Added subset and index types to reaction identifier

[bdeadman]

As proposed in #766.

Added SUBSET and INDEX types to reaction identifiers. This allows grouped and ordered relationships between reactions to be described.

Example usage:
A dataset containing several stages [SUBSET] of a DOE campaign, where the run number [INDEX] is explicitly documented.

An HTE dataset containing multiple plates [SUBSET] with well ids [INDEX] designating the spatial relationship of reactions in the plate.

[bdeadman]

I need to check my setup and compile the pb wrappers again. It looks like protoc-gen-ts (installed via npm using instructions at https://github.com/thesayyn/protoc-gen-ts) is backing up the old dataset_pb.ts and reaction_pb.ts files to dataset_pb.d.ts and equivalent, and creating a new dataset.ts and reaction_pb.ts file with the new enum options, but missing the ORD copyright statement.

[connorcoley]

In the vessel message, we already have position/row/column to fill the role of the “index” you’re proposing. I think those are redundant but please correct me if not! Sent from my phone; please excuse my brevity

…

On Wed, Mar 19, 2025 at 13:23 Ben Deadman ***@***.***> wrote: I need to check my setup and compile the pb wrappers again. It looks like protoc-gen-ts (installed via npm using instructions at https://github.com/thesayyn/protoc-gen-ts) is backing up the old dataset_pb.ts and reaction_pb.ts files to dataset_pb.d.ts and equivalent, and creating a new dataset.ts and reaction_pb.ts file with the new enum options, but missing the ORD copyright statement. — Reply to this email directly, view it on GitHub <#767 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABAEXJUA5GHS654BTMYU7W32VGRZDAVCNFSM6AAAAABZLMBZROVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMZXGQ3TCNBXG4> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> [image: bdeadman]*bdeadman* left a comment (open-reaction-database/ord-schema#767) <#767 (comment)> I need to check my setup and compile the pb wrappers again. It looks like protoc-gen-ts (installed via npm using instructions at https://github.com/thesayyn/protoc-gen-ts) is backing up the old dataset_pb.ts and reaction_pb.ts files to dataset_pb.d.ts and equivalent, and creating a new dataset.ts and reaction_pb.ts file with the new enum options, but missing the ORD copyright statement. — Reply to this email directly, view it on GitHub <#767 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABAEXJUA5GHS654BTMYU7W32VGRZDAVCNFSM6AAAAABZLMBZROVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMZXGQ3TCNBXG4> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[bdeadman]

I agree that reactor position/row/column is a better place to put the positional identifiers. I'll have to start using and recommending that label! I do however still think there is a use case for an index or order identifier for reactions performed in an explicit order, but where the exact times are not known. I'm currently making CUSTOM identifiers for 'run number' in a DOE and transient flow datasets.