base64.c 10 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310
  1. /*
  2. * Copyright (c) 1996-1999 by Internet Software Consortium.
  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 INTERNET SOFTWARE CONSORTIUM DISCLAIMS
  9. * ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES
  10. * OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL INTERNET SOFTWARE
  11. * CONSORTIUM BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
  12. * DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
  13. * PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
  14. * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
  15. * SOFTWARE.
  16. */
  17. /*
  18. * Portions Copyright (c) 1995 by International Business Machines, Inc.
  19. *
  20. * International Business Machines, Inc. (hereinafter called IBM) grants
  21. * permission under its copyrights to use, copy, modify, and distribute this
  22. * Software with or without fee, provided that the above copyright notice and
  23. * all paragraphs of this notice appear in all copies, and that the name of IBM
  24. * not be used in connection with the marketing of any product incorporating
  25. * the Software or modifications thereof, without specific, written prior
  26. * permission.
  27. *
  28. * To the extent it has a right to do so, IBM grants an immunity from suit
  29. * under its patents, if any, for the use, sale or manufacture of products to
  30. * the extent that such products are used for performing Domain Name System
  31. * dynamic updates in TCP/IP networks by means of the Software. No immunity is
  32. * granted for any product per se or for any other function of any product.
  33. *
  34. * THE SOFTWARE IS PROVIDED "AS IS", AND IBM DISCLAIMS ALL WARRANTIES,
  35. * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
  36. * PARTICULAR PURPOSE. IN NO EVENT SHALL IBM BE LIABLE FOR ANY SPECIAL,
  37. * DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER ARISING
  38. * OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE, EVEN
  39. * IF IBM IS APPRISED OF THE POSSIBILITY OF SUCH DAMAGES.
  40. */
  41. #include <sys/types.h>
  42. #include <sys/param.h>
  43. #include <sys/socket.h>
  44. #include <netinet/in.h>
  45. #include <arpa/inet.h>
  46. #include <arpa/nameser.h>
  47. #include <ctype.h>
  48. #include <resolv.h>
  49. #include <stdio.h>
  50. #include <stdlib.h>
  51. #include <string.h>
  52. #define Assert(Cond) if (!(Cond)) abort()
  53. static const char Base64[] =
  54. "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
  55. static const char Pad64 = '=';
  56. /* (From RFC1521 and draft-ietf-dnssec-secext-03.txt)
  57. The following encoding technique is taken from RFC 1521 by Borenstein
  58. and Freed. It is reproduced here in a slightly edited form for
  59. convenience.
  60. A 65-character subset of US-ASCII is used, enabling 6 bits to be
  61. represented per printable character. (The extra 65th character, "=",
  62. is used to signify a special processing function.)
  63. The encoding process represents 24-bit groups of input bits as output
  64. strings of 4 encoded characters. Proceeding from left to right, a
  65. 24-bit input group is formed by concatenating 3 8-bit input groups.
  66. These 24 bits are then treated as 4 concatenated 6-bit groups, each
  67. of which is translated into a single digit in the base64 alphabet.
  68. Each 6-bit group is used as an index into an array of 64 printable
  69. characters. The character referenced by the index is placed in the
  70. output string.
  71. Table 1: The Base64 Alphabet
  72. Value Encoding Value Encoding Value Encoding Value Encoding
  73. 0 A 17 R 34 i 51 z
  74. 1 B 18 S 35 j 52 0
  75. 2 C 19 T 36 k 53 1
  76. 3 D 20 U 37 l 54 2
  77. 4 E 21 V 38 m 55 3
  78. 5 F 22 W 39 n 56 4
  79. 6 G 23 X 40 o 57 5
  80. 7 H 24 Y 41 p 58 6
  81. 8 I 25 Z 42 q 59 7
  82. 9 J 26 a 43 r 60 8
  83. 10 K 27 b 44 s 61 9
  84. 11 L 28 c 45 t 62 +
  85. 12 M 29 d 46 u 63 /
  86. 13 N 30 e 47 v
  87. 14 O 31 f 48 w (pad) =
  88. 15 P 32 g 49 x
  89. 16 Q 33 h 50 y
  90. Special processing is performed if fewer than 24 bits are available
  91. at the end of the data being encoded. A full encoding quantum is
  92. always completed at the end of a quantity. When fewer than 24 input
  93. bits are available in an input group, zero bits are added (on the
  94. right) to form an integral number of 6-bit groups. Padding at the
  95. end of the data is performed using the '=' character.
  96. Since all base64 input is an integral number of octets, only the
  97. -------------------------------------------------
  98. following cases can arise:
  99. (1) the final quantum of encoding input is an integral
  100. multiple of 24 bits; here, the final unit of encoded
  101. output will be an integral multiple of 4 characters
  102. with no "=" padding,
  103. (2) the final quantum of encoding input is exactly 8 bits;
  104. here, the final unit of encoded output will be two
  105. characters followed by two "=" padding characters, or
  106. (3) the final quantum of encoding input is exactly 16 bits;
  107. here, the final unit of encoded output will be three
  108. characters followed by one "=" padding character.
