summaryrefslogtreecommitdiffstats
path: root/term.c
blob: b0f6fb0df0d7a95a84710e6fd860bfe639379ba4 (plain) (blame)
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
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
/*	$Id$ */
/*
 * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
 * Copyright (c) 2010-2018 Ingo Schwarze <schwarze@openbsd.org>
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES
 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 */
#include "config.h"

#include <sys/types.h>

#include <assert.h>
#include <ctype.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "mandoc.h"
#include "mandoc_aux.h"
#include "out.h"
#include "term.h"
#include "main.h"

static	size_t		 cond_width(const struct termp *, int, int *);
static	void		 adjbuf(struct termp_col *, size_t);
static	void		 bufferc(struct termp *, char);
static	void		 encode(struct termp *, const char *, size_t);
static	void		 encode1(struct termp *, int);
static	void		 endline(struct termp *);


void
term_setcol(struct termp *p, size_t maxtcol)
{
	if (maxtcol > p->maxtcol) {
		p->tcols = mandoc_recallocarray(p->tcols,
		    p->maxtcol, maxtcol, sizeof(*p->tcols));
		p->maxtcol = maxtcol;
	}
	p->lasttcol = maxtcol - 1;
	p->tcol = p->tcols;
}

void
term_free(struct termp *p)
{
	for (p->tcol = p->tcols; p->tcol < p->tcols + p->maxtcol; p->tcol++)
		free(p->tcol->buf);
	free(p->tcols);
	free(p->fontq);
	free(p);
}

void
term_begin(struct termp *p, term_margin head,
		term_margin foot, const struct roff_meta *arg)
{

	p->headf = head;
	p->footf = foot;
	p->argf = arg;
	(*p->begin)(p);
}

void
term_end(struct termp *p)
{

	(*p->end)(p);
}

/*
 * Flush a chunk of text.  By default, break the output line each time
 * the right margin is reached, and continue output on the next line
 * at the same offset as the chunk itself.  By default, also break the
 * output line at the end of the chunk.
 * The following flags may be specified:
 *
 *  - TERMP_NOBREAK: Do not break the output line at the right margin,
 *    but only at the max right margin.  Also, do not break the output
 *    line at the end of the chunk, such that the next call can pad to
 *    the next column.  However, if less than p->trailspace blanks,
 *    which can be 0, 1, or 2, remain to the right margin, the line
 *    will be broken.
 *  - TERMP_BRTRSP: Consider trailing whitespace significant
 *    when deciding whether the chunk fits or not.
 *  - TERMP_BRIND: If the chunk does not fit and the output line has
 *    to be broken, start the next line at the right margin instead
 *    of at the offset.  Used together with TERMP_NOBREAK for the tags
 *    in various kinds of tagged lists.
 *  - TERMP_HANG: Do not break the output line at the right margin,
 *    append the next chunk after it even if this one is too long.
 *    To be used together with TERMP_NOBREAK.
 *  - TERMP_NOPAD: Start writing at the current position,
 *    do not pad with blank characters up to the offset.
 */
void
term_flushln(struct termp *p)
{
	size_t		 vis;   /* current visual position on output */
	size_t		 vbl;   /* number of blanks to prepend to output */
	size_t		 vend;	/* end of word visual position on output */
	size_t		 bp;    /* visual right border position */
	size_t		 dv;    /* temporary for visual pos calculations */
	size_t		 j;     /* temporary loop index for p->tcol->buf */
	size_t		 jhy;	/* last hyph before overflow w/r/t j */
	size_t		 maxvis; /* output position of visible boundary */
	int		 ntab;	/* number of tabs to prepend */
	int		 breakline; /* after this word */

	vbl = (p->flags & TERMP_NOPAD) || p->tcol->offset < p->viscol ?
	    0 : p->tcol->offset - p->viscol;
	if (p->minbl && vbl < p->minbl)
		vbl = p->minbl;
	maxvis = p->tcol->rmargin > p->viscol + vbl ?
	    p->tcol->rmargin - p->viscol - vbl : 0;
	bp = !(p->flags & TERMP_NOBREAK) ? maxvis :
	    p->maxrmargin > p->viscol + vbl ?
	    p->maxrmargin - p->viscol - vbl : 0;
	vis = vend = 0;

	if ((p->flags & TERMP_MULTICOL) == 0)
		p->tcol->col = 0;
	while (p->tcol->col < p->tcol->lastcol) {

