Java AnalysisEngineFactory.createEngine方法代码示例

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Override
public void beforeTest() throws UIMAException {
	super.beforeTest();

	final ExternalResourceDescription tokensDesc = ExternalResourceFactory.createExternalResourceDescription(
			"lexica",
			ClearNlpLexica.class);

	final AnalysisEngineDescription tokeniserDesc = AnalysisEngineFactory.createEngineDescription(
			ClearNlpTokeniser.class,
			"lexica",
			tokensDesc);

	tokeniserAe = AnalysisEngineFactory.createEngine(tokeniserDesc);

	final AnalysisEngineDescription parserDesc = AnalysisEngineFactory.createEngineDescription(ClearNlpParser.class,
			"lexica",
			tokensDesc);

	ae = AnalysisEngineFactory.createEngine(parserDesc);
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Before
public void setUp() throws ResourceInitializationException, ResourceAccessException {
	// Create a description of an external resource - a fongo instance, in the same way we would
	// have created a shared mongo resource
	final ExternalResourceDescription erd = ExternalResourceFactory.createExternalResourceDescription(
			SharedFongoResource.class, "fongo.collection", "test", "fongo.data", "[]");

	// Create the analysis engine
	final AnalysisEngineDescription aed = AnalysisEngineFactory.createEngineDescription(MongoPatternSaver.class,
			MongoPatternSaver.KEY_MONGO, erd,
			"collection", "test");
	ae = AnalysisEngineFactory.createEngine(aed);
	ae.initialize(new CustomResourceSpecifier_impl(), Collections.emptyMap());

	sfr = (SharedFongoResource) ae.getUimaContext()
			.getResourceObject(MongoPatternSaver.KEY_MONGO);

}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
private static AnalysisEngine createDictionaryAnnotatorEngine() throws Exception {
    AggregateBuilder builder = new AggregateBuilder();

    builder.add(AnalysisEngineFactory.createEngineDescription(SimpleTokenizer.class,
            UimaUtil.SENTENCE_TYPE_PARAMETER, Sentence.class.getName(),
            UimaUtil.TOKEN_TYPE_PARAMETER, Token.class.getName()));

    builder.add(AnalysisEngineFactory.createEngineDescription(DictionaryAnnotator.class,
            DictionaryAnnotator.PARAM_DICTIONARY_LOCATION, "classpath:benchmark-dictionary.csv",
            DictionaryAnnotator.PARAM_TOKENIZER_CLASS, SimpleOpenNlpTokenizer.class.getName(),
            DictionaryAnnotator.PARAM_ANNOTATION_TYPE, DictionaryEntry.class.getName(),
            DictionaryAnnotator.PARAM_CSV_SEPARATOR, ";",
            DictionaryAnnotator.PARAM_DICTIONARY_CASE_SENSITIVE, true,
            DictionaryAnnotator.PARAM_DICTIONARY_ACCENT_SENSITIVE, true));
    AnalysisEngine engine = AnalysisEngineFactory.createEngine(builder.createAggregateDescription());
    return engine;
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
private JCas process(AnalysisEngineDescription dictionaryDescription, String text) {
    try {
        AggregateBuilder builder = new AggregateBuilder();

        builder.add(AnalysisEngineFactory.createEngineDescription(SimpleTokenizer.class,
                UimaUtil.SENTENCE_TYPE_PARAMETER, "uima.tcas.DocumentAnnotation",
                UimaUtil.TOKEN_TYPE_PARAMETER, Token.class.getName()));

        builder.add(dictionaryDescription);
        AnalysisEngine engine = AnalysisEngineFactory.createEngine(builder.createAggregateDescription());
        JCas jcas = engine.newJCas();
        jcas.setDocumentText(text);
        engine.process(jcas);
        return jcas;
    } catch (Exception e) {
        throw new RuntimeException("Failed to create UIMA engine", e);
    }
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testCaseInsensitive() throws Exception{
	AnalysisEngine regexAE = AnalysisEngineFactory.createEngine(Custom.class, Custom.PARAM_TYPE, UK_GOV_DSTL_BALEEN_TYPES_COMMON_PERSON, Custom.PARAM_PATTERN, DIGIT_REGEX, Custom.PARAM_CASE_SENSITIVE, false);
	
	jCas.setDocumentText(TEXT);
	regexAE.process(jCas);
	
	assertEquals(2, JCasUtil.select(jCas, Person.class).size());
	
