Examples with PlainLiteralImpl org.apache.clerezza.commons.rdf.impl.utils.PlainLiteralImpl used on opensource projects

Example 16 with PlainLiteralImpl

use of org.apache.clerezza.commons.rdf.impl.utils.PlainLiteralImpl in project stanbol by apache.

the class CeliLemmatizerEnhancementEngineTest method testCompleteMorphoAnalysis.

@Test
public void testCompleteMorphoAnalysis() throws Exception {
    ContentItem ci = wrapAsContentItem(TERM);
    //add a simple triple to statically define the language of the test
    //content
    ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE, new PlainLiteralImpl("it")));
    CeliLemmatizerEnhancementEngine morphoAnalysisEngine = initEngine(true);
    try {
        morphoAnalysisEngine.computeEnhancements(ci);
    } catch (EngineException e) {
        RemoteServiceHelper.checkServiceUnavailable(e);
        return;
    }
    TestUtils.logEnhancements(ci);
    //validate enhancements
    HashMap<IRI, RDFTerm> expectedValues = new HashMap<IRI, RDFTerm>();
    expectedValues.put(Properties.ENHANCER_EXTRACTED_FROM, ci.getUri());
    expectedValues.put(Properties.DC_CREATOR, LiteralFactory.getInstance().createTypedLiteral(morphoAnalysisEngine.getClass().getName()));
    Iterator<Triple> textAnnotationIterator = ci.getMetadata().filter(null, RDF_TYPE, ENHANCER_TEXTANNOTATION);
    // test if a textAnnotation is present
    //assertTrue(textAnnotationIterator.hasNext()); 
    //  -> this might be used to test that there are no TextAnnotations
    int textAnnotationCount = 0;
    while (textAnnotationIterator.hasNext()) {
        IRI textAnnotation = (IRI) textAnnotationIterator.next().getSubject();
        // test if selected Text is added
        validateTextAnnotation(ci.getMetadata(), textAnnotation, TERM, expectedValues);
        textAnnotationCount++;
        //perform additional tests for "hasMorphologicalFeature" and "hasLemmaForm"
        validateMorphoFeatureProperty(ci.getMetadata(), textAnnotation);
    }
    log.info("{} TextAnnotations found and validated ...", textAnnotationCount);
    int entityAnnoNum = validateAllEntityAnnotations(ci.getMetadata(), expectedValues);
    //no EntityAnnotations expected
    Assert.assertEquals("No EntityAnnotations expected by this test", 0, entityAnnoNum);
    shutdownEngine(morphoAnalysisEngine);
}

Example 17 with PlainLiteralImpl

use of org.apache.clerezza.commons.rdf.impl.utils.PlainLiteralImpl in project stanbol by apache.

the class CeliLemmatizerEnhancementEngineTest method testEngine.

@Test
public void testEngine() throws Exception {
    ContentItem ci = wrapAsContentItem(TEXT);
    //add a simple triple to statically define the language of the test
    //content
    ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE, new PlainLiteralImpl("it")));
    //unit test should not depend on each other (if possible)
    //CeliLanguageIdentifierEnhancementEngineTest.addEnanchements(ci);
    CeliLemmatizerEnhancementEngine morphoAnalysisEngine = initEngine(false);
    try {
        morphoAnalysisEngine.computeEnhancements(ci);
    } catch (EngineException e) {
        RemoteServiceHelper.checkServiceUnavailable(e);
        return;
    }
    TestUtils.logEnhancements(ci);
    //validate enhancement
    HashMap<IRI, RDFTerm> expectedValues = new HashMap<IRI, RDFTerm>();
    expectedValues.put(Properties.ENHANCER_EXTRACTED_FROM, ci.getUri());
    expectedValues.put(Properties.DC_CREATOR, LiteralFactory.getInstance().createTypedLiteral(morphoAnalysisEngine.getClass().getName()));
    Iterator<Triple> lemmaTextAnnotationIterator = ci.getMetadata().filter(null, RDF_TYPE, ENHANCER_TEXTANNOTATION);
    assertTrue("A TextAnnotation is expected by this Test", lemmaTextAnnotationIterator.hasNext());
    BlankNodeOrIRI lemmaTextAnnotation = lemmaTextAnnotationIterator.next().getSubject();
    assertTrue("TextAnnoations MUST BE IRIs!", lemmaTextAnnotation instanceof IRI);
    assertFalse("Only a single TextAnnotation is expected by this Test", lemmaTextAnnotationIterator.hasNext());
    //validate the enhancement metadata
    validateEnhancement(ci.getMetadata(), (IRI) lemmaTextAnnotation, expectedValues);
    //validate the lemma form TextAnnotation
    int lemmaForms = validateLemmaFormProperty(ci.getMetadata(), lemmaTextAnnotation, "it");
    assertTrue("Only a single LemmaForm property is expected if '" + MORPHOLOGICAL_ANALYSIS + "=false'", lemmaForms == 1);
    shutdownEngine(morphoAnalysisEngine);
}

