Menu
▾
▴
dl-learner-svn
|
From: <sk...@us...> - 2007-10-08 12:06:02
|
Revision: 180
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=180&view=rev
Author: sknappe
Date: 2007-10-08 05:05:59 -0700 (Mon, 08 Oct 2007)
Log Message:
-----------
Sparql Modul in org.dllearner.kb ?\195?\188bertragen
Added Paths:
-----------
trunk/src/dl-learner/org/dllearner/kb/Cache.java
trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java
trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java
trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java
trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java
Added: trunk/src/dl-learner/org/dllearner/kb/Cache.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/Cache.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/Cache.java 2007-10-08 12:05:59 UTC (rev 180)
@@ -0,0 +1,210 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.FileOutputStream;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.io.Serializable;
+import java.net.URLEncoder;
+/**
+ *
+ * This is a primitive cache.
+ * The objects of this class can be either the cache itself or just on entry in the cache
+ *
+ * the cache remembers: a timestamp, the original sparql-query, the result
+ * key is the subject http://dbpedia.org/resource/Angela_Merkel which is first urlencoded
+ * and so serves as the hash for the filename.
+ * Cache validates if timestamp too old and Sparql-Query the same
+ * before returning the SPARQL xml-result
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class Cache implements Serializable{
+
+
+ final static long serialVersionUID=104;
+ transient String basedir="";
+ transient String fileending=".cache";
+ long timestamp;
+ String content="";
+ long daysoffreshness=15;
+ long multiplier=24*60*60*1000;//h m s ms
+ String sparqlquery="";
+
+
+ /**
+ * Constructor for the cache itself.
+ * Called once at the beginning
+ *
+ * @param path Where the base path to the cache is
+ */
+ public Cache(String path){
+ this.basedir=path+File.separator;
+ if(!new File(path).exists())
+ {System.out.println(new File(path).mkdir());;}
+
+ }
+
+//
+ /**
+ * Constructor for single cache object(one entry)
+ *
+ * @param content the sparql xml result
+ * @param sparql the sparql query
+ */
+ public Cache(String content, String sparql){
+ this.content=content;
+ this.sparqlquery=sparql;
+ this.timestamp=System.currentTimeMillis();
+ }
+
+
+
+ /**
+ * use only on the cache object describing the cache itself
+ *
+ * @param key the individual
+ * @param sparql the sparql query
+ * @return the cached sparql result or null
+ */
+ public String get(String key, String sparql){
+ //System.out.println("get From "+key);
+ String ret=null;
+ try{
+ Cache c =readFromFile(makeFilename(key)) ;
+ if(c==null)return null;
+ //System.out.println(" file found");
+ if(!c.checkFreshness())return null;
+ //System.out.println("fresh");
+ if(!c.validate(sparql))return null;
+ //System.out.println("valid");
+ ret=c.content;
+ }catch (Exception e) {e.printStackTrace();}
+ return ret;
+ }
+
+ /**
+ *
+ * constructor for single cache object(one entry)
+ *
+ * @param key the individual
+ * @param content the sparql result
+ * @param sparql the sparql query
+ */
+ public void put(String key, String content, String sparql){
+ //System.out.println("put into "+key);
+ Cache c=new Cache(content,sparql);
+ putIntoFile(makeFilename(key), c);
+ }
+
+
+ /**
+ * to normalize the filenames
+ *
+ * @param key
+ * @return
+ */
+ String makeFilename(String key){
+ String ret="";
+ try{
+ ret=basedir+URLEncoder.encode(key, "UTF-8")+fileending;
+ }catch (Exception e) {e.printStackTrace();}
+ return ret;
+ }
+
+ /**
+ * how old is the result
+ * @return
+ */
+ boolean checkFreshness(){
+ if((System.currentTimeMillis()-this.timestamp)<=(daysoffreshness*multiplier))
+ //fresh
+ return true;
+ else return false;
+ }
+
+
+ /**
+ * some sparql query
+ * @param sparql
+ * @return
+ */
+ boolean validate(String sparql){
+ if(this.sparqlquery.equals(sparql))
+ //valid
+ return true;
+ else return false;
+ }
+
+ /**
+ * makes a new file if none exists
+ * @param Filename
+ */
+ public void checkFile(String Filename){
+ if(!new File(Filename).exists()){
+ try{
+ new File(Filename).createNewFile();
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+
+ }
+
+ /**
+ * internal saving function
+ * puts a cache object into a file
+ *
+ * @param Filename
+ * @param content
+ */
+ public void putIntoFile(String Filename,Cache content){
+ try{
+ //FileWriter fw=new FileWriter(new File(Filename),true);
+ FileOutputStream fos = new FileOutputStream( Filename , false );
+ ObjectOutputStream o = new ObjectOutputStream( fos );
+ o.writeObject( content );
+ fos.flush();
+ fos.close();
+ }catch (Exception e) {System.out.println("Not in cache creating: "+Filename);}
+ }
+
+ /**
+ * internal retrieval function
+ *
+ * @param Filename
+ * @return one entry object
+ */
+ public Cache readFromFile(String Filename){
+ Cache content=null;
+ try{
+ FileInputStream fos = new FileInputStream( Filename );
+ ObjectInputStream o = new ObjectInputStream( fos );
+ content=(Cache)o.readObject();
+ //FileReader fr=new FileReader(new File(Filename,"r"));
+ //BufferedReader br=new BufferedReader(fr);
+ }catch (Exception e) {}
+ return content;
+
+ }
+}
Added: trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java 2007-10-08 12:05:59 UTC (rev 180)
@@ -0,0 +1,365 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+import java.net.InetAddress;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.Iterator;
+
+/**
+ * This class collects the ontology from dbpedia,
+ * everything is saved in hashsets, so the doublettes are taken care of
+ *
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class OntologyCollector {
+
+ boolean print_flag=false;
+ SimpleHTTPRequest s;
+ QueryMaker q;
+ Cache c;
+ InetAddress ia;
+ SparqlFilter sf;
+ String[] subjectList;
+ int numberOfRecursions;
+ HashSet<String> properties;
+ HashSet<String> classes;
+ HashSet<String> instances;
+ HashSet<String> triples;
+
+ // some namespaces
+ String subclass="http://www.w3.org/2000/01/rdf-schema#subClassOf";
+ String type="http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
+ String objectProperty="http://www.w3.org/2002/07/owl#ObjectProperty";
+ String classns="http://www.w3.org/2002/07/owl#Class";
+ String thing="http://www.w3.org/2002/07/owl#Thing";
+
+
+ String[] defaultClasses={
+ "http://dbpedia.org/class/yago",
+ "http://dbpedia.org/resource/Category:",
+ "http://dbpedia.org/resource/Template:",
+ "http://www.w3.org/2004/02/skos/core",
+ "http://dbpedia.org/class/"}; //TODO FEHLER hier fehlt yago
+
+
+ /**
+ *
+ *
+ * @param subjectList
+ * @param numberOfRecursions
+ * @param filterMode
+ * @param FilterPredList
+ * @param FilterObjList
+ * @param defClasses
+ */
+ public OntologyCollector(String[] subjectList,int numberOfRecursions,
+ int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses){
+ this.subjectList=subjectList;
+ this.numberOfRecursions=numberOfRecursions;
+
+ this.s=new SimpleHTTPRequest();
+ this.q=new QueryMaker();
+ this.c=new Cache("cache");
+ if(defClasses!=null && defClasses.length>0 ){
+ this.defaultClasses=defClasses;
+ }
+
+ try{
+ this.sf=new SparqlFilter(filterMode,FilterPredList,FilterObjList);
+ this.ia=InetAddress.getByName("dbpedia.openlinksw.com");
+ //this.fw=new FileWriter(new File(System.currentTimeMillis()+".nt"),true);
+ this.properties=new HashSet<String>();
+ this.classes=new HashSet<String>();
+ this.instances=new HashSet<String>();
+ this.triples=new HashSet<String>();
+ //this.all=new HashSet<String>();
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+ /**
+ * first collects the ontology
+ * then types everything so it becomes owl-dl
+ *
+ * @return all triples in n-triple format
+ */
+ public String collectOntology(){
+ getRecursiveList(subjectList, numberOfRecursions);
+ finalize();
+ String ret="";
+ for (Iterator<String> iter = triples.iterator(); iter.hasNext();) {
+ ret += iter.next();
+
+ }
+ return ret;
+ }
+
+ /**
+ * calls getRecursive for each subject in list
+ * @param subjects
+ * @param NumberofRecursions
+ */
+ public void getRecursiveList(String[] subjects,int NumberofRecursions){
+ for (int i = 0; i < subjects.length; i++) {
+ getRecursive(subjects[i], NumberofRecursions);
+
+ }
+
+ }
+
+ /**
+ * gets all triples until numberofrecursion-- gets 0
+ *
+ * @param StartingSubject
+ * @param NumberofRecursions
+ */
+ public void getRecursive(String StartingSubject,int NumberofRecursions){
+ System.out.print("SparqlModul: Depth: "+NumberofRecursions+" @ "+StartingSubject+" ");
+ if(NumberofRecursions<=0)
+ { return;
+ }
+ else {NumberofRecursions--;}
+ //System.out.println(NumberofRecursions);
+ try{
+
+ String sparql=q.makeQueryFilter(StartingSubject,this.sf);
+ p(sparql);
+ p("*******************");
+ // checks cache
+ String FromCache=c.get(StartingSubject, sparql);
+ String xml;
+ // if not in cache get it from dbpedia
+ if(FromCache==null){
+ xml=s.sendAndReceive(ia, 8890, sparql);
+ c.put(StartingSubject, xml, sparql);
+ System.out.print("\n");
+ }
+ else{
+ xml=FromCache;
+ System.out.println("FROM CACHE");
+ }
+ p(xml);
+ p("***********************");
+ // get new Subjects
+ String[] newSubjects=processResult(StartingSubject,xml);
+
+ for (int i = 0; (i < newSubjects.length)&& NumberofRecursions!=0; i++) {
+ getRecursive(newSubjects[i], NumberofRecursions);
+ }
+
+ //System.out.println(xml);
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+
+ /**
+ * process the sparql result xml in a simple manner
+ *
+ *
+ * @param subject
+ * @param xml
+ * @return list of new individuals
+ */
+ public String[] processResult(String subject,String xml){
+ //TODO if result is empty, catch exceptions
+ String one="<binding name=\"predicate\"><uri>";
+ String two="<binding name=\"object\"><uri>";
+ String end="</uri></binding>";
+ String predtmp="";
+ String objtmp="";
+ ArrayList<String> al=new ArrayList<String>();
+
+ while(xml.indexOf(one)!=-1){
+ //get pred
+ xml=xml.substring(xml.indexOf(one)+one.length());
+ predtmp=xml.substring(0,xml.indexOf(end));
+ //getobj
+ xml=xml.substring(xml.indexOf(two)+two.length());
+ objtmp=xml.substring(0,xml.indexOf(end));
+
+ // writes the triples and resources in the hashsets
+ // also fills the arraylist al
+ processTriples(subject, predtmp, objtmp,al);
+ //System.out.println(al.size());
+
+ }
+
+ // convert al to list
+ Object[] o=al.toArray();
+ String[] ret=new String[o.length];
+ for (int i = 0; i < o.length; i++) {
+ ret[i]=(String)o[i];
+ }
+ return ret;
+ //return (String[])al.toArray();
+ //System.out.println(xml);
+ }
+
+
+
+ /**
+ *
+ * writes the triples and resources in the hashsets
+ * also fills the arraylist al with new individals for further processing
+ * @param s
+ * @param p
+ * @param o
+ * @param al
+ */
+ public void processTriples(String s,String p, String o,ArrayList<String> al){
+ // the next two lines bump out some inconsistencies within dbpedia
+ String t="/Category";
+ if(s.equals(t) || o.equals(t))return ;
+
+ if(sf.mode==2)
+ {
+ if( o.startsWith("http://dbpedia.org/resource/Category:")
+ &&
+ !p.startsWith("http://www.w3.org/2004/02/skos/core")
+ )
+ {return;}
+ if(p.equals("http://www.w3.org/2004/02/skos/core#broader")){
+ p=subclass;
+ }
+ else if(p.equals("http://www.w3.org/2004/02/skos/core#subject")){
+ p=type;
+ }
+ else {}
+ }
+
+ //save for further processing
+ al.add(o);
+
+ // type classes
+ if(isClass(o)){
+ classes.add(o);
+ if(isClass(s))p=subclass;
+ else p=type;
+ }
+ else {
+ instances.add(o);
+ this.properties.add(p);
+ }
+
+
+
+ //maketriples
+ try{
+ this.triples.add(makeTriples(s, p, o));
+ //fw.write(makeTriples(subject, predtmp, objtmp));
+ }catch (Exception e) {e.printStackTrace();}
+
+
+ return;
+ }
+//
+ /**
+ * also makes subclass property between classes
+ *
+ * @param s
+ * @param p
+ * @param o
+ * @return triple in the n triple notation
+ */
+ public String makeTriples(String s,String p, String o){
+ //s=replaceNamespace(s);
+ //p=replaceNamespace(p);
+ //o=replaceNamespace(o);
+ String ret="";
+ ret="<"+s+"> <"+p+"> <"+o+">.\n";
+ return ret;
+ }
+
+ /**
+ * decides if an object is treated as a class
+ *
+ * @param obj
+ * @return true if obj is in the defaultClassesList
+ */
+ public boolean isClass(String obj){
+
+ boolean retval=false;
+ for (String defclass : defaultClasses) {
+ if(obj.contains(defclass))retval=true;
+ }
+ return retval;
+ }
+
+
+ /**
+ * @see java.lang.Object#finalize()
+ */
+ @Override
+ public void finalize(){
+ typeProperties();
+ typeClasses();
+ typeInstances();
+ }
+
+ public void typeProperties(){
+ String rdfns="http://www.w3.org/1999/02/22-rdf-syntax-ns";
+ String owlns="http://www.w3.org/2002/07/owl";
+ Iterator<String> it=properties.iterator();
+ String current="";
+ while (it.hasNext()){
+ try{
+ current=it.next();
+ if(current.equals(subclass))continue;
+ if(current.contains(rdfns)||current.contains(owlns)){/*DO NOTHING*/}
+ else {this.triples.add(makeTriples(current,type,objectProperty));}
+ }catch (Exception e) {}
+
+ }
+ }
+ public void typeClasses(){
+ Iterator<String> it=classes.iterator();
+ String current="";
+ while (it.hasNext()){
+ try{
+ current=it.next();
+ this.triples.add(makeTriples(current,type,classns));
+ }catch (Exception e) {}
+ }
+ }
+ public void typeInstances(){
+ Iterator<String> it=instances.iterator();
+ String current="";
+ while (it.hasNext()){
+ try{
+ current=it.next();
+ this.triples.add(makeTriples(current,type,thing));
+ }catch (Exception e) {}
+ }
+ }
+
+ /**
+ * debug print turn on print_flag
+ * @param s
+ */
+ public void p(String s){
+ if(print_flag)
+ System.out.println(s);
+ }
+
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java 2007-10-08 12:05:59 UTC (rev 180)
@@ -0,0 +1,123 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+/**
+ *
+ * This class produces sparql queries
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class QueryMaker {
+ //Good
+ /*public static String owl ="http://www.w3.org/2002/07/owl#";
+ public static String xsd="http://www.w3.org/2001/XMLSchema#";
+ public static String rdfs="http://www.w3.org/2000/01/rdf-schema#";
+ public static String rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";
+ public static String base="http://dbpedia.org/resource/";
+ public static String dbpedia2="http://dbpedia.org/property/";
+ public static String dbpedia="http://dbpedia.org/";
+
+
+ //BAD
+ public static String skos="http://www.w3.org/2004/02/skos/core#";
+ public static String foaf="http://xmlns.com/foaf/0.1/";
+ public static String dc="http://purl.org/dc/elements/1.1/";
+ public static String foreign="http://dbpedia.org/property/wikipage-";
+ public static String sameAs="http://www.w3.org/2002/07/owl#sameAs";
+ public static String reference="http://dbpedia.org/property/reference";*/
+
+ int tempyago=0;
+
+ /**
+ * reads all the options and makes the sparql query accordingly
+ * @param subject
+ * @param sf special object encapsulating all options
+ * @return sparql query
+ */
+ public String makeQueryFilter(String subject, SparqlFilter sf){
+
+
+ String Filter="";
+ if(!sf.useLiterals)Filter+="!isLiteral(?object))";
+ for (String p : sf.getPredFilter()) {
+ Filter+="\n" + filterPredicate(p);
+ }
+ for (String o : sf.getObjFilter()) {
+ Filter+="\n" + filterObject(o);
+ }
+
+
+ String ret=
+ "SELECT * WHERE { \n" +
+ "<"+
+ subject+
+
+ "> ?predicate ?object.\n" +
+ "FILTER( \n" +
+ "(" +Filter+").}";
+ //System.out.println(ret);
+ return ret;
+ }
+
+
+ /*public String makeQueryDefault(String subject){
+ String ret=
+ "SELECT * WHERE { \n" +
+ "<"+
+ subject+
+
+ "> ?predicate ?object.\n" +
+ "FILTER( \n" +
+ "(!isLiteral(?object))" +
+ "\n" + filterPredicate(skos)+
+ //"\n" + filterObject(skos)+
+ "\n" + filterPredicate(foaf)+
+ "\n" + filterObject(foaf)+
+ "\n" + filterPredicate(foreign)+
+ "\n" + filterPredicate(sameAs)+
+ "\n" + filterPredicate(reference)+
+ ")." +
+ " }";
+
+ //System.out.println(ret);
+ return ret;
+}*/
+
+ /**
+ * add a new object filter
+ * (objects are filtered out of sparql result)
+ * @param ns namespace
+ * @return
+ */
+ public String filterObject(String ns){
+ return "&&( !regex((?object), '"+ns+"') )";
+ }
+ /**
+ * add a new object filter
+ * (objects are filtered out of sparql result)
+ * @param ns namespace
+ * * @return
+ */
+ public String filterPredicate(String ns){
+ return "&&( !regex(str(?predicate), '"+ns+"') )";
+ }
+}
Added: trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java 2007-10-08 12:05:59 UTC (rev 180)
@@ -0,0 +1,126 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+import java.io.BufferedInputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.net.InetAddress;
+import java.net.Socket;
+import java.net.URLEncoder;
+
+
+
+/**
+ *
+ * This class makes a simple http request
+ * the dbpedia url is hardcoded
+ *
+ * @author Sebastian Hellmann
+ *
+ *
+ */
+public class SimpleHTTPRequest {
+ static final char value[]={13,10};
+ static final String cut=new String(value);
+
+
+
+
+ public String sendAndReceive(InetAddress ia, int port, String sparql){
+ String retval="";
+ //
+
+ byte resp[]=null;
+
+ try{
+ Socket SparqlServer=new Socket(ia,port);
+ String request=makeHeader(sparql);
+ // send request
+ (SparqlServer.getOutputStream()).write(request.getBytes());
+
+ //get Response
+ resp=readBuffer(new BufferedInputStream(SparqlServer.getInputStream()));
+ retval=new String(resp);
+ retval=subtractResponseHeader(retval);
+ //retval="||"+retval;
+
+ SparqlServer.close();
+
+
+
+ }
+ catch(Exception e){e.printStackTrace();}
+ //System.out.println("got it");
+ return retval;
+
+ }//down
+
+ public static byte[] readBuffer(InputStream IS)
+ throws IOException{
+ byte buffer[] = new byte[0xffff];
+ int nbytes=0;
+ byte resp[]=new byte[0];
+ while ((nbytes=IS.read(buffer))!=-1) {
+ byte tmp[]=new byte[resp.length+nbytes];
+ int i=0;
+ for (;i<resp.length;i++){
+ tmp[i]=resp[i];
+ }
+ for(int a=0;a<nbytes;a++,i++){
+ tmp[i]=buffer[a];
+ }
+ resp=tmp;
+ }
+ return resp;
+ }
+
+ public String subtractResponseHeader(String in){
+ //System.out.println(in.indexOf(cut+""+cut));
+ return in.substring(in.indexOf(cut+""+cut)+4);
+
+
+ }
+
+ public String makeHeader(String query){
+
+
+ String RequestHeader="";
+ try{
+
+ RequestHeader="GET /sparql?default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
+ //"SELECT%20%2A%20WHERE%20%7B%20%3Chttp%3A%2F%2Fdbpedia.org%2Fresource%2FAristotle%3E%20%3Fa%20%3Fb%20%7D%20" +
+ URLEncoder.encode(query, "UTF-8")+
+ //query+// URLencode
+ "&format=application%2Fsparql-results%2Bxml HTTP/1.1"+cut+
+ "Host: dbpedia.openlinksw.com"+cut+
+ "Connection: close"+cut+
+ //"Accept-Encoding: gzip"+cut+
+ "Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5"+cut+
+ "Accept-Language: de-de,de;q=0.8,en-us;q=0.5,en;q=0.3"+cut+
+ "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7"+cut+
+ "User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24"+cut+
+ cut;
+ }catch (Exception e) {e.printStackTrace();}
+ return RequestHeader;
+
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java 2007-10-08 12:05:59 UTC (rev 180)
@@ -0,0 +1,119 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+/**
+ *
+ *
+ * encapsulates all the options
+ * see the documentation for more help
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class SparqlFilter {
+ public int mode=0;
+ // 0 yago, 1 only cat, 2 skos+cat
+ String[] PredFilter=null;
+ String[] ObjFilter=null;
+ boolean useLiterals=false;
+
+
+ String[] yagoPredFilterDefault={
+ "http://www.w3.org/2004/02/skos/core",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/property/wikipage-",
+ "http://www.w3.org/2002/07/owl#sameAs",
+ "http://dbpedia.org/property/reference" };
+ String[] yagoObjFilterDefault={
+ "http://dbpedia.org/resource/Category:Articles_",
+ "http://dbpedia.org/resource/Category:Wikipedia_",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/resource/Category",
+ "http://dbpedia.org/resource/Template",
+ "http://upload.wikimedia.org/wikipedia/commons"};
+
+ String[] onlyCatPredFilterDefault={
+ "http://www.w3.org/2004/02/skos/core",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/property/wikipage-",
+ "http://www.w3.org/2002/07/owl#sameAs",
+ "http://dbpedia.org/property/reference" };
+ String[] onlyCatObjFilterDefault={
+ "http://dbpedia.org/resource/Category:Articles_",
+ "http://dbpedia.org/resource/Category:Wikipedia_",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/class/yago",
+ "http://dbpedia.org/resource/Template",
+ "http://upload.wikimedia.org/wikipedia/commons"};
+
+ String[] skosPredFilterDefault={
+ "http://www.w3.org/2004/02/skos/core#narrower",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/property/wikipage-",
+ "http://www.w3.org/2002/07/owl#sameAs",
+ "http://dbpedia.org/property/reference" };
+ String[] skosObjFilterDefault={
+ "http://dbpedia.org/resource/Category:Articles_",
+ "http://dbpedia.org/resource/Category:Wikipedia_",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/class/yago",
+ "http://dbpedia.org/resource/Template",
+ "http://upload.wikimedia.org/wikipedia/commons"};
+
+ public SparqlFilter(int mode, String[] pred, String[] obj) {
+ if (mode==-1 && (pred==null || pred.length==0 || obj==null||obj.length==0))
+ {mode=0;}
+
+ switch (mode){
+ case 0: //yago
+ ObjFilter=yagoObjFilterDefault;
+ PredFilter=yagoPredFilterDefault;
+ break;
+ case 1: // only Categories
+ ObjFilter=onlyCatObjFilterDefault;
+ PredFilter=onlyCatPredFilterDefault;
+ break;
+ case 2: // there are some other changes to, which are made directly in other functions
+ ObjFilter=skosObjFilterDefault;
+ PredFilter=skosPredFilterDefault;
+ break;
+ default:
+ ObjFilter=obj;
+ PredFilter=pred;
+ break;
+
+ }}
+ public SparqlFilter(int mode, String[] pred, String[] obj,boolean uselits) throws Exception{
+ this(mode, pred,obj);
+ this.useLiterals=uselits;
+ }
+
+ public String[] getObjFilter(){
+ return this.ObjFilter;
+ }
+ public String[] getPredFilter(){
+ return this.PredFilter;
+ }
+
+
+
+ }
+
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <sk...@us...> - 2007-10-10 09:01:59
|
Revision: 205
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=205&view=rev
Author: sknappe
Date: 2007-10-10 02:01:57 -0700 (Wed, 10 Oct 2007)
Log Message:
-----------
option url is now used
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java
Modified: trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java 2007-10-09 15:52:15 UTC (rev 204)
+++ trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java 2007-10-10 09:01:57 UTC (rev 205)
@@ -20,6 +20,7 @@
package org.dllearner.kb;
import java.net.InetAddress;
+import java.net.URL;
import java.util.ArrayList;
import java.util.HashSet;
import java.util.Iterator;
@@ -75,7 +76,7 @@
* @param defClasses
*/
public OntologyCollector(String[] subjectList,int numberOfRecursions,
- int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses, String format){
+ int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses, String format, URL url){
this.subjectList=subjectList;
this.numberOfRecursions=numberOfRecursions;
this.format=format;
@@ -88,7 +89,7 @@
try{
this.sf=new SparqlFilter(filterMode,FilterPredList,FilterObjList);
- this.ia=InetAddress.getByName("dbpedia.openlinksw.com");
+ this.ia=InetAddress.getByName(url.getHost());
//this.fw=new FileWriter(new File(System.currentTimeMillis()+".nt"),true);
this.properties=new HashSet<String>();
this.classes=new HashSet<String>();
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java 2007-10-09 15:52:15 UTC (rev 204)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java 2007-10-10 09:01:57 UTC (rev 205)
@@ -125,7 +125,7 @@
System.out.println("SparqlModul: Collecting Ontology");
String[] a=new String[0];
OntologyCollector oc=new OntologyCollector(instances.toArray(a), numberOfRecursions,
- filterMode, Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format);
+ filterMode, Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format,url);
String ont=oc.collectOntology();
if (format.equals("N-TRIPLES")||dumpToFile){
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <sk...@us...> - 2007-10-10 12:37:17
|
Revision: 209
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=209&view=rev
Author: sknappe
Date: 2007-10-10 05:37:11 -0700 (Wed, 10 Oct 2007)
Log Message:
-----------
Renamed classes to begin with Sparqle and deleted SimpleHttpRequest
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java
Added Paths:
-----------
trunk/src/dl-learner/org/dllearner/kb/SparqlCache.java
trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java
trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java
Removed Paths:
-------------
trunk/src/dl-learner/org/dllearner/kb/Cache.java
trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java
trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java
trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java
Deleted: trunk/src/dl-learner/org/dllearner/kb/Cache.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/Cache.java 2007-10-10 12:28:02 UTC (rev 208)
+++ trunk/src/dl-learner/org/dllearner/kb/Cache.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -1,210 +0,0 @@
-/**
- * Copyright (C) 2007, Sebastian Hellmann
- *
- * This file is part of DL-Learner.
- *
- * DL-Learner is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * DL-Learner is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <http://www.gnu.org/licenses/>.
- *
- */
-package org.dllearner.kb;
-
-import java.io.File;
-import java.io.FileInputStream;
-import java.io.FileOutputStream;
-import java.io.ObjectInputStream;
-import java.io.ObjectOutputStream;
-import java.io.Serializable;
-import java.net.URLEncoder;
-/**
- *
- * This is a primitive cache.
- * The objects of this class can be either the cache itself or just on entry in the cache
- *
- * the cache remembers: a timestamp, the original sparql-query, the result
- * key is the subject http://dbpedia.org/resource/Angela_Merkel which is first urlencoded
- * and so serves as the hash for the filename.
- * Cache validates if timestamp too old and Sparql-Query the same
- * before returning the SPARQL xml-result
- *
- * @author Sebastian Hellmann
- *
- */
-public class Cache implements Serializable{
-
-
- final static long serialVersionUID=104;
- transient String basedir="";
- transient String fileending=".cache";
- long timestamp;
- String content="";
- long daysoffreshness=15;
- long multiplier=24*60*60*1000;//h m s ms
- String sparqlquery="";
-
-
- /**
- * Constructor for the cache itself.
- * Called once at the beginning
- *
- * @param path Where the base path to the cache is
- */
- public Cache(String path){
- this.basedir=path+File.separator;
- if(!new File(path).exists())
- {System.out.println(new File(path).mkdir());;}
-
- }
-
-//
- /**
- * Constructor for single cache object(one entry)
- *
- * @param content the sparql xml result
- * @param sparql the sparql query
- */
- public Cache(String content, String sparql){
- this.content=content;
- this.sparqlquery=sparql;
- this.timestamp=System.currentTimeMillis();
- }
-
-
-
- /**
- * use only on the cache object describing the cache itself
- *
- * @param key the individual
- * @param sparql the sparql query
- * @return the cached sparql result or null
- */
- public String get(String key, String sparql){
- //System.out.println("get From "+key);
- String ret=null;
- try{
- Cache c =readFromFile(makeFilename(key)) ;
- if(c==null)return null;
- //System.out.println(" file found");
- if(!c.checkFreshness())return null;
- //System.out.println("fresh");
- if(!c.validate(sparql))return null;
- //System.out.println("valid");
- ret=c.content;
- }catch (Exception e) {e.printStackTrace();}
- return ret;
- }
-
- /**
- *
- * constructor for single cache object(one entry)
- *
- * @param key the individual
- * @param content the sparql result
- * @param sparql the sparql query
- */
- public void put(String key, String content, String sparql){
- //System.out.println("put into "+key);
- Cache c=new Cache(content,sparql);
- putIntoFile(makeFilename(key), c);
- }
-
-
- /**
- * to normalize the filenames
- *
- * @param key
- * @return
- */
- String makeFilename(String key){
- String ret="";
- try{
- ret=basedir+URLEncoder.encode(key, "UTF-8")+fileending;
- }catch (Exception e) {e.printStackTrace();}
- return ret;
- }
-
- /**
- * how old is the result
- * @return
- */
- boolean checkFreshness(){
- if((System.currentTimeMillis()-this.timestamp)<=(daysoffreshness*multiplier))
- //fresh
- return true;
- else return false;
- }
-
-
- /**
- * some sparql query
- * @param sparql
- * @return
- */
- boolean validate(String sparql){
- if(this.sparqlquery.equals(sparql))
- //valid
- return true;
- else return false;
- }
-
- /**
- * makes a new file if none exists
- * @param Filename
- */
- public void checkFile(String Filename){
- if(!new File(Filename).exists()){
- try{
- new File(Filename).createNewFile();
- }catch (Exception e) {e.printStackTrace();}
-
- }
-
- }
-
- /**
- * internal saving function
- * puts a cache object into a file
- *
- * @param Filename
- * @param content
- */
- public void putIntoFile(String Filename,Cache content){
- try{
- //FileWriter fw=new FileWriter(new File(Filename),true);
- FileOutputStream fos = new FileOutputStream( Filename , false );
- ObjectOutputStream o = new ObjectOutputStream( fos );
- o.writeObject( content );
- fos.flush();
- fos.close();
- }catch (Exception e) {System.out.println("Not in cache creating: "+Filename);}
- }
-
- /**
- * internal retrieval function
- *
- * @param Filename
- * @return one entry object
- */
- public Cache readFromFile(String Filename){
- Cache content=null;
- try{
- FileInputStream fos = new FileInputStream( Filename );
- ObjectInputStream o = new ObjectInputStream( fos );
- content=(Cache)o.readObject();
- //FileReader fr=new FileReader(new File(Filename,"r"));
- //BufferedReader br=new BufferedReader(fr);
- }catch (Exception e) {}
- return content;
-
- }
-}
Deleted: trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java 2007-10-10 12:28:02 UTC (rev 208)
+++ trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -1,422 +0,0 @@
-/**
- * Copyright (C) 2007, Sebastian Hellmann
- *
- * This file is part of DL-Learner.
- *
- * DL-Learner is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * DL-Learner is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <http://www.gnu.org/licenses/>.