		/*
		 * Handle literal tab characters: collapse all
		 * subsequent tabs into a single huge set of spaces.
		 */

		ntab = 0;
		while (p->tcol->col < p->tcol->lastcol &&
		    p->tcol->buf[p->tcol->col] == '\t') {
			vend = term_tab_next(vis);
			vbl += vend - vis;
			vis = vend;
			ntab++;
			p->tcol->col++;
		}

		/*
		 * Count up visible word characters.  Control sequences
		 * (starting with the CSI) aren't counted.  A space
		 * generates a non-printing word, which is valid (the
		 * space is printed according to regular spacing rules).
		 */

		jhy = 0;
		breakline = 0;
		for (j = p->tcol->col; j < p->tcol->lastcol; j++) {
			if (p->tcol->buf[j] == '\n') {
				if ((p->flags & TERMP_BRIND) == 0)
					breakline = 1;
				continue;
			}
			if (p->tcol->buf[j] == ' ' || p->tcol->buf[j] == '\t')
				break;

			/* Back over the last printed character. */
			if (p->tcol->buf[j] == '\b') {
				assert(j);
				vend -= (*p->width)(p, p->tcol->buf[j - 1]);
				continue;
			}

			/* Regular word. */
			/* Break at the hyphen point if we overrun. */
			if (vend > vis && vend < bp &&
			    (p->tcol->buf[j] == ASCII_HYPH||
			     p->tcol->buf[j] == ASCII_BREAK))
				jhy = j;

			/*
			 * Hyphenation now decided, put back a real
			 * hyphen such that we get the correct width.
			 */
			if (p->tcol->buf[j] == ASCII_HYPH)
				p->tcol->buf[j] = '-';

			vend += (*p->width)(p, p->tcol->buf[j]);
		}

		/*
		 * Find out whether we would exceed the right margin.
		 * If so, break to the next line.
		 */

		if (vend > bp && jhy == 0 && vis > 0 &&
		    (p->flags & TERMP_BRNEVER) == 0) {
			if (p->flags & TERMP_MULTICOL)
				return;

			endline(p);
			vend -= vis;

			/* Use pending tabs on the new line. */

			vbl = 0;
			while (ntab--)
				vbl = term_tab_next(vbl);

			/* Re-establish indentation. */

			if (p->flags & TERMP_BRIND)
				vbl += p->tcol->rmargin;
			else
				vbl += p->tcol->offset;
			maxvis = p->tcol->rmargin > vbl ?
			    p->tcol->rmargin - vbl : 0;
			bp = !(p->flags & TERMP_NOBREAK) ? maxvis :
			    p->maxrmargin > vbl ?  p->maxrmargin - vbl : 0;
		}

		/*
		 * Write out the rest of the word.
		 */

		for ( ; p->tcol->col < p->tcol->lastcol; p->tcol->col++) {
			if (vend > bp && jhy > 0 && p->tcol->col > jhy)
				break;
			if (p->tcol->buf[p->tcol->col] == '\n')
				continue;
			if (p->tcol->buf[p->tcol->col] == '\t')
				break;
			if (p->tcol->buf[p->tcol->col] == ' ') {
				j = p->tcol->col;
				while (p->tcol->col < p->tcol->lastcol &&
				    p->tcol->buf[p->tcol->col] == ' ')
					p->tcol->col++;
				dv = (p->tcol->col - j) * (*p->width)(p, ' ');
				vbl += dv;
				vend += dv;
				break;
			}
			if (p->tcol->buf[p->tcol->col] == ASCII_NBRSP) {
				vbl += (*p->width)(p, ' ');
				continue;
			}
			if (p->tcol->buf[p->tcol->col] == ASCII_BREAK)
				continue;

			/*
			 * Now we definitely know there will be
			 * printable characters to output,
			 * so write preceding white space now.
			 */
			if (vbl) {
				(*p->advance)(p, vbl);
				p->viscol += vbl;
				vbl = 0;
			}

			(*p->letter)(p, p->tcol->buf[p->tcol->col]);
			if (p->tcol->buf[p->tcol->col] == '\b')
				p->viscol -= (*p->width)(p,
				    p->tcol->buf[p->tcol->col - 1]);
			else
				p->viscol += (*p->width)(p,
				    p->tcol->buf[p->tcol->col]);
		}
		vis = vend;

		if (breakline == 0)
			continue;

		/* Explicitly requested output line break. */

		if (p->flags & TERMP_MULTICOL)
			return;

		endline(p);
		breakline = 0;
		vis = vend = 0;

		/* Re-establish indentation. */

		vbl = p->tcol->offset;
		maxvis = p->tcol->rmargin > vbl ?
		    p->tcol->rmargin - vbl : 0;
		bp = !(p->flags & TERMP_NOBREAK) ? maxvis :
		    p->maxrmargin > vbl ?  p->maxrmargin - vbl : 0;
	}

	/*
	 * If there was trailing white space, it was not printed;
	 * so reset the cursor position accordingly.
	 */

	if (vis > vbl)
		vis -= vbl;
	else
		vis = 0;

	p->col = p->tcol->col = p->tcol->lastcol = 0;
	p->minbl = p->trailspace;
	p->flags &= ~(TERMP_BACKAFTER | TERMP_BACKBEFORE | TERMP_NOPAD);

	if (p->flags & TERMP_MULTICOL)
		return;

	/* Trailing whitespace is significant in some columns. */

	if (vis && vbl && (TERMP_BRTRSP & p->flags))
		vis += vbl;

	/* If the column was overrun, break the line. */
	if ((p->flags & TERMP_NOBREAK) == 0 ||
	    ((p->flags & TERMP_HANG) == 0 &&
	     vis + p->trailspace * (*p->width)(p, ' ') > maxvis))
		endline(p);
}

static void
endline(struct termp *p)
{
	if ((p->flags & (TERMP_NEWMC | TERMP_ENDMC)) == TERMP_ENDMC) {
		p->mc = NULL;
		p->flags &= ~TERMP_ENDMC;
	}
	if (p->mc != NULL) {
		if (p->viscol && p->maxrmargin >= p->viscol)
			(*p->advance)(p, p->maxrmargin - p->viscol + 1);
		p->flags |= TERMP_NOBUF | TERMP_NOSPACE;
		term_word(p, p->mc);
		p->flags &= ~(TERMP_NOBUF | TERMP_NEWMC);
	}
	p->viscol = 0;
	p->minbl = 0;
	(*p->endline)(p);
}

/*
 * A newline only breaks an existing line; it won't assert vertical
 * space.  All data in the output buffer is flushed prior to the newline
 * assertion.
 */
void
term_newln(struct termp *p)
{

	p->flags |= TERMP_NOSPACE;
	if (p->tcol->lastcol || p->viscol)
		term_flushln(p);
}

/*
 * Asserts a vertical space (a full, empty line-break between lines).
 * Note that if used twice, this will cause two blank spaces and so on.
 * All data in the output buffer is flushed prior to the newline
 * assertion.
 */
void
term_vspace(struct termp *p)
{

	term_newln(p);
	p->viscol = 0;
	p->minbl = 0;
	if (0 < p->skipvsp)
		p->skipvsp--;
	else
		(*p->endline)(p);
}

/* Swap current and previous font; for \fP and .ft P */
void
term_fontlast(struct termp *p)
{
	enum termfont	 f;

	f = p->fontl;
	p->fontl = p->fontq[p->fonti];
	p->fontq[p->fonti] = f;
}

/* Set font, save current, discard previous; for \f, .ft, .B etc. */
void
term_fontrepl(struct termp *p, enum termfont f)
{

	p->fontl = p->fontq[p->fonti];
	p->fontq[p->fonti] = f;
}

/* Set font, save previous. */
void
term_fontpush(struct termp *p, enum termfont f)
{

	p->fontl = p->fontq[p->fonti];
	if (++p->fonti == p->fontsz) {
		p->fontsz += 8;
		p->fontq = mandoc_reallocarray(p->fontq,
		    p->fontsz, sizeof(*p->fontq));
	}
	p->fontq[p->fonti] = f;
}

