Example usage for org.apache.solr.analysis TokenizerChain getCharFilterFactories

Introduction

In this page you can find the example usage for org.apache.solr.analysis TokenizerChain getCharFilterFactories.

Prototype

public CharFilterFactory[] getCharFilterFactories() 

Source Link

Usage

From source file:com.sindicetech.siren.solr.schema.ConciseJsonField.java

License:Open Source License

/**
 * Append the mandatory SIREn filters for the concise model, i.e.,
 * {@link com.sindicetech.siren.solr.analysis.DatatypeAnalyzerFilterFactory},
 * {@link com.sindicetech.siren.solr.analysis.PathEncodingFilterFactory},
 * {@link com.sindicetech.siren.solr.analysis.PositionAttributeFilterFactory} and
 * {@link com.sindicetech.siren.solr.analysis.SirenPayloadFilterFactory}, to the tokenizer chain.
 *
 * @see ExtendedJsonField#appendSirenFilters(org.apache.lucene.analysis.Analyzer, java.util.Map)
 *///from   w w w.  j  av  a 2s .co m
@Override
protected Analyzer appendSirenFilters(final Analyzer analyzer, final Map<String, Datatype> datatypes) {
    if (!(analyzer instanceof TokenizerChain)) {
        throw new SolrException(SolrException.ErrorCode.SERVER_ERROR,
                "Invalid index analyzer '" + analyzer.getClass() + "' received");
    }

    final TokenizerChain chain = (TokenizerChain) analyzer;
    // copy the existing list of token filters
    final TokenFilterFactory[] old = chain.getTokenFilterFactories();
    final TokenFilterFactory[] filterFactories = new TokenFilterFactory[old.length + 4];
    System.arraycopy(old, 0, filterFactories, 0, old.length);
    // append the datatype analyzer filter factory
    final DatatypeAnalyzerFilterFactory datatypeFactory = new DatatypeAnalyzerFilterFactory(
            new HashMap<String, String>());
    datatypeFactory.register(datatypes);
    filterFactories[old.length] = datatypeFactory;
    // append the path encoding filter factory
    filterFactories[old.length + 1] = new PathEncodingFilterFactory(new HashMap<String, String>());
    // append the position attribute filter factory
    filterFactories[old.length + 2] = new PositionAttributeFilterFactory(new HashMap<String, String>());
    // append the siren payload filter factory
    filterFactories[old.length + 3] = new SirenPayloadFilterFactory(new HashMap<String, String>());
    // create a new tokenizer chain with the updated list of filter factories
    return new TokenizerChain(chain.getCharFilterFactories(), chain.getTokenizerFactory(), filterFactories);
}

---

From source file:com.sindicetech.siren.solr.schema.ExtendedJsonField.java

License:Open Source License

/**
 * Append the mandatory SIREn filters, i.e.,
 * {@link DatatypeAnalyzerFilterFactory},
 * {@link PositionAttributeFilterFactory} and
 * {@link SirenPayloadFilterFactory}, to the tokenizer chain.
 * <br/>//w  ww .  ja v  a2 s.  c  om
 * The first time this is called, it will create a
 * {@link com.sindicetech.siren.solr.analysis.DatatypeAnalyzerFilterFactory} with no datatype registered. The datatypes
 * will be loaded and registered later, when {@link #inform(org.apache.lucene.analysis.util.ResourceLoader)} is
 * called.
 * <br/>
 * This is necessary to avoid having to call {@link org.apache.solr.schema.IndexSchema#refreshAnalyzers()}.
 * The {@link org.apache.solr.schema.IndexSchema} will have a reference to the SIREn field's analyzer, and
 * to the {@link com.sindicetech.siren.solr.analysis.DatatypeAnalyzerFilterFactory}. When the datatypes will be loaded,
 * we will access this reference, and register the datatypes.
 */
protected Analyzer appendSirenFilters(final Analyzer analyzer, final Map<String, Datatype> datatypes) {
    if (!(analyzer instanceof TokenizerChain)) {
        throw new SolrException(SolrException.ErrorCode.SERVER_ERROR,
                "Invalid index analyzer '" + analyzer.getClass() + "' received");
    }

    final TokenizerChain chain = (TokenizerChain) analyzer;
    // copy the existing list of token filters
    final TokenFilterFactory[] old = chain.getTokenFilterFactories();
    final TokenFilterFactory[] filterFactories = new TokenFilterFactory[old.length + 3];
    System.arraycopy(old, 0, filterFactories, 0, old.length);
    // append the datatype analyzer filter factory
    final DatatypeAnalyzerFilterFactory datatypeFactory = new DatatypeAnalyzerFilterFactory(
            new HashMap<String, String>());
    datatypeFactory.register(datatypes);
    filterFactories[old.length] = datatypeFactory;
    // append the position attribute filter factory
    filterFactories[old.length + 1] = new PositionAttributeFilterFactory(new HashMap<String, String>());
    // append the siren payload filter factory
    filterFactories[old.length + 2] = new SirenPayloadFilterFactory(new HashMap<String, String>());
    // create a new tokenizer chain with the updated list of filter factories
    return new TokenizerChain(chain.getCharFilterFactories(), chain.getTokenizerFactory(), filterFactories);
}

---

From source file:org.alfresco.solr.AlfrescoFieldType.java

License:Open Source License

private Analyzer constructMultiTermAnalyzer(Analyzer queryAnalyzer) {
    if (queryAnalyzer == null)
        return null;

    if (!(queryAnalyzer instanceof TokenizerChain)) {
        return new KeywordAnalyzer();
    }//  ww  w . j av  a 2 s  .  c  o  m

    TokenizerChain tc = (TokenizerChain) queryAnalyzer;
    MultiTermChainBuilder builder = new MultiTermChainBuilder();

    CharFilterFactory[] charFactories = tc.getCharFilterFactories();
    if (charFactories != null) {
        for (CharFilterFactory fact : charFactories) {
            builder.add(fact);
        }
    }

    builder.add(tc.getTokenizerFactory());

    for (TokenFilterFactory fact : tc.getTokenFilterFactories()) {
        builder.add(fact);
    }

    return builder.build();
}

---