Jena-arq Tutorial

Jena

A Semantic Web Framework for Java Amna Basharat Spring 2009

Outline What is Jena Packages Jena Example SPARQL SPARQL Example My project DEMO

What is Jena Jena is a Java framework for building Semantic Web

applications includes

a RDF API reading and writing RDF in

RDF/XML, N3 and N-Triples an OWL API in-memory and persistent storage SPARQL query engine http://jena.sourceforge.net/

Packages Jena includes Jena and ARQ a package providing API about RDF and SPARQL Jena CVS a tool used to manage changes within source code Joseki a RDF publishing server Eyeball an "RDF lint" for checking RDF/OWL models

ARQ - Application API

8

Introduction – ARQ Application API The application API is in the

package com.hp.hpl.jena.query.



Other packages contain various parts of the

system (execution engine, parsers, testing etc).



Most applications will only need to use the

9

main package.  Only applications wishing to programmatically build queries or modify the behaviour of the query engine need to use the others packages directly.

Key Classes of ARQ Application API  The package com.hp.hpl.jena.query is the main application package.  Query  a class that represents the application query.  It is a container for all the details of the query.  Objects of class Query are normally created by calling one of the

  



10

methods of QueryFactory methods which provide access to the various parsers. QueryExecution  represents one execution of a query. QueryExecutionFactory  a place to get QueryExecution instances DatasetFactory  a place to make datasets, including making a DataSource (an updatable Dataset) For SELECT queries:  QuerySolution - A single solution to the query  ResultSet - All the QuerySolutions. An iterator.  ResultSetFormatter - turn a ResultSet into various forms; into text,

Creating Queries

11

SELECT queries The basic steps in making a SELECT

12

query are outlined in the example below. A query is created from a string using the QueryFactory. The query and model or RDF dataset to be queried are then passed to QueryExecutionFactory to produce an instance of a query execution. Result are handled in a loop and finally the query execution is closed.

Example Code Structure

t com.hp.hpl.jena.query.* ;

model = ... ; g queryString = " .... " ; query = QueryFactory.create(queryString) ; Execution qexec = QueryExecutionFactory.create(query, model) ;

tSet results = qexec.execSelect() ; ; results.hasNext() ; )

Solution soln = results.nextSolution() ; de x = soln.get("varName") ; // Get a result variable by name. rce r = soln.getResource("VarR") ; // Get a result variable - must be a res al l = soln.getLiteral("VarL") ; // Get a result variable - must be a liter

ally { qexec.close() ; }

13

Simplifying the Query Formulation and Execution  It is important to cleanly close the query execution

when finished.

  System resources connected to persistent storage

may need to be released.

  The step of creating a query and then a query import com.hp.hpl .jenabe .query .* ; execution can reduced to one step in some

common cases:

 

14

Model model = ... ; String queryString = " .... " ; QueryExecution qexec = QueryExecutionFactory.create(queryString, model) ; try { ResultSet results = qexec.execSelect() ; . . . } finally { qexec.close() ; }

Example: Formatting a ResultSet Instead of a loop to deal with each row in

the result set, the application can call an operation of the ResultSetFormatter. This is what the command line applications do. Example: processing results to produce a simple text presentation: 

15

ResultSetFormatter fmt = new ResultSetFormatter(results, query) ; fmt.printAll(System.out) ; OR ResultSetFormatter.out(System.out, results, query) ;

Example: Processing results The results are objects from the Jena

RDF API and API calls, which do not modify the model, can be mixed with for ( ; results.hasNext() ; ) query results processing:  { // Access variables: soln.get("x") ; RDFNode n = soln.get("x") ; // "x" is a variable in the query // If you need to test the thing returned if ( n.isLiteral() ) ((Literal)n).getLexicalForm() ; if ( n.isResource() ) { Resource r = (Resource)n ; if ( ! r.isAnon() ) { ... r.getURI() ... }}}

16

CONSTRUCT Queries CONSTRUCT queries return a single

RDF graph. As usual, the query execution should be closed after use. Query query = QueryFactory.create(queryString) ; QueryExecution qexec = QueryExecutionFactory.create(query, model) ; Model resultModel = qexec.execConstruct() ; qexec.close() ;

17

DESCRIBE Queries DESCRIBE queries return a single RDF

graph.  Different handlers for the DESCRIBE operation can be loaded by added by the application. Query query = QueryFactory.create(queryString) ; QueryExecution qexec = QueryExecutionFactory.create(query, model) ; Model resultModel = qexec.execDescribe() ; qexec.close() ;

18

ASK Queries The operation Query.execAsk() returns

a boolean value indicating whether the query pattern matched the graph or dataset or not. Query query = QueryFactory.create(queryString) ; QueryExecution qexec = QueryExecutionFactory.create(query, model) ; boolean result = qexec.execAsk() ; qexec.close()

19

Formatting XML results The ResultSetFormatter class has

methods to write out the  SPARQL Query Results XML Format. See ResultSetFormatter.outputAsXML method. 