- *
- */
-package org.dllearner.kb;
-
-import java.io.BufferedReader;
-import java.io.IOException;
-import java.io.InputStream;
-import java.io.InputStreamReader;
-import java.io.OutputStream;
-import java.io.OutputStreamWriter;
-import java.net.HttpURLConnection;
-import java.net.InetAddress;
-import java.net.URL;
-import java.net.URLEncoder;
-import java.util.ArrayList;
-import java.util.HashSet;
-import java.util.Iterator;
-
-import org.dllearner.utilities.Files;
-
-/**
- * This class collects the ontology from dbpedia,
- * everything is saved in hashsets, so the doublettes are taken care of
- *
- *
- * @author Sebastian Hellmann
- *
- */
-public class OntologyCollector {
-
- boolean print_flag=false;
- SimpleHTTPRequest s;
- QueryMaker q;
- Cache c;
- URL url;
- SparqlFilter sf;
- String[] subjectList;
- int numberOfRecursions;
- HashSet<String> properties;
- HashSet<String> classes;
- HashSet<String> instances;
- HashSet<String> triples;
- String format;
- static final char value[]={13,10};
- static final String cut=new String(value);
-
- // some namespaces
- String subclass="http://www.w3.org/2000/01/rdf-schema#subClassOf";
- String type="http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
- String objectProperty="http://www.w3.org/2002/07/owl#ObjectProperty";
- String classns="http://www.w3.org/2002/07/owl#Class";
- String thing="http://www.w3.org/2002/07/owl#Thing";
-
-
- String[] defaultClasses={
- "http://dbpedia.org/class/yago",
- "http://dbpedia.org/resource/Category:",
- "http://dbpedia.org/resource/Template:",
- "http://www.w3.org/2004/02/skos/core",
- "http://dbpedia.org/class/"}; //TODO FEHLER hier fehlt yago
-
-
- /**
- *
- *
- * @param subjectList
- * @param numberOfRecursions
- * @param filterMode
- * @param FilterPredList
- * @param FilterObjList
- * @param defClasses
- */
- public OntologyCollector(String[] subjectList,int numberOfRecursions,
- int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses, String format, URL url){
- this.subjectList=subjectList;
- this.numberOfRecursions=numberOfRecursions;
- this.format=format;
- this.s=new SimpleHTTPRequest();
- this.q=new QueryMaker();
- this.c=new Cache("cache");
- if(defClasses!=null && defClasses.length>0 ){
- this.defaultClasses=defClasses;
- }
-
- try{
- this.sf=new SparqlFilter(filterMode,FilterPredList,FilterObjList);
- this.url=url;
- this.properties=new HashSet<String>();
- this.classes=new HashSet<String>();
- this.instances=new HashSet<String>();
- this.triples=new HashSet<String>();
- }catch (Exception e) {e.printStackTrace();}
-
- }
- /**
- * first collects the ontology
- * then types everything so it becomes owl-dl
- *
- * @return all triples in n-triple format
- */
- public String collectOntology(){
- getRecursiveList(subjectList, numberOfRecursions);
- finalize();
- String ret="";
- for (Iterator<String> iter = triples.iterator(); iter.hasNext();) {
- ret += iter.next();
-
- }
- return ret;
- }
-
- /**
- * calls getRecursive for each subject in list
- * @param subjects
- * @param NumberofRecursions
- */
- public void getRecursiveList(String[] subjects,int NumberofRecursions){
- for (int i = 0; i < subjects.length; i++) {
- getRecursive(subjects[i], NumberofRecursions);
-
- }
-
- }
-
- /**
- * gets all triples until numberofrecursion-- gets 0
- *
- * @param StartingSubject
- * @param NumberofRecursions
- */
- public void getRecursive(String StartingSubject,int NumberofRecursions){
- System.out.print("SparqlModul: Depth: "+NumberofRecursions+" @ "+StartingSubject+" ");
- if(NumberofRecursions<=0)
- { return;
- }
- else {NumberofRecursions--;}
- //System.out.println(NumberofRecursions);
- try{
-
- String sparql=q.makeQueryFilter(StartingSubject,this.sf);
- p(sparql);
- p("*******************");
- // checks cache
- String FromCache=c.get(StartingSubject, sparql);
- String xml;
- // if not in cache get it from dbpedia
- if(FromCache==null){
- xml=sendAndReceive(sparql);
- c.put(StartingSubject, xml, sparql);
- System.out.print("\n");
- }
- else{
- xml=FromCache;
- System.out.println("FROM CACHE");
- }
- p(xml);
- p("***********************");
- // get new Subjects
- String[] newSubjects=processResult(StartingSubject,xml);
-
- for (int i = 0; (i < newSubjects.length)&& NumberofRecursions!=0; i++) {
- getRecursive(newSubjects[i], NumberofRecursions);
- }
-
- //System.out.println(xml);
- }catch (Exception e) {e.printStackTrace();}
-
- }
-
- /**
- * process the sparql result xml in a simple manner
- *
- *
- * @param subject
- * @param xml
- * @return list of new individuals
- */
- public String[] processResult(String subject,String xml){
- //TODO if result is empty, catch exceptions
- String one="<binding name=\"predicate\"><uri>";
- String two="<binding name=\"object\"><uri>";
- String end="</uri></binding>";
- String predtmp="";
- String objtmp="";
- ArrayList<String> al=new ArrayList<String>();
-
- while(xml.indexOf(one)!=-1){
- //get pred
- xml=xml.substring(xml.indexOf(one)+one.length());
- predtmp=xml.substring(0,xml.indexOf(end));
- //getobj
- xml=xml.substring(xml.indexOf(two)+two.length());
- objtmp=xml.substring(0,xml.indexOf(end));
-
- // writes the triples and resources in the hashsets
- // also fills the arraylist al
- processTriples(subject, predtmp, objtmp,al);
- //System.out.println(al.size());
-
- }
-
- // convert al to list
- Object[] o=al.toArray();
- String[] ret=new String[o.length];
- for (int i = 0; i < o.length; i++) {
- ret[i]=(String)o[i];
- }
- return ret;
- //return (String[])al.toArray();
- //System.out.println(xml);
- }
-
-
-
- /**
- *
- * writes the triples and resources in the hashsets
- * also fills the arraylist al with new individals for further processing
- * @param s
- * @param p
- * @param o
- * @param al
- */
- public void processTriples(String s,String p, String o,ArrayList<String> al){
- // the next two lines bump out some inconsistencies within dbpedia
- String t="/Category";
- if(s.equals(t) || o.equals(t))return ;
-
- if(sf.mode==2)
- {
- if( o.startsWith("http://dbpedia.org/resource/Category:")
- &&
- !p.startsWith("http://www.w3.org/2004/02/skos/core")
- )
- {return;}
- if(p.equals("http://www.w3.org/2004/02/skos/core#broader")){
- p=subclass;
- }
- else if(p.equals("http://www.w3.org/2004/02/skos/core#subject")){
- p=type;
- }
- else {}
- }
-
- //save for further processing
- al.add(o);
-
- // type classes
- if(isClass(o)){
- classes.add(o);
- if(isClass(s))p=subclass;
- else p=type;
- }
- else {
- instances.add(o);
- this.properties.add(p);
- }
-
-
-
- //maketriples
- try{
- this.triples.add(makeTriples(s, p, o));
- //fw.write(makeTriples(subject, predtmp, objtmp));
- }catch (Exception e) {e.printStackTrace();}
-
-
- return;
- }
-//
- /**
- * also makes subclass property between classes
- *
- * @param s
- * @param p
- * @param o
- * @return triple in the n triple notation
- */
- public String makeTriples(String s,String p, String o){
- String ret="";
- if (format.equals("N-TRIPLES")) ret="<"+s+"> <"+p+"> <"+o+">.\n";
- else if (format.equals("KB")){
- if (p.equals("http://www.w3.org/1999/02/22-rdf-syntax-ns#type")) ret="\""+o+"\"(\""+s+"\").\n";
- else ret="\""+p+"\"(\""+s+"\",\""+o+"\").\n";
- }
- return ret;
- }
-
- /**
- * decides if an object is treated as a class
- *
- * @param obj
- * @return true if obj is in the defaultClassesList
- */
- public boolean isClass(String obj){
-
- boolean retval=false;
- for (String defclass : defaultClasses) {
- if(obj.contains(defclass))retval=true;
- }
- return retval;
- }
-
-
- /**
- * @see java.lang.Object#finalize()
- */
- @Override
- public void finalize(){
- typeProperties();
- typeClasses();
- typeInstances();
- }
-
- public void typeProperties(){
- String rdfns="http://www.w3.org/1999/02/22-rdf-syntax-ns";
- String owlns="http://www.w3.org/2002/07/owl";
- Iterator<String> it=properties.iterator();
- String current="";
- while (it.hasNext()){
- try{
- current=it.next();
- if(current.equals(subclass))continue;
- if(current.contains(rdfns)||current.contains(owlns)){/*DO NOTHING*/}
- else {this.triples.add(makeTriples(current,type,objectProperty));}
- }catch (Exception e) {}
-
- }
- }
- public void typeClasses(){
- Iterator<String> it=classes.iterator();
- String current="";
- while (it.hasNext()){
- try{
- current=it.next();
- this.triples.add(makeTriples(current,type,classns));
- }catch (Exception e) {}
- }
- }
- public void typeInstances(){
- Iterator<String> it=instances.iterator();
- String current="";
- while (it.hasNext()){
- try{
- current=it.next();
- this.triples.add(makeTriples(current,type,thing));
- }catch (Exception e) {}
- }
- }
-
- /**
- * debug print turn on print_flag
- * @param s
- */
- public void p(String s){
- if(print_flag)
- System.out.println(s);
- }
-
- private String sendAndReceive(String sparql) {
- StringBuilder answer = new StringBuilder();
-
- // String an Sparql-Endpoint schicken
- HttpURLConnection connection;
-
- try {
- connection = (HttpURLConnection) url.openConnection();
- connection.setDoOutput(true);
-
- connection.addRequestProperty("Host", "dbpedia.openlinksw.com");
- connection.addRequestProperty("Connection","close");
- connection.addRequestProperty("Accept","text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5");
- connection.addRequestProperty("Accept-Language","de-de,de;q=0.8,en-us;q=0.5,en;q=0.3");
- connection.addRequestProperty("Accept-Charset","ISO-8859-1,utf-8;q=0.7,*;q=0.7");
- connection.addRequestProperty("User-Agent","Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
-
- OutputStream os = connection.getOutputStream();
- OutputStreamWriter osw = new OutputStreamWriter(os);
- osw.write("default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
- URLEncoder.encode(sparql, "UTF-8")+
- "&format=application%2Fsparql-results%2Bxml");
- osw.close();
-
- // receive answer
- InputStream is = connection.getInputStream();
- InputStreamReader isr = new InputStreamReader(is);
- BufferedReader br = new BufferedReader(isr);
-
- String line;
- do {
- line = br.readLine();
- if(line!=null)
- answer.append(line);
- } while (line != null);
-
- br.close();
-
- } catch (IOException e) {
- System.out.println("Communication problem with Sparql Server.");
- System.exit(0);
- }
-
- return answer.toString();
- }
-}
Deleted: trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java 2007-10-10 12:28:02 UTC (rev 208)
+++ trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -1,123 +0,0 @@
-/**
- * Copyright (C) 2007, Sebastian Hellmann
- *
- * This file is part of DL-Learner.
- *
- * DL-Learner is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * DL-Learner is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <http://www.gnu.org/licenses/>.
- *
- */
-package org.dllearner.kb;
-
-/**
- *
- * This class produces sparql queries
- *
- * @author Sebastian Hellmann
- *
- */
-public class QueryMaker {
- //Good
- /*public static String owl ="http://www.w3.org/2002/07/owl#";
- public static String xsd="http://www.w3.org/2001/XMLSchema#";
- public static String rdfs="http://www.w3.org/2000/01/rdf-schema#";
- public static String rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";
- public static String base="http://dbpedia.org/resource/";
- public static String dbpedia2="http://dbpedia.org/property/";
- public static String dbpedia="http://dbpedia.org/";
-
-
- //BAD
- public static String skos="http://www.w3.org/2004/02/skos/core#";
- public static String foaf="http://xmlns.com/foaf/0.1/";
- public static String dc="http://purl.org/dc/elements/1.1/";
- public static String foreign="http://dbpedia.org/property/wikipage-";
- public static String sameAs="http://www.w3.org/2002/07/owl#sameAs";
- public static String reference="http://dbpedia.org/property/reference";*/
-
- int tempyago=0;
-
- /**
- * reads all the options and makes the sparql query accordingly
- * @param subject
- * @param sf special object encapsulating all options
- * @return sparql query
- */
- public String makeQueryFilter(String subject, SparqlFilter sf){
-
-
- String Filter="";
- if(!sf.useLiterals)Filter+="!isLiteral(?object))";
- for (String p : sf.getPredFilter()) {
- Filter+="\n" + filterPredicate(p);
- }
- for (String o : sf.getObjFilter()) {
- Filter+="\n" + filterObject(o);
- }
-
-
- String ret=
- "SELECT * WHERE { \n" +
- "<"+
- subject+
-
- "> ?predicate ?object.\n" +
- "FILTER( \n" +
- "(" +Filter+").}";
- //System.out.println(ret);
- return ret;
- }
-
-
- /*public String makeQueryDefault(String subject){
- String ret=
- "SELECT * WHERE { \n" +
- "<"+
- subject+
-
- "> ?predicate ?object.\n" +
- "FILTER( \n" +
- "(!isLiteral(?object))" +
- "\n" + filterPredicate(skos)+
- //"\n" + filterObject(skos)+
- "\n" + filterPredicate(foaf)+
- "\n" + filterObject(foaf)+
- "\n" + filterPredicate(foreign)+
- "\n" + filterPredicate(sameAs)+
- "\n" + filterPredicate(reference)+
- ")." +
- " }";
-
- //System.out.println(ret);
- return ret;
-}*/
-
- /**
- * add a new object filter
- * (objects are filtered out of sparql result)
- * @param ns namespace
- * @return
- */
- public String filterObject(String ns){
- return "&&( !regex((?object), '"+ns+"') )";
- }
- /**
- * add a new object filter
- * (objects are filtered out of sparql result)
- * @param ns namespace
- * * @return
- */
- public String filterPredicate(String ns){
- return "&&( !regex(str(?predicate), '"+ns+"') )";
- }
-}
Deleted: trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java 2007-10-10 12:28:02 UTC (rev 208)
+++ trunk/src/dl-learner/org/dllearner/kb/SimpleHTTPRequest.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -1,126 +0,0 @@
-/**
- * Copyright (C) 2007, Sebastian Hellmann
- *
- * This file is part of DL-Learner.
- *
- * DL-Learner is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * DL-Learner is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <http://www.gnu.org/licenses/>.
- *
- */
-package org.dllearner.kb;
-
-import java.io.BufferedInputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.net.InetAddress;
-import java.net.Socket;
-import java.net.URLEncoder;
-
-
-
-/**
- *
- * This class makes a simple http request
- * the dbpedia url is hardcoded
- *
- * @author Sebastian Hellmann
- *
- *
- */
-public class SimpleHTTPRequest {
- static final char value[]={13,10};
- static final String cut=new String(value);
-
-
-
-
- public String sendAndReceive(InetAddress ia, int port, String sparql){
- String retval="";
- //
-
- byte resp[]=null;
-
- try{
- Socket SparqlServer=new Socket(ia,port);
- String request=makeHeader(sparql);
- // send request
- (SparqlServer.getOutputStream()).write(request.getBytes());
-
- //get Response
- resp=readBuffer(new BufferedInputStream(SparqlServer.getInputStream()));
- retval=new String(resp);
- retval=subtractResponseHeader(retval);
- //retval="||"+retval;
-
- SparqlServer.close();
-
-
-
- }
- catch(Exception e){e.printStackTrace();}
- //System.out.println("got it");
- return retval;
-
- }//down
-
- public static byte[] readBuffer(InputStream IS)
- throws IOException{
- byte buffer[] = new byte[0xffff];
- int nbytes=0;
- byte resp[]=new byte[0];
- while ((nbytes=IS.read(buffer))!=-1) {
- byte tmp[]=new byte[resp.length+nbytes];
- int i=0;
- for (;i<resp.length;i++){
- tmp[i]=resp[i];
- }
- for(int a=0;a<nbytes;a++,i++){
- tmp[i]=buffer[a];
- }
- resp=tmp;
- }
- return resp;
- }
-
- public String subtractResponseHeader(String in){
- //System.out.println(in.indexOf(cut+""+cut));
- return in.substring(in.indexOf(cut+""+cut)+4);
-
-
- }
-
- public String makeHeader(String query){
-
-
- String RequestHeader="";
- try{
-
- RequestHeader="GET /sparql?default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
- //"SELECT%20%2A%20WHERE%20%7B%20%3Chttp%3A%2F%2Fdbpedia.org%2Fresource%2FAristotle%3E%20%3Fa%20%3Fb%20%7D%20" +
- URLEncoder.encode(query, "UTF-8")+
- //query+// URLencode
- "&format=application%2Fsparql-results%2Bxml HTTP/1.1"+cut+
- "Host: dbpedia.openlinksw.com"+cut+
- "Connection: close"+cut+
- //"Accept-Encoding: gzip"+cut+
- "Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5"+cut+
- "Accept-Language: de-de,de;q=0.8,en-us;q=0.5,en;q=0.3"+cut+
- "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7"+cut+
- "User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24"+cut+
- cut;
- }catch (Exception e) {e.printStackTrace();}
- return RequestHeader;
-
- }
-
-}
Copied: trunk/src/dl-learner/org/dllearner/kb/SparqlCache.java (from rev 205, trunk/src/dl-learner/org/dllearner/kb/Cache.java)
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlCache.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlCache.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -0,0 +1,210 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.FileOutputStream;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.io.Serializable;
+import java.net.URLEncoder;
+/**
+ *
+ * This is a primitive cache.
+ * The objects of this class can be either the cache itself or just on entry in the cache
+ *
+ * the cache remembers: a timestamp, the original sparql-query, the result
+ * key is the subject http://dbpedia.org/resource/Angela_Merkel which is first urlencoded
+ * and so serves as the hash for the filename.
+ * Cache validates if timestamp too old and Sparql-Query the same
+ * before returning the SPARQL xml-result
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class SparqlCache implements Serializable{
+
+
+ final static long serialVersionUID=104;
+ transient String basedir="";
+ transient String fileending=".cache";
+ long timestamp;
+ String content="";
+ long daysoffreshness=15;
+ long multiplier=24*60*60*1000;//h m s ms
+ String sparqlquery="";
+
+
+ /**
+ * Constructor for the cache itself.
+ * Called once at the beginning
+ *
+ * @param path Where the base path to the cache is
+ */
+ public SparqlCache(String path){
+ this.basedir=path+File.separator;
+ if(!new File(path).exists())
+ {System.out.println(new File(path).mkdir());;}
+
+ }
+
+//
+ /**
+ * Constructor for single cache object(one entry)
+ *
+ * @param content the sparql xml result
+ * @param sparql the sparql query
+ */
+ public SparqlCache(String content, String sparql){
+ this.content=content;
+ this.sparqlquery=sparql;
+ this.timestamp=System.currentTimeMillis();
+ }
+
+
+
+ /**
+ * use only on the cache object describing the cache itself
+ *
+ * @param key the individual
+ * @param sparql the sparql query
+ * @return the cached sparql result or null
+ */
+ public String get(String key, String sparql){
+ //System.out.println("get From "+key);
+ String ret=null;
+ try{
+ SparqlCache c =readFromFile(makeFilename(key)) ;
+ if(c==null)return null;
+ //System.out.println(" file found");
+ if(!c.checkFreshness())return null;
+ //System.out.println("fresh");
+ if(!c.validate(sparql))return null;
+ //System.out.println("valid");
+ ret=c.content;
+ }catch (Exception e) {e.printStackTrace();}
+ return ret;
+ }
+
+ /**
+ *
+ * constructor for single cache object(one entry)
+ *
+ * @param key the individual
+ * @param content the sparql result
+ * @param sparql the sparql query
+ */
+ public void put(String key, String content, String sparql){
+ //System.out.println("put into "+key);
+ SparqlCache c=new SparqlCache(content,sparql);
+ putIntoFile(makeFilename(key), c);
+ }
+
+
+ /**
+ * to normalize the filenames
+ *
+ * @param key
+ * @return
+ */
+ String makeFilename(String key){
+ String ret="";
+ try{
+ ret=basedir+URLEncoder.encode(key, "UTF-8")+fileending;
+ }catch (Exception e) {e.printStackTrace();}
+ return ret;
+ }
+
+ /**
+ * how old is the result
+ * @return
+ */
+ boolean checkFreshness(){
+ if((System.currentTimeMillis()-this.timestamp)<=(daysoffreshness*multiplier))
+ //fresh
+ return true;
+ else return false;
+ }
+
+
+ /**
+ * some sparql query
+ * @param sparql
+ * @return
+ */
+ boolean validate(String sparql){
+ if(this.sparqlquery.equals(sparql))
+ //valid
+ return true;
+ else return false;
+ }
+
+ /**
+ * makes a new file if none exists
+ * @param Filename
+ */
+ public void checkFile(String Filename){
+ if(!new File(Filename).exists()){
+ try{
+ new File(Filename).createNewFile();
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+
+ }
+
+ /**
+ * internal saving function
+ * puts a cache object into a file
+ *
+ * @param Filename
+ * @param content
+ */
+ public void putIntoFile(String Filename,SparqlCache content){
+ try{
+ //FileWriter fw=new FileWriter(new File(Filename),true);
+ FileOutputStream fos = new FileOutputStream( Filename , false );
+ ObjectOutputStream o = new ObjectOutputStream( fos );
+ o.writeObject( content );
+ fos.flush();
+ fos.close();
+ }catch (Exception e) {System.out.println("Not in cache creating: "+Filename);}
+ }
+
+ /**
+ * internal retrieval function
+ *
+ * @param Filename
+ * @return one entry object
+ */
+ public SparqlCache readFromFile(String Filename){
+ SparqlCache content=null;
+ try{
+ FileInputStream fos = new FileInputStream( Filename );
+ ObjectInputStream o = new ObjectInputStream( fos );
+ content=(SparqlCache)o.readObject();
+ //FileReader fr=new FileReader(new File(Filename,"r"));
+ //BufferedReader br=new BufferedReader(fr);
+ }catch (Exception e) {}
+ return content;
+
+ }
+}
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java 2007-10-10 12:28:02 UTC (rev 208)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -124,7 +124,7 @@
public void init() {
System.out.println("SparqlModul: Collecting Ontology");
String[] a=new String[0];
- OntologyCollector oc=new OntologyCollector(instances.toArray(a), numberOfRecursions,
+ SparqlOntologyCollector oc=new SparqlOntologyCollector(instances.toArray(a), numberOfRecursions,
filterMode, Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format,url);
String ont=oc.collectOntology();
Copied: trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java (from rev 208, trunk/src/dl-learner/org/dllearner/kb/OntologyCollector.java)
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -0,0 +1,420 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+import java.io.BufferedReader;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.OutputStream;
+import java.io.OutputStreamWriter;
+import java.net.HttpURLConnection;
+import java.net.InetAddress;
+import java.net.URL;
+import java.net.URLEncoder;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.Iterator;
+
+import org.dllearner.utilities.Files;
+
+/**
+ * This class collects the ontology from dbpedia,
+ * everything is saved in hashsets, so the doublettes are taken care of
+ *
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class SparqlOntologyCollector {
+
+ boolean print_flag=false;
+ SparqlQueryMaker q;
+ SparqlCache c;
+ URL url;
+ SparqlFilter sf;
+ String[] subjectList;
+ int numberOfRecursions;
+ HashSet<String> properties;
+ HashSet<String> classes;
+ HashSet<String> instances;
+ HashSet<String> triples;
+ String format;
+ static final char value[]={13,10};
+ static final String cut=new String(value);
+
+ // some namespaces
+ String subclass="http://www.w3.org/2000/01/rdf-schema#subClassOf";
+ String type="http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
+ String objectProperty="http://www.w3.org/2002/07/owl#ObjectProperty";
+ String classns="http://www.w3.org/2002/07/owl#Class";
+ String thing="http://www.w3.org/2002/07/owl#Thing";
+
+
+ String[] defaultClasses={
+ "http://dbpedia.org/class/yago",
+ "http://dbpedia.org/resource/Category:",
+ "http://dbpedia.org/resource/Template:",
+ "http://www.w3.org/2004/02/skos/core",
+ "http://dbpedia.org/class/"}; //TODO FEHLER hier fehlt yago
+
+
+ /**
+ *
+ *
+ * @param subjectList
+ * @param numberOfRecursions
+ * @param filterMode
+ * @param FilterPredList
+ * @param FilterObjList
+ * @param defClasses
+ */
+ public SparqlOntologyCollector(String[] subjectList,int numberOfRecursions,
+ int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses, String format, URL url){
+ this.subjectList=subjectList;
+ this.numberOfRecursions=numberOfRecursions;
+ this.format=format;
+ this.q=new SparqlQueryMaker();
+ this.c=new SparqlCache("cache");
+ if(defClasses!=null && defClasses.length>0 ){
+ this.defaultClasses=defClasses;
+ }
+
+ try{
+ this.sf=new SparqlFilter(filterMode,FilterPredList,FilterObjList);
+ this.url=url;
+ this.properties=new HashSet<String>();
+ this.classes=new HashSet<String>();
+ this.instances=new HashSet<String>();
+ this.triples=new HashSet<String>();
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+ /**
+ * first collects the ontology
+ * then types everything so it becomes owl-dl
+ *
+ * @return all triples in n-triple format
+ */
+ public String collectOntology(){
+ getRecursiveList(subjectList, numberOfRecursions);
+ finalize();
+ String ret="";
+ for (Iterator<String> iter = triples.iterator(); iter.hasNext();) {
+ ret += iter.next();
+
+ }
+ return ret;
+ }
+
+ /**
+ * calls getRecursive for each subject in list
+ * @param subjects
+ * @param NumberofRecursions
+ */
+ public void getRecursiveList(String[] subjects,int NumberofRecursions){
+ for (int i = 0; i < subjects.length; i++) {
+ getRecursive(subjects[i], NumberofRecursions);
+
+ }
+
+ }
+
+ /**
+ * gets all triples until numberofrecursion-- gets 0
+ *
+ * @param StartingSubject
+ * @param NumberofRecursions
+ */
+ public void getRecursive(String StartingSubject,int NumberofRecursions){
+ System.out.print("SparqlModul: Depth: "+NumberofRecursions+" @ "+StartingSubject+" ");
+ if(NumberofRecursions<=0)
+ { return;
+ }
+ else {NumberofRecursions--;}
+ //System.out.println(NumberofRecursions);
+ try{
+
+ String sparql=q.makeQueryFilter(StartingSubject,this.sf);
+ p(sparql);
+ p("*******************");
+ // checks cache
+ String FromCache=c.get(StartingSubject, sparql);
+ String xml;
+ // if not in cache get it from dbpedia
+ if(FromCache==null){
+ xml=sendAndReceive(sparql);
+ c.put(StartingSubject, xml, sparql);
+ System.out.print("\n");
+ }
+ else{
+ xml=FromCache;
+ System.out.println("FROM CACHE");
+ }
+ p(xml);
+ p("***********************");
+ // get new Subjects
+ String[] newSubjects=processResult(StartingSubject,xml);
+
+ for (int i = 0; (i < newSubjects.length)&& NumberofRecursions!=0; i++) {
+ getRecursive(newSubjects[i], NumberofRecursions);
+ }
+
+ //System.out.println(xml);
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+
+ /**
+ * process the sparql result xml in a simple manner
+ *
+ *
+ * @param subject
+ * @param xml
+ * @return list of new individuals
+ */
+ public String[] processResult(String subject,String xml){
+ //TODO if result is empty, catch exceptions
+ String one="<binding name=\"predicate\"><uri>";
+ String two="<binding name=\"object\"><uri>";
+ String end="</uri></binding>";
+ String predtmp="";
+ String objtmp="";
+ ArrayList<String> al=new ArrayList<String>();
+
+ while(xml.indexOf(one)!=-1){
+ //get pred
+ xml=xml.substring(xml.indexOf(one)+one.length());
+ predtmp=xml.substring(0,xml.indexOf(end));
+ //getobj
+ xml=xml.substring(xml.indexOf(two)+two.length());
+ objtmp=xml.substring(0,xml.indexOf(end));
+
+ // writes the triples and resources in the hashsets
+ // also fills the arraylist al
+ processTriples(subject, predtmp, objtmp,al);
+ //System.out.println(al.size());
+
+ }
+
+ // convert al to list
+ Object[] o=al.toArray();
+ String[] ret=new String[o.length];
+ for (int i = 0; i < o.length; i++) {
+ ret[i]=(String)o[i];
+ }
+ return ret;
+ //return (String[])al.toArray();
+ //System.out.println(xml);
+ }
+
+
+
+ /**
+ *
+ * writes the triples and resources in the hashsets
+ * also fills the arraylist al with new individals for further processing
+ * @param s
+ * @param p
+ * @param o
+ * @param al
+ */
+ public void processTriples(String s,String p, String o,ArrayList<String> al){
+ // the next two lines bump out some inconsistencies within dbpedia
+ String t="/Category";
+ if(s.equals(t) || o.equals(t))return ;
+
+ if(sf.mode==2)
+ {
+ if( o.startsWith("http://dbpedia.org/resource/Category:")
+ &&
+ !p.startsWith("http://www.w3.org/2004/02/skos/core")
+ )
+ {return;}
+ if(p.equals("http://www.w3.org/2004/02/skos/core#broader")){
+ p=subclass;
+ }
+ else if(p.equals("http://www.w3.org/2004/02/skos/core#subject")){
+ p=type;
+ }
+ else {}
+ }
+
+ //save for further processing
+ al.add(o);
+
+ // type classes
+ if(isClass(o)){
+ classes.add(o);
+ if(isClass(s))p=subclass;
+ else p=type;
+ }
+ else {
+ instances.add(o);
+ this.properties.add(p);
+ }
+
+
+
+ //maketriples
+ try{
+ this.triples.add(makeTriples(s, p, o));
+ //fw.write(makeTriples(subject, predtmp, objtmp));
+ }catch (Exception e) {e.printStackTrace();}
+
+
+ return;
+ }
+//
+ /**
+ * also makes subclass property between classes
+ *
+ * @param s
+ * @param p
+ * @param o
+ * @return triple in the n triple notation
+ */
+ public String makeTriples(String s,String p, String o){
+ String ret="";
+ if (format.equals("N-TRIPLES")) ret="<"+s+"> <"+p+"> <"+o+">.\n";
+ else if (format.equals("KB")){
+ if (p.equals("http://www.w3.org/1999/02/22-rdf-syntax-ns#type")) ret="\""+o+"\"(\""+s+"\").\n";
+ else ret="\""+p+"\"(\""+s+"\",\""+o+"\").\n";
+ }
+ return ret;
+ }
+
+ /**
+ * decides if an object is treated as a class
+ *
+ * @param obj
+ * @return true if obj is in the defaultClassesList
+ */
+ public boolean isClass(String obj){
+
+ boolean retval=false;
+ for (String defclass : defaultClasses) {
+ if(obj.contains(defclass))retval=true;
+ }
+ return retval;
+ }
+
+
+ /**
+ * @see java.lang.Object#finalize()
+ */
+ @Override
+ public void finalize(){
+ typeProperties();
+ typeClasses();
+ typeInstances();
+ }
+
+ public void typeProperties(){
+ String rdfns="http://www.w3.org/1999/02/22-rdf-syntax-ns";
+ String owlns="http://www.w3.org/2002/07/owl";
+ Iterator<String> it=properties.iterator();
+ String current="";
+ while (it.hasNext()){
+ try{
+ current=it.next();
+ if(current.equals(subclass))continue;
+ if(current.contains(rdfns)||current.contains(owlns)){/*DO NOTHING*/}
+ else {this.triples.add(makeTriples(current,type,objectProperty));}
+ }catch (Exception e) {}
+
+ }
+ }
+ public void typeClasses(){
+ Iterator<String> it=classes.iterator();
+ String current="";
+ while (it.hasNext()){
+ try{
+ current=it.next();
+ this.triples.add(makeTriples(current,type,classns));
+ }catch (Exception e) {}
+ }
+ }
+ public void typeInstances(){
+ Iterator<String> it=instances.iterator();
+ String current="";
+ while (it.hasNext()){
+ try{
+ current=it.next();
+ this.triples.add(makeTriples(current,type,thing));
+ }catch (Exception e) {}
+ }
+ }
+
+ /**
+ * debug print turn on print_flag
+ * @param s
+ */
+ public void p(String s){
+ if(print_flag)
+ System.out.println(s);
+ }
+
+ private String sendAndReceive(String sparql) {
+ StringBuilder answer = new StringBuilder();
+
+ // String an Sparql-Endpoint schicken
+ HttpURLConnection connection;
+
+ try {
+ connection = (HttpURLConnection) url.openConnection();
+ connection.setDoOutput(true);
+
+ connection.addRequestProperty("Host", "dbpedia.openlinksw.com");
+ connection.addRequestProperty("Connection","close");
+ connection.addRequestProperty("Accept","text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5");
+ connection.addRequestProperty("Accept-Language","de-de,de;q=0.8,en-us;q=0.5,en;q=0.3");
+ connection.addRequestProperty("Accept-Charset","ISO-8859-1,utf-8;q=0.7,*;q=0.7");
+ connection.addRequestProperty("User-Agent","Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
+
+ OutputStream os = connection.getOutputStream();
+ OutputStreamWriter osw = new OutputStreamWriter(os);
+ osw.write("default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
+ URLEncoder.encode(sparql, "UTF-8")+
+ "&format=application%2Fsparql-results%2Bxml");
+ osw.close();
+
+ // receive answer
+ InputStream is = connection.getInputStream();
+ InputStreamReader isr = new InputStreamReader(is);
+ BufferedReader br = new BufferedReader(isr);
+
+ String line;
+ do {
+ line = br.readLine();
+ if(line!=null)
+ answer.append(line);
+ } while (line != null);
+
+ br.close();
+
+ } catch (IOException e) {
+ System.out.println("Communication problem with Sparql Server.");
+ System.exit(0);
+ }
+
+ return answer.toString();
+ }
+}
Copied: trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java (from rev 205, trunk/src/dl-learner/org/dllearner/kb/QueryMaker.java)
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java 2007-10-10 12:37:11 UTC (rev 209)
@@ -0,0 +1,123 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+/**
+ *
+ * This class produces sparql queries
+ *
+ * @author Sebastian Hellmann
+ *
+ */
+public class SparqlQueryMaker {
+ //Good
+ /*public static String owl ="http://www.w3.org/2002/07/owl#";
+ public static String xsd="http://www.w3.org/2001/XMLSchema#";
+ public static String rdfs="http://www.w3.org/2000/01/rdf-schema#";
+ public static String rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";
+ public static String base="http://dbpedia.org/resource/";
+ public static String dbpedia2="http://dbpedia.org/property/";
+ public static String dbpedia="http://dbpedia.org/";
+
+
+ //BAD
+ public static String skos="http://www.w3.org/2004/02/skos/core#";
+ public static String foaf="http://xmlns.com/foaf/0.1/";
+ public static String dc="http://purl.org/dc/elements/1.1/";
