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 ca.uhn.hl7v2.hoh.util.repackage; 19 20 import java.io.UnsupportedEncodingException; 21 import java.math.BigInteger; 22 import java.nio.charset.Charset; 23 import java.nio.charset.StandardCharsets; 24 import java.nio.charset.UnsupportedCharsetException; 25 26 /** 27 * Provides Base64 encoding and decoding as defined by <a 28 * href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>. 29 * 30 * <p> 31 * This class implements section <cite>6.8. Base64 32 * Content-Transfer-Encoding</cite> from RFC 2045 <cite>Multipurpose Internet 33 * Mail Extensions (MIME) Part One: Format of Internet Message Bodies</cite> by 34 * Freed and Borenstein. 35 * </p> 36 * <p> 37 * The class can be parameterized in the following manner with various 38 * constructors: 39 * <ul> 40 * <li>URL-safe mode: Default off.</li> 41 * <li>Line length: Default 76. Line length that aren't multiples of 4 will 42 * still essentially end up being multiples of 4 in the encoded data. 43 * <li>Line separator: Default is CRLF ("\r\n")</li> 44 * </ul> 45 * </p> 46 * <p> 47 * Since this class operates directly on byte streams, and not character 48 * streams, it is hard-coded to only encode/decode character encodings which are 49 * compatible with the lower 127 ASCII chart (ISO-8859-1, Windows-1252, UTF-8, 50 * etc). 51 * </p> 52 * <p> 53 * This class is thread-safe. 54 * </p> 55 * 56 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a> 57 * @author Note that this class has been repackaged from Apache Commons-Codec 58 * and is distributed under the terms of the Apache Software License, 59 * version 2.0 60 */ 61 public class Base64 { 62 63 public static void main(String[] args) { 64 65 System.out.println("basic " + encodeBase64String("cgta:d@3r$@TTg2446yhhh2h4".getBytes())); 66 67 } 68 69 /** 70 * BASE32 characters are 6 bits in length. They are formed by taking a block 71 * of 3 octets to form a 24-bit string, which is converted into 4 BASE64 72 * characters. 73 */ 74 private static final int BITS_PER_ENCODED_BYTE = 6; 75 private static final int BYTES_PER_UNENCODED_BLOCK = 3; 76 private static final int BYTES_PER_ENCODED_BLOCK = 4; 77 78 /** 79 * Chunk separator per RFC 2045 section 2.1. 80 * 81 * <p> 82 * N.B. The next major release may break compatibility and make this field 83 * private. 84 * </p> 85 * 86 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 87 * 2.1</a> 88 */ 89 static final byte[] CHUNK_SEPARATOR = { '\r', '\n' }; 90 91 /** 92 * This array is a lookup table that translates 6-bit positive integer index 93 * values into their "Base64 Alphabet" equivalents as specified in Table 1 94 * of RFC 2045. 95 * 96 * Thanks to "commons" project in ws.apache.org for this code. 97 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 98 */ 99 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', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 100 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/' }; 101 102 /** 103 * This is a copy of the STANDARD_ENCODE_TABLE above, but with + and / 104 * changed to - and _ to make the encoded Base64 results more URL-SAFE. This 105 * table is only used when the Base64's mode is set to URL-SAFE. 106 */ 107 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', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 108 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_' }; 109 110 /** 111 * This array is a lookup table that translates Unicode characters drawn 112 * from the "Base64 Alphabet" (as specified in Table 1 of RFC 2045) into 113 * their 6-bit positive integer equivalents. Characters that are not in the 114 * Base64 alphabet but fall within the bounds of the array are translated to 115 * -1. 116 * 117 * Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This 118 * means decoder seamlessly handles both URL_SAFE and STANDARD base64. (The 119 * encoder, on the other hand, needs to know ahead of time what to emit). 120 * 121 * Thanks to "commons" project in ws.apache.org for this code. 122 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 123 */ 124 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, -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, 125 61, -1, -1, -1, -1, -1, -1, -1, 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, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 }; 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[] 134 // methods on Base64. 135 // The private member fields below are used with the new streaming approach, 136 // which requires 137 // some state be preserved between calls of encode() and decode(). 138 139 /** 140 * Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE 141 * above remains static because it is able to decode both STANDARD and 142 * URL_SAFE streams, but the encodeTable must be a member variable so we can 143 * switch between the two modes. 144 */ 145 private byte[] encodeTable; 146 147 // Only one decode table currently; keep for consistency with Base32 code 148 private final byte[] decodeTable = DECODE_TABLE; 149 150 /** 151 * Line separator for encoding. Not used when decoding. Only used if 152 * lineLength > 0. 153 */ 154 private byte[] lineSeparator; 155 156 /** 157 * Convenience variable to help us determine when our buffer is going to run 158 * out of room and needs resizing. 159 * <code>decodeSize = 3 + lineSeparator.length;</code> 160 */ 161 private int decodeSize; 162 163 /** 164 * Convenience variable to help us determine when our buffer is going to run 165 * out of room and needs resizing. 166 * <code>encodeSize = 4 + lineSeparator.length;</code> 167 */ 168 private int encodeSize; 169 170 /** 171 * Creates a Base64 codec used for decoding (all modes) and encoding in 172 * URL-unsafe mode. 173 * <p> 174 * When encoding the line length is 0 (no chunking), and the encoding table 175 * is STANDARD_ENCODE_TABLE. 176 * </p> 177 * 178 * <p> 179 * When decoding all variants are supported. 180 * </p> 181 */ 182 public Base64() { 183 this(0); 184 } 185 186 /** 187 * Creates a Base64 codec used for decoding (all modes) and encoding in the 188 * given URL-safe mode. 189 * <p> 190 * When encoding the line length is 76, the line separator is CRLF, and the 191 * encoding table is STANDARD_ENCODE_TABLE. 192 * </p> 193 * 194 * <p> 195 * When decoding all variants are supported. 196 * </p> 197 * 198 * @param urlSafe 199 * if {@code true}, URL-safe encoding is used. In most cases this 200 * should be set to {@code false}. 201 * @since 1.4 202 */ 203 public Base64(boolean urlSafe) { 204 this(MIME_CHUNK_SIZE, CHUNK_SEPARATOR, urlSafe); 205 } 206 207 /** 208 * Creates a Base64 codec used for decoding (all modes) and encoding in 209 * URL-unsafe mode. 210 * <p> 211 * When encoding the line length is given in the constructor, the line 212 * separator is CRLF, and the encoding table is STANDARD_ENCODE_TABLE. 213 * </p> 214 * <p> 215 * Line lengths that aren't multiples of 4 will still essentially end up 216 * being multiples of 4 in the encoded data. 217 * </p> 218 * <p> 219 * When decoding all variants are supported. 220 * </p> 221 * 222 * @param lineLength 223 * Each line of encoded data will be at most of the given length 224 * (rounded down to nearest multiple of 4). If lineLength <= 0, 225 * then the output will not be divided into lines (chunks). 226 * Ignored when decoding. 227 * @since 1.4 228 */ 229 public Base64(int lineLength) { 230 this(lineLength, CHUNK_SEPARATOR); 231 } 232 233 /** 234 * Creates a Base64 codec used for decoding (all modes) and encoding in 235 * URL-unsafe mode. 236 * <p> 237 * When encoding the line length and line separator are given in the 238 * constructor, and the encoding table is STANDARD_ENCODE_TABLE. 239 * </p> 240 * <p> 241 * Line lengths that aren't multiples of 4 will still essentially end up 242 * being multiples of 4 in the encoded data. 243 * </p> 244 * <p> 245 * When decoding all variants are supported. 246 * </p> 247 * 248 * @param lineLength 249 * Each line of encoded data will be at most of the given length 250 * (rounded down to nearest multiple of 4). If lineLength <= 0, 251 * then the output will not be divided into lines (chunks). 252 * Ignored when decoding. 253 * @param lineSeparator 254 * Each line of encoded data will end with this sequence of 255 * bytes. 256 * @throws IllegalArgumentException 257 * Thrown when the provided lineSeparator included some base64 258 * characters. 