1   /**
2    * Distribution License:
3    * JSword is free software; you can redistribute it and/or modify it under
4    * the terms of the GNU Lesser General Public License, version 2.1 as published by
5    * the Free Software Foundation. This program is distributed in the hope
6    * that it will be useful, but WITHOUT ANY WARRANTY; without even the
7    * implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
8    * See the GNU Lesser General Public License for more details.
9    *
10   * The License is available on the internet at:
11   *       http://www.gnu.org/copyleft/lgpl.html
12   * or by writing to:
13   *      Free Software Foundation, Inc.
14   *      59 Temple Place - Suite 330
15   *      Boston, MA 02111-1307, USA
16   *
17   * Copyright: 2007
18   *     The copyright to this program is held by it's authors.
19   *
20   * ID: $Id: KeyAnalyzer.java 1376 2007-06-01 18:27:01Z dmsmith $
21   */
22  package org.crosswire.jsword.index.lucene.analysis;
23  
24  import java.io.IOException;
25  import java.io.Reader;
26  
27  import org.apache.lucene.analysis.StopFilter;
28  import org.apache.lucene.analysis.TokenStream;
29  import org.apache.lucene.analysis.standard.StandardTokenizer;
30  import org.apache.lucene.analysis.th.ThaiWordFilter;
31  import org.apache.lucene.util.Version;
32  
33  /**
34   * Tokenization using ThaiWordFilter. It uses java.text.BreakIterator to break
35   * words. Stemming: Not implemented
36   * 
37   * @see gnu.lgpl.License for license details.<br>
38   *      The copyright to this program is held by it's authors.
39   * @author sijo cherian [sijocherian at yahoo dot com]
40   */
41  public class ThaiLuceneAnalyzer extends AbstractBookAnalyzer {
42  
43      public ThaiLuceneAnalyzer() {
44      }
45  
46      @Override
47      public TokenStream tokenStream(String fieldName, Reader reader) {
48          TokenStream ts = new StandardTokenizer(matchVersion, reader);
49          ts = new ThaiWordFilter(ts);
50          if (doStopWords && stopSet != null) {
51              ts = new StopFilter(false, ts, stopSet);
52          }
53          return ts;
54      }
55  
56      /* (non-Javadoc)
57       * @see org.apache.lucene.analysis.Analyzer#reusableTokenStream(java.lang.String, java.io.Reader)
58       */
59      @Override
60      public TokenStream reusableTokenStream(String fieldName, Reader reader) throws IOException {
61          SavedStreams streams = (SavedStreams) getPreviousTokenStream();
62          if (streams == null) {
63              streams = new SavedStreams(new StandardTokenizer(matchVersion, reader));
64              streams.setResult(new ThaiWordFilter(streams.getResult()));
65  
66              if (doStopWords && stopSet != null) {
67                  streams.setResult(new StopFilter(StopFilter.getEnablePositionIncrementsVersionDefault(matchVersion), streams.getResult(), stopSet));
68              }
69  
70              setPreviousTokenStream(streams);
71          } else {
72              streams.getSource().reset(reader);
73              streams.getResult().reset(); // reset the ThaiWordFilter's state
74          }
75          return streams.getResult();
76      }
77  
78      private final Version matchVersion = Version.LUCENE_29;
79  }
80