Uncommented existing test_cli for "oaei-ordo-hp"

[hrshdhgd]

The test for parsing oaei-ordo-hp.rdf was commented. This PR just uncomments it and wires is properly in test_cli.py to verify if it passes.

[hrshdhgd]

@matentzn : ok, so I'm running :sssom parse tests/data/oaei-ordo-hp.rdf -o test.tsv --input-format alignment-api-xml

and the test.tsv begins

# curie_map:
#   HP: http://purl.obolibrary.org/obo/HP_
#   orphanet.ordo: http://www.orpha.net/ORDO/Orphanet_
#   owl: http://www.w3.org/2002/07/owl#
#   rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns#
#   rdfs: http://www.w3.org/2000/01/rdf-schema#
#   semapv: https://w3id.org/semapv/
#   skos: http://www.w3.org/2004/02/skos/core#
#   sssom: https://w3id.org/sssom/
# license: https://w3id.org/sssom/license/unspecified
# mapping_set_id: https://w3id.org/sssom/mappings/105b111d-1f45-44e3-bb6d-8c33670537b7
# object_source: http://purl.obolibrary.org/hp.owl
# subject_source: http://purl.obolibrary.org/ordo.owl
subject_id	predicate_id	object_id	mapping_justification	confidence
orphanet.ordo:100047	owl:equivalentClass	HP:0100681	semapv:UnspecifiedMatching	0.27449
orphanet.ordo:100069	owl:equivalentClass	HP:0030219	semapv:UnspecifiedMatching	0.875
orphanet.ordo:100088	owl:equivalentClass	HP:0002890	semapv:UnspecifiedMatching	0.875

Is the object_source value incorrect? The reason I ask is test_conversion.py is throwing an error stating that the value is incorrect.

[matentzn]

What is your plan here re curification if an URI prefix is not in the prefix map?

[hrshdhgd]

I may have to either brute-force the CURIE or think if this info is secured via a curie_map in one of the yaml files?

[matentzn]

I would vote for leaving the URIs as is if the URI prefix is not known to the prefix map. We should produce a single warning at the end of the pipeline which lists the first 5 URIs that didn't have a proper CURIE, and log a warning "the extracted file is not valid SSSOM because some of the URIs could not be converted. Please provide appropriate URI prefixes and repeat the process".

Charlie prefers to not process the node at all if the URI prefix is unknown; this is an alternative, but only if we can overwrite it with a flag, and then we are back to the first point.

[hrshdhgd]

That is the problem. The sssom definition for subject_source and object_source is EntityReference. Does this mean only CURIEs? uriOrCurie might be a better solution?

[hrshdhgd]

Oh you don't want it to be a URI at all.

[matentzn]

I think a URI is a curie, where the prefix is http and the URI expansion http so everwhere where there is an entity reference, you should be able to drop a URI.

Its fine to write a URI, just also write the warning I mention above.

[matentzn]

What is missing in this PR?

[hrshdhgd]

What is missing in this PR?

Just the curiefication

[matentzn]

Looks great, two minor comments!