| 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 |
## ISSUE: HTML5 revision 967 says that the encoding layer MUST NOT |
require IO::Handle; |
|
## strip BOM and the HTML layer MUST ignore it. Whether we can do it |
|
|
## is not yet clear. |
|
|
## "{U+FEFF}..." in UTF-16BE/UTF-16LE is three or four characters? |
|
|
## "{U+FEFF}..." in GB18030? |
|
|
|
|
|
my $permitted_slash_tag_name = { |
|
|
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 |
|
eventsource => MISC_SPECIAL_EL, |
| 183 |
|
fieldset => MISC_SPECIAL_EL, |
| 184 |
|
figure => MISC_SPECIAL_EL, |
| 185 |
|
font => FORMATTING_EL, |
| 186 |
|
footer => MISC_SPECIAL_EL, |
| 187 |
|
form => FORM_EL, |
| 188 |
|
frame => MISC_SPECIAL_EL, |
| 189 |
|
frameset => FRAMESET_EL, |
| 190 |
|
h1 => HEADING_EL, |
| 191 |
|
h2 => HEADING_EL, |
| 192 |
|
h3 => HEADING_EL, |
| 193 |
|
h4 => HEADING_EL, |
| 194 |
|
h5 => HEADING_EL, |
| 195 |
|
h6 => HEADING_EL, |
| 196 |
|
head => MISC_SPECIAL_EL, |
| 197 |
|
header => 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 |
|
li => LI_EL, |
| 207 |
|
link => MISC_SPECIAL_EL, |
| 208 |
|
listing => MISC_SPECIAL_EL, |
| 209 |
|
marquee => MISC_SCOPING_EL, |
| 210 |
|
menu => MISC_SPECIAL_EL, |
| 211 |
|
meta => MISC_SPECIAL_EL, |
| 212 |
|
nav => MISC_SPECIAL_EL, |
| 213 |
|
nobr => NOBR_EL, |
| 214 |
|
noembed => MISC_SPECIAL_EL, |
| 215 |
|
noframes => MISC_SPECIAL_EL, |
| 216 |
|
noscript => MISC_SPECIAL_EL, |
| 217 |
|
object => MISC_SCOPING_EL, |
| 218 |
|
ol => MISC_SPECIAL_EL, |
| 219 |
|
optgroup => OPTGROUP_EL, |
| 220 |
|
option => OPTION_EL, |
| 221 |
|
p => P_EL, |
| 222 |
|
param => MISC_SPECIAL_EL, |
| 223 |
|
plaintext => MISC_SPECIAL_EL, |
| 224 |
|
pre => MISC_SPECIAL_EL, |
| 225 |
|
rp => RUBY_COMPONENT_EL, |
| 226 |
|
rt => RUBY_COMPONENT_EL, |
| 227 |
|
ruby => RUBY_EL, |
| 228 |
|
s => FORMATTING_EL, |
| 229 |
|
script => MISC_SPECIAL_EL, |
| 230 |
|
select => SELECT_EL, |
| 231 |
|
section => MISC_SPECIAL_EL, |
| 232 |
|
small => FORMATTING_EL, |
| 233 |
|
spacer => MISC_SPECIAL_EL, |
| 234 |
|
strike => FORMATTING_EL, |
| 235 |
|
strong => FORMATTING_EL, |
| 236 |
|
style => MISC_SPECIAL_EL, |
| 237 |
|
table => TABLE_EL, |
| 238 |
|
tbody => TABLE_ROW_GROUP_EL, |
| 239 |
|
td => TABLE_CELL_EL, |
| 240 |
|
textarea => MISC_SPECIAL_EL, |
| 241 |
|
tfoot => TABLE_ROW_GROUP_EL, |
| 242 |
|
th => TABLE_CELL_EL, |
| 243 |
|
thead => TABLE_ROW_GROUP_EL, |
| 244 |
|
title => MISC_SPECIAL_EL, |
| 245 |
|
tr => TABLE_ROW_EL, |
| 246 |
|
tt => FORMATTING_EL, |
| 247 |
|
u => FORMATTING_EL, |
| 248 |
|
ul => MISC_SPECIAL_EL, |
| 249 |
|
wbr => MISC_SPECIAL_EL, |
| 250 |
}; |
}; |
| 251 |
my $formatting_category = { |
|
| 252 |
a => 1, b => 1, big => 1, em => 1, font => 1, i => 1, nobr => 1, |
my $el_category_f = { |
| 253 |
s => 1, small => 1, strile => 1, strong => 1, tt => 1, u => 1, |
$MML_NS => { |
| 254 |
|
'annotation-xml' => MML_AXML_EL, |
| 255 |
|
mi => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 256 |
|
mo => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 257 |
|
mn => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 258 |
|
ms => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 259 |
|
mtext => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 260 |
|
}, |
| 261 |
|
$SVG_NS => { |
| 262 |
|
foreignObject => SCOPING_EL | FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 263 |
|
desc => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 264 |
|
title => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 265 |
|
}, |
| 266 |
|
## NOTE: In addition, FOREIGN_EL is set to non-HTML elements. |
| 267 |
}; |
}; |
|
# $phrasing_category: all other elements |
|
| 268 |
|
|
| 269 |
sub parse_string ($$$;$) { |
my $svg_attr_name = { |
| 270 |
my $self = shift->new; |
attributename => 'attributeName', |
| 271 |
my $s = \$_[0]; |
attributetype => 'attributeType', |
| 272 |
$self->{document} = $_[1]; |
basefrequency => 'baseFrequency', |
| 273 |
|
baseprofile => 'baseProfile', |
| 274 |
|
calcmode => 'calcMode', |
| 275 |
|
clippathunits => 'clipPathUnits', |
| 276 |
|
contentscripttype => 'contentScriptType', |
| 277 |
|
contentstyletype => 'contentStyleType', |
| 278 |
|
diffuseconstant => 'diffuseConstant', |
| 279 |
|
edgemode => 'edgeMode', |
| 280 |
|
externalresourcesrequired => 'externalResourcesRequired', |
| 281 |
|
filterres => 'filterRes', |
| 282 |
|
filterunits => 'filterUnits', |
| 283 |
|
glyphref => 'glyphRef', |
| 284 |
|
gradienttransform => 'gradientTransform', |
| 285 |
|
gradientunits => 'gradientUnits', |
| 286 |
|
kernelmatrix => 'kernelMatrix', |
| 287 |
|
kernelunitlength => 'kernelUnitLength', |
| 288 |
|
keypoints => 'keyPoints', |
| 289 |
|
keysplines => 'keySplines', |
| 290 |
|
keytimes => 'keyTimes', |
| 291 |
|
lengthadjust => 'lengthAdjust', |
| 292 |
|
limitingconeangle => 'limitingConeAngle', |
| 293 |
|
markerheight => 'markerHeight', |
| 294 |
|
markerunits => 'markerUnits', |
| 295 |
|
markerwidth => 'markerWidth', |
| 296 |
|
maskcontentunits => 'maskContentUnits', |
| 297 |
|
maskunits => 'maskUnits', |
| 298 |
|
numoctaves => 'numOctaves', |
| 299 |
|
pathlength => 'pathLength', |
| 300 |
|
patterncontentunits => 'patternContentUnits', |
| 301 |
|
patterntransform => 'patternTransform', |
| 302 |
|
patternunits => 'patternUnits', |
| 303 |
|
pointsatx => 'pointsAtX', |
| 304 |
|
pointsaty => 'pointsAtY', |
| 305 |
|
pointsatz => 'pointsAtZ', |
| 306 |
|
preservealpha => 'preserveAlpha', |
| 307 |
|
preserveaspectratio => 'preserveAspectRatio', |
| 308 |
|
primitiveunits => 'primitiveUnits', |
| 309 |
|
refx => 'refX', |
| 310 |
|
refy => 'refY', |
| 311 |
|
repeatcount => 'repeatCount', |
| 312 |
|
repeatdur => 'repeatDur', |
| 313 |
|
requiredextensions => 'requiredExtensions', |
| 314 |
|
requiredfeatures => 'requiredFeatures', |
| 315 |
|
specularconstant => 'specularConstant', |
| 316 |
|
specularexponent => 'specularExponent', |
| 317 |
|
spreadmethod => 'spreadMethod', |
| 318 |
|
startoffset => 'startOffset', |
| 319 |
|
stddeviation => 'stdDeviation', |
| 320 |
|
stitchtiles => 'stitchTiles', |
| 321 |
|
surfacescale => 'surfaceScale', |
| 322 |
|
systemlanguage => 'systemLanguage', |
| 323 |
|
tablevalues => 'tableValues', |
| 324 |
|
targetx => 'targetX', |
| 325 |
|
targety => 'targetY', |
| 326 |
|
textlength => 'textLength', |
| 327 |
|
viewbox => 'viewBox', |
| 328 |
|
viewtarget => 'viewTarget', |
| 329 |
|
xchannelselector => 'xChannelSelector', |
| 330 |
|
ychannelselector => 'yChannelSelector', |
| 331 |
|
zoomandpan => 'zoomAndPan', |
| 332 |
|
}; |
| 333 |
|
|
| 334 |
## NOTE: |set_inner_html| copies most of this method's code |
my $foreign_attr_xname = { |
| 335 |
|
'xlink:actuate' => [$XLINK_NS, ['xlink', 'actuate']], |
| 336 |
|
'xlink:arcrole' => [$XLINK_NS, ['xlink', 'arcrole']], |
| 337 |
|
'xlink:href' => [$XLINK_NS, ['xlink', 'href']], |
| 338 |
|
'xlink:role' => [$XLINK_NS, ['xlink', 'role']], |
| 339 |
|
'xlink:show' => [$XLINK_NS, ['xlink', 'show']], |
| 340 |
|
'xlink:title' => [$XLINK_NS, ['xlink', 'title']], |
| 341 |
|
'xlink:type' => [$XLINK_NS, ['xlink', 'type']], |
| 342 |
|
'xml:base' => [$XML_NS, ['xml', 'base']], |
| 343 |
|
'xml:lang' => [$XML_NS, ['xml', 'lang']], |
| 344 |
|
'xml:space' => [$XML_NS, ['xml', 'space']], |
| 345 |
|
'xmlns' => [$XMLNS_NS, [undef, 'xmlns']], |
| 346 |
|
'xmlns:xlink' => [$XMLNS_NS, ['xmlns', 'xlink']], |
| 347 |
|
}; |
| 348 |
|
|
| 349 |
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; |
|
| 350 |
|
|
| 351 |
pop @{$self->{prev_input_character}}; |
## TODO: Invoke the reset algorithm when a resettable element is |
| 352 |
unshift @{$self->{prev_input_character}}, $self->{next_input_character}; |
## created (cf. HTML5 revision 2259). |
| 353 |
|
|
| 354 |
$self->{next_input_character} = -1 and return if $i >= length $$s; |
sub parse_byte_string ($$$$;$) { |
| 355 |
$self->{next_input_character} = ord substr $$s, $i++, 1; |
my $self = shift; |
| 356 |
$column++; |
my $charset_name = shift; |
| 357 |
|
open my $input, '<', ref $_[0] ? $_[0] : \($_[0]); |
| 358 |
if ($self->{next_input_character} == 0x000A) { # LF |
return $self->parse_byte_stream ($charset_name, $input, @_[1..$#_]); |
| 359 |
$line++; |
} # parse_byte_string |
| 360 |
$column = 0; |
|
| 361 |
} elsif ($self->{next_input_character} == 0x000D) { # CR |
sub parse_byte_stream ($$$$;$$) { |
| 362 |
$i++ if substr ($$s, $i, 1) eq "\x0A"; |
# my ($self, $charset_name, $byte_stream, $doc, $onerror, $get_wrapper) = @_; |
| 363 |
$self->{next_input_character} = 0x000A; # LF # MUST |
my $self = ref $_[0] ? shift : shift->new; |
| 364 |
$line++; |
my $charset_name = shift; |
| 365 |
$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; |
|
| 366 |
|
|
| 367 |
my $onerror = $_[2] || sub { |
my $onerror = $_[2] || sub { |
| 368 |
my (%opt) = @_; |
my (%opt) = @_; |
| 369 |
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); |
|
| 370 |
}; |
}; |
| 371 |
|
$self->{parse_error} = $onerror; # updated later by parse_char_string |
| 372 |
|
|
| 373 |
$self->_initialize_tokenizer; |
my $get_wrapper = $_[3] || sub ($) { |
| 374 |
$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 { |
|
|
# |
|
| 375 |
}; |
}; |
|
return $self; |
|
|
} # new |
|
| 376 |
|
|
| 377 |
sub CM_ENTITY () { 0b001 } # & markup in data |
## HTML5 encoding sniffing algorithm |
| 378 |
sub CM_LIMITED_MARKUP () { 0b010 } # < markup in data (limited) |
require Message::Charset::Info; |
| 379 |
sub CM_FULL_MARKUP () { 0b100 } # < markup in data (any) |
my $charset; |
| 380 |
|
my $buffer; |
| 381 |
sub PLAINTEXT_CONTENT_MODEL () { 0 } |
my ($char_stream, $e_status); |
| 382 |
sub CDATA_CONTENT_MODEL () { CM_LIMITED_MARKUP } |
|
| 383 |
sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP } |
SNIFFING: { |
| 384 |
sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP } |
## NOTE: By setting |allow_fallback| option true when the |
| 385 |
|
## |get_decode_handle| method is invoked, we ignore what the HTML5 |
| 386 |
|
## spec requires, i.e. unsupported encoding should be ignored. |
| 387 |
|
## TODO: We should not do this unless the parser is invoked |
| 388 |
|
## in the conformance checking mode, in which this behavior |
| 389 |
|
## would be useful. |
| 390 |
|
|
| 391 |
## Implementations MUST act as if state machine in the spec |
## Step 1 |
| 392 |
|
if (defined $charset_name) { |
| 393 |
|
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 394 |
|
## TODO: Is this ok? Transfer protocol's parameter should be |
| 395 |
|
## interpreted in its semantics? |
| 396 |
|
|
| 397 |
|
($char_stream, $e_status) = $charset->get_decode_handle |
| 398 |
|
($byte_stream, allow_error_reporting => 1, |
| 399 |
|
allow_fallback => 1); |
| 400 |
|
if ($char_stream) { |
| 401 |
|
$self->{confident} = 1; |
| 402 |
|
last SNIFFING; |
| 403 |
|
} else { |
| 404 |
|
!!!parse-error (type => 'charset:not supported', |
| 405 |
|
layer => 'encode', |
| 406 |
|
line => 1, column => 1, |
| 407 |
|
value => $charset_name, |
| 408 |
|
level => $self->{level}->{uncertain}); |
| 409 |
|
} |
| 410 |
|
} |
| 411 |
|
|
| 412 |
sub _initialize_tokenizer ($) { |
## Step 2 |
| 413 |
my $self = shift; |
my $byte_buffer = ''; |
| 414 |
$self->{state} = 'data'; # MUST |
for (1..1024) { |
| 415 |
$self->{content_model} = PCDATA_CONTENT_MODEL; # be |
my $char = $byte_stream->getc; |
| 416 |
undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE |
last unless defined $char; |
| 417 |
undef $self->{current_attribute}; |
$byte_buffer .= $char; |
| 418 |
undef $self->{last_emitted_start_tag_name}; |
} ## TODO: timeout |
|
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. |
|
| 419 |
|
|
| 420 |
sub _get_next_token ($) { |
## Step 3 |
| 421 |
my $self = shift; |
if ($byte_buffer =~ /^\xFE\xFF/) { |
| 422 |
if (@{$self->{token}}) { |
$charset = Message::Charset::Info->get_by_html_name ('utf-16be'); |
| 423 |
return shift @{$self->{token}}; |
($char_stream, $e_status) = $charset->get_decode_handle |
| 424 |
} |
($byte_stream, allow_error_reporting => 1, |
| 425 |
|
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 426 |
|
$self->{confident} = 1; |
| 427 |
|
last SNIFFING; |
| 428 |
|
} elsif ($byte_buffer =~ /^\xFF\xFE/) { |
| 429 |
|
$charset = Message::Charset::Info->get_by_html_name ('utf-16le'); |
| 430 |
|
($char_stream, $e_status) = $charset->get_decode_handle |
| 431 |
|
($byte_stream, allow_error_reporting => 1, |
| 432 |
|
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 433 |
|
$self->{confident} = 1; |
| 434 |
|
last SNIFFING; |
| 435 |
|
} elsif ($byte_buffer =~ /^\xEF\xBB\xBF/) { |
| 436 |
|
$charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 437 |
|
($char_stream, $e_status) = $charset->get_decode_handle |
| 438 |
|
($byte_stream, allow_error_reporting => 1, |
| 439 |
|
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 440 |
|
$self->{confident} = 1; |
| 441 |
|
last SNIFFING; |
| 442 |
|
} |
| 443 |
|
|
| 444 |
A: { |
## Step 4 |
| 445 |
if ($self->{state} eq 'data') { |
## TODO: <meta charset> |
|
if ($self->{next_input_character} == 0x0026) { # & |
|
|
if ($self->{content_model} & CM_ENTITY) { # PCDATA | RCDATA |
|
|
$self->{state} = 'entity data'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($self->{next_input_character} == 0x002D) { # - |
|
|
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | 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} & CM_FULL_MARKUP or # PCDATA |
|
|
(($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA |
|
|
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} & CM_LIMITED_MARKUP)) { # RCDATA | 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 (0); |
|
| 446 |
|
|
| 447 |
$self->{state} = 'data'; |
## Step 5 |
| 448 |
# next-input-character is already done |
## TODO: from history |
| 449 |
|
|
| 450 |
unless (defined $token) { |
## Step 6 |
| 451 |
!!!emit ({type => 'character', data => '&'}); |
require Whatpm::Charset::UniversalCharDet; |
| 452 |
} else { |
$charset_name = Whatpm::Charset::UniversalCharDet->detect_byte_string |
| 453 |
!!!emit ($token); |
($byte_buffer); |
| 454 |
|
if (defined $charset_name) { |
| 455 |
|
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 456 |
|
|
| 457 |
|
require Whatpm::Charset::DecodeHandle; |
| 458 |
|
$buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
| 459 |
|
($byte_stream); |
| 460 |
|
($char_stream, $e_status) = $charset->get_decode_handle |
| 461 |
|
($buffer, allow_error_reporting => 1, |
| 462 |
|
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 463 |
|
if ($char_stream) { |
| 464 |
|
$buffer->{buffer} = $byte_buffer; |
| 465 |
|
!!!parse-error (type => 'sniffing:chardet', |
| 466 |
|
text => $charset_name, |
| 467 |
|
level => $self->{level}->{info}, |
| 468 |
|
layer => 'encode', |
| 469 |
|
line => 1, column => 1); |
| 470 |
|
$self->{confident} = 0; |
| 471 |
|
last SNIFFING; |
| 472 |
} |
} |
| 473 |
|
} |
| 474 |
|
|
| 475 |
redo A; |
## Step 7: default |
| 476 |
} elsif ($self->{state} eq 'tag open') { |
## TODO: Make this configurable. |
| 477 |
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
$charset = Message::Charset::Info->get_by_html_name ('windows-1252'); |
| 478 |
if ($self->{next_input_character} == 0x002F) { # / |
## NOTE: We choose |windows-1252| here, since |utf-8| should be |
| 479 |
!!!next-input-character; |
## detectable in the step 6. |
| 480 |
$self->{state} = 'close tag open'; |
require Whatpm::Charset::DecodeHandle; |
| 481 |
redo A; |
$buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
| 482 |
} else { |
($byte_stream); |
| 483 |
## reconsume |
($char_stream, $e_status) |
| 484 |
$self->{state} = 'data'; |
= $charset->get_decode_handle ($buffer, |
| 485 |
|
allow_error_reporting => 1, |
| 486 |
!!!emit ({type => 'character', data => '<'}); |
allow_fallback => 1, |
| 487 |
|
byte_buffer => \$byte_buffer); |
| 488 |
redo A; |
$buffer->{buffer} = $byte_buffer; |
| 489 |
} |
!!!parse-error (type => 'sniffing:default', |
| 490 |
} elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA |
text => 'windows-1252', |
| 491 |
if ($self->{next_input_character} == 0x0021) { # ! |
level => $self->{level}->{info}, |
| 492 |
$self->{state} = 'markup declaration open'; |
line => 1, column => 1, |
| 493 |
!!!next-input-character; |
layer => 'encode'); |
| 494 |
redo A; |
