|
|
@@ -0,0 +1,758 @@
|
|
|
1
|
+package net.miginfocom;
|
|
|
2
|
+
|
|
|
3
|
+import java.util.Arrays;
|
|
|
4
|
+
|
|
|
5
|
+/**
|
|
|
6
|
+ * A very fast and memory efficient class to encode and decode to and from
|
|
|
7
|
+ * BASE64 in full accordance
|
|
|
8
|
+ * with RFC 2045.<br>
|
|
|
9
|
+ * <br>
|
|
|
10
|
+ * On Windows XP sp1 with 1.4.2_04 and later ;), this encoder and decoder is
|
|
|
11
|
+ * about 10 times faster
|
|
|
12
|
+ * on small arrays (10 - 1000 bytes) and 2-3 times as fast on larger arrays
|
|
|
13
|
+ * (10000 - 1000000 bytes)
|
|
|
14
|
+ * compared to <code>sun.misc.Encoder()/Decoder()</code>.<br>
|
|
|
15
|
+ * <br>
|
|
|
16
|
+ *
|
|
|
17
|
+ * On byte arrays the encoder is about 20% faster than Jakarta Commons Base64
|
|
|
18
|
+ * Codec for encode and
|
|
|
19
|
+ * about 50% faster for decoding large arrays. This implementation is about
|
|
|
20
|
+ * twice as fast on very small
|
|
|
21
|
+ * arrays (< 30 bytes). If source/destination is a <code>String</code> this
|
|
|
22
|
+ * version is about three times as fast due to the fact that the Commons Codec
|
|
|
23
|
+ * result has to be recoded
|
|
|
24
|
+ * to a <code>String</code> from <code>byte[]</code>, which is very expensive.<br>
|
|
|
25
|
+ * <br>
|
|
|
26
|
+ *
|
|
|
27
|
+ * This encode/decode algorithm doesn't create any temporary arrays as many
|
|
|
28
|
+ * other codecs do, it only
|
|
|
29
|
+ * allocates the resulting array. This produces less garbage and it is possible
|
|
|
30
|
+ * to handle arrays twice
|
|
|
31
|
+ * as large as algorithms that create a temporary array. (E.g. Jakarta Commons
|
|
|
32
|
+ * Codec). It is unknown
|
|
|
33
|
+ * whether Sun's <code>sun.misc.Encoder()/Decoder()</code> produce temporary
|
|
|
34
|
+ * arrays but since performance
|
|
|
35
|
+ * is quite low it probably does.<br>
|
|
|
36
|
+ * <br>
|
|
|
37
|
+ *
|
|
|
38
|
+ * The encoder produces the same output as the Sun one except that the Sun's
|
|
|
39
|
+ * encoder appends
|
|
|
40
|
+ * a trailing line separator if the last character isn't a pad. Unclear why but
|
|
|
41
|
+ * it only adds to the
|
|
|
42
|
+ * length and is probably a side effect. Both are in conformance with RFC 2045
|
|
|
43
|
+ * though.<br>
|
|
|
44
|
+ * Commons codec seem to always att a trailing line separator.<br>
|
|
|
45
|
+ * <br>
|
|
|
46
|
+ *
|
|
|
47
|
+ * <b>Note!</b>
|
|
|
48
|
+ * The encode/decode method pairs (types) come in three versions with the
|
|
|
49
|
+ * <b>exact</b> same algorithm and
|
|
|
50
|
+ * thus a lot of code redundancy. This is to not create any temporary arrays for
|
|
|
51
|
+ * transcoding to/from different
|
|
|
52
|
+ * format types. The methods not used can simply be commented out.<br>
|
|
|
53
|
+ * <br>
|
|
|
54
|
+ *
|
|
|
55
|
+ * There is also a "fast" version of all decode methods that works the same way
|
|
|
56
|
+ * as the normal ones, but
|
|
|
57
|
+ * har a few demands on the decoded input. Normally though, these fast verions
|
|
|
58
|
+ * should be used if the source if
|
|
|
59
|
+ * the input is known and it hasn't bee tampered with.<br>
|
|
|
60
|
+ * <br>
|
|
|
61
|
+ *
|
|
|
62
|
+ * If you find the code useful or you find a bug, please send me a note at
|
|
|
63
|
+ * base64 @ miginfocom . com.
|
|
|
64
|
+ *
|
|
|
65
|
+ * Licence (BSD):
|
|
|
66
|
+ * ==============
|
|
|
67
|
+ *
|
|
|
68
|
+ * Copyright (c) 2004, Mikael Grev, MiG InfoCom AB. (base64 @ miginfocom . com)
|
|
|
69
|
+ * All rights reserved.
|
|
|
70
|
+ *
|
|
|
71
|
+ * Redistribution and use in source and binary forms, with or without
|
|
|
72
|
+ * modification,
|
|
|
73
|
+ * are permitted provided that the following conditions are met:
|
|
|
74
|
+ * Redistributions of source code must retain the above copyright notice, this
|
|
|
75
|
+ * list
|
|
|
76
|
+ * of conditions and the following disclaimer.
|
|
|
77
|
+ * Redistributions in binary form must reproduce the above copyright notice,
|
|
|
78
|
+ * this
|
|
|
79
|
+ * list of conditions and the following disclaimer in the documentation and/or
|
|
|
80
|
+ * other
|
|
|
81
|
+ * materials provided with the distribution.
|
|
|
82
|
+ * Neither the name of the MiG InfoCom AB nor the names of its contributors may
|
|
|
83
|
+ * be
|
|
|
84
|
+ * used to endorse or promote products derived from this software without
|
|
|
85
|
+ * specific
|
|
|
86
|
+ * prior written permission.
|
|
|
87
|
+ *
|
|
|
88
|
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
|
|
89
|
+ * AND
|
|
|
90
|
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
|
|
|
91
|
+ * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
|
|
92
|
+ * DISCLAIMED.
