001 /*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements. See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License. You may obtain a copy of the License at
008 *
009 * http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017
018 package org.apache.commons.codec.binary;
019
020 import java.math.BigInteger;
021
022 /**
023 * Provides Base64 encoding and decoding as defined by <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>.
024 *
025 * <p>
026 * This class implements section <cite>6.8. Base64 Content-Transfer-Encoding</cite> from RFC 2045 <cite>Multipurpose
027 * Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</cite> by Freed and Borenstein.
028 * </p>
029 * <p>
030 * The class can be parameterized in the following manner with various constructors:
031 * <ul>
032 * <li>URL-safe mode: Default off.</li>
033 * <li>Line length: Default 76. Line length that aren't multiples of 4 will still essentially end up being multiples of
034 * 4 in the encoded data.
035 * <li>Line separator: Default is CRLF ("\r\n")</li>
036 * </ul>
037 * </p>
038 * <p>
039 * Since this class operates directly on byte streams, and not character streams, it is hard-coded to only encode/decode
040 * character encodings which are compatible with the lower 127 ASCII chart (ISO-8859-1, Windows-1252, UTF-8, etc).
041 * </p>
042 * <p>
043 * This class is not thread-safe. Each thread should use its own instance.
044 * </p>
045 *
046 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>
047 * @author Apache Software Foundation
048 * @since 1.0
049 * @version $Revision: 1080712 $
050 */
051 public class Base64 extends BaseNCodec {
052
053 /**
054 * BASE32 characters are 6 bits in length.
055 * They are formed by taking a block of 3 octets to form a 24-bit string,
056 * which is converted into 4 BASE64 characters.
057 */
058 private static final int BITS_PER_ENCODED_BYTE = 6;
059 private static final int BYTES_PER_UNENCODED_BLOCK = 3;
060 private static final int BYTES_PER_ENCODED_BLOCK = 4;
061
062 /**
063 * Chunk separator per RFC 2045 section 2.1.
064 *
065 * <p>
066 * N.B. The next major release may break compatibility and make this field private.
067 * </p>
068 *
069 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a>
070 */
071 static final byte[] CHUNK_SEPARATOR = {'\r', '\n'};
072
073 /**
074 * This array is a lookup table that translates 6-bit positive integer index values into their "Base64 Alphabet"
075 * equivalents as specified in Table 1 of RFC 2045.
076 *
077 * Thanks to "commons" project in ws.apache.org for this code.
078 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
079 */
080 private static final byte[] STANDARD_ENCODE_TABLE = {
081 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M',
082 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z',
083 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm',
084 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
085 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/'
086 };
087
088 /**
089 * This is a copy of the STANDARD_ENCODE_TABLE above, but with + and /
090 * changed to - and _ to make the encoded Base64 results more URL-SAFE.
091 * This table is only used when the Base64's mode is set to URL-SAFE.
092 */
093 private static final byte[] URL_SAFE_ENCODE_TABLE = {
094 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M',
095 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z',
096 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm',
097 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
098 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_'
099 };
100
101 /**
102 * This array is a lookup table that translates Unicode characters drawn from the "Base64 Alphabet" (as specified in
103 * Table 1 of RFC 2045) into their 6-bit positive integer equivalents. Characters that are not in the Base64
104 * alphabet but fall within the bounds of the array are translated to -1.
105 *
106 * Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This means decoder seamlessly handles both
107 * URL_SAFE and STANDARD base64. (The encoder, on the other hand, needs to know ahead of time what to emit).
108 *
109 * Thanks to "commons" project in ws.apache.org for this code.
110 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
111 */
112 private static final byte[] DECODE_TABLE = {
113 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
114 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
115 -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, 62, -1, 63, 52, 53, 54,
116 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, -1, 0, 1, 2, 3, 4,
117 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23,
118 24, 25, -1, -1, -1, -1, 63, -1, 26, 27, 28, 29, 30, 31, 32, 33, 34,
119 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51
120 };
121
122 /**
123 * Base64 uses 6-bit fields.
