A JSON-LD context can be used to specify the range of a property. E.g., the following stats that the range of rdf:value consists of integers:
{
"@context": {
"rdf": "http://www.w3.org/1999/02/22-rdf-syntax-ns#",
"xsd": "http://www.w3.org/2001/XMLSchema#",
"rdf:value": { "@type": "xsd:integer" }
},
"rdf:value": "1"
}
In RDF modeling it is common to use different ranges for different uses of rdf:value. E.g., the following expresses that an object costs €2,50 and has temperature 28.2 ℃ (using Turtle notation):
_:1 ex:price [ rdf:value "2.50"^^xsd:decimal ; ex:unit ex:euros ] ;
ex:temperature [ rdf:value "28.2"^^xsd:float ; ex:unit ex:degreesCelsius ] .
How do I describe this in terms of a JSON-LD context? It seems to me that I need property paths (borrowing a concept from SPARQL) as keys, specifically the following for the current example:
"ex:price/rdf:value": "xsd:decimal"
"ex:temperature/rdf:value": "xsd:float"
Is there a way to specify this in JSON-LD?
You can also nest @context to specialize/override properties. To take your example:
{
"@context": {
"rdf": "http://www.w3.org/1999/02/22-rdf-syntax-ns#",
"xsd": "http://www.w3.org/2001/XMLSchema#",
"rdf:value": { "@type": "xsd:integexr" }
},
"rdf:value": "1",
"ex:price": {
"@context": {
"rdf:value": { "@type": "xsd:float"}
},
"rdf:value": "35.3"
},
"ex:temperature": {
"@context": {
"rdf:value": { "@type": "xsd:decimal"}
},
"rdf:value": "2.50"
}
}
You can experiment with this in the JSON-LD Playground.
Another approach is to use custom properties that all map to one @id (rdf:value) but with different datatypes:
{
"@context": {
"rdf": "http://www.w3.org/1999/02/22-rdf-syntax-ns#",
"xsd": "http://www.w3.org/2001/XMLSchema#",
"value_integer": {
"@id": "rdf:value",
"@type": "xsd:integer"
},
"value_float": {
"@id": "rdf:value",
"@type": "xsd:float"
},
"value_decimal": {
"@id": "rdf:value",
"@type": "xsd:decimal"
}
},
"value_integer": "1",
"ex:price": {
"value_decimal": "35.3"
},
"ex:temperature": {
"value_float": "2.50"
}
}