001/* 002 * Licensed to the Apache Software Foundation (ASF) under one or more 003 * contributor license agreements. See the NOTICE file distributed with 004 * this work for additional information regarding copyright ownership. 005 * The ASF licenses this file to You under the Apache License, Version 2.0 006 * (the "License"); you may not use this file except in compliance with 007 * the License. You may obtain a copy of the License at 008 * 009 * https://www.apache.org/licenses/LICENSE-2.0 010 * 011 * Unless required by applicable law or agreed to in writing, software 012 * distributed under the License is distributed on an "AS IS" BASIS, 013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 014 * See the License for the specific language governing permissions and 015 * limitations under the License. 016 */ 017 018package org.apache.commons.codec.binary; 019 020import java.math.BigInteger; 021import java.util.ArrayList; 022import java.util.Arrays; 023import java.util.Objects; 024import java.util.function.Supplier; 025 026import org.apache.commons.codec.BinaryDecoder; 027import org.apache.commons.codec.BinaryEncoder; 028import org.apache.commons.codec.CodecPolicy; 029import org.apache.commons.codec.DecoderException; 030import org.apache.commons.codec.EncoderException; 031 032/** 033 * Abstract superclass for Base-N encoders and decoders. 034 * 035 * <p> 036 * This class is thread-safe. 037 * </p> 038 * <p> 039 * You can set the decoding behavior when the input bytes contain leftover trailing bits that cannot be created by a valid encoding. These can be bits that are 040 * unused from the final character or entire characters. The default mode is lenient decoding. 041 * </p> 042 * <ul> 043 * <li>Lenient: Any trailing bits are composed into 8-bit bytes where possible. The remainder are discarded.</li> 044 * <li>Strict: The decoding will raise an {@link IllegalArgumentException} if trailing bits are not part of a valid encoding. Any unused bits from the final 045 * character must be zero. Impossible counts of entire final characters are not allowed.</li> 046 * </ul> 047 * <p> 048 * When strict decoding is enabled it is expected that the decoded bytes will be re-encoded to a byte array that matches the original, i.e. no changes occur on 049 * the final character. This requires that the input bytes use the same padding and alphabet as the encoder. 050 * </p> 051 */ 052public abstract class BaseNCodec implements BinaryEncoder, BinaryDecoder { 053 054 /** 055 * Builds {@link Base64} instances. 056 * 057 * @param <T> The codec type to build. 058 * @param <B> The codec builder subtype. 059 * @since 1.17.0 060 */ 061 public abstract static class AbstractBuilder<T, B extends AbstractBuilder<T, B>> implements Supplier<T> { 062 063 /** 064 * Clones the given array or returns a default array if the array is null. 065 * 066 * @param array The array to test and clone if not null. 067 * @param defaultArray The default array to return if the array is null. 068 * @return A clone of the array or the default array if the array is null. 069 */ 070 static byte[] clone(final byte[] array, final byte[] defaultArray) { 071 return array != null ? array.clone() : defaultArray; 072 } 073 074 private int unencodedBlockSize; 075 private int encodedBlockSize; 076 private CodecPolicy decodingPolicy = DECODING_POLICY_DEFAULT; 077 private int lineLength; 078 private byte[] lineSeparator = CHUNK_SEPARATOR; 079 private final byte[] defaultEncodeTable; 080 private byte[] encodeTable; 081 private byte[] decodeTable; 082 083 /** Padding byte. */ 084 private byte padding = PAD_DEFAULT; 085 086 AbstractBuilder(final byte[] defaultEncodeTable) { 087 this.defaultEncodeTable = defaultEncodeTable; 088 this.encodeTable = defaultEncodeTable; 089 } 090 091 /** 092 * Returns this instance typed as the subclass type {@code B}. 093 * <p> 094 * This is the same as the expression: 095 * </p> 096 * 097 * <pre> 098 * (B) this 099 * </pre> 100 * 101 * @return {@code this} instance typed as the subclass type {@code B}. 