ceba1986 |
1 | /* -*-c-*- |
2 | * |
fde30071 |
3 | * $Id: base64.c,v 1.5 2002/01/13 13:26:16 mdw Exp $ |
ceba1986 |
4 | * |
5 | * Base64 encoding and decoding. |
6 | * |
7 | * (c) 1997 Straylight/Edgeware |
8 | */ |
9 | |
10 | /*----- Licensing notice --------------------------------------------------* |
11 | * |
12 | * This file is part of the mLib utilities library. |
13 | * |
14 | * mLib is free software; you can redistribute it and/or modify |
15 | * it under the terms of the GNU Library General Public License as |
16 | * published by the Free Software Foundation; either version 2 of the |
17 | * License, or (at your option) any later version. |
18 | * |
19 | * mLib is distributed in the hope that it will be useful, |
20 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
21 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
22 | * GNU Library General Public License for more details. |
23 | * |
24 | * You should have received a copy of the GNU Library General Public |
25 | * License along with mLib; if not, write to the Free |
26 | * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, |
27 | * MA 02111-1307, USA. |
28 | */ |
29 | |
30 | /*----- Revision history --------------------------------------------------* |
31 | * |
32 | * $Log: base64.c,v $ |
fde30071 |
33 | * Revision 1.5 2002/01/13 13:26:16 mdw |
34 | * Change names for internal tables. |
35 | * |
a226592f |
36 | * Revision 1.4 1999/10/15 21:08:46 mdw |
37 | * Change support for erroneous Base64 streams with length 1 mod 4. |
38 | * |
0bfb1431 |
39 | * Revision 1.3 1999/05/21 22:14:30 mdw |
40 | * Take advantage of the new dynamic string macros. |
41 | * |
c601c4f1 |
42 | * Revision 1.2 1999/05/18 21:45:27 mdw |
43 | * Allow Base64 encode and decode of arbitrary rubbish. |
44 | * |
ceba1986 |
45 | * Revision 1.1 1999/05/17 20:35:00 mdw |
46 | * Base64 encoding and decoding support. |
47 | * |
48 | */ |
49 | |
50 | /*----- Header files ------------------------------------------------------*/ |
51 | |
52 | #include <stdio.h> |
53 | #include <stdlib.h> |
54 | #include <string.h> |
55 | |
56 | #include "base64.h" |
57 | #include "dstr.h" |
58 | |
59 | /*----- Important tables --------------------------------------------------*/ |
60 | |
fde30071 |
61 | static const char encodemap[] = { "ABCDEFGHIJKLMNOPQRSTUVWXYZ" |
62 | "abcdefghijklmnopqrstuvwxyz" |
63 | "0123456789+/" }; |
ceba1986 |
64 | |
fde30071 |
65 | static const signed char decodemap[] = { |
ceba1986 |
66 | -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 0x */ |
67 | -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, /* 1x */ |
68 | -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -1, -1, 63, /* 2x */ |
69 | 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, /* 3x */ |
70 | -1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, /* 4x */ |
71 | 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, -1, /* 5x */ |
72 | -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36 ,37, 38, 39, 40, /* 6x */ |
73 | 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1 /* 7x */ |
74 | }; |
75 | |
76 | /*----- Main code ---------------------------------------------------------*/ |
77 | |
78 | /* --- @base64_encode@ --- * |
79 | * |
80 | * Arguments: @base64_ctx *ctx@ = pointer to a context block |
c601c4f1 |
81 | * @const void *p@ = pointer to a source buffer |
