Created
August 14, 2024 16:40
-
-
Save s3rgeym/3ef98366c8a655953cf8b2343df1be6f to your computer and use it in GitHub Desktop.
git index parser
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python | |
# https://git-scm.com/docs/index-format | |
# https://github.com/sbp/gin/blob/master/gin | |
from __future__ import annotations | |
import codecs | |
import struct | |
from dataclasses import dataclass | |
from typing import BinaryIO, ClassVar, Iterable, Type | |
@dataclass | |
class GiEntry: | |
ctime_s: int | |
ctime_ns: int | |
mtime_s: int | |
mtime_ns: int | |
dev: int | |
ino: int | |
mode: int | |
uid: int | |
gid: int | |
file_size: int | |
sha1: str | |
assume_valid: bool | |
extended: bool | |
stage: int | |
extended_flags: int | |
filename: str | |
ENTRY_HEADER_STRUCT: ClassVar[struct.Struct] = struct.Struct("!10I") | |
@classmethod | |
def parse(cls: Type[GiEntry], fp: BinaryIO, version: int) -> GiEntry: | |
# Читаем заголовок записи | |
( | |
ctime_s, | |
ctime_ns, | |
mtime_s, | |
mtime_ns, | |
dev, | |
ino, | |
mode, | |
uid, | |
gid, | |
file_size, | |
) = cls.ENTRY_HEADER_STRUCT.unpack( | |
fp.read(cls.ENTRY_HEADER_STRUCT.size) | |
) | |
# Читаем SHA1 и флаги | |
sha1 = codecs.encode(fp.read(20), "hex").decode() | |
flags = struct.unpack("!H", fp.read(2))[0] | |
assume_valid = bool(flags & 0x8000) | |
extended = bool(flags & 0x4000) | |
stage = (flags & 0x3000) >> 12 | |
name_len = flags & 0xFFF | |
extended_flags = 0 | |
if extended and version > 2: | |
extended_flags = struct.unpack("!B", fp.read(1))[0] | |
entry_len = ( | |
cls.ENTRY_HEADER_STRUCT.size + 20 + 2 + (1 if extended else 0) | |
) | |
if name_len < 0xFFF: | |
filename = fp.read(name_len) | |
entry_len += name_len | |
else: | |
# Читаем null-терминированное имя файла | |
filename = b"".join(iter(lambda: fp.read(1), b"\x00")) | |
entry_len += len(filename) + 1 | |
filename = filename.decode(errors="replace") | |
padding = (8 - (entry_len % 8)) or 8 | |
fp.seek(padding, 1) | |
return cls( | |
ctime_s=ctime_s, | |
ctime_ns=ctime_ns, | |
mtime_s=mtime_s, | |
mtime_ns=mtime_ns, | |
dev=dev, | |
ino=ino, | |
mode=mode, | |
uid=uid, | |
gid=gid, | |
file_size=file_size, | |
sha1=sha1, | |
assume_valid=assume_valid, | |
extended=extended, | |
stage=stage, | |
extended_flags=extended_flags, | |
filename=filename, | |
) | |
@dataclass | |
class GitIndex: | |
version: int | |
entries: list[GiEntry] | |
HEADER_STRUCT: ClassVar[struct.Struct] = struct.Struct("!4s2I") | |
@classmethod | |
def parse(cls, fp: BinaryIO) -> GitIndex: | |
# Читаем заголовок файла индекса | |
signature, version, num_entries = cls.HEADER_STRUCT.unpack( | |
fp.read(cls.HEADER_STRUCT.size) | |
) | |
assert signature == b"DIRC", "Invalid index signature" | |
assert version in (2, 3, 4), "Unsupported index version" | |
assert num_entries >= 0, "Number of entries must be non-negative" | |
# Читаем записи | |
entries = [GiEntry.parse(fp, version) for _ in range(num_entries)] | |
return cls(version=version, entries=entries) | |
def __iter__(self) -> Iterable[GiEntry]: | |
return iter(self.entries) | |
if __name__ == "__main__": | |
import argparse | |
p = argparse.ArgumentParser() | |
p.add_argument("input", type=argparse.FileType("rb")) | |
args = p.parse_args() | |
for entry in GitIndex.parse(args.input): | |
print(entry) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment