https://bugs.kde.org/show_bug.cgi?id=520512

--- Comment #5 from Osamu Goto <[email protected]> ---
(In reply to Not My Real Name from comment #4)
> (In reply to Osamu Goto from comment #3)
> > - Environment: Fedora 44 / Fedora RPMs
> > - Reproduction: Reproducible with LANG=ja_JP.UTF-8, but initially could not
> > reproduce with en_US.UTF-8.
> > - Workaround: Disabling ktranscript.so avoids the issue.
> > - Backtrace: Contains 'malloc_printerr' and 
> > 'QV4::ExecutableCompilationUnit'.
> 
> 
> Just to add some context regarding the locale: I originally encountered and
> reported this crash while using the Finnish locale (fi_FI.UTF-8). 
> This confirms the issue is not isolated strictly to the Japanese locale, but
> seems to affect other non-English translations as well. 
> Thank you for looking into this so quickly!

Thanks for the additional locale context (fi_FI.UTF-8).

Our independent verification on Fedora 44 aligns with a broader pattern:

- Reproduces with LANG=ja_JP.UTF-8 on exit (KTranscriptImp / ktranscript.so in
backtrace).
- Does NOT reproduce with LANG=en_US.UTF-8 in initial tests.
- Still reproduces with LANG=ja_JP.UTF-8 + LC_MESSAGES=en_US.UTF-8 (so not just
message catalog loading).
- Workaround: hiding ktranscript.so prevents crash while keeping non-English
UI.

Together with your fi_FI report, this suggests the bug is tied to ktranscript
being loaded for non-English locales in general, not Japanese specifically.
Component frameworks-ki18n may be more appropriate than systemsettings alone.

Versions: plasma-systemsettings 6.6.5, kf6-ki18n 6.26.0, Qt 6.11.1, Wayland.

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to