summaryrefslogtreecommitdiff
path: root/macros/texinfo/texinfo/info/scan.c
blob: a52117ed2c120fc5e549918a9ed5e18e32c01197 (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
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
/* scan.c -- scanning Info files and nodes

   Copyright 1993-2022 Free Software Foundation, Inc.

   This program is free software: you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation, either version 3 of the License, or
   (at your option) any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program.  If not, see <http://www.gnu.org/licenses/>.

   Originally written by Brian Fox. */

#include "info.h"
#include "session.h"
#include "scan.h"
#include "util.h"
#include "tag.h"

#include <langinfo.h>
#if HAVE_ICONV
# include <iconv.h>
#endif
#include <wchar.h>
#ifdef __MINGW32__
/* MinGW uses a replacement nl_langinfo, see pcterm.c.  */
# define nl_langinfo rpl_nl_langinfo
extern char * rpl_nl_langinfo (nl_item);
/* MinGW uses its own replacement wcwidth, see pcterm.c for the
   reasons.  Since Gnulib's wchar.h might redirect wcwidth to
   rpl_wcwidth, we explicitly undo that here.  */
#undef wcwidth
#endif

#ifdef __hpux
#define va_copy(ap1,ap2) memcpy((&ap1),(&ap2),sizeof(va_list))
#endif

/* Variable which holds the most recent filename parsed as a result of
   calling info_parse_xxx (). */
char *info_parsed_filename = NULL;

/* Variable which holds the most recent nodename parsed as a result of
   calling info_parse_xxx (). */
char *info_parsed_nodename = NULL;

/* Read a filename surrounded by "(" and ")", accounting for matching
   characters, and place it in *FILENAME if FILENAME is not null.  Return 
   length of read filename.  On error, set *FILENAME to null and return 0.  */
int
read_bracketed_filename (char *string, char **filename)
{
  register int i = 0;
  int count = 0; /* Level of nesting. */
  int first_close = -1; /* First ")" encountered. */

  if (*string != '(')
    return 0;

  string++;
  count = 1;
  for (i = 0; string[i]; i++)
    {
      if (string[i] == '(')
        count++;
      else if (string[i] == ')')
        {
          if (first_close == -1)
            first_close = i;

          count--;
          if (count == 0)
            break;
        } 
    }
  
  /* If string ended before brackets were balanced, take the first ")" as
     terminating the filename. */
  if (count > 0)
    {
      if (first_close == -1)
        {
          if (filename)
            *filename = 0;
          return 0;
        }
      i = first_close;
    }

  if (filename)
    {
      *filename = xcalloc (1, i + 1);
      memcpy (*filename, string, i);
    }

  return i + 2; /* Length of filename plus "(" and ")". */
}

/* Parse the filename and nodename out of STRING, saving in
   INFO_PARSED_FILENAME and INFO_PARSED_NODENAME.  These variables should not
   be freed by calling code.  If either is missing, the relevant variable is
   set to a null pointer. */ 
void
info_parse_node (char *string)
{
  int nodename_len;

  free (info_parsed_filename);
  free (info_parsed_nodename);
  info_parsed_filename = 0;
  info_parsed_nodename = 0;

  /* Special case of nothing passed.  Return nothing. */
  if (!string || !*string)
    return;

  string += skip_whitespace_and_newlines (string);

  string += read_bracketed_filename (string, &info_parsed_filename);

  /* Parse out nodename. */
  string += skip_whitespace_and_newlines (string);
  nodename_len = read_quoted_string (string, "", 0, &info_parsed_nodename);

  if (nodename_len != 0)
    {
      canonicalize_whitespace (info_parsed_nodename);
    }
}

/* Set *OUTPUT to a copy of the string starting at START and finishing at
   a character in TERMINATOR, unless START[0] == INFO_QUOTE, in which case
   copy string from START+1 until the next occurence of INFO_QUOTE.  If
   TERMINATOR is an empty string, finish at a null character.   LINES is
   the number of lines that the string can span.  If LINES is zero, there is no
   limit.  Return length of string including any quoting characters.  Return
   0 if input was invalid. */
long
read_quoted_string (char *start, char *terminator, int lines, char **output)
{
  long len;
  char *nl = 0, saved_char;

  if (lines)
    {
      int i;
      nl = start;
      for (i = 0; i < lines; i++)
        {
          nl = strchr (nl, '\n');
          if (!nl)
            break; /* End of input string reached. */
          nl++;
        }
      if (nl)
        {
          saved_char = *nl;
          *nl = '\0';
        }
    }

  if (start[0] != '\177')
    {
      len = strcspn (start, terminator);

      if (*terminator && !start[len])
        {
          len = 0;
          *output = 0;
        }
      else
        {
          *output = xmalloc (len + 1);
          strncpy (*output, start, len);
          (*output)[len] = '\0';
        }
    }
  else
    {
      len = strcspn (start + 1, "\177");

      if (*terminator && !(start + 1)[len])
        {
          /* No closing 177 byte. */
          len = 0;
          *output = 0;
        }
      else
        {
          *output = xmalloc (len + 1);
          strncpy (*output, start + 1, len);
          (*output)[len] = '\0';
          len += 2; /* Count the two 177 bytes. */
        }

    }

  if (nl)
    *nl = saved_char;
  return len;
}


/* **************************************************************** */
/*                                                                  */
/*                  Finding and Building Menus                      */
/*                                                                  */
/* **************************************************************** */

