summaryrefslogtreecommitdiff
path: root/makeinfo/html.c
blob: 2d25d3d70e8e1d5c0ffdfa91108f4a850a31c7cf (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
/* html.c -- html-related utilities.
   $Id: html.c,v 1.42 2008/05/19 18:26:47 karl Exp $

   Copyright (C) 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
   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/>.  */

#include "system.h"
#include "cmds.h"
#include "files.h"
#include "html.h"
#include "lang.h"
#include "makeinfo.h"
#include "node.h"
#include "sectioning.h"



/* Filename to which to write list of index entries */
char *internal_links_filename = NULL;
FILE *internal_links_stream = NULL;

/* Append CHAR to BUFFER, (re)allocating as necessary.  We don't handle
   null characters.  */

typedef struct
{
  unsigned size;    /* allocated */
  unsigned length;  /* used */
  char *buffer;
} buffer_type;

static buffer_type *
init_buffer (void)
{
  buffer_type *buf = xmalloc (sizeof (buffer_type));
  buf->length = 0;
  buf->size = 0;
  buf->buffer = NULL;

  return buf;
}

static void
append_char (buffer_type *buf, int c)
{
  buf->length++;
  if (buf->length >= buf->size)
    {
      buf->size += 100;
      buf->buffer = xrealloc (buf->buffer, buf->size);
    }
  buf->buffer[buf->length - 1] = c;
  buf->buffer[buf->length] = 0;
}

/* Read the cascading style-sheet file FILENAME.  Write out any @import
   commands, which must come first, by the definition of css.  If the
   file contains any actual css code following the @imports, return it;
   else return NULL.  */
static char *
process_css_file (char *filename)
{
  int c;
  int lastchar = 0;
  FILE *f;
  buffer_type *import_text = init_buffer ();
  buffer_type *inline_text = init_buffer ();
  unsigned lineno = 1;
  enum { null_state, comment_state, import_state, inline_state } state
    = null_state, prev_state;

  prev_state = null_state;

  /* read from stdin if `-' is the filename.  */
  f = STREQ (filename, "-") ? stdin : fopen (filename, "r");
  if (!f)
    {
      error (_("%s: could not open --css-file: %s"), progname, filename);
      return NULL;
    }

  /* Read the file.  The @import statements must come at the beginning,
     with only whitespace and comments allowed before any inline css code.  */
  while ((c = getc (f)) >= 0)
    {
      if (c == '\n')
        lineno++;

      switch (state)
        {
        case null_state: /* between things */
          if (c == '@')
            { /* Only @import and @charset should switch into
                 import_state, other @-commands, such as @media, should
                 put us into inline_state.  I don't think any other css
                 @-commands start with `i' or `c', although of course
                 this will break when such a command is defined.  */
              int nextchar = getc (f);
              if (nextchar == 'i' || nextchar == 'c')
                {
                  append_char (import_text, c);
                  append_char (import_text, nextchar);
                  state = import_state;
                }
              else
                {
                  ungetc (nextchar, f);  /* wasn't an @import */
                  state = inline_state;
                }
            }
          else if (c == '/')
            { /* possible start of a comment */
              int nextchar = getc (f);
              if (nextchar == '*')
                state = comment_state;
              else
                {
                  ungetc (nextchar, f); /* wasn't a comment */
                  state = inline_state;
                }
            }
          else if (isspace (c))
            ; /* skip whitespace; maybe should use c_isspace?  */

          else
            /* not an @import, not a comment, not whitespace: we must
               have started the inline text.  */
            state = inline_state;

          if (state == inline_state)
            append_char (inline_text, c);

          if (state != null_state)
            prev_state = null_state;
          break;

        case comment_state:
          if (c == '/' && lastchar == '*')
            state = prev_state;  /* end of comment */
          break;  /* else ignore this comment char */

        case import_state:
          append_char (import_text, c);  /* include this import char */
          if (c == ';')
            { /* done with @import */
              append_char (import_text, '\n');  /* make the output nice */
              state = null_state;
              prev_state = import_state;
            }
          break;

        case inline_state:
          /* No harm in writing out comments, so don't bother parsing
             them out, just append everything.  */
          append_char (inline_text, c);
          break;
        }

      lastchar = c;
    }

  fclose (f);  /* Even closing stdin should be ok, can't read it more
                  than once?  */
  