|
|
|
93
|
+ * IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY
|
|
|
94
|
+ * DIRECT,
|
|
|
95
|
+ * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
|
|
|
96
|
+ * (INCLUDING,
|
|
|
97
|
+ * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
|
98
|
+ * DATA,
|
|
|
99
|
+ * OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
|
|
|
100
|
+ * LIABILITY,
|
|
|
101
|
+ * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
|
|
|
102
|
+ * OTHERWISE)
|
|
|
103
|
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
|
|
104
|
+ * POSSIBILITY
|
|
|
105
|
+ * OF SUCH DAMAGE.
|
|
|
106
|
+ *
|
|
|
107
|
+ * @version 2.2
|
|
|
108
|
+ * @author Mikael Grev
|
|
|
109
|
+ * Date: 2004-aug-02
|
|
|
110
|
+ * Time: 11:31:11
|
|
|
111
|
+ */
|
|
|
112
|
+
|
|
|
113
|
+public class Base64 {
|
|
|
114
|
+ private static final char[] CA = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
|
|
|
115
|
+ .toCharArray();
|
|
|
116
|
+ private static final int[] IA = new int[256];
|
|
|
117
|
+ static {
|
|
|
118
|
+ Arrays.fill(IA, -1);
|
|
|
119
|
+ for (int i = 0, iS = CA.length; i < iS; i++) {
|
|
|
120
|
+ IA[CA[i]] = i;
|
|
|
121
|
+ }
|
|
|
122
|
+ IA['='] = 0;
|
|
|
123
|
+ }
|
|
|
124
|
+
|
|
|
125
|
+ // ****************************************************************************************
|
|
|
126
|
+ // * char[] version
|
|
|
127
|
+ // ****************************************************************************************
|
|
|
128
|
+
|
|
|
129
|
+ /**
|
|
|
130
|
+ * Encodes a raw byte array into a BASE64 <code>char[]</code> representation
|
|
|
131
|
+ * i accordance with RFC 2045.
|
|
|
132
|
+ *
|
|
|
133
|
+ * @param sArr
|
|
|
134
|
+ * The bytes to convert. If <code>null</code> or length 0 an
|
|
|
135
|
+ * empty array will be returned.
|
|
|
136
|
+ * @param lineSep
|
|
|
137
|
+ * Optional "\r\n" after 76 characters, unless end of file.<br>
|
|
|
138
|
+ * No line separator will be in breach of RFC 2045 which
|
|
|
139
|
+ * specifies max 76 per line but will be a
|
|
|
140
|
+ * little faster.
|
|
|
141
|
+ * @return A BASE64 encoded array. Never <code>null</code>.
|
|
|
142
|
+ */
|
|
|
143
|
+ public static final char[] encodeToChar(final byte[] sArr,
|
|
|
144
|
+ final boolean lineSep) {
|
|
|
145
|
+ // Check special case
|
|
|
146
|
+ final int sLen = sArr != null ? sArr.length : 0;
|
|
|
147
|
+ if (sLen == 0) {
|
|
|
148
|
+ return new char[0];
|
|
|
149
|
+ }
|
|
|
150
|
+
|
|
|
151
|
+ final int eLen = sLen / 3 * 3; // Length of even 24-bits.
|
|
|
152
|
+ final int cCnt = (sLen - 1) / 3 + 1 << 2; // Returned character count
|
|
|
153
|
+ final int dLen = cCnt + (lineSep ? (cCnt - 1) / 76 << 1 : 0); // Length
|
|
|
154
|
+ // of
|
|
|
155
|
+ // returned
|
|
|
156
|
+ // array
|
|
|
157
|
+ final char[] dArr = new char[dLen];
|
|
|
158
|
+
|
|
|
159
|
+ // Encode even 24-bits
|
|
|
160
|
+ for (int s = 0, d = 0, cc = 0; s < eLen;) {
|
|
|
161
|
+ // Copy next three bytes into lower 24 bits of int, paying attension
|
|
|
162
|
+ // to sign.
|
|
|
163
|
+ final int i = (sArr[s++] & 0xff) << 16 | (sArr[s++] & 0xff) << 8
|
|
|
164
|
+ | sArr[s++] & 0xff;
|
|
|
165
|
+
|
|
|
166
|
+ // Encode the int into four chars
|
|
|
167
|
+ dArr[d++] = CA[i >>> 18 & 0x3f];
|
|
|
168
|
+ dArr[d++] = CA[i >>> 12 & 0x3f];
|
|
|
169
|
+ dArr[d++] = CA[i >>> 6 & 0x3f];
|
|
|
170
|
+ dArr[d++] = CA[i & 0x3f];
|
|
|
171
|
+
|
|
|
172
|
+ // Add optional line separator
|
|
|
173
|
+ if (lineSep && ++cc == 19 && d < dLen - 2) {
|
|
|
174
|
+ dArr[d++] = '\r';
|
|
|
175
|
+ dArr[d++] = '\n';
|
|
|
176
|
+ cc = 0;
|
|
|
177
|
+ }
|
|
|
178
|
+ }
|
|
|
179
|
+
|
|
|
180
|
+ // Pad and encode last bits if source isn't even 24 bits.
|
|
|
181
|
+ final int left = sLen - eLen; // 0 - 2.
