DataOps-platform/deployment/compose/reproducible_archive.py

#!/usr/bin/env python3
"""Create a byte-reproducible tar.gz from one directory tree."""

from __future__ import annotations

import argparse
import gzip
import os
import tarfile
from pathlib import Path


def create_archive(source: Path, output: Path, source_date_epoch: int) -> None:
    source = source.resolve()
    output.parent.mkdir(parents=True, exist_ok=True)
    paths = [source, *sorted(source.rglob("*"), key=lambda path: path.relative_to(source).as_posix())]
    with output.open("wb") as raw:
        with gzip.GzipFile(filename="", mode="wb", fileobj=raw, mtime=source_date_epoch) as compressed:
            with tarfile.open(fileobj=compressed, mode="w", format=tarfile.PAX_FORMAT) as archive:
                for path in paths:
                    arcname = source.name if path == source else f"{source.name}/{path.relative_to(source).as_posix()}"
                    info = archive.gettarinfo(str(path), arcname=arcname)
                    info.uid = 0
                    info.gid = 0
                    info.uname = "root"
                    info.gname = "root"
                    info.mtime = source_date_epoch
                    info.pax_headers = {}
                    if info.isdir():
                        info.mode = 0o755
                        archive.addfile(info)
                    elif info.isfile():
                        executable = bool(os.stat(path).st_mode & 0o111)
                        info.mode = 0o755 if executable else 0o644
                        with path.open("rb") as handle:
                            archive.addfile(info, handle)
                    elif info.issym():
                        archive.addfile(info)
                    else:
                        raise ValueError(f"unsupported archive entry: {path}")


def main() -> int:
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--source", type=Path, required=True)
    parser.add_argument("--output", type=Path, required=True)
    parser.add_argument("--source-date-epoch", type=int, required=True)
    args = parser.parse_args()
    create_archive(args.source, args.output, args.source_date_epoch)
    return 0


if __name__ == "__main__":
    raise SystemExit(main())