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 * https://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17
18 package org.apache.commons.codec.binary;
19
20 import java.util.Arrays;
21 import java.util.Objects;
22 import java.util.function.Supplier;
23
24 import org.apache.commons.codec.BinaryDecoder;
25 import org.apache.commons.codec.BinaryEncoder;
26 import org.apache.commons.codec.CodecPolicy;
27 import org.apache.commons.codec.DecoderException;
28 import org.apache.commons.codec.EncoderException;
29
30 /**
31 * Abstract superclass for Base-N encoders and decoders.
32 *
33 * <p>
34 * This class is thread-safe.
35 * </p>
36 * <p>
37 * 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
38 * unused from the final character or entire characters. The default mode is lenient decoding.
39 * </p>
40 * <ul>
41 * <li>Lenient: Any trailing bits are composed into 8-bit bytes where possible. The remainder are discarded.
42 * <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
43 * character must be zero. Impossible counts of entire final characters are not allowed.
44 * </ul>
45 * <p>
46 * 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
47 * the final character. This requires that the input bytes use the same padding and alphabet as the encoder.
48 * </p>
49 */
50 public abstract class BaseNCodec implements BinaryEncoder, BinaryDecoder {
51
52 /**
53 * Builds {@link Base64} instances.
54 *
55 * @param <T> the codec type to build.
56 * @param <B> the codec builder subtype.
57 * @since 1.17.0
58 */
59 public abstract static class AbstractBuilder<T, B extends AbstractBuilder<T, B>> implements Supplier<T> {
60
61 private int unencodedBlockSize;
62 private int encodedBlockSize;
63 private CodecPolicy decodingPolicy = DECODING_POLICY_DEFAULT;
64 private int lineLength;
65 private byte[] lineSeparator = CHUNK_SEPARATOR;
66 private final byte[] defaultEncodeTable;
67 private byte[] encodeTable;
68 private byte[] decodeTable;
69
70 /** Padding byte. */
71 private byte padding = PAD_DEFAULT;
72
73 AbstractBuilder(final byte[] defaultEncodeTable) {
74 this.defaultEncodeTable = defaultEncodeTable;
75 this.encodeTable = defaultEncodeTable;
76 }
77
78 /**
79 * Returns this instance typed as the subclass type {@code B}.
80 * <p>
81 * This is the same as the expression:
82 * </p>
83 *
84 * <pre>
85 * (B) this
86 * </pre>
87 *
88 * @return {@code this} instance typed as the subclass type {@code B}.
89 */
90 @SuppressWarnings("unchecked")
91 B asThis() {
92 return (B) this;
93 }
94
95 byte[] getDecodeTable() {
96 return decodeTable;
97 }
98
99 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 * Placeholder for the bytes we're dealing with for our based logic. Bitwise operations store and extract the encoding or decoding from this variable.
252 */
253 long lbitWorkArea;
254 /**
255 * Buffer for streaming.
256 */
257 byte[] buffer;
258 /**
259 * Position where next character should be written in the buffer.
260 */
261 int pos;
262 /**
263 * Position where next character should be read from the buffer.
264 */
265 int readPos;
266 /**
267 * Boolean flag to indicate the EOF has been reached. Once EOF has been reached, this object becomes useless, and must be thrown away.
268 */
269 boolean eof;
270 /**
271 * 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
272 * goes beyond lineLength (if lineLength > 0).
273 */
274 int currentLinePos;
275 /**
276 * 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.
277 */
278 int modulus;
279
280 /**
281 * Returns a String useful for debugging (especially within a debugger.)
282 *
283 * @return a String useful for debugging.
284 */
285 @Override
286 public String toString() {
287 return String.format("%s[buffer=%s, currentLinePos=%s, eof=%s, ibitWorkArea=%s, lbitWorkArea=%s, " + "modulus=%s, pos=%s, readPos=%s]",
288 this.getClass().getSimpleName(), Arrays.toString(buffer), currentLinePos, eof, ibitWorkArea, lbitWorkArea, modulus, pos, readPos);
289 }
290 }
291
292 /**
293 * End-of-file marker.
294 *
295 * @since 1.7
296 */
297 static final int EOF = -1;
298 /**
299 * MIME chunk size per RFC 2045 section 6.8.
300 *
301 * <p>
302 * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any equal signs.
303 * </p>
304 *
305 * @see <a href="https://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 6.8</a>
306 */
307 public static final int MIME_CHUNK_SIZE = 76;
308 /**
309 * PEM chunk size per RFC 1421 section 4.3.2.4.