$self->{confident} = 0; |
| 495 |
} elsif ($self->{next_input_character} == 0x002F) { # / |
} # SNIFFING |
| 496 |
$self->{state} = 'close tag open'; |
|
| 497 |
!!!next-input-character; |
if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 498 |
redo A; |
$self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 499 |
} elsif (0x0041 <= $self->{next_input_character} and |
!!!parse-error (type => 'chardecode:fallback', |
| 500 |
$self->{next_input_character} <= 0x005A) { # A..Z |
#text => $self->{input_encoding}, |
| 501 |
$self->{current_token} |
level => $self->{level}->{uncertain}, |
| 502 |
= {type => 'start tag', |
line => 1, column => 1, |
| 503 |
tag_name => chr ($self->{next_input_character} + 0x0020)}; |
layer => 'encode'); |
| 504 |
$self->{state} = 'tag name'; |
} elsif (not ($e_status & |
| 505 |
!!!next-input-character; |
Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) { |
| 506 |
redo A; |
$self->{input_encoding} = $charset->get_iana_name; |
| 507 |
} elsif (0x0061 <= $self->{next_input_character} and |
!!!parse-error (type => 'chardecode:no error', |
| 508 |
$self->{next_input_character} <= 0x007A) { # a..z |
text => $self->{input_encoding}, |
| 509 |
$self->{current_token} = {type => 'start tag', |
level => $self->{level}->{uncertain}, |
| 510 |
tag_name => chr ($self->{next_input_character})}; |
line => 1, column => 1, |
| 511 |
$self->{state} = 'tag name'; |
layer => 'encode'); |
| 512 |
!!!next-input-character; |
} else { |
| 513 |
redo A; |
$self->{input_encoding} = $charset->get_iana_name; |
| 514 |
} 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 |
|
|
|
|
|
!!!emit ({type => 'character', data => '<'}); |
|
| 515 |
|
|
| 516 |
redo A; |
$self->{change_encoding} = sub { |
| 517 |
} |
my $self = shift; |
| 518 |
} else { |
$charset_name = shift; |
| 519 |
die "$0: $self->{content_model} in tag open"; |
my $token = shift; |
|
} |
|
|
} elsif ($self->{state} eq 'close tag open') { |
|
|
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
|
if (defined $self->{last_emitted_start_tag_name}) { |
|
|
## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564> |
|
|
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'; |
|
| 520 |
|
|
| 521 |
!!!emit ({type => 'character', data => '</'}); |
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 522 |
|
($char_stream, $e_status) = $charset->get_decode_handle |
| 523 |
redo A; |
($byte_stream, allow_error_reporting => 1, allow_fallback => 1, |
| 524 |
} |
byte_buffer => \ $buffer->{buffer}); |
| 525 |
} |
|
| 526 |
push @next_char, $self->{next_input_character}; |
if ($char_stream) { # if supported |
| 527 |
|
## "Change the encoding" algorithm: |
|
unless ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020 or # SP |
|
|
$self->{next_input_character} == 0x003E or # > |
|
|
$self->{next_input_character} == 0x002F or # / |
|
|
$self->{next_input_character} == -1) { |
|
|
$self->{next_input_character} = shift @next_char; # reconsume |
|
|
!!!back-next-input-character (@next_char); |
|
|
$self->{state} = 'data'; |
|
|
!!!emit ({type => 'character', data => '</'}); |
|
|
redo A; |
|
|
} else { |
|
|
$self->{next_input_character} = shift @next_char; |
|
|
!!!back-next-input-character (@next_char); |
|
|
# and consume... |
|
|
} |
|
|
} else { |
|
|
## No start tag token has ever been emitted |
|
|
# next-input-character is already done |
|
|
$self->{state} = 'data'; |
|
|
!!!emit ({type => 'character', data => '</'}); |
|
|
redo A; |
|
|
} |
|
|
} |
|
| 528 |
|
|
| 529 |
if (0x0041 <= $self->{next_input_character} and |
## Step 1 |
| 530 |
$self->{next_input_character} <= 0x005A) { # A..Z |
if (defined $self->{input_encoding} and |
| 531 |
$self->{current_token} = {type => 'end tag', |
$self->{input_encoding} eq $charset_name) { |
| 532 |
tag_name => chr ($self->{next_input_character} + 0x0020)}; |
!!!parse-error (type => 'charset label:matching', |
| 533 |
$self->{state} = 'tag name'; |
text => $charset_name, |
| 534 |
!!!next-input-character; |
level => $self->{level}->{info}); |
| 535 |
redo A; |
$self->{confident} = 1; |
| 536 |
} elsif (0x0061 <= $self->{next_input_character} and |
return; |
|
$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->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
|
|
$self->{current_token}->{tag_name} .= chr ($self->{next_input_character} + 0x0020); |
|
|
# start tag or end tag |
|
|
## Stay in this state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x003E and # > |
|
|
$self->{current_token}->{type} eq 'start tag' and |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
|
|
# permitted slash |
|
|
# |
|
|
} else { |
|
|
!!!parse-error (type => 'nestc'); |
|
|
} |
|
|
$self->{state} = 'before attribute name'; |
|
|
# next-input-character is already done |
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_token}->{tag_name} .= chr $self->{next_input_character}; |
|
|
# start tag or end tag |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'before attribute name') { |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020) { # SP |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), |
|
|
value => ''}; |
|
|
$self->{state} = 'attribute name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x003E and # > |
|
|
$self->{current_token}->{type} eq 'start tag' and |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
|
|
# permitted slash |
|
|
# |
|
|
} else { |
|
|
!!!parse-error (type => 'nestc'); |
|
|
} |
|
|
## Stay in the state |
|
|
# next-input-character is already done |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character}), |
|
|
value => ''}; |
|
|
$self->{state} = 'attribute name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'attribute name') { |
|
|
my $before_leave = sub { |
|
|
if (exists $self->{current_token}->{attributes} # start tag or end tag |
|
|
->{$self->{current_attribute}->{name}}) { # MUST |
|
|
!!!parse-error (type => 'duplicate attribute:'.$self->{current_attribute}->{name}); |
|
|
## 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->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
|
|
$self->{current_attribute}->{name} .= chr ($self->{next_input_character} + 0x0020); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
|
|
$before_leave->(); |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x003E and # > |
|
|
$self->{current_token}->{type} eq 'start tag' and |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
|
|
# permitted slash |
|
|
# |
|
|
} else { |
|
|
!!!parse-error (type => 'nestc'); |
|
|
} |
|
|
$self->{state} = 'before attribute name'; |
|
|
# next-input-character is already done |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
$before_leave->(); |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute}->{name} .= chr ($self->{next_input_character}); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'after attribute name') { |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020) { # SP |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003D) { # = |
|
|
$self->{state} = 'before attribute value'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), |
|
|
value => ''}; |
|
|
$self->{state} = 'attribute name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x003E and # > |
|
|
$self->{current_token}->{type} eq 'start tag' and |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
|
|
# permitted slash |
|
|
# |
|
|
} else { |
|
|
!!!parse-error (type => 'nestc'); |
|
|
## TODO: Different error type for <aa / bb> than <aa/> |
|
|
} |
|
|
$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->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character}), |
|
|
value => ''}; |
|
|
$self->{state} = 'attribute name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'before attribute value') { |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020) { # SP |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0022) { # " |
|
|
$self->{state} = 'attribute value (double-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0026) { # & |
|
|
$self->{state} = 'attribute value (unquoted)'; |
|
|
## reconsume |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0027) { # ' |
|
|
$self->{state} = 'attribute value (single-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute}->{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->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute}->{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->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'attribute value (unquoted)') { |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # HT |
|
|
$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} == 0x0026) { # & |
|
|
$self->{last_attribute_value_state} = 'attribute value (unquoted)'; |
|
|
$self->{state} = 'entity in attribute value'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
|
|
$self->{current_token}->{first_start_tag} |
|
|
= not defined $self->{last_emitted_start_tag_name}; |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{current_token}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
| 537 |
} |
} |
|
} elsif ($self->{state} eq 'entity in attribute value') { |
|
|
my $token = $self->_tokenize_attempt_to_consume_an_entity (1); |
|
| 538 |
|
|
| 539 |
unless (defined $token) { |
## Step 2 (HTML5 revision 3205) |
| 540 |
$self->{current_attribute}->{value} .= '&'; |
if (defined $self->{input_encoding} and |
| 541 |
} else { |
Message::Charset::Info->get_by_html_name ($self->{input_encoding}) |
| 542 |
$self->{current_attribute}->{value} .= $token->{data}; |
->{category} & Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 543 |
## ISSUE: spec says "append the returned character token to the current attribute's value" |
$self->{confident} = 1; |
| 544 |
|
return; |
| 545 |
} |
} |
| 546 |
|
|
| 547 |
$self->{state} = $self->{last_attribute_value_state}; |
## Step 3 |
| 548 |
# next-input-character is already done |
if ($charset->{category} & |
| 549 |
redo A; |
Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 550 |
} elsif ($self->{state} eq 'bogus comment') { |
$charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 551 |
## (only happen if PCDATA state) |
($char_stream, $e_status) = $charset->get_decode_handle |
| 552 |
|
($byte_stream, |
| 553 |
my $token = {type => 'comment', data => ''}; |
byte_buffer => \ $buffer->{buffer}); |
| 554 |
|
} |
| 555 |
BC: { |
$charset_name = $charset->get_iana_name; |
| 556 |
if ($self->{next_input_character} == 0x003E) { # > |
|
| 557 |
$self->{state} = 'data'; |
!!!parse-error (type => 'charset label detected', |
| 558 |
!!!next-input-character; |
text => $self->{input_encoding}, |
| 559 |
|
value => $charset_name, |
| 560 |
!!!emit ($token); |
level => $self->{level}->{warn}, |
| 561 |
|
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}; |
|
| 562 |
|
|
| 563 |
if ($self->{next_input_character} == 0x002D) { # - |
## Step 4 |
| 564 |
!!!next-input-character; |
# if (can) { |
| 565 |
push @next_char, $self->{next_input_character}; |
## change the encoding on the fly. |
| 566 |
if ($self->{next_input_character} == 0x002D) { # - |
#$self->{confident} = 1; |
| 567 |
$self->{current_token} = {type => 'comment', data => ''}; |
#return; |
| 568 |
$self->{state} = 'comment start'; |
# } |
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{next_input_character} == 0x0044 or # D |
|
|
$self->{next_input_character} == 0x0064) { # d |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_input_character}; |
|
|
if ($self->{next_input_character} == 0x004F or # O |
|
|
$self->{next_input_character} == 0x006F) { # o |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_input_character}; |
|
|
if ($self->{next_input_character} == 0x0043 or # C |
|
|
$self->{next_input_character} == 0x0063) { # c |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_input_character}; |
|
|
if ($self->{next_input_character} == 0x0054 or # T |
|
|
$self->{next_input_character} == 0x0074) { # t |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_input_character}; |
|
|
if ($self->{next_input_character} == 0x0059 or # Y |
|
|
$self->{next_input_character} == 0x0079) { # y |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_input_character}; |
|
|
if ($self->{next_input_character} == 0x0050 or # P |
|
|
$self->{next_input_character} == 0x0070) { # p |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_input_character}; |
|
|
if ($self->{next_input_character} == 0x0045 or # E |
|
|
$self->{next_input_character} == 0x0065) { # e |
|
|
## ISSUE: What a stupid code this is! |
|
|
$self->{state} = 'DOCTYPE'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'bogus comment'); |
|
|
$self->{next_input_character} = shift @next_char; |
|
|
!!!back-next-input-character (@next_char); |
|
|
$self->{state} = 'bogus comment'; |
|
|
redo A; |
|
| 569 |
|
|
| 570 |
## ISSUE: typos in spec: chacacters, is is a parse error |
## Step 5 |
| 571 |
## 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 (); |
| 572 |
} elsif ($self->{state} eq 'comment start') { |
} |
| 573 |
if ($self->{next_input_character} == 0x002D) { # - |
}; # $self->{change_encoding} |
|
$self->{state} = 'comment start dash'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
!!!parse-error (type => 'bogus comment'); |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # comment |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
| 574 |
|
|
| 575 |
!!!emit ($self->{current_token}); # comment |
my $char_onerror = sub { |
| 576 |
|
my (undef, $type, %opt) = @_; |
| 577 |
|
!!!parse-error (layer => 'encode', |
| 578 |
|
line => $self->{line}, column => $self->{column} + 1, |
| 579 |
|
%opt, type => $type); |
| 580 |
|
if ($opt{octets}) { |
| 581 |
|
${$opt{octets}} = "\x{FFFD}"; # relacement character |
| 582 |
|
} |
| 583 |
|
}; |
| 584 |
|
|
| 585 |
redo A; |
my $wrapped_char_stream = $get_wrapper->($char_stream); |
| 586 |
} else { |
$wrapped_char_stream->onerror ($char_onerror); |
|
$self->{current_token}->{data} # comment |
|
|
.= chr ($self->{next_input_character}); |
|
|
$self->{state} = 'comment'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'comment start dash') { |
|
|
if ($self->{next_input_character} == 0x002D) { # - |
|
|
$self->{state} = 'comment end'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
!!!parse-error (type => 'bogus comment'); |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # comment |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
| 587 |
|
|
| 588 |
!!!emit ($self->{current_token}); # comment |
my @args = ($_[1], $_[2]); # $doc, $onerror - $get_wrapper = undef; |
| 589 |
|
my $return; |
| 590 |
|
try { |
| 591 |
|
$return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 592 |
|
} catch Whatpm::HTML::RestartParser with { |
| 593 |
|
## NOTE: Invoked after {change_encoding}. |
| 594 |
|
|
| 595 |
|
if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 596 |
|
$self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 597 |
|
!!!parse-error (type => 'chardecode:fallback', |
| 598 |
|
level => $self->{level}->{uncertain}, |
| 599 |
|
#text => $self->{input_encoding}, |
| 600 |
|
line => 1, column => 1, |
| 601 |
|
layer => 'encode'); |
| 602 |
|
} elsif (not ($e_status & |
| 603 |
|
Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) { |
| 604 |
|
$self->{input_encoding} = $charset->get_iana_name; |
| 605 |
|
!!!parse-error (type => 'chardecode:no error', |
| 606 |
|
text => $self->{input_encoding}, |
| 607 |
|
level => $self->{level}->{uncertain}, |
| 608 |
|
line => 1, column => 1, |
| 609 |
|
layer => 'encode'); |
| 610 |
|
} else { |
| 611 |
|
$self->{input_encoding} = $charset->get_iana_name; |
| 612 |
|
} |
| 613 |
|
$self->{confident} = 1; |
| 614 |
|
|
| 615 |
redo A; |
$wrapped_char_stream = $get_wrapper->($char_stream); |
| 616 |
} else { |
$wrapped_char_stream->onerror ($char_onerror); |
|
$self->{current_token}->{data} # comment |
|
|
.= '-' . chr ($self->{next_input_character}); |
|
|
$self->{state} = 'comment'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'comment') { |
|
|
if ($self->{next_input_character} == 0x002D) { # - |
|
|
$self->{state} = 'comment end dash'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
| 617 |
|
|
| 618 |
!!!emit ($self->{current_token}); # comment |
$return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 619 |
|
}; |
| 620 |
