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 *      http://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;
021
022import org.apache.commons.codec.BinaryDecoder;
023import org.apache.commons.codec.BinaryEncoder;
024import org.apache.commons.codec.DecoderException;
025import org.apache.commons.codec.EncoderException;
026
027/**
028 * Abstract superclass for Base-N encoders and decoders.
029 *
030 * <p>
031 * This class is thread-safe.
032 * </p>
033 *
034 * @version $Id: BaseNCodec.java 1563226 2014-01-31 19:38:06Z ggregory $
035 */
036public abstract class BaseNCodec implements BinaryEncoder, BinaryDecoder {
037
038    /**
039     * Holds thread context so classes can be thread-safe.
040     *
041     * This class is not itself thread-safe; each thread must allocate its own copy.
042     *
043     * @since 1.7
044     */
045    static class Context {
046
047        /**
048         * Place holder for the bytes we're dealing with for our based logic.
049         * Bitwise operations store and extract the encoding or decoding from this variable.
050         */
051        int ibitWorkArea;
052
053        /**
054         * Place holder for the bytes we're dealing with for our based logic.
055         * Bitwise operations store and extract the encoding or decoding from this variable.
056         */
057        long lbitWorkArea;
058
059        /**
060         * Buffer for streaming.
061         */
062        byte[] buffer;
063
064        /**
065         * Position where next character should be written in the buffer.
066         */
067        int pos;
068
069        /**
070         * Position where next character should be read from the buffer.
071         */
072        int readPos;
073
074        /**
075         * Boolean flag to indicate the EOF has been reached. Once EOF has been reached, this object becomes useless,
076         * and must be thrown away.
077         */
078        boolean eof;
079
080        /**
081         * Variable tracks how many characters have been written to the current line. Only used when encoding. We use
082         * it to make sure each encoded line never goes beyond lineLength (if lineLength &gt; 0).
083         */
084        int currentLinePos;
085
086        /**
087         * Writes to the buffer only occur after every 3/5 reads when encoding, and every 4/8 reads when decoding. This
088         * variable helps track that.
089         */
090        int modulus;
091
092        Context() {
093        }
094
095        /**
096         * Returns a String useful for debugging (especially within a debugger.)
097         *
098         * @return a String useful for debugging.
099         */
100        @SuppressWarnings("boxing") // OK to ignore boxing here
101        @Override
102        public String toString() {
103            return String.format("%s[buffer=%s, currentLinePos=%s, eof=%s, ibitWorkArea=%s, lbitWorkArea=%s, " +
104                    "modulus=%s, pos=%s, readPos=%s]", this.getClass().getSimpleName(), Arrays.toString(buffer),
105                    currentLinePos, eof, ibitWorkArea, lbitWorkArea, modulus, pos, readPos);
106        }
107    }
108
109    /**
110     * EOF
111     *
112     * @since 1.7
113     */
114    static final int EOF = -1;
115
116    /**
117     *  MIME chunk size per RFC 2045 section 6.8.
118     *
119     * <p>
120     * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any
121     * equal signs.
122     * </p>
123     *
124     * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 6.8</a>
125     */
126    public static final int MIME_CHUNK_SIZE = 76;
127
128    /**
129     * PEM chunk size per RFC 1421 section 4.3.2.4.
130     *
131     * <p>
132     * The {@value} character limit does not count the trailing CRLF, but counts all other characters, including any
133     * equal signs.
134     * </p>
135     *
136     * @see <a href="http://tools.ietf.org/html/rfc1421">RFC 1421 section 4.3.2.4</a>
137     */
138    public static final int PEM_CHUNK_SIZE = 64;
139
140    private static final int DEFAULT_BUFFER_RESIZE_FACTOR = 2;
141
142    /**
143     * Defines the default buffer size - currently {@value}
144     * - must be large enough for at least one encoded block+separator
145     */
146    private static final int DEFAULT_BUFFER_SIZE = 8192;
147
148    /** Mask used to extract 8 bits, used in decoding bytes */
149    protected static final int MASK_8BITS = 0xff;
150
151    /**
152     * Byte used to pad output.