310 *
311 * <p>
312 * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any equal signs.
313 * </p>
314 *
315 * @see <a href="https://tools.ietf.org/html/rfc1421">RFC 1421 section 4.3.2.4</a>
316 */
317 public static final int PEM_CHUNK_SIZE = 64;
318 private static final int DEFAULT_BUFFER_RESIZE_FACTOR = 2;
319 /**
320 * Defines the default buffer size - currently {@value} - must be large enough for at least one encoded block+separator
321 */
322 private static final int DEFAULT_BUFFER_SIZE = 8192;
323 /**
324 * The maximum size buffer to allocate.
325 *
326 * <p>
327 * This is set to the same size used in the JDK {@link java.util.ArrayList}:
328 * </p>
329 * <blockquote> Some VMs reserve some header words in an array. Attempts to allocate larger arrays may result in OutOfMemoryError: Requested array size
330 * exceeds VM limit. </blockquote>
331 */
332 private static final int MAX_BUFFER_SIZE = Integer.MAX_VALUE - 8;
333 /** Mask used to extract 8 bits, used in decoding bytes */
334 protected static final int MASK_8BITS = 0xff;
335 /**
336 * Byte used to pad output.
337 */
338 protected static final byte PAD_DEFAULT = '='; // Allow static access to default
339 /**
340 * The default decoding policy.
341 *
342 * @since 1.15
343 */
344 protected static final CodecPolicy DECODING_POLICY_DEFAULT = CodecPolicy.LENIENT;
345 /**
346 * Chunk separator per RFC 2045 section 2.1.
347 *
348 * @see <a href="https://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a>
349 */
350 static final byte[] CHUNK_SEPARATOR = { '\r', '\n' };
351 /**
352 * The empty byte array.
353 */
354 static final byte[] EMPTY_BYTE_ARRAY = {};
355
356 /**
357 * Create a positive capacity at least as large the minimum required capacity. If the minimum capacity is negative then this throws an OutOfMemoryError as
358 * no array can be allocated.
359 *
360 * @param minCapacity the minimum capacity
361 * @return the capacity
362 * @throws OutOfMemoryError if the {@code minCapacity} is negative
363 */
364 private static int createPositiveCapacity(final int minCapacity) {
365 if (minCapacity < 0) {
366 // overflow
367 throw new OutOfMemoryError("Unable to allocate array size: " + (minCapacity & 0xffffffffL));
368 }
369 // This is called when we require buffer expansion to a very big array.
370 // Use the conservative maximum buffer size if possible, otherwise the biggest required.
371 //
372 // Note: In this situation JDK 1.8 java.util.ArrayList returns Integer.MAX_VALUE.
373 // This excludes some VMs that can exceed MAX_BUFFER_SIZE but not allocate a full
374 // Integer.MAX_VALUE length array.
375 // The result is that we may have to allocate an array of this size more than once if
376 // the capacity must be expanded again.
377 return Math.max(minCapacity, MAX_BUFFER_SIZE);
378 }
379
380 /**
381 * Gets a copy of the chunk separator per RFC 2045 section 2.1.
382 *
383 * @return the chunk separator
384 * @see <a href="https://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a>
385 * @since 1.15
386 */
387 public static byte[] getChunkSeparator() {
388 return CHUNK_SEPARATOR.clone();
389 }
390
391 /**
392 * Gets the array length or 0 if null.
393 *
394 * @param array the array or null.
395 * @return the array length or 0 if null.
396 */
397 static int getLength(final byte[] array) {
398 return array == null ? 0 : array.length;
399 }
400
401 /**
402 * Checks if a byte value is whitespace or not.
403 *
404 * @param byteToCheck the byte to check
405 * @return true if byte is whitespace, false otherwise
406 * @see Character#isWhitespace(int)
407 * @deprecated Use {@link Character#isWhitespace(int)}.
408 */
409 @Deprecated
410 protected static boolean isWhiteSpace(final byte byteToCheck) {
411 return Character.isWhitespace(byteToCheck);
412 }
413
414 /**
415 * Increases our buffer by the {@link #DEFAULT_BUFFER_RESIZE_FACTOR}.
416 *
417 * @param context the context to be used
418 * @param minCapacity the minimum required capacity
419 * @return the resized byte[] buffer
420 * @throws OutOfMemoryError if the {@code minCapacity} is negative
421 */
422 private static byte[] resizeBuffer(final Context context, final int minCapacity) {
423 // Overflow-conscious code treats the min and new capacity as unsigned.