	Person p1 = JCasUtil.selectByIndex(jCas, Person.class, 0);
	assertNotNull(p1);
	assertEquals(P123, p1.getCoveredText());
	assertEquals(P123, p1.getValue());
	
	Person p2 = JCasUtil.selectByIndex(jCas, Person.class, 1);
	assertNotNull(p2);
	assertEquals(P456, p2.getCoveredText());
	assertEquals(P456, p2.getValue());
	
	regexAE.destroy();
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testLineBreak() throws UIMAException, IOException {
	AnalysisEngine consumer = AnalysisEngineFactory.createEngine(Html5.class, TypeSystemSingleton.getTypeSystemDescriptionInstance(), Html5.PARAM_OUTPUT_FOLDER,
			outputFolder.getPath());

	DocumentAnnotation da = (DocumentAnnotation) jCas.getDocumentAnnotationFs();
	da.setSourceUri("multiline.txt");
	
	jCas.setDocumentText("His name was James\n\nBond.");
	Person p = new Person(jCas, 13, 24);
	p.addToIndexes();
	
	consumer.process(jCas);

	File f = new File(outputFolder, "multiline.txt.html");
	assertTrue(f.exists());

	assertTrue(Files.toString(f, StandardCharsets.UTF_8).contains("data-referent=\"\">James\n\nBond</span>"));
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void test() throws Exception {
  AnalysisEngine engine = AnalysisEngineFactory.createEngine(PosTagger.getDescription("en"));
  this.tokenBuilder.buildTokens(
      this.jCas,
      "The brown fox jumped quickly over the lazy dog.",
      "The brown fox jumped quickly over the lazy dog .");
  engine.process(this.jCas);

  List<String> expected = Arrays.asList("DT JJ NN VBD RB IN DT JJ NN .".split(" "));
  List<String> actual = new ArrayList<String>();
  for (Token token : JCasUtil.select(this.jCas, Token.class)) {
    actual.add(token.getPos());
  }
  Assert.assertEquals(expected, actual);
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testOutputEmpty() throws Exception {
  final AnalysisEngine consumer =
      AnalysisEngineFactory.createEngine(StructuralHtml.class, Html5.PARAM_OUTPUT_FOLDER,
          outputFolder.getPath(), StructuralHtml.PARAM_OUTPUT_EMPTY_TAGS, true);
  final DocumentAnnotation da = (DocumentAnnotation) jCas.getDocumentAnnotationFs();
  da.setSourceUri("test.txt");


  jCas.setDocumentText("Example document: ''");

  final Paragraph d = new Paragraph(jCas);
  d.setBegin(19);
  d.setEnd(19);
  d.addToIndexes();

  consumer.process(jCas);

  final File f = new File(outputFolder, "test.txt.html");
  assertTrue(f.exists());

  System.out.println(Jsoup.parse(f, "UTF-8").html());

  assertEquals(Jsoup.parse(f, "UTF-8").html().replaceAll("\\s*", ""),
      EXPECTED_EMPTY.replaceAll("\\s*", ""));
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testCoNLL2005WriterDescriptor() throws UIMAException {
  try {
    AnalysisEngineFactory.createEngine(Conll2005Writer.class);
    Assert.fail("expected exception without output file parameter");
  } catch (ResourceInitializationException e) {
  }

  File outputFile = new File(outputDirectory, "dev-set-result.txt");

  AnalysisEngine engine = AnalysisEngineFactory.createEngine(
      Conll2005Writer.class,
      Conll2005Writer.PARAM_OUTPUT_FILE,
      outputFile.getPath());

  engine.collectionProcessComplete();
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testCaseSensitive() throws Exception{
	ExternalResourceDescription erd = ExternalResourceFactory.createExternalResourceDescription(FILE_GAZETTEER, SharedFileResource.class);
	AnalysisEngineDescription aed = AnalysisEngineFactory.createEngineDescription(File.class, FILE_GAZETTEER, erd, FILE_NAME, getClass().getResource(GAZETTEER_TXT).getPath(), TYPE, LOCATION, "caseSensitive", true);

	AnalysisEngine ae = AnalysisEngineFactory.createEngine(aed);

	jCas.setDocumentText("This text mentions New York and Paris in upper case and new york in lower case");

	ae.process(jCas);

