ChunkyByteArray


1   /*
2    * The Apache Software License, Version 1.1
3    *
4    *
5    * Copyright (c) 1999 The Apache Software Foundation.  All rights 
6    * reserved.
7    *
8    * Redistribution and use in source and binary forms, with or without
9    * modification, are permitted provided that the following conditions
10   * are met:
11   *
12   * 1. Redistributions of source code must retain the above copyright
13   *    notice, this list of conditions and the following disclaimer. 
14   *
15   * 2. Redistributions in binary form must reproduce the above copyright
16   *    notice, this list of conditions and the following disclaimer in
17   *    the documentation and/or other materials provided with the
18   *    distribution.
19   *
20   * 3. The end-user documentation included with the redistribution,
21   *    if any, must include the following acknowledgment:  
22   *       "This product includes software developed by the
23   *        Apache Software Foundation (http://www.apache.org/)."
24   *    Alternately, this acknowledgment may appear in the software itself,
25   *    if and wherever such third-party acknowledgments normally appear.
26   *
27   * 4. The names "Xerces" and "Apache Software Foundation" must
28   *    not be used to endorse or promote products derived from this
29   *    software without prior written permission. For written 
30   *    permission, please contact apache@apache.org.
31   *
32   * 5. Products derived from this software may not be called "Apache",
33   *    nor may "Apache" appear in their name, without prior written
34   *    permission of the Apache Software Foundation.
35   *
36   * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
37   * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
38   * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
39   * DISCLAIMED.  IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
40   * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
41   * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
42   * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
43   * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
44   * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
45   * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
46   * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
47   * SUCH DAMAGE.
48   * ====================================================================
49   *
50   * This software consists of voluntary contributions made by many
51   * individuals on behalf of the Apache Software Foundation and was
52   * originally based on software copyright (c) 1999, International
53   * Business Machines, Inc., http://www.apache.org.  For more
54   * information on the Apache Software Foundation, please see
55   * <http://www.apache.org/>.
56   */
57  
58  package org.enhydra.apache.xerces.utils;
59  
60  import java.io.IOException  ;
61  import java.io.InputStream  ;
62  
63  /**
64   * This class is used for accessing the data provided by an InputStream.
65   *
66   * There are two ways in which this class is used.  The first occurs
67   * when we are prescanning the start of the stream to determine the
68   * encoding being used.  Since we do not require that the stream be
69   * positionable, we wrap it with an instance of this class.  The first
70   * "chunk" of the file is read and the data may be accessed directly
71   * using the byteAt(offset) method.  After we have determined the
72   * encoding of the byte stream, the instance of this class is passed
73   * on to the EntityReader that will process the data for the scanner.
74   *
75   * At this point, the reader may continue to access this instance using
76   * the byteAt method, which will continue to read the contents into
77   * chunks as required until end of input.  An example of this is the
78   * UCSReader.
79   *
80   * Alternatively, the reader may access this instance as an InputStream
81   * which will first return any data that has been reading into the
82   * chunks, and will then return the remaining data from the original
83   * InputStream directly.
84   *
85   * @version
86   */
87  public final class ChunkyByteArray extends InputStream   {
88  
89      /**
90       * Constructor
91       *
92       * Reads the first chunk.
93       *
94       * @param is The input stream containing the data of the entity.
95       */
96      public ChunkyByteArray(InputStream   is) throws IOException   {
97          fInputStream = is;
98          fill();
99      }
100 
101     /**
102      * Read a byte.
103      *
104      * @return The next byte of the input data or -1 if there is no more data.
105      */
106     public int read() throws IOException   {
107         if (fData == null)
108             return fInputStream == null ? -1 : fInputStream.read();
109         int b = (int)(fData[0][fOffset]);
110         if (++fOffset == fLength) {
111             fData = null;
112             if (fLength < CHUNK_SIZE)
113                 fInputStream = null;
114         }
115         return b;
116     }
117 
118     /**
119      * Read bytes.
120      *
121      * @param buffer The destination for the bytes returned.  If null, then
122      *               the data will discarded instead of returned.
123      * @param offset The offset within the buffer where the first returned
124      *               byte should be placed.
