I just can't understand why make patchcheck is complaining about whitespace in inspect.py 😕

I just can't understand why make patchcheck is complaining about whitespace in inspect.py 😕

There is some pre-existing trailing whitespace in the file (around lines 1648 and 1686). make patchcheck runs only on changed files and we don't have a check to make sure it passes on unchanged files (sorry for that). You should probably just fix the trailing whitespace.

Oh, easy to fix now that I know what's happening! Thanks for the quick response 🤩

(Perhaps the patchcheck job could also run a git diff so that contributors can see the problem?)

This seems like it should be uncontroversial enough (famous last words — I'm sure someone is going to show up complaining that they use invalid signature objects as some load-bearing part of their workflow).

I'm sure someone is going to show up complaining that they use invalid signature objects as some load-bearing part of their workflow

Sorry for making you wait so long.

So, for many years we were dynamically generating functions based on interface described in YAML files. With Python 3.9 our tests started to fail. Turns out we had some cases where a parameter with name in was being generated. I wonder whether the actual generated low-level functions worked in this case.