summaryrefslogtreecommitdiffstats
path: root/src/encoding.c
blob: 9dbb9dd93ebf641e452caaeb4539f500316fdc80 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
/*
 * Copyright (c) Ian F. Darwin 1986-1995.
 * Software written by Ian F. Darwin and others;
 * maintained 1995-present by Christos Zoulas and others.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice immediately at the beginning of the file, without modification,
 *    this list of conditions, and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 */
/*
 * Encoding -- determine the character encoding of a text file.
 *
 * Joerg Wunsch <joerg@freebsd.org> wrote the original support for 8-bit
 * international characters.
 */

#include "file.h"

#ifndef	lint
FILE_RCSID("@(#)$File: encoding.c,v 1.42 2022/12/26 17:31:14 christos Exp $")
#endif	/* lint */

#include "magic.h"
#include <string.h>
#include <stdlib.h>


file_private int looks_ascii(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private int looks_utf8_with_BOM(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private int looks_utf7(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private int looks_ucs16(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private int looks_ucs32(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private int looks_latin1(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private int looks_extended(const unsigned char *, size_t, file_unichar_t *,
    size_t *);
file_private void from_ebcdic(const unsigned char *, size_t, unsigned char *);

#ifdef DEBUG_ENCODING
#define DPRINTF(a) printf a
#else
#define DPRINTF(a)
#endif

/*
 * Try to determine whether text is in some character code we can
 * identify.  Each of these tests, if it succeeds, will leave
 * the text converted into one-file_unichar_t-per-character Unicode in
 * ubuf, and the number of characters converted in ulen.
 */
file_protected int
file_encoding(struct magic_set *ms, const struct buffer *b,
    file_unichar_t **ubuf, size_t *ulen, const char **code,
    const char **code_mime, const char **type)
{
	const unsigned char *buf = CAST(const unsigned char *, b->fbuf);
	size_t nbytes = b->flen;
	size_t mlen;
	int rv = 1, ucs_type;
	file_unichar_t *udefbuf;
	size_t udeflen;

	if (ubuf == NULL)
		ubuf = &udefbuf;
	if (ulen == NULL)
		ulen = &udeflen;

	*type = "text";
	*ulen = 0;
	*code = "unknown";
	*code_mime = "binary";

	if (nbytes > ms->encoding_max)
		nbytes = ms->encoding_max;

	mlen = (nbytes + 1) * sizeof((*ubuf)[0]);
	*ubuf = CAST(file_unichar_t *, calloc(CAST(size_t, 1), mlen));
	if (*ubuf == NULL) {
		file_oomem(ms, mlen);
		goto done;
	}
	if (looks_ascii(buf, nbytes, *ubuf, ulen)) {
		if (looks_utf7(buf, nbytes, *ubuf, ulen) > 0) {
			DPRINTF(("utf-7 %" SIZE_T_FORMAT "u\n", *ulen));
			*code = "Unicode text, UTF-7";
			*code_mime = "utf-7";
		} else {
			DPRINTF(("ascii %" SIZE_T_FORMAT "u\n", *ulen));
			*code = "ASCII";
			*code_mime = "us-ascii";
		}
	} else if (looks_utf8_with_BOM(buf, nbytes, *ubuf, ulen) > 0) {
		DPRINTF(("utf8/bom %" SIZE_T_FORMAT "u\n", *ulen));
		*code = "Unicode text, UTF-8 (with BOM)";
		*code_mime = "utf-8";
	} else if (file_looks_utf8(buf, nbytes, *ubuf, ulen) > 1) {
		DPRINTF(("utf8 %" SIZE_T_FORMAT "u\n", *ulen));
		*code = "Unicode text, UTF-8";
		*code_mime = "utf-8";
	} else if ((ucs_type = looks_ucs32(buf, nbytes, *ubuf, ulen)) != 0) {
		if (ucs_type == 1) {
			*code = "Unicode text, UTF-32, little-endian";
			*code_mime = "utf-32le";
		} else {
			*code = "Unicode text, UTF-32, big-endian";
			*code_mime = "utf-32be";
		}
		DPRINTF(("ucs32 %" SIZE_T_FORMAT "u\n", *ulen));
	} else if ((ucs_type = looks_ucs16(buf, nbytes, *ubuf, ulen)) != 0) {
		if (ucs_type == 1) {
			*code = "Unicode text, UTF-16, little-endian";
			*code_mime = "utf-16le";
		} else {
			*code = "Unicode text, UTF-16, big-endian";
			*code_mime = "utf-16be";
		}
		DPRINTF(("ucs16 %" SIZE_T_FORMAT "u\n", *ulen));
	} else if (looks_latin1(buf, nbytes, *ubuf, ulen)) {
		DPRINTF(("latin1 %" SIZE_T_FORMAT "u\n", *ulen));
		*code = "ISO-8859";
		*code_mime = "iso-8859-1";
	} else if (looks_extended(buf, nbytes, *ubuf, ulen)) {
		DPRINTF(("extended %" SIZE_T_FORMAT "u\n", *ulen));
		*code = "Non-ISO extended-ASCII";
		*code_mime = "unknown-8bit";
	} else {
		unsigned char *nbuf;

