dear all, I changed the datasets, but instead of using the remote one as default local graph, I simply removed a triple (:c a foaf:Person) in the default local graph and I added another one to the remote endpoint (:c foaf:knows :a), so BINDINGS is useful. Carlos 2011/8/2 Gregory Williams <greg@evilfunhouse.com> > On Aug 1, 2011, at 1:32 PM, Carlos Buil Aranda wrote: > > > I updated the fed document with the changes you sugested. Regarding the > BINDINGS section I completed it a bit more with the queries missing. > > I don't think the BINDINGS example in section 2.4 is particularly helpful. > The example shows a situation where the BINDINGS clause isn't actually > restricting the results that are sent back from the remote endpoint because > the data on that endpoint is a subset of the BINDINGS values. I think it > would be better if the data was reversed: :a and :b in the local dataset and > :a, :b, and :c on the remote endpoint, with the BINDINGS clause restricting > the pattern matching on the remote endpoint to just :a and :b. > > thanks, > .greg > > >
Received on Thursday, 4 August 2011 14:21:14 UTC