Skip to content

hyprland 0.57.0-3.4: resend an on-screen keyboard's replaced keymap to the input method - #762

Open
ikegam wants to merge 1 commit into
derekross:mainfrom
ikegam:hyprland-ime-keymap
Open

ikegam wants to merge 1 commit into
derekross:mainfrom
ikegam:hyprland-ime-keymap

Conversation

@ikegam

@ikegam ikegam commented Nov 14, 2026 •

Copy link
Copy Markdown
Contributor

For Japanese input on the phone. This brings back #692; the shell side is #763.

The input method's keyboard grab sent a keymap only when the keyboard changed. squeekboard replaces its virtual keyboard's keymap on a layout change, so fcitx5 holding the grab kept the old one, and squeekboard's space reached it as "l". The patch compares the keymap too.

  • Tested on my 7 Pro with fcitx5 and Mozc: space converts now (before, "l").
  • tools/pr-check passes. Only the input-method grab path changes; with no input method holding a grab nothing is different.

🤖 Generated with Claude Code

…o the input method

The input method's keyboard grab sent a keymap only when the keyboard
changed. squeekboard replaces its virtual keyboard's keymap on a layout
change, so fcitx5 holding the grab kept the old one, and squeekboard's
space reached it as "l". Compare the keymap too.

On the 7 Pro with fcitx5 and Mozc: space converts now (before, "l").

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@ikegam
ikegam force-pushed the hyprland-ime-keymap branch from 8f22076 to cede25f Compare November 14, 2026 11:18
ikegam added a commit to ikegam/omarchy-mobile that referenced this pull request Nov 14, 2026
…le-ime)

`omamobile-ime on` hands Hyprland's one input method to fcitx5 with Mozc,
and `off` gives it back to squeekboard. squeekboard's keys reach fcitx5
through the keyboard grab, so romaji converts with Space and commits with
Enter. Mozc converts only while the keyboard's language (the globe key) is
"Japanese (romaji)", a copy of the US layout that `on` adds to the list;
with any other language fcitx5 passes keys through as typed. At login
omamobile-ime.service restarts squeekboard once fcitx5 holds the input
method, then follows the language. Off by default, and inert while off.

With it on, squeekboard no longer sees text fields (double-tap the home
strip), candidates take no taps, and the Kana layout can't be used (it
needs more keycodes than a keymap has). Needs hyprland 0.57.0-3.4 (derekross#762).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KgnCmDwY8r4r1jCa2SVYLN
ikegam added a commit to ikegam/omarchy-mobile that referenced this pull request Nov 14, 2026
…le-ime)

`omamobile-ime on` hands Hyprland's one input method to fcitx5 with Mozc,
and `off` gives it back to squeekboard. squeekboard's keys reach fcitx5
through the keyboard grab, so romaji converts with Space and commits with
Enter. Mozc converts only while the keyboard's language (the globe key) is
"Japanese (romaji)", a copy of the US layout that `on` adds to the list;
with any other language fcitx5 passes keys through as typed. At login
omamobile-ime.service restarts squeekboard once fcitx5 holds the input
method, then follows the language. Off by default, and inert while off.

With it on, squeekboard no longer sees text fields (double-tap the home
strip), candidates take no taps, and the Kana layout can't be used (it
needs more keycodes than a keymap has). Needs hyprland 0.57.0-3.4 (derekross#762).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KgnCmDwY8r4r1jCa2SVYLN
ikegam added a commit to ikegam/omarchy-mobile that referenced this pull request Nov 15, 2026
…le-ime)

`omamobile-ime on` hands Hyprland's one input method to fcitx5 with Mozc,
and `off` gives it back to squeekboard. squeekboard's keys reach fcitx5
through the keyboard grab, so romaji converts with Space and commits with
Enter. Mozc converts only while the keyboard's language (the globe key) is
"Japanese (romaji)", a copy of the US layout that `on` adds to the list;
with any other language fcitx5 passes keys through as typed. At login
omamobile-ime.service restarts squeekboard once fcitx5 holds the input
method, then follows the language. Off by default, and inert while off.

With it on, squeekboard no longer sees text fields (double-tap the home
strip), candidates take no taps, and the Kana layout can't be used (it
needs more keycodes than a keymap has). Needs hyprland 0.57.0-3.4 (derekross#762).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KgnCmDwY8r4r1jCa2SVYLN
@derekross

Copy link
Copy Markdown
Owner

Thank you, Hiroyuki. This is a clean fix: the grab only compared the keyboard, squeekboard keeps the same virtual keyboard and swaps its keymap, so fcitx5 kept decoding with the old one. Comparing the keymap as well is exactly the right place for it.

I checked it against the 0.57.0 release: the sums match, all six patches apply with --fuzz=0, and it also stacks cleanly with the three other Hyprland PRs that are open (#753, #785, #786), in any order. The changed function only runs while an input method holds a keyboard grab, so on a phone without fcitx5 it should do nothing at all.

Because it changes the compositor every phone runs, I'll run it on my Pixel 7 Pro before merging, together with the other three in one build, so that the phone only needs one Hyprland install. I'd like this one to go in first and keep 0.57.0-3.4, since #763 depends on it and says so; the others will move up a number.

One small thing, if you have it to hand: the pacman -Q hyprland and uname -r of the 7 Pro run you mention, in a comment. No need to change anything in the branch.

@ikegam

ikegam commented Nov 15, 2026 •

Copy link
Copy Markdown
Contributor Author

Thanks for checking it against the release and the other three, Derek. The 7 Pro run was on:

hyprland 0.57.0-3.4
7.3.0-rc6-omamobile-push1-gd2983420874d-dirty

(origin/omamobile at 82f31d3a7 with linux-omarchy-mobile#232 and your two patches.)

@derekross

Copy link
Copy Markdown
Owner

Thank you, Hiroyuki, that's the build line I was after. Nothing more is needed from you on this one. I've now applied it with the other three open Hyprland changes in every order they could merge, and the result is the same compositor each time, so it goes in first as 0.57.0-3.4, as planned, once my Pixel 7 Pro has run the combined build. #763 follows it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants