Wessel
Wessel
@gabrieldemarmiesse definitely! I've posted my message on the thread. Thanks :)
Hey @gabrieldemarmiesse! Do you have a specific test case that the linter should pick up, but doesn't? In the other PR, I've implemented your suggestion to add `get_overloads` inside the...
@gabrieldemarmiesse That makes sense! Let's do that.
Hmm, well, if we can make pyright happy but can’t satisfy mypy, perhaps that’s already worthwhile testing for. How about something like the following? ```python def test_methods(): # E: mypy(overloaded...
@PhilipVinc I agree that it's noisy, but it's also already uncovered some issues, so it does have significant utility. How about an opt-out mechanism, where the warning is supressed if...
E.g., this currently happens on `master`: ```python In [1]: from plum import dispatch ...: ...: ...: @dispatch.multi((int, float)) ...: def f(x: int, y: float, z: float = 1): ...: pass...
@ambv Thanks for responding to this issue. :) Yes, I'm still interested in the name `plum` on PyPI.
Amazing. Thank you!
Hey @nstarman! Thanks for creating a PR. :) I'm not entirely sure about this change. For a function, `function._f` corresponds to the method that was given whenever the function was...
@nstarman What would you think of changing the pattern to something like this? ```python @dispatch1.chain(dispatch2) def f(x: int, y: float): return x + y ``` Then, internally, the `chain` method...