1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * http://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17
18 package com.zone.weixin4j.base64;
19
20 import com.zone.weixin4j.util.ServerToolkits;
21
22 import java.math.BigInteger;
23
24 /**
25 * <p>
26 * <font color="red">reference of apache pivot</font>
27 * </p>
28 *
29 * Provides Base64 encoding and decoding as defined by <a
30 * href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>.
31 *
32 * <p>
33 * This class implements section <cite>6.8. Base64
34 * Content-Transfer-Encoding</cite> from RFC 2045 <cite>Multipurpose Internet
35 * Mail Extensions (MIME) Part One: Format of Internet Message Bodies</cite> by
36 * Freed and Borenstein.
37 * </p>
38 * <p>
39 * The class can be parameterized in the following manner with various
40 * constructors:
41 * <ul>
42 * <li>URL-safe mode: Default off.</li>
43 * <li>Line length: Default 76. Line length that aren't multiples of 4 will
44 * still essentially end up being multiples of 4 in the encoded data.
45 * <li>Line separator: Default is CRLF ("\r\n")</li>
46 * </ul>
47 * </p>
48 * <p>
49 * Since this class operates directly on byte streams, and not character
50 * streams, it is hard-coded to only encode/decode character encodings which are
51 * compatible with the lower 127 ASCII chart (ISO-8859-1, Windows-1252, UTF-8,
52 * etc).
53 * </p>
54 * <p>
55 * This class is thread-safe.
56 * </p>
57 *
58 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>
59 * @since 1.0
60 * @version $Id: Base64.java 1447577 2013-02-19 02:45:18Z julius $
61 */
62 public class Base64 extends BaseNCodec {
63
64 /**
65 * BASE32 characters are 6 bits in length. They are formed by taking a block
66 * of 3 octets to form a 24-bit string, which is converted into 4 BASE64
67 * characters.
68 */
69 private static final int BITS_PER_ENCODED_BYTE = 6;
70 private static final int BYTES_PER_UNENCODED_BLOCK = 3;
71 private static final int BYTES_PER_ENCODED_BLOCK = 4;
72
73 /**
74 * Chunk separator per RFC 2045 section 2.1.
75 *
76 * <p>
77 * N.B. The next major release may break compatibility and make this field
78 * private.
79 * </p>
80 *
81 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section
82 * 2.1</a>
83 */
84 static final byte[] CHUNK_SEPARATOR = { '\r', '\n' };
85
86 /**
87 * This array is a lookup table that translates 6-bit positive integer index
88 * values into their "Base64 Alphabet" equivalents as specified in Table 1
89 * of RFC 2045.
90 *
91 * Thanks to "commons" project in ws.apache.org for this code.
92 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
93 */
94 private static final byte[] STANDARD_ENCODE_TABLE = { 'A', 'B', 'C', 'D',
95 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q',
96 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd',
97 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q',
98 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', '0', '1', '2', '3',
99 '4', '5', '6', '7', '8', '9', '+', '/' };
100
101 /**
102 * This is a copy of the STANDARD_ENCODE_TABLE above, but with + and /
103 * changed to - and _ to make the encoded Base64 results more URL-SAFE. This
104 * table is only used when the Base64's mode is set to URL-SAFE.
105 */
106 private static final byte[] URL_SAFE_ENCODE_TABLE = { 'A', 'B', 'C', 'D',
107 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q',
108 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd',
109 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q',
110 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', '0', '1', '2', '3',
111 '4', '5', '6', '7', '8', '9', '-', '_' };
112
113 /**
114 * This array is a lookup table that translates Unicode characters drawn
115 * from the "Base64 Alphabet" (as specified in Table 1 of RFC 2045) into
116 * their 6-bit positive integer equivalents. Characters that are not in the
117 * Base64 alphabet but fall within the bounds of the array are translated to
118 * -1.
