[Port KF5] Why we get so many segfault...

Olivier Churlaud olivier at churlaud.com
Thu Oct 22 18:03:45 UTC 2015


Hi,

Le 22/10/2015 18:22, Aditya Sharma a écrit :
>
>     Hello,
>
>     Sorry for not being able to get back to you on IRC. I have been
>     busy with another project of mine at my university for quite some
>     time now.
>
>     If a method was virtual before and is pure virtual now then one
>     can't call that function as the method is of an abstract class now
>     in Qt5 (and hence can't be instantiated). Just removing these
>     function calls won't help either as the classes that derive from
>     these will also remain an abstract class until these pure virtual
>     functions are reimplemented in the derived classes.
>
>     So the calls to these specific functions will automatically turn
>     up as errors during the compilation process (thus automatically
>     highlighting the need to take care of the now pure virtual
>     function). Can you please explain why then is it needed to use
>     pragma in these cases?
>
>     Regards,
>     Aroonav.
>
>
> Apart from what Aroonav said, I'd like to add that if there were these 
> errors, then they would have appeared at compilation time only. Since 
> each derived class *needs* to define the function that it's deriving 
> from the abstract base class.
>
You are right, the problem wasn't this... I figured out after, that it 
was because of the Qt::UserRole that changed in Qt5.

Now I tracked the segmentation fault to 
/src/browsers/CollectionTreeItemModelBase.cpp, in
void CollectionTreeItemModelBase::listForLevel(int level, 
Collections::QueryMaker * qm, CollectionTreeItem * parent)


It appears that removing the last line (QTimer) removes the fault and 
that the slot is never triggered. I cannot go further for now. If you 
have any ideas.. please tell.

Cheers!
Olivier

> Regards
> Aditya
>
>
>
> _______________________________________________
> Amarok-devel mailing list
> Amarok-devel at kde.org
> https://mail.kde.org/mailman/listinfo/amarok-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/amarok-devel/attachments/20151022/2ebb366c/attachment.html>


More information about the Amarok-devel mailing list