Fuseki and SDB

[Paul Taylor <pa...@yahoo.com>]

Hello,

I have been working with Fuseki and TDB for quite some time, however this weekend I have been trying to setup Fuseki with SDB. In particular, I have my application to open a connection to a local MySQL server and then to create a new database called "sdb_data". Then I create a Store object programmatically etc., a very brief snapshot of the Java code:

 StoreDesc storeDesc = new StoreDesc(DEFAULT_LAYOUT, getDBType());

 sdbStore = SDBFactory.connectStore(getSDBConnection(), storeDesc);


 ...
 sdbStore.getTableFormatter().create();

I can verify that the SDB Store is created successfully because I can see that the database "sdb_data" is created with all the necessary tables inside to store RDF. Also, I can load/query an RDF-Graph in the store without any problems or exceptions. Note that for the purposes of my application I have many RDF-Graphs stored as Named Graphs, that means that I will never have any data to the default graph.

What I would like to do is to be able to connect to the SDB store just created and query the store with Sparql from Fuseki. The problem that I have is that when I ask a query to get a list of all the Named Graphs in the Store it returns an error in console. The strange thing is that if I run this query from the Java application programmatically it works, so it must be the setup of Fuseki using the .ttl file.

SELECT DISTINCT ?g
WHERE {
  GRAPH ?g {
    ?s ?p ?o .
  }
}

Error I am getting (Error 500: Server Error):

15:51:19 INFO  [1] Query =   SELECT DISTINCT ?g  WHERE {    GRAPH ?g {  ?s ?p  ?o .    }  }
15:51:19 WARN  [1] RC = 500 : null java.lang.NullPointerException
        at com.hp.hpl.jena.sparql.engine.main.iterator.QueryIterGraph$QueryIterGraphInner.buildIterator(QueryIterGraph.java:175)
        at com.hp.hpl.jena.sparql.engine.main.iterator.QueryIterGraph$QueryIterGraphInner.nextIterator(QueryIterGraph.java:155)
        at com.hp.hpl.jena.sparql.engine.main.iterator.QueryIterGraph$QueryIterGraphInner.hasNextBinding(QueryIterGraph.java:118)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorBase.hasNext(QueryIteratorBase.java:112)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIterRepeatApply.hasNextBinding(QueryIterRepeatApply.java:81)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorBase.hasNext(QueryIteratorBase.java:112)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIterConvert.hasNextBinding(QueryIterConvert.java:59)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorBase.hasNext(QueryIteratorBase.java:112)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIterDistinctReduced.hasNextBinding(QueryIterDistinctReduced.java:54)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorBase.hasNext(QueryIteratorBase.java:112)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorWrapper.hasNextBinding(QueryIteratorWrapper.java:40)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorBase.hasNext(QueryIteratorBase.java:112)
        at com.hp.hpl.jena.sparql.engine.iterator.QueryIteratorWrapper.hasNextBinding(QueryIteratorWrapper.java:40)


Config setup:

The .ttl config file for Fuseki is as below, and just to provide as much info as I can, I am using: 
 jena-fuseki-0.2.7, jena-sdb-1.3.6.jar, jena-arq-2.10.1.jar, jena-tdb-0.10.1.jar, jena-core-2.10.1.jar, mysql-connector-java-5.1.22.jar.

Config file:

@prefix fuseki:  <http://jena.apache.org/fuseki#> .
@prefix rdf:     <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix rdfs:    <http://www.w3.org/2000/01/rdf-schema#> .
@prefix tdb:     <http://jena.hpl.hp.com/2008/tdb#> .
@prefix sdb:     <http://jena.hpl.hp.com/2007/sdb#> .
@prefix ja:      <http://jena.hpl.hp.com/2005/11/Assembler#> .
@prefix :        <#> .

   ##Setup a Fuseki server
   [] rdf:type fuseki:Server ;

   ##Service to start
   fuseki:services (
      <#sdb_service>
   ) .

   ## SDB Store
   [] ja:loadClass "com.hp.hpl.jena.sdb.SDB" .

   sdb:DatasetStore rdfs:subClassOf ja:RDFDataset .
   sdb:Model rdfs:subClassOf  ja:Model .

   <#sdb_service>  rdf:type fuseki:Service ;
     rdfs:label    "SDB Service (RW)" ;
     fuseki:name   "metadata" ;
     fuseki:serviceQuery            "query" ;

     fuseki:serviceQuery            "sparql" ;
     fuseki:serviceUpdate           "update" ;
     fuseki:serviceReadGraphStore   "get" ;            

     fuseki:dataset  <#dataset1> ;       
    .

   <#dataset1> rdf:type sdb:DatasetStore ;
     sdb:store <#store>;

     .

  ##Connection to MySQL
   <#store> rdf:type sdb:Store  ;
    rdfs:label "SDB Store" ;
    sdb:layout "layout2" ;
    sdb:engine "InnoDB" ;
    sdb:connection
     [  rdf:type sdb:SDBConnection ;
      sdb:sdbType "MySQL" ;
      sdb:sdbHost "localhost" ;
      sdb:sdbName "sdb_data";
      sdb:sdbUser "root";
      sdb:sdbPassword "******";
      sdb:driver "com.mysql.jdbc.Driver";
     ]
   .

Many Thanks,

Paul