[Interest] I can't find QQuickStyle class

Hi guys… as the doc say (http://doc.qt.io/qt-5/qtquickcontrols2-styles.html), I can do something like this: #include #include #include int main(int argc, char *argv[]) { QGuiApplication ::setAttribute(Qt

Re: [Interest] What don't you like about Qt?

I was hired on to my new job because of my Qt4 experience, and my first QML related task was to convert a list view to a editable tree view with drag and drop. It was not a good experience. The TreeView control provided in Qt was full of bugs and I ended up pulling its private C++ model adaptor

Re: [Interest] QObject::dumpObjectTree() & const

On segunda-feira, 3 de outubro de 2016 14:31:06 CEST Giuseppe D'Angelo wrote: > Il 03/10/2016 13:18, Thiago Macieira ha scritto: > > Because they don't need to be. You can call any function from the > > debugger, > > regardless of actual constness of the object. > > > > They were designed to be

Re: [Interest] Qt3D not deleting buffers

Hi Sean, thanks for your fast reply. I dont know what is a JIRA. Anyway, I have prepared a very simple example where I can reproduce one of the problems I was explaining in my previous message. I go through the code: The main creates the QApplication, a SceneManager and a timer to call

Re: [Interest] [Qt3D] Wrong value passed to the shader when using a QMatrix value

I found the issue : https://bugreports.qt.io/browse/QTBUG-56337 I think that is the kind of bug I can fix, I suppose that I have to follow guidelines from http://wiki.qt.io/Portal:Contributors ? 2016-10-03 15:06 GMT+02:00 Xavier Bigand : > Hi, > > I am trying to pass

[Interest] [Qt3D] Wrong value passed to the shader when using a QMatrix value

Hi, I am trying to pass a QMatrix object to my shader, but instead of receiving an identity matrix all field are null. Here is my parameter declaration : mTexCoordTransform(new Qt3DRender::QParameter(QStringLiteral("u_texCoordTransform"), QMatrix())) And here my uniform declaration in the

Re: [Interest] QObject::dumpObjectTree() & const

Il 03/10/2016 13:18, Thiago Macieira ha scritto: > Because they don't need to be. You can call any function from the debugger, > regardless of actual constness of the object. > > They were designed to be called from the debugger itself, never by your code. Come on, that's not a justification

[Interest] Qt high-DPI configuration patches for testing

Hi all, Lately I’ve spent some time on high-DPI configuration options for Qt. This is a somewhat complex area, with many different setups in the field. I’d like to expose the patches to a wider audience to make sure we implement something that works as well as possible. Please feel free to

Re: [Interest] Qt3D not deleting buffers

Hi, could you file a JIRA please ideally along with a small test case that reproduces the issue. Many thanks! Sean On Monday 03 October 2016 13:40:36 Juan José Casafranca wrote: > Hello! > > I'm posting to act some questions about Qt3D. > > I'm trying to show some information for my

[Interest] Qt3D not deleting buffers

Hello! I'm posting to act some questions about Qt3D. I'm trying to show some information for my program. This information consist on a regular grid (I'm painting the lines that define the voxels), a distance map (in each voxel vertex I store a distance value) and the sources for the distance

Re: [Interest] QObject::dumpObjectTree() & const

On domingo, 2 de outubro de 2016 20:05:49 MDT Andy wrote: > Why aren't QObject::dumpObjectTree() and QObject::dumpObjectInfo() const? > Surely they shouldn't be modifying the state of the object. Because they don't need to be. You can call any function from the debugger, regardless of actual

Re: [Interest] What don't you like about Qt?

It's ironic in a way that every major graphical toolkit (and with many large software projects in general) that I've worked with over decades now, the attitude has commonly seemed to have been that "new" is better than "stable". The end result is a product full of both older and newer unstable

Re: [Interest] [OS X] 10.12 SDK breaks qt5-qtconnectivity build (on 10.11)

On Monday October 03 2016 09:21:28 Mike Krus wrote: > went for a more all encompassing (or brutal) approach here: > https://codereview.qt-project.org/#/c/172612/ > > Re: testing with older SDKs: Samuel's conditional check works on 10.9 with the

Re: [Interest] [Qt3D] Best way to do operations before rendering

Hi, I think that I didn't explain correctly how I use the uber shader, it's a shader that will change with properties of the material. Our shader is rebuild when lighting, engine settings,... changed so it might be rebuild at anytime for already existing objects. To avoid to regenerate the shader

Re: [Interest] [OS X] 10.12 SDK breaks qt5-qtconnectivity build (on 10.11)

HI went for a more all encompassing (or brutal) approach here: https://codereview.qt-project.org/#/c/172612/ Mike > On 2 Oct 2016, at 21:51, René J.V. Bertin wrote: > > On Sunday October 02 2016 21:26:13 Samuel Gaist

Re: [Interest] [Qt3D] Best way to do operations before rendering

Hi, a QFrameAction is for code that should be executed once per frame not a one time thing. It's mainly useful for prototyping as it can be quite expensive due the backend/frontend synchronisation it entails. To have some operation completed before rendering a part of your scene graph, just

Re: [Interest] What don't you like about Qt?

1. New features (quantity) are priorized over bug fixing (quality). Suggestions are almost sensless. I reduced writing bug reports and totally gave up writing suggestions due to this. 2. Widgets have too low priority. In general new fancy features are priorized above bread-and-butter features