Improve type annotations (add more and fix wrong ones)

[oprypin]

The origins of these are three-fold:

• Merging in stubs from https://github.com/python/typeshed/tree/main/stubs/Markdown using "merge-pyi"
  • Note: we can consider these annotations to be the important ones because it's what people have been adding according to their own need
• Double-checking around places where stubs were already added from the above, particularly conflicts with annotations that got added in this repository already
  • Taking the opportunity to declare a generic "Registry of T" class
• Running "mypy" and eliminating the most glaring errors it reported

[waylan]

So there appears to be a couple code changes included in this (I don't mean adjustments to whitespace/line length). Could you please remove those? They should be submitted separately.

[oprypin]

I only recognized one change that's clearly not about type annotations and reverted it.

I'm not sure if you also mean the NamedTuple change (which I think is on topic, as it's just the way to spell the same thing but with type annotations) or if there was something else.

[waylan]

I was thinking of the NameTuples, but you are correct, they do have type annotations. So nevermind on that.

[waylan]

Thanks for this. Some of the wrong annotations I completely missed.

I'll merge this after a changelog entry is added.

[oprypin]

That is done

[oprypin]

For future reference:

1. The main reason that nobody complained about the current type annotations is probably that they aren't actually used by type checkers. Unless a file named py.typed exists, the dominant type checker "mypy" ignores the types and asks you to install https://pypi.org/project/types-Markdown/ anyway.

2. If you are interested in a fully correct type-checked codebase (according to "mypy") as well as official external-facing type annotations support (by adding py.typed), let me know anytime, I can take up that task and help maintain this going forward. Otherwise that's all for now.

[waylan]

Note that you have a conflict with the changelog (presumably with #1392).

Regarding a fully correct type-checked codebase (according to "mypy") that is probably a reasonable goal. Personally, I haven't really explored what's involved given that this is a rather new thing for Python. So contributions are welcome.

[oprypin]

Note that you have a conflict with the changelog

Hm? I had already merged to resolve the conflict. There was just the spelling check failing

Regarding a fully correct type-checked codebase (according to "mypy") that is probably a reasonable goal.

Great, so I'll continue looking into this soon (in a separate PR)

[pawamoy]

This PR almost closes #1389 🙂

[oprypin]

Oh I didn't see that!
No there are still like 100 mypy errors so maybe not "almost" 😅

[pawamoy]

Oh OK I thought there was just the py.typed file missing 😅