]> src.bluestatic.org Git - macgdbp.git/blob - Source/base64.c
Move the callTable back into DebuggerProcessor.
[macgdbp.git] / Source / base64.c
1 /* base64.c -- Encode binary data using printable characters.
2 Copyright (C) 1999, 2000, 2001, 2004, 2005, 2006 Free Software
3 Foundation, Inc.
4
5 This program is free software; you can redistribute it and/or modify
6 it under the terms of the GNU General Public License as published by
7 the Free Software Foundation; either version 2, or (at your option)
8 any later version.
9
10 This program is distributed in the hope that it will be useful,
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 GNU General Public License for more details.
14
15 You should have received a copy of the GNU General Public License
16 along with this program; if not, write to the Free Software Foundation,
17 Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
18
19 /* Written by Simon Josefsson. Partially adapted from GNU MailUtils
20 * (mailbox/filter_trans.c, as of 2004-11-28). Improved by review
21 * from Paul Eggert, Bruno Haible, and Stepan Kasal.
22 *
23 * See also RFC 3548 <http://www.ietf.org/rfc/rfc3548.txt>.
24 *
25 * Be careful with error checking. Here is how you would typically
26 * use these functions:
27 *
28 * bool ok = base64_decode_alloc (in, inlen, &out, &outlen);
29 * if (!ok)
30 * FAIL: input was not valid base64
31 * if (out == NULL)
32 * FAIL: memory allocation error
33 * OK: data in OUT/OUTLEN
34 *
35 * size_t outlen = base64_encode_alloc (in, inlen, &out);
36 * if (out == NULL && outlen == 0 && inlen != 0)
37 * FAIL: input too long
38 * if (out == NULL)
39 * FAIL: memory allocation error
40 * OK: data in OUT/OUTLEN.
41 *
42 */
43
44 /* Get prototype. */
45 #include "base64.h"
46
47 /* Get malloc. */
48 #include <stdlib.h>
49
50 /* Get UCHAR_MAX. */
51 #include <limits.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 static const char b64str[64] =
69 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
70
71 while (inlen && outlen)
72 {
73 *out++ = b64str[(to_uchar (in[0]) >> 2) & 0x3f];
74 if (!--outlen)
75 break;
76 *out++ = b64str[((to_uchar (in[0]) << 4)
77 + (--inlen ? to_uchar (in[1]) >> 4 : 0))
78 & 0x3f];
79 if (!--outlen)
80 break;
81 *out++ =
82 (inlen
83 ? b64str[((to_uchar (in[1]) << 2)
84 + (--inlen ? to_uchar (in[2]) >> 6 : 0))
85 & 0x3f]
86 : '=');
87 if (!--outlen)
88 break;
89 *out++ = inlen ? b64str[to_uchar (in[2]) & 0x3f] : '=';
90 if (!--outlen)
91 break;
92 if (inlen)
93 inlen--;
94 if (inlen)
95 in += 3;
96 }
97
98 if (outlen)
99 *out = '\0';
100 }
101
102 /* Allocate a buffer and store zero terminated base64 encoded data
103 from array IN of size INLEN, returning BASE64_LENGTH(INLEN), i.e.,
104 the length of the encoded data, excluding the terminating zero. On
105 return, the OUT variable will hold a pointer to newly allocated
106 memory that must be deallocated by the caller. If output string
107 length would overflow, 0 is returned and OUT is set to NULL. If
108 memory allocation failed, OUT is set to NULL, and the return value
109 indicates length of the requested memory block, i.e.,
110 BASE64_LENGTH(inlen) + 1. */
111 size_t
112 base64_encode_alloc (const char *in, size_t inlen, char **out)
113 {
114 size_t outlen = 1 + BASE64_LENGTH (inlen);
115
116 /* Check for overflow in outlen computation.
117 *
118 * If there is no overflow, outlen >= inlen.
119 *
120 * If the operation (inlen + 2) overflows then it yields at most +1, so
121 * outlen is 0.
122 *
123 * If the multiplication overflows, we lose at least half of the
124 * correct value, so the result is < ((inlen + 2) / 3) * 2, which is
125 * less than (inlen + 2) * 0.66667, which is less than inlen as soon as
126 * (inlen > 4).
