1 package org.apache.lucene.analysis.fr;
4 * Licensed to the Apache Software Foundation (ASF) under one or more
5 * contributor license agreements. See the NOTICE file distributed with
6 * this work for additional information regarding copyright ownership.
7 * The ASF licenses this file to You under the Apache License, Version 2.0
8 * (the "License"); you may not use this file except in compliance with
9 * the License. You may obtain a copy of the License at
11 * http://www.apache.org/licenses/LICENSE-2.0
13 * Unless required by applicable law or agreed to in writing, software
14 * distributed under the License is distributed on an "AS IS" BASIS,
15 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16 * See the License for the specific language governing permissions and
17 * limitations under the License.
20 import org.apache.lucene.analysis.Analyzer;
21 import org.apache.lucene.analysis.CharArraySet;
22 import org.apache.lucene.analysis.LowerCaseFilter;
23 import org.apache.lucene.analysis.KeywordMarkerFilter;
24 import org.apache.lucene.analysis.StopFilter;
25 import org.apache.lucene.analysis.StopwordAnalyzerBase;
26 import org.apache.lucene.analysis.TokenStream;
27 import org.apache.lucene.analysis.Tokenizer;
28 import org.apache.lucene.analysis.WordlistLoader;
29 import org.apache.lucene.analysis.snowball.SnowballFilter;
30 import org.apache.lucene.analysis.standard.StandardFilter;
31 import org.apache.lucene.analysis.standard.StandardTokenizer;
32 import org.apache.lucene.analysis.standard.StandardAnalyzer; // for javadoc
33 import org.apache.lucene.util.Version;
36 import java.io.IOException;
37 import java.io.Reader;
38 import java.util.Arrays;
39 import java.util.HashSet;
40 import java.util.Collections;
45 * {@link Analyzer} for French language.
47 * Supports an external list of stopwords (words that
48 * will not be indexed at all) and an external list of exclusions (word that will
49 * not be stemmed, but indexed).
50 * A default set of stopwords is used unless an alternative list is specified, but the
51 * exclusion list is empty by default.
55 * <p>You must specify the required {@link Version}
56 * compatibility when creating FrenchAnalyzer:
58 * <li> As of 3.1, Snowball stemming is done with SnowballFilter,
59 * LowerCaseFilter is used prior to StopFilter, and ElisionFilter and
60 * Snowball stopwords are used by default.
61 * <li> As of 2.9, StopFilter preserves position
65 * <p><b>NOTE</b>: This class uses the same {@link Version}
66 * dependent settings as {@link StandardAnalyzer}.</p>
68 public final class FrenchAnalyzer extends StopwordAnalyzerBase {
71 * Extended list of typical French stopwords.
72 * @deprecated use {@link #getDefaultStopSet()} instead
75 public final static String[] FRENCH_STOP_WORDS = {
76 "a", "afin", "ai", "ainsi", "après", "attendu", "au", "aujourd", "auquel", "aussi",
77 "autre", "autres", "aux", "auxquelles", "auxquels", "avait", "avant", "avec", "avoir",
78 "c", "car", "ce", "ceci", "cela", "celle", "celles", "celui", "cependant", "certain",
79 "certaine", "certaines", "certains", "ces", "cet", "cette", "ceux", "chez", "ci",
80 "combien", "comme", "comment", "concernant", "contre", "d", "dans", "de", "debout",
81 "dedans", "dehors", "delà", "depuis", "derrière", "des", "désormais", "desquelles",
82 "desquels", "dessous", "dessus", "devant", "devers", "devra", "divers", "diverse",
83 "diverses", "doit", "donc", "dont", "du", "duquel", "durant", "dès", "elle", "elles",
84 "en", "entre", "environ", "est", "et", "etc", "etre", "eu", "eux", "excepté", "hormis",
85 "hors", "hélas", "hui", "il", "ils", "j", "je", "jusqu", "jusque", "l", "la", "laquelle",
86 "le", "lequel", "les", "lesquelles", "lesquels", "leur", "leurs", "lorsque", "lui", "là",
87 "ma", "mais", "malgré", "me", "merci", "mes", "mien", "mienne", "miennes", "miens", "moi",
88 "moins", "mon", "moyennant", "même", "mêmes", "n", "ne", "ni", "non", "nos", "notre",
89 "nous", "néanmoins", "nôtre", "nôtres", "on", "ont", "ou", "outre", "où", "par", "parmi",
90 "partant", "pas", "passé", "pendant", "plein", "plus", "plusieurs", "pour", "pourquoi",
91 "proche", "près", "puisque", "qu", "quand", "que", "quel", "quelle", "quelles", "quels",
92 "qui", "quoi", "quoique", "revoici", "revoilà", "s", "sa", "sans", "sauf", "se", "selon",
93 "seront", "ses", "si", "sien", "sienne", "siennes", "siens", "sinon", "soi", "soit",
94 "son", "sont", "sous", "suivant", "sur", "ta", "te", "tes", "tien", "tienne", "tiennes",
95 "tiens", "toi", "ton", "tous", "tout", "toute", "toutes", "tu", "un", "une", "va", "vers",
96 "voici", "voilà", "vos", "votre", "vous", "vu", "vôtre", "vôtres", "y", "à", "ça", "ès",
100 /** File containing default French stopwords. */
101 public final static String DEFAULT_STOPWORD_FILE = "french_stop.txt";
104 * Contains words that should be indexed but not stemmed.
106 private Set<?> excltable = Collections.<Object>emptySet();
109 * Returns an unmodifiable instance of the default stop-words set.
