Bug 515350 - Cannot bring up IBus
Summary: Cannot bring up IBus
Keywords:
Status: CLOSED DUPLICATE of bug 517001
Alias: None
Product: Fedora
Classification: Fedora
Component: gtk2
Version: rawhide
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: Matthias Clasen
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
: 515036 (view as bug list)
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2009-08-03 20:00 UTC by Pete Zaitcev
Modified: 2009-08-27 06:47 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2009-08-27 06:47:53 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
.xsession-errors (9.11 KB, text/plain)
2009-08-04 04:09 UTC, Pete Zaitcev
no flags Details

Description Pete Zaitcev 2009-08-03 20:00:45 UTC
Description of problem:

When I hit Ctrl-S, nothing happens.

Version-Release number of selected component (if applicable):

ibus-1.2.0.20090723-2.fc12.x86_64

How reproducible:

100%

Steps to Reproduce:
1. Enable IBus with Anthy
  
Actual results:

No input method

Expected results:

Hiragana

Additional info:

This is probably not IBus per se, but something else, because SCIM broke
over the weekend in exactly the same fashion.

Comment 1 Akira TAGOH 2009-08-04 03:43:41 UTC
1. Have you used im-chooser to enable IBus?
2. Can you see ibus blue icon at the systray?
3. When you run im-chooser again, is the check box turned on?

If not for any of the above questions, please attach your .xsession-errors and make sure if $HOME/.xinputrc points to ibus.conf.

Comment 2 Pete Zaitcev 2009-08-04 04:07:20 UTC
1. Yes, I have.
2. The icon is there. I can right-click on it and that brings IBus
   preferences dialog.
3. The "Enable input method feature" checkbox is on.

[zaitcev@niphredil ~]$ ls -ld .xinputrc 
lrwxrwxrwx 1 zaitcev zaitcev 33 2009-07-31 19:55 .xinputrc -> /etc/X11/xinit/xinput.d/ibus.conf
[zaitcev@niphredil ~]$ 

There's nothing interesting in .xsession-errors, but I'll attach it
just in case.

Comment 3 Pete Zaitcev 2009-08-04 04:09:18 UTC
Created attachment 356093 [details]
.xsession-errors

BTW, just in case too:

[zaitcev@niphredil ~]$ ps auxww| grep ibus
zaitcev   5792  0.1  0.1 136864  3148 ?        S    22:05   0:00 /usr/bin/ibus-daemon --xim
zaitcev   5793  0.1  0.1 141432  3472 ?        S    22:05   0:00 /usr/libexec/ibus-gconf
zaitcev   5794  0.7  1.2 295636 24228 ?        S    22:05   0:00 python /usr/share/ibus/ui/gtk/main.py
zaitcev   5796  0.0  0.3 218488  6328 ?        S    22:05   0:00 /usr/libexec/ibus-x11 --kill-daemon
zaitcev   5797  0.6  0.6 208264 11876 ?        S    22:05   0:00 python /usr/share/ibus-anthy/engine/main.py --ibus
zaitcev   5818  0.0  0.0  89068   832 pts/2    S+   22:06   0:00 grep ibus
[zaitcev@niphredil ~]$

Comment 4 Peng Huang 2009-08-04 04:23:25 UTC
Please try below command, and tell us if the ibus can work in gedit. thanks
GTK_IM_MODULE=ibus gedit

Comment 5 Mamoru TASAKA 2009-08-04 04:27:06 UTC
Is this related to bug 514720?

Comment 6 Pete Zaitcev 2009-08-04 04:38:21 UTC
The "GTK_IM_MODULE=ibus gedit" has no effect: Ctrl-Space does not work in it.

Regarding bug 514720, I cannot tell for sure, but I know that nothing
segfaults here. My kernel logs all segfaults and I only have one such
(from libvirtd at system boot-up). So there are no crashes in X or GNOME.

Comment 7 Peng Huang 2009-08-04 05:20:27 UTC
Please upload /etc/gtk-2.0/{i386, i686 or x86_64}-redhat-linux-gnu/gtk.immodules.

Comment 8 Pete Zaitcev 2009-08-04 15:04:58 UTC
I see, it's empty:

[zaitcev@niphredil ~]$ ls  -l /etc/gtk*/*/gtk.immodules 
-rw-r--r-- 1 root root 587 2009-08-03 12:50 /etc/gtk-2.0/i386-redhat-linux-gnu/gtk.immodules
-rw-r--r-- 1 root root   0 2009-08-03 12:49 /etc/gtk-2.0/x86_64-redhat-linux-gnu/gtk.immodules
[zaitcev@niphredil ~]$ 

And the i386 one only contains references to SCIM:

[zaitcev@niphredil ~]$ cat /etc/gtk-2.0/i386-redhat-linux-gnu/gtk.immodules
# GTK+ Input Method Modules file
# Automatically generated file, do not edit
# Created by /usr/bin/gtk-query-immodules-2.0-32 from gtk+-2.17.6
#
# ModulesPath = /root/.gtk-2.0/2.10.0/i386-redhat-linux-gnu/immodules:/root/.gtk-2.0/2.10.0/immodules:/root/.gtk-2.0/i386-redhat-linux-gnu/immodules:/root/.gtk-2.0/immodules:/usr/lib/gtk-2.0/2.10.0/i386-redhat-linux-gnu/immodules:/usr/lib/gtk-2.0/2.10.0/immodules:/usr/lib/gtk-2.0/i386-redhat-linux-gnu/immodules:/usr/lib/gtk-2.0/immodules
#
"/usr/lib/gtk-2.0/immodules/im-scim-bridge.so" 
"scim-bridge" "SCIM Bridge Input Method" "" "" "" 

[zaitcev@niphredil ~]$

Comment 9 Pete Zaitcev 2009-08-04 17:23:27 UTC
If I run gtk-query-immodules-2.0-64, it crashes. However, if I run it
piped to terminal, it produes the file gtk.immodules, then crashes.
If I run it with output redirected, it crashes without producing
the output. I speculate that it happens because stdio buffers output
if it's not directed to terminal, and the program crashes before
exit() flushes the output.