119 *
120 * Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This
121 * means decoder seamlessly handles both URL_SAFE and STANDARD base64. (The
122 * encoder, on the other hand, needs to know ahead of time what to emit).
123 *
124 * Thanks to "commons" project in ws.apache.org for this code.
125 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
126 */
127 private static final byte[] DECODE_TABLE = { -1, -1, -1, -1, -1, -1, -1,
128 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
129 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
130 -1, -1, 62, -1, 62, -1, 63, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61,
131 -1, -1, -1, -1, -1, -1, -1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
132 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1,
133 -1, 63, -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39,
134 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 };
135
136 /**
137 * Base64 uses 6-bit fields.
138 */
139 /** Mask used to extract 6 bits, used when encoding */
140 private static final int MASK_6BITS = 0x3f;
141
142 // The static final fields above are used for the original static byte[]
143 // methods on Base64.
144 // The private member fields below are used with the new streaming approach,
145 // which requires
146 // some state be preserved between calls of encode() and decode().
147
148 /**
149 * Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE
150 * above remains static because it is able to decode both STANDARD and
151 * URL_SAFE streams, but the encodeTable must be a member variable so we can
152 * switch between the two modes.
153 */
154 private final byte[] encodeTable;
155
156 // Only one decode table currently; keep for consistency with Base32 code
157 private final byte[] decodeTable = DECODE_TABLE;
158
159 /**
160 * Line separator for encoding. Not used when decoding. Only used if
161 * lineLength > 0.
162 */
163 private final byte[] lineSeparator;
164
165 /**
166 * Convenience variable to help us determine when our buffer is going to run
167 * out of room and needs resizing.
168 * <code>decodeSize = 3 + lineSeparator.length;</code>
169 */
170 private final int decodeSize;
171
172 /**
173 * Convenience variable to help us determine when our buffer is going to run
174 * out of room and needs resizing.
175 * <code>encodeSize = 4 + lineSeparator.length;</code>
176 */
177 private final int encodeSize;
178
179 /**
180 * Creates a Base64 codec used for decoding (all modes) and encoding in
181 * URL-unsafe mode.
182 * <p>
183 * When encoding the line length is 0 (no chunking), and the encoding table
184 * is STANDARD_ENCODE_TABLE.
185 * </p>
186 *
187 * <p>
188 * When decoding all variants are supported.
189 * </p>
190 */
191 public Base64() {
192 this(0);
193 }
194
195 /**
196 * Creates a Base64 codec used for decoding (all modes) and encoding in the
197 * given URL-safe mode.
198 * <p>
199 * When encoding the line length is 76, the line separator is CRLF, and the
200 * encoding table is STANDARD_ENCODE_TABLE.
201 * </p>
202 *
203 * <p>
204 * When decoding all variants are supported.
205 * </p>
206 *
207 * @param urlSafe
208 * if {@code true}, URL-safe encoding is used. In most cases this
209 * should be set to {@code false}.
210 * @since 1.4
211 */
212 public Base64(final boolean urlSafe) {
213 this(MIME_CHUNK_SIZE, CHUNK_SEPARATOR, urlSafe);
214 }
215
216 /**
217 * Creates a Base64 codec used for decoding (all modes) and encoding in
218 * URL-unsafe mode.
219 * <p>
220 * When encoding the line length is given in the constructor, the line
221 * separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE.
222 * </p>
223 * <p>
224 * Line lengths that aren't multiples of 4 will still essentially end up
225 * being multiples of 4 in the encoded data.
226 * </p>
227 * <p>
228 * When decoding all variants are supported.
229 * </p>
230 *
231 * @param lineLength
232 * Each line of encoded data will be at most of the given length
233 * (rounded down to nearest multiple of 4). If lineLength <= 0,
234 * then the output will not be divided into lines (chunks).
235 * Ignored when decoding.
