1 /*
2 * $Header: /home/cvs/jakarta-commons/httpclient/src/java/org/apache/commons/httpclient/ChunkedInputStream.java,v 1.16 2003/05/08 17:33:51 olegk Exp $
3 * $Revision: 1.16 $
4 * $Date: 2003/05/08 17:33:51 $
5 *
6 * ====================================================================
7 *
8 * The Apache Software License, Version 1.1
9 *
10 * Copyright (c) 2002-2003 The Apache Software Foundation. All rights
11 * reserved.
12 *
13 * Redistribution and use in source and binary forms, with or without
14 * modification, are permitted provided that the following conditions
15 * are met:
16 *
17 * 1. Redistributions of source code must retain the above copyright
18 * notice, this list of conditions and the following disclaimer.
19 *
20 * 2. Redistributions in binary form must reproduce the above copyright
21 * notice, this list of conditions and the following disclaimer in
22 * the documentation and/or other materials provided with the
23 * distribution.
24 *
25 * 3. The end-user documentation included with the redistribution, if
26 * any, must include the following acknowlegement:
27 * "This product includes software developed by the
28 * Apache Software Foundation (http://www.apache.org/)."
29 * Alternately, this acknowlegement may appear in the software itself,
30 * if and wherever such third-party acknowlegements normally appear.
31 *
32 * 4. The names "The Jakarta Project", "Commons", and "Apache Software
33 * Foundation" must not be used to endorse or promote products derived
34 * from this software without prior written permission. For written
35 * permission, please contact apache@apache.org.
36 *
37 * 5. Products derived from this software may not be called "Apache"
38 * nor may "Apache" appear in their names without prior written
39 * permission of the Apache Group.
40 *
41 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
42 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
43 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
44 * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
45 * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
46 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
47 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
48 * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
49 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
50 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
51 * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
52 * SUCH DAMAGE.
53 * ====================================================================
54 *
55 * This software consists of voluntary contributions made by many
56 * individuals on behalf of the Apache Software Foundation. For more
57 * information on the Apache Software Foundation, please see
58 * <http://www.apache.org/>.
59 *
60 * [Additional notices, if required by prior licensing conditions]
61 *
62 */
63
64 package org.apache.commons.httpclient;
65
66 import java.io.ByteArrayOutputStream;
67 import java.io.IOException;
68 import java.io.InputStream;
69
70
71 /***
72 * <p>Transparently coalesces chunks of a HTTP stream that uses
73 * Transfer-Encoding chunked.</p>
74 *
75 * <p>Note that this class NEVER closes the underlying stream, even when close
76 * gets called. Instead, it will read until the "end" of its chunking on close,
77 * which allows for the seamless invocation of subsequent HTTP 1.1 calls, while
78 * not requiring the client to remember to read the entire contents of the
79 * response.</p>
80 *
81 * @see ResponseInputStream
82 *
83 * @author Ortwin Gl�ck
84 * @author Sean C. Sullivan
85 * @author Martin Elwin
86 * @author Eric Johnson
87 * @author <a href="mailto:mbowler@GargoyleSoftware.com">Mike Bowler</a>
88 * @author Michael Becke
89 * @author <a href="mailto:oleg@ural.ru">Oleg Kalnichevski</a>
90 *
91 * @since 2.0
92 *
93 */
94
95 public class ChunkedInputStream extends InputStream {
96 /*** The inputstream that we're wrapping */
97 private InputStream in;
98
99 /*** The chunk size */
100 private int chunkSize;
101
102 /*** The current position within the current chunk */
103 private int pos;
104
105 /*** True if we'are at the beginning of stream */
106 private boolean bof = true;
107
108 /*** True if we've reached the end of stream */
109 private boolean eof = false;
110
111 /*** True if this stream is closed */
112 private boolean closed = false;
113
114 /*** The method that this stream came from */
115 private HttpMethod method;
116
117 /***
118 *
119 *
120 * @param in must be non-null
121 * @param method must be non-null
122 *
123 * @throws IOException If an IO error occurs
124 */
125 public ChunkedInputStream(
126 final InputStream in, final HttpMethod method) throws IOException {
127
128 if (in == null) {
129 throw new IllegalArgumentException("InputStream parameter may not be null");
130 }
131 if (method == null) {
132 throw new IllegalArgumentException("HttpMethod parameter may not be null");
133 }
134 this.in = in;
135 this.method = method;
136 this.pos = 0;
137 }
138
139 /***
140 * <p> Returns all the data in a chunked stream in coalesced form. A chunk
141 * is followed by a CRLF. The method returns -1 as soon as a chunksize of 0
142 * is detected.</p>
143 *
144 * <p> Trailer headers are read automcatically at the end of the stream and
145 * can be obtained with the getResponseFooters() method.</p>
146 *
147 * @return -1 of the end of the stream has been reached or the next data
148 * byte
149 * @throws IOException If an IO problem occurs
150 *
151 * @see HttpMethod#getResponseFooters()
152 */
153 public int read() throws IOException {
154
155 if (closed) {
156 throw new IOException("Attempted read from closed stream.");
157 }
158 if (eof) {
159 return -1;
160 }
161 if (pos >= chunkSize) {
162 nextChunk();
163 if (eof) {
164 return -1;
165 }
166 }
167 pos++;
168 return in.read();
169 }
170
171 /***
172 * Read some bytes from the stream.
173 * @param b The byte array that will hold the contents from the stream.
174 * @param off The offset into the byte array at which bytes will start to be
175 * placed.
176 * @param len the maximum number of bytes that can be returned.