  /* Reached the end of the file.  We should not be still in a comment.  */
  if (state == comment_state)
    warning (_("%s:%d: --css-file ended in comment"), filename, lineno);

  /* Write the @import text, if any.  */
  if (import_text->buffer)
    {
      add_word (import_text->buffer);
      free (import_text->buffer);
      free (import_text);
    }

  /* We're wasting the buffer struct memory, but so what.  */
  return inline_text->buffer;
}

HSTACK *htmlstack = NULL;

/* See html.h.  */
int html_title_written = 0;

void
html_output_head (void)
{
  static const char *html_title = NULL;
  char *encoding = current_document_encoding ();

  /* The <title> should not have markup, so use text_expansion.  */
  if (!html_title)
    html_title = escape_string (title ?
        text_expansion (title) : (char *) gdt("Untitled"));

  /* Make sure this is the very first string of the output document.  */
  output_paragraph_offset = 0;

  add_html_block_elt_args ("<html lang=\"%s\">\n<head>\n",
      language_table[language_code].abbrev);

  /* When splitting, add current node's name to title if it's available
     and not Top.  */
  if (splitting && current_node && !STREQ (current_node, "Top"))
    add_word_args ("<title>%s - %s</title>\n",
        escape_string (xstrdup (current_node)), html_title);
  else
    add_word_args ("<title>%s</title>\n",  html_title);

  add_word ("<meta http-equiv=\"Content-Type\" content=\"text/html");
  if (encoding && *encoding)
    add_word_args ("; charset=%s", encoding);
                   
  add_word ("\">\n");

  if (!document_description)
    document_description = html_title;

  add_word_args ("<meta name=\"description\" content=\"%s\">\n",
                 document_description);
  add_word_args ("<meta name=\"generator\" content=\"makeinfo %s\">\n",
                 VERSION);

  /* Navigation bar links.  */
  if (!splitting)
    add_word ("<link title=\"Top\" rel=\"top\" href=\"#Top\">\n");
  else if (tag_table)
    {
      /* Always put a top link.  */
      add_word ("<link title=\"Top\" rel=\"start\" href=\"index.html#Top\">\n");

      /* We already have a top link, avoid duplication.  */
      if (tag_table->up && !STREQ (tag_table->up, "Top"))
        add_link (tag_table->up, "rel=\"up\"");

      if (tag_table->prev)
        add_link (tag_table->prev, "rel=\"prev\"");

      if (tag_table->next)
        add_link (tag_table->next, "rel=\"next\"");

      /* fixxme: Look for a way to put links to various indices in the
         document.  Also possible candidates to be added here are First and
         Last links.  */
    }
  else
    {
      /* We are splitting, but we neither have a tag_table.  So this must be
         index.html.  So put a link to Top. */
      add_word ("<link title=\"Top\" rel=\"start\" href=\"#Top\">\n");
    }

  add_word ("<link href=\"http://www.gnu.org/software/texinfo/\" \
rel=\"generator-home\" title=\"Texinfo Homepage\">\n");

  if (copying_text)
    { /* It is not ideal that we include the html markup here within
         <head>, so we use text_expansion.  */
      insert_string ("<!--\n");
      insert_string (text_expansion (copying_text));
      insert_string ("-->\n");
    }

  /* Put the style definitions in a comment for the sake of browsers
     that don't support <style>.  */
  add_word ("<meta http-equiv=\"Content-Style-Type\" content=\"text/css\">\n");
  add_word ("<style type=\"text/css\"><!--\n");

  {
    char *css_inline = NULL;

    if (css_include)
      /* This writes out any @import commands from the --css-file,
         and returns any actual css code following the imports.  */
      css_inline = process_css_file (css_include);

    /* This seems cleaner than adding <br>'s at the end of each line for
       these "roman" displays.  It's hardly the end of the world if the
       browser doesn't do <style>s, in any case; they'll just come out in
       typewriter.  */
#define CSS_FONT_INHERIT "font-family:inherit"
    add_word_args ("  pre.display { %s }\n", CSS_FONT_INHERIT);
    add_word_args ("  pre.format  { %s }\n", CSS_FONT_INHERIT);

