Parent Directory
|
Revision Log
|
Patch
| revision 1.24 by wakaba, Sat Jun 23 16:42:43 2007 UTC | revision 1.235 by wakaba, Sun Sep 6 08:02:54 2009 UTC | |
|---|---|---|
| # | Line 1 | Line 1 |
| 1 | package Whatpm::HTML; | package Whatpm::HTML; |
| 2 | use strict; | use strict; |
| 3 | our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; | our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
| 4 | use Error qw(:try); | |
| 5 | ||
| 6 | use Whatpm::HTML::Tokenizer; | |
| 7 | ||
| 8 | ## NOTE: This module don't check all HTML5 parse errors; character | |
| 9 | ## encoding related parse errors are expected to be handled by relevant | |
| 10 | ## modules. | |
| 11 | ## Parse errors for control characters that are not allowed in HTML5 | |
| 12 | ## documents, for surrogate code points, and for noncharacter code | |
| 13 | ## points, as well as U+FFFD substitions for characters whose code points | |
| 14 | ## is higher than U+10FFFF may be detected by combining the parser with | |
| 15 | ## the checker implemented by Whatpm::Charset::UnicodeChecker (for its | |
| 16 | ## usage example, see |t/HTML-tree.t| in the Whatpm package or the | |
| 17 | ## WebHACC::Language::HTML module in the WebHACC package). | |
| 18 | ||
| 19 | ## ISSUE: | ## ISSUE: |
| 20 | ## var doc = implementation.createDocument (null, null, null); | ## var doc = implementation.createDocument (null, null, null); |
| 21 | ## doc.write (''); | ## doc.write (''); |
| 22 | ## alert (doc.compatMode); | ## alert (doc.compatMode); |
| 23 | ||
| 24 | my $permitted_slash_tag_name = { | require IO::Handle; |
| base => 1, | ||
| link => 1, | ||
| meta => 1, | ||
| hr => 1, | ||
| br => 1, | ||
| img=> 1, | ||
| embed => 1, | ||
| param => 1, | ||
| area => 1, | ||
| col => 1, | ||
| input => 1, | ||
| }; | ||
| 25 | ||
| 26 | my $c1_entity_char = { | ## Namespace URLs |
| 27 | 0x80 => 0x20AC, | |
| 28 | 0x81 => 0xFFFD, | my $HTML_NS = q<http://www.w3.org/1999/xhtml>; |
| 29 | 0x82 => 0x201A, | my $MML_NS = q<http://www.w3.org/1998/Math/MathML>; |
| 30 | 0x83 => 0x0192, | my $SVG_NS = q<http://www.w3.org/2000/svg>; |
| 31 | 0x84 => 0x201E, | my $XLINK_NS = q<http://www.w3.org/1999/xlink>; |
| 32 | 0x85 => 0x2026, | my $XML_NS = q<http://www.w3.org/XML/1998/namespace>; |
| 33 | 0x86 => 0x2020, | my $XMLNS_NS = q<http://www.w3.org/2000/xmlns/>; |
| 34 | 0x87 => 0x2021, | |
| 35 | 0x88 => 0x02C6, | ## Element categories |
| 36 | 0x89 => 0x2030, | |
| 37 | 0x8A => 0x0160, | ## Bits 12-15 |
| 38 | 0x8B => 0x2039, | sub SPECIAL_EL () { 0b1_000000000000000 } |
| 39 | 0x8C => 0x0152, | sub SCOPING_EL () { 0b1_00000000000000 } |
| 40 | 0x8D => 0xFFFD, | sub FORMATTING_EL () { 0b1_0000000000000 } |
| 41 | 0x8E => 0x017D, | sub PHRASING_EL () { 0b1_000000000000 } |
| 42 | 0x8F => 0xFFFD, | |
| 43 | 0x90 => 0xFFFD, | ## Bits 10-11 |
| 44 | 0x91 => 0x2018, | #sub FOREIGN_EL () { 0b1_00000000000 } # see Whatpm::HTML::Tokenizer |
| 45 | 0x92 => 0x2019, | sub FOREIGN_FLOW_CONTENT_EL () { 0b1_0000000000 } |
| 46 | 0x93 => 0x201C, | |
| 47 | 0x94 => 0x201D, | ## Bits 6-9 |
| 48 | 0x95 => 0x2022, | sub TABLE_SCOPING_EL () { 0b1_000000000 } |
| 49 | 0x96 => 0x2013, | sub TABLE_ROWS_SCOPING_EL () { 0b1_00000000 } |
| 50 | 0x97 => 0x2014, | sub TABLE_ROW_SCOPING_EL () { 0b1_0000000 } |
| 51 | 0x98 => 0x02DC, | sub TABLE_ROWS_EL () { 0b1_000000 } |
| 52 | 0x99 => 0x2122, | |
| 53 | 0x9A => 0x0161, | ## Bit 5 |
| 54 | 0x9B => 0x203A, | sub ADDRESS_DIV_P_EL () { 0b1_00000 } |
| 55 | 0x9C => 0x0153, | |
| 56 | 0x9D => 0xFFFD, | ## NOTE: Used in </body> and EOF algorithms. |
| 57 | 0x9E => 0x017E, | ## Bit 4 |
| 58 | 0x9F => 0x0178, | sub ALL_END_TAG_OPTIONAL_EL () { 0b1_0000 } |
| 59 | }; # $c1_entity_char | |
| 60 | ## NOTE: Used in "generate implied end tags" algorithm. | |
| 61 | my $special_category = { | ## NOTE: There is a code where a modified version of |
| 62 | address => 1, area => 1, base => 1, basefont => 1, bgsound => 1, | ## END_TAG_OPTIONAL_EL is used in "generate implied end tags" |
| 63 | blockquote => 1, body => 1, br => 1, center => 1, col => 1, colgroup => 1, | ## implementation (search for the algorithm name). |
| 64 | dd => 1, dir => 1, div => 1, dl => 1, dt => 1, embed => 1, fieldset => 1, | ## Bit 3 |
| 65 | form => 1, frame => 1, frameset => 1, h1 => 1, h2 => 1, h3 => 1, | sub END_TAG_OPTIONAL_EL () { 0b1_000 } |
| 66 | h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, iframe => 1, image => 1, | |
| 67 | img => 1, input => 1, isindex => 1, li => 1, link => 1, listing => 1, | ## Bits 0-2 |
| 68 | menu => 1, meta => 1, noembed => 1, noframes => 1, noscript => 1, | |
| 69 | ol => 1, optgroup => 1, option => 1, p => 1, param => 1, plaintext => 1, | sub MISC_SPECIAL_EL () { SPECIAL_EL | 0b000 } |
| 70 | pre => 1, script => 1, select => 1, spacer => 1, style => 1, tbody => 1, | sub FORM_EL () { SPECIAL_EL | 0b001 } |
| 71 | textarea => 1, tfoot => 1, thead => 1, title => 1, tr => 1, ul => 1, wbr => 1, | sub FRAMESET_EL () { SPECIAL_EL | 0b010 } |
| 72 | }; | sub HEADING_EL () { SPECIAL_EL | 0b011 } |
| 73 | my $scoping_category = { | sub SELECT_EL () { SPECIAL_EL | 0b100 } |
| 74 | button => 1, caption => 1, html => 1, marquee => 1, object => 1, | sub SCRIPT_EL () { SPECIAL_EL | 0b101 } |
| 75 | table => 1, td => 1, th => 1, | |
| 76 | sub ADDRESS_DIV_EL () { SPECIAL_EL | ADDRESS_DIV_P_EL | 0b001 } | |
| 77 | sub BODY_EL () { SPECIAL_EL | ALL_END_TAG_OPTIONAL_EL | 0b001 } | |
| 78 | ||
| 79 | sub DTDD_EL () { | |
| 80 | SPECIAL_EL | | |
| 81 | END_TAG_OPTIONAL_EL | | |
| 82 | ALL_END_TAG_OPTIONAL_EL | | |
| 83 | 0b010 | |
| 84 | } | |
| 85 | sub LI_EL () { | |
| 86 | SPECIAL_EL | | |
| 87 | END_TAG_OPTIONAL_EL | | |
| 88 | ALL_END_TAG_OPTIONAL_EL | | |
| 89 | 0b100 | |
| 90 | } | |
| 91 | sub P_EL () { | |
| 92 | SPECIAL_EL | | |
| 93 | ADDRESS_DIV_P_EL | | |
| 94 | END_TAG_OPTIONAL_EL | | |
| 95 | ALL_END_TAG_OPTIONAL_EL | | |
| 96 | 0b001 | |
| 97 | } | |
| 98 | ||
| 99 | sub TABLE_ROW_EL () { | |
| 100 | SPECIAL_EL | | |
| 101 | TABLE_ROWS_EL | | |
| 102 | TABLE_ROW_SCOPING_EL | | |
| 103 | ALL_END_TAG_OPTIONAL_EL | | |
| 104 | 0b001 | |
| 105 | } | |
| 106 | sub TABLE_ROW_GROUP_EL () { | |
| 107 | SPECIAL_EL | | |
| 108 | TABLE_ROWS_EL | | |
| 109 | TABLE_ROWS_SCOPING_EL | | |
| 110 | ALL_END_TAG_OPTIONAL_EL | | |
| 111 | 0b001 | |
| 112 | } | |
| 113 | ||
| 114 | sub MISC_SCOPING_EL () { SCOPING_EL | 0b000 } | |
| 115 | sub BUTTON_EL () { SCOPING_EL | 0b001 } | |
| 116 | sub CAPTION_EL () { SCOPING_EL | 0b010 } | |
| 117 | sub HTML_EL () { | |
| 118 | SCOPING_EL | | |
| 119 | TABLE_SCOPING_EL | | |
| 120 | TABLE_ROWS_SCOPING_EL | | |
| 121 | TABLE_ROW_SCOPING_EL | | |
| 122 | ALL_END_TAG_OPTIONAL_EL | | |
| 123 | 0b001 | |
| 124 | } | |
| 125 | sub TABLE_EL () { | |
| 126 | SCOPING_EL | | |
| 127 | TABLE_ROWS_EL | | |
| 128 | TABLE_SCOPING_EL | | |
| 129 | 0b001 | |
| 130 | } | |
| 131 | sub TABLE_CELL_EL () { | |
| 132 | SCOPING_EL | | |
| 133 | TABLE_ROW_SCOPING_EL | | |
| 134 | ALL_END_TAG_OPTIONAL_EL | | |
| 135 | 0b001 | |
| 136 | } | |
| 137 | ||
| 138 | sub MISC_FORMATTING_EL () { FORMATTING_EL | 0b000 } | |
| 139 | sub A_EL () { FORMATTING_EL | 0b001 } | |
| 140 | sub NOBR_EL () { FORMATTING_EL | 0b010 } | |
| 141 | ||
| 142 | sub RUBY_EL () { PHRASING_EL | 0b001 } | |
| 143 | ||
| 144 | ## ISSUE: ALL_END_TAG_OPTIONAL_EL? | |
| 145 | sub OPTGROUP_EL () { PHRASING_EL | END_TAG_OPTIONAL_EL | 0b001 } | |
| 146 | sub OPTION_EL () { PHRASING_EL | END_TAG_OPTIONAL_EL | 0b010 } | |
| 147 | sub RUBY_COMPONENT_EL () { PHRASING_EL | END_TAG_OPTIONAL_EL | 0b100 } | |
| 148 | ||
| 149 | sub MML_AXML_EL () { PHRASING_EL | FOREIGN_EL | 0b001 } | |
| 150 | ||
| 151 | my $el_category = { | |
| 152 | a => A_EL, | |
| 153 | address => ADDRESS_DIV_EL, | |
| 154 | applet => MISC_SCOPING_EL, | |
| 155 | area => MISC_SPECIAL_EL, | |
| 156 | article => MISC_SPECIAL_EL, | |
| 157 | aside => MISC_SPECIAL_EL, | |
| 158 | b => FORMATTING_EL, | |
| 159 | base => MISC_SPECIAL_EL, | |
| 160 | basefont => MISC_SPECIAL_EL, | |
| 161 | bgsound => MISC_SPECIAL_EL, | |
| 162 | big => FORMATTING_EL, | |
| 163 | blockquote => MISC_SPECIAL_EL, | |
| 164 | body => BODY_EL, | |
| 165 | br => MISC_SPECIAL_EL, | |
| 166 | button => BUTTON_EL, | |
| 167 | caption => CAPTION_EL, | |
| 168 | center => MISC_SPECIAL_EL, | |
| 169 | col => MISC_SPECIAL_EL, | |
| 170 | colgroup => MISC_SPECIAL_EL, | |
| 171 | command => MISC_SPECIAL_EL, | |
| 172 | datagrid => MISC_SPECIAL_EL, | |
| 173 | dd => DTDD_EL, | |
| 174 | details => MISC_SPECIAL_EL, | |
| 175 | dialog => MISC_SPECIAL_EL, | |
| 176 | dir => MISC_SPECIAL_EL, | |
| 177 | div => ADDRESS_DIV_EL, | |
| 178 | dl => MISC_SPECIAL_EL, | |
| 179 | dt => DTDD_EL, | |
| 180 | em => FORMATTING_EL, | |
| 181 | embed => MISC_SPECIAL_EL, | |
| 182 | fieldset => MISC_SPECIAL_EL, | |
| 183 | figure => MISC_SPECIAL_EL, | |
| 184 | font => FORMATTING_EL, | |
| 185 | footer => MISC_SPECIAL_EL, | |
| 186 | form => FORM_EL, | |
| 187 | frame => MISC_SPECIAL_EL, | |
| 188 | frameset => FRAMESET_EL, | |
| 189 | h1 => HEADING_EL, | |
| 190 | h2 => HEADING_EL, | |
| 191 | h3 => HEADING_EL, | |
| 192 | h4 => HEADING_EL, | |
| 193 | h5 => HEADING_EL, | |
| 194 | h6 => HEADING_EL, | |
| 195 | head => MISC_SPECIAL_EL, | |
| 196 | header => MISC_SPECIAL_EL, | |
| 197 | hr => MISC_SPECIAL_EL, | |
| 198 | html => HTML_EL, | |
| 199 | i => FORMATTING_EL, | |
| 200 | iframe => MISC_SPECIAL_EL, | |
| 201 | img => MISC_SPECIAL_EL, | |
| 202 | #image => MISC_SPECIAL_EL, ## NOTE: Commented out in the spec. | |
| 203 | input => MISC_SPECIAL_EL, | |
| 204 | isindex => MISC_SPECIAL_EL, | |
| 205 | ## XXX keygen? (Whether a void element is in Special or not does not | |
| 206 | ## affect to the processing, however.) | |
| 207 | li => LI_EL, | |
| 208 | link => MISC_SPECIAL_EL, | |
| 209 | listing => MISC_SPECIAL_EL, | |
| 210 | marquee => MISC_SCOPING_EL, | |
| 211 | menu => MISC_SPECIAL_EL, | |
| 212 | meta => MISC_SPECIAL_EL, | |
| 213 | nav => MISC_SPECIAL_EL, | |
| 214 | nobr => NOBR_EL, | |
| 215 | noembed => MISC_SPECIAL_EL, | |
| 216 | noframes => MISC_SPECIAL_EL, | |
| 217 | noscript => MISC_SPECIAL_EL, | |
| 218 | object => MISC_SCOPING_EL, | |
| 219 | ol => MISC_SPECIAL_EL, | |
| 220 | optgroup => OPTGROUP_EL, | |
| 221 | option => OPTION_EL, | |
| 222 | p => P_EL, | |
| 223 | param => MISC_SPECIAL_EL, | |
| 224 | plaintext => MISC_SPECIAL_EL, | |
| 225 | pre => MISC_SPECIAL_EL, | |
| 226 | rp => RUBY_COMPONENT_EL, | |
| 227 | rt => RUBY_COMPONENT_EL, | |
| 228 | ruby => RUBY_EL, | |
| 229 | s => FORMATTING_EL, | |
| 230 | script => MISC_SPECIAL_EL, | |
| 231 | select => SELECT_EL, | |
| 232 | section => MISC_SPECIAL_EL, | |
| 233 | small => FORMATTING_EL, | |
| 234 | spacer => MISC_SPECIAL_EL, | |
| 235 | strike => FORMATTING_EL, | |
| 236 | strong => FORMATTING_EL, | |
| 237 | style => MISC_SPECIAL_EL, | |
| 238 | table => TABLE_EL, | |
| 239 | tbody => TABLE_ROW_GROUP_EL, | |
| 240 | td => TABLE_CELL_EL, | |
| 241 | textarea => MISC_SPECIAL_EL, | |
| 242 | tfoot => TABLE_ROW_GROUP_EL, | |
| 243 | th => TABLE_CELL_EL, | |
| 244 | thead => TABLE_ROW_GROUP_EL, | |
| 245 | title => MISC_SPECIAL_EL, | |
| 246 | tr => TABLE_ROW_EL, | |
| 247 | tt => FORMATTING_EL, | |
| 248 | u => FORMATTING_EL, | |
| 249 | ul => MISC_SPECIAL_EL, | |
| 250 | wbr => MISC_SPECIAL_EL, | |
| 251 | }; | }; |
| 252 | my $formatting_category = { | |
| 253 | a => 1, b => 1, big => 1, em => 1, font => 1, i => 1, nobr => 1, | my $el_category_f = { |
| 254 | s => 1, small => 1, strile => 1, strong => 1, tt => 1, u => 1, | $MML_NS => { |
| 255 | 'annotation-xml' => MML_AXML_EL, | |
| 256 | mi => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 257 | mo => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 258 | mn => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 259 | ms => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 260 | mtext => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 261 | }, | |
| 262 | $SVG_NS => { | |
| 263 | foreignObject => SCOPING_EL | FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 264 | desc => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 265 | title => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 266 | }, | |
| 267 | ## NOTE: In addition, FOREIGN_EL is set to non-HTML elements. | |
| 268 | }; | }; |
| # $phrasing_category: all other elements | ||
| 269 | ||
| 270 | sub parse_string ($$$;$) { | my $svg_attr_name = { |
| 271 | my $self = shift->new; | attributename => 'attributeName', |
| 272 | my $s = \$_[0]; | attributetype => 'attributeType', |
| 273 | $self->{document} = $_[1]; | basefrequency => 'baseFrequency', |
| 274 | baseprofile => 'baseProfile', | |
| 275 | calcmode => 'calcMode', | |
| 276 | clippathunits => 'clipPathUnits', | |
| 277 | contentscripttype => 'contentScriptType', | |
| 278 | contentstyletype => 'contentStyleType', | |
| 279 | diffuseconstant => 'diffuseConstant', | |
| 280 | edgemode => 'edgeMode', | |
| 281 | externalresourcesrequired => 'externalResourcesRequired', | |
| 282 | filterres => 'filterRes', | |
| 283 | filterunits => 'filterUnits', | |
| 284 | glyphref => 'glyphRef', | |
| 285 | gradienttransform => 'gradientTransform', | |
| 286 | gradientunits => 'gradientUnits', | |
| 287 | kernelmatrix => 'kernelMatrix', | |
| 288 | kernelunitlength => 'kernelUnitLength', | |
| 289 | keypoints => 'keyPoints', | |
| 290 | keysplines => 'keySplines', | |
| 291 | keytimes => 'keyTimes', | |
| 292 | lengthadjust => 'lengthAdjust', | |
| 293 | limitingconeangle => 'limitingConeAngle', | |
| 294 | markerheight => 'markerHeight', | |
| 295 | markerunits => 'markerUnits', | |
| 296 | markerwidth => 'markerWidth', | |
| 297 | maskcontentunits => 'maskContentUnits', | |
| 298 | maskunits => 'maskUnits', | |
| 299 | numoctaves => 'numOctaves', | |
| 300 | pathlength => 'pathLength', | |
| 301 | patterncontentunits => 'patternContentUnits', | |
| 302 | patterntransform => 'patternTransform', | |
| 303 | patternunits => 'patternUnits', | |
| 304 | pointsatx => 'pointsAtX', | |
| 305 | pointsaty => 'pointsAtY', | |
| 306 | pointsatz => 'pointsAtZ', | |
| 307 | preservealpha => 'preserveAlpha', | |
| 308 | preserveaspectratio => 'preserveAspectRatio', | |
| 309 | primitiveunits => 'primitiveUnits', | |
| 310 | refx => 'refX', | |
| 311 | refy => 'refY', | |
| 312 | repeatcount => 'repeatCount', | |
| 313 | repeatdur => 'repeatDur', | |
| 314 | requiredextensions => 'requiredExtensions', | |
| 315 | requiredfeatures => 'requiredFeatures', | |
| 316 | specularconstant => 'specularConstant', | |
| 317 | specularexponent => 'specularExponent', | |
| 318 | spreadmethod => 'spreadMethod', | |
| 319 | startoffset => 'startOffset', | |
| 320 | stddeviation => 'stdDeviation', | |
| 321 | stitchtiles => 'stitchTiles', | |
| 322 | surfacescale => 'surfaceScale', | |
| 323 | systemlanguage => 'systemLanguage', | |
| 324 | tablevalues => 'tableValues', | |
| 325 | targetx => 'targetX', | |
| 326 | targety => 'targetY', | |
| 327 | textlength => 'textLength', | |
| 328 | viewbox => 'viewBox', | |
| 329 | viewtarget => 'viewTarget', | |
| 330 | xchannelselector => 'xChannelSelector', | |
| 331 | ychannelselector => 'yChannelSelector', | |
| 332 | zoomandpan => 'zoomAndPan', | |
| 333 | }; | |
| 334 | ||
| 335 | ## NOTE: |set_inner_html| copies most of this method's code | my $foreign_attr_xname = { |
| 336 | 'xlink:actuate' => [$XLINK_NS, ['xlink', 'actuate']], | |
| 337 | 'xlink:arcrole' => [$XLINK_NS, ['xlink', 'arcrole']], | |
| 338 | 'xlink:href' => [$XLINK_NS, ['xlink', 'href']], | |
| 339 | 'xlink:role' => [$XLINK_NS, ['xlink', 'role']], | |
| 340 | 'xlink:show' => [$XLINK_NS, ['xlink', 'show']], | |
| 341 | 'xlink:title' => [$XLINK_NS, ['xlink', 'title']], | |
| 342 | 'xlink:type' => [$XLINK_NS, ['xlink', 'type']], | |
| 343 | 'xml:base' => [$XML_NS, ['xml', 'base']], | |
| 344 | 'xml:lang' => [$XML_NS, ['xml', 'lang']], | |
| 345 | 'xml:space' => [$XML_NS, ['xml', 'space']], | |
| 346 | 'xmlns' => [$XMLNS_NS, [undef, 'xmlns']], | |
| 347 | 'xmlns:xlink' => [$XMLNS_NS, ['xmlns', 'xlink']], | |
| 348 | }; | |
| 349 | ||
| 350 | my $i = 0; | ## ISSUE: xmlns:xlink="non-xlink-ns" is not an error. |
| my $line = 1; | ||
| my $column = 0; | ||
| $self->{set_next_input_character} = sub { | ||
| my $self = shift; | ||
| 351 | ||
| 352 | pop @{$self->{prev_input_character}}; | ## TODO: Invoke the reset algorithm when a resettable element is |
| 353 | unshift @{$self->{prev_input_character}}, $self->{next_input_character}; | ## created (cf. HTML5 revision 2259). |
| 354 | ||
| 355 | $self->{next_input_character} = -1 and return if $i >= length $$s; | sub parse_byte_string ($$$$;$) { |
| 356 | $self->{next_input_character} = ord substr $$s, $i++, 1; | my $self = shift; |
| 357 | $column++; | my $charset_name = shift; |
| 358 | open my $input, '<', ref $_[0] ? $_[0] : \($_[0]); | |
| 359 | if ($self->{next_input_character} == 0x000A) { # LF | return $self->parse_byte_stream ($charset_name, $input, @_[1..$#_]); |
| 360 | $line++; | } # parse_byte_string |
| 361 | $column = 0; | |
| 362 | } elsif ($self->{next_input_character} == 0x000D) { # CR | sub parse_byte_stream ($$$$;$$) { |
| 363 | $i++ if substr ($$s, $i, 1) eq "\x0A"; | # my ($self, $charset_name, $byte_stream, $doc, $onerror, $get_wrapper) = @_; |
| 364 | $self->{next_input_character} = 0x000A; # LF # MUST | my $self = ref $_[0] ? shift : shift->new; |
| 365 | $line++; | my $charset_name = shift; |
| 366 | $column = 0; | my $byte_stream = $_[0]; |
| } elsif ($self->{next_input_character} > 0x10FFFF) { | ||
| $self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST | ||
| } elsif ($self->{next_input_character} == 0x0000) { # NULL | ||
| !!!parse-error (type => 'NULL'); | ||
| $self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST | ||
| } | ||
| }; | ||
| $self->{prev_input_character} = [-1, -1, -1]; | ||
| $self->{next_input_character} = -1; | ||
| 367 | ||
| 368 | my $onerror = $_[2] || sub { | my $onerror = $_[2] || sub { |
| 369 | my (%opt) = @_; | my (%opt) = @_; |
| 370 | warn "Parse error ($opt{type}) at line $opt{line} column $opt{column}\n"; | warn "Parse error ($opt{type})\n"; |
| }; | ||
| $self->{parse_error} = sub { | ||
| $onerror->(@_, line => $line, column => $column); | ||
| 371 | }; | }; |
| 372 | $self->{parse_error} = $onerror; # updated later by parse_char_string | |
| 373 | ||
| 374 | $self->_initialize_tokenizer; | my $get_wrapper = $_[3] || sub ($) { |
| 375 | $self->_initialize_tree_constructor; | return $_[0]; # $_[0] = byte stream handle, returned = arg to char handle |
| $self->_construct_tree; | ||
| $self->_terminate_tree_constructor; | ||
| return $self->{document}; | ||
| } # parse_string | ||
| sub new ($) { | ||
| my $class = shift; | ||
| my $self = bless {}, $class; | ||
| $self->{set_next_input_character} = sub { | ||
| $self->{next_input_character} = -1; | ||
| }; | ||
| $self->{parse_error} = sub { | ||
| # | ||
| 376 | }; | }; |
| return $self; | ||
| } # new | ||
| ## Implementations MUST act as if state machine in the spec | ||
| sub _initialize_tokenizer ($) { | ||
| my $self = shift; | ||
| $self->{state} = 'data'; # MUST | ||
| $self->{content_model_flag} = 'PCDATA'; # be | ||
| undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE | ||
| undef $self->{current_attribute}; | ||
| undef $self->{last_emitted_start_tag_name}; | ||
| undef $self->{last_attribute_value_state}; | ||
| $self->{char} = []; | ||
| # $self->{next_input_character} | ||
| !!!next-input-character; | ||
| $self->{token} = []; | ||
| # $self->{escape} | ||
| } # _initialize_tokenizer | ||
| ## A token has: | ||
| ## ->{type} eq 'DOCTYPE', 'start tag', 'end tag', 'comment', | ||
| ## 'character', or 'end-of-file' | ||
| ## ->{name} (DOCTYPE, start tag (tag name), end tag (tag name)) | ||
| ## ->{public_identifier} (DOCTYPE) | ||
| ## ->{system_identifier} (DOCTYPE) | ||
| ## ->{correct} == 1 or 0 (DOCTYPE) | ||
| ## ->{attributes} isa HASH (start tag, end tag) | ||
| ## ->{data} (comment, character) | ||
| ## Emitted token MUST immediately be handled by the tree construction state. | ||
| ## Before each step, UA MAY check to see if either one of the scripts in | ||
| ## "list of scripts that will execute as soon as possible" or the first | ||
| ## script in the "list of scripts that will execute asynchronously", | ||
| ## has completed loading. If one has, then it MUST be executed | ||
| ## and removed from the list. | ||
| sub _get_next_token ($) { | ||
| my $self = shift; | ||
| if (@{$self->{token}}) { | ||
| return shift @{$self->{token}}; | ||
| } | ||
| A: { | ||
| if ($self->{state} eq 'data') { | ||
| if ($self->{next_input_character} == 0x0026) { # & | ||
| if ($self->{content_model_flag} eq 'PCDATA' or | ||
| $self->{content_model_flag} eq 'RCDATA') { | ||
| $self->{state} = 'entity data'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } else { | ||
| # | ||
| } | ||
| } elsif ($self->{next_input_character} == 0x002D) { # - | ||
| if ($self->{content_model_flag} eq 'RCDATA' or | ||
| $self->{content_model_flag} eq 'CDATA') { | ||
| unless ($self->{escape}) { | ||
| if ($self->{prev_input_character}->[0] == 0x002D and # - | ||
| $self->{prev_input_character}->[1] == 0x0021 and # ! | ||
| $self->{prev_input_character}->[2] == 0x003C) { # < | ||
| $self->{escape} = 1; | ||
| } | ||
| } | ||
| } | ||
| # | ||
| } elsif ($self->{next_input_character} == 0x003C) { # < | ||
| if ($self->{content_model_flag} eq 'PCDATA' or | ||
| (($self->{content_model_flag} eq 'CDATA' or | ||
| $self->{content_model_flag} eq 'RCDATA') and | ||
| not $self->{escape})) { | ||
| $self->{state} = 'tag open'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } else { | ||
| # | ||
| } | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| if ($self->{escape} and | ||
| ($self->{content_model_flag} eq 'RCDATA' or | ||
| $self->{content_model_flag} eq 'CDATA')) { | ||
| if ($self->{prev_input_character}->[0] == 0x002D and # - | ||
| $self->{prev_input_character}->[1] == 0x002D) { # - | ||
| delete $self->{escape}; | ||
| } | ||
| } | ||
| # | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!emit ({type => 'end-of-file'}); | ||
| last A; ## TODO: ok? | ||
| } | ||
| # Anything else | ||
| my $token = {type => 'character', | ||
| data => chr $self->{next_input_character}}; | ||
| ## Stay in the data state | ||
| !!!next-input-character; | ||
| !!!emit ($token); | ||
| redo A; | ||
| } elsif ($self->{state} eq 'entity data') { | ||
| ## (cannot happen in CDATA state) | ||
| my $token = $self->_tokenize_attempt_to_consume_an_entity; | ||
| $self->{state} = 'data'; | ||
| # next-input-character is already done | ||
| unless (defined $token) { | ||
| !!!emit ({type => 'character', data => '&'}); | ||
| } else { | ||
| !!!emit ($token); | ||
| } | ||
| 377 | ||
| 378 | redo A; | ## HTML5 encoding sniffing algorithm |
| 379 | } elsif ($self->{state} eq 'tag open') { | require Message::Charset::Info; |
| 380 | if ($self->{content_model_flag} eq 'RCDATA' or | my $charset; |
| 381 | $self->{content_model_flag} eq 'CDATA') { | my $buffer; |
| 382 | if ($self->{next_input_character} == 0x002F) { # / | my ($char_stream, $e_status); |
| 383 | !!!next-input-character; | |
| 384 | $self->{state} = 'close tag open'; | SNIFFING: { |
| 385 | redo A; | ## NOTE: By setting |allow_fallback| option true when the |
| 386 | } else { | ## |get_decode_handle| method is invoked, we ignore what the HTML5 |
| 387 | ## reconsume | ## spec requires, i.e. unsupported encoding should be ignored. |
| 388 | $self->{state} = 'data'; | ## TODO: We should not do this unless the parser is invoked |
| 389 | ## in the conformance checking mode, in which this behavior | |
| 390 | !!!emit ({type => 'character', data => '<'}); | ## would be useful. |
| redo A; | ||
| } | ||
| } elsif ($self->{content_model_flag} eq 'PCDATA') { | ||
| if ($self->{next_input_character} == 0x0021) { # ! | ||
| $self->{state} = 'markup declaration open'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x002F) { # / | ||
| $self->{state} = 'close tag open'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x005A) { # A..Z | ||
| $self->{current_token} | ||
| = {type => 'start tag', | ||
| tag_name => chr ($self->{next_input_character} + 0x0020)}; | ||
| $self->{state} = 'tag name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif (0x0061 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x007A) { # a..z | ||
| $self->{current_token} = {type => 'start tag', | ||
| tag_name => chr ($self->{next_input_character})}; | ||
| $self->{state} = 'tag name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| !!!parse-error (type => 'empty start tag'); | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ({type => 'character', data => '<>'}); | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003F) { # ? | ||
| !!!parse-error (type => 'pio'); | ||
| $self->{state} = 'bogus comment'; | ||
| ## $self->{next_input_character} is intentionally left as is | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'bare stago'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 391 | ||
| 392 | !!!emit ({type => 'character', data => '<'}); | ## Step 1 |
| 393 | if (defined $charset_name) { | |
| 394 | redo A; | $charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 395 | } | ## TODO: Is this ok? Transfer protocol's parameter should be |
| 396 | ## interpreted in its semantics? | |
| 397 | ||
| 398 | ($char_stream, $e_status) = $charset->get_decode_handle | |
| 399 | ($byte_stream, allow_error_reporting => 1, | |
| 400 | allow_fallback => 1); | |
| 401 | if ($char_stream) { | |
| 402 | $self->{confident} = 1; | |
| 403 | last SNIFFING; | |
| 404 | } else { | } else { |
| 405 | die "$0: $self->{content_model_flag}: Unknown content model flag"; | !!!parse-error (type => 'charset:not supported', |
| 406 | layer => 'encode', | |
| 407 | line => 1, column => 1, | |
| 408 | value => $charset_name, | |
| 409 | level => $self->{level}->{uncertain}); | |
| 410 | } | } |
| 411 | } elsif ($self->{state} eq 'close tag open') { | } |
| if ($self->{content_model_flag} eq 'RCDATA' or | ||
| $self->{content_model_flag} eq 'CDATA') { | ||
| if (defined $self->{last_emitted_start_tag_name}) { | ||
| my @next_char; | ||
| TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) { | ||
| push @next_char, $self->{next_input_character}; | ||
| my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1); | ||
| my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c; | ||
| if ($self->{next_input_character} == $c or $self->{next_input_character} == $C) { | ||
| !!!next-input-character; | ||
| next TAGNAME; | ||
| } else { | ||
| $self->{next_input_character} = shift @next_char; # reconsume | ||
| !!!back-next-input-character (@next_char); | ||
| $self->{state} = 'data'; | ||
| !!!emit ({type => 'character', data => '</'}); | ||
| redo A; | ||
| } | ||
| } | ||
| push @next_char, $self->{next_input_character}; | ||
| unless ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020 or # SP | ||
| $self->{next_input_character} == 0x003E or # > | ||
| $self->{next_input_character} == 0x002F or # / | ||
| $self->{next_input_character} == -1) { | ||
| $self->{next_input_character} = shift @next_char; # reconsume | ||
| !!!back-next-input-character (@next_char); | ||
| $self->{state} = 'data'; | ||
| !!!emit ({type => 'character', data => '</'}); | ||
| redo A; | ||
| } else { | ||
| $self->{next_input_character} = shift @next_char; | ||
| !!!back-next-input-character (@next_char); | ||
| # and consume... | ||
| } | ||
| } else { | ||
| ## No start tag token has ever been emitted | ||
| # next-input-character is already done | ||
| $self->{state} = 'data'; | ||
| !!!emit ({type => 'character', data => '</'}); | ||
| redo A; | ||
| } | ||
| } | ||
| if (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x005A) { # A..Z | ||
| $self->{current_token} = {type => 'end tag', | ||
| tag_name => chr ($self->{next_input_character} + 0x0020)}; | ||
| $self->{state} = 'tag name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif (0x0061 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x007A) { # a..z | ||
| $self->{current_token} = {type => 'end tag', | ||
| tag_name => chr ($self->{next_input_character})}; | ||
| $self->{state} = 'tag name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| !!!parse-error (type => 'empty end tag'); | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'bare etago'); | ||
| $self->{state} = 'data'; | ||
| # reconsume | ||
| !!!emit ({type => 'character', data => '</'}); | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'bogus end tag'); | ||
| $self->{state} = 'bogus comment'; | ||
| ## $self->{next_input_character} is intentionally left as is | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'tag name') { | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| $self->{state} = 'before attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } elsif (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x005A) { # A..Z | ||
| $self->{current_token}->{tag_name} .= chr ($self->{next_input_character} + 0x0020); | ||
| # start tag or end tag | ||
| ## Stay in this state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed tag'); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| # reconsume | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x002F) { # / | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x003E and # > | ||
| $self->{current_token}->{type} eq 'start tag' and | ||
| $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { | ||
| # permitted slash | ||
| # | ||
| } else { | ||
| !!!parse-error (type => 'nestc'); | ||
| } | ||
| $self->{state} = 'before attribute name'; | ||
| # next-input-character is already done | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{tag_name} .= chr $self->{next_input_character}; | ||
| # start tag or end tag | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'before attribute name') { | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } elsif (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x005A) { # A..Z | ||
| $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), | ||
| value => ''}; | ||
| $self->{state} = 'attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x002F) { # / | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x003E and # > | ||
| $self->{current_token}->{type} eq 'start tag' and | ||
| $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { | ||
| # permitted slash | ||
| # | ||
| } else { | ||
| !!!parse-error (type => 'nestc'); | ||
| } | ||
| ## Stay in the state | ||
| # next-input-character is already done | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed tag'); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| # reconsume | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } else { | ||
| $self->{current_attribute} = {name => chr ($self->{next_input_character}), | ||
| value => ''}; | ||
| $self->{state} = 'attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'attribute name') { | ||
| my $before_leave = sub { | ||
| if (exists $self->{current_token}->{attributes} # start tag or end tag | ||
| ->{$self->{current_attribute}->{name}}) { # MUST | ||
| !!!parse-error (type => 'dupulicate attribute'); | ||
| ## Discard $self->{current_attribute} # MUST | ||
| } else { | ||
| $self->{current_token}->{attributes}->{$self->{current_attribute}->{name}} | ||
| = $self->{current_attribute}; | ||
| } | ||
| }; # $before_leave | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| $before_leave->(); | ||
| $self->{state} = 'after attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003D) { # = | ||
| $before_leave->(); | ||
| $self->{state} = 'before attribute value'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| $before_leave->(); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } elsif (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x005A) { # A..Z | ||
| $self->{current_attribute}->{name} .= chr ($self->{next_input_character} + 0x0020); | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x002F) { # / | ||
| $before_leave->(); | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x003E and # > | ||
