CGAL users discussion list

[Ben Supnik <>]

Hi Y'all,

I have a general polygon set that I want to calculate a striaght 
skeleton for.  Due to the complexity of the layout, I tried the trick of 
using an eact pred/inexact construction kernel, then convert.

But when I do this I get a clearly bogus skeleton...skeleton edges 
crossing each other and going outside the polygon area entirely.  Using 
the exact pred/exact construction kernel does not have this problem (but 
is 48 times slower).

Now here's the confession: the polygon boundaries come from boolean 
operations on a general_polygon_set, and the input data is _not_ very 
good.  I'm having to round down via CGAL::to_double to input curves into 
the inexact kernel, and it is possible that I am creating a bogus 
contour that way. :-(

But...when I run polygon_2.is_simple() on the input curves, they all say 
they're fine.  And some of the areas that have a broken skeleton look 
not-very-complex to me...they shouldn't be up against the limits of 
precision.

Any hints as to what's going on?  This is an off-line process so I can 
just use the slow kernel, but the processing time is going to get a bit 
out of control.

cheers
ben


--
Scenery Home Page: http://scenery.x-plane.com/
Scenery blog: http://xplanescenery.blogspot.com/
Plugin SDK: http://www.xsquawkbox.net/xpsdk/
X-Plane Wiki: http://wiki.x-plane.com/
Scenery mailing list: 

Developer mailing list: