09:08 <wlb> wayland-protocols/main: dcz * Start an input-method protocol https://gitlab.freedesktop.org/wayland/wayland-protocols/commit/4da04536e882 experimental/xx-input-method/ README input-method-unstable-v2.xml

09:08 <wlb> wayland-protocols/main: dcz * experimental/input-method: Rename to xx_input_method_v2 https://gitlab.freedesktop.org/wayland/wayland-protocols/commit/8d1ccbdebe63 experimental/xx-input-method/xx-input-method-v2.xml

09:08 <wlb> wayland-protocols/main: dcz * experimental/input-method: Implement only text-input counterparts https://gitlab.freedesktop.org/wayland/wayland-protocols/commit/73b611579946 experimental/xx-input-method/xx-input-method-v2.xml

09:09 <wlb> wayland-protocols Merge request !397 merged \o/ (Start an input-method protocol https://gitlab.freedesktop.org/wayland/wayland-protocols/-/merge_requests/397)

09:51 <tokyo4j> dcz_: is it intentional that the protocol is named input_method_experimental_v2 and manager interface is named xx_input_method_manager_v2, but IM interface is named xx_input_method_v1?

09:51 <wlb> weston Merge request !1761 opened by Rémi Bernon (rbernon) xwayland: Activate the X root window when a Wayland window is activated https://gitlab.freedesktop.org/wayland/weston/-/merge_requests/1761

09:53 <pq> kekekeks, that's not the only wart about new_id in events. They also race against the (parent) object being destroyed.

12:42 <wlb> wayland-protocols Merge request !407 opened by DorotaC (dcz) input-method: Define a popup https://gitlab.freedesktop.org/wayland/wayland-protocols/-/merge_requests/407

15:30 <emersion> hwentlan_, pq: thoughts on intel's request for 32-but LUTs?

15:30 <emersion> bit*

16:29 <kekekeks> Quick question: it's possible to "atomically" set a dispath queue when creating a proxy with new_id by using a wrapper for the factory object. Is it possible to atomically set _dispatcher_?

16:37 <kekekeks> Right now there are two sort-of soft-contradicting requirements: 1) dispatcher has to be set while the new-object's owner thread/queue is not dispatching events 2) proxy wrapper has to be used while factory object's queue is not attempting to destroy wl_proxy

16:46 <kekekeks> I guess it's currently only valid to call object-returning methods from the thread that "owns" the queue designated for returned object

16:47 <kekekeks> Is there something that would prevent adding an API for atomically setting a listener?

17:13 <dwt> I really need to have a thorough look through the colorops patches when I'm more awake. We have some weird stuff going on around LUTs. But I think it might be weird enough to just require a new op type anyway.

19:03 <wlb> wayland/main: Simon Ser * build: bump version to 1.23.90 for the RC1 release https://gitlab.freedesktop.org/wayland/wayland/commit/62cd0990e843 meson.build

19:03 <wlb> wayland New tag: 1.23.91 https://gitlab.freedesktop.org/wayland/wayland/tags/1.23.91

19:18 <wlb> wayland.freedesktop.org/main: Simon Ser * releases: add Wayland 1.23.91 https://gitlab.freedesktop.org/wayland/wayland.freedesktop.org/commit/d06dad7ef912 releases.html