/* Get the entry associated with LABEL in the menu of NODE.  Return a
   pointer to the ENTRY if found, or null.  Return value should not
   be freed by caller.  If SLOPPY, allow initial matches, like
   "Buffers" for a LABEL "buffer". */
REFERENCE *
info_get_menu_entry_by_label (NODE *node, char *label, int sloppy) 
{
  register int i;
  int best_guess = -1;
  REFERENCE *entry;
  REFERENCE **references = node->references;

  if (!references)
    return 0;

  for (i = 0; (entry = references[i]); i++)
    {
      if (entry->type != REFERENCE_MENU_ITEM)
        continue;
      if (mbscasecmp (label, entry->label) == 0)
        return entry; /* Exact, case-insensitive match. */
      else if (sloppy && best_guess == -1
               && (mbsncasecmp (entry->label, label, strlen (label)) == 0))
        best_guess = i;
    }

  if (sloppy && best_guess != -1)
    return references[best_guess];

  return 0;
}

/* A utility function for concatenating REFERENCE **.  Returns a new
   REFERENCE ** which is the concatenation of REF1 and REF2.  */
REFERENCE **
info_concatenate_references (REFERENCE **ref1, REFERENCE **ref2)
{
  register int i, j;
  REFERENCE **result;
  int size = 0;

  /* Get the total size of the slots that we will need. */
  if (ref1)
    {
      for (i = 0; ref1[i]; i++);
      size += i;
    }

  if (ref2)
    {
      for (i = 0; ref2[i]; i++);
      size += i;
    }

  result = xmalloc ((1 + size) * sizeof (REFERENCE *));

  /* Copy the contents over. */

  j = 0;
  if (ref1)
    {
      for (i = 0; ref1[i]; i++)
        result[j++] = ref1[i];
    }

  if (ref2)
    {
      for (i = 0; ref2[i]; i++)
        result[j++] = ref2[i];
    }

  result[j] = NULL;
  return result;
}

/* Copy a reference structure.  Copy each field into new memory.  */
REFERENCE *
info_copy_reference (REFERENCE *src)
{
  REFERENCE *dest = xmalloc (sizeof (REFERENCE));
  dest->label = src->label ? xstrdup (src->label) : NULL;
  dest->filename = src->filename ? xstrdup (src->filename) : NULL;
  dest->nodename = src->nodename ? xstrdup (src->nodename) : NULL;
  dest->start = src->start;
  dest->end = src->end;
  dest->line_number = src->line_number;
  dest->type = src->type;
  
  return dest;
}

/* Copy a list of references, copying in reference in turn with
   info_copy_reference. */
REFERENCE **
info_copy_references (REFERENCE **ref1)
{
  int i;
  REFERENCE **result;
  int size;

  if (!ref1)
    return 0;

  /* Get the total size of the slots that we will need. */
  for (i = 0; ref1[i]; i++);
  size = i;

  result = xmalloc ((1 + size) * sizeof (REFERENCE *));

  /* Copy the contents over. */
  for (i = 0; ref1[i]; i++)
    result[i] = info_copy_reference (ref1[i]);
  result[i] = NULL;

  return result;
}

void
info_reference_free (REFERENCE *ref)
{
  if (ref)
    {
      free (ref->label);
      free (ref->filename);
      free (ref->nodename);
      free (ref);
    }
}

/* Free the data associated with REFERENCES. */
void
info_free_references (REFERENCE **references)
{
  register int i;
  REFERENCE *entry;

  if (references)
    {
      for (i = 0; references && (entry = references[i]); i++)
        info_reference_free (entry);

      free (references);
    }
}

/* Return new REFERENCE with filename and nodename fields set. */
REFERENCE *
info_new_reference (char *filename, char *nodename)
{
  REFERENCE *r = xmalloc (sizeof (REFERENCE));
  r->label = 0;
  r->filename = filename ? xstrdup (filename) : 0;
  r->nodename = nodename ? xstrdup (nodename) : 0;
  r->start = 0;
  r->end = 0;
  r->line_number = 0;
  r->type = 0;
  return r;
}


/* Search for sequences of whitespace or newlines in STRING, replacing
   all such sequences with just a single space.  Remove whitespace from
   start and end of string. */
void
canonicalize_whitespace (char *string)
{
  register int i, j;
  int len, whitespace_found, whitespace_loc = 0;
  char *temp;

  if (!string)
    return;

  len = strlen (string);
  temp = xmalloc (1 + len);

  /* Search for sequences of whitespace or newlines.  Replace all such
     sequences in the string with just a single space. */

  whitespace_found = 0;
  for (i = 0, j = 0; string[i]; i++)
    {
      if (whitespace_or_newline (string[i]))
        {
          whitespace_found++;
          whitespace_loc = i;
          continue;
        }
      else
        {
          if (whitespace_found && whitespace_loc)
            {
              whitespace_found = 0;

              /* Suppress whitespace at start of string. */
              if (j)
                temp[j++] = ' ';
            }

          temp[j++] = string[i];
        }
    }

  /* Kill trailing whitespace. */
  if (j && whitespace (temp[j - 1]))
    j--;

  temp[j] = '\0';
  strcpy (string, temp);
  free (temp);
}



/* **************************************************************** */
/*                                                                  */
/*                          Scanning node                           */
/*                                                                  */
/* **************************************************************** */

/* Whether to strip syntax from the text of nodes. */
int preprocess_nodes_p;

/* Whether contents of nodes should be rewritten. */
static int rewrite_p;

/* inptr is moved forward through the body of a node. */
static char *inptr;

/* Pointer to first byte of node (after node separator). */
static char *input_start;

/* Number of bytes in node contents. */
static size_t input_length;

struct text_buffer output_buf;

/* Pointer into a tags table for the file to the anchor we need to adjust as
   a result of byte counts changing due to character encoding conversion or
   inserted/deleted text. */
static TAG **anchor_to_adjust;
/* Offset within file buffer of first byte of node, used for anchor
   adjustment. */
static int node_offset;

/* Difference so far between the number of bytes input in the file and
   bytes output.  Used to adjust the values of anchors in nodes. */
static long int output_bytes_difference;

/* Whether we are converting the character encoding of the file. */
static int convert_encoding_p;

#if HAVE_ICONV

/* Whether text in file is encoded in UTF-8. */
static int file_is_in_utf8;

/* Used for conversion from file encoding to output encoding. */
static iconv_t iconv_to_output;

/* Conversion from file encoding to UTF-8. */
static iconv_t iconv_to_utf8;

#endif /* HAVE_ICONV */

void
init_conversion (FILE_BUFFER *fb)
{
  char *target_encoding;

  convert_encoding_p = 0;

