CGAL users discussion list

[Marc Alexa <>]

Thanks! I’d say it’s still a little confusing that one casts to the element handles and the other doesn’t. Plus for 3D triangulations it works the same for both iterators.

I needed it because rather than extending the base classes for attaching information I like to create a hash map to int and than use the indices, like so:

    int nv = 0;

    CGAL::Unique_hash_map<T::Vertex_handle, int> VH;

    T::Finite_vertex_iterator vit=t.finite_vertices_begin();

    for (; vit != t.finite_vertices_end(); ++vit)

        VH[vit] = nv++; // error here: no known conversion

This works for the non-finite iterators. Which, frankly, took me some time to figure out. I believe this construction can also be found in the CGAL source code, at least for 3D triangulations. 

I guess you are saying that

VH[vit.base()] = nv++;

does the trick, correct? This is good to know. 

Best,

Marc

On 28. May 2020, at 08:00, Marc Glisse <> wrote:

On Wed, 27 May 2020, Marc Alexa  wrote:

I am having trouble to use the _finite_ element iterators in the dD triangulation package. I’d expect that I can extract vertex resp. full cell handles from the iterators, but I doesn’t work in the ways I tried.

They are boost::filtered_iterator, so you can use .base() for that, but why do you need it?

--
Marc Glisse

--
You are currently subscribed to cgal-discuss.
To unsubscribe or access the archives, go to
https://sympa.inria.fr/sympa/info/cgal-discuss