Java Code Examples for org.apache.uima.fit.factory.CollectionReaderFactory#createReader()

@Test
   public void test() throws Exception {

CollectionReader cr = CollectionReaderFactory.createReader(
	BioNLPGeniaEventsCollectionReader.class);

int i = 0;
while (cr.hasNext()) {
    CAS cas = CasCreationUtils.createCas(cr
	    .getProcessingResourceMetaData());
    cr.getNext(cas);

    // if (createHtml)
    // viewer.createHtml(cas.getJCas(), cas.getTypeSystem(),
    // styleMapFile, new File("target/" + i));

    i++;
}
cr.close();
assertEquals(259, i);

   }
 

@Test
public void testCount() throws Exception {

    CollectionReader cr = CollectionReaderFactory.createReader(
            BioNLPGeniaEventsCollectionReader.class,
            BlueUima.PARAM_INPUT_DIRECTORY, TEST_DIR);

    int i = 0;
    while (cr.hasNext()) {
        CAS cas = CasCreationUtils.createCas(cr
                .getProcessingResourceMetaData());
        cr.getNext(cas);
        LOG.debug(To.string("cas nr " + i, cas.getJCas()));
        i++;
    }
    cr.close();
    assertEquals(3, i);
}
 

@Test
public void testHasNextLooping() throws Exception {

  ExternalResourceDescription contentExtractor =
      ExternalResourceFactory.createNamedResourceDescription(
          KEY_CONTENT_EXTRACTOR, FakeBaleenContentExtractor.class);

  DummyBaleenCollectionReader cr =
      (DummyBaleenCollectionReader)
          CollectionReaderFactory.createReader(
              DummyBaleenCollectionReader.class,
              BaleenCollectionReader.KEY_CONTENT_EXTRACTOR,
              contentExtractor);

  while (cr.hasNext()) {
    JCas jCas = JCasSingleton.getJCasInstance();
    cr.getNext(jCas.getCas());
  }

  cr.destroy();
}
 

@Test
public void testTestCorpus() throws Exception {

    CollectionReader cr = CollectionReaderFactory.createReader(
            Biocreative2GeneCollectionReader.class, BlueUima.PARAM_MODE,
            "test");

    CAS cas = CasCreationUtils
            .createCas(cr.getProcessingResourceMetaData());
    cr.getNext(cas);

    Collection<BioEntityMention> genes = JCasUtil.select(cas.getJCas(),
            BioEntityMention.class);
    assertEquals(2, genes.size());

    cr.close();
}
 

private BaleenCollectionReader createReciever() throws ResourceInitializationException {
  return (BaleenCollectionReader)
      CollectionReaderFactory.createReader(
          ActiveMQTransportReceiver.class,
          TypeSystemSingleton.getTypeSystemDescriptionInstance(),
          SharedActiveMQResource.RESOURCE_KEY,
          mqerd,
          KEY_CONTENT_EXTRACTOR,
          ceerd);
}
 

public static CollectionReader getCR(String path)
		throws ResourceInitializationException, FileNotFoundException {
	File testcaseDir = ResourceHelper.getFile(path);

	return CollectionReaderFactory.createReader(
			PubmedArchiveCollectionReader.class,
			BlueUima.PARAM_INPUT_DIRECTORY, testcaseDir.getAbsolutePath());
}
 

public static CollectionReader getCR(String query, int nrResults)
        throws ResourceInitializationException {
    return CollectionReaderFactory.createReader(
            PubmedWebServiceCollectionReader.class, 
            BlueUima.PARAM_MAX_NR_RESULTS, nrResults,//
            BlueUima.PARAM_QUERY, query);
}
 

private BaleenCollectionReader createReciever() throws ResourceInitializationException {
  return (BaleenCollectionReader)
      CollectionReaderFactory.createReader(
          RedisTransportReceiver.class,
          TypeSystemSingleton.getTypeSystemDescriptionInstance(),
          SharedRedisResource.RESOURCE_KEY,
          erd,
          KEY_CONTENT_EXTRACTOR,
          ExternalResourceFactory.createNamedResourceDescription(
              KEY_CONTENT_EXTRACTOR, FakeBaleenContentExtractor.class));
}
 

private BaleenCollectionReader createReciever() throws ResourceInitializationException {
  return (BaleenCollectionReader)
      CollectionReaderFactory.createReader(
          RabbitMQTransportReceiver.class,
          TypeSystemSingleton.getTypeSystemDescriptionInstance(),
          SharedRabbitMQResource.RESOURCE_KEY,
          erd,
          KEY_CONTENT_EXTRACTOR,
          ExternalResourceFactory.createNamedResourceDescription(
              KEY_CONTENT_EXTRACTOR, FakeBaleenContentExtractor.class));
}
 