236 * @since 1.4
237 */
238 public Base64(final int lineLength) {
239 this(lineLength, CHUNK_SEPARATOR);
240 }
241
242 /**
243 * Creates a Base64 codec used for decoding (all modes) and encoding in
244 * URL-unsafe mode.
245 * <p>
246 * When encoding the line length and line separator are given in the
247 * constructor, and the encoding table is STANDARD_ENCODE_TABLE.
248 * </p>
249 * <p>
250 * Line lengths that aren't multiples of 4 will still essentially end up
251 * being multiples of 4 in the encoded data.
252 * </p>
253 * <p>
254 * When decoding all variants are supported.
255 * </p>
256 *
257 * @param lineLength
258 * Each line of encoded data will be at most of the given length
259 * (rounded down to nearest multiple of 4). If lineLength <= 0,
260 * then the output will not be divided into lines (chunks).
261 * Ignored when decoding.
262 * @param lineSeparator
263 * Each line of encoded data will end with this sequence of
264 * bytes.
265 * @throws IllegalArgumentException
266 * Thrown when the provided lineSeparator included some base64
267 * characters.
268 * @since 1.4
269 */
270 public Base64(final int lineLength, final byte[] lineSeparator) {
271 this(lineLength, lineSeparator, false);
272 }
273
274 /**
275 * Creates a Base64 codec used for decoding (all modes) and encoding in
276 * URL-unsafe mode.
277 * <p>
278 * When encoding the line length and line separator are given in the
279 * constructor, and the encoding table is STANDARD_ENCODE_TABLE.
280 * </p>
281 * <p>
282 * Line lengths that aren't multiples of 4 will still essentially end up
283 * being multiples of 4 in the encoded data.
284 * </p>
285 * <p>
286 * When decoding all variants are supported.
287 * </p>
288 *
289 * @param lineLength
290 * Each line of encoded data will be at most of the given length
291 * (rounded down to nearest multiple of 4). If lineLength <= 0,
292 * then the output will not be divided into lines (chunks).
293 * Ignored when decoding.
294 * @param lineSeparator
295 * Each line of encoded data will end with this sequence of
296 * bytes.
297 * @param urlSafe
298 * Instead of emitting '+' and '/' we emit '-' and '_'
299 * respectively. urlSafe is only applied to encode operations.
300 * Decoding seamlessly handles both modes. <b>Note: no padding is
301 * added when using the URL-safe alphabet.</b>
302 * @throws IllegalArgumentException
303 * The provided lineSeparator included some base64 characters.
304 * That's not going to work!
305 * @since 1.4
306 */
307 public Base64(final int lineLength, final byte[] lineSeparator,
308 final boolean urlSafe) {
309 super(BYTES_PER_UNENCODED_BLOCK, BYTES_PER_ENCODED_BLOCK, lineLength,
310 lineSeparator == null ? 0 : lineSeparator.length);
311 // TODO could be simplified if there is no requirement to reject invalid
312 // line sep when length <=0
313 // @see test case Base64Test.testConstructors()
314 if (lineSeparator != null) {
315 if (containsAlphabetOrPad(lineSeparator)) {
316 final String sep = ServerToolkits.newStringUtf8(lineSeparator);
317 throw new IllegalArgumentException(
318 "lineSeparator must not contain base64 characters: ["
319 + sep + "]");
320 }
321 if (lineLength > 0) { // null line-sep forces no chunking rather
322 // than throwing IAE
323 this.encodeSize = BYTES_PER_ENCODED_BLOCK
324 + lineSeparator.length;
325 this.lineSeparator = new byte[lineSeparator.length];
326 System.arraycopy(lineSeparator, 0, this.lineSeparator, 0,
327 lineSeparator.length);
328 } else {
329 this.encodeSize = BYTES_PER_ENCODED_BLOCK;
330 this.lineSeparator = null;
331 }
332 } else {
333 this.encodeSize = BYTES_PER_ENCODED_BLOCK;
334 this.lineSeparator = null;
335 }
336 this.decodeSize = this.encodeSize - 1;
337 this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE
338 : STANDARD_ENCODE_TABLE;
339 }
340
341 /**
342 * Returns our current encode mode. True if we're URL-SAFE, false otherwise.