124 */
125 /** Mask used to extract 6 bits, used when encoding */
126 private static final int MASK_6BITS = 0x3f;
127
128 // The static final fields above are used for the original static byte[] methods on Base64.
129 // The private member fields below are used with the new streaming approach, which requires
130 // some state be preserved between calls of encode() and decode().
131
132 /**
133 * Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE above remains static because it is able
134 * to decode both STANDARD and URL_SAFE streams, but the encodeTable must be a member variable so we can switch
135 * between the two modes.
136 */
137 private final byte[] encodeTable;
138
139 // Only one decode table currently; keep for consistency with Base32 code
140 private final byte[] decodeTable = DECODE_TABLE;
141
142 /**
143 * Line separator for encoding. Not used when decoding. Only used if lineLength > 0.
144 */
145 private final byte[] lineSeparator;
146
147 /**
148 * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing.
149 * <code>decodeSize = 3 + lineSeparator.length;</code>
150 */
151 private final int decodeSize;
152
153 /**
154 * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing.
155 * <code>encodeSize = 4 + lineSeparator.length;</code>
156 */
157 private final int encodeSize;
158
159 /**
160 * Place holder for the bytes we're dealing with for our based logic.
161 * Bitwise operations store and extract the encoding or decoding from this variable.
162 */
163 private int bitWorkArea;
164
165 /**
166 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
167 * <p>
168 * When encoding the line length is 0 (no chunking), and the encoding table is STANDARD_ENCODE_TABLE.
169 * </p>
170 *
171 * <p>
172 * When decoding all variants are supported.
173 * </p>
174 */
175 public Base64() {
176 this(0);
177 }
178
179 /**
180 * Creates a Base64 codec used for decoding (all modes) and encoding in the given URL-safe mode.
181 * <p>
182 * When encoding the line length is 76, the line separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE.
183 * </p>
184 *
185 * <p>
186 * When decoding all variants are supported.
187 * </p>
188 *
189 * @param urlSafe
190 * if <code>true</code>, URL-safe encoding is used. In most cases this should be set to
191 * <code>false</code>.
192 * @since 1.4
193 */
194 public Base64(boolean urlSafe) {
195 this(MIME_CHUNK_SIZE, CHUNK_SEPARATOR, urlSafe);
196 }
197
198 /**
199 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
200 * <p>
201 * When encoding the line length is given in the constructor, the line separator is CRLF, and the encoding table is
202 * STANDARD_ENCODE_TABLE.
203 * </p>
204 * <p>
205 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
206 * </p>
207 * <p>
208 * When decoding all variants are supported.
209 * </p>
210 *
211 * @param lineLength
212 * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of 4).
213 * If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when decoding.
214 * @since 1.4
215 */
216 public Base64(int lineLength) {
217 this(lineLength, CHUNK_SEPARATOR);
218 }
219
220 /**
221 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
222 * <p>
223 * When encoding the line length and line separator are given in the constructor, and the encoding table is
224 * STANDARD_ENCODE_TABLE.
225 * </p>
226 * <p>
227 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
228 * </p>
229 * <p>
230 * When decoding all variants are supported.
231 * </p>
232 *
233 * @param lineLength
234 * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of 4).
235 * If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when decoding.
236 * @param lineSeparator
237 * Each line of encoded data will end with this sequence of bytes.
238 * @throws IllegalArgumentException
239 * Thrown when the provided lineSeparator included some base64 characters.
240 * @since 1.4
241 */
242 public Base64(int lineLength, byte[] lineSeparator) {
243 this(lineLength, lineSeparator, false);
244 }
245
246 /**
247 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
248 * <p>
249 * When encoding the line length and line separator are given in the constructor, and the encoding table is
250 * STANDARD_ENCODE_TABLE.
