- From: Ian Davis <ian.davis@talis.com>
- Date: Wed, 24 Jan 2007 15:07:58 +0000
- To: Harry Halpin <hhalpin@ibiblio.org>
- CC: Dan Connolly <connolly@w3.org>, Ben Adida <ben@adida.net>, Fabien Gandon <Fabien.Gandon@sophia.inria.fr>, public-grddl-wg@w3.org
On 24/01/2007 00:58, Harry Halpin wrote: > > The use of SPARQL inequality does not work unless datatypes are > consistent. Therefore unless Embedded RDF supports data-types we should > probably move the example currently in the primer to RDFa, and test to > see if Fabien's XSLT supports use of XML Schema datatypes in RDFa. > I agree this might be a problem and it's a limitation of eRDF - no datatyping. The glean-hcal transform adds in rdf:datatype attributes for dtstart and dtend so there's a basic incompatibility there. I don't think there's a quick fix for eRDF so either we amend the primer scenarios or we look for alternate expression such as RDFa. One way might be to use Sparql constructor functions[1] and rewrite the FILTER as: FILTER ( xs:dateTime(?start1) = xs:dateTime(?start2) && xs:dateTime(?stop1) = xs:dateTime(?stop2) && ?url1 != ?url2 && ?location1=?location2) . Would that work? Ian [1] http://www.w3.org/TR/rdf-sparql-query/#FunctionMapping --
Received on Wednesday, 24 January 2007 15:48:51 UTC