CGAL users discussion list

[Sebastien Loriot <>]

Sebastien Loriot wrote:

Jean Pierre Charalambos wrote:

Hi!

any thought regarding my conversion issue? I have been not able to fix it :(

Best,

Jean Pierre



On Tue, Feb 17, 2009 at 6:59 PM, Jean Pierre Charalambos
 wrote:
  

Hi cgal users,

I'm new to cgal 3.4 and I want to compute the minimal bounding sphere of a set
of spheres. So I just tried to adapt the min_sphere_of_spheres_d_3.cpp
example, but I have a problem when trying to convert the sphere types (radius
and center) to double types. I suppose this is a basic question, but I have
read the faqs and I googled a lot without success. My code is:

typedef CGAL::Gmpq FT;
typedef CGAL::Cartesian<FT> K;
typedef CGAL::Min_sphere_of_spheres_d_traits_3<K,FT> Traits;
typedef CGAL::Min_sphere_of_spheres_d<Traits> Min_sphere;
typedef K::Point_3  Point;
typedef Traits::Sphere  Sphere;

//after checking in the spheres in ms and computing the minimal sphere (just
like in the example) I attempted the conversion as follows:

radius = CGAL::to_double(ms.radius());

However, I got the following compilation error:

cannot convert 'CGAL::Null_tag' to 'double' in assignment

If I use "typedef double FT" (instead of typedef CGAL::Gmpq FT) and retrieve
the radius with "radius = ms.radius()" the code compiles (and it actually runs
perfectly in cgal3.3), but I get the following errror at running time:

terminate called after throwing an instance of 'CGAL::Assertion_exception'
 what():  CGAL ERROR: assertion violation!
Expr: -CGAL_IA_MUL(-1.1, 10.1) != CGAL_IA_MUL(1.1, 10.1)
File: /usr/include/CGAL/Interval_nt.h
Line: 154
Explanation: Wrong rounding: did you forget the -frounding-math option if you
use GCC?
Aborted

Thank you in advanced for your suggestions,

Jean Pierre
--
You don't become a failure until you're satisfied with being one.


    

Hello,

According to
http://www.cgal.org/Manual/3.4/doc_html/cgal_manual/Bounding_volumes_ref/Class_Min_sphere_of_spheres_d.html
and in particular the text after Min_sphere_of_spheres_d<Traits>::Result, the Result of radius() should in your case be a pair<FT,FT> (a,b) representing a number a+b*sqrt(t), where t is given by member function disciminant().


Sebastien

Sorry my answer was half complete.
Then, one solution of your problem is to apply the to_double to a,b and t.

Sebastien.