Skip to Content.
Sympa Menu

cgal-discuss - Re: [cgal-discuss] Cartesian kernel with my own Point - Take 2

Subject: CGAL users discussion list

List archive

Re: [cgal-discuss] Cartesian kernel with my own Point - Take 2


Chronological Thread 
  • From: "Sebastien Loriot (GeometryFactory)" <>
  • To:
  • Subject: Re: [cgal-discuss] Cartesian kernel with my own Point - Take 2
  • Date: Thu, 10 May 2012 09:41:20 +0200

I know it is not user friendly at all, but you really need to follow
what is described here:

http://www.cgal.org/Manual/latest/doc_html/cgal_manual/Kernel_23/Chapter_main.html#Section_11.5

Your inheritance is not correct.

Sebastien.

On 05/09/2012 12:30 AM, Tomislav Maric wrote:
Hi everyone,

I tried to inherit simply from the CGAL::Cartesian kernel and provide
the functors that the Concept requires, instead of using filtered kernel
(I don't get how this works, but I'm reading through it). Still, even
after providing the functors, I get the same error again, when I'm
trying to read the Polyhedron_3 from an .off file. Everything else
works. The problematic Kernel_23 modified example is attached.

The OFF file scanner obviously expects a type R defined within the Point
P. I don't see how I can provide this in any way from outside of the
Point class. It's not trying to access R through a proxy, like a trait
class, it asks for R directly.

Q1:

Does this mean that if I want to use my Point class, I need to write a
specialization of the

template < class Point> inline
Point&
file_scan_vertex( File_scanner_OFF& scanner, Point& p)

for my Point? Is this the proper solution for this problem?


Q2:

Why are the point coordinates constructed using the RT (Ring Type)?

p = Point( RT(x), RT(y), RT(z));

(it's the same for the homogeneous representation)


Thanks for the help!

Best regards,
Tomislav


Attachment: Kernel_23.tgz
Description: application/compressed-tar




Archive powered by MHonArc 2.6.16.

Top of Page