Java Code Examples for org.apache.jena.tdb.TDBFactory#createDataset()

/**
 *
 * @throws Exception
 */
@BeforeClass
public static void setUpClass() throws Exception {
    final String typeName = "SparqlSourceType";
    if ( ! DataSourceTypesRegistry.isRegistered(typeName) ) {
        DataSourceTypesRegistry.register( typeName,
                                          new SparqlDataSourceType() );
    }

    String tmpdir = System.getProperty("java.io.tmpdir");
    jena = new File(tmpdir, "ldf-sparql-test");
    jena.mkdir();
    
    dataset = TDBFactory.createDataset(jena.getAbsolutePath());

    Model model = dataset.getDefaultModel();
    InputStream in = ClassLoader.getSystemResourceAsStream("demo.nt");
    RDFDataMgr.read(model, in, Lang.NTRIPLES);

    // Dynamically-generated port comes from pom.xml configuration: build-helper-maven-plugin
    int fusekiPort = Integer.parseInt(System.getProperty("fuseki.port"));

    // Create Fuseki, loaded with the test dataset
    fuseki = FusekiServer.create().setPort(fusekiPort).add("/ds", dataset).build();
    fuseki.start();

    // Everything is in place, now create the LDF datasource                
    JsonObject config = createConfig("sparql test", "sparql test",
                                     typeName);
    
    JsonObject settings = new JsonObject();
    settings.addProperty("endpoint", "http://localhost:" + fusekiPort + "/ds");
    config.add("settings", settings);

    setDatasource(DataSourceFactory.create(config));
}
 

private Dataset buildBaseDataset() {
	Dataset jenaData;
	
	if (StringUtils.isNotBlank(jenaConfig.getAssemblerFile())) {
		LOGGER.debug("Building dataset from assembler file {}", jenaConfig.getAssemblerFile());
		jenaData = DatasetFactory.assemble(jenaConfig.getAssemblerFile(), jenaConfig.getAssemblerDataset());
	} else if (StringUtils.isNotBlank(jenaConfig.getTdbPath())) {
		LOGGER.debug("Building dataset from TDB data at {}", jenaConfig.getTdbPath());
		jenaData = TDBFactory.createDataset(jenaConfig.getTdbPath());
	} else {
		LOGGER.debug("Building dataset from ontology URI {}", jenaConfig.getOntologyUri());
		FileManager fileManager = FileManager.get();
		Model model = fileManager.loadModel(jenaConfig.getOntologyUri());
		
		// Build the base dataset backed by the model loaded from the URI
		jenaData = DatasetFactory.create(model);
	}
	
	return jenaData;
}
 

public static void main(String... argv)
{
    // Direct way: Make a TDB-back Jena model in the named directory.
    String directory = "MyDatabases/DB1" ;
    Dataset dataset = TDBFactory.createDataset(directory) ;
    
    // Potentially expensive query.
    String sparqlQueryString = "SELECT (count(*) AS ?count) { ?s ?p ?o }" ;
    // See http://incubator.apache.org/jena/documentation/query/app_api.html
    
    Query query = QueryFactory.create(sparqlQueryString) ;
    QueryExecution qexec = QueryExecutionFactory.create(query, dataset) ;
    try {
      ResultSet results = qexec.execSelect() ;
      for ( ; results.hasNext() ; )
      {
          QuerySolution soln = results.nextSolution() ;
          int count = soln.getLiteral("count").getInt() ;
          System.out.println("count = "+count) ;
      }
    } finally { qexec.close() ; }

    // Close the dataset.
    dataset.close();
    
}
 

public static void main(String... argv)
{
    // Direct way: Make a TDB-back Jena model in the named directory.
    String directory = "MyDatabases/DB1" ;
    Dataset dataset = TDBFactory.createDataset(directory) ;
    
    // Potentially expensive query.
    String sparqlQueryString = "SELECT (count(*) AS ?count) { ?s ?p ?o }" ;
    // See http://incubator.apache.org/jena/documentation/query/app_api.html
    
    Query query = QueryFactory.create(sparqlQueryString) ;
    QueryExecution qexec = QueryExecutionFactory.create(query, dataset) ;
    ResultSet results = qexec.execSelect() ;
    ResultSetFormatter.out(results) ;
    qexec.close() ;

    dataset.close();
}
 

/**
 * Creates a dataset description with FTS support for the RDFS label property.
 */
private static Dataset createFusekiFTSDataset()
{
    Dataset ds1 = TDBFactory.createDataset();
    Directory dir = new RAMDirectory();
    EntityDefinition eDef = new EntityDefinition("iri", "text");
    eDef.setPrimaryPredicate(org.apache.jena.vocabulary.RDFS.label);
    TextIndex tidx = new TextIndexLucene(dir, new TextIndexConfig(eDef));
    Dataset ds = TextDatasetFactory.create(ds1, tidx);
    return ds;
}
 