251 * </p>
252 * <p>
253 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
254 * </p>
255 * <p>
256 * When decoding all variants are supported.
257 * </p>
258 *
259 * @param lineLength
260 * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of 4).
261 * If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when decoding.
262 * @param lineSeparator
263 * Each line of encoded data will end with this sequence of bytes.
264 * @param urlSafe
265 * Instead of emitting '+' and '/' we emit '-' and '_' respectively. urlSafe is only applied to encode
266 * operations. Decoding seamlessly handles both modes.
267 * @throws IllegalArgumentException
268 * The provided lineSeparator included some base64 characters. That's not going to work!
269 * @since 1.4
270 */
271 public Base64(int lineLength, byte[] lineSeparator, boolean urlSafe) {
272 super(BYTES_PER_UNENCODED_BLOCK, BYTES_PER_ENCODED_BLOCK,
273 lineLength,
274 lineSeparator == null ? 0 : lineSeparator.length);
275 // TODO could be simplified if there is no requirement to reject invalid line sep when length <=0
276 // @see test case Base64Test.testConstructors()
277 if (lineSeparator != null) {
278 if (containsAlphabetOrPad(lineSeparator)) {
279 String sep = StringUtils.newStringUtf8(lineSeparator);
280 throw new IllegalArgumentException("lineSeparator must not contain base64 characters: [" + sep + "]");
281 }
282 if (lineLength > 0){ // null line-sep forces no chunking rather than throwing IAE
283 this.encodeSize = BYTES_PER_ENCODED_BLOCK + lineSeparator.length;
284 this.lineSeparator = new byte[lineSeparator.length];
285 System.arraycopy(lineSeparator, 0, this.lineSeparator, 0, lineSeparator.length);
286 } else {
287 this.encodeSize = BYTES_PER_ENCODED_BLOCK;
288 this.lineSeparator = null;
289 }
290 } else {
291 this.encodeSize = BYTES_PER_ENCODED_BLOCK;
292 this.lineSeparator = null;
293 }
294 this.decodeSize = this.encodeSize - 1;
295 this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE : STANDARD_ENCODE_TABLE;
296 }
297
298 /**
299 * Returns our current encode mode. True if we're URL-SAFE, false otherwise.
300 *
301 * @return true if we're in URL-SAFE mode, false otherwise.
302 * @since 1.4
303 */
304 public boolean isUrlSafe() {
305 return this.encodeTable == URL_SAFE_ENCODE_TABLE;
306 }
307
308 /**
309 * <p>
310 * Encodes all of the provided data, starting at inPos, for inAvail bytes. Must be called at least twice: once with
311 * the data to encode, and once with inAvail set to "-1" to alert encoder that EOF has been reached, so flush last
312 * remaining bytes (if not multiple of 3).
313 * </p>
314 * <p>
315 * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach.
316 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
317 * </p>
318 *
319 * @param in
320 * byte[] array of binary data to base64 encode.
321 * @param inPos
322 * Position to start reading data from.
323 * @param inAvail
324 * Amount of bytes available from input for encoding.
325 */
326 void encode(byte[] in, int inPos, int inAvail) {
327 if (eof) {
328 return;
329 }
330 // inAvail < 0 is how we're informed of EOF in the underlying data we're
331 // encoding.
332 if (inAvail < 0) {
333 eof = true;
334 if (0 == modulus && lineLength == 0) {
335 return; // no leftovers to process and not using chunking
336 }
337 ensureBufferSize(encodeSize);
338 int savedPos = pos;
339 switch (modulus) { // 0-2
340 case 1 : // 8 bits = 6 + 2
341 buffer[pos++] = encodeTable[(bitWorkArea >> 2) & MASK_6BITS]; // top 6 bits
342 buffer[pos++] = encodeTable[(bitWorkArea << 4) & MASK_6BITS]; // remaining 2
343 // URL-SAFE skips the padding to further reduce size.
