summaryrefslogtreecommitdiffstats
path: root/src/html-url.c
blob: 53d3a934fd551ffb1e50f65b5b928e807c4e88dd (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
/* Collect URLs from HTML source.
   Copyright (C) 1998-2012, 2015, 2018-2022 Free Software Foundation,
   Inc.

This file is part of GNU Wget.

GNU Wget 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.

GNU Wget 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 Wget.  If not, see <http://www.gnu.org/licenses/>.

Additional permission under GNU GPL version 3 section 7

If you modify this program, or any covered work, by linking or
combining it with the OpenSSL project's OpenSSL library (or a
modified version of that library), containing parts covered by the
terms of the OpenSSL or SSLeay licenses, the Free Software Foundation
grants you additional permission to convey the resulting work.
Corresponding Source for a non-source form of such a combination
shall include the source code for the parts of OpenSSL used as well
as that of the covered work.  */

#include "wget.h"

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

#include "exits.h"
#include "html-parse.h"
#include "url.h"
#include "utils.h"
#include "hash.h"
#include "convert.h"
#include "recur.h"
#include "html-url.h"
#include "css-url.h"
#include "c-strcase.h"

typedef void (*tag_handler_t) (int, struct taginfo *, struct map_context *);

#define DECLARE_TAG_HANDLER(fun)                                \
  static void fun (int, struct taginfo *, struct map_context *)

DECLARE_TAG_HANDLER (tag_find_urls);
DECLARE_TAG_HANDLER (tag_handle_base);
DECLARE_TAG_HANDLER (tag_handle_form);
DECLARE_TAG_HANDLER (tag_handle_img);
DECLARE_TAG_HANDLER (tag_handle_link);
DECLARE_TAG_HANDLER (tag_handle_meta);

enum {
  TAG_A,
  TAG_APPLET,
  TAG_AREA,
  TAG_BASE,
  TAG_BGSOUND,
  TAG_BODY,
  TAG_EMBED,
  TAG_FIG,
  TAG_FORM,
  TAG_FRAME,
  TAG_IFRAME,
  TAG_IMG,
  TAG_INPUT,
  TAG_LAYER,
  TAG_LINK,
  TAG_META,
  TAG_OBJECT,
  TAG_OVERLAY,
  TAG_SCRIPT,
  TAG_TABLE,
  TAG_TD,
  TAG_TH,
  TAG_VIDEO,
  TAG_AUDIO,
  TAG_SOURCE
};

/* The list of known tags and functions used for handling them.  Most
   tags are simply harvested for URLs. */
static struct known_tag {
  int tagid;
  const char *name;
  tag_handler_t handler;
} known_tags[] = {
  { TAG_A,       "a",           tag_find_urls },
  { TAG_APPLET,  "applet",      tag_find_urls },
  { TAG_AREA,    "area",        tag_find_urls },
  { TAG_BASE,    "base",        tag_handle_base },
  { TAG_BGSOUND, "bgsound",     tag_find_urls },
  { TAG_BODY,    "body",        tag_find_urls },
  { TAG_EMBED,   "embed",       tag_find_urls },
  { TAG_FIG,     "fig",         tag_find_urls },
  { TAG_FORM,    "form",        tag_handle_form },
  { TAG_FRAME,   "frame",       tag_find_urls },
  { TAG_IFRAME,  "iframe",      tag_find_urls },
  { TAG_IMG,     "img",         tag_handle_img },
  { TAG_INPUT,   "input",       tag_find_urls },
  { TAG_LAYER,   "layer",       tag_find_urls },
  { TAG_LINK,    "link",        tag_handle_link },
  { TAG_META,    "meta",        tag_handle_meta },
  { TAG_OBJECT,  "object",      tag_find_urls },
  { TAG_OVERLAY, "overlay",     tag_find_urls },
  { TAG_SCRIPT,  "script",      tag_find_urls },
  { TAG_TABLE,   "table",       tag_find_urls },
  { TAG_TD,      "td",          tag_find_urls },
  { TAG_TH,      "th",          tag_find_urls },
  { TAG_VIDEO,   "video",       tag_find_urls },
  { TAG_AUDIO,   "audio",       tag_find_urls },
  { TAG_SOURCE,  "source",      tag_find_urls }
};

