summaryrefslogtreecommitdiffstats
path: root/src/UCAux.c
blob: a10f6249b577fba6ee86269a27334c66a03a4a82 (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
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
/*
 * $LynxId: UCAux.c,v 1.59 2024/01/15 11:24:17 tom Exp $
 */
#include <HTUtils.h>

#include <HTCJK.h>
#include <UCMap.h>
#include <UCDefs.h>
#include <HTStream.h>
#include <UCAux.h>
#include <LYCharSets.h>
#include <LYCurses.h>
#include <LYStrings.h>

BOOL UCCanUniTranslateFrom(int from)
{
    if (from < 0)
	return NO;
#ifndef USE_JAPANESEUTF8_SUPPORT
    if (LYCharSet_UC[from].enc == UCT_ENC_CJK)
	return NO;
#endif
    if (!strcmp(LYCharSet_UC[from].MIMEname, "x-transparent"))
	return NO;

    /* others YES */
    return YES;
}

BOOL UCCanTranslateUniTo(int to)
{
    if (to < 0)
	return NO;

    return YES;			/* well at least some characters... */
}

BOOL UCCanTranslateFromTo(int from,
			  int to)
{
    if (from == to)
	return YES;
    if (from < 0 || to < 0)
	return NO;
    if (from == LATIN1)
	return UCCanTranslateUniTo(to);
    if (to == LATIN1 || LYCharSet_UC[to].enc == UCT_ENC_UTF8)
	return UCCanUniTranslateFrom(from);
    {
	const char *fromname = LYCharSet_UC[from].MIMEname;
	const char *toname = LYCharSet_UC[to].MIMEname;

	if (!strcmp(fromname, "x-transparent") ||
	    !strcmp(toname, "x-transparent")) {
	    return YES;		/* ??? */
	} else if (!strcmp(fromname, "us-ascii")) {
	    return YES;
	}
	if (LYCharSet_UC[from].enc == UCT_ENC_CJK) {
	    /*
	     * CJK mode may be off (i.e., !IS_CJK_TTY) because the current
	     * document is not CJK, but the check may be for capability in
	     * relation to another document, for which CJK mode might be turned
	     * on when retrieved.  Thus, when the from charset is CJK, check if
	     * the to charset is CJK, and return NO or YES in relation to that. 
	     * - FM
	     */
	    if (LYCharSet_UC[to].enc != UCT_ENC_CJK)
		return NO;
	    if ((!strcmp(toname, "euc-jp") ||
		 !strcmp(toname, "shift_jis")) &&
		(!strcmp(fromname, "euc-jp") ||
		 !strcmp(fromname, "shift_jis")))
		return YES;
	    /*
	     * The euc-cn and euc-kr charsets were handled by the (from == to)
	     * above, so we need not check those.  - FM
	     */
	    return NO;
	}
    }
    return YES;			/* others YES */
}

/*
 *  Returns YES if no translation necessary (because
 *  charsets are equal, are equivalent, etc.).
 */
BOOL UCNeedNotTranslate(int from,
			int to)
{
    const char *fromname;
    const char *toname;

    if (from == to)
	return YES;
    if (from < 0)
	return NO;		/* ??? */
    if (LYCharSet_UC[from].enc == UCT_ENC_7BIT) {
	return YES;		/* Only 7bit chars. */
    }
    fromname = LYCharSet_UC[from].MIMEname;
    if (!strcmp(fromname, "x-transparent") ||
	!strcmp(fromname, "us-ascii")) {
	return YES;
    }
    if (to < 0)
	return NO;		/* ??? */
    if (to == LATIN1) {
	if (LYCharSet_UC[from].codepoints & (UCT_CP_SUBSETOF_LAT1))
	    return YES;
    }
    toname = LYCharSet_UC[to].MIMEname;
    if (!strcmp(toname, "x-transparent")) {
	return YES;
    }
    if (LYCharSet_UC[to].enc == UCT_ENC_UTF8) {
	return NO;
    }
    if (from == LATIN1) {
	if (LYCharSet_UC[from].codepoints & (UCT_CP_SUPERSETOF_LAT1))
	    return YES;
    }
    if (LYCharSet_UC[from].enc == UCT_ENC_CJK) {
	if (!IS_CJK_TTY)	/* Use that global flag, for now. */
	    return NO;
	if (HTCJK == JAPANESE &&
	    (!strcmp(fromname, "euc-jp") ||
	     !strcmp(fromname, "shift_jis")))
	    return YES;		/* translate internally by lynx, no unicode */
	return NO;		/* If not handled by (from == to) above. */
    }
    return NO;
}

