@Override
public void execute() throws ExecutionException {
initBeforeExecute();
AnnotationSet tokensAndDependenciesAS = inputAS;
TreeIndex index =
new GateAwareTreeIndex(
tokensAndDependenciesAS.get(null, Utils.setFromArray(new String[] {"args"})));
QueryData data =
new QueryData(index, new GateAnnotationsNodeAttributes(tokensAndDependenciesAS));
Iterable<QueryMatch> results = queryObject.evaluate(data);
int queryMatchOrd = 0;
for (QueryMatch result : results) {
queryMatchOrd++;
for (NodeMatch match : result.getMatchingNodes()) {
String name = match.getQueryNode().getName();
if (name != null) {
Annotation matchingAnnot = tokensAndDependenciesAS.get(match.getNodeId());
FeatureMap fm = Factory.newFeatureMap();
fm.put("matchingNodeId", match.getNodeId());
fm.put(
"queryMatchId",
String.format("%s_%03d", buildQueryStringHash(getQueryString()), queryMatchOrd));
outputAS.add(matchingAnnot.getStartNode(), matchingAnnot.getEndNode(), name, fm);
}
}
}
}
protected AnnotationDiffer calculateDocumentDiff(Document document, String annotTypeName) {
AnnotationSet responsesIter = responseAS.get(annotTypeName);
if (getKeyAnnotationsAreInDocumentFeatures()) {
return DocumentFeaturesDiff.computeDiffWithDocFeatures(document, featureNames, responsesIter);
}
AnnotationSet keysIter = keyAS.get(annotTypeName);
AnnotationDiffer differ = new AnnotationDiffer();
differ.setSignificantFeaturesSet(new HashSet<String>(featureNames));
differ.calculateDiff(keysIter, responsesIter); // compare
return differ;
}
public void execute() throws ExecutionException {
// get all the annotations we need from the input AS
AnnotationSet inputAS =
inputAnnotationSet == null || inputAnnotationSet.trim().length() == 0
? document.getAnnotations()
: document.getAnnotations(inputAnnotationSet);
AnnotationSet outputAS =
outputAnnotationSet == null || outputAnnotationSet.trim().length() == 0
? document.getAnnotations()
: document.getAnnotations(outputAnnotationSet);
// no spans?
if (getSpanAnnotationType() == null | getSpanAnnotationType().equals("")) {
AnnotationSet inputs = inputAS.get(inputAnnotationType);
List<Annotation> list = new ArrayList<Annotation>();
list.addAll(inputs);
Collections.sort(list, new OffsetComparator());
// use window or normal
if (window == -1) generateNGrams(list, outputAS);
else generateNGramsOverWindow(list, outputAS);
} else {
// use the spans
AnnotationSet spans = inputAS.get(getSpanAnnotationType());
Iterator spaniter = spans.iterator();
while (spaniter.hasNext()) {
Annotation span = (Annotation) spaniter.next();
AnnotationSet inputs =
inputAS.get(
inputAnnotationType,
span.getStartNode().getOffset(),
span.getEndNode().getOffset());
List<Annotation> list = new ArrayList<Annotation>();
list.addAll(inputs);
Collections.sort(list, new OffsetComparator());
if (window == -1) generateNGrams(list, outputAS);
else generateNGramsOverWindow(list, outputAS);
}
}
}
/** Returns a list of annotations to be added to the Behemoth document from the GATE one * */
private List<com.digitalpebble.behemoth.Annotation> convertGATEAnnotationsToBehemoth(
AnnotationSet GATEAnnotionSet, com.digitalpebble.behemoth.BehemothDocument behemoth) {
List<com.digitalpebble.behemoth.Annotation> beheannotations =
new ArrayList<com.digitalpebble.behemoth.Annotation>();
AnnotationSet resultAS = GATEAnnotionSet.get(filters.getTypes());
// sort the GATE annotations
List<gate.Annotation> annotationList = new ArrayList<gate.Annotation>(resultAS);
Collections.sort(annotationList, new OffsetComparator());
Iterator<gate.Annotation> inputASIter = annotationList.iterator();
while (inputASIter.hasNext()) {
gate.Annotation source = inputASIter.next();
com.digitalpebble.behemoth.Annotation target = new com.digitalpebble.behemoth.Annotation();
target.setType(source.getType());
target.setStart(source.getStartNode().getOffset().longValue());
target.setEnd(source.getEndNode().getOffset().longValue());
// now do the features
// is the type listed?
Set<String> expectedfeatnames = filters.getFeatureFilters().get(source.getType());
if (expectedfeatnames != null) {
Iterator featurenames = source.getFeatures().keySet().iterator();
while (featurenames.hasNext()) {
// cast the feature name to a string which will be right in
// 99% of cases
String featurename = featurenames.next().toString();
// if this feature name is not wanted just ignore it
if (expectedfeatnames.contains(featurename) == false) continue;
// we know that we want to keep this feature
// let's see what the best way of representing the value
// would be
// TODO later => find a better way of mapping when not a
// string
Object originalvalue = source.getFeatures().get(featurename);
if (originalvalue == null) originalvalue = "null";
target.getFeatures().put(featurename, originalvalue.toString());
}
}
beheannotations.add(target);
}
return beheannotations;
}
@Override
public void execute() throws ExecutionException {
Document doc = getDocument();
AnnotationSet as = doc.getAnnotations(getAnnotationSetName());
AnnotationSet tocs = as.get(getTokenAnnotationTypeName());
try {
for (Annotation t : tocs) {
String content = Utils.stringFor(doc, t);
String val = getOrthographyValue(content);
if (val != null) t.getFeatures().put("orth", val);
}
} catch (Exception e) {
throw new ExecutionException(e);
}
}
protected void doExecute(Document theDocument) throws ExecutionException {
interrupted = false;
if (theDocument == null) {
throw new ExecutionException("No document to process!");
}
AnnotationSet outputAS = theDocument.getAnnotations(getOutputAnnotationSet());
if (containingType == null || containingType.isEmpty()) {
annotateText(document, outputAS, 0, document.getContent().size());
} else {
AnnotationSet inputAS = null;
if (inputASName == null || inputASName.isEmpty()) {
inputAS = theDocument.getAnnotations();
} else {
inputAS = theDocument.getAnnotations(inputASName);
}
AnnotationSet containingAnns = inputAS.get(containingType);
for (Annotation containingAnn : containingAnns) {
annotateText(
document, outputAS, gate.Utils.start(containingAnn), gate.Utils.end(containingAnn));
}
}
}
/**
* Rename annotation
*
* @param outputAS output annotation set
* @param oldType old annotation name
* @param newType new annotation name
*/
private void renameAnnotations(AnnotationSet outputAS, String oldType, String newType) {
AnnotationSet tmpAnatomyAS = outputAS.get(oldType);
for (Annotation tmpAnn : tmpAnatomyAS) {
Long startOffset = tmpAnn.getStartNode().getOffset();
Long endOffset = tmpAnn.getEndNode().getOffset();
AnnotationSet existingAS = outputAS.getCovering(newType, startOffset, endOffset);
// If we've already got an annotation of the same name in the same place, don't add a new one
// just delete the old one
if (existingAS.isEmpty()) {
FeatureMap tmpFm = tmpAnn.getFeatures();
FeatureMap fm = Factory.newFeatureMap();
fm.putAll(tmpFm);
try {
outputAS.add(startOffset, endOffset, newType, fm);
outputAS.remove(tmpAnn);
} catch (InvalidOffsetException ie) {
// shouldn't happen
}
} else {
outputAS.remove(tmpAnn);
}
}
}
/**
* @param inputAS input annotation set
* @param outputAS output annotation set
* @param term String matched
* @param startOffset match start offset
* @param endOffset match end offset
*/
private void addLookup(
AnnotationSet inputAS,
AnnotationSet outputAS,
String term,
String outputASType,
Long startOffset,
Long endOffset,
boolean useNounChunk) {
if (useNounChunk && nounChunkType != null && !nounChunkType.isEmpty()) {
AnnotationSet nounChunkAS = inputAS.getCovering(nounChunkType, startOffset, endOffset);
if (!nounChunkAS.isEmpty()) {
startOffset = nounChunkAS.firstNode().getOffset();
endOffset = nounChunkAS.lastNode().getOffset();
}
}
try {
AnnotationSet diseaseAS = inputAS.get(outputASType, startOffset, endOffset);
if (diseaseAS.isEmpty()) {
FeatureMap fm = Factory.newFeatureMap();
fm.put("match", term);
outputAS.add(startOffset, endOffset, outputASType, fm);
} else {
Annotation disease = diseaseAS.iterator().next();
FeatureMap fm = disease.getFeatures();
String meta = (String) fm.get("match");
if (meta != null) {
meta = meta + " " + term;
}
fm.put("match", meta);
}
} catch (InvalidOffsetException ie) {
// shouldn't happen
gate.util.Err.println(ie);
}
}
@Override
public void execute() throws ExecutionException {
interrupted = false;
// quit if setup failed
if (gracefulExit) {
gracefulExit("Plugin was not initialised correctly. Exiting gracefully ... ");
return;
}
AnnotationSet inputAS =
(inputASName == null || inputASName.trim().length() == 0)
? document.getAnnotations()
: document.getAnnotations(inputASName);
AnnotationSet outputAS =
(outputASName == null || outputASName.trim().length() == 0)
? document.getAnnotations()
: document.getAnnotations(outputASName);
AnnotationSet sentenceAS = null;
if (sentenceType != null && !sentenceType.isEmpty()) {
sentenceAS = inputAS.get(sentenceType);
}
// Document content
String docContent = document.getContent().toString();
int docLen = docContent.length();
// For matching purposes replace all whitespace characters with a single space
