本文整理匯總了Java中org.apache.uima.cas.CAS.reset方法的典型用法代碼示例。如果您正苦於以下問題：Java CAS.reset方法的具體用法？Java CAS.reset怎麽用？Java CAS.reset使用的例子？那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在類org.apache.uima.cas.CAS的用法示例。
在下文中一共展示了CAS.reset方法的2個代碼示例，這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚，您的評價將有助於係統推薦出更棒的Java代碼示例。
示例1: main
import org.apache.uima.cas.CAS; //導入方法依賴的package包/類
public static void main(String[] args) throws IOException, InvalidXMLException,
ResourceInitializationException, AnalysisEngineProcessException, CASException,
SAXException {
if (args.length != 2) {
System.err.println("Usage: XmiToOpenNlpTrainer <input folder with xmis> <output file>");
}
AnalysisEngineDescription descriptor = (AnalysisEngineDescription) createResourceCreationSpecifier(
new XMLInputSource(XmiToOpenNlpTrainer.class.getClassLoader().getResourceAsStream(
"org/ie4opendata/octroy/SimpleFrenchTokenAndSentenceAnnotator.xml"), new File(
".")), new Object[0]);
AnalysisEngine engine = AnalysisEngineFactory.createEngine(descriptor);
CAS cas = engine.newCAS();
PrintWriter pw = new PrintWriter(new FileWriter(args[1]));
for (File file : new File(args[0]).listFiles()) {
cas.reset();
InputStream is = new FileInputStream(file);
XmiCasDeserializer.deserialize(is, cas, true);
JCas jcas = cas.getJCas();
if (!JCasUtil.exists(jcas, DocumentAnnotation.class)) {
DocumentAnnotation documentAnnotation = new DocumentAnnotation(cas.getJCas());
documentAnnotation.setDocumentName(file.getName());
documentAnnotation.setClassified(false);
documentAnnotation.addToIndexes();
}
if (!JCasUtil.exists(jcas, Sentence.class))
engine.process(cas);
List<int[]> companyBoundaries = new ArrayList<int[]>();
for (Company company : JCasUtil.select(jcas, Company.class))
companyBoundaries.add(new int[] { company.getBegin(), company.getEnd() });
int currentCompany = 0;
boolean inCompany = false;
// one sentence per line, one token separated by spaces
for (Sentence sentence : JCasUtil.select(jcas, Sentence.class)) {
if (inCompany) {
if (sentence.getBegin() >= companyBoundaries.get(currentCompany)[1]) {
inCompany = false;
currentCompany++;
} else
pw.print(" <START:company> ");
}
for (Token token : JCasUtil.selectCovered(Token.class, sentence)) {
if (currentCompany < companyBoundaries.size()) {
if (inCompany) {
if (token.getBegin() >= companyBoundaries.get(currentCompany)[1]) {
pw.println(" <END> ");
inCompany = false;
currentCompany++;
}
} else {
if (token.getBegin() >= companyBoundaries.get(currentCompany)[0]) {
pw.print(" <START:company> ");
inCompany = true;
}
}
}
pw.print(token.getCoveredText() + " ");
}
if (inCompany) {
pw.println(" <END> ");
}
pw.println();
}
// each document separated by an empty line
pw.println();
}
pw.close();
}
示例2: main
import org.apache.uima.cas.CAS; //導入方法依賴的package包/類
public static void main(String[] args) throws IOException, InvalidXMLException, ResourceInitializationException,
AnalysisEngineProcessException, CASException {
if (args.length != 2) {
System.err.println("Usage: OpenNlpTrainerExtractor <input folder> <output file>");
}
AnalysisEngineDescription descriptor = (AnalysisEngineDescription) createResourceCreationSpecifier(
new XMLInputSource(OpenNlpTrainerExtractor.class.getClassLoader().getResourceAsStream(
"org/ie4opendata/octroy/SimpleFrenchTokenAndSentenceAnnotator.xml"), new File(".")),
new Object[0]);
AnalysisEngine engine = AnalysisEngineFactory.createEngine(descriptor);
CAS cas = engine.newCAS();
PrintWriter pw = new PrintWriter(new FileWriter(args[1]));
for (File file : new File(args[0]).listFiles()) {
BufferedReader br = new BufferedReader(new FileReader(file));
StringBuilder doc = new StringBuilder();
String line = br.readLine();
while (line != null) {
doc.append(line).append('\n');
line = br.readLine();
}
br.close();
cas.reset();
cas.setDocumentText(doc.toString());
cas.setDocumentLanguage("fr");
DocumentAnnotation documentAnnotation = new DocumentAnnotation(cas.getJCas());
documentAnnotation.setDocumentName(file.getName());
documentAnnotation.setClassified(false);
documentAnnotation.addToIndexes();
engine.process(cas);
// one sentence per line, one token separated by spaces
JCas jcas = cas.getJCas();
for (Sentence sentence : JCasUtil.select(jcas, Sentence.class)) {
for (Token token : JCasUtil.selectCovered(Token.class, sentence)) {
pw.print(token.getCoveredText() + " ");
}
pw.println();
}
// each document separated by an empty line
pw.println();
}
pw.close();
}