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 018package org.apache.commons.codec.binary; 019 020import 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 * </p> 032 * <ul> 033 * <li>URL-safe mode: Default off.</li> 034 * <li>Line length: Default 76. Line length that aren't multiples of 4 will still essentially end up being multiples of 035 * 4 in the encoded data. 036 * <li>Line separator: Default is CRLF ("\r\n")</li> 037 * </ul> 038 * <p> 039 * The URL-safe parameter is only applied to encode operations. Decoding seamlessly handles both modes. 040 * </p> 041 * <p> 042 * Since this class operates directly on byte streams, and not character streams, it is hard-coded to only 043 * encode/decode character encodings which are compatible with the lower 127 ASCII chart (ISO-8859-1, Windows-1252, 044 * UTF-8, etc). 045 * </p> 046 * <p> 047 * This class is thread-safe. 048 * </p> 049 * 050 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a> 051 * @since 1.0 052 * @version $Id: Base64.java 1789158 2017-03-28 15:04:58Z sebb $ 053 */ 054public class Base64 extends BaseNCodec { 055 056 /** 057 * BASE32 characters are 6 bits in length. 058 * They are formed by taking a block of 3 octets to form a 24-bit string, 059 * which is converted into 4 BASE64 characters. 060 */ 061 private static final int BITS_PER_ENCODED_BYTE = 6; 062 private static final int BYTES_PER_UNENCODED_BLOCK = 3; 063 private static final int BYTES_PER_ENCODED_BLOCK = 4; 064 065 /** 066 * Chunk separator per RFC 2045 section 2.1. 067 * 068 * <p> 069 * N.B. The next major release may break compatibility and make this field private. 070 * </p> 071 * 072 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a> 073 */ 074 static final byte[] CHUNK_SEPARATOR = {'\r', '\n'}; 075 076 /** 077 * This array is a lookup table that translates 6-bit positive integer index values into their "Base64 Alphabet" 078 * equivalents as specified in Table 1 of RFC 2045. 079 * 080 * Thanks to "commons" project in ws.apache.org for this code. 081 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 082 */ 083 private static final byte[] STANDARD_ENCODE_TABLE = { 084 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 085 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 086 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 087 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 088 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/' 089 }; 090 091 /** 092 * This is a copy of the STANDARD_ENCODE_TABLE above, but with + and / 093 * changed to - and _ to make the encoded Base64 results more URL-SAFE. 094 * This table is only used when the Base64's mode is set to URL-SAFE. 095 */ 096 private static final byte[] URL_SAFE_ENCODE_TABLE = { 097 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 098 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 099 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 100 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 101 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_' 102 }; 103 104 /** 105 * This array is a lookup table that translates Unicode characters drawn from the "Base64 Alphabet" (as specified 106 * in Table 1 of RFC 2045) into their 6-bit positive integer equivalents. Characters that are not in the Base64 107 * alphabet but fall within the bounds of the array are translated to -1. 108 * 109 * Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This means decoder seamlessly handles both 110 * URL_SAFE and STANDARD base64. (The encoder, on the other hand, needs to know ahead of time what to emit). 