		mlen = (nbytes + 1) * sizeof(nbuf[0]);
		if ((nbuf = CAST(unsigned char *, malloc(mlen))) == NULL) {
			file_oomem(ms, mlen);
			goto done;
		}
		from_ebcdic(buf, nbytes, nbuf);

		if (looks_ascii(nbuf, nbytes, *ubuf, ulen)) {
			DPRINTF(("ebcdic %" SIZE_T_FORMAT "u\n", *ulen));
			*code = "EBCDIC";
			*code_mime = "ebcdic";
		} else if (looks_latin1(nbuf, nbytes, *ubuf, ulen)) {
			DPRINTF(("ebcdic/international %" SIZE_T_FORMAT "u\n",
			    *ulen));
			*code = "International EBCDIC";
			*code_mime = "ebcdic";
		} else { /* Doesn't look like text at all */
			DPRINTF(("binary\n"));
			rv = 0;
			*type = "binary";
		}
		free(nbuf);
	}

 done:
	if (ubuf == &udefbuf)
		free(udefbuf);

	return rv;
}

/*
 * This table reflects a particular philosophy about what constitutes
 * "text," and there is room for disagreement about it.
 *
 * Version 3.31 of the file command considered a file to be ASCII if
 * each of its characters was approved by either the isascii() or
 * isalpha() function.  On most systems, this would mean that any
 * file consisting only of characters in the range 0x00 ... 0x7F
 * would be called ASCII text, but many systems might reasonably
 * consider some characters outside this range to be alphabetic,
 * so the file command would call such characters ASCII.  It might
 * have been more accurate to call this "considered textual on the
 * local system" than "ASCII."
 *
 * It considered a file to be "International language text" if each
 * of its characters was either an ASCII printing character (according
 * to the real ASCII standard, not the above test), a character in
 * the range 0x80 ... 0xFF, or one of the following control characters:
 * backspace, tab, line feed, vertical tab, form feed, carriage return,
 * escape.  No attempt was made to determine the language in which files
 * of this type were written.
 *
 *
 * The table below considers a file to be ASCII if all of its characters
 * are either ASCII printing characters (again, according to the X3.4
 * standard, not isascii()) or any of the following controls: bell,
 * backspace, tab, line feed, form feed, carriage return, esc, nextline.
 *
 * I include bell because some programs (particularly shell scripts)
 * use it literally, even though it is rare in normal text.  I exclude
 * vertical tab because it never seems to be used in real text.  I also
 * include, with hesitation, the X3.64/ECMA-43 control nextline (0x85),
 * because that's what the dd EBCDIC->ASCII table maps the EBCDIC newline
 * character to.  It might be more appropriate to include it in the 8859
 * set instead of the ASCII set, but it's got to be included in *something*
 * we recognize or EBCDIC files aren't going to be considered textual.
 * Some old Unix source files use SO/SI (^N/^O) to shift between Greek
 * and Latin characters, so these should possibly be allowed.  But they
 * make a real mess on VT100-style displays if they're not paired properly,
 * so we are probably better off not calling them text.
 *
 * A file is considered to be ISO-8859 text if its characters are all
 * either ASCII, according to the above definition, or printing characters
 * from the ISO-8859 8-bit extension, characters 0xA0 ... 0xFF.
 *
 * Finally, a file is considered to be international text from some other
 * character code if its characters are all either ISO-8859 (according to
 * the above definition) or characters in the range 0x80 ... 0x9F, which
 * ISO-8859 considers to be control characters but the IBM PC and Macintosh
 * consider to be printing characters.
 */

