1 /*
2 * Copyright (c) 1995, 2006, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 package java.io;
27
28 /**
29 * The <code>DataInput</code> interface provides
30 * for reading bytes from a binary stream and
31 * reconstructing from them data in any of
32 * the Java primitive types. There is also
33 * a
34 * facility for reconstructing a <code>String</code>
35 * from data in
36 * <a href="#modified-utf-8">modified UTF-8</a>
37 * format.
38 * <p>
39 * It is generally true of all the reading
40 * routines in this interface that if end of
41 * file is reached before the desired number
42 * of bytes has been read, an <code>EOFException</code>
43 * (which is a kind of <code>IOException</code>)
44 * is thrown. If any byte cannot be read for
45 * any reason other than end of file, an <code>IOException</code>
46 * other than <code>EOFException</code> is
47 * thrown. In particular, an <code>IOException</code>
48 * may be thrown if the input stream has been
49 * closed.
50 *
51 * <h4><a name="modified-utf-8">Modified UTF-8</a></h4>
52 * <p>
53 * Implementations of the DataInput and DataOutput interfaces represent
54 * Unicode strings in a format that is a slight modification of UTF-8.
55 * (For information regarding the standard UTF-8 format, see section
56 * <i>3.9 Unicode Encoding Forms</i> of <i>The Unicode Standard, Version
57 * 4.0</i>).
58 * Note that in the following tables, the most significant bit appears in the
59 * far left-hand column.
60 * <p>
61 * All characters in the range <code>'\u0001'</code> to
62 * <code>'\u007F'</code> are represented by a single byte:
63 *
64 * <blockquote>
65 * <table border="1" cellspacing="0" cellpadding="8" width="50%"
66 * summary="Bit values and bytes">
67 * <tr>
68 * <td></td>
69 * <th id="bit">Bit Values</th>
70 * </tr>
71 * <tr>
72 * <th id="byte1">Byte 1</th>
73 * <td>
74 * <table border="1" cellspacing="0" width="100%">
75 * <tr>
76 * <td width="12%"><center>0</center>
77 * <td colspan="7"><center>bits 6-0</center>
78 * </tr>
79 * </table>
80 * </td>
81 * </tr>
82 * </table>
83 * </blockquote>
84 *
85 * <p>
86 * The null character <code>'\u0000'</code> and characters in the
87 * range <code>'\u0080'</code> to <code>'\u07FF'</code> are
88 * represented by a pair of bytes:
89 *
90 * <blockquote>
91 * <table border="1" cellspacing="0" cellpadding="8" width="50%"
92 * summary="Bit values and bytes">
93 * <tr>
94 * <td></td>
95 * <th id="bit">Bit Values</th>
96 * </tr>
97 * <tr>
98 * <th id="byte1">Byte 1</th>
99 * <td>
100 * <table border="1" cellspacing="0" width="100%">
101 * <tr>
102 * <td width="12%"><center>1</center>
103 * <td width="13%"><center>1</center>
104 * <td width="12%"><center>0</center>
105 * <td colspan="5"><center>bits 10-6</center>
106 * </tr>
107 * </table>
108 * </td>
109 * </tr>
110 * <tr>
111 * <th id="byte2">Byte 2</th>
112 * <td>
113 * <table border="1" cellspacing="0" width="100%">
114 * <tr>
115 * <td width="12%"><center>1</center>
116 * <td width="13%"><center>0</center>
117 * <td colspan="6"><center>bits 5-0</center>
118 * </tr>
119 * </table>
120 * </td>
121 * </tr>
122 * </table>
123 * </blockquote>
124 *
125 * <br>
126 * <code>char</code> values in the range <code>'\u0800'</code> to
127 * <code>'\uFFFF'</code> are represented by three bytes:
128 *
129 * <blockquote>
130 * <table border="1" cellspacing="0" cellpadding="8" width="50%"
131 * summary="Bit values and bytes">
132 * <tr>
133 * <td></td>
134 * <th id="bit">Bit Values</th>
135 * </tr>
136 * <tr>
137 * <th id="byte1">Byte 1</th>
138 * <td>
139 * <table border="1" cellspacing="0" width="100%">
140 * <tr>
141 * <td width="12%"><center>1</center>
142 * <td width="13%"><center>1</center>
143 * <td width="12%"><center>1</center>
144 * <td width="13%"><center>0</center>
145 * <td colspan="4"><center>bits 15-12</center>
146 * </tr>
147 * </table>
148 * </td>
149 * </tr>
150 * <tr>
151 * <th id="byte2">Byte 2</th>
152 * <td>
153 * <table border="1" cellspacing="0" width="100%">
154 * <tr>
155 * <td width="12%"><center>1</center>
156 * <td width="13%"><center>0</center>
157 * <td colspan="6"><center>bits 11-6</center>
158 * </tr>
159 * </table>
160 * </td>
161 * </tr>
162 * <tr>
163 * <th id="byte3">Byte 3</th>
164 * <td>
165 * <table border="1" cellspacing="0" width="100%">
166 * <tr>
167 * <td width="12%"><center>1</center>
168 * <td width="13%"><center>0</center>
169 * <td colspan="6"><center>bits 5-0</center>
170 * </tr>
171 * </table>
172 * </td>
173 * </tr>
174 * </table>
175 * </blockquote>
176 *
177 * <p>
178 * The differences between this format and the
179 * standard UTF-8 format are the following:
180 * <ul>
181 * <li>The null byte <code>'\u0000'</code> is encoded in 2-byte format
182 * rather than 1-byte, so that the encoded strings never have
183 * embedded nulls.
