public String translate(
String sourceLang, String intermediateLang, String targetLang, String source) {
try {
File taeDescriptor = new File(descriptor);
// File inputFile = new File(source);
XMLInputSource in = new XMLInputSource(taeDescriptor);
ResourceSpecifier specifier = UIMAFramework.getXMLParser().parseResourceSpecifier(in);
AnalysisEngine tae = UIMAFramework.produceAnalysisEngine(specifier);
// String document = FileUtils.file2String(inputFile, "UTF-8");
JCas jcas = tae.newJCas();
jcas.setDocumentText(source);
jcas.setDocumentLanguage(sourceLang + "," + intermediateLang + "," + targetLang);
tae.process(jcas);
String result = getResult(jcas, sourceLang, targetLang);
return result;
} catch (Exception e) {
e.printStackTrace();
}
return null;
}
public void setCollectionReader() {
try {
CollectionReaderDescription collectionReaderDescription =
UIMAFramework.getXMLParser()
.parseCollectionReaderDescription(new XMLInputSource(collectionReaderDescriptor));
CollectionReader collectionReader =
UIMAFramework.produceCollectionReader(collectionReaderDescription);
uimaEEEngine.setCollectionReader(collectionReader);
} catch (ResourceInitializationException e) {
} catch (InvalidXMLException e) {
} catch (IOException e) {
}
}
@Override
public ConfigurationParameterSettings settings() throws Exception {
ConfigurationParameterSettings parameters =
this.getTermSuiteTool().getSettings().getMetaData().getConfigurationParameterSettings();
ConfigurationParameterSettings settings =
UIMAFramework.getResourceSpecifierFactory().createConfigurationParameterSettings();
settings.setParameterValue(
"Directory", (String) parameters.getParameterValue("OutputDirectory"));
return settings;
}
@Override
public void load(String id, InputStream inputStream) throws IOException {
if (!this.isLoaded(id)) {
this.getLoaded().add(id);
UIMAFramework.getLogger().log(Level.INFO, "Loading " + id);
Set<Entry<String, String>> entries = this.parse(inputStream);
for (Entry<String, String> entry : entries) {
String source = entry.getKey();
String target = entry.getValue();
this.add(source, target);
}
}
}
/**
* Constructor for the class.
*
* @param args command line arguments into the program - see class description
*/
public SimpleQuestionRunCPE(String args[]) throws Exception {
mStartTime = System.currentTimeMillis();
if (args.length == 0) {
args = new String[1];
args[0] = new String("src/main/resources/CpeQuestionDescriptor.xml");
}
// check command line args
if (args.length < 1) {
printUsageMessage();
System.exit(1);
}
// parse CPE descriptor
System.out.println("Parsing CPE Descriptor");
CpeDescription cpeDesc =
UIMAFramework.getXMLParser().parseCpeDescription(new XMLInputSource(args[0]));
// instantiate CPE
System.out.println("Instantiating CPE");
mCPE = UIMAFramework.produceCollectionProcessingEngine(cpeDesc);
// Create and register a Status Callback Listener
mCPE.addStatusCallbackListener(new StatusCallbackListenerImpl());
// Start Processing
System.out.println("Running CPE");
mCPE.process();
// Allow user to abort by pressing Enter
System.out.println("To abort processing, type \"abort\" and press enter.");
while (true) {
String line = new BufferedReader(new InputStreamReader(System.in)).readLine();
if ("abort".equals(line) && mCPE.isProcessing()) {
System.out.println("Aborting...");
mCPE.stop();
break;
}
}
}
/**
* Method that initializes all vital prerequisites, including POS Tagger
*
* @param language Language to be processed with this copy of HeidelTime
* @param typeToProcess Domain type to be processed
* @param outputType Output type
* @param configPath Path to the configuration file for HeidelTimeStandalone
* @param posTagger POS Tagger to use for preprocessing
* @param doIntervalTagging Whether or not to invoke the IntervalTagger
*/
public void initialize(
Language language,
DocumentType typeToProcess,
OutputType outputType,
String configPath,
POSTagger posTagger,
Boolean doIntervalTagging) {
logger.log(
Level.INFO, "HeidelTimeStandalone initialized with language " + this.language.getName());
// set the POS tagger
this.posTagger = posTagger;
// set doIntervalTagging flag
this.doIntervalTagging = doIntervalTagging;
// read in configuration in case it's not yet initialized
if (!Config.isInitialized()) {
if (configPath == null) readConfigFile(CLISwitch.CONFIGFILE.getValue().toString());
else readConfigFile(configPath);
}
try {
heidelTime = new HeidelTime();
heidelTime.initialize(
new UimaContextImpl(language, typeToProcess, CLISwitch.VERBOSITY2.getIsActive()));
logger.log(Level.INFO, "HeidelTime initialized");
} catch (Exception e) {
e.printStackTrace();
logger.log(Level.WARNING, "HeidelTime could not be initialized");
}
// Initialize JCas factory -------------
logger.log(Level.FINE, "Initializing JCas factory...");
try {
TypeSystemDescription[] descriptions =
new TypeSystemDescription[] {
UIMAFramework.getXMLParser()
.parseTypeSystemDescription(
new XMLInputSource(
this.getClass()
.getClassLoader()
.getResource(Config.get(Config.TYPESYSTEMHOME))))
};
jcasFactory = new JCasFactoryImpl(descriptions);
logger.log(Level.INFO, "JCas factory initialized");
} catch (Exception e) {
e.printStackTrace();
logger.log(Level.WARNING, "JCas factory could not be initialized");
}
}
public static void main(String[] args) throws Exception {
String sLine;
long startTime = System.currentTimeMillis();
URL descUrl =
VectorSpaceRetrieval.class.getResource(
"/descriptors/retrievalsystem/VectorSpaceRetrieval.xml");
if (descUrl == null) {
throw new IllegalArgumentException("Error opening VectorSpaceRetrieval.xml");
}
// create AnalysisEngine
XMLInputSource input = new XMLInputSource(descUrl);
AnalysisEngineDescription desc =
UIMAFramework.getXMLParser().parseAnalysisEngineDescription(input);
AnalysisEngine anAnalysisEngine = UIMAFramework.produceAnalysisEngine(desc);
CAS aCas = anAnalysisEngine.newCAS();
URL docUrl = VectorSpaceRetrieval.class.getResource("/data/documents.txt");
if (docUrl == null) {
throw new IllegalArgumentException("Error opening data/documents.txt");
}
BufferedReader br = new BufferedReader(new InputStreamReader(docUrl.openStream()));
while ((sLine = br.readLine()) != null) {
aCas.setDocumentText(sLine);
anAnalysisEngine.process(aCas);
aCas.reset();
}
br.close();
br = null;
anAnalysisEngine.collectionProcessComplete();
anAnalysisEngine.destroy();
long endTime = System.currentTimeMillis();
double totalTime = (endTime - startTime) / 1000.0;
System.out.println("Total time taken: " + totalTime);
}
@Test
public void test() throws Exception {
TypeSystemDescription tsd =
TypeSystemDescriptionFactory.createTypeSystemDescription("desc.types.test-TypeSystem");
CollectionReaderDescription colReaderDesc =
CollectionReaderFactory.createReaderDescription(
XmiCollectionReader.class, tsd,
XmiCollectionReader.PARAM_INPUTDIR, inputFileXMIDir);
// configure AE
XMLInputSource aeDescInput = new XMLInputSource(U2BAggregateDesc);
AnalysisEngineDescription aeDesc =
UIMAFramework.getXMLParser().parseAnalysisEngineDescription(aeDescInput);
SimplePipeline.runPipeline(colReaderDesc, aeDesc);
}
public static void main(String[] args)
throws IOException, InvalidXMLException, CASException, ResourceInitializationException,
SAXException {
JCas jcas = null;
// Leer el descriptor del anotador
XMLParser xmlParser = UIMAFramework.getXMLParser();
XMLInputSource in = new XMLInputSource("desc/ej4/MetricsAnnotatorPipeline.xml");
// Crear un AE en base al descriptor
AnalysisEngineDescription tsDesc = xmlParser.parseAnalysisEngineDescription(in);
// Obtener el CAS
jcas = CasCreationUtils.createCas(tsDesc).getJCas();
if (jcas != null) {
// De-serializar la anotacion de un fichero
FileInputStream inputStream = null;
inputStream = new FileInputStream("resources/annotation.xmi");
XmiCasDeserializer.deserialize(inputStream, jcas.getCas());
// Obtener el texto de la anotacion
String sofaString = jcas.getDocumentText();
System.out.println(sofaString);
// Usar las anotaciones del fichero
FSIterator it = jcas.getAnnotationIndex(Metric.type).iterator();
while (it.isValid()) {
Metric metric = (Metric) it.get();
Number number = metric.getNumber();
Unit unit = metric.getUnit();
Double value =
(number.getIsDouble())
? number.getAbsoluteDoubleValue()
: Double.valueOf(number.getAbsoluteIntegerValue());
System.out.println("===================");
System.out.println("Metric: " + metric.getCoveredText());
System.out.println("Real value: " + value * number.getSign() * unit.getMultiplier());
System.out.println("Base unit: " + unit.getBaseUnit());
it.moveToNext();
}
}
}
/**
* Called when the processing of a Document is completed. <br>
* The process status can be looked at and corresponding actions taken.