127 */
128 if (inlen > outlen)
129 {
130 *out = NULL;
131 return 0;
132 }
133
134 *out = malloc (outlen);
135 if (!*out)
136 return outlen;
137
138 base64_encode (in, inlen, *out, outlen);
139
140 return outlen - 1;
141 }
142
143 /* With this approach this file works independent of the charset used
144 (think EBCDIC). However, it does assume that the characters in the
145 Base64 alphabet (A-Za-z0-9+/) are encoded in 0..255. POSIX
146 1003.1-2001 require that char and unsigned char are 8-bit
147 quantities, though, taking care of that problem. But this may be a
148 potential problem on non-POSIX C99 platforms.
149
150 IBM C V6 for AIX mishandles "#define B64(x) ...'x'...", so use "_"
151 as the formal parameter rather than "x". */
152 #define B64(_) \
153 ((_) == 'A' ? 0 \
154 : (_) == 'B' ? 1 \
155 : (_) == 'C' ? 2 \
156 : (_) == 'D' ? 3 \
157 : (_) == 'E' ? 4 \
158 : (_) == 'F' ? 5 \
159 : (_) == 'G' ? 6 \
160 : (_) == 'H' ? 7 \
161 : (_) == 'I' ? 8 \
162 : (_) == 'J' ? 9 \
163 : (_) == 'K' ? 10 \
164 : (_) == 'L' ? 11 \
165 : (_) == 'M' ? 12 \
166 : (_) == 'N' ? 13 \
167 : (_) == 'O' ? 14 \
168 : (_) == 'P' ? 15 \
169 : (_) == 'Q' ? 16 \
170 : (_) == 'R' ? 17 \
171 : (_) == 'S' ? 18 \
172 : (_) == 'T' ? 19 \
173 : (_) == 'U' ? 20 \
174 : (_) == 'V' ? 21 \
175 : (_) == 'W' ? 22 \
176 : (_) == 'X' ? 23 \
177 : (_) == 'Y' ? 24 \
178 : (_) == 'Z' ? 25 \
179 : (_) == 'a' ? 26 \
180 : (_) == 'b' ? 27 \
181 : (_) == 'c' ? 28 \
182 : (_) == 'd' ? 29 \
183 : (_) == 'e' ? 30 \
184 : (_) == 'f' ? 31 \
185 : (_) == 'g' ? 32 \
186 : (_) == 'h' ? 33 \
187 : (_) == 'i' ? 34 \
188 : (_) == 'j' ? 35 \
189 : (_) == 'k' ? 36 \
190 : (_) == 'l' ? 37 \
191 : (_) == 'm' ? 38 \
192 : (_) == 'n' ? 39 \
193 : (_) == 'o' ? 40 \
194 : (_) == 'p' ? 41 \
195 : (_) == 'q' ? 42 \
196 : (_) == 'r' ? 43 \
197 : (_) == 's' ? 44 \
198 : (_) == 't' ? 45 \
199 : (_) == 'u' ? 46 \
200 : (_) == 'v' ? 47 \
201 : (_) == 'w' ? 48 \
202 : (_) == 'x' ? 49 \
203 : (_) == 'y' ? 50 \
204 : (_) == 'z' ? 51 \
205 : (_) == '0' ? 52 \
206 : (_) == '1' ? 53 \
207 : (_) == '2' ? 54 \
208 : (_) == '3' ? 55 \
209 : (_) == '4' ? 56 \
210 : (_) == '5' ? 57 \
211 : (_) == '6' ? 58 \
212 : (_) == '7' ? 59 \
213 : (_) == '8' ? 60 \
214 : (_) == '9' ? 61 \
215 : (_) == '+' ? 62 \
216 : (_) == '/' ? 63 \
217 : -1)
218
219 static const signed char b64[0x100] = {
220 B64 (0), B64 (1), B64 (2), B64 (3),
221 B64 (4), B64 (5), B64 (6), B64 (7),
222 B64 (8), B64 (9), B64 (10), B64 (11),
223 B64 (12), B64 (13), B64 (14), B64 (15),
224 B64 (16), B64 (17), B64 (18), B64 (19),
225 B64 (20), B64 (21), B64 (22), B64 (23),
226 B64 (24), B64 (25), B64 (26), B64 (27),
227 B64 (28), B64 (29), B64 (30), B64 (31),
228 B64 (32), B64 (33), B64 (34), B64 (35),
229 B64 (36), B64 (37), B64 (38), B64 (39),
230 B64 (40), B64 (41), B64 (42), B64 (43),
231 B64 (44), B64 (45), B64 (46), B64 (47),
232 B64 (48), B64 (49), B64 (50), B64 (51),
233 B64 (52), B64 (53), B64 (54), B64 (55),
234 B64 (56), B64 (57), B64 (58), B64 (59),
