Examples of org.apache.uima.util.XMLInputSource

org.apache.uima.util.XMLInputSource
An input source used by the {@link XMLParser} to read XML documents for parsing.
The application uses the {@link #XMLInputSource(File)} constructor to create anXMLInputSource from a descriptor File. Alternatively, if the source of the XML is not a file, the {@link #XMLInputSource(InputStream,File)} constructor may beused to read the XML from an input stream. The second argument to this constructor is the relative path base, to be used if the descriptor contains imports with relative paths. It is acceptable to set this to null if it is known that the descriptor does not contain any such imports.


  public void testProcessDelegateAnalysisEngineMetaData() throws Exception {
    try {
      // create aggregate analysis engine whose delegates each declare
      // type system, type priorities, and indexes
      XMLInputSource in = new XMLInputSource(JUnitExtension
              .getFile("TextAnalysisEngineImplTest/AggregateTaeForMergeTest.xml"));
      AnalysisEngineDescription desc = UIMAFramework.getXMLParser().parseAnalysisEngineDescription(in);
      AggregateAnalysisEngine_impl ae = new AggregateAnalysisEngine_impl();
      ae.initialize(desc, Collections.EMPTY_MAP);
      // initialize method automatically calls processDelegateAnalysisEngineMetaData()

View Full Code Here

      aggAe.initialize(aggDesc, null);
      aggAe.collectionProcessComplete(new ProcessTrace_impl());
      
      //test that fixedFlow order is used
      File descFile = JUnitExtension.getFile("TextAnalysisEngineImplTest/AggregateForCollectionProcessCompleteTest.xml");
      AnalysisEngineDescription cpcTestDesc = UIMAFramework.getXMLParser().parseAnalysisEngineDescription(new XMLInputSource(descFile));
      AnalysisEngine cpcTestAe = UIMAFramework.produceAnalysisEngine(cpcTestDesc);
      cpcTestAe.collectionProcessComplete();
      assertEquals("One", AnnotatorForCollectionProcessCompleteTest.lastValue);
    } catch (Exception e) {
      JUnitExtension.handleException(e);

View Full Code Here

  }


  public void testTypeSystemInit() throws Exception {
    try {
      AnalysisEngineDescription aggWithCcDesc = UIMAFramework.getXMLParser().parseAnalysisEngineDescription(
              new XMLInputSource(JUnitExtension
                      .getFile("TextAnalysisEngineImplTest/AggregateTaeWithCasConsumer.xml")));
      AggregateAnalysisEngine_impl aggAe = new AggregateAnalysisEngine_impl();
      aggAe.initialize(aggWithCcDesc, null);
      CAS tcas = aggAe.newCAS();
      tcas.setDocumentText("This is a test");

View Full Code Here

  public void testProcessAndOutputNewCASes() throws Exception {
    try {
      // primitive
      AnalysisEngineDescription segmenterDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension
                              .getFile("TextAnalysisEngineImplTest/NewlineSegmenter.xml")));
      AnalysisEngine ae = UIMAFramework.produceAnalysisEngine(segmenterDesc);
      CAS cas = ae.newCAS();
      cas.setDocumentText("Line one\nLine two\nLine three");
      CasIterator iter = ae.processAndOutputNewCASes(cas);
      assertTrue(iter.hasNext());
      CAS outCas = iter.next();
      assertEquals("Line one", outCas.getDocumentText());
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line two", outCas.getDocumentText());
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line three", outCas.getDocumentText());
      outCas.release();
      assertFalse(iter.hasNext());


      // aggregate
      AnalysisEngineDescription aggSegDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension
                              .getFile("TextAnalysisEngineImplTest/AggregateWithSegmenter.xml")));
      ae = UIMAFramework.produceAnalysisEngine(aggSegDesc);
      cas = ae.newCAS();
      cas.setDocumentText("Line one\nLine two\nLine three");
      iter = ae.processAndOutputNewCASes(cas);
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line one", outCas.getDocumentText());
      assertEquals("Line one", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line two", outCas.getDocumentText());
      assertEquals("Line two", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line three", outCas.getDocumentText());
      assertEquals("Line three", TestAnnotator.lastDocument);
      outCas.release();
      assertFalse(iter.hasNext());
      // Annotator should NOT get the original CAS according to the default flow
      assertEquals("Line three", TestAnnotator.lastDocument);


