QWidget::setTabOrder logging to stdout when dialogs are open

[sssoleileraaa]

STR

1. Click on "Print" or "Export" and see log message: QWidget::setTabOrder: 'first' and 'second' must be in the same window

STR 2

1. Open the client in Offline mode
2. Click on SIGNIN to open the login dialog and see log message: QWidget::setTabOrder: 'first' and 'second' must be in the same window

[ntoll]

@creviera ack.

[ntoll]

OK... I cannot reproduce this. Perhaps code in the client has changed in some way. However, I can reproduce #697 as reported by @kushal. In any case, I've looked into the cause of this error message and the only reference to it with any meaningful context is from 2009 (https://www.qtcentre.org/threads/18033-QWidget-setTabOrder-question) and an older version of Qt.

I'll see what I can salvage from the old conversation referenced above and check our code in the SourceList/SourceWidget classes. Something about setting the tab layout after calling setLayout.

[ntoll]

TL;DR: is this really worth the rabbit-hole exploration required to remove such warnings..? I suggest we mark as "won't fix" because the benefit will cost lots more investigation.

Further investigation reveals:

• We don't set any tab order related attributes relating to the SourceList or SourceWidget classes.
• We call setLayout very early in the life cycle of pretty much all our custom widgets.
• The source of the warning is this line of code in Qt: https://code.woboq.org/qt5/qtbase/src/widgets/kernel/qwidget.cpp.html#7015 Note how it's checking that the widgets share the same window().
• This window constant (https://doc.qt.io/qt-5/qwidget.html#window) returns the parent widget in the widget tree that is the "window" of the application.

I'm confused that window should be different for two SourceWidget widgets in the same SourceList.

My advice is to mark this is "won't fix" since it's only emitting to stdout (which the user won't see) and the logic surrounding the window feels like a can of worms.

[redshiftzero]

fair, I'll remove this ticket from the sprint and mark as low priority

[eloquence]

Could there be any relation with #352?

[sssoleileraaa]

I can no longer repro the original STRs but we saw the same behavior when switching between sources in the sourcelist, which I just confirmed #1261 fixes. Since #1261 has been merged, I'm closing this issue.

[gonzalo-bulnes]

I hadn't seen this issue! Yes, it does look like #1261 a lot. I understood well the ordering requirement @creviera, so if helpful I'm happy to walk you or someone else through it if it shows up again 🙂