235 B64 (60), B64 (61), B64 (62), B64 (63),
236 B64 (64), B64 (65), B64 (66), B64 (67),
237 B64 (68), B64 (69), B64 (70), B64 (71),
238 B64 (72), B64 (73), B64 (74), B64 (75),
239 B64 (76), B64 (77), B64 (78), B64 (79),
240 B64 (80), B64 (81), B64 (82), B64 (83),
241 B64 (84), B64 (85), B64 (86), B64 (87),
242 B64 (88), B64 (89), B64 (90), B64 (91),
243 B64 (92), B64 (93), B64 (94), B64 (95),
244 B64 (96), B64 (97), B64 (98), B64 (99),
245 B64 (100), B64 (101), B64 (102), B64 (103),
246 B64 (104), B64 (105), B64 (106), B64 (107),
247 B64 (108), B64 (109), B64 (110), B64 (111),
248 B64 (112), B64 (113), B64 (114), B64 (115),
249 B64 (116), B64 (117), B64 (118), B64 (119),
250 B64 (120), B64 (121), B64 (122), B64 (123),
251 B64 (124), B64 (125), B64 (126), B64 (127),
252 B64 (128), B64 (129), B64 (130), B64 (131),
253 B64 (132), B64 (133), B64 (134), B64 (135),
254 B64 (136), B64 (137), B64 (138), B64 (139),
255 B64 (140), B64 (141), B64 (142), B64 (143),
256 B64 (144), B64 (145), B64 (146), B64 (147),
257 B64 (148), B64 (149), B64 (150), B64 (151),
258 B64 (152), B64 (153), B64 (154), B64 (155),
259 B64 (156), B64 (157), B64 (158), B64 (159),
260 B64 (160), B64 (161), B64 (162), B64 (163),
261 B64 (164), B64 (165), B64 (166), B64 (167),
262 B64 (168), B64 (169), B64 (170), B64 (171),
263 B64 (172), B64 (173), B64 (174), B64 (175),
264 B64 (176), B64 (177), B64 (178), B64 (179),
265 B64 (180), B64 (181), B64 (182), B64 (183),
266 B64 (184), B64 (185), B64 (186), B64 (187),
267 B64 (188), B64 (189), B64 (190), B64 (191),
268 B64 (192), B64 (193), B64 (194), B64 (195),
269 B64 (196), B64 (197), B64 (198), B64 (199),
270 B64 (200), B64 (201), B64 (202), B64 (203),
271 B64 (204), B64 (205), B64 (206), B64 (207),
272 B64 (208), B64 (209), B64 (210), B64 (211),
273 B64 (212), B64 (213), B64 (214), B64 (215),
274 B64 (216), B64 (217), B64 (218), B64 (219),
275 B64 (220), B64 (221), B64 (222), B64 (223),
276 B64 (224), B64 (225), B64 (226), B64 (227),
277 B64 (228), B64 (229), B64 (230), B64 (231),
278 B64 (232), B64 (233), B64 (234), B64 (235),
279 B64 (236), B64 (237), B64 (238), B64 (239),
280 B64 (240), B64 (241), B64 (242), B64 (243),
281 B64 (244), B64 (245), B64 (246), B64 (247),
282 B64 (248), B64 (249), B64 (250), B64 (251),
283 B64 (252), B64 (253), B64 (254), B64 (255)
284 };
285
286 #if UCHAR_MAX == 255
287 # define uchar_in_range(c) true
288 #else
289 # define uchar_in_range(c) ((c) <= 255)
290 #endif
291
292 /* Return true if CH is a character from the Base64 alphabet, and
293 false otherwise. Note that '=' is padding and not considered to be
294 part of the alphabet. */
295 bool
296 isbase64 (char ch)
297 {
298 return uchar_in_range (to_uchar (ch)) && 0 <= b64[to_uchar (ch)];
299 }
300
301 /* Decode base64 encoded input array IN of length INLEN to output
302 array OUT that can hold *OUTLEN bytes. Return true if decoding was
303 successful, i.e. if the input was valid base64 data, false
