1 /*
2 * Copyright (C) 2009 J.A.Bezemer@opensourcepartners.nl
3 *
4 * Permission to use, copy, modify, and distribute this software for any
5 * purpose with or without fee is hereby granted, provided that the above
6 * copyright notice and this permission notice appear in all copies.
7 *
8 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15 */
16
17 /*
18 * raw 76543210 76543210 76543210 76543210 76543210 76543210 76543210
19 * enc 65432106 54321065 43210654 32106543 21065432 10654321 06543210
20 * ^ ^ ^ ^ ^ ^ ^ ^
21 *
22 * 0001 1 0001 1
23 * 0011 3 0011 3
24 * 0111 7 0111 7
25 * 1111 f 0110 6
26 * 1110 e 0100 4
27 * 1100 c
28 * 1000 8
29 */
30
31 #include <stdio.h>
32 #include <stdlib.h>
33 #include <string.h>
34
35 #include "encoding.h"
36 #include "base128.h"
37
38 #define BLKSIZE_RAW 7
39 #define BLKSIZE_ENC 8
40
41 /* Don't use '-' (restricted to middle of labels), prefer iso_8859-1
42 * accent chars since they might readily be entered in normal use,
43 * don't use 254-255 because of possible function overloading in DNS systems.
44 */
45 static const unsigned char cb128[] =
46 "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
47 "\274\275\276\277"
48 "\300\301\302\303\304\305\306\307\310\311\312\313\314\315\316\317"
49 "\320\321\322\323\324\325\326\327\330\331\332\333\334\335\336\337"
50 "\340\341\342\343\344\345\346\347\350\351\352\353\354\355\356\357"
51 "\360\361\362\363\364\365\366\367\370\371\372\373\374\375";
52 static unsigned char rev128[256];
53 static int reverse_init = 0;
54
55 static int base128_encode(char *, size_t *, const void *, size_t);
56 static int base128_decode(void *, size_t *, const char *, size_t);
57 static int base128_handles_dots();
58 static int base128_blksize_raw();
59 static int base128_blksize_enc();
60
61 static struct encoder base128_encoder =
62 {
63 "Base128",
64 base128_encode,
65 base128_decode,
66 base128_handles_dots,
67 base128_handles_dots,
68 base128_blksize_raw,
69 base128_blksize_enc
70 };
71
72 struct encoder
get_base128_encoder()73 *get_base128_encoder()
74 {
75 return &base128_encoder;
76 }
77
78 static int
base128_handles_dots()79 base128_handles_dots()
80 {
81 return 0;
82 }
83
84 static int
base128_blksize_raw()85 base128_blksize_raw()
86 {
87 return BLKSIZE_RAW;
88 }
89
90 static int
base128_blksize_enc()91 base128_blksize_enc()
92 {
93 return BLKSIZE_ENC;
94 }
95
96 inline static void
base128_reverse_init()97 base128_reverse_init()
98 {
99 int i;
100 unsigned char c;
101
102 if (!reverse_init) {
103 memset (rev128, 0, 256);
104 for (i = 0; i < 128; i++) {
105 c = cb128[i];
106 rev128[(int) c] = i;
107 }
108 reverse_init = 1;
109 }
110 }
111
112 static int
base128_encode(char * buf,size_t * buflen,const void * data,size_t size)113 base128_encode(char *buf, size_t *buflen, const void *data, size_t size)
114 /*
115 * Fills *buf with max. *buflen characters, encoding size bytes of *data.
116 *
117 * NOTE: *buf space should be at least 1 byte _more_ than *buflen
118 * to hold the trailing '\0'.
119 *
120 * return value : #bytes filled in buf (excluding \0)
121 * sets *buflen to : #bytes encoded from data
122 */
123 {
124 unsigned char *ubuf = (unsigned char *) buf;
125 unsigned char *udata = (unsigned char *) data;
126 int iout = 0; /* to-be-filled output char */
127 int iin = 0; /* one more than last input byte that can be
128 successfully decoded */
129
130 /* Note: Don't bother to optimize manually. GCC optimizes
131 better(!) when using simplistic array indexing. */
132
133 while (1) {
134 if (iout >= *buflen || iin >= size)
135 break;
136 ubuf[iout] = cb128[((udata[iin] & 0xfe) >> 1)];
137 iout++;
138
139 if (iout >= *buflen || iin >= size) {
140 iout--; /* previous char is useless */
141 break;
142 }
143 ubuf[iout] = cb128[((udata[iin] & 0x01) << 6) |
144 ((iin + 1 < size) ?
145 ((udata[iin + 1] & 0xfc) >> 2) : 0)];
146 iin++; /* 0 complete, iin=1 */
147 iout++;
148
149 if (iout >= *buflen || iin >= size)
150 break;
151 ubuf[iout] = cb128[((udata[iin] & 0x03) << 5) |
152 ((iin + 1 < size) ?
153 ((udata[iin + 1] & 0xf8) >> 3) : 0)];
154 iin++; /* 1 complete, iin=2 */
155 iout++;
156
157 if (iout >= *buflen || iin >= size)
158 break;
159 ubuf[iout] = cb128[((udata[iin] & 0x07) << 4) |
160 ((iin + 1 < size) ?