  /* Node being processed does not come from an Info file. */
  if (!fb)
    return;

#if !HAVE_ICONV
  return;
#else
  file_is_in_utf8 = 0;

  /* Don't process file if encoding is unknown. */
  if (!fb->encoding)
    return;

  /* Read name of character encoding from environment locale */
  target_encoding = nl_langinfo (CODESET);

  /* Don't convert the contents if the locale
     uses the same character encoding as the file */
  if (!strcasecmp(target_encoding, fb->encoding))
    return;

  /* Check if an iconv conversion from file locale to system
     locale exists */
  iconv_to_output = iconv_open (target_encoding, fb->encoding);
  if (iconv_to_output == (iconv_t) -1)
    return; /* Return if no conversion function implemented */

  if (   !strcasecmp ("UTF8",  fb->encoding)
      || !strcasecmp ("UTF-8", fb->encoding))
    file_is_in_utf8 = 1;

  if (!file_is_in_utf8)
    {
      iconv_to_utf8 = iconv_open ("UTF-8", fb->encoding);
      if (iconv_to_utf8 == (iconv_t) -1)
        {
          /* Return if no conversion function implemented */
          iconv_close (iconv_to_output);
          iconv_to_output = (iconv_t) -1;
          return; 
        }
    }

  convert_encoding_p = 1;
  rewrite_p = 1;
#endif /* HAVE_ICONV */
}

void close_conversion (void)
{
#if HAVE_ICONV
  if (convert_encoding_p)
    {
      iconv_close (iconv_to_output);
      iconv_to_output = (iconv_t) -1;
      if (!file_is_in_utf8) iconv_close (iconv_to_utf8);
    }
#endif
}

static void
init_output_stream (FILE_BUFFER *fb)
{
  init_conversion (fb);
  output_bytes_difference = 0;

  if (rewrite_p)
    text_buffer_init (&output_buf);
}

static size_t saved_offset;
static char *saved_inptr;
static long saved_difference;

void
save_conversion_state (void)
{
  saved_offset = text_buffer_off (&output_buf);
  saved_inptr = inptr;
  saved_difference = output_bytes_difference;
}

/* Go back to the saved state of the output stream. */
void
reset_conversion (void)
{
  text_buffer_off (&output_buf) = saved_offset;
  inptr = saved_inptr;
  output_bytes_difference = saved_difference;
}

/* Copy bytes from input to output with no encoding conversion. */
static void
copy_direct (long n)
{
  text_buffer_add_string (&output_buf, inptr, n);
  inptr += n;
}

/* Read one character at *FROM and write out a sequence
   of bytes representing that character in ASCII.  *FROM
   is advanced past the read character. */
static int
degrade_utf8 (char **from, size_t *from_left)
{
  static struct encoding_replacement
  {
    char *from_string;
    char *to_string;
  } er[] = {
    {"\xE2\x80\x98","'"}, /* Opening single quote */
    {"\xE2\x80\x99","'"}, /* Closing single quote */
    {"\xE2\x80\x9C","\""},/* Opening double quote */
    {"\xE2\x80\x9D","\""},/* Closing double quote */
    {"\xC2\xA9","(C)"},   /* Copyright symbol */
    {"\xC2\xBB",">>"},    /* Closing double angle brackets */

    {"\xE2\x86\x92","->"},/* Right arrow */
    {"\xE2\x87\x92","=>"},/* Right double arrow */
    {"\xE2\x8A\xA3","-|"},/* Print symbol */
    {"\xE2\x98\x85","-!-"}, /* Point symbol */
    {"\xE2\x86\xA6","==>"}, /* Expansion symbol */

    {"\xE2\x80\x90","-"},  /* Hyphen */
    {"\xE2\x80\x91","-"},  /* Non-breaking hyphen */
    {"\xE2\x80\x92","-"},  /* Figure dash */
    {"\xE2\x80\x93","-"},  /* En dash */
    {"\xE2\x80\x94","--"},  /* Em dash */
    {"\xE2\x88\x92","-"},  /* Minus sign */
    {"\xE2\x80\xA6","..."},  /* Ellipsis */
    {"\xE2\x80\xA2","*"},  /* Bullet */

    {"\xC3\xA0","a`"},   /* Lower case letter a with grave accent */
    {"\xC3\xA2","a^"},   /* Lower case letter a with circumflex */
    {"\xC3\xA4","a\""},  /* Lower case letter a with diaeresis */
    {"\xC3\xA6","ae"},   /* Lower case letter ae ligature */
    {"\xC3\xA9","e'"},   /* Lower case letter e with acute accent */
    {"\xC3\xA8","e`"},   /* Lower case letter e with grave accent */
    {"\xC3\xAA","e^"},   /* Lower case letter e with circumflex */
    {"\xC3\xAB","e\""},  /* Lower case letter e with diaeresis */
    {"\xC3\xB6","o\""},  /* Lower case letter o with diaeresis */
    {"\xC3\xBC","u\""},  /* Lower case letter u with diaeresis */
    {"\xC3\x84", "A\""},  /* Upper case letter A with diaeresis. */
    {"\xC3\x96", "O\""},  /* Upper case letter O with diaeresis. */
    {"\xC3\x9c", "U\""},  /* Upper case letter U with diaeresis. */

    {"\xC3\xB1","n~"},  /* Lower case letter n with tilde */
    {"\xC3\x87","C,"},  /* Upper case letter C with cedilla */
    {"\xC3\xA7","c,"},  /* Lower case letter c with cedilla */
    {"\xC3\x9f","ss"},  /* Lower case letter sharp s */

    {0, 0}
  };

  struct encoding_replacement *erp;

  for (erp = er; erp->from_string != 0; erp++)
    {
      /* Avoid reading past end of input. */
      int width = strlen (erp->from_string);
      if (width > *from_left)
        continue;

      if (!strncmp (erp->from_string, *from, width))
        {
          text_buffer_add_string (&output_buf, erp->to_string,
                                  strlen(erp->to_string));
          *from += width;
          *from_left -= width;
          return 1;
        }
    }

