summaryrefslogtreecommitdiff
path: root/src/inlines.c
blob: 6f0adef42b3ca569a4a720def4f244001771ee6d (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
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
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
#include <stdlib.h>
#include <string.h>
#include <stdio.h>

#include "cmark_ctype.h"
#include "config.h"
#include "node.h"
#include "parser.h"
#include "references.h"
#include "cmark.h"
#include "houdini.h"
#include "utf8.h"
#include "scanners.h"
#include "inlines.h"

static const char *EMDASH = "\xE2\x80\x94";
static const char *ENDASH = "\xE2\x80\x93";
static const char *ELLIPSES = "\xE2\x80\xA6";
static const char *LEFTDOUBLEQUOTE = "\xE2\x80\x9C";
static const char *RIGHTDOUBLEQUOTE = "\xE2\x80\x9D";
static const char *LEFTSINGLEQUOTE = "\xE2\x80\x98";
static const char *RIGHTSINGLEQUOTE = "\xE2\x80\x99";

// Macros for creating various kinds of simple.
#define make_str(mem, s) make_literal(mem, CMARK_NODE_TEXT, s)
#define make_code(mem, s) make_literal(mem, CMARK_NODE_CODE, s)
#define make_raw_html(mem, s) make_literal(mem, CMARK_NODE_HTML_INLINE, s)
#define make_linebreak(mem) make_simple(mem, CMARK_NODE_LINEBREAK)
#define make_softbreak(mem) make_simple(mem, CMARK_NODE_SOFTBREAK)
#define make_emph(mem) make_simple(mem, CMARK_NODE_EMPH)
#define make_strong(mem) make_simple(mem, CMARK_NODE_STRONG)

typedef struct delimiter {
  struct delimiter *previous;
  struct delimiter *next;
  cmark_node *inl_text;
  bufsize_t position;
  unsigned char delim_char;
  bool can_open;
  bool can_close;
} delimiter;

typedef struct bracket {
	struct bracket *previous;
	struct delimiter *previous_delimiter;
	cmark_node *inl_text;
	bufsize_t position;
  bool image;
  bool active;
  bool bracket_after;
} bracket;

typedef struct {
  cmark_mem *mem;
  cmark_chunk input;
  bufsize_t pos;
  cmark_reference_map *refmap;
  delimiter *last_delim;
  bracket   *last_bracket;
} subject;

static CMARK_INLINE bool S_is_line_end_char(char c) {
  return (c == '\n' || c == '\r');
}

static delimiter *S_insert_emph(subject *subj, delimiter *opener,
                                delimiter *closer);

static int parse_inline(subject *subj, cmark_node *parent, int options);

static void subject_from_buf(cmark_mem *mem, subject *e, cmark_strbuf *buffer,
                             cmark_reference_map *refmap);
static bufsize_t subject_find_special_char(subject *subj, int options);

// Create an inline with a literal string value.
static CMARK_INLINE cmark_node *make_literal(cmark_mem *mem, cmark_node_type t, cmark_chunk s) {
  cmark_node *e = (cmark_node *)mem->calloc(1, sizeof(*e));
  cmark_strbuf_init(mem, &e->content, 0);
  e->type = t;
  e->as.literal = s;
  return e;
}

// Create an inline with no value.
static CMARK_INLINE cmark_node *make_simple(cmark_mem *mem, cmark_node_type t) {
  cmark_node *e = (cmark_node *)mem->calloc(1, sizeof(*e));
  cmark_strbuf_init(mem, &e->content, 0);
  e->type = t;
  return e;
}

// Like make_str, but parses entities.
static cmark_node *make_str_with_entities(cmark_mem *mem, cmark_chunk *content) {
  cmark_strbuf unescaped = CMARK_BUF_INIT(mem);

  if (houdini_unescape_html(&unescaped, content->data, content->len)) {
    return make_str(mem, cmark_chunk_buf_detach(&unescaped));
  } else {
    return make_str(mem, *content);
  }
}

// Duplicate a chunk by creating a copy of the buffer not by reusing the
// buffer like cmark_chunk_dup does.
static cmark_chunk chunk_clone(cmark_mem *mem, cmark_chunk *src) {
  cmark_chunk c;
  bufsize_t len = src->len;

  c.len = len;
  c.data = (unsigned char *)mem->calloc(len + 1, 1);
  c.alloc = 1;
  memcpy(c.data, src->data, len);
  c.data[len] = '\0';

  return c;
}

static cmark_chunk cmark_clean_autolink(cmark_mem *mem, cmark_chunk *url, int is_email) {
  cmark_strbuf buf = CMARK_BUF_INIT(mem);

  cmark_chunk_trim(url);

  if (url->len == 0) {
    cmark_chunk result = CMARK_CHUNK_EMPTY;
    return result;
  }

