Greetings,

I am using an algorithm to process points from a point cloud that requires me to iterate over all the points twice (unfortunately). So, this begs the question: What is the proper way of resetting or restarting a pciterate loop? Do I have to close the handle (pcclose), and reopen the point again or is there some other way to simply loop again without having to pcclose/pcopen again? (To call the pcopen twice seems…unnecessary.)

Regards,
Van Aarde.

There isn't any reason why you can't call pcopen more than one time. The AXYZ SSS shaders in the Houdini distribution do this.

Hi Wolfwood,

Thanks for the reply .

I wasn't saying (or thinking) that theres anything wrong with doing it that way. I was merely trying to find it if there is a more *optimal* way of doing it.
Now, I not familiar with the internal workings of the point cloud functions so this is how I picture it would work:

pcopen() looks up a bunch of points within the specified radius from a given point. These points are stored in memory (??) and an pciterate can be used to iterate over them. Now, to iterate over them a second time, I would have to close the handle, pcopen() again (which does the whole lookup thing again), stores the points in memory and only then I iterate over them again.

All I want to do is iterate over the points without having to do the pcopen lookup again: something like:
pcreset(handle); <– reset the iterator to the beginning of the list.

Unless, of course, if the pcopen/pciterate functions are so super optimal that it is not necessary at all…then I'm okay with it .

Hmmmm. Despite what the help says I seem to be able have multiple pciterates without any ill effects.

I have had mixed results with multiple loops at the same time. I guess it really just depends what you're looking to get out of the loop.


-j

nanocell
pcopen() looks up a bunch of points within the specified radius from a given point. These points are stored in memory (??) and an pciterate can be used to iterate over them. Now, to iterate over them a second time, I would have to close the handle, pcopen() again (which does the whole lookup thing again), stores the points in memory and only then I iterate over them again.

I don't know the details of how pcopen() was implemented either, but I'm guessing it just builds a new kd-tree on the same point data. It seems you're thinking about always using a single handle for both iterations, but you shouldn't (at least not if they're nested). Anyway… I had posted some more information in your odForce incarnation of this thread [forums.odforce.net] (sorry, didn't know you'd posted here as well).

My intention was actually to use the loops sequentially (not nested). I would never use the same iterator fo a nested loop! Blasphemy! I'll specify it next time.

Nevertheless. I double posted, here (for Houdini developer input on the pcopen implementation) and on odForce for any other advice .

Okay, back to odForce!