/* Flush to make the saved pointer current again. */
void
term_fontpopq(struct termp *p, int i)
{

	assert(i >= 0);
	if (p->fonti > i)
		p->fonti = i;
}

/* Pop one font off the stack. */
void
term_fontpop(struct termp *p)
{

	assert(p->fonti);
	p->fonti--;
}

/*
 * Handle pwords, partial words, which may be either a single word or a
 * phrase that cannot be broken down (such as a literal string).  This
 * handles word styling.
 */
void
term_word(struct termp *p, const char *word)
{
	struct roffsu	 su;
	const char	 nbrsp[2] = { ASCII_NBRSP, 0 };
	const char	*seq, *cp;
	int		 sz, uc;
	size_t		 csz, lsz, ssz;
	enum mandoc_esc	 esc;

	if ((p->flags & TERMP_NOBUF) == 0) {
		if ((p->flags & TERMP_NOSPACE) == 0) {
			if ((p->flags & TERMP_KEEP) == 0) {
				bufferc(p, ' ');
				if (p->flags & TERMP_SENTENCE)
					bufferc(p, ' ');
			} else
				bufferc(p, ASCII_NBRSP);
		}
		if (p->flags & TERMP_PREKEEP)
			p->flags |= TERMP_KEEP;
		if (p->flags & TERMP_NONOSPACE)
			p->flags |= TERMP_NOSPACE;
		else
			p->flags &= ~TERMP_NOSPACE;
		p->flags &= ~(TERMP_SENTENCE | TERMP_NONEWLINE);
		p->skipvsp = 0;
	}

	while ('\0' != *word) {
		if ('\\' != *word) {
			if (TERMP_NBRWORD & p->flags) {
				if (' ' == *word) {
					encode(p, nbrsp, 1);
					word++;
					continue;
				}
				ssz = strcspn(word, "\\ ");
			} else
				ssz = strcspn(word, "\\");
			encode(p, word, ssz);
			word += (int)ssz;
			continue;
		}

		word++;
		esc = mandoc_escape(&word, &seq, &sz);
		switch (esc) {
		case ESCAPE_UNICODE:
			uc = mchars_num2uc(seq + 1, sz - 1);
			break;
		case ESCAPE_NUMBERED:
			uc = mchars_num2char(seq, sz);
			if (uc < 0)
				continue;
			break;
		case ESCAPE_SPECIAL:
			if (p->enc == TERMENC_ASCII) {
				cp = mchars_spec2str(seq, sz, &ssz);
				if (cp != NULL)
					encode(p, cp, ssz);
			} else {
				uc = mchars_spec2cp(seq, sz);
				if (uc > 0)
					encode1(p, uc);
			}
			continue;
		case ESCAPE_UNDEF:
			uc = *seq;
			break;
		case ESCAPE_FONTBOLD:
			term_fontrepl(p, TERMFONT_BOLD);
			continue;
		case ESCAPE_FONTITALIC:
			term_fontrepl(p, TERMFONT_UNDER);
			continue;
		case ESCAPE_FONTBI:
			term_fontrepl(p, TERMFONT_BI);
			continue;
		case ESCAPE_FONT:
		case ESCAPE_FONTCW:
		case ESCAPE_FONTROMAN:
			term_fontrepl(p, TERMFONT_NONE);
			continue;
		case ESCAPE_FONTPREV:
			term_fontlast(p);
			continue;
		case ESCAPE_BREAK:
			bufferc(p, '\n');
			continue;
		case ESCAPE_NOSPACE:
			if (p->flags & TERMP_BACKAFTER)
				p->flags &= ~TERMP_BACKAFTER;
			else if (*word == '\0')
				p->flags |= (TERMP_NOSPACE | TERMP_NONEWLINE);
			continue;
		case ESCAPE_DEVICE:
			if (p->type == TERMTYPE_PDF)
				encode(p, "pdf", 3);
			else if (p->type == TERMTYPE_PS)
				encode(p, "ps", 2);
			else if (p->enc == TERMENC_ASCII)
				encode(p, "ascii", 5);
			else
				encode(p, "utf8", 4);
			continue;
		case ESCAPE_HORIZ:
			if (*seq == '|') {
				seq++;
				uc = -p->col;
			} else
				uc = 0;
			if (a2roffsu(seq, &su, SCALE_EM) == NULL)
				continue;
			uc += term_hen(p, &su);
			if (uc > 0)
				while (uc-- > 0)
					bufferc(p, ASCII_NBRSP);
			else if (p->col > (size_t)(-uc))
				p->col += uc;
			else {
				uc += p->col;
				p->col = 0;
				if (p->tcol->offset > (size_t)(-uc)) {
					p->ti += uc;
					p->tcol->offset += uc;
				} else {
					p->ti -= p->tcol->offset;
					p->tcol->offset = 0;
				}
			}
			continue;
		case ESCAPE_HLINE:
			if ((cp = a2roffsu(seq, &su, SCALE_EM)) == NULL)
				continue;
			uc = term_hen(p, &su);
			if (uc <= 0) {
				if (p->tcol->rmargin <= p->tcol->offset)
					continue;
				lsz = p->tcol->rmargin - p->tcol->offset;
			} else
				lsz = uc;
			if (*cp == seq[-1])
				uc = -1;
			else if (*cp == '\\') {
				seq = cp + 1;
				esc = mandoc_escape(&seq, &cp, &sz);
				switch (esc) {
				case ESCAPE_UNICODE:
					uc = mchars_num2uc(cp + 1, sz - 1);
					break;
				case ESCAPE_NUMBERED:
					uc = mchars_num2char(cp, sz);
					break;
				case ESCAPE_SPECIAL:
					uc = mchars_spec2cp(cp, sz);
					break;
				case ESCAPE_UNDEF:
					uc = *seq;
					break;
				default:
					uc = -1;
					break;
				}
			} else
				uc = *cp;
			if (uc < 0x20 || (uc > 0x7E && uc < 0xA0))
				uc = '_';
			if (p->enc == TERMENC_ASCII) {
				cp = ascii_uc2str(uc);
				csz = term_strlen(p, cp);
				ssz = strlen(cp);
			} else
				csz = (*p->width)(p, uc);
			while (lsz >= csz) {
				if (p->enc == TERMENC_ASCII)
					encode(p, cp, ssz);
				else
					encode1(p, uc);
				lsz -= csz;
			}
			continue;
		case ESCAPE_SKIPCHAR:
			p->flags |= TERMP_BACKAFTER;
			continue;
		case ESCAPE_OVERSTRIKE:
			cp = seq + sz;
			while (seq < cp) {
				if (*seq == '\\') {
					mandoc_escape(&seq, NULL, NULL);
					continue;
				}
				encode1(p, *seq++);
				if (seq < cp) {
					if (p->flags & TERMP_BACKBEFORE)
						p->flags |= TERMP_BACKAFTER;
					else
						p->flags |= TERMP_BACKBEFORE;
				}
			}
			/* Trim trailing backspace/blank pair. */
			if (p->tcol->lastcol > 2 &&
			    (p->tcol->buf[p->tcol->lastcol - 1] == ' ' ||
			     p->tcol->buf[p->tcol->lastcol - 1] == '\t'))
				p->tcol->lastcol -= 2;
			if (p->col > p->tcol->lastcol)
				p->col = p->tcol->lastcol;
			continue;
		default:
			continue;
		}