/*
 *  The idea here is that any stage of the stream pipe which is interested
 *  in some charset dependent processing will call this function.
 *  Given input and output charsets, this function will set various flags
 *  in a UCTransParams structure that _suggest_ to the caller what to do.
 *
 *  Should be called once when a stage starts processing text (and the
 *  input and output charsets are known), or whenever one of input or
 *  output charsets has changed (e.g., by SGML.c stage after HTML.c stage
 *  has processed a META tag).
 *  The global flags (LYRawMode, HTPassEightBitRaw etc.) are currently
 *  not taken into account here (except for HTCJK, somewhat), it's still
 *  up to the caller to do something about them. - KW
 */
void UCSetTransParams(UCTransParams * pT, int cs_in,
		      const LYUCcharset *p_in,
		      int cs_out,
		      const LYUCcharset *p_out)
{
    CTRACE((tfp, "UCSetTransParams: from %s(%d) to %s(%d)\n",
	    p_in->MIMEname, UCGetLYhndl_byMIME(p_in->MIMEname),
	    p_out->MIMEname, UCGetLYhndl_byMIME(p_out->MIMEname)));

    /*
     * Initialize this element to FALSE, and set it TRUE below if we're dealing
     * with VISCII.  - FM
     */
    pT->trans_C0_to_uni = FALSE;

    /*
     * The "transparent" display character set is a "super raw mode".  - FM
     */
    pT->transp = (BOOL) (!strcmp(p_in->MIMEname, "x-transparent") ||
			 !strcmp(p_out->MIMEname, "x-transparent"));

    /*
     * UCS-2 is handled as a special case in SGML_write().
     */
    pT->ucs_mode = 0;

    if (pT->transp) {
	/*
	 * Set up the structure for "transparent".  - FM
	 */
	pT->do_cjk = FALSE;
	pT->decode_utf8 = FALSE;
	pT->output_utf8 = FALSE;	/* We may, but won't know about it. - KW */
	pT->do_8bitraw = TRUE;
	pT->use_raw_char_in = TRUE;
	pT->strip_raw_char_in = FALSE;
	pT->pass_160_173_raw = TRUE;
	pT->repl_translated_C0 = (BOOL) (p_out->enc == UCT_ENC_8BIT_C0);
	pT->trans_C0_to_uni = (BOOL) (p_in->enc == UCT_ENC_8BIT_C0 ||
				      p_out->enc == UCT_ENC_8BIT_C0);
    } else {
	/*
	 * Initialize local flags.  - FM
	 */
	BOOL intm_ucs = FALSE;
	BOOL use_ucs = FALSE;

