/[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 1.213 - (hide annotations) (download)
Sun Aug 30 07:51:24 2009 UTC (15 years, 10 months ago) by wakaba
Branch: MAIN
Changes since 1.212: +26 -0 lines
++ whatpm/t/dom-conformance/ChangeLog	30 Aug 2009 07:47:50 -0000
	* html-texts-1.dat: Added test cases for nested |progress|,
	|time|, and |meter| elements (HTML5 revision 3692).

2009-08-30  Wakaba  <wakaba@suika.fam.cx>

++ whatpm/Whatpm/ContentChecker/ChangeLog	30 Aug 2009 07:48:15 -0000
	* HTML.pm: Disallow nested |progress|, |meter|, and |time|
	elements (HTML5 revision 3692).

2009-08-30  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 1.174 ## --- Feature Status ---
10    
11 wakaba 1.187 sub FEATURE_HTML5_REC () {
12 wakaba 1.154 ## NOTE: Part of HTML5, the implemented status.
13 wakaba 1.89 Whatpm::ContentChecker::FEATURE_STATUS_REC |
14     Whatpm::ContentChecker::FEATURE_ALLOWED
15 wakaba 1.187
16     ## Strictly speaking, HTML5's "implemented and widely deployed"
17     ## status does not necessarily satisfy the condition for
18     ## FEATURE_STATUS_REC, since there is no test cases for most of
19     ## features marked as "implemented" in HTML5. Nevertheless, we
20     ## special-case HTML5's this status as if that had passed the CR
21     ## phase, considering HTML's history.
22 wakaba 1.89 }
23 wakaba 1.187
24 wakaba 1.154 sub FEATURE_HTML5_CR () {
25     ## NOTE: Part of HTML5, the awaiting implementation feedback status.
26     Whatpm::ContentChecker::FEATURE_STATUS_CR |
27     Whatpm::ContentChecker::FEATURE_ALLOWED
28     }
29 wakaba 1.54 sub FEATURE_HTML5_LC () {
30 wakaba 1.154 ## NOTE: Part of HTML5, the last call of comments status.
31 wakaba 1.54 Whatpm::ContentChecker::FEATURE_STATUS_LC |
32     Whatpm::ContentChecker::FEATURE_ALLOWED
33     }
34     sub FEATURE_HTML5_AT_RISK () {
35 wakaba 1.154 ## NOTE: Part of HTML5, but in the being considered for removal
36     ## status.
37 wakaba 1.54 Whatpm::ContentChecker::FEATURE_STATUS_WD |
38     Whatpm::ContentChecker::FEATURE_ALLOWED
39     }
40     sub FEATURE_HTML5_WD () {
41 wakaba 1.154 ## NOTE: Part of HTML5, the working draft status.
42 wakaba 1.54 Whatpm::ContentChecker::FEATURE_STATUS_WD |
43     Whatpm::ContentChecker::FEATURE_ALLOWED
44     }
45     sub FEATURE_HTML5_FD () {
46 wakaba 1.154 ## NOTE: Part of HTML5, the first draft status.
47 wakaba 1.54 Whatpm::ContentChecker::FEATURE_STATUS_WD |
48     Whatpm::ContentChecker::FEATURE_ALLOWED
49     }
50     sub FEATURE_HTML5_DEFAULT () {
51 wakaba 1.154 ## NOTE: Part of HTML5, but not annotated.
52 wakaba 1.54 Whatpm::ContentChecker::FEATURE_STATUS_WD |
53     Whatpm::ContentChecker::FEATURE_ALLOWED
54 wakaba 1.49 }
55 wakaba 1.54 sub FEATURE_HTML5_DROPPED () {
56 wakaba 1.154 ## NOTE: Was part of HTML5, in a status before the last call of
57     ## comments, but then dropped.
58 wakaba 1.49 Whatpm::ContentChecker::FEATURE_STATUS_WD
59     }
60 wakaba 1.118 sub FEATURE_HTML5_LC_DROPPED () {
61 wakaba 1.154 ## NOTE: Was part of HTML5, in the last call of comments status, but
62     ## then dropped.
63 wakaba 1.118 Whatpm::ContentChecker::FEATURE_STATUS_LC
64     }
65 wakaba 1.154
66 wakaba 1.119 sub FEATURE_WF2X () {
67 wakaba 1.154 ## NOTE: Defined in WF2 (whether deprecated or not) and then
68     ## incorporated into the HTML5 spec.
69 wakaba 1.119 Whatpm::ContentChecker::FEATURE_STATUS_LC
70     }
71 wakaba 1.54 sub FEATURE_WF2 () {
72 wakaba 1.154 ## NOTE: Features introduced or modified in WF2, which were not
73     ## merged into HTML5.
74 wakaba 1.119 Whatpm::ContentChecker::FEATURE_STATUS_LC
75 wakaba 1.54 }
76 wakaba 1.126 sub FEATURE_WF2_INFORMATIVE () {
77 wakaba 1.154 ## NOTE: Features mentioned in WF2's informative appendix A, which
78     ## were not merged into HTML5.
79 wakaba 1.56 Whatpm::ContentChecker::FEATURE_STATUS_LC
80     }
81 wakaba 1.49
82 wakaba 1.154 sub FEATURE_RDFA_REC () {
83     Whatpm::ContentChecker::FEATURE_STATUS_REC
84 wakaba 1.121 }
85 wakaba 1.154 sub FEATURE_RDFA_LC_DROPPED () {
86     ## NOTE: The feature that was defined in a RDFa last call working
87     ## draft, but then dropped.
88 wakaba 1.61 Whatpm::ContentChecker::FEATURE_STATUS_LC
89     }
90 wakaba 1.58
91     ## NOTE: XHTML Role LCWD has almost no information on how the |role|
92     ## attribute can be used- the only requirements for that matter is:
93     ## "the attribute MUST be referenced using its namespace-qualified form" (and
94     ## this is a host language conformance!).
95 wakaba 1.82 sub FEATURE_ROLE_LC () {
96     Whatpm::ContentChecker::FEATURE_STATUS_LC
97     }
98    
99     sub FEATURE_XHTML2_ED () {
100 wakaba 1.154 ## NOTE: XHTML 2.0 Editor's Draft, in which the namespace URI is
101     ## "http://www.w3.org/1999/xhtml".
102 wakaba 1.82 Whatpm::ContentChecker::FEATURE_STATUS_WD
103     }
104 wakaba 1.58
105 wakaba 1.55 sub FEATURE_XHTMLBASIC11_CR () {
106 wakaba 1.154 ## NOTE: XHTML Basic 1.1 Recommendation, new features (not in XHTML
107     ## M12N).
108     Whatpm::ContentChecker::FEATURE_STATUS_REC
109 wakaba 1.55 }
110     sub FEATURE_XHTMLBASIC11_CR_DEPRECATED () {
111 wakaba 1.154 ## NOTE: XHTML Basic 1.1 Recommendation, new but deprecated
112     ## features.
113     Whatpm::ContentChecker::FEATURE_STATUS_REC |
114 wakaba 1.55 Whatpm::ContentChecker::FEATURE_DEPRECATED_INFO
115     }
116    
117 wakaba 1.154 sub FEATURE_RUBY_REC () {
118     Whatpm::ContentChecker::FEATURE_STATUS_CR
119 wakaba 1.82 }
120    
121 wakaba 1.154 sub FEATURE_M12N11_LC () {
122     ## NOTE: XHTML M12N 1.1 Recommendation, new features (not in 1.0).
123     Whatpm::ContentChecker::FEATURE_STATUS_REC;
124 wakaba 1.99 }
125    
126 wakaba 1.49 ## NOTE: M12N10 status is based on its abstract module definition.
127     ## It contains a number of problems. (However, again, it's a REC!)
128 wakaba 1.54 sub FEATURE_M12N10_REC () {
129 wakaba 1.154 ## NOTE: Oh, XHTML m12n 1.0 passed the CR phase! W3C Process sucks!
130 wakaba 1.54 Whatpm::ContentChecker::FEATURE_STATUS_REC
131     }
132     sub FEATURE_M12N10_REC_DEPRECATED () {
133     Whatpm::ContentChecker::FEATURE_STATUS_REC |
134     Whatpm::ContentChecker::FEATURE_DEPRECATED_INFO
135     }
136 wakaba 1.49
137     ## NOTE: XHTML10 status is based on its transitional and frameset DTDs
138     ## (second edition). Only missing attributes from M12N10 abstract
139     ## definition are added.
140 wakaba 1.54 sub FEATURE_XHTML10_REC () {
141     Whatpm::ContentChecker::FEATURE_STATUS_CR
142     }
143    
144 wakaba 1.61 ## NOTE: Diff from HTML4.
145     sub FEATURE_ISOHTML_PREPARATION () { ## Informative documentation
146     Whatpm::ContentChecker::FEATURE_STATUS_CR
147     }
148 wakaba 1.58
149 wakaba 1.49 ## NOTE: HTML4 status is based on its transitional and frameset DTDs (HTML
150     ## 4.01). Only missing attributes from XHTML10 are added.
151 wakaba 1.54 sub FEATURE_HTML4_REC_RESERVED () {
152     Whatpm::ContentChecker::FEATURE_STATUS_WD
153     }
154    
155     ## TODO: According to HTML4 definition, authors SHOULD use style sheets
156     ## rather than presentational attributes (deprecated or not deprecated).
157 wakaba 1.48
158 wakaba 1.61 ## NOTE: Diff from HTML4.
159     sub FEATURE_HTML32_REC_OBSOLETE () {
160     Whatpm::ContentChecker::FEATURE_STATUS_CR |
161     Whatpm::ContentChecker::FEATURE_DEPRECATED_SHOULD
162     ## NOTE: Lowercase normative "should".
163     }
164    
165     sub FEATURE_RFC2659 () { ## Experimental RFC
166     Whatpm::ContentChecker::FEATURE_STATUS_CR
167     }
168    
169     ## NOTE: HTML 2.x - diff from HTML 2.0 and not in newer versions.
170     sub FEATURE_HTML2X_RFC () { ## Proposed Standard, obsolete
171     Whatpm::ContentChecker::FEATURE_STATUS_CR
172     }
173    
174     ## NOTE: Diff from HTML 2.0.
175     sub FEATURE_RFC1942 () { ## Experimental RFC, obsolete
176     Whatpm::ContentChecker::FEATURE_STATUS_CR
177     }
178    
179     ## NOTE: Diff from HTML 3.2.
180     sub FEATURE_HTML20_RFC () { ## Proposed Standard, obsolete
181     Whatpm::ContentChecker::FEATURE_STATUS_CR
182     }
183 wakaba 1.58
184 wakaba 1.174 ## --- Content Model ---
185    
186 wakaba 1.29 ## December 2007 HTML5 Classification
187    
188     my $HTMLMetadataContent = {
189     $HTML_NS => {
190     title => 1, base => 1, link => 1, style => 1, script => 1, noscript => 1,
191 wakaba 1.118 'event-source' => 1, eventsource => 1,
192     command => 1, datatemplate => 1,
193 wakaba 1.29 ## NOTE: A |meta| with no |name| element is not allowed as
194     ## a metadata content other than |head| element.
195     meta => 1,
196     },
197     ## NOTE: RDF is mentioned in the HTML5 spec.
198     ## TODO: Other RDF elements?
199     q<http://www.w3.org/1999/02/22-rdf-syntax-ns#> => {RDF => 1},
200     };
201    
202 wakaba 1.72 my $HTMLFlowContent = {
203 wakaba 1.29 $HTML_NS => {
204     section => 1, nav => 1, article => 1, blockquote => 1, aside => 1,
205 wakaba 1.195 h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, hgroup => 1,
206     header => 1,
207 wakaba 1.29 footer => 1, address => 1, p => 1, hr => 1, dialog => 1, pre => 1,
208 wakaba 1.207 ol => 1, ul => 1, dl => 1, menu => 1, figure => 1, table => 1,
209 wakaba 1.119 form => 1, fieldset => 1,
210 wakaba 1.72 details => 1, ## ISSUE: "Flow element" in spec.
211     datagrid => 1, ## ISSUE: "Flow element" in spec.
212 wakaba 1.29 datatemplate => 1,
213     div => 1, ## ISSUE: No category in spec.
214     ## NOTE: |style| is only allowed if |scoped| attribute is specified.
215     ## Additionally, it must be before any other element or
216     ## non-inter-element-whitespace text node.
217     style => 1,
218    
219 wakaba 1.193 ## These phrasing content are also categorized as flow content.
220 wakaba 1.38 br => 1, q => 1, cite => 1, em => 1, strong => 1, small => 1, mark => 1,
221 wakaba 1.29 dfn => 1, abbr => 1, time => 1, progress => 1, meter => 1, code => 1,
222     var => 1, samp => 1, kbd => 1, sub => 1, sup => 1, span => 1, i => 1,
223 wakaba 1.99 b => 1, bdo => 1, ruby => 1,
224 wakaba 1.118 script => 1, noscript => 1, 'event-source' => 1, eventsource => 1,
225     command => 1, bb => 1,
226 wakaba 1.119 input => 1, button => 1, label => 1, select => 1, datalist => 1,
227 wakaba 1.194 textarea => 1, keygen => 1, output => 1,
228     datagrid => 1,
229 wakaba 1.29 ## NOTE: |area| is allowed only as a descendant of |map|.
230     area => 1,
231 wakaba 1.193
232     ## Flow/phrasing content whose content model is transparent.
233 wakaba 1.196 a => 1, ins => 1, del => 1, font => 1, map => 1,
234 wakaba 1.29
235 wakaba 1.193 ## These embeded content are also categorized as flow content.
236 wakaba 1.29 img => 1, iframe => 1, embed => 1, object => 1, video => 1, audio => 1,
237     canvas => 1,
238     },
239    
240 wakaba 1.193 ## These embedded content are also categorized as flow content.
241 wakaba 1.29 q<http://www.w3.org/1998/Math/MathML> => {math => 1},
242     q<http://www.w3.org/2000/svg> => {svg => 1},
243 wakaba 1.193
244     ## And, non-inter-element-whitespace text nodes.
245     }; # $HTMLFlowContent
246 wakaba 1.29
247 wakaba 1.58 my $HTMLSectioningContent = {
248 wakaba 1.57 $HTML_NS => {
249     section => 1, nav => 1, article => 1, aside => 1,
250     },
251 wakaba 1.205 }; # $HTMLSectioningContent
252 wakaba 1.57
253 wakaba 1.58 my $HTMLSectioningRoot = {
254 wakaba 1.29 $HTML_NS => {
255 wakaba 1.58 blockquote => 1, datagrid => 1, figure => 1, td => 1,
256 wakaba 1.29 },
257     };
258    
259     my $HTMLHeadingContent = {
260     $HTML_NS => {
261 wakaba 1.195 h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, hgroup => 1,
262 wakaba 1.29 },
263     };
264    
265     my $HTMLPhrasingContent = {
266 wakaba 1.72 ## NOTE: All phrasing content is also flow content.
267 wakaba 1.29 $HTML_NS => {
268 wakaba 1.38 br => 1, q => 1, cite => 1, em => 1, strong => 1, small => 1, mark => 1,
269 wakaba 1.29 dfn => 1, abbr => 1, time => 1, progress => 1, meter => 1, code => 1,
270     var => 1, samp => 1, kbd => 1, sub => 1, sup => 1, span => 1, i => 1,
271 wakaba 1.99 b => 1, bdo => 1, ruby => 1,
272 wakaba 1.118 script => 1, noscript => 1, 'event-source' => 1, eventsource => 1,
273     command => 1, bb => 1,
274 wakaba 1.119 input => 1, button => 1, label => 1, select => 1, datalist => 1,
275 wakaba 1.194 textarea => 1, keygen => 1, output => 1,
276     datagrid => 1,
277 wakaba 1.29 ## NOTE: |area| is allowed only as a descendant of |map|.
278     area => 1,
279    
280     ## NOTE: Transparent.
281 wakaba 1.196 a => 1, ins => 1, del => 1, font => 1, map => 1,
282 wakaba 1.29
283 wakaba 1.193 ## These embedded content is also categorized as phrasing content.
284 wakaba 1.29 img => 1, iframe => 1, embed => 1, object => 1, video => 1, audio => 1,
285     canvas => 1,
286     },
287    
288 wakaba 1.193 ## These embedded content is also categorized as phrasing content.
289 wakaba 1.29 q<http://www.w3.org/1998/Math/MathML> => {math => 1},
290     q<http://www.w3.org/2000/svg> => {svg => 1},
291    
292 wakaba 1.193 ## And, non-inter-element-whitespace text nodes.
293     }; # $HTMLPhrasingContent
294 wakaba 1.29
295 wakaba 1.40 ## $HTMLEmbeddedContent: See Whatpm::ContentChecker.
296 wakaba 1.29
297     my $HTMLInteractiveContent = {
298     $HTML_NS => {
299     a => 1,
300 wakaba 1.202 label => 1, button => 1, select => 1, textarea => 1,
301     keygen => 1, details => 1,
302     datagrid => 1, bb => 1, ## dropped
303     iframe => 1, embed => 1,
304    
305     ## NOTE: When the |usemap| attribute is specified.
306     img => 1, object => 1,
307    
308     ## NOTE: When "type=hidden" attribute is not specified.
309     input => 1,
310 wakaba 1.130
311     ## NOTE: When "controls" attribute is specified.
312     video => 1, audio => 1,
313    
314     ## NOTE: When "type=toolbar" attribute is specified.
315     menu => 1,
316 wakaba 1.29 },
317 wakaba 1.202 }; # $HTMLInteractiveContent
318 wakaba 1.29
319 wakaba 1.139 ## NOTE: Labelable form-associated element.
320     my $LabelableFAE = {
321     $HTML_NS => {
322 wakaba 1.194 input => 1, button => 1, select => 1, textarea => 1, keygen => 1,
323 wakaba 1.139 },
324     };
325    
326 wakaba 1.192 ## Check whether the labelable form-associated element is allowed to
327     ## place there or not and mark the element ID, if any, might be used
328     ## in the |for| attribute of a |label| element.
329     my $FAECheckStart = sub {
330     my ($self, $item, $element_state) = @_;
331    
332     $element_state->{id_type} = 'labelable';
333     }; # $FAECheckStart
334     my $FAECheckAttrs2 = sub {
335     my ($self, $item, $element_state) = @_;
336    
337     ## This must be done in "check_attrs2" phase since it requires the
338     ## |id| attribute of the element, if any, reflected to the
339     ## |$self->{id}| hash.
340    
341     CHK: {
342     if ($self->{flag}->{has_label} and $self->{flag}->{has_labelable}) {
343     my $for = $self->{flag}->{label_for};
344     if (defined $for) {
345     my $id_attrs = $self->{id}->{$for};
346     if ($id_attrs and $id_attrs->[0]) {
347     my $el = $id_attrs->[0]->owner_element;
348     if ($el and $el eq $item->{node}) {
349     ## Even if there is an ancestor |label| element with its
350     ## |for| attribute specified, the attribute value
351     ## identifies THIS element, then there is no problem.
352     last CHK;
353     }
354     }
355     }
356    
357     $self->{onerror}->(node => $item->{node},
358     type => 'multiple labelable fae',
359     level => $self->{level}->{must});
360     } else {
361     $self->{flag}->{has_labelable} = 2;
362     }
363     } # CHK
364     }; # $FAECheckAttrs2
365    
366 wakaba 1.130 our $IsInHTMLInteractiveContent; # See Whatpm::ContentChecker.
367    
368 wakaba 1.36 ## NOTE: $HTMLTransparentElements: See Whatpm::ContentChecker.
369     ## NOTE: Semi-transparent elements: See Whatpm::ContentChecker.
370    
371     ## -- Common attribute syntacx checkers
372    
373 wakaba 1.1 our $AttrChecker;
374 wakaba 1.82 our $AttrStatus;
375 wakaba 1.1
376     my $GetHTMLEnumeratedAttrChecker = sub {
377     my $states = shift; # {value => conforming ? 1 : -1}
378     return sub {
379     my ($self, $attr) = @_;
380     my $value = lc $attr->value; ## TODO: ASCII case insensitibility?
381     if ($states->{$value} > 0) {
382     #
383     } elsif ($states->{$value}) {
384 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'enumerated:non-conforming',
385     level => $self->{level}->{must});
386 wakaba 1.1 } else {
387 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'enumerated:invalid',
388     level => $self->{level}->{must});
389 wakaba 1.1 }
390     };
391     }; # $GetHTMLEnumeratedAttrChecker
392    
393     my $GetHTMLBooleanAttrChecker = sub {
394     my $local_name = shift;
395     return sub {
396     my ($self, $attr) = @_;
397 wakaba 1.88 my $value = lc $attr->value; ## TODO: case
398 wakaba 1.1 unless ($value eq $local_name or $value eq '') {
399 wakaba 1.88 $self->{onerror}->(node => $attr, type => 'boolean:invalid',
400 wakaba 1.104 level => $self->{level}->{must});
401 wakaba 1.1 }
402     };
403     }; # $GetHTMLBooleanAttrChecker
404    
405 wakaba 1.8 ## Unordered set of space-separated tokens
406 wakaba 1.92 my $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker = sub {
407     my $allowed_words = shift;
408     return sub {
409     my ($self, $attr) = @_;
410     my %word;
411 wakaba 1.132 for my $word (grep {length $_}
412     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
413 wakaba 1.92 unless ($word{$word}) {
414     $word{$word} = 1;
415     if (not defined $allowed_words or
416     $allowed_words->{$word}) {
417     #
418     } else {
419 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'word not allowed',
420 wakaba 1.92 value => $word,
421 wakaba 1.104 level => $self->{level}->{must});
422 wakaba 1.92 }
423     } else {
424 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate token',
425     value => $word,
426     level => $self->{level}->{must});
427 wakaba 1.92 }
428 wakaba 1.8 }
429 wakaba 1.92 };
430     }; # $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker
431 wakaba 1.8
432 wakaba 1.132 ## |rel| attribute (set of space separated tokens,
433 wakaba 1.1 ## whose allowed values are defined by the section on link types)
434     my $HTMLLinkTypesAttrChecker = sub {
435 wakaba 1.66 my ($a_or_area, $todo, $self, $attr, $item, $element_state) = @_;
436 wakaba 1.1 my %word;
437 wakaba 1.132 for my $word (grep {length $_}
438     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
439 wakaba 1.1 unless ($word{$word}) {
440     $word{$word} = 1;
441 wakaba 1.18 } elsif ($word eq 'up') {
442     #
443 wakaba 1.1 } else {
444 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate token',
445     value => $word,
446     level => $self->{level}->{must});
447 wakaba 1.1 }
448     }
449     ## NOTE: Case sensitive match (since HTML5 spec does not say link
450     ## types are case-insensitive and it says "The value should not
451     ## be confusingly similar to any other defined value (e.g.
452     ## differing only in case).").
453     ## NOTE: Though there is no explicit "MUST NOT" for undefined values,
454     ## "MAY"s and "only ... MAY" restrict non-standard non-registered
455     ## values to be used conformingly.
456 wakaba 1.66
457     my $is_hyperlink;
458     my $is_resource;
459 wakaba 1.1 require Whatpm::_LinkTypeList;
460     our $LinkType;
461     for my $word (keys %word) {
462     my $def = $LinkType->{$word};
463     if (defined $def) {
464     if ($def->{status} eq 'accepted') {
465     if (defined $def->{effect}->[$a_or_area]) {
466     #
467     } else {
468     $self->{onerror}->(node => $attr,
469 wakaba 1.104 type => 'link type:bad context',
470     value => $word,
471 wakaba 1.110 level => $self->{level}->{must});
472 wakaba 1.1 }
473     } elsif ($def->{status} eq 'proposal') {
474 wakaba 1.104 $self->{onerror}->(node => $attr,
475     type => 'link type:proposed',
476     value => $word,
477     level => $self->{level}->{should});
478 wakaba 1.20 if (defined $def->{effect}->[$a_or_area]) {
479     #
480     } else {
481     $self->{onerror}->(node => $attr,
482 wakaba 1.104 type => 'link type:bad context',
483     value => $word,
484     level => $self->{level}->{must});
485 wakaba 1.20 }
486 wakaba 1.1 } else { # rejected or synonym
487     $self->{onerror}->(node => $attr,
488 wakaba 1.104 type => 'link type:non-conforming',
489     value => $word,
490     level => $self->{level}->{must});
491 wakaba 1.1 }
492 wakaba 1.4 if (defined $def->{effect}->[$a_or_area]) {
493     if ($word eq 'alternate') {
494     #
495     } elsif ($def->{effect}->[$a_or_area] eq 'hyperlink') {
496 wakaba 1.66 $is_hyperlink = 1;
497 wakaba 1.4 }
498     }
499 wakaba 1.1 if ($def->{unique}) {
500     unless ($self->{has_link_type}->{$word}) {
501     $self->{has_link_type}->{$word} = 1;
502     } else {
503     $self->{onerror}->(node => $attr,
504 wakaba 1.104 type => 'link type:duplicate',
505     value => $word,
506     level => $self->{level}->{must});
507 wakaba 1.1 }
508     }
509 wakaba 1.66
510     if (defined $def->{effect}->[$a_or_area] and $word ne 'alternate') {
511     $is_hyperlink = 1 if $def->{effect}->[$a_or_area] eq 'hyperlink';
512     $is_resource = 1 if $def->{effect}->[$a_or_area] eq 'external resource';
513     }
514 wakaba 1.1 } else {
515 wakaba 1.104 $self->{onerror}->(node => $attr,
516     type => 'unknown link type',
517     value => $word,
518     level => $self->{level}->{uncertain});
519 wakaba 1.1 }
520     }
521 wakaba 1.66 $is_hyperlink = 1 if $word{alternate} and not $word{stylesheet};
522 wakaba 1.1 ## TODO: The Pingback 1.0 specification, which is referenced by HTML5,
523     ## says that using both X-Pingback: header field and HTML
524     ## <link rel=pingback> is deprecated and if both appears they
525     ## SHOULD contain exactly the same value.
526     ## ISSUE: Pingback 1.0 specification defines the exact representation
527     ## of its link element, which cannot be tested by the current arch.
528     ## ISSUE: Pingback 1.0 specification says that the document MUST NOT
529     ## include any string that matches to the pattern for the rel=pingback link,
530     ## which again inpossible to test.
531     ## ISSUE: rel=pingback href MUST NOT include entities other than predefined 4.
532 wakaba 1.12
533     ## NOTE: <link rel="up index"><link rel="up up index"> is not an error.
534 wakaba 1.17 ## NOTE: We can't check "If the page is part of multiple hierarchies,
535     ## then they SHOULD be described in different paragraphs.".
536 wakaba 1.66
537     $todo->{has_hyperlink_link_type} = 1 if $is_hyperlink;
538     if ($is_hyperlink or $a_or_area) {
539     $element_state->{uri_info}->{href}->{type}->{hyperlink} = 1;
540     }
541     if ($is_resource and not $a_or_area) {
542     $element_state->{uri_info}->{href}->{type}->{resource} = 1;
543     }
544 wakaba 1.96
545     $element_state->{link_rel} = \%word;
546 wakaba 1.1 }; # $HTMLLinkTypesAttrChecker
547 wakaba 1.20
548     ## 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."
549 wakaba 1.1
550     ## URI (or IRI)
551     my $HTMLURIAttrChecker = sub {
552 wakaba 1.66 my ($self, $attr, $item, $element_state) = @_;
553 wakaba 1.1 ## ISSUE: Relative references are allowed? (RFC 3987 "IRI" is an absolute reference with optional fragment identifier.)
554     my $value = $attr->value;
555     Whatpm::URIChecker->check_iri_reference ($value, sub {
556 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
557 wakaba 1.106 }), $self->{level};
558 wakaba 1.17 $self->{has_uri_attr} = 1; ## TODO: <html manifest>
559 wakaba 1.66
560     my $attr_name = $attr->name;
561     $element_state->{uri_info}->{$attr_name}->{node} = $attr;
562     ## TODO: absolute
563     push @{$self->{return}->{uri}->{$value} ||= []},
564     $element_state->{uri_info}->{$attr_name};
565 wakaba 1.1 }; # $HTMLURIAttrChecker
566    
567     ## A space separated list of one or more URIs (or IRIs)
568     my $HTMLSpaceURIsAttrChecker = sub {
569     my ($self, $attr) = @_;
570 wakaba 1.66
571     my $type = {ping => 'action',
572     profile => 'namespace',
573     archive => 'resource'}->{$attr->name};
574    
575 wakaba 1.1 my $i = 0;
576 wakaba 1.132 for my $value (split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
577 wakaba 1.1 Whatpm::URIChecker->check_iri_reference ($value, sub {
578 wakaba 1.104 $self->{onerror}->(value => $value, @_, node => $attr, index => $i);
579 wakaba 1.106 }, $self->{level});
580 wakaba 1.66
581     ## TODO: absolute
582     push @{$self->{return}->{uri}->{$value} ||= []},
583 wakaba 1.67 {node => $attr, type => {$type => 1}};
584 wakaba 1.66
585 wakaba 1.1 $i++;
586     }
587 wakaba 1.67 ## ISSUE: Relative references? (especially, in profile="")
588 wakaba 1.1 ## ISSUE: Leading or trailing white spaces are conformant?
589     ## ISSUE: A sequence of white space characters are conformant?
590     ## ISSUE: A zero-length string is conformant? (It does contain a relative reference, i.e. same as base URI.)
591 wakaba 1.132 ## ISSUE: What is "space"?
592 wakaba 1.1 ## NOTE: Duplication seems not an error.
593 wakaba 1.4 $self->{has_uri_attr} = 1;
594 wakaba 1.1 }; # $HTMLSpaceURIsAttrChecker
595    
596 wakaba 1.156 my $ValidEmailAddress;
597     {
598     my $atext = qr[[A-Za-z0-9!#\$%&'*+/=?^_`{|}~-]];
599 wakaba 1.200 my $dot_atom = qr/$atext+(?>\.$atext+)*/o;
600     $ValidEmailAddress = qr/$dot_atom\@$dot_atom/o;
601 wakaba 1.156 }
602    
603 wakaba 1.168 ## Valid global date and time.
604     my $GetDateTimeAttrChecker = sub ($) {
605     my $type = shift;
606     return sub {
607     my ($self, $attr, $item, $element_state) = @_;
608    
609     my $range_error;
610    
611     require Message::Date;
612     my $dp = Message::Date->new;
613     $dp->{level} = $self->{level};
614     $dp->{onerror} = sub {
615     my %opt = @_;
616     unless ($opt{type} eq 'date value not supported') {
617     $self->{onerror}->(%opt, node => $attr);
618     $range_error = '';
619     }
620     };
621    
622     my $method = 'parse_' . $type;
623     my $d = $dp->$method ($attr->value);
624     $element_state->{date_value}->{$attr->name} = $d || $range_error;
625     };
626     }; # $GetDateTimeAttrChecker
627 wakaba 1.1
628     my $HTMLIntegerAttrChecker = sub {
629     my ($self, $attr) = @_;
630     my $value = $attr->value;
631     unless ($value =~ /\A-?[0-9]+\z/) {
632 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'integer:syntax error',
633     level => $self->{level}->{must});
634 wakaba 1.1 }
635     }; # $HTMLIntegerAttrChecker
636    
637     my $GetHTMLNonNegativeIntegerAttrChecker = sub {
638     my $range_check = shift;
639     return sub {
640     my ($self, $attr) = @_;
641     my $value = $attr->value;
642     if ($value =~ /\A[0-9]+\z/) {
643     unless ($range_check->($value + 0)) {
644 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'nninteger:out of range',
645     level => $self->{level}->{must});
646 wakaba 1.1 }
647     } else {
648     $self->{onerror}->(node => $attr,
649 wakaba 1.104 type => 'nninteger:syntax error',
650     level => $self->{level}->{must});
651 wakaba 1.1 }
652     };
653     }; # $GetHTMLNonNegativeIntegerAttrChecker
654    
655 wakaba 1.203 ## "Valid floating point number".
656 wakaba 1.1 my $GetHTMLFloatingPointNumberAttrChecker = sub {
657     my $range_check = shift;
658     return sub {
659 wakaba 1.168 my ($self, $attr, $item, $element_state) = @_;
660 wakaba 1.1 my $value = $attr->value;
661 wakaba 1.203 if ($value =~ /
662     \A
663     (-?) # $1
664     ([0-9]+) # $2
665     (?>(\.[0-9]+))? # $3
666     (?>[Ee] ([+-]?[0-9]+) )? # $4
667     \z
668     /x) {
669     my $num = (defined $3 ? $2 . $3 : $2) + 0;
670     $num = -$num if $1;
671     $num *= 10 ** ($4 + 0) if $4; # $4 can be "-0", but no problem.
672     if ($range_check->($num)) {
673     $element_state->{number_value}->{$attr->name} = $num;
674 wakaba 1.168 } else {
675 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'float:out of range',
676     level => $self->{level}->{must});
677 wakaba 1.1 }
678     } else {
679     $self->{onerror}->(node => $attr,
680 wakaba 1.104 type => 'float:syntax error',
681     level => $self->{level}->{must});
682 wakaba 1.1 }
683     };
684 wakaba 1.203 }; # $GetHTMLFloatingPointNumberAttrChecker
685 wakaba 1.144
686 wakaba 1.203 my $PositiveFloatingPointNumberAttrChecker
687     = $GetHTMLFloatingPointNumberAttrChecker->(sub { $_[0] > 0 });
688 wakaba 1.1
689 wakaba 1.148 my $StepAttrChecker = sub {
690     ## NOTE: A valid floating point number (> 0), or ASCII
691     ## case-insensitive "any".
692    
693     my ($self, $attr) = @_;
694     my $value = $attr->value;
695 wakaba 1.203 if ($value =~ /\A[Aa][Nn][Yy]\z/) {
696 wakaba 1.148 #
697     } else {
698 wakaba 1.203 $PositiveFloatingPointNumberAttrChecker->(@_);
699 wakaba 1.148 }
700     }; # $StepAttrChecker
701    
702 wakaba 1.86 ## HTML4 %Length;
703     my $HTMLLengthAttrChecker = sub {
704     my ($self, $attr) = @_;
705     my $value = $attr->value;
706     unless ($value =~ /\A[0-9]+%?\z/) {
707     $self->{onerror}->(node => $attr, type => 'length:syntax error',
708 wakaba 1.104 level => $self->{level}->{must});
709 wakaba 1.86 }
710    
711     ## NOTE: HTML4 definition is too vague - it does not define the syntax
712     ## of percentage value at all (!).
713     }; # $HTMLLengthAttrChecker
714    
715 wakaba 1.161 my $MIMEToken = qr/[\x21\x23-\x27\x2A\x2B\x2D\x2E\x30-\x39\x41-\x5A\x5E-\x7E]+/;
716     my $TypeOrSubtype = qr/[A-Za-z0-9!#\$&.+^_-]{1,127}/; # RFC 4288
717     my $IMTNoParameter = qr[($TypeOrSubtype)/($TypeOrSubtype)];
718    
719 wakaba 1.1 ## "A valid MIME type, optionally with parameters. [RFC 2046]"
720     ## ISSUE: RFC 2046 does not define syntax of media types.
721     ## ISSUE: The definition of "a valid MIME type" is unknown.
722     ## Syntactical correctness?
723     my $HTMLIMTAttrChecker = sub {
724     my ($self, $attr) = @_;
725     my $value = $attr->value;
726     ## ISSUE: RFC 2045 Content-Type header field allows insertion
727     ## of LWS/comments between tokens. Is it allowed in HTML? Maybe no.
728     ## ISSUE: RFC 2231 extension? Maybe no.
729     my $lws0 = qr/(?>(?>\x0D\x0A)?[\x09\x20])*/;
730     my $qs = qr/"(?>[\x00-\x0C\x0E-\x21\x23-\x5B\x5D-\x7E]|\x0D\x0A[\x09\x20]|\x5C[\x00-\x7F])*"/;
731 wakaba 1.161 if ($value =~ m#\A$lws0($MIMEToken)$lws0/$lws0($MIMEToken)$lws0((?>;$lws0$MIMEToken$lws0=$lws0(?>$MIMEToken|$qs)$lws0)*)\z#) {
732 wakaba 1.1 my @type = ($1, $2);
733     my $param = $3;
734 wakaba 1.161 while ($param =~ s/^;$lws0($MIMEToken)$lws0=$lws0(?>($MIMEToken)|($qs))$lws0//) {
735 wakaba 1.1 if (defined $2) {
736     push @type, $1 => $2;
737     } else {
738     my $n = $1;
739 wakaba 1.152 my $v = $3;
740 wakaba 1.1 $v =~ s/\\(.)/$1/gs;
741 wakaba 1.152 push @type, $n => substr ($v, 1, length ($v) - 2);
742 wakaba 1.1 }
743     }
744     require Whatpm::IMTChecker;
745 wakaba 1.109 my $ic = Whatpm::IMTChecker->new;
746     $ic->{level} = $self->{level};
747     $ic->check_imt (sub {
748 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
749 wakaba 1.1 }, @type);
750     } else {
751 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'IMT:syntax error',
752     level => $self->{level}->{must});
753 wakaba 1.1 }
754     }; # $HTMLIMTAttrChecker
755    
756     my $HTMLLanguageTagAttrChecker = sub {
757 wakaba 1.7 ## NOTE: See also $AtomLanguageTagAttrChecker in Atom.pm.
758    
759 wakaba 1.1 my ($self, $attr) = @_;
760 wakaba 1.6 my $value = $attr->value;
761     require Whatpm::LangTag;
762     Whatpm::LangTag->check_rfc3066_language_tag ($value, sub {
763 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
764 wakaba 1.106 }, $self->{level});
765 wakaba 1.1 ## ISSUE: RFC 4646 (3066bis)?
766 wakaba 1.6
767     ## TODO: testdata
768 wakaba 1.1 }; # $HTMLLanguageTagAttrChecker
769    
770     ## "A valid media query [MQ]"
771     my $HTMLMQAttrChecker = sub {
772     my ($self, $attr) = @_;
773 wakaba 1.104 $self->{onerror}->(node => $attr,
774     type => 'media query',
775     level => $self->{level}->{uncertain});
776 wakaba 1.1 ## ISSUE: What is "a valid media query"?
777     }; # $HTMLMQAttrChecker
778    
779     my $HTMLEventHandlerAttrChecker = sub {
780     my ($self, $attr) = @_;
781 wakaba 1.104 $self->{onerror}->(node => $attr,
782     type => 'event handler',
783     level => $self->{level}->{uncertain});
784 wakaba 1.1 ## TODO: MUST contain valid ECMAScript code matching the
785     ## ECMAScript |FunctionBody| production. [ECMA262]
786     ## ISSUE: MUST be ES3? E4X? ES4? JS1.x?
787     ## ISSUE: Automatic semicolon insertion does not apply?
788     ## ISSUE: Other script languages?
789     }; # $HTMLEventHandlerAttrChecker
790    
791 wakaba 1.136 my $HTMLFormAttrChecker = sub {
792     my ($self, $attr) = @_;
793    
794     ## NOTE: MUST be the ID of a |form| element.
795    
796     my $value = $attr->value;
797 wakaba 1.138 push @{$self->{idref}}, ['form', $value => $attr];
798 wakaba 1.136
799     ## ISSUE: <form id=""><input form=""> (empty ID)?
800     }; # $HTMLFormAttrChecker
801    
802 wakaba 1.158 my $ListAttrChecker = sub {
803     my ($self, $attr) = @_;
804    
805     ## NOTE: MUST be the ID of a |datalist| element.
806    
807     push @{$self->{idref}}, ['datalist', $attr->value, $attr];
808    
809     ## TODO: Warn violation to control-dependent restrictions. For
810     ## example, |<input type=url maxlength=10 list=a> <datalist
811     ## id=a><option value=nonurlandtoolong></datalist>| should be
812     ## warned.
813     }; # $ListAttrChecker
814    
815 wakaba 1.160 my $PatternAttrChecker = sub {
816     my ($self, $attr) = @_;
817     $self->{onsubdoc}->({s => $attr->value,
818     container_node => $attr,
819     media_type => 'text/x-regexp-js',
820     is_char_string => 1});
821 wakaba 1.161
822     ## ISSUE: "value must match the Pattern production of ECMA 262's
823     ## grammar" - no additional constraints (e.g. {n,m} then n>=m).
824    
825     ## TODO: Warn if @value does not match @pattern.
826 wakaba 1.160 }; # $PatternAttrChecker
827    
828 wakaba 1.161 my $AcceptAttrChecker = sub {
829     my ($self, $attr) = @_;
830    
831     my $value = $attr->value;
832     $value =~ tr/A-Z/a-z/; ## ASCII case-insensitive
833 wakaba 1.200
834     ## A set of comma-separated tokens.
835 wakaba 1.161 my @value = length $value ? split /,/, $value, -1 : ('');
836 wakaba 1.200
837 wakaba 1.161 my %has_value;
838     for my $v (@value) {
839 wakaba 1.200 $v =~ s/^[\x09\x0A\x0C\x0D\x20]+//;
840     $v =~ s/[\x09\x0A\x0C\x0D\x20]+\z//;
841    
842 wakaba 1.161 if ($has_value{$v}) {
843     $self->{onerror}->(node => $attr,
844     type => 'duplicate token',
845     value => $v,
846     level => $self->{level}->{must});
847     next;
848     }
849     $has_value{$v} = 1;
850    
851     if ($v eq 'audio/*' or $v eq 'video/*' or $v eq 'image/*') {
852     #
853     } elsif ($v =~ m[\A$IMTNoParameter\z]) {
854     ## ISSUE: HTML5 references RFC 2046, but maybe HTML5 should
855     ## define its own syntax citing RFC 4288.
856    
857     ## NOTE: Parameters not allowed.
858     require Whatpm::IMTChecker;
859     my $ic = Whatpm::IMTChecker->new;
860     $ic->{level} = $self->{level};
861     $ic->check_imt (sub {
862     $self->{onerror}->(@_, node => $attr);
863     }, $1, $2);
864     } else {
865     $self->{onerror}->(node => $attr,
866     type => 'IMTnp:syntax error', ## TODOC: type
867     value => $v,
868     level => $self->{level}->{must});
869     }
870     }
871     }; # $AcceptAttrChecker
872    
873 wakaba 1.165 my $FormControlNameAttrChecker = sub {
874     my ($self, $attr) = @_;
875    
876     unless (length $attr->value) {
877     $self->{onerror}->(node => $attr,
878     type => 'empty control name', ## TODOC: type
879     level => $self->{level}->{must});
880     }
881    
882     ## NOTE: No uniqueness constraint.
883     }; # $FormControlNameAttrChecker
884    
885     my $AutofocusAttrChecker = sub {
886     my ($self, $attr) = @_;
887    
888     $GetHTMLBooleanAttrChecker->('autofocus')->(@_);
889    
890     if ($self->{has_autofocus}) {
891     $self->{onerror}->(node => $attr,
892     type => 'duplicate autofocus', ## TODOC: type
893     level => $self->{level}->{must});
894     }
895     $self->{has_autofocus} = 1;
896     }; # $AutofocusAttrChekcer
897    
898 wakaba 1.1 my $HTMLUsemapAttrChecker = sub {
899     my ($self, $attr) = @_;
900 wakaba 1.100 ## MUST be a valid hash-name reference to a |map| element.
901 wakaba 1.1 my $value = $attr->value;
902     if ($value =~ s/^#//) {
903 wakaba 1.100 ## NOTE: |usemap="#"| is conforming, though it identifies no |map| element
904     ## according to the "rules for parsing a hash-name reference" algorithm.
905     ## The document is non-conforming anyway, since |<map name="">| (empty
906     ## name) is non-conforming.
907 wakaba 1.1 push @{$self->{usemap}}, [$value => $attr];
908     } else {
909 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'hashref:syntax error',
910     level => $self->{level}->{must});
911 wakaba 1.1 }
912 wakaba 1.100 ## NOTE: Space characters in hash-name references are conforming.
913 wakaba 1.1 ## ISSUE: UA algorithm for matching is case-insensitive; IDs only different in cases should be reported
914     }; # $HTMLUsemapAttrChecker
915    
916 wakaba 1.76 ## Valid browsing context name
917     my $HTMLBrowsingContextNameAttrChecker = sub {
918     my ($self, $attr) = @_;
919     my $value = $attr->value;
920     if ($value =~ /^_/) {
921     $self->{onerror}->(node => $attr, type => 'window name:reserved',
922 wakaba 1.104 level => $self->{level}->{must},
923 wakaba 1.76 value => $value);
924     } elsif (length $value) {
925     #
926     } else {
927     $self->{onerror}->(node => $attr, type => 'window name:empty',
928 wakaba 1.104 level => $self->{level}->{must});
929 wakaba 1.76 }
930     }; # $HTMLBrowsingContextNameAttrChecker
931    
932     ## Valid browsing context name or keyword
933 wakaba 1.1 my $HTMLTargetAttrChecker = sub {
934     my ($self, $attr) = @_;
935     my $value = $attr->value;
936     if ($value =~ /^_/) {
937     $value = lc $value; ## ISSUE: ASCII case-insentitive?
938     unless ({
939 wakaba 1.76 _blank => 1,_self => 1, _parent => 1, _top => 1,
940 wakaba 1.1 }->{$value}) {
941     $self->{onerror}->(node => $attr,
942 wakaba 1.76 type => 'window name:reserved',
943 wakaba 1.104 level => $self->{level}->{must},
944 wakaba 1.76 value => $value);
945 wakaba 1.1 }
946 wakaba 1.76 } elsif (length $value) {
947     #
948 wakaba 1.1 } else {
949 wakaba 1.76 $self->{onerror}->(node => $attr, type => 'window name:empty',
950 wakaba 1.104 level => $self->{level}->{must});
951 wakaba 1.1 }
952     }; # $HTMLTargetAttrChecker
953    
954 wakaba 1.23 my $HTMLSelectorsAttrChecker = sub {
955     my ($self, $attr) = @_;
956    
957     ## ISSUE: Namespace resolution?