Here's what gdb says (all debuginfos installed):

(gdb) where
#0  0x00007ffff2402620 in ?? ()
#1  0x00007ffff3b5cb32 in exit () from /lib64/libc.so.6
#2  0x00007ffff3b44b94 in __libc_start_main () from /lib64/libc.so.6
#3  0x0000000000400fb9 in _start ()
(gdb) quit

Comment 10 Pete Zaitcev 2009-08-04 20:53:25 UTC
*** Bug 515036 has been marked as a duplicate of this bug. ***

Comment 11 Peng Huang 2009-08-05 23:07:30 UTC
I think this bug is in gtk2.

Comment 12 Matthias Clasen 2009-08-06 04:22:38 UTC
Not a very useful stacktrace, unfortunately. It doesn't segfault here, of course...

Comment 13 Pete Zaitcev 2009-08-13 23:40:48 UTC
I have two systems, gtk-query-immodues-2.0-64 crashes on one (above)
and does not crash on the other. Resulting file on the one that does NOT
crash:

# GTK+ Input Method Modules file
# Automatically generated file, do not edit
# Created by /usr/bin/gtk-query-immodules-2.0-64 from gtk+-2.17.6
#
# ModulesPath = /root/.gtk-2.0/2.10.0/x86_64-redhat-linux-gnu/immodules:/root/.gtk-2.0/2.10.0/immodules:/root/.gtk-2.0/x86_64-redhat-linux-gnu/immodules:/root/.gtk-2.0/immodules:/usr/lib64/gtk-2.0/2.10.0/x86_64-redhat-linux-gnu/immodules:/usr/lib64/gtk-2.0/2.10.0/immodules:/usr/lib64/gtk-2.0/x86_64-redhat-linux-gnu/immodules:/usr/lib64/gtk-2.0/immodules
#
"/usr/lib64/gtk-2.0/2.10.0/immodules/im-ibus.so" 
"ibus" "IBus (Intelligent Input Bus)" "ibus" "" "ja:ko:zh" 

I can only say that
 1. the system does not have SCIM installed
 2. the file is shorter, so maybe the longer one overflows a buffer somewhere

Comment 14 Pete Zaitcev 2009-08-13 23:43:36 UTC
Indeed, if I remove all SCIM packages, gtk-query-immodules-2.0-64 runs
to completion without a crash.

rpm -e scim-bridge-0.4.16-2.fc12.x86_64 scim-tables-russian-0.5.9-2.fc12.x86_64 scim-1.4.9-2.fc12.x86_64 scim-bridge-gtk-0.4.16-2.fc12.i686 scim-libs-1.4.9-2.fc12.i686 scim-libs-1.4.9-2.fc12.x86_64 scim-bridge-gtk-0.4.16-2.fc12.x86_64 scim-gtk-1.4.9-2.fc12.x86_64 scim-qtimm-0.9.4-13.fc12.x86_64 scim-anthy-1.2.7-3.fc12.x86_64 scim-tables-0.5.9-2.fc12.x86_64 scim-lang-japanese-1.4.9-2.fc12.x86_64

Comment 15 Pete Zaitcev 2009-08-13 23:46:08 UTC
This is how gtk.immodules would've looked if not for crash:

# GTK+ Input Method Modules file
# Automatically generated file, do not edit
# Created by gtk-query-immodules-2.0-64 from gtk+-2.17.7
#
# ModulesPath = /root/.gtk-2.0/2.10.0/x86_64-redhat-linux-gnu/immodules:/root/.gtk-2.0/2.10.0/immodules:/root/.gtk-2.0/x86_64-redhat-linux-gnu/immodules:/root/.gtk-2.0/immodules:/usr/lib64/gtk-2.0/2.10.0/x86_64-redhat-linux-gnu/immodules:/usr/lib64/gtk-2.0/2.10.0/immodules:/usr/lib64/gtk-2.0/x86_64-redhat-linux-gnu/immodules:/usr/lib64/gtk-2.0/immodules
#
"/usr/lib64/gtk-2.0/2.10.0/immodules/im-ibus.so" 
"ibus" "IBus (Intelligent Input Bus)" "ibus" "" "ja:ko:zh" 

"/usr/lib64/gtk-2.0/immodules/im-scim-bridge.so" 
"scim-bridge" "SCIM Bridge Input Method" "" "" "" 

"/usr/lib64/gtk-2.0/immodules/im-scim.so" 
"scim" "SCIM Input Method" "scim" "/usr/share/locale" ""

Comment 16 Mamoru TASAKA 2009-08-14 02:40:19 UTC
I guess the real cause of this issue is bug 517001.

Comment 17 Mamoru TASAKA 2009-08-27 06:47:53 UTC

*** This bug has been marked as a duplicate of bug 517001 ***


Note You need to log in before you can comment on or make changes to this bug.