      // nested aggregate
      AnalysisEngineDescription nestedAggSegDesc = UIMAFramework
              .getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(
                              JUnitExtension
                                      .getFile("TextAnalysisEngineImplTest/AggregateContainingAggregateSegmenter.xml")));
      ae = UIMAFramework.produceAnalysisEngine(nestedAggSegDesc);
      cas = ae.newCAS();
      cas.setDocumentText("Line one\nLine two\nLine three");
      iter = ae.processAndOutputNewCASes(cas);
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line one", outCas.getDocumentText());
      assertEquals("Line one", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line two", outCas.getDocumentText());
      assertEquals("Line two", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line three", outCas.getDocumentText());
      assertEquals("Line three", TestAnnotator.lastDocument);
      outCas.release();
      assertFalse(iter.hasNext());
      // Annotator should NOT get the original CAS according to the default flow
      assertEquals("Line three", TestAnnotator.lastDocument);


      // two segmenters
      AnalysisEngineDescription twoSegDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension
                              .getFile("TextAnalysisEngineImplTest/AggregateWith2Segmenters.xml")));
      ae = UIMAFramework.produceAnalysisEngine(twoSegDesc);
      cas = ae.newCAS();
      cas.setDocumentText("One\tTwo\nThree\tFour");
      iter = ae.processAndOutputNewCASes(cas);
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("One", outCas.getDocumentText());
      assertEquals("One", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Two", outCas.getDocumentText());
      assertEquals("Two", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Three", outCas.getDocumentText());
      assertEquals("Three", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Four", outCas.getDocumentText());
      assertEquals("Four", TestAnnotator.lastDocument);
      outCas.release();
      assertFalse(iter.hasNext());
      // Annotator should NOT get the original CAS according to the default flow
      assertEquals("Four", TestAnnotator.lastDocument);


      // dropping segments
      aggSegDesc = UIMAFramework.getXMLParser().parseAnalysisEngineDescription(
              new XMLInputSource(JUnitExtension
                      .getFile("TextAnalysisEngineImplTest/AggregateSegmenterForDropTest.xml")));
      ae = UIMAFramework.produceAnalysisEngine(aggSegDesc);
      cas = ae.newCAS();
      cas.setDocumentText("Line one\nDROP\nLine two\nDROP\nLine three");
      // results should be the same as the first aggregate segmenter test.
      // segmetns whose text is DROP should not be output.
      iter = ae.processAndOutputNewCASes(cas);
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line one", outCas.getDocumentText());
      assertEquals("Line one", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line two", outCas.getDocumentText());
      assertEquals("Line two", TestAnnotator.lastDocument);
      outCas.release();
      assertTrue(iter.hasNext());
      outCas = iter.next();
      assertEquals("Line three", outCas.getDocumentText());
      assertEquals("Line three", TestAnnotator.lastDocument);
      outCas.release();
      assertFalse(iter.hasNext());
      // Annotator should NOT get the original CAS according to the default flow
      assertEquals("Line three", TestAnnotator.lastDocument);
      
      //with ParallelStep
      AnalysisEngineDescription desc = UIMAFramework.getXMLParser().parseAnalysisEngineDescription(
        new XMLInputSource(JUnitExtension.getFile("TextAnalysisEngineImplTest/AggregateForParallelStepCasMultiplierTest.xml")));
      ae = UIMAFramework.produceAnalysisEngine(desc);
      cas.reset();
      cas.setDocumentText("One\tTwo\nThree\tFour");
      iter = ae.processAndOutputNewCASes(cas);
      Set<String> expectedOutputs = new HashSet<String>();
      expectedOutputs.add("One");
      expectedOutputs.add("Two\nThree");
      expectedOutputs.add("Four");
      expectedOutputs.add("One\tTwo");
      expectedOutputs.add("Three\tFour");
      while (iter.hasNext()) {
        outCas = iter.next();
        assertTrue(expectedOutputs.remove(outCas.getDocumentText()));        
        outCas.release();
      }
      assertTrue(expectedOutputs.isEmpty());


      
      // test aggregate with 2 AEs sharing resource manager
      AnalysisEngineDescription aggregateSegDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension
                              .getFile("TextAnalysisEngineImplTest/AggregateWithSegmenter.xml")));
      