259 * @since 1.4 260 */ 261 public Base64(int lineLength, byte[] lineSeparator) { 262 this(lineLength, lineSeparator, false); 263 } 264 265 /** 266 * Creates a Base64 codec used for decoding (all modes) and encoding in 267 * URL-unsafe mode. 268 * <p> 269 * When encoding the line length and line separator are given in the 270 * constructor, and the encoding table is STANDARD_ENCODE_TABLE. 271 * </p> 272 * <p> 273 * Line lengths that aren't multiples of 4 will still essentially end up 274 * being multiples of 4 in the encoded data. 275 * </p> 276 * <p> 277 * When decoding all variants are supported. 278 * </p> 279 * 280 * @param lineLength 281 * Each line of encoded data will be at most of the given length 282 * (rounded down to nearest multiple of 4). If lineLength <= 0, 283 * then the output will not be divided into lines (chunks). 284 * Ignored when decoding. 285 * @param lineSeparator 286 * Each line of encoded data will end with this sequence of 287 * bytes. 288 * @param urlSafe 289 * Instead of emitting '+' and '/' we emit '-' and '_' 290 * respectively. urlSafe is only applied to encode operations. 291 * Decoding seamlessly handles both modes. 292 * @throws IllegalArgumentException 293 * The provided lineSeparator included some base64 characters. 294 * That's not going to work! 295 * @since 1.4 296 */ 297 public Base64(int lineLength, byte[] lineSeparator, boolean urlSafe) { 298 this(BYTES_PER_UNENCODED_BLOCK, BYTES_PER_ENCODED_BLOCK, lineLength, lineSeparator == null ? 0 : lineSeparator.length); 299 // TODO could be simplified if there is no requirement to reject invalid 300 // line sep when length <=0 301 // @see test case Base64Test.testConstructors() 302 if (lineSeparator != null) { 303 if (containsAlphabetOrPad(lineSeparator)) { 304 String sep = StringUtils.newStringUtf8(lineSeparator); 305 throw new IllegalArgumentException("lineSeparator must not contain base64 characters: [" + sep + "]"); 306 } 307 if (lineLength > 0) { // null line-sep forces no chunking rather 308 // than throwing IAE 309 this.encodeSize = BYTES_PER_ENCODED_BLOCK + lineSeparator.length; 310 this.lineSeparator = new byte[lineSeparator.length]; 311 System.arraycopy(lineSeparator, 0, this.lineSeparator, 0, lineSeparator.length); 312 } else { 313 this.encodeSize = BYTES_PER_ENCODED_BLOCK; 314 this.lineSeparator = null; 315 } 316 } else { 317 this.encodeSize = BYTES_PER_ENCODED_BLOCK; 318 this.lineSeparator = null; 319 } 320 this.decodeSize = this.encodeSize - 1; 321 this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE : STANDARD_ENCODE_TABLE; 322 } 323 324 /** 325 * Returns our current encode mode. True if we're URL-SAFE, false otherwise. 326 * 327 * @return true if we're in URL-SAFE mode, false otherwise. 328 * @since 1.4 329 */ 330 public boolean isUrlSafe() { 331 return this.encodeTable == URL_SAFE_ENCODE_TABLE; 332 } 333 334 /** 335 * <p> 336 * Encodes all of the provided data, starting at inPos, for inAvail bytes. 337 * Must be called at least twice: once with the data to encode, and once 338 * with inAvail set to "-1" to alert encoder that EOF has been reached, so 339 * flush last remaining bytes (if not multiple of 3). 340 * </p> 341 * <p> 342 * Thanks to "commons" project in ws.apache.org for the bitwise operations, 343 * and general approach. 344 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 345 * </p> 346 * 347 * @param in 348 * byte[] array of binary data to base64 encode. 349 * @param inPos 350 * Position to start reading data from. 351 * @param inAvail 352 * Amount of bytes available from input for encoding. 353 * @param context 354 * the context to be used 355 */ 356 void encode(byte[] in, int inPos, int inAvail, Context context) { 357 if (context.eof) { 358 return; 359 } 360 // inAvail < 0 is how we're informed of EOF in the underlying data we're 361 // encoding. 362 if (inAvail < 0) { 363 context.eof = true; 364 if (0 == context.modulus && lineLength == 0) { 365 return; // no leftovers to process and not using chunking 366 } 367 ensureBufferSize(encodeSize, context); 368 int savedPos = context.pos; 369 switch (context.modulus) { // 0-2 370 case 1: // 8 bits = 6 + 2 371 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 2) & MASK_6BITS]; // top 372 // 6 373 // bits 374 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 4) & MASK_6BITS]; // remaining 375 // 2 376 // URL-SAFE skips the padding to further reduce size. 377 if (encodeTable == STANDARD_ENCODE_TABLE) { 378 context.buffer[context.pos++] = PAD; 379 context.buffer[context.pos++] = PAD; 380 } 381 break; 382 383 case 2: // 16 bits = 6 + 6 + 4 384 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 10) & MASK_6BITS]; 385 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 4) & MASK_6BITS]; 386 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 2) & MASK_6BITS]; 387 // URL-SAFE skips the padding to further reduce size. 388 if (encodeTable == STANDARD_ENCODE_TABLE) { 389 context.buffer[context.pos++] = PAD; 390 } 391 break; 392 } 393 context.currentLinePos += context.pos - savedPos; // keep track of 394 // current line 395 // position 396 // if currentPos == 0 we are at the start of a line, so don't add 397 // CRLF 398 if (lineLength > 0 && context.currentLinePos > 0) { 399 System.arraycopy(lineSeparator, 0, context.buffer, context.pos, lineSeparator.length); 400 context.pos += lineSeparator.length; 401 } 402 } else { 403 for (int i = 0; i < inAvail; i++) { 404 ensureBufferSize(encodeSize, context); 405 context.modulus = (context.modulus + 1) % BYTES_PER_UNENCODED_BLOCK; 406 int b = in[inPos++]; 407 if (b < 0) { 408 b += 256; 409 } 410 context.ibitWorkArea = (context.ibitWorkArea << 8) + b; // BITS_PER_BYTE 411 if (0 == context.modulus) { // 3 bytes = 24 bits = 4 * 6 bits to 412 // extract 413 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 18) & MASK_6BITS]; 414 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 12) & MASK_6BITS]; 415 context.buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 6) & MASK_6BITS]; 416 context.buffer[context.pos++] = encodeTable[context.ibitWorkArea & MASK_6BITS]; 417 context.currentLinePos += BYTES_PER_ENCODED_BLOCK; 418 if (lineLength > 0 && lineLength <= context.currentLinePos) { 419 System.arraycopy(lineSeparator, 0, context.buffer, context.pos, lineSeparator.length); 420 context.pos += lineSeparator.length; 421 context.currentLinePos = 0; 422 } 423 } 424 } 425 } 426 } 427 428 /** 429 * <p> 430 * Decodes all of the provided data, starting at inPos, for inAvail bytes. 431 * Should be called at least twice: once with the data to decode, and once 432 * with inAvail set to "-1" to alert decoder that EOF has been reached. The 433 * "-1" call is not necessary when decoding, but it doesn't hurt, either. 434 * </p> 435 * <p> 436 * Ignores all non-base64 characters. This is how chunked (e.g. 76 437 * character) data is handled, since CR and LF are silently ignored, but has 438 * implications for other bytes, too. This method subscribes to the 439 * garbage-in, garbage-out philosophy: it will not check the provided data 440 * for validity. 441 * </p> 442 * <p> 443 * Thanks to "commons" project in ws.apache.org for the bitwise operations, 444 * and general approach. 445 * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ 446 * </p> 447 * 448 * @param in 449 * byte[] array of ascii data to base64 decode. 450 * @param inPos 451 * Position to start reading data from. 452 * @param inAvail 453 * Amount of bytes available from input for encoding. 454 * @param context 455 * the context to be used 456 */ 457 void decode(byte[] in, int inPos, int inAvail, Context context) { 458 if (context.eof) { 459 return; 460 } 461 if (inAvail < 0) { 462 context.eof = true; 463 } 464 for (int i = 0; i < inAvail; i++) { 465 ensureBufferSize(decodeSize, context); 466 byte b = in[inPos++]; 467 if (b == PAD) { 468 // We're done. 469 context.eof = true; 470 break; 471 } else { 472 if (b >= 0 && b < DECODE_TABLE.length) { 473 int result = DECODE_TABLE[b]; 474 if (result >= 0) { 475 context.modulus = (context.modulus + 1) % BYTES_PER_ENCODED_BLOCK; 476 context.ibitWorkArea = (context.ibitWorkArea << BITS_PER_ENCODED_BYTE) + result; 477 if (context.modulus == 0) { 478 context.buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 16) & MASK_8BITS); 479 context.buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS); 480 context.buffer[context.pos++] = (byte) (context.ibitWorkArea & MASK_8BITS); 481 } 482 } 483 } 484 } 485 } 486 487 // Two forms of EOF as far as base64 decoder is concerned: actual 488 // EOF (-1) and first time '=' character is encountered in stream. 