|
Bugzilla – Full Text Bug Listing |
| Summary: | Ctrl-Space doesn't enable Chinese input. | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE 11.0 | Reporter: | Xin Wei Hu <xwhu> |
| Component: | Other | Assignee: | Dirk Mueller <dmueller> |
| Status: | RESOLVED FIXED | QA Contact: | E-mail List <qa-bugs> |
| Severity: | Major | ||
| Priority: | P5 - None | CC: | ast, coolo, forgotten_tfBQIqYkT6, helios_reds, kde-maintainers, shinkichi.yamazaki, swyear, tiwai |
| Version: | RC 2 | Flags: | coolo:
SHIP_STOPPER-
|
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Other | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
| Attachments: | Patch for libqt4 to delay input method initializing until chosen | ||
|
Description
Xin Wei Hu
2008-06-09 13:23:58 UTC
Strange, nothing has changed in our scim packages recently. Which desktop are you using? It happens with GNOME, KDE3 and KDE4. It turns out the scim is started as root in the first place. After killing the scim processes, and start it as the current user, everything back to normal. I can reproduce that with RC4. So far I can confirm it's a problem of KDM4. Because in qt4 imsw-multi will load all input method plugins, and initialize all input methods. When scim-bridge is initialized, the scim and scim-bridge daemon will be launched. So in qt4, no matter what input method will be used, all input method will be initialized. When kdm_greet is launched, the scim will be automatically launched. And the scim of root user will block ctrl-space. So I make the patch to delay input method initializing until the input method is chosen. Created attachment 222635 [details]
Patch for libqt4 to delay input method initializing until chosen
The attached patch fixes the issue. Thank you very much! Testing ... Yes, this patch helps. libqt4 with that patch submitted to STABLE/Factory. As this is really a showstopper for 11.0 for CJK, can we do an online update please? there is an accessor for the slaves array in the header file (accessor slave() ) which is not patched by this patch. therefore it is unsafe to apply. disabling input method support in kdm4 is not a good idea to resolve this issue? I think disabling input method support in kdm4 would be OK as well. *** Bug 372858 has been marked as a duplicate of this bug. *** Can we fix this with an online update, please? yup, will do that for 11.0 submitted for 11.0 released |