102 */ 103 @SuppressWarnings("unchecked") 104 B asThis() { 105 return (B) this; 106 } 107 108 byte[] getDecodeTable() { 109 return decodeTable; 110 } 111 112 CodecPolicy getDecodingPolicy() { 113 return decodingPolicy; 114 } 115 116 int getEncodedBlockSize() { 117 return encodedBlockSize; 118 } 119 120 byte[] getEncodeTable() { 121 return encodeTable; 122 } 123 124 int getLineLength() { 125 return lineLength; 126 } 127 128 byte[] getLineSeparator() { 129 return lineSeparator; 130 } 131 132 byte getPadding() { 133 return padding; 134 } 135 136 int getUnencodedBlockSize() { 137 return unencodedBlockSize; 138 } 139 140 /** 141 * Sets the decode table. 142 * 143 * @param decodeTable The decode table. 144 * @return {@code this} instance. 145 * @since 1.20.0 146 */ 147 public B setDecodeTable(final byte[] decodeTable) { 148 this.decodeTable = clone(decodeTable, null); 149 return asThis(); 150 } 151 152 /** 153 * Sets the decode table. 154 * 155 * @param decodeTable The decode table, null resets to the default. 156 * @return {@code this} instance. 157 */ 158 B setDecodeTableRaw(final byte[] decodeTable) { 159 this.decodeTable = decodeTable; 160 return asThis(); 161 } 162 163 /** 164 * Sets the decoding policy. 165 * 166 * @param decodingPolicy The decoding policy, null resets to the default. 167 * @return {@code this} instance. 168 */ 169 public B setDecodingPolicy(final CodecPolicy decodingPolicy) { 170 this.decodingPolicy = decodingPolicy != null ? decodingPolicy : DECODING_POLICY_DEFAULT; 171 return asThis(); 172 } 173 174 /** 175 * Sets the encoded block size, subclasses normally set this on construction. 176 * 177 * @param encodedBlockSize The encoded block size, subclasses normally set this on construction. 178 * @return {@code this} instance. 179 */ 180 B setEncodedBlockSize(final int encodedBlockSize) { 181 this.encodedBlockSize = gte0(encodedBlockSize); 182 return asThis(); 183 } 184 185 /** 186 * Sets the encode table. 187 * 188 * @param encodeTable The encode table, null resets to the default. 189 * @return {@code this} instance. 190 */ 191 public B setEncodeTable(final byte... encodeTable) { 192 this.encodeTable = clone(encodeTable, defaultEncodeTable); 193 return asThis(); 194 } 195 196 /** 197 * Sets the encode table. 198 * 199 * @param encodeTable The encode table, null resets to the default. 200 * @return {@code this} instance. 201 */ 202 B setEncodeTableRaw(final byte... encodeTable) { 203 this.encodeTable = encodeTable != null ? encodeTable : defaultEncodeTable; 204 return asThis(); 205 } 206 207 /** 208 * Sets the line length. 209 * 210 * @param lineLength The line length, less than 0 resets to the default. 211 * @return {@code this} instance. 212 */ 213 public B setLineLength(final int lineLength) { 214 this.lineLength = Math.max(0, lineLength); 215 return asThis(); 216 } 217 218 /** 219 * Sets the line separator. 220 * 221 * @param lineSeparator The line separator, null resets to the default. 222 * @return {@code this} instance. 223 */ 224 public B setLineSeparator(final byte... lineSeparator) { 225 this.lineSeparator = clone(lineSeparator , CHUNK_SEPARATOR); 226 return asThis(); 227 } 228 229 /** 230 * Sets the padding byte. 231 * 232 * @param padding The padding byte. 233 * @return {@code this} instance. 234 */ 235 public B setPadding(final byte padding) { 236 this.padding = padding; 237 return asThis(); 238 } 239 240 /** 241 * Sets the unencoded block size, subclasses normally set this on construction. 242 * 243 * @param unencodedBlockSize The unencoded block size, subclasses normally set this on construction. 244 * @return {@code this} instance. 245 */ 246 B setUnencodedBlockSize(final int unencodedBlockSize) { 247 this.unencodedBlockSize = gte0(unencodedBlockSize); 248 return asThis(); 249 } 250 } 251 252 /** 253 * Holds thread context so classes can be thread-safe. 254 * 255 * This class is not itself thread-safe; each thread must allocate its own copy. 256 */ 257 static class Context { 258 259 /** 260 * Placeholder for the bytes we're dealing with for our based logic. Bitwise operations store and extract the encoding or decoding from this variable. 261 */ 262 int ibitWorkArea; 263 264 /** 265 * Placeholder for the bytes we're dealing with for our based logic. Bitwise operations store and extract the encoding or decoding from this variable. 266 */ 267 long lbitWorkArea; 268 269 /** 270 * Buffer for streaming. 271 */ 272 byte[] buffer; 273 274 /** 275 * Position where next character should be written in the buffer. 276 */ 277 int pos; 278 279 /** 280 * Position where next character should be read from the buffer. 281 */ 282 int readPos; 283 284 /** 285 * Boolean flag to indicate the EOF has been reached. Once EOF has been reached, this object becomes useless, and must be thrown away. 286 */ 287 boolean eof; 288 289 /** 290 * 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 291 * goes beyond lineLength (if lineLength > 0). 292 */ 293 int currentLinePos; 294 295 /** 296 * Writes to the buffer only occur after every 3/5 reads when encoding, and every 4/8 reads when decoding. This variable helps track that. 297 */ 298 int modulus; 299 300 /** 301 * Returns a String useful for debugging (especially within a debugger.) 302 * 303 * @return A String useful for debugging. 304 */ 305 @Override 306 public String toString() { 307 return String.format("%s[buffer=%s, currentLinePos=%s, eof=%s, ibitWorkArea=%s, lbitWorkArea=%s, " + "modulus=%s, pos=%s, readPos=%s]", 308 this.getClass().getSimpleName(), Arrays.toString(buffer), currentLinePos, eof, ibitWorkArea, lbitWorkArea, modulus, pos, readPos); 309 } 310 } 311 312 /** 313 * End-of-file marker. 314 * 315 * @since 1.7 316 */ 317 static final int EOF = -1; 318 319 /** 320 * MIME chunk size per RFC 2045 section 6.8. 321 * 322 * <p> 323 * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any equal signs. 324 * </p> 325 * 326 * @see <a href="https://www.ietf.org/rfc/rfc2045">RFC 2045 section 6.8</a> 327 */ 328 public static final int MIME_CHUNK_SIZE = 76; 329 330 /** 331 * PEM chunk size per RFC 1421 section 4.3.2.4. 332 * 333 * <p> 334 * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any equal signs. 335 * </p> 336 * 337 * @see <a href="https://tools.ietf.org/html/rfc1421">RFC 1421 section 4.3.2.4</a> 338 */ 339 public static final int PEM_CHUNK_SIZE = 64; 340 private static final int DEFAULT_BUFFER_RESIZE_FACTOR = 2; 341 342 /** 343 * Defines the default buffer size - currently {@value} - must be large enough for at least one encoded block+separator 344 */ 345 private static final int DEFAULT_BUFFER_SIZE = 8192; 346 347 /** 348 * The maximum size buffer to allocate. 349 * 350 * <p> 351 * This is set to the same size used in the JDK {@link ArrayList}: 352 * </p> 353 * <blockquote> Some VMs reserve some header words in an array. Attempts to allocate larger arrays may result in OutOfMemoryError: Requested array size 354 * exceeds VM limit. </blockquote> 355 */ 356 private static final int MAX_BUFFER_SIZE = Integer.MAX_VALUE - 8; 357 358 /** Mask used to extract 8 bits, used in decoding bytes */ 359 protected static final int MASK_8BITS = 0xff; 360 361 /** 362 * Byte used to pad output. 363 */ 364 protected static final byte PAD_DEFAULT = '='; // Allow static access to default 365 366 /** 367 * The default decoding policy. 368 * 369 * @since 1.15 370 */ 371 protected static final CodecPolicy DECODING_POLICY_DEFAULT = CodecPolicy.LENIENT; 372 373 /** 374 * Chunk separator per RFC 2045 section 2.1. 375 * 376 * @see <a href="https://www.ietf.org/rfc/rfc2045">RFC 2045 section 2.1</a> 377 */ 378 static final byte[] CHUNK_SEPARATOR = { '\r', '\n' }; 379 380 /** 381 * The empty byte array. 382 */ 383 static final byte[] EMPTY_BYTE_ARRAY = {}; 384 385 /** 386 * Create a positive capacity at least as large the minimum required capacity. If the minimum capacity is negative then this throws an OutOfMemoryError as 387 * no array can be allocated. 388 * 389 * @param minCapacity The minimum capacity. 390 * @return The capacity. 391 * @throws OutOfMemoryError if the {@code minCapacity} is negative. 392 */ 393 private static int createPositiveCapacity(final int minCapacity) { 394 if (minCapacity < 0) { 395 // overflow 396 throw new OutOfMemoryError("Unable to allocate array size: " + (minCapacity & 0xffffffffL)); 397 } 398 // This is called when we require buffer expansion to a very big array. 399 // Use the conservative maximum buffer size if possible, otherwise the biggest required. 400 // 401 // Note: In this situation JDK 1.8 java.util.ArrayList returns Integer.MAX_VALUE. 402 // This excludes some VMs that can exceed MAX_BUFFER_SIZE but not allocate a full 403 // Integer.MAX_VALUE length array. 404 // The result is that we may have to allocate an array of this size more than once if 405 // the capacity must be expanded again. 406 return Math.max(minCapacity, MAX_BUFFER_SIZE); 407 } 408 409 /** 410 * Gets a copy of the chunk separator per RFC 2045 section 2.1. 411 * 412 * @return The chunk separator. 413 * @see <a href="https://www.ietf.org/rfc/rfc2045">RFC 2045 section 2.1</a> 414 * @since 1.15 415 */ 416 public static byte[] getChunkSeparator() { 417 return CHUNK_SEPARATOR.clone(); 418 } 419 420 private static int gte0(final int value) { 421 if (value < 0) { 422 throw new IllegalArgumentException("value must be greater than or equal to 0."); 423 } 424 return value; 425 } 426 427 /** 428 * Checks if a byte value is whitespace or not. 429 * 430 * @param byteToCheck The byte to check. 431 * @return true if byte is whitespace, false otherwise. 432 * @see Character#isWhitespace(int) 433 * @deprecated Use {@link Character#isWhitespace(int)}. 434 */ 435 @Deprecated 436 protected static boolean isWhiteSpace(final byte byteToCheck) { 437 return Character.isWhitespace(byteToCheck); 438 } 439 440 /** 441 * Increases our buffer by the {@link #DEFAULT_BUFFER_RESIZE_FACTOR}. 442 * 443 * @param context The context to be used. 444 * @param minCapacity The minimum required capacity. 445 * @return The resized byte[] buffer. 446 * @throws OutOfMemoryError if the {@code minCapacity} is negative. 447 */ 448 private static byte[] resizeBuffer(final Context context, final int minCapacity) { 449 // Overflow-conscious code treats the min and new capacity as unsigned. 450 final int oldCapacity = context.buffer.length; 451 int newCapacity = oldCapacity * DEFAULT_BUFFER_RESIZE_FACTOR; 452 if (Integer.compareUnsigned(newCapacity, minCapacity) < 0) { 453 newCapacity = minCapacity; 454 } 455 if (Integer.compareUnsigned(newCapacity, MAX_BUFFER_SIZE) > 0) { 456 newCapacity = createPositiveCapacity(minCapacity); 457 } 458 final byte[] b = Arrays.copyOf(context.buffer, newCapacity); 459 context.buffer = b; 460 return b; 461 } 462 463 /** 464 * Returns a byte-array representation of a {@code BigInteger} without sign bit. 465 * <p> 466 * The value {@link BigInteger#ZERO} maps to an empty array. 467 * </p> 468 * 469 * @param bigInt {@code BigInteger} to be converted. 470 * @return A byte array representation of the BigInteger parameter. 471 */ 472 static byte[] toUnsignedBytes(final BigInteger value) { 473 byte[] unsigned = value.equals(BigInteger.ZERO) ? EMPTY_BYTE_ARRAY : value.toByteArray(); 474 if (unsigned.length > 0 && unsigned[0] == 0) { 475 final byte[] tmp = new byte[unsigned.length - 1]; 476 System.arraycopy(unsigned, 1, tmp, 0, tmp.length); 477 unsigned = tmp; 478 } 479 return unsigned; 480 } 481 482 /** 483 * Deprecated: Will be removed in 2.0. 484 * <p> 485 * Instance variable just in case it needs to vary later 486 * </p> 487 * 488 * @deprecated Use {@link #pad}. Will be removed in 2.0. 489 */ 490 @Deprecated 491 protected final byte PAD = PAD_DEFAULT; 492 493 /** Pad byte. Instance variable just in case it needs to vary later. */ 494 protected final byte pad; 495 496 /** Number of bytes in each full block of unencoded data, for example 4 for Base64 and 5 for Base32 */ 497 private final int unencodedBlockSize; 498 499 /** Number of bytes in each full block of encoded data, for example 3 for Base64 and 8 for Base32 */ 500 private final int encodedBlockSize; 501 502 /** 503 * Chunksize for encoding. Not used when decoding. A value of zero or less implies no chunking of the encoded data. Rounded down to the nearest multiple of 504 * encodedBlockSize. 505 */ 506 protected final int lineLength; 507 508 /** 509 * Size of chunk separator. Not used unless {@link #lineLength} > 0. 510 */ 511 private final int chunkSeparatorLength; 512 513 /** 514 * Defines the decoding behavior when the input bytes contain leftover trailing bits that cannot be created by a valid encoding. These can be bits that are 515 * unused from the final character or entire characters. The default mode is lenient decoding. Set this to {@code true} to enable strict decoding. 516 * <ul> 517 * <li>Lenient: Any trailing bits are composed into 8-bit bytes where possible. The remainder are discarded.</li> 518 * <li>Strict: The decoding will raise an {@link IllegalArgumentException} if trailing bits are not part of a valid encoding. Any unused bits from the final 519 * character must be zero. Impossible counts of entire final characters are not allowed.</li> 520 * </ul> 521 * <p> 522 * When strict decoding is enabled it is expected that the decoded bytes will be re-encoded to a byte array that matches the original, i.e. no changes occur 523 * on the final character. This requires that the input bytes use the same padding and alphabet as the encoder. 524 * </p> 525 */ 526 private final CodecPolicy decodingPolicy; 527 528 /** 529 * Decode table to use. 530 */ 531 final byte[] decodeTable; 532 533 /** 534 * Encode table. 535 */ 536 final byte[] encodeTable; 537 538 /** 539 * Constructs a new instance for a subclass. 540 * 541 * @param builder How to build this portion of the instance. 542 * @since 1.20.0 543 */ 544 protected BaseNCodec(final AbstractBuilder<?, ?> builder) { 545 this.unencodedBlockSize = gte0(builder.unencodedBlockSize); 546 this.encodedBlockSize = gte0(builder.encodedBlockSize); 547 final boolean useChunking = builder.lineLength > 0 && builder.lineSeparator.length > 0; 548 this.lineLength = useChunking ? builder.lineLength / builder.encodedBlockSize * builder.encodedBlockSize : 0; 549 this.chunkSeparatorLength = builder.lineSeparator.length; 550 this.pad = builder.padding; 551 this.decodingPolicy = Objects.requireNonNull(builder.decodingPolicy, "codecPolicy"); 552 this.encodeTable = Objects.requireNonNull(builder.getEncodeTable(), "builder.getEncodeTable()"); 553 this.decodeTable = builder.getDecodeTable(); 554 } 555 556 /** 557 * Constructs a new instance. 558 * <p> 559 * Note {@code lineLength} is rounded down to the nearest multiple of the encoded block size. If {@code chunkSeparatorLength} is zero, then chunking is 560 * disabled. 561 * </p> 562 * 563 * @param unencodedBlockSize The size of an unencoded block (for example Base64 = 3). 564 * @param encodedBlockSize The size of an encoded block (for example Base64 = 4). 565 * @param lineLength if > 0, use chunking with a length {@code lineLength}. 566 * @param chunkSeparatorLength The chunk separator length, if relevant. 567 * @deprecated Use {@link BaseNCodec#BaseNCodec(AbstractBuilder)}. 568 */ 569 @Deprecated 570 protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize, final int lineLength, final int chunkSeparatorLength) { 571 this(unencodedBlockSize, encodedBlockSize, lineLength, chunkSeparatorLength, PAD_DEFAULT); 572 } 573 574 /** 575 * Constructs a new instance. 576 * <p> 577 * Note {@code lineLength} is rounded down to the nearest multiple of the encoded block size. If {@code chunkSeparatorLength} is zero, then chunking is 578 * disabled. 579 * </p> 580 * 581 * @param unencodedBlockSize The size of an unencoded block (for example Base64 = 3). 582 * @param encodedBlockSize The size of an encoded block (for example Base64 = 4). 583 * @param lineLength if > 0, use chunking with a length {@code lineLength}. 584 * @param chunkSeparatorLength The chunk separator length, if relevant. 585 * @param pad byte used as padding byte. 586 * @deprecated Use {@link BaseNCodec#BaseNCodec(AbstractBuilder)}. 587 */ 588 @Deprecated 589 protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize, final int lineLength, final int chunkSeparatorLength, final byte pad) { 590 this(unencodedBlockSize, encodedBlockSize, lineLength, chunkSeparatorLength, pad, DECODING_POLICY_DEFAULT); 591 } 592 593 /** 594 * Constructs a new instance. 595 * <p> 596 * Note {@code lineLength} is rounded down to the nearest multiple of the encoded block size. If {@code chunkSeparatorLength} is zero, then chunking is 597 * disabled. 598 * </p> 599 * 600 * @param unencodedBlockSize The size of an unencoded block (for example Base64 = 3). 601 * @param encodedBlockSize The size of an encoded block (for example Base64 = 4). 602 * @param lineLength if > 0, use chunking with a length {@code lineLength}. 603 * @param chunkSeparatorLength The chunk separator length, if relevant. 