Example 18 with PlainLiteralImpl

use of org.apache.clerezza.commons.rdf.impl.utils.PlainLiteralImpl in project stanbol by apache.

the class CeliNamedEntityExtractionEnhancementEngineTest method testInput.

private void testInput(String txt, String lang) throws EngineException, IOException {
    ContentItem ci = wrapAsContentItem(txt);
    try {
        //add a simple triple to statically define the language of the test content
        ci.getMetadata().add(new TripleImpl(ci.getUri(), DC_LANGUAGE, new PlainLiteralImpl(lang)));
        nerEngine.computeEnhancements(ci);
        TestUtils.logEnhancements(ci);
        HashMap<IRI, RDFTerm> expectedValues = new HashMap<IRI, RDFTerm>();
        expectedValues.put(Properties.ENHANCER_EXTRACTED_FROM, ci.getUri());
        expectedValues.put(Properties.DC_CREATOR, LiteralFactory.getInstance().createTypedLiteral(nerEngine.getClass().getName()));
        int textAnnoNum = validateAllTextAnnotations(ci.getMetadata(), txt, expectedValues);
        log.info(textAnnoNum + " TextAnnotations found ...");
        int entityAnnoNum = EnhancementStructureHelper.validateAllEntityAnnotations(ci.getMetadata(), expectedValues);
        log.info(entityAnnoNum + " EntityAnnotations found ...");
    } catch (EngineException e) {
        RemoteServiceHelper.checkServiceUnavailable(e);
    }
}

Example 19 with PlainLiteralImpl

use of org.apache.clerezza.commons.rdf.impl.utils.PlainLiteralImpl in project stanbol by apache.

the class CeliNamedEntityExtractionEnhancementEngine method computeEnhancements.

