- From: Steve Harris <swh@ecs.soton.ac.uk>
- Date: Wed, 4 Aug 2010 14:03:43 +0100
- To: Matthew Pocock <matthew.pocock@ncl.ac.uk>
- Cc: Lee Feigenbaum <lee@thefigtrees.net>, "public-sparql-dev@w3.org" <public-sparql-dev@w3.org>
Matthew,
Have you looked at the complexity limiting in 4store? This may be the source of the difference.
- Steve
On 2010-08-04, at 13:57, Matthew Pocock wrote:
> Hi,
>
> Using Joseki's web form, I've run the same queries. In this case, I get the same number of results for:
>
> ?a rt:sim ?b . ?b rt:bi_to ?c . ?a rt:bi_to ?c
>
> and the equivalent query using OPTIONAL/BOUND/FILTER:
>
> ?a rt:sim ?b . ?b rt:bi_to ?c
> . OPTIONAL { ?a rt:bi_to ?c2 . FILTER(?c = ?c2) }
> . FILTER (BOUND(?c2))
>
> at 2,734 hits each. This differs from 4store which appears to be returning 5,800 hits for the first (simple) query but 2,734 for the latter (complex). By eye, it looks like may of the 'extra' results from 4store contain nulls for ?b which smells buggy to me.
>
> Matthew
Received on Wednesday, 4 August 2010 13:04:31 UTC