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 org.apache.commons.net.util; 19 20 import java.math.BigInteger; 21 import java.nio.charset.StandardCharsets; 22 import java.util.Objects; 23 24 /** 25 * Provides Base64 encoding and decoding as defined by RFC 2045. 26 * 27 * <p> 28 * This class implements section <cite>6.8. Base64 Content-Transfer-Encoding</cite> from RFC 2045 <cite>Multipurpose Internet Mail Extensions (MIME) Part One: 29 * Format of Internet Message Bodies</cite> by Freed and Borenstein. 30 * </p> 31 * <p> 32 * The class can be parameterized in the following manner with various constructors: 33 * <ul> 34 * <li>URL-safe mode: Default off.</li> 35 * <li>Line length: Default 76. Line length that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. 36 * <li>Line separator: Default is CRLF ("\r\n")</li> 37 * </ul> 38 * <p> 39 * Since this class operates directly on byte streams, and not character streams, it is hard-coded to only encode/decode character encodings which are 40 * compatible with the lower 127 ASCII chart (ISO-8859-1, Windows-1252, UTF-8, etc). 41 * </p> 42 * 43 * @deprecated Use {@link java.util.Base64}. 44 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a> 45 * @since 2.2 46 */ 47 @Deprecated 48 public class Base64 { 49 private static final int DEFAULT_BUFFER_RESIZE_FACTOR = 2; 50 51 private static final int DEFAULT_BUFFER_SIZE = 8192; 52 53 /** 54 * Chunk size per RFC 2045 section 6.8. 55 * 56 * <p> 57 * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any equal signs. 58 * </p> 59 * 60 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 6.8</a> 61 */ 62 static final int CHUNK_SIZE = 76; 63 64 /** 65 * Chunk separator per RFC 2045 section 2.1. 66 * 67 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a> 68 */ 69 private static final byte[] CHUNK_SEPARATOR = { '\r', '\n' }; 70 71 /** 72 * This array is a lookup table that translates 6-bit positive integer index values into their "Base64 Alphabet" equivalents as specified in Table 1 of RFC 73 * 2045. 74 * 75 * Thanks to "commons" project in ws.apache.org for <a href="http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/">this code</a>. 76 */ 77 private static final byte[] STANDARD_ENCODE_TABLE = { 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 78 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 79 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/' }; 80 81 /** 82 * This is a copy of the STANDARD_ENCODE_TABLE above, but with + and / changed to - and _ to make the encoded Base64 results more URL-SAFE. This table is 83 * only used when the Base64's mode is set to URL-SAFE. 84 */ 85 private static final byte[] URL_SAFE_ENCODE_TABLE = { 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 86 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 87 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_' }; 88 89 /** 90 * Byte used to pad output. 91 */ 92 private static final byte PAD = '='; 93 94 /** 95 * This array is a lookup table that translates Unicode characters drawn from the "Base64 Alphabet" (as specified in Table 1 of RFC 2045) into their 6-bit 96 * positive integer equivalents. Characters that are not in the Base64 alphabet but fall within the bounds of the array are translated to -1. 97 * 98 * Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This means decoder seamlessly handles both URL_SAFE and STANDARD base64. (The 99 * encoder, on the other hand, needs to know ahead of time what to emit). 100 * 101 * Thanks to "commons" project in ws.apache.org for <a href="http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/">this code</a> 102 */ 103 private static final byte[] DECODE_TABLE = { -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 104 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, 62, -1, 63, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, -1, 105 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, 63, -1, 26, 27, 28, 29, 30, 31, 32, 106 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 }; 107 108 /** Mask used to extract 6 bits, used when encoding */ 109 private static final int MASK_6BITS = 0x3f; 110 111 /** Mask used to extract 8 bits, used in decoding base64 bytes */ 112 private static final int MASK_8BITS = 0xff; 113 114 // The static final fields above are used for the original static byte[] methods on Base64. 