424 final int oldCapacity = context.buffer.length;
425 int newCapacity = oldCapacity * DEFAULT_BUFFER_RESIZE_FACTOR;
426 if (Integer.compareUnsigned(newCapacity, minCapacity) < 0) {
427 newCapacity = minCapacity;
428 }
429 if (Integer.compareUnsigned(newCapacity, MAX_BUFFER_SIZE) > 0) {
430 newCapacity = createPositiveCapacity(minCapacity);
431 }
432 final byte[] b = Arrays.copyOf(context.buffer, newCapacity);
433 context.buffer = b;
434 return b;
435 }
436
437 /**
438 * Deprecated: Will be removed in 2.0.
439 * <p>
440 * Instance variable just in case it needs to vary later
441 * </p>
442 *
443 * @deprecated Use {@link #pad}. Will be removed in 2.0.
444 */
445 @Deprecated
446 protected final byte PAD = PAD_DEFAULT;
447 /** Pad byte. Instance variable just in case it needs to vary later. */
448 protected final byte pad;
449 /** Number of bytes in each full block of unencoded data, for example 4 for Base64 and 5 for Base32 */
450 private final int unencodedBlockSize;
451 /** Number of bytes in each full block of encoded data, for example 3 for Base64 and 8 for Base32 */
452 private final int encodedBlockSize;
453 /**
454 * 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
455 * encodedBlockSize.
456 */
457 protected final int lineLength;
458 /**
459 * Size of chunk separator. Not used unless {@link #lineLength} > 0.
460 */
461 private final int chunkSeparatorLength;
462 /**
463 * 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
464 * unused from the final character or entire characters. The default mode is lenient decoding. Set this to {@code true} to enable strict decoding.
465 * <ul>
466 * <li>Lenient: Any trailing bits are composed into 8-bit bytes where possible. The remainder are discarded.
467 * <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
468 * character must be zero. Impossible counts of entire final characters are not allowed.
469 * </ul>
470 * <p>
471 * 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
472 * on the final character. This requires that the input bytes use the same padding and alphabet as the encoder.
473 * </p>
474 */
475 private final CodecPolicy decodingPolicy;
476
477 /**
478 * Decode table to use.
479 */
480 final byte[] decodeTable;
481
482 /**
483 * Encode table.
484 */
485 final byte[] encodeTable;
486
487 /**
488 * Constructs a new instance for a subclass.
489 *
490 * @param builder How to build this portion of the instance.
491 * @since 1.20.0
492 */
493 protected BaseNCodec(final AbstractBuilder<?, ?> builder) {
494 this.unencodedBlockSize = builder.unencodedBlockSize;
495 this.encodedBlockSize = builder.encodedBlockSize;
496 final boolean useChunking = builder.lineLength > 0 && builder.lineSeparator.length > 0;
497 this.lineLength = useChunking ? builder.lineLength / builder.encodedBlockSize * builder.encodedBlockSize : 0;
498 this.chunkSeparatorLength = builder.lineSeparator.length;
499 this.pad = builder.padding;
500 this.decodingPolicy = Objects.requireNonNull(builder.decodingPolicy, "codecPolicy");
501 this.encodeTable = Objects.requireNonNull(builder.getEncodeTable(), "builder.getEncodeTable()");
502 this.decodeTable = builder.getDecodeTable();
503 }
504
505 /**
506 * Constructs a new instance.
507 * <p>
508 * Note {@code lineLength} is rounded down to the nearest multiple of the encoded block size. If {@code chunkSeparatorLength} is zero, then chunking is
509 * disabled.
510 * </p>
511 *
512 * @param unencodedBlockSize the size of an unencoded block (for example Base64 = 3).
513 * @param encodedBlockSize the size of an encoded block (for example Base64 = 4).
514 * @param lineLength if > 0, use chunking with a length {@code lineLength}.
515 * @param chunkSeparatorLength the chunk separator length, if relevant.
516 * @deprecated Use {@link BaseNCodec#BaseNCodec(AbstractBuilder)}.
517 */
518 @Deprecated
519 protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize, final int lineLength, final int chunkSeparatorLength) {
520 this(unencodedBlockSize, encodedBlockSize, lineLength, chunkSeparatorLength, PAD_DEFAULT);
521 }
522
523 /**
524 * Constructs a new instance.