    /* Alternatively, we could do <font size=-1> in insertion.c, but this
       way makes it easier to override.  */
#define CSS_FONT_SMALLER "font-size:smaller"
    add_word_args ("  pre.smalldisplay { %s; %s }\n", CSS_FONT_INHERIT,
                   CSS_FONT_SMALLER);
    add_word_args ("  pre.smallformat  { %s; %s }\n", CSS_FONT_INHERIT,
                   CSS_FONT_SMALLER);
    add_word_args ("  pre.smallexample { %s }\n", CSS_FONT_SMALLER);
    add_word_args ("  pre.smalllisp    { %s }\n", CSS_FONT_SMALLER);

    /* Since HTML doesn't have a sc element, we use span with a bit of
       CSS spice instead.  */
#define CSS_FONT_SMALL_CAPS "font-variant:small-caps"
    add_word_args ("  span.sc    { %s }\n", CSS_FONT_SMALL_CAPS);

    /* Roman (default) font class, closest we can come.  */
#define CSS_FONT_ROMAN "font-family:serif; font-weight:normal;"
    add_word_args ("  span.roman { %s } \n", CSS_FONT_ROMAN);

    /* Sans serif font class.  */
#define CSS_FONT_SANSSERIF "font-family:sans-serif; font-weight:normal;"
    add_word_args ("  span.sansserif { %s } \n", CSS_FONT_SANSSERIF);

    /* Write out any css code from the user's --css-file.  */
    if (css_inline)
      insert_string (css_inline);

    add_word ("--></style>\n");
  }
  if (css_ref)
    add_word_args ("<link rel=\"stylesheet\" type=\"text/css\" href=\"%s\">\n",
	            css_ref);
  
  add_word ("</head>\n<body>\n");

  if (title && !html_title_written && titlepage_cmd_present)
    {
      add_word_args ("<h1 class=\"settitle\">%s</h1>\n", html_title);
      html_title_written = 1;
    }

  free (encoding);
}

/* Escape HTML special characters in the string if necessary,
   returning a pointer to a possibly newly-allocated one. */
char *
escape_string (char *string)
{
  char *newstring;
  int i = 0, newlen = 0;

  do
    {
      /* Find how much to allocate. */
      switch (string[i])
        {
        case '"':
          newlen += 6;          /* `&quot;' */
          break;
        case '&':
          newlen += 5;          /* `&amp;' */
          break;
        case '<':
        case '>':
          newlen += 4;          /* `&lt;', `&gt;' */
          break;
        default:
          newlen++;
        }
    }
  while (string[i++]);

  if (newlen == i) return string; /* Already OK. */

  newstring = xmalloc (newlen);
  i = 0;
  do
    {
      switch (string[i])
        {
        case '"':
          strcpy (newstring, "&quot;");
          newstring += 6;
          break;
        case '&':
          strcpy (newstring, "&amp;");
          newstring += 5;
          break;
        case '<':
          strcpy (newstring, "&lt;");
          newstring += 4;
          break;
        case '>':
          strcpy (newstring, "&gt;");
          newstring += 4;
          break;
        default:
          newstring[0] = string[i];
          newstring++;
        }
    }
  while (string[i++]);

  return newstring - newlen;
}

/* Save current tag.  */
static void
push_tag (char *tag, char *attribs)
{
  HSTACK *newstack = xmalloc (sizeof (HSTACK));

  newstack->tag = tag;
  newstack->attribs = xstrdup (attribs);
  newstack->next = htmlstack;
  htmlstack = newstack;
}

/* Get last tag.  */
static void
pop_tag (void)
{
  HSTACK *tos = htmlstack;

  if (!tos)
    {
      line_error (_("[unexpected] no html tag to pop"));
      return;
    }

  free (htmlstack->attribs);

  htmlstack = htmlstack->next;
  free (tos);
}

/* Check if tag is an empty or a whitespace only element.
   If so, remove it, keeping whitespace intact.  */
int
rollback_empty_tag (char *tag)
{
  int check_position = output_paragraph_offset;
  int taglen = strlen (tag);
  int rollback_happened = 0;
  char *contents = "";			/* FIXME (ptr to constant, later
  					   assigned to malloc'd address).
					 */
  char *contents_canon_white = "";