958    
959     my $value = $attr->value;
960    
961     require Whatpm::CSS::SelectorsParser;
962     my $p = Whatpm::CSS::SelectorsParser->new;
963     $p->{pseudo_class}->{$_} = 1 for qw/
964     active checked disabled empty enabled first-child first-of-type
965     focus hover indeterminate last-child last-of-type link only-child
966     only-of-type root target visited
967     lang nth-child nth-last-child nth-of-type nth-last-of-type not
968     -manakai-contains -manakai-current
969     /;
970    
971     $p->{pseudo_element}->{$_} = 1 for qw/
972     after before first-letter first-line
973     /;
974    
975 wakaba 1.104 $p->{level} = $self->{level};
976 wakaba 1.23 $p->{onerror} = sub {
977 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
978 wakaba 1.23 };
979     $p->parse_string ($value);
980     }; # $HTMLSelectorsAttrChecker
981    
982 wakaba 1.129 my $HTMLCharsetChecker = sub ($$$;$) {
983     my ($charset_value, $self, $attr, $ascii_compat) = @_;
984    
985     ## NOTE: This code is used for |charset=""| attributes, |charset=|
986     ## portion of the |content=""| attributes, and |accept-charset=""|
987     ## attributes.
988 wakaba 1.91
989     ## NOTE: Though the case-sensitivility of |charset| attribute value
990     ## is not explicitly spelled in the HTML5 spec, the Character Set
991     ## registry of IANA, which is referenced from HTML5 spec, says that
992     ## charset name is case-insensitive.
993     $charset_value =~ tr/A-Z/a-z/; ## NOTE: ASCII Case-insensitive.
994    
995     require Message::Charset::Info;
996     my $charset = $Message::Charset::Info::IANACharset->{$charset_value};
997    
998     ## ISSUE: What is "valid character encoding name"? Syntactically valid?
999     ## Syntactically valid and registered? What about x-charset names?
1000     unless (Message::Charset::Info::is_syntactically_valid_iana_charset_name
1001     ($charset_value)) {
1002     $self->{onerror}->(node => $attr,
1003 wakaba 1.104 type => 'charset:syntax error',
1004     value => $charset_value,
1005     level => $self->{level}->{must});
1006 wakaba 1.91 }
1007    
1008     if ($charset) {
1009     ## ISSUE: What is "the preferred name for that encoding" (for a charset
1010     ## with no "preferred MIME name" label)?
1011     my $charset_status = $charset->{iana_names}->{$charset_value} || 0;
1012     if (($charset_status &
1013     Message::Charset::Info::PREFERRED_CHARSET_NAME ())
1014     != Message::Charset::Info::PREFERRED_CHARSET_NAME ()) {
1015     $self->{onerror}->(node => $attr,
1016 wakaba 1.104 type => 'charset:not preferred',
1017     value => $charset_value,
1018     level => $self->{level}->{must});
1019 wakaba 1.91 }
1020 wakaba 1.129
1021 wakaba 1.91 if (($charset_status &
1022     Message::Charset::Info::REGISTERED_CHARSET_NAME ())
1023     != Message::Charset::Info::REGISTERED_CHARSET_NAME ()) {
1024     if ($charset_value =~ /^x-/) {
1025     $self->{onerror}->(node => $attr,
1026 wakaba 1.104 type => 'charset:private',
1027     value => $charset_value,
1028     level => $self->{level}->{good});
1029 wakaba 1.91 } else {
1030     $self->{onerror}->(node => $attr,
1031 wakaba 1.104 type => 'charset:not registered',
1032     value => $charset_value,
1033     level => $self->{level}->{good});
1034 wakaba 1.91 }
1035     }
1036 wakaba 1.129
1037     if ($ascii_compat) {
1038     if ($charset->{category} &
1039     Message::Charset::Info::CHARSET_CATEGORY_ASCII_COMPAT ()) {
1040     #
1041     } else {
1042     $self->{onerror}->(node => $attr,
1043     type => 'charset:not ascii compat',
1044     value => $charset_value,
1045     level => $self->{level}->{must});
1046     }
1047     }
1048    
1049 wakaba 1.91 ## TODO: non-preferred-name error for following cases:
1050     } elsif ($charset_value =~ /^x-/) {
1051     $self->{onerror}->(node => $attr,
1052 wakaba 1.104 type => 'charset:private',
1053     value => $charset_value,
1054     level => $self->{level}->{good});
1055 wakaba 1.129
1056     ## NOTE: Whether this is an ASCII-compatible character encoding or
1057     ## not is unknown.
1058 wakaba 1.91 } else {
1059     $self->{onerror}->(node => $attr,
1060 wakaba 1.104 type => 'charset:not registered',
1061     value => $charset_value,
1062     level => $self->{level}->{good});
1063 wakaba 1.129
1064     ## NOTE: Whether this is an ASCII-compatible character encoding or
1065     ## not is unknown.
1066 wakaba 1.91 }
1067    
1068     return ($charset, $charset_value);
1069     }; # $HTMLCharsetChecker
1070    
1071 wakaba 1.129 ## NOTE: "An ordered set of space-separated tokens" where "each token
1072     ## MUST be the preferred name of an ASCII-compatible character
1073     ## encoding".
1074     my $HTMLCharsetsAttrChecker = sub {
1075     my ($self, $attr) = @_;
1076    
1077     ## ISSUE: "ordered set of space-separated tokens" is not defined.
1078    
1079 wakaba 1.132 my @value = grep {length $_} split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value;
1080 wakaba 1.129
1081 wakaba 1.176 ## XXX
1082 wakaba 1.129 ## ISSUE: Uniqueness is not enforced.
1083    
1084     for my $charset (@value) {
1085     $HTMLCharsetChecker->($charset, $self, $attr, 1);
1086     }
1087    
1088     ## ISSUE: Shift_JIS is ASCII-compatible? What about ISO-2022-JP?
1089     }; # $HTMLCharsetsAttrChecker
1090    
1091 wakaba 1.68 my $HTMLColorAttrChecker = sub {
1092     my ($self, $attr) = @_;
1093    
1094     ## NOTE: HTML4 "color" or |%Color;|
1095    
1096     my $value = $attr->value;
1097    
1098     if ($value !~ /\A(?>#[0-9A-F]+|black|silver|gray|white|maroon|red|purple|fuchsia|green|lime|olive|yellow|navy|blue|teal|aqua)\z/i) {
1099 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'color:syntax error',
1100 wakaba 1.105 level => $self->{level}->{html4_fact});
1101 wakaba 1.68 }
1102    
1103     ## TODO: HTML4 has some guideline on usage of color.
1104     }; # $HTMLColorAttrChecker
1105    
1106 wakaba 1.79 my $HTMLRefOrTemplateAttrChecker = sub {
1107     my ($self, $attr) = @_;
1108     $HTMLURIAttrChecker->(@_);
1109    
1110     my $attr_name = $attr->name;
1111    
1112     if ($attr_name eq 'ref') {
1113     unless ($attr->owner_element->has_attribute_ns (undef, 'template')) {
1114     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1115 wakaba 1.104 level => $self->{level}->{must});
1116 wakaba 1.79 }
1117     }
1118 wakaba 1.155
1119     require Message::URL;
1120 wakaba 1.79 my $doc = $attr->owner_document;
1121     my $doc_uri = $doc->document_uri;
1122 wakaba 1.155 my $uri = Message::URL->new_abs ($attr->value, $doc_uri);
1123 wakaba 1.79 my $no_frag_uri = $uri->clone;
1124     $no_frag_uri->uri_fragment (undef);
1125     if ((defined $doc_uri and $doc_uri eq $no_frag_uri) or
1126     (not defined $doc_uri and $no_frag_uri eq '')) {
1127     my $fragid = $uri->uri_fragment;
1128     if (defined $fragid) {
1129     push @{$self->{$attr_name}}, [$fragid => $attr];
1130     } else {
1131     DOCEL: {
1132     last DOCEL unless $attr_name eq 'template';
1133    
1134     my $docel = $doc->document_element;
1135     if ($docel) {
1136     my $nsuri = $docel->namespace_uri;
1137     if (defined $nsuri and $nsuri eq $HTML_NS) {
1138     if ($docel->manakai_local_name eq 'datatemplate') {
1139     last DOCEL;
1140     }
1141     }
1142     }
1143    
1144     $self->{onerror}->(node => $attr, type => 'template:not template',
1145 wakaba 1.104 level => $self->{level}->{must});
1146 wakaba 1.79 } # DOCEL
1147     }
1148     } else {
1149 wakaba 1.208 ## An external document is referenced.
1150    
1151     ## NOTE: Maybe the same-policy restriction should be posed to the
1152     ## referenced document, but the spec did not define such
1153     ## requirements and the entire feature has already been dropped
1154     ## from the spec anyway.
1155    
1156     ## XXXresource:
1157     ## - The document MUST be an HTML or XML document.
1158     ## - If there is a fragment identifier, it MUST point a part of the doc.
1159     ## - If the attribute is |template|, the pointed part MUST be a
1160     ## |datatemplat| element.
1161     ## - If no fragment identifier is specified, the root element MUST be
1162     ## a |datatemplate| element when the attribute is |template|.
1163 wakaba 1.79 }
1164     }; # $HTMLRefOrTemplateAttrChecker
1165    
1166 wakaba 1.83 my $HTMLRepeatIndexAttrChecker = sub {
1167     my ($self, $attr) = @_;
1168    
1169     if (defined $attr->namespace_uri) {
1170     my $oe = $attr->owner_element;
1171     my $oe_nsuri = $oe->namespace_uri;
1172 wakaba 1.128 if (defined $oe_nsuri or $oe_nsuri eq $HTML_NS) { ## TODO: wrong?
1173 wakaba 1.83 $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1174 wakaba 1.104 level => $self->{level}->{must});
1175 wakaba 1.83 }
1176     }
1177    
1178     $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 })->(@_);
1179     }; # $HTMLRepeatIndexAttrChecker
1180    
1181 wakaba 1.179 my $PlaceholderAttrChecker = sub {
1182     my ($self, $attr) = @_;
1183     if ($attr->value =~ /[\x0D\x0A]/) {
1184     $self->{onerror}->(node => $attr,
1185     type => 'newline in value', ## TODOC: type
1186     level => $self->{level}->{must});
1187     }
1188     }; # $PlaceholderAttrChecker
1189    
1190 wakaba 1.1 my $HTMLAttrChecker = {
1191 wakaba 1.176 accesskey => sub {
1192     my ($self, $attr) = @_;
1193    
1194     ## "Ordered set of unique space-separated tokens"
1195    
1196     my %keys;
1197     my @keys = grep {length} split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value;
1198    
1199     for my $key (@keys) {
1200     unless ($keys{$key}) {
1201     $keys{$key} = 1;
1202     if (length $key != 1) {
1203     $self->{onerror}->(node => $attr, type => 'char:syntax error',
1204     value => $key,
1205     level => $self->{level}->{must});
1206     }
1207     } else {
1208     $self->{onerror}->(node => $attr, type => 'duplicate token',
1209     value => $key,
1210     level => $self->{level}->{must});
1211     }
1212     }
1213     }, # accesskey
1214    
1215 wakaba 1.58 ## TODO: aria-* ## TODO: svg:*/@aria-* [HTML5ROLE] -> [STATES]
1216 wakaba 1.1 id => sub {
1217 wakaba 1.135 my ($self, $attr, $item, $element_state) = @_;
1218 wakaba 1.1 my $value = $attr->value;
1219     if (length $value > 0) {
1220     if ($self->{id}->{$value}) {
1221 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate ID',
1222     level => $self->{level}->{must});
1223 wakaba 1.1 push @{$self->{id}->{$value}}, $attr;
1224     } else {
1225     $self->{id}->{$value} = [$attr];
1226 wakaba 1.135 $self->{id_type}->{$value} = $element_state->{id_type} || '';
1227 wakaba 1.1 }
1228 wakaba 1.132 if ($value =~ /[\x09\x0A\x0C\x0D\x20]/) {
1229 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'space in ID',
1230     level => $self->{level}->{must});
1231 wakaba 1.1 }
1232     } else {
1233     ## NOTE: MUST contain at least one character
1234 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'empty attribute value',
1235     level => $self->{level}->{must});
1236 wakaba 1.1 }
1237     },
1238     title => sub {}, ## NOTE: No conformance creteria
1239     lang => sub {
1240     my ($self, $attr) = @_;
1241 wakaba 1.6 my $value = $attr->value;
1242     if ($value eq '') {
1243     #
1244     } else {
1245     require Whatpm::LangTag;
1246     Whatpm::LangTag->check_rfc3066_language_tag ($value, sub {
1247 wakaba 1.104 $self->{onerror}->(@_, node => $attr);
1248 wakaba 1.106 }, $self->{level});
1249 wakaba 1.6 }
1250 wakaba 1.1 ## ISSUE: RFC 4646 (3066bis)?
1251 wakaba 1.6
1252     ## TODO: test data
1253 wakaba 1.111
1254     ## NOTE: Inconsistency between |lang| and |xml:lang| attributes are
1255     ## non-conforming. Such errors are detected by the checkers of
1256     ## |{}xml:lang| and |{xml}:lang| attributes.
1257 wakaba 1.1 },
1258     dir => $GetHTMLEnumeratedAttrChecker->({ltr => 1, rtl => 1}),
1259     class => sub {
1260     my ($self, $attr) = @_;
1261 wakaba 1.132
1262     ## NOTE: "Unordered set of unique space-separated tokens".
1263    
1264 wakaba 1.1 my %word;
1265 wakaba 1.132 for my $word (grep {length $_}
1266     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
1267 wakaba 1.1 unless ($word{$word}) {
1268     $word{$word} = 1;
1269     push @{$self->{return}->{class}->{$word}||=[]}, $attr;
1270     } else {
1271 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'duplicate token',
1272     value => $word,
1273     level => $self->{level}->{must});
1274 wakaba 1.1 }
1275     }
1276     },
1277 wakaba 1.63 contenteditable => $GetHTMLEnumeratedAttrChecker->({
1278     true => 1, false => 1, '' => 1,
1279     }),
1280 wakaba 1.1 contextmenu => sub {
1281     my ($self, $attr) = @_;
1282     my $value = $attr->value;
1283 wakaba 1.138 push @{$self->{idref}}, ['menu', $value => $attr];
1284 wakaba 1.1 ## ISSUE: "The value must be the ID of a menu element in the DOM."
1285     ## What is "in the DOM"? A menu Element node that is not part
1286     ## of the Document tree is in the DOM? A menu Element node that
1287     ## belong to another Document tree is in the DOM?
1288     },
1289 wakaba 1.115 hidden => $GetHTMLBooleanAttrChecker->('hidden'),
1290 wakaba 1.60 irrelevant => $GetHTMLBooleanAttrChecker->('irrelevant'),
1291 wakaba 1.79 ref => $HTMLRefOrTemplateAttrChecker,
1292     registrationmark => sub {
1293     my ($self, $attr, $item, $element_state) = @_;
1294    
1295     ## NOTE: Any value is conforming.
1296    
1297     if ($self->{flag}->{in_rule}) {
1298     my $el = $attr->owner_element;
1299     my $ln = $el->manakai_local_name;
1300     if ($ln eq 'nest' or
1301     ($ln eq 'rule' and not $element_state->{in_rule_original})) {
1302     my $nsuri = $el->namespace_uri;
1303     if (defined $nsuri and $nsuri eq $HTML_NS) {
1304     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1305 wakaba 1.104 level => $self->{level}->{must});
1306 wakaba 1.79 }
1307     }
1308     } else {
1309     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1310 wakaba 1.104 level => $self->{level}->{must});
1311 wakaba 1.79 }
1312     },
1313 wakaba 1.80 repeat => sub {
1314     my ($self, $attr) = @_;
1315 wakaba 1.83
1316     if (defined $attr->namespace_uri) {
1317     my $oe = $attr->owner_element;
1318     my $oe_nsuri = $oe->namespace_uri;
1319     if (defined $oe_nsuri or $oe_nsuri eq $HTML_NS) {
1320     $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1321 wakaba 1.104 level => $self->{level}->{must});
1322 wakaba 1.83 }
1323     }
1324    
1325 wakaba 1.80 my $value = $attr->value;
1326     if ($value eq 'template') {
1327     #
1328     } elsif ($value =~ /\A-?[0-9]+\z/) {
1329     #
1330     } else {
1331     $self->{onerror}->(node => $attr, type => 'repeat:syntax error',
1332 wakaba 1.104 level => $self->{level}->{must});
1333 wakaba 1.80 }
1334    
1335 wakaba 1.208 ## NOTE: Where this attribute is allowed to set was not clearly
1336     ## defined in Web Forms 2.0. The spec said that "Repetition
1337     ## templates may occur anywhere", which might imply the attribute
1338     ## can be specified to any element, but its primary implication
1339     ## would be that the template can be appear in any hierarchy in
1340     ## the document structure. Anyway, the feature has been removed
1341     ## from the HTML5 spec.
1342 wakaba 1.80 },
1343 wakaba 1.83 'repeat-min' => $HTMLRepeatIndexAttrChecker,
1344     'repeat-max' => $HTMLRepeatIndexAttrChecker,
1345     'repeat-start' => $HTMLRepeatIndexAttrChecker,
1346 wakaba 1.80 'repeat-template' => sub {
1347 wakaba 1.83 my ($self, $attr) = @_;
1348    
1349     if (defined $attr->namespace_uri) {
1350     my $oe = $attr->owner_element;
1351     my $oe_nsuri = $oe->namespace_uri;
1352 wakaba 1.208 if (defined $oe_nsuri and $oe_nsuri eq $HTML_NS) {
1353 wakaba 1.83 $self->{onerror}->(node => $attr, type => 'attribute not allowed',
1354 wakaba 1.104 level => $self->{level}->{must});
1355 wakaba 1.83 }
1356     }
1357 wakaba 1.208
1358     ## NOTE: In the Web Forms 2.0 specification, this attribute had no
1359     ## author requirement. In addition, though the spec said that
1360     ## repetition blocks MAY have this attribute specified, it did not
1361     ## explicitly prohibit the attribute specified on an element that
1362     ## is not a repetition block. In anyway, the repetition template
1363     ## feature has been removed from the HTML5 specification.
1364 wakaba 1.80 },
1365 wakaba 1.58 ## TODO: role [HTML5ROLE] ## TODO: global @role [XHTML1ROLE]
1366 wakaba 1.184 spellcheck => $GetHTMLEnumeratedAttrChecker->({
1367     true => 1, false => 1, '' => 1,
1368     }),
1369 wakaba 1.128 style => sub {
1370     my ($self, $attr) = @_;
1371    
1372     $self->{onsubdoc}->({s => $attr->value,
1373     container_node => $attr,
1374     media_type => 'text/x-css-inline',
1375     is_char_string => 1});
1376    
1377     ## NOTE: "... MUST still be comprehensible and usable if those
1378     ## attributes were removed" is a semantic requirement, it cannot
1379     ## be tested.
1380     },
1381 wakaba 1.74 tabindex => $HTMLIntegerAttrChecker,
1382 wakaba 1.79 template => $HTMLRefOrTemplateAttrChecker,
1383 wakaba 1.208
1384     ## The |xml:lang| attribute in the null namespace, which is
1385     ## different from the |lang| attribute in the XML's namespace.
1386 wakaba 1.111 'xml:lang' => sub {
1387     my ($self, $attr) = @_;
1388    
1389     if ($attr->owner_document->manakai_is_html) {
1390     $self->{onerror}->(type => 'in HTML:xml:lang',
1391     level => $self->{level}->{info},
1392     node => $attr);
1393     ## NOTE: This is not an error, but the attribute will be ignored.
1394     } else {
1395     $self->{onerror}->(type => 'in XML:xml:lang',
1396     level => $self->{level}->{html5_no_may},
1397     node => $attr);
1398     ## TODO: We need to add test for this error.
1399     }
1400    
1401     my $lang_attr = $attr->owner_element->get_attribute_node_ns
1402     (undef, 'lang');
1403     if ($lang_attr) {
1404     my $lang_attr_value = $lang_attr->value;
1405     $lang_attr_value =~ tr/A-Z/a-z/; ## ASCII case-insensitive
1406     my $value = $attr->value;
1407     $value =~ tr/A-Z/a-z/; ## ASCII case-insensitive
1408     if ($lang_attr_value ne $value) {
1409     $self->{onerror}->(type => 'xml:lang ne lang',
1410     level => $self->{level}->{must},
1411     node => $attr);
1412     }
1413     } else {
1414     $self->{onerror}->(type => 'xml:lang not allowed',
1415     level => $self->{level}->{must},
1416     node => $attr);
1417     ## TODO: We need to add test for <x {xml}:lang {}xml:lang>.
1418     }
1419     },
1420 wakaba 1.208
1421     ## The |xmlns| attribute in the null namespace, which is different
1422     ## from the |xmlns| attribute in the XMLNS namespace.
1423 wakaba 1.74 xmlns => sub {
1424     my ($self, $attr) = @_;
1425     my $value = $attr->value;
1426     unless ($value eq $HTML_NS) {
1427 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'invalid attribute value',
1428     level => $self->{level}->{must});
1429 wakaba 1.74 ## TODO: Should be new "bad namespace" error?
1430     }
1431     unless ($attr->owner_document->manakai_is_html) {
1432 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'in XML:xmlns',
1433     level => $self->{level}->{must});
1434 wakaba 1.74 ## TODO: Test
1435     }
1436    
1437     ## TODO: Should be resolved?
1438     push @{$self->{return}->{uri}->{$value} ||= []},
1439     {node => $attr, type => {namespace => 1}};
1440     },
1441 wakaba 1.1 };
1442    
1443 wakaba 1.49 my %HTMLAttrStatus = (
1444 wakaba 1.176 accesskey => FEATURE_HTML5_FD,
1445 wakaba 1.187 class => FEATURE_HTML5_LC,
1446     contenteditable => FEATURE_HTML5_REC,
1447 wakaba 1.50 contextmenu => FEATURE_HTML5_WD,
1448 wakaba 1.187 dir => FEATURE_HTML5_REC,
1449 wakaba 1.50 draggable => FEATURE_HTML5_LC,
1450 wakaba 1.187 hidden => FEATURE_HTML5_LC,
1451     id => FEATURE_HTML5_REC,
1452 wakaba 1.115 irrelevant => FEATURE_HTML5_DROPPED,
1453 wakaba 1.187 lang => FEATURE_HTML5_REC,
1454 wakaba 1.208 ref => FEATURE_HTML5_DROPPED,
1455     registrationmark => FEATURE_HTML5_DROPPED,
1456 wakaba 1.60 repeat => FEATURE_WF2,
1457     'repeat-max' => FEATURE_WF2,
1458     'repeat-min' => FEATURE_WF2,
1459     'repeat-start' => FEATURE_WF2,
1460     'repeat-template' => FEATURE_WF2,
1461 wakaba 1.154 role => 0,
1462 wakaba 1.184 spellcheck => FEATURE_HTML5_WD,
1463 wakaba 1.187 style => FEATURE_HTML5_REC,
1464 wakaba 1.50 tabindex => FEATURE_HTML5_DEFAULT,
1465 wakaba 1.208 template => FEATURE_HTML5_DROPPED,
1466 wakaba 1.187 title => FEATURE_HTML5_REC,
1467 wakaba 1.154 xmlns => FEATURE_HTML5_WD,
1468 wakaba 1.49 );
1469    
1470     my %HTMLM12NCommonAttrStatus = (
1471 wakaba 1.154 about => FEATURE_RDFA_REC,
1472 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
1473 wakaba 1.154 content => FEATURE_RDFA_REC,
1474     datatype => FEATURE_RDFA_REC,
1475 wakaba 1.187 dir => FEATURE_HTML5_REC,
1476 wakaba 1.154 href => FEATURE_RDFA_REC,
1477 wakaba 1.187 id => FEATURE_HTML5_REC,
1478 wakaba 1.154 instanceof => FEATURE_RDFA_LC_DROPPED,
1479 wakaba 1.50 onclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1480     ondblclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1481     onmousedown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1482     onmouseup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1483     onmouseover => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1484     onmousemove => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1485     onmouseout => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1486     onkeypress => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1487     onkeydown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1488     onkeyup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
1489 wakaba 1.154 property => FEATURE_RDFA_REC,
1490     rel => FEATURE_RDFA_REC,
1491     resource => FEATURE_RDFA_REC,
1492     rev => FEATURE_RDFA_REC,
1493 wakaba 1.153 #style => FEATURE_HTML5_WD | FEATURE_XHTMLBASIC11_CR_DEPRECATED |
1494 wakaba 1.78 # FEATURE_M12N10_REC,
1495 wakaba 1.187 style => FEATURE_HTML5_REC,
1496     title => FEATURE_HTML5_REC,
1497 wakaba 1.154 typeof => FEATURE_RDFA_REC,
1498 wakaba 1.49 );
1499    
1500 wakaba 1.82 my %XHTML2CommonAttrStatus = (
1501     ## Core
1502 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_XHTML2_ED,
1503     id => FEATURE_HTML5_REC,
1504 wakaba 1.82 #xml:id
1505     layout => FEATURE_XHTML2_ED,
1506 wakaba 1.187 title => FEATURE_HTML5_REC,
1507 wakaba 1.82
1508     ## Hypertext
1509     cite => FEATURE_XHTML2_ED,
1510     href => FEATURE_XHTML2_ED,
1511     hreflang => FEATURE_XHTML2_ED,
1512     hrefmedia => FEATURE_XHTML2_ED,
1513     hreftype => FEATURE_XHTML2_ED,
1514     nextfocus => FEATURE_XHTML2_ED,
1515     prevfocus => FEATURE_XHTML2_ED,
1516     target => FEATURE_XHTML2_ED,
1517     #xml:base
1518    
1519     ## I18N
1520     #xml:lang
1521    
1522     ## Bi-directional
1523 wakaba 1.187 dir => FEATURE_HTML5_REC,
1524 wakaba 1.82
1525     ## Edit
1526     edit => FEATURE_XHTML2_ED,
1527     datetime => FEATURE_XHTML2_ED,
1528    
1529     ## Embedding
1530     encoding => FEATURE_XHTML2_ED,
1531     src => FEATURE_XHTML2_ED,
1532     srctype => FEATURE_XHTML2_ED,
1533    
1534     ## Image Map
1535     usemap => FEATURE_XHTML2_ED,
1536     ismap => FEATURE_XHTML2_ED,
1537     shape => FEATURE_XHTML2_ED,
1538     coords => FEATURE_XHTML2_ED,
1539    
1540     ## Media
1541     media => FEATURE_XHTML2_ED,
1542    
1543     ## Metadata
1544     about => FEATURE_XHTML2_ED,
1545     content => FEATURE_XHTML2_ED,
1546     datatype => FEATURE_XHTML2_ED,
1547     instanceof => FEATURE_XHTML2_ED,
1548     property => FEATURE_XHTML2_ED,
1549     rel => FEATURE_XHTML2_ED,
1550     resource => FEATURE_XHTML2_ED,
1551     rev => FEATURE_XHTML2_ED,
1552    
1553     ## Role
1554 wakaba 1.154 role => FEATURE_XHTML2_ED,
1555 wakaba 1.82
1556     ## Style
1557 wakaba 1.187 style => FEATURE_HTML5_REC,
1558 wakaba 1.82 );
1559    
1560     my %HTMLM12NXHTML2CommonAttrStatus = (
1561     %HTMLM12NCommonAttrStatus,
1562     %XHTML2CommonAttrStatus,
1563    
1564 wakaba 1.154 about => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1565 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1566 wakaba 1.154 content => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1567     datatype => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1568 wakaba 1.187 dir => FEATURE_HTML5_REC,
1569 wakaba 1.154 href => FEATURE_RDFA_REC,
1570 wakaba 1.187 id => FEATURE_HTML5_REC,
1571 wakaba 1.154 instanceof => FEATURE_RDFA_LC_DROPPED | FEATURE_XHTML2_ED,
1572     property => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1573     rel => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1574     resource => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1575     rev => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
1576 wakaba 1.153 #style => FEATURE_HTML5_WD | FEATURE_XHTMLBASIC11_CR_DEPRECATED |
1577 wakaba 1.82 # FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1578 wakaba 1.187 style => FEATURE_HTML5_REC,
1579     title => FEATURE_HTML5_REC,
1580 wakaba 1.154 typeof => FEATURE_RDFA_REC,
1581 wakaba 1.82 );
1582    
1583 wakaba 1.1 for (qw/
1584 wakaba 1.188 onabort onblur onchange onclick oncontextmenu
1585 wakaba 1.1 ondblclick ondrag ondragend ondragenter ondragleave ondragover
1586     ondragstart ondrop onerror onfocus onkeydown onkeypress
1587 wakaba 1.180 onkeyup onload onmousedown onmousemove onmouseout
1588 wakaba 1.188 onmouseover onmouseup onmousewheel onscroll onselect
1589     onsubmit
1590 wakaba 1.1 /) {
1591     $HTMLAttrChecker->{$_} = $HTMLEventHandlerAttrChecker;
1592 wakaba 1.50 $HTMLAttrStatus{$_} = FEATURE_HTML5_DEFAULT;
1593 wakaba 1.1 }
1594    
1595 wakaba 1.170 for (qw/
1596 wakaba 1.188 onbeforeunload onhashchange onresize onstorage onunload
1597 wakaba 1.170 ondataunavailable
1598 wakaba 1.180 onmessage
1599 wakaba 1.170 /) {
1600     $HTMLAttrChecker->{$_} = $HTMLEventHandlerAttrChecker;
1601     $HTMLAttrStatus{$_} = FEATURE_HTML5_DROPPED;
1602     }
1603    
1604 wakaba 1.82 ## NOTE: Non-standard global attributes in the HTML namespace.
1605     $AttrChecker->{$HTML_NS}->{''} = sub {}; # no syntactical checks
1606     $AttrStatus->{$HTML_NS}->{''} = 0; # disallowed and not part of any standard
1607    
1608     $AttrStatus->{$HTML_NS}->{active} = FEATURE_HTML5_DROPPED;
1609     for (qw/repeat repeat-max repeat-min repeat-start repeat-template/) {
1610     $AttrChecker->{$HTML_NS}->{$_} = $HTMLAttrChecker->{$_};
1611     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_WF2;
1612     }
1613    
1614 wakaba 1.120 for (qw/about content datatype property rel resource rev/) {
1615 wakaba 1.154 $AttrStatus->{$HTML_NS}->{$_} = FEATURE_RDFA_REC | FEATURE_XHTML2_ED;
1616 wakaba 1.82 }
1617 wakaba 1.154 $AttrStatus->{$HTML_NS}->{instanceof} = FEATURE_RDFA_LC_DROPPED | FEATURE_XHTML2_ED;
1618     $AttrStatus->{$HTML_NS}->{typeof} = FEATURE_RDFA_REC;
1619 wakaba 1.82 $AttrStatus->{$HTML_NS}->{role} = FEATURE_ROLE_LC;
1620     for (qw/cite coords datetime edit encoding href hreflang hrefmedia hreftype
1621     ismap layout media nextfocus prevfocus shape src srctype style
1622     target usemap/) {
1623     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_XHTML2_ED;
1624     }
1625     for (qw/class dir id title/) {
1626     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_M12N11_LC | FEATURE_XHTML2_ED;
1627     }
1628     for (qw/onclick ondblclick onmousedown onmouseup onmouseover onmousemove
1629     onmouseout onkeypress onkeydown onkeyup/) {
1630     $AttrStatus->{$HTML_NS}->{$_} = FEATURE_M12N11_LC;
1631     }
1632    
1633 wakaba 1.73 my $HTMLDatasetAttrChecker = sub {
1634     ## NOTE: "Authors should ... when the attributes are ignored and
1635     ## any associated CSS dropped, the page is still usable." (semantic
1636     ## constraint.)
1637     }; # $HTMLDatasetAttrChecker
1638    
1639 wakaba 1.187 my $HTMLDatasetAttrStatus = FEATURE_HTML5_LC;
1640 wakaba 1.73
1641 wakaba 1.1 my $GetHTMLAttrsChecker = sub {
1642     my $element_specific_checker = shift;
1643 wakaba 1.49 my $element_specific_status = shift;
1644 wakaba 1.1 return sub {
1645 wakaba 1.40 my ($self, $item, $element_state) = @_;
1646     for my $attr (@{$item->{node}->attributes}) {
1647 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
1648     $attr_ns = '' unless defined $attr_ns;
1649     my $attr_ln = $attr->manakai_local_name;
1650     my $checker;
1651 wakaba 1.73 my $status;
1652 wakaba 1.1 if ($attr_ns eq '') {
1653 wakaba 1.122 if ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
1654     $attr_ln !~ /[A-Z]/) {
1655 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
1656     $status = $HTMLDatasetAttrStatus;
1657     } else {
1658     $checker = $element_specific_checker->{$attr_ln}
1659     || $HTMLAttrChecker->{$attr_ln};
1660     $status = $element_specific_status->{$attr_ln};
1661     }
1662 wakaba 1.1 }
1663     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
1664 wakaba 1.40 || $AttrChecker->{$attr_ns}->{''};
1665 wakaba 1.82 $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
1666     || $AttrStatus->{$attr_ns}->{''};
1667     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
1668 wakaba 1.1 if ($checker) {
1669 wakaba 1.66 $checker->($self, $attr, $item, $element_state);
1670 wakaba 1.62 } elsif ($attr_ns eq '' and not $element_specific_status->{$attr_ln}) {
1671 wakaba 1.54 #
1672 wakaba 1.1 } else {
1673 wakaba 1.104 $self->{onerror}->(node => $attr,
1674     type => 'unknown attribute',
1675     level => $self->{level}->{uncertain});
1676 wakaba 1.49 ## ISSUE: No conformance createria for unknown attributes in the spec
1677     }
1678 wakaba 1.82 $self->_attr_status_info ($attr, $status);
1679 wakaba 1.1 }
1680     };
1681     }; # $GetHTMLAttrsChecker
1682    
1683 wakaba 1.40 my %HTMLChecker = (
1684     %Whatpm::ContentChecker::AnyChecker,
1685 wakaba 1.79 check_start => sub {
1686     my ($self, $item, $element_state) = @_;
1687    
1688     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
1689     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
1690     },
1691 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, \%HTMLAttrStatus),
1692 wakaba 1.40 );
1693    
1694     my %HTMLEmptyChecker = (
1695     %HTMLChecker,
1696     check_child_element => sub {
1697     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1698     $child_is_transparent, $element_state) = @_;
1699 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1700     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1701 wakaba 1.40 $self->{onerror}->(node => $child_el,
1702     type => 'element not allowed:minus',
1703 wakaba 1.104 level => $self->{level}->{must});
1704 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1705     #
1706     } else {
1707     $self->{onerror}->(node => $child_el,
1708     type => 'element not allowed:empty',
1709 wakaba 1.104 level => $self->{level}->{must});
1710 wakaba 1.40 }
1711     },
1712     check_child_text => sub {
1713     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1714     if ($has_significant) {
1715     $self->{onerror}->(node => $child_node,
1716     type => 'character not allowed:empty',
1717 wakaba 1.104 level => $self->{level}->{must});
1718 wakaba 1.40 }
1719     },
1720     );
1721    
1722     my %HTMLTextChecker = (
1723     %HTMLChecker,
1724     check_child_element => sub {
1725     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1726     $child_is_transparent, $element_state) = @_;
1727 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1728     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1729 wakaba 1.40 $self->{onerror}->(node => $child_el,
1730     type => 'element not allowed:minus',
1731 wakaba 1.104 level => $self->{level}->{must});
1732 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1733     #
1734     } else {
1735 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed:text',
1736     level => $self->{level}->{must});
1737 wakaba 1.40 }
1738     },
1739     );
1740    
1741 wakaba 1.72 my %HTMLFlowContentChecker = (
1742 wakaba 1.40 %HTMLChecker,
1743     check_child_element => sub {
1744     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1745     $child_is_transparent, $element_state) = @_;
1746 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1747     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1748 wakaba 1.40 $self->{onerror}->(node => $child_el,
1749     type => 'element not allowed:minus',
1750 wakaba 1.104 level => $self->{level}->{must});
1751 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1752     #
1753     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'style') {
1754     if ($element_state->{has_non_style} or
1755     not $child_el->has_attribute_ns (undef, 'scoped')) {
1756 wakaba 1.104 $self->{onerror}->(node => $child_el,
1757 wakaba 1.72 type => 'element not allowed:flow style',
1758 wakaba 1.104 level => $self->{level}->{must});
1759 wakaba 1.40 }
1760 wakaba 1.72 } elsif ($HTMLFlowContent->{$child_nsuri}->{$child_ln}) {
1761 wakaba 1.43 $element_state->{has_non_style} = 1 unless $child_is_transparent;
1762 wakaba 1.40 } else {
1763     $element_state->{has_non_style} = 1;
1764 wakaba 1.104 $self->{onerror}->(node => $child_el,
1765 wakaba 1.72 type => 'element not allowed:flow',
1766 wakaba 1.104 level => $self->{level}->{must})
1767 wakaba 1.40 }
1768     },
1769     check_child_text => sub {
1770     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1771     if ($has_significant) {
1772     $element_state->{has_non_style} = 1;
1773     }
1774     },
1775     check_end => sub {
1776     my ($self, $item, $element_state) = @_;
1777 wakaba 1.95 ## NOTE: A modified copy of the code below is in |datagrid| checker.
1778 wakaba 1.40 if ($element_state->{has_significant}) {
1779 wakaba 1.46 $item->{real_parent_state}->{has_significant} = 1;
1780 wakaba 1.40 } elsif ($item->{transparent}) {
1781     #
1782     } else {
1783     $self->{onerror}->(node => $item->{node},
1784 wakaba 1.104 level => $self->{level}->{should},
1785 wakaba 1.40 type => 'no significant content');
1786     }
1787     },
1788     );
1789    
1790     my %HTMLPhrasingContentChecker = (
1791     %HTMLChecker,
1792     check_child_element => sub {
1793     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1794     $child_is_transparent, $element_state) = @_;
1795 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1796     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1797 wakaba 1.40 $self->{onerror}->(node => $child_el,
1798     type => 'element not allowed:minus',
1799 wakaba 1.104 level => $self->{level}->{must});
1800 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1801     #
1802     } elsif ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
1803     #
1804     } else {
1805     $self->{onerror}->(node => $child_el,
1806     type => 'element not allowed:phrasing',
1807 wakaba 1.104 level => $self->{level}->{must});
1808 wakaba 1.40 }
1809     },
1810 wakaba 1.72 check_end => $HTMLFlowContentChecker{check_end},
1811 wakaba 1.40 ## NOTE: The definition for |li| assumes that the only differences
1812 wakaba 1.72 ## between flow and phrasing content checkers are |check_child_element|
1813 wakaba 1.40 ## and |check_child_text|.
1814     );
1815    
1816 wakaba 1.72 my %HTMLTransparentChecker = %HTMLFlowContentChecker;
1817 wakaba 1.40 ## ISSUE: Significant content rule should be applied to transparent element
1818 wakaba 1.46 ## with parent?
1819 wakaba 1.40
1820 wakaba 1.1 our $Element;
1821     our $ElementDefault;
1822    
1823     $Element->{$HTML_NS}->{''} = {
1824 wakaba 1.40 %HTMLChecker,
1825 wakaba 1.1 };
1826    
1827     $Element->{$HTML_NS}->{html} = {
1828 wakaba 1.187 status => FEATURE_HTML5_REC,
1829 wakaba 1.1 is_root => 1,
1830 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
1831 wakaba 1.16 manifest => $HTMLURIAttrChecker,
1832 wakaba 1.67 version => sub {
1833     ## NOTE: According to HTML4 prose, this is a "cdata" attribute.
1834     ## Though DTDs of various versions of HTML define the attribute
1835     ## as |#FIXED|, this conformance checker does no check for
1836     ## the attribute value, since what kind of check should be done
1837     ## is unknown.
1838     },
1839 wakaba 1.49 }, {
1840     %HTMLAttrStatus,
1841 wakaba 1.82 %XHTML2CommonAttrStatus,
1842 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_HTML2X_RFC,
1843     dir => FEATURE_HTML5_REC,
1844     id => FEATURE_HTML5_REC,
1845     lang => FEATURE_HTML5_REC,
1846 wakaba 1.153 manifest => FEATURE_HTML5_WD,
1847 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
1848 wakaba 1.82 version => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
1849 wakaba 1.1 }),
1850 wakaba 1.40 check_start => sub {
1851     my ($self, $item, $element_state) = @_;
1852     $element_state->{phase} = 'before head';
1853 wakaba 1.79
1854 wakaba 1.66 $element_state->{uri_info}->{manifest}->{type}->{resource} = 1;
1855 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
1856     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
1857 wakaba 1.40 },
1858     check_child_element => sub {
1859     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1860     $child_is_transparent, $element_state) = @_;
1861 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1862     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1863 wakaba 1.40 $self->{onerror}->(node => $child_el,
1864     type => 'element not allowed:minus',
1865 wakaba 1.104 level => $self->{level}->{must});
1866 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1867     #
1868     } elsif ($element_state->{phase} eq 'before head') {
1869     if ($child_nsuri eq $HTML_NS and $child_ln eq 'head') {
1870     $element_state->{phase} = 'after head';
1871     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'body') {
1872     $self->{onerror}->(node => $child_el,
1873 wakaba 1.104 type => 'ps element missing',
1874     text => 'head',
1875     level => $self->{level}->{must});
1876 wakaba 1.40 $element_state->{phase} = 'after body';
1877     } else {
1878     $self->{onerror}->(node => $child_el,
1879 wakaba 1.104 type => 'element not allowed',
1880     level => $self->{level}->{must});
1881 wakaba 1.40 }
1882     } elsif ($element_state->{phase} eq 'after head') {
1883     if ($child_nsuri eq $HTML_NS and $child_ln eq 'body') {
1884     $element_state->{phase} = 'after body';
1885     } else {
1886     $self->{onerror}->(node => $child_el,
1887 wakaba 1.104 type => 'element not allowed',
1888     level => $self->{level}->{must});
1889 wakaba 1.40 }
1890     } elsif ($element_state->{phase} eq 'after body') {
1891     $self->{onerror}->(node => $child_el,
1892 wakaba 1.104 type => 'element not allowed',
1893     level => $self->{level}->{must});
1894 wakaba 1.40 } else {
1895     die "check_child_element: Bad |html| phase: $element_state->{phase}";
1896     }
1897     },
1898     check_child_text => sub {
1899     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1900     if ($has_significant) {
1901     $self->{onerror}->(node => $child_node,
1902 wakaba 1.104 type => 'character not allowed',
1903     level => $self->{level}->{must});
1904 wakaba 1.40 }
1905     },
1906     check_end => sub {
1907     my ($self, $item, $element_state) = @_;
1908     if ($element_state->{phase} eq 'after body') {
1909     #
1910     } elsif ($element_state->{phase} eq 'before head') {
1911     $self->{onerror}->(node => $item->{node},
1912 wakaba 1.104 type => 'child element missing',
1913     text => 'head',
1914     level => $self->{level}->{must});
1915 wakaba 1.40 $self->{onerror}->(node => $item->{node},
1916 wakaba 1.104 type => 'child element missing',
1917     text => 'body',
1918     level => $self->{level}->{must});
1919 wakaba 1.40 } elsif ($element_state->{phase} eq 'after head') {
1920     $self->{onerror}->(node => $item->{node},
1921 wakaba 1.104 type => 'child element missing',
1922     text => 'body',
1923     level => $self->{level}->{must});
1924 wakaba 1.40 } else {
1925     die "check_end: Bad |html| phase: $element_state->{phase}";
1926     }
1927 wakaba 1.1
1928 wakaba 1.40 $HTMLChecker{check_end}->(@_);
1929     },
1930     };
1931 wakaba 1.25
1932 wakaba 1.40 $Element->{$HTML_NS}->{head} = {
1933 wakaba 1.187 status => FEATURE_HTML5_REC,
1934 wakaba 1.67 check_attrs => $GetHTMLAttrsChecker->({
1935     profile => $HTMLSpaceURIsAttrChecker, ## NOTE: MUST be profile URIs.
1936     }, {
1937 wakaba 1.49 %HTMLAttrStatus,
1938 wakaba 1.82 %XHTML2CommonAttrStatus,
1939 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_HTML2X_RFC,
1940     dir => FEATURE_HTML5_REC,
1941     id => FEATURE_HTML5_REC,
1942     lang => FEATURE_HTML5_REC,
1943 wakaba 1.49 profile => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC,
1944     }),
1945 wakaba 1.40 check_child_element => sub {
1946     my ($self, $item, $child_el, $child_nsuri, $child_ln,
1947     $child_is_transparent, $element_state) = @_;
1948 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
1949     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
1950 wakaba 1.40 $self->{onerror}->(node => $child_el,
1951     type => 'element not allowed:minus',
1952 wakaba 1.104 level => $self->{level}->{must});
1953 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
1954     #
1955     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'title') {
1956     unless ($element_state->{has_title}) {
1957     $element_state->{has_title} = 1;
1958     } else {
1959     $self->{onerror}->(node => $child_el,
1960     type => 'element not allowed:head title',
1961 wakaba 1.104 level => $self->{level}->{must});
1962 wakaba 1.40 }
1963     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'style') {
1964     if ($child_el->has_attribute_ns (undef, 'scoped')) {
1965     $self->{onerror}->(node => $child_el,
1966     type => 'element not allowed:head style',
1967 wakaba 1.104 level => $self->{level}->{must});
1968 wakaba 1.1 }
1969 wakaba 1.40 } elsif ($HTMLMetadataContent->{$child_nsuri}->{$child_ln}) {
1970     #
1971    
1972     ## NOTE: |meta| is a metadata content. However, strictly speaking,
1973     ## a |meta| element with none of |charset|, |name|,
1974     ## or |http-equiv| attribute is not allowed. It is non-conforming
1975     ## anyway.
1976 wakaba 1.56
1977     ## TODO: |form| MUST be empty and in XML [WF2].