/* tag_url_attributes documents which attributes of which tags contain
   URLs to harvest.  It is used by tag_find_urls.  */

/* Defines for the FLAGS. */

/* The link is "inline", i.e. needs to be retrieved for this document
   to be correctly rendered.  Inline links include inlined images,
   stylesheets, children frames, etc.  */
#define ATTR_INLINE     1

/* The link is expected to yield HTML contents.  It's important not to
   try to follow HTML obtained by following e.g. <img src="...">
   regardless of content-type.  Doing this causes infinite loops for
   "images" that return non-404 error pages with links to the same
   image.  */
#define ATTR_HTML       2

/* For tags handled by tag_find_urls: attributes that contain URLs to
   download. */
static struct {
  int tagid;
  const char *attr_name;
  int flags;
} tag_url_attributes[] = {
  { TAG_A,              "href",         ATTR_HTML },
  { TAG_APPLET,         "code",         ATTR_INLINE },
  { TAG_AREA,           "href",         ATTR_HTML },
  { TAG_BGSOUND,        "src",          ATTR_INLINE },
  { TAG_BODY,           "background",   ATTR_INLINE },
  { TAG_EMBED,          "href",         ATTR_HTML },
  { TAG_EMBED,          "src",          ATTR_INLINE | ATTR_HTML },
  { TAG_FIG,            "src",          ATTR_INLINE },
  { TAG_FRAME,          "src",          ATTR_INLINE | ATTR_HTML },
  { TAG_IFRAME,         "src",          ATTR_INLINE | ATTR_HTML },
  { TAG_IMG,            "href",         ATTR_INLINE },
  { TAG_IMG,            "lowsrc",       ATTR_INLINE },
  { TAG_IMG,            "src",          ATTR_INLINE },
  { TAG_INPUT,          "src",          ATTR_INLINE },
  { TAG_LAYER,          "src",          ATTR_INLINE | ATTR_HTML },
  { TAG_OBJECT,         "data",         ATTR_INLINE },
  { TAG_OVERLAY,        "src",          ATTR_INLINE | ATTR_HTML },
  { TAG_SCRIPT,         "src",          ATTR_INLINE },
  { TAG_TABLE,          "background",   ATTR_INLINE },
  { TAG_TD,             "background",   ATTR_INLINE },
  { TAG_TH,             "background",   ATTR_INLINE },
  { TAG_VIDEO,          "src",          ATTR_INLINE },
  { TAG_VIDEO,          "poster",       ATTR_INLINE },
  { TAG_AUDIO,          "src",          ATTR_INLINE },
  { TAG_AUDIO,          "poster",       ATTR_INLINE },
  { TAG_SOURCE,         "src",          ATTR_INLINE }
};

/* The lists of interesting tags and attributes are built dynamically,
   from the information above.  However, some places in the code refer
   to the attributes not mentioned here.  We add them manually.  */
static const char *additional_attributes[] = {
  "rel",                        /* used by tag_handle_link  */
  "type",                       /* used by tag_handle_link  */
  "http-equiv",                 /* used by tag_handle_meta  */
  "name",                       /* used by tag_handle_meta  */
  "content",                    /* used by tag_handle_meta  */
  "action",                     /* used by tag_handle_form  */
  "style",                      /* used by check_style_attr */
  "srcset",                     /* used by tag_handle_img */
};

static struct hash_table *interesting_tags;
static struct hash_table *interesting_attributes;

/* Will contains the (last) charset found in 'http-equiv=content-type'
   meta tags  */
static char *meta_charset;

static void
init_interesting (void)
{
  /* Init the variables interesting_tags and interesting_attributes
     that are used by the HTML parser to know which tags and
     attributes we're interested in.  We initialize this only once,
     for performance reasons.