604 * @param pad byte used as padding byte. 605 * @param decodingPolicy Decoding policy. 606 * @since 1.15 607 * @deprecated Use {@link BaseNCodec#BaseNCodec(AbstractBuilder)}. 608 */ 609 @Deprecated 610 protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize, final int lineLength, final int chunkSeparatorLength, final byte pad, 611 final CodecPolicy decodingPolicy) { 612 this.unencodedBlockSize = unencodedBlockSize; 613 this.encodedBlockSize = encodedBlockSize; 614 final boolean useChunking = lineLength > 0 && chunkSeparatorLength > 0; 615 this.lineLength = useChunking ? lineLength / encodedBlockSize * encodedBlockSize : 0; 616 this.chunkSeparatorLength = chunkSeparatorLength; 617 this.pad = pad; 618 this.decodingPolicy = Objects.requireNonNull(decodingPolicy, "codecPolicy"); 619 this.encodeTable = null; 620 this.decodeTable = null; 621 } 622 623 /** 624 * Returns the amount of buffered data available for reading. 625 * 626 * @param context The context to be used. 627 * @return The amount of buffered data available for reading. 628 */ 629 int available(final Context context) { // package protected for access from I/O streams 630 return hasData(context) ? context.pos - context.readPos : 0; 631 } 632 633 /** 634 * Tests a given byte array to see if it contains any characters within the alphabet or PAD. 635 * 636 * Intended for use in checking line-ending arrays. 637 * 638 * @param arrayOctet byte array to test. 639 * @return {@code true} if any byte is a valid character in the alphabet or PAD; {@code false} otherwise. 640 */ 641 protected boolean containsAlphabetOrPad(final byte[] arrayOctet) { 642 if (arrayOctet != null) { 643 for (final byte element : arrayOctet) { 644 if (pad == element || isInAlphabet(element)) { 645 return true; 646 } 647 } 648 } 649 return false; 650 } 651 652 /** 653 * Decodes a byte[] containing characters in the Base-N alphabet. 654 * 655 * @param array A byte array containing Base-N character data. 656 * @return A byte array containing binary data. 657 */ 658 @Override 659 public byte[] decode(final byte[] array) { 660 if (BinaryCodec.isEmpty(array)) { 661 return array; 662 } 663 final Context context = new Context(); 664 decode(array, 0, array.length, context); 665 decode(array, 0, EOF, context); // Notify decoder of EOF. 666 final byte[] result = new byte[context.pos]; 667 readResults(result, 0, result.length, context); 668 return result; 669 } 670 671 // package protected for access from I/O streams 672 abstract void decode(byte[] array, int i, int length, Context context); 673 674 /** 675 * Decodes an Object using the Base-N algorithm. This method is provided in order to satisfy the requirements of the Decoder interface, and will throw a 676 * DecoderException if the supplied object is not of type byte[] or String. 677 * 678 * @param obj Object to decode. 679 * @return An object (of type byte[]) containing the binary data which corresponds to the byte[] or String supplied. 680 * @throws DecoderException if the parameter supplied is not of type byte[]. 681 */ 682 @Override 683 public Object decode(final Object obj) throws DecoderException { 684 if (obj instanceof byte[]) { 685 return decode((byte[]) obj); 686 } 687 if (obj instanceof String) { 688 return decode((String) obj); 689 } 690 throw new DecoderException("Parameter supplied to Base-N decode is not a byte[] or a String"); 691 } 692 693 /** 694 * Decodes a String containing characters in the Base-N alphabet. 695 * 696 * @param array A String containing Base-N character data. 697 * @return A byte array containing binary data. 698 */ 699 public byte[] decode(final String array) { 700 return decode(StringUtils.getBytesUtf8(array)); 701 } 702 703 /** 704 * Encodes a byte[] containing binary data, into a byte[] containing characters in the alphabet. 705 * 706 * @param array A byte array containing binary data. 707 * @return A byte array containing only the base N alphabetic character data. 708 */ 709 @Override 710 public byte[] encode(final byte[] array) { 711 if (BinaryCodec.isEmpty(array)) { 712 return array; 713 } 714 return encode(array, 0, array.length); 715 } 716 717 /** 718 * Encodes a byte[] containing binary data, into a byte[] containing characters in the alphabet. 