| $self->{current_token}->{type} eq 'start tag' and | ||
| $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { | ||
| # permitted slash | ||
| # | ||
| } else { | ||
| !!!parse-error (type => 'nestc'); | ||
| } | ||
| $self->{state} = 'before attribute name'; | ||
| # next-input-character is already done | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed tag'); | ||
| $before_leave->(); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| # reconsume | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } else { | ||
| $self->{current_attribute}->{name} .= chr ($self->{next_input_character}); | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'after attribute name') { | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003D) { # = | ||
| $self->{state} = 'before attribute value'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } elsif (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x005A) { # A..Z | ||
| $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), | ||
| value => ''}; | ||
| $self->{state} = 'attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x002F) { # / | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x003E and # > | ||
| $self->{current_token}->{type} eq 'start tag' and | ||
| $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { | ||
| # permitted slash | ||
| # | ||
| } else { | ||
| !!!parse-error (type => 'nestc'); | ||
| } | ||
| $self->{state} = 'before attribute name'; | ||
| # next-input-character is already done | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed tag'); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| # reconsume | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } else { | ||
| $self->{current_attribute} = {name => chr ($self->{next_input_character}), | ||
| value => ''}; | ||
| $self->{state} = 'attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'before attribute value') { | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0022) { # " | ||
| $self->{state} = 'attribute value (double-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0026) { # & | ||
| $self->{state} = 'attribute value (unquoted)'; | ||
| ## reconsume | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0027) { # ' | ||
| $self->{state} = 'attribute value (single-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed tag'); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 412 | ||
| 413 | !!!emit ($self->{current_token}); # start tag or end tag | ## Step 2 |
| 414 | my $byte_buffer = ''; | |
| 415 | for (1..1024) { | |
| 416 | my $char = $byte_stream->getc; | |
| 417 | last unless defined $char; | |
| 418 | $byte_buffer .= $char; | |
| 419 | } ## TODO: timeout | |
| 420 | ||
| 421 | redo A; | ## Step 3 |
| 422 | } else { | if ($byte_buffer =~ /^\xFE\xFF/) { |
| 423 | $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | $charset = Message::Charset::Info->get_by_html_name ('utf-16be'); |
| 424 | $self->{state} = 'attribute value (unquoted)'; | ($char_stream, $e_status) = $charset->get_decode_handle |
| 425 | !!!next-input-character; | ($byte_stream, allow_error_reporting => 1, |
| 426 | redo A; | allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 427 | } | $self->{confident} = 1; |
| 428 | } elsif ($self->{state} eq 'attribute value (double-quoted)') { | last SNIFFING; |
| 429 | if ($self->{next_input_character} == 0x0022) { # " | } elsif ($byte_buffer =~ /^\xFF\xFE/) { |
| 430 | $self->{state} = 'before attribute name'; | $charset = Message::Charset::Info->get_by_html_name ('utf-16le'); |
| 431 | !!!next-input-character; | ($char_stream, $e_status) = $charset->get_decode_handle |
| 432 | redo A; | ($byte_stream, allow_error_reporting => 1, |
| 433 | } elsif ($self->{next_input_character} == 0x0026) { # & | allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 434 | $self->{last_attribute_value_state} = 'attribute value (double-quoted)'; | $self->{confident} = 1; |
| 435 | $self->{state} = 'entity in attribute value'; | last SNIFFING; |
| 436 | !!!next-input-character; | } elsif ($byte_buffer =~ /^\xEF\xBB\xBF/) { |
| 437 | redo A; | $charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 438 | } elsif ($self->{next_input_character} == -1) { | ($char_stream, $e_status) = $charset->get_decode_handle |
| 439 | !!!parse-error (type => 'unclosed attribute value'); | ($byte_stream, allow_error_reporting => 1, |
| 440 | if ($self->{current_token}->{type} eq 'start tag') { | allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 441 | $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | $self->{confident} = 1; |
| 442 | } elsif ($self->{current_token}->{type} eq 'end tag') { | last SNIFFING; |
| 443 | $self->{content_model_flag} = 'PCDATA'; # MUST | } |
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 444 | ||
| 445 | !!!emit ($self->{current_token}); # start tag or end tag | ## Step 4 |
| 446 | ## TODO: <meta charset> | |
| 447 | ||
| 448 | redo A; | ## Step 5 |
| 449 | } else { | ## TODO: from history |
| $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'attribute value (single-quoted)') { | ||
| if ($self->{next_input_character} == 0x0027) { # ' | ||
| $self->{state} = 'before attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0026) { # & | ||
| $self->{last_attribute_value_state} = 'attribute value (single-quoted)'; | ||
| $self->{state} = 'entity in attribute value'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed attribute value'); | ||
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 450 | ||
| 451 | !!!emit ($self->{current_token}); # start tag or end tag | ## Step 6 |
| 452 | require Whatpm::Charset::UniversalCharDet; | |
| 453 | $charset_name = Whatpm::Charset::UniversalCharDet->detect_byte_string | |
| 454 | ($byte_buffer); | |
| 455 | if (defined $charset_name) { | |
| 456 | $charset = Message::Charset::Info->get_by_html_name ($charset_name); | |
| 457 | ||
| 458 | require Whatpm::Charset::DecodeHandle; | |
| 459 | $buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new | |
| 460 | ($byte_stream); | |
| 461 | ($char_stream, $e_status) = $charset->get_decode_handle | |
| 462 | ($buffer, allow_error_reporting => 1, | |
| 463 | allow_fallback => 1, byte_buffer => \$byte_buffer); | |
| 464 | if ($char_stream) { | |
| 465 | $buffer->{buffer} = $byte_buffer; | |
| 466 | !!!parse-error (type => 'sniffing:chardet', | |
| 467 | text => $charset_name, | |
| 468 | level => $self->{level}->{info}, | |
| 469 | layer => 'encode', | |
| 470 | line => 1, column => 1); | |
| 471 | $self->{confident} = 0; | |
| 472 | last SNIFFING; | |
| 473 | } | |
| 474 | } | |
| 475 | ||
| 476 | redo A; | ## Step 7: default |
| 477 | } else { | ## TODO: Make this configurable. |
| 478 | $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | $charset = Message::Charset::Info->get_by_html_name ('windows-1252'); |
| 479 | ## Stay in the state | ## NOTE: We choose |windows-1252| here, since |utf-8| should be |
| 480 | !!!next-input-character; | ## detectable in the step 6. |
| 481 | redo A; | require Whatpm::Charset::DecodeHandle; |
| 482 | } | $buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
| 483 | } elsif ($self->{state} eq 'attribute value (unquoted)') { | ($byte_stream); |
| 484 | if ($self->{next_input_character} == 0x0009 or # HT | ($char_stream, $e_status) |
| 485 | $self->{next_input_character} == 0x000A or # LF | = $charset->get_decode_handle ($buffer, |
| 486 | $self->{next_input_character} == 0x000B or # HT | allow_error_reporting => 1, |
| 487 | $self->{next_input_character} == 0x000C or # FF | allow_fallback => 1, |
| 488 | $self->{next_input_character} == 0x0020) { # SP | byte_buffer => \$byte_buffer); |
| 489 | $self->{state} = 'before attribute name'; | $buffer->{buffer} = $byte_buffer; |
| 490 | !!!next-input-character; | !!!parse-error (type => 'sniffing:default', |
| 491 | redo A; | text => 'windows-1252', |
| 492 | } elsif ($self->{next_input_character} == 0x0026) { # & | level => $self->{level}->{info}, |
| 493 | $self->{last_attribute_value_state} = 'attribute value (unquoted)'; | line => 1, column => 1, |
| 494 | $self->{state} = 'entity in attribute value'; | layer => 'encode'); |
| 495 | !!!next-input-character; | $self->{confident} = 0; |
| 496 | redo A; | } # SNIFFING |
| 497 | } elsif ($self->{next_input_character} == 0x003E) { # > | |
| 498 | if ($self->{current_token}->{type} eq 'start tag') { | if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 499 | $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | $self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 500 | } elsif ($self->{current_token}->{type} eq 'end tag') { | !!!parse-error (type => 'chardecode:fallback', |
| 501 | $self->{content_model_flag} = 'PCDATA'; # MUST | #text => $self->{input_encoding}, |
| 502 | if ($self->{current_token}->{attributes}) { | level => $self->{level}->{uncertain}, |
| 503 | !!!parse-error (type => 'end tag attribute'); | line => 1, column => 1, |
| 504 | } | layer => 'encode'); |
| 505 | } else { | } elsif (not ($e_status & |
| 506 | die "$0: $self->{current_token}->{type}: Unknown token type"; | Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) { |
| 507 | } | $self->{input_encoding} = $charset->get_iana_name; |
| 508 | $self->{state} = 'data'; | !!!parse-error (type => 'chardecode:no error', |
| 509 | !!!next-input-character; | text => $self->{input_encoding}, |
| 510 | level => $self->{level}->{uncertain}, | |
| 511 | !!!emit ($self->{current_token}); # start tag or end tag | line => 1, column => 1, |
| 512 | layer => 'encode'); | |
| 513 | redo A; | } else { |
| 514 | } elsif ($self->{next_input_character} == -1) { | $self->{input_encoding} = $charset->get_iana_name; |
| 515 | !!!parse-error (type => 'unclosed tag'); | } |
| if ($self->{current_token}->{type} eq 'start tag') { | ||
| $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; | ||
| } elsif ($self->{current_token}->{type} eq 'end tag') { | ||
| $self->{content_model_flag} = 'PCDATA'; # MUST | ||
| if ($self->{current_token}->{attributes}) { | ||
| !!!parse-error (type => 'end tag attribute'); | ||
| } | ||
| } else { | ||
| die "$0: $self->{current_token}->{type}: Unknown token type"; | ||
| } | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 516 | ||
| 517 | !!!emit ($self->{current_token}); # start tag or end tag | $self->{change_encoding} = sub { |
| 518 | my $self = shift; | |
| 519 | $charset_name = shift; | |
| 520 | my $token = shift; | |
| 521 | ||
| 522 | redo A; | $charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 523 | } else { | ($char_stream, $e_status) = $charset->get_decode_handle |
| 524 | $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | ($byte_stream, allow_error_reporting => 1, allow_fallback => 1, |
| 525 | ## Stay in the state | byte_buffer => \ $buffer->{buffer}); |
| 526 | !!!next-input-character; | |
| 527 | redo A; | if ($char_stream) { # if supported |
| 528 | ## "Change the encoding" algorithm: | |
| 529 | ||
| 530 | ## Step 1 | |
| 531 | if (defined $self->{input_encoding} and | |
| 532 | $self->{input_encoding} eq $charset_name) { | |
| 533 | !!!parse-error (type => 'charset label:matching', | |
| 534 | text => $charset_name, | |
| 535 | level => $self->{level}->{info}); | |
| 536 | $self->{confident} = 1; | |
| 537 | return; | |
| 538 | } | } |
| } elsif ($self->{state} eq 'entity in attribute value') { | ||
| my $token = $self->_tokenize_attempt_to_consume_an_entity; | ||
| 539 | ||
| 540 | unless (defined $token) { | ## Step 2 (HTML5 revision 3205) |
| 541 | $self->{current_attribute}->{value} .= '&'; | if (defined $self->{input_encoding} and |
| 542 | } else { | Message::Charset::Info->get_by_html_name ($self->{input_encoding}) |
| 543 | $self->{current_attribute}->{value} .= $token->{data}; | ->{category} & Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 544 | ## ISSUE: spec says "append the returned character token to the current attribute's value" | $self->{confident} = 1; |
| 545 | return; | |
| 546 | } | } |
| 547 | ||
| 548 | $self->{state} = $self->{last_attribute_value_state}; | ## Step 3 |
| 549 | # next-input-character is already done | if ($charset->{category} & |
| 550 | redo A; | Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 551 | } elsif ($self->{state} eq 'bogus comment') { | $charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 552 | ## (only happen if PCDATA state) | ($char_stream, $e_status) = $charset->get_decode_handle |
| 553 | ($byte_stream, | |
| 554 | byte_buffer => \ $buffer->{buffer}); | |
| 555 | } | |
| 556 | $charset_name = $charset->get_iana_name; | |
| 557 | ||
| 558 | !!!parse-error (type => 'charset label detected', | |
| 559 | text => $self->{input_encoding}, | |
| 560 | value => $charset_name, | |
| 561 | level => $self->{level}->{warn}, | |
| 562 | token => $token); | |
| 563 | ||
| 564 | my $token = {type => 'comment', data => ''}; | ## Step 4 |
| 565 | # if (can) { | |
| 566 | BC: { | ## change the encoding on the fly. |
| 567 | if ($self->{next_input_character} == 0x003E) { # > | #$self->{confident} = 1; |
| 568 | $self->{state} = 'data'; | #return; |
| 569 | !!!next-input-character; | # } |
| !!!emit ($token); | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| !!!emit ($token); | ||
| redo A; | ||
| } else { | ||
| $token->{data} .= chr ($self->{next_input_character}); | ||
| !!!next-input-character; | ||
| redo BC; | ||
| } | ||
| } # BC | ||
| } elsif ($self->{state} eq 'markup declaration open') { | ||
| ## (only happen if PCDATA state) | ||
| my @next_char; | ||
| push @next_char, $self->{next_input_character}; | ||
| 570 | ||
| 571 | if ($self->{next_input_character} == 0x002D) { # - | ## Step 5 |
| 572 | !!!next-input-character; | throw Whatpm::HTML::RestartParser (); |
| 573 | push @next_char, $self->{next_input_character}; | } |
| 574 | if ($self->{next_input_character} == 0x002D) { # - | }; # $self->{change_encoding} |
| $self->{current_token} = {type => 'comment', data => ''}; | ||
| $self->{state} = 'comment start'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{next_input_character} == 0x0044 or # D | ||
| $self->{next_input_character} == 0x0064) { # d | ||
| !!!next-input-character; | ||
| push @next_char, $self->{next_input_character}; | ||
| if ($self->{next_input_character} == 0x004F or # O | ||
| $self->{next_input_character} == 0x006F) { # o | ||
| !!!next-input-character; | ||
| push @next_char, $self->{next_input_character}; | ||
| if ($self->{next_input_character} == 0x0043 or # C | ||
| $self->{next_input_character} == 0x0063) { # c | ||
| !!!next-input-character; | ||
| push @next_char, $self->{next_input_character}; | ||
| if ($self->{next_input_character} == 0x0054 or # T | ||
| $self->{next_input_character} == 0x0074) { # t | ||
| !!!next-input-character; | ||
| push @next_char, $self->{next_input_character}; | ||
| if ($self->{next_input_character} == 0x0059 or # Y | ||
| $self->{next_input_character} == 0x0079) { # y | ||
| !!!next-input-character; | ||
| push @next_char, $self->{next_input_character}; | ||
| if ($self->{next_input_character} == 0x0050 or # P | ||
| $self->{next_input_character} == 0x0070) { # p | ||
| !!!next-input-character; | ||
| push @next_char, $self->{next_input_character}; | ||
| if ($self->{next_input_character} == 0x0045 or # E | ||
| $self->{next_input_character} == 0x0065) { # e | ||
| ## ISSUE: What a stupid code this is! | ||
| $self->{state} = 'DOCTYPE'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| 575 | ||
| 576 | !!!parse-error (type => 'bogus comment open'); | my $char_onerror = sub { |
| 577 | $self->{next_input_character} = shift @next_char; | my (undef, $type, %opt) = @_; |
| 578 | !!!back-next-input-character (@next_char); | !!!parse-error (layer => 'encode', |
| 579 | $self->{state} = 'bogus comment'; | line => $self->{line}, column => $self->{column} + 1, |
| 580 | redo A; | %opt, type => $type); |
| 581 | if ($opt{octets}) { | |
| 582 | ## ISSUE: typos in spec: chacacters, is is a parse error | ${$opt{octets}} = "\x{FFFD}"; # relacement character |
| 583 | ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it? | } |
| 584 | } elsif ($self->{state} eq 'comment start') { | }; |
| if ($self->{next_input_character} == 0x002D) { # - | ||
| $self->{state} = 'comment start dash'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| !!!parse-error (type => 'bogus comment'); | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # comment | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed comment'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 585 | ||
| 586 | !!!emit ($self->{current_token}); # comment | my $wrapped_char_stream = $get_wrapper->($char_stream); |
| 587 | $wrapped_char_stream->onerror ($char_onerror); | |
| 588 | ||
| 589 | redo A; | my @args = ($_[1], $_[2]); # $doc, $onerror - $get_wrapper = undef; |
| 590 | } else { | my $return; |
| 591 | $self->{current_token}->{data} # comment | try { |
| 592 | .= chr ($self->{next_input_character}); | $return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 593 | $self->{state} = 'comment'; | } catch Whatpm::HTML::RestartParser with { |
| 594 | !!!next-input-character; | ## NOTE: Invoked after {change_encoding}. |
| 595 | redo A; | |
| 596 | } | if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 597 | } elsif ($self->{state} eq 'comment start dash') { | $self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 598 | if ($self->{next_input_character} == 0x002D) { # - | !!!parse-error (type => 'chardecode:fallback', |
| 599 | $self->{state} = 'comment end'; | level => $self->{level}->{uncertain}, |
| 600 | !!!next-input-character; | #text => $self->{input_encoding}, |
| 601 | redo A; | line => 1, column => 1, |
| 602 | } elsif ($self->{next_input_character} == 0x003E) { # > | layer => 'encode'); |
| 603 | !!!parse-error (type => 'bogus comment'); | } elsif (not ($e_status & |
| 604 | $self->{state} = 'data'; | Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) { |
| 605 | !!!next-input-character; | $self->{input_encoding} = $charset->get_iana_name; |
| 606 | !!!parse-error (type => 'chardecode:no error', | |
| 607 | !!!emit ($self->{current_token}); # comment | text => $self->{input_encoding}, |
| 608 | level => $self->{level}->{uncertain}, | |
| 609 | redo A; | line => 1, column => 1, |
| 610 | } elsif ($self->{next_input_character} == -1) { | layer => 'encode'); |
| 611 | !!!parse-error (type => 'unclosed comment'); | } else { |
| 612 | $self->{state} = 'data'; | $self->{input_encoding} = $charset->get_iana_name; |
| 613 | ## reconsume | } |
| 614 | $self->{confident} = 1; | |
| 615 | ||
| 616 | !!!emit ($self->{current_token}); # comment | $wrapped_char_stream = $get_wrapper->($char_stream); |
| 617 | $wrapped_char_stream->onerror ($char_onerror); | |
| 618 | ||
| 619 | redo A; | $return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 620 | } else { | }; |
| 621 | $self->{current_token}->{data} # comment | return $return; |
| 622 | .= chr ($self->{next_input_character}); | } # parse_byte_stream |
| $self->{state} = 'comment'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'comment') { | ||
| if ($self->{next_input_character} == 0x002D) { # - | ||
| $self->{state} = 'comment end dash'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed comment'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 623 | ||
| 624 | !!!emit ($self->{current_token}); # comment | ## NOTE: HTML5 spec says that the encoding layer MUST NOT strip BOM |
| 625 | ## and the HTML layer MUST ignore it. However, we does strip BOM in | |
| 626 | ## the encoding layer and the HTML layer does not ignore any U+FEFF, | |
| 627 | ## because the core part of our HTML parser expects a string of character, | |
| 628 | ## not a string of bytes or code units or anything which might contain a BOM. | |
| 629 | ## Therefore, any parser interface that accepts a string of bytes, | |
| 630 | ## such as |parse_byte_string| in this module, must ensure that it does | |
| 631 | ## strip the BOM and never strip any ZWNBSP. | |
| 632 | ||
| 633 | redo A; | sub parse_char_string ($$$;$$) { |
| 634 | } else { | #my ($self, $s, $doc, $onerror, $get_wrapper) = @_; |
| 635 | $self->{current_token}->{data} .= chr ($self->{next_input_character}); # comment | my $self = shift; |
| 636 | ## Stay in the state | my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 637 | !!!next-input-character; | require Whatpm::Charset::DecodeHandle; |
| 638 | redo A; | my $input = Whatpm::Charset::DecodeHandle::CharString->new ($s); |
| 639 | } | return $self->parse_char_stream ($input, @_[1..$#_]); |
| 640 | } elsif ($self->{state} eq 'comment end dash') { | } # parse_char_string |
| 641 | if ($self->{next_input_character} == 0x002D) { # - | *parse_string = \&parse_char_string; ## NOTE: Alias for backward compatibility. |
| 642 | $self->{state} = 'comment end'; | |
| 643 | !!!next-input-character; | sub parse_char_stream ($$$;$$) { |
| 644 | redo A; | my $self = ref $_[0] ? shift : shift->new; |
| 645 | } elsif ($self->{next_input_character} == -1) { | my $input = $_[0]; |
| 646 | !!!parse-error (type => 'unclosed comment'); | $self->{document} = $_[1]; |
| 647 | $self->{state} = 'data'; | @{$self->{document}->child_nodes} = (); |
| ## reconsume | ||
| 648 | ||
| 649 | !!!emit ($self->{current_token}); # comment | ## NOTE: |set_inner_html| copies most of this method's code |
| 650 | ||
| 651 | redo A; | ## Confidence: irrelevant. |
| 652 | } else { | $self->{confident} = 1 unless exists $self->{confident}; |
| $self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment | ||
| $self->{state} = 'comment'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'comment end') { | ||
| if ($self->{next_input_character} == 0x003E) { # > | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # comment | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x002D) { # - | ||
| !!!parse-error (type => 'dash in comment'); | ||
| $self->{current_token}->{data} .= '-'; # comment | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed comment'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 653 | ||
| 654 | !!!emit ($self->{current_token}); # comment | $self->{document}->input_encoding ($self->{input_encoding}) |
| 655 | if defined $self->{input_encoding}; | |
| 656 | ## TODO: |{input_encoding}| is needless? | |
| 657 | ||
| 658 | $self->{line_prev} = $self->{line} = 1; | |
| 659 | $self->{column_prev} = -1; | |
| 660 | $self->{column} = 0; | |
| 661 | $self->{set_nc} = sub { | |
| 662 | my $self = shift; | |
| 663 | ||
| 664 | redo A; | my $char = ''; |
| 665 | } else { | if (defined $self->{next_nc}) { |
| 666 | !!!parse-error (type => 'dash in comment'); | $char = $self->{next_nc}; |
| 667 | $self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment | delete $self->{next_nc}; |
| 668 | $self->{state} = 'comment'; | $self->{nc} = ord $char; |
| 669 | !!!next-input-character; | } else { |
| 670 | redo A; | $self->{char_buffer} = ''; |
| 671 | } | $self->{char_buffer_pos} = 0; |
| } elsif ($self->{state} eq 'DOCTYPE') { | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| $self->{state} = 'before DOCTYPE name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'no space before DOCTYPE name'); | ||
| $self->{state} = 'before DOCTYPE name'; | ||
| ## reconsume | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'before DOCTYPE name') { | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| !!!parse-error (type => 'no DOCTYPE name'); | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ({type => 'DOCTYPE'}); # incorrect | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'no DOCTYPE name'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 672 | ||
| 673 | !!!emit ({type => 'DOCTYPE'}); # incorrect | my $count = $input->manakai_read_until |
| 674 | ($self->{char_buffer}, qr/[^\x00\x0A\x0D]/, $self->{char_buffer_pos}); | |
| 675 | if ($count) { | |
| 676 | $self->{line_prev} = $self->{line}; | |
| 677 | $self->{column_prev} = $self->{column}; | |
| 678 | $self->{column}++; | |
| 679 | $self->{nc} | |
| 680 | = ord substr ($self->{char_buffer}, $self->{char_buffer_pos}++, 1); | |
| 681 | return; | |
| 682 | } | |
| 683 | ||
| 684 | redo A; | if ($input->read ($char, 1)) { |
| 685 | $self->{nc} = ord $char; | |
| 686 | } else { | } else { |
| 687 | $self->{current_token} | $self->{nc} = -1; |
| 688 | = {type => 'DOCTYPE', | return; |
| 689 | name => chr ($self->{next_input_character}), | } |
| 690 | correct => 1}; | } |
| ## ISSUE: "Set the token's name name to the" in the spec | ||
| $self->{state} = 'DOCTYPE name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'DOCTYPE name') { | ||
| ## ISSUE: Redundant "First," in the spec. | ||
| if ($self->{next_input_character} == 0x0009 or # HT | ||
| $self->{next_input_character} == 0x000A or # LF | ||
| $self->{next_input_character} == 0x000B or # VT | ||
| $self->{next_input_character} == 0x000C or # FF | ||
| $self->{next_input_character} == 0x0020) { # SP | ||
| $self->{state} = 'after DOCTYPE name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| 691 | ||
| 692 | redo A; | ($self->{line_prev}, $self->{column_prev}) |
| 693 | } else { | = ($self->{line}, $self->{column}); |
| 694 | $self->{current_token}->{name} | $self->{column}++; |
| 695 | .= chr ($self->{next_input_character}); # DOCTYPE | |
| 696 | ## Stay in the state | if ($self->{nc} == 0x000A) { # LF |
| 697 | !!!next-input-character; | !!!cp ('j1'); |
| 698 | redo A; | $self->{line}++; |
| 699 | } | $self->{column} = 0; |
| 700 | } elsif ($self->{state} eq 'after DOCTYPE name') { | } elsif ($self->{nc} == 0x000D) { # CR |
| 701 | if ($self->{next_input_character} == 0x0009 or # HT | !!!cp ('j2'); |
| 702 | $self->{next_input_character} == 0x000A or # LF | ## TODO: support for abort/streaming |
| 703 | $self->{next_input_character} == 0x000B or # VT | my $next = ''; |
| 704 | $self->{next_input_character} == 0x000C or # FF | if ($input->read ($next, 1) and $next ne "\x0A") { |
| 705 | $self->{next_input_character} == 0x0020) { # SP | $self->{next_nc} = $next; |
| 706 | ## Stay in the state | } |
| 707 | !!!next-input-character; | $self->{nc} = 0x000A; # LF # MUST |
| 708 | redo A; | $self->{line}++; |
| 709 | } elsif ($self->{next_input_character} == 0x003E) { # > | $self->{column} = 0; |
| 710 | $self->{state} = 'data'; | } elsif ($self->{nc} == 0x0000) { # NULL |
| 711 | !!!next-input-character; | !!!cp ('j4'); |
| 712 | !!!parse-error (type => 'NULL'); | |
| 713 | !!!emit ($self->{current_token}); # DOCTYPE | $self->{nc} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 714 | } | |
| 715 | redo A; | }; |
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0050 or # P | ||
| $self->{next_input_character} == 0x0070) { # p | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0055 or # U | ||
| $self->{next_input_character} == 0x0075) { # u | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0042 or # B | ||
| $self->{next_input_character} == 0x0062) { # b | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x004C or # L | ||
| $self->{next_input_character} == 0x006C) { # l | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0049 or # I | ||
| $self->{next_input_character} == 0x0069) { # i | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0043 or # C | ||
| $self->{next_input_character} == 0x0063) { # c | ||
| $self->{state} = 'before DOCTYPE public identifier'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| 716 | ||
| 717 | # | $self->{read_until} = sub { |
| 718 | } elsif ($self->{next_input_character} == 0x0053 or # S | #my ($scalar, $specials_range, $offset) = @_; |
| 719 | $self->{next_input_character} == 0x0073) { # s | return 0 if defined $self->{next_nc}; |
| 720 | !!!next-input-character; | |
| 721 | if ($self->{next_input_character} == 0x0059 or # Y | my $pattern = qr/[^$_[1]\x00\x0A\x0D]/; |
| 722 | $self->{next_input_character} == 0x0079) { # y | my $offset = $_[2] || 0; |
| 723 | !!!next-input-character; | |
| 724 | if ($self->{next_input_character} == 0x0053 or # S | if ($self->{char_buffer_pos} < length $self->{char_buffer}) { |
| 725 | $self->{next_input_character} == 0x0073) { # s | pos ($self->{char_buffer}) = $self->{char_buffer_pos}; |
| 726 | !!!next-input-character; | if ($self->{char_buffer} =~ /\G(?>$pattern)+/) { |
| 727 | if ($self->{next_input_character} == 0x0054 or # T | substr ($_[0], $offset) |
| 728 | $self->{next_input_character} == 0x0074) { # t | = substr ($self->{char_buffer}, $-[0], $+[0] - $-[0]); |
| 729 | !!!next-input-character; | my $count = $+[0] - $-[0]; |
| 730 | if ($self->{next_input_character} == 0x0045 or # E | if ($count) { |
| 731 | $self->{next_input_character} == 0x0065) { # e | $self->{column} += $count; |
| 732 | !!!next-input-character; | $self->{char_buffer_pos} += $count; |
| 733 | if ($self->{next_input_character} == 0x004D or # M | $self->{line_prev} = $self->{line}; |
| 734 | $self->{next_input_character} == 0x006D) { # m | $self->{column_prev} = $self->{column} - 1; |
| 735 | $self->{state} = 'before DOCTYPE system identifier'; | $self->{nc} = -1; |
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } | ||
| } | ||
| } | ||
| 736 | } | } |
| 737 | return $count; | |
| # | ||
| 738 | } else { | } else { |
| 739 | !!!next-input-character; | return 0; |
| # | ||
| 740 | } | } |
| 741 | } else { | |
| 742 | my $count = $input->manakai_read_until ($_[0], $pattern, $_[2]); | |
| 743 | if ($count) { | |
| 744 | $self->{column} += $count; | |
| 745 | $self->{line_prev} = $self->{line}; | |
| 746 | $self->{column_prev} = $self->{column} - 1; | |
| 747 | $self->{nc} = -1; | |
| 748 | } | |
| 749 | return $count; | |
| 750 | } | |
| 751 | }; # $self->{read_until} | |
| 752 | ||
| 753 | !!!parse-error (type => 'string after DOCTYPE name'); | my $onerror = $_[2] || sub { |
| 754 | $self->{state} = 'bogus DOCTYPE'; | my (%opt) = @_; |
| 755 | # next-input-character is already done | my $line = $opt{token} ? $opt{token}->{line} : $opt{line}; |
| 756 | redo A; | my $column = $opt{token} ? $opt{token}->{column} : $opt{column}; |
| 757 | } elsif ($self->{state} eq 'before DOCTYPE public identifier') { | warn "Parse error ($opt{type}) at line $line column $column\n"; |
| 758 | if ({ | }; |
| 759 | 0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, | $self->{parse_error} = sub { |
| 760 | #0x000D => 1, # HT, LF, VT, FF, SP, CR | $onerror->(line => $self->{line}, column => $self->{column}, @_); |
| 761 | }->{$self->{next_input_character}}) { | }; |
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} eq 0x0022) { # " | ||
| $self->{current_token}->{public_identifier} = ''; # DOCTYPE | ||
| $self->{state} = 'DOCTYPE public identifier (double-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} eq 0x0027) { # ' | ||
| $self->{current_token}->{public_identifier} = ''; # DOCTYPE | ||
| $self->{state} = 'DOCTYPE public identifier (single-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} eq 0x003E) { # > | ||
| !!!parse-error (type => 'no PUBLIC literal'); | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'string after PUBLIC'); | ||
| $self->{state} = 'bogus DOCTYPE'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'DOCTYPE public identifier (double-quoted)') { | ||
| if ($self->{next_input_character} == 0x0022) { # " | ||
| $self->{state} = 'after DOCTYPE public identifier'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed PUBLIC literal'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{public_identifier} # DOCTYPE | ||
| .= chr $self->{next_input_character}; | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'DOCTYPE public identifier (single-quoted)') { | ||
| if ($self->{next_input_character} == 0x0027) { # ' | ||
| $self->{state} = 'after DOCTYPE public identifier'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed PUBLIC literal'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{public_identifier} # DOCTYPE | ||
| .= chr $self->{next_input_character}; | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'after DOCTYPE public identifier') { | ||
| if ({ | ||
| 0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, | ||
| #0x000D => 1, # HT, LF, VT, FF, SP, CR | ||
| }->{$self->{next_input_character}}) { | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0022) { # " | ||
| $self->{current_token}->{system_identifier} = ''; # DOCTYPE | ||
| $self->{state} = 'DOCTYPE system identifier (double-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0027) { # ' | ||
| $self->{current_token}->{system_identifier} = ''; # DOCTYPE | ||
| $self->{state} = 'DOCTYPE system identifier (single-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## recomsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'string after PUBLIC literal'); | ||
| $self->{state} = 'bogus DOCTYPE'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'before DOCTYPE system identifier') { | ||
| if ({ | ||
| 0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, | ||
| #0x000D => 1, # HT, LF, VT, FF, SP, CR | ||
| }->{$self->{next_input_character}}) { | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0022) { # " | ||
| $self->{current_token}->{system_identifier} = ''; # DOCTYPE | ||
| $self->{state} = 'DOCTYPE system identifier (double-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0027) { # ' | ||
| $self->{current_token}->{system_identifier} = ''; # DOCTYPE | ||
| $self->{state} = 'DOCTYPE system identifier (single-quoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| !!!parse-error (type => 'no SYSTEM literal'); | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## recomsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'string after PUBLIC literal'); | ||
| $self->{state} = 'bogus DOCTYPE'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'DOCTYPE system identifier (double-quoted)') { | ||
| if ($self->{next_input_character} == 0x0022) { # " | ||
| $self->{state} = 'after DOCTYPE system identifier'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed SYSTEM literal'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{system_identifier} # DOCTYPE | ||
| .= chr $self->{next_input_character}; | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'DOCTYPE system identifier (single-quoted)') { | ||
| if ($self->{next_input_character} == 0x0027) { # ' | ||
| $self->{state} = 'after DOCTYPE system identifier'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed SYSTEM literal'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{system_identifier} # DOCTYPE | ||
| .= chr $self->{next_input_character}; | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'after DOCTYPE system identifier') { | ||
| if ({ | ||
| 0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, | ||
| #0x000D => 1, # HT, LF, VT, FF, SP, CR | ||
| }->{$self->{next_input_character}}) { | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x003E) { # > | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## recomsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'string after SYSTEM literal'); | ||
| $self->{state} = 'bogus DOCTYPE'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'bogus DOCTYPE') { | ||
| if ($self->{next_input_character} == 0x003E) { # > | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| 762 | ||
| 763 | delete $self->{current_token}->{correct}; | my $char_onerror = sub { |
| 764 | !!!emit ($self->{current_token}); # DOCTYPE | my (undef, $type, %opt) = @_; |
| 765 | !!!parse-error (layer => 'encode', | |
| 766 | line => $self->{line}, column => $self->{column} + 1, | |
| 767 | %opt, type => $type); | |
| 768 | }; # $char_onerror | |
| 769 | ||
| 770 | if ($_[3]) { | |
| 771 | $input = $_[3]->($input); | |
| 772 | $input->onerror ($char_onerror); | |
| 773 | } else { | |
| 774 | $input->onerror ($char_onerror) unless defined $input->onerror; | |
| 775 | } | |
| 776 | ||
| 777 | redo A; | $self->_initialize_tokenizer; |
| 778 | } else { | $self->_initialize_tree_constructor; |
| 779 | ## Stay in the state | $self->_construct_tree; |
| 780 | !!!next-input-character; | $self->_terminate_tree_constructor; |
| redo A; | ||
| } | ||
| } else { | ||
| die "$0: $self->{state}: Unknown state"; | ||
| } | ||
| } # A | ||
| 781 | ||
| 782 | die "$0: _get_next_token: unexpected case"; | delete $self->{parse_error}; # remove loop |
| } # _get_next_token | ||
| 783 | ||
| 784 | sub _tokenize_attempt_to_consume_an_entity ($) { | return $self->{document}; |
| 785 | my $self = shift; | } # parse_char_stream |
| 786 | ||
| 787 | if ({ | sub new ($) { |
| 788 | 0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF, | my $class = shift; |
| 789 | 0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR | my $self = bless { |
| 790 | }->{$self->{next_input_character}}) { | level => {must => 'm', |
| 791 | ## Don't consume | should => 's', |
| 792 | ## No error | warn => 'w', |
| 793 | return undef; | info => 'i', |
| 794 | } elsif ($self->{next_input_character} == 0x0023) { # # | uncertain => 'u'}, |
| 795 | !!!next-input-character; | }, $class; |
| 796 | if ($self->{next_input_character} == 0x0078 or # x | $self->{set_nc} = sub { |
| 797 | $self->{next_input_character} == 0x0058) { # X | $self->{nc} = -1; |
| 798 | my $num; | }; |
| 799 | X: { | $self->{parse_error} = sub { |
| 800 | my $x_char = $self->{next_input_character}; | # |
| 801 | !!!next-input-character; | }; |
| 802 | if (0x0030 <= $self->{next_input_character} and | $self->{change_encoding} = sub { |
| 803 | $self->{next_input_character} <= 0x0039) { # 0..9 | # if ($_[0] is a supported encoding) { |
| 804 | $num ||= 0; | # run "change the encoding" algorithm; |
| 805 | $num *= 0x10; | # throw Whatpm::HTML::RestartParser (charset => $new_encoding); |
| 806 | $num += $self->{next_input_character} - 0x0030; | # } |
| 807 | redo X; | }; |
| 808 | } elsif (0x0061 <= $self->{next_input_character} and | $self->{application_cache_selection} = sub { |
| 809 | $self->{next_input_character} <= 0x0066) { # a..f | # |
| 810 | ## ISSUE: the spec says U+0078, which is apparently incorrect | }; |
| 811 | $num ||= 0; | return $self; |
| 812 | $num *= 0x10; | } # new |
| $num += $self->{next_input_character} - 0x0060 + 9; | ||
| redo X; | ||
| } elsif (0x0041 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x0046) { # A..F | ||
| ## ISSUE: the spec says U+0058, which is apparently incorrect | ||
| $num ||= 0; | ||
| $num *= 0x10; | ||
| $num += $self->{next_input_character} - 0x0040 + 9; | ||
| redo X; | ||
| } elsif (not defined $num) { # no hexadecimal digit | ||
| !!!parse-error (type => 'bare hcro'); | ||
| $self->{next_input_character} = 0x0023; # # | ||
| !!!back-next-input-character ($x_char); | ||
| return undef; | ||
| } elsif ($self->{next_input_character} == 0x003B) { # ; | ||
| !!!next-input-character; | ||
| } else { | ||
| !!!parse-error (type => 'no refc'); | ||
| } | ||
| ## TODO: check the definition for |a valid Unicode character|. | ||
| ## <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2006-December/thread.html#8189> | ||
| if ($num > 1114111 or $num == 0) { | ||
| $num = 0xFFFD; # REPLACEMENT CHARACTER | ||
| ## ISSUE: Why this is not an error? | ||
| } elsif (0x80 <= $num and $num <= 0x9F) { | ||
| !!!parse-error (type => sprintf 'c1 entity:U+%04X', $num); | ||
| $num = $c1_entity_char->{$num}; | ||
| } | ||
| return {type => 'character', data => chr $num}; | ||
| } # X | ||
| } elsif (0x0030 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x0039) { # 0..9 | ||
| my $code = $self->{next_input_character} - 0x0030; | ||
| !!!next-input-character; | ||
| while (0x0030 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x0039) { # 0..9 | ||
| $code *= 10; | ||
| $code += $self->{next_input_character} - 0x0030; | ||
| !!!next-input-character; | ||
| } | ||
| 813 | ||
| 814 | if ($self->{next_input_character} == 0x003B) { # ; | ## Insertion modes |
| !!!next-input-character; | ||
| } else { | ||
| !!!parse-error (type => 'no refc'); | ||
| } | ||
| 815 | ||
| 816 | ## TODO: check the definition for |a valid Unicode character|. | sub AFTER_HTML_IMS () { 0b100 } |
| 817 | if ($code > 1114111 or $code == 0) { | sub HEAD_IMS () { 0b1000 } |
| 818 | $code = 0xFFFD; # REPLACEMENT CHARACTER | sub BODY_IMS () { 0b10000 } |
| 819 | ## ISSUE: Why this is not an error? | sub BODY_TABLE_IMS () { 0b100000 } |
| 820 | } elsif (0x80 <= $code and $code <= 0x9F) { | sub TABLE_IMS () { 0b1000000 } |
| 821 | !!!parse-error (type => sprintf 'c1 entity:U+%04X', $code); | sub ROW_IMS () { 0b10000000 } |
| 822 | $code = $c1_entity_char->{$code}; | sub BODY_AFTER_IMS () { 0b100000000 } |
| 823 | } | sub FRAME_IMS () { 0b1000000000 } |
| 824 | sub SELECT_IMS () { 0b10000000000 } | |
| 825 | return {type => 'character', data => chr $code}; | #sub IN_FOREIGN_CONTENT_IM () { 0b100000000000 } # see Whatpm::HTML::Tokenizer |
| 826 | } else { | ## NOTE: "in foreign content" insertion mode is special; it is combined |
| 827 | !!!parse-error (type => 'bare nero'); | ## with the secondary insertion mode. In this parser, they are stored |
| 828 | !!!back-next-input-character ($self->{next_input_character}); | ## together in the bit-or'ed form. |
| 829 | $self->{next_input_character} = 0x0023; # # | sub IN_CDATA_RCDATA_IM () { 0b1000000000000 } |
| 830 | return undef; | ## NOTE: "in CDATA/RCDATA" insertion mode is also special; it is |
| 831 | } | ## combined with the original insertion mode. In thie parser, |
| 832 | } elsif ((0x0041 <= $self->{next_input_character} and | ## they are stored together in the bit-or'ed form. |
| 833 | $self->{next_input_character} <= 0x005A) or | |
| 834 | (0x0061 <= $self->{next_input_character} and | sub IM_MASK () { 0b11111111111 } |
| 835 | $self->{next_input_character} <= 0x007A)) { | |
| 836 | my $entity_name = chr $self->{next_input_character}; | ## NOTE: "initial" and "before html" insertion modes have no constants. |
| 837 | !!!next-input-character; | |
| 838 | ## NOTE: "after after body" insertion mode. | |
| 839 | my $value = $entity_name; | sub AFTER_HTML_BODY_IM () { AFTER_HTML_IMS | BODY_AFTER_IMS } |
| 840 | my $match; | |
| 841 | require Whatpm::_NamedEntityList; | ## NOTE: "after after frameset" insertion mode. |
| 842 | our $EntityChar; | sub AFTER_HTML_FRAMESET_IM () { AFTER_HTML_IMS | FRAME_IMS } |
| 843 | ||
| 844 | while (length $entity_name < 10 and | sub IN_HEAD_IM () { HEAD_IMS | 0b00 } |
| 845 | ## NOTE: Some number greater than the maximum length of entity name | sub IN_HEAD_NOSCRIPT_IM () { HEAD_IMS | 0b01 } |
| 846 | ((0x0041 <= $self->{next_input_character} and # a | sub AFTER_HEAD_IM () { HEAD_IMS | 0b10 } |
| 847 | $self->{next_input_character} <= 0x005A) or # x | sub BEFORE_HEAD_IM () { HEAD_IMS | 0b11 } |
| 848 | (0x0061 <= $self->{next_input_character} and # a | sub IN_BODY_IM () { BODY_IMS } |
| 849 | $self->{next_input_character} <= 0x007A) or # z | sub IN_CELL_IM () { BODY_IMS | BODY_TABLE_IMS | 0b01 } |
| 850 | (0x0030 <= $self->{next_input_character} and # 0 | sub IN_CAPTION_IM () { BODY_IMS | BODY_TABLE_IMS | 0b10 } |
| 851 | $self->{next_input_character} <= 0x0039) or # 9 | sub IN_ROW_IM () { TABLE_IMS | ROW_IMS | 0b01 } |
| 852 | $self->{next_input_character} == 0x003B)) { # ; | sub IN_TABLE_BODY_IM () { TABLE_IMS | ROW_IMS | 0b10 } |
| 853 | $entity_name .= chr $self->{next_input_character}; | sub IN_TABLE_IM () { TABLE_IMS } |
| 854 | if (defined $EntityChar->{$entity_name}) { | sub AFTER_BODY_IM () { BODY_AFTER_IMS } |
| 855 | $value = $EntityChar->{$entity_name}; | sub IN_FRAMESET_IM () { FRAME_IMS | 0b01 } |
| 856 | if ($self->{next_input_character} == 0x003B) { # ; | sub AFTER_FRAMESET_IM () { FRAME_IMS | 0b10 } |
| 857 | $match = 1; | sub IN_SELECT_IM () { SELECT_IMS | 0b01 } |
| 858 | !!!next-input-character; | sub IN_SELECT_IN_TABLE_IM () { SELECT_IMS | 0b10 } |
| 859 | last; | sub IN_COLUMN_GROUP_IM () { 0b10 } |
| } else { | ||
| $match = -1; | ||
| } | ||
| } else { | ||
| $value .= chr $self->{next_input_character}; | ||
| } | ||
| !!!next-input-character; | ||
| } | ||
| if ($match > 0) { | ||
| return {type => 'character', data => $value}; | ||
| } elsif ($match < 0) { | ||
| !!!parse-error (type => 'refc'); | ||
| return {type => 'character', data => $value}; | ||
| } else { | ||
| !!!parse-error (type => 'bare ero'); | ||
| ## NOTE: No characters are consumed in the spec. | ||
| !!!back-token ({type => 'character', data => $value}); | ||
| return undef; | ||
| } | ||
| } else { | ||
| ## no characters are consumed | ||
| !!!parse-error (type => 'bare ero'); | ||
| return undef; | ||
| } | ||
| } # _tokenize_attempt_to_consume_an_entity | ||
| 860 | ||
| 861 | sub _initialize_tree_constructor ($) { | sub _initialize_tree_constructor ($) { |
| 862 | my $self = shift; | my $self = shift; |
| # | Line 1728 sub _initialize_tree_constructor ($) { | Line 865 sub _initialize_tree_constructor ($) { |
| 865 | ## TODO: Turn mutation events off # MUST | ## TODO: Turn mutation events off # MUST |
| 866 | ## TODO: Turn loose Document option (manakai extension) on | ## TODO: Turn loose Document option (manakai extension) on |
| 867 | $self->{document}->manakai_is_html (1); # MUST | $self->{document}->manakai_is_html (1); # MUST |
| 868 | $self->{document}->set_user_data (manakai_source_line => 1); | |
| 869 | $self->{document}->set_user_data (manakai_source_column => 1); | |
| 870 | } # _initialize_tree_constructor | } # _initialize_tree_constructor |
| 871 | ||
| 872 | sub _terminate_tree_constructor ($) { | sub _terminate_tree_constructor ($) { |
| # | Line 1747 sub _construct_tree ($) { | Line 886 sub _construct_tree ($) { |
| 886 | ## When an interactive UA render the $self->{document} available | ## When an interactive UA render the $self->{document} available |
| 887 | ## to the user, or when it begin accepting user input, are | ## to the user, or when it begin accepting user input, are |
| 888 | ## not defined. | ## not defined. |
| ## Append a character: collect it and all subsequent consecutive | ||
| ## characters and insert one Text node whose data is concatenation | ||
| ## of all those characters. # MUST | ||
| 889 | ||
| 890 | !!!next-token; | !!!next-token; |
| 891 | ||
| $self->{insertion_mode} = 'before head'; | ||
| 892 | undef $self->{form_element}; | undef $self->{form_element}; |
| 893 | undef $self->{head_element}; | undef $self->{head_element}; |
| 894 | undef $self->{head_element_inserted}; | |
| 895 | $self->{open_elements} = []; | $self->{open_elements} = []; |
| 896 | undef $self->{inner_html_node}; | undef $self->{inner_html_node}; |
| 897 | undef $self->{ignore_newline}; | |
| 898 | ||
| 899 | ## NOTE: The "initial" insertion mode. | |
| 900 | $self->_tree_construction_initial; # MUST | $self->_tree_construction_initial; # MUST |
| 901 | ||
| 902 | ## NOTE: The "before html" insertion mode. | |
| 903 | $self->_tree_construction_root_element; | $self->_tree_construction_root_element; |
| 904 | $self->{insertion_mode} = BEFORE_HEAD_IM; | |
| 905 | ||
| 906 | ## NOTE: The "before head" insertion mode and so on. | |
| 907 | $self->_tree_construction_main; | $self->_tree_construction_main; |
| 908 | } # _construct_tree | } # _construct_tree |
| 909 | ||
| 910 | sub _tree_construction_initial ($) { | sub _tree_construction_initial ($) { |
| 911 | my $self = shift; | my $self = shift; |
| 912 | ||
| 913 | ## NOTE: "initial" insertion mode | |
| 914 | ||
| 915 | INITIAL: { | INITIAL: { |
| 916 | if ($token->{type} eq 'DOCTYPE') { | if ($token->{type} == DOCTYPE_TOKEN) { |
| 917 | ## NOTE: Conformance checkers MAY, instead of reporting "not HTML5" | ## NOTE: Conformance checkers MAY, instead of reporting "not |
| 918 | ## error, switch to a conformance checking mode for another | ## HTML5" error, switch to a conformance checking mode for |
| 919 | ## language. | ## another language. (We don't support such mode switchings; it |
| 920 | ## is nonsense to do anything different from what browsers do.) | |
| 921 | my $doctype_name = $token->{name}; | my $doctype_name = $token->{name}; |
| 922 | $doctype_name = '' unless defined $doctype_name; | $doctype_name = '' unless defined $doctype_name; |
| 923 | $doctype_name =~ tr/a-z/A-Z/; | my $doctype = $self->{document}->create_document_type_definition |
| 924 | if (not defined $token->{name} or # <!DOCTYPE> | ($doctype_name); |
| 925 | defined $token->{public_identifier} or | |
| 926 | defined $token->{system_identifier}) { | $doctype_name =~ tr/A-Z/a-z/; # ASCII case-insensitive |
| 927 | !!!parse-error (type => 'not HTML5'); | if ($doctype_name ne 'html') { |
| 928 | } elsif ($doctype_name ne 'HTML') { | !!!cp ('t1'); |
| 929 | ## ISSUE: ASCII case-insensitive? (in fact it does not matter) | !!!parse-error (type => 'not HTML5', token => $token); |
| 930 | !!!parse-error (type => 'not HTML5'); | } elsif (defined $token->{pubid}) { |
| 931 | !!!cp ('t2'); | |
| 932 | ## XXX Obsolete permitted DOCTYPEs | |
| 933 | !!!parse-error (type => 'not HTML5', token => $token); | |
| 934 | } elsif (defined $token->{sysid}) { | |
| 935 | if ($token->{sysid} eq 'about:legacy-compat') { | |
| 936 | !!!cp ('t1.2'); ## <!DOCTYPE HTML SYSTEM "about:legacy-compat"> | |
| 937 | !!!parse-error (type => 'XSLT-compat', token => $token, | |
| 938 | level => $self->{level}->{should}); | |
| 939 | } else { | |
| 940 | !!!parse-error (type => 'not HTML5', token => $token); | |
| 941 | } | |
| 942 | } else { ## <!DOCTYPE HTML> | |
| 943 | !!!cp ('t3'); | |
| 944 | # | |
| 945 | } | } |
| 946 | ||
| 947 | my $doctype = $self->{document}->create_document_type_definition | ## NOTE: Default value for both |public_id| and |system_id| attributes |
| 948 | ($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)? | ## are empty strings, so that we don't set any value in missing cases. |
| 949 | $doctype->public_id ($token->{public_identifier}) | $doctype->public_id ($token->{pubid}) if defined $token->{pubid}; |
| 950 | if defined $token->{public_identifier}; | $doctype->system_id ($token->{sysid}) if defined $token->{sysid}; |
| 951 | $doctype->system_id ($token->{system_identifier}) | |
| if defined $token->{system_identifier}; | ||
| 952 | ## NOTE: Other DocumentType attributes are null or empty lists. | ## NOTE: Other DocumentType attributes are null or empty lists. |
| 953 | ## ISSUE: internalSubset = null?? | ## In Firefox3, |internalSubset| attribute is set to the empty |
| 954 | ## string, while |null| is an allowed value for the attribute | |
| 955 | ## according to DOM3 Core. | |
| 956 | $self->{document}->append_child ($doctype); | $self->{document}->append_child ($doctype); |
| 957 | ||
| 958 | if (not $token->{correct} or $doctype_name ne 'HTML') { | if ($token->{quirks} or $doctype_name ne 'html') { |
| 959 | !!!cp ('t4'); | |
| 960 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 961 | } elsif (defined $token->{public_identifier}) { | } elsif (defined $token->{pubid}) { |
| 962 | my $pubid = $token->{public_identifier}; | my $pubid = $token->{pubid}; |
| 963 | $pubid =~ tr/a-z/A-z/; | $pubid =~ tr/a-z/A-z/; |
| 964 | if ({ | my $prefix = [ |
| 965 | "+//SILMARIL//DTD HTML PRO V0R11 19970101//EN" => 1, | "+//SILMARIL//DTD HTML PRO V0R11 19970101//", |
| 966 | "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1, | "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//", |
| 967 | "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1, | "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//", |
| 968 | "-//IETF//DTD HTML 2.0 LEVEL 1//EN" => 1, | "-//IETF//DTD HTML 2.0 LEVEL 1//", |
| 969 | "-//IETF//DTD HTML 2.0 LEVEL 2//EN" => 1, | "-//IETF//DTD HTML 2.0 LEVEL 2//", |
| 970 | "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//EN" => 1, | "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//", |
| 971 | "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//EN" => 1, | "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//", |
| 972 | "-//IETF//DTD HTML 2.0 STRICT//EN" => 1, | "-//IETF//DTD HTML 2.0 STRICT//", |
| 973 | "-//IETF//DTD HTML 2.0//EN" => 1, | "-//IETF//DTD HTML 2.0//", |
| 974 | "-//IETF//DTD HTML 2.1E//EN" => 1, | "-//IETF//DTD HTML 2.1E//", |
| 975 | "-//IETF//DTD HTML 3.0//EN" => 1, | "-//IETF//DTD HTML 3.0//", |
| 976 | "-//IETF//DTD HTML 3.0//EN//" => 1, | "-//IETF//DTD HTML 3.2 FINAL//", |
| 977 | "-//IETF//DTD HTML 3.2 FINAL//EN" => 1, | "-//IETF//DTD HTML 3.2//", |
| 978 | "-//IETF//DTD HTML 3.2//EN" => 1, | "-//IETF//DTD HTML 3//", |
| 979 | "-//IETF//DTD HTML 3//EN" => 1, | "-//IETF//DTD HTML LEVEL 0//", |
| 980 | "-//IETF//DTD HTML LEVEL 0//EN" => 1, | "-//IETF//DTD HTML LEVEL 1//", |
| 981 | "-//IETF//DTD HTML LEVEL 0//EN//2.0" => 1, | "-//IETF//DTD HTML LEVEL 2//", |
| 982 | "-//IETF//DTD HTML LEVEL 1//EN" => 1, | "-//IETF//DTD HTML LEVEL 3//", |
| 983 | "-//IETF//DTD HTML LEVEL 1//EN//2.0" => 1, | "-//IETF//DTD HTML STRICT LEVEL 0//", |
| 984 | "-//IETF//DTD HTML LEVEL 2//EN" => 1, | "-//IETF//DTD HTML STRICT LEVEL 1//", |
| 985 | "-//IETF//DTD HTML LEVEL 2//EN//2.0" => 1, | "-//IETF//DTD HTML STRICT LEVEL 2//", |
| 986 | "-//IETF//DTD HTML LEVEL 3//EN" => 1, | "-//IETF//DTD HTML STRICT LEVEL 3//", |
| 987 | "-//IETF//DTD HTML LEVEL 3//EN//3.0" => 1, | "-//IETF//DTD HTML STRICT//", |
| 988 | "-//IETF//DTD HTML STRICT LEVEL 0//EN" => 1, | "-//IETF//DTD HTML//", |
| 989 | "-//IETF//DTD HTML STRICT LEVEL 0//EN//2.0" => 1, | "-//METRIUS//DTD METRIUS PRESENTATIONAL//", |
| 990 | "-//IETF//DTD HTML STRICT LEVEL 1//EN" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//", |
| 991 | "-//IETF//DTD HTML STRICT LEVEL 1//EN//2.0" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//", |
| 992 | "-//IETF//DTD HTML STRICT LEVEL 2//EN" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//", |
| 993 | "-//IETF//DTD HTML STRICT LEVEL 2//EN//2.0" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//", |
| 994 | "-//IETF//DTD HTML STRICT LEVEL 3//EN" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//", |
| 995 | "-//IETF//DTD HTML STRICT LEVEL 3//EN//3.0" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//", |
| 996 | "-//IETF//DTD HTML STRICT//EN" => 1, | "-//NETSCAPE COMM. CORP.//DTD HTML//", |
| 997 | "-//IETF//DTD HTML STRICT//EN//2.0" => 1, | "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//", |
| 998 | "-//IETF//DTD HTML STRICT//EN//3.0" => 1, | "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//", |
| 999 | "-//IETF//DTD HTML//EN" => 1, | "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//", |
| 1000 | "-//IETF//DTD HTML//EN//2.0" => 1, | "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//", |
| 1001 | "-//IETF//DTD HTML//EN//3.0" => 1, | "-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//", |
| 1002 | "-//METRIUS//DTD METRIUS PRESENTATIONAL//EN" => 1, | "-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//", |
| 1003 | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//EN" => 1, | "-//SPYGLASS//DTD HTML 2.0 EXTENDED//", |
| 1004 | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//EN" => 1, | "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//", |
| 1005 | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//EN" => 1, | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//", |
| 1006 | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//EN" => 1, | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//", |
| 1007 | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//EN" => 1, | "-//W3C//DTD HTML 3 1995-03-24//", |
| 1008 | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//EN" => 1, | "-//W3C//DTD HTML 3.2 DRAFT//", |
| 1009 | "-//NETSCAPE COMM. CORP.//DTD HTML//EN" => 1, | "-//W3C//DTD HTML 3.2 FINAL//", |
| 1010 | "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1, | "-//W3C//DTD HTML 3.2//", |
| 1011 | "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1, | "-//W3C//DTD HTML 3.2S DRAFT//", |
| 1012 | "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1, | "-//W3C//DTD HTML 4.0 FRAMESET//", |
| 1013 | "-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1, | "-//W3C//DTD HTML 4.0 TRANSITIONAL//", |
| 1014 | "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1, | "-//W3C//DTD HTML EXPERIMETNAL 19960712//", |
| 1015 | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1, | "-//W3C//DTD HTML EXPERIMENTAL 970421//", |
| 1016 | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//EN" => 1, | "-//W3C//DTD W3 HTML//", |
| 1017 | "-//W3C//DTD HTML 3 1995-03-24//EN" => 1, | "-//W3O//DTD W3 HTML 3.0//", |
| 1018 | "-//W3C//DTD HTML 3.2 DRAFT//EN" => 1, | "-//WEBTECHS//DTD MOZILLA HTML 2.0//", |
| 1019 | "-//W3C//DTD HTML 3.2 FINAL//EN" => 1, | "-//WEBTECHS//DTD MOZILLA HTML//", |
| 1020 | "-//W3C//DTD HTML 3.2//EN" => 1, | ]; # $prefix |
| 1021 | "-//W3C//DTD HTML 3.2S DRAFT//EN" => 1, | my $match; |
| 1022 | "-//W3C//DTD HTML 4.0 FRAMESET//EN" => 1, | for (@$prefix) { |
| 1023 | "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN" => 1, | if (substr ($prefix, 0, length $_) eq $_) { |
| 1024 | "-//W3C//DTD HTML EXPERIMETNAL 19960712//EN" => 1, | $match = 1; |
| 1025 | "-//W3C//DTD HTML EXPERIMENTAL 970421//EN" => 1, | last; |
| 1026 | "-//W3C//DTD W3 HTML//EN" => 1, | } |
| 1027 | "-//W3O//DTD W3 HTML 3.0//EN" => 1, | } |
| 1028 | "-//W3O//DTD W3 HTML 3.0//EN//" => 1, | if ($match or |
| 1029 | "-//W3O//DTD W3 HTML STRICT 3.0//EN//" => 1, | $pubid eq "-//W3O//DTD W3 HTML STRICT 3.0//EN//" or |
| 1030 | "-//WEBTECHS//DTD MOZILLA HTML 2.0//EN" => 1, | $pubid eq "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" or |
| 1031 | "-//WEBTECHS//DTD MOZILLA HTML//EN" => 1, | $pubid eq "HTML") { |
| 1032 | "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" => 1, | !!!cp ('t5'); |
| "HTML" => 1, | ||
| }->{$pubid}) { | ||
| 1033 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1034 | } elsif ($pubid eq "-//W3C//DTD HTML 4.01 FRAMESET//EN" or | } elsif ($pubid =~ m[^-//W3C//DTD HTML 4.01 FRAMESET//] or |
| 1035 | $pubid eq "-//W3C//DTD HTML 4.01 TRANSITIONAL//EN") { | $pubid =~ m[^-//W3C//DTD HTML 4.01 TRANSITIONAL//]) { |
| 1036 | if (defined $token->{system_identifier}) { | if (defined $token->{sysid}) { |
| 1037 | !!!cp ('t6'); | |
| 1038 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1039 | } else { | } else { |
| 1040 | !!!cp ('t7'); | |
| 1041 | $self->{document}->manakai_compat_mode ('limited quirks'); | $self->{document}->manakai_compat_mode ('limited quirks'); |
| 1042 | } | } |
| 1043 | } elsif ($pubid eq "-//W3C//DTD XHTML 1.0 Frameset//EN" or | } elsif ($pubid =~ m[^-//W3C//DTD XHTML 1.0 FRAMESET//] or |
| 1044 | $pubid eq "-//W3C//DTD XHTML 1.0 Transitional//EN") { | $pubid =~ m[^-//W3C//DTD XHTML 1.0 TRANSITIONAL//]) { |
| 1045 | !!!cp ('t8'); | |
| 1046 | $self->{document}->manakai_compat_mode ('limited quirks'); | $self->{document}->manakai_compat_mode ('limited quirks'); |
| 1047 | } else { | |
| 1048 | !!!cp ('t9'); | |
| 1049 | } | } |
| 1050 | } else { | |
| 1051 | !!!cp ('t10'); | |
| 1052 | } | } |
| 1053 | if (defined $token->{system_identifier}) { | if (defined $token->{sysid}) { |
| 1054 | my $sysid = $token->{system_identifier}; | my $sysid = $token->{sysid}; |
| 1055 | $sysid =~ tr/A-Z/a-z/; | $sysid =~ tr/A-Z/a-z/; |
| 1056 | if ($sysid eq "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd") { | if ($sysid eq "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd") { |
| 1057 | ## NOTE: Ensure that |PUBLIC "(limited quirks)" "(quirks)"| is | |
| 1058 | ## marked as quirks. | |
| 1059 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1060 | !!!cp ('t11'); | |
| 1061 | } else { | |
| 1062 | !!!cp ('t12'); | |
| 1063 | } | } |
| 1064 | } else { | |
| 1065 | !!!cp ('t13'); | |
| 1066 | } | } |
| 1067 | ||
| 1068 | ## Go to the root element phase. | ## Go to the "before html" insertion mode. |
| 1069 | !!!next-token; | !!!next-token; |
| 1070 | return; | return; |
| 1071 | } elsif ({ | } elsif ({ |
| 1072 | 'start tag' => 1, | START_TAG_TOKEN, 1, |
| 1073 | 'end tag' => 1, | END_TAG_TOKEN, 1, |
| 1074 | 'end-of-file' => 1, | END_OF_FILE_TOKEN, 1, |
| 1075 | }->{$token->{type}}) { | }->{$token->{type}}) { |
| 1076 | !!!parse-error (type => 'no DOCTYPE'); | !!!cp ('t14'); |
| 1077 | !!!parse-error (type => 'no DOCTYPE', token => $token); | |
| 1078 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1079 | ## Go to the root element phase | ## Go to the "before html" insertion mode. |
| 1080 | ## reprocess | ## reprocess |
| 1081 | !!!ack-later; | |
| 1082 | return; | return; |
| 1083 | } elsif ($token->{type} eq 'character') { | } elsif ($token->{type} == CHARACTER_TOKEN) { |
| 1084 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { # \x0D | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1085 | ## Ignore the token | ## Ignore the token |
| 1086 | ||
| 1087 | unless (length $token->{data}) { | unless (length $token->{data}) { |
| 1088 | ## Stay in the phase | !!!cp ('t15'); |
| 1089 | ## Stay in the insertion mode. | |
| 1090 | !!!next-token; | !!!next-token; |
| 1091 | redo INITIAL; | redo INITIAL; |
| 1092 | } else { | |
| 1093 | !!!cp ('t16'); | |
| 1094 | } | } |
| 1095 | } else { | |
| 1096 | !!!cp ('t17'); | |
| 1097 | } | } |
| 1098 | ||
| 1099 | !!!parse-error (type => 'no DOCTYPE'); | !!!parse-error (type => 'no DOCTYPE', token => $token); |
| 1100 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1101 | ## Go to the root element phase | ## Go to the "before html" insertion mode. |
| 1102 | ## reprocess | ## reprocess |
| 1103 | return; | return; |
| 1104 | } elsif ($token->{type} eq 'comment') { | } elsif ($token->{type} == COMMENT_TOKEN) { |
| 1105 | !!!cp ('t18'); | |
| 1106 | my $comment = $self->{document}->create_comment ($token->{data}); | my $comment = $self->{document}->create_comment ($token->{data}); |
| 1107 | $self->{document}->append_child ($comment); | $self->{document}->append_child ($comment); |
| 1108 | ||
| 1109 | ## Stay in the phase. | ## Stay in the insertion mode. |
| 1110 | !!!next-token; | !!!next-token; |
| 1111 | redo INITIAL; | redo INITIAL; |
| 1112 | } else { | } else { |
| 1113 | die "$0: $token->{type}: Unknown token"; | die "$0: $token->{type}: Unknown token type"; |
| 1114 | } | } |
| 1115 | } # INITIAL | } # INITIAL |
| 1116 | ||
| 1117 | die "$0: _tree_construction_initial: This should be never reached"; | |
| 1118 | } # _tree_construction_initial | } # _tree_construction_initial |
| 1119 | ||
| 1120 | sub _tree_construction_root_element ($) { | sub _tree_construction_root_element ($) { |
| 1121 | my $self = shift; | my $self = shift; |
| 1122 | ||
| 1123 | ## NOTE: "before html" insertion mode. | |
| 1124 | ||
| 1125 | B: { | B: { |
| 1126 | if ($token->{type} eq 'DOCTYPE') { | if ($token->{type} == DOCTYPE_TOKEN) { |
| 1127 | !!!parse-error (type => 'in html:#DOCTYPE'); | !!!cp ('t19'); |
| 1128 | !!!parse-error (type => 'in html:#DOCTYPE', token => $token); | |
| 1129 | ## Ignore the token | ## Ignore the token |
| 1130 | ## Stay in the phase | ## Stay in the insertion mode. |
| 1131 | !!!next-token; | !!!next-token; |
| 1132 | redo B; | redo B; |
| 1133 | } elsif ($token->{type} eq 'comment') { | } elsif ($token->{type} == COMMENT_TOKEN) { |
| 1134 | !!!cp ('t20'); | |
| 1135 | my $comment = $self->{document}->create_comment ($token->{data}); | my $comment = $self->{document}->create_comment ($token->{data}); |
| 1136 | $self->{document}->append_child ($comment); | $self->{document}->append_child ($comment); |
| 1137 | ## Stay in the phase | ## Stay in the insertion mode. |
| 1138 | !!!next-token; | !!!next-token; |
| 1139 | redo B; | redo B; |
| 1140 | } elsif ($token->{type} eq 'character') { | } elsif ($token->{type} == CHARACTER_TOKEN) { |
| 1141 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1142 | $self->{document}->manakai_append_text ($1); | ## Ignore the token. |
| 1143 | ## ISSUE: DOM3 Core does not allow Document > Text | |
| 1144 | unless (length $token->{data}) { | unless (length $token->{data}) { |
| 1145 | ## Stay in the phase | !!!cp ('t21'); |
| 1146 | ## Stay in the insertion mode. | |
| 1147 | !!!next-token; | !!!next-token; |
| 1148 | redo B; | redo B; |
| 1149 | } else { | |
| 1150 | !!!cp ('t22'); | |
| 1151 | } | } |
| 1152 | } else { | |
| 1153 | !!!cp ('t23'); | |
| 1154 | } | } |
| 1155 | ||
| 1156 | $self->{application_cache_selection}->(undef); | |
| 1157 | ||
| 1158 | # | # |
| 1159 | } elsif ($token->{type} == START_TAG_TOKEN) { | |