184 * <li>Only the 1-byte, 2-byte, and 3-byte formats are used.
185 * <li><a href="../lang/Character.html#unicode">Supplementary characters</a>
186 * are represented in the form of surrogate pairs.
187 * </ul>
188 * @author Frank Yellin
189 * @see java.io.DataInputStream
190 * @see java.io.DataOutput
191 * @since JDK1.0
192 */
193 public
194 interface DataInput {
195 /**
196 * Reads some bytes from an input
197 * stream and stores them into the buffer
198 * array <code>b</code>. The number of bytes
199 * read is equal
200 * to the length of <code>b</code>.
201 * <p>
202 * This method blocks until one of the
203 * following conditions occurs:<p>
204 * <ul>
205 * <li><code>b.length</code>
206 * bytes of input data are available, in which
207 * case a normal return is made.
208 *
209 * <li>End of
210 * file is detected, in which case an <code>EOFException</code>
211 * is thrown.
212 *
213 * <li>An I/O error occurs, in
214 * which case an <code>IOException</code> other
215 * than <code>EOFException</code> is thrown.
216 * </ul>
217 * <p>
218 * If <code>b</code> is <code>null</code>,
219 * a <code>NullPointerException</code> is thrown.
220 * If <code>b.length</code> is zero, then
221 * no bytes are read. Otherwise, the first
222 * byte read is stored into element <code>b[0]</code>,
223 * the next one into <code>b[1]</code>, and
224 * so on.
225 * If an exception is thrown from
226 * this method, then it may be that some but
227 * not all bytes of <code>b</code> have been
228 * updated with data from the input stream.
229 *
230 * @param b the buffer into which the data is read.
231 * @exception EOFException if this stream reaches the end before reading
232 * all the bytes.
233 * @exception IOException if an I/O error occurs.
234 */
235 void readFully(byte b[]) throws IOException;
236
237 /**
238 *
239 * Reads <code>len</code>
240 * bytes from
241 * an input stream.
242 * <p>
243 * This method
244 * blocks until one of the following conditions
245 * occurs:<p>
246 * <ul>
247 * <li><code>len</code> bytes
248 * of input data are available, in which case
249 * a normal return is made.
250 *
251 * <li>End of file
252 * is detected, in which case an <code>EOFException</code>
253 * is thrown.
254 *
255 * <li>An I/O error occurs, in
256 * which case an <code>IOException</code> other
257 * than <code>EOFException</code> is thrown.
258 * </ul>
259 * <p>
260 * If <code>b</code> is <code>null</code>,
261 * a <code>NullPointerException</code> is thrown.
262 * If <code>off</code> is negative, or <code>len</code>
263 * is negative, or <code>off+len</code> is
264 * greater than the length of the array <code>b</code>,
265 * then an <code>IndexOutOfBoundsException</code>
266 * is thrown.
267 * If <code>len</code> is zero,
268 * then no bytes are read. Otherwise, the first
269 * byte read is stored into element <code>b[off]</code>,
270 * the next one into <code>b[off+1]</code>,
271 * and so on. The number of bytes read is,
272 * at most, equal to <code>len</code>.
273 *
274 * @param b the buffer into which the data is read.
275 * @param off an int specifying the offset into the data.
276 * @param len an int specifying the number of bytes to read.
277 * @exception EOFException if this stream reaches the end before reading
278 * all the bytes.
