Hi,
I have confirmed that QGraphicsView::viewportEvent() returns true for
all QGestureEvents as long as the QGraphicsView is enabled (see lines
2770 to 2782 of gui/graphicsview/qgraphicsview.cpp from Qt 4.6.3).
>From my understand, I think a more correct implementation would be to
only return true if QGraphicsScene::event() returned true. That way it
truly "Returns true to indicate to the event system that the event has
been handled, and needs no further processing." Any thoughts on this
or a workaround?
Thank you,
Sean
On Tue, Jul 13, 2010 at 9:55 AM, Sean Hayes <[email protected]> wrote:
> Hi,
>
> Thank you for your help. It may be the case that the
> QGraphicsView::viewportEvent(pEvent) is called first and redirects the
> QGestureEvent to the QGraphicsObject. However, from my testing the
> QGraphicsView::viewportEvent(pEvent) always returns true for
> QGestureEvents, event when they were not used by a QGraphicsObject or
> a QGraphicsItem. Likewise, the QEvent::isAccepted() always returns
> true for QGestureEvents. I have tested this using the modified
> function that follows while performing gestures directly on the view
> (i.e., not on a QGraphicsObject or QGraphicsItem). I theorize that Qt
> my accept all gestures in the view if viewport()->grabGesture() is
> called for that gesture. Does anyone have any ideas of a work around
> that would allow me to only use a gesture in the viewport if it was
> not consumed by a QGraphicsObject?
>
> bool MyGraphicsView::viewportEvent(QEvent *pEvent) {
> bool used = QGraphicsView::viewportEvent(pEvent) &&
> pEvent->isAccepted();
> if(!used && pEvent->type() == QEvent::Gesture) {
> std::cout <<
> "Gesture Event Received from viewportEvent()" <<
> std::endl;
> pEvent->accept();
> return true;
> }
>
> return used;
> }
>
> Regards,
> Sean
>
> On Tue, Jul 13, 2010 at 9:11 AM, Marcus Tillmanns
> <[email protected]> wrote:
>> Hi,
>>
>> I would guess that it first has to go through
>> QGraphicsView::viewportEvent(pEvent); ( because it redirects the event to
>> the correct GraphicsObject ). I'd guess that
>> QGraphicsView::viewportEvent(pEvent); will return whether the event was
>> consumed or not.
>>
>> Cheers,
>> Marcus
>>
>>
>> On [DATE], "Sean Hayes" <[ADDRESS]> wrote:
>>
>>> I am using Qt 4.6.3 with Windows 7 with Multi-Touch hardware. I am
>>> creating a QGraphicsObject that receives QGestureEvents using the
>>> following function:
>>>
>>> bool MyGraphicsObject::sceneEvent(QEvent* pEvent) {
>>> switch (pEvent->type()) {
>>> case QEvent::TouchBegin: // Needed to get around potential Qt bug
>>> pEvent->accept();
>>> return true;
>>> case QEvent::Gesture:
>>> std::cout << "Gesture Event Received from sceneEvent()" << std::endl;
>>> pGestureEvent->accept();
>>> return true;
>>> }
>>>
>>> return QMapItem::sceneEvent(pEvent);
>>> }
>>>
>>> However, those events are still getting passed to viewport because my
>>> QGraphicsView is receiving them as well via:
>>>
>>> bool MyGraphicsView::viewportEvent(QEvent *pEvent) {
>>> if(pEvent->type() == QEvent::Gesture) {
>>> std::cout << "Gesture Event Received from
>>> viewportEvent()" << std::endl;
>>> pEvent->accept();
>>> return true;
>>> }
>>>
>>> return QGraphicsView::viewportEvent(pEvent);
>>> }
>>>
>>> It seems to me that if the QGraphicsItem receives a gesture, the
>>> gesture should not be passed to the viewpoint unless the QGraphicsItem
>>> passes the event to its super class. Is that correct? Is there a way
>>> for me to make this happen?
>>>
>>> Thank you for your help.
>>> _______________________________________________
>>> Qt4-preview-feedback mailing list
>>> [email protected]
>>> http://lists.trolltech.com/mailman/listinfo/qt4-preview-feedback
>>
>> Marcus Tillmanns
>> Manager Software Engineering | P&S Creative Enthusiasts
>>
>> Avid
>> Birkenstrasse 15
>> 28195 Bremen
>> Germany
>> [email protected]
>>
>> t +49 421 4337599
>>
>> We're Avid. Learn more at www.avid.com
>>
>> Avid Development GmbH; Paul-Heyse-Str. 29, D-80336 München
>> Amtsgericht München HRB 155339, USt. ID. Nr. DE 814232828
>> Geschäftsführer: Paige Parisi, Jochen Pielage, Kenneth A. Sexton
>>
>>
>
_______________________________________________
Qt4-preview-feedback mailing list
[email protected]
http://lists.trolltech.com/mailman/listinfo/qt4-preview-feedback