private BaleenCollectionReader createReciever() throws ResourceInitializationException {
  return (BaleenCollectionReader)
      CollectionReaderFactory.createReader(
          KafkaTransportReceiver.class,
          TypeSystemSingleton.getTypeSystemDescriptionInstance(),
          SharedKafkaResource.RESOURCE_KEY,
          erd,
          KEY_CONTENT_EXTRACTOR,
          ExternalResourceFactory.createNamedResourceDescription(
              KEY_CONTENT_EXTRACTOR, FakeBaleenContentExtractor.class));
}
 

private MemoryTransportReceiver createReciever() throws ResourceInitializationException {
  return (MemoryTransportReceiver)
      CollectionReaderFactory.createReader(
          MemoryTransportReceiver.class,
          TypeSystemSingleton.getTypeSystemDescriptionInstance(),
          KEY_CONTENT_EXTRACTOR,
          ExternalResourceFactory.createNamedResourceDescription(
              KEY_CONTENT_EXTRACTOR, FakeBaleenContentExtractor.class),
          SharedMemoryQueueResource.RESOURCE_KEY,
          erd);
}
 

protected BaleenCollectionReader getCollectionReader(Object... args)
    throws ResourceInitializationException {
  Object[] argumentWithExtractor =
      ImmutableList.builder()
          .add(KEY_CONTENT_EXTRACTOR)
          .add(contentExtractor)
          .addAll(Arrays.asList(args))
          .build()
          .toArray();
  return (BaleenCollectionReader)
      CollectionReaderFactory.createReader(
          readerClass,
          TypeSystemSingleton.getTypeSystemDescriptionInstance(),
          argumentWithExtractor);
}
 

/**
 * Tests if anno file is correctly rendered for a given document
 */
@Test
public void testRender() throws Exception
{
    String file = "src/test/resources/tcf04-karin-wl.xml";
    String pdftxt = new Scanner(
        new File("src/test/resources/rendererTestPdfExtract.txt")).useDelimiter("\\Z").next();

    CAS cas = JCasFactory.createJCas().getCas();
    CollectionReader reader = CollectionReaderFactory.createReader(TcfReader.class,
        TcfReader.PARAM_SOURCE_LOCATION, file);
    reader.getNext(cas);

    AnnotatorState state = new AnnotatorStateImpl(Mode.ANNOTATION);
    state.setPagingStrategy(new SentenceOrientedPagingStrategy());
    state.getPreferences().setWindowSize(10);
    state.setProject(project);

    VDocument vdoc = new VDocument();
    preRenderer.render(vdoc, 0, cas.getDocumentText().length(), cas,
            schemaService.listAnnotationLayer(project));

    PdfExtractFile pdfExtractFile = new PdfExtractFile(pdftxt, new HashMap<>());
    PdfAnnoRenderer renderer = new PdfAnnoRenderer(schemaService,
            new ColoringServiceImpl(schemaService));
    PdfAnnoModel annoFile = renderer.render(state, vdoc, cas.getDocumentText(), pdfExtractFile,
            0);

    assertThat(annoFile.getAnnoFileContent())
        .isEqualToNormalizingNewlines(contentOf(
                new File("src/test/resources/rendererTestAnnoFile.anno"), UTF_8));
}
 

public T create(Object... args) throws ResourceInitializationException {
  return (T) CollectionReaderFactory.createReader(clazz, args);
}
 

public static void main(String[] args) throws UIMAException, IOException {
  // Choosing different location depending on whether we are in the actual uimaFIT source tree
  // or in the extracted examples from the binary distribution.
  String samplePosFileName;
  if (new File("src/main/resources").exists()) {
      samplePosFileName = "src/main/resources/org/apache/uima/fit/examples/pos/sample-gold.txt";
  }
  else {
      samplePosFileName = "src/org/apache/uima/fit/examples/pos/sample-gold.txt";
  }

  // The lineReader simply copies the lines from the input file into the
  // default view - one line per CAS
  CollectionReader lineReader = CollectionReaderFactory.createReader(LineReader.class,
          LineReader.PARAM_INPUT_FILE, samplePosFileName);

  AggregateBuilder builder = new AggregateBuilder();