      ResourceManager rsrcMgr = UIMAFramework.newDefaultResourceManager();
      Map<String, Object> params = new HashMap<String, Object>();
      AnalysisEngine ae1 = UIMAFramework.produceAnalysisEngine(aggregateSegDesc, rsrcMgr, params);

View Full Code Here

    try {
      // aggregate
      AnalysisEngineDescription aggSegDesc = UIMAFramework
              .getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(
                              JUnitExtension
                                      .getFile("TextAnalysisEngineImplTest/AggregateWithSegmenterForErrorTest.xml")));
      AnalysisEngine ae = UIMAFramework.produceAnalysisEngine(aggSegDesc);
      
      CAS cas = ae.newCAS();
      for (int i = 0; i < 2; i++) // verify we can do this more than once
      {
        FlowControllerForErrorTest.reset();
        cas.setDocumentText("Line one\nLine two\nERROR");
        CasIterator iter = ae.processAndOutputNewCASes(cas);
        assertTrue(iter.hasNext());
        CAS outCas = iter.next();
        assertEquals("Line one", outCas.getDocumentText());
        outCas.release();
        assertTrue(iter.hasNext());
        outCas = iter.next();
        assertEquals("Line two", outCas.getDocumentText());
        outCas.release();
        try {
          UIMAFramework.getLogger().setLevel(Level.OFF);  // Suppress logging of expected exception
          assertTrue(iter.hasNext());
          outCas = iter.next();
          fail(); // the above should throw an exception
        } catch (AnalysisEngineProcessException e) {
          UIMAFramework.getLogger().setLevel(Level.INFO); // Restore to apparent default of INFO
        }
        //check that FlowController was notified twice, once for the 
        //segment's flow and once for the complete document's flow
        assertEquals(2, FlowControllerForErrorTest.abortedDocuments.size());
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("ERROR"));
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("Line one\nLine two\nERROR"));


        cas.reset();
      }


      // nested aggregate
      AnalysisEngineDescription nestedAggSegDesc = UIMAFramework
              .getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(
                              JUnitExtension
                                      .getFile("TextAnalysisEngineImplTest/NestedAggregateSegmenterForErrorTest.xml")));
      ae = UIMAFramework.produceAnalysisEngine(nestedAggSegDesc);
      cas = ae.newCAS();
      for (int i = 0; i < 2; i++) // verify we can do this more than once
      {
        FlowControllerForErrorTest.reset();
        cas.setDocumentText("Line one\nLine two\nERROR");
        CasIterator iter = ae.processAndOutputNewCASes(cas);
        assertTrue(iter.hasNext());
        CAS outCas = iter.next();
        assertEquals("Line one", outCas.getDocumentText());
        outCas.release();
        assertTrue(iter.hasNext());
        outCas = iter.next();
        assertEquals("Line two", outCas.getDocumentText());
        outCas.release();
        try {
          UIMAFramework.getLogger().setLevel(Level.OFF);  // Suppress logging of expected exception
          assertTrue(iter.hasNext());
          outCas = iter.next();
          fail(); // the above should throw an exception
        } catch (AnalysisEngineProcessException e) {
          UIMAFramework.getLogger().setLevel(Level.INFO); // Restore to apparent default of INFO
        }
        //check that FlowController was notified three times, once for the 
        //segment's flow and twice for the complete document's flow (once
        //in each aggregate)
        assertEquals(3, FlowControllerForErrorTest.abortedDocuments.size());
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("ERROR"));
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("Line one\nLine two\nERROR"));
        FlowControllerForErrorTest.abortedDocuments.remove("Line one\nLine two\nERROR");
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("Line one\nLine two\nERROR"));
        
