/*
* HLTTutorial.java
* Copyright (c) 2007- , DERI, National University of Ireland, Galway.
* Brian Davis, 12 December 2007
*/
package ie.deri.nlp;
/**
*
* @author bridav
*
*/
import gate.Annotation;
import gate.AnnotationSet;
import gate.Corpus;
import gate.Document;
import gate.Factory;
import gate.FeatureMap;
import gate.Gate;
import gate.ProcessingResource;
import gate.creole.SerialAnalyserController;
import gate.creole.Transducer;
import gate.creole.gazetteer.DefaultGazetteer;
import gate.util.GateException;
import gate.util.InvalidOffsetException;
import java.io.BufferedReader;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.FileReader;
import java.io.IOException;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.Iterator;
public class HLTTutorial {
public HLTTutorial() {
try {
initializeGate();
} catch (MalformedURLException ex) {
ex.printStackTrace();
} catch (GateException ex) {
ex.printStackTrace();
}
}
private final String fileSeparator = System.getProperty("file.separator");
public static String newline =("\n");
public static final String NS = "http://sw.deri.org/2007/12/fin/spec#";
public static final String FID_DATA_ORG = "http://sw.deri.org/2007/12/fin/inst/organisations/";
public static final String contextID="<http://sw.deri.org/2007/12/fin/sources/";
public String context;
public Document doc;
public static String contextData;
public static boolean hasURIContext=false;
private static int firstFile = 0;
public static String filename;
public SerialAnalyserController _annieController;
private void initializeGate() throws GateException, MalformedURLException {
System.out.println("SYSENV: "+System.getenv("GATE_HOME"));
System.setProperty("gate.home", System.getenv("GATE_HOME"));
System.setProperty("gate.site.config", System.getProperty("gate.home")
+ fileSeparator + "gate.xml");
System.setProperty("gate.user.config", System.getProperty("gate.site.config"));
Gate.init();
// register plugin directories
Gate.getCreoleRegister().registerDirectories(
new File(Gate.getGateHome().toString() + fileSeparator
+ "plugins" + fileSeparator + "ANNIE").toURL());
Gate.getCreoleRegister().registerDirectories(
new File(Gate.getGateHome().toString() + fileSeparator
+ "plugins" + fileSeparator + "Tools").toURL());
//Gate.getCreoleRegister().registerDirectories(
// new File(Gate.getGateHome().toString() + fileSeparator
// + "plugins" + fileSeparator + "Jape_Compiler").toURL());
_annieController = (SerialAnalyserController) Factory
.createResource("gate.creole.SerialAnalyserController", Factory
.newFeatureMap(), Factory.newFeatureMap(), "ANNIE_"
+ Gate.genSym());
FeatureMap params= Factory.newFeatureMap();
params.put("keepOriginalMarkupsAS",true);
ProcessingResource pr = (ProcessingResource) Factory.createResource("gate.creole.annotdelete.AnnotationDeletePR",
params);
_annieController.add(pr);
System.err.println("Loaded: Document Reset PR");
params = Factory.newFeatureMap();
pr = (ProcessingResource) Factory.createResource(
"gate.creole.tokeniser.SimpleTokeniser", params);
_annieController.add(pr);
System.err.println("Loaded: Tokeniser");
params = Factory.newFeatureMap();
pr = (ProcessingResource) Factory.createResource(
"gate.creole.splitter.SentenceSplitter", params);
_annieController.add(pr);
System.err.println("Loaded: Sentence Splitter");
params = Factory.newFeatureMap();
params.put("baseSentenceAnnotationType", "Sentence");
params.put("baseTokenAnnotationType", "Token");
params.put("outputAnnotationType", "Token");
pr = (ProcessingResource) Factory.createResource(
"gate.creole.POSTagger", params);
_annieController.add(pr);
System.err.println("Loaded: POS tagger");
params = Factory.newFeatureMap();
URL u=new File("ie/deri/nlp/LanguageResources/gazetteers/mylists.def").toURL();
params.put("listsURL", u);
DefaultGazetteer gazetteer = (DefaultGazetteer) Factory.createResource("gate.creole.gazetteer.DefaultGazetteer", params);
_annieController.add(gazetteer);
System.err.println("Loaded: Gazetteer");
u = new File("ie/deri/nlp/LanguageResources/JAPE/main.jape").toURL();
params = Factory.newFeatureMap();
params.put("grammarURL", u);
Transducer relationExtractionTransducer = (Transducer)Factory
.createResource("gate.creole.Transducer",params);
_annieController.add(relationExtractionTransducer);
System.err.println("Loaded: Relation Extraction JAPE grammars, using JAPE Transducer");
}
public static void main(String args[]) {
HLTTutorial _tutorial=new HLTTutorial();
try{
File contentDir = new File(args[0]);
File[] list = contentDir.listFiles();
_tutorial.extractRelations(list);
} catch (Exception ex) {
ex.printStackTrace();
}
}
private void extractRelations(File[]files) throws Exception {
// set corpus
//CorpusController application(CorpusController)PersistenceManager.loadObjectFromFile(gappFile);
Corpus corpus = Factory.newCorpus("BatchProcessApp Corpus");
_annieController.setCorpus(corpus);
for(int i = firstFile; i < files.length; i++) {
// load the document (using the specified encoding if one was given)
File file = files[i];
filename=file.getName();
System.out.println("currentfile count is"+i);
if(file.isFile() && file.length()!=0)
{
System.out.print("Processing document " + file + "...");
doc = Factory.newDocument(getContents(file));
corpus.add(doc);
_annieController.execute();
