> The program will not maintain the same position between window re-appearances.

If by position you mean window position, you can use Settings to store it:

https://doc.qt.io/qt-6/qml-qt-labs-settings-settings.html#details

> I'd also like to ask, is there any way to get rid of the "special"  class 
> function used only to invoke the other function? It really sounds like 
> something that can be improved.

If you mean the invokable that does a queued invoke of doStyleChange, you could 
probably shift it to QML instead:

    onClicked: Qt.callLater(function() { App.styleChange(text); })

So you could merge doStyleChange into styleChange, but you still need the 
queued call.

From: Interest <interest-boun...@qt-project.org> on behalf of randomslap--- via 
Interest <interest@qt-project.org>
Date: Wednesday, 15 June 2022 at 03:09
To: Interest <interest@qt-project.org>
Subject: [Interest] QML Runtime Style Change
Hello again

    IMPORTANT: This mail is an UPDATE to the "Change QML Application Style on 
Runtime" mails.

I should also note I run on GNOME Wayland sessions.

    I finally managed to create a stupidly small application which is able to 
change to a whole different style only by renewing the QML engines. I send an 
image that shows all the code used as well as the app running and a compressed 
file containing all relevant files.

-> What was the problem:
    Indeed there a was a need for a queued connection as it was suggested but 
that wasn't causing the warning at that time. It was rather the engine probably 
causing weird reactions. The solution was to simply change the engine to a 
pointer and delete and re-new every time I want to change the app's style. 
Also, the .close() function was causing a Wayland protocol fatal error but when 
I removed '()' it didn't complain. The engine.exit(0); might also not be 
necessary.

-> Important notes:
    The program will not maintain the same position between window 
re-appearances. Items that are not bound by C++ data won't also maintain their 
states between re-appearances either.

    I just hope I can still reach a point to implement my idea even in a 
compositor and solve the above problems. I also hope memory is not leaking but 
I did not catch any related processes on my system monitor ( maybe I missed 
something ).

    I'd also like to ask, is there any way to get rid of the "special"  class 
function used only to invoke the other function? It really sounds like 
something that can be improved.

Thank you for helping me. I'd appreciate any ideas to improve this "concept" if 
possible.
_______________________________________________
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest

Reply via email to