reduce dependence on any_of in range constraints - all planned proces…

[sierra-moxon]

…ses can be used in a 'was_generated_by' biosample slot

[github-actions]

PR Preview Action v1.4.8
🚀 Deployed preview to https://microbiomedata.github.io/nmdc-schema/pr-preview/pr-2246/
on branch gh-pages at 2024-11-01 22:22 UTC

[aclum]

Is this sufficient or do we need a grouping class?

[sierra-moxon]

@aclum—I looked at the descendants of PlannedProcess and we probably should discuss on a metadata call. From the definitions of the child classes of PlannedProcess, it seems that any of them could be the value of was_generated_by even if we don't specifically have data for them yet, but if we need to refine definitions to make this more restrictive, we certainly can. For this PR I tried to keep it simple and just "go up a level" in the hierarchy for the range constraint.

[sierra-moxon]

from modeling call:

• @kheal - would like to review so that we can see slot_usage is correctly documented for the usages of was_generated_by (if we were don't do this, it might be messier in viz)

[sierra-moxon]

from DM call:
the grouping here is that these two create a "dataobject", e.g. a "DataOutputter" class.

[turbomam]

DataEmittingProcess?

[kheal]

@sierra-moxon - I think we should put a pin in trying to resolve the any_ofs related to WorkflowExecution until #2238 is addressed. I suspect that when we add ranges to the slot_usages, it will create several more any_of instances regarding the WorkflowExecution classes.

[sierra-moxon]

Can we document the new ones we need to add? It would be my preference if we could avoid adding to the any_of paradigm.