|
|
|
182
|
+ if (left > 0) {
|
|
|
183
|
+ // Prepare the int
|
|
|
184
|
+ final int i = (sArr[eLen] & 0xff) << 10
|
|
|
185
|
+ | (left == 2 ? (sArr[sLen - 1] & 0xff) << 2 : 0);
|
|
|
186
|
+
|
|
|
187
|
+ // Set last four chars
|
|
|
188
|
+ dArr[dLen - 4] = CA[i >> 12];
|
|
|
189
|
+ dArr[dLen - 3] = CA[i >>> 6 & 0x3f];
|
|
|
190
|
+ dArr[dLen - 2] = left == 2 ? CA[i & 0x3f] : '=';
|
|
|
191
|
+ dArr[dLen - 1] = '=';
|
|
|
192
|
+ }
|
|
|
193
|
+ return dArr;
|
|
|
194
|
+ }
|
|
|
195
|
+
|
|
|
196
|
+ /**
|
|
|
197
|
+ * Decodes a BASE64 encoded char array. All illegal characters will be
|
|
|
198
|
+ * ignored and can handle both arrays with
|
|
|
199
|
+ * and without line separators.
|
|
|
200
|
+ *
|
|
|
201
|
+ * @param sArr
|
|
|
202
|
+ * The source array. <code>null</code> or length 0 will return an
|
|
|
203
|
+ * empty array.
|
|
|
204
|
+ * @return The decoded array of bytes. May be of length 0. Will be
|
|
|
205
|
+ * <code>null</code> if the legal characters
|
|
|
206
|
+ * (including '=') isn't divideable by 4. (I.e. definitely
|
|
|
207
|
+ * corrupted).
|
|
|
208
|
+ */
|
|
|
209
|
+ public static final byte[] decode(final char[] sArr) {
|
|
|
210
|
+ // Check special case
|
|
|
211
|
+ final int sLen = sArr != null ? sArr.length : 0;
|
|
|
212
|
+ if (sLen == 0) {
|
|
|
213
|
+ return new byte[0];
|
|
|
214
|
+ }
|
|
|
215
|
+
|
|
|
216
|
+ // Count illegal characters (including '\r', '\n') to know what size the
|
|
|
217
|
+ // returned array will be,
|
|
|
218
|
+ // so we don't have to reallocate & copy it later.
|
|
|
219
|
+ int sepCnt = 0; // Number of separator characters. (Actually illegal
|
|
|
220
|
+ // characters, but that's a bonus...)
|
|
|
221
|
+ for (int i = 0; i < sLen; i++) {
|
|
|
222
|
+ if (IA[sArr[i]] < 0) {
|
|
|
223
|
+ sepCnt++;
|
|
|
224
|
+ }
|
|
|
225
|
+ }
|
|
|
226
|
+
|
|
|
227
|
+ // Check so that legal chars (including '=') are evenly divideable by 4
|
|
|
228
|
+ // as specified in RFC 2045.
|
|
|
229
|
+ if ((sLen - sepCnt) % 4 != 0) {
|
|
|
230
|
+ return null;
|
|
|
231
|
+ }
|
|
|
232
|
+
|
|
|
233
|
+ int pad = 0;
|
|
|
234
|
+ for (int i = sLen; i > 1 && IA[sArr[--i]] <= 0;) {
|
|
|
235
|
+ if (sArr[i] == '=') {
|
|
|
236
|
+ pad++;
|
|
|
237
|
+ }
|
|
|
238
|
+ }
|
|
|
239
|
+
|
|
|
240
|
+ final int len = ((sLen - sepCnt) * 6 >> 3) - pad;
|
|
|
241
|
+
|
|
|
242
|
+ final byte[] dArr = new byte[len]; // Preallocate byte[] of exact length
|
|
|
243
|
+
|
|
|
244
|
+ for (int s = 0, d = 0; d < len;) {
|
|
|
245
|
+ // Assemble three bytes into an int from four "valid" characters.
|
|
|
246
|
+ int i = 0;
|
|
|
247
|
+ for (int j = 0; j < 4; j++) { // j only increased if a valid char
|
|
|
248
|
+ // was found.
|
|
|
249
|
+ final int c = IA[sArr[s++]];
|
|
|
250
|
+ if (c >= 0) {
|
|
|
251
|
+ i |= c << 18 - j * 6;
|
|
|
252
|
+ } else {
|
|
|
253
|
+ j--;
|
|
|
254
|
+ }
|
|
|
255
|
+ }
|
|
|
256
|
+ // Add the bytes
|
|
|
257
|
+ dArr[d++] = (byte) (i >> 16);
|
|
|
258
|
+ if (d < len) {
|
|
|
259
|
+ dArr[d++] = (byte) (i >> 8);
|
|
|
260
|
+ if (d < len) {
|
|
|
261
|
+ dArr[d++] = (byte) i;
|
|
|
262
|
+ }
|
|
|
263
|
+ }
|
|
|
264
|
+ }
|
|
|
265
|
+ return dArr;
|
|
|
266
|
+ }
|
|
|
267
|
+
|
|
|
268
|
+ /**
|
|
|
269
|
+ * Decodes a BASE64 encoded char array that is known to be resonably well
|
|
|
270
|
+ * formatted. The method is about twice as
|
|
|
271
|
+ * fast as {@link #decode(char[])}. The preconditions are:<br>
|
|
|
272
|
+ * + The array must have a line length of 76 chars OR no line separators at
|
|
|
273
|
+ * all (one line).<br>
|
|
|
274
|
+ * + Line separator must be "\r\n", as specified in RFC 2045
|
|
|
275
|
+ * + The array must not contain illegal characters within the encoded string<br>
|
|
|
276
|
+ * + The array CAN have illegal characters at the beginning and end, those
|
|
|
277
|
+ * will be dealt with appropriately.<br>
|
|
|
278
|
+ *
|
|
|
279
|
+ * @param sArr
|
|
|
280
|
+ * The source array. Length 0 will return an empty array.
|
|
|
281
|
+ * <code>null</code> will throw an exception.
|
|
|
282
|
+ * @return The decoded array of bytes. May be of length 0.
