CGAL users discussion list

[Juan Carlos Lopez Alfonso <>]

Hi Sebastien:

Thank you for all your answers. I will use your code to solve my problem, because I only need to have a jacobian different to 0, even when it is very close to 0. 

if I use this instruction: double exact_jacobian_double = exact_jacobian.to_double();

my jacobian is very close to zero but not equal to zero, and for this reason I can divide by this value without errors.

On the other hand, In december I was in INRIA and is a very nice place, I will have the honor to stay for  four months from March and I will be a very interesting oportunity for my research.

Thank you for your help, and for this user list

Juan Carlos

2012/1/20 Sebastien Loriot (GeometryFactory) <>

Try running the attached example, and you'll see you points are not coplanar.

In CGAL, we have nothing to change the predicates to consider almost
coplanar points as coplanar.
In your case, I remember that points are coming from parallel slices.
My last advice will be that you should round your points from each slice so that they exactly lie on a plane.

Please also do read this FAQ entry:
http://www.cgal.org/FAQ.html#inexact_NT

Sebastien.

On 01/20/2012 11:47 AM, Juan Carlos Lopez Alfonso wrote:

Hi Oliver:

Thank you for your answer. But, maybe you dont understand my point.

How is possible that this points are not coplanar when the jacobian is
exactly equal to 0? and you dont answer my question:  How Can I say to
CGAL when computes the triangualtion that these points are coplanar?  at
least with my criteria!!!

On the other hand, I cant include or remove points, and I need that the
jacobian of each thetraedron is different to 0. I am using MS 10.0 and
the jacobian is equal to 0, how is possible that in CGAL these points
are not coplanar? I cant understand this issue.

In my code I will use the Jacobian and in some case some expressions are
divide by the jacobian, and for this reason I need to have a jacobian
different to 0, while in c++ the jacobian is 0 for CGAL is not.

Regards
Juan Carlos

On Fri, Jan 20, 2012 at 11:32 AM, Olivier Devillers

< <mailto:>> wrote:

   Le 1/20/12 11:13 AM, Juan Carlos Lopez Alfonso a écrit :

       Hi There:

       I have this concrete example:
       ..........

       - How is possible that in CGAL these points are not coplanar?
       because these points are coplanar.

   as already explained your points ARE NOT coplanar,
   they are just almost coplanar (and since the volume is quite small
   you round it t zero)


       - In order that I can't remove points, How Can I say to CGAL
       when computes the triangualtion that these points are coplanar?

   THEY ARE NOT.
   If you can compute another triangulation, but it will not be the
   Delaunay triangulation.


       - I will use the resultant triangulation in other software where
       the jacobian is computed and I can't use the CGAL library. Is
       possible to change the CGAL test of coplanar points when the
       triangulation is computed?

   if you cannot remove points, may be you can add (e.g. using the
   meshing package).

   Olivier


   --
   You are currently subscribed to cgal-discuss.
   To unsubscribe or access the archives, go to

   https://lists-sop.inria.fr/__wws/info/cgal-discuss
   <https://lists-sop.inria.fr/wws/info/cgal-discuss>

--
You are currently subscribed to cgal-discuss.
To unsubscribe or access the archives, go to
https://lists-sop.inria.fr/wws/info/cgal-discuss