		/*
		 * Common handling for Unicode and numbered
		 * character escape sequences.
		 */

		if (p->enc == TERMENC_ASCII) {
			cp = ascii_uc2str(uc);
			encode(p, cp, strlen(cp));
		} else {
			if ((uc < 0x20 && uc != 0x09) ||
			    (uc > 0x7E && uc < 0xA0))
				uc = 0xFFFD;
			encode1(p, uc);
		}
	}
	p->flags &= ~TERMP_NBRWORD;
}

static void
adjbuf(struct termp_col *c, size_t sz)
{
	if (c->maxcols == 0)
		c->maxcols = 1024;
	while (c->maxcols <= sz)
		c->maxcols <<= 2;
	c->buf = mandoc_reallocarray(c->buf, c->maxcols, sizeof(*c->buf));
}

static void
bufferc(struct termp *p, char c)
{
	if (p->flags & TERMP_NOBUF) {
		(*p->letter)(p, c);
		return;
	}
	if (p->col + 1 >= p->tcol->maxcols)
		adjbuf(p->tcol, p->col + 1);
	if (p->tcol->lastcol <= p->col || (c != ' ' && c != ASCII_NBRSP))
		p->tcol->buf[p->col] = c;
	if (p->tcol->lastcol < ++p->col)
		p->tcol->lastcol = p->col;
}

