Parent Directory
|
Revision Log
|
Patch
| revision 1.22 by wakaba, Sat Jun 23 14:55:45 2007 UTC | revision 1.237 by wakaba, Sun Sep 6 08:29:32 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 | hgroup => MISC_SPECIAL_EL, | |
| 198 | hr => MISC_SPECIAL_EL, | |
| 199 | html => HTML_EL, | |
| 200 | i => FORMATTING_EL, | |
| 201 | iframe => MISC_SPECIAL_EL, | |
| 202 | img => MISC_SPECIAL_EL, | |
| 203 | #image => MISC_SPECIAL_EL, ## NOTE: Commented out in the spec. | |
| 204 | input => MISC_SPECIAL_EL, | |
| 205 | isindex => MISC_SPECIAL_EL, | |
| 206 | ## XXX keygen? (Whether a void element is in Special or not does not | |
| 207 | ## affect to the processing, however.) | |
| 208 | li => LI_EL, | |
| 209 | link => MISC_SPECIAL_EL, | |
| 210 | listing => MISC_SPECIAL_EL, | |
| 211 | marquee => MISC_SCOPING_EL, | |
| 212 | menu => MISC_SPECIAL_EL, | |
| 213 | meta => MISC_SPECIAL_EL, | |
| 214 | nav => MISC_SPECIAL_EL, | |
| 215 | nobr => NOBR_EL, | |
| 216 | noembed => MISC_SPECIAL_EL, | |
| 217 | noframes => MISC_SPECIAL_EL, | |
| 218 | noscript => MISC_SPECIAL_EL, | |
| 219 | object => MISC_SCOPING_EL, | |
| 220 | ol => MISC_SPECIAL_EL, | |
| 221 | optgroup => OPTGROUP_EL, | |
| 222 | option => OPTION_EL, | |
| 223 | p => P_EL, | |
| 224 | param => MISC_SPECIAL_EL, | |
| 225 | plaintext => MISC_SPECIAL_EL, | |
| 226 | pre => MISC_SPECIAL_EL, | |
| 227 | rp => RUBY_COMPONENT_EL, | |
| 228 | rt => RUBY_COMPONENT_EL, | |
| 229 | ruby => RUBY_EL, | |
| 230 | s => FORMATTING_EL, | |
| 231 | script => MISC_SPECIAL_EL, | |
| 232 | select => SELECT_EL, | |
| 233 | section => MISC_SPECIAL_EL, | |
| 234 | small => FORMATTING_EL, | |
| 235 | spacer => MISC_SPECIAL_EL, | |
| 236 | strike => FORMATTING_EL, | |
| 237 | strong => FORMATTING_EL, | |
| 238 | style => MISC_SPECIAL_EL, | |
| 239 | table => TABLE_EL, | |
| 240 | tbody => TABLE_ROW_GROUP_EL, | |
| 241 | td => TABLE_CELL_EL, | |
| 242 | textarea => MISC_SPECIAL_EL, | |
| 243 | tfoot => TABLE_ROW_GROUP_EL, | |
| 244 | th => TABLE_CELL_EL, | |
| 245 | thead => TABLE_ROW_GROUP_EL, | |
| 246 | title => MISC_SPECIAL_EL, | |
| 247 | tr => TABLE_ROW_EL, | |
| 248 | tt => FORMATTING_EL, | |
| 249 | u => FORMATTING_EL, | |
| 250 | ul => MISC_SPECIAL_EL, | |
| 251 | wbr => MISC_SPECIAL_EL, | |
| 252 | xmp => MISC_SPECIAL_EL, | |
| 253 | }; | }; |
| 254 | my $formatting_category = { | |
| 255 | a => 1, b => 1, big => 1, em => 1, font => 1, i => 1, nobr => 1, | my $el_category_f = { |
| 256 | s => 1, small => 1, strile => 1, strong => 1, tt => 1, u => 1, | $MML_NS => { |
| 257 | 'annotation-xml' => MML_AXML_EL, | |
| 258 | mi => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 259 | mo => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 260 | mn => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 261 | ms => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 262 | mtext => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 263 | }, | |
| 264 | $SVG_NS => { | |
| 265 | foreignObject => SCOPING_EL | FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 266 | desc => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 267 | title => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, | |
| 268 | }, | |
| 269 | ## NOTE: In addition, FOREIGN_EL is set to non-HTML elements. | |
| 270 | }; | }; |
| # $phrasing_category: all other elements | ||
| 271 | ||
| 272 | sub parse_string ($$$;$) { | my $svg_attr_name = { |
| 273 | my $self = shift->new; | attributename => 'attributeName', |
| 274 | my $s = \$_[0]; | attributetype => 'attributeType', |
| 275 | $self->{document} = $_[1]; | basefrequency => 'baseFrequency', |
| 276 | baseprofile => 'baseProfile', | |
| 277 | calcmode => 'calcMode', | |
| 278 | clippathunits => 'clipPathUnits', | |
| 279 | contentscripttype => 'contentScriptType', | |
| 280 | contentstyletype => 'contentStyleType', | |
| 281 | diffuseconstant => 'diffuseConstant', | |
| 282 | edgemode => 'edgeMode', | |
| 283 | externalresourcesrequired => 'externalResourcesRequired', | |
| 284 | filterres => 'filterRes', | |
| 285 | filterunits => 'filterUnits', | |
| 286 | glyphref => 'glyphRef', | |
| 287 | gradienttransform => 'gradientTransform', | |
| 288 | gradientunits => 'gradientUnits', | |
| 289 | kernelmatrix => 'kernelMatrix', | |
| 290 | kernelunitlength => 'kernelUnitLength', | |
| 291 | keypoints => 'keyPoints', | |
| 292 | keysplines => 'keySplines', | |
| 293 | keytimes => 'keyTimes', | |
| 294 | lengthadjust => 'lengthAdjust', | |
| 295 | limitingconeangle => 'limitingConeAngle', | |
| 296 | markerheight => 'markerHeight', | |
| 297 | markerunits => 'markerUnits', | |
| 298 | markerwidth => 'markerWidth', | |
| 299 | maskcontentunits => 'maskContentUnits', | |
| 300 | maskunits => 'maskUnits', | |
| 301 | numoctaves => 'numOctaves', | |
| 302 | pathlength => 'pathLength', | |
| 303 | patterncontentunits => 'patternContentUnits', | |
| 304 | patterntransform => 'patternTransform', | |
| 305 | patternunits => 'patternUnits', | |
| 306 | pointsatx => 'pointsAtX', | |
| 307 | pointsaty => 'pointsAtY', | |
| 308 | pointsatz => 'pointsAtZ', | |
| 309 | preservealpha => 'preserveAlpha', | |
| 310 | preserveaspectratio => 'preserveAspectRatio', | |
| 311 | primitiveunits => 'primitiveUnits', | |
| 312 | refx => 'refX', | |
| 313 | refy => 'refY', | |
| 314 | repeatcount => 'repeatCount', | |
| 315 | repeatdur => 'repeatDur', | |
| 316 | requiredextensions => 'requiredExtensions', | |
| 317 | requiredfeatures => 'requiredFeatures', | |
| 318 | specularconstant => 'specularConstant', | |
| 319 | specularexponent => 'specularExponent', | |
| 320 | spreadmethod => 'spreadMethod', | |
| 321 | startoffset => 'startOffset', | |
| 322 | stddeviation => 'stdDeviation', | |
| 323 | stitchtiles => 'stitchTiles', | |
| 324 | surfacescale => 'surfaceScale', | |
| 325 | systemlanguage => 'systemLanguage', | |
| 326 | tablevalues => 'tableValues', | |
| 327 | targetx => 'targetX', | |
| 328 | targety => 'targetY', | |
| 329 | textlength => 'textLength', | |
| 330 | viewbox => 'viewBox', | |
| 331 | viewtarget => 'viewTarget', | |
| 332 | xchannelselector => 'xChannelSelector', | |
| 333 | ychannelselector => 'yChannelSelector', | |
| 334 | zoomandpan => 'zoomAndPan', | |
| 335 | }; | |
| 336 | ||
| 337 | ## NOTE: |set_inner_html| copies most of this method's code | my $foreign_attr_xname = { |
| 338 | 'xlink:actuate' => [$XLINK_NS, ['xlink', 'actuate']], | |
| 339 | 'xlink:arcrole' => [$XLINK_NS, ['xlink', 'arcrole']], | |
| 340 | 'xlink:href' => [$XLINK_NS, ['xlink', 'href']], | |
| 341 | 'xlink:role' => [$XLINK_NS, ['xlink', 'role']], | |
| 342 | 'xlink:show' => [$XLINK_NS, ['xlink', 'show']], | |
| 343 | 'xlink:title' => [$XLINK_NS, ['xlink', 'title']], | |
| 344 | 'xlink:type' => [$XLINK_NS, ['xlink', 'type']], | |
| 345 | 'xml:base' => [$XML_NS, ['xml', 'base']], | |
| 346 | 'xml:lang' => [$XML_NS, ['xml', 'lang']], | |
| 347 | 'xml:space' => [$XML_NS, ['xml', 'space']], | |
| 348 | 'xmlns' => [$XMLNS_NS, [undef, 'xmlns']], | |
| 349 | 'xmlns:xlink' => [$XMLNS_NS, ['xmlns', 'xlink']], | |
| 350 | }; | |
| 351 | ||
| 352 | 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; | ||
| 353 | ||
| 354 | pop @{$self->{prev_input_character}}; | ## TODO: Invoke the reset algorithm when a resettable element is |
| 355 | unshift @{$self->{prev_input_character}}, $self->{next_input_character}; | ## created (cf. HTML5 revision 2259). |
| 356 | ||
| 357 | $self->{next_input_character} = -1 and return if $i >= length $$s; | sub parse_byte_string ($$$$;$) { |
| 358 | $self->{next_input_character} = ord substr $$s, $i++, 1; | my $self = shift; |
| 359 | $column++; | my $charset_name = shift; |
| 360 | open my $input, '<', ref $_[0] ? $_[0] : \($_[0]); | |
| 361 | if ($self->{next_input_character} == 0x000A) { # LF | return $self->parse_byte_stream ($charset_name, $input, @_[1..$#_]); |
| 362 | $line++; | } # parse_byte_string |
| 363 | $column = 0; | |
| 364 | } elsif ($self->{next_input_character} == 0x000D) { # CR | sub parse_byte_stream ($$$$;$$) { |
| 365 | $i++ if substr ($$s, $i, 1) eq "\x0A"; | # my ($self, $charset_name, $byte_stream, $doc, $onerror, $get_wrapper) = @_; |
| 366 | $self->{next_input_character} = 0x000A; # LF # MUST | my $self = ref $_[0] ? shift : shift->new; |
| 367 | $line++; | my $charset_name = shift; |
| 368 | $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; | ||
| 369 | ||
| 370 | my $onerror = $_[2] || sub { | my $onerror = $_[2] || sub { |
| 371 | my (%opt) = @_; | my (%opt) = @_; |
| 372 | 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); | ||
| 373 | }; | }; |
| 374 | $self->{parse_error} = $onerror; # updated later by parse_char_string | |
| 375 | ||
| 376 | $self->_initialize_tokenizer; | my $get_wrapper = $_[3] || sub ($) { |
| 377 | $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 { | ||
| # | ||
| 378 | }; | }; |
| 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; | ||
| 379 | ||
| 380 | $self->{state} = 'data'; | ## HTML5 encoding sniffing algorithm |
| 381 | # next-input-character is already done | require Message::Charset::Info; |
| 382 | my $charset; | |
| 383 | my $buffer; | |
| 384 | my ($char_stream, $e_status); | |
| 385 | ||
| 386 | SNIFFING: { | |
| 387 | ## NOTE: By setting |allow_fallback| option true when the | |
| 388 | ## |get_decode_handle| method is invoked, we ignore what the HTML5 | |
| 389 | ## spec requires, i.e. unsupported encoding should be ignored. | |
| 390 | ## TODO: We should not do this unless the parser is invoked | |
| 391 | ## in the conformance checking mode, in which this behavior | |
| 392 | ## would be useful. | |
| 393 | ||
| 394 | unless (defined $token) { | ## Step 1 |
| 395 | !!!emit ({type => 'character', data => '&'}); | if (defined $charset_name) { |
| 396 | $charset = Message::Charset::Info->get_by_html_name ($charset_name); | |
| 397 | ## TODO: Is this ok? Transfer protocol's parameter should be | |
| 398 | ## interpreted in its semantics? | |
| 399 | ||
| 400 | ($char_stream, $e_status) = $charset->get_decode_handle | |
| 401 | ($byte_stream, allow_error_reporting => 1, | |
| 402 | allow_fallback => 1); | |
| 403 | if ($char_stream) { | |
| 404 | $self->{confident} = 1; | |
| 405 | last SNIFFING; | |
| 406 | } else { | } else { |
| 407 | !!!emit ($token); | !!!parse-error (type => 'charset:not supported', |
| 408 | layer => 'encode', | |
| 409 | line => 1, column => 1, | |
| 410 | value => $charset_name, | |
| 411 | level => $self->{level}->{uncertain}); | |
| 412 | } | } |
| 413 | } | |
| 414 | ||
| 415 | redo A; | ## Step 2 |
| 416 | } elsif ($self->{state} eq 'tag open') { | my $byte_buffer = ''; |
| 417 | if ($self->{content_model_flag} eq 'RCDATA' or | for (1..1024) { |
| 418 | $self->{content_model_flag} eq 'CDATA') { | my $char = $byte_stream->getc; |
| 419 | if ($self->{next_input_character} == 0x002F) { # / | last unless defined $char; |
| 420 | !!!next-input-character; | $byte_buffer .= $char; |
| 421 | $self->{state} = 'close tag open'; | } ## TODO: timeout |
| redo A; | ||
| } else { | ||
| ## reconsume | ||
| $self->{state} = 'data'; | ||
| !!!emit ({type => 'character', data => '<'}); | ||
| 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 | ||
| 422 | ||
| 423 | !!!emit ({type => 'character', data => '<'}); | ## Step 3 |
| 424 | if ($byte_buffer =~ /^\xFE\xFF/) { | |
| 425 | $charset = Message::Charset::Info->get_by_html_name ('utf-16be'); | |
| 426 | ($char_stream, $e_status) = $charset->get_decode_handle | |
| 427 | ($byte_stream, allow_error_reporting => 1, | |
| 428 | allow_fallback => 1, byte_buffer => \$byte_buffer); | |
| 429 | $self->{confident} = 1; | |
| 430 | last SNIFFING; | |
| 431 | } elsif ($byte_buffer =~ /^\xFF\xFE/) { | |
| 432 | $charset = Message::Charset::Info->get_by_html_name ('utf-16le'); | |
| 433 | ($char_stream, $e_status) = $charset->get_decode_handle | |
| 434 | ($byte_stream, allow_error_reporting => 1, | |
| 435 | allow_fallback => 1, byte_buffer => \$byte_buffer); | |
| 436 | $self->{confident} = 1; | |
| 437 | last SNIFFING; | |
| 438 | } elsif ($byte_buffer =~ /^\xEF\xBB\xBF/) { | |
| 439 | $charset = Message::Charset::Info->get_by_html_name ('utf-8'); | |
| 440 | ($char_stream, $e_status) = $charset->get_decode_handle | |
| 441 | ($byte_stream, allow_error_reporting => 1, | |
| 442 | allow_fallback => 1, byte_buffer => \$byte_buffer); | |
| 443 | $self->{confident} = 1; | |
| 444 | last SNIFFING; | |
| 445 | } | |
| 446 | ||
| 447 | redo A; | ## Step 4 |
| 448 | } | ## TODO: <meta charset> |
| } else { | ||
| die "$0: $self->{content_model_flag}: Unknown content model flag"; | ||
| } | ||
| } elsif ($self->{state} eq 'close tag open') { | ||
| if ($self->{content_model_flag} eq 'RCDATA' or | ||
| $self->{content_model_flag} eq 'CDATA') { | ||
| 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'; | ||
| 449 | ||
| 450 | !!!emit ({type => 'character', data => '</'}); | ## Step 5 |
| 451 | ## TODO: from history | |
| 452 | ||
| 453 | redo A; | ## Step 6 |
| 454 | } | require Whatpm::Charset::UniversalCharDet; |
| 455 | } | $charset_name = Whatpm::Charset::UniversalCharDet->detect_byte_string |
| 456 | push @next_char, $self->{next_input_character}; | ($byte_buffer); |
| 457 | if (defined $charset_name) { | |
| 458 | unless ($self->{next_input_character} == 0x0009 or # HT | $charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 459 | $self->{next_input_character} == 0x000A or # LF | |
| 460 | $self->{next_input_character} == 0x000B or # VT | require Whatpm::Charset::DecodeHandle; |
| 461 | $self->{next_input_character} == 0x000C or # FF | $buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
| 462 | $self->{next_input_character} == 0x0020 or # SP | ($byte_stream); |
| 463 | $self->{next_input_character} == 0x003E or # > | ($char_stream, $e_status) = $charset->get_decode_handle |
| 464 | $self->{next_input_character} == 0x002F or # / | ($buffer, allow_error_reporting => 1, |
| 465 | $self->{next_input_character} == -1) { | allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 466 | $self->{next_input_character} = shift @next_char; # reconsume | if ($char_stream) { |
| 467 | !!!back-next-input-character (@next_char); | $buffer->{buffer} = $byte_buffer; |
| 468 | $self->{state} = 'data'; | !!!parse-error (type => 'sniffing:chardet', |
| 469 | text => $charset_name, | |
| 470 | !!!emit ({type => 'character', data => '</'}); | level => $self->{level}->{info}, |
| 471 | layer => 'encode', | |
| 472 | redo A; | line => 1, column => 1); |
| 473 | } else { | $self->{confident} = 0; |
| 474 | $self->{next_input_character} = shift @next_char; | last SNIFFING; |
| !!!back-next-input-character (@next_char); | ||
| # and consume... | ||
| } | ||
| 475 | } | } |
| 476 | } | |
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } else { | ||
| $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | ||
| $self->{state} = 'attribute value (unquoted)'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'attribute value (double-quoted)') { | ||
| if ($self->{next_input_character} == 0x0022) { # " | ||
| $self->{state} = 'before attribute name'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == 0x0026) { # & | ||
| $self->{last_attribute_value_state} = 'attribute value (double-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 | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } else { | ||
| $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 | ||
| !!!emit ($self->{current_token}); # start tag or end tag | ||
| undef $self->{current_token}; | ||
| 477 | ||
| 478 | redo A; | ## Step 7: default |
| 479 | } else { | ## TODO: Make this configurable. |
| 480 | $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | $charset = Message::Charset::Info->get_by_html_name ('windows-1252'); |
| 481 | ## Stay in the state | ## NOTE: We choose |windows-1252| here, since |utf-8| should be |
| 482 | !!!next-input-character; | ## detectable in the step 6. |
| 483 | redo A; | require Whatpm::Charset::DecodeHandle; |
| 484 | } | $buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
| 485 | } elsif ($self->{state} eq 'attribute value (unquoted)') { | ($byte_stream); |
| 486 | if ($self->{next_input_character} == 0x0009 or # HT | ($char_stream, $e_status) |
| 487 | $self->{next_input_character} == 0x000A or # LF | = $charset->get_decode_handle ($buffer, |
| 488 | $self->{next_input_character} == 0x000B or # HT | allow_error_reporting => 1, |
| 489 | $self->{next_input_character} == 0x000C or # FF | allow_fallback => 1, |
| 490 | $self->{next_input_character} == 0x0020) { # SP | byte_buffer => \$byte_buffer); |
| 491 | $self->{state} = 'before attribute name'; | $buffer->{buffer} = $byte_buffer; |
| 492 | !!!next-input-character; | !!!parse-error (type => 'sniffing:default', |
| 493 | redo A; | text => 'windows-1252', |
| 494 | } elsif ($self->{next_input_character} == 0x0026) { # & | level => $self->{level}->{info}, |
| 495 | $self->{last_attribute_value_state} = 'attribute value (unquoted)'; | line => 1, column => 1, |
| 496 | $self->{state} = 'entity in attribute value'; | layer => 'encode'); |
| 497 | !!!next-input-character; | $self->{confident} = 0; |
| 498 | redo A; | } # SNIFFING |