|
return $return; |
| 621 |
|
} # parse_byte_stream |
| 622 |
|
|
| 623 |
redo A; |
## NOTE: HTML5 spec says that the encoding layer MUST NOT strip BOM |
| 624 |
} else { |
## and the HTML layer MUST ignore it. However, we does strip BOM in |
| 625 |
$self->{current_token}->{data} .= chr ($self->{next_input_character}); # comment |
## the encoding layer and the HTML layer does not ignore any U+FEFF, |
| 626 |
## Stay in the state |
## because the core part of our HTML parser expects a string of character, |
| 627 |
!!!next-input-character; |
## not a string of bytes or code units or anything which might contain a BOM. |
| 628 |
redo A; |
## Therefore, any parser interface that accepts a string of bytes, |
| 629 |
} |
## such as |parse_byte_string| in this module, must ensure that it does |
| 630 |
} elsif ($self->{state} eq 'comment end dash') { |
## strip the BOM and never strip any ZWNBSP. |
|
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 |
|
| 631 |
|
|
| 632 |
!!!emit ($self->{current_token}); # comment |
sub parse_char_string ($$$;$$) { |
| 633 |
|
#my ($self, $s, $doc, $onerror, $get_wrapper) = @_; |
| 634 |
|
my $self = shift; |
| 635 |
|
my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 636 |
|
require Whatpm::Charset::DecodeHandle; |
| 637 |
|
my $input = Whatpm::Charset::DecodeHandle::CharString->new ($s); |
| 638 |
|
return $self->parse_char_stream ($input, @_[1..$#_]); |
| 639 |
|
} # parse_char_string |
| 640 |
|
*parse_string = \&parse_char_string; ## NOTE: Alias for backward compatibility. |
| 641 |
|
|
| 642 |
|
sub parse_char_stream ($$$;$$) { |
| 643 |
|
my $self = ref $_[0] ? shift : shift->new; |
| 644 |
|
my $input = $_[0]; |
| 645 |
|
$self->{document} = $_[1]; |
| 646 |
|
@{$self->{document}->child_nodes} = (); |
| 647 |
|
|
| 648 |
redo A; |
## NOTE: |set_inner_html| copies most of this method's code |
|
} 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 |
|
|
|
|
|
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'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
| 649 |
|
|
| 650 |
!!!emit ($self->{current_token}); # comment |
$self->{confident} = 1 unless exists $self->{confident}; |
| 651 |
|
$self->{document}->input_encoding ($self->{input_encoding}) |
| 652 |
|
if defined $self->{input_encoding}; |
| 653 |
|
## TODO: |{input_encoding}| is needless? |
| 654 |
|
|
| 655 |
|
$self->{line_prev} = $self->{line} = 1; |
| 656 |
|
$self->{column_prev} = -1; |
| 657 |
|
$self->{column} = 0; |
| 658 |
|
$self->{set_nc} = sub { |
| 659 |
|
my $self = shift; |
| 660 |
|
|
| 661 |
redo A; |
my $char = ''; |
| 662 |
} else { |
if (defined $self->{next_nc}) { |
| 663 |
!!!parse-error (type => 'dash in comment'); |
$char = $self->{next_nc}; |
| 664 |
$self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment |
delete $self->{next_nc}; |
| 665 |
$self->{state} = 'comment'; |
$self->{nc} = ord $char; |
| 666 |
!!!next-input-character; |
} else { |
| 667 |
redo A; |
$self->{char_buffer} = ''; |
| 668 |
} |
$self->{char_buffer_pos} = 0; |
|
} elsif ($self->{state} eq 'DOCTYPE') { |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020) { # SP |
|
|
$self->{state} = 'before DOCTYPE name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'no space before DOCTYPE name'); |
|
|
$self->{state} = 'before DOCTYPE name'; |
|
|
## reconsume |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'before DOCTYPE name') { |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020) { # SP |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
!!!parse-error (type => 'no DOCTYPE name'); |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ({type => 'DOCTYPE'}); # incorrect |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'no DOCTYPE name'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
| 669 |
|
|
| 670 |
!!!emit ({type => 'DOCTYPE'}); # incorrect |
my $count = $input->manakai_read_until |
| 671 |
|
($self->{char_buffer}, qr/[^\x00\x0A\x0D]/, $self->{char_buffer_pos}); |
| 672 |
|
if ($count) { |
| 673 |
|
$self->{line_prev} = $self->{line}; |
| 674 |
|
$self->{column_prev} = $self->{column}; |
| 675 |
|
$self->{column}++; |
| 676 |
|
$self->{nc} |
| 677 |
|
= ord substr ($self->{char_buffer}, $self->{char_buffer_pos}++, 1); |
| 678 |
|
return; |
| 679 |
|
} |
| 680 |
|
|
| 681 |
redo A; |
if ($input->read ($char, 1)) { |
| 682 |
|
$self->{nc} = ord $char; |
| 683 |
} else { |
} else { |
| 684 |
$self->{current_token} |
$self->{nc} = -1; |
| 685 |
= {type => 'DOCTYPE', |
return; |
| 686 |
name => chr ($self->{next_input_character}), |
} |
| 687 |
correct => 1}; |
} |
|
## ISSUE: "Set the token's name name to the" in the spec |
|
|
$self->{state} = 'DOCTYPE name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'DOCTYPE name') { |
|
|
## ISSUE: Redundant "First," in the spec. |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
|
|
$self->{next_input_character} == 0x000A or # LF |
|
|
$self->{next_input_character} == 0x000B or # VT |
|
|
$self->{next_input_character} == 0x000C or # FF |
|
|
$self->{next_input_character} == 0x0020) { # SP |
|
|
$self->{state} = 'after DOCTYPE name'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
| 688 |
|
|
| 689 |
redo A; |
($self->{line_prev}, $self->{column_prev}) |
| 690 |
} else { |
= ($self->{line}, $self->{column}); |
| 691 |
$self->{current_token}->{name} |
$self->{column}++; |
| 692 |
.= chr ($self->{next_input_character}); # DOCTYPE |
|
| 693 |
## Stay in the state |
if ($self->{nc} == 0x000A) { # LF |
| 694 |
!!!next-input-character; |
!!!cp ('j1'); |
| 695 |
redo A; |
$self->{line}++; |
| 696 |
} |
$self->{column} = 0; |
| 697 |
} elsif ($self->{state} eq 'after DOCTYPE name') { |
} elsif ($self->{nc} == 0x000D) { # CR |
| 698 |
if ($self->{next_input_character} == 0x0009 or # HT |
!!!cp ('j2'); |
| 699 |
$self->{next_input_character} == 0x000A or # LF |
## TODO: support for abort/streaming |
| 700 |
$self->{next_input_character} == 0x000B or # VT |
my $next = ''; |
| 701 |
$self->{next_input_character} == 0x000C or # FF |
if ($input->read ($next, 1) and $next ne "\x0A") { |
| 702 |
$self->{next_input_character} == 0x0020) { # SP |
$self->{next_nc} = $next; |
| 703 |
## Stay in the state |
} |
| 704 |
!!!next-input-character; |
$self->{nc} = 0x000A; # LF # MUST |
| 705 |
redo A; |
$self->{line}++; |
| 706 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
$self->{column} = 0; |
| 707 |
$self->{state} = 'data'; |
} elsif ($self->{nc} == 0x0000) { # NULL |
| 708 |
!!!next-input-character; |
!!!cp ('j4'); |
| 709 |
|
!!!parse-error (type => 'NULL'); |
| 710 |
!!!emit ($self->{current_token}); # DOCTYPE |
$self->{nc} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 711 |
|
} |
| 712 |
redo A; |
}; |
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0050 or # P |
|
|
$self->{next_input_character} == 0x0070) { # p |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x0055 or # U |
|
|
$self->{next_input_character} == 0x0075) { # u |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x0042 or # B |
|
|
$self->{next_input_character} == 0x0062) { # b |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x004C or # L |
|
|
$self->{next_input_character} == 0x006C) { # l |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x0049 or # I |
|
|
$self->{next_input_character} == 0x0069) { # i |
|
|
!!!next-input-character; |
|
|
if ($self->{next_input_character} == 0x0043 or # C |
|
|
$self->{next_input_character} == 0x0063) { # c |
|
|
$self->{state} = 'before DOCTYPE public identifier'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
| 713 |
|
|
| 714 |
# |
$self->{read_until} = sub { |
| 715 |
} elsif ($self->{next_input_character} == 0x0053 or # S |
#my ($scalar, $specials_range, $offset) = @_; |
| 716 |
$self->{next_input_character} == 0x0073) { # s |
return 0 if defined $self->{next_nc}; |
| 717 |
!!!next-input-character; |
|
| 718 |
if ($self->{next_input_character} == 0x0059 or # Y |
my $pattern = qr/[^$_[1]\x00\x0A\x0D]/; |
| 719 |
$self->{next_input_character} == 0x0079) { # y |
my $offset = $_[2] || 0; |
| 720 |
!!!next-input-character; |
|
| 721 |
if ($self->{next_input_character} == 0x0053 or # S |
if ($self->{char_buffer_pos} < length $self->{char_buffer}) { |
| 722 |
$self->{next_input_character} == 0x0073) { # s |
pos ($self->{char_buffer}) = $self->{char_buffer_pos}; |
| 723 |
!!!next-input-character; |
if ($self->{char_buffer} =~ /\G(?>$pattern)+/) { |
| 724 |
if ($self->{next_input_character} == 0x0054 or # T |
substr ($_[0], $offset) |
| 725 |
$self->{next_input_character} == 0x0074) { # t |
= substr ($self->{char_buffer}, $-[0], $+[0] - $-[0]); |
| 726 |
!!!next-input-character; |
my $count = $+[0] - $-[0]; |
| 727 |
if ($self->{next_input_character} == 0x0045 or # E |
if ($count) { |
| 728 |
$self->{next_input_character} == 0x0065) { # e |
$self->{column} += $count; |
| 729 |
!!!next-input-character; |
$self->{char_buffer_pos} += $count; |
| 730 |
if ($self->{next_input_character} == 0x004D or # M |
$self->{line_prev} = $self->{line}; |
| 731 |
$self->{next_input_character} == 0x006D) { # m |
$self->{column_prev} = $self->{column} - 1; |
| 732 |
$self->{state} = 'before DOCTYPE system identifier'; |
$self->{nc} = -1; |
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
| 733 |
} |
} |
| 734 |
|
return $count; |
|
# |
|
| 735 |
} else { |
} else { |
| 736 |
!!!next-input-character; |
return 0; |
|
# |
|
| 737 |
} |
} |
| 738 |
|
} else { |
| 739 |
|
my $count = $input->manakai_read_until ($_[0], $pattern, $_[2]); |
| 740 |
|
if ($count) { |
| 741 |
|
$self->{column} += $count; |
| 742 |
|
$self->{line_prev} = $self->{line}; |
| 743 |
|
$self->{column_prev} = $self->{column} - 1; |
| 744 |
|
$self->{nc} = -1; |
| 745 |
|
} |
| 746 |
|
return $count; |
| 747 |
|
} |
| 748 |
|
}; # $self->{read_until} |
| 749 |
|
|
| 750 |
!!!parse-error (type => 'string after DOCTYPE name'); |
my $onerror = $_[2] || sub { |
| 751 |
$self->{state} = 'bogus DOCTYPE'; |
my (%opt) = @_; |
| 752 |
# next-input-character is already done |
my $line = $opt{token} ? $opt{token}->{line} : $opt{line}; |
| 753 |
redo A; |
my $column = $opt{token} ? $opt{token}->{column} : $opt{column}; |
| 754 |
} elsif ($self->{state} eq 'before DOCTYPE public identifier') { |
warn "Parse error ($opt{type}) at line $line column $column\n"; |
| 755 |
if ({ |
}; |
| 756 |
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, |
$self->{parse_error} = sub { |
| 757 |
#0x000D => 1, # HT, LF, VT, FF, SP, CR |
$onerror->(line => $self->{line}, column => $self->{column}, @_); |
| 758 |
}->{$self->{next_input_character}}) { |
}; |
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} eq 0x0022) { # " |
|
|
$self->{current_token}->{public_identifier} = ''; # DOCTYPE |
|
|
$self->{state} = 'DOCTYPE public identifier (double-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} eq 0x0027) { # ' |
|
|
$self->{current_token}->{public_identifier} = ''; # DOCTYPE |
|
|
$self->{state} = 'DOCTYPE public identifier (single-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} eq 0x003E) { # > |
|
|
!!!parse-error (type => 'no PUBLIC literal'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'string after PUBLIC'); |
|
|
$self->{state} = 'bogus DOCTYPE'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'DOCTYPE public identifier (double-quoted)') { |
|
|
if ($self->{next_input_character} == 0x0022) { # " |
|
|
$self->{state} = 'after DOCTYPE public identifier'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_token}->{public_identifier} # DOCTYPE |
|
|
.= chr $self->{next_input_character}; |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'DOCTYPE public identifier (single-quoted)') { |
|
|
if ($self->{next_input_character} == 0x0027) { # ' |
|
|
$self->{state} = 'after DOCTYPE public identifier'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_token}->{public_identifier} # DOCTYPE |
|
|
.= chr $self->{next_input_character}; |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'after DOCTYPE public identifier') { |
|
|
if ({ |
|
|
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, |
|
|
#0x000D => 1, # HT, LF, VT, FF, SP, CR |
|
|
}->{$self->{next_input_character}}) { |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0022) { # " |
|
|
$self->{current_token}->{system_identifier} = ''; # DOCTYPE |
|
|
$self->{state} = 'DOCTYPE system identifier (double-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0027) { # ' |
|
|
$self->{current_token}->{system_identifier} = ''; # DOCTYPE |
|
|
$self->{state} = 'DOCTYPE system identifier (single-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'string after PUBLIC literal'); |
|
|
$self->{state} = 'bogus DOCTYPE'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'before DOCTYPE system identifier') { |
|
|
if ({ |
|
|
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, |
|
|
#0x000D => 1, # HT, LF, VT, FF, SP, CR |
|
|
}->{$self->{next_input_character}}) { |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0022) { # " |
|
|
$self->{current_token}->{system_identifier} = ''; # DOCTYPE |
|
|
$self->{state} = 'DOCTYPE system identifier (double-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x0027) { # ' |
|
|
$self->{current_token}->{system_identifier} = ''; # DOCTYPE |
|
|
$self->{state} = 'DOCTYPE system identifier (single-quoted)'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
!!!parse-error (type => 'no SYSTEM literal'); |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'string after SYSTEM'); |
|
|
$self->{state} = 'bogus DOCTYPE'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'DOCTYPE system identifier (double-quoted)') { |
|
|
if ($self->{next_input_character} == 0x0022) { # " |
|
|
$self->{state} = 'after DOCTYPE system identifier'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed SYSTEM literal'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_token}->{system_identifier} # DOCTYPE |
|
|
.= chr $self->{next_input_character}; |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'DOCTYPE system identifier (single-quoted)') { |
|
|
if ($self->{next_input_character} == 0x0027) { # ' |
|
|
$self->{state} = 'after DOCTYPE system identifier'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed SYSTEM literal'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
$self->{current_token}->{system_identifier} # DOCTYPE |
|
|
.= chr $self->{next_input_character}; |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'after DOCTYPE system identifier') { |
|
|
if ({ |
|
|
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, |
|
|
#0x000D => 1, # HT, LF, VT, FF, SP, CR |
|
|
}->{$self->{next_input_character}}) { |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'string after SYSTEM literal'); |
|
|
$self->{state} = 'bogus DOCTYPE'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} eq 'bogus DOCTYPE') { |
|
|
if ($self->{next_input_character} == 0x003E) { # > |
|
|
$self->{state} = 'data'; |
|
|
!!!next-input-character; |
|
|
|
|
|
delete $self->{current_token}->{correct}; |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_input_character} == -1) { |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = 'data'; |
|
|
## reconsume |
|
| 759 |
|
|
| 760 |
delete $self->{current_token}->{correct}; |
my $char_onerror = sub { |
| 761 |
!!!emit ($self->{current_token}); # DOCTYPE |
my (undef, $type, %opt) = @_; |
| 762 |
|
!!!parse-error (layer => 'encode', |
| 763 |
|
line => $self->{line}, column => $self->{column} + 1, |
| 764 |
|
%opt, type => $type); |
| 765 |
|
}; # $char_onerror |
| 766 |
|
|
| 767 |
|
if ($_[3]) { |
| 768 |
|
$input = $_[3]->($input); |
| 769 |
|
$input->onerror ($char_onerror); |
| 770 |
|
} else { |
| 771 |
|
$input->onerror ($char_onerror) unless defined $input->onerror; |
| 772 |
|
} |
| 773 |
|
|
| 774 |
redo A; |
$self->_initialize_tokenizer; |
| 775 |
} else { |
$self->_initialize_tree_constructor; |
| 776 |
## Stay in the state |
$self->_construct_tree; |
| 777 |
!!!next-input-character; |
$self->_terminate_tree_constructor; |
|
redo A; |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{state}: Unknown state"; |
|
|
} |
|
|
} # A |
|
| 778 |
|
|
| 779 |
die "$0: _get_next_token: unexpected case"; |
delete $self->{parse_error}; # remove loop |
|
} # _get_next_token |
|
| 780 |
|
|
| 781 |
sub _tokenize_attempt_to_consume_an_entity ($$) { |
