1 /*
2 * Copyright (c) 1995, 2013, 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} 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}
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}
43 * (which is a kind of {@code IOException})
44 * is thrown. If any byte cannot be read for
45 * any reason other than end of file, an {@code IOException}
46 * other than {@code EOFException} is
47 * thrown. In particular, an {@code IOException}
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 '\u005Cu0001'} to
62 * {@code '\u005Cu007F'} 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_a">Bit Values</th>
70 * </tr>
71 * <tr>
72 * <th id="byte1_a">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 '\u005Cu0000'} and characters in the
87 * range {@code '\u005Cu0080'} to {@code '\u005Cu07FF'} 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_b">Bit Values</th>
96 * </tr>
97 * <tr>
98 * <th id="byte1_b">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_a">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} values in the range {@code '\u005Cu0800'} to
127 * {@code '\u005CuFFFF'} 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_c">Bit Values</th>
135 * </tr>
136 * <tr>
137 * <th id="byte1_c">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_b">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 '\u005Cu0000'} 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}. The number of bytes
199 * read is equal
200 * to the length of {@code b}.
201 * <p>
202 * This method blocks until one of the
203 * following conditions occurs:<p>
204 * <ul>
205 * <li>{@code b.length}
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}
211 * is thrown.
212 *
213 * <li>An I/O error occurs, in
214 * which case an {@code IOException} other
215 * than {@code EOFException} is thrown.
216 * </ul>
217 * <p>
218 * If {@code b} is {@code null},
219 * a {@code NullPointerException} is thrown.
220 * If {@code b.length} is zero, then
221 * no bytes are read. Otherwise, the first
222 * byte read is stored into element {@code b[0]},
223 * the next one into {@code b[1]}, 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} 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}
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} 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}
253 * is thrown.
254 *
255 * <li>An I/O error occurs, in
256 * which case an {@code IOException} other
257 * than {@code EOFException} is thrown.
258 * </ul>
259 * <p>
260 * If {@code b} is {@code null},
261 * a {@code NullPointerException} is thrown.
262 * If {@code off} is negative, or {@code len}
263 * is negative, or {@code off+len} is
264 * greater than the length of the array {@code b},
265 * then an {@code IndexOutOfBoundsException}
266 * is thrown.
267 * If {@code len} is zero,
268 * then no bytes are read. Otherwise, the first
269 * byte read is stored into element {@code b[off]},
270 * the next one into {@code b[off+1]},
271 * and so on. The number of bytes read is,
272 * at most, equal to {@code len}.
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} 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} bytes
294 * have been skipped is
295 * only one possibility.
296 * This method never throws an {@code EOFException}.
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} if that byte is nonzero,
309 * {@code false} if that byte is zero.
310 * This method is suitable for reading
311 * the byte written by the {@code writeBoolean}
312 * method of interface {@code DataOutput}.
313 *
314 * @return the {@code boolean} 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} through {@code 127},
325 * inclusive.
326 * This method is suitable for
327 * reading the byte written by the {@code writeByte}
328 * method of interface {@code DataOutput}.
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}, and returns
340 * the result, which is therefore in the range
341 * {@code 0}
342 * through {@code 255}.
343 * This method is suitable for reading
344 * the byte written by the {@code writeByte}
345 * method of interface {@code DataOutput}
346 * if the argument to {@code writeByte}
347 * was intended to be a value in the range
348 * {@code 0} through {@code 255}.
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} value. Let {@code a}
360 * be the first byte read and {@code b}
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} method of
369 * interface {@code DataOutput}.
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} value in the range {@code 0}
381 * through {@code 65535}. Let {@code a}
382 * be the first byte read and
383 * {@code b}
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} method
389 * of interface {@code DataOutput} if
390 * the argument to {@code writeShort}
391 * was intended to be a value in the range
392 * {@code 0} through {@code 65535}.
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} value.
403 * Let {@code a}
404 * be the first byte read and {@code b}
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} method of interface
412 * {@code DataOutput}.
