vi
vi
> I'm not entirely sure this change is worth it. Why should we make this change? Because of this: >> there technically hasn't been a non-2.x release I especially noticed...
@onerandomusername I've moved the legacy changelogs to a separate page, and linked that at the bottom of the main changelog page. The two migration pages are linked in the v0.10.0...
> However, moving the changelog would be nice, and it would be appreciated if we could implement that. Could you revert the versionchanged 1.x changes from this pr so we...
What makes setuptools unsuitable for our purposes? Or rather, how are the other build backends better, is it actually worth it to switch? I currently don't see an issue with...
Deduplicating code is definitely a great idea, but we might lose out on type-safety depending on how it's implemented (though the current code paths are already sprinkled with `typing.Any` so...
Hmm, found another bug: ```py from typing import NamedTuple import disnake from disnake.ext import commands class H(NamedTuple): abc: int @commands.register_injection def create_h(inter: disnake.CommandInter, param: str) -> H: return H(int(param)) @create_h.autocomplete("param")...
> Due to the cache bust that only affects select files and shouldn't be relevant to #722, I think. It might be worth consideration to actually do that for all...
> I don't think we should fully remove the warning. Instead of displaying the message content intent warning, we should instead use a warning to use InteractionBot and None as...
> This requires a lot more work, due to the http handling essentially being duplicated in webhook/async_.py From what I can tell, the purpose of this is to avoid exceptionally...
> @shiftinv could you rebase this pr on master? Done.