Describe CWL workflows as RDF

[vemonet]

cwltool offers the possiblity to print a representation of the CWL workflow and tools using --print-rdf
But the generated triples are not really nice, using the absolute local filepath to describe the CWL workflows.

We already implemented the step required to do it, and it works fine on GraphDB, but Virtuoso doesn't accept those triples

Uploading the RDF automatically generated by CWL to describe the workflow and its steps works on GraphDB, but fails on Virtuoso

From my understanding CWL generates the RDF using schema-salad
Error message:

at org.eclipse.rdf4j.http.client.SPARQLProtocolSession.execute(SPARQLProtocolSession.java:1123)
        at org.eclipse.rdf4j.http.client.SPARQLProtocolSession.executeNoContent(SPARQLProtocolSession.java:1073)
        at org.eclipse.rdf4j.http.client.SPARQLProtocolSession.sendUpdate(SPARQLProtocolSession.java:407)
        at org.eclipse.rdf4j.repository.sparql.query.SPARQLUpdate.execute(SPARQLUpdate.java:43)
        ... 5 more
[job step6-upload-cwl-rdf] completed permanentFail

We should look into a better way to represent it

See with @rcelebi to implement FAIR workflows recommendations