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
35 #include "file.h"
36
37 #ifndef lint
38 FILE_RCSID("@(#)$File: encoding.c,v 1.20 2019/04/15 16:48:41 christos Exp $")
39 #endif /* lint */
40
41 #include "magic.h"
42 #include <string.h>
43 #include <memory.h>
44 #include <stdlib.h>
45
46
47 private int looks_ascii(const unsigned char *, size_t, unichar *, size_t *);
48 private int looks_utf8_with_BOM(const unsigned char *, size_t, unichar *,
49 size_t *);
50 private int looks_utf7(const unsigned char *, size_t, unichar *, size_t *);
51 private int looks_ucs16(const unsigned char *, size_t, unichar *, size_t *);
52 private int looks_ucs32(const unsigned char *, size_t, unichar *, size_t *);
53 private int looks_latin1(const unsigned char *, size_t, unichar *, size_t *);
54 private int looks_extended(const unsigned char *, size_t, unichar *, size_t *);
55 private void from_ebcdic(const unsigned char *, size_t, unsigned char *);
56
57 #ifdef DEBUG_ENCODING
58 #define DPRINTF(a) printf a
59 #else
60 #define DPRINTF(a)
61 #endif
62
63 /*
64 * Try to determine whether text is in some character code we can
65 * identify. Each of these tests, if it succeeds, will leave
66 * the text converted into one-unichar-per-character Unicode in
67 * ubuf, and the number of characters converted in ulen.
68 */
69 protected int
file_encoding(struct magic_set * ms,const struct buffer * b,unichar ** ubuf,size_t * ulen,const char ** code,const char ** code_mime,const char ** type)70 file_encoding(struct magic_set *ms, const struct buffer *b, unichar **ubuf,
71 size_t *ulen, const char **code, const char **code_mime, const char **type)
72 {
73 const unsigned char *buf = CAST(const unsigned char *, b->fbuf);
74 size_t nbytes = b->flen > 64*1024 ? 64*1024 : b->flen;
75 size_t mlen;
76 int rv = 1, ucs_type;
77 unsigned char *nbuf = NULL;
78 unichar *udefbuf;
79 size_t udeflen;
80
81 if (ubuf == NULL)
82 ubuf = &udefbuf;
83 if (ulen == NULL)
84 ulen = &udeflen;
85
86 *type = "text";
87 *ulen = 0;
88 *code = "unknown";
89 *code_mime = "binary";
90
91 mlen = (nbytes + 1) * sizeof((*ubuf)[0]);
92 if ((*ubuf = CAST(unichar *, ecalloc(CAST(size_t, 1), mlen))) == NULL) {
93 file_oomem(ms, mlen);
94 goto done;
95 }
96 mlen = (nbytes + 1) * sizeof(nbuf[0]);
97 if ((nbuf = CAST(unsigned char *,
98 ecalloc(CAST(size_t, 1), mlen))) == NULL) {
99 file_oomem(ms, mlen);
100 goto done;
101 }
102
103 if (looks_ascii(buf, nbytes, *ubuf, ulen)) {
104 if (looks_utf7(buf, nbytes, *ubuf, ulen) > 0) {
105 DPRINTF(("utf-7 %" SIZE_T_FORMAT "u\n", *ulen));
106 *code = "UTF-7 Unicode";
107 *code_mime = "utf-7";
108 } else {
109 DPRINTF(("ascii %" SIZE_T_FORMAT "u\n", *ulen));
110 *code = "ASCII";
111 *code_mime = "us-ascii";
112 }
113 } else if (looks_utf8_with_BOM(buf, nbytes, *ubuf, ulen) > 0) {
114 DPRINTF(("utf8/bom %" SIZE_T_FORMAT "u\n", *ulen));
115 *code = "UTF-8 Unicode (with BOM)";
116 *code_mime = "utf-8";
117 } else if (file_looks_utf8(buf, nbytes, *ubuf, ulen) > 1) {
118 DPRINTF(("utf8 %" SIZE_T_FORMAT "u\n", *ulen));
119 *code = "UTF-8 Unicode";
120 *code_mime = "utf-8";
121 } else if ((ucs_type = looks_ucs32(buf, nbytes, *ubuf, ulen)) != 0) {
122 if (ucs_type == 1) {
123 *code = "Little-endian UTF-32 Unicode";
124 *code_mime = "utf-32le";
125 } else {
126 *code = "Big-endian UTF-32 Unicode";
127 *code_mime = "utf-32be";
128 }
129 DPRINTF(("ucs32 %" SIZE_T_FORMAT "u\n", *ulen));