489 // This approach makes the '=' padding characters completely optional. 490 if (context.eof && context.modulus != 0) { 491 ensureBufferSize(decodeSize, context); 492 493 // We have some spare bits remaining 494 // Output all whole multiples of 8 bits and ignore the rest 495 switch (context.modulus) { 496 // case 1: // 6 bits - ignore entirely 497 // break; 498 case 2: // 12 bits = 8 + 4 499 context.ibitWorkArea = context.ibitWorkArea >> 4; // dump the 500 // extra 4 501 // bits 502 context.buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS); 503 break; 504 case 3: // 18 bits = 8 + 8 + 2 505 context.ibitWorkArea = context.ibitWorkArea >> 2; // dump 2 bits 506 context.buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS); 507 context.buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS); 508 break; 509 } 510 } 511 } 512 513 /** 514 * Tests a given byte array to see if it contains only valid characters 515 * within the Base64 alphabet. Currently the method treats whitespace as 516 * valid. 517 * 518 * @param arrayOctet 519 * byte array to test 520 * @return {@code true} if all bytes are valid characters in the Base64 521 * alphabet or if the byte array is empty; {@code false}, otherwise 522 * @deprecated 1.5 Use {@link #isBase64(byte[])}, will be removed in 2.0. 523 */ 524 @Deprecated 525 public static boolean isArrayByteBase64(byte[] arrayOctet) { 526 return isBase64(arrayOctet); 527 } 528 529 /** 530 * Returns whether or not the <code>octet</code> is in the base 64 alphabet. 531 * 532 * @param octet 533 * The value to test 534 * @return {@code true} if the value is defined in the the base 64 alphabet, 535 * {@code false} otherwise. 536 * @since 1.4 537 */ 538 public static boolean isBase64(byte octet) { 539 return octet == PAD_DEFAULT || (octet >= 0 && octet < DECODE_TABLE.length && DECODE_TABLE[octet] != -1); 540 } 541 542 /** 543 * Tests a given String to see if it contains only valid characters within 544 * the Base64 alphabet. Currently the method treats whitespace as valid. 545 * 546 * @param base64 547 * String to test 548 * @return {@code true} if all characters in the String are valid characters 549 * in the Base64 alphabet or if the String is empty; {@code false}, 550 * otherwise 551 * @since 1.5 552 */ 553 public static boolean isBase64(String base64) { 554 return isBase64(StringUtils.getBytesUtf8(base64)); 555 } 556 557 /** 558 * Tests a given byte array to see if it contains only valid characters 559 * within the Base64 alphabet. Currently the method treats whitespace as 560 * valid. 561 * 562 * @param arrayOctet 563 * byte array to test 564 * @return {@code true} if all bytes are valid characters in the Base64 565 * alphabet or if the byte array is empty; {@code false}, otherwise 566 * @since 1.5 567 */ 568 public static boolean isBase64(byte[] arrayOctet) { 569 for (byte b : arrayOctet) { 570 if (!isBase64(b) && !isWhiteSpace(b)) { 571 return false; 572 } 573 } 574 return true; 575 } 576 577 /** 578 * Encodes binary data using the base64 algorithm but does not chunk the 579 * output. 580 * 581 * @param binaryData 582 * binary data to encode 583 * @return byte[] containing Base64 characters in their UTF-8 584 * representation. 585 */ 586 public static byte[] encodeBase64(byte[] binaryData) { 587 return encodeBase64(binaryData, false); 588 } 589 590 /** 591 * Encodes binary data using the base64 algorithm but does not chunk the 592 * output. 593 * 594 * NOTE: We changed the behaviour of this method from multi-line chunking 595 * (commons-codec-1.4) to single-line non-chunking (commons-codec-1.5). 596 * 597 * @param binaryData 598 * binary data to encode 599 * @return String containing Base64 characters. 600 * @since 1.4 (NOTE: 1.4 chunked the output, whereas 1.5 does not). 601 */ 602 public static String encodeBase64String(byte[] binaryData) { 603 return StringUtils.newStringUtf8(encodeBase64(binaryData, false)); 604 } 605 606 /** 607 * Encodes binary data using a URL-safe variation of the base64 algorithm 608 * but does not chunk the output. The url-safe variation emits - and _ 609 * instead of + and / characters. 610 * 611 * @param binaryData 612 * binary data to encode 613 * @return byte[] containing Base64 characters in their UTF-8 614 * representation. 615 * @since 1.4 616 */ 617 public static byte[] encodeBase64URLSafe(byte[] binaryData) { 618 return encodeBase64(binaryData, false, true); 619 } 620 621 /** 622 * Encodes binary data using a URL-safe variation of the base64 algorithm 623 * but does not chunk the output. The url-safe variation emits - and _ 624 * instead of + and / characters. 625 * 626 * @param binaryData 627 * binary data to encode 628 * @return String containing Base64 characters 629 * @since 1.4 630 */ 631 public static String encodeBase64URLSafeString(byte[] binaryData) { 632 return StringUtils.newStringUtf8(encodeBase64(binaryData, false, true)); 633 } 634 635 /** 636 * Encodes binary data using the base64 algorithm and chunks the encoded 637 * output into 76 character blocks 638 * 639 * @param binaryData 640 * binary data to encode 641 * @return Base64 characters chunked in 76 character blocks 642 */ 643 public static byte[] encodeBase64Chunked(byte[] binaryData) { 644 return encodeBase64(binaryData, true); 645 } 646 647 /** 648 * Encodes binary data using the base64 algorithm, optionally chunking the 649 * output into 76 character blocks. 650 * 651 * @param binaryData 652 * Array containing binary data to encode. 653 * @param isChunked 654 * if {@code true} this encoder will chunk the base64 output into 655 * 76 character blocks 656 * @return Base64-encoded data. 657 * @throws IllegalArgumentException 658 * Thrown when the input array needs an output array bigger than 659 * {@link Integer#MAX_VALUE} 660 */ 661 public static byte[] encodeBase64(byte[] binaryData, boolean isChunked) { 662 return encodeBase64(binaryData, isChunked, false); 663 } 664 665 /** 666 * Encodes binary data using the base64 algorithm, optionally chunking the 667 * output into 76 character blocks. 668 * 669 * @param binaryData 670 * Array containing binary data to encode. 671 * @param isChunked 672 * if {@code true} this encoder will chunk the base64 output into 673 * 76 character blocks 674 * @param urlSafe 675 * if {@code true} this encoder will emit - and _ instead of the 676 * usual + and / characters. 677 * @return Base64-encoded data. 678 * @throws IllegalArgumentException 679 * Thrown when the input array needs an output array bigger than 680 * {@link Integer#MAX_VALUE} 681 * @since 1.4 682 */ 683 public static byte[] encodeBase64(byte[] binaryData, boolean isChunked, boolean urlSafe) { 684 return encodeBase64(binaryData, isChunked, urlSafe, Integer.MAX_VALUE); 685 } 686 687 /** 688 * Encodes binary data using the base64 algorithm, optionally chunking the 689 * output into 76 character blocks. 690 * 691 * @param binaryData 692 * Array containing binary data to encode. 693 * @param isChunked 694 * if {@code true} this encoder will chunk the base64 output into 695 * 76 character blocks 696 * @param urlSafe 697 * if {@code true} this encoder will emit - and _ instead of the 698 * usual + and / characters. 699 * @param maxResultSize 700 * The maximum result size to accept. 701 * @return Base64-encoded data. 702 * @throws IllegalArgumentException 703 * Thrown when the input array needs an output array bigger than 704 * maxResultSize 705 * @since 1.4 706 */ 707 public static byte[] encodeBase64(byte[] binaryData, boolean isChunked, boolean urlSafe, int maxResultSize) { 708 if (binaryData == null || binaryData.length == 0) { 709 return binaryData; 710 } 711 712 // Create this so can use the super-class method 713 // Also ensures that the same roundings are performed by the ctor and 714 // the code 715 Base64 b64 = isChunked ? new Base64package/Base64.html#Base64">Base64(urlSafe) : new Base64(0, CHUNK_SEPARATOR, urlSafe); 716 long len = b64.getEncodedLength(binaryData); 717 if (len > maxResultSize) { 718 throw new IllegalArgumentException("Input array too big, the output array would be bigger (" + len + ") than the specified maximum size of " + maxResultSize); 719 } 720 721 return b64.encode(binaryData); 722 } 723 724 /** 725 * Decodes a Base64 String into octets 726 * 727 * @param base64String 728 * String containing Base64 data 729 * @return Array containing decoded data. 730 * @since 1.4 731 */ 732 public static byte[] decodeBase64(String base64String) { 733 return new Base64().decode(base64String); 734 } 735 736 /** 737 * Decodes Base64 data into