|
|
|
283
|
+ */
|
|
|
284
|
+ public static final byte[] decodeFast(final char[] sArr) {
|
|
|
285
|
+ // Check special case
|
|
|
286
|
+ final int sLen = sArr.length;
|
|
|
287
|
+ if (sLen == 0) {
|
|
|
288
|
+ return new byte[0];
|
|
|
289
|
+ }
|
|
|
290
|
+
|
|
|
291
|
+ int sIx = 0, eIx = sLen - 1; // Start and end index after trimming.
|
|
|
292
|
+
|
|
|
293
|
+ // Trim illegal chars from start
|
|
|
294
|
+ while (sIx < eIx && IA[sArr[sIx]] < 0) {
|
|
|
295
|
+ sIx++;
|
|
|
296
|
+ }
|
|
|
297
|
+
|
|
|
298
|
+ // Trim illegal chars from end
|
|
|
299
|
+ while (eIx > 0 && IA[sArr[eIx]] < 0) {
|
|
|
300
|
+ eIx--;
|
|
|
301
|
+ }
|
|
|
302
|
+
|
|
|
303
|
+ // get the padding count (=) (0, 1 or 2)
|
|
|
304
|
+ final int pad = sArr[eIx] == '=' ? (sArr[eIx - 1] == '=' ? 2 : 1) : 0; // Count
|
|
|
305
|
+ // '='
|
|
|
306
|
+ // at
|
|
|
307
|
+ // end.
|
|
|
308
|
+ final int cCnt = eIx - sIx + 1; // Content count including possible
|
|
|
309
|
+ // separators
|
|
|
310
|
+ final int sepCnt = sLen > 76 ? (sArr[76] == '\r' ? cCnt / 78 : 0) << 1
|
|
|
311
|
+ : 0;
|
|
|
312
|
+
|
|
|
313
|
+ final int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of
|
|
|
314
|
+ // decoded bytes
|
|
|
315
|
+ final byte[] dArr = new byte[len]; // Preallocate byte[] of exact length
|
|
|
316
|
+
|
|
|
317
|
+ // Decode all but the last 0 - 2 bytes.
|
|
|
318
|
+ int d = 0;
|
|
|
319
|
+ for (int cc = 0, eLen = len / 3 * 3; d < eLen;) {
|
|
|
320
|
+ // Assemble three bytes into an int from four "valid" characters.
|
|
|
321
|
+ final int i = IA[sArr[sIx++]] << 18 | IA[sArr[sIx++]] << 12
|
|
|
322
|
+ | IA[sArr[sIx++]] << 6 | IA[sArr[sIx++]];
|
|
|
323
|
+
|
|
|
324
|
+ // Add the bytes
|
|
|
325
|
+ dArr[d++] = (byte) (i >> 16);
|
|
|
326
|
+ dArr[d++] = (byte) (i >> 8);
|
|
|
327
|
+ dArr[d++] = (byte) i;
|
|
|
328
|
+
|
|
|
329
|
+ // If line separator, jump over it.
|
|
|
330
|
+ if (sepCnt > 0 && ++cc == 19) {
|
|
|
331
|
+ sIx += 2;
|
|
|
332
|
+ cc = 0;
|
|
|
333
|
+ }
|
|
|
334
|
+ }
|
|
|
335
|
+
|
|
|
336
|
+ if (d < len) {
|
|
|
337
|
+ // Decode last 1-3 bytes (incl '=') into 1-3 bytes
|
|
|
338
|
+ int i = 0;
|
|
|
339
|
+ for (int j = 0; sIx <= eIx - pad; j++) {
|
|
|
340
|
+ i |= IA[sArr[sIx++]] << 18 - j * 6;
|
|
|
341
|
+ }
|
|
|
342
|
+
|
|
|
343
|
+ for (int r = 16; d < len; r -= 8) {
|
|
|
344
|
+ dArr[d++] = (byte) (i >> r);
|
|
|
345
|
+ }
|
|
|
346
|
+ }
|
|
|
347
|
+
|
|
|
348
|
+ return dArr;
|
|
|
349
|
+ }
|
|
|
350
|
+
|
|
|
351
|
+ // ****************************************************************************************
|
|
|
352
|
+ // * byte[] version
|
|
|
353
|
+ // ****************************************************************************************
|
|
|
354
|
+
|
|
|
355
|
+ /**
|
|
|
356
|
+ * Encodes a raw byte array into a BASE64 <code>byte[]</code> representation
|
|
|
357
|
+ * i accordance with RFC 2045.
|
|
|
358
|
+ *
|
|
|
359
|
+ * @param sArr
|
|
|
360
|
+ * The bytes to convert. If <code>null</code> or length 0 an
|
|
|
361
|
+ * empty array will be returned.
|
|
|
362
|
+ * @param lineSep
|
|
|
363
|
+ * Optional "\r\n" after 76 characters, unless end of file.<br>
|
|
|
364
|
+ * No line separator will be in breach of RFC 2045 which
|
|
|
365
|
+ * specifies max 76 per line but will be a
|
|
|
366
|
+ * little faster.
|
|
|
367
|
+ * @return A BASE64 encoded array. Never <code>null</code>.
|
|
|
368
|
+ */
|
|
|
369
|
+ public static final byte[] encodeToByte(final byte[] sArr,
|
|
|
370
|
+ final boolean lineSep) {
|
|
|
371
|
+ // Check special case
|
|
|
372
|
+ final int sLen = sArr != null ? sArr.length : 0;
|
|
|
373
|
+ if (sLen == 0) {
|
|
|
374
|
+ return new byte[0];
|
|
|
375
|
+ }
|
|
|
376
|
+
|
|
|
377
|
+ final int eLen = sLen / 3 * 3; // Length of even 24-bits.