413 *
414 * @return the {@code char} 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} value. Let {@code a-d}
424 * be the first through fourth bytes read. The value returned is:
425 * <p><pre><code>
426 * (((a & 0xff) << 24) | ((b & 0xff) << 16) |
427 *  ((c & 0xff) << 8) | (d & 0xff))
428 * </code></pre>
429 * This method is suitable
430 * for reading bytes written by the {@code writeInt}
431 * method of interface {@code DataOutput}.
432 *
433 * @return the {@code int} value read.
434 * @exception EOFException if this stream reaches the end before reading
435 * all the bytes.
436 * @exception IOException if an I/O error occurs.
437 */
438 int readInt() throws IOException;
439
440 /**
441 * Reads eight input bytes and returns
442 * a {@code long} value. Let {@code a-h}
443 * be the first through eighth bytes read.
444 * The value returned is:
445 * <p><pre><code>
446 * (((long)(a & 0xff) << 56) |
447 * ((long)(b & 0xff) << 48) |
448 * ((long)(c & 0xff) << 40) |
449 * ((long)(d & 0xff) << 32) |
450 * ((long)(e & 0xff) << 24) |
451 * ((long)(f & 0xff) << 16) |
452 * ((long)(g & 0xff) << 8) |
453 * ((long)(h & 0xff)))
454 * </code></pre>
455 * <p>
456 * This method is suitable
457 * for reading bytes written by the {@code writeLong}
458 * method of interface {@code DataOutput}.
459 *
460 * @return the {@code long} value read.
461 * @exception EOFException if this stream reaches the end before reading
462 * all the bytes.
463 * @exception IOException if an I/O error occurs.
464 */
465 long readLong() throws IOException;
466
467 /**
468 * Reads four input bytes and returns
469 * a {@code float} value. It does this
470 * by first constructing an {@code int}
471 * value in exactly the manner
472 * of the {@code readInt}
473 * method, then converting this {@code int}
474 * value to a {@code float} in
475 * exactly the manner of the method {@code Float.intBitsToFloat}.
476 * This method is suitable for reading
477 * bytes written by the {@code writeFloat}
478 * method of interface {@code DataOutput}.
479 *
480 * @return the {@code float} value read.
481 * @exception EOFException if this stream reaches the end before reading
482 * all the bytes.
483 * @exception IOException if an I/O error occurs.
484 */
485 float readFloat() throws IOException;
486
487 /**
488 * Reads eight input bytes and returns
489 * a {@code double} value. It does this
490 * by first constructing a {@code long}
491 * value in exactly the manner
492 * of the {@code readlong}
493 * method, then converting this {@code long}
494 * value to a {@code double} in exactly
495 * the manner of the method {@code Double.longBitsToDouble}.
496 * This method is suitable for reading
497 * bytes written by the {@code writeDouble}
498 * method of interface {@code DataOutput}.
499 *
500 * @return the {@code double} value read.
501 * @exception EOFException if this stream reaches the end before reading
502 * all the bytes.
503 * @exception IOException if an I/O error occurs.
504 */
505 double readDouble() throws IOException;
506
507 /**
508 * Reads the next line of text from the input stream.
509 * It reads successive bytes, converting
510 * each byte separately into a character,
511 * until it encounters a line terminator or
512 * end of
513 * file; the characters read are then
514 * returned as a {@code String}. Note
515 * that because this
516 * method processes bytes,
517 * it does not support input of the full Unicode
518 * character set.
519 * <p>
520 * If end of file is encountered
521 * before even one byte can be read, then {@code null}
522 * is returned. Otherwise, each byte that is
523 * read is converted to type {@code char}
524 * by zero-extension. If the character {@code '\n'}
525 * is encountered, it is discarded and reading
526 * ceases. If the character {@code '\r'}
527 * is encountered, it is discarded and, if
528 * the following byte converts  to the
529 * character {@code '\n'}, then that is
530 * discarded also; reading then ceases. If
531 * end of file is encountered before either
532 * of the characters {@code '\n'} and
533 * {@code '\r'} is encountered, reading
534 * ceases. Once reading has ceased, a {@code String}
535 * is returned that contains all the characters
536 * read and not discarded, taken in order.
