CGAL users discussion list

[Dennis Schieferdecker <>]

Hi,
I am trying to compute exact circular arrangements with CGAL but even
with a simple example I run into trouble:

I use three circles at (0.01, 0.00), (-0.01, 0.00) and (0.00, 0.01) with
a radius of 0.01 each. Thus the circles intersect at (0.00, 0.00),
(0.01, 0.01) and (-0.01, 0.01). After computing the arrangement, I
iterate through the border (i.e. the halfedges) of each face and print
the source and target of each edge. Here I saw two things, that should
not happen with exact calculations:
First, there are edges which seem to have the same source and target,
but when comparing them with the appropriate equal-predicate they are
reported to be different. Secondly, e.g. instead of a location at (0.00,
0.00) a location at (0.00, -2.50766e-11) is reported.

Are the circular arrangements currently not implemented for exact
calculation or am I doing something wrong? (I am using the
CGAL::Exact_circular_kernel_2 and the CGAL::Arr_circle_segment_traits_2)

bye,
Dennis

-- 
Dennis Schieferdecker
Universität Karlsruhe (TH)           | Fon  : +49 (721) 608-6781
Institut für Theoretische Informatik | Fax  : +49 (721) 608-3088
Am Fasanengarten 5, Zimmer 220       |
D-76131 Karlsruhe, Germany           | Email: 

-----------------------------------------------------------------