Brick NodeShapes that seemingly won't validate anything

[danie...@csiro.au]

Hi everyone,

I've been looking at shapes in 1.3 and 1.4RC. I think I've found a category of shape that is not going to ever be tested against any model nodes.

Here's a query...

####################

SELECT DISTINCT

  ?shape ?isRangeOf

WHERE {

 ?shape a sh:NodeShape .

 OPTIONAL{ ?isRangeOf rdfs:range ?shape } # does a property try to associate with the shape via range? (If so, this won't help targeting.)

 FILTER (

     # the shape has some triple that marks it as a validation shape, rather than an inferencing rule shape...

     EXISTS { ?shape sh:property ?p }

     ||

     EXISTS { ?shape sh:and ?and }

     ||

     EXISTS { ?shape sh:or ?or }

     ||

     EXISTS { ?shape sh:parameter ?parameter } 

     ||

     EXISTS { ?shape sh:sparql ?sparql }

  )

   

  # it IS a class though, so it's conceivable that you could declare something to be an instance of the shape, or have something inferenced to become it...

  FILTER(

     EXISTS { ?shape rdf:type owl:Class }

     ||

     EXISTS { ?shape rdf:type rdfs:Class }

  )

   

  # ...but nothing (other than rdfs:range maybe) points to the shape. It's an island, all on its own...

  FILTER NOT EXISTS {

    ?s1 ?p2 ?shape

        FILTER(?p2 != rdfs:range)

  }

              

 # ...and there's no way for instance to match this shape by virtue of their type as brick:Entity subclass, or targeting, or SHACL-AF features, or whatever...

 FILTER (

   !(

     EXISTS { ?shape ((^owl:equivalentClass) | rdfs:subClassOf)* brick:EntityPropertyValue }

     ||

     EXISTS { ?shape ((^owl:equivalentClass) | rdfs:subClassOf)* brick:Entity }

     ||

     EXISTS { ?shape sh:target ?explicitTarget }

     ||

     EXISTS { ?shape sh:targetClass ?tc }

     ||

     EXISTS { ?shape sh:targetSubjectsOf ?ts }

     ||

     EXISTS { ?shape sh:targetObjectsOf ?to }

     ||

     EXISTS { ?propertyShape sh:node ?shape }

     ||

     EXISTS { ?viaClass sh:class ?shape }

     ||

     EXISTS { ?viaCondition sh:condition ?shape }

     ||

     EXISTS { ?viaObject sh:object ?shape }

   )

 )

}

ORDER BY ?shape

####################

This will find things like bsh:CO2_ConcentrationShape (1.3) or CO2_ConcentrationQuantityShape (1.4). The problem is that I can't see how a SHACL engine will ever come to test such a shape on any target. Have I misunderstood something about how SHACL works, or is this an oversight in the Brick schema? 

Best regards, 
Daniel Hugo.

[danie...@csiro.au]

Hi again, 

Sorry, I should have mentioned scale... I found 121 shapes like this in 1.3, and 99 in 1.4RC. 

Best regards, 
Daniel Hugo.