	/*
	 * Set this element if we want to treat the input as CJK.  - FM
	 */
	pT->do_cjk = (BOOL) ((p_in->enc == UCT_ENC_CJK) &&
			     (
				 IS_CJK_TTY
#ifdef EXP_CHINESEUTF8_SUPPORT
				 || !strcmp(p_in->MIMEname, "euc-cn")
				 || !strcmp(p_in->MIMEname, "big5")
				 || !strcmp(p_in->MIMEname, "euc-kr")
#endif
			     )
	    );
	/*
	 * Set these elements based on whether we are dealing with UTF-8.  - FM
	 */
	pT->decode_utf8 = (BOOL) (p_in->enc == UCT_ENC_UTF8);
	pT->output_utf8 = (BOOL) (p_out->enc == UCT_ENC_UTF8);
	if (pT->do_cjk) {
	    /*
	     * Set up the structure for a CJK input with
	     * a CJK output (IS_CJK_TTY).  - FM
	     */
	    pT->trans_to_uni = FALSE;
#ifdef EXP_CHINESEUTF8_SUPPORT
	    if (!strcmp(p_in->MIMEname, "euc-cn") ||
		!strcmp(p_in->MIMEname, "big5") ||
		!strcmp(p_in->MIMEname, "euc-kr")) {
		pT->trans_to_uni = (BOOL) UCCanUniTranslateFrom(cs_in);
	    }
#endif
	    pT->do_8bitraw = FALSE;
	    pT->pass_160_173_raw = TRUE;
	    pT->use_raw_char_in = FALSE;	/* Not used for CJK. - KW */
	    pT->repl_translated_C0 = FALSE;
	    pT->trans_from_uni = FALSE;		/* Not used for CJK. - KW */
	} else {
	    /*
	     * Set up for all other charset combinations.  The intm_ucs flag is
	     * set TRUE if the input charset is iso-8859-1 or UTF-8, or largely
	     * equivalent to them, i.e., if we have UCS without having to do a
	     * table translation.
	     */
	    intm_ucs = (BOOL) (cs_in == LATIN1 || pT->decode_utf8 ||
			       (p_in->codepoints &
				(UCT_CP_SUBSETOF_LAT1 | UCT_CP_SUBSETOF_UCS2)));
	    /*
	     * pT->trans_to_uni is set TRUE if we do not have that as input
	     * already, and we can translate to Unicode.  Note that UTF-8
	     * always is converted to Unicode in functions that use the
	     * transformation structure, so it is treated as already Unicode
	     * here.
	     */
	    pT->trans_to_uni = (BOOL) (!intm_ucs &&
				       UCCanUniTranslateFrom(cs_in));
	    /*
	     * We set this if we are translating to Unicode and what normally
	     * are low value control characters in fact are encoding octets for
	     * the input charset (presently, this applies to VISCII).  - FM
	     */
	    pT->trans_C0_to_uni = (BOOL) (pT->trans_to_uni &&
					  p_in->enc == UCT_ENC_8BIT_C0);
	    /*
	     * We set this, presently, for VISCII.  - FM
	     */
	    pT->repl_translated_C0 = (BOOL) (p_out->enc == UCT_ENC_8BIT_C0);
	    /*
	     * Currently unused for any charset combination.
	     * Should always be FALSE
	     */
	    pT->strip_raw_char_in = FALSE;
	    /*
	     * use_ucs should be set TRUE if we have or will create Unicode
	     * values for input octets or UTF multibytes.  - FM
	     */
	    use_ucs = (BOOL) (intm_ucs || pT->trans_to_uni);
	    /*
	     * This is set TRUE if use_ucs was set FALSE.  It is complementary
	     * to the HTPassEightBitRaw flag, which is set TRUE or FALSE
	     * elsewhere based on the raw mode setting in relation to the
	     * current Display Character Set.  - FM
	     */
	    pT->do_8bitraw = (BOOL) (!use_ucs);
	    /*
	     * This is set TRUE when 160 and 173 should not be treated as nbsp
	     * and shy, respectively.  - FM
	     */
	    pT->pass_160_173_raw = (BOOL) (!use_ucs &&
					   !(p_in->like8859 & UCT_R_8859SPECL));
	    /*
	     * This is set when the input and output charsets match, and they
	     * are not ones which should go through a Unicode translation
	     * process anyway.  - FM
	     */
	    pT->use_raw_char_in = (BOOL) (!pT->output_utf8 &&
					  cs_in == cs_out &&
					  !pT->trans_C0_to_uni);
	    /*
	     * This should be set TRUE when we expect to have done translation
	     * to Unicode or had the equivalent as input, can translate it to
	     * our output charset, and normally want to do so.  The latter
	     * depends on the pT->do_8bitraw and pT->use_raw_char_in values set
	     * above, but also on HTPassEightBitRaw in any functions which use
	     * the transformation structure..  - FM
	     */
	    pT->trans_from_uni = (BOOL) (use_ucs && !pT->do_8bitraw &&
					 !pT->use_raw_char_in &&
					 UCCanTranslateUniTo(cs_out));
	}
    }
    CTRACE((tfp, "UCSetTransParams (done):\n"));
    CTRACE((tfp, "  transp:             %d\n", pT->transp));
    CTRACE((tfp, "  do_cjk:             %d\n", pT->do_cjk));
    CTRACE((tfp, "  decode_utf8:        %d\n", pT->decode_utf8));
    CTRACE((tfp, "  output_utf8:        %d\n", pT->output_utf8));
    CTRACE((tfp, "  do_8bitraw:         %d\n", pT->do_8bitraw));
    CTRACE((tfp, "  use_raw_char_in:    %d\n", pT->use_raw_char_in));
    CTRACE((tfp, "  strip_raw_char_in:  %d\n", pT->strip_raw_char_in));
    CTRACE((tfp, "  pass_160_173_raw:   %d\n", pT->pass_160_173_raw));
    CTRACE((tfp, "  trans_to_uni:       %d\n", pT->trans_to_uni));
    CTRACE((tfp, "  trans_C0_to_uni:    %d\n", pT->trans_C0_to_uni));
    CTRACE((tfp, "  repl_translated_C0: %d\n", pT->repl_translated_C0));
    CTRACE((tfp, "  trans_from_uni:     %d\n", pT->trans_from_uni));
}

