This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Scenario: - Create Quote sample - Open quote.cc file - Wait file is parsed (green square in corner) - Close file - Open file again ==> File is not reparsed. Semantic highlighting is broken. Mark Occurrence is broken.
the issue is that after closing the editor EditorRegistry.componentList() still contains editor => we do not detach. In fact the situation is even worse. When file is opened the second time => EditorRegistry.componentList() has two entries about the same file close file/open file => EditorRegistry.componentList() has three entries and so on. Looks like EditorRegistry.notifyClose(c) is not closed when needed. At least I do not see such events when turn on logging -J-Dorg.netbeans.api.editor.EditorRegistry.level=FINEST
FYI, issue is not reproducible in Beta and in trunk from 21 Oct We didn't have any code changes in our affected components.
Hi, I think I might maybe a similar problem. I had ACE library project open, alongside my custom project, when I tried to resolve a method from ACE API, in this case activate() from ACE_Task_Base.cpp, the parser wouldn't parse file. As I close the ACE project, everything was ok. The problem is that I have the ACE library installed in my system + I had the ACE project open in the workspace, I don't know if this helps... Thanks, Rolando
I'm investigating...
<rant> Caused by #481670dcdf35, which was caused by #b4dd900f3229, which itself was backed out and reapplied at least once and was a fix for 'Slowness detected' sort of a problem. It's a pity that fixing one 'Slowness detected' can lead to several P1 regressions. Maybe it's time to start fixing real problems reported by real people. </rant> Anyway, as long as openide.text is under JET's supervision I'm going to backout any change in this code that is not done by mslama or myself and is not a fix for P1 regression/showstopper. For those tempted, it's QA's call what is or is not a showstopper. http://hg.netbeans.org/jet-main/rev/6365cb5ab7d0
Thanks, now it works
> Anyway, as long as openide.text is under JET's supervision I'm going to backout any change in this code that is not done > by mslama or myself and is not a fix for P1 regression/showstopper. For those tempted, it's QA's call what is or is not > a showstopper. Agreed with Vita. Looks like previous changeset caused a lot of regressions, so be careful - at least for NB 6.8 .
Integrated into 'main-golden', will be available in build *200910301401* on http://bits.netbeans.org/dev/nightly/ (upload may still be in progress) Changeset: http://hg.netbeans.org/main-golden/rev/6365cb5ab7d0 User: Vita Stejskal <vstejskal@netbeans.org> Log: #175576: remove TopComponent's children later so that EditorRegistry can find its JEditorPane
Integrated into 'main-golden', will be available in build *200910310201* on http://bits.netbeans.org/dev/nightly/ (upload may still be in progress) Changeset: http://hg.netbeans.org/main-golden/rev/4e102577737e User: Vladimir Voskresensky <vv159170@netbeans.org> Log: investigating IZ#175576: Semantic Highlighting and Mark Occurrence are broken are broken if user reopens file