| 499 | } elsif ($self->{next_input_character} == 0x003E) { # > | |
| 500 | if ($self->{current_token}->{type} eq 'start tag') { | if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 501 | $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? |
| 502 | } elsif ($self->{current_token}->{type} eq 'end tag') { | !!!parse-error (type => 'chardecode:fallback', |
| 503 | $self->{content_model_flag} = 'PCDATA'; # MUST | #text => $self->{input_encoding}, |
| 504 | if ($self->{current_token}->{attributes}) { | level => $self->{level}->{uncertain}, |
| 505 | !!!parse-error (type => 'end tag attribute'); | line => 1, column => 1, |
| 506 | } | layer => 'encode'); |
| 507 | } else { | } elsif (not ($e_status & |
| 508 | die "$0: $self->{current_token}->{type}: Unknown token type"; | Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) { |
| 509 | } | $self->{input_encoding} = $charset->get_iana_name; |
| 510 | $self->{state} = 'data'; | !!!parse-error (type => 'chardecode:no error', |
| 511 | !!!next-input-character; | text => $self->{input_encoding}, |
| 512 | level => $self->{level}->{uncertain}, | |
| 513 | !!!emit ($self->{current_token}); # start tag or end tag | line => 1, column => 1, |
| 514 | undef $self->{current_token}; | layer => 'encode'); |
| 515 | } else { | |
| 516 | redo A; | $self->{input_encoding} = $charset->get_iana_name; |
| 517 | } 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 | ||
| 518 | ||
| 519 | !!!emit ($self->{current_token}); # start tag or end tag | $self->{change_encoding} = sub { |
| 520 | undef $self->{current_token}; | my $self = shift; |
| 521 | $charset_name = shift; | |
| 522 | my $token = shift; | |
| 523 | ||
| 524 | redo A; | $charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 525 | } else { | ($char_stream, $e_status) = $charset->get_decode_handle |
| 526 | $self->{current_attribute}->{value} .= chr ($self->{next_input_character}); | ($byte_stream, allow_error_reporting => 1, allow_fallback => 1, |
| 527 | ## Stay in the state | byte_buffer => \ $buffer->{buffer}); |
| 528 | !!!next-input-character; | |
| 529 | redo A; | if ($char_stream) { # if supported |
| 530 | ## "Change the encoding" algorithm: | |
| 531 | ||
| 532 | ## Step 1 | |
| 533 | if (defined $self->{input_encoding} and | |
| 534 | $self->{input_encoding} eq $charset_name) { | |
| 535 | !!!parse-error (type => 'charset label:matching', | |
| 536 | text => $charset_name, | |
| 537 | level => $self->{level}->{info}); | |
| 538 | $self->{confident} = 1; | |
| 539 | return; | |
| 540 | } | } |
| } elsif ($self->{state} eq 'entity in attribute value') { | ||
| my $token = $self->_tokenize_attempt_to_consume_an_entity; | ||
| 541 | ||
| 542 | unless (defined $token) { | ## Step 2 (HTML5 revision 3205) |
| 543 | $self->{current_attribute}->{value} .= '&'; | if (defined $self->{input_encoding} and |
| 544 | } else { | Message::Charset::Info->get_by_html_name ($self->{input_encoding}) |
| 545 | $self->{current_attribute}->{value} .= $token->{data}; | ->{category} & Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 546 | ## ISSUE: spec says "append the returned character token to the current attribute's value" | $self->{confident} = 1; |
| 547 | return; | |
| 548 | } | } |
| 549 | ||
| 550 | $self->{state} = $self->{last_attribute_value_state}; | ## Step 3 |
| 551 | # next-input-character is already done | if ($charset->{category} & |
| 552 | redo A; | Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 553 | } elsif ($self->{state} eq 'bogus comment') { | $charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 554 | ## (only happen if PCDATA state) | ($char_stream, $e_status) = $charset->get_decode_handle |
| 555 | ($byte_stream, | |
| 556 | my $token = {type => 'comment', data => ''}; | byte_buffer => \ $buffer->{buffer}); |
| 557 | } | |
| 558 | BC: { | $charset_name = $charset->get_iana_name; |
| 559 | if ($self->{next_input_character} == 0x003E) { # > | |
| 560 | $self->{state} = 'data'; | !!!parse-error (type => 'charset label detected', |
| 561 | !!!next-input-character; | text => $self->{input_encoding}, |
| 562 | value => $charset_name, | |
| 563 | !!!emit ($token); | level => $self->{level}->{warn}, |
| 564 | token => $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}; | ||
| 565 | ||
| 566 | if ($self->{next_input_character} == 0x002D) { # - | ## Step 4 |
| 567 | !!!next-input-character; | # if (can) { |
| 568 | push @next_char, $self->{next_input_character}; | ## change the encoding on the fly. |
| 569 | if ($self->{next_input_character} == 0x002D) { # - | #$self->{confident} = 1; |
| 570 | $self->{current_token} = {type => 'comment', data => ''}; | #return; |
| 571 | $self->{state} = 'comment'; | # } |
| !!!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; | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| !!!parse-error (type => 'bogus comment open'); | ||
| $self->{next_input_character} = shift @next_char; | ||
| !!!back-next-input-character (@next_char); | ||
| $self->{state} = 'bogus comment'; | ||
| redo A; | ||
| 572 | ||
| 573 | ## ISSUE: typos in spec: chacacters, is is a parse error | ## Step 5 |
| 574 | ## 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? | throw Whatpm::HTML::RestartParser (); |
| 575 | } elsif ($self->{state} eq 'comment') { | } |
| 576 | if ($self->{next_input_character} == 0x002D) { # - | }; # $self->{change_encoding} |
| $self->{state} = 'comment dash'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed comment'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| !!!emit ($self->{current_token}); # comment | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{data} .= chr ($self->{next_input_character}); # comment | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'comment dash') { | ||
| if ($self->{next_input_character} == 0x002D) { # - | ||
| $self->{state} = 'comment end'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed comment'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| !!!emit ($self->{current_token}); # comment | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } else { | ||
| $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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| !!!emit ($self->{current_token}); # comment | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } else { | ||
| !!!parse-error (type => 'dash in comment'); | ||
| $self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment | ||
| $self->{state} = 'comment'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } 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 | ||
| !!!emit ({type => 'DOCTYPE'}); # incorrect | ||
| redo A; | ||
| } else { | ||
| $self->{current_token} | ||
| = {type => 'DOCTYPE', | ||
| name => chr ($self->{next_input_character}), | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } else { | ||
| $self->{current_token}->{name} | ||
| .= chr ($self->{next_input_character}); # DOCTYPE | ||
| ## Stay in the state | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } elsif ($self->{state} eq 'after 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) { # > | ||
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 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; | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| # | ||
| } elsif ($self->{next_input_character} == 0x0053 or # S | ||
| $self->{next_input_character} == 0x0073) { # s | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0059 or # Y | ||
| $self->{next_input_character} == 0x0079) { # y | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0053 or # S | ||
| $self->{next_input_character} == 0x0073) { # s | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0054 or # T | ||
| $self->{next_input_character} == 0x0074) { # t | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0045 or # E | ||
| $self->{next_input_character} == 0x0065) { # e | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x004D or # M | ||
| $self->{next_input_character} == 0x006D) { # m | ||
| $self->{state} = 'before DOCTYPE system identifier'; | ||
| !!!next-input-character; | ||
| redo A; | ||
| } | ||
| } | ||
| } | ||
| } | ||
| } | ||
| 577 | ||
| 578 | # | my $char_onerror = sub { |
| 579 | } else { | my (undef, $type, %opt) = @_; |
| 580 | !!!next-input-character; | !!!parse-error (layer => 'encode', |
| 581 | # | line => $self->{line}, column => $self->{column} + 1, |
| 582 | } | %opt, type => $type); |
| 583 | if ($opt{octets}) { | |
| 584 | ${$opt{octets}} = "\x{FFFD}"; # relacement character | |
| 585 | } | |
| 586 | }; | |
| 587 | ||
| 588 | !!!parse-error (type => 'string after DOCTYPE name'); | my $wrapped_char_stream = $get_wrapper->($char_stream); |
| 589 | $self->{state} = 'bogus DOCTYPE'; | $wrapped_char_stream->onerror ($char_onerror); |
| # next-input-character is already done | ||
| redo A; | ||
| } elsif ($self->{state} eq 'before 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} 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 590 | ||
| 591 | redo A; | my @args = ($_[1], $_[2]); # $doc, $onerror - $get_wrapper = undef; |
| 592 | } else { | my $return; |
| 593 | !!!parse-error (type => 'string after PUBLIC'); | try { |
| 594 | $self->{state} = 'bogus DOCTYPE'; | $return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 595 | !!!next-input-character; | } catch Whatpm::HTML::RestartParser with { |
| 596 | redo A; | ## NOTE: Invoked after {change_encoding}. |
| 597 | } | |
| 598 | } elsif ($self->{state} eq 'DOCTYPE public identifier (double-quoted)') { | if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 599 | if ($self->{next_input_character} == 0x0022) { # " | $self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 600 | $self->{state} = 'after DOCTYPE public identifier'; | !!!parse-error (type => 'chardecode:fallback', |
| 601 | !!!next-input-character; | level => $self->{level}->{uncertain}, |
| 602 | redo A; | #text => $self->{input_encoding}, |
| 603 | } elsif ($self->{next_input_character} == -1) { | line => 1, column => 1, |
| 604 | !!!parse-error (type => 'unclosed PUBLIC literal'); | layer => 'encode'); |
| 605 | } elsif (not ($e_status & | |
| 606 | $self->{state} = 'data'; | Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) { |
| 607 | ## reconsume | $self->{input_encoding} = $charset->get_iana_name; |
| 608 | !!!parse-error (type => 'chardecode:no error', | |
| 609 | delete $self->{current_token}->{correct}; | text => $self->{input_encoding}, |
| 610 | !!!emit ($self->{current_token}); # DOCTYPE | level => $self->{level}->{uncertain}, |
| 611 | undef $self->{current_token}; | line => 1, column => 1, |
| 612 | layer => 'encode'); | |
| 613 | } else { | |
| 614 | $self->{input_encoding} = $charset->get_iana_name; | |
| 615 | } | |
| 616 | $self->{confident} = 1; | |
| 617 | ||
| 618 | redo A; | $wrapped_char_stream = $get_wrapper->($char_stream); |
| 619 | } else { | $wrapped_char_stream->onerror ($char_onerror); |
| $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 | ||
| undef $self->{current_token}; | ||
| 620 | ||
| 621 | redo A; | $return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 622 | } else { | }; |
| 623 | $self->{current_token}->{public_identifier} # DOCTYPE | return $return; |
| 624 | .= chr $self->{next_input_character}; | } # parse_byte_stream |
| ## 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 625 | ||
| 626 | redo A; | ## NOTE: HTML5 spec says that the encoding layer MUST NOT strip BOM |
| 627 | } else { | ## and the HTML layer MUST ignore it. However, we does strip BOM in |
| 628 | !!!parse-error (type => 'string after PUBLIC literal'); | ## the encoding layer and the HTML layer does not ignore any U+FEFF, |
| 629 | $self->{state} = 'bogus DOCTYPE'; | ## because the core part of our HTML parser expects a string of character, |
| 630 | !!!next-input-character; | ## not a string of bytes or code units or anything which might contain a BOM. |
| 631 | redo A; | ## Therefore, any parser interface that accepts a string of bytes, |
| 632 | } | ## such as |parse_byte_string| in this module, must ensure that it does |
| 633 | } elsif ($self->{state} eq 'before DOCTYPE system identifier') { | ## strip the BOM and never strip any ZWNBSP. |
| 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 | ||
| undef $self->{current_token}; | ||
| 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 | ||
| undef $self->{current_token}; | ||
| 634 | ||
| 635 | redo A; | sub parse_char_string ($$$;$$) { |
| 636 | } else { | #my ($self, $s, $doc, $onerror, $get_wrapper) = @_; |
| 637 | !!!parse-error (type => 'string after PUBLIC literal'); | my $self = shift; |
| 638 | $self->{state} = 'bogus DOCTYPE'; | my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 639 | !!!next-input-character; | require Whatpm::Charset::DecodeHandle; |
| 640 | redo A; | my $input = Whatpm::Charset::DecodeHandle::CharString->new ($s); |
| 641 | } | return $self->parse_char_stream ($input, @_[1..$#_]); |
| 642 | } elsif ($self->{state} eq 'DOCTYPE system identifier (double-quoted)') { | } # parse_char_string |
| 643 | if ($self->{next_input_character} == 0x0022) { # " | *parse_string = \&parse_char_string; ## NOTE: Alias for backward compatibility. |
| 644 | $self->{state} = 'after DOCTYPE system identifier'; | |
| 645 | !!!next-input-character; | sub parse_char_stream ($$$;$$) { |
| 646 | redo A; | my $self = ref $_[0] ? shift : shift->new; |
| 647 | } elsif ($self->{next_input_character} == -1) { | my $input = $_[0]; |
| 648 | !!!parse-error (type => 'unclosed SYSTEM literal'); | $self->{document} = $_[1]; |
| 649 | @{$self->{document}->child_nodes} = (); | |
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| undef $self->{current_token}; | ||
| 650 | ||
| 651 | redo A; | ## NOTE: |set_inner_html| copies most of this method's code |
| } 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 | ||
| undef $self->{current_token}; | ||
| 652 | ||
| 653 | redo A; | ## Confidence: irrelevant. |
| 654 | } else { | $self->{confident} = 1 unless exists $self->{confident}; |
| $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 | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## recomsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| undef $self->{current_token}; | ||
| 655 | ||
| 656 | redo A; | $self->{document}->input_encoding ($self->{input_encoding}) |
| 657 | } else { | if defined $self->{input_encoding}; |
| 658 | !!!parse-error (type => 'string after SYSTEM literal'); | ## TODO: |{input_encoding}| is needless? |
| 659 | $self->{state} = 'bogus DOCTYPE'; | |
| 660 | !!!next-input-character; | $self->{line_prev} = $self->{line} = 1; |
| 661 | redo A; | $self->{column_prev} = -1; |
| 662 | } | $self->{column} = 0; |
| 663 | } elsif ($self->{state} eq 'bogus DOCTYPE') { | $self->{set_nc} = sub { |
| 664 | if ($self->{next_input_character} == 0x003E) { # > | my $self = shift; |
| $self->{state} = 'data'; | ||
| !!!next-input-character; | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| undef $self->{current_token}; | ||
| redo A; | ||
| } elsif ($self->{next_input_character} == -1) { | ||
| !!!parse-error (type => 'unclosed DOCTYPE'); | ||
| $self->{state} = 'data'; | ||
| ## reconsume | ||
| delete $self->{current_token}->{correct}; | ||
| !!!emit ($self->{current_token}); # DOCTYPE | ||
| undef $self->{current_token}; | ||
| 665 | ||
| 666 | redo A; | my $char = ''; |
| 667 | } else { | if (defined $self->{next_nc}) { |
| 668 | ## Stay in the state | $char = $self->{next_nc}; |
| 669 | !!!next-input-character; | delete $self->{next_nc}; |
| 670 | redo A; | $self->{nc} = ord $char; |
| } | ||
| 671 | } else { | } else { |
| 672 | die "$0: $self->{state}: Unknown state"; | $self->{char_buffer} = ''; |
| 673 | } | $self->{char_buffer_pos} = 0; |
| } # A | ||
| 674 | ||
| 675 | die "$0: _get_next_token: unexpected case"; | my $count = $input->manakai_read_until |
| 676 | } # _get_next_token | ($self->{char_buffer}, qr/[^\x00\x0A\x0D]/, $self->{char_buffer_pos}); |
| 677 | if ($count) { | |
| 678 | sub _tokenize_attempt_to_consume_an_entity ($) { | $self->{line_prev} = $self->{line}; |
| 679 | my $self = shift; | $self->{column_prev} = $self->{column}; |
| 680 | $self->{column}++; | |
| 681 | if ({ | $self->{nc} |
| 682 | 0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF, | = ord substr ($self->{char_buffer}, $self->{char_buffer_pos}++, 1); |
| 683 | 0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR | return; |
| }->{$self->{next_input_character}}) { | ||
| ## Don't consume | ||
| ## No error | ||
| return undef; | ||
| } elsif ($self->{next_input_character} == 0x0023) { # # | ||
| !!!next-input-character; | ||
| if ($self->{next_input_character} == 0x0078 or # x | ||
| $self->{next_input_character} == 0x0058) { # X | ||
| my $num; | ||
| X: { | ||
| my $x_char = $self->{next_input_character}; | ||
| !!!next-input-character; | ||
| if (0x0030 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x0039) { # 0..9 | ||
| $num ||= 0; | ||
| $num *= 0x10; | ||
| $num += $self->{next_input_character} - 0x0030; | ||
| redo X; | ||
| } elsif (0x0061 <= $self->{next_input_character} and | ||
| $self->{next_input_character} <= 0x0066) { # a..f | ||
