Retrieve GitHub repository archive and extract it to the expected directory.

checkout.py

# /// script
# requires-python = ">=3"
# dependencies = []
# ///

from __future__ import annotations

import shutil
import tarfile
import urllib.request
from pathlib import Path
from tempfile import TemporaryDirectory
from typing import TYPE_CHECKING

if TYPE_CHECKING:
    from http.client import HTTPResponse

TARGET_TAG = "v3.11.12"


def main() -> None:
    """Retrieve GitHub repository archive and extract it to the expected directory."""
    owner = "aio-libs"
    repo = "aiohttp"
    url = f"https://github.com/{owner}/{repo}/archive/refs/tags/{TARGET_TAG}.tar.gz"

    downloads_dir = Path("downloads")
    expected_dir = downloads_dir / repo

    response: HTTPResponse
    with TemporaryDirectory() as temp_dir:
        with (
            urllib.request.urlopen(url) as response,
            tarfile.open(fileobj=response, mode="r:gz") as tar,
        ):
            tar.extractall(temp_dir, filter="data")

        remote_filename = response.headers.get_filename()
        if remote_filename is None:
            raise FileNotFoundError

        predicted_dirname = remote_filename.removesuffix(".tar.gz")
        predicted_dir = Path(temp_dir, predicted_dirname)
        if not predicted_dir.is_dir():
            raise FileNotFoundError

        if expected_dir.exists():
            shutil.rmtree(expected_dir)

        expected_dir.parent.mkdir(parents=True, exist_ok=True)
        predicted_dir.rename(expected_dir)

    print(f"Downloaded {repo} {TARGET_TAG} to {expected_dir}")


if __name__ == "__main__":
    main()