  /* Failing this, just print a question mark.  Maybe we should use SUB
     (^Z) (ASCII substitute character code) instead, or pass through the
     original bytes. */
  text_buffer_add_string (&output_buf, "?", 1);

  /* Ideally we would advance one UTF-8 character.  This would
     require knowing its length in bytes. */
  (*from)++;
  (*from_left)--;

  return 0;
}

/* Convert N bytes from input to output encoding and write to
   output buffer.  Return number of bytes over N written. */
static int
copy_converting (long n)
{
#if !HAVE_ICONV
  return 0;
#else
  size_t bytes_left, orig_bytes_left;
  int extra_at_end;
  size_t iconv_ret;
  long output_start;

  size_t utf8_char_free; 
  char utf8_char[4]; /* Maximum 4 bytes in a UTF-8 character */
  char *utf8_char_ptr, *orig_inptr;
  size_t i;
  
  /* Use n as an estimate of how many bytes will be required
     in target encoding. */
  text_buffer_alloc (&output_buf, (size_t) n);

  output_start = text_buffer_off (&output_buf);
  bytes_left = n;
  extra_at_end = 0;
  while (1)
    {
      iconv_ret = text_buffer_iconv (&output_buf, iconv_to_output,
                                     (ICONV_CONST char **)&inptr, &bytes_left);

      /* Make sure libiconv flushes out the last converted character.
	 This is required when the conversion is stateful, in which
	 case libiconv might not output the last character, waiting to
	 see whether it should be combined with the next one.  */
      if (iconv_ret != (size_t) -1
	  && text_buffer_iconv (&output_buf, iconv_to_output,
				NULL, NULL) != (size_t) -1)
        /* Success: all of input converted. */
        break;

      /* There's been an error while converting. */
      switch (errno)
        {
        case EINVAL:
          /* Incomplete byte sequence at end of input buffer.  Try to read
             more. */

          /* input_length - 2 is offset of last-but-one byte within input.
             This checks if there is at least one more byte within node
             contents. */
          if (inptr - input_start + (bytes_left - 1) <= input_length - 2)
            {
              bytes_left++;
              extra_at_end++;
            }
          else
            {
              copy_direct (bytes_left);
              bytes_left = 0;
            }
          continue;
        default: /* Unknown error */
          info_error (_("Error converting file character encoding"));

          /* Skip past current input and hope we don't get an
             error next time. */
          inptr += bytes_left;
          return 0;
        case EILSEQ:
          /* Byte sequence in input not recognized.  Degrade to ASCII.  */
          break;
        }

      /* Flush any waiting input in iconv_to_output and enter the
         default shift state. */
      text_buffer_iconv (&output_buf, iconv_to_output, NULL, NULL);
      
      if (file_is_in_utf8)
        {
          degrade_utf8 (&inptr, &bytes_left);
          continue;     
        }

      /* If file is not in UTF-8, we degrade to ASCII in two steps:
         first convert the character to UTF-8, then look up a replacement
         string.  Note that mixing iconv_to_output and iconv_to_utf8
         on the same input may not work well if the input encoding
         is stateful.  We could deal with this by always converting to
         UTF-8 first; then we could mix conversions on the UTF-8 stream. */

      /* We want to read exactly one character.  Do this by
         restricting size of output buffer. */
      utf8_char_ptr = utf8_char;
      orig_inptr = inptr;
      orig_bytes_left = bytes_left;
      for (i = 1; i <= 4; i++)
        {
          utf8_char_free = i;
          errno = 0;
          iconv_ret = iconv (iconv_to_utf8, (ICONV_CONST char **)&inptr,
                             &bytes_left, &utf8_char_ptr, &utf8_char_free);
          if ((iconv_ret == (size_t) -1 && errno != E2BIG)
              /* If we managed to convert a character: */
              || utf8_char_ptr > utf8_char)
            break;
        }

      /* errno == E2BIG if iconv ran out of output buffer,
         which is expected. */
      if (iconv_ret == (size_t) -1 && errno != E2BIG)
	{
	  /* Character is not recognized.  Copy a single byte.  */
	  inptr = orig_inptr;	/* iconv might have incremented inptr  */
	  copy_direct (1);
	  bytes_left = orig_bytes_left - 1;
	}
      else
        {
          utf8_char_ptr = utf8_char;
          /* i is width of UTF-8 character */
          degrade_utf8 (&utf8_char_ptr, &i);
	  /* If we are done, make sure iconv flushes the last character.  */
	  if (bytes_left <= 0)
	    {
	      utf8_char_ptr = utf8_char;
	      i = 4;
	      iconv (iconv_to_utf8, NULL, NULL,
		     &utf8_char_ptr, &utf8_char_free);
	      if (utf8_char_ptr > utf8_char)
		{
		  utf8_char_ptr = utf8_char;
		  degrade_utf8 (&utf8_char_ptr, &i);
		}
	    }
        }
    }

  /* Must cast because the difference between unsigned size_t is always
     positive. */
  output_bytes_difference +=
    n - ((signed long) text_buffer_off (&output_buf) - output_start);

  return extra_at_end;
#endif /* HAVE_ICONV */
}

/* Functions below are named from the perspective of the preprocess_nodes_p
   flag being on. */

/* Copy text from input node contents, possibly converting the
   character encoding and adjusting anchor offsets at the same time. */
static void
copy_input_to_output (long n)
{
  if (rewrite_p)
    {
      long bytes_left;

      bytes_left = n;
      while (bytes_left > 0)
        {
          if (!convert_encoding_p)
            {
              copy_direct (bytes_left);
              bytes_left = 0;
            }
          else
            {
              long bytes_to_convert;
              long extra_written;

              bytes_to_convert = bytes_left;

              if (anchor_to_adjust)
                {
                  /* Check there is an anchor in the input. */
                  long first_anchor =
                    (*anchor_to_adjust)->nodestart - node_offset;

                  if (first_anchor < 0)
                    anchor_to_adjust = 0; /* error in input file */
                  else if (first_anchor < (inptr-input_start) + bytes_left)
                    {
                      /* Convert enough to pass the first anchor in input. */
                      bytes_to_convert = first_anchor - (inptr-input_start)+1;
                      if (bytes_to_convert < 0)
                        {
                          bytes_to_convert = bytes_left;
                          anchor_to_adjust = 0;
                        }
                    }
                }

