Avoiding intersection of geometries

hi guys
i have an algorithm that scales a geometry to random sizes and arranges them side by side without any intersections.
i am wondering if there is any easier and shorter way to getting this result, i think it should be a way using of galapagos or hoopsnake, but i dont have any idea how its gonna be,
cheers.


lookin for a shorter way.gh (70.1 KB)

Here’s one try.

There is one thing I could not figure out:

  • how to pull the X length of the arch from the geometry :thinking:

scale_arches_randomly.gh (46.0 KB)

This is my try using bounding box, works quite well on the x axis and for this geometry. (in the picture it misses a flatten after the last merge, updated the file)

lookin for a shorter way_re.gh (71.4 KB)

2 Likes

Hi Ivelin
thanks for your try!:heart:
i got x length of the geometry from deq brep, select outside vertice and then distance,
but @danielbent used deq boundingbox and deqbox that works better!
your definition looks interesting but does not consider X domain that looks to achieve (500 unit) with 10 geometry.
or i dont understand it well!:roll_eyes:

thanks joseph for your reply
i saw your replys before and i know you as a pro in this field.
but i think you dont get exactly my purpose,
i dont get yours! :sweat_smile:

thanks daniel!
well done!
im really excited about your definition.
its something like my way but much correcter and shorter!!
thanks again:)

just one another question, is there any chance to get this result with galapagos?

I don’t think galapagos is useful in this case. You use galapagos when you don’t know your input parameters, while in your case you know it’s random scale and moved so they align. But hey, no harm trying to use it if you have the time.

1 Like

Nice. Same idea, simplified and aligned with X = 0:


space_along_curve_2019Jul16a.gh (21.7 KB)

2 Likes

@Joseph_Oster Nice, thanks for the simplification of my solution :slight_smile:

brilliant!
thanks a lot dear joseph.