Classes in this File | Line Coverage | Branch Coverage | Complexity | ||||
Base64 |
|
| 3.3461538461538463;3.346 |
1 | /* | |
2 | * Licensed to the Apache Software Foundation (ASF) under one or more | |
3 | * contributor license agreements. See the NOTICE file distributed with | |
4 | * this work for additional information regarding copyright ownership. | |
5 | * The ASF licenses this file to You under the Apache License, Version 2.0 | |
6 | * (the "License"); you may not use this file except in compliance with | |
7 | * the License. You may obtain a copy of the License at | |
8 | * | |
9 | * http://www.apache.org/licenses/LICENSE-2.0 | |
10 | * | |
11 | * Unless required by applicable law or agreed to in writing, software | |
12 | * distributed under the License is distributed on an "AS IS" BASIS, | |
13 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | |
14 | * See the License for the specific language governing permissions and | |
15 | * limitations under the License. | |
16 | */ | |
17 | ||
18 | package org.apache.commons.codec.binary; | |
19 | ||
20 | import java.math.BigInteger; | |
21 | ||
22 | /** | |
23 | * Provides Base64 encoding and decoding as defined by <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>. | |
24 | * | |
25 | * <p> | |
26 | * This class implements section <cite>6.8. Base64 Content-Transfer-Encoding</cite> from RFC 2045 <cite>Multipurpose | |
27 | * Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</cite> by Freed and Borenstein. | |
28 | * </p> | |
29 | * <p> | |
30 | * The class can be parameterized in the following manner with various constructors: | |
31 | * <ul> | |
32 | * <li>URL-safe mode: Default off.</li> | |
33 | * <li>Line length: Default 76. Line length that aren't multiples of 4 will still essentially end up being multiples of | |
34 | * 4 in the encoded data. | |
35 | * <li>Line separator: Default is CRLF ("\r\n")</li> | |
36 | * </ul> | |
37 | * </p> | |
38 | * <p> | |
39 | * Since this class operates directly on byte streams, and not character streams, it is hard-coded to only | |
40 | * encode/decode character encodings which are compatible with the lower 127 ASCII chart (ISO-8859-1, Windows-1252, | |
41 | * UTF-8, etc). | |
42 | * </p> | |
43 | * <p> | |
44 | * This class is thread-safe. | |
45 | * </p> | |
46 | * | |
47 | * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a> | |
48 | * @since 1.0 | |
49 | * @version $Id$ | |
50 | */ | |
51 | public class Base64 extends BaseNCodec { | |
52 | ||
53 | /** | |
54 | * BASE32 characters are 6 bits in length. | |
55 | * They are formed by taking a block of 3 octets to form a 24-bit string, | |
56 | * which is converted into 4 BASE64 characters. | |
57 | */ | |
58 | private static final int BITS_PER_ENCODED_BYTE = 6; | |
59 | private static final int BYTES_PER_UNENCODED_BLOCK = 3; | |
60 | private static final int BYTES_PER_ENCODED_BLOCK = 4; | |
61 | ||
62 | /** | |
63 | * Chunk separator per RFC 2045 section 2.1. | |
64 | * | |
65 | * <p> | |
66 | * N.B. The next major release may break compatibility and make this field private. | |
67 | * </p> | |
68 | * | |
69 | * @see <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045 section 2.1</a> | |
70 | */ | |
71 | 1 | static final byte[] CHUNK_SEPARATOR = {'\r', '\n'}; |
72 | ||
73 | /** | |
74 | * This array is a lookup table that translates 6-bit positive integer index values into their "Base64 Alphabet" | |
75 | * equivalents as specified in Table 1 of RFC 2045. | |
76 | * | |
77 | * Thanks to "commons" project in ws.apache.org for this code. | |
78 | * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ | |
79 | */ | |
80 | 1 | private static final byte[] STANDARD_ENCODE_TABLE = { |
81 | 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', | |
82 | 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', | |
83 | 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', | |
84 | 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', | |
85 | '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/' | |
86 | }; | |
87 | ||
88 | /** | |
89 | * This is a copy of the STANDARD_ENCODE_TABLE above, but with + and / | |
90 | * changed to - and _ to make the encoded Base64 results more URL-SAFE. | |