130 } else if ((ucs_type = looks_ucs16(buf, nbytes, *ubuf, ulen)) != 0) {
131 if (ucs_type == 1) {
132 *code = "Little-endian UTF-16 Unicode";
133 *code_mime = "utf-16le";
134 } else {
135 *code = "Big-endian UTF-16 Unicode";
136 *code_mime = "utf-16be";
137 }
138 DPRINTF(("ucs16 %" SIZE_T_FORMAT "u\n", *ulen));
139 } else if (looks_latin1(buf, nbytes, *ubuf, ulen)) {
140 DPRINTF(("latin1 %" SIZE_T_FORMAT "u\n", *ulen));
141 *code = "ISO-8859";
142 *code_mime = "iso-8859-1";
143 } else if (looks_extended(buf, nbytes, *ubuf, ulen)) {
144 DPRINTF(("extended %" SIZE_T_FORMAT "u\n", *ulen));
145 *code = "Non-ISO extended-ASCII";
146 *code_mime = "unknown-8bit";
147 } else {
148 from_ebcdic(buf, nbytes, nbuf);
149
150 if (looks_ascii(nbuf, nbytes, *ubuf, ulen)) {
151 DPRINTF(("ebcdic %" SIZE_T_FORMAT "u\n", *ulen));
152 *code = "EBCDIC";
153 *code_mime = "ebcdic";
154 } else if (looks_latin1(nbuf, nbytes, *ubuf, ulen)) {
155 DPRINTF(("ebcdic/international %" SIZE_T_FORMAT "u\n",
156 *ulen));
157 *code = "International EBCDIC";
158 *code_mime = "ebcdic";
159 } else { /* Doesn't look like text at all */
160 DPRINTF(("binary\n"));
161 rv = 0;
162 *type = "binary";
163 }
164 }
165
166 done:
167 efree(nbuf);
168 if (ubuf == &udefbuf)
169 efree(udefbuf);
170
171 return rv;
172 }
173
174 /*
175 * This table reflects a particular philosophy about what constitutes
176 * "text," and there is room for disagreement about it.
177 *
178 * Version 3.31 of the file command considered a file to be ASCII if
179 * each of its characters was approved by either the isascii() or
180 * isalpha() function. On most systems, this would mean that any
181 * file consisting only of characters in the range 0x00 ... 0x7F
182 * would be called ASCII text, but many systems might reasonably
183 * consider some characters outside this range to be alphabetic,
184 * so the file command would call such characters ASCII. It might
185 * have been more accurate to call this "considered textual on the
186 * local system" than "ASCII."
187 *
188 * It considered a file to be "International language text" if each
189 * of its characters was either an ASCII printing character (according
190 * to the real ASCII standard, not the above test), a character in
191 * the range 0x80 ... 0xFF, or one of the following control characters:
192 * backspace, tab, line feed, vertical tab, form feed, carriage return,
193 * escape. No attempt was made to determine the language in which files
194 * of this type were written.
195 *
196 *
197 * The table below considers a file to be ASCII if all of its characters
198 * are either ASCII printing characters (again, according to the X3.4
199 * standard, not isascii()) or any of the following controls: bell,
200 * backspace, tab, line feed, form feed, carriage return, esc, nextline.
201 *
202 * I include bell because some programs (particularly shell scripts)
203 * use it literally, even though it is rare in normal text. I exclude
204 * vertical tab because it never seems to be used in real text. I also
205 * include, with hesitation, the X3.64/ECMA-43 control nextline (0x85),
206 * because that's what the dd EBCDIC->ASCII table maps the EBCDIC newline
207 * character to. It might be more appropriate to include it in the 8859
208 * set instead of the ASCII set, but it's got to be included in *something*
209 * we recognize or EBCDIC files aren't going to be considered textual.
