STREAM_END",
+ "LZMA_NO_CHECK",
+ "LZMA_UNSUPPORTED_CHECK",
+ "LZMA_GET_CHECK",
+ "LZMA_MEM_ERROR",
+ "LZMA_MEMLIMIT_ERROR",
+ "LZMA_FORMAT_ERROR",
+ "LZMA_OPTIONS_ERROR",
+ "LZMA_DATA_ERROR",
+ "LZMA_BUF_ERROR",
+ "LZMA_PROG_ERROR"
+ };
+
+ return msgs[ret];
+}
+
+
+static inline bool
+coder_loop(lzma_stream *strm, uint8_t *in, size_t in_size,
+ uint8_t *out, size_t out_size,
+ lzma_ret expected_ret, lzma_action finishing_action)
+{
+ size_t in_left = in_size;
+ size_t out_left = out_size > 0 ? out_size + 1 : 0;
+ lzma_action action = LZMA_RUN;
+ lzma_ret ret;
+
+ strm->next_in = NULL;
+ strm->avail_in = 0;
+ strm->next_out = NULL;
+ strm->avail_out = 0;
+
+ while (true) {
+ if (in_left > 0) {
+ if (--in_left == 0)
+ action = finishing_action;
+
+ strm->next_in = in++;
+ strm->avail_in = 1;
+ }
+
+ if (out_left > 0) {
+ --out_left;
+ strm->next_out = out++;
+ strm->avail_out = 1;
+ }
+
+ ret = lzma_code(strm, action);
+ if (ret != LZMA_OK)
+ break;
+ }
+
+ bool error = false;
+
+ if (ret != expected_ret)
+ error = true;
+
+ if (expected_ret == LZMA_STREAM_END) {
+ if (strm->total_in != in_size || strm->total_out != out_size)
+ error = true;
+ } else {
+ if (strm->total_in != in_size || strm->total_out != out_size)
+ error = true;
+ }
+
+ return error;
+}
+
+
+static inline bool
+decoder_loop_ret(lzma_stream *strm, uint8_t *in, size_t in_size,
+ lzma_ret expected_ret)
+{
+ return coder_loop(strm, in, in_size, NULL, 0, expected_ret, LZMA_RUN);
+}
+
+
+static inline bool
+decoder_loop(lzma_stream *strm, uint8_t *in, size_t in_size)
+{
+ return coder_loop(strm, in, in_size, NULL, 0,
+ LZMA_STREAM_END, LZMA_RUN);
+}
+
+#endif
diff --git a/version.sh b/version.sh
new file mode 100755
index 0000000..40d0493
--- /dev/null
+++ b/version.sh
@@ -0,0 +1,24 @@
+#!/bin/sh
+#
+#############################################################################
+#
+# Get the version string from version.h and print it out without
+# trailing newline. This makes it suitable for use in configure.ac.
+#
+#############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+#############################################################################
+
+sed -n 's/LZMA_VERSION_STABILITY_ALPHA/alpha/
+ s/LZMA_VERSION_STABILITY_BETA/beta/
+ s/LZMA_VERSION_STABILITY_STABLE//
+ s/^#define LZMA_VERSION_[MPS][AIT][AJNT][A-Z]* //p' \
+ src/liblzma/api/lzma/version.h \
+ | tr '\n' '|' \
+ | sed 's/|/./; s/|/./; s/|//g' \
+ | tr -d '\n'
diff --git a/windows/INSTALL-Windows.txt b/windows/INSTALL-Windows.txt
new file mode 100644
index 0000000..b5ac963
--- /dev/null
+++ b/windows/INSTALL-Windows.txt
@@ -0,0 +1,131 @@
+
+Building XZ Utils on Windows
+============================
+
+Introduction
+------------
+
+ This document explains shortly where to get and how to install the
+ build tool that are needed to build XZ Utils on Windows. The final
+ binary package will be standalone in sense that it will depend only
+ on DLLs that are included in all Windows installations.
+
+ These instructions don't apply to Cygwin. XZ Utils can be built under
+ Cygwin in the same way as many other packages.