343 *
344 * @return true if we're in URL-SAFE mode, false otherwise.
345 * @since 1.4
346 */
347 public boolean isUrlSafe() {
348 return this.encodeTable == URL_SAFE_ENCODE_TABLE;
349 }
350
351 /**
352 * <p>
353 * Encodes all of the provided data, starting at inPos, for inAvail bytes.
354 * Must be called at least twice: once with the data to encode, and once
355 * with inAvail set to "-1" to alert encoder that EOF has been reached, to
356 * flush last remaining bytes (if not multiple of 3).
357 * </p>
358 * <p>
359 * <b>Note: no padding is added when encoding using the URL-safe
360 * alphabet.</b>
361 * </p>
362 * <p>
363 * Thanks to "commons" project in ws.apache.org for the bitwise operations,
364 * and general approach.
365 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
366 * </p>
367 *
368 * @param in
369 * byte[] array of binary data to base64 encode.
370 * @param inPos
371 * Position to start reading data from.
372 * @param inAvail
373 * Amount of bytes available from input for encoding.
374 * @param context
375 * the context to be used
376 */
377 @Override
378 void encode(final byte[] in, int inPos, final int inAvail,
379 final Context context) {
380 if (context.eof) {
381 return;
382 }
383 // inAvail < 0 is how we're informed of EOF in the underlying data we're
384 // encoding.
385 if (inAvail < 0) {
386 context.eof = true;
387 if (0 == context.modulus && lineLength == 0) {
388 return; // no leftovers to process and not using chunking
389 }
390 final byte[] buffer = ensureBufferSize(encodeSize, context);
391 final int savedPos = context.pos;
392 switch (context.modulus) { // 0-2
393 case 0: // nothing to do here
394 break;
395 case 1: // 8 bits = 6 + 2
396 // top 6 bits:
397 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 2)
398 & MASK_6BITS];
399 // remaining 2:
400 buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 4)
401 & MASK_6BITS];
402 // URL-SAFE skips the padding to further reduce size.
403 if (encodeTable == STANDARD_ENCODE_TABLE) {
404 buffer[context.pos++] = PAD;
405 buffer[context.pos++] = PAD;
406 }
407 break;
408
409 case 2: // 16 bits = 6 + 6 + 4
410 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 10)
411 & MASK_6BITS];
412 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 4)
413 & MASK_6BITS];
414 buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 2)
415 & MASK_6BITS];
416 // URL-SAFE skips the padding to further reduce size.
417 if (encodeTable == STANDARD_ENCODE_TABLE) {
418 buffer[context.pos++] = PAD;
419 }
420 break;
421 default:
422 throw new IllegalStateException("Impossible modulus "
423 + context.modulus);
424 }
425 context.currentLinePos += context.pos - savedPos; // keep track of
426 // current line
427 // position
428 // if currentPos == 0 we are at the start of a line, so don't add
429 // CRLF
430 if (lineLength > 0 && context.currentLinePos > 0) {
431 System.arraycopy(lineSeparator, 0, buffer, context.pos,
432 lineSeparator.length);
433 context.pos += lineSeparator.length;
434 }
435 } else {
436 for (int i = 0; i < inAvail; i++) {
437 final byte[] buffer = ensureBufferSize(encodeSize, context);
438 context.modulus = (context.modulus + 1)
439 % BYTES_PER_UNENCODED_BLOCK;
440 int b = in[inPos++];
441 if (b < 0) {
442 b += 256;
443 }
444 context.ibitWorkArea = (context.ibitWorkArea << 8) + b; // BITS_PER_BYTE
445 if (0 == context.modulus) { // 3 bytes = 24 bits = 4 * 6 bits to
446 // extract
447 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 18)
448 & MASK_6BITS];
449 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 12)
450 & MASK_6BITS];
451 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 6)
452 & MASK_6BITS];
453 buffer[context.pos++] = encodeTable[context.ibitWorkArea
454 & MASK_6BITS];
455 context.currentLinePos += BYTES_PER_ENCODED_BLOCK;
456 if (lineLength > 0 && lineLength <= context.currentLinePos) {
457 System.arraycopy(lineSeparator, 0, buffer, context.pos,
458 lineSeparator.length);
459 context.pos += lineSeparator.length;
460 context.currentLinePos = 0;
461 }
462 }
463 }
464 }
465 }
466
467 /**
468 * <p>
469 * Decodes all of the provided data, starting at inPos, for inAvail bytes.
