1 |
/* DataInputStream.java -- FilteredInputStream that implements DataInput |
2 |
Copyright (C) 1998, 1999, 2000, 2001, 2003 Free Software Foundation |
3 |
|
4 |
This file is part of GNU Classpath. |
5 |
|
6 |
GNU Classpath is free software; you can redistribute it and/or modify |
7 |
it under the terms of the GNU General Public License as published by |
8 |
the Free Software Foundation; either version 2, or (at your option) |
9 |
any later version. |
10 |
|
11 |
GNU Classpath is distributed in the hope that it will be useful, but |
12 |
WITHOUT ANY WARRANTY; without even the implied warranty of |
13 |
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
14 |
General Public License for more details. |
15 |
|
16 |
You should have received a copy of the GNU General Public License |
17 |
along with GNU Classpath; see the file COPYING. If not, write to the |
18 |
Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA |
19 |
02111-1307 USA. |
20 |
|
21 |
Linking this library statically or dynamically with other modules is |
22 |
making a combined work based on this library. Thus, the terms and |
23 |
conditions of the GNU General Public License cover the whole |
24 |
combination. |
25 |
|
26 |
As a special exception, the copyright holders of this library give you |
27 |
permission to link this library with independent modules to produce an |
28 |
executable, regardless of the license terms of these independent |
29 |
modules, and to copy and distribute the resulting executable under |
30 |
terms of your choice, provided that you also meet, for each linked |
31 |
independent module, the terms and conditions of the license of that |
32 |
module. An independent module is a module which is not derived from |
33 |
or based on this library. If you modify this library, you may extend |
34 |
this exception to your version of the library, but you are not |
35 |
obligated to do so. If you do not wish to do so, delete this |
36 |
exception statement from your version. */ |
37 |
|
38 |
package java.io; |
39 |
|
40 |
/* Written using "Java Class Libraries", 2nd edition, ISBN 0-201-31002-3 |
41 |
* "The Java Language Specification", ISBN 0-201-63451-1 |
42 |
* plus online API docs for JDK 1.2 beta from http://www.javasoft.com. |
43 |
* Status: Believed complete and correct. |
44 |
*/ |
45 |
|
46 |
/** |
47 |
* This subclass of <code>FilteredInputStream</code> implements the |
48 |
* <code>DataInput</code> interface that provides method for reading primitive |
49 |
* Java data types from a stream. |
50 |
* |
51 |
* @see DataInput |
52 |
* |
53 |
* @author Warren Levy <warrenl@cygnus.com> |
54 |
* @author Aaron M. Renn <arenn@urbanophile.com> |
55 |
* @date October 20, 1998. |
56 |
*/ |
57 |
public class DataInputStream extends FilterInputStream implements DataInput |
58 |
{ |
59 |
// readLine() hack to ensure that an '\r' not followed by an '\n' is |
60 |
// handled correctly. If set, readLine() will ignore the first char it sees |
61 |
// if that char is a '\n' |
62 |
boolean ignoreInitialNewline = false; |
63 |
|
64 |
// Byte buffer, used to make primitive read calls more efficient. |
65 |
byte[] buf = new byte [8]; |
66 |
|
67 |
/** |
68 |
* This constructor initializes a new <code>DataInputStream</code> |
69 |
* to read from the specified subordinate stream. |
70 |
* |
71 |
* @param in The subordinate <code>InputStream</code> to read from |
72 |
*/ |
73 |
public DataInputStream (InputStream in) |
74 |
{ |
75 |
super (in); |
76 |
} |
77 |
|
78 |
/** |
79 |
* This method reads bytes from the underlying stream into the specified |
80 |
* byte array buffer. It will attempt to fill the buffer completely, but |
81 |
* may return a short count if there is insufficient data remaining to be |
82 |
* read to fill the buffer. |
83 |
* |
84 |
* @param b The buffer into which bytes will be read. |
85 |
* |
86 |
* @return The actual number of bytes read, or -1 if end of stream reached |
87 |
* before reading any bytes. |
88 |
* |
89 |
* @exception IOException If an error occurs. |
90 |
*/ |
91 |
public final int read (byte[] b) throws IOException |
