- From: Seaborne, Andy <andy.seaborne@hp.com>
- Date: Thu, 26 Oct 2006 11:25:53 +0100
- To: "Arjohn Kampman" <arjohn.kampman@aduna-software.com>
- Cc: <public-rdf-dawg-comments@w3.org>
-------- Original Message -------- > From: Arjohn Kampman <> > Date: 23 October 2006 16:15 > > Dear RDF DAWG, > > If I understand the SPARQL grammar correctly, it allows white space to > be used between the string part of RDF literals and the optional > language tag or datatype following it. Is this reading correct? And if > so, is this intended to be like this? Neither N-Triples nor Turtle > allow this. > > -- > Arjohn Kampman, Senior Software Engineer Aduna - Guided Exploration > www.aduna-software.com Arjohn, Your reading is correct (same for datatype with ^^ as well). The reason is that separate tokens are generated for lexical form, language tag and datatype for a constant literal in a query. Having the three part separately can be helpful in building whatever the internal representation of an RDF literal is. Lexers will return the matching string - if it is the whole of the literal, lang or datatype, then that whole string needs to be split up. This includes the prefixed name for the datatype so turning prefixed names into IRI would need to happen in two places. Currently it can done in one place, the IRIref rule). Lang tags could be treated differently to datatypes. Andy
Received on Thursday, 26 October 2006 10:26:07 UTC