525 * <p>
526 * Note {@code lineLength} is rounded down to the nearest multiple of the encoded block size. If {@code chunkSeparatorLength} is zero, then chunking is
527 * disabled.
528 * </p>
529 *
530 * @param unencodedBlockSize the size of an unencoded block (for example Base64 = 3).
531 * @param encodedBlockSize the size of an encoded block (for example Base64 = 4).
532 * @param lineLength if > 0, use chunking with a length {@code lineLength}.
533 * @param chunkSeparatorLength the chunk separator length, if relevant.
534 * @param pad byte used as padding byte.
535 * @deprecated Use {@link BaseNCodec#BaseNCodec(AbstractBuilder)}.
536 */
537 @Deprecated
538 protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize, final int lineLength, final int chunkSeparatorLength, final byte pad) {
539 this(unencodedBlockSize, encodedBlockSize, lineLength, chunkSeparatorLength, pad, DECODING_POLICY_DEFAULT);
540 }
541
542 /**
543 * Constructs a new instance.
544 * <p>
545 * Note {@code lineLength} is rounded down to the nearest multiple of the encoded block size. If {@code chunkSeparatorLength} is zero, then chunking is
546 * disabled.
547 * </p>
548 *
549 * @param unencodedBlockSize the size of an unencoded block (for example Base64 = 3).
550 * @param encodedBlockSize the size of an encoded block (for example Base64 = 4).
551 * @param lineLength if > 0, use chunking with a length {@code lineLength}.
552 * @param chunkSeparatorLength the chunk separator length, if relevant.
553 * @param pad byte used as padding byte.
554 * @param decodingPolicy Decoding policy.
555 * @since 1.15
556 * @deprecated Use {@link BaseNCodec#BaseNCodec(AbstractBuilder)}.
557 */
558 @Deprecated
559 protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize, final int lineLength, final int chunkSeparatorLength, final byte pad,
560 final CodecPolicy decodingPolicy) {
561 this.unencodedBlockSize = unencodedBlockSize;
562 this.encodedBlockSize = encodedBlockSize;
563 final boolean useChunking = lineLength > 0 && chunkSeparatorLength > 0;
564 this.lineLength = useChunking ? lineLength / encodedBlockSize * encodedBlockSize : 0;
565 this.chunkSeparatorLength = chunkSeparatorLength;
566 this.pad = pad;
567 this.decodingPolicy = Objects.requireNonNull(decodingPolicy, "codecPolicy");
568 this.encodeTable = null;
569 this.decodeTable = null;
570 }
571
572 /**
573 * Returns the amount of buffered data available for reading.
574 *
575 * @param context the context to be used.
576 * @return The amount of buffered data available for reading.
577 */
578 int available(final Context context) { // package protected for access from I/O streams
579 return hasData(context) ? context.pos - context.readPos : 0;
580 }
581
582 /**
583 * Tests a given byte array to see if it contains any characters within the alphabet or PAD.
584 *
585 * Intended for use in checking line-ending arrays.
586 *
587 * @param arrayOctet byte array to test.
588 * @return {@code true} if any byte is a valid character in the alphabet or PAD; {@code false} otherwise.
589 */
590 protected boolean containsAlphabetOrPad(final byte[] arrayOctet) {
591 if (arrayOctet != null) {
592 for (final byte element : arrayOctet) {
593 if (pad == element || isInAlphabet(element)) {
594 return true;
595 }
596 }
597 }
598 return false;
599 }
600
601 /**
602 * Decodes a byte[] containing characters in the Base-N alphabet.
603 *
604 * @param array A byte array containing Base-N character data.
605 * @return a byte array containing binary data.
606 */
607 @Override
608 public byte[] decode(final byte[] array) {
609 if (BinaryCodec.isEmpty(array)) {
610 return array;
611 }
612 final Context context = new Context();
613 decode(array, 0, array.length, context);
614 decode(array, 0, EOF, context); // Notify decoder of EOF.
615 final byte[] result = new byte[context.pos];
616 readResults(result, 0, result.length, context);
617 return result;
618 }
619
620 // package protected for access from I/O streams
621 abstract void decode(byte[] array, int i, int length, Context context);
622
623 /**
624 * 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
625 * DecoderException if the supplied object is not of type byte[] or String.
626 *
627 * @param obj Object to decode.
628 * @return An object (of type byte[]) containing the binary data which corresponds to the byte[] or String supplied.
629 * @throws DecoderException if the parameter supplied is not of type byte[].
