[vtkusers] Single source shortest path class published

[Rasmus Reinhold Paulsen]

> -----Original Message-----
> From: vtkusers-admin at public.kitware.com
> [mailto:vtkusers-admin at public.kitware.com]On Behalf Of John Biddiscombe
> Sent: 4. april 2002 23:16

> I haven't tested the code, but did flip through it casually. Would it
> require a lot of work to alter the distance measure from physical {x,y,x}
> Euclidian to some function of the scalar value at a point - ie a sphere as
> in the example, but with different scalar values at each point
> rather than a
> distortion of shape. I don't need this, but it would be
> interesting to play
> with assigning random scalars to surfaces and making it traverse them in
> scalar space. (I say function of scalar value as there may be a vector
> quantity)

That will not require a lot of work and I have incorporated an experimental
switch in the code that does something of what you ask. I think the best way
to do it is to let the user supply an edge cost callback function that based
on the two vertice indeces returns a weight that can be based on the vertex
positions, the scalar values or both.
I just have to figure out have to make this funky callback mechanism behave
like I want.

Thanks for the comments!
Rasmus