Is it possible to create/edit/delete ontology objects from Pyspark transforms? If yes, then what would a dummy example transform look like?
Hi @tobyl,
Outside of updating the backing dataset of an object type (which can lead to conflicts as soon as there are edits on the objects), the only way to safely interact with objects from pyspark transforms is via the Actions API within an external transform.
Hope that helps.
If you’re looking for a smoother developer experience than trying to make Python Requests calls to the various ontology APIs directly, you could consider creating a Developer Console application that includes all the various objects types and action types that you may wish to interact with, generate a python OSDK package for it, then import and use the generated library in your python transform.