ceba1986 |
82 | * @size_t sz@ = size of the source buffer |
83 | * @dstr *d@ = pointer to destination string |
84 | * |
85 | * Returns: --- |
86 | * |
87 | * Use: Encodes a binary string in base64. To flush out the final |
88 | * few characters (if necessary), pass a null source pointer. |
89 | */ |
90 | |
91 | void base64_encode(base64_ctx *ctx, |
c601c4f1 |
92 | const void *p, size_t sz, |
ceba1986 |
93 | dstr *d) |
94 | { |
c601c4f1 |
95 | if (p) { |
ceba1986 |
96 | unsigned long acc = ctx->acc; |
97 | unsigned qsz = ctx->qsz; |
c601c4f1 |
98 | const unsigned char *src = p; |
ceba1986 |
99 | |
100 | while (sz) { |
101 | acc = (acc << 8) | *src++; |
102 | qsz++; |
103 | sz--; |
104 | if (qsz == 3) { |
fde30071 |
105 | DPUTC(d, encodemap[(acc >> 18) & 0x3f]); |
106 | DPUTC(d, encodemap[(acc >> 12) & 0x3f]); |
107 | DPUTC(d, encodemap[(acc >> 6) & 0x3f]); |
108 | DPUTC(d, encodemap[(acc >> 0) & 0x3f]); |
ceba1986 |
109 | ctx->lnlen += 4; |
110 | if (ctx->maxline && ctx->lnlen >= ctx->maxline) { |
111 | dstr_puts(d, ctx->indent); |
112 | ctx->lnlen = 0; |
113 | } |
114 | qsz = 0; |
115 | acc = 0; |
116 | } |
117 | } |
118 | |
119 | ctx->acc = acc; |
120 | ctx->qsz = qsz; |
121 | } else { |
122 | unsigned long acc = ctx->acc; |
123 | unsigned qsz = ctx->qsz; |
124 | |
125 | switch (qsz) { |
126 | case 0: |
127 | break; |
128 | case 1: |
129 | acc <<= 16; |
fde30071 |
130 | DPUTC(d, encodemap[(acc >> 18) & 0x3f]); |
131 | DPUTC(d, encodemap[(acc >> 12) & 0x3f]); |
c601c4f1 |
132 | DPUTC(d, '='); |
133 | DPUTC(d, '='); |
ceba1986 |
134 | ctx->lnlen += 4; |
135 | break; |
136 | case 2: |
137 | acc <<= 8; |
fde30071 |
138 | DPUTC(d, encodemap[(acc >> 18) & 0x3f]); |
139 | DPUTC(d, encodemap[(acc >> 12) & 0x3f]); |
140 | DPUTC(d, encodemap[(acc >> 6) & 0x3f]); |
c601c4f1 |
141 | DPUTC(d, '='); |
ceba1986 |
142 | ctx->lnlen += 4; |
143 | break; |
144 | } |
145 | ctx->qsz = 0; |
146 | ctx->acc = 0; |
147 | } |
148 | } |
149 | |
150 | /* --- @base64_decode@ --- * |
151 | * |
152 | * Arguments: @base64_ctx *ctx@ = pointer to a context block |
c601c4f1 |
153 | * @const void *p@ = pointer to a source buffer |
ceba1986 |
154 | * @size_t sz@ = size of the source buffer |
155 | * @dstr *d@ = pointer to destination string |
156 | * |
157 | * Returns: --- |
158 | * |
159 | * Use: Decodes a binary string in base64. To flush out the final |
160 | * few characters (if necessary), pass a null source pointer. |
161 | */ |
162 | |
163 | void base64_decode(base64_ctx *ctx, |
c601c4f1 |
164 | const void *p, size_t sz, |
ceba1986 |
165 | dstr *d) |
166 | { |
c601c4f1 |
167 | if (p) { |
ceba1986 |
168 | unsigned long acc = ctx->acc; |
169 | unsigned qsz = ctx->qsz; |
c601c4f1 |
170 | const char *src = p; |
ceba1986 |
171 | int ch; |
172 | |
173 | while (sz) { |
174 | |
175 | /* --- Get the next character and convert it --- */ |
176 | |
177 | ch = *src++; |
178 | if (ch >= 128 || ch < 0) |
179 | ch = -1; |
180 | else |
fde30071 |
181 | ch = decodemap[ch]; |
ceba1986 |
182 | sz--; |
183 | if (ch == -1) |
184 | continue; |
185 | |
186 | /* --- Bung it in the accumulator --- */ |
187 | |
188 | acc = (acc << 6) | ch; |
189 | qsz++; |
190 | |
191 | /* --- Maybe write out a completed triplet --- */ |
192 | |
193 | if (qsz == 4) { |
c601c4f1 |
194 | DPUTC(d, (acc >> 16) & 0xff); |