279 * @exception IOException if an I/O error occurs.
280 */
281 void readFully(byte b[], int off, int len) throws IOException;
282
283 /**
284 * Makes an attempt to skip over
285 * <code>n</code> bytes
286 * of data from the input
287 * stream, discarding the skipped bytes. However,
288 * it may skip
289 * over some smaller number of
290 * bytes, possibly zero. This may result from
291 * any of a
292 * number of conditions; reaching
293 * end of file before <code>n</code> bytes
294 * have been skipped is
295 * only one possibility.
296 * This method never throws an <code>EOFException</code>.
297 * The actual
298 * number of bytes skipped is returned.
299 *
300 * @param n the number of bytes to be skipped.
301 * @return the number of bytes actually skipped.
302 * @exception IOException if an I/O error occurs.
303 */
304 int skipBytes(int n) throws IOException;
305
306 /**
307 * Reads one input byte and returns
308 * <code>true</code> if that byte is nonzero,
309 * <code>false</code> if that byte is zero.
310 * This method is suitable for reading
311 * the byte written by the <code>writeBoolean</code>
312 * method of interface <code>DataOutput</code>.
313 *
314 * @return the <code>boolean</code> value read.
315 * @exception EOFException if this stream reaches the end before reading
316 * all the bytes.
317 * @exception IOException if an I/O error occurs.
318 */
319 boolean readBoolean() throws IOException;
320
321 /**
322 * Reads and returns one input byte.
323 * The byte is treated as a signed value in
324 * the range <code>-128</code> through <code>127</code>,
325 * inclusive.
326 * This method is suitable for
327 * reading the byte written by the <code>writeByte</code>
328 * method of interface <code>DataOutput</code>.
329 *
330 * @return the 8-bit value read.
331 * @exception EOFException if this stream reaches the end before reading
332 * all the bytes.
333 * @exception IOException if an I/O error occurs.
334 */
335 byte readByte() throws IOException;
336
337 /**
338 * Reads one input byte, zero-extends
339 * it to type <code>int</code>, and returns
340 * the result, which is therefore in the range
341 * <code>0</code>
342 * through <code>255</code>.
343 * This method is suitable for reading
344 * the byte written by the <code>writeByte</code>
345 * method of interface <code>DataOutput</code>
346 * if the argument to <code>writeByte</code>
347 * was intended to be a value in the range
348 * <code>0</code> through <code>255</code>.
349 *
350 * @return the unsigned 8-bit value read.
351 * @exception EOFException if this stream reaches the end before reading
352 * all the bytes.
353 * @exception IOException if an I/O error occurs.
354 */
355 int readUnsignedByte() throws IOException;
356
357 /**
358 * Reads two input bytes and returns
359 * a <code>short</code> value. Let <code>a</code>
360 * be the first byte read and <code>b</code>
361 * be the second byte. The value
362 * returned
363 * is:
364 * <p><pre><code>(short)((a << 8) | (b & 0xff))
365 * </code></pre>
366 * This method
367 * is suitable for reading the bytes written
368 * by the <code>writeShort</code> method of
369 * interface <code>DataOutput</code>.
370 *
371 * @return the 16-bit value read.
372 * @exception EOFException if this stream reaches the end before reading
373 * all the bytes.
374 * @exception IOException if an I/O error occurs.
375 */
376 short readShort() throws IOException;
377
378 /**
379 * Reads two input bytes and returns
380 * an <code>int</code> value in the range <code>0</code>
381 * through <code>65535</code>. Let <code>a</code>
382 * be the first byte read and
383 * <code>b</code>
384 * be the second byte. The value returned is:
385 * <p><pre><code>(((a & 0xff) << 8) | (b & 0xff))
386 * </code></pre>
387 * This method is suitable for reading the bytes
388 * written by the <code>writeShort</code> method
389 * of interface <code>DataOutput</code> if
390 * the argument to <code>writeShort</code>
391 * was intended to be a value in the range
392 * <code>0</code> through <code>65535</code>.
393 *
394 * @return the unsigned 16-bit value read.
395 * @exception EOFException if this stream reaches the end before reading
396 * all the bytes.
397 * @exception IOException if an I/O error occurs.
398 */
399 int readUnsignedShort() throws IOException;
400
401 /**
402 * Reads two input bytes and returns a <code>char</code> value.