719 * 720 * @param array A byte array containing binary data. 721 * @param offset initial offset of the subarray. 722 * @param length length of the subarray. 723 * @return A byte array containing only the base N alphabetic character data. 724 * @since 1.11 725 */ 726 public byte[] encode(final byte[] array, final int offset, final int length) { 727 if (BinaryCodec.isEmpty(array)) { 728 return array; 729 } 730 final Context context = new Context(); 731 encode(array, offset, length, context); 732 encode(array, offset, EOF, context); // Notify encoder of EOF. 733 final byte[] buf = new byte[context.pos - context.readPos]; 734 readResults(buf, 0, buf.length, context); 735 return buf; 736 } 737 738 // package protected for access from I/O streams 739 abstract void encode(byte[] array, int i, int length, Context context); 740 741 /** 742 * Encodes an Object using the Base-N algorithm. This method is provided in order to satisfy the requirements of the Encoder interface, and will throw an 743 * EncoderException if the supplied object is not of type byte[]. 744 * 745 * @param obj Object to encode. 746 * @return An object (of type byte[]) containing the Base-N encoded data which corresponds to the byte[] supplied. 747 * @throws EncoderException if the parameter supplied is not of type byte[]. 748 */ 749 @Override 750 public Object encode(final Object obj) throws EncoderException { 751 if (!(obj instanceof byte[])) { 752 throw new EncoderException("Parameter supplied to Base-N encode is not a byte[]"); 753 } 754 return encode((byte[]) obj); 755 } 756 757 /** 758 * Encodes a byte[] containing binary data, into a String containing characters in the appropriate alphabet. Uses UTF8 encoding. 759 * <p> 760 * This is a duplicate of {@link #encodeToString(byte[])}; it was merged during refactoring. 761 * </p> 762 * 763 * @param array A byte array containing binary data. 764 * @return String containing only character data in the appropriate alphabet. 765 * @since 1.5 766 */ 767 public String encodeAsString(final byte[] array) { 768 return StringUtils.newStringUtf8(encode(array)); 769 } 770 771 /** 772 * Encodes a byte[] containing binary data, into a String containing characters in the Base-N alphabet. Uses UTF8 encoding. 773 * 774 * @param array A byte array containing binary data. 775 * @return A String containing only Base-N character data. 776 */ 777 public String encodeToString(final byte[] array) { 778 return StringUtils.newStringUtf8(encode(array)); 779 } 780 781 /** 782 * Ensures that the buffer has room for {@code size} bytes 783 * 784 * @param size minimum spare space required. 785 * @param context The context to be used. 786 * @return The buffer. 787 */ 788 protected byte[] ensureBufferSize(final int size, final Context context) { 789 if (context.buffer == null) { 790 context.buffer = new byte[Math.max(size, getDefaultBufferSize())]; 791 context.pos = 0; 792 context.readPos = 0; 793 // Overflow-conscious: 794 // x + y > z == x + y - z > 0 795 } else if (context.pos + size - context.buffer.length > 0) { 796 return resizeBuffer(context, context.pos + size); 797 } 798 return context.buffer; 799 } 800 801 /** 802 * Gets the decoding behavior policy. 803 * 804 * <p> 805 * The default is lenient. If the decoding policy is strict, then decoding will raise an {@link IllegalArgumentException} if trailing bits are not part of a 806 * valid encoding. Decoding will compose trailing bits into 8-bit bytes and discard the remainder. 807 * </p> 808 * 809 * @return true if using strict decoding. 810 * @since 1.15 811 */ 812 public CodecPolicy getCodecPolicy() { 813 return decodingPolicy; 814 } 815 816 /** 817 * Gets the default buffer size. Can be overridden. 818 * 819 * @return The default buffer size. 820 */ 821 protected int getDefaultBufferSize() { 822 return DEFAULT_BUFFER_SIZE; 823 } 824 825 /** 826 * Gets the amount of space needed to encode the supplied array. 827 * 828 * @param array byte[] array which will later be encoded. 829 * @return amount of space needed to encode the supplied array. Returns a long since a max-len array will require > Integer.MAX_VALUE. 