[Fedora-i18n-bugs] [Bug 824307] New: unassigned keysym causes crash

bugzilla at redhat.com bugzilla at redhat.com
Wed May 23 09:02:55 UTC 2012


https://bugzilla.redhat.com/show_bug.cgi?id=824307

            Bug ID: 824307
        QA Contact: extras-qa at fedoraproject.org
          Severity: unspecified
           Version: rawhide
          Priority: unspecified
                CC: apatil at redhat.com, i18n-bugs at lists.fedoraproject.org
          Assignee: apatil at redhat.com
           Summary: unassigned keysym causes crash
        Regression: ---
      Story Points: ---
    Classification: Fedora
                OS: Unspecified
          Reporter: dueno at redhat.com
              Type: Bug
     Documentation: ---
          Hardware: Unspecified
        Mount Type: ---
            Status: NEW
         Component: ibus-hunspell-table
           Product: Fedora

Created attachment 586281
  --> https://bugzilla.redhat.com/attachment.cgi?id=586281&action=edit
patch

Description of problem:
When ibus-hunspell-table is enabled, typing key with unassigned keysym
(NoSymbol) will cause crash.

Version-Release number of selected component (if applicable):
ibus-hunspell-table-0.0.6-1.fc17

How reproducible:
100%

Steps to Reproduce:
1. xmodmap -e "keycode 96 = NoSymbol"
2. enable "English - US(Hunspell)" engine in IBus menu
3. type F12

Actual results:
The engine process crashes with the following message:

 process 6891: Array or variant type requires that type string be written, but
end_struct was written.
 The overall signature expected here was '(sa{sv}s(sa{sv}av))' and we are on
byte 8 of that signature.
  D-Bus not built with -rdynamic so unable to print a backtrace

Expected results:
No crash.

Additional info:
I noticed this when using international keyboard with more keys than US
keyboard.  Since IBus switches the system keyboard layout with "setxkbmap us",
some keys (here "|") are unmapped.

It seems that ASCII range check in _table_mode_process_key_event() is fragile
(patch attached).

-- 
You are receiving this mail because:
You are on the CC list for the bug.


More information about the i18n-bugs mailing list