Re: Can the default graph be defined as the union of all named graphs?

[Andy Seaborne]

On 30/04/15 11:46, Olivier Rossel wrote:
Please tell me if I am wrong:
As far as i understand, there is no single query that will, for sure,
list all triples of an endpoint once.
(no matter which endpoint it is, and no matter how it is configured).

This will do it:

SELECT DISTINCT ?s ?p ?o {
  { ?s ?p ?o }
  UNION
  { GRAPH ?g { ?s ?p ?o } }
}

and will do it for in any situation (if it executes - it is a 
potentially very expensive DISTINCT).  The query is trying all 
possibilities of where a 3-tuple of S,P,O is.

        Andy


On Wed, Apr 29, 2015 at 10:59 PM, Andy Seaborne <a...@apache.org> wrote:
In default union graph mode, the "graph = set of triples" is handled
correctly.  Triples appear only once.

SELECT * { ?s ?p ?o }

It is not the same as:

SELECT ?s ?p ?o {
   GRAPH ?g { ?s ?p ?o }
}

         Andy


On 29/04/15 18:04, Olivier Rossel wrote:

Then what is the standard query to list all triples of the store only
once?

On Wed, Apr 29, 2015 at 5:31 PM, Daniel Hernández <dan...@degu.cl> wrote:

Thanks! I have to ran queries with property property paths against an
TDB that have several named graphs. I need that named graphs look as
they where a single graph.

Daniel

On Wed, 2015-04-29 at 10:20 +0100, Andy Seaborne wrote:

On 28/04/15 22:34, Daniel Hernández wrote:

hi all,

I have a dataset that have millions of named graph and I want to
query a property path against the union of all. It is not practical
to list all named graphs with FROM clauses, because they are millions.
Thus, I wan't to know if it is possible to set fuseki to assume
that the default graph is the union of all named graphs.

Thanks,
Daniel


Yes (if TDB backed).

Either in the assembler for the TDB dataset, or "--set
tdb:unionDefaultGraph=true" what starting Fuseki.


https://jena.apache.org/documentation/tdb/configuration.html#query-of-the-union-of-named-graphs

Or query graph <urn:x-arq:UnionGraph>

        Andy