+ public static String foreign="http://dbpedia.org/property/wikipage-";
+ public static String sameAs="http://www.w3.org/2002/07/owl#sameAs";
+ public static String reference="http://dbpedia.org/property/reference";*/
+
+ int tempyago=0;
+
+ /**
+ * reads all the options and makes the sparql query accordingly
+ * @param subject
+ * @param sf special object encapsulating all options
+ * @return sparql query
+ */
+ public String makeQueryFilter(String subject, SparqlFilter sf){
+
+
+ String Filter="";
+ if(!sf.useLiterals)Filter+="!isLiteral(?object))";
+ for (String p : sf.getPredFilter()) {
+ Filter+="\n" + filterPredicate(p);
+ }
+ for (String o : sf.getObjFilter()) {
+ Filter+="\n" + filterObject(o);
+ }
+
+
+ String ret=
+ "SELECT * WHERE { \n" +
+ "<"+
+ subject+
+
+ "> ?predicate ?object.\n" +
+ "FILTER( \n" +
+ "(" +Filter+").}";
+ //System.out.println(ret);
+ return ret;
+ }
+
+
+ /*public String makeQueryDefault(String subject){
+ String ret=
+ "SELECT * WHERE { \n" +
+ "<"+
+ subject+
+
+ "> ?predicate ?object.\n" +
+ "FILTER( \n" +
+ "(!isLiteral(?object))" +
+ "\n" + filterPredicate(skos)+
+ //"\n" + filterObject(skos)+
+ "\n" + filterPredicate(foaf)+
+ "\n" + filterObject(foaf)+
+ "\n" + filterPredicate(foreign)+
+ "\n" + filterPredicate(sameAs)+
+ "\n" + filterPredicate(reference)+
+ ")." +
+ " }";
+
+ //System.out.println(ret);
+ return ret;
+}*/
+
+ /**
+ * add a new object filter
+ * (objects are filtered out of sparql result)
+ * @param ns namespace
+ * @return
+ */
+ public String filterObject(String ns){
+ return "&&( !regex((?object), '"+ns+"') )";
+ }
+ /**
+ * add a new object filter
+ * (objects are filtered out of sparql result)
+ * @param ns namespace
+ * * @return
+ */
+ public String filterPredicate(String ns){
+ return "&&( !regex(str(?predicate), '"+ns+"') )";
+ }
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <sk...@us...> - 2007-10-18 15:55:01
|
Revision: 234
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=234&view=rev
Author: sknappe
Date: 2007-10-18 08:54:56 -0700 (Thu, 18 Oct 2007)
Log Message:
-----------
Added functions for DBPedia-Navigator
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java
trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java
trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java
trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java 2007-10-18 15:54:15 UTC (rev 233)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpoint.java 2007-10-18 15:54:56 UTC (rev 234)
@@ -28,6 +28,8 @@
import java.util.Collection;
import java.util.LinkedList;
import java.util.Set;
+import java.util.StringTokenizer;
+import java.util.Vector;
import org.dllearner.core.BooleanConfigOption;
import org.dllearner.core.ConfigEntry;
@@ -56,7 +58,7 @@
private URL url;
private Set<String> instances;
- private URL ntFile;
+ private URL dumpFile;
private int numberOfRecursions;
private int filterMode;
private Set<String> predList;
@@ -65,6 +67,8 @@
private KB kb;
private String format;
private boolean dumpToFile;
+ private boolean useLits=false;
+ private String[] ontArray;
public static String getName() {
return "SPARQL Endpoint";
@@ -81,6 +85,7 @@
options.add(new StringSetConfigOption("classList","a class list"));
options.add(new StringConfigOption("format", "N-TRIPLES or KB format"));
options.add(new BooleanConfigOption("dumpToFile", "wether Ontology from DBPedia is written to a file or not"));
+ options.add(new BooleanConfigOption("useLits","use Literals in SPARQL query"));
return options;
}
@@ -114,6 +119,8 @@
format=(String)entry.getValue();
} else if(option.equals("dumpToFile")){
dumpToFile=(Boolean)entry.getValue();
+ } else if(option.equals("useLits")){
+ useLits=(Boolean)entry.getValue();
}
}
@@ -125,10 +132,16 @@
System.out.println("SparqlModul: Collecting Ontology");
String[] a=new String[0];
SparqlOntologyCollector oc=new SparqlOntologyCollector(instances.toArray(a), numberOfRecursions,
- filterMode, Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format,url);
- String ont=oc.collectOntology();
+ filterMode, Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format,url,useLits);
+ String ont="";
+ if (format.equals("Array")){
+ ontArray=oc.collectOntologyAsArray();
+ }
+ else{
+ ont=oc.collectOntology();
+ }
- if (format.equals("N-TRIPLES")||dumpToFile){
+ if (dumpToFile){
String filename=System.currentTimeMillis()+".nt";
String basedir="cache"+File.separator;
try{
@@ -140,7 +153,7 @@
fw.flush();
fw.close();
- ntFile=(new File(basedir+filename)).toURI().toURL();
+ dumpFile=(new File(basedir+filename)).toURI().toURL();
}catch (Exception e) {e.printStackTrace();}
}
if (format.equals("KB")) {
@@ -159,11 +172,24 @@
*/
@Override
public String toDIG(URI kbURI) {
- if (format.equals("N-TRIPLES")) return JenaOWLDIGConverter.getTellsString(ntFile, OntologyFileFormat.N_TRIPLES, kbURI);
+ if (format.equals("N-TRIPLES")) return JenaOWLDIGConverter.getTellsString(dumpFile, OntologyFileFormat.N_TRIPLES, kbURI);
else return DIGConverter.getDIGString(kb, kbURI).toString();
}
public URL getURL() {
return url;
}
+
+ public String[] getOntArray() {
+ return ontArray;
+ }
+
+ public String[] getSubjects(String label,int limit)
+ {
+ System.out.println("SparqlModul: Collecting Subjects");
+ SparqlOntologyCollector oc=new SparqlOntologyCollector(null, 1,0,null,null,null,null,url,false);
+ String[] ret=oc.getSubjectsFromLabel(label,limit);
+ System.out.println("SparqlModul: ****Finished");
+ return ret;
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java 2007-10-18 15:54:15 UTC (rev 233)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlFilter.java 2007-10-18 15:54:56 UTC (rev 234)
@@ -81,6 +81,7 @@
public SparqlFilter(int mode, String[] pred, String[] obj) {
if (mode==-1 && (pred==null || pred.length==0 || obj==null||obj.length==0))
{mode=0;}
+ else this.mode=mode;
switch (mode){
case 0: //yago
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java 2007-10-18 15:54:15 UTC (rev 233)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlOntologyCollector.java 2007-10-18 15:54:56 UTC (rev 234)
@@ -31,6 +31,8 @@
import java.util.ArrayList;
import java.util.HashSet;
import java.util.Iterator;
+import java.util.StringTokenizer;
+import java.util.Vector;
/**
@@ -85,7 +87,7 @@
* @param defClasses
*/
public SparqlOntologyCollector(String[] subjectList,int numberOfRecursions,
- int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses, String format, URL url){
+ int filterMode, String[] FilterPredList,String[] FilterObjList,String[] defClasses, String format, URL url, boolean useLits){
this.subjectList=subjectList;
this.numberOfRecursions=numberOfRecursions;
this.format=format;
@@ -93,10 +95,9 @@
this.c=new SparqlCache("cache");
if(defClasses!=null && defClasses.length>0 ){
this.defaultClasses=defClasses;
- }
-
+ }
try{
- this.sf=new SparqlFilter(filterMode,FilterPredList,FilterObjList);
+ this.sf=new SparqlFilter(filterMode,FilterPredList,FilterObjList,useLits);
this.url=url;
this.properties=new HashSet<String>();
this.classes=new HashSet<String>();
@@ -122,6 +123,32 @@
return ret;
}
+ public String[] collectOntologyAsArray(){
+ getRecursiveList(subjectList, numberOfRecursions);
+ if (sf.mode!=3) finalize();
+ String[] a=new String[0];
+ return triples.toArray(a);
+ }
+
+ public String[] getSubjectsFromLabel(String label, int limit){
+ System.out.println("Searching for Label: "+label);
+ String sparql=q.makeLabelQuery(label,limit);
+ String FromCache=c.get(label, sparql);
+ String xml;
+ // if not in cache get it from dbpedia
+ if(FromCache==null){
+ xml=sendAndReceive(sparql);
+ c.put(label, xml, sparql);
+ System.out.print("\n");
+ }
+ else{
+ xml=FromCache;
+ System.out.println("FROM CACHE");
+ }
+
+ return processSubjects(xml);
+ }
+
/**
* calls getRecursive for each subject in list
* @param subjects
@@ -191,20 +218,25 @@
public String[] processResult(String subject,String xml){
//TODO if result is empty, catch exceptions
String one="<binding name=\"predicate\"><uri>";
- String two="<binding name=\"object\"><uri>";
+ String two="<binding name=\"object\">";
String end="</uri></binding>";
String predtmp="";
String objtmp="";
ArrayList<String> al=new ArrayList<String>();
-
while(xml.indexOf(one)!=-1){
//get pred
xml=xml.substring(xml.indexOf(one)+one.length());
predtmp=xml.substring(0,xml.indexOf(end));
//getobj
xml=xml.substring(xml.indexOf(two)+two.length());
- objtmp=xml.substring(0,xml.indexOf(end));
+ if (xml.startsWith("<literal xml:lang=\"en\">")){
+ xml=xml.substring(xml.indexOf(">")+1);
+ objtmp=xml.substring(0,xml.indexOf("</literal>"));
+ }
+ else if (xml.startsWith("<uri>")) objtmp=xml.substring(5,xml.indexOf(end));
+ else continue;
+ System.out.println("Pred: "+predtmp+" Obj: "+objtmp);
// writes the triples and resources in the hashsets
// also fills the arraylist al
processTriples(subject, predtmp, objtmp,al);
@@ -280,6 +312,23 @@
return;
}
+
+ private String[] processSubjects(String xml){
+ Vector<String> vec=new Vector<String>();
+ String one="<binding name=\"subject\"><uri>";
+ String end="</uri></binding>";
+ String subject="";
+ while(xml.indexOf(one)!=-1){
+ //get subject
+ xml=xml.substring(xml.indexOf(one)+one.length());
+ subject=xml.substring(0,xml.indexOf(end));
+
+ System.out.println("Subject: "+subject);
+ vec.addElement(subject);
+ }
+ String[] a=new String[vec.size()];
+ return vec.toArray(a);
+ }
//
/**
* also makes subclass property between classes
@@ -296,6 +345,10 @@
if (p.equals("http://www.w3.org/1999/02/22-rdf-syntax-ns#type")) ret="\""+o+"\"(\""+s+"\").\n";
else ret="\""+p+"\"(\""+s+"\",\""+o+"\").\n";
}
+ else if (format.equals("Array"))
+ {
+ ret=s+"<"+p+"<"+o+"\n";
+ }
return ret;
}
@@ -375,7 +428,7 @@
// String an Sparql-Endpoint schicken
HttpURLConnection connection;
-
+
try {
connection = (HttpURLConnection) url.openConnection();
connection.setDoOutput(true);
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java 2007-10-18 15:54:15 UTC (rev 233)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlQueryMaker.java 2007-10-18 15:54:56 UTC (rev 234)
@@ -27,26 +27,7 @@
*
*/
public class SparqlQueryMaker {
- //Good
- /*public static String owl ="http://www.w3.org/2002/07/owl#";
- public static String xsd="http://www.w3.org/2001/XMLSchema#";
- public static String rdfs="http://www.w3.org/2000/01/rdf-schema#";
- public static String rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";
- public static String base="http://dbpedia.org/resource/";
- public static String dbpedia2="http://dbpedia.org/property/";
- public static String dbpedia="http://dbpedia.org/";
-
- //BAD
- public static String skos="http://www.w3.org/2004/02/skos/core#";
- public static String foaf="http://xmlns.com/foaf/0.1/";
- public static String dc="http://purl.org/dc/elements/1.1/";
- public static String foreign="http://dbpedia.org/property/wikipage-";
- public static String sameAs="http://www.w3.org/2002/07/owl#sameAs";
- public static String reference="http://dbpedia.org/property/reference";*/
-
- int tempyago=0;
-
/**
* reads all the options and makes the sparql query accordingly
* @param subject
@@ -57,7 +38,7 @@
String Filter="";
- if(!sf.useLiterals)Filter+="!isLiteral(?object))";
+ if(!sf.useLiterals)Filter+="!isLiteral(?object)";
for (String p : sf.getPredFilter()) {
Filter+="\n" + filterPredicate(p);
}
@@ -65,43 +46,19 @@
Filter+="\n" + filterObject(o);
}
-
String ret=
"SELECT * WHERE { \n" +
"<"+
subject+
-
- "> ?predicate ?object.\n" +
- "FILTER( \n" +
- "(" +Filter+").}";
+ "> ?predicate ?object.\n";
+ if (!(Filter.length()==0))
+ ret+="FILTER( \n" +
+ "(" +Filter+")).";
+ ret+="}";
//System.out.println(ret);
return ret;
}
-
- /*public String makeQueryDefault(String subject){
- String ret=
- "SELECT * WHERE { \n" +
- "<"+
- subject+
-
- "> ?predicate ?object.\n" +
- "FILTER( \n" +
- "(!isLiteral(?object))" +
- "\n" + filterPredicate(skos)+
- //"\n" + filterObject(skos)+
- "\n" + filterPredicate(foaf)+
- "\n" + filterObject(foaf)+
- "\n" + filterPredicate(foreign)+
- "\n" + filterPredicate(sameAs)+
- "\n" + filterPredicate(reference)+
- ")." +
- " }";
-
- //System.out.println(ret);
- return ret;
-}*/
-
/**
* add a new object filter
* (objects are filtered out of sparql result)
@@ -120,4 +77,11 @@
public String filterPredicate(String ns){
return "&&( !regex(str(?predicate), '"+ns+"') )";
}
+
+ public String makeLabelQuery(String label,int limit){
+ return "PREFIX rdfs:<http://www.w3.org/2000/01/rdf-schema#>\n"+
+ "SELECT DISTINCT ?subject\n"+
+ "WHERE { ?subject rdfs:label ?object.FILTER regex(?object,\""+label+"\"@en)}\n"+
+ "LIMIT "+limit;
+ }
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <ku...@us...> - 2007-12-01 20:50:21
|
Revision: 293
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=293&view=rev
Author: kurzum
Date: 2007-12-01 12:50:16 -0800 (Sat, 01 Dec 2007)
Log Message:
-----------
all files
Added Paths:
-----------
trunk/src/dl-learner/org/dllearner/kb/extraction/
trunk/src/dl-learner/org/dllearner/kb/extraction/Configuration.java
trunk/src/dl-learner/org/dllearner/kb/extraction/ExtractionAlgorithm.java
trunk/src/dl-learner/org/dllearner/kb/extraction/Manager.java
trunk/src/dl-learner/org/dllearner/kb/extraction/Manipulator.java
trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlEndpoint.java
trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlQueryType.java
trunk/src/dl-learner/org/dllearner/kb/extraction/Test.java
trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/
trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/ClassNode.java
trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/InstanceNode.java
trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Node.java
trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/PropertyNode.java
trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Tupel.java
trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/
trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/Cache.java
trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SimpleHTTPRequest.java
trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlHTTPRequest.java
trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlQueryMaker.java
trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/TypedSparqlQuery.java
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/Configuration.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/Configuration.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/Configuration.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,202 @@
+package org.dllearner.kb.extraction;
+
+import java.io.File;
+import java.net.URI;
+import java.net.URL;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Map;
+import java.util.Set;
+
+import org.semanticweb.owl.apibinding.OWLManager;
+import org.semanticweb.owl.model.OWLConstant;
+import org.semanticweb.owl.model.OWLDataPropertyExpression;
+import org.semanticweb.owl.model.OWLIndividual;
+import org.semanticweb.owl.model.OWLObjectPropertyExpression;
+import org.semanticweb.owl.model.OWLOntology;
+import org.semanticweb.owl.model.OWLOntologyManager;
+
+public class Configuration {
+ private SparqlEndpoint SparqlEndpoint;
+ private SparqlQueryType SparqlQueryType;
+ private Manipulator Manipulator;
+
+
+ private Configuration(){
+ }
+ public Configuration(SparqlEndpoint SparqlEndpoint,SparqlQueryType SparqlQueryType){
+ this.SparqlEndpoint=SparqlEndpoint;
+ this.SparqlQueryType=SparqlQueryType;
+ }
+
+
+ public static Configuration getConfiguration(URI uri){
+ //public static String getTellsString(URL file, URI kbURI){//throws OWLOntologyCreationException{
+ Configuration ret=new Configuration();
+ try{
+ String file="config/config.owl";
+
+
+ File f= new File(file);
+ String fileURL="file:///"+f.getAbsolutePath();
+ URL u=new URL(fileURL);
+ /* Load an ontology from a physical URI */
+ OWLOntologyManager manager = OWLManager.createOWLOntologyManager();
+ OWLOntology ontology = manager.loadOntologyFromPhysicalURI(u.toURI());
+ //System.out.println( ontology.containsIndividualReference(uri));
+ //OWLIndividualImpl ind=new OWLIndividualImpl();
+ //System.out.println(ontology.getReferencedIndividuals());
+ Set<OWLIndividual> s=ontology.getReferencedIndividuals();
+ //System.out.println(ontology.getReferencedClasses());
+ //Set<OWLIndividualAxiom> s= ontology.getIndividualAxioms();
+ Iterator<OWLIndividual> it=s.iterator();
+ while (it.hasNext()){
+ OWLIndividual tmp=(OWLIndividual)it.next();
+ //tmp.getURI()
+ if(tmp.getURI().equals(uri)){
+ OWLIndividual[] arr=getIndividualsForProperty("hasSparqlEndpoint",tmp.getObjectPropertyValues(ontology));
+ OWLIndividual sEndpoint=arr[0];
+ ret.SparqlEndpoint=makeEndpoint(sEndpoint, ontology);
+ arr=getIndividualsForProperty("hasTypedQuery",tmp.getObjectPropertyValues(ontology));
+ OWLIndividual typedQuery=arr[0];
+ ret.SparqlQueryType=makeSparqlQueryType(typedQuery, ontology);
+
+ }
+ //{hasSparqlEndpoint=[dbpediaEndpoint]}
+ }
+
+ ret.Manipulator=makeManipulator();
+ }catch (Exception e) {e.printStackTrace();}
+
+ return ret;
+ }
+
+
+
+
+
+ public static OWLIndividual[] getIndividualsForProperty(String propertyname,Map<OWLObjectPropertyExpression, Set<OWLIndividual>> m){
+ Set<OWLObjectPropertyExpression> s=m.keySet();
+
+ Iterator<OWLObjectPropertyExpression> it=s.iterator();
+ while (it.hasNext()){
+ OWLObjectPropertyExpression tmp=(OWLObjectPropertyExpression)it.next();
+ //System.out.println(tmp);
+ //System.out.println(propertyname);
+ if(tmp.toString().equals(propertyname))
+ {
+ Object[] arr=((Set<OWLIndividual>)m.get(tmp)).toArray() ;
+ OWLIndividual[] o=new OWLIndividual[arr.length];
+ for (int i = 0; i < o.length; i++) {
+ o[i]=(OWLIndividual)arr[i];
+ }
+
+ return o;}
+ }
+ return null;
+
+ }
+
+ public static String getFirstValueForDataProperty(String propertyname,Map<OWLDataPropertyExpression, Set<OWLConstant>> m){
+ return getValuesForDataProperty(propertyname, m)[0];
+ }
+
+ public static String[] getValuesForDataProperty(String propertyname,Map<OWLDataPropertyExpression, Set<OWLConstant>> m){
+ Set<OWLDataPropertyExpression> s=m.keySet();
+
+ Iterator<OWLDataPropertyExpression> it=s.iterator();
+ while (it.hasNext()){
+ OWLDataPropertyExpression tmp=(OWLDataPropertyExpression)it.next();
+ if(tmp.toString().equals(propertyname))
+ {
+ Object[] arr=((Set<OWLConstant>)m.get(tmp)).toArray() ;
+ String[] str=new String[arr.length];
+ for (int i = 0; i < str.length; i++) {
+ str[i]=((OWLConstant)arr[i]).getLiteral();
+ }
+ return str;}
+ }
+ return null;
+
+ }
+
+ public static SparqlEndpoint makeEndpoint(OWLIndividual sEndpoint,OWLOntology o){
+ String host=getFirstValueForDataProperty("hasHost",sEndpoint.getDataPropertyValues(o));
+ String port=getFirstValueForDataProperty("hasPort",sEndpoint.getDataPropertyValues(o));
+ String hasAfterGET=getFirstValueForDataProperty("hasAfterGET",sEndpoint.getDataPropertyValues(o));
+ String hasQueryParameter=getFirstValueForDataProperty("hasQueryParameter",sEndpoint.getDataPropertyValues(o));
+ OWLIndividual[] para=getIndividualsForProperty("hasGETParameter",sEndpoint.getObjectPropertyValues(o));
+ //System.out.println("test");
+ HashMap<String,String> parameters=new HashMap<String,String>();
+ if(para==null)return new SparqlEndpoint( host, port, hasAfterGET, hasQueryParameter, parameters);
+ for (OWLIndividual p : para) {
+ //System.out.println("test2");
+ String a1=getFirstValueForDataProperty("hasParameterName",p.getDataPropertyValues(o));
+ String a2=getFirstValueForDataProperty("hasParameterContent",p.getDataPropertyValues(o));
+ parameters.put(a1, a2);
+ }
+ //System.out.println("test2");
+ //System.out.println(host+port+ hasAfterGET+ hasQueryParameter+ parameters);
+ return new SparqlEndpoint( host, port, hasAfterGET, hasQueryParameter, parameters);
+
+
+
+ }
+
+ public static SparqlQueryType makeSparqlQueryType(OWLIndividual typedQuery,OWLOntology o){
+ String useLiterals=getFirstValueForDataProperty("usesLiterals",typedQuery.getDataPropertyValues(o));
+ String hasMode=getFirstValueForDataProperty("hasMode",typedQuery.getDataPropertyValues(o));
+ //String hasAfterGET=getValuesForDataProperty("hasAfterGET",sEndpoint.getDataPropertyValues(o));
+ //String hasQueryParameter=getValuesForDataProperty("hasQueryParameter",sEndpoint.getDataPropertyValues(o));
+ OWLIndividual[] objFilter=getIndividualsForProperty("hasObjectFilterSet",typedQuery.getObjectPropertyValues(o));
+ OWLIndividual[] predFilter=getIndividualsForProperty("hasPredicateFilterSet",typedQuery.getObjectPropertyValues(o));
+
+ Set<String> objectFilter=new HashSet<String>();
+ Set<String> predicateFilter=new HashSet<String>();
+
+ for (OWLIndividual of : objFilter) {
+ String[] tmp=getValuesForDataProperty("filtersURI",of.getDataPropertyValues(o));
+ for (String s : tmp){
+ objectFilter.add(s);
+
+ }
+ }
+
+ for (OWLIndividual pf : predFilter) {
+ String[] tmp=getValuesForDataProperty("filtersURI",pf.getDataPropertyValues(o));
+ for (String s : tmp){
+ predicateFilter.add(s);
+
+ }
+ }
+ //System.out.println(predicateFilter);
+ //System.out.println(hasMode+objectFilter+predicateFilter+useLiterals);
+ return new SparqlQueryType(hasMode,objectFilter,predicateFilter,useLiterals);
+
+
+
+ }
+
+ public static Manipulator makeManipulator() {
+ return new Manipulator();
+ }
+
+
+ public Manipulator getManipulator() {
+ return this.Manipulator;
+ }
+
+ public SparqlEndpoint getSparqlEndpoint() {
+ return SparqlEndpoint;
+ }
+
+
+
+ public SparqlQueryType getSparqlQueryType() {
+ return SparqlQueryType;
+ }
+
+
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/ExtractionAlgorithm.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/ExtractionAlgorithm.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/ExtractionAlgorithm.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,74 @@
+package org.dllearner.kb.extraction;
+
+import java.net.URI;
+import java.util.Vector;
+
+import org.dllearner.kb.extraction.datastructures.InstanceNode;
+import org.dllearner.kb.extraction.datastructures.Node;
+import org.dllearner.kb.extraction.sparql.TypedSparqlQuery;
+
+public class ExtractionAlgorithm {
+
+ private Configuration Configuration;
+ private Manipulator Manipulator;
+ private int recursiondepth=2;
+ private boolean getAllBackground=true;
+
+
+
+ public ExtractionAlgorithm (Configuration Configuration){
+ this.Configuration=Configuration;
+ this.Manipulator=Configuration.getManipulator();
+
+ }
+
+ public Node getFirstNode(URI u){
+ return new InstanceNode(u);
+ }
+ public Vector<Node> expandAll(URI[] u,TypedSparqlQuery tsp){
+ Vector<Node> v=new Vector<Node>();
+ for(URI one:u){
+ v.add(expandNode(one, tsp));
+ }
+ return v;
+ }
+
+
+ public Node expandNode(URI u, TypedSparqlQuery tsp){
+ Node n=getFirstNode(u);
+ Vector<Node> v=new Vector<Node>();
+ v.add(n);
+ System.out.println("StartVector: "+v);
+ // n.expand(tsp, this.Manipulator);
+ //Vector<Node> second=
+ for(int x=1;x<=this.recursiondepth;x++){
+
+ Vector<Node>tmp=new Vector<Node>();
+ while (v.size()>0) {
+ Node tmpNode=v.remove(0);
+ System.out.println("Expanding "+tmpNode);
+ Vector<Node> tmpVec=tmpNode.expand(tsp, this.Manipulator);
+
+ tmp.addAll(tmpVec);
+ }
+ v=tmp;
+ System.out.println("Rec: "+x+" with "+v);
+ }
+ if(this.getAllBackground){
+ Vector<Node> classes=new Vector<Node>();
+ for(Node one:v){
+ if(one.isClass()) {classes.add(one);}
+ }
+ while(classes.size()>0){
+ System.out.println(classes.size());
+ classes.addAll(classes.remove(0).expand(tsp, this.Manipulator));
+ }
+
+ }
+ return n;
+
+ }
+
+
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/Manager.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/Manager.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/Manager.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,67 @@
+package org.dllearner.kb.extraction;
+
+import java.net.URI;
+import java.util.HashSet;
+import java.util.Set;
+
+import org.dllearner.kb.extraction.datastructures.Node;
+import org.dllearner.kb.extraction.sparql.TypedSparqlQuery;
+
+public class Manager {
+
+ private Configuration Configuration;
+ private TypedSparqlQuery TypedSparqlQuery;
+ private ExtractionAlgorithm ExtractionAlgorithm;
+
+
+ public void usePredefinedConfiguration(URI uri){
+
+ this.Configuration=org.dllearner.kb.extraction.Configuration.getConfiguration(uri);
+ this.TypedSparqlQuery=new TypedSparqlQuery(Configuration);
+ this.ExtractionAlgorithm=new ExtractionAlgorithm(Configuration);
+ }
+
+ public void useConfiguration(SparqlQueryType SparqlQueryType, SparqlEndpoint SparqlEndpoint){
+
+ this.Configuration=new Configuration(SparqlEndpoint,SparqlQueryType);
+ this.TypedSparqlQuery=new TypedSparqlQuery(Configuration);
+ this.ExtractionAlgorithm=new ExtractionAlgorithm(Configuration);
+ }
+
+ public String extract(URI uri){
+ //this.TypedSparqlQuery.query(uri);
+ //System.out.println(ExtractionAlgorithm.getFirstNode(uri));
+ System.out.println("Start extracting");
+ Node n=this.ExtractionAlgorithm.expandNode(uri, this.TypedSparqlQuery);
+ Set<String> s=n.toNTriple();
+ String nt="";
+ for(String str:s){
+ nt+=str+"\n";
+ }
+ return nt;
+ }
+
+ public String extract(Set<String> instances){
+ //this.TypedSparqlQuery.query(uri);
+ //System.out.println(ExtractionAlgorithm.getFirstNode(uri));
+ System.out.println("Start extracting");
+ Set<String> ret=new HashSet<String>();
+
+
+ for(String one:instances){
+ try{
+ Node n=this.ExtractionAlgorithm.expandNode(new URI(one),this.TypedSparqlQuery);
+ ret.addAll(n.toNTriple());
+ }catch (Exception e) {e.printStackTrace();}
+ }
+
+
+ String nt="";
+ for(String str:ret){
+ nt+=str+"\n";
+ }
+ return nt;
+ }
+
+
+}
\ No newline at end of file
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/Manipulator.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/Manipulator.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/Manipulator.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,55 @@
+package org.dllearner.kb.extraction;
+
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+
+import org.dllearner.kb.extraction.datastructures.Node;
+import org.dllearner.kb.extraction.datastructures.Tupel;
+
+public class Manipulator {
+ public String subclass="http://www.w3.org/2000/01/rdf-schema#subClassOf";
+ public String type="http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
+
+ String objectProperty="http://www.w3.org/2002/07/owl#ObjectProperty";
+ String classns="http://www.w3.org/2002/07/owl#Class";
+ String thing="http://www.w3.org/2002/07/owl#Thing";
+
+ Set<String> classproperties;
+
+
+ String[] defaultClasses={
+ "http://dbpedia.org/class/yago",
+ "http://dbpedia.org/resource/Category:",
+ "http://dbpedia.org/resource/Template:",
+ "http://www.w3.org/2004/02/skos/core",
+ "http://dbpedia.org/class/"}; //TODO FEHLER hier fehlt yago
+
+ public Manipulator(){
+ Set<String> classproperties=new HashSet<String>();
+ classproperties.add(subclass);
+
+ }
+
+ public Set<Tupel> check(Set<Tupel> s,Node node){
+ Set<Tupel> toRemove=new HashSet<Tupel>();
+ Iterator<Tupel> it=s.iterator();
+ while(it.hasNext()){
+ Tupel t=(Tupel)it.next();
+ //all classes with owl:type class
+ if(t.a.equals(this.type) && t.b.equals(this.classns)&& node.isClass() )
+ {toRemove.add(t);};
+ // all with type class
+ if( t.b.equals(this.classns) && node.isClass() )
+ {toRemove.add(t);};
+ // all instances with owl:type thing
+ if(t.a.equals(this.type) && t.b.equals(this.thing)&& node.isInstance() )
+ {toRemove.add(t);};
+
+ }
+ s.removeAll(toRemove);
+
+ return s;
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlEndpoint.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlEndpoint.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,69 @@
+package org.dllearner.kb.extraction;
+
+import java.util.HashMap;
+
+public class SparqlEndpoint {
+
+ String host;
+ int port;
+ String hasAfterGET;
+ String hasQueryParameter;
+ String hasURL;
+ public HashMap<String,String> parameters=new HashMap<String,String>();
+ public SparqlEndpoint(String host, String port, String hasAfterGET, String hasQueryParameter, HashMap<String, String> parameters) {
+ super();
+ this.host = host;
+ this.port = Integer.parseInt(port);
+ this.hasAfterGET = hasAfterGET;
+ this.hasQueryParameter = hasQueryParameter;
+ this.parameters = parameters;
+ }
+
+ public SparqlEndpoint(String host, int port, String hasURL, HashMap<String, String> parameters) {
+ super();
+ this.port=port;
+ this.host=host;
+ this.hasURL = hasURL;
+ this.hasQueryParameter = "query";
+ this.parameters = parameters;
+ }
+ public String getHasAfterGET() {
+ return hasAfterGET;
+ }
+ public void setHasAfterGET(String hasAfterGET) {
+ this.hasAfterGET = hasAfterGET;
+ }
+ public String getHasQueryParameter() {
+ return hasQueryParameter;
+ }
+ public void setHasQueryParameter(String hasQueryParameter) {
+ this.hasQueryParameter = hasQueryParameter;
+ }
+ public String getHost() {
+ return host;
+ }
+ public void setHost(String host) {
+ this.host = host;
+ }
+ public HashMap<String, String> getParameters() {
+ return parameters;
+ }
+ public void setParameters(HashMap<String, String> parameters) {
+ this.parameters = parameters;
+ }
+ public int getPort() {
+ return port;
+ }
+ public void setPort(int port) {
+ this.port = port;
+ }
+
+
+
+
+ /*sparql?default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
+ //"SELECT%20%2A%20WHERE%20%7B%20%3Chttp%3A%2F%2Fdbpedia.org%2Fresource%2FAristotle%3E%20%3Fa%20%3Fb%20%7D%20" +
+ URLEncoder.encode(query, "UTF-8")+
+ //query+// URLencode
+ "&format=application%2Fsparql-results%2Bxml*/
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlQueryType.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlQueryType.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/SparqlQueryType.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,68 @@
+package org.dllearner.kb.extraction;
+
+import java.util.Set;
+
+public class SparqlQueryType {
+
+ private String mode="forbid";
+ private String[] objectfilterlist={
+ "http://dbpedia.org/resource/Category:Articles_",
+ "http://dbpedia.org/resource/Category:Wikipedia_",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/resource/Category",
+ "http://dbpedia.org/resource/Template",
+ "http://upload.wikimedia.org/wikipedia/commons"};
+ private String[] predicatefilterlist={
+ "http://www.w3.org/2004/02/skos/core",
+ "http://xmlns.com/foaf/0.1/",
+ "http://dbpedia.org/property/wikipage-",
+ "http://www.w3.org/2002/07/owl#sameAs",
+ "http://dbpedia.org/property/reference" };
+ private boolean literals=false;
+
+ public SparqlQueryType(String mode, String[] obectfilterlist, String[] predicatefilterlist, boolean literals) {
+ super();
+ this.mode = mode;
+ this.objectfilterlist = obectfilterlist;
+ this.predicatefilterlist = predicatefilterlist;
+ this.literals = literals;
+ }
+
+ public SparqlQueryType(String mode, Set<String> objectfilterlist, Set<String> predicatefilterlist, String literals) {
+ super();
+ this.mode = mode;
+ this.literals = (literals.equals("true"))?true:false;
+
+ Object[] arr=objectfilterlist.toArray();
+ Object[] arr2=predicatefilterlist.toArray();
+ this.objectfilterlist = new String[arr.length];
+ this.predicatefilterlist = new String[arr2.length];
+ for (int i = 0; i < arr.length; i++) {
+ this.objectfilterlist[i]=(String)arr[i];
+ }
+ for (int i = 0; i < arr2.length; i++) {
+ this.predicatefilterlist[i]=(String)arr2[i];
+ }
+
+
+ }
+
+ public boolean isLiterals() {
+ return literals;
+ }
+
+ public String getMode() {
+ return mode;
+ }
+
+ public String[] getObjectfilterlist() {
+ return objectfilterlist;
+ }
+
+ public String[] getPredicatefilterlist() {
+ return predicatefilterlist;
+ }
+
+
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/Test.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/Test.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/Test.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,31 @@
+package org.dllearner.kb.extraction;
+
+import java.io.File;
+import java.io.FileWriter;
+import java.net.URI;
+
+public class Test {
+
+
+ public static void main(String[] args) {
+ System.out.println("Start");
+ String test2="http://www.extraction.org/config#dbpediatest";
+ String test="http://www.extraction.org/config#localjoseki";
+ try{
+ URI u=new URI(test);
+ Manager m=new Manager();
+ m.usePredefinedConfiguration(u);
+
+
+ URI u2=new URI("http://dbpedia.org/resource/Angela_Merkel");
+
+ String filename=System.currentTimeMillis()+".nt";
+ FileWriter fw=new FileWriter(new File(filename),true);
+ fw.write(m.extract(u2));
+ fw.flush();
+ fw.close();
+
+ }catch (Exception e) {e.printStackTrace();}
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/ClassNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/ClassNode.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/ClassNode.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,60 @@
+package org.dllearner.kb.extraction.datastructures;
+
+import java.net.URI;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+import java.util.Vector;
+
+import org.dllearner.kb.extraction.Manipulator;
+import org.dllearner.kb.extraction.sparql.TypedSparqlQuery;