210 * Some old Unix source files use SO/SI (^N/^O) to shift between Greek
211 * and Latin characters, so these should possibly be allowed. But they
212 * make a real mess on VT100-style displays if they're not paired properly,
213 * so we are probably better off not calling them text.
214 *
215 * A file is considered to be ISO-8859 text if its characters are all
216 * either ASCII, according to the above definition, or printing characters
217 * from the ISO-8859 8-bit extension, characters 0xA0 ... 0xFF.
218 *
219 * Finally, a file is considered to be international text from some other
220 * character code if its characters are all either ISO-8859 (according to
221 * the above definition) or characters in the range 0x80 ... 0x9F, which
222 * ISO-8859 considers to be control characters but the IBM PC and Macintosh
223 * consider to be printing characters.
224 */
225
226 #define F 0 /* character never appears in text */
227 #define T 1 /* character appears in plain ASCII text */
228 #define I 2 /* character appears in ISO-8859 text */
229 #define X 3 /* character appears in non-ISO extended ASCII (Mac, IBM PC) */
230
231 private char text_chars[256] = {
232 /* BEL BS HT LF VT FF CR */
233 F, F, F, F, F, F, F, T, T, T, T, T, T, T, F, F, /* 0x0X */
234 /* ESC */
235 F, F, F, F, F, F, F, F, F, F, F, T, F, F, F, F, /* 0x1X */
236 T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x2X */
237 T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x3X */
238 T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x4X */
239 T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x5X */
240 T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x6X */
241 T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, F, /* 0x7X */
242 /* NEL */
243 X, X, X, X, X, T, X, X, X, X, X, X, X, X, X, X, /* 0x8X */
244 X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, /* 0x9X */
245 I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xaX */
246 I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xbX */
247 I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xcX */
248 I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xdX */
249 I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xeX */
250 I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I /* 0xfX */
251 };
252
253 private int
looks_ascii(const unsigned char * buf,size_t nbytes,unichar * ubuf,size_t * ulen)254 looks_ascii(const unsigned char *buf, size_t nbytes, unichar *ubuf,
255 size_t *ulen)
256 {
257 size_t i;
258
259 *ulen = 0;
260
261 for (i = 0; i < nbytes; i++) {
262 int t = text_chars[buf[i]];
263
264 if (t != T)
265 return 0;
266
267 ubuf[(*ulen)++] = buf[i];
268 }
269
270 return 1;
271 }
272
273 private int
looks_latin1(const unsigned char * buf,size_t nbytes,unichar * ubuf,size_t * ulen)274 looks_latin1(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
275 {
276 size_t i;
277
278 *ulen = 0;
279
280 for (i = 0; i < nbytes; i++) {
281 int t = text_chars[buf[i]];
282
283 if (t != T && t != I)
284 return 0;
285
286 ubuf[(*ulen)++] = buf[i];
287 }
288
289 return 1;
290 }
291
292 private int
looks_extended(const unsigned char * buf,size_t nbytes,unichar * ubuf,size_t * ulen)293 looks_extended(const unsigned char *buf, size_t nbytes, unichar *ubuf,
294 size_t *ulen)
295 {
296 size_t i;
297
298 *ulen = 0;
299
300 for (i = 0; i < nbytes; i++) {
301 int t = text_chars[buf[i]];
302
303 if (t != T && t != I && t != X)
304 return 0;
305
306 ubuf[(*ulen)++] = buf[i];
307 }
308
309 return 1;
310 }
311
312 /*
313 * Decide whether some text looks like UTF-8. Returns:
314 *
315 * -1: invalid UTF-8
316 * 0: uses odd control characters, so doesn't look like text
317 * 1: 7-bit text
318 * 2: definitely UTF-8 text (valid high-bit set bytes)
319 *
320 * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen;
321 * ubuf must be big enough!
