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