- From: Richard Newman <r.newman@reading.ac.uk>
- Date: Wed, 21 Feb 2007 11:38:26 -0800
- To: "Stephane Fellah" <stephanef@imagemattersllc.com>
- Cc: <public-rdf-dawg-comments@w3.org>
Sorry for omitting that; that's already described in the SPARQL operator selection, which defines < for dateTimes, numeric values, booleans, and simple literals (strings). The additional ordering constraints are only necessary when the more specific ordering rules do not kick in. When two xsd:booleans are compared, the comparison never reaches the literal comparison stage, because op:boolean-less-than(A, B) is used instead. SPARQL defines additional ordering constraints for when these do not apply, e.g., for unknown datatypes, or for comparing an xsd:boolean to a URI. See <http://www.w3.org/2001/sw/DataAccess/rq23/rq25.html#modOrderBy> and <http://www.w3.org/2001/sw/DataAccess/rq23/ rq25.html#OperatorMapping> -R On 21 Feb 2007, at 11:26 AM, Stephane Fellah wrote: > > Richard, > > I agree with the order you suggest however I think prior doing the > comparison on the lexical form, we need to do a comparison on the > value > space (date, numbers, boolean). If the value space comparison returns > equality, then a lexical form comparison needs to be done. > In this case we ensure we have a "semantic"/mathematical ordering > of literal > value, not only syntactic. > > Stephane
Received on Wednesday, 21 February 2007 19:38:52 UTC