161 ((udata[iin + 1] & 0xf0) >> 4) : 0)];
162 iin++; /* 2 complete, iin=3 */
163 iout++;
164
165 if (iout >= *buflen || iin >= size)
166 break;
167 ubuf[iout] = cb128[((udata[iin] & 0x0f) << 3) |
168 ((iin + 1 < size) ?
169 ((udata[iin + 1] & 0xe0) >> 5) : 0)];
170 iin++; /* 3 complete, iin=4 */
171 iout++;
172
173 if (iout >= *buflen || iin >= size)
174 break;
175 ubuf[iout] = cb128[((udata[iin] & 0x1f) << 2) |
176 ((iin + 1 < size) ?
177 ((udata[iin + 1] & 0xc0) >> 6) : 0)];
178 iin++; /* 4 complete, iin=5 */
179 iout++;
180
181 if (iout >= *buflen || iin >= size)
182 break;
183 ubuf[iout] = cb128[((udata[iin] & 0x3f) << 1) |
184 ((iin + 1 < size) ?
185 ((udata[iin + 1] & 0x80) >> 7) : 0)];
186 iin++; /* 5 complete, iin=6 */
187 iout++;
188
189 if (iout >= *buflen || iin >= size)
190 break;
191 ubuf[iout] = cb128[(udata[iin] & 0x7f)];
192 iin++; /* 6 complete, iin=7 */
193 iout++;
194 }
195
196 ubuf[iout] = '\0';
197
198 /* store number of bytes from data that was used */
199 *buflen = iin;
200
201 return iout;
202 }
203
204 #define REV128(x) rev128[(int) (x)]
205
206 static int
base128_decode(void * buf,size_t * buflen,const char * str,size_t slen)207 base128_decode(void *buf, size_t *buflen, const char *str, size_t slen)
208 /*
209 * Fills *buf with max. *buflen bytes, decoded from slen chars in *str.
210 * Decoding stops early when *str contains \0.
211 * Illegal encoded chars are assumed to decode to zero.
212 *
213 * NOTE: *buf space should be at least 1 byte _more_ than *buflen
214 * to hold a trailing '\0' that is added (though *buf will usually
215 * contain full-binary data).
216 *
217 * return value : #bytes filled in buf (excluding \0)
218 */
219 {
220 unsigned char *ustr = (unsigned char *) str;
221 unsigned char *ubuf = (unsigned char *) buf;
222 int iout = 0; /* to-be-filled output byte */
223 int iin = 0; /* next input char to use in decoding */
224
225 base128_reverse_init ();
226
227 /* Note: Don't bother to optimize manually. GCC optimizes
228 better(!) when using simplistic array indexing. */
229
230 while (1) {
231 if (iout >= *buflen || iin + 1 >= slen ||
232 str[iin] == '\0' || str[iin + 1] == '\0')
233 break;
234 ubuf[iout] = ((REV128(ustr[iin]) & 0x7f) << 1) |
235 ((REV128(ustr[iin + 1]) & 0x40) >> 6);
236 iin++; /* 0 used up, iin=1 */
237 iout++;
238
239 if (iout >= *buflen || iin + 1 >= slen ||
240 str[iin] == '\0' || str[iin + 1] == '\0')
241 break;
242 ubuf[iout] = ((REV128(ustr[iin]) & 0x3f) << 2) |
243 ((REV128(ustr[iin + 1]) & 0x60) >> 5);
244 iin++; /* 1 used up, iin=2 */
245 iout++;
246
247 if (iout >= *buflen || iin + 1 >= slen ||
248 str[iin] == '\0' || str[iin + 1] == '\0')
249 break;
250 ubuf[iout] = ((REV128(ustr[iin]) & 0x1f) << 3) |
251 ((REV128(ustr[iin + 1]) & 0x70) >> 4);
252 iin++; /* 2 used up, iin=3 */
253 iout++;
254
255 if (iout >= *buflen || iin + 1 >= slen ||
256 str[iin] == '\0' || str[iin + 1] == '\0')
257 break;
258 ubuf[iout] = ((REV128(ustr[iin]) & 0x0f) << 4) |
259 ((REV128(ustr[iin + 1]) & 0x78) >> 3);
260 iin++; /* 3 used up, iin=4 */
261 iout++;
262
263 if (iout >= *buflen || iin + 1 >= slen ||
264 str[iin] == '\0' || str[iin + 1] == '\0')
265 break;
266 ubuf[iout] = ((REV128(ustr[iin]) & 0x07) << 5) |
267 ((REV128(ustr[iin + 1]) & 0x7c) >> 2);
268 iin++; /* 4 used up, iin=5 */
269 iout++;
270
271 if (iout >= *buflen || iin + 1 >= slen ||
272 str[iin] == '\0' || str[iin + 1] == '\0')
273 break;
274 ubuf[iout] = ((REV128(ustr[iin]) & 0x03) << 6) |
275 ((REV128(ustr[iin + 1]) & 0x7e) >> 1);
276 iin++; /* 5 used up, iin=6 */
277 iout++;
278
279 if (iout >= *buflen || iin + 1 >= slen ||
280 str[iin] == '\0' || str[iin + 1] == '\0')
281 break;
282 ubuf[iout] = ((REV128(ustr[iin]) & 0x01) << 7) |
283 ((REV128(ustr[iin + 1]) & 0x7f));
284 iin += 2; /* 6,7 used up, iin=8 */
285 iout++;
286 }
287
288 ubuf[iout] = '\0';
289
290 return iout;
291 }
292