  if (is_email)
    cmark_strbuf_puts(&buf, "mailto:");

  houdini_unescape_html_f(&buf, url->data, url->len);
  return cmark_chunk_buf_detach(&buf);
}

static CMARK_INLINE cmark_node *make_autolink(cmark_mem *mem, cmark_chunk url, int is_email) {
  cmark_node *link = make_simple(mem, CMARK_NODE_LINK);
  link->as.link.url = cmark_clean_autolink(mem, &url, is_email);
  link->as.link.title = cmark_chunk_literal("");
  cmark_node_append_child(link, make_str_with_entities(mem, &url));
  return link;
}

static void subject_from_buf(cmark_mem *mem, subject *e, cmark_strbuf *buffer,
                             cmark_reference_map *refmap) {
  e->mem = mem;
  e->input.data = buffer->ptr;
  e->input.len = buffer->size;
  e->input.alloc = 0;
  e->pos = 0;
  e->refmap = refmap;
  e->last_delim = NULL;
  e->last_bracket = NULL;
}

static CMARK_INLINE int isbacktick(int c) { return (c == '`'); }

static CMARK_INLINE unsigned char peek_char(subject *subj) {
  // NULL bytes should have been stripped out by now.  If they're
  // present, it's a programming error:
  assert(!(subj->pos < subj->input.len && subj->input.data[subj->pos] == 0));
  return (subj->pos < subj->input.len) ? subj->input.data[subj->pos] : 0;
}

static CMARK_INLINE unsigned char peek_at(subject *subj, bufsize_t pos) {
  return subj->input.data[pos];
}

// Return true if there are more characters in the subject.
static CMARK_INLINE int is_eof(subject *subj) {
  return (subj->pos >= subj->input.len);
}

// Advance the subject.  Doesn't check for eof.
#define advance(subj) (subj)->pos += 1

static CMARK_INLINE bool skip_spaces(subject *subj) {
  bool skipped = false;
  while (peek_char(subj) == ' ' || peek_char(subj) == '\t') {
    advance(subj);
    skipped = true;
  }
  return skipped;
}

static CMARK_INLINE bool skip_line_end(subject *subj) {
  bool seen_line_end_char = false;
  if (peek_char(subj) == '\r') {
    advance(subj);
    seen_line_end_char = true;
  }
  if (peek_char(subj) == '\n') {
    advance(subj);
    seen_line_end_char = true;
  }
  return seen_line_end_char || is_eof(subj);
}

// Take characters while a predicate holds, and return a string.
static CMARK_INLINE cmark_chunk take_while(subject *subj, int (*f)(int)) {
  unsigned char c;
  bufsize_t startpos = subj->pos;
  bufsize_t len = 0;

  while ((c = peek_char(subj)) && (*f)(c)) {
    advance(subj);
    len++;
  }

  return cmark_chunk_dup(&subj->input, startpos, len);
}

// Try to process a backtick code span that began with a
// span of ticks of length openticklength length (already
// parsed).  Return 0 if you don't find matching closing
// backticks, otherwise return the position in the subject
// after the closing backticks.
static bufsize_t scan_to_closing_backticks(subject *subj,
                                           bufsize_t openticklength) {
  // read non backticks
  unsigned char c;
  while ((c = peek_char(subj)) && c != '`') {
    advance(subj);
  }
  if (is_eof(subj)) {
    return 0; // did not find closing ticks, return 0
  }
  bufsize_t numticks = 0;
  while (peek_char(subj) == '`') {
    advance(subj);
    numticks++;
  }
  if (numticks != openticklength) {
    return (scan_to_closing_backticks(subj, openticklength));
  }
  return (subj->pos);
}

// Parse backtick code section or raw backticks, return an inline.
// Assumes that the subject has a backtick at the current position.
static cmark_node *handle_backticks(subject *subj) {
  cmark_chunk openticks = take_while(subj, isbacktick);
  bufsize_t startpos = subj->pos;
  bufsize_t endpos = scan_to_closing_backticks(subj, openticks.len);

  if (endpos == 0) {      // not found
    subj->pos = startpos; // rewind
    return make_str(subj->mem, openticks);
  } else {
    cmark_strbuf buf = CMARK_BUF_INIT(subj->mem);

    cmark_strbuf_set(&buf, subj->input.data + startpos,
                     endpos - startpos - openticks.len);
    cmark_strbuf_trim(&buf);
    cmark_strbuf_normalize_whitespace(&buf);

    return make_code(subj->mem, cmark_chunk_buf_detach(&buf));
  }
}

// Scan ***, **, or * and return number scanned, or 0.
// Advances position.
static int scan_delims(subject *subj, unsigned char c, bool *can_open,
                       bool *can_close) {
  int numdelims = 0;
  bufsize_t before_char_pos;
  int32_t after_char = 0;
  int32_t before_char = 0;
  int len;
  bool left_flanking, right_flanking;

