- From: Seaborne, Andy <andy.seaborne@hp.com>
- Date: Fri, 8 May 2009 18:28:33 +0000
- To: Lee Feigenbaum <lee@thefigtrees.net>, SPARQL Working Group <public-rdf-dawg@w3.org>
> -----Original Message----- > From: public-rdf-dawg-request@w3.org [mailto:public-rdf-dawg- > request@w3.org] On Behalf Of Lee Feigenbaum > Sent: 8 May 2009 19:22 > To: SPARQL Working Group > Subject: More subquery questions - BASE & PREFIX > > (Can you tell that I'm gearing up for that implementation I promised?) > > The following questions are mainly for existing implementors of > subquery, though if you have a strong opinion on how these points should > be specified, I'd be pleased to hear them. > > Do existing subquery implementations allow subqueries to define its own > base uri? (i.e. include a BASE clause) Do the subqueries inherit the > BASE of the parent? > > Similarly for PREFIX, can a subquery define its own prefixes? do they > override prefixes in the parent query? is it additive? (i.e. if I have > PREFIX p1: <...> in the parent and PREFIX p2: <...> in the subquery, can > I use both p1: and p2: throughout the subquery?) > > Lee In ARQ, subquery happens by allowing grammar rule "[5] SelectQuery" appear inside {} in a group (rule 20). ARQ should have also ban DatasetClause clause (FROM/FROM NAMED) in some way - it is merely ignored. So the prologue (BASE and PREFIX rules 2/3/4), are scoped for the whole query and they do not occur in the subquery. Andy
Received on Friday, 8 May 2009 18:29:50 UTC