Bug 122252

Summary: hit ctrl-d instead of ctrl-s, quanta crashed
Product: [Unmaintained] quanta Reporter: Mathieu Jobin <opensource>
Component: generalAssignee: András Manţia <amantia>
Status: RESOLVED WORKSFORME    
Severity: crash    
Priority: NOR    
Version: unspecified   
Target Milestone: ---   
Platform: unspecified   
OS: Linux   
Latest Commit: Version Fixed In:
Sentry Crash Report:
Attachments: file that I suspect make quanta crash

Description Mathieu Jobin 2006-02-18 23:07:53 UTC
Version:           unknown (using KDE 3.5.1, Gentoo)
Compiler:          gcc version 3.3.6 (Gentoo 3.3.6, ssp-3.3.6-1.0, pie-8.7.8)
OS:                Linux (i686) release 2.6.15-suspend2-r6-justbudget

hit ctrl-d instead of ctrl-s, quanta crashed

Using host libthread_db library "/lib/libthread_db.so.1".
`system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
[Thread debugging using libthread_db enabled]
[New Thread 16384 (LWP 8862)]
[KCrash handler]
#6  0x0817c337 in AreaStruct (this=0xbf972fc0, area=@0x1c) at tag.h:59
#7  0x0817c17a in Tag::area (this=0x0) at tag.h:113
#8  0x0817a92b in SAParser::slotParseNodeInDetail (this=0x8580238)
    at saparser.cpp:896
#9  0x0817b019 in SAParser::qt_invoke (this=0x8580238, _id=4, _o=0xbf9730d0)
    at saparser.moc:115
#10 0xb64a46ac in QObject::activate_signal (this=0x85a76b0, clist=0x851d1a0, 
    o=0xbf9730d0) at qobject.cpp:2355
#11 0xb64a4504 in QObject::activate_signal (this=0x85a76b0, signal=2)
    at qobject.cpp:2324
#12 0xb6889c64 in QTimer::timeout (this=0x85a76b0) at moc_qtimer.cpp:82
#13 0xb64d009e in QTimer::event (this=0x85a76b0, e=0xbf9733f0)
    at qtimer.cpp:219
#14 0xb6434235 in QApplication::internalNotify (this=0x83a3f18, 
    receiver=0x85a76b0, e=0xbf9733f0) at qapplication.cpp:2635
#15 0xb6433478 in QApplication::notify (this=0x83a3f18, receiver=0x85a76b0, 
    e=0xbf9733f0) at qapplication.cpp:2358
#16 0xb6c7ccff in KApplication::notify (this=0x83a3f18, receiver=0x85a76b0, 
    event=0xbf9733f0) at kapplication.cpp:550
#17 0xb7e755e9 in QApplication::sendEvent (receiver=0x1c, event=0xbf972fc0)
    at qapplication.h:491
#18 0xb641f75f in QEventLoop::activateTimers (this=0x84795e0)
    at qeventloop_unix.cpp:556
#19 0xb63cfd5d in QEventLoop::processEvents (this=0x84795e0, flags=4)
    at qeventloop_x11.cpp:389
#20 0xb644b7d9 in QEventLoop::enterLoop (this=0x84795e0) at qeventloop.cpp:198
#21 0xb644b6f2 in QEventLoop::exec (this=0x84795e0) at qeventloop.cpp:145
#22 0xb64343d7 in QApplication::exec (this=0x83a3f18) at qapplication.cpp:2758
#23 0x080b3a59 in main (argc=1, argv=0xbf9737e4) at main.cpp:212
Comment 1 Mathieu Jobin 2006-02-22 06:28:20 UTC
this happen very often (10 times / day)