docContent = docContent.replaceAll("[\\s\\xA0\\u2007\\u202F]", " ");
fireStatusChanged("Locating anatomy, disease and procedure mentions in " + document.getName());
fireProgressChanged(0);
if (sentenceAS != null) {
for (Annotation sentence : sentenceAS) {
Long sentStartOffset = sentence.getStartNode().getOffset();
Long sentEndOffset = sentence.getEndNode().getOffset();
// Converting the sentence to lower case prevents the need to use case-insenstive regex
// matching, which should give a small performance boost
String sentenceContent =
docContent
.substring(sentStartOffset.intValue(), sentEndOffset.intValue())
.toLowerCase(Locale.ENGLISH);
if (diseaseType != null && !diseaseType.isEmpty()) {
doMatch(
patternMap.get("disease_suffix"),
sentenceContent,
inputAS,
outputAS,
"suffDisease",
sentStartOffset,
docLen);
doMatch(
patternMap.get("disease_abbrevs"),
sentenceContent,
inputAS,
outputAS,
"preDisease",
sentStartOffset,
docLen);
doMatch(
patternMap.get("disease_named_syndrome"),
sentenceContent,
inputAS,
outputAS,
"namedDisease",
sentStartOffset,
docLen);
doMatch(
patternMap.get("disease_sense"),
sentenceContent,
inputAS,
outputAS,
"tmpDiseaseSense",
sentStartOffset,
docLen);
doMatch(
patternMap.get("disease_sense_context"),
sentenceContent,
inputAS,
outputAS,
"tmpDiseaseSenseContext",
sentStartOffset,
docLen);
doMatch(
patternMap.get("disease_generic_context"),
sentenceContent,
inputAS,
outputAS,
"poDisease",
sentStartOffset,
docLen);
doMatch(
patternMap.get("disease_anatomy_context"),
sentenceContent,
inputAS,
outputAS,
"tmpDisease",
sentStartOffset,
docLen);
}
if (procedureType != null && !procedureType.isEmpty()) {
doMatch(
patternMap.get("procedure_suffix"),
sentenceContent,
inputAS,
outputAS,
"poProcedure",
sentStartOffset,
docLen);
doMatch(
patternMap.get("procedure_key"),
sentenceContent,
inputAS,
outputAS,
"poProcedure",
sentStartOffset,
docLen);
doMatch(
patternMap.get("procedure_anatomy_context"),
sentenceContent,
inputAS,
outputAS,
"tmpProcedure",
sentStartOffset,
docLen);
}
if (symptomType != null && !symptomType.isEmpty()) {
doMatch(
patternMap.get("symptom_key"),
sentenceContent,
inputAS,
outputAS,
"poSymptom",
sentStartOffset,
docLen);
}
if (testType != null && !testType.isEmpty()) {
doMatch(
patternMap.get("test_key"),
sentenceContent,
inputAS,
outputAS,
"poTest",
sentStartOffset,
docLen);
}
if (anatomyType != null && !anatomyType.isEmpty()) {
doMatch(
patternMap.get("anatomy_suffix_adjective"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatSuffAdj",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_suffix"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatSuff",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_prefix"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatPre",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_position"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatPos",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_space_region_junction"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatSpace",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_part_adjective"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatAdj",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_latin_noun"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatLatin",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_muscle"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatMuscle",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_part"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatPart",
sentStartOffset,
docLen);
doMatch(
patternMap.get("anatomy_fluid"),
sentenceContent,
inputAS,
outputAS,
"tmpAnatFluid",
sentStartOffset,
docLen);
}
}
// Run JAPE transducer to clean up the output
fireStatusChanged(
"Processing anatomical, disease and procedure mentions in " + document.getName());
try {
japeTransducer.setDocument(document);
japeTransducer.setInputASName(inputASName);
japeTransducer.setOutputASName(outputASName);
japeTransducer.addProgressListener(this);
japeTransducer.execute();
} catch (ExecutionException re) {
gate.util.Err.println("Unable to run " + japeURL);
gracefulExit = true;
} finally {
japeTransducer.setDocument(null);
}
// rename temporary annotations
if (!debug) {
renameAnnotations(outputAS, "tmpAnatomicalTerm", anatomyType);
renameAnnotations(outputAS, "suffDisease", diseaseType);
renameAnnotations(outputAS, "poDisease", diseaseType);
renameAnnotations(outputAS, "preDisease", diseaseType);
renameAnnotations(outputAS, "poProcedure", procedureType);
renameAnnotations(outputAS, "poSymptom", symptomType);
renameAnnotations(outputAS, "poTest", testType);
}
} else {
gracefulExit("No sentences to process!");
}
// want list of disease key words plus symptoms such as oedema? or just diseases
fireProcessFinished();
} // end execute()
/**
* The main entry point. First we parse the command line options (see usage() method for details),
* then we take all remaining command line parameters to be file names to process. Each file is
* loaded, processed using the application and the results written to the output file
* (inputFile.out.xml).