322 */
323 protected int
file_looks_utf8(const unsigned char * buf,size_t nbytes,unichar * ubuf,size_t * ulen)324 file_looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
325 {
326 size_t i;
327 int n;
328 unichar c;
329 int gotone = 0, ctrl = 0;
330
331 if (ubuf)
332 *ulen = 0;
333
334 for (i = 0; i < nbytes; i++) {
335 if ((buf[i] & 0x80) == 0) { /* 0xxxxxxx is plain ASCII */
336 /*
337 * Even if the whole file is valid UTF-8 sequences,
338 * still reject it if it uses weird control characters.
339 */
340
341 if (text_chars[buf[i]] != T)
342 ctrl = 1;
343
344 if (ubuf)
345 ubuf[(*ulen)++] = buf[i];
346 } else if ((buf[i] & 0x40) == 0) { /* 10xxxxxx never 1st byte */
347 return -1;
348 } else { /* 11xxxxxx begins UTF-8 */
349 int following;
350
351 if ((buf[i] & 0x20) == 0) { /* 110xxxxx */
352 c = buf[i] & 0x1f;
353 following = 1;
354 } else if ((buf[i] & 0x10) == 0) { /* 1110xxxx */
355 c = buf[i] & 0x0f;
356 following = 2;
357 } else if ((buf[i] & 0x08) == 0) { /* 11110xxx */
358 c = buf[i] & 0x07;
359 following = 3;
360 } else if ((buf[i] & 0x04) == 0) { /* 111110xx */
361 c = buf[i] & 0x03;
362 following = 4;
363 } else if ((buf[i] & 0x02) == 0) { /* 1111110x */
364 c = buf[i] & 0x01;
365 following = 5;
366 } else
367 return -1;
368
369 for (n = 0; n < following; n++) {
370 i++;
371 if (i >= nbytes)
372 goto done;
373
374 if ((buf[i] & 0x80) == 0 || (buf[i] & 0x40))
375 return -1;
376
377 c = (c << 6) + (buf[i] & 0x3f);
378 }
379
380 if (ubuf)
381 ubuf[(*ulen)++] = c;
382 gotone = 1;
383 }
384 }
385 done:
386 return ctrl ? 0 : (gotone ? 2 : 1);
387 }
388
389 /*
390 * Decide whether some text looks like UTF-8 with BOM. If there is no
391 * BOM, return -1; otherwise return the result of looks_utf8 on the
392 * rest of the text.
393 */
394 private int
looks_utf8_with_BOM(const unsigned char * buf,size_t nbytes,unichar * ubuf,size_t * ulen)395 looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes, unichar *ubuf,
396 size_t *ulen)
397 {
398 if (nbytes > 3 && buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf)
399 return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen);
400 else
401 return -1;
402 }
403
404 private int
looks_utf7(const unsigned char * buf,size_t nbytes,unichar * ubuf,size_t * ulen)405 looks_utf7(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
406 {
407 if (nbytes > 4 && buf[0] == '+' && buf[1] == '/' && buf[2] == 'v')
408 switch (buf[3]) {
409 case '8':
410 case '9':
411 case '+':
412 case '/':
413 if (ubuf)
414 *ulen = 0;
415 return 1;
416 default:
417 return -1;
418 }
419 else
420 return -1;
421 }
422
423 private int
looks_ucs16(const unsigned char * bf,size_t nbytes,unichar * ubf,size_t * ulen)424 looks_ucs16(const unsigned char *bf, size_t nbytes, unichar *ubf,
425 size_t *ulen)
426 {
427 int bigend;
428 size_t i;
429
430 if (nbytes < 2)
431 return 0;
432
433 if (bf[0] == 0xff && bf[1] == 0xfe)
434 bigend = 0;
435 else if (bf[0] == 0xfe && bf[1] == 0xff)
436 bigend = 1;
437 else
438 return 0;
439
440 *ulen = 0;
441
442 for (i = 2; i + 1 < nbytes; i += 2) {
443 /* XXX fix to properly handle chars > 65536 */
444
445 if (bigend)
446 ubf[(*ulen)++] = bf[i + 1]
447 | (CAST(unichar, bf[i]) << 8);
448 else
449 ubf[(*ulen)++] = bf[i]
450 | (CAST(unichar, bf[i + 1]) << 8);
451
452 if (ubf[*ulen - 1] == 0xfffe)
453 return 0;
454 if (ubf[*ulen - 1] < 128 &&
455 text_chars[CAST(size_t, ubf[*ulen - 1])] != T)
456 return 0;
457 }
458
459 return 1 + bigend;
460 }
461
462 private int
looks_ucs32(const unsigned char * bf,size_t nbytes,unichar * ubf,size_t * ulen)463 looks_ucs32(const unsigned char *bf, size_t nbytes, unichar *ubf,
464 size_t *ulen)
465 {
466 int bigend;
467 size_t i;
468
469 if (nbytes < 4)
470 return 0;
471
472 if (bf[0] == 0xff && bf[1] == 0xfe && bf[2] == 0 && bf[3] == 0)