91 | * This table is only used when the Base64's mode is set to URL-SAFE. | |
92 | */ | |
93 | 1 | private static final byte[] URL_SAFE_ENCODE_TABLE = { |
94 | 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', | |
95 | 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', | |
96 | 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', | |
97 | 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', | |
98 | '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_' | |
99 | }; | |
100 | ||
101 | /** | |
102 | * This array is a lookup table that translates Unicode characters drawn from the "Base64 Alphabet" (as specified | |
103 | * in Table 1 of RFC 2045) into their 6-bit positive integer equivalents. Characters that are not in the Base64 | |
104 | * alphabet but fall within the bounds of the array are translated to -1. | |
105 | * | |
106 | * Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This means decoder seamlessly handles both | |
107 | * URL_SAFE and STANDARD base64. (The encoder, on the other hand, needs to know ahead of time what to emit). | |
108 | * | |
109 | * Thanks to "commons" project in ws.apache.org for this code. | |
110 | * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ | |
111 | */ | |
112 | 1 | private static final byte[] DECODE_TABLE = { |
113 | -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, | |
114 | -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, | |
115 | -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, 62, -1, 63, 52, 53, 54, | |
116 | 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, -1, 0, 1, 2, 3, 4, | |
117 | 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, | |
118 | 24, 25, -1, -1, -1, -1, 63, -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, | |
119 | 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 | |
120 | }; | |
121 | ||
122 | /** | |
123 | * Base64 uses 6-bit fields. | |
124 | */ | |
125 | /** Mask used to extract 6 bits, used when encoding */ | |
126 | private static final int MASK_6BITS = 0x3f; | |
127 | ||
128 | // The static final fields above are used for the original static byte[] methods on Base64. | |
129 | // The private member fields below are used with the new streaming approach, which requires | |
130 | // some state be preserved between calls of encode() and decode(). | |
131 | ||
132 | /** | |
133 | * Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE above remains static because it is able | |
134 | * to decode both STANDARD and URL_SAFE streams, but the encodeTable must be a member variable so we can switch | |
135 | * between the two modes. | |
136 | */ | |
137 | private final byte[] encodeTable; | |
138 | ||
139 | // Only one decode table currently; keep for consistency with Base32 code | |
140 | 16841 | private final byte[] decodeTable = DECODE_TABLE; |
141 | ||
142 | /** | |
143 | * Line separator for encoding. Not used when decoding. Only used if lineLength > 0. | |
144 | */ | |
145 | private final byte[] lineSeparator; | |
146 | ||
147 | /** | |
148 | * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing. | |
149 | * <code>decodeSize = 3 + lineSeparator.length;</code> | |
150 | */ | |
151 | private final int decodeSize; | |
152 | ||
153 | /** | |
154 | * Convenience variable to help us determine when our buffer is going to run out of room and needs resizing. | |
155 | * <code>encodeSize = 4 + lineSeparator.length;</code> | |
156 | */ | |
157 | private final int encodeSize; | |
158 | ||
159 | /** | |
160 | * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. | |
161 | * <p> | |
162 | * When encoding the line length is 0 (no chunking), and the encoding table is STANDARD_ENCODE_TABLE. | |
163 | * </p> | |
164 | * | |
165 | * <p> | |
166 | * When decoding all variants are supported. | |
167 | * </p> | |
168 | */ | |
169 | public Base64() { | |
170 | 969 | this(0); |
171 | 969 | } |
172 | ||
173 | /** | |
174 | * Creates a Base64 codec used for decoding (all modes) and encoding in the given URL-safe mode. | |
175 | * <p> | |
176 | * When encoding the line length is 76, the line separator is CRLF, and the encoding table is | |
177 | * STANDARD_ENCODE_TABLE. | |
178 | * </p> | |
179 | * | |
180 | * <p> | |
181 | * When decoding all variants are supported. | |