return $self->{document}; |
| 782 |
my ($self, $in_attr) = @_; |
} # parse_char_stream |
| 783 |
|
|
| 784 |
if ({ |
sub new ($) { |
| 785 |
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF, |
my $class = shift; |
| 786 |
0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR |
my $self = bless { |
| 787 |
}->{$self->{next_input_character}}) { |
level => {must => 'm', |
| 788 |
## Don't consume |
should => 's', |
| 789 |
## No error |
warn => 'w', |
| 790 |
return undef; |
info => 'i', |
| 791 |
} elsif ($self->{next_input_character} == 0x0023) { # # |
uncertain => 'u'}, |
| 792 |
!!!next-input-character; |
}, $class; |
| 793 |
if ($self->{next_input_character} == 0x0078 or # x |
$self->{set_nc} = sub { |
| 794 |
$self->{next_input_character} == 0x0058) { # X |
$self->{nc} = -1; |
| 795 |
my $code; |
}; |
| 796 |
X: { |
$self->{parse_error} = sub { |
| 797 |
my $x_char = $self->{next_input_character}; |
# |
| 798 |
!!!next-input-character; |
}; |
| 799 |
if (0x0030 <= $self->{next_input_character} and |
$self->{change_encoding} = sub { |
| 800 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
# if ($_[0] is a supported encoding) { |
| 801 |
$code ||= 0; |
# run "change the encoding" algorithm; |
| 802 |
$code *= 0x10; |
# throw Whatpm::HTML::RestartParser (charset => $new_encoding); |
| 803 |
$code += $self->{next_input_character} - 0x0030; |
# } |
| 804 |
redo X; |
}; |
| 805 |
} elsif (0x0061 <= $self->{next_input_character} and |
$self->{application_cache_selection} = sub { |
| 806 |
$self->{next_input_character} <= 0x0066) { # a..f |
# |
| 807 |
$code ||= 0; |
}; |
| 808 |
$code *= 0x10; |
return $self; |
| 809 |
$code += $self->{next_input_character} - 0x0060 + 9; |
} # new |
|
redo X; |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
|
|
$self->{next_input_character} <= 0x0046) { # A..F |
|
|
$code ||= 0; |
|
|
$code *= 0x10; |
|
|
$code += $self->{next_input_character} - 0x0040 + 9; |
|
|
redo X; |
|
|
} elsif (not defined $code) { # no hexadecimal digit |
|
|
!!!parse-error (type => 'bare hcro'); |
|
|
!!!back-next-input-character ($x_char, $self->{next_input_character}); |
|
|
$self->{next_input_character} = 0x0023; # # |
|
|
return undef; |
|
|
} elsif ($self->{next_input_character} == 0x003B) { # ; |
|
|
!!!next-input-character; |
|
|
} else { |
|
|
!!!parse-error (type => 'no refc'); |
|
|
} |
|
|
|
|
|
if ($code == 0 or (0xD800 <= $code and $code <= 0xDFFF)) { |
|
|
!!!parse-error (type => sprintf 'invalid character reference:U+%04X', $code); |
|
|
$code = 0xFFFD; |
|
|
} elsif ($code > 0x10FFFF) { |
|
|
!!!parse-error (type => sprintf 'invalid character reference:U-%08X', $code); |
|
|
$code = 0xFFFD; |
|
|
} elsif ($code == 0x000D) { |
|
|
!!!parse-error (type => 'CR character reference'); |
|
|
$code = 0x000A; |
|
|
} elsif (0x80 <= $code and $code <= 0x9F) { |
|
|
!!!parse-error (type => sprintf 'C1 character reference:U+%04X', $code); |
|
|
$code = $c1_entity_char->{$code}; |
|
|
} |
|
|
|
|
|
return {type => 'character', data => chr $code}; |
|
|
} # 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; |
|
|
} |
|
| 810 |
|
|
| 811 |
if ($self->{next_input_character} == 0x003B) { # ; |
## Insertion modes |
|
!!!next-input-character; |
|
|
} else { |
|
|
!!!parse-error (type => 'no refc'); |
|
|
} |
|
| 812 |
|
|
| 813 |
if ($code == 0 or (0xD800 <= $code and $code <= 0xDFFF)) { |
sub AFTER_HTML_IMS () { 0b100 } |
| 814 |
!!!parse-error (type => sprintf 'invalid character reference:U+%04X', $code); |
sub HEAD_IMS () { 0b1000 } |
| 815 |
$code = 0xFFFD; |
sub BODY_IMS () { 0b10000 } |
| 816 |
} elsif ($code > 0x10FFFF) { |
sub BODY_TABLE_IMS () { 0b100000 } |
| 817 |
!!!parse-error (type => sprintf 'invalid character reference:U-%08X', $code); |
sub TABLE_IMS () { 0b1000000 } |
| 818 |
$code = 0xFFFD; |
sub ROW_IMS () { 0b10000000 } |
| 819 |
} elsif ($code == 0x000D) { |
sub BODY_AFTER_IMS () { 0b100000000 } |
| 820 |
!!!parse-error (type => 'CR character reference'); |
sub FRAME_IMS () { 0b1000000000 } |
| 821 |
$code = 0x000A; |
sub SELECT_IMS () { 0b10000000000 } |
| 822 |
} elsif (0x80 <= $code and $code <= 0x9F) { |
#sub IN_FOREIGN_CONTENT_IM () { 0b100000000000 } # see Whatpm::HTML::Tokenizer |
| 823 |
!!!parse-error (type => sprintf 'C1 character reference:U+%04X', $code); |
## NOTE: "in foreign content" insertion mode is special; it is combined |
| 824 |
$code = $c1_entity_char->{$code}; |
## with the secondary insertion mode. In this parser, they are stored |
| 825 |
} |
## together in the bit-or'ed form. |
| 826 |
|
sub IN_CDATA_RCDATA_IM () { 0b1000000000000 } |
| 827 |
return {type => 'character', data => chr $code}; |
## NOTE: "in CDATA/RCDATA" insertion mode is also special; it is |
| 828 |
} else { |
## combined with the original insertion mode. In thie parser, |
| 829 |
!!!parse-error (type => 'bare nero'); |
## they are stored together in the bit-or'ed form. |
| 830 |
!!!back-next-input-character ($self->{next_input_character}); |
|
| 831 |
$self->{next_input_character} = 0x0023; # # |
sub IM_MASK () { 0b11111111111 } |
| 832 |
return undef; |
|
| 833 |
} |
## NOTE: "initial" and "before html" insertion modes have no constants. |
| 834 |
} elsif ((0x0041 <= $self->{next_input_character} and |
|
| 835 |
$self->{next_input_character} <= 0x005A) or |
## NOTE: "after after body" insertion mode. |
| 836 |
(0x0061 <= $self->{next_input_character} and |
sub AFTER_HTML_BODY_IM () { AFTER_HTML_IMS | BODY_AFTER_IMS } |
| 837 |
$self->{next_input_character} <= 0x007A)) { |
|
| 838 |
my $entity_name = chr $self->{next_input_character}; |
## NOTE: "after after frameset" insertion mode. |
| 839 |
!!!next-input-character; |
sub AFTER_HTML_FRAMESET_IM () { AFTER_HTML_IMS | FRAME_IMS } |
| 840 |
|
|
| 841 |
my $value = $entity_name; |
sub IN_HEAD_IM () { HEAD_IMS | 0b00 } |
| 842 |
my $match = 0; |
sub IN_HEAD_NOSCRIPT_IM () { HEAD_IMS | 0b01 } |
| 843 |
require Whatpm::_NamedEntityList; |
sub AFTER_HEAD_IM () { HEAD_IMS | 0b10 } |
| 844 |
our $EntityChar; |
sub BEFORE_HEAD_IM () { HEAD_IMS | 0b11 } |
| 845 |
|
sub IN_BODY_IM () { BODY_IMS } |
| 846 |
while (length $entity_name < 10 and |
sub IN_CELL_IM () { BODY_IMS | BODY_TABLE_IMS | 0b01 } |
| 847 |
## NOTE: Some number greater than the maximum length of entity name |
sub IN_CAPTION_IM () { BODY_IMS | BODY_TABLE_IMS | 0b10 } |
| 848 |
((0x0041 <= $self->{next_input_character} and # a |
sub IN_ROW_IM () { TABLE_IMS | ROW_IMS | 0b01 } |
| 849 |
$self->{next_input_character} <= 0x005A) or # x |
sub IN_TABLE_BODY_IM () { TABLE_IMS | ROW_IMS | 0b10 } |
| 850 |
(0x0061 <= $self->{next_input_character} and # a |
sub IN_TABLE_IM () { TABLE_IMS } |
| 851 |
$self->{next_input_character} <= 0x007A) or # z |
sub AFTER_BODY_IM () { BODY_AFTER_IMS } |
| 852 |
(0x0030 <= $self->{next_input_character} and # 0 |
sub IN_FRAMESET_IM () { FRAME_IMS | 0b01 } |
| 853 |
$self->{next_input_character} <= 0x0039) or # 9 |
sub AFTER_FRAMESET_IM () { FRAME_IMS | 0b10 } |
| 854 |
$self->{next_input_character} == 0x003B)) { # ; |
sub IN_SELECT_IM () { SELECT_IMS | 0b01 } |
| 855 |
$entity_name .= chr $self->{next_input_character}; |
sub IN_SELECT_IN_TABLE_IM () { SELECT_IMS | 0b10 } |
| 856 |
if (defined $EntityChar->{$entity_name}) { |
sub IN_COLUMN_GROUP_IM () { 0b10 } |
|
if ($self->{next_input_character} == 0x003B) { # ; |
|
|
$value = $EntityChar->{$entity_name}; |
|
|
$match = 1; |
|
|
!!!next-input-character; |
|
|
last; |
|
|
} else { |
|
|
$value = $EntityChar->{$entity_name}; |
|
|
$match = -1; |
|
|
!!!next-input-character; |
|
|
} |
|
|
} else { |
|
|
$value .= chr $self->{next_input_character}; |
|
|
$match *= 2; |
|
|
!!!next-input-character; |
|
|
} |
|
|
} |
|
|
|
|
|
if ($match > 0) { |
|
|
return {type => 'character', data => $value}; |
|
|
} elsif ($match < 0) { |
|
|
!!!parse-error (type => 'no refc'); |
|
|
if ($in_attr and $match < -1) { |
|
|
return {type => 'character', data => '&'.$entity_name}; |
|
|
} else { |
|
|
return {type => 'character', data => $value}; |
|
|
} |
|
|
} else { |
|
|
!!!parse-error (type => 'bare ero'); |
|
|
## NOTE: No characters are consumed in the spec. |
|
|
return {type => 'character', data => '&'.$value}; |
|
|
} |
|
|
} else { |
|
|
## no characters are consumed |
|
|
!!!parse-error (type => 'bare ero'); |
|
|
return undef; |
|
|
} |
|
|
} # _tokenize_attempt_to_consume_an_entity |
|
| 857 |
|
|
| 858 |
sub _initialize_tree_constructor ($) { |
sub _initialize_tree_constructor ($) { |
| 859 |
my $self = shift; |
my $self = shift; |
| 862 |
## TODO: Turn mutation events off # MUST |
## TODO: Turn mutation events off # MUST |
| 863 |
## TODO: Turn loose Document option (manakai extension) on |
## TODO: Turn loose Document option (manakai extension) on |
| 864 |
$self->{document}->manakai_is_html (1); # MUST |
$self->{document}->manakai_is_html (1); # MUST |
| 865 |
|
$self->{document}->set_user_data (manakai_source_line => 1); |
| 866 |
|
$self->{document}->set_user_data (manakai_source_column => 1); |
| 867 |
} # _initialize_tree_constructor |
} # _initialize_tree_constructor |
| 868 |
|
|
| 869 |
sub _terminate_tree_constructor ($) { |
sub _terminate_tree_constructor ($) { |
| 883 |
## When an interactive UA render the $self->{document} available |
## When an interactive UA render the $self->{document} available |
| 884 |
## to the user, or when it begin accepting user input, are |
## to the user, or when it begin accepting user input, are |
| 885 |
## 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 |
|
| 886 |
|
|
| 887 |
!!!next-token; |
!!!next-token; |
| 888 |
|
|
|
$self->{insertion_mode} = 'before head'; |
|
| 889 |
undef $self->{form_element}; |
undef $self->{form_element}; |
| 890 |
undef $self->{head_element}; |
undef $self->{head_element}; |
| 891 |
|
undef $self->{head_element_inserted}; |
| 892 |
$self->{open_elements} = []; |
$self->{open_elements} = []; |
| 893 |
undef $self->{inner_html_node}; |
undef $self->{inner_html_node}; |
| 894 |
|
undef $self->{ignore_newline}; |
| 895 |
|
|
| 896 |
|
## NOTE: The "initial" insertion mode. |
| 897 |
$self->_tree_construction_initial; # MUST |
$self->_tree_construction_initial; # MUST |
| 898 |
|
|
| 899 |
|
## NOTE: The "before html" insertion mode. |
| 900 |
$self->_tree_construction_root_element; |
$self->_tree_construction_root_element; |
| 901 |
|
$self->{insertion_mode} = BEFORE_HEAD_IM; |
| 902 |
|
|
| 903 |
|
## NOTE: The "before head" insertion mode and so on. |
| 904 |
$self->_tree_construction_main; |
$self->_tree_construction_main; |
| 905 |
} # _construct_tree |
} # _construct_tree |
| 906 |
|
|
| 907 |
sub _tree_construction_initial ($) { |
sub _tree_construction_initial ($) { |
| 908 |
my $self = shift; |
my $self = shift; |
| 909 |
|
|
| 910 |
|
## NOTE: "initial" insertion mode |
| 911 |
|
|
| 912 |
INITIAL: { |
INITIAL: { |
| 913 |
if ($token->{type} eq 'DOCTYPE') { |
if ($token->{type} == DOCTYPE_TOKEN) { |
| 914 |
## NOTE: Conformance checkers MAY, instead of reporting "not HTML5" |
## NOTE: Conformance checkers MAY, instead of reporting "not HTML5" |
| 915 |
## error, switch to a conformance checking mode for another |
## error, switch to a conformance checking mode for another |
| 916 |
## language. |
## language. |
| 917 |
my $doctype_name = $token->{name}; |
my $doctype_name = $token->{name}; |
| 918 |
$doctype_name = '' unless defined $doctype_name; |
$doctype_name = '' unless defined $doctype_name; |
| 919 |
$doctype_name =~ tr/a-z/A-Z/; |
$doctype_name =~ tr/a-z/A-Z/; # ASCII case-insensitive |
| 920 |
if (not defined $token->{name} or # <!DOCTYPE> |
if (not defined $token->{name} or # <!DOCTYPE> |
| 921 |
defined $token->{public_identifier} or |
defined $token->{sysid}) { |
| 922 |
defined $token->{system_identifier}) { |
!!!cp ('t1'); |
| 923 |
!!!parse-error (type => 'not HTML5'); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 924 |
} elsif ($doctype_name ne 'HTML') { |
} elsif ($doctype_name ne 'HTML') { |
| 925 |
## ISSUE: ASCII case-insensitive? (in fact it does not matter) |
!!!cp ('t2'); |
| 926 |
!!!parse-error (type => 'not HTML5'); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 927 |
|
} elsif (defined $token->{pubid}) { |
| 928 |
|
if ($token->{pubid} eq 'XSLT-compat') { |
| 929 |
|
!!!cp ('t1.2'); |
| 930 |
|
!!!parse-error (type => 'XSLT-compat', token => $token, |
| 931 |
|
level => $self->{level}->{should}); |
| 932 |
|
} else { |
| 933 |
|
!!!parse-error (type => 'not HTML5', token => $token); |
| 934 |
|
} |
| 935 |
|
} else { |
| 936 |
|
!!!cp ('t3'); |
| 937 |
|
# |
| 938 |
} |
} |
| 939 |
|
|
| 940 |
my $doctype = $self->{document}->create_document_type_definition |
my $doctype = $self->{document}->create_document_type_definition |
| 941 |
($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)? |
($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)? |
| 942 |
$doctype->public_id ($token->{public_identifier}) |
## NOTE: Default value for both |public_id| and |system_id| attributes |
| 943 |
if defined $token->{public_identifier}; |
## are empty strings, so that we don't set any value in missing cases. |
| 944 |
$doctype->system_id ($token->{system_identifier}) |
$doctype->public_id ($token->{pubid}) if defined $token->{pubid}; |
| 945 |
if defined $token->{system_identifier}; |
$doctype->system_id ($token->{sysid}) if defined $token->{sysid}; |
| 946 |
## NOTE: Other DocumentType attributes are null or empty lists. |
## NOTE: Other DocumentType attributes are null or empty lists. |
| 947 |
## ISSUE: internalSubset = null?? |
## In Firefox3, |internalSubset| attribute is set to the empty |
| 948 |
|
## string, while |null| is an allowed value for the attribute |
| 949 |
|
## according to DOM3 Core. |
| 950 |
$self->{document}->append_child ($doctype); |
$self->{document}->append_child ($doctype); |
| 951 |
|
|
| 952 |
if (not $token->{correct} or $doctype_name ne 'HTML') { |
if ($token->{quirks} or $doctype_name ne 'HTML') { |
| 953 |
|
!!!cp ('t4'); |
| 954 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 955 |
} elsif (defined $token->{public_identifier}) { |
} elsif (defined $token->{pubid}) { |
| 956 |
my $pubid = $token->{public_identifier}; |
my $pubid = $token->{pubid}; |
| 957 |
$pubid =~ tr/a-z/A-z/; |
$pubid =~ tr/a-z/A-z/; |
| 958 |
if ({ |
my $prefix = [ |
| 959 |
"+//SILMARIL//DTD HTML PRO V0R11 19970101//EN" => 1, |
"+//SILMARIL//DTD HTML PRO V0R11 19970101//", |
| 960 |
"-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1, |
"-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//", |
| 961 |
"-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1, |
"-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//", |
| 962 |
"-//IETF//DTD HTML 2.0 LEVEL 1//EN" => 1, |
"-//IETF//DTD HTML 2.0 LEVEL 1//", |
| 963 |
"-//IETF//DTD HTML 2.0 LEVEL 2//EN" => 1, |
"-//IETF//DTD HTML 2.0 LEVEL 2//", |
| 964 |
"-//IETF//DTD HTML 2.0 STRICT LEVEL 1//EN" => 1, |
"-//IETF//DTD HTML 2.0 STRICT LEVEL 1//", |
| 965 |
"-//IETF//DTD HTML 2.0 STRICT LEVEL 2//EN" => 1, |
"-//IETF//DTD HTML 2.0 STRICT LEVEL 2//", |
| 966 |
"-//IETF//DTD HTML 2.0 STRICT//EN" => 1, |
"-//IETF//DTD HTML 2.0 STRICT//", |
| 967 |
"-//IETF//DTD HTML 2.0//EN" => 1, |
"-//IETF//DTD HTML 2.0//", |
| 968 |
"-//IETF//DTD HTML 2.1E//EN" => 1, |
"-//IETF//DTD HTML 2.1E//", |
| 969 |
"-//IETF//DTD HTML 3.0//EN" => 1, |
"-//IETF//DTD HTML 3.0//", |
| 970 |
"-//IETF//DTD HTML 3.0//EN//" => 1, |
"-//IETF//DTD HTML 3.2 FINAL//", |
| 971 |
"-//IETF//DTD HTML 3.2 FINAL//EN" => 1, |
"-//IETF//DTD HTML 3.2//", |
| 972 |
"-//IETF//DTD HTML 3.2//EN" => 1, |
"-//IETF//DTD HTML 3//", |
| 973 |
"-//IETF//DTD HTML 3//EN" => 1, |
"-//IETF//DTD HTML LEVEL 0//", |
| 974 |
"-//IETF//DTD HTML LEVEL 0//EN" => 1, |
"-//IETF//DTD HTML LEVEL 1//", |
| 975 |
"-//IETF//DTD HTML LEVEL 0//EN//2.0" => 1, |
"-//IETF//DTD HTML LEVEL 2//", |
| 976 |
"-//IETF//DTD HTML LEVEL 1//EN" => 1, |
"-//IETF//DTD HTML LEVEL 3//", |
| 977 |
"-//IETF//DTD HTML LEVEL 1//EN//2.0" => 1, |
"-//IETF//DTD HTML STRICT LEVEL 0//", |
| 978 |
"-//IETF//DTD HTML LEVEL 2//EN" => 1, |
"-//IETF//DTD HTML STRICT LEVEL 1//", |
| 979 |
"-//IETF//DTD HTML LEVEL 2//EN//2.0" => 1, |
"-//IETF//DTD HTML STRICT LEVEL 2//", |
| 980 |
"-//IETF//DTD HTML LEVEL 3//EN" => 1, |
"-//IETF//DTD HTML STRICT LEVEL 3//", |
| 981 |
"-//IETF//DTD HTML LEVEL 3//EN//3.0" => 1, |
