src/objects/string.h

// Copyright 2017 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#ifndef V8_OBJECTS_STRING_H_
#define V8_OBJECTS_STRING_H_

#include "src/base/bits.h"
#include "src/objects/name.h"
#include "src/unicode-decoder.h"

// Has to be the last include (doesn't have include guards):
#include "src/objects/object-macros.h"

namespace v8 {
namespace internal {

class BigInt;

enum AllowNullsFlag { ALLOW_NULLS, DISALLOW_NULLS };
enum RobustnessFlag { ROBUST_STRING_TRAVERSAL, FAST_STRING_TRAVERSAL };

// The characteristics of a string are stored in its map.  Retrieving these
// few bits of information is moderately expensive, involving two memory
// loads where the second is dependent on the first.  To improve efficiency
// the shape of the string is given its own class so that it can be retrieved
// once and used for several string operations.  A StringShape is small enough
// to be passed by value and is immutable, but be aware that flattening a
// string can potentially alter its shape.  Also be aware that a GC caused by
// something else can alter the shape of a string due to ConsString
// shortcutting.  Keeping these restrictions in mind has proven to be error-
// prone and so we no longer put StringShapes in variables unless there is a
// concrete performance benefit at that particular point in the code.
class StringShape BASE_EMBEDDED {
 public:
  inline explicit StringShape(const String* s);
  inline explicit StringShape(Map* s);
  inline explicit StringShape(InstanceType t);
  inline bool IsSequential();
  inline bool IsExternal();
  inline bool IsCons();
  inline bool IsSliced();
  inline bool IsThin();
  inline bool IsIndirect();
  inline bool IsExternalOneByte();
  inline bool IsExternalTwoByte();
  inline bool IsSequentialOneByte();
  inline bool IsSequentialTwoByte();
  inline bool IsInternalized();
  inline StringRepresentationTag representation_tag();
  inline uint32_t encoding_tag();
  inline uint32_t full_representation_tag();
  inline bool HasOnlyOneByteChars();
#ifdef DEBUG
  inline uint32_t type() { return type_; }
  inline void invalidate() { valid_ = false; }
  inline bool valid() { return valid_; }
#else
  inline void invalidate() {}
#endif

 private:
  uint32_t type_;
#ifdef DEBUG
  inline void set_valid() { valid_ = true; }
  bool valid_;
#else
  inline void set_valid() {}
#endif
};

// The String abstract class captures JavaScript string values:
//
// Ecma-262:
//  4.3.16 String Value
//    A string value is a member of the type String and is a finite
//    ordered sequence of zero or more 16-bit unsigned integer values.
//
// All string values have a length field.
class String : public Name {
 public:
  enum Encoding { ONE_BYTE_ENCODING, TWO_BYTE_ENCODING };

  class SubStringRange {
   public:
    explicit inline SubStringRange(String* string, int first = 0,
                                   int length = -1);
    class iterator;
    inline iterator begin();
    inline iterator end();

   private:
    String* string_;
    int first_;
    int length_;
  };

  // Representation of the flat content of a String.
  // A non-flat string doesn't have flat content.
  // A flat string has content that's encoded as a sequence of either
  // one-byte chars or two-byte UC16.
  // Returned by String::GetFlatContent().
  class FlatContent {
   public:
    // Returns true if the string is flat and this structure contains content.
    bool IsFlat() const { return state_ != NON_FLAT; }
    // Returns true if the structure contains one-byte content.
    bool IsOneByte() const { return state_ == ONE_BYTE; }
    // Returns true if the structure contains two-byte content.
    bool IsTwoByte() const { return state_ == TWO_BYTE; }