111 * 112 * Thanks to "commons" project in ws.apache.org for this code. 113 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 114 */ 115 private static final byte[] DECODE_TABLE = { 116 // 0 1 2 3 4 5 6 7 8 9 A B C D E F 117 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 00-0f 118 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 10-1f 119 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, 62, -1, 63, // 20-2f + - / 120 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, // 30-3f 0-9 121 -1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, // 40-4f A-O 122 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, 63, // 50-5f P-Z _ 123 -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, // 60-6f a-o 124 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 // 70-7a p-z 125 }; 126 127 /** 128 * Base64 uses 6-bit fields. 129 */ 130 /** Mask used to extract 6 bits, used when encoding */ 131 private static final int MASK_6BITS = 0x3f; 132 133 // The static final fields above are used for the original static byte[] methods on Base64. 134 // The private member fields below are used with the new streaming approach, which requires 135 // some state be preserved between calls of encode() and decode(). 136 137 /** 138 * Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE above remains static because it is able 139 * to decode both STANDARD and URL_SAFE streams, but the encodeTable must be a member variable so we can switch 140 * between the two modes. 141 */ 142 private final byte[] encodeTable; 143 144 // Only one decode table currently; keep for consistency with Base32 code 145 private final byte[] decodeTable = DECODE_TABLE; 146 147 /** 148 * Line separator for encoding. Not used when decoding. Only used if lineLength > 0. 149 */ 150 private final byte[] lineSeparator; 151 152 /** 153 * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing. 154 * <code>decodeSize = 3 + lineSeparator.length;</code> 155 */ 156 private final int decodeSize; 157 158 /** 159 * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing. 160 * <code>encodeSize = 4 + lineSeparator.length;</code> 161 */ 162 private final int encodeSize; 163 164 /** 165 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 166 * <p> 167 * When encoding the line length is 0 (no chunking), and the encoding table is STANDARD_ENCODE_TABLE. 168 * </p> 169 * 170 * <p> 171 * When decoding all variants are supported. 172 * </p> 173 */ 174 public Base64() { 175 this(0); 176 } 177 178 /** 179 * Creates a Base64 codec used for decoding (all modes) and encoding in the given URL-safe mode. 180 * <p> 181 * When encoding the line length is 76, the line separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE. 182 * </p> 183 * 184 * <p> 185 * When decoding all variants are supported. 186 * </p> 187 * 188 * @param urlSafe 189 * if <code>true</code>, URL-safe encoding is used. In most cases this should be set to 190 * <code>false</code>. 191 * @since 1.4 192 */ 193 public Base64(final boolean urlSafe) { 194 this(MIME_CHUNK_SIZE, CHUNK_SEPARATOR, urlSafe); 195 } 196 197 /** 198 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 199 * <p> 200 * When encoding the line length is given in the constructor, the line separator is CRLF, and the encoding table is 201 * STANDARD_ENCODE_TABLE. 202 * </p> 203 * <p> 204 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. 205 * </p> 206 * <p> 207 * When decoding all variants are supported. 208 * </p> 209 * 210 * @param lineLength 211 * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of 212 * 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when 213 * decoding. 214 * @since 1.4 215 */ 216 public Base64(final 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 235 * 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when 236 * decoding. 237 * @param lineSeparator 238 * Each line of encoded data will end with this sequence of bytes. 239 * @throws IllegalArgumentException 240 * Thrown when the provided lineSeparator included some base64 characters. 241 * @since 1.4 242 */ 243 public Base64(final int lineLength, final byte[] lineSeparator) { 244 this(lineLength, lineSeparator, false); 245 } 246 247 /** 248 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 249 * <p> 250 * When encoding the line length and line separator are given in the constructor, and the encoding table is 251 * STANDARD_ENCODE_TABLE. 