  if (subj->pos == 0) {
    before_char = 10;
  } else {
    before_char_pos = subj->pos - 1;
    // walk back to the beginning of the UTF_8 sequence:
    while (peek_at(subj, before_char_pos) >> 6 == 2 && before_char_pos > 0) {
      before_char_pos -= 1;
    }
    len = cmark_utf8proc_iterate(subj->input.data + before_char_pos,
                                 subj->pos - before_char_pos, &before_char);
    if (len == -1) {
      before_char = 10;
    }
  }

  if (c == '\'' || c == '"') {
    numdelims++;
    advance(subj); // limit to 1 delim for quotes
  } else {
    while (peek_char(subj) == c) {
      numdelims++;
      advance(subj);
    }
  }

  len = cmark_utf8proc_iterate(subj->input.data + subj->pos,
                               subj->input.len - subj->pos, &after_char);
  if (len == -1) {
    after_char = 10;
  }
  left_flanking = numdelims > 0 && !cmark_utf8proc_is_space(after_char) &&
                  !(cmark_utf8proc_is_punctuation(after_char) &&
                    !cmark_utf8proc_is_space(before_char) &&
                    !cmark_utf8proc_is_punctuation(before_char));
  right_flanking = numdelims > 0 && !cmark_utf8proc_is_space(before_char) &&
                   !(cmark_utf8proc_is_punctuation(before_char) &&
                     !cmark_utf8proc_is_space(after_char) &&
                     !cmark_utf8proc_is_punctuation(after_char));
  if (c == '_') {
    *can_open = left_flanking &&
                (!right_flanking || cmark_utf8proc_is_punctuation(before_char));
    *can_close = right_flanking &&
                 (!left_flanking || cmark_utf8proc_is_punctuation(after_char));
  } else if (c == '\'' || c == '"') {
    *can_open = left_flanking && !right_flanking;
    *can_close = right_flanking;
  } else {
    *can_open = left_flanking;
    *can_close = right_flanking;
  }
  return numdelims;
}

/*
static void print_delimiters(subject *subj)
{
        delimiter *delim;
        delim = subj->last_delim;
        while (delim != NULL) {
                printf("Item at stack pos %p, text pos %d: %d %d %d next(%p)
prev(%p)\n",
                       (void*)delim, delim->position, delim->delim_char,
                       delim->can_open, delim->can_close,
                       (void*)delim->next, (void*)delim->previous);
                delim = delim->previous;
        }
}
*/

static void remove_delimiter(subject *subj, delimiter *delim) {
  if (delim == NULL)
    return;
  if (delim->next == NULL) {
    // end of list:
    assert(delim == subj->last_delim);
    subj->last_delim = delim->previous;
  } else {
    delim->next->previous = delim->previous;
  }
  if (delim->previous != NULL) {
    delim->previous->next = delim->next;
  }
  free(delim);
}

static void pop_bracket(subject *subj) {
  bracket *b;
  if (subj->last_bracket == NULL)
    return;
  b = subj->last_bracket;
  subj->last_bracket = subj->last_bracket->previous;
  free(b);
}


static void push_delimiter(subject *subj, unsigned char c, bool can_open,
                           bool can_close, cmark_node *inl_text) {
  delimiter *delim = (delimiter *)subj->mem->calloc(1, sizeof(delimiter));
  delim->delim_char = c;
  delim->can_open = can_open;
  delim->can_close = can_close;
  delim->inl_text = inl_text;
  delim->previous = subj->last_delim;
  delim->next = NULL;
  if (delim->previous != NULL) {
    delim->previous->next = delim;
  }
  delim->position = subj->pos;
  subj->last_delim = delim;
}

static void push_bracket(subject *subj, bool image, cmark_node *inl_text) {
  bracket *b = (bracket *)subj->mem->calloc(1, sizeof(bracket));
  if (subj->last_bracket != NULL) {
    subj->last_bracket->bracket_after = true;
  }
  b->image = image;
  b->active = true;
  b->inl_text = inl_text;
  b->previous = subj->last_bracket;
  b->previous_delimiter = subj->last_delim;
  b->position = subj->pos;
  b->bracket_after = false;
  subj->last_bracket = b;
}