Using host libthread_db library "/lib/libthread_db.so.1".
`system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
[Thread debugging using libthread_db enabled]
[New Thread 16384 (LWP 10203)]
[KCrash handler]
#6  0x0817c337 in AreaStruct (this=0xbfada6d0, area=@0x1c) at tag.h:59
#7  0x0817c17a in Tag::area (this=0x0) at tag.h:113
#8  0x0817a92b in SAParser::slotParseNodeInDetail (this=0x84d8ee0)
    at saparser.cpp:896
#9  0x0817b019 in SAParser::qt_invoke (this=0x84d8ee0, _id=4, _o=0xbfada7e0)
    at saparser.moc:115
#10 0xb650b6ac in QObject::activate_signal (this=0x84cddf8, clist=0x858d280, 
    o=0xbfada7e0) at qobject.cpp:2355
#11 0xb650b504 in QObject::activate_signal (this=0x84cddf8, signal=2)
    at qobject.cpp:2324
#12 0xb68f0c64 in QTimer::timeout (this=0x84cddf8) at moc_qtimer.cpp:82
#13 0xb653709e in QTimer::event (this=0x84cddf8, e=0xbfadab00)
    at qtimer.cpp:219
#14 0xb649b235 in QApplication::internalNotify (this=0x83a3f18, 
    receiver=0x84cddf8, e=0xbfadab00) at qapplication.cpp:2635
#15 0xb649a478 in QApplication::notify (this=0x83a3f18, receiver=0x84cddf8, 
    e=0xbfadab00) at qapplication.cpp:2358
#16 0xb6ce3cff in KApplication::notify (this=0x83a3f18, receiver=0x84cddf8, 
    event=0xbfadab00) at kapplication.cpp:550
#17 0xb7edc5e9 in QApplication::sendEvent (receiver=0x1c, event=0xbfada6d0)
    at qapplication.h:491
#18 0xb648675f in QEventLoop::activateTimers (this=0x8479618)
    at qeventloop_unix.cpp:556
#19 0xb6436d5d in QEventLoop::processEvents (this=0x8479618, flags=4)
    at qeventloop_x11.cpp:389
#20 0xb64b27d9 in QEventLoop::enterLoop (this=0x8479618) at qeventloop.cpp:198
#21 0xb64b26f2 in QEventLoop::exec (this=0x8479618) at qeventloop.cpp:145
#22 0xb649b3d7 in QApplication::exec (this=0x83a3f18) at qapplication.cpp:2758
#23 0x080b3a59 in main (argc=1, argv=0xbfadaef4) at main.cpp:212

Comment 2 András Manţia 2006-02-22 11:01:48 UTC
CTRL-D is the shortcut to insert a comment. Does the crash happen if you 
are in a certain part of the document when you press CTRL-D? Altough 
your backtrace is quite good, I would like to see an example file which 
can be used to reproduce the bug.
Comment 3 Mathieu Jobin 2006-02-22 12:15:45 UTC
actually, I first notice it while pressing CTRL-D
but it seems to happen after saving as well, and also while typing....

I edit .rb/.rxml/.rhtml files with Quanta, mostly.

Comment 4 András Manţia 2006-02-22 12:58:48 UTC
I would really appreciate to have an example file and a way to 
reproduce. For some reason a pointer becomes null, which should not 
happen, but I am unable to trigger the bug myself.
Comment 5 Mathieu Jobin 2006-02-23 06:38:30 UTC
Created attachment 14820 [details]
file that I suspect make quanta crash

i just deleted the three "ok" lines, and I saved.

it crashed

i'm not totally sure that what cause it, but its the last thing I remember
having done.
Comment 6 András Manţia 2006-02-28 22:38:17 UTC
SVN commit 514630 by amantia:

This should finnally fix a grave crash bug in Quanta, which appeared due to a node tree corruption.The patch does other things as well:
- stop background detailed parsing if the text was changed (this caused the corruption)
- parse only once after the user stopped typing, for real. The intention was there since several versions but DID NOT work! Now this should really make typing faster.

I will mark all related bugs as fixed. Reporters, in case you still see the crash or slow typing with Quanta 3.5.2 (or a recent  - after this commit - snapshot/svn version), reopen the report or open a new one.

BUG: 121280, 122475, 122252, 120983

 M  +1 -1      parsers/node.cpp  
 M  +23 -7     parsers/parser.cpp  
 M  +1 -1      parsers/parsercommon.cpp  
 M  +2 -0      parsers/saparser.cpp  
 M  +9 -4      src/document.cpp  
 M  +1 -0      src/document.h  


--- branches/KDE/3.5/kdewebdev/quanta/parsers/node.cpp #514629:514630
@@ -52,7 +52,7 @@
   //If it has crashed here, the Node doesn't exist anymore.
   // If it has crashed the next line, it is a GroupElements bug.
   //FIXME: Andras: or it is a VPL undo/redo bug...
-/*  Q_ASSERT(tag); */ //disable as on some distros it causes a real assert=crash
+  Q_ASSERT(tag);  //disable as on some distros it causes a real assert=crash
   if (tag)
     tag->setCleanStrBuilt(false);
 
