KickJava   Java API By Example, From Geeks To Geeks.

Java > Open Source Codes > org > jahia > services > search > analyzer > StandardAnalyzer


1 package org.jahia.services.search.analyzer;
2
3 /* ====================================================================
4  * The Apache Software License, Version 1.1
5  *
6  * Copyright (c) 2001 The Apache Software Foundation. All rights
7  * reserved.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions
11  * are met:
12  *
13  * 1. Redistributions of source code must retain the above copyright
14  * notice, this list of conditions and the following disclaimer.
15  *
16  * 2. Redistributions in binary form must reproduce the above copyright
17  * notice, this list of conditions and the following disclaimer in
18  * the documentation and/or other materials provided with the
19  * distribution.
20  *
21  * 3. The end-user documentation included with the redistribution,
22  * if any, must include the following acknowledgment:
23  * "This product includes software developed by the
24  * Apache Software Foundation (http://www.apache.org/)."
25  * Alternately, this acknowledgment may appear in the software itself,
26  * if and wherever such third-party acknowledgments normally appear.
27  *
28  * 4. The names "Apache" and "Apache Software Foundation" and
29  * "Apache Lucene" must not be used to endorse or promote products
30  * derived from this software without prior written permission. For
31  * written permission, please contact apache@apache.org.
32  *
33  * 5. Products derived from this software may not be called "Apache",
34  * "Apache Lucene", nor may "Apache" appear in their name, without
35  * prior written permission of the Apache Software Foundation.
36  *
37  * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
38  * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
39  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
40  * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
41  * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
42  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
43  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
44  * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
45  * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
46  * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
47  * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
48  * SUCH DAMAGE.
49  * ====================================================================
50  *
51  * This software consists of voluntary contributions made by many
52  * individuals on behalf of the Apache Software Foundation. For more
53  * information on the Apache Software Foundation, please see
54  * <http://www.apache.org/>.
55  */
56
57 import org.apache.lucene.analysis.*;
58 import java.io.Reader JavaDoc;
59 import java.util.Hashtable JavaDoc;
60
61 /**
62  * Filters {@link StandardTokenizer} with {@link StandardFilter}, {@link
63  * LowerCaseFilter} and {@link StopFilter}.
64  *
65  * @version $Id: StandardAnalyzer.java 9130 2005-05-26 10:20:36Z knguyen $
66  */
67 public class StandardAnalyzer extends Analyzer {
68     private Hashtable JavaDoc stopTable;
69
70     private boolean indexeAnalyzer = false;
71
72     /** An array containing some common English words that are usually not
73     useful for searching. */
74     public static final String JavaDoc[] STOP_WORDS = StopAnalyzer.ENGLISH_STOP_WORDS;
75
76     /** Builds an analyzer. */
77     public StandardAnalyzer() {
78         this(false);
79     }
80
81     /** Builds an analyzer. */
82     public StandardAnalyzer(boolean indexAnalyzer) {
83         this(STOP_WORDS);
84         this.indexeAnalyzer = indexAnalyzer;
85     }
86
87     /** Builds an analyzer with the given stop words. */
88     public StandardAnalyzer(String JavaDoc[] stopWords) {
89         this(stopWords,false);
90     }
91
92     public StandardAnalyzer(String JavaDoc[] stopWords, boolean indexAnalyzer) {
93         stopTable = StopFilter.makeStopTable(stopWords);
94         this.indexeAnalyzer = indexAnalyzer;
95     }
96
97     /** Constructs a {@link StandardTokenizer} filtered by a {@link
98     StandardFilter}, a {@link LowerCaseFilter} and a {@link StopFilter}. */
99     public TokenStream tokenStream(String JavaDoc fieldName, Reader JavaDoc reader) {
100         TokenStream result = new StandardTokenizer(reader);
101         result = new StandardFilter(result);
102         result = new LowerCaseFilter(result);
103         result = new StopFilter(result, stopTable);
104         result = new LanguageIndependantFilter(result);
105         if ( this.indexeAnalyzer ){
106             result = new TokenWithQuoteFilter(result);
107         }
108         return result;
109     }
110
111
112 }
113
Popular Tags