/[suikacvs]/markup/html/whatpm/Whatpm/ContentChecker/HTML.pm
Suika

Contents of /markup/html/whatpm/Whatpm/ContentChecker/HTML.pm

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.135 - (hide annotations) (download)
Sun Sep 21 11:55:49 2008 UTC (16 years, 10 months ago) by wakaba
Branch: MAIN
Changes since 1.134: +10 -55 lines
++ whatpm/Whatpm/ChangeLog	21 Sep 2008 11:41:39 -0000
	* ContentChecker.pm: |contextmenu| validness is now checked using
	|id| and |id_type| properties, and |menu| property is removed.

2008-09-21  Wakaba  <wakaba@suika.fam.cx>

++ whatpm/Whatpm/ContentChecker/ChangeLog	21 Sep 2008 11:44:45 -0000
	* HTML.pm: |id| attribute checkers for the |menu| and |map|
	elements are removed; the |id| attribute of those elements is now
	checked by the global |id| attribute.

2008-09-21  Wakaba  <wakaba@suika.fam.cx>

1 wakaba 1.1 package Whatpm::ContentChecker;
2     use strict;
3     require Whatpm::ContentChecker;
4    
5 wakaba 1.117 use Char::Class::XML qw/InXML_NCNameStartChar10 InXMLNCNameChar10/;
6    
7 wakaba 1.1 my $HTML_NS = q<http://www.w3.org/1999/xhtml>;
8    
9 wakaba 1.58 sub FEATURE_HTML5_ROLE () {
10     Whatpm::ContentChecker::FEATURE_STATUS_WD
11     ## TODO: svg:*/@role
12     }
13    
14 wakaba 1.89 sub FEATURE_HTML5_COMPLETE () {
15     Whatpm::ContentChecker::FEATURE_STATUS_REC |
16     Whatpm::ContentChecker::FEATURE_ALLOWED
17     }
18    
19 wakaba 1.54 sub FEATURE_HTML5_LC () {
20     Whatpm::ContentChecker::FEATURE_STATUS_LC |
21     Whatpm::ContentChecker::FEATURE_ALLOWED
22     }
23     sub FEATURE_HTML5_AT_RISK () {
24     Whatpm::ContentChecker::FEATURE_STATUS_WD |
25     Whatpm::ContentChecker::FEATURE_ALLOWED
26     }
27     sub FEATURE_HTML5_WD () {
28     Whatpm::ContentChecker::FEATURE_STATUS_WD |
29     Whatpm::ContentChecker::FEATURE_ALLOWED
30     }
31     sub FEATURE_HTML5_FD () {
32     Whatpm::ContentChecker::FEATURE_STATUS_WD |
33     Whatpm::ContentChecker::FEATURE_ALLOWED
34     }
35     sub FEATURE_HTML5_DEFAULT () {
36     Whatpm::ContentChecker::FEATURE_STATUS_WD |
37     Whatpm::ContentChecker::FEATURE_ALLOWED
38 wakaba 1.49 }
39 wakaba 1.54 sub FEATURE_HTML5_DROPPED () {
40     ## NOTE: Was part of HTML5, but was dropped.
41 wakaba 1.49 Whatpm::ContentChecker::FEATURE_STATUS_WD
42     }
43 wakaba 1.118 sub FEATURE_HTML5_LC_DROPPED () {
44     ## NOTE: Was part of HTML5, but was dropped.
45     Whatpm::ContentChecker::FEATURE_STATUS_LC
46     }
47 wakaba 1.119 sub FEATURE_WF2X () {
48     ## NOTE: Defined in WF2 and then incorporated into the HTML5 spec.
49     Whatpm::ContentChecker::FEATURE_STATUS_LC
50     }
51 wakaba 1.54 sub FEATURE_WF2 () {
52 wakaba 1.119 ## NOTE: Part of WF2, still not added to the HTML5 spec.
53     Whatpm::ContentChecker::FEATURE_STATUS_LC
54 wakaba 1.54 }
55 wakaba 1.126 sub FEATURE_WF2_INFORMATIVE () {
56     ## NOTE: Part of WF2 Appendix A, still not added to the HTML5 spec.
57     Whatpm::ContentChecker::FEATURE_STATUS_LC
58     }
59 wakaba 1.119 sub FEATURE_WF2X_DEPRECATED () {
60 wakaba 1.56 Whatpm::ContentChecker::FEATURE_STATUS_LC
61 wakaba 1.119 ## NOTE: MUST NOT be used according to WF2.
62 wakaba 1.56 }
63 wakaba 1.49
64 wakaba 1.61 ## NOTE: Metainformation Attributes Module by W3C XHTML2 WG.
65 wakaba 1.121 sub FEATURE_RDFA_PR () {
66     Whatpm::ContentChecker::FEATURE_STATUS_CR
67     ## NOTE: Diff from FEATURE_RDFA_CR
68     }
69 wakaba 1.120 sub FEATURE_RDFA_CR () {
70     Whatpm::ContentChecker::FEATURE_STATUS_CR
71     }
72 wakaba 1.61 sub FEATURE_RDFA_LC () {
73     Whatpm::ContentChecker::FEATURE_STATUS_LC
74     }
75 wakaba 1.58
76 wakaba 1.82 sub FEATURE_RDFA_ED () {
77     Whatpm::ContentChecker::FEATURE_STATUS_WD
78     }
79    
80 wakaba 1.58 ## NOTE: XHTML Role LCWD has almost no information on how the |role|
81     ## attribute can be used- the only requirements for that matter is:
82     ## "the attribute MUST be referenced using its namespace-qualified form" (and
83     ## this is a host language conformance!).
84 wakaba 1.82 sub FEATURE_ROLE_LC () {
85     Whatpm::ContentChecker::FEATURE_STATUS_LC
86     }
87    
88     sub FEATURE_XHTML2_ED () {
89     Whatpm::ContentChecker::FEATURE_STATUS_WD
90     }
91 wakaba 1.58
92 wakaba 1.55 sub FEATURE_XHTMLBASIC11_CR () {
93     ## NOTE: Only additions to M12N10_REC are marked.
94     Whatpm::ContentChecker::FEATURE_STATUS_CR
95     }
96     sub FEATURE_XHTMLBASIC11_CR_DEPRECATED () {
97     Whatpm::ContentChecker::FEATURE_STATUS_CR |
98     Whatpm::ContentChecker::FEATURE_DEPRECATED_INFO
99     }
100    
101 wakaba 1.82 sub FEATURE_M12N11_LC () {
102     Whatpm::ContentChecker::FEATURE_STATUS_LC
103     }
104    
105 wakaba 1.99 sub FEATURE_RUBY_REC () {
106     Whatpm::ContentChecker::FEATURE_STATUS_CR
107     }
108    
109 wakaba 1.49 ## NOTE: M12N10 status is based on its abstract module definition.
110     ## It contains a number of problems. (However, again, it's a REC!)
111 wakaba 1.54 sub FEATURE_M12N10_REC () {
112     ## NOTE: Oh, XHTML m12n 1.0 passed the CR phase! W3C Process suck!
113     Whatpm::ContentChecker::FEATURE_STATUS_REC
114     }
115     sub FEATURE_M12N10_REC_DEPRECATED () {
116     Whatpm::ContentChecker::FEATURE_STATUS_REC |
117     Whatpm::ContentChecker::FEATURE_DEPRECATED_INFO
118     }
119 wakaba 1.120 ## NOTE: XHTML M12N 1.1 is in the PR status at the time of writing and no
120 wakaba 1.58 ## addition from 1.0.
121 wakaba 1.49
122     ## NOTE: XHTML10 status is based on its transitional and frameset DTDs
123     ## (second edition). Only missing attributes from M12N10 abstract
124     ## definition are added.
125 wakaba 1.54 sub FEATURE_XHTML10_REC () {
126     Whatpm::ContentChecker::FEATURE_STATUS_CR
127     }
128    
129 wakaba 1.61 ## NOTE: Diff from HTML4.
130     sub FEATURE_ISOHTML_PREPARATION () { ## Informative documentation
131     Whatpm::ContentChecker::FEATURE_STATUS_CR
132     }
133 wakaba 1.58
134 wakaba 1.49 ## NOTE: HTML4 status is based on its transitional and frameset DTDs (HTML
135     ## 4.01). Only missing attributes from XHTML10 are added.
136 wakaba 1.54 sub FEATURE_HTML4_REC_RESERVED () {
137     Whatpm::ContentChecker::FEATURE_STATUS_WD
138     }
139    
140     ## TODO: According to HTML4 definition, authors SHOULD use style sheets
141     ## rather than presentational attributes (deprecated or not deprecated).
142 wakaba 1.48
143 wakaba 1.61 ## NOTE: Diff from HTML4.
144     sub FEATURE_HTML32_REC_OBSOLETE () {
145     Whatpm::ContentChecker::FEATURE_STATUS_CR |
146     Whatpm::ContentChecker::FEATURE_DEPRECATED_SHOULD
147     ## NOTE: Lowercase normative "should".
148     }
149    
150     sub FEATURE_RFC2659 () { ## Experimental RFC
151     Whatpm::ContentChecker::FEATURE_STATUS_CR
152     }
153    
154     ## NOTE: HTML 2.x - diff from HTML 2.0 and not in newer versions.
155     sub FEATURE_HTML2X_RFC () { ## Proposed Standard, obsolete
156     Whatpm::ContentChecker::FEATURE_STATUS_CR
157     }
158    
159     ## NOTE: Diff from HTML 2.0.
160     sub FEATURE_RFC1942 () { ## Experimental RFC, obsolete
161     Whatpm::ContentChecker::FEATURE_STATUS_CR
162     }
163    
164     ## NOTE: Diff from HTML 3.2.
165     sub FEATURE_HTML20_RFC () { ## Proposed Standard, obsolete
166     Whatpm::ContentChecker::FEATURE_STATUS_CR
167     }
168 wakaba 1.58
169 wakaba 1.29 ## December 2007 HTML5 Classification
170    
171     my $HTMLMetadataContent = {
172     $HTML_NS => {
173     title => 1, base => 1, link => 1, style => 1, script => 1, noscript => 1,
174 wakaba 1.118 'event-source' => 1, eventsource => 1,
175     command => 1, datatemplate => 1,
176 wakaba 1.29 ## NOTE: A |meta| with no |name| element is not allowed as
177     ## a metadata content other than |head| element.
178     meta => 1,
179     },
180     ## NOTE: RDF is mentioned in the HTML5 spec.
181     ## TODO: Other RDF elements?
182     q<http://www.w3.org/1999/02/22-rdf-syntax-ns#> => {RDF => 1},
183     };
184    
185 wakaba 1.72 my $HTMLFlowContent = {
186 wakaba 1.29 $HTML_NS => {
187     section => 1, nav => 1, article => 1, blockquote => 1, aside => 1,
188     h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, header => 1,
189     footer => 1, address => 1, p => 1, hr => 1, dialog => 1, pre => 1,
190     ol => 1, ul => 1, dl => 1, figure => 1, map => 1, table => 1,
191 wakaba 1.119 form => 1, fieldset => 1,
192 wakaba 1.72 details => 1, ## ISSUE: "Flow element" in spec.
193     datagrid => 1, ## ISSUE: "Flow element" in spec.
194 wakaba 1.29 datatemplate => 1,
195     div => 1, ## ISSUE: No category in spec.
196     ## NOTE: |style| is only allowed if |scoped| attribute is specified.
197     ## Additionally, it must be before any other element or
198     ## non-inter-element-whitespace text node.
199     style => 1,
200    
201 wakaba 1.38 br => 1, q => 1, cite => 1, em => 1, strong => 1, small => 1, mark => 1,
202 wakaba 1.29 dfn => 1, abbr => 1, time => 1, progress => 1, meter => 1, code => 1,
203     var => 1, samp => 1, kbd => 1, sub => 1, sup => 1, span => 1, i => 1,
204 wakaba 1.99 b => 1, bdo => 1, ruby => 1,
205 wakaba 1.118 script => 1, noscript => 1, 'event-source' => 1, eventsource => 1,
206     command => 1, bb => 1,
207 wakaba 1.119 input => 1, button => 1, label => 1, select => 1, datalist => 1,
208 wakaba 1.121 textarea => 1, output => 1,
209 wakaba 1.29 datagrid => 1, ## ISSUE: "Interactive element" in the spec.
210     ## NOTE: |area| is allowed only as a descendant of |map|.
211     area => 1,
212    
213 wakaba 1.124 ## NOTE: Transparent.
214     a => 1, ins => 1, del => 1, font => 1,
215 wakaba 1.29
216 wakaba 1.72 ## NOTE: If there is a |menu| ancestor, phrasing. Otherwise, flow.
217 wakaba 1.29 menu => 1,
218    
219     img => 1, iframe => 1, embed => 1, object => 1, video => 1, audio => 1,
220     canvas => 1,
221     },
222    
223     ## NOTE: Embedded
224     q<http://www.w3.org/1998/Math/MathML> => {math => 1},
225     q<http://www.w3.org/2000/svg> => {svg => 1},
226     };
227    
228 wakaba 1.58 my $HTMLSectioningContent = {
229 wakaba 1.57 $HTML_NS => {
230     section => 1, nav => 1, article => 1, aside => 1,
231     ## NOTE: |body| is only allowed in |html| element.
232     body => 1,
233     },
234     };
235    
236 wakaba 1.58 my $HTMLSectioningRoot = {
237 wakaba 1.29 $HTML_NS => {
238 wakaba 1.58 blockquote => 1, datagrid => 1, figure => 1, td => 1,
239 wakaba 1.29 },
240     };
241    
242     my $HTMLHeadingContent = {
243     $HTML_NS => {
244     h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, header => 1,
245     },
246     };
247    
248     my $HTMLPhrasingContent = {
249 wakaba 1.72 ## NOTE: All phrasing content is also flow content.
250 wakaba 1.29 $HTML_NS => {
251 wakaba 1.38 br => 1, q => 1, cite => 1, em => 1, strong => 1, small => 1, mark => 1,
252 wakaba 1.29 dfn => 1, abbr => 1, time => 1, progress => 1, meter => 1, code => 1,
253     var => 1, samp => 1, kbd => 1, sub => 1, sup => 1, span => 1, i => 1,
254 wakaba 1.99 b => 1, bdo => 1, ruby => 1,
255 wakaba 1.118 script => 1, noscript => 1, 'event-source' => 1, eventsource => 1,
256     command => 1, bb => 1,
257 wakaba 1.119 input => 1, button => 1, label => 1, select => 1, datalist => 1,
258 wakaba 1.121 textarea => 1, output => 1,
259 wakaba 1.29 datagrid => 1, ## ISSUE: "Interactive element" in the spec.
260     ## NOTE: |area| is allowed only as a descendant of |map|.
261     area => 1,
262    
263     ## NOTE: Transparent.
264 wakaba 1.124 a => 1, ins => 1, del => 1, font => 1,
265 wakaba 1.29
266 wakaba 1.72 ## NOTE: If there is a |menu| ancestor, phrasing. Otherwise, flow.
267 wakaba 1.29 menu => 1,
268    
269     img => 1, iframe => 1, embed => 1, object => 1, video => 1, audio => 1,
270     canvas => 1,
271     },
272    
273     ## NOTE: Embedded
274     q<http://www.w3.org/1998/Math/MathML> => {math => 1},
275     q<http://www.w3.org/2000/svg> => {svg => 1},
276    
277     ## NOTE: And non-inter-element-whitespace text nodes.
278     };
279    
280 wakaba 1.40 ## $HTMLEmbeddedContent: See Whatpm::ContentChecker.
281 wakaba 1.29
282     my $HTMLInteractiveContent = {
283     $HTML_NS => {
284     a => 1,
285 wakaba 1.130 label => 1, input => 1, button => 1, select => 1, textarea => 1,
286     details => 1, datagrid => 1, bb => 1,
287    
288     ## NOTE: When "controls" attribute is specified.
289     video => 1, audio => 1,
290    
291     ## NOTE: When "type=toolbar" attribute is specified.
292     menu => 1,
293 wakaba 1.29 },
294     };
295    
296 wakaba 1.130 our $IsInHTMLInteractiveContent; # See Whatpm::ContentChecker.
297    
298 wakaba 1.36 ## NOTE: $HTMLTransparentElements: See Whatpm::ContentChecker.
299     ## NOTE: Semi-transparent elements: See Whatpm::ContentChecker.
300    
301     ## -- Common attribute syntacx checkers
302    
303 wakaba 1.1 our $AttrChecker;
304 wakaba 1.82 our $AttrStatus;
305 wakaba 1.1
306     my $GetHTMLEnumeratedAttrChecker = sub {
307     my $states = shift; # {value => conforming ? 1 : -1}
308     return sub {
309     my ($self, $attr) = @_;
310     my $value = lc $attr->value; ## TODO: ASCII case insensitibility?
311     if ($states->{$value} > 0) {
312     #
313     } elsif ($states->{$value}) {
314 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'enumerated:non-conforming',
315     level => $self->{level}->{must});
316 wakaba 1.1 } else {
317 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'enumerated:invalid',
318     level => $self->{level}->{must});
319 wakaba 1.1 }
320     };
321     }; # $GetHTMLEnumeratedAttrChecker
322    
323     my $GetHTMLBooleanAttrChecker = sub {
324     my $local_name = shift;
325     return sub {
326     my ($self, $attr) = @_;
327 wakaba 1.88 my $value = lc $attr->value; ## TODO: case
328 wakaba 1.1 unless ($value eq $local_name or $value eq '') {
329 wakaba 1.88 $self->{onerror}->(node => $attr, type => 'boolean:invalid',
330 wakaba 1.104 level => $self->{level}->{must});
331 wakaba 1.1 }
332     };
333     }; # $GetHTMLBooleanAttrChecker
334    
335 wakaba 1.8 ## Unordered set of space-separated tokens
336 wakaba 1.92 my $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker = sub {
337     my $allowed_words = shift;
338     return sub {
339     my ($self, $attr) = @_;
340     my %word;
341 wakaba 1.132 for my $word (grep {length $_}
342     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
343 wakaba 1.92 unless ($word{$word}) {
344     $word{$word} = 1;
345     if (not defined $allowed_words or
346     $allowed_words->{$word}) {
347     #
348     } else {
349 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'word not allowed',
350 wakaba 1.92 value => $word,
351 wakaba 1.104 level => $self->{level}->{must});
352 wakaba 1.92 }
353     } else {
354 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate token',
355     value => $word,
356     level => $self->{level}->{must});
357 wakaba 1.92 }
358 wakaba 1.8 }
359 wakaba 1.92 };
360     }; # $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker
361 wakaba 1.8
362 wakaba 1.132 ## |rel| attribute (set of space separated tokens,
363 wakaba 1.1 ## whose allowed values are defined by the section on link types)
364     my $HTMLLinkTypesAttrChecker = sub {
365 wakaba 1.66 my ($a_or_area, $todo, $self, $attr, $item, $element_state) = @_;
366 wakaba 1.1 my %word;
367 wakaba 1.132 for my $word (grep {length $_}
368     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
369 wakaba 1.1 unless ($word{$word}) {
370     $word{$word} = 1;
371 wakaba 1.18 } elsif ($word eq 'up') {
372     #
373 wakaba 1.1 } else {
374 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate token',
375     value => $word,
376     level => $self->{level}->{must});
377 wakaba 1.1 }
378     }
379     ## NOTE: Case sensitive match (since HTML5 spec does not say link
380     ## types are case-insensitive and it says "The value should not
381     ## be confusingly similar to any other defined value (e.g.
382     ## differing only in case).").
383     ## NOTE: Though there is no explicit "MUST NOT" for undefined values,
384     ## "MAY"s and "only ... MAY" restrict non-standard non-registered
385     ## values to be used conformingly.
386 wakaba 1.66
387     my $is_hyperlink;
388     my $is_resource;
389 wakaba 1.1 require Whatpm::_LinkTypeList;
390     our $LinkType;
391     for my $word (keys %word) {
392     my $def = $LinkType->{$word};
393     if (defined $def) {
394     if ($def->{status} eq 'accepted') {
395     if (defined $def->{effect}->[$a_or_area]) {
396     #
397     } else {
398     $self->{onerror}->(node => $attr,
399 wakaba 1.104 type => 'link type:bad context',
400     value => $word,
401 wakaba 1.110 level => $self->{level}->{must});
402 wakaba 1.1 }
403     } elsif ($def->{status} eq 'proposal') {
404 wakaba 1.104 $self->{onerror}->(node => $attr,
405     type => 'link type:proposed',
406     value => $word,
407     level => $self->{level}->{should});
408 wakaba 1.20 if (defined $def->{effect}->[$a_or_area]) {
409     #
410     } else {
411     $self->{onerror}->(node => $attr,
412 wakaba 1.104 type => 'link type:bad context',
413     value => $word,
414     level => $self->{level}->{must});
415 wakaba 1.20 }
416 wakaba 1.1 } else { # rejected or synonym
417     $self->{onerror}->(node => $attr,
418 wakaba 1.104 type => 'link type:non-conforming',
419     value => $word,
420     level => $self->{level}->{must});
421 wakaba 1.1 }
422 wakaba 1.4 if (defined $def->{effect}->[$a_or_area]) {
423     if ($word eq 'alternate') {
424     #
425     } elsif ($def->{effect}->[$a_or_area] eq 'hyperlink') {
426 wakaba 1.66 $is_hyperlink = 1;
427 wakaba 1.4 }
428     }
429 wakaba 1.1 if ($def->{unique}) {
430     unless ($self->{has_link_type}->{$word}) {
431     $self->{has_link_type}->{$word} = 1;
432     } else {
433     $self->{onerror}->(node => $attr,
434 wakaba 1.104 type => 'link type:duplicate',
435     value => $word,
436     level => $self->{level}->{must});
437 wakaba 1.1 }
438     }
439 wakaba 1.66
440     if (defined $def->{effect}->[$a_or_area] and $word ne 'alternate') {
441     $is_hyperlink = 1 if $def->{effect}->[$a_or_area] eq 'hyperlink';
442     $is_resource = 1 if $def->{effect}->[$a_or_area] eq 'external resource';
443     }
444 wakaba 1.1 } else {
445 wakaba 1.104 $self->{onerror}->(node => $attr,
446     type => 'unknown link type',
447     value => $word,
448     level => $self->{level}->{uncertain});
449 wakaba 1.1 }
450     }
451 wakaba 1.66 $is_hyperlink = 1 if $word{alternate} and not $word{stylesheet};
452 wakaba 1.1 ## TODO: The Pingback 1.0 specification, which is referenced by HTML5,
453     ## says that using both X-Pingback: header field and HTML
454     ## <link rel=pingback> is deprecated and if both appears they
455     ## SHOULD contain exactly the same value.
456     ## ISSUE: Pingback 1.0 specification defines the exact representation
457     ## of its link element, which cannot be tested by the current arch.
458     ## ISSUE: Pingback 1.0 specification says that the document MUST NOT
459     ## include any string that matches to the pattern for the rel=pingback link,
460     ## which again inpossible to test.
461     ## ISSUE: rel=pingback href MUST NOT include entities other than predefined 4.
462 wakaba 1.12
463     ## NOTE: <link rel="up index"><link rel="up up index"> is not an error.
464 wakaba 1.17 ## NOTE: We can't check "If the page is part of multiple hierarchies,
465     ## then they SHOULD be described in different paragraphs.".
466 wakaba 1.66
467     $todo->{has_hyperlink_link_type} = 1 if $is_hyperlink;
468     if ($is_hyperlink or $a_or_area) {
469     $element_state->{uri_info}->{href}->{type}->{hyperlink} = 1;
470     }
471     if ($is_resource and not $a_or_area) {
472     $element_state->{uri_info}->{href}->{type}->{resource} = 1;
473     }
474 wakaba 1.96
475     $element_state->{link_rel} = \%word;
476 wakaba 1.1 }; # $HTMLLinkTypesAttrChecker
477 wakaba 1.20
478     ## TODO: "When an author uses a new type not defined by either this specification or the Wiki page, conformance checkers should offer to add the value to the Wiki, with the details described above, with the "proposal" status."
479 wakaba 1.1
480     ## URI (or IRI)
481     my $HTMLURIAttrChecker = sub {
482 wakaba 1.66 my ($self, $attr, $item, $element_state) = @_;
483 wakaba 1.1 ## ISSUE: Relative references are allowed? (RFC 3987 "IRI" is an absolute reference with optional fragment identifier.)
484     my $value = $attr->value;
485     Whatpm::URIChecker->check_iri_reference ($value, sub {
486 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
487 wakaba 1.106 }), $self->{level};
488 wakaba 1.17 $self->{has_uri_attr} = 1; ## TODO: <html manifest>
489 wakaba 1.66
490     my $attr_name = $attr->name;
491     $element_state->{uri_info}->{$attr_name}->{node} = $attr;
492     ## TODO: absolute
493     push @{$self->{return}->{uri}->{$value} ||= []},
494     $element_state->{uri_info}->{$attr_name};
495 wakaba 1.1 }; # $HTMLURIAttrChecker
496    
497     ## A space separated list of one or more URIs (or IRIs)
498     my $HTMLSpaceURIsAttrChecker = sub {
499     my ($self, $attr) = @_;
500 wakaba 1.66
501     my $type = {ping => 'action',
502     profile => 'namespace',
503     archive => 'resource'}->{$attr->name};
504    
505 wakaba 1.1 my $i = 0;
506 wakaba 1.132 for my $value (split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
507 wakaba 1.1 Whatpm::URIChecker->check_iri_reference ($value, sub {
508 wakaba 1.104 $self->{onerror}->(value => $value, @_, node => $attr, index => $i);
509 wakaba 1.106 }, $self->{level});
510 wakaba 1.66
511     ## TODO: absolute
512     push @{$self->{return}->{uri}->{$value} ||= []},
513 wakaba 1.67 {node => $attr, type => {$type => 1}};
514 wakaba 1.66
515 wakaba 1.1 $i++;
516     }
517 wakaba 1.67 ## ISSUE: Relative references? (especially, in profile="")
518 wakaba 1.1 ## ISSUE: Leading or trailing white spaces are conformant?
519     ## ISSUE: A sequence of white space characters are conformant?
520     ## ISSUE: A zero-length string is conformant? (It does contain a relative reference, i.e. same as base URI.)
521 wakaba 1.132 ## ISSUE: What is "space"?
522 wakaba 1.1 ## NOTE: Duplication seems not an error.
523 wakaba 1.4 $self->{has_uri_attr} = 1;
524 wakaba 1.1 }; # $HTMLSpaceURIsAttrChecker
525    
526 wakaba 1.132 ## NOTE: "Valid datetime".
527 wakaba 1.1 my $HTMLDatetimeAttrChecker = sub {
528     my ($self, $attr) = @_;
529     my $value = $attr->value;
530     ## ISSUE: "space", not "space character" (in parsing algorihtm, "space character")
531 wakaba 1.132 if ($value =~ /\A([0-9]{4})-([0-9]{2})-([0-9]{2})
532     (?>[\x09\x0A\x0C\x0D\x20]+
533     (?>T[\x09\x0A\x0C\x0D\x20]*)?|T[\x09\x0A\x0C\x0D\x20]*)
534     ([0-9]{2}):([0-9]{2})(?>:([0-9]{2}))?(?>\.([0-9]+))?
535     [\x09\x0A\x0C\x0D\x20]*
536     (?>Z|[+-]([0-9]{2}):([0-9]{2}))\z/x) {
537 wakaba 1.1 my ($y, $M, $d, $h, $m, $s, $f, $zh, $zm)
538     = ($1, $2, $3, $4, $5, $6, $7, $8, $9);
539     if (0 < $M and $M < 13) { ## ISSUE: This is not explicitly specified (though in parsing algorithm)
540 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'datetime:bad day',
541     level => $self->{level}->{must})
542 wakaba 1.1 if $d < 1 or
543     $d > [0, 31, 29, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]->[$M];
544 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'datetime:bad day',
545     level => $self->{level}->{must})
546 wakaba 1.1 if $M == 2 and $d == 29 and
547     not ($y % 400 == 0 or ($y % 4 == 0 and $y % 100 != 0));
548     } else {
549 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'datetime:bad month',
550     level => $self->{level}->{must});
551 wakaba 1.1 }
552 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'datetime:bad hour',
553     level => $self->{level}->{must}) if $h > 23;
554     $self->{onerror}->(node => $attr, type => 'datetime:bad minute',
555     level => $self->{level}->{must}) if $m > 59;
556     $self->{onerror}->(node => $attr, type => 'datetime:bad second',
557     level => $self->{level}->{must})
558 wakaba 1.1 if defined $s and $s > 59;
559 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'datetime:bad timezone hour',
560     level => $self->{level}->{must}) if $zh > 23;
561     $self->{onerror}->(node => $attr, type => 'datetime:bad timezone minute',
562     level => $self->{level}->{must}) if $zm > 59;
563 wakaba 1.1 ## ISSUE: Maybe timezone -00:00 should have same semantics as in RFC 3339.
564     } else {
565 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'datetime:syntax error',
566     level => $self->{level}->{must});
567 wakaba 1.1 }
568     }; # $HTMLDatetimeAttrChecker
569    
570     my $HTMLIntegerAttrChecker = sub {
571     my ($self, $attr) = @_;
572     my $value = $attr->value;
573     unless ($value =~ /\A-?[0-9]+\z/) {
574 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'integer:syntax error',
575     level => $self->{level}->{must});
576 wakaba 1.1 }
577     }; # $HTMLIntegerAttrChecker
578    
579     my $GetHTMLNonNegativeIntegerAttrChecker = sub {
580     my $range_check = shift;
581     return sub {
582     my ($self, $attr) = @_;
583     my $value = $attr->value;
584     if ($value =~ /\A[0-9]+\z/) {
585     unless ($range_check->($value + 0)) {
586 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'nninteger:out of range',
587     level => $self->{level}->{must});
588 wakaba 1.1 }
589     } else {
590     $self->{onerror}->(node => $attr,
591 wakaba 1.104 type => 'nninteger:syntax error',
592     level => $self->{level}->{must});
593 wakaba 1.1 }
594     };
595     }; # $GetHTMLNonNegativeIntegerAttrChecker
596    
597     my $GetHTMLFloatingPointNumberAttrChecker = sub {
598     my $range_check = shift;
599     return sub {
600     my ($self, $attr) = @_;
601     my $value = $attr->value;
602 wakaba 1.90 if ($value =~ /\A-?[0-9]+(?>\.[0-9]*)?\z/ or
603     $value =~ /\A-?\.[0-9]+\z/) {
604 wakaba 1.1 unless ($range_check->($value + 0)) {
605 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'float:out of range',
606     level => $self->{level}->{must});
607 wakaba 1.1 }
608     } else {
609     $self->{onerror}->(node => $attr,
610 wakaba 1.104 type => 'float:syntax error',
611     level => $self->{level}->{must});
612 wakaba 1.1 }
613     };
614     }; # $GetHTMLFloatingPointNumberAttrChecker
615    
616 wakaba 1.86 ## HTML4 %Length;
617     my $HTMLLengthAttrChecker = sub {
618     my ($self, $attr) = @_;
619     my $value = $attr->value;
620     unless ($value =~ /\A[0-9]+%?\z/) {
621     $self->{onerror}->(node => $attr, type => 'length:syntax error',
622 wakaba 1.104 level => $self->{level}->{must});
623 wakaba 1.86 }
624    
625     ## NOTE: HTML4 definition is too vague - it does not define the syntax
626     ## of percentage value at all (!).
627     }; # $HTMLLengthAttrChecker
628    
629 wakaba 1.1 ## "A valid MIME type, optionally with parameters. [RFC 2046]"
630     ## ISSUE: RFC 2046 does not define syntax of media types.
631     ## ISSUE: The definition of "a valid MIME type" is unknown.
632     ## Syntactical correctness?
633     my $HTMLIMTAttrChecker = sub {
634     my ($self, $attr) = @_;
635     my $value = $attr->value;
636     ## ISSUE: RFC 2045 Content-Type header field allows insertion
637     ## of LWS/comments between tokens. Is it allowed in HTML? Maybe no.
638     ## ISSUE: RFC 2231 extension? Maybe no.
639     my $lws0 = qr/(?>(?>\x0D\x0A)?[\x09\x20])*/;
640     my $token = qr/[\x21\x23-\x27\x2A\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+/;
641     my $qs = qr/"(?>[\x00-\x0C\x0E-\x21\x23-\x5B\x5D-\x7E]|\x0D\x0A[\x09\x20]|\x5C[\x00-\x7F])*"/;
642     if ($value =~ m#\A$lws0($token)$lws0/$lws0($token)$lws0((?>;$lws0$token$lws0=$lws0(?>$token|$qs)$lws0)*)\z#) {
643     my @type = ($1, $2);
644     my $param = $3;
645     while ($param =~ s/^;$lws0($token)$lws0=$lws0(?>($token)|($qs))$lws0//) {
646     if (defined $2) {
647     push @type, $1 => $2;
648     } else {
649     my $n = $1;
650     my $v = $2;
651     $v =~ s/\\(.)/$1/gs;
652     push @type, $n => $v;
653     }
654     }
655     require Whatpm::IMTChecker;
656 wakaba 1.109 my $ic = Whatpm::IMTChecker->new;
657     $ic->{level} = $self->{level};
658     $ic->check_imt (sub {
659 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
660 wakaba 1.1 }, @type);
661     } else {
662 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'IMT:syntax error',
663     level => $self->{level}->{must});
664 wakaba 1.1 }
665     }; # $HTMLIMTAttrChecker
666    
667     my $HTMLLanguageTagAttrChecker = sub {
668 wakaba 1.7 ## NOTE: See also $AtomLanguageTagAttrChecker in Atom.pm.
669    
670 wakaba 1.1 my ($self, $attr) = @_;
671 wakaba 1.6 my $value = $attr->value;
672     require Whatpm::LangTag;
673     Whatpm::LangTag->check_rfc3066_language_tag ($value, sub {
674 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
675 wakaba 1.106 }, $self->{level});
676 wakaba 1.1 ## ISSUE: RFC 4646 (3066bis)?
677 wakaba 1.6
678     ## TODO: testdata
679 wakaba 1.1 }; # $HTMLLanguageTagAttrChecker
680    
681     ## "A valid media query [MQ]"
682     my $HTMLMQAttrChecker = sub {
683     my ($self, $attr) = @_;
684 wakaba 1.104 $self->{onerror}->(node => $attr,
685     type => 'media query',
686     level => $self->{level}->{uncertain});
687 wakaba 1.1 ## ISSUE: What is "a valid media query"?
688     }; # $HTMLMQAttrChecker
689    
690     my $HTMLEventHandlerAttrChecker = sub {
691     my ($self, $attr) = @_;
692 wakaba 1.104 $self->{onerror}->(node => $attr,
693     type => 'event handler',
694     level => $self->{level}->{uncertain});
695 wakaba 1.1 ## TODO: MUST contain valid ECMAScript code matching the
696     ## ECMAScript |FunctionBody| production. [ECMA262]
697     ## ISSUE: MUST be ES3? E4X? ES4? JS1.x?
698     ## ISSUE: Automatic semicolon insertion does not apply?
699     ## ISSUE: Other script languages?
700     }; # $HTMLEventHandlerAttrChecker
701    
702     my $HTMLUsemapAttrChecker = sub {
703     my ($self, $attr) = @_;
704 wakaba 1.100 ## MUST be a valid hash-name reference to a |map| element.
705 wakaba 1.1 my $value = $attr->value;
706     if ($value =~ s/^#//) {
707 wakaba 1.100 ## NOTE: |usemap="#"| is conforming, though it identifies no |map| element
708     ## according to the "rules for parsing a hash-name reference" algorithm.
709     ## The document is non-conforming anyway, since |<map name="">| (empty
710     ## name) is non-conforming.
711 wakaba 1.1 push @{$self->{usemap}}, [$value => $attr];
712     } else {
713 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'hashref:syntax error',
714     level => $self->{level}->{must});
715 wakaba 1.1 }
716 wakaba 1.100 ## NOTE: Space characters in hash-name references are conforming.
717 wakaba 1.1 ## ISSUE: UA algorithm for matching is case-insensitive; IDs only different in cases should be reported
718     }; # $HTMLUsemapAttrChecker
719    
720 wakaba 1.76 ## Valid browsing context name
721     my $HTMLBrowsingContextNameAttrChecker = sub {
722     my ($self, $attr) = @_;
723     my $value = $attr->value;
724     if ($value =~ /^_/) {
725     $self->{onerror}->(node => $attr, type => 'window name:reserved',
726 wakaba 1.104 level => $self->{level}->{must},
727 wakaba 1.76 value => $value);
728     } elsif (length $value) {
729     #
730     } else {
731     $self->{onerror}->(node => $attr, type => 'window name:empty',
732 wakaba 1.104 level => $self->{level}->{must});
733 wakaba 1.76 }
734     }; # $HTMLBrowsingContextNameAttrChecker
735    
736     ## Valid browsing context name or keyword
737 wakaba 1.1 my $HTMLTargetAttrChecker = sub {
738     my ($self, $attr) = @_;
739     my $value = $attr->value;
740     if ($value =~ /^_/) {
741     $value = lc $value; ## ISSUE: ASCII case-insentitive?
