KickJava   Java API By Example, From Geeks To Geeks.

Java > Open Source Codes > org > apache > lucene > index > FieldsReader


1 package org.apache.lucene.index;
2
3 /**
4  * Copyright 2004 The Apache Software Foundation
5  *
6  * Licensed under the Apache License, Version 2.0 (the "License");
7  * you may not use this file except in compliance with the License.
8  * You may obtain a copy of the License at
9  *
10  * http://www.apache.org/licenses/LICENSE-2.0
11  *
12  * Unless required by applicable law or agreed to in writing, software
13  * distributed under the License is distributed on an "AS IS" BASIS,
14  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15  * See the License for the specific language governing permissions and
16  * limitations under the License.
17  */
18
19 import java.io.ByteArrayOutputStream JavaDoc;
20 import java.io.IOException JavaDoc;
21 import java.util.zip.DataFormatException JavaDoc;
22 import java.util.zip.Inflater JavaDoc;
23
24 import org.apache.lucene.document.Document;
25 import org.apache.lucene.document.Field;
26 import org.apache.lucene.store.Directory;
27 import org.apache.lucene.store.IndexInput;
28
29 /**
30  * Class responsible for access to stored document fields.
31  *
32  * It uses <segment>.fdt and <segment>.fdx; files.
33  *
34  * @version $Id: FieldsReader.java 329524 2005-10-30 05:38:46Z yonik $
35  */
36 final class FieldsReader {
37   private FieldInfos fieldInfos;
38   private IndexInput fieldsStream;
39   private IndexInput indexStream;
40   private int size;
41
42   FieldsReader(Directory d, String JavaDoc segment, FieldInfos fn) throws IOException JavaDoc {
43     fieldInfos = fn;
44
45     fieldsStream = d.openInput(segment + ".fdt");
46     indexStream = d.openInput(segment + ".fdx");
47
48     size = (int)(indexStream.length() / 8);
49   }
50
51   final void close() throws IOException JavaDoc {
52     fieldsStream.close();
53     indexStream.close();
54   }
55
56   final int size() {
57     return size;
58   }
59
60   final Document doc(int n) throws IOException JavaDoc {
61     indexStream.seek(n * 8L);
62     long position = indexStream.readLong();
63     fieldsStream.seek(position);
64
65     Document doc = new Document();
66     int numFields = fieldsStream.readVInt();
67     for (int i = 0; i < numFields; i++) {
68       int fieldNumber = fieldsStream.readVInt();
69       FieldInfo fi = fieldInfos.fieldInfo(fieldNumber);
70
71       byte bits = fieldsStream.readByte();
72       
73       boolean compressed = (bits & FieldsWriter.FIELD_IS_COMPRESSED) != 0;
74       boolean tokenize = (bits & FieldsWriter.FIELD_IS_TOKENIZED) != 0;
75       
76       if ((bits & FieldsWriter.FIELD_IS_BINARY) != 0) {
77         final byte[] b = new byte[fieldsStream.readVInt()];
78         fieldsStream.readBytes(b, 0, b.length);
79         if (compressed)
80           doc.add(new Field(fi.name, uncompress(b), Field.Store.COMPRESS));
81         else
82           doc.add(new Field(fi.name, b, Field.Store.YES));
83       }
84       else {
85         Field.Index index;
86         Field.Store store = Field.Store.YES;
87         
88         if (fi.isIndexed && tokenize)
89           index = Field.Index.TOKENIZED;
90         else if (fi.isIndexed && !tokenize)
91           index = Field.Index.UN_TOKENIZED;
92         else
93           index = Field.Index.NO;
94         
95         Field.TermVector termVector = null;
96         if (fi.storeTermVector) {
97           if (fi.storeOffsetWithTermVector) {
98             if (fi.storePositionWithTermVector) {
99               termVector = Field.TermVector.WITH_POSITIONS_OFFSETS;
100             }
101             else {
102               termVector = Field.TermVector.WITH_OFFSETS;
103             }
104           }
105           else if (fi.storePositionWithTermVector) {
106             termVector = Field.TermVector.WITH_POSITIONS;
107           }
108           else {
109             termVector = Field.TermVector.YES;
110           }
111         }
112         else {
113           termVector = Field.TermVector.NO;
114         }
115         
116         if (compressed) {
117           store = Field.Store.COMPRESS;
118           final byte[] b = new byte[fieldsStream.readVInt()];
119           fieldsStream.readBytes(b, 0, b.length);
120           Field f = new Field(fi.name, // field name
121 new String JavaDoc(uncompress(b), "UTF-8"), // uncompress the value and add as string
122 store,
123               index,
124               termVector);
125           f.setOmitNorms(fi.omitNorms);
126           doc.add(f);
127         }
128         else {
129           Field f = new Field(fi.name, // name
130 fieldsStream.readString(), // read value
131 store,
132                 index,
133                 termVector);
134           f.setOmitNorms(fi.omitNorms);
135           doc.add(f);
136         }
137       }
138     }
139
140     return doc;
141   }
142   
143   private final byte[] uncompress(final byte[] input)
144     throws IOException JavaDoc
145   {
146   
147     Inflater JavaDoc decompressor = new Inflater JavaDoc();
148     decompressor.setInput(input);
149   
150     // Create an expandable byte array to hold the decompressed data
151 ByteArrayOutputStream JavaDoc bos = new ByteArrayOutputStream JavaDoc(input.length);
152   
153     // Decompress the data
154 byte[] buf = new byte[1024];
155     while (!decompressor.finished()) {
156       try {
157         int count = decompressor.inflate(buf);
158         bos.write(buf, 0, count);
159       }
160       catch (DataFormatException JavaDoc e) {
161         // this will happen if the field is not compressed
162 throw new IOException JavaDoc ("field data are in wrong format: " + e.toString());
163       }
164     }
165   
166     decompressor.end();
167     
168     // Get the decompressed data
169 return bos.toByteArray();
170   }
171 }
172
Popular Tags