--- branches/KDE/3.5/kdewebdev/quanta/parsers/parser.cpp #514629:514630
@@ -43,8 +43,11 @@
 #include "quantacommon.h"
 #include "document.h"
 #include "qextfileinfo.h"
+
+
 #include "kafkacommon.h"
 #include "undoredo.h"
+
 #include "dtds.h"
 #include "structtreetag.h"
 
@@ -515,10 +518,12 @@
  // clearGroups();
   if (baseNode)
   {
-    delete baseNode;
-    baseNode = 0L;
-    m_node = 0L;
+     //kdDebug(24000) << "baseNode before delete = " << baseNode << endl;
+     //ParserCommon::coutTree(m_node, 2);
+     delete baseNode;
+     baseNode = 0L;
   }
+  m_node = 0L;
 
   Node *lastNode;
   write = w;
@@ -657,6 +662,7 @@
 }
 void Parser::logReparse(NodeModifsSet *modifs, Document *w)
 {
+
   NodeModif *modif;
   if (baseNode)
   {
@@ -796,6 +802,7 @@
   Node *node = firstNode;
   bool closesPrevious = false;
   NodeModif *modif;
+
   //delete all the nodes between the firstNode and lastNode
   while (node && node != lastNode )
   {
@@ -829,8 +836,11 @@
     node->parent = 0L;
     node->next = 0L;
     node->prev = 0L;
+    
+    //delete node;
     node->detachNode();
     modif->setNode(node);
+
     node = 0L;
     i = 0;
     j = 0;
@@ -936,9 +946,11 @@
 
       }
     }
+
     modif->setChildrenMovedUp(i);
     modif->setNeighboursMovedDown(j);
     modifs->addNodeModif(modif);
+
     node = nextNode;
 
  //   kdDebug(24000)<< "Node removed!" << endl;
@@ -961,7 +973,6 @@
  NodeModifsSet *modifs = new NodeModifsSet();
  NodeModif *modif;
 
-
 // kdDebug(24000)<< "Node *Parser::rebuild()" << endl;
  modifs->setIsModifiedAfter(w->isModified());
 
@@ -989,9 +1000,10 @@
         (area.eLine < area.bLine || (area.eLine == area.bLine && area.eCol <= area.bCol)) //something strange has happened, like moving text with D&D inside the editor
       )
    {
-     logReparse(modifs, w);
-     m_saParser->setParsingEnabled(saParserEnabled);
-     return parse(w);
+      logReparse(modifs, w);
+       m_saParser->setParsingEnabled(saParserEnabled);
+       Node *n = parse(w, true);
+       return n;
    }
 
    kdDebug(24000) << QString("Invalid area: %1,%2,%3,%4").arg(area.bLine).arg(area.bCol).arg(area.eLine).arg(area.eCol) << "\n";
@@ -1071,6 +1083,7 @@
         modif = new NodeModif();
         modif->setType(NodeModif::NodeRemoved);
         modif->setLocation(kafkaCommon::getLocation(lastNode));
+
         if(lastInserted->prev)
           lastInserted->prev->next = 0L;
         if(lastInserted->parent && lastInserted->parent->child == lastInserted)
@@ -1079,6 +1092,8 @@
         lastInserted->next = 0L;
         lastInserted->parent = 0L;
         lastInserted->child = 0L;
+//        delete lastInserted;
+
         lastInserted->detachNode();
         modif->setNode(lastInserted);
         modifs->addNodeModif(modif);
@@ -1185,6 +1200,7 @@
 /*   kdDebug(24000)<< "END"<< endl;
    ParserCommon::coutTree(baseNode,  2);
    kdDebug(24000)<< "************* End User Modification *****************" << endl;*/
+
    w->docUndoRedo->addNewModifsSet(modifs, undoRedo::SourceModif);
  }
   kdDebug(24000) << "Rebuild: " << t.elapsed() << " ms \n";
--- branches/KDE/3.5/kdewebdev/quanta/parsers/parsercommon.cpp #514629:514630
@@ -220,7 +220,7 @@
             else
                     output+= node->tag->tagStr().replace('\n'," ");
             kdDebug(24000) << output <<" (" << node->tag->type << ") at pos " <<
-                    bLine << ":" << bCol << " - " << eLine << ":" << eCol << " This: "<< node << " Parent: " << node->parent << " Prev: " << node->prev << " Next: " << node->next << " Child: " << node->child << endl;
+                    bLine << ":" << bCol << " - " << eLine << ":" << eCol << " This: "<< node << " Parent: " << node->parent << " Prev: " << node->prev << " Next: " << node->next << " Child: " << node->child << " Tag:" << node->tag << endl;
  /*           for(j = 0; j < node->tag->attrCount(); j++)
             {
                     kdDebug(24000)<< " attr" << j << " " <<
--- branches/KDE/3.5/kdewebdev/quanta/parsers/saparser.cpp #514629:514630
@@ -734,6 +734,7 @@
         Node *n = m_lastParsedNode;
         if (m_useNext)
         {
+//           kdDebug(24000) << "m_lastParsedNode: " << m_lastParsedNode << endl;
           n->next = s_next;
           if (s_next)
             s_next->prev = n;
@@ -827,6 +828,7 @@
   {
       if (m_useNext && s_currentNode)
       {
+//           kdDebug(24000) << "s_currentNode: " << s_currentNode << endl;
         Node *n = s_currentNode;
         n->next = s_next;
         if (s_next)
--- branches/KDE/3.5/kdewebdev/quanta/src/document.cpp #514629:514630
@@ -194,6 +194,7 @@
   dtdName = Project::ref()->defaultDTD();
   reparseEnabled = true;
   repaintEnabled = true;
+  delayedTextChangedEnabled = true;
   docUndoRedo = new undoRedo(this);
 