252 * </p> 253 * <p> 254 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. 255 * </p> 256 * <p> 257 * When decoding all variants are supported. 258 * </p> 259 * 260 * @param lineLength 261 * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of 262 * 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when 263 * decoding. 264 * @param lineSeparator 265 * Each line of encoded data will end with this sequence of bytes. 266 * @param urlSafe 267 * Instead of emitting '+' and '/' we emit '-' and '_' respectively. urlSafe is only applied to encode 268 * operations. Decoding seamlessly handles both modes. 269 * <b>Note: no padding is added when using the URL-safe alphabet.</b> 270 * @throws IllegalArgumentException 271 * The provided lineSeparator included some base64 characters. That's not going to work! 272 * @since 1.4 273 */ 274 public Base64(final int lineLength, final byte[] lineSeparator, final boolean urlSafe) { 275 super(BYTES_PER_UNENCODED_BLOCK, BYTES_PER_ENCODED_BLOCK, 276 lineLength, 277 lineSeparator == null ? 0 : lineSeparator.length); 278 // TODO could be simplified if there is no requirement to reject invalid line sep when length <=0 279 // @see test case Base64Test.testConstructors() 280 if (lineSeparator != null) { 281 if (containsAlphabetOrPad(lineSeparator)) { 282 final String sep = StringUtils.newStringUtf8(lineSeparator); 283 throw new IllegalArgumentException("lineSeparator must not contain base64 characters: [" + sep + "]"); 284 } 285 if (lineLength > 0){ // null line-sep forces no chunking rather than throwing IAE 286 this.encodeSize = BYTES_PER_ENCODED_BLOCK + lineSeparator.length; 287 this.lineSeparator = new byte[lineSeparator.length]; 288 System.arraycopy(lineSeparator, 0, this.lineSeparator, 0, lineSeparator.length); 289 } else { 290 this.encodeSize = BYTES_PER_ENCODED_BLOCK; 291 this.lineSeparator = null; 292 } 293 } else { 294 this.encodeSize = BYTES_PER_ENCODED_BLOCK; 295 this.lineSeparator = null; 296 } 297 this.decodeSize = this.encodeSize - 1; 298 this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE : STANDARD_ENCODE_TABLE; 299 } 300 301 /** 302 * Returns our current encode mode. True if we're URL-SAFE, false otherwise. 303 * 304 * @return true if we're in URL-SAFE mode, false otherwise. 305 * @since 1.4 306 */ 307 public boolean isUrlSafe() { 308 return this.encodeTable == URL_SAFE_ENCODE_TABLE; 309 } 310 311 /** 312 * <p> 313 * Encodes all of the provided data, starting at inPos, for inAvail bytes. Must be called at least twice: once with 314 * the data to encode, and once with inAvail set to "-1" to alert encoder that EOF has been reached, to flush last 315 * remaining bytes (if not multiple of 3). 316 * </p> 317 * <p><b>Note: no padding is added when encoding using the URL-safe alphabet.</b></p> 318 * <p> 319 * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach. 320 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 321 * </p> 322 * 323 * @param in 324 * byte[] array of binary data to base64 encode. 325 * @param inPos 326 * Position to start reading data from. 327 * @param inAvail 328 * Amount of bytes available from input for encoding. 329 * @param context 330 * the context to be used 331 */ 332 @Override 333 void encode(final byte[] in, int inPos, final int inAvail, final Context context) { 334 if (context.eof) { 335 return; 336 } 337 // inAvail < 0 is how we're informed of EOF in the underlying data we're 338 // encoding. 