octets 738 * 739 * @param base64Data 740 * Byte array containing Base64 data 741 * @return Array containing decoded data. 742 */ 743 public static byte[] decodeBase64(byte[] base64Data) { 744 return new Base64().decode(base64Data); 745 } 746 747 // Implementation of the Encoder Interface 748 749 // Implementation of integer encoding used for crypto 750 /** 751 * Decodes a byte64-encoded integer according to crypto standards such as 752 * W3C's XML-Signature 753 * 754 * @param pArray 755 * a byte array containing base64 character data 756 * @return A BigInteger 757 * @since 1.4 758 */ 759 public static BigInteger decodeInteger(byte[] pArray) { 760 return new BigInteger(1, decodeBase64(pArray)); 761 } 762 763 /** 764 * Encodes to a byte64-encoded integer according to crypto standards such as 765 * W3C's XML-Signature 766 * 767 * @param bigInt 768 * a BigInteger 769 * @return A byte array containing base64 character data 770 * @throws NullPointerException 771 * if null is passed in 772 * @since 1.4 773 */ 774 public static byte[] encodeInteger(BigInteger bigInt) { 775 if (bigInt == null) { 776 throw new NullPointerException("encodeInteger called with null parameter"); 777 } 778 return encodeBase64(toIntegerBytes(bigInt), false); 779 } 780 781 /** 782 * Returns a byte-array representation of a <code>BigInteger</code> without 783 * sign bit. 784 * 785 * @param bigInt 786 * <code>BigInteger</code> to be converted 787 * @return a byte array representation of the BigInteger parameter 788 */ 789 static byte[] toIntegerBytes(BigInteger bigInt) { 790 int bitlen = bigInt.bitLength(); 791 // round bitlen 792 bitlen = ((bitlen + 7) >> 3) << 3; 793 byte[] bigBytes = bigInt.toByteArray(); 794 795 if (((bigInt.bitLength() % 8) != 0) && (((bigInt.bitLength() / 8) + 1) == (bitlen / 8))) { 796 return bigBytes; 797 } 798 // set up params for copying everything but sign bit 799 int startSrc = 0; 800 int len = bigBytes.length; 801 802 // if bigInt is exactly byte-aligned, just skip signbit in copy 803 if ((bigInt.bitLength() % 8) == 0) { 804 startSrc = 1; 805 len--; 806 } 807 int startDst = bitlen / 8 - len; // to pad w/ nulls as per spec 808 byte[] resizedBytes = new byte[bitlen / 8]; 809 System.arraycopy(bigBytes, startSrc, resizedBytes, startDst, len); 810 return resizedBytes; 811 } 812 813 /** 814 * Returns whether or not the <code>octet</code> is in the Base32 alphabet. 815 * 816 * @param octet 817 * The value to test 818 * @return {@code true} if the value is defined in the the Base32 alphabet 819 * {@code false} otherwise. 820 */ 821 protected boolean isInAlphabet(byte octet) { 822 return octet >= 0 && octet < decodeTable.length && decodeTable[octet] != -1; 823 } 824 825 /** 826 * Holds thread context so classes can be thread-safe. 827 * 828 * This class is not itself thread-safe; each thread must allocate its own 829 * copy. 830 * 831 * @since 1.7 832 */ 833 static class Context { 834 835 /** 836 * Place holder for the bytes we're dealing with for our based logic. 837 * Bitwise operations store and extract the encoding or decoding from 838 * this variable. 839 */ 840 int ibitWorkArea; 841 842 /** 843 * Place holder for the bytes we're dealing with for our based logic. 844 * Bitwise operations store and extract the encoding or decoding from 845 * this variable. 846 */ 847 long lbitWorkArea; 848 849 /** 850 * Buffer for streaming. 851 */ 852 byte[] buffer; 853 854 /** 855 * Position where next character should be written in the buffer. 856 */ 857 int pos; 858 859 /** 860 * Position where next character should be read from the buffer. 861 */ 862 int readPos; 863 864 /** 865 * Boolean flag to indicate the EOF has been reached. Once EOF has been 866 * reached, this object becomes useless, and must be thrown away. 867 */ 868 boolean eof; 869 870 /** 871 * Variable tracks how many characters have been written to the current 872 * line. Only used when encoding. We use it to make sure each encoded 873 * line never goes beyond lineLength (if lineLength > 0). 874 */ 875 int currentLinePos; 876 877 /** 878 * Writes to the buffer only occur after every 3/5 reads when encoding, 879 * and every 4/8 reads when decoding. This variable helps track that. 880 */ 881 int modulus; 882 883 Context() { 884 } 885 } 886 887 /** 888 * EOF 889 * 890 * @since 1.7 891 */ 892 static final int EOF = -1; 893 894 /** 895 * MIME chunk size per RFC 2045 section 6.8. 896 * 897 * <p> 898 * The {@value} character limit does not count the trailing CRLF, but counts 899 * all other characters, including any equal signs. 900 * </p> 901 * 902 * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 903 * 6.8</a> 904 */ 905 public static final int MIME_CHUNK_SIZE = 76; 906 907 /** 908 * PEM chunk size per RFC 1421 section 4.3.2.4. 909 * 910 * <p> 911 * The {@value} character limit does not count the trailing CRLF, but counts 912 * all other characters, including any equal signs. 913 * </p> 914 * 915 * @see <a href="http://tools.ietf.org/html/rfc1421">RFC 1421 section 916 * 4.3.2.4</a> 917 */ 918 public static final int PEM_CHUNK_SIZE = 64; 919 920 private static final int DEFAULT_BUFFER_RESIZE_FACTOR = 2; 921 922 /** 923 * Defines the default buffer size - currently {@value} - must be large 924 * enough for at least one encoded block+separator 925 */ 926 private static final int DEFAULT_BUFFER_SIZE = 8192; 927 928 /** Mask used to extract 8 bits, used in decoding bytes */ 929 protected static final int MASK_8BITS = 0xff; 930 931 /** 932 * Byte used to pad output. 933 */ 934 protected static final byte PAD_DEFAULT = '='; // Allow static access to 935 // default 936 937 protected final byte PAD = PAD_DEFAULT; // instance variable just in case it 938 // needs to vary later 939 940 /** 941 * Number of bytes in each full block of unencoded data, e.g. 4 for Base64 942 * and 5 for Base32 943 */ 944 private final int unencodedBlockSize; 945 946 /** 947 * Number of bytes in each full block of encoded data, e.g. 3 for Base64 and 948 * 8 for Base32 949 */ 950 private final int encodedBlockSize; 951 952 /** 953 * Chunksize for encoding. Not used when decoding. A value of zero or less 954 * implies no chunking of the encoded data. Rounded down to nearest multiple 955 * of encodedBlockSize. 956 */ 957 protected final int lineLength; 958 959 /** 960 * Size of chunk separator. Not used unless {@link #lineLength} > 0. 961 */ 962 private final int chunkSeparatorLength; 963 964 /** 965 * Note <code>lineLength</code> is rounded down to the nearest multiple of 966 * {@link #encodedBlockSize} If <code>chunkSeparatorLength</code> is zero, 967 * then chunking is disabled. 968 * 969 * @param unencodedBlockSize 970 * the size of an unencoded block (e.g. Base64 = 3) 971 * @param encodedBlockSize 972 * the size of an encoded block (e.g. Base64 = 4) 973 * @param lineLength 974 * if > 0, use chunking with a length <code>lineLength</code> 975 * @param chunkSeparatorLength 976 * the chunk separator length, if relevant 977 */ 978 protected Base64(int unencodedBlockSize, int encodedBlockSize, int lineLength, int chunkSeparatorLength) { 979 this.unencodedBlockSize = unencodedBlockSize; 980 this.encodedBlockSize = encodedBlockSize; 981 this.lineLength = (lineLength > 0 && chunkSeparatorLength > 0) ? (lineLength / encodedBlockSize) * encodedBlockSize : 0; 982 this.chunkSeparatorLength = chunkSeparatorLength; 983 } 984 985 /** 986 * Returns true if this object has buffered data for reading. 987 * 988 * @param context 989 * the context to be used 990 * @return true if there is data still available for reading. 991 */ 992 boolean hasData(Context context) { // package protected for access from I/O 993 // streams 994 return context.buffer != null; 995 } 996 997 /** 998 * Returns the amount of buffered data available for reading. 999 * 1000 * @param context 1001 * the context to be used 1002 * @return The amount of buffered data available for reading. 1003 */ 1004 int available(Context context) { // package protected for access from I/O 1005 // streams 1006 return context.buffer != null ? context.pos - context.readPos : 0; 1007 } 1008 1009 /** 1010 * Get the default buffer size. Can be overridden. 1011 * 1012 * @return {@link #DEFAULT_BUFFER_SIZE} 1013 */ 1014 protected int getDefaultBufferSize() { 1015 return DEFAULT_BUFFER_SIZE; 1016 } 1017 1018 /** 1019 * Increases our buffer by the {@link #DEFAULT_BUFFER_RESIZE_FACTOR}. 