Post B7QgpH9tozcqlSkGYK by dnkl@social.treehouse.systems
(DIR) More posts by dnkl@social.treehouse.systems
(DIR) Post #B7Qg00PaKKGa3WdP96 by dnkl@social.treehouse.systems
0 likes, 1 repeats
#foot just might try to play nice with compositors doing close-animations. It has been reported that this is glitchy on at least kwin (but I assume it's not limited to kwin).Presumably because foot unmaps all surfaces before destroying the toplevel and all other surfaces. This is being done to ensure a consistent state within foot, with no dangling pointers to the soon-to-be-destroyed terminal instance.If you've experienced visual glitches in the close-animations, do feel free to test https://codeberg.org/dnkl/foot/pulls/2388.Besides fixing the glitches, it should:Not crashNot result in any warnings or errors (on stderr) not seen before the MRNot result in any Wayland protocol errors (seen on stdout/stderr)
(DIR) Post #B7QgfD8ZuxnbOfKOrQ by dnkl@social.treehouse.systems
0 likes, 0 repeats
@vyivel I can only assume they trigger the animation on destroy, not unmap. Why? No idea. I agree with you though.This is why a said "might", and one reason the MR is work-in-progress.If there's a reason for kwin's behavior, I'd like to hear it. It would make me more likely to merge the MR.
(DIR) Post #B7QgpH9tozcqlSkGYK by dnkl@social.treehouse.systems
0 likes, 0 repeats
Feedback on whether close-animations already work correctly on your compositor is also very welcome.If it turns out this is kwin only, and that kwin should be able to animate this correctly with foot's current shutdown behavior, then the MR will obviously not be merged.
(DIR) Post #B7RDVTfisufEGlnW8e by YaLTeR@mastodon.online
0 likes, 0 repeats
@dnkl haven't seen any issues with foot on niri. Btw I've had to deal with a similar problem before: alacritty (winit) also used to unmap CSD subsurfaces before closing. So I had to store the snapshot of the toplevel whenever any subsurface unmaps. Mutter (GNOME Shell) also handles this fine.
(DIR) Post #B7RqEQQnZp8fVYg1BY by dnkl@social.treehouse.systems
0 likes, 0 repeats
@vyivel seems kwin had issues with this before': https://bugs.kde.org/show_bug.cgi?id=478297.Supposedly, it has been fixed. But that was years ago. Plenty of time for a regression...?