Hi
On Wed, Jun 20, 2012 at 2:02 PM, seralf <[email protected]> wrote:
> Hi Rupert, this is really interesting... is it possibile to have some
> simple example?
>
This is how it should work (should because I have only tested this
with Jena TDB because I have no virtuoso server running on my machine)
0) Have a virtuoso server running
1) install the two virtuoso bundles [2] in your OSGI environment
2) configure the virtuoso server (e.g. via the Config tab of the Felix
WebConsole)
3) create an instance of "Apache Stanbol Entityhub Yard: Clerezza Yard
Configuration"(org.apache.stanbol.entityhub.yard.clerezza.impl.ClerezzaYard)
* make sure that the configured "Graph URI" points to an existing
Graph on the virtuoso server
* check for a INFO level logging such as "... (re)use existing Graph
<{graph-uri}> for Yard {clerezza-yard-name}"
4) configure a "Apache Stanbol Entityhub Cache Configuration"
(org.apache.stanbol.entityhub.yard.clerezza.impl.ClerezzaYard)
* just add the ID of the created ClerezzaYard in the "Yard" field of the Cach
5) Configure a "Apache Stanbol Entityhub Referenced Site
Configuration"(Apache Stanbol Entityhub Referenced Site Configuration)
* choose a ID (used in the RESTful Service endpoint URL),
* add a name
* set Cache Strategy to "All (prepared cache ..)"
* set Cache ID to the ID of the create ClerezzaYard
After that a ReferencedSite based on the RDF data in the Virtuoso RDF
graph should be available via the entityhub.
NOTE that this feature should be used for demonstration purpose only
as there are still a lot of things to do. Most important I will need
to allow the use of Virtuoso specific SPARQL queries - currently label
based searches will use REGEX filters that are much to slow.
If you have a Virtuoso Server running and you want only use it with
the apache stanbol Entityhub you can also start at step (5) and
configure
* ID, name same as above
* Access URI: The CoolURI or SPARQL endpoint (what you prefer)
* Dereferencer Implementation: depending on the configured Access URI
* Query Service URI: the SPARQL endpoint
* Searcher Implementation: SPARQL+Virtuoso extensions
* Cache Strategy: never
* Cache ID: leave empty
This will directly connect to your Virtuoso endpoint and allow Stanbol
read only Access to the data exposed via the SPARQL interface.
best
Rupert
>
> 2012/6/20 Rupert Westenthaler <[email protected]>
>
>> Hi Sebastian
>>
>> With the implementation of STANBOL-622 [1] it is now possible to
>> configure a ClerezzaYard to use an already existing RDF graph with the
>> Entityhub. So if you add the clerezza-virtuoso bundles [2] to the OSGI
>> environment running Apache Stanbol you can now provide a configuration
>> that allows you to use existing RDF graphs managed by Virtuoso (the
>> default Triple Store used by the LOD2 [3] stack) with the Apache
>> Entityhub.
>>
>> If you have any questions or need some help please contact me. Please
>> keep the stanbol-dev mailing list in the loop as also others might be
>> interested in this
>>
>> best
>> Rupert
>>
>>
>> [1] https://issues.apache.org/jira/browse/STANBOL-662
>> [2]
>> https://github.com/enridaga/clerezza-virtuoso/tree/master/clerezza-virtuoso
>> [3] http://lod2.eu/
>>
>> --
>> | Rupert Westenthaler [email protected]
>> | Bodenlehenstraße 11 ++43-699-11108907
>> | A-5500 Bischofshofen
>>
--
| Rupert Westenthaler [email protected]
| Bodenlehenstraße 11 ++43-699-11108907
| A-5500 Bischofshofen
