GNOME Bugzilla – Bug 332026
First input character on calendar view is always ASCII
Last modified: 2013-08-26 11:18:34 UTC
I'm using SCIM for my favorite Input Method and it has a feature to share Input Context between applications, so-called the global IC mode. on this mode, Bug#264485 (CJK input gets replicated in Evolution Cal) disappears. but it introduces another problem. it is, even if I activate IM on a time slot and input some Japanese character say and click on another slot, first character is always ASCII. but next one is native character even without press ctrl+space say - ctrl+space is the default key to activate IM on SCIM. So it means IC recognizes IM is activated now. This is a bad behavior because this behavior enforces people to press backspace to delete the unexpected character once. How to reproduce this problem: 1. make sure if scim and scim-anthy works say 2. run scim-setup and go to FrondEnd->Global Setup and turn on "Share the same input method among all applications" checkbox. press OK button then. 3. run evolution with LANG=ja_JP.UTF-8 locale say, and open a calendar view. 4. click a timeslot somewhere and input a say. 5. press ctrl+space and make sure the scim panel appears and shows an icon like "あ". 6. type aka 7. click on another timeslot and type aka again. Actual Result: it shows like <u>aか</u> Expected Result: it should be <u>あか</u> step 4 is even a bug IMHO because I can't input Japanese from the first though.
Downstream bug - https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=182247
Created attachment 78679 [details] [review] Patch for solving bug
copying from downstream: Comment #11 From Matthew Barnes on 2006-11-06 Patch applied to evolution-2.9.1-3.fc7. Comment #13 From A S Alam on 2007-04-27 04:21 with ja_JP, aka is working ok, when goto send slot, it is shown without ASCII. tested with following version: evolution-2.10.1-4.fc7 so is this safe to commit to upstream trunk? matthew, what do you think?
Created attachment 94459 [details] [review] Patch from RHEL5 Here's the final patch we wound up going with for RHEL5. I just verified that it applies cleanly and fixes the bug. Also passed our internal QA and all that. Any objections to me committing this?
Matthew, one little observation: won't there be a regression in week/month views? I guess, similar changes for (e_week_view_do_key_press), (e_week_view_start_editing_event) would fix that :-)
Fix has been committed to svn HEAD.
To confirm, has any released Evolution been applied this fix? I can still see this problem on 2.21.5
the patch mentioned here has been committed and is included in 2.21.5: http://svn.gnome.org/viewvc/evolution/trunk/calendar/gui/e-day-view.c?r1=34156&r2=34204 so perhaps it did not fix this issue.
Bumping version to a stable release.
Using GNOME 3.8 and Russian hardware keyboard input via the dropdown in the top bar (IBus), I cannot reproduce the problem anymore. First character is Russian. Closing as obsolete, but please reopen if this is still an issue in a recent Evolution version (3.8 or 3.6). Thanks!