1 package org.apache.lucene.index;
3 import org.apache.lucene.util.StringHelper;
5 import java.io.IOException;
6 import java.util.ArrayList;
9 * Licensed under the Apache License, Version 2.0 (the "License");
10 * you may not use this file except in compliance with the License.
11 * You may obtain a copy of the License at
13 * http://www.apache.org/licenses/LICENSE-2.0
15 * Unless required by applicable law or agreed to in writing, software
16 * distributed under the License is distributed on an "AS IS" BASIS,
17 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
18 * See the License for the specific language governing permissions and
19 * limitations under the License.
25 * Transparent access to the vector space model,
26 * either via TermFreqVector or by resolving it from the inverted index.
28 * Resolving a term vector from a large index can be a time consuming process.
30 * Warning! This class is not thread safe!
32 public class TermVectorAccessor {
34 public TermVectorAccessor() {
38 * Instance reused to save garbage collector some time
40 private TermVectorMapperDecorator decoratedMapper = new TermVectorMapperDecorator();
44 * Visits the TermVectorMapper and populates it with terms available for a given document,
45 * either via a vector created at index time or by resolving them from the inverted index.
47 * @param indexReader Index source
48 * @param documentNumber Source document to access
49 * @param fieldName Field to resolve
50 * @param mapper Mapper to be mapped with data
53 public void accept(IndexReader indexReader, int documentNumber, String fieldName, TermVectorMapper mapper) throws IOException {
55 fieldName = StringHelper.intern(fieldName);
57 decoratedMapper.decorated = mapper;
58 decoratedMapper.termVectorStored = false;
60 indexReader.getTermFreqVector(documentNumber, fieldName, decoratedMapper);
62 if (!decoratedMapper.termVectorStored) {
63 mapper.setDocumentNumber(documentNumber);
64 build(indexReader, fieldName, mapper, documentNumber);
68 /** Instance reused to save garbage collector some time */
69 private List<String> tokens;
71 /** Instance reused to save garbage collector some time */
72 private List<int[]> positions;
74 /** Instance reused to save garbage collector some time */
75 private List<Integer> frequencies;
79 * Populates the mapper with terms available for the given field in a document
80 * by resolving the inverted index.
83 * @param field interned field name
85 * @param documentNumber
88 private void build(IndexReader indexReader, String field, TermVectorMapper mapper, int documentNumber) throws IOException {
91 tokens = new ArrayList<String>(500);
92 positions = new ArrayList<int[]>(500);
93 frequencies = new ArrayList<Integer>(500);
100 TermEnum termEnum = indexReader.terms(new Term(field, ""));
101 if (termEnum.term() != null) {
102 while (termEnum.term().field() == field) {
103 TermPositions termPositions = indexReader.termPositions(termEnum.term());
104 if (termPositions.skipTo(documentNumber)) {
106 frequencies.add(Integer.valueOf(termPositions.freq()));
107 tokens.add(termEnum.term().text());
110 if (!mapper.isIgnoringPositions()) {
111 int[] positions = new int[termPositions.freq()];
112 for (int i = 0; i < positions.length; i++) {
113 positions[i] = termPositions.nextPosition();
115 this.positions.add(positions);
120 termPositions.close();
121 if (!termEnum.next()) {
125 mapper.setDocumentNumber(documentNumber);
126 mapper.setExpectations(field, tokens.size(), false, !mapper.isIgnoringPositions());
127 for (int i = 0; i < tokens.size(); i++) {
128 mapper.map(tokens.get(i), frequencies.get(i).intValue(), (TermVectorOffsetInfo[]) null, positions.get(i));
137 private static class TermVectorMapperDecorator extends TermVectorMapper {
139 private TermVectorMapper decorated;
142 public boolean isIgnoringPositions() {
143 return decorated.isIgnoringPositions();
147 public boolean isIgnoringOffsets() {
148 return decorated.isIgnoringOffsets();
151 private boolean termVectorStored = false;
154 public void setExpectations(String field, int numTerms, boolean storeOffsets, boolean storePositions) {
155 decorated.setExpectations(field, numTerms, storeOffsets, storePositions);
156 termVectorStored = true;
160 public void map(String term, int frequency, TermVectorOffsetInfo[] offsets, int[] positions) {
161 decorated.map(term, frequency, offsets, positions);
165 public void setDocumentNumber(int documentNumber) {
166 decorated.setDocumentNumber(documentNumber);