#define F 0   /* character never appears in text */
#define T 1   /* character appears in plain ASCII text */
#define I 2   /* character appears in ISO-8859 text */
#define X 3   /* character appears in non-ISO extended ASCII (Mac, IBM PC) */

file_private char text_chars[256] = {
	/*                  BEL BS HT LF VT FF CR    */
	F, F, F, F, F, F, F, T, T, T, T, T, T, T, F, F,  /* 0x0X */
	/*                              ESC          */
	F, F, F, F, F, F, F, F, F, F, F, T, F, F, F, F,  /* 0x1X */
	T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x2X */
	T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x3X */
	T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x4X */
	T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x5X */
	T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T,  /* 0x6X */
	T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, F,  /* 0x7X */
	/*            NEL                            */
	X, X, X, X, X, T, X, X, X, X, X, X, X, X, X, X,  /* 0x8X */
	X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, X,  /* 0x9X */
	I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xaX */
	I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xbX */
	I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xcX */
	I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xdX */
	I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I,  /* 0xeX */
	I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I   /* 0xfX */
};

#define LOOKS(NAME, COND) \
file_private int \
looks_ ## NAME(const unsigned char *buf, size_t nbytes, file_unichar_t *ubuf, \
    size_t *ulen) \
{ \
	size_t i; \
\
	*ulen = 0; \
\
	for (i = 0; i < nbytes; i++) { \
		int t = text_chars[buf[i]]; \
\
		if (COND) \
			return 0; \
\
		ubuf[(*ulen)++] = buf[i]; \
	} \
	return 1; \
}

LOOKS(ascii, t != T)
LOOKS(latin1, t != T && t != I)
LOOKS(extended, t != T && t != I && t != X)

/*
 * Decide whether some text looks like UTF-8. Returns:
 *
 *     -1: invalid UTF-8
 *      0: uses odd control characters, so doesn't look like text
 *      1: 7-bit text
 *      2: definitely UTF-8 text (valid high-bit set bytes)
 *
 * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen;
 * ubuf must be big enough!
 */

// from: https://golang.org/src/unicode/utf8/utf8.go

#define	XX 0xF1 // invalid: size 1
#define	AS 0xF0 // ASCII: size 1
#define	S1 0x02 // accept 0, size 2
#define	S2 0x13 // accept 1, size 3
#define	S3 0x03 // accept 0, size 3
#define	S4 0x23 // accept 2, size 3
#define	S5 0x34 // accept 3, size 4
#define	S6 0x04 // accept 0, size 4
#define	S7 0x44 // accept 4, size 4

#define LOCB 0x80
#define HICB 0xBF

// first is information about the first byte in a UTF-8 sequence.
static const uint8_t first[] = {
    //   1   2   3   4   5   6   7   8   9   A   B   C   D   E   F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x00-0x0F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x10-0x1F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x20-0x2F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x30-0x3F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x40-0x4F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x50-0x5F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x60-0x6F
    AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, AS, // 0x70-0x7F
    //   1   2   3   4   5   6   7   8   9   A   B   C   D   E   F
    XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, // 0x80-0x8F
    XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, // 0x90-0x9F
    XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, // 0xA0-0xAF
    XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, // 0xB0-0xBF
    XX, XX, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, // 0xC0-0xCF
    S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, S1, // 0xD0-0xDF
    S2, S3, S3, S3, S3, S3, S3, S3, S3, S3, S3, S3, S3, S4, S3, S3, // 0xE0-0xEF
    S5, S6, S6, S6, S7, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, XX, // 0xF0-0xFF
};