630 */
631 @Override
632 public Object decode(final Object obj) throws DecoderException {
633 if (obj instanceof byte[]) {
634 return decode((byte[]) obj);
635 }
636 if (obj instanceof String) {
637 return decode((String) obj);
638 }
639 throw new DecoderException("Parameter supplied to Base-N decode is not a byte[] or a String");
640 }
641
642 /**
643 * Decodes a String containing characters in the Base-N alphabet.
644 *
645 * @param array A String containing Base-N character data.
646 * @return a byte array containing binary data.
647 */
648 public byte[] decode(final String array) {
649 return decode(StringUtils.getBytesUtf8(array));
650 }
651
652 /**
653 * Encodes a byte[] containing binary data, into a byte[] containing characters in the alphabet.
654 *
655 * @param array a byte array containing binary data.
656 * @return A byte array containing only the base N alphabetic character data.
657 */
658 @Override
659 public byte[] encode(final byte[] array) {
660 if (BinaryCodec.isEmpty(array)) {
661 return array;
662 }
663 return encode(array, 0, array.length);
664 }
665
666 /**
667 * Encodes a byte[] containing binary data, into a byte[] containing characters in the alphabet.
668 *
669 * @param array a byte array containing binary data.
670 * @param offset initial offset of the subarray.
671 * @param length length of the subarray.
672 * @return A byte array containing only the base N alphabetic character data.
673 * @since 1.11
674 */
675 public byte[] encode(final byte[] array, final int offset, final int length) {
676 if (BinaryCodec.isEmpty(array)) {
677 return array;
678 }
679 final Context context = new Context();
680 encode(array, offset, length, context);
681 encode(array, offset, EOF, context); // Notify encoder of EOF.
682 final byte[] buf = new byte[context.pos - context.readPos];
683 readResults(buf, 0, buf.length, context);
684 return buf;
685 }
686
687 // package protected for access from I/O streams
688 abstract void encode(byte[] array, int i, int length, Context context);
689
690 /**
691 * 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
692 * EncoderException if the supplied object is not of type byte[].
693 *
694 * @param obj Object to encode.
695 * @return An object (of type byte[]) containing the Base-N encoded data which corresponds to the byte[] supplied.
696 * @throws EncoderException if the parameter supplied is not of type byte[].
697 */
698 @Override
699 public Object encode(final Object obj) throws EncoderException {
700 if (!(obj instanceof byte[])) {
701 throw new EncoderException("Parameter supplied to Base-N encode is not a byte[]");
702 }
703 return encode((byte[]) obj);
704 }
705
706 /**
707 * Encodes a byte[] containing binary data, into a String containing characters in the appropriate alphabet. Uses UTF8 encoding.
708 * <p>
709 * This is a duplicate of {@link #encodeToString(byte[])}; it was merged during refactoring.
710 * </p>
711 *
712 * @param array a byte array containing binary data.
713 * @return String containing only character data in the appropriate alphabet.
714 * @since 1.5
715 */
716 public String encodeAsString(final byte[] array) {
717 return StringUtils.newStringUtf8(encode(array));
718 }
719
720 /**
721 * Encodes a byte[] containing binary data, into a String containing characters in the Base-N alphabet. Uses UTF8 encoding.
722 *
723 * @param array a byte array containing binary data.
724 * @return A String containing only Base-N character data.
725 */
726 public String encodeToString(final byte[] array) {
727 return StringUtils.newStringUtf8(encode(array));
728 }
729
730 /**
731 * Ensures that the buffer has room for {@code size} bytes
732 *
733 * @param size minimum spare space required.
734 * @param context the context to be used.
735 * @return the buffer
736 */
737 protected byte[] ensureBufferSize(final int size, final Context context) {
738 if (context.buffer == null) {
739 context.buffer = new byte[Math.max(size, getDefaultBufferSize())];
740 context.pos = 0;
741 context.readPos = 0;
742 // Overflow-conscious:
743 // x + y > z == x + y - z > 0
744 } else if (context.pos + size - context.buffer.length > 0) {
745 return resizeBuffer(context, context.pos + size);
746 }
747 return context.buffer;
748 }
749
750 /**
751 * Gets the decoding behavior policy.
752 *
753 * <p>
754 * 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
755 * valid encoding. Decoding will compose trailing bits into 8-bit bytes and discard the remainder.
756 * </p>
757 *
758 * @return true if using strict decoding.
