Issue 106901 - Mac:crash when insert latin char with "Character viewer"
Summary: Mac:crash when insert latin char with "Character viewer"
Status: CLOSED FIXED
Alias: None
Product: General
Classification: Code
Component: www (show other issues)
Version: OOO310m19
Hardware: Mac Mac OS X, all
: P3 Trivial (vote)
Target Milestone: OOo 3.2
Assignee: h.ilter
QA Contact: issues@framework
URL:
Keywords: aqua
Depends on:
Blocks: 99999
  Show dependency tree
 
Reported: 2009-11-13 14:59 UTC by ed_baudry
Modified: 2017-05-20 10:29 UTC (History)
3 users (show)

See Also:
Issue Type: DEFECT
Latest Confirmation in: ---
Developer Difficulty: ---

Attachments
wrokaround (1.34 KB, patch)
2009-11-17 13:50 UTC, philipp.lohmann 		no flags Details | Diff
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this issue.
Description ed_baudry 2009-11-13 14:59:18 UTC 
Hi,

The crash has been noticed on Mac os X Leopard and Snow Leopard with Calc and
Writer, at least... with OpenOffice 3.1.1 (OOO310m19(Build:9420))

To Reproduce it:
->Start the system
->Open OpenOffice Writer
->Write few charaters with the keyboard
->Launch the "Character View". (This system tool can be enabled from "System
preferences/Language & text/Input Sources", then check the box near "Character
viewer")
->Into "Character viewer", select the "Latin" charset
->select the "A" character and press the "insert" button
->Then, OpenOffice crashes

Other informations:
->It has been reproduced with many palette applications, like "Character
Viewer", IPAPalette, or with the palette application we are working on.
->OpenOffice doesn't crash if:
  -> You don't enter at least one character with the keyboard before inserting
one with the palette application
  -> You don't try to insert Latin characters. We made lost's of tests and it
appears that if the unicode code point of the character you try to insert is
greater than "0x00ff" (UTF16), then OpenOffice doen't crash

Thanks
Edgard Baudry
Comment 1 Mathias_Bauer 2009-11-13 15:32:41 UTC 
Herbert, please take over.
Comment 2 hdu@apache.org 2009-11-13 16:01:24 UTC 
reassigning to the IME-expert
Comment 3 philipp.lohmann 2009-11-17 13:44:01 UTC 
confirm
Comment 4 philipp.lohmann 2009-11-17 13:49:01 UTC 
A workaround is to return a nonempty marked range in this case (where we are not
actually working on a key event). Why this is the case I do not quite
understand. Committed the workaround in CWS ooo32gsl06.

ekato: if you're interested you might want to have a look at the patch I will
committed as a fix.
Comment 5 philipp.lohmann 2009-11-17 13:50:35 UTC 
Created attachment 66153 [details]
wrokaround
Comment 6 philipp.lohmann 2009-11-18 10:25:21 UTC 
please verify in CWS ooo32gsl06
Comment 7 h.ilter 2009-11-23 14:21:33 UTC 
Verified with cws ooo32gsl06 = ok