encoding.c 17 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503
  1. /*
  2. * Copyright (c) Ian F. Darwin 1986-1995.
  3. * Software written by Ian F. Darwin and others;
  4. * maintained 1995-present by Christos Zoulas and others.
  5. *
  6. * Redistribution and use in source and binary forms, with or without
  7. * modification, are permitted provided that the following conditions
  8. * are met:
  9. * 1. Redistributions of source code must retain the above copyright
  10. * notice immediately at the beginning of the file, without modification,
  11. * this list of conditions, and the following disclaimer.
  12. * 2. Redistributions in binary form must reproduce the above copyright
  13. * notice, this list of conditions and the following disclaimer in the
  14. * documentation and/or other materials provided with the distribution.
  15. *
  16. * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  17. * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  18. * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  19. * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
  20. * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  21. * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  22. * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  23. * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  24. * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  25. * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  26. * SUCH DAMAGE.
  27. */
  28. /*
  29. * Encoding -- determine the character encoding of a text file.
  30. *
  31. * Joerg Wunsch <joerg@freebsd.org> wrote the original support for 8-bit
  32. * international characters.
  33. */
  34. #include "file.h"
  35. #ifndef lint
  36. FILE_RCSID("@(#)$File: encoding.c,v 1.9 2013/11/19 20:45:50 christos Exp $")
  37. #endif /* lint */
  38. #include "magic.h"
  39. #include <string.h>
  40. #include <memory.h>
  41. #include <stdlib.h>
  42. private int looks_ascii(const unsigned char *, size_t, unichar *, size_t *);
  43. private int looks_utf8_with_BOM(const unsigned char *, size_t, unichar *,
  44. size_t *);
  45. private int looks_ucs16(const unsigned char *, size_t, unichar *, size_t *);
  46. private int looks_latin1(const unsigned char *, size_t, unichar *, size_t *);
  47. private int looks_extended(const unsigned char *, size_t, unichar *, size_t *);
  48. private void from_ebcdic(const unsigned char *, size_t, unsigned char *);
  49. #ifdef DEBUG_ENCODING
  50. #define DPRINTF(a) printf a
  51. #else
  52. #define DPRINTF(a)
  53. #endif
  54. /*
  55. * Try to determine whether text is in some character code we can
  56. * identify. Each of these tests, if it succeeds, will leave
  57. * the text converted into one-unichar-per-character Unicode in
  58. * ubuf, and the number of characters converted in ulen.
  59. */
  60. protected int
  61. file_encoding(struct magic_set *ms, const unsigned char *buf, size_t nbytes, unichar **ubuf, size_t *ulen, const char **code, const char **code_mime, const char **type)
  62. {
  63. size_t mlen;
  64. int rv = 1, ucs_type;
  65. unsigned char *nbuf = NULL;
  66. *type = "text";
  67. *ulen = 0;
  68. *code = "unknown";
  69. *code_mime = "binary";
  70. mlen = (nbytes + 1) * sizeof((*ubuf)[0]);
  71. if ((*ubuf = CAST(unichar *, calloc((size_t)1, mlen))) == NULL) {
  72. file_oomem(ms, mlen);
  73. goto done;
  74. }
  75. mlen = (nbytes + 1) * sizeof(nbuf[0]);
  76. if ((nbuf = CAST(unsigned char *, calloc((size_t)1, mlen))) == NULL) {
  77. file_oomem(ms, mlen);
  78. goto done;
  79. }
  80. if (looks_ascii(buf, nbytes, *ubuf, ulen)) {
  81. DPRINTF(("ascii %" SIZE_T_FORMAT "u\n", *ulen));
  82. *code = "ASCII";
  83. *code_mime = "us-ascii";
  84. } else if (looks_utf8_with_BOM(buf, nbytes, *ubuf, ulen) > 0) {
  85. DPRINTF(("utf8/bom %" SIZE_T_FORMAT "u\n", *ulen));
  86. *code = "UTF-8 Unicode (with BOM)";
  87. *code_mime = "utf-8";
  88. } else if (file_looks_utf8(buf, nbytes, *ubuf, ulen) > 1) {
  89. DPRINTF(("utf8 %" SIZE_T_FORMAT "u\n", *ulen));
  90. *code = "UTF-8 Unicode";
  91. *code_mime = "utf-8";
  92. } else if ((ucs_type = looks_ucs16(buf, nbytes, *ubuf, ulen)) != 0) {
  93. if (ucs_type == 1) {
  94. *code = "Little-endian UTF-16 Unicode";
  95. *code_mime = "utf-16le";
