CGAL users discussion list

[Martin Baeker <>]

Hi Priyank,

thanks for the mail. I'm a bit confused, though, because the manual
states that a Polyhedron consists of halfedges - what exactly do you
mean by loosing half-edge information? I can write a polyhedron to an
off and read it back in - so far, it seems that I only loose on
additional information like the plane - I presume, I'll loose
everything that is marked as optional in the Polyhedron_3-section
"Types for Tagging Optional Features"

BTW, why do you write a renderer? Isn't there is a QT-renderer
available (it is for Nef-Polyhedra), or is yours more powerful? In
that case, I'd be interested in having a look...

Thanks,

Martin.



> Hi Martin,
>
> Im new to CGAL myself and playing with polyhedron for an algorithm that I am 
> trying to implement. So, take my advise with a pinch of salt!
>
> From what I understand, when you write a polyhedron to a .off file and read 
> it back in..I think you loose out on the half-edge data structure 
> information. Now, if you don't need the half-edge data structure later on, 
> its okay to use .off format. But, if you want to really use the half edge 
> information later on, you will have to come up with a file format (or maybe 
> already exists out there that is geared towards half-edge ds). Eitherways, I 
> am writing a small renderer for polyhedron, and if you want I can share the 
> code with you when I'm done with it. Its in Qt, so that should take care of 
> cross-platform issues.
>
> Hope that helped some.
> /Priyank
>
> On 3/16/2010 7:48 AM, Martin Baeker wrote:
> > Hi Andreas,
> >
> > thanks for the answer - so it seems the only way to do it is to
> > actually cycle over all the vertices of the facet and create a plane
> > from that? I still do not really understand why there is no standard
> > function doing this, but surely I can program it myself.
> >
> > Thanks again,
> >
> > Martin.
> >
> >
> >
> > > Hello Martin,
> > >
> > > If I understand it right, you would expect that the plane
> > > equation gets recomputed when the off file is loaded.
> > >
> > > I don't see that this is supported by the off file format: 
> > > http://www.geomview.org/docs/oogltour.html
> > >
> > > Also it would not make sense, that CGAL computes the plane
> > > equation after loading, as you could not have computed
> > > it when you wrote it, or have stored yet another plane.
> > >
> > > One solution would be to have an eoff format
> > > ( I just come up with this name, with the 'e'
> > > standing for extended), which writes/reads whatever
> > > information added to vertices and faces to the stream.
> > >
> > > best regards,
> > >
> > > andreas
> > >
> > > On 16/03/2010 11:43, Martin Baeker wrote:
> > > > Dear all,
> > > >
> > > > I'm getting a bit desperate while trying to work with Polyhedra-stuff.
> > > >
> > > > The manual says that I can access the plane that belongs to a facet with
> > > > Plane_3& f.plane () if Supports_facet_plane == CGAL::Tag_true
> > > >
> > > > However, I do not understand how this really works.
> > > >
> > > > When I create a Polyhedron, e.g., with the make_cube_3-routine that is
> > > > provided in the examples, I can access the plane corresponding to each
> > > > facet without problems.
> > > > However, when I write this Polyhedron to an off-file and read it in
> > > > again, I cannot access its planes anymore.
> > > >
> > > > Here is a simple test program to show this:
> > > >
> > > > void do_stuff(Polyhedron & pcube)
> > > > {
> > > > for (Halfedge_iterator h=pcube.halfedges_begin(); h !=
> > > > pcube.halfedges_end(); ++h)
> > > > {
> > > > Facet face=*(h->facet());
> > > > std::cout << face.plane().orthogonal_vector() << std::endl;
> > > > }
> > > > }
> > > >
> > > > int main() {
> > > > Polyhedron pcube;
> > > > make_cube_3(pcube);
> > > > std::ofstream os1("poly2.off" ) ;
> > > > os1 << pcube ;
> > > > os1.close();
> > > > do_stuff(pcube);
> > > > std::cout << "\n";
> > > > Polyhedron pcube2;
> > > > std::ifstream is1("poly2.off" ) ;
> > > > is1 >> pcube2;
> > > > do_stuff(pcube2);
> > > > }
> > > >
> > > > The output for pcube is correct; for pcube2, all orthogonal vectors are
> > > > output as 0/1 0/1 0/1.
> > > >
> > > >
> > > > Is there a way to make sure that I can always access the plane
> > > > belonging to a facet? If not, why not? Could I do it by just taking
> > > > three points from the facet and create a plane with them?
> > > >
> > > >
> > > > Any help will be greatly appreciated,
> > > >
> > > > Martin.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Priv.-Doz. Dr. Martin Bäker
> > > > Institut für Werkstoffe
> > > > Technische Universität Braunschweig
> > > > Langer Kamp 8
> > > > 38106 Braunschweig
> > > > Germany
> > > > Tel.: 00-49-531-391-3073
> > > > Fax 00-49-531-391-3058
> > > > e-mail 
> > > > <>
> > >
> > >
> > > --
> > > You are currently subscribed to cgal-discuss.
> > > To unsubscribe or access the archives, go to
> > > https://lists-sop.inria.fr/wws/info/cgal-discuss
> >
> >                    Priv.-Doz. Dr. Martin Bäker
> >                    Institut für Werkstoffe
> >                    Technische Universität Braunschweig
> >                    Langer Kamp 8
> >                    38106 Braunschweig
> >                    Germany
> >                    Tel.: 00-49-531-391-3073
> >                    Fax   00-49-531-391-3058
> >                    e-mail 
> > <>
>
> --
> You are currently subscribed to cgal-discuss.
> To unsubscribe or access the archives, go to
> https://lists-sop.inria.fr/wws/info/cgal-discuss

                   Priv.-Doz. Dr. Martin Bäker
                   Institut für Werkstoffe
                   Technische Universität Braunschweig
                   Langer Kamp 8
                   38106 Braunschweig
                   Germany
                   Tel.: 00-49-531-391-3073
                   Fax   00-49-531-391-3058
                   e-mail 
<>