153     */
154    protected static final byte PAD_DEFAULT = '='; // Allow static access to default
155
156    protected final byte PAD = PAD_DEFAULT; // instance variable just in case it needs to vary later
157
158    /** Number of bytes in each full block of unencoded data, e.g. 4 for Base64 and 5 for Base32 */
159    private final int unencodedBlockSize;
160
161    /** Number of bytes in each full block of encoded data, e.g. 3 for Base64 and 8 for Base32 */
162    private final int encodedBlockSize;
163
164    /**
165     * Chunksize for encoding. Not used when decoding.
166     * A value of zero or less implies no chunking of the encoded data.
167     * Rounded down to nearest multiple of encodedBlockSize.
168     */
169    protected final int lineLength;
170
171    /**
172     * Size of chunk separator. Not used unless {@link #lineLength} &gt; 0.
173     */
174    private final int chunkSeparatorLength;
175
176    /**
177     * Note <code>lineLength</code> is rounded down to the nearest multiple of {@link #encodedBlockSize}
178     * If <code>chunkSeparatorLength</code> is zero, then chunking is disabled.
179     * @param unencodedBlockSize the size of an unencoded block (e.g. Base64 = 3)
180     * @param encodedBlockSize the size of an encoded block (e.g. Base64 = 4)
181     * @param lineLength if &gt; 0, use chunking with a length <code>lineLength</code>
182     * @param chunkSeparatorLength the chunk separator length, if relevant
183     */
184    protected BaseNCodec(final int unencodedBlockSize, final int encodedBlockSize,
185                         final int lineLength, final int chunkSeparatorLength) {
186        this.unencodedBlockSize = unencodedBlockSize;
187        this.encodedBlockSize = encodedBlockSize;
188        final boolean useChunking = lineLength > 0 && chunkSeparatorLength > 0;
189        this.lineLength = useChunking ? (lineLength / encodedBlockSize) * encodedBlockSize : 0;
190        this.chunkSeparatorLength = chunkSeparatorLength;
191    }
192
193    /**
194     * Returns true if this object has buffered data for reading.
195     *
196     * @param context the context to be used
197     * @return true if there is data still available for reading.
198     */
199    boolean hasData(final Context context) {  // package protected for access from I/O streams
200        return context.buffer != null;
201    }
202
203    /**
204     * Returns the amount of buffered data available for reading.
205     *
206     * @param context the context to be used
207     * @return The amount of buffered data available for reading.
208     */
209    int available(final Context context) {  // package protected for access from I/O streams
210        return context.buffer != null ? context.pos - context.readPos : 0;
211    }
212
213    /**
214     * Get the default buffer size. Can be overridden.
215     *
216     * @return {@link #DEFAULT_BUFFER_SIZE}
217     */
218    protected int getDefaultBufferSize() {
219        return DEFAULT_BUFFER_SIZE;
220    }
221
222    /**
223     * Increases our buffer by the {@link #DEFAULT_BUFFER_RESIZE_FACTOR}.
224     * @param context the context to be used
225     */
226    private byte[] resizeBuffer(final Context context) {
227        if (context.buffer == null) {
228            context.buffer = new byte[getDefaultBufferSize()];
229            context.pos = 0;
230            context.readPos = 0;
231        } else {
232            final byte[] b = new byte[context.buffer.length * DEFAULT_BUFFER_RESIZE_FACTOR];
233            System.arraycopy(context.buffer, 0, b, 0, context.buffer.length);
234            context.buffer = b;
235        }
236        return context.buffer;
237    }
238
239    /**
240     * Ensure that the buffer has room for <code>size</code> bytes
241     *
242     * @param size minimum spare space required
243     * @param context the context to be used
244     */
245    protected byte[] ensureBufferSize(final int size, final Context context){
246        if ((context.buffer == null) || (context.buffer.length < context.pos + size)){
247            return resizeBuffer(context);
248        }
249        return context.buffer;
250    }
251
252    /**
253     * Extracts buffered data into the provided byte[] array, starting at position bPos, up to a maximum of bAvail
254     * bytes. Returns how many bytes were actually extracted.
255     * <p>
256     * Package protected for access from I/O streams.
257     *
258     * @param b
259     *            byte[] array to extract the buffered data into.
260     * @param bPos
261     *            position in byte[] array to start extraction at.
262     * @param bAvail
263     *            amount of bytes we're allowed to extract. We may extract fewer (if fewer are available).