     Here we also make sure that what we put in interesting_tags
     matches the user's preferences as specified through --ignore-tags
     and --follow-tags.  */

  size_t i;
  interesting_tags = make_nocase_string_hash_table (countof (known_tags));

  /* First, add all the tags we know hot to handle, mapped to their
     respective entries in known_tags.  */
  for (i = 0; i < countof (known_tags); i++)
    hash_table_put (interesting_tags, known_tags[i].name, known_tags + i);

  /* Then remove the tags ignored through --ignore-tags.  */
  if (opt.ignore_tags)
    {
      char **ignored;
      for (ignored = opt.ignore_tags; *ignored; ignored++)
        hash_table_remove (interesting_tags, *ignored);
    }

  /* If --follow-tags is specified, use only those tags.  */
  if (opt.follow_tags)
    {
      /* Create a new table intersecting --follow-tags and known_tags,
         and use it as interesting_tags.  */
      struct hash_table *intersect = make_nocase_string_hash_table (0);
      char **followed;
      for (followed = opt.follow_tags; *followed; followed++)
        {
          struct known_tag *t = hash_table_get (interesting_tags, *followed);
          if (!t)
            continue;           /* ignore unknown --follow-tags entries. */
          hash_table_put (intersect, *followed, t);
        }
      hash_table_destroy (interesting_tags);
      interesting_tags = intersect;
    }

  /* Add the attributes we care about. */
  interesting_attributes = make_nocase_string_hash_table (10);
  for (i = 0; i < countof (additional_attributes); i++)
    hash_table_put (interesting_attributes, additional_attributes[i], "1");
  for (i = 0; i < countof (tag_url_attributes); i++)
    hash_table_put (interesting_attributes,
                    tag_url_attributes[i].attr_name, "1");
}

/* Find the value of attribute named NAME in the taginfo TAG.  If the
   attribute is not present, return NULL.  If ATTRIND is non-NULL, the
   index of the attribute in TAG will be stored there.  */

static char *
find_attr (struct taginfo *tag, const char *name, int *attrind)
{
  int i;
  for (i = 0; i < tag->nattrs; i++)
    if (!c_strcasecmp (tag->attrs[i].name, name))
      {
        if (attrind)
          *attrind = i;
        return tag->attrs[i].value;
      }
  return NULL;
}

/* used for calls to append_url */
#define ATTR_POS(tag, attrind, ctx) \
 (tag->attrs[attrind].value_raw_beginning - ctx->text)
#define ATTR_SIZE(tag, attrind) \
 (tag->attrs[attrind].value_raw_size)

/* Append LINK_URI to the urlpos structure that is being built.

   LINK_URI will be merged with the current document base.
*/

struct urlpos *
append_url (const char *link_uri, int position, int size,
            struct map_context *ctx)
{
  int link_has_scheme = url_has_scheme (link_uri);
  struct urlpos *newel;
  const char *base = ctx->base ? ctx->base : ctx->parent_base;
  struct url *url;

  struct iri *iri = iri_new ();
  set_uri_encoding (iri, opt.locale, true);
  iri->utf8_encode = true;

  if (!base)
    {
      DEBUGP (("%s: no base, merge will use \"%s\".\n",
               ctx->document_file, link_uri));

      if (!link_has_scheme)
        {
          /* Base URL is unavailable, and the link does not have a
             location attached to it -- we have to give up.  Since
             this can only happen when using `--force-html -i', print
             a warning.  */
          logprintf (LOG_NOTQUIET,
                     _("%s: Cannot resolve incomplete link %s.\n"),
                     ctx->document_file, link_uri);
          iri_free (iri);
          return NULL;
        }

      url = url_parse (link_uri, NULL, iri, false);
      if (!url)
        {
          DEBUGP (("%s: link \"%s\" doesn't parse.\n",
                   ctx->document_file, link_uri));
          iri_free (iri);
          return NULL;
        }
    }
  else
    {
      /* Merge BASE with LINK_URI, but also make sure the result is
         canonicalized, i.e. that "../" have been resolved.
         (parse_url will do that for us.) */

      char *complete_uri = uri_merge (base, link_uri);