  // The goldTagger parses the data in the default view into Token objects
  // along with their part-of-speech tags which will be added to the
  // GOLD_VIEW
  AnalysisEngineDescription goldTagger = AnalysisEngineFactory.createEngineDescription(
          GoldTagger.class);
  builder.add(goldTagger);

  // The textCopier creates the SYSTEM_VIEW and set the text of this view
  // to that of the text found in GOLD_VIEW
  AnalysisEngineDescription textCopier = AnalysisEngineFactory.createEngineDescription(
          ViewTextCopierAnnotator.class,
          ViewTextCopierAnnotator.PARAM_SOURCE_VIEW_NAME, ViewNames.GOLD_VIEW,
          ViewTextCopierAnnotator.PARAM_DESTINATION_VIEW_NAME, ViewNames.SYSTEM_VIEW);
  builder.add(textCopier);

  // The sentenceAndTokenCopier copies Token and Sentence annotations in
  // the GOLD_VIEW into the SYSTEM_VIEW
  AnalysisEngineDescription sentenceAndTokenCopier = AnalysisEngineFactory
          .createEngineDescription(SentenceAndTokenCopier.class);
  builder.add(sentenceAndTokenCopier, ViewNames.VIEW1, ViewNames.GOLD_VIEW, ViewNames.VIEW2,
          ViewNames.SYSTEM_VIEW);

  // The baselineTagger is run on the SYSTEM_VIEW
  AnalysisEngineDescription baselineTagger = AnalysisEngineFactory.createEngineDescription(
          BaselineTagger.class);
  builder.add(baselineTagger, CAS.NAME_DEFAULT_SOFA, ViewNames.SYSTEM_VIEW);

  // The evaluator will compare the part-of-speech tags in the SYSTEM_VIEW
  // with those in the GOLD_VIEW
  AnalysisEngineDescription evaluator = AnalysisEngineFactory.createEngineDescription(
          Evaluator.class);
  builder.add(evaluator);

  // The xWriter writes out the contents of each CAS (one per sentence) to
  // an XMI file. It is instructive to open one of these
  // XMI files in the CAS Visual Debugger and look at the contents of each
  // view.
  AnalysisEngineDescription xWriter = AnalysisEngineFactory.createEngineDescription(
          XmiWriter.class, XmiWriter.PARAM_OUTPUT_DIRECTORY, "target/examples/pos/xmi");
  builder.add(xWriter);

  // runs the collection reader and the aggregate AE.
  SimplePipeline.runPipeline(lineReader, builder.createAggregate());
}
 

public T create() throws ResourceInitializationException {
  return (T) CollectionReaderFactory.createReader(clazz);
}
 

@Test
public void test()
    throws Exception
{
    String targetFolder = "target/test-output/" + testContext.getTestOutputFolderName();
    
    CollectionReader reader = CollectionReaderFactory.createReader(
            WebannoTsv3Reader.class,
            WebannoTsv3Reader.PARAM_SOURCE_LOCATION, "src/test/resources/tsv3/",
            WebannoTsv3Reader.PARAM_PATTERNS, "coref.tsv");

    List<String> slotFeatures = new ArrayList<>();
    List<String> slotTargets = new ArrayList<>();
    List<String> linkTypes = new ArrayList<>();
    List<String> spanLayers = new ArrayList<>();
    spanLayers.add(NamedEntity.class.getName());
    spanLayers.add(POS.class.getName());
    spanLayers.add(Lemma.class.getName());
    List<String> chainLayers = new ArrayList<>();
    chainLayers.add("de.tudarmstadt.ukp.dkpro.core.api.coref.type.Coreference");
    List<String> relationLayers = new ArrayList<>();
    relationLayers.add(Dependency.class.getName());

    AnalysisEngineDescription writer = createEngineDescription(
            WebannoTsv3Writer.class,
            WebannoTsv3Writer.PARAM_TARGET_LOCATION, targetFolder,
            WebannoTsv3Writer.PARAM_STRIP_EXTENSION, true, 
            WebannoTsv3Writer.PARAM_OVERWRITE, true, 
            WebannoTsv3Writer.PARAM_SPAN_LAYERS, spanLayers, 
            WebannoTsv3Writer.PARAM_SLOT_FEATS, slotFeatures,
            WebannoTsv3Writer.PARAM_SLOT_TARGETS, slotTargets,
            WebannoTsv3Writer.PARAM_LINK_TYPES, linkTypes, 
            WebannoTsv3Writer.PARAM_CHAIN_LAYERS, chainLayers,
            WebannoTsv3Writer.PARAM_RELATION_LAYERS, relationLayers);

    runPipeline(reader, writer);