344 if (encodeTable == STANDARD_ENCODE_TABLE) {
345 buffer[pos++] = PAD;
346 buffer[pos++] = PAD;
347 }
348 break;
349
350 case 2 : // 16 bits = 6 + 6 + 4
351 buffer[pos++] = encodeTable[(bitWorkArea >> 10) & MASK_6BITS];
352 buffer[pos++] = encodeTable[(bitWorkArea >> 4) & MASK_6BITS];
353 buffer[pos++] = encodeTable[(bitWorkArea << 2) & MASK_6BITS];
354 // URL-SAFE skips the padding to further reduce size.
355 if (encodeTable == STANDARD_ENCODE_TABLE) {
356 buffer[pos++] = PAD;
357 }
358 break;
359 }
360 currentLinePos += pos - savedPos; // keep track of current line position
361 // if currentPos == 0 we are at the start of a line, so don't add CRLF
362 if (lineLength > 0 && currentLinePos > 0) {
363 System.arraycopy(lineSeparator, 0, buffer, pos, lineSeparator.length);
364 pos += lineSeparator.length;
365 }
366 } else {
367 for (int i = 0; i < inAvail; i++) {
368 ensureBufferSize(encodeSize);
369 modulus = (modulus+1) % BYTES_PER_UNENCODED_BLOCK;
370 int b = in[inPos++];
371 if (b < 0) {
372 b += 256;
373 }
374 bitWorkArea = (bitWorkArea << 8) + b; // BITS_PER_BYTE
375 if (0 == modulus) { // 3 bytes = 24 bits = 4 * 6 bits to extract
376 buffer[pos++] = encodeTable[(bitWorkArea >> 18) & MASK_6BITS];
377 buffer[pos++] = encodeTable[(bitWorkArea >> 12) & MASK_6BITS];
378 buffer[pos++] = encodeTable[(bitWorkArea >> 6) & MASK_6BITS];
379 buffer[pos++] = encodeTable[bitWorkArea & MASK_6BITS];
380 currentLinePos += BYTES_PER_ENCODED_BLOCK;
381 if (lineLength > 0 && lineLength <= currentLinePos) {
382 System.arraycopy(lineSeparator, 0, buffer, pos, lineSeparator.length);
383 pos += lineSeparator.length;
384 currentLinePos = 0;
385 }
386 }
387 }
388 }
389 }
390
391 /**
392 * <p>
393 * Decodes all of the provided data, starting at inPos, for inAvail bytes. Should be called at least twice: once
394 * with the data to decode, and once with inAvail set to "-1" to alert decoder that EOF has been reached. The "-1"
395 * call is not necessary when decoding, but it doesn't hurt, either.
396 * </p>
397 * <p>
398 * Ignores all non-base64 characters. This is how chunked (e.g. 76 character) data is handled, since CR and LF are
399 * silently ignored, but has implications for other bytes, too. This method subscribes to the garbage-in,
400 * garbage-out philosophy: it will not check the provided data for validity.
401 * </p>
402 * <p>
403 * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach.
404 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
405 * </p>
406 *
407 * @param in
408 * byte[] array of ascii data to base64 decode.
409 * @param inPos
410 * Position to start reading data from.
411 * @param inAvail
412 * Amount of bytes available from input for encoding.
413 */
414 void decode(byte[] in, int inPos, int inAvail) {
415 if (eof) {
416 return;
417 }
418 if (inAvail < 0) {
419 eof = true;
420 }
421 for (int i = 0; i < inAvail; i++) {
422 ensureBufferSize(decodeSize);
423 byte b = in[inPos++];
424 if (b == PAD) {
425 // We're done.