"-//IETF//DTD HTML STRICT//", |
| 982 |
"-//IETF//DTD HTML STRICT LEVEL 0//EN" => 1, |
"-//IETF//DTD HTML//", |
| 983 |
"-//IETF//DTD HTML STRICT LEVEL 0//EN//2.0" => 1, |
"-//METRIUS//DTD METRIUS PRESENTATIONAL//", |
| 984 |
"-//IETF//DTD HTML STRICT LEVEL 1//EN" => 1, |
"-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//", |
| 985 |
"-//IETF//DTD HTML STRICT LEVEL 1//EN//2.0" => 1, |
"-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//", |
| 986 |
"-//IETF//DTD HTML STRICT LEVEL 2//EN" => 1, |
"-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//", |
| 987 |
"-//IETF//DTD HTML STRICT LEVEL 2//EN//2.0" => 1, |
"-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//", |
| 988 |
"-//IETF//DTD HTML STRICT LEVEL 3//EN" => 1, |
"-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//", |
| 989 |
"-//IETF//DTD HTML STRICT LEVEL 3//EN//3.0" => 1, |
"-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//", |
| 990 |
"-//IETF//DTD HTML STRICT//EN" => 1, |
"-//NETSCAPE COMM. CORP.//DTD HTML//", |
| 991 |
"-//IETF//DTD HTML STRICT//EN//2.0" => 1, |
"-//NETSCAPE COMM. CORP.//DTD STRICT HTML//", |
| 992 |
"-//IETF//DTD HTML STRICT//EN//3.0" => 1, |
"-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//", |
| 993 |
"-//IETF//DTD HTML//EN" => 1, |
"-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//", |
| 994 |
"-//IETF//DTD HTML//EN//2.0" => 1, |
"-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//", |
| 995 |
"-//IETF//DTD HTML//EN//3.0" => 1, |
"-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//", |
| 996 |
"-//METRIUS//DTD METRIUS PRESENTATIONAL//EN" => 1, |
"-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//", |
| 997 |
"-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//EN" => 1, |
"-//SPYGLASS//DTD HTML 2.0 EXTENDED//", |
| 998 |
"-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//EN" => 1, |
"-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//", |
| 999 |
"-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//EN" => 1, |
"-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//", |
| 1000 |
"-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//EN" => 1, |
"-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//", |
| 1001 |
"-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//EN" => 1, |
"-//W3C//DTD HTML 3 1995-03-24//", |
| 1002 |
"-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//EN" => 1, |
"-//W3C//DTD HTML 3.2 DRAFT//", |
| 1003 |
"-//NETSCAPE COMM. CORP.//DTD HTML//EN" => 1, |
"-//W3C//DTD HTML 3.2 FINAL//", |
| 1004 |
"-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1, |
"-//W3C//DTD HTML 3.2//", |
| 1005 |
"-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1, |
"-//W3C//DTD HTML 3.2S DRAFT//", |
| 1006 |
"-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1, |
"-//W3C//DTD HTML 4.0 FRAMESET//", |
| 1007 |
"-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1, |
"-//W3C//DTD HTML 4.0 TRANSITIONAL//", |
| 1008 |
"-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1, |
"-//W3C//DTD HTML EXPERIMETNAL 19960712//", |
| 1009 |
"-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1, |
"-//W3C//DTD HTML EXPERIMENTAL 970421//", |
| 1010 |
"-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//EN" => 1, |
"-//W3C//DTD W3 HTML//", |
| 1011 |
"-//W3C//DTD HTML 3 1995-03-24//EN" => 1, |
"-//W3O//DTD W3 HTML 3.0//", |
| 1012 |
"-//W3C//DTD HTML 3.2 DRAFT//EN" => 1, |
"-//WEBTECHS//DTD MOZILLA HTML 2.0//", |
| 1013 |
"-//W3C//DTD HTML 3.2 FINAL//EN" => 1, |
"-//WEBTECHS//DTD MOZILLA HTML//", |
| 1014 |
"-//W3C//DTD HTML 3.2//EN" => 1, |
]; # $prefix |
| 1015 |
"-//W3C//DTD HTML 3.2S DRAFT//EN" => 1, |
my $match; |
| 1016 |
"-//W3C//DTD HTML 4.0 FRAMESET//EN" => 1, |
for (@$prefix) { |
| 1017 |
"-//W3C//DTD HTML 4.0 TRANSITIONAL//EN" => 1, |
if (substr ($prefix, 0, length $_) eq $_) { |
| 1018 |
"-//W3C//DTD HTML EXPERIMETNAL 19960712//EN" => 1, |
$match = 1; |
| 1019 |
"-//W3C//DTD HTML EXPERIMENTAL 970421//EN" => 1, |
last; |
| 1020 |
"-//W3C//DTD W3 HTML//EN" => 1, |
} |
| 1021 |
"-//W3O//DTD W3 HTML 3.0//EN" => 1, |
} |
| 1022 |
"-//W3O//DTD W3 HTML 3.0//EN//" => 1, |
if ($match or |
| 1023 |
"-//W3O//DTD W3 HTML STRICT 3.0//EN//" => 1, |
$pubid eq "-//W3O//DTD W3 HTML STRICT 3.0//EN//" or |
| 1024 |
"-//WEBTECHS//DTD MOZILLA HTML 2.0//EN" => 1, |
$pubid eq "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" or |
| 1025 |
"-//WEBTECHS//DTD MOZILLA HTML//EN" => 1, |
$pubid eq "HTML") { |
| 1026 |
"-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" => 1, |
!!!cp ('t5'); |
|
"HTML" => 1, |
|
|
}->{$pubid}) { |
|
| 1027 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 1028 |
} elsif ($pubid eq "-//W3C//DTD HTML 4.01 FRAMESET//EN" or |
} elsif ($pubid =~ m[^-//W3C//DTD HTML 4.01 FRAMESET//] or |
| 1029 |
$pubid eq "-//W3C//DTD HTML 4.01 TRANSITIONAL//EN") { |
$pubid =~ m[^-//W3C//DTD HTML 4.01 TRANSITIONAL//]) { |
| 1030 |
if (defined $token->{system_identifier}) { |
if (defined $token->{sysid}) { |
| 1031 |
|
!!!cp ('t6'); |
| 1032 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 1033 |
} else { |
} else { |
| 1034 |
|
!!!cp ('t7'); |
| 1035 |
$self->{document}->manakai_compat_mode ('limited quirks'); |
$self->{document}->manakai_compat_mode ('limited quirks'); |
| 1036 |
} |
} |
| 1037 |
} elsif ($pubid eq "-//W3C//DTD XHTML 1.0 Frameset//EN" or |
} elsif ($pubid =~ m[^-//W3C//DTD XHTML 1.0 FRAMESET//] or |
| 1038 |
$pubid eq "-//W3C//DTD XHTML 1.0 Transitional//EN") { |
$pubid =~ m[^-//W3C//DTD XHTML 1.0 TRANSITIONAL//]) { |
| 1039 |
|
!!!cp ('t8'); |
| 1040 |
$self->{document}->manakai_compat_mode ('limited quirks'); |
$self->{document}->manakai_compat_mode ('limited quirks'); |
| 1041 |
|
} else { |
| 1042 |
|
!!!cp ('t9'); |
| 1043 |
} |
} |
| 1044 |
|
} else { |
| 1045 |
|
!!!cp ('t10'); |
| 1046 |
} |
} |
| 1047 |
if (defined $token->{system_identifier}) { |
if (defined $token->{sysid}) { |
| 1048 |
my $sysid = $token->{system_identifier}; |
my $sysid = $token->{sysid}; |
| 1049 |
$sysid =~ tr/A-Z/a-z/; |
$sysid =~ tr/A-Z/a-z/; |
| 1050 |
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") { |
| 1051 |
|
## NOTE: Ensure that |PUBLIC "(limited quirks)" "(quirks)"| is |
| 1052 |
|
## marked as quirks. |
| 1053 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 1054 |
|
!!!cp ('t11'); |
| 1055 |
|
} else { |
| 1056 |
|
!!!cp ('t12'); |
| 1057 |
} |
} |
| 1058 |
|
} else { |
| 1059 |
|
!!!cp ('t13'); |
| 1060 |
} |
} |
| 1061 |
|
|
| 1062 |
## Go to the root element phase. |
## Go to the "before html" insertion mode. |
| 1063 |
!!!next-token; |
!!!next-token; |
| 1064 |
return; |
return; |
| 1065 |
} elsif ({ |
} elsif ({ |
| 1066 |
'start tag' => 1, |
START_TAG_TOKEN, 1, |
| 1067 |
'end tag' => 1, |
END_TAG_TOKEN, 1, |
| 1068 |
'end-of-file' => 1, |
END_OF_FILE_TOKEN, 1, |
| 1069 |
}->{$token->{type}}) { |
}->{$token->{type}}) { |
| 1070 |
!!!parse-error (type => 'no DOCTYPE'); |
!!!cp ('t14'); |
| 1071 |
|
!!!parse-error (type => 'no DOCTYPE', token => $token); |
| 1072 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 1073 |
## Go to the root element phase |
## Go to the "before html" insertion mode. |
| 1074 |
## reprocess |
## reprocess |
| 1075 |
|
!!!ack-later; |
| 1076 |
return; |
return; |
| 1077 |
} elsif ($token->{type} eq 'character') { |
} elsif ($token->{type} == CHARACTER_TOKEN) { |
| 1078 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { # \x0D |
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1079 |
## Ignore the token |
## Ignore the token |
| 1080 |
|
|
| 1081 |
unless (length $token->{data}) { |
unless (length $token->{data}) { |
| 1082 |
## Stay in the phase |
!!!cp ('t15'); |
| 1083 |
|
## Stay in the insertion mode. |
| 1084 |
!!!next-token; |
!!!next-token; |
| 1085 |
redo INITIAL; |
redo INITIAL; |
| 1086 |
|
} else { |
| 1087 |
|
!!!cp ('t16'); |
| 1088 |
} |
} |
| 1089 |
|
} else { |
| 1090 |
|
!!!cp ('t17'); |
| 1091 |
} |
} |
| 1092 |
|
|
| 1093 |
!!!parse-error (type => 'no DOCTYPE'); |
!!!parse-error (type => 'no DOCTYPE', token => $token); |
| 1094 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 1095 |
## Go to the root element phase |
## Go to the "before html" insertion mode. |
| 1096 |
## reprocess |
## reprocess |
| 1097 |
return; |
return; |
| 1098 |
} elsif ($token->{type} eq 'comment') { |
} elsif ($token->{type} == COMMENT_TOKEN) { |
| 1099 |
|
!!!cp ('t18'); |
| 1100 |
my $comment = $self->{document}->create_comment ($token->{data}); |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 1101 |
$self->{document}->append_child ($comment); |
$self->{document}->append_child ($comment); |
| 1102 |
|
|
| 1103 |
## Stay in the phase. |
## Stay in the insertion mode. |
| 1104 |
!!!next-token; |
!!!next-token; |
| 1105 |
redo INITIAL; |
redo INITIAL; |
| 1106 |
} else { |
} else { |
| 1107 |
die "$0: $token->{type}: Unknown token"; |
die "$0: $token->{type}: Unknown token type"; |
| 1108 |
} |
} |
| 1109 |
} # INITIAL |
} # INITIAL |
| 1110 |
|
|
| 1111 |
|
die "$0: _tree_construction_initial: This should be never reached"; |
| 1112 |
} # _tree_construction_initial |
} # _tree_construction_initial |
| 1113 |
|
|
| 1114 |
sub _tree_construction_root_element ($) { |
sub _tree_construction_root_element ($) { |
| 1115 |
my $self = shift; |
my $self = shift; |
| 1116 |
|
|
| 1117 |
|
## NOTE: "before html" insertion mode. |
| 1118 |
|
|
| 1119 |
B: { |
B: { |
| 1120 |
if ($token->{type} eq 'DOCTYPE') { |
if ($token->{type} == DOCTYPE_TOKEN) { |
| 1121 |
!!!parse-error (type => 'in html:#DOCTYPE'); |
!!!cp ('t19'); |
| 1122 |
|
!!!parse-error (type => 'in html:#DOCTYPE', token => $token); |
| 1123 |
## Ignore the token |
## Ignore the token |
| 1124 |
## Stay in the phase |
## Stay in the insertion mode. |
| 1125 |
!!!next-token; |
!!!next-token; |
| 1126 |
redo B; |
redo B; |
| 1127 |
} elsif ($token->{type} eq 'comment') { |
} elsif ($token->{type} == COMMENT_TOKEN) { |
| 1128 |
|
!!!cp ('t20'); |
| 1129 |
my $comment = $self->{document}->create_comment ($token->{data}); |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 1130 |
$self->{document}->append_child ($comment); |
$self->{document}->append_child ($comment); |
| 1131 |
## Stay in the phase |
## Stay in the insertion mode. |
| 1132 |
!!!next-token; |
!!!next-token; |
| 1133 |
redo B; |
redo B; |
| 1134 |
} elsif ($token->{type} eq 'character') { |
} elsif ($token->{type} == CHARACTER_TOKEN) { |
| 1135 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { # \x0D |
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1136 |
## Ignore the token. |
## Ignore the token. |
| 1137 |
|
|
| 1138 |
unless (length $token->{data}) { |
unless (length $token->{data}) { |
| 1139 |
## Stay in the phase |
!!!cp ('t21'); |
| 1140 |
|
## Stay in the insertion mode. |
| 1141 |
!!!next-token; |
!!!next-token; |
| 1142 |
redo B; |
redo B; |
| 1143 |
|
} else { |
| 1144 |
|
!!!cp ('t22'); |
| 1145 |
} |
} |
| 1146 |
|
} else { |
| 1147 |
|
!!!cp ('t23'); |
| 1148 |
} |
} |
| 1149 |
|
|
| 1150 |
|
$self->{application_cache_selection}->(undef); |
| 1151 |
|
|
| 1152 |
# |
# |
| 1153 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 1154 |
|
if ($token->{tag_name} eq 'html') { |
| 1155 |
|
my $root_element; |
| 1156 |
|
!!!create-element ($root_element, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); |
| 1157 |
|
$self->{document}->append_child ($root_element); |
| 1158 |
|
push @{$self->{open_elements}}, |
| 1159 |
|
[$root_element, $el_category->{html}]; |
| 1160 |
|
|
| 1161 |
|
if ($token->{attributes}->{manifest}) { |
| 1162 |
|
!!!cp ('t24'); |
| 1163 |
|
$self->{application_cache_selection} |
| 1164 |
|
->($token->{attributes}->{manifest}->{value}); |
| 1165 |
|
## ISSUE: Spec is unclear on relative references. |
| 1166 |
|
## According to Hixie (#whatwg 2008-03-19), it should be |
| 1167 |
|
## resolved against the base URI of the document in HTML |
| 1168 |
|
## or xml:base of the element in XHTML. |
| 1169 |
|
} else { |
| 1170 |
|
!!!cp ('t25'); |
| 1171 |
|
$self->{application_cache_selection}->(undef); |
| 1172 |
|
} |
| 1173 |
|
|
| 1174 |
|
!!!nack ('t25c'); |
| 1175 |
|
|
| 1176 |
|
!!!next-token; |
| 1177 |
|
return; ## Go to the "before head" insertion mode. |
| 1178 |
|
} else { |
| 1179 |
|
!!!cp ('t25.1'); |
| 1180 |
|
# |
| 1181 |
|
} |
| 1182 |
} elsif ({ |
} elsif ({ |
| 1183 |
'start tag' => 1, |
END_TAG_TOKEN, 1, |
| 1184 |
'end tag' => 1, |
END_OF_FILE_TOKEN, 1, |
|
'end-of-file' => 1, |
|
| 1185 |
}->{$token->{type}}) { |
}->{$token->{type}}) { |
| 1186 |
## ISSUE: There is an issue in the spec |
!!!cp ('t26'); |
| 1187 |
# |
# |
| 1188 |
} else { |
} else { |
| 1189 |
die "$0: $token->{type}: Unknown token"; |
die "$0: $token->{type}: Unknown token type"; |
| 1190 |
} |
} |
| 1191 |
my $root_element; !!!create-element ($root_element, 'html'); |
|
| 1192 |
$self->{document}->append_child ($root_element); |
my $root_element; |
| 1193 |
push @{$self->{open_elements}}, [$root_element, 'html']; |
!!!create-element ($root_element, $HTML_NS, 'html',, $token); |
| 1194 |
## reprocess |
$self->{document}->append_child ($root_element); |
| 1195 |
#redo B; |
push @{$self->{open_elements}}, [$root_element, $el_category->{html}]; |
| 1196 |
return; ## Go to the main phase. |
|
| 1197 |
|
$self->{application_cache_selection}->(undef); |
| 1198 |
|
|
| 1199 |
|
## NOTE: Reprocess the token. |
| 1200 |
|
!!!ack-later; |
| 1201 |
|
return; ## Go to the "before head" insertion mode. |
| 1202 |
} # B |
} # B |
| 1203 |
|
|
| 1204 |
|
die "$0: _tree_construction_root_element: This should never be reached"; |
| 1205 |
} # _tree_construction_root_element |
} # _tree_construction_root_element |
| 1206 |
|
|
| 1207 |
sub _reset_insertion_mode ($) { |
sub _reset_insertion_mode ($) { |
| 1216 |
|
|
| 1217 |
## Step 3 |
## Step 3 |
| 1218 |
S3: { |
S3: { |
|
## ISSUE: Oops! "If node is the first node in the stack of open |
|
|
## elements, then set last to true. If the context element of the |
|
|
## HTML fragment parsing algorithm is neither a td element nor a |
|
|
## th element, then set node to the context element. (fragment case)": |
|
|
## The second "if" is in the scope of the first "if"!? |
|
| 1219 |
if ($self->{open_elements}->[0]->[0] eq $node->[0]) { |
if ($self->{open_elements}->[0]->[0] eq $node->[0]) { |
| 1220 |
$last = 1; |
$last = 1; |
| 1221 |
if (defined $self->{inner_html_node}) { |
if (defined $self->{inner_html_node}) { |
| 1222 |
if ($self->{inner_html_node}->[1] eq 'td' or |
!!!cp ('t28'); |
| 1223 |
$self->{inner_html_node}->[1] eq 'th') { |
$node = $self->{inner_html_node}; |
| 1224 |
# |
} else { |
| 1225 |
} else { |
die "_reset_insertion_mode: t27"; |
|
$node = $self->{inner_html_node}; |
|
|
} |
|
| 1226 |
} |
} |
| 1227 |
} |
} |
| 1228 |
|
|
| 1229 |
## Step 4..13 |
## Step 4..14 |
| 1230 |
my $new_mode = { |
my $new_mode; |
| 1231 |
select => 'in select', |
if ($node->[1] & FOREIGN_EL) { |
| 1232 |
td => 'in cell', |
!!!cp ('t28.1'); |
| 1233 |
th => 'in cell', |
## NOTE: Strictly spaking, the line below only applies to MathML and |
| 1234 |
tr => 'in row', |
## SVG elements. Currently the HTML syntax supports only MathML and |
| 1235 |
tbody => 'in table body', |
## SVG elements as foreigners. |
| 1236 |
thead => 'in table head', |
$new_mode = IN_BODY_IM | IN_FOREIGN_CONTENT_IM; |
| 1237 |
tfoot => 'in table foot', |
} elsif ($node->[1] == TABLE_CELL_EL) { |
| 1238 |
caption => 'in caption', |
if ($last) { |
| 1239 |
colgroup => 'in column group', |
!!!cp ('t28.2'); |
| 1240 |
table => 'in table', |
# |
| 1241 |
head => 'in body', # not in head! |
} else { |
| 1242 |
body => 'in body', |
!!!cp ('t28.3'); |
| 1243 |
frameset => 'in frameset', |
$new_mode = IN_CELL_IM; |
| 1244 |
}->{$node->[1]}; |
} |
| 1245 |
|
} else { |
| 1246 |
|
!!!cp ('t28.4'); |
| 1247 |
|
$new_mode = { |
| 1248 |
|
select => IN_SELECT_IM, |
| 1249 |
|
## NOTE: |option| and |optgroup| do not set |
| 1250 |
|
## insertion mode to "in select" by themselves. |
| 1251 |
|
tr => IN_ROW_IM, |
| 1252 |
|
tbody => IN_TABLE_BODY_IM, |
| 1253 |
|
thead => IN_TABLE_BODY_IM, |
| 1254 |
|
tfoot => IN_TABLE_BODY_IM, |
| 1255 |
|
caption => IN_CAPTION_IM, |
| 1256 |
|
colgroup => IN_COLUMN_GROUP_IM, |
| 1257 |
|
table => IN_TABLE_IM, |
| 1258 |
|
head => IN_BODY_IM, # not in head! |
| 1259 |
|
body => IN_BODY_IM, |