   //each document remember wheter it has a entry in quantarc
@@ -2264,11 +2265,13 @@
 void Document::slotTextChanged()
 {
   changed = true;
-  if (reparseEnabled)
+  parser->setSAParserEnabled(false); //disable special area parsing if the text was changed.
+  if (reparseEnabled && delayedTextChangedEnabled)
   {
     kdDebug(24000) << "Delayed text changed called." << endl;
     //delay the handling, otherwise we may get wrong values for (line,column)
     QTimer::singleShot(0, this, SLOT(slotDelayedTextChanged()));
+    delayedTextChangedEnabled = false;  
   }
 }
 
@@ -2276,13 +2279,14 @@
 {
    if (!forced && typingInProgress)
    {
-    // kdDebug(24000) << "Reparsing delayed!" << endl;
+     kdDebug(24000) << "Reparsing delayed!" << endl;
      parser->setParsingNeeded(true);
      QTimer::singleShot(1000, this, SLOT(slotDelayedTextChanged()));
      reparseEnabled = false;
+     delayedTextChangedEnabled = false;	
      return;
    }
-
+	
     uint line, column;
     QString oldNodeName = "";
     Node *node;
@@ -2311,7 +2315,7 @@
         }
       }
     }
-
+    parser->setSAParserEnabled(true); //enable special area parsing, it was disabled in slotTextChanged()
     baseNode = parser->rebuild(this);
     if (qConfig.updateClosingTags && currentNode)
     {
@@ -2408,6 +2412,7 @@
       StructTreeView::ref()->slotReparse(this, baseNode , qConfig.expandLevel);
     }
     reparseEnabled = true;
+   delayedTextChangedEnabled = true;	
 }
 
 /** Returns list of values for attribute */
--- branches/KDE/3.5/kdewebdev/quanta/src/document.h #514629:514630
@@ -310,6 +310,7 @@
   bool hintRequested;
   bool reparseEnabled;
   bool repaintEnabled;
+  bool delayedTextChangedEnabled;
   /** True if the document is dirty (has been modified outside). */
   bool m_dirty;
   QString m_md5sum;
Comment 7 Mathieu Jobin 2006-04-04 05:07:39 UTC
quanta still crash quite often while editing....

somekool@krypton ~/Sites/projeckt001 $ quanta --version
Qt: 3.3.4
KDE: 3.5.2
Quanta: 3.5.2