473 bigend = 0;
474 else if (bf[0] == 0 && bf[1] == 0 && bf[2] == 0xfe && bf[3] == 0xff)
475 bigend = 1;
476 else
477 return 0;
478
479 *ulen = 0;
480
481 for (i = 4; i + 3 < nbytes; i += 4) {
482 /* XXX fix to properly handle chars > 65536 */
483
484 if (bigend)
485 ubf[(*ulen)++] = CAST(unichar, bf[i + 3])
486 | (CAST(unichar, bf[i + 2]) << 8)
487 | (CAST(unichar, bf[i + 1]) << 16)
488 | (CAST(unichar, bf[i]) << 24);
489 else
490 ubf[(*ulen)++] = CAST(unichar, bf[i + 0])
491 | (CAST(unichar, bf[i + 1]) << 8)
492 | (CAST(unichar, bf[i + 2]) << 16)
493 | (CAST(unichar, bf[i + 3]) << 24);
494
495 if (ubf[*ulen - 1] == 0xfffe)
496 return 0;
497 if (ubf[*ulen - 1] < 128 &&
498 text_chars[CAST(size_t, ubf[*ulen - 1])] != T)
499 return 0;
500 }
501
502 return 1 + bigend;
503 }
504 #undef F
505 #undef T
506 #undef I
507 #undef X
508
509 /*
510 * This table maps each EBCDIC character to an (8-bit extended) ASCII
511 * character, as specified in the rationale for the dd(1) command in
512 * draft 11.2 (September, 1991) of the POSIX P1003.2 standard.
513 *
514 * Unfortunately it does not seem to correspond exactly to any of the
515 * five variants of EBCDIC documented in IBM's _Enterprise Systems
516 * Architecture/390: Principles of Operation_, SA22-7201-06, Seventh
517 * Edition, July, 1999, pp. I-1 - I-4.
518 *
519 * Fortunately, though, all versions of EBCDIC, including this one, agree
520 * on most of the printing characters that also appear in (7-bit) ASCII.
521 * Of these, only '|', '!', '~', '^', '[', and ']' are in question at all.
522 *
523 * Fortunately too, there is general agreement that codes 0x00 through
524 * 0x3F represent control characters, 0x41 a nonbreaking space, and the
525 * remainder printing characters.
526 *
527 * This is sufficient to allow us to identify EBCDIC text and to distinguish
528 * between old-style and internationalized examples of text.
529 */
530
531 private unsigned char ebcdic_to_ascii[] = {
532 0, 1, 2, 3, 156, 9, 134, 127, 151, 141, 142, 11, 12, 13, 14, 15,
533 16, 17, 18, 19, 157, 133, 8, 135, 24, 25, 146, 143, 28, 29, 30, 31,
534 128, 129, 130, 131, 132, 10, 23, 27, 136, 137, 138, 139, 140, 5, 6, 7,
535 144, 145, 22, 147, 148, 149, 150, 4, 152, 153, 154, 155, 20, 21, 158, 26,
536 ' ', 160, 161, 162, 163, 164, 165, 166, 167, 168, 213, '.', '<', '(', '+', '|',
537 '&', 169, 170, 171, 172, 173, 174, 175, 176, 177, '!', '$', '*', ')', ';', '~',
538 '-', '/', 178, 179, 180, 181, 182, 183, 184, 185, 203, ',', '%', '_', '>', '?',
539 186, 187, 188, 189, 190, 191, 192, 193, 194, '`', ':', '#', '@', '\'','=', '"',
540 195, 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 196, 197, 198, 199, 200, 201,
541 202, 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', '^', 204, 205, 206, 207, 208,
542 209, 229, 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 210, 211, 212, '[', 214, 215,
543 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, ']', 230, 231,
544 '{', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 232, 233, 234, 235, 236, 237,
545 '}', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 238, 239, 240, 241, 242, 243,
546 '\\',159, 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 244, 245, 246, 247, 248, 249,
547 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 250, 251, 252, 253, 254, 255
548 };
549
550 #ifdef notdef
551 /*
552 * The following EBCDIC-to-ASCII table may relate more closely to reality,
553 * or at least to modern reality. It comes from
554 *
555 * http://ftp.s390.ibm.com/products/oe/bpxqp9.html
556 *
557 * and maps the characters of EBCDIC code page 1047 (the code used for
558 * Unix-derived software on IBM's 390 systems) to the corresponding
559 * characters from ISO 8859-1.
