Ideed yes, it was my bad. Thanks a lot for all those precisions. Best.
Le 8 oct. 2014 à 16:29, Joshua TAYLOR <[email protected]> a écrit : > On Wed, Oct 8, 2014 at 10:10 AM, Damian Steer <[email protected]> wrote: >> On 08/10/14 15:06, Julien Plu wrote: >>> I think I found a little bug in the SPARQL parser for Jena 2.12.0. When I >>> run the query : >>> >>> SELECT count(distinct ?s) as ?count WHERE {?s ?p ?o} >> >> You need parentheses around the ... as ...: >> >> SELECT (count(distinct ?s) as ?count) WHERE {?s ?p ?o} > > Damien Steer's answer is correct. Virtuoso (the endpoint that DBpedia > uses) allows a number of things that aren't actually allowed by the > SPARQL standard. There's a SPARQL validator at sparql.org that might > be useful. It's based on Jena, though, so if there *were* a bug, the > validator might not catch it. When in doubt, though, you can check > the SPARQL Grammar [1]. In this case, the important rule > > [9] SelectClause ::= 'SELECT' ( 'DISTINCT' | 'REDUCED' )? ( ( Var > | ( '(' Expression 'AS' Var ')' ) )+ | '*' ) > > which shows that parentheses are needed around the "expression as var" > construction. > > //JT > > [1] http://www.w3.org/TR/sparql11-query/#sparqlGrammar > > > > > -- > Joshua Taylor, http://www.cs.rpi.edu/~tayloj/