195 | DPUTC(d, (acc >> 8) & 0xff); |
196 | DPUTC(d, (acc >> 0) & 0xff); |
ceba1986 |
197 | acc = 0; |
198 | qsz = 0; |
199 | } |
200 | } |
201 | |
202 | ctx->acc = acc; |
203 | ctx->qsz = qsz; |
204 | } else { |
205 | |
206 | /* --- Notes about the tail-end bits --- * |
207 | * |
208 | * Ending Base64 decoding is messy. The reference I'm using to define |
209 | * the encoding, RFC1521 section 5.2, is a little hazy on exactly what to |
210 | * do at the end. It explains that I'm meant to ignore spurious `=' |
211 | * characters, and points out that I'm not guaranteed to see anything |
212 | * interesting at the end. I'll play safe here, and ignore all `=' |
213 | * characters, relying on my client to work out when to stop feeding me |
214 | * data. I'll use the queue size to work out how many tail-end bytes |
215 | * I ought to write. |
216 | */ |
217 | |
218 | unsigned long acc = ctx->acc; |
219 | unsigned qsz = ctx->qsz; |
220 | |
a226592f |
221 | /* --- Now fiddle with everything else --- * |
222 | * |
223 | * There's a bodge here for invalid encodings which have only one hextet |
224 | * in the final group. I'm not sure this is really worth having, but it |
225 | * might save some unexpected behaviour. (Not that you won't still get |
226 | * unexpected behaviour if the stream is completely empty, of course.) |
227 | */ |
ceba1986 |
228 | |
a226592f |
229 | if (qsz) { |
230 | acc <<= 6 * (4 - qsz); |
231 | qsz *= 6; |
232 | if (qsz < 8) |
233 | qsz = 8; |
234 | while (qsz >= 8) { |
235 | DPUTC(d, (acc >> 16) & 0xff); |
236 | acc <<= 8; |
237 | qsz -= 8; |
238 | } |
ceba1986 |
239 | } |
240 | |
241 | /* --- That seems to be good enough --- */ |
242 | |
243 | ctx->qsz = 0; |
244 | ctx->acc = 0; |
245 | } |
246 | } |
247 | |
248 | /* --- @base64_init@ --- * |
249 | * |
250 | * Arguments: @base64_ctx *ctx@ = pointer to context block to initialize |
251 | * |
252 | * Returns: --- |
253 | * |
254 | * Use: Initializes a base64 context properly. |
255 | */ |
256 | |
257 | void base64_init(base64_ctx *ctx) |
258 | { |
259 | ctx->acc = 0; |
260 | ctx->qsz = 0; |
261 | ctx->lnlen = 0; |
262 | ctx->indent = "\n"; |
263 | ctx->maxline = 72; |
264 | } |
265 | |
266 | /*----- Test driver code --------------------------------------------------*/ |
267 | |
268 | #ifdef TEST_RIG |
269 | |
270 | int main(int argc, char *argv[]) |
271 | { |
272 | unsigned char buf[BUFSIZ]; |
0bfb1431 |
273 | dstr d = DSTR_INIT; |
ceba1986 |
274 | base64_ctx ctx; |
275 | void (*proc)(base64_ctx *, const unsigned char *, size_t, dstr *); |
276 | size_t sz; |
277 | |
ceba1986 |
278 | base64_init(&ctx); |
279 | |
280 | if (argc > 1 && strcmp(argv[1], "-d") == 0) |
281 | proc = base64_decode; |
282 | else { |
283 | proc = base64_encode; |
284 | putchar('\t'); |
285 | ctx.indent = "\n\t"; |
286 | ctx.maxline = 64; |
287 | } |
288 | |
289 | do { |
290 | sz = fread(buf, 1, sizeof(buf), stdin); |
291 | if (sz) { |
292 | proc(&ctx, buf, sz, &d); |
293 | dstr_write(&d, stdout); |
294 | dstr_destroy(&d); |
295 | } |
296 | } while (sz == sizeof(buf)); |
297 | |
298 | proc(&ctx, 0, 0, &d); |
299 | dstr_write(&d, stdout); |
300 | |
301 | if (proc == base64_encode) |
302 | putchar('\n'); |
303 | |
304 | return (0); |
305 | } |
306 | |
307 | #endif |
308 | |
309 | /*----- That's all, folks -------------------------------------------------*/ |