| ## ISSUE: the spec says U+0078, which is apparently incorrect | ||
| $num ||= 0; | ||
| $num *= 0x10; | ||
| $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 += $self->{next_input_character} - 0x0030; | ||
| !!!next-input-character; | ||
| 684 | } | } |
| 685 | ||
| 686 | if ($self->{next_input_character} == 0x003B) { # ; | if ($input->read ($char, 1)) { |
| 687 | !!!next-input-character; | $self->{nc} = ord $char; |
| 688 | } else { | } else { |
| 689 | !!!parse-error (type => 'no refc'); | $self->{nc} = -1; |
| 690 | return; | |
| 691 | } | } |
| 692 | } | |
| 693 | ||
| 694 | ## TODO: check the definition for |a valid Unicode character|. | ($self->{line_prev}, $self->{column_prev}) |
| 695 | if ($code > 1114111 or $code == 0) { | = ($self->{line}, $self->{column}); |
| 696 | $code = 0xFFFD; # REPLACEMENT CHARACTER | $self->{column}++; |
| 697 | ## ISSUE: Why this is not an error? | |
| 698 | } elsif (0x80 <= $code and $code <= 0x9F) { | if ($self->{nc} == 0x000A) { # LF |
| 699 | !!!parse-error (type => sprintf 'c1 entity:U+%04X', $code); | !!!cp ('j1'); |
| 700 | $code = $c1_entity_char->{$code}; | $self->{line}++; |
| 701 | } | $self->{column} = 0; |
| 702 | } elsif ($self->{nc} == 0x000D) { # CR | |
| 703 | return {type => 'character', data => chr $code}; | !!!cp ('j2'); |
| 704 | } else { | ## TODO: support for abort/streaming |
| 705 | !!!parse-error (type => 'bare nero'); | my $next = ''; |
| 706 | !!!back-next-input-character ($self->{next_input_character}); | if ($input->read ($next, 1) and $next ne "\x0A") { |
| 707 | $self->{next_input_character} = 0x0023; # # | $self->{next_nc} = $next; |
| 708 | return undef; | } |
| 709 | $self->{nc} = 0x000A; # LF # MUST | |
| 710 | $self->{line}++; | |
| 711 | $self->{column} = 0; | |
| 712 | } elsif ($self->{nc} == 0x0000) { # NULL | |
| 713 | !!!cp ('j4'); | |
| 714 | !!!parse-error (type => 'NULL'); | |
| 715 | $self->{nc} = 0xFFFD; # REPLACEMENT CHARACTER # MUST | |
| 716 | } | } |
| 717 | } elsif ((0x0041 <= $self->{next_input_character} and | }; |
| 718 | $self->{next_input_character} <= 0x005A) or | |
| 719 | (0x0061 <= $self->{next_input_character} and | $self->{read_until} = sub { |
| 720 | $self->{next_input_character} <= 0x007A)) { | #my ($scalar, $specials_range, $offset) = @_; |
| 721 | my $entity_name = chr $self->{next_input_character}; | return 0 if defined $self->{next_nc}; |
| 722 | !!!next-input-character; | |
| 723 | my $pattern = qr/[^$_[1]\x00\x0A\x0D]/; | |
| 724 | my $value = $entity_name; | my $offset = $_[2] || 0; |
| 725 | my $match; | |
| 726 | require Whatpm::_NamedEntityList; | if ($self->{char_buffer_pos} < length $self->{char_buffer}) { |
| 727 | our $EntityChar; | pos ($self->{char_buffer}) = $self->{char_buffer_pos}; |
| 728 | if ($self->{char_buffer} =~ /\G(?>$pattern)+/) { | |
| 729 | while (length $entity_name < 10 and | substr ($_[0], $offset) |
| 730 | ## NOTE: Some number greater than the maximum length of entity name | = substr ($self->{char_buffer}, $-[0], $+[0] - $-[0]); |
| 731 | ((0x0041 <= $self->{next_input_character} and # a | my $count = $+[0] - $-[0]; |
| 732 | $self->{next_input_character} <= 0x005A) or # x | if ($count) { |
| 733 | (0x0061 <= $self->{next_input_character} and # a | $self->{column} += $count; |
| 734 | $self->{next_input_character} <= 0x007A) or # z | $self->{char_buffer_pos} += $count; |
| 735 | (0x0030 <= $self->{next_input_character} and # 0 | $self->{line_prev} = $self->{line}; |
| 736 | $self->{next_input_character} <= 0x0039) or # 9 | $self->{column_prev} = $self->{column} - 1; |
| 737 | $self->{next_input_character} == 0x003B)) { # ; | $self->{nc} = -1; |
| $entity_name .= chr $self->{next_input_character}; | ||
| if (defined $EntityChar->{$entity_name}) { | ||
| $value = $EntityChar->{$entity_name}; | ||
| if ($self->{next_input_character} == 0x003B) { # ; | ||
| $match = 1; | ||
| !!!next-input-character; | ||
| last; | ||
| } else { | ||
| $match = -1; | ||
| 738 | } | } |
| 739 | return $count; | |
| 740 | } else { | } else { |
| 741 | $value .= chr $self->{next_input_character}; | return 0; |
| 742 | } | } |
| !!!next-input-character; | ||
| } | ||
| if ($match > 0) { | ||
| return {type => 'character', data => $value}; | ||
| } elsif ($match < 0) { | ||
| !!!parse-error (type => 'refc'); | ||
| return {type => 'character', data => $value}; | ||
| 743 | } else { | } else { |
| 744 | !!!parse-error (type => 'bare ero'); | my $count = $input->manakai_read_until ($_[0], $pattern, $_[2]); |
| 745 | ## NOTE: No characters are consumed in the spec. | if ($count) { |
| 746 | !!!back-token ({type => 'character', data => $value}); | $self->{column} += $count; |
| 747 | return undef; | $self->{line_prev} = $self->{line}; |
| 748 | $self->{column_prev} = $self->{column} - 1; | |
| 749 | $self->{nc} = -1; | |
| 750 | } | |
| 751 | return $count; | |
| 752 | } | } |
| 753 | }; # $self->{read_until} | |
| 754 | ||
| 755 | my $onerror = $_[2] || sub { | |
| 756 | my (%opt) = @_; | |
| 757 | my $line = $opt{token} ? $opt{token}->{line} : $opt{line}; | |
| 758 | my $column = $opt{token} ? $opt{token}->{column} : $opt{column}; | |
| 759 | warn "Parse error ($opt{type}) at line $line column $column\n"; | |
| 760 | }; | |
| 761 | $self->{parse_error} = sub { | |
| 762 | $onerror->(line => $self->{line}, column => $self->{column}, @_); | |
| 763 | }; | |
| 764 | ||
| 765 | my $char_onerror = sub { | |
| 766 | my (undef, $type, %opt) = @_; | |
| 767 | !!!parse-error (layer => 'encode', | |
| 768 | line => $self->{line}, column => $self->{column} + 1, | |
| 769 | %opt, type => $type); | |
| 770 | }; # $char_onerror | |
| 771 | ||
| 772 | if ($_[3]) { | |
| 773 | $input = $_[3]->($input); | |
| 774 | $input->onerror ($char_onerror); | |
| 775 | } else { | } else { |
| 776 | ## no characters are consumed | $input->onerror ($char_onerror) unless defined $input->onerror; |
| !!!parse-error (type => 'bare ero'); | ||
| return undef; | ||
| 777 | } | } |
| 778 | } # _tokenize_attempt_to_consume_an_entity | |
| 779 | $self->_initialize_tokenizer; | |
| 780 | $self->_initialize_tree_constructor; | |
| 781 | $self->_construct_tree; | |
| 782 | $self->_terminate_tree_constructor; | |
| 783 | ||
| 784 | delete $self->{parse_error}; # remove loop | |
| 785 | ||
| 786 | return $self->{document}; | |
| 787 | } # parse_char_stream | |
| 788 | ||
| 789 | sub new ($) { | |
| 790 | my $class = shift; | |
| 791 | my $self = bless { | |
| 792 | level => {must => 'm', | |
| 793 | should => 's', | |
| 794 | warn => 'w', | |
| 795 | info => 'i', | |
| 796 | uncertain => 'u'}, | |
| 797 | }, $class; | |
| 798 | $self->{set_nc} = sub { | |
| 799 | $self->{nc} = -1; | |
| 800 | }; | |
| 801 | $self->{parse_error} = sub { | |
| 802 | # | |
| 803 | }; | |
| 804 | $self->{change_encoding} = sub { | |
| 805 | # if ($_[0] is a supported encoding) { | |
| 806 | # run "change the encoding" algorithm; | |
| 807 | # throw Whatpm::HTML::RestartParser (charset => $new_encoding); | |
| 808 | # } | |
| 809 | }; | |
| 810 | $self->{application_cache_selection} = sub { | |
| 811 | # | |
| 812 | }; | |
| 813 | return $self; | |
| 814 | } # new | |
| 815 | ||
| 816 | ## Insertion modes | |
| 817 | ||
| 818 | sub AFTER_HTML_IMS () { 0b100 } | |
| 819 | sub HEAD_IMS () { 0b1000 } | |
| 820 | sub BODY_IMS () { 0b10000 } | |
| 821 | sub BODY_TABLE_IMS () { 0b100000 } | |
| 822 | sub TABLE_IMS () { 0b1000000 } | |
| 823 | sub ROW_IMS () { 0b10000000 } | |
| 824 | sub BODY_AFTER_IMS () { 0b100000000 } | |
| 825 | sub FRAME_IMS () { 0b1000000000 } | |
| 826 | sub SELECT_IMS () { 0b10000000000 } | |
| 827 | #sub IN_FOREIGN_CONTENT_IM () { 0b100000000000 } # see Whatpm::HTML::Tokenizer | |
| 828 | ## NOTE: "in foreign content" insertion mode is special; it is combined | |
| 829 | ## with the secondary insertion mode. In this parser, they are stored | |
| 830 | ## together in the bit-or'ed form. | |
| 831 | sub IN_CDATA_RCDATA_IM () { 0b1000000000000 } | |
| 832 | ## NOTE: "in CDATA/RCDATA" insertion mode is also special; it is | |
| 833 | ## combined with the original insertion mode. In thie parser, | |
| 834 | ## they are stored together in the bit-or'ed form. | |
| 835 | ||
| 836 | sub IM_MASK () { 0b11111111111 } | |
| 837 | ||
| 838 | ## NOTE: "initial" and "before html" insertion modes have no constants. | |
| 839 | ||
| 840 | ## NOTE: "after after body" insertion mode. | |
| 841 | sub AFTER_HTML_BODY_IM () { AFTER_HTML_IMS | BODY_AFTER_IMS } | |
| 842 | ||
| 843 | ## NOTE: "after after frameset" insertion mode. | |
| 844 | sub AFTER_HTML_FRAMESET_IM () { AFTER_HTML_IMS | FRAME_IMS } | |
| 845 | ||
| 846 | sub IN_HEAD_IM () { HEAD_IMS | 0b00 } | |
| 847 | sub IN_HEAD_NOSCRIPT_IM () { HEAD_IMS | 0b01 } | |
| 848 | sub AFTER_HEAD_IM () { HEAD_IMS | 0b10 } | |
| 849 | sub BEFORE_HEAD_IM () { HEAD_IMS | 0b11 } | |
| 850 | sub IN_BODY_IM () { BODY_IMS } | |
| 851 | sub IN_CELL_IM () { BODY_IMS | BODY_TABLE_IMS | 0b01 } | |
| 852 | sub IN_CAPTION_IM () { BODY_IMS | BODY_TABLE_IMS | 0b10 } | |
| 853 | sub IN_ROW_IM () { TABLE_IMS | ROW_IMS | 0b01 } | |
| 854 | sub IN_TABLE_BODY_IM () { TABLE_IMS | ROW_IMS | 0b10 } | |
| 855 | sub IN_TABLE_IM () { TABLE_IMS } | |
| 856 | sub AFTER_BODY_IM () { BODY_AFTER_IMS } | |
| 857 | sub IN_FRAMESET_IM () { FRAME_IMS | 0b01 } | |
| 858 | sub AFTER_FRAMESET_IM () { FRAME_IMS | 0b10 } | |
| 859 | sub IN_SELECT_IM () { SELECT_IMS | 0b01 } | |
| 860 | sub IN_SELECT_IN_TABLE_IM () { SELECT_IMS | 0b10 } | |
| 861 | sub IN_COLUMN_GROUP_IM () { 0b10 } | |
| 862 | ||
| 863 | sub _initialize_tree_constructor ($) { | sub _initialize_tree_constructor ($) { |
| 864 | my $self = shift; | my $self = shift; |
| # | Line 1702 sub _initialize_tree_constructor ($) { | Line 867 sub _initialize_tree_constructor ($) { |
| 867 | ## TODO: Turn mutation events off # MUST | ## TODO: Turn mutation events off # MUST |
| 868 | ## TODO: Turn loose Document option (manakai extension) on | ## TODO: Turn loose Document option (manakai extension) on |
| 869 | $self->{document}->manakai_is_html (1); # MUST | $self->{document}->manakai_is_html (1); # MUST |
| 870 | $self->{document}->set_user_data (manakai_source_line => 1); | |
| 871 | $self->{document}->set_user_data (manakai_source_column => 1); | |
| 872 | } # _initialize_tree_constructor | } # _initialize_tree_constructor |
| 873 | ||
| 874 | sub _terminate_tree_constructor ($) { | sub _terminate_tree_constructor ($) { |
| # | Line 1721 sub _construct_tree ($) { | Line 888 sub _construct_tree ($) { |
| 888 | ## When an interactive UA render the $self->{document} available | ## When an interactive UA render the $self->{document} available |
| 889 | ## to the user, or when it begin accepting user input, are | ## to the user, or when it begin accepting user input, are |
| 890 | ## 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 | ||
| 891 | ||
| 892 | !!!next-token; | !!!next-token; |
| 893 | ||
| $self->{insertion_mode} = 'before head'; | ||
| 894 | undef $self->{form_element}; | undef $self->{form_element}; |
| 895 | undef $self->{head_element}; | undef $self->{head_element}; |
| 896 | undef $self->{head_element_inserted}; | |
| 897 | $self->{open_elements} = []; | $self->{open_elements} = []; |
| 898 | undef $self->{inner_html_node}; | undef $self->{inner_html_node}; |
| 899 | undef $self->{ignore_newline}; | |
| 900 | ||
| 901 | ## NOTE: The "initial" insertion mode. | |
| 902 | $self->_tree_construction_initial; # MUST | $self->_tree_construction_initial; # MUST |
| 903 | ||
| 904 | ## NOTE: The "before html" insertion mode. | |
| 905 | $self->_tree_construction_root_element; | $self->_tree_construction_root_element; |
| 906 | $self->{insertion_mode} = BEFORE_HEAD_IM; | |
| 907 | ||
| 908 | ## NOTE: The "before head" insertion mode and so on. | |
| 909 | $self->_tree_construction_main; | $self->_tree_construction_main; |
| 910 | } # _construct_tree | } # _construct_tree |
| 911 | ||
| 912 | sub _tree_construction_initial ($) { | sub _tree_construction_initial ($) { |
| 913 | my $self = shift; | my $self = shift; |
| 914 | ||
| 915 | ## NOTE: "initial" insertion mode | |
| 916 | ||
| 917 | INITIAL: { | INITIAL: { |
| 918 | if ($token->{type} eq 'DOCTYPE') { | if ($token->{type} == DOCTYPE_TOKEN) { |
| 919 | ## NOTE: Conformance checkers MAY, instead of reporting "not HTML5" | ## NOTE: Conformance checkers MAY, instead of reporting "not |
| 920 | ## error, switch to a conformance checking mode for another | ## HTML5" error, switch to a conformance checking mode for |
| 921 | ## language. | ## another language. (We don't support such mode switchings; it |
| 922 | ## is nonsense to do anything different from what browsers do.) | |
| 923 | my $doctype_name = $token->{name}; | my $doctype_name = $token->{name}; |
| 924 | $doctype_name = '' unless defined $doctype_name; | $doctype_name = '' unless defined $doctype_name; |
| 925 | $doctype_name =~ tr/a-z/A-Z/; | my $doctype = $self->{document}->create_document_type_definition |
| 926 | if (not defined $token->{name} or # <!DOCTYPE> | ($doctype_name); |
| 927 | defined $token->{public_identifier} or | |
| 928 | defined $token->{system_identifier}) { | $doctype_name =~ tr/A-Z/a-z/; # ASCII case-insensitive |
| 929 | !!!parse-error (type => 'not HTML5'); | if ($doctype_name ne 'html') { |
| 930 | } elsif ($doctype_name ne 'HTML') { | !!!cp ('t1'); |
| 931 | ## ISSUE: ASCII case-insensitive? (in fact it does not matter) | !!!parse-error (type => 'not HTML5', token => $token); |
| 932 | !!!parse-error (type => 'not HTML5'); | } elsif (defined $token->{pubid}) { |
| 933 | !!!cp ('t2'); | |
| 934 | ## XXX Obsolete permitted DOCTYPEs | |
| 935 | !!!parse-error (type => 'not HTML5', token => $token); | |
| 936 | } elsif (defined $token->{sysid}) { | |
| 937 | if ($token->{sysid} eq 'about:legacy-compat') { | |
| 938 | !!!cp ('t1.2'); ## <!DOCTYPE HTML SYSTEM "about:legacy-compat"> | |
| 939 | !!!parse-error (type => 'XSLT-compat', token => $token, | |
| 940 | level => $self->{level}->{should}); | |
| 941 | } else { | |
| 942 | !!!parse-error (type => 'not HTML5', token => $token); | |
| 943 | } | |
| 944 | } else { ## <!DOCTYPE HTML> | |
| 945 | !!!cp ('t3'); | |
| 946 | # | |
| 947 | } | } |
| 948 | ||
| 949 | my $doctype = $self->{document}->create_document_type_definition | ## NOTE: Default value for both |public_id| and |system_id| attributes |
| 950 | ($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)? | ## are empty strings, so that we don't set any value in missing cases. |
| 951 | $doctype->public_id ($token->{public_identifier}) | $doctype->public_id ($token->{pubid}) if defined $token->{pubid}; |
| 952 | if defined $token->{public_identifier}; | $doctype->system_id ($token->{sysid}) if defined $token->{sysid}; |
| 953 | $doctype->system_id ($token->{system_identifier}) | |
| if defined $token->{system_identifier}; | ||
| 954 | ## NOTE: Other DocumentType attributes are null or empty lists. | ## NOTE: Other DocumentType attributes are null or empty lists. |
| 955 | ## ISSUE: internalSubset = null?? | ## In Firefox3, |internalSubset| attribute is set to the empty |
| 956 | ## string, while |null| is an allowed value for the attribute | |
| 957 | ## according to DOM3 Core. | |
| 958 | $self->{document}->append_child ($doctype); | $self->{document}->append_child ($doctype); |
| 959 | ||
| 960 | if (not $token->{correct} or $doctype_name ne 'HTML') { | if ($token->{quirks} or $doctype_name ne 'html') { |
| 961 | !!!cp ('t4'); | |
| 962 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 963 | } elsif (defined $token->{public_identifier}) { | } elsif (defined $token->{pubid}) { |
| 964 | my $pubid = $token->{public_identifier}; | my $pubid = $token->{pubid}; |
| 965 | $pubid =~ tr/a-z/A-z/; | $pubid =~ tr/a-z/A-z/; |
| 966 | if ({ | my $prefix = [ |
| 967 | "+//SILMARIL//DTD HTML PRO V0R11 19970101//EN" => 1, | "+//SILMARIL//DTD HTML PRO V0R11 19970101//", |
| 968 | "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1, | "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//", |
| 969 | "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1, | "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//", |
| 970 | "-//IETF//DTD HTML 2.0 LEVEL 1//EN" => 1, | "-//IETF//DTD HTML 2.0 LEVEL 1//", |
| 971 | "-//IETF//DTD HTML 2.0 LEVEL 2//EN" => 1, | "-//IETF//DTD HTML 2.0 LEVEL 2//", |
| 972 | "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//EN" => 1, | "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//", |
| 973 | "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//EN" => 1, | "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//", |
| 974 | "-//IETF//DTD HTML 2.0 STRICT//EN" => 1, | "-//IETF//DTD HTML 2.0 STRICT//", |
| 975 | "-//IETF//DTD HTML 2.0//EN" => 1, | "-//IETF//DTD HTML 2.0//", |
| 976 | "-//IETF//DTD HTML 2.1E//EN" => 1, | "-//IETF//DTD HTML 2.1E//", |
| 977 | "-//IETF//DTD HTML 3.0//EN" => 1, | "-//IETF//DTD HTML 3.0//", |
| 978 | "-//IETF//DTD HTML 3.0//EN//" => 1, | "-//IETF//DTD HTML 3.2 FINAL//", |
| 979 | "-//IETF//DTD HTML 3.2 FINAL//EN" => 1, | "-//IETF//DTD HTML 3.2//", |
| 980 | "-//IETF//DTD HTML 3.2//EN" => 1, | "-//IETF//DTD HTML 3//", |
| 981 | "-//IETF//DTD HTML 3//EN" => 1, | "-//IETF//DTD HTML LEVEL 0//", |
