Nice one ! Thanks :-)

Hi and welcome to devnet, Did you already follow this pinned thread about accessing the bug report system ? In between, I updated the bug report with your findings but don't hesitate to try again and add more context as needed.

I've added a task for that: https://qt-project.atlassian.net/browse/QTBUG-151640

Unlikely, the internal mechanism is different, and I furthermore cannot reproduce any crash with OP's sample. @Qoissi Please provide more detail - which version of Qt and which version of PyQt? Is the crash due to a Python exception or in native code? Can you provide a stack trace?

Hi, and welcome! @BRNany said in CMAKE_AUTORCC was removed from the default cmake project?: My question is: WHY ISN'T AUTORCC ENABLED BY DEFAULT? If I create a project using "CMake with Qt 5 compatibility", the AUTORCC is already enabled by default... @SGaist is correct. CMAKE_AUTORCC is discouraged now: https://qt-project.atlassian.net/browse/QTBUG-87643 https://qt-project.atlassian.net/browse/QTBUG-141364 explicity process each .qrc file using qt_add_resources(). The recommended approach in Qt 6 is to NOT have a *.qrc file at all. Instead, add your files directly to qt_add_resources().

Welcome to the forum! It would useful if you can expand on your use case and motivation for running multiple applications under X11 with no window manager at all, not even twm. A minimal reproducible setup would also be required for it to ever progress. It is quite possible that this is the result of a bug in the xcb platform plugin, but it is also quite an unusual setup (Qt can work on X11 without a WM, but it is more intended for kiosk like setups).

Indeed. I filed https://qt-project.atlassian.net/browse/QTBUG-151608 on the matter.

Hi and welcome to devnet, AFAICT, this one should be included. Usually the license part of the module documentation gives that information and if a module belongs to a more specific commercial product such as Qt for Automotive it is also stated in the documentation. Since you are a commercial customer don't hesitate to contact the Qt Company directly to clear your doubts. With that said, that module has been deprecated in favor of Qt Graphs that you might want to consider as well.

@Christian-Ehrlicher , I added below line in main.cpp but nothing changed in Debian 12. QApplication::setStyle(QStyleFactory::create("Fusion")); MainWindow w; Regards, Mucip:)

@nnpyro Try Qt Creator 21. There are fixes for this use case.

@ziller thank you very much. cd qtcreator SIGNING_IDENTITY=- ./scripts/build.py --src . --build ../build_qtcreator --qt-path ../install_qt --llvm-path ../libclang --no-sbom --no-dmg --no-zip This works. cd build_qtcreator cmake -G Ninja "-DCMAKE_PREFIX_PATH=../install_qt;../libclang" ../qtcreator cmake --build . --parallel cmake --install . --prefix ../install_qtcreator cmake --install . --prefix ../install_qtcreator --component Dependencies This terminates with errors: ERROR: codesign verification error: ERROR: "/Users/valentingorelov/Projects/qt/install_qtcreator/Qt Creator.app: code has no resources but signature indicates they must be present\nIn subcomponent: /Users/valentingorelov/Projects/qt/install_qtcreator/Qt Creator.app/Contents/Frameworks/QtQuick3DXr.framework\n" but the resulting app bundle works on my machine. I suppose the reason may be the documentation which has not been built. A small remark about what I am trying to achieve. I do not need a fully signed relocatable app bundle which can be installed on any machine. I just want to try to develop a small feature for Qt Creator. But this feature depends on Info.plist and will not work if I run Qt Creator from the build directory. If I understand correctly, the idea behind just cmake --install . was to achieve exactly this: non-relocatable app bundle for debug purposes which does not contain dependencies. And it works if I build Qt Creator against the Qt from Homebrew. But when Qt Creator is build against self-built Qt, this approach does not work and I need to build a full-fledged app bundle with all the dependencies. cmake --install . --prefix ../install_qtcreator --component Dependencies step takes a lot of time and it is not convenient for debugging.

Thank you! Very little useful information unfortunately. I get this upon initialization, but nothing else (even when updating the render target): [18] Creating QRhi with backend Vulkan for window 0x55beb55752a0 (wflags 0x1) Graphics API debug/validation layers: 0 Debug markers: 0 Timestamps: 0 Prefer software device: 0 Shader/pipeline cache collection: 1 [18] Using existing native Vulkan physical device 0x55beb59dcc80 device 0x55beb5acb7c0 graphics queue family index 0 [18] Initializing QRhi Vulkan backend 0x55beb559cef0 with flags 20 [18] Enabled instance extensions: [18] VK_KHR_get_physical_device_properties2 [18] VK_EXT_swapchain_colorspace [18] Using imported physical device 'NVIDIA GeForce RTX 5070 Laptop GPU' 610.228.256 (api 1.4.341 vendor 0x10DE device 0x2D18 type 2) [18] 280 device extensions available [18] Using imported device 0x55beb5acb7c0 [18] Using queue family index 0 and queue index 0 [18] Created QRhi 0x55bec060c310 for window 0x55beb55752a0 [18] Attempting to seed pipeline cache for QRhi 0x55bec060c310 from '/home/znurre/.cache/qtpipelinecache-x86_64-little_endian-lp64/qqpc_vulkan' [18] Created pipeline cache with initial data of 291394 bytes [18] rhi texture atlas dimensions: 2048x1024

@Jim-G said in Selection highlight is a thin blue bar at the left of the cell.: So why does setting alternatingRowColors cause the change in highlight? Qt tries to use the OS default widgets first. Alternating row colors is not supported by the OS directly. So, Qt will draw it from scratch and usually tries to mimic the OS look. Seems like it did not get updated, yet.

Hi, Can you provide a minimal compilable example that shows this behaviour ?