CGAL users discussion list

[Sylvain Pion <>]

BBB HHH wrote:
> Hello,
>
> I am using CGAL 3.4 release with VS2005 SP1 under Windows XP.
>
> I used the join member function of the CGAL::Polygon_set_2<Kernel> class 
> template with an exact Kernel (such as Simple_cartesian<CGAL::Gmpq>), my 
> program works fine. But when i tried to use a built-in number type_based 
> Kernel (i.e. the Kernel Simple_cartesian<double>) i had the following 
> assertion.
>
> CGAL error: precondition violation!
> Expression : comp_f(object, parentP->object) != SMALLER
> File       : c:\program files\cgal-3.4\include\cgal\multiset.h
> Line       : 2128
> Explanation:
> Refer to the bug-reporting instructions at 
> http://www.cgal.org/bug_report.html
> CGAL error: assertion violation!
> Expression : (m_statusLine.size() == 0)
> File       : c:\program 
> files\cgal-3.4\include\cgal\sweep_line_2\basic_sweep_line_2_impl.h
> Line       : 232
> Explanation:
> Refer to the bug-reporting instructions at 
> http://www.cgal.org/bug_report.html
>
> I want to know if this is a CGAL bug or not and how to resolve it.

The CGAL code is not robust when you plug an inexact number type
like double.  This is the infamous robustness probem (see the FAQ,
the web site...).  So, don't use double.

--
Sylvain Pion
INRIA Sophia-Antipolis
Geometrica Project-Team
CGAL, http://cgal.org/