@Override
public void computeEnhancements(ContentItem ci) throws EngineException {
    Entry<IRI, Blob> contentPart = ContentItemHelper.getBlob(ci, SUPPORTED_MIMTYPES);
    if (contentPart == null) {
        throw new IllegalStateException("No ContentPart with Mimetype '" + TEXT_PLAIN_MIMETYPE + "' found for ContentItem " + ci.getUri() + ": This is also checked in the canEnhance method! -> This " + "indicated an Bug in the implementation of the " + "EnhancementJobManager!");
    }
    String text = "";
    try {
        text = ContentItemHelper.getText(contentPart.getValue());
    } catch (IOException e) {
        throw new InvalidContentException(this, ci, e);
    }
    if (text.trim().length() == 0) {
        log.info("No text contained in ContentPart {" + contentPart.getKey() + "} of ContentItem {" + ci.getUri() + "}");
        return;
    }
    String language = EnhancementEngineHelper.getLanguage(ci);
    if (language == null) {
        throw new IllegalStateException("Unable to extract Language for " + "ContentItem " + ci.getUri() + ": This is also checked in the canEnhance " + "method! -> This indicated an Bug in the implementation of the " + "EnhancementJobManager!");
    }
    //used for the palin literals in TextAnnotations
    Language lang = new Language(language);
    try {
        List<NamedEntity> lista = this.client.extractEntities(text, language);
        LiteralFactory literalFactory = LiteralFactory.getInstance();
        Graph g = ci.getMetadata();
        for (NamedEntity ne : lista) {
            try {
                IRI textAnnotation = EnhancementEngineHelper.createTextEnhancement(ci, this);
                //add selected text as PlainLiteral in the language extracted from the text
                g.add(new TripleImpl(textAnnotation, ENHANCER_SELECTED_TEXT, new PlainLiteralImpl(ne.getFormKind(), lang)));
                g.add(new TripleImpl(textAnnotation, DC_TYPE, getEntityRefForType(ne.type)));
                if (ne.getFrom() != null && ne.getTo() != null) {
                    g.add(new TripleImpl(textAnnotation, ENHANCER_START, literalFactory.createTypedLiteral(ne.getFrom().intValue())));
                    g.add(new TripleImpl(textAnnotation, ENHANCER_END, literalFactory.createTypedLiteral(ne.getTo().intValue())));
                    g.add(new TripleImpl(textAnnotation, ENHANCER_SELECTION_CONTEXT, new PlainLiteralImpl(getSelectionContext(text, ne.getFormKind(), ne.getFrom().intValue()), lang)));
                }
            } catch (NoConvertorException e) {
                log.error(e.getMessage(), e);
            }
        }
    } catch (IOException e) {
        throw new EngineException("Error while calling the CELI NER (Named Entity Recognition)" + " service (configured URL: " + serviceURL + ")!", e);
    } catch (SOAPException e) {
        throw new EngineException("Error wile encoding/decoding the request/" + "response to the CELI NER (Named Entity Recognition) service!", e);
    }
}

Example 20 with PlainLiteralImpl

use of org.apache.clerezza.commons.rdf.impl.utils.PlainLiteralImpl in project stanbol by apache.

the class SpotlightEngineUtils method createEntityAnnotation.

/**
	 * Creates a fise:EntityAnnotation for the parsed parameters and
     * adds it the the {@link ContentItem#getMetadata()}. <p>
     * This method assumes a write lock on the parsed content item.
	 * @param resource the candidate resource
	 * @param engine the engine
	 * @param ci the content item
	 * @param textAnnotation the fise:TextAnnotation to dc:relate the
	 * created fise:EntityAnnotation
	 * @return the URI of the created fise:TextAnnotation
	 */
public static IRI createEntityAnnotation(CandidateResource resource, EnhancementEngine engine, ContentItem ci, IRI textAnnotation) {
    IRI entityAnnotation = EnhancementEngineHelper.createEntityEnhancement(ci, engine);
    Graph model = ci.getMetadata();
    Literal label = new PlainLiteralImpl(resource.label, new Language("en"));
    model.add(new TripleImpl(entityAnnotation, DC_RELATION, textAnnotation));
    model.add(new TripleImpl(entityAnnotation, ENHANCER_ENTITY_LABEL, label));
    model.add(new TripleImpl(entityAnnotation, ENHANCER_ENTITY_REFERENCE, resource.getUri()));
    model.add(new TripleImpl(entityAnnotation, PROPERTY_CONTEXTUAL_SCORE, literalFactory.createTypedLiteral(resource.contextualScore)));
    model.add(new TripleImpl(entityAnnotation, PROPERTY_PERCENTAGE_OF_SECOND_RANK, literalFactory.createTypedLiteral(resource.percentageOfSecondRank)));
    model.add(new TripleImpl(entityAnnotation, PROPERTY_SUPPORT, literalFactory.createTypedLiteral(resource.support)));
    model.add(new TripleImpl(entityAnnotation, PROPERTY_PRIOR_SCORE, literalFactory.createTypedLiteral(resource.priorScore)));
    model.add(new TripleImpl(entityAnnotation, PROPERTY_FINAL_SCORE, literalFactory.createTypedLiteral(resource.finalScore)));
    return entityAnnotation;
}