Hi Specklers,
applicationId and location (point, line, curve) are possible to read from the Deconstruct Node but there is no way to add/ modify them through GH.
the docs cover some good ways to mutate existing objects to ensure a consistent applicationID is assigned to GH objects:
I have a few questions, ideas on this:
- in v2 i would often define the application ID from the logic within the scripts data structure, Hash of a
unique_obj_type+pathwhich would often be fairly consistent across iterations and where it isn’t probably not concerned about diffing?
-
Loading the last published model in and using the passthrough nodes is a good approach. a few issues i see is that you still need a way to re associate the new geometry with the previous loaded, probably via a closest point match or something similar. depending on the changes this could be simple or not.
-
There are likely scenarios where you want to take revit geometry and say split it into parts for analysis. This would create 3 new/ random application_ids for the new objects each time through GH. however there is a consistent logic to how these could be calculated if they could be set through GH. original_application_id + 1,2,3…
On the location data. I’m happy that we can create dataobjects constructed from multiple geometry (Brep and the base line geometry) to so they are both stored and accessible.
However there could well be times when you don’t really want the baseCurve etc rendered in the viewer but stored for future interoperability/ use but not presented.
Geometry can’t be added to properties so arn’t suitable.
A higher level object that exposes this i expect is the solution. which leads on to in v2 there was quite an extensive array of Object Kits for GH. Is there any intention currently to develop this for V3/ any developer guidance on how this could be implemented.
Sorry possible a bit stuck in my ways from v2 but hopefully some useful items for discussion.