92 |
{ |
93 |
return in.read (b, 0, b.length); |
94 |
} |
95 |
|
96 |
/** |
97 |
* This method reads bytes from the underlying stream into the specified |
98 |
* byte array buffer. It will attempt to read <code>len</code> bytes and |
99 |
* will start storing them at position <code>off</code> into the buffer. |
100 |
* This method can return a short count if there is insufficient data |
101 |
* remaining to be read to complete the desired read length. |
102 |
* |
103 |
* @param b The buffer into which bytes will be read. |
104 |
* @param off The offset into the buffer to start storing bytes. |
105 |
* @param len The requested number of bytes to read. |
106 |
* |
107 |
* @return The actual number of bytes read, or -1 if end of stream reached |
108 |
* before reading any bytes. |
109 |
* |
110 |
* @exception IOException If an error occurs. |
111 |
*/ |
112 |
public final int read (byte[] b, int off, int len) throws IOException |
113 |
{ |
114 |
return in.read (b, off, len); |
115 |
} |
116 |
|
117 |
/** |
118 |
* This method reads a Java boolean value from an input stream. It does |
119 |
* so by reading a single byte of data. If that byte is zero, then the |
120 |
* value returned is <code>false</code>. If the byte is non-zero, then |
121 |
* the value returned is <code>true</code>. |
122 |
* <p> |
123 |
* This method can read a <code>boolean</code> written by an object |
124 |
* implementing the <code>writeBoolean()</code> method in the |
125 |
* <code>DataOutput</code> interface. |
126 |
* |
127 |
* @return The <code>boolean</code> value read |
128 |
* |
129 |
* @exception EOFException If end of file is reached before reading |
130 |
* the boolean |
131 |
* @exception IOException If any other error occurs |
132 |
* |
133 |
* @see DataOutput#writeBoolean |
134 |
*/ |
135 |
public final boolean readBoolean () throws IOException |
136 |
{ |
137 |
return convertToBoolean (in.read ()); |
138 |
} |
139 |
|
140 |
/** |
141 |
* This method reads a Java byte value from an input stream. The value |
142 |
* is in the range of -128 to 127. |
143 |
* <p> |
144 |
* This method can read a <code>byte</code> written by an object |
145 |
* implementing the <code>writeByte()</code> method in the |
146 |
* <code>DataOutput</code> interface. |
147 |
* |
148 |
* @return The <code>byte</code> value read |
149 |
* |
150 |
* @exception EOFException If end of file is reached before reading the byte |
151 |
* @exception IOException If any other error occurs |
152 |
* |
153 |
* @see DataOutput#writeByte |
154 |
*/ |
155 |
public final byte readByte () throws IOException |
156 |
{ |
157 |
return convertToByte (in.read ()); |
158 |
} |
159 |
|
160 |
/** |
161 |
* This method reads a Java <code>char</code> value from an input stream. |
162 |
* It operates by reading two bytes from the stream and converting them to |
163 |
* a single 16-bit Java <code>char</code>. The two bytes are stored most |
164 |
* significant byte first (i.e., "big endian") regardless of the native |
165 |
* host byte ordering. |
166 |
* <p> |
167 |
* As an example, if <code>byte1</code> and <code>byte2</code> |
168 |
* represent the first and second byte read from the stream |
169 |
* respectively, they will be transformed to a <code>char</code> in |
170 |
* the following manner: |
171 |
* <p> |
172 |
* <code>(char)(((byte1 & 0xFF) << 8) | (byte2 & 0xFF)</code> |
173 |
* <p> |
174 |
* This method can read a <code>char</code> written by an object |
175 |
* implementing the <code>writeChar()</code> method in the |
176 |
* <code>DataOutput</code> interface. |
177 |
* |
178 |
* @return The <code>char</code> value read |
179 |
* |
180 |
* @exception EOFException If end of file is reached before reading the char |
181 |
* @exception IOException If any other error occurs |
182 |
* |
183 |
* @see DataOutput#writeChar |
184 |
*/ |
185 |
public final char readChar () throws IOException |
186 |
{ |
187 |
readFully (buf, 0, 2); |
188 |
return convertToChar (buf); |
189 |
} |
190 |
|
191 |
/** |
192 |
* This method reads a Java double value from an input stream. It operates |
193 |
