The other post got taken down so I’ll post a sample geometry script here. So for Model 1, index 8 and 9 is identical to Model 2’s, index 6 and 10 with index 9 being in the opposite direction. But index 10 only exists in model 2. If I compare the vectors through, index 8 and 9 from Model 1 will match index 10 but for my situation I need it to not exhibit a match. Hopefully that makes sense.
sample_script.gh (13.5 KB)
Yes, that’s exactly the problem 
“As for the list of vectors that share the same direction, they will have repeats because vectors do not have anchors. They are a single point value that when measured from the origin represent direction and magnitude.” For my situation it needs to map in space and map in vector.
Well, then you just need to group the data first for direction using vector comparison, then further sort by location using the midpoint. But as I said earlier, you could just compare endpoints to find same lines. Does the direction of same lines matter? I.e., are lines that are the same but have their direction flipped supposed to be considered the same?
I’ve got multiple lines connecting into 1 point at times so the end point method wouldn’t be feasible. The directions matter for my project, unfortunately. I’m working with a lot more data than the sample script geometry I’ve provided as well so my heads messy with the logic.
… plugging your sample geometry into my previous solution works just fine. I sent you a video where I plug your sample geometry into that script to prove it.
It won’t matter if multiple lines connect at the same point. If two lines being compared do not share the same start point and the same endpoint, they are not the same line!
I think I’ve found a solution using your suggestion of end points! I’m going to test it on a larger scaled model to ensure that I’ve set it up w/o any issues 
Good. If that didn’t work, here is a robust solution:
that can take any number of models and compare between them for unique lines. This took longer to finish than I care to admit. I don’t know if it could be refactored better. Certainly, it could be expanded to output an index mapping to the original set of lines. In a corner of the canvas you’ll find it reduced to a cluster:
Oh wow! Thank you! I’ll take a look