// acceptRange gives the range of valid values for the second byte in a UTF-8
// sequence.
struct accept_range {
	uint8_t lo; // lowest value for second byte.
	uint8_t hi; // highest value for second byte.
} accept_ranges[16] = {
// acceptRanges has size 16 to avoid bounds checks in the code that uses it.
	{ LOCB, HICB },
	{ 0xA0, HICB },
	{ LOCB, 0x9F },
	{ 0x90, HICB },
	{ LOCB, 0x8F },
};

file_protected int
file_looks_utf8(const unsigned char *buf, size_t nbytes, file_unichar_t *ubuf,
    size_t *ulen)
{
	size_t i;
	int n;
	file_unichar_t c;
	int gotone = 0, ctrl = 0;

	if (ubuf)
		*ulen = 0;

	for (i = 0; i < nbytes; i++) {
		if ((buf[i] & 0x80) == 0) {	   /* 0xxxxxxx is plain ASCII */
			/*
			 * Even if the whole file is valid UTF-8 sequences,
			 * still reject it if it uses weird control characters.
			 */

			if (text_chars[buf[i]] != T)
				ctrl = 1;

			if (ubuf)
				ubuf[(*ulen)++] = buf[i];
		} else if ((buf[i] & 0x40) == 0) { /* 10xxxxxx never 1st byte */
			return -1;
		} else {			   /* 11xxxxxx begins UTF-8 */
			int following;
			uint8_t x = first[buf[i]];
			const struct accept_range *ar =
			    &accept_ranges[(unsigned int)x >> 4];
			if (x == XX)
				return -1;

			if ((buf[i] & 0x20) == 0) {		/* 110xxxxx */
				c = buf[i] & 0x1f;
				following = 1;
			} else if ((buf[i] & 0x10) == 0) {	/* 1110xxxx */
				c = buf[i] & 0x0f;
				following = 2;
			} else if ((buf[i] & 0x08) == 0) {	/* 11110xxx */
				c = buf[i] & 0x07;
				following = 3;
			} else if ((buf[i] & 0x04) == 0) {	/* 111110xx */
				c = buf[i] & 0x03;
				following = 4;
			} else if ((buf[i] & 0x02) == 0) {	/* 1111110x */
				c = buf[i] & 0x01;
				following = 5;
			} else
				return -1;

			for (n = 0; n < following; n++) {
				i++;
				if (i >= nbytes)
					goto done;

				if (n == 0 &&
				     (buf[i] < ar->lo || buf[i] > ar->hi))
					return -1;

				if ((buf[i] & 0x80) == 0 || (buf[i] & 0x40))
					return -1;

				c = (c << 6) + (buf[i] & 0x3f);
			}

			if (ubuf)
				ubuf[(*ulen)++] = c;
			gotone = 1;
		}
	}
done:
	return ctrl ? 0 : (gotone ? 2 : 1);
}

/*
 * Decide whether some text looks like UTF-8 with BOM. If there is no
 * BOM, return -1; otherwise return the result of looks_utf8 on the
 * rest of the text.
 */
file_private int
looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes,
    file_unichar_t *ubuf, size_t *ulen)
{
	if (nbytes > 3 && buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf)
		return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen);
	else
		return -1;
}

file_private int
looks_utf7(const unsigned char *buf, size_t nbytes, file_unichar_t *ubuf,
    size_t *ulen)
{
	if (nbytes > 4 && buf[0] == '+' && buf[1] == '/' && buf[2] == 'v')
		switch (buf[3]) {
		case '8':
		case '9':
		case '+':
		case '/':
			if (ubuf)
				*ulen = 0;
			return 1;
		default:
			return -1;
		}
	else
		return -1;
}

#define UCS16_NOCHAR(c) ((c) >= 0xfdd0 && (c) <= 0xfdef)
#define UCS16_HISURR(c) ((c) >= 0xd800 && (c) <= 0xdbff)
#define UCS16_LOSURR(c) ((c) >= 0xdc00 && (c) <= 0xdfff)

file_private int
looks_ucs16(const unsigned char *bf, size_t nbytes, file_unichar_t *ubf,
    size_t *ulen)
{
	int bigend;
	uint32_t hi;
	size_t i;

	if (nbytes < 2)
		return 0;

	if (bf[0] == 0xff && bf[1] == 0xfe)
		bigend = 0;
	else if (bf[0] == 0xfe && bf[1] == 0xff)
		bigend = 1;
	else
		return 0;

