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