+
+ These instructions don't apply to MinGW and MSYS developers either,
+ who may want to package XZ Utils for MinGW or MSYS distributions.
+ You know who you are, and will probably use quite different configure
+ options etc. than what is described here.
+
+
+Installing the toolchain(s)
+---------------------------
+
+ Some of the following is needed:
+ - MSYS is always needed to use the GNU Autotools based build system.
+ - MinGW builds 32-bit x86 binaries.
+ - MingW-w32 builds 32-bit x86 executables too.
+ - MinGW-w64 builds 64-bit x86-64 binaries.
+
+ So you need to pick between MinGW and MinGW-w32 when building
+ 32-bit version. You don't need both.
+
+ You might find 7-Zip <http://7-zip.org/> handy when extracting
+ some files (especially the .tar.lzma files). The ready-made
+ build script will also use 7-Zip to create the distributable
+ .zip and .7z files.
+
+ I used the following directory structure but you can use whatever
+ you want. Just note that I will use these in my examples. Each of
+ these should have a subdirectory "bin":
+
+ C:\devel\tools\msys
+ C:\devel\tools\mingw
+ C:\devel\tools\mingw-w32
+ C:\devel\tools\mingw-w64
+
+
+Installing MSYS
+
+ You can download MSYS from MinGW's Sourceforge page:
+
+ http://sourceforge.net/projects/mingw/files/
+
+ It's under "MSYS Base System". I recommend using MSYS 1.0.11
+ (MSYS-1.0.11.exe or msysCORE-1.0.11-bin.tar.gz) because that
+ package includes all the required tools. At least some of the
+ later versions include only a subset and thus you would need to
+ download the rest separately. The old version will work fine for
+ building XZ Utils.
+
+ You can use either the .exe or .tar.gz package. I prefer .tar.gz,
+ because it can be extracted into any directory and later removed
+ without worrying about uninstallers.
+
+
+Installing MinGW
+
+ You can download the required packages from MinGW's Sourceforge page:
+
+ http://sourceforge.net/projects/mingw/files/
+
+ These version numbers were the latest when I wrote this document, but
+ you probably should pick the latest versions:
+
+ MinGW Runtime -> mingwrt-3.17-mingw32-dev.tar.gz
+ MinGW API for MS-Windows -> w32api-3.14-mingw32-dev.tar.gz
+ GNU Binutils -> binutils-2.20-1-bin.tar.gz
+ GCC Version 4 -> gcc-full-4.4.0-mingw32-bin-2.tar.lzma
+
+ The full GCC package is quite big, but if you want a smaller
+ download, you will need to download more than one file, so I'm
+ using the full package in this document for simplicity.
+
+ Extract the packages in the above order, possibly overwriting files
+ from packages that were extracted earlier.
+
+
+Installing MinGW-w32 or MinGW-w64
+
+ You can find the latest MinGW-w32 and MinGW-w64 builds here:
+
+ http://sourceforge.net/projects/mingw-w64/files/
+
+ Locate the appropriate files:
+
+ Toolchains targeting Win32 -> mingw-w32-*-mingw*.zip
+ Toolchains targeting Win64 -> mingw-w64-*-mingw*.zip
+
+ I don't know what is the most recommended one. I used sezero's
+ versions from "Personal Builds", since they seemed to have
+ a stable GCC (judging from the GCC version number only).
+
+ If you will install both MinGW-w32 and MinGW-w64, remember to
+ extract them into different directories.
+
+
+Building XZ Utils
+-----------------
+
+ Start MSYS by going to the directory C:\devel\tools\msys and running
+ msys.bat there (double-click or use command prompt). It will start
+ at "home" directory, which is C:\devel\tools\msys\home\YourUserName.
+
+ If you have xz-5.x.x.tar.gz in C:\devel, you should be able to build
+ it now with the following commands:
+
+ cd /c/devel
+ tar xzf xz-5.x.x.tar.gz
+ cd xz-5.x.x
+ sh windows/build.sh
+
+ If you used some other directory than C:\devel\tools for the build
+ tools, edit the variables near the beginning of build.sh first.