177 * @return The number of bytes returned or -1 if the end of stream has been
178 * reached.
179 * @see java.io.InputStream#read(byte[], int, int)
180 * @throws IOException if an IO problem occurs.
181 */
182 public int read (byte[] b, int off, int len) throws IOException {
183
184 if (closed) {
185 throw new IOException("Attempted read from closed stream.");
186 }
187
188 if (eof) {
189 return -1;
190 }
191 if (pos >= chunkSize) {
192 nextChunk();
193 if (eof) {
194 return -1;
195 }
196 }
197 len = Math.min(len, chunkSize - pos);
198 int count = in.read(b, off, len);
199 pos += count;
200 return count;
201 }
202
203 /***
204 * Read some bytes from the stream.
205 * @param b The byte array that will hold the contents from the stream.
206 * @return The number of bytes returned or -1 if the end of stream has been
207 * reached.
208 * @see java.io.InputStream#read(byte[])
209 * @throws IOException if an IO problem occurs.
210 */
211 public int read (byte[] b) throws IOException {
212 return read(b, 0, b.length);
213 }
214
215 /***
216 * Read the CRLF terminator.
217 * @throws IOException If an IO error occurs.
218 */
219 private void readCRLF() throws IOException {
220 int cr = in.read();
221 int lf = in.read();
222 if ((cr != '\r') || (lf != '\n')) {
223 throw new IOException(
224 "CRLF expected at end of chunk: " + cr + "/" + lf);
225 }
226 }
227
228
229 /***
230 * Read the next chunk.
231 * @throws IOException If an IO error occurs.
232 */
233 private void nextChunk() throws IOException {
234 if (!bof) {
235 readCRLF();
236 }
237 chunkSize = getChunkSizeFromInputStream(in);
238 bof = false;
239 pos = 0;
240 if (chunkSize == 0) {
241 eof = true;
242 parseTrailerHeaders();
243 }
244 }
245
246 /***
247 * Expects the stream to start with a chunksize in hex with optional
248 * comments after a semicolon. The line must end with a CRLF: "a3; some
249 * comment\r\n" Positions the stream at the start of the next line.
250 *
251 * @param in The new input stream.
252 * @param required <tt>true<tt/> if a valid chunk must be present,
253 * <tt>false<tt/> otherwise.
254 *
255 * @return the chunk size as integer
256 *
257 * @throws IOException when the chunk size could not be parsed
258 */
259 private static int getChunkSizeFromInputStream(final InputStream in)
260 throws IOException {
261
262 ByteArrayOutputStream baos = new ByteArrayOutputStream();
263 // States: 0=normal, 1=\r was scanned, 2=inside quoted string, -1=end
264 int state = 0;
265 while (state != -1) {
266 int b = in.read();
267 if (b == -1) {
268 throw new IOException("chunked stream ended unexpectedly");
269 }
270 switch (state) {
271 case 0:
272 switch (b) {
273 case '\r':
274 state = 1;
275 break;
276 case '\"':
277 state = 2;
278 /* fall through */
279 default:
280 baos.write(b);
281 }
282 break;
283
284 case 1:
285 if (b == '\n') {
286 state = -1;
287 } else {
288 // this was not CRLF
289 throw new IOException("Protocol violation: Unexpected"
290 + " single newline character in chunk size");
291 }
292 break;
293
294 case 2:
295 switch (b) {
296 case '//':
297 b = in.read();
298 baos.write(b);
299 break;
300 case '\"':
301 state = 0;
302 /* fall through */
303 default:
304 baos.write(b);
305 }
306 break;
307 default: throw new RuntimeException("assertion failed");
308 }
309 }
310
311 //parse data
312 String dataString = HttpConstants.getString(baos.toByteArray());
313 int separator = dataString.indexOf(';');
314 dataString = (separator > 0)
315 ? dataString.substring(0, separator).trim()
316 : dataString.trim();
317
318 int result;
319 try {
320 result = Integer.parseInt(dataString.trim(), 16);
321 } catch (NumberFormatException e) {
322 throw new IOException ("Bad chunk size: " + dataString);
323 }
324 return result;
325 }
326
327 /***
328 * Reads and stores the Trailer headers.
329 * @throws IOException If an IO problem occurs
330 */
331 private void parseTrailerHeaders() throws IOException {
332 Header[] footers = HttpParser.parseHeaders(in);
333
334 for (int i = 0; i < footers.length; i++) {
335 method.addResponseFooter(footers[i]);
336 }
337 }
338
339 /***
340 * Upon close, this reads the remainder of the chunked message,
341 * leaving the underlying socket at a position to start reading the
342 * next response without scanning.
343 * @throws IOException If an IO problem occurs.
344 */
345 public void close() throws IOException {
346 if (!closed) {
347 try {
348 if (!eof) {
349 exhaustInputStream(this);
350 }
351 } finally {
352 eof = true;
353 closed = true;
354 }
355 }
356 }
357
358 /***
359 * Exhaust an input stream, reading until EOF has been encountered.
360 *
361 * <p>Note that this function is intended as a non-public utility.
362 * This is a little weird, but it seemed silly to make a utility
363 * class for this one function, so instead it is just static and
364 * shared that way.</p>
365 *
366 * @param inStream The {@link InputStream} to exhaust.
367 * @throws IOException If an IO problem occurs
368 */
369 static void exhaustInputStream(InputStream inStream) throws IOException {
370 // read and discard the remainder of the message
371 byte buffer[] = new byte[1024];
372 while (inStream.read(buffer) >= 0) {
373 ;
374 }
375 }
376 }
This page was automatically generated by Maven