// Assumes the subject has a c at the current position.
static cmark_node *handle_delim(subject *subj, unsigned char c, bool smart) {
  bufsize_t numdelims;
  cmark_node *inl_text;
  bool can_open, can_close;
  cmark_chunk contents;

  numdelims = scan_delims(subj, c, &can_open, &can_close);

  if (c == '\'' && smart) {
    contents = cmark_chunk_literal(RIGHTSINGLEQUOTE);
  } else if (c == '"' && smart) {
    contents =
        cmark_chunk_literal(can_close ? RIGHTDOUBLEQUOTE : LEFTDOUBLEQUOTE);
  } else {
    contents = cmark_chunk_dup(&subj->input, subj->pos - numdelims, numdelims);
  }

  inl_text = make_str(subj->mem, contents);

  if ((can_open || can_close) && (!(c == '\'' || c == '"') || smart)) {
    push_delimiter(subj, c, can_open, can_close, inl_text);
  }

  return inl_text;
}

// Assumes we have a hyphen at the current position.
static cmark_node *handle_hyphen(subject *subj, bool smart) {
  int startpos = subj->pos;

  advance(subj);

  if (!smart || peek_char(subj) != '-') {
    return make_str(subj->mem, cmark_chunk_literal("-"));
  }

  while (smart && peek_char(subj) == '-') {
    advance(subj);
  }

  int numhyphens = subj->pos - startpos;
  int en_count = 0;
  int em_count = 0;
  int i;
  cmark_strbuf buf = CMARK_BUF_INIT(subj->mem);

  if (numhyphens % 3 == 0) { // if divisible by 3, use all em dashes
    em_count = numhyphens / 3;
  } else if (numhyphens % 2 == 0) { // if divisible by 2, use all en dashes
    en_count = numhyphens / 2;
  } else if (numhyphens % 3 == 2) { // use one en dash at end
    en_count = 1;
    em_count = (numhyphens - 2) / 3;
  } else { // use two en dashes at the end
    en_count = 2;
    em_count = (numhyphens - 4) / 3;
  }

  for (i = em_count; i > 0; i--) {
    cmark_strbuf_puts(&buf, EMDASH);
  }

  for (i = en_count; i > 0; i--) {
    cmark_strbuf_puts(&buf, ENDASH);
  }

  return make_str(subj->mem, cmark_chunk_buf_detach(&buf));
}

// Assumes we have a period at the current position.
static cmark_node *handle_period(subject *subj, bool smart) {
  advance(subj);
  if (smart && peek_char(subj) == '.') {
    advance(subj);
    if (peek_char(subj) == '.') {
      advance(subj);
      return make_str(subj->mem, cmark_chunk_literal(ELLIPSES));
    } else {
      return make_str(subj->mem, cmark_chunk_literal(".."));
    }
  } else {
    return make_str(subj->mem, cmark_chunk_literal("."));
  }
}

static void process_emphasis(subject *subj, delimiter *stack_bottom) {
  delimiter *closer = subj->last_delim;
  delimiter *opener;
  delimiter *old_closer;
  bool opener_found;
  delimiter *openers_bottom[128];

  // initialize openers_bottom:
  openers_bottom['*'] = stack_bottom;
  openers_bottom['_'] = stack_bottom;
  openers_bottom['\''] = stack_bottom;
  openers_bottom['"'] = stack_bottom;

  // move back to first relevant delim.
  while (closer != NULL && closer->previous != stack_bottom) {
    closer = closer->previous;
  }

  // now move forward, looking for closers, and handling each
  while (closer != NULL) {
    if (closer->can_close) {
      // Now look backwards for first matching opener:
      opener = closer->previous;
      opener_found = false;
      while (opener != NULL && opener != stack_bottom &&
             opener != openers_bottom[closer->delim_char]) {
        if (opener->delim_char == closer->delim_char && opener->can_open) {
          opener_found = true;
          break;
        }
        opener = opener->previous;
      }
      old_closer = closer;
      if (closer->delim_char == '*' || closer->delim_char == '_') {
        if (opener_found) {
          closer = S_insert_emph(subj, opener, closer);
        } else {
          closer = closer->next;
        }
      } else if (closer->delim_char == '\'') {
        cmark_chunk_free(subj->mem, &closer->inl_text->as.literal);
        closer->inl_text->as.literal = cmark_chunk_literal(RIGHTSINGLEQUOTE);
        if (opener_found) {
          cmark_chunk_free(subj->mem, &opener->inl_text->as.literal);
          opener->inl_text->as.literal = cmark_chunk_literal(LEFTSINGLEQUOTE);
        }
        closer = closer->next;
      } else if (closer->delim_char == '"') {
        cmark_chunk_free(subj->mem, &closer->inl_text->as.literal);
        closer->inl_text->as.literal = cmark_chunk_literal(RIGHTDOUBLEQUOTE);
        if (opener_found) {
          cmark_chunk_free(subj->mem, &opener->inl_text->as.literal);
          opener->inl_text->as.literal = cmark_chunk_literal(LEFTDOUBLEQUOTE);
        }
        closer = closer->next;
      }
      if (!opener_found) {
        // set lower bound for future searches for openers:
        openers_bottom[old_closer->delim_char] = old_closer->previous;
        if (!old_closer->can_open) {
          // we can remove a closer that can't be an
          // opener, once we've seen there's no
          // matching opener:
          remove_delimiter(subj, old_closer);
        }
      }
    } else {
      closer = closer->next;
    }
  }
  // free all delimiters in list until stack_bottom:
  while (subj->last_delim != stack_bottom) {
    remove_delimiter(subj, subj->last_delim);
  }
}

