SPARQL: Specify datatype in INSERT query - sparql

Question

Let's same I want to add via a SPARQL query these triples to my graph
<?xml version="1.0"?>
<rdf:RDF
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:my="http://example.com/ontology#">
<rdf:Description rdf:about="http:/example.com/entity/thisEntity">
<rdf:type rdf:resource="http://example.com/ontology#Identification" />
<my:someID rdf:datatype="http://www.w3.org/2001/XMLSchema#string">003</my:someID>
<my:hasSource rdf:resource="http:/example.com/entity/thisSource" />
</rdf:Description>
</rdf:RDF>
How am I supposed to specify the data type of the third triple?
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX my: <http://example.com/ontology#>
INSERT DATA {
<http:/example.com/entity/thisEntity> rdf:type <http://example.com/ontology#Identification>;
my:hasSource <http:/example.com/entity/thisSource>;
my:someID "003"#xsd:string.
}

Answer 1

You're almost there. The # is for language tags. Datatypes are appended by using the ^^ separator instead. So use "003"^^xsd:string.
You will also need to add the namespace prefix to your update:
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
As an aside: in RDF 1.1, xsd:string is the default datatype for any literal that does not have an explicitly defined datatype nor a language tag. So "003" and "003"^^xsd:string are both the same, string-typed, literal.