    // Return the one byte content of the string. Only use if IsOneByte()
    // returns true.
    Vector<const uint8_t> ToOneByteVector() const {
      DCHECK_EQ(ONE_BYTE, state_);
      return Vector<const uint8_t>(onebyte_start, length_);
    }
    // Return the two-byte content of the string. Only use if IsTwoByte()
    // returns true.
    Vector<const uc16> ToUC16Vector() const {
      DCHECK_EQ(TWO_BYTE, state_);
      return Vector<const uc16>(twobyte_start, length_);
    }

    uc16 Get(int i) const {
      DCHECK(i < length_);
      DCHECK(state_ != NON_FLAT);
      if (state_ == ONE_BYTE) return onebyte_start[i];
      return twobyte_start[i];
    }

    bool UsesSameString(const FlatContent& other) const {
      return onebyte_start == other.onebyte_start;
    }

   private:
    enum State { NON_FLAT, ONE_BYTE, TWO_BYTE };

    // Constructors only used by String::GetFlatContent().
    explicit FlatContent(const uint8_t* start, int length)
        : onebyte_start(start), length_(length), state_(ONE_BYTE) {}
    explicit FlatContent(const uc16* start, int length)
        : twobyte_start(start), length_(length), state_(TWO_BYTE) {}
    FlatContent() : onebyte_start(nullptr), length_(0), state_(NON_FLAT) {}

    union {
      const uint8_t* onebyte_start;
      const uc16* twobyte_start;
    };
    int length_;
    State state_;

    friend class String;
    friend class IterableSubString;
  };

  template <typename Char>
  V8_INLINE Vector<const Char> GetCharVector();

  // Get and set the length of the string.
  inline int length() const;
  inline void set_length(int value);

  // Get and set the length of the string using acquire loads and release
  // stores.
  inline int synchronized_length() const;
  inline void synchronized_set_length(int value);

  // Returns whether this string has only one-byte chars, i.e. all of them can
  // be one-byte encoded.  This might be the case even if the string is
  // two-byte.  Such strings may appear when the embedder prefers
  // two-byte external representations even for one-byte data.
  inline bool IsOneByteRepresentation() const;
  inline bool IsTwoByteRepresentation() const;

  // Cons and slices have an encoding flag that may not represent the actual
  // encoding of the underlying string.  This is taken into account here.
  // Requires: this->IsFlat()
  inline bool IsOneByteRepresentationUnderneath();
  inline bool IsTwoByteRepresentationUnderneath();

  // NOTE: this should be considered only a hint.  False negatives are
  // possible.
  inline bool HasOnlyOneByteChars();

  // Get and set individual two byte chars in the string.
  inline void Set(int index, uint16_t value);
  // Get individual two byte char in the string.  Repeated calls
  // to this method are not efficient unless the string is flat.
  V8_INLINE uint16_t Get(int index);

  // ES6 section 7.1.3.1 ToNumber Applied to the String Type
  static Handle<Object> ToNumber(Isolate* isolate, Handle<String> subject);

  // Flattens the string.  Checks first inline to see if it is
  // necessary.  Does nothing if the string is not a cons string.
  // Flattening allocates a sequential string with the same data as
  // the given string and mutates the cons string to a degenerate
  // form, where the first component is the new sequential string and
  // the second component is the empty string.  If allocation fails,
  // this function returns a failure.  If flattening succeeds, this
  // function returns the sequential string that is now the first
  // component of the cons string.
  //
  // Degenerate cons strings are handled specially by the garbage
  // collector (see IsShortcutCandidate).

  static inline Handle<String> Flatten(Isolate* isolate, Handle<String> string,
                                       PretenureFlag pretenure = NOT_TENURED);

  // Tries to return the content of a flat string as a structure holding either
  // a flat vector of char or of uc16.
  // If the string isn't flat, and therefore doesn't have flat content, the
  // returned structure will report so, and can't provide a vector of either
  // kind.
  FlatContent GetFlatContent();

  // Returns the parent of a sliced string or first part of a flat cons string.
  // Requires: StringShape(this).IsIndirect() && this->IsFlat()
  inline String* GetUnderlying();

