A dynamic tiling Wayland compositor
Breaking changes
- window/layer rule regexes now require a full match (not any match) to trigger.
- cursor:dumb_copy is gone in favor of cursor:use_cpu_buffer. This should allow no-downsides Nvidia hardware cursors. Please note it’s experimental.
Notes for packagers
- New deps: hyprgraphics, re2
New features:
- binds: add option to allow fullscreening a pinned window (#8526)
- config: add ‘force’ option for ‘cursor:warp_on_change_workspace’ (#8681)
- core: Add support for hyprqtutils’ update screen (#8651)
- core: add a few festive splashes
- core: move colorspace handling to oklab (#8635)
- dispatchers: Add an option to prioritize focus change within groups with movefocus (#8601)
- hooks: add pre connected/disconnected monitor events (#8503)
- hyprctl: add an inhibitingIdle field to windows
- hyprctl: add directScanout to hyprctl monitors
- hyprctl: add json output on hyprctl -j plugins list (#8480)
- input: add warp_back_after_non_mouse_input
- logging: Add some context to config error logs (#8326)
- makefile: add stub to discourage direct make
- pointer: add drm dumb buffers for cursors (#8399)
- renderer: add lockdead_screen_delay (#8467)
- renderer: add option to blur IME popups (#8521)
- version: add link versions for other utils (#8619)
- windowrules: add rules for mouse and touchpad scroll factors (#8655)
- xwayland: Support cross DnD from Wayland (#8708)
I wonder if i3 is more, just because it’s been around forever, but I could certainly see Hyprland as having the most growth among standalone options.
I would think i3 is slowly dying because sway is it’s Wayland replacement. But wrong, i3 has a stable fanbase.
Sway, from what I understand, hasn’t really made a ton of improvements that would entice the X11 holdouts to switch. Not that it’s bad, but if it’s virtually the same as X11—and that’s what someone knows best—why would they bother switching (I would, but I also try to use Wayland whenever possible).