| 982 | "-//IETF//DTD HTML LEVEL 0//EN" => 1, | "-//IETF//DTD HTML LEVEL 1//", |
| 983 | "-//IETF//DTD HTML LEVEL 0//EN//2.0" => 1, | "-//IETF//DTD HTML LEVEL 2//", |
| 984 | "-//IETF//DTD HTML LEVEL 1//EN" => 1, | "-//IETF//DTD HTML LEVEL 3//", |
| 985 | "-//IETF//DTD HTML LEVEL 1//EN//2.0" => 1, | "-//IETF//DTD HTML STRICT LEVEL 0//", |
| 986 | "-//IETF//DTD HTML LEVEL 2//EN" => 1, | "-//IETF//DTD HTML STRICT LEVEL 1//", |
| 987 | "-//IETF//DTD HTML LEVEL 2//EN//2.0" => 1, | "-//IETF//DTD HTML STRICT LEVEL 2//", |
| 988 | "-//IETF//DTD HTML LEVEL 3//EN" => 1, | "-//IETF//DTD HTML STRICT LEVEL 3//", |
| 989 | "-//IETF//DTD HTML LEVEL 3//EN//3.0" => 1, | "-//IETF//DTD HTML STRICT//", |
| 990 | "-//IETF//DTD HTML STRICT LEVEL 0//EN" => 1, | "-//IETF//DTD HTML//", |
| 991 | "-//IETF//DTD HTML STRICT LEVEL 0//EN//2.0" => 1, | "-//METRIUS//DTD METRIUS PRESENTATIONAL//", |
| 992 | "-//IETF//DTD HTML STRICT LEVEL 1//EN" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//", |
| 993 | "-//IETF//DTD HTML STRICT LEVEL 1//EN//2.0" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//", |
| 994 | "-//IETF//DTD HTML STRICT LEVEL 2//EN" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//", |
| 995 | "-//IETF//DTD HTML STRICT LEVEL 2//EN//2.0" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//", |
| 996 | "-//IETF//DTD HTML STRICT LEVEL 3//EN" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//", |
| 997 | "-//IETF//DTD HTML STRICT LEVEL 3//EN//3.0" => 1, | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//", |
| 998 | "-//IETF//DTD HTML STRICT//EN" => 1, | "-//NETSCAPE COMM. CORP.//DTD HTML//", |
| 999 | "-//IETF//DTD HTML STRICT//EN//2.0" => 1, | "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//", |
| 1000 | "-//IETF//DTD HTML STRICT//EN//3.0" => 1, | "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//", |
| 1001 | "-//IETF//DTD HTML//EN" => 1, | "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//", |
| 1002 | "-//IETF//DTD HTML//EN//2.0" => 1, | "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//", |
| 1003 | "-//IETF//DTD HTML//EN//3.0" => 1, | "-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//", |
| 1004 | "-//METRIUS//DTD METRIUS PRESENTATIONAL//EN" => 1, | "-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//", |
| 1005 | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//EN" => 1, | "-//SPYGLASS//DTD HTML 2.0 EXTENDED//", |
| 1006 | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//EN" => 1, | "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//", |
| 1007 | "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//EN" => 1, | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//", |
| 1008 | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//EN" => 1, | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//", |
| 1009 | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//EN" => 1, | "-//W3C//DTD HTML 3 1995-03-24//", |
| 1010 | "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//EN" => 1, | "-//W3C//DTD HTML 3.2 DRAFT//", |
| 1011 | "-//NETSCAPE COMM. CORP.//DTD HTML//EN" => 1, | "-//W3C//DTD HTML 3.2 FINAL//", |
| 1012 | "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1, | "-//W3C//DTD HTML 3.2//", |
| 1013 | "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1, | "-//W3C//DTD HTML 3.2S DRAFT//", |
| 1014 | "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1, | "-//W3C//DTD HTML 4.0 FRAMESET//", |
| 1015 | "-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1, | "-//W3C//DTD HTML 4.0 TRANSITIONAL//", |
| 1016 | "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1, | "-//W3C//DTD HTML EXPERIMETNAL 19960712//", |
| 1017 | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1, | "-//W3C//DTD HTML EXPERIMENTAL 970421//", |
| 1018 | "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//EN" => 1, | "-//W3C//DTD W3 HTML//", |
| 1019 | "-//W3C//DTD HTML 3 1995-03-24//EN" => 1, | "-//W3O//DTD W3 HTML 3.0//", |
| 1020 | "-//W3C//DTD HTML 3.2 DRAFT//EN" => 1, | "-//WEBTECHS//DTD MOZILLA HTML 2.0//", |
| 1021 | "-//W3C//DTD HTML 3.2 FINAL//EN" => 1, | "-//WEBTECHS//DTD MOZILLA HTML//", |
| 1022 | "-//W3C//DTD HTML 3.2//EN" => 1, | ]; # $prefix |
| 1023 | "-//W3C//DTD HTML 3.2S DRAFT//EN" => 1, | my $match; |
| 1024 | "-//W3C//DTD HTML 4.0 FRAMESET//EN" => 1, | for (@$prefix) { |
| 1025 | "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN" => 1, | if (substr ($prefix, 0, length $_) eq $_) { |
| 1026 | "-//W3C//DTD HTML EXPERIMETNAL 19960712//EN" => 1, | $match = 1; |
| 1027 | "-//W3C//DTD HTML EXPERIMENTAL 970421//EN" => 1, | last; |
| 1028 | "-//W3C//DTD W3 HTML//EN" => 1, | } |
| 1029 | "-//W3O//DTD W3 HTML 3.0//EN" => 1, | } |
| 1030 | "-//W3O//DTD W3 HTML 3.0//EN//" => 1, | if ($match or |
| 1031 | "-//W3O//DTD W3 HTML STRICT 3.0//EN//" => 1, | $pubid eq "-//W3O//DTD W3 HTML STRICT 3.0//EN//" or |
| 1032 | "-//WEBTECHS//DTD MOZILLA HTML 2.0//EN" => 1, | $pubid eq "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" or |
| 1033 | "-//WEBTECHS//DTD MOZILLA HTML//EN" => 1, | $pubid eq "HTML") { |
| 1034 | "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" => 1, | !!!cp ('t5'); |
| "HTML" => 1, | ||
| }->{$pubid}) { | ||
| 1035 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1036 | } elsif ($pubid eq "-//W3C//DTD HTML 4.01 FRAMESET//EN" or | } elsif ($pubid =~ m[^-//W3C//DTD HTML 4.01 FRAMESET//] or |
| 1037 | $pubid eq "-//W3C//DTD HTML 4.01 TRANSITIONAL//EN") { | $pubid =~ m[^-//W3C//DTD HTML 4.01 TRANSITIONAL//]) { |
| 1038 | if (defined $token->{system_identifier}) { | if (defined $token->{sysid}) { |
| 1039 | !!!cp ('t6'); | |
| 1040 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1041 | } else { | } else { |
| 1042 | !!!cp ('t7'); | |
| 1043 | $self->{document}->manakai_compat_mode ('limited quirks'); | $self->{document}->manakai_compat_mode ('limited quirks'); |
| 1044 | } | } |
| 1045 | } elsif ($pubid eq "-//W3C//DTD XHTML 1.0 Frameset//EN" or | } elsif ($pubid =~ m[^-//W3C//DTD XHTML 1.0 FRAMESET//] or |
| 1046 | $pubid eq "-//W3C//DTD XHTML 1.0 Transitional//EN") { | $pubid =~ m[^-//W3C//DTD XHTML 1.0 TRANSITIONAL//]) { |
| 1047 | !!!cp ('t8'); | |
| 1048 | $self->{document}->manakai_compat_mode ('limited quirks'); | $self->{document}->manakai_compat_mode ('limited quirks'); |
| 1049 | } else { | |
| 1050 | !!!cp ('t9'); | |
| 1051 | } | } |
| 1052 | } else { | |
| 1053 | !!!cp ('t10'); | |
| 1054 | } | } |
| 1055 | if (defined $token->{system_identifier}) { | if (defined $token->{sysid}) { |
| 1056 | my $sysid = $token->{system_identifier}; | my $sysid = $token->{sysid}; |
| 1057 | $sysid =~ tr/A-Z/a-z/; | $sysid =~ tr/A-Z/a-z/; |
| 1058 | if ($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") { |
| 1059 | ## NOTE: Ensure that |PUBLIC "(limited quirks)" "(quirks)"| is | |
| 1060 | ## marked as quirks. | |
| 1061 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1062 | !!!cp ('t11'); | |
| 1063 | } else { | |
| 1064 | !!!cp ('t12'); | |
| 1065 | } | } |
| 1066 | } else { | |
| 1067 | !!!cp ('t13'); | |
| 1068 | } | } |
| 1069 | ||
| 1070 | ## Go to the root element phase. | ## Go to the "before html" insertion mode. |
| 1071 | !!!next-token; | !!!next-token; |
| 1072 | return; | return; |
| 1073 | } elsif ({ | } elsif ({ |
| 1074 | 'start tag' => 1, | START_TAG_TOKEN, 1, |
| 1075 | 'end tag' => 1, | END_TAG_TOKEN, 1, |
| 1076 | 'end-of-file' => 1, | END_OF_FILE_TOKEN, 1, |
| 1077 | }->{$token->{type}}) { | }->{$token->{type}}) { |
| 1078 | !!!parse-error (type => 'no DOCTYPE'); | !!!cp ('t14'); |
| 1079 | !!!parse-error (type => 'no DOCTYPE', token => $token); | |
| 1080 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1081 | ## Go to the root element phase | ## Go to the "before html" insertion mode. |
| 1082 | ## reprocess | ## reprocess |
| 1083 | !!!ack-later; | |
| 1084 | return; | return; |
| 1085 | } elsif ($token->{type} eq 'character') { | } elsif ($token->{type} == CHARACTER_TOKEN) { |
| 1086 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { # \x0D | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1087 | ## Ignore the token | ## Ignore the token |
| 1088 | ||
| 1089 | unless (length $token->{data}) { | unless (length $token->{data}) { |
| 1090 | ## Stay in the phase | !!!cp ('t15'); |
| 1091 | ## Stay in the insertion mode. | |
| 1092 | !!!next-token; | !!!next-token; |
| 1093 | redo INITIAL; | redo INITIAL; |
| 1094 | } else { | |
| 1095 | !!!cp ('t16'); | |
| 1096 | } | } |
| 1097 | } else { | |
| 1098 | !!!cp ('t17'); | |
| 1099 | } | } |
| 1100 | ||
| 1101 | !!!parse-error (type => 'no DOCTYPE'); | !!!parse-error (type => 'no DOCTYPE', token => $token); |
| 1102 | $self->{document}->manakai_compat_mode ('quirks'); | $self->{document}->manakai_compat_mode ('quirks'); |
| 1103 | ## Go to the root element phase | ## Go to the "before html" insertion mode. |
| 1104 | ## reprocess | ## reprocess |
| 1105 | return; | return; |
| 1106 | } elsif ($token->{type} eq 'comment') { | } elsif ($token->{type} == COMMENT_TOKEN) { |
| 1107 | !!!cp ('t18'); | |
| 1108 | my $comment = $self->{document}->create_comment ($token->{data}); | my $comment = $self->{document}->create_comment ($token->{data}); |
| 1109 | $self->{document}->append_child ($comment); | $self->{document}->append_child ($comment); |
| 1110 | ||
| 1111 | ## Stay in the phase. | ## Stay in the insertion mode. |
| 1112 | !!!next-token; | !!!next-token; |
| 1113 | redo INITIAL; | redo INITIAL; |
| 1114 | } else { | } else { |
| 1115 | die "$0: $token->{type}: Unknown token"; | die "$0: $token->{type}: Unknown token type"; |
| 1116 | } | } |
| 1117 | } # INITIAL | } # INITIAL |
| 1118 | ||
| 1119 | die "$0: _tree_construction_initial: This should be never reached"; | |
| 1120 | } # _tree_construction_initial | } # _tree_construction_initial |
| 1121 | ||
| 1122 | sub _tree_construction_root_element ($) { | sub _tree_construction_root_element ($) { |
| 1123 | my $self = shift; | my $self = shift; |
| 1124 | ||
| 1125 | ## NOTE: "before html" insertion mode. | |
| 1126 | ||
| 1127 | B: { | B: { |
| 1128 | if ($token->{type} eq 'DOCTYPE') { | if ($token->{type} == DOCTYPE_TOKEN) { |
| 1129 | !!!parse-error (type => 'in html:#DOCTYPE'); | !!!cp ('t19'); |
| 1130 | !!!parse-error (type => 'in html:#DOCTYPE', token => $token); | |
| 1131 | ## Ignore the token | ## Ignore the token |
| 1132 | ## Stay in the phase | ## Stay in the insertion mode. |
| 1133 | !!!next-token; | !!!next-token; |
| 1134 | redo B; | redo B; |
| 1135 | } elsif ($token->{type} eq 'comment') { | } elsif ($token->{type} == COMMENT_TOKEN) { |
| 1136 | !!!cp ('t20'); | |
| 1137 | my $comment = $self->{document}->create_comment ($token->{data}); | my $comment = $self->{document}->create_comment ($token->{data}); |
| 1138 | $self->{document}->append_child ($comment); | $self->{document}->append_child ($comment); |
| 1139 | ## Stay in the phase | ## Stay in the insertion mode. |
| 1140 | !!!next-token; | !!!next-token; |
| 1141 | redo B; | redo B; |
| 1142 | } elsif ($token->{type} eq 'character') { | } elsif ($token->{type} == CHARACTER_TOKEN) { |
| 1143 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1144 | $self->{document}->manakai_append_text ($1); | ## Ignore the token. |
| 1145 | ## ISSUE: DOM3 Core does not allow Document > Text | |
| 1146 | unless (length $token->{data}) { | unless (length $token->{data}) { |
| 1147 | ## Stay in the phase | !!!cp ('t21'); |
| 1148 | ## Stay in the insertion mode. | |
| 1149 | !!!next-token; | !!!next-token; |
| 1150 | redo B; | redo B; |
| 1151 | } else { | |
| 1152 | !!!cp ('t22'); | |
| 1153 | } | } |
| 1154 | } else { | |
| 1155 | !!!cp ('t23'); | |
| 1156 | } | } |
| 1157 | ||
| 1158 | $self->{application_cache_selection}->(undef); | |
| 1159 | ||
| 1160 | # | # |
| 1161 | } elsif ($token->{type} == START_TAG_TOKEN) { | |
| 1162 | if ($token->{tag_name} eq 'html') { | |
| 1163 | my $root_element; | |
| 1164 | !!!create-element ($root_element, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); | |
| 1165 | $self->{document}->append_child ($root_element); | |
| 1166 | push @{$self->{open_elements}}, | |
| 1167 | [$root_element, $el_category->{html}]; | |
| 1168 | ||
| 1169 | if ($token->{attributes}->{manifest}) { | |
| 1170 | !!!cp ('t24'); | |
| 1171 | $self->{application_cache_selection} | |
| 1172 | ->($token->{attributes}->{manifest}->{value}); | |
| 1173 | ## ISSUE: Spec is unclear on relative references. | |
| 1174 | ## According to Hixie (#whatwg 2008-03-19), it should be | |
| 1175 | ## resolved against the base URI of the document in HTML | |
| 1176 | ## or xml:base of the element in XHTML. | |
| 1177 | } else { | |
| 1178 | !!!cp ('t25'); | |
| 1179 | $self->{application_cache_selection}->(undef); | |
| 1180 | } | |
| 1181 | ||
| 1182 | !!!nack ('t25c'); | |
| 1183 | ||
| 1184 | !!!next-token; | |
| 1185 | return; ## Go to the "before head" insertion mode. | |
| 1186 | } else { | |
| 1187 | !!!cp ('t25.1'); | |
| 1188 | # | |
| 1189 | } | |
| 1190 | } elsif ({ | } elsif ({ |
| 1191 | 'start tag' => 1, | END_TAG_TOKEN, 1, |
| 1192 | 'end tag' => 1, | END_OF_FILE_TOKEN, 1, |
| 'end-of-file' => 1, | ||
| 1193 | }->{$token->{type}}) { | }->{$token->{type}}) { |
| 1194 | ## ISSUE: There is an issue in the spec | !!!cp ('t26'); |
| 1195 | # | # |
| 1196 | } else { | } else { |
| 1197 | die "$0: $token->{type}: Unknown token"; | die "$0: $token->{type}: Unknown token type"; |
| 1198 | } | } |
| 1199 | my $root_element; !!!create-element ($root_element, 'html'); | |
| 1200 | $self->{document}->append_child ($root_element); | my $root_element; |
| 1201 | push @{$self->{open_elements}}, [$root_element, 'html']; | !!!create-element ($root_element, $HTML_NS, 'html',, $token); |
| 1202 | #$phase = 'main'; | $self->{document}->append_child ($root_element); |
| 1203 | ## reprocess | push @{$self->{open_elements}}, [$root_element, $el_category->{html}]; |
| 1204 | #redo B; | |
| 1205 | return; | $self->{application_cache_selection}->(undef); |
| 1206 | ||
| 1207 | ## NOTE: Reprocess the token. | |
| 1208 | !!!ack-later; | |
| 1209 | return; ## Go to the "before head" insertion mode. | |
| 1210 | } # B | } # B |
| 1211 | ||
| 1212 | die "$0: _tree_construction_root_element: This should never be reached"; | |
| 1213 | } # _tree_construction_root_element | } # _tree_construction_root_element |
| 1214 | ||
| 1215 | sub _reset_insertion_mode ($) { | sub _reset_insertion_mode ($) { |
| # | Line 1965 sub _reset_insertion_mode ($) { | Line 1224 sub _reset_insertion_mode ($) { |
| 1224 | ||
| 1225 | ## Step 3 | ## Step 3 |
| 1226 | S3: { | S3: { |
| 1227 | $last = 1 if $self->{open_elements}->[0]->[0] eq $node->[0]; | if ($self->{open_elements}->[0]->[0] eq $node->[0]) { |
| 1228 | if (defined $self->{inner_html_node}) { | $last = 1; |
| 1229 | if ($self->{inner_html_node}->[1] eq 'td' or | if (defined $self->{inner_html_node}) { |
| 1230 | $self->{inner_html_node}->[1] eq 'th') { | !!!cp ('t28'); |
| 1231 | $node = $self->{inner_html_node}; | |
| 1232 | } else { | |
| 1233 | die "_reset_insertion_mode: t27"; | |
| 1234 | } | |
| 1235 | } | |
| 1236 | ||
| 1237 | ## Step 4..14 | |
| 1238 | my $new_mode; | |
| 1239 | if ($node->[1] & FOREIGN_EL) { | |
| 1240 | !!!cp ('t28.1'); | |
| 1241 | ## NOTE: Strictly spaking, the line below only applies to MathML and | |
| 1242 | ## SVG elements. Currently the HTML syntax supports only MathML and | |
| 1243 | ## SVG elements as foreigners. | |
| 1244 | $new_mode = IN_BODY_IM | IN_FOREIGN_CONTENT_IM; | |
| 1245 | } elsif ($node->[1] == TABLE_CELL_EL) { | |
| 1246 | if ($last) { | |
| 1247 | !!!cp ('t28.2'); | |
| 1248 | # | # |
| 1249 | } else { | } else { |
| 1250 | $node = $self->{inner_html_node}; | !!!cp ('t28.3'); |
| 1251 | $new_mode = IN_CELL_IM; | |
| 1252 | } | } |
| 1253 | } else { | |
| 1254 | !!!cp ('t28.4'); | |
| 1255 | $new_mode = { | |
| 1256 | select => IN_SELECT_IM, | |
| 1257 | ## NOTE: |option| and |optgroup| do not set | |
| 1258 | ## insertion mode to "in select" by themselves. | |
| 1259 | tr => IN_ROW_IM, | |
| 1260 | tbody => IN_TABLE_BODY_IM, | |
| 1261 | thead => IN_TABLE_BODY_IM, | |
| 1262 | tfoot => IN_TABLE_BODY_IM, | |
| 1263 | caption => IN_CAPTION_IM, | |
| 1264 | colgroup => IN_COLUMN_GROUP_IM, | |
| 1265 | table => IN_TABLE_IM, | |
| 1266 | head => IN_BODY_IM, # not in head! | |
| 1267 | body => IN_BODY_IM, | |
| 1268 | frameset => IN_FRAMESET_IM, | |
| 1269 | }->{$node->[0]->manakai_local_name}; | |
| 1270 | } | } |
| ## 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]}; | ||
| 1271 | $self->{insertion_mode} = $new_mode and return if defined $new_mode; | $self->{insertion_mode} = $new_mode and return if defined $new_mode; |
| 1272 | ||
| 1273 | ## Step 14 | ## Step 15 |
| 1274 | if ($node->[1] eq 'html') { | if ($node->[1] == HTML_EL) { |
| 1275 | unless (defined $self->{head_element}) { | unless (defined $self->{head_element}) { |
| 1276 | $self->{insertion_mode} = 'before head'; | !!!cp ('t29'); |
| 1277 | $self->{insertion_mode} = BEFORE_HEAD_IM; | |
| 1278 | } else { | } else { |
| 1279 | $self->{insertion_mode} = 'after head'; | ## ISSUE: Can this state be reached? |
| 1280 | !!!cp ('t30'); | |