759 * @since 1.15
760 */
761 public CodecPolicy getCodecPolicy() {
762 return decodingPolicy;
763 }
764
765 /**
766 * Gets the default buffer size. Can be overridden.
767 *
768 * @return the default buffer size.
769 */
770 protected int getDefaultBufferSize() {
771 return DEFAULT_BUFFER_SIZE;
772 }
773
774 /**
775 * Gets the amount of space needed to encode the supplied array.
776 *
777 * @param array byte[] array which will later be encoded.
778 * @return amount of space needed to encode the supplied array. Returns a long since a max-len array will require > Integer.MAX_VALUE.
779 */
780 public long getEncodedLength(final byte[] array) {
781 // Calculate non-chunked size - rounded up to allow for padding
782 // cast to long is needed to avoid possibility of overflow
783 long len = (array.length + unencodedBlockSize - 1) / unencodedBlockSize * (long) encodedBlockSize;
784 if (lineLength > 0) { // We're using chunking
785 // Round up to nearest multiple
786 len += (len + lineLength - 1) / lineLength * chunkSeparatorLength;
787 }
788 return len;
789 }
790
791 /**
792 * Tests whether this object has buffered data for reading.
793 *
794 * @param context the context to be used.
795 * @return true if there is data still available for reading.
796 */
797 boolean hasData(final Context context) { // package protected for access from I/O streams
798 return context.pos > context.readPos;
799 }
800
801 /**
802 * Tests whether or not the {@code octet} is in the current alphabet. Does not allow whitespace or pad.
803 *
804 * @param value The value to test.
805 * @return {@code true} if the value is defined in the current alphabet, {@code false} otherwise.
806 */
807 protected abstract boolean isInAlphabet(byte value);
808
809 /**
810 * 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.
811 *
812 * @param arrayOctet byte array to test.
813 * @param allowWSPad if {@code true}, then whitespace and PAD are also allowed.
814 * @return {@code true} if all bytes are valid characters in the alphabet or if the byte array is empty; {@code false}, otherwise.
815 */
816 public boolean isInAlphabet(final byte[] arrayOctet, final boolean allowWSPad) {
817 for (final byte octet : arrayOctet) {
818 if (!isInAlphabet(octet) && (!allowWSPad || octet != pad && !Character.isWhitespace(octet))) {
819 return false;
820 }
821 }
822 return true;
823 }
824
825 /**
826 * Tests a given String to see if it contains only valid characters within the alphabet. The method treats whitespace and PAD as valid.
827 *
828 * @param basen String to test.
829 * @return {@code true} if all characters in the String are valid characters in the alphabet or if the String is empty; {@code false}, otherwise.
830 * @see #isInAlphabet(byte[], boolean)
831 */
832 public boolean isInAlphabet(final String basen) {
833 return isInAlphabet(StringUtils.getBytesUtf8(basen), true);
834 }
835
836 /**
837 * Tests true if decoding behavior is strict. Decoding will raise an {@link IllegalArgumentException} if trailing bits are not part of a valid encoding.
838 *
839 * <p>
840 * The default is false for lenient decoding. Decoding will compose trailing bits into 8-bit bytes and discard the remainder.
841 * </p>
842 *
843 * @return true if using strict decoding.
844 * @since 1.15
845 */
846 public boolean isStrictDecoding() {
847 return decodingPolicy == CodecPolicy.STRICT;
848 }
849
850 /**
851 * 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
852 * extracted.
853 * <p>
854 * Package private for access from I/O streams.
855 * </p>
856 *
857 * @param b byte[] array to extract the buffered data into.
858 * @param bPos position in byte[] array to start extraction at.
859 * @param bAvail amount of bytes we're allowed to extract. We may extract fewer (if fewer are available).
860 * @param context the context to be used.
861 * @return The number of bytes successfully extracted into the provided byte[] array.
862 */
863 int readResults(final byte[] b, final int bPos, final int bAvail, final Context context) {
864 if (hasData(context)) {
865 final int len = Math.min(available(context), bAvail);
866 System.arraycopy(context.buffer, context.readPos, b, bPos, len);
867 context.readPos += len;
868 if (!hasData(context)) {
869 // All data read.
870 // Reset position markers but do not set buffer to null to allow its reuse.
871 // hasData(context) will still return false, and this method will return 0 until
872 // more data is available, or -1 if EOF.
873 context.pos = context.readPos = 0;
874 }
875 return len;
876 }
877 return context.eof ? EOF : 0;
878 }
879 }