et.py

#!/usr/bin/env python3
import argparse
import glob
import os
import tarfile
from multiprocessing import cpu_count, Pool

from tqdm import tqdm


def extract_tgz(path):
    """
    Extract a single .tgz archive into a folder named after the archive
    (without its .tgz extension), using an explicit filter to avoid
    the Python-3.14 deprecation warning.
    """
    out_dir = os.path.splitext(path)[0]
    os.makedirs(out_dir, exist_ok=True)
    with tarfile.open(path, "r:gz") as tar:
        # the filter=lambda ti: ti simply returns each TarInfo unchanged
        tar.extractall(path=out_dir, filter=lambda ti: ti)
    return path


def main():
    p = argparse.ArgumentParser(
        description="Extract multiple .tgz files in parallel with a progress bar."
    )
    p.add_argument(
        "pattern",
        nargs="?",
        default="*.tgz",
        help="glob pattern for .tgz files (default: '*.tgz')",
    )
    p.add_argument(
        "-j",
        "--jobs",
        type=int,
        default=cpu_count(),
        help="number of parallel workers (default: number of CPU cores)",
    )
    args = p.parse_args()

    files = sorted(glob.glob(args.pattern))
    if not files:
        print(f"No files matched pattern {args.pattern!r}")
        return

    with Pool(processes=args.jobs) as pool:
        for _ in tqdm(
            pool.imap_unordered(extract_tgz, files),
            total=len(files),
            desc="Extracting",
            unit="archive",
        ):
            pass


if __name__ == "__main__":
    main()