| 1160 | if ($token->{tag_name} eq 'html') { | |
| 1161 | my $root_element; | |
| 1162 | !!!create-element ($root_element, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); | |
| 1163 | $self->{document}->append_child ($root_element); | |
| 1164 | push @{$self->{open_elements}}, | |
| 1165 | [$root_element, $el_category->{html}]; | |
| 1166 | ||
| 1167 | if ($token->{attributes}->{manifest}) { | |
| 1168 | !!!cp ('t24'); | |
| 1169 | $self->{application_cache_selection} | |
| 1170 | ->($token->{attributes}->{manifest}->{value}); | |
| 1171 | ## ISSUE: Spec is unclear on relative references. | |
| 1172 | ## According to Hixie (#whatwg 2008-03-19), it should be | |
| 1173 | ## resolved against the base URI of the document in HTML | |
| 1174 | ## or xml:base of the element in XHTML. | |
| 1175 | } else { | |
| 1176 | !!!cp ('t25'); | |
| 1177 | $self->{application_cache_selection}->(undef); | |
| 1178 | } | |
| 1179 | ||
| 1180 | !!!nack ('t25c'); | |
| 1181 | ||
| 1182 | !!!next-token; | |
| 1183 | return; ## Go to the "before head" insertion mode. | |
| 1184 | } else { | |
| 1185 | !!!cp ('t25.1'); | |
| 1186 | # | |
| 1187 | } | |
| 1188 | } elsif ({ | } elsif ({ |
| 1189 | 'start tag' => 1, | END_TAG_TOKEN, 1, |
| 1190 | 'end tag' => 1, | END_OF_FILE_TOKEN, 1, |
| 'end-of-file' => 1, | ||
| 1191 | }->{$token->{type}}) { | }->{$token->{type}}) { |
| 1192 | ## ISSUE: There is an issue in the spec | !!!cp ('t26'); |
| 1193 | # | # |
| 1194 | } else { | } else { |
| 1195 | die "$0: $token->{type}: Unknown token"; | die "$0: $token->{type}: Unknown token type"; |
| 1196 | } | } |
| 1197 | my $root_element; !!!create-element ($root_element, 'html'); | |
| 1198 | $self->{document}->append_child ($root_element); | my $root_element; |
| 1199 | push @{$self->{open_elements}}, [$root_element, 'html']; | !!!create-element ($root_element, $HTML_NS, 'html',, $token); |
| 1200 | #$phase = 'main'; | $self->{document}->append_child ($root_element); |
| 1201 | ## reprocess | push @{$self->{open_elements}}, [$root_element, $el_category->{html}]; |
| 1202 | #redo B; | |
| 1203 | return; | $self->{application_cache_selection}->(undef); |
| 1204 | ||
| 1205 | ## NOTE: Reprocess the token. | |
| 1206 | !!!ack-later; | |
| 1207 | return; ## Go to the "before head" insertion mode. | |
| 1208 | } # B | } # B |
| 1209 | ||
| 1210 | die "$0: _tree_construction_root_element: This should never be reached"; | |
| 1211 | } # _tree_construction_root_element | } # _tree_construction_root_element |
| 1212 | ||
| 1213 | sub _reset_insertion_mode ($) { | sub _reset_insertion_mode ($) { |
| # | Line 1991 sub _reset_insertion_mode ($) { | Line 1222 sub _reset_insertion_mode ($) { |
| 1222 | ||
| 1223 | ## Step 3 | ## Step 3 |
| 1224 | S3: { | S3: { |
| 1225 | $last = 1 if $self->{open_elements}->[0]->[0] eq $node->[0]; | if ($self->{open_elements}->[0]->[0] eq $node->[0]) { |
| 1226 | if (defined $self->{inner_html_node}) { | $last = 1; |
| 1227 | if ($self->{inner_html_node}->[1] eq 'td' or | if (defined $self->{inner_html_node}) { |
| 1228 | $self->{inner_html_node}->[1] eq 'th') { | !!!cp ('t28'); |
| 1229 | $node = $self->{inner_html_node}; | |
| 1230 | } else { | |
| 1231 | die "_reset_insertion_mode: t27"; | |
| 1232 | } | |
| 1233 | } | |
| 1234 | ||
| 1235 | ## Step 4..14 | |
| 1236 | my $new_mode; | |
| 1237 | if ($node->[1] & FOREIGN_EL) { | |
| 1238 | !!!cp ('t28.1'); | |
| 1239 | ## NOTE: Strictly spaking, the line below only applies to MathML and | |
| 1240 | ## SVG elements. Currently the HTML syntax supports only MathML and | |
| 1241 | ## SVG elements as foreigners. | |
| 1242 | $new_mode = IN_BODY_IM | IN_FOREIGN_CONTENT_IM; | |
| 1243 | } elsif ($node->[1] == TABLE_CELL_EL) { | |
| 1244 | if ($last) { | |
| 1245 | !!!cp ('t28.2'); | |
| 1246 | # | # |
| 1247 | } else { | } else { |
| 1248 | $node = $self->{inner_html_node}; | !!!cp ('t28.3'); |
| 1249 | $new_mode = IN_CELL_IM; | |
| 1250 | } | } |
| 1251 | } else { | |
| 1252 | !!!cp ('t28.4'); | |
| 1253 | $new_mode = { | |
| 1254 | select => IN_SELECT_IM, | |
| 1255 | ## NOTE: |option| and |optgroup| do not set | |
| 1256 | ## insertion mode to "in select" by themselves. | |
| 1257 | tr => IN_ROW_IM, | |
| 1258 | tbody => IN_TABLE_BODY_IM, | |
| 1259 | thead => IN_TABLE_BODY_IM, | |
| 1260 | tfoot => IN_TABLE_BODY_IM, | |
| 1261 | caption => IN_CAPTION_IM, | |
| 1262 | colgroup => IN_COLUMN_GROUP_IM, | |
| 1263 | table => IN_TABLE_IM, | |
| 1264 | head => IN_BODY_IM, # not in head! | |
| 1265 | body => IN_BODY_IM, | |
| 1266 | frameset => IN_FRAMESET_IM, | |
| 1267 | }->{$node->[0]->manakai_local_name}; | |
| 1268 | } | } |
| ## Step 4..13 | ||
| my $new_mode = { | ||
| select => 'in select', | ||
| td => 'in cell', | ||
| th => 'in cell', | ||
| tr => 'in row', | ||
| tbody => 'in table body', | ||
| thead => 'in table head', | ||
| tfoot => 'in table foot', | ||
| caption => 'in caption', | ||
| colgroup => 'in column group', | ||
| table => 'in table', | ||
| head => 'in body', # not in head! | ||
| body => 'in body', | ||
| frameset => 'in frameset', | ||
| }->{$node->[1]}; | ||
| 1269 | $self->{insertion_mode} = $new_mode and return if defined $new_mode; | $self->{insertion_mode} = $new_mode and return if defined $new_mode; |
| 1270 | ||
| 1271 | ## Step 14 | ## Step 15 |
| 1272 | if ($node->[1] eq 'html') { | if ($node->[1] == HTML_EL) { |
| 1273 | unless (defined $self->{head_element}) { | unless (defined $self->{head_element}) { |
| 1274 | $self->{insertion_mode} = 'before head'; | !!!cp ('t29'); |
| 1275 | $self->{insertion_mode} = BEFORE_HEAD_IM; | |
| 1276 | } else { | } else { |
| 1277 | $self->{insertion_mode} = 'after head'; | ## ISSUE: Can this state be reached? |
| 1278 | !!!cp ('t30'); | |
| 1279 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 1280 | } | } |
| 1281 | return; | return; |
| 1282 | } else { | |
| 1283 | !!!cp ('t31'); | |
| 1284 | } | } |
| 1285 | ||
| ## Step 15 | ||
| $self->{insertion_mode} = 'in body' and return if $last; | ||
| 1286 | ## Step 16 | ## Step 16 |
| 1287 | $self->{insertion_mode} = IN_BODY_IM and return if $last; | |
| 1288 | ||
| 1289 | ## Step 17 | |
| 1290 | $i--; | $i--; |
| 1291 | $node = $self->{open_elements}->[$i]; | $node = $self->{open_elements}->[$i]; |
| 1292 | ||
| 1293 | ## Step 17 | ## Step 18 |
| 1294 | redo S3; | redo S3; |
| 1295 | } # S3 | } # S3 |
| 1296 | ||
| 1297 | die "$0: _reset_insertion_mode: This line should never be reached"; | |
| 1298 | } # _reset_insertion_mode | } # _reset_insertion_mode |
| 1299 | ||
| 1300 | sub _tree_construction_main ($) { | sub _tree_construction_main ($) { |
| 1301 | my $self = shift; | my $self = shift; |
| 1302 | ||
| my $phase = 'main'; | ||
| 1303 | my $active_formatting_elements = []; | my $active_formatting_elements = []; |
| 1304 | ||
| 1305 | my $reconstruct_active_formatting_elements = sub { # MUST | my $reconstruct_active_formatting_elements = sub { # MUST |
| # | Line 2062 sub _tree_construction_main ($) { | Line 1316 sub _tree_construction_main ($) { |
| 1316 | return if $entry->[0] eq '#marker'; | return if $entry->[0] eq '#marker'; |
| 1317 | for (@{$self->{open_elements}}) { | for (@{$self->{open_elements}}) { |
| 1318 | if ($entry->[0] eq $_->[0]) { | if ($entry->[0] eq $_->[0]) { |
| 1319 | !!!cp ('t32'); | |
| 1320 | return; | return; |
| 1321 | } | } |
| 1322 | } | } |
| # | Line 2076 sub _tree_construction_main ($) { | Line 1331 sub _tree_construction_main ($) { |
| 1331 | ||
| 1332 | ## Step 6 | ## Step 6 |
| 1333 | if ($entry->[0] eq '#marker') { | if ($entry->[0] eq '#marker') { |
| 1334 | !!!cp ('t33_1'); | |
| 1335 | # | # |
| 1336 | } else { | } else { |
| 1337 | my $in_open_elements; | my $in_open_elements; |
| 1338 | OE: for (@{$self->{open_elements}}) { | OE: for (@{$self->{open_elements}}) { |
| 1339 | if ($entry->[0] eq $_->[0]) { | if ($entry->[0] eq $_->[0]) { |
| 1340 | !!!cp ('t33'); | |
| 1341 | $in_open_elements = 1; | $in_open_elements = 1; |
| 1342 | last OE; | last OE; |
| 1343 | } | } |
| 1344 | } | } |
| 1345 | if ($in_open_elements) { | if ($in_open_elements) { |
| 1346 | !!!cp ('t34'); | |
| 1347 | # | # |
| 1348 | } else { | } else { |
| 1349 | ## NOTE: <!DOCTYPE HTML><p><b><i><u></p> <p>X | |
| 1350 | !!!cp ('t35'); | |
| 1351 | redo S4; | redo S4; |
| 1352 | } | } |
| 1353 | } | } |
| # | Line 2110 sub _tree_construction_main ($) { | Line 1370 sub _tree_construction_main ($) { |
| 1370 | ||
| 1371 | ## Step 11 | ## Step 11 |
| 1372 | unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { | unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { |
| 1373 | !!!cp ('t36'); | |
| 1374 | ## Step 7' | ## Step 7' |
| 1375 | $i++; | $i++; |
| 1376 | $entry = $active_formatting_elements->[$i]; | $entry = $active_formatting_elements->[$i]; |
| 1377 | ||
| 1378 | redo S7; | redo S7; |
| 1379 | } | } |
| 1380 | ||
| 1381 | !!!cp ('t37'); | |
| 1382 | } # S7 | } # S7 |
| 1383 | }; # $reconstruct_active_formatting_elements | }; # $reconstruct_active_formatting_elements |
| 1384 | ||
| 1385 | my $clear_up_to_marker = sub { | my $clear_up_to_marker = sub { |
| 1386 | for (reverse 0..$#$active_formatting_elements) { | for (reverse 0..$#$active_formatting_elements) { |
| 1387 | if ($active_formatting_elements->[$_]->[0] eq '#marker') { | if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1388 | !!!cp ('t38'); | |
| 1389 | splice @$active_formatting_elements, $_; | splice @$active_formatting_elements, $_; |
| 1390 | return; | return; |
| 1391 | } | } |
| 1392 | } | } |
| 1393 | ||
| 1394 | !!!cp ('t39'); | |
| 1395 | }; # $clear_up_to_marker | }; # $clear_up_to_marker |
| 1396 | ||
| 1397 | my $style_start_tag = sub { | my $insert; |
| 1398 | my $style_el; !!!create-element ($style_el, 'style', $token->{attributes}); | |
| 1399 | ## $self->{insertion_mode} eq 'in head' and ... (always true) | my $parse_rcdata = sub ($) { |
| 1400 | (($self->{insertion_mode} eq 'in head' and defined $self->{head_element}) | my ($content_model_flag) = @_; |
| 1401 | ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | |
| 1402 | ->append_child ($style_el); | ## Step 1 |
| 1403 | $self->{content_model_flag} = 'CDATA'; | my $start_tag_name = $token->{tag_name}; |
| 1404 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); | |
| 1405 | ||
| 1406 | ## Step 2 | |
| 1407 | $self->{content_model} = $content_model_flag; # CDATA or RCDATA | |
| 1408 | delete $self->{escape}; # MUST | delete $self->{escape}; # MUST |
| 1409 | ||
| 1410 | my $text = ''; | ## Step 3, 4 |
| 1411 | !!!next-token; | $self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
| 1412 | while ($token->{type} eq 'character') { | |
| 1413 | $text .= $token->{data}; | !!!nack ('t40.1'); |
| !!!next-token; | ||
| } # stop if non-character token or tokenizer stops tokenising | ||
| if (length $text) { | ||
| $style_el->manakai_append_text ($text); | ||
| } | ||
| $self->{content_model_flag} = 'PCDATA'; | ||
| if ($token->{type} eq 'end tag' and $token->{tag_name} eq 'style') { | ||
| ## Ignore the token | ||
| } else { | ||
| !!!parse-error (type => 'in CDATA:#'.$token->{type}); | ||
| ## ISSUE: And ignore? | ||
| } | ||
| 1414 | !!!next-token; | !!!next-token; |
| 1415 | }; # $style_start_tag | }; # $parse_rcdata |
| 1416 | ||
| 1417 | my $script_start_tag = sub { | my $script_start_tag = sub () { |
| 1418 | ## Step 1 | |
| 1419 | my $script_el; | my $script_el; |
| 1420 | !!!create-element ($script_el, 'script', $token->{attributes}); | !!!create-element ($script_el, $HTML_NS, 'script', $token->{attributes}, $token); |
| 1421 | ||
| 1422 | ## Step 2 | |
| 1423 | ## TODO: mark as "parser-inserted" | ## TODO: mark as "parser-inserted" |
| 1424 | ||
| 1425 | $self->{content_model_flag} = 'CDATA'; | ## Step 3 |
| 1426 | ## TODO: Mark as "already executed", if ... | |
| 1427 | ||
| 1428 | ## Step 4 (HTML5 revision 2702) | |
| 1429 | $insert->($script_el); | |
| 1430 | push @{$self->{open_elements}}, [$script_el, $el_category->{script}]; | |
| 1431 | ||
| 1432 | ## Step 5 | |
| 1433 | $self->{content_model} = CDATA_CONTENT_MODEL; | |
| 1434 | delete $self->{escape}; # MUST | delete $self->{escape}; # MUST |
| my $text = ''; | ||
| !!!next-token; | ||
| while ($token->{type} eq 'character') { | ||
| $text .= $token->{data}; | ||
| !!!next-token; | ||
| } # stop if non-character token or tokenizer stops tokenising | ||
| if (length $text) { | ||
| $script_el->manakai_append_text ($text); | ||
| } | ||
| $self->{content_model_flag} = 'PCDATA'; | ||
| 1435 | ||
| 1436 | if ($token->{type} eq 'end tag' and | ## Step 6-7 |
| 1437 | $token->{tag_name} eq 'script') { | $self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
| 1438 | ## Ignore the token | |
| 1439 | } else { | !!!nack ('t40.2'); |
| !!!parse-error (type => 'in CDATA:#'.$token->{type}); | ||
| ## ISSUE: And ignore? | ||
| ## TODO: mark as "already executed" | ||
| } | ||
| if (defined $self->{inner_html_node}) { | ||
| ## TODO: mark as "already executed" | ||
| } else { | ||
| ## TODO: $old_insertion_point = current insertion point | ||
| ## TODO: insertion point = just before the next input character | ||
| (($self->{insertion_mode} eq 'in head' and defined $self->{head_element}) | ||
| ? $self->{head_element} : $self->{open_elements}->[-1]->[0])->append_child ($script_el); | ||
| ## TODO: insertion point = $old_insertion_point (might be "undefined") | ||
| ## TODO: if there is a script that will execute as soon as the parser resume, then... | ||
| } | ||
| 1440 | !!!next-token; | !!!next-token; |
| 1441 | }; # $script_start_tag | }; # $script_start_tag |
| 1442 | ||
| 1443 | ## NOTE: $open_tables->[-1]->[0] is the "current table" element node. | |
| 1444 | ## NOTE: $open_tables->[-1]->[1] is the "tainted" flag (OBSOLETE; unused). | |
| 1445 | ## NOTE: $open_tables->[-1]->[2] is set false when non-Text node inserted. | |
| 1446 | my $open_tables = [[$self->{open_elements}->[0]->[0]]]; | |
| 1447 | ||
| 1448 | my $formatting_end_tag = sub { | my $formatting_end_tag = sub { |
| 1449 | my $tag_name = shift; | my $end_tag_token = shift; |
| 1450 | my $tag_name = $end_tag_token->{tag_name}; | |
| 1451 | ||
| 1452 | ## NOTE: The adoption agency algorithm (AAA). | |
| 1453 | ||
| 1454 | FET: { | FET: { |
| 1455 | ## Step 1 | ## Step 1 |
| 1456 | my $formatting_element; | my $formatting_element; |
| 1457 | my $formatting_element_i_in_active; | my $formatting_element_i_in_active; |
| 1458 | AFE: for (reverse 0..$#$active_formatting_elements) { | AFE: for (reverse 0..$#$active_formatting_elements) { |
| 1459 | if ($active_formatting_elements->[$_]->[1] eq $tag_name) { | if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1460 | !!!cp ('t52'); | |
| 1461 | last AFE; | |
| 1462 | } elsif ($active_formatting_elements->[$_]->[0]->manakai_local_name | |
| 1463 | eq $tag_name) { | |
| 1464 | !!!cp ('t51'); | |
| 1465 | $formatting_element = $active_formatting_elements->[$_]; | $formatting_element = $active_formatting_elements->[$_]; |
| 1466 | $formatting_element_i_in_active = $_; | $formatting_element_i_in_active = $_; |
| 1467 | last AFE; | last AFE; |
| } elsif ($active_formatting_elements->[$_]->[0] eq '#marker') { | ||
| last AFE; | ||
| 1468 | } | } |
| 1469 | } # AFE | } # AFE |
| 1470 | unless (defined $formatting_element) { | unless (defined $formatting_element) { |
| 1471 | !!!parse-error (type => 'unmatched end tag:'.$tag_name); | !!!cp ('t53'); |
| 1472 | !!!parse-error (type => 'unmatched end tag', text => $tag_name, token => $end_tag_token); | |
| 1473 | ## Ignore the token | ## Ignore the token |
| 1474 | !!!next-token; | !!!next-token; |
| 1475 | return; | return; |
| # | Line 2233 sub _tree_construction_main ($) { | Line 1481 sub _tree_construction_main ($) { |
| 1481 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 1482 | if ($node->[0] eq $formatting_element->[0]) { | if ($node->[0] eq $formatting_element->[0]) { |
| 1483 | if ($in_scope) { | if ($in_scope) { |
| 1484 | !!!cp ('t54'); | |
| 1485 | $formatting_element_i_in_open = $_; | $formatting_element_i_in_open = $_; |
| 1486 | last INSCOPE; | last INSCOPE; |
| 1487 | } else { # in open elements but not in scope | } else { # in open elements but not in scope |
| 1488 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t55'); |
| 1489 | !!!parse-error (type => 'unmatched end tag', | |
| 1490 | text => $token->{tag_name}, | |
| 1491 | token => $end_tag_token); | |
| 1492 | ## Ignore the token | ## Ignore the token |
| 1493 | !!!next-token; | !!!next-token; |
| 1494 | return; | return; |
| 1495 | } | } |
| 1496 | } elsif ({ | } elsif ($node->[1] & SCOPING_EL) { |
| 1497 | table => 1, caption => 1, td => 1, th => 1, | !!!cp ('t56'); |
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| 1498 | $in_scope = 0; | $in_scope = 0; |
| 1499 | } | } |
| 1500 | } # INSCOPE | } # INSCOPE |
| 1501 | unless (defined $formatting_element_i_in_open) { | unless (defined $formatting_element_i_in_open) { |
| 1502 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t57'); |
| 1503 | !!!parse-error (type => 'unmatched end tag', | |
| 1504 | text => $token->{tag_name}, | |
| 1505 | token => $end_tag_token); | |
| 1506 | pop @$active_formatting_elements; # $formatting_element | pop @$active_formatting_elements; # $formatting_element |
| 1507 | !!!next-token; ## TODO: ok? | !!!next-token; ## TODO: ok? |
| 1508 | return; | return; |
| 1509 | } | } |
| 1510 | if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { | if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { |
| 1511 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | !!!cp ('t58'); |
| 1512 | !!!parse-error (type => 'not closed', | |
| 1513 | text => $self->{open_elements}->[-1]->[0] | |
| 1514 | ->manakai_local_name, | |
| 1515 | token => $end_tag_token); | |
| 1516 | } | } |
| 1517 | ||
| 1518 | ## Step 2 | ## Step 2 |
| # | Line 2263 sub _tree_construction_main ($) { | Line 1520 sub _tree_construction_main ($) { |
| 1520 | my $furthest_block_i_in_open; | my $furthest_block_i_in_open; |
| 1521 | OE: for (reverse 0..$#{$self->{open_elements}}) { | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1522 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 1523 | if (not $formatting_category->{$node->[1]} and | if (not ($node->[1] & FORMATTING_EL) and |
| 1524 | #not $phrasing_category->{$node->[1]} and | #not $phrasing_category->{$node->[1]} and |
| 1525 | ($special_category->{$node->[1]} or | ($node->[1] & SPECIAL_EL or |
| 1526 | $scoping_category->{$node->[1]})) { | $node->[1] & SCOPING_EL)) { ## Scoping is redundant, maybe |
| 1527 | !!!cp ('t59'); | |
| 1528 | $furthest_block = $node; | $furthest_block = $node; |
| 1529 | $furthest_block_i_in_open = $_; | $furthest_block_i_in_open = $_; |
| 1530 | ## NOTE: The topmost (eldest) node. | |
| 1531 | } elsif ($node->[0] eq $formatting_element->[0]) { | } elsif ($node->[0] eq $formatting_element->[0]) { |
| 1532 | !!!cp ('t60'); | |
| 1533 | last OE; | last OE; |
| 1534 | } | } |
| 1535 | } # OE | } # OE |
| 1536 | ||
| 1537 | ## Step 3 | ## Step 3 |
| 1538 | unless (defined $furthest_block) { # MUST | unless (defined $furthest_block) { # MUST |
| 1539 | !!!cp ('t61'); | |
| 1540 | splice @{$self->{open_elements}}, $formatting_element_i_in_open; | splice @{$self->{open_elements}}, $formatting_element_i_in_open; |
| 1541 | splice @$active_formatting_elements, $formatting_element_i_in_active, 1; | splice @$active_formatting_elements, $formatting_element_i_in_active, 1; |
| 1542 | !!!next-token; | !!!next-token; |
| # | Line 2288 sub _tree_construction_main ($) { | Line 1549 sub _tree_construction_main ($) { |
| 1549 | ## Step 5 | ## Step 5 |
| 1550 | my $furthest_block_parent = $furthest_block->[0]->parent_node; | my $furthest_block_parent = $furthest_block->[0]->parent_node; |
| 1551 | if (defined $furthest_block_parent) { | if (defined $furthest_block_parent) { |
| 1552 | !!!cp ('t62'); | |
| 1553 | $furthest_block_parent->remove_child ($furthest_block->[0]); | $furthest_block_parent->remove_child ($furthest_block->[0]); |
| 1554 | } | } |
| 1555 | ||
| # | Line 2310 sub _tree_construction_main ($) { | Line 1572 sub _tree_construction_main ($) { |
| 1572 | S7S2: { | S7S2: { |
| 1573 | for (reverse 0..$#$active_formatting_elements) { | for (reverse 0..$#$active_formatting_elements) { |
| 1574 | if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { | if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
| 1575 | !!!cp ('t63'); | |
| 1576 | $node_i_in_active = $_; | $node_i_in_active = $_; |
| 1577 | last S7S2; | last S7S2; |
| 1578 | } | } |
| # | Line 2323 sub _tree_construction_main ($) { | Line 1586 sub _tree_construction_main ($) { |
| 1586 | ||
| 1587 | ## Step 4 | ## Step 4 |
| 1588 | if ($last_node->[0] eq $furthest_block->[0]) { | if ($last_node->[0] eq $furthest_block->[0]) { |
| 1589 | !!!cp ('t64'); | |
| 1590 | $bookmark_prev_el = $node->[0]; | $bookmark_prev_el = $node->[0]; |
| 1591 | } | } |
| 1592 | ||
| 1593 | ## Step 5 | ## Step 5 |
| 1594 | if ($node->[0]->has_child_nodes ()) { | if ($node->[0]->has_child_nodes ()) { |
| 1595 | !!!cp ('t65'); | |
| 1596 | my $clone = [$node->[0]->clone_node (0), $node->[1]]; | my $clone = [$node->[0]->clone_node (0), $node->[1]]; |
| 1597 | $active_formatting_elements->[$node_i_in_active] = $clone; | $active_formatting_elements->[$node_i_in_active] = $clone; |
| 1598 | $self->{open_elements}->[$node_i_in_open] = $clone; | $self->{open_elements}->[$node_i_in_open] = $clone; |
| # | Line 2345 sub _tree_construction_main ($) { | Line 1610 sub _tree_construction_main ($) { |
| 1610 | } # S7 | } # S7 |
| 1611 | ||
| 1612 | ## Step 8 | ## Step 8 |
| 1613 | $common_ancestor_node->[0]->append_child ($last_node->[0]); | if ($common_ancestor_node->[1] & TABLE_ROWS_EL) { |
| 1614 | ## Foster parenting. | |
| 1615 | my $foster_parent_element; | |
| 1616 | my $next_sibling; | |
| 1617 | OE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 1618 | if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { | |
| 1619 | !!!cp ('t65.2'); | |
| 1620 | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; | |
| 1621 | $next_sibling = $self->{open_elements}->[$_]->[0]; | |
| 1622 | undef $next_sibling | |
| 1623 | unless $next_sibling->parent_node eq $foster_parent_element; | |
| 1624 | last OE; | |
| 1625 | } | |
| 1626 | } # OE | |
| 1627 | $foster_parent_element ||= $self->{open_elements}->[0]->[0]; | |
| 1628 | ||
| 1629 | $foster_parent_element->insert_before ($last_node->[0], $next_sibling); | |
| 1630 | $open_tables->[-1]->[1] = 1; # tainted | |
| 1631 | } else { | |
| 1632 | !!!cp ('t65.3'); | |
| 1633 | $common_ancestor_node->[0]->append_child ($last_node->[0]); | |
| 1634 | } | |
| 1635 | ||
| 1636 | ## Step 9 | ## Step 9 |
| 1637 | my $clone = [$formatting_element->[0]->clone_node (0), | my $clone = [$formatting_element->[0]->clone_node (0), |
| # | Line 2362 sub _tree_construction_main ($) { | Line 1648 sub _tree_construction_main ($) { |
| 1648 | my $i; | my $i; |
| 1649 | AFE: for (reverse 0..$#$active_formatting_elements) { | AFE: for (reverse 0..$#$active_formatting_elements) { |
| 1650 | if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { | if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { |
| 1651 | !!!cp ('t66'); | |
| 1652 | splice @$active_formatting_elements, $_, 1; | splice @$active_formatting_elements, $_, 1; |
| 1653 | $i-- and last AFE if defined $i; | $i-- and last AFE if defined $i; |
| 1654 | } elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { | } elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { |
| 1655 | !!!cp ('t67'); | |
| 1656 | $i = $_; | $i = $_; |
| 1657 | } | } |
| 1658 | } # AFE | } # AFE |
| # | Line 2374 sub _tree_construction_main ($) { | Line 1662 sub _tree_construction_main ($) { |
| 1662 | undef $i; | undef $i; |
| 1663 | OE: for (reverse 0..$#{$self->{open_elements}}) { | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1664 | if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { | if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { |
| 1665 | !!!cp ('t68'); | |
| 1666 | splice @{$self->{open_elements}}, $_, 1; | splice @{$self->{open_elements}}, $_, 1; |
| 1667 | $i-- and last OE if defined $i; | $i-- and last OE if defined $i; |
| 1668 | } elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { | } elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { |
| 1669 | !!!cp ('t69'); | |
| 1670 | $i = $_; | $i = $_; |
| 1671 | } | } |
| 1672 | } # OE | } # OE |
| 1673 | splice @{$self->{open_elements}}, $i + 1, 1, $clone; | splice @{$self->{open_elements}}, $i + 1, 0, $clone; |
| 1674 | ||
| 1675 | ## Step 14 | ## Step 14 |
| 1676 | redo FET; | redo FET; |
| 1677 | } # FET | } # FET |
| 1678 | }; # $formatting_end_tag | }; # $formatting_end_tag |
| 1679 | ||
| 1680 | my $insert_to_current = sub { | $insert = my $insert_to_current = sub { |
| 1681 | $self->{open_elements}->[-1]->[0]->append_child (shift); | $self->{open_elements}->[-1]->[0]->append_child ($_[0]); |
| 1682 | }; # $insert_to_current | }; # $insert_to_current |
| 1683 | ||
| 1684 | ## Foster parenting. Note that there are three "foster parenting" | |
| 1685 | ## code in the parser: for elements (this one), for texts, and for | |
| 1686 | ## elements in the AAA code. | |
| 1687 | my $insert_to_foster = sub { | my $insert_to_foster = sub { |
| 1688 | my $child = shift; | my $child = shift; |
| 1689 | if ({ | if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1690 | table => 1, tbody => 1, tfoot => 1, | # MUST |
| 1691 | thead => 1, tr => 1, | my $foster_parent_element; |
| 1692 | }->{$self->{open_elements}->[-1]->[1]}) { | my $next_sibling; |
| 1693 | # MUST | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1694 | my $foster_parent_element; | if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1695 | my $next_sibling; | !!!cp ('t71'); |
| 1696 | OE: for (reverse 0..$#{$self->{open_elements}}) { | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1697 | if ($self->{open_elements}->[$_]->[1] eq 'table') { | $next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1698 | my $parent = $self->{open_elements}->[$_]->[0]->parent_node; | undef $next_sibling |
| 1699 | if (defined $parent and $parent->node_type == 1) { | unless $next_sibling->parent_node eq $foster_parent_element; |
| 1700 | $foster_parent_element = $parent; | last OE; |
| 1701 | $next_sibling = $self->{open_elements}->[$_]->[0]; | } |
| 1702 | } else { | } # OE |
| 1703 | $foster_parent_element | $foster_parent_element ||= $self->{open_elements}->[0]->[0]; |
| 1704 | = $self->{open_elements}->[$_ - 1]->[0]; | |
| 1705 | } | $foster_parent_element->insert_before ($child, $next_sibling); |
| 1706 | last OE; | $open_tables->[-1]->[1] = 1; # tainted |
| 1707 | } | } else { |
| 1708 | } # OE | !!!cp ('t72'); |
| 1709 | $foster_parent_element = $self->{open_elements}->[0]->[0] | $self->{open_elements}->[-1]->[0]->append_child ($child); |
| 1710 | unless defined $foster_parent_element; | } |
| $foster_parent_element->insert_before | ||
| ($child, $next_sibling); | ||
| } else { | ||
| $self->{open_elements}->[-1]->[0]->append_child ($child); | ||
| } | ||
| 1711 | }; # $insert_to_foster | }; # $insert_to_foster |
| 1712 | ||
| 1713 | my $in_body = sub { | ## NOTE: Insert a character (MUST): When a character is inserted, if |
| 1714 | my $insert = shift; | ## the last node that was inserted by the parser is a Text node and |
| 1715 | if ($token->{type} eq 'start tag') { | ## the character has to be inserted after that node, then the |
| 1716 | if ($token->{tag_name} eq 'script') { | ## character is appended to the Text node. However, if any other |
| 1717 | $script_start_tag->(); | ## node is inserted by the parser, then a new Text node is created |
| 1718 | return; | ## and the character is appended as that Text node. If I'm not |
| 1719 | } elsif ($token->{tag_name} eq 'style') { | ## wrong, for a parser with scripting disabled, there are only two |
| 1720 | $style_start_tag->(); | ## cases where this occurs. One is the case where an element node |
| 1721 | return; | ## is inserted to the |head| element. This is covered by using the |
| 1722 | } elsif ({ | ## |$self->{head_element_inserted}| flag. Another is the case where |
| 1723 | base => 1, link => 1, meta => 1, | ## an element or comment is inserted into the |table| subtree while |
| 1724 | }->{$token->{tag_name}}) { | ## foster parenting happens. This is covered by using the [2] flag |
| 1725 | ## NOTE: This is an "as if in head" code clone | ## of the |$open_tables| structure. All other cases are handled |
| 1726 | my $el; | ## simply by calling |manakai_append_text| method. |
| 1727 | !!!create-element ($el, $token->{tag_name}, $token->{attributes}); | |
| 1728 | if ($self->{insertion_mode} eq 'in head' and | ## TODO: |<body><script>document.write("a<br>"); |
| 1729 | defined $self->{head_element}) { | ## document.body.removeChild (document.body.lastChild); |
| 1730 | $self->{head_element}->append_child ($el); | ## document.write ("b")</script>| |
| 1731 | } else { | |
| 1732 | $insert->($el); | B: while (1) { |
| 1733 | } | |
| 1734 | ## The "in table text" insertion mode. | |
| 1735 | !!!next-token; | if ($self->{insertion_mode} & TABLE_IMS and |
| 1736 | return; | not $self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and |
| 1737 | } elsif ($token->{tag_name} eq 'title') { | not $self->{insertion_mode} & IN_CDATA_RCDATA_IM) { |
| 1738 | !!!parse-error (type => 'in body:title'); | C: { |
| 1739 | ## NOTE: There is an "as if in head" code clone | my $s; |
| 1740 | my $title_el; | if ($token->{type} == CHARACTER_TOKEN) { |
| 1741 | !!!create-element ($title_el, 'title', $token->{attributes}); | !!!cp ('t194'); |
| 1742 | (defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | $self->{pending_chars} ||= []; |
| 1743 | ->append_child ($title_el); | push @{$self->{pending_chars}}, $token; |
| $self->{content_model_flag} = 'RCDATA'; | ||
| delete $self->{escape}; # MUST | ||
| my $text = ''; | ||
| !!!next-token; | ||
| while ($token->{type} eq 'character') { | ||
| $text .= $token->{data}; | ||
| 1744 | !!!next-token; | !!!next-token; |
| 1745 | } | next B; |
| if (length $text) { | ||
| $title_el->manakai_append_text ($text); | ||
| } | ||
| $self->{content_model_flag} = 'PCDATA'; | ||
| if ($token->{type} eq 'end tag' and | ||
| $token->{tag_name} eq 'title') { | ||
| ## Ignore the token | ||
| 1746 | } else { | } else { |
| 1747 | !!!parse-error (type => 'in RCDATA:#'.$token->{type}); | if ($self->{pending_chars}) { |
| 1748 | ## ISSUE: And ignore? | $s = join '', map { $_->{data} } @{$self->{pending_chars}}; |
| 1749 | } | delete $self->{pending_chars}; |
| 1750 | !!!next-token; | if ($s =~ /[^\x09\x0A\x0C\x0D\x20]/) { |
| 1751 | return; | !!!cp ('t195'); |
| 1752 | } elsif ($token->{tag_name} eq 'body') { | # |
| 1753 | !!!parse-error (type => 'in body:body'); | } else { |
| 1754 | !!!cp ('t195.1'); | |
| 1755 | if (@{$self->{open_elements}} == 1 or | #$self->{open_elements}->[-1]->[0]->manakai_append_text ($s); |
| 1756 | $self->{open_elements}->[1]->[1] ne 'body') { | $self->{open_elements}->[-1]->[0]->append_child |
| 1757 | ## Ignore the token | ($self->{document}->create_text_node ($s)); |
| 1758 | } else { | last C; |
| my $body_el = $self->{open_elements}->[1]->[0]; | ||
| for my $attr_name (keys %{$token->{attributes}}) { | ||
| unless ($body_el->has_attribute_ns (undef, $attr_name)) { | ||
| $body_el->set_attribute_ns | ||
| (undef, [undef, $attr_name], | ||
| $token->{attributes}->{$attr_name}->{value}); | ||
| 1759 | } | } |
| 1760 | } else { | |
| 1761 | !!!cp ('t195.2'); | |
| 1762 | last C; | |
| 1763 | } | } |
| 1764 | } | } |
| 1765 | !!!next-token; | |
| 1766 | return; | ## Foster parenting. |
| 1767 | } elsif ({ | !!!parse-error (type => 'in table:#text', token => $token); |
| 1768 | address => 1, blockquote => 1, center => 1, dir => 1, | |
| 1769 | div => 1, dl => 1, fieldset => 1, listing => 1, | ## NOTE: As if in body, but insert into the foster parent element. |
| 1770 | menu => 1, ol => 1, p => 1, ul => 1, | $reconstruct_active_formatting_elements->($insert_to_foster); |