              /* copy_converting may read more than bytes_to_convert
                 bytes if its input ends in an incomplete byte sequence. */
              extra_written = copy_converting (bytes_to_convert);

              bytes_left -= bytes_to_convert + extra_written;
            }

          /* Check if we have gone past any anchors and
             adjust with output_bytes_difference. */
          if (anchor_to_adjust)
            while ((*anchor_to_adjust)->nodestart - node_offset
                   <= inptr - input_start)
              {
                (*anchor_to_adjust)->nodestart_adjusted
                   = (*anchor_to_adjust)->nodestart - output_bytes_difference;

                anchor_to_adjust++;
                if (!*anchor_to_adjust
                    || (*anchor_to_adjust)->cache.nodelen != 0)
                  {
                    anchor_to_adjust = 0;
                    break;
                  }
              }
        }
    }
  else
    inptr += n;
}

static void
skip_input (long n)
{
  if (preprocess_nodes_p)
    {
      inptr += n;
      output_bytes_difference += n;
    }
  else if (rewrite_p)
    {
      /* We are expanding tags only.  Do not skip input. */
      copy_input_to_output (n);
    }
  else
    {
      inptr += n;
    }
}

static void
write_extra_bytes_to_output (char *input, long n)
{
  if (preprocess_nodes_p)
    {
      text_buffer_add_string (&output_buf, input, n);
      output_bytes_difference -= n;
    }
}

/* Like write_extra_bytes_to_output, but writes bytes even when
   preprocess_nodes=Off. */
static void
write_tag_contents (char *input, long n)
{
  if (rewrite_p)
    {
      text_buffer_add_string (&output_buf, input, n);
      output_bytes_difference -= n;
    }
}

/* Like skip_input, but skip even when !preprocess_nodes_p. */
static void
skip_tag_contents (long n)
{
  if (rewrite_p)
    {
      inptr += n;
      output_bytes_difference += n;
    }
}

/* Read first line of node and set next, prev and up. */
static void
parse_top_node_line (NODE *node)
{
  char **store_in = 0;
  char *nodename;
  char *ptr;
  int value_length;

  /* If the first line is empty, leave it in.  This is the case
     in the index-apropos window. */
  if (*node->contents == '\n')
    return;

  node->next = node->prev = node->up = 0;
  ptr = node->contents;

  while (1)
    {
      store_in = 0;

      ptr += skip_whitespace (ptr);

      /* Check what field we are looking at */
      if (!strncasecmp (ptr, INFO_FILE_LABEL, strlen(INFO_FILE_LABEL)))
        {
          ptr += strlen (INFO_FILE_LABEL);
        }
      else if (!strncasecmp (ptr, INFO_NODE_LABEL, strlen(INFO_NODE_LABEL)))
        {
          ptr += strlen (INFO_NODE_LABEL);
        }
      else if (!strncasecmp (ptr, INFO_PREV_LABEL, strlen(INFO_PREV_LABEL)))
        {
          ptr += strlen (INFO_PREV_LABEL);
          store_in = &node->prev;
        }
      else if (!strncasecmp (ptr, INFO_ALTPREV_LABEL, 
                             strlen(INFO_ALTPREV_LABEL)))
        {
          ptr += strlen (INFO_ALTPREV_LABEL);
          store_in = &node->prev;
        }
      else if (!strncasecmp (ptr, INFO_NEXT_LABEL, strlen(INFO_NEXT_LABEL)))
        {
          ptr += strlen (INFO_NEXT_LABEL);
          store_in = &node->next;
        }
      else if (!strncasecmp (ptr, INFO_UP_LABEL, strlen(INFO_UP_LABEL)))
        {
          ptr += strlen (INFO_UP_LABEL);
          store_in = &node->up;
        }
      else 
        {
          store_in = 0;
          /* Not recognized - code below will skip to next comma */
        }
      ptr += skip_whitespace (ptr);

      /* Get length of a bracketed filename component. */
      if (*ptr != '(')
        value_length = 0;
      else
        value_length = read_bracketed_filename (ptr, 0);

      /* Get length of node name, or filename if following "File:".  Note 
         that .  is not included in the second argument here in order to 
         support this character in file names. */
      value_length += read_quoted_string (ptr + value_length,
                                          "\n\r\t,", 1, &nodename);
      if (store_in)
        {
          *store_in = xmalloc (value_length + 1);
          strncpy (*store_in, ptr, value_length);
          (*store_in)[value_length] = '\0';
        }

      free (nodename);
      ptr += value_length;

      if (*ptr == '\n' || !*ptr)
        break;

      ptr += 1; /* Point after field terminator */
    }
}

/* Output, replace or hide text introducing a reference.  INPTR starts on
   the first byte of a sequence introducing a reference and finishes on the
   first (non-whitespace) byte of the reference label. */
static int
scan_reference_marker (REFERENCE *entry, int in_parentheses)
{
  /* When preprocess_nodes is Off, we position the cursor on
     the "*" when moving between references. */
  if (!preprocess_nodes_p)
    {
      if (rewrite_p)
        entry->start = text_buffer_off(&output_buf);
      else
        entry->start = inptr - input_start;
    }

  /* Check what we found based on first character of match */
  if (inptr[0] == '\n')
    {
      entry->type = REFERENCE_MENU_ITEM;
      if (!preprocess_nodes_p)
        entry->start++;
    }
  else
    entry->type = REFERENCE_XREF;

  if (entry->type == REFERENCE_MENU_ITEM)
    copy_input_to_output (strlen ("\n* "));
  else
    {
      /* Only match "*Note" if it is followed by a whitespace character so that 
         it will not be recognized if, e.g., it is surrounded in inverted 
         commas. */
      if (!strchr (" \t\r\n", inptr[strlen ("*Note")]))
        {
          copy_input_to_output (strlen ("*Note:"));
          return 0;
        }