  // String relational comparison, implemented according to ES6 section 7.2.11
  // Abstract Relational Comparison (step 5): The comparison of Strings uses a
  // simple lexicographic ordering on sequences of code unit values. There is no
  // attempt to use the more complex, semantically oriented definitions of
  // character or string equality and collating order defined in the Unicode
  // specification. Therefore String values that are canonically equal according
  // to the Unicode standard could test as unequal. In effect this algorithm
  // assumes that both Strings are already in normalized form. Also, note that
  // for strings containing supplementary characters, lexicographic ordering on
  // sequences of UTF-16 code unit values differs from that on sequences of code
  // point values.
  V8_WARN_UNUSED_RESULT static ComparisonResult Compare(Isolate* isolate,
                                                        Handle<String> x,
                                                        Handle<String> y);

  // Perform ES6 21.1.3.8, including checking arguments.
  static Object* IndexOf(Isolate* isolate, Handle<Object> receiver,
                         Handle<Object> search, Handle<Object> position);
  // Perform string match of pattern on subject, starting at start index.
  // Caller must ensure that 0 <= start_index <= sub->length(), as this does not
  // check any arguments.
  static int IndexOf(Isolate* isolate, Handle<String> receiver,
                     Handle<String> search, int start_index);

  static Object* LastIndexOf(Isolate* isolate, Handle<Object> receiver,
                             Handle<Object> search, Handle<Object> position);

  // Encapsulates logic related to a match and its capture groups as required
  // by GetSubstitution.
  class Match {
   public:
    virtual Handle<String> GetMatch() = 0;
    virtual Handle<String> GetPrefix() = 0;
    virtual Handle<String> GetSuffix() = 0;

    // A named capture can be invalid (if it is not specified in the pattern),
    // unmatched (specified but not matched in the current string), and matched.
    enum CaptureState { INVALID, UNMATCHED, MATCHED };

    virtual int CaptureCount() = 0;
    virtual bool HasNamedCaptures() = 0;
    virtual MaybeHandle<String> GetCapture(int i, bool* capture_exists) = 0;
    virtual MaybeHandle<String> GetNamedCapture(Handle<String> name,
                                                CaptureState* state) = 0;

    virtual ~Match() {}
  };

  // ES#sec-getsubstitution
  // GetSubstitution(matched, str, position, captures, replacement)
  // Expand the $-expressions in the string and return a new string with
  // the result.
  // A {start_index} can be passed to specify where to start scanning the
  // replacement string.
  V8_WARN_UNUSED_RESULT static MaybeHandle<String> GetSubstitution(
      Isolate* isolate, Match* match, Handle<String> replacement,
      int start_index = 0);

  // String equality operations.
  inline bool Equals(String* other);
  inline static bool Equals(Isolate* isolate, Handle<String> one,
                            Handle<String> two);
  bool IsUtf8EqualTo(Vector<const char> str, bool allow_prefix_match = false);

  // Dispatches to Is{One,Two}ByteEqualTo.
  template <typename Char>
  bool IsEqualTo(Vector<const Char> str);

  bool IsOneByteEqualTo(Vector<const uint8_t> str);
  bool IsTwoByteEqualTo(Vector<const uc16> str);

  // Return a UTF8 representation of the string.  The string is null
  // terminated but may optionally contain nulls.  Length is returned
  // in length_output if length_output is not a null pointer  The string
  // should be nearly flat, otherwise the performance of this method may
  // be very slow (quadratic in the length).  Setting robustness_flag to
  // ROBUST_STRING_TRAVERSAL invokes behaviour that is robust  This means it
  // handles unexpected data without causing assert failures and it does not
  // do any heap allocations.  This is useful when printing stack traces.
  std::unique_ptr<char[]> ToCString(AllowNullsFlag allow_nulls,
                                    RobustnessFlag robustness_flag, int offset,
                                    int length, int* length_output = 0);
  std::unique_ptr<char[]> ToCString(
      AllowNullsFlag allow_nulls = DISALLOW_NULLS,
      RobustnessFlag robustness_flag = FAST_STRING_TRAVERSAL,
      int* length_output = 0);

  bool ComputeArrayIndex(uint32_t* index);

  // Externalization.
  bool MakeExternal(v8::String::ExternalStringResource* resource);
  bool MakeExternal(v8::String::ExternalOneByteStringResource* resource);
  bool SupportsExternalization();

