CGAL users discussion list

[Efi Fogel <>]

1. The  CGAL::Exact_predicates_exact_constructions_kernel is defined as Lazy_kernel<Simple_cartesian<Epeck_ft> > (see Exact_predicates_exact_constructions_kernel.h).

You can try to replace both occurrences of CGAL::Cartesian<SOMETHING> with Lazy_kernel<Simple_cartesian<SMETHING> >, but I haven't tried it with this particular traits.

2. Usually, a traits that supports a limited type of curves would be more efficient then a traits that supports a broader type.

For example, the Arr_segment_traits_2 is more efficient then the Arr_algebraic_segment_traits_2 for handling segments.

However, the Arr_conic_traits_2.h is a bit old and the Arr_algebraic_segment_traits_2 might beat it for handling conics in certain cases.

   ____  _        ____             _
  /_____/_) o    /__________  __  //
 (____ (   (    (    (_/ (_/-(-'_(/
                         _/

On 17 February 2018 at 21:01, Stuart Hungerford <> wrote:

On Sun, Feb 18, 2018 at 1:46 AM, Efi Fogel <> wrote:

> 1. I guess it does, but why not using one of the predefined kernel (see
> below)?
>
> #include <CGAL/Exact_predicates_exact_constructions_kernel.h>
>
> typedef CGAL::Exact_predicates_exact_constructions_kernel   Linear_kernel;
>
> 2. The points and curve types defined by the circular kernel and the points
> and curves defined by the Arr_circle_segment_traits_2 traits are the same.
> They use coordinates represented by a type called CGAL: :Sqrt_extension. You
> can read more about it in the manual.)

Thanks for the pointers Efi -- much appreciated.

Does that approach of using a predefined kernel also hold for conic curves?

Following the example in
https://doc.cgal.org/latest/Arrangement_on_surface_2/Arrangement_on_surface_2_2conics_8cpp-example.html,
the example uses explicit choices of number type etc.

I'm asking so I can create ellipses and elliptic arcs using CGAL.
(Even if I have to create the bounding box and affine transformation
methods myself).

Thanks,

Stu

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