|
|
|
378
|
+ final int cCnt = (sLen - 1) / 3 + 1 << 2; // Returned character count
|
|
|
379
|
+ final int dLen = cCnt + (lineSep ? (cCnt - 1) / 76 << 1 : 0); // Length
|
|
|
380
|
+ // of
|
|
|
381
|
+ // returned
|
|
|
382
|
+ // array
|
|
|
383
|
+ final byte[] dArr = new byte[dLen];
|
|
|
384
|
+
|
|
|
385
|
+ // Encode even 24-bits
|
|
|
386
|
+ for (int s = 0, d = 0, cc = 0; s < eLen;) {
|
|
|
387
|
+ // Copy next three bytes into lower 24 bits of int, paying attension
|
|
|
388
|
+ // to sign.
|
|
|
389
|
+ final int i = (sArr[s++] & 0xff) << 16 | (sArr[s++] & 0xff) << 8
|
|
|
390
|
+ | sArr[s++] & 0xff;
|
|
|
391
|
+
|
|
|
392
|
+ // Encode the int into four chars
|
|
|
393
|
+ dArr[d++] = (byte) CA[i >>> 18 & 0x3f];
|
|
|
394
|
+ dArr[d++] = (byte) CA[i >>> 12 & 0x3f];
|
|
|
395
|
+ dArr[d++] = (byte) CA[i >>> 6 & 0x3f];
|
|
|
396
|
+ dArr[d++] = (byte) CA[i & 0x3f];
|
|
|
397
|
+
|
|
|
398
|
+ // Add optional line separator
|
|
|
399
|
+ if (lineSep && ++cc == 19 && d < dLen - 2) {
|
|
|
400
|
+ dArr[d++] = '\r';
|
|
|
401
|
+ dArr[d++] = '\n';
|
|
|
402
|
+ cc = 0;
|
|
|
403
|
+ }
|
|
|
404
|
+ }
|
|
|
405
|
+
|
|
|
406
|
+ // Pad and encode last bits if source isn't an even 24 bits.
|
|
|
407
|
+ final int left = sLen - eLen; // 0 - 2.
|
|
|
408
|
+ if (left > 0) {
|
|
|
409
|
+ // Prepare the int
|
|
|
410
|
+ final int i = (sArr[eLen] & 0xff) << 10
|
|
|
411
|
+ | (left == 2 ? (sArr[sLen - 1] & 0xff) << 2 : 0);
|
|
|
412
|
+
|
|
|
413
|
+ // Set last four chars
|
|
|
414
|
+ dArr[dLen - 4] = (byte) CA[i >> 12];
|
|
|
415
|
+ dArr[dLen - 3] = (byte) CA[i >>> 6 & 0x3f];
|
|
|
416
|
+ dArr[dLen - 2] = left == 2 ? (byte) CA[i & 0x3f] : (byte) '=';
|
|
|
417
|
+ dArr[dLen - 1] = '=';
|
|
|
418
|
+ }
|
|
|
419
|
+ return dArr;
|
|
|
420
|
+ }
|
|
|
421
|
+
|
|
|
422
|
+ /**
|
|
|
423
|
+ * Decodes a BASE64 encoded byte array. All illegal characters will be
|
|
|
424
|
+ * ignored and can handle both arrays with
|
|
|
425
|
+ * and without line separators.
|
|
|
426
|
+ *
|
|
|
427
|
+ * @param sArr
|
|
|
428
|
+ * The source array. Length 0 will return an empty array.
|
|
|
429
|
+ * <code>null</code> will throw an exception.
|
|
|
430
|
+ * @return The decoded array of bytes. May be of length 0. Will be
|
|
|
431
|
+ * <code>null</code> if the legal characters
|
|
|
432
|
+ * (including '=') isn't divideable by 4. (I.e. definitely
|
|
|
433
|
+ * corrupted).
|
|
|
434
|
+ */
|
|
|
435
|
+ public static final byte[] decode(final byte[] sArr) {
|
|
|
436
|
+ // Check special case
|
|
|
437
|
+ final int sLen = sArr.length;
|
|
|
438
|
+
|
|
|
439
|
+ // Count illegal characters (including '\r', '\n') to know what size the
|
|
|
440
|
+ // returned array will be,
|
|
|
441
|
+ // so we don't have to reallocate & copy it later.
|
|
|
442
|
+ int sepCnt = 0; // Number of separator characters. (Actually illegal
|
|
|
443
|
+ // characters, but that's a bonus...)
|
|
|
444
|
+ for (int i = 0; i < sLen; i++) {
|
|
|
445
|
+ if (IA[sArr[i] & 0xff] < 0) {
|
|
|
446
|
+ sepCnt++;
|
|
|
447
|
+ }
|
|
|
448
|
+ }
|
|
|
449
|
+
|
|
|
450
|
+ // Check so that legal chars (including '=') are evenly divideable by 4
|
|
|
451
|
+ // as specified in RFC 2045.
|
|
|
452
|
+ if ((sLen - sepCnt) % 4 != 0) {
|
|
|
453
|
+ return null;
|
|
|
454
|
+ }
|
|
|
455
|
+
|
|
|
456
|
+ int pad = 0;
|
|
|
457
|
+ for (int i = sLen; i > 1 && IA[sArr[--i] & 0xff] <= 0;) {
|
|
|
458
|
+ if (sArr[i] == '=') {
|
|
|
459
|
+ pad++;
|
|
|
460
|
+ }
|
|
|
461
|
+ }
|
|
|
462
|
+
|
|
|
463
|
+ final int len = ((sLen - sepCnt) * 6 >> 3) - pad;
|
|
|
464
|
+
|
|
|
465
|
+ final byte[] dArr = new byte[len]; // Preallocate byte[] of exact length
|
|
|
466
|
+
|
|
|
467
|
+ for (int s = 0, d = 0; d < len;) {
|
|
|
468
|
+ // Assemble three bytes into an int from four "valid" characters.
|
|
|
469
|
+ int i = 0;
|
|
|
470
|
+ for (int j = 0; j < 4; j++) { // j only increased if a valid char
|
|
|
471
|
+ // was found.
