type slot assertion on planned NEON process classes and processed sample

[sujaypatil96]

Fixes #1027 and #1030

[sujaypatil96]

Agenda item for 8/16 call with @turbomam:

• decide how we want to constrain the range of type slot on the NEON PlannedProcess and ProcessedSample classes.

[aclum]

@turbomam can we get this reviewed and merged? @sujaypatil96 is there corresponding migration code?

[turbomam]

This can be closed, but we need to refactor "type" slots ASAP. I have an issue for this but haven't found it yet.

[aclum]

@turbomam are you saying you'd like to close instead of merge this pull request where @sujaypatil96 has addressed some of the type slot issues?

[turbomam]

I was hoping to

• prioritize applying type, as an inherited type-designating slot (whose range would be identifiers of nmdc-schema classes) to root classes
• remove all other NMDC slots with "type" in their name. (We would keep any MIxS slots that have "type" in the name)
• replace the removed slots with xyz_category slots, whose ranges would be enumerated strings

Maybe I'm not the best judge of when to do that. If it's beneficial to start populating the string type slot for the classes in the issue/PR now, then yes, we can merge it.

[aclum]

If we can address your first bullet 'prioritize applying type, as an inherited type-designating slot (whose range would be identifiers of nmdc-schema classes) to root classes' with the december refactor we should do that in favor on merging this in. let's leave this open until we decided @lamccue @mslarae13

[mslarae13]

I'll defer to @lamccue for the prioritization & if there's time to target type for December.

As for getting it done, @turbomam can we get a list of all the slots that have "type" in the name/title that shouldn't so we can evaluate how many slots we need to change?

[lamccue]

I don't have a good understanding of the level of effort required, or the impact the change would have within the schema, of the first point, i.e.,:

1. prioritize applying type, as an inherited type-designating slot (whose range would be identifiers of nmdc-schema classes) to root classes

[mslarae13]

Thanks @lamccue ... @turbomam can we get a list of all the slots that have "type" in the name/title that shouldn't so we can evaluate how many slots we need to change?

[aclum]

Just a note that we can separate concerns here wrt to what gets done during the December refactor, we can add slot 'type' on all classes and set a range to be a class independent of changing the name of slots which currently have the word type in them (ie data_object_type).

[aclum]

Im going to close this PR since @turbomam took care of what is addressed in this PR globally in the berkeley-schema-fy24 fork. I unlinked the nmdc-schema ticket so that would remain open.