1978 wakaba 1.40 } else {
1979     $self->{onerror}->(node => $child_el,
1980     type => 'element not allowed:metadata',
1981 wakaba 1.104 level => $self->{level}->{must});
1982 wakaba 1.40 }
1983     $element_state->{in_head_original} = $self->{flag}->{in_head};
1984     $self->{flag}->{in_head} = 1;
1985     },
1986     check_child_text => sub {
1987     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
1988     if ($has_significant) {
1989 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
1990     level => $self->{level}->{must});
1991 wakaba 1.1 }
1992 wakaba 1.40 },
1993     check_end => sub {
1994     my ($self, $item, $element_state) = @_;
1995     unless ($element_state->{has_title}) {
1996     $self->{onerror}->(node => $item->{node},
1997 wakaba 1.104 type => 'child element missing',
1998     text => 'title',
1999 wakaba 1.105 level => $self->{level}->{must});
2000 wakaba 1.1 }
2001 wakaba 1.40 $self->{flag}->{in_head} = $element_state->{in_head_original};
2002 wakaba 1.1
2003 wakaba 1.40 $HTMLChecker{check_end}->(@_);
2004 wakaba 1.1 },
2005     };
2006    
2007 wakaba 1.40 $Element->{$HTML_NS}->{title} = {
2008     %HTMLTextChecker,
2009 wakaba 1.187 status => FEATURE_HTML5_REC,
2010 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
2011     %HTMLAttrStatus,
2012 wakaba 1.82 %XHTML2CommonAttrStatus,
2013 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_HTML2X_RFC,
2014     dir => FEATURE_HTML5_REC,
2015     id => FEATURE_HTML5_REC,
2016     lang => FEATURE_HTML5_REC,
2017 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2018 wakaba 1.49 }),
2019 wakaba 1.40 };
2020 wakaba 1.1
2021 wakaba 1.40 $Element->{$HTML_NS}->{base} = {
2022 wakaba 1.153 status => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
2023 wakaba 1.40 %HTMLEmptyChecker,
2024     check_attrs => sub {
2025     my ($self, $item, $element_state) = @_;
2026 wakaba 1.1
2027 wakaba 1.40 if ($self->{has_base}) {
2028     $self->{onerror}->(node => $item->{node},
2029 wakaba 1.104 type => 'element not allowed:base',
2030     level => $self->{level}->{must});
2031 wakaba 1.40 } else {
2032     $self->{has_base} = 1;
2033 wakaba 1.29 }
2034    
2035 wakaba 1.40 my $has_href = $item->{node}->has_attribute_ns (undef, 'href');
2036     my $has_target = $item->{node}->has_attribute_ns (undef, 'target');
2037 wakaba 1.14
2038     if ($self->{has_uri_attr} and $has_href) {
2039 wakaba 1.4 ## ISSUE: Are these examples conforming?
2040     ## <head profile="a b c"><base href> (except for |profile|'s
2041     ## non-conformance)
2042     ## <title xml:base="relative"/><base href/> (maybe it should be)
2043     ## <unknown xmlns="relative"/><base href/> (assuming that
2044     ## |{relative}:unknown| is allowed before XHTML |base| (unlikely, though))
2045     ## <style>@import 'relative';</style><base href>
2046     ## <script>location.href = 'relative';</script><base href>
2047 wakaba 1.14 ## NOTE: <html manifest=".."><head><base href=""/> is conforming as
2048     ## an exception.
2049 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2050 wakaba 1.104 type => 'basehref after URL attribute',
2051     level => $self->{level}->{must});
2052 wakaba 1.4 }
2053 wakaba 1.14 if ($self->{has_hyperlink_element} and $has_target) {
2054 wakaba 1.4 ## ISSUE: Are these examples conforming?
2055     ## <head><title xlink:href=""/><base target="name"/></head>
2056     ## <xbl:xbl>...<svg:a href=""/>...</xbl:xbl><base target="name"/>
2057     ## (assuming that |xbl:xbl| is allowed before |base|)
2058     ## NOTE: These are non-conformant anyway because of |head|'s content model:
2059     ## <link href=""/><base target="name"/>
2060     ## <link rel=unknown href=""><base target=name>
2061 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2062 wakaba 1.104 type => 'basetarget after hyperlink',
2063     level => $self->{level}->{must});
2064 wakaba 1.4 }
2065    
2066 wakaba 1.14 if (not $has_href and not $has_target) {
2067 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2068 wakaba 1.104 type => 'attribute missing:href|target',
2069     level => $self->{level}->{must});
2070 wakaba 1.14 }
2071    
2072 wakaba 1.66 $element_state->{uri_info}->{href}->{type}->{base} = 1;
2073    
2074 wakaba 1.4 return $GetHTMLAttrsChecker->({
2075     href => $HTMLURIAttrChecker,
2076     target => $HTMLTargetAttrChecker,
2077 wakaba 1.49 }, {
2078     %HTMLAttrStatus,
2079 wakaba 1.153 href => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
2080 wakaba 1.187 id => FEATURE_HTML5_REC,
2081 wakaba 1.153 target => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
2082 wakaba 1.40 })->($self, $item, $element_state);
2083 wakaba 1.4 },
2084 wakaba 1.1 };
2085    
2086     $Element->{$HTML_NS}->{link} = {
2087 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2088 wakaba 1.40 %HTMLEmptyChecker,
2089     check_attrs => sub {
2090     my ($self, $item, $element_state) = @_;
2091 wakaba 1.96 my $sizes_attr;
2092 wakaba 1.1 $GetHTMLAttrsChecker->({
2093 wakaba 1.91 charset => sub {
2094     my ($self, $attr) = @_;
2095     $HTMLCharsetChecker->($attr->value, @_);
2096     },
2097 wakaba 1.1 href => $HTMLURIAttrChecker,
2098 wakaba 1.40 rel => sub { $HTMLLinkTypesAttrChecker->(0, $item, @_) },
2099 wakaba 1.92 rev => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->(),
2100 wakaba 1.1 media => $HTMLMQAttrChecker,
2101     hreflang => $HTMLLanguageTagAttrChecker,
2102 wakaba 1.96 sizes => sub {
2103     my ($self, $attr) = @_;
2104     $sizes_attr = $attr;
2105     my %word;
2106     for my $word (grep {length $_}
2107 wakaba 1.132 split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
2108 wakaba 1.96 unless ($word{$word}) {
2109     $word{$word} = 1;
2110     if ($word eq 'any' or $word =~ /\A[1-9][0-9]*x[1-9][0-9]*\z/) {
2111     #
2112     } else {
2113     $self->{onerror}->(node => $attr,
2114 wakaba 1.104 type => 'sizes:syntax error',
2115 wakaba 1.96 value => $word,
2116 wakaba 1.104 level => $self->{level}->{must});
2117 wakaba 1.96 }
2118     } else {
2119     $self->{onerror}->(node => $attr, type => 'duplicate token',
2120     value => $word,
2121 wakaba 1.104 level => $self->{level}->{must});
2122 wakaba 1.96 }
2123     }
2124     },
2125 wakaba 1.70 target => $HTMLTargetAttrChecker,
2126 wakaba 1.1 type => $HTMLIMTAttrChecker,
2127     ## NOTE: Though |title| has special semantics,
2128     ## syntactically same as the |title| as global attribute.
2129 wakaba 1.49 }, {
2130     %HTMLAttrStatus,
2131 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2132 wakaba 1.91 charset => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC,
2133     ## NOTE: |charset| attribute had been part of HTML5 spec though
2134     ## it had been commented out.
2135 wakaba 1.154 href => FEATURE_HTML5_LC | FEATURE_RDFA_REC | FEATURE_XHTML2_ED |
2136 wakaba 1.82 FEATURE_M12N10_REC,
2137 wakaba 1.153 hreflang => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2138 wakaba 1.187 lang => FEATURE_HTML5_REC,
2139 wakaba 1.153 media => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2140 wakaba 1.61 methods => FEATURE_HTML20_RFC,
2141 wakaba 1.154 rel => FEATURE_HTML5_LC | FEATURE_RDFA_REC | FEATURE_XHTML2_ED |
2142 wakaba 1.153 FEATURE_M12N10_REC,
2143 wakaba 1.154 rev => FEATURE_RDFA_REC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2144 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
2145 wakaba 1.153 sizes => FEATURE_HTML5_LC,
2146 wakaba 1.82 target => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2147 wakaba 1.153 # title: HTML5_WD | HTML5_LC | ...
2148     type => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
2149 wakaba 1.61 urn => FEATURE_HTML20_RFC,
2150 wakaba 1.40 })->($self, $item, $element_state);
2151 wakaba 1.96
2152 wakaba 1.40 if ($item->{node}->has_attribute_ns (undef, 'href')) {
2153     $self->{has_hyperlink_element} = 1 if $item->{has_hyperlink_link_type};
2154 wakaba 1.4 } else {
2155 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2156 wakaba 1.104 type => 'attribute missing',
2157     text => 'href',
2158     level => $self->{level}->{must});
2159 wakaba 1.1 }
2160 wakaba 1.96
2161 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'rel')) {
2162     $self->{onerror}->(node => $item->{node},
2163 wakaba 1.104 type => 'attribute missing',
2164     text => 'rel',
2165     level => $self->{level}->{must});
2166 wakaba 1.96 }
2167    
2168     if ($sizes_attr and not $element_state->{link_rel}->{icon}) {
2169     $self->{onerror}->(node => $sizes_attr,
2170     type => 'attribute not allowed',
2171 wakaba 1.104 level => $self->{level}->{must});
2172 wakaba 1.1 }
2173 wakaba 1.116
2174     if ($element_state->{link_rel}->{alternate} and
2175     $element_state->{link_rel}->{stylesheet}) {
2176     my $title_attr = $item->{node}->get_attribute_node_ns (undef, 'title');
2177     unless ($title_attr) {
2178     $self->{onerror}->(node => $item->{node},
2179     type => 'attribute missing',
2180     text => 'title',
2181     level => $self->{level}->{must});
2182     } elsif ($title_attr->value eq '') {
2183     $self->{onerror}->(node => $title_attr,
2184     type => 'empty style sheet title',
2185     level => $self->{level}->{must});
2186     }
2187     }
2188 wakaba 1.1 },
2189     };
2190    
2191     $Element->{$HTML_NS}->{meta} = {
2192 wakaba 1.153 status => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2193 wakaba 1.40 %HTMLEmptyChecker,
2194     check_attrs => sub {
2195     my ($self, $item, $element_state) = @_;
2196 wakaba 1.1 my $name_attr;
2197     my $http_equiv_attr;
2198     my $charset_attr;
2199     my $content_attr;
2200 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
2201 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
2202     $attr_ns = '' unless defined $attr_ns;
2203     my $attr_ln = $attr->manakai_local_name;
2204     my $checker;
2205 wakaba 1.73 my $status;
2206 wakaba 1.1 if ($attr_ns eq '') {
2207 wakaba 1.73 $status = {
2208     %HTMLAttrStatus,
2209 wakaba 1.82 %XHTML2CommonAttrStatus,
2210 wakaba 1.153 charset => FEATURE_HTML5_WD,
2211     content => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2212 wakaba 1.187 dir => FEATURE_HTML5_REC,
2213 wakaba 1.153 'http-equiv' => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
2214 wakaba 1.187 id => FEATURE_HTML5_REC,
2215     lang => FEATURE_HTML5_REC,
2216 wakaba 1.153 name => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
2217 wakaba 1.73 scheme => FEATURE_M12N10_REC,
2218     }->{$attr_ln};
2219    
2220 wakaba 1.1 if ($attr_ln eq 'content') {
2221     $content_attr = $attr;
2222     $checker = 1;
2223     } elsif ($attr_ln eq 'name') {
2224     $name_attr = $attr;
2225     $checker = 1;
2226     } elsif ($attr_ln eq 'http-equiv') {
2227     $http_equiv_attr = $attr;
2228     $checker = 1;
2229     } elsif ($attr_ln eq 'charset') {
2230     $charset_attr = $attr;
2231     $checker = 1;
2232 wakaba 1.67 } elsif ($attr_ln eq 'scheme') {
2233 wakaba 1.71 ## NOTE: <http://suika.fam.cx/2007/html/standards#html-meta-scheme>
2234 wakaba 1.67 $checker = sub {};
2235 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
2236     $attr_ln !~ /[A-Z]/) {
2237 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
2238     $status = $HTMLDatasetAttrStatus;
2239 wakaba 1.1 } else {
2240     $checker = $HTMLAttrChecker->{$attr_ln}
2241 wakaba 1.67 || $AttrChecker->{$attr_ns}->{$attr_ln}
2242 wakaba 1.1 || $AttrChecker->{$attr_ns}->{''};
2243     }
2244     } else {
2245     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
2246 wakaba 1.82 || $AttrChecker->{$attr_ns}->{''};
2247     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
2248     || $AttrStatus->{$attr_ns}->{''};
2249     $status = FEATURE_ALLOWED if not defined $status;
2250 wakaba 1.1 }
2251 wakaba 1.62
2252 wakaba 1.1 if ($checker) {
2253 wakaba 1.66 $checker->($self, $attr, $item, $element_state) if ref $checker;
2254 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
2255 wakaba 1.54 #
2256 wakaba 1.1 } else {
2257 wakaba 1.104 $self->{onerror}->(node => $attr,
2258     type => 'unknown attribute',
2259     level => $self->{level}->{uncertain});
2260 wakaba 1.49 ## ISSUE: No conformance createria for unknown attributes in the spec
2261     }
2262    
2263 wakaba 1.82 $self->_attr_status_info ($attr, $status);
2264 wakaba 1.1 }
2265    
2266     if (defined $name_attr) {
2267     if (defined $http_equiv_attr) {
2268     $self->{onerror}->(node => $http_equiv_attr,
2269 wakaba 1.104 type => 'attribute not allowed',
2270     level => $self->{level}->{must});
2271 wakaba 1.1 } elsif (defined $charset_attr) {
2272     $self->{onerror}->(node => $charset_attr,
2273 wakaba 1.104 type => 'attribute not allowed',
2274     level => $self->{level}->{must});
2275 wakaba 1.1 }
2276     my $metadata_name = $name_attr->value;
2277     my $metadata_value;
2278     if (defined $content_attr) {
2279     $metadata_value = $content_attr->value;
2280     } else {
2281 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2282 wakaba 1.104 type => 'attribute missing',
2283     text => 'content',
2284     level => $self->{level}->{must});
2285 wakaba 1.1 $metadata_value = '';
2286     }
2287     } elsif (defined $http_equiv_attr) {
2288     if (defined $charset_attr) {
2289     $self->{onerror}->(node => $charset_attr,
2290 wakaba 1.104 type => 'attribute not allowed',
2291     level => $self->{level}->{must});
2292 wakaba 1.1 }
2293     unless (defined $content_attr) {
2294 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2295 wakaba 1.104 type => 'attribute missing',
2296     text => 'content',
2297     level => $self->{level}->{must});
2298 wakaba 1.1 }
2299     } elsif (defined $charset_attr) {
2300     if (defined $content_attr) {
2301     $self->{onerror}->(node => $content_attr,
2302 wakaba 1.104 type => 'attribute not allowed',
2303     level => $self->{level}->{must});
2304 wakaba 1.1 }
2305     } else {
2306     if (defined $content_attr) {
2307     $self->{onerror}->(node => $content_attr,
2308 wakaba 1.104 type => 'attribute not allowed',
2309     level => $self->{level}->{must});
2310 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2311 wakaba 1.104 type => 'attribute missing:name|http-equiv',
2312     level => $self->{level}->{must});
2313 wakaba 1.1 } else {
2314 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2315 wakaba 1.104 type => 'attribute missing:name|http-equiv|charset',
2316     level => $self->{level}->{must});
2317 wakaba 1.1 }
2318     }
2319    
2320 wakaba 1.32 my $check_charset_decl = sub () {
2321 wakaba 1.40 my $parent = $item->{node}->manakai_parent_element;
2322 wakaba 1.29 if ($parent and $parent eq $parent->owner_document->manakai_head) {
2323     for my $el (@{$parent->child_nodes}) {
2324     next unless $el->node_type == 1; # ELEMENT_NODE
2325 wakaba 1.40 unless ($el eq $item->{node}) {
2326 wakaba 1.29 ## NOTE: Not the first child element.
2327 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2328 wakaba 1.32 type => 'element not allowed:meta charset',
2329 wakaba 1.104 level => $self->{level}->{must});
2330 wakaba 1.29 }
2331     last;
2332     ## NOTE: Entity references are not supported.
2333     }
2334     } else {
2335 wakaba 1.40 $self->{onerror}->(node => $item->{node},
2336 wakaba 1.32 type => 'element not allowed:meta charset',
2337 wakaba 1.104 level => $self->{level}->{must});
2338 wakaba 1.29 }
2339 wakaba 1.32 }; # $check_charset_decl
2340 wakaba 1.21
2341 wakaba 1.32 my $check_charset = sub ($$) {
2342     my ($attr, $charset_value) = @_;
2343 wakaba 1.21
2344 wakaba 1.91 my $charset;
2345     ($charset, $charset_value)
2346     = $HTMLCharsetChecker->($charset_value, $self, $attr);
2347    
2348 wakaba 1.40 my $ic = $item->{node}->owner_document->input_encoding;
2349 wakaba 1.21 if (defined $ic) {
2350     ## TODO: Test for this case
2351     my $ic_charset = $Message::Charset::Info::IANACharset->{$ic};
2352     if ($charset ne $ic_charset) {
2353 wakaba 1.32 $self->{onerror}->(node => $attr,
2354 wakaba 1.104 type => 'mismatched charset name',
2355 wakaba 1.106 text => $ic,
2356 wakaba 1.104 value => $charset_value,
2357     level => $self->{level}->{must});
2358 wakaba 1.21 }
2359     } else {
2360     ## NOTE: MUST, but not checkable, since the document is not originally
2361     ## in serialized form (or the parser does not preserve the input
2362     ## encoding information).
2363 wakaba 1.32 $self->{onerror}->(node => $attr,
2364 wakaba 1.104 type => 'mismatched charset name not checked',
2365     value => $charset_value,
2366     level => $self->{level}->{uncertain});
2367 wakaba 1.21 }
2368    
2369 wakaba 1.32 if ($attr->get_user_data ('manakai_has_reference')) {
2370     $self->{onerror}->(node => $attr,
2371 wakaba 1.104 type => 'charref in charset',
2372     level => $self->{level}->{must},
2373     layer => 'syntax');
2374 wakaba 1.22 }
2375 wakaba 1.32 }; # $check_charset
2376    
2377     ## TODO: metadata conformance
2378    
2379 wakaba 1.204 ## -- The |http-equiv| attribute (pragmas)
2380 wakaba 1.32 if (defined $http_equiv_attr) { ## An enumerated attribute
2381 wakaba 1.204 my $keyword = $http_equiv_attr->value;
2382     $keyword =~ tr/A-Z/a-z/; ## ASCII case-insensitive.
2383 wakaba 1.33
2384 wakaba 1.85 if ($self->{has_http_equiv}->{$keyword}) {
2385     $self->{onerror}->(type => 'duplicate http-equiv', value => $keyword,
2386     node => $http_equiv_attr,
2387 wakaba 1.104 level => $self->{level}->{must});
2388 wakaba 1.85 } else {
2389     $self->{has_http_equiv}->{$keyword} = 1;
2390     }
2391    
2392     if ($keyword eq 'content-type') {
2393 wakaba 1.58 ## TODO: refs in "text/html; charset=" are not disallowed since rev.1275.
2394 wakaba 1.33
2395 wakaba 1.32 $check_charset_decl->();
2396 wakaba 1.182
2397     unless ($item->{node}->owner_document->manakai_is_html) {
2398     $self->{onerror}->(node => $item->{node},
2399     type => 'in XML:charset',
2400     level => $self->{level}->{must});
2401     }
2402    
2403 wakaba 1.32 if ($content_attr) {
2404     my $content = $content_attr->value;
2405 wakaba 1.58 if ($content =~ m!^[Tt][Ee][Xx][Tt]/[Hh][Tt][Mm][Ll];
2406 wakaba 1.132 [\x09\x0A\x0C\x0D\x20]*[Cc][Hh][Aa][Rr][Ss][Ee][Tt]
2407 wakaba 1.58 =(.+)\z!sx) {
2408 wakaba 1.32 $check_charset->($content_attr, $1);
2409     } else {
2410     $self->{onerror}->(node => $content_attr,
2411     type => 'meta content-type syntax error',
2412 wakaba 1.104 level => $self->{level}->{must});
2413 wakaba 1.85 }
2414     }
2415     } elsif ($keyword eq 'default-style') {
2416 wakaba 1.204 ## XXX no author requirement in the spec
2417    
2418 wakaba 1.85 } elsif ($keyword eq 'refresh') {
2419     if ($content_attr) {
2420     my $content = $content_attr->value;
2421     if ($content =~ /\A[0-9]+\z/) {
2422     ## NOTE: Valid non-negative integer.
2423     #
2424 wakaba 1.132 } elsif ($content =~ s/\A[0-9]+;[\x09\x0A\x0C\x0D\x20]+[Uu][Rr][Ll]=//) {
2425 wakaba 1.204 ## XXXURL
2426 wakaba 1.85 Whatpm::URIChecker->check_iri_reference ($content, sub {
2427 wakaba 1.104 $self->{onerror}->(value => $content, @_, node => $content_attr);
2428 wakaba 1.106 }, $self->{level});
2429 wakaba 1.204 $self->{has_uri_attr} = 1; ## NOTE: One of "attributes with URLs".
2430 wakaba 1.85
2431     $element_state->{uri_info}->{content}->{node} = $content_attr;
2432     $element_state->{uri_info}->{content}->{type}->{hyperlink} = 1;
2433 wakaba 1.204 ## XXXTODO: absolute
2434 wakaba 1.85 push @{$self->{return}->{uri}->{$content} ||= []},
2435     $element_state->{uri_info}->{content};
2436     } else {
2437     $self->{onerror}->(node => $content_attr,
2438     type => 'refresh:syntax error',
2439 wakaba 1.104 level => $self->{level}->{must});
2440 wakaba 1.32 }
2441     }
2442 wakaba 1.204 } elsif ($keyword eq 'content-language') {
2443     if ($content_attr) {
2444     my $content = $content_attr->value;
2445     require Whatpm::LangTag;
2446     ## XXX In fact what the spec requires is "BCP 47 langauge code".
2447     Whatpm::LangTag->check_rfc3066_language_tag ($content, sub {
2448     $self->{onerror}->(@_, node => $content_attr);
2449     }, $self->{level});
2450     }
2451    
2452     ## XXX This is conforming but obsolete.
2453 wakaba 1.32 } else {
2454 wakaba 1.204 ## NOTE: |Content-Style-Type| and |Content-Script-Type|
2455     ## pragmas are listed in the table of the spec in the
2456     ## commented-out form, but there is no author requirement
2457     ## (even commented-out one isn't there).
2458    
2459     ## NOTE: Pragma extensions are listed in
2460     ## <http://wiki.whatwg.org/wiki/PragmaExtensions>. At the
2461     ## time of writing, no extension has been registered yet.
2462    
2463 wakaba 1.32 $self->{onerror}->(node => $http_equiv_attr,
2464 wakaba 1.104 type => 'enumerated:invalid',
2465     level => $self->{level}->{must});
2466 wakaba 1.32 }
2467     }
2468    
2469     if (defined $charset_attr) {
2470 wakaba 1.182 my $value = $charset_attr->value;
2471    
2472 wakaba 1.32 $check_charset_decl->();
2473 wakaba 1.182 $check_charset->($charset_attr, $value);
2474    
2475     if (not $item->{node}->owner_document->manakai_is_html and
2476     not $value =~ /\A[Uu][Tt][Ff]-8\z/) {
2477     $self->{onerror}->(node => $item->{node},
2478     type => 'in XML:charset',
2479     level => $self->{level}->{must});
2480     }
2481 wakaba 1.1 }
2482     },
2483     };
2484    
2485     $Element->{$HTML_NS}->{style} = {
2486 wakaba 1.153 status => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2487 wakaba 1.40 %HTMLChecker,
2488     check_attrs => $GetHTMLAttrsChecker->({
2489 wakaba 1.1 type => $HTMLIMTAttrChecker, ## TODO: MUST be a styling language
2490     media => $HTMLMQAttrChecker,
2491     scoped => $GetHTMLBooleanAttrChecker->('scoped'),
2492     ## NOTE: |title| has special semantics for |style|s, but is syntactically
2493     ## not different
2494 wakaba 1.49 }, {
2495     %HTMLAttrStatus,
2496 wakaba 1.82 %XHTML2CommonAttrStatus,
2497 wakaba 1.187 dir => FEATURE_HTML5_REC,
2498 wakaba 1.82 disabled => FEATURE_XHTML2_ED,
2499 wakaba 1.154 href => FEATURE_RDFA_REC | FEATURE_XHTML2_ED,
2500 wakaba 1.187 id => FEATURE_HTML5_REC,
2501     lang => FEATURE_HTML5_REC,
2502 wakaba 1.153 media => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2503     scoped => FEATURE_HTML5_FD,
2504 wakaba 1.187 title => FEATURE_HTML5_REC,
2505 wakaba 1.153 type => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2506 wakaba 1.1 }),
2507 wakaba 1.40 check_start => sub {
2508     my ($self, $item, $element_state) = @_;
2509    
2510 wakaba 1.27 ## NOTE: |html:style| itself has no conformance creteria on content model.
2511 wakaba 1.40 my $type = $item->{node}->get_attribute_ns (undef, 'type');
2512 wakaba 1.93 $type = 'text/css' unless defined $type;
2513     if ($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]) {
2514     $type = "$1/$2";
2515     $type =~ tr/A-Z/a-z/; ## NOTE: ASCII case-insensitive
2516     } else {
2517     ## NOTE: We don't know how parameters are handled by UAs. According to
2518     ## HTML5 specification, <style> with unknown parameters in |type=""|
2519     ## must be ignored.
2520     undef $type;
2521     }
2522     if (not defined $type) {
2523     $element_state->{allow_element} = 1; # invalid type=""
2524     } elsif ($type eq 'text/css') {
2525 wakaba 1.40 $element_state->{allow_element} = 0;
2526 wakaba 1.93 #} elsif ($type =~ m![/+][Xx][Mm][Ll]\z!) {
2527     # ## NOTE: There is no definition for "XML-based styling language" in HTML5
2528     # $element_state->{allow_element} = 1;
2529 wakaba 1.40 } else {
2530     $element_state->{allow_element} = 1; # unknown
2531     }
2532 wakaba 1.93 $element_state->{style_type} = $type;
2533 wakaba 1.79
2534     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2535     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2536 wakaba 1.107
2537     $element_state->{text} = '';
2538 wakaba 1.40 },
2539     check_child_element => sub {
2540     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2541     $child_is_transparent, $element_state) = @_;
2542 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2543     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2544 wakaba 1.40 $self->{onerror}->(node => $child_el,
2545     type => 'element not allowed:minus',
2546 wakaba 1.104 level => $self->{level}->{must});
2547 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2548     #
2549     } elsif ($element_state->{allow_element}) {
2550     #
2551     } else {
2552 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2553     level => $self->{level}->{must});
2554 wakaba 1.40 }
2555     },
2556     check_child_text => sub {
2557     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2558 wakaba 1.115 $element_state->{text} .= $child_node->data;
2559 wakaba 1.40 },
2560     check_end => sub {
2561     my ($self, $item, $element_state) = @_;
2562 wakaba 1.93 if (not defined $element_state->{style_type}) {
2563     ## NOTE: Invalid type=""
2564     #
2565     } elsif ($element_state->{style_type} eq 'text/css') {
2566 wakaba 1.40 $self->{onsubdoc}->({s => $element_state->{text},
2567     container_node => $item->{node},
2568 wakaba 1.28 media_type => 'text/css', is_char_string => 1});
2569 wakaba 1.93 } elsif ($element_state->{style_type} =~ m![+/][Xx][Mm][Ll]\z!) {
2570     ## NOTE: XML content should be checked by THIS instance of checker
2571     ## as part of normal tree validation. However, we don't know of any
2572     ## XML-based styling language that can be used in HTML <style> element,
2573     ## such that we throw a "style language not supported" error.
2574 wakaba 1.104 $self->{onerror}->(node => $item->{node},
2575     type => 'XML style lang',
2576     text => $element_state->{style_type},
2577     level => $self->{level}->{uncertain});
2578 wakaba 1.93 } else {
2579     ## NOTE: Should we raise some kind of error for,
2580     ## say, <style type="text/plaion">?
2581     $self->{onsubdoc}->({s => $element_state->{text},
2582     container_node => $item->{node},
2583     media_type => $element_state->{style_type},
2584     is_char_string => 1});
2585 wakaba 1.27 }
2586 wakaba 1.40
2587     $HTMLChecker{check_end}->(@_);
2588 wakaba 1.1 },
2589     };
2590 wakaba 1.25 ## ISSUE: Relationship to significant content check?
2591 wakaba 1.1
2592     $Element->{$HTML_NS}->{body} = {
2593 wakaba 1.72 %HTMLFlowContentChecker,
2594 wakaba 1.187 status => FEATURE_HTML5_REC,
2595 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2596     alink => $HTMLColorAttrChecker,
2597     background => $HTMLURIAttrChecker,
2598     bgcolor => $HTMLColorAttrChecker,
2599     link => $HTMLColorAttrChecker,
2600 wakaba 1.188 onafterprint => $HTMLEventHandlerAttrChecker,
2601     onbeforeprint => $HTMLEventHandlerAttrChecker,
2602     onbeforeunload => $HTMLEventHandlerAttrChecker,
2603     onblur => $HTMLEventHandlerAttrChecker,
2604     onerror => $HTMLEventHandlerAttrChecker,
2605     onfocus => $HTMLEventHandlerAttrChecker,
2606     onhashchange => $HTMLEventHandlerAttrChecker,
2607     onload => $HTMLEventHandlerAttrChecker,
2608     onmessage => $HTMLEventHandlerAttrChecker,
2609     onoffline => $HTMLEventHandlerAttrChecker,
2610     ononline => $HTMLEventHandlerAttrChecker,
2611 wakaba 1.186 onpopstate => $HTMLEventHandlerAttrChecker,
2612 wakaba 1.188 onredo => $HTMLEventHandlerAttrChecker,
2613     onresize => $HTMLEventHandlerAttrChecker,
2614     onstorage => $HTMLEventHandlerAttrChecker,
2615     onundo => $HTMLEventHandlerAttrChecker,
2616     onunload => $HTMLEventHandlerAttrChecker,
2617 wakaba 1.68 text => $HTMLColorAttrChecker,
2618     vlink => $HTMLColorAttrChecker,
2619     }, {
2620 wakaba 1.49 %HTMLAttrStatus,
2621 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2622 wakaba 1.49 alink => FEATURE_M12N10_REC_DEPRECATED,
2623     background => FEATURE_M12N10_REC_DEPRECATED,
2624     bgcolor => FEATURE_M12N10_REC_DEPRECATED,
2625 wakaba 1.187 lang => FEATURE_HTML5_REC,
2626 wakaba 1.49 link => FEATURE_M12N10_REC_DEPRECATED,
2627 wakaba 1.188 onafterprint => FEATURE_HTML5_LC,
2628     onbeforeprint => FEATURE_HTML5_LC,
2629     onbeforeunload => FEATURE_HTML5_LC,
2630     onblur => FEATURE_HTML5_LC,
2631     onerror => FEATURE_HTML5_LC,
2632     onfocus => FEATURE_HTML5_LC,
2633     onhashchange => FEATURE_HTML5_LC,
2634     onload => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
2635     onmessage => FEATURE_HTML5_LC,
2636     onoffline => FEATURE_HTML5_LC,
2637     ononline => FEATURE_HTML5_LC,
2638 wakaba 1.186 onpopstate => FEATURE_HTML5_LC,
2639 wakaba 1.188 onredo => FEATURE_HTML5_LC,
2640     onresize => FEATURE_HTML5_LC,
2641     onstorage => FEATURE_HTML5_LC,
2642     onundo => FEATURE_HTML5_LC,
2643     onunload => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
2644 wakaba 1.49 text => FEATURE_M12N10_REC_DEPRECATED,
2645     vlink => FEATURE_M12N10_REC_DEPRECATED,
2646     }),
2647 wakaba 1.68 check_start => sub {
2648     my ($self, $item, $element_state) = @_;
2649    
2650     $element_state->{uri_info}->{background}->{type}->{embedded} = 1;
2651 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2652     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2653 wakaba 1.68 },
2654 wakaba 1.1 };
2655    
2656     $Element->{$HTML_NS}->{section} = {
2657 wakaba 1.72 %HTMLFlowContentChecker,
2658 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED,
2659 wakaba 1.82 check_attrs => $GetHTMLAttrsChecker->({
2660 wakaba 1.189 cite => $HTMLURIAttrChecker,
2661 wakaba 1.82 }, {
2662     %HTMLAttrStatus,
2663     %XHTML2CommonAttrStatus,
2664 wakaba 1.189 cite => FEATURE_HTML5_DROPPED | FEATURE_XHTML2_ED,
2665 wakaba 1.82 }),
2666 wakaba 1.1 };
2667    
2668     $Element->{$HTML_NS}->{nav} = {
2669 wakaba 1.153 status => FEATURE_HTML5_LC,
2670 wakaba 1.72 %HTMLFlowContentChecker,
2671 wakaba 1.1 };
2672    
2673     $Element->{$HTML_NS}->{article} = {
2674 wakaba 1.174 %HTMLFlowContentChecker,
2675 wakaba 1.153 status => FEATURE_HTML5_LC,
2676 wakaba 1.174 check_attrs => $GetHTMLAttrsChecker->({
2677 wakaba 1.189 cite => $HTMLURIAttrChecker,
2678 wakaba 1.174 pubdate => $GetDateTimeAttrChecker->('global_date_and_time_string'),
2679     }, {
2680     %HTMLAttrStatus,
2681 wakaba 1.189 cite => FEATURE_HTML5_DROPPED,
2682 wakaba 1.174 pubdate => FEATURE_HTML5_LC,
2683     }),
2684     }; # article
2685 wakaba 1.1
2686     $Element->{$HTML_NS}->{blockquote} = {
2687 wakaba 1.154 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2688 wakaba 1.72 %HTMLFlowContentChecker,
2689 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
2690 wakaba 1.1 cite => $HTMLURIAttrChecker,
2691 wakaba 1.49 }, {
2692     %HTMLAttrStatus,
2693 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2694 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2695 wakaba 1.154 cite => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
2696 wakaba 1.187 lang => FEATURE_HTML5_REC,
2697 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2698 wakaba 1.1 }),
2699 wakaba 1.66 check_start => sub {
2700     my ($self, $item, $element_state) = @_;
2701    
2702     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
2703 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2704     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2705 wakaba 1.66 },
2706 wakaba 1.1 };
2707    
2708     $Element->{$HTML_NS}->{aside} = {
2709 wakaba 1.153 status => FEATURE_HTML5_LC,
2710 wakaba 1.72 %HTMLFlowContentChecker,
2711 wakaba 1.1 };
2712    
2713     $Element->{$HTML_NS}->{h1} = {
2714 wakaba 1.40 %HTMLPhrasingContentChecker,
2715 wakaba 1.187 status => FEATURE_HTML5_REC,
2716 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2717     align => $GetHTMLEnumeratedAttrChecker->({
2718     left => 1, center => 1, right => 1, justify => 1,
2719     }),
2720     }, {
2721 wakaba 1.49 %HTMLAttrStatus,
2722 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2723 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
2724 wakaba 1.187 lang => FEATURE_HTML5_REC,
2725 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2726 wakaba 1.49 }),
2727 wakaba 1.40 check_start => sub {
2728     my ($self, $item, $element_state) = @_;
2729     $self->{flag}->{has_hn} = 1;
2730 wakaba 1.79
2731     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2732     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2733 wakaba 1.1 },
2734     };
2735    
2736 wakaba 1.40 $Element->{$HTML_NS}->{h2} = {%{$Element->{$HTML_NS}->{h1}}};
2737 wakaba 1.1
2738 wakaba 1.40 $Element->{$HTML_NS}->{h3} = {%{$Element->{$HTML_NS}->{h1}}};
2739 wakaba 1.1
2740 wakaba 1.40 $Element->{$HTML_NS}->{h4} = {%{$Element->{$HTML_NS}->{h1}}};
2741 wakaba 1.1
2742 wakaba 1.40 $Element->{$HTML_NS}->{h5} = {%{$Element->{$HTML_NS}->{h1}}};
2743 wakaba 1.1
2744 wakaba 1.40 $Element->{$HTML_NS}->{h6} = {%{$Element->{$HTML_NS}->{h1}}};
2745 wakaba 1.1
2746 wakaba 1.29 ## TODO: Explicit sectioning is "encouraged".
2747 wakaba 1.174
2748 wakaba 1.195 $Element->{$HTML_NS}->{hgroup} = {
2749     %HTMLChecker,
2750     status => FEATURE_HTML5_LC,
2751     check_child_element => sub {
2752     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2753     $child_is_transparent, $element_state, $element_state2) = @_;
2754     if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2755     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2756     $self->{onerror}->(node => $child_el,
2757     type => 'element not allowed:minus',
2758     level => $self->{level}->{must});
2759     if ($child_nsuri eq $HTML_NS and $child_ln =~ /\Ah[1-6]\z/) {
2760     $element_state2->{has_hn} = 1;
2761     }
2762     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2763     #
2764     } elsif ($child_nsuri eq $HTML_NS and $child_ln =~ /\Ah[1-6]\z/) {
2765     ## NOTE: Use $element_state2 instead of $element_state here so
2766     ## that the |h2| element in |<hgroup><ins><h2>| is not counted
2767     ## as an |h2| of the |hgroup| element.
2768     $element_state2->{has_hn} = 1;
2769     } else {
2770     $self->{onerror}->(node => $child_el, type => 'element not allowed',
2771     level => $self->{level}->{must});
2772     }
2773     }, # check_child_element
2774     check_child_text => sub {
2775     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2776     if ($has_significant) {
2777     $self->{onerror}->(node => $child_node, type => 'character not allowed',
2778     level => $self->{level}->{must});
2779     }
2780     }, # check_child_text
2781     check_end => sub {
2782     my ($self, $item, $element_state) = @_;
2783     unless ($element_state->{has_hn}) {
2784     $self->{onerror}->(node => $item->{node},
2785     type => 'element missing:hn',
2786     level => $self->{level}->{must});
2787     }
2788    
2789     $HTMLChecker{check_end}->(@_);
2790     }, # check_end
2791     }; # hgroup
2792 wakaba 1.29
2793 wakaba 1.1 $Element->{$HTML_NS}->{header} = {
2794 wakaba 1.195 %HTMLFlowContentChecker,
2795 wakaba 1.153 status => FEATURE_HTML5_LC,
2796 wakaba 1.40 check_start => sub {
2797     my ($self, $item, $element_state) = @_;
2798     $self->_add_minus_elements ($element_state,
2799 wakaba 1.195 {$HTML_NS => {qw/header 1 footer 1/}});
2800 wakaba 1.40 $element_state->{has_hn_original} = $self->{flag}->{has_hn};
2801     $self->{flag}->{has_hn} = 0;
2802 wakaba 1.79
2803     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2804     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2805 wakaba 1.195 }, # check_start
2806 wakaba 1.40 check_end => sub {
2807     my ($self, $item, $element_state) = @_;
2808     $self->_remove_minus_elements ($element_state);
2809     unless ($self->{flag}->{has_hn}) {
2810     $self->{onerror}->(node => $item->{node},
2811 wakaba 1.104 type => 'element missing:hn',
2812 wakaba 1.195 level => $self->{level}->{warn});
2813 wakaba 1.40 }
2814     $self->{flag}->{has_hn} ||= $element_state->{has_hn_original};
2815 wakaba 1.1
2816 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
2817 wakaba 1.195 }, # check_end
2818     }; # header
2819 wakaba 1.1
2820     $Element->{$HTML_NS}->{footer} = {
2821 wakaba 1.153 status => FEATURE_HTML5_LC,
2822 wakaba 1.72 %HTMLFlowContentChecker,
2823 wakaba 1.40 check_start => sub {
2824     my ($self, $item, $element_state) = @_;
2825     $self->_add_minus_elements ($element_state,
2826 wakaba 1.177 {$HTML_NS => {header => 1, footer => 1}},
2827 wakaba 1.58 $HTMLSectioningContent,
2828 wakaba 1.57 $HTMLHeadingContent);
2829 wakaba 1.79
2830     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2831     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2832 wakaba 1.40 },
2833     check_end => sub {
2834     my ($self, $item, $element_state) = @_;
2835     $self->_remove_minus_elements ($element_state);
2836 wakaba 1.1
2837 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
2838 wakaba 1.1 },
2839     };
2840    
2841     $Element->{$HTML_NS}->{address} = {
2842 wakaba 1.72 %HTMLFlowContentChecker,
2843 wakaba 1.187 status => FEATURE_HTML5_REC,
2844 wakaba 1.110 check_attrs => $GetHTMLAttrsChecker->({
2845     ## TODO: add test
2846     #align => $GetHTMLEnumeratedAttrChecker->({
2847     # left => 1, center => 1, right => 1, justify => 1,
2848     #}),
2849     }, {
2850 wakaba 1.49 %HTMLAttrStatus,
2851 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2852 wakaba 1.61 align => FEATURE_HTML2X_RFC,
2853 wakaba 1.187 lang => FEATURE_HTML5_REC,
2854 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2855     sdapref => FEATURE_HTML20_RFC,
2856 wakaba 1.49 }),
2857 wakaba 1.40 check_start => sub {
2858     my ($self, $item, $element_state) = @_;
2859 wakaba 1.177 $self->_add_minus_elements
2860     ($element_state,
2861     {$HTML_NS => {header => 1, footer => 1, address => 1}},
2862     $HTMLSectioningContent, $HTMLHeadingContent);
2863 wakaba 1.79
2864     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2865     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2866 wakaba 1.40 },
2867     check_end => sub {
2868     my ($self, $item, $element_state) = @_;
2869     $self->_remove_minus_elements ($element_state);
2870 wakaba 1.29
2871 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
2872 wakaba 1.29 },
2873 wakaba 1.1 };
2874    
2875     $Element->{$HTML_NS}->{p} = {
2876 wakaba 1.40 %HTMLPhrasingContentChecker,
2877 wakaba 1.187 status => FEATURE_HTML5_REC,
2878 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2879     align => $GetHTMLEnumeratedAttrChecker->({
2880     left => 1, center => 1, right => 1, justify => 1,
2881     }),
2882     }, {
2883 wakaba 1.49 %HTMLAttrStatus,
2884 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
2885 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
2886 wakaba 1.187 lang => FEATURE_HTML5_REC,
2887 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
2888 wakaba 1.49 }),
2889 wakaba 1.1 };
2890    
2891     $Element->{$HTML_NS}->{hr} = {
2892 wakaba 1.40 %HTMLEmptyChecker,
2893 wakaba 1.187 status => FEATURE_HTML5_REC,
2894 wakaba 1.70 check_attrs => $GetHTMLAttrsChecker->({
2895     ## TODO: HTML4 |align|, |noshade|, |size|, |width|
2896     }, {
2897 wakaba 1.49 %HTMLAttrStatus,
2898     %HTMLM12NCommonAttrStatus,
2899     align => FEATURE_M12N10_REC_DEPRECATED,
2900 wakaba 1.187 lang => FEATURE_HTML5_REC,
2901 wakaba 1.49 noshade => FEATURE_M12N10_REC_DEPRECATED,
2902 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
2903 wakaba 1.49 size => FEATURE_M12N10_REC_DEPRECATED,
2904     width => FEATURE_M12N10_REC_DEPRECATED,
2905     }),
2906 wakaba 1.1 };
2907    
2908     $Element->{$HTML_NS}->{br} = {
2909 wakaba 1.40 %HTMLEmptyChecker,
2910 wakaba 1.187 status => FEATURE_HTML5_REC,
2911 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
2912     clear => $GetHTMLEnumeratedAttrChecker->({
2913     left => 1, all => 1, right => 1, none => 1,
2914     }),
2915     }, {
2916 wakaba 1.49 %HTMLAttrStatus,
2917 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
2918 wakaba 1.49 clear => FEATURE_M12N10_REC_DEPRECATED,
2919 wakaba 1.187 id => FEATURE_HTML5_REC,
2920 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
2921 wakaba 1.187 style => FEATURE_HTML5_REC,
2922     title => FEATURE_HTML5_REC,
2923 wakaba 1.49 }),
2924 wakaba 1.29 ## NOTE: Blank line MUST NOT be used for presentation purpose.
2925     ## (This requirement is semantic so that we cannot check.)
2926 wakaba 1.1 };
2927    
2928     $Element->{$HTML_NS}->{dialog} = {
2929 wakaba 1.153 status => FEATURE_HTML5_WD,
2930 wakaba 1.40 %HTMLChecker,
2931     check_start => sub {
2932     my ($self, $item, $element_state) = @_;
2933     $element_state->{phase} = 'before dt';
2934 wakaba 1.79
2935     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
2936     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
2937 wakaba 1.40 },
2938     check_child_element => sub {
2939     my ($self, $item, $child_el, $child_nsuri, $child_ln,
2940     $child_is_transparent, $element_state) = @_;
2941 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
2942     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
2943 wakaba 1.40 $self->{onerror}->(node => $child_el,
2944     type => 'element not allowed:minus',
2945 wakaba 1.104 level => $self->{level}->{must});
2946 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
2947     #
2948     } elsif ($element_state->{phase} eq 'before dt') {
2949     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2950     $element_state->{phase} = 'before dd';
2951     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2952     $self->{onerror}
2953 wakaba 1.104 ->(node => $child_el, type => 'ps element missing',
2954     text => 'dt',
2955     level => $self->{level}->{must});
2956 wakaba 1.40 $element_state->{phase} = 'before dt';
2957     } else {
2958 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2959     level => $self->{level}->{must});
2960 wakaba 1.40 }
2961     } elsif ($element_state->{phase} eq 'before dd') {
2962     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
2963     $element_state->{phase} = 'before dt';
2964     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
2965     $self->{onerror}
2966 wakaba 1.104 ->(node => $child_el, type => 'ps element missing',
2967     text => 'dd',
2968     level => $self->{level}->{must});
2969 wakaba 1.40 $element_state->{phase} = 'before dd';
2970     } else {
2971 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
2972     level => $self->{level}->{must});
2973 wakaba 1.1 }
2974 wakaba 1.40 } else {
2975     die "check_child_element: Bad |dialog| phase: $element_state->{phase}";
2976     }
2977     },
2978     check_child_text => sub {
2979     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
2980     if ($has_significant) {
2981 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
2982     level => $self->{level}->{must});
2983 wakaba 1.1 }
2984 wakaba 1.40 },
2985     check_end => sub {
2986     my ($self, $item, $element_state) = @_;
2987     if ($element_state->{phase} eq 'before dd') {
2988     $self->{onerror}->(node => $item->{node},
2989 wakaba 1.104 type => 'child element missing',
2990     text => 'dd',
2991     level => $self->{level}->{must});
2992 wakaba 1.1 }
2993 wakaba 1.40
2994     $HTMLChecker{check_end}->(@_);
2995 wakaba 1.1 },
2996     };
2997    
2998     $Element->{$HTML_NS}->{pre} = {
2999 wakaba 1.40 %HTMLPhrasingContentChecker,
3000 wakaba 1.187 status => FEATURE_HTML5_REC,
3001 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
3002     width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
3003     }, {
3004 wakaba 1.49 %HTMLAttrStatus,
3005 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3006 wakaba 1.187 lang => FEATURE_HTML5_REC,
3007 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3008 wakaba 1.49 width => FEATURE_M12N10_REC_DEPRECATED,
3009     }),
3010 wakaba 1.101 check_end => sub {
3011     my ($self, $item, $element_state) = @_;
3012    
3013     ## TODO: Flag to enable/disable IDL checking?
