CGAL users discussion list

[Efi Fogel <>]

This is intentional. The operation performed is regularized. That is, 
the result is the closure of the non-regularized operation. So in your 
words you may think of the vertices of the hole in the resulting polygon 
as being "epsilon" distance from the outer boundary.


 wrote:
> Hello,
>
> CGAL symmetric difference seems to give wrong results on this input:
>
> polygon 1: (axis aligned square)
>
> 0 0     
> 10 0    
> 10 10   
> 0 10
>
> Polygon 2:  (45 degrees aligned square that touches the edges of polygon 1)
>
> 5 0     
> 10 5    
> 5 10    
> 0 5
>
> The proper symmetric difference is four triangles, but the result given by the 
> symmetric_difference() operation gives polygon 1 as the outer boundary, and polygon 2 
> as its inner holes, as if there is an "epsilon" difference between the 
> vertices of polygon 2 and the edges of polygon 1. Am I doing it wrong, or is this 
> intentional?
>
> Thanks in advance,
> Amir Vaxman.
>   

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