Hi Robert, just quick answer (as I found an internet place on my holiday 
trip...)

- starting point as Vec3d - not a problem for me

- direction (vec3d) instead of end point (vec4d) - at the end.... it is not a 
problem for me. I started to think whether it would not give us trouble to 
switch LineSegmentIntersector by RayIntersector in OSG code. Two cases, I 
considered:
During IntersectionVisitor::apply: LineSegmentIntersector is transformed by 
MatrixTransform and should be RayIntersector afterwards (as MatrixTransform 
contained special matrix).
Second case, osgViewer::View::computeIntersections would need to consider 
which intersector it should use. But this should, hopefully, be simple, 
although I am not sure before implementing it.

John


On Friday 07 of June 2013 11:42:57 Robert Osfield wrote:
> Thanks for the extra insight John.  I think I'll need to re-read a few
> more times and do some background research get my head around the
> issue.
> 
> I am starting to wonder if perhaps RayIntersector shouldn't just have
> a Vec3d start and direction rather than offer the possibility of using
> Vec4d for positions, this way we avoid the issues with putting both
> points at infinity, users will also be more used to thinking of ray as
> start and direciton.
> 
> Robert.
> _______________________________________________
> osg-submissions mailing list
> [email protected]
> http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.
> org
_______________________________________________
osg-submissions mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org

Reply via email to