*/
public static void main(String[] args) throws Exception {
parseCommandLine(args);
// initialise GATE - this must be done before calling any GATE APIs
Gate.init();
// load the saved application
CorpusController application =
(CorpusController) PersistenceManager.loadObjectFromFile(gappFile);
// Create a Corpus to use. We recycle the same Corpus object for each
// iteration. The string parameter to newCorpus() is simply the
// GATE-internal name to use for the corpus. It has no particular
// significance.
ArrayList<String> files = getFilesFromDir(inputDir);
gate.Corpus corpus = createCorpus(files);
// Corpus corpus = Factory.newCorpus("BatchProcessApp Corpus");
application.setCorpus(corpus);
System.out.println("Processing " + files.size() + " files");
// process the files one by one
for (int i = 0; i < files.size(); i++) {
// load the document (using the specified encoding if one was given)
File docFile = new File(files.get(i));
System.out.print("Processing document " + docFile + " (" + i + ") ...");
Document doc = Factory.newDocument(docFile.toURL(), encoding);
// put the document in the corpus
corpus.add(doc);
// run the application
application.execute();
// remove the document from the corpus again
corpus.clear();
String docXMLString = null;
// if we want to just write out specific annotation types, we must
// extract the annotations into a Set
if (annotTypesToWrite != null) {
// Create a temporary Set to hold the annotations we wish to write out
Set annotationsToWrite = new HashSet();
// we only extract annotations from the default (unnamed) AnnotationSet
// in this example
AnnotationSet defaultAnnots = doc.getAnnotations("Output");
Iterator annotTypesIt = annotTypesToWrite.iterator();
while (annotTypesIt.hasNext()) {
// extract all the annotations of each requested type and add them to
// the temporary set
AnnotationSet annotsOfThisType = defaultAnnots.get((String) annotTypesIt.next());
if (annotsOfThisType != null) {
annotationsToWrite.addAll(annotsOfThisType);
}
}
// create the XML string using these annotations
docXMLString = doc.toXml(annotationsToWrite, true);
}
// otherwise, just write out the whole document as GateXML
else {
docXMLString = doc.toXml();
}
// Release the document, as it is no longer needed
Factory.deleteResource(doc);
// output the XML to <inputFile>.out.xml
System.out.println("Writing file " + docFile.getName());
String outputFileName = docFile.getName() + ".out.xml";
// File outputFile = new File(docFile.getParentFile(), outputFileName);
File outputFile = new File(new File(outputDir).getAbsolutePath(), outputFileName);
// Write output files using the same encoding as the original
FileOutputStream fos = new FileOutputStream(outputFile);
BufferedOutputStream bos = new BufferedOutputStream(fos);
OutputStreamWriter out;
if (encoding == null) {
out = new OutputStreamWriter(bos);
} else {
out = new OutputStreamWriter(bos, encoding);
}
out.write(docXMLString);
out.close();
System.out.println("done");
} // for each file
System.out.println("All done");
} // void main(String[] args)
private RetObj ProcessRecords() throws Exception {
// Create a Corpus to use. We recycle the same Corpus object for each
// iteration.
Corpus corpus = Factory.newCorpus("BatchProcessApp Corpus");
this.application.setCorpus(corpus);
// object for returned data
List<String> processedlines = new ArrayList<String>();
List<String> processedText = new ArrayList<String>();
for (int record_num = 0; record_num < this.recs.size(); ++record_num) {
/*if( record_num % Math.ceil(((double) this.recs.size())/10.0) == 0)
System.out.println("Thread " + this.threadID + ": "+ ((int) ((double)record_num)/((double) this.recs.size())*100.0 ) +"% complete.");
*/
// first, split title from body and get embedded age in title..