339 if (inAvail < 0) { 340 context.eof = true; 341 if (0 == context.modulus && lineLength == 0) { 342 return; // no leftovers to process and not using chunking 343 } 344 final byte[] buffer = ensureBufferSize(encodeSize, context); 345 final int savedPos = context.pos; 346 switch (context.modulus) { // 0-2 347 case 0 : // nothing to do here 348 break; 349 case 1 : // 8 bits = 6 + 2 350 // top 6 bits: 351 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 2) & MASK_6BITS]; 352 // remaining 2: 353 buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 4) & MASK_6BITS]; 354 // URL-SAFE skips the padding to further reduce size. 355 if (encodeTable == STANDARD_ENCODE_TABLE) { 356 buffer[context.pos++] = pad; 357 buffer[context.pos++] = pad; 358 } 359 break; 360 361 case 2 : // 16 bits = 6 + 6 + 4 362 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 10) & MASK_6BITS]; 363 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 4) & MASK_6BITS]; 364 buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 2) & MASK_6BITS]; 365 // URL-SAFE skips the padding to further reduce size. 366 if (encodeTable == STANDARD_ENCODE_TABLE) { 367 buffer[context.pos++] = pad; 368 } 369 break; 370 default: 371 throw new IllegalStateException("Impossible modulus "+context.modulus); 372 } 373 context.currentLinePos += context.pos - savedPos; // keep track of current line position 374 // if currentPos == 0 we are at the start of a line, so don't add CRLF 375 if (lineLength > 0 && context.currentLinePos > 0) { 376 System.arraycopy(lineSeparator, 0, buffer, context.pos, lineSeparator.length); 377 context.pos += lineSeparator.length; 378 } 379 } else { 380 for (int i = 0; i < inAvail; i++) { 381 final byte[] buffer = ensureBufferSize(encodeSize, context); 382 context.modulus = (context.modulus+1) % BYTES_PER_UNENCODED_BLOCK; 383 int b = in[inPos++]; 384 if (b < 0) { 385 b += 256; 386 } 387 context.ibitWorkArea = (context.ibitWorkArea << 8) + b; // BITS_PER_BYTE 388 if (0 == context.modulus) { // 3 bytes = 24 bits = 4 * 6 bits to extract 389 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 18) & MASK_6BITS]; 390 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 12) & MASK_6BITS]; 391 buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 6) & MASK_6BITS]; 392 buffer[context.pos++] = encodeTable[context.ibitWorkArea & MASK_6BITS]; 393 context.currentLinePos += BYTES_PER_ENCODED_BLOCK; 394 if (lineLength > 0 && lineLength <= context.currentLinePos) { 395 System.arraycopy(lineSeparator, 0, buffer, context.pos, lineSeparator.length); 396 context.pos += lineSeparator.length; 397 context.currentLinePos = 0; 398 } 399 } 400 } 401 } 402 } 403 404 /** 405 * <p> 406 * Decodes all of the provided data, starting at inPos, for inAvail bytes. Should be called at least twice: once 407 * with the data to decode, and once with inAvail set to "-1" to alert decoder that EOF has been reached. The "-1" 408 * call is not necessary when decoding, but it doesn't hurt, either. 409 * </p> 410 * <p> 411 * Ignores all non-base64 characters. This is how chunked (e.g. 76 character) data is handled, since CR and LF are 412 * silently ignored, but has implications for other bytes, too. This method subscribes to the garbage-in, 413 * garbage-out philosophy: it will not check the provided data for validity. 414 * </p> 415 * <p> 416 * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach. 417 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 418 * </p> 419 * 420 * @param in 421 * byte[] array of ascii data to base64 decode. 422 * @param inPos 423 * Position to start reading data from. 424 * @param inAvail 425 * Amount of bytes available from input for encoding. 426 * @param context 427 * the context to be used 428 */ 429 @Override 430 void decode(final byte[] in, int inPos, final int inAvail, final Context context) { 431 if (context.eof) { 432 return; 433 } 434 if (inAvail < 0) { 435 context.eof = true; 436 } 437 for (int i = 0; i < inAvail; i++) { 438 final byte[] buffer = ensureBufferSize(decodeSize, context); 439 final byte b = in[inPos++]; 440 if (b == pad) { 441 // We're done. 442 context.eof = true; 443 break; 444 } 445 if (b >= 0 && b < DECODE_TABLE.length) { 446 final int result = DECODE_TABLE[b]; 447 if (result >= 0) { 448 context.modulus = (context.modulus+1) % BYTES_PER_ENCODED_BLOCK; 449 context.ibitWorkArea = (context.ibitWorkArea << BITS_PER_ENCODED_BYTE) + result; 450 if (context.modulus == 0) { 451 buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 16) & MASK_8BITS); 452 buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS); 453 buffer[context.pos++] = (byte) (context.ibitWorkArea & MASK_8BITS); 454 } 455 } 456 } 457 } 458 459 // Two forms of EOF as far as base64 decoder is concerned: actual 460 // EOF (-1) and first time '=' character is encountered in stream. 