      DEBUGP (("%s: merge(%s, %s) -> %s\n",
               quotearg_n_style (0, escape_quoting_style, ctx->document_file),
               quote_n (1, base),
               quote_n (2, link_uri),
               quotearg_n_style (3, escape_quoting_style, complete_uri)));

      url = url_parse (complete_uri, NULL, iri, false);
      if (!url)
        {
          DEBUGP (("%s: merged link \"%s\" doesn't parse.\n",
                   ctx->document_file, complete_uri));
          xfree (complete_uri);
          iri_free (iri);
          return NULL;
        }
      xfree (complete_uri);
    }

  iri_free (iri);

  DEBUGP (("appending %s to urlpos.\n", quote (url->url)));

  newel = xnew0 (struct urlpos);
  newel->url = url;
  newel->pos = position;
  newel->size = size;

  /* A URL is relative if the host is not named, and the name does not
     start with `/'.  */
  if (!link_has_scheme && *link_uri != '/')
    newel->link_relative_p = 1;
  else if (link_has_scheme)
    newel->link_complete_p = 1;

  /* Append the new URL maintaining the order by position.  */
  if (ctx->head == NULL)
    ctx->head = newel;
  else
    {
      struct urlpos *it, *prev = NULL;

      it = ctx->head;
      while (it && position > it->pos)
        {
          prev = it;
          it = it->next;
        }

      newel->next = it;

      if (prev)
        prev->next = newel;
      else
        ctx->head = newel;
    }

  return newel;
}

static void
check_style_attr (struct taginfo *tag, struct map_context *ctx)
{
  int attrind;
  int raw_start;
  int raw_len;
  char *style = find_attr (tag, "style", &attrind);
  if (!style)
    return;

  /* raw pos and raw size include the quotes, skip them when they are
     present.  */
  raw_start = ATTR_POS (tag, attrind, ctx);
  raw_len  = ATTR_SIZE (tag, attrind);
  if( *(char *)(ctx->text + raw_start) == '\''
      || *(char *)(ctx->text + raw_start) == '"')
    {
      raw_start += 1;
      raw_len -= 2;
    }

  if(raw_len <= 0)
       return;

  get_urls_css (ctx, raw_start, raw_len);
}

/* All the tag_* functions are called from collect_tags_mapper, as
   specified by KNOWN_TAGS.  */

/* Default tag handler: collect URLs from attributes specified for
   this tag by tag_url_attributes.  */

static void
tag_find_urls (int tagid, struct taginfo *tag, struct map_context *ctx)
{
  size_t i;
  int attrind;
  int first = -1;

  for (i = 0; i < countof (tag_url_attributes); i++)
    if (tag_url_attributes[i].tagid == tagid)
      {
        /* We've found the index of tag_url_attributes where the
           attributes of our tag begin.  */
        first = i;
        break;
      }
  assert (first != -1);

  /* Loop over the "interesting" attributes of this tag.  In this
     example, it will loop over "src" and "lowsrc".

       <img src="foo.png" lowsrc="bar.png">

     This has to be done in the outer loop so that the attributes are
     processed in the same order in which they appear in the page.
     This is required when converting links.  */

  for (attrind = 0; attrind < tag->nattrs; attrind++)
    {
      /* Find whether TAG/ATTRIND is a combination that contains a
         URL. */
      char *link = tag->attrs[attrind].value;
      const size_t size = countof (tag_url_attributes);