String title_age = "-1";
String sep = "..THIS IS MY SEPARATION STRING..";
String title = "";
String body = this.recs.get(record_num);
Boolean trimmed = false;
int age_end = body.indexOf(",> ");
if (age_end >= 0 && age_end < body.length()) {
int age_start = body.lastIndexOf("-", age_end);
if (age_start >= 0 && age_start < age_end) {
title_age = body.substring(age_start + 1, age_end).trim();
if (!isInteger(title_age)) title_age = "-1";
else {
title = body.substring(0, age_start);
body = body.substring(age_end + 2, body.length());
body = title + sep + body;
trimmed = true;
}
}
if (!trimmed) {
title = body.substring(0, age_end);
body = body.substring(age_end + 2, body.length());
body = title + sep + body;
trimmed = true;
}
}
// --------------------
org.jsoup.nodes.Document htmldoc =
Jsoup.parseBodyFragment(body.replaceAll("COMMA_GOES_HERE", ","));
Elements links = htmldoc.select("a[href]");
Elements media = htmldoc.select("[src]");
Elements imports = htmldoc.select("link[href]");
processedText.add(htmldoc.text().replace(sep, " "));
Document doc = Factory.newDocument(htmldoc.text());
// put the document in the corpus
corpus.add(doc);
// run the application
this.application.execute();
// remove the document from the corpus again
corpus.clear();
// extract annotations
String line = "";
AnnotationSet Annots = doc.getAnnotations("");
Integer FirstPersonCount = 0, ThirdPersonCount = 0;
AnnotationSet FirstPerson = Annots.get("FirstPerson");
if (FirstPerson != null) FirstPersonCount = FirstPerson.size();
AnnotationSet ThirdPerson = Annots.get("ThirdPerson");
if (ThirdPerson != null) ThirdPersonCount = ThirdPerson.size();
line += FirstPersonCount.toString() + "," + ThirdPersonCount.toString() + ",";
AnnotationSet Names = Annots.get("Name");
if (Names == null || Names.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = Names.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("name");
if (Feat != null) line += Feat.toString();
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet Age = Annots.get("Age");
if (Age == null || Age.size() < 1) line += title_age + ",";
else {
Iterator<Annotation> Iter = Age.inDocumentOrder().iterator();
line += title_age + ";";
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("age");
if (Feat != null) line += Feat.toString();
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet Cost = Annots.get("Cost");
if (Cost == null || Cost.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = Cost.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("value");
if (Feat != null) line += Feat.toString();
else line += "none";
line += "/";
Feat = Ann.getFeatures().get("target_value");
if (Feat != null) line += Feat.toString();
else line += "none";
line += "/";
Feat = Ann.getFeatures().get("target_type");
if (Feat != null) line += Feat.toString();
else line += "none";
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet height = Annots.get("height");
if (height == null || height.size() < 1) line += ",,";
else {
String ft = "";
String inch = "";
Iterator<Annotation> Iter = height.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("feet");
if (Feat != null) ft += Feat.toString();
else ft += "none";
Feat = Ann.getFeatures().get("inches");
if (Feat != null) inch += Feat.toString();
else inch += "none";
if (Iter.hasNext()) {
ft += ";";
inch += ";";
}
}
line += ft + "," + inch + ",";
}
AnnotationSet weight = Annots.get("weight");
if (weight == null || weight.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = weight.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("pounds");
if (Feat != null) line += Feat.toString();
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet measurement = Annots.get("measurement");
if (measurement == null || measurement.size() < 1) line += ",,,,";
else {
String cup = "";
String chest = "";
String waist = "";
String hip = "";
Iterator<Annotation> Iter = measurement.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("cup");
if (Feat != null) cup += Feat.toString();
else cup += "none";
Feat = Ann.getFeatures().get("chest");
if (Feat != null) chest += Feat.toString();
else chest += "none";
Feat = Ann.getFeatures().get("waist");
if (Feat != null) waist += Feat.toString();
else waist += "none";
Feat = Ann.getFeatures().get("hip");
if (Feat != null) hip += Feat.toString();
else hip += "none";
if (Iter.hasNext()) {
cup += ";";
chest += ";";
waist += ";";
hip += ";";
}
}
line += cup + "," + chest + "," + waist + "," + hip + ",";
}
AnnotationSet Ethnicity = Annots.get("Ethnicity");
if (Ethnicity == null || Ethnicity.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = Ethnicity.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("ethnicity");
if (Feat != null)
line += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ");
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet SkinColor = Annots.get("SkinColor");