403 * Let <code>a</code>
404 * be the first byte read and <code>b</code>
405 * be the second byte. The value
406 * returned is:
407 * <p><pre><code>(char)((a << 8) | (b & 0xff))
408 * </code></pre>
409 * This method
410 * is suitable for reading bytes written by
411 * the <code>writeChar</code> method of interface
412 * <code>DataOutput</code>.
413 *
414 * @return the <code>char</code> value read.
415 * @exception EOFException if this stream reaches the end before reading
416 * all the bytes.
417 * @exception IOException if an I/O error occurs.
418 */
419 char readChar() throws IOException;
420
421 /**
422 * Reads four input bytes and returns an
423 * <code>int</code> value. Let <code>a-d</code>
424 * be the first through fourth bytes read. The value returned is:
425 * <p><pre>
426 * <code>
427 * (((a & 0xff) << 24) | ((b & 0xff) << 16) |
428 *  ((c & 0xff) << 8) | (d & 0xff))
429 * </code></pre>
430 * This method is suitable
431 * for reading bytes written by the <code>writeInt</code>
432 * method of interface <code>DataOutput</code>.
433 *
434 * @return the <code>int</code> value read.
435 * @exception EOFException if this stream reaches the end before reading
436 * all the bytes.
437 * @exception IOException if an I/O error occurs.
438 */
439 int readInt() throws IOException;
440
441 /**
442 * Reads eight input bytes and returns
443 * a <code>long</code> value. Let <code>a-h</code>
444 * be the first through eighth bytes read.
445 * The value returned is:
446 * <p><pre> <code>
447 * (((long)(a & 0xff) << 56) |
448 * ((long)(b & 0xff) << 48) |
449 * ((long)(c & 0xff) << 40) |
450 * ((long)(d & 0xff) << 32) |
451 * ((long)(e & 0xff) << 24) |
452 * ((long)(f & 0xff) << 16) |
453 * ((long)(g & 0xff) << 8) |
454 * ((long)(h & 0xff)))
455 * </code></pre>
456 * <p>
457 * This method is suitable
458 * for reading bytes written by the <code>writeLong</code>
459 * method of interface <code>DataOutput</code>.
460 *
461 * @return the <code>long</code> value read.
462 * @exception EOFException if this stream reaches the end before reading
463 * all the bytes.
464 * @exception IOException if an I/O error occurs.
465 */
466 long readLong() throws IOException;
467
468 /**
469 * Reads four input bytes and returns
470 * a <code>float</code> value. It does this
471 * by first constructing an <code>int</code>
472 * value in exactly the manner
473 * of the <code>readInt</code>
474 * method, then converting this <code>int</code>
475 * value to a <code>float</code> in
476 * exactly the manner of the method <code>Float.intBitsToFloat</code>.
477 * This method is suitable for reading
478 * bytes written by the <code>writeFloat</code>
479 * method of interface <code>DataOutput</code>.
480 *
481 * @return the <code>float</code> value read.
482 * @exception EOFException if this stream reaches the end before reading
483 * all the bytes.
484 * @exception IOException if an I/O error occurs.
485 */
486 float readFloat() throws IOException;
487
488 /**
489 * Reads eight input bytes and returns
490 * a <code>double</code> value. It does this
491 * by first constructing a <code>long</code>
492 * value in exactly the manner
493 * of the <code>readlong</code>
494 * method, then converting this <code>long</code>
495 * value to a <code>double</code> in exactly
496 * the manner of the method <code>Double.longBitsToDouble</code>.
497 * This method is suitable for reading
498 * bytes written by the <code>writeDouble</code>
499 * method of interface <code>DataOutput</code>.
500 *
501 * @return the <code>double</code> value read.
502 * @exception EOFException if this stream reaches the end before reading
503 * all the bytes.
504 * @exception IOException if an I/O error occurs.
505 */
506 double readDouble() throws IOException;
507
508 /**
509 * Reads the next line of text from the input stream.
510 * It reads successive bytes, converting
511 * each byte separately into a character,
512 * until it encounters a line terminator or
513 * end of
514 * file; the characters read are then
515 * returned as a <code>String</code>. Note
516 * that because this
517 * method processes bytes,
518 * it does not support input of the full Unicode
519 * character set.