      /* If you're cringing at the inefficiency of the nested loops,
         remember that they both iterate over a very small number of
         items.  The worst-case inner loop is for the IMG tag, which
         has three attributes.  */
      for (i = first; i < size && tag_url_attributes[i].tagid == tagid; i++)
        {
          if (0 == strcasecmp (tag->attrs[attrind].name,
                               tag_url_attributes[i].attr_name))
            {
              struct urlpos *up = append_url (link, ATTR_POS(tag,attrind,ctx),
                                              ATTR_SIZE(tag,attrind), ctx);
              if (up)
                {
                  int flags = tag_url_attributes[i].flags;
                  if (flags & ATTR_INLINE)
                    up->link_inline_p = 1;
                  if (flags & ATTR_HTML)
                    up->link_expect_html = 1;
                }
            }
        }
    }
}

/* Handle the BASE tag, for <base href=...>. */

static void
tag_handle_base (int tagid _GL_UNUSED, struct taginfo *tag, struct map_context *ctx)
{
  struct urlpos *base_urlpos;
  int attrind;
  char *newbase = find_attr (tag, "href", &attrind);
  if (!newbase)
    return;

  base_urlpos = append_url (newbase, ATTR_POS(tag,attrind,ctx),
                            ATTR_SIZE(tag,attrind), ctx);
  if (!base_urlpos)
    return;
  base_urlpos->ignore_when_downloading = 1;
  base_urlpos->link_base_p = 1;

  xfree (ctx->base);
  if (ctx->parent_base)
    ctx->base = uri_merge (ctx->parent_base, newbase);
  else
    ctx->base = xstrdup (newbase);
}

/* Mark the URL found in <form action=...> for conversion. */

static void
tag_handle_form (int tagid _GL_UNUSED, struct taginfo *tag, struct map_context *ctx)
{
  int attrind;
  char *action = find_attr (tag, "action", &attrind);

  if (action)
    {
      struct urlpos *up = append_url (action, ATTR_POS(tag,attrind,ctx),
                                      ATTR_SIZE(tag,attrind), ctx);
      if (up)
        up->ignore_when_downloading = 1;
    }
}

/* Handle the LINK tag.  It requires special handling because how its
   links will be followed in -p mode depends on the REL attribute.  */

static void
tag_handle_link (int tagid _GL_UNUSED, struct taginfo *tag, struct map_context *ctx)
{
  int attrind;
  char *href = find_attr (tag, "href", &attrind);

  /* All <link href="..."> link references are external, except those
     known not to be, such as style sheet and shortcut icon:

     <link rel="stylesheet" href="..."> or <link rel="alternate stylesheet" href="...">
     <link rel="shortcut icon" href="..."> or <link rel="icon" href="...">
  */
  if (href)
    {
      struct urlpos *up = append_url (href, ATTR_POS(tag,attrind,ctx),
                                      ATTR_SIZE(tag,attrind), ctx);
      if (up)
        {
          char *rel = find_attr (tag, "rel", NULL);
          if (rel)
            {
              if (0 == c_strcasecmp (rel, "stylesheet") || 0 == c_strcasecmp (rel, "alternate stylesheet"))
                {
                  up->link_inline_p = 1;
                  up->link_expect_css = 1;
                }
              else if (0 == c_strcasecmp (rel, "shortcut icon") || 0 == c_strcasecmp (rel, "icon"))
                {
                  up->link_inline_p = 1;
                }
              else if (0 == c_strcasecmp (rel, "manifest"))
                {
                  up->link_inline_p = 1;
                }
              else
                {
                  /* The external ones usually point to HTML pages, such as
                     <link rel="next" href="...">
                     except when the type attribute says otherwise:
                     <link rel="alternate" type="application/rss+xml" href=".../?feed=rss2" />
                  */
                  char *type = find_attr (tag, "type", NULL);
                  if (!type || c_strcasecmp (type, "text/html") == 0)
                    up->link_expect_html = 1;
                }
            }
        }
    }
}