115 // The private member fields below are used with the new streaming approach, which requires 116 // some state be preserved between calls of encode() and decode(). 117 118 /** 119 * Tests a given byte array to see if it contains any valid character within the Base64 alphabet. 120 * 121 * @param arrayOctet byte array to test 122 * @return {@code true} if any byte is a valid character in the Base64 alphabet; {@code false} otherwise 123 */ 124 private static boolean containsBase64Byte(final byte[] arrayOctet) { 125 for (final byte element : arrayOctet) { 126 if (isBase64(element)) { 127 return true; 128 } 129 } 130 return false; 131 } 132 133 /** 134 * Decodes Base64 data into octets. 135 * 136 * @param base64Data Byte array containing Base64 data 137 * @return Array containing decoded data. 138 */ 139 public static byte[] decodeBase64(final byte[] base64Data) { 140 return new Base64().decode(base64Data); 141 } 142 143 /** 144 * Decodes a Base64 String into octets. 145 * 146 * @param base64String String containing Base64 data 147 * @return Array containing decoded data. 148 * @since 1.4 149 */ 150 public static byte[] decodeBase64(final String base64String) { 151 return new Base64().decode(base64String); 152 } 153 154 // Implementation of integer encoding used for crypto 155 /** 156 * Decodes a byte64-encoded integer according to crypto standards such as W3C's XML-Signature 157 * 158 * @param pArray a byte array containing base64 character data 159 * @return A BigInteger 160 * @since 1.4 161 */ 162 public static BigInteger decodeInteger(final byte[] pArray) { 163 return new BigInteger(1, decodeBase64(pArray)); 164 } 165 166 /** 167 * Encodes binary data using the base64 algorithm but does not chunk the output. 168 * 169 * @param binaryData binary data to encode 170 * @return byte[] containing Base64 characters in their UTF-8 representation. 171 */ 172 public static byte[] encodeBase64(final byte[] binaryData) { 173 return encodeBase64(binaryData, false); 174 } 175 176 /** 177 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. 178 * 179 * @param binaryData Array containing binary data to encode. 180 * @param isChunked if {@code true} this encoder will chunk the base64 output into 76 character blocks 181 * @return Base64-encoded data. 182 * @throws IllegalArgumentException Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE} 183 */ 184 public static byte[] encodeBase64(final byte[] binaryData, final boolean isChunked) { 185 return encodeBase64(binaryData, isChunked, false); 186 } 187 188 /** 189 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. 190 * 191 * @param binaryData Array containing binary data to encode. 192 * @param isChunked if {@code true} this encoder will chunk the base64 output into 76 character blocks 193 * @param urlSafe if {@code true} this encoder will emit - and _ instead of the usual + and / characters. 194 * @return Base64-encoded data. 195 * @throws IllegalArgumentException Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE} 196 * @since 1.4 197 */ 198 public static byte[] encodeBase64(final byte[] binaryData, final boolean isChunked, final boolean urlSafe) { 199 return encodeBase64(binaryData, isChunked, urlSafe, Integer.MAX_VALUE); 200 } 201 202 /** 203 * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. 204 * 205 * @param binaryData Array containing binary data to encode. 206 * @param isChunked if {@code true} this encoder will chunk the base64 output into 76 character blocks 207 * @param urlSafe if {@code true} this encoder will emit - and _ instead of the usual + and / characters. 208 * @param maxResultSize The maximum result size to accept. 