/*
 *  This function initializes the transformation
 *  structure by setting all its elements to
 *  FALSE. - KW
 */
void UCTransParams_clear(UCTransParams * pT)
{
    pT->transp = FALSE;
    pT->do_cjk = FALSE;
    pT->decode_utf8 = FALSE;
    pT->output_utf8 = FALSE;
    pT->do_8bitraw = FALSE;
    pT->use_raw_char_in = FALSE;
    pT->strip_raw_char_in = FALSE;
    pT->pass_160_173_raw = FALSE;
    pT->trans_to_uni = FALSE;
    pT->trans_C0_to_uni = FALSE;
    pT->repl_translated_C0 = FALSE;
    pT->trans_from_uni = FALSE;
}

/*
 * If terminal is in UTF-8 mode, it probably cannot understand box drawing
 * chars as the 8-bit (n)curses handles them.  (This may also be true for other
 * display character sets, but isn't currently checked.) In that case set the
 * chars for horizontal and vertical drawing chars to displayable ASCII chars
 * if '0' was requested.  They'll stay as they are otherwise.  -KW, TD
 *
 * If we're able to obtain a character set based on the locale settings,
 * assume that the user has setup $TERM and the fonts already so line-drawing
 * works.
 */
void UCSetBoxChars(int cset,
		   int *pvert_out,
		   int *phori_out,
		   int vert_in,
		   int hori_in)
{
    BOOL fix_lines = FALSE;