* by first reading a <code>long</code> value from the stream by calling the |
194 |
* <code>readLong()</code> method in this interface, then converts |
195 |
* that <code>long</code> to a <code>double</code> using the |
196 |
* <code>longBitsToDouble</code> method in the class |
197 |
* <code>java.lang.Double</code> |
198 |
* <p> |
199 |
* This method can read a <code>double</code> written by an object |
200 |
* implementing the <code>writeDouble()</code> method in the |
201 |
* <code>DataOutput</code> interface. |
202 |
* |
203 |
* @return The <code>double</code> value read |
204 |
* |
205 |
* @exception EOFException If end of file is reached before reading |
206 |
* the double |
207 |
* @exception IOException If any other error occurs |
208 |
* |
209 |
* @see DataOutput#writeDouble |
210 |
* @see java.lang.Double#longBitsToDouble |
211 |
*/ |
212 |
public final double readDouble () throws IOException |
213 |
{ |
214 |
return Double.longBitsToDouble (readLong ()); |
215 |
} |
216 |
|
217 |
/** |
218 |
* This method reads a Java float value from an input stream. It |
219 |
* operates by first reading an <code>int</code> value from the |
220 |
* stream by calling the <code>readInt()</code> method in this |
221 |
* interface, then converts that <code>int</code> to a |
222 |
* <code>float</code> using the <code>intBitsToFloat</code> method |
223 |
* in the class <code>java.lang.Float</code> |
224 |
* <p> |
225 |
* This method can read a <code>float</code> written by an object |
226 |
* implementing the <code>writeFloat()</code> method in the |
227 |
* <code>DataOutput</code> interface. |
228 |
* |
229 |
* @return The <code>float</code> value read |
230 |
* |
231 |
* @exception EOFException If end of file is reached before reading the float |
232 |
* @exception IOException If any other error occurs |
233 |
* |
234 |
* @see DataOutput#writeFloat |
235 |
* @see java.lang.Float#intBitsToFloat |
236 |
*/ |
237 |
public final float readFloat () throws IOException |
238 |
{ |
239 |
return Float.intBitsToFloat (readInt ()); |
240 |
} |
241 |
|
242 |
/** |
243 |
* This method reads raw bytes into the passed array until the array is |
244 |
* full. Note that this method blocks until the data is available and |
245 |
* throws an exception if there is not enough data left in the stream to |
246 |
* fill the buffer. Note also that zero length buffers are permitted. |
247 |
* In this case, the method will return immediately without reading any |
248 |
* bytes from the stream. |
249 |
* |
250 |
* @param b The buffer into which to read the data |
251 |
* |
252 |
* @exception EOFException If end of file is reached before filling the |
253 |
* buffer |
254 |
* @exception IOException If any other error occurs |
255 |
*/ |
256 |
public final void readFully (byte[] b) throws IOException |
257 |
{ |
258 |
readFully (b, 0, b.length); |
259 |
} |
260 |
|
261 |
/** |
262 |
* This method reads raw bytes into the passed array <code>buf</code> |
263 |
* starting |
264 |
* <code>offset</code> bytes into the buffer. The number of bytes read |
265 |
* will be |
266 |
* exactly <code>len</code>. Note that this method blocks until the data is |
267 |
* available and throws an exception if there is not enough data left in |
268 |
* the stream to read <code>len</code> bytes. Note also that zero length |
269 |
* buffers are permitted. In this case, the method will return immediately |
270 |
* without reading any bytes from the stream. |
271 |
* |
272 |
* @param buf The buffer into which to read the data |
273 |
* @param offset The offset into the buffer to start storing data |
274 |
* @param len The number of bytes to read into the buffer |
275 |
* |
276 |
* @exception EOFException If end of file is reached before filling the |
277 |
* buffer |
278 |
* @exception IOException If any other error occurs |
279 |
*/ |
280 |
public final void readFully (byte[] b, int off, int len) throws IOException |
281 |
{ |
282 |
while (len > 0) |
283 |
{ |
284 |
// in.read will block until some data is available. |
285 |
int numread = in.read (b, off, len); |
286 |
if (numread < 0) |
