Projecting a pattern definition on a cubic shell

Replies to This Discussion

Permalink Reply by peter fotiadis on February 11, 2016 at 3:24am

Well ... by "projecting" you mean apply some selection rules to  (all or some) pts u/v divisions (in trees) related with the BrepFaces  eh? (your "cube" in this occasion).

Then you need to do something with the affected points (the "pattern") - maybe move them- and "combine" with the unchanged ones and do ...  some "distorted" surface and the likes.

If the above are true:

1.Spend some minutes with this attached - although it does the job (the "pattern" selection) via code and thus it could serve nothing to you other than an indication ...  but the concept is(?) the same(?).

2. 3 cases are included to get the gist of how it works (start with the 1d pts tree).

3. Notice that although the divisions (case cube) are square ... well ... observe the top/bottom "collections" (if the "pattern" is NOT square as well).

Attachments:

• Patterns_in_1d_trees_V2.gh, 138 KB

Permalink Reply by Taraneh on February 12, 2016 at 11:23am

Actually not projecting but putting it on a all 6 faces of the box in a way that i can use my control points which i made on my 3D pattern...
I tried to figure it out by bounding box and box morph, but the pattern comes correctly just on one face and on the other faces there's just a little part of my pattern , and the other problem is that in this way i dont know how to keep my control point for the pattern on each of my cube faces ...

Permalink Reply by peter fotiadis on February 12, 2016 at 11:57am

Judging from your answer ... er ... it's obvious that the def provided doesn't help you (wonder why: it does what you want). BTW: 3D pattern? what means this? (patterns on division points are rather "2D").

To give you a hint more: after dividing ALL the BrepFaces (the def does this) you select (ditto) points according some logic (let's call it" "pattern") and then you do whatever you want with these ... and after that you "combine" them with the other (not modified) points and finally you create some surface by points.

However: IF the edges of the Brep are desired to stay "the same" (or you want a closed Brep at the end: by joining all the newly created surfaces) the points must lie "inside" the divisions  (i.e. exclude the first and last branch and the first and last point per remaining branches).

I could very easily do it but ... solely via code and ... I guess ... that this is not what you want. 

BTW: Have you completely got the gist of the DataTrees? Without that it's rather impossible to do that type of stuff.