    if (cset >= 0) {
#ifndef WIDEC_CURSES
	if (LYCharSet_UC[cset].enc == UCT_ENC_UTF8) {
	    fix_lines = TRUE;
	}
#endif
	/*
	 * If we've identified a charset that works, require it.
	 * This is important if we have loaded a font, which would
	 * confuse curses.
	 */
	/* US-ASCII vs Latin-1 is safe (usually) */
	if ((cset == US_ASCII
	     || cset == LATIN1)
	    && (linedrawing_char_set == US_ASCII
		|| linedrawing_char_set == LATIN1)) {
#if (defined(FANCY_CURSES) && defined(A_ALTCHARSET)) || defined(USE_SLANG)
	    vert_in = 0;
	    hori_in = 0;
#else
	    ;
#endif
	}
#ifdef EXP_CHARTRANS_AUTOSWITCH
#if defined(NCURSES_VERSION) || defined(HAVE_TIGETSTR)
	else {
	    static BOOL first = TRUE;
	    static int last_cset = -99;
	    static BOOL last_result = TRUE;
	    /* *INDENT-OFF* */
	    static struct {
		int mapping;
		UCode_t internal;
		int external;
	    } table[] = {
		{ 'j', 0x2518, 0 }, /* BOX DRAWINGS LIGHT UP AND LEFT */
		{ 'k', 0x2510, 0 }, /* BOX DRAWINGS LIGHT DOWN AND LEFT */
		{ 'l', 0x250c, 0 }, /* BOX DRAWINGS LIGHT DOWN AND RIGHT */
		{ 'm', 0x2514, 0 }, /* BOX DRAWINGS LIGHT UP AND RIGHT */
		{ 'n', 0x253c, 0 }, /* BOX DRAWINGS LIGHT VERTICAL AND HORIZONTAL */
		{ 'q', 0x2500, 0 }, /* BOX DRAWINGS LIGHT HORIZONTAL */
		{ 't', 0x251c, 0 }, /* BOX DRAWINGS LIGHT VERTICAL AND RIGHT */
		{ 'u', 0x2524, 0 }, /* BOX DRAWINGS LIGHT VERTICAL AND LEFT */
		{ 'v', 0x2534, 0 }, /* BOX DRAWINGS LIGHT UP AND HORIZONTAL */
		{ 'w', 0x252c, 0 }, /* BOX DRAWINGS LIGHT DOWN AND HORIZONTAL */
		{ 'x', 0x2502, 0 }, /* BOX DRAWINGS LIGHT VERTICAL */
	    };
	    /* *INDENT-ON* */

	    unsigned n;

	    if (first) {
		static char acsc_name[] = "acsc";
		char *map = tigetstr(acsc_name);

		if (map != 0) {
		    CTRACE((tfp, "build terminal line-drawing map\n"));
		    while (map[0] != 0 && map[1] != 0) {
			for (n = 0; n < TABLESIZE(table); ++n) {
			    if (table[n].mapping == map[0]) {
				table[n].external = UCH(map[1]);
				CTRACE((tfp,
					"  map[%c] %#" PRI_UCode_t " -> %#x\n",
					table[n].mapping,
					CAST_UCode_t (table[n].internal),
					(unsigned)table[n].external));
				break;
			    }
			}
			map += 2;
		    }
		}
		first = FALSE;
	    }

	    if (cset == last_cset) {
		fix_lines = last_result;
	    } else if (cset == UTF8_handle) {
		last_result = FALSE;
		last_cset = cset;
	    } else {
		CTRACE((tfp, "check terminal line-drawing map\n"));
		for (n = 0; n < TABLESIZE(table); ++n) {
		    int test = UCTransUniChar(table[n].internal, cset);

		    if (test != table[n].external) {
			CTRACE((tfp,
				"line-drawing map %c mismatch (have %#x, want %#x)\n",
				table[n].mapping,
				(unsigned) test,
				(unsigned) table[n].external));
			fix_lines = TRUE;
			break;
		    }
		}
		last_result = fix_lines;
		last_cset = cset;
	    }
	}
#else
	else if (cset != linedrawing_char_set && linedrawing_char_set >= 0) {
	    fix_lines = TRUE;
	}
#endif
#endif
    }
    if (fix_lines) {
	if (!vert_in)
	    vert_in = '|';
	if (!hori_in)
	    hori_in = '-';
    }
    *pvert_out = vert_in;
    *phori_out = hori_in;
}

/*
 *  Given an output target HTStream* (can also be a HTStructured* via
 *  typecast), the target stream's put_character method, and a Unicode
 *  character,  CPutUtf8_charstring() will either output the UTF8
 *  encoding of the Unicode and return YES, or do nothing and return
 *  NO (if conversion would be unnecessary or the Unicode character is
 *  considered invalid).
 *
 *  [Could be used more generally, but is currently only used for &#nnnnn
 *  stuff - generation of UTF8 from 8-bit encoded charsets not yet done
 *  by SGML.c etc.]
 */
#define PUTC(ch) ((*myPutc)(target, (char)(ch)))
#define PUTC2(ch) ((*myPutc)(target,(char)(0x80|(0x3f &(ch)))))