209 * @return Base64-encoded data. 210 * @throws IllegalArgumentException Thrown when the input array needs an output array bigger than maxResultSize 211 * @since 1.4 212 */ 213 public static byte[] encodeBase64(final byte[] binaryData, final boolean isChunked, final boolean urlSafe, final int maxResultSize) { 214 if (binaryData == null || binaryData.length == 0) { 215 return binaryData; 216 } 217 218 final long len = getEncodeLength(binaryData, isChunked ? CHUNK_SIZE : 0, isChunked ? CHUNK_SEPARATOR : NetConstants.EMPTY_BTYE_ARRAY); 219 if (len > maxResultSize) { 220 throw new IllegalArgumentException( 221 "Input array too big, the output array would be bigger (" + len + ") than the specified maxium size of " + maxResultSize); 222 } 223 224 final Base64 b64 = isChunked ? new Base64(urlSafe) : new Base64(0, CHUNK_SEPARATOR, urlSafe); 225 return b64.encode(binaryData); 226 } 227 228 /** 229 * Encodes binary data using the base64 algorithm and chunks the encoded output into 76 character blocks 230 * 231 * @param binaryData binary data to encode 232 * @return Base64 characters chunked in 76 character blocks 233 */ 234 public static byte[] encodeBase64Chunked(final byte[] binaryData) { 235 return encodeBase64(binaryData, true); 236 } 237 238 /** 239 * Encodes binary data using the base64 algorithm into 76 character blocks separated by CRLF. 240 * <p> 241 * For a non-chunking version, see {@link #encodeBase64StringUnChunked(byte[])}. 242 * 243 * @param binaryData binary data to encode 244 * @return String containing Base64 characters. 245 * @since 1.4 246 */ 247 public static String encodeBase64String(final byte[] binaryData) { 248 return newStringUtf8(encodeBase64(binaryData, true)); 249 } 250 251 /** 252 * Encodes binary data using the base64 algorithm. 253 * 254 * @param binaryData binary data to encode 255 * @param useChunking whether to split the output into chunks 256 * @return String containing Base64 characters. 257 * @since 3.2 258 */ 259 public static String encodeBase64String(final byte[] binaryData, final boolean useChunking) { 260 return newStringUtf8(encodeBase64(binaryData, useChunking)); 261 } 262 263 /** 264 * Encodes binary data using the base64 algorithm, without using chunking. 265 * <p> 266 * For a chunking version, see {@link #encodeBase64String(byte[])}. 267 * 268 * @param binaryData binary data to encode 269 * @return String containing Base64 characters. 270 * @since 3.2 271 */ 272 public static String encodeBase64StringUnChunked(final byte[] binaryData) { 273 return newStringUtf8(encodeBase64(binaryData, false)); 274 } 275 276 /** 277 * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The url-safe variation emits - and _ instead of + 278 * and / characters. 279 * 280 * @param binaryData binary data to encode 281 * @return byte[] containing Base64 characters in their UTF-8 representation. 282 * @since 1.4 283 */ 284 public static byte[] encodeBase64URLSafe(final byte[] binaryData) { 285 return encodeBase64(binaryData, false, true); 286 } 287 288 /** 289 * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The url-safe variation emits - and _ instead of + 290 * and / characters. 291 * 292 * @param binaryData binary data to encode 293 * @return String containing Base64 characters 294 * @since 1.4 295 */ 296 public static String encodeBase64URLSafeString(final byte[] binaryData) { 297 return newStringUtf8(encodeBase64(binaryData, false, true)); 298 } 299 300 /** 301 * Encodes to a byte64-encoded integer according to crypto standards such as W3C's XML-Signature 302 * 303 * @param bigInt a BigInteger 304 * @return A byte array containing base64 character data 305 * @throws NullPointerException if null is passed in 306 * @since 1.4 307 */ 308 public static byte[] encodeInteger(final BigInteger bigInt) { 309 return encodeBase64(toIntegerBytes(bigInt), false); 310 } 311 312 /** 313 * Pre-calculates the amount of space needed to base64-encode the supplied array. 314 * 315 * @param pArray byte[] array which will later be encoded 316 * @param chunkSize line-length of the output (<= 0 means no chunking) between each chunkSeparator (e.g. CRLF). 317 * @param chunkSeparator the sequence of bytes used to separate chunks of output (e.g. CRLF). 318 * 319 * @return amount of space needed to encode the supplied array. Returns a long since a max-len array will require Integer.MAX_VALUE + 33%. 320 */ 321 private static long getEncodeLength(final byte[] pArray, int chunkSize, final byte[] chunkSeparator) { 322 // base64 always encodes to multiples of 4. 