182 | * </p> | |
183 | * | |
184 | * @param urlSafe | |
185 | * if {@code true}, URL-safe encoding is used. In most cases this should be set to {@code false}. | |
186 | * @since 1.4 | |
187 | */ | |
188 | public Base64(boolean urlSafe) { | |
189 | 7352 | this(MIME_CHUNK_SIZE, CHUNK_SEPARATOR, urlSafe); |
190 | 7352 | } |
191 | ||
192 | /** | |
193 | * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. | |
194 | * <p> | |
195 | * When encoding the line length is given in the constructor, the line separator is CRLF, and the encoding table is | |
196 | * STANDARD_ENCODE_TABLE. | |
197 | * </p> | |
198 | * <p> | |
199 | * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. | |
200 | * </p> | |
201 | * <p> | |
202 | * When decoding all variants are supported. | |
203 | * </p> | |
204 | * | |
205 | * @param lineLength | |
206 | * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of | |
207 | * 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when | |
208 | * decoding. | |
209 | * @since 1.4 | |
210 | */ | |
211 | public Base64(int lineLength) { | |
212 | 971 | this(lineLength, CHUNK_SEPARATOR); |
213 | 971 | } |
214 | ||
215 | /** | |
216 | * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. | |
217 | * <p> | |
218 | * When encoding the line length and line separator are given in the constructor, and the encoding table is | |
219 | * STANDARD_ENCODE_TABLE. | |
220 | * </p> | |
221 | * <p> | |
222 | * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. | |
223 | * </p> | |
224 | * <p> | |
225 | * When decoding all variants are supported. | |
226 | * </p> | |
227 | * | |
228 | * @param lineLength | |
229 | * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of | |
230 | * 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when | |
231 | * decoding. | |
232 | * @param lineSeparator | |
233 | * Each line of encoded data will end with this sequence of bytes. | |
234 | * @throws IllegalArgumentException | |
235 | * Thrown when the provided lineSeparator included some base64 characters. | |
236 | * @since 1.4 | |
237 | */ | |
238 | public Base64(int lineLength, byte[] lineSeparator) { | |
239 | 7894 | this(lineLength, lineSeparator, false); |
240 | 7890 | } |
241 | ||
242 | /** | |
243 | * Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode. | |
244 | * <p> | |
245 | * When encoding the line length and line separator are given in the constructor, and the encoding table is | |
246 | * STANDARD_ENCODE_TABLE. | |
247 | * </p> | |
248 | * <p> | |
249 | * Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data. | |
250 | * </p> | |
251 | * <p> | |
252 | * When decoding all variants are supported. | |
253 | * </p> | |
254 | * | |
255 | * @param lineLength | |
256 | * Each line of encoded data will be at most of the given length (rounded down to nearest multiple of | |
257 | * 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when | |
258 | * decoding. | |
259 | * @param lineSeparator | |
260 | * Each line of encoded data will end with this sequence of bytes. | |
261 | * @param urlSafe | |
262 | * Instead of emitting '+' and '/' we emit '-' and '_' respectively. urlSafe is only applied to encode | |
263 | * operations. Decoding seamlessly handles both modes. | |
264 | * @throws IllegalArgumentException | |
265 | * The provided lineSeparator included some base64 characters. That's not going to work! | |
266 | * @since 1.4 | |
267 | */ | |
268 | public Base64(int lineLength, byte[] lineSeparator, boolean urlSafe) { | |
269 | 16841 | super(BYTES_PER_UNENCODED_BLOCK, BYTES_PER_ENCODED_BLOCK, |
270 | lineLength, | |
271 | lineSeparator == null ? 0 : lineSeparator.length); | |
272 | // TODO could be simplified if there is no requirement to reject invalid line sep when length <=0 | |
273 | // @see test case Base64Test.testConstructors() | |
274 | 16841 | if (lineSeparator != null) { |
275 | 16838 | if (containsAlphabetOrPad(lineSeparator)) { |
276 | 4 | String sep = StringUtils.newStringUtf8(lineSeparator); |
277 | 4 | throw new IllegalArgumentException("lineSeparator must not contain base64 characters: [" + sep + "]"); |
278 | } | |
279 | 16834 | if (lineLength > 0){ // null line-sep forces no chunking rather than throwing IAE |
280 | 7581 | this.encodeSize = BYTES_PER_ENCODED_BLOCK + lineSeparator.length; |
281 | 7581 | this.lineSeparator = new byte[lineSeparator.length]; |
282 | 7581 | System.arraycopy(lineSeparator, 0, this.lineSeparator, 0, lineSeparator.length); |
283 | } else { | |
284 | 9253 | this.encodeSize = BYTES_PER_ENCODED_BLOCK; |
285 | 9253 | this.lineSeparator = null; |
286 | } | |
287 | } else { | |
288 | 3 | this.encodeSize = BYTES_PER_ENCODED_BLOCK; |
289 | 3 | this.lineSeparator = null; |
290 | } | |
291 | 16837 | this.decodeSize = this.encodeSize - 1; |
292 | 16837 | this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE : STANDARD_ENCODE_TABLE; |
293 | 16837 | } |
294 | ||
295 | /** | |
296 | * Returns our current encode mode. True if we're URL-SAFE, false otherwise. | |
297 | * | |
298 | * @return true if we're in URL-SAFE mode, false otherwise. | |
299 | * @since 1.4 | |
300 | */ | |
301 | public boolean isUrlSafe() { | |
302 | 2 | return this.encodeTable == URL_SAFE_ENCODE_TABLE; |
303 | } | |
304 | ||
305 | /** | |
306 | * <p> | |
307 | * Encodes all of the provided data, starting at inPos, for inAvail bytes. Must be called at least twice: once with | |
308 | * the data to encode, and once with inAvail set to "-1" to alert encoder that EOF has been reached, so flush last | |
309 | * remaining bytes (if not multiple of 3). | |
310 | * </p> | |
311 | * <p> | |
312 | * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach. | |
313 | * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ | |
314 | * </p> | |
315 | * | |
316 | * @param in | |
317 | * byte[] array of binary data to base64 encode. | |
318 | * @param inPos | |
319 | * Position to start reading data from. | |
320 | * @param inAvail | |
321 | * Amount of bytes available from input for encoding. | |
322 | * @param context | |
323 | * the context to be used | |
324 | */ | |
325 | @Override | |
326 | void encode(byte[] in, int inPos, int inAvail, Context context) { | |
327 | 103298 | if (context.eof) { |
328 | 17053 | return; |
329 | } | |
330 | // inAvail < 0 is how we're informed of EOF in the underlying data we're | |
331 | // encoding. | |
332 | 86245 | if (inAvail < 0) { |
333 | 8972 | context.eof = true; |
334 | 8972 | if (0 == context.modulus && lineLength == 0) { |
335 | 2672 | return; // no leftovers to process and not using chunking |
336 | } | |
337 | 6300 | final byte[] buffer = ensureBufferSize(encodeSize, context); |
338 | 6300 | final int savedPos = context.pos; |
339 | 6300 | switch (context.modulus) { // 0-2 |
340 | case 0 : // nothing to do here | |
341 | 186 | break; |
342 | case 1 : // 8 bits = 6 + 2 | |
343 | // top 6 bits: | |
344 | 3121 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 2) & MASK_6BITS]; |
345 | // remaining 2: | |
346 | 3121 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 4) & MASK_6BITS]; |
347 | // URL-SAFE skips the padding to further reduce size. | |
348 | 3121 | if (encodeTable == STANDARD_ENCODE_TABLE) { |
349 | 3066 | buffer[context.pos++] = PAD; |
350 | 3066 | buffer[context.pos++] = PAD; |
351 | } | |
352 | break; | |
353 | ||
354 | case 2 : // 16 bits = 6 + 6 + 4 | |
355 | 2993 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 10) & MASK_6BITS]; |
356 | 2993 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 4) & MASK_6BITS]; |
357 | 2993 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 2) & MASK_6BITS]; |
358 | // URL-SAFE skips the padding to further reduce size. | |
359 | 2993 | if (encodeTable == STANDARD_ENCODE_TABLE) { |
360 | 2942 | buffer[context.pos++] = PAD; |
361 | } | |
362 | break; | |
363 | default: | |
364 | 0 | throw new IllegalStateException("Impossible modulus "+context.modulus); |
365 | } | |
366 | 6300 | context.currentLinePos += context.pos - savedPos; // keep track of current line position |
367 | // if currentPos == 0 we are at the start of a line, so don't add CRLF | |
368 | 6300 | if (lineLength > 0 && context.currentLinePos > 0) { |
369 | 401 | System.arraycopy(lineSeparator, 0, buffer, context.pos, lineSeparator.length); |
370 | 401 | context.pos += lineSeparator.length; |
371 | } | |
372 | 6300 | } else { |
373 | 795205 | for (int i = 0; i < inAvail; i++) { |
374 | 717932 | final byte[] buffer = ensureBufferSize(encodeSize, context); |
375 | 717932 | context.modulus = (context.modulus+1) % BYTES_PER_UNENCODED_BLOCK; |
376 | 717932 | int b = in[inPos++]; |
377 | 717932 | if (b < 0) { |
378 | 354816 | b += 256; |
379 | } | |
380 | 717932 | context.ibitWorkArea = (context.ibitWorkArea << 8) + b; // BITS_PER_BYTE |
381 | 717932 | if (0 == context.modulus) { // 3 bytes = 24 bits = 4 * 6 bits to extract |
382 | 236275 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 18) & MASK_6BITS]; |
383 | 236275 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 12) & MASK_6BITS]; |
384 | 236275 | buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 6) & MASK_6BITS]; |
385 | 236275 | buffer[context.pos++] = encodeTable[context.ibitWorkArea & MASK_6BITS]; |
386 | 236275 | context.currentLinePos += BYTES_PER_ENCODED_BLOCK; |
387 | 236275 | if (lineLength > 0 && lineLength <= context.currentLinePos) { |
388 | 1136 | System.arraycopy(lineSeparator, 0, buffer, context.pos, lineSeparator.length); |
389 | 1136 | context.pos += lineSeparator.length; |
390 | 1136 | context.currentLinePos = 0; |
391 | } | |
392 | } | |
393 | } | |
394 | } | |
395 | 83573 | } |
396 | ||
397 | /** | |
398 | * <p> | |
399 | * Decodes all of the provided data, starting at inPos, for inAvail bytes. Should be called at least twice: once | |
400 | * with the data to decode, and once with inAvail set to "-1" to alert decoder that EOF has been reached. The "-1" | |
401 | * call is not necessary when decoding, but it doesn't hurt, either. | |
402 | * </p> | |
403 | * <p> | |
404 | * Ignores all non-base64 characters. This is how chunked (e.g. 76 character) data is handled, since CR and LF are | |
405 | * silently ignored, but has implications for other bytes, too. This method subscribes to the garbage-in, | |
406 | * garbage-out philosophy: it will not check the provided data for validity. | |
407 | * </p> | |
408 | * <p> | |
409 | * Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach. | |
410 | * http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/ | |
411 | * </p> | |
412 | * | |
413 | * @param in | |
414 | * byte[] array of ascii data to base64 decode. | |
415 | * @param inPos | |
416 | * Position to start reading data from. | |
417 | * @param inAvail | |
418 | * Amount of bytes available from input for encoding. | |
419 | * @param context | |
420 | * the context to be used | |
421 | */ | |
422 | @Override | |
423 | void decode(byte[] in, int inPos, int inAvail, Context context) { | |
424 | 113518 | if (context.eof) { |
425 | 20361 | return; |
426 | } | |
427 | 93157 | if (inAvail < 0) { |
428 | 2700 | context.eof = true; |
429 | } | |
430 | 1008203 | for (int i = 0; i < inAvail; i++) { |
431 | 920564 | final byte[] buffer = ensureBufferSize(decodeSize, context); |
432 | 920564 | final byte b = in[inPos++]; |
433 | 920564 | if (b == PAD) { |
434 | // We're done. | |
435 | 5518 | context.eof = true; |
436 | 5518 | break; |
437 | } else { | |
438 | 915046 | if (b >= 0 && b < DECODE_TABLE.length) { |
439 | 915032 | final int result = DECODE_TABLE[b]; |
440 | 915032 | if (result >= 0) { |
441 | 912737 | context.modulus = (context.modulus+1) % BYTES_PER_ENCODED_BLOCK; |
442 | 912737 | context.ibitWorkArea = (context.ibitWorkArea << BITS_PER_ENCODED_BYTE) + result; |
443 | 912737 | if (context.modulus == 0) { |
444 | 224664 | buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 16) & MASK_8BITS); |
445 | 224664 | buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS); |
446 | 224664 | buffer[context.pos++] = (byte) (context.ibitWorkArea & MASK_8BITS); |
447 | } | |
448 | } | |
449 | } | |
450 | } | |
451 | } | |
452 | ||
453 | // Two forms of EOF as far as base64 decoder is concerned: actual | |
454 | // EOF (-1) and first time '=' character is encountered in stream. | |
455 | // This approach makes the '=' padding characters completely optional. | |
456 | 93157 | if (context.eof && context.modulus != 0) { |
457 | 5618 | final byte[] buffer = ensureBufferSize(decodeSize, context); |
458 | ||
459 | // We have some spare bits remaining | |
460 | // Output all whole multiples of 8 bits and ignore the rest | |
461 | 5618 | switch (context.modulus) { |
462 | // case 0 : // impossible, as excluded above | |
463 | case 1 : // 6 bits - ignore entirely | |
464 | // TODO not currently tested; perhaps it is impossible? | |
465 | 0 | break; |
466 | case 2 : // 12 bits = 8 + 4 | |
467 | 2773 | context.ibitWorkArea = context.ibitWorkArea >> 4; // dump the extra 4 bits |
468 | 2773 | buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS); |
469 | 2773 | break; |
470 | case 3 : // 18 bits = 8 + 8 + 2 | |
471 | 2845 | context.ibitWorkArea = context.ibitWorkArea >> 2; // dump 2 bits |
472 | 2845 | buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS); |
473 | 2845 | buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS); |
474 | 2845 | break; |
475 | default: | |
476 | 0 | throw new IllegalStateException("Impossible modulus "+context.modulus); |
477 | } | |
478 | } | |
479 | 93157 | } |
480 | ||
481 | /** | |
482 | * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently the | |
483 | * method treats whitespace as valid. | |
484 | * | |
485 | * @param arrayOctet | |
486 | * byte array to test | |
487 | * @return {@code true} if all bytes are valid characters in the Base64 alphabet or if the byte array is empty; | |
488 | * {@code false}, otherwise | |
489 | * @deprecated 1.5 Use {@link #isBase64(byte[])}, will be removed in 2.0. | |
490 | */ | |
491 | @Deprecated | |
492 | public static boolean isArrayByteBase64(byte[] arrayOctet) { | |
493 | 0 | return isBase64(arrayOctet); |
494 | } | |
495 | ||
496 | /** | |
497 | * Returns whether or not the <code>octet</code> is in the base 64 alphabet. | |
498 | * | |
499 | * @param octet | |
500 | * The value to test | |
501 | * @return {@code true} if the value is defined in the the base 64 alphabet, {@code false} otherwise. | |
502 | * @since 1.4 | |
503 | */ | |
504 | public static boolean isBase64(byte octet) { | |
505 | 32534 | return octet == PAD_DEFAULT || (octet >= 0 && octet < DECODE_TABLE.length && DECODE_TABLE[octet] != -1); |
506 | } | |
507 | ||
508 | /** | |
509 | * Tests a given String to see if it contains only valid characters within the Base64 alphabet. Currently the | |
510 | * method treats whitespace as valid. | |
511 | * | |
512 | * @param base64 | |
513 | * String to test | |
514 | * @return {@code true} if all characters in the String are valid characters in the Base64 alphabet or if | |
515 | * the String is empty; {@code false}, otherwise | |
516 | * @since 1.5 | |
517 | */ | |
518 | public static boolean isBase64(String base64) { | |
519 | 4 | return isBase64(StringUtils.getBytesUtf8(base64)); |
520 | } | |
521 | ||
522 | /** | |
523 | * Tests a given byte array to see if it contains only valid characters within the Base64 alphabet. Currently the | |
524 | * method treats whitespace as valid. | |
525 | * | |
526 | * @param arrayOctet | |
527 | * byte array to test | |
528 | * @return {@code true} if all bytes are valid characters in the Base64 alphabet or if the byte array is empty; | |
529 | * {@code false}, otherwise | |
530 | * @since 1.5 | |
531 | */ | |
532 | public static boolean isBase64(byte[] arrayOctet) { | |
533 | 32557 | for (int i = 0; i < arrayOctet.length; i++) { |
534 | 32534 | if (!isBase64(arrayOctet[i]) && !isWhiteSpace(arrayOctet[i])) { |
535 | 10 | return false; |
536 | } | |
537 | } | |
538 | 22 | return true; |
539 | } | |
540 | ||
541 | /** | |
542 | * Encodes binary data using the base64 algorithm but does not chunk the output. | |
543 | * | |
544 | * @param binaryData | |
545 | * binary data to encode | |
546 | * @return byte[] containing Base64 characters in their UTF-8 representation. | |
547 | */ | |
548 | public static byte[] encodeBase64(byte[] binaryData) { | |
549 | 1421 | return encodeBase64(binaryData, false); |
550 | } | |
551 | ||
552 | /** | |
553 | * Encodes binary data using the base64 algorithm but does not chunk the output. | |
554 | * | |
555 | * NOTE: We changed the behaviour of this method from multi-line chunking (commons-codec-1.4) to | |
556 | * single-line non-chunking (commons-codec-1.5). | |
557 | * | |
558 | * @param binaryData | |
559 | * binary data to encode | |
560 | * @return String containing Base64 characters. | |
561 | * @since 1.4 (NOTE: 1.4 chunked the output, whereas 1.5 does not). | |
562 | */ | |
563 | public static String encodeBase64String(byte[] binaryData) { | |
564 | 25 | return StringUtils.newStringUtf8(encodeBase64(binaryData, false)); |
565 | } | |
566 | ||
567 | /** | |
568 | * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The | |
569 | * url-safe variation emits - and _ instead of + and / characters. | |
570 | * | |
571 | * @param binaryData | |
572 | * binary data to encode | |
573 | * @return byte[] containing Base64 characters in their UTF-8 representation. | |
574 | * @since 1.4 | |
575 | */ | |
576 | public static byte[] encodeBase64URLSafe(byte[] binaryData) { | |
577 | 155 | return encodeBase64(binaryData, false, true); |
578 | } | |
579 | ||
580 | /** | |
581 | * Encodes binary data using a URL-safe variation of the base64 algorithm but does not chunk the output. The | |
582 | * url-safe variation emits - and _ instead of + and / characters. | |
583 | * | |
584 | * @param binaryData | |
585 | * binary data to encode | |
586 | * @return String containing Base64 characters | |
587 | * @since 1.4 | |
588 | */ | |
589 | public static String encodeBase64URLSafeString(byte[] binaryData) { | |
590 | 1 | return StringUtils.newStringUtf8(encodeBase64(binaryData, false, true)); |
591 | } | |
592 | ||
593 | /** | |
594 | * Encodes binary data using the base64 algorithm and chunks the encoded output into 76 character blocks | |
595 | * | |
596 | * @param binaryData | |
597 | * binary data to encode | |
598 | * @return Base64 characters chunked in 76 character blocks | |
599 | */ | |
600 | public static byte[] encodeBase64Chunked(byte[] binaryData) { | |
601 | 267 | return encodeBase64(binaryData, true); |
602 | } | |
603 | ||
604 | /** | |
605 | * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. | |
606 | * | |
607 | * @param binaryData | |
608 | * Array containing binary data to encode. | |
609 | * @param isChunked | |
610 | * if {@code true} this encoder will chunk the base64 output into 76 character blocks | |
611 | * @return Base64-encoded data. | |
612 | * @throws IllegalArgumentException | |
613 | * Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE} | |
614 | */ | |
615 | public static byte[] encodeBase64(byte[] binaryData, boolean isChunked) { | |
616 | 1718 | return encodeBase64(binaryData, isChunked, false); |
617 | } | |
618 | ||
619 | /** | |
620 | * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. | |
621 | * | |
622 | * @param binaryData | |
623 | * Array containing binary data to encode. | |
624 | * @param isChunked | |
625 | * if {@code true} this encoder will chunk the base64 output into 76 character blocks | |
626 | * @param urlSafe | |
627 | * if {@code true} this encoder will emit - and _ instead of the usual + and / characters. | |
628 | * @return Base64-encoded data. | |
629 | * @throws IllegalArgumentException | |
630 | * Thrown when the input array needs an output array bigger than {@link Integer#MAX_VALUE} | |
631 | * @since 1.4 | |
632 | */ | |
633 | public static byte[] encodeBase64(byte[] binaryData, boolean isChunked, boolean urlSafe) { | |
634 | 1874 | return encodeBase64(binaryData, isChunked, urlSafe, Integer.MAX_VALUE); |
635 | } | |
636 | ||
637 | /** | |
638 | * Encodes binary data using the base64 algorithm, optionally chunking the output into 76 character blocks. | |
639 | * | |
640 | * @param binaryData | |
641 | * Array containing binary data to encode. | |
642 | * @param isChunked | |
643 | * if {@code true} this encoder will chunk the base64 output into 76 character blocks | |
644 | * @param urlSafe | |
645 | * if {@code true} this encoder will emit - and _ instead of the usual + and / characters. | |
646 | * @param maxResultSize | |
647 | * The maximum result size to accept. | |
648 | * @return Base64-encoded data. | |
649 | * @throws IllegalArgumentException | |
650 | * Thrown when the input array needs an output array bigger than maxResultSize | |
651 | * @since 1.4 | |
652 | */ | |
653 | public static byte[] encodeBase64(byte[] binaryData, boolean isChunked, boolean urlSafe, int maxResultSize) { | |
654 | 1879 | if (binaryData == null || binaryData.length == 0) { |
655 | 15 | return binaryData; |
656 | } | |
657 | ||
658 | // Create this so can use the super-class method | |
659 | // Also ensures that the same roundings are performed by the ctor and the code | |
660 | 1864 | Base64 b64 = isChunked ? new Base64(urlSafe) : new Base64(0, CHUNK_SEPARATOR, urlSafe); |
661 | 1864 | long len = b64.getEncodedLength(binaryData); |
662 | 1864 | if (len > maxResultSize) { |
663 | 4 | throw new IllegalArgumentException("Input array too big, the output array would be bigger (" + |
664 | len + | |
665 | ") than the specified maximum size of " + | |
666 | maxResultSize); | |
667 | } | |
668 | ||
669 | 1860 | return b64.encode(binaryData); |
670 | } | |
671 | ||
672 | /** | |
673 | * Decodes a Base64 String into octets | |
674 | * | |
675 | * @param base64String | |
676 | * String containing Base64 data | |
677 | * @return Array containing decoded data. | |
678 | * @since 1.4 | |
679 | */ | |
680 | public static byte[] decodeBase64(String base64String) { | |
681 | 33 | return new Base64().decode(base64String); |
682 | } | |
683 | ||
684 | /** | |
685 | * Decodes Base64 data into octets | |
686 | * | |
687 | * @param base64Data | |
688 | * Byte array containing Base64 data | |
689 | * @return Array containing decoded data. | |
690 | */ | |
691 | public static byte[] decodeBase64(byte[] base64Data) { | |
692 | 924 | return new Base64().decode(base64Data); |
693 | } | |
694 | ||
695 | // Implementation of the Encoder Interface | |
696 | ||
697 | // Implementation of integer encoding used for crypto | |
698 | /** | |
699 | * Decodes a byte64-encoded integer according to crypto standards such as W3C's XML-Signature | |
700 | * | |
701 | * @param pArray | |
702 | * a byte array containing base64 character data | |
703 | * @return A BigInteger | |
704 | * @since 1.4 | |
705 | */ | |
706 | public static BigInteger decodeInteger(byte[] pArray) { | |
707 | 4 | return new BigInteger(1, decodeBase64(pArray)); |
708 | } | |
709 | ||
710 | /** | |
711 | * Encodes to a byte64-encoded integer according to crypto standards such as W3C's XML-Signature | |
712 | * | |
713 | * @param bigInt | |
714 | * a BigInteger | |
715 | * @return A byte array containing base64 character data | |
716 | * @throws NullPointerException | |
717 | * if null is passed in | |
718 | * @since 1.4 | |
719 | */ | |
720 | public static byte[] encodeInteger(BigInteger bigInt) { | |
721 | 5 | if (bigInt == null) { |
722 | 1 | throw new NullPointerException("encodeInteger called with null parameter"); |
723 | } | |
724 | 4 | return encodeBase64(toIntegerBytes(bigInt), false); |
725 | } | |
726 | ||
727 | /** | |
728 | * Returns a byte-array representation of a <code>BigInteger</code> without sign bit. | |
729 | * | |
730 | * @param bigInt | |
731 | * <code>BigInteger</code> to be converted | |
732 | * @return a byte array representation of the BigInteger parameter | |
733 | */ | |
734 | static byte[] toIntegerBytes(BigInteger bigInt) { | |
735 | 4 | int bitlen = bigInt.bitLength(); |
736 | // round bitlen | |
737 | 4 | bitlen = ((bitlen + 7) >> 3) << 3; |
738 | 4 | byte[] bigBytes = bigInt.toByteArray(); |
739 | ||
740 | 4 | if (((bigInt.bitLength() % 8) != 0) && (((bigInt.bitLength() / 8) + 1) == (bitlen / 8))) { |
741 | 2 | return bigBytes; |
742 | } | |
743 | // set up params for copying everything but sign bit | |
744 | 2 | int startSrc = 0; |
745 | 2 | int len = bigBytes.length; |
746 | ||
747 | // if bigInt is exactly byte-aligned, just skip signbit in copy | |
748 | 2 | if ((bigInt.bitLength() % 8) == 0) { |
749 | 2 | startSrc = 1; |
750 | 2 | len--; |
751 | } | |
752 | 2 | int startDst = bitlen / 8 - len; // to pad w/ nulls as per spec |
753 | 2 | byte[] resizedBytes = new byte[bitlen / 8]; |
754 | 2 | System.arraycopy(bigBytes, startSrc, resizedBytes, startDst, len); |
755 | 2 | return resizedBytes; |
756 | } | |
757 | ||
758 | /** | |
759 | * Returns whether or not the <code>octet</code> is in the Base32 alphabet. | |
760 | * | |
761 | * @param octet | |
762 | * The value to test | |
763 | * @return {@code true} if the value is defined in the the Base32 alphabet {@code false} otherwise. | |
764 | */ | |
765 | @Override | |
766 | protected boolean isInAlphabet(byte octet) { | |
767 | 26939 | return octet >= 0 && octet < decodeTable.length && decodeTable[octet] != -1; |
768 | } | |
769 | ||
770 | } |