+
+ If you want to build manually, read the buildit() function in
+ build.sh. Look especially at the latter configure invocation.
+
+ Be patient. Running configure and other scripts used by the build
+ system is (very) slow under Windows.
+
diff --git a/windows/README-Windows.txt b/windows/README-Windows.txt
new file mode 100644
index 0000000..536d08f
--- /dev/null
+++ b/windows/README-Windows.txt
@@ -0,0 +1,115 @@
+
+XZ Utils for Windows
+====================
+
+Introduction
+------------
+
+ This package includes command line tools (xz.exe and a few others)
+ and the liblzma compression library from XZ Utils. You can find the
+ latest version and full source code from <http://tukaani.org/xz/>.
+
+ The parts of the XZ Utils source code, that are relevant to this
+ binary package, are in the public domain. XZ Utils have been built
+ for this package with MinGW and linked statically against the MinGW
+ runtime and w32api packages.
+
+ FIXME: Add license info about MinGW runtime and w32api.
+
+
+Package contents
+----------------
+
+ All executables and libraries in this package require msvcrt.dll.
+ It's included in all recent Windows versions. On Windows 95 it
+ might be missing, but once you get it somewhere, XZ Utils should
+ run even on Windows 95.
+
+ There are two different versions of the executable and library files.
+ There is one directory for each type of binaries:
+
+ bin_i486 32-bit x86 (i486 and up), Windows 95 and later
+ bin_x86-64 64-bit x86-64, Windows XP and later
+
+ Each of the above directories have the following files:
+
+ *.exe Command line tools. (It's useless to double-click
+ these; use the command prompt instead.) These have
+ been linked statically against liblzma, so they
+ don't require liblzma.dll. Thus, you can copy e.g.
+ xz.exe to a directory that is in PATH without copying
+ any other files from this package.
+
+ liblzma.dll Shared version of the liblzma compression library.
+ This file is mostly useful to developers, although
+ some non-developers might use it to upgrade their
+ copy of liblzma.
+
+ liblzma.a Static version of the liblzma compression library.
+ This file is useful only for developers.
+
+ The rest of the directories contain architecture-independent files:
+
+ doc Documentation in the plain text (TXT) format. The
+ manuals of the command line tools are provided also
+ in the PDF format. liblzma.def is in this directory
+ too.
+
+ include C header files for liblzma. These should be
+ compatible with most C and C++ compilers. If you
+ have problems, try to fix it and send your fixes
+ upstream, or at least report a bug, thanks.
+
+
+Linking against liblzma
+-----------------------
+
+MinGW
+
+ If you use MinGW, linking against liblzma.dll or liblzma.a should
+ be straightforward. You don't need an import library to link
+ against liblzma.dll, and for static linking, you don't need to
+ worry about the LZMA_API_STATIC macro.
+
+ Note that the MinGW distribution includes liblzma. If you are
+ building packages that will be part of the MinGW distribution, you
+ probably should use the version of liblzma shipped in MinGW instead
+ of this package.
+
+
+Microsoft Visual C++
+
+ To link against liblzma.dll, you need to create an import library
+ first. You need the "lib" command from MSVC and liblzma.def from
+ the "doc" directory of this package. Here is the command that works
+ on 32-bit x86:
+
+ lib /def:liblzma.def /out:liblzma.lib /machine:ix86
+
+ On x86-64, the /machine argument has to naturally be changed:
+
+ lib /def:liblzma.def /out:liblzma.lib /machine:x64
+
+ Linking against static liblzma should work too. Rename liblzma.a
+ to e.g. liblzma_static.lib and tell MSVC to link against it. You
+ also need to tell lzma.h to not use __declspec(dllimport) by defining
+ the macro LZMA_API_STATIC. You can do it either in the C/C++ code
+
+ #define LZMA_API_STATIC
+ #include <lzma.h>
+
+ or by adding it to compiler options.