/**
 * Given a path to an RDF/XML or TTL file and a RDF language, load the file as the default model
 * of a TDB dataset backed by a directory to improve processing time. Return the new dataset.
 *
 * <p>WARNING - this creates a directory at given tdbDir location!
 *
 * @param inputPath input path of RDF/XML or TTL file
 * @param tdbDir location to put TDB mappings
 * @return Dataset instantiated with triples
 * @throws JenaException if TDB directory can't be written to
 */
public static Dataset loadToTDBDataset(String inputPath, String tdbDir) throws JenaException {
  Dataset dataset;
  if (new File(tdbDir).isDirectory()) {
    dataset = TDBFactory.createDataset(tdbDir);
    if (!dataset.isEmpty()) {
      return dataset;
    }
  }
  dataset = TDBFactory.createDataset(tdbDir);
  logger.debug(String.format("Parsing input '%s' to dataset", inputPath));
  // Track parsing time
  long start = System.nanoTime();
  Model m;
  dataset.begin(ReadWrite.WRITE);
  try {
    m = dataset.getDefaultModel();
    FileManager.get().readModel(m, inputPath);
    dataset.commit();
  } catch (JenaException e) {
    dataset.abort();
    dataset.end();
    dataset.close();
    throw new JenaException(String.format(syntaxError, inputPath));
  } finally {
    dataset.end();
  }
  long time = (System.nanoTime() - start) / 1000000000;
  logger.debug(String.format("Parsing complete - took %s seconds", String.valueOf(time)));
  return dataset;
}
 

/**
 *
 * @throws Exception
 */
@BeforeClass
public static void setUpClass() throws Exception {
    final String typeName = "JenaSourceType";
    if ( ! DataSourceTypesRegistry.isRegistered(typeName) ) {
        DataSourceTypesRegistry.register( typeName,
                                          new JenaTDBDataSourceType() );
    }

    String tmpdir = System.getProperty("java.io.tmpdir");
    jena = new File(tmpdir, "ldf-jena-test");
    jena.mkdir();
    
    dataset = TDBFactory.createDataset(jena.getAbsolutePath());

    Model model = dataset.getDefaultModel();
    InputStream in = ClassLoader.getSystemResourceAsStream("demo.nt");
    RDFDataMgr.read(model, in, Lang.NTRIPLES);

    // Everything is in place, now create the LDF datasource                
    JsonObject config = createConfig("jena tdb test", "jena tdb test",
                                     typeName);
    
    JsonObject settings = new JsonObject();
    settings.addProperty("directory", jena.getAbsolutePath());
    config.add("settings", settings);

    setDatasource(DataSourceFactory.create(config));
}
 

public static void main(String... argv)
{
    String directory = "MyDatabases/DB1" ;
    Dataset dataset = TDBFactory.createDataset(directory) ;

    // Start WRITE transaction. 
    //   It's possible to read from the datet inside the write transaction.

    //   An application can have other Datasets, in the same JVM, 
    //   tied to the same TDB database performing read
    //   transactions concurrently. If another write transaction
    //   starts, the call of dataset.begin(WRITE) blocks until
    //   existing writer finishes.
    
    dataset.begin(ReadWrite.WRITE) ;
    try
    {
        GraphStore graphStore = GraphStoreFactory.create(dataset) ;
        // Do a SPARQL Update.
        String sparqlUpdateString = StrUtils.strjoinNL(
             "PREFIX . <http://example/>",
             "INSERT { :s :p ?now } WHERE { BIND(now() AS ?now) }"
             ) ;

        execUpdate(sparqlUpdateString, graphStore) ;
        dataset.commit() ;
        // Or call .abort()
        
    } finally
    {
        // Notify the end of the transaction.
        // The transaction was finished at the point .commit or .abort was called.
        // .end will force an abort() if no previous call to .commit() or .abort()
        // has occurred, so .end() help manage track the state of the transaction.
        // .end() can be called multiple times for the same .begin(WRITE)
        dataset.end() ;
    }
}
 

public static void main(String... argv)
{
    String directory = "MyDatabases/DB1" ;
    Dataset dataset = TDBFactory.createDataset(directory) ;

    // Start READ transaction. 
    //   No updates or changes to the dataset are possible while this
    //   dataset is used for a read transaction.
    //   An application can have other Datasets, in the same JVM, 
    //   tied to the same TDB database performing read or write
    //   transactions concurrently.
    
    dataset.begin(ReadWrite.READ) ;
    try
    {
        // Do some queries
        String sparqlQueryString1 = "SELECT (count(*) AS ?count) { ?s ?p ?o }" ;
        execQuery(sparqlQueryString1, dataset) ;
        
        String sparqlQueryString2 = "SELECT * { ?s ?p ?o }" ;
        execQuery(sparqlQueryString2, dataset) ;
        
        // Can also call dataset.abort() or dataset.commit() here 
    } finally
    {
        // Notify the end of the READ transaction.
        // Any use of dataset.abort() or dataset.commit() or dataset.end()
        // .end() can be called multiple times for the same .begin(READ)
        dataset.end() ;
    }
}
 

public static void main(String... argv)
{
    // Direct way: Make a TDB-back Jena model in the named directory.
    String directory = "MyDatabases/DB1" ;
    Dataset ds = TDBFactory.createDataset(directory) ;
    Model model = ds.getDefaultModel() ;
    
    // ... do work ...
    