static delimiter *S_insert_emph(subject *subj, delimiter *opener,
                                delimiter *closer) {
  delimiter *delim, *tmp_delim;
  bufsize_t use_delims;
  cmark_node *opener_inl = opener->inl_text;
  cmark_node *closer_inl = closer->inl_text;
  bufsize_t opener_num_chars = opener_inl->as.literal.len;
  bufsize_t closer_num_chars = closer_inl->as.literal.len;
  cmark_node *tmp, *tmpnext, *emph;

  // calculate the actual number of characters used from this closer
  if (closer_num_chars < 3 || opener_num_chars < 3) {
    use_delims = closer_num_chars <= opener_num_chars ? closer_num_chars
                                                      : opener_num_chars;
  } else { // closer and opener both have >= 3 characters
    use_delims = closer_num_chars % 2 == 0 ? 2 : 1;
  }

  // remove used characters from associated inlines.
  opener_num_chars -= use_delims;
  closer_num_chars -= use_delims;
  opener_inl->as.literal.len = opener_num_chars;
  closer_inl->as.literal.len = closer_num_chars;

  // free delimiters between opener and closer
  delim = closer->previous;
  while (delim != NULL && delim != opener) {
    tmp_delim = delim->previous;
    remove_delimiter(subj, delim);
    delim = tmp_delim;
  }

  // create new emph or strong, and splice it in to our inlines
  // between the opener and closer
  emph = use_delims == 1 ? make_emph(subj->mem) : make_strong(subj->mem);

  tmp = opener_inl->next;
  while (tmp && tmp != closer_inl) {
    tmpnext = tmp->next;
    cmark_node_append_child(emph, tmp);
    tmp = tmpnext;
  }
  cmark_node_insert_after(opener_inl, emph);

  // if opener has 0 characters, remove it and its associated inline
  if (opener_num_chars == 0) {
    cmark_node_free(opener_inl);
    remove_delimiter(subj, opener);
  }

  // if closer has 0 characters, remove it and its associated inline
  if (closer_num_chars == 0) {
    // remove empty closer inline
    cmark_node_free(closer_inl);
    // remove closer from list
    tmp_delim = closer->next;
    remove_delimiter(subj, closer);
    closer = tmp_delim;
  }

  return closer;
}

// Parse backslash-escape or just a backslash, returning an inline.
static cmark_node *handle_backslash(subject *subj) {
  advance(subj);
  unsigned char nextchar = peek_char(subj);
  if (cmark_ispunct(
          nextchar)) { // only ascii symbols and newline can be escaped
    advance(subj);
    return make_str(subj->mem, cmark_chunk_dup(&subj->input, subj->pos - 1, 1));
  } else if (!is_eof(subj) && skip_line_end(subj)) {
    return make_linebreak(subj->mem);
  } else {
    return make_str(subj->mem, cmark_chunk_literal("\\"));
  }
}

// Parse an entity or a regular "&" string.
// Assumes the subject has an '&' character at the current position.
static cmark_node *handle_entity(subject *subj) {
  cmark_strbuf ent = CMARK_BUF_INIT(subj->mem);
  bufsize_t len;

  advance(subj);

  len = houdini_unescape_ent(&ent, subj->input.data + subj->pos,
                             subj->input.len - subj->pos);

  if (len == 0)
    return make_str(subj->mem, cmark_chunk_literal("&"));

  subj->pos += len;
  return make_str(subj->mem, cmark_chunk_buf_detach(&ent));
}

// Clean a URL: remove surrounding whitespace and surrounding <>,
// and remove \ that escape punctuation.
cmark_chunk cmark_clean_url(cmark_mem *mem, cmark_chunk *url) {
  cmark_strbuf buf = CMARK_BUF_INIT(mem);

  cmark_chunk_trim(url);

  if (url->len == 0) {
    cmark_chunk result = CMARK_CHUNK_EMPTY;
    return result;
  }

  if (url->data[0] == '<' && url->data[url->len - 1] == '>') {
    houdini_unescape_html_f(&buf, url->data + 1, url->len - 2);
  } else {
    houdini_unescape_html_f(&buf, url->data, url->len);
  }

  cmark_strbuf_unescape(&buf);
  return cmark_chunk_buf_detach(&buf);
}

cmark_chunk cmark_clean_title(cmark_mem *mem, cmark_chunk *title) {
  cmark_strbuf buf = CMARK_BUF_INIT(mem);
  unsigned char first, last;

  if (title->len == 0) {
    cmark_chunk result = CMARK_CHUNK_EMPTY;
    return result;
  }

  first = title->data[0];
  last = title->data[title->len - 1];