        cas.reset();
      }


      // 2 segmenters
      AnalysisEngineDescription twoSegDesc = UIMAFramework
              .getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(
                              JUnitExtension
                                      .getFile("TextAnalysisEngineImplTest/AggregateWith2SegmentersForErrorTest.xml")));
      ae = UIMAFramework.produceAnalysisEngine(twoSegDesc);
      cas = ae.newCAS();
      for (int i = 0; i < 2; i++) // verify we can do this more than once
      {
        FlowControllerForErrorTest.abortedDocuments.clear();
        cas.setDocumentText("One\tTwo\nThree\tERROR");
        CasIterator iter = ae.processAndOutputNewCASes(cas);
        assertTrue(iter.hasNext());
        CAS outCas = iter.next();
        assertEquals("One", outCas.getDocumentText());
        outCas.release();
        assertTrue(iter.hasNext());
        outCas = iter.next();
        assertEquals("Two", outCas.getDocumentText());
        outCas.release();
        assertTrue(iter.hasNext());
        outCas = iter.next();
        assertEquals("Three", outCas.getDocumentText());
        outCas.release();
        try {
          UIMAFramework.getLogger().setLevel(Level.OFF);  // Suppress logging of expected exception
          assertTrue(iter.hasNext());
          outCas = iter.next();
          fail(); // the above should throw an exception
        } catch (AnalysisEngineProcessException e) {
          UIMAFramework.getLogger().setLevel(Level.INFO); // Restore to apparent default of INFO
        }
        //check that FlowController was notified three times, once for each level of granularity
        assertEquals(3, FlowControllerForErrorTest.abortedDocuments.size());
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("ERROR"));
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("Three\tERROR"));
        assertTrue(FlowControllerForErrorTest.abortedDocuments.contains("One\tTwo\nThree\tERROR"));
        
        cas.reset();
      }


      // segmenter that requests too many CASes
      AnalysisEngineDescription segmenterDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension
                              .getFile("TextAnalysisEngineImplTest/BadSegmenter.xml")));
      ae = UIMAFramework.produceAnalysisEngine(segmenterDesc);
      cas = ae.newCAS();
      cas.setDocumentText("Line one\nLine two\nLine three");
      CasIterator iter = ae.processAndOutputNewCASes(cas);
      assertTrue(iter.hasNext());
      CAS outCas = iter.next(); // first call OK
      outCas.release();
      assertTrue(iter.hasNext());
      // next call should fail with AnalysisEngineProcessException
      try {
        UIMAFramework.getLogger().setLevel(Level.OFF);  // Suppress logging of expected exception
        iter.next();
        fail(); // should not get here
      } catch (AnalysisEngineProcessException e) {
        UIMAFramework.getLogger().setLevel(Level.INFO); // Restore to apparent default of INFO
      }
      
      // bad segmenter in an aggregate
      AnalysisEngineDescription aggWithBadSegmenterDesc = UIMAFramework.getXMLParser()
      .parseAnalysisEngineDescription(
              new XMLInputSource(JUnitExtension
                      .getFile("TextAnalysisEngineImplTest/AggregateWithBadSegmenterForErrorTest.xml")));
      ae = UIMAFramework.produceAnalysisEngine(aggWithBadSegmenterDesc);
      FlowControllerForErrorTest.reset();
      cas = ae.newCAS();
      cas.setDocumentText("Line one\nLine two\nLine three");

View Full Code Here

  public void testResegment() throws Exception {
    try {
      // primitive
      AnalysisEngineDescription segmenterDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension
                              .getFile("TextAnalysisEngineImplTest/NewlineResegmenter.xml")));
      AnalysisEngine ae = UIMAFramework.produceAnalysisEngine(segmenterDesc);
      CAS inputCas1 = ae.newCAS();
      Type sdiType = inputCas1.getTypeSystem().getType(
              "org.apache.uima.examples.SourceDocumentInformation");

View Full Code Here

  public void testProcessWithError() throws Exception {
    try {
      //This test uses an aggregate AE fails if the document text is set to "ERROR".
      AnalysisEngineDescription aeDesc = UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(
                      new XMLInputSource(JUnitExtension.getFile("TextAnalysisEngineImplTest/AggregateForErrorTest.xml")));
      AnalysisEngine ae = UIMAFramework.produceAnalysisEngine(aeDesc);
      FlowControllerForErrorTest.reset();
      CAS cas = ae.newCAS();
      //try document that should succeed
      cas.setDocumentText("This is OK");

