Base64


1   /*
2    * Copyright 2001-2004 The Apache Software Foundation.
3    * 
4    * Licensed under the Apache License, Version 2.0 (the "License");
5    * you may not use this file except in compliance with the License.
6    * You may obtain a copy of the License at
7    * 
8    *      http://www.apache.org/licenses/LICENSE-2.0
9    * 
10   * Unless required by applicable law or agreed to in writing, software
11   * distributed under the License is distributed on an "AS IS" BASIS,
12   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13   * See the License for the specific language governing permissions and
14   * limitations under the License.
15   */ 
16  
17  package org.apache.commons.codec.binary;
18  
19  import org.apache.commons.codec.BinaryDecoder;
20  import org.apache.commons.codec.BinaryEncoder;
21  import org.apache.commons.codec.DecoderException;
22  import org.apache.commons.codec.EncoderException;
23  
24  /**
25   * Provides Base64 encoding and decoding as defined by RFC 2045.
26   * 
27   * <p>This class implements section <cite>6.8. Base64 Content-Transfer-Encoding</cite> 
28   * from RFC 2045 <cite>Multipurpose Internet Mail Extensions (MIME) Part One: 
29   * Format of Internet Message Bodies</cite> by Freed and Borenstein.</p> 
30   *
31   * @see <a HREF="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>
32   * @author Apache Software Foundation
33   * @since 1.0-dev
34   * @version $Id: Base64.java,v 1.20 2004/05/24 00:21:24 ggregory Exp $
35   */
36  public class Base64 implements BinaryEncoder, BinaryDecoder {
37  
38      /**
39       * Chunk size per RFC 2045 section 6.8.
40       * 
41       * <p>The {@value} character limit does not count the trailing CRLF, but counts 
42       * all other characters, including any equal signs.</p>
43       * 
44       * @see <a HREF="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 6.8</a>
45       */
46      static final int CHUNK_SIZE = 76;
47  
48      /**
49       * Chunk separator per RFC 2045 section 2.1.
50       * 
51       * @see <a HREF="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a>
52       */
53      static final byte[] CHUNK_SEPARATOR = "\r\n".getBytes();
54  
55      /**
56       * The base length.
57       */
58      static final int BASELENGTH = 255;
59  
60      /**
61       * Lookup length.
62       */
63      static final int LOOKUPLENGTH = 64;
64  
65      /**
66       * Used to calculate the number of bits in a byte.
67       */
68      static final int EIGHTBIT = 8;
69  
70      /**
71       * Used when encoding something which has fewer than 24 bits.
72       */
73      static final int SIXTEENBIT = 16;
74  
75      /**
76       * Used to determine how many bits data contains.
77       */
78      static final int TWENTYFOURBITGROUP = 24;
79  
80      /**
81       * Used to get the number of Quadruples.
82       */
83      static final int FOURBYTE = 4;
84  
85      /**
86       * Used to test the sign of a byte.
87       */
88      static final int SIGN = -128;
89      
90      /**
91       * Byte used to pad output.
92       */
93      static final byte PAD = (byte) '=';
94  
95      // Create arrays to hold the base64 characters and a 
96      // lookup for base64 chars
97      private static byte[] base64Alphabet = new byte[BASELENGTH];
98      private static byte[] lookUpBase64Alphabet = new byte[LOOKUPLENGTH];
99  
100     // Populating the lookup and character arrays
101     static {
102         for (int i = 0; i < BASELENGTH; i++) {
103             base64Alphabet[i] = (byte) -1;
104         }
105         for (int i = 'Z'; i >= 'A'; i--) {
106             base64Alphabet[i] = (byte) (i - 'A');
107         }
108         for (int i = 'z'; i >= 'a'; i--) {
109             base64Alphabet[i] = (byte) (i - 'a' + 26);
110         }
111         for (int i = '9'; i >= '0'; i--) {
112             base64Alphabet[i] = (byte) (i - '0' + 52);
113         }
114 
115         base64Alphabet['+'] = 62;
116         base64Alphabet['/'] = 63;
117 
118         for (int i = 0; i <= 25; i++) {
119             lookUpBase64Alphabet[i] = (byte) ('A' + i);
120         }
121 
122         for (int i = 26, j = 0; i <= 51; i++, j++) {
123             lookUpBase64Alphabet[i] = (byte) ('a' + j);
124         }
125 
126         for (int i = 52, j = 0; i <= 61; i++, j++) {
127             lookUpBase64Alphabet[i] = (byte) ('0' + j);
128         }
129 
130         lookUpBase64Alphabet[62] = (byte) '+';
131         lookUpBase64Alphabet[63] = (byte) '/';
132     }
133 
134     private static boolean isBase64(byte octect) {
135         if (octect == PAD) {
136             return true;
137         } else if (base64Alphabet[octect] == -1) {
138             return false;
139         } else {
140             return true;
141         }
142     }
143 
144     /**
145      * Tests a given byte array to see if it contains
146      * only valid characters within the Base64 alphabet.
147      *
148      * @param arrayOctect byte array to test
149      * @return true if all bytes are valid characters in the Base64
150      *         alphabet or if the byte array is empty; false, otherwise
151      */
152     public static boolean isArrayByteBase64(byte[] arrayOctect) {
153 
154         arrayOctect = discardWhitespace(arrayOctect);
155 
156         int length = arrayOctect.length;
157         if (length == 0) {
158             // shouldn't a 0 length array be valid base64 data?
159             // return false;
160             return true;
161         }
162         for (int i = 0; i < length; i++) {
163             if (!isBase64(arrayOctect[i])) {
164                 return false;
165             }
166         }
167         return true;
168     }
169 
170     /**
171      * Encodes binary data using the base64 algorithm but
172      * does not chunk the output.
173      *
174      * @param binaryData binary data to encode
175      * @return Base64 characters
176      */
177     public static byte[] encodeBase64(byte[] binaryData) {
178         return encodeBase64(binaryData, false);
179     }
180 
181     /**
182      * Encodes binary data using the base64 algorithm and chunks
183      * the encoded output into 76 character blocks
184      *
185      * @param binaryData binary data to encode
186      * @return Base64 characters chunked in 76 character blocks
187      */
188     public static byte[] encodeBase64Chunked(byte[] binaryData) {
189         return encodeBase64(binaryData, true);
190     }
191 
192 
193     /**
194      * Decodes an Object using the base64 algorithm.  This method
195      * is provided in order to satisfy the requirements of the
196      * Decoder interface, and will throw a DecoderException if the
197      * supplied object is not of type byte[].
198      *
199      * @param pObject Object to decode
200      * @return An object (of type byte[]) containing the 
201      *         binary data which corresponds to the byte[] supplied.
202      * @throws DecoderException if the parameter supplied is not
203      *                          of type byte[]
204      */
205     public Object   decode(Object   pObject) throws DecoderException {
206         if (!(pObject instanceof byte[])) {
207             throw new DecoderException("Parameter supplied to Base64 decode is not a byte[]");
208         }
209         return decode((byte[]) pObject);
210     }
211 
212     /**
213      * Decodes a byte[] containing containing
214      * characters in the Base64 alphabet.
215      *
216      * @param pArray A byte array containing Base64 character data
217      * @return a byte array containing binary data
218      */
219     public byte[] decode(byte[] pArray) {
220         return decodeBase64(pArray);
221     }
222 
223     /**
224      * Encodes binary data using the base64 algorithm, optionally
225      * chunking the output into 76 character blocks.
226      *
227      * @param binaryData Array containing binary data to encode.
228      * @param isChunked if isChunked is true this encoder will chunk
229      *                  the base64 output into 76 character blocks
230      * @return Base64-encoded data.