      /* Cross-references can be generated by four different Texinfo
         commands.  @inforef and @xref output "*Note " in Info format,
         and "See" in HTML and print.  @ref and @pxref output "*note "
         in Info format, and either nothing at all or "see" in HTML
         and print.  Unfortunately, there is no easy way to distinguish
         between these latter two cases. */
      /* TODO: Internationalize these strings, but only if we know the
         language of the document. */
      if (inptr[1] == 'N')
        {
          write_extra_bytes_to_output ("See", 3);
          in_parentheses = 1;
        }
      else if (in_parentheses)
        {
          write_extra_bytes_to_output ("see", 3);
          /* Only output the "see" for input like "(*note ...)", which
             would have come from a use of @pxref.  We used to output "see" for 
             "*note" in more circumstances, with a list of words where to
             suppress it (to avoid "see *note" turning into "see see"), but
             such a list can't be complete or reliable.  It's better to remove 
             it with more enthusiasm, then if the document writer wants a "see"
             to appear, they can add one themselves. */
        }

      skip_input (strlen ("*Note"));
      if (!in_parentheses)
        skip_input (skip_whitespace (inptr));
    }

  /* Copy any white space before label. */
  copy_input_to_output (skip_whitespace_and_newlines (inptr));

  return 1;
}

/* Output reference label and update ENTRY.  INPTR should be on the first
   non-whitespace byte of label when this function is called.  It is left
   at the first character after the colon terminating the label.  Return 0 if
   invalid syntax is encountered. */
static int
scan_reference_label (REFERENCE *entry, int in_index)
{
  int max_lines;
  int len, label_len = 0;

  /* Handle case of cross-reference like (FILE)NODE::. */
  if (inptr[0] == '(' && !in_index)
    label_len = read_bracketed_filename (inptr, &entry->filename);

  /* Search forward to ":" to get label name.  Cross-references may have
     a newline in the middle. */
  if (entry->type == REFERENCE_MENU_ITEM)
    max_lines = 1;
  else
    max_lines = 2;
  if (!in_index || inptr[label_len] == '\177')
    {
      len = read_quoted_string (inptr + label_len, ":", max_lines,
                                &entry->nodename);
      canonicalize_whitespace (entry->nodename);
      if (!len)
        return 0; /* Input invalid. */
      label_len += len;
    }
  else
    {
      /* If in an index node, go forward to the last colon on the line
         (not preceded by a newline, NUL or DEL).  This is in order to
         support index entries containing colons.  This should work fine
         as long as the node name does not contain a colon as well. */

      char *p;
      int n, m = 0;
      p = inptr + label_len;

      while (1)
        {
          n = strcspn (p, ":\n\177");
          if (p[n] == ':')
            {
              m += n + 1;
              p += n + 1;
              continue;
            }
          break;
        }
      if (m == 0)
        return 0; /* no : found */
      label_len += m - 1;
    }

#if HAVE_ICONV
  if (iconv_to_output != (iconv_t) -1 && iconv_to_output != (iconv_t) 0)
    {
      static struct text_buffer label_text;
      size_t iconv_ret;
      size_t inbytesleft = label_len;
      char *p = inptr;
      text_buffer_reset (&label_text);
      text_buffer_alloc (&label_text, label_len);

      while (1)
        {
          iconv_ret = text_buffer_iconv (&label_text, iconv_to_output,
                                         (ICONV_CONST char **)&p,
                                         &inbytesleft);

          /* Make sure libiconv flushes out the last converted character. */
          if (iconv_ret != (size_t) -1
                && text_buffer_iconv (&label_text, iconv_to_output,
                       NULL, NULL) != (size_t) -1)
            break; /* Success: all of input converted. */

          /* There's been an error while converting. */
          goto no_convert;
        }

      text_buffer_add_char (&label_text, '\0');
      entry->label = strdup (label_text.base);
    }
  else
#endif
    {
  no_convert:
      entry->label = xmalloc (label_len + 1);
      memcpy (entry->label, inptr, label_len);
      entry->label[label_len] = '\0';
    }
  canonicalize_whitespace (entry->label);

  if (preprocess_nodes_p)
    entry->start = text_buffer_off (&output_buf);

  /* Write text of label. */
  copy_input_to_output (label_len);

  if (rewrite_p)
    entry->end = text_buffer_off (&output_buf);
  else
    entry->end = inptr - input_start;

  /* Colon after label. */
  if (*inptr)
    skip_input (1);
  /* Don't mess up the margin of a menu description. */
  if (entry->type == REFERENCE_MENU_ITEM)
    write_extra_bytes_to_output (" ", 1);

  return 1;
}

/* INPTR should be at the first character after the colon
   terminating the label.  Return 0 on syntax error. */
static int
scan_reference_target (REFERENCE *entry, NODE *node, int in_parentheses)
{
  int i;

  /* This entry continues with a specific target.  Parse the
     file name and node name from the specification. */

  if (entry->type == REFERENCE_XREF)
    {
      int length = 0; /* Length of specification */
      char *target_start = inptr;
      char *nl_off = 0;
      int space_at_start_of_line = 0;

      length += skip_whitespace_and_newlines (inptr);

      length += read_bracketed_filename (inptr + length, &entry->filename);

      length += skip_whitespace_and_newlines (inptr + length);

      /* Get the node name. */
      length += read_quoted_string (inptr + length, ",.", 2, &entry->nodename);

      skip_input (length);

      /* Check if there is a newline in the target. */
      nl_off = strchr (target_start, '\n');
      if (nl_off)
        {
          if (nl_off < inptr)
            space_at_start_of_line = skip_whitespace (nl_off + 1);
          else
            nl_off = 0;
        }
      canonicalize_whitespace (entry->nodename);

      if (entry->filename)
        {
          /* Heuristic of whether it's worth outputing a newline before the
             filename.  This checks whether the newline appears more
             than half way through the text, and therefore which side is
             longer. */
          if (nl_off
              && nl_off < target_start + (length - space_at_start_of_line) / 2)
            {
              int i;
              write_extra_bytes_to_output ("\n", 1);

              for (i = 0; i < space_at_start_of_line; i++)
                write_extra_bytes_to_output (" ", 1);
              skip_input (strspn (inptr, " "));
              nl_off = 0;
            }
          else

          if (*inptr != '\n')
            {
              write_extra_bytes_to_output (" ", 1);
            }
          write_extra_bytes_to_output ("(", 1);
          write_extra_bytes_to_output (entry->filename,
                                       strlen (entry->filename));
          write_extra_bytes_to_output (" manual)",
                                       strlen (" manual)"));
        }
      