  // Conversion.
  inline bool AsArrayIndex(uint32_t* index);
  uint32_t inline ToValidIndex(Object* number);

  // Trimming.
  enum TrimMode { kTrim, kTrimStart, kTrimEnd };
  static Handle<String> Trim(Isolate* isolate, Handle<String> string,
                             TrimMode mode);

  DECL_CAST(String)

  void PrintOn(FILE* out);

  // For use during stack traces.  Performs rudimentary sanity check.
  bool LooksValid();

  // Dispatched behavior.
  void StringShortPrint(StringStream* accumulator, bool show_details = true);
  void PrintUC16(std::ostream& os, int start = 0, int end = -1);  // NOLINT
#if defined(DEBUG) || defined(OBJECT_PRINT)
  char* ToAsciiArray();
#endif
  DECL_PRINTER(String)
  DECL_VERIFIER(String)

  inline bool IsFlat();

  // Layout description.
  static const int kLengthOffset = Name::kSize;
  static const int kSize = kLengthOffset + kPointerSize;

  // Max char codes.
  static const int32_t kMaxOneByteCharCode = unibrow::Latin1::kMaxChar;
  static const uint32_t kMaxOneByteCharCodeU = unibrow::Latin1::kMaxChar;
  static const int kMaxUtf16CodeUnit = 0xffff;
  static const uint32_t kMaxUtf16CodeUnitU = kMaxUtf16CodeUnit;
  static const uc32 kMaxCodePoint = 0x10ffff;

  // Maximal string length.
  // The max length is different on 32 and 64 bit platforms. Max length for a
  // 32-bit platform is ~268.4M chars. On 64-bit platforms, max length is
  // ~1.073B chars. The limit on 64-bit is so that SeqTwoByteString::kMaxSize
  // can fit in a 32bit int: 2^31 - 1 is the max positive int, minus one bit as
  // each char needs two bytes, subtract 24 bytes for the string header size.

  // See include/v8.h for the definition.
  static const int kMaxLength = v8::String::kMaxLength;
  static_assert(kMaxLength <= (Smi::kMaxValue / 2 - kSize),
                "Unexpected max String length");

  // Max length for computing hash. For strings longer than this limit the
  // string length is used as the hash value.
  static const int kMaxHashCalcLength = 16383;

  // Limit for truncation in short printing.
  static const int kMaxShortPrintLength = 1024;

  // Support for regular expressions.
  const uc16* GetTwoByteData(unsigned start);

  // Helper function for flattening strings.
  template <typename sinkchar>
  static void WriteToFlat(String* source, sinkchar* sink, int from, int to);

  // The return value may point to the first aligned word containing the first
  // non-one-byte character, rather than directly to the non-one-byte character.
  // If the return value is >= the passed length, the entire string was
  // one-byte.
  static inline int NonAsciiStart(const char* chars, int length) {
    const char* start = chars;
    const char* limit = chars + length;

    if (length >= kIntptrSize) {
      // Check unaligned bytes.
      while (!IsAligned(reinterpret_cast<intptr_t>(chars), sizeof(uintptr_t))) {
        if (static_cast<uint8_t>(*chars) > unibrow::Utf8::kMaxOneByteChar) {
          return static_cast<int>(chars - start);
        }
        ++chars;
      }
      // Check aligned words.
      DCHECK_EQ(unibrow::Utf8::kMaxOneByteChar, 0x7F);
      const uintptr_t non_one_byte_mask = kUintptrAllBitsSet / 0xFF * 0x80;
      while (chars + sizeof(uintptr_t) <= limit) {
        if (*reinterpret_cast<const uintptr_t*>(chars) & non_one_byte_mask) {
          return static_cast<int>(chars - start);
        }
        chars += sizeof(uintptr_t);
      }
    }
    // Check remaining unaligned bytes.
    while (chars < limit) {
      if (static_cast<uint8_t>(*chars) > unibrow::Utf8::kMaxOneByteChar) {
        return static_cast<int>(chars - start);
      }
      ++chars;
    }

    return static_cast<int>(chars - start);
  }

  static inline bool IsAscii(const char* chars, int length) {
    return NonAsciiStart(chars, length) >= length;
  }