| 1771 | pre => 1, | |
| 1772 | }->{$token->{tag_name}}) { | if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1773 | ## has a p element in scope | # MUST |
| 1774 | INSCOPE: for (reverse @{$self->{open_elements}}) { | my $foster_parent_element; |
| 1775 | if ($_->[1] eq 'p') { | my $next_sibling; |
| 1776 | !!!back-token; | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1777 | $token = {type => 'end tag', tag_name => 'p'}; | if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1778 | return; | !!!cp ('t197'); |
| 1779 | } elsif ({ | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1780 | table => 1, caption => 1, td => 1, th => 1, | $next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1781 | button => 1, marquee => 1, object => 1, html => 1, | undef $next_sibling |
| 1782 | }->{$_->[1]}) { | unless $next_sibling->parent_node eq $foster_parent_element; |
| 1783 | last INSCOPE; | last OE; |
| 1784 | } | } |
| 1785 | } # INSCOPE | } # OE |
| 1786 | $foster_parent_element ||= $self->{open_elements}->[0]->[0]; | |
| 1787 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | |
| 1788 | if ($token->{tag_name} eq 'pre') { | !!!cp ('t199'); |
| 1789 | !!!next-token; | $foster_parent_element->insert_before |
| 1790 | if ($token->{type} eq 'character') { | ($self->{document}->create_text_node ($s), $next_sibling); |
| 1791 | $token->{data} =~ s/^\x0A//; | |
| 1792 | unless (length $token->{data}) { | $open_tables->[-1]->[1] = 1; # tainted |
| 1793 | !!!next-token; | $open_tables->[-1]->[2] = 1; # ~node inserted |
| 1794 | } | } else { |
| 1795 | } | ## NOTE: Fragment case or in a foster parent'ed element |
| 1796 | } else { | ## (e.g. |<table><span>a|). In fragment case, whether the |
| 1797 | !!!next-token; | ## character is appended to existing node or a new node is |
| 1798 | ## created is irrelevant, since the foster parent'ed nodes | |
| 1799 | ## are discarded and fragment parsing does not invoke any | |
| 1800 | ## script. | |
| 1801 | !!!cp ('t200'); | |
| 1802 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($s); | |
| 1803 | } | |
| 1804 | } # C | |
| 1805 | } # TABLE_IMS | |
| 1806 | ||
| 1807 | if ($token->{type} == DOCTYPE_TOKEN) { | |
| 1808 | !!!cp ('t73'); | |
| 1809 | !!!parse-error (type => 'in html:#DOCTYPE', token => $token); | |
| 1810 | ## Ignore the token | |
| 1811 | ## Stay in the phase | |
| 1812 | !!!next-token; | |
| 1813 | next B; | |
| 1814 | } elsif ($token->{type} == START_TAG_TOKEN and | |
| 1815 | $token->{tag_name} eq 'html') { | |
| 1816 | if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { | |
| 1817 | !!!cp ('t79'); | |
| 1818 | !!!parse-error (type => 'after html', text => 'html', token => $token); | |
| 1819 | $self->{insertion_mode} = AFTER_BODY_IM; | |
| 1820 | } elsif ($self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) { | |
| 1821 | !!!cp ('t80'); | |
| 1822 | !!!parse-error (type => 'after html', text => 'html', token => $token); | |
| 1823 | $self->{insertion_mode} = AFTER_FRAMESET_IM; | |
| 1824 | } else { | |
| 1825 | !!!cp ('t81'); | |
| 1826 | } | |
| 1827 | ||
| 1828 | !!!cp ('t82'); | |
| 1829 | !!!parse-error (type => 'not first start tag', token => $token); | |
| 1830 | my $top_el = $self->{open_elements}->[0]->[0]; | |
| 1831 | for my $attr_name (keys %{$token->{attributes}}) { | |
| 1832 | unless ($top_el->has_attribute_ns (undef, $attr_name)) { | |
| 1833 | !!!cp ('t84'); | |
| 1834 | $top_el->set_attribute_ns | |
| 1835 | (undef, [undef, $attr_name], | |
| 1836 | $token->{attributes}->{$attr_name}->{value}); | |
| 1837 | } | } |
| 1838 | return; | } |
| 1839 | } elsif ($token->{tag_name} eq 'form') { | !!!nack ('t84.1'); |
| 1840 | if (defined $self->{form_element}) { | !!!next-token; |
| 1841 | !!!parse-error (type => 'in form:form'); | next B; |
| 1842 | ## Ignore the token | } elsif ($token->{type} == COMMENT_TOKEN) { |
| 1843 | !!!next-token; | my $comment = $self->{document}->create_comment ($token->{data}); |
| 1844 | return; | if ($self->{insertion_mode} & AFTER_HTML_IMS) { |
| 1845 | !!!cp ('t85'); | |
| 1846 | $self->{document}->append_child ($comment); | |
| 1847 | } elsif ($self->{insertion_mode} == AFTER_BODY_IM) { | |
| 1848 | !!!cp ('t86'); | |
| 1849 | $self->{open_elements}->[0]->[0]->append_child ($comment); | |
| 1850 | } else { | |
| 1851 | !!!cp ('t87'); | |
| 1852 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | |
| 1853 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 1854 | } | |
| 1855 | !!!next-token; | |
| 1856 | next B; | |
| 1857 | } elsif ($self->{insertion_mode} & IN_CDATA_RCDATA_IM) { | |
| 1858 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 1859 | $token->{data} =~ s/^\x0A// if $self->{ignore_newline}; | |
| 1860 | delete $self->{ignore_newline}; | |
| 1861 | ||
| 1862 | if (length $token->{data}) { | |
| 1863 | !!!cp ('t43'); | |
| 1864 | $self->{open_elements}->[-1]->[0]->manakai_append_text | |
| 1865 | ($token->{data}); | |
| 1866 | } else { | } else { |
| 1867 | ## has a p element in scope | !!!cp ('t43.1'); |
| INSCOPE: for (reverse @{$self->{open_elements}}) { | ||
| if ($_->[1] eq 'p') { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'p'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$_->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| $self->{form_element} = $self->{open_elements}->[-1]->[0]; | ||
| !!!next-token; | ||
| return; | ||
| 1868 | } | } |
| } elsif ($token->{tag_name} eq 'li') { | ||
| ## has a p element in scope | ||
| INSCOPE: for (reverse @{$self->{open_elements}}) { | ||
| if ($_->[1] eq 'p') { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'p'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$_->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| ## Step 1 | ||
| my $i = -1; | ||
| my $node = $self->{open_elements}->[$i]; | ||
| LI: { | ||
| ## Step 2 | ||
| if ($node->[1] eq 'li') { | ||
| if ($i != -1) { | ||
| !!!parse-error (type => 'end tag missing:'. | ||
| $self->{open_elements}->[-1]->[1]); | ||
| ## TODO: test | ||
| } | ||
| splice @{$self->{open_elements}}, $i; | ||
| last LI; | ||
| } | ||
| ## Step 3 | ||
| if (not $formatting_category->{$node->[1]} and | ||
| #not $phrasing_category->{$node->[1]} and | ||
| ($special_category->{$node->[1]} or | ||
| $scoping_category->{$node->[1]}) and | ||
| $node->[1] ne 'address' and $node->[1] ne 'div') { | ||
| last LI; | ||
| } | ||
| ## Step 4 | ||
| $i--; | ||
| $node = $self->{open_elements}->[$i]; | ||
| redo LI; | ||
| } # LI | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| 1869 | !!!next-token; | !!!next-token; |
| 1870 | return; | next B; |
| 1871 | } elsif ($token->{tag_name} eq 'dd' or $token->{tag_name} eq 'dt') { | } elsif ($token->{type} == END_TAG_TOKEN) { |
| 1872 | ## has a p element in scope | delete $self->{ignore_newline}; |
| 1873 | INSCOPE: for (reverse @{$self->{open_elements}}) { | |
| 1874 | if ($_->[1] eq 'p') { | if ($token->{tag_name} eq 'script') { |
| 1875 | !!!back-token; | !!!cp ('t50'); |
| $token = {type => 'end tag', tag_name => 'p'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$_->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| ## Step 1 | ||
| my $i = -1; | ||
| my $node = $self->{open_elements}->[$i]; | ||
| LI: { | ||
| ## Step 2 | ||
| if ($node->[1] eq 'dt' or $node->[1] eq 'dd') { | ||
| if ($i != -1) { | ||
| !!!parse-error (type => 'end tag missing:'. | ||
| $self->{open_elements}->[-1]->[1]); | ||
| ## TODO: test | ||
| } | ||
| splice @{$self->{open_elements}}, $i; | ||
| last LI; | ||
| } | ||
| ## Step 3 | ||
| if (not $formatting_category->{$node->[1]} and | ||
| #not $phrasing_category->{$node->[1]} and | ||
| ($special_category->{$node->[1]} or | ||
| $scoping_category->{$node->[1]}) and | ||
| $node->[1] ne 'address' and $node->[1] ne 'div') { | ||
| last LI; | ||
| } | ||
| ## Step 4 | ||
| $i--; | ||
| $node = $self->{open_elements}->[$i]; | ||
| redo LI; | ||
| } # LI | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'plaintext') { | ||
| ## has a p element in scope | ||
| INSCOPE: for (reverse @{$self->{open_elements}}) { | ||
| if ($_->[1] eq 'p') { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'p'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$_->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| $self->{content_model_flag} = 'PLAINTEXT'; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ({ | ||
| h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, | ||
| }->{$token->{tag_name}}) { | ||
| ## has a p element in scope | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ($node->[1] eq 'p') { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'p'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| ## NOTE: See <http://html5.org/tools/web-apps-tracker?from=925&to=926> | ||
| ## has an element in scope | ||
| #my $i; | ||
| #INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| # my $node = $self->{open_elements}->[$_]; | ||
| # if ({ | ||
| # h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, | ||
| # }->{$node->[1]}) { | ||
| # $i = $_; | ||
| # last INSCOPE; | ||
| # } elsif ({ | ||
| # table => 1, caption => 1, td => 1, th => 1, | ||
| # button => 1, marquee => 1, object => 1, html => 1, | ||
| # }->{$node->[1]}) { | ||
| # last INSCOPE; | ||
| # } | ||
| #} # INSCOPE | ||
| # | ||
| #if (defined $i) { | ||
| # !!! parse-error (type => 'in hn:hn'); | ||
| # splice @{$self->{open_elements}}, $i; | ||
| #} | ||
| 1876 | ||
| 1877 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ## Para 1-2 |
| 1878 | my $script = pop @{$self->{open_elements}}; | |
| 1879 | ||
| 1880 | !!!next-token; | ## Para 3 |
| 1881 | return; | $self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| } elsif ($token->{tag_name} eq 'a') { | ||
| AFE: for my $i (reverse 0..$#$active_formatting_elements) { | ||
| my $node = $active_formatting_elements->[$i]; | ||
| if ($node->[1] eq 'a') { | ||
| !!!parse-error (type => 'in a:a'); | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'a'}; | ||
| $formatting_end_tag->($token->{tag_name}); | ||
| AFE2: for (reverse 0..$#$active_formatting_elements) { | ||
| if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { | ||
| splice @$active_formatting_elements, $_, 1; | ||
| last AFE2; | ||
| } | ||
| } # AFE2 | ||
| OE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| if ($self->{open_elements}->[$_]->[0] eq $node->[0]) { | ||
| splice @{$self->{open_elements}}, $_, 1; | ||
| last OE; | ||
| } | ||
| } # OE | ||
| last AFE; | ||
| } elsif ($node->[0] eq '#marker') { | ||
| last AFE; | ||
| } | ||
| } # AFE | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| 1882 | ||
| 1883 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ## Para 4 |
| 1884 | push @$active_formatting_elements, $self->{open_elements}->[-1]; | ## TODO: $old_insertion_point = $current_insertion_point; |
| 1885 | ## TODO: $current_insertion_point = just before $self->{nc}; | |
| 1886 | ||
| 1887 | ## Para 5 | |
| 1888 | ## TODO: Run the $script->[0]. | |
| 1889 | ||
| 1890 | ## Para 6 | |
| 1891 | ## TODO: $current_insertion_point = $old_insertion_point; | |
| 1892 | ||
| 1893 | ## Para 7 | |
| 1894 | ## TODO: if ($pending_external_script) { | |
| 1895 | ## TODO: ... | |
| 1896 | ## TODO: } | |
| 1897 | ||
| 1898 | !!!next-token; | !!!next-token; |
| 1899 | return; | next B; |
| 1900 | } elsif ({ | } else { |
| 1901 | b => 1, big => 1, em => 1, font => 1, i => 1, | !!!cp ('t42'); |
| 1902 | s => 1, small => 1, strile => 1, | |
| 1903 | strong => 1, tt => 1, u => 1, | pop @{$self->{open_elements}}; |
| }->{$token->{tag_name}}) { | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| push @$active_formatting_elements, $self->{open_elements}->[-1]; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'nobr') { | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| ## has a |nobr| element in scope | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ($node->[1] eq 'nobr') { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'nobr'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| push @$active_formatting_elements, $self->{open_elements}->[-1]; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'button') { | ||
| ## has a button element in scope | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ($node->[1] eq 'button') { | ||
| !!!parse-error (type => 'in button:button'); | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'button'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| push @$active_formatting_elements, ['#marker', '']; | ||
| 1904 | ||
| 1905 | !!!next-token; | $self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1906 | return; | !!!next-token; |
| 1907 | } elsif ($token->{tag_name} eq 'marquee' or | next B; |
| $token->{tag_name} eq 'object') { | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| push @$active_formatting_elements, ['#marker', '']; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'xmp') { | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| $self->{content_model_flag} = 'CDATA'; | ||
| delete $self->{escape}; # MUST | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'table') { | ||
| ## has a p element in scope | ||
| INSCOPE: for (reverse @{$self->{open_elements}}) { | ||
| if ($_->[1] eq 'p') { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => 'p'}; | ||
| return; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$_->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| $self->{insertion_mode} = 'in table'; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ({ | ||
| area => 1, basefont => 1, bgsound => 1, br => 1, | ||
| embed => 1, img => 1, param => 1, spacer => 1, wbr => 1, | ||
| image => 1, | ||
| }->{$token->{tag_name}}) { | ||
| if ($token->{tag_name} eq 'image') { | ||
| !!!parse-error (type => 'image'); | ||
| $token->{tag_name} = 'img'; | ||
| 1908 | } | } |
| 1909 | } elsif ($token->{type} == END_OF_FILE_TOKEN) { | |
| 1910 | $reconstruct_active_formatting_elements->($insert_to_current); | delete $self->{ignore_newline}; |
| 1911 | ||
| 1912 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | !!!cp ('t44'); |
| 1913 | !!!parse-error (type => 'not closed', | |
| 1914 | text => $self->{open_elements}->[-1]->[0] | |
| 1915 | ->manakai_local_name, | |
| 1916 | token => $token); | |
| 1917 | ||
| 1918 | #if ($self->{open_elements}->[-1]->[1] == SCRIPT_EL) { | |
| 1919 | # ## TODO: Mark as "already executed" | |
| 1920 | #} | |
| 1921 | ||
| 1922 | pop @{$self->{open_elements}}; | pop @{$self->{open_elements}}; |
| 1923 | ||
| 1924 | $self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; | |
| 1925 | ## Reprocess. | |
| 1926 | next B; | |
| 1927 | } else { | |
| 1928 | die "$0: $token->{type}: In CDATA/RCDATA: Unknown token type"; | |
| 1929 | } | |
| 1930 | } elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM) { | |
| 1931 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 1932 | !!!cp ('t87.1'); | |
| 1933 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | |
| 1934 | !!!next-token; | !!!next-token; |
| 1935 | return; | next B; |
| 1936 | } elsif ($token->{tag_name} eq 'hr') { | } elsif ($token->{type} == START_TAG_TOKEN) { |
| 1937 | ## has a p element in scope | if ((not {mglyph => 1, malignmark => 1}->{$token->{tag_name}} and |
| 1938 | INSCOPE: for (reverse @{$self->{open_elements}}) { | $self->{open_elements}->[-1]->[1] & FOREIGN_FLOW_CONTENT_EL) or |
| 1939 | if ($_->[1] eq 'p') { | not ($self->{open_elements}->[-1]->[1] & FOREIGN_EL) or |
| 1940 | !!!back-token; | ($token->{tag_name} eq 'svg' and |
| 1941 | $token = {type => 'end tag', tag_name => 'p'}; | $self->{open_elements}->[-1]->[1] == MML_AXML_EL)) { |
| 1942 | return; | ## NOTE: "using the rules for secondary insertion mode"then"continue" |
| 1943 | } elsif ({ | !!!cp ('t87.2'); |
| 1944 | table => 1, caption => 1, td => 1, th => 1, | # |
| 1945 | button => 1, marquee => 1, object => 1, html => 1, | } elsif ({ |
| 1946 | }->{$_->[1]}) { | b => 1, big => 1, blockquote => 1, body => 1, br => 1, |
| 1947 | last INSCOPE; | center => 1, code => 1, dd => 1, div => 1, dl => 1, dt => 1, |
| 1948 | em => 1, embed => 1, h1 => 1, h2 => 1, h3 => 1, | |
| 1949 | h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, i => 1, | |
| 1950 | img => 1, li => 1, listing => 1, menu => 1, meta => 1, | |
| 1951 | nobr => 1, ol => 1, p => 1, pre => 1, ruby => 1, s => 1, | |
| 1952 | small => 1, span => 1, strong => 1, strike => 1, sub => 1, | |
| 1953 | sup => 1, table => 1, tt => 1, u => 1, ul => 1, var => 1, | |
| 1954 | }->{$token->{tag_name}} or | |
| 1955 | ($token->{tag_name} eq 'font' and | |
| 1956 | ($token->{attributes}->{color} or | |
| 1957 | $token->{attributes}->{face} or | |
| 1958 | $token->{attributes}->{size}))) { | |
| 1959 | !!!cp ('t87.2'); | |
| 1960 | !!!parse-error (type => 'not closed', | |
| 1961 | text => $self->{open_elements}->[-1]->[0] | |
| 1962 | ->manakai_local_name, | |
| 1963 | token => $token); | |
| 1964 | ||
| 1965 | pop @{$self->{open_elements}} | |
| 1966 | while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; | |
| 1967 | ||
| 1968 | $self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; | |
| 1969 | ## Reprocess. | |
| 1970 | next B; | |
| 1971 | } else { | |
| 1972 | my $nsuri = $self->{open_elements}->[-1]->[0]->namespace_uri; | |
| 1973 | my $tag_name = $token->{tag_name}; | |
| 1974 | if ($nsuri eq $SVG_NS) { | |
| 1975 | $tag_name = { | |
| 1976 | altglyph => 'altGlyph', | |
| 1977 | altglyphdef => 'altGlyphDef', | |
| 1978 | altglyphitem => 'altGlyphItem', | |
| 1979 | animatecolor => 'animateColor', | |
| 1980 | animatemotion => 'animateMotion', | |
| 1981 | animatetransform => 'animateTransform', | |
| 1982 | clippath => 'clipPath', | |
| 1983 | feblend => 'feBlend', | |
| 1984 | fecolormatrix => 'feColorMatrix', | |
| 1985 | fecomponenttransfer => 'feComponentTransfer', | |
| 1986 | fecomposite => 'feComposite', | |
| 1987 | feconvolvematrix => 'feConvolveMatrix', | |
| 1988 | fediffuselighting => 'feDiffuseLighting', | |
| 1989 | fedisplacementmap => 'feDisplacementMap', | |
| 1990 | fedistantlight => 'feDistantLight', | |
| 1991 | feflood => 'feFlood', | |
| 1992 | fefunca => 'feFuncA', | |
| 1993 | fefuncb => 'feFuncB', | |
| 1994 | fefuncg => 'feFuncG', | |
| 1995 | fefuncr => 'feFuncR', | |
| 1996 | fegaussianblur => 'feGaussianBlur', | |
| 1997 | feimage => 'feImage', | |
| 1998 | femerge => 'feMerge', | |
| 1999 | femergenode => 'feMergeNode', | |
| 2000 | femorphology => 'feMorphology', | |
| 2001 | feoffset => 'feOffset', | |
| 2002 | fepointlight => 'fePointLight', | |
| 2003 | fespecularlighting => 'feSpecularLighting', | |
| 2004 | fespotlight => 'feSpotLight', | |
| 2005 | fetile => 'feTile', | |
| 2006 | feturbulence => 'feTurbulence', | |
| 2007 | foreignobject => 'foreignObject', | |
| 2008 | glyphref => 'glyphRef', | |
| 2009 | lineargradient => 'linearGradient', | |
| 2010 | radialgradient => 'radialGradient', | |
| 2011 | #solidcolor => 'solidColor', ## NOTE: Commented in spec (SVG1.2) | |
| 2012 | textpath => 'textPath', | |
| 2013 | }->{$tag_name} || $tag_name; | |
| 2014 | } | } |
| 2015 | } # INSCOPE | |
| 2016 | ## "adjust SVG attributes" (SVG only) - done in insert-element-f | |
| 2017 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | |
| 2018 | pop @{$self->{open_elements}}; | ## "adjust foreign attributes" - done in insert-element-f |
| 2019 | ||
| 2020 | !!!next-token; | !!!insert-element-f ($nsuri, $tag_name, $token->{attributes}, $token); |
| 2021 | return; | |
| 2022 | } elsif ($token->{tag_name} eq 'input') { | if ($self->{self_closing}) { |
| 2023 | $reconstruct_active_formatting_elements->($insert_to_current); | pop @{$self->{open_elements}}; |
| 2024 | !!!ack ('t87.3'); | |
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| ## TODO: associate with $self->{form_element} if defined | ||
| pop @{$self->{open_elements}}; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'isindex') { | ||
| !!!parse-error (type => 'isindex'); | ||
| if (defined $self->{form_element}) { | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| return; | ||
| } else { | ||
| my $at = $token->{attributes}; | ||
| my $form_attrs; | ||
| $form_attrs->{action} = $at->{action} if $at->{action}; | ||
| my $prompt_attr = $at->{prompt}; | ||
| $at->{name} = {name => 'name', value => 'isindex'}; | ||
| delete $at->{action}; | ||
| delete $at->{prompt}; | ||
| my @tokens = ( | ||
| {type => 'start tag', tag_name => 'form', | ||
| attributes => $form_attrs}, | ||
| {type => 'start tag', tag_name => 'hr'}, | ||
| {type => 'start tag', tag_name => 'p'}, | ||
| {type => 'start tag', tag_name => 'label'}, | ||
| ); | ||
| if ($prompt_attr) { | ||
| push @tokens, {type => 'character', data => $prompt_attr->{value}}; | ||
| 2025 | } else { | } else { |
| 2026 | push @tokens, {type => 'character', | !!!cp ('t87.4'); |
| data => 'This is a searchable index. Insert your search keywords here: '}; # SHOULD | ||
| ## TODO: make this configurable | ||
| 2027 | } | } |
| 2028 | push @tokens, | |
| {type => 'start tag', tag_name => 'input', attributes => $at}, | ||
| #{type => 'character', data => ''}, # SHOULD | ||
| {type => 'end tag', tag_name => 'label'}, | ||
| {type => 'end tag', tag_name => 'p'}, | ||
| {type => 'start tag', tag_name => 'hr'}, | ||
| {type => 'end tag', tag_name => 'form'}; | ||
| $token = shift @tokens; | ||
| !!!back-token (@tokens); | ||
| return; | ||
| } | ||
| } elsif ({ | ||
| textarea => 1, | ||
| iframe => 1, | ||
| noembed => 1, | ||
| noframes => 1, | ||
| noscript => 0, ## TODO: 1 if scripting is enabled | ||
| }->{$token->{tag_name}}) { | ||
| my $tag_name = $token->{tag_name}; | ||
| my $el; | ||
| !!!create-element ($el, $token->{tag_name}, $token->{attributes}); | ||
| if ($token->{tag_name} eq 'textarea') { | ||
| ## TODO: $self->{form_element} if defined | ||
| $self->{content_model_flag} = 'RCDATA'; | ||
| } else { | ||
| $self->{content_model_flag} = 'CDATA'; | ||
| } | ||
| delete $self->{escape}; # MUST | ||
| $insert->($el); | ||
| my $text = ''; | ||
| if ($token->{tag_name} eq 'textarea') { | ||
| !!!next-token; | ||
| if ($token->{type} eq 'character') { | ||
| $token->{data} =~ s/^\x0A//; | ||
| unless (length $token->{data}) { | ||
| !!!next-token; | ||
| } | ||
| } | ||
| } else { | ||
| !!!next-token; | ||
| } | ||
| while ($token->{type} eq 'character') { | ||
| $text .= $token->{data}; | ||
| 2029 | !!!next-token; | !!!next-token; |
| 2030 | next B; | |
| 2031 | } | } |
| 2032 | if (length $text) { | } elsif ($token->{type} == END_TAG_TOKEN) { |
| 2033 | $el->manakai_append_text ($text); | ## NOTE: "using the rules for secondary insertion mode" then "continue" |
| 2034 | } | if ($token->{tag_name} eq 'script') { |
| 2035 | !!!cp ('t87.41'); | |
| 2036 | $self->{content_model_flag} = 'PCDATA'; | # |
| 2037 | ## XXXscript: Execute script here. | |
| if ($token->{type} eq 'end tag' and | ||
| $token->{tag_name} eq $tag_name) { | ||
| ## Ignore the token | ||
| 2038 | } else { | } else { |
| 2039 | if ($token->{tag_name} eq 'textarea') { | !!!cp ('t87.5'); |
| 2040 | !!!parse-error (type => 'in RCDATA:#'.$token->{type}); | # |
| } else { | ||
| !!!parse-error (type => 'in CDATA:#'.$token->{type}); | ||
| } | ||
| ## ISSUE: And ignore? | ||
| 2041 | } | } |
| 2042 | !!!next-token; | } elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 2043 | return; | !!!cp ('t87.6'); |
| 2044 | } elsif ($token->{tag_name} eq 'select') { | !!!parse-error (type => 'not closed', |
| 2045 | $reconstruct_active_formatting_elements->($insert_to_current); | text => $self->{open_elements}->[-1]->[0] |
| 2046 | ->manakai_local_name, | |
| 2047 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | token => $token); |
| 2048 | ||
| 2049 | $self->{insertion_mode} = 'in select'; | pop @{$self->{open_elements}} |
| 2050 | !!!next-token; | while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; |
| 2051 | return; | |
| 2052 | } elsif ({ | ## NOTE: |<span><svg>| ... two parse errors, |<svg>| ... a parse error. |
| 2053 | caption => 1, col => 1, colgroup => 1, frame => 1, | |
| 2054 | frameset => 1, head => 1, option => 1, optgroup => 1, | $self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
| 2055 | tbody => 1, td => 1, tfoot => 1, th => 1, | ## Reprocess. |
| 2056 | thead => 1, tr => 1, | next B; |
| }->{$token->{tag_name}}) { | ||
| !!!parse-error (type => 'in body:'.$token->{tag_name}); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| return; | ||
| ## ISSUE: An issue on HTML5 new elements in the spec. | ||
| 2057 | } else { | } else { |
| 2058 | $reconstruct_active_formatting_elements->($insert_to_current); | die "$0: $token->{type}: Unknown token type"; |
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| !!!next-token; | ||
| return; | ||
| 2059 | } | } |
| 2060 | } elsif ($token->{type} eq 'end tag') { | } |
| if ($token->{tag_name} eq 'body') { | ||
| if (@{$self->{open_elements}} > 1 and | ||
| $self->{open_elements}->[1]->[1] eq 'body') { | ||
| for (@{$self->{open_elements}}) { | ||
| unless ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, td => 1, | ||
| th => 1, tr => 1, body => 1, html => 1, | ||
| }->{$_->[1]}) { | ||
| !!!parse-error (type => 'not closed:'.$_->[1]); | ||
| } | ||
| } | ||
| 2061 | ||
| 2062 | $self->{insertion_mode} = 'after body'; | if ($self->{insertion_mode} & HEAD_IMS) { |
| 2063 | !!!next-token; | if ($token->{type} == CHARACTER_TOKEN) { |
| 2064 | return; | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 2065 | } else { | unless ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2066 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | if ($self->{head_element_inserted}) { |
| 2067 | ## Ignore the token | !!!cp ('t88.3'); |
| 2068 | !!!next-token; | $self->{open_elements}->[-1]->[0]->append_child |
| 2069 | return; | ($self->{document}->create_text_node ($1)); |
| 2070 | } | delete $self->{head_element_inserted}; |
| 2071 | } elsif ($token->{tag_name} eq 'html') { | ## NOTE: |</head> <link> | |
| 2072 | if (@{$self->{open_elements}} > 1 and $self->{open_elements}->[1]->[1] eq 'body') { | # |
| 2073 | ## ISSUE: There is an issue in the spec. | } else { |
| 2074 | if ($self->{open_elements}->[-1]->[1] ne 'body') { | !!!cp ('t88.2'); |
| 2075 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[1]->[1]); | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 2076 | } | ## NOTE: |</head>  | |
| 2077 | $self->{insertion_mode} = 'after body'; | # |
| ## reprocess | ||
| return; | ||
| } else { | ||
| !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| return; | ||
| } | ||
| } elsif ({ | ||
| address => 1, blockquote => 1, center => 1, dir => 1, | ||
| div => 1, dl => 1, fieldset => 1, listing => 1, | ||
| menu => 1, ol => 1, pre => 1, ul => 1, | ||
| p => 1, | ||
| dd => 1, dt => 1, li => 1, | ||
| button => 1, marquee => 1, object => 1, | ||
| }->{$token->{tag_name}}) { | ||
| ## has an element in scope | ||
| my $i; | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ($node->[1] eq $token->{tag_name}) { | ||
| ## generate implied end tags | ||
| if ({ | ||
| dd => ($token->{tag_name} ne 'dd'), | ||
| dt => ($token->{tag_name} ne 'dt'), | ||
| li => ($token->{tag_name} ne 'li'), | ||
| p => ($token->{tag_name} ne 'p'), | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| return; | ||
| 2078 | } | } |
| 2079 | $i = $_; | } else { |
| 2080 | last INSCOPE unless $token->{tag_name} eq 'p'; | !!!cp ('t88.1'); |
| 2081 | } elsif ({ | ## Ignore the token. |
| 2082 | table => 1, caption => 1, td => 1, th => 1, | # |
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| 2083 | } | } |
| 2084 | } # INSCOPE | unless (length $token->{data}) { |
| 2085 | !!!cp ('t88'); | |
| 2086 | if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { | !!!next-token; |
| 2087 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | next B; |
| } | ||
| splice @{$self->{open_elements}}, $i if defined $i; | ||
| $clear_up_to_marker->() | ||
| if { | ||
| button => 1, marquee => 1, object => 1, | ||
| }->{$token->{tag_name}}; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'form') { | ||
| ## has an element in scope | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ($node->[1] eq $token->{tag_name}) { | ||
| ## generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| return; | ||
| } | ||
| last INSCOPE; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| 2088 | } | } |
| 2089 | } # INSCOPE | ## TODO: set $token->{column} appropriately |
| if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) { | ||
| pop @{$self->{open_elements}}; | ||
| } else { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| 2090 | } | } |
| 2091 | ||
| 2092 | undef $self->{form_element}; | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2093 | !!!next-token; | !!!cp ('t89'); |
| 2094 | return; | ## As if <head> |
| 2095 | } elsif ({ | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2096 | h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2097 | }->{$token->{tag_name}}) { | push @{$self->{open_elements}}, |
| 2098 | ## has an element in scope | [$self->{head_element}, $el_category->{head}]; |
| my $i; | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ({ | ||
| h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, | ||
| }->{$node->[1]}) { | ||
| ## generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| return; | ||
| } | ||
| $i = $_; | ||
| last INSCOPE; | ||
| } elsif ({ | ||
| table => 1, caption => 1, td => 1, th => 1, | ||
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| } | ||
| splice @{$self->{open_elements}}, $i if defined $i; | ||
| !!!next-token; | ||
| return; | ||
| } elsif ({ | ||
| a => 1, | ||
| b => 1, big => 1, em => 1, font => 1, i => 1, | ||
| nobr => 1, s => 1, small => 1, strile => 1, | ||
| strong => 1, tt => 1, u => 1, | ||
| }->{$token->{tag_name}}) { | ||
| $formatting_end_tag->($token->{tag_name}); | ||
| ## TODO: <http://html5.org/tools/web-apps-tracker?from=883&to=884> | ||
| return; | ||
| } elsif ({ | ||
| caption => 1, col => 1, colgroup => 1, frame => 1, | ||
| frameset => 1, head => 1, option => 1, optgroup => 1, | ||
| tbody => 1, td => 1, tfoot => 1, th => 1, | ||
| thead => 1, tr => 1, | ||
| area => 1, basefont => 1, bgsound => 1, br => 1, | ||
| embed => 1, hr => 1, iframe => 1, image => 1, | ||
| img => 1, input => 1, isindex => 1, noembed => 1, | ||
| noframes => 1, param => 1, select => 1, spacer => 1, | ||
| table => 1, textarea => 1, wbr => 1, | ||
| noscript => 0, ## TODO: if scripting is enabled | ||
| }->{$token->{tag_name}}) { | ||
| !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| return; | ||
| ## ISSUE: Issue on HTML5 new elements in spec | ||
| } else { | ||
| ## Step 1 | ||
| my $node_i = -1; | ||
| my $node = $self->{open_elements}->[$node_i]; | ||
| 2099 | ||
| 2100 | ## Step 2 | ## Reprocess in the "in head" insertion mode... |
| 2101 | S2: { | pop @{$self->{open_elements}}; |
| 2102 | if ($node->[1] eq $token->{tag_name}) { | |
| 2103 | ## Step 1 | ## Reprocess in the "after head" insertion mode... |
| 2104 | ## generate implied end tags | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2105 | if ({ | !!!cp ('t90'); |
| 2106 | dd => 1, dt => 1, li => 1, p => 1, | ## As if </noscript> |
| 2107 | td => 1, th => 1, tr => 1, | pop @{$self->{open_elements}}; |
| 2108 | }->{$self->{open_elements}->[-1]->[1]}) { | !!!parse-error (type => 'in noscript:#text', token => $token); |
| 2109 | !!!back-token; | |
| 2110 | $token = {type => 'end tag', | ## Reprocess in the "in head" insertion mode... |
| 2111 | tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ## As if </head> |
| 2112 | return; | pop @{$self->{open_elements}}; |
| 2113 | } | |
| 2114 | ## Reprocess in the "after head" insertion mode... | |
| 2115 | ## Step 2 | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2116 | if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { | !!!cp ('t91'); |
| 2117 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | pop @{$self->{open_elements}}; |
| 2118 | } | |
| 2119 | ## Reprocess in the "after head" insertion mode... | |
| 2120 | ## Step 3 | } else { |