461 // This approach makes the '=' padding characters completely optional. 462 if (context.eof && context.modulus != 0) { 463 final byte[] buffer = ensureBufferSize(decodeSize, context); 464 465 // We have some spare bits remaining 466 // Output all whole multiples of 8 bits and ignore the rest 467 switch (context.modulus) { 468// case 0 : // impossible, as excluded above 469 case 1 : // 6 bits - ignore entirely 470 // TODO not currently tested; perhaps it is impossible? 471 break; 472 case 2 : // 12 bits = 8 + 4 473 context.ibitWorkArea = context.ibitWorkArea >> 4; // dump the extra 4 bits 474 buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS); 475 break; 476 case 3 : // 18 bits = 8 + 8 + 2 477 context.ibitWorkArea = context.ibitWorkArea >> 2; // dump 2 bits 478 buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS); 479 buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS); 480 break; 481 default: 482 throw new IllegalStateException("Impossible modulus "+context.modulus); 483 } 484 } 485 } 486 487 /** 488 * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently the 489 * method treats whitespace as valid. 490 * 491 * @param arrayOctet 492 * byte array to test 493 * @return <code>true</code> if all bytes are valid characters in the Base64 alphabet or if the byte array is empty; 494 * <code>false</code>, otherwise 495 * @deprecated 1.5 Use {@link #isBase64(byte[])}, will be removed in 2.0. 496 */ 497 @Deprecated 498 public static boolean isArrayByteBase64(final byte[] arrayOctet) { 499 return isBase64(arrayOctet); 500 } 501 502 /** 503 * Returns whether or not the <code>octet</code> is in the base 64 alphabet. 504 * 505 * @param octet 506 * The value to test 507 * @return <code>true</code> if the value is defined in the the base 64 alphabet, <code>false</code> otherwise. 508 * @since 1.4 509 */ 510 public static boolean isBase64(final byte octet) { 511 return octet == PAD_DEFAULT || (octet >= 0 && octet < DECODE_TABLE.length && DECODE_TABLE[octet] != -1); 512 } 513 514 /** 515 * Tests a given String to see if it contains only valid characters within the Base64 alphabet. Currently the 516 * method treats whitespace as valid. 517 * 518 * @param base64 519 * String to test 520 * @return <code>true</code> if all characters in the String are valid characters in the Base64 alphabet or if 521 * the String is empty; <code>false</code>, otherwise 522 * @since 1.5 523 */ 524 public static boolean isBase64(final String base64) { 525 return isBase64(StringUtils.getBytesUtf8(base64)); 526 } 527 528 /** 529 * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently the 530 * method treats whitespace as valid. 531 * 532 * @param arrayOctet 533 * byte array to test 534 * @return <code>true</code> if all bytes are valid characters in the Base64 alphabet or if the byte array is empty; 535 * <code>false</code>, otherwise 536 * @since 1.5 537 */ 538 public static boolean isBase64(final byte[] arrayOctet) { 539 for (int i = 0; i < arrayOctet.length; i++) { 540 if (!isBase64(arrayOctet[i]) && !isWhiteSpace(arrayOctet[i])) { 541 return false; 542 } 543 } 544 return true; 545 } 546 547 /** 548 * Encodes binary data using the base64 algorithm but does not chunk the output. 549 * 550 * @param binaryData 551 * binary data to encode 552 * @return byte[] containing Base64 characters in their UTF-8 representation. 553 */ 554 public static byte[] encodeBase64(final byte[] binaryData) { 555 return encodeBase64(binaryData, false); 556 } 557 558 /** 559 * Encodes binary data using the base64 algorithm but does not chunk the output. 