470 * Should be called at least twice: once with the data to decode, and once
471 * with inAvail set to "-1" to alert decoder that EOF has been reached. The
472 * "-1" call is not necessary when decoding, but it doesn't hurt, either.
473 * </p>
474 * <p>
475 * Ignores all non-base64 characters. This is how chunked (e.g. 76
476 * character) data is handled, since CR and LF are silently ignored, but has
477 * implications for other bytes, too. This method subscribes to the
478 * garbage-in, garbage-out philosophy: it will not check the provided data
479 * for validity.
480 * </p>
481 * <p>
482 * Thanks to "commons" project in ws.apache.org for the bitwise operations,
483 * and general approach.
484 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
485 * </p>
486 *
487 * @param in
488 * byte[] array of ascii data to base64 decode.
489 * @param inPos
490 * Position to start reading data from.
491 * @param inAvail
492 * Amount of bytes available from input for encoding.
493 * @param context
494 * the context to be used
495 */
496 @Override
497 void decode(final byte[] in, int inPos, final int inAvail,
498 final Context context) {
499 if (context.eof) {
500 return;
501 }
502 if (inAvail < 0) {
503 context.eof = true;
504 }
505 for (int i = 0; i < inAvail; i++) {
506 final byte[] buffer = ensureBufferSize(decodeSize, context);
507 final byte b = in[inPos++];
508 if (b == PAD) {
509 // We're done.
510 context.eof = true;
511 break;
512 } else {
513 if (b >= 0 && b < DECODE_TABLE.length) {
514 final int result = DECODE_TABLE[b];
515 if (result >= 0) {
516 context.modulus = (context.modulus + 1)
517 % BYTES_PER_ENCODED_BLOCK;
518 context.ibitWorkArea = (context.ibitWorkArea << BITS_PER_ENCODED_BYTE)
519 + result;
520 if (context.modulus == 0) {
521 buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 16) & MASK_8BITS);
522 buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS);
523 buffer[context.pos++] = (byte) (context.ibitWorkArea & MASK_8BITS);
524 }
525 }
526 }
527 }
528 }
529
530 // Two forms of EOF as far as base64 decoder is concerned: actual
531 // EOF (-1) and first time '=' character is encountered in stream.
532 // This approach makes the '=' padding characters completely optional.
533 if (context.eof && context.modulus != 0) {
534 final byte[] buffer = ensureBufferSize(decodeSize, context);
535
536 // We have some spare bits remaining
537 // Output all whole multiples of 8 bits and ignore the rest
538 switch (context.modulus) {
539 // case 0 : // impossible, as excluded above
540 case 1: // 6 bits - ignore entirely
541 // TODO not currently tested; perhaps it is impossible?
542 break;
543 case 2: // 12 bits = 8 + 4
544 context.ibitWorkArea = context.ibitWorkArea >> 4; // dump the
545 // extra 4
546 // bits
547 buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS);
548 break;
549 case 3: // 18 bits = 8 + 8 + 2
550 context.ibitWorkArea = context.ibitWorkArea >> 2; // dump 2 bits
551 buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS);
552 buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS);
553 break;
554 default:
555 throw new IllegalStateException("Impossible modulus "
556 + context.modulus);
557 }
558 }
559 }
560
561 /**
562 * Tests a given byte array to see if it contains only valid characters
563 * within the Base64 alphabet. Currently the method treats whitespace as
564 * valid.
