CGAL users discussion list

[Bai <>]

Thank you very much for Menelaos Karavelas's advice, I was busy during last
three days and has just back to lab.
I tried Gmpq, this really improved the efficiency and time spent on my cases
reduced to 2/3. Here is my result on 50 segments on polyline:
        GMpq insertSegments()
        1.89  
        
        MP_Float insertSegments()
        2.61

        GMpq insertPolyline()
        1.89
        
        MP_Float insertPolyline()
        2.594

However, this is only performed on Segment Delaunay Graph. 
I tried 600 segments on SDG(segment Delaunay Graph) and VD (Voronoi diagram)
respectively.
The results performed on SDG is ok, the is_valid() function returns true as
well.
However, VD is not good as expected. The is_valid() function did not finish
within 20 miniutes. And I interrputed the validating process manually.
I think Vu, Khuong's analysis is reasonable, there is somthing wrong with
the Voronoi Diagram Adapter, how do you think?

Another issue is proposed in my application, I still need some advice on the
DCEL structure in CGAL. 
It seems that the HalfEdge structure only gives the source and target point
, with out indicating what 
type the edge is, such as line segment, line ,or parabolic arc, any way to
obtain the detail information 
of the halfedges in DCEL?

Thank you.

-- 
View this message in context: 
http://cgal-discuss.949826.n4.nabble.com/Problems-on-segment-Voronoi-diagram-tp2023020p2068548.html
Sent from the cgal-discuss mailing list archive at Nabble.com.