560 * 561 * NOTE: We changed the behaviour of this method from multi-line chunking (commons-codec-1.4) to 562 * single-line non-chunking (commons-codec-1.5). 563 * 564 * @param binaryData 565 * binary data to encode 566 * @return String containing Base64 characters. 567 * @since 1.4 (NOTE: 1.4 chunked the output, whereas 1.5 does not). 568 */ 569 public static String encodeBase64String(final byte[] binaryData) { 570 return StringUtils.newStringUsAscii(encodeBase64(binaryData, false)); 571 } 572 573 /** 574 * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The 575 * url-safe variation emits - and _ instead of + and / characters. 576 * <b>Note: no padding is added.</b> 577 * @param binaryData 578 * binary data to encode 579 * @return byte[] containing Base64 characters in their UTF-8 representation. 580 * @since 1.4 581 */ 582 public static byte[] encodeBase64URLSafe(final byte[] binaryData) { 583 return encodeBase64(binaryData, false, true); 584 } 585 586 /** 587 * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The 588 * url-safe variation emits - and _ instead of + and / characters. 589 * <b>Note: no padding is added.</b> 590 * @param binaryData 591 * binary data to encode 592 * @return String containing Base64 characters 593 * @since 1.4 594 */ 595 public static String encodeBase64URLSafeString(final byte[] binaryData) { 596 return StringUtils.newStringUsAscii(encodeBase64(binaryData, false, true)); 597 } 598 599 /** 600 * Encodes binary data using the base64 algorithm and chunks the encoded output into 76 character blocks 601 * 602 * @param binaryData 603 * binary data to encode 604 * @return Base64 characters chunked in 76 character blocks 605 */ 606 public static byte[] encodeBase64Chunked(final byte[] binaryData) { 607 return encodeBase64(binaryData, true); 608 } 609 610 /** 611 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. 612 * 613 * @param binaryData 614 * Array containing binary data to encode. 615 * @param isChunked 616 * if <code>true</code> this encoder will chunk the base64 output into 76 character blocks 617 * @return Base64-encoded data. 618 * @throws IllegalArgumentException 619 * Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE} 620 */ 621 public static byte[] encodeBase64(final byte[] binaryData, final boolean isChunked) { 622 return encodeBase64(binaryData, isChunked, false); 623 } 624 625 /** 626 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. 627 * 628 * @param binaryData 629 * Array containing binary data to encode. 630 * @param isChunked 631 * if <code>true</code> this encoder will chunk the base64 output into 76 character blocks 632 * @param urlSafe 633 * if <code>true</code> this encoder will emit - and _ instead of the usual + and / characters. 634 * <b>Note: no padding is added when encoding using the URL-safe alphabet.</b> 635 * @return Base64-encoded data. 636 * @throws IllegalArgumentException 637 * Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE} 638 * @since 1.4 639 */ 640 public static byte[] encodeBase64(final byte[] binaryData, final boolean isChunked, final boolean urlSafe) { 641 return encodeBase64(binaryData, isChunked, urlSafe, Integer.MAX_VALUE); 642 } 643 644 /** 645 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. 646 * 647 * @param binaryData 648 * Array containing binary data to encode. 649 * @param isChunked 650 * if <code>true</code> this encoder will chunk the base64 output into 76 character blocks 651 * @param urlSafe 652 * if <code>true</code> this encoder will emit - and _ instead of the usual + and / characters. 653 * <b>Note: no padding is added when encoding using the URL-safe alphabet.</b> 654 * @param maxResultSize 655 * The maximum result size to accept. 