How can I extract the data created by a particle combing operation (on hair particles, for instance) to file (even if some Python code is required) in a 3-component RGB normal image map to UV space (when the unwrapping has already been done for the geometry), and then (that's a second question, I guess) feed it back to the particle system without having to recomb again?
I do not have an answer for you, although I can provide you with a few nuggets of info that will come in handy.
Here is some python code that shows you how to dump the coordinates of the root and tip of the hairs:
import bpy def dump_particles(ps): for p in ps.particles: print(p.location) #print(p.fuv) print(p.hair_keys[-1].co_local) print("_") dump_particles(bpy.context.active_object.particle_systems)
This makes calculating the position of the hair in UV coordinates much harder (although you could probably reverse-engineer it from the location if you are willing to do a lot of analytic geometry).
Even once you do figure out where the hair is in UV space, you'll need to choose how to interpolate to compute the values for the image you want to use as a normal map. (Maybe compute the dual of a voronoi diagram and use barycentric coordinates again).