287 |
throw new EOFException (); |
288 |
len -= numread; |
289 |
off += numread; |
290 |
} |
291 |
} |
292 |
|
293 |
/** |
294 |
* This method reads a Java <code>int</code> value from an input stream |
295 |
* It operates by reading four bytes from the stream and converting them to |
296 |
* a single Java <code>int</code>. The bytes are stored most |
297 |
* significant byte first (i.e., "big endian") regardless of the native |
298 |
* host byte ordering. |
299 |
* <p> |
300 |
* As an example, if <code>byte1</code> through <code>byte4</code> represent |
301 |
* the first four bytes read from the stream, they will be |
302 |
* transformed to an <code>int</code> in the following manner: |
303 |
* <p> |
304 |
* <code>(int)(((byte1 & 0xFF) << 24) + ((byte2 & 0xFF) << 16) + |
305 |
* ((byte3 & 0xFF)<< 8) + (byte4 & 0xFF)))</code> |
306 |
* <p> |
307 |
* The value returned is in the range of -2147483648 to 2147483647. |
308 |
* <p> |
309 |
* This method can read an <code>int</code> written by an object |
310 |
* implementing the <code>writeInt()</code> method in the |
311 |
* <code>DataOutput</code> interface. |
312 |
* |
313 |
* @return The <code>int</code> value read |
314 |
* |
315 |
* @exception EOFException If end of file is reached before reading the int |
316 |
* @exception IOException If any other error occurs |
317 |
* |
318 |
* @see DataOutput#writeInt |
319 |
*/ |
320 |
public final int readInt () throws IOException |
321 |
{ |
322 |
readFully (buf, 0, 4); |
323 |
return convertToInt (buf); |
324 |
} |
325 |
|
326 |
/** |
327 |
* This method reads the next line of text data from an input |
328 |
* stream. It operates by reading bytes and converting those bytes |
329 |
* to <code>char</code> values by treating the byte read as the low |
330 |
* eight bits of the <code>char</code> and using 0 as the high eight |
331 |
* bits. Because of this, it does not support the full 16-bit |
332 |
* Unicode character set. |
333 |
* <p> |
334 |
* The reading of bytes ends when either the end of file or a line |
335 |
* terminator is encountered. The bytes read are then returned as a |
336 |
* <code>String</code> A line terminator is a byte sequence |
337 |
* consisting of either <code>\r</code>, <code>\n</code> or |
338 |
* <code>\r\n</code>. These termination charaters are discarded and |
339 |
* are not returned as part of the string. |
340 |
* <p> |
341 |
* This method can read data that was written by an object implementing the |
342 |
* <code>writeLine()</code> method in <code>DataOutput</code>. |
343 |
* |
344 |
* @return The line read as a <code>String</code> |
345 |
* |
346 |
* @exception IOException If an error occurs |
347 |
* |
348 |
* @see DataOutput |
349 |
* |
350 |
* @deprecated |
351 |
*/ |
352 |
public final String readLine () throws IOException |
353 |
{ |
354 |
StringBuffer strb = new StringBuffer (); |
355 |
|
356 |
readloop: while (true) |
357 |
{ |
358 |
int c = 0; |
359 |
char ch = ' '; |
360 |
boolean getnext = true; |
361 |
while (getnext) |
362 |
{ |
363 |
getnext = false; |
364 |
c = in.read(); |
365 |
if (c < 0) // got an EOF |
366 |
return strb.length () > 0 ? strb.toString () : null; |
367 |
ch = (char) c; |
368 |
if ((ch &= 0xFF) == '\n') |
369 |
// hack to correctly handle '\r\n' sequences |
370 |
if (ignoreInitialNewline) |
371 |
{ |
372 |
ignoreInitialNewline = false; |
373 |
getnext = true; |
374 |
} |
375 |
else |
376 |
break readloop; |
377 |
} |
378 |
|
379 |
if (ch == '\r') |
380 |
{ |
381 |
// FIXME: The following code tries to adjust the stream back one |
382 |
// character if the next char read is '\n'. As a last resort, |
383 |
// it tries to mark the position before reading but the bottom |
384 |
// line is that it is possible that this method will not properly |
385 |
// deal with a '\r' '\n' combination thus not fulfilling the |
386 |
// DataInput contract for readLine. It's not a particularly |
387 |
// safe approach threadwise since it is unsynchronized and |
388 |
// since it might mark an input stream behind the users back. |
389 |
// Along the same vein it could try the same thing for |