  /* If output_paragraph is empty, we cannot rollback :-\  */
  if (output_paragraph_offset <= 0)
    return 0;

  /* Find the end of the previous tag.  */
  while (check_position > 0 && output_paragraph[check_position-1] != '>')
    check_position--;

  /* Save stuff between tag's end to output_paragraph's end.  */
  if (check_position != output_paragraph_offset)
    {
      contents = xmalloc (output_paragraph_offset - check_position + 1);
      memcpy (contents, output_paragraph + check_position,
          output_paragraph_offset - check_position);

      contents[output_paragraph_offset - check_position] = '\0';

      contents_canon_white = xstrdup (contents);
      canon_white (contents_canon_white);
    }

  /* Find the start of the previous tag.  */
  while (check_position > 0 && output_paragraph[check_position-1] != '<')
    check_position--;

  /* Check to see if this is the tag.  */
  if (strncmp ((char *) output_paragraph + check_position, tag, taglen) == 0
      && (whitespace (output_paragraph[check_position + taglen])
          || output_paragraph[check_position + taglen] == '>'))
    {
      if (!contents_canon_white || !*contents_canon_white)
        {
          /* Empty content after whitespace removal, so roll it back.  */
          output_paragraph_offset = check_position - 1;
          rollback_happened = 1;

          /* Original contents may not be empty (whitespace.)  */
          if (contents && *contents)
            {
              insert_string (contents);
              free (contents);
            }
        }
    }

  return rollback_happened;
}

/* Open or close TAG according to START_OR_END. */
void
#if defined (VA_FPRINTF) && __STDC__
insert_html_tag_with_attribute (int start_or_end, char *tag, char *format, ...)
#else
insert_html_tag_with_attribute (start_or_end, tag, format, va_alist)
     int start_or_end;
     char *tag;
     char *format;
     va_dcl
#endif
{
  char *old_tag = NULL;
  char *old_attribs = NULL;
  char formatted_attribs[2000]; /* xx no fixed limits */
  int do_return = 0;
  extern int in_html_elt;

  if (start_or_end != START)
    pop_tag ();

  if (htmlstack)
    {
      old_tag = htmlstack->tag;
      old_attribs = htmlstack->attribs;
    }
  
  if (format)
    {
#ifdef VA_SPRINTF
      va_list ap;
#endif

      VA_START (ap, format);
#ifdef VA_SPRINTF
      VA_SPRINTF (formatted_attribs, format, ap);
#else
      sprintf (formatted_attribs, format, a1, a2, a3, a4, a5, a6, a7, a8);
#endif
      va_end (ap);
    }
  else
    formatted_attribs[0] = '\0';

  /* Exception: can nest multiple spans.  */
  if (htmlstack
      && STREQ (htmlstack->tag, tag)
      && !(STREQ (tag, "span") && STREQ (old_attribs, formatted_attribs)))
    do_return = 1;

  if (start_or_end == START)
    push_tag (tag, formatted_attribs);

  if (do_return)
    return;

  in_html_elt++;

  /* texinfo.tex doesn't support more than one font attribute
     at the same time.  */
  if ((start_or_end == START) && old_tag && *old_tag
      && !STREQ (old_tag, "samp")
      && !rollback_empty_tag (old_tag))
    add_word_args ("</%s>", old_tag);

  if (*tag)
    {
      if (start_or_end == START)
        add_word_args (format ? "<%s %s>" : "<%s>", tag, formatted_attribs);
      else if (STREQ (tag, "samp") || !rollback_empty_tag (tag))
        /* Insert close tag only if we didn't rollback,
           in which case the opening tag is removed.  */
        add_word_args ("</%s>", tag);
    }

  if ((start_or_end != START) && old_tag && *old_tag && !STREQ (old_tag, "samp"))
    add_word_args (strlen (old_attribs) > 0 ? "<%s %s>" : "<%s>",
        old_tag, old_attribs);

  in_html_elt--;
}

void
insert_html_tag (int start_or_end, char *tag)
{
  insert_html_tag_with_attribute (start_or_end, tag, NULL);
}

