Shatter output inconsistency, how to solve?

I am having some trouble with the way [shatter] delivers/numbers/organizes the shattered outputs.

When I [Shatter] a series of rotated circles into 3 items per layer, I get as a result helix-like outputs (in orange).

but

after some modification to those circles, I try to shatter again and I get a total different outcome (in orange)

I need to get the [shatter] result be like in the first image (helix-like arrangement) but after the modifications I previously mentioned.

I've been trying to solve this in many different ways but I am giving up. Any suggestions?

Attachments:

current.PNG, 144 KB
goal.PNG, 138 KB
column.gh, 21 KB

Replies are closed for this discussion.

Replies to This Discussion

Permalink Reply by Hannes Löschke on March 15, 2014 at 12:19pm

you may realize, that whatever "modifcations" you applied, may have changed the tree structure of your data.

So in order for anybody to be able to really help you, a file with your definition would be good.

Permalink Reply by Dario Vidal on March 16, 2014 at 2:05am

Indeed, I realize there is a change in the data structure but I can't see where or how to modify it.

I've attached the definition, appreciate any suggestions.

Permalink Reply by Hannes Löschke on March 16, 2014 at 6:38am

Okay, your problem here is the join component, not data structure. Join doesn't guarantee to preserve the order of the segments to join. That is, if there are several segments after the joining, you cannot assume that the first resulting segment is associated to the first item from the input list.

Looks like this also affects the seam position of the closed curve. So the start of the joined ring isn't always where the first arc in the list started.

I tried to adjust the seam to where the inner rings seam would be, but that doesn't seem to produce reliable results either. You might have to find a better way to adjust the seam or to join/shatter the segments.

Attachments:

column_seam.gh, 20 KB

Permalink Reply by Dario Vidal on March 21, 2014 at 4:32am

I'm trying to find a way to reorder/reconfigure the start point of each layer (joined curve) so the evaluation of the curve would be 'oriented' in the same way throughout the entire column, containing 3 points per curve. The seam component you suggest is promising but there are still some issues.

Thanks for your suggestion, but is there any other way to join and shatter the segments? I might have to modify the design of the column due to the modelling issues, and that is always a downside.

Permalink Reply by Danny Boyes on March 21, 2014 at 5:53am