Using host libthread_db library "/lib/libthread_db.so.1".
`system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
[Thread debugging using libthread_db enabled]
[New Thread 16384 (LWP 10563)]
[KCrash handler]
#6  0x0817f90f in AreaStruct (this=0xbf8786f0, area=@0x1c) at tag.h:59
#7  0x0817f752 in Tag::area (this=0x0) at tag.h:113
#8  0x0817df03 in SAParser::slotParseNodeInDetail (this=0x84f7798)
    at saparser.cpp:902
#9  0x0817e5f1 in SAParser::qt_invoke (this=0x84f7798, _id=4, _o=0xbf878800)
    at saparser.moc:115
#10 0xb64926ac in QObject::activate_signal (this=0x84d3da8, clist=0x84fc4c0, 
    o=0xbf878800) at qobject.cpp:2355
#11 0xb6492504 in QObject::activate_signal (this=0x84d3da8, signal=2)
    at qobject.cpp:2324
#12 0xb6877c64 in QTimer::timeout (this=0x84d3da8) at moc_qtimer.cpp:82
#13 0xb64be09e in QTimer::event (this=0x84d3da8, e=0xbf878b20)
    at qtimer.cpp:219
#14 0xb6422235 in QApplication::internalNotify (this=0x83a9cf8, 
    receiver=0x84d3da8, e=0xbf878b20) at qapplication.cpp:2635
#15 0xb6421478 in QApplication::notify (this=0x83a9cf8, receiver=0x84d3da8, 
    e=0xbf878b20) at qapplication.cpp:2358
#16 0xb6c6adaf in KApplication::notify (this=0x83a9cf8, receiver=0x84d3da8, 
    event=0xbf878b20) at kapplication.cpp:550
#17 0xb7e795e9 in QApplication::sendEvent (receiver=0x1c, event=0xbf8786f0)
    at qapplication.h:491
#18 0xb640d75f in QEventLoop::activateTimers (this=0x847f000)
    at qeventloop_unix.cpp:556
#19 0xb63bdd5d in QEventLoop::processEvents (this=0x847f000, flags=4)
    at qeventloop_x11.cpp:389
#20 0xb64397d9 in QEventLoop::enterLoop (this=0x847f000) at qeventloop.cpp:198
#21 0xb64396f2 in QEventLoop::exec (this=0x847f000) at qeventloop.cpp:145
#22 0xb64223d7 in QApplication::exec (this=0x83a9cf8) at qapplication.cpp:2758
#23 0x080b465d in main (argc=1, argv=0xbf878f14) at main.cpp:212

Comment 8 András Manţia 2006-04-04 12:25:42 UTC
Strange that this still happens. Can you give a testcase for Quanta 
3.5.2?
Comment 9 Mathieu Jobin 2006-04-04 13:50:04 UTC
i'd say its hard to reproduce, the file is a big HTML file with few <%= %> tags and the base HTML has been created with dreamweaver, so its all messy.
Comment 10 Mathieu Jobin 2006-04-05 04:37:19 UTC
just play by fixing the indent of an ugly HTML file,... it should make Quanta crash pretty easily.

Comment 11 Mathieu Jobin 2006-04-28 05:15:17 UTC
Quanta did not crash in a long time, but it just did.

running latest from KDE 3.5.2

I was replacing an h2 by an h1 in a .rhtml file.

here is the backtrace. I hope it can be helpful.

Using host libthread_db library "/lib/libthread_db.so.1".
`system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
[Thread debugging using libthread_db enabled]
[New Thread 16384 (LWP 9016)]
[KCrash handler]
#6  0x081768ff in Parser::rebuild (this=0x8594cb8, w=0x95578e0)
    at parser.cpp:1122
#7  0x080c0746 in Document::slotDelayedTextChanged (this=0x95578e0)
    at document.cpp:2320
#8  0x080c4e5d in Document::qt_invoke (this=0x95578e0, _id=50, _o=0xbfd3df40)
    at document.moc:249
#9  0xb64566ac in QObject::activate_signal (this=0x8595ff8, clist=0x95279f0, 
    o=0xbfd3df40) at qobject.cpp:2355
#10 0xb6839b47 in QSignal::signal (this=0x8595ff8, t0=@0x8596020)
    at moc_qsignal.cpp:100
#11 0xb647856b in QSignal::activate (this=0x8595ff8) at qsignal.cpp:212
#12 0xb6482478 in QSingleShotTimer::event (this=0x8595fd0) at qtimer.cpp:286
#13 0xb63e6235 in QApplication::internalNotify (this=0x83a9cf8, 
    receiver=0x8595fd0, e=0xbfd3e2a0) at qapplication.cpp:2635
#14 0xb63e5478 in QApplication::notify (this=0x83a9cf8, receiver=0x8595fd0, 
    e=0xbfd3e2a0) at qapplication.cpp:2358
#15 0xb6c2eddf in KApplication::notify (this=0x83a9cf8, receiver=0x8595fd0, 
    event=0xbfd3e2a0) at kapplication.cpp:550
#16 0xb7e3f5e9 in QApplication::sendEvent (receiver=0x0, event=0x955bcd0)
    at qapplication.h:491
#17 0xb63d175f in QEventLoop::activateTimers (this=0x847fd98)
    at qeventloop_unix.cpp:556
#18 0xb6381d5d in QEventLoop::processEvents (this=0x847fd98, flags=4)
    at qeventloop_x11.cpp:389
#19 0xb63fd7d9 in QEventLoop::enterLoop (this=0x847fd98) at qeventloop.cpp:198
#20 0xb63fd6f2 in QEventLoop::exec (this=0x847fd98) at qeventloop.cpp:145
#21 0xb63e63d7 in QApplication::exec (this=0x83a9cf8) at qapplication.cpp:2758
#22 0x080b465d in main (argc=1, argv=0xbfd3e694) at main.cpp:212
Comment 12 Mathieu Jobin 2006-04-28 10:44:36 UTC
quanta crash again while editing HTML files.
I think it happens more often when I replace a selection.
I really often select some text and type to replace it. not just type something.
I think it happens in this situation.

