CGAL users discussion list

["Laurent Rineau (GeometryFactory)" <>]

On Tuesday 23 March 2010 11:04:35 Simon Perkins wrote:
> Hi Laurent
> 
> I've added the assertion but the problem still occurs. Here is the code. It
> produces the two different precondition errors on both CGAL 3.5.1 and 3.6.
> The call to refine_Delaunay_mesh_2 is on line 62 of trigen.cpp.

Congratulation! You have found a bug in Mesh_2! Your data set is quite 
strange, because you insert in a cdt all finite edges of another one. That 
uncommon data set triggers a bug in the detection of clusters of constrained 
edges.

I will send you a fix of include/CGAL/Mesh_2/Clusters.h before the end of the 
week.

Do you agree that your (random) data set is inserted in the CGAL test suite? 
Being uncommon, it is interesting!

-- 
Laurent Rineau, PhD
R&D Engineer at GeometryFactory           http://www.geometryfactory.com/
Release Manager of the CGAL Project       http://www.cgal.org/