231      */
232     public static byte[] encodeBase64(byte[] binaryData, boolean isChunked) {
233         int lengthDataBits = binaryData.length * EIGHTBIT;
234         int fewerThan24bits = lengthDataBits % TWENTYFOURBITGROUP;
235         int numberTriplets = lengthDataBits / TWENTYFOURBITGROUP;
236         byte encodedData[] = null;
237         int encodedDataLength = 0;
238         int nbrChunks = 0;
239 
240         if (fewerThan24bits != 0) {
241             //data not divisible by 24 bit
242             encodedDataLength = (numberTriplets + 1) * 4;
243         } else {
244             // 16 or 8 bit
245             encodedDataLength = numberTriplets * 4;
246         }
247 
248         // If the output is to be "chunked" into 76 character sections, 
249         // for compliance with RFC 2045 MIME, then it is important to 
250         // allow for extra length to account for the separator(s)
251         if (isChunked) {
252 
253             nbrChunks =
254                 (CHUNK_SEPARATOR.length == 0 ? 0 : (int) Math.ceil((float) encodedDataLength / CHUNK_SIZE));
255             encodedDataLength += nbrChunks * CHUNK_SEPARATOR.length;
256         }
257 
258         encodedData = new byte[encodedDataLength];
259 
260         byte k = 0, l = 0, b1 = 0, b2 = 0, b3 = 0;
261 
262         int encodedIndex = 0;
263         int dataIndex = 0;
264         int i = 0;
265         int nextSeparatorIndex = CHUNK_SIZE;
266         int chunksSoFar = 0;
267 
268         //log.debug("number of triplets = " + numberTriplets);
269         for (i = 0; i < numberTriplets; i++) {
270             dataIndex = i * 3;
271             b1 = binaryData[dataIndex];
272             b2 = binaryData[dataIndex + 1];
273             b3 = binaryData[dataIndex + 2];
274 
275             //log.debug("b1= " + b1 +", b2= " + b2 + ", b3= " + b3);
276 
277             l = (byte) (b2 & 0x0f);
278             k = (byte) (b1 & 0x03);
279 
280             byte val1 =
281                 ((b1 & SIGN) == 0) ? (byte) (b1 >> 2) : (byte) ((b1) >> 2 ^ 0xc0);
282             byte val2 =
283                 ((b2 & SIGN) == 0) ? (byte) (b2 >> 4) : (byte) ((b2) >> 4 ^ 0xf0);
284             byte val3 =
285                 ((b3 & SIGN) == 0) ? (byte) (b3 >> 6) : (byte) ((b3) >> 6 ^ 0xfc);
286 
287             encodedData[encodedIndex] = lookUpBase64Alphabet[val1];
288             //log.debug( "val2 = " + val2 );
289             //log.debug( "k4   = " + (k<<4) );
290             //log.debug(  "vak  = " + (val2 | (k<<4)) );
291             encodedData[encodedIndex + 1] =
292                 lookUpBase64Alphabet[val2 | (k << 4)];
293             encodedData[encodedIndex + 2] =
294                 lookUpBase64Alphabet[(l << 2) | val3];
295             encodedData[encodedIndex + 3] = lookUpBase64Alphabet[b3 & 0x3f];
296 
297             encodedIndex += 4;
298 
299             // If we are chunking, let's put a chunk separator down.