	// should match "new york" and "Paris", but not "New York"
	assertEquals(2, JCasUtil.select(jCas, Location.class).size());
	Location l1 = JCasUtil.selectByIndex(jCas, Location.class, 0);
	Location l2 = JCasUtil.selectByIndex(jCas, Location.class, 1);
	assertEquals("Paris", l1.getValue());
	assertEquals("new york", l2.getValue());

	ae.destroy();
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testWindowClassNames() throws Exception {
  String text = "I bought a lamp. I love lamp. Lamps are great!";
  this.jCas.setDocumentText(text);
  Sentence window = new Sentence(this.jCas, 0, 30);
  window.addToIndexes();

  AnalysisEngineDescription desc = SentenceAnnotator.getDescription();
  ConfigurationParameterFactory.addConfigurationParameter(
      desc,
      SentenceAnnotator.PARAM_WINDOW_CLASS_NAMES,
      new String[] { "org.cleartk.token.type.Sentence" });
  AnalysisEngine engine = AnalysisEngineFactory.createEngine(desc);
  engine.process(this.jCas);
  engine.collectionProcessComplete();

  Collection<Sentence> sentences = JCasUtil.select(this.jCas, Sentence.class);
  Iterator<Sentence> sentenceIter = sentences.iterator();
  assertEquals(3, sentences.size());
  assertEquals(window, sentenceIter.next());
  assertEquals("I bought a lamp.", sentenceIter.next().getCoveredText());
  assertEquals("I love lamp.", sentenceIter.next().getCoveredText());
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testNoExistingReferentsMerge() throws Exception{
	AnalysisEngine ae = AnalysisEngineFactory.createEngine(CorefBrackets.class, CorefBrackets.PARAM_MERGE_REFERENTS, true);
	jCas.setDocumentText("James (Jimmy) visited Thomas and Ben");
	
	Person p1 = new Person(jCas, 0, 5);
	p1.addToIndexes();
	
	Person p2 = new Person(jCas, 7, 12);
	p2.addToIndexes();
	
	Person p3 = new Person(jCas, 22, 28);
	p3.addToIndexes();
	
	Person p4 = new Person(jCas, 33, 36);
	p4.addToIndexes();
	
	ae.process(jCas);
	
	assertNotNull(p1.getReferent());
	assertEquals(p1.getReferent(), p2.getReferent());
	assertNotEquals(p1.getReferent(), p3.getReferent());
	assertNotEquals(p1.getReferent(), p4.getReferent());
	assertNull(p3.getReferent());
	assertNull(p4.getReferent());
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Test
public void testDescriptor() throws UIMAException {
  try {
    AnalysisEngineFactory.createEngine(TreebankAligningAnnotator.class);
    Assert.fail("expected exception with TreebankDirectory unspecified");
  } catch (ResourceInitializationException e) {
  }

  String treebankPath = "data/treebank/wsj";
  AnalysisEngine engine = AnalysisEngineFactory.createEngine(
      TreebankAligningAnnotator.class,
      TreebankAligningAnnotator.PARAM_TREEBANK_DIRECTORY_NAME,
      treebankPath);

  Object treebankDirectory = engine.getConfigParameterValue(TreebankAligningAnnotator.PARAM_TREEBANK_DIRECTORY_NAME);
  Assert.assertEquals(treebankPath, treebankDirectory);

  engine.collectionProcessComplete();
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
/**
 * Constructor, creates an AnalysisEngine with a CRF classifier.
 * @param modelDirectory the directory where the model is located
 */
public CrfClassifier(String modelDirectory) {
    try {
        classifier = AnalysisEngineFactory.createEngine(AspectAnnotator.class,
                GenericJarClassifierFactory.PARAM_CLASSIFIER_JAR_PATH,
                modelDirectory + "model.jar");
    } catch (ResourceInitializationException e) {
        e.printStackTrace();
    }
}
 

import org.apache.uima.fit.factory.AnalysisEngineFactory; //导入方法依赖的package包/类
@Override
public void beforeTest() throws UIMAException {
	super.beforeTest();

	final ExternalResourceDescription tokensDesc = ExternalResourceFactory.createExternalResourceDescription(
			"lexica",
			ClearNlpLexica.class);

	final AnalysisEngineDescription desc = AnalysisEngineFactory.createEngineDescription(ClearNlpTokeniser.class,
			"lexica",
			tokensDesc);

	ae = AnalysisEngineFactory.createEngine(desc);
}