NIHVIVO-3974 preventing reproduction of axioms with blank nodes in filegraphs and NIHVIVO-3958 applying query syntax leniency to RDFServiceSparql

webapp/src/edu/cornell/mannlib/vitro/webapp/rdfservice/impl/RDFServiceImpl.java

@@ -3,6 +3,7 @@
 package edu.cornell.mannlib.vitro.webapp.rdfservice.impl;
 
 import java.io.ByteArrayInputStream;
+import java.util.Arrays;
 import java.util.Iterator;
 import java.util.List;
 import java.util.concurrent.CopyOnWriteArrayList;
@@ -13,7 +14,11 @@ import org.apache.commons.logging.LogFactory;
 import com.hp.hpl.jena.graph.Graph;
 import com.hp.hpl.jena.graph.Node;
 import com.hp.hpl.jena.graph.Triple;
+import com.hp.hpl.jena.query.Query;
+import com.hp.hpl.jena.query.QueryFactory;
+import com.hp.hpl.jena.query.QueryParseException;
 import com.hp.hpl.jena.query.ResultSet;
+import com.hp.hpl.jena.query.Syntax;
 import com.hp.hpl.jena.rdf.model.Model;
 import com.hp.hpl.jena.rdf.model.ModelFactory;
 import com.hp.hpl.jena.rdf.model.Statement;
@@ -31,7 +36,7 @@ public abstract class RDFServiceImpl implements RDFService {
 	
 	private static final Log log = LogFactory.getLog(RDFServiceImpl.class);
 	protected static final String BNODE_ROOT_QUERY = 
-	        "SELECT ?s WHERE { ?s ?p ?o OPTIONAL { ?ss ?pp ?s } FILTER (!isBlank(?s) || !bound(?ss)) }";
+	        "SELECT DISTINCT ?s WHERE { ?s ?p ?o OPTIONAL { ?ss ?pp ?s } FILTER (!isBlank(?s) || !bound(?ss)) }";
 	
 	protected String defaultWriteGraphURI;
 	protected List<ChangeListener> registeredListeners = new CopyOnWriteArrayList<ChangeListener>();
@@ -261,4 +266,24 @@ public abstract class RDFServiceImpl implements RDFService {
         result[1] = nonBlankNodeModel;
         return result;
     }
+    
+    protected Query createQuery(String queryString) {
+        List<Syntax> syntaxes = Arrays.asList(
+                Syntax.defaultQuerySyntax, Syntax.syntaxSPARQL_11, 
+                Syntax.syntaxSPARQL_10, Syntax.syntaxSPARQL, Syntax.syntaxARQ);
+        Query q = null;
+        Iterator<Syntax> syntaxIt = syntaxes.iterator(); 
+        while (q == null) {
+            Syntax syntax = syntaxIt.next();
+            try {
+               q = QueryFactory.create(queryString, syntax);  
+            } catch (QueryParseException e) {
+               if (!syntaxIt.hasNext()) {
+                   throw(e);
+               }
+            }
+        }
+        return q;
+    }
+    
 }

webapp/src/edu/cornell/mannlib/vitro/webapp/rdfservice/impl/jena/RDFServiceJena.java

@@ -359,25 +359,6 @@ public abstract class RDFServiceJena extends RDFServiceImpl implements RDFServic
         super.notifyListeners(triple, operation, graphURI);
     }
     
-    protected Query createQuery(String queryString) {
-        List<Syntax> syntaxes = Arrays.asList(
-                Syntax.defaultQuerySyntax, Syntax.syntaxSPARQL_11, 
-                Syntax.syntaxSPARQL_10, Syntax.syntaxSPARQL, Syntax.syntaxARQ);
-        Query q = null;
-        Iterator<Syntax> syntaxIt = syntaxes.iterator(); 
-        while (q == null) {
-            Syntax syntax = syntaxIt.next();
-            try {
-               q = QueryFactory.create(queryString, syntax);  
-            } catch (QueryParseException e) {
-               if (!syntaxIt.hasNext()) {
-                   throw(e);
-               }
-            }
-        }
-        return q;
-    }
-    
     protected QueryExecution createQueryExecution(String queryString, Query q, Dataset d) {
         return QueryExecutionFactory.create(q, d);
     }

webapp/src/edu/cornell/mannlib/vitro/webapp/rdfservice/impl/sparql/RDFServiceSparql.java