520 * <p>
521 * If end of file is encountered
522 * before even one byte can be read, then <code>null</code>
523 * is returned. Otherwise, each byte that is
524 * read is converted to type <code>char</code>
525 * by zero-extension. If the character <code>'\n'</code>
526 * is encountered, it is discarded and reading
527 * ceases. If the character <code>'\r'</code>
528 * is encountered, it is discarded and, if
529 * the following byte converts  to the
530 * character <code>'\n'</code>, then that is
531 * discarded also; reading then ceases. If
532 * end of file is encountered before either
533 * of the characters <code>'\n'</code> and
534 * <code>'\r'</code> is encountered, reading
535 * ceases. Once reading has ceased, a <code>String</code>
536 * is returned that contains all the characters
537 * read and not discarded, taken in order.
538 * Note that every character in this string
539 * will have a value less than <code>\u0100</code>,
540 * that is, <code>(char)256</code>.
541 *
542 * @return the next line of text from the input stream,
543 * or <CODE>null</CODE> if the end of file is
544 * encountered before a byte can be read.
545 * @exception IOException if an I/O error occurs.
546 */
547 String readLine() throws IOException;
548
549 /**
550 * Reads in a string that has been encoded using a
551 * <a href="#modified-utf-8">modified UTF-8</a>
552 * format.
553 * The general contract of <code>readUTF</code>
554 * is that it reads a representation of a Unicode
555 * character string encoded in modified
556 * UTF-8 format; this string of characters
557 * is then returned as a <code>String</code>.
558 * <p>
559 * First, two bytes are read and used to
560 * construct an unsigned 16-bit integer in
561 * exactly the manner of the <code>readUnsignedShort</code>
562 * method . This integer value is called the
563 * <i>UTF length</i> and specifies the number
564 * of additional bytes to be read. These bytes
565 * are then converted to characters by considering
566 * them in groups. The length of each group
567 * is computed from the value of the first
568 * byte of the group. The byte following a
569 * group, if any, is the first byte of the
570 * next group.
571 * <p>
572 * If the first byte of a group
573 * matches the bit pattern <code>0xxxxxxx</code>
574 * (where <code>x</code> means "may be <code>0</code>
575 * or <code>1</code>"), then the group consists
576 * of just that byte. The byte is zero-extended
577 * to form a character.
578 * <p>
579 * If the first byte
580 * of a group matches the bit pattern <code>110xxxxx</code>,
581 * then the group consists of that byte <code>a</code>
582 * and a second byte <code>b</code>. If there
583 * is no byte <code>b</code> (because byte
584 * <code>a</code> was the last of the bytes
585 * to be read), or if byte <code>b</code> does
586 * not match the bit pattern <code>10xxxxxx</code>,
587 * then a <code>UTFDataFormatException</code>
588 * is thrown. Otherwise, the group is converted
589 * to the character:<p>
590 * <pre><code>(char)(((a& 0x1F) << 6) | (b & 0x3F))
591 * </code></pre>
592 * If the first byte of a group
593 * matches the bit pattern <code>1110xxxx</code>,
594 * then the group consists of that byte <code>a</code>
595 * and two more bytes <code>b</code> and <code>c</code>.
596 * If there is no byte <code>c</code> (because
597 * byte <code>a</code> was one of the last
598 * two of the bytes to be read), or either
599 * byte <code>b</code> or byte <code>c</code>
600 * does not match the bit pattern <code>10xxxxxx</code>,
601 * then a <code>UTFDataFormatException</code>
602 * is thrown. Otherwise, the group is converted
603 * to the character:<p>
604 * <pre><code>
605 * (char)(((a & 0x0F) << 12) | ((b & 0x3F) << 6) | (c & 0x3F))
606 * </code></pre>
607 * If the first byte of a group matches the
608 * pattern <code>1111xxxx</code> or the pattern
609 * <code>10xxxxxx</code>, then a <code>UTFDataFormatException</code>
610 * is thrown.
611 * <p>
612 * If end of file is encountered
613 * at any time during this entire process,
614 * then an <code>EOFException</code> is thrown.
615 * <p>
616 * After every group has been converted to
617 * a character by this process, the characters
618 * are gathered, in the same order in which
619 * their corresponding groups were read from
620 * the input stream, to form a <code>String</code>,
621 * which is returned.
622 * <p>
623 * The <code>writeUTF</code>
624 * method of interface <code>DataOutput</code>
625 * may be used to write data that is suitable
626 * for reading by this method.
627 * @return a Unicode string.
628 * @exception EOFException if this stream reaches the end
629 * before reading all the bytes.
630 * @exception IOException if an I/O error occurs.
631 * @exception UTFDataFormatException if the bytes do not represent a
632 * valid modified UTF-8 encoding of a string.
633 */
634 String readUTF() throws IOException;
635 }