algorithmic modeling for Rhino

Sorting and grouping of points lying on the lines of the grid

Welcome colleagues

Please tell me whether it is possible as that sort points lie on the grid, so that they are in pairs as shown in figure !?



Views: 895


Replies are closed for this discussion.

Replies to This Discussion


I would strongly recommend some further study on DataTrees.

Also I would strongly (and emphatically) recommend the proper way, he he

I've sampled the 2 possible subdivision "methods" together : can you see the difference? (just exploit the order in the triPolys DataTree when the gate is open for the Mesh way).

PS: Skype me (we need to spend "some" minutes on these mysterious matters).

best, Peter


Anyway ... because it's you, he he

Is this what are you trying to achieve?

PS: Skype me


And (because it's you), he he


Yikes : WHAT I was thinking? (nothing apparently, he he)

Please accept my apologies (PUT this new C# into the other def)


Again: Sorry


Feeling guilty for my terrible mistake (don't forget to put the "C#NewAC" in your other scripts) I've prepared a small tutorial : what "maps" to what? (that's the 1M question, he he)


Peter thank you very much for your answers !!! With me is due !!!
Can you please suggest how from my code to assign the parameter in your code !? To pass the point of my code your !?


Sweet Jesus (you are a tough client, he he). This little C# gets 6 pts in a given branch and makes 3 subbranches each containing 2 items (your pairs) - kinda like the GH path mapper. But why you insist on that thing since mine does what you want with no issues? (or maybe you want something other: like rearranging the triads kinda like a surf subdivision ?).

OK, I'll give a look on it (P.S. I haven't examine your def in depth since you did some very CPU taxing stuff ... and thus I've "translated" your requirement into a much simplified def that works 100(?) times faster(??)).

BUT ... if you did bad things (like a tree with variable items per branch) > remind me to kill you when we meet (no mercy).

have ready some red pepper Vodka (and some caviar - obviously).

he, he

And of course you did bad things :

1. What is this? What is the meaning of sampling variable amount of items in a datatree that is supposedly a collection of "triad" Lines (i.e. 3 items per branch).

2. And this? 3K Pts for what?

honestly picked at random algorithm (((

Hmm .. I'll pass this "random" thingy (but we are Engineers you know, he he).

1. I publicly admit that I'm lost with your definition: too chaotic for my primitive mind (AND WHY you don't use mine?).

2. Added the equivalent of C# with mapper (2 options in fact). Of course if you feed mapper with garbage > garbage is what you'll get.

3. Added hints about the most critical thing in MERO systems: node connectivity (but I prefer to do this with code: I feel far more "in command").

4. But ... another critical thing is ... well the obvious: relying solely to some mesh triangulation method ... you can't control the triads (some are heading east some west). We don't want that don't we?


OK, here's the deal:

Allow me to do this entirely with code. Why may you ask (same question comes in some minds in this Noble Forum, he he).

Well... to make a very long story short I'll use only 2 arguments:

1. You are targeting CATIA don't you? (not exactly tomorrow but ... soon) and/or SolidWorks (hello C# haven't we met before?).

2. You MUST deal with nested block instances instead of what you are trying to do right now (I'm talking about the real MERO things not abstract Lines and points). This is not doable with GH components I'm afraid (but it's rather easy with code).

3. You MUST deal with RDBMS in order to keep track with what's going on in your company per project per case per designer (who sells that bolt? what's his cat name? is he a reliable supplier? what I'm doing in life? ... that sort of "queries"). At this point: CATIA is 1% CAD things and 99% PLM stuff (Product Life cycle Management). We do want that since it's 21st century running don't we?.

I hear you: but these are 3 arguments ... indeed but ... hey who's counting? he he.


A. This def attached has a very simple C# that gets mesh Pts and makes a nice U/V style collection of points (DataTree in plain English).

B. Then we go to that umbrella sticks thingy: we can calculate anything (already the thing does "some") plus your collections of divided points (with the right way, he he) VS a given node: you said (Skype) that you want to calculate angles with these (from 2 to 6)  in mind: obvious since you are doing real-life MERO things.

C. Then we could calculate the appropriate Planes for PlaneToPlane transformations: get a nested instance definition (the red things that you've showed to me yesterday) placed at 0,0,0 (Plane.WorldXY) and put in in every Plane collection related with every node (clash defection is an obvious must).

Case resolved, closed: what about that Vodka?

More in Skype







  • Add Photos
  • View All

© 2019   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service