264     * @param context
265     *            the context to be used
266     * @return The number of bytes successfully extracted into the provided byte[] array.
267     */
268    int readResults(final byte[] b, final int bPos, final int bAvail, final Context context) {
269        if (context.buffer != null) {
270            final int len = Math.min(available(context), bAvail);
271            System.arraycopy(context.buffer, context.readPos, b, bPos, len);
272            context.readPos += len;
273            if (context.readPos >= context.pos) {
274                context.buffer = null; // so hasData() will return false, and this method can return -1
275            }
276            return len;
277        }
278        return context.eof ? EOF : 0;
279    }
280
281    /**
282     * Checks if a byte value is whitespace or not.
283     * Whitespace is taken to mean: space, tab, CR, LF
284     * @param byteToCheck
285     *            the byte to check
286     * @return true if byte is whitespace, false otherwise
287     */
288    protected static boolean isWhiteSpace(final byte byteToCheck) {
289        switch (byteToCheck) {
290            case ' ' :
291            case '\n' :
292            case '\r' :
293            case '\t' :
294                return true;
295            default :
296                return false;
297        }
298    }
299
300    /**
301     * Encodes an Object using the Base-N algorithm. This method is provided in order to satisfy the requirements of
302     * the Encoder interface, and will throw an EncoderException if the supplied object is not of type byte[].
303     *
304     * @param obj
305     *            Object to encode
306     * @return An object (of type byte[]) containing the Base-N encoded data which corresponds to the byte[] supplied.
307     * @throws EncoderException
308     *             if the parameter supplied is not of type byte[]
309     */
310    @Override
311    public Object encode(final Object obj) throws EncoderException {
312        if (!(obj instanceof byte[])) {
313            throw new EncoderException("Parameter supplied to Base-N encode is not a byte[]");
314        }
315        return encode((byte[]) obj);
316    }
317
318    /**
319     * Encodes a byte[] containing binary data, into a String containing characters in the Base-N alphabet.
320     * Uses UTF8 encoding.
321     *
322     * @param pArray
323     *            a byte array containing binary data
324     * @return A String containing only Base-N character data
325     */
326    public String encodeToString(final byte[] pArray) {
327        return StringUtils.newStringUtf8(encode(pArray));
328    }
329
330    /**
331     * Encodes a byte[] containing binary data, into a String containing characters in the appropriate alphabet.
332     * Uses UTF8 encoding.
333     *
334     * @param pArray a byte array containing binary data
335     * @return String containing only character data in the appropriate alphabet.
336    */
337    public String encodeAsString(final byte[] pArray){
338        return StringUtils.newStringUtf8(encode(pArray));
339    }
340
341    /**
342     * Decodes an Object using the Base-N algorithm. This method is provided in order to satisfy the requirements of
343     * the Decoder interface, and will throw a DecoderException if the supplied object is not of type byte[] or String.
344     *
345     * @param obj
346     *            Object to decode
347     * @return An object (of type byte[]) containing the binary data which corresponds to the byte[] or String
348     *         supplied.
349     * @throws DecoderException
350     *             if the parameter supplied is not of type byte[]
351     */
352    @Override
353    public Object decode(final Object obj) throws DecoderException {
354        if (obj instanceof byte[]) {
355            return decode((byte[]) obj);
356        } else if (obj instanceof String) {
357            return decode((String) obj);
358        } else {
359            throw new DecoderException("Parameter supplied to Base-N decode is not a byte[] or a String");
360        }
361    }
362
363    /**
364     * Decodes a String containing characters in the Base-N alphabet.
365     *
366     * @param pArray
367     *            A String containing Base-N character data
368     * @return a byte array containing binary data
369     */
370    public byte[] decode(final String pArray) {
371        return decode(StringUtils.getBytesUtf8(pArray));
372    }
373
374    /**
375     * Decodes a byte[] containing characters in the Base-N alphabet.
376     *
377     * @param pArray
378     *            A byte array containing Base-N character data
379     * @return a byte array containing binary data
380     */
381    @Override
382    public byte[] decode(final byte[] pArray) {
383        if (pArray == null || pArray.length == 0) {
384            return pArray;
385        }
386        final Context context = new Context();
387        decode(pArray, 0, pArray.length, context);
388        decode(pArray, 0, EOF, context); // Notify decoder of EOF.