      /* Hide terminating punctuation if we are in a reference
         like "(*note Label:(file)node.)". */
      if (in_parentheses && inptr[0] == '.')
        skip_input (1);

      /* Copy any terminating punctuation before the optional newline. */
      copy_input_to_output (strspn (inptr, ".),"));

      /* Output a newline if one is needed.  Don't do it at the end of
         a paragraph. */
      if (nl_off && *inptr != '\n')
        { 
          int i;

          write_extra_bytes_to_output ("\n", 1);
          for (i = 0; i < space_at_start_of_line; i++)
            write_extra_bytes_to_output (" ", 1);
          skip_input (strspn (inptr, " "));
        }
    }
  else /* entry->type == REFERENCE_MENU_ITEM */
    {
      int line_len;
      int length = 0; /* Length of specification */

      length = skip_whitespace (inptr);
      length += read_bracketed_filename (inptr + length, &entry->filename);
      length += strspn (inptr + length, " ");

      /* Get the node name. */
      length += read_quoted_string (inptr + length, ",.\t\n", 2, 
                                    &entry->nodename);
      if (inptr[length] == '.') /* A '.' terminating the entry. */
        length++;
      canonicalize_whitespace (entry->nodename);

      if (node->flags & N_IsDir)
        {
          /* Set line_len to length of line so far. */

          char *linestart;
          linestart = memrchr (input_start, '\n', inptr - input_start);
          if (!linestart)
            linestart = input_start;
          else
            linestart++; /* Point to first character after newline. */
          line_len = inptr - linestart;
        }

      if (node->flags & N_IsIndex)
        /* Show the name of the node the index entry refers to. */
        copy_input_to_output (length);
      else
        {
          skip_input (length);

          if ((node->flags & N_IsDir) && inptr[strspn (inptr, " ")] == '\n')
            {
              /* For a dir node, if there is no more text in this line,
                 check if there is a menu entry description in the next
                 line to the right of the end of the label, and display it
                 in this line. */
              skip_input (strspn (inptr, " "));
              if (line_len <= strspn (inptr + 1, " "))
                skip_input (1 + line_len);
            }
          else
            {
              for (i = 0; i < length; i++)
                write_extra_bytes_to_output (" ", 1);
            }
        }

      /* Parse "(line ...)" part of menus, if any.  */
      {
        char *lineptr = inptr;
        /* Skip any whitespace first, and then a newline in case the item
           was so long to contain the ``(line ...)'' string in the same
           physical line.  */
        lineptr += skip_whitespace (inptr);
        if (*lineptr == '\n')
          lineptr += 1 + skip_whitespace (lineptr + 1);

        if (!strncmp (lineptr, "(line ", strlen ("(line ")))
          {
            lineptr += strlen ("(line ");
            entry->line_number = strtol (lineptr, 0, 0);
          }
        else
          entry->line_number = 0;
      }
    }

  return 1;
}

/* BASE is earlier in a block of allocated memory than PTR, and the block
   extends until at least BASE + LEN - 1.  Return PTR[INDEX], unless this
   could be outside the allocated block, in which case return 0. */
static char
safe_string_index (char *ptr, long index, char *base, long len)
{
  long offset = ptr - base;

  if (   offset + index < 0
      || offset + index >= len)
    return 0;

  return ptr[index];
}

/* Process an in index marker ("^@^H[index^@^H]") or an image marker
   ("^@^H[image ...^@^H]"). */
static void
scan_info_tag (NODE *node, int *in_index, FILE_BUFFER *fb)
{
  char *p, *p1;
  struct text_buffer *expansion = xmalloc (sizeof (struct text_buffer));

  p = inptr;
  p1 = p;

  text_buffer_init (expansion);

  if (tag_expand (&p1, input_start + input_length, expansion, in_index))
    {
      if (*in_index)
        node->flags |= N_IsIndex;

      if (!rewrite_p)
        {
          rewrite_p = 1;
          init_output_stream (fb);

          /* Put inptr back to start so that
             copy_input_to_output below gets all
             preceding contents. */
          inptr = node->contents;
        }

      /* Write out up to tag. */
      copy_input_to_output (p - inptr);

      write_tag_contents (text_buffer_base (expansion),
                          text_buffer_off (expansion));
      /* Skip past body of tag. */
      skip_tag_contents (p1 - inptr);
    }
  else
    {
      /* It was not a valid tag. */ 
      copy_input_to_output (p - inptr + 1);
    }

  text_buffer_free (expansion);
  free (expansion);
}

#define looking_at_string(contents, string) \
  (!strncasecmp (contents, string, strlen (string)))

static char *
forward_to_info_syntax (char *contents)
{
  /* Loop until just before the end of the input.  The '- 3' prevents us
     accessing memory after the end of the input, and none of the strings we 
     are looking for are shorter than 3 bytes. */
  while (contents < input_start + input_length - 3)
    {
      /* Menu entry comes first to optimize for the case of looking through a 
         long index node. */
      if (looking_at_string (contents, INFO_MENU_ENTRY_LABEL)
          || looking_at_string (contents, INFO_XREF_LABEL)
          || !memcmp (contents, "\0\b[", 3))
        return contents;
      contents++;
    }
  return 0;
}

/* Scan contents of NODE, recording cross-references and similar.

   Convert character encoding of node contents to that of the user if the two 
   are known to be different.  If PREPROCESS_NODES_P == 1, remove Info syntax 
   in contents.