3014 wakaba 1.145 my $class = $item->{node}->get_attribute_ns (undef, 'class');
3015 wakaba 1.102 if ($class =~ /\bidl(?>-code)?\b/) { ## TODO: use classList.has
3016     ## NOTE: pre.idl: WHATWG, XHR, Selectors API, CSSOM specs
3017     ## NOTE: pre.code > code.idl-code: WebIDL spec
3018     ## NOTE: pre.idl-code: DOM1 spec
3019     ## NOTE: div.idl-code > pre: DOM, ProgressEvent specs
3020     ## NOTE: pre.schema: ReSpec-generated specs
3021 wakaba 1.101 $self->{onsubdoc}->({s => $item->{node}->text_content,
3022     container_node => $item->{node},
3023     media_type => 'text/x-webidl',
3024     is_char_string => 1});
3025     }
3026    
3027 wakaba 1.110 $HTMLPhrasingContentChecker{check_end}->(@_);
3028 wakaba 1.101 },
3029 wakaba 1.1 };
3030    
3031     $Element->{$HTML_NS}->{ol} = {
3032 wakaba 1.40 %HTMLChecker,
3033 wakaba 1.153 status => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3034 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
3035 wakaba 1.68 compact => $GetHTMLBooleanAttrChecker->('compact'),
3036 wakaba 1.69 reversed => $GetHTMLBooleanAttrChecker->('reversed'),
3037 wakaba 1.1 start => $HTMLIntegerAttrChecker,
3038 wakaba 1.69 ## TODO: HTML4 |type|
3039 wakaba 1.49 }, {
3040     %HTMLAttrStatus,
3041 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3042 wakaba 1.61 align => FEATURE_HTML2X_RFC,
3043 wakaba 1.49 compact => FEATURE_M12N10_REC_DEPRECATED,
3044 wakaba 1.187 lang => FEATURE_HTML5_REC,
3045 wakaba 1.153 reversed => FEATURE_HTML5_WD,
3046 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3047 wakaba 1.153 #start => FEATURE_HTML5_WD | FEATURE_M12N10_REC_DEPRECATED,
3048     start => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
3049 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
3050 wakaba 1.1 }),
3051 wakaba 1.40 check_child_element => sub {
3052     my ($self, $item, $child_el, $child_nsuri, $child_ln,
3053     $child_is_transparent, $element_state) = @_;
3054 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
3055     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
3056 wakaba 1.40 $self->{onerror}->(node => $child_el,
3057     type => 'element not allowed:minus',
3058 wakaba 1.104 level => $self->{level}->{must});
3059 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
3060     #
3061     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'li') {
3062     #
3063     } else {
3064 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
3065     level => $self->{level}->{must});
3066 wakaba 1.1 }
3067 wakaba 1.40 },
3068     check_child_text => sub {
3069     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
3070     if ($has_significant) {
3071 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
3072     level => $self->{level}->{must});
3073 wakaba 1.1 }
3074     },
3075     };
3076    
3077     $Element->{$HTML_NS}->{ul} = {
3078 wakaba 1.40 %{$Element->{$HTML_NS}->{ol}},
3079 wakaba 1.187 status => FEATURE_HTML5_REC,
3080 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
3081     compact => $GetHTMLBooleanAttrChecker->('compact'),
3082 wakaba 1.69 ## TODO: HTML4 |type|
3083     ## TODO: sdaform, align
3084 wakaba 1.68 }, {
3085 wakaba 1.49 %HTMLAttrStatus,
3086 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3087 wakaba 1.61 align => FEATURE_HTML2X_RFC,
3088 wakaba 1.49 compact => FEATURE_M12N10_REC_DEPRECATED,
3089 wakaba 1.187 lang => FEATURE_HTML5_REC,
3090 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3091 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
3092     }),
3093 wakaba 1.1 };
3094    
3095 wakaba 1.64 $Element->{$HTML_NS}->{dir} = {
3096     ## TODO: %block; is not allowed [HTML4] ## TODO: Empty list allowed?
3097     %{$Element->{$HTML_NS}->{ul}},
3098     status => FEATURE_M12N10_REC_DEPRECATED,
3099 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
3100     compact => $GetHTMLBooleanAttrChecker->('compact'),
3101     }, {
3102 wakaba 1.64 %HTMLAttrStatus,
3103     %HTMLM12NCommonAttrStatus,
3104     align => FEATURE_HTML2X_RFC,
3105     compact => FEATURE_M12N10_REC_DEPRECATED,
3106 wakaba 1.187 lang => FEATURE_HTML5_REC,
3107 wakaba 1.64 sdaform => FEATURE_HTML20_RFC,
3108     sdapref => FEATURE_HTML20_RFC,
3109     }),
3110     };
3111    
3112 wakaba 1.1 $Element->{$HTML_NS}->{li} = {
3113 wakaba 1.72 %HTMLFlowContentChecker,
3114 wakaba 1.187 status => FEATURE_HTML5_REC,
3115 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
3116 wakaba 1.69 ## TODO: HTML4 |type|
3117 wakaba 1.49 value => sub {
3118 wakaba 1.1 my ($self, $attr) = @_;
3119 wakaba 1.152
3120     my $parent_is_ol;
3121 wakaba 1.1 my $parent = $attr->owner_element->manakai_parent_element;
3122     if (defined $parent) {
3123     my $parent_ns = $parent->namespace_uri;
3124     $parent_ns = '' unless defined $parent_ns;
3125     my $parent_ln = $parent->manakai_local_name;
3126 wakaba 1.152 $parent_is_ol = ($parent_ns eq $HTML_NS and $parent_ln eq 'ol');
3127     }
3128    
3129     unless ($parent_is_ol) {
3130     ## ISSUE: No "MUST" in the spec.
3131     $self->{onerror}->(node => $attr,
3132     type => 'non-ol li value',
3133     level => $self->{level}->{html5_fact});
3134 wakaba 1.1 }
3135 wakaba 1.152
3136 wakaba 1.1 $HTMLIntegerAttrChecker->($self, $attr);
3137 wakaba 1.131 },
3138 wakaba 1.49 }, {
3139     %HTMLAttrStatus,
3140 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3141 wakaba 1.61 align => FEATURE_HTML2X_RFC,
3142 wakaba 1.187 lang => FEATURE_HTML5_REC,
3143 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3144 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
3145 wakaba 1.154 #value => FEATURE_HTML5_LC | FEATURE_XHTMLBASIC11_CR |
3146 wakaba 1.55 # FEATURE_M12N10_REC_DEPRECATED,
3147 wakaba 1.154 value => FEATURE_HTML5_LC | FEATURE_XHTML2_ED |
3148 wakaba 1.82 FEATURE_XHTMLBASIC11_CR | FEATURE_M12N10_REC,
3149 wakaba 1.207 }), # check_attrs
3150 wakaba 1.40 check_child_element => sub {
3151     my ($self, $item, $child_el, $child_nsuri, $child_ln,
3152     $child_is_transparent, $element_state) = @_;
3153 wakaba 1.207 if (0) {
3154     ## XXXTODO: In <dir> element, then ...
3155 wakaba 1.40 $HTMLPhrasingContentChecker{check_child_element}->(@_);
3156     } else {
3157 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
3158 wakaba 1.40 }
3159 wakaba 1.207 }, # check_child_element
3160 wakaba 1.40 check_child_text => sub {
3161     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
3162 wakaba 1.207 if (0) {
3163     ## XXXTODO: In <dir> element, then ...
3164 wakaba 1.40 $HTMLPhrasingContentChecker{check_child_text}->(@_);
3165 wakaba 1.1 } else {
3166 wakaba 1.72 $HTMLFlowContentChecker{check_child_text}->(@_);
3167 wakaba 1.1 }
3168 wakaba 1.207 }, # check_child_text
3169     }; # li
3170 wakaba 1.1
3171     $Element->{$HTML_NS}->{dl} = {
3172 wakaba 1.40 %HTMLChecker,
3173 wakaba 1.187 status => FEATURE_HTML5_REC,
3174 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
3175     compact => $GetHTMLBooleanAttrChecker->('compact'),
3176     }, {
3177 wakaba 1.49 %HTMLAttrStatus,
3178 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3179 wakaba 1.49 compact => FEATURE_M12N10_REC_DEPRECATED,
3180 wakaba 1.187 lang => FEATURE_HTML5_REC,
3181 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3182     sdapref => FEATURE_HTML20_RFC,
3183 wakaba 1.49 type => FEATURE_M12N10_REC_DEPRECATED,
3184     }),
3185 wakaba 1.40 check_start => sub {
3186     my ($self, $item, $element_state) = @_;
3187     $element_state->{phase} = 'before dt';
3188 wakaba 1.79
3189     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3190     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3191 wakaba 1.40 },
3192     check_child_element => sub {
3193     my ($self, $item, $child_el, $child_nsuri, $child_ln,
3194     $child_is_transparent, $element_state) = @_;
3195 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
3196     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
3197 wakaba 1.40 $self->{onerror}->(node => $child_el,
3198     type => 'element not allowed:minus',
3199 wakaba 1.104 level => $self->{level}->{must});
3200 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
3201     #
3202     } elsif ($element_state->{phase} eq 'in dds') {
3203     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
3204     #$element_state->{phase} = 'in dds';
3205     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
3206     $element_state->{phase} = 'in dts';
3207     } else {
3208 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
3209     level => $self->{level}->{must});
3210 wakaba 1.40 }
3211     } elsif ($element_state->{phase} eq 'in dts') {
3212     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
3213     #$element_state->{phase} = 'in dts';
3214     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
3215     $element_state->{phase} = 'in dds';
3216     } else {
3217 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
3218     level => $self->{level}->{must});
3219 wakaba 1.40 }
3220     } elsif ($element_state->{phase} eq 'before dt') {
3221     if ($child_nsuri eq $HTML_NS and $child_ln eq 'dt') {
3222     $element_state->{phase} = 'in dts';
3223     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'dd') {
3224     $self->{onerror}
3225 wakaba 1.104 ->(node => $child_el, type => 'ps element missing',
3226     text => 'dt',
3227     level => $self->{level}->{must});
3228 wakaba 1.40 $element_state->{phase} = 'in dds';
3229     } else {
3230 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
3231     level => $self->{level}->{must});
3232 wakaba 1.1 }
3233 wakaba 1.40 } else {
3234     die "check_child_element: Bad |dl| phase: $element_state->{phase}";
3235 wakaba 1.1 }
3236 wakaba 1.40 },
3237     check_child_text => sub {
3238     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
3239     if ($has_significant) {
3240 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
3241     level => $self->{level}->{must});
3242 wakaba 1.40 }
3243     },
3244     check_end => sub {
3245     my ($self, $item, $element_state) = @_;
3246     if ($element_state->{phase} eq 'in dts') {
3247     $self->{onerror}->(node => $item->{node},
3248 wakaba 1.104 type => 'child element missing',
3249     text => 'dd',
3250     level => $self->{level}->{must});
3251 wakaba 1.1 }
3252    
3253 wakaba 1.40 $HTMLChecker{check_end}->(@_);
3254 wakaba 1.1 },
3255     };
3256    
3257     $Element->{$HTML_NS}->{dt} = {
3258 wakaba 1.40 %HTMLPhrasingContentChecker,
3259 wakaba 1.187 status => FEATURE_HTML5_REC,
3260 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3261     %HTMLAttrStatus,
3262 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3263 wakaba 1.187 lang => FEATURE_HTML5_REC,
3264 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3265 wakaba 1.49 }),
3266 wakaba 1.1 };
3267    
3268     $Element->{$HTML_NS}->{dd} = {
3269 wakaba 1.72 %HTMLFlowContentChecker,
3270 wakaba 1.187 status => FEATURE_HTML5_REC,
3271 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3272     %HTMLAttrStatus,
3273 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3274 wakaba 1.187 lang => FEATURE_HTML5_REC,
3275 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3276 wakaba 1.49 }),
3277 wakaba 1.1 };
3278    
3279     $Element->{$HTML_NS}->{a} = {
3280 wakaba 1.123 %HTMLTransparentChecker,
3281 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3282 wakaba 1.40 check_attrs => sub {
3283     my ($self, $item, $element_state) = @_;
3284 wakaba 1.1 my %attr;
3285 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
3286 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
3287     $attr_ns = '' unless defined $attr_ns;
3288     my $attr_ln = $attr->manakai_local_name;
3289     my $checker;
3290 wakaba 1.73 my $status;
3291 wakaba 1.1 if ($attr_ns eq '') {
3292 wakaba 1.73 $status = {
3293     %HTMLAttrStatus,
3294 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3295 wakaba 1.176 accesskey => FEATURE_M12N10_REC | FEATURE_HTML5_FD,
3296 wakaba 1.73 charset => FEATURE_M12N10_REC,
3297 wakaba 1.82 coords => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3298 wakaba 1.73 cryptopts => FEATURE_RFC2659,
3299     dn => FEATURE_RFC2659,
3300 wakaba 1.154 href => FEATURE_HTML5_WD | FEATURE_RDFA_REC | FEATURE_XHTML2_ED |
3301 wakaba 1.153 FEATURE_M12N10_REC,
3302     hreflang => FEATURE_HTML5_WD | FEATURE_XHTML2_ED |
3303     FEATURE_M12N10_REC,
3304 wakaba 1.187 lang => FEATURE_HTML5_REC,
3305 wakaba 1.153 media => FEATURE_HTML5_WD | FEATURE_XHTML2_ED,
3306 wakaba 1.73 methods => FEATURE_HTML20_RFC,
3307     name => FEATURE_M12N10_REC_DEPRECATED,
3308     nonce => FEATURE_RFC2659,
3309     onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3310     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3311 wakaba 1.153 ping => FEATURE_HTML5_WD,
3312 wakaba 1.154 rel => FEATURE_HTML5_WD | FEATURE_RDFA_REC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3313     rev => FEATURE_RDFA_REC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3314 wakaba 1.73 sdapref => FEATURE_HTML20_RFC,
3315 wakaba 1.82 shape => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3316 wakaba 1.73 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
3317 wakaba 1.153 target => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3318     type => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
3319 wakaba 1.73 urn => FEATURE_HTML20_RFC,
3320     }->{$attr_ln};
3321    
3322 wakaba 1.1 $checker = {
3323 wakaba 1.91 charset => sub {
3324     my ($self, $attr) = @_;
3325     $HTMLCharsetChecker->($attr->value, @_);
3326     },
3327 wakaba 1.70 ## TODO: HTML4 |coords|
3328 wakaba 1.1 target => $HTMLTargetAttrChecker,
3329     href => $HTMLURIAttrChecker,
3330     ping => $HTMLSpaceURIsAttrChecker,
3331 wakaba 1.40 rel => sub { $HTMLLinkTypesAttrChecker->(1, $item, @_) },
3332 wakaba 1.92 rev => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->(),
3333 wakaba 1.70 ## TODO: HTML4 |shape|
3334 wakaba 1.1 media => $HTMLMQAttrChecker,
3335 wakaba 1.70 ## TODO: HTML4/XHTML1 |name|
3336 wakaba 1.1 hreflang => $HTMLLanguageTagAttrChecker,
3337     type => $HTMLIMTAttrChecker,
3338     }->{$attr_ln};
3339     if ($checker) {
3340     $attr{$attr_ln} = $attr;
3341 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
3342     $attr_ln !~ /[A-Z]/) {
3343 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
3344     $status = $HTMLDatasetAttrStatus;
3345 wakaba 1.1 } else {
3346     $checker = $HTMLAttrChecker->{$attr_ln};
3347     }
3348     }
3349     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
3350     || $AttrChecker->{$attr_ns}->{''};
3351 wakaba 1.82 $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
3352     || $AttrStatus->{$attr_ns}->{''};
3353     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
3354 wakaba 1.62
3355 wakaba 1.1 if ($checker) {
3356 wakaba 1.66 $checker->($self, $attr, $item, $element_state) if ref $checker;
3357 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
3358 wakaba 1.54 #
3359 wakaba 1.1 } else {
3360 wakaba 1.104 $self->{onerror}->(node => $attr,
3361     type => 'unknown attribute',
3362     level => $self->{level}->{uncertain});
3363 wakaba 1.50 ## ISSUE: No conformance createria for unknown attributes in the spec
3364 wakaba 1.1 }
3365 wakaba 1.49
3366 wakaba 1.82 $self->_attr_status_info ($attr, $status);
3367 wakaba 1.1 }
3368    
3369 wakaba 1.40 $element_state->{in_a_href_original} = $self->{flag}->{in_a_href};
3370 wakaba 1.4 if (defined $attr{href}) {
3371     $self->{has_hyperlink_element} = 1;
3372 wakaba 1.40 $self->{flag}->{in_a_href} = 1;
3373 wakaba 1.4 } else {
3374 wakaba 1.1 for (qw/target ping rel media hreflang type/) {
3375     if (defined $attr{$_}) {
3376     $self->{onerror}->(node => $attr{$_},
3377 wakaba 1.104 type => 'attribute not allowed',
3378     level => $self->{level}->{must});
3379 wakaba 1.1 }
3380     }
3381     }
3382 wakaba 1.66
3383     $element_state->{uri_info}->{href}->{type}->{hyperlink} = 1;
3384 wakaba 1.1 },
3385 wakaba 1.40 check_start => sub {
3386     my ($self, $item, $element_state) = @_;
3387     $self->_add_minus_elements ($element_state, $HTMLInteractiveContent);
3388 wakaba 1.79
3389     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3390     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3391 wakaba 1.40 },
3392     check_end => sub {
3393     my ($self, $item, $element_state) = @_;
3394     $self->_remove_minus_elements ($element_state);
3395 wakaba 1.59 delete $self->{flag}->{in_a_href}
3396     unless $element_state->{in_a_href_original};
3397 wakaba 1.1
3398 wakaba 1.123 $HTMLTransparentChecker{check_end}->(@_);
3399 wakaba 1.1 },
3400     };
3401    
3402     $Element->{$HTML_NS}->{q} = {
3403 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3404 wakaba 1.40 %HTMLPhrasingContentChecker,
3405     check_attrs => $GetHTMLAttrsChecker->({
3406 wakaba 1.50 cite => $HTMLURIAttrChecker,
3407     }, {
3408 wakaba 1.49 %HTMLAttrStatus,
3409 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3410 wakaba 1.153 cite => FEATURE_HTML5_AT_RISK | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3411 wakaba 1.187 lang => FEATURE_HTML5_REC,
3412 wakaba 1.61 sdapref => FEATURE_HTML2X_RFC,
3413     sdasuff => FEATURE_HTML2X_RFC,
3414 wakaba 1.1 }),
3415 wakaba 1.66 check_start => sub {
3416     my ($self, $item, $element_state) = @_;
3417    
3418     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
3419 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3420     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3421 wakaba 1.66 },
3422 wakaba 1.1 };
3423 wakaba 1.75 ## TODO: "Quotation punctuation (such as quotation marks), if any, must be
3424     ## placed inside the <code>q</code> element." Though we cannot test the
3425     ## element against this requirement since it incluides a semantic bit,
3426     ## it might be possible to inform of the existence of quotation marks OUTSIDE
3427     ## the |q| element.
3428 wakaba 1.1
3429     $Element->{$HTML_NS}->{cite} = {
3430 wakaba 1.40 %HTMLPhrasingContentChecker,
3431 wakaba 1.187 status => FEATURE_HTML5_REC,
3432 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3433     %HTMLAttrStatus,
3434 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3435 wakaba 1.187 lang => FEATURE_HTML5_REC,
3436 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3437 wakaba 1.49 }),
3438 wakaba 1.1 };
3439    
3440     $Element->{$HTML_NS}->{em} = {
3441 wakaba 1.40 %HTMLPhrasingContentChecker,
3442 wakaba 1.187 status => FEATURE_HTML5_REC,
3443 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3444     %HTMLAttrStatus,
3445 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3446 wakaba 1.187 lang => FEATURE_HTML5_REC,
3447 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3448 wakaba 1.49 }),
3449 wakaba 1.1 };
3450    
3451     $Element->{$HTML_NS}->{strong} = {
3452 wakaba 1.40 %HTMLPhrasingContentChecker,
3453 wakaba 1.187 status => FEATURE_HTML5_REC,
3454 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3455     %HTMLAttrStatus,
3456 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3457 wakaba 1.187 lang => FEATURE_HTML5_REC,
3458 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3459 wakaba 1.49 }),
3460 wakaba 1.1 };
3461    
3462     $Element->{$HTML_NS}->{small} = {
3463 wakaba 1.40 %HTMLPhrasingContentChecker,
3464 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
3465 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3466     %HTMLAttrStatus,
3467     %HTMLM12NCommonAttrStatus,
3468 wakaba 1.187 lang => FEATURE_HTML5_REC,
3469 wakaba 1.49 }),
3470 wakaba 1.1 };
3471    
3472 wakaba 1.51 $Element->{$HTML_NS}->{big} = {
3473     %HTMLPhrasingContentChecker,
3474     status => FEATURE_M12N10_REC,
3475     check_attrs => $GetHTMLAttrsChecker->({}, {
3476     %HTMLAttrStatus,
3477     %HTMLM12NCommonAttrStatus,
3478 wakaba 1.187 lang => FEATURE_HTML5_REC,
3479 wakaba 1.51 }),
3480     };
3481    
3482 wakaba 1.38 $Element->{$HTML_NS}->{mark} = {
3483 wakaba 1.187 status => FEATURE_HTML5_WD,
3484 wakaba 1.40 %HTMLPhrasingContentChecker,
3485 wakaba 1.1 };
3486    
3487     $Element->{$HTML_NS}->{dfn} = {
3488 wakaba 1.40 %HTMLPhrasingContentChecker,
3489 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
3490 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3491     %HTMLAttrStatus,
3492 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3493 wakaba 1.187 lang => FEATURE_HTML5_REC,
3494 wakaba 1.49 }),
3495 wakaba 1.40 check_start => sub {
3496     my ($self, $item, $element_state) = @_;
3497     $self->_add_minus_elements ($element_state, {$HTML_NS => {dfn => 1}});
3498 wakaba 1.1
3499 wakaba 1.40 my $node = $item->{node};
3500 wakaba 1.1 my $term = $node->get_attribute_ns (undef, 'title');
3501     unless (defined $term) {
3502     for my $child (@{$node->child_nodes}) {
3503     if ($child->node_type == 1) { # ELEMENT_NODE
3504     if (defined $term) {
3505     undef $term;
3506     last;
3507     } elsif ($child->manakai_local_name eq 'abbr') {
3508     my $nsuri = $child->namespace_uri;
3509     if (defined $nsuri and $nsuri eq $HTML_NS) {
3510     my $attr = $child->get_attribute_node_ns (undef, 'title');
3511     if ($attr) {
3512     $term = $attr->value;
3513     }
3514     }
3515     }
3516     } elsif ($child->node_type == 3 or $child->node_type == 4) {
3517     ## TEXT_NODE or CDATA_SECTION_NODE
3518 wakaba 1.132 if ($child->data =~ /\A[\x09\x0A\x0C\x0D\x20]+\z/) { # Inter-element whitespace
3519 wakaba 1.1 next;
3520     }
3521     undef $term;
3522     last;
3523     }
3524     }
3525     unless (defined $term) {
3526     $term = $node->text_content;
3527     }
3528     }
3529     if ($self->{term}->{$term}) {
3530     push @{$self->{term}->{$term}}, $node;
3531     } else {
3532     $self->{term}->{$term} = [$node];
3533     }
3534 wakaba 1.77 ## ISSUE: The HTML5 definition for the defined term does not work with
3535     ## |ruby| unless |dfn| has |title|.
3536 wakaba 1.79
3537     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
3538     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
3539 wakaba 1.40 },
3540     check_end => sub {
3541     my ($self, $item, $element_state) = @_;
3542     $self->_remove_minus_elements ($element_state);
3543 wakaba 1.1
3544 wakaba 1.40 $HTMLPhrasingContentChecker{check_end}->(@_);
3545 wakaba 1.1 },
3546     };
3547    
3548     $Element->{$HTML_NS}->{abbr} = {
3549 wakaba 1.40 %HTMLPhrasingContentChecker,
3550 wakaba 1.187 status => FEATURE_HTML5_REC,
3551 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3552     %HTMLAttrStatus,
3553 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3554     full => FEATURE_XHTML2_ED,
3555 wakaba 1.187 lang => FEATURE_HTML5_REC,
3556 wakaba 1.49 }),
3557 wakaba 1.77 ## NOTE: "If an abbreviation is pluralised, the expansion's grammatical
3558     ## number (plural vs singular) must match the grammatical number of the
3559     ## contents of the element." Though this can be checked by machine,
3560     ## it requires language-specific knowledge and dictionary, such that
3561     ## we don't support the check of the requirement.
3562     ## ISSUE: Is <abbr title="Cascading Style Sheets">CSS</abbr> conforming?
3563 wakaba 1.49 };
3564    
3565     $Element->{$HTML_NS}->{acronym} = {
3566     %HTMLPhrasingContentChecker,
3567     status => FEATURE_M12N10_REC,
3568     check_attrs => $GetHTMLAttrsChecker->({}, {
3569     %HTMLAttrStatus,
3570     %HTMLM12NCommonAttrStatus,
3571 wakaba 1.187 lang => FEATURE_HTML5_REC,
3572 wakaba 1.49 }),
3573 wakaba 1.1 };
3574    
3575     $Element->{$HTML_NS}->{time} = {
3576 wakaba 1.187 status => FEATURE_HTML5_WD,
3577 wakaba 1.40 %HTMLPhrasingContentChecker,
3578     check_attrs => $GetHTMLAttrsChecker->({
3579 wakaba 1.1 datetime => sub { 1 }, # checked in |checker|
3580 wakaba 1.49 }, {
3581     %HTMLAttrStatus,
3582     %HTMLM12NCommonAttrStatus,
3583 wakaba 1.72 datetime => FEATURE_HTML5_FD,
3584 wakaba 1.201 }), # check_attrs
3585 wakaba 1.213 check_start => sub {
3586     my ($self, $item, $element_state) = @_;
3587     $self->_add_minus_elements ($element_state, {$HTML_NS => {time => 1}});
3588    
3589     $HTMLPhrasingContentChecker{check_start}->(@_);
3590     }, # check_start
3591 wakaba 1.40 check_end => sub {
3592     my ($self, $item, $element_state) = @_;
3593 wakaba 1.213 $self->_remove_minus_elements ($element_state);
3594 wakaba 1.1
3595 wakaba 1.201 ## XXX Maybe we should move this code out somewhere (maybe
3596     ## Message::Date) such that we can reuse this code in other places
3597     ## (e.g. HTMLTimeElement implementation).
3598    
3599     ## "Vaguer moments in time" or "valid date or time string".
3600 wakaba 1.40 my $attr = $item->{node}->get_attribute_node_ns (undef, 'datetime');
3601 wakaba 1.1 my $input;
3602     my $reg_sp;
3603     my $input_node;
3604     if ($attr) {
3605     $input = $attr->value;
3606 wakaba 1.201 $reg_sp = qr/[\x09\x0A\x0C\x0D\x20]/;
3607 wakaba 1.1 $input_node = $attr;
3608     } else {
3609 wakaba 1.40 $input = $item->{node}->text_content;
3610 wakaba 1.201 $reg_sp = qr/\p{WhiteSpace}/;
3611 wakaba 1.40 $input_node = $item->{node};
3612 wakaba 1.1 }
3613    
3614     my $hour;
3615     my $minute;
3616     my $second;
3617     if ($input =~ /
3618     \A
3619 wakaba 1.201 $reg_sp*
3620 wakaba 1.1 ([0-9]+) # 1
3621     (?>
3622     -([0-9]+) # 2
3623 wakaba 1.112 -((?>[0-9]+)) # 3 # Use (?>) such that yyyy-mm-ddhh:mm does not match
3624 wakaba 1.201 $reg_sp*
3625 wakaba 1.1 (?>
3626 wakaba 1.201 (?>
3627     T
3628     $reg_sp*
3629     )?
3630     ([0-9]+) # 4
3631     :([0-9]+) # 5
3632     (?>
3633     :([0-9]+(?>\.[0-9]+)?) # 6
3634     )?
3635     $reg_sp*
3636     (?>
3637     Z
3638     $reg_sp*
3639     |
3640     [+-]([0-9]+):([0-9]+) # 7, 8
3641     $reg_sp*
3642     )?
3643 wakaba 1.1 )?
3644     \z
3645     |
3646     :([0-9]+) # 9
3647 wakaba 1.201 (?:
3648     :([0-9]+(?>\.[0-9]+)?) # 10
3649 wakaba 1.1 )?
3650 wakaba 1.201 $reg_sp*
3651 wakaba 1.112 \z
3652 wakaba 1.1 )
3653     /x) {
3654 wakaba 1.201 my $has_syntax_error;
3655 wakaba 1.1 if (defined $2) { ## YYYY-MM-DD T? hh:mm
3656     if (length $1 != 4 or length $2 != 2 or length $3 != 2 or
3657 wakaba 1.201 (defined $4 and length $4 != 2) or
3658     (defined $5 and length $5 != 2)) {
3659 wakaba 1.1 $self->{onerror}->(node => $input_node,
3660 wakaba 1.104 type => 'dateortime:syntax error',
3661     level => $self->{level}->{must});
3662 wakaba 1.201 $has_syntax_error = 1;
3663 wakaba 1.1 }
3664    
3665     if (1 <= $2 and $2 <= 12) {
3666 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad day',
3667     level => $self->{level}->{must})
3668 wakaba 1.1 if $3 < 1 or
3669     $3 > [0, 31, 29, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31]->[$2];
3670 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad day',
3671     level => $self->{level}->{must})
3672 wakaba 1.1 if $2 == 2 and $3 == 29 and
3673     not ($1 % 400 == 0 or ($1 % 4 == 0 and $1 % 100 != 0));
3674     } else {
3675     $self->{onerror}->(node => $input_node,
3676 wakaba 1.104 type => 'datetime:bad month',
3677     level => $self->{level}->{must});
3678 wakaba 1.1 }
3679 wakaba 1.201 $self->{onerror}->(node => $input_node,
3680     type => 'datetime:bad year',
3681     level => $self->{level}->{must})
3682     if $1 == 0;
3683 wakaba 1.1
3684     ($hour, $minute, $second) = ($4, $5, $6);
3685    
3686     if (defined $7) { ## [+-]hh:mm
3687     if (length $7 != 2 or length $8 != 2) {
3688     $self->{onerror}->(node => $input_node,
3689 wakaba 1.104 type => 'dateortime:syntax error',
3690     level => $self->{level}->{must});
3691 wakaba 1.201 $has_syntax_error = 1;
3692 wakaba 1.1 }
3693    
3694     $self->{onerror}->(node => $input_node,
3695 wakaba 1.104 type => 'datetime:bad timezone hour',
3696     level => $self->{level}->{must})
3697 wakaba 1.1 if $7 > 23;
3698     $self->{onerror}->(node => $input_node,
3699 wakaba 1.104 type => 'datetime:bad timezone minute',
3700     level => $self->{level}->{must})
3701 wakaba 1.1 if $8 > 59;
3702     }
3703     } else { ## hh:mm
3704     if (length $1 != 2 or length $9 != 2) {
3705     $self->{onerror}->(node => $input_node,
3706 wakaba 1.104 type => qq'dateortime:syntax error',
3707     level => $self->{level}->{must});
3708 wakaba 1.201 $has_syntax_error = 1;
3709 wakaba 1.1 }
3710    
3711     ($hour, $minute, $second) = ($1, $9, $10);
3712     }
3713    
3714 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad hour',
3715     level => $self->{level}->{must}) if $hour > 23;
3716     $self->{onerror}->(node => $input_node, type => 'datetime:bad minute',
3717     level => $self->{level}->{must}) if $minute > 59;
3718 wakaba 1.1
3719     if (defined $second) { ## s
3720     ## NOTE: Integer part of second don't have to have length of two.
3721    
3722     if (substr ($second, 0, 1) eq '.') {
3723     $self->{onerror}->(node => $input_node,
3724 wakaba 1.104 type => 'dateortime:syntax error',
3725     level => $self->{level}->{must});
3726 wakaba 1.201 $has_syntax_error = 1;
3727 wakaba 1.1 }
3728    
3729 wakaba 1.104 $self->{onerror}->(node => $input_node, type => 'datetime:bad second',
3730     level => $self->{level}->{must}) if $second >= 60;
3731 wakaba 1.201 }
3732    
3733     unless ($has_syntax_error) {
3734     $input =~ s/\A$reg_sp+//;
3735     $input =~ s/$reg_sp+\z//;
3736     if ($input =~ /$reg_sp+/) {
3737     $self->{onerror}->(node => $input_node,
3738     type => 'dateortime:syntax error',
3739     level => $self->{level}->{must});
3740     }
3741     }
3742 wakaba 1.1 } else {
3743     $self->{onerror}->(node => $input_node,
3744 wakaba 1.104 type => 'dateortime:syntax error',
3745     level => $self->{level}->{must});
3746 wakaba 1.1 }
3747    
3748 wakaba 1.40 $HTMLPhrasingContentChecker{check_end}->(@_);
3749 wakaba 1.201 }, # check_end
3750     }; # time
3751 wakaba 1.1
3752     $Element->{$HTML_NS}->{meter} = { ## TODO: "The recommended way of giving the value is to include it as contents of the element"
3753 wakaba 1.77 ## TODO: value inequalities (HTML5 revision 1463)
3754 wakaba 1.113 ## TODO: content checking
3755     ## TODO: content or value must contain number (rev 2053)
3756 wakaba 1.40 %HTMLPhrasingContentChecker,
3757 wakaba 1.187 status => FEATURE_HTML5_WD,
3758 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
3759 wakaba 1.1 value => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3760     min => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3761     low => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3762     high => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3763     max => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3764     optimum => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
3765 wakaba 1.50 }, {
3766     %HTMLAttrStatus,
3767     high => FEATURE_HTML5_DEFAULT,
3768     low => FEATURE_HTML5_DEFAULT,
3769     max => FEATURE_HTML5_DEFAULT,
3770     min => FEATURE_HTML5_DEFAULT,
3771     optimum => FEATURE_HTML5_DEFAULT,
3772     value => FEATURE_HTML5_DEFAULT,
3773 wakaba 1.211 }), # check_attrs
3774 wakaba 1.213 check_start => sub {
3775     my ($self, $item, $element_state) = @_;
3776     $self->_add_minus_elements ($element_state, {$HTML_NS => {meter => 1}});
3777    
3778     $HTMLPhrasingContentChecker{check_start}->(@_);
3779     }, # check_start
3780 wakaba 1.211 check_end => sub {
3781     my ($self, $item, $element_state) = @_;
3782 wakaba 1.213 $self->_remove_minus_elements ($element_state);
3783 wakaba 1.1
3784 wakaba 1.211 ## XXX Work in progress
3785    
3786     my $tc = $item->{node}->text_content;
3787     my $n1;
3788     my $denominator;
3789     my $n2;
3790     if ($tc =~ s/^([0-9.]+)//) {
3791     $n1 = $1;
3792    
3793     if ($tc =~ s/^[^0-9.]+([0-9.]+)//) {
3794     $n2 = $1;
3795     } elsif ($tc =~ s/^([\x{0025}\x{066A}\x{FE6A}\x{FF05}\x{2030}\x{2031}])//) {
3796     $denominator = $1;
3797     }
3798     }
3799    
3800     if ($tc =~ /[0-9.]/) {
3801     undef $n1;
3802     undef $n2;
3803     }
3804    
3805     $HTMLPhrasingContentChecker{check_end}->(@_);
3806     }, # check_end
3807     }; # meter
3808    
3809     $Element->{$HTML_NS}->{progress} = {
3810 wakaba 1.40 %HTMLPhrasingContentChecker,
3811 wakaba 1.187 status => FEATURE_HTML5_WD,
3812 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
3813 wakaba 1.211 value => sub { }, ## checked in |check_attrs2|
3814     max => sub { }, ## checked in |check_attrs2|
3815 wakaba 1.50 }, {
3816     %HTMLAttrStatus,
3817     max => FEATURE_HTML5_DEFAULT,
3818     value => FEATURE_HTML5_DEFAULT,
3819 wakaba 1.211 }), # check_attrs
3820     check_attrs2 => sub {
3821     my ($self, $item, $element_state) = @_;
3822    
3823     my $max = 1;
3824     my $max_attr = $item->{node}->get_attribute_node_ns (undef, 'max');
3825     if ($max_attr) {
3826     $GetHTMLFloatingPointNumberAttrChecker->(sub {
3827     my $num = $_[0];
3828     $max = $num;
3829     return $num > 0; ## >, not >=
3830     })->($self, $max_attr);
3831     }
3832    
3833     my $value_attr = $item->{node}->get_attribute_node_ns (undef, 'value');
3834     if ($value_attr) {
3835     $self->{onerror}->(node => $value_attr,
3836     type => 'attribute not allowed',
3837     text => 'value',
3838     level => $self->{level}->{should}); # RECOMMENDED
3839    
3840     $GetHTMLFloatingPointNumberAttrChecker->(sub {
3841     my $num = $_[0];
3842    
3843     unless ($num <= $max) {
3844     $self->{onerror}->(node => $value_attr,
3845     type => 'progress value out of range',
3846     value => $max, # XXX document error type
3847     level => $self->{level}->{must});
3848     }
3849    
3850     return $num >= 0; ## >=, not >
3851     })->($self, $value_attr);
3852     }
3853     }, # check_attrs2
3854     # XXX warn if the value from the content is greater than |max|
3855     # attribute value.
3856     # XXX warn if the element content does not contain one or two numbers.
3857 wakaba 1.213 check_start => sub {
3858     my ($self, $item, $element_state) = @_;
3859     $self->_add_minus_elements ($element_state, {$HTML_NS => {progress => 1}});
3860    
3861     $HTMLPhrasingContentChecker{check_start}->(@_);
3862     }, # check_start
3863     check_end => sub {
3864     my ($self, $item, $element_state) = @_;
3865     $self->_remove_minus_elements ($element_state);
3866    
3867     $HTMLPhrasingContentChecker{check_end}->(@_);
3868     }, # check_end
3869 wakaba 1.211 }; # progress
3870 wakaba 1.1
3871     $Element->{$HTML_NS}->{code} = {
3872 wakaba 1.40 %HTMLPhrasingContentChecker,
3873 wakaba 1.187 status => FEATURE_HTML5_REC,
3874 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3875     %HTMLAttrStatus,
3876 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3877 wakaba 1.187 lang => FEATURE_HTML5_REC,
3878 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3879 wakaba 1.49 }),
3880 wakaba 1.1 };
3881    
3882     $Element->{$HTML_NS}->{var} = {
3883 wakaba 1.40 %HTMLPhrasingContentChecker,
3884 wakaba 1.187 status => FEATURE_HTML5_REC,
3885 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3886     %HTMLAttrStatus,
3887 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3888 wakaba 1.187 lang => FEATURE_HTML5_REC,
3889 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3890 wakaba 1.49 }),
3891 wakaba 1.1 };
3892    
3893     $Element->{$HTML_NS}->{samp} = {
3894 wakaba 1.40 %HTMLPhrasingContentChecker,
3895 wakaba 1.187 status => FEATURE_HTML5_REC,
3896 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3897     %HTMLAttrStatus,
3898 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3899 wakaba 1.187 lang => FEATURE_HTML5_REC,
3900 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3901 wakaba 1.49 }),
3902 wakaba 1.1 };
3903    
3904     $Element->{$HTML_NS}->{kbd} = {
3905 wakaba 1.40 %HTMLPhrasingContentChecker,
3906 wakaba 1.187 status => FEATURE_HTML5_REC,
3907 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3908     %HTMLAttrStatus,
3909 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3910 wakaba 1.187 lang => FEATURE_HTML5_REC,
3911 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3912 wakaba 1.49 }),
3913 wakaba 1.1 };
3914    
3915     $Element->{$HTML_NS}->{sub} = {
3916 wakaba 1.40 %HTMLPhrasingContentChecker,
3917 wakaba 1.187 status => FEATURE_HTML5_REC,
3918 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3919     %HTMLAttrStatus,
3920 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3921 wakaba 1.187 lang => FEATURE_HTML5_REC,
3922 wakaba 1.61 sdapref => FEATURE_HTML2X_RFC,
3923 wakaba 1.49 }),
3924 wakaba 1.1 };
3925    
3926 wakaba 1.51 $Element->{$HTML_NS}->{sup} = $Element->{$HTML_NS}->{sub};
3927 wakaba 1.1
3928     $Element->{$HTML_NS}->{span} = {
3929 wakaba 1.40 %HTMLPhrasingContentChecker,
3930 wakaba 1.187 status => FEATURE_HTML5_REC,
3931 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3932     %HTMLAttrStatus,
3933 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
3934 wakaba 1.49 datafld => FEATURE_HTML4_REC_RESERVED,
3935     dataformatas => FEATURE_HTML4_REC_RESERVED,
3936     datasrc => FEATURE_HTML4_REC_RESERVED,
3937 wakaba 1.187 lang => FEATURE_HTML5_REC,
3938 wakaba 1.61 sdaform => FEATURE_HTML2X_RFC,
3939 wakaba 1.49 }),
3940 wakaba 1.1 };
3941    
3942 wakaba 1.205 # XXX Warning for "authors are encouraged to consider whether other
3943     # elements might be more applicable"
3944 wakaba 1.1 $Element->{$HTML_NS}->{i} = {
3945 wakaba 1.40 %HTMLPhrasingContentChecker,
3946 wakaba 1.187 status => FEATURE_HTML5_REC,
3947 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3948     %HTMLAttrStatus,
3949     %HTMLM12NCommonAttrStatus,
3950 wakaba 1.187 lang => FEATURE_HTML5_REC,
3951 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3952 wakaba 1.49 }),
3953 wakaba 1.1 };
3954    
3955 wakaba 1.51 $Element->{$HTML_NS}->{b} = $Element->{$HTML_NS}->{i};
3956    
3957 wakaba 1.61 $Element->{$HTML_NS}->{tt} = {
3958     %HTMLPhrasingContentChecker,
3959     status => FEATURE_M12N10_REC,
3960     check_attrs => $GetHTMLAttrsChecker->({}, {
3961     %HTMLAttrStatus,
3962     %HTMLM12NCommonAttrStatus,
3963 wakaba 1.187 lang => FEATURE_HTML5_REC,
3964 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
3965     }),
3966     };
3967 wakaba 1.51
3968     $Element->{$HTML_NS}->{s} = {
3969 wakaba 1.40 %HTMLPhrasingContentChecker,
3970 wakaba 1.51 status => FEATURE_M12N10_REC_DEPRECATED,
3971 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
3972     %HTMLAttrStatus,
3973     %HTMLM12NCommonAttrStatus,
3974 wakaba 1.187 lang => FEATURE_HTML5_REC,
3975 wakaba 1.49 }),
3976 wakaba 1.1 };
3977    
3978 wakaba 1.51 $Element->{$HTML_NS}->{strike} = $Element->{$HTML_NS}->{s};
3979    
3980     $Element->{$HTML_NS}->{u} = $Element->{$HTML_NS}->{s};
3981    
3982 wakaba 1.1 $Element->{$HTML_NS}->{bdo} = {
3983 wakaba 1.40 %HTMLPhrasingContentChecker,
3984 wakaba 1.187 status => FEATURE_HTML5_REC,
3985 wakaba 1.40 check_attrs => sub {
3986     my ($self, $item, $element_state) = @_;
3987 wakaba 1.49 $GetHTMLAttrsChecker->({}, {
3988     %HTMLAttrStatus,
3989 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
3990     dir => FEATURE_HTML5_REC,
3991     id => FEATURE_HTML5_REC,
3992     style => FEATURE_HTML5_REC,
3993     title => FEATURE_HTML5_REC,
3994     lang => FEATURE_HTML5_REC,
3995 wakaba 1.61 sdapref => FEATURE_HTML2X_RFC,
3996     sdasuff => FEATURE_HTML2X_RFC,
3997 wakaba 1.49 })->($self, $item, $element_state);
3998 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'dir')) {
3999     $self->{onerror}->(node => $item->{node},
4000 wakaba 1.104 type => 'attribute missing',
4001     text => 'dir',
4002     level => $self->{level}->{must});
4003 wakaba 1.1 }
4004     },
4005     ## ISSUE: The spec does not directly say that |dir| is a enumerated attr.
