Package org.carrot2.core

Examples of org.carrot2.core.Document


    }
    return solrStopWords.get(fieldName);
  }

  public ILexicalData getLexicalData(LanguageCode languageCode) {
    final ILexicalData carrot2LexicalData = carrot2LexicalDataFactory
        .getLexicalData(languageCode);

    return new ILexicalData() {
      public boolean isStopLabel(CharSequence word) {
        // Nothing in Solr maps to the concept of a stop label,
        // so return Carrot2's default here.
        return carrot2LexicalData.isStopLabel(word);
      }

      public boolean isCommonWord(MutableCharArray word) {
        // Loop over the fields involved in clustering first
        for (String fieldName : fieldNames) {
          for (CharArraySet stopWords : getSolrStopWordsForField(fieldName)) {
            if (stopWords.contains(word)) {
              return true;
            }
          }
        }
        // Check default Carrot2 stop words too
        return carrot2LexicalData.isCommonWord(word);
      }
    };
  }
View Full Code Here


      return;
    }

    // Test with Maltese so that the English clustering performed in other tests
    // is not affected by the test stopwords and stoplabels.
    ILexicalData lexicalData = preprocessing.lexicalDataFactory
        .getLexicalData(LanguageCode.MALTESE);

    for (String word : wordsToCheck.split(",")) {
      if (!lexicalData.isCommonWord(new MutableCharArray(word))
          && !lexicalData.isStopLabel(word)) {
        clusters.add(new Cluster(word));
      }
    }
  }
View Full Code Here

      private final MutableCharArray tempCharSequence;
      private final Class<?> tokenFilterClass;

      private ChineseTokenizer() throws Exception {
        this.tempCharSequence = new MutableCharArray(new char[0]);

        // As Smart Chinese is not available during compile time,
        // we need to resort to reflection.
        final Class<?> tokenizerClass = ReflectionUtils.classForName(
            "org.apache.lucene.analysis.cn.smart.SentenceTokenizer", false);
View Full Code Here

    // is not affected by the test stopwords and stoplabels.
    ILexicalData lexicalData = preprocessing.lexicalDataFactory
        .getLexicalData(LanguageCode.MALTESE);

    for (String word : wordsToCheck.split(",")) {
      if (!lexicalData.isCommonWord(new MutableCharArray(word))
          && !lexicalData.isStopLabel(word)) {
        clusters.add(new Cluster(word));
      }
    }
  }
View Full Code Here

          }
        },
       
        // Using the class loader directly because this time we want to omit the prefix
        new ClassLoaderLocator(core.getResourceLoader().getClassLoader())));
   
    this.controller.init(initAttributes);
    this.idFieldName = core.getSchema().getUniqueKeyField().getName();

    // Make sure the requested Carrot2 clustering algorithm class is available
View Full Code Here

                    .defaultLanguage(LanguageCode.ENGLISH);


                    File resourcesDir = new File(environment.configFile(), "carrot2/resources");

                    ResourceLookup resourceLookup = new ResourceLookup(new DirLocator(resourcesDir));

                    DefaultLexicalDataFactoryDescriptor.attributeBuilder(attributes)
                    .mergeResources(true);
                    LexicalDataLoaderDescriptor.attributeBuilder(attributes)
                    .resourceLookup(resourceLookup);
View Full Code Here

                  + carrot2ResourcesDir);
              final InputStream resourceStream = resourceLoader
                  .openResource(carrot2ResourcesDir + "/" + resource);
             
              log.info(resource + " loaded from " + carrot2ResourcesDir);
              final IResource foundResource = new IResource() {
                public InputStream open() throws IOException {
                  return resourceStream;
                }
              };
              return new IResource[] { foundResource };
View Full Code Here

    initAttributes.put("solrIndexSchema", core.getSchema());

    // Customize Carrot2's resource lookup to first look for resources
    // using Solr's resource loader. If that fails, try loading from the classpath.
    DefaultLexicalDataFactoryDescriptor.attributeBuilder(initAttributes)
        .resourceLookup(new ResourceLookup(new IResourceLocator() {
          public IResource[] getAll(final String resource) {
            final SolrResourceLoader resourceLoader = core.getResourceLoader();
            final String carrot2ResourcesDir = resourceLoader.getConfigDir()
                + initParams.get(CarrotParams.LEXICAL_RESOURCES_DIR, CARROT_RESOURCES_PREFIX);
            try {
View Full Code Here

    initAttributes.put("solrIndexSchema", core.getSchema());

    // Customize Carrot2's resource lookup to first look for resources
    // using Solr's resource loader. If that fails, try loading from the classpath.
    DefaultLexicalDataFactoryDescriptor.attributeBuilder(initAttributes)
        .resourceLookup(new ResourceLookup(new IResourceLocator() {
          public IResource[] getAll(final String resource) {
            final SolrResourceLoader resourceLoader = core.getResourceLoader();
            final String carrot2ResourcesDir = resourceLoader.getConfigDir()
                + initParams.get(CarrotParams.LEXICAL_RESOURCES_DIR, CARROT_RESOURCES_PREFIX);
            try {
View Full Code Here

                    .defaultLanguage(LanguageCode.ENGLISH);


                    File resourcesDir = new File(environment.configFile(), "carrot2/resources");

                    ResourceLookup resourceLookup = new ResourceLookup(new DirLocator(resourcesDir));

                    DefaultLexicalDataFactoryDescriptor.attributeBuilder(attributes)
                    .mergeResources(true);
                    LexicalDataLoaderDescriptor.attributeBuilder(attributes)
                    .resourceLookup(resourceLookup);
View Full Code Here

TOP

Related Classes of org.carrot2.core.Document

Copyright © 2018 www.massapicom. All rights reserved.
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.