- From: Steve Harris <steve.harris@garlik.com>
- Date: Wed, 11 Oct 2006 16:22:39 +0100
- To: "Seaborne, Andy" <andy.seaborne@hp.com>
- Cc: RDF Data Access Working Group <public-rdf-dawg@w3.org>
On 11 Oct 2006, at 15:34, Seaborne, Andy wrote:
>
> After the telecon formally ended, we were talking about BGP's and
> entailment. Just for the record, the example I gave was:
>
> Data:
>
> :x :p 1 .
> :x :p 2 .
>
> Query:
>
> SELECT * { ?x :p [] }
>
> It's the same point as:
> http://www.w3.org/2001/sw/DataAccess/tests/data/rdfSemantics/query-
> se-bN
> ode-type-var.rq
>
>
> [2006/10/10 16:50] AndyS: Example: :x :p 1 . :x :p 2 . { ?x :p [] } =>
> projection => counting is undef
> [2006/10/10 16:52] ericP: | ?x |
> [2006/10/10 16:52] ericP: | :x |
> [2006/10/10 16:52] ericP: | :x |
>
> The question I raised was whether an implementation is wrong if it
> returns two results (both ?x = :x) given we don't define counting
> except
> in the presence of DISTINCT. I don't see why an implementation should
> be forced in this one case to reduce to one result.
Certainly both my implementations produce :x twice without DISTINCT -
I would be surprised to see anything else. Though, that might be my
database inclination showing through.
- Steve
Received on Wednesday, 11 October 2006 15:22:53 UTC