    CollectionReader reader1 = CollectionReaderFactory.createReader(
            WebannoTsv3Reader.class,
            WebannoTsv3Reader.PARAM_SOURCE_LOCATION, "src/test/resources/tsv3/",
            WebannoTsv3Reader.PARAM_PATTERNS, "coref.tsv");

    CollectionReader reader2 = CollectionReaderFactory.createReader(
            WebannoTsv3Reader.class,
            WebannoTsv3Reader.PARAM_SOURCE_LOCATION, targetFolder,
            WebannoTsv3Reader.PARAM_PATTERNS, "coref.tsv");

    CAS cas1 = JCasFactory.createJCas().getCas();
    reader1.getNext(cas1);

    CAS cas2 = JCasFactory.createJCas().getCas();
    reader2.getNext(cas2);

    assertEquals(JCasUtil.select(cas2.getJCas(), Token.class).size(),
            JCasUtil.select(cas1.getJCas(), Token.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), POS.class).size(),
            JCasUtil.select(cas1.getJCas(), POS.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), Lemma.class).size(),
            JCasUtil.select(cas1.getJCas(), Lemma.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), NamedEntity.class).size(),
            JCasUtil.select(cas1.getJCas(), NamedEntity.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), Sentence.class).size(),
            JCasUtil.select(cas1.getJCas(), Sentence.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), Dependency.class).size(),
            JCasUtil.select(cas1.getJCas(), Dependency.class).size());
}
 

@Test
public void test()
    throws Exception
{
    String targetFolder = "target/test-output/" + testContext.getTestOutputFolderName();
    
    CollectionReader reader = CollectionReaderFactory.createReader(
            WebannoTsv3XReader.class,
            WebannoTsv3XReader.PARAM_SOURCE_LOCATION, "src/test/resources/tsv3/",
            WebannoTsv3XReader.PARAM_PATTERNS, "coref.tsv");

    AnalysisEngineDescription writer = createEngineDescription(
            WebannoTsv3XWriter.class,
            WebannoTsv3XWriter.PARAM_TARGET_LOCATION, targetFolder,
            WebannoTsv3XWriter.PARAM_STRIP_EXTENSION, true,
            WebannoTsv3XWriter.PARAM_OVERWRITE, true);

    runPipeline(reader, writer);

    CollectionReader reader1 = CollectionReaderFactory.createReader(
            WebannoTsv3XReader.class,
            WebannoTsv3XReader.PARAM_SOURCE_LOCATION, "src/test/resources/tsv3/",
            WebannoTsv3XReader.PARAM_PATTERNS, "coref.tsv");

    CollectionReader reader2 = CollectionReaderFactory.createReader(
            WebannoTsv3XReader.class,
            WebannoTsv3XReader.PARAM_SOURCE_LOCATION, targetFolder,
            WebannoTsv3XReader.PARAM_PATTERNS, "coref.tsv");

    CAS cas1 = JCasFactory.createJCas().getCas();
    reader1.getNext(cas1);

    CAS cas2 = JCasFactory.createJCas().getCas();
    reader2.getNext(cas2);

    assertEquals(JCasUtil.select(cas2.getJCas(), Token.class).size(),
            JCasUtil.select(cas1.getJCas(), Token.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), POS.class).size(),
            JCasUtil.select(cas1.getJCas(), POS.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), Lemma.class).size(),
            JCasUtil.select(cas1.getJCas(), Lemma.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), NamedEntity.class).size(),
            JCasUtil.select(cas1.getJCas(), NamedEntity.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), Sentence.class).size(),
            JCasUtil.select(cas1.getJCas(), Sentence.class).size());
    assertEquals(JCasUtil.select(cas2.getJCas(), Dependency.class).size(),
            JCasUtil.select(cas1.getJCas(), Dependency.class).size());
}
 

/**
 * <pre>
 * GENE.eval
 * P00001606T0076|14 33|alkaline phosphatases
 * P00001606T0076|37 50|5-nucleotidase
 * 
 * train.in:
 * P00001606T0076 Comparison with alkaline phosphatases and 5-nucleotidase
 * </pre>
 */
@Test
public void testTrainCorpus() throws Exception {

    CollectionReader cr = CollectionReaderFactory.createReader(
            Biocreative2GeneCollectionReader.class,
            BlueUima.PARAM_MODE, "train");

