CGAL users discussion list

[Marc Glisse <>]

On Fri, 8 Mar 2024, CGAL_problem  wrote:

> My codes are as follows :

Please always provide a complete testcase, with #include and function, to 
make it easier for us to reproduce the issue (and preferably as an 
attachment or a github gist because some email programs do unspeakable 
things to the main text).

> 2. why the order of input finite vertices are changed?

CGAL first sorts the points according to some particular order (though in 
Triangulation_3 that's only done for Delaunay and not for a plain 
triangulation).

Points are then stored in a triangulation where they don't have any order. 
What you get is the order in which a graph traversal happens to visit 
those vertices.

If you want to remember information about vertices (say an int 
corresponding to their initial index), the doc has information about how 
to achieve that.

> 3. The expected results would contain the following tetrahedra, but I can not 
> find them in the output file.

There are several possible triangulations, there is no reason to expect 
one in particular.

> > > > > The results in the output file are as follows:   
> 3
> 8
> 3 1 2 -3
> 3 1 -2 -3
> 3 -1 -2 -3
> 3 1 2 3
> 3 -1 2 -3
> 3 1 -2 3
> 3 -1 -2 3
> 3 -1 2 3
> 3 1 2 -3

Notice that it announces 8 points and then proceeds to print 9. My guess 
is that one (the first) is the infinite vertex, which does not have a 
point associated to it, so the coordinates it prints are read from a 
random place in memory (happens to be the point of another vertex here). 
Maybe we could improve that in CGAL, I didn't check what Triangulation_3 
does.

Does that help interpret the output you get?

--
Marc Glisse