General and Desktop

Thank you guys for your feedback. Yes, I think the problem is probably in the Test function. I know that also Qt is switching to CMake, but I wonder if there is a temporary solution to fix this problem in some other way

@isar yes i also tried that but thought it was a bit overkill. Apparently update() is not enough always. My test was actually made in Designer and i just use textChange signal to see where i typed and it was easy to get same effect that setting focus back to Group after i hide widget1 would still focus the hidden line edit and type to it. So for me, it smells as a bug as hidden widgets should not get keyboard focus. And the GroupBox clearly set focus to a hidden lineedit. :)

As you can see there is a difference. I would guess the content-length:0 comes from the empty QByteArray() you pass to the sendCustomRequest() function. Why the server behaves differently is out of our scope (and Qt's scope)

@Komerad Since it's a hobby project, may I suggest you look at https://github.com/. This would allow you for free to maintain all the versions of your code you choose to deposit up on the ethercloud. Apart from giving you "free backup" of every version of your work, it also allows you to do things like compare new versions against old versions when you're trying to figure consequences of changes in your code. It's all integrated from within Qt Creator via Git/GitHub. If you are, say, Windows a little tutorial is https://cpb-us-w2.wpmucdn.com/u.osu.edu/dist/7/11881/files/2018/02/QtCreatorGitTutorial-2ahr2m0.pdf, or there are plenty of others around. It takes a few minutes to set up, but once you have done so you will want to use this always. At least then if for whatever reason you "lose" stuff again, like you're worried about and don't know why, you won't actually lose any work.

Hi @mrjj , I try to understand on QImage. for now I have implemented it on QPicture and I don't know if you can capture the color type from QPicture functions. can you? yes it is true the complexity of understanding what type of design it is, I thought there is an easy way to capture the type of design. maybe if anyone knows how to do it. would be very helpful. thank you very much. I await other solutions

If this is what you need: [image: 1d71d407-19b9-499f-bd1a-1d6b87b29694.png] [image: f798902c-b218-4267-be06-4541b81a9f92.png]

@SGaist Great! That is what I need. Thank you so much.

@VRonin You are a god. Finally I have finished the implementation thanks to your second option. I have tried (all possible combinations) and it works. I didn't know about Qt::QueuedConnection and the option to concatenate signals to the Qt event queue. This is going to be very useful to me. You have saved me and it is not the first time, I have seen your answers in other threads and they have helped me too. Thanks from Spain and sorry if my English is bad.

Yup, looks like you've spotted it -- in particular, the specular term is materialSpecularColor * lightStrength * pow(cosAlpha, 10) That's the part that will contribute to the shininess of the material. And, materialSpecularColor comes from highp vec3 materialSpecularColor = lightColor.rgb; which is to say that the materialSpecularColor in that shader is not really a property of the material since the value is just from the lighting. Not really the clearest name if you are trying to pick it apart. This type of look is traditionally used for a plastic type material which has white highlights for white lights regardless of the color of the material. If you want something a bit more metallic looking, you can multiply the lightcolor.rgb by some actual material color. This lighting model is a bit of an old school aphysical hack, so you may want to look into to more modern physically based lighting models if you are looking for more realism. But the classic specular + diffuse model is way simpler to understand even if it isn't super accurate to the real world.

And The Qt version you are using in your Kit is also 5.14.2 ? By the way, do you have a minimal compilable examine that shows that behaviour ?

@SGaist On "Ok" button I have hide(); auto res = mSessionProgress->exec(); show(); On "Cancel" of Progress dialog I just have: connect(ui->btnCancel, &QPushButton::clicked, this, ={ accept(); });

I have a QListWidget* in ui->barra_scroll and I feel very smooth with this... QScrollBar *qsb = ui->barra_scroll->verticalScrollBar(); qsb->setSingleStep(5);

Hi, Once you have Xcode installed, you can use macports or homebrew to install additional tools (or more recent versions of bash ;-)) Then you pretty much get the same feeling as on Linux. You have to change Ctrl for Cmd when you are outside of a terminal and then you should feel at home.

@IceMichael Hi. I have not seen any option for fakevim saying "use normal selection" so i dont think so. [image: 4jsLf4.png]