323 chunkSize = chunkSize / 4 * 4; 324 325 long len = pArray.length * 4 / 3; 326 final long mod = len % 4; 327 if (mod != 0) { 328 len += 4 - mod; 329 } 330 if (chunkSize > 0) { 331 final boolean lenChunksPerfectly = len % chunkSize == 0; 332 len += len / chunkSize * chunkSeparator.length; 333 if (!lenChunksPerfectly) { 334 len += chunkSeparator.length; 335 } 336 } 337 return len; 338 } 339 340 /** 341 * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently, the method treats whitespace as valid. 342 * 343 * @param arrayOctet byte array to test 344 * @return {@code true} if all bytes are valid characters in the Base64 alphabet or if the byte array is empty; false, otherwise 345 */ 346 public static boolean isArrayByteBase64(final byte[] arrayOctet) { 347 for (final byte element : arrayOctet) { 348 if (!isBase64(element) && !isWhiteSpace(element)) { 349 return false; 350 } 351 } 352 return true; 353 } 354 355 /** 356 * Returns whether or not the <code>octet</code> is in the base 64 alphabet. 357 * 358 * @param octet The value to test 359 * @return {@code true} if the value is defined in the base 64 alphabet, {@code false} otherwise. 360 * @since 1.4 361 */ 362 public static boolean isBase64(final byte octet) { 363 return octet == PAD || octet >= 0 && octet < DECODE_TABLE.length && DECODE_TABLE[octet] != -1; 364 } 365 366 /** 367 * Checks if a byte value is whitespace or not. 368 * 369 * @param byteToCheck the byte to check 370 * @return true if byte is whitespace, false otherwise 371 */ 372 private static boolean isWhiteSpace(final byte byteToCheck) { 373 switch (byteToCheck) { 374 case ' ': 375 case '\n': 376 case '\r': 377 case '\t': 378 return true; 379 default: 380 return false; 381 } 382 } 383 384 private static String newStringUtf8(final byte[] encode) { 385 return new String(encode, StandardCharsets.UTF_8); 386 } 387 388 /** 389 * Returns a byte-array representation of a <code>BigInteger</code> without sign bit. 390 * 391 * @param bigInt <code>BigInteger</code> to be converted 392 * @return a byte array representation of the BigInteger parameter 393 */ 394 static byte[] toIntegerBytes(final BigInteger bigInt) { 395 Objects.requireNonNull(bigInt, "bigInt"); 396 int bitlen = bigInt.bitLength(); 397 // round bitlen 398 bitlen = bitlen + 7 >> 3 << 3; 399 final byte[] bigBytes = bigInt.toByteArray(); 400 401 if (bigInt.bitLength() % 8 != 0 && bigInt.bitLength() / 8 + 1 == bitlen / 8) { 402 return bigBytes; 403 } 404 // set up params for copying everything but sign bit 405 int startSrc = 0; 406 int len = bigBytes.length; 407 408 // if bigInt is exactly byte-aligned, just skip signbit in copy 409 if (bigInt.bitLength() % 8 == 0) { 410 startSrc = 1; 411 len--; 412 } 413 final int startDst = bitlen / 8 - len; // to pad w/ nulls as per spec 414 final byte[] resizedBytes = new byte[bitlen / 8]; 415 System.arraycopy(bigBytes, startSrc, resizedBytes, startDst, len); 416 return resizedBytes; 417 } 418 419 /** 420 * Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE above remains static because it is able to decode both STANDARD and URL_SAFE 421 * streams, but the encodeTable must be a member variable, so we can switch between the two modes. 422 */ 423 private final byte[] encodeTable; 424 425 /** 426 * Line length for encoding. Not used when decoding. A value of zero or less implies no chunking of the base64 encoded data. 427 */ 428 private final int lineLength; 429 430 /** 431 * Line separator for encoding. Not used when decoding. Only used if lineLength > 0. 432 */ 433 private final byte[] lineSeparator; 434 435 /** 436 * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing. 437 * <code>decodeSize = 3 + lineSeparator.length;</code> 438 */ 439 private final int decodeSize; 440 441 /** 442 * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing. 443 * <code>encodeSize = 4 + lineSeparator.length;</code> 444 */ 445 private final int encodeSize; 446 447 /** 448 * Buffer for streaming. 449 */ 450 private byte[] buffer; 451 452 /** 453 * Position where next character should be written in the buffer. 454 */ 455 private int pos; 456 457 /** 458 * Position where next character should be read from the buffer. 