assert in MultiFilterSortModel::invalidate() on first run

Tomaz Canabrava tcanabrava at kde.org
Sat Oct 10 15:34:31 PDT 2015


Taking a look at it now.
Em 10 de out de 2015 18:49, "Lubomir I. Ivanov" <neolit123 at gmail.com>
escreveu:

> steps:
> - open Subsurface
> - press CTRL+F
> - press CTRL+F again (i.e. close/clear the MultiFilter widget)
>
> ASSERT failure in QPersistentModelIndex::~QPersistentModelIndex:
> "persistent model indexes corrupted", file
> itemmodels\qabstractitemmodel.cpp, line 557
>
> Qt = 5.5.0.
>
> i can't be reproduced with the Win32 daily because it's not using the
> 'd' versions of Qt.
> some time ago i've noticed that Qt completely skips such asserts on
> Linux (and probably OSX), so this one is likely the 4-5th
> index-assert, only i can see for some reason...
>
> rough stack trace in reverse order:
> MainWindow::on_actionFilterTags_triggered()
> MultiFilter::closeFilter()
> MultiFilterSortModel::clearFilter();
> MultiFilterSortModel::myInvalidate();
> MultiFilterSortModel::invalidate() <----------
>
> i don't have a fix because i wouldn't touch this overly complicated
> and overly engineered Qt model-space-cosmos-universal-proxy-crazyness
> with a 10 foot pole. i mean, i probably can submit some sort of
> half-baked fix but it won't be the correct solution.
>
> no offense to Tomaz, of course...who is the only person who seems to
> juggle this Qt madness in a sane way.
>
> lubomir
> --
> _______________________________________________
> subsurface mailing list
> subsurface at subsurface-divelog.org
> http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20151010/211c9e66/attachment.html>


More information about the subsurface mailing list