[Development] Qt 5.13 & 5.14 add device-independent pixels to device-dependent

Mark De Wit mark.dewit at iesve.com
Fri Mar 27 12:49:23 CET 2020


> From: Development <development-bounces at qt-project.org> On Behalf Of
> Morten Sørvig
> Sent: 18 February 2020 13:38
> To: Qt development mailing list <development at qt-project.org>
> Subject: Re: [Development] Qt 5.13 & 5.14 add device-independent pixels to
> device-dependent
> 
> > On 14 Feb 2020, at 21:46, Thiago Macieira <thiago.macieira at intel.com>
> wrote:
> >
> > On Friday, 14 February 2020 07:18:10 PST Thiago Macieira wrote:
> >> On Thursday, 13 February 2020 14:13:52 PST Christoph Cullmann wrote:
> >>> I had in Kate/Konsole and Co. broken repainting as soon as winId()
> >>> was called on any non-native widget, see:
> >>>
> >>> https://bugs.kde.org/show_bug.cgi?id=411965
> >>>
> >>> If this is unrelated, just ignore me ;=)
> >>
> >> It's probably very related. Let me see if I can modify one of our
> >> examples to cause this problem.
> >
> > Confirmed with examples/widgets/mainwindows/sdi.
> >
> > This shows *both* issues:
> > 1) when you launch ./sdi and use the menu, it shows up in the wrong
> > screen
> > 2) when you resize the window to expand, it produces drawing artifacts
> 
> Thanks Christoph and Thiago; I can reproduce both issues.

Our application is also blocked from Qt upgrades by the winId() issue, but in our case it appears to be linked to our use of QDockWidget and QTabWidgets.

We've created https://bugreports.qt.io/browse/QTBUG-83131 with a sample to reproduce.  We even tried delaying our call to winId() by 5 seconds to ensure that all the windows have been created and drawn, but corruption still happens...

Mark


More information about the Development mailing list