  static inline bool IsAscii(const uint8_t* chars, int length) {
    return NonAsciiStart(reinterpret_cast<const char*>(chars), length) >=
           length;
  }

  static inline int NonOneByteStart(const uc16* chars, int length) {
    const uc16* limit = chars + length;
    const uc16* start = chars;
    while (chars < limit) {
      if (*chars > kMaxOneByteCharCodeU) return static_cast<int>(chars - start);
      ++chars;
    }
    return static_cast<int>(chars - start);
  }

  static inline bool IsOneByte(const uc16* chars, int length) {
    return NonOneByteStart(chars, length) >= length;
  }

  template <class Visitor>
  static inline ConsString* VisitFlat(Visitor* visitor, String* string,
                                      int offset = 0);

  static Handle<FixedArray> CalculateLineEnds(Isolate* isolate,
                                              Handle<String> string,
                                              bool include_ending_line);

 private:
  friend class Name;
  friend class StringTableInsertionKey;
  friend class InternalizedStringKey;

  static Handle<String> SlowFlatten(Isolate* isolate, Handle<ConsString> cons,
                                    PretenureFlag tenure);

  // Slow case of String::Equals.  This implementation works on any strings
  // but it is most efficient on strings that are almost flat.
  bool SlowEquals(String* other);

  static bool SlowEquals(Isolate* isolate, Handle<String> one,
                         Handle<String> two);

  // Slow case of AsArrayIndex.
  V8_EXPORT_PRIVATE bool SlowAsArrayIndex(uint32_t* index);

  // Compute and set the hash code.
  uint32_t ComputeAndSetHash(Isolate* isolate);

  DISALLOW_IMPLICIT_CONSTRUCTORS(String);
};

// The SeqString abstract class captures sequential string values.
class SeqString : public String {
 public:
  DECL_CAST(SeqString)

  // Layout description.
  static const int kHeaderSize = String::kSize;

  // Truncate the string in-place if possible and return the result.
  // In case of new_length == 0, the empty string is returned without
  // truncating the original string.
  V8_WARN_UNUSED_RESULT static Handle<String> Truncate(Handle<SeqString> string,
                                                       int new_length);

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(SeqString);
};

// The OneByteString class captures sequential one-byte string objects.
// Each character in the OneByteString is an one-byte character.
class SeqOneByteString : public SeqString {
 public:
  static const bool kHasOneByteEncoding = true;

  // Dispatched behavior.
  inline uint16_t SeqOneByteStringGet(int index);
  inline void SeqOneByteStringSet(int index, uint16_t value);

  // Get the address of the characters in this string.
  inline Address GetCharsAddress();

  inline uint8_t* GetChars();

  // Clear uninitialized padding space. This ensures that the snapshot content
  // is deterministic.
  void clear_padding();

  DECL_CAST(SeqOneByteString)

  // Garbage collection support.  This method is called by the
  // garbage collector to compute the actual size of an OneByteString
  // instance.
  inline int SeqOneByteStringSize(InstanceType instance_type);

  // Computes the size for an OneByteString instance of a given length.
  static int SizeFor(int length) {
    return OBJECT_POINTER_ALIGN(kHeaderSize + length * kCharSize);
  }

  // Maximal memory usage for a single sequential one-byte string.
  static const int kMaxCharsSize = kMaxLength;
  static const int kMaxSize = OBJECT_POINTER_ALIGN(kMaxCharsSize + kHeaderSize);
  STATIC_ASSERT((kMaxSize - kHeaderSize) >= String::kMaxLength);

  class BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(SeqOneByteString);
};

// The TwoByteString class captures sequential unicode string objects.
// Each character in the TwoByteString is a two-byte uint16_t.
class SeqTwoByteString : public SeqString {
 public:
  static const bool kHasOneByteEncoding = false;

  // Dispatched behavior.
  inline uint16_t SeqTwoByteStringGet(int index);
  inline void SeqTwoByteStringSet(int index, uint16_t value);

