Version: 3.0.90-CVS (using KDE 3.2.90 (CVS >= 20040117), compiled sources) Compiler: gcc version 3.3.3 20040125 (prerelease) (Debian) OS: Linux (i686) release 2.4.23-1-k7 The following is a strange one, but crashes kdevelop reproducable for me: 1. load any project (I have tested php, c++) with at least two tabs (using IDEAl mode) 2. click multiple times with the right mouse button on one of the tabs (left or right tab, active or non-active is not important). Do that fast and slowly move the mouse around (still over the tab) and move it over the close-button KDevelop than crashes for me with the following bt: [New Thread 16384 (LWP 2503)] 0x41a63bd8 in waitpid () from /lib/libpthread.so.0 #0 0x41a63bd8 in waitpid () from /lib/libpthread.so.0 #1 0x41004e98 in __JCR_LIST__ () from /opt/kde-new/lib/libkdecore.so.4 #2 0x41a62815 in __pthread_sighandler () from /lib/libpthread.so.0 #3 0x41bc64a8 in __libc_sigaction () from /lib/libc.so.6 #4 0x41316aab in QApplication::notify(QObject*, QEvent*) () from /opt/qt-new/lib/libqt-mt.so.3 #5 0x40eccdfc in KApplication::notify(QObject*, QEvent*) (this=0xbffffa40, receiver=0x8654a10, event=0xbffff5c0) at kapplication.cpp:506 #6 0x41318d75 in qt_dispatchEnterLeave(QWidget*, QWidget*) () from /opt/qt-new/lib/libqt-mt.so.3 #7 0x412ad4ea in QApplication::x11ProcessEvent(_XEvent*) () from /opt/qt-new/lib/libqt-mt.so.3 #8 0x412c35b1 in QEventLoop::processEvents(unsigned) () from /opt/qt-new/lib/libqt-mt.so.3 #9 0x41329658 in QEventLoop::enterLoop() () from /opt/qt-new/lib/libqt-mt.so.3 #10 0x41329508 in QEventLoop::exec() () from /opt/qt-new/lib/libqt-mt.so.3 #11 0x413176f1 in QApplication::exec() () from /opt/qt-new/lib/libqt-mt.so.3 #12 0x08066b9b in main (argc=1, argv=0xbffffbb4) at main.cpp:128 I am afraid it's not too useful--perhaps a Qt bug? Qt unfortunately is compiled without debug info. Testing the same thing with konqueror does not crash konqueror. While trying to reproduce that I sometimes only had to click 2-3 times, sometimes up to 20 times, but every time kdevelop crashed. Also I am quite sure that I have seen for a very short time a context menue showing up exactly when the crash happened (looking like one of these you get in konqueror). Strange...
"Don't do that." Sorry, couldn't resist.. ;) I was able to reproduce it though. I even noticed a popup menu.
Canot reproduce.
Still around. Just managed it again.
Maybe comparing of Qt versions would help.I'm using current CVS version (>=20040808), Qt version 3.3.2. I can't reproduce this bug, but I'm not sure, if I understand the description of reproducing.
Just checked again and the bug seems to be gone. I use latest CVS, that is Qt: 3.3.2, KDE: 3.3.0
I can still do it. It seems harder to trigger now, because I had to be at it awhile, but it's still there. :(
Don't know why I never tried it in Quanta before, but the crash is there as well. (Tried Quanta 3.3.0.) So.. not ours. KMDI? KTabWidget? Reassigning to kdelibs.
I couldn't reproduce this with a recent cvs checkout after KDE 3.4 alpha was released. Please reopen this bug if you can still reproduce with kdevelop 3.1.90. Thanks.
*** Bug 109334 has been marked as a duplicate of this bug. ***
I've posted a new bug (incorrectly for akregator and already marked as a duplicate), but the bug is still there. A more consistant way to reproduce this bug is to have an application that uses tabs and that also has close buttons on each tab. Akregator has this as an option, as does kdevelop3 (tabbed pages mode, use close on hover: yes, delayed, show icons on document tabs). The easier and more consistant way to reproduce the bug is to hover the mouse over the icon on the tab, wait till it shows the greyed out close button (which is immediate), then right click and hold the button down. Wait till the greyed out button turns into a clickable close icon, then move the cursor off the button while holding down the right mouse button. I'm not the original reporter for this bug so I can't reopen it. This is all tested with recent software: KDE 3.4.0 kdevelop 3.2.0 quanta 3.4.0 akregator 1.0 Qt 3.3.3
*** Bug 111597 has been marked as a duplicate of this bug. ***