psiotwo / ontology-version-extractor Goto Github PK

• create a CLI with two commands - extract (to fetch the data) and transform (to generate the HTML/CSV output)
• produce a TTL intermediate representation of the extracted data.

As per @matentzn's request:

Categorize version IRIs into the following groups:

• Standard OBO version IRI 1: http://purl.obolibrary.org/obo/ID/releases/YYYY-MM-DD/ID.owl
• Standard OBO version IRI 2: http://purl.obolibrary.org/obo/ID/YYYY-MM-DD/ID.owl
• Non-standard OBO version IRI: http://purl.obolibrary.org/obo/ID/ARBITRARY-VERSION/ID.owl
• Non-OBO version IRI: any IRI (https://bio.scai.fraunhofer.de/ontology/epilepsy/2021-05-28/EPIO.owl)
• No version IRI, but version INFO
• No version information at all
• Other

Currently only head of the ontology is fetched. However, sometimes the versionInfo and the versionIri is present at the end of the document.
A/C:

• support fetching also the tail of the document
• join both to perfom extraction on head+tail together
• try to use range request wherever possible

Currently, version IRI is extracted/generated.

To support more refined comparison and also more user-friendly visualization, a version (number, date) should be extracted too.

Ontology with unresolvable imports is unprocessable, like in CIDO-ontology/cido#41

As an ontology curator, I want to be informed that my ontology has unresolvable imports.

As an ontology user, I want to be informed, without downloading it (it might be huge)