Chore: Fully deprecate objc2::rc::{Id, WeakId}

[PaulDance]

As part of #617.

With this, 0.6 should be in a good position to be delivered, or are there still a lot of things to do? What I see in https://github.com/madsmtm/objc2/milestone/6 seems that it could be optional while the important changes that are already completed would be nice if published and back-porting things I would need in a fork would be cumbersome.

[madsmtm]

As far as I can tell, this should be the last remaining part required in order to close #617.

I think I'd also like to figure out #[method_id(...)], I've added mention of #457 to the issue description.

With this, 0.6 should be in a good position to be delivered, or are there still a lot of things to do? What I see in https://github.com/madsmtm/objc2/milestone/6 seems that it could be optional while the important changes that are already completed would be nice if published and back-porting things I would need in a fork would be cumbersome.

Ugh, I'm so bad at deciding when to release, in my defence things are complicated by us having a fairly large amount of dependents, and thus infrequent breaking versions are desirable for the ecosystem in general. But you are right, we are close, the target date I've set for the milestone (end of the year) should be accurate now, the remaining stuff that I really want to get in is:

• Proper enum prefix stripping, having to write NSWindowSharingType::NSWindowSharingNone is really ugly when NSWindowSharingType::None could do.
• The cargo features stuff (removing "all" in favour of enabling things by default + std features shouldn't enable sub-features), since it's a breaking change.
• The autogenerated thunks for functions, since it's also a breaking change, and it's needed for nicer Metal integration (allows objc2-metal to really shine over metal-rs).

(So if you want to explore some of those, then feel free to ;) )

[madsmtm]

Thanks!

[PaulDance]

I think I'd also like to figure out #[method_id(...)], I've added mention of #457 to the issue description.

OK, changed the description accordingly.

I'm so bad at deciding when to release

Most definitely understandable. Another way to see things though is that breaking a lot at the same time makes transitioning harder, while more frequent but more focused breakage is somewhat more manageable. Tough choices overall.

the target date I've set for the milestone (end of the year) should be accurate now

That is more than acceptable, especially considering vacations and stuff. It should also align nicely with a development I will take on again.

So if you want to explore some of those, then feel free to ;)

I just might, but maybe only the first two 😉

Thanks!

You're welcome!