| 2121 | splice @{$self->{open_elements}}, $node_i; | !!!cp ('t92'); |
| 2122 | } | |
| 2123 | ||
| 2124 | ## "after head" insertion mode | |
| 2125 | ## As if <body> | |
| 2126 | !!!insert-element ('body',, $token); | |
| 2127 | $self->{insertion_mode} = IN_BODY_IM; | |
| 2128 | ## reprocess | |
| 2129 | next B; | |
| 2130 | } elsif ($token->{type} == START_TAG_TOKEN) { | |
| 2131 | if ($token->{tag_name} eq 'head') { | |
| 2132 | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { | |
| 2133 | !!!cp ('t93'); | |
| 2134 | !!!create-element ($self->{head_element}, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); | |
| 2135 | $self->{open_elements}->[-1]->[0]->append_child | |
| 2136 | ($self->{head_element}); | |
| 2137 | push @{$self->{open_elements}}, | |
| 2138 | [$self->{head_element}, $el_category->{head}]; | |
| 2139 | $self->{insertion_mode} = IN_HEAD_IM; | |
| 2140 | !!!nack ('t93.1'); | |
| 2141 | !!!next-token; | !!!next-token; |
| 2142 | last S2; | next B; |
| 2143 | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2144 | !!!cp ('t93.2'); | |
| 2145 | !!!parse-error (type => 'after head', text => 'head', | |
| 2146 | token => $token); | |
| 2147 | ## Ignore the token | |
| 2148 | !!!nack ('t93.3'); | |
| 2149 | !!!next-token; | |
| 2150 | next B; | |
| 2151 | } else { | } else { |
| 2152 | ## Step 3 | !!!cp ('t95'); |
| 2153 | if (not $formatting_category->{$node->[1]} and | !!!parse-error (type => 'in head:head', |
| 2154 | #not $phrasing_category->{$node->[1]} and | token => $token); # or in head noscript |
| 2155 | ($special_category->{$node->[1]} or | ## Ignore the token |
| 2156 | $scoping_category->{$node->[1]})) { | !!!nack ('t95.1'); |
| 2157 | !!!parse-error (type => 'not closed:'.$node->[1]); | !!!next-token; |
| 2158 | ## Ignore the token | next B; |
| !!!next-token; | ||
| last S2; | ||
| } | ||
| 2159 | } | } |
| 2160 | } elsif ($self->{insertion_mode} == BEFORE_HEAD_IM) { | |
| 2161 | !!!cp ('t96'); | |
| 2162 | ## As if <head> | |
| 2163 | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); | |
| 2164 | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | |
| 2165 | push @{$self->{open_elements}}, | |
| 2166 | [$self->{head_element}, $el_category->{head}]; | |
| 2167 | ||
| 2168 | $self->{insertion_mode} = IN_HEAD_IM; | |
| 2169 | ## Reprocess in the "in head" insertion mode... | |
| 2170 | } else { | |
| 2171 | !!!cp ('t97'); | |
| 2172 | } | |
| 2173 | ||
| 2174 | if ($token->{tag_name} eq 'base') { | |
| 2175 | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2176 | !!!cp ('t98'); | |
| 2177 | ## As if </noscript> | |
| 2178 | pop @{$self->{open_elements}}; | |
| 2179 | !!!parse-error (type => 'in noscript', text => 'base', | |
| 2180 | token => $token); | |
| 2181 | ||
| 2182 | ## Step 4 | $self->{insertion_mode} = IN_HEAD_IM; |
| 2183 | $node_i--; | ## Reprocess in the "in head" insertion mode... |
| 2184 | $node = $self->{open_elements}->[$node_i]; | } else { |
| 2185 | !!!cp ('t99'); | |
| ## Step 5; | ||
| redo S2; | ||
| } # S2 | ||
| return; | ||
| } | ||
| } | ||
| }; # $in_body | ||
| B: { | ||
| if ($phase eq 'main') { | ||
| if ($token->{type} eq 'DOCTYPE') { | ||
| !!!parse-error (type => 'in html:#DOCTYPE'); | ||
| ## Ignore the token | ||
| ## Stay in the phase | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'start tag' and | ||
| $token->{tag_name} eq 'html') { | ||
| ## TODO: unless it is the first start tag token, parse-error | ||
| my $top_el = $self->{open_elements}->[0]->[0]; | ||
| for my $attr_name (keys %{$token->{attributes}}) { | ||
| unless ($top_el->has_attribute_ns (undef, $attr_name)) { | ||
| $top_el->set_attribute_ns | ||
| (undef, [undef, $attr_name], | ||
| $token->{attributes}->{$attr_name}->{value}); | ||
| 2186 | } | } |
| } | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'end-of-file') { | ||
| ## Generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', tag_name => $self->{open_elements}->[-1]->[1]}; | ||
| redo B; | ||
| } | ||
| if (@{$self->{open_elements}} > 2 or | ||
| (@{$self->{open_elements}} == 2 and $self->{open_elements}->[1]->[1] ne 'body')) { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| } elsif (defined $self->{inner_html_node} and | ||
| @{$self->{open_elements}} > 1 and | ||
| $self->{open_elements}->[1]->[1] ne 'body') { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| } | ||
| 2187 | ||
| 2188 | ## Stop parsing | ## NOTE: There is a "as if in head" code clone. |
| 2189 | last B; | if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2190 | !!!cp ('t100'); | |
| 2191 | !!!parse-error (type => 'after head', | |
| 2192 | text => $token->{tag_name}, token => $token); | |
| 2193 | push @{$self->{open_elements}}, | |
| 2194 | [$self->{head_element}, $el_category->{head}]; | |
| 2195 | $self->{head_element_inserted} = 1; | |
| 2196 | } else { | |
| 2197 | !!!cp ('t101'); | |
| 2198 | } | |
| 2199 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2200 | pop @{$self->{open_elements}}; | |
| 2201 | pop @{$self->{open_elements}} # <head> | |
| 2202 | if $self->{insertion_mode} == AFTER_HEAD_IM; | |
| 2203 | !!!nack ('t101.1'); | |
| 2204 | !!!next-token; | |
| 2205 | next B; | |
| 2206 | } elsif ($token->{tag_name} eq 'link') { | |
| 2207 | ## NOTE: There is a "as if in head" code clone. | |
| 2208 | if ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2209 | !!!cp ('t102'); | |
| 2210 | !!!parse-error (type => 'after head', | |
| 2211 | text => $token->{tag_name}, token => $token); | |
| 2212 | push @{$self->{open_elements}}, | |
| 2213 | [$self->{head_element}, $el_category->{head}]; | |
| 2214 | $self->{head_element_inserted} = 1; | |
| 2215 | } else { | |
| 2216 | !!!cp ('t103'); | |
| 2217 | } | |
| 2218 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2219 | pop @{$self->{open_elements}}; | |
| 2220 | pop @{$self->{open_elements}} # <head> | |
| 2221 | if $self->{insertion_mode} == AFTER_HEAD_IM; | |
| 2222 | !!!ack ('t103.1'); | |
| 2223 | !!!next-token; | |
| 2224 | next B; | |
| 2225 | } elsif ($token->{tag_name} eq 'command') { | |
| 2226 | if ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2227 | ## NOTE: If the insertion mode at the time of the emission | |
| 2228 | ## of the token was "before head", $self->{insertion_mode} | |
| 2229 | ## is already changed to |IN_HEAD_IM|. | |
| 2230 | ||
| 2231 | ## ISSUE: There is an issue in the spec. | ## NOTE: There is a "as if in head" code clone. |
| 2232 | } else { | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2233 | if ($self->{insertion_mode} eq 'before head') { | pop @{$self->{open_elements}}; |
| 2234 | if ($token->{type} eq 'character') { | pop @{$self->{open_elements}} # <head> |
| 2235 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2236 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | !!!ack ('t103.2'); |
| unless (length $token->{data}) { | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| } | ||
| ## As if <head> | ||
| !!!create-element ($self->{head_element}, 'head'); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | ||
| push @{$self->{open_elements}}, [$self->{head_element}, 'head']; | ||
| $self->{insertion_mode} = 'in head'; | ||
| ## reprocess | ||
| redo B; | ||
| } elsif ($token->{type} eq 'comment') { | ||
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| 2237 | !!!next-token; | !!!next-token; |
| 2238 | redo B; | next B; |
| } elsif ($token->{type} eq 'start tag') { | ||
| my $attr = $token->{tag_name} eq 'head' ? $token->{attributes} : {}; | ||
| !!!create-element ($self->{head_element}, 'head', $attr); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | ||
| push @{$self->{open_elements}}, [$self->{head_element}, 'head']; | ||
| $self->{insertion_mode} = 'in head'; | ||
| if ($token->{tag_name} eq 'head') { | ||
| !!!next-token; | ||
| #} elsif ({ | ||
| # base => 1, link => 1, meta => 1, | ||
| # script => 1, style => 1, title => 1, | ||
| # }->{$token->{tag_name}}) { | ||
| # ## reprocess | ||
| } else { | ||
| ## reprocess | ||
| } | ||
| redo B; | ||
| } elsif ($token->{type} eq 'end tag') { | ||
| if ({head => 1, body => 1, html => 1}->{$token->{tag_name}}) { | ||
| ## As if <head> | ||
| !!!create-element ($self->{head_element}, 'head'); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | ||
| push @{$self->{open_elements}}, [$self->{head_element}, 'head']; | ||
| $self->{insertion_mode} = 'in head'; | ||
| ## reprocess | ||
| redo B; | ||
| } else { | ||
| !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | ||
| ## Ignore the token ## ISSUE: An issue in the spec. | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| 2239 | } else { | } else { |
| 2240 | die "$0: $token->{type}: Unknown type"; | ## NOTE: "in head noscript" or "after head" insertion mode |
| 2241 | ## - in these cases, these tags are treated as same as | |
| 2242 | ## normal in-body tags. | |
| 2243 | !!!cp ('t103.3'); | |
| 2244 | # | |
| 2245 | } | } |
| 2246 | } elsif ($self->{insertion_mode} eq 'in head') { | } elsif ($token->{tag_name} eq 'meta') { |
| 2247 | if ($token->{type} eq 'character') { | ## NOTE: There is a "as if in head" code clone. |
| 2248 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2249 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | !!!cp ('t104'); |
| 2250 | unless (length $token->{data}) { | !!!parse-error (type => 'after head', |
| 2251 | !!!next-token; | text => $token->{tag_name}, token => $token); |
| 2252 | redo B; | push @{$self->{open_elements}}, |
| 2253 | [$self->{head_element}, $el_category->{head}]; | |
| 2254 | $self->{head_element_inserted} = 1; | |
| 2255 | } else { | |
| 2256 | !!!cp ('t105'); | |
| 2257 | } | |
| 2258 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2259 | my $meta_el = pop @{$self->{open_elements}}; | |
| 2260 | ||
| 2261 | unless ($self->{confident}) { | |
| 2262 | if ($token->{attributes}->{charset}) { | |
| 2263 | !!!cp ('t106'); | |
| 2264 | ## NOTE: Whether the encoding is supported or not is handled | |
| 2265 | ## in the {change_encoding} callback. | |
| 2266 | $self->{change_encoding} | |
| 2267 | ->($self, $token->{attributes}->{charset}->{value}, | |
| 2268 | $token); | |
| 2269 | ||
| 2270 | $meta_el->[0]->get_attribute_node_ns (undef, 'charset') | |
| 2271 | ->set_user_data (manakai_has_reference => | |
| 2272 | $token->{attributes}->{charset} | |
| 2273 | ->{has_reference}); | |
| 2274 | } elsif ($token->{attributes}->{content}) { | |
| 2275 | if ($token->{attributes}->{content}->{value} | |
| 2276 | =~ /[Cc][Hh][Aa][Rr][Ss][Ee][Tt] | |
| 2277 | [\x09\x0A\x0C\x0D\x20]*= | |
| 2278 | [\x09\x0A\x0C\x0D\x20]*(?>"([^"]*)"|'([^']*)'| | |
| 2279 | ([^"'\x09\x0A\x0C\x0D\x20] | |
| 2280 | [^\x09\x0A\x0C\x0D\x20\x3B]*))/x) { | |
| 2281 | !!!cp ('t107'); | |
| 2282 | ## NOTE: Whether the encoding is supported or not is handled | |
| 2283 | ## in the {change_encoding} callback. | |
| 2284 | $self->{change_encoding} | |
| 2285 | ->($self, defined $1 ? $1 : defined $2 ? $2 : $3, | |
| 2286 | $token); | |
| 2287 | $meta_el->[0]->get_attribute_node_ns (undef, 'content') | |
| 2288 | ->set_user_data (manakai_has_reference => | |
| 2289 | $token->{attributes}->{content} | |
| 2290 | ->{has_reference}); | |
| 2291 | } else { | |
| 2292 | !!!cp ('t108'); | |
| 2293 | } | |
| 2294 | } | |
| 2295 | } else { | |
| 2296 | if ($token->{attributes}->{charset}) { | |
| 2297 | !!!cp ('t109'); | |
| 2298 | $meta_el->[0]->get_attribute_node_ns (undef, 'charset') | |
| 2299 | ->set_user_data (manakai_has_reference => | |
| 2300 | $token->{attributes}->{charset} | |
| 2301 | ->{has_reference}); | |
| 2302 | } | |
| 2303 | if ($token->{attributes}->{content}) { | |
| 2304 | !!!cp ('t110'); | |
| 2305 | $meta_el->[0]->get_attribute_node_ns (undef, 'content') | |
| 2306 | ->set_user_data (manakai_has_reference => | |
| 2307 | $token->{attributes}->{content} | |
| 2308 | ->{has_reference}); | |
| 2309 | } | |
| 2310 | } | } |
| } | ||
| # | ||
| } elsif ($token->{type} eq 'comment') { | ||
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'start tag') { | ||
| if ($token->{tag_name} eq 'title') { | ||
| ## NOTE: There is an "as if in head" code clone | ||
| my $title_el; | ||
| !!!create-element ($title_el, 'title', $token->{attributes}); | ||
| (defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | ||
| ->append_child ($title_el); | ||
| $self->{content_model_flag} = 'RCDATA'; | ||
| delete $self->{escape}; # MUST | ||
| 2311 | ||
| 2312 | my $text = ''; | pop @{$self->{open_elements}} # <head> |
| 2313 | if $self->{insertion_mode} == AFTER_HEAD_IM; | |
| 2314 | !!!ack ('t110.1'); | |
| 2315 | !!!next-token; | !!!next-token; |
| 2316 | while ($token->{type} eq 'character') { | next B; |
| 2317 | $text .= $token->{data}; | } elsif ($token->{tag_name} eq 'title') { |
| 2318 | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2319 | !!!cp ('t111'); | |
| 2320 | ## As if </noscript> | |
| 2321 | pop @{$self->{open_elements}}; | |
| 2322 | !!!parse-error (type => 'in noscript', text => 'title', | |
| 2323 | token => $token); | |
| 2324 | ||
| 2325 | $self->{insertion_mode} = IN_HEAD_IM; | |
| 2326 | ## Reprocess in the "in head" insertion mode... | |
| 2327 | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2328 | !!!cp ('t112'); | |
| 2329 | !!!parse-error (type => 'after head', | |
| 2330 | text => $token->{tag_name}, token => $token); | |
| 2331 | push @{$self->{open_elements}}, | |
| 2332 | [$self->{head_element}, $el_category->{head}]; | |
| 2333 | $self->{head_element_inserted} = 1; | |
| 2334 | } else { | |
| 2335 | !!!cp ('t113'); | |
| 2336 | } | |
| 2337 | ||
| 2338 | ## NOTE: There is a "as if in head" code clone. | |
| 2339 | $parse_rcdata->(RCDATA_CONTENT_MODEL); | |
| 2340 | ||
| 2341 | ## NOTE: At this point the stack of open elements contain | |
| 2342 | ## the |head| element (index == -2) and the |script| element | |
| 2343 | ## (index == -1). In the "after head" insertion mode the | |
| 2344 | ## |head| element is inserted only for the purpose of | |
| 2345 | ## providing the context for the |script| element, and | |
| 2346 | ## therefore we can now and have to remove the element from | |
| 2347 | ## the stack. | |
| 2348 | splice @{$self->{open_elements}}, -2, 1, () # <head> | |
| 2349 | if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; | |
| 2350 | next B; | |
| 2351 | } elsif ($token->{tag_name} eq 'style' or | |
| 2352 | $token->{tag_name} eq 'noframes') { | |
| 2353 | ## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and | |
| 2354 | ## insertion mode IN_HEAD_IM) | |
| 2355 | ## NOTE: There is a "as if in head" code clone. | |
| 2356 | if ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2357 | !!!cp ('t114'); | |
| 2358 | !!!parse-error (type => 'after head', | |
| 2359 | text => $token->{tag_name}, token => $token); | |
| 2360 | push @{$self->{open_elements}}, | |
| 2361 | [$self->{head_element}, $el_category->{head}]; | |
| 2362 | $self->{head_element_inserted} = 1; | |
| 2363 | } else { | |
| 2364 | !!!cp ('t115'); | |
| 2365 | } | |
| 2366 | $parse_rcdata->(CDATA_CONTENT_MODEL); | |
| 2367 | ## ISSUE: A spec bug [Bug 6038] | |
| 2368 | splice @{$self->{open_elements}}, -2, 1, () # <head> | |
| 2369 | if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; | |
| 2370 | next B; | |
| 2371 | } elsif ($token->{tag_name} eq 'noscript') { | |
| 2372 | if ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2373 | !!!cp ('t116'); | |
| 2374 | ## NOTE: and scripting is disalbed | |
| 2375 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2376 | $self->{insertion_mode} = IN_HEAD_NOSCRIPT_IM; | |
| 2377 | !!!nack ('t116.1'); | |
| 2378 | !!!next-token; | !!!next-token; |
| 2379 | } | next B; |
| 2380 | if (length $text) { | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2381 | $title_el->manakai_append_text ($text); | !!!cp ('t117'); |
| 2382 | } | !!!parse-error (type => 'in noscript', text => 'noscript', |
| 2383 | token => $token); | |
| $self->{content_model_flag} = 'PCDATA'; | ||
| if ($token->{type} eq 'end tag' and | ||
| $token->{tag_name} eq 'title') { | ||
| 2384 | ## Ignore the token | ## Ignore the token |
| 2385 | !!!nack ('t117.1'); | |
| 2386 | !!!next-token; | |
| 2387 | next B; | |
| 2388 | } else { | } else { |
| 2389 | !!!parse-error (type => 'in RCDATA:#'.$token->{type}); | !!!cp ('t118'); |
| 2390 | ## ISSUE: And ignore? | # |
| 2391 | } | } |
| 2392 | !!!next-token; | } elsif ($token->{tag_name} eq 'script') { |
| 2393 | redo B; | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2394 | } elsif ($token->{tag_name} eq 'style') { | !!!cp ('t119'); |
| 2395 | $style_start_tag->(); | ## As if </noscript> |
| 2396 | redo B; | pop @{$self->{open_elements}}; |
| 2397 | } elsif ($token->{tag_name} eq 'script') { | !!!parse-error (type => 'in noscript', text => 'script', |
| 2398 | $script_start_tag->(); | token => $token); |
| 2399 | redo B; | |
| 2400 | } elsif ({base => 1, link => 1, meta => 1}->{$token->{tag_name}}) { | $self->{insertion_mode} = IN_HEAD_IM; |
| 2401 | ## NOTE: There are "as if in head" code clones | ## Reprocess in the "in head" insertion mode... |
| 2402 | my $el; | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2403 | !!!create-element ($el, $token->{tag_name}, $token->{attributes}); | !!!cp ('t120'); |
| 2404 | if ($self->{insertion_mode} eq 'in head' and | !!!parse-error (type => 'after head', |
| 2405 | defined $self->{head_element}) { | text => $token->{tag_name}, token => $token); |
| 2406 | $self->{head_element}->append_child ($el); | push @{$self->{open_elements}}, |
| 2407 | [$self->{head_element}, $el_category->{head}]; | |
| 2408 | $self->{head_element_inserted} = 1; | |
| 2409 | } else { | |
| 2410 | !!!cp ('t121'); | |
| 2411 | } | |
| 2412 | ||
| 2413 | ## NOTE: There is a "as if in head" code clone. | |
| 2414 | $script_start_tag->(); | |
| 2415 | ## ISSUE: A spec bug [Bug 6038] | |
| 2416 | splice @{$self->{open_elements}}, -2, 1 # <head> | |
| 2417 | if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; | |
| 2418 | next B; | |
| 2419 | } elsif ($token->{tag_name} eq 'body' or | |
| 2420 | $token->{tag_name} eq 'frameset') { | |
| 2421 | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2422 | !!!cp ('t122'); | |
| 2423 | ## As if </noscript> | |
| 2424 | pop @{$self->{open_elements}}; | |
| 2425 | !!!parse-error (type => 'in noscript', | |
| 2426 | text => $token->{tag_name}, token => $token); | |
| 2427 | ||
| 2428 | ## Reprocess in the "in head" insertion mode... | |
| 2429 | ## As if </head> | |
| 2430 | pop @{$self->{open_elements}}; | |
| 2431 | ||
| 2432 | ## Reprocess in the "after head" insertion mode... | |
| 2433 | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2434 | !!!cp ('t124'); | |
| 2435 | pop @{$self->{open_elements}}; | |
| 2436 | ||
| 2437 | ## Reprocess in the "after head" insertion mode... | |
| 2438 | } else { | } else { |
| 2439 | $self->{open_elements}->[-1]->[0]->append_child ($el); | !!!cp ('t125'); |
| 2440 | } | } |
| 2441 | ||
| 2442 | ## "after head" insertion mode | |
| 2443 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2444 | if ($token->{tag_name} eq 'body') { | |
| 2445 | !!!cp ('t126'); | |
| 2446 | $self->{insertion_mode} = IN_BODY_IM; | |
| 2447 | } elsif ($token->{tag_name} eq 'frameset') { | |
| 2448 | !!!cp ('t127'); | |
| 2449 | $self->{insertion_mode} = IN_FRAMESET_IM; | |
| 2450 | } else { | |
| 2451 | die "$0: tag name: $self->{tag_name}"; | |
| 2452 | } | |
| 2453 | !!!nack ('t127.1'); | |
| 2454 | !!!next-token; | !!!next-token; |
| 2455 | redo B; | next B; |
| } elsif ($token->{tag_name} eq 'head') { | ||
| !!!parse-error (type => 'in head:head'); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| redo B; | ||
| 2456 | } else { | } else { |
| 2457 | !!!cp ('t128'); | |
| 2458 | # | # |
| 2459 | } | } |
| 2460 | } elsif ($token->{type} eq 'end tag') { | |
| 2461 | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2462 | !!!cp ('t129'); | |
| 2463 | ## As if </noscript> | |
| 2464 | pop @{$self->{open_elements}}; | |
| 2465 | !!!parse-error (type => 'in noscript:/', | |
| 2466 | text => $token->{tag_name}, token => $token); | |
| 2467 | ||
| 2468 | ## Reprocess in the "in head" insertion mode... | |
| 2469 | ## As if </head> | |
| 2470 | pop @{$self->{open_elements}}; | |
| 2471 | ||
| 2472 | ## Reprocess in the "after head" insertion mode... | |
| 2473 | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2474 | !!!cp ('t130'); | |
| 2475 | ## As if </head> | |
| 2476 | pop @{$self->{open_elements}}; | |
| 2477 | ||
| 2478 | ## Reprocess in the "after head" insertion mode... | |
| 2479 | } else { | |
| 2480 | !!!cp ('t131'); | |
| 2481 | } | |
| 2482 | ||
| 2483 | ## "after head" insertion mode | |
| 2484 | ## As if <body> | |
| 2485 | !!!insert-element ('body',, $token); | |
| 2486 | $self->{insertion_mode} = IN_BODY_IM; | |
| 2487 | ## reprocess | |
| 2488 | !!!ack-later; | |
| 2489 | next B; | |
| 2490 | } elsif ($token->{type} == END_TAG_TOKEN) { | |
| 2491 | if ($token->{tag_name} eq 'head') { | if ($token->{tag_name} eq 'head') { |
| 2492 | if ($self->{open_elements}->[-1]->[1] eq 'head') { | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2493 | !!!cp ('t132'); | |
| 2494 | ## As if <head> | |
| 2495 | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); | |
| 2496 | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | |
| 2497 | push @{$self->{open_elements}}, | |
| 2498 | [$self->{head_element}, $el_category->{head}]; | |
| 2499 | ||
| 2500 | ## Reprocess in the "in head" insertion mode... | |
| 2501 | pop @{$self->{open_elements}}; | |
| 2502 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 2503 | !!!next-token; | |
| 2504 | next B; | |
| 2505 | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2506 | !!!cp ('t133'); | |
| 2507 | ## As if </noscript> | |
| 2508 | pop @{$self->{open_elements}}; | pop @{$self->{open_elements}}; |
| 2509 | !!!parse-error (type => 'in noscript:/', | |
| 2510 | text => 'head', token => $token); | |
| 2511 | ||
| 2512 | ## Reprocess in the "in head" insertion mode... | |
| 2513 | pop @{$self->{open_elements}}; | |
| 2514 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 2515 | !!!next-token; | |
| 2516 | next B; | |
| 2517 | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2518 | !!!cp ('t134'); | |
| 2519 | pop @{$self->{open_elements}}; | |
| 2520 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 2521 | !!!next-token; | |
| 2522 | next B; | |
| 2523 | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2524 | !!!cp ('t134.1'); | |
| 2525 | !!!parse-error (type => 'unmatched end tag', text => 'head', | |
| 2526 | token => $token); | |
| 2527 | ## Ignore the token | |
| 2528 | !!!next-token; | |
| 2529 | next B; | |
| 2530 | } else { | } else { |
| 2531 | !!!parse-error (type => 'unmatched end tag:head'); | die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2532 | } | } |
| 2533 | $self->{insertion_mode} = 'after head'; | } elsif ($token->{tag_name} eq 'noscript') { |
| 2534 | !!!next-token; | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2535 | redo B; | !!!cp ('t136'); |
| 2536 | } elsif ($token->{tag_name} eq 'body' or | pop @{$self->{open_elements}}; |
| 2537 | $token->{tag_name} eq 'html') { | $self->{insertion_mode} = IN_HEAD_IM; |
| 2538 | # | !!!next-token; |
| 2539 | } else { | next B; |
| 2540 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | } elsif ($self->{insertion_mode} == BEFORE_HEAD_IM or |
| 2541 | $self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2542 | !!!cp ('t137'); | |
| 2543 | !!!parse-error (type => 'unmatched end tag', | |
| 2544 | text => 'noscript', token => $token); | |
| 2545 | ## Ignore the token ## ISSUE: An issue in the spec. | |
| 2546 | !!!next-token; | |
| 2547 | next B; | |
| 2548 | } else { | |
| 2549 | !!!cp ('t138'); | |
| 2550 | # | |
| 2551 | } | |
| 2552 | } elsif ({ | |
| 2553 | body => 1, html => 1, | |
| 2554 | }->{$token->{tag_name}}) { | |
| 2555 | ## TODO: This branch is entirely redundant. | |
| 2556 | if ($self->{insertion_mode} == BEFORE_HEAD_IM or | |
| 2557 | $self->{insertion_mode} == IN_HEAD_IM or | |
| 2558 | $self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2559 | !!!cp ('t140'); | |
| 2560 | !!!parse-error (type => 'unmatched end tag', | |
| 2561 | text => $token->{tag_name}, token => $token); | |
| 2562 | ## Ignore the token | |
| 2563 | !!!next-token; | |
| 2564 | next B; | |
| 2565 | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2566 | !!!cp ('t140.1'); | |
| 2567 | !!!parse-error (type => 'unmatched end tag', | |
| 2568 | text => $token->{tag_name}, token => $token); | |
| 2569 | ## Ignore the token | |
| 2570 | !!!next-token; | |
| 2571 | next B; | |
| 2572 | } else { | |
| 2573 | die "$0: $self->{insertion_mode}: Unknown insertion mode"; | |
| 2574 | } | |
| 2575 | } elsif ($token->{tag_name} eq 'p') { | |
| 2576 | !!!cp ('t142'); | |
| 2577 | !!!parse-error (type => 'unmatched end tag', | |
| 2578 | text => $token->{tag_name}, token => $token); | |
| 2579 | ## Ignore the token | ## Ignore the token |
| 2580 | !!!next-token; | !!!next-token; |
| 2581 | redo B; | next B; |
| 2582 | } | } elsif ($token->{tag_name} eq 'br') { |
| 2583 | } else { | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2584 | # | !!!cp ('t142.2'); |
| 2585 | } | ## (before head) as if <head>, (in head) as if </head> |
| 2586 | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); | |
| 2587 | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | |
| 2588 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 2589 | ||
| 2590 | ## Reprocess in the "after head" insertion mode... | |
| 2591 | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2592 | !!!cp ('t143.2'); | |
| 2593 | ## As if </head> | |
| 2594 | pop @{$self->{open_elements}}; | |
| 2595 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 2596 | ||
| 2597 | ## Reprocess in the "after head" insertion mode... | |
| 2598 | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2599 | !!!cp ('t143.3'); | |
| 2600 | ## NOTE: Two parse errors for <head><noscript></br> | |
| 2601 | !!!parse-error (type => 'unmatched end tag', | |
| 2602 | text => 'br', token => $token); | |
| 2603 | ## As if </noscript> | |
| 2604 | pop @{$self->{open_elements}}; | |
| 2605 | $self->{insertion_mode} = IN_HEAD_IM; | |
| 2606 | ||
| 2607 | if ($self->{open_elements}->[-1]->[1] eq 'head') { | ## Reprocess in the "in head" insertion mode... |
| 2608 | ## As if </head> | ## As if </head> |
| 2609 | pop @{$self->{open_elements}}; | pop @{$self->{open_elements}}; |
| 2610 | } | $self->{insertion_mode} = AFTER_HEAD_IM; |
| $self->{insertion_mode} = 'after head'; | ||
| ## reprocess | ||
| redo B; | ||
| 2611 | ||
| 2612 | ## ISSUE: An issue in the spec. | ## Reprocess in the "after head" insertion mode... |
| 2613 | } elsif ($self->{insertion_mode} eq 'after head') { | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2614 | if ($token->{type} eq 'character') { | !!!cp ('t143.4'); |
| if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | ||
| $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | ||
| unless (length $token->{data}) { | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| } | ||
| 2615 | # | # |
| 2616 | } elsif ($token->{type} eq 'comment') { | } else { |
| 2617 | my $comment = $self->{document}->create_comment ($token->{data}); | die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2618 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | } |
| 2619 | !!!next-token; | |
| 2620 | redo B; | # |
| 2621 | } elsif ($token->{type} eq 'start tag') { | } else { ## Other end tags |
| 2622 | if ($token->{tag_name} eq 'body') { | !!!cp ('t145'); |
| 2623 | !!!insert-element ('body', $token->{attributes}); | !!!parse-error (type => 'unmatched end tag', |
| 2624 | $self->{insertion_mode} = 'in body'; | text => $token->{tag_name}, token => $token); |
| 2625 | ## Ignore the token | |
| 2626 | !!!next-token; | !!!next-token; |
| 2627 | redo B; | next B; |
| 2628 | } elsif ($token->{tag_name} eq 'frameset') { | } |
| 2629 | !!!insert-element ('frameset', $token->{attributes}); | |
| 2630 | $self->{insertion_mode} = 'in frameset'; | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2631 | !!!cp ('t146'); | |
| 2632 | ## As if </noscript> | |
| 2633 | pop @{$self->{open_elements}}; | |
| 2634 | !!!parse-error (type => 'in noscript:/', | |
| 2635 | text => $token->{tag_name}, token => $token); | |
| 2636 | ||
| 2637 | ## Reprocess in the "in head" insertion mode... | |
| 2638 | ## As if </head> | |
| 2639 | pop @{$self->{open_elements}}; | |
| 2640 | ||
| 2641 | ## Reprocess in the "after head" insertion mode... | |
| 2642 | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2643 | !!!cp ('t147'); | |
| 2644 | ## As if </head> | |
| 2645 | pop @{$self->{open_elements}}; | |
| 2646 | ||
| 2647 | ## Reprocess in the "after head" insertion mode... | |
| 2648 | } elsif ($self->{insertion_mode} == BEFORE_HEAD_IM) { | |
| 2649 | ## ISSUE: This case cannot be reached? | |
| 2650 | !!!cp ('t148'); | |
| 2651 | !!!parse-error (type => 'unmatched end tag', | |
| 2652 | text => $token->{tag_name}, token => $token); | |
| 2653 | ## Ignore the token ## ISSUE: An issue in the spec. | |
| 2654 | !!!next-token; | !!!next-token; |
| 2655 | redo B; | next B; |
| } elsif ({ | ||
| base => 1, link => 1, meta => 1, | ||
| script => 1, style => 1, title => 1, | ||
| }->{$token->{tag_name}}) { | ||
| !!!parse-error (type => 'after head:'.$token->{tag_name}); | ||
| $self->{insertion_mode} = 'in head'; | ||
| ## reprocess | ||
| redo B; | ||
| 2656 | } else { | } else { |
| 2657 | # | !!!cp ('t149'); |
| 2658 | } | } |
| 2659 | } else { | |
| 2660 | # | ## "after head" insertion mode |
| 2661 | } | ## As if <body> |
| 2662 | !!!insert-element ('body',, $token); | |
| 2663 | $self->{insertion_mode} = IN_BODY_IM; | |
| 2664 | ## reprocess | |
| 2665 | next B; | |
| 2666 | } elsif ($token->{type} == END_OF_FILE_TOKEN) { | |
| 2667 | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { | |
| 2668 | !!!cp ('t149.1'); | |
| 2669 | ||
| 2670 | ## NOTE: As if <head> | |
| 2671 | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); | |
| 2672 | $self->{open_elements}->[-1]->[0]->append_child | |
| 2673 | ($self->{head_element}); | |
| 2674 | #push @{$self->{open_elements}}, | |
| 2675 | # [$self->{head_element}, $el_category->{head}]; | |
| 2676 | #$self->{insertion_mode} = IN_HEAD_IM; | |
| 2677 | ## NOTE: Reprocess. | |
| 2678 | ||
| 2679 | ## NOTE: As if </head> | |
| 2680 | #pop @{$self->{open_elements}}; | |
| 2681 | #$self->{insertion_mode} = IN_AFTER_HEAD_IM; | |
| 2682 | ## NOTE: Reprocess. | |
| 2683 | ||
| 2684 | ## As if <body> | # |
| 2685 | !!!insert-element ('body'); | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2686 | $self->{insertion_mode} = 'in body'; | !!!cp ('t149.2'); |
| 2687 | ## reprocess | |
| 2688 | redo B; | ## NOTE: As if </head> |
| 2689 | } elsif ($self->{insertion_mode} eq 'in body') { | pop @{$self->{open_elements}}; |
| 2690 | if ($token->{type} eq 'character') { | #$self->{insertion_mode} = IN_AFTER_HEAD_IM; |
| 2691 | ## NOTE: Reprocess. | |
| 2692 | ||
| 2693 | # | |
| 2694 | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2695 | !!!cp ('t149.3'); | |
| 2696 | ||
| 2697 | !!!parse-error (type => 'in noscript:#eof', token => $token); | |
| 2698 | ||
| 2699 | ## As if </noscript> | |
| 2700 | pop @{$self->{open_elements}}; | |
| 2701 | #$self->{insertion_mode} = IN_HEAD_IM; | |
| 2702 | ## NOTE: Reprocess. | |
| 2703 | ||
| 2704 | ## NOTE: As if </head> | |
| 2705 | pop @{$self->{open_elements}}; | |
| 2706 | #$self->{insertion_mode} = IN_AFTER_HEAD_IM; | |
| 2707 | ## NOTE: Reprocess. | |
| 2708 | ||
| 2709 | # | |
| 2710 | } else { | |
| 2711 | !!!cp ('t149.4'); | |
| 2712 | # | |
| 2713 | } | |
| 2714 | ||
| 2715 | ## NOTE: As if <body> | |
| 2716 | !!!insert-element ('body',, $token); | |
| 2717 | $self->{insertion_mode} = IN_BODY_IM; | |
| 2718 | ## NOTE: Reprocess. | |
| 2719 | next B; | |
| 2720 | } else { | |
| 2721 | die "$0: $token->{type}: Unknown token type"; | |
| 2722 | } | |
| 2723 | } elsif ($self->{insertion_mode} & BODY_IMS) { | |