1020 * 1021 * @param context 1022 * the context to be used 1023 */ 1024 private void resizeBuffer(Context context) { 1025 if (context.buffer == null) { 1026 context.buffer = new byte[getDefaultBufferSize()]; 1027 context.pos = 0; 1028 context.readPos = 0; 1029 } else { 1030 byte[] b = new byte[context.buffer.length * DEFAULT_BUFFER_RESIZE_FACTOR]; 1031 System.arraycopy(context.buffer, 0, b, 0, context.buffer.length); 1032 context.buffer = b; 1033 } 1034 } 1035 1036 /** 1037 * Ensure that the buffer has room for <code>size</code> bytes 1038 * 1039 * @param size 1040 * minimum spare space required 1041 * @param context 1042 * the context to be used 1043 */ 1044 protected void ensureBufferSize(int size, Context context) { 1045 if ((context.buffer == null) || (context.buffer.length < context.pos + size)) { 1046 resizeBuffer(context); 1047 } 1048 } 1049 1050 /** 1051 * Extracts buffered data into the provided byte[] array, starting at 1052 * position bPos, up to a maximum of bAvail bytes. Returns how many bytes 1053 * were actually extracted. 1054 * 1055 * @param b 1056 * byte[] array to extract the buffered data into. 1057 * @param bPos 1058 * position in byte[] array to start extraction at. 1059 * @param bAvail 1060 * amount of bytes we're allowed to extract. We may extract fewer 1061 * (if fewer are available). 1062 * @param context 1063 * the context to be used 1064 * @return The number of bytes successfully extracted into the provided 1065 * byte[] array. 1066 */ 1067 int readResults(byte[] b, int bPos, int bAvail, Context context) { // package 1068 // protected 1069 // for 1070 // access 1071 // from 1072 // I/O 1073 // streams 1074 if (context.buffer != null) { 1075 int len = Math.min(available(context), bAvail); 1076 System.arraycopy(context.buffer, context.readPos, b, bPos, len); 1077 context.readPos += len; 1078 if (context.readPos >= context.pos) { 1079 context.buffer = null; // so hasData() will return false, and 1080 // this method can return -1 1081 } 1082 return len; 1083 } 1084 return context.eof ? EOF : 0; 1085 } 1086 1087 /** 1088 * Checks if a byte value is whitespace or not. Whitespace is taken to mean: 1089 * space, tab, CR, LF 1090 * 1091 * @param byteToCheck 1092 * the byte to check 1093 * @return true if byte is whitespace, false otherwise 1094 */ 1095 protected static boolean isWhiteSpace(byte byteToCheck) { 1096 switch (byteToCheck) { 1097 case ' ': 1098 case '\n': 1099 case '\r': 1100 case '\t': 1101 return true; 1102 default: 1103 return false; 1104 } 1105 } 1106 1107 /** 1108 * Encodes an Object using the Base-N algorithm. This method is provided in 1109 * order to satisfy the requirements of the Encoder interface, and will 1110 * throw an EncoderException if the supplied object is not of type byte[]. 1111 * 1112 * @param obj 1113 * Object to encode 1114 * @return An object (of type byte[]) containing the Base-N encoded data 1115 * which corresponds to the byte[] supplied. 1116 * @throws Exception 1117 * if the parameter supplied is not of type byte[] 1118 */ 1119 public Object encode(Object obj) throws Exception { 1120 if (!(obj instanceof byte[])) { 1121 throw new Exception("Parameter supplied to Base-N encode is not a byte[]"); 1122 } 1123 return encode((byte[]) obj); 1124 } 1125 1126 /** 1127 * Encodes a byte[] containing binary data, into a String containing 1128 * characters in the Base-N alphabet. Uses UTF8 encoding. 1129 * 1130 * @param pArray 1131 * a byte array containing binary data 1132 * @return A String containing only Base-N character data 1133 */ 1134 public String encodeToString(byte[] pArray) { 1135 return StringUtils.newStringUtf8(encode(pArray)); 1136 } 1137 1138 /** 1139 * Encodes a byte[] containing binary data, into a String containing 1140 * characters in the appropriate alphabet. Uses UTF8 encoding. 1141 * 1142 * @param pArray 1143 * a byte array containing binary data 1144 * @return String containing only character data in the appropriate 1145 * alphabet. 1146 */ 1147 public String encodeAsString(byte[] pArray) { 1148 return StringUtils.newStringUtf8(encode(pArray)); 1149 } 1150 1151 /** 1152 * Decodes an Object using the Base-N algorithm. This method is provided in 1153 * order to satisfy the requirements of the Decoder interface, and will 1154 * throw a DecoderException if the supplied object is not of type byte[] or 1155 * String. 1156 * 1157 * @param obj 1158 * Object to decode 1159 * @return An object (of type byte[]) containing the binary data which 1160 * corresponds to the byte[] or String supplied. 1161 * @throws Exception 1162 * if the parameter supplied is not of type byte[] 1163 */ 1164 public Object decode(Object obj) throws Exception { 1165 if (obj instanceof byte[]) { 1166 return decode((byte[]) obj); 1167 } else if (obj instanceof String) { 1168 return decode((String) obj); 1169 } else { 1170 throw new Exception("Parameter supplied to Base-N decode is not a byte[] or a String"); 1171 } 1172 } 1173 1174 /** 1175 * Decodes a String containing characters in the Base-N alphabet. 1176 * 1177 * @param pArray 1178 * A String containing Base-N character data 1179 * @return a byte array containing binary data 1180 */ 1181 public byte[] decode(String pArray) { 1182 return decode(StringUtils.getBytesUtf8(pArray)); 1183 } 1184 1185 /** 1186 * Decodes a byte[] containing characters in the Base-N alphabet. 1187 * 1188 * @param pArray 1189 * A byte array containing Base-N character data 1190 * @return a byte array containing binary data 1191 */ 1192 public byte[] decode(byte[] pArray) { 1193 Context context = new Context(); 1194 if (pArray == null || pArray.length == 0) { 1195 return pArray; 1196 } 1197 decode(pArray, 0, pArray.length, context); 1198 decode(pArray, 0, EOF, context); // Notify decoder of EOF. 1199 byte[] result = new byte[context.pos]; 1200 readResults(result, 0, result.length, context); 1201 return result; 1202 } 1203 1204 /** 1205 * Encodes a byte[] containing binary data, into a byte[] containing 1206 * characters in the alphabet. 1207 * 1208 * @param pArray 1209 * a byte array containing binary data 1210 * @return A byte array containing only the basen alphabetic character data 1211 */ 1212 public byte[] encode(byte[] pArray) { 1213 Context context = new Context(); 1214 if (pArray == null || pArray.length == 0) { 1215 return pArray; 1216 } 1217 encode(pArray, 0, pArray.length, context); 1218 encode(pArray, 0, EOF, context); // Notify encoder of EOF. 1219 byte[] buf = new byte[context.pos - context.readPos]; 1220 readResults(buf, 0, buf.length, context); 1221 return buf; 1222 } 1223 1224 /** 1225 * Tests a given byte array to see if it contains only valid characters 1226 * within the alphabet. The method optionally treats whitespace and pad as 1227 * valid. 1228 * 1229 * @param arrayOctet 1230 * byte array to test 1231 * @param allowWSPad 1232 * if {@code true}, then whitespace and PAD are also allowed 1233 * 1234 * @return {@code true} if all bytes are valid characters in the alphabet or 1235 * if the byte array is empty; {@code false}, otherwise 1236 */ 1237 public boolean isInAlphabet(byte[] arrayOctet, boolean allowWSPad) { 1238 for (byte b : arrayOctet) { 1239 if (!isInAlphabet(b) && (!allowWSPad || (b != PAD) && !isWhiteSpace(b))) { 1240 return false; 1241 } 1242 } 1243 return true; 1244 } 1245 1246 /** 1247 * Tests a given String to see if it contains only valid characters within 1248 * the alphabet. The method treats whitespace and PAD as valid. 1249 * 1250 * @param basen 1251 * String to test 1252 * @return {@code true} if all characters in the String are valid characters 1253 * in the alphabet or if the String is empty; {@code false}, 1254 * otherwise 1255 * @see #isInAlphabet(byte[], boolean) 1256 */ 1257 public boolean isInAlphabet(String basen) { 1258 return isInAlphabet(StringUtils.getBytesUtf8(basen), true); 1259 } 1260 1261 /** 1262 * Tests a given byte array to see if it contains any characters within the 1263 * alphabet or PAD. 1264 * 1265 * Intended for use in checking line-ending arrays 1266 * 1267 * @param arrayOctet 1268 * byte array to test 1269 * @return {@code true} if any byte is a valid character in the alphabet or 1270 * PAD; {@code false} otherwise 1271 */ 1272 protected boolean containsAlphabetOrPad(byte[] arrayOctet) { 1273 if (arrayOctet == null) { 1274 return false; 1275 } 1276 for (byte element : arrayOctet) { 1277 if (PAD == element || isInAlphabet(element)) { 1278 return true; 1279 } 1280 } 1281 return false; 1282 } 1283 1284 /** 1285 * Calculates the amount of space needed to encode the supplied array. 