| 1281 | $self->{insertion_mode} = AFTER_HEAD_IM; | |
| 1282 | } | } |
| 1283 | return; | return; |
| 1284 | } else { | |
| 1285 | !!!cp ('t31'); | |
| 1286 | } | } |
| 1287 | ||
| ## Step 15 | ||
| $self->{insertion_mode} = 'in body' and return if $last; | ||
| 1288 | ## Step 16 | ## Step 16 |
| 1289 | $self->{insertion_mode} = IN_BODY_IM and return if $last; | |
| 1290 | ||
| 1291 | ## Step 17 | |
| 1292 | $i--; | $i--; |
| 1293 | $node = $self->{open_elements}->[$i]; | $node = $self->{open_elements}->[$i]; |
| 1294 | ||
| 1295 | ## Step 17 | ## Step 18 |
| 1296 | redo S3; | redo S3; |
| 1297 | } # S3 | } # S3 |
| 1298 | ||
| 1299 | die "$0: _reset_insertion_mode: This line should never be reached"; | |
| 1300 | } # _reset_insertion_mode | } # _reset_insertion_mode |
| 1301 | ||
| 1302 | sub _tree_construction_main ($) { | sub _tree_construction_main ($) { |
| 1303 | my $self = shift; | my $self = shift; |
| 1304 | ||
| my $phase = 'main'; | ||
| 1305 | my $active_formatting_elements = []; | my $active_formatting_elements = []; |
| 1306 | ||
| 1307 | my $reconstruct_active_formatting_elements = sub { # MUST | my $reconstruct_active_formatting_elements = sub { # MUST |
| # | Line 2036 sub _tree_construction_main ($) { | Line 1318 sub _tree_construction_main ($) { |
| 1318 | return if $entry->[0] eq '#marker'; | return if $entry->[0] eq '#marker'; |
| 1319 | for (@{$self->{open_elements}}) { | for (@{$self->{open_elements}}) { |
| 1320 | if ($entry->[0] eq $_->[0]) { | if ($entry->[0] eq $_->[0]) { |
| 1321 | !!!cp ('t32'); | |
| 1322 | return; | return; |
| 1323 | } | } |
| 1324 | } | } |
| # | Line 2050 sub _tree_construction_main ($) { | Line 1333 sub _tree_construction_main ($) { |
| 1333 | ||
| 1334 | ## Step 6 | ## Step 6 |
| 1335 | if ($entry->[0] eq '#marker') { | if ($entry->[0] eq '#marker') { |
| 1336 | !!!cp ('t33_1'); | |
| 1337 | # | # |
| 1338 | } else { | } else { |
| 1339 | my $in_open_elements; | my $in_open_elements; |
| 1340 | OE: for (@{$self->{open_elements}}) { | OE: for (@{$self->{open_elements}}) { |
| 1341 | if ($entry->[0] eq $_->[0]) { | if ($entry->[0] eq $_->[0]) { |
| 1342 | !!!cp ('t33'); | |
| 1343 | $in_open_elements = 1; | $in_open_elements = 1; |
| 1344 | last OE; | last OE; |
| 1345 | } | } |
| 1346 | } | } |
| 1347 | if ($in_open_elements) { | if ($in_open_elements) { |
| 1348 | !!!cp ('t34'); | |
| 1349 | # | # |
| 1350 | } else { | } else { |
| 1351 | ## NOTE: <!DOCTYPE HTML><p><b><i><u></p> <p>X | |
| 1352 | !!!cp ('t35'); | |
| 1353 | redo S4; | redo S4; |
| 1354 | } | } |
| 1355 | } | } |
| # | Line 2084 sub _tree_construction_main ($) { | Line 1372 sub _tree_construction_main ($) { |
| 1372 | ||
| 1373 | ## Step 11 | ## Step 11 |
| 1374 | unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { | unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { |
| 1375 | !!!cp ('t36'); | |
| 1376 | ## Step 7' | ## Step 7' |
| 1377 | $i++; | $i++; |
| 1378 | $entry = $active_formatting_elements->[$i]; | $entry = $active_formatting_elements->[$i]; |
| 1379 | ||
| 1380 | redo S7; | redo S7; |
| 1381 | } | } |
| 1382 | ||
| 1383 | !!!cp ('t37'); | |
| 1384 | } # S7 | } # S7 |
| 1385 | }; # $reconstruct_active_formatting_elements | }; # $reconstruct_active_formatting_elements |
| 1386 | ||
| 1387 | my $clear_up_to_marker = sub { | my $clear_up_to_marker = sub { |
| 1388 | for (reverse 0..$#$active_formatting_elements) { | for (reverse 0..$#$active_formatting_elements) { |
| 1389 | if ($active_formatting_elements->[$_]->[0] eq '#marker') { | if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1390 | !!!cp ('t38'); | |
| 1391 | splice @$active_formatting_elements, $_; | splice @$active_formatting_elements, $_; |
| 1392 | return; | return; |
| 1393 | } | } |
| 1394 | } | } |
| 1395 | ||
| 1396 | !!!cp ('t39'); | |
| 1397 | }; # $clear_up_to_marker | }; # $clear_up_to_marker |
| 1398 | ||
| 1399 | my $style_start_tag = sub { | my $insert; |
| 1400 | my $style_el; !!!create-element ($style_el, 'style', $token->{attributes}); | |
| 1401 | ## $self->{insertion_mode} eq 'in head' and ... (always true) | my $parse_rcdata = sub ($) { |
| 1402 | (($self->{insertion_mode} eq 'in head' and defined $self->{head_element}) | my ($content_model_flag) = @_; |
| 1403 | ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | |
| 1404 | ->append_child ($style_el); | ## Step 1 |
| 1405 | $self->{content_model_flag} = 'CDATA'; | my $start_tag_name = $token->{tag_name}; |
| 1406 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); | |
| 1407 | ||
| 1408 | ## Step 2 | |
| 1409 | $self->{content_model} = $content_model_flag; # CDATA or RCDATA | |
| 1410 | delete $self->{escape}; # MUST | delete $self->{escape}; # MUST |
| 1411 | ||
| 1412 | my $text = ''; | ## Step 3, 4 |
| 1413 | !!!next-token; | $self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
| 1414 | while ($token->{type} eq 'character') { | |
| 1415 | $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? | ||
| } | ||
| 1416 | !!!next-token; | !!!next-token; |
| 1417 | }; # $style_start_tag | }; # $parse_rcdata |
| 1418 | ||
| 1419 | my $script_start_tag = sub { | my $script_start_tag = sub () { |
| 1420 | ## Step 1 | |
| 1421 | my $script_el; | my $script_el; |
| 1422 | !!!create-element ($script_el, 'script', $token->{attributes}); | !!!create-element ($script_el, $HTML_NS, 'script', $token->{attributes}, $token); |
| 1423 | ||
| 1424 | ## Step 2 | |
| 1425 | ## TODO: mark as "parser-inserted" | ## TODO: mark as "parser-inserted" |
| 1426 | ||
| 1427 | $self->{content_model_flag} = 'CDATA'; | ## Step 3 |
| 1428 | ## TODO: Mark as "already executed", if ... | |
| 1429 | ||
| 1430 | ## Step 4 (HTML5 revision 2702) | |
| 1431 | $insert->($script_el); | |
| 1432 | push @{$self->{open_elements}}, [$script_el, $el_category->{script}]; | |
| 1433 | ||
| 1434 | ## Step 5 | |
| 1435 | $self->{content_model} = CDATA_CONTENT_MODEL; | |
| 1436 | 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); | ||
| } | ||
| $self->{content_model_flag} = 'PCDATA'; | ||
| 1437 | ||
| 1438 | if ($token->{type} eq 'end tag' and | ## Step 6-7 |
| 1439 | $token->{tag_name} eq 'script') { | $self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
| 1440 | ## Ignore the token | |
| 1441 | } else { | !!!nack ('t40.2'); |
| !!!parse-error (type => 'in CDATA:#'.$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... | ||
| } | ||
| 1442 | !!!next-token; | !!!next-token; |
| 1443 | }; # $script_start_tag | }; # $script_start_tag |
| 1444 | ||
| 1445 | ## NOTE: $open_tables->[-1]->[0] is the "current table" element node. | |
| 1446 | ## NOTE: $open_tables->[-1]->[1] is the "tainted" flag (OBSOLETE; unused). | |
| 1447 | ## NOTE: $open_tables->[-1]->[2] is set false when non-Text node inserted. | |
| 1448 | my $open_tables = [[$self->{open_elements}->[0]->[0]]]; | |
| 1449 | ||
| 1450 | my $formatting_end_tag = sub { | my $formatting_end_tag = sub { |
| 1451 | my $tag_name = shift; | my $end_tag_token = shift; |
| 1452 | my $tag_name = $end_tag_token->{tag_name}; | |
| 1453 | ||
| 1454 | ## NOTE: The adoption agency algorithm (AAA). | |
| 1455 | ||
| 1456 | FET: { | FET: { |
| 1457 | ## Step 1 | ## Step 1 |
| 1458 | my $formatting_element; | my $formatting_element; |
| 1459 | my $formatting_element_i_in_active; | my $formatting_element_i_in_active; |
| 1460 | AFE: for (reverse 0..$#$active_formatting_elements) { | AFE: for (reverse 0..$#$active_formatting_elements) { |
| 1461 | if ($active_formatting_elements->[$_]->[1] eq $tag_name) { | if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1462 | !!!cp ('t52'); | |
| 1463 | last AFE; | |
| 1464 | } elsif ($active_formatting_elements->[$_]->[0]->manakai_local_name | |
| 1465 | eq $tag_name) { | |
| 1466 | !!!cp ('t51'); | |
| 1467 | $formatting_element = $active_formatting_elements->[$_]; | $formatting_element = $active_formatting_elements->[$_]; |
| 1468 | $formatting_element_i_in_active = $_; | $formatting_element_i_in_active = $_; |
| 1469 | last AFE; | last AFE; |
| } elsif ($active_formatting_elements->[$_]->[0] eq '#marker') { | ||
| last AFE; | ||
| 1470 | } | } |
| 1471 | } # AFE | } # AFE |
| 1472 | unless (defined $formatting_element) { | unless (defined $formatting_element) { |
| 1473 | !!!parse-error (type => 'unmatched end tag:'.$tag_name); | !!!cp ('t53'); |
| 1474 | !!!parse-error (type => 'unmatched end tag', text => $tag_name, token => $end_tag_token); | |
| 1475 | ## Ignore the token | ## Ignore the token |
| 1476 | !!!next-token; | !!!next-token; |
| 1477 | return; | return; |
| # | Line 2207 sub _tree_construction_main ($) { | Line 1483 sub _tree_construction_main ($) { |
| 1483 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 1484 | if ($node->[0] eq $formatting_element->[0]) { | if ($node->[0] eq $formatting_element->[0]) { |
| 1485 | if ($in_scope) { | if ($in_scope) { |
| 1486 | !!!cp ('t54'); | |
| 1487 | $formatting_element_i_in_open = $_; | $formatting_element_i_in_open = $_; |
| 1488 | last INSCOPE; | last INSCOPE; |
| 1489 | } else { # in open elements but not in scope | } else { # in open elements but not in scope |
| 1490 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t55'); |
| 1491 | !!!parse-error (type => 'unmatched end tag', | |
| 1492 | text => $token->{tag_name}, | |
| 1493 | token => $end_tag_token); | |
| 1494 | ## Ignore the token | ## Ignore the token |
| 1495 | !!!next-token; | !!!next-token; |
| 1496 | return; | return; |
| 1497 | } | } |
| 1498 | } elsif ({ | } elsif ($node->[1] & SCOPING_EL) { |
| 1499 | table => 1, caption => 1, td => 1, th => 1, | !!!cp ('t56'); |
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| 1500 | $in_scope = 0; | $in_scope = 0; |
| 1501 | } | } |
| 1502 | } # INSCOPE | } # INSCOPE |
| 1503 | unless (defined $formatting_element_i_in_open) { | unless (defined $formatting_element_i_in_open) { |
| 1504 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | !!!cp ('t57'); |
| 1505 | !!!parse-error (type => 'unmatched end tag', | |
| 1506 | text => $token->{tag_name}, | |
| 1507 | token => $end_tag_token); | |
| 1508 | pop @$active_formatting_elements; # $formatting_element | pop @$active_formatting_elements; # $formatting_element |
| 1509 | !!!next-token; ## TODO: ok? | !!!next-token; ## TODO: ok? |
| 1510 | return; | return; |
| 1511 | } | } |
| 1512 | if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { | if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { |
| 1513 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | !!!cp ('t58'); |
| 1514 | !!!parse-error (type => 'not closed', | |
| 1515 | text => $self->{open_elements}->[-1]->[0] | |
| 1516 | ->manakai_local_name, | |
| 1517 | token => $end_tag_token); | |
| 1518 | } | } |
| 1519 | ||
| 1520 | ## Step 2 | ## Step 2 |
| # | Line 2237 sub _tree_construction_main ($) { | Line 1522 sub _tree_construction_main ($) { |
| 1522 | my $furthest_block_i_in_open; | my $furthest_block_i_in_open; |
| 1523 | OE: for (reverse 0..$#{$self->{open_elements}}) { | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1524 | my $node = $self->{open_elements}->[$_]; | my $node = $self->{open_elements}->[$_]; |
| 1525 | if (not $formatting_category->{$node->[1]} and | if (not ($node->[1] & FORMATTING_EL) and |
| 1526 | #not $phrasing_category->{$node->[1]} and | #not $phrasing_category->{$node->[1]} and |
| 1527 | ($special_category->{$node->[1]} or | ($node->[1] & SPECIAL_EL or |
| 1528 | $scoping_category->{$node->[1]})) { | $node->[1] & SCOPING_EL)) { ## Scoping is redundant, maybe |
| 1529 | !!!cp ('t59'); | |
| 1530 | $furthest_block = $node; | $furthest_block = $node; |
| 1531 | $furthest_block_i_in_open = $_; | $furthest_block_i_in_open = $_; |
| 1532 | ## NOTE: The topmost (eldest) node. | |
| 1533 | } elsif ($node->[0] eq $formatting_element->[0]) { | } elsif ($node->[0] eq $formatting_element->[0]) { |
| 1534 | !!!cp ('t60'); | |
| 1535 | last OE; | last OE; |
| 1536 | } | } |
| 1537 | } # OE | } # OE |
| 1538 | ||
| 1539 | ## Step 3 | ## Step 3 |
| 1540 | unless (defined $furthest_block) { # MUST | unless (defined $furthest_block) { # MUST |
| 1541 | !!!cp ('t61'); | |
| 1542 | splice @{$self->{open_elements}}, $formatting_element_i_in_open; | splice @{$self->{open_elements}}, $formatting_element_i_in_open; |
| 1543 | splice @$active_formatting_elements, $formatting_element_i_in_active, 1; | splice @$active_formatting_elements, $formatting_element_i_in_active, 1; |
| 1544 | !!!next-token; | !!!next-token; |
| # | Line 2262 sub _tree_construction_main ($) { | Line 1551 sub _tree_construction_main ($) { |
| 1551 | ## Step 5 | ## Step 5 |
| 1552 | my $furthest_block_parent = $furthest_block->[0]->parent_node; | my $furthest_block_parent = $furthest_block->[0]->parent_node; |
| 1553 | if (defined $furthest_block_parent) { | if (defined $furthest_block_parent) { |
| 1554 | !!!cp ('t62'); | |
| 1555 | $furthest_block_parent->remove_child ($furthest_block->[0]); | $furthest_block_parent->remove_child ($furthest_block->[0]); |
| 1556 | } | } |
| 1557 | ||
| # | Line 2284 sub _tree_construction_main ($) { | Line 1574 sub _tree_construction_main ($) { |
| 1574 | S7S2: { | S7S2: { |
| 1575 | for (reverse 0..$#$active_formatting_elements) { | for (reverse 0..$#$active_formatting_elements) { |
| 1576 | if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { | if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
| 1577 | !!!cp ('t63'); | |
| 1578 | $node_i_in_active = $_; | $node_i_in_active = $_; |
| 1579 | last S7S2; | last S7S2; |
| 1580 | } | } |
| # | Line 2297 sub _tree_construction_main ($) { | Line 1588 sub _tree_construction_main ($) { |
| 1588 | ||
| 1589 | ## Step 4 | ## Step 4 |
| 1590 | if ($last_node->[0] eq $furthest_block->[0]) { | if ($last_node->[0] eq $furthest_block->[0]) { |
| 1591 | !!!cp ('t64'); | |
| 1592 | $bookmark_prev_el = $node->[0]; | $bookmark_prev_el = $node->[0]; |
| 1593 | } | } |
| 1594 | ||
| 1595 | ## Step 5 | ## Step 5 |
| 1596 | if ($node->[0]->has_child_nodes ()) { | if ($node->[0]->has_child_nodes ()) { |
| 1597 | !!!cp ('t65'); | |
| 1598 | my $clone = [$node->[0]->clone_node (0), $node->[1]]; | my $clone = [$node->[0]->clone_node (0), $node->[1]]; |
| 1599 | $active_formatting_elements->[$node_i_in_active] = $clone; | $active_formatting_elements->[$node_i_in_active] = $clone; |
| 1600 | $self->{open_elements}->[$node_i_in_open] = $clone; | $self->{open_elements}->[$node_i_in_open] = $clone; |
| # | Line 2319 sub _tree_construction_main ($) { | Line 1612 sub _tree_construction_main ($) { |
| 1612 | } # S7 | } # S7 |
| 1613 | ||
| 1614 | ## Step 8 | ## Step 8 |
| 1615 | $common_ancestor_node->[0]->append_child ($last_node->[0]); | if ($common_ancestor_node->[1] & TABLE_ROWS_EL) { |
| 1616 | ## Foster parenting. | |
| 1617 | my $foster_parent_element; | |
| 1618 | my $next_sibling; | |
| 1619 | OE: for (reverse 0..$#{$self->{open_elements}}) { | |
| 1620 | if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { | |
| 1621 | !!!cp ('t65.2'); | |
| 1622 | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; | |
| 1623 | $next_sibling = $self->{open_elements}->[$_]->[0]; | |
| 1624 | undef $next_sibling | |
| 1625 | unless $next_sibling->parent_node eq $foster_parent_element; | |
| 1626 | last OE; | |
| 1627 | } | |
| 1628 | } # OE | |
| 1629 | $foster_parent_element ||= $self->{open_elements}->[0]->[0]; | |
| 1630 | ||
| 1631 | $foster_parent_element->insert_before ($last_node->[0], $next_sibling); | |
| 1632 | $open_tables->[-1]->[1] = 1; # tainted | |
| 1633 | } else { | |
| 1634 | !!!cp ('t65.3'); | |
| 1635 | $common_ancestor_node->[0]->append_child ($last_node->[0]); | |
| 1636 | } | |
| 1637 | ||
| 1638 | ## Step 9 | ## Step 9 |
| 1639 | my $clone = [$formatting_element->[0]->clone_node (0), | my $clone = [$formatting_element->[0]->clone_node (0), |
| # | Line 2336 sub _tree_construction_main ($) { | Line 1650 sub _tree_construction_main ($) { |
| 1650 | my $i; | my $i; |
| 1651 | AFE: for (reverse 0..$#$active_formatting_elements) { | AFE: for (reverse 0..$#$active_formatting_elements) { |
| 1652 | if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { | if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { |
| 1653 | !!!cp ('t66'); | |
| 1654 | splice @$active_formatting_elements, $_, 1; | splice @$active_formatting_elements, $_, 1; |
| 1655 | $i-- and last AFE if defined $i; | $i-- and last AFE if defined $i; |
| 1656 | } elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { | } elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { |
| 1657 | !!!cp ('t67'); | |
| 1658 | $i = $_; | $i = $_; |
| 1659 | } | } |
| 1660 | } # AFE | } # AFE |
| # | Line 2348 sub _tree_construction_main ($) { | Line 1664 sub _tree_construction_main ($) { |
| 1664 | undef $i; | undef $i; |
| 1665 | OE: for (reverse 0..$#{$self->{open_elements}}) { | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1666 | if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { | if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { |
| 1667 | !!!cp ('t68'); | |
| 1668 | splice @{$self->{open_elements}}, $_, 1; | splice @{$self->{open_elements}}, $_, 1; |
| 1669 | $i-- and last OE if defined $i; | $i-- and last OE if defined $i; |
