If the goal of gram is to focus on stability it seems reasonable to work to reduce and eventually eliminate the monorepo coupling between gram's GPUI and gram itself. This would provide a great starting point for a community-focused fork of GPUI, following gram's principles but intended for general use.
Port Gram to WGPUI (GPUI / WGPU rewrite) or move to separate project? #76
Another thought i just had is that if we "beat" Zed to improving gpui for general use both gram and gram gpui stand to benefit from projects choosing to be downstream of us rather than zed
Yes! I really like this idea. Split it out, switch to WGPU and stabilize.
We’ll need to rename it. But yeah, sounds great to me.
I would also like to move the UI component library out so other tools can use it more easily.
There are already forks of GPUI; maybe use them or reach out to them?
https://github.com/Far-Beyond-Pulsar/WGPUI
And maybe some others ^^'
There's also an official one by the Zed team, but it's kinda dead
https://github.com/gpui-ce/gpui-ce
And, this, but might be ... a bit new?
https://github.com/ahkohd/ggpui
That one is probably the most stable https://github.com/Far-Beyond-Pulsar/WGPUI
If we could move to WGPUI that would be great, would solve the WGPU transition as well.
Worth noting that a PR to switch to wgpu without switching to an external fork of gpui has been merged.
@ItsHarper wrote in #76 (comment):
Worth noting that a PR to switch to wgpu without switching to an external fork of gpui has been merged.
I am still thinking that the best solution long term would be to switch to an external fork if possible, and I would want to contribute any fixes or discoveries in the WGPU move there if/when doing the switch.
No due date set.
No dependencies set.
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?