  // remove surrounding quotes if any:
  if ((first == '\'' && last == '\'') || (first == '(' && last == ')') ||
      (first == '"' && last == '"')) {
    houdini_unescape_html_f(&buf, title->data + 1, title->len - 2);
  } else {
    houdini_unescape_html_f(&buf, title->data, title->len);
  }

  cmark_strbuf_unescape(&buf);
  return cmark_chunk_buf_detach(&buf);
}

// Parse an autolink or HTML tag.
// Assumes the subject has a '<' character at the current position.
static cmark_node *handle_pointy_brace(subject *subj) {
  bufsize_t matchlen = 0;
  cmark_chunk contents;

  advance(subj); // advance past first <

  // first try to match a URL autolink
  matchlen = scan_autolink_uri(&subj->input, subj->pos);
  if (matchlen > 0) {
    contents = cmark_chunk_dup(&subj->input, subj->pos, matchlen - 1);
    subj->pos += matchlen;

    return make_autolink(subj->mem, contents, 0);
  }

  // next try to match an email autolink
  matchlen = scan_autolink_email(&subj->input, subj->pos);
  if (matchlen > 0) {
    contents = cmark_chunk_dup(&subj->input, subj->pos, matchlen - 1);
    subj->pos += matchlen;

    return make_autolink(subj->mem, contents, 1);
  }

  // finally, try to match an html tag
  matchlen = scan_html_tag(&subj->input, subj->pos);
  if (matchlen > 0) {
    contents = cmark_chunk_dup(&subj->input, subj->pos - 1, matchlen + 1);
    subj->pos += matchlen;
    return make_raw_html(subj->mem, contents);
  }

  // if nothing matches, just return the opening <:
  return make_str(subj->mem, cmark_chunk_literal("<"));
}

// Parse a link label.  Returns 1 if successful.
// Note:  unescaped brackets are not allowed in labels.
// The label begins with `[` and ends with the first `]` character
// encountered.  Backticks in labels do not start code spans.
static int link_label(subject *subj, cmark_chunk *raw_label) {
  bufsize_t startpos = subj->pos;
  int length = 0;
  unsigned char c;

  // advance past [
  if (peek_char(subj) == '[') {
    advance(subj);
  } else {
    return 0;
  }

  while ((c = peek_char(subj)) && c != '[' && c != ']') {
    if (c == '\\') {
      advance(subj);
      length++;
      if (cmark_ispunct(peek_char(subj))) {
        advance(subj);
        length++;
      }
    } else {
      advance(subj);
      length++;
    }
    if (length > MAX_LINK_LABEL_LENGTH) {
      goto noMatch;
    }
  }

  if (c == ']') { // match found
    *raw_label =
        cmark_chunk_dup(&subj->input, startpos + 1, subj->pos - (startpos + 1));
    cmark_chunk_trim(raw_label);
    advance(subj); // advance past ]
    return 1;
  }

noMatch:
  subj->pos = startpos; // rewind
  return 0;
}

// Return a link, an image, or a literal close bracket.
static cmark_node *handle_close_bracket(subject *subj) {
  bufsize_t initial_pos;
  bufsize_t starturl, endurl, starttitle, endtitle, endall;
  bufsize_t n;
  bufsize_t sps;
  cmark_reference *ref = NULL;
  cmark_chunk url_chunk, title_chunk;
  cmark_chunk url, title;
  bracket *opener;
  cmark_node *inl;
  cmark_chunk raw_label;
  int found_label;
  cmark_node *tmp, *tmpnext;
  bool is_image;

  advance(subj); // advance past ]
  initial_pos = subj->pos;

  // get last [ or ![
  opener = subj->last_bracket;

  if (opener == NULL) {
    return make_str(subj->mem, cmark_chunk_literal("]"));
  }

  if (!opener->active) {
    // take delimiter off stack
    pop_bracket(subj);
    return make_str(subj->mem, cmark_chunk_literal("]"));
  }

  // If we got here, we matched a potential link/image text.
  // Now we check to see if it's a link/image.
  is_image = opener->image;

  // First, look for an inline link.
  if (peek_char(subj) == '(' &&
      ((sps = scan_spacechars(&subj->input, subj->pos + 1)) > -1) &&
      ((n = scan_link_url(&subj->input, subj->pos + 1 + sps)) > -1)) {