+
+public class ClassNode extends Node{
+ Set<PropertyNode> properties=new HashSet<PropertyNode>();
+
+ public ClassNode(URI u) {
+ super(u);
+ this.type="class";
+ }
+
+ @Override
+ public Vector<Node> expand(TypedSparqlQuery tsq,Manipulator m){
+ Set<Tupel> s =tsq.query(this.URI);
+ s=m.check(s, this);
+ Vector<Node> Nodes=new Vector<Node>();
+ // Manipulation
+
+ Iterator<Tupel> it=s.iterator();
+ while(it.hasNext()){
+ Tupel t=(Tupel)it.next();
+ try{
+ if(t.a.equals(m.type) || t.a.equals(m.subclass)){
+ ClassNode tmp=new ClassNode(new URI(t.b));
+ properties.add(new PropertyNode(new URI(m.subclass),this,tmp));
+ Nodes.add(tmp);
+ }
+ }catch (Exception e) {System.out.println(t);e.printStackTrace();}
+
+ }
+ return Nodes;
+ }
+
+ @Override
+ public boolean isClass(){
+ return true;
+ }
+
+ @Override
+ public Set<String> toNTriple(){
+ Set<String> s= new HashSet<String>();
+ s.add("<"+this.URI+"><"+"http://www.w3.org/1999/02/22-rdf-syntax-ns#type"+"><"+"http://www.w3.org/2002/07/owl#Class"+">.");
+
+ for (PropertyNode one:properties){
+ s.add("<"+this.URI+"><"+one.getURI()+"><"+one.getB().getURI()+">.");
+ s.addAll(one.getB().toNTriple());
+ }
+
+ return s;
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/InstanceNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/InstanceNode.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/InstanceNode.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,77 @@
+package org.dllearner.kb.extraction.datastructures;
+
+import java.net.URI;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+import java.util.Vector;
+
+import org.dllearner.kb.extraction.Manipulator;
+import org.dllearner.kb.extraction.sparql.TypedSparqlQuery;
+
+public class InstanceNode extends Node{
+
+ Set<ClassNode> classes=new HashSet<ClassNode>();
+ Set<Tupel> datatypes=new HashSet<Tupel>();
+ Set<PropertyNode> properties=new HashSet<PropertyNode>();
+
+ public InstanceNode(URI u) {
+ super(u);
+ this.type="instance";
+
+ }
+
+ @Override
+ public Vector<Node> expand(TypedSparqlQuery tsq,Manipulator m){
+
+ Set<Tupel> s =tsq.query(this.URI);
+ // Manipulation
+ m.check(s,this);
+ Vector<Node> Nodes=new Vector<Node>();
+
+ Iterator<Tupel> it=s.iterator();
+ while(it.hasNext()){
+ Tupel t=(Tupel)it.next();
+
+ try{
+ if(t.a.equals(m.type)){
+ ClassNode tmp=new ClassNode(new URI(t.b));
+ classes.add(tmp);
+ Nodes.add(tmp);
+ }else {
+ InstanceNode tmp=new InstanceNode(new URI(t.b));
+ properties.add(new PropertyNode(new URI(t.a),this,tmp));
+ Nodes.add(tmp);
+
+ }
+ }catch (Exception e) {
+ System.out.println("Problem with: "+t);
+ e.printStackTrace();}
+
+ }
+ this.expanded=true;
+ return Nodes;
+ }
+ @Override
+ public boolean isInstance(){
+ return true;
+ }
+
+ @Override
+ public Set<String> toNTriple(){
+ Set<String> s= new HashSet<String>();
+ s.add("<"+this.URI+"><"+"http://www.w3.org/1999/02/22-rdf-syntax-ns#type"+"><"+"http://www.w3.org/2002/07/owl#Thing"+">.");
+ for (ClassNode one:classes){
+ s.add("<"+this.URI+"><"+"http://www.w3.org/1999/02/22-rdf-syntax-ns#type"+"><"+one.getURI()+">.");
+ s.addAll(one.toNTriple());
+ }
+ for (PropertyNode one:properties){
+ s.add("<"+this.URI+"><"+one.getURI()+"><"+one.getB().getURI()+">.");
+ s.addAll(one.toNTriple());
+ s.addAll(one.getB().toNTriple());
+ }
+
+ return s;
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Node.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Node.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Node.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,54 @@
+package org.dllearner.kb.extraction.datastructures;
+
+import java.net.URI;
+import java.util.Set;
+import java.util.Vector;
+
+import org.dllearner.kb.extraction.Manipulator;
+import org.dllearner.kb.extraction.sparql.TypedSparqlQuery;
+
+public abstract class Node {
+ URI URI;
+ protected String type;
+ protected boolean expanded=false;
+
+ //Hashtable<String,Node> classes=new Hashtable<String,Node>();
+ //Hashtable<String,Node> instances=new Hashtable<String,Node>();;
+ //Hashtable<String,Node> datatype=new Hashtable<String,Node>();;
+
+ public Node (URI u){
+ this.URI=u;
+
+ }
+
+ /*public void checkConsistency(){
+ if (type.equals("class") && ( instances.size()>0 || datatype.size()>0)){
+ System.out.println("Warning, inconsistent:"+this.toString());
+ }
+
+ }*/
+
+ public abstract Vector<Node> expand(TypedSparqlQuery tsq,Manipulator m);
+ public abstract Set<String> toNTriple();
+
+ @Override
+ public String toString(){
+ return "Node: "+URI+":"+type;
+
+ }
+
+
+ public boolean isClass(){
+ return false;
+ }
+ public boolean isInstance(){
+ return false;
+ }
+ public boolean isProperty(){
+ return false;
+ }
+ public URI getURI() {
+ return URI;
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/PropertyNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/PropertyNode.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/PropertyNode.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,74 @@
+package org.dllearner.kb.extraction.datastructures;
+
+import java.net.URI;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+import java.util.Vector;
+
+import org.dllearner.kb.extraction.Manipulator;
+import org.dllearner.kb.extraction.sparql.TypedSparqlQuery;
+
+public class PropertyNode extends Node{
+
+ private Node a;
+ private Node b;
+ private Set<String> SpecialTypes;
+
+
+ public PropertyNode(URI u) {
+ super(u);
+ this.type="property";
+
+ }
+
+ public PropertyNode(URI u,Node a, Node b) {
+ super(u);
+ this.type="property";
+ this.a=a;
+ this.b=b;
+ this.SpecialTypes=new HashSet<String>();
+ }
+
+ @Override
+ public Vector<Node> expand(TypedSparqlQuery tsq,Manipulator m){
+ Set<Tupel> s =tsq.query(this.URI);
+ Vector<Node> Nodes=new Vector<Node>();
+ // Manipulation
+
+ Iterator<Tupel> it=s.iterator();
+ while(it.hasNext()){
+ Tupel t=(Tupel)it.next();
+ try{
+ if(t.a.equals(m.type) ){
+ SpecialTypes.add(t.b);
+ }
+ }catch (Exception e) {System.out.println(t);e.printStackTrace();}
+
+ }
+ return Nodes;
+ }
+ @Override
+ public boolean isProperty(){
+ return true;
+ }
+ public Node getB(){
+ return this.b;
+ }
+
+ @Override
+ public Set<String> toNTriple(){
+ Set<String> s= new HashSet<String>();
+ s.add("<"+this.URI+"><"+"http://www.w3.org/1999/02/22-rdf-syntax-ns#type"+"><"+"http://www.w3.org/2002/07/owl#ObjectProperty"+">.");
+ for (String one:SpecialTypes){
+ s.add("<"+this.URI+"><"+"http://www.w3.org/1999/02/22-rdf-syntax-ns#type"+"><"+one+">.");
+
+ }
+
+ return s;
+ }
+
+
+
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Tupel.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Tupel.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/datastructures/Tupel.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,25 @@
+package org.dllearner.kb.extraction.datastructures;
+
+public class Tupel {
+
+ public String a;
+ public String b;
+
+ public Tupel(String a, String b) {
+ this.a = a;
+ this.b = b;
+ }
+
+ @Override
+ public String toString(){
+ return "<"+a+"|"+b+">";
+ }
+
+ public boolean equals(Tupel t){
+ if(this.a.equals(t.a)&& this.b.equals(t.b))return true;
+ else return false;
+ }
+
+
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/Cache.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/Cache.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/Cache.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,115 @@
+package org.dllearner.kb.extraction.sparql;
+
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.FileOutputStream;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.io.Serializable;
+import java.net.URLEncoder;
+
+public class Cache implements Serializable{
+ // Object can be the cache itself
+ // or a cache object(one entry)
+
+ final static long serialVersionUID=104;
+ transient String basedir="";
+ transient String fileending=".cache";
+ long timestamp;
+ String content="";
+ long daysoffreshness=15;
+ long multiplier=24*60*60*1000;//h m s ms
+ String sparqlquery="";
+
+ //constructor for the cache itself
+ public Cache(String path){
+ this.basedir=path+File.separator;
+ if(!new File(path).exists())
+ {System.out.println(new File(path).mkdir());;}
+
+ }
+
+// constructor for single cache object(one entry)
+ public Cache(String c, String sparql){
+ this.content=c;
+ this.sparqlquery=sparql;
+ this.timestamp=System.currentTimeMillis();
+ }
+
+
+ public String get(String key, String sparql){
+ //System.out.println("get From "+key);
+ String ret=null;
+ try{
+ Cache c =readFromFile(makeFilename(key)) ;
+ if(c==null)return null;
+ //System.out.println(" file found");
+ if(!c.checkFreshness())return null;
+ //System.out.println("fresh");
+ if(!c.validate(sparql))return null;
+ //System.out.println("valid");
+ ret=c.content;
+ }catch (Exception e) {e.printStackTrace();}
+ return ret;
+ };
+ public void put(String key, String content, String sparql){
+ //System.out.println("put into "+key);
+ Cache c=new Cache(content,sparql);
+ putIntoFile(makeFilename(key), c);
+ }
+
+
+ String makeFilename(String key){
+ String ret="";
+ try{
+ ret=basedir+URLEncoder.encode(key, "UTF-8")+fileending;
+ }catch (Exception e) {e.printStackTrace();}
+ return ret;
+ }
+ boolean checkFreshness(){
+ if((System.currentTimeMillis()-this.timestamp)<=(daysoffreshness*multiplier))
+ //fresh
+ return true;
+ else return false;
+ }
+ boolean validate(String sparql){
+ if(this.sparqlquery.equals(sparql))
+ //valid
+ return true;
+ else return false;
+ }
+
+ public void checkFile(String Filename){
+ if(!new File(Filename).exists()){
+ try{
+ new File(Filename).createNewFile();
+ }catch (Exception e) {e.printStackTrace();}
+
+ }
+
+ }
+
+ public void putIntoFile(String Filename,Cache content){
+ try{
+ //FileWriter fw=new FileWriter(new File(Filename),true);
+ FileOutputStream fos = new FileOutputStream( Filename , false );
+ ObjectOutputStream o = new ObjectOutputStream( fos );
+ o.writeObject( content );
+ fos.flush();
+ fos.close();
+ }catch (Exception e) {System.out.println("Not in cache creating: "+Filename);}
+ }
+
+ public Cache readFromFile(String Filename){
+ Cache content=null;
+ try{
+ FileInputStream fos = new FileInputStream( Filename );
+ ObjectInputStream o = new ObjectInputStream( fos );
+ content=(Cache)o.readObject();
+ //FileReader fr=new FileReader(new File(Filename,"r"));
+ //BufferedReader br=new BufferedReader(fr);
+ }catch (Exception e) {}
+ return content;
+
+ }
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SimpleHTTPRequest.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SimpleHTTPRequest.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SimpleHTTPRequest.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,129 @@
+package org.dllearner.kb.extraction.sparql;
+
+import java.io.BufferedInputStream;
+import java.io.BufferedReader;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.OutputStream;
+import java.io.OutputStreamWriter;
+import java.net.HttpURLConnection;
+import java.net.InetAddress;
+import java.net.Socket;
+import java.net.URL;
+import java.net.URLEncoder;
+
+
+
+public class SimpleHTTPRequest {
+ static final char value[]={13,10};
+ static final String cut=new String(value);
+ private InetAddress ia;
+ private int port;
+
+
+
+ public SimpleHTTPRequest(InetAddress ia, int port) {
+ super();
+ this.ia = ia;
+ this.port = port;
+
+ }
+
+ public String sendAndReceive( String content){
+ String retval="";
+ //
+
+ byte resp[]=null;
+
+ try{
+ Socket SparqlServer=new Socket(this.ia,this.port);
+ //String request=makeHeader(content);
+ // send request
+ (SparqlServer.getOutputStream()).write(content.getBytes());
+
+ //get Response
+ resp=readBuffer(new BufferedInputStream(SparqlServer.getInputStream()));
+ retval=new String(resp);
+ retval=subtractResponseHeader(retval);
+ //retval="||"+retval;
+
+ SparqlServer.close();
+
+
+
+ }
+ catch(Exception e){e.printStackTrace();}
+ //System.out.println("got it");
+ return retval;
+
+ }//down
+
+ public static byte[] readBuffer(InputStream IS)
+ throws IOException{
+ byte buffer[] = new byte[0xffff];
+ int nbytes=0;
+ byte resp[]=new byte[0];
+ while ((nbytes=IS.read(buffer))!=-1) {
+ byte tmp[]=new byte[resp.length+nbytes];
+ int i=0;
+ for (;i<resp.length;i++){
+ tmp[i]=resp[i];
+ }
+ for(int a=0;a<nbytes;a++,i++){
+ tmp[i]=buffer[a];
+ }
+ resp=tmp;
+ }
+ return resp;
+ }
+
+ public String subtractResponseHeader(String in){
+ //System.out.println(in.indexOf(cut+""+cut));
+ return in.substring(in.indexOf(cut+""+cut)+4);
+
+
+ }
+
+
+ private String sendAndReceive2(String sparql,URL url) throws IOException{
+ StringBuilder answer = new StringBuilder();
+
+ // String an Sparql-Endpoint schicken
+ HttpURLConnection connection;
+
+ connection = (HttpURLConnection) url.openConnection();
+ connection.setDoOutput(true);
+
+ connection.addRequestProperty("Host", "dbpedia.openlinksw.com");
+ connection.addRequestProperty("Connection","close");
+ connection.addRequestProperty("Accept","text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5");
+ connection.addRequestProperty("Accept-Language","de-de,de;q=0.8,en-us;q=0.5,en;q=0.3");
+ connection.addRequestProperty("Accept-Charset","utf-8;q=1.0");
+ connection.addRequestProperty("User-Agent","Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
+
+ OutputStream os = connection.getOutputStream();
+ OutputStreamWriter osw = new OutputStreamWriter(os);
+ osw.write("default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
+ URLEncoder.encode(sparql, "UTF-8")+
+ "&format=application%2Fsparql-results%2Bxml");
+ osw.close();
+
+ // receive answer
+ InputStream is = connection.getInputStream();
+ InputStreamReader isr = new InputStreamReader(is,"UTF-8");
+ BufferedReader br = new BufferedReader(isr);
+
+ String line;
+ do {
+ line = br.readLine();
+ if(line!=null)
+ answer.append(line);
+ } while (line != null);
+
+ br.close();
+
+ return answer.toString();
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlHTTPRequest.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlHTTPRequest.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlHTTPRequest.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,82 @@
+package org.dllearner.kb.extraction.sparql;
+
+import java.net.InetAddress;
+import java.net.URLEncoder;
+import java.util.Iterator;
+import java.util.Set;
+
+import org.dllearner.kb.extraction.SparqlEndpoint;
+
+
+
+public class SparqlHTTPRequest {
+ static final char value[]={13,10};
+ static final String cut=new String(value);
+
+ private SparqlEndpoint SparqlEndpoint;
+ private SimpleHTTPRequest SimpleHTTPRequest;
+
+
+ public SparqlHTTPRequest(SparqlEndpoint SparqlEndpoint){
+ this.SparqlEndpoint=SparqlEndpoint;
+ InetAddress ia=null;
+ try{
+ ia=InetAddress.getByName(SparqlEndpoint.getHost());
+ }catch (Exception e) {e.printStackTrace();}
+ this.SimpleHTTPRequest=new SimpleHTTPRequest(ia,SparqlEndpoint.getPort());
+
+ }
+
+
+ public String sendAndReceiveSPARQL( String sparql){
+
+ //System.out.println(sparql);
+ String content=makeContent(sparql );
+ //System.out.println(content);
+ String ret= this.SimpleHTTPRequest.sendAndReceive(content);
+ //System.out.println(ret);
+
+ //this.sendAndReceiveSPARQL("SELECT * WHERE {?a ?b ?c} LIMIT 10");
+
+ return ret;
+
+
+ }//down
+
+
+
+
+ public String makeContent(String query){
+
+
+ String RequestHeader="";
+ try{
+
+ RequestHeader="GET ";
+ RequestHeader+=SparqlEndpoint.getHasAfterGET()+"?";
+ // parameters
+ Set<String> s =SparqlEndpoint.getParameters().keySet();
+ Iterator<String> it=s.iterator();
+ while (it.hasNext()) {
+ String element = (String) it.next();
+ RequestHeader+=""+URLEncoder.encode(element, "UTF-8")+"="+
+ URLEncoder.encode(SparqlEndpoint.getParameters().get(element), "UTF-8")+"&";
+ }
+ RequestHeader+=""+SparqlEndpoint.getHasQueryParameter()+"="+URLEncoder.encode(query, "UTF-8");
+ RequestHeader+=" HTTP/1.1"+cut;
+ RequestHeader+="Host: "+SparqlEndpoint.getHost()+cut;
+
+ RequestHeader+=
+ "Connection: close"+cut+
+ //"Accept-Encoding: gzip"+cut+
+ "Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5"+cut+
+ "Accept-Language: de-de,de;q=0.8,en-us;q=0.5,en;q=0.3"+cut+
+ "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7"+cut+
+ "User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24"+cut+
+ cut;
+ }catch (Exception e) {e.printStackTrace();}
+ return RequestHeader;
+
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlQueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlQueryMaker.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/SparqlQueryMaker.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,46 @@
+package org.dllearner.kb.extraction.sparql;
+
+import org.dllearner.kb.extraction.SparqlQueryType;
+
+
+
+public class SparqlQueryMaker {
+
+ private SparqlQueryType SparqlQueryType;
+
+ public SparqlQueryMaker(SparqlQueryType SparqlQueryType){
+ this.SparqlQueryType=SparqlQueryType;
+ }
+
+ public String makeQueryUsingFilters(String subject){
+ String lineend="\n";
+
+ String Filter="";
+ if(!this.SparqlQueryType.isLiterals())Filter+="!isLiteral(?object))";
+ for (String p : this.SparqlQueryType.getPredicatefilterlist()) {
+ Filter+=lineend + filterPredicate(p);
+ }
+ for (String o : this.SparqlQueryType.getObjectfilterlist()) {
+ Filter+=lineend + filterObject(o);
+ }
+
+
+ String ret=
+ "SELECT * WHERE { "+lineend +
+ "<"+
+ subject+
+ "> ?predicate ?object. "+ lineend+
+ "FILTER( "+lineend +
+ "(" +Filter+").}";
+ //System.out.println(ret);
+ return ret;
+ }
+
+
+ public String filterObject(String ns){
+ return "&&( !regex(str(?object), '"+ns+"') )";
+ }
+ public String filterPredicate(String ns){
+ return "&&( !regex(str(?predicate), '"+ns+"') )";
+ }
+}
Added: trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/TypedSparqlQuery.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/TypedSparqlQuery.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/extraction/sparql/TypedSparqlQuery.java 2007-12-01 20:50:16 UTC (rev 293)
@@ -0,0 +1,83 @@
+package org.dllearner.kb.extraction.sparql;
+
+import java.net.URI;
+import java.util.HashSet;
+import java.util.Set;
+
+import org.dllearner.kb.extraction.Configuration;
+import org.dllearner.kb.extraction.datastructures.Tupel;
+
+
+public class TypedSparqlQuery {
+ private Configuration Configuration;
+ private SparqlHTTPRequest SparqlHTTPRequest;
+ private SparqlQueryMaker SparqlQueryMaker;
+ Cache Cache;
+
+ public TypedSparqlQuery(Configuration Configuration){
+ this.Configuration=Configuration;
+ this.SparqlHTTPRequest=new SparqlHTTPRequest(Configuration.getSparqlEndpoint());
+ this.SparqlQueryMaker=new SparqlQueryMaker(Configuration.getSparqlQueryType());
+ this.Cache=new Cache("cache");
+ }
+
+ public Set<Tupel> query(URI u){
+
+ //getQuery
+ String sparql=SparqlQueryMaker.makeQueryUsingFilters(u.toString());
+
+ // check cache
+ String FromCache=this.Cache.get(u.toString(), sparql);
+ FromCache=null;
+ String xml;
+ // if not in cache get it from EndPoint
+ if(FromCache==null){
+ xml=this.SparqlHTTPRequest.sendAndReceiveSPARQL(sparql);
+ //this.Cache.put(u.toString(), xml, sparql);
+ System.out.print("\n");
+ }
+ else{
+ xml=FromCache;
+ System.out.println("FROM CACHE");
+ }
+
+ //System.out.println(xml);
+ //process XML
+ Set<Tupel> s=this.processResult(xml);
+ try {
+ System.out.println("retrieved "+s.size()+" tupels");
+ } catch (Exception e) { }
+ return s;
+ }
+
+ public Set<Tupel> processResult(String xml){
+
+ Set<Tupel> ret=new HashSet<Tupel>();
+ //TODO if result is empty, catch exceptions
+ String one="<binding name=\"predicate\">";
+ String two="<binding name=\"object\">";
+ String uridel="<uri>";
+ String end="</uri>";
+ String predtmp="";
+ String objtmp="";
+
+ while(xml.indexOf(one)!=-1){
+ //get pred
+ xml=xml.substring(xml.indexOf(one)+one.length());
+ xml=xml.substring(xml.indexOf(uridel)+uridel.length());
+ predtmp=xml.substring(0,xml.indexOf(end));
+
+ //getobj
+ xml=xml.substring(xml.indexOf(two)+two.length());
+ xml=xml.substring(xml.indexOf(uridel)+uridel.length());
+ objtmp=xml.substring(0,xml.indexOf(end));
+ ret.add(new Tupel(predtmp,objtmp));
+ //System.out.println(new Tupel(predtmp,objtmp));
+ }
+
+
+ return ret;
+
+ }
+
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <ku...@us...> - 2007-12-02 13:30:02
|
Revision: 298
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=298&view=rev
Author: kurzum
Date: 2007-12-02 05:29:58 -0800 (Sun, 02 Dec 2007)
Log Message:
-----------
working, filtermode gives a predefined set wchich is stored in kb.sparql.Predefined. Manual configuration might not work yet.
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlHTTPRequest.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Test.java
trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
Added Paths:
-----------
trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedConfigurations.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SpecificSparqlEndpoint.java
Removed Paths:
-------------
trunk/src/dl-learner/org/dllearner/kb/sparql/SimpleHTTPRequest.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlEndpoint.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -27,6 +27,7 @@
import java.net.URI;
import java.net.URL;
import java.util.Collection;
+import java.util.HashMap;
import java.util.LinkedList;
import java.util.Set;
@@ -42,12 +43,11 @@
import org.dllearner.core.config.StringSetConfigOption;
import org.dllearner.core.dl.KB;
import org.dllearner.kb.sparql.Manager;
-import org.dllearner.kb.sparql.SparqlEndpoint;
import org.dllearner.kb.sparql.SparqlQueryType;
+import org.dllearner.kb.sparql.SpecificSparqlEndpoint;
import org.dllearner.parser.KBParser;
import org.dllearner.reasoning.DIGConverter;
import org.dllearner.reasoning.JenaOWLDIGConverter;
-import org.dllearner.utilities.Datastructures;
/**
* Represents a SPARQL Endpoint.
@@ -59,7 +59,8 @@
public class SparqlEndpointRestructured extends KnowledgeSource {
//ConfigOptions
- private URL url;
+ private URL url;
+ String host;
private Set<String> instances;
private URL dumpFile;
private int numberOfRecursions;
@@ -70,6 +71,7 @@
private String format;
private boolean dumpToFile;
private boolean useLits=false;
+ private boolean getAllBackground=false;
/**
* Holds the results of the calculateSubjects method
@@ -120,8 +122,9 @@
* @return
*/
public static Collection<ConfigOption<?>> createConfigOptions() {
- Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
+ Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
options.add(new StringConfigOption("url", "URL of SPARQL Endpoint"));
+ options.add(new StringConfigOption("host", "host of SPARQL Endpoint"));
options.add(new StringSetConfigOption("instances","relevant instances e.g. positive and negative examples in a learning problem"));
options.add(new IntegerConfigOption("numberOfRecursions","number of Recursions, the Sparql-Endpoint is asked"));
options.add(new IntegerConfigOption("filterMode","the mode of the SPARQL Filter"));
@@ -131,6 +134,7 @@
options.add(new StringConfigOption("format", "N-TRIPLES or KB format"));
options.add(new BooleanConfigOption("dumpToFile", "wether Ontology from DBPedia is written to a file or not"));
options.add(new BooleanConfigOption("useLits","use Literals in SPARQL query"));
+ options.add(new BooleanConfigOption("getAllBackground","get"));
return options;
}
@@ -148,6 +152,8 @@
} catch (MalformedURLException e) {
throw new InvalidConfigOptionValueException(entry.getOption(), entry.getValue(),"malformed URL " + s);
}
+ } else if(option.equals("host")) {
+ host = (String) entry.getValue();
} else if(option.equals("instances")) {
instances = (Set<String>) entry.getValue();
} else if(option.equals("numberOfRecursions")){
@@ -166,6 +172,8 @@
dumpToFile=(Boolean)entry.getValue();
} else if(option.equals("useLits")){
useLits=(Boolean)entry.getValue();
+ } else if(option.equals("getAllBackground")){
+ getAllBackground=(Boolean)entry.getValue();
}
}
@@ -182,15 +190,16 @@
//Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format,url,useLits);
Manager m=new Manager();
if(filterMode==0){
- try{
- m.usePredefinedConfiguration(new URI("http://www.extraction.org/config#dbpediatest"));
- }catch (Exception e) {e.printStackTrace();}
+
+ m.usePredefinedConfiguration(filterMode);
+
}
else{
- //SparqlQueryType sqt=new SparqlQueryType("forbid", objList,predList,useLits+"");
- //SparqlEndpoint se=new SparqlEndpoint();
- //m.useConfiguration(SparqlQueryType, SparqlEndpoint)
+ SparqlQueryType sqt=new SparqlQueryType("forbid", objList,predList,useLits+"");
+ SpecificSparqlEndpoint se=new SpecificSparqlEndpoint(url, host, new HashMap<String, String>());
+ m.useConfiguration(sqt, se,numberOfRecursions,getAllBackground);
}
+
try
{
String ont=m.extract(instances);
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -18,16 +18,26 @@
import org.semanticweb.owl.model.OWLOntologyManager;
public class Configuration {
- private SparqlEndpoint SparqlEndpoint;
+ private SpecificSparqlEndpoint SparqlEndpoint;
private SparqlQueryType SparqlQueryType;
private Manipulator Manipulator;
-
+ private int recursiondepth = 2;
+ private boolean getAllBackground = true;
+
private Configuration() {
}
- public Configuration(SparqlEndpoint SparqlEndpoint, SparqlQueryType SparqlQueryType) {
+
+ public Configuration(SpecificSparqlEndpoint SparqlEndpoint,
+ SparqlQueryType SparqlQueryType, int recursiondepth,
+ boolean getAllBackground) {
this.SparqlEndpoint = SparqlEndpoint;
this.SparqlQueryType = SparqlQueryType;
+ this.Manipulator=new Manipulator();
+ this.recursiondepth = recursiondepth;
+ this.getAllBackground = getAllBackground;
+
+
}
public static Configuration getConfiguration(URI uri) {
@@ -123,7 +133,7 @@
}
- public static SparqlEndpoint makeEndpoint(OWLIndividual sEndpoint, OWLOntology o) {
+ public static SpecificSparqlEndpoint makeEndpoint(OWLIndividual sEndpoint, OWLOntology o) {
String host = getFirstValueForDataProperty("hasHost", sEndpoint.getDataPropertyValues(o));
String port = getFirstValueForDataProperty("hasPort", sEndpoint.getDataPropertyValues(o));
String hasAfterGET = getFirstValueForDataProperty("hasAfterGET", sEndpoint
@@ -135,7 +145,7 @@
// System.out.println("test");
HashMap<String, String> parameters = new HashMap<String, String>();
if (para == null)
- return new SparqlEndpoint(host, port, hasAfterGET, hasQueryParameter, parameters);
+ return null;//new SpecificSparqlEndpoint(host, port, hasAfterGET, hasQueryParameter, parameters);
for (OWLIndividual p : para) {
// System.out.println("test2");
String a1 = getFirstValueForDataProperty("hasParameterName", p.getDataPropertyValues(o));
@@ -146,7 +156,7 @@
// System.out.println("test2");
// System.out.println(host+port+ hasAfterGET+ hasQueryParameter+
// parameters);
- return new SparqlEndpoint(host, port, hasAfterGET, hasQueryParameter, parameters);
+ return null;//new SpecificSparqlEndpoint(host, port, hasAfterGET, hasQueryParameter, parameters);
}
@@ -196,7 +206,7 @@
return this.Manipulator;
}
- public SparqlEndpoint getSparqlEndpoint() {
+ public SpecificSparqlEndpoint getSparqlEndpoint() {
return SparqlEndpoint;
}
@@ -204,4 +214,14 @@
return SparqlQueryType;
}
+
+ public boolean isGetAllBackground() {
+ return getAllBackground;
+ }
+
+
+ public int getRecursiondepth() {
+ return recursiondepth;
+ }
+
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -13,6 +13,8 @@
public ExtractionAlgorithm(Configuration Configuration) {
this.Configuration = Configuration;
this.Manipulator = Configuration.getManipulator();
+ this.recursiondepth=Configuration.getRecursiondepth();
+ this.getAllBackground=Configuration.isGetAllBackground();
}
@@ -41,6 +43,7 @@
while (v.size() > 0) {
Node tmpNode = v.remove(0);
System.out.println("Expanding " + tmpNode);
+ //System.out.println(this.Manipulator);
Vector<Node> tmpVec = tmpNode.expand(tsp, this.Manipulator);
tmp.addAll(tmpVec);
@@ -57,7 +60,9 @@
}
while (classes.size() > 0) {
System.out.println(classes.size());
- classes.addAll(classes.remove(0).expand(tsp, this.Manipulator));
+ Node next=classes.remove(0);
+ System.out.println(next);
+ classes.addAll(next.expand(tsp, this.Manipulator));
}
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -24,6 +24,7 @@
Set<Tupel> s = tsq.query(this.URI);
// Manipulation
m.check(s, this);
+ //System.out.println("fffffff"+m);
Vector<Node> Nodes = new Vector<Node>();
Iterator<Tupel> it = s.iterator();
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -10,16 +10,23 @@
private TypedSparqlQuery TypedSparqlQuery;
private ExtractionAlgorithm ExtractionAlgorithm;
- public void usePredefinedConfiguration(URI uri) {
+ /*public void usePredefinedConfiguration(URI uri) {
this.Configuration = org.dllearner.kb.sparql.Configuration.getConfiguration(uri);
this.TypedSparqlQuery = new TypedSparqlQuery(Configuration);
this.ExtractionAlgorithm = new ExtractionAlgorithm(Configuration);
+ }*/
+
+ public void usePredefinedConfiguration(int i) {
+
+ this.Configuration = PredefinedConfigurations.get(i);
+ this.TypedSparqlQuery = new TypedSparqlQuery(Configuration);
+ this.ExtractionAlgorithm = new ExtractionAlgorithm(Configuration);
}
- public void useConfiguration(SparqlQueryType SparqlQueryType, SparqlEndpoint SparqlEndpoint) {
+ public void useConfiguration(SparqlQueryType SparqlQueryType, SpecificSparqlEndpoint SparqlEndpoint, int recursiondepth,boolean getAllBackground) {
- this.Configuration = new Configuration(SparqlEndpoint, SparqlQueryType);
+ this.Configuration = new Configuration(SparqlEndpoint, SparqlQueryType,recursiondepth,getAllBackground);
this.TypedSparqlQuery = new TypedSparqlQuery(Configuration);
this.ExtractionAlgorithm = new ExtractionAlgorithm(Configuration);
}
Added: trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedConfigurations.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedConfigurations.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedConfigurations.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -0,0 +1,59 @@
+package org.dllearner.kb.sparql;
+
+import java.net.URL;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Set;
+
+public class PredefinedConfigurations {
+
+ public static Configuration get(int i){
+
+ switch (i){
+ case 0: return dbpediaYago();
+
+ }
+ return null;
+ }
+
+ public static Configuration dbpediaYago(){
+ URL u=null;
+ HashMap<String, String>m=new HashMap<String, String>();
+ m.put("default-graph-uri","http://dbpedia.org");
+ m.put("format","application/sparql-results.xml");
+ try{
+ u=new URL("http://dbpedia.openlinksw.com:8890/sparql");
+ }catch (Exception e) {e.printStackTrace();}
+ SpecificSparqlEndpoint sse=new SpecificSparqlEndpoint(
+ u,"dbpedia.openlinksw.com",m);
+ //System.out.println(u);
+ Set<String>pred=new HashSet<String>();
+ pred.add("http://www.w3.org/2004/02/skos/core");
+ pred.add("http://www.w3.org/2002/07/owl#sameAs");
+ pred.add("http://xmlns.com/foaf/0.1/");
+ pred.add("http://dbpedia.org/property/reference");
+ pred.add("http://dbpedia.org/property/website");
+ pred.add("http://dbpedia.org/property/wikipage");
+
+
+ Set<String>obj=new HashSet<String>();
+ obj.add("http://dbpedia.org/resource/Category:Wikipedia_");
+ obj.add("http://dbpedia.org/resource/Category:Articles_");
+ obj.add("http://xmlns.com/foaf/0.1/");
+ obj.add("http://upload.wikimedia.org/wikipedia/commons");
+ obj.add("http://upload.wikimedia.org/wikipedia");
+ obj.add("http://www.geonames.org");
+ obj.add("http://www.w3.org/2006/03/wn/wn20/instances/synset");
+ obj.add("http://www4.wiwiss.fu-berlin.de/flickrwrappr");
+ obj.add("http://www.w3.org/2004/02/skos/core");
+
+ SparqlQueryType sqt=new SparqlQueryType("forbid",obj,pred,"false");
+
+
+
+ return new Configuration(sse,sqt,2,true);
+
+ }
+
+
+}
Deleted: trunk/src/dl-learner/org/dllearner/kb/sparql/SimpleHTTPRequest.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SimpleHTTPRequest.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SimpleHTTPRequest.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -1,128 +0,0 @@
-package org.dllearner.kb.sparql;
-
-import java.io.BufferedInputStream;
-import java.io.BufferedReader;
-import java.io.IOException;
-import java.io.InputStream;
-import java.io.InputStreamReader;
-import java.io.OutputStream;
-import java.io.OutputStreamWriter;
-import java.net.HttpURLConnection;
-import java.net.InetAddress;
-import java.net.Socket;
-import java.net.URL;
-import java.net.URLEncoder;
-
-public class SimpleHTTPRequest {
- static final char value[] = { 13, 10 };
- static final String cut = new String(value);
- private InetAddress ia;
- private int port;
-
- public SimpleHTTPRequest(InetAddress ia, int port) {
- super();
- this.ia = ia;
- this.port = port;
-
- }
-
- public String sendAndReceive(String content) {
- String retval = "";
- //
-
- byte resp[] = null;
-
- try {
- Socket SparqlServer = new Socket(this.ia, this.port);
- // String request=makeHeader(content);
- // send request
- (SparqlServer.getOutputStream()).write(content.getBytes());
-
- // get Response
- resp = readBuffer(new BufferedInputStream(SparqlServer.getInputStream()));
- retval = new String(resp);
- retval = subtractResponseHeader(retval);
- // retval="||"+retval;
-
- SparqlServer.close();
-
- } catch (Exception e) {