110 * @return an unmodifiable instance of the default stop-words set.
112 public static Set<?> getDefaultStopSet(){
113 return DefaultSetHolder.DEFAULT_STOP_SET;
116 private static class DefaultSetHolder {
117 /** @deprecated remove this in Lucene 5.0 */
119 static final Set<?> DEFAULT_STOP_SET_30 = CharArraySet
120 .unmodifiableSet(new CharArraySet(Version.LUCENE_CURRENT, Arrays.asList(FRENCH_STOP_WORDS),
122 static final Set<?> DEFAULT_STOP_SET;
126 WordlistLoader.getSnowballWordSet(SnowballFilter.class, DEFAULT_STOPWORD_FILE);
127 } catch (IOException ex) {
128 // default set should always be present as it is part of the
129 // distribution (JAR)
130 throw new RuntimeException("Unable to load default stopword set");
136 * Builds an analyzer with the default stop words ({@link #getDefaultStopSet}).
138 public FrenchAnalyzer(Version matchVersion) {
140 matchVersion.onOrAfter(Version.LUCENE_31) ? DefaultSetHolder.DEFAULT_STOP_SET
141 : DefaultSetHolder.DEFAULT_STOP_SET_30);
145 * Builds an analyzer with the given stop words
147 * @param matchVersion
148 * lucene compatibility version
152 public FrenchAnalyzer(Version matchVersion, Set<?> stopwords){
153 this(matchVersion, stopwords, CharArraySet.EMPTY_SET);
157 * Builds an analyzer with the given stop words
159 * @param matchVersion
160 * lucene compatibility version
163 * @param stemExclutionSet
164 * a stemming exclusion set
166 public FrenchAnalyzer(Version matchVersion, Set<?> stopwords,
167 Set<?> stemExclutionSet) {
168 super(matchVersion, stopwords);
169 this.excltable = CharArraySet.unmodifiableSet(CharArraySet
170 .copy(matchVersion, stemExclutionSet));
175 * Builds an analyzer with the given stop words.
176 * @deprecated use {@link #FrenchAnalyzer(Version, Set)} instead
179 public FrenchAnalyzer(Version matchVersion, String... stopwords) {
180 this(matchVersion, StopFilter.makeStopSet(matchVersion, stopwords));
184 * Builds an analyzer with the given stop words.
185 * @throws IOException
186 * @deprecated use {@link #FrenchAnalyzer(Version, Set)} instead
189 public FrenchAnalyzer(Version matchVersion, File stopwords) throws IOException {
190 this(matchVersion, WordlistLoader.getWordSet(stopwords));
194 * Builds an exclusionlist from an array of Strings.
195 * @deprecated use {@link #FrenchAnalyzer(Version, Set, Set)} instead
198 public void setStemExclusionTable(String... exclusionlist) {
199 excltable = StopFilter.makeStopSet(matchVersion, exclusionlist);
200 setPreviousTokenStream(null); // force a new stemmer to be created
204 * Builds an exclusionlist from a Map.
205 * @deprecated use {@link #FrenchAnalyzer(Version, Set, Set)} instead
208 public void setStemExclusionTable(Map<?,?> exclusionlist) {
209 excltable = new HashSet<Object>(exclusionlist.keySet());
210 setPreviousTokenStream(null); // force a new stemmer to be created
214 * Builds an exclusionlist from the words contained in the given file.
215 * @throws IOException
216 * @deprecated use {@link #FrenchAnalyzer(Version, Set, Set)} instead
219 public void setStemExclusionTable(File exclusionlist) throws IOException {
220 excltable = new HashSet<Object>(WordlistLoader.getWordSet(exclusionlist));
221 setPreviousTokenStream(null); // force a new stemmer to be created
226 * {@link org.apache.lucene.analysis.ReusableAnalyzerBase.TokenStreamComponents}
227 * used to tokenize all the text in the provided {@link Reader}.
229 * @return {@link org.apache.lucene.analysis.ReusableAnalyzerBase.TokenStreamComponents}
230 * built from a {@link StandardTokenizer} filtered with
231 * {@link StandardFilter}, {@link ElisionFilter},
232 * {@link LowerCaseFilter}, {@link StopFilter},
233 * {@link KeywordMarkerFilter} if a stem exclusion set is
234 * provided, and {@link SnowballFilter}
237 protected TokenStreamComponents createComponents(String fieldName,
239 if (matchVersion.onOrAfter(Version.LUCENE_31)) {
240 final Tokenizer source = new StandardTokenizer(matchVersion, reader);
241 TokenStream result = new StandardFilter(matchVersion, source);
242 result = new ElisionFilter(matchVersion, result);
243 result = new LowerCaseFilter(matchVersion, result);
244 result = new StopFilter(matchVersion, result, stopwords);
245 if(!excltable.isEmpty())
246 result = new KeywordMarkerFilter(result, excltable);
247 result = new SnowballFilter(result, new org.tartarus.snowball.ext.FrenchStemmer());
248 return new TokenStreamComponents(source, result);
250 final Tokenizer source = new StandardTokenizer(matchVersion, reader);
251 TokenStream result = new StandardFilter(matchVersion, source);
252 result = new StopFilter(matchVersion, result, stopwords);
253 if(!excltable.isEmpty())
254 result = new KeywordMarkerFilter(result, excltable);
255 result = new FrenchStemFilter(result);
256 // Convert to lowercase after stemming!
257 return new TokenStreamComponents(source, new LowerCaseFilter(matchVersion, result));