| 1670 | } elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { | } elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { |
| 1671 | !!!cp ('t69'); | |
| 1672 | $i = $_; | $i = $_; |
| 1673 | } | } |
| 1674 | } # OE | } # OE |
| 1675 | splice @{$self->{open_elements}}, $i + 1, 1, $clone; | splice @{$self->{open_elements}}, $i + 1, 0, $clone; |
| 1676 | ||
| 1677 | ## Step 14 | ## Step 14 |
| 1678 | redo FET; | redo FET; |
| 1679 | } # FET | } # FET |
| 1680 | }; # $formatting_end_tag | }; # $formatting_end_tag |
| 1681 | ||
| 1682 | my $insert_to_current = sub { | $insert = my $insert_to_current = sub { |
| 1683 | $self->{open_elements}->[-1]->[0]->append_child (shift); | $self->{open_elements}->[-1]->[0]->append_child ($_[0]); |
| 1684 | }; # $insert_to_current | }; # $insert_to_current |
| 1685 | ||
| 1686 | ## Foster parenting. Note that there are three "foster parenting" | |
| 1687 | ## code in the parser: for elements (this one), for texts, and for | |
| 1688 | ## elements in the AAA code. | |
| 1689 | my $insert_to_foster = sub { | my $insert_to_foster = sub { |
| 1690 | my $child = shift; | my $child = shift; |
| 1691 | if ({ | if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1692 | table => 1, tbody => 1, tfoot => 1, | # MUST |
| 1693 | thead => 1, tr => 1, | my $foster_parent_element; |
| 1694 | }->{$self->{open_elements}->[-1]->[1]}) { | my $next_sibling; |
| 1695 | # MUST | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1696 | my $foster_parent_element; | if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1697 | my $next_sibling; | !!!cp ('t71'); |
| 1698 | OE: for (reverse 0..$#{$self->{open_elements}}) { | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1699 | if ($self->{open_elements}->[$_]->[1] eq 'table') { | $next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1700 | my $parent = $self->{open_elements}->[$_]->[0]->parent_node; | undef $next_sibling |
| 1701 | if (defined $parent and $parent->node_type == 1) { | unless $next_sibling->parent_node eq $foster_parent_element; |
| 1702 | $foster_parent_element = $parent; | last OE; |
| 1703 | $next_sibling = $self->{open_elements}->[$_]->[0]; | } |
| 1704 | } else { | } # OE |
| 1705 | $foster_parent_element | $foster_parent_element ||= $self->{open_elements}->[0]->[0]; |
| 1706 | = $self->{open_elements}->[$_ - 1]->[0]; | |
| 1707 | } | $foster_parent_element->insert_before ($child, $next_sibling); |
| 1708 | last OE; | $open_tables->[-1]->[1] = 1; # tainted |
| 1709 | } | } else { |
| 1710 | } # OE | !!!cp ('t72'); |
| 1711 | $foster_parent_element = $self->{open_elements}->[0]->[0] | $self->{open_elements}->[-1]->[0]->append_child ($child); |
| 1712 | unless defined $foster_parent_element; | } |
| $foster_parent_element->insert_before | ||
| ($child, $next_sibling); | ||
| } else { | ||
| $self->{open_elements}->[-1]->[0]->append_child ($child); | ||
| } | ||
| 1713 | }; # $insert_to_foster | }; # $insert_to_foster |
| 1714 | ||
| 1715 | my $in_body = sub { | ## NOTE: Insert a character (MUST): When a character is inserted, if |
| 1716 | my $insert = shift; | ## the last node that was inserted by the parser is a Text node and |
| 1717 | if ($token->{type} eq 'start tag') { | ## the character has to be inserted after that node, then the |
| 1718 | if ($token->{tag_name} eq 'script') { | ## character is appended to the Text node. However, if any other |
| 1719 | $script_start_tag->(); | ## node is inserted by the parser, then a new Text node is created |
| 1720 | return; | ## and the character is appended as that Text node. If I'm not |
| 1721 | } elsif ($token->{tag_name} eq 'style') { | ## wrong, for a parser with scripting disabled, there are only two |
| 1722 | $style_start_tag->(); | ## cases where this occurs. One is the case where an element node |
| 1723 | return; | ## is inserted to the |head| element. This is covered by using the |
| 1724 | } elsif ({ | ## |$self->{head_element_inserted}| flag. Another is the case where |
| 1725 | base => 1, link => 1, meta => 1, | ## an element or comment is inserted into the |table| subtree while |
| 1726 | }->{$token->{tag_name}}) { | ## foster parenting happens. This is covered by using the [2] flag |
| 1727 | !!!parse-error (type => 'in body:'.$token->{tag_name}); | ## of the |$open_tables| structure. All other cases are handled |
| 1728 | ## NOTE: This is an "as if in head" code clone | ## simply by calling |manakai_append_text| method. |
| 1729 | my $el; | |
| 1730 | !!!create-element ($el, $token->{tag_name}, $token->{attributes}); | ## TODO: |<body><script>document.write("a<br>"); |
| 1731 | if (defined $self->{head_element}) { | ## document.body.removeChild (document.body.lastChild); |
| 1732 | $self->{head_element}->append_child ($el); | ## document.write ("b")</script>| |
| 1733 | } else { | |
| 1734 | $insert->($el); | B: while (1) { |
| 1735 | } | |
| 1736 | ## The "in table text" insertion mode. | |
| 1737 | !!!next-token; | if ($self->{insertion_mode} & TABLE_IMS and |
| 1738 | return; | not $self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and |
| 1739 | } elsif ($token->{tag_name} eq 'title') { | not $self->{insertion_mode} & IN_CDATA_RCDATA_IM) { |
| 1740 | !!!parse-error (type => 'in body:title'); | C: { |
| 1741 | ## NOTE: There is an "as if in head" code clone | my $s; |
| 1742 | my $title_el; | if ($token->{type} == CHARACTER_TOKEN) { |
| 1743 | !!!create-element ($title_el, 'title', $token->{attributes}); | !!!cp ('t194'); |
| 1744 | (defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | $self->{pending_chars} ||= []; |
| 1745 | ->append_child ($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}; | ||
| 1746 | !!!next-token; | !!!next-token; |
| 1747 | } | next B; |
| if (length $text) { | ||
| $title_el->manakai_append_text ($text); | ||
| } | ||
| $self->{content_model_flag} = 'PCDATA'; | ||
| if ($token->{type} eq 'end tag' and | ||
| $token->{tag_name} eq 'title') { | ||
| ## Ignore the token | ||
| } else { | ||
| !!!parse-error (type => 'in RCDATA:#'.$token->{type}); | ||
| ## ISSUE: And ignore? | ||
| } | ||
| !!!next-token; | ||
| return; | ||
| } elsif ($token->{tag_name} eq 'body') { | ||
| !!!parse-error (type => 'in body:body'); | ||
| if (@{$self->{open_elements}} == 1 or | ||
| $self->{open_elements}->[1]->[1] ne 'body') { | ||
| ## Ignore the token | ||
| 1748 | } else { | } else { |
| 1749 | my $body_el = $self->{open_elements}->[1]->[0]; | if ($self->{pending_chars}) { |
| 1750 | for my $attr_name (keys %{$token->{attributes}}) { | $s = join '', map { $_->{data} } @{$self->{pending_chars}}; |
| 1751 | unless ($body_el->has_attribute_ns (undef, $attr_name)) { | delete $self->{pending_chars}; |
| 1752 | $body_el->set_attribute_ns | if ($s =~ /[^\x09\x0A\x0C\x0D\x20]/) { |
| 1753 | (undef, [undef, $attr_name], | !!!cp ('t195'); |
| 1754 | $token->{attributes}->{$attr_name}->{value}); | # |
| 1755 | } else { | |
| 1756 | !!!cp ('t195.1'); | |
| 1757 | #$self->{open_elements}->[-1]->[0]->manakai_append_text ($s); | |
| 1758 | $self->{open_elements}->[-1]->[0]->append_child | |
| 1759 | ($self->{document}->create_text_node ($s)); | |
| 1760 | last C; | |
| 1761 | } | } |
| 1762 | } else { | |
| 1763 | !!!cp ('t195.2'); | |
| 1764 | last C; | |
| 1765 | } | } |
| 1766 | } | } |
| 1767 | !!!next-token; | |
| 1768 | return; | ## Foster parenting. |
| 1769 | } elsif ({ | !!!parse-error (type => 'in table:#text', token => $token); |
| 1770 | address => 1, blockquote => 1, center => 1, dir => 1, | |
| 1771 | div => 1, dl => 1, fieldset => 1, listing => 1, | ## NOTE: As if in body, but insert into the foster parent element. |
| 1772 | menu => 1, ol => 1, p => 1, ul => 1, | $reconstruct_active_formatting_elements->($insert_to_foster); |
| 1773 | pre => 1, | |
| 1774 | }->{$token->{tag_name}}) { | if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1775 | ## has a p element in scope | # MUST |
| 1776 | INSCOPE: for (reverse @{$self->{open_elements}}) { | my $foster_parent_element; |
| 1777 | if ($_->[1] eq 'p') { | my $next_sibling; |
| 1778 | !!!back-token; | OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1779 | $token = {type => 'end tag', tag_name => 'p'}; | if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1780 | return; | !!!cp ('t197'); |
| 1781 | } elsif ({ | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1782 | table => 1, caption => 1, td => 1, th => 1, | $next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1783 | button => 1, marquee => 1, object => 1, html => 1, | undef $next_sibling |
| 1784 | }->{$_->[1]}) { | unless $next_sibling->parent_node eq $foster_parent_element; |
| 1785 | last INSCOPE; | last OE; |
| 1786 | } | } |
| 1787 | } # INSCOPE | } # OE |
| 1788 | $foster_parent_element ||= $self->{open_elements}->[0]->[0]; | |
| 1789 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | |
| 1790 | if ($token->{tag_name} eq 'pre') { | !!!cp ('t199'); |
| 1791 | !!!next-token; | $foster_parent_element->insert_before |
| 1792 | if ($token->{type} eq 'character') { | ($self->{document}->create_text_node ($s), $next_sibling); |
| 1793 | $token->{data} =~ s/^\x0A//; | |
| 1794 | unless (length $token->{data}) { | $open_tables->[-1]->[1] = 1; # tainted |
| 1795 | !!!next-token; | $open_tables->[-1]->[2] = 1; # ~node inserted |
| 1796 | } | } else { |
| 1797 | } | ## NOTE: Fragment case or in a foster parent'ed element |
| 1798 | } else { | ## (e.g. |<table><span>a|). In fragment case, whether the |
| 1799 | !!!next-token; | ## character is appended to existing node or a new node is |
| 1800 | ## created is irrelevant, since the foster parent'ed nodes | |
| 1801 | ## are discarded and fragment parsing does not invoke any | |
| 1802 | ## script. | |
| 1803 | !!!cp ('t200'); | |
| 1804 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($s); | |
| 1805 | } | |
| 1806 | } # C | |
| 1807 | } # TABLE_IMS | |
| 1808 | ||
| 1809 | if ($token->{type} == DOCTYPE_TOKEN) { | |
| 1810 | !!!cp ('t73'); | |
| 1811 | !!!parse-error (type => 'in html:#DOCTYPE', token => $token); | |
| 1812 | ## Ignore the token | |
| 1813 | ## Stay in the phase | |
| 1814 | !!!next-token; | |
| 1815 | next B; | |
| 1816 | } elsif ($token->{type} == START_TAG_TOKEN and | |
| 1817 | $token->{tag_name} eq 'html') { | |
| 1818 | if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { | |
| 1819 | !!!cp ('t79'); | |
| 1820 | !!!parse-error (type => 'after html', text => 'html', token => $token); | |
| 1821 | $self->{insertion_mode} = AFTER_BODY_IM; | |
| 1822 | } elsif ($self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) { | |
| 1823 | !!!cp ('t80'); | |
| 1824 | !!!parse-error (type => 'after html', text => 'html', token => $token); | |
| 1825 | $self->{insertion_mode} = AFTER_FRAMESET_IM; | |
| 1826 | } else { | |
| 1827 | !!!cp ('t81'); | |
| 1828 | } | |
| 1829 | ||
| 1830 | !!!cp ('t82'); | |
| 1831 | !!!parse-error (type => 'not first start tag', token => $token); | |
| 1832 | my $top_el = $self->{open_elements}->[0]->[0]; | |
| 1833 | for my $attr_name (keys %{$token->{attributes}}) { | |
| 1834 | unless ($top_el->has_attribute_ns (undef, $attr_name)) { | |
| 1835 | !!!cp ('t84'); | |
| 1836 | $top_el->set_attribute_ns | |
| 1837 | (undef, [undef, $attr_name], | |
| 1838 | $token->{attributes}->{$attr_name}->{value}); | |
| 1839 | } | } |
| 1840 | return; | } |
| 1841 | } elsif ($token->{tag_name} eq 'form') { | !!!nack ('t84.1'); |
| 1842 | if (defined $self->{form_element}) { | !!!next-token; |
| 1843 | !!!parse-error (type => 'in form:form'); | next B; |
| 1844 | ## Ignore the token | } elsif ($token->{type} == COMMENT_TOKEN) { |
| 1845 | !!!next-token; | my $comment = $self->{document}->create_comment ($token->{data}); |
| 1846 | return; | if ($self->{insertion_mode} & AFTER_HTML_IMS) { |
| 1847 | !!!cp ('t85'); | |
| 1848 | $self->{document}->append_child ($comment); | |
| 1849 | } elsif ($self->{insertion_mode} == AFTER_BODY_IM) { | |
| 1850 | !!!cp ('t86'); | |
| 1851 | $self->{open_elements}->[0]->[0]->append_child ($comment); | |
| 1852 | } else { | |
| 1853 | !!!cp ('t87'); | |
| 1854 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | |
| 1855 | $open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted | |
| 1856 | } | |
| 1857 | !!!next-token; | |
| 1858 | next B; | |
| 1859 | } elsif ($self->{insertion_mode} & IN_CDATA_RCDATA_IM) { | |
| 1860 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 1861 | $token->{data} =~ s/^\x0A// if $self->{ignore_newline}; | |
| 1862 | delete $self->{ignore_newline}; | |
| 1863 | ||
| 1864 | if (length $token->{data}) { | |
| 1865 | !!!cp ('t43'); | |
| 1866 | $self->{open_elements}->[-1]->[0]->manakai_append_text | |
| 1867 | ($token->{data}); | |
| 1868 | } else { | } else { |
| 1869 | ## 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} = $self->{open_elements}->[-1]->[0]; | ||
| !!!next-token; | ||
| return; | ||
| 1870 | } | } |
| } elsif ($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}, $token->{attributes}); | ||
| 1871 | !!!next-token; | !!!next-token; |
| 1872 | return; | next B; |
| 1873 | } elsif ($token->{tag_name} eq 'dd' or $token->{tag_name} eq 'dt') { | } elsif ($token->{type} == END_TAG_TOKEN) { |
| 1874 | ## has a p element in scope | delete $self->{ignore_newline}; |
| 1875 | INSCOPE: for (reverse @{$self->{open_elements}}) { | |
| 1876 | if ($_->[1] eq 'p') { | if ($token->{tag_name} eq 'script') { |
| 1877 | !!!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 | ||
| ## 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}}, $i; | ||
| } | ||
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| 1878 | ||
| 1879 | !!!next-token; | ## Para 1-2 |
| 1880 | return; | my $script = pop @{$self->{open_elements}}; |
| } 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 | ||
| 1881 | ||
| 1882 | $reconstruct_active_formatting_elements->($insert_to_current); | ## Para 3 |
| 1883 | $self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; | |
| 1884 | ||
| 1885 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ## Para 4 |
| 1886 | push @$active_formatting_elements, $self->{open_elements}->[-1]; | ## TODO: $old_insertion_point = $current_insertion_point; |
| 1887 | ## TODO: $current_insertion_point = just before $self->{nc}; | |
| 1888 | ||
| 1889 | ## Para 5 | |
| 1890 | ## TODO: Run the $script->[0]. | |
| 1891 | ||
| 1892 | ## Para 6 | |
| 1893 | ## TODO: $current_insertion_point = $old_insertion_point; | |
| 1894 | ||
| 1895 | ## Para 7 | |
| 1896 | ## TODO: if ($pending_external_script) { | |
| 1897 | ## TODO: ... | |
| 1898 | ## TODO: } | |
| 1899 | ||
| 1900 | !!!next-token; | !!!next-token; |
| 1901 | return; | next B; |
| 1902 | } elsif ({ | } else { |
| 1903 | b => 1, big => 1, em => 1, font => 1, i => 1, | !!!cp ('t42'); |
| 1904 | s => 1, small => 1, strile => 1, | |
| 1905 | 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', '']; | ||
| 1906 | ||
| 1907 | !!!next-token; | $self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1908 | return; | !!!next-token; |
| 1909 | } elsif ($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'); | ||
| $token->{tag_name} = 'img'; | ||
| 1910 | } | } |
| 1911 | } elsif ($token->{type} == END_OF_FILE_TOKEN) { | |
| 1912 | $reconstruct_active_formatting_elements->($insert_to_current); | delete $self->{ignore_newline}; |
| 1913 | ||
| 1914 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | !!!cp ('t44'); |
| 1915 | !!!parse-error (type => 'not closed', | |
| 1916 | text => $self->{open_elements}->[-1]->[0] | |
| 1917 | ->manakai_local_name, | |
| 1918 | token => $token); | |
| 1919 | ||
| 1920 | #if ($self->{open_elements}->[-1]->[1] == SCRIPT_EL) { | |
| 1921 | # ## TODO: Mark as "already executed" | |
| 1922 | #} | |
| 1923 | ||
| 1924 | pop @{$self->{open_elements}}; | pop @{$self->{open_elements}}; |
| 1925 | ||
| 1926 | $self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; | |
| 1927 | ## Reprocess. | |
| 1928 | next B; | |
| 1929 | } else { | |
| 1930 | die "$0: $token->{type}: In CDATA/RCDATA: Unknown token type"; | |
| 1931 | } | |
| 1932 | } elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM) { | |
| 1933 | if ($token->{type} == CHARACTER_TOKEN) { | |
| 1934 | !!!cp ('t87.1'); | |
| 1935 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | |
| 1936 | !!!next-token; | !!!next-token; |
| 1937 | return; | next B; |
| 1938 | } elsif ($token->{tag_name} eq 'hr') { | } elsif ($token->{type} == START_TAG_TOKEN) { |
| 1939 | ## has a p element in scope | if ((not {mglyph => 1, malignmark => 1}->{$token->{tag_name}} and |
| 1940 | INSCOPE: for (reverse @{$self->{open_elements}}) { | $self->{open_elements}->[-1]->[1] & FOREIGN_FLOW_CONTENT_EL) or |
| 1941 | if ($_->[1] eq 'p') { | not ($self->{open_elements}->[-1]->[1] & FOREIGN_EL) or |
| 1942 | !!!back-token; | ($token->{tag_name} eq 'svg' and |
| 1943 | $token = {type => 'end tag', tag_name => 'p'}; | $self->{open_elements}->[-1]->[1] == MML_AXML_EL)) { |
| 1944 | return; | ## NOTE: "using the rules for secondary insertion mode"then"continue" |
| 1945 | } elsif ({ | !!!cp ('t87.2'); |
| 1946 | table => 1, caption => 1, td => 1, th => 1, | # |
| 1947 | button => 1, marquee => 1, object => 1, html => 1, | } elsif ({ |
