CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

On 02/16/2016 03:24 PM, Frédérik Paradis wrote:
> Hi everyone,
Hello Frédérik,

> I'm currently developing a module for CGAL for the Well-separated pair
> decomposition (WSPD) (See [1]). It is actually the first time I use CGAL
> so I do not fully understand yet the overall architecture of CGAL. So I
> have a couple of questions about how my module should work.
>
> For now, I developed the WSPD with the Split tree in 2D using the class
> Point_container.
>
> I developed it in 2D first because I wanted to develop an Ipelet for
> Ipe, but I want to develop it in d dimension. So, the first questions is
> when do you do many module for 2, 3 and dimension and when do you do
> only one module for d dimension?
There are several factors to take into account:
- would there be a significant speed up when using dimension specific 
implementation?
- can the implementation be made easily dimension generic?



> I'm currently using the class Point_container which seems really tied
> with the Kd-tree. Should I use that or implement the splitting part
> myself? Also, this class take in a parameter an iterator of that typedef:
>
> typedef std::vector<const Point_d*> Point_vector;
>
> It seems contrary to the coding rules of CGAL which has led me to
> believe it was more an internal class for the Split tree than anything else.
This class is not documented so it cannot appear in examples but it can
certainly be used in the implementation of algorithms of CGAL. Actually
it is even better to reuse something existing than creating an almost
identical duplicate. At the same time if you need some improvements
or new functions in this class they should be welcome.


> Also, when computing the WSPD, it requires the comparison of the
> distance between two circles and a factor times the radius of the two
> circles. For now, I'm just using CGAL::to_double and the sqrt function.
> Is there any other way a should do that?

I would need a more precise definition to answer correctly. However,
there is the traits Root_of_traits that depending on the number type
used will provide an exact representation of algebraic numbers of degree 
2  when using an exact number type.

http://doc.cgal.org/latest/Number_types/classCGAL_1_1Root__of__traits.html

http://doc.cgal.org/latest/Number_types/classCGAL_1_1Sqrt__extension.html

Sebastien.

> That's all for my questions but I will surely have more later.
>
> Thank you.
>
> Frédérik
>
>
> [1] http://people.scs.carleton.ca/~michiel/aa-handbook.pdf