4006     };
4007    
4008 wakaba 1.99 $Element->{$HTML_NS}->{ruby} = {
4009     %HTMLPhrasingContentChecker,
4010 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_RUBY_REC,
4011 wakaba 1.99 check_attrs => $GetHTMLAttrsChecker->({}, {
4012     %HTMLAttrStatus,
4013     %HTMLM12NXHTML2CommonAttrStatus, # XHTML 1.1 & XHTML 2.0 & XHTML+RDFa 1.0
4014 wakaba 1.187 lang => FEATURE_HTML5_REC,
4015 wakaba 1.99 }),
4016     check_start => sub {
4017     my ($self, $item, $element_state) = @_;
4018    
4019     $element_state->{phase} = 'before-rb';
4020     #$element_state->{has_sig}
4021 wakaba 1.100
4022     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4023     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4024 wakaba 1.99 },
4025     ## NOTE: (phrasing, (rt | (rp, rt, rp)))+
4026     check_child_element => sub {
4027     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4028     $child_is_transparent, $element_state) = @_;
4029 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4030     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4031 wakaba 1.99 $self->{onerror}->(node => $child_el,
4032     type => 'element not allowed:minus',
4033 wakaba 1.104 level => $self->{level}->{must});
4034 wakaba 1.99 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4035     #
4036     } elsif ($element_state->{phase} eq 'before-rb') {
4037     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
4038     $element_state->{phase} = 'in-rb';
4039     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
4040     $self->{onerror}->(node => $child_el,
4041 wakaba 1.104 level => $self->{level}->{should},
4042     type => 'no significant content before');
4043 wakaba 1.99 $element_state->{phase} = 'after-rt';
4044     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
4045     $self->{onerror}->(node => $child_el,
4046 wakaba 1.104 level => $self->{level}->{should},
4047     type => 'no significant content before');
4048 wakaba 1.99 $element_state->{phase} = 'after-rp1';
4049     } else {
4050     $self->{onerror}->(node => $child_el,
4051 wakaba 1.104 type => 'element not allowed:ruby base',
4052     level => $self->{level}->{must});
4053 wakaba 1.99 $element_state->{phase} = 'in-rb';
4054     }
4055     } elsif ($element_state->{phase} eq 'in-rb') {
4056     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
4057     #$element_state->{phase} = 'in-rb';
4058     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
4059     unless ($element_state->{has_significant}) {
4060     $self->{onerror}->(node => $child_el,
4061 wakaba 1.104 level => $self->{level}->{should},
4062     type => 'no significant content before');
4063 wakaba 1.99 }
4064     $element_state->{phase} = 'after-rt';
4065     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
4066     unless ($element_state->{has_significant}) {
4067     $self->{onerror}->(node => $child_el,
4068 wakaba 1.104 level => $self->{level}->{should},
4069     type => 'no significant content before');
4070 wakaba 1.99 }
4071     $element_state->{phase} = 'after-rp1';
4072     } else {
4073     $self->{onerror}->(node => $child_el,
4074 wakaba 1.104 type => 'element not allowed:ruby base',
4075     level => $self->{level}->{must});
4076 wakaba 1.99 #$element_state->{phase} = 'in-rb';
4077     }
4078     } elsif ($element_state->{phase} eq 'after-rt') {
4079     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
4080     if ($element_state->{has_significant}) {
4081     $element_state->{has_sig} = 1;
4082     delete $element_state->{has_significant};
4083     }
4084     $element_state->{phase} = 'in-rb';
4085     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
4086     $self->{onerror}->(node => $child_el,
4087 wakaba 1.104 level => $self->{level}->{should},
4088     type => 'no significant content before');
4089 wakaba 1.99 $element_state->{phase} = 'after-rp1';
4090     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
4091     $self->{onerror}->(node => $child_el,
4092 wakaba 1.104 level => $self->{level}->{should},
4093     type => 'no significant content before');
4094 wakaba 1.99 #$element_state->{phase} = 'after-rt';
4095     } else {
4096     $self->{onerror}->(node => $child_el,
4097 wakaba 1.104 type => 'element not allowed:ruby base',
4098     level => $self->{level}->{must});
4099 wakaba 1.99 if ($element_state->{has_significant}) {
4100     $element_state->{has_sig} = 1;
4101     delete $element_state->{has_significant};
4102     }
4103     $element_state->{phase} = 'in-rb';
4104     }
4105     } elsif ($element_state->{phase} eq 'after-rp1') {
4106     if ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
4107     $element_state->{phase} = 'after-rp-rt';
4108     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
4109     $self->{onerror}->(node => $child_el,
4110 wakaba 1.104 type => 'ps element missing',
4111     text => 'rt',
4112     level => $self->{level}->{must});
4113 wakaba 1.99 $element_state->{phase} = 'after-rp2';
4114     } else {
4115     $self->{onerror}->(node => $child_el,
4116 wakaba 1.104 type => 'ps element missing',
4117     text => 'rt',
4118     level => $self->{level}->{must});
4119 wakaba 1.99 $self->{onerror}->(node => $child_el,
4120 wakaba 1.104 type => 'ps element missing',
4121     text => 'rp',
4122     level => $self->{level}->{must});
4123 wakaba 1.99 unless ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
4124     $self->{onerror}->(node => $child_el,
4125 wakaba 1.104 type => 'element not allowed:ruby base',
4126     level => $self->{level}->{must});
4127 wakaba 1.99 }
4128     if ($element_state->{has_significant}) {
4129     $element_state->{has_sig} = 1;
4130     delete $element_state->{has_significant};
4131     }
4132     $element_state->{phase} = 'in-rb';
4133     }
4134     } elsif ($element_state->{phase} eq 'after-rp-rt') {
4135     if ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
4136     $element_state->{phase} = 'after-rp2';
4137     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
4138     $self->{onerror}->(node => $child_el,
4139 wakaba 1.104 type => 'ps element missing',
4140     text => 'rp',
4141     level => $self->{level}->{must});
4142 wakaba 1.99 $self->{onerror}->(node => $child_el,
4143 wakaba 1.104 level => $self->{level}->{should},
4144     type => 'no significant content before');
4145 wakaba 1.99 $element_state->{phase} = 'after-rt';
4146     } else {
4147     $self->{onerror}->(node => $child_el,
4148 wakaba 1.104 type => 'ps element missing',
4149     text => 'rp',
4150     level => $self->{level}->{must});
4151 wakaba 1.99 unless ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
4152     $self->{onerror}->(node => $child_el,
4153 wakaba 1.104 type => 'element not allowed:ruby base',
4154     level => $self->{level}->{must});
4155 wakaba 1.99 }
4156     if ($element_state->{has_significant}) {
4157     $element_state->{has_sig} = 1;
4158     delete $element_state->{has_significant};
4159     }
4160     $element_state->{phase} = 'in-rb';
4161     }
4162     } elsif ($element_state->{phase} eq 'after-rp2') {
4163     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
4164     if ($element_state->{has_significant}) {
4165     $element_state->{has_sig} = 1;
4166     delete $element_state->{has_significant};
4167     }
4168     $element_state->{phase} = 'in-rb';
4169     } elsif ($child_ln eq 'rt' and $child_nsuri eq $HTML_NS) {
4170     $self->{onerror}->(node => $child_el,
4171 wakaba 1.104 level => $self->{level}->{should},
4172     type => 'no significant content before');
4173 wakaba 1.99 $element_state->{phase} = 'after-rt';
4174     } elsif ($child_ln eq 'rp' and $child_nsuri eq $HTML_NS) {
4175     $self->{onerror}->(node => $child_el,
4176 wakaba 1.104 level => $self->{level}->{should},
4177     type => 'no significant content before');
4178 wakaba 1.99 $element_state->{phase} = 'after-rp1';
4179     } else {
4180     $self->{onerror}->(node => $child_el,
4181 wakaba 1.104 type => 'element not allowed:ruby base',
4182     level => $self->{level}->{must});
4183 wakaba 1.99 if ($element_state->{has_significant}) {
4184     $element_state->{has_sig} = 1;
4185     delete $element_state->{has_significant};
4186     }
4187     $element_state->{phase} = 'in-rb';
4188     }
4189     } else {
4190     die "check_child_element: Bad |ruby| phase: $element_state->{phase}";
4191     }
4192     },
4193     check_child_text => sub {
4194     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4195     if ($has_significant) {
4196     if ($element_state->{phase} eq 'before-rb') {
4197     $element_state->{phase} = 'in-rb';
4198     } elsif ($element_state->{phase} eq 'in-rb') {
4199     #
4200     } elsif ($element_state->{phase} eq 'after-rt' or
4201     $element_state->{phase} eq 'after-rp2') {
4202     $element_state->{phase} = 'in-rb';
4203     } elsif ($element_state->{phase} eq 'after-rp1') {
4204     $self->{onerror}->(node => $child_node,
4205 wakaba 1.104 type => 'ps element missing',
4206     text => 'rt',
4207     level => $self->{level}->{must});
4208 wakaba 1.99 $self->{onerror}->(node => $child_node,
4209 wakaba 1.104 type => 'ps element missing',
4210     text => 'rp',
4211     level => $self->{level}->{must});
4212 wakaba 1.99 $element_state->{phase} = 'in-rb';
4213     } elsif ($element_state->{phase} eq 'after-rp-rt') {
4214     $self->{onerror}->(node => $child_node,
4215 wakaba 1.104 type => 'ps element missing',
4216     text => 'rp',
4217     level => $self->{level}->{must});
4218 wakaba 1.99 $element_state->{phase} = 'in-rb';
4219     } else {
4220     die "check_child_text: Bad |ruby| phase: $element_state->{phase}";
4221     }
4222     }
4223     },
4224     check_end => sub {
4225     my ($self, $item, $element_state) = @_;
4226     $self->_remove_minus_elements ($element_state);
4227    
4228     if ($element_state->{phase} eq 'before-rb') {
4229     $self->{onerror}->(node => $item->{node},
4230 wakaba 1.104 level => $self->{level}->{should},
4231 wakaba 1.99 type => 'no significant content');
4232     $self->{onerror}->(node => $item->{node},
4233 wakaba 1.104 type => 'element missing',
4234     text => 'rt',
4235     level => $self->{level}->{must});
4236 wakaba 1.99 } elsif ($element_state->{phase} eq 'in-rb') {
4237     unless ($element_state->{has_significant}) {
4238     $self->{onerror}->(node => $item->{node},
4239 wakaba 1.104 level => $self->{level}->{should},
4240     type => 'no significant content at the end');
4241 wakaba 1.99 }
4242     $self->{onerror}->(node => $item->{node},
4243 wakaba 1.104 type => 'element missing',
4244     text => 'rt',
4245     level => $self->{level}->{must});
4246 wakaba 1.99 } elsif ($element_state->{phase} eq 'after-rt' or
4247     $element_state->{phase} eq 'after-rp2') {
4248     #
4249     } elsif ($element_state->{phase} eq 'after-rp1') {
4250     $self->{onerror}->(node => $item->{node},
4251 wakaba 1.104 type => 'element missing',
4252     text => 'rt',
4253     level => $self->{level}->{must});
4254 wakaba 1.99 $self->{onerror}->(node => $item->{node},
4255 wakaba 1.104 type => 'element missing',
4256     text => 'rp',
4257     level => $self->{level}->{must});
4258 wakaba 1.99 } elsif ($element_state->{phase} eq 'after-rp-rt') {
4259     $self->{onerror}->(node => $item->{node},
4260 wakaba 1.104 type => 'element missing',
4261     text => 'rp',
4262     level => $self->{level}->{must});
4263 wakaba 1.99 } else {
4264     die "check_child_text: Bad |ruby| phase: $element_state->{phase}";
4265     }
4266    
4267     ## NOTE: A modified version of |check_end| of %AnyChecker.
4268     if ($element_state->{has_significant} or $element_state->{has_sig}) {
4269     $item->{real_parent_state}->{has_significant} = 1;
4270     }
4271     },
4272     };
4273    
4274     $Element->{$HTML_NS}->{rt} = {
4275     %HTMLPhrasingContentChecker,
4276 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_RUBY_REC,
4277 wakaba 1.99 check_attrs => $GetHTMLAttrsChecker->({}, {
4278     %HTMLAttrStatus,
4279     %HTMLM12NXHTML2CommonAttrStatus,
4280 wakaba 1.187 lang => FEATURE_HTML5_REC,
4281 wakaba 1.99 }),
4282     };
4283    
4284     $Element->{$HTML_NS}->{rp} = {
4285 wakaba 1.171 %HTMLPhrasingContentChecker,
4286 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_RUBY_REC,
4287 wakaba 1.99 check_attrs => $GetHTMLAttrsChecker->({}, {
4288     %HTMLAttrStatus,
4289     %HTMLM12NXHTML2CommonAttrStatus,
4290 wakaba 1.187 lang => FEATURE_HTML5_REC,
4291 wakaba 1.99 }),
4292 wakaba 1.171 }; # rp
4293 wakaba 1.99
4294 wakaba 1.29 =pod
4295    
4296     ## TODO:
4297    
4298     +
4299     + <p>Partly because of the confusion described above, authors are
4300     + strongly recommended to always mark up all paragraphs with the
4301     + <code>p</code> element, and to not have any <code>ins</code> or
4302     + <code>del</code> elements that cross across any <span
4303     + title="paragraph">implied paragraphs</span>.</p>
4304     +
4305     (An informative note)
4306    
4307     <p><code>ins</code> elements should not cross <span
4308     + title="paragraph">implied paragraph</span> boundaries.</p>
4309     (normative)
4310    
4311     + <p><code>del</code> elements should not cross <span
4312     + title="paragraph">implied paragraph</span> boundaries.</p>
4313     (normative)
4314    
4315     =cut
4316    
4317 wakaba 1.1 $Element->{$HTML_NS}->{ins} = {
4318 wakaba 1.40 %HTMLTransparentChecker,
4319 wakaba 1.187 status => FEATURE_HTML5_REC,
4320 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4321 wakaba 1.1 cite => $HTMLURIAttrChecker,
4322 wakaba 1.168 datetime => $GetDateTimeAttrChecker->('global_date_and_time_string'),
4323 wakaba 1.49 }, {
4324     %HTMLAttrStatus,
4325     %HTMLM12NCommonAttrStatus,
4326 wakaba 1.50 cite => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4327 wakaba 1.153 datetime => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4328 wakaba 1.187 lang => FEATURE_HTML5_REC,
4329 wakaba 1.1 }),
4330 wakaba 1.66 check_start => sub {
4331     my ($self, $item, $element_state) = @_;
4332    
4333     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
4334 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4335     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4336 wakaba 1.66 },
4337 wakaba 1.1 };
4338    
4339     $Element->{$HTML_NS}->{del} = {
4340 wakaba 1.40 %HTMLTransparentChecker,
4341 wakaba 1.187 status => FEATURE_HTML5_REC,
4342 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4343 wakaba 1.1 cite => $HTMLURIAttrChecker,
4344 wakaba 1.168 datetime => $GetDateTimeAttrChecker->('global_date_and_time_string'),
4345 wakaba 1.49 }, {
4346     %HTMLAttrStatus,
4347     %HTMLM12NCommonAttrStatus,
4348 wakaba 1.50 cite => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4349 wakaba 1.153 datetime => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4350 wakaba 1.187 lang => FEATURE_HTML5_REC,
4351 wakaba 1.1 }),
4352 wakaba 1.40 check_end => sub {
4353     my ($self, $item, $element_state) = @_;
4354     if ($element_state->{has_significant}) {
4355     ## NOTE: Significantness flag does not propagate.
4356     } elsif ($item->{transparent}) {
4357     #
4358     } else {
4359     $self->{onerror}->(node => $item->{node},
4360 wakaba 1.104 level => $self->{level}->{should},
4361 wakaba 1.40 type => 'no significant content');
4362     }
4363 wakaba 1.1 },
4364 wakaba 1.66 check_start => sub {
4365     my ($self, $item, $element_state) = @_;
4366    
4367     $element_state->{uri_info}->{cite}->{type}->{cite} = 1;
4368 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4369     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4370 wakaba 1.66 },
4371 wakaba 1.1 };
4372    
4373 wakaba 1.35 $Element->{$HTML_NS}->{figure} = {
4374 wakaba 1.72 %HTMLFlowContentChecker,
4375 wakaba 1.153 status => FEATURE_HTML5_WD,
4376 wakaba 1.72 ## NOTE: legend, Flow | Flow, legend?
4377 wakaba 1.41 check_child_element => sub {
4378     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4379     $child_is_transparent, $element_state) = @_;
4380 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4381     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4382 wakaba 1.41 $self->{onerror}->(node => $child_el,
4383     type => 'element not allowed:minus',
4384 wakaba 1.104 level => $self->{level}->{must});
4385 wakaba 1.41 $element_state->{has_non_legend} = 1;
4386 wakaba 1.181 $element_state->{has_non_table} = 1;
4387 wakaba 1.41 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4388 wakaba 1.181 $element_state->{has_non_table} = 1;
4389 wakaba 1.41 } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'legend') {
4390     if ($element_state->{has_legend_at_first}) {
4391     $self->{onerror}->(node => $child_el,
4392     type => 'element not allowed:figure legend',
4393 wakaba 1.104 level => $self->{level}->{must});
4394 wakaba 1.41 } elsif ($element_state->{has_legend}) {
4395     $self->{onerror}->(node => $element_state->{has_legend},
4396     type => 'element not allowed:figure legend',
4397 wakaba 1.104 level => $self->{level}->{must});
4398 wakaba 1.41 $element_state->{has_legend} = $child_el;
4399     } elsif ($element_state->{has_non_legend}) {
4400     $element_state->{has_legend} = $child_el;
4401     } else {
4402     $element_state->{has_legend_at_first} = 1;
4403 wakaba 1.35 }
4404 wakaba 1.41 delete $element_state->{has_non_legend};
4405     } else {
4406 wakaba 1.181 if ($child_nsuri eq $HTML_NS and $child_ln eq 'table') {
4407     $element_state->{has_table}++;
4408     } else {
4409     $element_state->{has_non_table}++;
4410     }
4411 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
4412 wakaba 1.43 $element_state->{has_non_legend} = 1 unless $child_is_transparent;
4413 wakaba 1.41 }
4414     },
4415     check_child_text => sub {
4416     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4417     if ($has_significant) {
4418     $element_state->{has_non_legend} = 1;
4419 wakaba 1.181 $element_state->{has_non_table}++;
4420 wakaba 1.35 }
4421 wakaba 1.170
4422     $element_state->{in_figure} = 1;
4423 wakaba 1.41 },
4424     check_end => sub {
4425     my ($self, $item, $element_state) = @_;
4426 wakaba 1.35
4427 wakaba 1.41 if ($element_state->{has_legend_at_first}) {
4428     #
4429     } elsif ($element_state->{has_legend}) {
4430     if ($element_state->{has_non_legend}) {
4431     $self->{onerror}->(node => $element_state->{has_legend},
4432 wakaba 1.35 type => 'element not allowed:figure legend',
4433 wakaba 1.104 level => $self->{level}->{must});
4434 wakaba 1.35 }
4435     }
4436 wakaba 1.41
4437 wakaba 1.181 if (($element_state->{has_table} || 0) == 1 and
4438     not $element_state->{has_non_table} and
4439     $element_state->{table_caption_element}) {
4440     $self->{onerror}->(node => $element_state->{table_caption_element},
4441     type => 'element not allowed',
4442     level => $self->{level}->{should});
4443     }
4444    
4445 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
4446 wakaba 1.41 ## ISSUE: |<figure><legend>aa</legend></figure>| should be an error?
4447 wakaba 1.35 },
4448     };
4449 wakaba 1.8 ## TODO: Test for <nest/> in <figure/>
4450 wakaba 1.1
4451 wakaba 1.92 my $AttrCheckerNotImplemented = sub {
4452     my ($self, $attr) = @_;
4453 wakaba 1.104 $self->{onerror}->(node => $attr,
4454     type => 'unknown attribute',
4455     level => $self->{level}->{uncertain});
4456 wakaba 1.92 };
4457    
4458 wakaba 1.1 $Element->{$HTML_NS}->{img} = {
4459 wakaba 1.40 %HTMLEmptyChecker,
4460 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4461 wakaba 1.40 check_attrs => sub {
4462     my ($self, $item, $element_state) = @_;
4463 wakaba 1.1 $GetHTMLAttrsChecker->({
4464 wakaba 1.70 align => $GetHTMLEnumeratedAttrChecker->({
4465     bottom => 1, middle => 1, top => 1, left => 1, right => 1,
4466     }),
4467 wakaba 1.1 alt => sub { }, ## NOTE: No syntactical requirement
4468 wakaba 1.70 border => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4469 wakaba 1.1 src => $HTMLURIAttrChecker,
4470     usemap => $HTMLUsemapAttrChecker,
4471 wakaba 1.70 hspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4472 wakaba 1.1 ismap => sub {
4473 wakaba 1.40 my ($self, $attr, $parent_item) = @_;
4474     if (not $self->{flag}->{in_a_href}) {
4475 wakaba 1.15 $self->{onerror}->(node => $attr,
4476 wakaba 1.59 type => 'attribute not allowed:ismap',
4477 wakaba 1.104 level => $self->{level}->{must});
4478 wakaba 1.1 }
4479 wakaba 1.40 $GetHTMLBooleanAttrChecker->('ismap')->($self, $attr, $parent_item);
4480 wakaba 1.1 },
4481 wakaba 1.70 longdesc => $HTMLURIAttrChecker,
4482     ## TODO: HTML4 |name|
4483 wakaba 1.178 height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4484 wakaba 1.70 vspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4485 wakaba 1.178 width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4486 wakaba 1.49 }, {
4487     %HTMLAttrStatus,
4488 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4489 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
4490 wakaba 1.153 alt => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4491 wakaba 1.49 border => FEATURE_M12N10_REC_DEPRECATED,
4492 wakaba 1.153 height => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4493 wakaba 1.49 hspace => FEATURE_M12N10_REC_DEPRECATED,
4494 wakaba 1.153 ismap => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4495 wakaba 1.187 lang => FEATURE_HTML5_REC,
4496 wakaba 1.49 longdesc => FEATURE_M12N10_REC,
4497     name => FEATURE_M12N10_REC_DEPRECATED,
4498 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
4499 wakaba 1.153 src => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4500     usemap => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4501 wakaba 1.49 vspace => FEATURE_M12N10_REC_DEPRECATED,
4502 wakaba 1.153 width => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4503 wakaba 1.66 })->($self, $item, $element_state);
4504 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'alt')) {
4505     $self->{onerror}->(node => $item->{node},
4506 wakaba 1.104 type => 'attribute missing',
4507     text => 'alt',
4508     level => $self->{level}->{should});
4509 wakaba 1.114 ## TODO: ...
4510 wakaba 1.1 }
4511 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'src')) {
4512     $self->{onerror}->(node => $item->{node},
4513 wakaba 1.104 type => 'attribute missing',
4514     text => 'src',
4515     level => $self->{level}->{must});
4516 wakaba 1.1 }
4517 wakaba 1.66
4518 wakaba 1.114 ## TODO: external resource check
4519    
4520 wakaba 1.66 $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4521     $element_state->{uri_info}->{lowsrc}->{type}->{embedded} = 1;
4522     $element_state->{uri_info}->{dynsrc}->{type}->{embedded} = 1;
4523     $element_state->{uri_info}->{longdesc}->{type}->{cite} = 1;
4524 wakaba 1.1 },
4525     };
4526    
4527     $Element->{$HTML_NS}->{iframe} = {
4528 wakaba 1.205 %HTMLTextChecker, # XXX content model restriction
4529 wakaba 1.114 status => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4530 wakaba 1.49 ## NOTE: Not part of M12N10 Strict
4531 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4532 wakaba 1.178 height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4533 wakaba 1.76 name => $HTMLBrowsingContextNameAttrChecker,
4534 wakaba 1.92 sandbox => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->({
4535     'allow-same-origin' => 1, 'allow-forms' => 1, 'allow-scripts' => 1,
4536     }),
4537     seemless => $GetHTMLBooleanAttrChecker->('seemless'),
4538 wakaba 1.1 src => $HTMLURIAttrChecker,
4539 wakaba 1.178 width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4540 wakaba 1.49 }, {
4541     %HTMLAttrStatus,
4542     %HTMLM12NCommonAttrStatus,
4543     align => FEATURE_XHTML10_REC,
4544 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4545 wakaba 1.49 frameborder => FEATURE_M12N10_REC,
4546 wakaba 1.153 height => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4547 wakaba 1.187 id => FEATURE_HTML5_REC,
4548 wakaba 1.49 longdesc => FEATURE_M12N10_REC,
4549     marginheight => FEATURE_M12N10_REC,
4550     marginwidth => FEATURE_M12N10_REC,
4551 wakaba 1.114 #name => FEATURE_HTML5_WD | FEATURE_M12N10_REC_DEPRECATED,
4552     name => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4553     sandbox => FEATURE_HTML5_WD,
4554 wakaba 1.49 scrolling => FEATURE_M12N10_REC,
4555 wakaba 1.114 seemless => FEATURE_HTML5_WD,
4556     src => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4557 wakaba 1.187 title => FEATURE_HTML5_REC,
4558 wakaba 1.153 width => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4559 wakaba 1.1 }),
4560 wakaba 1.66 check_start => sub {
4561     my ($self, $item, $element_state) = @_;
4562    
4563     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4564 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4565     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4566 wakaba 1.66 },
4567 wakaba 1.40 };
4568    
4569 wakaba 1.1 $Element->{$HTML_NS}->{embed} = {
4570 wakaba 1.40 %HTMLEmptyChecker,
4571 wakaba 1.98 status => FEATURE_HTML5_WD,
4572 wakaba 1.40 check_attrs => sub {
4573     my ($self, $item, $element_state) = @_;
4574 wakaba 1.1 my $has_src;
4575 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
4576 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
4577     $attr_ns = '' unless defined $attr_ns;
4578     my $attr_ln = $attr->manakai_local_name;
4579     my $checker;
4580 wakaba 1.73
4581     my $status = {
4582     %HTMLAttrStatus,
4583 wakaba 1.153 height => FEATURE_HTML5_LC,
4584 wakaba 1.98 src => FEATURE_HTML5_WD,
4585     type => FEATURE_HTML5_WD,
4586 wakaba 1.153 width => FEATURE_HTML5_LC,
4587 wakaba 1.73 }->{$attr_ln};
4588    
4589 wakaba 1.1 if ($attr_ns eq '') {
4590     if ($attr_ln eq 'src') {
4591     $checker = $HTMLURIAttrChecker;
4592     $has_src = 1;
4593     } elsif ($attr_ln eq 'type') {
4594     $checker = $HTMLIMTAttrChecker;
4595 wakaba 1.92 } elsif ($attr_ln eq 'width' or $attr_ln eq 'height') {
4596 wakaba 1.178 $checker = $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 });
4597 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
4598     $attr_ln !~ /[A-Z]/) {
4599 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
4600     $status = $HTMLDatasetAttrStatus;
4601 wakaba 1.117 } elsif ($attr_ln !~ /^[Xx][Mm][Ll]/ and
4602 wakaba 1.118 $attr_ln !~ /[A-Z]/ and
4603 wakaba 1.117 $attr_ln =~ /\A\p{InXML_NCNameStartChar10}\p{InXMLNCNameChar10}*\z/) {
4604 wakaba 1.1 $checker = $HTMLAttrChecker->{$attr_ln}
4605     || sub { }; ## NOTE: Any local attribute is ok.
4606 wakaba 1.98 $status = FEATURE_HTML5_WD | FEATURE_ALLOWED;
4607 wakaba 1.117 } else {
4608     $checker = $HTMLAttrChecker->{$attr_ln};
4609 wakaba 1.1 }
4610     }
4611     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
4612 wakaba 1.82 || $AttrChecker->{$attr_ns}->{''};
4613     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
4614     || $AttrStatus->{$attr_ns}->{''};
4615     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
4616 wakaba 1.62
4617 wakaba 1.1 if ($checker) {
4618 wakaba 1.66 $checker->($self, $attr, $item, $element_state);
4619 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
4620 wakaba 1.54 #
4621 wakaba 1.1 } else {
4622 wakaba 1.104 $self->{onerror}->(node => $attr,
4623     type => 'unknown attribute',
4624     level => $self->{level}->{uncertain});
4625 wakaba 1.50 ## ISSUE: No conformance createria for global attributes in the spec
4626     }
4627    
4628 wakaba 1.82 $self->_attr_status_info ($attr, $status);
4629 wakaba 1.1 }
4630    
4631     unless ($has_src) {
4632 wakaba 1.40 $self->{onerror}->(node => $item->{node},
4633 wakaba 1.104 type => 'attribute missing',
4634     text => 'src',
4635 wakaba 1.114 level => $self->{level}->{info});
4636     ## NOTE: <embed> without src="" is allowed since revision 1929.
4637     ## We issues an informational message since <embed> w/o src=""
4638     ## is likely an authoring error.
4639 wakaba 1.1 }
4640 wakaba 1.114
4641     ## TODO: external resource check
4642 wakaba 1.66
4643     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4644 wakaba 1.1 },
4645     };
4646    
4647 wakaba 1.49 ## TODO:
4648     ## {applet} FEATURE_M12N10_REC_DEPRECATED
4649     ## class, id, title, alt, archive, code, codebase, height, object, width name style,hspace,vspace(xhtml10)
4650    
4651 wakaba 1.1 $Element->{$HTML_NS}->{object} = {
4652 wakaba 1.40 %HTMLTransparentChecker,
4653 wakaba 1.153 status => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4654 wakaba 1.40 check_attrs => sub {
4655     my ($self, $item, $element_state) = @_;
4656 wakaba 1.1 $GetHTMLAttrsChecker->({
4657 wakaba 1.70 align => $GetHTMLEnumeratedAttrChecker->({
4658     bottom => 1, middle => 1, top => 1, left => 1, right => 1,
4659     }),
4660     archive => $HTMLSpaceURIsAttrChecker,
4661     ## TODO: Relative to @codebase
4662     border => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4663     classid => $HTMLURIAttrChecker,
4664     codebase => $HTMLURIAttrChecker,
4665     codetype => $HTMLIMTAttrChecker,
4666     ## TODO: "RECOMMENDED when |classid| is specified" [HTML4]
4667 wakaba 1.1 data => $HTMLURIAttrChecker,
4668 wakaba 1.70 declare => $GetHTMLBooleanAttrChecker->('declare'),
4669     ## NOTE: "The object MUST be instantiated by a subsequent OBJECT ..."
4670     ## [HTML4] but we don't know how to test this.
4671 wakaba 1.167 form => $HTMLFormAttrChecker,
4672 wakaba 1.178 height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4673 wakaba 1.70 hspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4674 wakaba 1.76 name => $HTMLBrowsingContextNameAttrChecker,
4675 wakaba 1.167 ## NOTE: |name| attribute of the |object| element defines
4676     ## the name of the browsing context created by the element,
4677     ## if any, but is also used as the form control name of the
4678     ## form control provided by the plugin, if any.
4679 wakaba 1.70 standby => sub {}, ## NOTE: %Text; in HTML4
4680 wakaba 1.1 type => $HTMLIMTAttrChecker,
4681     usemap => $HTMLUsemapAttrChecker,
4682 wakaba 1.70 vspace => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4683 wakaba 1.178 width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4684 wakaba 1.49 }, {
4685     %HTMLAttrStatus,
4686 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
4687 wakaba 1.49 align => FEATURE_XHTML10_REC,
4688 wakaba 1.82 archive => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4689 wakaba 1.49 border => FEATURE_XHTML10_REC,
4690     classid => FEATURE_M12N10_REC,
4691     codebase => FEATURE_M12N10_REC,
4692     codetype => FEATURE_M12N10_REC,
4693 wakaba 1.82 'content-length' => FEATURE_XHTML2_ED,
4694 wakaba 1.153 data => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4695 wakaba 1.49 datafld => FEATURE_HTML4_REC_RESERVED,
4696     dataformatas => FEATURE_HTML4_REC_RESERVED,
4697     datasrc => FEATURE_HTML4_REC_RESERVED,
4698 wakaba 1.82 declare => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4699 wakaba 1.187 form => FEATURE_HTML5_LC,
4700 wakaba 1.153 height => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4701 wakaba 1.49 hspace => FEATURE_XHTML10_REC,
4702 wakaba 1.187 lang => FEATURE_HTML5_REC,
4703 wakaba 1.153 name => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4704 wakaba 1.49 standby => FEATURE_M12N10_REC,
4705 wakaba 1.50 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
4706 wakaba 1.153 type => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
4707     usemap => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4708 wakaba 1.49 vspace => FEATURE_XHTML10_REC,
4709 wakaba 1.153 width => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
4710 wakaba 1.66 })->($self, $item, $element_state);
4711 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'data')) {
4712     unless ($item->{node}->has_attribute_ns (undef, 'type')) {
4713     $self->{onerror}->(node => $item->{node},
4714 wakaba 1.104 type => 'attribute missing:data|type',
4715     level => $self->{level}->{must});
4716 wakaba 1.1 }
4717     }
4718 wakaba 1.66
4719     $element_state->{uri_info}->{data}->{type}->{embedded} = 1;
4720     $element_state->{uri_info}->{classid}->{type}->{embedded} = 1;
4721     $element_state->{uri_info}->{codebase}->{type}->{base} = 1;
4722     ## TODO: archive
4723     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
4724 wakaba 1.1 },
4725 wakaba 1.72 ## NOTE: param*, transparent (Flow)
4726 wakaba 1.41 check_child_element => sub {
4727     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4728     $child_is_transparent, $element_state) = @_;
4729 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4730     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4731 wakaba 1.41 $self->{onerror}->(node => $child_el,
4732     type => 'element not allowed:minus',
4733 wakaba 1.104 level => $self->{level}->{must});
4734 wakaba 1.41 $element_state->{has_non_legend} = 1;
4735     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4736     #
4737     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'param') {
4738     if ($element_state->{has_non_param}) {
4739 wakaba 1.104 $self->{onerror}->(node => $child_el,
4740 wakaba 1.72 type => 'element not allowed:flow',
4741 wakaba 1.104 level => $self->{level}->{must});
4742 wakaba 1.39 }
4743 wakaba 1.41 } else {
4744 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
4745 wakaba 1.41 $element_state->{has_non_param} = 1;
4746 wakaba 1.39 }
4747 wakaba 1.25 },
4748 wakaba 1.41 check_child_text => sub {
4749     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4750     if ($has_significant) {
4751     $element_state->{has_non_param} = 1;
4752     }
4753 wakaba 1.42 },
4754     check_end => sub {
4755     my ($self, $item, $element_state) = @_;
4756     if ($element_state->{has_significant}) {
4757 wakaba 1.46 $item->{real_parent_state}->{has_significant} = 1;
4758 wakaba 1.42 } elsif ($item->{node}->manakai_parent_element) {
4759     ## NOTE: Transparent.
4760     } else {
4761     $self->{onerror}->(node => $item->{node},
4762 wakaba 1.104 level => $self->{level}->{should},
4763 wakaba 1.42 type => 'no significant content');
4764     }
4765     },
4766 wakaba 1.8 ## TODO: Tests for <nest/> in <object/>
4767 wakaba 1.1 };
4768 wakaba 1.41 ## ISSUE: Is |<menu><object data><li>aa</li></object></menu>| conforming?
4769     ## What about |<section><object data><style scoped></style>x</object></section>|?
4770     ## |<section><ins></ins><object data><style scoped></style>x</object></section>|?
4771 wakaba 1.1
4772     $Element->{$HTML_NS}->{param} = {
4773 wakaba 1.40 %HTMLEmptyChecker,
4774 wakaba 1.187 status => FEATURE_HTML5_REC,
4775 wakaba 1.40 check_attrs => sub {
4776     my ($self, $item, $element_state) = @_;
4777 wakaba 1.1 $GetHTMLAttrsChecker->({
4778     name => sub { },
4779 wakaba 1.70 type => $HTMLIMTAttrChecker,
4780 wakaba 1.1 value => sub { },
4781 wakaba 1.70 valuetype => $GetHTMLEnumeratedAttrChecker->({
4782     data => 1, ref => 1, object => 1,
4783     }),
4784 wakaba 1.49 }, {
4785     %HTMLAttrStatus,
4786 wakaba 1.154 href => FEATURE_RDFA_REC,
4787 wakaba 1.187 id => FEATURE_HTML5_REC,
4788 wakaba 1.153 name => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4789 wakaba 1.82 type => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4790 wakaba 1.153 value => FEATURE_HTML5_WD | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4791 wakaba 1.82 valuetype => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
4792 wakaba 1.66 })->(@_);
4793 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'name')) {
4794     $self->{onerror}->(node => $item->{node},
4795 wakaba 1.104 type => 'attribute missing',
4796     text => 'name',
4797     level => $self->{level}->{must});
4798 wakaba 1.1 }
4799 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'value')) {
4800     $self->{onerror}->(node => $item->{node},
4801 wakaba 1.104 type => 'attribute missing',
4802     text => 'value',
4803     level => $self->{level}->{must});
4804 wakaba 1.1 }
4805     },
4806     };
4807    
4808     $Element->{$HTML_NS}->{video} = {
4809 wakaba 1.40 %HTMLTransparentChecker,
4810 wakaba 1.48 status => FEATURE_HTML5_LC,
4811 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4812 wakaba 1.183 autobuffer => $GetHTMLBooleanAttrChecker->('autobuffer'),
4813 wakaba 1.205 autoplay => sub {
4814     my ($self, $attr) = @_;
4815    
4816     ## "Authors are also encouraged to consider not using the
4817     ## automatic playback behavior at all" according to HTML5.
4818     $self->{onerror}->(node => $attr,
4819     type => 'attribute not allowed',
4820     level => $self->{level}->{warn});
4821    
4822     $GetHTMLBooleanAttrChecker->('autoplay')->(@_);
4823     },
4824 wakaba 1.1 controls => $GetHTMLBooleanAttrChecker->('controls'),
4825 wakaba 1.209 end => sub { },
4826     height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4827     loop => $GetHTMLBooleanAttrChecker->('loop'),
4828     loopend => sub { },
4829     loopstart => sub { },
4830     playcount => sub { },
4831 wakaba 1.59 poster => $HTMLURIAttrChecker,
4832 wakaba 1.209 src => $HTMLURIAttrChecker,
4833     start => sub { },
4834 wakaba 1.178 width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4835 wakaba 1.209
4836     ## NOTE: |start|, |end|, |loopstart|, |loopend|, and |playcount|
4837     ## attributes has been deleted from the spec before the exact
4838     ## author requirement is defined.
4839 wakaba 1.50 }, {
4840     %HTMLAttrStatus,
4841 wakaba 1.183 autobuffer => FEATURE_HTML5_LC,
4842 wakaba 1.50 autoplay => FEATURE_HTML5_LC,
4843     controls => FEATURE_HTML5_LC,
4844 wakaba 1.209 end => FEATURE_HTML5_DROPPED,
4845 wakaba 1.50 height => FEATURE_HTML5_LC,
4846 wakaba 1.209 loop => FEATURE_HTML5_LC,
4847     loopend => FEATURE_HTML5_DROPPED,
4848     loopstart => FEATURE_HTML5_DROPPED,
4849     playcount => FEATURE_HTML5_DROPPED,
4850 wakaba 1.50 poster => FEATURE_HTML5_LC,
4851     src => FEATURE_HTML5_LC,
4852 wakaba 1.209 start => FEATURE_HTML5_DROPPED,
4853 wakaba 1.50 width => FEATURE_HTML5_LC,
4854 wakaba 1.209 }), # check_attrs
4855 wakaba 1.42 check_start => sub {
4856     my ($self, $item, $element_state) = @_;
4857     $element_state->{allow_source}
4858     = not $item->{node}->has_attribute_ns (undef, 'src');
4859     $element_state->{has_source} ||= $element_state->{allow_source} * -1;
4860     ## NOTE: It might be set true by |check_element|.
4861 wakaba 1.66
4862     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4863     $element_state->{uri_info}->{poster}->{type}->{embedded} = 1;
4864 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
4865     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
4866 wakaba 1.42 },
4867     check_child_element => sub {
4868     my ($self, $item, $child_el, $child_nsuri, $child_ln,
4869     $child_is_transparent, $element_state) = @_;
4870 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
4871     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
4872 wakaba 1.42 $self->{onerror}->(node => $child_el,
4873     type => 'element not allowed:minus',
4874 wakaba 1.104 level => $self->{level}->{must});
4875 wakaba 1.42 delete $element_state->{allow_source};
4876     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
4877     #
4878     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'source') {
4879 wakaba 1.45 unless ($element_state->{allow_source}) {
4880 wakaba 1.104 $self->{onerror}->(node => $child_el,
4881 wakaba 1.72 type => 'element not allowed:flow',
4882 wakaba 1.104 level => $self->{level}->{must});
4883 wakaba 1.42 }
4884 wakaba 1.45 $element_state->{has_source} = 1;
4885 wakaba 1.1 } else {
4886 wakaba 1.42 delete $element_state->{allow_source};
4887 wakaba 1.72 $HTMLFlowContentChecker{check_child_element}->(@_);
4888 wakaba 1.42 }
4889     },
4890     check_child_text => sub {
4891     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
4892     if ($has_significant) {
4893     delete $element_state->{allow_source};
4894     }
4895 wakaba 1.72 $HTMLFlowContentChecker{check_child_text}->(@_);
4896 wakaba 1.42 },
4897     check_end => sub {
4898     my ($self, $item, $element_state) = @_;
4899     if ($element_state->{has_source} == -1) {
4900     $self->{onerror}->(node => $item->{node},
4901 wakaba 1.104 type => 'child element missing',
4902     text => 'source',
4903     level => $self->{level}->{must});
4904 wakaba 1.1 }
4905 wakaba 1.42
4906     $Element->{$HTML_NS}->{object}->{check_end}->(@_);
4907 wakaba 1.1 },
4908 wakaba 1.209 }; # video
4909 wakaba 1.1
4910     $Element->{$HTML_NS}->{audio} = {
4911 wakaba 1.40 %{$Element->{$HTML_NS}->{video}},
4912 wakaba 1.48 status => FEATURE_HTML5_LC,
4913 wakaba 1.42 check_attrs => $GetHTMLAttrsChecker->({
4914 wakaba 1.183 autobuffer => $GetHTMLBooleanAttrChecker->('autobuffer'),
4915 wakaba 1.205 autoplay => sub {
4916     my ($self, $attr) = @_;
4917    
4918     ## "Authors are also encouraged to consider not using the
4919     ## automatic playback behavior at all" according to HTML5.
4920     $self->{onerror}->(node => $attr,
4921     type => 'attribute not allowed',
4922     level => $self->{level}->{warn});
4923    
4924     $GetHTMLBooleanAttrChecker->('autoplay')->(@_);
4925     },
4926 wakaba 1.42 controls => $GetHTMLBooleanAttrChecker->('controls'),
4927 wakaba 1.209 end => sub { },
4928     loop => $GetHTMLBooleanAttrChecker->('loop'),
4929     loopend => sub { },
4930     loopstart => sub { },
4931     playcount => sub { },
4932     src => $HTMLURIAttrChecker,
4933     start => sub { },
4934    
4935     ## NOTE: |start|, |end|, |loopstart|, |loopend|, and |playcount|
4936     ## attributes has been deleted from the spec before the exact
4937     ## author requirement is defined.
4938 wakaba 1.50 }, {
4939     %HTMLAttrStatus,
4940 wakaba 1.183 autobuffer => FEATURE_HTML5_LC,
4941 wakaba 1.50 autoplay => FEATURE_HTML5_LC,
4942     controls => FEATURE_HTML5_LC,
4943 wakaba 1.209 end => FEATURE_HTML5_DROPPED,
4944     loop => FEATURE_HTML5_LC,
4945     loopend => FEATURE_HTML5_DROPPED,
4946     loopstart => FEATURE_HTML5_DROPPED,
4947     playcount => FEATURE_HTML5_DROPPED,
4948 wakaba 1.50 src => FEATURE_HTML5_LC,
4949 wakaba 1.209 start => FEATURE_HTML5_DROPPED,
4950     }), # check_attrs
4951     }; # audio
4952 wakaba 1.1
4953     $Element->{$HTML_NS}->{source} = {
4954 wakaba 1.40 %HTMLEmptyChecker,
4955 wakaba 1.153 status => FEATURE_HTML5_LC,
4956 wakaba 1.40 check_attrs => sub {
4957     my ($self, $item, $element_state) = @_;
4958 wakaba 1.1 $GetHTMLAttrsChecker->({
4959 wakaba 1.90 media => $HTMLMQAttrChecker,
4960 wakaba 1.206 pixelratio => $PositiveFloatingPointNumberAttrChecker,
4961     src => $HTMLURIAttrChecker,
4962 wakaba 1.1 type => $HTMLIMTAttrChecker,
4963 wakaba 1.50 }, {
4964     %HTMLAttrStatus,
4965 wakaba 1.153 media => FEATURE_HTML5_LC,
4966 wakaba 1.206 pixelratio => FEATURE_HTML5_DROPPED,
4967 wakaba 1.153 src => FEATURE_HTML5_LC,
4968     type => FEATURE_HTML5_LC,
4969 wakaba 1.66 })->(@_);
4970 wakaba 1.40 unless ($item->{node}->has_attribute_ns (undef, 'src')) {
4971     $self->{onerror}->(node => $item->{node},
4972 wakaba 1.104 type => 'attribute missing',
4973     text => 'src',
4974     level => $self->{level}->{must});
4975 wakaba 1.1 }
4976 wakaba 1.66
4977     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
4978 wakaba 1.206
4979     ## NOTE: The |pixelratio| attribute should have been forbidden
4980     ## when the parent of the |source| element is an |audio| element,
4981     ## but the attribute itself has been dropped from the spec.
4982 wakaba 1.1 },
4983 wakaba 1.206 }; # source
4984 wakaba 1.1
4985     $Element->{$HTML_NS}->{canvas} = {
4986 wakaba 1.40 %HTMLTransparentChecker,
4987 wakaba 1.187 status => FEATURE_HTML5_REC,
4988 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
4989 wakaba 1.1 height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4990     width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
4991 wakaba 1.50 }, {
4992     %HTMLAttrStatus,
4993 wakaba 1.187 height => FEATURE_HTML5_REC,
4994     width => FEATURE_HTML5_REC,
4995 wakaba 1.1 }),
4996 wakaba 1.178
4997     # Authors MUST provide alternative content (HTML5 revision 2868) -
4998     # This requirement cannot be checked, since the alternative content
4999     # might be placed outside of the element.