/* Output an HTML <link> to the filename for NODE, including the
   other string as extra attributes. */

void
add_link (char *nodename, char *attributes)
{
  if (nodename)
    {
      char *escaped_nodename;
      add_html_elt ("<link ");
      add_word_args ("%s", attributes);
      add_word_args (" href=\"");
      add_anchor_name (nodename, 1);
      escaped_nodename = escape_string (nodename);
      add_word_args ("\" title=\"%s\">\n", escaped_nodename);
      if (escaped_nodename != nodename)
        free (escaped_nodename);
    }
}

/* Copy a name with characters escaped as appropriate for an anchor
   name, i.e., escape URL special characters with our _00hh convention.
   (See the manual for details on the new scheme.) */
   
char *
escaped_anchor_name (const char *name)
{
  /* The factor 5 in the next allocation allows all chars to be expanded.  */
  char *res = xmalloc (5 * strlen (name) + 1);
  char *d = res;
  
  for (; *name; name++)
    {
      if (cr_or_whitespace (*name))
        *d++ = '-';
      else if (! URL_SAFE_CHAR (*name))
        {
          sprintf (d, "_00%x", (unsigned char) *name);
          /* do this manually since sprintf returns char * on
             SunOS 4 and other old systems.  */
          while (*d)
            d++;
        }
      else
        *d++ = *name;
    }
  *d = 0;
  return res;
}

/* Output NAME with characters escaped as appropriate for an anchor
   name, i.e., escape URL special characters with our _00hh convention
   if OLD is zero.  (See the manual for details on the new scheme.)
   
   If OLD is nonzero, generate the node name with the 4.6-and-earlier
   convention of %hh (and more special characters output as-is, notably
   - and *).  This is only so that external references to old names can
   still work with HTML generated by the new makeinfo; the gcc folks
   needed this.  Our own HTML does not refer to these names.  */

void
add_escaped_anchor_name (char *name, int old)
{
  canon_white (name);

  if (!old && !strchr ("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ",
                       *name))
    { /* XHTML does not allow anything but an ASCII letter to start an
         identifier.  Therefore kludge in this constant string if we
         have a nonletter.  */
      add_word ("g_t");
    }
  if (!old) 
    {
      char *expanded = escaped_anchor_name (name);
      add_word (expanded);
      free (expanded);
    }
  else for (; *name; name++)
    {
      if (cr_or_whitespace (*name))
        add_char ('-');

      else if (!URL_SAFE_CHAR (*name) && !OLD_URL_SAFE_CHAR (*name))
        /* Cast so characters with the high bit set are treated as >128,
           for example o-umlaut should be 246, not -10.  */ 
        add_word_args ("%%%x", (unsigned char) *name);
      else
        add_char (*name);
    }
}

/* Insert the text for the name of a reference in an HTML anchor
   appropriate for NODENAME.
   
   If HREF is zero, generate text for name= in the new node name
     conversion convention.
   If HREF is negative, generate text for name= in the old convention.
   If HREF is positive, generate the name for an href= attribute, i.e.,
     including the `#' if it's an internal reference.   */
void
add_anchor_name (char *nodename, int href)
{
  if (href > 0)
    {
      if (splitting)
	add_url_name (nodename, href);
      add_char ('#');
    }
  /* Always add NODENAME, so that the reference would pinpoint the
     exact node on its file.  This is so several nodes could share the
     same file, in case of file-name clashes, but also for more
     accurate browser positioning.  */
  if (mbscasecmp (nodename, "(dir)") == 0)
    /* Strip the parens, but keep the original letter-case.  */
    add_word_args ("%.3s", nodename + 1);
  else if (mbscasecmp (nodename, "top") == 0)
    add_word ("Top");
  else
    add_escaped_anchor_name (nodename, href < 0);
}

/* Insert the text for the name of a reference in an HTML url, aprropriate
   for NODENAME */
void
add_url_name (char *nodename, int href)
{
    add_nodename_to_filename (nodename, href);
}

/* Convert non [A-Za-z0-9] characters depending on the command line options given.
   If --transliterate-file-names is specified, these are replaced with their ASCII
   phonetic transliteration. Otherwise, _00xx notation is used, where xx means the
   hexadecimal representation of the ASCII character.  Also convert spaces and
   newlines to dashes.  */
static void
fix_filename (char *filename)
{
  int i;
  int len = strlen (filename);
  char *oldname = xstrdup (filename);