|
|
|
472
|
+ final int c = IA[sArr[s++] & 0xff];
|
|
|
473
|
+ if (c >= 0) {
|
|
|
474
|
+ i |= c << 18 - j * 6;
|
|
|
475
|
+ } else {
|
|
|
476
|
+ j--;
|
|
|
477
|
+ }
|
|
|
478
|
+ }
|
|
|
479
|
+
|
|
|
480
|
+ // Add the bytes
|
|
|
481
|
+ dArr[d++] = (byte) (i >> 16);
|
|
|
482
|
+ if (d < len) {
|
|
|
483
|
+ dArr[d++] = (byte) (i >> 8);
|
|
|
484
|
+ if (d < len) {
|
|
|
485
|
+ dArr[d++] = (byte) i;
|
|
|
486
|
+ }
|
|
|
487
|
+ }
|
|
|
488
|
+ }
|
|
|
489
|
+
|
|
|
490
|
+ return dArr;
|
|
|
491
|
+ }
|
|
|
492
|
+
|
|
|
493
|
+ /**
|
|
|
494
|
+ * Decodes a BASE64 encoded byte array that is known to be resonably well
|
|
|
495
|
+ * formatted. The method is about twice as
|
|
|
496
|
+ * fast as {@link #decode(byte[])}. The preconditions are:<br>
|
|
|
497
|
+ * + The array must have a line length of 76 chars OR no line separators at
|
|
|
498
|
+ * all (one line).<br>
|
|
|
499
|
+ * + Line separator must be "\r\n", as specified in RFC 2045
|
|
|
500
|
+ * + The array must not contain illegal characters within the encoded string<br>
|
|
|
501
|
+ * + The array CAN have illegal characters at the beginning and end, those
|
|
|
502
|
+ * will be dealt with appropriately.<br>
|
|
|
503
|
+ *
|
|
|
504
|
+ * @param sArr
|
|
|
505
|
+ * The source array. Length 0 will return an empty array.
|
|
|
506
|
+ * <code>null</code> will throw an exception.
|
|
|
507
|
+ * @return The decoded array of bytes. May be of length 0.
|
|
|
508
|
+ */
|
|
|
509
|
+ public static final byte[] decodeFast(final byte[] sArr) {
|
|
|
510
|
+ // Check special case
|
|
|
511
|
+ final int sLen = sArr.length;
|
|
|
512
|
+ if (sLen == 0) {
|
|
|
513
|
+ return new byte[0];
|
|
|
514
|
+ }
|
|
|
515
|
+
|
|
|
516
|
+ int sIx = 0, eIx = sLen - 1; // Start and end index after trimming.
|
|
|
517
|
+
|
|
|
518
|
+ // Trim illegal chars from start
|
|
|
519
|
+ while (sIx < eIx && IA[sArr[sIx] & 0xff] < 0) {
|
|
|
520
|
+ sIx++;
|
|
|
521
|
+ }
|
|
|
522
|
+
|
|
|
523
|
+ // Trim illegal chars from end
|
|
|
524
|
+ while (eIx > 0 && IA[sArr[eIx] & 0xff] < 0) {
|
|
|
525
|
+ eIx--;
|
|
|
526
|
+ }
|
|
|
527
|
+
|
|
|
528
|
+ // get the padding count (=) (0, 1 or 2)
|
|
|
529
|
+ final int pad = sArr[eIx] == '=' ? (sArr[eIx - 1] == '=' ? 2 : 1) : 0; // Count
|
|
|
530
|
+ // '='
|
|
|
531
|
+ // at
|
|
|
532
|
+ // end.
|
|
|
533
|
+ final int cCnt = eIx - sIx + 1; // Content count including possible
|
|
|
534
|
+ // separators
|
|
|
535
|
+ final int sepCnt = sLen > 76 ? (sArr[76] == '\r' ? cCnt / 78 : 0) << 1
|
|
|
536
|
+ : 0;
|
|
|
537
|
+
|
|
|
538
|
+ final int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of
|
|
|
539
|
+ // decoded bytes
|
|
|
540
|
+ final byte[] dArr = new byte[len]; // Preallocate byte[] of exact length
|
|
|
541
|
+
|
|
|
542
|
+ // Decode all but the last 0 - 2 bytes.
|
|
|
543
|
+ int d = 0;
|
|
|
544
|
+ for (int cc = 0, eLen = len / 3 * 3; d < eLen;) {
|
|
|
545
|
+ // Assemble three bytes into an int from four "valid" characters.
|
|
|
546
|
+ final int i = IA[sArr[sIx++]] << 18 | IA[sArr[sIx++]] << 12
|
|
|
547
|
+ | IA[sArr[sIx++]] << 6 | IA[sArr[sIx++]];
|
|
|
548
|
+
|
|
|
549
|
+ // Add the bytes
|
|
|
550
|
+ dArr[d++] = (byte) (i >> 16);
|
|
|
551
|
+ dArr[d++] = (byte) (i >> 8);
|
|
|
552
|
+ dArr[d++] = (byte) i;
|
|
|
553
|
+
|
|
|
554
|
+ // If line separator, jump over it.