5000     }; # canvas
5001 wakaba 1.1
5002     $Element->{$HTML_NS}->{map} = {
5003 wakaba 1.72 %HTMLFlowContentChecker,
5004 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5005 wakaba 1.40 check_attrs => sub {
5006     my ($self, $item, $element_state) = @_;
5007 wakaba 1.100 my $has_name;
5008 wakaba 1.4 $GetHTMLAttrsChecker->({
5009 wakaba 1.100 name => sub {
5010     my ($self, $attr) = @_;
5011     my $value = $attr->value;
5012     if (length $value) {
5013 wakaba 1.210 if ($value =~ /[\x09\x0A\x0C\x0D\x20]/) {
5014     $self->{onerror}->(node => $attr, type => 'space in map name',
5015     level => $self->{level}->{must}); ## XXX documentation
5016     }
5017    
5018     ## XXXNOTE: Duplication is not non-conforming.
5019 wakaba 1.100 } else {
5020     $self->{onerror}->(node => $attr,
5021     type => 'empty attribute value',
5022 wakaba 1.104 level => $self->{level}->{must});
5023 wakaba 1.100 }
5024 wakaba 1.4 $self->{map}->{$value} ||= $attr;
5025 wakaba 1.100 $has_name = [$value, $attr];
5026 wakaba 1.4 },
5027 wakaba 1.49 }, {
5028     %HTMLAttrStatus,
5029 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5030     dir => FEATURE_HTML5_REC,
5031     id => FEATURE_HTML5_REC,
5032     lang => FEATURE_HTML5_REC,
5033 wakaba 1.153 #name => FEATURE_HTML5_LC | FEATURE_M12N10_REC_DEPRECATED,
5034     name => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5035 wakaba 1.50 onclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5036     ondblclick => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5037     onmousedown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5038     onmouseup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5039     onmouseover => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5040     onmousemove => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5041     onmouseout => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5042     onkeypress => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5043     onkeydown => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5044     onkeyup => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5045 wakaba 1.187 title => FEATURE_HTML5_REC,
5046 wakaba 1.66 })->(@_);
5047 wakaba 1.100
5048 wakaba 1.135 if ($has_name) {
5049 wakaba 1.145 my $id = $item->{node}->get_attribute_ns (undef, 'id');
5050 wakaba 1.135 if (defined $id and $has_name->[0] ne $id) {
5051 wakaba 1.155 $self->{onerror}
5052     ->(node => $item->{node}->get_attribute_node_ns (undef, 'id'),
5053     type => 'id ne name',
5054     level => $self->{level}->{must});
5055 wakaba 1.100 }
5056 wakaba 1.135 } else {
5057 wakaba 1.100 $self->{onerror}->(node => $item->{node},
5058 wakaba 1.104 type => 'attribute missing',
5059     text => 'name',
5060     level => $self->{level}->{must});
5061 wakaba 1.100 }
5062 wakaba 1.4 },
5063 wakaba 1.59 check_start => sub {
5064     my ($self, $item, $element_state) = @_;
5065     $element_state->{in_map_original} = $self->{flag}->{in_map};
5066 wakaba 1.137 $self->{flag}->{in_map} = [@{$self->{flag}->{in_map} or []}, {}];
5067     ## NOTE: |{in_map}| is a reference to the array which contains
5068     ## hash references. Hashes are corresponding to the opening
5069     ## |map| elements and each of them contains the key-value
5070     ## pairs corresponding to the absolute URLs for the processed
5071     ## |area| elements in the |map| element corresponding to the
5072     ## hash. The key represents the resource (## TODO: use
5073     ## absolute URL), while the value represents whether there is
5074     ## an |area| element whose |alt| attribute is specified to a
5075     ## non-empty value. If there IS such an |area| element for
5076     ## the resource specified by the key, then the value is set to
5077     ## zero (|0|). Otherwise, if there is no such an |area|
5078     ## element but there is any |area| element with the empty
5079     ## |alt=""| attribute, then the value contains an array
5080     ## reference that contains all of such |area| elements.
5081 wakaba 1.79
5082     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5083     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5084 wakaba 1.59 },
5085     check_end => sub {
5086     my ($self, $item, $element_state) = @_;
5087 wakaba 1.137
5088     for (keys %{$self->{flag}->{in_map}->[-1]}) {
5089     my $nodes = $self->{flag}->{in_map}->[-1]->{$_};
5090     next unless $nodes;
5091     for (@$nodes) {
5092     $self->{onerror}->(type => 'empty area alt',
5093     node => $_,
5094     level => $self->{level}->{html5_no_may});
5095     }
5096     }
5097    
5098     $self->{flag}->{in_map} = $element_state->{in_map_original};
5099    
5100 wakaba 1.72 $HTMLFlowContentChecker{check_end}->(@_);
5101 wakaba 1.59 },
5102 wakaba 1.1 };
5103    
5104     $Element->{$HTML_NS}->{area} = {
5105 wakaba 1.40 %HTMLEmptyChecker,
5106 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5107 wakaba 1.40 check_attrs => sub {
5108     my ($self, $item, $element_state) = @_;
5109 wakaba 1.1 my %attr;
5110     my $coords;
5111 wakaba 1.40 for my $attr (@{$item->{node}->attributes}) {
5112 wakaba 1.1 my $attr_ns = $attr->namespace_uri;
5113     $attr_ns = '' unless defined $attr_ns;
5114     my $attr_ln = $attr->manakai_local_name;
5115     my $checker;
5116 wakaba 1.73 my $status;
5117 wakaba 1.1 if ($attr_ns eq '') {
5118 wakaba 1.73 $status = {
5119     %HTMLAttrStatus,
5120     %HTMLM12NCommonAttrStatus,
5121 wakaba 1.176 accesskey => FEATURE_M12N10_REC | FEATURE_HTML5_FD,
5122 wakaba 1.153 alt => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5123     coords => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5124 wakaba 1.154 href => FEATURE_HTML5_WD | FEATURE_RDFA_REC | FEATURE_M12N10_REC,
5125 wakaba 1.153 hreflang => FEATURE_HTML5_WD,
5126 wakaba 1.187 lang => FEATURE_HTML5_REC,
5127 wakaba 1.154 media => FEATURE_HTML5_WD,
5128 wakaba 1.73 nohref => FEATURE_M12N10_REC,
5129     onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5130     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5131 wakaba 1.153 ping => FEATURE_HTML5_WD,
5132 wakaba 1.154 rel => FEATURE_HTML5_WD | FEATURE_RDFA_REC,
5133 wakaba 1.153 shape => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5134 wakaba 1.73 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5135 wakaba 1.153 target => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
5136     type => FEATURE_HTML5_WD,
5137 wakaba 1.73 }->{$attr_ln};
5138    
5139 wakaba 1.1 $checker = {
5140 wakaba 1.153 alt => sub {
5141     ## NOTE: Checked later.
5142     },
5143 wakaba 1.1 shape => $GetHTMLEnumeratedAttrChecker->({
5144     circ => -1, circle => 1,
5145     default => 1,
5146     poly => 1, polygon => -1,
5147     rect => 1, rectangle => -1,
5148     }),
5149     coords => sub {
5150     my ($self, $attr) = @_;
5151     my $value = $attr->value;
5152     if ($value =~ /\A-?[0-9]+(?>,-?[0-9]+)*\z/) {
5153     $coords = [split /,/, $value];
5154     } else {
5155     $self->{onerror}->(node => $attr,
5156 wakaba 1.104 type => 'coords:syntax error',
5157     level => $self->{level}->{must});
5158 wakaba 1.1 }
5159     },
5160 wakaba 1.70 nohref => $GetHTMLBooleanAttrChecker->('nohref'),
5161     target => $HTMLTargetAttrChecker,
5162 wakaba 1.1 href => $HTMLURIAttrChecker,
5163     ping => $HTMLSpaceURIsAttrChecker,
5164 wakaba 1.40 rel => sub { $HTMLLinkTypesAttrChecker->(1, $item, @_) },
5165 wakaba 1.1 media => $HTMLMQAttrChecker,
5166     hreflang => $HTMLLanguageTagAttrChecker,
5167     type => $HTMLIMTAttrChecker,
5168     }->{$attr_ln};
5169     if ($checker) {
5170     $attr{$attr_ln} = $attr;
5171 wakaba 1.122 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
5172     $attr_ln !~ /[A-Z]/) {
5173 wakaba 1.73 $checker = $HTMLDatasetAttrChecker;
5174     $status = $HTMLDatasetAttrStatus;
5175 wakaba 1.1 } else {
5176     $checker = $HTMLAttrChecker->{$attr_ln};
5177     }
5178     }
5179     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
5180 wakaba 1.82 || $AttrChecker->{$attr_ns}->{''};
5181     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
5182     || $AttrStatus->{$attr_ns}->{''};
5183     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
5184 wakaba 1.62
5185 wakaba 1.1 if ($checker) {
5186 wakaba 1.66 $checker->($self, $attr, $item, $element_state) if ref $checker;
5187 wakaba 1.62 } elsif ($attr_ns eq '' and not $status) {
5188 wakaba 1.54 #
5189 wakaba 1.1 } else {
5190 wakaba 1.104 $self->{onerror}->(node => $attr,
5191     type => 'unknown attribute',
5192     level => $self->{level}->{uncertain});
5193 wakaba 1.1 ## ISSUE: No comformance createria for unknown attributes in the spec
5194     }
5195 wakaba 1.49
5196 wakaba 1.82 $self->_attr_status_info ($attr, $status);
5197 wakaba 1.1 }
5198    
5199     if (defined $attr{href}) {
5200 wakaba 1.4 $self->{has_hyperlink_element} = 1;
5201 wakaba 1.137 if (defined $attr{alt}) {
5202     my $url = $attr{href}->value; ## TODO: resolve
5203     if (length $attr{alt}->value) {
5204     for (@{$self->{flag}->{in_map} or []}) {
5205     $_->{$url} = 0;
5206     }
5207     } else {
5208     ## NOTE: Empty |alt=""|. If there is another |area| element
5209     ## with the same |href=""| and that |area| elemnet's
5210     ## |alt=""| attribute is not an empty string, then this
5211     ## is conforming.
5212     for (@{$self->{flag}->{in_map} or []}) {
5213     push @{$_->{$url} ||= []}, $attr{alt}
5214     unless exists $_->{$url} and not $_->{$url};
5215     }
5216     }
5217     } else {
5218 wakaba 1.40 $self->{onerror}->(node => $item->{node},
5219 wakaba 1.104 type => 'attribute missing',
5220     text => 'alt',
5221     level => $self->{level}->{must});
5222 wakaba 1.1 }
5223     } else {
5224     for (qw/target ping rel media hreflang type alt/) {
5225     if (defined $attr{$_}) {
5226     $self->{onerror}->(node => $attr{$_},
5227 wakaba 1.104 type => 'attribute not allowed',
5228     level => $self->{level}->{must});
5229 wakaba 1.1 }
5230     }
5231     }
5232    
5233     my $shape = 'rectangle';
5234     if (defined $attr{shape}) {
5235     $shape = {
5236     circ => 'circle', circle => 'circle',
5237     default => 'default',
5238     poly => 'polygon', polygon => 'polygon',
5239     rect => 'rectangle', rectangle => 'rectangle',
5240     }->{lc $attr{shape}->value} || 'rectangle';
5241     ## TODO: ASCII lowercase?
5242     }
5243    
5244     if ($shape eq 'circle') {
5245     if (defined $attr{coords}) {
5246     if (defined $coords) {
5247     if (@$coords == 3) {
5248     if ($coords->[2] < 0) {
5249     $self->{onerror}->(node => $attr{coords},
5250 wakaba 1.104 type => 'coords:out of range',
5251     index => 2,
5252     value => $coords->[2],
5253     level => $self->{level}->{must});
5254 wakaba 1.1 }
5255     } else {
5256     $self->{onerror}->(node => $attr{coords},
5257 wakaba 1.104 type => 'coords:number not 3',
5258     text => 0+@$coords,
5259     level => $self->{level}->{must});
5260 wakaba 1.1 }
5261     } else {
5262     ## NOTE: A syntax error has been reported.
5263     }
5264     } else {
5265 wakaba 1.40 $self->{onerror}->(node => $item->{node},
5266 wakaba 1.104 type => 'attribute missing',
5267     text => 'coords',
5268     level => $self->{level}->{must});
5269 wakaba 1.1 }
5270     } elsif ($shape eq 'default') {
5271     if (defined $attr{coords}) {
5272     $self->{onerror}->(node => $attr{coords},
5273 wakaba 1.104 type => 'attribute not allowed',
5274     level => $self->{level}->{must});
5275 wakaba 1.1 }
5276     } elsif ($shape eq 'polygon') {
5277     if (defined $attr{coords}) {
5278     if (defined $coords) {
5279     if (@$coords >= 6) {
5280     unless (@$coords % 2 == 0) {
5281     $self->{onerror}->(node => $attr{coords},
5282 wakaba 1.104 type => 'coords:number not even',
5283     text => 0+@$coords,
5284     level => $self->{level}->{must});
5285 wakaba 1.1 }
5286     } else {
5287     $self->{onerror}->(node => $attr{coords},
5288 wakaba 1.104 type => 'coords:number lt 6',
5289     text => 0+@$coords,
5290     level => $self->{level}->{must});
5291 wakaba 1.1 }
5292     } else {
5293     ## NOTE: A syntax error has been reported.
5294     }
5295     } else {
5296 wakaba 1.40 $self->{onerror}->(node => $item->{node},
5297 wakaba 1.104 type => 'attribute missing',
5298     text => 'coords',
5299     level => $self->{level}->{must});
5300 wakaba 1.1 }
5301     } elsif ($shape eq 'rectangle') {
5302     if (defined $attr{coords}) {
5303     if (defined $coords) {
5304     if (@$coords == 4) {
5305     unless ($coords->[0] < $coords->[2]) {
5306     $self->{onerror}->(node => $attr{coords},
5307 wakaba 1.104 type => 'coords:out of range',
5308     index => 0,
5309     value => $coords->[0],
5310     level => $self->{level}->{must});
5311 wakaba 1.1 }
5312     unless ($coords->[1] < $coords->[3]) {
5313     $self->{onerror}->(node => $attr{coords},
5314 wakaba 1.104 type => 'coords:out of range',
5315     index => 1,
5316     value => $coords->[1],
5317     level => $self->{level}->{must});
5318 wakaba 1.1 }
5319     } else {
5320     $self->{onerror}->(node => $attr{coords},
5321 wakaba 1.104 type => 'coords:number not 4',
5322     text => 0+@$coords,
5323     level => $self->{level}->{must});
5324 wakaba 1.1 }
5325     } else {
5326     ## NOTE: A syntax error has been reported.
5327     }
5328     } else {
5329 wakaba 1.40 $self->{onerror}->(node => $item->{node},
5330 wakaba 1.104 type => 'attribute missing',
5331     text => 'coords',
5332     level => $self->{level}->{must});
5333 wakaba 1.1 }
5334     }
5335 wakaba 1.66
5336     $element_state->{uri_info}->{href}->{type}->{hyperlink} = 1;
5337 wakaba 1.1 },
5338 wakaba 1.59 check_start => sub {
5339     my ($self, $item, $element_state) = @_;
5340     unless ($self->{flag}->{in_map} or
5341     not $item->{node}->manakai_parent_element) {
5342     $self->{onerror}->(node => $item->{node},
5343     type => 'element not allowed:area',
5344 wakaba 1.104 level => $self->{level}->{must});
5345 wakaba 1.59 }
5346 wakaba 1.79
5347     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5348     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5349 wakaba 1.59 },
5350 wakaba 1.1 };
5351    
5352     $Element->{$HTML_NS}->{table} = {
5353 wakaba 1.40 %HTMLChecker,
5354 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5355 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
5356 wakaba 1.86 cellpadding => $HTMLLengthAttrChecker,
5357     cellspacing => $HTMLLengthAttrChecker,
5358 wakaba 1.69 frame => $GetHTMLEnumeratedAttrChecker->({
5359     void => 1, above => 1, below => 1, hsides => 1, vsides => 1,
5360     lhs => 1, rhs => 1, box => 1, border => 1,
5361     }),
5362     rules => $GetHTMLEnumeratedAttrChecker->({
5363     none => 1, groups => 1, rows => 1, cols => 1, all => 1,
5364     }),
5365     summary => sub {}, ## NOTE: %Text; in HTML4.
5366     width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }), ## %Pixels;
5367     }, {
5368 wakaba 1.49 %HTMLAttrStatus,
5369 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5370 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
5371     bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5372     border => FEATURE_M12N10_REC,
5373     cellpadding => FEATURE_M12N10_REC,
5374     cellspacing => FEATURE_M12N10_REC,
5375 wakaba 1.61 cols => FEATURE_RFC1942,
5376 wakaba 1.49 datafld => FEATURE_HTML4_REC_RESERVED,
5377     dataformatas => FEATURE_HTML4_REC_RESERVED,
5378     datapagesize => FEATURE_M12N10_REC,
5379     datasrc => FEATURE_HTML4_REC_RESERVED,
5380     frame => FEATURE_M12N10_REC,
5381 wakaba 1.187 lang => FEATURE_HTML5_REC,
5382 wakaba 1.49 rules => FEATURE_M12N10_REC,
5383     summary => FEATURE_M12N10_REC,
5384     width => FEATURE_M12N10_REC,
5385     }),
5386 wakaba 1.40 check_start => sub {
5387     my ($self, $item, $element_state) = @_;
5388     $element_state->{phase} = 'before caption';
5389 wakaba 1.66
5390     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
5391 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5392     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5393 wakaba 1.40 },
5394     check_child_element => sub {
5395     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5396     $child_is_transparent, $element_state) = @_;
5397 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5398     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5399 wakaba 1.40 $self->{onerror}->(node => $child_el,
5400     type => 'element not allowed:minus',
5401 wakaba 1.104 level => $self->{level}->{must});
5402 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5403     #
5404     } elsif ($element_state->{phase} eq 'in tbodys') {
5405     if ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
5406     #$element_state->{phase} = 'in tbodys';
5407     } elsif (not $element_state->{has_tfoot} and
5408     $child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
5409     $element_state->{phase} = 'after tfoot';
5410     $element_state->{has_tfoot} = 1;
5411     } else {
5412 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5413     level => $self->{level}->{must});
5414 wakaba 1.40 }
5415     } elsif ($element_state->{phase} eq 'in trs') {
5416     if ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
5417     #$element_state->{phase} = 'in trs';
5418     } elsif (not $element_state->{has_tfoot} and
5419     $child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
5420     $element_state->{phase} = 'after tfoot';
5421     $element_state->{has_tfoot} = 1;
5422     } else {
5423 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5424     level => $self->{level}->{must});
5425 wakaba 1.40 }
5426     } elsif ($element_state->{phase} eq 'after thead') {
5427     if ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
5428     $element_state->{phase} = 'in tbodys';
5429     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
5430     $element_state->{phase} = 'in trs';
5431     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
5432     $element_state->{phase} = 'in tbodys';
5433     $element_state->{has_tfoot} = 1;
5434     } else {
5435 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5436     level => $self->{level}->{must});
5437 wakaba 1.40 }
5438     } elsif ($element_state->{phase} eq 'in colgroup') {
5439     if ($child_nsuri eq $HTML_NS and $child_ln eq 'colgroup') {
5440     $element_state->{phase} = 'in colgroup';
5441     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'thead') {
5442     $element_state->{phase} = 'after thead';
5443     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
5444     $element_state->{phase} = 'in tbodys';
5445     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
5446     $element_state->{phase} = 'in trs';
5447     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
5448     $element_state->{phase} = 'in tbodys';
5449     $element_state->{has_tfoot} = 1;
5450     } else {
5451 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5452     level => $self->{level}->{must});
5453 wakaba 1.40 }
5454     } elsif ($element_state->{phase} eq 'before caption') {
5455     if ($child_nsuri eq $HTML_NS and $child_ln eq 'caption') {
5456 wakaba 1.181 $item->{parent_state}->{table_caption_element} = $child_el;
5457 wakaba 1.40 $element_state->{phase} = 'in colgroup';
5458     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'colgroup') {
5459     $element_state->{phase} = 'in colgroup';
5460     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'thead') {
5461     $element_state->{phase} = 'after thead';
5462     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tbody') {
5463     $element_state->{phase} = 'in tbodys';
5464     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
5465     $element_state->{phase} = 'in trs';
5466     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tfoot') {
5467     $element_state->{phase} = 'in tbodys';
5468     $element_state->{has_tfoot} = 1;
5469     } else {
5470 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5471     level => $self->{level}->{must});
5472 wakaba 1.40 }
5473     } elsif ($element_state->{phase} eq 'after tfoot') {
5474 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5475     level => $self->{level}->{must});
5476 wakaba 1.40 } else {
5477     die "check_child_element: Bad |table| phase: $element_state->{phase}";
5478     }
5479     },
5480     check_child_text => sub {
5481     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5482     if ($has_significant) {
5483 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5484     level => $self->{level}->{must});
5485 wakaba 1.1 }
5486 wakaba 1.40 },
5487     check_end => sub {
5488     my ($self, $item, $element_state) = @_;
5489 wakaba 1.1
5490     ## Table model errors
5491     require Whatpm::HTMLTable;
5492 wakaba 1.87 my $table = Whatpm::HTMLTable->form_table ($item->{node}, sub {
5493 wakaba 1.104 $self->{onerror}->(@_);
5494     }, $self->{level});
5495 wakaba 1.87 Whatpm::HTMLTable->assign_header
5496 wakaba 1.104 ($table, $self->{onerror}, $self->{level});
5497 wakaba 1.87 push @{$self->{return}->{table}}, $table;
5498 wakaba 1.1
5499 wakaba 1.40 $HTMLChecker{check_end}->(@_);
5500 wakaba 1.1 },
5501     };
5502    
5503     $Element->{$HTML_NS}->{caption} = {
5504 wakaba 1.169 %HTMLFlowContentChecker,
5505 wakaba 1.187 status => FEATURE_HTML5_REC,
5506 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
5507     align => $GetHTMLEnumeratedAttrChecker->({
5508     top => 1, bottom => 1, left => 1, right => 1,
5509     }),
5510     }, {
5511 wakaba 1.49 %HTMLAttrStatus,
5512 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5513 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
5514 wakaba 1.187 lang => FEATURE_HTML5_REC,
5515 wakaba 1.49 }),
5516 wakaba 1.169 check_start => sub {
5517     my ($self, $item, $element_state) = @_;
5518     $self->_add_minus_elements ($element_state, {$HTML_NS => {table => 1}});
5519    
5520     $HTMLFlowContentChecker{check_start}->(@_);
5521     },
5522     check_end => sub {
5523     my ($self, $item, $element_state) = @_;
5524     $self->_remove_minus_elements ($element_state);
5525    
5526     $HTMLFlowContentChecker{check_end}->(@_);
5527     },
5528     }; # caption
5529 wakaba 1.1
5530 wakaba 1.69 my %cellalign = (
5531     ## HTML4 %cellhalign;
5532 wakaba 1.70 align => $GetHTMLEnumeratedAttrChecker->({
5533     left => 1, center => 1, right => 1, justify => 1, char => 1,
5534     }),
5535     char => sub {
5536     my ($self, $attr) = @_;
5537 wakaba 1.69
5538 wakaba 1.70 ## NOTE: "character" or |%Character;| in HTML4.
5539    
5540     my $value = $attr->value;
5541     if (length $value != 1) {
5542     $self->{onerror}->(node => $attr, type => 'char:syntax error',
5543 wakaba 1.105 level => $self->{level}->{html4_fact});
5544 wakaba 1.70 }
5545     },
5546 wakaba 1.86 charoff => $HTMLLengthAttrChecker,
5547    
5548 wakaba 1.69 ## HTML4 %cellvalign;
5549 wakaba 1.70 valign => $GetHTMLEnumeratedAttrChecker->({
5550     top => 1, middle => 1, bottom => 1, baseline => 1,
5551     }),
5552 wakaba 1.69 );
5553    
5554 wakaba 1.1 $Element->{$HTML_NS}->{colgroup} = {
5555 wakaba 1.40 %HTMLEmptyChecker,
5556 wakaba 1.187 status => FEATURE_HTML5_REC,
5557 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5558 wakaba 1.69 %cellalign,
5559 wakaba 1.1 span => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5560     ## NOTE: Defined only if "the |colgroup| element contains no |col| elements"
5561     ## TODO: "attribute not supported" if |col|.
5562     ## ISSUE: MUST NOT if any |col|?
5563     ## ISSUE: MUST NOT for |<colgroup span="1"><any><col/></any></colgroup>| (though non-conforming)?
5564 wakaba 1.49 }, {
5565     %HTMLAttrStatus,
5566 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5567 wakaba 1.49 align => FEATURE_M12N10_REC,
5568     char => FEATURE_M12N10_REC,
5569     charoff => FEATURE_M12N10_REC,
5570 wakaba 1.187 lang => FEATURE_HTML5_REC,
5571 wakaba 1.153 span => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5572 wakaba 1.49 valign => FEATURE_M12N10_REC,
5573     width => FEATURE_M12N10_REC,
5574 wakaba 1.1 }),
5575 wakaba 1.40 check_child_element => sub {
5576     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5577     $child_is_transparent, $element_state) = @_;
5578 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5579     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5580 wakaba 1.40 $self->{onerror}->(node => $child_el,
5581     type => 'element not allowed:minus',
5582 wakaba 1.104 level => $self->{level}->{must});
5583 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5584     #
5585     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'col') {
5586     #
5587     } else {
5588 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5589     level => $self->{level}->{must});
5590 wakaba 1.40 }
5591     },
5592     check_child_text => sub {
5593     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5594     if ($has_significant) {
5595 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5596     level => $self->{level}->{must});
5597 wakaba 1.1 }
5598     },
5599     };
5600    
5601     $Element->{$HTML_NS}->{col} = {
5602 wakaba 1.40 %HTMLEmptyChecker,
5603 wakaba 1.187 status => FEATURE_HTML5_REC,
5604 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5605 wakaba 1.69 %cellalign,
5606 wakaba 1.1 span => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5607 wakaba 1.49 }, {
5608     %HTMLAttrStatus,
5609 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5610 wakaba 1.49 align => FEATURE_M12N10_REC,
5611     char => FEATURE_M12N10_REC,
5612     charoff => FEATURE_M12N10_REC,
5613 wakaba 1.187 lang => FEATURE_HTML5_REC,
5614 wakaba 1.153 span => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5615 wakaba 1.49 valign => FEATURE_M12N10_REC,
5616     width => FEATURE_M12N10_REC,
5617 wakaba 1.1 }),
5618     };
5619    
5620     $Element->{$HTML_NS}->{tbody} = {
5621 wakaba 1.40 %HTMLChecker,
5622 wakaba 1.187 status => FEATURE_HTML5_REC,
5623 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
5624     %cellalign,
5625     }, {
5626 wakaba 1.49 %HTMLAttrStatus,
5627 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5628 wakaba 1.49 align => FEATURE_M12N10_REC,
5629     char => FEATURE_M12N10_REC,
5630     charoff => FEATURE_M12N10_REC,
5631 wakaba 1.187 lang => FEATURE_HTML5_REC,
5632 wakaba 1.49 valign => FEATURE_M12N10_REC,
5633     }),
5634 wakaba 1.40 check_child_element => sub {
5635     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5636     $child_is_transparent, $element_state) = @_;
5637 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5638     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5639 wakaba 1.40 $self->{onerror}->(node => $child_el,
5640     type => 'element not allowed:minus',
5641 wakaba 1.104 level => $self->{level}->{must});
5642 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5643     #
5644     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'tr') {
5645 wakaba 1.84 #
5646 wakaba 1.40 } else {
5647 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5648     level => $self->{level}->{must});
5649 wakaba 1.40 }
5650     },
5651     check_child_text => sub {
5652     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5653     if ($has_significant) {
5654 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5655     level => $self->{level}->{must});
5656 wakaba 1.1 }
5657 wakaba 1.40 },
5658 wakaba 1.1 };
5659    
5660     $Element->{$HTML_NS}->{thead} = {
5661 wakaba 1.40 %{$Element->{$HTML_NS}->{tbody}},
5662 wakaba 1.1 };
5663    
5664     $Element->{$HTML_NS}->{tfoot} = {
5665 wakaba 1.40 %{$Element->{$HTML_NS}->{tbody}},
5666 wakaba 1.1 };
5667    
5668     $Element->{$HTML_NS}->{tr} = {
5669 wakaba 1.40 %HTMLChecker,
5670 wakaba 1.187 status => FEATURE_HTML5_REC,
5671 wakaba 1.69 check_attrs => $GetHTMLAttrsChecker->({
5672     %cellalign,
5673     bgcolor => $HTMLColorAttrChecker,
5674     }, {
5675 wakaba 1.49 %HTMLAttrStatus,
5676 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5677 wakaba 1.49 align => FEATURE_M12N10_REC,
5678     bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5679     char => FEATURE_M12N10_REC,
5680     charoff => FEATURE_M12N10_REC,
5681 wakaba 1.187 lang => FEATURE_HTML5_REC,
5682 wakaba 1.49 valign => FEATURE_M12N10_REC,
5683     }),
5684 wakaba 1.40 check_child_element => sub {
5685     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5686     $child_is_transparent, $element_state) = @_;
5687 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5688     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5689 wakaba 1.40 $self->{onerror}->(node => $child_el,
5690     type => 'element not allowed:minus',
5691 wakaba 1.104 level => $self->{level}->{must});
5692 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5693     #
5694     } elsif ($child_nsuri eq $HTML_NS and
5695     ($child_ln eq 'td' or $child_ln eq 'th')) {
5696 wakaba 1.84 #
5697 wakaba 1.40 } else {
5698 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
5699     level => $self->{level}->{must});
5700 wakaba 1.40 }
5701     },
5702     check_child_text => sub {
5703     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5704     if ($has_significant) {
5705 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
5706     level => $self->{level}->{must});
5707 wakaba 1.1 }
5708     },
5709     };
5710    
5711     $Element->{$HTML_NS}->{td} = {
5712 wakaba 1.72 %HTMLFlowContentChecker,
5713 wakaba 1.187 status => FEATURE_HTML5_REC,
5714 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5715 wakaba 1.69 %cellalign,
5716     abbr => sub {}, ## NOTE: HTML4 %Text; and SHOULD be short.
5717     axis => sub {}, ## NOTE: HTML4 "cdata", comma-separated
5718     bgcolor => $HTMLColorAttrChecker,
5719 wakaba 1.1 colspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5720 wakaba 1.87 headers => sub {
5721     ## NOTE: Will be checked by Whatpm::HTMLTable->assign_header.
5722     ## Though that method does not check the |headers| attribute of a
5723     ## |td| element if the element does not form a table, in that case
5724     ## the |td| element is non-conforming anyway.
5725     },
5726 wakaba 1.69 nowrap => $GetHTMLBooleanAttrChecker->('nowrap'),
5727 wakaba 1.1 rowspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5728 wakaba 1.69 scope => $GetHTMLEnumeratedAttrChecker
5729     ->({row => 1, col => 1, rowgroup => 1, colgroup => 1}),
5730 wakaba 1.49 }, {
5731     %HTMLAttrStatus,
5732 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5733     abbr => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5734 wakaba 1.49 align => FEATURE_M12N10_REC,
5735 wakaba 1.82 axis => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5736 wakaba 1.49 bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5737     char => FEATURE_M12N10_REC,
5738     charoff => FEATURE_M12N10_REC,
5739 wakaba 1.153 colspan => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5740 wakaba 1.187 headers => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5741 wakaba 1.49 height => FEATURE_M12N10_REC_DEPRECATED,
5742 wakaba 1.187 lang => FEATURE_HTML5_REC,
5743 wakaba 1.49 nowrap => FEATURE_M12N10_REC_DEPRECATED,
5744 wakaba 1.153 rowspan => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5745 wakaba 1.82 scope => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5746 wakaba 1.49 valign => FEATURE_M12N10_REC,
5747     width => FEATURE_M12N10_REC_DEPRECATED,
5748 wakaba 1.1 }),
5749     };
5750    
5751     $Element->{$HTML_NS}->{th} = {
5752 wakaba 1.40 %HTMLPhrasingContentChecker,
5753 wakaba 1.187 status => FEATURE_HTML5_REC,
5754 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
5755 wakaba 1.69 %cellalign,
5756     abbr => sub {}, ## NOTE: HTML4 %Text; and SHOULD be short.
5757     axis => sub {}, ## NOTE: HTML4 "cdata", comma-separated
5758     bgcolor => $HTMLColorAttrChecker,
5759 wakaba 1.1 colspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5760 wakaba 1.87 ## TODO: HTML4(?) |headers|
5761 wakaba 1.69 nowrap => $GetHTMLBooleanAttrChecker->('nowrap'),
5762 wakaba 1.1 rowspan => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
5763     scope => $GetHTMLEnumeratedAttrChecker
5764     ->({row => 1, col => 1, rowgroup => 1, colgroup => 1}),
5765 wakaba 1.49 }, {
5766     %HTMLAttrStatus,
5767 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
5768     abbr => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5769 wakaba 1.49 align => FEATURE_M12N10_REC,
5770 wakaba 1.82 axis => FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5771 wakaba 1.49 bgcolor => FEATURE_M12N10_REC_DEPRECATED,
5772     char => FEATURE_M12N10_REC,
5773     charoff => FEATURE_M12N10_REC,
5774 wakaba 1.153 colspan => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5775 wakaba 1.187 headers => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5776 wakaba 1.49 height => FEATURE_M12N10_REC_DEPRECATED,
5777 wakaba 1.187 lang => FEATURE_HTML5_REC,
5778 wakaba 1.49 nowrap => FEATURE_M12N10_REC_DEPRECATED,
5779 wakaba 1.153 rowspan => FEATURE_HTML5_LC | FEATURE_XHTML2_ED | FEATURE_M12N10_REC,
5780 wakaba 1.187 scope => FEATURE_HTML5_REC,
5781 wakaba 1.49 valign => FEATURE_M12N10_REC,
5782     width => FEATURE_M12N10_REC_DEPRECATED,
5783 wakaba 1.1 }),
5784     };
5785    
5786 wakaba 1.52 $Element->{$HTML_NS}->{form} = {
5787 wakaba 1.121 %HTMLFlowContentChecker,
5788 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_WF2X | FEATURE_M12N10_REC,
5789 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
5790 wakaba 1.161 accept => $AcceptAttrChecker,
5791 wakaba 1.129 'accept-charset' => $HTMLCharsetsAttrChecker,
5792 wakaba 1.166 action => $HTMLURIAttrChecker, ## TODO: Warn if submission is not defined for the scheme
5793 wakaba 1.185 autocomplete => $GetHTMLEnumeratedAttrChecker->({
5794     on => 1, off => 1,
5795     }),
5796 wakaba 1.56 data => $HTMLURIAttrChecker, ## TODO: MUST point ... [WF2]
5797 wakaba 1.166 enctype => $GetHTMLEnumeratedAttrChecker->({
5798     'application/x-www-form-urlencoded' => 1,
5799     'multipart/form-data' => 1,
5800     'text/plain' => 1,
5801     }),
5802 wakaba 1.56 method => $GetHTMLEnumeratedAttrChecker->({
5803     get => 1, post => 1, put => 1, delete => 1,
5804     }),
5805 wakaba 1.133 name => sub {
5806     my ($self, $attr) = @_;
5807    
5808     my $value = $attr->value;
5809     if ($value eq '') {
5810     $self->{onerror}->(type => 'empty form name',
5811     node => $attr,
5812     level => $self->{level}->{must});
5813     } else {
5814     if ($self->{form}->{$value}) {
5815     $self->{onerror}->(type => 'duplicate form name',
5816     node => $attr,
5817     value => $value,
5818     level => $self->{level}->{must});
5819     } else {
5820     $self->{form}->{$value} = 1;
5821     }
5822     }
5823     },
5824 wakaba 1.166 novalidate => $GetHTMLBooleanAttrChecker->('novalidate'),
5825 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
5826     onforminput => $HTMLEventHandlerAttrChecker,
5827 wakaba 1.56 onreceived => $HTMLEventHandlerAttrChecker,
5828     replace => $GetHTMLEnumeratedAttrChecker->({document => 1, values => 1}),
5829 wakaba 1.52 target => $HTMLTargetAttrChecker,
5830     }, {
5831     %HTMLAttrStatus,
5832     %HTMLM12NCommonAttrStatus,
5833 wakaba 1.161 accept => FEATURE_HTML5_DROPPED | FEATURE_WF2X | FEATURE_M12N10_REC,
5834 wakaba 1.187 'accept-charset' => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
5835 wakaba 1.119 action => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5836 wakaba 1.185 autocomplete => FEATURE_HTML5_WD,
5837 wakaba 1.56 data => FEATURE_WF2,
5838 wakaba 1.119 enctype => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5839 wakaba 1.187 lang => FEATURE_HTML5_REC,
5840 wakaba 1.119 method => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5841     #name => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC_DEPRECATED,
5842 wakaba 1.187 name => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
5843 wakaba 1.207 novalidate => FEATURE_HTML5_LC,
5844 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
5845     onforminput => FEATURE_WF2_INFORMATIVE,
5846 wakaba 1.56 onreceived => FEATURE_WF2,
5847 wakaba 1.52 onreset => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5848     onsubmit => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5849 wakaba 1.56 replace => FEATURE_WF2,
5850 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
5851     sdasuff => FEATURE_HTML20_RFC,
5852 wakaba 1.119 target => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5853 wakaba 1.52 }),
5854 wakaba 1.66 check_start => sub {
5855     my ($self, $item, $element_state) = @_;
5856 wakaba 1.121 $self->_add_minus_elements ($element_state, {$HTML_NS => {form => 1}});
5857 wakaba 1.66
5858     $element_state->{uri_info}->{action}->{type}->{action} = 1;
5859     $element_state->{uri_info}->{data}->{type}->{resource} = 1;
5860 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
5861     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
5862 wakaba 1.136 $element_state->{id_type} = 'form';
5863 wakaba 1.66 },
5864 wakaba 1.121 check_end => sub {
5865     my ($self, $item, $element_state) = @_;
5866     $self->_remove_minus_elements ($element_state);
5867    
5868     $HTMLFlowContentChecker{check_end}->(@_);
5869     },
5870 wakaba 1.185 }; # form
5871 wakaba 1.52
5872     $Element->{$HTML_NS}->{fieldset} = {
5873 wakaba 1.134 %HTMLFlowContentChecker,
5874 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_WF2X | FEATURE_M12N10_REC,
5875 wakaba 1.56 check_attrs => $GetHTMLAttrsChecker->({
5876     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
5877 wakaba 1.136 form => $HTMLFormAttrChecker,
5878 wakaba 1.165 name => $FormControlNameAttrChecker,
5879 wakaba 1.56 }, {
5880 wakaba 1.52 %HTMLAttrStatus,
5881     %HTMLM12NCommonAttrStatus,
5882 wakaba 1.187 disabled => FEATURE_HTML5_WD | FEATURE_WF2X,
5883     form => FEATURE_HTML5_LC | FEATURE_WF2X,
5884     lang => FEATURE_HTML5_REC,
5885     name => FEATURE_HTML5_LC,
5886 wakaba 1.52 }),
5887 wakaba 1.134 ## NOTE: legend, Flow
5888     check_child_element => sub {
5889     my ($self, $item, $child_el, $child_nsuri, $child_ln,
5890     $child_is_transparent, $element_state) = @_;
5891     if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
5892     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
5893     $self->{onerror}->(node => $child_el,
5894     type => 'element not allowed:minus',
5895     level => $self->{level}->{must});
5896     $element_state->{has_non_legend} = 1;
5897     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
5898     #
5899     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'legend') {
5900     if ($element_state->{has_non_legend}) {
5901     $self->{onerror}->(node => $child_el,
5902     type => 'element not allowed:details legend',
5903     level => $self->{level}->{must});
5904     }
5905     $element_state->{has_legend} = 1;
5906     $element_state->{has_non_legend} = 1;
5907     } else {
5908     $HTMLFlowContentChecker{check_child_element}->(@_);
5909     $element_state->{has_non_legend} = 1 unless $child_is_transparent;
5910     ## TODO:
5911 wakaba 1.167 ## |<fieldset><object><legend>xx</legend></object>..</fieldset>|
5912 wakaba 1.134 ## should be an error, since |object| is allowed as flow,
5913     ## therefore |details| part of the content model does not match.
5914     }
5915     },
5916     check_child_text => sub {
5917     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
5918     if ($has_significant) {
5919     $element_state->{has_non_legend} = 1;
5920     }
5921     },
5922     check_end => sub {
5923     my ($self, $item, $element_state) = @_;
5924    
5925     unless ($element_state->{has_legend}) {
5926     $self->{onerror}->(node => $item->{node},
5927     type => 'child element missing',
5928     text => 'legend',
5929     level => $self->{level}->{must});
5930     }
5931    
5932     $HTMLFlowContentChecker{check_end}->(@_);
5933 wakaba 1.167 ## ISSUE: |<fieldset><legend>aa</legend></fieldset>| error?
5934 wakaba 1.134 },
5935     ## NOTE: This definition is partially reused by |details| element's
5936     ## checker.
5937 wakaba 1.52 };
5938    
5939     $Element->{$HTML_NS}->{input} = {
5940 wakaba 1.119 %HTMLEmptyChecker,
5941 wakaba 1.187 status => FEATURE_HTML5_WD | FEATURE_WF2X | FEATURE_M12N10_REC,
5942 wakaba 1.140 check_attrs => sub {
5943     my ($self, $item, $element_state) = @_;
5944 wakaba 1.142
5945 wakaba 1.145 my $state = $item->{node}->get_attribute_ns (undef, 'type');
5946 wakaba 1.142 $state = 'text' unless defined $state;
5947     $state =~ tr/A-Z/a-z/; ## ASCII case-insensitive
5948    
5949 wakaba 1.140 for my $attr (@{$item->{node}->attributes}) {
5950     my $attr_ns = $attr->namespace_uri;
5951     $attr_ns = '' unless defined $attr_ns;
5952     my $attr_ln = $attr->manakai_local_name;
5953     my $checker;
5954     my $status;
5955     if ($attr_ns eq '') {
5956     $status =
5957     {
5958     %HTMLAttrStatus,
5959     %HTMLM12NCommonAttrStatus,
5960     accept => FEATURE_HTML5_DEFAULT | FEATURE_WF2X | FEATURE_M12N10_REC,
5961     'accept-charset' => FEATURE_HTML2X_RFC,
5962 wakaba 1.176 accesskey => FEATURE_M12N10_REC | FEATURE_HTML5_FD,
5963 wakaba 1.198 action => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
5964 wakaba 1.140 align => FEATURE_M12N10_REC_DEPRECATED,
5965     alt => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5966 wakaba 1.185 autocomplete => FEATURE_HTML5_LC | FEATURE_WF2X,
5967 wakaba 1.187 autofocus => FEATURE_HTML5_LC | FEATURE_WF2X,
5968     checked => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
5969 wakaba 1.140 datafld => FEATURE_HTML4_REC_RESERVED,
5970     dataformatas => FEATURE_HTML4_REC_RESERVED,
5971     datasrc => FEATURE_HTML4_REC_RESERVED,
5972 wakaba 1.187 disabled => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
5973 wakaba 1.198 enctype => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
5974 wakaba 1.187 form => FEATURE_HTML5_LC | FEATURE_WF2X,
5975 wakaba 1.198 formaction => FEATURE_HTML5_LC,
5976     formenctype => FEATURE_HTML5_LC,
5977     formmethod => FEATURE_HTML5_LC,
5978     formnovalidate => FEATURE_HTML5_LC,
5979     formtarget => FEATURE_HTML5_LC,
5980 wakaba 1.178 height => FEATURE_HTML5_LC,
5981 wakaba 1.150 inputmode => FEATURE_HTML5_DROPPED | FEATURE_WF2X |
5982     FEATURE_XHTMLBASIC11_CR,
5983 wakaba 1.140 ismap => FEATURE_M12N10_REC,
5984 wakaba 1.187 lang => FEATURE_HTML5_REC,
5985     list => FEATURE_HTML5_LC | FEATURE_WF2X,
5986     max => FEATURE_HTML5_LC | FEATURE_WF2X,
5987     maxlength => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
5988 wakaba 1.198 method => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
5989 wakaba 1.187 min => FEATURE_HTML5_LC | FEATURE_WF2X,
5990     multiple => FEATURE_HTML5_LC,
5991     name => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
5992 wakaba 1.198 novalidate => FEATURE_HTML5_DROPPED,
5993 wakaba 1.140 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5994     onchange => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5995     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
5996     onformchange => FEATURE_WF2_INFORMATIVE,
5997     onforminput => FEATURE_WF2_INFORMATIVE,
5998     oninput => FEATURE_WF2,
5999     oninvalid => FEATURE_WF2,
6000     onselect => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6001 wakaba 1.187 pattern => FEATURE_HTML5_LC | FEATURE_WF2X,
6002     placeholder => FEATURE_HTML5_LC,
6003     readonly => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6004 wakaba 1.140 replace => FEATURE_WF2,
6005 wakaba 1.187 required => FEATURE_HTML5_LC | FEATURE_WF2X,
6006 wakaba 1.140 sdapref => FEATURE_HTML20_RFC,
6007 wakaba 1.187 size => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6008 wakaba 1.140 src => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6009 wakaba 1.187 step => FEATURE_HTML5_LC | FEATURE_WF2X,
6010 wakaba 1.140 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6011 wakaba 1.198 target => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
6012 wakaba 1.208 template => FEATURE_HTML5_DROPPED | FEATURE_WF2,
6013 wakaba 1.187 type => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
6014 wakaba 1.140 usemap => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC,
6015 wakaba 1.187 value => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
6016 wakaba 1.178 width => FEATURE_HTML5_LC,
6017 wakaba 1.140 }->{$attr_ln};
6018    
6019     $checker =
6020     {
6021 wakaba 1.141 ## NOTE: Value of an empty string means that the attribute is only
6022     ## applicable for a specific set of states.
