Shatter - strange behaviour

Hello,

the shatter components just stole quite some time from me due to some very strange behaviour. I reduced my problem to the following: There are some lines given in a data tree but I do not provide a splitting parameter for each. So if there is no data given for a particular branch, no shattering should be done. However, the shatter components then pulls data from the next branch. I don’t think this is intended behaviour or am I missing something here?

Shatter - strange behaviour.gh (13.7 KB)

There would be a data branch matching rule is going on just like “Longest list Matching”. This could be a workaround using Split Tree.


Shatter - strange behaviour_re.gh (15.6 KB)

Shatter won’t work if you don’t provide a “t” parameter, if it is working anyway it is because it is picking the parameter from another branch.
If you have just one parameter for each line you can work with lists:

1 Like

@HS_Kim @maje90 Thanks a lot for your thoughts. Right now I do filtering based on if there is any intersection and then pipe only the curves with intersections. Afterwards I have to merge for later processing. So quite some components needed to shift around shatter.
As seen in the picture, shatter can handle zero length branches. Probably smart picking 0. So for me this “picking parameters from another branch” threw me off in debugging and I think it is creating more trouble than solving something. Or you think there is any benefit to it?

Ok, sorry, I have to correct myself. Null items are proccessed as aspected. Zero length branches are ignored, so no “smart picking” on those either.