304 otherwise. If *OUTLEN is too small, as many bytes as possible will
305 be written to OUT. On return, *OUTLEN holds the length of decoded
306 bytes in OUT. Note that as soon as any non-alphabet characters are
307 encountered, decoding is stopped and false is returned. This means
308 that, when applicable, you must remove any line terminators that is
309 part of the data stream before calling this function. */
310 bool
311 base64_decode (const char *restrict in, size_t inlen,
312 char *restrict out, size_t *outlen)
313 {
314 size_t outleft = *outlen;
315
316 while (inlen >= 2)
317 {
318 if (!isbase64 (in[0]) || !isbase64 (in[1]))
319 break;
320
321 if (outleft)
322 {
323 *out++ = ((b64[to_uchar (in[0])] << 2)
324 | (b64[to_uchar (in[1])] >> 4));
325 outleft--;
326 }
327
328 if (inlen == 2)
329 break;
330
331 if (in[2] == '=')
332 {
333 if (inlen != 4)
334 break;
335
336 if (in[3] != '=')
337 break;
338
339 }
340 else
341 {
342 if (!isbase64 (in[2]))
343 break;
344
345 if (outleft)
346 {
347 *out++ = (((b64[to_uchar (in[1])] << 4) & 0xf0)
348 | (b64[to_uchar (in[2])] >> 2));
349 outleft--;
350 }
351
352 if (inlen == 3)
353 break;
354
355 if (in[3] == '=')
356 {
357 if (inlen != 4)
358 break;
359 }
360 else
361 {
362 if (!isbase64 (in[3]))
363 break;
364
365 if (outleft)
366 {
367 *out++ = (((b64[to_uchar (in[2])] << 6) & 0xc0)
368 | b64[to_uchar (in[3])]);
369 outleft--;
370 }
371 }
372 }
373
374 in += 4;
375 inlen -= 4;
376 }
377
378 *outlen -= outleft;
379
380 if (inlen != 0)
381 return false;
382
383 return true;
384 }
385
386 /* Allocate an output buffer in *OUT, and decode the base64 encoded
387 data stored in IN of size INLEN to the *OUT buffer. On return, the
388 size of the decoded data is stored in *OUTLEN. OUTLEN may be NULL,
389 if the caller is not interested in the decoded length. *OUT may be
390 NULL to indicate an out of memory error, in which case *OUTLEN
391 contains the size of the memory block needed. The function returns
392 true on successful decoding and memory allocation errors. (Use the
393 *OUT and *OUTLEN parameters to differentiate between successful
394 decoding and memory error.) The function returns false if the
395 input was invalid, in which case *OUT is NULL and *OUTLEN is
396 undefined. */
397 bool
398 base64_decode_alloc (const char *in, size_t inlen, char **out,
399 size_t *outlen)
400 {
401 /* This may allocate a few bytes too much, depending on input,
402 but it's not worth the extra CPU time to compute the exact amount.
403 The exact amount is 3 * inlen / 4, minus 1 if the input ends
404 with "=" and minus another 1 if the input ends with "==".
405 Dividing before multiplying avoids the possibility of overflow. */
406 size_t needlen = 3 * (inlen / 4) + 2;
407
408 *out = malloc (needlen);
409 if (!*out)
410 return true;
411
412 if (!base64_decode (in, inlen, *out, &needlen))
413 {
414 free (*out);
415 *out = NULL;
416 return false;
417 }
418
419 if (outlen)
420 *outlen = needlen;
421
422 return true;
423 }