    // Close the dataset.
    ds.close();
    
}
 

/** Example setup - in-memory dataset with two graphs, one triple in each */
private static Dataset setup()
{
    Dataset ds = TDBFactory.createDataset() ;
    DatasetGraph dsg = ds.asDatasetGraph() ;
    Quad q1 = SSE.parseQuad("(<http://example/g1> <http://example/s> <http://example/p> <http://example/o1>)") ;
    Quad q2 = SSE.parseQuad("(<http://example/g2> <http://example/s> <http://example/p> <http://example/o2>)") ;
    dsg.add(q1) ;
    dsg.add(q2) ;
    return ds ;
}
 

public JenaTDBBasedRequestProcessorForTPFs(File tdbdir, String defaultGraph) {
    this.defaultGraph = defaultGraph;
    this.tdb = TDBFactory.createDataset(tdbdir.getAbsolutePath());
}
 

public static void main(String[] args) throws Exception {
    /// turn off the "No BGP optimizer warning"
    TDB.setOptimizerWarningFlag(false);

    final IRIFactory iriFactory = IRIFactory.semanticWebImplementation();

    final String DATASET_DIR_NAME = "data0";
    final Dataset data0 = TDBFactory.createDataset( DATASET_DIR_NAME );

    // show the currently registered names
    for (Iterator<String> it = data0.listNames(); it.hasNext(); ) {
        out.println("NAME="+it.next());
    }

    out.println("getting named model...");
    /// this is the OWL portion
    final Model model = data0.getNamedModel( MY_NS );
    out.println("Model := "+model);

    out.println("getting graph...");
    /// this is the DATA in that MODEL
    final Graph graph = model.getGraph();
    out.println("Graph := "+graph);

    if (graph.isEmpty()) {
        final Resource product1 = model.createResource(
                iriFactory.construct( MY_NS +"product/1" )
                    .toString() );

        final Property hasName = model.createProperty( MY_NS, "#hasName");
        final Statement stmt = model.createStatement(
                product1, hasName, model.createLiteral("Beach Ball","en") );
        out.println("Statement = " + stmt);

        model.add(stmt);

        // just for fun
        out.println("Triple := " + stmt.asTriple().toString());
    } else {
        out.println("Graph is not Empty; it has "+graph.size()+" Statements");
        long t0, t1;
        t0 = System.currentTimeMillis();
        final Query q = QueryFactory.create(
                "PREFIX exns: <"+MY_NS+"#>\n"+
                "PREFIX exprod: <"+MY_NS+"product/>\n"+
                " SELECT * "
                // if you don't provide the Model to the
                // QueryExecutionFactory below, then you'll need
                // to specify the FROM;
                // you *can* always specify it, if you want
                // +" FROM <"+MY_NS+">\n"
                // +" WHERE { ?node <"+MY_NS+"#hasName> ?name }"
                // +" WHERE { ?node exns:hasName ?name }"
                // +" WHERE { exprod:1 exns:hasName ?name }"
                +" WHERE { ?res ?pred ?obj }"
                );
        out.println("Query := "+q);
        t1 = System.currentTimeMillis();
        out.println("QueryFactory.TIME="+(t1 - t0));

        t0 = System.currentTimeMillis();
        final QueryExecution qExec = QueryExecutionFactory
                // if you query the whole DataSet,
                // you have to provide a FROM in the SparQL
                //.create(q, data0);
                .create(q, model);
        t1 = System.currentTimeMillis();
        out.println("QueryExecutionFactory.TIME="+(t1 - t0));

        try {
            t0 = System.currentTimeMillis();
            ResultSet rs = qExec.execSelect();
            t1 = System.currentTimeMillis();
            out.println("executeSelect.TIME="+(t1 - t0));
            while (rs.hasNext()) {
                QuerySolution sol = rs.next();
                out.println("Solution := "+sol);
                for (Iterator<String> names = sol.varNames(); names.hasNext(); ) {
                    final String name = names.next();
                    out.println("\t"+name+" := "+sol.get(name));
                }
            }
        } finally {
            qExec.close();
        }
    }
    out.println("closing graph");
    graph.close();
    out.println("closing model");
    model.close();
    //out.println("closing DataSetGraph");
    //dsg.close();
    out.println("closing DataSet");
    data0.close();
}