public static void showJCas(JCas jcas) {
FSIterator<Annotation> it = jcas.getAnnotationIndex().iterator();
Map<String, MutableInt> counters = new TreeMap<String, MutableInt>();
int total = 0;
while (it.hasNext()) {
total += 1;
String annoType = "rien";
try {
Annotation annotation = (Annotation) it.next();
annoType = annotation.getType().getName();
} catch (NullPointerException e) {
it.moveToNext();
annoType = e.getClass().getCanonicalName();
}
if (counters.get(annoType) == null) {
counters.put(annoType, new MutableInt(1));
} else {
counters.get(annoType).increment();
}
}
System.out.println(
"Total annotation in JCas (ID: " + System.identityHashCode(jcas) + "): " + total);
for (String annoType : counters.keySet()) {
System.out.println(annoType + ": " + counters.get(annoType));
}
}
private static String getResult(JCas jcas, String sourceLang, String interLang) {
FSIterator iter = jcas.getAnnotationIndex(Target.type).iterator();
String result = "";
while (iter.isValid()) {
FeatureStructure fs = iter.get();
Target transText = (Target) fs;
result = transText.getContent();
iter.moveToNext();
}
return result;
}
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();
}
}
}
/** Copied and modified from {@link org.apache.uima.util.CasToInlineXml} */
private static String toXML(CAS cas, AnnotationsToElements converter) throws SAXException {
ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
XMLSerializer sax2xml = new XMLSerializer(byteArrayOutputStream, false);
// get document text
String docText = cas.getDocumentText();
char[] docCharArray = docText.toCharArray();
// get iterator over annotations sorted by increasing start position and
// decreasing end position
FSIterator<AnnotationFS> iterator = cas.getAnnotationIndex().iterator();
// This is basically a recursive algorithm that has had the recursion
// removed through the use of an explicit Stack. We iterate over the
// annotations, and if an annotation contains other annotations, we
// push the parent annotation on the stack, process the children, and
// then come back to the parent later.
List<AnnotationFS> stack = new ArrayList<AnnotationFS>();
int pos = 0;
ContentHandler handler = sax2xml.getContentHandler();
handler.startDocument();
// write the start tag
converter.startRootElement(handler);
// now use null is a placeholder for this artificial Document annotation
AnnotationFS curAnnot = null;
while (iterator.isValid()) {
AnnotationFS nextAnnot = iterator.get();
if (curAnnot == null || nextAnnot.getBegin() < curAnnot.getEnd()) {
// nextAnnot's start point is within the span of curAnnot
if (curAnnot == null || nextAnnot.getEnd() <= curAnnot.getEnd()) // crossover span check
{
// nextAnnot is contained within curAnnot
// write text between current pos and beginning of nextAnnot
try {
handler.characters(docCharArray, pos, nextAnnot.getBegin() - pos);
pos = nextAnnot.getBegin();
converter.startAnnotationElement(nextAnnot, handler);
// push parent annotation on stack
stack.add(curAnnot);
// move on to next annotation
curAnnot = nextAnnot;
} catch (StringIndexOutOfBoundsException e) {
System.err.println(
"Invalid annotation range: "
+ nextAnnot.getBegin()
+ ","
+ nextAnnot.getEnd()
+ " in document of length "
+ docText.length());
}
}
iterator.moveToNext();
} else {
// nextAnnot begins after curAnnot ends
// write text between current pos and end of curAnnot
try {
handler.characters(docCharArray, pos, curAnnot.getEnd() - pos);
pos = curAnnot.getEnd();
} catch (StringIndexOutOfBoundsException e) {
System.err.println(
"Invalid annotation range: "
+ curAnnot.getBegin()
+ ","
+ curAnnot.getEnd()
+ " in document of length "
+ docText.length());
}
converter.endAnnotationElement(curAnnot, handler);
// pop next containing annotation off stack
curAnnot = stack.remove(stack.size() - 1);
}
}
// finished writing all start tags, now finish up
if (curAnnot != null) {
try {
handler.characters(docCharArray, pos, curAnnot.getEnd() - pos);
pos = curAnnot.getEnd();
} catch (StringIndexOutOfBoundsException e) {
System.err.println(
"Invalid annotation range: "
+ curAnnot.getBegin()
+ ","
+ curAnnot.getEnd()
+ "in document of length "
+ docText.length());
}
converter.endAnnotationElement(curAnnot, handler);
while (!stack.isEmpty()) {
curAnnot = stack.remove(stack.size() - 1); // pop
if (curAnnot == null) {
break;
}
try {
handler.characters(docCharArray, pos, curAnnot.getEnd() - pos);
pos = curAnnot.getEnd();
} catch (StringIndexOutOfBoundsException e) {
System.err.println(
"Invalid annotation range: "
+ curAnnot.getBegin()
+ ","
+ curAnnot.getEnd()
+ "in document of length "
+ docText.length());
}
converter.endAnnotationElement(curAnnot, handler);
}
}
if (pos < docCharArray.length) {
handler.characters(docCharArray, pos, docCharArray.length - pos);
}
converter.endRootElement(handler);
handler.endDocument();
// return XML string
return new String(byteArrayOutputStream.toByteArray());
}