| 1948 | }->{$_->[1]}) { | b => 1, big => 1, blockquote => 1, body => 1, br => 1, |
| 1949 | last INSCOPE; | center => 1, code => 1, dd => 1, div => 1, dl => 1, dt => 1, |
| 1950 | em => 1, embed => 1, h1 => 1, h2 => 1, h3 => 1, | |
| 1951 | h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, i => 1, | |
| 1952 | img => 1, li => 1, listing => 1, menu => 1, meta => 1, | |
| 1953 | nobr => 1, ol => 1, p => 1, pre => 1, ruby => 1, s => 1, | |
| 1954 | small => 1, span => 1, strong => 1, strike => 1, sub => 1, | |
| 1955 | sup => 1, table => 1, tt => 1, u => 1, ul => 1, var => 1, | |
| 1956 | }->{$token->{tag_name}} or | |
| 1957 | ($token->{tag_name} eq 'font' and | |
| 1958 | ($token->{attributes}->{color} or | |
| 1959 | $token->{attributes}->{face} or | |
| 1960 | $token->{attributes}->{size}))) { | |
| 1961 | !!!cp ('t87.2'); | |
| 1962 | !!!parse-error (type => 'not closed', | |
| 1963 | text => $self->{open_elements}->[-1]->[0] | |
| 1964 | ->manakai_local_name, | |
| 1965 | token => $token); | |
| 1966 | ||
| 1967 | pop @{$self->{open_elements}} | |
| 1968 | while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; | |
| 1969 | ||
| 1970 | $self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; | |
| 1971 | ## Reprocess. | |
| 1972 | next B; | |
| 1973 | } else { | |
| 1974 | my $nsuri = $self->{open_elements}->[-1]->[0]->namespace_uri; | |
| 1975 | my $tag_name = $token->{tag_name}; | |
| 1976 | if ($nsuri eq $SVG_NS) { | |
| 1977 | $tag_name = { | |
| 1978 | altglyph => 'altGlyph', | |
| 1979 | altglyphdef => 'altGlyphDef', | |
| 1980 | altglyphitem => 'altGlyphItem', | |
| 1981 | animatecolor => 'animateColor', | |
| 1982 | animatemotion => 'animateMotion', | |
| 1983 | animatetransform => 'animateTransform', | |
| 1984 | clippath => 'clipPath', | |
| 1985 | feblend => 'feBlend', | |
| 1986 | fecolormatrix => 'feColorMatrix', | |
| 1987 | fecomponenttransfer => 'feComponentTransfer', | |
| 1988 | fecomposite => 'feComposite', | |
| 1989 | feconvolvematrix => 'feConvolveMatrix', | |
| 1990 | fediffuselighting => 'feDiffuseLighting', | |
| 1991 | fedisplacementmap => 'feDisplacementMap', | |
| 1992 | fedistantlight => 'feDistantLight', | |
| 1993 | feflood => 'feFlood', | |
| 1994 | fefunca => 'feFuncA', | |
| 1995 | fefuncb => 'feFuncB', | |
| 1996 | fefuncg => 'feFuncG', | |
| 1997 | fefuncr => 'feFuncR', | |
| 1998 | fegaussianblur => 'feGaussianBlur', | |
| 1999 | feimage => 'feImage', | |
| 2000 | femerge => 'feMerge', | |
| 2001 | femergenode => 'feMergeNode', | |
| 2002 | femorphology => 'feMorphology', | |
| 2003 | feoffset => 'feOffset', | |
| 2004 | fepointlight => 'fePointLight', | |
| 2005 | fespecularlighting => 'feSpecularLighting', | |
| 2006 | fespotlight => 'feSpotLight', | |
| 2007 | fetile => 'feTile', | |
| 2008 | feturbulence => 'feTurbulence', | |
| 2009 | foreignobject => 'foreignObject', | |
| 2010 | glyphref => 'glyphRef', | |
| 2011 | lineargradient => 'linearGradient', | |
| 2012 | radialgradient => 'radialGradient', | |
| 2013 | #solidcolor => 'solidColor', ## NOTE: Commented in spec (SVG1.2) | |
| 2014 | textpath => 'textPath', | |
| 2015 | }->{$tag_name} || $tag_name; | |
| 2016 | } | } |
| 2017 | } # INSCOPE | |
| 2018 | ## "adjust SVG attributes" (SVG only) - done in insert-element-f | |
| 2019 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | |
| 2020 | pop @{$self->{open_elements}}; | ## "adjust foreign attributes" - done in insert-element-f |
| 2021 | ||
| 2022 | !!!next-token; | !!!insert-element-f ($nsuri, $tag_name, $token->{attributes}, $token); |
| 2023 | return; | |
| 2024 | } elsif ($token->{tag_name} eq 'input') { | if ($self->{self_closing}) { |
| 2025 | $reconstruct_active_formatting_elements->($insert_to_current); | pop @{$self->{open_elements}}; |
| 2026 | !!!ack ('t87.3'); | |
| !!!insert-element-t ($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}}; | ||
| 2027 | } else { | } else { |
| 2028 | push @tokens, {type => 'character', | !!!cp ('t87.4'); |
| data => 'This is a searchable index. Insert your search keywords here: '}; # SHOULD | ||
| ## TODO: make this configurable | ||
| } | ||
| 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 $token->{data}) { | ||
| !!!next-token; | ||
| } | ||
| 2029 | } | } |
| 2030 | } else { | |
| !!!next-token; | ||
| } | ||
| while ($token->{type} eq 'character') { | ||
| $text .= $token->{data}; | ||
| 2031 | !!!next-token; | !!!next-token; |
| 2032 | next B; | |
| 2033 | } | } |
| 2034 | if (length $text) { | } elsif ($token->{type} == END_TAG_TOKEN) { |
| 2035 | $el->manakai_append_text ($text); | ## NOTE: "using the rules for secondary insertion mode" then "continue" |
| 2036 | } | if ($token->{tag_name} eq 'script') { |
| 2037 | !!!cp ('t87.41'); | |
| 2038 | $self->{content_model_flag} = 'PCDATA'; | # |
| 2039 | ## XXXscript: Execute script here. | |
| if ($token->{type} eq 'end tag' and | ||
| $token->{tag_name} eq $tag_name) { | ||
| ## Ignore the token | ||
| 2040 | } else { | } else { |
| 2041 | if ($token->{tag_name} eq 'textarea') { | !!!cp ('t87.5'); |
| 2042 | !!!parse-error (type => 'in RCDATA:#'.$token->{type}); | # |
| } else { | ||
| !!!parse-error (type => 'in CDATA:#'.$token->{type}); | ||
| } | ||
| ## ISSUE: And ignore? | ||
| 2043 | } | } |
| 2044 | !!!next-token; | } elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 2045 | return; | !!!cp ('t87.6'); |
| 2046 | } elsif ($token->{tag_name} eq 'select') { | !!!parse-error (type => 'not closed', |
| 2047 | $reconstruct_active_formatting_elements->($insert_to_current); | text => $self->{open_elements}->[-1]->[0] |
| 2048 | ->manakai_local_name, | |
| 2049 | !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | token => $token); |
| 2050 | ||
| 2051 | $self->{insertion_mode} = 'in select'; | pop @{$self->{open_elements}} |
| 2052 | !!!next-token; | while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; |
| 2053 | return; | |
| 2054 | } elsif ({ | ## NOTE: |<span><svg>| ... two parse errors, |<svg>| ... a parse error. |
| 2055 | caption => 1, col => 1, colgroup => 1, frame => 1, | |
| 2056 | frameset => 1, head => 1, option => 1, optgroup => 1, | $self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
| 2057 | tbody => 1, td => 1, tfoot => 1, th => 1, | ## Reprocess. |
| 2058 | thead => 1, tr => 1, | next B; |
| }->{$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. | ||
| 2059 | } else { | } else { |
| 2060 | $reconstruct_active_formatting_elements->($insert_to_current); | die "$0: $token->{type}: Unknown token type"; |
| !!!insert-element-t ($token->{tag_name}, $token->{attributes}); | ||
| !!!next-token; | ||
| return; | ||
| 2061 | } | } |
| 2062 | } elsif ($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]); | ||
| } | ||
| } | ||
| 2063 | ||
| 2064 | $self->{insertion_mode} = 'after body'; | if ($self->{insertion_mode} & HEAD_IMS) { |
| 2065 | !!!next-token; | if ($token->{type} == CHARACTER_TOKEN) { |
| 2066 | return; | if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 2067 | } else { | unless ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2068 | !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | if ($self->{head_element_inserted}) { |
| 2069 | ## Ignore the token | !!!cp ('t88.3'); |
| 2070 | !!!next-token; | $self->{open_elements}->[-1]->[0]->append_child |
| 2071 | return; | ($self->{document}->create_text_node ($1)); |
| 2072 | } | delete $self->{head_element_inserted}; |
| 2073 | } elsif ($token->{tag_name} eq 'html') { | ## NOTE: |</head> <link> | |
| 2074 | if (@{$self->{open_elements}} > 1 and $self->{open_elements}->[1]->[1] eq 'body') { | # |
| 2075 | ## ISSUE: There is an issue in the spec. | } else { |
| 2076 | if ($self->{open_elements}->[-1]->[1] ne 'body') { | !!!cp ('t88.2'); |
| 2077 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[1]->[1]); | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 2078 | } | ## NOTE: |</head>  | |
| 2079 | $self->{insertion_mode} = 'after body'; | # |
| ## reprocess | ||
| return; | ||
| } else { | ||
| !!!parse-error (type => 'unmatched end tag:'.$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; | ||
| 2080 | } | } |
| 2081 | $i = $_; | } else { |
| 2082 | last INSCOPE unless $token->{tag_name} eq 'p'; | !!!cp ('t88.1'); |
| 2083 | } elsif ({ | ## Ignore the token. |
| 2084 | table => 1, caption => 1, td => 1, th => 1, | # |
| button => 1, marquee => 1, object => 1, html => 1, | ||
| }->{$node->[1]}) { | ||
| last INSCOPE; | ||
| 2085 | } | } |
| 2086 | } # INSCOPE | unless (length $token->{data}) { |
| 2087 | !!!cp ('t88'); | |
| 2088 | if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { | !!!next-token; |
| 2089 | !!!parse-error (type => 'not closed:'.$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; | ||
| 2090 | } | } |
| 2091 | } # 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]); | ||
| 2092 | } | } |
| 2093 | ||
| 2094 | undef $self->{form_element}; | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2095 | !!!next-token; | !!!cp ('t89'); |
| 2096 | return; | ## As if <head> |
| 2097 | } elsif ({ | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2098 | h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2099 | }->{$token->{tag_name}}) { | push @{$self->{open_elements}}, |
| 2100 | ## 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]; | ||
| 2101 | ||
| 2102 | ## Step 2 | ## Reprocess in the "in head" insertion mode... |
| 2103 | S2: { | pop @{$self->{open_elements}}; |
| if ($node->[1] eq $token->{tag_name}) { | ||
| ## Step 1 | ||
| ## generate implied end tags | ||
| if ({ | ||
| dd => 1, dt => 1, li => 1, p => 1, | ||
| td => 1, th => 1, tr => 1, | ||
| }->{$self->{open_elements}->[-1]->[1]}) { | ||
| !!!back-token; | ||
| $token = {type => 'end tag', | ||
| tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | ||
| return; | ||
| } | ||
| ## Step 2 | ||
| if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| } | ||
| ## Step 3 | ||
| splice @{$self->{open_elements}}, $node_i; | ||
| 2104 | ||
| 2105 | !!!next-token; | ## Reprocess in the "after head" insertion mode... |
| 2106 | last S2; | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2107 | } else { | !!!cp ('t90'); |
| 2108 | ## Step 3 | ## As if </noscript> |
| 2109 | if (not $formatting_category->{$node->[1]} and | pop @{$self->{open_elements}}; |
| 2110 | #not $phrasing_category->{$node->[1]} and | !!!parse-error (type => 'in noscript:#text', token => $token); |
| ($special_category->{$node->[1]} or | ||
| $scoping_category->{$node->[1]})) { | ||
| !!!parse-error (type => 'not closed:'.$node->[1]); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| last S2; | ||
| } | ||
| } | ||
| ## Step 4 | ||
| $node_i--; | ||
| $node = $self->{open_elements}->[$node_i]; | ||
| 2111 | ||
| 2112 | ## Step 5; | ## Reprocess in the "in head" insertion mode... |
| 2113 | redo S2; | ## As if </head> |
| 2114 | } # S2 | pop @{$self->{open_elements}}; |
| return; | ||
| } | ||
| } | ||
| }; # $in_body | ||
| 2115 | ||
| 2116 | B: { | ## Reprocess in the "after head" insertion mode... |
| 2117 | if ($phase eq 'main') { | } elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2118 | if ($token->{type} eq 'DOCTYPE') { | !!!cp ('t91'); |
| 2119 | !!!parse-error (type => 'in html:#DOCTYPE'); | pop @{$self->{open_elements}}; |
| ## Ignore the token | ||
| ## Stay in the phase | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($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}); | ||
| } | ||
| } | ||
| !!!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]); | ||
| } | ||
| 2120 | ||
| 2121 | ## Stop parsing | ## Reprocess in the "after head" insertion mode... |
| 2122 | last B; | } else { |
| 2123 | !!!cp ('t92'); | |
| 2124 | } | |
| 2125 | ||
| 2126 | ## ISSUE: There is an issue in the spec. | ## "after head" insertion mode |
| 2127 | } else { | ## As if <body> |
| 2128 | if ($self->{insertion_mode} eq 'before head') { | !!!insert-element ('body',, $token); |
| 2129 | if ($token->{type} eq 'character') { | $self->{insertion_mode} = IN_BODY_IM; |
| 2130 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | ## reprocess |
| 2131 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | next B; |
| 2132 | unless (length $token->{data}) { | } elsif ($token->{type} == START_TAG_TOKEN) { |
| 2133 | !!!next-token; | if ($token->{tag_name} eq 'head') { |
| 2134 | redo B; | if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2135 | } | !!!cp ('t93'); |
| 2136 | } | !!!create-element ($self->{head_element}, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); |
| 2137 | ## As if <head> | $self->{open_elements}->[-1]->[0]->append_child |
| 2138 | !!!create-element ($self->{head_element}, 'head'); | ($self->{head_element}); |
| 2139 | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | push @{$self->{open_elements}}, |
| 2140 | push @{$self->{open_elements}}, [$self->{head_element}, 'head']; | [$self->{head_element}, $el_category->{head}]; |
| 2141 | $self->{insertion_mode} = 'in head'; | $self->{insertion_mode} = IN_HEAD_IM; |
| 2142 | ## reprocess | !!!nack ('t93.1'); |
| redo B; | ||
| } elsif ($token->{type} eq 'comment') { | ||
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| 2143 | !!!next-token; | !!!next-token; |
| 2144 | redo B; | next B; |
| 2145 | } elsif ($token->{type} eq 'start tag') { | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2146 | my $attr = $token->{tag_name} eq 'head' ? $token->{attributes} : {}; | !!!cp ('t93.2'); |
| 2147 | !!!create-element ($self->{head_element}, 'head', $attr); | !!!parse-error (type => 'after head', text => 'head', |
| 2148 | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); | token => $token); |
| 2149 | push @{$self->{open_elements}}, [$self->{head_element}, 'head']; | ## Ignore the token |
| 2150 | $self->{insertion_mode} = 'in head'; | !!!nack ('t93.3'); |
| 2151 | if ($token->{tag_name} eq 'head') { | !!!next-token; |
| 2152 | !!!next-token; | next B; |
| #} 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; | ||
| } | ||
| 2153 | } else { | } else { |
| 2154 | die "$0: $token->{type}: Unknown type"; | !!!cp ('t95'); |
| 2155 | } | !!!parse-error (type => 'in head:head', |
| 2156 | } elsif ($self->{insertion_mode} eq 'in head') { | token => $token); # or in head noscript |
| 2157 | if ($token->{type} eq 'character') { | ## Ignore the token |
| 2158 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | !!!nack ('t95.1'); |
| $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | ||
| unless (length $token->{data}) { | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| } | ||
| # | ||
| } elsif ($token->{type} eq 'comment') { | ||
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| 2159 | !!!next-token; | !!!next-token; |
| 2160 | redo B; | next B; |
| 2161 | } elsif ($token->{type} eq 'start tag') { | } |
| 2162 | if ($token->{tag_name} eq 'title') { | } elsif ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2163 | ## NOTE: There is an "as if in head" code clone | !!!cp ('t96'); |
| 2164 | my $title_el; | ## As if <head> |
| 2165 | !!!create-element ($title_el, 'title', $token->{attributes}); | !!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2166 | (defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | $self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2167 | ->append_child ($title_el); | push @{$self->{open_elements}}, |
| 2168 | $self->{content_model_flag} = 'RCDATA'; | [$self->{head_element}, $el_category->{head}]; |
| 2169 | delete $self->{escape}; # MUST | |
| 2170 | $self->{insertion_mode} = IN_HEAD_IM; | |
| 2171 | my $text = ''; | ## Reprocess in the "in head" insertion mode... |
| 2172 | !!!next-token; | } else { |
| 2173 | while ($token->{type} eq 'character') { | !!!cp ('t97'); |
| 2174 | $text .= $token->{data}; | } |
| 2175 | !!!next-token; | |
| 2176 | } | if ($token->{tag_name} eq 'base') { |
| 2177 | if (length $text) { | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2178 | $title_el->manakai_append_text ($text); | !!!cp ('t98'); |
| 2179 | } | ## As if </noscript> |
| 2180 | pop @{$self->{open_elements}}; | |
| 2181 | $self->{content_model_flag} = 'PCDATA'; | !!!parse-error (type => 'in noscript', text => 'base', |
| 2182 | token => $token); | |
| 2183 | if ($token->{type} eq 'end tag' and | |
| 2184 | $token->{tag_name} eq 'title') { | $self->{insertion_mode} = IN_HEAD_IM; |
| 2185 | ## Ignore the token | ## Reprocess in the "in head" insertion mode... |
| } else { | ||
| !!!parse-error (type => 'in RCDATA:#'.$token->{type}); | ||
| ## ISSUE: And ignore? | ||
| } | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{tag_name} eq 'style') { | ||
| $style_start_tag->(); | ||
| redo B; | ||
| } elsif ($token->{tag_name} eq 'script') { | ||
| $script_start_tag->(); | ||
| redo B; | ||
| } elsif ({base => 1, link => 1, meta => 1}->{$token->{tag_name}}) { | ||
| ## NOTE: There are "as if in head" code clones | ||
| my $el; | ||
| !!!create-element ($el, $token->{tag_name}, $token->{attributes}); | ||
| (defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) | ||
| ->append_child ($el); | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{tag_name} eq 'head') { | ||