	*ulen = 0;
	hi = 0;

	for (i = 2; i + 1 < nbytes; i += 2) {
		uint32_t uc;

		if (bigend)
			uc = CAST(uint32_t,
			    bf[i + 1] | (CAST(file_unichar_t, bf[i]) << 8));
		else
			uc = CAST(uint32_t,
			    bf[i] | (CAST(file_unichar_t, bf[i + 1]) << 8));

		uc &= 0xffff;

		switch (uc) {
		case 0xfffe:
		case 0xffff:
			return 0;
		default:
			if (UCS16_NOCHAR(uc))
				return 0;
			break;
		}
		if (hi) {
			if (!UCS16_LOSURR(uc))
				return 0;
			uc = 0x10000 + 0x400 * (hi - 1) + (uc - 0xdc00);
			hi = 0;
		}
		if (uc < 128 && text_chars[CAST(size_t, uc)] != T)
			return 0;
		ubf[(*ulen)++] = uc;
		if (UCS16_HISURR(uc))
			hi = uc - 0xd800 + 1;
		if (UCS16_LOSURR(uc))
			return 0;
	}

	return 1 + bigend;
}

file_private int
looks_ucs32(const unsigned char *bf, size_t nbytes, file_unichar_t *ubf,
    size_t *ulen)
{
	int bigend;
	size_t i;

	if (nbytes < 4)
		return 0;

	if (bf[0] == 0xff && bf[1] == 0xfe && bf[2] == 0 && bf[3] == 0)
		bigend = 0;
	else if (bf[0] == 0 && bf[1] == 0 && bf[2] == 0xfe && bf[3] == 0xff)
		bigend = 1;
	else
		return 0;

	*ulen = 0;

	for (i = 4; i + 3 < nbytes; i += 4) {
		/* XXX fix to properly handle chars > 65536 */

		if (bigend)
			ubf[(*ulen)++] = CAST(file_unichar_t, bf[i + 3])
			    | (CAST(file_unichar_t, bf[i + 2]) << 8)
			    | (CAST(file_unichar_t, bf[i + 1]) << 16)
			    | (CAST(file_unichar_t, bf[i]) << 24);
		else
			ubf[(*ulen)++] = CAST(file_unichar_t, bf[i + 0])
			    | (CAST(file_unichar_t, bf[i + 1]) << 8) 
			    | (CAST(file_unichar_t, bf[i + 2]) << 16)
			    | (CAST(file_unichar_t, bf[i + 3]) << 24);

		if (ubf[*ulen - 1] == 0xfffe)
			return 0;
		if (ubf[*ulen - 1] < 128 &&
		    text_chars[CAST(size_t, ubf[*ulen - 1])] != T)
			return 0;
	}

	return 1 + bigend;
}
#undef F
#undef T
#undef I
#undef X

/*
 * This table maps each EBCDIC character to an (8-bit extended) ASCII
 * character, as specified in the rationale for the dd(1) command in
 * draft 11.2 (September, 1991) of the POSIX P1003.2 standard.
 *
 * Unfortunately it does not seem to correspond exactly to any of the
 * five variants of EBCDIC documented in IBM's _Enterprise Systems
 * Architecture/390: Principles of Operation_, SA22-7201-06, Seventh
 * Edition, July, 1999, pp. I-1 - I-4.
 *
 * Fortunately, though, all versions of EBCDIC, including this one, agree
 * on most of the printing characters that also appear in (7-bit) ASCII.
 * Of these, only '|', '!', '~', '^', '[', and ']' are in question at all.
 *
 * Fortunately too, there is general agreement that codes 0x00 through
 * 0x3F represent control characters, 0x41 a nonbreaking space, and the
 * remainder printing characters.
 *
 * This is sufficient to allow us to identify EBCDIC text and to distinguish
 * between old-style and internationalized examples of text.
 */