    // try to parse an explicit link:
    starturl = subj->pos + 1 + sps; // after (
    endurl = starturl + n;
    starttitle = endurl + scan_spacechars(&subj->input, endurl);

    // ensure there are spaces btw url and title
    endtitle = (starttitle == endurl)
                   ? starttitle
                   : starttitle + scan_link_title(&subj->input, starttitle);

    endall = endtitle + scan_spacechars(&subj->input, endtitle);

    if (peek_at(subj, endall) == ')') {
      subj->pos = endall + 1;

      url_chunk = cmark_chunk_dup(&subj->input, starturl, endurl - starturl);
      title_chunk =
          cmark_chunk_dup(&subj->input, starttitle, endtitle - starttitle);
      url = cmark_clean_url(subj->mem, &url_chunk);
      title = cmark_clean_title(subj->mem, &title_chunk);
      cmark_chunk_free(subj->mem, &url_chunk);
      cmark_chunk_free(subj->mem, &title_chunk);
      goto match;

    } else {
      goto noMatch;
    }
  }

  // Next, look for a following [link label] that matches in refmap.
  // skip spaces
  raw_label = cmark_chunk_literal("");
  found_label = link_label(subj, &raw_label);
  if (!found_label) {
    // If we have a shortcut reference link, back up
    // to before the spacse we skipped.
    subj->pos = initial_pos;
  }

  if ((!found_label || raw_label.len == 0) && !opener->bracket_after) {
    cmark_chunk_free(subj->mem, &raw_label);
    raw_label = cmark_chunk_dup(&subj->input, opener->position,
                                initial_pos - opener->position - 1);
    found_label = true;
  }

  if (found_label) {
    ref = cmark_reference_lookup(subj->refmap, &raw_label);
    cmark_chunk_free(subj->mem, &raw_label);
  }

  if (ref != NULL) { // found
    url = chunk_clone(subj->mem, &ref->url);
    title = chunk_clone(subj->mem, &ref->title);
    goto match;
  } else {
    goto noMatch;
  }

noMatch:
  // If we fall through to here, it means we didn't match a link:
  pop_bracket(subj); // remove this opener from delimiter list
  subj->pos = initial_pos;
  return make_str(subj->mem, cmark_chunk_literal("]"));

match:
  inl = make_simple(subj->mem, is_image ? CMARK_NODE_IMAGE : CMARK_NODE_LINK);
  inl->as.link.url = url;
  inl->as.link.title = title;
  cmark_node_insert_before(opener->inl_text, inl);
  // Add link text:
  tmp = opener->inl_text->next;
  while (tmp) {
    tmpnext = tmp->next;
    cmark_node_append_child(inl, tmp);
    tmp = tmpnext;
  }

  // Free the bracket [:
  cmark_node_free(opener->inl_text);

  process_emphasis(subj, opener->previous_delimiter);
  pop_bracket(subj);

  // Now, if we have a link, we also want to deactivate earlier link
  // delimiters. (This code can be removed if we decide to allow links
  // inside links.)
  if (!is_image) {
    opener = subj->last_bracket;
    while (opener != NULL) {
      if (!opener->image) {
        if (!opener->active) {
          break;
        } else {
          opener->active = false;
        }
      }
      opener = opener->previous;
    }
  }

  return NULL;
}

// Parse a hard or soft linebreak, returning an inline.
// Assumes the subject has a cr or newline at the current position.
static cmark_node *handle_newline(subject *subj) {
  bufsize_t nlpos = subj->pos;
  // skip over cr, crlf, or lf:
  if (peek_at(subj, subj->pos) == '\r') {
    advance(subj);
  }
  if (peek_at(subj, subj->pos) == '\n') {
    advance(subj);
  }
  // skip spaces at beginning of line
  skip_spaces(subj);
  if (nlpos > 1 && peek_at(subj, nlpos - 1) == ' ' &&
      peek_at(subj, nlpos - 2) == ' ') {
    return make_linebreak(subj->mem);
  } else {
    return make_softbreak(subj->mem);
  }
}

static bufsize_t subject_find_special_char(subject *subj, int options) {
  // "\r\n\\`&_*[]<!"
  static const int8_t SPECIAL_CHARS[256] = {
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1,
      1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};