| 1260 |
|
frameset => IN_FRAMESET_IM, |
| 1261 |
|
}->{$node->[0]->manakai_local_name}; |
| 1262 |
|
} |
| 1263 |
$self->{insertion_mode} = $new_mode and return if defined $new_mode; |
$self->{insertion_mode} = $new_mode and return if defined $new_mode; |
| 1264 |
|
|
| 1265 |
## Step 14 |
## Step 15 |
| 1266 |
if ($node->[1] eq 'html') { |
if ($node->[1] == HTML_EL) { |
| 1267 |
unless (defined $self->{head_element}) { |
unless (defined $self->{head_element}) { |
| 1268 |
$self->{insertion_mode} = 'before head'; |
!!!cp ('t29'); |
| 1269 |
|
$self->{insertion_mode} = BEFORE_HEAD_IM; |
| 1270 |
} else { |
} else { |
| 1271 |
$self->{insertion_mode} = 'after head'; |
## ISSUE: Can this state be reached? |
| 1272 |
|
!!!cp ('t30'); |
| 1273 |
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 1274 |
} |
} |
| 1275 |
return; |
return; |
| 1276 |
|
} else { |
| 1277 |
|
!!!cp ('t31'); |
| 1278 |
} |
} |
| 1279 |
|
|
|
## Step 15 |
|
|
$self->{insertion_mode} = 'in body' and return if $last; |
|
|
|
|
| 1280 |
## Step 16 |
## Step 16 |
| 1281 |
|
$self->{insertion_mode} = IN_BODY_IM and return if $last; |
| 1282 |
|
|
| 1283 |
|
## Step 17 |
| 1284 |
$i--; |
$i--; |
| 1285 |
$node = $self->{open_elements}->[$i]; |
$node = $self->{open_elements}->[$i]; |
| 1286 |
|
|
| 1287 |
## Step 17 |
## Step 18 |
| 1288 |
redo S3; |
redo S3; |
| 1289 |
} # S3 |
} # S3 |
| 1290 |
|
|
| 1291 |
|
die "$0: _reset_insertion_mode: This line should never be reached"; |
| 1292 |
} # _reset_insertion_mode |
} # _reset_insertion_mode |
| 1293 |
|
|
| 1294 |
sub _tree_construction_main ($) { |
sub _tree_construction_main ($) { |
| 1295 |
my $self = shift; |
my $self = shift; |
| 1296 |
|
|
|
my $previous_insertion_mode; |
|
|
|
|
| 1297 |
my $active_formatting_elements = []; |
my $active_formatting_elements = []; |
| 1298 |
|
|
| 1299 |
my $reconstruct_active_formatting_elements = sub { # MUST |
my $reconstruct_active_formatting_elements = sub { # MUST |
| 1310 |
return if $entry->[0] eq '#marker'; |
return if $entry->[0] eq '#marker'; |
| 1311 |
for (@{$self->{open_elements}}) { |
for (@{$self->{open_elements}}) { |
| 1312 |
if ($entry->[0] eq $_->[0]) { |
if ($entry->[0] eq $_->[0]) { |
| 1313 |
|
!!!cp ('t32'); |
| 1314 |
return; |
return; |
| 1315 |
} |
} |
| 1316 |
} |
} |
| 1325 |
|
|
| 1326 |
## Step 6 |
## Step 6 |
| 1327 |
if ($entry->[0] eq '#marker') { |
if ($entry->[0] eq '#marker') { |
| 1328 |
|
!!!cp ('t33_1'); |
| 1329 |
# |
# |
| 1330 |
} else { |
} else { |
| 1331 |
my $in_open_elements; |
my $in_open_elements; |
| 1332 |
OE: for (@{$self->{open_elements}}) { |
OE: for (@{$self->{open_elements}}) { |
| 1333 |
if ($entry->[0] eq $_->[0]) { |
if ($entry->[0] eq $_->[0]) { |
| 1334 |
|
!!!cp ('t33'); |
| 1335 |
$in_open_elements = 1; |
$in_open_elements = 1; |
| 1336 |
last OE; |
last OE; |
| 1337 |
} |
} |
| 1338 |
} |
} |
| 1339 |
if ($in_open_elements) { |
if ($in_open_elements) { |
| 1340 |
|
!!!cp ('t34'); |
| 1341 |
# |
# |
| 1342 |
} else { |
} else { |
| 1343 |
|
## NOTE: <!DOCTYPE HTML><p><b><i><u></p> <p>X |
| 1344 |
|
!!!cp ('t35'); |
| 1345 |
redo S4; |
redo S4; |
| 1346 |
} |
} |
| 1347 |
} |
} |
| 1364 |
|
|
| 1365 |
## Step 11 |
## Step 11 |
| 1366 |
unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { |
unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { |
| 1367 |
|
!!!cp ('t36'); |
| 1368 |
## Step 7' |
## Step 7' |
| 1369 |
$i++; |
$i++; |
| 1370 |
$entry = $active_formatting_elements->[$i]; |
$entry = $active_formatting_elements->[$i]; |
| 1371 |
|
|
| 1372 |
redo S7; |
redo S7; |
| 1373 |
} |
} |
| 1374 |
|
|
| 1375 |
|
!!!cp ('t37'); |
| 1376 |
} # S7 |
} # S7 |
| 1377 |
}; # $reconstruct_active_formatting_elements |
}; # $reconstruct_active_formatting_elements |
| 1378 |
|
|
| 1379 |
my $clear_up_to_marker = sub { |
my $clear_up_to_marker = sub { |
| 1380 |
for (reverse 0..$#$active_formatting_elements) { |
for (reverse 0..$#$active_formatting_elements) { |
| 1381 |
if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1382 |
|
!!!cp ('t38'); |
| 1383 |
splice @$active_formatting_elements, $_; |
splice @$active_formatting_elements, $_; |
| 1384 |
return; |
return; |
| 1385 |
} |
} |
| 1386 |
} |
} |
| 1387 |
|
|
| 1388 |
|
!!!cp ('t39'); |
| 1389 |
}; # $clear_up_to_marker |
}; # $clear_up_to_marker |
| 1390 |
|
|
| 1391 |
my $parse_rcdata = sub ($$) { |
my $insert; |
| 1392 |
my ($content_model_flag, $insert) = @_; |
|
| 1393 |
|
my $parse_rcdata = sub ($) { |
| 1394 |
|
my ($content_model_flag) = @_; |
| 1395 |
|
|
| 1396 |
## Step 1 |
## Step 1 |
| 1397 |
my $start_tag_name = $token->{tag_name}; |
my $start_tag_name = $token->{tag_name}; |
| 1398 |
my $el; |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
|
!!!create-element ($el, $start_tag_name, $token->{attributes}); |
|
| 1399 |
|
|
| 1400 |
## Step 2 |
## Step 2 |
|
$insert->($el); # /context node/->append_child ($el) |
|
|
|
|
|
## Step 3 |
|
| 1401 |
$self->{content_model} = $content_model_flag; # CDATA or RCDATA |
$self->{content_model} = $content_model_flag; # CDATA or RCDATA |
| 1402 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
| 1403 |
|
|
| 1404 |
## Step 4 |
## Step 3, 4 |
| 1405 |
my $text = ''; |
$self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
|
!!!next-token; |
|
|
while ($token->{type} eq 'character') { # or until stop tokenizing |
|
|
$text .= $token->{data}; |
|
|
!!!next-token; |
|
|
} |
|
|
|
|
|
## Step 5 |
|
|
if (length $text) { |
|
|
my $text = $self->{document}->create_text_node ($text); |
|
|
$el->append_child ($text); |
|
|
} |
|
|
|
|
|
## Step 6 |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; |
|
| 1406 |
|
|
| 1407 |
## Step 7 |
!!!nack ('t40.1'); |
|
if ($token->{type} eq 'end tag' and $token->{tag_name} eq $start_tag_name) { |
|
|
## Ignore the token |
|
|
} elsif ($content_model_flag == CDATA_CONTENT_MODEL) { |
|
|
!!!parse-error (type => 'in CDATA:#'.$token->{type}); |
|
|
} elsif ($content_model_flag == RCDATA_CONTENT_MODEL) { |
|
|
!!!parse-error (type => 'in RCDATA:#'.$token->{type}); |
|
|
} else { |
|
|
die "$0: $content_model_flag in parse_rcdata"; |
|
|
} |
|
| 1408 |
!!!next-token; |
!!!next-token; |
| 1409 |
}; # $parse_rcdata |
}; # $parse_rcdata |
| 1410 |
|
|
| 1411 |
my $script_start_tag = sub ($) { |
my $script_start_tag = sub () { |
| 1412 |
my $insert = $_[0]; |
## Step 1 |
| 1413 |
my $script_el; |
my $script_el; |
| 1414 |
!!!create-element ($script_el, 'script', $token->{attributes}); |
!!!create-element ($script_el, $HTML_NS, 'script', $token->{attributes}, $token); |
| 1415 |
|
|
| 1416 |
|
## Step 2 |
| 1417 |
## TODO: mark as "parser-inserted" |
## TODO: mark as "parser-inserted" |
| 1418 |
|
|
| 1419 |
|
## Step 3 |
| 1420 |
|
## TODO: Mark as "already executed", if ... |
| 1421 |
|
|
| 1422 |
|
## Step 4 |
| 1423 |
|
$insert->($script_el); |
| 1424 |
|
|
| 1425 |
|
## ISSUE: $script_el is not put into the stack |
| 1426 |
|
push @{$self->{open_elements}}, [$script_el, $el_category->{script}]; |
| 1427 |
|
|
| 1428 |
|
## Step 5 |
| 1429 |
$self->{content_model} = CDATA_CONTENT_MODEL; |
$self->{content_model} = CDATA_CONTENT_MODEL; |
| 1430 |
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); |
|
|
} |
|
|
|
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; |
|
| 1431 |
|
|
| 1432 |
if ($token->{type} eq 'end tag' and |
## Step 6-7 |
| 1433 |
$token->{tag_name} eq 'script') { |
$self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
|
## Ignore the token |
|
|
} else { |
|
|
!!!parse-error (type => 'in CDATA:#'.$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 |
|
| 1434 |
|
|
| 1435 |
$insert->($script_el); |
!!!nack ('t40.2'); |
|
|
|
|
## 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... |
|
|
} |
|
|
|
|
| 1436 |
!!!next-token; |
!!!next-token; |
| 1437 |
}; # $script_start_tag |
}; # $script_start_tag |
| 1438 |
|
|
| 1439 |
|
## NOTE: $open_tables->[-1]->[0] is the "current table" element node. |
| 1440 |
|
## NOTE: $open_tables->[-1]->[1] is the "tainted" flag. |
| 1441 |
|
## NOTE: $open_tables->[-1]->[2] is set false when non-Text node inserted. |
| 1442 |
|
my $open_tables = [[$self->{open_elements}->[0]->[0]]]; |
| 1443 |
|
|
| 1444 |
my $formatting_end_tag = sub { |
my $formatting_end_tag = sub { |
| 1445 |
my $tag_name = shift; |
my $end_tag_token = shift; |
| 1446 |
|
my $tag_name = $end_tag_token->{tag_name}; |
| 1447 |
|
|
| 1448 |
|
## NOTE: The adoption agency algorithm (AAA). |
| 1449 |
|
|
| 1450 |
FET: { |
FET: { |
| 1451 |
## Step 1 |
## Step 1 |
| 1452 |
my $formatting_element; |
my $formatting_element; |
| 1453 |
my $formatting_element_i_in_active; |
my $formatting_element_i_in_active; |
| 1454 |
AFE: for (reverse 0..$#$active_formatting_elements) { |
AFE: for (reverse 0..$#$active_formatting_elements) { |
| 1455 |
if ($active_formatting_elements->[$_]->[1] eq $tag_name) { |
if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1456 |
|
!!!cp ('t52'); |
| 1457 |
|
last AFE; |
| 1458 |
|
} elsif ($active_formatting_elements->[$_]->[0]->manakai_local_name |
| 1459 |
|
eq $tag_name) { |
| 1460 |
|
!!!cp ('t51'); |
| 1461 |
$formatting_element = $active_formatting_elements->[$_]; |
$formatting_element = $active_formatting_elements->[$_]; |
| 1462 |
$formatting_element_i_in_active = $_; |
$formatting_element_i_in_active = $_; |
| 1463 |
last AFE; |
last AFE; |
|
} elsif ($active_formatting_elements->[$_]->[0] eq '#marker') { |
|
|
last AFE; |
|
| 1464 |
} |
} |
| 1465 |
} # AFE |
} # AFE |
| 1466 |
unless (defined $formatting_element) { |
unless (defined $formatting_element) { |
| 1467 |
!!!parse-error (type => 'unmatched end tag:'.$tag_name); |
!!!cp ('t53'); |
| 1468 |
|
!!!parse-error (type => 'unmatched end tag', text => $tag_name, token => $end_tag_token); |
| 1469 |
## Ignore the token |
## Ignore the token |
| 1470 |
!!!next-token; |
!!!next-token; |
| 1471 |
return; |
return; |
| 1477 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 1478 |
if ($node->[0] eq $formatting_element->[0]) { |
if ($node->[0] eq $formatting_element->[0]) { |
| 1479 |
if ($in_scope) { |
if ($in_scope) { |
| 1480 |
|
!!!cp ('t54'); |
| 1481 |
$formatting_element_i_in_open = $_; |
$formatting_element_i_in_open = $_; |
| 1482 |
last INSCOPE; |
last INSCOPE; |
| 1483 |
} else { # in open elements but not in scope |
} else { # in open elements but not in scope |
| 1484 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t55'); |
| 1485 |
|
!!!parse-error (type => 'unmatched end tag', |
| 1486 |
|
text => $token->{tag_name}, |
| 1487 |
|
token => $end_tag_token); |
| 1488 |
## Ignore the token |
## Ignore the token |
| 1489 |
!!!next-token; |
!!!next-token; |
| 1490 |
return; |
return; |
| 1491 |
} |
} |
| 1492 |
} elsif ({ |
} elsif ($node->[1] & SCOPING_EL) { |
| 1493 |
table => 1, caption => 1, td => 1, th => 1, |
!!!cp ('t56'); |
|
button => 1, marquee => 1, object => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
| 1494 |
$in_scope = 0; |
$in_scope = 0; |
| 1495 |
} |
} |
| 1496 |
} # INSCOPE |
} # INSCOPE |
| 1497 |
unless (defined $formatting_element_i_in_open) { |
unless (defined $formatting_element_i_in_open) { |
| 1498 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t57'); |
| 1499 |
|
!!!parse-error (type => 'unmatched end tag', |
| 1500 |
|
text => $token->{tag_name}, |
| 1501 |
|
token => $end_tag_token); |
| 1502 |
pop @$active_formatting_elements; # $formatting_element |
pop @$active_formatting_elements; # $formatting_element |
| 1503 |
!!!next-token; ## TODO: ok? |
!!!next-token; ## TODO: ok? |
| 1504 |
return; |
return; |
| 1505 |
} |
} |
| 1506 |
if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { |
if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { |
| 1507 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t58'); |
| 1508 |
|
!!!parse-error (type => 'not closed', |
| 1509 |
|
text => $self->{open_elements}->[-1]->[0] |
| 1510 |
|
->manakai_local_name, |
| 1511 |
|
token => $end_tag_token); |
| 1512 |
} |
} |
| 1513 |
|
|
| 1514 |
## Step 2 |
## Step 2 |
| 1516 |
my $furthest_block_i_in_open; |
my $furthest_block_i_in_open; |
| 1517 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1518 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 1519 |
if (not $formatting_category->{$node->[1]} and |
if (not ($node->[1] & FORMATTING_EL) and |
| 1520 |
#not $phrasing_category->{$node->[1]} and |
#not $phrasing_category->{$node->[1]} and |
| 1521 |
($special_category->{$node->[1]} or |
($node->[1] & SPECIAL_EL or |
| 1522 |
$scoping_category->{$node->[1]})) { |
$node->[1] & SCOPING_EL)) { ## Scoping is redundant, maybe |
| 1523 |
|
!!!cp ('t59'); |
| 1524 |
$furthest_block = $node; |
$furthest_block = $node; |
| 1525 |
$furthest_block_i_in_open = $_; |
$furthest_block_i_in_open = $_; |
| 1526 |
|
## NOTE: The topmost (eldest) node. |
| 1527 |
} elsif ($node->[0] eq $formatting_element->[0]) { |
} elsif ($node->[0] eq $formatting_element->[0]) { |
| 1528 |
|
!!!cp ('t60'); |
| 1529 |
last OE; |
last OE; |
| 1530 |
} |
} |
| 1531 |
} # OE |
} # OE |
| 1532 |
|
|
| 1533 |
## Step 3 |
## Step 3 |
| 1534 |
unless (defined $furthest_block) { # MUST |
unless (defined $furthest_block) { # MUST |
| 1535 |
|
!!!cp ('t61'); |
| 1536 |
splice @{$self->{open_elements}}, $formatting_element_i_in_open; |
splice @{$self->{open_elements}}, $formatting_element_i_in_open; |
| 1537 |
splice @$active_formatting_elements, $formatting_element_i_in_active, 1; |
splice @$active_formatting_elements, $formatting_element_i_in_active, 1; |
| 1538 |
!!!next-token; |
!!!next-token; |
| 1545 |
## Step 5 |
## Step 5 |
| 1546 |
my $furthest_block_parent = $furthest_block->[0]->parent_node; |
my $furthest_block_parent = $furthest_block->[0]->parent_node; |
| 1547 |
if (defined $furthest_block_parent) { |
if (defined $furthest_block_parent) { |
| 1548 |
|
!!!cp ('t62'); |
| 1549 |
$furthest_block_parent->remove_child ($furthest_block->[0]); |
$furthest_block_parent->remove_child ($furthest_block->[0]); |
| 1550 |
} |
} |
| 1551 |
|
|
| 1568 |
S7S2: { |
S7S2: { |
| 1569 |
for (reverse 0..$#$active_formatting_elements) { |
for (reverse 0..$#$active_formatting_elements) { |
| 1570 |
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
| 1571 |
|
!!!cp ('t63'); |
| 1572 |
$node_i_in_active = $_; |
$node_i_in_active = $_; |
| 1573 |
last S7S2; |
last S7S2; |
| 1574 |
} |
} |
| 1582 |
|
|
| 1583 |
## Step 4 |
## Step 4 |
| 1584 |
if ($last_node->[0] eq $furthest_block->[0]) { |
if ($last_node->[0] eq $furthest_block->[0]) { |
| 1585 |
|
!!!cp ('t64'); |
| 1586 |
$bookmark_prev_el = $node->[0]; |
$bookmark_prev_el = $node->[0]; |
| 1587 |
} |
} |
| 1588 |
|
|
| 1589 |
## Step 5 |
## Step 5 |
| 1590 |
if ($node->[0]->has_child_nodes ()) { |
if ($node->[0]->has_child_nodes ()) { |
| 1591 |
|
!!!cp ('t65'); |
| 1592 |
my $clone = [$node->[0]->clone_node (0), $node->[1]]; |
my $clone = [$node->[0]->clone_node (0), $node->[1]]; |
| 1593 |
$active_formatting_elements->[$node_i_in_active] = $clone; |
$active_formatting_elements->[$node_i_in_active] = $clone; |
| 1594 |
$self->{open_elements}->[$node_i_in_open] = $clone; |
$self->{open_elements}->[$node_i_in_open] = $clone; |
| 1606 |
} # S7 |
} # S7 |
| 1607 |
|
|
| 1608 |
## Step 8 |
## Step 8 |
| 1609 |
$common_ancestor_node->[0]->append_child ($last_node->[0]); |
if ($common_ancestor_node->[1] & TABLE_ROWS_EL) { |
| 1610 |
|
my $foster_parent_element; |
| 1611 |
|
my $next_sibling; |
| 1612 |
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1613 |
|
if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1614 |
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 1615 |
|
if (defined $parent and $parent->node_type == 1) { |
| 1616 |
|
!!!cp ('t65.1'); |
| 1617 |
|
$foster_parent_element = $parent; |
| 1618 |
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1619 |
|
} else { |
| 1620 |
|
!!!cp ('t65.2'); |
| 1621 |
|
$foster_parent_element |
| 1622 |
|
= $self->{open_elements}->[$_ - 1]->[0]; |
| 1623 |
|
} |
| 1624 |
|
last OE; |
| 1625 |
|
} |
| 1626 |
|
} # OE |
| 1627 |
|
$foster_parent_element = $self->{open_elements}->[0]->[0] |
| 1628 |
|
unless defined $foster_parent_element; |
| 1629 |
|
$foster_parent_element->insert_before ($last_node->[0], $next_sibling); |
| 1630 |
|
$open_tables->[-1]->[1] = 1; # tainted |
| 1631 |
|
} else { |
| 1632 |
|
!!!cp ('t65.3'); |