  // Get the address of the characters in this string.
  inline Address GetCharsAddress();

  inline uc16* GetChars();

  // Clear uninitialized padding space. This ensures that the snapshot content
  // is deterministic.
  void clear_padding();

  // For regexp code.
  const uint16_t* SeqTwoByteStringGetData(unsigned start);

  DECL_CAST(SeqTwoByteString)

  // Garbage collection support.  This method is called by the
  // garbage collector to compute the actual size of a TwoByteString
  // instance.
  inline int SeqTwoByteStringSize(InstanceType instance_type);

  // Computes the size for a TwoByteString instance of a given length.
  static int SizeFor(int length) {
    return OBJECT_POINTER_ALIGN(kHeaderSize + length * kShortSize);
  }

  // Maximal memory usage for a single sequential two-byte string.
  static const int kMaxCharsSize = kMaxLength * 2;
  static const int kMaxSize = OBJECT_POINTER_ALIGN(kMaxCharsSize + kHeaderSize);
  STATIC_ASSERT(static_cast<int>((kMaxSize - kHeaderSize) / sizeof(uint16_t)) >=
                String::kMaxLength);

  class BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(SeqTwoByteString);
};

// The ConsString class describes string values built by using the
// addition operator on strings.  A ConsString is a pair where the
// first and second components are pointers to other string values.
// One or both components of a ConsString can be pointers to other
// ConsStrings, creating a binary tree of ConsStrings where the leaves
// are non-ConsString string values.  The string value represented by
// a ConsString can be obtained by concatenating the leaf string
// values in a left-to-right depth-first traversal of the tree.
class ConsString : public String {
 public:
  // First string of the cons cell.
  inline String* first();
  // Doesn't check that the result is a string, even in debug mode.  This is
  // useful during GC where the mark bits confuse the checks.
  inline Object* unchecked_first();
  inline void set_first(Isolate* isolate, String* first,
                        WriteBarrierMode mode = UPDATE_WRITE_BARRIER);

  // Second string of the cons cell.
  inline String* second();
  // Doesn't check that the result is a string, even in debug mode.  This is
  // useful during GC where the mark bits confuse the checks.
  inline Object* unchecked_second();
  inline void set_second(Isolate* isolate, String* second,
                         WriteBarrierMode mode = UPDATE_WRITE_BARRIER);

  // Dispatched behavior.
  V8_EXPORT_PRIVATE uint16_t ConsStringGet(int index);

  DECL_CAST(ConsString)

  // Layout description.
  static const int kFirstOffset = POINTER_SIZE_ALIGN(String::kSize);
  static const int kSecondOffset = kFirstOffset + kPointerSize;
  static const int kSize = kSecondOffset + kPointerSize;

  // Minimum length for a cons string.
  static const int kMinLength = 13;

  typedef FixedBodyDescriptor<kFirstOffset, kSecondOffset + kPointerSize, kSize>
      BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

  DECL_VERIFIER(ConsString)

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(ConsString);
};

// The ThinString class describes string objects that are just references
// to another string object. They are used for in-place internalization when
// the original string cannot actually be internalized in-place: in these
// cases, the original string is converted to a ThinString pointing at its
// internalized version (which is allocated as a new object).
// In terms of memory layout and most algorithms operating on strings,
// ThinStrings can be thought of as "one-part cons strings".
class ThinString : public String {
 public:
  // Actual string that this ThinString refers to.
  inline String* actual() const;
  inline HeapObject* unchecked_actual() const;
  inline void set_actual(String* s,
                         WriteBarrierMode mode = UPDATE_WRITE_BARRIER);

  V8_EXPORT_PRIVATE uint16_t ThinStringGet(int index);

  DECL_CAST(ThinString)
  DECL_VERIFIER(ThinString)

  // Layout description.
  static const int kActualOffset = String::kSize;
  static const int kSize = kActualOffset + kPointerSize;

  typedef FixedBodyDescriptor<kActualOffset, kSize, kSize> BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

 private:
  DISALLOW_COPY_AND_ASSIGN(ThinString);
};

