CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

On 03/23/2012 02:34 AM, Zohar wrote:
> 1. This is our current thread:
>
> http://cgal-discuss.949826.n4.nabble.com/Convert-C3t3-to-Polyhedron-td4467182.html
>
> In the first post I gave a link by Lehtonen Matti, and commented about it,
> and I believe you gave me an older link of the same guy. I pointed out that
> he uses a different interface, and even after making the changes it still
> didn't work.
Sorry I looked for a link in a previous post but could not find it.

> 2. I saw the facets enum in a complex, but the whole doc is a bit obscure
> about c3t3. And yes, if I spend a few days on this, I believe I could write
> something. It would have been easier if I knew how to detect if a facet
> belongs to the boundary, and how to discern between the different domains of
> the complex. I might be mixing definitions, since I didn't fully understood
> the doc. I need to read it again, and a depiction like in the polyhedron doc
> would have surely helped . Since I don't need it urgently, I prefer to look
> for an already made code - someone said he would fix the one on your link.
If you look at the file CGAL/IO/Complex_3_in_triangulation_3_to_vtk.h,
you will see that it's easy to either write it to an off or write a 
polyhedron modifier to extract the boundary on the mesh,

To get the subdomain of a facet f, use f.first->subdomain_index().


> 3. I am more curious about the difference between the two mesh generation
> algorithms, the one for c2t3 and the one for c3t3. Theuy both seem to rely
> on [BO05], which basically uses 3D restricted delaunay. As such I also
> wondered, if the c3t3 would have generated a surface (polyhedron), would it
> be, as expected, the same surface generated by c2t3?

> Why the c2t3 is still buggy,
Why are you saying that? do you have a bug report?

> why the c3t3 doesn't have an official polyhedron converter, and
Mainly because a c3t3 contains information about a volumetric mesh, not
a surface mesh. I agree in some case it can be useful.


> why each (c3t3, c2t3) is in a different package?
Same as above, they are not made for the same purpose.
However you are right that a c2t3 can be represented within a c3t3.
In a near future, the surface mesher package will reuse some
functionality of the volume mesher,

Sebastien.

> --
> View this message in context: 
> http://cgal-discuss.949826.n4.nabble.com/Convert-C3t3-to-Polyhedron-tp4467182p4497593.html
> Sent from the cgal-discuss mailing list archive at Nabble.com.