426 eof = true;
427 break;
428 } else {
429 if (b >= 0 && b < DECODE_TABLE.length) {
430 int result = DECODE_TABLE[b];
431 if (result >= 0) {
432 modulus = (modulus+1) % BYTES_PER_ENCODED_BLOCK;
433 bitWorkArea = (bitWorkArea << BITS_PER_ENCODED_BYTE) + result;
434 if (modulus == 0) {
435 buffer[pos++] = (byte) ((bitWorkArea >> 16) & MASK_8BITS);
436 buffer[pos++] = (byte) ((bitWorkArea >> 8) & MASK_8BITS);
437 buffer[pos++] = (byte) (bitWorkArea & MASK_8BITS);
438 }
439 }
440 }
441 }
442 }
443
444 // Two forms of EOF as far as base64 decoder is concerned: actual
445 // EOF (-1) and first time '=' character is encountered in stream.
446 // This approach makes the '=' padding characters completely optional.
447 if (eof && modulus != 0) {
448 ensureBufferSize(decodeSize);
449
450 // We have some spare bits remaining
451 // Output all whole multiples of 8 bits and ignore the rest
452 switch (modulus) {
453 // case 1: // 6 bits - ignore entirely
454 // break;
455 case 2 : // 12 bits = 8 + 4
456 bitWorkArea = bitWorkArea >> 4; // dump the extra 4 bits
457 buffer[pos++] = (byte) ((bitWorkArea) & MASK_8BITS);
458 break;
459 case 3 : // 18 bits = 8 + 8 + 2
460 bitWorkArea = bitWorkArea >> 2; // dump 2 bits
461 buffer[pos++] = (byte) ((bitWorkArea >> 8) & MASK_8BITS);
462 buffer[pos++] = (byte) ((bitWorkArea) & MASK_8BITS);
463 break;
464 }
465 }
466 }
467
468 /**
469 * Returns whether or not the <code>octet</code> is in the base 64 alphabet.
470 *
471 * @param octet
472 * The value to test
473 * @return <code>true</code> if the value is defined in the the base 64 alphabet, <code>false</code> otherwise.
474 * @since 1.4
475 */
476 public static boolean isBase64(byte octet) {
477 return octet == PAD_DEFAULT || (octet >= 0 && octet < DECODE_TABLE.length && DECODE_TABLE[octet] != -1);
478 }
479
480 /**
481 * Tests a given String to see if it contains only valid characters within the Base64 alphabet. Currently the
482 * method treats whitespace as valid.
483 *
484 * @param base64
485 * String to test
486 * @return <code>true</code> if all characters in the String are valid characters in the Base64 alphabet or if
487 * the String is empty; <code>false</code>, otherwise
488 * @since 1.5
489 */
490 public static boolean isBase64(String base64) {
491 return isBase64(StringUtils.getBytesUtf8(base64));
492 }
493
494 /**
495 * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently the
496 * method treats whitespace as valid.
497 *
498 * @param arrayOctet
499 * byte array to test
500 * @return <code>true</code> if all bytes are valid characters in the Base64 alphabet or if the byte array is empty;
501 * <code>false</code>, otherwise
502 * @deprecated 1.5 Use {@link #isBase64(byte[])}, will be removed in 2.0.
503 */
504 public static boolean isArrayByteBase64(byte[] arrayOctet) {
505 return isBase64(arrayOctet);
506 }
507
508 /**
509 * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently the
510 * method treats whitespace as valid.
511 *
512 * @param arrayOctet
513 * byte array to test
514 * @return <code>true</code> if all bytes are valid characters in the Base64 alphabet or if the byte array is empty;
515 * <code>false</code>, otherwise
516 * @since 1.5
517 */
518 public static boolean isBase64(byte[] arrayOctet) {
519 for (int i = 0; i < arrayOctet.length; i++) {
520 if (!isBase64(arrayOctet[i]) && !isWhiteSpace(arrayOctet[i])) {
521 return false;
522 }
523 }
524 return true;
525 }
526
527 /**
528 * Encodes binary data using the base64 algorithm but does not chunk the output.
529 *
530 * @param binaryData
531 * binary data to encode
532 * @return byte[] containing Base64 characters in their UTF-8 representation.