300             if (isChunked) {
301                 // this assumes that CHUNK_SIZE % 4 == 0
302                 if (encodedIndex == nextSeparatorIndex) {
303                     System.arraycopy(
304                         CHUNK_SEPARATOR,
305                         0,
306                         encodedData,
307                         encodedIndex,
308                         CHUNK_SEPARATOR.length);
309                     chunksSoFar++;
310                     nextSeparatorIndex =
311                         (CHUNK_SIZE * (chunksSoFar + 1)) + 
312                         (chunksSoFar * CHUNK_SEPARATOR.length);
313                     encodedIndex += CHUNK_SEPARATOR.length;
314                 }
315             }
316         }
317 
318         // form integral number of 6-bit groups
319         dataIndex = i * 3;
320 
321         if (fewerThan24bits == EIGHTBIT) {
322             b1 = binaryData[dataIndex];
323             k = (byte) (b1 & 0x03);
324             //log.debug("b1=" + b1);
325             //log.debug("b1<<2 = " + (b1>>2) );
326             byte val1 =
327                 ((b1 & SIGN) == 0) ? (byte) (b1 >> 2) : (byte) ((b1) >> 2 ^ 0xc0);
328             encodedData[encodedIndex] = lookUpBase64Alphabet[val1];
329             encodedData[encodedIndex + 1] = lookUpBase64Alphabet[k << 4];
330             encodedData[encodedIndex + 2] = PAD;
331             encodedData[encodedIndex + 3] = PAD;
332         } else if (fewerThan24bits == SIXTEENBIT) {
333 
334             b1 = binaryData[dataIndex];
335             b2 = binaryData[dataIndex + 1];
336             l = (byte) (b2 & 0x0f);
337             k = (byte) (b1 & 0x03);
338 
339             byte val1 =
340                 ((b1 & SIGN) == 0) ? (byte) (b1 >> 2) : (byte) ((b1) >> 2 ^ 0xc0);
341             byte val2 =
342                 ((b2 & SIGN) == 0) ? (byte) (b2 >> 4) : (byte) ((b2) >> 4 ^ 0xf0);
343 
344             encodedData[encodedIndex] = lookUpBase64Alphabet[val1];
345             encodedData[encodedIndex + 1] =
346                 lookUpBase64Alphabet[val2 | (k << 4)];
347             encodedData[encodedIndex + 2] = lookUpBase64Alphabet[l << 2];
348             encodedData[encodedIndex + 3] = PAD;
349         }
350 
351         if (isChunked) {
352             // we also add a separator to the end of the final chunk.
353             if (chunksSoFar < nbrChunks) {
354                 System.arraycopy(
355                     CHUNK_SEPARATOR,
356                     0,
357                     encodedData,
358                     encodedDataLength - CHUNK_SEPARATOR.length,
359                     CHUNK_SEPARATOR.length);
360             }
361         }
362 
363         return encodedData;
364     }
365 
366     /**
367      * Decodes Base64 data into octects
368      *
369      * @param base64Data Byte array containing Base64 data
370      * @return Array containing decoded data.
371      */
372     public static byte[] decodeBase64(byte[] base64Data) {
373         // RFC 2045 requires that we discard ALL non-Base64 characters
374         base64Data = discardNonBase64(base64Data);
375 
376         // handle the edge case, so we don't have to worry about it later
377         if (base64Data.length == 0) {
378             return new byte[0];
379         }
380 
381         int numberQuadruple = base64Data.length / FOURBYTE;
382         byte decodedData[] = null;
383         byte b1 = 0, b2 = 0, b3 = 0, b4 = 0, marker0 = 0, marker1 = 0;
384 
385         // Throw away anything not in base64Data
386 
387         int encodedIndex = 0;
388         int dataIndex = 0;
389         {
390             // this sizes the output array properly - rlw
391             int lastData = base64Data.length;
392             // ignore the '=' padding
393             while (base64Data[lastData - 1] == PAD) {
394                 if (--lastData == 0) {
395                     return new byte[0];
396                 }
397             }
398             decodedData = new byte[lastData - numberQuadruple];
399         }
400         
401         for (int i = 0; i < numberQuadruple; i++) {
402             dataIndex = i * 4;
403             marker0 = base64Data[dataIndex + 2];
404             marker1 = base64Data[dataIndex + 3];
405             
406             b1 = base64Alphabet[base64Data[dataIndex]];
407             b2 = base64Alphabet[base64Data[dataIndex + 1]];
408             
409             if (marker0 != PAD && marker1 != PAD) {
410                 //No PAD e.g 3cQl
411                 b3 = base64Alphabet[marker0];
412                 b4 = base64Alphabet[marker1];
413                 
414                 decodedData[encodedIndex] = (byte) (b1 << 2 | b2 >> 4);
415                 decodedData[encodedIndex + 1] =
416                     (byte) (((b2 & 0xf) << 4) | ((b3 >> 2) & 0xf));
417                 decodedData[encodedIndex + 2] = (byte) (b3 << 6 | b4);
418             } else if (marker0 == PAD) {
419                 //Two PAD e.g. 3c[Pad][Pad]
420                 decodedData[encodedIndex] = (byte) (b1 << 2 | b2 >> 4);
421             } else if (marker1 == PAD) {
422                 //One PAD e.g. 3cQ[Pad]
423                 b3 = base64Alphabet[marker0];
424                 
425                 decodedData[encodedIndex] = (byte) (b1 << 2 | b2 >> 4);
426                 decodedData[encodedIndex + 1] =
427                     (byte) (((b2 & 0xf) << 4) | ((b3 >> 2) & 0xf));
428             }
429             encodedIndex += 3;
430         }
431         return decodedData;
432     }
433     
434     /**
435      * Discards any whitespace from a base-64 encoded block.