| 2724 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 2725 | !!!cp ('t150'); | |
| 2726 | ## NOTE: There is a code clone of "character in body". | ## NOTE: There is a code clone of "character in body". |
| 2727 | $reconstruct_active_formatting_elements->($insert_to_current); | $reconstruct_active_formatting_elements->($insert_to_current); |
| 2728 | ||
| 2729 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 2730 | ||
| 2731 | !!!next-token; | !!!next-token; |
| 2732 | redo B; | next B; |
| 2733 | } elsif ($token->{type} eq 'comment') { | } elsif ($token->{type} == START_TAG_TOKEN) { |
| 2734 | ## NOTE: There is a code clone of "comment in body". | if ({ |
| 2735 | my $comment = $self->{document}->create_comment ($token->{data}); | caption => 1, col => 1, colgroup => 1, tbody => 1, |
| 2736 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
| 2737 | !!!next-token; | }->{$token->{tag_name}}) { |
| 2738 | redo B; | if (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2739 | } else { | ## have an element in table scope |
| 2740 | $in_body->($insert_to_current); | for (reverse 0..$#{$self->{open_elements}}) { |
| 2741 | redo B; | my $node = $self->{open_elements}->[$_]; |
| 2742 | } | if ($node->[1] == TABLE_CELL_EL) { |
| 2743 | } elsif ($self->{insertion_mode} eq 'in table') { | !!!cp ('t151'); |
| 2744 | if ($token->{type} eq 'character') { | |
| 2745 | ## NOTE: There are "character in table" code clones. | ## Close the cell |
| 2746 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | !!!back-token; # <x> |
| 2747 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | $token = {type => END_TAG_TOKEN, |
| 2748 | tag_name => $node->[0]->manakai_local_name, | |
| 2749 | unless (length $token->{data}) { | line => $token->{line}, |
| 2750 | column => $token->{column}}; | |
| 2751 | next B; | |
| 2752 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 2753 | !!!cp ('t152'); | |
| 2754 | ## ISSUE: This case can never be reached, maybe. | |
| 2755 | last; | |
| 2756 | } | |
| 2757 | } | |
| 2758 | ||
| 2759 | !!!cp ('t153'); | |
| 2760 | !!!parse-error (type => 'start tag not allowed', | |
| 2761 | text => $token->{tag_name}, token => $token); | |
| 2762 | ## Ignore the token | |
| 2763 | !!!nack ('t153.1'); | |
| 2764 | !!!next-token; | !!!next-token; |
| 2765 | redo B; | next B; |
| 2766 | } | } elsif (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2767 | } | !!!parse-error (type => 'not closed', text => 'caption', |
| 2768 | token => $token); | |
| 2769 | ||
| 2770 | ## NOTE: As if </caption>. | |
| 2771 | ## have a table element in table scope | |
| 2772 | my $i; | |
| 2773 | INSCOPE: { | |
| 2774 | for (reverse 0..$#{$self->{open_elements}}) { | |
| 2775 | my $node = $self->{open_elements}->[$_]; | |
| 2776 | if ($node->[1] == CAPTION_EL) { | |
| 2777 | !!!cp ('t155'); | |
| 2778 | $i = $_; | |
| 2779 | last INSCOPE; | |
| 2780 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 2781 | !!!cp ('t156'); | |
| 2782 | last; | |
| 2783 | } | |
| 2784 | } | |
| 2785 | ||
| 2786 | !!!parse-error (type => 'in table:#character'); | !!!cp ('t157'); |
| 2787 | !!!parse-error (type => 'start tag not allowed', | |
| 2788 | text => $token->{tag_name}, token => $token); | |
| 2789 | ## Ignore the token | |
| 2790 | !!!nack ('t157.1'); | |
| 2791 | !!!next-token; | |
| 2792 | next B; | |
| 2793 | } # INSCOPE | |
| 2794 | ||
| 2795 | ## generate implied end tags | |
| 2796 | while ($self->{open_elements}->[-1]->[1] | |
| 2797 | & END_TAG_OPTIONAL_EL) { | |
| 2798 | !!!cp ('t158'); | |
| 2799 | pop @{$self->{open_elements}}; | |
| 2800 | } | |
| 2801 | ||
| 2802 | ## As if in body, but insert into foster parent element | unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2803 | ## ISSUE: Spec says that "whenever a node would be inserted | !!!cp ('t159'); |
| 2804 | ## into the current node" while characters might not be | !!!parse-error (type => 'not closed', |
| 2805 | ## result in a new Text node. | text => $self->{open_elements}->[-1]->[0] |
| 2806 | $reconstruct_active_formatting_elements->($insert_to_foster); | ->manakai_local_name, |
| 2807 | token => $token); | |
| 2808 | if ({ | } else { |
| 2809 | table => 1, tbody => 1, tfoot => 1, | !!!cp ('t160'); |
| thead => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| # MUST | ||
| my $foster_parent_element; | ||
| my $next_sibling; | ||
| my $prev_sibling; | ||
| OE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| if ($self->{open_elements}->[$_]->[1] eq 'table') { | ||
| my $parent = $self->{open_elements}->[$_]->[0]->parent_node; | ||
| if (defined $parent and $parent->node_type == 1) { | ||
| $foster_parent_element = $parent; | ||
| $next_sibling = $self->{open_elements}->[$_]->[0]; | ||
| $prev_sibling = $next_sibling->previous_sibling; | ||
| } else { | ||
| $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; | ||
| $prev_sibling = $foster_parent_element->last_child; | ||
| } | ||
| last OE; | ||
| 2810 | } | } |
| 2811 | } # OE | |
| 2812 | $foster_parent_element = $self->{open_elements}->[0]->[0] and | splice @{$self->{open_elements}}, $i; |
| 2813 | $prev_sibling = $foster_parent_element->last_child | |
| 2814 | unless defined $foster_parent_element; | $clear_up_to_marker->(); |
| 2815 | if (defined $prev_sibling and | |
| 2816 | $prev_sibling->node_type == 3) { | $self->{insertion_mode} = IN_TABLE_IM; |
| 2817 | $prev_sibling->manakai_append_text ($token->{data}); | |
| 2818 | ## reprocess | |
| 2819 | !!!ack-later; | |
| 2820 | next B; | |
| 2821 | } else { | } else { |
| 2822 | $foster_parent_element->insert_before | !!!cp ('t161'); |
| 2823 | ($self->{document}->create_text_node ($token->{data}), | # |
| $next_sibling); | ||
| 2824 | } | } |
| 2825 | } else { | } else { |
| 2826 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | !!!cp ('t162'); |
| 2827 | # | |
| 2828 | } | } |
| 2829 | } elsif ($token->{type} == END_TAG_TOKEN) { | |
| 2830 | !!!next-token; | if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
| 2831 | redo B; | if (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2832 | } elsif ($token->{type} eq 'comment') { | ## have an element in table scope |
| 2833 | my $comment = $self->{document}->create_comment ($token->{data}); | my $i; |
| 2834 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2835 | !!!next-token; | my $node = $self->{open_elements}->[$_]; |
| 2836 | redo B; | if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 2837 | } elsif ($token->{type} eq 'start tag') { | !!!cp ('t163'); |
| 2838 | if ({ | $i = $_; |
| 2839 | caption => 1, | last INSCOPE; |
| 2840 | colgroup => 1, | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2841 | tbody => 1, tfoot => 1, thead => 1, | !!!cp ('t164'); |
| 2842 | }->{$token->{tag_name}}) { | last INSCOPE; |
| 2843 | ## Clear back to table context | } |
| 2844 | while ($self->{open_elements}->[-1]->[1] ne 'table' and | } # INSCOPE |
| 2845 | $self->{open_elements}->[-1]->[1] ne 'html') { | unless (defined $i) { |
| 2846 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | !!!cp ('t165'); |
| 2847 | pop @{$self->{open_elements}}; | !!!parse-error (type => 'unmatched end tag', |
| 2848 | } | text => $token->{tag_name}, |
| 2849 | token => $token); | |
| 2850 | ## Ignore the token | |
| 2851 | !!!next-token; | |
| 2852 | next B; | |
| 2853 | } | |
| 2854 | ||
| 2855 | ## generate implied end tags | |
| 2856 | while ($self->{open_elements}->[-1]->[1] | |
| 2857 | & END_TAG_OPTIONAL_EL) { | |
| 2858 | !!!cp ('t166'); | |
| 2859 | pop @{$self->{open_elements}}; | |
| 2860 | } | |
| 2861 | ||
| 2862 | push @$active_formatting_elements, ['#marker', ''] | if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
| 2863 | if $token->{tag_name} eq 'caption'; | ne $token->{tag_name}) { |
| 2864 | !!!cp ('t167'); | |
| 2865 | !!!parse-error (type => 'not closed', | |
| 2866 | text => $self->{open_elements}->[-1]->[0] | |
| 2867 | ->manakai_local_name, | |
| 2868 | token => $token); | |
| 2869 | } else { | |
| 2870 | !!!cp ('t168'); | |
| 2871 | } | |
| 2872 | ||
| 2873 | splice @{$self->{open_elements}}, $i; | |
| 2874 | ||
| 2875 | $clear_up_to_marker->(); | |
| 2876 | ||
| 2877 | $self->{insertion_mode} = IN_ROW_IM; | |
| 2878 | ||
| 2879 | !!!next-token; | |
| 2880 | next B; | |
| 2881 | } elsif (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { | |
| 2882 | !!!cp ('t169'); | |
| 2883 | !!!parse-error (type => 'unmatched end tag', | |
| 2884 | text => $token->{tag_name}, token => $token); | |
| 2885 | ## Ignore the token | |
| 2886 | !!!next-token; | |
| 2887 | next B; | |
| 2888 | } else { | |
| 2889 | !!!cp ('t170'); | |
| 2890 | # | |
| 2891 | } | |
| 2892 | } elsif ($token->{tag_name} eq 'caption') { | |
| 2893 | if (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { | |
| 2894 | ## have a table element in table scope | |
| 2895 | my $i; | |
| 2896 | INSCOPE: { | |
| 2897 | for (reverse 0..$#{$self->{open_elements}}) { | |
| 2898 | my $node = $self->{open_elements}->[$_]; | |
| 2899 | if ($node->[1] == CAPTION_EL) { | |
| 2900 | !!!cp ('t171'); | |
| 2901 | $i = $_; | |
| 2902 | last INSCOPE; | |
| 2903 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 2904 | !!!cp ('t172'); | |
| 2905 | last; | |
| 2906 | } | |
| 2907 | } | |
| 2908 | ||
| 2909 | !!!insert-element ($token->{tag_name}, $token->{attributes}); | !!!cp ('t173'); |
| 2910 | $self->{insertion_mode} = { | !!!parse-error (type => 'unmatched end tag', |
| 2911 | caption => 'in caption', | text => $token->{tag_name}, token => $token); |
| 2912 | colgroup => 'in column group', | ## Ignore the token |
| 2913 | tbody => 'in table body', | !!!next-token; |
| 2914 | tfoot => 'in table body', | next B; |
| 2915 | thead => 'in table body', | } # INSCOPE |
| 2916 | }->{$token->{tag_name}}; | |
| 2917 | !!!next-token; | ## generate implied end tags |
| 2918 | redo B; | while ($self->{open_elements}->[-1]->[1] |
| 2919 | } elsif ({ | & END_TAG_OPTIONAL_EL) { |
| 2920 | col => 1, | !!!cp ('t174'); |
| 2921 | td => 1, th => 1, tr => 1, | pop @{$self->{open_elements}}; |
| 2922 | }->{$token->{tag_name}}) { | } |
| 2923 | ## Clear back to table context | |
| 2924 | while ($self->{open_elements}->[-1]->[1] ne 'table' and | unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2925 | $self->{open_elements}->[-1]->[1] ne 'html') { | !!!cp ('t175'); |
| 2926 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | !!!parse-error (type => 'not closed', |
| 2927 | pop @{$self->{open_elements}}; | text => $self->{open_elements}->[-1]->[0] |
| 2928 | ->manakai_local_name, | |
| 2929 | token => $token); | |
| 2930 | } else { | |
| 2931 | !!!cp ('t176'); | |
| 2932 | } | |
| 2933 | ||
| 2934 | splice @{$self->{open_elements}}, $i; | |
| 2935 | ||
| 2936 | $clear_up_to_marker->(); | |
| 2937 | ||
| 2938 | $self->{insertion_mode} = IN_TABLE_IM; | |
| 2939 | ||
| 2940 | !!!next-token; | |
| 2941 | next B; | |
| 2942 | } elsif (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { | |
| 2943 | !!!cp ('t177'); | |
| 2944 | !!!parse-error (type => 'unmatched end tag', | |
| 2945 | text => $token->{tag_name}, token => $token); | |
| 2946 | ## Ignore the token | |
| 2947 | !!!next-token; | |
| 2948 | next B; | |
| 2949 | } else { | |
| 2950 | !!!cp ('t178'); | |
| 2951 | # | |
| 2952 | } | } |
| 2953 | } elsif ({ | |
| 2954 | table => 1, tbody => 1, tfoot => 1, | |
| 2955 | thead => 1, tr => 1, | |
| 2956 | }->{$token->{tag_name}} and | |
| 2957 | ($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { | |
| 2958 | ## have an element in table scope | |
| 2959 | my $i; | |
| 2960 | my $tn; | |
| 2961 | INSCOPE: { | |
| 2962 | for (reverse 0..$#{$self->{open_elements}}) { | |
| 2963 | my $node = $self->{open_elements}->[$_]; | |
| 2964 | if ($node->[0]->manakai_local_name eq $token->{tag_name}) { | |
| 2965 | !!!cp ('t179'); | |
| 2966 | $i = $_; | |
| 2967 | ||
| 2968 | ## Close the cell | |
| 2969 | !!!back-token; # </x> | |
| 2970 | $token = {type => END_TAG_TOKEN, tag_name => $tn, | |
| 2971 | line => $token->{line}, | |
| 2972 | column => $token->{column}}; | |
| 2973 | next B; | |
| 2974 | } elsif ($node->[1] == TABLE_CELL_EL) { | |
| 2975 | !!!cp ('t180'); | |
| 2976 | $tn = $node->[0]->manakai_local_name; | |
| 2977 | ## NOTE: There is exactly one |td| or |th| element | |
| 2978 | ## in scope in the stack of open elements by definition. | |
| 2979 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 2980 | ## ISSUE: Can this be reached? | |
| 2981 | !!!cp ('t181'); | |
| 2982 | last; | |
| 2983 | } | |
| 2984 | } | |
| 2985 | ||
| 2986 | !!!insert-element ($token->{tag_name} eq 'col' ? 'colgroup' : 'tbody'); | !!!cp ('t182'); |
| 2987 | $self->{insertion_mode} = $token->{tag_name} eq 'col' | !!!parse-error (type => 'unmatched end tag', |
| 2988 | ? 'in column group' : 'in table body'; | text => $token->{tag_name}, token => $token); |
| 2989 | ## reprocess | ## Ignore the token |
| 2990 | redo B; | !!!next-token; |
| 2991 | } elsif ($token->{tag_name} eq 'table') { | next B; |
| 2992 | ## NOTE: There are code clones for this "table in table" | } # INSCOPE |
| 2993 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | } elsif ($token->{tag_name} eq 'table' and |
| 2994 | ($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { | |
| 2995 | !!!parse-error (type => 'not closed', text => 'caption', | |
| 2996 | token => $token); | |
| 2997 | ||
| 2998 | ## As if </table> | ## As if </caption> |
| 2999 | ## have a table element in table scope | ## have a table element in table scope |
| 3000 | my $i; | my $i; |
| 3001 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3002 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 3003 | if ($node->[1] eq 'table') { | if ($node->[1] == CAPTION_EL) { |
| 3004 | !!!cp ('t184'); | |
| 3005 | $i = $_; | $i = $_; |
| 3006 | last INSCOPE; | last INSCOPE; |
| 3007 | } elsif ({ | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3008 | table => 1, html => 1, | !!!cp ('t185'); |
| }->{$node->[1]}) { | ||
| 3009 | last INSCOPE; | last INSCOPE; |
| 3010 | } | } |
| 3011 | } # INSCOPE | } # INSCOPE |
| 3012 | unless (defined $i) { | unless (defined $i) { |
| 3013 | !!!parse-error (type => 'unmatched end tag:table'); | !!!cp ('t186'); |
| 3014 | ## Ignore tokens </table><table> | ## TODO: Wrong error type? |
| 3015 | !!!parse-error (type => 'unmatched end tag', | |
| 3016 | text => 'caption', token => $token); | |
| 3017 | ## Ignore the token | |
| 3018 | !!!next-token; | !!!next-token; |
| 3019 | redo B; | next B; |
| 3020 | } | } |
| 3021 | ||
| 3022 | ## generate implied end tags | ## generate implied end tags |
| 3023 | if ({ | while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 3024 | dd => 1, dt => 1, li => 1, p => 1, | !!!cp ('t187'); |
| 3025 | td => 1, th => 1, tr => 1, | pop @{$self->{open_elements}}; |
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; # <table> | ||
| $token = {type => 'end tag', tag_name => 'table'}; | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| redo B; | ||
| 3026 | } | } |
| 3027 | ||
| 3028 | if ($self->{open_elements}->[-1]->[1] ne 'table') { | unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 3029 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | !!!cp ('t188'); |
| 3030 | !!!parse-error (type => 'not closed', | |
| 3031 | text => $self->{open_elements}->[-1]->[0] | |
| 3032 | ->manakai_local_name, | |
| 3033 | token => $token); | |
| 3034 | } else { | |
| 3035 | !!!cp ('t189'); | |
| 3036 | } | } |
| 3037 | ||
| 3038 | splice @{$self->{open_elements}}, $i; | splice @{$self->{open_elements}}, $i; |
| 3039 | ||
| 3040 | $self->_reset_insertion_mode; | $clear_up_to_marker->(); |
| 3041 | ||
| 3042 | $self->{insertion_mode} = IN_TABLE_IM; | |
| 3043 | ||
| 3044 | ## reprocess | ## reprocess |
| 3045 | redo B; | next B; |
| 3046 | } elsif ({ | |
| 3047 | body => 1, col => 1, colgroup => 1, html => 1, | |
| 3048 | }->{$token->{tag_name}}) { | |
| 3049 | if ($self->{insertion_mode} & BODY_TABLE_IMS) { | |
| 3050 | !!!cp ('t190'); | |
| 3051 | !!!parse-error (type => 'unmatched end tag', | |
| 3052 | text => $token->{tag_name}, token => $token); | |
| 3053 | ## Ignore the token | |
| 3054 | !!!next-token; | |
| 3055 | next B; | |
| 3056 | } else { | |
| 3057 | !!!cp ('t191'); | |
| 3058 | # | |
| 3059 | } | |
| 3060 | } elsif ({ | |
| 3061 | tbody => 1, tfoot => 1, | |
| 3062 | thead => 1, tr => 1, | |
| 3063 | }->{$token->{tag_name}} and | |
| 3064 | ($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { | |
| 3065 | !!!cp ('t192'); | |
| 3066 | !!!parse-error (type => 'unmatched end tag', | |
| 3067 | text => $token->{tag_name}, token => $token); | |
| 3068 | ## Ignore the token | |
| 3069 | !!!next-token; | |
| 3070 | next B; | |
| 3071 | } else { | |
| 3072 | !!!cp ('t193'); | |
| 3073 | # | |
| 3074 | } | |
| 3075 | } elsif ($token->{type} == END_OF_FILE_TOKEN) { | |
| 3076 | for my $entry (@{$self->{open_elements}}) { | |
| 3077 | unless ($entry->[1] & ALL_END_TAG_OPTIONAL_EL) { | |
| 3078 | !!!cp ('t75'); | |
| 3079 | !!!parse-error (type => 'in body:#eof', token => $token); | |
| 3080 | last; | |
| 3081 | } | |
| 3082 | } | |
| 3083 | ||
| 3084 | ## Stop parsing. | |
| 3085 | last B; | |
| 3086 | } else { | |
| 3087 | die "$0: $token->{type}: Unknown token type"; | |
| 3088 | } | |
| 3089 | ||
| 3090 | $insert = $insert_to_current; | |
| 3091 | # | |
| 3092 | } elsif ($self->{insertion_mode} & TABLE_IMS) { | |
| 3093 | if ($token->{type} == START_TAG_TOKEN) { | |
| 3094 | if ({ | |
| 3095 | tr => (($self->{insertion_mode} & IM_MASK) != IN_ROW_IM), | |
| 3096 | th => 1, td => 1, | |
| 3097 | }->{$token->{tag_name}}) { | |
| 3098 | if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_IM) { | |
| 3099 | ## Clear back to table context | |
| 3100 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3101 | & TABLE_SCOPING_EL)) { | |
| 3102 | !!!cp ('t201'); | |
| 3103 | pop @{$self->{open_elements}}; | |
| 3104 | } | |
| 3105 | ||
| 3106 | !!!insert-element ('tbody',, $token); | |
| 3107 | $self->{insertion_mode} = IN_TABLE_BODY_IM; | |
| 3108 | ## reprocess in the "in table body" insertion mode... | |
| 3109 | } | |
| 3110 | ||
| 3111 | if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { | |
| 3112 | unless ($token->{tag_name} eq 'tr') { | |
| 3113 | !!!cp ('t202'); | |
| 3114 | !!!parse-error (type => 'missing start tag:tr', token => $token); | |
| 3115 | } | |
| 3116 | ||
| 3117 | ## Clear back to table body context | |
| 3118 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3119 | & TABLE_ROWS_SCOPING_EL)) { | |
| 3120 | !!!cp ('t203'); | |
| 3121 | ## ISSUE: Can this case be reached? | |
| 3122 | pop @{$self->{open_elements}}; | |
| 3123 | } | |
| 3124 | ||
| 3125 | $self->{insertion_mode} = IN_ROW_IM; | |
| 3126 | if ($token->{tag_name} eq 'tr') { | |
| 3127 | !!!cp ('t204'); | |
| 3128 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 3129 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3130 | !!!nack ('t204'); | |
| 3131 | !!!next-token; | |
| 3132 | next B; | |
| 3133 | } else { | } else { |
| 3134 | # | !!!cp ('t205'); |
| 3135 | !!!insert-element ('tr',, $token); | |
| 3136 | ## reprocess in the "in row" insertion mode | |
| 3137 | } | |
| 3138 | } else { | |
| 3139 | !!!cp ('t206'); | |
| 3140 | } | |
| 3141 | ||
| 3142 | ## Clear back to table row context | |
| 3143 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3144 | & TABLE_ROW_SCOPING_EL)) { | |
| 3145 | !!!cp ('t207'); | |
| 3146 | pop @{$self->{open_elements}}; | |
| 3147 | } | |
| 3148 | ||
| 3149 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 3150 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3151 | $self->{insertion_mode} = IN_CELL_IM; | |
| 3152 | ||
| 3153 | push @$active_formatting_elements, ['#marker', '']; | |
| 3154 | ||
| 3155 | !!!nack ('t207.1'); | |
| 3156 | !!!next-token; | |
| 3157 | next B; | |
| 3158 | } elsif ({ | |
| 3159 | caption => 1, col => 1, colgroup => 1, | |
| 3160 | tbody => 1, tfoot => 1, thead => 1, | |
| 3161 | tr => 1, # $self->{insertion_mode} == IN_ROW_IM | |
| 3162 | }->{$token->{tag_name}}) { | |
| 3163 | if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { | |
| 3164 | ## As if </tr> | |
| 3165 | ## have an element in table scope | |
| 3166 | my $i; | |
| 3167 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3168 | my $node = $self->{open_elements}->[$_]; | |
| 3169 | if ($node->[1] == TABLE_ROW_EL) { | |
| 3170 | !!!cp ('t208'); | |
| 3171 | $i = $_; | |
| 3172 | last INSCOPE; | |
| 3173 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3174 | !!!cp ('t209'); | |
| 3175 | last INSCOPE; | |
| 3176 | } | |
| 3177 | } # INSCOPE | |
| 3178 | unless (defined $i) { | |
| 3179 | !!!cp ('t210'); | |
| 3180 | ## TODO: This type is wrong. | |
| 3181 | !!!parse-error (type => 'unmacthed end tag', | |
| 3182 | text => $token->{tag_name}, token => $token); | |
| 3183 | ## Ignore the token | |
| 3184 | !!!nack ('t210.1'); | |
| 3185 | !!!next-token; | |
| 3186 | next B; | |
| 3187 | } | |
| 3188 | ||
| 3189 | ## Clear back to table row context | |
| 3190 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3191 | & TABLE_ROW_SCOPING_EL)) { | |
| 3192 | !!!cp ('t211'); | |
| 3193 | ## ISSUE: Can this case be reached? | |
| 3194 | pop @{$self->{open_elements}}; | |
| 3195 | } | |
| 3196 | ||
| 3197 | pop @{$self->{open_elements}}; # tr | |
| 3198 | $self->{insertion_mode} = IN_TABLE_BODY_IM; | |
| 3199 | if ($token->{tag_name} eq 'tr') { | |
| 3200 | !!!cp ('t212'); | |
| 3201 | ## reprocess | |
| 3202 | !!!ack-later; | |
| 3203 | next B; | |
| 3204 | } else { | |
| 3205 | !!!cp ('t213'); | |
| 3206 | ## reprocess in the "in table body" insertion mode... | |
| 3207 | } | |
| 3208 | } | |
| 3209 | ||
| 3210 | if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { | |
| 3211 | ## have an element in table scope | |
| 3212 | my $i; | |
| 3213 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3214 | my $node = $self->{open_elements}->[$_]; | |
| 3215 | if ($node->[1] == TABLE_ROW_GROUP_EL) { | |
| 3216 | !!!cp ('t214'); | |
| 3217 | $i = $_; | |
| 3218 | last INSCOPE; | |
| 3219 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3220 | !!!cp ('t215'); | |
| 3221 | last INSCOPE; | |
| 3222 | } | |
| 3223 | } # INSCOPE | |
| 3224 | unless (defined $i) { | |
| 3225 | !!!cp ('t216'); | |
| 3226 | ## TODO: This erorr type is wrong. | |
| 3227 | !!!parse-error (type => 'unmatched end tag', | |
| 3228 | text => $token->{tag_name}, token => $token); | |
| 3229 | ## Ignore the token | |
| 3230 | !!!nack ('t216.1'); | |
| 3231 | !!!next-token; | |
| 3232 | next B; | |
| 3233 | } | |
| 3234 | ||
| 3235 | ## Clear back to table body context | |
| 3236 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3237 | & TABLE_ROWS_SCOPING_EL)) { | |
| 3238 | !!!cp ('t217'); | |
| 3239 | ## ISSUE: Can this state be reached? | |
| 3240 | pop @{$self->{open_elements}}; | |
| 3241 | } | |
| 3242 | ||
| 3243 | ## As if <{current node}> | |
| 3244 | ## have an element in table scope | |
| 3245 | ## true by definition | |
| 3246 | ||
| 3247 | ## Clear back to table body context | |
| 3248 | ## nop by definition | |
| 3249 | ||
| 3250 | pop @{$self->{open_elements}}; | |
| 3251 | $self->{insertion_mode} = IN_TABLE_IM; | |
| 3252 | ## reprocess in "in table" insertion mode... | |
| 3253 | } else { | |
| 3254 | !!!cp ('t218'); | |
| 3255 | } | |
| 3256 | ||
| 3257 | if ($token->{tag_name} eq 'col') { | |
| 3258 | ## Clear back to table context | |
| 3259 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3260 | & TABLE_SCOPING_EL)) { | |
| 3261 | !!!cp ('t219'); | |
| 3262 | ## ISSUE: Can this state be reached? | |
| 3263 | pop @{$self->{open_elements}}; | |
| 3264 | } | } |
| 3265 | } elsif ($token->{type} eq 'end tag') { | |
| 3266 | if ($token->{tag_name} eq 'table') { | !!!insert-element ('colgroup',, $token); |
| 3267 | $self->{insertion_mode} = IN_COLUMN_GROUP_IM; | |
| 3268 | ## reprocess | |
| 3269 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3270 | !!!ack-later; | |
| 3271 | next B; | |
| 3272 | } elsif ({ | |
| 3273 | caption => 1, | |
| 3274 | colgroup => 1, | |
| 3275 | tbody => 1, tfoot => 1, thead => 1, | |
| 3276 | }->{$token->{tag_name}}) { | |
| 3277 | ## Clear back to table context | |
| 3278 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3279 | & TABLE_SCOPING_EL)) { | |
| 3280 | !!!cp ('t220'); | |
| 3281 | ## ISSUE: Can this state be reached? | |
| 3282 | pop @{$self->{open_elements}}; | |
| 3283 | } | |
| 3284 | ||
| 3285 | push @$active_formatting_elements, ['#marker', ''] | |
| 3286 | if $token->{tag_name} eq 'caption'; | |
| 3287 | ||
| 3288 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 3289 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3290 | $self->{insertion_mode} = { | |
| 3291 | caption => IN_CAPTION_IM, | |
| 3292 | colgroup => IN_COLUMN_GROUP_IM, | |
| 3293 | tbody => IN_TABLE_BODY_IM, | |
| 3294 | tfoot => IN_TABLE_BODY_IM, | |
| 3295 | thead => IN_TABLE_BODY_IM, | |
| 3296 | }->{$token->{tag_name}}; | |
| 3297 | !!!next-token; | |
| 3298 | !!!nack ('t220.1'); | |
| 3299 | next B; | |
| 3300 | } else { | |
| 3301 | die "$0: in table: <>: $token->{tag_name}"; | |
| 3302 | } | |
| 3303 | } elsif ($token->{tag_name} eq 'table') { | |
| 3304 | !!!parse-error (type => 'not closed', | |
| 3305 | text => $self->{open_elements}->[-1]->[0] | |
| 3306 | ->manakai_local_name, | |
| 3307 | token => $token); | |
| 3308 | ||
| 3309 | ## As if </table> | |
| 3310 | ## have a table element in table scope | ## have a table element in table scope |
| 3311 | my $i; | my $i; |
| 3312 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3313 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 3314 | if ($node->[1] eq $token->{tag_name}) { | if ($node->[1] == TABLE_EL) { |
| 3315 | !!!cp ('t221'); | |
| 3316 | $i = $_; | $i = $_; |
| 3317 | last INSCOPE; | last INSCOPE; |
| 3318 | } elsif ({ | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3319 | table => 1, html => 1, | !!!cp ('t222'); |
| }->{$node->[1]}) { | ||
| 3320 | last INSCOPE; | last INSCOPE; |
| 3321 | } | } |
| 3322 | } # INSCOPE | } # INSCOPE |
| 3323 | unless (defined $i) { | unless (defined $i) { |
| 3324 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t223'); |
| 3325 | ## Ignore the token | ## TODO: The following is wrong, maybe. |
| 3326 | !!!parse-error (type => 'unmatched end tag', text => 'table', | |
| 3327 | token => $token); | |
| 3328 | ## Ignore tokens </table><table> | |
| 3329 | !!!nack ('t223.1'); | |
| 3330 | !!!next-token; | !!!next-token; |
| 3331 | redo B; | next B; |
| 3332 | } | } |
| 3333 | ||
| 3334 | ## TODO: Followings are removed from the latest spec. | |
| 3335 | ## generate implied end tags | ## generate implied end tags |
| 3336 | if ({ | while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 3337 | dd => 1, dt => 1, li => 1, p => 1, | !!!cp ('t224'); |
| 3338 | td => 1, th => 1, tr => 1, | pop @{$self->{open_elements}}; |
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| redo B; | ||
| 3339 | } | } |
| 3340 | ||
| 3341 | if ($self->{open_elements}->[-1]->[1] ne 'table') { | unless ($self->{open_elements}->[-1]->[1] == TABLE_EL) { |
| 3342 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | !!!cp ('t225'); |
| 3343 | ## NOTE: |<table><tr><table>| | |
| 3344 | !!!parse-error (type => 'not closed', | |
| 3345 | text => $self->{open_elements}->[-1]->[0] | |
| 3346 | ->manakai_local_name, | |
| 3347 | token => $token); | |
| 3348 | } else { | |
| 3349 | !!!cp ('t226'); | |
| 3350 | } | } |
| 3351 | ||
| 3352 | splice @{$self->{open_elements}}, $i; | splice @{$self->{open_elements}}, $i; |
| 3353 | pop @{$open_tables}; | |
| 3354 | ||
| 3355 | $self->_reset_insertion_mode; | $self->_reset_insertion_mode; |
| 3356 | ||
| 3357 | ## reprocess | |
| 3358 | !!!ack-later; | |
| 3359 | next B; | |
| 3360 | } elsif ($token->{tag_name} eq 'style') { | |
| 3361 | !!!cp ('t227.8'); | |
| 3362 | ## NOTE: This is a "as if in head" code clone. | |
| 3363 | $parse_rcdata->(CDATA_CONTENT_MODEL); | |
| 3364 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3365 | next B; | |
| 3366 | } elsif ($token->{tag_name} eq 'script') { | |
| 3367 | !!!cp ('t227.6'); | |
| 3368 | ## NOTE: This is a "as if in head" code clone. | |
| 3369 | $script_start_tag->(); | |
| 3370 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3371 | next B; | |
| 3372 | } elsif ($token->{tag_name} eq 'input') { | |
| 3373 | if ($token->{attributes}->{type}) { | |
| 3374 | my $type = $token->{attributes}->{type}->{value}; | |
| 3375 | $type =~ tr/A-Z/a-z/; ## ASCII case-insensitive. | |
| 3376 | if ($type eq 'hidden') { | |
| 3377 | !!!cp ('t227.3'); | |
| 3378 | !!!parse-error (type => 'in table', | |
| 3379 | text => $token->{tag_name}, token => $token); | |
| 3380 | ||
| 3381 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 3382 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 3383 | ||
| 3384 | ## TODO: form element pointer | |
| 3385 | ||
| 3386 | pop @{$self->{open_elements}}; | |
| 3387 | ||
| 3388 | !!!next-token; | !!!next-token; |
| 3389 | redo B; | !!!ack ('t227.2.1'); |
| 3390 | } elsif ({ | next B; |
| body => 1, caption => 1, col => 1, colgroup => 1, | ||
| html => 1, tbody => 1, td => 1, tfoot => 1, th => 1, | ||
| thead => 1, tr => 1, | ||
| }->{$token->{tag_name}}) { | ||
| !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| redo B; | ||
| 3391 | } else { | } else { |
| 3392 | !!!cp ('t227.1'); | |
| 3393 | # | # |
| 3394 | } | } |
| 3395 | } else { | } else { |
| 3396 | !!!cp ('t227.4'); | |
| 3397 | # | # |
| 3398 | } | } |
| 3399 | } else { | |
| 3400 | !!!cp ('t227'); | |
| 3401 | # | |
| 3402 | } | |
| 3403 | ||
| 3404 | !!!parse-error (type => 'in table:'.$token->{tag_name}); | !!!parse-error (type => 'in table', text => $token->{tag_name}, |
| 3405 | $in_body->($insert_to_foster); | token => $token); |
| redo B; | ||
| } elsif ($self->{insertion_mode} eq 'in caption') { | ||
| if ($token->{type} eq 'character') { | ||
| ## NOTE: This is a code clone of "character in body". | ||
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'comment') { | ||
| ## NOTE: This is a code clone of "comment in body". | ||
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'start tag') { | ||
| if ({ | ||
| caption => 1, col => 1, colgroup => 1, tbody => 1, | ||
| td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, | ||
| }->{$token->{tag_name}}) { | ||
| !!!parse-error (type => 'not closed:caption'); | ||
| 3406 | ||
| 3407 | ## As if </caption> | $insert = $insert_to_foster; |
| 3408 | ## have a table element in table scope | # |
| 3409 | } elsif ($token->{type} == END_TAG_TOKEN) { | |
| 3410 | if ($token->{tag_name} eq 'tr' and | |
| 3411 | ($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { | |
| 3412 | ## have an element in table scope | |
| 3413 | my $i; | my $i; |
| 3414 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3415 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 3416 | if ($node->[1] eq 'caption') { | if ($node->[1] == TABLE_ROW_EL) { |
| 3417 | !!!cp ('t228'); | |
| 3418 | $i = $_; | $i = $_; |
| 3419 | last INSCOPE; | last INSCOPE; |
| 3420 | } elsif ({ | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3421 | table => 1, html => 1, | !!!cp ('t229'); |
| }->{$node->[1]}) { | ||
| 3422 | last INSCOPE; | last INSCOPE; |
| 3423 | } | } |
| 3424 | } # INSCOPE | } # INSCOPE |
| 3425 | unless (defined $i) { | unless (defined $i) { |
| 3426 | !!!parse-error (type => 'unmatched end tag:caption'); | !!!cp ('t230'); |
| 3427 | !!!parse-error (type => 'unmatched end tag', | |
| 3428 | text => $token->{tag_name}, token => $token); | |
| 3429 | ## Ignore the token | ## Ignore the token |
| 3430 | !!!nack ('t230.1'); | |
| 3431 | !!!next-token; | !!!next-token; |
| 3432 | redo B; | next B; |
| 3433 | } | } else { |
| 3434 | !!!cp ('t232'); | |
| ## generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; # <?> | ||
| $token = {type => 'end tag', tag_name => 'caption'}; | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| redo B; | ||
| 3435 | } | } |
| 3436 | ||