533 */
534 public static byte[] encodeBase64(byte[] binaryData) {
535 return encodeBase64(binaryData, false);
536 }
537
538 /**
539 * Encodes binary data using the base64 algorithm but does not chunk the output.
540 *
541 * NOTE: We changed the behaviour of this method from multi-line chunking (commons-codec-1.4) to
542 * single-line non-chunking (commons-codec-1.5).
543 *
544 * @param binaryData
545 * binary data to encode
546 * @return String containing Base64 characters.
547 * @since 1.4 (NOTE: 1.4 chunked the output, whereas 1.5 does not).
548 */
549 public static String encodeBase64String(byte[] binaryData) {
550 return StringUtils.newStringUtf8(encodeBase64(binaryData, false));
551 }
552
553 /**
554 * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The
555 * url-safe variation emits - and _ instead of + and / characters.
556 *
557 * @param binaryData
558 * binary data to encode
559 * @return byte[] containing Base64 characters in their UTF-8 representation.
560 * @since 1.4
561 */
562 public static byte[] encodeBase64URLSafe(byte[] binaryData) {
563 return encodeBase64(binaryData, false, true);
564 }
565
566 /**
567 * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The
568 * url-safe variation emits - and _ instead of + and / characters.
569 *
570 * @param binaryData
571 * binary data to encode
572 * @return String containing Base64 characters
573 * @since 1.4
574 */
575 public static String encodeBase64URLSafeString(byte[] binaryData) {
576 return StringUtils.newStringUtf8(encodeBase64(binaryData, false, true));
577 }
578
579 /**
580 * Encodes binary data using the base64 algorithm and chunks the encoded output into 76 character blocks
581 *
582 * @param binaryData
583 * binary data to encode
584 * @return Base64 characters chunked in 76 character blocks
585 */
586 public static byte[] encodeBase64Chunked(byte[] binaryData) {
587 return encodeBase64(binaryData, true);
588 }
589
590 /**
591 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks.
592 *
593 * @param binaryData
594 * Array containing binary data to encode.
595 * @param isChunked
596 * if <code>true</code> this encoder will chunk the base64 output into 76 character blocks
597 * @return Base64-encoded data.
598 * @throws IllegalArgumentException
599 * Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE}
600 */
601 public static byte[] encodeBase64(byte[] binaryData, boolean isChunked) {
602 return encodeBase64(binaryData, isChunked, false);
603 }
604
605 /**
606 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks.
607 *
608 * @param binaryData
609 * Array containing binary data to encode.
610 * @param isChunked
611 * if <code>true</code> this encoder will chunk the base64 output into 76 character blocks
612 * @param urlSafe
613 * if <code>true</code> this encoder will emit - and _ instead of the usual + and / characters.
614 * @return Base64-encoded data.
615 * @throws IllegalArgumentException
616 * Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE}
617 * @since 1.4
618 */
619 public static byte[] encodeBase64(byte[] binaryData, boolean isChunked, boolean urlSafe) {
620 return encodeBase64(binaryData, isChunked, urlSafe, Integer.MAX_VALUE);
621 }
622
623 /**
624 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks.
625 *
626 * @param binaryData
627 * Array containing binary data to encode.
628 * @param isChunked
629 * if <code>true</code> this encoder will chunk the base64 output into 76 character blocks
630 * @param urlSafe
631 * if <code>true</code> this encoder will emit - and _ instead of the usual + and / characters.
632 * @param maxResultSize
633 * The maximum result size to accept.
634 * @return Base64-encoded data.