6023 wakaba 1.142 accept => '',
6024 wakaba 1.149 'accept-charset' => $HTMLCharsetsAttrChecker,
6025     ## NOTE: To which states it applies is not defined in RFC 2070.
6026 wakaba 1.142 action => '',
6027 wakaba 1.150 align => '',
6028 wakaba 1.141 alt => '',
6029 wakaba 1.142 autocomplete => '',
6030 wakaba 1.165 autofocus => $AutofocusAttrChecker,
6031     ## NOTE: <input type=hidden disabled> is not disallowed.
6032 wakaba 1.142 checked => '',
6033     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6034 wakaba 1.165 ## NOTE: <input type=hidden disabled> is not disallowed.
6035 wakaba 1.142 enctype => '',
6036     form => $HTMLFormAttrChecker,
6037 wakaba 1.198 formaction => '',
6038     formenctype => '',
6039     formmethod => '',
6040     formnovalidate => '',
6041     formtarget => '',
6042 wakaba 1.178 height => '',
6043 wakaba 1.150 inputmode => '',
6044     ismap => '', ## NOTE: "MUST" be type=image [HTML4]
6045 wakaba 1.142 list => '',
6046     max => '',
6047     maxlength => '',
6048     method => '',
6049     min => '',
6050 wakaba 1.156 multiple => '',
6051 wakaba 1.165 name => $FormControlNameAttrChecker,
6052 wakaba 1.166 novalidate => '',
6053 wakaba 1.149 onformchange => $HTMLEventHandlerAttrChecker, # [WF2]
6054     onforminput => $HTMLEventHandlerAttrChecker, # [WF2]
6055     oninput => $HTMLEventHandlerAttrChecker, # [WF2]
6056     oninvalid => $HTMLEventHandlerAttrChecker, # [WF2]
6057     ## TODO: tests for four attributes above
6058 wakaba 1.142 pattern => '',
6059 wakaba 1.156 placeholder => '',
6060 wakaba 1.142 readonly => '',
6061 wakaba 1.150 replace => '',
6062 wakaba 1.142 required => '',
6063     size => '',
6064     src => '',
6065     step => '',
6066     target => '',
6067 wakaba 1.140 type => $GetHTMLEnumeratedAttrChecker->({
6068 wakaba 1.156 hidden => 1, text => 1, search => 1, url => 1,
6069 wakaba 1.193 tel => 1, email => 1, password => 1,
6070 wakaba 1.141 datetime => 1, date => 1, month => 1, week => 1, time => 1,
6071 wakaba 1.157 'datetime-local' => 1, number => 1, range => 1, color => 1,
6072     checkbox => 1,
6073 wakaba 1.141 radio => 1, file => 1, submit => 1, image => 1, reset => 1,
6074     button => 1,
6075 wakaba 1.140 }),
6076 wakaba 1.151 usemap => '',
6077 wakaba 1.142 value => '',
6078 wakaba 1.178 width => '',
6079 wakaba 1.140 }->{$attr_ln};
6080 wakaba 1.141
6081     ## State-dependent checkers
6082     unless ($checker) {
6083     if ($state eq 'hidden') {
6084     $checker =
6085     {
6086 wakaba 1.142 value => sub {
6087     my ($self, $attr, $item, $element_state) = @_;
6088 wakaba 1.145 my $name = $item->{node}->get_attribute_ns (undef, 'name');
6089 wakaba 1.142 if (defined $name and $name eq '_charset_') { ## case-sensitive
6090     $self->{onerror}->(node => $attr,
6091     type => '_charset_ value',
6092     level => $self->{level}->{must});
6093     }
6094     },
6095 wakaba 1.141 }->{$attr_ln} || $checker;
6096 wakaba 1.142 ## TODO: Warn if no name attribute?
6097     ## TODO: Warn if name!=_charset_ and no value attribute?
6098 wakaba 1.168 } elsif ({
6099     datetime => 1, date => 1, month => 1, time => 1,
6100     week => 1, 'datetime-local' => 1,
6101     }->{$state}) {
6102     my $v = {
6103     datetime => ['global_date_and_time_string'],
6104     date => ['date_string'],
6105     month => ['month_string'],
6106     week => ['week_string'],
6107     time => ['time_string'],
6108     'datetime-local' => ['local_date_and_time_string'],
6109     }->{$state};
6110 wakaba 1.144 $checker =
6111     {
6112 wakaba 1.145 autocomplete => $GetHTMLEnumeratedAttrChecker->({
6113     on => 1, off => 1,
6114     }),
6115 wakaba 1.158 list => $ListAttrChecker,
6116 wakaba 1.168 min => $GetDateTimeAttrChecker->($v->[0]),
6117     max => $GetDateTimeAttrChecker->($v->[0]),
6118 wakaba 1.145 readonly => $GetHTMLBooleanAttrChecker->('readonly'),
6119 wakaba 1.146 required => $GetHTMLBooleanAttrChecker->('required'),
6120 wakaba 1.148 step => $StepAttrChecker,
6121 wakaba 1.168 value => $GetDateTimeAttrChecker->($v->[0]),
6122 wakaba 1.144 }->{$attr_ln} || $checker;
6123 wakaba 1.205
6124     ## XXX Maybe it is better to check min <= value <= max
6125     ## relation is hold for convinience?
6126 wakaba 1.144 } elsif ($state eq 'number') {
6127     $checker =
6128     {
6129 wakaba 1.145 autocomplete => $GetHTMLEnumeratedAttrChecker->({
6130     on => 1, off => 1,
6131     }),
6132 wakaba 1.158 list => $ListAttrChecker,
6133 wakaba 1.144 max => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
6134     min => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
6135 wakaba 1.145 readonly => $GetHTMLBooleanAttrChecker->('readonly'),
6136 wakaba 1.146 required => $GetHTMLBooleanAttrChecker->('required'),
6137 wakaba 1.148 step => $StepAttrChecker,
6138 wakaba 1.168 value => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
6139 wakaba 1.144 }->{$attr_ln} || $checker;
6140     } elsif ($state eq 'range') {
6141     $checker =
6142     {
6143 wakaba 1.145 autocomplete => $GetHTMLEnumeratedAttrChecker->({
6144     on => 1, off => 1,
6145     }),
6146 wakaba 1.158 list => $ListAttrChecker,
6147 wakaba 1.144 max => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
6148     min => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
6149 wakaba 1.148 step => $StepAttrChecker,
6150 wakaba 1.168 value => $GetHTMLFloatingPointNumberAttrChecker->(sub { 1 }),
6151 wakaba 1.144 }->{$attr_ln} || $checker;
6152 wakaba 1.157 } elsif ($state eq 'color') {
6153     $checker =
6154     {
6155     autocomplete => $GetHTMLEnumeratedAttrChecker->({
6156     on => 1, off => 1,
6157     }),
6158 wakaba 1.158 list => $ListAttrChecker,
6159 wakaba 1.157 value => sub {
6160     my ($self, $attr) = @_;
6161     unless ($attr->value =~ /\A#[0-9A-Fa-f]{6}\z/) {
6162     $self->{onerror}->(node => $attr,
6163     type => 'scolor:syntax error', ## TODOC: type
6164     level => $self->{level}->{must});
6165     }
6166     },
6167     }->{$attr_ln} || $checker;
6168 wakaba 1.144 } elsif ($state eq 'checkbox' or $state eq 'radio') {
6169     $checker =
6170     {
6171 wakaba 1.149 checked => $GetHTMLBooleanAttrChecker->('checked'),
6172     ## TODO: tests
6173 wakaba 1.146 required => $GetHTMLBooleanAttrChecker->('required'),
6174 wakaba 1.144 value => sub { }, ## NOTE: No restriction.
6175     }->{$attr_ln} || $checker;
6176     ## TODO: There MUST be another input type=radio with same
6177     ## name (Radio state).
6178     ## ISSUE: There should be exactly one type=radio with checked?
6179     } elsif ($state eq 'file') {
6180     $checker =
6181     {
6182 wakaba 1.161 accept => $AcceptAttrChecker,
6183 wakaba 1.168 ## max (default 1) & min (default 0) [WF2]: Dropped by HTML5.
6184 wakaba 1.159 multiple => $GetHTMLBooleanAttrChecker->('multiple'),
6185 wakaba 1.146 required => $GetHTMLBooleanAttrChecker->('required'),
6186 wakaba 1.144 }->{$attr_ln} || $checker;
6187     } elsif ($state eq 'submit') {
6188     $checker =
6189     {
6190 wakaba 1.149 action => $HTMLURIAttrChecker,
6191 wakaba 1.166 enctype => $GetHTMLEnumeratedAttrChecker->({
6192     'application/x-www-form-urlencoded' => 1,
6193     'multipart/form-data' => 1,
6194     'text/plain' => 1,
6195     }),
6196 wakaba 1.198 formaction => $HTMLURIAttrChecker,
6197     formenctype => $GetHTMLEnumeratedAttrChecker->({
6198     'application/x-www-form-urlencoded' => 1,
6199     'multipart/form-data' => 1,
6200     'text/plain' => 1,
6201     }),
6202     formmethod => $GetHTMLEnumeratedAttrChecker->({
6203     get => 1, post => 1, put => 1, delete => 1,
6204     }),
6205     formnovalidate => $GetHTMLBooleanAttrChecker->('formnovalidate'),
6206     formtarget => $HTMLTargetAttrChecker,
6207 wakaba 1.149 method => $GetHTMLEnumeratedAttrChecker->({
6208     get => 1, post => 1, put => 1, delete => 1,
6209     }),
6210 wakaba 1.166 novalidate => $GetHTMLBooleanAttrChecker->('novalidate'),
6211 wakaba 1.149 replace => $GetHTMLEnumeratedAttrChecker->({
6212     document => 1, values => 1,
6213     }),
6214     target => $HTMLTargetAttrChecker,
6215 wakaba 1.144 value => sub { }, ## NOTE: No restriction.
6216     }->{$attr_ln} || $checker;
6217     } elsif ($state eq 'image') {
6218     $checker =
6219     {
6220 wakaba 1.149 action => $HTMLURIAttrChecker,
6221     align => $GetHTMLEnumeratedAttrChecker->({
6222     top => 1, middle => 1, bottom => 1, left => 1, right => 1,
6223     }),
6224 wakaba 1.144 alt => sub {
6225     my ($self, $attr) = @_;
6226     my $value = $attr->value;
6227     unless (length $value) {
6228     $self->{onerror}->(node => $attr,
6229     type => 'empty anchor image alt',
6230     level => $self->{level}->{must});
6231     }
6232     },
6233 wakaba 1.166 enctype => $GetHTMLEnumeratedAttrChecker->({
6234     'application/x-www-form-urlencoded' => 1,
6235     'multipart/form-data' => 1,
6236     'text/plain' => 1,
6237     }),
6238 wakaba 1.198 formaction => $HTMLURIAttrChecker,
6239     formenctype => $GetHTMLEnumeratedAttrChecker->({
6240     'application/x-www-form-urlencoded' => 1,
6241     'multipart/form-data' => 1,
6242     'text/plain' => 1,
6243     }),
6244     formmethod => $GetHTMLEnumeratedAttrChecker->({
6245     get => 1, post => 1, put => 1, delete => 1,
6246     }),
6247     formnovalidate => $GetHTMLBooleanAttrChecker->('formnovalidate'),
6248     formtarget => $HTMLTargetAttrChecker,
6249 wakaba 1.178 height => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
6250 wakaba 1.149 ismap => $GetHTMLBooleanAttrChecker->('ismap'),
6251     method => $GetHTMLEnumeratedAttrChecker->({
6252     get => 1, post => 1, put => 1, delete => 1,
6253     }),
6254 wakaba 1.166 novalidate => $GetHTMLBooleanAttrChecker->('novalidate'),
6255 wakaba 1.149 replace => $GetHTMLEnumeratedAttrChecker->({
6256     document => 1, values => 1,
6257     }),
6258 wakaba 1.144 src => $HTMLURIAttrChecker,
6259     ## TODO: There is requirements on the referenced resource.
6260 wakaba 1.149 target => $HTMLTargetAttrChecker,
6261     usemap => $HTMLUsemapAttrChecker,
6262 wakaba 1.178 width => $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 }),
6263 wakaba 1.144 }->{$attr_ln} || $checker;
6264     ## TODO: alt & src are required.
6265     } elsif ({
6266     reset => 1, button => 1,
6267     ## NOTE: From Web Forms 2.0:
6268     remove => 1, 'move-up' => 1, 'move-down' => 1,
6269     add => 1,
6270     }->{$state}) {
6271     $checker =
6272     {
6273     ## NOTE: According to Web Forms 2.0, |input| attribute
6274     ## has |template| attribute to support the |add| button
6275     ## type (as part of the repetition template feature). It
6276     ## conflicts with the |template| global attribute
6277     ## introduced as part of the data template feature.
6278     ## NOTE: |template| attribute as defined in Web Forms 2.0
6279     ## has no author requirement.
6280     value => sub { }, ## NOTE: No restriction.
6281     }->{$attr_ln} || $checker;
6282 wakaba 1.193 } else { # Text, Search, E-mail, URL, Telephone, Password
6283 wakaba 1.141 $checker =
6284     {
6285 wakaba 1.145 autocomplete => $GetHTMLEnumeratedAttrChecker->({
6286     on => 1, off => 1,
6287     }),
6288 wakaba 1.149 ## TODO: inputmode [WF2]
6289 wakaba 1.158 list => $ListAttrChecker,
6290 wakaba 1.147 maxlength => sub {
6291     my ($self, $attr, $item, $element_state) = @_;
6292    
6293     $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 })->(@_);
6294    
6295 wakaba 1.165 if ($attr->value =~ /^[\x09\x0A\x0C\x0D\x20]*([0-9]+)/) {
6296 wakaba 1.147 ## NOTE: Applying the rules for parsing non-negative
6297     ## integers results in a number.
6298     my $max_allowed_value_length = 0+$1;
6299    
6300     my $value = $item->{node}->get_attribute_ns (undef, 'value');
6301     if (defined $value) {
6302     my $codepoint_length = length $value;
6303 wakaba 1.162
6304 wakaba 1.147 if ($codepoint_length > $max_allowed_value_length) {
6305     $self->{onerror}
6306     ->(node => $item->{node}
6307     ->get_attribute_node_ns (undef, 'value'),
6308     type => 'value too long',
6309     level => $self->{level}->{must});
6310     }
6311     }
6312     }
6313     },
6314 wakaba 1.160 pattern => $PatternAttrChecker,
6315 wakaba 1.179 placeholder => $PlaceholderAttrChecker,
6316 wakaba 1.145 readonly => $GetHTMLBooleanAttrChecker->('readonly'),
6317 wakaba 1.146 required => $GetHTMLBooleanAttrChecker->('required'),
6318 wakaba 1.147 size => $GetHTMLNonNegativeIntegerAttrChecker->(sub {shift > 0}),
6319 wakaba 1.143 value => sub {
6320 wakaba 1.156 my ($self, $attr, $item, $element_state) = @_;
6321     if ($state eq 'url') {
6322 wakaba 1.205 ## XXX MUST be absolute IRI.
6323 wakaba 1.156 $HTMLURIAttrChecker->(@_);
6324     } elsif ($state eq 'email') {
6325     if ($item->{node}->has_attribute_ns (undef, 'multiple')) {
6326 wakaba 1.200 ## A set of comma-separated tokens.
6327 wakaba 1.156 my @addr = split /,/, $attr->value, -1;
6328     @addr = ('') unless @addr;
6329     for (@addr) {
6330 wakaba 1.165 s/\A[\x09\x0A\x0C\x0D\x20]+//;
6331     s/[\x09\x0A\x0C\x0D\x20]\z//;
6332 wakaba 1.200
6333     unless (/\A$ValidEmailAddress\z/o) {
6334 wakaba 1.156 $self->{onerror}->(node => $attr,
6335     type => 'email:syntax error', ## TODO: type
6336     value => $_,
6337     level => $self->{level}->{must});
6338     }
6339     }
6340     } else {
6341     unless ($attr->value =~ /\A$ValidEmailAddress\z/) {
6342     $self->{onerror}->(node => $attr,
6343     type => 'email:syntax error', ## TODO: type
6344     level => $self->{level}->{must});
6345     }
6346     }
6347     } else {
6348     if ($attr->value =~ /[\x0D\x0A]/) {
6349     $self->{onerror}->(node => $attr,
6350     type => 'newline in value', ## TODO: type
6351     level => $self->{level}->{must});
6352     }
6353     }
6354 wakaba 1.143 },
6355 wakaba 1.141 }->{$attr_ln} || $checker;
6356 wakaba 1.147 $checker = '' if $state eq 'password' and $attr_ln eq 'list';
6357 wakaba 1.156 $checker = $GetHTMLBooleanAttrChecker->('multiple')
6358     if $state eq 'email' and $attr_ln eq 'multiple';
6359 wakaba 1.161
6360     if ($item->{node}->has_attribute_ns (undef, 'pattern') and
6361     not $item->{node}->has_attribute_ns (undef, 'title')) {
6362     $self->{onerror}->(node => $item->{node},
6363     type => 'attribute missing',
6364     text => 'title',
6365     level => $self->{level}->{should});
6366     }
6367 wakaba 1.141 }
6368     }
6369    
6370     if (defined $checker) {
6371     if ($checker eq '') {
6372     $checker = sub {
6373     my ($self, $attr) = @_;
6374     $self->{onerror}->(node => $attr,
6375     type => 'input attr not applicable',
6376     text => $state,
6377     level => $self->{level}->{must});
6378     };
6379     }
6380 wakaba 1.140 } elsif ($attr_ln =~ /^data-\p{InXMLNCNameChar10}+\z/ and
6381     $attr_ln !~ /[A-Z]/) {
6382     $checker = $HTMLDatasetAttrChecker;
6383     $status = $HTMLDatasetAttrStatus;
6384     } else {
6385     $checker = $HTMLAttrChecker->{$attr_ln};
6386     }
6387     }
6388     $checker ||= $AttrChecker->{$attr_ns}->{$attr_ln}
6389     || $AttrChecker->{$attr_ns}->{''};
6390     $status ||= $AttrStatus->{$attr_ns}->{$attr_ln}
6391     || $AttrStatus->{$attr_ns}->{''};
6392     $status = FEATURE_ALLOWED if not defined $status and length $attr_ns;
6393 wakaba 1.157
6394 wakaba 1.140 if ($checker) {
6395     $checker->($self, $attr, $item, $element_state) if ref $checker;
6396     } elsif ($attr_ns eq '' and not $status) {
6397     #
6398     } else {
6399     $self->{onerror}->(node => $attr,
6400     type => 'unknown attribute',
6401     level => $self->{level}->{uncertain});
6402     ## ISSUE: No comformance createria for unknown attributes in the spec
6403     }
6404    
6405     $self->_attr_status_info ($attr, $status);
6406     }
6407 wakaba 1.168
6408     ## ISSUE: -0/+0
6409    
6410     if ($state eq 'range') {
6411     $element_state->{number_value}->{min} ||= 0;
6412     $element_state->{number_value}->{max} = 100
6413     unless defined $element_state->{number_value}->{max};
6414     }
6415    
6416     if (defined $element_state->{date_value}->{min} or
6417     defined $element_state->{date_value}->{max}) {
6418     my $min_value = $element_state->{date_value}->{min};
6419     my $max_value = $element_state->{date_value}->{max};
6420     my $value_value = $element_state->{date_value}->{value};
6421    
6422     if (defined $min_value and $min_value eq '' and
6423     (defined $max_value or defined $value_value)) {
6424     my $min = $item->{node}->get_attribute_node_ns (undef, 'min');
6425     $self->{onerror}->(node => $min,
6426     type => 'date value not supported', ## TODOC: type
6427     value => $min->value,
6428     level => $self->{level}->{unsupported});
6429     undef $min_value;
6430     }
6431     if (defined $max_value and $max_value eq '' and
6432     (defined $max_value or defined $value_value)) {
6433     my $max = $item->{node}->get_attribute_node_ns (undef, 'max');
6434     $self->{onerror}->(node => $max,
6435     type => 'date value not supported', ## TODOC: type
6436     value => $max->value,
6437     level => $self->{level}->{unsupported});
6438     undef $max_value;
6439     }
6440     if (defined $value_value and $value_value eq '' and
6441     (defined $max_value or defined $min_value)) {
6442     my $value = $item->{node}->get_attribute_node_ns (undef, 'value');
6443     $self->{onerror}->(node => $value,
6444     type => 'date value not supported', ## TODOC: type
6445     value => $value->value,
6446     level => $self->{level}->{unsupported});
6447     undef $value_value;
6448     }
6449    
6450     if (defined $min_value and defined $max_value) {
6451     if ($min_value->to_html5_number > $max_value->to_html5_number) {
6452     my $max = $item->{node}->get_attribute_node_ns (undef, 'max');
6453     $self->{onerror}->(node => $max,
6454     type => 'max lt min', ## TODOC: type
6455     level => $self->{level}->{must});
6456     }
6457     }
6458    
6459     if (defined $min_value and defined $value_value) {
6460     if ($min_value->to_html5_number > $value_value->to_html5_number) {
6461     my $value = $item->{node}->get_attribute_node_ns (undef, 'value');
6462     $self->{onerror}->(node => $value,
6463     type => 'value lt min', ## TODOC: type
6464     level => $self->{level}->{warn});
6465     ## NOTE: Not an error.
6466     }
6467     }
6468    
6469     if (defined $max_value and defined $value_value) {
6470     if ($max_value->to_html5_number < $value_value->to_html5_number) {
6471     my $value = $item->{node}->get_attribute_node_ns (undef, 'value');
6472     $self->{onerror}->(node => $value,
6473     type => 'value gt max', ## TODOC: type
6474     level => $self->{level}->{warn});
6475     ## NOTE: Not an error.
6476     }
6477     }
6478     } elsif (defined $element_state->{number_value}->{min} or
6479     defined $element_state->{number_value}->{max}) {
6480     my $min_value = $element_state->{number_value}->{min};
6481     my $max_value = $element_state->{number_value}->{max};
6482     my $value_value = $element_state->{number_value}->{value};
6483    
6484     if (defined $min_value and defined $max_value) {
6485     if ($min_value > $max_value) {
6486     my $max = $item->{node}->get_attribute_node_ns (undef, 'max');
6487     $self->{onerror}->(node => $max,
6488     type => 'max lt min', ## TODOC: type
6489     level => $self->{level}->{must});
6490     }
6491     }
6492    
6493     if (defined $min_value and defined $value_value) {
6494     if ($min_value > $value_value) {
6495     my $value = $item->{node}->get_attribute_node_ns (undef, 'value');
6496     $self->{onerror}->(node => $value,
6497     type => 'value lt min', ## TODOC: type
6498     level => $self->{level}->{warn});
6499     ## NOTE: Not an error.
6500     }
6501     }
6502    
6503     if (defined $max_value and defined $value_value) {
6504     if ($max_value < $value_value) {
6505     my $value = $item->{node}->get_attribute_node_ns (undef, 'value');
6506     $self->{onerror}->(node => $value,
6507     type => 'value gt max', ## TODOC: type
6508     level => $self->{level}->{warn});
6509     ## NOTE: Not an error.
6510     }
6511     }
6512     }
6513 wakaba 1.150
6514 wakaba 1.168 ## TODO: Warn unless value = min * x where x is an integer.
6515    
6516 wakaba 1.150 $element_state->{uri_info}->{action}->{type}->{action} = 1;
6517 wakaba 1.198 $element_state->{uri_info}->{action}->{type}->{formaction} = 1;
6518 wakaba 1.150 $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
6519     $element_state->{uri_info}->{src}->{type}->{embedded} = 1;
6520     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6521     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6522 wakaba 1.192 }, # check_attrs
6523 wakaba 1.66 check_start => sub {
6524     my ($self, $item, $element_state) = @_;
6525 wakaba 1.192 $FAECheckStart->($self, $item, $element_state);
6526     }, # check_start
6527     check_attrs2 => sub {
6528     my ($self, $item, $element_state) = @_;
6529     $FAECheckAttrs2->($self, $item, $element_state);
6530     }, # check_attrs2
6531     }; # input
6532 wakaba 1.52
6533 wakaba 1.178 ## XXXresource: Dimension attributes have requirements on width and
6534     ## height of referenced resource.
6535 wakaba 1.80
6536 wakaba 1.52 $Element->{$HTML_NS}->{button} = {
6537 wakaba 1.202 %HTMLPhrasingContentChecker,
6538 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6539 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6540 wakaba 1.199 ## XXXISSUE: In HTML5, no "MUST NOT" for using |action|, |method|,
6541 wakaba 1.165 ## |enctype|, |target|, and |novalidate| with non-|submit|-|type|
6542     ## |button| elements.
6543 wakaba 1.56 action => $HTMLURIAttrChecker,
6544 wakaba 1.165 autofocus => $AutofocusAttrChecker,
6545 wakaba 1.52 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6546 wakaba 1.166 enctype => $GetHTMLEnumeratedAttrChecker->({
6547     'application/x-www-form-urlencoded' => 1,
6548     'multipart/form-data' => 1,
6549     'text/plain' => 1,
6550     }),
6551 wakaba 1.136 form => $HTMLFormAttrChecker,
6552 wakaba 1.199 formaction => $HTMLURIAttrChecker,
6553     formenctype => $GetHTMLEnumeratedAttrChecker->({
6554     'application/x-www-form-urlencoded' => 1,
6555     'multipart/form-data' => 1,
6556     'text/plain' => 1,
6557     }),
6558     formmethod => $GetHTMLEnumeratedAttrChecker->({
6559     get => 1, post => 1, put => 1, delete => 1,
6560     }),
6561     formnovalidate => $GetHTMLBooleanAttrChecker->('formnovalidate'),
6562     formtarget => $HTMLTargetAttrChecker,
6563 wakaba 1.56 method => $GetHTMLEnumeratedAttrChecker->({
6564     get => 1, post => 1, put => 1, delete => 1,
6565     }),
6566 wakaba 1.165 name => $FormControlNameAttrChecker,
6567 wakaba 1.166 novalidate => $GetHTMLBooleanAttrChecker->('novalidate'),
6568 wakaba 1.162 onformchange => $HTMLEventHandlerAttrChecker, ## TODO: tests
6569     onforminput => $HTMLEventHandlerAttrChecker, ## TODO: tests
6570 wakaba 1.56 replace => $GetHTMLEnumeratedAttrChecker->({document => 1, values => 1}),
6571     target => $HTMLTargetAttrChecker,
6572 wakaba 1.80 ## NOTE: According to Web Forms 2.0, |button| attribute has |template|
6573     ## attribute to support the |add| button type (as part of repetition
6574     ## template feature). It conflicts with the |template| global attribute
6575     ## introduced as part of the data template feature.
6576     ## NOTE: |template| attribute as defined in Web Forms 2.0 has no
6577     ## author requirement.
6578 wakaba 1.52 type => $GetHTMLEnumeratedAttrChecker->({
6579     button => 1, submit => 1, reset => 1,
6580     }),
6581 wakaba 1.162 value => sub {}, ## NOTE: No restriction.
6582 wakaba 1.52 }, {
6583     %HTMLAttrStatus,
6584     %HTMLM12NCommonAttrStatus,
6585 wakaba 1.176 accesskey => FEATURE_M12N10_REC | FEATURE_HTML5_FD,
6586 wakaba 1.199 action => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
6587 wakaba 1.187 autofocus => FEATURE_HTML5_LC | FEATURE_WF2X,
6588 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
6589     dataformatas => FEATURE_HTML4_REC_RESERVED,
6590     datasrc => FEATURE_HTML4_REC_RESERVED,
6591 wakaba 1.187 disabled => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6592 wakaba 1.199 enctype => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
6593 wakaba 1.187 form => FEATURE_HTML5_LC | FEATURE_WF2X,
6594 wakaba 1.199 formaction => FEATURE_HTML5_LC,
6595     formenctype => FEATURE_HTML5_LC,
6596     formmethod => FEATURE_HTML5_LC,
6597     formnovalidate => FEATURE_HTML5_LC,
6598     formtarget => FEATURE_HTML5_LC,
6599 wakaba 1.187 lang => FEATURE_HTML5_REC,
6600 wakaba 1.199 method => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
6601 wakaba 1.187 name => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6602 wakaba 1.199 novalidate => FEATURE_HTML5_DROPPED,
6603 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6604     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6605 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
6606     onforminput => FEATURE_WF2_INFORMATIVE,
6607 wakaba 1.56 replace => FEATURE_WF2,
6608 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6609 wakaba 1.199 target => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
6610 wakaba 1.208 template => FEATURE_HTML5_DROPPED | FEATURE_WF2,
6611 wakaba 1.187 type => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6612     value => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6613 wakaba 1.199 }), # check_attrs
6614 wakaba 1.66 check_start => sub {
6615     my ($self, $item, $element_state) = @_;
6616 wakaba 1.202 $self->_add_minus_elements ($element_state, $HTMLInteractiveContent);
6617 wakaba 1.192 $FAECheckStart->($self, $item, $element_state);
6618 wakaba 1.162
6619 wakaba 1.202 ## XXXISSUE: "The value attribute must not be present unless the
6620     ## form [content] attribute is present.": Wrong? Maybe it should
6621     ## also be allowed when there is an ancestor |form| element.
6622 wakaba 1.139
6623 wakaba 1.66 $element_state->{uri_info}->{action}->{type}->{action} = 1;
6624 wakaba 1.202 $element_state->{uri_info}->{formaction}->{type}->{action} = 1;
6625 wakaba 1.66 $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
6626 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6627     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6628 wakaba 1.192 }, # check_start
6629     check_attrs2 => sub {
6630     my ($self, $item, $element_state) = @_;
6631     $FAECheckAttrs2->($self, $item, $element_state);
6632     }, # check_attrs2
6633 wakaba 1.202 check_end => sub {
6634     my ($self, $item, $element_state) = @_;
6635     $self->_remove_minus_elements ($element_state);
6636    
6637     $HTMLPhrasingContentChecker{check_end}->(@_);
6638     }, # check_end
6639 wakaba 1.199 }; # button
6640 wakaba 1.52
6641     $Element->{$HTML_NS}->{label} = {
6642 wakaba 1.139 %HTMLPhrasingContentChecker,
6643 wakaba 1.187 status => FEATURE_HTML5_REC,
6644 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6645 wakaba 1.138 for => sub {
6646     my ($self, $attr) = @_;
6647    
6648     ## NOTE: MUST be an ID of a labelable element.
6649    
6650     push @{$self->{idref}}, ['labelable', $attr->value, $attr];
6651     },
6652 wakaba 1.136 form => $HTMLFormAttrChecker,
6653 wakaba 1.52 }, {
6654     %HTMLAttrStatus,
6655 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
6656 wakaba 1.176 accesskey => FEATURE_HTML5_FD | FEATURE_WF2 | FEATURE_M12N10_REC,
6657 wakaba 1.187 for => FEATURE_HTML5_REC,
6658     form => FEATURE_HTML5_LC,
6659     lang => FEATURE_HTML5_REC,
6660 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6661     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6662     }),
6663 wakaba 1.139 check_start => sub {
6664     my ($self, $item, $element_state) = @_;
6665     $self->_add_minus_elements ($element_state, {$HTML_NS => {label => 1}});
6666    
6667 wakaba 1.192 ## If $self->{flag}->{has_label} is true, then there is at least
6668     ## an ancestor |label| element.
6669    
6670     ## If $self->{flag}->{has_labelable} is equal to 1, then there is
6671     ## an ancestor |label| element with its |for| attribute specified.
6672     ## If the value is equal to 2, then there is an ancestor |label|
6673     ## element with its |for| attribute unspecified but there is an
6674     ## associated form control element.
6675    
6676 wakaba 1.139 $element_state->{has_label_original} = $self->{flag}->{has_label};
6677 wakaba 1.192 $element_state->{has_labelable_original} = $self->{flag}->{has_labelable};
6678     $element_state->{label_for_original} = $self->{flag}->{label_for};
6679    
6680 wakaba 1.139 $self->{flag}->{has_label} = 1;
6681 wakaba 1.155 $self->{flag}->{has_labelable}
6682     = $item->{node}->has_attribute_ns (undef, 'for') ? 1 : 0;
6683 wakaba 1.192 $self->{flag}->{label_for}
6684     = $item->{node}->get_attribute_ns (undef, 'for');
6685 wakaba 1.139
6686     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6687     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6688     },
6689     check_end => sub {
6690     my ($self, $item, $element_state) = @_;
6691     $self->_remove_minus_elements ($element_state);
6692    
6693     if ($self->{flag}->{has_labelable} == 1) { # has for="" but no labelable
6694     $self->{flag}->{has_labelable}
6695     = $element_state->{has_labelable_original};
6696     }
6697     delete $self->{flag}->{has_label}
6698     unless $element_state->{has_label_original};
6699 wakaba 1.192 $self->{flag}->{label_for} = $element_state->{label_for_original};
6700    
6701 wakaba 1.139 ## TODO: Warn if no labelable descendant? <input type=hidden>?
6702    
6703     ## NOTE: |<label for=a><input id=a></label>| is non-conforming.
6704    
6705     $HTMLPhrasingContentChecker{check_end}->(@_);
6706     },
6707 wakaba 1.192 }; # label
6708 wakaba 1.52
6709     $Element->{$HTML_NS}->{select} = {
6710 wakaba 1.121 %HTMLChecker,
6711 wakaba 1.163 ## ISSUE: HTML5 has no requirement like these:
6712 wakaba 1.52 ## TODO: author should SELECTED at least one OPTION in non-MULTIPLE case [HTML4].
6713     ## TODO: more than one OPTION with SELECTED in non-MULTIPLE case is "error" [HTML4]
6714 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6715 wakaba 1.56 is_root => 1, ## TODO: SHOULD NOT in application/xhtml+xml [WF2]
6716 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6717 wakaba 1.165 autofocus => $AutofocusAttrChecker,
6718 wakaba 1.52 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6719 wakaba 1.56 data => $HTMLURIAttrChecker, ## TODO: MUST point ... [WF2]
6720 wakaba 1.136 form => $HTMLFormAttrChecker,
6721 wakaba 1.52 multiple => $GetHTMLBooleanAttrChecker->('multiple'),
6722 wakaba 1.165 name => $FormControlNameAttrChecker,
6723 wakaba 1.163 ## TODO: tests for on*
6724 wakaba 1.126 onformchange => $HTMLEventHandlerAttrChecker,
6725     onforminput => $HTMLEventHandlerAttrChecker,
6726     oninput => $HTMLEventHandlerAttrChecker,
6727 wakaba 1.56 oninvalid => $HTMLEventHandlerAttrChecker,
6728 wakaba 1.163 size => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
6729 wakaba 1.52 }, {
6730     %HTMLAttrStatus,
6731     %HTMLM12NCommonAttrStatus,
6732 wakaba 1.176 accesskey => FEATURE_HTML5_FD | FEATURE_WF2,
6733 wakaba 1.187 autofocus => FEATURE_HTML5_LC | FEATURE_WF2X,
6734 wakaba 1.56 data => FEATURE_WF2,
6735 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
6736     dataformatas => FEATURE_HTML4_REC_RESERVED,
6737     datasrc => FEATURE_HTML4_REC_RESERVED,
6738 wakaba 1.187 disabled => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6739     form => FEATURE_HTML5_LC | FEATURE_WF2X,
6740     lang => FEATURE_HTML5_REC,
6741     multiple => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6742     name => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6743 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6744     onchange => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6745 wakaba 1.126 onformchange => FEATURE_WF2_INFORMATIVE,
6746     onforminput => FEATURE_WF2_INFORMATIVE,
6747 wakaba 1.52 onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6748 wakaba 1.126 oninput => FEATURE_WF2,
6749 wakaba 1.56 oninvalid => FEATURE_WF2,
6750 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
6751     sdapref => FEATURE_HTML20_RFC,
6752 wakaba 1.187 size => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6753 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
6754     }),
6755 wakaba 1.66 check_start => sub {
6756     my ($self, $item, $element_state) = @_;
6757 wakaba 1.192 $FAECheckStart->($self, $item, $element_state);
6758 wakaba 1.66
6759     $element_state->{uri_info}->{data}->{type}->{resource} = 1;
6760     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
6761 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6762     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6763 wakaba 1.192 }, # check_start
6764     check_attrs2 => sub {
6765     my ($self, $item, $element_state) = @_;
6766     $FAECheckAttrs2->($self, $item, $element_state);
6767     }, # check_attrs2
6768 wakaba 1.121 check_child_element => sub {
6769 wakaba 1.163 ## NOTE: (option | optgroup)*
6770    
6771 wakaba 1.121 my ($self, $item, $child_el, $child_nsuri, $child_ln,
6772     $child_is_transparent, $element_state) = @_;
6773 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6774     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6775 wakaba 1.121 $self->{onerror}->(node => $child_el,
6776     type => 'element not allowed:minus',
6777     level => $self->{level}->{must});
6778     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6779     #
6780     } elsif ($child_nsuri eq $HTML_NS and
6781     {
6782     option => 1, optgroup => 1,
6783     }->{$child_ln}) {
6784     #
6785     } else {
6786     $self->{onerror}->(node => $child_el, type => 'element not allowed',
6787     level => $self->{level}->{must});
6788     }
6789     },
6790     check_child_text => sub {
6791     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6792     if ($has_significant) {
6793     $self->{onerror}->(node => $child_node, type => 'character not allowed',
6794     level => $self->{level}->{must});
6795     }
6796     },
6797 wakaba 1.52 };
6798 wakaba 1.1
6799 wakaba 1.52 $Element->{$HTML_NS}->{datalist} = {
6800 wakaba 1.121 %HTMLPhrasingContentChecker,
6801 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X,
6802 wakaba 1.56 check_attrs => $GetHTMLAttrsChecker->({
6803     data => $HTMLURIAttrChecker, ## TODO: MUST point ... [WF2]
6804     }, {
6805 wakaba 1.52 %HTMLAttrStatus,
6806 wakaba 1.56 data => FEATURE_WF2,
6807 wakaba 1.52 }),
6808 wakaba 1.66 check_start => sub {
6809     my ($self, $item, $element_state) = @_;
6810    
6811 wakaba 1.121 $element_state->{phase} = 'any'; # any | phrasing | option
6812    
6813 wakaba 1.66 $element_state->{uri_info}->{data}->{type}->{resource} = 1;
6814 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
6815     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
6816 wakaba 1.158
6817     $element_state->{id_type} = 'datalist';
6818 wakaba 1.66 },
6819 wakaba 1.121 ## NOTE: phrasing | option*
6820     check_child_element => sub {
6821     my ($self, $item, $child_el, $child_nsuri, $child_ln,
6822     $child_is_transparent, $element_state) = @_;
6823 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6824     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6825 wakaba 1.121 $self->{onerror}->(node => $child_el,
6826     type => 'element not allowed:minus',
6827     level => $self->{level}->{must});
6828     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6829     #
6830     } elsif ($element_state->{phase} eq 'phrasing') {
6831     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
6832     #
6833     } else {
6834     $self->{onerror}->(node => $child_el,
6835     type => 'element not allowed:phrasing',
6836     level => $self->{level}->{must});
6837     }
6838     } elsif ($element_state->{phase} eq 'option') {
6839     if ($child_nsuri eq $HTML_NS and $child_ln eq 'option') {
6840     #
6841     } else {
6842     $self->{onerror}->(node => $child_el,
6843     type => 'element not allowed',
6844     level => $self->{level}->{must});
6845     }
6846     } elsif ($element_state->{phase} eq 'any') {
6847     if ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
6848     $element_state->{phase} = 'phrasing';
6849     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'option') {
6850     $element_state->{phase} = 'option';
6851     } else {
6852     $self->{onerror}->(node => $child_el,
6853     type => 'element not allowed',
6854     level => $self->{level}->{must});
6855     }
6856     } else {
6857     die "check_child_element: Bad |datalist| phase: $element_state->{phase}";
6858     }
6859     },
6860     check_child_text => sub {
6861     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6862     if ($has_significant) {
6863     if ($element_state->{phase} eq 'phrasing') {
6864     #
6865     } elsif ($element_state->{phase} eq 'any') {
6866     $element_state->{phase} = 'phrasing';
6867     } else {
6868     $self->{onerror}->(node => $child_node,
6869     type => 'character not allowed',
6870     level => $self->{level}->{must});
6871     }
6872     }
6873     },
6874     check_end => sub {
6875     my ($self, $item, $element_state) = @_;
6876     if ($element_state->{phase} eq 'phrasing') {
6877     if ($element_state->{has_significant}) {
6878     $item->{real_parent_state}->{has_significant} = 1;
6879     } elsif ($item->{transparent}) {
6880     #
6881     } else {
6882     $self->{onerror}->(node => $item->{node},
6883     type => 'no significant content',
6884     level => $self->{level}->{should});
6885     }
6886     } else {
6887     ## NOTE: Since the content model explicitly allows a |datalist| element
6888     ## being empty, we don't raise "no significant content" error for this
6889     ## element when there is no element. (We should raise an error for
6890     ## |<datalist><br></datalist>|, however.)
6891     ## NOTE: As a side-effect, when the |datalist| element only contains
6892     ## non-conforming content, then the |phase| flag has not changed from
6893     ## |any|, no "no significant content" error is raised neither.
6894     $HTMLChecker{check_end}->(@_);
6895     }
6896     },
6897 wakaba 1.52 };
6898 wakaba 1.49
6899 wakaba 1.52 $Element->{$HTML_NS}->{optgroup} = {
6900 wakaba 1.121 %HTMLChecker,
6901 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6902 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6903     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6904 wakaba 1.164 label => sub {},
6905 wakaba 1.52 }, {
6906     %HTMLAttrStatus,
6907     %HTMLM12NCommonAttrStatus,
6908 wakaba 1.187 disabled => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6909     label => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6910     lang => FEATURE_HTML5_REC,
6911 wakaba 1.52 }),
6912 wakaba 1.164 check_attrs2 => sub {
6913     my ($self, $item, $element_state) = @_;
6914    
6915     unless ($item->{node}->has_attribute_ns (undef, 'label')) {
6916     $self->{onerror}->(node => $item->{node},
6917     type => 'attribute missing',
6918     text => 'label',
6919     level => $self->{level}->{must});
6920     }
6921     },
6922 wakaba 1.121 check_child_element => sub {
6923     my ($self, $item, $child_el, $child_nsuri, $child_ln,
6924     $child_is_transparent, $element_state) = @_;
6925 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
6926     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
6927 wakaba 1.121 $self->{onerror}->(node => $child_el,
6928     type => 'element not allowed:minus',
6929     level => $self->{level}->{must});
6930     } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
6931     #
6932     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'option') {
6933     #
6934     } else {
6935     $self->{onerror}->(node => $child_el, type => 'element not allowed',
6936     level => $self->{level}->{must});
6937     }
6938     },
6939     check_child_text => sub {
6940     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
6941     if ($has_significant) {
6942     $self->{onerror}->(node => $child_node, type => 'character not allowed',
6943     level => $self->{level}->{must});
6944     }
6945     },
6946 wakaba 1.52 };
6947    
6948     $Element->{$HTML_NS}->{option} = {
6949     %HTMLTextChecker,
6950 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6951 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6952     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6953 wakaba 1.164 label => sub {}, ## NOTE: No restriction.
6954     selected => $GetHTMLBooleanAttrChecker->('selected'), ## ISSUE: Not a "boolean attribute"
6955     value => sub {}, ## NOTE: No restriction.
6956 wakaba 1.52 }, {
6957     %HTMLAttrStatus,
6958     %HTMLM12NCommonAttrStatus,
6959 wakaba 1.187 disabled => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6960     label => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6961     lang => FEATURE_HTML5_REC,
6962 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
6963     sdapref => FEATURE_HTML20_RFC,
6964 wakaba 1.187 selected => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6965     value => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
6966 wakaba 1.52 }),
6967     };
6968 wakaba 1.49
6969 wakaba 1.52 $Element->{$HTML_NS}->{textarea} = {
6970     %HTMLTextChecker,
6971 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
6972 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
6973 wakaba 1.164 accept => $HTMLIMTAttrChecker, ## TODO: MUST be a text-based type [WF2]
6974 wakaba 1.165 autofocus => $AutofocusAttrChecker,
6975 wakaba 1.164 cols => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
6976 wakaba 1.52 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
6977 wakaba 1.136 form => $HTMLFormAttrChecker,
6978 wakaba 1.56 ## TODO: inputmode [WF2]
6979 wakaba 1.164 maxlength => sub {
6980     my ($self, $attr, $item, $element_state) = @_;
6981    
6982     $GetHTMLNonNegativeIntegerAttrChecker->(sub { 1 })->(@_);
6983    
6984 wakaba 1.165 if ($attr->value =~ /^[\x09\x0A\x0C\x0D\x20]*([0-9]+)/) {
6985 wakaba 1.164 ## NOTE: Applying the rules for parsing non-negative integers
6986     ## results in a number.
6987     my $max_allowed_value_length = 0+$1;
6988    
6989     ## ISSUE: "The the purposes of this requirement," (typo)
6990    
6991     ## ISSUE: This constraint is applied w/o CRLF normalization to
6992     ## |value| attribute, but w/ CRLF normalization to
6993     ## concept-value.