560 *
561 * If this table is used instead of the above one, some of the special
562 * cases for the NEL character can be taken out of the code.
563 */
564
565 private unsigned char ebcdic_1047_to_8859[] = {
566 0x00,0x01,0x02,0x03,0x9C,0x09,0x86,0x7F,0x97,0x8D,0x8E,0x0B,0x0C,0x0D,0x0E,0x0F,
567 0x10,0x11,0x12,0x13,0x9D,0x0A,0x08,0x87,0x18,0x19,0x92,0x8F,0x1C,0x1D,0x1E,0x1F,
568 0x80,0x81,0x82,0x83,0x84,0x85,0x17,0x1B,0x88,0x89,0x8A,0x8B,0x8C,0x05,0x06,0x07,
569 0x90,0x91,0x16,0x93,0x94,0x95,0x96,0x04,0x98,0x99,0x9A,0x9B,0x14,0x15,0x9E,0x1A,
570 0x20,0xA0,0xE2,0xE4,0xE0,0xE1,0xE3,0xE5,0xE7,0xF1,0xA2,0x2E,0x3C,0x28,0x2B,0x7C,
571 0x26,0xE9,0xEA,0xEB,0xE8,0xED,0xEE,0xEF,0xEC,0xDF,0x21,0x24,0x2A,0x29,0x3B,0x5E,
572 0x2D,0x2F,0xC2,0xC4,0xC0,0xC1,0xC3,0xC5,0xC7,0xD1,0xA6,0x2C,0x25,0x5F,0x3E,0x3F,
573 0xF8,0xC9,0xCA,0xCB,0xC8,0xCD,0xCE,0xCF,0xCC,0x60,0x3A,0x23,0x40,0x27,0x3D,0x22,
574 0xD8,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0xAB,0xBB,0xF0,0xFD,0xFE,0xB1,
575 0xB0,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F,0x70,0x71,0x72,0xAA,0xBA,0xE6,0xB8,0xC6,0xA4,
576 0xB5,0x7E,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0xA1,0xBF,0xD0,0x5B,0xDE,0xAE,
577 0xAC,0xA3,0xA5,0xB7,0xA9,0xA7,0xB6,0xBC,0xBD,0xBE,0xDD,0xA8,0xAF,0x5D,0xB4,0xD7,
578 0x7B,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0xAD,0xF4,0xF6,0xF2,0xF3,0xF5,
579 0x7D,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F,0x50,0x51,0x52,0xB9,0xFB,0xFC,0xF9,0xFA,0xFF,
580 0x5C,0xF7,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0xB2,0xD4,0xD6,0xD2,0xD3,0xD5,
581 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0xB3,0xDB,0xDC,0xD9,0xDA,0x9F
582 };
583 #endif
584
585 /*
586 * Copy buf[0 ... nbytes-1] into out[], translating EBCDIC to ASCII.
587 */
588 private void
from_ebcdic(const unsigned char * buf,size_t nbytes,unsigned char * out)589 from_ebcdic(const unsigned char *buf, size_t nbytes, unsigned char *out)
590 {
591 size_t i;
592
593 for (i = 0; i < nbytes; i++) {
594 out[i] = ebcdic_to_ascii[buf[i]];
595 }
596 }
597