let me submit you another Back Trace ;)

by the way, apart a few bugs, Quanta is an outstanding application. I use it all the time, its excellent.


Using host libthread_db library "/lib/libthread_db.so.1".
`system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
[Thread debugging using libthread_db enabled]
[New Thread 16384 (LWP 9948)]
[KCrash handler]
#6  0x0817f90f in AreaStruct (this=0xbfa8cec0, area=@0x1c) at tag.h:59
#7  0x0817f752 in Tag::area (this=0x0) at tag.h:113
#8  0x0817df03 in SAParser::slotParseNodeInDetail (this=0x85a7368)
    at saparser.cpp:902
#9  0x0817e5f1 in SAParser::qt_invoke (this=0x85a7368, _id=4, _o=0xbfa8cfd0)
    at saparser.moc:115
#10 0xb64a66ac in QObject::activate_signal (this=0x84fe8f8, clist=0x84f7178, 
    o=0xbfa8cfd0) at qobject.cpp:2355
#11 0xb64a6504 in QObject::activate_signal (this=0x84fe8f8, signal=2)
    at qobject.cpp:2324
#12 0xb688bc64 in QTimer::timeout (this=0x84fe8f8) at moc_qtimer.cpp:82
#13 0xb64d209e in QTimer::event (this=0x84fe8f8, e=0xbfa8d2f0)
    at qtimer.cpp:219
#14 0xb6436235 in QApplication::internalNotify (this=0x83a9cf8, 
    receiver=0x84fe8f8, e=0xbfa8d2f0) at qapplication.cpp:2635
#15 0xb6435478 in QApplication::notify (this=0x83a9cf8, receiver=0x84fe8f8, 
    e=0xbfa8d2f0) at qapplication.cpp:2358
#16 0xb6c7eddf in KApplication::notify (this=0x83a9cf8, receiver=0x84fe8f8, 
    event=0xbfa8d2f0) at kapplication.cpp:550
#17 0xb7e8f5e9 in QApplication::sendEvent (receiver=0x1c, event=0xbfa8cec0)
    at qapplication.h:491
#18 0xb642175f in QEventLoop::activateTimers (this=0x847fcf0)
    at qeventloop_unix.cpp:556
#19 0xb63d1d5d in QEventLoop::processEvents (this=0x847fcf0, flags=4)
    at qeventloop_x11.cpp:389
#20 0xb644d7d9 in QEventLoop::enterLoop (this=0x847fcf0) at qeventloop.cpp:198
#21 0xb644d6f2 in QEventLoop::exec (this=0x847fcf0) at qeventloop.cpp:145
#22 0xb64363d7 in QApplication::exec (this=0x83a9cf8) at qapplication.cpp:2758
#23 0x080b465d in main (argc=1, argv=0xbfa8d6e4) at main.cpp:212


Comment 13 András Manţia 2006-05-19 14:45:18 UTC
But if I don't know how to reproduce this will be not fixed...
Comment 14 András Manţia 2006-09-02 14:53:37 UTC
Unfortunately without a way to reproduce it, the chances to solve are 
close to 0.
Comment 15 András Manţia 2007-10-03 13:11:26 UTC
Cannot reproduce...