390 |
// ByteArrayInputStream and PushbackInputStream, but that is |
391 |
// probably overkill since this is deprecated & BufferedInputStream |
392 |
// is the most likely type of input stream. |
393 |
// |
394 |
// The alternative is to somehow push back the next byte if it |
395 |
// isn't a '\n' or to have the reading methods of this class |
396 |
// keep track of whether the last byte read was '\r' by readLine |
397 |
// and then skip the very next byte if it is '\n'. Either way, |
398 |
// this would increase the complexity of the non-deprecated methods |
399 |
// and since it is undesirable to make non-deprecated methods |
400 |
// less efficient, the following seems like the most reasonable |
401 |
// approach. |
402 |
int next_c = 0; |
403 |
char next_ch = ' '; |
404 |
if (in instanceof BufferedInputStream) |
405 |
{ |
406 |
next_c = in.read(); |
407 |
next_ch = (char) (next_c & 0xFF); |
408 |
if ((next_ch != '\n') && (next_c >= 0)) |
409 |
{ |
410 |
BufferedInputStream bin = (BufferedInputStream) in; |
411 |
if (bin.pos > 0) |
412 |
bin.pos--; |
413 |
} |
414 |
} |
415 |
else if (markSupported()) |
416 |
{ |
417 |
next_c = in.read(); |
418 |
next_ch = (char) (next_c & 0xFF); |
419 |
if ((next_ch != '\n') && (next_c >= 0)) |
420 |
{ |
421 |
mark(1); |
422 |
if ((in.read() & 0xFF) != '\n') |
423 |
reset(); |
424 |
} |
425 |
} |
426 |
// In order to catch cases where 'in' isn't a BufferedInputStream |
427 |
// and doesn't support mark() (such as reading from a Socket), set |
428 |
// a flag that instructs readLine() to ignore the first character |
429 |
// it sees _if_ that character is a '\n'. |
430 |
else ignoreInitialNewline = true; |
431 |
break; |
432 |
} |
433 |
strb.append(ch); |
434 |
} |
435 |
|
436 |
return strb.length() > 0 ? strb.toString() : ""; |
437 |
} |
438 |
|
439 |
/** |
440 |
* This method reads a Java <code>long</code> value from an input stream |
441 |
* It operates by reading eight bytes from the stream and converting them to |
442 |
* a single Java <code>long</code>. The bytes are stored most |
443 |
* significant byte first (i.e., "big endian") regardless of the native |
444 |
* host byte ordering. |
445 |
* <p> |
446 |
* As an example, if <code>byte1</code> through <code>byte8</code> represent |
447 |
* the first eight bytes read from the stream, they will be |
448 |
* transformed to an <code>long</code> in the following manner: |
449 |
* <p> |
450 |
* <code>(long)(((byte1 & 0xFF) << 56) + ((byte2 & 0xFF) << 48) + |
451 |
* ((byte3 & 0xFF) << 40) + ((byte4 & 0xFF) << 32) + |
452 |
* ((byte5 & 0xFF) << 24) + ((byte6 & 0xFF) << 16) + |
453 |
* ((byte7 & 0xFF) << 8) + (byte8 & 0xFF))) |
454 |
* </code> |
455 |
* <p> |
456 |
* The value returned is in the range of -9223372036854775808 to |
457 |
* 9223372036854775807. |
458 |
* <p> |
459 |
* This method can read an <code>long</code> written by an object |
460 |
* implementing the <code>writeLong()</code> method in the |
461 |
* <code>DataOutput</code> interface. |
462 |
* |
463 |
* @return The <code>long</code> value read |
464 |
* |
465 |
* @exception EOFException If end of file is reached before reading the long |
466 |
* @exception IOException If any other error occurs |
467 |
* |
468 |
* @see DataOutput#writeLong |
469 |
*/ |
470 |
public final long readLong () throws IOException |
471 |
{ |
472 |
readFully (buf, 0, 8); |
473 |
return convertToLong (buf); |
474 |
} |
475 |
|
476 |
/** |
477 |
* This method reads a signed 16-bit value into a Java in from the |
478 |
* stream. It operates by reading two bytes from the stream and |
479 |
* converting them to a single 16-bit Java <code>short</code>. The |
480 |
* two bytes are stored most significant byte first (i.e., "big |
481 |
* endian") regardless of the native host byte ordering. |
482 |
* <p> |
483 |
* As an example, if <code>byte1</code> and <code>byte2</code> |
484 |
* represent the first and second byte read from the stream |
485 |
* respectively, they will be transformed to a <code>short</code>. in |
486 |
* the following manner: |
487 |
* <p> |
488 |