  96. } else {
  97. *code = "Big-endian UTF-16 Unicode";
  98. *code_mime = "utf-16be";
  99. }
  100. DPRINTF(("ucs16 %" SIZE_T_FORMAT "u\n", *ulen));
  101. } else if (looks_latin1(buf, nbytes, *ubuf, ulen)) {
  102. DPRINTF(("latin1 %" SIZE_T_FORMAT "u\n", *ulen));
  103. *code = "ISO-8859";
  104. *code_mime = "iso-8859-1";
  105. } else if (looks_extended(buf, nbytes, *ubuf, ulen)) {
  106. DPRINTF(("extended %" SIZE_T_FORMAT "u\n", *ulen));
  107. *code = "Non-ISO extended-ASCII";
  108. *code_mime = "unknown-8bit";
  109. } else {
  110. from_ebcdic(buf, nbytes, nbuf);
  111. if (looks_ascii(nbuf, nbytes, *ubuf, ulen)) {
  112. DPRINTF(("ebcdic %" SIZE_T_FORMAT "u\n", *ulen));
  113. *code = "EBCDIC";
  114. *code_mime = "ebcdic";
  115. } else if (looks_latin1(nbuf, nbytes, *ubuf, ulen)) {
  116. DPRINTF(("ebcdic/international %" SIZE_T_FORMAT "u\n",
  117. *ulen));
  118. *code = "International EBCDIC";
  119. *code_mime = "ebcdic";
  120. } else { /* Doesn't look like text at all */
  121. DPRINTF(("binary\n"));
  122. rv = 0;
  123. *type = "binary";
  124. }
  125. }
  126. done:
  127. free(nbuf);
  128. return rv;
  129. }
  130. /*
  131. * This table reflects a particular philosophy about what constitutes
  132. * "text," and there is room for disagreement about it.
  133. *
  134. * Version 3.31 of the file command considered a file to be ASCII if
  135. * each of its characters was approved by either the isascii() or
  136. * isalpha() function. On most systems, this would mean that any
  137. * file consisting only of characters in the range 0x00 ... 0x7F
  138. * would be called ASCII text, but many systems might reasonably
  139. * consider some characters outside this range to be alphabetic,
  140. * so the file command would call such characters ASCII. It might
  141. * have been more accurate to call this "considered textual on the
  142. * local system" than "ASCII."
  143. *
  144. * It considered a file to be "International language text" if each
  145. * of its characters was either an ASCII printing character (according
  146. * to the real ASCII standard, not the above test), a character in
  147. * the range 0x80 ... 0xFF, or one of the following control characters:
  148. * backspace, tab, line feed, vertical tab, form feed, carriage return,
  149. * escape. No attempt was made to determine the language in which files
  150. * of this type were written.
  151. *
  152. *
  153. * The table below considers a file to be ASCII if all of its characters
  154. * are either ASCII printing characters (again, according to the X3.4
  155. * standard, not isascii()) or any of the following controls: bell,
  156. * backspace, tab, line feed, form feed, carriage return, esc, nextline.
  157. *
  158. * I include bell because some programs (particularly shell scripts)
  159. * use it literally, even though it is rare in normal text. I exclude
  160. * vertical tab because it never seems to be used in real text. I also
  161. * include, with hesitation, the X3.64/ECMA-43 control nextline (0x85),
  162. * because that's what the dd EBCDIC->ASCII table maps the EBCDIC newline
  163. * character to. It might be more appropriate to include it in the 8859
  164. * set instead of the ASCII set, but it's got to be included in *something*
  165. * we recognize or EBCDIC files aren't going to be considered textual.
  166. * Some old Unix source files use SO/SI (^N/^O) to shift between Greek
  167. * and Latin characters, so these should possibly be allowed. But they
  168. * make a real mess on VT100-style displays if they're not paired properly,
  169. * so we are probably better off not calling them text.
  170. *
  171. * A file is considered to be ISO-8859 text if its characters are all
  172. * either ASCII, according to the above definition, or printing characters
  173. * from the ISO-8859 8-bit extension, characters 0xA0 ... 0xFF.
  174. *
  175. * Finally, a file is considered to be international text from some other
  176. * character code if its characters are all either ISO-8859 (according to
  177. * the above definition) or characters in the range 0x80 ... 0x9F, which
  178. * ISO-8859 considers to be control characters but the IBM PC and Macintosh
  179. * consider to be printing characters.