436      *
437      * @param data The base-64 encoded data to discard the whitespace
438      * from.
439      * @return The data, less whitespace (see RFC 2045).
440      */
441     static byte[] discardWhitespace(byte[] data) {
442         byte groomedData[] = new byte[data.length];
443         int bytesCopied = 0;
444         
445         for (int i = 0; i < data.length; i++) {
446             switch (data[i]) {
447             case (byte) ' ' :
448             case (byte) '\n' :
449             case (byte) '\r' :
450             case (byte) '\t' :
451                     break;
452             default:
453                     groomedData[bytesCopied++] = data[i];
454             }
455         }
456 
457         byte packedData[] = new byte[bytesCopied];
458 
459         System.arraycopy(groomedData, 0, packedData, 0, bytesCopied);
460 
461         return packedData;
462     }
463 
464     /**
465      * Discards any characters outside of the base64 alphabet, per
466      * the requirements on page 25 of RFC 2045 - "Any characters
467      * outside of the base64 alphabet are to be ignored in base64
468      * encoded data."
469      *
470      * @param data The base-64 encoded data to groom
471      * @return The data, less non-base64 characters (see RFC 2045).
472      */
473     static byte[] discardNonBase64(byte[] data) {
474         byte groomedData[] = new byte[data.length];
475         int bytesCopied = 0;
476 
477         for (int i = 0; i < data.length; i++) {
478             if (isBase64(data[i])) {
479                 groomedData[bytesCopied++] = data[i];
480             }
481         }
482 
483         byte packedData[] = new byte[bytesCopied];
484 
485         System.arraycopy(groomedData, 0, packedData, 0, bytesCopied);
486 
487         return packedData;
488     }
489 
490 
491     // Implementation of the Encoder Interface
492 
493     /**
494      * Encodes an Object using the base64 algorithm.  This method
495      * is provided in order to satisfy the requirements of the
496      * Encoder interface, and will throw an EncoderException if the
497      * supplied object is not of type byte[].
498      *
499      * @param pObject Object to encode
500      * @return An object (of type byte[]) containing the 
501      *         base64 encoded data which corresponds to the byte[] supplied.
502      * @throws EncoderException if the parameter supplied is not
503      *                          of type byte[]
504      */
505     public Object   encode(Object   pObject) throws EncoderException {
506         if (!(pObject instanceof byte[])) {
507             throw new EncoderException(
508                 "Parameter supplied to Base64 encode is not a byte[]");
509         }
510         return encode((byte[]) pObject);
511     }
512 
513     /**
514      * Encodes a byte[] containing binary data, into a byte[] containing
515      * characters in the Base64 alphabet.
516      *
517      * @param pArray a byte array containing binary data
518      * @return A byte array containing only Base64 character data
519      */
520     public byte[] encode(byte[] pArray) {
521         return encodeBase64(pArray, false);
522     }
523 
524 }
525
A to Z: JavaDoc & Examples Daily Java News & Articles Open Source Projects Open Source Codes Free Computer Books Remove Frame
Popular Tags