if (SkinColor == null || SkinColor.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = SkinColor.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("color");
if (Feat != null)
line += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ");
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet EyeColor = Annots.get("EyeColor");
if (EyeColor == null || EyeColor.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = EyeColor.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("color");
if (Feat != null)
line += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ");
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet HairColor = Annots.get("HairColor");
if (HairColor == null || HairColor.size() < 1) line += ",";
else {
Iterator<Annotation> Iter = HairColor.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("color");
if (Feat != null)
line += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ");
if (Iter.hasNext()) line += ";";
}
line += ",";
}
AnnotationSet Restriction = Annots.get("Restriction");
if (Restriction == null || Restriction.size() < 1) line += ",,,";
else {
String type = "";
String ethnicity = "";
String age = "";
Iterator<Annotation> Iter = Restriction.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("type");
if (Feat != null) type += Feat.toString();
else type += "none";
Feat = Ann.getFeatures().get("ethnicity");
if (Feat != null)
ethnicity += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ");
else ethnicity += "none";
Feat = Ann.getFeatures().get("age");
if (Feat != null) age += Feat.toString();
else age += "none";
if (Iter.hasNext()) {
type += ";";
ethnicity += ";";
age += ";";
}
}
line += type + "," + ethnicity + "," + age + ",";
}
AnnotationSet Phone = Annots.get("PhoneNumber");
if (Phone == null || Phone.size() < 1) line += ",,,";
else {
String value = "";
String state = "";
String city = "";
Iterator<Annotation> Iter = Phone.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("value");
if (Feat != null) value += Feat.toString();
else value += "none";
Feat = Ann.getFeatures().get("state");
if (Feat != null) state += Feat.toString();
else state += "none";
Feat = Ann.getFeatures().get("area");
if (Feat != null)
city += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ");
else city += "none";
if (Iter.hasNext()) {
value += ";";
state += ";";
city += ";";
}
}
line += value + "," + state + "," + city + ",";
}
String Emails = "";
AnnotationSet Email = Annots.get("Email");
if (Email == null || Email.size() < 1) Emails = "";
else {
Iterator<Annotation> Iter = Email.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("email");
if (Feat != null)
Emails += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ") + ";";
}
}
if (links != null) {
for (Element l : links) {
String href = l.attr("abs:href");
if (href == null) continue;
if (href.length() > 7 && href.substring(0, 7).toLowerCase().equals("mailto:")) {
Emails +=
href.substring(7, href.length()).replaceAll(",", " ").replaceAll(";", " ") + ";";
}
}
}
if (Emails.length() > 0 && Emails.substring(Emails.length() - 1, Emails.length()).equals(";"))
Emails = Emails.substring(0, Emails.length() - 1);
line += Emails + ",";
String Urls = "";
AnnotationSet Url = Annots.get("Url");
if (Url == null || Url.size() < 1) Urls = "";
else {
Iterator<Annotation> Iter = Url.inDocumentOrder().iterator();
while (Iter.hasNext()) {
Annotation Ann = Iter.next();
Object Feat = Ann.getFeatures().get("url");
if (Feat != null)
Urls += Feat.toString().toLowerCase().replaceAll(",", " ").replaceAll(";", " ") + ";";
}
}
if (links != null) {
for (Element l : links) {
String href = l.attr("abs:href");
if (href == null) continue;
if (href.length() <= 7 || !href.substring(0, 7).toLowerCase().equals("mailto:")) {
Urls += href.replaceAll(",", " ").replaceAll(";", " ") + ";";
}
}
}
if (imports != null) {
for (Element l : imports) {
String href = l.attr("abs:href");
if (href == null) continue;
Urls += href.replaceAll(",", " ").replaceAll(";", " ") + ";";
}
}
if (Urls.length() > 0 && Urls.substring(Urls.length() - 1, Urls.length()).equals(";"))
Urls = Urls.substring(0, Urls.length() - 1);
line += Urls + ",";
String Medias = "";
if (media != null) {
for (Element l : media) {
String src = l.attr("abs:src");
if (src == null) continue;
Medias += src.replaceAll(",", " ").replaceAll(";", " ") + ";";
}
}
if (Medias.length() > 0 && Medias.substring(Medias.length() - 1, Medias.length()).equals(";"))
Medias = Medias.substring(0, Medias.length() - 1);
line += Medias;
processedlines.add(line);
// Release the document, as it is no longer needed
Factory.deleteResource(doc);
}
Factory.deleteResource(corpus);
RetObj out = new RetObj(processedlines, processedText);
return out;
}
/**
* Run from the command-line, with a list of URLs as argument.