corpus.clear();
AnnotationSet defaultAnnotSet = doc.getAnnotations();
AnnotationSet companytoCompany1 = defaultAnnotSet.get("companytoCompanyRelation1");
if(!companytoCompany1.isEmpty())
{toRDF(companytoCompany1, 1);}
AnnotationSet companytoCompany1a = defaultAnnotSet.get("companytoCompanyRelation1a");
if(!companytoCompany1a.isEmpty())
{toRDF(companytoCompany1a, 7);}
AnnotationSet companytoCompany2 = defaultAnnotSet.get("companytoCompanyRelation2");
if(!companytoCompany2.isEmpty())
{toRDF(companytoCompany2, 2);}
doc.cleanup();
if (doc!=null) Factory.deleteResource(doc);
System.gc();
}//endif
else{
System.out.print("Document " + file + " is empty! Will ignore and continue to next file!");
}
}//endfor
/** **************************CLEANUP************************************* */
Factory.deleteResource(_annieController);
_annieController=null;
System.out.println("Processing Complete");
}// execute
/**
* Print to stdout the triple
* @param relationAnnotationSet
* @param content
*/
private void toRDF(AnnotationSet relationAnnotationSet, int code) {
//get the company
Iterator annoIter = relationAnnotationSet.iterator();
//iterate over annotations
while(annoIter.hasNext()){
//get the annotation
Annotation a = (Annotation) annoIter.next();
//
//debug
//System.out.println(a.toString());
FeatureMap features = a.getFeatures();
//get subject
Annotation subjectAnn = (Annotation)features.get("Company1");
String subject="empty";
try {
subject = doc.getContent().getContent(subjectAnn.getStartNode().getOffset(),subjectAnn.getEndNode().getOffset()).toString();
} catch (InvalidOffsetException e1) {
// TODO Auto-generated catch block
e1.printStackTrace();
}
Annotation predicateAnno=null;
String predicate="empty";
switch(code)
{
case 1:
predicateAnno = (Annotation)features.get("ownerOf");
if(predicateAnno!=null)
{predicate = predicateAnno.getType();}
case 2:
predicateAnno = (Annotation)features.get("customerOf");
if(predicateAnno!=null)
{predicate = predicateAnno.getType();}
case 3:
predicateAnno = (Annotation)features.get("supplierOf");
if(predicateAnno!=null)
{predicate = predicateAnno.getType();}
case 4:
if(predicateAnno!=null)
{predicate = predicateAnno.getType();}
predicateAnno = (Annotation)features.get("lenderOf");
case 5:
predicateAnno = (Annotation)features.get("competitorOf");
if(predicateAnno!=null)
{predicate = predicateAnno.getType();}
case 7:
predicateAnno = (Annotation)features.get("bidFor");
if(predicateAnno!=null)
{predicate = predicateAnno.getType();}
}
//System.out.println("\n#"+predicateAnno+"\n");
//String predicate = predicateAnno.getType();
//get objectAnnoyation
Annotation objectAnn = (Annotation)features.get("Company2");
String object="empty";
try {
object = doc.getContent().getContent(objectAnn.getStartNode().getOffset(),objectAnn.getEndNode().getOffset()).toString();
} catch (InvalidOffsetException e1) {
// TODO Auto-generated catch block
e1.printStackTrace();
}
if(subject.equals("empty")|| predicate.equals("empty")|| object.equals("empty"))
{
System.err.println("Null triples + "+subject+" "+predicate+" "+object );
}
if(subject.equals(object))
{
System.err.println("Duplicate triples + "+subject+" "+predicate+" "+object );
}
else
{
String subjectID=getOrgNameSpace(subject);
String predicateID=getPredNameSpace(predicate);
String objectID=getOrgNameSpace(object);
if(subjectID.equals("empty"))
{System.err.println("Error:"+ subject);}
else if(objectID.equals("empty"))
{System.err.println(" Error could not find name:"+ object);}
else{
System.out.print(subjectID+" "+predicateID+" "+objectID+" ."+newline);
System.out.println("**********EndofTriple***********");
}
}
}
}
static String getOrgNameSpace(String name){
name = name.toLowerCase();
name = stripNonAlphanumeric(name);
return "<"+FID_DATA_ORG+ name+">";
}
private static String getPredNameSpace(String name){
if (name.equals("ownerOfPassive"))
{name="ownerOf";}
name = name.toLowerCase();
name = stripNonAlphanumeric(name);
return "<"+NS+name+">";
}
private static String stripNonAlphanumeric(String s) {
return s.replaceAll("[^a-zA-Z0-9]", "");
}
/**
* Depending on the performance of Gate we should maybe think about a hack
*
* -big file: breaking into chunk of 1000 lines
*
*
* @param aFile
* @return
*/
public String getContents(File aFile) {
// ...checks on aFile are elided
StringBuffer contents = new StringBuffer();
// declared here only to make visible to finally clause
BufferedReader input = null;
try {
// use buffering, reading one line at a time
// FileReader always assumes default encoding is OK!
input = new BufferedReader(new FileReader(aFile));
String line = null; // not declared within while loop
/*
* readLine is a bit quirky : it returns the content of a line MINUS
* the newline. it returns null only for the END of the stream. it
* returns an empty String if two newlines appear in a row.
*/
while ((line = input.readLine()) != null) {
contents.append(line);
contents.append(System.getProperty("line.separator"));
}
} catch (FileNotFoundException ex) {
ex.printStackTrace();
} catch (IOException ex) {
ex.printStackTrace();
} finally {
try {
if (input != null) {
// flush and close both "input" and its underlying
// FileReader
input.close();
}
} catch (IOException ex) {
ex.printStackTrace();
}
}
return contents.toString();
}
} // class
