Classes in this File | Line Coverage | Branch Coverage | Complexity | ||||
PersianLuceneAnalyzer |
|
| 2.6666666666666665;2.667 |
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 or later | |
5 | * as published by the Free Software Foundation. This program is distributed | |
6 | * in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even | |
7 | * the 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 | * © CrossWire Bible Society, 2009 - 2016 | |
18 | * | |
19 | */ | |
20 | package org.crosswire.jsword.index.lucene.analysis; | |
21 | ||
22 | import java.io.IOException; | |
23 | import java.io.Reader; | |
24 | ||
25 | import org.apache.lucene.analysis.LowerCaseFilter; | |
26 | import org.apache.lucene.analysis.StopFilter; | |
27 | import org.apache.lucene.analysis.TokenStream; | |
28 | import org.apache.lucene.analysis.ar.ArabicLetterTokenizer; | |
29 | import org.apache.lucene.analysis.ar.ArabicNormalizationFilter; | |
30 | import org.apache.lucene.analysis.fa.PersianAnalyzer; | |
31 | import org.apache.lucene.analysis.fa.PersianNormalizationFilter; | |
32 | import org.apache.lucene.util.Version; | |
33 | ||
34 | /** | |
35 | * An Analyzer whose {@link TokenStream} is built from a | |
36 | * {@link ArabicLetterTokenizer} filtered with {@link LowerCaseFilter}, | |
37 | * {@link ArabicNormalizationFilter}, {@link PersianNormalizationFilter} and | |
38 | * Persian {@link StopFilter} (optional) | |
39 | * | |
40 | * @see gnu.lgpl.License The GNU Lesser General Public License for details. | |
41 | * @author DM Smith | |
42 | */ | |
43 | public class PersianLuceneAnalyzer extends AbstractBookAnalyzer { | |
44 | 0 | public PersianLuceneAnalyzer() { |
45 | 0 | stopSet = PersianAnalyzer.getDefaultStopSet(); |
46 | 0 | } |
47 | ||
48 | /* | |
49 | * (non-Javadoc) | |
50 | * | |
51 | * @see org.apache.lucene.analysis.Analyzer#tokenStream(java.lang.String, | |
52 | * java.io.Reader) | |
53 | */ | |
54 | @Override | |
55 | public final TokenStream tokenStream(String fieldName, Reader reader) { | |
56 | 0 | TokenStream result = new ArabicLetterTokenizer(reader); |
57 | 0 | result = new LowerCaseFilter(result); |
58 | 0 | result = new ArabicNormalizationFilter(result); |
59 | /* additional persian-specific normalization */ | |
60 | 0 | result = new PersianNormalizationFilter(result); |
61 | /* | |
62 | * the order here is important: the stop set is normalized with the | |
63 | * above! | |
64 | */ | |
65 | 0 | if (doStopWords && stopSet != null) { |
66 | 0 | result = new StopFilter(StopFilter.getEnablePositionIncrementsVersionDefault(matchVersion), result, stopSet); |
67 | } | |
68 | ||
69 | 0 | return result; |
70 | } | |
71 | ||
72 | /** | |
73 | * Returns a (possibly reused) {@link TokenStream} which tokenizes all the | |
74 | * text in the provided {@link Reader}. | |
75 | * | |
76 | * @return A {@link TokenStream} built from a {@link ArabicLetterTokenizer} | |
77 | * filtered with {@link LowerCaseFilter}, | |
78 | * {@link ArabicNormalizationFilter}, | |
79 | * {@link PersianNormalizationFilter} and Persian Stop words | |
80 | */ | |
81 | @Override | |
82 | public TokenStream reusableTokenStream(String fieldName, Reader reader) throws IOException { | |
83 | 0 | SavedStreams streams = (SavedStreams) getPreviousTokenStream(); |
84 | 0 | if (streams == null) { |
85 | 0 | streams = new SavedStreams(new ArabicLetterTokenizer(reader)); |
86 | 0 | streams.setResult(new LowerCaseFilter(streams.getResult())); |
87 | 0 | streams.setResult(new ArabicNormalizationFilter(streams.getResult())); |
88 | /* additional persian-specific normalization */ | |
89 | 0 | streams.setResult(new PersianNormalizationFilter(streams.getResult())); |
90 | /* | |
91 | * the order here is important: the stop set is normalized with the | |
92 | * above! | |
93 | */ | |
94 | 0 | if (doStopWords && stopSet != null) { |
95 | 0 | streams.setResult(new StopFilter(false, streams.getResult(), stopSet)); |
96 | } | |
97 | 0 | setPreviousTokenStream(streams); |
98 | } else { | |
99 | 0 | streams.getSource().reset(reader); |
100 | } | |
101 | 0 | return streams.getResult(); |
102 | } | |
103 | 0 | private final Version matchVersion = Version.LUCENE_29; |
104 | } |