| 3437 | if ($self->{open_elements}->[-1]->[1] ne 'caption') { | ## Clear back to table row context |
| 3438 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | while (not ($self->{open_elements}->[-1]->[1] |
| 3439 | & TABLE_ROW_SCOPING_EL)) { | |
| 3440 | !!!cp ('t231'); | |
| 3441 | ## ISSUE: Can this state be reached? | |
| 3442 | pop @{$self->{open_elements}}; | |
| 3443 | } | } |
| 3444 | ||
| 3445 | splice @{$self->{open_elements}}, $i; | pop @{$self->{open_elements}}; # tr |
| 3446 | $self->{insertion_mode} = IN_TABLE_BODY_IM; | |
| 3447 | $clear_up_to_marker->(); | !!!next-token; |
| 3448 | !!!nack ('t231.1'); | |
| 3449 | next B; | |
| 3450 | } elsif ($token->{tag_name} eq 'table') { | |
| 3451 | if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { | |
| 3452 | ## As if </tr> | |
| 3453 | ## have an element in table scope | |
| 3454 | my $i; | |
| 3455 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3456 | my $node = $self->{open_elements}->[$_]; | |
| 3457 | if ($node->[1] == TABLE_ROW_EL) { | |
| 3458 | !!!cp ('t233'); | |
| 3459 | $i = $_; | |
| 3460 | last INSCOPE; | |
| 3461 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3462 | !!!cp ('t234'); | |
| 3463 | last INSCOPE; | |
| 3464 | } | |
| 3465 | } # INSCOPE | |
| 3466 | unless (defined $i) { | |
| 3467 | !!!cp ('t235'); | |
| 3468 | ## TODO: The following is wrong. | |
| 3469 | !!!parse-error (type => 'unmatched end tag', | |
| 3470 | text => $token->{type}, token => $token); | |
| 3471 | ## Ignore the token | |
| 3472 | !!!nack ('t236.1'); | |
| 3473 | !!!next-token; | |
| 3474 | next B; | |
| 3475 | } | |
| 3476 | ||
| 3477 | ## Clear back to table row context | |
| 3478 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3479 | & TABLE_ROW_SCOPING_EL)) { | |
| 3480 | !!!cp ('t236'); | |
| 3481 | ## ISSUE: Can this state be reached? | |
| 3482 | pop @{$self->{open_elements}}; | |
| 3483 | } | |
| 3484 | ||
| 3485 | pop @{$self->{open_elements}}; # tr | |
| 3486 | $self->{insertion_mode} = IN_TABLE_BODY_IM; | |
| 3487 | ## reprocess in the "in table body" insertion mode... | |
| 3488 | } | |
| 3489 | ||
| 3490 | if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { | |
| 3491 | ## have an element in table scope | |
| 3492 | my $i; | |
| 3493 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3494 | my $node = $self->{open_elements}->[$_]; | |
| 3495 | if ($node->[1] == TABLE_ROW_GROUP_EL) { | |
| 3496 | !!!cp ('t237'); | |
| 3497 | $i = $_; | |
| 3498 | last INSCOPE; | |
| 3499 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3500 | !!!cp ('t238'); | |
| 3501 | last INSCOPE; | |
| 3502 | } | |
| 3503 | } # INSCOPE | |
| 3504 | unless (defined $i) { | |
| 3505 | !!!cp ('t239'); | |
| 3506 | !!!parse-error (type => 'unmatched end tag', | |
| 3507 | text => $token->{tag_name}, token => $token); | |
| 3508 | ## Ignore the token | |
| 3509 | !!!nack ('t239.1'); | |
| 3510 | !!!next-token; | |
| 3511 | next B; | |
| 3512 | } | |
| 3513 | ||
| 3514 | ## Clear back to table body context | |
| 3515 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3516 | & TABLE_ROWS_SCOPING_EL)) { | |
| 3517 | !!!cp ('t240'); | |
| 3518 | pop @{$self->{open_elements}}; | |
| 3519 | } | |
| 3520 | ||
| 3521 | ## As if <{current node}> | |
| 3522 | ## have an element in table scope | |
| 3523 | ## true by definition | |
| 3524 | ||
| 3525 | ## Clear back to table body context | |
| 3526 | ## nop by definition | |
| 3527 | ||
| 3528 | pop @{$self->{open_elements}}; | |
| 3529 | $self->{insertion_mode} = IN_TABLE_IM; | |
| 3530 | ## reprocess in the "in table" insertion mode... | |
| 3531 | } | |
| 3532 | ||
| 3533 | $self->{insertion_mode} = 'in table'; | ## NOTE: </table> in the "in table" insertion mode. |
| 3534 | ## When you edit the code fragment below, please ensure that | |
| 3535 | ## the code for <table> in the "in table" insertion mode | |
| 3536 | ## is synced with it. | |
| 3537 | ||
| ## reprocess | ||
| redo B; | ||
| } else { | ||
| # | ||
| } | ||
| } elsif ($token->{type} eq 'end tag') { | ||
| if ($token->{tag_name} eq 'caption') { | ||
| 3538 | ## have a table element in table scope | ## have a table element in table scope |
| 3539 | my $i; | my $i; |
| 3540 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3541 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 3542 | if ($node->[1] eq $token->{tag_name}) { | if ($node->[1] == TABLE_EL) { |
| 3543 | !!!cp ('t241'); | |
| 3544 | $i = $_; | $i = $_; |
| 3545 | last INSCOPE; | last INSCOPE; |
| 3546 | } elsif ({ | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3547 | table => 1, html => 1, | !!!cp ('t242'); |
| }->{$node->[1]}) { | ||
| 3548 | last INSCOPE; | last INSCOPE; |
| 3549 | } | } |
| 3550 | } # INSCOPE | } # INSCOPE |
| 3551 | unless (defined $i) { | unless (defined $i) { |
| 3552 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t243'); |
| 3553 | !!!parse-error (type => 'unmatched end tag', | |
| 3554 | text => $token->{tag_name}, token => $token); | |
| 3555 | ## Ignore the token | ## Ignore the token |
| 3556 | !!!nack ('t243.1'); | |
| 3557 | !!!next-token; | !!!next-token; |
| 3558 | redo B; | next B; |
| } | ||
| ## generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| redo B; | ||
| 3559 | } | } |
| 3560 | ||
| if ($self->{open_elements}->[-1]->[1] ne 'caption') { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| } | ||
| 3561 | splice @{$self->{open_elements}}, $i; | splice @{$self->{open_elements}}, $i; |
| 3562 | pop @{$open_tables}; | |
| 3563 | $clear_up_to_marker->(); | |
| 3564 | $self->_reset_insertion_mode; | |
| 3565 | $self->{insertion_mode} = 'in table'; | |
| 3566 | !!!next-token; | !!!next-token; |
| 3567 | redo B; | next B; |
| 3568 | } elsif ($token->{tag_name} eq 'table') { | } elsif ({ |
| 3569 | !!!parse-error (type => 'not closed:caption'); | tbody => 1, tfoot => 1, thead => 1, |
| 3570 | }->{$token->{tag_name}} and | |
| 3571 | $self->{insertion_mode} & ROW_IMS) { | |
| 3572 | if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { | |
| 3573 | ## have an element in table scope | |
| 3574 | my $i; | |
| 3575 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3576 | my $node = $self->{open_elements}->[$_]; | |
| 3577 | if ($node->[0]->manakai_local_name eq $token->{tag_name}) { | |
| 3578 | !!!cp ('t247'); | |
| 3579 | $i = $_; | |
| 3580 | last INSCOPE; | |
| 3581 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3582 | !!!cp ('t248'); | |
| 3583 | last INSCOPE; | |
| 3584 | } | |
| 3585 | } # INSCOPE | |
| 3586 | unless (defined $i) { | |
| 3587 | !!!cp ('t249'); | |
| 3588 | !!!parse-error (type => 'unmatched end tag', | |
| 3589 | text => $token->{tag_name}, token => $token); | |
| 3590 | ## Ignore the token | |
| 3591 | !!!nack ('t249.1'); | |
| 3592 | !!!next-token; | |
| 3593 | next B; | |
| 3594 | } | |
| 3595 | ||
| 3596 | ## As if </tr> | |
| 3597 | ## have an element in table scope | |
| 3598 | my $i; | |
| 3599 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3600 | my $node = $self->{open_elements}->[$_]; | |
| 3601 | if ($node->[1] == TABLE_ROW_EL) { | |
| 3602 | !!!cp ('t250'); | |
| 3603 | $i = $_; | |
| 3604 | last INSCOPE; | |
| 3605 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3606 | !!!cp ('t251'); | |
| 3607 | last INSCOPE; | |
| 3608 | } | |
| 3609 | } # INSCOPE | |
| 3610 | unless (defined $i) { | |
| 3611 | !!!cp ('t252'); | |
| 3612 | !!!parse-error (type => 'unmatched end tag', | |
| 3613 | text => 'tr', token => $token); | |
| 3614 | ## Ignore the token | |
| 3615 | !!!nack ('t252.1'); | |
| 3616 | !!!next-token; | |
| 3617 | next B; | |
| 3618 | } | |
| 3619 | ||
| 3620 | ## Clear back to table row context | |
| 3621 | while (not ($self->{open_elements}->[-1]->[1] | |
| 3622 | & TABLE_ROW_SCOPING_EL)) { | |
| 3623 | !!!cp ('t253'); | |
| 3624 | ## ISSUE: Can this case be reached? | |
| 3625 | pop @{$self->{open_elements}}; | |
| 3626 | } | |
| 3627 | ||
| 3628 | pop @{$self->{open_elements}}; # tr | |
| 3629 | $self->{insertion_mode} = IN_TABLE_BODY_IM; | |
| 3630 | ## reprocess in the "in table body" insertion mode... | |
| 3631 | } | |
| 3632 | ||
| 3633 | ## As if </caption> | ## have an element in table scope |
| ## have a table element in table scope | ||
| 3634 | my $i; | my $i; |
| 3635 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3636 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 3637 | if ($node->[1] eq 'caption') { | if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 3638 | !!!cp ('t254'); | |
| 3639 | $i = $_; | $i = $_; |
| 3640 | last INSCOPE; | last INSCOPE; |
| 3641 | } elsif ({ | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3642 | table => 1, html => 1, | !!!cp ('t255'); |
| }->{$node->[1]}) { | ||
| 3643 | last INSCOPE; | last INSCOPE; |
| 3644 | } | } |
| 3645 | } # INSCOPE | } # INSCOPE |
| 3646 | unless (defined $i) { | unless (defined $i) { |
| 3647 | !!!parse-error (type => 'unmatched end tag:caption'); | !!!cp ('t256'); |
| 3648 | !!!parse-error (type => 'unmatched end tag', | |
| 3649 | text => $token->{tag_name}, token => $token); | |
| 3650 | ## Ignore the token | ## Ignore the token |
| 3651 | !!!nack ('t256.1'); | |
| 3652 | !!!next-token; | !!!next-token; |
| 3653 | redo B; | next B; |
| } | ||
| ## generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; # </table> | ||
| $token = {type => 'end tag', tag_name => 'caption'}; | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| redo B; | ||
| 3654 | } | } |
| 3655 | ||
| 3656 | if ($self->{open_elements}->[-1]->[1] ne 'caption') { | ## Clear back to table body context |
| 3657 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | while (not ($self->{open_elements}->[-1]->[1] |
| 3658 | & TABLE_ROWS_SCOPING_EL)) { | |
| 3659 | !!!cp ('t257'); | |
| 3660 | ## ISSUE: Can this case be reached? | |
| 3661 | pop @{$self->{open_elements}}; | |
| 3662 | } | } |
| 3663 | ||
| 3664 | splice @{$self->{open_elements}}, $i; | pop @{$self->{open_elements}}; |
| 3665 | $self->{insertion_mode} = IN_TABLE_IM; | |
| 3666 | $clear_up_to_marker->(); | !!!nack ('t257.1'); |
| 3667 | !!!next-token; | |
| 3668 | $self->{insertion_mode} = 'in table'; | next B; |
| ## reprocess | ||
| redo B; | ||
| 3669 | } elsif ({ | } elsif ({ |
| 3670 | body => 1, col => 1, colgroup => 1, | body => 1, caption => 1, col => 1, colgroup => 1, |
| 3671 | html => 1, tbody => 1, td => 1, tfoot => 1, | html => 1, td => 1, th => 1, |
| 3672 | th => 1, thead => 1, tr => 1, | tr => 1, # $self->{insertion_mode} == IN_ROW_IM |
| 3673 | tbody => 1, tfoot => 1, thead => 1, # $self->{insertion_mode} == IN_TABLE_IM | |
| 3674 | }->{$token->{tag_name}}) { | }->{$token->{tag_name}}) { |
| 3675 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t258'); |
| 3676 | ## Ignore the token | !!!parse-error (type => 'unmatched end tag', |
| 3677 | redo B; | text => $token->{tag_name}, token => $token); |
| 3678 | } else { | ## Ignore the token |
| 3679 | # | !!!nack ('t258.1'); |
| 3680 | } | !!!next-token; |
| 3681 | } else { | next B; |
| 3682 | # | } else { |
| 3683 | } | !!!cp ('t259'); |
| 3684 | !!!parse-error (type => 'in table:/', | |
| 3685 | $in_body->($insert_to_current); | text => $token->{tag_name}, token => $token); |
| 3686 | redo B; | |
| 3687 | } elsif ($self->{insertion_mode} eq 'in column group') { | $insert = $insert_to_foster; |
| 3688 | if ($token->{type} eq 'character') { | # |
| 3689 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | } |
| 3690 | } elsif ($token->{type} == END_OF_FILE_TOKEN) { | |
| 3691 | unless ($self->{open_elements}->[-1]->[1] == HTML_EL and | |
| 3692 | @{$self->{open_elements}} == 1) { # redundant, maybe | |
| 3693 | !!!parse-error (type => 'in body:#eof', token => $token); | |
| 3694 | !!!cp ('t259.1'); | |
| 3695 | # | |
| 3696 | } else { | |
| 3697 | !!!cp ('t259.2'); | |
| 3698 | # | |
| 3699 | } | |
| 3700 | ||
| 3701 | ## Stop parsing | |
| 3702 | last B; | |
| 3703 | } else { | |
| 3704 | die "$0: $token->{type}: Unknown token type"; | |
| 3705 | } | |
| 3706 | } elsif (($self->{insertion_mode} & IM_MASK) == IN_COLUMN_GROUP_IM) { | |
| 3707 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 3708 | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { | |
| 3709 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3710 | unless (length $token->{data}) { | unless (length $token->{data}) { |
| 3711 | !!!cp ('t260'); | |
| 3712 | !!!next-token; | !!!next-token; |
| 3713 | redo B; | next B; |
| 3714 | } | } |
| 3715 | } | } |
| 3716 | ||
| 3717 | !!!cp ('t261'); | |
| 3718 | # | # |
| 3719 | } elsif ($token->{type} eq 'comment') { | } elsif ($token->{type} == START_TAG_TOKEN) { |
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'start tag') { | ||
| 3720 | if ($token->{tag_name} eq 'col') { | if ($token->{tag_name} eq 'col') { |
| 3721 | !!!insert-element ($token->{tag_name}, $token->{attributes}); | !!!cp ('t262'); |
| 3722 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 3723 | pop @{$self->{open_elements}}; | pop @{$self->{open_elements}}; |
| 3724 | !!!ack ('t262.1'); | |
| 3725 | !!!next-token; | !!!next-token; |
| 3726 | redo B; | next B; |
| 3727 | } else { | } else { |
| 3728 | !!!cp ('t263'); | |
| 3729 | # | # |
| 3730 | } | } |
| 3731 | } elsif ($token->{type} eq 'end tag') { | } elsif ($token->{type} == END_TAG_TOKEN) { |
| 3732 | if ($token->{tag_name} eq 'colgroup') { | if ($token->{tag_name} eq 'colgroup') { |
| 3733 | if ($self->{open_elements}->[-1]->[1] eq 'html') { | if ($self->{open_elements}->[-1]->[1] == HTML_EL) { |
| 3734 | !!!parse-error (type => 'unmatched end tag:colgroup'); | !!!cp ('t264'); |
| 3735 | !!!parse-error (type => 'unmatched end tag', | |
| 3736 | text => 'colgroup', token => $token); | |
| 3737 | ## Ignore the token | ## Ignore the token |
| 3738 | !!!next-token; | !!!next-token; |
| 3739 | redo B; | next B; |
| 3740 | } else { | } else { |
| 3741 | !!!cp ('t265'); | |
| 3742 | pop @{$self->{open_elements}}; # colgroup | pop @{$self->{open_elements}}; # colgroup |
| 3743 | $self->{insertion_mode} = 'in table'; | $self->{insertion_mode} = IN_TABLE_IM; |
| 3744 | !!!next-token; | !!!next-token; |
| 3745 | redo B; | next B; |
| 3746 | } | } |
| 3747 | } elsif ($token->{tag_name} eq 'col') { | } elsif ($token->{tag_name} eq 'col') { |
| 3748 | !!!parse-error (type => 'unmatched end tag:col'); | !!!cp ('t266'); |
| 3749 | !!!parse-error (type => 'unmatched end tag', | |
| 3750 | text => 'col', token => $token); | |
| 3751 | ## Ignore the token | ## Ignore the token |
| 3752 | !!!next-token; | !!!next-token; |
| 3753 | redo B; | next B; |
| 3754 | } else { | } else { |
| 3755 | !!!cp ('t267'); | |
| 3756 | # | # |
| 3757 | } | } |
| 3758 | } else { | } elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 3759 | # | if ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 3760 | } | @{$self->{open_elements}} == 1) { # redundant, maybe |
| 3761 | !!!cp ('t270.2'); | |
| 3762 | ## Stop parsing. | |
| 3763 | last B; | |
| 3764 | } else { | |
| 3765 | ## NOTE: As if </colgroup>. | |
| 3766 | !!!cp ('t270.1'); | |
| 3767 | pop @{$self->{open_elements}}; # colgroup | |
| 3768 | $self->{insertion_mode} = IN_TABLE_IM; | |
| 3769 | ## Reprocess. | |
| 3770 | next B; | |
| 3771 | } | |
| 3772 | } else { | |
| 3773 | die "$0: $token->{type}: Unknown token type"; | |
| 3774 | } | |
| 3775 | ||
| 3776 | ## As if </colgroup> | ## As if </colgroup> |
| 3777 | if ($self->{open_elements}->[-1]->[1] eq 'html') { | if ($self->{open_elements}->[-1]->[1] == HTML_EL) { |
| 3778 | !!!parse-error (type => 'unmatched end tag:colgroup'); | !!!cp ('t269'); |
| 3779 | ## TODO: Wrong error type? | |
| 3780 | !!!parse-error (type => 'unmatched end tag', | |
| 3781 | text => 'colgroup', token => $token); | |
| 3782 | ## Ignore the token | ## Ignore the token |
| 3783 | !!!nack ('t269.1'); | |
| 3784 | !!!next-token; | !!!next-token; |
| 3785 | redo B; | next B; |
| 3786 | } else { | } else { |
| 3787 | !!!cp ('t270'); | |
| 3788 | pop @{$self->{open_elements}}; # colgroup | pop @{$self->{open_elements}}; # colgroup |
| 3789 | $self->{insertion_mode} = 'in table'; | $self->{insertion_mode} = IN_TABLE_IM; |
| 3790 | !!!ack-later; | |
| 3791 | ## reprocess | ## reprocess |
| 3792 | redo B; | next B; |
| 3793 | } | |
| 3794 | } elsif ($self->{insertion_mode} & SELECT_IMS) { | |
| 3795 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 3796 | !!!cp ('t271'); | |
| 3797 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | |
| 3798 | !!!next-token; | |
| 3799 | next B; | |
| 3800 | } elsif ($token->{type} == START_TAG_TOKEN) { | |
| 3801 | if ($token->{tag_name} eq 'option') { | |
| 3802 | if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { | |
| 3803 | !!!cp ('t272'); | |
| 3804 | ## As if </option> | |
| 3805 | pop @{$self->{open_elements}}; | |
| 3806 | } else { | |
| 3807 | !!!cp ('t273'); | |
| 3808 | } | } |
| } elsif ($self->{insertion_mode} eq 'in table body') { | ||
| if ($token->{type} eq 'character') { | ||
| ## NOTE: This is a "character in table" code clone. | ||
| if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | ||
| $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | ||
| unless (length $token->{data}) { | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| } | ||
| 3809 | ||
| 3810 | !!!parse-error (type => 'in table:#character'); | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3811 | !!!nack ('t273.1'); | |
| 3812 | !!!next-token; | |
| 3813 | next B; | |
| 3814 | } elsif ($token->{tag_name} eq 'optgroup') { | |
| 3815 | if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { | |
| 3816 | !!!cp ('t274'); | |
| 3817 | ## As if </option> | |
| 3818 | pop @{$self->{open_elements}}; | |
| 3819 | } else { | |
| 3820 | !!!cp ('t275'); | |
| 3821 | } | |
| 3822 | ||
| 3823 | ## As if in body, but insert into foster parent element | if ($self->{open_elements}->[-1]->[1] == OPTGROUP_EL) { |
| 3824 | ## ISSUE: Spec says that "whenever a node would be inserted | !!!cp ('t276'); |
| 3825 | ## into the current node" while characters might not be | ## As if </optgroup> |
| 3826 | ## result in a new Text node. | pop @{$self->{open_elements}}; |
| 3827 | $reconstruct_active_formatting_elements->($insert_to_foster); | } else { |
| 3828 | !!!cp ('t277'); | |
| 3829 | } | |
| 3830 | ||
| 3831 | if ({ | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3832 | table => 1, tbody => 1, tfoot => 1, | !!!nack ('t277.1'); |
| 3833 | thead => 1, tr => 1, | !!!next-token; |
| 3834 | }->{$self->{open_elements}->[-1]->[1]}) { | next B; |
| 3835 | # MUST | } elsif ({ |
| 3836 | my $foster_parent_element; | select => 1, input => 1, textarea => 1, keygen => 1, |
| 3837 | my $next_sibling; | }->{$token->{tag_name}} or |
| 3838 | my $prev_sibling; | (($self->{insertion_mode} & IM_MASK) |
| 3839 | OE: for (reverse 0..$#{$self->{open_elements}}) { | == IN_SELECT_IN_TABLE_IM and |
| 3840 | if ($self->{open_elements}->[$_]->[1] eq 'table') { | { |
| 3841 | my $parent = $self->{open_elements}->[$_]->[0]->parent_node; | caption => 1, table => 1, |
| 3842 | if (defined $parent and $parent->node_type == 1) { | tbody => 1, tfoot => 1, thead => 1, |
| 3843 | $foster_parent_element = $parent; | tr => 1, td => 1, th => 1, |
| 3844 | $next_sibling = $self->{open_elements}->[$_]->[0]; | }->{$token->{tag_name}})) { |
| 3845 | $prev_sibling = $next_sibling->previous_sibling; | |
| 3846 | } else { | ## 1. Parse error. |
| 3847 | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; | if ($token->{tag_name} eq 'select') { |
| 3848 | $prev_sibling = $foster_parent_element->last_child; | !!!parse-error (type => 'select in select', ## XXX: documentation |
| 3849 | } | token => $token); |
| 3850 | last OE; | } else { |
| 3851 | } | !!!parse-error (type => 'not closed', text => 'select', |
| 3852 | } # OE | token => $token); |
| 3853 | $foster_parent_element = $self->{open_elements}->[0]->[0] and | } |
| 3854 | $prev_sibling = $foster_parent_element->last_child | |
| 3855 | unless defined $foster_parent_element; | ## 2./<select>-1. Unless "have an element in table scope" (select): |
| 3856 | if (defined $prev_sibling and | my $i; |
| 3857 | $prev_sibling->node_type == 3) { | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3858 | $prev_sibling->manakai_append_text ($token->{data}); | my $node = $self->{open_elements}->[$_]; |
| 3859 | } else { | if ($node->[1] == SELECT_EL) { |
| 3860 | $foster_parent_element->insert_before | !!!cp ('t278'); |
| 3861 | ($self->{document}->create_text_node ($token->{data}), | $i = $_; |
| 3862 | $next_sibling); | last INSCOPE; |
| 3863 | } | } elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3864 | } else { | !!!cp ('t279'); |
| 3865 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | last INSCOPE; |
| 3866 | } | } |
| 3867 | } # INSCOPE | |
| 3868 | !!!next-token; | unless (defined $i) { |
| 3869 | redo B; | !!!cp ('t280'); |
| 3870 | } elsif ($token->{type} eq 'comment') { | if ($token->{tag_name} eq 'select') { |
| 3871 | ## Copied from 'in table' | ## NOTE: This error would be raised when |
| 3872 | my $comment = $self->{document}->create_comment ($token->{data}); | ## |select.innerHTML = '<select>'| is executed; in this |
| 3873 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | ## case two errors, "select in select" and "unmatched |
| 3874 | ## end tags" are reported to the user, the latter might | |
| 3875 | ## be confusing but this is what the spec requires. | |
| 3876 | !!!parse-error (type => 'unmatched end tag', | |
| 3877 | text => 'select', | |
| 3878 | token => $token); | |
| 3879 | } | |
| 3880 | ## Ignore the token. | |
| 3881 | !!!nack ('t280.1'); | |
| 3882 | !!!next-token; | !!!next-token; |
| 3883 | redo B; | next B; |
| 3884 | } elsif ($token->{type} eq 'start tag') { | } |
| if ({ | ||
| tr => 1, | ||
| th => 1, td => 1, | ||
| }->{$token->{tag_name}}) { | ||
| unless ($token->{tag_name} eq 'tr') { | ||
| !!!parse-error (type => 'missing start tag:tr'); | ||
| } | ||
| 3885 | ||
| 3886 | ## Clear back to table body context | ## 3. Otherwise, as if there were <select>: |
| while (not { | ||
| tbody => 1, tfoot => 1, thead => 1, html => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| pop @{$self->{open_elements}}; | ||
| } | ||
| 3887 | ||
| 3888 | $self->{insertion_mode} = 'in row'; | !!!cp ('t281'); |
| 3889 | if ($token->{tag_name} eq 'tr') { | splice @{$self->{open_elements}}, $i; |
| !!!insert-element ($token->{tag_name}, $token->{attributes}); | ||
| !!!next-token; | ||
| } else { | ||
| !!!insert-element ('tr'); | ||
| ## reprocess | ||
| } | ||
| redo B; | ||
| } elsif ({ | ||
| caption => 1, col => 1, colgroup => 1, | ||
| tbody => 1, tfoot => 1, thead => 1, | ||
| }->{$token->{tag_name}}) { | ||
| ## have an element in table scope | ||
| my $i; | ||
| INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ||
| my $node = $self->{open_elements}->[$_]; | ||
| if ({ | ||
| tbody => 1, thead => 1, tfoot => 1, | ||
| }->{$node->[1]}) { | ||
| $i = $_; | ||
| last INSCOPE; | ||
| } elsif ({ | ||
| table => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| } | ||
| } # INSCOPE | ||
| unless (defined $i) { | ||
| !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| 3890 | ||
| 3891 | ## Clear back to table body context | $self->_reset_insertion_mode; |
| while (not { | ||
| tbody => 1, tfoot => 1, thead => 1, html => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| pop @{$self->{open_elements}}; | ||
| } | ||
| 3892 | ||
| 3893 | ## As if <{current node}> | if ($token->{tag_name} eq 'select') { |
| 3894 | ## have an element in table scope | !!!nack ('t281.2'); |
| 3895 | ## true by definition | !!!next-token; |
| 3896 | next B; | |
| 3897 | } else { | |
| 3898 | !!!cp ('t281.1'); | |
| 3899 | !!!ack-later; | |
| 3900 | ## Reprocess the token. | |
| 3901 | next B; | |
| 3902 | } | |
| 3903 | } elsif ($token->{tag_name} eq 'script') { | |
| 3904 | !!!cp ('t281.3'); | |
| 3905 | ## NOTE: This is an "as if in head" code clone | |
| 3906 | $script_start_tag->(); | |
| 3907 | next B; | |
| 3908 | } else { | |
| 3909 | !!!cp ('t282'); | |
| 3910 | !!!parse-error (type => 'in select', | |
| 3911 | text => $token->{tag_name}, token => $token); | |
| 3912 | ## Ignore the token | |
| 3913 | !!!nack ('t282.1'); | |
| 3914 | !!!next-token; | |
| 3915 | next B; | |
| 3916 | } | |
| 3917 | } elsif ($token->{type} == END_TAG_TOKEN) { | |
| 3918 | if ($token->{tag_name} eq 'optgroup') { | |
| 3919 | if ($self->{open_elements}->[-1]->[1] == OPTION_EL and | |
| 3920 | $self->{open_elements}->[-2]->[1] == OPTGROUP_EL) { | |
| 3921 | !!!cp ('t283'); | |
| 3922 | ## As if </option> | |
| 3923 | splice @{$self->{open_elements}}, -2; | |
| 3924 | } elsif ($self->{open_elements}->[-1]->[1] == OPTGROUP_EL) { | |
| 3925 | !!!cp ('t284'); | |
| 3926 | pop @{$self->{open_elements}}; | |
| 3927 | } else { | |
| 3928 | !!!cp ('t285'); | |
| 3929 | !!!parse-error (type => 'unmatched end tag', | |
| 3930 | text => $token->{tag_name}, token => $token); | |
| 3931 | ## Ignore the token | |
| 3932 | } | |
| 3933 | !!!nack ('t285.1'); | |
| 3934 | !!!next-token; | |
| 3935 | next B; | |
| 3936 | } elsif ($token->{tag_name} eq 'option') { | |
| 3937 | if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { | |
| 3938 | !!!cp ('t286'); | |
| 3939 | pop @{$self->{open_elements}}; | |
| 3940 | } else { | |
| 3941 | !!!cp ('t287'); | |
| 3942 | !!!parse-error (type => 'unmatched end tag', | |
| 3943 | text => $token->{tag_name}, token => $token); | |
| 3944 | ## Ignore the token | |
| 3945 | } | |
| 3946 | !!!nack ('t287.1'); | |
| 3947 | !!!next-token; | |
| 3948 | next B; | |
| 3949 | } elsif ($token->{tag_name} eq 'select') { | |
| 3950 | ## have an element in table scope | |
| 3951 | my $i; | |
| 3952 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3953 | my $node = $self->{open_elements}->[$_]; | |
| 3954 | if ($node->[1] == SELECT_EL) { | |
| 3955 | !!!cp ('t288'); | |
| 3956 | $i = $_; | |
| 3957 | last INSCOPE; | |
| 3958 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 3959 | !!!cp ('t289'); | |
| 3960 | last INSCOPE; | |
| 3961 | } | |
| 3962 | } # INSCOPE | |
| 3963 | unless (defined $i) { | |
| 3964 | !!!cp ('t290'); | |
| 3965 | !!!parse-error (type => 'unmatched end tag', | |
| 3966 | text => $token->{tag_name}, token => $token); | |
| 3967 | ## Ignore the token | |
| 3968 | !!!nack ('t290.1'); | |
| 3969 | !!!next-token; | |
| 3970 | next B; | |
| 3971 | } | |
| 3972 | ||
| 3973 | !!!cp ('t291'); | |
| 3974 | splice @{$self->{open_elements}}, $i; | |
| 3975 | ||
| 3976 | ## Clear back to table body context | $self->_reset_insertion_mode; |
| ## nop by definition | ||
| 3977 | ||
| 3978 | pop @{$self->{open_elements}}; | !!!nack ('t291.1'); |
| 3979 | $self->{insertion_mode} = 'in table'; | !!!next-token; |
| 3980 | ## reprocess | next B; |
| 3981 | redo B; | } elsif (($self->{insertion_mode} & IM_MASK) |
| 3982 | } elsif ($token->{tag_name} eq 'table') { | == IN_SELECT_IN_TABLE_IM and |
| 3983 | ## NOTE: This is a code clone of "table in table" | { |
| 3984 | !!!parse-error (type => 'not closed:table'); | caption => 1, table => 1, tbody => 1, |
| 3985 | tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, | |
| 3986 | }->{$token->{tag_name}}) { | |
| 3987 | ## TODO: The following is wrong? | |
| 3988 | !!!parse-error (type => 'unmatched end tag', | |
| 3989 | text => $token->{tag_name}, token => $token); | |
| 3990 | ||
| 3991 | ## have an element in table scope | |
| 3992 | my $i; | |
| 3993 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 3994 | my $node = $self->{open_elements}->[$_]; | |
| 3995 | if ($node->[0]->manakai_local_name eq $token->{tag_name}) { | |
| 3996 | !!!cp ('t292'); | |
| 3997 | $i = $_; | |
| 3998 | last INSCOPE; | |
| 3999 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 4000 | !!!cp ('t293'); | |
| 4001 | last INSCOPE; | |
| 4002 | } | |
| 4003 | } # INSCOPE | |
| 4004 | unless (defined $i) { | |
| 4005 | !!!cp ('t294'); | |
| 4006 | ## Ignore the token | |
| 4007 | !!!nack ('t294.1'); | |
| 4008 | !!!next-token; | |
| 4009 | next B; | |
| 4010 | } | |
| 4011 | ||
| 4012 | ## As if </select> | |
| 4013 | ## have an element in table scope | |
| 4014 | undef $i; | |
| 4015 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 4016 | my $node = $self->{open_elements}->[$_]; | |
| 4017 | if ($node->[1] == SELECT_EL) { | |
| 4018 | !!!cp ('t295'); | |
| 4019 | $i = $_; | |
| 4020 | last INSCOPE; | |
| 4021 | } elsif ($node->[1] & TABLE_SCOPING_EL) { | |
| 4022 | ## ISSUE: Can this state be reached? | |
| 4023 | !!!cp ('t296'); | |
| 4024 | last INSCOPE; | |
| 4025 | } | |
| 4026 | } # INSCOPE | |
| 4027 | unless (defined $i) { | |
| 4028 | !!!cp ('t297'); | |
| 4029 | ## TODO: The following error type is correct? | |
| 4030 | !!!parse-error (type => 'unmatched end tag', | |
| 4031 | text => 'select', token => $token); | |
| 4032 | ## Ignore the </select> token | |
| 4033 | !!!nack ('t297.1'); | |
| 4034 | !!!next-token; ## TODO: ok? | |
| 4035 | next B; | |
| 4036 | } | |
| 4037 | ||
| 4038 | !!!cp ('t298'); | |
| 4039 | splice @{$self->{open_elements}}, $i; | |
| 4040 | ||
| 4041 | ## As if </table> | $self->_reset_insertion_mode; |
| 4042 | ## have a table element in table scope | |
| 4043 | my $i; | !!!ack-later; |
| 4044 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ## reprocess |
| 4045 | my $node = $self->{open_elements}->[$_]; | next B; |
| 4046 | if ($node->[1] eq 'table') { | } else { |
| 4047 | $i = $_; | !!!cp ('t299'); |
| 4048 | last INSCOPE; | !!!parse-error (type => 'in select:/', |
| 4049 | } elsif ({ | text => $token->{tag_name}, token => $token); |
| 4050 | table => 1, html => 1, | ## Ignore the token |
| 4051 | }->{$node->[1]}) { | !!!nack ('t299.3'); |
| 4052 | last INSCOPE; | !!!next-token; |
| 4053 | } | next B; |
| 4054 | } # INSCOPE | } |
| 4055 | unless (defined $i) { | } elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 4056 | !!!parse-error (type => 'unmatched end tag:table'); | unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4057 | ## Ignore tokens </table><table> | @{$self->{open_elements}} == 1) { # redundant, maybe |
| 4058 | !!!next-token; | !!!cp ('t299.1'); |
| 4059 | redo B; | !!!parse-error (type => 'in body:#eof', token => $token); |
| 4060 | } | } else { |
| 4061 | !!!cp ('t299.2'); | |
| 4062 | } | |
| 4063 | ||
| 4064 | ## Stop parsing. | |
| 4065 | last B; | |
| 4066 | } else { | |
| 4067 | die "$0: $token->{type}: Unknown token type"; | |
| 4068 | } | |
| 4069 | } elsif ($self->{insertion_mode} & BODY_AFTER_IMS) { | |
| 4070 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 4071 | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { | |
| 4072 | my $data = $1; | |
| 4073 | ## As if in body | |
| 4074 | $reconstruct_active_formatting_elements->($insert_to_current); | |
| 4075 | ||
| 4076 | ## generate implied end tags | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 4077 | if ({ | |
| 4078 | dd => 1, dt => 1, li => 1, p => 1, | unless (length $token->{data}) { |
| 4079 | td => 1, th => 1, tr => 1, | !!!cp ('t300'); |
| 4080 | }->{$self->{open_elements}->[-1]->[1]}) { | !!!next-token; |
| 4081 | !!!back-token; # <table> | next B; |
| 4082 | $token = {type => 'end tag', tag_name => 'table'}; | } |
| 4083 | !!!back-token; | } |
| 4084 | $token = {type => 'end tag', | |
| 4085 | tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { |
| 4086 | redo B; | !!!cp ('t301'); |
| 4087 | } | !!!parse-error (type => 'after html:#text', token => $token); |
| 4088 | # | |
| 4089 | } else { | |
| 4090 | !!!cp ('t302'); | |
| 4091 | ## "after body" insertion mode | |
| 4092 | !!!parse-error (type => 'after body:#text', token => $token); | |
| 4093 | # | |
| 4094 | } | |
| 4095 | ||
| 4096 | if ($self->{open_elements}->[-1]->[1] ne 'table') { | $self->{insertion_mode} = IN_BODY_IM; |
| 4097 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ## reprocess |
| 4098 | } | next B; |
| 4099 | } elsif ($token->{type} == START_TAG_TOKEN) { | |
| 4100 | if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { | |
| 4101 | !!!cp ('t303'); | |
| 4102 | !!!parse-error (type => 'after html', | |
| 4103 | &nbs |