// The Sliced String class describes strings that are substrings of another
// sequential string.  The motivation is to save time and memory when creating
// a substring.  A Sliced String is described as a pointer to the parent,
// the offset from the start of the parent string and the length.  Using
// a Sliced String therefore requires unpacking of the parent string and
// adding the offset to the start address.  A substring of a Sliced String
// are not nested since the double indirection is simplified when creating
// such a substring.
// Currently missing features are:
//  - handling externalized parent strings
//  - external strings as parent
//  - truncating sliced string to enable otherwise unneeded parent to be GC'ed.
class SlicedString : public String {
 public:
  inline String* parent();
  inline void set_parent(Isolate* isolate, String* parent,
                         WriteBarrierMode mode = UPDATE_WRITE_BARRIER);
  inline int offset() const;
  inline void set_offset(int offset);

  // Dispatched behavior.
  V8_EXPORT_PRIVATE uint16_t SlicedStringGet(int index);

  DECL_CAST(SlicedString)

  // Layout description.
  static const int kParentOffset = POINTER_SIZE_ALIGN(String::kSize);
  static const int kOffsetOffset = kParentOffset + kPointerSize;
  static const int kSize = kOffsetOffset + kPointerSize;

  // Minimum length for a sliced string.
  static const int kMinLength = 13;

  typedef FixedBodyDescriptor<kParentOffset, kOffsetOffset + kPointerSize,
                              kSize>
      BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

  DECL_VERIFIER(SlicedString)

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(SlicedString);
};

// The ExternalString class describes string values that are backed by
// a string resource that lies outside the V8 heap.  ExternalStrings
// consist of the length field common to all strings, a pointer to the
// external resource.  It is important to ensure (externally) that the
// resource is not deallocated while the ExternalString is live in the
// V8 heap.
//
// The API expects that all ExternalStrings are created through the
// API.  Therefore, ExternalStrings should not be used internally.
class ExternalString : public String {
 public:
  DECL_CAST(ExternalString)

  // Layout description.
  static const int kResourceOffset = POINTER_SIZE_ALIGN(String::kSize);
  static const int kShortSize = kResourceOffset + kPointerSize;
  static const int kResourceDataOffset = kResourceOffset + kPointerSize;
  static const int kSize = kResourceDataOffset + kPointerSize;

  // Return whether external string is short (data pointer is not cached).
  inline bool is_short() const;
  // Size in bytes of the external payload.
  int ExternalPayloadSize() const;

  // Used in the serializer/deserializer.
  inline Address resource_as_address();
  inline void set_address_as_resource(Address address);
  inline uint32_t resource_as_uint32();
  inline void set_uint32_as_resource(uint32_t value);

  STATIC_ASSERT(kResourceOffset == Internals::kStringResourceOffset);

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalString);
};

// The ExternalOneByteString class is an external string backed by an
// one-byte string.
class ExternalOneByteString : public ExternalString {
 public:
  static const bool kHasOneByteEncoding = true;

  typedef v8::String::ExternalOneByteStringResource Resource;

  // The underlying resource.
  inline const Resource* resource();

  // It is assumed that the previous resource is null. If it is not null, then
  // it is the responsability of the caller the handle the previous resource.
  inline void SetResource(Isolate* isolate, const Resource* buffer);
  // Used only during serialization.
  inline void set_resource(const Resource* buffer);

  // Update the pointer cache to the external character array.
  // The cached pointer is always valid, as the external character array does =
  // not move during lifetime.  Deserialization is the only exception, after
  // which the pointer cache has to be refreshed.
  inline void update_data_cache();

  inline const uint8_t* GetChars();

  // Dispatched behavior.
  inline uint16_t ExternalOneByteStringGet(int index);

  DECL_CAST(ExternalOneByteString)

  class BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalOneByteString);
};

// The ExternalTwoByteString class is an external string backed by a UTF-16
// encoded string.
class ExternalTwoByteString : public ExternalString {
 public:
  static const bool kHasOneByteEncoding = false;

  typedef v8::String::ExternalStringResource Resource;

  // The underlying string resource.
  inline const Resource* resource();