6994     my $value = $item->{node}->text_content;
6995     if (defined $value) {
6996     my $codepoint_length = length $value;
6997    
6998     if ($codepoint_length > $max_allowed_value_length) {
6999     $self->{onerror}->(node => $item->{node},
7000     type => 'value too long',
7001     level => $self->{level}->{must});
7002     }
7003     }
7004     }
7005     },
7006 wakaba 1.165 name => $FormControlNameAttrChecker,
7007 wakaba 1.164 onformchange => $HTMLEventHandlerAttrChecker, ## TODO: tests
7008     onforminput => $HTMLEventHandlerAttrChecker, ## TODO: tests
7009     oninput => $HTMLEventHandlerAttrChecker, ## TODO: tests
7010 wakaba 1.161 pattern => $PatternAttrChecker,
7011 wakaba 1.179 placeholder => $PlaceholderAttrChecker,
7012 wakaba 1.52 readonly => $GetHTMLBooleanAttrChecker->('readonly'),
7013 wakaba 1.56 required => $GetHTMLBooleanAttrChecker->('required'),
7014 wakaba 1.164 rows => $GetHTMLNonNegativeIntegerAttrChecker->(sub { shift > 0 }),
7015     oninput => $HTMLEventHandlerAttrChecker, ## TODO: tests
7016     oninvalid => $HTMLEventHandlerAttrChecker, ## TODO: tests
7017 wakaba 1.161 ## NOTE: |title| had special semantics if |pattern| was specified [WF2].
7018 wakaba 1.56 wrap => $GetHTMLEnumeratedAttrChecker->({soft => 1, hard => 1}),
7019 wakaba 1.52 }, {
7020     %HTMLAttrStatus,
7021     %HTMLM12NCommonAttrStatus,
7022 wakaba 1.164 accept => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
7023 wakaba 1.61 'accept-charset' => FEATURE_HTML2X_RFC,
7024 wakaba 1.176 accesskey => FEATURE_HTML5_FD | FEATURE_M12N10_REC,
7025 wakaba 1.187 autofocus => FEATURE_HTML5_LC | FEATURE_WF2X,
7026     cols => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
7027 wakaba 1.52 datafld => FEATURE_HTML4_REC_RESERVED,
7028 wakaba 1.49 dataformatas => FEATURE_HTML4_REC_RESERVED,
7029     datasrc => FEATURE_HTML4_REC_RESERVED,
7030 wakaba 1.187 disabled => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
7031     form => FEATURE_HTML5_LC | FEATURE_WF2X,
7032 wakaba 1.164 inputmode => FEATURE_HTML5_DROPPED | FEATURE_WF2X | FEATURE_XHTMLBASIC11_CR,
7033 wakaba 1.187 lang => FEATURE_HTML5_REC,
7034 wakaba 1.121 maxlength => FEATURE_HTML5_DEFAULT | FEATURE_WF2X,
7035 wakaba 1.187 name => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
7036 wakaba 1.52 onblur => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
7037     onchange => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
7038     onfocus => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
7039 wakaba 1.164 onformchange => FEATURE_WF2_INFORMATIVE, ## TODO: tests
7040     onforminput => FEATURE_WF2_INFORMATIVE, ## TODO: tests
7041     oninput => FEATURE_WF2, ## TODO: tests
7042     oninvalid => FEATURE_WF2, ## TODO: tests
7043 wakaba 1.52 onselect => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
7044 wakaba 1.161 pattern => FEATURE_HTML5_DROPPED | FEATURE_WF2X,
7045 wakaba 1.179 placeholder => FEATURE_HTML5_LC,
7046 wakaba 1.187 readonly => FEATURE_HTML5_LC | FEATURE_WF2X | FEATURE_M12N10_REC,
7047     required => FEATURE_HTML5_LC | FEATURE_WF2X,
7048     rows => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
7049 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
7050     sdapref => FEATURE_HTML20_RFC,
7051 wakaba 1.52 tabindex => FEATURE_HTML5_DEFAULT | FEATURE_M12N10_REC,
7052 wakaba 1.187 wrap => FEATURE_HTML5_LC | FEATURE_WF2X,
7053 wakaba 1.52 }),
7054 wakaba 1.66 check_start => sub {
7055     my ($self, $item, $element_state) = @_;
7056 wakaba 1.192 $FAECheckStart->($self, $item, $element_state);
7057 wakaba 1.164
7058     $element_state->{uri_info}->{data}->{type}->{resource} = 1;
7059     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7060     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7061     },
7062     check_attrs2 => sub {
7063     my ($self, $item, $element_state) = @_;
7064 wakaba 1.66
7065 wakaba 1.161 if ($item->{node}->has_attribute_ns (undef, 'pattern') and
7066     not $item->{node}->has_attribute_ns (undef, 'title')) {
7067     ## NOTE: WF2 (dropped by HTML5)
7068     $self->{onerror}->(node => $item->{node},
7069     type => 'attribute missing',
7070     text => 'title',
7071     level => $self->{level}->{should});
7072     }
7073    
7074 wakaba 1.164 unless ($item->{node}->has_attribute_ns (undef, 'cols')) {
7075     my $wrap = $item->{node}->get_attribute_ns (undef, 'wrap');
7076     if (defined $wrap) {
7077     $wrap =~ tr/A-Z/a-z/; ## ASCII case-insensitive
7078     if ($wrap eq 'hard') {
7079     $self->{onerror}->(node => $item->{node},
7080     type => 'attribute missing',
7081     text => 'cols',
7082     level => $self->{level}->{must});
7083     }
7084     }
7085     }
7086 wakaba 1.192
7087     $FAECheckAttrs2->($self, $item, $element_state);
7088     }, # check_attrs2
7089 wakaba 1.194 }; # textarea
7090    
7091     $Element->{$HTML_NS}->{keygen} = {
7092     %HTMLEmptyChecker,
7093     status => FEATURE_HTML5_FD,
7094     check_attrs => $GetHTMLAttrsChecker->({
7095     autofocus => $AutofocusAttrChecker,
7096     challenge => sub { }, ## No constraints.
7097     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
7098     form => $HTMLFormAttrChecker,
7099     keytype => $GetHTMLEnumeratedAttrChecker->({rsa => 1}),
7100     name => $FormControlNameAttrChecker,
7101     }, {
7102     %HTMLAttrStatus,
7103     autofocus => FEATURE_HTML5_LC,
7104     challenge => FEATURE_HTML5_FD,
7105     disabled => FEATURE_HTML5_LC,
7106     form => FEATURE_HTML5_LC,
7107     keytype => FEATURE_HTML5_FD,
7108     name => FEATURE_HTML5_LC,
7109     }), # check_attrs
7110     check_start => sub {
7111     my ($self, $item, $element_state) = @_;
7112     $FAECheckStart->($self, $item, $element_state);
7113    
7114     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7115     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7116     }, # check_start
7117     check_attrs2 => sub {
7118     my ($self, $item, $element_state) = @_;
7119     $FAECheckAttrs2->($self, $item, $element_state);
7120     }, # check_attrs2
7121     }; # keygen
7122 wakaba 1.49
7123 wakaba 1.52 $Element->{$HTML_NS}->{output} = {
7124 wakaba 1.121 %HTMLPhrasingContentChecker,
7125 wakaba 1.187 status => FEATURE_HTML5_LC | FEATURE_WF2X,
7126 wakaba 1.56 check_attrs => $GetHTMLAttrsChecker->({
7127 wakaba 1.165 for => sub {
7128     my ($self, $attr) = @_;
7129    
7130     ## NOTE: "Unordered set of unique space-separated tokens".
7131    
7132     my %word;
7133     for my $word (grep {length $_}
7134     split /[\x09\x0A\x0C\x0D\x20]+/, $attr->value) {
7135     unless ($word{$word}) {
7136     $word{$word} = 1;
7137     push @{$self->{idref}}, ['any', $word, $attr];
7138     } else {
7139     $self->{onerror}->(node => $attr, type => 'duplicate token',
7140     value => $word,
7141     level => $self->{level}->{must});
7142     }
7143     }
7144     },
7145 wakaba 1.136 form => $HTMLFormAttrChecker,
7146 wakaba 1.165 name => $FormControlNameAttrChecker,
7147     onformchange => $HTMLEventHandlerAttrChecker, ## TODO: tests
7148     onforminput => $HTMLEventHandlerAttrChecker, ## TODO: tests
7149 wakaba 1.56 }, {
7150 wakaba 1.52 %HTMLAttrStatus,
7151 wakaba 1.187 for => FEATURE_HTML5_LC | FEATURE_WF2X,
7152     form => FEATURE_HTML5_LC | FEATURE_WF2X,
7153     name => FEATURE_HTML5_LC | FEATURE_WF2X,
7154 wakaba 1.56 onchange => FEATURE_HTML5_DEFAULT | FEATURE_WF2,
7155     onformchange => FEATURE_WF2,
7156     onforminput => FEATURE_WF2,
7157 wakaba 1.52 }),
7158     };
7159    
7160     $Element->{$HTML_NS}->{isindex} = {
7161     %HTMLEmptyChecker,
7162 wakaba 1.54 status => FEATURE_M12N10_REC_DEPRECATED |
7163     Whatpm::ContentChecker::FEATURE_DEPRECATED_SHOULD, ## [HTML4]
7164 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
7165     prompt => sub {}, ## NOTE: Text [M12N]
7166     }, {
7167     %HTMLAttrStatus,
7168 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
7169     dir => FEATURE_HTML5_REC,
7170     id => FEATURE_HTML5_REC,
7171     lang => FEATURE_HTML5_REC,
7172 wakaba 1.52 prompt => FEATURE_M12N10_REC_DEPRECATED,
7173 wakaba 1.61 sdapref => FEATURE_HTML20_RFC,
7174 wakaba 1.187 style => FEATURE_HTML5_REC,
7175     title => FEATURE_HTML5_REC,
7176 wakaba 1.52 }),
7177     ## TODO: Tests
7178     ## TODO: Tests for <nest/> in <isindex>
7179 wakaba 1.66 check_start => sub {
7180     my ($self, $item, $element_state) = @_;
7181    
7182     $element_state->{uri_info}->{action}->{type}->{action} = 1;
7183 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7184     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7185 wakaba 1.66 },
7186 wakaba 1.52 };
7187 wakaba 1.49
7188 wakaba 1.1 $Element->{$HTML_NS}->{script} = {
7189 wakaba 1.40 %HTMLChecker,
7190 wakaba 1.153 status => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
7191 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7192 wakaba 1.91 charset => sub {
7193     my ($self, $attr) = @_;
7194    
7195     unless ($attr->owner_element->has_attribute_ns (undef, 'src')) {
7196     $self->{onerror}->(type => 'attribute not allowed',
7197     node => $attr,
7198 wakaba 1.104 level => $self->{level}->{must});
7199 wakaba 1.91 }
7200    
7201 wakaba 1.188 ## XXXresource: MUST match the charset of the referenced
7202     ## resource (HTML5 revision 2967).
7203    
7204 wakaba 1.91 $HTMLCharsetChecker->($attr->value, @_);
7205     },
7206 wakaba 1.86 language => sub {}, ## NOTE: No syntax constraint according to HTML4.
7207 wakaba 1.91 src => $HTMLURIAttrChecker, ## TODO: pointed resource MUST be in type of type="" (resource error)
7208 wakaba 1.1 defer => $GetHTMLBooleanAttrChecker->('defer'),
7209     async => $GetHTMLBooleanAttrChecker->('async'),
7210 wakaba 1.91 type => $HTMLIMTAttrChecker, ## TODO: MUST NOT: |charset=""| parameter
7211 wakaba 1.49 }, {
7212     %HTMLAttrStatus,
7213 wakaba 1.153 async => FEATURE_HTML5_WD,
7214     charset => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
7215     defer => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
7216 wakaba 1.49 event => FEATURE_HTML4_REC_RESERVED,
7217     for => FEATURE_HTML4_REC_RESERVED,
7218 wakaba 1.154 href => FEATURE_RDFA_REC,
7219 wakaba 1.187 id => FEATURE_HTML5_REC,
7220 wakaba 1.49 language => FEATURE_M12N10_REC_DEPRECATED,
7221 wakaba 1.153 src => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
7222     type => FEATURE_HTML5_WD | FEATURE_M12N10_REC,
7223 wakaba 1.9 }),
7224 wakaba 1.191 check_attrs2 => sub {
7225     my ($self, $item, $element_state) = @_;
7226    
7227     my $el = $item->{node};
7228     if ($el->has_attribute_ns (undef, 'defer') and
7229     not $el->has_attribute_ns (undef, 'src')) {
7230     $self->{onerror}->(node => $el,
7231     type => 'attribute missing',
7232     text => 'src',
7233     level => $self->{level}->{must});
7234     }
7235     },
7236 wakaba 1.40 check_start => sub {
7237     my ($self, $item, $element_state) = @_;
7238 wakaba 1.1
7239 wakaba 1.40 if ($item->{node}->has_attribute_ns (undef, 'src')) {
7240     $element_state->{must_be_empty} = 1;
7241 wakaba 1.1 } else {
7242     ## NOTE: No content model conformance in HTML5 spec.
7243 wakaba 1.40 my $type = $item->{node}->get_attribute_ns (undef, 'type');
7244     my $language = $item->{node}->get_attribute_ns (undef, 'language');
7245 wakaba 1.1 if ((defined $type and $type eq '') or
7246     (defined $language and $language eq '')) {
7247     $type = 'text/javascript';
7248     } elsif (defined $type) {
7249     #
7250     } elsif (defined $language) {
7251     $type = 'text/' . $language;
7252     } else {
7253     $type = 'text/javascript';
7254     }
7255 wakaba 1.93
7256     if ($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)]) {
7257     $type = "$1/$2";
7258     $type =~ tr/A-Z/a-z/; ## NOTE: ASCII case-insensitive
7259     ## TODO: Though we strip prameter here, it should not be ignored for the purpose of conformance checking...
7260     }
7261     $element_state->{script_type} = $type;
7262 wakaba 1.40 }
7263 wakaba 1.66
7264     $element_state->{uri_info}->{src}->{type}->{resource} = 1;
7265 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7266     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7267 wakaba 1.107
7268     $element_state->{text} = '';
7269 wakaba 1.40 },
7270     check_child_element => sub {
7271     my ($self, $item, $child_el, $child_nsuri, $child_ln,
7272     $child_is_transparent, $element_state) = @_;
7273 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
7274     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
7275 wakaba 1.40 $self->{onerror}->(node => $child_el,
7276     type => 'element not allowed:minus',
7277 wakaba 1.104 level => $self->{level}->{must});
7278 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
7279     #
7280     } else {
7281     if ($element_state->{must_be_empty}) {
7282     $self->{onerror}->(node => $child_el,
7283 wakaba 1.104 type => 'element not allowed:empty',
7284     level => $self->{level}->{must});
7285 wakaba 1.40 }
7286     }
7287     },
7288     check_child_text => sub {
7289     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
7290     if ($has_significant and
7291     $element_state->{must_be_empty}) {
7292     $self->{onerror}->(node => $child_node,
7293 wakaba 1.104 type => 'character not allowed:empty',
7294     level => $self->{level}->{must});
7295 wakaba 1.40 }
7296 wakaba 1.115 $element_state->{text} .= $child_node->data;
7297 wakaba 1.40 },
7298     check_end => sub {
7299     my ($self, $item, $element_state) = @_;
7300     unless ($element_state->{must_be_empty}) {
7301 wakaba 1.93 if ($element_state->{script_type} =~ m![+/][Xx][Mm][Ll]\z!) {
7302     ## NOTE: XML content should be checked by THIS instance of checker
7303     ## as part of normal tree validation.
7304 wakaba 1.104 $self->{onerror}->(node => $item->{node},
7305     type => 'XML script lang',
7306     text => $element_state->{script_type},
7307     level => $self->{level}->{uncertain});
7308     ## ISSUE: Should we raise some kind of error for
7309     ## <script type="text/xml">aaaaa</script>?
7310     ## NOTE: ^^^ This is why we throw an "uncertain" error.
7311 wakaba 1.93 } else {
7312     $self->{onsubdoc}->({s => $element_state->{text},
7313     container_node => $item->{node},
7314     media_type => $element_state->{script_type},
7315     is_char_string => 1});
7316     }
7317 wakaba 1.40
7318     $HTMLChecker{check_end}->(@_);
7319 wakaba 1.1 }
7320     },
7321 wakaba 1.91 ## TODO: There MUST be |type| unless the script type is JavaScript. (resource error)
7322     ## NOTE: "When used to include script data, the script data must be embedded
7323     ## inline, the format of the data must be given using the type attribute,
7324     ## and the src attribute must not be specified." - not testable.
7325     ## TODO: It would be possible to err <script type=text/plain src=...>
7326 wakaba 1.1 };
7327 wakaba 1.25 ## ISSUE: Significant check and text child node
7328 wakaba 1.1
7329     ## NOTE: When script is disabled.
7330     $Element->{$HTML_NS}->{noscript} = {
7331 wakaba 1.40 %HTMLTransparentChecker,
7332 wakaba 1.187 status => FEATURE_HTML5_REC,
7333 wakaba 1.49 check_attrs => $GetHTMLAttrsChecker->({}, {
7334     %HTMLAttrStatus,
7335     %HTMLM12NCommonAttrStatus,
7336 wakaba 1.187 lang => FEATURE_HTML5_REC,
7337 wakaba 1.49 }),
7338 wakaba 1.40 check_start => sub {
7339     my ($self, $item, $element_state) = @_;
7340 wakaba 1.3
7341 wakaba 1.40 unless ($item->{node}->owner_document->manakai_is_html) {
7342 wakaba 1.104 $self->{onerror}->(node => $item->{node}, type => 'in XML:noscript',
7343     level => $self->{level}->{must});
7344 wakaba 1.3 }
7345    
7346 wakaba 1.40 unless ($self->{flag}->{in_head}) {
7347     $self->_add_minus_elements ($element_state,
7348     {$HTML_NS => {noscript => 1}});
7349     }
7350 wakaba 1.79
7351     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7352     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7353 wakaba 1.3 },
7354 wakaba 1.40 check_child_element => sub {
7355     my ($self, $item, $child_el, $child_nsuri, $child_ln,
7356     $child_is_transparent, $element_state) = @_;
7357     if ($self->{flag}->{in_head}) {
7358 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
7359     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
7360 wakaba 1.40 $self->{onerror}->(node => $child_el,
7361     type => 'element not allowed:minus',
7362 wakaba 1.104 level => $self->{level}->{must});
7363 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
7364     #
7365     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'link') {
7366     #
7367     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'style') {
7368     if ($child_el->has_attribute_ns (undef, 'scoped')) {
7369     $self->{onerror}->(node => $child_el,
7370     type => 'element not allowed:head noscript',
7371 wakaba 1.104 level => $self->{level}->{must});
7372 wakaba 1.40 }
7373     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'meta') {
7374 wakaba 1.47 my $http_equiv_attr
7375     = $child_el->get_attribute_node_ns (undef, 'http-equiv');
7376     if ($http_equiv_attr) {
7377     ## TODO: case
7378     if (lc $http_equiv_attr->value eq 'content-type') {
7379 wakaba 1.40 $self->{onerror}->(node => $child_el,
7380 wakaba 1.34 type => 'element not allowed:head noscript',
7381 wakaba 1.104 level => $self->{level}->{must});
7382 wakaba 1.47 } else {
7383     #
7384 wakaba 1.3 }
7385 wakaba 1.47 } else {
7386     $self->{onerror}->(node => $child_el,
7387     type => 'element not allowed:head noscript',
7388 wakaba 1.104 level => $self->{level}->{must});
7389 wakaba 1.3 }
7390 wakaba 1.40 } else {
7391     $self->{onerror}->(node => $child_el,
7392     type => 'element not allowed:head noscript',
7393 wakaba 1.104 level => $self->{level}->{must});
7394 wakaba 1.40 }
7395     } else {
7396     $HTMLTransparentChecker{check_child_element}->(@_);
7397     }
7398     },
7399     check_child_text => sub {
7400     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
7401     if ($self->{flag}->{in_head}) {
7402     if ($has_significant) {
7403     $self->{onerror}->(node => $child_node,
7404 wakaba 1.104 type => 'character not allowed',
7405     level => $self->{level}->{must});
7406 wakaba 1.3 }
7407     } else {
7408 wakaba 1.40 $HTMLTransparentChecker{check_child_text}->(@_);
7409     }
7410     },
7411     check_end => sub {
7412     my ($self, $item, $element_state) = @_;
7413     $self->_remove_minus_elements ($element_state);
7414     if ($self->{flag}->{in_head}) {
7415     $HTMLChecker{check_end}->(@_);
7416     } else {
7417     $HTMLPhrasingContentChecker{check_end}->(@_);
7418 wakaba 1.3 }
7419 wakaba 1.1 },
7420     };
7421 wakaba 1.3 ## ISSUE: Scripting is disabled: <head><noscript><html a></noscript></head>
7422 wakaba 1.1
7423     $Element->{$HTML_NS}->{'event-source'} = {
7424 wakaba 1.40 %HTMLEmptyChecker,
7425 wakaba 1.118 status => FEATURE_HTML5_LC_DROPPED,
7426     check_attrs => $GetHTMLAttrsChecker->({
7427     src => $HTMLURIAttrChecker,
7428     }, {
7429     %HTMLAttrStatus,
7430     src => FEATURE_HTML5_LC_DROPPED,
7431     }),
7432     check_start => sub {
7433     my ($self, $item, $element_state) = @_;
7434    
7435     $element_state->{uri_info}->{src}->{type}->{resource} = 1;
7436     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7437     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7438     },
7439     };
7440    
7441     $Element->{$HTML_NS}->{eventsource} = {
7442     %HTMLEmptyChecker,
7443 wakaba 1.180 status => FEATURE_HTML5_DROPPED,
7444 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7445 wakaba 1.1 src => $HTMLURIAttrChecker,
7446 wakaba 1.50 }, {
7447     %HTMLAttrStatus,
7448 wakaba 1.180 src => FEATURE_HTML5_DROPPED,
7449 wakaba 1.1 }),
7450 wakaba 1.66 check_start => sub {
7451     my ($self, $item, $element_state) = @_;
7452    
7453     $element_state->{uri_info}->{src}->{type}->{resource} = 1;
7454 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7455     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7456 wakaba 1.66 },
7457 wakaba 1.1 };
7458    
7459     $Element->{$HTML_NS}->{details} = {
7460 wakaba 1.134 %{$Element->{$HTML_NS}->{fieldset}},
7461 wakaba 1.153 status => FEATURE_HTML5_LC,
7462 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7463 wakaba 1.1 open => $GetHTMLBooleanAttrChecker->('open'),
7464 wakaba 1.50 }, {
7465     %HTMLAttrStatus,
7466 wakaba 1.153 open => FEATURE_HTML5_LC,
7467 wakaba 1.1 }),
7468     };
7469    
7470     $Element->{$HTML_NS}->{datagrid} = {
7471 wakaba 1.72 %HTMLFlowContentChecker,
7472 wakaba 1.197 status => FEATURE_HTML5_DROPPED,
7473 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7474 wakaba 1.1 disabled => $GetHTMLBooleanAttrChecker->('disabled'),
7475     multiple => $GetHTMLBooleanAttrChecker->('multiple'),
7476 wakaba 1.50 }, {
7477     %HTMLAttrStatus,
7478 wakaba 1.197 disabled => FEATURE_HTML5_DROPPED,
7479     multiple => FEATURE_HTML5_DROPPED,
7480     }), # check_attrs
7481     }; # datagrid
7482 wakaba 1.1
7483     $Element->{$HTML_NS}->{command} = {
7484 wakaba 1.40 %HTMLEmptyChecker,
7485 wakaba 1.48 status => FEATURE_HTML5_WD,
7486 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7487 wakaba 1.1 checked => $GetHTMLBooleanAttrChecker->('checked'),
7488     default => $GetHTMLBooleanAttrChecker->('default'),
7489     disabled => $GetHTMLBooleanAttrChecker->('disabled'),
7490     icon => $HTMLURIAttrChecker,
7491 wakaba 1.212 label => sub { }, ## NOTE: No requirement
7492     radiogroup => sub { }, ## NOTE: No requirement for the value
7493     type => $GetHTMLEnumeratedAttrChecker->({
7494     command => 1, checkbox => 1, radio => 1,
7495     }),
7496 wakaba 1.50 }, {
7497     %HTMLAttrStatus,
7498     checked => FEATURE_HTML5_WD,
7499 wakaba 1.175 default => FEATURE_HTML5_DROPPED, # HTML5 revision 3067
7500 wakaba 1.50 disabled => FEATURE_HTML5_WD,
7501     icon => FEATURE_HTML5_WD,
7502     label => FEATURE_HTML5_WD,
7503     radiogroup => FEATURE_HTML5_WD,
7504     type => FEATURE_HTML5_WD,
7505 wakaba 1.212 }), # check_attrs
7506     check_attrs2 => sub {
7507     my ($self, $item, $element_state) = @_;
7508    
7509     my $type = $item->{node}->get_attribute_ns (undef, 'type') || '';
7510     $type =~ tr/A-Z/a-z/; ## ASCII case-insensitive.
7511     $type = 'command' unless $type eq 'radio' or $type eq 'checkbox';
7512    
7513     unless ($type eq 'radio') {
7514     my $rg_attr = $item->{node}->get_attribute_node_ns (undef, 'radiogroup');
7515     if ($rg_attr) {
7516     $self->{onerror}->(node => $rg_attr,
7517     type => 'attribute not allowed:radiogroup',
7518     level => $self->{level}->{must});
7519     }
7520     }
7521    
7522     unless ($type eq 'checkbox' or $type eq 'radio') {
7523     my $cd_attr = $item->{node}->get_attribute_node_ns (undef, 'checked');
7524     if ($cd_attr) {
7525     $self->{onerror}->(node => $cd_attr,
7526     type => 'attribute not allowed:checked',
7527     level => $self->{level}->{must});
7528     }
7529     }
7530    
7531     unless ($type eq 'command') {
7532     my $def_attr = $item->{node}->get_attribute_node_ns (undef, 'default');
7533     if ($def_attr) {
7534     ## HTML5 revision 2415
7535     $self->{onerror}->(node => $def_attr,
7536     type => 'attribute not allowed:default',
7537     level => $self->{level}->{must});
7538     }
7539     }
7540     }, # check_attrs2
7541 wakaba 1.66 check_start => sub {
7542     my ($self, $item, $element_state) = @_;
7543    
7544     $element_state->{uri_info}->{icon}->{type}->{embedded} = 1;
7545 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7546     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7547 wakaba 1.212 }, # check_start
7548     }; # command
7549 wakaba 1.115
7550     $Element->{$HTML_NS}->{bb} = {
7551     %HTMLPhrasingContentChecker,
7552 wakaba 1.190 status => FEATURE_HTML5_DROPPED,
7553 wakaba 1.115 check_attrs => $GetHTMLAttrsChecker->({
7554     type => $GetHTMLEnumeratedAttrChecker->({makeapp => 1}),
7555     }, {
7556     %HTMLAttrStatus,
7557 wakaba 1.190 type => FEATURE_HTML5_DROPPED,
7558 wakaba 1.115 }),
7559 wakaba 1.130 check_start => sub {
7560     my ($self, $item, $element_state) = @_;
7561     $self->_add_minus_elements ($element_state, $HTMLInteractiveContent);
7562    
7563     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7564     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7565     },
7566     check_end => sub {
7567     my ($self, $item, $element_state) = @_;
7568     $self->_remove_minus_elements ($element_state);
7569    
7570     $HTMLTransparentChecker{check_end}->(@_);
7571     },
7572 wakaba 1.1 };
7573    
7574     $Element->{$HTML_NS}->{menu} = {
7575 wakaba 1.40 %HTMLPhrasingContentChecker,
7576 wakaba 1.54 #status => FEATURE_M12N10_REC_DEPRECATED | FEATURE_HTML5_WD,
7577     status => FEATURE_M12N10_REC | FEATURE_HTML5_WD,
7578     ## NOTE: We don't want any |menu| element warned as deprecated.
7579 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7580 wakaba 1.1 autosubmit => $GetHTMLBooleanAttrChecker->('autosubmit'),
7581 wakaba 1.68 compact => $GetHTMLBooleanAttrChecker->('compact'),
7582 wakaba 1.135 ## ISSUE: <menu id=""><p contextmenu=""> match? (In the current
7583     ## implementation, it does not match.)
7584 wakaba 1.1 label => sub { }, ## NOTE: No conformance creteria
7585     type => $GetHTMLEnumeratedAttrChecker->({context => 1, toolbar => 1}),
7586 wakaba 1.49 }, {
7587     %HTMLAttrStatus,
7588     %HTMLM12NCommonAttrStatus,
7589 wakaba 1.61 align => FEATURE_HTML2X_RFC,
7590 wakaba 1.113 autosubmit => FEATURE_HTML5_DROPPED,
7591 wakaba 1.49 compat => FEATURE_M12N10_REC_DEPRECATED,
7592 wakaba 1.50 label => FEATURE_HTML5_WD,
7593 wakaba 1.187 lang => FEATURE_HTML5_REC,
7594 wakaba 1.61 sdaform => FEATURE_HTML20_RFC,
7595     sdapref => FEATURE_HTML20_RFC,
7596 wakaba 1.50 type => FEATURE_HTML5_WD,
7597 wakaba 1.207 }), # check_attrs
7598 wakaba 1.40 check_start => sub {
7599     my ($self, $item, $element_state) = @_;
7600     $element_state->{phase} = 'li or phrasing';
7601 wakaba 1.79
7602     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7603     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7604 wakaba 1.135 $element_state->{id_type} = 'menu';
7605 wakaba 1.207 }, # check_start
7606 wakaba 1.40 check_child_element => sub {
7607     my ($self, $item, $child_el, $child_nsuri, $child_ln,
7608     $child_is_transparent, $element_state) = @_;
7609 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
7610     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
7611 wakaba 1.40 $self->{onerror}->(node => $child_el,
7612     type => 'element not allowed:minus',
7613 wakaba 1.104 level => $self->{level}->{must});
7614 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
7615     #
7616     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'li') {
7617     if ($element_state->{phase} eq 'li') {
7618     #
7619     } elsif ($element_state->{phase} eq 'li or phrasing') {
7620     $element_state->{phase} = 'li';
7621     } else {
7622 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
7623     level => $self->{level}->{must});
7624 wakaba 1.40 }
7625     } elsif ($HTMLPhrasingContent->{$child_nsuri}->{$child_ln}) {
7626     if ($element_state->{phase} eq 'phrasing') {
7627     #
7628     } elsif ($element_state->{phase} eq 'li or phrasing') {
7629     $element_state->{phase} = 'phrasing';
7630     } else {
7631 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
7632     level => $self->{level}->{must});
7633 wakaba 1.40 }
7634     } else {
7635 wakaba 1.104 $self->{onerror}->(node => $child_el, type => 'element not allowed',
7636     level => $self->{level}->{must});
7637 wakaba 1.40 }
7638 wakaba 1.207 }, # check_child_element
7639 wakaba 1.40 check_child_text => sub {
7640     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
7641     if ($has_significant) {
7642     if ($element_state->{phase} eq 'phrasing') {
7643     #
7644     } elsif ($element_state->{phase} eq 'li or phrasing') {
7645     $element_state->{phase} = 'phrasing';
7646     } else {
7647     $self->{onerror}->(node => $child_node,
7648 wakaba 1.104 type => 'character not allowed',
7649     level => $self->{level}->{must});
7650 wakaba 1.1 }
7651     }
7652 wakaba 1.207 }, # check_child_text
7653 wakaba 1.40 check_end => sub {
7654     my ($self, $item, $element_state) = @_;
7655     if ($element_state->{phase} eq 'li') {
7656     $HTMLChecker{check_end}->(@_);
7657     } else { # 'phrasing' or 'li or phrasing'
7658     $HTMLPhrasingContentChecker{check_end}->(@_);
7659 wakaba 1.1 }
7660 wakaba 1.207 }, # check_end
7661     }; # menu
7662 wakaba 1.8
7663     $Element->{$HTML_NS}->{datatemplate} = {
7664 wakaba 1.40 %HTMLChecker,
7665 wakaba 1.208 status => FEATURE_HTML5_DROPPED,
7666 wakaba 1.40 check_child_element => sub {
7667     my ($self, $item, $child_el, $child_nsuri, $child_ln,
7668     $child_is_transparent, $element_state) = @_;
7669 wakaba 1.130 if ($self->{minus_elements}->{$child_nsuri}->{$child_ln} and
7670     $IsInHTMLInteractiveContent->($child_el, $child_nsuri, $child_ln)) {
7671 wakaba 1.40 $self->{onerror}->(node => $child_el,
7672     type => 'element not allowed:minus',
7673 wakaba 1.104 level => $self->{level}->{must});
7674 wakaba 1.40 } elsif ($self->{plus_elements}->{$child_nsuri}->{$child_ln}) {
7675     #
7676     } elsif ($child_nsuri eq $HTML_NS and $child_ln eq 'rule') {
7677     #
7678     } else {
7679     $self->{onerror}->(node => $child_el,
7680 wakaba 1.104 type => 'element not allowed:datatemplate',
7681     level => $self->{level}->{must});
7682 wakaba 1.40 }
7683     },
7684     check_child_text => sub {
7685     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
7686     if ($has_significant) {
7687 wakaba 1.104 $self->{onerror}->(node => $child_node, type => 'character not allowed',
7688     level => $self->{level}->{must});
7689 wakaba 1.8 }
7690     },
7691     is_xml_root => 1,
7692     };
7693    
7694     $Element->{$HTML_NS}->{rule} = {
7695 wakaba 1.40 %HTMLChecker,
7696 wakaba 1.208 status => FEATURE_HTML5_DROPPED,
7697 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7698 wakaba 1.23 condition => $HTMLSelectorsAttrChecker,
7699 wakaba 1.92 mode => $GetHTMLUnorderedUniqueSetOfSpaceSeparatedTokensAttrChecker->(),
7700 wakaba 1.50 }, {
7701     %HTMLAttrStatus,
7702 wakaba 1.208 condition => FEATURE_HTML5_DROPPED,
7703     mode => FEATURE_HTML5_DROPPED,
7704 wakaba 1.8 }),
7705 wakaba 1.40 check_start => sub {
7706     my ($self, $item, $element_state) = @_;
7707 wakaba 1.79
7708 wakaba 1.40 $self->_add_plus_elements ($element_state, {$HTML_NS => {nest => 1}});
7709 wakaba 1.79 $element_state->{in_rule_original} = $self->{flag}->{in_rule};
7710     $self->{flag}->{in_rule} = 1;
7711    
7712     $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7713     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7714 wakaba 1.40 },
7715     check_child_element => sub { },
7716     check_child_text => sub { },
7717     check_end => sub {
7718     my ($self, $item, $element_state) = @_;
7719 wakaba 1.79
7720 wakaba 1.40 $self->_remove_plus_elements ($element_state);
7721 wakaba 1.79 delete $self->{flag}->{in_rule} unless $element_state->{in_rule_original};
7722    
7723 wakaba 1.40 $HTMLChecker{check_end}->(@_);
7724 wakaba 1.8 },
7725     ## NOTE: "MAY be anything that, when the parent |datatemplate|
7726     ## is applied to some conforming data, results in a conforming DOM tree.":
7727     ## We don't check against this.
7728     };
7729    
7730     $Element->{$HTML_NS}->{nest} = {
7731 wakaba 1.40 %HTMLEmptyChecker,
7732 wakaba 1.208 status => FEATURE_HTML5_DROPPED,
7733 wakaba 1.40 check_attrs => $GetHTMLAttrsChecker->({
7734 wakaba 1.23 filter => $HTMLSelectorsAttrChecker,
7735     mode => sub {
7736     my ($self, $attr) = @_;
7737     my $value = $attr->value;
7738 wakaba 1.132 if ($value !~ /\A[^\x09\x0A\x0C\x0D\x20]+\z/) {
7739 wakaba 1.104 $self->{onerror}->(node => $attr, type => 'mode:syntax error',
7740     level => $self->{level}->{must});
7741 wakaba 1.23 }
7742     },
7743 wakaba 1.50 }, {
7744     %HTMLAttrStatus,
7745 wakaba 1.208 filter => FEATURE_HTML5_DROPPED,
7746     mode => FEATURE_HTML5_DROPPED,
7747 wakaba 1.8 }),
7748 wakaba 1.1 };
7749    
7750     $Element->{$HTML_NS}->{legend} = {
7751 wakaba 1.40 %HTMLPhrasingContentChecker,
7752 wakaba 1.153 status => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
7753 wakaba 1.52 check_attrs => $GetHTMLAttrsChecker->({
7754 wakaba 1.195 # XXX
7755 wakaba 1.52 # align => $GetHTMLEnumeratedAttrChecker->({
7756     # top => 1, bottom => 1, left => 1, right => 1,
7757     # }),
7758 wakaba 1.167 form => $HTMLFormAttrChecker,
7759 wakaba 1.52 }, {
7760 wakaba 1.49 %HTMLAttrStatus,
7761     %HTMLM12NCommonAttrStatus,
7762 wakaba 1.176 accesskey => FEATURE_HTML5_FD | FEATURE_M12N10_REC,
7763 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
7764 wakaba 1.167 form => FEATURE_HTML5_DROPPED,
7765 wakaba 1.187 lang => FEATURE_HTML5_REC,
7766 wakaba 1.49 }),
7767 wakaba 1.170 check_child_element => sub {
7768     my ($self, $item, $child_el, $child_nsuri, $child_ln,
7769     $child_is_transparent, $element_state) = @_;
7770 wakaba 1.195 ## XXX This does not work for |<legned><ins><blockquote>|
7771 wakaba 1.170 if ($item->{parent_state}->{in_figure}) {
7772     $HTMLFlowContentChecker{check_child_element}->(@_);
7773     } else {
7774     $HTMLPhrasingContentChecker{check_child_element}->(@_);
7775     }
7776     },
7777     check_child_text => sub {
7778     my ($self, $item, $child_node, $has_significant, $element_state) = @_;
7779     if ($item->{parent_state}->{in_figure}) {
7780     $HTMLFlowContentChecker{check_child_text}->(@_);
7781     } else {
7782     $HTMLPhrasingContentChecker{check_child_text}->(@_);
7783     }
7784     },
7785     check_start => sub {
7786     my ($self, $item, $element_state) = @_;
7787     $self->_add_minus_elements ($element_state, {$HTML_NS => {figure => 1}});
7788    
7789     $HTMLFlowContentChecker{check_start}->(@_);
7790     },
7791     check_end => sub {
7792     my ($self, $item, $element_state) = @_;
7793     $self->_remove_minus_elements ($element_state);
7794    
7795     $HTMLFlowContentChecker{check_end}->(@_);
7796     },
7797     }; # legend
7798 wakaba 1.1
7799     $Element->{$HTML_NS}->{div} = {
7800 wakaba 1.72 %HTMLFlowContentChecker,
7801 wakaba 1.187 status => FEATURE_HTML5_REC,
7802 wakaba 1.68 check_attrs => $GetHTMLAttrsChecker->({
7803     align => $GetHTMLEnumeratedAttrChecker->({
7804     left => 1, center => 1, right => 1, justify => 1,
7805     }),
7806     }, {
7807 wakaba 1.49 %HTMLAttrStatus,
7808 wakaba 1.82 %HTMLM12NXHTML2CommonAttrStatus,
7809 wakaba 1.49 align => FEATURE_M12N10_REC_DEPRECATED,
7810     datafld => FEATURE_HTML4_REC_RESERVED,
7811     dataformatas => FEATURE_HTML4_REC_RESERVED,
7812     datasrc => FEATURE_HTML4_REC_RESERVED,
7813 wakaba 1.187 lang => FEATURE_HTML5_REC,
7814 wakaba 1.49 }),
7815 wakaba 1.66 check_start => sub {
7816     my ($self, $item, $element_state) = @_;
7817    
7818     $element_state->{uri_info}->{datasrc}->{type}->{resource} = 1;
7819 wakaba 1.79 $element_state->{uri_info}->{template}->{type}->{resource} = 1;
7820     $element_state->{uri_info}->{ref}->{type}->{resource} = 1;
7821 wakaba 1.66 },
7822 wakaba 1.1 };
7823    
7824 wakaba 1.64 $Element->{$HTML_NS}->{center} = {
7825 wakaba 1.72 %HTMLFlowContentChecker,
7826 wakaba 1.64 status => FEATURE_M12N10_REC_DEPRECATED,
7827     check_attrs => $GetHTMLAttrsChecker->({}, {
7828     %HTMLAttrStatus,
7829     %HTMLM12NCommonAttrStatus,
7830 wakaba 1.187 lang => FEATURE_HTML5_REC,
7831 wakaba 1.64 }),
7832     };
7833    
7834 wakaba 1.1 $Element->{$HTML_NS}->{font} = {
7835 wakaba 1.40 %HTMLTransparentChecker,
7836 wakaba 1.78 status => FEATURE_HTML5_DROPPED | FEATURE_M12N10_REC_DEPRECATED,
7837 wakaba 1.70 check_attrs => $GetHTMLAttrsChecker->({
7838     ## TODO: HTML4 |size|, |color|, |face|
7839 wakaba 1.49 }, {
7840     %HTMLAttrStatus,
7841 wakaba 1.187 class => FEATURE_HTML5_LC | FEATURE_M12N10_REC,
7842 wakaba 1.49 color => FEATURE_M12N10_REC_DEPRECATED,
7843 wakaba 1.187 dir => FEATURE_HTML5_REC,
7844 wakaba 1.49 face => FEATURE_M12N10_REC_DEPRECATED,
7845 wakaba 1.187 id => FEATURE_HTML5_REC,
7846     lang => FEATURE_HTML5_REC,
7847 wakaba 1.49 size => FEATURE_M12N10_REC_DEPRECATED,
7848 wakaba 1.187 style => FEATURE_HTML5_REC,
7849     title => FEATURE_HTML5_REC,
7850 wakaba 1.49 }),
7851 wakaba 1.78 ## NOTE: When the |font| element was defined in the HTML5 specification,
7852     ## it is allowed only in a document with the WYSIWYG signature. The
7853     ## checker does not check whether there is the signature, since the
7854     ## signature is dropped, too, and has never been implemented. (In addition,
7855     ## for any |font| element an "element not defined" error is raised anyway,
7856     ## such that we don't have to raise an additional error.)
7857 wakaba 1.1 };
7858 wakaba 1.49
7859 wakaba 1.64 $Element->{$HTML_NS}->{basefont} = {
7860     %HTMLEmptyChecker,
7861     status => FEATURE_M12N10_REC_DEPRECATED,
7862     check_attrs => $GetHTMLAttrsChecker->({
7863     ## TODO: color, face, size
7864     }, {
7865     %HTMLAttrStatus,
7866     color => FEATURE_M12N10_REC_DEPRECATED,
7867     face => FEATURE_M12N10_REC_DEPRECATED,
7868 wakaba 1.187 id => FEATURE_HTML5_REC,
7869 wakaba 1.64 size => FEATURE_M12N10_REC_DEPRECATED,
7870     }),
7871 wakaba 1.188 }; # basefont
7872 wakaba 1.64
7873 wakaba 1.49 ## TODO: frameset FEATURE_M12N10_REC
7874 wakaba 1.188 ## class title id cols rows style(x10)
7875    
7876     $Element->{$HTML_NS}->{frameset} = {
7877     %HTMLEmptyChecker, # XXX
7878     status => FEATURE_M12N10_REC,
7879     check_attrs => $GetHTMLAttrsChecker->({
7880     ## XXX
7881     onafterprint => $HTMLEventHandlerAttrChecker,
7882     onbeforeprint => $HTMLEventHandlerAttrChecker,
7883     onbeforeunload => $HTMLEventHandlerAttrChecker,
7884     onblur => $HTMLEventHandlerAttrChecker,
7885     onerror => $HTMLEventHandlerAttrChecker,
7886     onfocus => $HTMLEventHandlerAttrChecker,
7887     onhashchange => $HTMLEventHandlerAttrChecker,
7888     onload => $HTMLEventHandlerAttrChecker,
7889     onmessage => $HTMLEventHandlerAttrChecker,
7890     onoffline => $HTMLEventHandlerAttrChecker,
7891     ononline => $HTMLEventHandlerAttrChecker,
7892     onpopstate => $HTMLEventHandlerAttrChecker,
7893     onredo => $HTMLEventHandlerAttrChecker,
7894     onresize => $HTMLEventHandlerAttrChecker,
7895     onstorage => $HTMLEventHandlerAttrChecker,
7896     onundo => $HTMLEventHandlerAttrChecker,
7897     onunload => $HTMLEventHandlerAttrChecker,
7898     }, {
7899     %HTMLAttrStatus,
7900     ## XXX
7901     onload => FEATURE_M12N10_REC,
7902     onunload => FEATURE_M12N10_REC,
7903     }),
7904     }; # frameset
7905    
7906 wakaba 1.49 ## frame frameborder longdesc marginheight marginwidth noresize scrolling src name(deprecated) class,id,title,style(x10)
7907     ## noframes Common, lang(xhtml10)
7908    
7909 wakaba 1.100 ## TODO: CR: rbc rtc @rbspan (M12NXHTML2Common)
7910 wakaba 1.56
7911 wakaba 1.61 ## TODO: xmp, listing, plaintext FEATURE_HTML32_REC_OBSOLETE
7912     ## TODO: ^^^ lang, dir, id, class [HTML 2.x] sdaform [HTML 2.0]
7913     ## xmp, listing sdapref[HTML2,0]
7914    
7915 wakaba 1.56 =pod
7916    
7917 wakaba 1.61 HTML 2.0 nextid @n
7918    
7919     RFC 2659: CERTS CRYPTOPTS
7920    
7921     ISO-HTML: pre-html, divN
7922 wakaba 1.82
7923     XHTML2: blockcode (Common), h (Common), separator (Common), l (Common),
7924     di (Common), nl (Common), handler (Common, type), standby (Common),
7925     summary (Common)
7926    
7927 wakaba 1.97 Access & XHTML2: access (LC)
7928 wakaba 1.82
7929     XML Events & XForms (for XHTML2 support; very, very low priority)
7930 wakaba 1.61
7931 wakaba 1.56 =cut
7932 wakaba 1.61
7933     ## NOTE: Where RFC 2659 allows additional attributes is unclear.
7934     ## We added them only to |a|. |link| and |form| might also allow them
7935     ## in theory.
7936 wakaba 1.1
7937     $Whatpm::ContentChecker::Namespace->{$HTML_NS}->{loaded} = 1;
7938    
7939     1;

admin@suikawiki.org
ViewVC Help
Powered by ViewVC 1.1.24