BOOL UCPutUtf8_charstring(HTStream *target, putc_func_t *myPutc, UCode_t code)
{
    if (code < 128)
	return NO;		/* indicate to caller we didn't handle it */
    else if (code < 0x800L) {
	PUTC(0xc0 | (code >> 6));
	PUTC2(code);
    } else if (code < 0x10000L) {
	PUTC(0xe0 | (code >> 12));
	PUTC2(code >> 6);
	PUTC2(code);
    } else if (code < 0x200000L) {
	PUTC(0xf0 | (code >> 18));
	PUTC2(code >> 12);
	PUTC2(code >> 6);
	PUTC2(code);
    } else if (code < 0x4000000L) {
	PUTC(0xf8 | (code >> 24));
	PUTC2(code >> 18);
	PUTC2(code >> 12);
	PUTC2(code >> 6);
	PUTC2(code);
    } else if (code <= 0x7fffffffL) {
	PUTC(0xfc | (code >> 30));
	PUTC2(code >> 24);
	PUTC2(code >> 18);
	PUTC2(code >> 12);
	PUTC2(code >> 6);
	PUTC2(code);
    } else
	return NO;
    return YES;
}

/*
 *  This function converts a Unicode (UCode_t) value
 *  to a multibyte UTF-8 character, which is loaded
 *  into the buffer received as an argument.  The
 *  buffer should be large enough to hold at least
 *  seven characters (but should be declared as 8
 *  to minimize byte alignment problems with some
 *  compilers). - FM
 */
BOOL UCConvertUniToUtf8(UCode_t code, char *buffer)
{
    char *ch = buffer;

    if (!ch)
	return NO;

    if (code <= 0 || code > 0x7fffffffL) {
	*ch = '\0';
	return NO;
    }

    if (code < 0x800L) {
	*ch++ = (char) (0xc0 | (code >> 6));
	*ch++ = (char) (0x80 | (0x3f & (code)));
	*ch = '\0';
    } else if (code < 0x10000L) {
	*ch++ = (char) (0xe0 | (code >> 12));
	*ch++ = (char) (0x80 | (0x3f & (code >> 6)));
	*ch++ = (char) (0x80 | (0x3f & (code)));
	*ch = '\0';
    } else if (code < 0x200000L) {
	*ch++ = (char) (0xf0 | (code >> 18));
	*ch++ = (char) (0x80 | (0x3f & (code >> 12)));
	*ch++ = (char) (0x80 | (0x3f & (code >> 6)));
	*ch++ = (char) (0x80 | (0x3f & (code)));
	*ch = '\0';
    } else if (code < 0x4000000L) {
	*ch++ = (char) (0xf8 | (code >> 24));
	*ch++ = (char) (0x80 | (0x3f & (code >> 18)));
	*ch++ = (char) (0x80 | (0x3f & (code >> 12)));
	*ch++ = (char) (0x80 | (0x3f & (code >> 6)));
	*ch++ = (char) (0x80 | (0x3f & (code)));
	*ch = '\0';
    } else {
	*ch++ = (char) (0xfc | (code >> 30));
	*ch++ = (char) (0x80 | (0x3f & (code >> 24)));
	*ch++ = (char) (0x80 | (0x3f & (code >> 18)));
	*ch++ = (char) (0x80 | (0x3f & (code >> 12)));
	*ch++ = (char) (0x80 | (0x3f & (code >> 6)));
	*ch++ = (char) (0x80 | (0x3f & (code)));
	*ch = '\0';
    }
    return YES;
}