1286 * 1287 * @param pArray 1288 * byte[] array which will later be encoded 1289 * 1290 * @return amount of space needed to encoded the supplied array. Returns a 1291 * long since a max-len array will require > Integer.MAX_VALUE 1292 */ 1293 public long getEncodedLength(byte[] pArray) { 1294 // Calculate non-chunked size - rounded up to allow for padding 1295 // cast to long is needed to avoid possibility of overflow 1296 long len = ((pArray.length + unencodedBlockSize - 1) / unencodedBlockSize) * (long) encodedBlockSize; 1297 if (lineLength > 0) { // We're using chunking 1298 // Round up to nearest multiple 1299 len += ((len + lineLength - 1) / lineLength) * chunkSeparatorLength; 1300 } 1301 return len; 1302 } 1303 1304 /** 1305 * Character encoding names required of every implementation of the Java 1306 * platform. 1307 * 1308 * From the Java documentation <a href= 1309 * "http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html" 1310 * >Standard charsets</a>: 1311 * <p> 1312 * <cite>Every implementation of the Java platform is required to support 1313 * the following character encodings. Consult the release documentation for 1314 * your implementation to see if any other encodings are supported. Consult 1315 * the release documentation for your implementation to see if any other 1316 * encodings are supported. </cite> 1317 * </p> 1318 * 1319 * <ul> 1320 * <li><code>US-ASCII</code><br/> 1321 * Seven-bit ASCII, a.k.a. ISO646-US, a.k.a. the Basic Latin block of the 1322 * Unicode character set.</li> 1323 * <li><code>ISO-8859-1</code><br/> 1324 * ISO Latin Alphabet No. 1, a.k.a. ISO-LATIN-1.</li> 1325 * <li><code>UTF-8</code><br/> 1326 * Eight-bit Unicode Transformation Format.</li> 1327 * <li><code>UTF-16BE</code><br/> 1328 * Sixteen-bit Unicode Transformation Format, big-endian byte order.</li> 1329 * <li><code>UTF-16LE</code><br/> 1330 * Sixteen-bit Unicode Transformation Format, little-endian byte order.</li> 1331 * <li><code>UTF-16</code><br/> 1332 * Sixteen-bit Unicode Transformation Format, byte order specified by a 1333 * mandatory initial byte-order mark (either order accepted on input, 1334 * big-endian used on output.)</li> 1335 * </ul> 1336 * 1337 * This perhaps would best belong in the [lang] project. Even if a similar 1338 * interface is defined in [lang], it is not foreseen that [codec] would be 1339 * made to depend on [lang]. 1340 * 1341 * <p> 1342 * This class is immutable and thread-safe. 1343 * </p> 1344 * 1345 * @see <a 1346 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1347 * charsets</a> 1348 * @since 1.4 1349 * @version $Id$ 1350 */ 1351 public static class CharEncoding { 1352 /** 1353 * CharEncodingISO Latin Alphabet No. 1, a.k.a. ISO-LATIN-1. </p> 1354 * <p> 1355 * Every implementation of the Java platform is required to support this 1356 * character encoding. 1357 * </p> 1358 * 1359 * @see <a 1360 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1361 * charsets</a> 1362 */ 1363 public static final String ISO_8859_1 = "ISO-8859-1"; 1364 1365 /** 1366 * <p> 1367 * Seven-bit ASCII, also known as ISO646-US, also known as the Basic 1368 * Latin block of the Unicode character set. 1369 * </p> 1370 * <p> 1371 * Every implementation of the Java platform is required to support this 1372 * character encoding. 1373 * </p> 1374 * 1375 * @see <a 1376 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1377 * charsets</a> 1378 */ 1379 public static final String US_ASCII = "US-ASCII"; 1380 1381 /** 1382 * <p> 1383 * Sixteen-bit Unicode Transformation Format, The byte order specified 1384 * by a mandatory initial byte-order mark (either order accepted on 1385 * input, big-endian used on output) 1386 * </p> 1387 * <p> 1388 * Every implementation of the Java platform is required to support this 1389 * character encoding. 1390 * </p> 1391 * 1392 * @see <a 1393 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1394 * charsets</a> 1395 */ 1396 public static final String UTF_16 = "UTF-16"; 1397 1398 /** 1399 * <p> 1400 * Sixteen-bit Unicode Transformation Format, big-endian byte order. 1401 * </p> 1402 * <p> 1403 * Every implementation of the Java platform is required to support this 1404 * character encoding. 1405 * </p> 1406 * 1407 * @see <a 1408 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1409 * charsets</a> 1410 */ 1411 public static final String UTF_16BE = "UTF-16BE"; 1412 1413 /** 1414 * <p> 1415 * Sixteen-bit Unicode Transformation Format, little-endian byte order. 1416 * </p> 1417 * <p> 1418 * Every implementation of the Java platform is required to support this 1419 * character encoding. 1420 * </p> 1421 * 1422 * @see <a 1423 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1424 * charsets</a> 1425 */ 1426 public static final String UTF_16LE = "UTF-16LE"; 1427 1428 /** 1429 * <p> 1430 * Eight-bit Unicode Transformation Format. 1431 * </p> 1432 * <p> 1433 * Every implementation of the Java platform is required to support this 1434 * character encoding. 1435 * </p> 1436 * 1437 * @see <a 1438 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1439 * charsets</a> 1440 */ 1441 public static final String UTF_8 = "UTF-8"; 1442 } 1443 1444 /** 1445 * Charsets required of every implementation of the Java platform. 1446 * 1447 * From the Java documentation <a href= 1448 * "http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html" 1449 * >Standard charsets</a>: 1450 * <p> 1451 * <cite>Every implementation of the Java platform is required to support 1452 * the following character encodings. Consult the release documentation for 1453 * your implementation to see if any other encodings are supported. Consult 1454 * the release documentation for your implementation to see if any other 1455 * encodings are supported. </cite> 1456 * </p> 1457 * 1458 * <ul> 1459 * <li><code>US-ASCII</code><br/> 1460 * Seven-bit ASCII, a.k.a. ISO646-US, a.k.a. the Basic Latin block of the 1461 * Unicode character set.</li> 1462 * <li><code>ISO-8859-1</code><br/> 1463 * ISO Latin Alphabet No. 1, a.k.a. ISO-LATIN-1.</li> 1464 * <li><code>UTF-8</code><br/> 1465 * Eight-bit Unicode Transformation Format.</li> 1466 * <li><code>UTF-16BE</code><br/> 1467 * Sixteen-bit Unicode Transformation Format, big-endian byte order.</li> 1468 * <li><code>UTF-16LE</code><br/> 1469 * Sixteen-bit Unicode Transformation Format, little-endian byte order.</li> 1470 * <li><code>UTF-16</code><br/> 1471 * Sixteen-bit Unicode Transformation Format, byte order specified by a 1472 * mandatory initial byte-order mark (either order accepted on input, 1473 * big-endian used on output.)</li> 1474 * </ul> 1475 * 1476 * This perhaps would best belong in the Commons Lang project. Even if a 1477 * similar class is defined in Commons Lang, it is not foreseen that Commons 1478 * Codec would be made to depend on Commons Lang. 1479 * 1480 * <p> 1481 * This class is immutable and thread-safe. 1482 * </p> 1483 * 1484 * @see <a 1485 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1486 * charsets</a> 1487 * @since 1.7 1488 * @version $Id: CharEncoding.java 1173287 2011-09-20 18:16:19Z ggregory $ 1489 */ 1490 public static class Charsets { 1491 1492 // 1493 // This class should only contain Charset instances for required 1494 // encodings. This guarantees that it will load correctly and 1495 // without delay on all Java platforms. 1496 // 1497 1498 /** 1499 * Returns the given Charset or the default Charset if the given Charset 1500 * is null. 1501 * 1502 * @param charset 1503 * A charset or null. 1504 * @return the given Charset or the default Charset if the given Charset 1505 * is null 1506 */ 1507 public static Charset toCharset(Charset charset) { 1508 return charset == null ? Charset.defaultCharset() : charset; 1509 } 1510 1511 /** 1512 * Returns a Charset for the named charset. If the name is null, return 1513 * the default Charset. 1514 * 1515 * @param charset 1516 * The name of the requested charset, may be null. 1517 * @return a Charset for the named charset 1518 * @throws UnsupportedCharsetException 1519 * If the named charset is unavailable 1520 */ 1521 public static Charset toCharset(String charset) { 1522 return charset == null ? Charset.defaultCharset() : Charset.forName(charset); 1523 } 1524 1525 /** 1526 * CharEncodingISO Latin Alphabet No. 1, a.k.a. ISO-LATIN-1. </p> 1527 * <p> 1528 * Every implementation of the Java platform is required to support this 1529 * character encoding. 