|
|
|
555
|
+ if (sepCnt > 0 && ++cc == 19) {
|
|
|
556
|
+ sIx += 2;
|
|
|
557
|
+ cc = 0;
|
|
|
558
|
+ }
|
|
|
559
|
+ }
|
|
|
560
|
+
|
|
|
561
|
+ if (d < len) {
|
|
|
562
|
+ // Decode last 1-3 bytes (incl '=') into 1-3 bytes
|
|
|
563
|
+ int i = 0;
|
|
|
564
|
+ for (int j = 0; sIx <= eIx - pad; j++) {
|
|
|
565
|
+ i |= IA[sArr[sIx++]] << 18 - j * 6;
|
|
|
566
|
+ }
|
|
|
567
|
+
|
|
|
568
|
+ for (int r = 16; d < len; r -= 8) {
|
|
|
569
|
+ dArr[d++] = (byte) (i >> r);
|
|
|
570
|
+ }
|
|
|
571
|
+ }
|
|
|
572
|
+
|
|
|
573
|
+ return dArr;
|
|
|
574
|
+ }
|
|
|
575
|
+
|
|
|
576
|
+ // ****************************************************************************************
|
|
|
577
|
+ // * String version
|
|
|
578
|
+ // ****************************************************************************************
|
|
|
579
|
+
|
|
|
580
|
+ /**
|
|
|
581
|
+ * Encodes a raw byte array into a BASE64 <code>String</code> representation
|
|
|
582
|
+ * i accordance with RFC 2045.
|
|
|
583
|
+ *
|
|
|
584
|
+ * @param sArr
|
|
|
585
|
+ * The bytes to convert. If <code>null</code> or length 0 an
|
|
|
586
|
+ * empty array will be returned.
|
|
|
587
|
+ * @param lineSep
|
|
|
588
|
+ * Optional "\r\n" after 76 characters, unless end of file.<br>
|
|
|
589
|
+ * No line separator will be in breach of RFC 2045 which
|
|
|
590
|
+ * specifies max 76 per line but will be a
|
|
|
591
|
+ * little faster.
|
|
|
592
|
+ * @return A BASE64 encoded array. Never <code>null</code>.
|
|
|
593
|
+ */
|
|
|
594
|
+ public static final String encodeToString(final byte[] sArr,
|
|
|
595
|
+ final boolean lineSep) {
|
|
|
596
|
+ // Reuse char[] since we can't create a String incrementally anyway and
|
|
|
597
|
+ // StringBuffer/Builder would be slower.
|
|
|
598
|
+ return new String(encodeToChar(sArr, lineSep));
|
|
|
599
|
+ }
|
|
|
600
|
+
|
|
|
601
|
+ /**
|
|
|
602
|
+ * Decodes a BASE64 encoded <code>String</code>. All illegal characters will
|
|
|
603
|
+ * be ignored and can handle both strings with
|
|
|
604
|
+ * and without line separators.<br>
|
|
|
605
|
+ * <b>Note!</b> It can be up to about 2x the speed to call
|
|
|
606
|
+ * <code>decode(str.toCharArray())</code> instead. That
|
|
|
607
|
+ * will create a temporary array though. This version will use
|
|
|
608
|
+ * <code>str.charAt(i)</code> to iterate the string.
|
|
|
609
|
+ *
|
|
|
610
|
+ * @param str
|
|
|
611
|
+ * The source string. <code>null</code> or length 0 will return
|
|
|
612
|
+ * an empty array.
|
|
|
613
|
+ * @return The decoded array of bytes. May be of length 0. Will be
|
|
|
614
|
+ * <code>null</code> if the legal characters
|
|
|
615
|
+ * (including '=') isn't divideable by 4. (I.e. definitely
|
|
|
616
|
+ * corrupted).
|
|
|
617
|
+ */
|
|
|
618
|
+ public static final byte[] decode(final String str) {
|
|
|
619
|
+ // Check special case
|
|
|
620
|
+ final int sLen = str != null ? str.length() : 0;
|
|
|
621
|
+ if (sLen == 0) {
|
|
|
622
|
+ return new byte[0];
|
|
|
623
|
+ }
|
|
|
624
|
+
|
|
|
625
|
+ // Count illegal characters (including '\r', '\n') to know what size the
|
|
|
626
|
+ // returned array will be,
|
|
|
627
|
+ // so we don't have to reallocate & copy it later.
|
|
|
628
|
+ int sepCnt = 0; // Number of separator characters. (Actually illegal
|
|
|
629
|
+ // characters, but that's a bonus...)
|
|
|
630
|
+ for (int i = 0; i < sLen; i++) {
|
|
|
631
|
+ if (IA[str.charAt(i)] < 0) {
|
|
|
632
|
+ sepCnt++;
|
|
|
633
|
+ }
|
|
|
634
|
+ }
|
|
|
635
|
+
|
|
|
636
|
+ // Check so that legal chars (including '=') are evenly divideable by 4
|
|
|
637
|
+ // as specified in RFC 2045.
|
|
|
638
|
+ if ((sLen - sepCnt) % 4 != 0) {
|
|
|
639
|
+ return null;
|
|
|
640
|
+ }
|
|
|
641
|
+
|
|
|
642
|
+ // Count '=' at end
|
|
|
643
|
+ int pad = 0;
|
|
|
644
|
+ for (int i = sLen; i > 1 && IA[str.charAt(--i)] <= 0;) {
|
|
|
645
|
+ if (str.charAt(i) == '=') {
|
|
|
646
|
+ pad++;
|
|
|
647
|
+ }
|
|
|
648
|
+ }
|
|
|
649
|
+
|
|
|
650
|
+ final int len = ((sLen - sepCnt) * 6 >> 3) - pad;
|
|
|
651
|
+
|
|
|
652
|
+ final byte[] dArr = new byte[len]; // Preallocate byte[] of exact length
|
|
|
653
|
+
|
|
|
654
|
+ for (int s = 0, d = 0; d < len;) {
|
|
|
655
|
+ // Assemble three bytes into an int from four "valid" characters.
|
|
|
656
|
+ int i = 0;
|
|
|
657
|
+ for (int j = 0; j < 4; j++) { // j only increased if a valid char
|
|
|
658
|
+ // was found.
|
|
|
659
|
+ final int c = IA[str.charAt(s++)];
|
|
|
660
|
+ if (c >= 0) {
|
|
|
661
|
+ i |= c << 18 - j * 6;
|
|
|
662
|
+ } else {
|
|
|
663
|
+ j--;
|
|
|
664
|
+ }
|
|
|
665
|
+ }
|
|
|
666
|
+ // Add the bytes
|
|
|
667
|
+ dArr[d++] = (byte) (i >> 16);
|
|
|
668
|
+ if (d < len) {
|
|
|
669
|
+ dArr[d++] = (byte) (i >> 8);
|
|
|
670
|
+ if (d < len) {
|
|
|
671
|
+ dArr[d++] = (byte) i;
|
|
|
672
|
+ }
|
|
|
673
|
+ }
|
|
|
674
|
+ }
|
|
|
675
|
+ return dArr;
|
|
|
676
|
+ }
|
|
|
677
|
+
|
|
|
678
|
+ /**
|
|
|
679
|
+ * Decodes a BASE64 encoded string that is known to be resonably well
|
|
|
680
|
+ * formatted. The method is about twice as
|
|
|
681
|
+ * fast as {@link #decode(String)}. The preconditions are:<br>
|
|
|
682
|
+ * + The array must have a line length of 76 chars OR no line separators at
|
|
|
683
|
+ * all (one line).<br>
|
|
|
684
|
+ * + Line separator must be "\r\n", as specified in RFC 2045
|
|
|
685
|
+ * + The array must not contain illegal characters within the encoded string<br>
|
|
|
686
|
+ * + The array CAN have illegal characters at the beginning and end, those
|
|
|
687
|
+ * will be dealt with appropriately.<br>
|
|
|
688
|
+ *
|
|
|
689
|
+ * @param s
|
|
|
690
|
+ * The source string. Length 0 will return an empty array.
|
|
|
691
|
+ * <code>null</code> will throw an exception.
|
|
|
692
|
+ * @return The decoded array of bytes. May be of length 0.
|
|
|
693
|
+ */
|
|
|
694
|
+ public static final byte[] decodeFast(final String s) {
|
|
|
695
|
+ // Check special case
|
|
|
696
|
+ final int sLen = s.length();
|
|
|
697
|
+ if (sLen == 0) {
|
|
|
698
|
+ return new byte[0];
|
|
|
699
|
+ }
|
|
|
700
|
+
|
|
|
701
|
+ int sIx = 0, eIx = sLen - 1; // Start and end index after trimming.
|
|
|
702
|
+
|
|
|
703
|
+ // Trim illegal chars from start
|
|
|
704
|
+ while (sIx < eIx && IA[s.charAt(sIx) & 0xff] < 0) {
|
|
|
705
|
+ sIx++;
|
|
|
706
|
+ }
|
|
|
707
|
+
|
|
|
708
|
+ // Trim illegal chars from end
|
|
|
709
|
+ while (eIx > 0 && IA[s.charAt(eIx) & 0xff] < 0) {
|
|
|
710
|
+ eIx--;
|
|
|
711
|
+ }
|
|
|
712
|
+
|
|
|
713
|
+ // get the padding count (=) (0, 1 or 2)
|
|
|
714
|
+ final int pad = s.charAt(eIx) == '=' ? (s.charAt(eIx - 1) == '=' ? 2
|
|
|
715
|
+ : 1) : 0; // Count '=' at end.
|
|
|
716
|
+ final int cCnt = eIx - sIx + 1; // Content count including possible
|
|
|
717
|
+ // separators
|
|
|
718
|
+ final int sepCnt = sLen > 76 ? (s.charAt(76) == '\r' ? cCnt / 78 : 0) << 1
|
|
|
719
|
+ : 0;
|
|
|
720
|
+
|
|
|
721
|
+ final int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of
|
|
|
722
|
+ // decoded bytes
|
|
|
723
|
+ final byte[] dArr = new byte[len]; // Preallocate byte[] of exact length
|
|
|
724
|
+
|
|
|
725
|
+ // Decode all but the last 0 - 2 bytes.
|
|
|
726
|
+ int d = 0;
|
|
|
727
|
+ for (int cc = 0, eLen = len / 3 * 3; d < eLen;) {
|
|
|
728
|
+ // Assemble three bytes into an int from four "valid" characters.
|
|
|
729
|
+ final int i = IA[s.charAt(sIx++)] << 18 | IA[s.charAt(sIx++)] << 12
|
|
|
730
|
+ | IA[s.charAt(sIx++)] << 6 | IA[s.charAt(sIx++)];
|
|
|
731
|
+
|
|
|
732
|
+ // Add the bytes
|
|
|
733
|
+ dArr[d++] = (byte) (i >> 16);
|
|
|
734
|
+ dArr[d++] = (byte) (i >> 8);
|
|
|
735
|
+ dArr[d++] = (byte) i;
|
|
|
736
|
+
|
|
|
737
|
+ // If line separator, jump over it.
|
|
|
738
|
+ if (sepCnt > 0 && ++cc == 19) {
|
|
|
739
|
+ sIx += 2;
|
|
|
740
|
+ cc = 0;
|
|
|
741
|
+ }
|
|
|
742
|
+ }
|
|
|
743
|
+
|
|
|
744
|
+ if (d < len) {
|
|
|
745
|
+ // Decode last 1-3 bytes (incl '=') into 1-3 bytes
|
|
|
746
|
+ int i = 0;
|
|
|
747
|
+ for (int j = 0; sIx <= eIx - pad; j++) {
|
|
|
748
|
+ i |= IA[s.charAt(sIx++)] << 18 - j * 6;
|
|
|
749
|
+ }
|
|
|
750
|
+
|
|
|
751
|
+ for (int r = 16; d < len; r -= 8) {
|
|
|
752
|
+ dArr[d++] = (byte) (i >> r);
|
|
|
753
|
+ }
|
|
|
754
|
+ }
|
|
|
755
|
+
|
|
|
756
|
+ return dArr;
|
|
|
757
|
+ }
|
|
|
758
|
+}
|
