CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

The pb comes from the fact that a point from the homogeneous kernel is
read from a stream as x y z w.

That is if the points in the off are 3 doubles they cannot be
interpreted as an homogeneous point.

Is there any reason why you need this kernel rather than 
CGAL::Exact_predicates_exact_constructions_kernel?

Sebastien.

On 03/13/2015 02:32 PM, KHartmann wrote:
> Hi Sebastion,
>
> Could you tell me how you have defined "Kernel" in your sample code (shown
> at bottom)?
>
> When I define it as CGAL::Homogeneous<CGAL::Gmpz>, I get thousands of
> run-time warnings telling me that the data being read-in is "double" when it
> should be "integer":
> ----------------------------------------------------------------------
> Line       : 102
> Explanation: Gmpz constructed from non-integer double value
> Refer to the bug-reporting instructions at
> http://www.cgal.org/bug_report.html
> CGAL warning: check violation!
> Expression : is_integer(d)
> -----------------------------------------------------------------------
>
> And then eventually the program crashes.
>
> When I use any other type, compilation fails.
> (CGAL::Gmpzf, CGAL::Gmpfr...)
>
> My test data is "s.off" in this location:
>
> https://docs.google.com/file/d/0B4PDrM6_sr2UdzFJYTZYOWdTb2VJTkp5WmVuXzAxdw/edit
>
> Thanks,
> Kevin Hartmann
>
> ------- Sebastion's sample code ------------------
> #include <CGAL/IO/Polyhedron_iostream.h>
>
> std::ofstream file("myoff.off");
> Polyhedron_3<Kernel> P;
> file >> P;
>
> Nef_polyhedron_3<Kernel> nef(P);
> -----------------------------------------------
>
> (ofstream should be ifstream)
>
>
>
>
> --
> View this message in context: 
> http://cgal-discuss.949826.n4.nabble.com/Why-do-the-OFF-to-nef-3-take-so-much-time-tp4440642p4660580.html
> Sent from the cgal-discuss mailing list archive at Nabble.com.