  109. */
  110. int
  111. b64_ntop(u_char const *src, size_t srclength, char *target, size_t targsize) {
  112. size_t datalength = 0;
  113. u_char input[3];
  114. u_char output[4];
  115. size_t i;
  116. while (2 < srclength) {
  117. input[0] = *src++;
  118. input[1] = *src++;
  119. input[2] = *src++;
  120. srclength -= 3;
  121. output[0] = input[0] >> 2;
  122. output[1] = ((input[0] & 0x03) << 4) + (input[1] >> 4);
  123. output[2] = ((input[1] & 0x0f) << 2) + (input[2] >> 6);
  124. output[3] = input[2] & 0x3f;
  125. Assert(output[0] < 64);
  126. Assert(output[1] < 64);
  127. Assert(output[2] < 64);
  128. Assert(output[3] < 64);
  129. if (datalength + 4 > targsize)
  130. return (-1);
  131. target[datalength++] = Base64[output[0]];
  132. target[datalength++] = Base64[output[1]];
  133. target[datalength++] = Base64[output[2]];
  134. target[datalength++] = Base64[output[3]];
  135. }
  136. /* Now we worry about padding. */
  137. if (0 != srclength) {
  138. /* Get what's left. */
  139. input[0] = input[1] = input[2] = '\0';
  140. for (i = 0; i < srclength; i++)
  141. input[i] = *src++;
  142. output[0] = input[0] >> 2;
  143. output[1] = ((input[0] & 0x03) << 4) + (input[1] >> 4);
  144. output[2] = ((input[1] & 0x0f) << 2) + (input[2] >> 6);
  145. Assert(output[0] < 64);
  146. Assert(output[1] < 64);
  147. Assert(output[2] < 64);
  148. if (datalength + 4 > targsize)
  149. return (-1);
  150. target[datalength++] = Base64[output[0]];
  151. target[datalength++] = Base64[output[1]];
  152. if (srclength == 1)
  153. target[datalength++] = Pad64;
  154. else
  155. target[datalength++] = Base64[output[2]];
  156. target[datalength++] = Pad64;
  157. }
  158. if (datalength >= targsize)
  159. return (-1);
  160. target[datalength] = '\0'; /* Returned value doesn't count \0. */
  161. return (datalength);
  162. }
  163. libresolv_hidden_def (b64_ntop)
  164. /* skips all whitespace anywhere.
  165. converts characters, four at a time, starting at (or after)
  166. src from base - 64 numbers into three 8 bit bytes in the target area.
  167. it returns the number of data bytes stored at the target, or -1 on error.
  168. */
  169. int
  170. b64_pton (char const *src, u_char *target, size_t targsize)
  171. {
  172. int tarindex, state, ch;
  173. char *pos;
  174. state = 0;
  175. tarindex = 0;
  176. while ((ch = *src++) != '\0') {
  177. if (isspace(ch)) /* Skip whitespace anywhere. */
  178. continue;
  179. if (ch == Pad64)
  180. break;
  181. pos = strchr(Base64, ch);
  182. if (pos == 0) /* A non-base64 character. */
  183. return (-1);
  184. switch (state) {
  185. case 0:
  186. if (target) {
  187. if ((size_t)tarindex >= targsize)
  188. return (-1);
  189. target[tarindex] = (pos - Base64) << 2;
  190. }
  191. state = 1;
  192. break;
  193. case 1:
  194. if (target) {
  195. if ((size_t)tarindex + 1 >= targsize)
  196. return (-1);
  197. target[tarindex] |= (pos - Base64) >> 4;
  198. target[tarindex+1] = ((pos - Base64) & 0x0f)
  199. << 4 ;
  200. }
  201. tarindex++;
  202. state = 2;
  203. break;
  204. case 2:
  205. if (target) {
  206. if ((size_t)tarindex + 1 >= targsize)
  207. return (-1);
  208. target[tarindex] |= (pos - Base64) >> 2;
  209. target[tarindex+1] = ((pos - Base64) & 0x03)
  210. << 6;
  211. }
  212. tarindex++;
  213. state = 3;
  214. break;
  215. case 3:
  216. if (target) {
  217. if ((size_t)tarindex >= targsize)
  218. return (-1);
  219. target[tarindex] |= (pos - Base64);
  220. }
  221. tarindex++;
  222. state = 0;
  223. break;
  224. default:
  225. abort();
  226. }
  227. }
  228. /*
  229. * We are done decoding Base-64 chars. Let's see if we ended
  230. * on a byte boundary, and/or with erroneous trailing characters.
  231. */
  232. if (ch == Pad64) { /* We got a pad char. */
  233. ch = *src++; /* Skip it, get next. */
  234. switch (state) {
  235. case 0: /* Invalid = in first position */
  236. case 1: /* Invalid = in second position */
  237. return (-1);
  238. case 2: /* Valid, means one byte of info */
  239. /* Skip any number of spaces. */
  240. for ((void)NULL; ch != '\0'; ch = *src++)
  241. if (!isspace(ch))
  242. break;
  243. /* Make sure there is another trailing = sign. */
  244. if (ch != Pad64)
  245. return (-1);
  246. ch = *src++; /* Skip the = */
  247. /* Fall through to "single trailing =" case. */
  248. /* FALLTHROUGH */
  249. case 3: /* Valid, means two bytes of info */
  250. /*
  251. * We know this char is an =. Is there anything but
  252. * whitespace after it?
  253. */
  254. for ((void)NULL; ch != '\0'; ch = *src++)
  255. if (!isspace(ch))
  256. return (-1);
  257. /*
  258. * Now make sure for cases 2 and 3 that the "extra"
  259. * bits that slopped past the last full byte were
  260. * zeros. If we don't check them, they become a
  261. * subliminal channel.
  262. */
  263. if (target && target[tarindex] != 0)
  264. return (-1);
  265. }
  266. } else {
  267. /*
  268. * We ended by seeing the end of the string. Make sure we
  269. * have no partial bytes lying around.
  270. */
  271. if (state != 0)
  272. return (-1);
  273. }
  274. return (tarindex);
  275. }