CGAL users discussion list

During the post-union simplification phase, I run into a problem in
snc_sm_overlayer around line 399.

    set_face(v, *(UF.find(Pitem[v->incident_sface()])));

Isolated svertex v has no incident sface (i.e. incident_sface() ==
SFace_handle()). This isolated vertex with no incident_sface appears to be
some kind of temporary byproduct of simplification. I notice that this
situation occurs very frequently in cases where the nef union succeeds
(this is the first problem I've had in several hundred runs). What isn't
common is that Vitem[v] == NULL.

Both input nefs are verified as is_valid. It's a little difficult to
provide a repro, as the input nefs are non-manifold and use a custom
marking scheme (non-binary). But perhaps somebody has an idea on how to
debug this?

Regards,
Fred