@kshegunov Thanks. I can reproduce the same behavior with a single thread.

#include <QtCore> struct Object: QObject{ Q_OBJECT public slots: void foo(){ qDebug() << __LINE__; while(entered) qApp->processEvents(); qDebug() << __LINE__; QEventLoop loop; QTimer::singleShot(1000, &loop, SLOT(quit())); entered = true; loop.exec(); entered = false; qDebug() << __LINE__; } private: bool entered = false; }; int main(int argc, char *argv[]) { QCoreApplication a(argc, argv); Object w; QTimer::singleShot(0, &w, SLOT(foo())); QTimer::singleShot(0, &w, SLOT(foo())); QTimer::singleShot(3000, &a, SLOT(quit())); a.exec(); qDebug() << __LINE__; } #include "main.moc"

In fact, it has nothing to do with multi-threads. I have never studied Qt source code. It seems exec will not return before all slots finish. So, the code actually creates a dead lock.

Well, I have a custom shortcut shC applied to Action/Menu which is global (in every window), but I can not register that into window A, or window B. I mean, I am on OS X, where there is only one menubar that is shared across the multiple windows, so I have to dispatch them, right?

I mean, for example, Qt Creator has a shortcut "Run" which is invokable from many windows. But I can not make two menubars and/or having two actions, right?

Anyway solved it. I needed to call grabKeyboard function from the MainWindow class, so this->grabKeyboard() is the line I needed to add when key_0 button is pressed so that MainWindow doesn't lose the keyboard focus, and then when released I needed to add the line this->releaseKeyboard() to resume normal behaviour, that is, other widgets can have the keyboard focus.

@Walux said:

Probably not as good , so i'll use yours .

just for clarification:
Actually this is exactly the same approach, just a step later.
(After the graphics view translated and forwarded it's received events to it's scene).

@mrjj

It's alright , i figured it out ;)

Huge thanks for your help :)

Topic->setAsClosed(true);

I tried, but it doesn't work.

Hi and welcome to devnet,

What backtrace do you get from the crash ?

Sounds pretty crazy...

@SGaist
When I finish up my code the obvious run and if it crashes test will tell for sure.
Unfortunately, the documentation is very cryptic of such issues ... :)

i guess this issue will be solved along with your other issue.

Here's the solution. You have to do it in Javascript. Ensure jQuery is loaded in your web pages (even the IFRAME/FRAME ones) and then apply this code:

<script type="text/javascript"> $(document).ready(function(){ // Disable backspace key except in fields. $(document).keydown(function(e) { var elid = $(document.activeElement).is('INPUT, TEXTAREA') ; if (e.keyCode === 8 && !elid) { if(e.ctrlKey) { window.history.back(); } else { // disable backspace e.preventDefault(); return false; } } }); }); </script>

SOURCE: http://stackoverflow.com/a/17278620/105539

@jsulm I have declared a pointer to my class to have access at the instance and it works! I found this methode not really good with static member but for the moment I have juste this solution.

Hi,

QWidget::mousePressEvent is probably what you are looking for

@SGaist
I have tried to add a focusOut event in my loop but it doesn't detect it ...

if event.type() == event.FocusOut: print "Widget focusOut"

Hi,

What you do is to send your event to the event dispatcher for it to dispatch said event to the correct object but at the same time you're setting your filter between the dispatcher and the rest of your application. Thus you are seeing the "impossible"

Hope it helps

Hi,

You are missing a return False in your filterEvent function see the C++ documentation for more details

Pushing again.

I had an old mouse that gave irratic scroll wheel signals. It still worked but I did notice it jumpy when trying to scoll windows on occasion (all windows, not just Qt). I don't know what you are using but I would make sure the problem is not hardware.

@tomma

Thanks a lot! That was great input.
While reading about uinput the idea of writing to the event itself instead crossed me.
I implemented a function that generates the various (basic) events of a touch screen and writing it directly to event0.
It works perfect so far.

You could subclass QGraphicItems and implement the behavior only once there.

Another thing to consider is using tooltip or asking the user to click on the QGraphicItem to trigger the pop-up instead of hovering. I haven't seem much software that open pop-up without the user asking for it, I would be annoyed by pop-up openning if it's not clear that my action opened it.

That doesn't really help. I know the difference between the 2 devices.
The issue seems to be in the path and maybe the resulting timing Qt (QWS to be more precise) gets the XY position compared to the path over which it gets the click events. This seems not to travel the same way from DFB to QWS.
Still have no solution for this.

I found solution. i remplemented keyPressEvent from QTextEdit class

void MyTextEdit::keyPressEvent(QKeyEvent *e) { if(e->key() == 16777220){ //enter clicked return; } return QTextEdit::keyPressEvent(e); }

Yes that is correct. Every of the 8 custom gestures defines its own set of data in addition to the QGestureState, which is set by the according gesture recognizer.

I now implemented an abstract superclass for all these gestures with a method getGestureData(). This method basically wraps all the data in a QMap<QString,QVariant> to make a general gesture treatment possible. The custom event GestureEvent now takes gesture type, the Qt::GestureState and the mentioned gesture data map.
It was a little bit circumstantial, but now it works.

Thanks again,
Malte

@Wieland ok, then I'll secretly add you to the admin list one day, when I forget that you didn't want to be there. Or tell me when you feel more at home, and I'll add rights :)

The translations have been done at different times and might not have been updated when the original has been updated.

@SGaist I got it. Thanks for the reply. Cheers