| 1633 |
|
$common_ancestor_node->[0]->append_child ($last_node->[0]); |
| 1634 |
|
} |
| 1635 |
|
|
| 1636 |
## Step 9 |
## Step 9 |
| 1637 |
my $clone = [$formatting_element->[0]->clone_node (0), |
my $clone = [$formatting_element->[0]->clone_node (0), |
| 1648 |
my $i; |
my $i; |
| 1649 |
AFE: for (reverse 0..$#$active_formatting_elements) { |
AFE: for (reverse 0..$#$active_formatting_elements) { |
| 1650 |
if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { |
if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { |
| 1651 |
|
!!!cp ('t66'); |
| 1652 |
splice @$active_formatting_elements, $_, 1; |
splice @$active_formatting_elements, $_, 1; |
| 1653 |
$i-- and last AFE if defined $i; |
$i-- and last AFE if defined $i; |
| 1654 |
} elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { |
} elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { |
| 1655 |
|
!!!cp ('t67'); |
| 1656 |
$i = $_; |
$i = $_; |
| 1657 |
} |
} |
| 1658 |
} # AFE |
} # AFE |
| 1662 |
undef $i; |
undef $i; |
| 1663 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1664 |
if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { |
if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { |
| 1665 |
|
!!!cp ('t68'); |
| 1666 |
splice @{$self->{open_elements}}, $_, 1; |
splice @{$self->{open_elements}}, $_, 1; |
| 1667 |
$i-- and last OE if defined $i; |
$i-- and last OE if defined $i; |
| 1668 |
} elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { |
} elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { |
| 1669 |
|
!!!cp ('t69'); |
| 1670 |
$i = $_; |
$i = $_; |
| 1671 |
} |
} |
| 1672 |
} # OE |
} # OE |
| 1673 |
splice @{$self->{open_elements}}, $i + 1, 1, $clone; |
splice @{$self->{open_elements}}, $i + 1, 0, $clone; |
| 1674 |
|
|
| 1675 |
## Step 14 |
## Step 14 |
| 1676 |
redo FET; |
redo FET; |
| 1677 |
} # FET |
} # FET |
| 1678 |
}; # $formatting_end_tag |
}; # $formatting_end_tag |
| 1679 |
|
|
| 1680 |
my $insert_to_current = sub { |
$insert = my $insert_to_current = sub { |
| 1681 |
$self->{open_elements}->[-1]->[0]->append_child ($_[0]); |
$self->{open_elements}->[-1]->[0]->append_child ($_[0]); |
| 1682 |
}; # $insert_to_current |
}; # $insert_to_current |
| 1683 |
|
|
| 1684 |
my $insert_to_foster = sub { |
my $insert_to_foster = sub { |
| 1685 |
my $child = shift; |
my $child = shift; |
| 1686 |
if ({ |
if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1687 |
table => 1, tbody => 1, tfoot => 1, |
# MUST |
| 1688 |
thead => 1, tr => 1, |
my $foster_parent_element; |
| 1689 |
}->{$self->{open_elements}->[-1]->[1]}) { |
my $next_sibling; |
| 1690 |
# MUST |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1691 |
my $foster_parent_element; |
if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
|
my $next_sibling; |
|
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
|
| 1692 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 1693 |
if (defined $parent and $parent->node_type == 1) { |
if (defined $parent and $parent->node_type == 1) { |
| 1694 |
|
!!!cp ('t70'); |
| 1695 |
$foster_parent_element = $parent; |
$foster_parent_element = $parent; |
| 1696 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1697 |
} else { |
} else { |
| 1698 |
|
!!!cp ('t71'); |
| 1699 |
$foster_parent_element |
$foster_parent_element |
| 1700 |
= $self->{open_elements}->[$_ - 1]->[0]; |
= $self->{open_elements}->[$_ - 1]->[0]; |
| 1701 |
} |
} |
| 1706 |
unless defined $foster_parent_element; |
unless defined $foster_parent_element; |
| 1707 |
$foster_parent_element->insert_before |
$foster_parent_element->insert_before |
| 1708 |
($child, $next_sibling); |
($child, $next_sibling); |
| 1709 |
} else { |
$open_tables->[-1]->[1] = 1; # tainted |
| 1710 |
$self->{open_elements}->[-1]->[0]->append_child ($child); |
} else { |
| 1711 |
} |
!!!cp ('t72'); |
| 1712 |
|
$self->{open_elements}->[-1]->[0]->append_child ($child); |
| 1713 |
|
} |
| 1714 |
}; # $insert_to_foster |
}; # $insert_to_foster |
| 1715 |
|
|
| 1716 |
my $in_body = sub { |
## NOTE: Insert a character (MUST): When a character is inserted, if |
| 1717 |
my $insert = shift; |
## the last node that was inserted by the parser is a Text node and |
| 1718 |
if ($token->{type} eq 'start tag') { |
## the character has to be inserted after that node, then the |
| 1719 |
if ($token->{tag_name} eq 'script') { |
## character is appended to the Text node. However, if any other |
| 1720 |
## NOTE: This is an "as if in head" code clone |
## node is inserted by the parser, then a new Text node is created |
| 1721 |
$script_start_tag->($insert); |
## and the character is appended as that Text node. If I'm not |
| 1722 |
return; |
## wrong, for a parser with scripting disabled, there are only two |
| 1723 |
} elsif ($token->{tag_name} eq 'style') { |
## cases where this occurs. One is the case where an element node |
| 1724 |
## NOTE: This is an "as if in head" code clone |
## is inserted to the |head| element. This is covered by using the |
| 1725 |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
## |$self->{head_element_inserted}| flag. Another is the case where |
| 1726 |
return; |
## an element or comment is inserted into the |table| subtree while |
| 1727 |
} elsif ({ |
## foster parenting happens. This is covered by using the [2] flag |
| 1728 |
base => 1, link => 1, |
## of the |$open_tables| structure. All other cases are handled |
| 1729 |
}->{$token->{tag_name}}) { |
## simply by calling |manakai_append_text| method. |
| 1730 |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
|
| 1731 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
## TODO: |<body><script>document.write("a<br>"); |
| 1732 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
## document.body.removeChild (document.body.lastChild); |
| 1733 |
!!!next-token; |
## document.write ("b")</script>| |
| 1734 |
return; |
|
| 1735 |
} elsif ($token->{tag_name} eq 'meta') { |
B: while (1) { |
| 1736 |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
if ($token->{type} == DOCTYPE_TOKEN) { |
| 1737 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
!!!cp ('t73'); |
| 1738 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
!!!parse-error (type => 'in html:#DOCTYPE', token => $token); |
| 1739 |
|
## Ignore the token |
| 1740 |
|
## Stay in the phase |
| 1741 |
|
!!!next-token; |
| 1742 |
|
next B; |
| 1743 |
|
} elsif ($token->{type} == START_TAG_TOKEN and |
| 1744 |
|
$token->{tag_name} eq 'html') { |
| 1745 |
|
if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { |
| 1746 |
|
!!!cp ('t79'); |
| 1747 |
|
!!!parse-error (type => 'after html', text => 'html', token => $token); |
| 1748 |
|
$self->{insertion_mode} = AFTER_BODY_IM; |
| 1749 |
|
} elsif ($self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) { |
| 1750 |
|
!!!cp ('t80'); |
| 1751 |
|
!!!parse-error (type => 'after html', text => 'html', token => $token); |
| 1752 |
|
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
| 1753 |
|
} else { |
| 1754 |
|
!!!cp ('t81'); |
| 1755 |
|
} |
| 1756 |
|
|
| 1757 |
unless ($self->{confident}) { |
!!!cp ('t82'); |
| 1758 |
my $charset; |
!!!parse-error (type => 'not first start tag', token => $token); |
| 1759 |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
my $top_el = $self->{open_elements}->[0]->[0]; |
| 1760 |
$charset = $token->{attributes}->{charset}->{value}; |
for my $attr_name (keys %{$token->{attributes}}) { |
| 1761 |
} |
unless ($top_el->has_attribute_ns (undef, $attr_name)) { |
| 1762 |
if ($token->{attributes}->{'http-equiv'}) { |
!!!cp ('t84'); |
| 1763 |
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
$top_el->set_attribute_ns |
| 1764 |
if ($token->{attributes}->{'http-equiv'}->{value} |
(undef, [undef, $attr_name], |
| 1765 |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
$token->{attributes}->{$attr_name}->{value}); |
|
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
|
|
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
|
|
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
|
|
} ## TODO: And if supported |
|
|
} |
|
|
## TODO: Change the encoding |
|
| 1766 |
} |
} |
| 1767 |
|
} |
| 1768 |
!!!next-token; |
!!!nack ('t84.1'); |
| 1769 |
return; |
!!!next-token; |
| 1770 |
} elsif ($token->{tag_name} eq 'title') { |
next B; |
| 1771 |
!!!parse-error (type => 'in body:title'); |
} elsif ($token->{type} == COMMENT_TOKEN) { |
| 1772 |
## NOTE: This is an "as if in head" code clone |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 1773 |
$parse_rcdata->(RCDATA_CONTENT_MODEL, sub { |
if ($self->{insertion_mode} & AFTER_HTML_IMS) { |
| 1774 |
if (defined $self->{head_element}) { |
!!!cp ('t85'); |
| 1775 |
$self->{head_element}->append_child ($_[0]); |
$self->{document}->append_child ($comment); |
| 1776 |
} else { |
} elsif ($self->{insertion_mode} == AFTER_BODY_IM) { |
| 1777 |
$insert->($_[0]); |
!!!cp ('t86'); |
| 1778 |
} |
$self->{open_elements}->[0]->[0]->append_child ($comment); |
| 1779 |
}); |
} else { |
| 1780 |
return; |
!!!cp ('t87'); |
| 1781 |
} elsif ($token->{tag_name} eq 'body') { |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 1782 |
!!!parse-error (type => 'in body:body'); |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 1783 |
|
} |
| 1784 |
if (@{$self->{open_elements}} == 1 or |
!!!next-token; |
| 1785 |
$self->{open_elements}->[1]->[1] ne 'body') { |
next B; |
| 1786 |
## Ignore the token |
} elsif ($self->{insertion_mode} & IN_CDATA_RCDATA_IM) { |
| 1787 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 1788 |
|
$token->{data} =~ s/^\x0A// if $self->{ignore_newline}; |
| 1789 |
|
delete $self->{ignore_newline}; |
| 1790 |
|
|
| 1791 |
|
if (length $token->{data}) { |
| 1792 |
|
!!!cp ('t43'); |
| 1793 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text |
| 1794 |
|
($token->{data}); |
| 1795 |
} else { |
} else { |
| 1796 |
my $body_el = $self->{open_elements}->[1]->[0]; |
!!!cp ('t43.1'); |
|
for my $attr_name (keys %{$token->{attributes}}) { |
|
|
unless ($body_el->has_attribute_ns (undef, $attr_name)) { |
|
|
$body_el->set_attribute_ns |
|
|
(undef, [undef, $attr_name], |
|
|
$token->{attributes}->{$attr_name}->{value}); |
|
|
} |
|
|
} |
|
| 1797 |
} |
} |
| 1798 |
!!!next-token; |
!!!next-token; |
| 1799 |
return; |
next B; |
| 1800 |
} elsif ({ |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 1801 |
address => 1, blockquote => 1, center => 1, dir => 1, |
delete $self->{ignore_newline}; |
| 1802 |
div => 1, dl => 1, fieldset => 1, listing => 1, |
|
| 1803 |
menu => 1, ol => 1, p => 1, ul => 1, |
if ($token->{tag_name} eq 'script') { |
| 1804 |
pre => 1, |
!!!cp ('t50'); |
|
}->{$token->{tag_name}}) { |
|
|
## has a p element in scope |
|
|
INSCOPE: for (reverse @{$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 |
|
| 1805 |
|
|
| 1806 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
## Para 1-2 |
| 1807 |
if ($token->{tag_name} eq 'pre') { |
my $script = pop @{$self->{open_elements}}; |
| 1808 |
!!!next-token; |
|
| 1809 |
if ($token->{type} eq 'character') { |
## Para 3 |
| 1810 |
$token->{data} =~ s/^\x0A//; |
$self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1811 |
unless (length $token->{data}) { |
|
| 1812 |
!!!next-token; |
## Para 4 |
| 1813 |
} |
## TODO: $old_insertion_point = $current_insertion_point; |
| 1814 |
} |
## TODO: $current_insertion_point = just before $self->{nc}; |
| 1815 |
} else { |
|
| 1816 |
!!!next-token; |
## Para 5 |
| 1817 |
} |
## TODO: Run the $script->[0]. |
| 1818 |
return; |
|
| 1819 |
} elsif ($token->{tag_name} eq 'form') { |
## Para 6 |
| 1820 |
if (defined $self->{form_element}) { |
## TODO: $current_insertion_point = $old_insertion_point; |
| 1821 |
!!!parse-error (type => 'in form:form'); |
|
| 1822 |
## Ignore the token |
## Para 7 |
| 1823 |
|
## TODO: if ($pending_external_script) { |
| 1824 |
|
## TODO: ... |
| 1825 |
|
## TODO: } |
| 1826 |
|
|
| 1827 |
!!!next-token; |
!!!next-token; |
| 1828 |
return; |
next B; |
| 1829 |
} else { |
} else { |
| 1830 |
## has a p element in scope |
!!!cp ('t42'); |
| 1831 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
|
| 1832 |
if ($_->[1] eq 'p') { |
pop @{$self->{open_elements}}; |
| 1833 |
!!!back-token; |
|
| 1834 |
$token = {type => 'end tag', tag_name => 'p'}; |
$self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
|
return; |
|
|
} elsif ({ |
|
|
table => 1, caption => 1, td => 1, th => 1, |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
|
|
}->{$_->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
|
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
|
|
$self->{form_element} = $self->{open_elements}->[-1]->[0]; |
|
| 1835 |
!!!next-token; |
!!!next-token; |
| 1836 |
return; |
next B; |
| 1837 |
} |
} |
| 1838 |
} elsif ($token->{tag_name} eq 'li') { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 1839 |
## has a p element in scope |
delete $self->{ignore_newline}; |
| 1840 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
|
| 1841 |
if ($_->[1] eq 'p') { |
!!!cp ('t44'); |
| 1842 |
!!!back-token; |
!!!parse-error (type => 'not closed', |
| 1843 |
$token = {type => 'end tag', tag_name => 'p'}; |
text => $self->{open_elements}->[-1]->[0] |
| 1844 |
return; |
->manakai_local_name, |
| 1845 |
} elsif ({ |
token => $token); |
| 1846 |
table => 1, caption => 1, td => 1, th => 1, |
|
| 1847 |
button => 1, marquee => 1, object => 1, html => 1, |
#if ($self->{open_elements}->[-1]->[1] == SCRIPT_EL) { |
| 1848 |
}->{$_->[1]}) { |
# ## TODO: Mark as "already executed" |
|
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:'. |
|
|
$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
splice @{$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 'dd' or $token->{tag_name} eq 'dt') { |
|
|
## has a p element in scope |
|
|
INSCOPE: for (reverse @{$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 'dt' or $node->[1] eq 'dd') { |
|
|
if ($i != -1) { |
|
|
!!!parse-error (type => 'end tag missing:'. |
|
|
$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
splice @{$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}); |
|
|
|
|
|
$self->{content_model} = PLAINTEXT_CONTENT_MODEL; |
|
|
|
|
|
!!!next-token; |
|
|
return; |
|
|
} elsif ({ |
|
|
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## has a p element in scope |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'p') { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
|
|
return; |
|
|
} elsif ({ |
|
|
table => 1, caption => 1, td => 1, th => 1, |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
|
|
|
## NOTE: See <http://html5.org/tools/web-apps-tracker?from=925&to=926> |
|
|
## has an element in scope |
|
|
#my $i; |
|
|
#INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
# my $node = $self->{open_elements}->[$_]; |
|
|
# if ({ |
|
|
# h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
|
|
# }->{$node->[1]}) { |
|
|
# $i = $_; |
|
|
# last INSCOPE; |
|
|
# } elsif ({ |
|
|
# table => 1, caption => 1, td => 1, th => 1, |
|
|
# button => 1, marquee => 1, object => 1, html => 1, |
|
|
# }->{$node->[1]}) { |
|
|
# last INSCOPE; |
|
|
# } |
|
|
#} # INSCOPE |
|
|
# |
|
|
#if (defined $i) { |
|
|
# !!! parse-error (type => 'in hn:hn'); |
|
|
# splice @{$self->{open_elements}}, $i; |
|
| 1849 |
#} |
#} |
| 1850 |
|
|
| 1851 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
pop @{$self->{open_elements}}; |
| 1852 |
|
|
| 1853 |
|
$self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1854 |
|
## Reprocess. |
| 1855 |
|
next B; |
| 1856 |
|
} else { |
| 1857 |
|
die "$0: $token->{type}: In CDATA/RCDATA: Unknown token type"; |
| 1858 |
|
} |
| 1859 |
|
} elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM) { |
| 1860 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 1861 |
|
!!!cp ('t87.1'); |
| 1862 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 1863 |
!!!next-token; |
!!!next-token; |
| 1864 |
return; |
next B; |
| 1865 |
} elsif ($token->{tag_name} eq 'a') { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 1866 |
AFE: for my $i (reverse 0..$#$active_formatting_elements) { |
if ((not {mglyph => 1, malignmark => 1}->{$token->{tag_name}} and |
| 1867 |
my $node = $active_formatting_elements->[$i]; |
$self->{open_elements}->[-1]->[1] & FOREIGN_FLOW_CONTENT_EL) or |
| 1868 |
if ($node->[1] eq 'a') { |
