Awesome

SARQ - Free Text Indexing for SPARQL

SARQ is a combination of ARQ and Solr. It gives ARQ the ability to perform free text searches using a remote Solr server. Lucene indexes in Solr are additional information for accessing the RDF graph, not storage for the graph itself.

This is experimental (and unsupported).

How to use it

This is how you build an index from a Jena Model:

IndexBuilderModel builder = new IndexBuilderString("http://127.0.0.1:8983/solr/sarq");
builder.indexStatements(model.listStatements());
builder.commit();

This is how you configure ARQ to use Solr:

SARQ.setDefaultIndex(builder.getSolrServer());

This is an example of a SPARQL query using the sarq:search property function:

PREFIX sarq:     <http://openjena.org/SARQ/property#>
SELECT * WHERE {
    ?doc ?p ?lit .
    (?lit ?score ) sarq:search "+text" .
}

Acknowledgement

The design and part of the code has been taken from LARQ, see:

• http://openjena.org/ARQ/lucene-arq.html

TODO

• Fix the failing test and add more tests. [DONE]
• Double check the id as unique key, does it make sense? [DONE]
• Consider using multiValue="true" when index by subject, is it possible/useful?
• Test with Solr, add id=x title=foo then add id=x title=bar... if title is multi value field, what happens?
• Add required="true" in the schema.xml where appropriate?
• Add custom rank field and how to combined it with existing rank?
• Investigate allowDups=true|false, when is it appropriate/safe to use?
• Use EmbeddedSolrServer for testing
• ...