459 */ 460 private int readPos; 461 462 /** 463 * Variable tracks how many characters have been written to the current line. Only used when encoding. We use it to make sure each encoded line never goes 464 * beyond lineLength (if lineLength > 0). 465 */ 466 private int currentLinePos; 467 468 /** 469 * Writes to the buffer only occur after every 3 reads when encoding, an every 4 reads when decoding. This variable helps track that. 470 */ 471 private int modulus; 472 473 /** 474 * Boolean flag to indicate the EOF has been reached. Once EOF has been reached, this Base64 object becomes useless, and must be thrown away. 475 */ 476 private boolean eof; 477 478 /** 479 * Placeholder for the 3 bytes we're dealing with for our base64 logic. Bitwise operations store and extract the base64 encoding or decoding from this 480 * variable. 481 */ 482 private int x; 483 484 /** 485 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 486 * <p> 487 * When encoding the line length is 76, the line separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE. 488 * </p> 489 * 490 * <p> 491 * When decoding all variants are supported. 492 * </p> 493 */ 494 public Base64() { 495 this(false); 496 } 497 498 /** 499 * Creates a Base64 codec used for decoding (all modes) and encoding in the given URL-safe mode. 500 * <p> 501 * When encoding the line length is 76, the line separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE. 502 * </p> 503 * 504 * <p> 505 * When decoding all variants are supported. 506 * </p> 507 * 508 * @param urlSafe if {@code true}, URL-safe encoding is used. In most cases this should be set to {@code false}. 509 * @since 1.4 510 */ 511 public Base64(final boolean urlSafe) { 512 this(CHUNK_SIZE, CHUNK_SEPARATOR, urlSafe); 513 } 514 515 /** 516 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 517 * <p> 518 * When encoding the line length is given in the constructor, the line separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE. 519 * </p> 520 * <p> 521 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. 522 * </p> 523 * <p> 524 * When decoding all variants are supported. 525 * </p> 526 * 527 * @param lineLength Each line of encoded data will be at most of the given length (rounded down to the nearest multiple of 4). 528 * If {@code lineLength <= 0}, then the output will not be divided into lines (chunks). Ignored when decoding. 529 * @since 1.4 530 */ 531 public Base64(final int lineLength) { 532 this(lineLength, CHUNK_SEPARATOR); 533 } 534 535 /** 536 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 537 * <p> 538 * When encoding the line length and line separator are given in the constructor, and the encoding table is STANDARD_ENCODE_TABLE. 539 * </p> 540 * <p> 541 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. 542 * </p> 543 * <p> 544 * When decoding all variants are supported. 545 * </p> 546 * 547 * @param lineLength Each line of encoded data will be at most of the given length (rounded down to the nearest multiple of 4). 548 * If {@code lineLength <= 0}, then the output will not be divided into lines (chunks). Ignored when decoding. 549 * @param lineSeparator Each line of encoded data will end with this sequence of bytes. 550 * @throws IllegalArgumentException Thrown when the provided lineSeparator included some base64 characters. 551 * @since 1.4 552 */ 553 public Base64(final int lineLength, final byte[] lineSeparator) { 554 this(lineLength, lineSeparator, false); 555 } 556 557 /** 558 * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. 559 * <p> 560 * When encoding the line length and line separator are given in the constructor, and the encoding table is STANDARD_ENCODE_TABLE. 561 * </p> 562 * <p> 563 * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. 564 * </p> 565 * <p> 566 * When decoding all variants are supported. 567 * </p> 568 * 569 * @param lineLength Each line of encoded data will be at most of the given length (rounded down to the nearest multiple of 4). 570 * If {@code lineLength <= 0}, then the output will not be divided into lines (chunks). Ignored when decoding. 571 * @param lineSeparator Each line of encoded data will end with this sequence of bytes. 