/* Handle the META tag.  This requires special handling because of the
   refresh feature and because of robot exclusion.  */

static void
tag_handle_meta (int tagid _GL_UNUSED, struct taginfo *tag, struct map_context *ctx)
{
  char *name = find_attr (tag, "name", NULL);
  char *http_equiv = find_attr (tag, "http-equiv", NULL);

  if (http_equiv && 0 == c_strcasecmp (http_equiv, "refresh"))
    {
      /* Some pages use a META tag to specify that the page be
         refreshed by a new page after a given number of seconds.  The
         general format for this is:

           <meta http-equiv=Refresh content="NUMBER; URL=index2.html">

         So we just need to skip past the "NUMBER; URL=" garbage to
         get to the URL.  */

      struct urlpos *entry;
      int attrind;
      int timeout;
      char *p;

      char *refresh = find_attr (tag, "content", &attrind);
      if (!refresh)
        return;

      timeout = strtol(refresh, &p, 10);

      if (timeout < 0 || *p++ != ';')
        return;

      while (c_isspace (*p))
        ++p;
      if (!(   c_toupper (*p)       == 'U'
            && c_toupper (*(p + 1)) == 'R'
            && c_toupper (*(p + 2)) == 'L'
            &&          *(p + 3)  == '='))
        return;
      p += 4;
      while (c_isspace (*p))
        ++p;

      entry = append_url (p, ATTR_POS(tag,attrind,ctx),
                          ATTR_SIZE(tag,attrind), ctx);
      if (entry)
        {
          entry->link_refresh_p = 1;
          entry->refresh_timeout = timeout;
          entry->link_expect_html = 1;
        }
    }
  else if (http_equiv && 0 == c_strcasecmp (http_equiv, "content-type"))
    {
      /* Handle stuff like:
         <meta http-equiv="Content-Type" content="text/html; charset=CHARSET"> */

      char *mcharset;
      char *content = find_attr (tag, "content", NULL);
      if (!content)
        return;

      mcharset = parse_charset (content);
      if (!mcharset)
        return;

      xfree (meta_charset);
      meta_charset = mcharset;
    }
  else if (name && 0 == c_strcasecmp (name, "robots"))
    {
      /* Handle stuff like:
         <meta name="robots" content="index,nofollow"> */
      char *content = find_attr (tag, "content", NULL);
      if (!content)
        return;
      if (!c_strcasecmp (content, "none"))
        ctx->nofollow = true;
      else
        {
          while (*content)
            {
              char *end;
              /* Skip any initial whitespace. */
              content += strspn (content, " \f\n\r\t\v");
              /* Find the next occurrence of ',' or whitespace,
               * or the end of the string.  */
              end = content + strcspn (content, ", \f\n\r\t\v");
              if (!c_strncasecmp (content, "nofollow", end - content))
                ctx->nofollow = true;
              /* Skip past the next comma, if any. */
              if (*end == ',')
                ++end;
              else
                {
                  end = strchr (end, ',');
                  if (end)
                    ++end;
                  else
                    end = content + strlen (content);
                }
              content = end;
            }
        }
    }
}

/* Handle the IMG tag.  This requires special handling for the srcset attr,
   while the traditional src/lowsrc/href attributes can be handled generically.
*/

static void
tag_handle_img (int tagid, struct taginfo *tag, struct map_context *ctx) {
  int attrind;
  char *srcset;

  /* Use the generic approach for the attributes without special syntax. */
  tag_find_urls(tagid, tag, ctx);

  srcset = find_attr (tag, "srcset", &attrind);
  if (srcset)
    {
      /* These are relative to the input text. */
      int base_ind = ATTR_POS (tag,attrind,ctx);
      int size = strlen (srcset);

      /* These are relative to srcset. */
      int offset, url_start, url_end;

      /* Make sure to line up base_ind with srcset[0], not outside quotes. */
      if (ctx->text[base_ind] == '"' || ctx->text[base_ind] == '\'')
        ++base_ind;

      offset = 0;
      while (offset < size)
        {
          bool has_descriptor = true;

          /* Skip over initial whitespace and commas. Note there is no \v
            in HTML5 whitespace. */
          url_start = offset + strspn (srcset + offset, " \f\n\r\t,");

          if (url_start == size)
            return;

          /* URL is any non-whitespace chars (including commas) - but with
             trailing commas removed. */
          url_end = url_start + strcspn (srcset + url_start, " \f\n\r\t");
          while ((url_end - 1) > url_start && srcset[url_end - 1] == ',')
            {
              has_descriptor = false;
              --url_end;
            }

          if (url_end > url_start)
            {
              char *url_text = strdupdelim (srcset + url_start,
                                            srcset + url_end);
              struct urlpos *up = append_url (url_text, base_ind + url_start,
                                              url_end - url_start, ctx);
              if (up)
                {
                  up->link_inline_p = 1;
                  up->link_noquote_html_p = 1;
                }
              xfree (url_text);
            }

          /* If the URL wasn't terminated by a , there may also be a descriptor
             which we just skip. */
          if (has_descriptor)
            {
              /* This is comma-terminated, except there may be one level of
                 parentheses escaping that. */
              bool in_paren = false;
              for (offset = url_end; offset < size; ++offset)
                {
                  char c = srcset[offset];
                  if (c == '(')
                    in_paren = true;
                  else if (c == ')' && in_paren)
                    in_paren = false;
                  else if (c == ',' && !in_paren)
                    break;
                }
            }
          else
            offset = url_end;
        }
    }
}

/* Dispatch the tag handler appropriate for the tag we're mapping
   over.  See known_tags[] for definition of tag handlers.  */

static void
collect_tags_mapper (struct taginfo *tag, void *arg)
{
  struct map_context *ctx = (struct map_context *)arg;

  /* Find the tag in our table of tags.  This must not fail because
     map_html_tags only returns tags found in interesting_tags.

     I've changed this for now, I'm passing NULL as interesting_tags
     to map_html_tags.  This way we can check all tags for a style
     attribute.
  */
  struct known_tag *t = hash_table_get (interesting_tags, tag->name);

  if (t != NULL)
    t->handler (t->tagid, tag, ctx);

  check_style_attr (tag, ctx);

  if (tag->end_tag_p && (0 == c_strcasecmp (tag->name, "style"))
      && tag->contents_begin && tag->contents_end
      && tag->contents_begin <= tag->contents_end)
  {
    /* parse contents */
    get_urls_css (ctx, tag->contents_begin - ctx->text,
                  tag->contents_end - tag->contents_begin);
  }
}

/* Analyze HTML tags FILE and construct a list of URLs referenced from
   it.  It merges relative links in FILE with URL.  It is aware of
   <base href=...> and does the right thing.  */

struct urlpos *
get_urls_html_fm (const char *file, const struct file_memory *fm,
                    const char *url, bool *meta_disallow_follow,
                    struct iri *iri)
{
  struct map_context ctx;
  int flags;

  ctx.text = fm->content;
  ctx.head = NULL;
  ctx.base = NULL;
  ctx.parent_base = url ? url : opt.base_href;
  ctx.document_file = file;
  ctx.nofollow = false;

  if (!interesting_tags)
    init_interesting ();

  /* Specify MHT_TRIM_VALUES because of buggy HTML generators that
     generate <a href=" foo"> instead of <a href="foo"> (browsers
     ignore spaces as well.)  If you really mean space, use &32; or
     %20.  MHT_TRIM_VALUES also causes squashing of embedded newlines,
     e.g. in <img src="foo.[newline]html">.  Such newlines are also
     ignored by IE and Mozilla and are presumably introduced by
     writing HTML with editors that force word wrap.  */
  flags = MHT_TRIM_VALUES;
  if (opt.strict_comments)
    flags |= MHT_STRICT_COMMENTS;

  /* the NULL here used to be interesting_tags */
  map_html_tags (fm->content, fm->length, collect_tags_mapper, &ctx, flags,
                 NULL, interesting_attributes);

#ifdef ENABLE_IRI
  /* Meta charset is only valid if there was no HTTP header Content-Type charset. */
  /* This is true for HTTP 1.0 and 1.1. */
  if (iri && !iri->content_encoding && meta_charset)
    set_content_encoding (iri, meta_charset);
#endif
  xfree (meta_charset);

  DEBUGP (("nofollow in %s: %d\n", file, ctx.nofollow));

  if (meta_disallow_follow)
    *meta_disallow_follow = ctx.nofollow;

  xfree (ctx.base);
  return ctx.head;
}

struct urlpos *
get_urls_html (const char *file, const char *url, bool *meta_disallow_follow,
                 struct iri *iri)
{
  struct urlpos *urls;
  struct file_memory *fm;

  fm = wget_read_file (file);
  if (!fm)
    {
      logprintf (LOG_NOTQUIET, "%s: %s\n", file, strerror (errno));
      return NULL;
    }
  DEBUGP (("Loaded %s (size %s).\n", file, number_to_static_string (fm->length)));

  urls = get_urls_html_fm (file, fm, url, meta_disallow_follow, iri);
  wget_read_file_free (fm);
  return urls;
}

/* This doesn't really have anything to do with HTML, but it's similar
   to get_urls_html, so we put it here.  */

struct urlpos *
get_urls_file (const char *file)
{
  struct file_memory *fm;
  struct urlpos *head, *tail;
  const char *text, *text_end;

  /* Load the file.  */
  fm = wget_read_file (file);
  if (!fm)
    {
      logprintf (LOG_NOTQUIET, "%s: %s\n", file, strerror (errno));
      return NULL;
    }
  DEBUGP (("Loaded %s (size %s).\n", file, number_to_static_string (fm->length)));

  head = tail = NULL;
  text = fm->content;
  text_end = fm->content + fm->length;
  while (text < text_end)
    {
      int up_error_code;
      char *url_text;
      char *new_url;
      struct urlpos *entry;
      struct url *url;

      const char *line_beg = text;
      const char *line_end = memchr (text, '\n', text_end - text);
      if (!line_end)
        line_end = text_end;
      else
        ++line_end;
      text = line_end;

      /* Strip whitespace from the beginning and end of line. */
      while (line_beg < line_end && c_isspace (*line_beg))
        ++line_beg;
      while (line_end > line_beg && c_isspace (*(line_end - 1)))
        --line_end;

      if (line_beg == line_end)
        continue;

      /* The URL is in the [line_beg, line_end) region. */

      /* We must copy the URL to a zero-terminated string, and we
         can't use alloca because we're in a loop.  *sigh*.  */
      url_text = strdupdelim (line_beg, line_end);

      if (opt.base_href)
        {
          /* Merge opt.base_href with URL. */
          char *merged = uri_merge (opt.base_href, url_text);
          xfree (url_text);
          url_text = merged;
        }

      new_url = rewrite_shorthand_url (url_text);
      if (new_url)
        {
          xfree (url_text);
          url_text = new_url;
        }

      url = url_parse (url_text, &up_error_code, NULL, false);
      if (!url)
        {
          char *error = url_error (url_text, up_error_code);
          logprintf (LOG_NOTQUIET, _("%s: Invalid URL %s: %s\n"),
                     file, url_text, error);
          xfree (url_text);
          xfree (error);
          inform_exit_status (URLERROR);
          continue;
        }
      xfree (url_text);

      entry = xnew0 (struct urlpos);
      entry->url = url;

      if (!head)
        head = entry;
      else
        tail->next = entry;
      tail = entry;
    }
  wget_read_file_free (fm);
  return head;
}

#if defined DEBUG_MALLOC || defined TESTING
void
cleanup_html_url (void)
{
  /* Destroy the hash tables.  The hash table keys and values are not
     allocated by this code, so we don't need to free them here.  */
  if (interesting_tags)
    hash_table_destroy (interesting_tags);
  if (interesting_attributes)
    hash_table_destroy (interesting_attributes);
}
#endif