The query engine breaks the SPARQL query down into various insturctions
(joins, etc.) and then performs the calls on the various models it is
querying, the models do some delegaion to the graph.  So a query like (it's
been a while since I wrote any SPARQL so bear with me if this is not a
correct statement)
SELECT ?a WHERE {
?a your:predicate ?foo ;
?foo a your:type

will probably create 2 queries to your graph

graphBaseFind( <ANY, a, your:type> )
followed by
graphBaseFind( < ANY, your:predicate ANY> )

but I am not certain about the exact statements.  Anyway, the query engine
in conjunction with the higher level classes/objects breaks the query down
into basic "find me this" statements which the higher level objects then
put back together unsing joins, filters, etc as per the SPARQL standard.

Claude


On Thu, Apr 3, 2014 at 11:27 PM, Shijie Jin -X (shijjin - AAP3 INC at
Cisco) <[email protected]> wrote:

> Thanks for your answer. I¹m implementing a read-only system, so I may just
> need to implement its graphBaseFind method. But one more question, when I
> look into parameter of graphBaseFind, it only provide TripleMatch, where
> does this TripleMatch come from? Is it represented for one statement of
> SPARQL query?
>
> On 4/3/14, 2:28 PM, "Claude Warren" <[email protected]> wrote:
>
> >I'm not familiar with HBase and it's query capability, so I am going to
> >assume that it has a query language that works something like SQL and go
> >with the standard answer:
> >extend GraphBase (com.hp.hpl.jena.graph.impl.GraphBase) as a starting
> >point.
> >
> >It is not the algebra but it will present the HBase data as a graph and it
> >will plug into the query engine and a number of other components.
> >
> >Claude
> >
> >
> >On Thu, Apr 3, 2014 at 6:12 PM, Shijie Jin -X (shijjin - AAP3 INC at
> >Cisco)
> ><[email protected]> wrote:
> >
> >> Hi All:
> >>
> >> I¹m trying to extend the ARQ execution so it can convert SPARQL query to
> >> Hbase ³scan" operation. But after reading the document(
> >> http://jena.sourceforge.net/ARQ/arq-query-eval.html) and
> >> example(src-examples/arq.examples.engine.MyQueryEngine), I still can¹t
> >> figure out where should I write my code to turn the SPARQL query (or
> >>SPARQL
> >> algebra) into Hbase query.
> >>
> >> And I¹v also read the source code of D2RQ, but it uses the old version
> >> Jena framework which has a lot of difference compare to the latest
> >>version.
> >>
> >> So anyone can give me some hints how to extend the ARQ execution
> >> correctly? Is there any good document or source code that I should look
> >> into?
> >>
> >> Thank you,
> >> Shijie
> >>
> >
> >
> >
> >--
> >I like: Like Like - The likeliest place on the
> >web<http://like-like.xenei.com>
> >LinkedIn: http://www.linkedin.com/in/claudewarren
>
>


-- 
I like: Like Like - The likeliest place on the web<http://like-like.xenei.com>
LinkedIn: http://www.linkedin.com/in/claudewarren

Reply via email to