  *filename = '\0';

  for (i = 0; i < len; i++)
    {
      const char *p = lang_transliterate_char (oldname[i]);

      if (p)
	strcat (filename, p);
      else if (cr_or_whitespace (oldname[i]))
        strcat (filename, "-");
      else if (URL_SAFE_CHAR (oldname[i]))
        strncat (filename, (char *) oldname + i, 1);
      else
        {
          char *hexchar = xmalloc (6 * sizeof (char));
          sprintf (hexchar, "_00%x", (unsigned char) oldname[i]);
          strcat (filename, hexchar);
          free (hexchar);
        }

      /* Check if we are nearing boundaries.  */
      if (strlen (filename) >= PATH_MAX - 20)
        break;
    }

  free (oldname);
}

/* As we can't look-up a (forward-referenced) nodes' html filename
   from the tentry, we take the easy way out.  We assume that
   nodenames are unique, and generate the html filename from the
   nodename, that's always known.  */
static char *
nodename_to_filename_1 (char *nodename, int href)
{
  char *p;
  char *filename;
  char dirname[PATH_MAX];

  if (mbscasecmp (nodename, "Top") == 0)
    {
      /* We want to convert references to the Top node into
	 "index.html#Top".  */
      if (href)
	filename = xstrdup ("index.html"); /* "#Top" is added by our callers */
      else
	filename = xstrdup ("Top");
    }
  else if (mbscasecmp (nodename, "(dir)") == 0)
    /* We want to convert references to the (dir) node into
       "../index.html".  */
    filename = xstrdup ("../index.html");
  else
    {
      filename = xmalloc (PATH_MAX);
      dirname[0] = '\0';
      *filename = '\0';

      /* Check for external reference: ``(info-document)node-name''
	 Assume this node lives at: ``../info-document/node-name.html''

	 We need to handle the special case (sigh): ``(info-document)'',
	 ie, an external top-node, which should translate to:
	 ``../info-document/info-document.html'' */

      p = nodename;
      if (*nodename == '(')
	{
	  int length;

	  p = strchr (nodename, ')');
	  if (p == NULL)
	    {
	      line_error (_("[unexpected] invalid node name: `%s'"), nodename);
	      xexit (1);
	    }

	  length = p - nodename - 1;
	  if (length > 5 &&
	      FILENAME_CMPN (p - 5, ".info", 5) == 0)
	    length -= 5;
	  /* This is for DOS, and also for Windows and GNU/Linux
	     systems that might have Info files copied from a DOS 8+3
	     filesystem.  */
	  if (length > 4 &&
	      FILENAME_CMPN (p - 4, ".inf", 4) == 0)
	    length -= 4;
	  strcpy (filename, "../");
	  strncpy (dirname, nodename + 1, length);
	  *(dirname + length) = '\0';
	  fix_filename (dirname);
	  strcat (filename, dirname);
	  strcat (filename, "/");
	  p++;
	}

      /* In the case of just (info-document), there will be nothing
	 remaining, and we will refer to ../info-document/, which will
	 work fine.  */
      strcat (filename, p);
      if (*p)
	{
	  /* Hmm */
	  fix_filename (filename + strlen (filename) - strlen (p));
	  strcat (filename, ".html");
	}
    }

  /* Produce a file name suitable for the underlying filesystem.  */
  normalize_filename (filename);

#if 0
  /* We add ``#Nodified-filename'' anchor to external references to be
     prepared for non-split HTML support.  Maybe drop this. */
  if (href && *dirname)
    {
      strcat (filename, "#");
      strcat (filename, p);
      /* Hmm, again */
      fix_filename (filename + strlen (filename) - strlen (p));
    }
#endif

  return filename;
}

/* If necessary, ie, if current filename != filename of node, output
   the node name.  */
void
add_nodename_to_filename (char *nodename, int href)
{
  /* for now, don't check: always output filename */
  char *filename = nodename_to_filename_1 (nodename, href);
  add_word (filename);
  free (filename);
}

char *
nodename_to_filename (char *nodename)
{
  /* The callers of nodename_to_filename use the result to produce
     <a href=, so call nodename_to_filename_1 with last arg non-zero.  */
  return nodename_to_filename_1 (nodename, 1);
}