  180. */
  181. #define F 0 /* character never appears in text */
  182. #define T 1 /* character appears in plain ASCII text */
  183. #define I 2 /* character appears in ISO-8859 text */
  184. #define X 3 /* character appears in non-ISO extended ASCII (Mac, IBM PC) */
  185. private char text_chars[256] = {
  186. /* BEL BS HT LF FF CR */
  187. F, F, F, F, F, F, F, T, T, T, T, F, T, T, F, F, /* 0x0X */
  188. /* ESC */
  189. F, F, F, F, F, F, F, F, F, F, F, T, F, F, F, F, /* 0x1X */
  190. T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x2X */
  191. T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x3X */
  192. T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x4X */
  193. T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x5X */
  194. T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x6X */
  195. T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, F, /* 0x7X */
  196. /* NEL */
  197. X, X, X, X, X, T, X, X, X, X, X, X, X, X, X, X, /* 0x8X */
  198. X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, /* 0x9X */
  199. I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xaX */
  200. I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xbX */
  201. I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xcX */
  202. I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xdX */
  203. I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xeX */
  204. I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I /* 0xfX */
  205. };
  206. private int
  207. looks_ascii(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  208. size_t *ulen)
  209. {
  210. size_t i;
  211. *ulen = 0;
  212. for (i = 0; i < nbytes; i++) {
  213. int t = text_chars[buf[i]];
  214. if (t != T)
  215. return 0;
  216. ubuf[(*ulen)++] = buf[i];
  217. }
  218. return 1;
  219. }
  220. private int
  221. looks_latin1(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
  222. {
  223. size_t i;
  224. *ulen = 0;
  225. for (i = 0; i < nbytes; i++) {
  226. int t = text_chars[buf[i]];
  227. if (t != T && t != I)
  228. return 0;
  229. ubuf[(*ulen)++] = buf[i];
  230. }
  231. return 1;
  232. }
  233. private int
  234. looks_extended(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  235. size_t *ulen)
  236. {
  237. size_t i;
  238. *ulen = 0;
  239. for (i = 0; i < nbytes; i++) {
  240. int t = text_chars[buf[i]];
  241. if (t != T && t != I && t != X)
  242. return 0;
  243. ubuf[(*ulen)++] = buf[i];
  244. }
  245. return 1;
  246. }
  247. /*
  248. * Decide whether some text looks like UTF-8. Returns:
  249. *
  250. * -1: invalid UTF-8
  251. * 0: uses odd control characters, so doesn't look like text
  252. * 1: 7-bit text
  253. * 2: definitely UTF-8 text (valid high-bit set bytes)
  254. *
  255. * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen;
  256. * ubuf must be big enough!
  257. */
  258. protected int
  259. file_looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
  260. {
  261. size_t i;
  262. int n;
  263. unichar c;
  264. int gotone = 0, ctrl = 0;
  265. if (ubuf)
  266. *ulen = 0;
  267. for (i = 0; i < nbytes; i++) {
  268. if ((buf[i] & 0x80) == 0) { /* 0xxxxxxx is plain ASCII */
  269. /*
  270. * Even if the whole file is valid UTF-8 sequences,
  271. * still reject it if it uses weird control characters.
  272. */
  273. if (text_chars[buf[i]] != T)
  274. ctrl = 1;
  275. if (ubuf)
  276. ubuf[(*ulen)++] = buf[i];
  277. } else if ((buf[i] & 0x40) == 0) { /* 10xxxxxx never 1st byte */
  278. return -1;
  279. } else { /* 11xxxxxx begins UTF-8 */
  280. int following;
  281. if ((buf[i] & 0x20) == 0) { /* 110xxxxx */
  282. c = buf[i] & 0x1f;
  283. following = 1;
  284. } else if ((buf[i] & 0x10) == 0) { /* 1110xxxx */
  285. c = buf[i] & 0x0f;
  286. following = 2;
  287. } else if ((buf[i] & 0x08) == 0) { /* 11110xxx */
  288. c = buf[i] & 0x07;
  289. following = 3;
  290. } else if ((buf[i] & 0x04) == 0) { /* 111110xx */
  291. c = buf[i] & 0x03;
  292. following = 4;
  293. } else if ((buf[i] & 0x02) == 0) { /* 1111110x */
  294. c = buf[i] & 0x01;
  295. following = 5;
  296. } else
  297. return -1;
  298. for (n = 0; n < following; n++) {
  299. i++;
  300. if (i >= nbytes)
  301. goto done;
  302. if ((buf[i] & 0x80) == 0 || (buf[i] & 0x40))
  303. return -1;
  304. c = (c << 6) + (buf[i] & 0x3f);
  305. }
  306. if (ubuf)
  307. ubuf[(*ulen)++] = c;
  308. gotone = 1;
  309. }
  310. }
  311. done:
  312. return ctrl ? 0 : (gotone ? 2 : 1);
  313. }
  314. /*
  315. * Decide whether some text looks like UTF-8 with BOM. If there is no
  316. * BOM, return -1; otherwise return the result of looks_utf8 on the
  317. * rest of the text.