572 * @param urlSafe Instead of emitting '+' and '/' we emit '-' and '_' respectively. urlSafe is only applied to encode operations. Decoding seamlessly 573 * handles both modes. 574 * @throws IllegalArgumentException The provided lineSeparator included some base64 characters. That's not going to work! 575 * @since 1.4 576 */ 577 public Base64(int lineLength, byte[] lineSeparator, final boolean urlSafe) { 578 if (lineSeparator == null) { 579 lineLength = 0; // disable chunk-separating 580 lineSeparator = NetConstants.EMPTY_BTYE_ARRAY; // this just gets ignored 581 } 582 this.lineLength = lineLength > 0 ? lineLength / 4 * 4 : 0; 583 this.lineSeparator = new byte[lineSeparator.length]; 584 System.arraycopy(lineSeparator, 0, this.lineSeparator, 0, lineSeparator.length); 585 if (lineLength > 0) { 586 this.encodeSize = 4 + lineSeparator.length; 587 } else { 588 this.encodeSize = 4; 589 } 590 this.decodeSize = this.encodeSize - 1; 591 if (containsBase64Byte(lineSeparator)) { 592 final String sep = newStringUtf8(lineSeparator); 593 throw new IllegalArgumentException("lineSeperator must not contain base64 characters: [" + sep + "]"); 594 } 595 this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE : STANDARD_ENCODE_TABLE; 596 } 597 598 /** 599 * Returns the amount of buffered data available for reading. 600 * 601 * @return The amount of buffered data available for reading. 602 */ 603 int avail() { 604 return buffer != null ? pos - readPos : 0; 605 } 606 607 /** 608 * Decodes a byte array containing characters in the Base64 alphabet. 609 * 610 * @param pArray A byte array containing Base64 character data 611 * @return a byte array containing binary data; will return {@code null} if provided byte array is {@code null}. 612 */ 613 public byte[] decode(final byte[] pArray) { 614 reset(); 615 if (pArray == null || pArray.length == 0) { 616 return pArray; 617 } 618 final long len = pArray.length * 3 / 4; 619 final byte[] buf = new byte[(int) len]; 620 setInitialBuffer(buf, 0, buf.length); 621 decode(pArray, 0, pArray.length); 622 decode(pArray, 0, -1); // Notify decoder of EOF. 623 624 // Would be nice to just return buf (like we sometimes do in the encode 625 // logic), but we have no idea what the line-length was (could even be 626 // variable). So we cannot determine ahead of time exactly how big an 627 // array is necessary. Hence, the need to construct a 2nd byte array to 628 // hold the final result: 629 630 final byte[] result = new byte[pos]; 631 readResults(result, 0, result.length); 632 return result; 633 } 634 635 /** 636 * <p> 637 * Decodes all of the provided data, starting at inPos, for inAvail bytes. Should be called at least twice: once with the data to decode, and once with 638 * inAvail set to "-1" to alert decoder that EOF has been reached. The "-1" call is not necessary when decoding, but it doesn't hurt, either. 639 * </p> 640 * <p> 641 * Ignores all non-base64 characters. This is how chunked (e.g. 76 character) data is handled, since CR and LF are silently ignored, but has implications 642 * for other bytes, too. This method subscribes to the garbage-in, garbage-out philosophy: it will not check the provided data for validity. 643 * </p> 644 * <p> 645 * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach. 646 * <p> 647 * See: <a href="http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/"> 648 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 649 * </a> 650 * 651 * </p> 652 * 653 * @param in byte[] array of ascii data to base64 decode. 654 * @param inPos Position to start reading data from. 655 * @param inAvail Amount of bytes available from input for encoding. 656 */ 657 void decode(final byte[] in, int inPos, final int inAvail) { 658 if (eof) { 659 return; 660 } 661 if (inAvail < 0) { 662 eof = true; 663 } 664 for (int i = 0; i < inAvail; i++) { 665 if (buffer == null || buffer.length - pos < decodeSize) { 666 resizeBuffer(); 667 } 668 final byte b = in[inPos++]; 669 if (b == PAD) { 670 // We're done. 671 eof = true; 672 break; 673 } 674 if (b >= 0 && b < DECODE_TABLE.length) { 675 final int result = DECODE_TABLE[b]; 676 if (result >= 0) { 677 modulus = ++modulus % 4; 678 x = (x << 6) + result; 679 if (modulus == 0) { 680 buffer[pos++] = (byte) (x >> 16 & MASK_8BITS); 681 buffer[pos++] = (byte) (x >> 8 & MASK_8BITS); 682 buffer[pos++] = (byte) (x & MASK_8BITS); 683 } 684 } 685 } 686 } 687 688 // Two forms of EOF as far as base64 decoder is concerned: actual 689 // EOF (-1) and first time '=' character is encountered in stream. 