- From: Andy Seaborne <andy.seaborne@talis.com>
- Date: Wed, 03 Feb 2010 14:02:51 +0000
- To: Steve Harris <steve.harris@garlik.com>
- CC: SPARQL Working Group <public-rdf-dawg@w3.org>
On 03/02/2010 12:01 PM, Steve Harris wrote: > On 3 Feb 2010, at 09:39, Andy Seaborne wrote: > >> To get to a final syntax, the things I know need doing to [1] are: >> >> + SPARQL Update syntax >> >> + Mandatory AS (it's optional currently) >> >> + ! in property paths, if we include that. >> >> + DISTINCT in all built-in aggregates >> >> + Finalize the aggregate keywords. >> >> + TF-LIB: Constructors for terms [2] >> >> IRI(string) -> IRI > > Should that be > IRI(string or IRI) > to make it the parallel of STR() and presumably there will be an alias > URI(...)? > > Otherwise you'd have to do IRI(STR(?x)) to be sure you'd get a URI out > from URI in. Seems reasonable - n harm extending the function signature. Otherwise it's: IF(isURI(?x),?x, URI(?x)) >> BNODE() -> fresh blank node >> BNODE(string) -> same blank node as other use of BNODE(string) >> LITERAL(str) -> >> LITERAL(str, IRI) -> >> LITERAL(str, string) -> >> >> + TF-LIB: IN and BETWEEN >> (COALESCE and IF are already in the grammar) > > I've never used BETWEEN, so I'm not sure what it does. As it's just syntactic sugar, and the strangeness of the boundary, I'm happy not to have it. I just recall it being mentioned. Andy > > - Steve >
Received on Wednesday, 3 February 2010 14:02:53 UTC