*
* @param aCas CAS corresponding to the completed processing
* @param aStatus EntityProcessStatus that holds the status of all the events for aEntity
*/
public void entityProcessComplete(CAS aCas, EntityProcessStatus aStatus) {
// if there is an error above the individual document level,
// an entityProcessStatus is created with a null value for entity
if (aCas == null) {
for (int i = 0; i < aStatus.getFailedComponentNames().size(); i++) {
LOG.info("[{}] FailedComponentNames", aStatus.getFailedComponentNames().get(i));
}
for (int i = 0; i < aStatus.getExceptions().size(); i++) {
LOG.info("[{}] Exceptions", aStatus.getExceptions().get(i));
}
return;
}
try {
entityCount++;
// FIXME int dataSize = 0;
// // get size here
// Type t = aCas.getTypeSystem().getType("uima.cpm.FileLocation");
// Feature f = t.getFeatureByBaseName("DocumentSize");
// FSIterator fsI = aCas.getAnnotationIndex(t).iterator();
// if (fsI.isValid()) {
// dataSize = fsI.get().getIntValue(f);
// }
//
// size += dataSize;
// to handle exceptions occured in any of the components for the
// entity
if (aStatus.isException()) {
for (int q = 0; q < aStatus.getExceptions().size(); q++) {
Exception e = (Exception) aStatus.getExceptions().get(q);
e.printStackTrace();
}
}
} catch (Exception io) {
UIMAFramework.getLogger(this.getClass()).log(Level.WARNING, "", io);
}
}
@Test
public void test() throws Exception {
String html = "<Parent>\n";
html += "<Child1>Some content</Child1>\n";
html += "<Child2 attribute=“someValue” />\n";
html += "<Child3>More content.</Child3>\n";
html += "</Parent>\n";
URL urlA = HtmlAnnotator.class.getClassLoader().getResource("HtmlAnnotator.xml");
if (urlA == null) {
urlA =
HtmlAnnotator.class
.getClassLoader()
.getResource("org/apache/uima/ruta/engine/HtmlAnnotator.xml");
}
URL urlC = HtmlAnnotator.class.getClassLoader().getResource("HtmlConverter.xml");
if (urlC == null) {
urlC =
HtmlAnnotator.class
.getClassLoader()
.getResource("org/apache/uima/ruta/engine/HtmlConverter.xml");
}
XMLInputSource inA = new XMLInputSource(urlA);
ResourceSpecifier specifierA = UIMAFramework.getXMLParser().parseResourceSpecifier(inA);
AnalysisEngine aeA = UIMAFramework.produceAnalysisEngine(specifierA);
aeA.setConfigParameterValue(HtmlAnnotator.PARAM_ONLY_CONTENT, false);
aeA.reconfigure();
XMLInputSource inC = new XMLInputSource(urlC);
ResourceSpecifier specifierC = UIMAFramework.getXMLParser().parseResourceSpecifier(inC);
AnalysisEngine aeC = UIMAFramework.produceAnalysisEngine(specifierC);
aeC.setConfigParameterValue(HtmlConverter.PARAM_SKIP_WHITESPACES, false);
aeC.setConfigParameterValue(HtmlConverter.PARAM_PROCESS_ALL, true);
aeC.setConfigParameterValue(
HtmlConverter.PARAM_GAP_INDUCING_TAGS, new String[] {"child1", "child2", "child3"});
aeC.setConfigParameterValue(HtmlConverter.PARAM_GAP_TEXT, "$");
aeC.reconfigure();
CAS cas = aeA.newCAS();
Type tagType = cas.getTypeSystem().getType(HtmlAnnotator.NAMESPACE + "TAG");
AnnotationIndex<AnnotationFS> ai = null;
FSIterator<AnnotationFS> iterator = null;
cas.setDocumentText(html);
aeA.process(cas);
aeC.process(cas);
CAS plainTextCas = cas.getView(HtmlConverter.DEFAULT_MODIFIED_VIEW);
