ReparamSrf kind of screwing everything up

In this example, the ReparamSrf component not only gets rid of branches containing breps instead of just ignoring them, but it also gets rid of a branch containing a surface, for no apparent reason.

Pufferfish reparameterize and screw up.gh (62.8 KB)

Hi @osuire dunno how I missed this post, you can tag me next time.

You are correct there is a bug there for this unusual case, I likely just forgot to check for nulls (will have to open up the code and check). I just tested a similar situation with nulls in branches with the other reparam components and there isn’t this issue. What is happening is the null branches cause the component to fail all the way through. So yes the last surface is missing, but it isn’t randomly, its just cuz it is after the nulls before it which caused it to fail. (sometimes when making all these components stupid things like that can be missed and it is not a usual case I would have tested it for, my bad).

For now you can just use the reparam output on surface components as you have (my reparam components are just for ppl like seeing the components, similar to flatten component existing).

Alternatively, you can change the order of operations and use the repram surface before the nulls exist (but this wont get rid of the breps for you).

Thanks for reporting the bug!