1530 * </p> 1531 * 1532 * @see <a 1533 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1534 * charsets</a> 1535 */ 1536 public static final Charset ISO_8859_1 = StandardCharsets.ISO_8859_1; 1537 1538 /** 1539 * <p> 1540 * Seven-bit ASCII, also known as ISO646-US, also known as the Basic 1541 * Latin block of the Unicode character set. 1542 * </p> 1543 * <p> 1544 * Every implementation of the Java platform is required to support this 1545 * character encoding. 1546 * </p> 1547 * 1548 * @see <a 1549 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1550 * charsets</a> 1551 */ 1552 public static final Charset US_ASCII = StandardCharsets.US_ASCII; 1553 1554 /** 1555 * <p> 1556 * Sixteen-bit Unicode Transformation Format, The byte order specified 1557 * by a mandatory initial byte-order mark (either order accepted on 1558 * input, big-endian used on output) 1559 * </p> 1560 * <p> 1561 * Every implementation of the Java platform is required to support this 1562 * character encoding. 1563 * </p> 1564 * 1565 * @see <a 1566 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1567 * charsets</a> 1568 */ 1569 public static final Charset UTF_16 = StandardCharsets.UTF_16; 1570 1571 /** 1572 * <p> 1573 * Sixteen-bit Unicode Transformation Format, big-endian byte order. 1574 * </p> 1575 * <p> 1576 * Every implementation of the Java platform is required to support this 1577 * character encoding. 1578 * </p> 1579 * 1580 * @see <a 1581 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1582 * charsets</a> 1583 */ 1584 public static final Charset UTF_16BE = StandardCharsets.UTF_16BE; 1585 1586 /** 1587 * <p> 1588 * Sixteen-bit Unicode Transformation Format, little-endian byte order. 1589 * </p> 1590 * <p> 1591 * Every implementation of the Java platform is required to support this 1592 * character encoding. 1593 * </p> 1594 * 1595 * @see <a 1596 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1597 * charsets</a> 1598 */ 1599 public static final Charset UTF_16LE = StandardCharsets.UTF_16LE; 1600 1601 /** 1602 * <p> 1603 * Eight-bit Unicode Transformation Format. 1604 * </p> 1605 * <p> 1606 * Every implementation of the Java platform is required to support this 1607 * character encoding. 1608 * </p> 1609 * 1610 * @see <a 1611 * href="http://docs.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1612 * charsets</a> 1613 */ 1614 public static final Charset UTF_8 = StandardCharsets.UTF_8; 1615 } 1616 1617 /** 1618 * Converts String to and from bytes using the encodings required by the 1619 * Java specification. These encodings are specified in <a href= 1620 * "http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html" 1621 * >Standard charsets</a> 1622 * 1623 * <p> 1624 * This class is immutable and thread-safe. 1625 * </p> 1626 * 1627 * @see CharEncoding 1628 * @see <a 1629 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1630 * charsets</a> 1631 * @version $Id$ 1632 * @since 1.4 1633 */ 1634 public static class StringUtils { 1635 1636 /** 1637 * Calls {@link String#getBytes(Charset)} 1638 * 1639 * @param string 1640 * The string to encode (if null, return null). 1641 * @param charset 1642 * The {@link Charset} to encode the {@code String} 1643 * @return the encoded bytes 1644 */ 1645 private static byte[] getBytes(String string, Charset charset) { 1646 if (string == null) { 1647 return null; 1648 } 1649 return string.getBytes(charset); 1650 } 1651 1652 /** 1653 * Encodes the given string into a sequence of bytes using the 1654 * ISO-8859-1 charset, storing the result into a new byte array. 1655 * 1656 * @param string 1657 * the String to encode, may be {@code null} 1658 * @return encoded bytes, or {@code null} if the input string was 1659 * {@code null} 1660 * @throws NullPointerException 1661 * Thrown if {@link Charsets#ISO_8859_1} is not initialized, 1662 * which should never happen since it is required by the 1663 * Java platform specification. 1664 * @since As of 1.7, throws {@link NullPointerException} instead of 1665 * UnsupportedEncodingException 1666 * @see <a 1667 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1668 * charsets</a> 1669 * @see #getBytesUnchecked(String, String) 1670 */ 1671 public static byte[] getBytesIso8859_1(String string) { 1672 return getBytes(string, Charsets.ISO_8859_1); 1673 } 1674 1675 /** 1676 * Encodes the given string into a sequence of bytes using the named 1677 * charset, storing the result into a new byte array. 1678 * <p> 1679 * This method catches {@link UnsupportedEncodingException} and rethrows 1680 * it as {@link IllegalStateException}, which should never happen for a 1681 * required charset name. Use this method when the encoding is required 1682 * to be in the JRE. 1683 * </p> 1684 * 1685 * @param string 1686 * the String to encode, may be {@code null} 1687 * @param charsetName 1688 * The name of a required {@link java.nio.charset.Charset} 1689 * @return encoded bytes, or {@code null} if the input string was 1690 * {@code null} 1691 * @throws IllegalStateException 1692 * Thrown when a {@link UnsupportedEncodingException} is 1693 * caught, which should never happen for a required charset 1694 * name. 1695 * @see CharEncoding 1696 * @see String#getBytes(String) 1697 */ 1698 public static byte[] getBytesUnchecked(String string, String charsetName) { 1699 if (string == null) { 1700 return null; 1701 } 1702 try { 1703 return string.getBytes(charsetName); 1704 } catch (UnsupportedEncodingException e) { 1705 throw StringUtils.newIllegalStateException(charsetName, e); 1706 } 1707 } 1708 1709 /** 1710 * Encodes the given string into a sequence of bytes using the US-ASCII 1711 * charset, storing the result into a new byte array. 1712 * 1713 * @param string 1714 * the String to encode, may be {@code null} 1715 * @return encoded bytes, or {@code null} if the input string was 1716 * {@code null} 1717 * @throws NullPointerException 1718 * Thrown if {@link Charsets#US_ASCII} is not initialized, 1719 * which should never happen since it is required by the 1720 * Java platform specification. 1721 * @since As of 1.7, throws {@link NullPointerException} instead of 1722 * UnsupportedEncodingException 1723 * @see <a 1724 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1725 * charsets</a> 1726 * @see #getBytesUnchecked(String, String) 1727 */ 1728 public static byte[] getBytesUsAscii(String string) { 1729 return getBytes(string, Charsets.US_ASCII); 1730 } 1731 1732 /** 1733 * Encodes the given string into a sequence of bytes using the UTF-16 1734 * charset, storing the result into a new byte array. 1735 * 1736 * @param string 1737 * the String to encode, may be {@code null} 1738 * @return encoded bytes, or {@code null} if the input string was 1739 * {@code null} 1740 * @throws NullPointerException 1741 * Thrown if {@link Charsets#UTF_16} is not initialized, 1742 * which should never happen since it is required by the 1743 * Java platform specification. 1744 * @since As of 1.7, throws {@link NullPointerException} instead of 1745 * UnsupportedEncodingException 1746 * @see <a 1747 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1748 * charsets</a> 1749 * @see #getBytesUnchecked(String, String) 1750 */ 1751 public static byte[] getBytesUtf16(String string) { 1752 return getBytes(string, Charsets.UTF_16); 1753 } 1754 1755 /** 1756 * Encodes the given string into a sequence of bytes using the UTF-16BE 1757 * charset, storing the result into a new byte array. 1758 * 1759 * @param string 1760 * the String to encode, may be {@code null} 1761 * @return encoded bytes, or {@code null} if the input string was 1762 * {@code null} 1763 * @throws NullPointerException 1764 * Thrown if {@link Charsets#UTF_16BE} is not initialized, 1765 * which should never happen since it is required by the 1766 * Java platform specification. 1767 * @since As of 1.7, throws {@link NullPointerException} instead of 1768 * UnsupportedEncodingException 1769 * @see <a 1770 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1771 * charsets</a> 1772 * @see #getBytesUnchecked(String, String) 1773 */ 1774 public static byte[] getBytesUtf16Be(String string) { 1775 return getBytes(string, Charsets.UTF_16BE); 1776 } 1777 1778 /** 1779 * Encodes the given string into a sequence of bytes using the UTF-16LE 1780 * charset, storing the result into a new byte array. 1781 * 1782 * @param string 1783 * the String to encode, may be {@code null} 1784 * @return encoded bytes, or {@code null} if the input string was 1785 * {@code null} 1786 * @throws NullPointerException 1787 * Thrown if {@link Charsets#UTF_16LE} is not initialized, 1788 * which should never happen since it is required by the 1789 * Java platform specification. 