assertEquals("$Some content$$More content.", plainTextCas.getDocumentText());
ai = plainTextCas.getAnnotationIndex(tagType);
iterator = ai.iterator();
assertEquals(4, ai.size());
assertEquals("$Some content$$More content.", iterator.next().getCoveredText());
assertEquals("$Some content", iterator.next().getCoveredText());
assertEquals("$", iterator.next().getCoveredText());
assertEquals("$More content.", iterator.next().getCoveredText());
cas.release();
}
@Test
public void testExpandOffsets() throws Exception {
String html = "<Parent>\n";
html += "<Child1>Some content</Child1>\n";
html += "<Child2 attribute=“someValue” />\n";
html += "<Child3>More content.</Child3>\n";
html += "</Parent>\n";
URL urlA = HtmlAnnotator.class.getClassLoader().getResource("HtmlAnnotator.xml");
if (urlA == null) {
urlA =
HtmlAnnotator.class
.getClassLoader()
.getResource("org/apache/uima/ruta/engine/HtmlAnnotator.xml");
}
URL urlC = HtmlAnnotator.class.getClassLoader().getResource("HtmlConverter.xml");
if (urlC == null) {
urlC =
HtmlAnnotator.class
.getClassLoader()
.getResource("org/apache/uima/ruta/engine/HtmlConverter.xml");
}
XMLInputSource inA = new XMLInputSource(urlA);
ResourceSpecifier specifierA = UIMAFramework.getXMLParser().parseResourceSpecifier(inA);
AnalysisEngine aeA = UIMAFramework.produceAnalysisEngine(specifierA);
aeA.setConfigParameterValue(HtmlAnnotator.PARAM_ONLY_CONTENT, false);
aeA.reconfigure();
XMLInputSource inC = new XMLInputSource(urlC);
ResourceSpecifier specifierC = UIMAFramework.getXMLParser().parseResourceSpecifier(inC);
AnalysisEngine aeC = UIMAFramework.produceAnalysisEngine(specifierC);
aeC.setConfigParameterValue(HtmlConverter.PARAM_SKIP_WHITESPACES, false);
aeC.setConfigParameterValue(HtmlConverter.PARAM_PROCESS_ALL, true);
aeC.setConfigParameterValue(HtmlConverter.PARAM_EXPAND_OFFSETS, true);
aeC.reconfigure();
CAS cas = aeA.newCAS();
Type tagType = cas.getTypeSystem().getType(HtmlAnnotator.NAMESPACE + "TAG");
Feature expandedFeature = tagType.getFeatureByBaseName("expandedOffsets");
AnnotationIndex<AnnotationFS> ai = null;
FSIterator<AnnotationFS> iterator = null;
cas.setDocumentText(html);
aeA.process(cas);
aeC.process(cas);
CAS plainTextCas = cas.getView(HtmlConverter.DEFAULT_MODIFIED_VIEW);
assertEquals("Some contentMore content.", plainTextCas.getDocumentText());
ai = plainTextCas.getAnnotationIndex(tagType);
iterator = ai.iterator();
assertEquals(4, ai.size());
AnnotationFS next = null;
next = iterator.next();
assertEquals(false, next.getBooleanValue(expandedFeature));
assertEquals("Some contentMore content.", next.getCoveredText());
next = iterator.next();
assertEquals(false, next.getBooleanValue(expandedFeature));
assertEquals("Some content", next.getCoveredText());
next = iterator.next();
boolean b1 = next.getBooleanValue(expandedFeature);
assertEquals("More content.", next.getCoveredText());
next = iterator.next();
boolean b2 = next.getBooleanValue(expandedFeature);
assertEquals("More content.", next.getCoveredText());
// for one of these two annotation (with same offsets) the feature must be set to true
assertEquals(true, b1 || b2);
cas.release();
}
