bpo-46101: argparse bug fix when using parents & subcommands

[Lucas-C]

Here is some minimal code reproducing the issue:

import argparse

common_opts_parser = argparse.ArgumentParser(add_help=False)
common_opts_parser.add_argument("--endpoint", choices=("prod", "dev"), default="dev")

parser = argparse.ArgumentParser(parents=[common_opts_parser])
subparsers = parser.add_subparsers(required=True)

subcmd_cmd = subparsers.add_parser("subcmd", parents=[common_opts_parser])
subcmd_cmd.add_argument("--debug", action="store_true")

print(parser.parse_args())

Everything works fine / as expected when specifying the common optional arg last:

$ ./bug_repro.py subcmd --endpoint=dev
Namespace(endpoint='dev', debug=False)

However when specifying the --endpoint between the program and the subcommand, the value provided is ignored:

$ ./bug_repro.py --endpoint=dev subcmd
Namespace(endpoint=None, debug=False)

Tested under Windows / Python 3.11.0a0 with:

python -m test -v -m test_common_opts test_argparse

https://bugs.python.org/issue46101

[the-knights-who-say-ni]

Hello, and thanks for your contribution!

I'm a bot set up to make sure that the project can legally accept this contribution by verifying everyone involved has signed the PSF contributor agreement (CLA).

Recognized GitHub username

We couldn't find a bugs.python.org (b.p.o) account corresponding to the following GitHub usernames:

@creich

This might be simply due to a missing "GitHub Name" entry in one's b.p.o account settings. This is necessary for legal reasons before we can look at this contribution. Please follow the steps outlined in the CPython devguide to rectify this issue.

You can check yourself to see if the CLA has been received.

Thanks again for the contribution, we look forward to reviewing it!

[Lucas-C]

Please not that the @the-knights-who-say-ni bot got my username wrong 😉

I have already signed the CLA agreement.

[Lucas-C]

This PR is ready for review:
the CI pipeline is passing, and I dot not plan to add commits.

[github-actions]

This PR is stale because it has been open for 30 days with no activity.

[Lucas-C]

Closing as @paul.j3 rejected the Python bug: https://bugs.python.org/issue46101