*
* <p><B>NOTE:</B><br>
* This code will run with all the documents in memory - if you want to unload each from memory
* after use, add code to store the corpus in a DataStore.
*/
public static void main(String args[]) throws GateException, IOException {
// initialise the GATE library
Out.prln("Initialising GATE...");
Gate.init();
Out.prln("...GATE initialised");
// initialise ANNIE (this may take several minutes)
StandAloneAnnie annie = new StandAloneAnnie();
annie.initAnnie();
// create a GATE corpus and add a document for each command-line
// argument
Corpus corpus = Factory.newCorpus("StandAloneAnnie corpus");
for (int i = 0; i < args.length; i++) {
URL u = new URL(args[i]);
FeatureMap params = Factory.newFeatureMap();
params.put("sourceUrl", u);
params.put("preserveOriginalContent", new Boolean(true));
params.put("collectRepositioningInfo", new Boolean(true));
Out.prln("Creating doc for " + u);
Document doc = (Document) Factory.createResource("gate.corpora.DocumentImpl", params);
corpus.add(doc);
} // for each of args
// tell the pipeline about the corpus and run it
annie.setCorpus(corpus);
annie.execute();
// for each document, get an XML document with the
// person and location names added
Iterator iter = corpus.iterator();
int count = 0;
String startTagPart_1 = "<span GateID=\"";
String startTagPart_2 = "\" title=\"";
String startTagPart_3 = "\" style=\"background:Red;\">";
String endTag = "</span>";
while (iter.hasNext()) {
Document doc = (Document) iter.next();
AnnotationSet defaultAnnotSet = doc.getAnnotations();
Set annotTypesRequired = new HashSet();
annotTypesRequired.add("Person");
annotTypesRequired.add("Location");
Set<Annotation> peopleAndPlaces =
new HashSet<Annotation>(defaultAnnotSet.get(annotTypesRequired));
FeatureMap features = doc.getFeatures();
String originalContent =
(String) features.get(GateConstants.ORIGINAL_DOCUMENT_CONTENT_FEATURE_NAME);
RepositioningInfo info =
(RepositioningInfo) features.get(GateConstants.DOCUMENT_REPOSITIONING_INFO_FEATURE_NAME);
++count;
File file = new File("StANNIE_" + count + ".HTML");
Out.prln("File name: '" + file.getAbsolutePath() + "'");
if (originalContent != null && info != null) {
Out.prln("OrigContent and reposInfo existing. Generate file...");
Iterator it = peopleAndPlaces.iterator();
Annotation currAnnot;
SortedAnnotationList sortedAnnotations = new SortedAnnotationList();
while (it.hasNext()) {
currAnnot = (Annotation) it.next();
sortedAnnotations.addSortedExclusive(currAnnot);
} // while
StringBuffer editableContent = new StringBuffer(originalContent);
long insertPositionEnd;
long insertPositionStart;
// insert anotation tags backward
Out.prln("Unsorted annotations count: " + peopleAndPlaces.size());
Out.prln("Sorted annotations count: " + sortedAnnotations.size());
for (int i = sortedAnnotations.size() - 1; i >= 0; --i) {
currAnnot = (Annotation) sortedAnnotations.get(i);
insertPositionStart = currAnnot.getStartNode().getOffset().longValue();
insertPositionStart = info.getOriginalPos(insertPositionStart);
insertPositionEnd = currAnnot.getEndNode().getOffset().longValue();
insertPositionEnd = info.getOriginalPos(insertPositionEnd, true);
if (insertPositionEnd != -1 && insertPositionStart != -1) {
editableContent.insert((int) insertPositionEnd, endTag);
editableContent.insert((int) insertPositionStart, startTagPart_3);
editableContent.insert((int) insertPositionStart, currAnnot.getType());
editableContent.insert((int) insertPositionStart, startTagPart_2);
editableContent.insert((int) insertPositionStart, currAnnot.getId().toString());
editableContent.insert((int) insertPositionStart, startTagPart_1);
} // if
} // for
FileWriter writer = new FileWriter(file);
writer.write(editableContent.toString());
writer.close();
} // if - should generate
else if (originalContent != null) {
Out.prln("OrigContent existing. Generate file...");
Iterator it = peopleAndPlaces.iterator();
Annotation currAnnot;
SortedAnnotationList sortedAnnotations = new SortedAnnotationList();
while (it.hasNext()) {
currAnnot = (Annotation) it.next();
sortedAnnotations.addSortedExclusive(currAnnot);
} // while
StringBuffer editableContent = new StringBuffer(originalContent);
long insertPositionEnd;