- e.printStackTrace();
- }
- // System.out.println("got it");
- return retval;
-
- }// down
-
- public static byte[] readBuffer(InputStream IS) throws IOException {
- byte buffer[] = new byte[0xffff];
- int nbytes = 0;
- byte resp[] = new byte[0];
- while ((nbytes = IS.read(buffer)) != -1) {
- byte tmp[] = new byte[resp.length + nbytes];
- int i = 0;
- for (; i < resp.length; i++) {
- tmp[i] = resp[i];
- }
- for (int a = 0; a < nbytes; a++, i++) {
- tmp[i] = buffer[a];
- }
- resp = tmp;
- }
- return resp;
- }
-
- public String subtractResponseHeader(String in) {
- // System.out.println(in.indexOf(cut+""+cut));
- return in.substring(in.indexOf(cut + "" + cut) + 4);
-
- }
-
- private String sendAndReceive2(String sparql, URL url) throws IOException {
- StringBuilder answer = new StringBuilder();
-
- // String an Sparql-Endpoint schicken
- HttpURLConnection connection;
-
- connection = (HttpURLConnection) url.openConnection();
- connection.setDoOutput(true);
-
- connection.addRequestProperty("Host", "dbpedia.openlinksw.com");
- connection.addRequestProperty("Connection", "close");
- connection
- .addRequestProperty(
- "Accept",
- "text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5");
- connection.addRequestProperty("Accept-Language", "de-de,de;q=0.8,en-us;q=0.5,en;q=0.3");
- connection.addRequestProperty("Accept-Charset", "utf-8;q=1.0");
- connection
- .addRequestProperty(
- "User-Agent",
- "Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
-
- OutputStream os = connection.getOutputStream();
- OutputStreamWriter osw = new OutputStreamWriter(os);
- osw
- .write("default-graph-uri=http%3A%2F%2Fdbpedia.org&query="
- + URLEncoder.encode(sparql, "UTF-8")
- + "&format=application%2Fsparql-results%2Bxml");
- osw.close();
-
- // receive answer
- InputStream is = connection.getInputStream();
- InputStreamReader isr = new InputStreamReader(is, "UTF-8");
- BufferedReader br = new BufferedReader(isr);
-
- String line;
- do {
- line = br.readLine();
- if (line != null)
- answer.append(line);
- } while (line != null);
-
- br.close();
-
- return answer.toString();
- }
-
-}
Deleted: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlEndpoint.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlEndpoint.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -1,79 +0,0 @@
-package org.dllearner.kb.sparql;
-
-import java.util.HashMap;
-
-public class SparqlEndpoint {
-
- String host;
- int port;
- String hasAfterGET;
- String hasQueryParameter;
- String hasURL;
- public HashMap<String, String> parameters = new HashMap<String, String>();
-
- public SparqlEndpoint(String host, String port, String hasAfterGET, String hasQueryParameter,
- HashMap<String, String> parameters) {
- super();
- this.host = host;
- this.port = Integer.parseInt(port);
- this.hasAfterGET = hasAfterGET;
- this.hasQueryParameter = hasQueryParameter;
- this.parameters = parameters;
- }
-
- public SparqlEndpoint(String host, int port, String hasURL, HashMap<String, String> parameters) {
- super();
- this.port = port;
- this.host = host;
- this.hasURL = hasURL;
- this.hasQueryParameter = "query";
- this.parameters = parameters;
- }
-
- public String getHasAfterGET() {
- return hasAfterGET;
- }
-
- public void setHasAfterGET(String hasAfterGET) {
- this.hasAfterGET = hasAfterGET;
- }
-
- public String getHasQueryParameter() {
- return hasQueryParameter;
- }
-
- public void setHasQueryParameter(String hasQueryParameter) {
- this.hasQueryParameter = hasQueryParameter;
- }
-
- public String getHost() {
- return host;
- }
-
- public void setHost(String host) {
- this.host = host;
- }
-
- public HashMap<String, String> getParameters() {
- return parameters;
- }
-
- public void setParameters(HashMap<String, String> parameters) {
- this.parameters = parameters;
- }
-
- public int getPort() {
- return port;
- }
-
- public void setPort(int port) {
- this.port = port;
- }
-
- /*
- * sparql?default-graph-uri=http%3A%2F%2Fdbpedia.org&query=" +
- * //"SELECT%20%2A%20WHERE%20%7B%20%3Chttp%3A%2F%2Fdbpedia.org%2Fresource%2FAristotle%3E%20%3Fa%20%3Fb%20%7D%20" +
- * URLEncoder.encode(query, "UTF-8")+ //query+// URLencode
- * "&format=application%2Fsparql-results%2Bxml
- */
-}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlHTTPRequest.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlHTTPRequest.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlHTTPRequest.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -1,81 +1,105 @@
package org.dllearner.kb.sparql;
+import java.io.BufferedReader;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.OutputStream;
+import java.io.OutputStreamWriter;
+import java.net.HttpURLConnection;
import java.net.InetAddress;
import java.net.URLEncoder;
import java.util.Iterator;
import java.util.Set;
-public class SparqlHTTPRequest {
- static final char value[] = { 13, 10 };
- static final String cut = new String(value);
- private SparqlEndpoint SparqlEndpoint;
- private SimpleHTTPRequest SimpleHTTPRequest;
- public SparqlHTTPRequest(SparqlEndpoint SparqlEndpoint) {
- this.SparqlEndpoint = SparqlEndpoint;
- InetAddress ia = null;
- try {
- ia = InetAddress.getByName(SparqlEndpoint.getHost());
- } catch (Exception e) {
- e.printStackTrace();
- }
- this.SimpleHTTPRequest = new SimpleHTTPRequest(ia, SparqlEndpoint.getPort());
-
+public class SparqlHTTPRequest {
+
+
+ private SpecificSparqlEndpoint SparqlEndpoint;
+
+
+
+ public SparqlHTTPRequest(SpecificSparqlEndpoint SparqlEndpoint){
+ this.SparqlEndpoint=SparqlEndpoint;
+
+
}
-
- public String sendAndReceiveSPARQL(String sparql) {
-
- // System.out.println(sparql);
- String content = makeContent(sparql);
- // System.out.println(content);
- String ret = this.SimpleHTTPRequest.sendAndReceive(content);
- // System.out.println(ret);
-
- // this.sendAndReceiveSPARQL("SELECT * WHERE {?a ?b ?c} LIMIT 10");
-
+
+
+ public String sendAndReceiveSPARQL( String sparql){
+ String ret= "";
+ try{
+ //System.out.println(sparql);
+
+ //System.out.println(content);
+
+ ret=this.sendAndReceive(sparql);
+ //System.out.println(ret);
+
+ //this.sendAndReceiveSPARQL("SELECT * WHERE {?a ?b ?c} LIMIT 10");
+ }catch (Exception e) {e.printStackTrace();}
return ret;
-
- }// down
-
- public String makeContent(String query) {
-
- String RequestHeader = "";
- try {
-
- RequestHeader = "GET ";
- RequestHeader += SparqlEndpoint.getHasAfterGET() + "?";
- // parameters
- Set<String> s = SparqlEndpoint.getParameters().keySet();
- Iterator<String> it = s.iterator();
- while (it.hasNext()) {
- String element = (String) it.next();
- RequestHeader += "" + URLEncoder.encode(element, "UTF-8") + "="
- + URLEncoder.encode(SparqlEndpoint.getParameters().get(element), "UTF-8")
- + "&";
- }
- RequestHeader += "" + SparqlEndpoint.getHasQueryParameter() + "="
- + URLEncoder.encode(query, "UTF-8");
- RequestHeader += " HTTP/1.1" + cut;
- RequestHeader += "Host: " + SparqlEndpoint.getHost() + cut;
-
- RequestHeader += "Connection: close"
- + cut
- +
- // "Accept-Encoding: gzip"+cut+
- "Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5"
- + cut
- + "Accept-Language: de-de,de;q=0.8,en-us;q=0.5,en;q=0.3"
- + cut
- + "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7"
- + cut
- + "User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24"
- + cut + cut;
- } catch (Exception e) {
- e.printStackTrace();
+
+
+ }//down
+
+
+
+
+
+
+ private String sendAndReceive(String sparql) throws IOException{
+ StringBuilder answer = new StringBuilder();
+
+ // String an Sparql-Endpoint schicken
+ HttpURLConnection connection;
+
+ connection = (HttpURLConnection) this.SparqlEndpoint.getURL().openConnection();
+ connection.setDoOutput(true);
+
+ connection.addRequestProperty("Host", this.SparqlEndpoint.getHost());
+ connection.addRequestProperty("Connection","close");
+ connection.addRequestProperty("Accept","text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5");
+ connection.addRequestProperty("Accept-Language","de-de,de;q=0.8,en-us;q=0.5,en;q=0.3");
+ connection.addRequestProperty("Accept-Charset","utf-8;q=1.0");
+ connection.addRequestProperty("User-Agent","Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
+
+ OutputStream os = connection.getOutputStream();
+ OutputStreamWriter osw = new OutputStreamWriter(os);
+
+ Set<String> s =SparqlEndpoint.getParameters().keySet();
+ Iterator<String> it=s.iterator();
+ String FullURI="";
+ while (it.hasNext()) {
+ String element = (String) it.next();
+ FullURI+=""+URLEncoder.encode(element, "UTF-8")+"="+
+ URLEncoder.encode(SparqlEndpoint.getParameters().get(element), "UTF-8")+"&";
}
- return RequestHeader;
-
+ //System.out.println(FullURI);
+ FullURI+=""+SparqlEndpoint.getHasQueryParameter()+"="+URLEncoder.encode(sparql, "UTF-8");
+
+
+ osw.write(FullURI);
+ osw.close();
+
+ // receive answer
+ InputStream is = connection.getInputStream();
+ InputStreamReader isr = new InputStreamReader(is,"UTF-8");
+ BufferedReader br = new BufferedReader(isr);
+
+ String line;
+ do {
+ line = br.readLine();
+ if(line!=null)
+ answer.append(line);
+ } while (line != null);
+
+ br.close();
+
+ return answer.toString();
}
+
+}
-}
Copied: trunk/src/dl-learner/org/dllearner/kb/sparql/SpecificSparqlEndpoint.java (from rev 296, trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlEndpoint.java)
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SpecificSparqlEndpoint.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SpecificSparqlEndpoint.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -0,0 +1,53 @@
+package org.dllearner.kb.sparql;
+
+import java.net.URL;
+import java.util.HashMap;
+
+public class SpecificSparqlEndpoint {
+
+
+
+ String host;
+ String hasQueryParameter;
+ URL URL;
+ public HashMap<String, String> parameters = new HashMap<String, String>();
+
+
+ public SpecificSparqlEndpoint(URL url,String host, HashMap<String, String> parameters) {
+ super();
+ this.host=host;
+ this.URL = url;
+ this.hasQueryParameter = "query";
+ this.parameters = parameters;
+ }
+
+
+ public String getHasQueryParameter() {
+ return hasQueryParameter;
+ }
+
+ public void setHasQueryParameter(String hasQueryParameter) {
+ this.hasQueryParameter = hasQueryParameter;
+ }
+
+ public String getHost() {
+ return host;
+ }
+
+ public void setHost(String host) {
+ this.host = host;
+ }
+
+ public HashMap<String, String> getParameters() {
+ return parameters;
+ }
+
+ public void setParameters(HashMap<String, String> parameters) {
+ this.parameters = parameters;
+ }
+
+ public URL getURL() {
+ return this.URL;
+ }
+
+}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Test.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Test.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Test.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -13,7 +13,7 @@
try {
URI u = new URI(test);
Manager m = new Manager();
- m.usePredefinedConfiguration(u);
+ //m.usePredefinedConfiguration(u);
URI u2 = new URI("http://dbpedia.org/resource/Angela_Merkel");
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java 2007-12-02 12:03:41 UTC (rev 297)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java 2007-12-02 13:29:58 UTC (rev 298)
@@ -24,12 +24,14 @@
// check cache
String FromCache = this.Cache.get(u.toString(), sparql);
- FromCache = null;
+
String xml;
// if not in cache get it from EndPoint
if (FromCache == null) {
xml = this.SparqlHTTPRequest.sendAndReceiveSPARQL(sparql);
- // this.Cache.put(u.toString(), xml, sparql);
+ //System.out.println(sparql);
+ //System.out.println(xml);
+ this.Cache.put(u.toString(), xml, sparql);
System.out.print("\n");
} else {
xml = FromCache;
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <ku...@us...> - 2007-12-03 14:45:16
|
Revision: 310
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=310&view=rev
Author: kurzum
Date: 2007-12-03 06:45:11 -0800 (Mon, 03 Dec 2007)
Log Message:
-----------
changed and Formatted everything
Delete old Cache files
Examples will follow
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SpecificSparqlEndpoint.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Test.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Tupel.java
trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -43,6 +43,7 @@
import org.dllearner.core.config.StringSetConfigOption;
import org.dllearner.core.dl.KB;
import org.dllearner.kb.sparql.Manager;
+import org.dllearner.kb.sparql.Manipulator;
import org.dllearner.kb.sparql.PredefinedEndpoint;
import org.dllearner.kb.sparql.PredefinedFilter;
import org.dllearner.kb.sparql.SparqlQueryType;
@@ -52,94 +53,108 @@
import org.dllearner.reasoning.JenaOWLDIGConverter;
/**
- * Represents a SPARQL Endpoint.
- * TODO: Is it necessary to create a class DBpediaSparqlEndpoint?
+ * Represents a SPARQL Endpoint. TODO: Is it necessary to create a class
+ * DBpediaSparqlEndpoint?
*
* @author Jens Lehmann
* @author Sebastian Knappe
*/
public class SparqlEndpointRestructured extends KnowledgeSource {
-
- //ConfigOptions
- private URL url;
+
+ // ConfigOptions
+ private URL url;
String host;
private Set<String> instances;
private URL dumpFile;
private int numberOfRecursions;
- private int predefinedFilter=0;
- private int predefinedEndpoint=0;
+ private int predefinedFilter = 0;
+ private int predefinedEndpoint = 0;
private Set<String> predList;
private Set<String> objList;
private Set<String> classList;
private String format;
private boolean dumpToFile;
- private boolean useLits=false;
- private boolean getAllBackground=false;
-
+ private boolean useLits = false;
+ private boolean getAllBackground = false;
+
+ private boolean learnDomain = false;
+ private String role;
+ private String blankNodeIdentifier = "bnode";
+
/**
- * Holds the results of the calculateSubjects method
+ * Holds the results of the calculateSubjects method
*/
private String[] subjects;
-
+
/**
- * Holds the results of the calculateTriples method
+ * Holds the results of the calculateTriples method
*/
private String[] triples;
-
+
/**
- * Holds the results of the calculateConceptSubjects method
+ * Holds the results of the calculateConceptSubjects method
*/
private String[] conceptSubjects;
-
+
/**
* if a method is running this becomes true
*/
- private boolean subjectThreadRunning=false;
-
- private boolean triplesThreadRunning=false;
-
- private boolean conceptThreadRunning=false;
-
+ private boolean subjectThreadRunning = false;
+
+ private boolean triplesThreadRunning = false;
+
+ private boolean conceptThreadRunning = false;
+
/**
* the Thread that is running a method
*/
private Thread subjectThread;
-
+
private Thread triplesThread;
-
+
private Thread conceptThread;
-
- //received ontology as array, used if format=Array(an element of the
- //array consists of the subject, predicate and object separated by '<'
+
+ // received ontology as array, used if format=Array(an element of the
+ // array consists of the subject, predicate and object separated by '<'
private String[] ontArray;
-
- //received ontology as KB, the internal format
+
+ // received ontology as KB, the internal format
private KB kb;
-
+
public static String getName() {
return "SPARQL Endpoint Restructured";
}
-
+
/**
* sets the ConfigOptions for this KnowledgeSource
+ *
* @return
*/
public static Collection<ConfigOption<?>> createConfigOptions() {
- Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
+ Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
options.add(new StringConfigOption("url", "URL of SPARQL Endpoint"));
options.add(new StringConfigOption("host", "host of SPARQL Endpoint"));
- options.add(new StringSetConfigOption("instances","relevant instances e.g. positive and negative examples in a learning problem"));
- options.add(new IntegerConfigOption("numberOfRecursions","number of Recursions, the Sparql-Endpoint is asked"));
- options.add(new IntegerConfigOption("predefinedFilter","the mode of the SPARQL Filter"));
- options.add(new IntegerConfigOption("predefinedEndpoint","the mode of the SPARQL Filter"));
-
- options.add(new StringSetConfigOption("predList","a predicate list"));
- options.add(new StringSetConfigOption("objList","an object list"));
- options.add(new StringSetConfigOption("classList","a class list"));
+ options.add(new StringSetConfigOption("instances",
+ "relevant instances e.g. positive and negative examples in a learning problem"));
+ options.add(new IntegerConfigOption("numberOfRecursions",
+ "number of Recursions, the Sparql-Endpoint is asked"));
+ options.add(new IntegerConfigOption("predefinedFilter", "the mode of the SPARQL Filter"));
+ options.add(new IntegerConfigOption("predefinedEndpoint", "the mode of the SPARQL Filter"));
+
+ options.add(new StringSetConfigOption("predList", "a predicate list"));
+ options.add(new StringSetConfigOption("objList", "an object list"));
+ options.add(new StringSetConfigOption("classList", "a class list"));
options.add(new StringConfigOption("format", "N-TRIPLES or KB format"));
- options.add(new BooleanConfigOption("dumpToFile", "wether Ontology from DBPedia is written to a file or not"));
- options.add(new BooleanConfigOption("useLits","use Literals in SPARQL query"));
- options.add(new BooleanConfigOption("getAllBackground","get"));
+ options.add(new BooleanConfigOption("dumpToFile",
+ "wether Ontology from DBPedia is written to a file or not"));
+ options.add(new BooleanConfigOption("useLits", "use Literals in SPARQL query"));
+ options.add(new BooleanConfigOption("getAllBackground", "get"));
+
+ options.add(new BooleanConfigOption("learnDomain", "learns the Domain for a Role"));
+ options.add(new StringConfigOption("role", "role to learn Domain from"));
+ options.add(new StringConfigOption("blankNodeIdentifier",
+ "used to identify blanknodes in Tripels"));
+
return options;
}
@@ -147,7 +162,7 @@
* @see org.dllearner.core.Component#applyConfigEntry(org.dllearner.core.ConfigEntry)
*/
@Override
- @SuppressWarnings({"unchecked"})
+ @SuppressWarnings( { "unchecked" })
public <T> void applyConfigEntry(ConfigEntry<T> entry) throws InvalidConfigOptionValueException {
String option = entry.getOptionName();
if (option.equals("url")) {
@@ -155,103 +170,124 @@
try {
url = new URL(s);
} catch (MalformedURLException e) {
- throw new InvalidConfigOptionValueException(entry.getOption(), entry.getValue(),"malformed URL " + s);
+ throw new InvalidConfigOptionValueException(entry.getOption(), entry.getValue(),
+ "malformed URL " + s);
}
- } else if(option.equals("host")) {
+ } else if (option.equals("host")) {
host = (String) entry.getValue();
- } else if(option.equals("instances")) {
+ } else if (option.equals("instances")) {
instances = (Set<String>) entry.getValue();
- } else if(option.equals("numberOfRecursions")){
- numberOfRecursions=(Integer)entry.getValue();
- } else if(option.equals("predList")) {
+ } else if (option.equals("numberOfRecursions")) {
+ numberOfRecursions = (Integer) entry.getValue();
+ } else if (option.equals("predList")) {
predList = (Set<String>) entry.getValue();
- } else if(option.equals("objList")) {
+ } else if (option.equals("objList")) {
objList = (Set<String>) entry.getValue();
- } else if(option.equals("classList")) {
+ } else if (option.equals("classList")) {
classList = (Set<String>) entry.getValue();
- } else if(option.equals("predefinedEndpoint")){
- predefinedEndpoint=(Integer)entry.getValue();
- } else if(option.equals("predefinedFilter")){
- predefinedFilter=(Integer)entry.getValue();
- } else if(option.equals("format")){
- format=(String)entry.getValue();
- } else if(option.equals("dumpToFile")){
- dumpToFile=(Boolean)entry.getValue();
- } else if(option.equals("useLits")){
- useLits=(Boolean)entry.getValue();
- } else if(option.equals("getAllBackground")){
- getAllBackground=(Boolean)entry.getValue();
+ } else if (option.equals("predefinedEndpoint")) {
+ predefinedEndpoint = (Integer) entry.getValue();
+ } else if (option.equals("predefinedFilter")) {
+ predefinedFilter = (Integer) entry.getValue();
+ } else if (option.equals("format")) {
+ format = (String) entry.getValue();
+ } else if (option.equals("dumpToFile")) {
+ dumpToFile = (Boolean) entry.getValue();
+ } else if (option.equals("useLits")) {
+ useLits = (Boolean) entry.getValue();
+ } else if (option.equals("getAllBackground")) {
+ getAllBackground = (Boolean) entry.getValue();
+ } else if (option.equals("learnDomain")) {
+ learnDomain = (Boolean) entry.getValue();
+ } else if (option.equals("role")) {
+ role = (String) entry.getValue();
+ } else if (option.equals("blankNodeIdentifier")) {
+ blankNodeIdentifier = (String) entry.getValue();
}
+
}
- /* (non-Javadoc)
+ /*
+ * (non-Javadoc)
+ *
* @see org.dllearner.core.Component#init()
*/
@Override
public void init() {
System.out.println("SparqlModul: Collecting Ontology");
- //SparqlOntologyCollector oc=
-
-
- //new SparqlOntologyCollector(Datastructures.setToArray(instances), numberOfRecursions, filterMode,
- //Datastructures.setToArray(predList),Datastructures.setToArray( objList),Datastructures.setToArray(classList),format,url,useLits);
- Manager m=new Manager();
- SpecificSparqlEndpoint sse=null;
- SparqlQueryType sqt=null;
+ // SparqlOntologyCollector oc=
+ // new SparqlOntologyCollector(Datastructures.setToArray(instances),
+ // numberOfRecursions, filterMode,
+ // Datastructures.setToArray(predList),Datastructures.setToArray(
+ // objList),Datastructures.setToArray(classList),format,url,useLits);
+ Manager m = new Manager();
+ SpecificSparqlEndpoint sse = null;
+ SparqlQueryType sqt = null;
+ // get Options for Manipulator
+ Manipulator man = new Manipulator(blankNodeIdentifier);
HashMap<String, String> parameters = new HashMap<String, String>();
parameters.put("default-graph-uri", "http://dbpedia.org");
parameters.put("format", "application/sparql-results.xml");
-
- if(predefinedEndpoint>=1){
- sse=PredefinedEndpoint.getEndpoint(predefinedEndpoint);
-
+
+ // get Options for endpoints
+ if (predefinedEndpoint >= 1) {
+ sse = PredefinedEndpoint.getEndpoint(predefinedEndpoint);
+ } else {
+ sse = new SpecificSparqlEndpoint(url, host, parameters);
}
- else{
- SpecificSparqlEndpoint se=new SpecificSparqlEndpoint(url, host, parameters);
-
+
+ // get Options for Filters
+ if (predefinedFilter >= 1) {
+ sqt = PredefinedFilter.getFilter(predefinedFilter);
+
+ } else {
+ sqt = new SparqlQueryType("forbid", objList, predList, useLits + "");
}
-
- if(predefinedFilter>=1){
- sqt=PredefinedFilter.getFilter(predefinedFilter);
-
+ // give everything to the manager
+ m.useConfiguration(sqt, sse, man, numberOfRecursions, getAllBackground);
+ try {
+ String ont = "";
+ // used to learn a domain of a role
+ if (learnDomain) {
+ instances = m.getDomainInstancesForRole(role);
+ // ad the role to the filter(a solution is always EXISTS
+ // role.TOP)
+ m.addPredicateFilter(role);
+ System.out.println(instances);
}
- else{
- sqt=new SparqlQueryType("forbid", objList,predList,useLits+"");
-
-
-
- }
- m.useConfiguration(sqt, sse,numberOfRecursions,getAllBackground);
- try
- {
- String ont=m.extract(instances);
-
- if (dumpToFile){
- String filename=System.currentTimeMillis()+".nt";
- String basedir="cache"+File.separator;
- try{
- if(!new File(basedir).exists())
+ // the actual extraction is started here
+ ont = m.extract(instances);
+
+ if (dumpToFile) {
+ String filename = System.currentTimeMillis() + ".nt";
+ String basedir = "cache" + File.separator;
+ try {
+ if (!new File(basedir).exists())
new File(basedir).mkdir();
-
- FileWriter fw=new FileWriter(new File(basedir+filename),true);
+
+ FileWriter fw = new FileWriter(new File(basedir + filename), true);
fw.write(ont);
fw.flush();
fw.close();
-
- dumpFile=(new File(basedir+filename)).toURI().toURL();
- }catch (Exception e) {e.printStackTrace();}
+
+ dumpFile = (new File(basedir + filename)).toURI().toURL();
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
}
if (format.equals("KB")) {
- try{
- kb=KBParser.parseKBFile(new StringReader(ont));
- } catch(Exception e) {e.printStackTrace();}
+ try {
+ kb = KBParser.parseKBFile(new StringReader(ont));
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
}
- }catch(Exception e) {
+ } catch (Exception e) {
e.printStackTrace();
}
System.out.println("SparqlModul: ****Finished");
}
-
+
/*
* (non-Javadoc)
*
@@ -259,97 +295,94 @@
*/
@Override
public String toDIG(URI kbURI) {
- if (format.equals("N-TRIPLES")) return JenaOWLDIGConverter.getTellsString(dumpFile, OntologyFormat.N_TRIPLES, kbURI);
- else return DIGConverter.getDIGString(kb, kbURI).toString();
+ if (format.equals("N-TRIPLES"))
+ return JenaOWLDIGConverter.getTellsString(dumpFile, OntologyFormat.N_TRIPLES, kbURI);
+ else
+ return DIGConverter.getDIGString(kb, kbURI).toString();
}
- /* (non-Javadoc)
- * @see org.dllearner.core.KnowledgeSource#export(java.io.File, org.dllearner.core.OntologyFormat)
+ /*
+ * (non-Javadoc)
+ *
+ * @see org.dllearner.core.KnowledgeSource#export(java.io.File,
+ * org.dllearner.core.OntologyFormat)
*/
@Override
public void export(File file, OntologyFormat format) throws OntologyFormatUnsupportedException {
- // currently no export functions implemented, so we just throw an exception
+ // currently no export functions implemented, so we just throw an
+ // exception
throw new OntologyFormatUnsupportedException("export", format);
}
-
+
public URL getURL() {
return url;
}
-
+
public String[] getOntArray() {
return ontArray;
}
-
- public void calculateSubjects(String label,int limit)
- {
+
+ public void calculateSubjects(String label, int limit) {
System.out.println("SparqlModul: Collecting Subjects");
- SparqlOntologyCollector oc=new SparqlOntologyCollector(url);
- try{
- subjects=oc.getSubjectsFromLabel(label,limit);
- }catch (IOException e){
- subjects=new String[1];
- subjects[0]="[Error]Sparql Endpoint could not be reached.";
+ SparqlOntologyCollector oc = new SparqlOntologyCollector(url);
+ try {
+ subjects = oc.getSubjectsFromLabel(label, limit);
+ } catch (IOException e) {
+ subjects = new String[1];
+ subjects[0] = "[Error]Sparql Endpoint could not be reached.";
}
System.out.println("SparqlModul: ****Finished");
}
-
+
public void calculateTriples(String subject) {
System.out.println("SparqlModul: Collecting Triples");
- SparqlOntologyCollector oc=new SparqlOntologyCollector(url);
- try{
- triples=oc.collectTriples(subject);
- }catch (IOException e){
- triples=new String[1];
- triples[0]="[Error]Sparql Endpoint could not be reached.";
+ SparqlOntologyCollector oc = new SparqlOntologyCollector(url);
+ try {
+ triples = oc.collectTriples(subject);
+ } catch (IOException e) {
+ triples = new String[1];
+ triples[0] = "[Error]Sparql Endpoint could not be reached.";
}
System.out.println("SparqlModul: ****Finished");
}
-
- public void calculateConceptSubjects(String concept)
- {
+
+ public void calculateConceptSubjects(String concept) {
System.out.println("SparqlModul: Collecting Subjects");
- SparqlOntologyCollector oc=new SparqlOntologyCollector(url);
- try{
- conceptSubjects=oc.getSubjectsFromConcept(concept);
- }catch (IOException e){
- conceptSubjects=new String[1];
- conceptSubjects[0]="[Error]Sparql Endpoint could not be reached.";
+ SparqlOntologyCollector oc = new SparqlOntologyCollector(url);
+ try {
+ conceptSubjects = oc.getSubjectsFromConcept(concept);
+ } catch (IOException e) {
+ conceptSubjects = new String[1];
+ conceptSubjects[0] = "[Error]Sparql Endpoint could not be reached.";
}
System.out.println("SparqlModul: ****Finished");
}
-
- public boolean subjectThreadIsRunning()
- {
+
+ public boolean subjectThreadIsRunning() {
return subjectThreadRunning;
}
-
- public void setSubjectThreadRunning(boolean bool)
- {
- subjectThreadRunning=bool;
+
+ public void setSubjectThreadRunning(boolean bool) {
+ subjectThreadRunning = bool;
}
-
- public boolean triplesThreadIsRunning()
- {
+
+ public boolean triplesThreadIsRunning() {
return triplesThreadRunning;
}
-
- public void setTriplesThreadRunning(boolean bool)
- {
- triplesThreadRunning=bool;
+
+ public void setTriplesThreadRunning(boolean bool) {
+ triplesThreadRunning = bool;
}
-
- public boolean conceptThreadIsRunning()
- {
+
+ public boolean conceptThreadIsRunning() {
return conceptThreadRunning;
}
-
- public void setConceptThreadRunning(boolean bool)
- {
- conceptThreadRunning=bool;
+
+ public void setConceptThreadRunning(boolean bool) {
+ conceptThreadRunning = bool;
}
-
- public String[] getSubjects()
- {
+
+ public String[] getSubjects() {
return subjects;
}
@@ -377,13 +410,11 @@
this.conceptThread = conceptThread;
}
- public String[] getTriples()
- {
+ public String[] getTriples() {
return triples;
}
-
- public String[] getConceptSubjects()
- {
+
+ public String[] getConceptSubjects() {
return conceptSubjects;
}
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.io.File;
@@ -7,19 +26,20 @@
import java.io.ObjectOutputStream;
import java.io.Serializable;
import java.net.URLEncoder;
+import java.util.HashMap;
public class Cache implements Serializable {
// Object can be the cache itself
// or a cache object(one entry)
+ // it now uses a hashmap and can contain different queries at once
final static long serialVersionUID = 104;
transient String basedir = "";
transient String fileending = ".cache";
long timestamp;
- String content = "";
long daysoffreshness = 15;
long multiplier = 24 * 60 * 60 * 1000;// h m s ms
- String sparqlquery = "";
+ private HashMap<String, String> hm;
// constructor for the cache itself
public Cache(String path) {
@@ -28,14 +48,15 @@
System.out.println(new File(path).mkdir());
;
}
-
}
// constructor for single cache object(one entry)
- public Cache(String c, String sparql) {
- this.content = c;
- this.sparqlquery = sparql;
+ public Cache(String sparql, String content) {
+ // this.content = c;
+ // this.sparqlquery = sparql;
this.timestamp = System.currentTimeMillis();
+ this.hm = new HashMap<String, String>();
+ hm.put(sparql, content);
}
public String get(String key, String sparql) {
@@ -49,20 +70,32 @@
if (!c.checkFreshness())
return null;
// System.out.println("fresh");
- if (!c.validate(sparql))
+ String xml = "";
+ try {
+ xml = c.hm.get(sparql);
+ } catch (Exception e) {
return null;
+ }
+ return xml;
// System.out.println("valid");
- ret = c.content;
+ // ret = c.content;
} catch (Exception e) {
e.printStackTrace();
}
return ret;
};
- public void put(String key, String content, String sparql) {
+ public void put(String key, String sparql, String content) {
// System.out.println("put into "+key);
- Cache c = new Cache(content, sparql);
- putIntoFile(makeFilename(key), c);
+ Cache c = readFromFile(makeFilename(key));
+ if (c == null) {
+ c = new Cache(sparql, content);
+ putIntoFile(makeFilename(key), c);
+ } else {
+ c.hm.put(sparql, content);
+ putIntoFile(makeFilename(key), c);
+ }
+
}
String makeFilename(String key) {
@@ -83,14 +116,6 @@
return false;
}
- boolean validate(String sparql) {
- if (this.sparqlquery.equals(sparql))
- // valid
- return true;
- else
- return false;
- }
-
public void checkFile(String Filename) {
if (!new File(Filename).exists()) {
try {
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URI;
@@ -6,6 +25,7 @@
import java.util.Set;
import java.util.Vector;
+// is a node in the graph that is a class
public class ClassNode extends Node {
Set<PropertyNode> properties = new HashSet<PropertyNode>();
@@ -29,6 +49,17 @@
ClassNode tmp = new ClassNode(new URI(t.b));
properties.add(new PropertyNode(new URI(m.subclass), this, tmp));
Nodes.add(tmp);
+ } else {
+ ClassNode tmp = new ClassNode(new URI(t.b));
+ properties.add(new PropertyNode(new URI(t.a), this, tmp));
+ // System.out.println(m.blankNodeIdentifier);
+ // System.out.println("XXXXX"+t.b);
+ if (t.b.startsWith(m.blankNodeIdentifier)) {
+ tmp.expand(tsq, m);
+ System.out.println(m.blankNodeIdentifier);
+ System.out.println("XXXXX" + t.b);
+ }
+ // Nodes.add(tmp);
}
} catch (Exception e) {
System.out.println(t);
@@ -40,6 +71,12 @@
}
@Override
+ public Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m) {
+ // TODO return type doesn't make sense
+ return new Vector<Node>();
+ }
+
+ @Override
public Set<String> toNTriple() {
Set<String> s = new HashSet<String>();
s.add("<" + this.uri + "><" + "http://www.w3.org/1999/02/22-rdf-syntax-ns#type" + "><"
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Configuration.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,227 +1,71 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
-import java.io.File;
-import java.net.URI;
-import java.net.URL;
-import java.util.HashMap;
-import java.util.HashSet;
-import java.util.Iterator;
-import java.util.Map;
-import java.util.Set;
-
-import org.semanticweb.owl.apibinding.OWLManager;
-import org.semanticweb.owl.model.OWLConstant;
-import org.semanticweb.owl.model.OWLDataPropertyExpression;
-import org.semanticweb.owl.model.OWLIndividual;
-import org.semanticweb.owl.model.OWLObjectPropertyExpression;
-import org.semanticweb.owl.model.OWLOntology;
-import org.semanticweb.owl.model.OWLOntologyManager;
-
+//stores all configuration settings
public class Configuration {
- private SpecificSparqlEndpoint SparqlEndpoint;
- private SparqlQueryType SparqlQueryType;
- private Manipulator Manipulator;