742     unless ({
743 wakaba 1.76 _blank => 1,_self => 1, _parent => 1, _top => 1,
744 wakaba 1.1 }->{$value}) {
745     $self->{onerror}->(node => $attr,
746 wakaba 1.76 type => 'window name:reserved',
747 wakaba 1.104 level => $self->{level}->{must},
748 wakaba 1.76 value => $value);
749 wakaba 1.1 }
750 wakaba 1.76 } elsif (length $value) {
751     #
752 wakaba 1.1 } else {
753 wakaba 1.76 $self->{onerror}->(node => $attr, type => 'window name:empty',
754 wakaba 1.104 level => $self->{level}->{must});
755 wakaba 1.1 }
756     }; # $HTMLTargetAttrChecker
757    
758 wakaba 1.23 my $HTMLSelectorsAttrChecker = sub {
759     my ($self, $attr) = @_;
760    
761     ## ISSUE: Namespace resolution?
762    
763     my $value = $attr->value;
764    
765     require Whatpm::CSS::SelectorsParser;
766     my $p = Whatpm::CSS::SelectorsParser->new;
767     $p->{pseudo_class}->{$_} = 1 for qw/
768     active checked disabled empty enabled first-child first-of-type
769     focus hover indeterminate last-child last-of-type link only-child
770     only-of-type root target visited
771     lang nth-child nth-last-child nth-of-type nth-last-of-type not
772     -manakai-contains -manakai-current
773     /;
774    
775     $p->{pseudo_element}->{$_} = 1 for qw/
776     after before first-letter first-line
777     /;
778    
779 wakaba 1.104 $p->{level} = $self->{level};
780 wakaba 1.23 $p->{onerror} = sub {
781 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
782 wakaba 1.23 };
783     $p->parse_string ($value);
784     }; # $HTMLSelectorsAttrChecker
785    
786 wakaba 1.66 my $HTMLAccesskeyAttrChecker = sub {
787     my ($self, $attr) = @_;
788    
789     ## NOTE: "character" or |%Character;| in HTML4.
790    
791     my $value = $attr->value;
792     if (length $value != 1) {
793     $self->{onerror}->(node => $attr, type => 'char:syntax error',
794 wakaba 1.105 level => $self->{level}->{html4_fact});
795 wakaba 1.66 }
796    
797     ## NOTE: "Note. Authors should consider the input method of the expected
798     ## reader when specifying an accesskey." [HTML4] This is hard to implement,
799     ## since it depends on keyboard and so on.
800     ## NOTE: "We recommend that authors include the access key in label text
801     ## or wherever the access key is to apply." [HTML4] (informative)
802     }; # $HTMLAccesskeyAttrChecker
803    
804 wakaba 1.129 my $HTMLCharsetChecker = sub ($$$;$) {
805     my ($charset_value, $self, $attr, $ascii_compat) = @_;
806    
807     ## NOTE: This code is used for |charset=""| attributes, |charset=|
808     ## portion of the |content=""| attributes, and |accept-charset=""|
809     ## attributes.
810 wakaba 1.91
811     ## NOTE: Though the case-sensitivility of |charset| attribute value
812     ## is not explicitly spelled in the HTML5 spec, the Character Set
813     ## registry of IANA, which is referenced from HTML5 spec, says that
814     ## charset name is case-insensitive.
815     $charset_value =~ tr/A-Z/a-z/; ## NOTE: ASCII Case-insensitive.
816    
817     require Message::Charset::Info;
818     my $charset = $Message::Charset::Info::IANACharset->{$charset_value};
819    
820     ## ISSUE: What is "valid character encoding name"? Syntactically valid?
821     ## Syntactically valid and registered? What about x-charset names?
822     unless (Message::Charset::Info::is_syntactically_valid_iana_charset_name
823     ($charset_value)) {
824     $self->{onerror}->(node => $attr,
825 wakaba 1.104 type => 'charset:syntax error',
826     value => $charset_value,
827     level => $self->{level}->{must});
828 wakaba 1.91 }
829    
830     if ($charset) {
831     ## ISSUE: What is "the preferred name for that encoding" (for a charset
832     ## with no "preferred MIME name" label)?
833     my $charset_status = $charset->{iana_names}->{$charset_value} || 0;
834     if (($charset_status &
835     Message::Charset::Info::PREFERRED_CHARSET_NAME ())
836     != Message::Charset::Info::PREFERRED_CHARSET_NAME ()) {
837     $self->{onerror}->(node => $attr,
838 wakaba 1.104 type => 'charset:not preferred',
839     value => $charset_value,
840     level => $self->{level}->{must});
841 wakaba 1.91 }
842 wakaba 1.129
843 wakaba 1.91 if (($charset_status &
844     Message::Charset::Info::REGISTERED_CHARSET_NAME ())
845     != Message::Charset::Info::REGISTERED_CHARSET_NAME ()) {
846     if ($charset_value =~ /^x-/) {
847     $self->{onerror}->(node => $attr,
848 wakaba 1.104 type => 'charset:private',
849     value => $charset_value,
850     level => $self->{level}->{good});
851 wakaba 1.91 } else {
852     $self->{onerror}->(node => $attr,
853 wakaba 1.104 type => 'charset:not registered',
854     value => $charset_value,
855     level => $self->{level}->{good});
856 wakaba 1.91 }
857     }
858 wakaba 1.129
859     if ($ascii_compat) {
860     if ($charset->{category} &
861     Message::Charset::Info::CHARSET_CATEGORY_ASCII_COMPAT ()) {
862     #
863     } else {
864     $self->{onerror}->(node => $attr,
865     type => 'charset:not ascii compat',
866     value => $charset_value,
867     level => $self->{level}->{must});
868     }
869     }
870    
871 wakaba 1.91 ## TODO: non-preferred-name error for following cases:
872     } elsif ($charset_value =~ /^x-/) {
873     $self->{onerror}->(node => $attr,
874 wakaba 1.104 type => 'charset:private',
875     value => $charset_value,
876     level => $self->{level}->{good});
877 wakaba 1.129
878     ## NOTE: Whether this is an ASCII-compatible character encoding or
879     ## not is unknown.
880 wakaba 1.91 } else {
881     $self->{onerror}->(node => $attr,
882 wakaba 1.104 type => 'charset:not registered',
883     value => $charset_value,
884     level => $self->{level}->{good});
885 wakaba 1.129
886     ## NOTE: Whether this is an ASCII-compatible character encoding or
887     ## not is unknown.
888 wakaba 1.91 }
889    
890     return ($charset, $charset_value);
891     }; # $HTMLCharsetChecker
892    
893 wakaba 1.129 ## NOTE: "An ordered set of space-separated tokens" where "each token
894     ## MUST be the preferred name of an ASCII-compatible character
895     ## encoding".
896     my $HTMLCharsetsAttrChecker = sub {
897     my ($self, $attr) = @_;
898    
899     ## ISSUE: "ordered set of space-separated tokens" is not defined.
900    
901 wakaba 1.132 my @value = grep {length $_} split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value;
902 wakaba 1.129
903     ## ISSUE: Uniqueness is not enforced.
904    
905     for my $charset (@value) {
906     $HTMLCharsetChecker->($charset, $self, $attr, 1);
907     }
908    
909     ## ISSUE: Shift_JIS is ASCII-compatible? What about ISO-2022-JP?
910     }; # $HTMLCharsetsAttrChecker
911    
912 wakaba 1.68 my $HTMLColorAttrChecker = sub {
913     my ($self, $attr) = @_;
914    
915     ## NOTE: HTML4 "color" or |%Color;|
916    
917     my $value = $attr->value;
918    
919     if ($value !~ /\A(?>#[0-9A-F]+|black|silver|gray|white|maroon|red|purple|fuchsia|green|lime|olive|yellow|navy|blue|teal|aqua)\z/i) {
920 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'color:syntax error',
921 wakaba 1.105 level => $self->{level}->{html4_fact});
922 wakaba 1.68 }
923    
924     ## TODO: HTML4 has some guideline on usage of color.
925     }; # $HTMLColorAttrChecker
926    
927 wakaba 1.79 my $HTMLRefOrTemplateAttrChecker = sub {
928     my ($self, $attr) = @_;
929     $HTMLURIAttrChecker->(@_);
930    
931     my $attr_name = $attr->name;
932    
933     if ($attr_name eq 'ref') {
934     unless ($attr->owner_element->has_attribute_ns (undef, 'template')) {
935     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
936 wakaba 1.104 level => $self->{level}->{must});
937 wakaba 1.79 }
938     }
939    
940     my $doc = $attr->owner_document;
941     my $doc_uri = $doc->document_uri;
942     my $uri = $doc->implementation
943     ->create_uri_reference ($attr->value)
944     ->get_absolute_reference ($doc_uri);
945     my $no_frag_uri = $uri->clone;
946     $no_frag_uri->uri_fragment (undef);
947     if ((defined $doc_uri and $doc_uri eq $no_frag_uri) or
948     (not defined $doc_uri and $no_frag_uri eq '')) {
949     my $fragid = $uri->uri_fragment;
950     if (defined $fragid) {
951     push @{$self->{$attr_name}}, [$fragid => $attr];
952     } else {
953     DOCEL: {
954     last DOCEL unless $attr_name eq 'template';
955    
956     my $docel = $doc->document_element;
957     if ($docel) {
958     my $nsuri = $docel->namespace_uri;
959     if (defined $nsuri and $nsuri eq $HTML_NS) {
960     if ($docel->manakai_local_name eq 'datatemplate') {
961     last DOCEL;
962     }
963     }
964     }
965    
966     $self->{onerror}->(node => $attr, type => 'template:not template',
967 wakaba 1.104 level => $self->{level}->{must});
968 wakaba 1.79 } # DOCEL
969     }
970     } else {
971     ## TODO: An external document is referenced.
972     ## The document MUST be an HTML or XML document.
973     ## If there is a fragment identifier, it MUST point a part of the doc.
974     ## If the attribute is |template|, the pointed part MUST be a
975     ## |datatemplat| element.
976     ## If no fragment identifier is specified, the root element MUST be
977     ## a |datatemplate| element when the attribute is |template|.
978     }
979     }; # $HTMLRefOrTemplateAttrChecker
980    
981 wakaba 1.83 my $HTMLRepeatIndexAttrChecker = sub {
982     my ($self, $attr) = @_;
983    
984     if (defined $attr->namespace_uri) {
985     my $oe = $attr->owner_element;
986     my $oe_nsuri = $oe->namespace_uri;
987 wakaba 1.128 if (defined $oe_nsuri or $oe_nsuri eq $HTML_NS) { ## TODO: wrong?
988 wakaba 1.83 $self->{onerror}->(node => $attr, type => 'attribute not allowed',
989 wakaba 1.104 level => $self->{level}->{must});
990 wakaba 1.83 }
991     }
992    
993     $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 })->(@_);
994     }; # $HTMLRepeatIndexAttrChecker
995    
996 wakaba 1.1 my $HTMLAttrChecker = {
997 wakaba 1.58 ## TODO: aria-* ## TODO: svg:*/@aria-* [HTML5ROLE] -> [STATES]
998 wakaba 1.1 id => sub {
999 wakaba 1.135 my ($self, $attr, $item, $element_state) = @_;
1000 wakaba 1.1 my $value = $attr->value;
1001     if (length $value > 0) {
1002     if ($self->{id}->{$value}) {
1003 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate ID',
1004     level => $self->{level}->{must});
1005 wakaba 1.1 push @{$self->{id}->{$value}}, $attr;
1006     } else {
1007     $self->{id}->{$value} = [$attr];
1008 wakaba 1.135 $self->{id_type}->{$value} = $element_state->{id_type} || '';
1009 wakaba 1.1 }
1010 wakaba 1.132 if ($value =~ /[\x09\x0A\x0C\x0D\x20]/) {
1011 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'space in ID',
1012     level => $self->{level}->{must});
1013 wakaba 1.1 }
1014     } else {
1015     ## NOTE: MUST contain at least one character
1016 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'empty attribute value',
1017     level => $self->{level}->{must});
1018 wakaba 1.1 }
1019     },
1020     title => sub {}, ## NOTE: No conformance creteria
1021     lang => sub {
1022     my ($self, $attr) = @_;
1023 wakaba 1.6 my $value = $attr->value;
1024     if ($value eq '') {
1025     #
1026     } else {
1027     require Whatpm::LangTag;
1028     Whatpm::LangTag->check_rfc3066_language_tag ($value, sub {
1029 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
1030 wakaba 1.106 }, $self->{level});
1031 wakaba 1.6 }
1032 wakaba 1.1 ## ISSUE: RFC 4646 (3066bis)?
1033 wakaba 1.6
1034     ## TODO: test data
1035 wakaba 1.111
1036     ## NOTE: Inconsistency between |lang| and |xml:lang| attributes are
1037     ## non-conforming. Such errors are detected by the checkers of
1038     ## |{}xml:lang| and |{xml}:lang| attributes.
1039 wakaba 1.1 },
1040     dir => $GetHTMLEnumeratedAttrChecker->({ltr => 1, rtl => 1}),
1041     class => sub {
1042     my ($self, $attr) = @_;
1043 wakaba 1.132
1044     ## NOTE: "Unordered set of unique space-separated tokens".
1045    
1046 wakaba 1.1 my %word;
1047 wakaba 1.132 for my $word (grep {length $_}
1048     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
1049 wakaba 1.1 unless ($word{$word}) {
1050     $word{$word} = 1;
1051     push @{$self->{return}->{class}->{$word}||=[]}, $attr;
1052     } else {
1053 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate token',
1054     value => $word,
1055     level => $self->{level}->{must});
1056 wakaba 1.1 }
1057     }
1058     },
1059 wakaba 1.63 contenteditable => $GetHTMLEnumeratedAttrChecker->({
1060     true => 1, false => 1, '' => 1,
1061     }),
1062 wakaba 1.1 contextmenu => sub {
1063     my ($self, $attr) = @_;
1064     my $value = $attr->value;
1065     push @{$self->{contextmenu}}, [$value => $attr];
1066     ## ISSUE: "The value must be the ID of a menu element in the DOM."
1067     ## What is "in the DOM"? A menu Element node that is not part
1068     ## of the Document tree is in the DOM? A menu Element node that
1069     ## belong to another Document tree is in the DOM?
1070     },
1071 wakaba 1.115 hidden => $GetHTMLBooleanAttrChecker->('hidden'),
1072 wakaba 1.60 irrelevant => $GetHTMLBooleanAttrChecker->('irrelevant'),
1073 wakaba 1.79 ref => $HTMLRefOrTemplateAttrChecker,
1074     registrationmark => sub {
1075     my ($self, $attr, $item, $element_state) = @_;
1076    
1077     ## NOTE: Any value is conforming.
1078    
1079     if ($self->{flag}->{in_rule}) {
1080     my $el = $attr->owner_element;
1081     my $ln = $el->manakai_local_name;
1082     if ($ln eq 'nest' or
1083     ($ln eq 'rule' and not $element_state->{in_rule_original})) {
1084     my $nsuri = $el->namespace_uri;
1085     if (defined $nsuri and $nsuri eq $HTML_NS) {
1086     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1087 wakaba 1.104 level => $self->{level}->{must});
1088 wakaba 1.79 }
1089     }
1090     } else {
1091     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1092 wakaba 1.104 level => $self->{level}->{must});
1093 wakaba 1.79 }
1094     },
1095 wakaba 1.80 repeat => sub {
1096     my ($self, $attr) = @_;
1097 wakaba 1.83
1098     if (defined $attr->namespace_uri) {
1099     my $oe = $attr->owner_element;
1100     my $oe_nsuri = $oe->namespace_uri;
1101     if (defined $oe_nsuri or $oe_nsuri eq $HTML_NS) {
1102     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1103 wakaba 1.104 level => $self->{level}->{must});
1104 wakaba 1.83 }
1105     }
1106    
1107 wakaba 1.80 my $value = $attr->value;
1108     if ($value eq 'template') {
1109     #
1110     } elsif ($value =~ /\A-?[0-9]+\z/) {
1111     #
1112     } else {
1113     $self->{onerror}->(node => $attr, type => 'repeat:syntax error',
1114 wakaba 1.104 level => $self->{level}->{must});
1115 wakaba 1.80 }
1116    
1117     ## ISSUE: "Repetition templates may occur anywhere." Does that mean
1118     ## that the attribute MAY be specified to any element, or that the
1119     ## element with that attribute (i.e. a repetition template) can be
1120     ## inserted anywhere in a document tree?
1121     },
1122 wakaba 1.83 'repeat-min' => $HTMLRepeatIndexAttrChecker,
1123     'repeat-max' => $HTMLRepeatIndexAttrChecker,
1124     'repeat-start' => $HTMLRepeatIndexAttrChecker,
1125 wakaba 1.80 'repeat-template' => sub {
1126 wakaba 1.83 my ($self, $attr) = @_;
1127    
1128     if (defined $attr->namespace_uri) {
1129     my $oe = $attr->owner_element;
1130     my $oe_nsuri = $oe->namespace_uri;
1131 wakaba 1.128 if (defined $oe_nsuri or $oe_nsuri eq $HTML_NS) { ## TODO: This condition is wrong?
1132 wakaba 1.83 $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1133 wakaba 1.104 level => $self->{level}->{must});
1134 wakaba 1.83 }
1135     }
1136    
1137 wakaba 1.80 ## ISSUE: This attribute has no conformance requirement.
1138     ## ISSUE: Repetition blocks MAY have this attribute. Then, is the
1139     ## attribute allowed on an element that is not a repetition block?
1140     },
1141 wakaba 1.58 ## TODO: role [HTML5ROLE] ## TODO: global @role [XHTML1ROLE]
1142 wakaba 1.128 style => sub {
1143     my ($self, $attr) = @_;
1144    
1145     $self->{onsubdoc}->({s => $attr->value,
1146     container_node => $attr,
1147     media_type => 'text/x-css-inline',
1148     is_char_string => 1});
1149    
1150     ## NOTE: "... MUST still be comprehensible and usable if those
1151     ## attributes were removed" is a semantic requirement, it cannot
1152     ## be tested.
1153     },
1154 wakaba 1.74 tabindex => $HTMLIntegerAttrChecker,
1155 wakaba 1.79 template => $HTMLRefOrTemplateAttrChecker,
1156 wakaba 1.111 'xml:lang' => sub {
1157     my ($self, $attr) = @_;
1158    
1159     if ($attr->owner_document->manakai_is_html) {
1160     $self->{onerror}->(type => 'in HTML:xml:lang',
1161     level => $self->{level}->{info},
1162     node => $attr);
1163     ## NOTE: This is not an error, but the attribute will be ignored.
1164     } else {
1165     $self->{onerror}->(type => 'in XML:xml:lang',
1166     level => $self->{level}->{html5_no_may},
1167     node => $attr);
1168     ## TODO: We need to add test for this error.
1169     }
1170    
1171     my $lang_attr = $attr->owner_element->get_attribute_node_ns
1172     (undef, 'lang');
1173     if ($lang_attr) {
1174     my $lang_attr_value = $lang_attr->value;
1175     $lang_attr_value =~ tr/A-Z/a-z/; ## ASCII case-insensitive
1176     my $value = $attr->value;
1177     $value =~ tr/A-Z/a-z/; ## ASCII case-insensitive
1178     if ($lang_attr_value ne $value) {
1179     $self->{onerror}->(type => 'xml:lang ne lang',
1180     level => $self->{level}->{must},
1181     node => $attr);
1182     }
1183     } else {
1184     $self->{onerror}->(type => 'xml:lang not allowed',
1185     level => $self->{level}->{must},
1186     node => $attr);
1187     ## TODO: We need to add test for <x {xml}:lang {}xml:lang>.
1188     }
1189     },
1190 wakaba 1.74 xmlns => sub {
1191     my ($self, $attr) = @_;
1192     my $value = $attr->value;
1193     unless ($value eq $HTML_NS) {
1194 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'invalid attribute value',
1195     level => $self->{level}->{must});
1196 wakaba 1.74 ## TODO: Should be new "bad namespace" error?
1197     }
1198     unless ($attr->owner_document->manakai_is_html) {
1199 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'in XML:xmlns',
1200     level => $self->{level}->{must});
1201 wakaba 1.74 ## TODO: Test
1202     }
1203    
1204     ## TODO: Should be resolved?
1205     push @{$self->{return}->{uri}->{$value} ||= []},
1206     {node => $attr, type => {namespace => 1}};
1207     },
1208 wakaba 1.1 };
1209    
1210 wakaba 1.79 ## ISSUE: Shouldn't the same-origin policy applied to the datatemplate feature?
1211    
1212 wakaba 1.49 my %HTMLAttrStatus = (
1213 wakaba 1.50 class => FEATURE_HTML5_DEFAULT,
1214     contenteditable => FEATURE_HTML5_DEFAULT,
1215     contextmenu => FEATURE_HTML5_WD,
1216     dir => FEATURE_HTML5_DEFAULT,
1217     draggable => FEATURE_HTML5_LC,
1218 wakaba 1.115 hidden => FEATURE_HTML5_DEFAULT,
1219 wakaba 1.50 id => FEATURE_HTML5_DEFAULT,
1220 wakaba 1.115 irrelevant => FEATURE_HTML5_DROPPED,
1221 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT,
1222     ref => FEATURE_HTML5_AT_RISK,
1223     registrationmark => FEATURE_HTML5_AT_RISK,
1224 wakaba 1.60 repeat => FEATURE_WF2,
1225     'repeat-max' => FEATURE_WF2,
1226     'repeat-min' => FEATURE_WF2,
1227     'repeat-start' => FEATURE_WF2,
1228     'repeat-template' => FEATURE_WF2,
1229 wakaba 1.58 role => FEATURE_HTML5_ROLE,
1230 wakaba 1.78 style => FEATURE_HTML5_DEFAULT,
1231 wakaba 1.50 tabindex => FEATURE_HTML5_DEFAULT,
1232     template => FEATURE_HTML5_AT_RISK,
1233     title => FEATURE_HTML5_DEFAULT,
1234 wakaba 1.74 xmlns => FEATURE_HTML5_DEFAULT,
1235 wakaba 1.49 );
1236    
1237     my %HTMLM12NCommonAttrStatus = (
1238 wakaba 1.120 about => FEATURE_RDFA_CR,
1239 wakaba 1.50 class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1240 wakaba 1.120 content => FEATURE_RDFA_CR,
1241     datatype => FEATURE_RDFA_CR,
1242 wakaba 1.50 dir => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1243 wakaba 1.121 href => FEATURE_RDFA_CR,
1244 wakaba 1.50 id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1245 wakaba 1.61 instanceof => FEATURE_RDFA_LC,
1246 wakaba 1.50 onclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1247     ondblclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1248     onmousedown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1249     onmouseup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1250     onmouseover => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1251     onmousemove => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1252     onmouseout => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1253     onkeypress => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1254     onkeydown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1255     onkeyup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1256 wakaba 1.120 property => FEATURE_RDFA_CR,
1257     rel => FEATURE_RDFA_CR,
1258     resource => FEATURE_RDFA_CR,
1259     rev => FEATURE_RDFA_CR,
1260 wakaba 1.78 #style => FEATURE_HTML5_DEFAULT | FEATURE_XHTMLBASIC11_CR_DEPRECATED |
1261     # FEATURE_M12N10_REC,
1262     style => FEATURE_HTML5_DEFAULT | FEATURE_XHTMLBASIC11_CR |
1263 wakaba 1.55 FEATURE_M12N10_REC,
1264 wakaba 1.50 title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1265 wakaba 1.120 typeof => FEATURE_RDFA_CR,
1266 wakaba 1.49 );
1267    
1268 wakaba 1.82 my %XHTML2CommonAttrStatus = (
1269     ## Core
1270 wakaba 1.108 class => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED,
1271     id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED,
1272 wakaba 1.82 #xml:id
1273     layout => FEATURE_XHTML2_ED,
1274 wakaba 1.108 title => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED,
1275 wakaba 1.82
1276     ## Hypertext
1277     cite => FEATURE_XHTML2_ED,
1278     href => FEATURE_XHTML2_ED,
1279     hreflang => FEATURE_XHTML2_ED,
1280     hrefmedia => FEATURE_XHTML2_ED,
1281     hreftype => FEATURE_XHTML2_ED,
1282     nextfocus => FEATURE_XHTML2_ED,
1283     prevfocus => FEATURE_XHTML2_ED,
1284     target => FEATURE_XHTML2_ED,
1285     #xml:base
1286    
1287     ## I18N
1288     #xml:lang
1289    
1290     ## Bi-directional
1291 wakaba 1.108 dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED,
1292 wakaba 1.82
1293     ## Edit
1294     edit => FEATURE_XHTML2_ED,
1295     datetime => FEATURE_XHTML2_ED,
1296    
1297     ## Embedding
1298     encoding => FEATURE_XHTML2_ED,
1299     src => FEATURE_XHTML2_ED,
1300     srctype => FEATURE_XHTML2_ED,
1301    
1302     ## Image Map
1303     usemap => FEATURE_XHTML2_ED,
1304     ismap => FEATURE_XHTML2_ED,
1305     shape => FEATURE_XHTML2_ED,
1306     coords => FEATURE_XHTML2_ED,
1307    
1308     ## Media
1309     media => FEATURE_XHTML2_ED,
1310    
1311     ## Metadata
1312     about => FEATURE_XHTML2_ED,
1313     content => FEATURE_XHTML2_ED,
1314     datatype => FEATURE_XHTML2_ED,
1315     instanceof => FEATURE_XHTML2_ED,
1316     property => FEATURE_XHTML2_ED,
1317     rel => FEATURE_XHTML2_ED,
1318     resource => FEATURE_XHTML2_ED,
1319     rev => FEATURE_XHTML2_ED,
1320    
1321     ## Role
1322 wakaba 1.108 role => FEATURE_HTML5_ROLE | FEATURE_XHTML2_ED,
1323 wakaba 1.82
1324     ## Style
1325 wakaba 1.108 style => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED, # "strongly discouraged"
1326 wakaba 1.82 );
1327    
1328     my %HTMLM12NXHTML2CommonAttrStatus = (
1329     %HTMLM12NCommonAttrStatus,
1330     %XHTML2CommonAttrStatus,
1331    
1332 wakaba 1.120 about => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1333 wakaba 1.82 class => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1334 wakaba 1.120 content => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1335     datatype => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1336 wakaba 1.82 dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1337 wakaba 1.121 href => FEATURE_RDFA_CR,
1338 wakaba 1.82 id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1339     instanceof => FEATURE_RDFA_LC | FEATURE_XHTML2_ED,
1340 wakaba 1.120 property => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1341     rel => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1342     resource => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1343     rev => FEATURE_RDFA_CR | FEATURE_XHTML2_ED,
1344 wakaba 1.82 #style => FEATURE_HTML5_DEFAULT | FEATURE_XHTMLBASIC11_CR_DEPRECATED |
1345     # FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1346     style => FEATURE_HTML5_DEFAULT | FEATURE_XHTMLBASIC11_CR |
1347     FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1348     title => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1349 wakaba 1.120 typeof => FEATURE_RDFA_CR,
1350 wakaba 1.82 );
1351    
1352 wakaba 1.1 for (qw/
1353     onabort onbeforeunload onblur onchange onclick oncontextmenu
1354     ondblclick ondrag ondragend ondragenter ondragleave ondragover
1355     ondragstart ondrop onerror onfocus onkeydown onkeypress
1356     onkeyup onload onmessage onmousedown onmousemove onmouseout
1357     onmouseover onmouseup onmousewheel onresize onscroll onselect
1358 wakaba 1.77 onstorage onsubmit onunload
1359 wakaba 1.1 /) {
1360     $HTMLAttrChecker->{$_} = $HTMLEventHandlerAttrChecker;
1361 wakaba 1.50 $HTMLAttrStatus{$_} = FEATURE_HTML5_DEFAULT;
1362 wakaba 1.1 }
1363    
1364 wakaba 1.82 ## NOTE: Non-standard global attributes in the HTML namespace.
1365     $AttrChecker->{$HTML_NS}->{''} = sub {}; # no syntactical checks
1366     $AttrStatus->{$HTML_NS}->{''} = 0; # disallowed and not part of any standard
1367    
1368     $AttrStatus->{$HTML_NS}->{active} = FEATURE_HTML5_DROPPED;
1369     for (qw/repeat repeat-max repeat-min repeat-start repeat-template/) {
1370     $AttrChecker->{$HTML_NS}->{$_} = $HTMLAttrChecker->{$_};
1371     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_WF2;
1372     }
1373    
1374 wakaba 1.120 for (qw/about content datatype property rel resource rev/) {
1375     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_RDFA_CR | FEATURE_XHTML2_ED;
1376 wakaba 1.82 }
1377 wakaba 1.120 $AttrStatus->{$HTML_NS}->{instanceof} = FEATURE_RDFA_LC | FEATURE_XHTML2_ED;
1378     $AttrStatus->{$HTML_NS}->{typeof} = FEATURE_RDFA_CR;
1379 wakaba 1.82 $AttrStatus->{$HTML_NS}->{role} = FEATURE_ROLE_LC;
1380     for (qw/cite coords datetime edit encoding href hreflang hrefmedia hreftype
1381     ismap layout media nextfocus prevfocus shape src srctype style
1382     target usemap/) {
1383     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_XHTML2_ED;
1384     }
1385     for (qw/class dir id title/) {
1386     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_M12N11_LC | FEATURE_XHTML2_ED;
1387     }
1388     for (qw/onclick ondblclick onmousedown onmouseup onmouseover onmousemove
1389     onmouseout onkeypress onkeydown onkeyup/) {
1390     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_M12N11_LC;
1391     }
1392    
1393 wakaba 1.73 my $HTMLDatasetAttrChecker = sub {
1394     ## NOTE: "Authors should ... when the attributes are ignored and
1395     ## any associated CSS dropped, the page is still usable." (semantic
1396     ## constraint.)
1397     }; # $HTMLDatasetAttrChecker
1398    
1399     my $HTMLDatasetAttrStatus = FEATURE_HTML5_DEFAULT;
1400    
1401 wakaba 1.1 my $GetHTMLAttrsChecker = sub {
1402     my $element_specific_checker = shift;
1403 wakaba 1.49 my $element_specific_status = shift;
1404 wakaba 1.1 return sub {
1405 wakaba 1.40 my ($self, $item, $element_state) = @_;
1406     for my $attr (@{$item->{node}->attributes}) {
1407 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
1408     $attr_ns = '' unless defined $attr_ns;
1409     my $attr_ln = $attr->manakai_local_name;
1410     my $checker;
1411 wakaba 1.73 my $status;
1412 wakaba 1.1 if ($attr_ns eq '') {
1413 wakaba 1.122 if ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
1414     $attr_ln !~ /[A-Z]/) {
1415 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
1416     $status = $HTMLDatasetAttrStatus;
1417     } else {
1418     $checker = $element_specific_checker->{$attr_ln}
1419     || $HTMLAttrChecker->{$attr_ln};
1420     $status = $element_specific_status->{$attr_ln};
1421     }
1422 wakaba 1.1 }
1423     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
1424 wakaba 1.40 || $AttrChecker->{$attr_ns}->{''};
1425 wakaba 1.82 $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
1426     || $AttrStatus->{$attr_ns}->{''};
1427     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
1428 wakaba 1.1 if ($checker) {
1429 wakaba 1.66 $checker->($self, $attr, $item, $element_state);
1430 wakaba 1.62 } elsif ($attr_ns eq '' and not $element_specific_status->{$attr_ln}) {
1431 wakaba 1.54 #
1432 wakaba 1.1 } else {
1433 wakaba 1.104 $self->{onerror}->(node => $attr,
1434     type => 'unknown attribute',
1435     level => $self->{level}->{uncertain});
1436 wakaba 1.49 ## ISSUE: No conformance createria for unknown attributes in the spec
1437     }
1438 wakaba 1.82 $self->_attr_status_info ($attr, $status);
1439 wakaba 1.1 }
1440     };
1441     }; # $GetHTMLAttrsChecker
1442    
1443 wakaba 1.40 my %HTMLChecker = (
1444     %Whatpm::ContentChecker::AnyChecker,
1445 wakaba 1.79 check_start => sub {
1446     my ($self, $item, $element_state) = @_;
1447    
1448     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
1449     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
1450     },
1451 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, \%HTMLAttrStatus),
1452 wakaba 1.40 );
1453    
1454     my %HTMLEmptyChecker = (
1455     %HTMLChecker,
1456     check_child_element => sub {
1457     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1458     $child_is_transparent, $element_state) = @_;
1459 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1460     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1461 wakaba 1.40 $self->{onerror}->(node => $child_el,
1462     type => 'element not allowed:minus',
1463 wakaba 1.104 level => $self->{level}->{must});
1464 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1465     #
1466     } else {
1467     $self->{onerror}->(node => $child_el,
1468     type => 'element not allowed:empty',
1469 wakaba 1.104 level => $self->{level}->{must});
1470 wakaba 1.40 }
1471     },
1472     check_child_text => sub {
1473     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1474     if ($has_significant) {
1475     $self->{onerror}->(node => $child_node,
1476     type => 'character not allowed:empty',
1477 wakaba 1.104 level => $self->{level}->{must});
1478 wakaba 1.40 }
1479     },
1480     );
1481    
1482     my %HTMLTextChecker = (
1483     %HTMLChecker,
1484     check_child_element => sub {
1485     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1486     $child_is_transparent, $element_state) = @_;
1487 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1488     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1489 wakaba 1.40 $self->{onerror}->(node => $child_el,
1490     type => 'element not allowed:minus',
1491 wakaba 1.104 level => $self->{level}->{must});
1492 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1493     #
1494     } else {
1495 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed:text',
1496     level => $self->{level}->{must});
1497 wakaba 1.40 }
1498     },
1499     );
1500    
1501 wakaba 1.72 my %HTMLFlowContentChecker = (
1502 wakaba 1.40 %HTMLChecker,
1503     check_child_element => sub {
1504     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1505     $child_is_transparent, $element_state) = @_;
1506 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1507     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1508 wakaba 1.40 $self->{onerror}->(node => $child_el,
1509     type => 'element not allowed:minus',
1510 wakaba 1.104 level => $self->{level}->{must});
1511 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1512     #
1513     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'style') {
1514     if ($element_state->{has_non_style} or
1515     not $child_el->has_attribute_ns (undef, 'scoped')) {
1516 wakaba 1.104 $self->{onerror}->(node => $child_el,
1517 wakaba 1.72 type => 'element not allowed:flow style',
1518 wakaba 1.104 level => $self->{level}->{must});
1519 wakaba 1.40 }
1520 wakaba 1.72 } elsif ($HTMLFlowContent->{$child_nsuri}->{$child_ln}) {
1521 wakaba 1.43 $element_state->{has_non_style} = 1 unless $child_is_transparent;
1522 wakaba 1.40 } else {
1523     $element_state->{has_non_style} = 1;
1524 wakaba 1.104 $self->{onerror}->(node => $child_el,
1525 wakaba 1.72 type => 'element not allowed:flow',
1526 wakaba 1.104 level => $self->{level}->{must})
1527 wakaba 1.40 }
1528     },
1529     check_child_text => sub {
1530     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1531     if ($has_significant) {
1532     $element_state->{has_non_style} = 1;
1533     }
1534     },
1535     check_end => sub {
1536     my ($self, $item, $element_state) = @_;
1537 wakaba 1.95 ## NOTE: A modified copy of the code below is in |datagrid| checker.
1538 wakaba 1.40 if ($element_state->{has_significant}) {
1539 wakaba 1.46 $item->{real_parent_state}->{has_significant} = 1;
1540 wakaba 1.40 } elsif ($item->{transparent}) {
1541     #
1542     } else {
1543     $self->{onerror}->(node => $item->{node},
1544 wakaba 1.104 level => $self->{level}->{should},
1545 wakaba 1.40 type => 'no significant content');
1546     }
1547     },
1548     );
1549    
1550     my %HTMLPhrasingContentChecker = (
1551     %HTMLChecker,
1552     check_child_element => sub {
1553     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1554     $child_is_transparent, $element_state) = @_;
1555 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1556     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1557 wakaba 1.40 $self->{onerror}->(node => $child_el,
1558     type => 'element not allowed:minus',
1559 wakaba 1.104 level => $self->{level}->{must});
1560 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1561     #
1562     } elsif ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
1563     #
1564     } else {
1565     $self->{onerror}->(node => $child_el,
1566     type => 'element not allowed:phrasing',
1567 wakaba 1.104 level => $self->{level}->{must});
1568 wakaba 1.40 }
1569     },
1570 wakaba 1.72 check_end => $HTMLFlowContentChecker{check_end},
1571 wakaba 1.40 ## NOTE: The definition for |li| assumes that the only differences
1572 wakaba 1.72 ## between flow and phrasing content checkers are |check_child_element|
1573 wakaba 1.40 ## and |check_child_text|.
1574     );
1575    
1576 wakaba 1.72 my %HTMLTransparentChecker = %HTMLFlowContentChecker;
1577 wakaba 1.40 ## ISSUE: Significant content rule should be applied to transparent element
1578 wakaba 1.46 ## with parent?
1579 wakaba 1.40
1580 wakaba 1.1 our $Element;
1581     our $ElementDefault;
1582    
1583     $Element->{$HTML_NS}->{''} = {
1584 wakaba 1.40 %HTMLChecker,
1585 wakaba 1.1 };
1586    
1587     $Element->{$HTML_NS}->{html} = {
1588 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1589 wakaba 1.1 is_root => 1,
1590 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
1591 wakaba 1.16 manifest => $HTMLURIAttrChecker,
1592 wakaba 1.67 version => sub {
1593     ## NOTE: According to HTML4 prose, this is a "cdata" attribute.
1594     ## Though DTDs of various versions of HTML define the attribute
1595     ## as |#FIXED|, this conformance checker does no check for
1596     ## the attribute value, since what kind of check should be done
1597     ## is unknown.
1598     },
1599 wakaba 1.49 }, {
1600     %HTMLAttrStatus,
1601 wakaba 1.82 %XHTML2CommonAttrStatus,
1602     class => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_HTML2X_RFC,
1603     dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1604     id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_XHTML10_REC,
1605 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
1606     manifest => FEATURE_HTML5_DEFAULT,
1607 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
1608 wakaba 1.82 version => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1609 wakaba 1.1 }),
1610 wakaba 1.40 check_start => sub {
1611     my ($self, $item, $element_state) = @_;
1612     $element_state->{phase} = 'before head';
1613 wakaba 1.79
1614 wakaba 1.66 $element_state->{uri_info}->{manifest}->{type}->{resource} = 1;
1615 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
1616     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
1617 wakaba 1.40 },
1618     check_child_element => sub {
1619     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1620     $child_is_transparent, $element_state) = @_;
1621 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1622     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1623 wakaba 1.40 $self->{onerror}->(node => $child_el,
1624     type => 'element not allowed:minus',
1625 wakaba 1.104 level => $self->{level}->{must});
1626 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1627     #
1628     } elsif ($element_state->{phase} eq 'before head') {
1629     if ($child_nsuri eq $HTML_NS and $child_ln eq 'head') {
1630     $element_state->{phase} = 'after head';
1631     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'body') {
1632     $self->{onerror}->(node => $child_el,
1633 wakaba 1.104 type => 'ps element missing',
1634     text => 'head',
1635     level => $self->{level}->{must});
1636 wakaba 1.40 $element_state->{phase} = 'after body';
1637     } else {
1638     $self->{onerror}->(node => $child_el,
1639 wakaba 1.104 type => 'element not allowed',
1640     level => $self->{level}->{must});
1641 wakaba 1.40 }
1642     } elsif ($element_state->{phase} eq 'after head') {
1643     if ($child_nsuri eq $HTML_NS and $child_ln eq 'body') {
1644     $element_state->{phase} = 'after body';
1645     } else {
1646     $self->{onerror}->(node => $child_el,
1647 wakaba 1.104 type => 'element not allowed',
1648     level => $self->{level}->{must});
1649 wakaba 1.40 }
1650     } elsif ($element_state->{phase} eq 'after body') {
1651     $self->{onerror}->(node => $child_el,
1652 wakaba 1.104 type => 'element not allowed',
1653     level => $self->{level}->{must});
1654 wakaba 1.40 } else {
1655     die "check_child_element: Bad |html| phase: $element_state->{phase}";
1656     }
1657     },
1658     check_child_text => sub {
1659     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1660     if ($has_significant) {
1661     $self->{onerror}->(node => $child_node,
1662 wakaba 1.104 type => 'character not allowed',
1663     level => $self->{level}->{must});
1664 wakaba 1.40 }
1665     },
1666     check_end => sub {
1667     my ($self, $item, $element_state) = @_;
1668     if ($element_state->{phase} eq 'after body') {
1669     #
1670     } elsif ($element_state->{phase} eq 'before head') {
1671     $self->{onerror}->(node => $item->{node},
1672 wakaba 1.104 type => 'child element missing',
1673     text => 'head',
1674     level => $self->{level}->{must});
1675 wakaba 1.40 $self->{onerror}->(node => $item->{node},
1676 wakaba 1.104 type => 'child element missing',
1677     text => 'body',
1678     level => $self->{level}->{must});
1679 wakaba 1.40 } elsif ($element_state->{phase} eq 'after head') {
1680     $self->{onerror}->(node => $item->{node},
1681 wakaba 1.104 type => 'child element missing',
1682     text => 'body',
1683     level => $self->{level}->{must});
1684 wakaba 1.40 } else {
1685     die "check_end: Bad |html| phase: $element_state->{phase}";
1686     }
1687 wakaba 1.1
1688 wakaba 1.40 $HTMLChecker{check_end}->(@_);
1689     },
1690     };
1691 wakaba 1.25
1692 wakaba 1.40 $Element->{$HTML_NS}->{head} = {
1693 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1694 wakaba 1.67 check_attrs => $GetHTMLAttrsChecker->({
1695     profile => $HTMLSpaceURIsAttrChecker, ## NOTE: MUST be profile URIs.
1696     }, {
1697 wakaba 1.49 %HTMLAttrStatus,
1698 wakaba 1.82 %XHTML2CommonAttrStatus,
1699     class => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_HTML2X_RFC,
1700     dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1701     id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_XHTML10_REC,
1702 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
1703 wakaba 1.49 profile => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC,
1704     }),
1705 wakaba 1.40 check_child_element => sub {
1706     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1707     $child_is_transparent, $element_state) = @_;
1708 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1709     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1710 wakaba 1.40 $self->{onerror}->(node => $child_el,
1711     type => 'element not allowed:minus',
1712 wakaba 1.104 level => $self->{level}->{must});
1713 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1714     #
1715     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'title') {
1716     unless ($element_state->{has_title}) {
1717     $element_state->{has_title} = 1;
1718     } else {
1719     $self->{onerror}->(node => $child_el,
1720     type => 'element not allowed:head title',
1721 wakaba 1.104 level => $self->{level}->{must});
1722 wakaba 1.40 }
1723     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'style') {
1724     if ($child_el->has_attribute_ns (undef, 'scoped')) {
1725     $self->{onerror}->(node => $child_el,
1726     type => 'element not allowed:head style',
1727 wakaba 1.104 level => $self->{level}->{must});
1728 wakaba 1.1 }
1729 wakaba 1.40 } elsif ($HTMLMetadataContent->{$child_nsuri}->{$child_ln}) {
1730     #
1731    
1732     ## NOTE: |meta| is a metadata content. However, strictly speaking,
1733     ## a |meta| element with none of |charset|, |name|,
1734     ## or |http-equiv| attribute is not allowed. It is non-conforming
1735     ## anyway.
1736 wakaba 1.56
1737     ## TODO: |form| MUST be empty and in XML [WF2].
1738 wakaba 1.40 } else {
1739     $self->{onerror}->(node => $child_el,
1740     type => 'element not allowed:metadata',
1741 wakaba 1.104 level => $self->{level}->{must});
1742 wakaba 1.40 }
1743     $element_state->{in_head_original} = $self->{flag}->{in_head};
1744     $self->{flag}->{in_head} = 1;
1745     },
1746     check_child_text => sub {
1747     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1748     if ($has_significant) {
1749 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
1750     level => $self->{level}->{must});
1751 wakaba 1.1 }
1752 wakaba 1.40 },
1753     check_end => sub {
1754     my ($self, $item, $element_state) = @_;
1755     unless ($element_state->{has_title}) {
1756     $self->{onerror}->(node => $item->{node},
1757 wakaba 1.104 type => 'child element missing',
1758     text => 'title',
1759 wakaba 1.105 level => $self->{level}->{must});
1760 wakaba 1.1 }
1761 wakaba 1.40 $self->{flag}->{in_head} = $element_state->{in_head_original};
1762 wakaba 1.1
1763 wakaba 1.40 $HTMLChecker{check_end}->(@_);
1764 wakaba 1.1 },
1765     };
1766    
1767 wakaba 1.40 $Element->{$HTML_NS}->{title} = {
1768     %HTMLTextChecker,
1769 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1770 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
1771     %HTMLAttrStatus,
1772 wakaba 1.82 %XHTML2CommonAttrStatus,
1773     class => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_HTML2X_RFC,
1774     dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1775     id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_XHTML10_REC,
1776 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
1777 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
1778 wakaba 1.49 }),
1779 wakaba 1.40 };
1780 wakaba 1.1
1781 wakaba 1.40 $Element->{$HTML_NS}->{base} = {
1782 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1783 wakaba 1.40 %HTMLEmptyChecker,
1784     check_attrs => sub {
1785     my ($self, $item, $element_state) = @_;
1786 wakaba 1.1
1787 wakaba 1.40 if ($self->{has_base}) {
1788     $self->{onerror}->(node => $item->{node},
1789 wakaba 1.104 type => 'element not allowed:base',
1790     level => $self->{level}->{must});
1791 wakaba 1.40 } else {
1792     $self->{has_base} = 1;
1793 wakaba 1.29 }
1794    
1795 wakaba 1.40 my $has_href = $item->{node}->has_attribute_ns (undef, 'href');
1796     my $has_target = $item->{node}->has_attribute_ns (undef, 'target');
1797 wakaba 1.14
1798     if ($self->{has_uri_attr} and $has_href) {
1799 wakaba 1.4 ## ISSUE: Are these examples conforming?
1800     ## <head profile="a b c"><base href> (except for |profile|'s
1801     ## non-conformance)
1802     ## <title xml:base="relative"/><base href/> (maybe it should be)
1803     ## <unknown xmlns="relative"/><base href/> (assuming that
1804     ## |{relative}:unknown| is allowed before XHTML |base| (unlikely, though))
1805     ## <style>@import 'relative';</style><base href>
1806     ## <script>location.href = 'relative';</script><base href>
1807 wakaba 1.14 ## NOTE: <html manifest=".."><head><base href=""/> is conforming as
1808     ## an exception.
1809 wakaba 1.40 $self->{onerror}->(node => $item->{node},
1810 wakaba 1.104 type => 'basehref after URL attribute',
1811     level => $self->{level}->{must});
1812 wakaba 1.4 }
1813 wakaba 1.14 if ($self->{has_hyperlink_element} and $has_target) {
1814 wakaba 1.4 ## ISSUE: Are these examples conforming?
1815     ## <head><title xlink:href=""/><base target="name"/></head>
1816     ## <xbl:xbl>...<svg:a href=""/>...</xbl:xbl><base target="name"/>
1817     ## (assuming that |xbl:xbl| is allowed before |base|)
1818     ## NOTE: These are non-conformant anyway because of |head|'s content model:
1819     ## <link href=""/><base target="name"/>
1820     ## <link rel=unknown href=""><base target=name>
1821 wakaba 1.40 $self->{onerror}->(node => $item->{node},
1822 wakaba 1.104 type => 'basetarget after hyperlink',
1823     level => $self->{level}->{must});
1824 wakaba 1.4 }
1825    
1826 wakaba 1.14 if (not $has_href and not $has_target) {
1827 wakaba 1.40 $self->{onerror}->(node => $item->{node},
1828 wakaba 1.104 type => 'attribute missing:href|target',
1829     level => $self->{level}->{must});
1830 wakaba 1.14 }
1831    
1832 wakaba 1.66 $element_state->{uri_info}->{href}->{type}->{base} = 1;
1833    
1834 wakaba 1.4 return $GetHTMLAttrsChecker->({
1835     href => $HTMLURIAttrChecker,
1836     target => $HTMLTargetAttrChecker,
1837 wakaba 1.49 }, {
1838     %HTMLAttrStatus,
1839 wakaba 1.50 href => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1840     id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
1841     target => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1842 wakaba 1.40 })->($self, $item, $element_state);
1843 wakaba 1.4 },
1844 wakaba 1.1 };
1845    
1846     $Element->{$HTML_NS}->{link} = {
1847 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1848 wakaba 1.40 %HTMLEmptyChecker,
1849     check_attrs => sub {
1850     my ($self, $item, $element_state) = @_;
1851 wakaba 1.96 my $sizes_attr;
1852 wakaba 1.1 $GetHTMLAttrsChecker->({
1853 wakaba 1.91 charset => sub {
1854     my ($self, $attr) = @_;
1855     $HTMLCharsetChecker->($attr->value, @_);
1856     },
1857 wakaba 1.1 href => $HTMLURIAttrChecker,
1858 wakaba 1.40 rel => sub { $HTMLLinkTypesAttrChecker->(0, $item, @_) },
1859 wakaba 1.92 rev => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->(),
1860 wakaba 1.1 media => $HTMLMQAttrChecker,
1861     hreflang => $HTMLLanguageTagAttrChecker,
1862 wakaba 1.96 sizes => sub {
1863     my ($self, $attr) = @_;
1864     $sizes_attr = $attr;
1865     my %word;
1866     for my $word (grep {length $_}
1867 wakaba 1.132 split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
1868 wakaba 1.96 unless ($word{$word}) {
1869     $word{$word} = 1;
1870     if ($word eq 'any' or $word =~ /\A[1-9][0-9]*x[1-9][0-9]*\z/) {
1871     #
1872     } else {
1873     $self->{onerror}->(node => $attr,
1874 wakaba 1.104 type => 'sizes:syntax error',
1875 wakaba 1.96 value => $word,
1876 wakaba 1.104 level => $self->{level}->{must});
1877 wakaba 1.96 }
1878     } else {
1879     $self->{onerror}->(node => $attr, type => 'duplicate token',
1880     value => $word,
1881 wakaba 1.104 level => $self->{level}->{must});
1882 wakaba 1.96 }
1883     }
1884     },
1885 wakaba 1.70 target => $HTMLTargetAttrChecker,
1886 wakaba 1.1 type => $HTMLIMTAttrChecker,
1887     ## NOTE: Though |title| has special semantics,
1888     ## syntactically same as the |title| as global attribute.
1889 wakaba 1.49 }, {
1890     %HTMLAttrStatus,
1891 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
1892 wakaba 1.91 charset => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC,
1893     ## NOTE: |charset| attribute had been part of HTML5 spec though
1894     ## it had been commented out.
1895 wakaba 1.121 href => FEATURE_HTML5_DEFAULT | FEATURE_RDFA_CR | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1896 wakaba 1.82 hreflang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED |
1897     FEATURE_M12N10_REC,
1898 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
1899 wakaba 1.82 media => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1900 wakaba 1.61 methods => FEATURE_HTML20_RFC,
1901 wakaba 1.120 rel => FEATURE_HTML5_DEFAULT | FEATURE_RDFA_CR | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1902     rev => FEATURE_RDFA_CR | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1903 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
1904 wakaba 1.96 sizes => FEATURE_HTML5_DEFAULT,
1905 wakaba 1.82 target => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1906 wakaba 1.50 type => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1907 wakaba 1.61 urn => FEATURE_HTML20_RFC,
1908 wakaba 1.40 })->($self, $item, $element_state);
1909 wakaba 1.96
1910 wakaba 1.40 if ($item->{node}->has_attribute_ns (undef, 'href')) {
1911     $self->{has_hyperlink_element} = 1 if $item->{has_hyperlink_link_type};
1912 wakaba 1.4 } else {
1913 wakaba 1.40 $self->{onerror}->(node => $item->{node},
1914 wakaba 1.104 type => 'attribute missing',
1915     text => 'href',
1916     level => $self->{level}->{must});
1917 wakaba 1.1 }
1918 wakaba 1.96
1919 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'rel')) {
1920     $self->{onerror}->(node => $item->{node},
1921 wakaba 1.104 type => 'attribute missing',
1922     text => 'rel',
1923     level => $self->{level}->{must});
1924 wakaba 1.96 }
1925    
1926     if ($sizes_attr and not $element_state->{link_rel}->{icon}) {
1927     $self->{onerror}->(node => $sizes_attr,
1928     type => 'attribute not allowed',
1929 wakaba 1.104 level => $self->{level}->{must});
1930 wakaba 1.1 }
1931 wakaba 1.116
1932     if ($element_state->{link_rel}->{alternate} and
1933     $element_state->{link_rel}->{stylesheet}) {
1934     my $title_attr = $item->{node}->get_attribute_node_ns (undef, 'title');
1935     unless ($title_attr) {
1936     $self->{onerror}->(node => $item->{node},
1937     type => 'attribute missing',
1938     text => 'title',
1939     level => $self->{level}->{must});
1940     } elsif ($title_attr->value eq '') {
1941     $self->{onerror}->(node => $title_attr,
1942     type => 'empty style sheet title',
1943     level => $self->{level}->{must});
1944     }
1945     }
1946 wakaba 1.1 },
1947     };
1948    
1949     $Element->{$HTML_NS}->{meta} = {
1950 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1951 wakaba 1.40 %HTMLEmptyChecker,
1952     check_attrs => sub {
1953     my ($self, $item, $element_state) = @_;
1954 wakaba 1.1 my $name_attr;
1955     my $http_equiv_attr;
1956     my $charset_attr;
1957     my $content_attr;
1958 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
1959 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
1960     $attr_ns = '' unless defined $attr_ns;
1961     my $attr_ln = $attr->manakai_local_name;
1962     my $checker;
1963 wakaba 1.73 my $status;
1964 wakaba 1.1 if ($attr_ns eq '') {
1965 wakaba 1.73 $status = {
1966     %HTMLAttrStatus,
1967 wakaba 1.82 %XHTML2CommonAttrStatus,
1968 wakaba 1.73 charset => FEATURE_HTML5_DEFAULT,
1969 wakaba 1.82 content => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED |
1970     FEATURE_M12N10_REC,
1971     dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED |
1972     FEATURE_M12N10_REC,
1973 wakaba 1.73 'http-equiv' => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1974 wakaba 1.82 id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED |
1975     FEATURE_XHTML10_REC,
1976 wakaba 1.73 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
1977     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1978     scheme => FEATURE_M12N10_REC,
1979     }->{$attr_ln};
1980    
1981 wakaba 1.1 if ($attr_ln eq 'content') {
1982     $content_attr = $attr;
1983     $checker = 1;
1984     } elsif ($attr_ln eq 'name') {
1985     $name_attr = $attr;
1986     $checker = 1;
1987     } elsif ($attr_ln eq 'http-equiv') {
1988     $http_equiv_attr = $attr;
1989     $checker = 1;
1990     } elsif ($attr_ln eq 'charset') {
1991     $charset_attr = $attr;
1992     $checker = 1;
1993 wakaba 1.67 } elsif ($attr_ln eq 'scheme') {
1994 wakaba 1.71 ## NOTE: <http://suika.fam.cx/2007/html/standards#html-meta-scheme>
1995 wakaba 1.67 $checker = sub {};
1996 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
1997     $attr_ln !~ /[A-Z]/) {
1998 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
1999     $status = $HTMLDatasetAttrStatus;
2000 wakaba 1.1 } else {
2001     $checker = $HTMLAttrChecker->{$attr_ln}
2002 wakaba 1.67 || $AttrChecker->{$attr_ns}->{$attr_ln}
2003 wakaba 1.1 || $AttrChecker->{$attr_ns}->{''};
2004     }
2005     } else {
2006     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
2007 wakaba 1.82 || $AttrChecker->{$attr_ns}->{''};
2008     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
2009     || $AttrStatus->{$attr_ns}->{''};
2010     $status = FEATURE_ALLOWED if not defined $status;
2011 wakaba 1.1 }
2012 wakaba 1.62
2013 wakaba 1.1 if ($checker) {
2014 wakaba 1.66 $checker->($self, $attr, $item, $element_state) if ref $checker;
2015 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
2016 wakaba 1.54 #
2017 wakaba 1.1 } else {
2018 wakaba 1.104 $self->{onerror}->(node => $attr,
2019     type => 'unknown attribute',
2020     level => $self->{level}->{uncertain});
2021 wakaba 1.49 ## ISSUE: No conformance createria for unknown attributes in the spec
2022     }
2023    
2024 wakaba 1.82 $self->_attr_status_info ($attr, $status);
2025 wakaba 1.1 }
2026    
2027     if (defined $name_attr) {
2028     if (defined $http_equiv_attr) {
2029     $self->{onerror}->(node => $http_equiv_attr,
2030 wakaba 1.104 type => 'attribute not allowed',
2031     level => $self->{level}->{must});
2032 wakaba 1.1 } elsif (defined $charset_attr) {
2033     $self->{onerror}->(node => $charset_attr,
2034 wakaba 1.104 type => 'attribute not allowed',
2035     level => $self->{level}->{must});
2036 wakaba 1.1 }
2037     my $metadata_name = $name_attr->value;
2038     my $metadata_value;
2039     if (defined $content_attr) {
2040     $metadata_value = $content_attr->value;
2041     } else {
2042 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2043 wakaba 1.104 type => 'attribute missing',
2044     text => 'content',
2045     level => $self->{level}->{must});
2046 wakaba 1.1 $metadata_value = '';
2047     }
2048     } elsif (defined $http_equiv_attr) {
2049     if (defined $charset_attr) {
2050     $self->{onerror}->(node => $charset_attr,
2051 wakaba 1.104 type => 'attribute not allowed',
2052     level => $self->{level}->{must});
2053 wakaba 1.1 }
2054     unless (defined $content_attr) {
2055 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2056 wakaba 1.104 type => 'attribute missing',
2057     text => 'content',
2058     level => $self->{level}->{must});
2059 wakaba 1.1 }
2060     } elsif (defined $charset_attr) {
2061     if (defined $content_attr) {
2062     $self->{onerror}->(node => $content_attr,
2063 wakaba 1.104 type => 'attribute not allowed',
2064     level => $self->{level}->{must});
2065 wakaba 1.1 }
2066     } else {
2067     if (defined $content_attr) {
2068     $self->{onerror}->(node => $content_attr,
2069 wakaba 1.104 type => 'attribute not allowed',
2070     level => $self->{level}->{must});
2071 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2072 wakaba 1.104 type => 'attribute missing:name|http-equiv',
2073     level => $self->{level}->{must});
2074 wakaba 1.1 } else {
2075 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2076 wakaba 1.104 type => 'attribute missing:name|http-equiv|charset',
2077     level => $self->{level}->{must});
2078 wakaba 1.1 }
2079     }
2080    
2081 wakaba 1.32 my $check_charset_decl = sub () {
2082 wakaba 1.40 my $parent = $item->{node}->manakai_parent_element;
2083 wakaba 1.29 if ($parent and $parent eq $parent->owner_document->manakai_head) {
2084     for my $el (@{$parent->child_nodes}) {
2085     next unless $el->node_type == 1; # ELEMENT_NODE
2086 wakaba 1.40 unless ($el eq $item->{node}) {
2087 wakaba 1.29 ## NOTE: Not the first child element.
2088 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2089 wakaba 1.32 type => 'element not allowed:meta charset',
2090 wakaba 1.104 level => $self->{level}->{must});
2091 wakaba 1.29 }
2092     last;
2093     ## NOTE: Entity references are not supported.
2094     }
2095     } else {
2096 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2097 wakaba 1.32 type => 'element not allowed:meta charset',
2098 wakaba 1.104 level => $self->{level}->{must});
2099 wakaba 1.29 }
2100    
2101 wakaba 1.40 unless ($item->{node}->owner_document->manakai_is_html) {
2102     $self->{onerror}->(node => $item->{node},
2103 wakaba 1.32 type => 'in XML:charset',
2104 wakaba 1.104 level => $self->{level}->{must});
2105 wakaba 1.1 }
2106 wakaba 1.32 }; # $check_charset_decl
2107 wakaba 1.21
2108 wakaba 1.32 my $check_charset = sub ($$) {
2109     my ($attr, $charset_value) = @_;
2110 wakaba 1.21
2111 wakaba 1.91 my $charset;
2112     ($charset, $charset_value)
2113     = $HTMLCharsetChecker->($charset_value, $self, $attr);
2114    
2115 wakaba 1.40 my $ic = $item->{node}->owner_document->input_encoding;
2116 wakaba 1.21 if (defined $ic) {
2117     ## TODO: Test for this case
2118     my $ic_charset = $Message::Charset::Info::IANACharset->{$ic};
2119     if ($charset ne $ic_charset) {
2120 wakaba 1.32 $self->{onerror}->(node => $attr,
2121 wakaba 1.104 type => 'mismatched charset name',
2122 wakaba 1.106 text => $ic,
2123 wakaba 1.104 value => $charset_value,
2124     level => $self->{level}->{must});
2125 wakaba 1.21 }
2126     } else {
2127     ## NOTE: MUST, but not checkable, since the document is not originally
2128     ## in serialized form (or the parser does not preserve the input
2129     ## encoding information).
2130 wakaba 1.32 $self->{onerror}->(node => $attr,
2131 wakaba 1.104 type => 'mismatched charset name not checked',
2132     value => $charset_value,
2133     level => $self->{level}->{uncertain});
2134 wakaba 1.21 }
2135    
2136 wakaba 1.32 if ($attr->get_user_data ('manakai_has_reference')) {
2137     $self->{onerror}->(node => $attr,
2138 wakaba 1.104 type => 'charref in charset',
2139     level => $self->{level}->{must},
2140     layer => 'syntax');
2141 wakaba 1.22 }
2142 wakaba 1.32 }; # $check_charset
2143    
2144     ## TODO: metadata conformance
2145    
2146     ## TODO: pragma conformance
2147     if (defined $http_equiv_attr) { ## An enumerated attribute
2148     my $keyword = lc $http_equiv_attr->value; ## TODO: ascii case?
2149 wakaba 1.33
2150 wakaba 1.85 if ($self->{has_http_equiv}->{$keyword}) {
2151     $self->{onerror}->(type => 'duplicate http-equiv', value => $keyword,
2152     node => $http_equiv_attr,
2153 wakaba 1.104 level => $self->{level}->{must});
2154 wakaba 1.85 } else {
2155     $self->{has_http_equiv}->{$keyword} = 1;
2156     }
2157    
2158     if ($keyword eq 'content-type') {
2159 wakaba 1.58 ## TODO: refs in "text/html; charset=" are not disallowed since rev.1275.
2160 wakaba 1.33
2161 wakaba 1.32 $check_charset_decl->();
2162     if ($content_attr) {
2163     my $content = $content_attr->value;
2164 wakaba 1.58 if ($content =~ m!^[Tt][Ee][Xx][Tt]/[Hh][Tt][Mm][Ll];
2165 wakaba 1.132 [\x09\x0A\x0C\x0D\x20]*[Cc][Hh][Aa][Rr][Ss][Ee][Tt]
2166 wakaba 1.58 =(.+)\z!sx) {
2167 wakaba 1.32 $check_charset->($content_attr, $1);
2168     } else {
2169     $self->{onerror}->(node => $content_attr,
2170     type => 'meta content-type syntax error',
2171 wakaba 1.104 level => $self->{level}->{must});
2172 wakaba 1.85 }
2173     }
2174     } elsif ($keyword eq 'default-style') {
2175     ## ISSUE: Not defined yet in the spec.
2176     } elsif ($keyword eq 'refresh') {
2177     if ($content_attr) {
2178     my $content = $content_attr->value;
2179     if ($content =~ /\A[0-9]+\z/) {
2180     ## NOTE: Valid non-negative integer.
2181     #
2182 wakaba 1.132 } elsif ($content =~ s/\A[0-9]+;[\x09\x0A\x0C\x0D\x20]+[Uu][Rr][Ll]=//) {
2183 wakaba 1.85 ## ISSUE: Relative references are allowed? (RFC 3987 "IRI" is an absolute reference with optional fragment identifier.)
2184     Whatpm::URIChecker->check_iri_reference ($content, sub {
2185 wakaba 1.104 $self->{onerror}->(value => $content, @_, node => $content_attr);
2186 wakaba 1.106 }, $self->{level});
2187 wakaba 1.85 $self->{has_uri_attr} = 1; ## NOTE: One of "attributes with URIs".
2188    
2189     $element_state->{uri_info}->{content}->{node} = $content_attr;
2190     $element_state->{uri_info}->{content}->{type}->{hyperlink} = 1;
2191     ## TODO: absolute
2192     push @{$self->{return}->{uri}->{$content} ||= []},
2193     $element_state->{uri_info}->{content};
2194     } else {
2195     $self->{onerror}->(node => $content_attr,
2196     type => 'refresh:syntax error',
2197 wakaba 1.104 level => $self->{level}->{must});
2198 wakaba 1.32 }
2199     }
2200     } else {
2201     $self->{onerror}->(node => $http_equiv_attr,
2202 wakaba 1.104 type => 'enumerated:invalid',
2203     level => $self->{level}->{must});
2204 wakaba 1.32 }
2205     }
2206    
2207     if (defined $charset_attr) {
2208     $check_charset_decl->();
2209     $check_charset->($charset_attr, $charset_attr->value);
2210 wakaba 1.1 }
2211     },
2212     };
2213    
2214     $Element->{$HTML_NS}->{style} = {
2215 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2216 wakaba 1.40 %HTMLChecker,
2217     check_attrs => $GetHTMLAttrsChecker->({
2218 wakaba 1.1 type => $HTMLIMTAttrChecker, ## TODO: MUST be a styling language
2219     media => $HTMLMQAttrChecker,
2220     scoped => $GetHTMLBooleanAttrChecker->('scoped'),
2221     ## NOTE: |title| has special semantics for |style|s, but is syntactically
2222     ## not different
2223 wakaba 1.49 }, {
2224     %HTMLAttrStatus,
2225 wakaba 1.82 %XHTML2CommonAttrStatus,
2226     dir => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2227     disabled => FEATURE_XHTML2_ED,
2228 wakaba 1.121 href => FEATURE_RDFA_PR | FEATURE_XHTML2_ED,
2229 wakaba 1.82 id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_XHTML10_REC,
2230 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2231 wakaba 1.82 media => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2232 wakaba 1.50 scoped => FEATURE_HTML5_DEFAULT,
2233 wakaba 1.82 title => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2234     type => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2235 wakaba 1.1 }),
2236 wakaba 1.40 check_start => sub {
2237     my ($self, $item, $element_state) = @_;
2238    
2239 wakaba 1.27 ## NOTE: |html:style| itself has no conformance creteria on content model.
2240 wakaba 1.40 my $type = $item->{node}->get_attribute_ns (undef, 'type');
2241 wakaba 1.93 $type = 'text/css' unless defined $type;
2242     if ($type =~ m[\A(?>(?>\x0D\x0A)?[\x09\x20])*([\x21\x23-\x27\x2A\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+)(?>(?>\x0D\x0A)?[\x09\x20])*/(?>(?>\x0D\x0A)?[\x09\x20])*([\x21\x23-\x27\x2A\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+)(?>(?>\x0D\x0A)?[\x09\x20])*\z]) {
2243     $type = "$1/$2";
2244     $type =~ tr/A-Z/a-z/; ## NOTE: ASCII case-insensitive
2245     } else {
2246     ## NOTE: We don't know how parameters are handled by UAs. According to
2247     ## HTML5 specification, <style> with unknown parameters in |type=""|
2248     ## must be ignored.
2249     undef $type;
2250     }
2251     if (not defined $type) {
2252     $element_state->{allow_element} = 1; # invalid type=""
2253     } elsif ($type eq 'text/css') {
2254 wakaba 1.40 $element_state->{allow_element} = 0;
2255 wakaba 1.93 #} elsif ($type =~ m![/+][Xx][Mm][Ll]\z!) {
2256     # ## NOTE: There is no definition for "XML-based styling language" in HTML5
2257     # $element_state->{allow_element} = 1;
2258 wakaba 1.40 } else {
2259     $element_state->{allow_element} = 1; # unknown
2260     }
2261 wakaba 1.93 $element_state->{style_type} = $type;
2262 wakaba 1.79
2263     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2264     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2265 wakaba 1.107
2266     $element_state->{text} = '';
2267 wakaba 1.40 },
2268     check_child_element => sub {
2269     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2270     $child_is_transparent, $element_state) = @_;
2271 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2272     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2273 wakaba 1.40 $self->{onerror}->(node => $child_el,
2274     type => 'element not allowed:minus',
2275 wakaba 1.104 level => $self->{level}->{must});
2276 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2277     #
2278     } elsif ($element_state->{allow_element}) {
2279     #
2280     } else {
2281 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2282     level => $self->{level}->{must});
2283 wakaba 1.40 }
2284     },
2285     check_child_text => sub {
2286     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2287 wakaba 1.115 $element_state->{text} .= $child_node->data;
2288 wakaba 1.40 },
2289     check_end => sub {
2290     my ($self, $item, $element_state) = @_;
2291 wakaba 1.93 if (not defined $element_state->{style_type}) {
2292     ## NOTE: Invalid type=""
2293     #
2294     } elsif ($element_state->{style_type} eq 'text/css') {
2295 wakaba 1.40 $self->{onsubdoc}->({s => $element_state->{text},
2296     container_node => $item->{node},
2297 wakaba 1.28 media_type => 'text/css', is_char_string => 1});
2298 wakaba 1.93 } elsif ($element_state->{style_type} =~ m![+/][Xx][Mm][Ll]\z!) {
2299     ## NOTE: XML content should be checked by THIS instance of checker
2300     ## as part of normal tree validation. However, we don't know of any
2301     ## XML-based styling language that can be used in HTML <style> element,
2302     ## such that we throw a "style language not supported" error.
2303 wakaba 1.104 $self->{onerror}->(node => $item->{node},
2304     type => 'XML style lang',
2305     text => $element_state->{style_type},
2306     level => $self->{level}->{uncertain});
2307 wakaba 1.93 } else {
2308     ## NOTE: Should we raise some kind of error for,
2309     ## say, <style type="text/plaion">?
2310     $self->{onsubdoc}->({s => $element_state->{text},
2311     container_node => $item->{node},
2312     media_type => $element_state->{style_type},
2313     is_char_string => 1});
2314 wakaba 1.27 }
2315 wakaba 1.40
2316     $HTMLChecker{check_end}->(@_);
2317 wakaba 1.1 },
2318     };
2319 wakaba 1.25 ## ISSUE: Relationship to significant content check?
2320 wakaba 1.1
2321     $Element->{$HTML_NS}->{body} = {
2322 wakaba 1.72 %HTMLFlowContentChecker,
2323 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2324 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2325     alink => $HTMLColorAttrChecker,
2326     background => $HTMLURIAttrChecker,
2327     bgcolor => $HTMLColorAttrChecker,
2328     link => $HTMLColorAttrChecker,
2329     text => $HTMLColorAttrChecker,
2330     vlink => $HTMLColorAttrChecker,
2331     }, {
2332 wakaba 1.49 %HTMLAttrStatus,
2333 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2334 wakaba 1.49 alink => FEATURE_M12N10_REC_DEPRECATED,
2335     background => FEATURE_M12N10_REC_DEPRECATED,
2336     bgcolor => FEATURE_M12N10_REC_DEPRECATED,
2337 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2338 wakaba 1.49 link => FEATURE_M12N10_REC_DEPRECATED,
2339 wakaba 1.50 onload => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2340     onunload => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2341 wakaba 1.49 text => FEATURE_M12N10_REC_DEPRECATED,
2342     vlink => FEATURE_M12N10_REC_DEPRECATED,
2343     }),
2344 wakaba 1.68 check_start => sub {
2345     my ($self, $item, $element_state) = @_;
2346    
2347     $element_state->{uri_info}->{background}->{type}->{embedded} = 1;
2348 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2349     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2350 wakaba 1.68 },
2351 wakaba 1.1 };
2352    
2353     $Element->{$HTML_NS}->{section} = {
2354 wakaba 1.72 %HTMLFlowContentChecker,
2355 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED,
2356     check_attrs => $GetHTMLAttrsChecker->({
2357     }, {
2358     %HTMLAttrStatus,
2359     %XHTML2CommonAttrStatus,
2360     }),
2361 wakaba 1.1 };
2362    
2363     $Element->{$HTML_NS}->{nav} = {
2364 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
2365 wakaba 1.72 %HTMLFlowContentChecker,
2366 wakaba 1.1 };
2367    
2368     $Element->{$HTML_NS}->{article} = {
2369 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
2370 wakaba 1.72 %HTMLFlowContentChecker,
2371 wakaba 1.1 };
2372    
2373     $Element->{$HTML_NS}->{blockquote} = {
2374 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2375 wakaba 1.72 %HTMLFlowContentChecker,
2376 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
2377 wakaba 1.1 cite => $HTMLURIAttrChecker,
2378 wakaba 1.49 }, {
2379     %HTMLAttrStatus,
2380 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2381 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2382 wakaba 1.82 cite => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2383 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2384 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2385 wakaba 1.1 }),
2386 wakaba 1.66 check_start => sub {
2387     my ($self, $item, $element_state) = @_;
2388    
2389     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
2390 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2391     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2392 wakaba 1.66 },
2393 wakaba 1.1 };
2394    
2395     $Element->{$HTML_NS}->{aside} = {
2396 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
2397 wakaba 1.72 %HTMLFlowContentChecker,
2398 wakaba 1.1 };
2399    
2400     $Element->{$HTML_NS}->{h1} = {
2401 wakaba 1.40 %HTMLPhrasingContentChecker,
2402 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2403 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2404     align => $GetHTMLEnumeratedAttrChecker->({
2405     left => 1, center => 1, right => 1, justify => 1,
2406     }),
2407     }, {
2408 wakaba 1.49 %HTMLAttrStatus,
2409 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2410 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
2411 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2412 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2413 wakaba 1.49 }),
2414 wakaba 1.40 check_start => sub {
2415     my ($self, $item, $element_state) = @_;
2416     $self->{flag}->{has_hn} = 1;
2417 wakaba 1.79
2418     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2419     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2420 wakaba 1.1 },
2421     };
2422    
2423 wakaba 1.40 $Element->{$HTML_NS}->{h2} = {%{$Element->{$HTML_NS}->{h1}}};
2424 wakaba 1.1
2425 wakaba 1.40 $Element->{$HTML_NS}->{h3} = {%{$Element->{$HTML_NS}->{h1}}};
2426 wakaba 1.1
2427 wakaba 1.40 $Element->{$HTML_NS}->{h4} = {%{$Element->{$HTML_NS}->{h1}}};
2428 wakaba 1.1
2429 wakaba 1.40 $Element->{$HTML_NS}->{h5} = {%{$Element->{$HTML_NS}->{h1}}};
2430 wakaba 1.1
2431 wakaba 1.40 $Element->{$HTML_NS}->{h6} = {%{$Element->{$HTML_NS}->{h1}}};
2432 wakaba 1.1
2433 wakaba 1.29 ## TODO: Explicit sectioning is "encouraged".
2434    
2435 wakaba 1.1 $Element->{$HTML_NS}->{header} = {
2436 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
2437 wakaba 1.72 %HTMLFlowContentChecker,
2438 wakaba 1.40 check_start => sub {
2439     my ($self, $item, $element_state) = @_;
2440     $self->_add_minus_elements ($element_state,
2441     {$HTML_NS => {qw/header 1 footer 1/}},
2442 wakaba 1.58 $HTMLSectioningContent);
2443 wakaba 1.40 $element_state->{has_hn_original} = $self->{flag}->{has_hn};
2444     $self->{flag}->{has_hn} = 0;
2445 wakaba 1.79
2446     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2447     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2448 wakaba 1.40 },
2449     check_end => sub {
2450     my ($self, $item, $element_state) = @_;
2451     $self->_remove_minus_elements ($element_state);
2452     unless ($self->{flag}->{has_hn}) {
2453     $self->{onerror}->(node => $item->{node},
2454 wakaba 1.104 type => 'element missing:hn',
2455     level => $self->{level}->{must});
2456 wakaba 1.40 }
2457     $self->{flag}->{has_hn} ||= $element_state->{has_hn_original};
2458 wakaba 1.1
2459 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
2460 wakaba 1.1 },
2461 wakaba 1.40 ## ISSUE: <header><del><h1>...</h1></del></header> is conforming?
2462 wakaba 1.1 };
2463    
2464     $Element->{$HTML_NS}->{footer} = {
2465 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
2466 wakaba 1.72 %HTMLFlowContentChecker,
2467 wakaba 1.40 check_start => sub {
2468     my ($self, $item, $element_state) = @_;
2469     $self->_add_minus_elements ($element_state,
2470     {$HTML_NS => {footer => 1}},
2471 wakaba 1.58 $HTMLSectioningContent,
2472 wakaba 1.57 $HTMLHeadingContent);
2473 wakaba 1.79
2474     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2475     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2476 wakaba 1.40 },
2477     check_end => sub {
2478     my ($self, $item, $element_state) = @_;
2479     $self->_remove_minus_elements ($element_state);
2480 wakaba 1.1
2481 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
2482 wakaba 1.1 },
2483     };
2484    
2485     $Element->{$HTML_NS}->{address} = {
2486 wakaba 1.72 %HTMLFlowContentChecker,
2487 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2488 wakaba 1.110 check_attrs => $GetHTMLAttrsChecker->({
2489     ## TODO: add test
2490     #align => $GetHTMLEnumeratedAttrChecker->({
2491     # left => 1, center => 1, right => 1, justify => 1,
2492     #}),
2493     }, {
2494 wakaba 1.49 %HTMLAttrStatus,
2495 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2496 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2497 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2498 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2499     sdapref => FEATURE_HTML20_RFC,
2500 wakaba 1.49 }),
2501 wakaba 1.40 check_start => sub {
2502     my ($self, $item, $element_state) = @_;
2503     $self->_add_minus_elements ($element_state,
2504     {$HTML_NS => {footer => 1, address => 1}},
2505     $HTMLSectioningContent, $HTMLHeadingContent);
2506 wakaba 1.79
2507     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2508     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2509 wakaba 1.40 },
2510     check_end => sub {
2511     my ($self, $item, $element_state) = @_;
2512     $self->_remove_minus_elements ($element_state);
2513 wakaba 1.29
2514 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
2515 wakaba 1.29 },
2516 wakaba 1.1 };
2517    
2518     $Element->{$HTML_NS}->{p} = {
2519 wakaba 1.40 %HTMLPhrasingContentChecker,
2520 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2521 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2522     align => $GetHTMLEnumeratedAttrChecker->({
2523     left => 1, center => 1, right => 1, justify => 1,
2524     }),
2525     }, {
2526 wakaba 1.49 %HTMLAttrStatus,
2527 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2528 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
2529 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2530 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2531 wakaba 1.49 }),
2532 wakaba 1.1 };
2533    
2534     $Element->{$HTML_NS}->{hr} = {
2535 wakaba 1.40 %HTMLEmptyChecker,
2536 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2537 wakaba 1.70 check_attrs => $GetHTMLAttrsChecker->({
2538     ## TODO: HTML4 |align|, |noshade|, |size|, |width|
2539     }, {
2540 wakaba 1.49 %HTMLAttrStatus,
2541     %HTMLM12NCommonAttrStatus,
2542     align => FEATURE_M12N10_REC_DEPRECATED,
2543 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2544 wakaba 1.49 noshade => FEATURE_M12N10_REC_DEPRECATED,
2545 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
2546 wakaba 1.49 size => FEATURE_M12N10_REC_DEPRECATED,
2547     width => FEATURE_M12N10_REC_DEPRECATED,
2548     }),
2549 wakaba 1.1 };
2550    
2551     $Element->{$HTML_NS}->{br} = {
2552 wakaba 1.40 %HTMLEmptyChecker,
2553 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2554 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2555     clear => $GetHTMLEnumeratedAttrChecker->({
2556     left => 1, all => 1, right => 1, none => 1,
2557     }),
2558     }, {
2559 wakaba 1.49 %HTMLAttrStatus,
2560 wakaba 1.50 class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2561 wakaba 1.49 clear => FEATURE_M12N10_REC_DEPRECATED,
2562 wakaba 1.50 id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2563 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
2564 wakaba 1.78 style => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2565 wakaba 1.50 title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2566 wakaba 1.49 }),
2567 wakaba 1.29 ## NOTE: Blank line MUST NOT be used for presentation purpose.
2568     ## (This requirement is semantic so that we cannot check.)
2569 wakaba 1.1 };
2570    
2571     $Element->{$HTML_NS}->{dialog} = {
2572 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
2573 wakaba 1.40 %HTMLChecker,
2574     check_start => sub {
2575     my ($self, $item, $element_state) = @_;
2576     $element_state->{phase} = 'before dt';
2577 wakaba 1.79
2578     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2579     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2580 wakaba 1.40 },
2581     check_child_element => sub {
2582     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2583     $child_is_transparent, $element_state) = @_;
2584 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2585     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2586 wakaba 1.40 $self->{onerror}->(node => $child_el,
2587     type => 'element not allowed:minus',
2588 wakaba 1.104 level => $self->{level}->{must});
2589 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2590     #
2591     } elsif ($element_state->{phase} eq 'before dt') {
2592     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2593     $element_state->{phase} = 'before dd';
2594     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2595     $self->{onerror}
2596 wakaba 1.104 ->(node => $child_el, type => 'ps element missing',
2597     text => 'dt',
2598     level => $self->{level}->{must});
2599 wakaba 1.40 $element_state->{phase} = 'before dt';
2600     } else {
2601 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2602     level => $self->{level}->{must});
2603 wakaba 1.40 }
2604     } elsif ($element_state->{phase} eq 'before dd') {
2605     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2606     $element_state->{phase} = 'before dt';
2607     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2608     $self->{onerror}
2609 wakaba 1.104 ->(node => $child_el, type => 'ps element missing',
2610     text => 'dd',
2611     level => $self->{level}->{must});
2612 wakaba 1.40 $element_state->{phase} = 'before dd';
2613     } else {
2614 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2615     level => $self->{level}->{must});
2616 wakaba 1.1 }
2617 wakaba 1.40 } else {
2618     die "check_child_element: Bad |dialog| phase: $element_state->{phase}";
2619     }
2620     },
2621     check_child_text => sub {
2622     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2623     if ($has_significant) {
2624 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
2625     level => $self->{level}->{must});
2626 wakaba 1.1 }
2627 wakaba 1.40 },
2628     check_end => sub {
2629     my ($self, $item, $element_state) = @_;
2630     if ($element_state->{phase} eq 'before dd') {
2631     $self->{onerror}->(node => $item->{node},
2632 wakaba 1.104 type => 'child element missing',
2633     text => 'dd',
2634     level => $self->{level}->{must});
2635 wakaba 1.1 }
2636 wakaba 1.40
2637     $HTMLChecker{check_end}->(@_);
2638 wakaba 1.1 },
2639     };
2640    
2641     $Element->{$HTML_NS}->{pre} = {
2642 wakaba 1.40 %HTMLPhrasingContentChecker,
2643 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2644 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2645     width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
2646     }, {
2647 wakaba 1.49 %HTMLAttrStatus,
2648 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2649 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2650 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2651 wakaba 1.49 width => FEATURE_M12N10_REC_DEPRECATED,
2652     }),
2653 wakaba 1.101 check_end => sub {
2654     my ($self, $item, $element_state) = @_;
2655    
2656     ## TODO: Flag to enable/disable IDL checking?
2657     my $class = $item->{node}->get_attribute ('class');
2658 wakaba 1.102 if ($class =~ /\bidl(?>-code)?\b/) { ## TODO: use classList.has
2659     ## NOTE: pre.idl: WHATWG, XHR, Selectors API, CSSOM specs
2660     ## NOTE: pre.code > code.idl-code: WebIDL spec
2661     ## NOTE: pre.idl-code: DOM1 spec
2662     ## NOTE: div.idl-code > pre: DOM, ProgressEvent specs
2663     ## NOTE: pre.schema: ReSpec-generated specs
2664 wakaba 1.101 $self->{onsubdoc}->({s => $item->{node}->text_content,
2665     container_node => $item->{node},
2666     media_type => 'text/x-webidl',
2667     is_char_string => 1});
2668     }
2669    
2670 wakaba 1.110 $HTMLPhrasingContentChecker{check_end}->(@_);
2671 wakaba 1.101 },
2672 wakaba 1.1 };
2673    
2674     $Element->{$HTML_NS}->{ol} = {
2675 wakaba 1.40 %HTMLChecker,
2676 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2677 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
2678 wakaba 1.68 compact => $GetHTMLBooleanAttrChecker->('compact'),
2679 wakaba 1.69 reversed => $GetHTMLBooleanAttrChecker->('reversed'),
2680 wakaba 1.1 start => $HTMLIntegerAttrChecker,
2681 wakaba 1.69 ## TODO: HTML4 |type|
2682 wakaba 1.49 }, {
2683     %HTMLAttrStatus,
2684 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2685 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2686 wakaba 1.49 compact => FEATURE_M12N10_REC_DEPRECATED,
2687 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2688 wakaba 1.53 reversed => FEATURE_HTML5_DEFAULT,
2689 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2690 wakaba 1.54 #start => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC_DEPRECATED,
2691     start => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2692 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
2693 wakaba 1.1 }),
2694 wakaba 1.40 check_child_element => sub {
2695     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2696     $child_is_transparent, $element_state) = @_;
2697 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2698     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2699 wakaba 1.40 $self->{onerror}->(node => $child_el,
2700     type => 'element not allowed:minus',
2701 wakaba 1.104 level => $self->{level}->{must});
2702 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2703     #
2704     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'li') {
2705     #
2706     } else {
2707 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2708     level => $self->{level}->{must});
2709 wakaba 1.1 }
2710 wakaba 1.40 },
2711     check_child_text => sub {
2712     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2713     if ($has_significant) {
2714 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
2715     level => $self->{level}->{must});
2716 wakaba 1.1 }
2717     },
2718     };
2719    
2720     $Element->{$HTML_NS}->{ul} = {
2721 wakaba 1.40 %{$Element->{$HTML_NS}->{ol}},
2722 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2723 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2724     compact => $GetHTMLBooleanAttrChecker->('compact'),
2725 wakaba 1.69 ## TODO: HTML4 |type|
2726     ## TODO: sdaform, align
2727 wakaba 1.68 }, {
2728 wakaba 1.49 %HTMLAttrStatus,
2729 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2730 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2731 wakaba 1.49 compact => FEATURE_M12N10_REC_DEPRECATED,
2732 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2733 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2734 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
2735     }),
2736 wakaba 1.1 };
2737    
2738 wakaba 1.64 $Element->{$HTML_NS}->{dir} = {
2739     ## TODO: %block; is not allowed [HTML4] ## TODO: Empty list allowed?
2740     %{$Element->{$HTML_NS}->{ul}},
2741     status => FEATURE_M12N10_REC_DEPRECATED,
2742 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2743     compact => $GetHTMLBooleanAttrChecker->('compact'),
2744     }, {
2745 wakaba 1.64 %HTMLAttrStatus,
2746     %HTMLM12NCommonAttrStatus,
2747     align => FEATURE_HTML2X_RFC,
2748     compact => FEATURE_M12N10_REC_DEPRECATED,
2749     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2750     sdaform => FEATURE_HTML20_RFC,
2751     sdapref => FEATURE_HTML20_RFC,
2752     }),
2753     };
2754    
2755 wakaba 1.1 $Element->{$HTML_NS}->{li} = {
2756 wakaba 1.72 %HTMLFlowContentChecker,
2757 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2758 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
2759 wakaba 1.69 ## TODO: HTML4 |type|
2760 wakaba 1.49 value => sub {
2761 wakaba 1.1 my ($self, $attr) = @_;
2762     my $parent = $attr->owner_element->manakai_parent_element;
2763     if (defined $parent) {
2764     my $parent_ns = $parent->namespace_uri;
2765     $parent_ns = '' unless defined $parent_ns;
2766     my $parent_ln = $parent->manakai_local_name;
2767     unless ($parent_ns eq $HTML_NS and $parent_ln eq 'ol') {
2768 wakaba 1.131 ## ISSUE: ...
2769 wakaba 1.104 $self->{onerror}->(node => $attr,
2770     type => 'unknown attribute',
2771     level => $self->{level}->{uncertain});
2772 wakaba 1.1 }
2773     }
2774     $HTMLIntegerAttrChecker->($self, $attr);
2775 wakaba 1.131 },
2776 wakaba 1.49 }, {
2777     %HTMLAttrStatus,
2778 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2779 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2780 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2781 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2782 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
2783 wakaba 1.55 #value => FEATURE_HTML5_DEFAULT | FEATURE_XHTMLBASIC11_CR |
2784     # FEATURE_M12N10_REC_DEPRECATED,
2785 wakaba 1.82 value => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED |
2786     FEATURE_XHTMLBASIC11_CR | FEATURE_M12N10_REC,
2787 wakaba 1.1 }),
2788 wakaba 1.40 check_child_element => sub {
2789     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2790     $child_is_transparent, $element_state) = @_;
2791     if ($self->{flag}->{in_menu}) {
2792     $HTMLPhrasingContentChecker{check_child_element}->(@_);
2793     } else {
2794 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
2795 wakaba 1.40 }
2796     },
2797     check_child_text => sub {
2798     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2799     if ($self->{flag}->{in_menu}) {
2800     $HTMLPhrasingContentChecker{check_child_text}->(@_);
2801 wakaba 1.1 } else {
2802 wakaba 1.72 $HTMLFlowContentChecker{check_child_text}->(@_);
2803 wakaba 1.1 }
2804     },
2805     };
2806    
2807     $Element->{$HTML_NS}->{dl} = {
2808 wakaba 1.40 %HTMLChecker,
2809 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2810 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2811     compact => $GetHTMLBooleanAttrChecker->('compact'),
2812     }, {
2813 wakaba 1.49 %HTMLAttrStatus,
2814 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2815 wakaba 1.49 compact => FEATURE_M12N10_REC_DEPRECATED,
2816 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2817 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2818     sdapref => FEATURE_HTML20_RFC,
2819 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
2820     }),
2821 wakaba 1.40 check_start => sub {
2822     my ($self, $item, $element_state) = @_;
2823     $element_state->{phase} = 'before dt';
2824 wakaba 1.79
2825     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2826     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2827 wakaba 1.40 },
2828     check_child_element => sub {
2829     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2830     $child_is_transparent, $element_state) = @_;
2831 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2832     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2833 wakaba 1.40 $self->{onerror}->(node => $child_el,
2834     type => 'element not allowed:minus',
2835 wakaba 1.104 level => $self->{level}->{must});
2836 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2837     #
2838     } elsif ($element_state->{phase} eq 'in dds') {
2839     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2840     #$element_state->{phase} = 'in dds';
2841     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2842     $element_state->{phase} = 'in dts';
2843     } else {
2844 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2845     level => $self->{level}->{must});
2846 wakaba 1.40 }
2847     } elsif ($element_state->{phase} eq 'in dts') {
2848     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2849     #$element_state->{phase} = 'in dts';
2850     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2851     $element_state->{phase} = 'in dds';
2852     } else {
2853 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2854     level => $self->{level}->{must});
2855 wakaba 1.40 }
2856     } elsif ($element_state->{phase} eq 'before dt') {
2857     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2858     $element_state->{phase} = 'in dts';
2859     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2860     $self->{onerror}
2861 wakaba 1.104 ->(node => $child_el, type => 'ps element missing',
2862     text => 'dt',
2863     level => $self->{level}->{must});
2864 wakaba 1.40 $element_state->{phase} = 'in dds';
2865     } else {
2866 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2867     level => $self->{level}->{must});
2868 wakaba 1.1 }
2869 wakaba 1.40 } else {
2870     die "check_child_element: Bad |dl| phase: $element_state->{phase}";
2871 wakaba 1.1 }
2872 wakaba 1.40 },
2873     check_child_text => sub {
2874     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2875     if ($has_significant) {
2876 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
2877     level => $self->{level}->{must});
2878 wakaba 1.40 }
2879     },
2880     check_end => sub {
2881     my ($self, $item, $element_state) = @_;
2882     if ($element_state->{phase} eq 'in dts') {
2883     $self->{onerror}->(node => $item->{node},
2884 wakaba 1.104 type => 'child element missing',
2885     text => 'dd',
2886     level => $self->{level}->{must});
2887 wakaba 1.1 }
2888    
2889 wakaba 1.40 $HTMLChecker{check_end}->(@_);
2890 wakaba 1.1 },
2891     };
2892    
2893     $Element->{$HTML_NS}->{dt} = {
2894 wakaba 1.40 %HTMLPhrasingContentChecker,
2895 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2896 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
2897     %HTMLAttrStatus,
2898 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2899 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2900 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2901 wakaba 1.49 }),
2902 wakaba 1.1 };
2903    
2904     $Element->{$HTML_NS}->{dd} = {
2905 wakaba 1.72 %HTMLFlowContentChecker,
2906 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2907 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
2908     %HTMLAttrStatus,
2909 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2910 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2911 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2912 wakaba 1.49 }),
2913 wakaba 1.1 };
2914    
2915     $Element->{$HTML_NS}->{a} = {
2916 wakaba 1.123 %HTMLTransparentChecker,
2917 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2918 wakaba 1.40 check_attrs => sub {
2919     my ($self, $item, $element_state) = @_;
2920 wakaba 1.1 my %attr;
2921 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
2922 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
2923     $attr_ns = '' unless defined $attr_ns;
2924     my $attr_ln = $attr->manakai_local_name;
2925     my $checker;
2926 wakaba 1.73 my $status;
2927 wakaba 1.1 if ($attr_ns eq '') {
2928 wakaba 1.73 $status = {
2929     %HTMLAttrStatus,
2930 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2931 wakaba 1.73 accesskey => FEATURE_M12N10_REC,
2932     charset => FEATURE_M12N10_REC,
2933 wakaba 1.82 coords => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2934 wakaba 1.73 cryptopts => FEATURE_RFC2659,
2935     dn => FEATURE_RFC2659,
2936 wakaba 1.121 href => FEATURE_HTML5_DEFAULT | FEATURE_RDFA_CR | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2937 wakaba 1.82 hreflang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2938 wakaba 1.73 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
2939 wakaba 1.82 media => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED,
2940 wakaba 1.73 methods => FEATURE_HTML20_RFC,
2941     name => FEATURE_M12N10_REC_DEPRECATED,
2942     nonce => FEATURE_RFC2659,
2943     onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2944     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2945     ping => FEATURE_HTML5_DEFAULT,
2946 wakaba 1.120 rel => FEATURE_HTML5_DEFAULT | FEATURE_RDFA_CR | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2947     rev => FEATURE_RDFA_CR | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2948 wakaba 1.73 sdapref => FEATURE_HTML20_RFC,
2949 wakaba 1.82 shape => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2950 wakaba 1.73 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2951 wakaba 1.82 target => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2952 wakaba 1.73 type => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
2953     urn => FEATURE_HTML20_RFC,
2954     }->{$attr_ln};
2955    
2956 wakaba 1.1 $checker = {
2957 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
2958 wakaba 1.91 charset => sub {
2959     my ($self, $attr) = @_;
2960     $HTMLCharsetChecker->($attr->value, @_);
2961     },
2962 wakaba 1.70 ## TODO: HTML4 |coords|
2963 wakaba 1.1 target => $HTMLTargetAttrChecker,
2964     href => $HTMLURIAttrChecker,
2965     ping => $HTMLSpaceURIsAttrChecker,
2966 wakaba 1.40 rel => sub { $HTMLLinkTypesAttrChecker->(1, $item, @_) },
2967 wakaba 1.92 rev => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->(),
2968 wakaba 1.70 ## TODO: HTML4 |shape|
2969 wakaba 1.1 media => $HTMLMQAttrChecker,
2970 wakaba 1.70 ## TODO: HTML4/XHTML1 |name|
2971 wakaba 1.1 hreflang => $HTMLLanguageTagAttrChecker,
2972     type => $HTMLIMTAttrChecker,
2973     }->{$attr_ln};
2974     if ($checker) {
2975     $attr{$attr_ln} = $attr;
2976 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
2977     $attr_ln !~ /[A-Z]/) {
2978 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
2979     $status = $HTMLDatasetAttrStatus;
2980 wakaba 1.1 } else {
2981     $checker = $HTMLAttrChecker->{$attr_ln};
2982     }
2983     }
2984     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
2985     || $AttrChecker->{$attr_ns}->{''};
2986 wakaba 1.82 $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
2987     || $AttrStatus->{$attr_ns}->{''};
2988     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
2989 wakaba 1.62
2990 wakaba 1.1 if ($checker) {
2991 wakaba 1.66 $checker->($self, $attr, $item, $element_state) if ref $checker;
2992 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
2993 wakaba 1.54 #
2994 wakaba 1.1 } else {
2995 wakaba 1.104 $self->{onerror}->(node => $attr,
2996     type => 'unknown attribute',
2997     level => $self->{level}->{uncertain});
2998 wakaba 1.50 ## ISSUE: No conformance createria for unknown attributes in the spec
2999 wakaba 1.1 }
3000 wakaba 1.49
3001 wakaba 1.82 $self->_attr_status_info ($attr, $status);
3002 wakaba 1.1 }
3003    
3004 wakaba 1.40 $element_state->{in_a_href_original} = $self->{flag}->{in_a_href};
3005 wakaba 1.4 if (defined $attr{href}) {
3006     $self->{has_hyperlink_element} = 1;
3007 wakaba 1.40 $self->{flag}->{in_a_href} = 1;
3008 wakaba 1.4 } else {
3009 wakaba 1.1 for (qw/target ping rel media hreflang type/) {
3010     if (defined $attr{$_}) {
3011     $self->{onerror}->(node => $attr{$_},
3012 wakaba 1.104 type => 'attribute not allowed',
3013     level => $self->{level}->{must});
3014 wakaba 1.1 }
3015     }
3016     }
3017 wakaba 1.66
3018     $element_state->{uri_info}->{href}->{type}->{hyperlink} = 1;
3019 wakaba 1.1 },
3020 wakaba 1.40 check_start => sub {
3021     my ($self, $item, $element_state) = @_;
3022     $self->_add_minus_elements ($element_state, $HTMLInteractiveContent);
3023 wakaba 1.79
3024     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3025     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3026 wakaba 1.40 },
3027     check_end => sub {
3028     my ($self, $item, $element_state) = @_;
3029     $self->_remove_minus_elements ($element_state);
3030 wakaba 1.59 delete $self->{flag}->{in_a_href}
3031     unless $element_state->{in_a_href_original};
3032 wakaba 1.1
3033 wakaba 1.123 $HTMLTransparentChecker{check_end}->(@_);
3034 wakaba 1.1 },
3035     };
3036    
3037     $Element->{$HTML_NS}->{q} = {
3038 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3039 wakaba 1.40 %HTMLPhrasingContentChecker,
3040     check_attrs => $GetHTMLAttrsChecker->({
3041 wakaba 1.50 cite => $HTMLURIAttrChecker,
3042     }, {
3043 wakaba 1.49 %HTMLAttrStatus,
3044 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3045     cite => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3046 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3047 wakaba 1.61 sdapref => FEATURE_HTML2X_RFC,
3048     sdasuff => FEATURE_HTML2X_RFC,
3049 wakaba 1.1 }),
3050 wakaba 1.66 check_start => sub {
3051     my ($self, $item, $element_state) = @_;
3052    
3053     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
3054 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3055     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3056 wakaba 1.66 },
3057 wakaba 1.1 };
3058 wakaba 1.75 ## TODO: "Quotation punctuation (such as quotation marks), if any, must be
3059     ## placed inside the <code>q</code> element." Though we cannot test the
3060     ## element against this requirement since it incluides a semantic bit,
3061     ## it might be possible to inform of the existence of quotation marks OUTSIDE
3062     ## the |q| element.
3063 wakaba 1.1
3064     $Element->{$HTML_NS}->{cite} = {
3065 wakaba 1.40 %HTMLPhrasingContentChecker,
3066 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3067 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3068     %HTMLAttrStatus,
3069 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3070 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3071 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3072 wakaba 1.49 }),
3073 wakaba 1.1 };
3074    
3075     $Element->{$HTML_NS}->{em} = {
3076 wakaba 1.40 %HTMLPhrasingContentChecker,
3077 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3078 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3079     %HTMLAttrStatus,
3080 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3081 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3082 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3083 wakaba 1.49 }),
3084 wakaba 1.1 };
3085    
3086     $Element->{$HTML_NS}->{strong} = {
3087 wakaba 1.40 %HTMLPhrasingContentChecker,
3088 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3089 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3090     %HTMLAttrStatus,
3091 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3092 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3093 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3094 wakaba 1.49 }),
3095 wakaba 1.1 };
3096    
3097     $Element->{$HTML_NS}->{small} = {
3098 wakaba 1.40 %HTMLPhrasingContentChecker,
3099 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3100     check_attrs => $GetHTMLAttrsChecker->({}, {
3101     %HTMLAttrStatus,
3102     %HTMLM12NCommonAttrStatus,
3103 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3104 wakaba 1.49 }),
3105 wakaba 1.1 };
3106    
3107 wakaba 1.51 $Element->{$HTML_NS}->{big} = {
3108     %HTMLPhrasingContentChecker,
3109     status => FEATURE_M12N10_REC,
3110     check_attrs => $GetHTMLAttrsChecker->({}, {
3111     %HTMLAttrStatus,
3112     %HTMLM12NCommonAttrStatus,
3113     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3114     }),
3115     };
3116    
3117 wakaba 1.38 $Element->{$HTML_NS}->{mark} = {
3118 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
3119 wakaba 1.40 %HTMLPhrasingContentChecker,
3120 wakaba 1.1 };
3121    
3122     $Element->{$HTML_NS}->{dfn} = {
3123 wakaba 1.40 %HTMLPhrasingContentChecker,
3124 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3125 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3126     %HTMLAttrStatus,
3127 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3128 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3129 wakaba 1.49 }),
3130 wakaba 1.40 check_start => sub {
3131     my ($self, $item, $element_state) = @_;
3132     $self->_add_minus_elements ($element_state, {$HTML_NS => {dfn => 1}});
3133 wakaba 1.1
3134 wakaba 1.40 my $node = $item->{node};
3135 wakaba 1.1 my $term = $node->get_attribute_ns (undef, 'title');
3136     unless (defined $term) {
3137     for my $child (@{$node->child_nodes}) {
3138     if ($child->node_type == 1) { # ELEMENT_NODE
3139     if (defined $term) {
3140     undef $term;
3141     last;
3142     } elsif ($child->manakai_local_name eq 'abbr') {
3143     my $nsuri = $child->namespace_uri;
3144     if (defined $nsuri and $nsuri eq $HTML_NS) {
3145     my $attr = $child->get_attribute_node_ns (undef, 'title');
3146     if ($attr) {
3147     $term = $attr->value;
3148     }
3149     }
3150     }
3151     } elsif ($child->node_type == 3 or $child->node_type == 4) {
3152     ## TEXT_NODE or CDATA_SECTION_NODE
3153 wakaba 1.132 if ($child->data =~ /\A[\x09\x0A\x0C\x0D\x20]+\z/) { # Inter-element whitespace
3154 wakaba 1.1 next;
3155     }
3156     undef $term;
3157     last;
3158     }
3159     }
3160     unless (defined $term) {
3161     $term = $node->text_content;
3162     }
3163     }
3164     if ($self->{term}->{$term}) {
3165     push @{$self->{term}->{$term}}, $node;
3166     } else {
3167     $self->{term}->{$term} = [$node];
3168     }
3169 wakaba 1.77 ## ISSUE: The HTML5 definition for the defined term does not work with
3170     ## |ruby| unless |dfn| has |title|.
3171 wakaba 1.79
3172     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3173     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3174 wakaba 1.40 },
3175     check_end => sub {
3176     my ($self, $item, $element_state) = @_;
3177     $self->_remove_minus_elements ($element_state);
3178 wakaba 1.1
3179 wakaba 1.40 $HTMLPhrasingContentChecker{check_end}->(@_);
3180 wakaba 1.1 },
3181     };
3182    
3183     $Element->{$HTML_NS}->{abbr} = {
3184 wakaba 1.40 %HTMLPhrasingContentChecker,
3185 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3186 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3187     %HTMLAttrStatus,
3188 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3189     full => FEATURE_XHTML2_ED,
3190 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3191 wakaba 1.49 }),
3192 wakaba 1.77 ## NOTE: "If an abbreviation is pluralised, the expansion's grammatical
3193     ## number (plural vs singular) must match the grammatical number of the
3194     ## contents of the element." Though this can be checked by machine,
3195     ## it requires language-specific knowledge and dictionary, such that
3196     ## we don't support the check of the requirement.
3197     ## ISSUE: Is <abbr title="Cascading Style Sheets">CSS</abbr> conforming?
3198 wakaba 1.49 };
3199    
3200     $Element->{$HTML_NS}->{acronym} = {
3201     %HTMLPhrasingContentChecker,
3202     status => FEATURE_M12N10_REC,
3203     check_attrs => $GetHTMLAttrsChecker->({}, {
3204     %HTMLAttrStatus,
3205     %HTMLM12NCommonAttrStatus,
3206 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3207 wakaba 1.49 }),
3208 wakaba 1.1 };
3209    
3210     $Element->{$HTML_NS}->{time} = {
3211 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
3212 wakaba 1.40 %HTMLPhrasingContentChecker,
3213     check_attrs => $GetHTMLAttrsChecker->({
3214 wakaba 1.1 datetime => sub { 1 }, # checked in |checker|
3215 wakaba 1.49 }, {
3216     %HTMLAttrStatus,
3217     %HTMLM12NCommonAttrStatus,
3218 wakaba 1.72 datetime => FEATURE_HTML5_FD,
3219 wakaba 1.1 }),
3220     ## TODO: Write tests
3221 wakaba 1.40 check_end => sub {
3222     my ($self, $item, $element_state) = @_;
3223 wakaba 1.1
3224 wakaba 1.40 my $attr = $item->{node}->get_attribute_node_ns (undef, 'datetime');
3225 wakaba 1.1 my $input;
3226     my $reg_sp;
3227     my $input_node;
3228     if ($attr) {
3229     $input = $attr->value;
3230 wakaba 1.132 $reg_sp = qr/[\x09\x0A\x0C\x0D\x20]*/;
3231 wakaba 1.1 $input_node = $attr;
3232     } else {
3233 wakaba 1.40 $input = $item->{node}->text_content;
3234 wakaba 1.112 $reg_sp = qr/\p{WhiteSpace}*/;
3235 wakaba 1.40 $input_node = $item->{node};
3236 wakaba 1.1
3237     ## ISSUE: What is the definition for "successfully extracts a date
3238     ## or time"? If the algorithm says the string is invalid but
3239     ## return some date or time, is it "successfully"?
3240     }
3241    
3242     my $hour;
3243     my $minute;
3244     my $second;
3245     if ($input =~ /
3246     \A
3247 wakaba 1.112 $reg_sp
3248 wakaba 1.1 ([0-9]+) # 1
3249     (?>
3250     -([0-9]+) # 2
3251 wakaba 1.112 -((?>[0-9]+)) # 3 # Use (?>) such that yyyy-mm-ddhh:mm does not match
3252     $reg_sp
3253 wakaba 1.1 (?>
3254     T
3255 wakaba 1.112 $reg_sp
3256 wakaba 1.1 )?
3257     ([0-9]+) # 4
3258     :([0-9]+) # 5
3259     (?>
3260     :([0-9]+(?>\.[0-9]*)?|\.[0-9]*) # 6
3261     )?
3262 wakaba 1.112 $reg_sp
3263 wakaba 1.1 (?>
3264     Z
3265 wakaba 1.112 $reg_sp
3266 wakaba 1.1 |
3267     [+-]([0-9]+):([0-9]+) # 7, 8
3268 wakaba 1.112 $reg_sp
3269 wakaba 1.1 )?
3270     \z
3271     |
3272     :([0-9]+) # 9
3273     (?>
3274     :([0-9]+(?>\.[0-9]*)?|\.[0-9]*) # 10
3275     )?
3276 wakaba 1.112 $reg_sp
3277     \z
3278 wakaba 1.1 )
3279     /x) {
3280     if (defined $2) { ## YYYY-MM-DD T? hh:mm
3281     if (length $1 != 4 or length $2 != 2 or length $3 != 2 or
3282     length $4 != 2 or length $5 != 2) {
3283     $self->{onerror}->(node => $input_node,
3284 wakaba 1.104 type => 'dateortime:syntax error',
3285     level => $self->{level}->{must});
3286 wakaba 1.1 }
3287    
3288     if (1 <= $2 and $2 <= 12) {
3289 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad day',
3290     level => $self->{level}->{must})
3291 wakaba 1.1 if $3 < 1 or
3292     $3 > [0, 31, 29, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]->[$2];
3293 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad day',
3294     level => $self->{level}->{must})
3295 wakaba 1.1 if $2 == 2 and $3 == 29 and
3296     not ($1 % 400 == 0 or ($1 % 4 == 0 and $1 % 100 != 0));
3297     } else {
3298     $self->{onerror}->(node => $input_node,
3299 wakaba 1.104 type => 'datetime:bad month',
3300     level => $self->{level}->{must});
3301 wakaba 1.1 }
3302    
3303     ($hour, $minute, $second) = ($4, $5, $6);
3304    
3305     if (defined $7) { ## [+-]hh:mm
3306     if (length $7 != 2 or length $8 != 2) {
3307     $self->{onerror}->(node => $input_node,
3308 wakaba 1.104 type => 'dateortime:syntax error',
3309     level => $self->{level}->{must});
3310 wakaba 1.1 }
3311    
3312     $self->{onerror}->(node => $input_node,
3313 wakaba 1.104 type => 'datetime:bad timezone hour',
3314     level => $self->{level}->{must})
3315 wakaba 1.1 if $7 > 23;
3316     $self->{onerror}->(node => $input_node,
3317 wakaba 1.104 type => 'datetime:bad timezone minute',
3318     level => $self->{level}->{must})
3319 wakaba 1.1 if $8 > 59;
3320     }
3321     } else { ## hh:mm
3322     if (length $1 != 2 or length $9 != 2) {
3323     $self->{onerror}->(node => $input_node,
3324 wakaba 1.104 type => qq'dateortime:syntax error',
3325     level => $self->{level}->{must});
3326 wakaba 1.1 }
3327    
3328     ($hour, $minute, $second) = ($1, $9, $10);
3329     }
3330    
3331 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad hour',
3332     level => $self->{level}->{must}) if $hour > 23;
3333     $self->{onerror}->(node => $input_node, type => 'datetime:bad minute',
3334     level => $self->{level}->{must}) if $minute > 59;
3335 wakaba 1.1
3336     if (defined $second) { ## s
3337     ## NOTE: Integer part of second don't have to have length of two.
3338    
3339     if (substr ($second, 0, 1) eq '.') {
3340     $self->{onerror}->(node => $input_node,
3341 wakaba 1.104 type => 'dateortime:syntax error',
3342     level => $self->{level}->{must});
3343 wakaba 1.1 }
3344    
3345 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad second',
3346     level => $self->{level}->{must}) if $second >= 60;
3347 wakaba 1.1 }
3348     } else {
3349     $self->{onerror}->(node => $input_node,
3350 wakaba 1.104 type => 'dateortime:syntax error',
3351     level => $self->{level}->{must});
3352 wakaba 1.1 }
3353    
3354 wakaba 1.40 $HTMLPhrasingContentChecker{check_end}->(@_);
3355 wakaba 1.1 },
3356     };
3357    
3358     $Element->{$HTML_NS}->{meter} = { ## TODO: "The recommended way of giving the value is to include it as contents of the element"
3359 wakaba 1.77 ## TODO: value inequalities (HTML5 revision 1463)
3360 wakaba 1.113 ## TODO: content checking
3361     ## TODO: content or value must contain number (rev 2053)
3362 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
3363 wakaba 1.40 %HTMLPhrasingContentChecker,
3364     check_attrs => $GetHTMLAttrsChecker->({
3365 wakaba 1.1 value => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3366     min => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3367     low => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3368     high => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3369     max => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3370     optimum => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3371 wakaba 1.50 }, {
3372     %HTMLAttrStatus,
3373     high => FEATURE_HTML5_DEFAULT,
3374     low => FEATURE_HTML5_DEFAULT,
3375     max => FEATURE_HTML5_DEFAULT,
3376     min => FEATURE_HTML5_DEFAULT,
3377     optimum => FEATURE_HTML5_DEFAULT,
3378     value => FEATURE_HTML5_DEFAULT,
3379 wakaba 1.1 }),
3380     };
3381    
3382     $Element->{$HTML_NS}->{progress} = { ## TODO: recommended to use content
3383 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
3384 wakaba 1.40 %HTMLPhrasingContentChecker,
3385     check_attrs => $GetHTMLAttrsChecker->({
3386 wakaba 1.1 value => $GetHTMLFloatingPointNumberAttrChecker->(sub { shift >= 0 }),
3387     max => $GetHTMLFloatingPointNumberAttrChecker->(sub { shift > 0 }),
3388 wakaba 1.50 }, {
3389     %HTMLAttrStatus,
3390     max => FEATURE_HTML5_DEFAULT,
3391     value => FEATURE_HTML5_DEFAULT,
3392 wakaba 1.1 }),
3393     };
3394    
3395     $Element->{$HTML_NS}->{code} = {
3396 wakaba 1.40 %HTMLPhrasingContentChecker,
3397 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3398 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3399     %HTMLAttrStatus,
3400 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3401 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3402 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3403 wakaba 1.49 }),
3404 wakaba 1.1 };
3405    
3406     $Element->{$HTML_NS}->{var} = {
3407 wakaba 1.40 %HTMLPhrasingContentChecker,
3408 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3409 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3410     %HTMLAttrStatus,
3411 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3412 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3413 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3414 wakaba 1.49 }),
3415 wakaba 1.1 };
3416    
3417     $Element->{$HTML_NS}->{samp} = {
3418 wakaba 1.40 %HTMLPhrasingContentChecker,
3419 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3420 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3421     %HTMLAttrStatus,
3422 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3423 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3424 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3425 wakaba 1.49 }),
3426 wakaba 1.1 };
3427    
3428     $Element->{$HTML_NS}->{kbd} = {
3429 wakaba 1.40 %HTMLPhrasingContentChecker,
3430 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3431 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3432     %HTMLAttrStatus,
3433 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3434 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3435 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3436 wakaba 1.49 }),
3437 wakaba 1.1 };
3438    
3439     $Element->{$HTML_NS}->{sub} = {
3440 wakaba 1.40 %HTMLPhrasingContentChecker,
3441 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3442 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3443     %HTMLAttrStatus,
3444 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3445 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3446 wakaba 1.61 sdapref => FEATURE_HTML2X_RFC,
3447 wakaba 1.49 }),
3448 wakaba 1.1 };
3449    
3450 wakaba 1.51 $Element->{$HTML_NS}->{sup} = $Element->{$HTML_NS}->{sub};
3451 wakaba 1.1
3452     $Element->{$HTML_NS}->{span} = {
3453 wakaba 1.40 %HTMLPhrasingContentChecker,
3454 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3455 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3456     %HTMLAttrStatus,
3457 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3458 wakaba 1.49 datafld => FEATURE_HTML4_REC_RESERVED,
3459     dataformatas => FEATURE_HTML4_REC_RESERVED,
3460     datasrc => FEATURE_HTML4_REC_RESERVED,
3461 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3462 wakaba 1.61 sdaform => FEATURE_HTML2X_RFC,
3463 wakaba 1.49 }),
3464 wakaba 1.1 };
3465    
3466     $Element->{$HTML_NS}->{i} = {
3467 wakaba 1.40 %HTMLPhrasingContentChecker,
3468 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3469     check_attrs => $GetHTMLAttrsChecker->({}, {
3470     %HTMLAttrStatus,
3471     %HTMLM12NCommonAttrStatus,
3472 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3473 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3474 wakaba 1.49 }),
3475 wakaba 1.1 };
3476    
3477 wakaba 1.51 $Element->{$HTML_NS}->{b} = $Element->{$HTML_NS}->{i};
3478    
3479 wakaba 1.61 $Element->{$HTML_NS}->{tt} = {
3480     %HTMLPhrasingContentChecker,
3481     status => FEATURE_M12N10_REC,
3482     check_attrs => $GetHTMLAttrsChecker->({}, {
3483     %HTMLAttrStatus,
3484     %HTMLM12NCommonAttrStatus,
3485     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3486     sdaform => FEATURE_HTML20_RFC,
3487     }),
3488     };
3489 wakaba 1.51
3490     $Element->{$HTML_NS}->{s} = {
3491 wakaba 1.40 %HTMLPhrasingContentChecker,
3492 wakaba 1.51 status => FEATURE_M12N10_REC_DEPRECATED,
3493 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3494     %HTMLAttrStatus,
3495     %HTMLM12NCommonAttrStatus,
3496 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3497 wakaba 1.49 }),
3498 wakaba 1.1 };
3499    
3500 wakaba 1.51 $Element->{$HTML_NS}->{strike} = $Element->{$HTML_NS}->{s};
3501    
3502     $Element->{$HTML_NS}->{u} = $Element->{$HTML_NS}->{s};
3503    
3504 wakaba 1.1 $Element->{$HTML_NS}->{bdo} = {
3505 wakaba 1.40 %HTMLPhrasingContentChecker,
3506 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3507 wakaba 1.40 check_attrs => sub {
3508     my ($self, $item, $element_state) = @_;
3509 wakaba 1.49 $GetHTMLAttrsChecker->({}, {
3510     %HTMLAttrStatus,
3511 wakaba 1.50 class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3512     dir => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3513     id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3514 wakaba 1.78 style => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3515 wakaba 1.50 title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3516     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3517 wakaba 1.61 sdapref => FEATURE_HTML2X_RFC,
3518     sdasuff => FEATURE_HTML2X_RFC,
3519 wakaba 1.49 })->($self, $item, $element_state);
3520 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'dir')) {
3521     $self->{onerror}->(node => $item->{node},
3522 wakaba 1.104 type => 'attribute missing',
3523     text => 'dir',
3524     level => $self->{level}->{must});
3525 wakaba 1.1 }
3526     },
3527     ## ISSUE: The spec does not directly say that |dir| is a enumerated attr.
3528     };
3529    
3530 wakaba 1.99 $Element->{$HTML_NS}->{ruby} = {
3531     %HTMLPhrasingContentChecker,
3532     status => FEATURE_HTML5_DEFAULT | FEATURE_RUBY_REC,
3533     check_attrs => $GetHTMLAttrsChecker->({}, {
3534     %HTMLAttrStatus,
3535     %HTMLM12NXHTML2CommonAttrStatus, # XHTML 1.1 & XHTML 2.0 & XHTML+RDFa 1.0
3536     lang => FEATURE_HTML5_DEFAULT,
3537     }),
3538     check_start => sub {
3539     my ($self, $item, $element_state) = @_;
3540    
3541     $element_state->{phase} = 'before-rb';
3542     #$element_state->{has_sig}
3543 wakaba 1.100
3544     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3545     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3546 wakaba 1.99 },
3547     ## NOTE: (phrasing, (rt | (rp, rt, rp)))+
3548     check_child_element => sub {
3549     my ($self, $item, $child_el, $child_nsuri, $child_ln,
3550     $child_is_transparent, $element_state) = @_;
3551 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
3552     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
3553 wakaba 1.99 $self->{onerror}->(node => $child_el,
3554     type => 'element not allowed:minus',
3555 wakaba 1.104 level => $self->{level}->{must});
3556 wakaba 1.99 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
3557     #
3558     } elsif ($element_state->{phase} eq 'before-rb') {
3559     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
3560     $element_state->{phase} = 'in-rb';
3561     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
3562     $self->{onerror}->(node => $child_el,
3563 wakaba 1.104 level => $self->{level}->{should},
3564     type => 'no significant content before');
3565 wakaba 1.99 $element_state->{phase} = 'after-rt';
3566     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
3567     $self->{onerror}->(node => $child_el,
3568 wakaba 1.104 level => $self->{level}->{should},
3569     type => 'no significant content before');
3570 wakaba 1.99 $element_state->{phase} = 'after-rp1';
3571     } else {
3572     $self->{onerror}->(node => $child_el,
3573 wakaba 1.104 type => 'element not allowed:ruby base',
3574     level => $self->{level}->{must});
3575 wakaba 1.99 $element_state->{phase} = 'in-rb';
3576     }
3577     } elsif ($element_state->{phase} eq 'in-rb') {
3578     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
3579     #$element_state->{phase} = 'in-rb';
3580     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
3581     unless ($element_state->{has_significant}) {
3582     $self->{onerror}->(node => $child_el,
3583 wakaba 1.104 level => $self->{level}->{should},
3584     type => 'no significant content before');
3585 wakaba 1.99 }
3586     $element_state->{phase} = 'after-rt';
3587     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
3588     unless ($element_state->{has_significant}) {
3589     $self->{onerror}->(node => $child_el,
3590 wakaba 1.104 level => $self->{level}->{should},
3591     type => 'no significant content before');
3592 wakaba 1.99 }
3593     $element_state->{phase} = 'after-rp1';
3594     } else {
3595     $self->{onerror}->(node => $child_el,
3596 wakaba 1.104 type => 'element not allowed:ruby base',
3597     level => $self->{level}->{must});
3598 wakaba 1.99 #$element_state->{phase} = 'in-rb';
3599     }
3600     } elsif ($element_state->{phase} eq 'after-rt') {
3601     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
3602     if ($element_state->{has_significant}) {
3603     $element_state->{has_sig} = 1;
3604     delete $element_state->{has_significant};
3605     }
3606     $element_state->{phase} = 'in-rb';
3607     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
3608     $self->{onerror}->(node => $child_el,
3609 wakaba 1.104 level => $self->{level}->{should},
3610     type => 'no significant content before');
3611 wakaba 1.99 $element_state->{phase} = 'after-rp1';
3612     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
3613     $self->{onerror}->(node => $child_el,
3614 wakaba 1.104 level => $self->{level}->{should},
3615     type => 'no significant content before');
3616 wakaba 1.99 #$element_state->{phase} = 'after-rt';
3617     } else {
3618     $self->{onerror}->(node => $child_el,
3619 wakaba 1.104 type => 'element not allowed:ruby base',
3620     level => $self->{level}->{must});
3621 wakaba 1.99 if ($element_state->{has_significant}) {
3622     $element_state->{has_sig} = 1;
3623     delete $element_state->{has_significant};
3624     }
3625     $element_state->{phase} = 'in-rb';
3626     }
3627     } elsif ($element_state->{phase} eq 'after-rp1') {
3628     if ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
3629     $element_state->{phase} = 'after-rp-rt';
3630     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
3631     $self->{onerror}->(node => $child_el,
3632 wakaba 1.104 type => 'ps element missing',
3633     text => 'rt',
3634     level => $self->{level}->{must});
3635 wakaba 1.99 $element_state->{phase} = 'after-rp2';
3636     } else {
3637     $self->{onerror}->(node => $child_el,
3638 wakaba 1.104 type => 'ps element missing',
3639     text => 'rt',
3640     level => $self->{level}->{must});
3641 wakaba 1.99 $self->{onerror}->(node => $child_el,
3642 wakaba 1.104 type => 'ps element missing',
3643     text => 'rp',
3644     level => $self->{level}->{must});
3645 wakaba 1.99 unless ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
3646     $self->{onerror}->(node => $child_el,
3647 wakaba 1.104 type => 'element not allowed:ruby base',
3648     level => $self->{level}->{must});
3649 wakaba 1.99 }
3650     if ($element_state->{has_significant}) {
3651     $element_state->{has_sig} = 1;
3652     delete $element_state->{has_significant};
3653     }
3654     $element_state->{phase} = 'in-rb';
3655     }
3656     } elsif ($element_state->{phase} eq 'after-rp-rt') {
3657     if ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
3658     $element_state->{phase} = 'after-rp2';
3659     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
3660     $self->{onerror}->(node => $child_el,
3661 wakaba 1.104 type => 'ps element missing',
3662     text => 'rp',
3663     level => $self->{level}->{must});
3664 wakaba 1.99 $self->{onerror}->(node => $child_el,
3665 wakaba 1.104 level => $self->{level}->{should},
3666     type => 'no significant content before');
3667 wakaba 1.99 $element_state->{phase} = 'after-rt';
3668     } else {
3669     $self->{onerror}->(node => $child_el,
3670 wakaba 1.104 type => 'ps element missing',
3671     text => 'rp',
3672     level => $self->{level}->{must});
3673 wakaba 1.99 unless ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
3674     $self->{onerror}->(node => $child_el,
3675 wakaba 1.104 type => 'element not allowed:ruby base',
3676     level => $self->{level}->{must});
3677 wakaba 1.99 }
3678     if ($element_state->{has_significant}) {
3679     $element_state->{has_sig} = 1;
3680     delete $element_state->{has_significant};
3681     }
3682     $element_state->{phase} = 'in-rb';
3683     }
3684     } elsif ($element_state->{phase} eq 'after-rp2') {
3685     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
3686     if ($element_state->{has_significant}) {
3687     $element_state->{has_sig} = 1;
3688     delete $element_state->{has_significant};
3689     }
3690     $element_state->{phase} = 'in-rb';
3691     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
3692     $self->{onerror}->(node => $child_el,
3693 wakaba 1.104 level => $self->{level}->{should},
3694     type => 'no significant content before');
3695 wakaba 1.99 $element_state->{phase} = 'after-rt';
3696     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
3697     $self->{onerror}->(node => $child_el,
3698 wakaba 1.104 level => $self->{level}->{should},
3699     type => 'no significant content before');
3700 wakaba 1.99 $element_state->{phase} = 'after-rp1';
3701     } else {
3702     $self->{onerror}->(node => $child_el,
3703 wakaba 1.104 type => 'element not allowed:ruby base',
3704     level => $self->{level}->{must});
3705 wakaba 1.99 if ($element_state->{has_significant}) {
3706     $element_state->{has_sig} = 1;
3707     delete $element_state->{has_significant};
3708     }
3709     $element_state->{phase} = 'in-rb';
3710     }
3711     } else {
3712     die "check_child_element: Bad |ruby| phase: $element_state->{phase}";
3713     }
3714     },
3715     check_child_text => sub {
3716     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
3717     if ($has_significant) {
3718     if ($element_state->{phase} eq 'before-rb') {
3719     $element_state->{phase} = 'in-rb';
3720     } elsif ($element_state->{phase} eq 'in-rb') {
3721     #
3722     } elsif ($element_state->{phase} eq 'after-rt' or
3723     $element_state->{phase} eq 'after-rp2') {
3724     $element_state->{phase} = 'in-rb';
3725     } elsif ($element_state->{phase} eq 'after-rp1') {
3726     $self->{onerror}->(node => $child_node,
3727 wakaba 1.104 type => 'ps element missing',
3728     text => 'rt',
3729     level => $self->{level}->{must});
3730 wakaba 1.99 $self->{onerror}->(node => $child_node,
3731 wakaba 1.104 type => 'ps element missing',
3732     text => 'rp',
3733     level => $self->{level}->{must});
3734 wakaba 1.99 $element_state->{phase} = 'in-rb';
3735     } elsif ($element_state->{phase} eq 'after-rp-rt') {
3736     $self->{onerror}->(node => $child_node,
3737 wakaba 1.104 type => 'ps element missing',
3738     text => 'rp',
3739     level => $self->{level}->{must});
3740 wakaba 1.99 $element_state->{phase} = 'in-rb';
3741     } else {
3742     die "check_child_text: Bad |ruby| phase: $element_state->{phase}";
3743     }
3744     }
3745     },
3746     check_end => sub {
3747     my ($self, $item, $element_state) = @_;
3748     $self->_remove_minus_elements ($element_state);
3749    
3750     if ($element_state->{phase} eq 'before-rb') {
3751     $self->{onerror}->(node => $item->{node},
3752 wakaba 1.104 level => $self->{level}->{should},
3753 wakaba 1.99 type => 'no significant content');
3754     $self->{onerror}->(node => $item->{node},
3755 wakaba 1.104 type => 'element missing',
3756     text => 'rt',
3757     level => $self->{level}->{must});
3758 wakaba 1.99 } elsif ($element_state->{phase} eq 'in-rb') {
3759     unless ($element_state->{has_significant}) {
3760     $self->{onerror}->(node => $item->{node},
3761 wakaba 1.104 level => $self->{level}->{should},
3762     type => 'no significant content at the end');
3763 wakaba 1.99 }
3764     $self->{onerror}->(node => $item->{node},
3765 wakaba 1.104 type => 'element missing',
3766     text => 'rt',
3767     level => $self->{level}->{must});
3768 wakaba 1.99 } elsif ($element_state->{phase} eq 'after-rt' or
3769     $element_state->{phase} eq 'after-rp2') {
3770     #
3771     } elsif ($element_state->{phase} eq 'after-rp1') {
3772     $self->{onerror}->(node => $item->{node},
3773 wakaba 1.104 type => 'element missing',
3774     text => 'rt',
3775     level => $self->{level}->{must});
3776 wakaba 1.99 $self->{onerror}->(node => $item->{node},
3777 wakaba 1.104 type => 'element missing',
3778     text => 'rp',
3779     level => $self->{level}->{must});
3780 wakaba 1.99 } elsif ($element_state->{phase} eq 'after-rp-rt') {
3781     $self->{onerror}->(node => $item->{node},
3782 wakaba 1.104 type => 'element missing',
3783     text => 'rp',
3784     level => $self->{level}->{must});
3785 wakaba 1.99 } else {
3786     die "check_child_text: Bad |ruby| phase: $element_state->{phase}";
3787     }
3788    
3789     ## NOTE: A modified version of |check_end| of %AnyChecker.
3790     if ($element_state->{has_significant} or $element_state->{has_sig}) {
3791     $item->{real_parent_state}->{has_significant} = 1;
3792     }
3793     },
3794     };
3795    
3796     $Element->{$HTML_NS}->{rt} = {
3797     %HTMLPhrasingContentChecker,
3798     status => FEATURE_HTML5_DEFAULT | FEATURE_RUBY_REC,
3799     check_attrs => $GetHTMLAttrsChecker->({}, {
3800     %HTMLAttrStatus,
3801     %HTMLM12NXHTML2CommonAttrStatus,
3802     lang => FEATURE_HTML5_DEFAULT,
3803     }),
3804     };
3805    
3806     $Element->{$HTML_NS}->{rp} = {
3807     %HTMLTextChecker,
3808     status => FEATURE_HTML5_DEFAULT | FEATURE_RUBY_REC,
3809     check_attrs => $GetHTMLAttrsChecker->({}, {
3810     %HTMLAttrStatus,
3811     %HTMLM12NXHTML2CommonAttrStatus,
3812     lang => FEATURE_HTML5_DEFAULT,
3813     }),
3814 wakaba 1.100 check_start => sub {
3815 wakaba 1.99 my ($self, $item, $element_state) = @_;
3816     $element_state->{text} = '';
3817 wakaba 1.100
3818     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3819     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3820 wakaba 1.99 },
3821     check_child_text => sub {
3822     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
3823     if ($has_significant) {
3824     $element_state->{text} .= $child_node->data;
3825     ## NOTE: |<rp> <!---->(</rp>| is allowed.
3826     }
3827     },
3828     check_end => sub {
3829     my ($self, $item, $element_state) = @_;
3830    
3831     my $p_class = ($item->{parent_state} and
3832     $item->{parent_state}->{phase} and
3833     $item->{parent_state}->{phase} eq 'after-rp2')
3834     ? qr/\p{Pe}/ : qr/\p{Ps}/;
3835     if ($element_state->{text} =~ /\A$p_class\z/) {
3836 wakaba 1.132 #=~ /\A[\x09\x0A\x0C\x0D\x20]*${p_class}[\x09\x0A\x0C\x0D\x20]*\z/) {
3837 wakaba 1.99 #
3838     } else {
3839     $self->{onerror}->(node => $item->{node},
3840 wakaba 1.104 type => 'rp:syntax error',
3841     level => $self->{level}->{must});
3842 wakaba 1.99 }
3843    
3844     $HTMLTextChecker{check_end}->(@_);
3845     },
3846     };
3847    
3848 wakaba 1.29 =pod
3849    
3850     ## TODO:
3851    
3852     +
3853     + <p>Partly because of the confusion described above, authors are
3854     + strongly recommended to always mark up all paragraphs with the
3855     + <code>p</code> element, and to not have any <code>ins</code> or
3856     + <code>del</code> elements that cross across any <span
3857     + title="paragraph">implied paragraphs</span>.</p>
3858     +
3859     (An informative note)
3860    
3861     <p><code>ins</code> elements should not cross <span
3862     + title="paragraph">implied paragraph</span> boundaries.</p>
3863     (normative)
3864    
3865     + <p><code>del</code> elements should not cross <span
3866     + title="paragraph">implied paragraph</span> boundaries.</p>
3867     (normative)
3868    
3869     =cut
3870    
3871 wakaba 1.1 $Element->{$HTML_NS}->{ins} = {
3872 wakaba 1.40 %HTMLTransparentChecker,
3873 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3874 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
3875 wakaba 1.1 cite => $HTMLURIAttrChecker,
3876     datetime => $HTMLDatetimeAttrChecker,
3877 wakaba 1.49 }, {
3878     %HTMLAttrStatus,
3879     %HTMLM12NCommonAttrStatus,
3880 wakaba 1.50 cite => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3881 wakaba 1.72 datetime => FEATURE_HTML5_FD | FEATURE_M12N10_REC,
3882 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3883 wakaba 1.1 }),
3884 wakaba 1.66 check_start => sub {
3885     my ($self, $item, $element_state) = @_;
3886    
3887     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
3888 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3889     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3890 wakaba 1.66 },
3891 wakaba 1.1 };
3892    
3893     $Element->{$HTML_NS}->{del} = {
3894 wakaba 1.40 %HTMLTransparentChecker,
3895 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3896 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
3897 wakaba 1.1 cite => $HTMLURIAttrChecker,
3898     datetime => $HTMLDatetimeAttrChecker,
3899 wakaba 1.49 }, {
3900     %HTMLAttrStatus,
3901     %HTMLM12NCommonAttrStatus,
3902 wakaba 1.50 cite => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3903 wakaba 1.72 datetime => FEATURE_HTML5_FD | FEATURE_M12N10_REC,
3904 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
3905 wakaba 1.1 }),
3906 wakaba 1.40 check_end => sub {
3907     my ($self, $item, $element_state) = @_;
3908     if ($element_state->{has_significant}) {
3909     ## NOTE: Significantness flag does not propagate.
3910     } elsif ($item->{transparent}) {
3911     #
3912     } else {
3913     $self->{onerror}->(node => $item->{node},
3914 wakaba 1.104 level => $self->{level}->{should},
3915 wakaba 1.40 type => 'no significant content');
3916     }
3917 wakaba 1.1 },
3918 wakaba 1.66 check_start => sub {
3919     my ($self, $item, $element_state) = @_;
3920    
3921     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
3922 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3923     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3924 wakaba 1.66 },
3925 wakaba 1.1 };
3926    
3927 wakaba 1.35 $Element->{$HTML_NS}->{figure} = {
3928 wakaba 1.72 %HTMLFlowContentChecker,
3929 wakaba 1.48 status => FEATURE_HTML5_FD,
3930 wakaba 1.72 ## NOTE: legend, Flow | Flow, legend?
3931 wakaba 1.41 check_child_element => sub {
3932     my ($self, $item, $child_el, $child_nsuri, $child_ln,
3933     $child_is_transparent, $element_state) = @_;
3934 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
3935     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
3936 wakaba 1.41 $self->{onerror}->(node => $child_el,
3937     type => 'element not allowed:minus',
3938 wakaba 1.104 level => $self->{level}->{must});
3939 wakaba 1.41 $element_state->{has_non_legend} = 1;
3940     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
3941     #
3942     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'legend') {
3943     if ($element_state->{has_legend_at_first}) {
3944     $self->{onerror}->(node => $child_el,
3945     type => 'element not allowed:figure legend',
3946 wakaba 1.104 level => $self->{level}->{must});
3947 wakaba 1.41 } elsif ($element_state->{has_legend}) {
3948     $self->{onerror}->(node => $element_state->{has_legend},
3949     type => 'element not allowed:figure legend',
3950 wakaba 1.104 level => $self->{level}->{must});
3951 wakaba 1.41 $element_state->{has_legend} = $child_el;
3952     } elsif ($element_state->{has_non_legend}) {
3953     $element_state->{has_legend} = $child_el;
3954     } else {
3955     $element_state->{has_legend_at_first} = 1;
3956 wakaba 1.35 }
3957 wakaba 1.41 delete $element_state->{has_non_legend};
3958     } else {
3959 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
3960 wakaba 1.43 $element_state->{has_non_legend} = 1 unless $child_is_transparent;
3961 wakaba 1.41 }
3962     },
3963     check_child_text => sub {
3964     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
3965     if ($has_significant) {
3966     $element_state->{has_non_legend} = 1;
3967 wakaba 1.35 }
3968 wakaba 1.41 },
3969     check_end => sub {
3970     my ($self, $item, $element_state) = @_;
3971 wakaba 1.35
3972 wakaba 1.41 if ($element_state->{has_legend_at_first}) {
3973     #
3974     } elsif ($element_state->{has_legend}) {
3975     if ($element_state->{has_non_legend}) {
3976     $self->{onerror}->(node => $element_state->{has_legend},
3977 wakaba 1.35 type => 'element not allowed:figure legend',
3978 wakaba 1.104 level => $self->{level}->{must});
3979 wakaba 1.35 }
3980     }
3981 wakaba 1.41
3982 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
3983 wakaba 1.41 ## ISSUE: |<figure><legend>aa</legend></figure>| should be an error?
3984 wakaba 1.35 },
3985     };
3986 wakaba 1.8 ## TODO: Test for <nest/> in <figure/>
3987 wakaba 1.1
3988 wakaba 1.92 my $AttrCheckerNotImplemented = sub {
3989     my ($self, $attr) = @_;
3990 wakaba 1.104 $self->{onerror}->(node => $attr,
3991     type => 'unknown attribute',
3992     level => $self->{level}->{uncertain});
3993 wakaba 1.92 };
3994    
3995 wakaba 1.1 $Element->{$HTML_NS}->{img} = {
3996 wakaba 1.40 %HTMLEmptyChecker,
3997 wakaba 1.98 status => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3998 wakaba 1.40 check_attrs => sub {
3999     my ($self, $item, $element_state) = @_;
4000 wakaba 1.1 $GetHTMLAttrsChecker->({
4001 wakaba 1.70 align => $GetHTMLEnumeratedAttrChecker->({
4002     bottom => 1, middle => 1, top => 1, left => 1, right => 1,
4003     }),
4004 wakaba 1.1 alt => sub { }, ## NOTE: No syntactical requirement
4005 wakaba 1.70 border => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4006 wakaba 1.1 src => $HTMLURIAttrChecker,
4007     usemap => $HTMLUsemapAttrChecker,
4008 wakaba 1.70 hspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4009 wakaba 1.1 ismap => sub {
4010 wakaba 1.40 my ($self, $attr, $parent_item) = @_;
4011     if (not $self->{flag}->{in_a_href}) {
4012 wakaba 1.15 $self->{onerror}->(node => $attr,
4013 wakaba 1.59 type => 'attribute not allowed:ismap',
4014 wakaba 1.104 level => $self->{level}->{must});
4015 wakaba 1.1 }
4016 wakaba 1.40 $GetHTMLBooleanAttrChecker->('ismap')->($self, $attr, $parent_item);
4017 wakaba 1.1 },
4018 wakaba 1.70 longdesc => $HTMLURIAttrChecker,
4019     ## TODO: HTML4 |name|
4020 wakaba 1.92 height => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4021 wakaba 1.70 vspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4022 wakaba 1.92 width => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4023 wakaba 1.49 }, {
4024     %HTMLAttrStatus,
4025 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4026 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
4027 wakaba 1.98 alt => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4028 wakaba 1.49 border => FEATURE_M12N10_REC_DEPRECATED,
4029 wakaba 1.98 height => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4030 wakaba 1.49 hspace => FEATURE_M12N10_REC_DEPRECATED,
4031 wakaba 1.98 ismap => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4032 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4033 wakaba 1.49 longdesc => FEATURE_M12N10_REC,
4034     name => FEATURE_M12N10_REC_DEPRECATED,
4035 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
4036 wakaba 1.98 src => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4037     usemap => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4038 wakaba 1.49 vspace => FEATURE_M12N10_REC_DEPRECATED,
4039 wakaba 1.98 width => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4040 wakaba 1.66 })->($self, $item, $element_state);
4041 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'alt')) {
4042     $self->{onerror}->(node => $item->{node},
4043 wakaba 1.104 type => 'attribute missing',
4044     text => 'alt',
4045     level => $self->{level}->{should});
4046 wakaba 1.114 ## TODO: ...
4047 wakaba 1.1 }
4048 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'src')) {
4049     $self->{onerror}->(node => $item->{node},
4050 wakaba 1.104 type => 'attribute missing',
4051     text => 'src',
4052     level => $self->{level}->{must});
4053 wakaba 1.1 }
4054 wakaba 1.66
4055 wakaba 1.114 ## TODO: external resource check
4056    
4057 wakaba 1.66 $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4058     $element_state->{uri_info}->{lowsrc}->{type}->{embedded} = 1;
4059     $element_state->{uri_info}->{dynsrc}->{type}->{embedded} = 1;
4060     $element_state->{uri_info}->{longdesc}->{type}->{cite} = 1;
4061 wakaba 1.1 },
4062     };
4063    
4064     $Element->{$HTML_NS}->{iframe} = {
4065 wakaba 1.40 %HTMLTextChecker,
4066 wakaba 1.114 status => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4067 wakaba 1.49 ## NOTE: Not part of M12N10 Strict
4068 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4069 wakaba 1.92 height => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4070 wakaba 1.76 name => $HTMLBrowsingContextNameAttrChecker,
4071 wakaba 1.92 sandbox => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->({
4072     'allow-same-origin' => 1, 'allow-forms' => 1, 'allow-scripts' => 1,
4073     }),
4074     seemless => $GetHTMLBooleanAttrChecker->('seemless'),
4075 wakaba 1.1 src => $HTMLURIAttrChecker,
4076 wakaba 1.92 width => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4077 wakaba 1.49 }, {
4078     %HTMLAttrStatus,
4079     %HTMLM12NCommonAttrStatus,
4080     align => FEATURE_XHTML10_REC,
4081 wakaba 1.50 class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4082 wakaba 1.49 frameborder => FEATURE_M12N10_REC,
4083 wakaba 1.92 height => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4084 wakaba 1.50 id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4085 wakaba 1.49 longdesc => FEATURE_M12N10_REC,
4086     marginheight => FEATURE_M12N10_REC,
4087     marginwidth => FEATURE_M12N10_REC,
4088 wakaba 1.114 #name => FEATURE_HTML5_WD | FEATURE_M12N10_REC_DEPRECATED,
4089     name => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4090     sandbox => FEATURE_HTML5_WD,
4091 wakaba 1.49 scrolling => FEATURE_M12N10_REC,
4092 wakaba 1.114 seemless => FEATURE_HTML5_WD,
4093     src => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4094 wakaba 1.50 title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4095 wakaba 1.92 width => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4096 wakaba 1.1 }),
4097 wakaba 1.66 check_start => sub {
4098     my ($self, $item, $element_state) = @_;
4099    
4100     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4101 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4102     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4103 wakaba 1.66 },
4104 wakaba 1.40 };
4105    
4106 wakaba 1.1 $Element->{$HTML_NS}->{embed} = {
4107 wakaba 1.40 %HTMLEmptyChecker,
4108 wakaba 1.98 status => FEATURE_HTML5_WD,
4109 wakaba 1.40 check_attrs => sub {
4110     my ($self, $item, $element_state) = @_;
4111 wakaba 1.1 my $has_src;
4112 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
4113 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
4114     $attr_ns = '' unless defined $attr_ns;
4115     my $attr_ln = $attr->manakai_local_name;
4116     my $checker;
4117 wakaba 1.73
4118     my $status = {
4119     %HTMLAttrStatus,
4120     height => FEATURE_HTML5_DEFAULT,
4121 wakaba 1.98 src => FEATURE_HTML5_WD,
4122     type => FEATURE_HTML5_WD,
4123 wakaba 1.73 width => FEATURE_HTML5_DEFAULT,
4124     }->{$attr_ln};
4125    
4126 wakaba 1.1 if ($attr_ns eq '') {
4127     if ($attr_ln eq 'src') {
4128     $checker = $HTMLURIAttrChecker;
4129     $has_src = 1;
4130     } elsif ($attr_ln eq 'type') {
4131     $checker = $HTMLIMTAttrChecker;
4132 wakaba 1.92 } elsif ($attr_ln eq 'width' or $attr_ln eq 'height') {
4133     $checker = $AttrCheckerNotImplemented; ## TODO: because spec does not define them yet.
4134 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
4135     $attr_ln !~ /[A-Z]/) {
4136 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
4137     $status = $HTMLDatasetAttrStatus;
4138 wakaba 1.117 } elsif ($attr_ln !~ /^[Xx][Mm][Ll]/ and
4139 wakaba 1.118 $attr_ln !~ /[A-Z]/ and
4140 wakaba 1.117 $attr_ln =~ /\A\p{InXML_NCNameStartChar10}\p{InXMLNCNameChar10}*\z/) {
4141 wakaba 1.1 $checker = $HTMLAttrChecker->{$attr_ln}
4142     || sub { }; ## NOTE: Any local attribute is ok.
4143 wakaba 1.98 $status = FEATURE_HTML5_WD | FEATURE_ALLOWED;
4144 wakaba 1.117 } else {
4145     $checker = $HTMLAttrChecker->{$attr_ln};
4146 wakaba 1.1 }
4147     }
4148     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
4149 wakaba 1.82 || $AttrChecker->{$attr_ns}->{''};
4150     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
4151     || $AttrStatus->{$attr_ns}->{''};
4152     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
4153 wakaba 1.62
4154 wakaba 1.1 if ($checker) {
4155 wakaba 1.66 $checker->($self, $attr, $item, $element_state);
4156 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
4157 wakaba 1.54 #
4158 wakaba 1.1 } else {
4159 wakaba 1.104 $self->{onerror}->(node => $attr,
4160     type => 'unknown attribute',
4161     level => $self->{level}->{uncertain});
4162 wakaba 1.50 ## ISSUE: No conformance createria for global attributes in the spec
4163     }
4164    
4165 wakaba 1.82 $self->_attr_status_info ($attr, $status);
4166 wakaba 1.1 }
4167    
4168     unless ($has_src) {
4169 wakaba 1.40 $self->{onerror}->(node => $item->{node},
4170 wakaba 1.104 type => 'attribute missing',
4171     text => 'src',
4172 wakaba 1.114 level => $self->{level}->{info});
4173     ## NOTE: <embed> without src="" is allowed since revision 1929.
4174     ## We issues an informational message since <embed> w/o src=""
4175     ## is likely an authoring error.
4176 wakaba 1.1 }
4177 wakaba 1.114
4178     ## TODO: external resource check
4179 wakaba 1.66
4180     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4181 wakaba 1.1 },
4182     };
4183    
4184 wakaba 1.49 ## TODO:
4185     ## {applet} FEATURE_M12N10_REC_DEPRECATED
4186     ## class, id, title, alt, archive, code, codebase, height, object, width name style,hspace,vspace(xhtml10)
4187    
4188 wakaba 1.1 $Element->{$HTML_NS}->{object} = {
4189 wakaba 1.40 %HTMLTransparentChecker,
4190 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4191 wakaba 1.40 check_attrs => sub {
4192     my ($self, $item, $element_state) = @_;
4193 wakaba 1.1 $GetHTMLAttrsChecker->({
4194 wakaba 1.70 align => $GetHTMLEnumeratedAttrChecker->({
4195     bottom => 1, middle => 1, top => 1, left => 1, right => 1,
4196     }),
4197     archive => $HTMLSpaceURIsAttrChecker,
4198     ## TODO: Relative to @codebase
4199     border => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4200     classid => $HTMLURIAttrChecker,
4201     codebase => $HTMLURIAttrChecker,
4202     codetype => $HTMLIMTAttrChecker,
4203     ## TODO: "RECOMMENDED when |classid| is specified" [HTML4]
4204 wakaba 1.1 data => $HTMLURIAttrChecker,
4205 wakaba 1.70 declare => $GetHTMLBooleanAttrChecker->('declare'),
4206     ## NOTE: "The object MUST be instantiated by a subsequent OBJECT ..."
4207     ## [HTML4] but we don't know how to test this.
4208     hspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4209 wakaba 1.76 name => $HTMLBrowsingContextNameAttrChecker,
4210 wakaba 1.70 standby => sub {}, ## NOTE: %Text; in HTML4
4211 wakaba 1.1 type => $HTMLIMTAttrChecker,
4212     usemap => $HTMLUsemapAttrChecker,
4213 wakaba 1.92 height => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4214 wakaba 1.70 vspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4215 wakaba 1.92 width => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4216 wakaba 1.49 }, {
4217     %HTMLAttrStatus,
4218 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4219 wakaba 1.49 align => FEATURE_XHTML10_REC,
4220 wakaba 1.82 archive => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4221 wakaba 1.49 border => FEATURE_XHTML10_REC,
4222     classid => FEATURE_M12N10_REC,
4223     codebase => FEATURE_M12N10_REC,
4224     codetype => FEATURE_M12N10_REC,
4225 wakaba 1.82 'content-length' => FEATURE_XHTML2_ED,
4226 wakaba 1.50 data => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4227 wakaba 1.49 datafld => FEATURE_HTML4_REC_RESERVED,
4228     dataformatas => FEATURE_HTML4_REC_RESERVED,
4229     datasrc => FEATURE_HTML4_REC_RESERVED,
4230 wakaba 1.82 declare => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4231 wakaba 1.50 height => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4232 wakaba 1.49 hspace => FEATURE_XHTML10_REC,
4233 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4234 wakaba 1.76 name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4235 wakaba 1.49 standby => FEATURE_M12N10_REC,
4236 wakaba 1.50 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4237     type => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4238     usemap => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4239 wakaba 1.49 vspace => FEATURE_XHTML10_REC,
4240 wakaba 1.50 width => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4241 wakaba 1.66 })->($self, $item, $element_state);
4242 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'data')) {
4243     unless ($item->{node}->has_attribute_ns (undef, 'type')) {
4244     $self->{onerror}->(node => $item->{node},
4245 wakaba 1.104 type => 'attribute missing:data|type',
4246     level => $self->{level}->{must});
4247 wakaba 1.1 }
4248     }
4249 wakaba 1.66
4250     $element_state->{uri_info}->{data}->{type}->{embedded} = 1;
4251     $element_state->{uri_info}->{classid}->{type}->{embedded} = 1;
4252     $element_state->{uri_info}->{codebase}->{type}->{base} = 1;
4253     ## TODO: archive
4254     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
4255 wakaba 1.1 },
4256 wakaba 1.72 ## NOTE: param*, transparent (Flow)
4257 wakaba 1.41 check_child_element => sub {
4258     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4259     $child_is_transparent, $element_state) = @_;
4260 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4261     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4262 wakaba 1.41 $self->{onerror}->(node => $child_el,
4263     type => 'element not allowed:minus',
4264 wakaba 1.104 level => $self->{level}->{must});
4265 wakaba 1.41 $element_state->{has_non_legend} = 1;
4266     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4267     #
4268     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'param') {
4269     if ($element_state->{has_non_param}) {
4270 wakaba 1.104 $self->{onerror}->(node => $child_el,
4271 wakaba 1.72 type => 'element not allowed:flow',
4272 wakaba 1.104 level => $self->{level}->{must});
4273 wakaba 1.39 }
4274 wakaba 1.41 } else {
4275 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
4276 wakaba 1.41 $element_state->{has_non_param} = 1;
4277 wakaba 1.39 }
4278 wakaba 1.25 },
4279 wakaba 1.41 check_child_text => sub {
4280     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4281     if ($has_significant) {
4282     $element_state->{has_non_param} = 1;
4283     }
4284 wakaba 1.42 },
4285     check_end => sub {
4286     my ($self, $item, $element_state) = @_;
4287     if ($element_state->{has_significant}) {
4288 wakaba 1.46 $item->{real_parent_state}->{has_significant} = 1;
4289 wakaba 1.42 } elsif ($item->{node}->manakai_parent_element) {
4290     ## NOTE: Transparent.
4291     } else {
4292     $self->{onerror}->(node => $item->{node},
4293 wakaba 1.104 level => $self->{level}->{should},
4294 wakaba 1.42 type => 'no significant content');
4295     }
4296     },
4297 wakaba 1.8 ## TODO: Tests for <nest/> in <object/>
4298 wakaba 1.1 };
4299 wakaba 1.41 ## ISSUE: Is |<menu><object data><li>aa</li></object></menu>| conforming?
4300     ## What about |<section><object data><style scoped></style>x</object></section>|?
4301     ## |<section><ins></ins><object data><style scoped></style>x</object></section>|?
4302 wakaba 1.1
4303     $Element->{$HTML_NS}->{param} = {
4304 wakaba 1.40 %HTMLEmptyChecker,
4305 wakaba 1.94 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4306 wakaba 1.40 check_attrs => sub {
4307     my ($self, $item, $element_state) = @_;
4308 wakaba 1.1 $GetHTMLAttrsChecker->({
4309     name => sub { },
4310 wakaba 1.70 type => $HTMLIMTAttrChecker,
4311 wakaba 1.1 value => sub { },
4312 wakaba 1.70 valuetype => $GetHTMLEnumeratedAttrChecker->({
4313     data => 1, ref => 1, object => 1,
4314     }),
4315 wakaba 1.49 }, {
4316     %HTMLAttrStatus,
4317 wakaba 1.121 href => FEATURE_RDFA_PR,
4318 wakaba 1.50 id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4319 wakaba 1.94 name => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4320 wakaba 1.82 type => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4321 wakaba 1.94 value => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4322 wakaba 1.82 valuetype => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4323 wakaba 1.66 })->(@_);
4324 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'name')) {
4325     $self->{onerror}->(node => $item->{node},
4326 wakaba 1.104 type => 'attribute missing',
4327     text => 'name',
4328     level => $self->{level}->{must});
4329 wakaba 1.1 }
4330 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'value')) {
4331     $self->{onerror}->(node => $item->{node},
4332 wakaba 1.104 type => 'attribute missing',
4333     text => 'value',
4334     level => $self->{level}->{must});
4335 wakaba 1.1 }
4336     },
4337     };
4338    
4339     $Element->{$HTML_NS}->{video} = {
4340 wakaba 1.40 %HTMLTransparentChecker,
4341 wakaba 1.48 status => FEATURE_HTML5_LC,
4342 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4343 wakaba 1.1 src => $HTMLURIAttrChecker,
4344     ## TODO: start, loopstart, loopend, end
4345     ## ISSUE: they MUST be "value time offset"s. Value?
4346 wakaba 1.11 ## ISSUE: playcount has no conformance creteria
4347 wakaba 1.1 autoplay => $GetHTMLBooleanAttrChecker->('autoplay'),
4348     controls => $GetHTMLBooleanAttrChecker->('controls'),
4349 wakaba 1.59 poster => $HTMLURIAttrChecker,
4350 wakaba 1.92 height => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4351     width => $AttrCheckerNotImplemented, ## TODO: spec does not define yet
4352 wakaba 1.50 }, {
4353     %HTMLAttrStatus,
4354     autoplay => FEATURE_HTML5_LC,
4355     controls => FEATURE_HTML5_LC,
4356     end => FEATURE_HTML5_LC,
4357     height => FEATURE_HTML5_LC,
4358     loopend => FEATURE_HTML5_LC,
4359     loopstart => FEATURE_HTML5_LC,
4360     playcount => FEATURE_HTML5_LC,
4361     poster => FEATURE_HTML5_LC,
4362     src => FEATURE_HTML5_LC,
4363     start => FEATURE_HTML5_LC,
4364     width => FEATURE_HTML5_LC,
4365 wakaba 1.1 }),
4366 wakaba 1.42 check_start => sub {
4367     my ($self, $item, $element_state) = @_;
4368     $element_state->{allow_source}
4369     = not $item->{node}->has_attribute_ns (undef, 'src');
4370     $element_state->{has_source} ||= $element_state->{allow_source} * -1;
4371     ## NOTE: It might be set true by |check_element|.
4372 wakaba 1.66
4373     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4374     $element_state->{uri_info}->{poster}->{type}->{embedded} = 1;
4375 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4376     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4377 wakaba 1.42 },
4378     check_child_element => sub {
4379     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4380     $child_is_transparent, $element_state) = @_;
4381 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4382     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4383 wakaba 1.42 $self->{onerror}->(node => $child_el,
4384     type => 'element not allowed:minus',
4385 wakaba 1.104 level => $self->{level}->{must});
4386 wakaba 1.42 delete $element_state->{allow_source};
4387     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4388     #
4389     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'source') {
4390 wakaba 1.45 unless ($element_state->{allow_source}) {
4391 wakaba 1.104 $self->{onerror}->(node => $child_el,
4392 wakaba 1.72 type => 'element not allowed:flow',
4393 wakaba 1.104 level => $self->{level}->{must});
4394 wakaba 1.42 }
4395 wakaba 1.45 $element_state->{has_source} = 1;
4396 wakaba 1.1 } else {
4397 wakaba 1.42 delete $element_state->{allow_source};
4398 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
4399 wakaba 1.42 }
4400     },
4401     check_child_text => sub {
4402     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4403     if ($has_significant) {
4404     delete $element_state->{allow_source};
4405     }
4406 wakaba 1.72 $HTMLFlowContentChecker{check_child_text}->(@_);
4407 wakaba 1.42 },
4408     check_end => sub {
4409     my ($self, $item, $element_state) = @_;
4410     if ($element_state->{has_source} == -1) {
4411     $self->{onerror}->(node => $item->{node},
4412 wakaba 1.104 type => 'child element missing',
4413     text => 'source',
4414     level => $self->{level}->{must});
4415 wakaba 1.1 }
4416 wakaba 1.42
4417     $Element->{$HTML_NS}->{object}->{check_end}->(@_);
4418 wakaba 1.1 },
4419     };
4420    
4421     $Element->{$HTML_NS}->{audio} = {
4422 wakaba 1.40 %{$Element->{$HTML_NS}->{video}},
4423 wakaba 1.48 status => FEATURE_HTML5_LC,
4424 wakaba 1.42 check_attrs => $GetHTMLAttrsChecker->({
4425     src => $HTMLURIAttrChecker,
4426     ## TODO: start, loopstart, loopend, end
4427     ## ISSUE: they MUST be "value time offset"s. Value?
4428     ## ISSUE: playcount has no conformance creteria
4429     autoplay => $GetHTMLBooleanAttrChecker->('autoplay'),
4430     controls => $GetHTMLBooleanAttrChecker->('controls'),
4431 wakaba 1.50 }, {
4432     %HTMLAttrStatus,
4433     autoplay => FEATURE_HTML5_LC,
4434     controls => FEATURE_HTML5_LC,
4435     end => FEATURE_HTML5_LC,
4436     loopend => FEATURE_HTML5_LC,
4437     loopstart => FEATURE_HTML5_LC,
4438     playcount => FEATURE_HTML5_LC,
4439     src => FEATURE_HTML5_LC,
4440     start => FEATURE_HTML5_LC,
4441 wakaba 1.42 }),
4442 wakaba 1.1 };
4443    
4444     $Element->{$HTML_NS}->{source} = {
4445 wakaba 1.40 %HTMLEmptyChecker,
4446 wakaba 1.48 status => FEATURE_HTML5_DEFAULT,
4447 wakaba 1.40 check_attrs => sub {
4448     my ($self, $item, $element_state) = @_;
4449 wakaba 1.1 $GetHTMLAttrsChecker->({
4450 wakaba 1.90 media => $HTMLMQAttrChecker,
4451     pixelratio => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
4452     src => $HTMLURIAttrChecker, ## ISSUE: Negative or zero pixelratio=""
4453 wakaba 1.1 type => $HTMLIMTAttrChecker,
4454 wakaba 1.50 }, {
4455     %HTMLAttrStatus,
4456     media => FEATURE_HTML5_DEFAULT,
4457 wakaba 1.90 pixelratio => FEATURE_HTML5_DEFAULT,
4458 wakaba 1.50 src => FEATURE_HTML5_DEFAULT,
4459     type => FEATURE_HTML5_DEFAULT,
4460 wakaba 1.66 })->(@_);
4461 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'src')) {
4462     $self->{onerror}->(node => $item->{node},
4463 wakaba 1.104 type => 'attribute missing',
4464     text => 'src',
4465     level => $self->{level}->{must});
4466 wakaba 1.1 }
4467 wakaba 1.66
4468     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4469 wakaba 1.1 },
4470     };
4471    
4472     $Element->{$HTML_NS}->{canvas} = {
4473 wakaba 1.40 %HTMLTransparentChecker,
4474 wakaba 1.89 status => FEATURE_HTML5_COMPLETE,
4475 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4476 wakaba 1.1 height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4477     width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4478 wakaba 1.50 }, {
4479     %HTMLAttrStatus,
4480 wakaba 1.89 height => FEATURE_HTML5_COMPLETE,
4481     width => FEATURE_HTML5_COMPLETE,
4482 wakaba 1.1 }),
4483     };
4484    
4485     $Element->{$HTML_NS}->{map} = {
4486 wakaba 1.72 %HTMLFlowContentChecker,
4487 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4488 wakaba 1.40 check_attrs => sub {
4489     my ($self, $item, $element_state) = @_;
4490 wakaba 1.100 my $has_name;
4491 wakaba 1.4 $GetHTMLAttrsChecker->({
4492 wakaba 1.100 name => sub {
4493     my ($self, $attr) = @_;
4494     my $value = $attr->value;
4495     if (length $value) {
4496     ## NOTE: Duplication is not non-conforming.
4497     ## NOTE: Space characters are not non-conforming.
4498     #
4499     } else {
4500     $self->{onerror}->(node => $attr,
4501     type => 'empty attribute value',
4502 wakaba 1.104 level => $self->{level}->{must});
4503 wakaba 1.100 }
4504 wakaba 1.4 $self->{map}->{$value} ||= $attr;
4505 wakaba 1.100 $has_name = [$value, $attr];
4506 wakaba 1.4 },
4507 wakaba 1.49 }, {
4508     %HTMLAttrStatus,
4509 wakaba 1.50 class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4510     dir => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4511     id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4512     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4513 wakaba 1.100 #name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC_DEPRECATED,
4514     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4515 wakaba 1.50 onclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4516     ondblclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4517     onmousedown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4518     onmouseup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4519     onmouseover => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4520     onmousemove => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4521     onmouseout => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4522     onkeypress => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4523     onkeydown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4524     onkeyup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4525     title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4526 wakaba 1.66 })->(@_);
4527 wakaba 1.100
4528 wakaba 1.135 if ($has_name) {
4529     my $id = $item->{node}->get_attribute ('id');
4530     if (defined $id and $has_name->[0] ne $id) {
4531     $self->{onerror}->(node => $item->{node}->get_attribute_node ('id'),
4532 wakaba 1.104 type => 'id ne name',
4533     level => $self->{level}->{must});
4534 wakaba 1.100 }
4535 wakaba 1.135 } else {
4536 wakaba 1.100 $self->{onerror}->(node => $item->{node},
4537 wakaba 1.104 type => 'attribute missing',
4538     text => 'name',
4539     level => $self->{level}->{must});
4540 wakaba 1.100 }
4541 wakaba 1.4 },
4542 wakaba 1.59 check_start => sub {
4543     my ($self, $item, $element_state) = @_;
4544     $element_state->{in_map_original} = $self->{flag}->{in_map};
4545     $self->{flag}->{in_map} = 1;
4546 wakaba 1.79
4547     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4548     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4549 wakaba 1.59 },
4550     check_end => sub {
4551     my ($self, $item, $element_state) = @_;
4552     delete $self->{flag}->{in_map} unless $element_state->{in_map_original};
4553 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
4554 wakaba 1.59 },
4555 wakaba 1.1 };
4556    
4557     $Element->{$HTML_NS}->{area} = {
4558 wakaba 1.40 %HTMLEmptyChecker,
4559 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4560 wakaba 1.40 check_attrs => sub {
4561     my ($self, $item, $element_state) = @_;
4562 wakaba 1.1 my %attr;
4563     my $coords;
4564 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
4565 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
4566     $attr_ns = '' unless defined $attr_ns;
4567     my $attr_ln = $attr->manakai_local_name;
4568     my $checker;
4569 wakaba 1.73 my $status;
4570 wakaba 1.1 if ($attr_ns eq '') {
4571 wakaba 1.73 $status = {
4572     %HTMLAttrStatus,
4573     %HTMLM12NCommonAttrStatus,
4574     accesskey => FEATURE_M12N10_REC,
4575     alt => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4576     coords => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4577 wakaba 1.121 href => FEATURE_HTML5_DEFAULT | FEATURE_RDFA_CR | FEATURE_M12N10_REC,
4578 wakaba 1.73 hreflang => FEATURE_HTML5_DEFAULT,
4579     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4580     media => FEATURE_HTML5_DEFAULT,
4581     nohref => FEATURE_M12N10_REC,
4582     onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4583     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4584     ping => FEATURE_HTML5_DEFAULT,
4585 wakaba 1.120 rel => FEATURE_HTML5_DEFAULT | FEATURE_RDFA_CR,
4586 wakaba 1.73 shape => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4587     tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4588     target => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4589     type => FEATURE_HTML5_DEFAULT,
4590     }->{$attr_ln};
4591    
4592 wakaba 1.1 $checker = {
4593 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
4594 wakaba 1.1 alt => sub { },
4595     ## NOTE: |alt| value has no conformance creteria.
4596     shape => $GetHTMLEnumeratedAttrChecker->({
4597     circ => -1, circle => 1,
4598     default => 1,
4599     poly => 1, polygon => -1,
4600     rect => 1, rectangle => -1,
4601     }),
4602     coords => sub {
4603     my ($self, $attr) = @_;
4604     my $value = $attr->value;
4605     if ($value =~ /\A-?[0-9]+(?>,-?[0-9]+)*\z/) {
4606     $coords = [split /,/, $value];
4607     } else {
4608     $self->{onerror}->(node => $attr,
4609 wakaba 1.104 type => 'coords:syntax error',
4610     level => $self->{level}->{must});
4611 wakaba 1.1 }
4612     },
4613 wakaba 1.70 nohref => $GetHTMLBooleanAttrChecker->('nohref'),
4614     target => $HTMLTargetAttrChecker,
4615 wakaba 1.1 href => $HTMLURIAttrChecker,
4616     ping => $HTMLSpaceURIsAttrChecker,
4617 wakaba 1.40 rel => sub { $HTMLLinkTypesAttrChecker->(1, $item, @_) },
4618 wakaba 1.1 media => $HTMLMQAttrChecker,
4619     hreflang => $HTMLLanguageTagAttrChecker,
4620     type => $HTMLIMTAttrChecker,
4621     }->{$attr_ln};
4622     if ($checker) {
4623     $attr{$attr_ln} = $attr;
4624 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
4625     $attr_ln !~ /[A-Z]/) {
4626 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
4627     $status = $HTMLDatasetAttrStatus;
4628 wakaba 1.1 } else {
4629     $checker = $HTMLAttrChecker->{$attr_ln};
4630     }
4631     }
4632     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
4633 wakaba 1.82 || $AttrChecker->{$attr_ns}->{''};
4634     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
4635     || $AttrStatus->{$attr_ns}->{''};
4636     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
4637 wakaba 1.62
4638 wakaba 1.1 if ($checker) {
4639 wakaba 1.66 $checker->($self, $attr, $item, $element_state) if ref $checker;
4640 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
4641 wakaba 1.54 #
4642 wakaba 1.1 } else {
4643 wakaba 1.104 $self->{onerror}->(node => $attr,
4644     type => 'unknown attribute',
4645     level => $self->{level}->{uncertain});
4646 wakaba 1.1 ## ISSUE: No comformance createria for unknown attributes in the spec
4647     }
4648 wakaba 1.49
4649 wakaba 1.82 $self->_attr_status_info ($attr, $status);
4650 wakaba 1.1 }
4651    
4652     if (defined $attr{href}) {
4653 wakaba 1.4 $self->{has_hyperlink_element} = 1;
4654 wakaba 1.1 unless (defined $attr{alt}) {
4655 wakaba 1.40 $self->{onerror}->(node => $item->{node},
4656 wakaba 1.104 type => 'attribute missing',
4657     text => 'alt',
4658     level => $self->{level}->{must});
4659 wakaba 1.1 }
4660     } else {
4661     for (qw/target ping rel media hreflang type alt/) {
4662     if (defined $attr{$_}) {
4663     $self->{onerror}->(node => $attr{$_},
4664 wakaba 1.104 type => 'attribute not allowed',
4665     level => $self->{level}->{must});
4666 wakaba 1.1 }
4667     }
4668     }
4669    
4670     my $shape = 'rectangle';
4671     if (defined $attr{shape}) {
4672     $shape = {
4673     circ => 'circle', circle => 'circle',
4674     default => 'default',
4675     poly => 'polygon', polygon => 'polygon',
4676     rect => 'rectangle', rectangle => 'rectangle',
4677     }->{lc $attr{shape}->value} || 'rectangle';
4678     ## TODO: ASCII lowercase?
4679     }
4680    
4681     if ($shape eq 'circle') {
4682     if (defined $attr{coords}) {
4683     if (defined $coords) {
4684     if (@$coords == 3) {
4685     if ($coords->[2] < 0) {
4686     $self->{onerror}->(node => $attr{coords},
4687 wakaba 1.104 type => 'coords:out of range',
4688     index => 2,
4689     value => $coords->[2],
4690     level => $self->{level}->{must});
4691 wakaba 1.1 }
4692     } else {
4693     $self->{onerror}->(node => $attr{coords},
4694 wakaba 1.104 type => 'coords:number not 3',
4695     text => 0+@$coords,
4696     level => $self->{level}->{must});
4697 wakaba 1.1 }
4698     } else {
4699     ## NOTE: A syntax error has been reported.
4700     }
4701     } else {
4702 wakaba 1.40 $self->{onerror}->(node => $item->{node},
4703 wakaba 1.104 type => 'attribute missing',
4704     text => 'coords',
4705     level => $self->{level}->{must});
4706 wakaba 1.1 }
4707     } elsif ($shape eq 'default') {
4708     if (defined $attr{coords}) {
4709     $self->{onerror}->(node => $attr{coords},
4710 wakaba 1.104 type => 'attribute not allowed',
4711     level => $self->{level}->{must});
4712 wakaba 1.1 }
4713     } elsif ($shape eq 'polygon') {
4714     if (defined $attr{coords}) {
4715     if (defined $coords) {
4716     if (@$coords >= 6) {
4717     unless (@$coords % 2 == 0) {
4718     $self->{onerror}->(node => $attr{coords},
4719 wakaba 1.104 type => 'coords:number not even',
4720     text => 0+@$coords,
4721     level => $self->{level}->{must});
4722 wakaba 1.1 }
4723     } else {
4724     $self->{onerror}->(node => $attr{coords},
4725 wakaba 1.104 type => 'coords:number lt 6',
4726     text => 0+@$coords,
4727     level => $self->{level}->{must});
4728 wakaba 1.1 }
4729     } else {
4730     ## NOTE: A syntax error has been reported.
4731     }
4732     } else {
4733 wakaba 1.40 $self->{onerror}->(node => $item->{node},
4734 wakaba 1.104 type => 'attribute missing',
4735     text => 'coords',
4736     level => $self->{level}->{must});
4737 wakaba 1.1 }
4738     } elsif ($shape eq 'rectangle') {
4739     if (defined $attr{coords}) {
4740     if (defined $coords) {
4741     if (@$coords == 4) {
4742     unless ($coords->[0] < $coords->[2]) {
4743     $self->{onerror}->(node => $attr{coords},
4744 wakaba 1.104 type => 'coords:out of range',
4745     index => 0,
4746     value => $coords->[0],
4747     level => $self->{level}->{must});
4748 wakaba 1.1 }
4749     unless ($coords->[1] < $coords->[3]) {
4750     $self->{onerror}->(node => $attr{coords},
4751 wakaba 1.104 type => 'coords:out of range',
4752     index => 1,
4753     value => $coords->[1],
4754     level => $self->{level}->{must});
4755 wakaba 1.1 }
4756     } else {
4757     $self->{onerror}->(node => $attr{coords},
4758 wakaba 1.104 type => 'coords:number not 4',
4759     text => 0+@$coords,
4760     level => $self->{level}->{must});
4761 wakaba 1.1 }
4762     } else {
4763     ## NOTE: A syntax error has been reported.
4764     }
4765     } else {
4766 wakaba 1.40 $self->{onerror}->(node => $item->{node},
4767 wakaba 1.104 type => 'attribute missing',
4768     text => 'coords',
4769     level => $self->{level}->{must});
4770 wakaba 1.1 }
4771     }
4772 wakaba 1.66
4773     $element_state->{uri_info}->{href}->{type}->{hyperlink} = 1;
4774 wakaba 1.1 },
4775 wakaba 1.59 check_start => sub {
4776     my ($self, $item, $element_state) = @_;
4777     unless ($self->{flag}->{in_map} or
4778     not $item->{node}->manakai_parent_element) {
4779     $self->{onerror}->(node => $item->{node},
4780     type => 'element not allowed:area',
4781 wakaba 1.104 level => $self->{level}->{must});
4782 wakaba 1.59 }
4783 wakaba 1.79
4784     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4785     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4786 wakaba 1.59 },
4787 wakaba 1.1 };
4788    
4789     $Element->{$HTML_NS}->{table} = {
4790 wakaba 1.40 %HTMLChecker,
4791 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4792 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
4793 wakaba 1.86 cellpadding => $HTMLLengthAttrChecker,
4794     cellspacing => $HTMLLengthAttrChecker,
4795 wakaba 1.69 frame => $GetHTMLEnumeratedAttrChecker->({
4796     void => 1, above => 1, below => 1, hsides => 1, vsides => 1,
4797     lhs => 1, rhs => 1, box => 1, border => 1,
4798     }),
4799     rules => $GetHTMLEnumeratedAttrChecker->({
4800     none => 1, groups => 1, rows => 1, cols => 1, all => 1,
4801     }),
4802     summary => sub {}, ## NOTE: %Text; in HTML4.
4803     width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }), ## %Pixels;
4804     }, {
4805 wakaba 1.49 %HTMLAttrStatus,
4806 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4807 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
4808     bgcolor => FEATURE_M12N10_REC_DEPRECATED,
4809     border => FEATURE_M12N10_REC,
4810     cellpadding => FEATURE_M12N10_REC,
4811     cellspacing => FEATURE_M12N10_REC,
4812 wakaba 1.61 cols => FEATURE_RFC1942,
4813 wakaba 1.49 datafld => FEATURE_HTML4_REC_RESERVED,
4814     dataformatas => FEATURE_HTML4_REC_RESERVED,
4815     datapagesize => FEATURE_M12N10_REC,
4816     datasrc => FEATURE_HTML4_REC_RESERVED,
4817     frame => FEATURE_M12N10_REC,
4818 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4819 wakaba 1.49 rules => FEATURE_M12N10_REC,
4820     summary => FEATURE_M12N10_REC,
4821     width => FEATURE_M12N10_REC,
4822     }),
4823 wakaba 1.40 check_start => sub {
4824     my ($self, $item, $element_state) = @_;
4825     $element_state->{phase} = 'before caption';
4826 wakaba 1.66
4827     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
4828 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4829     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4830 wakaba 1.40 },
4831     check_child_element => sub {
4832     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4833     $child_is_transparent, $element_state) = @_;
4834 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4835     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4836 wakaba 1.40 $self->{onerror}->(node => $child_el,
4837     type => 'element not allowed:minus',
4838 wakaba 1.104 level => $self->{level}->{must});
4839 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4840     #
4841     } elsif ($element_state->{phase} eq 'in tbodys') {
4842     if ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
4843     #$element_state->{phase} = 'in tbodys';
4844     } elsif (not $element_state->{has_tfoot} and
4845     $child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
4846     $element_state->{phase} = 'after tfoot';
4847     $element_state->{has_tfoot} = 1;
4848     } else {
4849 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
4850     level => $self->{level}->{must});
4851 wakaba 1.40 }
4852     } elsif ($element_state->{phase} eq 'in trs') {
4853     if ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
4854     #$element_state->{phase} = 'in trs';
4855     } elsif (not $element_state->{has_tfoot} and
4856     $child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
4857     $element_state->{phase} = 'after tfoot';
4858     $element_state->{has_tfoot} = 1;
4859     } else {
4860 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
4861     level => $self->{level}->{must});
4862 wakaba 1.40 }
4863     } elsif ($element_state->{phase} eq 'after thead') {
4864     if ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
4865     $element_state->{phase} = 'in tbodys';
4866     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
4867     $element_state->{phase} = 'in trs';
4868     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
4869     $element_state->{phase} = 'in tbodys';
4870     $element_state->{has_tfoot} = 1;
4871     } else {
4872 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
4873     level => $self->{level}->{must});
4874 wakaba 1.40 }
4875     } elsif ($element_state->{phase} eq 'in colgroup') {
4876     if ($child_nsuri eq $HTML_NS and $child_ln eq 'colgroup') {
4877     $element_state->{phase} = 'in colgroup';
4878     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'thead') {
4879     $element_state->{phase} = 'after thead';
4880     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
4881     $element_state->{phase} = 'in tbodys';
4882     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
4883     $element_state->{phase} = 'in trs';
4884     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
4885     $element_state->{phase} = 'in tbodys';
4886     $element_state->{has_tfoot} = 1;
4887     } else {
4888 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
4889     level => $self->{level}->{must});
4890 wakaba 1.40 }
4891     } elsif ($element_state->{phase} eq 'before caption') {
4892     if ($child_nsuri eq $HTML_NS and $child_ln eq 'caption') {
4893     $element_state->{phase} = 'in colgroup';
4894     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'colgroup') {
4895     $element_state->{phase} = 'in colgroup';
4896     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'thead') {
4897     $element_state->{phase} = 'after thead';
4898     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
4899     $element_state->{phase} = 'in tbodys';
4900     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
4901     $element_state->{phase} = 'in trs';
4902     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
4903     $element_state->{phase} = 'in tbodys';
4904     $element_state->{has_tfoot} = 1;
4905     } else {
4906 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
4907     level => $self->{level}->{must});
4908 wakaba 1.40 }
4909     } elsif ($element_state->{phase} eq 'after tfoot') {
4910 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
4911     level => $self->{level}->{must});
4912 wakaba 1.40 } else {
4913     die "check_child_element: Bad |table| phase: $element_state->{phase}";
4914     }
4915     },
4916     check_child_text => sub {
4917     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4918     if ($has_significant) {
4919 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
4920     level => $self->{level}->{must});
4921 wakaba 1.1 }
4922 wakaba 1.40 },
4923     check_end => sub {
4924     my ($self, $item, $element_state) = @_;
4925 wakaba 1.1
4926     ## Table model errors
4927     require Whatpm::HTMLTable;
4928 wakaba 1.87 my $table = Whatpm::HTMLTable->form_table ($item->{node}, sub {
4929 wakaba 1.104 $self->{onerror}->(@_);
4930     }, $self->{level});
4931 wakaba 1.87 Whatpm::HTMLTable->assign_header
4932 wakaba 1.104 ($table, $self->{onerror}, $self->{level});
4933 wakaba 1.87 push @{$self->{return}->{table}}, $table;
4934 wakaba 1.1
4935 wakaba 1.40 $HTMLChecker{check_end}->(@_);
4936 wakaba 1.1 },
4937     };
4938    
4939     $Element->{$HTML_NS}->{caption} = {
4940 wakaba 1.40 %HTMLPhrasingContentChecker,
4941 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4942 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
4943     align => $GetHTMLEnumeratedAttrChecker->({
4944     top => 1, bottom => 1, left => 1, right => 1,
4945     }),
4946     }, {
4947 wakaba 1.49 %HTMLAttrStatus,
4948 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4949 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
4950 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4951 wakaba 1.49 }),
4952 wakaba 1.1 };
4953    
4954 wakaba 1.69 my %cellalign = (
4955     ## HTML4 %cellhalign;
4956 wakaba 1.70 align => $GetHTMLEnumeratedAttrChecker->({
4957     left => 1, center => 1, right => 1, justify => 1, char => 1,
4958     }),
4959     char => sub {
4960     my ($self, $attr) = @_;
4961 wakaba 1.69
4962 wakaba 1.70 ## NOTE: "character" or |%Character;| in HTML4.
4963    
4964     my $value = $attr->value;
4965     if (length $value != 1) {
4966     $self->{onerror}->(node => $attr, type => 'char:syntax error',
4967 wakaba 1.105 level => $self->{level}->{html4_fact});
4968 wakaba 1.70 }
4969     },
4970 wakaba 1.86 charoff => $HTMLLengthAttrChecker,
4971    
4972 wakaba 1.69 ## HTML4 %cellvalign;
4973 wakaba 1.70 valign => $GetHTMLEnumeratedAttrChecker->({
4974     top => 1, middle => 1, bottom => 1, baseline => 1,
4975     }),
4976 wakaba 1.69 );
4977    
4978 wakaba 1.1 $Element->{$HTML_NS}->{colgroup} = {
4979 wakaba 1.40 %HTMLEmptyChecker,
4980 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4981 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4982 wakaba 1.69 %cellalign,
4983 wakaba 1.1 span => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
4984     ## NOTE: Defined only if "the |colgroup| element contains no |col| elements"
4985     ## TODO: "attribute not supported" if |col|.
4986     ## ISSUE: MUST NOT if any |col|?
4987     ## ISSUE: MUST NOT for |<colgroup span="1"><any><col/></any></colgroup>| (though non-conforming)?
4988 wakaba 1.49 }, {
4989     %HTMLAttrStatus,
4990 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4991 wakaba 1.49 align => FEATURE_M12N10_REC,
4992     char => FEATURE_M12N10_REC,
4993     charoff => FEATURE_M12N10_REC,
4994 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
4995 wakaba 1.82 span => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4996 wakaba 1.49 valign => FEATURE_M12N10_REC,
4997     width => FEATURE_M12N10_REC,
4998 wakaba 1.1 }),
4999 wakaba 1.40 check_child_element => sub {
5000     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5001     $child_is_transparent, $element_state) = @_;
5002 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5003     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5004 wakaba 1.40 $self->{onerror}->(node => $child_el,
5005     type => 'element not allowed:minus',
5006 wakaba 1.104 level => $self->{level}->{must});
5007 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5008     #
5009     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'col') {
5010     #
5011     } else {
5012 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5013     level => $self->{level}->{must});
5014 wakaba 1.40 }
5015     },
5016     check_child_text => sub {
5017     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5018     if ($has_significant) {
5019 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5020     level => $self->{level}->{must});
5021 wakaba 1.1 }
5022     },
5023     };
5024    
5025     $Element->{$HTML_NS}->{col} = {
5026 wakaba 1.40 %HTMLEmptyChecker,
5027 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5028 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5029 wakaba 1.69 %cellalign,
5030 wakaba 1.1 span => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5031 wakaba 1.49 }, {
5032     %HTMLAttrStatus,
5033 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5034 wakaba 1.49 align => FEATURE_M12N10_REC,
5035     char => FEATURE_M12N10_REC,
5036     charoff => FEATURE_M12N10_REC,
5037 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5038 wakaba 1.82 span => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5039 wakaba 1.49 valign => FEATURE_M12N10_REC,
5040     width => FEATURE_M12N10_REC,
5041 wakaba 1.1 }),
5042     };
5043    
5044     $Element->{$HTML_NS}->{tbody} = {
5045 wakaba 1.40 %HTMLChecker,
5046 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5047 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
5048     %cellalign,
5049     }, {
5050 wakaba 1.49 %HTMLAttrStatus,
5051 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5052 wakaba 1.49 align => FEATURE_M12N10_REC,
5053     char => FEATURE_M12N10_REC,
5054     charoff => FEATURE_M12N10_REC,
5055 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5056 wakaba 1.49 valign => FEATURE_M12N10_REC,
5057     }),
5058 wakaba 1.40 check_child_element => sub {
5059     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5060     $child_is_transparent, $element_state) = @_;
5061 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5062     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5063 wakaba 1.40 $self->{onerror}->(node => $child_el,
5064     type => 'element not allowed:minus',
5065 wakaba 1.104 level => $self->{level}->{must});
5066 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5067     #
5068     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
5069 wakaba 1.84 #
5070 wakaba 1.40 } else {
5071 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5072     level => $self->{level}->{must});
5073 wakaba 1.40 }
5074     },
5075     check_child_text => sub {
5076     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5077     if ($has_significant) {
5078 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5079     level => $self->{level}->{must});
5080 wakaba 1.1 }
5081 wakaba 1.40 },
5082 wakaba 1.1 };
5083    
5084     $Element->{$HTML_NS}->{thead} = {
5085 wakaba 1.40 %{$Element->{$HTML_NS}->{tbody}},
5086 wakaba 1.1 };
5087    
5088     $Element->{$HTML_NS}->{tfoot} = {
5089 wakaba 1.40 %{$Element->{$HTML_NS}->{tbody}},
5090 wakaba 1.1 };
5091    
5092     $Element->{$HTML_NS}->{tr} = {
5093 wakaba 1.40 %HTMLChecker,
5094 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5095 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
5096     %cellalign,
5097     bgcolor => $HTMLColorAttrChecker,
5098     }, {
5099 wakaba 1.49 %HTMLAttrStatus,
5100 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5101 wakaba 1.49 align => FEATURE_M12N10_REC,
5102     bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5103     char => FEATURE_M12N10_REC,
5104     charoff => FEATURE_M12N10_REC,
5105 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5106 wakaba 1.49 valign => FEATURE_M12N10_REC,
5107     }),
5108 wakaba 1.40 check_child_element => sub {
5109     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5110     $child_is_transparent, $element_state) = @_;
5111 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5112     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5113 wakaba 1.40 $self->{onerror}->(node => $child_el,
5114     type => 'element not allowed:minus',
5115 wakaba 1.104 level => $self->{level}->{must});
5116 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5117     #
5118     } elsif ($child_nsuri eq $HTML_NS and
5119     ($child_ln eq 'td' or $child_ln eq 'th')) {
5120 wakaba 1.84 #
5121 wakaba 1.40 } else {
5122 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5123     level => $self->{level}->{must});
5124 wakaba 1.40 }
5125     },
5126     check_child_text => sub {
5127     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5128     if ($has_significant) {
5129 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5130     level => $self->{level}->{must});
5131 wakaba 1.1 }
5132     },
5133     };
5134    
5135     $Element->{$HTML_NS}->{td} = {
5136 wakaba 1.72 %HTMLFlowContentChecker,
5137 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5138 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5139 wakaba 1.69 %cellalign,
5140     abbr => sub {}, ## NOTE: HTML4 %Text; and SHOULD be short.
5141     axis => sub {}, ## NOTE: HTML4 "cdata", comma-separated
5142     bgcolor => $HTMLColorAttrChecker,
5143 wakaba 1.1 colspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5144 wakaba 1.87 headers => sub {
5145     ## NOTE: Will be checked by Whatpm::HTMLTable->assign_header.
5146     ## Though that method does not check the |headers| attribute of a
5147     ## |td| element if the element does not form a table, in that case
5148     ## the |td| element is non-conforming anyway.
5149     },
5150 wakaba 1.69 nowrap => $GetHTMLBooleanAttrChecker->('nowrap'),
5151 wakaba 1.1 rowspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5152 wakaba 1.69 scope => $GetHTMLEnumeratedAttrChecker
5153     ->({row => 1, col => 1, rowgroup => 1, colgroup => 1}),
5154 wakaba 1.49 }, {
5155     %HTMLAttrStatus,
5156 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5157     abbr => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5158 wakaba 1.49 align => FEATURE_M12N10_REC,
5159 wakaba 1.82 axis => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5160 wakaba 1.49 bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5161     char => FEATURE_M12N10_REC,
5162     charoff => FEATURE_M12N10_REC,
5163 wakaba 1.82 colspan => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5164 wakaba 1.87 headers => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5165 wakaba 1.49 height => FEATURE_M12N10_REC_DEPRECATED,
5166 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5167 wakaba 1.49 nowrap => FEATURE_M12N10_REC_DEPRECATED,
5168 wakaba 1.82 rowspan => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5169     scope => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5170 wakaba 1.49 valign => FEATURE_M12N10_REC,
5171     width => FEATURE_M12N10_REC_DEPRECATED,
5172 wakaba 1.1 }),
5173     };
5174    
5175     $Element->{$HTML_NS}->{th} = {
5176 wakaba 1.40 %HTMLPhrasingContentChecker,
5177 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5178 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5179 wakaba 1.69 %cellalign,
5180     abbr => sub {}, ## NOTE: HTML4 %Text; and SHOULD be short.
5181     axis => sub {}, ## NOTE: HTML4 "cdata", comma-separated
5182     bgcolor => $HTMLColorAttrChecker,
5183 wakaba 1.1 colspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5184 wakaba 1.87 ## TODO: HTML4(?) |headers|
5185 wakaba 1.69 nowrap => $GetHTMLBooleanAttrChecker->('nowrap'),
5186 wakaba 1.1 rowspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5187     scope => $GetHTMLEnumeratedAttrChecker
5188     ->({row => 1, col => 1, rowgroup => 1, colgroup => 1}),
5189 wakaba 1.49 }, {
5190     %HTMLAttrStatus,
5191 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5192     abbr => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5193 wakaba 1.49 align => FEATURE_M12N10_REC,
5194 wakaba 1.82 axis => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5195 wakaba 1.49 bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5196     char => FEATURE_M12N10_REC,
5197     charoff => FEATURE_M12N10_REC,
5198 wakaba 1.82 colspan => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5199     headers => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5200 wakaba 1.49 height => FEATURE_M12N10_REC_DEPRECATED,
5201 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5202 wakaba 1.49 nowrap => FEATURE_M12N10_REC_DEPRECATED,
5203 wakaba 1.82 rowspan => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5204     scope => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5205 wakaba 1.49 valign => FEATURE_M12N10_REC,
5206     width => FEATURE_M12N10_REC_DEPRECATED,
5207 wakaba 1.1 }),
5208     };
5209    
5210 wakaba 1.52 $Element->{$HTML_NS}->{form} = {
5211 wakaba 1.121 %HTMLFlowContentChecker,
5212 wakaba 1.119 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5213 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5214 wakaba 1.56 accept => $AttrCheckerNotImplemented, ## TODO: ContentTypes [WF2]
5215 wakaba 1.129 'accept-charset' => $HTMLCharsetsAttrChecker,
5216 wakaba 1.52 action => $HTMLURIAttrChecker, ## TODO: "User agent behavior for a value other than HTTP URI is undefined" [HTML4]
5217 wakaba 1.56 data => $HTMLURIAttrChecker, ## TODO: MUST point ... [WF2]
5218     enctype => $HTMLIMTAttrChecker, ## TODO: "multipart/form-data" should be used when type=file is used [HTML4] ## TODO: MUST NOT parameter [WF2]
5219     method => $GetHTMLEnumeratedAttrChecker->({
5220     get => 1, post => 1, put => 1, delete => 1,
5221     }),
5222 wakaba 1.52 ## NOTE: "get" SHOULD be used for idempotent submittion,
5223     ## "post" SHOULD be used otherwise [HTML4]. This cannot be tested.
5224 wakaba 1.133 name => sub {
5225     my ($self, $attr) = @_;
5226    
5227     my $value = $attr->value;
5228     if ($value eq '') {
5229     $self->{onerror}->(type => 'empty form name',
5230     node => $attr,
5231     level => $self->{level}->{must});
5232     } else {
5233     if ($self->{form}->{$value}) {
5234     $self->{onerror}->(type => 'duplicate form name',
5235     node => $attr,
5236     value => $value,
5237     level => $self->{level}->{must});
5238     } else {
5239     $self->{form}->{$value} = 1;
5240     }
5241     }
5242     },
5243 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5244     onforminput => $HTMLEventHandlerAttrChecker,
5245 wakaba 1.56 onreceived => $HTMLEventHandlerAttrChecker,
5246     replace => $GetHTMLEnumeratedAttrChecker->({document => 1, values => 1}),
5247 wakaba 1.52 target => $HTMLTargetAttrChecker,
5248 wakaba 1.121 ## TODO: Warn for combination whose behavior is not defined in HTML4/WF2.
5249 wakaba 1.52 }, {
5250     %HTMLAttrStatus,
5251     %HTMLM12NCommonAttrStatus,
5252 wakaba 1.119 accept => FEATURE_WF2X | FEATURE_M12N10_REC,
5253     'accept-charset' => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5254     action => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5255 wakaba 1.56 data => FEATURE_WF2,
5256 wakaba 1.119 enctype => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5257 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5258 wakaba 1.119 method => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5259     #name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC_DEPRECATED,
5260     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5261 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
5262     onforminput => FEATURE_WF2_INFORMATIVE,
5263 wakaba 1.56 onreceived => FEATURE_WF2,
5264 wakaba 1.52 onreset => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5265     onsubmit => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5266 wakaba 1.56 replace => FEATURE_WF2,
5267 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
5268     sdasuff => FEATURE_HTML20_RFC,
5269 wakaba 1.119 target => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5270 wakaba 1.52 }),
5271 wakaba 1.66 check_start => sub {
5272     my ($self, $item, $element_state) = @_;
5273 wakaba 1.121 $self->_add_minus_elements ($element_state, {$HTML_NS => {form => 1}});
5274 wakaba 1.66
5275     $element_state->{uri_info}->{action}->{type}->{action} = 1;
5276     $element_state->{uri_info}->{data}->{type}->{resource} = 1;
5277 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5278     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5279 wakaba 1.66 },
5280 wakaba 1.121 check_end => sub {
5281     my ($self, $item, $element_state) = @_;
5282     $self->_remove_minus_elements ($element_state);
5283    
5284     $HTMLFlowContentChecker{check_end}->(@_);
5285     },
5286 wakaba 1.52 };
5287    
5288     $Element->{$HTML_NS}->{fieldset} = {
5289 wakaba 1.134 %HTMLFlowContentChecker,
5290 wakaba 1.119 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5291 wakaba 1.56 check_attrs => $GetHTMLAttrsChecker->({
5292     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5293     ## TODO: form [WF2]
5294 wakaba 1.125 ## TODO: name [HTML5]
5295 wakaba 1.56 }, {
5296 wakaba 1.52 %HTMLAttrStatus,
5297     %HTMLM12NCommonAttrStatus,
5298 wakaba 1.119 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5299     form => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5300 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5301 wakaba 1.125 name => FEATURE_HTML5_DEFAULT,
5302 wakaba 1.52 }),
5303 wakaba 1.134 ## NOTE: legend, Flow
5304     check_child_element => sub {
5305     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5306     $child_is_transparent, $element_state) = @_;
5307     if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5308     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5309     $self->{onerror}->(node => $child_el,
5310     type => 'element not allowed:minus',
5311     level => $self->{level}->{must});
5312     $element_state->{has_non_legend} = 1;
5313     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5314     #
5315     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'legend') {
5316     if ($element_state->{has_non_legend}) {
5317     $self->{onerror}->(node => $child_el,
5318     type => 'element not allowed:details legend',
5319     level => $self->{level}->{must});
5320     }
5321     $element_state->{has_legend} = 1;
5322     $element_state->{has_non_legend} = 1;
5323     } else {
5324     $HTMLFlowContentChecker{check_child_element}->(@_);
5325     $element_state->{has_non_legend} = 1 unless $child_is_transparent;
5326     ## TODO:
5327     ## |<details><object><legend>xx</legend></object>..</details>|
5328     ## should be an error, since |object| is allowed as flow,
5329     ## therefore |details| part of the content model does not match.
5330     }
5331     },
5332     check_child_text => sub {
5333     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5334     if ($has_significant) {
5335     $element_state->{has_non_legend} = 1;
5336     }
5337     },
5338     check_end => sub {
5339     my ($self, $item, $element_state) = @_;
5340    
5341     unless ($element_state->{has_legend}) {
5342     $self->{onerror}->(node => $item->{node},
5343     type => 'child element missing',
5344     text => 'legend',
5345     level => $self->{level}->{must});
5346     }
5347    
5348     $HTMLFlowContentChecker{check_end}->(@_);
5349     ## ISSUE: |<details><legend>aa</legend></details>| error?
5350     },
5351     ## NOTE: This definition is partially reused by |details| element's
5352     ## checker.
5353 wakaba 1.52 ## TODO: Tests
5354     ## TODO: Tests for <nest/> in <fieldset>
5355     };
5356    
5357     $Element->{$HTML_NS}->{input} = {
5358 wakaba 1.119 %HTMLEmptyChecker,
5359     status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5360 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5361 wakaba 1.56 accept => $AttrCheckerNotImplemented, ## TODO: ContentTypes [WF2]
5362 wakaba 1.129 'accept-charset' => $HTMLCharsetsAttrChecker,
5363 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
5364 wakaba 1.56 action => $HTMLURIAttrChecker,
5365 wakaba 1.52 align => $GetHTMLEnumeratedAttrChecker->({
5366     top => 1, middle => 1, bottom => 1, left => 1, right => 1,
5367     }),
5368     alt => sub {}, ## NOTE: Text [M12N] ## TODO: |alt| should be provided for |type=image| [HTML4]
5369     ## NOTE: HTML4 has a "should" for accessibility, which cannot be tested
5370     ## here.
5371 wakaba 1.56 autocomplete => $GetHTMLEnumeratedAttrChecker->({on => 1, off => 1}),
5372     autofocus => $GetHTMLBooleanAttrChecker->('autofocus'),
5373 wakaba 1.52 checked => $GetHTMLBooleanAttrChecker->('checked'),
5374     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5375 wakaba 1.56 enctype => $HTMLIMTAttrChecker,
5376     ## TODO: form [WF2]
5377     ## TODO: inputmode [WF2]
5378 wakaba 1.52 ismap => $GetHTMLBooleanAttrChecker->('ismap'),
5379 wakaba 1.56 ## TODO: list [WF2]
5380     ## TODO: max [WF2]
5381 wakaba 1.52 maxlength => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
5382 wakaba 1.56 method => $GetHTMLEnumeratedAttrChecker->({
5383     get => 1, post => 1, put => 1, delete => 1,
5384     }),
5385     ## TODO: min [WF2]
5386 wakaba 1.52 name => sub {}, ## NOTE: CDATA [M12N]
5387 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5388     onforminput => $HTMLEventHandlerAttrChecker,
5389     oninput => $HTMLEventHandlerAttrChecker,
5390     oninvalid => $HTMLEventHandlerAttrChecker,
5391 wakaba 1.119 ## TODO: pattern
5392 wakaba 1.52 readonly => $GetHTMLBooleanAttrChecker->('readonly'),
5393 wakaba 1.56 replace => $GetHTMLEnumeratedAttrChecker->({document => 1, values => 1}),
5394     required => $GetHTMLBooleanAttrChecker->('required'),
5395 wakaba 1.52 size => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
5396     src => $HTMLURIAttrChecker,
5397 wakaba 1.56 ## TODO: step [WF2]
5398     target => $HTMLTargetAttrChecker,
5399 wakaba 1.80 ## NOTE: According to Web Forms 2.0, |input| attribute has |template|
5400     ## attribute to support the |add| button type (as part of repetition
5401     ## template feature). It conflicts with the |template| global attribute
5402     ## introduced as part of the data template feature.
5403     ## NOTE: |template| attribute as defined in Web Forms 2.0 has no
5404     ## author requirement.
5405 wakaba 1.52 type => $GetHTMLEnumeratedAttrChecker->({
5406     text => 1, password => 1, checkbox => 1, radio => 1, submit => 1,
5407     reset => 1, file => 1, hidden => 1, image => 1, button => 1,
5408 wakaba 1.56 ## [WF2]
5409     datatime => 1, 'datetime-local' => 1, date => 1, month => 1, week => 1,
5410     time => 1, number => 1, range => 1, email => 1, url => 1,
5411     add => 1, remove => 1, 'move-up' => 1, 'move-down' => 1,
5412 wakaba 1.52 }),
5413     usemap => $HTMLUsemapAttrChecker,
5414 wakaba 1.56 value => sub {}, ## NOTE: CDATA [M12N] ## TODO: "optional except when the type attribute has the value "radio" or "checkbox"" [HTML4] ## TODO: constraints [WF2]
5415     ## TODO: "authors should ensure that in each set of radio buttons that one is initially "on"." [HTML4] [WF2]
5416 wakaba 1.52 }, {
5417     %HTMLAttrStatus,
5418     %HTMLM12NCommonAttrStatus,
5419 wakaba 1.119 accept => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5420 wakaba 1.61 'accept-charset' => FEATURE_HTML2X_RFC,
5421 wakaba 1.52 accesskey => FEATURE_M12N10_REC,
5422 wakaba 1.119 action => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5423 wakaba 1.52 align => FEATURE_M12N10_REC_DEPRECATED,
5424 wakaba 1.119 alt => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5425     autocomplete => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5426     autofocus => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5427     checked => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5428 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
5429     dataformatas => FEATURE_HTML4_REC_RESERVED,
5430     datasrc => FEATURE_HTML4_REC_RESERVED,
5431 wakaba 1.119 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5432     enctype => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5433     form => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5434     inputmode => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_XHTMLBASIC11_CR,
5435 wakaba 1.52 ismap => FEATURE_M12N10_REC,
5436     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5437 wakaba 1.119 list => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5438     max => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5439     maxlength => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5440     method => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5441     min => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5442     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5443 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5444     onchange => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5445     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5446 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
5447     onforminput => FEATURE_WF2_INFORMATIVE,
5448     oninput => FEATURE_WF2,
5449     oninvalid => FEATURE_WF2,
5450 wakaba 1.52 onselect => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5451 wakaba 1.119 pattern => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5452     readonly => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5453 wakaba 1.65 replace => FEATURE_WF2,
5454 wakaba 1.119 required => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5455 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
5456 wakaba 1.119 size => FEATURE_HTML5_DEFAULT | FEATURE_WF2X_DEPRECATED | FEATURE_M12N10_REC,
5457     src => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5458     step => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5459 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5460 wakaba 1.119 target => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5461 wakaba 1.56 template => FEATURE_WF2,
5462 wakaba 1.119 type => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5463 wakaba 1.52 usemap => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC,
5464 wakaba 1.119 value => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5465 wakaba 1.52 }),
5466     ## TODO: Tests
5467     ## TODO: Tests for <nest/> in <input>
5468 wakaba 1.66 check_start => sub {
5469     my ($self, $item, $element_state) = @_;
5470    
5471     $element_state->{uri_info}->{action}->{type}->{action} = 1;
5472     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
5473     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
5474 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5475     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5476 wakaba 1.66 },
5477 wakaba 1.52 };
5478    
5479 wakaba 1.56 ## TODO: Form |name| attributes: MUST NOT conflict with RFC 3106 [WF2]
5480    
5481 wakaba 1.80 ## NOTE: "authors who are nesting repetition blocks should position such
5482     ## [repetition-block-related] buttons carefully to make clear which block a
5483     ## button applies to.": I have no idea how this can be tested.
5484    
5485 wakaba 1.52 $Element->{$HTML_NS}->{button} = {
5486 wakaba 1.119 %HTMLPhrasingContentChecker, ## ISSUE: -interactive?
5487     status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5488 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5489 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
5490 wakaba 1.56 action => $HTMLURIAttrChecker,
5491     autofocus => $GetHTMLBooleanAttrChecker->('autofocus'),
5492 wakaba 1.52 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5493 wakaba 1.56 ## TODO: form [WF2]
5494     method => $GetHTMLEnumeratedAttrChecker->({
5495     get => 1, post => 1, put => 1, delete => 1,
5496     }),
5497 wakaba 1.52 name => sub {}, ## NOTE: CDATA [M12N]
5498 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5499     onforminput => $HTMLEventHandlerAttrChecker,
5500 wakaba 1.56 replace => $GetHTMLEnumeratedAttrChecker->({document => 1, values => 1}),
5501     target => $HTMLTargetAttrChecker,
5502 wakaba 1.80 ## NOTE: According to Web Forms 2.0, |button| attribute has |template|
5503     ## attribute to support the |add| button type (as part of repetition
5504     ## template feature). It conflicts with the |template| global attribute
5505     ## introduced as part of the data template feature.
5506     ## NOTE: |template| attribute as defined in Web Forms 2.0 has no
5507     ## author requirement.
5508 wakaba 1.52 type => $GetHTMLEnumeratedAttrChecker->({
5509     button => 1, submit => 1, reset => 1,
5510 wakaba 1.80 ## [WF2]
5511     add => 1, remove => 1, 'move-up' => 1, 'move-down' => 1,
5512 wakaba 1.52 }),
5513     value => sub {}, ## NOTE: CDATA [M12N]
5514     }, {
5515     %HTMLAttrStatus,
5516     %HTMLM12NCommonAttrStatus,
5517     accesskey => FEATURE_M12N10_REC,
5518 wakaba 1.119 action => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5519     autofocus => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5520 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
5521     dataformatas => FEATURE_HTML4_REC_RESERVED,
5522     datasrc => FEATURE_HTML4_REC_RESERVED,
5523 wakaba 1.119 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5524     enctype => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5525     form => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5526 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5527 wakaba 1.119 method => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5528     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5529 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5530     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5531 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
5532     onforminput => FEATURE_WF2_INFORMATIVE,
5533 wakaba 1.56 replace => FEATURE_WF2,
5534 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5535 wakaba 1.119 target => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5536 wakaba 1.56 template => FEATURE_WF2,
5537 wakaba 1.119 type => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5538     value => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5539 wakaba 1.52 }),
5540 wakaba 1.66 check_start => sub {
5541     my ($self, $item, $element_state) = @_;
5542    
5543     $element_state->{uri_info}->{action}->{type}->{action} = 1;
5544     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
5545 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5546     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5547 wakaba 1.66 },
5548 wakaba 1.52 };
5549    
5550     $Element->{$HTML_NS}->{label} = {
5551 wakaba 1.119 %HTMLPhrasingContentChecker, ## ISSUE: -label? -interactive? At most one form control [WF2]?
5552     status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC
5553     | FEATURE_XHTML2_ED,
5554 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5555 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
5556 wakaba 1.52 for => $AttrCheckerNotImplemented, ## TODO: IDREF ## TODO: Must be |id| of control [HTML4] ## TODO: Or, "may only contain one control element"
5557 wakaba 1.119 ## TODO: form
5558 wakaba 1.52 }, {
5559     %HTMLAttrStatus,
5560 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5561 wakaba 1.56 accesskey => FEATURE_WF2 | FEATURE_M12N10_REC,
5562 wakaba 1.119 for => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5563     form => FEATURE_HTML5_DEFAULT,
5564 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5565     onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5566     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5567     }),
5568     ## TODO: Tests
5569     ## TODO: Tests for <nest/> in <label>
5570     };
5571    
5572     $Element->{$HTML_NS}->{select} = {
5573 wakaba 1.121 %HTMLChecker,
5574 wakaba 1.52 ## TODO: author should SELECTED at least one OPTION in non-MULTIPLE case [HTML4].
5575     ## TODO: more than one OPTION with SELECTED in non-MULTIPLE case is "error" [HTML4]
5576 wakaba 1.119 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5577 wakaba 1.56 is_root => 1, ## TODO: SHOULD NOT in application/xhtml+xml [WF2]
5578 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5579 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
5580 wakaba 1.56 autofocus => $GetHTMLBooleanAttrChecker->('autofocus'),
5581 wakaba 1.52 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5582 wakaba 1.56 data => $HTMLURIAttrChecker, ## TODO: MUST point ... [WF2]
5583     ## TODO: form [WF2]
5584 wakaba 1.52 multiple => $GetHTMLBooleanAttrChecker->('multiple'),
5585     name => sub {}, ## NOTE: CDATA [M12N]
5586 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5587     onforminput => $HTMLEventHandlerAttrChecker,
5588     oninput => $HTMLEventHandlerAttrChecker,
5589 wakaba 1.56 oninvalid => $HTMLEventHandlerAttrChecker,
5590 wakaba 1.52 size => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
5591     }, {
5592     %HTMLAttrStatus,
5593     %HTMLM12NCommonAttrStatus,
5594 wakaba 1.56 accesskey => FEATURE_WF2,
5595 wakaba 1.119 autofocus => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5596 wakaba 1.56 data => FEATURE_WF2,
5597 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
5598     dataformatas => FEATURE_HTML4_REC_RESERVED,
5599     datasrc => FEATURE_HTML4_REC_RESERVED,
5600 wakaba 1.119 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5601     form => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5602 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5603 wakaba 1.119 multiple => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5604     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5605 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5606     onchange => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5607 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
5608     onforminput => FEATURE_WF2_INFORMATIVE,
5609 wakaba 1.52 onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5610 wakaba 1.126 oninput => FEATURE_WF2,
5611 wakaba 1.56 oninvalid => FEATURE_WF2,
5612 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
5613     sdapref => FEATURE_HTML20_RFC,
5614 wakaba 1.119 size => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5615 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5616     }),
5617 wakaba 1.66 check_start => sub {
5618     my ($self, $item, $element_state) = @_;
5619    
5620     $element_state->{uri_info}->{data}->{type}->{resource} = 1;
5621     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
5622 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5623     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5624 wakaba 1.66 },
5625 wakaba 1.121 check_child_element => sub {
5626     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5627     $child_is_transparent, $element_state) = @_;
5628 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5629     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5630 wakaba 1.121 $self->{onerror}->(node => $child_el,
5631     type => 'element not allowed:minus',
5632     level => $self->{level}->{must});
5633     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5634     #
5635     } elsif ($child_nsuri eq $HTML_NS and
5636     {
5637     option => 1, optgroup => 1,
5638     }->{$child_ln}) {
5639     #
5640     } else {
5641     $self->{onerror}->(node => $child_el, type => 'element not allowed',
5642     level => $self->{level}->{must});
5643     }
5644     },
5645     check_child_text => sub {
5646     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5647     if ($has_significant) {
5648     $self->{onerror}->(node => $child_node, type => 'character not allowed',
5649     level => $self->{level}->{must});
5650     }
5651     },
5652 wakaba 1.52 };
5653 wakaba 1.1
5654 wakaba 1.52 $Element->{$HTML_NS}->{datalist} = {
5655 wakaba 1.121 %HTMLPhrasingContentChecker,
5656 wakaba 1.56 ## TODO: |option| child MUST be empty [WF2]
5657 wakaba 1.119 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5658 wakaba 1.56 check_attrs => $GetHTMLAttrsChecker->({
5659     data => $HTMLURIAttrChecker, ## TODO: MUST point ... [WF2]
5660     }, {
5661 wakaba 1.52 %HTMLAttrStatus,
5662 wakaba 1.56 data => FEATURE_WF2,
5663 wakaba 1.52 }),
5664 wakaba 1.66 check_start => sub {
5665     my ($self, $item, $element_state) = @_;
5666    
5667 wakaba 1.121 $element_state->{phase} = 'any'; # any | phrasing | option
5668    
5669 wakaba 1.66 $element_state->{uri_info}->{data}->{type}->{resource} = 1;
5670 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5671     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5672 wakaba 1.66 },
5673 wakaba 1.121 ## NOTE: phrasing | option*
5674     check_child_element => sub {
5675     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5676     $child_is_transparent, $element_state) = @_;
5677 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5678     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5679 wakaba 1.121 $self->{onerror}->(node => $child_el,
5680     type => 'element not allowed:minus',
5681     level => $self->{level}->{must});
5682     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5683     #
5684     } elsif ($element_state->{phase} eq 'phrasing') {
5685     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
5686     #
5687     } else {
5688     $self->{onerror}->(node => $child_el,
5689     type => 'element not allowed:phrasing',
5690     level => $self->{level}->{must});
5691     }
5692     } elsif ($element_state->{phase} eq 'option') {
5693     if ($child_nsuri eq $HTML_NS and $child_ln eq 'option') {
5694     #
5695     } else {
5696     $self->{onerror}->(node => $child_el,
5697     type => 'element not allowed',
5698     level => $self->{level}->{must});
5699     }
5700     } elsif ($element_state->{phase} eq 'any') {
5701     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
5702     $element_state->{phase} = 'phrasing';
5703     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'option') {
5704     $element_state->{phase} = 'option';
5705     } else {
5706     $self->{onerror}->(node => $child_el,
5707     type => 'element not allowed',
5708     level => $self->{level}->{must});
5709     }
5710     } else {
5711     die "check_child_element: Bad |datalist| phase: $element_state->{phase}";
5712     }
5713     },
5714     check_child_text => sub {
5715     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5716     if ($has_significant) {
5717     if ($element_state->{phase} eq 'phrasing') {
5718     #
5719     } elsif ($element_state->{phase} eq 'any') {
5720     $element_state->{phase} = 'phrasing';
5721     } else {
5722     $self->{onerror}->(node => $child_node,
5723     type => 'character not allowed',
5724     level => $self->{level}->{must});
5725     }
5726     }
5727     },
5728     check_end => sub {
5729     my ($self, $item, $element_state) = @_;
5730     if ($element_state->{phase} eq 'phrasing') {
5731     if ($element_state->{has_significant}) {
5732     $item->{real_parent_state}->{has_significant} = 1;
5733     } elsif ($item->{transparent}) {
5734     #
5735     } else {
5736     $self->{onerror}->(node => $item->{node},
5737     type => 'no significant content',
5738     level => $self->{level}->{should});
5739     }
5740     } else {
5741     ## NOTE: Since the content model explicitly allows a |datalist| element
5742     ## being empty, we don't raise "no significant content" error for this
5743     ## element when there is no element. (We should raise an error for
5744     ## |<datalist><br></datalist>|, however.)
5745     ## NOTE: As a side-effect, when the |datalist| element only contains
5746     ## non-conforming content, then the |phase| flag has not changed from
5747     ## |any|, no "no significant content" error is raised neither.
5748     $HTMLChecker{check_end}->(@_);
5749     }
5750     },
5751 wakaba 1.52 };
5752 wakaba 1.49
5753 wakaba 1.52 $Element->{$HTML_NS}->{optgroup} = {
5754 wakaba 1.121 %HTMLChecker,
5755 wakaba 1.119 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5756 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5757     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5758     label => sub {}, ## NOTE: Text [M12N] ## TODO: required
5759     }, {
5760     %HTMLAttrStatus,
5761     %HTMLM12NCommonAttrStatus,
5762 wakaba 1.119 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5763     label => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5764 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5765     }),
5766 wakaba 1.121 check_child_element => sub {
5767     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5768     $child_is_transparent, $element_state) = @_;
5769 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5770     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5771 wakaba 1.121 $self->{onerror}->(node => $child_el,
5772     type => 'element not allowed:minus',
5773     level => $self->{level}->{must});
5774     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5775     #
5776     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'option') {
5777     #
5778     } else {
5779     $self->{onerror}->(node => $child_el, type => 'element not allowed',
5780     level => $self->{level}->{must});
5781     }
5782     },
5783     check_child_text => sub {
5784     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5785     if ($has_significant) {
5786     $self->{onerror}->(node => $child_node, type => 'character not allowed',
5787     level => $self->{level}->{must});
5788     }
5789     },
5790 wakaba 1.52 };
5791    
5792     $Element->{$HTML_NS}->{option} = {
5793     %HTMLTextChecker,
5794 wakaba 1.119 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5795 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5796     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5797     label => sub {}, ## NOTE: Text [M12N]
5798     selected => $GetHTMLBooleanAttrChecker->('selected'),
5799     value => sub {}, ## NOTE: CDATA [M12N]
5800     }, {
5801     %HTMLAttrStatus,
5802     %HTMLM12NCommonAttrStatus,
5803 wakaba 1.119 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5804     label => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5805 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5806 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
5807     sdapref => FEATURE_HTML20_RFC,
5808 wakaba 1.119 selected => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5809     value => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5810 wakaba 1.52 }),
5811     };
5812 wakaba 1.49
5813 wakaba 1.52 $Element->{$HTML_NS}->{textarea} = {
5814     %HTMLTextChecker,
5815 wakaba 1.121 status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5816 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5817 wakaba 1.56 accept => $HTMLIMTAttrChecker, ## TODO: MUST be a text-based type
5818 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
5819 wakaba 1.56 autofocus => $GetHTMLBooleanAttrChecker->('autofocus'),
5820     cols => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }), ## TODO: SHOULD if wrap=hard [WF2]
5821 wakaba 1.52 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5822 wakaba 1.56 ## TODO: form [WF2]
5823     ## TODO: inputmode [WF2]
5824     maxlength => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
5825 wakaba 1.52 name => sub {}, ## NOTE: CDATA [M12N]
5826 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5827     onforminput => $HTMLEventHandlerAttrChecker,
5828 wakaba 1.127 oninput => $HTMLEventHandlerAttrChecker,
5829 wakaba 1.56 ## TODO: pattern [WF2] ## TODO: |title| special semantics
5830 wakaba 1.52 readonly => $GetHTMLBooleanAttrChecker->('readonly'),
5831 wakaba 1.56 required => $GetHTMLBooleanAttrChecker->('required'),
5832     rows => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
5833 wakaba 1.126 oninput => $HTMLEventHandlerAttrChecker,
5834 wakaba 1.56 oninvalid => $HTMLEventHandlerAttrChecker,
5835     wrap => $GetHTMLEnumeratedAttrChecker->({soft => 1, hard => 1}),
5836 wakaba 1.52 }, {
5837     %HTMLAttrStatus,
5838     %HTMLM12NCommonAttrStatus,
5839 wakaba 1.121 accept => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5840 wakaba 1.61 'accept-charset' => FEATURE_HTML2X_RFC,
5841 wakaba 1.52 accesskey => FEATURE_M12N10_REC,
5842 wakaba 1.121 autofocus => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5843     cols => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5844 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
5845 wakaba 1.49 dataformatas => FEATURE_HTML4_REC_RESERVED,
5846     datasrc => FEATURE_HTML4_REC_RESERVED,
5847 wakaba 1.121 disabled => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5848     form => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5849     inputmode => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_XHTMLBASIC11_CR,
5850 wakaba 1.52 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5851 wakaba 1.121 maxlength => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5852     name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5853 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5854     onchange => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5855     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5856 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
5857     onforminput => FEATURE_WF2_INFORMATIVE,
5858     oninput => FEATURE_WF2,
5859 wakaba 1.56 oninvalid => FEATURE_WF2,
5860 wakaba 1.52 onselect => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5861 wakaba 1.121 pattern => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5862     readonly => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5863     required => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5864     rows => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5865 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
5866     sdapref => FEATURE_HTML20_RFC,
5867 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5868 wakaba 1.121 wrap => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5869 wakaba 1.52 }),
5870 wakaba 1.66 check_start => sub {
5871     my ($self, $item, $element_state) = @_;
5872    
5873     $element_state->{uri_info}->{data}->{type}->{resource} = 1;
5874 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5875     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5876 wakaba 1.66 },
5877 wakaba 1.52 };
5878 wakaba 1.49
5879 wakaba 1.52 $Element->{$HTML_NS}->{output} = {
5880 wakaba 1.121 %HTMLPhrasingContentChecker,
5881     status => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5882 wakaba 1.56 check_attrs => $GetHTMLAttrsChecker->({
5883     ## TODO: for [WF2]
5884     ## TODO: form [WF2]
5885     ## TODO: name [WF2]
5886 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5887     onforminput => $HTMLEventHandlerAttrChecker,
5888 wakaba 1.56 }, {
5889 wakaba 1.52 %HTMLAttrStatus,
5890 wakaba 1.121 for => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5891     form => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5892     name => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
5893 wakaba 1.56 onchange => FEATURE_HTML5_DEFAULT | FEATURE_WF2,
5894     onformchange => FEATURE_WF2,
5895     onforminput => FEATURE_WF2,
5896 wakaba 1.52 }),
5897 wakaba 1.56 ## NOTE: "The output element should be used when ..." [WF2]
5898 wakaba 1.52 };
5899    
5900     $Element->{$HTML_NS}->{isindex} = {
5901     %HTMLEmptyChecker,
5902 wakaba 1.54 status => FEATURE_M12N10_REC_DEPRECATED |
5903     Whatpm::ContentChecker::FEATURE_DEPRECATED_SHOULD, ## [HTML4]
5904 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5905     prompt => sub {}, ## NOTE: Text [M12N]
5906     }, {
5907     %HTMLAttrStatus,
5908     class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5909     dir => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5910     id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5911     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5912     prompt => FEATURE_M12N10_REC_DEPRECATED,
5913 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
5914 wakaba 1.78 style => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5915 wakaba 1.52 title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5916     }),
5917     ## TODO: Tests
5918     ## TODO: Tests for <nest/> in <isindex>
5919 wakaba 1.66 check_start => sub {
5920     my ($self, $item, $element_state) = @_;
5921    
5922     $element_state->{uri_info}->{action}->{type}->{action} = 1;
5923 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5924     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5925 wakaba 1.66 },
5926 wakaba 1.52 };
5927 wakaba 1.49
5928 wakaba 1.1 $Element->{$HTML_NS}->{script} = {
5929 wakaba 1.40 %HTMLChecker,
5930 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5931 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5932 wakaba 1.91 charset => sub {
5933     my ($self, $attr) = @_;
5934    
5935     unless ($attr->owner_element->has_attribute_ns (undef, 'src')) {
5936     $self->{onerror}->(type => 'attribute not allowed',
5937     node => $attr,
5938 wakaba 1.104 level => $self->{level}->{must});
5939 wakaba 1.91 }
5940    
5941     $HTMLCharsetChecker->($attr->value, @_);
5942     },
5943 wakaba 1.86 language => sub {}, ## NOTE: No syntax constraint according to HTML4.
5944 wakaba 1.91 src => $HTMLURIAttrChecker, ## TODO: pointed resource MUST be in type of type="" (resource error)
5945 wakaba 1.1 defer => $GetHTMLBooleanAttrChecker->('defer'),
5946     async => $GetHTMLBooleanAttrChecker->('async'),
5947 wakaba 1.91 type => $HTMLIMTAttrChecker, ## TODO: MUST NOT: |charset=""| parameter
5948 wakaba 1.49 }, {
5949     %HTMLAttrStatus,
5950 wakaba 1.50 async => FEATURE_HTML5_DEFAULT,
5951 wakaba 1.91 charset => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5952 wakaba 1.50 defer => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5953 wakaba 1.49 event => FEATURE_HTML4_REC_RESERVED,
5954     for => FEATURE_HTML4_REC_RESERVED,
5955 wakaba 1.121 href => FEATURE_RDFA_PR,
5956 wakaba 1.50 id => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
5957 wakaba 1.49 language => FEATURE_M12N10_REC_DEPRECATED,
5958 wakaba 1.50 src => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5959     type => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5960 wakaba 1.9 }),
5961 wakaba 1.40 check_start => sub {
5962     my ($self, $item, $element_state) = @_;
5963 wakaba 1.1
5964 wakaba 1.40 if ($item->{node}->has_attribute_ns (undef, 'src')) {
5965     $element_state->{must_be_empty} = 1;
5966 wakaba 1.1 } else {
5967     ## NOTE: No content model conformance in HTML5 spec.
5968 wakaba 1.40 my $type = $item->{node}->get_attribute_ns (undef, 'type');
5969     my $language = $item->{node}->get_attribute_ns (undef, 'language');
5970 wakaba 1.1 if ((defined $type and $type eq '') or
5971     (defined $language and $language eq '')) {
5972     $type = 'text/javascript';
5973     } elsif (defined $type) {
5974     #
5975     } elsif (defined $language) {
5976     $type = 'text/' . $language;
5977     } else {
5978     $type = 'text/javascript';
5979     }
5980 wakaba 1.93
5981     if ($type =~ m[\A(?>(?>\x0D\x0A)?[\x09\x20])*([\x21\x23-\x27\x2A\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+)(?>(?>\x0D\x0A)?[\x09\x20])*/(?>(?>\x0D\x0A)?[\x09\x20])*([\x21\x23-\x27\x2A\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+)(?>(?>\x0D\x0A)?[\x09\x20])*(?>;|\z)]) {
5982     $type = "$1/$2";
5983     $type =~ tr/A-Z/a-z/; ## NOTE: ASCII case-insensitive
5984     ## TODO: Though we strip prameter here, it should not be ignored for the purpose of conformance checking...
5985     }
5986     $element_state->{script_type} = $type;
5987 wakaba 1.40 }
5988 wakaba 1.66
5989     $element_state->{uri_info}->{src}->{type}->{resource} = 1;
5990 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5991     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5992 wakaba 1.107
5993     $element_state->{text} = '';
5994 wakaba 1.40 },
5995     check_child_element => sub {
5996     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5997     $child_is_transparent, $element_state) = @_;
5998 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5999     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6000 wakaba 1.40 $self->{onerror}->(node => $child_el,
6001     type => 'element not allowed:minus',
6002 wakaba 1.104 level => $self->{level}->{must});
6003 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6004     #
6005     } else {
6006     if ($element_state->{must_be_empty}) {
6007     $self->{onerror}->(node => $child_el,
6008 wakaba 1.104 type => 'element not allowed:empty',
6009     level => $self->{level}->{must});
6010 wakaba 1.40 }
6011     }
6012     },
6013     check_child_text => sub {
6014     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6015     if ($has_significant and
6016     $element_state->{must_be_empty}) {
6017     $self->{onerror}->(node => $child_node,
6018 wakaba 1.104 type => 'character not allowed:empty',
6019     level => $self->{level}->{must});
6020 wakaba 1.40 }
6021 wakaba 1.115 $element_state->{text} .= $child_node->data;
6022 wakaba 1.40 },
6023     check_end => sub {
6024     my ($self, $item, $element_state) = @_;
6025     unless ($element_state->{must_be_empty}) {
6026 wakaba 1.93 if ($element_state->{script_type} =~ m![+/][Xx][Mm][Ll]\z!) {
6027     ## NOTE: XML content should be checked by THIS instance of checker
6028     ## as part of normal tree validation.
6029 wakaba 1.104 $self->{onerror}->(node => $item->{node},
6030     type => 'XML script lang',
6031     text => $element_state->{script_type},
6032     level => $self->{level}->{uncertain});
6033     ## ISSUE: Should we raise some kind of error for
6034     ## <script type="text/xml">aaaaa</script>?
6035     ## NOTE: ^^^ This is why we throw an "uncertain" error.
6036 wakaba 1.93 } else {
6037     $self->{onsubdoc}->({s => $element_state->{text},
6038     container_node => $item->{node},
6039     media_type => $element_state->{script_type},
6040     is_char_string => 1});
6041     }
6042 wakaba 1.40
6043     $HTMLChecker{check_end}->(@_);
6044 wakaba 1.1 }
6045     },
6046 wakaba 1.91 ## TODO: There MUST be |type| unless the script type is JavaScript. (resource error)
6047     ## NOTE: "When used to include script data, the script data must be embedded
6048     ## inline, the format of the data must be given using the type attribute,
6049     ## and the src attribute must not be specified." - not testable.
6050     ## TODO: It would be possible to err <script type=text/plain src=...>
6051 wakaba 1.1 };
6052 wakaba 1.25 ## ISSUE: Significant check and text child node
6053 wakaba 1.1
6054     ## NOTE: When script is disabled.
6055     $Element->{$HTML_NS}->{noscript} = {
6056 wakaba 1.40 %HTMLTransparentChecker,
6057 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6058     check_attrs => $GetHTMLAttrsChecker->({}, {
6059     %HTMLAttrStatus,
6060     %HTMLM12NCommonAttrStatus,
6061 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6062 wakaba 1.49 }),
6063 wakaba 1.40 check_start => sub {
6064     my ($self, $item, $element_state) = @_;
6065 wakaba 1.3
6066 wakaba 1.40 unless ($item->{node}->owner_document->manakai_is_html) {
6067 wakaba 1.104 $self->{onerror}->(node => $item->{node}, type => 'in XML:noscript',
6068     level => $self->{level}->{must});
6069 wakaba 1.3 }
6070    
6071 wakaba 1.40 unless ($self->{flag}->{in_head}) {
6072     $self->_add_minus_elements ($element_state,
6073     {$HTML_NS => {noscript => 1}});
6074     }
6075 wakaba 1.79
6076     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6077     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6078 wakaba 1.3 },
6079 wakaba 1.40 check_child_element => sub {
6080     my ($self, $item, $child_el, $child_nsuri, $child_ln,
6081     $child_is_transparent, $element_state) = @_;
6082     if ($self->{flag}->{in_head}) {
6083 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6084     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6085 wakaba 1.40 $self->{onerror}->(node => $child_el,
6086     type => 'element not allowed:minus',
6087 wakaba 1.104 level => $self->{level}->{must});
6088 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6089     #
6090     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'link') {
6091     #
6092     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'style') {
6093     if ($child_el->has_attribute_ns (undef, 'scoped')) {
6094     $self->{onerror}->(node => $child_el,
6095     type => 'element not allowed:head noscript',
6096 wakaba 1.104 level => $self->{level}->{must});
6097 wakaba 1.40 }
6098     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'meta') {
6099 wakaba 1.47 my $http_equiv_attr
6100     = $child_el->get_attribute_node_ns (undef, 'http-equiv');
6101     if ($http_equiv_attr) {
6102     ## TODO: case
6103     if (lc $http_equiv_attr->value eq 'content-type') {
6104 wakaba 1.40 $self->{onerror}->(node => $child_el,
6105 wakaba 1.34 type => 'element not allowed:head noscript',
6106 wakaba 1.104 level => $self->{level}->{must});
6107 wakaba 1.47 } else {
6108     #
6109 wakaba 1.3 }
6110 wakaba 1.47 } else {
6111     $self->{onerror}->(node => $child_el,
6112     type => 'element not allowed:head noscript',
6113 wakaba 1.104 level => $self->{level}->{must});
6114 wakaba 1.3 }
6115 wakaba 1.40 } else {
6116     $self->{onerror}->(node => $child_el,
6117     type => 'element not allowed:head noscript',
6118 wakaba 1.104 level => $self->{level}->{must});
6119 wakaba 1.40 }
6120     } else {
6121     $HTMLTransparentChecker{check_child_element}->(@_);
6122     }
6123     },
6124     check_child_text => sub {
6125     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6126     if ($self->{flag}->{in_head}) {
6127     if ($has_significant) {
6128     $self->{onerror}->(node => $child_node,
6129 wakaba 1.104 type => 'character not allowed',
6130     level => $self->{level}->{must});
6131 wakaba 1.3 }
6132     } else {
6133 wakaba 1.40 $HTMLTransparentChecker{check_child_text}->(@_);
6134     }
6135     },
6136     check_end => sub {
6137     my ($self, $item, $element_state) = @_;
6138     $self->_remove_minus_elements ($element_state);
6139     if ($self->{flag}->{in_head}) {
6140     $HTMLChecker{check_end}->(@_);
6141     } else {
6142     $HTMLPhrasingContentChecker{check_end}->(@_);
6143 wakaba 1.3 }
6144 wakaba 1.1 },
6145     };
6146 wakaba 1.3 ## ISSUE: Scripting is disabled: <head><noscript><html a></noscript></head>
6147 wakaba 1.1
6148     $Element->{$HTML_NS}->{'event-source'} = {
6149 wakaba 1.40 %HTMLEmptyChecker,
6150 wakaba 1.118 status => FEATURE_HTML5_LC_DROPPED,
6151     check_attrs => $GetHTMLAttrsChecker->({
6152     src => $HTMLURIAttrChecker,
6153     }, {
6154     %HTMLAttrStatus,
6155     src => FEATURE_HTML5_LC_DROPPED,
6156     }),
6157     check_start => sub {
6158     my ($self, $item, $element_state) = @_;
6159    
6160     $element_state->{uri_info}->{src}->{type}->{resource} = 1;
6161     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6162     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6163     },
6164     };
6165    
6166     $Element->{$HTML_NS}->{eventsource} = {
6167     %HTMLEmptyChecker,
6168     status => FEATURE_HTML5_DEFAULT,
6169 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6170 wakaba 1.1 src => $HTMLURIAttrChecker,
6171 wakaba 1.50 }, {
6172     %HTMLAttrStatus,
6173 wakaba 1.118 src => FEATURE_HTML5_DEFAULT,
6174 wakaba 1.1 }),
6175 wakaba 1.66 check_start => sub {
6176     my ($self, $item, $element_state) = @_;
6177    
6178     $element_state->{uri_info}->{src}->{type}->{resource} = 1;
6179 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6180     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6181 wakaba 1.66 },
6182 wakaba 1.1 };
6183    
6184     $Element->{$HTML_NS}->{details} = {
6185 wakaba 1.134 %{$Element->{$HTML_NS}->{fieldset}},
6186 wakaba 1.48 status => FEATURE_HTML5_WD,
6187 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6188 wakaba 1.1 open => $GetHTMLBooleanAttrChecker->('open'),
6189 wakaba 1.50 }, {
6190     %HTMLAttrStatus,
6191 wakaba 1.59 open => FEATURE_HTML5_WD,
6192 wakaba 1.1 }),
6193     };
6194    
6195     $Element->{$HTML_NS}->{datagrid} = {
6196 wakaba 1.72 %HTMLFlowContentChecker,
6197 wakaba 1.48 status => FEATURE_HTML5_WD,
6198 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6199 wakaba 1.1 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6200     multiple => $GetHTMLBooleanAttrChecker->('multiple'),
6201 wakaba 1.50 }, {
6202     %HTMLAttrStatus,
6203     disabled => FEATURE_HTML5_WD,
6204     multiple => FEATURE_HTML5_WD,
6205 wakaba 1.1 }),
6206 wakaba 1.40 check_start => sub {
6207     my ($self, $item, $element_state) = @_;
6208 wakaba 1.1
6209 wakaba 1.40 $self->_add_minus_elements ($element_state,
6210     {$HTML_NS => {a => 1, datagrid => 1}});
6211     $element_state->{phase} = 'any';
6212 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6213     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6214 wakaba 1.40 },
6215 wakaba 1.95 ## NOTE: Flow -(text* (table|select|datalist) Flow*) | table | select |
6216     ## datalist | Empty
6217 wakaba 1.40 check_child_element => sub {
6218     my ($self, $item, $child_el, $child_nsuri, $child_ln,
6219     $child_is_transparent, $element_state) = @_;
6220 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6221     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6222 wakaba 1.40 $self->{onerror}->(node => $child_el,
6223     type => 'element not allowed:minus',
6224 wakaba 1.104 level => $self->{level}->{must});
6225 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6226     #
6227 wakaba 1.72 } elsif ($element_state->{phase} eq 'flow') {
6228     if ($HTMLFlowContent->{$child_nsuri}->{$child_ln}) {
6229 wakaba 1.44 if (not $element_state->{has_element} and
6230 wakaba 1.40 $child_nsuri eq $HTML_NS and
6231 wakaba 1.95 {
6232     table => 1, select => 1, datalist => 1,
6233     }->{$child_ln}) {
6234 wakaba 1.40 $self->{onerror}->(node => $child_el,
6235 wakaba 1.104 type => 'element not allowed',
6236     level => $self->{level}->{must});
6237 wakaba 1.40 } else {
6238 wakaba 1.8 #
6239 wakaba 1.1 }
6240 wakaba 1.40 } else {
6241     $self->{onerror}->(node => $child_el,
6242 wakaba 1.121 type => 'element not allowed', ## TODO: :flow
6243 wakaba 1.104 level => $self->{level}->{must});
6244 wakaba 1.40 }
6245 wakaba 1.43 $element_state->{has_element} = 1;
6246 wakaba 1.40 } elsif ($element_state->{phase} eq 'any') {
6247     if ($child_nsuri eq $HTML_NS and
6248     {table => 1, select => 1, datalist => 1}->{$child_ln}) {
6249     $element_state->{phase} = 'none';
6250 wakaba 1.72 } elsif ($HTMLFlowContent->{$child_nsuri}->{$child_ln}) {
6251 wakaba 1.40 $element_state->{has_element} = 1;
6252 wakaba 1.72 $element_state->{phase} = 'flow';
6253 wakaba 1.40 } else {
6254     $self->{onerror}->(node => $child_el,
6255 wakaba 1.104 type => 'element not allowed',
6256     level => $self->{level}->{must});
6257 wakaba 1.40 }
6258     } elsif ($element_state->{phase} eq 'none') {
6259     $self->{onerror}->(node => $child_el,
6260 wakaba 1.104 type => 'element not allowed',
6261     level => $self->{level}->{must});
6262 wakaba 1.40 } else {
6263     die "check_child_element: Bad |datagrid| phase: $element_state->{phase}";
6264     }
6265     },
6266     check_child_text => sub {
6267     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6268     if ($has_significant) {
6269 wakaba 1.72 if ($element_state->{phase} eq 'flow') {
6270 wakaba 1.40 #
6271     } elsif ($element_state->{phase} eq 'any') {
6272 wakaba 1.72 $element_state->{phase} = 'flow';
6273 wakaba 1.40 } else {
6274     $self->{onerror}->(node => $child_node,
6275 wakaba 1.104 type => 'character not allowed',
6276     level => $self->{level}->{must});
6277 wakaba 1.1 }
6278     }
6279 wakaba 1.40 },
6280     check_end => sub {
6281     my ($self, $item, $element_state) = @_;
6282     $self->_remove_minus_elements ($element_state);
6283 wakaba 1.1
6284 wakaba 1.95 if ($element_state->{phase} eq 'flow') {
6285     if ($element_state->{has_significant}) {
6286     $item->{real_parent_state}->{has_significant} = 1;
6287     } elsif ($item->{transparent}) {
6288     #
6289     } else {
6290     $self->{onerror}->(node => $item->{node},
6291 wakaba 1.104 type => 'no significant content',
6292 wakaba 1.110 level => $self->{level}->{should});
6293 wakaba 1.95 }
6294     } else {
6295     ## NOTE: Since the content model explicitly allows a |datagird| element
6296     ## being empty, we don't raise "no significant content" error for this
6297     ## element when there is no element. (We should raise an error for
6298     ## |<datagrid><br></datagrid>|, however.)
6299     ## NOTE: As a side-effect, when the |datagrid| element only contains
6300     ## non-conforming content, then the |phase| flag has not changed from
6301     ## |any|, no "no significant content" error is raised neither.
6302     ## NOTE: Another side-effect of the current implementation:
6303     ## |<daragrid><datagrid/></datagrid>| has no "no significant content"
6304     ## error at all.
6305 wakaba 1.40 $HTMLChecker{check_end}->(@_);
6306     }
6307     },
6308 wakaba 1.1 };
6309    
6310     $Element->{$HTML_NS}->{command} = {
6311 wakaba 1.40 %HTMLEmptyChecker,
6312 wakaba 1.48 status => FEATURE_HTML5_WD,
6313 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6314 wakaba 1.1 checked => $GetHTMLBooleanAttrChecker->('checked'),
6315     default => $GetHTMLBooleanAttrChecker->('default'),
6316     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6317     icon => $HTMLURIAttrChecker,
6318     label => sub { }, ## NOTE: No conformance creteria
6319     radiogroup => sub { }, ## NOTE: No conformance creteria
6320     type => sub {
6321     my ($self, $attr) = @_;
6322     my $value = $attr->value;
6323     unless ({command => 1, checkbox => 1, radio => 1}->{$value}) {
6324 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'invalid attribute value',
6325     level => $self->{level}->{must});
6326 wakaba 1.1 }
6327     },
6328 wakaba 1.50 }, {
6329     %HTMLAttrStatus,
6330     checked => FEATURE_HTML5_WD,
6331     default => FEATURE_HTML5_WD,
6332     disabled => FEATURE_HTML5_WD,
6333     icon => FEATURE_HTML5_WD,
6334     label => FEATURE_HTML5_WD,
6335     radiogroup => FEATURE_HTML5_WD,
6336     type => FEATURE_HTML5_WD,
6337 wakaba 1.1 }),
6338 wakaba 1.66 check_start => sub {
6339     my ($self, $item, $element_state) = @_;
6340    
6341     $element_state->{uri_info}->{icon}->{type}->{embedded} = 1;
6342 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6343     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6344 wakaba 1.66 },
6345 wakaba 1.115 };
6346    
6347     $Element->{$HTML_NS}->{bb} = {
6348     %HTMLPhrasingContentChecker,
6349     status => FEATURE_HTML5_DEFAULT,
6350     check_attrs => $GetHTMLAttrsChecker->({
6351     type => $GetHTMLEnumeratedAttrChecker->({makeapp => 1}),
6352     }, {
6353     %HTMLAttrStatus,
6354     type => FEATURE_HTML5_DEFAULT,
6355     }),
6356 wakaba 1.130 check_start => sub {
6357     my ($self, $item, $element_state) = @_;
6358     $self->_add_minus_elements ($element_state, $HTMLInteractiveContent);
6359    
6360     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6361     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6362     },
6363     check_end => sub {
6364     my ($self, $item, $element_state) = @_;
6365     $self->_remove_minus_elements ($element_state);
6366    
6367     $HTMLTransparentChecker{check_end}->(@_);
6368     },
6369 wakaba 1.1 };
6370    
6371     $Element->{$HTML_NS}->{menu} = {
6372 wakaba 1.40 %HTMLPhrasingContentChecker,
6373 wakaba 1.54 #status => FEATURE_M12N10_REC_DEPRECATED | FEATURE_HTML5_WD,
6374     status => FEATURE_M12N10_REC | FEATURE_HTML5_WD,
6375     ## NOTE: We don't want any |menu| element warned as deprecated.
6376 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6377 wakaba 1.1 autosubmit => $GetHTMLBooleanAttrChecker->('autosubmit'),
6378 wakaba 1.68 compact => $GetHTMLBooleanAttrChecker->('compact'),
6379 wakaba 1.135 ## ISSUE: <menu id=""><p contextmenu=""> match? (In the current
6380     ## implementation, it does not match.)
6381 wakaba 1.1 label => sub { }, ## NOTE: No conformance creteria
6382     type => $GetHTMLEnumeratedAttrChecker->({context => 1, toolbar => 1}),
6383 wakaba 1.49 }, {
6384     %HTMLAttrStatus,
6385     %HTMLM12NCommonAttrStatus,
6386 wakaba 1.61 align => FEATURE_HTML2X_RFC,
6387 wakaba 1.113 autosubmit => FEATURE_HTML5_DROPPED,
6388 wakaba 1.49 compat => FEATURE_M12N10_REC_DEPRECATED,
6389 wakaba 1.50 label => FEATURE_HTML5_WD,
6390     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6391 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
6392     sdapref => FEATURE_HTML20_RFC,
6393 wakaba 1.50 type => FEATURE_HTML5_WD,
6394 wakaba 1.1 }),
6395 wakaba 1.40 check_start => sub {
6396     my ($self, $item, $element_state) = @_;
6397     $element_state->{phase} = 'li or phrasing';
6398     $element_state->{in_menu_original} = $self->{flag}->{in_menu};
6399     $self->{flag}->{in_menu} = 1;
6400 wakaba 1.79
6401     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6402     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6403 wakaba 1.135 $element_state->{id_type} = 'menu';
6404 wakaba 1.40 },
6405     check_child_element => sub {
6406     my ($self, $item, $child_el, $child_nsuri, $child_ln,
6407     $child_is_transparent, $element_state) = @_;
6408 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6409     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6410 wakaba 1.40 $self->{onerror}->(node => $child_el,
6411     type => 'element not allowed:minus',
6412 wakaba 1.104 level => $self->{level}->{must});
6413 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6414     #
6415     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'li') {
6416     if ($element_state->{phase} eq 'li') {
6417     #
6418     } elsif ($element_state->{phase} eq 'li or phrasing') {
6419     $element_state->{phase} = 'li';
6420     } else {
6421 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
6422     level => $self->{level}->{must});
6423 wakaba 1.40 }
6424     } elsif ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
6425     if ($element_state->{phase} eq 'phrasing') {
6426     #
6427     } elsif ($element_state->{phase} eq 'li or phrasing') {
6428     $element_state->{phase} = 'phrasing';
6429     } else {
6430 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
6431     level => $self->{level}->{must});
6432 wakaba 1.40 }
6433     } else {
6434 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
6435     level => $self->{level}->{must});
6436 wakaba 1.40 }
6437     },
6438     check_child_text => sub {
6439     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6440     if ($has_significant) {
6441     if ($element_state->{phase} eq 'phrasing') {
6442     #
6443     } elsif ($element_state->{phase} eq 'li or phrasing') {
6444     $element_state->{phase} = 'phrasing';
6445     } else {
6446     $self->{onerror}->(node => $child_node,
6447 wakaba 1.104 type => 'character not allowed',
6448     level => $self->{level}->{must});
6449 wakaba 1.1 }
6450     }
6451 wakaba 1.40 },
6452     check_end => sub {
6453     my ($self, $item, $element_state) = @_;
6454     delete $self->{flag}->{in_menu} unless $element_state->{in_menu_original};
6455    
6456     if ($element_state->{phase} eq 'li') {
6457     $HTMLChecker{check_end}->(@_);
6458     } else { # 'phrasing' or 'li or phrasing'
6459     $HTMLPhrasingContentChecker{check_end}->(@_);
6460 wakaba 1.1 }
6461     },
6462 wakaba 1.8 };
6463    
6464     $Element->{$HTML_NS}->{datatemplate} = {
6465 wakaba 1.40 %HTMLChecker,
6466 wakaba 1.48 status => FEATURE_HTML5_AT_RISK,
6467 wakaba 1.40 check_child_element => sub {
6468     my ($self, $item, $child_el, $child_nsuri, $child_ln,
6469     $child_is_transparent, $element_state) = @_;
6470 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6471     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6472 wakaba 1.40 $self->{onerror}->(node => $child_el,
6473     type => 'element not allowed:minus',
6474 wakaba 1.104 level => $self->{level}->{must});
6475 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6476     #
6477     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'rule') {
6478     #
6479     } else {
6480     $self->{onerror}->(node => $child_el,
6481 wakaba 1.104 type => 'element not allowed:datatemplate',
6482     level => $self->{level}->{must});
6483 wakaba 1.40 }
6484     },
6485     check_child_text => sub {
6486     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6487     if ($has_significant) {
6488 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
6489     level => $self->{level}->{must});
6490 wakaba 1.8 }
6491     },
6492     is_xml_root => 1,
6493     };
6494    
6495     $Element->{$HTML_NS}->{rule} = {
6496 wakaba 1.40 %HTMLChecker,
6497 wakaba 1.48 status => FEATURE_HTML5_AT_RISK,
6498 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6499 wakaba 1.23 condition => $HTMLSelectorsAttrChecker,
6500 wakaba 1.92 mode => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->(),
6501 wakaba 1.50 }, {
6502     %HTMLAttrStatus,
6503     condition => FEATURE_HTML5_AT_RISK,
6504     mode => FEATURE_HTML5_AT_RISK,
6505 wakaba 1.8 }),
6506 wakaba 1.40 check_start => sub {
6507     my ($self, $item, $element_state) = @_;
6508 wakaba 1.79
6509 wakaba 1.40 $self->_add_plus_elements ($element_state, {$HTML_NS => {nest => 1}});
6510 wakaba 1.79 $element_state->{in_rule_original} = $self->{flag}->{in_rule};
6511     $self->{flag}->{in_rule} = 1;
6512    
6513     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6514     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6515 wakaba 1.40 },
6516     check_child_element => sub { },
6517     check_child_text => sub { },
6518     check_end => sub {
6519     my ($self, $item, $element_state) = @_;
6520 wakaba 1.79
6521 wakaba 1.40 $self->_remove_plus_elements ($element_state);
6522 wakaba 1.79 delete $self->{flag}->{in_rule} unless $element_state->{in_rule_original};
6523    
6524 wakaba 1.40 $HTMLChecker{check_end}->(@_);
6525 wakaba 1.8 },
6526     ## NOTE: "MAY be anything that, when the parent |datatemplate|
6527     ## is applied to some conforming data, results in a conforming DOM tree.":
6528     ## We don't check against this.
6529     };
6530    
6531     $Element->{$HTML_NS}->{nest} = {
6532 wakaba 1.40 %HTMLEmptyChecker,
6533 wakaba 1.48 status => FEATURE_HTML5_AT_RISK,
6534 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
6535 wakaba 1.23 filter => $HTMLSelectorsAttrChecker,
6536     mode => sub {
6537     my ($self, $attr) = @_;
6538     my $value = $attr->value;
6539 wakaba 1.132 if ($value !~ /\A[^\x09\x0A\x0C\x0D\x20]+\z/) {
6540 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'mode:syntax error',
6541     level => $self->{level}->{must});
6542 wakaba 1.23 }
6543     },
6544 wakaba 1.50 }, {
6545     %HTMLAttrStatus,
6546     filter => FEATURE_HTML5_AT_RISK,
6547     mode => FEATURE_HTML5_AT_RISK,
6548 wakaba 1.8 }),
6549 wakaba 1.1 };
6550    
6551     $Element->{$HTML_NS}->{legend} = {
6552 wakaba 1.40 %HTMLPhrasingContentChecker,
6553 wakaba 1.49 status => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6554 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6555 wakaba 1.66 accesskey => $HTMLAccesskeyAttrChecker,
6556 wakaba 1.52 # align => $GetHTMLEnumeratedAttrChecker->({
6557     # top => 1, bottom => 1, left => 1, right => 1,
6558     # }),
6559     }, {
6560 wakaba 1.49 %HTMLAttrStatus,
6561     %HTMLM12NCommonAttrStatus,
6562     accesskey => FEATURE_M12N10_REC,
6563     align => FEATURE_M12N10_REC_DEPRECATED,
6564 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6565 wakaba 1.49 }),
6566 wakaba 1.1 };
6567    
6568     $Element->{$HTML_NS}->{div} = {
6569 wakaba 1.72 %HTMLFlowContentChecker,
6570 wakaba 1.82 status => FEATURE_HTML5_DEFAULT | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
6571 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
6572     align => $GetHTMLEnumeratedAttrChecker->({
6573     left => 1, center => 1, right => 1, justify => 1,
6574     }),
6575     }, {
6576 wakaba 1.49 %HTMLAttrStatus,
6577 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
6578 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
6579     datafld => FEATURE_HTML4_REC_RESERVED,
6580     dataformatas => FEATURE_HTML4_REC_RESERVED,
6581     datasrc => FEATURE_HTML4_REC_RESERVED,
6582 wakaba 1.50 lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6583 wakaba 1.49 }),
6584 wakaba 1.66 check_start => sub {
6585     my ($self, $item, $element_state) = @_;
6586    
6587     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
6588 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6589     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6590 wakaba 1.66 },
6591 wakaba 1.1 };
6592    
6593 wakaba 1.64 $Element->{$HTML_NS}->{center} = {
6594 wakaba 1.72 %HTMLFlowContentChecker,
6595 wakaba 1.64 status => FEATURE_M12N10_REC_DEPRECATED,
6596     check_attrs => $GetHTMLAttrsChecker->({}, {
6597     %HTMLAttrStatus,
6598     %HTMLM12NCommonAttrStatus,
6599     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6600     }),
6601     };
6602    
6603 wakaba 1.1 $Element->{$HTML_NS}->{font} = {
6604 wakaba 1.40 %HTMLTransparentChecker,
6605 wakaba 1.78 status => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC_DEPRECATED,
6606 wakaba 1.70 check_attrs => $GetHTMLAttrsChecker->({
6607     ## TODO: HTML4 |size|, |color|, |face|
6608 wakaba 1.49 }, {
6609     %HTMLAttrStatus,
6610 wakaba 1.50 class => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6611 wakaba 1.49 color => FEATURE_M12N10_REC_DEPRECATED,
6612 wakaba 1.50 dir => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6613 wakaba 1.49 face => FEATURE_M12N10_REC_DEPRECATED,
6614 wakaba 1.50 id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6615     lang => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6616 wakaba 1.49 size => FEATURE_M12N10_REC_DEPRECATED,
6617 wakaba 1.78 style => FEATURE_HTML5_DEFAULT | FEATURE_XHTML10_REC,
6618 wakaba 1.50 title => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6619 wakaba 1.49 }),
6620 wakaba 1.78 ## NOTE: When the |font| element was defined in the HTML5 specification,
6621     ## it is allowed only in a document with the WYSIWYG signature. The
6622     ## checker does not check whether there is the signature, since the
6623     ## signature is dropped, too, and has never been implemented. (In addition,
6624     ## for any |font| element an "element not defined" error is raised anyway,
6625     ## such that we don't have to raise an additional error.)
6626 wakaba 1.1 };
6627 wakaba 1.49
6628 wakaba 1.64 $Element->{$HTML_NS}->{basefont} = {
6629     %HTMLEmptyChecker,
6630     status => FEATURE_M12N10_REC_DEPRECATED,
6631     check_attrs => $GetHTMLAttrsChecker->({
6632     ## TODO: color, face, size
6633     }, {
6634     %HTMLAttrStatus,
6635     color => FEATURE_M12N10_REC_DEPRECATED,
6636     face => FEATURE_M12N10_REC_DEPRECATED,
6637     #id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC_DEPRECATED,
6638     id => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6639     size => FEATURE_M12N10_REC_DEPRECATED,
6640     }),
6641     };
6642    
6643 wakaba 1.49 ## TODO: frameset FEATURE_M12N10_REC
6644     ## class title id cols rows onload onunload style(x10)
6645     ## frame frameborder longdesc marginheight marginwidth noresize scrolling src name(deprecated) class,id,title,style(x10)
6646     ## noframes Common, lang(xhtml10)
6647    
6648 wakaba 1.100 ## TODO: CR: rbc rtc @rbspan (M12NXHTML2Common)
6649 wakaba 1.56
6650 wakaba 1.61 ## TODO: xmp, listing, plaintext FEATURE_HTML32_REC_OBSOLETE
6651     ## TODO: ^^^ lang, dir, id, class [HTML 2.x] sdaform [HTML 2.0]
6652     ## xmp, listing sdapref[HTML2,0]
6653    
6654 wakaba 1.56 =pod
6655    
6656     WF2: Documents MUST comply to [CHARMOD].
6657     WF2: Vencor extensions MUST NOT be used.
6658    
6659 wakaba 1.61 HTML 2.0 nextid @n
6660    
6661     RFC 2659: CERTS CRYPTOPTS
6662    
6663     ISO-HTML: pre-html, divN
6664 wakaba 1.82
6665     XHTML2: blockcode (Common), h (Common), separator (Common), l (Common),
6666     di (Common), nl (Common), handler (Common, type), standby (Common),
6667     summary (Common)
6668    
6669 wakaba 1.97 Access & XHTML2: access (LC)
6670 wakaba 1.82
6671     XML Events & XForms (for XHTML2 support; very, very low priority)
6672 wakaba 1.61
6673 wakaba 1.56 =cut
6674 wakaba 1.61
6675     ## NOTE: Where RFC 2659 allows additional attributes is unclear.
6676     ## We added them only to |a|. |link| and |form| might also allow them
6677     ## in theory.
6678 wakaba 1.1
6679     $Whatpm::ContentChecker::Namespace->{$HTML_NS}->{loaded} = 1;
6680    
6681     1;

admin@suikawiki.org
ViewVC Help
Powered by ViewVC 1.1.24