  318. */
  319. private int
  320. looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  321. size_t *ulen)
  322. {
  323. if (nbytes > 3 && buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf)
  324. return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen);
  325. else
  326. return -1;
  327. }
  328. private int
  329. looks_ucs16(const unsigned char *buf, size_t nbytes, unichar *ubuf,
  330. size_t *ulen)
  331. {
  332. int bigend;
  333. size_t i;
  334. if (nbytes < 2)
  335. return 0;
  336. if (buf[0] == 0xff && buf[1] == 0xfe)
  337. bigend = 0;
  338. else if (buf[0] == 0xfe && buf[1] == 0xff)
  339. bigend = 1;
  340. else
  341. return 0;
  342. *ulen = 0;
  343. for (i = 2; i + 1 < nbytes; i += 2) {
  344. /* XXX fix to properly handle chars > 65536 */
  345. if (bigend)
  346. ubuf[(*ulen)++] = buf[i + 1] + 256 * buf[i];
  347. else
  348. ubuf[(*ulen)++] = buf[i] + 256 * buf[i + 1];
  349. if (ubuf[*ulen - 1] == 0xfffe)
  350. return 0;
  351. if (ubuf[*ulen - 1] < 128 &&
  352. text_chars[(size_t)ubuf[*ulen - 1]] != T)
  353. return 0;
  354. }
  355. return 1 + bigend;
  356. }
  357. #undef F
  358. #undef T
  359. #undef I
  360. #undef X
  361. /*
  362. * This table maps each EBCDIC character to an (8-bit extended) ASCII
  363. * character, as specified in the rationale for the dd(1) command in
  364. * draft 11.2 (September, 1991) of the POSIX P1003.2 standard.
  365. *
  366. * Unfortunately it does not seem to correspond exactly to any of the
  367. * five variants of EBCDIC documented in IBM's _Enterprise Systems
  368. * Architecture/390: Principles of Operation_, SA22-7201-06, Seventh
  369. * Edition, July, 1999, pp. I-1 - I-4.
  370. *
  371. * Fortunately, though, all versions of EBCDIC, including this one, agree
  372. * on most of the printing characters that also appear in (7-bit) ASCII.
  373. * Of these, only '|', '!', '~', '^', '[', and ']' are in question at all.
  374. *
  375. * Fortunately too, there is general agreement that codes 0x00 through
  376. * 0x3F represent control characters, 0x41 a nonbreaking space, and the
  377. * remainder printing characters.
  378. *
  379. * This is sufficient to allow us to identify EBCDIC text and to distinguish
  380. * between old-style and internationalized examples of text.