    CAS cas = CasCreationUtils
            .createCas(cr.getProcessingResourceMetaData());
    cr.getNext(cas);

    Collection<BioEntityMention> genes = JCasUtil.select(cas.getJCas(),
            BioEntityMention.class);
    assertEquals(2, genes.size());
    Iterator<BioEntityMention> iterator = genes.iterator();
    BioEntityMention gene = iterator.next();
    assertEquals("alkaline phosphatases", gene.getCoveredText());
    gene = iterator.next();
    assertEquals("5-nucleotidase", gene.getCoveredText());

    // move to 'P00027739T0000 Serum gamma glutamyltransferase in the
    // diagnosis of liver disease in cattle.' to test ALTGENE annotations
    for (int i = 0; i < 11; i++) {
        cas = CasCreationUtils
                .createCas(cr.getProcessingResourceMetaData());
        cr.hasNext();
        cr.getNext(cas);
        Header header = JCasUtil.selectSingle(cas.getJCas(), Header.class);
        LOG.debug("docid:{}, text:{}", header.getDocId(),
                cas.getDocumentText());
    }

    genes = JCasUtil.select(cas.getJCas(), BioEntityMention.class);
    iterator = genes.iterator();

    // check the 2 alternate forms
    assertEquals(2, genes.size());
    gene = iterator.next();
    LOG.debug(gene.getCoveredText());
    assertEquals("Serum gamma glutamyltransferase", gene.getCoveredText());
    gene = iterator.next();
    LOG.debug(gene.getCoveredText());
    assertEquals("gamma glutamyltransferase", gene.getCoveredText());
}
 

/**
 * Tests if given offsets for PDFAnno can be converted to offsets for the document in INCEpTION
 */
@Test
public void testConvertToDocumentOffset() throws Exception
{
    String file = "src/test/resources/tcf04-karin-wl.xml";
    String pdftxt = new Scanner(
        new File("src/test/resources/rendererTestPdfExtract.txt")).useDelimiter("\\Z").next();
    PdfExtractFile pdfExtractFile = new PdfExtractFile(pdftxt, new HashMap<>());

    CAS cas = JCasFactory.createJCas().getCas();
    CollectionReader reader = CollectionReaderFactory.createReader(TcfReader.class,
        TcfReader.PARAM_SOURCE_LOCATION, file);
    reader.getNext(cas);

    AnnotatorState state = new AnnotatorStateImpl(Mode.ANNOTATION);
    state.setPagingStrategy(new SentenceOrientedPagingStrategy());
    state.getPreferences().setWindowSize(10);
    state.setProject(project);

    DocumentModel documentModel = new DocumentModel(cas.getDocumentText());
    // List of PDFAnno offsets
    // indices represent line numbers in the PDFExtractFile for the according character
    List<Offset> offsets = new ArrayList<>();
    offsets.add(new Offset(3, 3));
    offsets.add(new Offset(3, 4));
    offsets.add(new Offset(3, 5));
    offsets.add(new Offset(3, 6));
    offsets.add(new Offset(3, 7));
    offsets.add(new Offset(3, 8));
    offsets.add(new Offset(6, 8));
    offsets.add(new Offset(7, 7));
    offsets.add(new Offset(7, 8));
    offsets.add(new Offset(8, 8));
    offsets.add(new Offset(8, 13));
    offsets.add(new Offset(28, 28));
    offsets.add(new Offset(28, 30));
    offsets.add(new Offset(35, 38));
    // convert to offests for document in INCEpTION
    List<Offset> docOffsets =
        PdfAnnoRenderer.convertToDocumentOffsets(offsets, documentModel, pdfExtractFile);
    List<Offset> expectedOffsets = new ArrayList<>();
    expectedOffsets.add(new Offset(0, 0));
    expectedOffsets.add(new Offset(0, 1));
    expectedOffsets.add(new Offset(0, 2));
    expectedOffsets.add(new Offset(0, 3));
    expectedOffsets.add(new Offset(0, 4));
    expectedOffsets.add(new Offset(0, 6));
    expectedOffsets.add(new Offset(3, 6));
    expectedOffsets.add(new Offset(4, 4));
    expectedOffsets.add(new Offset(4, 6));
    expectedOffsets.add(new Offset(6, 6));
    expectedOffsets.add(new Offset(6, 11));
    expectedOffsets.add(new Offset(29, 29));
    expectedOffsets.add(new Offset(29, 31));
    expectedOffsets.add(new Offset(38, 41));
    assertThat(docOffsets).isEqualTo(expectedOffsets);
}