1 package org.apache.lucene.analysis.ru;
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 java.io.Reader;
21 import org.apache.lucene.analysis.CharTokenizer;
22 import org.apache.lucene.analysis.Tokenizer; // for javadocs
23 import org.apache.lucene.analysis.LetterTokenizer; // for javadocs
24 import org.apache.lucene.analysis.standard.StandardTokenizer; // for javadocs
25 import org.apache.lucene.util.AttributeSource;
26 import org.apache.lucene.util.Version;
29 * A RussianLetterTokenizer is a {@link Tokenizer} that extends {@link LetterTokenizer}
30 * by also allowing the basic Latin digits 0-9.
33 * You must specify the required {@link Version} compatibility when creating
34 * {@link RussianLetterTokenizer}:
36 * <li>As of 3.1, {@link CharTokenizer} uses an int based API to normalize and
37 * detect token characters. See {@link CharTokenizer#isTokenChar(int)} and
38 * {@link CharTokenizer#normalize(int)} for details.</li>
40 * @deprecated Use {@link StandardTokenizer} instead, which has the same functionality.
41 * This filter will be removed in Lucene 5.0
44 public class RussianLetterTokenizer extends CharTokenizer
46 private static final int DIGIT_0 = '0';
47 private static final int DIGIT_9 = '9';
50 * Construct a new RussianLetterTokenizer. * @param matchVersion Lucene version
51 * to match See {@link <a href="#version">above</a>}
54 * the input to split up into tokens
56 public RussianLetterTokenizer(Version matchVersion, Reader in) {
57 super(matchVersion, in);
61 * Construct a new RussianLetterTokenizer using a given {@link AttributeSource}.
64 * Lucene version to match See {@link <a href="#version">above</a>}
66 * the attribute source to use for this {@link Tokenizer}
68 * the input to split up into tokens
70 public RussianLetterTokenizer(Version matchVersion, AttributeSource source, Reader in) {
71 super(matchVersion, source, in);
75 * Construct a new RussianLetterTokenizer using a given
76 * {@link org.apache.lucene.util.AttributeSource.AttributeFactory}. * @param
77 * matchVersion Lucene version to match See
78 * {@link <a href="#version">above</a>}
81 * the attribute factory to use for this {@link Tokenizer}
83 * the input to split up into tokens
85 public RussianLetterTokenizer(Version matchVersion, AttributeFactory factory, Reader in) {
86 super(matchVersion, factory, in);
90 * Construct a new RussianLetterTokenizer.
92 * @deprecated use {@link #RussianLetterTokenizer(Version, Reader)} instead. This will
93 * be removed in Lucene 4.0.
96 public RussianLetterTokenizer(Reader in) {
101 * Construct a new RussianLetterTokenizer using a given {@link AttributeSource}.
103 * @deprecated use {@link #RussianLetterTokenizer(Version, AttributeSource, Reader)}
104 * instead. This will be removed in Lucene 4.0.
107 public RussianLetterTokenizer(AttributeSource source, Reader in) {
112 * Construct a new RussianLetterTokenizer using a given
113 * {@link org.apache.lucene.util.AttributeSource.AttributeFactory}.
115 * @deprecated use {@link #RussianLetterTokenizer(Version, AttributeSource.AttributeFactory, Reader)}
116 * instead. This will be removed in Lucene 4.0.
119 public RussianLetterTokenizer(AttributeFactory factory, Reader in) {
125 * Collects only characters which satisfy
126 * {@link Character#isLetter(int)}.
129 protected boolean isTokenChar(int c) {
130 return Character.isLetter(c) || (c >= DIGIT_0 && c <= DIGIT_9);