+ private SpecificSparqlEndpoint specificSparqlEndpoint;
+ private SparqlQueryType sparqlQueryType;
+ private Manipulator manipulator;
private int recursiondepth = 2;
private boolean getAllBackground = true;
-
+
private Configuration() {
}
-
- public Configuration(SpecificSparqlEndpoint SparqlEndpoint,
- SparqlQueryType SparqlQueryType, int recursiondepth,
+ public Configuration(SpecificSparqlEndpoint specificSparqlEndpoint,
+ SparqlQueryType sparqlQueryType, Manipulator manipulator, int recursiondepth,
boolean getAllBackground) {
- this.SparqlEndpoint = SparqlEndpoint;
- this.SparqlQueryType = SparqlQueryType;
- this.Manipulator=new Manipulator();
+ this.specificSparqlEndpoint = specificSparqlEndpoint;
+ this.sparqlQueryType = sparqlQueryType;
+ this.manipulator = manipulator;
this.recursiondepth = recursiondepth;
this.getAllBackground = getAllBackground;
-
}
- public static Configuration getConfiguration(URI uri) {
- // public static String getTellsString(URL file, URI kbURI){//throws
- // OWLOntologyCreationException{
- Configuration ret = new Configuration();
- try {
- String file = "config/config.owl";
+ public Configuration changeQueryType(SparqlQueryType sqt) {
+ // TODO must clone here
+ return new Configuration(this.specificSparqlEndpoint, sqt, this.manipulator,
+ this.recursiondepth, this.getAllBackground);
- File f = new File(file);
- String fileURL = "file:///" + f.getAbsolutePath();
- URL u = new URL(fileURL);
- /* Load an ontology from a physical URI */
- OWLOntologyManager manager = OWLManager.createOWLOntologyManager();
- OWLOntology ontology = manager.loadOntologyFromPhysicalURI(u.toURI());
- // System.out.println( ontology.containsIndividualReference(uri));
- // OWLIndividualImpl ind=new OWLIndividualImpl();
- // System.out.println(ontology.getReferencedIndividuals());
- Set<OWLIndividual> s = ontology.getReferencedIndividuals();
- // System.out.println(ontology.getReferencedClasses());
- // Set<OWLIndividualAxiom> s= ontology.getIndividualAxioms();
- Iterator<OWLIndividual> it = s.iterator();
- while (it.hasNext()) {
- OWLIndividual tmp = (OWLIndividual) it.next();
- // tmp.getURI()
- if (tmp.getURI().equals(uri)) {
- OWLIndividual[] arr = getIndividualsForProperty("hasSparqlEndpoint", tmp
- .getObjectPropertyValues(ontology));
- OWLIndividual sEndpoint = arr[0];
- ret.SparqlEndpoint = makeEndpoint(sEndpoint, ontology);
- arr = getIndividualsForProperty("hasTypedQuery", tmp
- .getObjectPropertyValues(ontology));
- OWLIndividual typedQuery = arr[0];
- ret.SparqlQueryType = makeSparqlQueryType(typedQuery, ontology);
-
- }
- // {hasSparqlEndpoint=[dbpediaEndpoint]}
- }
-
- ret.Manipulator = makeManipulator();
- } catch (Exception e) {
- e.printStackTrace();
- }
-
- return ret;
}
- public static OWLIndividual[] getIndividualsForProperty(String propertyname,
- Map<OWLObjectPropertyExpression, Set<OWLIndividual>> m) {
- Set<OWLObjectPropertyExpression> s = m.keySet();
-
- Iterator<OWLObjectPropertyExpression> it = s.iterator();
- while (it.hasNext()) {
- OWLObjectPropertyExpression tmp = (OWLObjectPropertyExpression) it.next();
- // System.out.println(tmp);
- // System.out.println(propertyname);
- if (tmp.toString().equals(propertyname)) {
- Object[] arr = ((Set<OWLIndividual>) m.get(tmp)).toArray();
- OWLIndividual[] o = new OWLIndividual[arr.length];
- for (int i = 0; i < o.length; i++) {
- o[i] = (OWLIndividual) arr[i];
- }
-
- return o;
- }
- }
- return null;
-
- }
-
- public static String getFirstValueForDataProperty(String propertyname,
- Map<OWLDataPropertyExpression, Set<OWLConstant>> m) {
- return getValuesForDataProperty(propertyname, m)[0];
- }
-
- public static String[] getValuesForDataProperty(String propertyname,
- Map<OWLDataPropertyExpression, Set<OWLConstant>> m) {
- Set<OWLDataPropertyExpression> s = m.keySet();
-
- Iterator<OWLDataPropertyExpression> it = s.iterator();
- while (it.hasNext()) {
- OWLDataPropertyExpression tmp = (OWLDataPropertyExpression) it.next();
- if (tmp.toString().equals(propertyname)) {
- Object[] arr = ((Set<OWLConstant>) m.get(tmp)).toArray();
- String[] str = new String[arr.length];
- for (int i = 0; i < str.length; i++) {
- str[i] = ((OWLConstant) arr[i]).getLiteral();
- }
- return str;
- }
- }
- return null;
-
- }
-
- public static SpecificSparqlEndpoint makeEndpoint(OWLIndividual sEndpoint, OWLOntology o) {
- String host = getFirstValueForDataProperty("hasHost", sEndpoint.getDataPropertyValues(o));
- String port = getFirstValueForDataProperty("hasPort", sEndpoint.getDataPropertyValues(o));
- String hasAfterGET = getFirstValueForDataProperty("hasAfterGET", sEndpoint
- .getDataPropertyValues(o));
- String hasQueryParameter = getFirstValueForDataProperty("hasQueryParameter", sEndpoint
- .getDataPropertyValues(o));
- OWLIndividual[] para = getIndividualsForProperty("hasGETParameter", sEndpoint
- .getObjectPropertyValues(o));
- // System.out.println("test");
- HashMap<String, String> parameters = new HashMap<String, String>();
- if (para == null)
- return null;//new SpecificSparqlEndpoint(host, port, hasAfterGET, hasQueryParameter, parameters);
- for (OWLIndividual p : para) {
- // System.out.println("test2");
- String a1 = getFirstValueForDataProperty("hasParameterName", p.getDataPropertyValues(o));
- String a2 = getFirstValueForDataProperty("hasParameterContent", p
- .getDataPropertyValues(o));
- parameters.put(a1, a2);
- }
- // System.out.println("test2");
- // System.out.println(host+port+ hasAfterGET+ hasQueryParameter+
- // parameters);
- return null;//new SpecificSparqlEndpoint(host, port, hasAfterGET, hasQueryParameter, parameters);
-
- }
-
- public static SparqlQueryType makeSparqlQueryType(OWLIndividual typedQuery, OWLOntology o) {
- String useLiterals = getFirstValueForDataProperty("usesLiterals", typedQuery
- .getDataPropertyValues(o));
- String hasMode = getFirstValueForDataProperty("hasMode", typedQuery
- .getDataPropertyValues(o));
- // String
- // hasAfterGET=getValuesForDataProperty("hasAfterGET",sEndpoint.getDataPropertyValues(o));
- // String
- // hasQueryParameter=getValuesForDataProperty("hasQueryParameter",sEndpoint.getDataPropertyValues(o));
- OWLIndividual[] objFilter = getIndividualsForProperty("hasObjectFilterSet", typedQuery
- .getObjectPropertyValues(o));
- OWLIndividual[] predFilter = getIndividualsForProperty("hasPredicateFilterSet", typedQuery
- .getObjectPropertyValues(o));
-
- Set<String> objectFilter = new HashSet<String>();
- Set<String> predicateFilter = new HashSet<String>();
-
- for (OWLIndividual of : objFilter) {
- String[] tmp = getValuesForDataProperty("filtersURI", of.getDataPropertyValues(o));
- for (String s : tmp) {
- objectFilter.add(s);
-
- }
- }
-
- for (OWLIndividual pf : predFilter) {
- String[] tmp = getValuesForDataProperty("filtersURI", pf.getDataPropertyValues(o));
- for (String s : tmp) {
- predicateFilter.add(s);
-
- }
- }
- // System.out.println(predicateFilter);
- // System.out.println(hasMode+objectFilter+predicateFilter+useLiterals);
- return new SparqlQueryType(hasMode, objectFilter, predicateFilter, useLiterals);
-
- }
-
- public static Manipulator makeManipulator() {
- return new Manipulator();
- }
-
public Manipulator getManipulator() {
- return this.Manipulator;
+ return this.manipulator;
}
public SpecificSparqlEndpoint getSparqlEndpoint() {
- return SparqlEndpoint;
+ return specificSparqlEndpoint;
}
public SparqlQueryType getSparqlQueryType() {
- return SparqlQueryType;
+ return sparqlQueryType;
}
-
public boolean isGetAllBackground() {
- return getAllBackground;
+ return getAllBackground;
}
-
public int getRecursiondepth() {
- return recursiondepth;
+ return recursiondepth;
}
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URI;
@@ -3,4 +22,5 @@
import java.util.Vector;
+// this class is used to extract the information recursively
public class ExtractionAlgorithm {
@@ -13,8 +33,8 @@
public ExtractionAlgorithm(Configuration Configuration) {
this.configuration = Configuration;
this.manipulator = Configuration.getManipulator();
- this.recursionDepth=Configuration.getRecursiondepth();
- this.getAllBackground=Configuration.isGetAllBackground();
+ this.recursionDepth = Configuration.getRecursiondepth();
+ this.getAllBackground = Configuration.isGetAllBackground();
}
@@ -43,13 +63,14 @@
while (v.size() > 0) {
Node tmpNode = v.remove(0);
System.out.println("Expanding " + tmpNode);
- //System.out.println(this.Manipulator);
+ // System.out.println(this.Manipulator);
Vector<Node> tmpVec = tmpNode.expand(tsp, manipulator);
tmp.addAll(tmpVec);
}
v = tmp;
- System.out.println("Recursion counter: " + x + " with " + v.size()+" Nodes remaining");
+ System.out
+ .println("Recursion counter: " + x + " with " + v.size() + " Nodes remaining");
}
if (this.getAllBackground) {
Vector<Node> classes = new Vector<Node>();
@@ -59,9 +80,9 @@
}
}
while (classes.size() > 0) {
- System.out.println("Remaining classes: "+classes.size());
- Node next=classes.remove(0);
- System.out.println("Expanding: "+next);
+ System.out.println("Remaining classes: " + classes.size());
+ Node next = classes.remove(0);
+ System.out.println("Expanding: " + next);
classes.addAll(next.expand(tsp, manipulator));
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URI;
@@ -6,6 +25,7 @@
import java.util.Set;
import java.util.Vector;
+// a node in the graph that is an instance
public class InstanceNode extends Node {
Set<ClassNode> classes = new HashSet<ClassNode>();
@@ -24,7 +44,7 @@
Set<Tupel> s = tsq.query(uri);
// Manipulation
m.check(s, this);
- //System.out.println("fffffff"+m);
+ // System.out.println("fffffff"+m);
Vector<Node> Nodes = new Vector<Node>();
Iterator<Tupel> it = s.iterator();
@@ -53,6 +73,14 @@
}
@Override
+ public Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m) {
+ for (PropertyNode one : properties) {
+ one.expandProperties(tsq, m);
+ }
+ return new Vector<Node>();
+ }
+
+ @Override
public Set<String> toNTriple() {
Set<String> s = new HashSet<String>();
s.add("<" + uri + "><" + "http://www.w3.org/1999/02/22-rdf-syntax-ns#type" + "><"
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URI;
@@ -4,21 +23,53 @@
import java.util.HashSet;
import java.util.Set;
+// an object of this class encapsulates everything
public class Manager {
private Configuration configuration;
private TypedSparqlQuery typedSparqlQuery;
private ExtractionAlgorithm extractionAlgorithm;
-
+ public void useConfiguration(SparqlQueryType SparqlQueryType,
+ SpecificSparqlEndpoint SparqlEndpoint, Manipulator manipulator, int recursiondepth,
+ boolean getAllBackground) {
- public void useConfiguration(SparqlQueryType SparqlQueryType, SpecificSparqlEndpoint SparqlEndpoint, int recursiondepth,boolean getAllBackground) {
-
- this.configuration = new Configuration(SparqlEndpoint, SparqlQueryType,recursiondepth,getAllBackground);
+ this.configuration = new Configuration(SparqlEndpoint, SparqlQueryType, manipulator,
+ recursiondepth, getAllBackground);
this.typedSparqlQuery = new TypedSparqlQuery(configuration);
this.extractionAlgorithm = new ExtractionAlgorithm(configuration);
}
+ public Set<String> getDomainInstancesForRole(String role) {
+ URI u = null;
+ try {
+ u = new URI(role);
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ Set<Tupel> t = this.typedSparqlQuery.getTupelsForRole(u);
+ Set<String> ret = new HashSet<String>();
+ for (Tupel one : t) {
+ ret.add(one.a);
+ }
+ return ret;
+ }
+
+ public Set<String> getRangeInstancesForRole(String role) {
+ URI u = null;
+ try {
+ u = new URI(role);
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ Set<Tupel> t = this.typedSparqlQuery.getTupelsForRole(u);
+ Set<String> ret = new HashSet<String>();
+ for (Tupel one : t) {
+ ret.add(one.b);
+ }
+ return ret;
+ }
+
public String extract(URI uri) {
// this.TypedSparqlQuery.query(uri);
// System.out.println(ExtractionAlgorithm.getFirstNode(uri));
@@ -54,4 +105,9 @@
return nt;
}
+ public void addPredicateFilter(String str) {
+ this.configuration.getSparqlQueryType().addPredicateFilter(str);
+
+ }
+
}
\ No newline at end of file
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.util.HashSet;
@@ -4,9 +23,11 @@
import java.util.Iterator;
import java.util.Set;
+// used to manipulate retrieved tupels, identify blanknodes, etc.
public class Manipulator {
public String subclass = "http://www.w3.org/2000/01/rdf-schema#subClassOf";
public String type = "http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
+ public String blankNodeIdentifier = "bnode";
String objectProperty = "http://www.w3.org/2002/07/owl#ObjectProperty";
String classns = "http://www.w3.org/2002/07/owl#Class";
@@ -16,13 +37,10 @@
String[] defaultClasses = { "http://dbpedia.org/class/yago",
"http://dbpedia.org/resource/Category:", "http://dbpedia.org/resource/Template:",
- "http://www.w3.org/2004/02/skos/core", "http://dbpedia.org/class/" }; // TODO
- // FEHLER
- // hier
- // fehlt
- // yago
+ "http://www.w3.org/2004/02/skos/core", "http://dbpedia.org/class/" };
- public Manipulator() {
+ public Manipulator(String blankNodeIdentifier) {
+ this.blankNodeIdentifier = blankNodeIdentifier;
Set<String> classproperties = new HashSet<String>();
classproperties.add(subclass);
@@ -33,21 +51,22 @@
Iterator<Tupel> it = s.iterator();
while (it.hasNext()) {
Tupel t = (Tupel) it.next();
+ // System.out.println(t);
// all classes with owl:type class
if (t.a.equals(type) && t.b.equals(classns) && node instanceof ClassNode) {
toRemove.add(t);
}
-
+
// all with type class
if (t.b.equals(classns) && node instanceof ClassNode) {
toRemove.add(t);
}
-
+
// all instances with owl:type thing
if (t.a.equals(type) && t.b.equals(thing) && node instanceof InstanceNode) {
toRemove.add(t);
}
-
+
}
s.removeAll(toRemove);
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URI;
@@ -4,29 +23,20 @@
import java.util.Set;
import java.util.Vector;
+// abstract class
public abstract class Node {
URI uri;
protected String type;
protected boolean expanded = false;
- // Hashtable<String,Node> classes=new Hashtable<String,Node>();
- // Hashtable<String,Node> instances=new Hashtable<String,Node>();;
- // Hashtable<String,Node> datatype=new Hashtable<String,Node>();;
-
public Node(URI u) {
this.uri = u;
-
}
- /*
- * public void checkConsistency(){ if (type.equals("class") && (
- * instances.size()>0 || datatype.size()>0)){ System.out.println("Warning,
- * inconsistent:"+this.toString()); }
- * }
- */
-
public abstract Vector<Node> expand(TypedSparqlQuery tsq, Manipulator m);
+ public abstract Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m);
+
public abstract Set<String> toNTriple();
@Override
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URL;
@@ -3,26 +22,41 @@
import java.util.HashMap;
+// holds some predefined endpoints
public class PredefinedEndpoint {
- public static SpecificSparqlEndpoint getEndpoint(int i) {
+ public static SpecificSparqlEndpoint getEndpoint(int i) {
- switch (i) {
- case 1:
- return dbpediaEndpoint();
+ switch (i) {
+ case 1:
+ return dbpediaEndpoint();
+ case 2:
+ return localJoseki();
+ }
+ return null;
+ }
+ public static SpecificSparqlEndpoint dbpediaEndpoint() {
+ URL u = null;
+ HashMap<String, String> m = new HashMap<String, String>();
+ m.put("default-graph-uri", "http://dbpedia.org");
+ m.put("format", "application/sparql-results.xml");
+ try {
+ u = new URL("http://dbpedia.openlinksw.com:8890/sparql");
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ return new SpecificSparqlEndpoint(u, "dbpedia.openlinksw.com", m);
}
- return null;
- }
-
- public static SpecificSparqlEndpoint dbpediaEndpoint(){
- URL u = null;
- HashMap<String, String> m = new HashMap<String, String>();
- m.put("default-graph-uri", "http://dbpedia.org");
- m.put("format", "application/sparql-results.xml");
- try {
- u = new URL("http://dbpedia.openlinksw.com:8890/sparql");
- } catch (Exception e) {
- e.printStackTrace();
+
+ public static SpecificSparqlEndpoint localJoseki() {
+ URL u = null;
+ HashMap<String, String> m = new HashMap<String, String>();
+ // m.put("default-graph-uri", "http://dbpedia.org");
+ // m.put("format", "application/sparql-results.xml");
+ try {
+ u = new URL("http://localhost:2020/books");
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ return new SpecificSparqlEndpoint(u, "localost", m);
}
- return new SpecificSparqlEndpoint(u, "dbpedia.openlinksw.com", m);
}
-}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.util.HashSet;
@@ -4,14 +23,15 @@
import java.util.Set;
public class PredefinedFilter {
-
+
public static SparqlQueryType getFilter(int i) {
switch (i) {
case 1:
return YagoFilter();
-
+ case 2:
+ return CategoriesOnly();
}
return null;
}
@@ -22,6 +42,7 @@
pred.add("http://www.w3.org/2004/02/skos/core");
pred.add("http://www.w3.org/2002/07/owl#sameAs");
pred.add("http://xmlns.com/foaf/0.1/");
+
pred.add("http://dbpedia.org/property/reference");
pred.add("http://dbpedia.org/property/website");
pred.add("http://dbpedia.org/property/wikipage");
@@ -39,4 +60,36 @@
return new SparqlQueryType("forbid", obj, pred, "false");
}
+
+ public static SparqlQueryType CategoriesOnly(){
+ Set<String> pred = new HashSet<String>();
+ pred.add("http://www.w3.org/2004/02/skos/core");
+ pred.add("http://www.w3.org/2002/07/owl#sameAs");
+ pred.add("http://xmlns.com/foaf/0.1/");
+
+ pred.add("http://dbpedia.org/property/reference");
+ pred.add("http://dbpedia.org/property/website");
+ pred.add("http://dbpedia.org/property/wikipage");
+
+ Set<String> obj = new HashSet<String>();
+ obj.add("http://dbpedia.org/resource/Category:Wikipedia_");
+ obj.add("http://dbpedia.org/resource/Category:Articles_");
+ obj.add("http://xmlns.com/foaf/0.1/");
+ obj.add("http://upload.wikimedia.org/wikipedia/commons");
+ obj.add("http://upload.wikimedia.org/wikipedia");
+
+ obj.add("http://www.geonames.org");
+ obj.add("http://www.w3.org/2006/03/wn/wn20/instances/synset");
+ obj.add("http://www4.wiwiss.fu-berlin.de/flickrwrappr");
+ obj.add("http://www.w3.org/2004/02/skos/core");
+
+ obj.add("http://dbpedia.org/class/yago");
+ obj.add("http://dbpedia.org/resource/Template");
+
+
+ return new SparqlQueryType("forbid", obj, pred, "false");
+ }
+
+
+
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.net.URI;
@@ -47,6 +66,12 @@
}
return Nodes;
}
+ @Override
+ public Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m) {
+ b.expandProperties(tsq, m);
+ return this.expand(tsq, m);
+ }
+
public Node getB() {
return b;
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,5 +1,25 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
+// can assemble sparql queries
public class SparqlQueryMaker {
private SparqlQueryType sparqlQueryType;
@@ -8,9 +28,27 @@
this.sparqlQueryType = SparqlQueryType;
}
- public String makeQueryUsingFilters(String subject) {
+ public String makeSubjectQueryUsingFilters(String subject) {
String lineend = "\n";
+ String Filter = internalFilterAssemblySubject();
+ String ret = "SELECT * WHERE { " + lineend + "<" + subject + "> ?predicate ?object. "
+ + lineend + "FILTER( " + lineend + "(" + Filter + ").}";
+ // System.out.println(ret);
+ return ret;
+ }
+ public String makeRoleQueryUsingFilters(String role) {
+ String lineend = "\n";
+ String Filter = internalFilterAssemblyRole();
+ String ret = "SELECT * WHERE { " + lineend + "?subject <" + role + "> ?object. " + lineend
+ + "FILTER( " + lineend + "(" + Filter + ").}";
+ // System.out.println(ret);
+
+ return ret;
+ }
+
+ private String internalFilterAssemblySubject() {
+ String lineend = "\n";
String Filter = "";
if (!this.sparqlQueryType.isLiterals())
Filter += "!isLiteral(?object))";
@@ -20,18 +58,32 @@
for (String o : sparqlQueryType.getObjectfilterlist()) {
Filter += lineend + filterObject(o);
}
+ return Filter;
+ }
- String ret = "SELECT * WHERE { " + lineend + "<" + subject + "> ?predicate ?object. "
- + lineend + "FILTER( " + lineend + "(" + Filter + ").}";
- // System.out.println(ret);
- return ret;
+ private String internalFilterAssemblyRole() {
+ String lineend = "\n";
+ String Filter = "";
+ if (!this.sparqlQueryType.isLiterals())
+ Filter += "!isLiteral(?object))";
+ for (String s : sparqlQueryType.getObjectfilterlist()) {
+ Filter += lineend + filterSubject(s);
+ }
+ for (String o : sparqlQueryType.getObjectfilterlist()) {
+ Filter += lineend + filterObject(o);
+ }
+ return Filter;
}
- public String filterObject(String ns) {
- return "&&( !regex(str(?object), '" + ns + "') )";
+ public String filterSubject(String ns) {
+ return "&&( !regex(str(?subject), '" + ns + "') )";
}
public String filterPredicate(String ns) {
return "&&( !regex(str(?predicate), '" + ns + "') )";
}
+
+ public String filterObject(String ns) {
+ return "&&( !regex(str(?object), '" + ns + "') )";
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java 2007-12-02 20:52:38 UTC (rev 309)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java 2007-12-03 14:45:11 UTC (rev 310)
@@ -1,3 +1,22 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
package org.dllearner.kb.sparql;
import java.util.Set;
@@ -2,4 +21,5 @@
+// is used to set the filter: configuration
public class SparqlQueryType {
-
+ // TODO make sets out of them
private String mode = "forbid";
@@ -58,4 +78,15 @@
return predicatefilterlist;
}
+ public void addPredicateFilter(String filter) {
+ String[] tmp = new String[predicatefilterlist.length + 1];
+ int i = 0;
+ for (; i < predicatefilterlist.length; i++) {
+ tmp[i] = predicatefilterlist[i];
+
+ }
+ tmp[i] = filter;
+
+ }
+
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SpecificSparqlEndpoint.java...
[truncated message content] |
|
From: <ku...@us...> - 2007-12-05 12:44:59
|
Revision: 325
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=325&view=rev
Author: kurzum
Date: 2007-12-05 04:44:53 -0800 (Wed, 05 Dec 2007)
Log Message:
-----------
added option breakSuperClassRetrievalAfter needed for SKOS
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-05 12:24:13 UTC (rev 324)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-05 12:44:53 UTC (rev 325)
@@ -80,6 +80,7 @@
private boolean dumpToFile = true;
private boolean useLits = false;
private boolean getAllSuperClasses = true;
+ private int breakSuperClassRetrievalAfter = 500;
private boolean learnDomain = false;
private String role;
@@ -166,8 +167,9 @@
options.add(new StringTupleListConfigOption("example", "example"));
options.add(new StringTupleListConfigOption("replacePredicate", "rule for replacing predicates"));
options.add(new StringTupleListConfigOption("replaceObject", "rule for replacing predicates"));
+ options.add(new IntegerConfigOption("breakSuperClassRetrievalAfter", "stops a cyclic hierarchy after specified number of classes"));
+
-
return options;
}
@@ -222,9 +224,10 @@
replacePredicate = (LinkedList)entry.getValue();
}else if (option.equals("replaceObject")) {
replaceObject = (LinkedList)entry.getValue();
-
+ }else if (option.equals("breakSuperClassRetrievalAfter")) {
+ breakSuperClassRetrievalAfter = (Integer) entry.getValue();
}
-
+
}
/*
@@ -244,7 +247,7 @@
SpecificSparqlEndpoint sse = null;
SparqlQueryType sqt = null;
// get Options for Manipulator
- Manipulator man = new Manipulator(blankNodeIdentifier,replacePredicate,replaceObject);
+ Manipulator man = new Manipulator(blankNodeIdentifier,breakSuperClassRetrievalAfter,replacePredicate,replaceObject);
HashMap<String, String> parameters = new HashMap<String, String>();
parameters.put("default-graph-uri", "http://dbpedia.org");
parameters.put("format", "application/sparql-results.xml");
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-05 12:24:13 UTC (rev 324)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-05 12:44:53 UTC (rev 325)
@@ -94,7 +94,7 @@
Node next = classes.remove(0);
System.out.println("Expanding: " + next);
classes.addAll(next.expand(tsp, manipulator));
- if (classes.size()>=500){break;}
+ if (classes.size()>=manipulator.breakSuperClassRetrievalAfter){break;}
}
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java 2007-12-05 12:24:13 UTC (rev 324)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java 2007-12-05 12:44:53 UTC (rev 325)
@@ -31,6 +31,7 @@
public String subclass = "http://www.w3.org/2000/01/rdf-schema#subClassOf";
public String type = "http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
public String blankNodeIdentifier = "bnode";
+ public int breakSuperClassRetrievalAfter=200;
public LinkedList<StringTuple> replacePredicate;
public LinkedList<StringTuple> replaceObject;
@@ -44,7 +45,7 @@
"http://dbpedia.org/resource/Category:", "http://dbpedia.org/resource/Template:",
"http://www.w3.org/2004/02/skos/core", "http://dbpedia.org/class/" };
- public Manipulator(String blankNodeIdentifier,LinkedList<StringTuple> replacePredicate,LinkedList<StringTuple> replaceObject) {
+ public Manipulator(String blankNodeIdentifier,int breakSuperClassRetrievalAfter,LinkedList<StringTuple> replacePredicate,LinkedList<StringTuple> replaceObject) {
this.blankNodeIdentifier = blankNodeIdentifier;
this.replaceObject=replaceObject;
this.replacePredicate=replacePredicate;
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <ku...@us...> - 2007-12-08 18:46:58
|
Revision: 331
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=331&view=rev
Author: kurzum
Date: 2007-12-08 10:46:56 -0800 (Sat, 08 Dec 2007)
Log Message:
-----------
some code changes mainly needed for role learning
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java
trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-05 16:18:15 UTC (rev 330)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-08 18:46:56 UTC (rev 331)
@@ -84,7 +84,8 @@
private boolean learnDomain = false;
private boolean learnRange = false;
- private String role;
+ private int numberOfInstancesUsedForRoleLearning=40;
+ private String role="";
private String blankNodeIdentifier = "bnode";
LinkedList<StringTuple> URIParameters = new LinkedList<StringTuple>();
@@ -170,8 +171,10 @@
options.add(new StringTupleListConfigOption("replacePredicate", "rule for replacing predicates"));
options.add(new StringTupleListConfigOption("replaceObject", "rule for replacing predicates"));
options.add(new IntegerConfigOption("breakSuperClassRetrievalAfter", "stops a cyclic hierarchy after specified number of classes"));
+ options.add(new IntegerConfigOption("numberOfInstancesUsedForRoleLearning", ""));
+
return options;
}
@@ -230,6 +233,8 @@
replaceObject = (LinkedList)entry.getValue();
}else if (option.equals("breakSuperClassRetrievalAfter")) {
breakSuperClassRetrievalAfter = (Integer) entry.getValue();
+ }else if (option.equals("numberOfInstancesUsedForRoleLearning")) {
+ numberOfInstancesUsedForRoleLearning = (Integer) entry.getValue();
}
}
@@ -279,12 +284,12 @@
// used to learn a domain of a role
if (learnDomain || learnRange) {
Set<String> pos=new HashSet<String>();
- Set<String> neg=new HashSet<String>();
+ //Set<String> neg=new HashSet<String>();
if(learnDomain){
pos = m.getDomainInstancesForRole(role);
- neg = m.getRangeInstancesForRole(role);
+ //neg = m.getRangeInstancesForRole(role);
}else if(learnRange){
- neg = m.getDomainInstancesForRole(role);
+ //neg = m.getDomainInstancesForRole(role);
pos = m.getRangeInstancesForRole(role);
}
//choose 30
@@ -293,29 +298,29 @@
Set<String> tmp=new HashSet<String>();
for(String one:pos){
tmp.add(one);
- if(tmp.size()>=5)break;
+ if(tmp.size()>=numberOfInstancesUsedForRoleLearning)break;
}
pos=tmp;
- System.out.println(pos.size());
+ System.out.println("Instances used: "+pos.size());
- tmp=new HashSet<String>();
+ /*tmp=new HashSet<String>();
for(String one:neg){
tmp.add(one);
if(tmp.size()>=5)break;
}
- neg=tmp;
+ neg=tmp;*/
instances=new HashSet<String>();
- instances.addAll( pos);
+ instances.addAll(pos);
- instances.addAll(neg);
+ //instances.addAll(neg);
for(String one:pos){
System.out.println("+\""+one+"\"");
}
- for(String one:neg){
+ /*for(String one:neg){
System.out.println("-\""+one+"\"");
- }
+ }*/
/*Random r= new Random();
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-05 16:18:15 UTC (rev 330)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-08 18:46:56 UTC (rev 331)
@@ -52,7 +52,7 @@
} catch (Exception e) {
e.printStackTrace();
}
- Set<StringTuple> t = this.typedSparqlQuery.getTupelsForRole(u);
+ Set<StringTuple> t = this.typedSparqlQuery.getTupelsForRole(u, true);
Set<String> ret = new HashSet<String>();
for (StringTuple one : t) {
@@ -68,7 +68,7 @@
} catch (Exception e) {
e.printStackTrace();
}
- Set<StringTuple> t = this.typedSparqlQuery.getTupelsForRole(u);
+ Set<StringTuple> t = this.typedSparqlQuery.getTupelsForRole(u,false);
Set<String> ret = new HashSet<String>();
for (StringTuple one : t) {
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java 2007-12-05 16:18:15 UTC (rev 330)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedEndpoint.java 2007-12-08 18:46:56 UTC (rev 331)
@@ -32,13 +32,13 @@
case 2:
return localJoseki();
case 3:
- return worldFactBook();
- case 4:
return govTrack();
+ case 4:
+ return revyu();
case 5:
- return revyu();
- case 6:
return myopenlink();
+ case 6:
+ return worldFactBook();
}
return null;
}
@@ -66,7 +66,7 @@
} catch (Exception e) {
e.printStackTrace();
}
- return new SpecificSparqlEndpoint(u, "localost", m);
+ return new SpecificSparqlEndpoint(u, "localhost", m);
}
public static SpecificSparqlEndpoint worldFactBook() {
URL u = null;
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java 2007-12-05 16:18:15 UTC (rev 330)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java 2007-12-08 18:46:56 UTC (rev 331)
@@ -21,7 +21,8 @@
// can assemble sparql queries
public class SparqlQueryMaker {
-
+ String lineend="\n";
+ boolean print_flag=false;
/* can make queries for subject, predicate, object
* according to the filter settings
* object not yet implemented
@@ -35,26 +36,47 @@
}
public String makeSubjectQueryUsingFilters(String subject) {
- String lineend = "\n";
+
String Filter = internalFilterAssemblySubject();
String ret = "SELECT * WHERE { " + lineend + "<" + subject + "> ?predicate ?object. "
+ lineend + "FILTER( " + lineend + "(" + Filter + ").}";
// System.out.println(ret);
+ //System.out.println(sparqlQueryType.getPredicatefilterlist().length);
return ret;
}
public String makeRoleQueryUsingFilters(String role) {
- String lineend = "\n";
+
String Filter = internalFilterAssemblyRole();
- String ret = "SELECT * WHERE { " + lineend + "?subject <" + role + "> ?object. " + lineend
+ String ret = "SELECT * WHERE { " + lineend + " ?subject <" + role + "> ?object. " + lineend
+ "FILTER( " + lineend + "(" + Filter + ").}";
// System.out.println(ret);
return ret;
}
+ public String makeRoleQueryUsingFilters(String role,boolean domain) {
+
+ String Filter = internalFilterAssemblyRole();
+ String ret="";
+ if(domain){
+ ret = "SELECT * WHERE { " + lineend +
+ "?subject <" + role + "> ?object; a []. " + lineend
+ + "FILTER( " + lineend + "(" + Filter + ").}";
+ // System.out.println(ret);
+ }else{
+ ret = "SELECT * WHERE { " + lineend +
+ "?object a [] . " +
+ "?subject <" + role + "> ?object . " + lineend
+ + "FILTER( " + lineend + "(" + Filter + ").}";
+
+ }
+ //System.out.println(ret);
+ return ret;
+ }
+
private String internalFilterAssemblySubject() {
- String lineend = "\n";
+
String Filter = "";
if (!this.sparqlQueryType.isLiterals())
Filter += "!isLiteral(?object))";
@@ -68,7 +90,7 @@
}
private String internalFilterAssemblyRole() {
- String lineend = "\n";
+
String Filter = "";
if (!this.sparqlQueryType.isLiterals())
Filter += "!isLiteral(?object))";
@@ -92,4 +114,10 @@
public String filterObject(String ns) {
return "&&( !regex(str(?object), '" + ns + "') )";
}
+
+ public void p(String str){
+ if(print_flag){
+ System.out.println(str);
+ }
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java 2007-12-05 16:18:15 UTC (rev 330)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryType.java 2007-12-08 18:46:56 UTC (rev 331)
@@ -85,9 +85,11 @@
int i = 0;
for (; i < predicatefilterlist.length; i++) {
tmp[i] = predicatefilterlist[i];
-
+ //System.out.println(tmp[i]);
}
tmp[i] = filter;
+ predicatefilterlist=tmp;
+ //System.out.println("added filter: "+filter);
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java 2007-12-05 16:18:15 UTC (rev 330)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java 2007-12-08 18:46:56 UTC (rev 331)
@@ -36,6 +36,8 @@
// can execute different queries
public class TypedSparqlQuery {
+ boolean print_flag=false;
+ boolean debug_no_cache=false;
private Configuration configuration;
// private SparqlHTTPRequest SparqlHTTPRequest;
private SparqlQueryMaker sparqlQueryMaker;
@@ -68,13 +70,25 @@
return s;
}
+ public Set<StringTuple> getTupelsForRole(URI u,boolean domain) {
+ // getQuery
+ String sparql = sparqlQueryMaker.makeRoleQueryUsingFilters(u.toString(),domain);
+
+ Set<StringTuple> s = cachedSparql(u, sparql, "subject", "object");
+ // System.out.println(s);
+ return s;
+
+ }
+
// uses a cache
private Set<StringTuple> cachedSparql(URI u, String sparql, String a, String b) {
// check cache
String FromCache = cache.get(u.toString(), sparql);
-
+ if(debug_no_cache) {
+ FromCache=null;
+ }
String xml = null;
// if not in cache get it from EndPoint
if (FromCache == null) {
