CGAL users discussion list

[Andreas Fabri <>]

Bernhard Kornberger wrote:
> > Hello Bernhard,
> >
> > You haven't computed the plane equation. See:
> >
> > http://www.cgal.org/Manual/3.3/doc_html/cgal_manual/Polyhedron/Chapter_main.html#Subsection_12.3.4 
> >
> >
> >
> > andreas
>
> Thank you, Andreas!
>
> This solves the problem. Sure, this is documented, but a programmer
> can not expect that behavior. And even worse, the function returns
> a wrong plane instead of some error message. Shouldn't the function
> have an assertion like assert(plane_is_valid()); ??

I admit that this isn't fool proof, and leaves space for improvement.


> FYI: The other problem where the software only crashes when it is
> started in valgrind is still there.

I just downloaded the tarball.  Can you please reduce the complexity
of the bug report example  to a single file with some includes and typedefs
in the beginning, and a main that generates a triangulation and inserts
some points and still crashes.

I could do it myself, but I think it is your task.


It seems that the points are the points of a cube. Do you observe the
same with simpler values than

double l=-2396.4;
double u=2595.59;
        

that is with 0 and 1



best

andreas


> Best
> Bernhard