565 *
566 * @param arrayOctet
567 * byte array to test
568 * @return {@code true} if all bytes are valid characters in the Base64
569 * alphabet or if the byte array is empty; {@code false}, otherwise
570 * @deprecated 1.5 Use {@link #isBase64(byte[])}, will be removed in 2.0.
571 */
572 @Deprecated
573 public static boolean isArrayByteBase64(final byte[] arrayOctet) {
574 return isBase64(arrayOctet);
575 }
576
577 /**
578 * Returns whether or not the <code>octet</code> is in the base 64 alphabet.
579 *
580 * @param octet
581 * The value to test
582 * @return {@code true} if the value is defined in the the base 64 alphabet,
583 * {@code false} otherwise.
584 * @since 1.4
585 */
586 public static boolean isBase64(final byte octet) {
587 return octet == PAD_DEFAULT
588 || (octet >= 0 && octet < DECODE_TABLE.length && DECODE_TABLE[octet] != -1);
589 }
590
591 /**
592 * Tests a given String to see if it contains only valid characters within
593 * the Base64 alphabet. Currently the method treats whitespace as valid.
594 *
595 * @param base64
596 * String to test
597 * @return {@code true} if all characters in the String are valid characters
598 * in the Base64 alphabet or if the String is empty; {@code false},
599 * otherwise
600 * @since 1.5
601 */
602 public static boolean isBase64(final String base64) {
603 return isBase64(ServerToolkits.getBytesUtf8(base64));
604 }
605
606 /**
607 * Tests a given byte array to see if it contains only valid characters
608 * within the Base64 alphabet. Currently the method treats whitespace as
609 * valid.
610 *
611 * @param arrayOctet
612 * byte array to test
613 * @return {@code true} if all bytes are valid characters in the Base64
614 * alphabet or if the byte array is empty; {@code false}, otherwise
615 * @since 1.5
616 */
617 public static boolean isBase64(final byte[] arrayOctet) {
618 for (int i = 0; i < arrayOctet.length; i++) {
619 if (!isBase64(arrayOctet[i]) && !isWhiteSpace(arrayOctet[i])) {
620 return false;
621 }
622 }
623 return true;
624 }
625
626 /**
627 * Encodes binary data using the base64 algorithm but does not chunk the
628 * output.
629 *
630 * @param binaryData
631 * binary data to encode
632 * @return byte[] containing Base64 characters in their UTF-8
633 * representation.
634 */
635 public static byte[] encodeBase64(final byte[] binaryData) {
636 return encodeBase64(binaryData, false);
637 }
638
639 /**
640 * Encodes binary data using the base64 algorithm but does not chunk the
641 * output.
642 *
643 * NOTE: We changed the behaviour of this method from multi-line chunking
644 * (commons-codec-1.4) to single-line non-chunking (commons-codec-1.5).
645 *
646 * @param binaryData
647 * binary data to encode
648 * @return String containing Base64 characters.
649 * @since 1.4 (NOTE: 1.4 chunked the output, whereas 1.5 does not).
650 */
651 public static String encodeBase64String(final byte[] binaryData) {
652 return ServerToolkits.newStringUtf8(encodeBase64(binaryData, false));
653 }
654
655 /**
656 * Encodes binary data using a URL-safe variation of the base64 algorithm
657 * but does not chunk the output. The url-safe variation emits - and _
658 * instead of + and / characters. <b>Note: no padding is added.</b>
659 *
660 * @param binaryData
661 * binary data to encode
662 * @return byte[] containing Base64 characters in their UTF-8
663 * representation.
