1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * http://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17 package org.apache.pdfbox.filter;
18
19 import java.io.ByteArrayOutputStream;
20 import java.io.IOException;
21 import java.io.InputStream;
22 import java.io.OutputStream;
23 import java.io.PushbackInputStream;
24 import java.io.StreamCorruptedException;
25
26 import org.apache.pdfbox.cos.COSDictionary;
27
28 import org.apache.pdfbox.io.NBitInputStream;
29 import org.apache.pdfbox.io.NBitOutputStream;
30
31 /**
32 * This is the used for the LZWDecode filter.
33 *
34 * @author <a href="mailto:ben@benlitchfield.com">Ben Litchfield</a>
35 * @version $Revision: 1.15 $
36 */
37 public class LZWFilter implements Filter
38 {
39
40 /**
41 * The LZW clear table code.
42 */
43 public static final long CLEAR_TABLE = 256;
44 /**
45 * The LZW end of data code.
46 */
47 public static final long EOD = 257;
48
49 /**
50 * {@inheritDoc}
51 */
52 public void decode( InputStream compressedData, OutputStream result, COSDictionary options, int filterIndex )
53 throws IOException
54 {
55 //log.debug("decode( )");
56 NBitInputStream in = null;
57 in = new NBitInputStream( compressedData );
58 in.setBitsInChunk( 9 );
59 LZWDictionary dic = new LZWDictionary();
60 byte firstByte = 0;
61 long nextCommand = 0;
62 while( (nextCommand = in.read() ) != EOD )
63 {
64 // log.debug( "decode - nextCommand=" + nextCommand + ", bitsInChunk: " + in.getBitsInChunk());
65
66 if( nextCommand == CLEAR_TABLE )
67 {
68 in.setBitsInChunk( 9 );
69 dic = new LZWDictionary();
70 }
71 else
72 {
73 byte[] data = dic.getData( nextCommand );
74 if( data == null )
75 {
76 dic.visit( firstByte );
77 data = dic.getData( nextCommand );
78 dic.clear();
79 }
80 if( data == null )
81 {
82 throw new StreamCorruptedException( "Error: data is null" );
83 }
84 dic.visit(data);
85
86 //log.debug( "decode - dic.getNextCode(): " + dic.getNextCode());
87
88 if( dic.getNextCode() >= 2047 )
89 {
90 in.setBitsInChunk( 12 );
91 }
92 else if( dic.getNextCode() >= 1023 )
93 {
94 in.setBitsInChunk( 11 );
95 }
96 else if( dic.getNextCode() >= 511 )
97 {
98 in.setBitsInChunk( 10 );
99 }
100 else
101 {
102 in.setBitsInChunk( 9 );
103 }
104 /**
105 if( in.getBitsInChunk() != dic.getCodeSize() )
106 {
107 in.unread( nextCommand );
108 in.setBitsInChunk( dic.getCodeSize() );
109 System.out.print( "Switching " + nextCommand + " to " );
110 nextCommand = in.read();
111 System.out.println( "" + nextCommand );
112 data = dic.getData( nextCommand );
113 }**/
114 firstByte = data[0];
115 result.write( data );
116 }
117 }
118 result.flush();
119 }
120
121
122 /**
123 * {@inheritDoc}
124 */
125 public void encode( InputStream rawData, OutputStream result, COSDictionary options, int filterIndex )
126 throws IOException
127 {
128 //log.debug("encode( )");
129 PushbackInputStream input = new PushbackInputStream( rawData, 4096 );
130 LZWDictionary dic = new LZWDictionary();
131 NBitOutputStream out = new NBitOutputStream( result );
132 out.setBitsInChunk( 9 ); //initially nine
133 out.write( CLEAR_TABLE );
134 ByteArrayOutputStream buffer = new ByteArrayOutputStream();
135 int byteRead = 0;
136 for( int i=0; (byteRead = input.read()) != -1; i++ )
137 {
138 //log.debug( "byteRead = '" + (char)byteRead + "' (0x" + Integer.toHexString(byteRead) + "), i=" + i);
139 buffer.write( byteRead );
140 dic.visit( (byte)byteRead );
141 out.setBitsInChunk( dic.getCodeSize() );
142
143 //log.debug( "Getting node '" + new String( buffer.toByteArray() ) + "', buffer.size = " + buffer.size() );
144 LZWNode node = dic.getNode( buffer.toByteArray() );
145 int nextByte = input.read();
146 if( nextByte != -1 )
147 {
148 //log.debug( "nextByte = '" + (char)nextByte + "' (0x" + Integer.toHexString(nextByte) + ")");
149 LZWNode next = node.getNode( (byte)nextByte );
150 if( next == null )
151 {
152 //log.debug("encode - No next node, writing node and resetting buffer (" +
153 // " node.getCode: " + node.getCode() + ")" +
154 // " bitsInChunk: " + out.getBitsInChunk() +
155 // ")");
156 out.write( node.getCode() );
157 buffer.reset();
158 }
159
160 input.unread( nextByte );
161 }
162 else
163 {
164 //log.debug("encode - EOF on lookahead: writing node, resetting buffer, and terminating read loop (" +
165 // " node.getCode: " + node.getCode() + ")" +
166 // " bitsInChunk: " + out.getBitsInChunk() +
167 // ")");
168 out.write( node.getCode() );
169 buffer.reset();
170 break;
171 }
172
173 if( dic.getNextCode() == 4096 )
174 {
175 //log.debug("encode - Clearing dictionary and unreading pending buffer data (" +
176 // " bitsInChunk: " + out.getBitsInChunk() +
177 // ")");
178 out.write( CLEAR_TABLE );
179 dic = new LZWDictionary();
180 input.unread( buffer.toByteArray() );
181 buffer.reset();
182 }
183 }
184
185 // Fix the code size based on the fact that we are writing the EOD
186 //
187 if( dic.getNextCode() >= 2047 )
188 {
189 out.setBitsInChunk( 12 );
190 }
191 else if( dic.getNextCode() >= 1023 )
192 {
193 out.setBitsInChunk( 11 );
194 }
195 else if( dic.getNextCode() >= 511 )
196 {
197 out.setBitsInChunk( 10 );
198 }
199 else
200 {
201 out.setBitsInChunk( 9 );
202 }
203
204 //log.debug("encode - Writing EOD (" +
205 // " bitsInChunk: " + out.getBitsInChunk() +
206 // ")");
207 out.write( EOD );
208 out.close();
209 result.flush();
210 }
211 }