/*
 * Get UCS character code for one character from UTF-8 encoded string.
 *
 * On entry:
 *	*ppuni should point to beginning of UTF-8 encoding character
 * On exit:
 *	*ppuni is advanced to point to the last byte of UTF-8 sequence,
 *		if there was a valid one; otherwise unchanged.
 * returns the UCS value
 * returns negative value on error (invalid UTF-8 sequence)
 */
UCode_t UCGetUniFromUtf8String(const char **ppuni)
{
    UCode_t uc_out = 0;
    const char *p = *ppuni;
    int utf_count, i;

    if (!(**ppuni & 0x80))
	return (UCode_t) **ppuni;	/* ASCII range character */
    else if (!(**ppuni & 0x40))
	return (-1);		/* not a valid UTF-8 start */
    if ((*p & 0xe0) == 0xc0) {
	utf_count = 1;
    } else if ((*p & 0xf0) == 0xe0) {
	utf_count = 2;
    } else if ((*p & 0xf8) == 0xf0) {
	utf_count = 3;
    } else if ((*p & 0xfc) == 0xf8) {
	utf_count = 4;
    } else if ((*p & 0xfe) == 0xfc) {
	utf_count = 5;
    } else {			/* garbage */
	return (-1);
    }
    for (p = *ppuni, i = 0; i < utf_count; i++) {
	if ((*(++p) & 0xc0) != 0x80)
	    return (-1);
    }
    p = *ppuni;
    switch (utf_count) {
    case 1:
	uc_out = (((*p & 0x1f) << 6) |
		  (*(p + 1) & 0x3f));
	break;
    case 2:
	uc_out = (((((*p & 0x0f) << 6) |
		    (*(p + 1) & 0x3f)) << 6) |
		  (*(p + 2) & 0x3f));
	break;
    case 3:
	uc_out = (((((((*p & 0x07) << 6) |
		      (*(p + 1) & 0x3f)) << 6) |
		    (*(p + 2) & 0x3f)) << 6) |
		  (*(p + 3) & 0x3f));
	break;
    case 4:
	uc_out = (((((((((*p & 0x03) << 6) |
			(*(p + 1) & 0x3f)) << 6) |
		      (*(p + 2) & 0x3f)) << 6) |
		    (*(p + 3) & 0x3f)) << 6) |
		  (*(p + 4) & 0x3f));
	break;
    case 5:
	uc_out = (((((((((((*p & 0x01) << 6) |
			  (*(p + 1) & 0x3f)) << 6) |
			(*(p + 2) & 0x3f)) << 6) |
		      (*(p + 3) & 0x3f)) << 6) |
		    (*(p + 4) & 0x3f)) << 6) |
		  (*(p + 5) & 0x3f));
	break;
    }
    *ppuni = p + utf_count;
    return uc_out;
}

/*
 * Combine UTF-8 into Unicode.  Incomplete characters are either ignored, or
 * returned as the UCS replacement character.
 */
dUTF8 HTDecodeUTF8(UTFDecodeState * me, int *c_in_out, UCode_t *result)
{
    dUTF8 rc = dUTF8_ok;
    int c = *c_in_out;
    unsigned uc = UCH(c);

    if (TOASCII(uc) > 127) {
	/*
	 * continue a multibyte character...
	 */
	if (me->utf_count > 0 && (TOASCII(c) & 0xc0) == 0x80) {
	    if (me->utf_count == 1) {
		int limit = (int) (me->utf_buf_p - me->utf_buf) + 1;
		int maybe = 0;