   If FB is non-null, it is the file containing the node, and TAG_PTR is an 
   offset into FB->tags.  If the node contents are rewritten, adjust anchors
   that occur in the node and store adjusted value as TAG->nodestart_adjusted, 
   otherwise simply copy TAG->nodestart to TAG->nodestart_adjusted for each 
   anchor in the node. */
void
scan_node_contents (NODE *node, FILE_BUFFER *fb, TAG **tag_ptr)
{
  int in_menu = 0;
  char *match;

  REFERENCE **refs = NULL;
  size_t refs_index = 0, refs_slots = 0;

  /* Whether an index tag was seen. */
  int in_index = 0;

  rewrite_p = preprocess_nodes_p;

  init_output_stream (fb);

  if (fb)
    {
      char *file_contents;

      /* Set anchor_to_adjust to first anchor in node, if any. */
      anchor_to_adjust = tag_ptr + 1;
      if (!*anchor_to_adjust)
        anchor_to_adjust = 0;
      else if (*anchor_to_adjust
               && (*anchor_to_adjust)->cache.nodelen != 0)
        anchor_to_adjust = 0;

      if (!node->subfile)
        file_contents = fb->contents;
      else
        {
          FILE_BUFFER *f = info_find_subfile (node->subfile);
          if (!f)
            return; /* This shouldn't happen. */
          file_contents = f->contents;
        }
      node_offset = (*tag_ptr)->nodestart
        + skip_node_separator (file_contents + (*tag_ptr)->nodestart);
    }
  else
    anchor_to_adjust = 0;

  /* Initialize refs to point to array of one null pointer in case
     there are no results.  This way we know if refs has been initialized
     even if it is empty. */
  refs = calloc (1, sizeof *refs);
  refs_slots = 1;

  parse_top_node_line (node);

  /* This should be the only time we assign to inptr in this function -
     all other assignment should be done with the helper functions above. */
  inptr = node->contents;
  input_start = node->contents;
  input_length = node->nodelen;


  while ((match = forward_to_info_syntax (inptr))
          && match < node->contents + node->nodelen)
    {
      int in_parentheses = 0;
      REFERENCE *entry;

      /* Write out up to match */
      copy_input_to_output (match - inptr); 

      if ((in_menu && match[0] == '\n') || match[0] == '*')
        {
          /* Menu entry or cross reference. */
          /* Create REFERENCE entity. */
          entry = info_new_reference (0, 0);

          if (safe_string_index (inptr, -1, input_start, input_length) == '('
             && safe_string_index (inptr, 1, input_start, input_length) == 'n')
            in_parentheses = 1;

          save_conversion_state ();
          
          if (!scan_reference_marker (entry, in_parentheses))
            goto not_a_reference;

          if (!scan_reference_label (entry, in_index))
            goto not_a_reference;

          /* If this reference entry continues with another ':' then the target
             of the reference is given by the label. */
          if (*inptr == ':')
            {
              int label_len;
              skip_input (1);
              if (entry->type == REFERENCE_MENU_ITEM)
                write_extra_bytes_to_output (" ", 1);

              /* Remove the DEL bytes from a label like "(FOO)^?BAR^?::". */
              label_len = strlen (entry->label);
              if (label_len >= 2 && entry->label[label_len - 1] == 0177)
                {
                  char *p = strchr (entry->label, '\177');
                  memmove (p, p + 1, label_len - (p - entry->label) - 1);
                  entry->label[label_len - 2] = '\0';
                }
            }
          else
            {
              /* Proceed to read the rest of the reference. */
              /* TODO: we should probably not allow references of the form 
                 "(file)node1:node2." or "(file1)node1:(file2)node2", so
                 bail out here if entry->filename is non-null. */

              free (entry->filename); entry->filename = 0;
              free (entry->nodename); entry->nodename = 0;
              if (!scan_reference_target (entry, node, in_parentheses))
                goto not_a_reference;
            }

          if (0)
            {
              char *cur_inptr;

not_a_reference:
              /* This is not a menu entry or reference.  Do not add to our 
                 list. */
              cur_inptr = inptr;
              reset_conversion ();
              copy_input_to_output (cur_inptr - inptr);

              info_reference_free (entry);
              continue;
            }

          add_pointer_to_array (entry, refs_index, refs, refs_slots, 50);
        }
      /* Was "* Menu:" seen?  If so, search for menu entries hereafter. */
      else if (!in_menu && !strncmp (match, INFO_MENU_LABEL,
                               strlen (INFO_MENU_LABEL)))
        {
          in_menu = 1;
          skip_input (strlen ("\n* Menu:"));
          if (*inptr == '\n')
            skip_input (strspn (inptr, "\n") - 1); /* Keep one newline. */

        }
      else if (match[0] == '\0') /* Info tag */
        {
          scan_info_tag (node, &in_index, fb);
        }
      else
        copy_input_to_output (1);
    }

  /* If we haven't accidentally gone past the end of the node, write
     out the rest of it. */
  if (inptr < node->contents + node->nodelen)
    copy_input_to_output ((node->contents + node->nodelen) - inptr); 

  /* Null to terminate buffer. */
  if (rewrite_p)
    text_buffer_add_string (&output_buf, "\0", 1);

  /* Free resources used in character encoding conversion. */
  close_conversion ();
  
  node->references = refs;

  if (rewrite_p)
    {
      if (node->flags & N_WasRewritten)
        free (node->contents);
      node->contents = text_buffer_base (&output_buf);
      node->flags |= N_WasRewritten;
 
      /* output_buf.off is the offset of the next character to be
         written.  Subtracting 1 gives the offset of our terminating
         null, that is, the length. */
      node->nodelen = text_buffer_off (&output_buf) - 1;
    }
  else if (fb && tag_ptr)
    {
      /* Set nodestart_adjusted for all of the anchors in this node. */
      tag_ptr++;
      while (*tag_ptr && (*tag_ptr)->cache.nodelen == 0)
        {
          (*tag_ptr)->nodestart_adjusted = (*tag_ptr)->nodestart
                                             - output_bytes_difference;
          tag_ptr++;
        }
    }
}