- From: Andy Seaborne <andy@apache.org>
- Date: Wed, 9 Mar 2016 21:47:59 +0000
- To: public-sparql-dev@w3.org
On 09/03/16 17:08, Bill Roberts wrote:
> Hi list
>
> I’m not sure of all the background to this, but do these changes to the SPARQL 1.1 tests mean that a SPARQL 1.1 implementation should treat "foo" and "foo"^^xsd:string as equal ?
>
> Thanks
>
> Bill
>
>
This isn't a SPARQL thing, it's an RDF thing. SPARQL 1.1 is strictly RDF
1.0 (SPARQL 1.1 REC predates RDF 1.1 REC).
DATATYPE("foo") is xsd:string even in SPARQL 1.0.
DATATYPE("foo"@en) is rdf:langString in SPARQL 1.1 as it made it in time.
Those were possible to do - flipping to "foo" not just equal to
"foo"^^xsd:string but being the same RDFterm was not because it's about RDF.
sameTerm("foo", "foo"^^xsd:string) is false for RDF 1.0 and true for RDF
1.1.
In keeping the SPARQL test suite up-to-date, the CG has applied RDF 1.1
to the tests. i.e. the tests the CG maintains are SPARQL 1.1 + RDF 1.1
system.
Andy
Received on Wednesday, 9 March 2016 21:48:29 UTC