537 * Note that every character in this string
538 * will have a value less than {@code \u005Cu0100},
539 * that is, {@code (char)256}.
540 *
541 * @return the next line of text from the input stream,
542 * or {@code null} if the end of file is
543 * encountered before a byte can be read.
544 * @exception IOException if an I/O error occurs.
545 */
546 String readLine() throws IOException;
547
548 /**
549 * Reads in a string that has been encoded using a
550 * <a href="#modified-utf-8">modified UTF-8</a>
551 * format.
552 * The general contract of {@code readUTF}
553 * is that it reads a representation of a Unicode
554 * character string encoded in modified
555 * UTF-8 format; this string of characters
556 * is then returned as a {@code String}.
557 * <p>
558 * First, two bytes are read and used to
559 * construct an unsigned 16-bit integer in
560 * exactly the manner of the {@code readUnsignedShort}
561 * method . This integer value is called the
562 * <i>UTF length</i> and specifies the number
563 * of additional bytes to be read. These bytes
564 * are then converted to characters by considering
565 * them in groups. The length of each group
566 * is computed from the value of the first
567 * byte of the group. The byte following a
568 * group, if any, is the first byte of the
569 * next group.
570 * <p>
571 * If the first byte of a group
572 * matches the bit pattern {@code 0xxxxxxx}
573 * (where {@code x} means "may be {@code 0}
574 * or {@code 1}"), then the group consists
575 * of just that byte. The byte is zero-extended
576 * to form a character.
577 * <p>
578 * If the first byte
579 * of a group matches the bit pattern {@code 110xxxxx},
580 * then the group consists of that byte {@code a}
581 * and a second byte {@code b}. If there
582 * is no byte {@code b} (because byte
583 * {@code a} was the last of the bytes
584 * to be read), or if byte {@code b} does
585 * not match the bit pattern {@code 10xxxxxx},
586 * then a {@code UTFDataFormatException}
587 * is thrown. Otherwise, the group is converted
588 * to the character:<p>
589 * <pre><code>(char)(((a& 0x1F) << 6) | (b & 0x3F))
590 * </code></pre>
591 * If the first byte of a group
592 * matches the bit pattern {@code 1110xxxx},
593 * then the group consists of that byte {@code a}
594 * and two more bytes {@code b} and {@code c}.
595 * If there is no byte {@code c} (because
596 * byte {@code a} was one of the last
597 * two of the bytes to be read), or either
598 * byte {@code b} or byte {@code c}
599 * does not match the bit pattern {@code 10xxxxxx},
600 * then a {@code UTFDataFormatException}
601 * is thrown. Otherwise, the group is converted
602 * to the character:<p>
603 * <pre><code>
604 * (char)(((a & 0x0F) << 12) | ((b & 0x3F) << 6) | (c & 0x3F))
605 * </code></pre>
606 * If the first byte of a group matches the
607 * pattern {@code 1111xxxx} or the pattern
608 * {@code 10xxxxxx}, then a {@code UTFDataFormatException}
609 * is thrown.
610 * <p>
611 * If end of file is encountered
612 * at any time during this entire process,
613 * then an {@code EOFException} is thrown.
614 * <p>
615 * After every group has been converted to
616 * a character by this process, the characters
617 * are gathered, in the same order in which
618 * their corresponding groups were read from
619 * the input stream, to form a {@code String},
620 * which is returned.
621 * <p>
622 * The {@code writeUTF}
623 * method of interface {@code DataOutput}
624 * may be used to write data that is suitable
625 * for reading by this method.
626 * @return a Unicode string.
627 * @exception EOFException if this stream reaches the end
628 * before reading all the bytes.
629 * @exception IOException if an I/O error occurs.
630 * @exception UTFDataFormatException if the bytes do not represent a
631 * valid modified UTF-8 encoding of a string.
632 */
633 String readUTF() throws IOException;
634 }