389        final byte[] result = new byte[context.pos];
390        readResults(result, 0, result.length, context);
391        return result;
392    }
393
394    /**
395     * Encodes a byte[] containing binary data, into a byte[] containing characters in the alphabet.
396     *
397     * @param pArray
398     *            a byte array containing binary data
399     * @return A byte array containing only the basen alphabetic character data
400     */
401    @Override
402    public byte[] encode(final byte[] pArray) {
403        if (pArray == null || pArray.length == 0) {
404            return pArray;
405        }
406        final Context context = new Context();
407        encode(pArray, 0, pArray.length, context);
408        encode(pArray, 0, EOF, context); // Notify encoder of EOF.
409        final byte[] buf = new byte[context.pos - context.readPos];
410        readResults(buf, 0, buf.length, context);
411        return buf;
412    }
413
414    // package protected for access from I/O streams
415    abstract void encode(byte[] pArray, int i, int length, Context context);
416
417    // package protected for access from I/O streams
418    abstract void decode(byte[] pArray, int i, int length, Context context);
419
420    /**
421     * Returns whether or not the <code>octet</code> is in the current alphabet.
422     * Does not allow whitespace or pad.
423     *
424     * @param value The value to test
425     *
426     * @return {@code true} if the value is defined in the current alphabet, {@code false} otherwise.
427     */
428    protected abstract boolean isInAlphabet(byte value);
429
430    /**
431     * Tests a given byte array to see if it contains only valid characters within the alphabet.
432     * The method optionally treats whitespace and pad as valid.
433     *
434     * @param arrayOctet byte array to test
435     * @param allowWSPad if {@code true}, then whitespace and PAD are also allowed
436     *
437     * @return {@code true} if all bytes are valid characters in the alphabet or if the byte array is empty;
438     *         {@code false}, otherwise
439     */
440    public boolean isInAlphabet(final byte[] arrayOctet, final boolean allowWSPad) {
441        for (int i = 0; i < arrayOctet.length; i++) {
442            if (!isInAlphabet(arrayOctet[i]) &&
443                    (!allowWSPad || (arrayOctet[i] != PAD) && !isWhiteSpace(arrayOctet[i]))) {
444                return false;
445            }
446        }
447        return true;
448    }
449
450    /**
451     * Tests a given String to see if it contains only valid characters within the alphabet.
452     * The method treats whitespace and PAD as valid.
453     *
454     * @param basen String to test
455     * @return {@code true} if all characters in the String are valid characters in the alphabet or if
456     *         the String is empty; {@code false}, otherwise
457     * @see #isInAlphabet(byte[], boolean)
458     */
459    public boolean isInAlphabet(final String basen) {
460        return isInAlphabet(StringUtils.getBytesUtf8(basen), true);
461    }
462
463    /**
464     * Tests a given byte array to see if it contains any characters within the alphabet or PAD.
465     *
466     * Intended for use in checking line-ending arrays
467     *
468     * @param arrayOctet
469     *            byte array to test
470     * @return {@code true} if any byte is a valid character in the alphabet or PAD; {@code false} otherwise
471     */
472    protected boolean containsAlphabetOrPad(final byte[] arrayOctet) {
473        if (arrayOctet == null) {
474            return false;
475        }
476        for (final byte element : arrayOctet) {
477            if (PAD == element || isInAlphabet(element)) {
478                return true;
479            }
480        }
481        return false;
482    }
483
484    /**
485     * Calculates the amount of space needed to encode the supplied array.
486     *
487     * @param pArray byte[] array which will later be encoded
488     *
489     * @return amount of space needed to encoded the supplied array.
490     * Returns a long since a max-len array will require &gt; Integer.MAX_VALUE
491     */
492    public long getEncodedLength(final byte[] pArray) {
493        // Calculate non-chunked size - rounded up to allow for padding
494        // cast to long is needed to avoid possibility of overflow
495        long len = ((pArray.length + unencodedBlockSize-1)  / unencodedBlockSize) * (long) encodedBlockSize;
496        if (lineLength > 0) { // We're using chunking
497            // Round up to nearest multiple
498            len += ((len + lineLength-1) / lineLength) * chunkSeparatorLength;
499        }
500        return len;
501    }
502}