@@ -84,9 +98,11 @@
// TODO Auto-generated catch block
e.printStackTrace();
}
- // System.out.println(sparql);
+ p(sparql);
// System.out.println(xml);
- cache.put(u.toString(), sparql, xml);
+ if(!debug_no_cache) {
+ cache.put(u.toString(), sparql, xml);
+ }
//System.out.print("\n");
} else {
xml = FromCache;
@@ -134,16 +150,20 @@
}
private String sendAndReceiveSPARQL(String sparql) throws IOException {
+ p("sendAndReceiveSPARQL");
StringBuilder answer = new StringBuilder();
+ //sparql="SELECT * WHERE {?a ?b ?c}LIMIT 10";
// String an Sparql-Endpoint schicken
HttpURLConnection connection;
SpecificSparqlEndpoint se = configuration.getSparqlEndpoint();
-
+ p("URL: "+se.getURL());
+ p("Host: "+se.getHost());
+
connection = (HttpURLConnection) se.getURL().openConnection();
connection.setDoOutput(true);
- connection.addRequestProperty("Host", se.getHost());
+ //connection.addRequestProperty("Host", se.getHost());
connection.addRequestProperty("Connection", "close");
connection
.addRequestProperty(
@@ -155,7 +175,7 @@
.addRequestProperty(
"User-Agent",
"Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
-
+
OutputStream os = connection.getOutputStream();
OutputStreamWriter osw = new OutputStreamWriter(os);
@@ -167,9 +187,9 @@
FullURI += "" + URLEncoder.encode(element, "UTF-8") + "="
+ URLEncoder.encode(se.getParameters().get(element), "UTF-8") + "&";
}
- // System.out.println(FullURI);
+
FullURI += "" + se.getHasQueryParameter() + "=" + URLEncoder.encode(sparql, "UTF-8");
-
+ p(FullURI);
osw.write(FullURI);
osw.close();
@@ -186,8 +206,13 @@
} while (line != null);
br.close();
-
+ p(answer.toString());
return answer.toString();
}
+ public void p(String str){
+ if(print_flag){
+ System.out.println(str);
+ }
+ }
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <ku...@us...> - 2007-12-14 20:16:54
|
Revision: 335
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=335&view=rev
Author: kurzum
Date: 2007-12-14 12:16:47 -0800 (Fri, 14 Dec 2007)
Log Message:
-----------
a lot of changes
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java
trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
Added Paths:
-----------
trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryClasses.java
trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryInterface.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -80,7 +80,7 @@
private boolean dumpToFile = true;
private boolean useLits = false;
private boolean getAllSuperClasses = true;
- private int breakSuperClassRetrievalAfter = 500;
+ private int breakSuperClassRetrievalAfter = 200;
private boolean learnDomain = false;
private boolean learnRange = false;
@@ -281,15 +281,16 @@
m.useConfiguration(sqt, sse, man, recursionDepth, getAllSuperClasses);
try {
String ont = "";
+ System.out.println(learnDomain);
// used to learn a domain of a role
if (learnDomain || learnRange) {
Set<String> pos=new HashSet<String>();
- //Set<String> neg=new HashSet<String>();
+ Set<String> neg=new HashSet<String>();
if(learnDomain){
pos = m.getDomainInstancesForRole(role);
- //neg = m.getRangeInstancesForRole(role);
+ neg = m.getRangeInstancesForRole(role);
}else if(learnRange){
- //neg = m.getDomainInstancesForRole(role);
+ neg = m.getDomainInstancesForRole(role);
pos = m.getRangeInstancesForRole(role);
}
//choose 30
@@ -303,24 +304,24 @@
pos=tmp;
System.out.println("Instances used: "+pos.size());
- /*tmp=new HashSet<String>();
+ tmp=new HashSet<String>();
for(String one:neg){
tmp.add(one);
- if(tmp.size()>=5)break;
+ if(tmp.size()>=numberOfInstancesUsedForRoleLearning)break;
}
- neg=tmp;*/
+ neg=tmp;
instances=new HashSet<String>();
instances.addAll(pos);
- //instances.addAll(neg);
+ instances.addAll(neg);
for(String one:pos){
System.out.println("+\""+one+"\"");
}
- /*for(String one:neg){
+ for(String one:neg){
System.out.println("-\""+one+"\"");
- }*/
+ }
/*Random r= new Random();
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Cache.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -32,6 +32,7 @@
// Object can be the cache itself
// or a cache object(one entry)
// it now uses a hashmap and can contain different queries at once
+ private HashMap<String, String> hm;
final static long serialVersionUID = 104;
transient String basedir = "";
@@ -39,7 +40,7 @@
long timestamp;
long daysoffreshness = 15;
long multiplier = 24 * 60 * 60 * 1000;// h m s ms
- private HashMap<String, String> hm;
+ private HashMap<String, String> inmem_cache;
// constructor for the cache itself
public Cache(String path) {
@@ -154,4 +155,5 @@
return content;
}
+
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -40,7 +40,7 @@
//expands all directly connected nodes
@Override
- public Vector<Node> expand(TypedSparqlQuery tsq, Manipulator m) {
+ public Vector<Node> expand(TypedSparqlQueryInterface tsq, Manipulator m) {
Set<StringTuple> s = tsq.query(this.uri);
// see manipulator
s = m.check(s, this);
@@ -83,7 +83,7 @@
// gets the types for properties recursively
@Override
- public Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m) {
+ public Vector<Node> expandProperties(TypedSparqlQueryInterface tsq, Manipulator m) {
// TODO return type doesn't make sense
return new Vector<Node>();
}
@@ -101,5 +101,16 @@
return s;
}
+
+ @Override
+ public int compareTo(Object n){
+ return super.compareTo((Node)n);
+ //
+ }
+ @Override
+ public int compareTo(Node n){
+ return super.compareTo(n);
+ //
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -20,6 +20,10 @@
package org.dllearner.kb.sparql;
import java.net.URI;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+import java.util.TreeSet;
import java.util.Vector;
// this class is used to extract the information recursively
@@ -29,6 +33,8 @@
private Manipulator manipulator;
private int recursionDepth = 2;
private boolean getAllBackground = true;
+ private boolean closeAfterRecursion = true;
+ private boolean print_flag=false;
public ExtractionAlgorithm(Configuration Configuration) {
this.configuration = Configuration;
@@ -42,7 +48,7 @@
return new InstanceNode(u);
}
- public Vector<Node> expandAll(URI[] u, TypedSparqlQuery tsp) {
+ public Vector<Node> expandAll(URI[] u, TypedSparqlQueryInterface tsp) {
Vector<Node> v = new Vector<Node>();
for (URI one : u) {
v.add(expandNode(one, tsp));
@@ -55,11 +61,12 @@
cave: the recursion is not a recursion anymore,
it was transformed to an iteration
*/
- public Node expandNode(URI u, TypedSparqlQuery tsp) {
+ public Node expandNode(URI u, TypedSparqlQueryInterface tsp) {
+ long time=System.currentTimeMillis();
Node n = getFirstNode(u);
Vector<Node> v = new Vector<Node>();
v.add(n);
- System.out.println("StartVector: " + v);
+ p("StartVector: " + v);
// n.expand(tsp, this.Manipulator);
// Vector<Node> second=
for (int x = 1; x <= recursionDepth; x++) {
@@ -67,7 +74,7 @@
Vector<Node> tmp = new Vector<Node>();
while (v.size() > 0) {
Node tmpNode = v.remove(0);
- System.out.println("Expanding " + tmpNode);
+ p("Expanding " + tmpNode);
// System.out.println(this.Manipulator);
// these are the new not expanded nodes
@@ -77,29 +84,74 @@
tmp.addAll(tmpVec);
}
v = tmp;
- System.out
- .println("Recursion counter: " + x + " with " + v.size() + " Nodes remaining");
+ System.out.println("Recursion counter: " + x +
+ " with " + v.size() + " Nodes remaining, needed: "
+ +(System.currentTimeMillis()-time));
+ time=System.currentTimeMillis();
}
+ HashSet<String> hadAlready=new HashSet<String>();
// gets All Class Nodes and expands them further
if (this.getAllBackground) {
+ //Set<Node> classes = new TreeSet<Node>();
Vector<Node> classes = new Vector<Node>();
+
+ Vector<Node> instances = new Vector<Node>();
for (Node one : v) {
if (one instanceof ClassNode) {
classes.add(one);
}
+ if (one instanceof InstanceNode) {
+ instances.add(one);
+ }
+
}
+ System.out.println(instances.size());
+ TypedSparqlQueryClasses tsqc=new TypedSparqlQueryClasses(configuration);
+ if(closeAfterRecursion){
+ while (instances.size() > 0) {
+ p("Getting classes for remaining instances: " + instances.size());
+ Node next = instances.remove(0);
+ p("Getting classes for: " + next);
+ classes.addAll(next.expand(tsqc, manipulator));
+ if (classes.size()>=manipulator.breakSuperClassRetrievalAfter){break;}
+ }
+ }
+ Vector<Node>tmp=new Vector<Node>();
+ int i=0;
while (classes.size() > 0) {
- System.out.println("Remaining classes: " + classes.size());
+ p("Remaining classes: " + classes.size());
+ //Iterator<Node> it=classes.iterator();
+ //Node next =(Node) it.next();
+ //classes.remove(next);
Node next = classes.remove(0);
- System.out.println("Expanding: " + next);
- classes.addAll(next.expand(tsp, manipulator));
- if (classes.size()>=manipulator.breakSuperClassRetrievalAfter){break;}
+
+ if(!hadAlready.contains(next.uri.toString())){
+ p("Expanding: " + next);
+ //System.out.println(hadAlready.size());
+ hadAlready.add(next.uri.toString());
+ tmp=next.expand(tsp, manipulator);
+ classes.addAll(tmp);
+ tmp=new Vector<Node>();
+ if(i % 50==0)System.out.println("got "+i+" extra classes, max: "+manipulator.breakSuperClassRetrievalAfter);
+ i++;
+ if (i>=manipulator.breakSuperClassRetrievalAfter){break;}
+ }
+ //System.out.println("Skipping");
+
+
+ //if (classes.size()>=manipulator.breakSuperClassRetrievalAfter){break;}
+
}
+ //System.out.println((System.currentTimeMillis()-time)+"");
}
return n;
}
+
+ void p(String s){
+ if(print_flag)System.out.println(s);
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -44,7 +44,7 @@
//expands all directly connected nodes
@Override
- public Vector<Node> expand(TypedSparqlQuery tsq, Manipulator m) {
+ public Vector<Node> expand(TypedSparqlQueryInterface tsq, Manipulator m) {
Set<StringTuple> s = tsq.query(uri);
// see Manipulator
@@ -80,7 +80,7 @@
// gets the types for properties recursively
@Override
- public Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m) {
+ public Vector<Node> expandProperties(TypedSparqlQueryInterface tsq, Manipulator m) {
for (PropertyNode one : properties) {
one.expandProperties(tsq, m);
}
@@ -105,5 +105,16 @@
return s;
}
+
+ @Override
+ public int compareTo(Object n){
+ return 0;
+ //
+ }
+ @Override
+ public int compareTo(Node n){
+ return super.compareTo(n);
+ //
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manager.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -52,7 +52,7 @@
} catch (Exception e) {
e.printStackTrace();
}
- Set<StringTuple> t = this.typedSparqlQuery.getTupelsForRole(u, true);
+ Set<StringTuple> t = ((TypedSparqlQuery)this.typedSparqlQuery).getTupelsForRole(u, true);
Set<String> ret = new HashSet<String>();
for (StringTuple one : t) {
@@ -68,7 +68,7 @@
} catch (Exception e) {
e.printStackTrace();
}
- Set<StringTuple> t = this.typedSparqlQuery.getTupelsForRole(u,false);
+ Set<StringTuple> t = ((TypedSparqlQuery)this.typedSparqlQuery).getTupelsForRole(u,false);
Set<String> ret = new HashSet<String>();
for (StringTuple one : t) {
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Manipulator.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -49,7 +49,7 @@
this.blankNodeIdentifier = blankNodeIdentifier;
this.replaceObject=replaceObject;
this.replacePredicate=replacePredicate;
-
+ this.breakSuperClassRetrievalAfter=breakSuperClassRetrievalAfter;
Set<String> classproperties = new HashSet<String>();
classproperties.add(subclass);
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -24,7 +24,7 @@
import java.util.Vector;
// abstract class
-public abstract class Node {
+public abstract class Node implements Comparable{
URI uri;
protected String type;
protected boolean expanded = false;
@@ -33,9 +33,9 @@
this.uri = u;
}
- public abstract Vector<Node> expand(TypedSparqlQuery tsq, Manipulator m);
+ public abstract Vector<Node> expand(TypedSparqlQueryInterface tsq, Manipulator m);
- public abstract Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m);
+ public abstract Vector<Node> expandProperties(TypedSparqlQueryInterface tsq, Manipulator m);
public abstract Set<String> toNTriple();
@@ -48,5 +48,12 @@
public URI getURI() {
return uri;
}
+ public boolean equals(Node n){
+ if(this.uri.equals(n.uri))return true;
+ else return false;
+ }
+ public int compareTo(Node n){
+ return this.uri.toString().compareTo(n.uri.toString());
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PredefinedFilter.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -32,6 +32,10 @@
return YagoFilter();
case 2:
return SKOS();
+ case 3:
+ return YAGOSKOS();
+ case 4:
+ return YagoSpecialHierarchy();
}
return null;
}
@@ -47,11 +51,12 @@
pred.add("http://dbpedia.org/property/website");
pred.add("http://dbpedia.org/property/wikipage");
pred.add("http://dbpedia.org/property/wikiPageUsesTemplate");
+ pred.add("http://dbpedia.org/property/relatedInstance");
-
Set<String> obj = new HashSet<String>();
- obj.add("http://dbpedia.org/resource/Category:Wikipedia_");
- obj.add("http://dbpedia.org/resource/Category:Articles_");
+ //obj.add("http://dbpedia.org/resource/Category:Wikipedia_");
+ //obj.add("http://dbpedia.org/resource/Category:Articles_");
+ obj.add("http://dbpedia.org/resource/Category:");
obj.add("http://dbpedia.org/resource/Template");
obj.add("http://xmlns.com/foaf/0.1/");
obj.add("http://upload.wikimedia.org/wikipedia/commons");
@@ -63,7 +68,36 @@
return new SparqlQueryType("forbid", obj, pred, "false");
}
+ public static SparqlQueryType YagoSpecialHierarchy(){
+ Set<String> pred = new HashSet<String>();
+ pred.add("http://www.w3.org/2004/02/skos/core");
+ pred.add("http://www.w3.org/2002/07/owl#sameAs");
+ pred.add("http://xmlns.com/foaf/0.1/");
+
+ pred.add("http://dbpedia.org/property/reference");
+ pred.add("http://dbpedia.org/property/website");
+ pred.add("http://dbpedia.org/property/wikipage");
+ pred.add("http://dbpedia.org/property/wikiPageUsesTemplate");
+ pred.add("http://dbpedia.org/property/relatedInstance");
+ pred.add("http://dbpedia.org/property/monarch");
+
+
+ Set<String> obj = new HashSet<String>();
+ obj.add("http://dbpedia.org/resource/Category:Wikipedia_");
+ obj.add("http://dbpedia.org/resource/Category:Articles_");
+ obj.add("http://dbpedia.org/resource/Template");
+ obj.add("http://xmlns.com/foaf/0.1/");
+ obj.add("http://upload.wikimedia.org/wikipedia/commons");
+ obj.add("http://upload.wikimedia.org/wikipedia");
+ obj.add("http://www.geonames.org");
+ obj.add("http://www.w3.org/2006/03/wn/wn20/instances/synset");
+ obj.add("http://www4.wiwiss.fu-berlin.de/flickrwrappr");
+ obj.add("http://www.w3.org/2004/02/skos/core");
+
+ return new SparqlQueryType("forbid", obj, pred, "false");
+ }
+
public static SparqlQueryType SKOS(){
Set<String> pred = new HashSet<String>();
//pred.add("http://www.w3.org/2004/02/skos/core");
@@ -94,6 +128,36 @@
return new SparqlQueryType("forbid", obj, pred, "false");
}
+ public static SparqlQueryType YAGOSKOS(){
+ Set<String> pred = new HashSet<String>();
+ //pred.add("http://www.w3.org/2004/02/skos/core");
+ pred.add("http://www.w3.org/2002/07/owl#sameAs");
+ pred.add("http://xmlns.com/foaf/0.1/");
+
+ pred.add("http://dbpedia.org/property/reference");
+ pred.add("http://dbpedia.org/property/website");
+ pred.add("http://dbpedia.org/property/wikipage");
+ //pred.add("http://www.w3.org/2004/02/skos/core#narrower");
+ pred.add("http://dbpedia.org/property/wikiPageUsesTemplate");
+
+ Set<String> obj = new HashSet<String>();
+ //obj.add("http://dbpedia.org/resource/Category:Wikipedia_");
+ //obj.add("http://dbpedia.org/resource/Category:Articles_");
+ obj.add("http://xmlns.com/foaf/0.1/");
+ obj.add("http://upload.wikimedia.org/wikipedia/commons");
+ obj.add("http://upload.wikimedia.org/wikipedia");
+
+ obj.add("http://www.geonames.org");
+ obj.add("http://www.w3.org/2006/03/wn/wn20/instances/synset");
+ obj.add("http://www4.wiwiss.fu-berlin.de/flickrwrappr");
+
+
+ //obj.add("http://dbpedia.org/class/yago");
+ obj.add("http://dbpedia.org/resource/Template");
+
+
+ return new SparqlQueryType("forbid", obj, pred, "false");
+ }
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -23,13 +23,11 @@
import java.util.HashSet;
import java.util.Iterator;
import java.util.Set;
-import java.util.SortedSet;
-import java.util.TreeSet;
import java.util.Vector;
import org.dllearner.utilities.StringTuple;
-public class PropertyNode extends Node {
+public class PropertyNode extends Node implements Comparable{
// the a and b part of a property
private Node a;
@@ -52,7 +50,7 @@
}
@Override
- public Vector<Node> expand(TypedSparqlQuery tsq, Manipulator m) {
+ public Vector<Node> expand(TypedSparqlQueryInterface tsq, Manipulator m) {
Set<StringTuple> s = tsq.query(uri);
Vector<Node> Nodes = new Vector<Node>();
// Manipulation
@@ -75,7 +73,7 @@
// gets the types for properties recursively
@Override
- public Vector<Node> expandProperties(TypedSparqlQuery tsq, Manipulator m) {
+ public Vector<Node> expandProperties(TypedSparqlQueryInterface tsq, Manipulator m) {
b.expandProperties(tsq, m);
return this.expand(tsq, m);
}
@@ -98,5 +96,21 @@
return s;
}
+
+ @Override
+ public boolean equals(Node n){
+ if(this.uri.equals(n.uri))return true;
+ else return false;
+ }
+ @Override
+ public int compareTo(Object n){
+ return 0;
+ //
+ }
+ @Override
+ public int compareTo(Node n){
+ return super.compareTo(n);
+ //
+ }
}
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlQueryMaker.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -61,13 +61,15 @@
if(domain){
ret = "SELECT * WHERE { " + lineend +
"?subject <" + role + "> ?object; a []. " + lineend
- + "FILTER( " + lineend + "(" + Filter + ").}";
+ + "FILTER( " + lineend + "(" + Filter + ").}" ;
+ //"ORDER BY ?subject";
// System.out.println(ret);
}else{
ret = "SELECT * WHERE { " + lineend +
"?object a [] . " +
"?subject <" + role + "> ?object . " + lineend
+ "FILTER( " + lineend + "(" + Filter + ").}";
+ //"ORDER BY ?object";
}
//System.out.println(ret);
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java 2007-12-08 19:37:08 UTC (rev 334)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQuery.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -35,9 +35,9 @@
import org.dllearner.utilities.StringTuple;
// can execute different queries
-public class TypedSparqlQuery {
+public class TypedSparqlQuery implements TypedSparqlQueryInterface{
boolean print_flag=false;
- boolean debug_no_cache=false;
+ boolean debug_no_cache=false;// true means no cahce is used
private Configuration configuration;
// private SparqlHTTPRequest SparqlHTTPRequest;
private SparqlQueryMaker sparqlQueryMaker;
@@ -106,7 +106,7 @@
//System.out.print("\n");
} else {
xml = FromCache;
- System.out.println("FROM CACHE");
+ //System.out.println("FROM CACHE");
}
// System.out.println(sparql);
@@ -114,7 +114,7 @@
// process XML
Set<StringTuple> s = processResult(xml, a, b);
try {
- System.out.println("retrieved " + s.size() + " tupels\n");
+ //System.out.println("retrieved " + s.size() + " tupels\n");
} catch (Exception e) {
}
return s;
@@ -124,30 +124,73 @@
Set<StringTuple> ret = new HashSet<StringTuple>();
// TODO if result is empty, catch exceptions
- String one = "<binding name=\"" + a + "\">";
- String two = "<binding name=\"" + b + "\">";
+ String resEnd="</result>";
+ String one = "binding name=\"" + a + "\"";
+ String two = "binding name=\"" + b + "\"";
+ String endbinding= "binding";
+ String uri="uri";
String uridel = "<uri>";
- String end = "</uri>";
+ String bnode = "<bnode>";
+ String uriend = "</uri>";
String predtmp = "";
String objtmp = "";
-
- while (xml.indexOf(one) != -1) {
+ //System.out.println(getNextResult(xml));
+ String nextResult="";
+ while ((nextResult=getNextResult( xml))!=null){
+ //System.out.println(xml.indexOf(resEnd));
+ //System.out.println(xml);
+ if(nextResult.indexOf(bnode)!=-1)
+ {xml=xml.substring(xml.indexOf(resEnd)+resEnd.length());continue;}
// get pred
- xml = xml.substring(xml.indexOf(one) + one.length());
- xml = xml.substring(xml.indexOf(uridel) + uridel.length());
- predtmp = xml.substring(0, xml.indexOf(end));
-
+ //predtmp = nextResult.substring(nextResult.indexOf(one) + one.length());
+ predtmp=getinTag(nextResult, one,endbinding);
+ predtmp=getinTag(predtmp, uri,uri);
+ //System.out.println(predtmp);
+
// getobj
- xml = xml.substring(xml.indexOf(two) + two.length());
- xml = xml.substring(xml.indexOf(uridel) + uridel.length());
- objtmp = xml.substring(0, xml.indexOf(end));
- ret.add(new StringTuple(predtmp, objtmp));
- // System.out.println(new Tupel(predtmp,objtmp));
+ objtmp=getinTag(nextResult, two,endbinding);
+ objtmp=getinTag(objtmp, uri,uri);
+ //System.out.println(objtmp);
+
+ StringTuple st=new StringTuple(predtmp, objtmp);
+ //System.out.println(st);
+ ret.add(st);
+ xml=xml.substring(xml.indexOf(resEnd)+resEnd.length());
+
}
+ /*while (xml.indexOf(one) != -1) {
+
+
+
+ // System.out.println(new Tupel(predtmp,objtmp));
+ }*/
+
return ret;
}
+
+ private String getNextResult(String xml){
+ String res1="<result>";
+ String res2="</result>";
+ if(xml.indexOf(res1)==-1)return null;
+ xml = xml.substring(xml.indexOf(res1) + res1.length());
+ xml = xml.substring(0,xml.indexOf(res2) );
+ //System.out.println(xml);
+ return xml;
+ }
+ private String getinTag(String xml, String starttag, String endtag){
+ String res1="<"+starttag+">";
+ //System.out.println(res1);
+ String res2="</"+endtag+">";
+ if(xml.indexOf(res1)==-1)return null;
+ xml = xml.substring(xml.indexOf(res1) + res1.length());
+ //System.out.println(xml);
+ xml = xml.substring(0,xml.indexOf(res2) );
+ //System.out.println(xml);
+
+ return xml;
+ }
private String sendAndReceiveSPARQL(String sparql) throws IOException {
p("sendAndReceiveSPARQL");
@@ -214,5 +257,7 @@
System.out.println(str);
}
}
+
+
}
Added: trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryClasses.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryClasses.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryClasses.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -0,0 +1,245 @@
+/**
+ * Copyright (C) 2007, Sebastian Hellmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb.sparql;
+
+import java.io.BufferedReader;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.OutputStream;
+import java.io.OutputStreamWriter;
+import java.net.HttpURLConnection;
+import java.net.URI;
+import java.net.URLEncoder;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+
+import org.dllearner.utilities.StringTuple;
+
+// can execute different queries
+public class TypedSparqlQueryClasses implements TypedSparqlQueryInterface{
+ boolean print_flag=false;
+ boolean debug_no_cache=false;
+ private Configuration configuration;
+ // private SparqlHTTPRequest SparqlHTTPRequest;
+ //private SparqlQueryMaker sparqlQueryMaker;
+ Cache cache;
+
+
+ public TypedSparqlQueryClasses(Configuration configuration) {
+ this.configuration = configuration;
+ this.cache = new Cache("cache");
+ }
+
+ // standard query get a tupels (p,o) for subject s
+ public Set<StringTuple> query(URI u) {
+
+ // getQuery
+ String sparql = "SELECT ?predicate ?object " +
+ "WHERE {" +
+ "<"+u.toString()+"> ?predicate ?object;" +
+ "a ?object . " +
+ " FILTER (!regex(str(?object),'http://xmlns.com/foaf/0.1/'))"+
+ "}";
+
+ return cachedSparql(u, sparql, "predicate", "object");
+
+ }
+
+
+
+ // uses a cache
+ private Set<StringTuple> cachedSparql(URI u, String sparql, String a, String b) {
+ // check cache
+ String FromCache = cache.get(u.toString(), sparql);
+ if(debug_no_cache) {
+ FromCache=null;
+ }
+ String xml = null;
+ // if not in cache get it from EndPoint
+ if (FromCache == null) {
+ try {
+ xml = sendAndReceiveSPARQL(sparql);
+ } catch (IOException e) {
+ // TODO Auto-generated catch block
+ e.printStackTrace();
+ }
+ p(sparql);
+ // System.out.println(xml);
+ if(!debug_no_cache) {
+ cache.put(u.toString(), sparql, xml);
+ }
+ //System.out.print("\n");
+ } else {
+ xml = FromCache;
+ //System.out.println("FROM CACHE");
+ }
+
+ // System.out.println(sparql);
+ // System.out.println(xml);
+ // process XML
+ Set<StringTuple> s = processResult(xml, a, b);
+ try {
+ //System.out.println("retrieved " + s.size() + " tupels\n");
+ } catch (Exception e) {
+ }
+ return s;
+ }
+
+ public Set<StringTuple> processResult(String xml, String a, String b) {
+
+ Set<StringTuple> ret = new HashSet<StringTuple>();
+ // TODO if result is empty, catch exceptions
+ String resEnd="</result>";
+ String one = "binding name=\"" + a + "\"";
+ String two = "binding name=\"" + b + "\"";
+ String endbinding= "binding";
+ String uri="uri";
+ String uridel = "<uri>";
+ String bnode = "<bnode>";
+ String uriend = "</uri>";
+ String predtmp = "";
+ String objtmp = "";
+ //System.out.println(getNextResult(xml));
+ String nextResult="";
+ while ((nextResult=getNextResult( xml))!=null){
+ //System.out.println(xml.indexOf(resEnd));
+ //System.out.println(xml);
+ if(nextResult.indexOf(bnode)!=-1)
+ {xml=xml.substring(xml.indexOf(resEnd)+resEnd.length());continue;}
+ // get pred
+ //predtmp = nextResult.substring(nextResult.indexOf(one) + one.length());
+ predtmp=getinTag(nextResult, one,endbinding);
+ predtmp=getinTag(predtmp, uri,uri);
+ //System.out.println(predtmp);
+
+ // getobj
+ objtmp=getinTag(nextResult, two,endbinding);
+ objtmp=getinTag(objtmp, uri,uri);
+ //System.out.println(objtmp);
+
+ StringTuple st=new StringTuple(predtmp, objtmp);
+ //System.out.println(st);
+ ret.add(st);
+ xml=xml.substring(xml.indexOf(resEnd)+resEnd.length());
+
+ }
+ /*while (xml.indexOf(one) != -1) {
+
+
+
+
+ // System.out.println(new Tupel(predtmp,objtmp));
+ }*/
+
+ return ret;
+
+ }
+
+ private String getNextResult(String xml){
+ String res1="<result>";
+ String res2="</result>";
+ if(xml.indexOf(res1)==-1)return null;
+ xml = xml.substring(xml.indexOf(res1) + res1.length());
+ xml = xml.substring(0,xml.indexOf(res2) );
+ //System.out.println(xml);
+ return xml;
+ }
+ private String getinTag(String xml, String starttag, String endtag){
+ String res1="<"+starttag+">";
+ //System.out.println(res1);
+ String res2="</"+endtag+">";
+ if(xml.indexOf(res1)==-1)return null;
+ xml = xml.substring(xml.indexOf(res1) + res1.length());
+ //System.out.println(xml);
+ xml = xml.substring(0,xml.indexOf(res2) );
+ //System.out.println(xml);
+
+ return xml;
+ }
+
+ private String sendAndReceiveSPARQL(String sparql) throws IOException {
+ p("sendAndReceiveSPARQL");
+ StringBuilder answer = new StringBuilder();
+ //sparql="SELECT * WHERE {?a ?b ?c}LIMIT 10";
+
+ // String an Sparql-Endpoint schicken
+ HttpURLConnection connection;
+ SpecificSparqlEndpoint se = configuration.getSparqlEndpoint();
+ p("URL: "+se.getURL());
+ p("Host: "+se.getHost());
+
+ connection = (HttpURLConnection) se.getURL().openConnection();
+ connection.setDoOutput(true);
+
+ //connection.addRequestProperty("Host", se.getHost());
+ connection.addRequestProperty("Connection", "close");
+ connection
+ .addRequestProperty(
+ "Accept",
+ "text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5");
+ connection.addRequestProperty("Accept-Language", "de-de,de;q=0.8,en-us;q=0.5,en;q=0.3");
+ connection.addRequestProperty("Accept-Charset", "utf-8;q=1.0");
+ connection
+ .addRequestProperty(
+ "User-Agent",
+ "Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4 Web-Sniffer/1.0.24");
+
+ OutputStream os = connection.getOutputStream();
+ OutputStreamWriter osw = new OutputStreamWriter(os);
+
+ Set<String> s = se.getParameters().keySet();
+ Iterator<String> it = s.iterator();
+ String FullURI = "";
+ while (it.hasNext()) {
+ String element = it.next();
+ FullURI += "" + URLEncoder.encode(element, "UTF-8") + "="
+ + URLEncoder.encode(se.getParameters().get(element), "UTF-8") + "&";
+ }
+
+ FullURI += "" + se.getHasQueryParameter() + "=" + URLEncoder.encode(sparql, "UTF-8");
+ p(FullURI);
+ osw.write(FullURI);
+ osw.close();
+
+ // receive answer
+ InputStream is = connection.getInputStream();
+ InputStreamReader isr = new InputStreamReader(is, "UTF-8");
+ BufferedReader br = new BufferedReader(isr);
+
+ String line;
+ do {
+ line = br.readLine();
+ if (line != null)
+ answer.append(line);
+ } while (line != null);
+
+ br.close();
+ p(answer.toString());
+ return answer.toString();
+ }
+ public void p(String str){
+ if(print_flag){
+ System.out.println(str);
+ }
+ }
+
+}
Added: trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryInterface.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryInterface.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/TypedSparqlQueryInterface.java 2007-12-14 20:16:47 UTC (rev 335)
@@ -0,0 +1,11 @@
+package org.dllearner.kb.sparql;
+
+import java.net.URI;
+import java.util.Set;
+
+import org.dllearner.utilities.StringTuple;
+
+public interface TypedSparqlQueryInterface {
+
+ public Set<StringTuple> query(URI u);
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <jen...@us...> - 2008-01-02 12:59:34
|
Revision: 336
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=336&view=rev
Author: jenslehmann
Date: 2008-01-02 04:59:17 -0800 (Wed, 02 Jan 2008)
Log Message:
-----------
fixed a couple of errors and warnings
Modified Paths:
--------------
trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java
trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java
Modified: trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2007-12-14 20:16:47 UTC (rev 335)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlEndpointRestructured.java 2008-01-02 12:59:17 UTC (rev 336)
@@ -30,7 +30,6 @@
import java.util.HashMap;
import java.util.HashSet;
import java.util.LinkedList;
-import java.util.Random;
import java.util.Set;
import org.dllearner.core.KnowledgeSource;
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java 2007-12-14 20:16:47 UTC (rev 335)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ClassNode.java 2008-01-02 12:59:17 UTC (rev 336)
@@ -23,8 +23,6 @@
import java.util.HashSet;
import java.util.Iterator;
import java.util.Set;
-import java.util.SortedSet;
-import java.util.TreeSet;
import java.util.Vector;
import org.dllearner.utilities.StringTuple;
@@ -103,11 +101,6 @@
}
@Override
- public int compareTo(Object n){
- return super.compareTo((Node)n);
- //
- }
- @Override
public int compareTo(Node n){
return super.compareTo(n);
//
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2007-12-14 20:16:47 UTC (rev 335)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/ExtractionAlgorithm.java 2008-01-02 12:59:17 UTC (rev 336)
@@ -21,9 +21,6 @@
import java.net.URI;
import java.util.HashSet;
-import java.util.Iterator;
-import java.util.Set;
-import java.util.TreeSet;
import java.util.Vector;
// this class is used to extract the information recursively
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2007-12-14 20:16:47 UTC (rev 335)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/InstanceNode.java 2008-01-02 12:59:17 UTC (rev 336)
@@ -23,8 +23,6 @@
import java.util.HashSet;
import java.util.Iterator;
import java.util.Set;
-import java.util.SortedSet;
-import java.util.TreeSet;
import java.util.Vector;
import org.dllearner.utilities.StringTuple;
@@ -107,11 +105,6 @@
}
@Override
- public int compareTo(Object n){
- return 0;
- //
- }
- @Override
public int compareTo(Node n){
return super.compareTo(n);
//
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java 2007-12-14 20:16:47 UTC (rev 335)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/Node.java 2008-01-02 12:59:17 UTC (rev 336)
@@ -24,7 +24,7 @@
import java.util.Vector;
// abstract class
-public abstract class Node implements Comparable{
+public abstract class Node implements Comparable<Node> {
URI uri;
protected String type;
protected boolean expanded = false;
Modified: trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java 2007-12-14 20:16:47 UTC (rev 335)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/PropertyNode.java 2008-01-02 12:59:17 UTC (rev 336)
@@ -27,7 +27,7 @@
import org.dllearner.utilities.StringTuple;
-public class PropertyNode extends Node implements Comparable{
+public class PropertyNode extends Node {
// the a and b part of a property
private Node a;
@@ -102,15 +102,10 @@
if(this.uri.equals(n.uri))return true;
else return false;
}
+
@Override
- public int compareTo(Object n){
- return 0;
- //
- }
- @Override
public int compareTo(Node n){
return super.compareTo(n);
- //
}
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <ku...@us...> - 2008-01-18 10:47:36
|
Revision: 388
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=388&view=rev
Author: kurzum
Date: 2008-01-18 02:47:32 -0800 (Fri, 18 Jan 2008)
Log Message:
-----------
renamed knowledgeSource
Added Paths:
-----------
trunk/src/dl-learner/org/dllearner/kb/SparqlKnowledgeSource.java
Removed Paths:
-------------
trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlKnowledgeSource.java
Copied: trunk/src/dl-learner/org/dllearner/kb/SparqlKnowledgeSource.java (from rev 386, trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlKnowledgeSource.java)
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/SparqlKnowledgeSource.java (rev 0)
+++ trunk/src/dl-learner/org/dllearner/kb/SparqlKnowledgeSource.java 2008-01-18 10:47:32 UTC (rev 388)
@@ -0,0 +1,524 @@
+/**
+ * Copyright (C) 2007, Jens Lehmann
+ *
+ * This file is part of DL-Learner.