  381. */
  382. private unsigned char ebcdic_to_ascii[] = {
  383. 0, 1, 2, 3, 156, 9, 134, 127, 151, 141, 142, 11, 12, 13, 14, 15,
  384. 16, 17, 18, 19, 157, 133, 8, 135, 24, 25, 146, 143, 28, 29, 30, 31,
  385. 128, 129, 130, 131, 132, 10, 23, 27, 136, 137, 138, 139, 140, 5, 6, 7,
  386. 144, 145, 22, 147, 148, 149, 150, 4, 152, 153, 154, 155, 20, 21, 158, 26,
  387. ' ', 160, 161, 162, 163, 164, 165, 166, 167, 168, 213, '.', '<', '(', '+', '|',
  388. '&', 169, 170, 171, 172, 173, 174, 175, 176, 177, '!', '$', '*', ')', ';', '~',
  389. '-', '/', 178, 179, 180, 181, 182, 183, 184, 185, 203, ',', '%', '_', '>', '?',
  390. 186, 187, 188, 189, 190, 191, 192, 193, 194, '`', ':', '#', '@', '\'','=', '"',
  391. 195, 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 196, 197, 198, 199, 200, 201,
  392. 202, 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', '^', 204, 205, 206, 207, 208,
  393. 209, 229, 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 210, 211, 212, '[', 214, 215,
  394. 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, ']', 230, 231,
  395. '{', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 232, 233, 234, 235, 236, 237,
  396. '}', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 238, 239, 240, 241, 242, 243,
  397. '\\',159, 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 244, 245, 246, 247, 248, 249,
  398. '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 250, 251, 252, 253, 254, 255
  399. };
  400. #ifdef notdef
  401. /*
  402. * The following EBCDIC-to-ASCII table may relate more closely to reality,
  403. * or at least to modern reality. It comes from
  404. *
  405. * http://ftp.s390.ibm.com/products/oe/bpxqp9.html
  406. *
  407. * and maps the characters of EBCDIC code page 1047 (the code used for
  408. * Unix-derived software on IBM's 390 systems) to the corresponding
  409. * characters from ISO 8859-1.
  410. *
  411. * If this table is used instead of the above one, some of the special
  412. * cases for the NEL character can be taken out of the code.
  413. */
  414. private unsigned char ebcdic_1047_to_8859[] = {
  415. 0x00,0x01,0x02,0x03,0x9C,0x09,0x86,0x7F,0x97,0x8D,0x8E,0x0B,0x0C,0x0D,0x0E,0x0F,
  416. 0x10,0x11,0x12,0x13,0x9D,0x0A,0x08,0x87,0x18,0x19,0x92,0x8F,0x1C,0x1D,0x1E,0x1F,
  417. 0x80,0x81,0x82,0x83,0x84,0x85,0x17,0x1B,0x88,0x89,0x8A,0x8B,0x8C,0x05,0x06,0x07,
  418. 0x90,0x91,0x16,0x93,0x94,0x95,0x96,0x04,0x98,0x99,0x9A,0x9B,0x14,0x15,0x9E,0x1A,
  419. 0x20,0xA0,0xE2,0xE4,0xE0,0xE1,0xE3,0xE5,0xE7,0xF1,0xA2,0x2E,0x3C,0x28,0x2B,0x7C,
  420. 0x26,0xE9,0xEA,0xEB,0xE8,0xED,0xEE,0xEF,0xEC,0xDF,0x21,0x24,0x2A,0x29,0x3B,0x5E,
  421. 0x2D,0x2F,0xC2,0xC4,0xC0,0xC1,0xC3,0xC5,0xC7,0xD1,0xA6,0x2C,0x25,0x5F,0x3E,0x3F,
  422. 0xF8,0xC9,0xCA,0xCB,0xC8,0xCD,0xCE,0xCF,0xCC,0x60,0x3A,0x23,0x40,0x27,0x3D,0x22,
  423. 0xD8,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0xAB,0xBB,0xF0,0xFD,0xFE,0xB1,
  424. 0xB0,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F,0x70,0x71,0x72,0xAA,0xBA,0xE6,0xB8,0xC6,0xA4,
  425. 0xB5,0x7E,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0xA1,0xBF,0xD0,0x5B,0xDE,0xAE,
  426. 0xAC,0xA3,0xA5,0xB7,0xA9,0xA7,0xB6,0xBC,0xBD,0xBE,0xDD,0xA8,0xAF,0x5D,0xB4,0xD7,
  427. 0x7B,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0xAD,0xF4,0xF6,0xF2,0xF3,0xF5,
  428. 0x7D,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F,0x50,0x51,0x52,0xB9,0xFB,0xFC,0xF9,0xFA,0xFF,
  429. 0x5C,0xF7,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0xB2,0xD4,0xD6,0xD2,0xD3,0xD5,
  430. 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0xB3,0xDB,0xDC,0xD9,0xDA,0x9F
  431. };
  432. #endif
  433. /*
  434. * Copy buf[0 ... nbytes-1] into out[], translating EBCDIC to ASCII.
  435. */
  436. private void
  437. from_ebcdic(const unsigned char *buf, size_t nbytes, unsigned char *out)
  438. {
  439. size_t i;
  440. for (i = 0; i < nbytes; i++) {
  441. out[i] = ebcdic_to_ascii[buf[i]];
  442. }
  443. }