664 * @since 1.4
665 */
666 public static byte[] encodeBase64URLSafe(final byte[] binaryData) {
667 return encodeBase64(binaryData, false, true);
668 }
669
670 /**
671 * Encodes binary data using a URL-safe variation of the base64 algorithm
672 * but does not chunk the output. The url-safe variation emits - and _
673 * instead of + and / characters. <b>Note: no padding is added.</b>
674 *
675 * @param binaryData
676 * binary data to encode
677 * @return String containing Base64 characters
678 * @since 1.4
679 */
680 public static String encodeBase64URLSafeString(final byte[] binaryData) {
681 return ServerToolkits.newStringUtf8(encodeBase64(binaryData, false, true));
682 }
683
684 /**
685 * Encodes binary data using the base64 algorithm and chunks the encoded
686 * output into 76 character blocks
687 *
688 * @param binaryData
689 * binary data to encode
690 * @return Base64 characters chunked in 76 character blocks
691 */
692 public static byte[] encodeBase64Chunked(final byte[] binaryData) {
693 return encodeBase64(binaryData, true);
694 }
695
696 /**
697 * Encodes binary data using the base64 algorithm, optionally chunking the
698 * output into 76 character blocks.
699 *
700 * @param binaryData
701 * Array containing binary data to encode.
702 * @param isChunked
703 * if {@code true} this encoder will chunk the base64 output into
704 * 76 character blocks
705 * @return Base64-encoded data.
706 * @throws IllegalArgumentException
707 * Thrown when the input array needs an output array bigger than
708 * {@link Integer#MAX_VALUE}
709 */
710 public static byte[] encodeBase64(final byte[] binaryData,
711 final boolean isChunked) {
712 return encodeBase64(binaryData, isChunked, false);
713 }
714
715 /**
716 * Encodes binary data using the base64 algorithm, optionally chunking the
717 * output into 76 character blocks.
718 *
719 * @param binaryData
720 * Array containing binary data to encode.
721 * @param isChunked
722 * if {@code true} this encoder will chunk the base64 output into
723 * 76 character blocks
724 * @param urlSafe
725 * if {@code true} this encoder will emit - and _ instead of the
726 * usual + and / characters. <b>Note: no padding is added when
727 * encoding using the URL-safe alphabet.</b>
728 * @return Base64-encoded data.
729 * @throws IllegalArgumentException
730 * Thrown when the input array needs an output array bigger than
731 * {@link Integer#MAX_VALUE}
732 * @since 1.4
733 */
734 public static byte[] encodeBase64(final byte[] binaryData,
735 final boolean isChunked, final boolean urlSafe) {
736 return encodeBase64(binaryData, isChunked, urlSafe, Integer.MAX_VALUE);
737 }
738
739 /**
740 * Encodes binary data using the base64 algorithm, optionally chunking the
741 * output into 76 character blocks.
742 *
743 * @param binaryData
744 * Array containing binary data to encode.
745 * @param isChunked
746 * if {@code true} this encoder will chunk the base64 output into
747 * 76 character blocks
748 * @param urlSafe
749 * if {@code true} this encoder will emit - and _ instead of the
750 * usual + and / characters. <b>Note: no padding is added when
751 * encoding using the URL-safe alphabet.</b>
752 * @param maxResultSize
753 * The maximum result size to accept.
754 * @return Base64-encoded data.