20

Datasets The examples above are all queries on a single

model. A SPARQL query is made on a dataset, which is a default graph and zero or more named graphs. String dftGraphURI = "file :default-graph.ttlusing " ; Datasets can be constructed List namedGraphURIs = new ArrayList() ; the DatasetFactory: namedGraphURIs .add("file:named-1.ttl") ; namedGraphURIs.add("file:named-2.ttl") ;

Query query = QueryFactory.create(queryString) ; Dataset dataset = DatsetFactory.create(dftGraphURI, namedGraphURIs) ; QueryExecution qExec = QueryExecutionFactory.create(query, dataset) ; try { ... } finally { qExec.close() ; }

21

Using Existing Models Already existing models can also be

used: A DataSource is an updatable dataset: DataSource dataSource = DatsetFactory.create() ; dataSource.setDefaultModel(model) ; dataSource.addNamedModel("http://example/named-1", modelX) ; dataSource.addNamedModel("http://example/named-2", modelY) ; QueryExecution qExec = QueryExecutionFactory.create(query, dataSource) ; 

22

Examples

23

Example 1

http://example.org/book# 1 DC.title

DC.description DC.description Advanced techniques for SPARQL

SPARQL - the book

A book about SPARQL

Example 1 public class Ex1 {  public static void main( String[] args )  {  Model m = ModelFactory.createDefaultModel() ; 

 

Resource r1 = m.createResource( "http://example.org/book#1" ) ;

  



r1.addProperty(DC.title, "SPARQL - the book" ) .addProperty(DC.description, "A book about SPARQL" ) .addProperty(DC.description, "Advanced techniques for SPARQL" );



m.write( System.out, "RDF/XML-ABBREV" );

 

}



}

Example 1   Advanced techniques for SPARQL  A book about SPARQL  SPARQL - the book   

Example 2 http://example.org/book# 1 DC.title

DC.description DC.description http://example.org/book# 2

SPARQL - the book A book about SPARQL

DC.description

http://example.org/book# 3 DC.title

Jena - an RDF framework for Java

DC.description

A book about Jena

DC.title Advanced techniques for SPARQL

public class Ex2 {  public static void main( String[] args )  {  Model m = ModelFactory.createDefaultModel() ; 

 





Resource r1 = m.createResource( "http://example.org/book#1" ) ; Resource r2 = m.createResource( "http://example.org/book#2" ) ; Resource r3 = m.createResource( "http://example.org/book#3" ) ;

  

SPARQL" ) 

r1.addProperty(DC.title, "SPARQL - the book" ) .addProperty(DC.description, "A book about .addProperty(DC.description, r2) ;

 

SPARQL" ) 

r2.addProperty(DC.title, "Advanced techniques for .addProperty(DC.description, r3) ;

 

for Java" ) 

Jena") ;

r3.addProperty(DC.title, "Jena - an RDF framework .addProperty(DC.description, "A book about

 

m.write( System.out, "RDF/XML-ABBREV" );

    A book about Jena  Jena - an RDF framework for Java    Advanced techniques for SPARQL     A book about SPARQL  SPARQL - the book 

SPARQL An RDF query language allows for a query to consist of

triple patterns, conjunctions, disjunctions, and optional patterns ?x queries across diverse data dc:description sources      

?y

PREFIX dc: dc:description SELECT ?title WHERE { ?x dc:description ?y . dc:title ?y dc:description ?z . ?title ?z dc:title ?title . }

?z

SPARQL in Example 2 http://example.org/book# 1 DC.title

DC.description DC.description http://example.org/book# 2

SPARQL - the book A book about SPARQL

DC.description

http://example.org/book# 3 DC.title

Jena - an RDF framework for Java

DC.description

A book about Jena

DC.title Advanced techniques for SPARQL

public class Ex3 {  public static void main( String[] args) {  Model model = createModel() ;  

   

String prolog = "PREFIX dc:< " + DC.getURI() + ">"; String queryString = prolog + "\n" + "SELECT ?title WHERE { ?x dc:description ?y . ?y dc:description ?z . ?z dc:title ?title }";