| !!!parse-error (type => 'in head:head'); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| redo B; | ||
| } else { | ||
| # | ||
| } | ||
| } elsif ($token->{type} eq 'end tag') { | ||
| if ($token->{tag_name} eq 'head') { | ||
| if ($self->{open_elements}->[-1]->[1] eq 'head') { | ||
| pop @{$self->{open_elements}}; | ||
| } else { | ||
| !!!parse-error (type => 'unmatched end tag:head'); | ||
| } | ||
| $self->{insertion_mode} = 'after head'; | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{tag_name} eq 'body' or | ||
| $token->{tag_name} eq 'html') { | ||
| # | ||
| } else { | ||
| !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); | ||
| ## Ignore the token | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| 2186 | } else { | } else { |
| 2187 | # | !!!cp ('t99'); |
| 2188 | } | } |
| 2189 | ||
| 2190 | if ($self->{open_elements}->[-1]->[1] eq 'head') { | ## NOTE: There is a "as if in head" code clone. |
| 2191 | ## As if </head> | if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2192 | pop @{$self->{open_elements}}; | !!!cp ('t100'); |
| 2193 | !!!parse-error (type => 'after head', | |
| 2194 | text => $token->{tag_name}, token => $token); | |
| 2195 | push @{$self->{open_elements}}, | |
| 2196 | [$self->{head_element}, $el_category->{head}]; | |
| 2197 | $self->{head_element_inserted} = 1; | |
| 2198 | } else { | |
| 2199 | !!!cp ('t101'); | |
| 2200 | } | } |
| 2201 | $self->{insertion_mode} = 'after head'; | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2202 | ## reprocess | pop @{$self->{open_elements}}; |
| 2203 | redo B; | pop @{$self->{open_elements}} # <head> |
| 2204 | if $self->{insertion_mode} == AFTER_HEAD_IM; | |
| 2205 | !!!nack ('t101.1'); | |
| 2206 | !!!next-token; | |
| 2207 | next B; | |
| 2208 | } elsif ($token->{tag_name} eq 'link') { | |
| 2209 | ## NOTE: There is a "as if in head" code clone. | |
| 2210 | if ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2211 | !!!cp ('t102'); | |
| 2212 | !!!parse-error (type => 'after head', | |
| 2213 | text => $token->{tag_name}, token => $token); | |
| 2214 | push @{$self->{open_elements}}, | |
| 2215 | [$self->{head_element}, $el_category->{head}]; | |
| 2216 | $self->{head_element_inserted} = 1; | |
| 2217 | } else { | |
| 2218 | !!!cp ('t103'); | |
| 2219 | } | |
| 2220 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2221 | pop @{$self->{open_elements}}; | |
| 2222 | pop @{$self->{open_elements}} # <head> | |
| 2223 | if $self->{insertion_mode} == AFTER_HEAD_IM; | |
| 2224 | !!!ack ('t103.1'); | |
| 2225 | !!!next-token; | |
| 2226 | next B; | |
| 2227 | } elsif ($token->{tag_name} eq 'command') { | |
| 2228 | if ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2229 | ## NOTE: If the insertion mode at the time of the emission | |
| 2230 | ## of the token was "before head", $self->{insertion_mode} | |
| 2231 | ## is already changed to |IN_HEAD_IM|. | |
| 2232 | ||
| 2233 | ## ISSUE: An issue in the spec. | ## NOTE: There is a "as if in head" code clone. |
| 2234 | } elsif ($self->{insertion_mode} eq 'after head') { | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2235 | if ($token->{type} eq 'character') { | pop @{$self->{open_elements}}; |
| 2236 | if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | pop @{$self->{open_elements}} # <head> |
| 2237 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2238 | unless (length $token->{data}) { | !!!ack ('t103.2'); |
| !!!next-token; | ||
| redo B; | ||
| } | ||
| } | ||
| # | ||
| } elsif ($token->{type} eq 'comment') { | ||
| my $comment = $self->{document}->create_comment ($token->{data}); | ||
| $self->{open_elements}->[-1]->[0]->append_child ($comment); | ||
| 2239 | !!!next-token; | !!!next-token; |
| 2240 | redo B; | next B; |
| } elsif ($token->{type} eq 'start tag') { | ||
| if ($token->{tag_name} eq 'body') { | ||
| !!!insert-element ('body', $token->{attributes}); | ||
| $self->{insertion_mode} = 'in body'; | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{tag_name} eq 'frameset') { | ||
| !!!insert-element ('frameset', $token->{attributes}); | ||
| $self->{insertion_mode} = 'in frameset'; | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ({ | ||
| base => 1, link => 1, meta => 1, | ||
| script => 1, style => 1, title => 1, | ||
| }->{$token->{tag_name}}) { | ||
| !!!parse-error (type => 'after head:'.$token->{tag_name}); | ||
| $self->{insertion_mode} = 'in head'; | ||
| ## reprocess | ||
| redo B; | ||
| } else { | ||
| # | ||
| } | ||
| 2241 | } else { | } else { |
| 2242 | ## NOTE: "in head noscript" or "after head" insertion mode | |
| 2243 | ## - in these cases, these tags are treated as same as | |
| 2244 | ## normal in-body tags. | |
| 2245 | !!!cp ('t103.3'); | |
| 2246 | # | # |
| 2247 | } | } |
| 2248 | } elsif ($token->{tag_name} eq 'meta') { | |
| 2249 | ## As if <body> | ## NOTE: There is a "as if in head" code clone. |
| 2250 | !!!insert-element ('body'); | if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2251 | $self->{insertion_mode} = 'in body'; | !!!cp ('t104'); |
| 2252 | ## reprocess | !!!parse-error (type => 'after head', |
| 2253 | redo B; | text => $token->{tag_name}, token => $token); |
| 2254 | } elsif ($self->{insertion_mode} eq 'in body') { | push @{$self->{open_elements}}, |
| 2255 | if ($token->{type} eq 'character') { | [$self->{head_element}, $el_category->{head}]; |
| 2256 | ## NOTE: There is a code clone of "character in body". | $self->{head_element_inserted} = 1; |
| $reconstruct_active_formatting_elements->($insert_to_current); | ||
| $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | ||
| !!!next-token; | ||
| redo B; | ||
| } elsif ($token->{type} eq 'comment') { | ||
| ## NOTE: There 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; | ||
| 2257 | } else { | } else { |
| 2258 | $in_body->($insert_to_current); | !!!cp ('t105'); |
| redo B; | ||
| 2259 | } | } |
| 2260 | } elsif ($self->{insertion_mode} eq 'in table') { | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2261 | if ($token->{type} eq 'character') { | my $meta_el = pop @{$self->{open_elements}}; |
| ## NOTE: There are "character in table" code clones. | ||
| if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { | ||
| $self->{open_elements}->[-1]->[0]->manakai_append_text ($1); | ||
| unless (length $token->{data}) { | ||
| !!!next-token; | ||
| redo B; | ||
| } | ||
| } | ||
| !!!parse-error (type => 'in table:#character'); | ||
| 2262 | ||
| 2263 | ## As if in body, but insert into foster parent element | unless ($self->{confident}) { |
| 2264 | ## ISSUE: Spec says that "whenever a node would be inserted | if ($token->{attributes}->{charset}) { |
| 2265 | ## into the current node" while characters might not be | !!!cp ('t106'); |
| 2266 | ## result in a new Text node. | ## NOTE: Whether the encoding is supported or not is handled |
| 2267 | $reconstruct_active_formatting_elements->($insert_to_foster); | ## in the {change_encoding} callback. |
| 2268 | $self->{change_encoding} | |
| 2269 | if ({ | ->($self, $token->{attributes}->{charset}->{value}, |
| 2270 | table => 1, tbody => 1, tfoot => 1, | $token); |
| 2271 | thead => 1, tr => 1, | |
| 2272 | }->{$self->{open_elements}->[-1]->[1]}) { | $meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2273 | # MUST | ->set_user_data (manakai_has_reference => |
| 2274 | my $foster_parent_element; | $token->{attributes}->{charset} |
| 2275 | my $next_sibling; | ->{has_reference}); |
| 2276 | my $prev_sibling; | } elsif ($token->{attributes}->{content}) { |
| 2277 | OE: for (reverse 0..$#{$self->{open_elements}}) { | if ($token->{attributes}->{content}->{value} |
| 2278 | if ($self->{open_elements}->[$_]->[1] eq 'table') { | =~ /[Cc][Hh][Aa][Rr][Ss][Ee][Tt] |
| 2279 | my $parent = $self->{open_elements}->[$_]->[0]->parent_node; | [\x09\x0A\x0C\x0D\x20]*= |
| 2280 | if (defined $parent and $parent->node_type == 1) { | [\x09\x0A\x0C\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 2281 | $foster_parent_element = $parent; | ([^"'\x09\x0A\x0C\x0D\x20] |
| 2282 | $next_sibling = $self->{open_elements}->[$_]->[0]; | [^\x09\x0A\x0C\x0D\x20\x3B]*))/x) { |
| 2283 | $prev_sibling = $next_sibling->previous_sibling; | !!!cp ('t107'); |
| 2284 | ## NOTE: Whether the encoding is supported or not is handled | |
| 2285 | ## in the {change_encoding} callback. | |
| 2286 | $self->{change_encoding} | |
| 2287 | ->($self, defined $1 ? $1 : defined $2 ? $2 : $3, | |
| 2288 | $token); | |
| 2289 | $meta_el->[0]->get_attribute_node_ns (undef, 'content') | |
| 2290 | ->set_user_data (manakai_has_reference => | |
| 2291 | $token->{attributes}->{content} | |
| 2292 | ->{has_reference}); | |
| 2293 | } else { | } else { |
| 2294 | $foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; | !!!cp ('t108'); |
| $prev_sibling = $foster_parent_element->last_child; | ||
| 2295 | } | } |
| last OE; | ||
| 2296 | } | } |
| } # OE | ||
| $foster_parent_element = $self->{open_elements}->[0]->[0] and | ||
| $prev_sibling = $foster_parent_element->last_child | ||
| unless defined $foster_parent_element; | ||
| if (defined $prev_sibling and | ||
| $prev_sibling->node_type == 3) { | ||
| $prev_sibling->manakai_append_text ($token->{data}); | ||
| 2297 | } else { | } else { |
| 2298 | $foster_parent_element->insert_before | if ($token->{attributes}->{charset}) { |
| 2299 | ($self->{document}->create_text_node ($token->{data}), | !!!cp ('t109'); |
| 2300 | $next_sibling); | $meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2301 | } | ->set_user_data (manakai_has_reference => |
| 2302 | } else { | $token->{attributes}->{charset} |
| 2303 | $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); | ->{has_reference}); |
| 2304 | } | } |
| 2305 | if ($token->{attributes}->{content}) { | |
| 2306 | !!!next-token; | !!!cp ('t110'); |
| 2307 | redo B; | $meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2308 | } elsif ($token->{type} eq 'comment') { | ->set_user_data (manakai_has_reference => |
| 2309 | my $comment = $self->{document}->create_comment ($token->{data}); | $token->{attributes}->{content} |
| 2310 | $self->{open_elements}->[-1]->[0]->append_child ($comment); | ->{has_reference}); |
| 2311 | !!!next-token; | } |
| redo B; | ||
| } elsif ($token->{type} eq 'start tag') { | ||
| if ({ | ||
| caption => 1, | ||
| colgroup => 1, | ||
| tbody => 1, tfoot => 1, thead => 1, | ||
| }->{$token->{tag_name}}) { | ||
| ## Clear back to table context | ||
| while ($self->{open_elements}->[-1]->[1] ne 'table' and | ||
| $self->{open_elements}->[-1]->[1] ne 'html') { | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| pop @{$self->{open_elements}}; | ||
| 2312 | } | } |
| 2313 | ||
| 2314 | push @$active_formatting_elements, ['#marker', ''] | pop @{$self->{open_elements}} # <head> |
| 2315 | if $token->{tag_name} eq 'caption'; | if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2316 | !!!ack ('t110.1'); | |
| !!!insert-element ($token->{tag_name}, $token->{attributes}); | ||
| $self->{insertion_mode} = { | ||
| caption => 'in caption', | ||
| colgroup => 'in column group', | ||
| tbody => 'in table body', | ||
| tfoot => 'in table body', | ||
| thead => 'in table body', | ||
| }->{$token->{tag_name}}; | ||
| 2317 | !!!next-token; | !!!next-token; |
| 2318 | redo B; | next B; |
| 2319 | } elsif ({ | } elsif ($token->{tag_name} eq 'title') { |
| 2320 | col => 1, | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2321 | td => 1, th => 1, tr => 1, | !!!cp ('t111'); |
| 2322 | }->{$token->{tag_name}}) { | ## As if </noscript> |
| 2323 | ## Clear back to table context | pop @{$self->{open_elements}}; |
| 2324 | while ($self->{open_elements}->[-1]->[1] ne 'table' and | !!!parse-error (type => 'in noscript', text => 'title', |
| 2325 | $self->{open_elements}->[-1]->[1] ne 'html') { | token => $token); |
| 2326 | !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | |
| 2327 | pop @{$self->{open_elements}}; | $self->{insertion_mode} = IN_HEAD_IM; |
| 2328 | } | ## Reprocess in the "in head" insertion mode... |
| 2329 | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { | |
| 2330 | !!!cp ('t112'); | |
| 2331 | !!!parse-error (type => 'after head', | |
| 2332 | text => $token->{tag_name}, token => $token); | |
| 2333 | push @{$self->{open_elements}}, | |
| 2334 | [$self->{head_element}, $el_category->{head}]; | |
| 2335 | $self->{head_element_inserted} = 1; | |
| 2336 | } else { | |
| 2337 | !!!cp ('t113'); | |
| 2338 | } | |
| 2339 | ||
| 2340 | !!!insert-element ($token->{tag_name} eq 'col' ? 'colgroup' : 'tbody'); | ## NOTE: There is a "as if in head" code clone. |
| 2341 | $self->{insertion_mode} = $token->{tag_name} eq 'col' | $parse_rcdata->(RCDATA_CONTENT_MODEL); |
| ? 'in column group' : 'in table body'; | ||
| ## reprocess | ||
| redo B; | ||
| } elsif ($token->{tag_name} eq 'table') { | ||
| ## NOTE: There are code clones for this "table in table" | ||
| !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); | ||
| 2342 | ||
| 2343 | ## As if </table> | ## NOTE: At this point the stack of open elements contain |
| 2344 | ## have a table element in table scope | ## the |head| element (index == -2) and the |script| element |
| 2345 | my $i; | ## (index == -1). In the "after head" insertion mode the |
| 2346 | INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { | ## |head| element is inserted only for the purpose of |
| 2347 | my $node = $self->{open_elements}->[$_]; | ## providing the context for the |script| element, and |
| 2348 | if ($node->[1] eq 'table') { | ## therefore we can now and have to remove the element from |
| 2349 | $i = $_; | ## the stack. |
| 2350 | last INSCOPE; | splice @{$self->{open_elements}}, -2, 1, () # <head> |
| 2351 | } elsif ({ | if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2352 | table => 1, html => 1, | next B; |
| 2353 | }->{$node->[1]}) { | } elsif ($token->{tag_name} eq 'style' or |
| 2354 | last INSCOPE; | $token->{tag_name} eq 'noframes') { |
| 2355 | } | ## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
| 2356 | } # INSCOPE | ## insertion mode IN_HEAD_IM) |
| 2357 | unless (defined $i) { | ## NOTE: There is a "as if in head" code clone. |
| 2358 | !!!parse-error (type => 'unmatched end tag:table'); | if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2359 | ## Ignore tokens </table><table> | !!!cp ('t114'); |
| 2360 | !!!parse-error (type => 'after head', | |
| 2361 | text => $token->{tag_name}, token => $token); | |
| 2362 | push @{$self->{open_elements}}, | |
| 2363 | [$self->{head_element}, $el_category->{head}]; | |
| 2364 | $self->{head_element_inserted} = 1; | |
| 2365 | } else { | |
| 2366 | !!!cp ('t115'); | |
| 2367 | } | |
| 2368 | $parse_rcdata->(CDATA_CONTENT_MODEL); | |
| 2369 | ## ISSUE: A spec bug [Bug 6038] | |
| 2370 | splice @{$self->{open_elements}}, -2, 1, () # <head> | |
| 2371 | if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; | |
| 2372 | next B; | |
| 2373 | } elsif ($token->{tag_name} eq 'noscript') { | |
| 2374 | if ($self->{insertion_mode} == IN_HEAD_IM) { | |
| 2375 | !!!cp ('t116'); | |
| 2376 | ## NOTE: and scripting is disalbed | |
| 2377 | !!!insert-element ($token->{tag_name}, $token->{attributes}, $token); | |
| 2378 | $self->{insertion_mode} = IN_HEAD_NOSCRIPT_IM; | |
| 2379 | !!!nack ('t116.1'); | |
| 2380 | !!!next-token; | !!!next-token; |
| 2381 | redo B; | next B; |
| 2382 | } elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { | |
| 2383 | !!!cp ('t117'); | |
| 2384 | !!!parse-error (type => 'in noscript', text => 'noscript', | |
| 2385 | token => $token); | |
| 2386 | ## Ignore the token | |
| 2387 | !!!nack ('t117.1'); | |
| 2388 | !!!next-token; | |
| 2389 | next B; | |
| 2390 | } else { | |
| 2391 | !!!cp ('t118'); | |
| 2392 | # | |
| 2393 | } | } |
| 2394 | } elsif ($token->{tag_name} eq 'script') { | |
| 2395 | ## generate implied end tags | if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2396 | if ({ | !!!cp ('t119'); |
| 2397 | dd => 1, dt => 1, li => 1, p => 1, | ## As if </noscript> |
| 2398 | td => 1, th => 1, tr => 1, | pop @{$self->{open_elements}}; |
| 2399 | }->{$self->{open_elements}->[-1]->[1]}) { | !!!parse-error (type => 'in noscript', text => 'script', |
| 2400 | !!!back-token; # <table> | token => $token); |
| 2401 | $token = {type => 'end tag', tag_name => 'table'}; | |
| 2402 | !!!back-token; | $self->{insertion_mode} = IN_HEAD_IM; |
| 2403 | $token = {type => 'end tag', | ## Reprocess in the "in head" insertion mode... |
| 2404 | tag_name => $self->{open_elements}->[-1]->[1]}; # MUST | } elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2405 | redo B; | !!!cp ('t120'); |
| 2406 | !!!parse-error (type => 'after head', | |
| 2407 | text => $token->{tag_name}, token => $token); | |
| 2408 | push @{$self->{open_elements}}, | |
| 2409 | [$self->{head_element}, $el_category->{head}]; | |
| 2410 | $self->{head_element_inserted} = 1; | |
| 2411 | } else { | |
| 2412 | !!!cp ('t121'); | |
| 2413 | } | |
| 2414 | ||
| 2415 | ## NOTE: There is a "as if in head" code clone. | |
| 2416 | $script_start_tag->(); | |
| 2417 | ## ISSUE: A spec bug [Bug 6038] | |
| 2418 | splice @{$self->{open_elements}}, -2, 1 # <head> | |
| 2419 |   |