CGAL users discussion list

[Monique Teillaud <>]

Olivier Devillers wrote:
>  a écrit :
> > Hello,
> > I would like to know what is the most efficient way to explore all 
> > edges of all
> > cells neighbor to a vertex v in a triangulation_3. Currently, I put 
> > incident
> > vertices of v into a vector ( by:
> > T.incident_vertices(v,back_inserter(neighbour_vertices)) ), I add v to 
> > the
> > vector, and I check all pairs of vertices in this vector to see if 
> > they are
> > connected by an edge in the triangulation or not( by: T.is_edge(...) ), I
> > wonder if there is any more efficient way to to it?
> > I would appreciate any help,
> > Ehsan Aganj
> >   
> from
>
> http://www.cgal.org/Manual/3.3/doc_html/cgal_manual/Triangulation_3_ref/Class_Triangulation_3.html 

> it seems that t.incident_edges is missing, (and is not what you want)

Note that the two following functionalities will be added in the coming 
release
(this does not answer your question, as noted by Olivier)

====================
template <class OutputIterator>
OutputIterator  t.incident_edges ( Vertex_handle v, OutputIterator edges)
		Copies all Edges incident to v to the output iterator edges. Returns 
the resulting output iterator.
Precondition:   t.dimension() >0, v Vertex_handle(), t.is_vertex(v).

template <class OutputIterator>
OutputIterator 	t.finite_incident_edges ( Vertex_handle v, 
OutputIterator edges)
		Copies all finite Edges incident to v to the output iterator edges. 
Returns the resulting output iterator.
Precondition:   t.dimension() >0, v Vertex_handle(), t.is_vertex(v).
====================

> a better workarround than yours is to take incident cells and take all 6 
> edges of these cells. In that way
> - an edge from v is obtained several time (6 on average)
> - an edge not containing v is obtained twice (probably with both 
> orientations if you do it carefully).
>
> removing duplicates is then needed.....