Re: Fixing krunner's threading

2010-03-03 Thread Thiago Macieira
Em Terça-feira 02 Março 2010, às 15:49:18, Thiago Macieira escreveu: > I'm actually raising the bar here. > > I'm turning on QT_FATAL_WARNINGS by default in Qt debug builds (starting > with Qt 4.7). After suggesting this to Qt developers, the overwhelming reaction was against it. So there will

Re: Fixing krunner's threading

2010-03-02 Thread Richard Moore
On Tue, Mar 2, 2010 at 2:49 PM, Thiago Macieira wrote: > I'm turning on QT_FATAL_WARNINGS by default in Qt debug builds (starting with > Qt 4.7). > > It's time people stopped ignoring Qt warnings, which indicate that there are > problems in your code. I can see these warnings coming out of plasma:

Re: Fixing krunner's threading

2010-03-02 Thread Aaron J. Seigo
On March 2, 2010, Thiago Macieira wrote: > are problems in your code. I can see these warnings coming out of plasma: > > Object::disconnect: Unexpected null parameter > QGraphicsScene::removeItem: cannot remove 0-item > > This from krunner: > QObject::killTimer: timers cannot be stopped from anot

Re: Fixing krunner's threading

2010-03-02 Thread Thiago Macieira
Em Segunda-feira 22 Fevereiro 2010, às 19:18:15, Andreas Hartmetz escreveu: > Now that many of the more prominent crashes in krunner have been fixed > I think it's time to attack something more "theoretic": runners run in > threads and often don't pay any attention to thread-safety of API they > us