656 * @return Base64-encoded data. 657 * @throws IllegalArgumentException 658 * Thrown when the input array needs an output array bigger than maxResultSize 659 * @since 1.4 660 */ 661 public static byte[] encodeBase64(final byte[] binaryData, final boolean isChunked, 662 final boolean urlSafe, final int maxResultSize) { 663 if (binaryData == null || binaryData.length == 0) { 664 return binaryData; 665 } 666 667 // Create this so can use the super-class method 668 // Also ensures that the same roundings are performed by the ctor and the code 669 final Base64 b64 = isChunked ? new Base64(urlSafe) : new Base64(0, CHUNK_SEPARATOR, urlSafe); 670 final long len = b64.getEncodedLength(binaryData); 671 if (len > maxResultSize) { 672 throw new IllegalArgumentException("Input array too big, the output array would be bigger (" + 673 len + 674 ") than the specified maximum size of " + 675 maxResultSize); 676 } 677 678 return b64.encode(binaryData); 679 } 680 681 /** 682 * Decodes a Base64 String into octets. 683 * <p> 684 * <b>Note:</b> this method seamlessly handles data encoded in URL-safe or normal mode. 685 * </p> 686 * 687 * @param base64String 688 * String containing Base64 data 689 * @return Array containing decoded data. 690 * @since 1.4 691 */ 692 public static byte[] decodeBase64(final String base64String) { 693 return new Base64().decode(base64String); 694 } 695 696 /** 697 * Decodes Base64 data into octets. 698 * <p> 699 * <b>Note:</b> this method seamlessly handles data encoded in URL-safe or normal mode. 700 * </p> 701 * 702 * @param base64Data 703 * Byte array containing Base64 data 704 * @return Array containing decoded data. 705 */ 706 public static byte[] decodeBase64(final byte[] base64Data) { 707 return new Base64().decode(base64Data); 708 } 709 710 // Implementation of the Encoder Interface 711 712 // Implementation of integer encoding used for crypto 713 /** 714 * Decodes a byte64-encoded integer according to crypto standards such as W3C's XML-Signature. 715 * 716 * @param pArray 717 * a byte array containing base64 character data 718 * @return A BigInteger 719 * @since 1.4 720 */ 721 public static BigInteger decodeInteger(final byte[] pArray) { 722 return new BigInteger(1, decodeBase64(pArray)); 723 } 724 725 /** 726 * Encodes to a byte64-encoded integer according to crypto standards such as W3C's XML-Signature. 727 * 728 * @param bigInt 729 * a BigInteger 730 * @return A byte array containing base64 character data 731 * @throws NullPointerException 732 * if null is passed in 733 * @since 1.4 734 */ 735 public static byte[] encodeInteger(final BigInteger bigInt) { 736 if (bigInt == null) { 737 throw new NullPointerException("encodeInteger called with null parameter"); 738 } 739 return encodeBase64(toIntegerBytes(bigInt), false); 740 } 741 742 /** 743 * Returns a byte-array representation of a <code>BigInteger</code> without sign bit. 744 * 745 * @param bigInt 746 * <code>BigInteger</code> to be converted 747 * @return a byte array representation of the BigInteger parameter 748 */ 749 static byte[] toIntegerBytes(final BigInteger bigInt) { 750 int bitlen = bigInt.bitLength(); 751 // round bitlen 752 bitlen = ((bitlen + 7) >> 3) << 3; 753 final byte[] bigBytes = bigInt.toByteArray(); 754 755 if (((bigInt.bitLength() % 8) != 0) && (((bigInt.bitLength() / 8) + 1) == (bitlen / 8))) { 756 return bigBytes; 757 } 758 // set up params for copying everything but sign bit 759 int startSrc = 0; 760 int len = bigBytes.length; 761 762 // if bigInt is exactly byte-aligned, just skip signbit in copy 763 if ((bigInt.bitLength() % 8) == 0) { 764 startSrc = 1; 765 len--; 766 } 767 final int startDst = bitlen / 8 - len; // to pad w/ nulls as per spec 768 final byte[] resizedBytes = new byte[bitlen / 8]; 769 System.arraycopy(bigBytes, startSrc, resizedBytes, startDst, len); 770 return resizedBytes; 771 } 772 773 /** 774 * Returns whether or not the <code>octet</code> is in the Base64 alphabet. 775 * 776 * @param octet 777 * The value to test 778 * @return <code>true</code> if the value is defined in the the Base64 alphabet <code>false</code> otherwise. 779 */ 780 @Override 781 protected boolean isInAlphabet(final byte octet) { 782 return octet >= 0 && octet < decodeTable.length && decodeTable[octet] != -1; 783 } 784 785}