+
+
+Other compilers
+
+ If you are using some other compiler, see its documentation how to
+ create an import library (if it is needed). If it is simple, I
+ might consider including the instructions here.
+
+
+Reporting bugs
+--------------
+
+ Report bugs to <lasse.collin@tukaani.org> (in English or Finnish).
+
diff --git a/windows/build.sh b/windows/build.sh
new file mode 100755
index 0000000..3c762dc
--- /dev/null
+++ b/windows/build.sh
@@ -0,0 +1,188 @@
+#!/bin/sh
+#
+###############################################################################
+#
+# Build a binary package on Windows with MinGW and MSYS
+#
+# Set the paths where MinGW, Mingw-w32, or MinGW-w64 are installed. If both
+# MinGW and MinGW-w32 are specified, MinGW will be used. If there is no
+# 32-bit or 64-bit compiler at all, it is simply skipped.
+#
+# Optionally, 7-Zip is used to create the final .zip and .7z packages.
+# If you have installed it in the default directory, this script should
+# find it automatically. Otherwise adjust the path manually.
+#
+# If you want to use a cross-compiler e.g. on GNU/Linux, this script won't
+# work out of the box. You need to omit "make check" commands and replace
+# u2d with some other tool to convert newlines from LF to CR+LF. You will
+# also need to pass the --host option to configure.
+#
+###############################################################################
+#
+# Author: Lasse Collin
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+MINGW_DIR=/c/devel/tools/mingw
+MINGW_W32_DIR=/c/devel/tools/mingw-w32
+MINGW_W64_DIR=/c/devel/tools/mingw-w64
+
+for SEVENZ_EXE in "$PROGRAMW6432/7-Zip/7z.exe" "$PROGRAMFILES/7-Zip/7z.exe" \
+ "/c/Program Files/7-Zip/7z.exe"
+do
+ [ -x "$SEVENZ_EXE" ] && break
+done
+
+
+# Abort immediately if something goes wrong.
+set -e
+
+# White spaces in directory names may break things so catch them immediately.
+case $(pwd) in
+ ' ' | ' ' | '
+') echo "Error: White space in the directory name" >&2; exit 1 ;;
+esac
+
+# This script can be run either at the top-level directory of the package
+# or in the same directory containing this script.
+if [ ! -f windows/build.sh ]; then
+ cd ..
+ if [ ! -f windows/build.sh ]; then
+ echo "You are in a wrong directory." >&2
+ exit 1
+ fi
+fi
+
+# Run configure and copy the binaries to the given directory.
+#
+# The first argument is the directory where to copy the binaries.
+# The rest of the arguments are passed to configure.
+buildit()
+{
+ DESTDIR=$1
+ BUILD=$2
+ CFLAGS=$3
+
+ # Clean up if it was already configured.
+ [ -f Makefile ] && make distclean
+
+ # Build the size-optimized binaries. Note that I don't want to
+ # provide size-optimized liblzma (shared nor static), because
+ # that isn't thread-safe now, and depending on bunch of things,
+ # maybe it will never be on Windows (pthreads-win32 helps but
+ # static liblzma might bit a bit tricky with it).
+ ./configure \
+ --prefix= \
+ --disable-nls \
+ --disable-threads \
+ --disable-shared \
+ --enable-small \
+ --build="$BUILD" \
+ CFLAGS="$CFLAGS -Os"
+ make check
+
+ mkdir -pv "$DESTDIR"
+ cp -v src/xzdec/{xz,lzma}dec.exe src/lzmainfo/lzmainfo.exe "$DESTDIR"
+
+ make distclean
+
+ # Build the normal speed-optimized binaries. Note that while
+ # --disable-threads has been documented to make some things
+ # thread-unsafe, it's not actually true with this combination
+ # of configure flags in XZ Utils 5.0.x. Things can (and probably
+ # will) change after 5.0.x, and this script will be updated too.