755 * @throws IllegalArgumentException
756 * Thrown when the input array needs an output array bigger than
757 * maxResultSize
758 * @since 1.4
759 */
760 public static byte[] encodeBase64(final byte[] binaryData,
761 final boolean isChunked, final boolean urlSafe,
762 final int maxResultSize) {
763 if (binaryData == null || binaryData.length == 0) {
764 return binaryData;
765 }
766
767 // Create this so can use the super-class method
768 // Also ensures that the same roundings are performed by the ctor and
769 // the code
770 final Base64 b64 = isChunked ? new Base64(urlSafe) : new Base64(0,
771 CHUNK_SEPARATOR, urlSafe);
772 final long len = b64.getEncodedLength(binaryData);
773 if (len > maxResultSize) {
774 throw new IllegalArgumentException(
775 "Input array too big, the output array would be bigger ("
776 + len + ") than the specified maximum size of "
777 + maxResultSize);
778 }
779
780 return b64.encode(binaryData);
781 }
782
783 /**
784 * Decodes a Base64 String into octets
785 *
786 * @param base64String
787 * String containing Base64 data
788 * @return Array containing decoded data.
789 * @since 1.4
790 */
791 public static byte[] decodeBase64(final String base64String) {
792 return new Base64().decode(base64String);
793 }
794
795 /**
796 * Decodes Base64 data into octets
797 *
798 * @param base64Data
799 * Byte array containing Base64 data
800 * @return Array containing decoded data.
801 */
802 public static byte[] decodeBase64(final byte[] base64Data) {
803 return new Base64().decode(base64Data);
804 }
805
806 // Implementation of the Encoder Interface
807
808 // Implementation of integer encoding used for crypto
809 /**
810 * Decodes a byte64-encoded integer according to crypto standards such as
811 * W3C's XML-Signature
812 *
813 * @param pArray
814 * a byte array containing base64 character data
815 * @return A BigInteger
816 * @since 1.4
817 */
818 public static BigInteger decodeInteger(final byte[] pArray) {
819 return new BigInteger(1, decodeBase64(pArray));
820 }
821
822 /**
823 * Encodes to a byte64-encoded integer according to crypto standards such as
824 * W3C's XML-Signature
825 *
826 * @param bigInt
827 * a BigInteger
828 * @return A byte array containing base64 character data
829 * @throws NullPointerException
830 * if null is passed in
831 * @since 1.4
832 */
833 public static byte[] encodeInteger(final BigInteger bigInt) {
834 if (bigInt == null) {
835 throw new NullPointerException(
836 "encodeInteger called with null parameter");
837 }
838 return encodeBase64(toIntegerBytes(bigInt), false);
839 }
840
841 /**
842 * Returns a byte-array representation of a <code>BigInteger</code> without
843 * sign bit.
844 *
845 * @param bigInt
846 * <code>BigInteger</code> to be converted
847 * @return a byte array representation of the BigInteger parameter
848 */
849 static byte[] toIntegerBytes(final BigInteger bigInt) {
850 int bitlen = bigInt.bitLength();
851 // round bitlen
852 bitlen = ((bitlen + 7) >> 3) << 3;
853 final byte[] bigBytes = bigInt.toByteArray();
854
855 if (((bigInt.bitLength() % 8) != 0)
856 && (((bigInt.bitLength() / 8) + 1) == (bitlen / 8))) {
857 return bigBytes;
858 }
859 // set up params for copying everything but sign bit
860 int startSrc = 0;
861 int len = bigBytes.length;
862
863 // if bigInt is exactly byte-aligned, just skip signbit in copy
864 if ((bigInt.bitLength() % 8) == 0) {
865 startSrc = 1;
866 len--;
867 }
868 final int startDst = bitlen / 8 - len; // to pad w/ nulls as per spec
869 final byte[] resizedBytes = new byte[bitlen / 8];
870 System.arraycopy(bigBytes, startSrc, resizedBytes, startDst, len);
871 return resizedBytes;
872 }
873
874 /**
875 * Returns whether or not the <code>octet</code> is in the Base64 alphabet.
876 *
877 * @param octet
878 * The value to test
879 * @return {@code true} if the value is defined in the the Base64 alphabet
880 * {@code false} otherwise.
881 */
882 @Override
883 protected boolean isInAlphabet(final byte octet) {
884 return octet >= 0 && octet < decodeTable.length
885 && decodeTable[octet] != -1;
886 }
887
888 }