Is this crate umaintained?

[MarijnS95]

Even though a release has been published just two days ago, lots of perfectly good PRs appear to be ignored, making it pretty much impossible to use this crate for anything more than the absolute basics. Such PRs include both soundness fixes and exposure of missing API surface.

Is there anything that can be done to pick up the pace and slack on the pull request backlog? This is important for us as we're working towards shipping crates and applications based on metal-rs. We'd hate to have to desert to publishing a fork with all the missing fixes and features because of this, but that looks to become our only option.

Note also that this lack of activity is actively discouraging further contributions. We have a lot more patches we'd like to upstream, but a big dependency stack in PRs is a nontrivial burden to handle - something that doesn't have to happen when PRs are actively reviewed and merged.

Thanks!

[grovesNL]

It's not unmaintained - just lagging a bit behind at the moment unfortunately. I could try to prioritize the open PRs soon if it's causing problems for anyone.

I've also been following the progress on objc2 bindings across the ecosystem a bit, in case we could switch over sometime (#241) and avoid most of the manual bindings.

[MarijnS95]

@grovesNL thanks for the update!

objc2 has been in consideration, but we have enough metal-rs code that migrating over is no small feat - even though I recognize it's likely the way forward. As said in the wgpu Matrix channel that I just found, it feels a bit harsh to request/force maintenance when a clear - autogenerated and maintained - alternative already exists.

It'd be nice to go over open PRs (and we'll open more for our backlog), and on the side I'll keep chipping away at seeing what objc2 can do for us.