View Full Code Here

    // Test that an aggregate can be copied preserving all comments and ordering of delegates
    XMLParser.ParsingOptions parsingOptions = new XMLParser.ParsingOptions(false);
    parsingOptions.preserveComments = true;
    XMLParser parser = UIMAFramework.getXMLParser();
    File inFile = JUnitExtension.getFile("TextAnalysisEngineImplTest/AggregateWithManyDelegates.xml");
    AnalysisEngineDescription desc = parser.parseAnalysisEngineDescription(new XMLInputSource(inFile), parsingOptions);


    // Write out descriptor
    File cloneFile = new File(inFile.getParentFile(), "CopyOfAggregateWithManyDelegates.xml");
    BufferedOutputStream os = new BufferedOutputStream(new FileOutputStream(cloneFile));
    XMLSerializer xmlSerializer = new XMLSerializer(true);
    xmlSerializer.setOutputStream(os);
    // set the amount to a value which will show up if used
    // indent should not be used because we're using a parser mode which preserves
    // comments and ignorable white space.
    xmlSerializer.setOutputProperty("{http://xml.apache.org/xslt}indent-amount", "4");
    ContentHandler contentHandler = xmlSerializer.getContentHandler();
    contentHandler.startDocument();
    desc.toXML(contentHandler, true);
    contentHandler.endDocument();
    os.close();
    
    long diff = cloneFile.length() - inFile.length();
    // One platform inserts a blank line and a final newline, so don't insist on perfection
    assertTrue("File size changed by "+diff+" should be no more than 2", diff >= -2 && diff <= 2);


    // Initialize all delegates and check the initialization order (should be declaration order)
    TestAnnotator2.allContexts = "";
    UIMAFramework.produceAnalysisEngine(desc);
    assertEquals("D/C/B/A/F/E/", TestAnnotator2.allContexts);
    
    // Check that copying aggregate preserved the order of the delegates
    desc = parser.parseAnalysisEngineDescription(new XMLInputSource(cloneFile), parsingOptions);
    TestAnnotator2.allContexts = "";
    UIMAFramework.produceAnalysisEngine(desc);
    assertEquals("D/C/B/A/F/E/", TestAnnotator2.allContexts);
//    cloneFile.delete();
  }

View Full Code Here

  
  public void testGetSofa() throws Exception {
    try {
      File typeSystemFile = JUnitExtension.getFile("ExampleCas/testTypeSystem.xml");
      TypeSystemDescription typeSystem = UIMAFramework.getXMLParser().parseTypeSystemDescription(
              new XMLInputSource(typeSystemFile));
      CAS newCas = CasCreationUtils.createCas(typeSystem, null, null);
      File xcasFile = JUnitExtension.getFile("ExampleCas/multiSofaCas.xml"); 
      XCASDeserializer.deserialize(new FileInputStream(xcasFile), newCas);
      JCas newJCas = newCas.getJCas();

View Full Code Here

      File baseDescriptorFile = JUnitExtension
              .getFile("ImportImplTest/subdir/subdir2/AggregateTaeForNestedImportTest.xml");
      File importedFile = JUnitExtension
              .getFile("ImportImplTest/subdir/PrimitiveTaeForNestedImportTest.xml");
      AnalysisEngineDescription_impl agg = (AnalysisEngineDescription_impl) UIMAFramework.getXMLParser()
              .parseAnalysisEngineDescription(new XMLInputSource(baseDescriptorFile));
      assertEquals(baseDescriptorFile.toURL(), agg.getSourceUrl());


      AnalysisEngineDescription_impl prim = (AnalysisEngineDescription_impl) agg.getDelegateAnalysisEngineSpecifiers()
              .get("Annotator1");
      assertEquals(importedFile.toURL(), prim.getSourceUrl());

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.uima.util.XMLInputSource

org.apache.clerezza.uima.utils.ExternalServicesFacade

org.apache.ctakes.core.cc.XmiWriterCasConsumerCtakes

org.apache.ctakes.core.cpe.CmdLineCpeRunner

org.apache.ctakes.relationextractor.eval.SHARPXMI

org.apache.ctakes.utils.test.TestUtil

org.apache.ctakes.ytex.tools.DBAnnotationViewerMain

org.apache.lucene.analysis.uima.ae.BasicAEProvider

org.apache.uima.aae.AsynchAECasManager_impl

org.apache.uima.aae.controller.AggregateAnalysisEngineController_impl

org.apache.uima.aae.deployment.impl.AEService_Impl

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.