@@ -13,9 +13,7 @@ import java.util.List;
 import org.apache.commons.httpclient.HttpClient;
 import org.apache.commons.httpclient.MultiThreadedHttpConnectionManager;
 import org.apache.commons.httpclient.NameValuePair;
-import org.apache.commons.httpclient.methods.ByteArrayRequestEntity;
 import org.apache.commons.httpclient.methods.PostMethod;
-import org.apache.commons.httpclient.methods.RequestEntity;
 import org.apache.commons.io.IOUtils;
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
@@ -49,7 +47,6 @@ import edu.cornell.mannlib.vitro.webapp.rdfservice.RDFServiceException;
 import edu.cornell.mannlib.vitro.webapp.rdfservice.impl.ChangeSetImpl;
 import edu.cornell.mannlib.vitro.webapp.rdfservice.impl.RDFServiceImpl;
 import edu.cornell.mannlib.vitro.webapp.rdfservice.impl.jena.ListeningGraph;
-import edu.cornell.mannlib.vitro.webapp.web.URLEncoder;
 
 /*
  * API to write, read, and update Vitro's RDF store, with support 
@@ -223,7 +220,7 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
 			                                RDFServiceImpl.ModelSerializationFormat resultFormat) throws RDFServiceException {
 		
 		Model model = ModelFactory.createDefaultModel();
-		Query query = QueryFactory.create(queryStr);
+		Query query = createQuery(queryStr);
 		QueryExecution qe = QueryExecutionFactory.sparqlService(readEndpointURI, query);
 		
 		try {
@@ -253,7 +250,7 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
 			                               RDFServiceImpl.ModelSerializationFormat resultFormat) throws RDFServiceException {
 		
 		Model model = ModelFactory.createDefaultModel();
-		Query query = QueryFactory.create(queryStr);
+		Query query = createQuery(queryStr);
 		QueryExecution qe = QueryExecutionFactory.sparqlService(readEndpointURI, query);
 		
 		try {
@@ -281,7 +278,7 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
 	@Override
 	public InputStream sparqlSelectQuery(String queryStr, RDFService.ResultFormat resultFormat) throws RDFServiceException {
 		
-        Query query = QueryFactory.create(queryStr);
+        Query query = createQuery(queryStr);
         QueryExecution qe = QueryExecutionFactory.sparqlService(readEndpointURI, query);
         
         try {
@@ -323,7 +320,7 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
 	@Override
 	public boolean sparqlAskQuery(String queryStr) throws RDFServiceException {
 		
-	    Query query = QueryFactory.create(queryStr);
+	    Query query = createQuery(queryStr);
 	    QueryExecution qe = QueryExecutionFactory.sparqlService(readEndpointURI, query);
 	    
 	    try {
@@ -465,7 +462,7 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
         }
     }
     
-    protected void executeUpdate(String updateString) throws RDFServiceException {    
+    protected void executeUpdate(String updateString) throws RDFServiceException {  
         try {
             PostMethod meth = new PostMethod(updateEndpointURI);
             try {
@@ -582,7 +579,7 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
 	@Override
 	protected boolean sparqlSelectQueryHasResults(String queryStr) throws RDFServiceException {
 		
-        Query query = QueryFactory.create(queryStr);
+        Query query = createQuery(queryStr);
         QueryExecution qe = QueryExecutionFactory.sparqlService(readEndpointURI, query);
         
         try {
@@ -638,6 +635,8 @@ public class RDFServiceSparql extends RDFServiceImpl implements RDFService {
         Model blankNodeModel = ModelFactory.createDefaultModel();
         blankNodeModel.add(blankNodeStatements);
         
+        
+        
         log.debug("update model size " + model.size());       
         log.debug("blank node model size " + blankNodeModel.size());
         

webapp/src/edu/cornell/mannlib/vitro/webapp/servlet/setup/FileGraphSetup.java

@@ -19,7 +19,6 @@ import javax.servlet.ServletContextListener;
 import org.apache.commons.logging.Log;
 import org.apache.commons.logging.LogFactory;
 
