Closed majidaldo closed 10 months ago
I believe it's because of the mistake that basically 99.5% of SHACL-SPARQL users are making, namely to misunderstand how sh:prefixes works.
sh:prefixes rdf:, brick: ;
sh:select "SELECT $this WHERE { $this rdf:type/owl:deprecated true }" ] ;
would look for prefix definitions the given URIs, not use them. But at rdf: there is no prefix declaration. In the brick schema, it needs to be
sh:prefixes <https://brickschema.org/schema/1.3/Brick>
because that is where the sh:declare triples are.
See https://www.w3.org/TR/shacl/#sparql-prefixes - this needs to be improved in SHACL 1.1 but it is what it is now.
Thanks!
I'm not sure if this is a question about the expected behavior of a shacl processor or an issue (perhaps related to #53 ).
I get an unresolved prefix error if I remove the below line. What the processor is doing is fetching the ontology from the internet instead of using the prefixes that are (already) in the file. At this time, the internet file did not have the prefixes so it was failing..
Here's the minimal shape file but it's not reproducible because of the web address. shapes.ttl
error if offending line removed.