utils/embed_header.py

#!/usr/bin/env python3
# Copyright (C) Microsoft Corporation. All rights reserved.
# This file is distributed under the University of Illinois Open Source License.
# See LICENSE.TXT for details.
"""Embed a file's contents into a generated C++ source as a StringRef.

Reads the input file as bytes and produces a C++ snippet of the form:

    llvm::StringRef Data =
      "<escaped line 1>"
      "<escaped line 2>"
      ...;

The generated file declares a single variable named ``Data`` at file
scope. The intent is for the file to be ``#include``-ed inside a wrapping
namespace by the consumer.
"""

import argparse
import os
import sys


def escape_byte(b):
    # Backslash and double-quote must always be escaped.
    if b == 0x5C:
        return "\\\\"
    if b == 0x22:
        return "\\\""
    if b == 0x09:
        return "\\t"
    if b == 0x0A:
        return "\\n"
    if b == 0x0D:
        return "\\r"
    # Printable ASCII passes through unchanged.
    if 0x20 <= b < 0x7F:
        return chr(b)
    # Use 3-digit octal escapes for everything else. Octal escapes
    # consume at most 3 digits, so they cannot accidentally absorb
    # subsequent characters in the literal.
    return "\\{:03o}".format(b)


def escape_line(line_bytes):
    return "".join(escape_byte(b) for b in line_bytes)


def generate(input_path, output_path):
    with open(input_path, "rb") as f:
        data = f.read()

    # Split on '\n' but preserve the newline at the end of each piece.
    pieces = []
    start = 0
    for i, b in enumerate(data):
        if b == 0x0A:
            pieces.append(data[start:i + 1])
            start = i + 1
    if start < len(data):
        pieces.append(data[start:])
    if not pieces:
        pieces = [b""]

    lines = []
    lines.append("// Generated by utils/embed_header.py. Do not edit.")
    lines.append("// Source: {}".format(os.path.basename(input_path)))
    lines.append("llvm::StringRef Data =")
    for piece in pieces:
        lines.append("  \"{}\"".format(escape_line(piece)))
    # Terminate the variable declaration.
    lines[-1] = lines[-1] + ";"
    lines.append("")

    output_text = "\n".join(lines)

    # Avoid touching the file if the contents are unchanged so that
    # downstream build steps are not invalidated unnecessarily.
    if os.path.exists(output_path):
        try:
            with open(output_path, "r", encoding="utf-8") as f:
                if f.read() == output_text:
                    return
        except OSError:
            pass

    out_dir = os.path.dirname(output_path)
    if out_dir and not os.path.isdir(out_dir):
        os.makedirs(out_dir, exist_ok=True)

    with open(output_path, "w", encoding="utf-8") as f:
        f.write(output_text)


def main(argv):
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("input", help="Path to input file to embed.")
    parser.add_argument("output", help="Path to generated C++ snippet.")
    args = parser.parse_args(argv)
    generate(args.input, args.output)
    return 0


if __name__ == "__main__":
    sys.exit(main(sys.argv[1:]))