690 // This approach makes the '=' padding characters completely optional. 691 if (eof && modulus != 0) { 692 x = x << 6; 693 switch (modulus) { 694 case 2: 695 x = x << 6; 696 buffer[pos++] = (byte) (x >> 16 & MASK_8BITS); 697 break; 698 case 3: 699 buffer[pos++] = (byte) (x >> 16 & MASK_8BITS); 700 buffer[pos++] = (byte) (x >> 8 & MASK_8BITS); 701 break; 702 default: 703 break; // other values ignored 704 } 705 } 706 } 707 708 /** 709 * Decodes a String containing characters in the Base64 alphabet. 710 * 711 * @param pArray A String containing Base64 character data, must not be {@code null} 712 * @return a byte array containing binary data 713 * @since 1.4 714 */ 715 public byte[] decode(final String pArray) { 716 return decode(getBytesUtf8(pArray)); 717 } 718 719 /** 720 * Encodes a byte[] containing binary data, into a byte[] containing characters in the Base64 alphabet. 721 * 722 * @param pArray a byte array containing binary data 723 * @return A byte array containing only Base64 character data 724 */ 725 public byte[] encode(final byte[] pArray) { 726 reset(); 727 if (pArray == null || pArray.length == 0) { 728 return pArray; 729 } 730 final long len = getEncodeLength(pArray, lineLength, lineSeparator); 731 byte[] buf = new byte[(int) len]; 732 setInitialBuffer(buf, 0, buf.length); 733 encode(pArray, 0, pArray.length); 734 encode(pArray, 0, -1); // Notify encoder of EOF. 735 // Encoder might have resized, even though it was unnecessary. 736 if (buffer != buf) { 737 readResults(buf, 0, buf.length); 738 } 739 // In URL-SAFE mode we skip the padding characters, so sometimes our 740 // final length is a bit smaller. 741 if (isUrlSafe() && pos < buf.length) { 742 final byte[] smallerBuf = new byte[pos]; 743 System.arraycopy(buf, 0, smallerBuf, 0, pos); 744 buf = smallerBuf; 745 } 746 return buf; 747 } 748 749 /** 750 * <p> 751 * Encodes all of the provided data, starting at inPos, for inAvail bytes. Must be called at least twice: once with the data to encode, and once with 752 * inAvail set to "-1" to alert encoder that EOF has been reached, so flush last remaining bytes (if not multiple of 3). 753 * </p> 754 * <p> 755 * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach. 756 * <p> 757 * See: <a href="http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/"> 758 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 759 * </a> 760 * </p> 761 * 762 * @param in byte[] array of binary data to base64 encode. 763 * @param inPos Position to start reading data from. 764 * @param inAvail Amount of bytes available from input for encoding. 765 */ 766 void encode(final byte[] in, int inPos, final int inAvail) { 767 if (eof) { 768 return; 769 } 770 // inAvail < 0 is how we're informed of EOF in the underlying data we're 771 // encoding. 772 if (inAvail < 0) { 773 eof = true; 774 if (buffer == null || buffer.length - pos < encodeSize) { 775 resizeBuffer(); 776 } 777 switch (modulus) { 778 case 1: 779 buffer[pos++] = encodeTable[x >> 2 & MASK_6BITS]; 780 buffer[pos++] = encodeTable[x << 4 & MASK_6BITS]; 781 // URL-SAFE skips the padding to further reduce size. 782 if (encodeTable == STANDARD_ENCODE_TABLE) { 783 buffer[pos++] = PAD; 784 buffer[pos++] = PAD; 785 } 786 break; 787 788 case 2: 789 buffer[pos++] = encodeTable[x >> 10 & MASK_6BITS]; 790 buffer[pos++] = encodeTable[x >> 4 & MASK_6BITS]; 791 buffer[pos++] = encodeTable[x << 2 & MASK_6BITS]; 792 // URL-SAFE skips the padding to further reduce size. 793 if (encodeTable == STANDARD_ENCODE_TABLE) { 794 buffer[pos++] = PAD; 795 } 796 break; 797 default: 798 break; // other values ignored 799 } 800 if (lineLength > 0 && pos > 0) { 801 System.arraycopy(lineSeparator, 0, buffer, pos, lineSeparator.length); 802 pos += lineSeparator.length; 803 } 804 } else { 805 for (int i = 0; i < inAvail; i++) { 806 