/*
 * See encode().
 * Do this for a single (probably unicode) value.
 * Does not check for non-decorated glyphs.
 */
static void
encode1(struct termp *p, int c)
{
	enum termfont	  f;

	if (p->flags & TERMP_NOBUF) {
		(*p->letter)(p, c);
		return;
	}

	if (p->col + 7 >= p->tcol->maxcols)
		adjbuf(p->tcol, p->col + 7);

	f = (c == ASCII_HYPH || c > 127 || isgraph(c)) ?
	    p->fontq[p->fonti] : TERMFONT_NONE;

	if (p->flags & TERMP_BACKBEFORE) {
		if (p->tcol->buf[p->col - 1] == ' ' ||
		    p->tcol->buf[p->col - 1] == '\t')
			p->col--;
		else
			p->tcol->buf[p->col++] = '\b';
		p->flags &= ~TERMP_BACKBEFORE;
	}
	if (f == TERMFONT_UNDER || f == TERMFONT_BI) {
		p->tcol->buf[p->col++] = '_';
		p->tcol->buf[p->col++] = '\b';
	}
	if (f == TERMFONT_BOLD || f == TERMFONT_BI) {
		if (c == ASCII_HYPH)
			p->tcol->buf[p->col++] = '-';
		else
			p->tcol->buf[p->col++] = c;
		p->tcol->buf[p->col++] = '\b';
	}
	if (p->tcol->lastcol <= p->col || (c != ' ' && c != ASCII_NBRSP))
		p->tcol->buf[p->col] = c;
	if (p->tcol->lastcol < ++p->col)
		p->tcol->lastcol = p->col;
	if (p->flags & TERMP_BACKAFTER) {
		p->flags |= TERMP_BACKBEFORE;
		p->flags &= ~TERMP_BACKAFTER;
	}
}

static void
encode(struct termp *p, const char *word, size_t sz)
{
	size_t		  i;

	if (p->flags & TERMP_NOBUF) {
		for (i = 0; i < sz; i++)
			(*p->letter)(p, word[i]);
		return;
	}

	if (p->col + 2 + (sz * 5) >= p->tcol->maxcols)
		adjbuf(p->tcol, p->col + 2 + (sz * 5));

	for (i = 0; i < sz; i++) {
		if (ASCII_HYPH == word[i] ||
		    isgraph((unsigned char)word[i]))
			encode1(p, word[i]);
		else {
			if (p->tcol->lastcol <= p->col ||
			    (word[i] != ' ' && word[i] != ASCII_NBRSP))
				p->tcol->buf[p->col] = word[i];
			p->col++;

			/*
			 * Postpone the effect of \z while handling
			 * an overstrike sequence from ascii_uc2str().
			 */

			if (word[i] == '\b' &&
			    (p->flags & TERMP_BACKBEFORE)) {
				p->flags &= ~TERMP_BACKBEFORE;
				p->flags |= TERMP_BACKAFTER;
			}
		}
	}
	if (p->tcol->lastcol < p->col)
		p->tcol->lastcol = p->col;
}

void
term_setwidth(struct termp *p, const char *wstr)
{
	struct roffsu	 su;
	int		 iop, width;

	iop = 0;
	width = 0;
	if (NULL != wstr) {
		switch (*wstr) {
		case '+':
			iop = 1;
			wstr++;
			break;
		case '-':
			iop = -1;
			wstr++;
			break;
		default:
			break;
		}
		if (a2roffsu(wstr, &su, SCALE_MAX) != NULL)
			width = term_hspan(p, &su);
		else
			iop = 0;
	}
	(*p->setwidth)(p, iop, width);
}

size_t
term_len(const struct termp *p, size_t sz)
{

	return (*p->width)(p, ' ') * sz;
}

static size_t
cond_width(const struct termp *p, int c, int *skip)
{

	if (*skip) {
		(*skip) = 0;
		return 0;
	} else
		return (*p->width)(p, c);
}

size_t
term_strlen(const struct termp *p, const char *cp)
{
	size_t		 sz, rsz, i;
	int		 ssz, skip, uc;
	const char	*seq, *rhs;
	enum mandoc_esc	 esc;
	static const char rej[] = { '\\', ASCII_NBRSP, ASCII_HYPH,
			ASCII_BREAK, '\0' };

