I was using a QPlainTextEdit. I had not noticed that there were several different type of controls for inputting text. QTextEdit does the same thing for me as QplainTextEdit. QLineEdit, however, does not change size on me. It also looks like it may be closer to what I actually need. I will change all of the QPlainTextEdit widgets to QLineEdit and see how that works.

Thanks for the reply! Yes, I agree this really seems to be a tvirt.vir issue, not Qt. That's why I wrote a standalone C program (outside Qt) to test the pipe communication. I’ve already tried: Sending the message DO#0#1 to \\.\pipe\\tubotest Triggering the EventoEnviar event after writing Still, nothing changes on the simulator side, and the input value shown in my DAQ console remains 0. I do have some documentation, and it says tvirt.vir listens for specific commands and events, but I’ll try to confirm exactly what sequence it expects. Thanks again. I’ll keep testing with the C version for now.

@JulsPower That does look more reasonable, i wasn't sure what the MS docs were telling you to do. Which implies you were in the Another possibility is that the issue has nothing to do with Qt or deployment but rather is a "bug" which only happens to show up in certain environments or circumstances. Such as a memory access issue. I had a quick glance at your WTSEnumerateSessionsExA() and it seemed OK to me, but you might want to try it standalone without any Qt stuff to make sure that works across environments. situation! The Qt and deployment changes simply showed or masked the problem. But you have still learnt what you should do to e.g. deploy to another machine, or be sure it works on yours outside of the Creator development environment which should be useful :)

Thank you so much for your support . it works now.

@aha_1980 said in How to switch off specific warnings in Qt Creator: @DiBosco Have you already seen https://stackoverflow.com/questions/22129383/removing-unused-parameters-warning-in-qtcreator ? For your case, it would be: QMAKE_CXXFLAGS_WARN_ON += -Wno-deprecated-copy Regards Ah, yes I had seen that. The English in the most upticked reply just made me go "What?". Just could not understand what they were saying, but I had tried: QMAKE_CXXFLAGS_WARN_ON += -Wdeprecated-copy Thinking why on earth would putting the warning on make it disappear, but out of desperation I gave it a go. Had I realised it needed -Wno-deprecated-copy, not -Wdeprecated-copy I'd've been fine hours ago. Many thanks for that, now doing just what I want. Also @JonB Thanks for your help too.

@J.Hilk said in how to set app-ico ?: shadow build because your not building inside your source file but all generated files are placed in a different folder As @J.Hilk said, I also think the issue is same, image is not in the shadow build folder @cuijg you can uncheck the shadow build checkbox and clean, build, run.

@Mizmas haha, looks funny :)

@J.Hilk thx for the offer... I already did changes - but it's not public yet give me some weeks to publish a first raw Qt 6.9 / CMake release then you can take a look if I missed something ;-)

My full blown test app now runs on the X86_64 simulator it looks awful its ui was not designed for phone use but all the bells gongs and whistles are there and information from my database is populating the widgets which is good. Time now to reflect on what I have learned and tidy up loose ends, thanks to those who helped on this journey.

@SimonSchroeder said in Using QList [] operator in const methods: Even if operator[](int) const returns a copy you have the exact same possible race condition in a multithreading environment. The only advantage is that it is fewer lines of code where this might occur. Multithreading seems an irrelevant point though because the competent programmer understands that accessing a shared resource needs a mutex, so if safe programming processes are observed then there is no race condition. re const vs non-const: totally agree and get it. It's the way things are...however, I do sometimes wish nothing was ever const and everything was mutable. :^)

It's all documented here and it works out of the box. I personally don't use -skip. I explicitly select the submodules I want to build. I always build w/o prefix into directories outside the source tree. E.g. like that: ../dev/configure -developer-build -xcb -nomake examples -make tests -submodules qtdeclarative ninja

@JKSH said in In win10 can't customization: Yes, the best way is via the import statement: import QtQuick.Controls.Basic -- this allows the Qt Quick Compiler to optimize your code. That optimization can't happen if you set the style at runtime. Thanks for this tip. I'll bear it in mind for later. I currently have my code running in both Qt 5.15 and 6.8 as I may or may not need to make the switch in the near future (out of my control unfortunately). The runtime selection is a pragmatic approach for me at the moment.

@hskoglund said in Qt6 and QT_ENABLE_HIGHDPI_SCALING: If you right-click on the .exe file, select Properties and in the Compability tab click the "Change high DPI settings" button, does any setting in that dialog box help the faulty scaling after QApplication is instantiated? Thanks for the suggestion but I think those settings effectively override the manifest (so startup only) and indeed I can change the setting on an exe so that it "supports" High DPI even when it doesn't and start it up and you can see scaling issues. Reverting it while it's running makes no difference. However, this is now moot as I decided to suck it up and change the legacy code to check for and support the system scaling. We have a fair number of these MFC apps and on some this will be a lot more work than others. This one is not too bad as the main client window is written using our Direct2D library which already supports such scaling. As expected Windows common controls support this out of the box and weirdly all our MFC dialogs also display scaled with the only apparent issue being the font text display appearing "thin" and somewhat faded. Not enough of a problem to care about.

I think it is necessary to port the Qt5 service library to Qt6. I haven't did it because the qt6 already built framework can't run on Windows 2016 machines and I have a lot of them. Qt6 requires Windows 10. So, up to now, I'm using Qt5.

Thanks for sharing the detailed steps — it looks like you're almost there. The error about the missing ABI usually means QtCreator isn’t detecting the correct architecture for deployment. Adding ``` ANDROID_ABIS = arm64-v8a Double-check that the Kit you're using is actually set up with the correct Qt for Android version and the appropriate architecture. Also, make sure the NDK path is correctly configured in the SDK settings — mismatches there can silently cause issues like this. Running on an emulator first is definitely a smart move. Hopefully with those tweaks, you'll be able to deploy without issues. Let us know how it goes!

@jsulm said in System tray crash in macOS 16: how is your question related to this thread? Forked now :)

fail logs:qt.core.plugin.factoryloader: Got keys from plugin meta data QList("windows") qt.core.plugin.factoryloader: checking directory path "C:/Temp/gui_hello/platforms" ... qt.core.library: "C:/Temp/gui_hello/_internal/PySide6/plugins/platforms/qwindows.dll" cannot load: Cannot load library C:\Temp\gui_hello_internal\PySide6\plugins\platforms\qwindows.dll: The specified procedure could not be found. qt.core.plugin.loader: QLibraryPrivate::loadPlugin failed on "C:/Temp/gui_hello/_internal/PySide6/plugins/platforms/qwindows.dll" : "Cannot load library C:\Temp\gui_hello\_internal\PySide6\plugins\platforms\qwindows.dll: The specified procedure could not be found."