+ *
+ * DL-Learner is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * DL-Learner is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+package org.dllearner.kb;
+
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+import java.net.MalformedURLException;
+import java.net.URI;
+import java.net.URL;
+import java.util.Collection;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.LinkedList;
+import java.util.Set;
+
+import org.dllearner.core.KnowledgeSource;
+import org.dllearner.core.OntologyFormat;
+import org.dllearner.core.OntologyFormatUnsupportedException;
+import org.dllearner.core.config.BooleanConfigOption;
+import org.dllearner.core.config.ConfigEntry;
+import org.dllearner.core.config.ConfigOption;
+import org.dllearner.core.config.IntegerConfigOption;
+import org.dllearner.core.config.InvalidConfigOptionValueException;
+import org.dllearner.core.config.StringConfigOption;
+import org.dllearner.core.config.StringSetConfigOption;
+import org.dllearner.core.config.StringTupleListConfigOption;
+import org.dllearner.core.dl.KB;
+import org.dllearner.kb.sparql.Manager;
+import org.dllearner.kb.sparql.Manipulator;
+import org.dllearner.kb.sparql.configuration.PredefinedEndpoint;
+import org.dllearner.kb.sparql.configuration.PredefinedFilter;
+import org.dllearner.kb.sparql.configuration.SparqlQueryType;
+import org.dllearner.kb.sparql.configuration.SpecificSparqlEndpoint;
+import org.dllearner.kb.sparql.old.*;
+import org.dllearner.parser.KBParser;
+import org.dllearner.reasoning.DIGConverter;
+import org.dllearner.reasoning.JenaOWLDIGConverter;
+import org.dllearner.utilities.StringTuple;
+
+/**
+ * Represents a SPARQL Endpoint.
+ *
+ * @author Jens Lehmann
+ * @author Sebastian Knappe
+ * @author Sebastian Hellmann
+ */
+public class SparqlKnowledgeSource extends KnowledgeSource {
+
+ // ConfigOptions
+ private URL url;
+ String host;
+ private Set<String> instances=new HashSet<String>();;
+ private URL dumpFile;
+ private int recursionDepth = 1;
+ private int predefinedFilter = 0;
+ private int predefinedEndpoint = 0;
+ private Set<String> predList=new HashSet<String>();
+ private Set<String> objList=new HashSet<String>();
+ // private Set<String> classList;
+ private String format = "N-TRIPLES";
+ private boolean dumpToFile = true;
+ private boolean useLits = false;
+ private boolean getAllSuperClasses = true;
+ private boolean closeAfterRecursion = true;
+ private int breakSuperClassRetrievalAfter = 200;
+
+ private boolean learnDomain = false;
+ private boolean learnRange = false;
+ private int numberOfInstancesUsedForRoleLearning=40;
+ private String role="";
+ private String blankNodeIdentifier = "bnode";
+
+ LinkedList<StringTuple> URIParameters = new LinkedList<StringTuple>();
+ LinkedList<StringTuple> replacePredicate = new LinkedList<StringTuple>();
+ LinkedList<StringTuple> replaceObject = new LinkedList<StringTuple>();
+
+ /**
+ * Holds the results of the calculateSubjects method
+ */
+ private String[] subjects;
+
+ /**
+ * Holds the results of the calculateTriples method
+ */
+ private String[] triples;
+
+ /**
+ * Holds the results of the calculateConceptSubjects method
+ */
+ private String[] conceptSubjects;
+
+ /**
+ * if a method is running this becomes true
+ */
+ private boolean subjectThreadRunning = false;
+
+ private boolean triplesThreadRunning = false;
+
+ private boolean conceptThreadRunning = false;
+
+ /**
+ * the Thread that is running a method
+ */
+ private Thread subjectThread;
+
+ private Thread triplesThread;
+
+ private Thread conceptThread;
+
+ // received ontology as array, used if format=Array(an element of the
+ // array consists of the subject, predicate and object separated by '<'
+ private String[] ontArray;
+
+ // received ontology as KB, the internal format
+ private KB kb;
+
+ public static String getName() {
+ return "SPARQL Endpoint Restructured";
+ }
+
+ /**
+ * sets the ConfigOptions for this KnowledgeSource
+ *
+ * @return
+ */
+ public static Collection<ConfigOption<?>> createConfigOptions() {
+ Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
+ options.add(new StringConfigOption("url", "URL of SPARQL Endpoint"));
+ options.add(new StringConfigOption("host", "host of SPARQL Endpoint"));
+ options.add(new StringSetConfigOption("instances",
+ "relevant instances e.g. positive and negative examples in a learning problem"));
+ options.add(new IntegerConfigOption("recursionDepth",
+ "recursion depth of KB fragment selection", 2));
+ options.add(new IntegerConfigOption("predefinedFilter", "the mode of the SPARQL Filter"));
+ options.add(new IntegerConfigOption("predefinedEndpoint", "the mode of the SPARQL Filter"));
+
+ options.add(new StringSetConfigOption("predList", "list of all ignored roles"));
+ options.add(new StringSetConfigOption("objList", "list of all ignored objects"));
+ options.add(new StringSetConfigOption("classList", "list of all ignored classes"));
+ options.add(new StringConfigOption("format", "N-TRIPLES or KB format", "N-TRIPLES"));
+ options.add(new BooleanConfigOption("dumpToFile",
+ "Specifies whether the extracted ontology is written to a file or not.", true));
+ options.add(new BooleanConfigOption("useLits", "use Literals in SPARQL query"));
+ options.add(new BooleanConfigOption("getAllSuperClasses", "If true then all superclasses are retrieved until the most general class (owl:Thing) is reached.", true));
+
+ options.add(new BooleanConfigOption("learnDomain", "learns the Domain for a Role"));
+ options.add(new BooleanConfigOption("learnRange", "learns the Range for a Role"));
+ options.add(new StringConfigOption("role", "role to learn Domain/Range from"));
+ options.add(new StringConfigOption("blankNodeIdentifier",
+ "used to identify blanknodes in Tripels"));
+
+ options.add(new StringTupleListConfigOption("example", "example"));
+ options.add(new StringTupleListConfigOption("replacePredicate", "rule for replacing predicates"));
+ options.add(new StringTupleListConfigOption("replaceObject", "rule for replacing predicates"));
+ options.add(new IntegerConfigOption("breakSuperClassRetrievalAfter", "stops a cyclic hierarchy after specified number of classes"));
+ options.add(new IntegerConfigOption("numberOfInstancesUsedForRoleLearning", ""));
+ options.add(new BooleanConfigOption("closeAfterRecursion", "gets all classes for all instances"));
+
+
+
+ return options;
+ }
+
+ /*
+ * @see org.dllearner.core.Component#applyConfigEntry(org.dllearner.core.ConfigEntry)
+ */
+ @Override
+ @SuppressWarnings( { "unchecked" })
+ public <T> void applyConfigEntry(ConfigEntry<T> entry) throws InvalidConfigOptionValueException {
+ String option = entry.getOptionName();
+ if (option.equals("url")) {
+ String s = (String) entry.getValue();
+ try {
+ url = new URL(s);
+ } catch (MalformedURLException e) {
+ throw new InvalidConfigOptionValueException(entry.getOption(), entry.getValue(),
+ "malformed URL " + s);
+ }
+ } else if (option.equals("host")) {
+ host = (String) entry.getValue();
+ } else if (option.equals("instances")) {
+ instances = (Set<String>) entry.getValue();
+ } else if (option.equals("recursionDepth")) {
+ recursionDepth = (Integer) entry.getValue();
+ } else if (option.equals("predList")) {
+ predList = (Set<String>) entry.getValue();
+ } else if (option.equals("objList")) {
+ objList = (Set<String>) entry.getValue();
+ //} else if (option.equals("classList")) {
+ // classList = (Set<String>) entry.getValue();
+ } else if (option.equals("predefinedEndpoint")) {
+ predefinedEndpoint = (Integer) entry.getValue();
+ } else if (option.equals("predefinedFilter")) {
+ predefinedFilter = (Integer) entry.getValue();
+ } else if (option.equals("format")) {
+ format = (String) entry.getValue();
+ } else if (option.equals("dumpToFile")) {
+ dumpToFile = (Boolean) entry.getValue();
+ } else if (option.equals("useLits")) {
+ useLits = (Boolean) entry.getValue();
+ } else if (option.equals("getAllSuperClasses")) {
+ getAllSuperClasses = (Boolean) entry.getValue();
+ } else if (option.equals("learnDomain")) {
+ learnDomain = (Boolean) entry.getValue();
+ }else if (option.equals("learnRange")) {
+ learnRange = (Boolean) entry.getValue();
+ } else if (option.equals("role")) {
+ role = (String) entry.getValue();
+ } else if (option.equals("blankNodeIdentifier")) {
+ blankNodeIdentifier = (String) entry.getValue();
+ } else if (option.equals("example")) {
+ //System.out.println(entry.getValue());
+ }else if (option.equals("replacePredicate")) {
+ replacePredicate = (LinkedList)entry.getValue();
+ }else if (option.equals("replaceObject")) {
+ replaceObject = (LinkedList)entry.getValue();
+ }else if (option.equals("breakSuperClassRetrievalAfter")) {
+ breakSuperClassRetrievalAfter = (Integer) entry.getValue();
+ }else if (option.equals("numberOfInstancesUsedForRoleLearning")) {
+ numberOfInstancesUsedForRoleLearning = (Integer) entry.getValue();
+ }else if (option.equals("closeAfterRecursion")) {
+ closeAfterRecursion = (Boolean) entry.getValue();
+ }
+
+ }
+
+ /*
+ * (non-Javadoc)
+ *
+ * @see org.dllearner.core.Component#init()
+ */
+ @Override
+ public void init() {
+ System.out.println("SparqlModul: Collecting Ontology");
+ // SparqlOntologyCollector oc=
+ // new SparqlOntologyCollector(Datastructures.setToArray(instances),
+ // numberOfRecursions, filterMode,
+ // Datastructures.setToArray(predList),Datastructures.setToArray(
+ // objList),Datastructures.setToArray(classList),format,url,useLits);
+ Manager m = new Manager();
+ SpecificSparqlEndpoint sse = null;
+ SparqlQueryType sqt = null;
+ // get Options for Manipulator
+ Manipulator man = new Manipulator(blankNodeIdentifier,breakSuperClassRetrievalAfter,replacePredicate,replaceObject);
+ HashMap<String, String> parameters = new HashMap<String, String>();
+ parameters.put("default-graph-uri", "http://dbpedia.org");
+ parameters.put("format", "application/sparql-results.xml");
+
+ // get Options for endpoints
+ if (predefinedEndpoint >= 1) {
+ sse = PredefinedEndpoint.getEndpoint(predefinedEndpoint);
+ } else {
+ sse = new SpecificSparqlEndpoint(url, host, parameters);
+ }
+
+ // get Options for Filters
+
+ if (predefinedFilter >= 1) {
+ sqt = PredefinedFilter.getFilter(predefinedFilter);
+
+ } else {
+ sqt = new SparqlQueryType("forbid", objList, predList, useLits + "");
+
+ }
+ // give everything to the manager
+ m.useConfiguration(sqt, sse, man, recursionDepth, getAllSuperClasses,closeAfterRecursion);
+ try {
+ String ont = "";
+ //System.out.println(learnDomain);
+ // used to learn a domain of a role
+ if (learnDomain || learnRange) {
+ Set<String> pos=new HashSet<String>();
+ Set<String> neg=new HashSet<String>();
+ if(learnDomain){
+ pos = m.getDomainInstancesForRole(role);
+ neg = m.getRangeInstancesForRole(role);
+ }else if(learnRange){
+ neg = m.getDomainInstancesForRole(role);
+ pos = m.getRangeInstancesForRole(role);
+ }
+ //choose 30
+
+
+ Set<String> tmp=new HashSet<String>();
+ for(String one:pos){
+ tmp.add(one);
+ if(tmp.size()>=numberOfInstancesUsedForRoleLearning)break;
+ }
+ pos=tmp;
+ System.out.println("Instances used: "+pos.size());
+
+ tmp=new HashSet<String>();
+ for(String one:neg){
+ tmp.add(one);
+ if(tmp.size()>=numberOfInstancesUsedForRoleLearning)break;
+ }
+ neg=tmp;
+
+ instances=new HashSet<String>();
+ instances.addAll(pos);
+
+ instances.addAll(neg);
+
+ for(String one:pos){
+ System.out.println("+\""+one+"\"");
+ }
+ for(String one:neg){
+ System.out.println("-\""+one+"\"");
+ }
+
+ /*Random r= new Random();
+
+
+ Object[] arr=instances.toArray();
+ while(instances.size()>=30){
+
+ }*/
+ // add the role to the filter(a solution is always EXISTS
+ // role.TOP)
+ m.addPredicateFilter(role);
+ //System.out.println(instances);
+ // THIS is a workaround
+
+ }
+ // the actual extraction is started here
+ ont = m.extract(instances);
+ System.out.println("Number of cached SPARQL queries: "+m.getConfiguration().numberOfCachedSparqlQueries);
+ System.out.println("Number of uncached SPARQL queries: "+m.getConfiguration().numberOfUncachedSparqlQueries);
+
+ System.out.println("Finished collecting Fragment");
+
+ if (dumpToFile) {
+ String filename = System.currentTimeMillis() + ".nt";
+ String basedir = "cache" + File.separator;
+ try {
+ if (!new File(basedir).exists())
+ new File(basedir).mkdir();
+
+ FileWriter fw = new FileWriter(new File(basedir + filename), true);
+ fw.write(ont);
+ fw.flush();
+ fw.close();
+
+ dumpFile = (new File(basedir + filename)).toURI().toURL();
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ }
+ if (format.equals("KB")) {
+ try {
+ //kb = KBParser.parseKBFile(new StringReader(ont));
+ kb=KBParser.parseKBFile(dumpFile);
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ }
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ System.out.println("SparqlModul: ****Finished");
+ }
+
+ /*
+ * (non-Javadoc)
+ *
+ * @see org.dllearner.core.KnowledgeSource#toDIG()
+ */
+ @Override
+ public String toDIG(URI kbURI) {
+ if (format.equals("N-TRIPLES"))
+ return JenaOWLDIGConverter.getTellsString(dumpFile, OntologyFormat.N_TRIPLES, kbURI);
+ else
+ return DIGConverter.getDIGString(kb, kbURI).toString();
+ }
+
+ /*
+ * (non-Javadoc)
+ *
+ * @see org.dllearner.core.KnowledgeSource#export(java.io.File,
+ * org.dllearner.core.OntologyFormat)
+ */
+ @Override
+ public void export(File file, OntologyFormat format) throws OntologyFormatUnsupportedException {
+ // currently no export functions implemented, so we just throw an
+ // exception
+ throw new OntologyFormatUnsupportedException("export", format);
+ }
+
+ public URL getURL() {
+ return url;
+ }
+
+ public String[] getOntArray() {
+ return ontArray;
+ }
+
+
+ /**
+ * TODO SparqlOntologyCollector needs to be removed
+ * @param label
+ * @param limit
+ */
+ public void calculateSubjects(String label, int limit) {
+ System.out.println("SparqlModul: Collecting Subjects");
+ oldSparqlOntologyCollector oc = new oldSparqlOntologyCollector(url);
+ try {
+ subjects = oc.getSubjectsFromLabel(label, limit);
+ } catch (IOException e) {
+ subjects = new String[1];
+ subjects[0] = "[Error]Sparql Endpoint could not be reached.";
+ }
+ System.out.println("SparqlModul: ****Finished");
+ }
+
+ /**
+ * TODO SparqlOntologyCollector needs to be removed
+ * @param subject
+ */
+ public void calculateTriples(String subject) {
+ System.out.println("SparqlModul: Collecting Triples");
+ oldSparqlOntologyCollector oc = new oldSparqlOntologyCollector(url);
+ try {
+ triples = oc.collectTriples(subject);
+ } catch (IOException e) {
+ triples = new String[1];
+ triples[0] = "[Error]Sparql Endpoint could not be reached.";
+ }
+ System.out.println("SparqlModul: ****Finished");
+ }
+
+ /**
+ * TODO SparqlOntologyCollector needs to be removed
+ * @param concept
+ */
+ public void calculateConceptSubjects(String concept) {
+ System.out.println("SparqlModul: Collecting Subjects");
+ oldSparqlOntologyCollector oc = new oldSparqlOntologyCollector(url);
+ try {
+ conceptSubjects = oc.getSubjectsFromConcept(concept);
+ } catch (IOException e) {
+ conceptSubjects = new String[1];
+ conceptSubjects[0] = "[Error]Sparql Endpoint could not be reached.";
+ }
+ System.out.println("SparqlModul: ****Finished");
+ }
+
+ public boolean subjectThreadIsRunning() {
+ return subjectThreadRunning;
+ }
+
+ public void setSubjectThreadRunning(boolean bool) {
+ subjectThreadRunning = bool;
+ }
+
+ public boolean triplesThreadIsRunning() {
+ return triplesThreadRunning;
+ }
+
+ public void setTriplesThreadRunning(boolean bool) {
+ triplesThreadRunning = bool;
+ }
+
+ public boolean conceptThreadIsRunning() {
+ return conceptThreadRunning;
+ }
+
+ public void setConceptThreadRunning(boolean bool) {
+ conceptThreadRunning = bool;
+ }
+
+ public String[] getSubjects() {
+ return subjects;
+ }
+
+ public Thread getSubjectThread() {
+ return subjectThread;
+ }
+
+ public void setSubjectThread(Thread subjectThread) {
+ this.subjectThread = subjectThread;
+ }
+
+ public Thread getTriplesThread() {
+ return triplesThread;
+ }
+
+ public void setTriplesThread(Thread triplesThread) {
+ this.triplesThread = triplesThread;
+ }
+
+ public Thread getConceptThread() {
+ return conceptThread;
+ }
+
+ public void setConceptThread(Thread conceptThread) {
+ this.conceptThread = conceptThread;
+ }
+
+ public String[] getTriples() {
+ return triples;
+ }
+
+ public String[] getConceptSubjects() {
+ return conceptSubjects;
+ }
+}
Deleted: trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlKnowledgeSource.java
===================================================================
--- trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlKnowledgeSource.java 2008-01-18 10:29:26 UTC (rev 387)
+++ trunk/src/dl-learner/org/dllearner/kb/sparql/SparqlKnowledgeSource.java 2008-01-18 10:47:32 UTC (rev 388)
@@ -1,522 +0,0 @@
-/**
- * Copyright (C) 2007, Jens Lehmann
- *
- * This file is part of DL-Learner.
- *
- * DL-Learner is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * DL-Learner is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <http://www.gnu.org/licenses/>.
- *
- */
-package org.dllearner.kb.sparql;
-
-import java.io.File;
-import java.io.FileWriter;
-import java.io.IOException;
-import java.net.MalformedURLException;
-import java.net.URI;
-import java.net.URL;
-import java.util.Collection;
-import java.util.HashMap;
-import java.util.HashSet;
-import java.util.LinkedList;
-import java.util.Set;
-
-import org.dllearner.core.KnowledgeSource;
-import org.dllearner.core.OntologyFormat;
-import org.dllearner.core.OntologyFormatUnsupportedException;
-import org.dllearner.core.config.BooleanConfigOption;
-import org.dllearner.core.config.ConfigEntry;
-import org.dllearner.core.config.ConfigOption;
-import org.dllearner.core.config.IntegerConfigOption;
-import org.dllearner.core.config.InvalidConfigOptionValueException;
-import org.dllearner.core.config.StringConfigOption;
-import org.dllearner.core.config.StringSetConfigOption;
-import org.dllearner.core.config.StringTupleListConfigOption;
-import org.dllearner.core.dl.KB;
-import org.dllearner.kb.sparql.configuration.PredefinedEndpoint;
-import org.dllearner.kb.sparql.configuration.PredefinedFilter;
-import org.dllearner.kb.sparql.configuration.SparqlQueryType;
-import org.dllearner.kb.sparql.configuration.SpecificSparqlEndpoint;
-import org.dllearner.kb.sparql.old.*;
-import org.dllearner.parser.KBParser;
-import org.dllearner.reasoning.DIGConverter;
-import org.dllearner.reasoning.JenaOWLDIGConverter;
-import org.dllearner.utilities.StringTuple;
-
-/**
- * Represents a SPARQL Endpoint.
- *
- * @author Jens Lehmann
- * @author Sebastian Knappe
- * @author Sebastian Hellmann
- */
-public class SparqlKnowledgeSource extends KnowledgeSource {
-
- // ConfigOptions
- private URL url;
- String host;
- private Set<String> instances=new HashSet<String>();;
- private URL dumpFile;
- private int recursionDepth = 1;
- private int predefinedFilter = 0;
- private int predefinedEndpoint = 0;
- private Set<String> predList=new HashSet<String>();
- private Set<String> objList=new HashSet<String>();
- // private Set<String> classList;
- private String format = "N-TRIPLES";
- private boolean dumpToFile = true;
- private boolean useLits = false;
- private boolean getAllSuperClasses = true;
- private boolean closeAfterRecursion = true;
- private int breakSuperClassRetrievalAfter = 200;
-
- private boolean learnDomain = false;
- private boolean learnRange = false;
- private int numberOfInstancesUsedForRoleLearning=40;
- private String role="";
- private String blankNodeIdentifier = "bnode";
-
- LinkedList<StringTuple> URIParameters = new LinkedList<StringTuple>();
- LinkedList<StringTuple> replacePredicate = new LinkedList<StringTuple>();
- LinkedList<StringTuple> replaceObject = new LinkedList<StringTuple>();
-
- /**
- * Holds the results of the calculateSubjects method
- */
- private String[] subjects;
-
- /**
- * Holds the results of the calculateTriples method
- */
- private String[] triples;
-
- /**
- * Holds the results of the calculateConceptSubjects method
- */
- private String[] conceptSubjects;
-
- /**
- * if a method is running this becomes true
- */
- private boolean subjectThreadRunning = false;
-
- private boolean triplesThreadRunning = false;
-
- private boolean conceptThreadRunning = false;
-
- /**
- * the Thread that is running a method
- */
- private Thread subjectThread;
-
- private Thread triplesThread;
-
- private Thread conceptThread;
-
- // received ontology as array, used if format=Array(an element of the
- // array consists of the subject, predicate and object separated by '<'
- private String[] ontArray;
-
- // received ontology as KB, the internal format
- private KB kb;
-
- public static String getName() {
- return "SPARQL Endpoint Restructured";
- }
-
- /**
- * sets the ConfigOptions for this KnowledgeSource
- *
- * @return
- */
- public static Collection<ConfigOption<?>> createConfigOptions() {
- Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
- options.add(new StringConfigOption("url", "URL of SPARQL Endpoint"));
- options.add(new StringConfigOption("host", "host of SPARQL Endpoint"));
- options.add(new StringSetConfigOption("instances",
- "relevant instances e.g. positive and negative examples in a learning problem"));
- options.add(new IntegerConfigOption("recursionDepth",
- "recursion depth of KB fragment selection", 2));
- options.add(new IntegerConfigOption("predefinedFilter", "the mode of the SPARQL Filter"));
- options.add(new IntegerConfigOption("predefinedEndpoint", "the mode of the SPARQL Filter"));
-
- options.add(new StringSetConfigOption("predList", "list of all ignored roles"));
- options.add(new StringSetConfigOption("objList", "list of all ignored objects"));
- options.add(new StringSetConfigOption("classList", "list of all ignored classes"));
- options.add(new StringConfigOption("format", "N-TRIPLES or KB format", "N-TRIPLES"));
- options.add(new BooleanConfigOption("dumpToFile",
- "Specifies whether the extracted ontology is written to a file or not.", true));
- options.add(new BooleanConfigOption("useLits", "use Literals in SPARQL query"));
- options.add(new BooleanConfigOption("getAllSuperClasses", "If true then all superclasses are retrieved until the most general class (owl:Thing) is reached.", true));
-
- options.add(new BooleanConfigOption("learnDomain", "learns the Domain for a Role"));
- options.add(new BooleanConfigOption("learnRange", "learns the Range for a Role"));
- options.add(new StringConfigOption("role", "role to learn Domain/Range from"));
- options.add(new StringConfigOption("blankNodeIdentifier",
- "used to identify blanknodes in Tripels"));
-
- options.add(new StringTupleListConfigOption("example", "example"));
- options.add(new StringTupleListConfigOption("replacePredicate", "rule for replacing predicates"));
- options.add(new StringTupleListConfigOption("replaceObject", "rule for replacing predicates"));
- options.add(new IntegerConfigOption("breakSuperClassRetrievalAfter", "stops a cyclic hierarchy after specified number of classes"));
- options.add(new IntegerConfigOption("numberOfInstancesUsedForRoleLearning", ""));
- options.add(new BooleanConfigOption("closeAfterRecursion", "gets all classes for all instances"));
-
-
-
- return options;
- }
-
- /*
- * @see org.dllearner.core.Component#applyConfigEntry(org.dllearner.core.ConfigEntry)
- */
- @Override
- @SuppressWarnings( { "unchecked" })
- public <T> void applyConfigEntry(ConfigEntry<T> entry) throws InvalidConfigOptionValueException {
- String option = entry.getOptionName();
- if (option.equals("url")) {
- String s = (String) entry.getValue();
- try {
- url = new URL(s);
- } catch (MalformedURLException e) {
- throw new InvalidConfigOptionValueException(entry.getOption(), entry.getValue(),
- "malformed URL " + s);
- }
- } else if (option.equals("host")) {
- host = (String) entry.getValue();
- } else if (option.equals("instances")) {
- instances = (Set<String>) entry.getValue();
- } else if (option.equals("recursionDepth")) {
- recursionDepth = (Integer) entry.getValue();
- } else if (option.equals("predList")) {
- predList = (Set<String>) entry.getValue();
- } else if (option.equals("objList")) {
- objList = (Set<String>) entry.getValue();
- //} else if (option.equals("classList")) {
- // classList = (Set<String>) entry.getValue();
- } else if (option.equals("predefinedEndpoint")) {
- predefinedEndpoint = (Integer) entry.getValue();
- } else if (option.equals("predefinedFilter")) {
- predefinedFilter = (Integer) entry.getValue();
- } else if (option.equals("format")) {
- format = (String) entry.getValue();
- } else if (option.equals("dumpToFile")) {
- dumpToFile = (Boolean) entry.getValue();
- } else if (option.equals("useLits")) {
- useLits = (Boolean) entry.getValue();
- } else if (option.equals("getAllSuperClasses")) {
- getAllSuperClasses = (Boolean) entry.getValue();
- } else if (option.equals("learnDomain")) {
- learnDomain = (Boolean) entry.getValue();
- }else if (option.equals("learnRange")) {
- learnRange = (Boolean) entry.getValue();
- } else if (option.equals("role")) {
- role = (String) entry.getValue();
- } else if (option.equals("blankNodeIdentifier")) {
- blankNodeIdentifier = (String) entry.getValue();
- } else if (option.equals("example")) {
- //System.out.println(entry.getValue());
- }else if (option.equals("replacePredicate")) {
- replacePredicate = (LinkedList)entry.getValue();
- }else if (option.equals("replaceObject")) {
- replaceObject = (LinkedList)entry.getValue();
- }else if (option.equals("breakSuperClassRetrievalAfter")) {
- breakSuperClassRetrievalAfter = (Integer) entry.getValue();
- }else if (option.equals("numberOfInstancesUsedForRoleLearning")) {
- numberOfInstancesUsedForRoleLearning = (Integer) entry.getValue();
- }else if (option.equals("closeAfterRecursion")) {
- closeAfterRecursion = (Boolean) entry.getValue();
- }
-
- }
-
- /*
- * (non-Javadoc)
- *
- * @see org.dllearner.core.Component#init()
- */
- @Override
- public void init() {
- System.out.println("SparqlModul: Collecting Ontology");
- // SparqlOntologyCollector oc=
- // new SparqlOntologyCollector(Datastructures.setToArray(instances),
- // numberOfRecursions, filterMode,
- // Datastructures.setToArray(predList),Datastructures.setToArray(
- // objList),Datastructures.setToArray(classList),format,url,useLits);
- Manager m = new Manager();
- SpecificSparqlEndpoint sse = null;
- SparqlQueryType sqt = null;
- // get Options for Manipulator
- Manipulator man = new Manipulator(blankNodeIdentifier,breakSuperClassRetrievalAfter,replacePredicate,replaceObject);
- HashMap<String, String> parameters = new HashMap<String, String>();
- parameters.put("default-graph-uri", "http://dbpedia.org");
- parameters.put("format", "application/sparql-results.xml");
-
- // get Options for endpoints
- if (predefinedEndpoint >= 1) {
- sse = PredefinedEndpoint.getEndpoint(predefinedEndpoint);
- } else {
- sse = new SpecificSparqlEndpoint(url, host, parameters);
- }
-
- // get Options for Filters
-
- if (predefinedFilter >= 1) {
- sqt = PredefinedFilter.getFilter(predefinedFilter);
-
- } else {
- sqt = new SparqlQueryType("forbid", objList, predList, useLits + "");
-
- }
- // give everything to the manager
- m.useConfiguration(sqt, sse, man, recursionDepth, getAllSuperClasses,closeAfterRecursion);
- try {
- String ont = "";
- //System.out.println(learnDomain);
- // used to learn a domain of a role
- if (learnDomain || learnRange) {
- Set<String> pos=new HashSet<String>();
- Set<String> neg=new HashSet<String>();
- if(learnDomain){
- pos = m.getDomainInstancesForRole(role);
- neg = m.getRangeInstancesForRole(role);
- }else if(learnRange){
- neg = m.getDomainInstancesForRole(role);
- pos = m.getRangeInstancesForRole(role);
- }
- //choose 30
-
-
- Set<String> tmp=new HashSet<String>();
- for(String one:pos){
- tmp.add(one);
- if(tmp.size()>=numberOfInstancesUsedForRoleLearning)break;
- }
- pos=tmp;
- System.out.println("Instances used: "+pos.size());
-
- tmp=new HashSet<String>();
- for(String one:neg){
- tmp.add(one);
- if(tmp.size()>=numberOfInstancesUsedForRoleLearning)break;
- }
- neg=tmp;
-
- instances=new HashSet<String>();
- instances.addAll(pos);
-
- instances.addAll(neg);
-
- for(String one:pos){
- System.out.println("+\""+one+"\"");
- }
- for(String one:neg){
- System.out.println("-\""+one+"\"");
- }
-
- /*Random r= new Random();
-
-
- Object[] arr=instances.toArray();
- while(instances.size()>=30){
-
- }*/
- // add the role to the filter(a solution is always EXISTS
- // role.TOP)
- m.addPredicateFilter(role);
- //System.out.println(instances);
- // THIS is a workaround
-
- }
- // the actual extraction is started here
- ont = m.extract(instances);
- System.out.println("Number of cached SPARQL queries: "+m.getConfiguration().numberOfCachedSparqlQueries);
- System.out.println("Number of uncached SPARQL queries: "+m.getConfiguration().numberOfUncachedSparqlQueries);
-
- System.out.println("Finished collecting Fragment");
-
- if (dumpToFile) {
- String filename = System.currentTimeMillis() + ".nt";
- String basedir = "cache" + File.separator;
- try {
- if (!new File(basedir).exists())
- new File(basedir).mkdir();
-
- FileWriter fw = new FileWriter(new File(basedir + filename), true);
- fw.write(ont);
- fw.flush();
- fw.close();
-
- dumpFile = (new File(basedir + filename)).toURI().toURL();
- } catch (Exception e) {
- e.printStackTrace();
- }
- }
- if (format.equals("KB")) {
- try {
- //kb = KBParser.parseKBFile(new StringReader(ont));
- kb=KBParser.parseKBFile(dumpFile);
- } catch (Exception e) {
- e.printStackTrace();
- }
- }
- } catch (Exception e) {
- e.printStackTrace();
- }
- System.out.println("SparqlModul: ****Finished");
- }
-
- /*
- * (non-Javadoc)
- *
- * @see org.dllearner.core.KnowledgeSource#toDIG()
- */
- @Override
- public String toDIG(URI kbURI) {
- if (format.equals("N-TRIPLES"))
- return JenaOWLDIGConverter.getTellsString(dumpFile, OntologyFormat.N_TRIPLES, kbURI);
- else
- return DIGConverter.getDIGString(kb, kbURI).toString();
- }
-
- /*
- * (non-Javadoc)
- *
- * @see org.dllearner.core.KnowledgeSource#export(java.io.File,
- * org.dllearner.core.OntologyFormat)
- */
- @Override
- public void export(File file, OntologyFormat format) throws OntologyFormatUnsupportedException {
- // currently no export functions implemented, so we just throw an
- // exception
- throw new OntologyFormatUnsupportedException("export", format);
- }
-
- public URL getURL() {
- return url;
- }
-
- public String[] getOntArray() {
- return ontArray;
- }
-
-
- /**
- * TODO SparqlOntologyCollector needs to be removed
- * @param label
- * @param limit
- */
- public void calculateSubjects(String label, int limit) {
- System.out.println("SparqlModul: Collecting Subjects");
- oldSparqlOntologyCollector oc = new oldSparqlOntologyCollector(url);
- try {
- subjects = oc.getSubjectsFromLabel(label, limit);
- } catch (IOException e) {
- subjects = new String[1];
- subjects[0] = "[Error]Sparql Endpoint could not be reached.";
- }
- System.out.println("SparqlModul: ****Finished");
- }
-
- /**
- * TODO SparqlOntologyCollector needs to be removed
- * @param subject
- */
- public void calculateTriples(String subject) {
- System.out.println("SparqlModul: Collecting Triples");
- oldSparqlOntologyCollector oc = new oldSparqlOntologyCollector(url);
- try {
- triples = oc.collectTriples(subject);
- } catch (IOException e) {
- triples = new String[1];
- triples[0] = "[Error]Sparql Endpoint could not be reached.";
- }
- System.out.println("SparqlModul: ****Finished");
- }
-
- /**
- * TODO SparqlOntologyCollector needs to be removed
- * @param concept
- */
- public void calculateConceptSubjects(String concept) {
- System.out.println("SparqlModul: Collecting Subjects");
- oldSparqlOntologyCollector oc = new oldSparqlOntologyCollector(url);
- try {
- conceptSubjects = oc.getSubjectsFromConcept(concept);
- } catch (IOException e) {
- conceptSubjects = new String[1];
- conceptSubjects[0] = "[Error]Sparql Endpoint could not be reached.";
- }
- System.out.println("SparqlModul: ****Finished");
- }
-
- public boolean subjectThreadIsRunning() {
- return subjectThreadRunning;
- }
-
- public void setSubjectThreadRunning(boolean bool) {
- subjectThreadRunning = bool;
- }
-
- public boolean triplesThreadIsRunning() {
- return triplesThreadRunning;
- }
-
- public void setTriplesThreadRunning(boolean bool) {
- triplesThreadRunning = bool;
- }
-
- public boolean conceptThreadIsRunning() {
- return conceptThreadRunning;
- }
-
- public void setConceptThreadRunning(boolean bool) {
- conceptThreadRunning = bool;
- }
-
- public String[] getSubjects() {
- return subjects;
- }
-
- public Thread getSubjectThread() {
- return subjectThread;
- }
-
- public void setSubjectThread(Thread subjectThread) {
- this.subjectThread = subjectThread;
- }
-
- public Thread getTriplesThread() {
- return triplesThread;
- }
-
- public void setTriplesThread(Thread triplesThread) {
- this.triplesThread = triplesThread;
- }
-
- public Thread getConceptThread() {
- return conceptThread;
- }
-
- public void setConceptThread(Thread conceptThread) {
- this.conceptThread = conceptThread;
- }
-
- public String[] getTriples() {
- return triples;
- }
-
- public String[] getConceptSubjects() {
- return conceptSubjects;
- }
-}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X