Planner issues 1#

Tomaz Canabrava tcanabrava at kde.org
Sun Jun 1 07:11:03 PDT 2014


ok - quick fix then:
if  ( isnan(x) || isnan(y) )
     setPos(0,0); // this will trigger one more recursion depth, but
it should fix.

On Sun, Jun 1, 2014 at 10:03 AM, Henrik Brautaset Aronsen
<subsurface at henrik.synth.no> wrote:
> Tomaz Canabrava wrote:
>>
>> infinite recursion on recalculate.
>> Qt should only send the itemChangeEvent *if* the position changed ( if
>> I setPos with the same pos() things shouldn't trigger itemChange) but
>> it seems that this is not true. Can you two put a qDebug()<<  pos();
>> on the recalculate to see if the positions are the same or if we are
>> getting something weirder on?
>
>
> OK, I reverted Dirk's commit e74914fdd0f439d3584657147c81539d2a863955 and
> inserted qDebug() << pos(); before line 44 of ruleritem.cpp.
>
> That gives me:
>
> QPointF(0, 0)
> QPointF(nan, 3)
> QPointF(nan, 3)
> QPointF(nan, 3)
> ... and a gazillion repeats of that last line before it crashes.
>
> Henrik
>


More information about the subsurface mailing list