First of all, I'd like to say that I love where this is going :)
I have a few modifications that I (personally) would prefer:

• is there a reason why _NAMES_MAPPING has raw integers instead of pygame.X event constants? I tried doing that, and it worked perfectly fine. In cursors.py, the dictionary also has pygame constants
• I think your stub situation can be improved: IMO there should not be _event.pyi, rather, there should be only event.pyi containing both from _event.c and from event.py (there is a stub file for cursors.py, even tho it's made in python!). The stubs are fake code, we don't need to tell the user that event is actually split between C and py, what matters is what is accessible in it. Again I tried that myself, added init, quit and modified a signature and the stubcheck.py with mypy passes for me, that's why I'm suggesting it.

Tell me what you think about this

@damusss Thanks for the comment - I'll update _NAMES_MAPPING shortly. As for _event.pyi, I'm using it as a roadmap of what's left to do, but I'm not going to completely remove pygame._event, just change some of the functionality, so probably this file will be included.

@damusss Thanks for the comment - I'll update _NAMES_MAPPING shortly. As for _event.pyi, I'm using it as a roadmap of what's left to do, but I'm not going to completely remove pygame._event, just change some of the functionality, so probably this file will be included.

Alright, I'll check the stub situation when you are closer to finish then. Nice work so far :>

(Why the commits duplicated after I resolved merge conflict?)

No actually why are part of the commits not yours?? what even happened? It looks like you are adding the deprecator decorator

I merged commits from upstream using git pull origin main --rebase and likely this --rebase is the cause of inclusion of these commits... I'll try to discard this merge and re-attempt doing it.

Done, even though they still appear in the conversation, I removed them from the commits tab (I guess that conversation side just logs every change and not exactly those that truly will be merged?).

I think I'll split this PR in two parts:

1. Migrating stuff related to Event class. (nearly done, c-api documentation still needed to be updated)
2. Migrating the monstrosity of events filtering. (not touched yet)