Alex Waygood

Huh, seems like the stubtest issue has also been magically fixed by one of our mypy upgrades since I originally filed this PR in October? Or maybe the fact that...

Broader questions about how `@overload` works, and possible ways of improving `@overload`, are somewhat off-topic for the typeshed issue tracker @Daverball :) https://discuss.python.org/c/typing/32 would be a better place to discuss...

> One possible solution is to omit the `LiteralString` overloads for these three methods. This would eliminate the type violation error in the common case, but it would introduce new...

The package doesn't support Python 3.10, and we run stubtest on 3.10 in CI. I'd rather not start running stubtest on 3.9 in CI for all stubs. It would probably...

> The official package hasn't had an update or a communication from the author since 2021. However newer forked versions of it are available with wheels for 3.10. (To be...

If a superclass says that it will accept a `datetime.time` for a parameter in some circumstances, but a subclass says that it doesn't accept a `datetime.time` under any circumstances for...

(Adding DO-NOT-MERGE for now, as I agree we should resolve #9487 first before tackling this :)

> In the meantime, @AlexWaygood , can we consider allowing [lxml-stubs](https://pypi.org/project/lxml-stubs/) in the stubuploader? SGTM but I'm not a stub-uploader maintainer :)

> Also, is there a list somewhere of those "stubs taken out of typeshed because incomplete"? I feel it shouldn't be too hard to add 'em back with modern stubgen....

You'll have to ask one of the other maintainers -- sounds like that all happened before I arrived on the typeshed scene!