Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

stubtest: ignore more dunder pos-only errors #12294

Merged
merged 1 commit into from Mar 5, 2022

Conversation

hauntsaninja
Copy link
Collaborator

@hauntsaninja hauntsaninja commented Mar 5, 2022

This isn't actually a reversion. This logic was asymmetrical for reasons
lost to time. Although I suspect it was this way because the delta
only a few errors on typeshed.

What changed was that as a result of #12203 we actually started checking
a lot more dunder methods. Previously, we only checked dunders if either:
a) it was a special dunder, like __init__ or __call__, or
b) the dunder was defined on the actual stub class itself

In particular, we started checking every dunder redefined at runtime
that the stub just inherited from object.

A possible intermediate option would be to not check positional-only
arguments in the case where a stub inherits the definition. I'll
experiment with that once typeshed CI is green.

This isn't actually a reversion. This logic was asymmetrical for reasons
lost to time. Although I suspect it was this way because it introduced
only a few errors on typeshed.

What changed was that as a result of python#12203 we actually started checking
a lot more dunder methods. Previously, we only checked dunders if either:
a) it was a special dunder, like __init__ or __call, or
b) the dunder was defined on the actual stub class itself

In particular, we started checking every dunder redefined at runtime
that the stub just inherited from object.

A possible intermediate option would be to not check positional-only
arguments in the case where a stub inherits the definition. I'll
experiment with that once typeshed CI is green.
@github-actions
Copy link
Contributor

github-actions bot commented Mar 5, 2022

According to mypy_primer, this change has no effect on the checked open source code. 🤖🎉

"__hash__",
"__getattr__", # resulting behaviour might be typed explicitly
"__setattr__", # defining this on a class can cause worse type checking
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, this one seems to be rare enough that I think I'd prefer to consider each case separately and allowlist them in typeshed. I can't think of an example where including it in the stub would improve type checking, but you never know?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants