|
5498
|
1 /* base64.c -- Encode binary data using printable characters. |
|
|
2 Copyright (C) 1999, 2000, 2001, 2004 Free Software Foundation, Inc. |
|
|
3 |
|
|
4 This program is free software; you can redistribute it and/or modify |
|
|
5 it under the terms of the GNU General Public License as published by |
|
|
6 the Free Software Foundation; either version 2, or (at your option) |
|
|
7 any later version. |
|
|
8 |
|
|
9 This program is distributed in the hope that it will be useful, |
|
|
10 but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
|
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
|
12 GNU General Public License for more details. |
|
|
13 |
|
|
14 You should have received a copy of the GNU General Public License |
|
|
15 along with this program; if not, write to the Free Software Foundation, |
|
|
16 Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ |
|
|
17 |
|
|
18 /* Written by Simon Josefsson. Partially adapted from GNU MailUtils |
|
|
19 * (mailbox/filter_trans.c, as of 2004-11-28). Improved by review |
|
|
20 * from Paul Eggert, Bruno Haible, and Stepan Kasal. |
|
|
21 * |
|
|
22 * See also RFC 3548 <http://www.ietf.org/rfc/rfc3548.txt>. |
|
|
23 * |
|
|
24 * Be careful with error checking. Here is how you would typically |
|
|
25 * use these functions: |
|
|
26 * |
|
|
27 * bool ok = base64_decode_alloc (in, inlen, &out, &outlen); |
|
|
28 * if (!ok) |
|
|
29 * FAIL: input was not valid base64 |
|
|
30 * if (out == NULL) |
|
|
31 * FAIL: memory allocation error |
|
|
32 * OK: data in OUT/OUTLEN |
|
|
33 * |
|
|
34 * size_t outlen = base64_encode_alloc (in, inlen, &out); |
|
|
35 * if (out == NULL && outlen == 0 && inlen != 0) |
|
|
36 * FAIL: input too long |
|
|
37 * if (out == NULL) |
|
|
38 * FAIL: memory allocation error |
|
|
39 * OK: data in OUT/LEN. |
|
|
40 * |
|
|
41 */ |
|
|
42 |
|
|
43 #ifdef HAVE_CONFIG_H |
|
|
44 # include <config.h> |
|
|
45 #endif |
|
|
46 |
|
|
47 /* Get prototype. */ |
|
|
48 #include "base64.h" |
|
|
49 |
|
|
50 /* Get malloc. */ |
|
|
51 #include <stdlib.h> |
|
|
52 |
|
|
53 /* C89 compliant way to cast 'char' to 'unsigned char'. */ |
|
|
54 static inline unsigned char |
|
|
55 to_uchar (char ch) |
|
|
56 { |
|
|
57 return ch; |
|
|
58 } |
|
|
59 |
|
|
60 /* Base64 encode IN array of size INLEN into OUT array of size OUTLEN. |
|
|
61 If OUTLEN is less than BASE64_LENGTH(INLEN), write as many bytes as |
|
|
62 possible. If OUTLEN is larger than BASE64_LENGTH(INLEN), also zero |
|
|
63 terminate the output buffer. */ |
|
|
64 void |
|
|
65 base64_encode (const char *restrict in, size_t inlen, |
|
|
66 char *restrict out, size_t outlen) |
|
|
67 { |
|
|
68 const char b64[64] = |
|
|
69 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; |
|
|
70 |
|
|
71 while (inlen && outlen) |
|
|
72 { |
|
|
73 *out++ = b64[to_uchar (in[0]) >> 2]; |
|
|
74 if (!--outlen) |
|
|
75 break; |
|
|
76 *out++ = b64[((to_uchar (in[0]) << 4) |
|
|
77 + (--inlen ? to_uchar (in[1]) >> 4 : 0)) & 0x3f]; |
|
|
78 if (!--outlen) |
|
|
79 break; |
|
|
80 *out++ = |
|
|
81 (inlen |
|
|
82 ? b64[((to_uchar (in[1]) << 2) |
|
|
83 + (--inlen ? to_uchar (in[2]) >> 6 : 0)) & 0x3f] : '='); |
|
|
84 if (!--outlen) |
|
|
85 break; |
|
|
86 *out++ = inlen ? b64[to_uchar (in[2]) & 0x3f] : '='; |
|
|
87 if (!--outlen) |
|
|
88 break; |
|
|
89 if (inlen) |
|
|
90 inlen--; |
|
|
91 in += 3; |
|
|
92 } |
|
|
93 |
|
|
94 if (outlen) |
|
|
95 *out = '\0'; |
|
|
96 } |
|
|
97 |
|
|
98 /* Allocate a buffer and store zero terminated base64 encoded data |
|
|
99 from array IN of size INLEN, returning BASE64_LENGTH(INLEN), i.e., |
|
|
100 the length of the encoded data, excluding the terminating zero. On |
|
|
101 return, the OUT variable will hold a pointer to newly allocated |
|
|
102 memory that must be deallocated by the caller. If output string |
|
|
103 length would overflow, 0 is returned and OUT is set to NULL. If |
|
|
104 memory allocation fail, OUT is set to NULL, and the return value |
|
|
105 indicate length of the requested memory block, i.e., |
|
|
106 BASE64_LENGTH(inlen) + 1. */ |
|
|
107 size_t |
|
|
108 base64_encode_alloc (const char *in, size_t inlen, char **out) |
|
|
109 { |
|
|
110 size_t outlen = 1 + BASE64_LENGTH (inlen); |
|
|
111 |
|
|
112 /* Check for overflow in outlen computation. |
|
|
113 * |
|
|
114 * If there is no overflow, outlen >= inlen. |
|
|
115 * |
|
|
116 * If the operation (inlen + 2) overflows then it yields at most +1, so |
|
|
117 * outlen is 0. |
|
|
118 * |
|
|
119 * If the multiplication overflows, we lose at least half of the |
|
|
120 * correct value, so the result is < ((inlen + 2) / 3) * 2, which is |
|
|
121 * less than (inlen + 2) * 0.66667, which is less than inlen as soon as |
|
|
122 * (inlen > 4). |
|
|
123 */ |
|
|
124 if (inlen > outlen) |
|
|
125 { |
|
|
126 *out = NULL; |
|
|
127 return 0; |
|
|
128 } |
|
|
129 |
|
|
130 *out = malloc (outlen); |
|
|
131 if (*out) |
|
|
132 base64_encode (in, inlen, *out, outlen); |
|
|
133 |
|
|
134 return outlen - 1; |
|
|
135 } |
|
|
136 |
|
|
137 /* With this approach this file works independent of the charset used |
|
|
138 (think EBCDIC). However, it does assume that the characters in the |
|
|
139 Base64 alphabet (A-Za-z0-9+/) are encoded in 0..255. POSIX |
|
|
140 1003.1-2001 require that char and unsigned char are 8-bit |
|
|
141 quantities, though, taking care of that problem. But this may be a |
|
|
142 potential problem on non-POSIX C99 platforms. */ |
|
|
143 #define B64(x) \ |
|
|
144 ((x) == 'A' ? 0 \ |
|
|
145 : (x) == 'B' ? 1 \ |
|
|
146 : (x) == 'C' ? 2 \ |
|
|
147 : (x) == 'D' ? 3 \ |
|
|
148 : (x) == 'E' ? 4 \ |
|
|
149 : (x) == 'F' ? 5 \ |
|
|
150 : (x) == 'G' ? 6 \ |
|
|
151 : (x) == 'H' ? 7 \ |
|
|
152 : (x) == 'I' ? 8 \ |
|
|
153 : (x) == 'J' ? 9 \ |
|
|
154 : (x) == 'K' ? 10 \ |
|
|
155 : (x) == 'L' ? 11 \ |
|
|
156 : (x) == 'M' ? 12 \ |
|
|
157 : (x) == 'N' ? 13 \ |
|
|
158 : (x) == 'O' ? 14 \ |
|
|
159 : (x) == 'P' ? 15 \ |
|
|
160 : (x) == 'Q' ? 16 \ |
|
|
161 : (x) == 'R' ? 17 \ |
|
|
162 : (x) == 'S' ? 18 \ |
|
|
163 : (x) == 'T' ? 19 \ |
|
|
164 : (x) == 'U' ? 20 \ |
|
|
165 : (x) == 'V' ? 21 \ |
|
|
166 : (x) == 'W' ? 22 \ |
|
|
167 : (x) == 'X' ? 23 \ |
|
|
168 : (x) == 'Y' ? 24 \ |
|
|
169 : (x) == 'Z' ? 25 \ |
|
|
170 : (x) == 'a' ? 26 \ |
|
|
171 : (x) == 'b' ? 27 \ |
|
|
172 : (x) == 'c' ? 28 \ |
|
|
173 : (x) == 'd' ? 29 \ |
|
|
174 : (x) == 'e' ? 30 \ |
|
|
175 : (x) == 'f' ? 31 \ |
|
|
176 : (x) == 'g' ? 32 \ |
|
|
177 : (x) == 'h' ? 33 \ |
|
|
178 : (x) == 'i' ? 34 \ |
|
|
179 : (x) == 'j' ? 35 \ |
|
|
180 : (x) == 'k' ? 36 \ |
|
|
181 : (x) == 'l' ? 37 \ |
|
|
182 : (x) == 'm' ? 38 \ |
|
|
183 : (x) == 'n' ? 39 \ |
|
|
184 : (x) == 'o' ? 40 \ |
|
|
185 : (x) == 'p' ? 41 \ |
|
|
186 : (x) == 'q' ? 42 \ |
|
|
187 : (x) == 'r' ? 43 \ |
|
|
188 : (x) == 's' ? 44 \ |
|
|
189 : (x) == 't' ? 45 \ |
|
|
190 : (x) == 'u' ? 46 \ |
|
|
191 : (x) == 'v' ? 47 \ |
|
|
192 : (x) == 'w' ? 48 \ |
|
|
193 : (x) == 'x' ? 49 \ |
|
|
194 : (x) == 'y' ? 50 \ |
|
|
195 : (x) == 'z' ? 51 \ |
|
|
196 : (x) == '0' ? 52 \ |
|
|
197 : (x) == '1' ? 53 \ |
|
|
198 : (x) == '2' ? 54 \ |
|
|
199 : (x) == '3' ? 55 \ |
|
|
200 : (x) == '4' ? 56 \ |
|
|
201 : (x) == '5' ? 57 \ |
|
|
202 : (x) == '6' ? 58 \ |
|
|
203 : (x) == '7' ? 59 \ |
|
|
204 : (x) == '8' ? 60 \ |
|
|
205 : (x) == '9' ? 61 \ |
|
|
206 : (x) == '+' ? 62 \ |
|
|
207 : (x) == '/' ? 63 \ |
|
|
208 : -1) |
|
|
209 |
|
|
210 static const signed char b64[0x100] = { |
|
|
211 B64 (0), B64 (1), B64 (2), B64 (3), |
|
|
212 B64 (4), B64 (5), B64 (6), B64 (7), |
|
|
213 B64 (8), B64 (9), B64 (10), B64 (11), |
|
|
214 B64 (12), B64 (13), B64 (14), B64 (15), |
|
|
215 B64 (16), B64 (17), B64 (18), B64 (19), |
|
|
216 B64 (20), B64 (21), B64 (22), B64 (23), |
|
|
217 B64 (24), B64 (25), B64 (26), B64 (27), |
|
|
218 B64 (28), B64 (29), B64 (30), B64 (31), |
|
|
219 B64 (32), B64 (33), B64 (34), B64 (35), |
|
|
220 B64 (36), B64 (37), B64 (38), B64 (39), |
|
|
221 B64 (40), B64 (41), B64 (42), B64 (43), |
|
|
222 B64 (44), B64 (45), B64 (46), B64 (47), |
|
|
223 B64 (48), B64 (49), B64 (50), B64 (51), |
|
|
224 B64 (52), B64 (53), B64 (54), B64 (55), |
|
|
225 B64 (56), B64 (57), B64 (58), B64 (59), |
|
|
226 B64 (60), B64 (61), B64 (62), B64 (63), |
|
|
227 B64 (64), B64 (65), B64 (66), B64 (67), |
|
|
228 B64 (68), B64 (69), B64 (70), B64 (71), |
|
|
229 B64 (72), B64 (73), B64 (74), B64 (75), |
|
|
230 B64 (76), B64 (77), B64 (78), B64 (79), |
|
|
231 B64 (80), B64 (81), B64 (82), B64 (83), |
|
|
232 B64 (84), B64 (85), B64 (86), B64 (87), |
|
|
233 B64 (88), B64 (89), B64 (90), B64 (91), |
|
|
234 B64 (92), B64 (93), B64 (94), B64 (95), |
|
|
235 B64 (96), B64 (97), B64 (98), B64 (99), |
|
|
236 B64 (100), B64 (101), B64 (102), B64 (103), |
|
|
237 B64 (104), B64 (105), B64 (106), B64 (107), |
|
|
238 B64 (108), B64 (109), B64 (110), B64 (111), |
|
|
239 B64 (112), B64 (113), B64 (114), B64 (115), |
|
|
240 B64 (116), B64 (117), B64 (118), B64 (119), |
|
|
241 B64 (120), B64 (121), B64 (122), B64 (123), |
|
|
242 B64 (124), B64 (125), B64 (126), B64 (127), |
|
|
243 B64 (128), B64 (129), B64 (130), B64 (131), |
|
|
244 B64 (132), B64 (133), B64 (134), B64 (135), |
|
|
245 B64 (136), B64 (137), B64 (138), B64 (139), |
|
|
246 B64 (140), B64 (141), B64 (142), B64 (143), |
|
|
247 B64 (144), B64 (145), B64 (146), B64 (147), |
|
|
248 B64 (148), B64 (149), B64 (150), B64 (151), |
|
|
249 B64 (152), B64 (153), B64 (154), B64 (155), |
|
|
250 B64 (156), B64 (157), B64 (158), B64 (159), |
|
|
251 B64 (160), B64 (161), B64 (162), B64 (163), |
|
|
252 B64 (164), B64 (165), B64 (166), B64 (167), |
|
|
253 B64 (168), B64 (169), B64 (170), B64 (171), |
|
|
254 B64 (172), B64 (173), B64 (174), B64 (175), |
|
|
255 B64 (176), B64 (177), B64 (178), B64 (179), |
|
|
256 B64 (180), B64 (181), B64 (182), B64 (183), |
|
|
257 B64 (184), B64 (185), B64 (186), B64 (187), |
|
|
258 B64 (188), B64 (189), B64 (190), B64 (191), |
|
|
259 B64 (192), B64 (193), B64 (194), B64 (195), |
|
|
260 B64 (196), B64 (197), B64 (198), B64 (199), |
|
|
261 B64 (200), B64 (201), B64 (202), B64 (203), |
|
|
262 B64 (204), B64 (205), B64 (206), B64 (207), |
|
|
263 B64 (208), B64 (209), B64 (210), B64 (211), |
|
|
264 B64 (212), B64 (213), B64 (214), B64 (215), |
|
|
265 B64 (216), B64 (217), B64 (218), B64 (219), |
|
|
266 B64 (220), B64 (221), B64 (222), B64 (223), |
|
|
267 B64 (224), B64 (225), B64 (226), B64 (227), |
|
|
268 B64 (228), B64 (229), B64 (230), B64 (231), |
|
|
269 B64 (232), B64 (233), B64 (234), B64 (235), |
|
|
270 B64 (236), B64 (237), B64 (238), B64 (239), |
|
|
271 B64 (240), B64 (241), B64 (242), B64 (243), |
|
|
272 B64 (244), B64 (245), B64 (246), B64 (247), |
|
|
273 B64 (248), B64 (249), B64 (250), B64 (251), |
|
|
274 B64 (252), B64 (253), B64 (254), B64 (255) |
|
|
275 }; |
|
|
276 |
|
|
277 static inline bool |
|
|
278 isb64 (unsigned char ch) |
|
|
279 { |
|
|
280 return ch <= 255 && 0 <= b64[ch]; |
|
|
281 } |
|
|
282 |
|
|
283 /* Decode base64 encoded input array IN of length INLEN to output |
|
|
284 array OUT that can hold *OUTLEN bytes. Return true if decoding was |
|
|
285 successful, i.e. if the input was valid base64 data, false |
|
|
286 otherwise. If *OUTLEN is too small, as many bytes as possible will |
|
|
287 be written to OUT. On return, *OUTLEN holds the length of decoded |
|
|
288 bytes in OUT. Note that as soon as any non-alphabet characters are |
|
|
289 encountered, decoding is stopped and false is returned. */ |
|
|
290 bool |
|
|
291 base64_decode (const char *restrict in, size_t inlen, |
|
|
292 char *restrict out, size_t * outlen) |
|
|
293 { |
|
|
294 size_t outleft = *outlen; |
|
|
295 |
|
|
296 while (inlen >= 2) |
|
|
297 { |
|
|
298 if (!isb64 (in[0]) || !isb64 (in[1])) |
|
|
299 break; |
|
|
300 |
|
|
301 if (outleft) |
|
|
302 { |
|
|
303 *out++ = ((b64[to_uchar (in[0])] << 2) |
|
|
304 | (b64[to_uchar (in[1])] >> 4)); |
|
|
305 outleft--; |
|
|
306 } |
|
|
307 |
|
|
308 if (inlen == 2) |
|
|
309 break; |
|
|
310 |
|
|
311 if (in[2] == '=') |
|
|
312 { |
|
|
313 if (inlen != 4) |
|
|
314 break; |
|
|
315 |
|
|
316 if (in[3] != '=') |
|
|
317 break; |
|
|
318 |
|
|
319 } |
|
|
320 else |
|
|
321 { |
|
|
322 if (!isb64 (in[2])) |
|
|
323 break; |
|
|
324 |
|
|
325 if (outleft) |
|
|
326 { |
|
|
327 *out++ = (((b64[to_uchar (in[1])] << 4) & 0xf0) |
|
|
328 | (b64[to_uchar (in[2])] >> 2)); |
|
|
329 outleft--; |
|
|
330 } |
|
|
331 |
|
|
332 if (inlen == 3) |
|
|
333 break; |
|
|
334 |
|
|
335 if (in[3] == '=') |
|
|
336 { |
|
|
337 if (inlen != 4) |
|
|
338 break; |
|
|
339 } |
|
|
340 else |
|
|
341 { |
|
|
342 if (!isb64 (in[3])) |
|
|
343 break; |
|
|
344 |
|
|
345 if (outleft) |
|
|
346 { |
|
|
347 *out++ = (((b64[to_uchar (in[2])] << 6) & 0xc0) |
|
|
348 | b64[to_uchar (in[3])]); |
|
|
349 outleft--; |
|
|
350 } |
|
|
351 } |
|
|
352 } |
|
|
353 |
|
|
354 in += 4; |
|
|
355 inlen -= 4; |
|
|
356 } |
|
|
357 |
|
|
358 *outlen -= outleft; |
|
|
359 |
|
|
360 if (inlen != 0) |
|
|
361 return false; |
|
|
362 |
|
|
363 return true; |
|
|
364 } |
|
|
365 |
|
|
366 /* Allocate an output buffer in *OUT, and decode the base64 encoded |
|
|
367 data stored in IN of size INLEN to the *OUT buffer. On return, the |
|
|
368 size of the decoded data is stored in *OUTLEN. OUTLEN may be NULL, |
|
|
369 if the caller is not interested in the decoded length. *OUT may be |
|
|
370 NULL to indicate an out of memory error, in which case *OUTLEN |
|
|
371 contain the size of the memory block needed. The function return |
|
|
372 true on successful decoding and memory allocation errors. (Use the |
|
|
373 *OUT and *OUTLEN parameters to differentiate between successful |
|
|
374 decoding and memory error.) The function return false if the input |
|
|
375 was invalid, in which case *OUT is NULL and *OUTLEN is |
|
|
376 undefined. */ |
|
|
377 bool |
|
|
378 base64_decode_alloc (const char *in, size_t inlen, char **out, |
|
|
379 size_t * outlen) |
|
|
380 { |
|
|
381 /* This may allocate a few bytes too much, depending on input, |
|
|
382 but it's not worth the extra CPU time to compute the exact amount. |
|
|
383 The exact amount is 3 * inlen / 4, minus 1 if the input ends |
|
|
384 with "=" and minus another 1 if the input ends with "==". |
|
|
385 Dividing before multiplying avoids the possibility of overflow. */ |
|
|
386 size_t needlen = 3 * (inlen / 4) + 2; |
|
|
387 |
|
|
388 *out = malloc (needlen); |
|
|
389 if (!*out) |
|
|
390 return true; |
|
|
391 |
|
|
392 if (!base64_decode (in, inlen, *out, &needlen)) |
|
|
393 { |
|
|
394 free (*out); |
|
|
395 *out = NULL; |
|
|
396 return false; |
|
|
397 } |
|
|
398 |
|
|
399 if (outlen) |
|
|
400 *outlen = needlen; |
|
|
401 |
|
|
402 return true; |
|
|
403 } |
