@aha_1980 : I just noticed something else: In the log file, there are lines saying "GDB 71101", presumably because it is using v7.11.1 (?!). But I have installed GDB v8.2 in both usr/bin and usr/local/bin. And I've set those locations in QtCreator's options pages for kits and for debuggers. There should be no way for QtCreator to find an older GDB version.

The GDB version in both usr/ folders above is 8.2, confirmed by running via terminal. And I did not find any older stray 7.11.1 versions that QtCreator could be picking up.

Another oddity. Is there any other options page where the GDB location needs to be set?

By build a debug version do you mean adding -DDEBUG -g to CFLAGS in my makefile?
Hooray! Adding the -DDEBUG -g flags to my makefile fixed it!
Thank you @aha_1980! I googled for this but nothing really came up, thanks!

@artudi54 Tested with gnome on arch and it does the same thing. I'd say this is a bug since d.show() never pulls the window back up.

I would file the bug with Qt.

To work around it I just made sure to call d.hide() before hiding the main window. Then after w.show() just call d.show() again.

@SGaist

And function update_vars() is:

151 static void 152 internal_function 153 update_vars (void) 154 { 155 __daylight = tz_rules[0].offset != tz_rules[1].offset; 156 __timezone = -tz_rules[0].offset; 157 __tzname[0] = (char *) tz_rules[0].name; 158 __tzname[1] = (char *) tz_rules[1].name; 159 160 /* Keep __tzname_cur_max up to date. */ 161 size_t len0 = strlen (__tzname[0]); 162 size_t len1 = strlen (__tzname[1]); 163 if (len0 > __tzname_cur_max) 164 __tzname_cur_max = len0; 165 if (len1 > __tzname_cur_max) 166 __tzname_cur_max = len1; 167 }

I created a SO question
and an issue

Hi,

I'd try with a custom QDialog and use the open method.

@jsulm We determined that the issue was javascript-related. The page was using javascript to perform the redirect, and iOS didn't like that.

Thanks for the help, though!

Please have a look at the bug report I just submitted on this issue: https://bugreports.qt.io/browse/QTBUG-61059

@seb21 I've now managed to fully integrate the QML maps into QWidget application but I also have a working Google Maps plugin with all map types supported.

I can share some code if you like?

I actually prefer the ESRI map implementation over Google Maps and am presently looking at the ArcGIS Runtime SDK stuff for Qt.

Thanks for your input jpnurmi!

Yes I'm using 5.8 and 5.9 beta is fixing this issue! Thanks for pointing it out! For some reason I didn't find this bug!

Best,
nx

This is not a known issue. Please report a bug and we'll take a look.

@SGaist said:

mkspecs/device/linux-rasp-pi2-g++

Shouldn't it maybe be mkspecs/device/linux-rasp-pi-g++ (without the 2)?
In his makefile there is this line:

INCPATH = -I../threadtest -I. -I/usr/local/qt5/mkspecs/devices/linux-rasp-pi-g++

@ghielmetti
Hi
Normally I do have the .h file in the project folder and point to that in the
Promote dialog.
Im not sure it will take a full path in the dialog but I never tried :)

Hi Guys,

Thanks, indeed this is due to the graphic card... I have 2 of them on his laptop and when I force the ati one, it solves the problem. However if I leave the built in Intel HD chip active as it is by default, it crashes.

Will look for some drive update for it, thanks !

@SGaist Hi, I'm running Windows 10, OpenGL 4.5.0

Hello!

I faced a similar problem, tried hacks with single QTimer and showEvent: timer was a solution, and showEvent -- wasn't. I also use Win10, but Qt5.5.1. Have you succeeded?

@Wieland said:

Qt.createQmlObject('import QtQuick 2.6; Rectangle {color: "red"; width: 100; height: 100; function nix(){ var arr=["function","two"]; console.log(arr)} Component.onCompleted:nix() }', mainWindow, "dynamicSnippet1");

Good to hear, must be solved then. Thanks!

Finally, I have fresh installed it with Offline Installer, and it's working as expected. It seems isolated, so maybe nothing.

Thank you for having tried ;)

Hi, because your application is also called QSettings, the #include <QSettings> gets confused and instead of including Qt's file, it ends up trying to compile your .exe file.

So either change to #include <qsettings.h> or rename your app to something else...

@Wilk said:

The reason that it did not work with QMainWindow or QWidget is that those classes does not implement fullScreen as full featured property - it can not be set with setFullScreen() slot.

I see, thanks for the precision.

@Wilk said:

Thanks alot

No problem, I hope your bug report will be helpful for other people.

@Sargamon
Once your problem is solved please use the Topic Tools button to mark as Solved.
Thank you!

@SGaist

Ok, thanks. I reported it (https://bugreports.qt.io/browse/QTCREATORBUG-15444)
Seems just cosmetic but nonetheless.

It's more of a limitation than a bug I think.

To OS, by default a window is a rectangle, no matter how you paint it, and stylesheets are just a way to paint it.
To cut out the corners you would need a mask. Something like:

auto frame = new QWidget(parent, Qt::Popup); frame->setStyleSheet("background-color: red; border: 1px solid green; border-radius: 6px;"); QPainterPath path; path.addRoundedRect(frame->rect(), 6, 6); frame->setMask(path.toFillPolygon().toPolygon()); frame->show();

The problem is that the mask is a bitmap (0-1), not an image with alpha, so you get a jagged edges, but for some radii it works pretty well.

@jsulm said:

You can check this by typing "groups" in a terminal.

it gave me "root"

@chrisadams No, there is no change at all with QML_NO_TOUCH_COMPRESSION enabled

Nevermind I found a solution:

QQmlApplicationEngine::setObjectOwnership(MY_OBJECT*, QQmlApplicationEngine::CppOwnership);

It's important to set this whenever returning a pointer in a Q_INVOKABLE function or else it might get garbage collected apparently.

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.

@koahnig
I think it really is a bug, since the same Qt-creator has always worked without this issue on my Ubuntu and Mint machines.
Thanks for the link, I check over.
Istvan

What version of Ubuntu are you running ?

Did you test with the version of Qt Creator from Ubuntu's repositories ?

I know, and I want to, but I don't think I can because I can't create a good compilable example that reproduces the error. If you simply create a dockWidget that has a listView and you try to scroll it is fine. Even though I tried to configure it the same way.
I am not sure what exactly in my app triggered this bug.

Hi,

Technically putting the Q_OBJECT macro without using any signals or slots does not really make sense. Anyway, you can create a templated QObject without using the Q_OBJECT macro.

The Qt Quarterly number 15 talks about that.

Hope it helps

I am having this exact same problem with Qt 5.6.0. After seeing this thread, it's apparent that this bug has been present for several years.