if (buffer == null || buffer.length - pos < encodeSize) { 807 resizeBuffer(); 808 } 809 modulus = ++modulus % 3; 810 int b = in[inPos++]; 811 if (b < 0) { 812 b += 256; 813 } 814 x = (x << 8) + b; 815 if (0 == modulus) { 816 buffer[pos++] = encodeTable[x >> 18 & MASK_6BITS]; 817 buffer[pos++] = encodeTable[x >> 12 & MASK_6BITS]; 818 buffer[pos++] = encodeTable[x >> 6 & MASK_6BITS]; 819 buffer[pos++] = encodeTable[x & MASK_6BITS]; 820 currentLinePos += 4; 821 if (lineLength > 0 && lineLength <= currentLinePos) { 822 System.arraycopy(lineSeparator, 0, buffer, pos, lineSeparator.length); 823 pos += lineSeparator.length; 824 currentLinePos = 0; 825 } 826 } 827 } 828 } 829 } 830 831 /** 832 * Encodes a byte[] containing binary data, into a String containing characters in the Base64 alphabet. 833 * 834 * @param pArray a byte array containing binary data 835 * @return A String containing only Base64 character data 836 * @since 1.4 837 */ 838 public String encodeToString(final byte[] pArray) { 839 return newStringUtf8(encode(pArray)); 840 } 841 842 private byte[] getBytesUtf8(final String pArray) { 843 return pArray.getBytes(StandardCharsets.UTF_8); 844 } 845 846 int getLineLength() { 847 return lineLength; 848 } 849 850 byte[] getLineSeparator() { 851 return lineSeparator.clone(); 852 } 853 854 /** 855 * Returns true if this Base64 object has buffered data for reading. 856 * 857 * @return true if there is Base64 object still available for reading. 858 */ 859 boolean hasData() { 860 return this.buffer != null; 861 } 862 863 /** 864 * Returns our current encode mode. True if we're URL-SAFE, false otherwise. 865 * 866 * @return true if we're in URL-SAFE mode, false otherwise. 867 * @since 1.4 868 */ 869 public boolean isUrlSafe() { 870 return this.encodeTable == URL_SAFE_ENCODE_TABLE; 871 } 872 873 /** 874 * Extracts buffered data into the provided byte[] array, starting at position bPos, up to a maximum of bAvail bytes. Returns how many bytes were actually 875 * extracted. 876 * 877 * @param b byte[] array to extract the buffered data into. 878 * @param bPos position in byte[] array to start extraction at. 879 * @param bAvail amount of bytes we're allowed to extract. We may extract fewer (if fewer are available). 880 * @return The number of bytes successfully extracted into the provided byte[] array. 881 */ 882 int readResults(final byte[] b, final int bPos, final int bAvail) { 883 if (buffer != null) { 884 final int len = Math.min(avail(), bAvail); 885 if (buffer != b) { 886 System.arraycopy(buffer, readPos, b, bPos, len); 887 readPos += len; 888 if (readPos >= pos) { 889 buffer = null; 890 } 891 } else { 892 // Re-using the original consumer's output array is only 893 // allowed for one round. 894 buffer = null; 895 } 896 return len; 897 } 898 return eof ? -1 : 0; 899 } 900 901 /** 902 * Resets this Base64 object to its initial newly constructed state. 903 */ 904 private void reset() { 905 buffer = null; 906 pos = 0; 907 readPos = 0; 908 currentLinePos = 0; 909 modulus = 0; 910 eof = false; 911 } 912 913 // Getters for use in testing 914 915 /** Doubles our buffer. */ 916 private void resizeBuffer() { 917 if (buffer == null) { 918 buffer = new byte[DEFAULT_BUFFER_SIZE]; 919 pos = 0; 920 readPos = 0; 921 } else { 922 final byte[] b = new byte[buffer.length * DEFAULT_BUFFER_RESIZE_FACTOR]; 923 System.arraycopy(buffer, 0, b, 0, buffer.length); 924 buffer = b; 925 } 926 } 927 928 /** 929 * Sets the streaming buffer. This is a small optimization where we try to buffer directly to the consumer's output array for one round (if the consumer 930 * calls this method first) instead of starting our own buffer. 931 * 932 * @param out byte[] array to buffer directly to. 933 * @param outPos Position to start buffering into. 934 * @param outAvail Amount of bytes available for direct buffering. 935 */ 936 void setInitialBuffer(final byte[] out, final int outPos, final int outAvail) { 937 // We can re-use consumer's original output array under 938 // special circumstances, saving on some System.arraycopy(). 939 if (out != null && out.length == outAvail) { 940 buffer = out; 941 pos = outPos; 942 readPos = outPos; 943 } 944 } 945 }