635 * @throws IllegalArgumentException
636 * Thrown when the input array needs an output array bigger than maxResultSize
637 * @since 1.4
638 */
639 public static byte[] encodeBase64(byte[] binaryData, boolean isChunked, boolean urlSafe, int maxResultSize) {
640 if (binaryData == null || binaryData.length == 0) {
641 return binaryData;
642 }
643
644 // Create this so can use the super-class method
645 // Also ensures that the same roundings are performed by the ctor and the code
646 Base64 b64 = isChunked ? new Base64(urlSafe) : new Base64(0, CHUNK_SEPARATOR, urlSafe);
647 long len = b64.getEncodedLength(binaryData);
648 if (len > maxResultSize) {
649 throw new IllegalArgumentException("Input array too big, the output array would be bigger (" +
650 len +
651 ") than the specified maximum size of " +
652 maxResultSize);
653 }
654
655 return b64.encode(binaryData);
656 }
657
658 /**
659 * Decodes a Base64 String into octets
660 *
661 * @param base64String
662 * String containing Base64 data
663 * @return Array containing decoded data.
664 * @since 1.4
665 */
666 public static byte[] decodeBase64(String base64String) {
667 return new Base64().decode(base64String);
668 }
669
670 /**
671 * Decodes Base64 data into octets
672 *
673 * @param base64Data
674 * Byte array containing Base64 data
675 * @return Array containing decoded data.
676 */
677 public static byte[] decodeBase64(byte[] base64Data) {
678 return new Base64().decode(base64Data);
679 }
680
681 // Implementation of the Encoder Interface
682
683 // Implementation of integer encoding used for crypto
684 /**
685 * Decodes a byte64-encoded integer according to crypto standards such as W3C's XML-Signature
686 *
687 * @param pArray
688 * a byte array containing base64 character data
689 * @return A BigInteger
690 * @since 1.4
691 */
692 public static BigInteger decodeInteger(byte[] pArray) {
693 return new BigInteger(1, decodeBase64(pArray));
694 }
695
696 /**
697 * Encodes to a byte64-encoded integer according to crypto standards such as W3C's XML-Signature
698 *
699 * @param bigInt
700 * a BigInteger
701 * @return A byte array containing base64 character data
702 * @throws NullPointerException
703 * if null is passed in
704 * @since 1.4
705 */
706 public static byte[] encodeInteger(BigInteger bigInt) {
707 if (bigInt == null) {
708 throw new NullPointerException("encodeInteger called with null parameter");
709 }
710 return encodeBase64(toIntegerBytes(bigInt), false);
711 }
712
713 /**
714 * Returns a byte-array representation of a <code>BigInteger</code> without sign bit.
715 *
716 * @param bigInt
717 * <code>BigInteger</code> to be converted
718 * @return a byte array representation of the BigInteger parameter
719 */
720 static byte[] toIntegerBytes(BigInteger bigInt) {
721 int bitlen = bigInt.bitLength();
722 // round bitlen
723 bitlen = ((bitlen + 7) >> 3) << 3;
724 byte[] bigBytes = bigInt.toByteArray();
725
726 if (((bigInt.bitLength() % 8) != 0) && (((bigInt.bitLength() / 8) + 1) == (bitlen / 8))) {
727 return bigBytes;
728 }
729 // set up params for copying everything but sign bit
730 int startSrc = 0;
731 int len = bigBytes.length;
732
733 // if bigInt is exactly byte-aligned, just skip signbit in copy
734 if ((bigInt.bitLength() % 8) == 0) {
735 startSrc = 1;
736 len--;
737 }
738 int startDst = bitlen / 8 - len; // to pad w/ nulls as per spec
739 byte[] resizedBytes = new byte[bitlen / 8];
740 System.arraycopy(bigBytes, startSrc, resizedBytes, startDst, len);
741 return resizedBytes;
742 }
743
744 /**
745 * Returns whether or not the <code>octet</code> is in the Base32 alphabet.
746 *
747 * @param octet
748 * The value to test
749 * @return <code>true</code> if the value is defined in the the Base32 alphabet <code>false</code> otherwise.
750 */
751 protected boolean isInAlphabet(byte octet) {
752 return octet >= 0 && octet < decodeTable.length && decodeTable[octet] != -1;
753 }
754
755 }