-import com.hp.hpl.jena.ontology.OntDocumentManager;
 import com.hp.hpl.jena.ontology.OntDocumentManager;
 import com.hp.hpl.jena.ontology.OntModel;
 import com.hp.hpl.jena.ontology.OntModelSpec;
@@ -29,7 +28,8 @@ import com.hp.hpl.jena.rdf.model.ModelFactory;
 
 import edu.cornell.mannlib.vitro.webapp.dao.jena.ModelContext;
 import edu.cornell.mannlib.vitro.webapp.dao.jena.OntModelSelector;
-import edu.cornell.mannlib.vitro.webapp.dao.jena.SparqlGraph;
+import edu.cornell.mannlib.vitro.webapp.dao.jena.RDFServiceModelMaker;
+import edu.cornell.mannlib.vitro.webapp.rdfservice.impl.RDFServiceUtils;
 
 // This ContextListener must run after the JenaDataSourceSetup ContextListener
 
@@ -50,9 +50,11 @@ public class FileGraphSetup implements ServletContextListener {
 
         try {
 
+            ServletContext ctx = sce.getServletContext();
             OntDocumentManager.getInstance().setProcessImports(true);
             baseOms = ModelContext.getBaseOntModelSelector(sce.getServletContext());
             Dataset dataset = JenaDataSourceSetupBase.getStartupDataset(sce.getServletContext());
+            RDFServiceModelMaker maker = new RDFServiceModelMaker(RDFServiceUtils.getRDFServiceFactory(ctx));
 
             // ABox files
             Set<String> pathSet = sce.getServletContext().getResourcePaths(PATH_ROOT + ABOX);
@@ -61,7 +63,7 @@ public class FileGraphSetup implements ServletContextListener {
 
             if (pathSet != null) {
                 OntModel aboxBaseModel = baseOms.getABoxModel();
-                aboxChanged = readGraphs(sce, pathSet, dataset, ABOX, aboxBaseModel);		
+                aboxChanged = readGraphs(sce, pathSet, maker, ABOX, aboxBaseModel);		
             }
 
             // TBox files
@@ -71,7 +73,7 @@ public class FileGraphSetup implements ServletContextListener {
 
             if (pathSet != null) {
                 OntModel tboxBaseModel = baseOms.getTBoxModel();
-                tboxChanged = readGraphs(sce, pathSet, dataset, TBOX, tboxBaseModel);
+                tboxChanged = readGraphs(sce, pathSet, maker, TBOX, tboxBaseModel);
             }
         } catch (ClassCastException cce) {
             String errMsg = "Unable to cast servlet context attribute to the appropriate type " + cce.getLocalizedMessage();
@@ -105,7 +107,7 @@ public class FileGraphSetup implements ServletContextListener {
      * Note: no connection needs to be maintained between the in-memory copy of the
      * graph and the DB copy.
      */
-    public boolean readGraphs(ServletContextEvent sce, Set<String> pathSet, Dataset dataset, String type, OntModel baseModel) {
+    public boolean readGraphs(ServletContextEvent sce, Set<String> pathSet, RDFServiceModelMaker dataset, String type, OntModel baseModel) {
 
         int count = 0;
 
@@ -169,10 +171,9 @@ public class FileGraphSetup implements ServletContextListener {
      * Otherwise, if a graph with the given name is in the DB and is isomorphic with
      * the graph that was read from the files system, then do nothing. 
      */
-    public boolean updateGraphInDB(Dataset dataset, Model fileModel, String type, String path) {
+    public boolean updateGraphInDB(RDFServiceModelMaker dataset, Model fileModel, String type, String path) {
-
         String graphURI = pathToURI(path,type);
-        Model dbModel = dataset.getNamedModel(graphURI);
+        Model dbModel = dataset.getModel(graphURI);
         boolean modelChanged = false;
         
         boolean isIsomorphic = dbModel.isIsomorphicWith(fileModel);

webapp/src/edu/cornell/mannlib/vitro/webapp/utils/jena/JenaOutputUtils.java

@@ -28,6 +28,10 @@ public class JenaOutputUtils {
 		
 		Map<String,String> prefixes = new HashMap<String,String>();
 		List<Ontology> ontologies = wadf.getOntologyDao().getAllOntologies();
+		// apparently this is null if empty
+		if (ontologies == null) {
+		    return;
+		}
 		Iterator<Ontology> iter = ontologies.iterator();
 		String namespace = null;
 		String prefix = null;