Issue 117193

Summary: [sw] crash in SwTxtNode::impl_FmtToTxtAttr
Product: Writer Reporter: mst.ooo
Component: codeAssignee: stefan.baltzer
Status: CLOSED FIXED QA Contact: issues@sw <issues>
Severity: Normal    
Priority: P2 CC: binbjguo, dtardon, issues, michael.ruess, stefan.baltzer
Version: OOo 3.3Keywords: regression
Target Milestone: 3.4.0   
Hardware: All   
OS: All   
Issue Type: DEFECT Latest Confirmation in: ---
Developer Difficulty: ---
Attachments:
Description Flags
Bugdoc to ease reproduction none

Description mst.ooo 2011-03-03 13:57:32 UTC
the stack trace id for this is: 1203697

managed to reproduce it:

1. split paragraph (both paras have default formatting)
2. first para: set italic on => para autostyle has italic
3. enter some text in first para, and in second para
4. first para: mark part of text, apply char style "Emphasis"
5. on the same text range marked in 4.:
   _additionally_ apply char style "Strong Emphasis" (shift+doubleclick)
6. join both paragraphs
7. crash with NULL pItem in RemovePresentAttrs::operator()

this is a regression from OOo 3.2, introduced by the fix for issue 108345.

after fixing this crash i've noticed an assertion on this scenario,
which is caused by a hint with end < start.
reason is a wrong comparison in lcl_CollectHintSpans,
which does not handle multiple hints with same start+end correctly.
damn... why haven't i tested this case back then...
Comment 1 mst.ooo 2011-03-03 14:17:25 UTC
fixed in CWS sw34bf04
http://hg.services.openoffice.org/hg/cws/sw34bf04/rev/96bf77a25739
Comment 2 stefan.baltzer 2011-03-16 16:09:10 UTC
I can not reproduce the crash with the given description. :-(
Comment 3 stefan.baltzer 2011-03-18 10:23:47 UTC
Created attachment 76142 [details]
Bugdoc to ease reproduction
Comment 4 stefan.baltzer 2011-03-18 10:26:47 UTC
Correcting priority to P2 (crash).
Thanks to OD, I could reproduce it now: In attached .odt, delete the paragraph end mark of first paragraph...
Verified in CWS sw34bf04.
Comment 5 Oliver-Rainer Wittmann 2012-06-13 12:22:22 UTC
getting rid of value "enhancement" for field "severity".
For enhancement the field "issue type" shall be used.
Comment 6 binguo 2012-06-20 08:29:00 UTC
Verified it on Aoo_Trunk_20120616.1800.1350879 and it does not reproduce, so close it as fixed.