* <code>(short)(((byte1 & 0xFF) << 8) | (byte2 & 0xFF))</code> |
489 |
* <p> |
490 |
* The value returned is in the range of -32768 to 32767. |
491 |
* <p> |
492 |
* This method can read a <code>short</code> written by an object |
493 |
* implementing the <code>writeShort()</code> method in the |
494 |
* <code>DataOutput</code> interface. |
495 |
* |
496 |
* @return The <code>short</code> value read |
497 |
* |
498 |
* @exception EOFException If end of file is reached before reading the value |
499 |
* @exception IOException If any other error occurs |
500 |
* |
501 |
* @see DataOutput#writeShort |
502 |
*/ |
503 |
public final short readShort () throws IOException |
504 |
{ |
505 |
readFully (buf, 0, 2); |
506 |
return convertToShort (buf); |
507 |
} |
508 |
|
509 |
/** |
510 |
* This method reads 8 unsigned bits into a Java <code>int</code> |
511 |
* value from the stream. The value returned is in the range of 0 to |
512 |
* 255. |
513 |
* <p> |
514 |
* This method can read an unsigned byte written by an object |
515 |
* implementing the <code>writeUnsignedByte()</code> method in the |
516 |
* <code>DataOutput</code> interface. |
517 |
* |
518 |
* @return The unsigned bytes value read as a Java <code>int</code>. |
519 |
* |
520 |
* @exception EOFException If end of file is reached before reading the value |
521 |
* @exception IOException If any other error occurs |
522 |
* |
523 |
* @see DataOutput#writeByte |
524 |
*/ |
525 |
public final int readUnsignedByte () throws IOException |
526 |
{ |
527 |
return convertToUnsignedByte (in.read ()); |
528 |
} |
529 |
|
530 |
/** |
531 |
* This method reads 16 unsigned bits into a Java int value from the stream. |
532 |
* It operates by reading two bytes from the stream and converting them to |
533 |
* a single Java <code>int</code> The two bytes are stored most |
534 |
* significant byte first (i.e., "big endian") regardless of the native |
535 |
* host byte ordering. |
536 |
* <p> |
537 |
* As an example, if <code>byte1</code> and <code>byte2</code> |
538 |
* represent the first and second byte read from the stream |
539 |
* respectively, they will be transformed to an <code>int</code> in |
540 |
* the following manner: |
541 |
* <p> |
542 |
* <code>(int)(((byte1 & 0xFF) << 8) + (byte2 & 0xFF))</code> |
543 |
* <p> |
544 |
* The value returned is in the range of 0 to 65535. |
545 |
* <p> |
546 |
* This method can read an unsigned short written by an object |
547 |
* implementing the <code>writeUnsignedShort()</code> method in the |
548 |
* <code>DataOutput</code> interface. |
549 |
* |
550 |
* @return The unsigned short value read as a Java <code>int</code> |
551 |
* |
552 |
* @exception EOFException If end of file is reached before reading the value |
553 |
* @exception IOException If any other error occurs |
554 |
* |
555 |
* @see DataOutput#writeShort |
556 |
*/ |
557 |
public final int readUnsignedShort () throws IOException |
558 |
{ |
559 |
readFully (buf, 0, 2); |
560 |
return convertToUnsignedShort (buf); |
561 |
} |
562 |
|
563 |
/** |
564 |
* This method reads a <code>String</code> from an input stream that |
565 |
* is encoded in a modified UTF-8 format. This format has a leading |
566 |
* two byte sequence that contains the remaining number of bytes to |
567 |
* read. This two byte sequence is read using the |
568 |
* <code>readUnsignedShort()</code> method of this interface. |
569 |
* <p> |
570 |
* After the number of remaining bytes have been determined, these |
571 |
* bytes are read an transformed into <code>char</code> values. |
572 |
* These <code>char</code> values are encoded in the stream using |
573 |
* either a one, two, or three byte format. The particular format |
574 |
* in use can be determined by examining the first byte read. |
575 |
* <p> |
576 |
* If the first byte has a high order bit of 0, then that character |
577 |
* consists on only one byte. This character value consists of |
578 |
* seven bits that are at positions 0 through 6 of the byte. As an |
579 |
* example, if <code>byte1</code> is the byte read from the stream, |
580 |