long insertPositionStart;
// insert anotation tags backward
Out.prln("Unsorted annotations count: " + peopleAndPlaces.size());
Out.prln("Sorted annotations count: " + sortedAnnotations.size());
for (int i = sortedAnnotations.size() - 1; i >= 0; --i) {
currAnnot = (Annotation) sortedAnnotations.get(i);
insertPositionStart = currAnnot.getStartNode().getOffset().longValue();
insertPositionEnd = currAnnot.getEndNode().getOffset().longValue();
if (insertPositionEnd != -1 && insertPositionStart != -1) {
editableContent.insert((int) insertPositionEnd, endTag);
editableContent.insert((int) insertPositionStart, startTagPart_3);
editableContent.insert((int) insertPositionStart, currAnnot.getType());
editableContent.insert((int) insertPositionStart, startTagPart_2);
editableContent.insert((int) insertPositionStart, currAnnot.getId().toString());
editableContent.insert((int) insertPositionStart, startTagPart_1);
} // if
} // for
FileWriter writer = new FileWriter(file);
writer.write(editableContent.toString());
writer.close();
} else {
Out.prln("Content : " + originalContent);
Out.prln("Repositioning: " + info);
}
String xmlDocument = doc.toXml(peopleAndPlaces, false);
String fileName = new String("StANNIE_toXML_" + count + ".HTML");
FileWriter writer = new FileWriter(fileName);
writer.write(xmlDocument);
writer.close();
} // for each doc
} // main
@Override
public void execute() throws ExecutionException {
AnnotationSet inputAS = document.getAnnotations(inputASName);
AnnotationSet outputAS = document.getAnnotations(outputASName);
AnnotationSet tagAS = document.getAnnotations(tagASName);
AnnotationSet annotsToTransfer = null;
boolean newID = copyAnnotations && inputAS.equals(outputAS);
mappings.clear();
// TODO clean this up so we don't have to repeat ourselves
if (configURL != null) {
BufferedReader in = null;
try {
in = new BomStrippingInputStreamReader(configURL.openStream());
String line = in.readLine();
while (line != null) {
if (!line.trim().equals("")) {
String[] data = line.split("=", 2);
String oldName = data[0].trim();
String newName = data.length == 2 ? data[1].trim() : null;
mappings.put(oldName, new Mapping(oldName, newName));
}
line = in.readLine();
}
} catch (IOException ioe) {
ioe.printStackTrace();
} finally {
IOUtils.closeQuietly(in);
}
} else if (annotationTypes != null) {
for (String type : annotationTypes) {
String[] data = type.split("=", 2);
String oldName = data[0].trim();
String newName = data.length == 2 ? data[1].trim() : null;
mappings.put(oldName, new Mapping(oldName, newName));
}
}
// else
// throw new
// ExecutionException("The annotation list and URL cannot both be null");
if (mappings.size() > 0) {
annotsToTransfer = inputAS.get(mappings.keySet());
} else {
// transfer everything
annotsToTransfer = inputAS.get();
}
// in case of no one annotation from some of annotationTypes
if (annotsToTransfer == null || annotsToTransfer.size() == 0) return;
// check if we have a BODY annotation
// if not, just copy all
if (textTagName == null || textTagName.equals("")) {
// remove from input set unless we copy only
if (!copyAnnotations) inputAS.removeAll(annotsToTransfer);
transferAnnotations(new ArrayList<Annotation>(annotsToTransfer), outputAS, newID);
return;
}
// get the BODY annotation
bodyAnnotations = tagAS.get(textTagName);
if (bodyAnnotations == null || bodyAnnotations.isEmpty()) {
// outputAS.addAll(inputAS);
if (transferAllUnlessFound) {
// remove from input set unless we copy only
if (!copyAnnotations) inputAS.removeAll(annotsToTransfer);
transferAnnotations(new ArrayList<Annotation>(annotsToTransfer), outputAS, newID);
}
return;
}
List<Annotation> annots2Move = new ArrayList<Annotation>();
Iterator<Annotation> bodyIter = bodyAnnotations.iterator();
while (bodyIter.hasNext()) {
Annotation bodyAnn = bodyIter.next();
Long start = bodyAnn.getStartNode().getOffset();
Long end = bodyAnn.getEndNode().getOffset();
// get all annotations we want transferred
AnnotationSet annots2Copy = annotsToTransfer.getContained(start, end);
// copy them to the new set and delete them from the old one
annots2Move.addAll(annots2Copy);
}
if (!copyAnnotations) inputAS.removeAll(annots2Move);
transferAnnotations(annots2Move, outputAS, newID);
}