xdavidwu
commented
2 years ago This works for my daily workflow. There are some more planned features in #1, but I personally don't need them and have no time to work on those yet. I would say that this project is not done, but already in a shape good enough.
As for alternatives, there is wlpinyin, which uses RIME engine instead and should also be capable of inputing traditional chinese with even more input methods. But note that wlpinyin, instead of a candidate panel, they use preedit buffer to show candidates in a text-based inline style. I have heared that there are also efforts in fcitx5 on making it wayland-native with input-method-unstable-v2.
There is also some alternatives that are not wayland-native but kind of works, like fcitx4 via its specific GTK IM module. With fcitx4, candidate panel will still be presented with X11 (via XWayland), and last time I checked XWayland still have no proper HiDPI support. Note that this set up will also encounter issues with candidate panel position.
About how to use these wayland-native input methods (those using input-method-unstable-v2, including wlchewing, wlpinyin and others), in general, make sure that you have environment variable GTK_IM_MODULE set to wayland, and you should be good to go. But note that last time I checked, there is still no client support of related protocol (text-input-unstable-v3) in Qt, and thus Qt apps won't work with these input methods.
Would love to get an update on the status of this project because I wish for this project to succeed.
ie. Any progress/roadmap? Are there any alternatives for wayland/sway?
And also if its possible to add documentation on how to test it/play around with it. Screenshots would also be appreciated so users can know what it looks like.
Thanks!