but I would be interested in this answer also :) As I remember, we did some project where we do changes for FTDI latency parameter: #ifdef Q_OS_WIN // Возвращает указатель реестра для устройства с заданным именем, // если оно является последовательным портом FTDI, либо 0. HKEY openFtdiDeviceParameters(const QString &portName) { const wchar_t *keypath = L"SYSTEM\\CurrentControlSet\\Enum\\FTDIBUS"; HKEY key; if (RegOpenKeyEx(HKEY_LOCAL_MACHINE, keypath, 0, KEY_READ, &key)) { // Порты FTDI отсутствуют. return 0; } // Просматриваем все порты FTDI. int idx = 0; HKEY readOnlyKey = 0; HKEY paramKey = 0; while (true) { if (paramKey) { RegCloseKey(paramKey); paramKey = 0; } DWORD subkeySize = 140; wchar_t subKey[140] = {0}; // Последовательный порт не является устройством FTDI. if (RegEnumKeyEx(key, idx++, subKey, &subkeySize, 0, 0, 0, 0)) { break; } QString paramPath(QString(QLatin1String("%1\\0000\\Device Parameters")).arg(QString::fromWCharArray(subKey))); if (RegOpenKeyEx(key, paramPath.toStdWString().c_str(), 0, KEY_READ, &readOnlyKey)) { continue; } DWORD regPortNameSize = 20; wchar_t regPortName[20] = {0}; DWORD type = 0; LONG result = RegQueryValueEx(readOnlyKey, L"PortName", 0, &type, (LPBYTE) &regPortName, &regPortNameSize); if (ERROR_SUCCESS == result) { QString name = QString::fromWCharArray(regPortName); if (portName == name) { RegOpenKeyEx(readOnlyKey, 0, 0, KEY_READ | KEY_SET_VALUE, &paramKey); } } RegCloseKey(readOnlyKey); if (paramKey) { break; } } RegCloseKey(key); return paramKey; } #endif and then: #ifdef Q_OS_WIN // Если выбранный порт является FTDI, устанавливаем его latency в 1. DWORD latency = 0; HKEY latencyKey = openFtdiDeviceParameters(name); DWORD dwType = 0; DWORD dwSize = sizeof(DWORD); bool foundFtdiDevice; if (latencyKey && !RegQueryValueEx(latencyKey, L"LatencyTimer", 0, &dwType, (LPBYTE) &latency, &dwSize) && dwType == REG_DWORD) { foundFtdiDevice = true; } else { foundFtdiDevice = false; } if (foundFtdiDevice) { if (latency != 1) { DWORD dwLatency = 1; RegSetValueEx(latencyKey, L"LatencyTimer", 0, REG_DWORD, (LPBYTE) &dwLatency, sizeof(DWORD)); RegCloseKey(latencyKey); } } #endif Here we did set the latency to 1 msec. PS: Sorry, all comments in russian (because it is copy-paste).

@Mihan said in Run a command with different result between in QProcess and in terminal: I think it should be added in wiki. You're more than welcomed to do that!

I've made the peace of code showing this behaviour a little simpler: import sys from PyQt5 import QtCore, QtGui, QtWidgets # Maybe you need to play with these constants PIXMAP_NUM = 1000 EVENTLOOP_RUN_NUM = 100 class Signals(QtCore.QObject): pixmaps = QtCore.pyqtSignal(list) class Worker(QtCore.QRunnable): def __init__(self): super().__init__() self.signals = Signals() @QtCore.pyqtSlot() def run(self): '''Make QPixmap object with different colours, add them to a list and emit the list''' c = QtGui.QColor colors = [c('red'), c('green'), c('blue'), c('cyan'), c('magenta'), c('yellow')] i = 0 pms = [] for _ in range(PIXMAP_NUM): pixmap = QtGui.QPixmap(200, 200) pixmap.fill(colors[i]) i += 1 if i == len(colors): i = 0 pms.append(pixmap) self.signals.pixmaps.emit(pms) class MyMainWindow(QtWidgets.QMainWindow): def __init__(self): super().__init__() self.setMinimumSize(QtCore.QSize(800, 600)) centralWidget = QtWidgets.QWidget() self.setCentralWidget(centralWidget) scrollArea = QtWidgets.QScrollArea() scrollArea.setWidgetResizable(True) imagesWidget = QtWidgets.QWidget() scrollArea.setWidget(imagesWidget) startButton = QtWidgets.QPushButton('Start') startButton.clicked.connect(self.run_worker) verticalLayout = QtWidgets.QVBoxLayout() verticalLayout.addWidget(scrollArea) verticalLayout.addWidget(startButton) centralWidget.setLayout(verticalLayout) self.imagesLayout = QtWidgets.QVBoxLayout() imagesWidget.setLayout(self.imagesLayout) def run_worker(self): threadpool = QtCore.QThreadPool.globalInstance() w = Worker() w.signals.pixmaps.connect(self.update_gui) threadpool.start(w) def update_gui(self, pixmaps): for i, img in enumerate(pixmaps): label = QtWidgets.QLabel() # The next line doesn't change the behaviour label.setMinimumSize(QtCore.QSize(200, 200)) label.setPixmap(img) self.imagesLayout.addWidget(label) # The next line doesn't change the behaviour self.updateGeometry() if i % EVENTLOOP_RUN_NUM == 0: QtCore.QCoreApplication.processEvents() if __name__ == '__main__': app = QtWidgets.QApplication(sys.argv) mw = MyMainWindow() mw.show() app.exec() This code reproduce the glitches on both the machines. I've noticed one thing. If we do not set the minimum size, label.setMinimumSize(QtCore.QSize(200, 200)) labels shrink. But if we do, they overlap (so, indeed, the minimum size is respected but imagesWidget is not big enough - cannot change its own size fast enough??? - to set the labels without overlapping???)

@jsulm Thanks so much. Solved.