Grasshopper

algorithmic modeling for Rhino

Hi everybody,

I have a problem with a list I need to obtain. Basically I want to get the list of points that have X values inside multiple domains, the problem is that I have 500+ points and 60+ domains.

I won't consider to do as many include as many domains I have and then use the "AND" command because it will take a lot of time, maybe you know the right way to do it.

Thanks

Views: 1262

Attachments:

Replies to This Discussion

Well I could consider doing it ... but your avatar conflicts with some other and THAT is a major issue

Plan B: use a pink hue.

he he

Lol, better now?

Well ... yes (Sorry about that ... but... these things matter, he he).

So, I'll look on your issue ASAP.

Hmm ... could you describe the goal of yours? (LEAVING aside domains et all) i.e. what are you trying to achieve?

For instance: I have a "canopy" type of stuff (a collection of the facet's edges, anyway) ... then I want to do this (say: making the Breps), this and that ...

I want to isolate two groups of curves: the vertical ones a on a list and the horizontal and arc shaped ones on another list

Well ... given this collection of surfaces (that can been "sorted" in x/y using their centroid) ... do you want to get the edges "along" x and the others "along" y? ("sorted as well": this is the left, this the right blah, blah).

PS: Of course we are talking about trees NOT lists.

And if yes ... describe the next step.

OK, here's the deal:

1. I'll do the sort thingy (sort centroids in X.Y and on sync sort the surfaces as well): technically this is an synchronous Array sort  ... and I'll prepare a proper data tree out of the edges (in left/right/bottom/top "order" as well). Each branch would derive from the sorted surfaces, each item (from 3 up to 4) it could be the surface edge.

2. Then you get your tree and do whatever you want: pick triangles or quads or both, pick based on surface area, offset the surfaces, project their centroids, extrude along the edges, offset edges on these (VERY SLOW: Avoid at any cost) or whatever. BTW: If I knew the "whatever" (PS: what means  "last post: what last post?") it could be rather better.

Note: extruding "different" profiles along these "sorted" edges is a giant rabbit hole - you'll see soon the reason.

BTW: Some sketch(es) anyone ?? (with the "extruded" things)

Ok let's see the result, thanks man

Good news: appears "working" (but is written at-once > meaning that a V2 is more than probable).

Bad news: it's C# (since I can think that way 100 times faster than using components) BUT IF IS what you want > I'll translate that freaky thing into native components .

Note: see comments and try to understand the "classification" on that @$@^$ east/west/north/south DataTree (I assume that you know well what these mysterious things are - if not notify: Skype is required).

Note: C# assumes that pieces (Breps) have min of 3 and max of 5 edges - apparently a far more "elaborated" way is required in order to deal with any situation.

Moral: roll on V2, he he

PS: Since the Dark Side is present in this Noble Forum ... you can use the White Option for your avatar (as Plan Z).

best, Lord of Darkness

Attachments:

Oops some 0,1 are swapped (blame Alzheimer). Discard V1

Attachments:

Close enough! I will not use the command I will need depending on the order of the elements but I will use it on the list of the curves, but that's not a problem obviously.

The problem is that I need two separate lists: one with the vertically oriented curves (red in the sketch) and one with the remainings (yellow in the sketch) (forgive my sketch, a wookie draw it).

I think that it will only require a little modification to the C# component. If you'll be able to do it in native it would mean that the force is with you!

Attachments:

Hmm ... but you DO have 2 separate Lists ... the only thing required is getting them out of the "proper" DataTree branches (the ones with the 2nd dimension 0 or 1). Anyway ...

Appears that ... it's time for the V2; but since there's the Brazilian F1 thingy going [at least the gossips right now] I'll do it after enjoying Lewis humiliating the opposition (I do hope, that is).

PS: sorting Breps (X and then Y) is not that bad idea in case that you want some "finer" control on these "pieces" (for instance: get me stuff from here up to there blah, blah - for testing or for other purpose).

PS: we do some reverse engineering here: these tri/quads (or more) are due to some trimmed surface subdivision right? so why bother with sort this sort that? Bring the donor surface(s) in order to simplify things.

Moral: go Lewis, go.

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service