Type Signatures for extension modules (PEP draft)

[davidhewitt]

@skirpichev

I would appreciate if you could point to a related discussion.

I can find the following:

Introspection capabilities are incomplete for C objects

… but I’m not sure right now if that got any further.

[skirpichev (Sergey B Kirpichev)]

@steve.dower

What they’ve opted for is to look for a .pyi file, and to be honest, I agree with them. Writing a .pyi file is much easier than trying to expose the information directly from the method. And so perhaps another alternative is teaching inspect to look for a .pyi file for native modules and read from that instead?

Does this idea require a PEP or I could just open a feature request issue?

I would worry about backward compatibility. While this shouldn’t break anything in the CPython itself, it introduces an incompatible change for external projects, which already use *.pyi files right now just for typing support. E.g. help() uses inspect’s signature, if available, to show signatures of callables, but we have also PEP 7, that instructs people to put “signature line” in docstring. Not sure how severe it is: we always can detect and exclude PEP 7 “signature” and prefer one from the inspect module, but…

@encukou

Then we’d also want teach Argument Clinic to write .pyi files. Or use them as input.

This is a separate issue. And doesn’t help for external projects, unless AC will become public.

[steve.dower]

@skirpichev

Does this idea require a PEP or I could just open a feature request issue?

I think it’s fine to open the feature request, and if anything comes up that requires a big decision it can become a PEP at that point. (e.g. whether to use a .pyi instead of a .py, where to pull docstrings from, etc. - these might have good answers already, I’m not sure).