Stanislav Terliakov

Okay, we've even had one such false positive in primer! https://github.com/enthought/comtypes/blob/430bbdca8c0a146139acb990146ecbe06bdbf803/comtypes/_post_coinit/misc.py#L143

It does, thanks! I expected this to be too niche to run a mypy-issues scan, but that was a wrong estimate:(

> I get different results when running tests versus checking a file manually: Likely typeshed definition of `tuple.__{r,}add__` is more complex than what's in our fixture (`[buitlins fixtures/tuple.pyi]`). That fixture...

* `xarray` should either always use `__hash__ = None` or `__hash__: ClassVar[None]`. Using `__hash__: None` everywhere should also work, I suppose, but doesn't have proper semantics. All errors are correct,...

@A5rocks nope, this is way more tricky: this only works because `gmtime` is a "special" C function. If you do the same with ```python import time def my_gmtime(t: float |...

> I think that error is fine. It's not clear whether we should error or not I think the message could be worded a bit differently to clarify that this...

Overload specification does not require any function body checks against overloaded signatures. Overloads are explicitly external to the function itself: foreign callers see them and don't see the implementation signature,...

And why on earth is this PR listed as fixing #7724? That one is much broader and doesn't seem to concern `__class__` assignments at all?

I'd suggest a simple heuristic: if the provided path ends with `.xml`, treat it as a filename. IMO it's more than reasonable to allow specifying a file path instead of...

I'm frankly not sure what part of the core team to ping here - reviewers of the original PR, @JukkaL? @randolf-scholz please also take a look if you have time,...