125      * @param length The maximum number of bytes to place in the buffer or discard.
126      * @return The number of bytes actually placed in the buffer or discarded.
127      */
128     public int read(byte buffer[], int offset, int length) throws IOException   {
129         int bytesLeft = fLength - fOffset;
130         if (bytesLeft == 0)
131             return fInputStream == null ? -1 : fInputStream.read(buffer, offset, length);
132         if (length <= 0)
133             return 0;
134         byte[] chunk = fData[0];
135         if (length >= bytesLeft) {
136             length = bytesLeft;
137             if (fLength < CHUNK_SIZE)
138                 fInputStream = null;
139         }
140         if (buffer == null) {
141             fOffset += length;
142             return length;
143         }
144         int stop = offset + length;
145         do {
146             buffer[offset++] = chunk[fOffset++];
147         } while (offset < stop);
148         return length;
149     }
150 
151     /**
152      * Reset position within the data stream back to
153      * the very beginning.
154      */
155     public void rewind() {
156         fOffset = 0;
157     }
158 
159     /**
160      * Return a byte of input data at the given offset.
161      *
162      * @param offset The offset in the data stream.
163      * @return The byte at the specified position within the data stream.
164      */
165     public byte byteAt(int offset) throws IOException   {
166         int chunk = offset >> CHUNK_SHIFT;
167         int index = offset & CHUNK_MASK;
168         try {
169             return fData[chunk][index];
170         } catch (NullPointerException   ex) {
171             // ignore -- let fill create new chunk
172         } catch (ArrayIndexOutOfBoundsException   e) {
173             // current chunk array is not big enough; resize
174             byte newdata[][] = new byte[fData.length * 2][];
175             System.arraycopy(fData, 0, newdata, 0, fData.length);
176             fData = newdata;
177         }
178         if (index == 0) {
179             fill();
180             return fData[chunk][index];
181         }
182         return 0;
183     }
184 
185     /**
186      * Test to see if an offset is at the end of the input data.
187      *
188      * @param offset A position in the data stream.
189      * @return <code>true</code> if the position is at the end of the data stream;
190      *         <code>false</code> otherwise.
191      */
192     public boolean atEOF(int offset) {
193         return(offset > fLength);
194     }
195 
196 
197 
198     /**
199      * Closes this input Stream
200      * 
201      * @exception IOException
202      */
203     public void close() throws IOException   {
204         if ( fInputStream != null ) {
205              fInputStream.close(); 
206              fInputStream = null; // Null it 
207         }
208     }
209 
210 
211     //
212     // Fill in the next chunk with additional data.
213     //
214     private void fill() throws IOException   {
215         int bufnum = fLength >> CHUNK_SHIFT;
216         byte[] data = new byte[CHUNK_SIZE];
217         fData[bufnum] = data;
218         int offset = 0;
219         int capacity = CHUNK_SIZE;
220         int result = 0;
221         do {
222             result = fInputStream.read(data, offset, capacity);
223             if (result == -1) {
224                 data[offset] = (byte)0xff;
225                 fInputStream.close();
226                 fInputStream = null;
227                 break;
228             }
229             if (result > 0) {
230                 fLength += result;
231                 offset += result;
232                 capacity -= result;
233             }
234         } while (capacity > 0);
235     }
236     //
237     // Chunk size constants
238     //
239     private static final int CHUNK_SHIFT = 14;           // 2^14 = 16k
240     private static final int CHUNK_SIZE = (1 << CHUNK_SHIFT);
241     private static final int CHUNK_MASK = CHUNK_SIZE - 1;
242     private static final int INITIAL_CHUNK_COUNT = (1 << (20 - CHUNK_SHIFT));   // 2^20 = 1m
243     //
244     // Instance variables
245     //
246     private InputStream   fInputStream = null;
247     private byte[][] fData = new byte[INITIAL_CHUNK_COUNT][];
248     private int fLength = 0;
249     private int fOffset = 0; // for read methods
250 }
251
A to Z: JavaDoc & Examples Daily Java News & Articles Open Source Projects Open Source Codes Free Computer Books Remove Frame
Popular Tags