830 */ 831 public long getEncodedLength(final byte[] array) { 832 // Calculate non-chunked size - rounded up to allow for padding 833 // cast to long is needed to avoid possibility of overflow 834 long len = (array.length + unencodedBlockSize - 1) / unencodedBlockSize * (long) encodedBlockSize; 835 if (lineLength > 0) { // We're using chunking 836 // Round up to nearest multiple 837 len += (len + lineLength - 1) / lineLength * chunkSeparatorLength; 838 } 839 return len; 840 } 841 842 /** 843 * Tests whether this object has buffered data for reading. 844 * 845 * @param context The context to be used. 846 * @return true if there is data still available for reading. 847 */ 848 boolean hasData(final Context context) { // package protected for access from I/O streams 849 return context.pos > context.readPos; 850 } 851 852 /** 853 * Tests whether or not the {@code octet} is in the current alphabet. Does not allow whitespace or pad. 854 * 855 * @param value The value to test. 856 * @return {@code true} if the value is defined in the current alphabet, {@code false} otherwise. 857 */ 858 protected abstract boolean isInAlphabet(byte value); 859 860 /** 861 * Tests a given byte array to see if it contains only valid characters within the alphabet. The method optionally treats whitespace and pad as valid. 862 * 863 * @param arrayOctet byte array to test. 864 * @param allowWhitespacePad if {@code true}, then whitespace and PAD are also allowed. 865 * @return {@code true} if all bytes are valid characters in the alphabet or if the byte array is empty; {@code false}, otherwise. 866 */ 867 public boolean isInAlphabet(final byte[] arrayOctet, final boolean allowWhitespacePad) { 868 for (final byte octet : arrayOctet) { 869 if (!isInAlphabet(octet) && (!allowWhitespacePad || octet != pad && !Character.isWhitespace(octet))) { 870 return false; 871 } 872 } 873 return true; 874 } 875 876 /** 877 * Tests a given String to see if it contains only valid characters within the alphabet. The method treats whitespace and PAD as valid. 878 * 879 * @param basen String to test. 880 * @return {@code true} if all characters in the String are valid characters in the alphabet or if the String is empty; {@code false}, otherwise. 881 * @see #isInAlphabet(byte[], boolean) 882 */ 883 public boolean isInAlphabet(final String basen) { 884 return isInAlphabet(StringUtils.getBytesUtf8(basen), true); 885 } 886 887 /** 888 * Tests true if decoding behavior is strict. Decoding will raise an {@link IllegalArgumentException} if trailing bits are not part of a valid encoding. 889 * 890 * <p> 891 * The default is false for lenient decoding. Decoding will compose trailing bits into 8-bit bytes and discard the remainder. 892 * </p> 893 * 894 * @return true if using strict decoding. 895 * @since 1.15 896 */ 897 public boolean isStrictDecoding() { 898 return decodingPolicy == CodecPolicy.STRICT; 899 } 900 901 /** 902 * Reads buffered data into the provided byte[] array, starting at position bPos, up to a maximum of bAvail bytes. Returns how many bytes were actually 903 * extracted. 904 * <p> 905 * Package private for access from I/O streams. 906 * </p> 907 * 908 * @param b byte[] array to extract the buffered data into. 909 * @param position position in byte[] array to start extraction at. 910 * @param available amount of bytes we're allowed to extract. We may extract fewer (if fewer are available). 911 * @param context The context to be used. 912 * @return The number of bytes successfully extracted into the provided byte[] array. 913 */ 914 int readResults(final byte[] b, final int position, final int available, final Context context) { 915 if (hasData(context)) { 916 final int len = Math.min(available(context), available); 917 System.arraycopy(context.buffer, context.readPos, b, position, len); 918 context.readPos += len; 919 if (!hasData(context)) { 920 // All data read. 921 // Reset position markers but do not set buffer to null to allow its reuse. 922 // hasData(context) will still return false, and this method will return 0 until 923 // more data is available, or -1 if EOF. 924 context.pos = context.readPos = 0; 925 } 926 return len; 927 } 928 return context.eof ? EOF : 0; 929 } 930}