* it would be converted to a <code>char</code> like so: |
581 |
* <p> |
582 |
* <code>(char)byte1</code> |
583 |
* <p> |
584 |
* If the first byte has 110 as its high order bits, then the |
585 |
* character consists of two bytes. The bits that make up the character |
586 |
* value are in positions 0 through 4 of the first byte and bit positions |
587 |
* 0 through 5 of the second byte. (The second byte should have |
588 |
* 10 as its high order bits). These values are in most significant |
589 |
* byte first (i.e., "big endian") order. |
590 |
* <p> |
591 |
* As an example, if <code>byte1</code> and <code>byte2</code> are |
592 |
* the first two bytes read respectively, and the high order bits of |
593 |
* them match the patterns which indicate a two byte character |
594 |
* encoding, then they would be converted to a Java |
595 |
* <code>char</code> like so: |
596 |
* <p> |
597 |
* <code>(char)(((byte1 & 0x1F) << 6) | (byte2 & 0x3F))</code> |
598 |
* <p> |
599 |
* If the first byte has a 1110 as its high order bits, then the |
600 |
* character consists of three bytes. The bits that make up the character |
601 |
* value are in positions 0 through 3 of the first byte and bit positions |
602 |
* 0 through 5 of the other two bytes. (The second and third bytes should |
603 |
* have 10 as their high order bits). These values are in most |
604 |
* significant byte first (i.e., "big endian") order. |
605 |
* <p> |
606 |
* As an example, if <code>byte1</code> <code>byte2</code> and |
607 |
* <code>byte3</code> are the three bytes read, and the high order |
608 |
* bits of them match the patterns which indicate a three byte |
609 |
* character encoding, then they would be converted to a Java |
610 |
* <code>char</code> like so: |
611 |
* <p> |
612 |
* <code>(char)(((byte1 & 0x0F) << 12) | ((byte2 & 0x3F) << 6) | |
613 |
* (byte3 & 0x3F))</code> |
614 |
* <p> |
615 |
* Note that all characters are encoded in the method that requires |
616 |
* the fewest number of bytes with the exception of the character |
617 |
* with the value of <code>\u0000</code> which is encoded as two |
618 |
* bytes. This is a modification of the UTF standard used to |
619 |
* prevent C language style <code>NUL</code> values from appearing |
620 |
* in the byte stream. |
621 |
* <p> |
622 |
* This method can read data that was written by an object implementing the |
623 |
* <code>writeUTF()</code> method in <code>DataOutput</code> |
624 |
* |
625 |
* @return The <code>String</code> read |
626 |
* |
627 |
* @exception EOFException If end of file is reached before reading |
628 |
* the String |
629 |
* @exception UTFDataFormatException If the data is not in UTF-8 format |
630 |
* @exception IOException If any other error occurs |
631 |
* |
632 |
* @see DataOutput#writeUTF |
633 |
*/ |
634 |
public final String readUTF () throws IOException |
635 |
{ |
636 |
return readUTF (this); |
637 |
} |
638 |
|
639 |
/** |
640 |
* This method reads a String encoded in UTF-8 format from the |
641 |
* specified <code>DataInput</code> source. |
642 |
* |
643 |
* @param in The <code>DataInput</code> source to read from |
644 |
* |
645 |
* @return The String read from the source |
646 |
* |
647 |
* @exception IOException If an error occurs |
648 |
* |
649 |
* @see DataInput#readUTF |
650 |
*/ |
651 |
public final static String readUTF (DataInput in) throws IOException |
652 |
{ |
653 |
final int UTFlen = in.readUnsignedShort (); |
654 |
byte[] buf = new byte [UTFlen]; |
655 |
|
656 |
// This blocks until the entire string is available rather than |
657 |
// doing partial processing on the bytes that are available and then |
658 |
// blocking. An advantage of the latter is that Exceptions |
659 |
// could be thrown earlier. The former is a bit cleaner. |
660 |
in.readFully (buf, 0, UTFlen); |
661 |
|
662 |
return convertFromUTF (buf); |
663 |
} |
664 |
|
665 |
/** |
666 |
* This method attempts to skip and discard the specified number of bytes |
667 |
* in the input stream. It may actually skip fewer bytes than requested. |
668 |
* This method will not skip any bytes if passed a negative number of bytes |