  // " ' . -
  static const char SMART_PUNCT_CHARS[] = {
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
      0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
  };

  bufsize_t n = subj->pos + 1;

  while (n < subj->input.len) {
    if (SPECIAL_CHARS[subj->input.data[n]])
      return n;
    if (options & CMARK_OPT_SMART && SMART_PUNCT_CHARS[subj->input.data[n]])
      return n;
    n++;
  }

  return subj->input.len;
}

// Parse an inline, advancing subject, and add it as a child of parent.
// Return 0 if no inline can be parsed, 1 otherwise.
static int parse_inline(subject *subj, cmark_node *parent, int options) {
  cmark_node *new_inl = NULL;
  cmark_chunk contents;
  unsigned char c;
  bufsize_t endpos;
  c = peek_char(subj);
  if (c == 0) {
    return 0;
  }
  switch (c) {
  case '\r':
  case '\n':
    new_inl = handle_newline(subj);
    break;
  case '`':
    new_inl = handle_backticks(subj);
    break;
  case '\\':
    new_inl = handle_backslash(subj);
    break;
  case '&':
    new_inl = handle_entity(subj);
    break;
  case '<':
    new_inl = handle_pointy_brace(subj);
    break;
  case '*':
  case '_':
  case '\'':
  case '"':
    new_inl = handle_delim(subj, c, (options & CMARK_OPT_SMART) != 0);
    break;
  case '-':
    new_inl = handle_hyphen(subj, (options & CMARK_OPT_SMART) != 0);
    break;
  case '.':
    new_inl = handle_period(subj, (options & CMARK_OPT_SMART) != 0);
    break;
  case '[':
    advance(subj);
    new_inl = make_str(subj->mem, cmark_chunk_literal("["));
    push_bracket(subj, false, new_inl);
    break;
  case ']':
    new_inl = handle_close_bracket(subj);
    break;
  case '!':
    advance(subj);
    if (peek_char(subj) == '[') {
      advance(subj);
      new_inl = make_str(subj->mem, cmark_chunk_literal("!["));
      push_bracket(subj, true, new_inl);
    } else {
      new_inl = make_str(subj->mem, cmark_chunk_literal("!"));
    }
    break;
  default:
    endpos = subject_find_special_char(subj, options);
    contents = cmark_chunk_dup(&subj->input, subj->pos, endpos - subj->pos);
    subj->pos = endpos;

    // if we're at a newline, strip trailing spaces.
    if (S_is_line_end_char(peek_char(subj))) {
      cmark_chunk_rtrim(&contents);
    }

    new_inl = make_str(subj->mem, contents);
  }
  if (new_inl != NULL) {
    cmark_node_append_child(parent, new_inl);
  }

  return 1;
}

// Parse inlines from parent's string_content, adding as children of parent.
extern void cmark_parse_inlines(cmark_mem *mem, cmark_node *parent, cmark_reference_map *refmap,
                                int options) {
  subject subj;
  subject_from_buf(mem, &subj, &parent->content, refmap);
  cmark_chunk_rtrim(&subj.input);

  while (!is_eof(&subj) && parse_inline(&subj, parent, options))
    ;

  process_emphasis(&subj, NULL);
  // free bracket stack
  while (subj.last_bracket) {
    pop_bracket(&subj);
  }
}

// Parse zero or more space characters, including at most one newline.
static void spnl(subject *subj) {
  skip_spaces(subj);
  if (skip_line_end(subj)) {
    skip_spaces(subj);
  }
}

// Parse reference.  Assumes string begins with '[' character.
// Modify refmap if a reference is encountered.
// Return 0 if no reference found, otherwise position of subject
// after reference is parsed.
bufsize_t cmark_parse_reference_inline(cmark_mem *mem, cmark_strbuf *input,
                                       cmark_reference_map *refmap) {
  subject subj;

  cmark_chunk lab;
  cmark_chunk url;
  cmark_chunk title;

  bufsize_t matchlen = 0;
  bufsize_t beforetitle;

  subject_from_buf(mem, &subj, input, NULL);

  // parse label:
  if (!link_label(&subj, &lab) || lab.len == 0)
    return 0;

  // colon:
  if (peek_char(&subj) == ':') {
    advance(&subj);
  } else {
    return 0;
  }

  // parse link url:
  spnl(&subj);
  matchlen = scan_link_url(&subj.input, subj.pos);
  if (matchlen) {
    url = cmark_chunk_dup(&subj.input, subj.pos, matchlen);
    subj.pos += matchlen;
  } else {
    return 0;
  }

  // parse optional link_title
  beforetitle = subj.pos;
  spnl(&subj);
  matchlen = scan_link_title(&subj.input, subj.pos);
  if (matchlen) {
    title = cmark_chunk_dup(&subj.input, subj.pos, matchlen);
    subj.pos += matchlen;
  } else {
    subj.pos = beforetitle;
    title = cmark_chunk_literal("");
  }

  // parse final spaces and newline:
  skip_spaces(&subj);
  if (!skip_line_end(&subj)) {
    if (matchlen) { // try rewinding before title
      subj.pos = beforetitle;
      skip_spaces(&subj);
      if (!skip_line_end(&subj)) {
        return 0;
      }
    } else {
      return 0;
    }
  }
  // insert reference into refmap
  cmark_reference_create(refmap, &lab, &url, &title);
  return subj.pos;
}