Issue 46948 - PDF export crashes for document with unicode surrogate pair
Summary: PDF export crashes for document with unicode surrogate pair
Status: CLOSED FIXED
Alias: None
Product: gsl
Classification: Code
Component: code (show other issues)
Version: OOo 2.0
Hardware: All Windows XP
: P3 Trivial (vote)
Target Milestone: OOo 2.0.1
Assignee: hdu@apache.org
QA Contact: issues@gsl
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-04-07 14:36 UTC by hdu@apache.org
Modified: 2005-10-19 15:09 UTC (History)
2 users (show)

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


Attachments

Note You need to log in before you can comment on or make changes to this issue.
Description hdu@apache.org 2005-04-07 14:36:17 UTC
Reproduction:
- install the latest Doulos font from http://scripts.sil.org/DoulosSILfont
- load the first sample document from issue 45983
- export this as PDF
=> OOo crashes in MultiSalLayout::GetNextGlyphs()
Comment 1 hdu@apache.org 2005-04-07 14:39:00 UTC
Found in SRC680m91 wntmsci10.pro.
Comment 2 hdu@apache.org 2005-05-17 14:23:09 UTC
Found it: An integer overflow resulted in a negative index value and accessing
of the value there causes the crash. The fix gets into CWS vcl39.
Comment 3 hdu@apache.org 2005-05-19 11:19:53 UTC
Also the fallback levels was allowed too deep because only one half of the
surrogate pair was recognized as unresolved.
Comment 4 hdu@apache.org 2005-06-15 11:18:52 UTC
HDU->US: please verify that the crash is fixed in CWS vcl39.

re-open issue and reassign to us@openoffice.org
Comment 5 hdu@apache.org 2005-06-15 11:18:55 UTC
reassign to us@openoffice.org
Comment 6 hdu@apache.org 2005-06-15 11:18:59 UTC
reset resolution to FIXED
Comment 7 ulf.stroehler 2005-08-25 16:52:53 UTC
US->HI: could you pls. verify this one on WinXP.
Looks good on Un*x. 
Thanks.

re-open issue and reassign to hi@openoffice.org
Comment 8 ulf.stroehler 2005-08-25 16:52:58 UTC
reassign to hi@openoffice.org
Comment 9 ulf.stroehler 2005-08-25 16:53:03 UTC
reset resolution to FIXED
Comment 10 h.ilter 2005-09-12 09:47:39 UTC
Verified with vcl39 = No crash -> OK
Comment 11 h.ilter 2005-10-11 13:48:57 UTC
Verified with 680m133_8966 = broken
Comment 12 h.ilter 2005-10-11 13:50:58 UTC
Back to HDU
Comment 13 hdu@apache.org 2005-10-18 15:53:06 UTC
With the increased number of existing GF alternatives the overflow is now
triggered after more fallback levels...
Comment 14 h.ilter 2005-10-19 15:07:46 UTC
I did an wrong comment on Tue Oct 11 05
Comment 15 h.ilter 2005-10-19 15:09:00 UTC
The pdf export still not crash with master 680m134_8969