669 |
* to skip. |
670 |
* |
671 |
* @param n The requested number of bytes to skip. |
672 |
* |
673 |
* @return The requested number of bytes to skip. |
674 |
* |
675 |
* @exception IOException If an error occurs. |
676 |
* @specnote The JDK docs claim that this returns the number of bytes |
677 |
* actually skipped. The JCL claims that this method can throw an |
678 |
* EOFException. Neither of these appear to be true in the JDK 1.3's |
679 |
* implementation. This tries to implement the actual JDK behaviour. |
680 |
*/ |
681 |
public final int skipBytes (int n) throws IOException |
682 |
{ |
683 |
if (n <= 0) |
684 |
return 0; |
685 |
try |
686 |
{ |
687 |
return (int) in.skip (n); |
688 |
} |
689 |
catch (EOFException x) |
690 |
{ |
691 |
// do nothing. |
692 |
} |
693 |
return n; |
694 |
} |
695 |
|
696 |
static boolean convertToBoolean (int b) throws EOFException |
697 |
{ |
698 |
if (b < 0) |
699 |
throw new EOFException (); |
700 |
|
701 |
return (b != 0); |
702 |
} |
703 |
|
704 |
static byte convertToByte (int i) throws EOFException |
705 |
{ |
706 |
if (i < 0) |
707 |
throw new EOFException (); |
708 |
|
709 |
return (byte) i; |
710 |
} |
711 |
|
712 |
static int convertToUnsignedByte (int i) throws EOFException |
713 |
{ |
714 |
if (i < 0) |
715 |
throw new EOFException (); |
716 |
|
717 |
return (i & 0xFF); |
718 |
} |
719 |
|
720 |
static char convertToChar (byte[] buf) |
721 |
{ |
722 |
return (char) ((buf [0] << 8) |
723 |
| (buf [1] & 0xff)); |
724 |
} |
725 |
|
726 |
static short convertToShort (byte[] buf) |
727 |
{ |
728 |
return (short) ((buf [0] << 8) |
729 |
| (buf [1] & 0xff)); |
730 |
} |
731 |
|
732 |
static int convertToUnsignedShort (byte[] buf) |
733 |
{ |
734 |
return (((buf [0] & 0xff) << 8) |
735 |
| (buf [1] & 0xff)); |
736 |
} |
737 |
|
738 |
static int convertToInt (byte[] buf) |
739 |
{ |
740 |
return (((buf [0] & 0xff) << 24) |
741 |
| ((buf [1] & 0xff) << 16) |
742 |
| ((buf [2] & 0xff) << 8) |
743 |
| (buf [3] & 0xff)); |
744 |
} |
745 |
|
746 |
static long convertToLong (byte[] buf) |
747 |
{ |
748 |
return (((long)(buf [0] & 0xff) << 56) | |
749 |
((long)(buf [1] & 0xff) << 48) | |
750 |
((long)(buf [2] & 0xff) << 40) | |
751 |
((long)(buf [3] & 0xff) << 32) | |
752 |
((long)(buf [4] & 0xff) << 24) | |
753 |
((long)(buf [5] & 0xff) << 16) | |
754 |
((long)(buf [6] & 0xff) << 8) | |
755 |
((long)(buf [7] & 0xff))); |
756 |
} |
757 |
|
758 |
// FIXME: This method should be re-thought. I suspect we have multiple |
759 |
// UTF-8 decoders floating around. We should use the standard charset |
760 |
// converters, maybe and adding a direct call into one of the new |
761 |
// NIO converters for a super-fast UTF8 decode. |
762 |
static String convertFromUTF (byte[] buf) |
763 |
throws EOFException, UTFDataFormatException |
764 |
{ |
765 |
// Give StringBuffer an initial estimated size to avoid |
766 |
// enlarge buffer frequently |
767 |
StringBuffer strbuf = new StringBuffer (buf.length / 2 + 2); |
768 |
|
769 |
for (int i = 0; i < buf.length; ) |
770 |
{ |
771 |
if ((buf [i] & 0x80) == 0) // bit pattern 0xxxxxxx |
772 |
strbuf.append ((char) (buf [i++] & 0xFF)); |
773 |
else if ((buf [i] & 0xE0) == 0xC0) // bit pattern 110xxxxx |
774 |
{ |
775 |
if (i + 1 >= buf.length |
776 |
|| (buf [i + 1] & 0xC0) != 0x80) |
777 |
throw new UTFDataFormatException (); |
778 |
|
779 |
strbuf.append((char) (((buf [i++] & 0x1F) << 6) |
780 |
| (buf [i++] & 0x3F))); |
781 |
} |
782 |
else if ((buf [i] & 0xF0) == 0xE0) // bit pattern 1110xxxx |
783 |
{ |
784 |
if (i + 2 >= buf.length |
785 |
|| (buf [i + 1] & 0xC0) != 0x80 |
786 |
|| (buf [i + 2] & 0xC0) != 0x80) |
787 |
throw new UTFDataFormatException (); |
788 |
|
789 |
strbuf.append ((char) (((buf [i++] & 0x0F) << 12) |
790 |
| ((buf [i++] & 0x3F) << 6) |
791 |
| (buf [i++] & 0x3F))); |
792 |
} |
793 |
else // must be ((buf [i] & 0xF0) == 0xF0 || (buf [i] & 0xC0) == 0x80) |
794 |
throw new UTFDataFormatException (); // bit patterns 1111xxxx or |
795 |
// 10xxxxxx |
796 |
} |
797 |
|
798 |
return strbuf.toString (); |
799 |
} |
800 |
} |