https://bugs.kde.org/show_bug.cgi?id=328248
--- Comment #11 from Simon Raffeiner <sturmf...@lieberbiber.de> --- darktable is one of the few applications which actually uses both colord and the X11 atoms. The darktable-cmstest command line utility is often used as a convenient shortcut to check which values are stored for which display in both colord and the X11 atoms, and because darktable is also one of the even fewer applications which try to handle multiple monitors correctly, it also serves as an example for how the enumeration of X displays and storing ICC profiles for multiple displays in X11 atoms might work in a real-world implementation. That doesn't mean I agree with the way darktable enumerates displays, but I have multiple multi-monitor setups and the way it handles displays seems to work in all cases. The patch for colord-kde committed in this bug report seems to line up with what darktable is expecting, so now there seem to be two real-world implementations of the specification which agree with each other. I reported the same bug against GNOME Color Manager in March, https://bugzilla.gnome.org/show_bug.cgi?id=794486. Quoting the maintainer: "I don't think it's possible to solve this bug. In my opinion, the ICC-profiles-in-X specification is just impossible to implement fully on a modern desktop stack." I think Wayland should be left out of the discussion since this bug report only concerns the X atoms, X will still be with us for many years even inside Wayland sessions, and I have the feeling that if applications even support system-wide color management (GIMP and RawTherapee don't) they seem to rely on the X11 atoms more than on colord. Genview and digiKam fall in the latter category, for example, as does Chrome. -- You are receiving this mail because: You are watching all bug changes.