Query query = QueryFactory.create( queryString ); QueryExecution qexec = QueryExecutionFactory.create( query, model );

 

System.out.println( "Titles: " ) ; try { ResultSet rs = qexec.execSelect();

  



for ( ; rs.hasNext() ; ) { QuerySolution rb = rs.nextSolution() ; RDFNode x = rb.get( "title" ) ;

   

if ( x.isLiteral() ) { Literal titleStr= ( Literal )x ; System.out.println( "

  

}

     

}



}

“ + titleStr ) ;

}

} finally qexec.close() ;

Titles: Jena - an RDF framework for Java

Q u e ry in g R e m o te S P A R Q L S e rv ice s

33

ARQ - Querying Remote SPARQL Services SPARQL is a query language and a 

remote access protocol. The remote access protocol can be used with plain HTTP or over  SOAP. See Joseki for an implementation of an RDF publishing server, using the SPARQL protocol (HTTP and SOAP). Joseki uses ARQ to provide SPARQL query access to Jena models, including Jena persistent models. ARQ includes a query engine capable of using the HTTP version. A version using SOAP is include in Joseki.  34

Firewalls and Proxies Don't forget to set the proxy for Java if you are

accessing a public server from behind a blocking firewall. Most home firewalls do not block outgoing requests; many corporate firewalls do block outgoing requests. If, to use your web browser, you need to set a proxy, you need to do so for a Java program. Simple examples include: -DsocksProxyHost=YourSocksServerDsocksProxyHost=YourSocksServer -DsocksProxyPort=portDhttp.proxyHost=WebProxy -Dhttp.proxyPort=Port This can be done in the application if it is done before any network connection are made: 35 System.setProperty("socksProxyHost", "socks.corp.com"); Consult the Java

From your application The QueryExecutionFactory has methods for

creating a QueryExecution object for remote use. QueryExecutionFactory.sparqlService

 These methods build a query execution object that

uses the query engine in com.hp.hpl.jena.sparql.engine.http.

 The remote request is made when

the execSelect, execConstruct, execDescribe or e xecAsk method is called.

 The results are held locally after remote execution

and can be processed as usual.



36

Protégé OWL API

37

OWL API Comparisons http://wiki.yoshtec.com/java-owl-api

38

A Comparison API-Name

Type

Generates Java Classes Yes

Tested

Jastor

Type 2

Sommer Type 2 Protege OWL-API 1 & 2

? Yes

Not Yet Yes

3.4 beta

BSD MPL

RDFReactor

Yes

Yes

4.5.2

BSD / LGPL ?

Kazuki ? Sesame and Elmo ?

? ?

Not Yet Not Yet

OWL API

Type 1

No

Yes

2.1.1

LGPL

Since it is a plain low level OWL API code looks horrible for this purpose

Jena

Type 1

No

Indirect

2.4

BSD-Style

Is used in a lot of different higher level APIs

Topaz JAOB

Type 2 Type 2

? Yes

No Yes

0.1

Apache 2.0 LGPL

39

Type 2

Version

Tried

License

Problems / Issues / Remarks

CPL

Unable to run with Jena 2.4 or Jena 2.5

none More research needed to correct OWL output

BSD

My own implementation

From the above Comparison,

ProtegeOWL API seems to be an OK Choice. 

40

How to use protege OWL API in Eclipse 1-Create Java Project

2-Goto Project->Properties-> Java Build Paths

41

Libraries tab, click add external JARS and go inside Protege directory ->select all jar files. inside Protege directory ->Plugins inside (edu.stanford.smi.protegex.owl) select all jar files. Also  Select protege.api from /protege/

ProtegeOWL API Tutorial By Amna Basharat Haider Fast - NU , Fall 2008

42

Tasks Creating Ontology Factory Publishing: Publishing new

Knowledge into Your OWL Ontology using Ontology Factory Querying/Knowledge Retrieval: Reasoning: Invoking a Reasoner throught the Code   43

Creating Ontology Factory

44

Publishing : Publishing new Knowledge into Your OWL Ontology using Ontology Factory

45

Querying / Knowledge Retrieval

JENA – ARQ Tutorial http :// jena . sourceforge . net / ARQ / documenta

46

An Introduction to RDF and the

Jena RDF API 

http://jena.sourceforge.net/tutorial/R

47

Jena Tutorial

http://jena.hpl.hp.com/juc2006/proce

48

R e a so n in g : In v o k in g a R e a so n e r th ro u g h t th e Code

49