1790 * @since As of 1.7, throws {@link NullPointerException} instead of 1791 * UnsupportedEncodingException 1792 * @see <a 1793 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1794 * charsets</a> 1795 * @see #getBytesUnchecked(String, String) 1796 */ 1797 public static byte[] getBytesUtf16Le(String string) { 1798 return getBytes(string, Charsets.UTF_16LE); 1799 } 1800 1801 /** 1802 * Encodes the given string into a sequence of bytes using the UTF-8 1803 * charset, storing the result into a new byte array. 1804 * 1805 * @param string 1806 * the String to encode, may be {@code null} 1807 * @return encoded bytes, or {@code null} if the input string was 1808 * {@code null} 1809 * @throws NullPointerException 1810 * Thrown if {@link Charsets#UTF_8} is not initialized, 1811 * which should never happen since it is required by the 1812 * Java platform specification. 1813 * @since As of 1.7, throws {@link NullPointerException} instead of 1814 * UnsupportedEncodingException 1815 * @see <a 1816 * href="http://download.oracle.com/javase/6/docs/api/java/nio/charset/Charset.html">Standard 1817 * charsets</a> 1818 * @see #getBytesUnchecked(String, String) 1819 */ 1820 public static byte[] getBytesUtf8(String string) { 1821 return getBytes(string, Charsets.UTF_8); 1822 } 1823 1824 private static IllegalStateException newIllegalStateException(String charsetName, UnsupportedEncodingException e) { 1825 return new IllegalStateException(charsetName + ": " + e); 1826 } 1827 1828 /** 1829 * Constructs a new <code>String</code> by decoding the specified array 1830 * of bytes using the given charset. 1831 * 1832 * @param bytes 1833 * The bytes to be decoded into characters 1834 * @param charset 1835 * The {@link Charset} to encode the {@code String} 1836 * @return A new <code>String</code> decoded from the specified array of 1837 * bytes using the given charset, or {@code null} if the input 1838 * byte array was {@code null}. 1839 * @throws NullPointerException 1840 * Thrown if {@link Charsets#UTF_8} is not initialized, 1841 * which should never happen since it is required by the 1842 * Java platform specification. 1843 */ 1844 private static String newString(byte[] bytes, Charset charset) { 1845 return bytes == null ? null : new String(bytes, charset); 1846 } 1847 1848 /** 1849 * Constructs a new <code>String</code> by decoding the specified array 1850 * of bytes using the given charset. 1851 * <p> 1852 * This method catches {@link UnsupportedEncodingException} and 1853 * re-throws it as {@link IllegalStateException}, which should never 1854 * happen for a required charset name. Use this method when the encoding 1855 * is required to be in the JRE. 1856 * </p> 1857 * 1858 * @param bytes 1859 * The bytes to be decoded into characters, may be 1860 * {@code null} 1861 * @param charsetName 1862 * The name of a required {@link java.nio.charset.Charset} 1863 * @return A new <code>String</code> decoded from the specified array of 1864 * bytes using the given charset, or {@code null} if the input 1865 * byte array was {@code null}. 1866 * @throws IllegalStateException 1867 * Thrown when a {@link UnsupportedEncodingException} is 1868 * caught, which should never happen for a required charset 1869 * name. 1870 * @see CharEncoding 1871 * @see String#String(byte[], String) 1872 */ 1873 public static String newString(byte[] bytes, String charsetName) { 1874 if (bytes == null) { 1875 return null; 1876 } 1877 try { 1878 return new String(bytes, charsetName); 1879 } catch (UnsupportedEncodingException e) { 1880 throw StringUtils.newIllegalStateException(charsetName, e); 1881 } 1882 } 1883 1884 /** 1885 * Constructs a new <code>String</code> by decoding the specified array 1886 * of bytes using the ISO-8859-1 charset. 1887 * 1888 * @param bytes 1889 * The bytes to be decoded into characters, may be 1890 * {@code null} 1891 * @return A new <code>String</code> decoded from the specified array of 1892 * bytes using the ISO-8859-1 charset, or {@code null} if the 1893 * input byte array was {@code null}. 1894 * @throws NullPointerException 1895 * Thrown if {@link Charsets#ISO_8859_1} is not initialized, 1896 * which should never happen since it is required by the 1897 * Java platform specification. 1898 * @since As of 1.7, throws {@link NullPointerException} instead of 1899 * UnsupportedEncodingException 1900 */ 1901 public static String newStringIso8859_1(byte[] bytes) { 1902 return new String(bytes, Charsets.ISO_8859_1); 1903 } 1904 1905 /** 1906 * Constructs a new <code>String</code> by decoding the specified array 1907 * of bytes using the US-ASCII charset. 1908 * 1909 * @param bytes 1910 * The bytes to be decoded into characters 1911 * @return A new <code>String</code> decoded from the specified array of 1912 * bytes using the US-ASCII charset, or {@code null} if the 1913 * input byte array was {@code null}. 1914 * @throws NullPointerException 1915 * Thrown if {@link Charsets#US_ASCII} is not initialized, 1916 * which should never happen since it is required by the 1917 * Java platform specification. 1918 * @since As of 1.7, throws {@link NullPointerException} instead of 1919 * UnsupportedEncodingException 1920 */ 1921 public static String newStringUsAscii(byte[] bytes) { 1922 return new String(bytes, Charsets.US_ASCII); 1923 } 1924 1925 /** 1926 * Constructs a new <code>String</code> by decoding the specified array 1927 * of bytes using the UTF-16 charset. 1928 * 1929 * @param bytes 1930 * The bytes to be decoded into characters 1931 * @return A new <code>String</code> decoded from the specified array of 1932 * bytes using the UTF-16 charset or {@code null} if the input 1933 * byte array was {@code null}. 1934 * @throws NullPointerException 1935 * Thrown if {@link Charsets#UTF_16} is not initialized, 1936 * which should never happen since it is required by the 1937 * Java platform specification. 1938 * @since As of 1.7, throws {@link NullPointerException} instead of 1939 * UnsupportedEncodingException 1940 */ 1941 public static String newStringUtf16(byte[] bytes) { 1942 return new String(bytes, Charsets.UTF_16); 1943 } 1944 1945 /** 1946 * Constructs a new <code>String</code> by decoding the specified array 1947 * of bytes using the UTF-16BE charset. 1948 * 1949 * @param bytes 1950 * The bytes to be decoded into characters 1951 * @return A new <code>String</code> decoded from the specified array of 1952 * bytes using the UTF-16BE charset, or {@code null} if the 1953 * input byte array was {@code null}. 1954 * @throws NullPointerException 1955 * Thrown if {@link Charsets#UTF_16BE} is not initialized, 1956 * which should never happen since it is required by the 1957 * Java platform specification. 1958 * @since As of 1.7, throws {@link NullPointerException} instead of 1959 * UnsupportedEncodingException 1960 */ 1961 public static String newStringUtf16Be(byte[] bytes) { 1962 return new String(bytes, Charsets.UTF_16BE); 1963 } 1964 1965 /** 1966 * Constructs a new <code>String</code> by decoding the specified array 1967 * of bytes using the UTF-16LE charset. 1968 * 1969 * @param bytes 1970 * The bytes to be decoded into characters 1971 * @return A new <code>String</code> decoded from the specified array of 1972 * bytes using the UTF-16LE charset, or {@code null} if the 1973 * input byte array was {@code null}. 1974 * @throws NullPointerException 1975 * Thrown if {@link Charsets#UTF_16LE} is not initialized, 1976 * which should never happen since it is required by the 1977 * Java platform specification. 1978 * @since As of 1.7, throws {@link NullPointerException} instead of 1979 * UnsupportedEncodingException 1980 */ 1981 public static String newStringUtf16Le(byte[] bytes) { 1982 return new String(bytes, Charsets.UTF_16LE); 1983 } 1984 1985 /** 1986 * Constructs a new <code>String</code> by decoding the specified array 1987 * of bytes using the UTF-8 charset. 1988 * 1989 * @param bytes 1990 * The bytes to be decoded into characters 1991 * @return A new <code>String</code> decoded from the specified array of 1992 * bytes using the UTF-8 charset, or {@code null} if the input 1993 * byte array was {@code null}. 1994 * @throws NullPointerException 1995 * Thrown if {@link Charsets#UTF_8} is not initialized, 1996 * which should never happen since it is required by the 1997 * Java platform specification. 1998 * @since As of 1.7, throws {@link NullPointerException} instead of 1999 * UnsupportedEncodingException 2000 */ 2001 public static String newStringUtf8(byte[] bytes) { 2002 return newString(bytes, Charsets.UTF_8); 2003 } 2004 2005 } 2006 2007 }