	/*
	 * Account for escaped sequences within string length
	 * calculations.  This follows the logic in term_word() as we
	 * must calculate the width of produced strings.
	 */

	sz = 0;
	skip = 0;
	while ('\0' != *cp) {
		rsz = strcspn(cp, rej);
		for (i = 0; i < rsz; i++)
			sz += cond_width(p, *cp++, &skip);

		switch (*cp) {
		case '\\':
			cp++;
			rhs = NULL;
			esc = mandoc_escape(&cp, &seq, &ssz);
			switch (esc) {
			case ESCAPE_UNICODE:
				uc = mchars_num2uc(seq + 1, ssz - 1);
				break;
			case ESCAPE_NUMBERED:
				uc = mchars_num2char(seq, ssz);
				if (uc < 0)
					continue;
				break;
			case ESCAPE_SPECIAL:
				if (p->enc == TERMENC_ASCII) {
					rhs = mchars_spec2str(seq, ssz, &rsz);
					if (rhs != NULL)
						break;
				} else {
					uc = mchars_spec2cp(seq, ssz);
					if (uc > 0)
						sz += cond_width(p, uc, &skip);
				}
				continue;
			case ESCAPE_UNDEF:
				uc = *seq;
				break;
			case ESCAPE_DEVICE:
				if (p->type == TERMTYPE_PDF) {
					rhs = "pdf";
					rsz = 3;
				} else if (p->type == TERMTYPE_PS) {
					rhs = "ps";
					rsz = 2;
				} else if (p->enc == TERMENC_ASCII) {
					rhs = "ascii";
					rsz = 5;
				} else {
					rhs = "utf8";
					rsz = 4;
				}
				break;
			case ESCAPE_SKIPCHAR:
				skip = 1;
				continue;
			case ESCAPE_OVERSTRIKE:
				rsz = 0;
				rhs = seq + ssz;
				while (seq < rhs) {
					if (*seq == '\\') {
						mandoc_escape(&seq, NULL, NULL);
						continue;
					}
					i = (*p->width)(p, *seq++);
					if (rsz < i)
						rsz = i;
				}
				sz += rsz;
				continue;
			default:
				continue;
			}

			/*
			 * Common handling for Unicode and numbered
			 * character escape sequences.
			 */

			if (rhs == NULL) {
				if (p->enc == TERMENC_ASCII) {
					rhs = ascii_uc2str(uc);
					rsz = strlen(rhs);
				} else {
					if ((uc < 0x20 && uc != 0x09) ||
					    (uc > 0x7E && uc < 0xA0))
						uc = 0xFFFD;
					sz += cond_width(p, uc, &skip);
					continue;
				}
			}

			if (skip) {
				skip = 0;
				break;
			}

			/*
			 * Common handling for all escape sequences
			 * printing more than one character.
			 */

			for (i = 0; i < rsz; i++)
				sz += (*p->width)(p, *rhs++);
			break;
		case ASCII_NBRSP:
			sz += cond_width(p, ' ', &skip);
			cp++;
			break;
		case ASCII_HYPH:
			sz += cond_width(p, '-', &skip);
			cp++;
			break;
		default:
			break;
		}
	}

	return sz;
}

int
term_vspan(const struct termp *p, const struct roffsu *su)
{
	double		 r;
	int		 ri;

	switch (su->unit) {
	case SCALE_BU:
		r = su->scale / 40.0;
		break;
	case SCALE_CM:
		r = su->scale * 6.0 / 2.54;
		break;
	case SCALE_FS:
		r = su->scale * 65536.0 / 40.0;
		break;
	case SCALE_IN:
		r = su->scale * 6.0;
		break;
	case SCALE_MM:
		r = su->scale * 0.006;
		break;
	case SCALE_PC:
		r = su->scale;
		break;
	case SCALE_PT:
		r = su->scale / 12.0;
		break;
	case SCALE_EN:
	case SCALE_EM:
		r = su->scale * 0.6;
		break;
	case SCALE_VS:
		r = su->scale;
		break;
	default:
		abort();
	}
	ri = r > 0.0 ? r + 0.4995 : r - 0.4995;
	return ri < 66 ? ri : 1;
}

/*
 * Convert a scaling width to basic units, rounding towards 0.
 */
int
term_hspan(const struct termp *p, const struct roffsu *su)
{

	return (*p->hspan)(p, su);
}

/*
 * Convert a scaling width to basic units, rounding to closest.
 */
int
term_hen(const struct termp *p, const struct roffsu *su)
{
	int bu;

	if ((bu = (*p->hspan)(p, su)) >= 0)
		return (bu + 11) / 24;
	else
		return -((-bu + 11) / 24);
}