  // It is assumed that the previous resource is null. If it is not null, then
  // it is the responsability of the caller the handle the previous resource.
  inline void SetResource(Isolate* isolate, const Resource* buffer);
  // Used only during serialization.
  inline void set_resource(const Resource* buffer);

  // Update the pointer cache to the external character array.
  // The cached pointer is always valid, as the external character array does =
  // not move during lifetime.  Deserialization is the only exception, after
  // which the pointer cache has to be refreshed.
  inline void update_data_cache();

  inline const uint16_t* GetChars();

  // Dispatched behavior.
  inline uint16_t ExternalTwoByteStringGet(int index);

  // For regexp code.
  inline const uint16_t* ExternalTwoByteStringGetData(unsigned start);

  DECL_CAST(ExternalTwoByteString)

  class BodyDescriptor;
  // No weak fields.
  typedef BodyDescriptor BodyDescriptorWeak;

 private:
  DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalTwoByteString);
};

// A flat string reader provides random access to the contents of a
// string independent of the character width of the string.  The handle
// must be valid as long as the reader is being used.
class FlatStringReader : public Relocatable {
 public:
  FlatStringReader(Isolate* isolate, Handle<String> str);
  FlatStringReader(Isolate* isolate, Vector<const char> input);
  void PostGarbageCollection();
  inline uc32 Get(int index);
  template <typename Char>
  inline Char Get(int index);
  int length() { return length_; }

 private:
  String** str_;
  bool is_one_byte_;
  int length_;
  const void* start_;
};

// This maintains an off-stack representation of the stack frames required
// to traverse a ConsString, allowing an entirely iterative and restartable
// traversal of the entire string
class ConsStringIterator {
 public:
  inline ConsStringIterator() {}
  inline explicit ConsStringIterator(ConsString* cons_string, int offset = 0) {
    Reset(cons_string, offset);
  }
  inline void Reset(ConsString* cons_string, int offset = 0) {
    depth_ = 0;
    // Next will always return nullptr.
    if (cons_string == nullptr) return;
    Initialize(cons_string, offset);
  }
  // Returns nullptr when complete.
  inline String* Next(int* offset_out) {
    *offset_out = 0;
    if (depth_ == 0) return nullptr;
    return Continue(offset_out);
  }

 private:
  static const int kStackSize = 32;
  // Use a mask instead of doing modulo operations for stack wrapping.
  static const int kDepthMask = kStackSize - 1;
  static_assert(base::bits::IsPowerOfTwo(kStackSize),
                "kStackSize must be power of two");
  static inline int OffsetForDepth(int depth);

  inline void PushLeft(ConsString* string);
  inline void PushRight(ConsString* string);
  inline void AdjustMaximumDepth();
  inline void Pop();
  inline bool StackBlown() { return maximum_depth_ - depth_ == kStackSize; }
  void Initialize(ConsString* cons_string, int offset);
  String* Continue(int* offset_out);
  String* NextLeaf(bool* blew_stack);
  String* Search(int* offset_out);

  // Stack must always contain only frames for which right traversal
  // has not yet been performed.
  ConsString* frames_[kStackSize];
  ConsString* root_;
  int depth_;
  int maximum_depth_;
  int consumed_;
  DISALLOW_COPY_AND_ASSIGN(ConsStringIterator);
};

class StringCharacterStream {
 public:
  inline explicit StringCharacterStream(String* string, int offset = 0);
  inline uint16_t GetNext();
  inline bool HasMore();
  inline void Reset(String* string, int offset = 0);
  inline void VisitOneByteString(const uint8_t* chars, int length);
  inline void VisitTwoByteString(const uint16_t* chars, int length);

 private:
  ConsStringIterator iter_;
  bool is_one_byte_;
  union {
    const uint8_t* buffer8_;
    const uint16_t* buffer16_;
  };
  const uint8_t* end_;
  DISALLOW_COPY_AND_ASSIGN(StringCharacterStream);
};

}  // namespace internal
}  // namespace v8

#include "src/objects/object-macros-undef.h"

#endif  // V8_OBJECTS_STRING_H_