Identifiers

Identifiers are intended to formalize representation, classification, testing, and validation of unique identifers used by external systems and platforms. Having this functionality embedded in datacatalog assists with data integration between it and data representations in other systems. Currently, the following identifiers are supported:

Platform	Identifier	Description	Module
Abaco	actorId	Identifies a distinct Abaco actor	datacatalog.identifiers.abaco.actorid
	executionId	Identifies a distinct execution of an Abaco actor	datacatalog.identifiers.abaco.execid
	nonceId	Identifies a distinct temporary API key for an Abaco actor	datacatalog.identifiers.abaco.nonceid
	appId	Identifies an Agave API app	datacatalog.identifiers.agave.appid
Agave	UUID	Internal, type-spaced identifier for an app, file, job, notification, metadata object, or system	datacatalog.identifiers.agave.uuids
Data Catalog	typedUUID	Internal, type-spaced identifier for any managed object in the system	datacatalog.identifiers.typeduuid

Add a new identifier type

Extending datacatalog.identifiers with a new type is a matter of cloning the template directory in the identifiers directory and adapting it to implement a new identifer. The, add unit tests, ensuring that the JSON schema defining the new identifier(s) is generated and validates properly.

Baseline

Each identifier module must implement the following:

• generate - create a new ID (if possible, otherwise returns mock())
• get_schemas - return a dictionary of JSON schemas
• mock - generate or return a valid ID that does not refer to a known entity
• validate - assess whether a string represents a valid instance of the type