Sebastian Rittau

My gut feeling says that we should disable the test even after the PR (as you also wrote), but maybe it's a good idea to do this after the next...

I've also removed the (old) second criterion. It was a left-over from the Python 2->3 transition (where we wanted to remove Python-2-only packages) and is in a way a special...

It would be great if we could somehow distinguish between tz-aware and naive `datetime`s using type checking and this is a clever idea. And yes, ideally this should be solved...

I'm unsure why the pyright regression tests fail. I can't see how the two ignored tests can match any of the overloads: ``` /home/runner/work/typeshed/typeshed/stdlib/@tests/test_cases/check_argparse.py:11:57 - error: Unnecessary "# type: ignore"...

Regarding the primer output: pylint is actually using `add_argument` in an unsafe manner. It defines "argument" classes that have choices, which are `list[str]` and a `type` that can be a...

@erictraut Could you have a look at https://github.com/python/typeshed/pull/12211/files? I'm not sure why pyright thinks the lines in question are okay. I don't think any potential should match, but maybe I'm...

That's interesting. mypy comes to a different conclusion (when using a slightly adapted example): ``` foo.py:10: error: Argument "type" to "add_argument" has incompatible type "type[int]"; expected "Callable[[str], str]" [arg-type] foo.py:11:...

It seems that this is currently inexpressible in a way that's compatible with all type checkers.

I'm very confused about the mypy error on win32. Can anyone shed a light on what I did wrong? https://github.com/python/typeshed/actions/runs/9547845403/job/26313973228?pr=12151

This is python/mypy#14309. I will disable "possibly-undefined" for now.