		/*
		 * Check for overlong sequences (from comment in xterm):
		 *   1100000x 10xxxxxx
		 *   11100000 100xxxxx 10xxxxxx
		 *   11110000 1000xxxx 10xxxxxx 10xxxxxx
		 *   11111000 10000xxx 10xxxxxx 10xxxxxx 10xxxxxx
		 *   11111100 100000xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
		 */
		switch (limit) {
		case 2:
		    maybe = (UCH(me->utf_buf[0]) & 0xfe) == 0xc0;
		    break;
		case 3:
		    maybe = ((UCH(me->utf_buf[0]) == 0xe0) &&
			     (UCH(me->utf_buf[1]) & 0xf0) == 0x80);
		    break;
		case 4:
		    maybe = ((UCH(me->utf_buf[0]) == 0xf0) &&
			     (UCH(me->utf_buf[1]) & 0xf8) == 0x80);
		    break;
		case 5:
		    maybe = ((UCH(me->utf_buf[0]) == 0xf8) &&
			     (UCH(me->utf_buf[1]) & 0xfd) == 0x80);
		    break;
		}
		if (maybe) {
		    while (limit-- > 2) {
			if ((UCH(me->utf_buf[limit]) & 0xc0) != 0x80) {
			    maybe = 0;
			    break;
			}
		    }
		    if (maybe) {
			me->utf_char = UCS_REPL;
		    }
		}
	    }
	    if (me->utf_char == UCS_REPL) {
		rc = dUTF8_err;
	    } else if (me->utf_char || ((uc & 0x7f) >> (7 - me->utf_count))) {
		me->utf_char = (me->utf_char << 6) | (TOASCII(c) & 0x3f);
		if ((me->utf_char >= 0xd800 &&
		     me->utf_char <= 0xdfff) ||
		    (me->utf_char == 0xfffe) ||
		    (me->utf_char == UCS_HIDE)) {
		    me->utf_char = UCS_REPL;
		    rc = dUTF8_err;
		}
	    } else {
		me->utf_char = UCS_REPL;
		rc = dUTF8_err;
	    }
	    me->utf_count--;
	    *(me->utf_buf_p) = (char) c;
	    (me->utf_buf_p)++;

	    if (me->utf_count == 0) {
		*(me->utf_buf_p) = '\0';
		*result = me->utf_char;
		if (*result < 256) {
		    *c_in_out = UCH(*result & 0xff);
		}
		switch (*result) {
		case 0x200e:	/* left-to-right mark */
		case 0x200f:	/* right-to-left mark */
		    /* lynx does not use these */
		    *result = '\0';
		    break;
		}
	    } else {
		rc = dUTF8_more;
	    }
	} else {
	    /*
	     * begin a multibyte character
	     */
	    rc = dUTF8_more;
	    me->utf_buf_p = me->utf_buf;
	    *(me->utf_buf_p) = (char) c;
	    (me->utf_buf_p)++;
	    if ((uc & 0xe0) == 0xc0) {
		me->utf_count = 1;
		me->utf_char = (uc & 0x1f);
	    } else if ((uc & 0xf0) == 0xe0) {
		me->utf_count = 2;
		me->utf_char = (uc & 0x0f);
	    } else if ((uc & 0xf8) == 0xf0) {
		me->utf_count = 3;
		me->utf_char = (uc & 0x07);
	    } else if ((uc & 0xfc) == 0xf8) {
		me->utf_count = 4;
		me->utf_char = (uc & 0x03);
	    } else if ((uc & 0xfe) == 0xfc) {
		me->utf_count = 5;
		me->utf_char = (uc & 0x01);
	    } else {
		me->utf_count = 0;
		me->utf_buf_p = me->utf_buf;
		*(me->utf_buf_p) = '\0';
		rc = dUTF8_err;
	    }
	}
    } else {
	me->utf_count = 0;
	me->utf_buf_p = me->utf_buf;
	*(me->utf_buf_p) = '\0';
    }

#if 0
    if (rc != dUTF8_ok) {
	CTRACE((tfp, "UTF8 %#x ->%#x %s\n",
		uc, UCH(*c_in_out),
		(rc == dUTF8_err) ? "err" : "more"));
    } else {
	if (*result > 127) {
	    CTRACE((tfp, "UTF8 %#x == %#x\n", uc, (int) *result));
	} else if (c != UCS_REPL && !isspace(c)) {
	    CTRACE((tfp, "CHAR %#x == %c (%#x)\n", uc, uc, (int) *result));
	}
    }
#endif
    return rc;
}