1 package org.apache.lucene.analysis.ngram;
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.Tokenizer;
21 import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
22 import org.apache.lucene.analysis.tokenattributes.OffsetAttribute;
23 import org.apache.lucene.util.AttributeSource;
25 import java.io.IOException;
26 import java.io.Reader;
29 * Tokenizes the input from an edge into n-grams of given size(s).
31 * This {@link Tokenizer} create n-grams from the beginning edge or ending edge of a input token.
32 * MaxGram can't be larger than 1024 because of limitation.
35 public final class EdgeNGramTokenizer extends Tokenizer {
36 public static final Side DEFAULT_SIDE = Side.FRONT;
37 public static final int DEFAULT_MAX_GRAM_SIZE = 1;
38 public static final int DEFAULT_MIN_GRAM_SIZE = 1;
40 private final CharTermAttribute termAtt = addAttribute(CharTermAttribute.class);
41 private final OffsetAttribute offsetAtt = addAttribute(OffsetAttribute.class);
43 /** Specifies which side of the input the n-gram should be generated from */
44 public static enum Side {
46 /** Get the n-gram from the front of the input */
49 public String getLabel() { return "front"; }
52 /** Get the n-gram from the end of the input */
55 public String getLabel() { return "back"; }
58 public abstract String getLabel();
60 // Get the appropriate Side from a string
61 public static Side getSide(String sideName) {
62 if (FRONT.getLabel().equals(sideName)) {
65 if (BACK.getLabel().equals(sideName)) {
76 private boolean started = false;
82 * Creates EdgeNGramTokenizer that can generate n-grams in the sizes of the given range
84 * @param input {@link Reader} holding the input to be tokenized
85 * @param side the {@link Side} from which to chop off an n-gram
86 * @param minGram the smallest n-gram to generate
87 * @param maxGram the largest n-gram to generate
89 public EdgeNGramTokenizer(Reader input, Side side, int minGram, int maxGram) {
91 init(side, minGram, maxGram);
95 * Creates EdgeNGramTokenizer that can generate n-grams in the sizes of the given range
97 * @param source {@link AttributeSource} to use
98 * @param input {@link Reader} holding the input to be tokenized
99 * @param side the {@link Side} from which to chop off an n-gram
100 * @param minGram the smallest n-gram to generate
101 * @param maxGram the largest n-gram to generate
103 public EdgeNGramTokenizer(AttributeSource source, Reader input, Side side, int minGram, int maxGram) {
104 super(source, input);
105 init(side, minGram, maxGram);
109 * Creates EdgeNGramTokenizer that can generate n-grams in the sizes of the given range
111 * @param factory {@link org.apache.lucene.util.AttributeSource.AttributeFactory} to use
112 * @param input {@link Reader} holding the input to be tokenized
113 * @param side the {@link Side} from which to chop off an n-gram
114 * @param minGram the smallest n-gram to generate
115 * @param maxGram the largest n-gram to generate
117 public EdgeNGramTokenizer(AttributeFactory factory, Reader input, Side side, int minGram, int maxGram) {
118 super(factory, input);
119 init(side, minGram, maxGram);
123 * Creates EdgeNGramTokenizer that can generate n-grams in the sizes of the given range
125 * @param input {@link Reader} holding the input to be tokenized
126 * @param sideLabel the name of the {@link Side} from which to chop off an n-gram
127 * @param minGram the smallest n-gram to generate
128 * @param maxGram the largest n-gram to generate
130 public EdgeNGramTokenizer(Reader input, String sideLabel, int minGram, int maxGram) {
131 this(input, Side.getSide(sideLabel), minGram, maxGram);
135 * Creates EdgeNGramTokenizer that can generate n-grams in the sizes of the given range
137 * @param source {@link AttributeSource} to use
138 * @param input {@link Reader} holding the input to be tokenized
139 * @param sideLabel the name of the {@link Side} from which to chop off an n-gram
140 * @param minGram the smallest n-gram to generate
141 * @param maxGram the largest n-gram to generate
143 public EdgeNGramTokenizer(AttributeSource source, Reader input, String sideLabel, int minGram, int maxGram) {
144 this(source, input, Side.getSide(sideLabel), minGram, maxGram);
148 * Creates EdgeNGramTokenizer that can generate n-grams in the sizes of the given range
150 * @param factory {@link org.apache.lucene.util.AttributeSource.AttributeFactory} to use
151 * @param input {@link Reader} holding the input to be tokenized
152 * @param sideLabel the name of the {@link Side} from which to chop off an n-gram
153 * @param minGram the smallest n-gram to generate
154 * @param maxGram the largest n-gram to generate
156 public EdgeNGramTokenizer(AttributeFactory factory, Reader input, String sideLabel, int minGram, int maxGram) {
157 this(factory, input, Side.getSide(sideLabel), minGram, maxGram);
160 private void init(Side side, int minGram, int maxGram) {
162 throw new IllegalArgumentException("sideLabel must be either front or back");
166 throw new IllegalArgumentException("minGram must be greater than zero");
169 if (minGram > maxGram) {
170 throw new IllegalArgumentException("minGram must not be greater than maxGram");
173 this.minGram = minGram;
174 this.maxGram = maxGram;
178 /** Returns the next token in the stream, or null at EOS. */
180 public final boolean incrementToken() throws IOException {
182 // if we are just starting, read the whole input
185 char[] chars = new char[1024];
186 int charsRead = input.read(chars);
187 inStr = new String(chars, 0, charsRead).trim(); // remove any leading or trailing spaces
188 inLen = inStr.length();
192 // if the remaining input is too short, we can't generate any n-grams
193 if (gramSize > inLen) {
197 // if we have hit the end of our n-gram size range, quit
198 if (gramSize > maxGram) {
202 // grab gramSize chars from front or back
203 int start = side == Side.FRONT ? 0 : inLen - gramSize;
204 int end = start + gramSize;
205 termAtt.setEmpty().append(inStr, start, end);
206 offsetAtt.setOffset(correctOffset(start), correctOffset(end));
212 public final void end() {
214 final int finalOffset = inLen;
215 this.offsetAtt.setOffset(finalOffset, finalOffset);
219 public void reset(Reader input) throws IOException {
225 public void reset() throws IOException {