file_private unsigned char ebcdic_to_ascii[] = {
  0,   1,   2,   3, 156,   9, 134, 127, 151, 141, 142,  11,  12,  13,  14,  15,
 16,  17,  18,  19, 157, 133,   8, 135,  24,  25, 146, 143,  28,  29,  30,  31,
128, 129, 130, 131, 132,  10,  23,  27, 136, 137, 138, 139, 140,   5,   6,   7,
144, 145,  22, 147, 148, 149, 150,   4, 152, 153, 154, 155,  20,  21, 158,  26,
' ', 160, 161, 162, 163, 164, 165, 166, 167, 168, 213, '.', '<', '(', '+', '|',
'&', 169, 170, 171, 172, 173, 174, 175, 176, 177, '!', '$', '*', ')', ';', '~',
'-', '/', 178, 179, 180, 181, 182, 183, 184, 185, 203, ',', '%', '_', '>', '?',
186, 187, 188, 189, 190, 191, 192, 193, 194, '`', ':', '#', '@', '\'','=', '"',
195, 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 196, 197, 198, 199, 200, 201,
202, 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', '^', 204, 205, 206, 207, 208,
209, 229, 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 210, 211, 212, '[', 214, 215,
216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, ']', 230, 231,
'{', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 232, 233, 234, 235, 236, 237,
'}', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 238, 239, 240, 241, 242, 243,
'\\',159, 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 244, 245, 246, 247, 248, 249,
'0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 250, 251, 252, 253, 254, 255
};

#ifdef notdef
/*
 * The following EBCDIC-to-ASCII table may relate more closely to reality,
 * or at least to modern reality.  It comes from
 *
 *   http://ftp.s390.ibm.com/products/oe/bpxqp9.html
 *
 * and maps the characters of EBCDIC code page 1047 (the code used for
 * Unix-derived software on IBM's 390 systems) to the corresponding
 * characters from ISO 8859-1.
 *
 * If this table is used instead of the above one, some of the special
 * cases for the NEL character can be taken out of the code.
 */

file_private unsigned char ebcdic_1047_to_8859[] = {
0x00,0x01,0x02,0x03,0x9C,0x09,0x86,0x7F,0x97,0x8D,0x8E,0x0B,0x0C,0x0D,0x0E,0x0F,
0x10,0x11,0x12,0x13,0x9D,0x0A,0x08,0x87,0x18,0x19,0x92,0x8F,0x1C,0x1D,0x1E,0x1F,
0x80,0x81,0x82,0x83,0x84,0x85,0x17,0x1B,0x88,0x89,0x8A,0x8B,0x8C,0x05,0x06,0x07,
0x90,0x91,0x16,0x93,0x94,0x95,0x96,0x04,0x98,0x99,0x9A,0x9B,0x14,0x15,0x9E,0x1A,
0x20,0xA0,0xE2,0xE4,0xE0,0xE1,0xE3,0xE5,0xE7,0xF1,0xA2,0x2E,0x3C,0x28,0x2B,0x7C,
0x26,0xE9,0xEA,0xEB,0xE8,0xED,0xEE,0xEF,0xEC,0xDF,0x21,0x24,0x2A,0x29,0x3B,0x5E,
0x2D,0x2F,0xC2,0xC4,0xC0,0xC1,0xC3,0xC5,0xC7,0xD1,0xA6,0x2C,0x25,0x5F,0x3E,0x3F,
0xF8,0xC9,0xCA,0xCB,0xC8,0xCD,0xCE,0xCF,0xCC,0x60,0x3A,0x23,0x40,0x27,0x3D,0x22,
0xD8,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0xAB,0xBB,0xF0,0xFD,0xFE,0xB1,
0xB0,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F,0x70,0x71,0x72,0xAA,0xBA,0xE6,0xB8,0xC6,0xA4,
0xB5,0x7E,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0xA1,0xBF,0xD0,0x5B,0xDE,0xAE,
0xAC,0xA3,0xA5,0xB7,0xA9,0xA7,0xB6,0xBC,0xBD,0xBE,0xDD,0xA8,0xAF,0x5D,0xB4,0xD7,
0x7B,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0xAD,0xF4,0xF6,0xF2,0xF3,0xF5,
0x7D,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F,0x50,0x51,0x52,0xB9,0xFB,0xFC,0xF9,0xFA,0xFF,
0x5C,0xF7,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0xB2,0xD4,0xD6,0xD2,0xD3,0xD5,
0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0xB3,0xDB,0xDC,0xD9,0xDA,0x9F
};
#endif

/*
 * Copy buf[0 ... nbytes-1] into out[], translating EBCDIC to ASCII.
 */
file_private void
from_ebcdic(const unsigned char *buf, size_t nbytes, unsigned char *out)
{
	size_t i;

	for (i = 0; i < nbytes; i++) {
		out[i] = ebcdic_to_ascii[buf[i]];
	}
}