+ ./configure \
+ --prefix= \
+ --disable-nls \
+ --disable-threads \
+ --enable-dynamic=no \
+ --build="$BUILD" \
+ CFLAGS="$CFLAGS -O2"
+ make check
+
+ cp -v src/xz/xz.exe src/liblzma/.libs/liblzma.a "$DESTDIR"
+ cp -v src/liblzma/.libs/liblzma-*.dll "$DESTDIR/liblzma.dll"
+
+ strip -v "$DESTDIR/"*
+}
+
+# Copy files and convert newlines from LF to CR+LF. Optinally add a suffix
+# to the destination filename.
+#
+# The first argument is the destination directory. The second argument is
+# the suffix to append to the filenames; use empty string if no extra suffix
+# is wanted. The rest of the arguments are actual the filenames.
+txtcp()
+{
+ DESTDIR=$1
+ SUFFIX=$2
+ shift 2
+ for SRCFILE; do
+ DESTFILE="$DESTDIR/${SRCFILE##*/}$SUFFIX"
+ echo "Converting \`$SRCFILE' -> \`$DESTFILE'"
+ u2d < "$SRCFILE" > "$DESTFILE"
+ done
+}
+
+# FIXME: Make sure that we don't get i686 or i586 code from the runtime.
+# Actually i586 would be fine, but i686 probably not if the idea is to
+# support even Win95.
+#
+# FIXME: Using i486 in the configure triplet may be wrong.
+if [ -d "$MINGW_DIR" ]; then
+ # 32-bit x86, Win95 or later, using MinGW
+ PATH=$MINGW_DIR/bin:$PATH \
+ buildit \
+ pkg/bin_i486 \
+ i486-pc-mingw32 \
+ '-march=i486 -mtune=generic'
+elif [ -d "$MINGW_W32_DIR" ]; then
+ # 32-bit x86, Win95 or later, using MinGW-w32
+ PATH=$MINGW_W32_DIR/bin:$MINGW_W32_DIR/i686-w64-mingw32/bin:$PATH \
+ buildit \
+ pkg/bin_i486 \
+ i486-w64-mingw32 \
+ '-march=i486 -mtune=generic'
+fi
+
+if [ -d "$MINGW_W64_DIR" ]; then
+ # 64-bit x86, WinXP or later, using MinGW-w64
+ PATH=$MINGW_W64_DIR/bin:$MINGW_W64_DIR/x86_64-w64-mingw32/bin:$PATH \
+ buildit \
+ pkg/bin_x86-64 \
+ x86_64-w64-mingw32 \
+ '-march=x86-64 -mtune=generic'
+fi
+
+# Copy the headers, the .def file, and the docs.
+# They are the same for all architectures and builds.
+mkdir -pv pkg/{include/lzma,doc/manuals}
+txtcp pkg/include "" src/liblzma/api/lzma.h
+txtcp pkg/include/lzma "" src/liblzma/api/lzma/*.h
+txtcp pkg/doc "" src/liblzma/liblzma.def
+txtcp pkg/doc .txt AUTHORS COPYING NEWS README THANKS TODO
+txtcp pkg/doc "" doc/*.txt
+txtcp pkg/doc/manuals "" doc/man/txt/{xz,xzdec,lzmainfo}.txt
+cp -v doc/man/pdf-*/{xz,xzdec,lzmainfo}-*.pdf pkg/doc/manuals
+txtcp pkg "" windows/README-Windows.txt
+
+# Create the package. This requires 7z.exe from 7-Zip. If it wasn't found,
+# this step is skipped and you have to zip it yourself.
+VER=$(sh version.sh)
+cd pkg
+if [ -x "$SEVENZ_EXE" ]; then
+ "$SEVENZ_EXE" a -tzip ../xz-$VER-windows.zip *
+ "$SEVENZ_EXE" a ../xz-$VER-windows.7z *
+else
+ echo
+ echo "NOTE: 7z.exe was not found. xz-$VER-windows.zip"
+ echo " and xz-$VER-windows.7z were not created."
+ echo " You can create them yourself from the pkg directory."
+fi
+
+echo
+echo "Build completed successfully."
+echo