not ($self->{open_elements}->[-1]->[1] & FOREIGN_EL) or |
| 1869 |
!!!parse-error (type => 'in a:a'); |
($token->{tag_name} eq 'svg' and |
| 1870 |
|
$self->{open_elements}->[-1]->[1] == MML_AXML_EL)) { |
| 1871 |
!!!back-token; |
## NOTE: "using the rules for secondary insertion mode"then"continue" |
| 1872 |
$token = {type => 'end tag', tag_name => 'a'}; |
!!!cp ('t87.2'); |
| 1873 |
$formatting_end_tag->($token->{tag_name}); |
# |
| 1874 |
|
} elsif ({ |
| 1875 |
AFE2: for (reverse 0..$#$active_formatting_elements) { |
b => 1, big => 1, blockquote => 1, body => 1, br => 1, |
| 1876 |
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
center => 1, code => 1, dd => 1, div => 1, dl => 1, dt => 1, |
| 1877 |
splice @$active_formatting_elements, $_, 1; |
em => 1, embed => 1, font => 1, h1 => 1, h2 => 1, h3 => 1, |
| 1878 |
last AFE2; |
h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, i => 1, |
| 1879 |
} |
img => 1, li => 1, listing => 1, menu => 1, meta => 1, |
| 1880 |
} # AFE2 |
nobr => 1, ol => 1, p => 1, pre => 1, ruby => 1, s => 1, |
| 1881 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
small => 1, span => 1, strong => 1, strike => 1, sub => 1, |
| 1882 |
if ($self->{open_elements}->[$_]->[0] eq $node->[0]) { |
sup => 1, table => 1, tt => 1, u => 1, ul => 1, var => 1, |
| 1883 |
splice @{$self->{open_elements}}, $_, 1; |
}->{$token->{tag_name}}) { |
| 1884 |
last OE; |
!!!cp ('t87.2'); |
| 1885 |
} |
!!!parse-error (type => 'not closed', |
| 1886 |
} # OE |
text => $self->{open_elements}->[-1]->[0] |
| 1887 |
last AFE; |
->manakai_local_name, |
| 1888 |
} elsif ($node->[0] eq '#marker') { |
token => $token); |
| 1889 |
last AFE; |
|
| 1890 |
|
pop @{$self->{open_elements}} |
| 1891 |
|
while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; |
| 1892 |
|
|
| 1893 |
|
$self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
| 1894 |
|
## Reprocess. |
| 1895 |
|
next B; |
| 1896 |
|
} else { |
| 1897 |
|
my $nsuri = $self->{open_elements}->[-1]->[0]->namespace_uri; |
| 1898 |
|
my $tag_name = $token->{tag_name}; |
| 1899 |
|
if ($nsuri eq $SVG_NS) { |
| 1900 |
|
$tag_name = { |
| 1901 |
|
altglyph => 'altGlyph', |
| 1902 |
|
altglyphdef => 'altGlyphDef', |
| 1903 |
|
altglyphitem => 'altGlyphItem', |
| 1904 |
|
animatecolor => 'animateColor', |
| 1905 |
|
animatemotion => 'animateMotion', |
| 1906 |
|
animatetransform => 'animateTransform', |
| 1907 |
|
clippath => 'clipPath', |
| 1908 |
|
feblend => 'feBlend', |
| 1909 |
|
fecolormatrix => 'feColorMatrix', |
| 1910 |
|
fecomponenttransfer => 'feComponentTransfer', |
| 1911 |
|
fecomposite => 'feComposite', |
| 1912 |
|
feconvolvematrix => 'feConvolveMatrix', |
| 1913 |
|
fediffuselighting => 'feDiffuseLighting', |
| 1914 |
|
fedisplacementmap => 'feDisplacementMap', |
| 1915 |
|
fedistantlight => 'feDistantLight', |
| 1916 |
|
feflood => 'feFlood', |
| 1917 |
|
fefunca => 'feFuncA', |
| 1918 |
|
fefuncb => 'feFuncB', |
| 1919 |
|
fefuncg => 'feFuncG', |
| 1920 |
|
fefuncr => 'feFuncR', |
| 1921 |
|
fegaussianblur => 'feGaussianBlur', |
| 1922 |
|
feimage => 'feImage', |
| 1923 |
|
femerge => 'feMerge', |
| 1924 |
|
femergenode => 'feMergeNode', |
| 1925 |
|
femorphology => 'feMorphology', |
| 1926 |
|
feoffset => 'feOffset', |
| 1927 |
|
fepointlight => 'fePointLight', |
| 1928 |
|
fespecularlighting => 'feSpecularLighting', |
| 1929 |
|
fespotlight => 'feSpotLight', |
| 1930 |
|
fetile => 'feTile', |
| 1931 |
|
feturbulence => 'feTurbulence', |
| 1932 |
|
foreignobject => 'foreignObject', |
| 1933 |
|
glyphref => 'glyphRef', |
| 1934 |
|
lineargradient => 'linearGradient', |
| 1935 |
|
radialgradient => 'radialGradient', |
| 1936 |
|
#solidcolor => 'solidColor', ## NOTE: Commented in spec (SVG1.2) |
| 1937 |
|
textpath => 'textPath', |
| 1938 |
|
}->{$tag_name} || $tag_name; |
| 1939 |
} |
} |
|
} # AFE |
|
|
|
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
| 1940 |
|
|
| 1941 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
## "adjust SVG attributes" (SVG only) - done in insert-element-f |
|
push @$active_formatting_elements, $self->{open_elements}->[-1]; |
|
| 1942 |
|
|
| 1943 |
!!!next-token; |
## "adjust foreign attributes" - done in insert-element-f |
|
return; |
|
|
} elsif ({ |
|
|
b => 1, big => 1, em => 1, font => 1, i => 1, |
|
|
s => 1, small => 1, strile => 1, |
|
|
strong => 1, tt => 1, u => 1, |
|
|
}->{$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); |
|
| 1944 |
|
|
| 1945 |
## has a |nobr| element in scope |
!!!insert-element-f ($nsuri, $tag_name, $token->{attributes}, $token); |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'nobr') { |
|
|
!!!parse-error (type => 'not closed: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', '']; |
|
| 1946 |
|
|
| 1947 |
!!!next-token; |
if ($self->{self_closing}) { |
| 1948 |
return; |
pop @{$self->{open_elements}}; |
| 1949 |
} elsif ($token->{tag_name} eq 'marquee' or |
!!!ack ('t87.3'); |
| 1950 |
$token->{tag_name} eq 'object') { |
} else { |
| 1951 |
$reconstruct_active_formatting_elements->($insert_to_current); |
!!!cp ('t87.4'); |
|
|
|
|
!!!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); |
|
|
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
|
|
return; |
|
|
} elsif ($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; |
|
| 1952 |
} |
} |
|
} # 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'; |
|
|
} |
|
| 1953 |
|
|
|
## NOTE: There is an "as if <br>" code clone. |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; |
|
|
|
|
|
!!!next-token; |
|
|
return; |
|
|
} elsif ($token->{tag_name} eq 'hr') { |
|
|
## has a p element in scope |
|
|
INSCOPE: for (reverse @{$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}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; |
|
|
|
|
|
!!!next-token; |
|
|
return; |
|
|
} elsif ($token->{tag_name} eq 'input') { |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
!!!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 |
|
| 1954 |
!!!next-token; |
!!!next-token; |
| 1955 |
return; |
next B; |
| 1956 |
} else { |
} |
| 1957 |
my $at = $token->{attributes}; |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 1958 |
my $form_attrs; |
## NOTE: "using the rules for secondary insertion mode" then "continue" |
| 1959 |
$form_attrs->{action} = $at->{action} if $at->{action}; |
!!!cp ('t87.5'); |
| 1960 |
my $prompt_attr = $at->{prompt}; |
# |
| 1961 |
$at->{name} = {name => 'name', value => 'isindex'}; |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 1962 |
delete $at->{action}; |
!!!cp ('t87.6'); |
| 1963 |
delete $at->{prompt}; |
!!!parse-error (type => 'not closed', |
| 1964 |
my @tokens = ( |
text => $self->{open_elements}->[-1]->[0] |
| 1965 |
{type => 'start tag', tag_name => 'form', |
->manakai_local_name, |
| 1966 |
attributes => $form_attrs}, |
token => $token); |
| 1967 |
{type => 'start tag', tag_name => 'hr'}, |
|
| 1968 |
{type => 'start tag', tag_name => 'p'}, |
pop @{$self->{open_elements}} |
| 1969 |
{type => 'start tag', tag_name => 'label'}, |
while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; |
| 1970 |
); |
|
| 1971 |
if ($prompt_attr) { |
## NOTE: |<span><svg>| ... two parse errors, |<svg>| ... a parse error. |
| 1972 |
push @tokens, {type => 'character', data => $prompt_attr->{value}}; |
|
| 1973 |
|
$self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
| 1974 |
|
## Reprocess. |
| 1975 |
|
next B; |
| 1976 |
|
} else { |
| 1977 |
|
die "$0: $token->{type}: Unknown token type"; |
| 1978 |
|
} |
| 1979 |
|
} |
| 1980 |
|
|
| 1981 |
|
if ($self->{insertion_mode} & HEAD_IMS) { |
| 1982 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 1983 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 1984 |
|
unless ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 1985 |
|
if ($self->{head_element_inserted}) { |
| 1986 |
|
!!!cp ('t88.3'); |
| 1987 |
|
$self->{open_elements}->[-1]->[0]->append_child |
| 1988 |
|
($self->{document}->create_text_node ($1)); |
| 1989 |
|
delete $self->{head_element_inserted}; |
| 1990 |
|
## NOTE: |</head> <link> | |
| 1991 |
|
# |
| 1992 |
|
} else { |
| 1993 |
|
!!!cp ('t88.2'); |
| 1994 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 1995 |
|
## NOTE: |</head>  | |
| 1996 |
|
# |
| 1997 |
|
} |
| 1998 |
} else { |
} else { |
| 1999 |
push @tokens, {type => 'character', |
!!!cp ('t88.1'); |
| 2000 |
data => 'This is a searchable index. Insert your search keywords here: '}; # SHOULD |
## Ignore the token. |
| 2001 |
## TODO: make this configurable |
# |
| 2002 |
} |
} |
|
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'}; |
|
|
$token = shift @tokens; |
|
|
!!!back-token (@tokens); |
|
|
return; |
|
|
} |
|
|
} elsif ($token->{tag_name} eq 'textarea') { |
|
|
my $tag_name = $token->{tag_name}; |
|
|
my $el; |
|
|
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
|
|
|
|
|
## TODO: $self->{form_element} if defined |
|
|
$self->{content_model} = RCDATA_CONTENT_MODEL; |
|
|
delete $self->{escape}; # MUST |
|
|
|
|
|
$insert->($el); |
|
|
|
|
|
my $text = ''; |
|
|
!!!next-token; |
|
|
if ($token->{type} eq 'character') { |
|
|
$token->{data} =~ s/^\x0A//; |
|
| 2003 |
unless (length $token->{data}) { |
unless (length $token->{data}) { |
| 2004 |
|
!!!cp ('t88'); |
| 2005 |
!!!next-token; |
!!!next-token; |
| 2006 |
|
next B; |
| 2007 |
} |
} |
| 2008 |
|
## TODO: set $token->{column} appropriately |
| 2009 |
} |
} |
|
while ($token->{type} eq 'character') { |
|
|
$text .= $token->{data}; |
|
|
!!!next-token; |
|
|
} |
|
|
if (length $text) { |
|
|
$el->manakai_append_text ($text); |
|
|
} |
|
|
|
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; |
|
|
|
|
|
if ($token->{type} eq 'end tag' and |
|
|
$token->{tag_name} eq $tag_name) { |
|
|
## Ignore the token |
|
|
} else { |
|
|
!!!parse-error (type => 'in RCDATA:#'.$token->{type}); |
|
|
} |
|
|
!!!next-token; |
|
|
return; |
|
|
} elsif ({ |
|
|
iframe => 1, |
|
|
noembed => 1, |
|
|
noframes => 1, |
|
|
noscript => 0, ## TODO: 1 if scripting is enabled |
|
|
}->{$token->{tag_name}}) { |
|
|
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
|
|
return; |
|
|
} elsif ($token->{tag_name} eq 'select') { |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
|
|
|
|
|
$self->{insertion_mode} = 'in select'; |
|
|
!!!next-token; |
|
|
return; |
|
|
} elsif ({ |
|
|
caption => 1, col => 1, colgroup => 1, frame => 1, |
|
|
frameset => 1, head => 1, option => 1, optgroup => 1, |
|
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$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. |
|
|
} else { |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
|
|
|
|
|
!!!next-token; |
|
|
return; |
|
|
} |
|
|
} elsif ($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, |
|
|
tbody => 1, tfoot => 1, thead => 1, |
|
|
}->{$_->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$_->[1]); |
|
|
} |
|
|
} |
|
| 2010 |
|
|
| 2011 |
$self->{insertion_mode} = 'after body'; |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2012 |
!!!next-token; |
!!!cp ('t89'); |
| 2013 |
return; |
## As if <head> |
| 2014 |
|
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2015 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2016 |
|
push @{$self->{open_elements}}, |
| 2017 |
|
[$self->{head_element}, $el_category->{head}]; |
| 2018 |
|
|
| 2019 |
|
## Reprocess in the "in head" insertion mode... |
| 2020 |
|
pop @{$self->{open_elements}}; |
| 2021 |
|
|
| 2022 |
|
## Reprocess in the "after head" insertion mode... |
| 2023 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2024 |
|
!!!cp ('t90'); |
| 2025 |
|
## As if </noscript> |
| 2026 |
|
pop @{$self->{open_elements}}; |
| 2027 |
|
!!!parse-error (type => 'in noscript:#text', token => $token); |
| 2028 |
|
|
| 2029 |
|
## Reprocess in the "in head" insertion mode... |
| 2030 |
|
## As if </head> |
| 2031 |
|
pop @{$self->{open_elements}}; |
| 2032 |
|
|
| 2033 |
|
## Reprocess in the "after head" insertion mode... |
| 2034 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2035 |
|
!!!cp ('t91'); |
| 2036 |
|
pop @{$self->{open_elements}}; |
| 2037 |
|
|
| 2038 |
|
## Reprocess in the "after head" insertion mode... |
| 2039 |
} else { |
} else { |
| 2040 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t92'); |
|
## Ignore the token |
|
|
!!!next-token; |
|
|
return; |
|
| 2041 |
} |
} |
| 2042 |
} elsif ($token->{tag_name} eq 'html') { |
|
| 2043 |
if (@{$self->{open_elements}} > 1 and $self->{open_elements}->[1]->[1] eq 'body') { |
## "after head" insertion mode |
| 2044 |
## ISSUE: There is an issue in the spec. |
## As if <body> |
| 2045 |
if ($self->{open_elements}->[-1]->[1] ne 'body') { |
!!!insert-element ('body',, $token); |
| 2046 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[1]->[1]); |
$self->{insertion_mode} = IN_BODY_IM; |
| 2047 |
|
## reprocess |
| 2048 |
|
next B; |
| 2049 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 2050 |
|
if ($token->{tag_name} eq 'head') { |
| 2051 |
|
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2052 |
|
!!!cp ('t93'); |
| 2053 |
|
!!!create-element ($self->{head_element}, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); |
| 2054 |
|
$self->{open_elements}->[-1]->[0]->append_child |
| 2055 |
|
($self->{head_element}); |
| 2056 |
|
push @{$self->{open_elements}}, |
| 2057 |
|
[$self->{head_element}, $el_category->{head}]; |
| 2058 |
|
$self->{insertion_mode} = IN_HEAD_IM; |
| 2059 |
|
!!!nack ('t93.1'); |
| 2060 |
|
!!!next-token; |
| 2061 |
|
next B; |
| 2062 |
|
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2063 |
|
!!!cp ('t93.2'); |
| 2064 |
|
!!!parse-error (type => 'after head', text => 'head', |
| 2065 |
|
token => $token); |
| 2066 |
|
## Ignore the token |
| 2067 |
|
!!!nack ('t93.3'); |
| 2068 |
|
!!!next-token; |
| 2069 |
|
next B; |
| 2070 |
|
} else { |
| 2071 |
|
!!!cp ('t95'); |
| 2072 |
|
!!!parse-error (type => 'in head:head', |
| 2073 |
|
token => $token); # or in head noscript |
| 2074 |
|
## Ignore the token |
| 2075 |
|
!!!nack ('t95.1'); |
| 2076 |
|
!!!next-token; |
| 2077 |
|
next B; |
| 2078 |
} |
} |
| 2079 |
$self->{insertion_mode} = 'after body'; |
} elsif ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2080 |
## reprocess |
!!!cp ('t96'); |
| 2081 |
return; |
## As if <head> |
| 2082 |
} else { |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2083 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2084 |
## Ignore the token |
push @{$self->{open_elements}}, |
| 2085 |
!!!next-token; |
[$self->{head_element}, $el_category->{head}]; |
| 2086 |
return; |
|
| 2087 |
} |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2088 |
} elsif ({ |
## Reprocess in the "in head" insertion mode... |
| 2089 |
address => 1, blockquote => 1, center => 1, dir => 1, |
} else { |
| 2090 |
div => 1, dl => 1, fieldset => 1, listing => 1, |
!!!cp ('t97'); |
| 2091 |
menu => 1, ol => 1, pre => 1, ul => 1, |
} |
| 2092 |
p => 1, |
|
| 2093 |
dd => 1, dt => 1, li => 1, |
if ($token->{tag_name} eq 'base') { |
| 2094 |
button => 1, marquee => 1, object => 1, |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2095 |
}->{$token->{tag_name}}) { |
!!!cp ('t98'); |
| 2096 |
## has an element in scope |
## As if </noscript> |
| 2097 |
my $i; |
pop @{$self->{open_elements}}; |
| 2098 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
!!!parse-error (type => 'in noscript', text => 'base', |
| 2099 |
my $node = $self->{open_elements}->[$_]; |
token => $token); |
| 2100 |
if ($node->[1] eq $token->{tag_name}) { |
|
| 2101 |
## generate implied end tags |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2102 |
if ({ |
## Reprocess in the "in head" insertion mode... |
| 2103 |
dd => ($token->{tag_name} ne 'dd'), |
} else { |
| 2104 |
dt => ($token->{tag_name} ne 'dt'), |
!!!cp ('t99'); |
|
li => ($token->{tag_name} ne 'li'), |
|
|
p => ($token->{tag_name} ne 'p'), |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
return; |
|
|
} |
|
|
$i = $_; |
|
|
last INSCOPE unless $token->{tag_name} eq 'p'; |
|
|
} elsif ({ |
|
|
table => 1, caption => 1, td => 1, th => 1, |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
| 2105 |
} |
} |
| 2106 |
} # INSCOPE |
|
| 2107 |
|
## NOTE: There is a "as if in head" code clone. |
| 2108 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2109 |
if (defined $i) { |
!!!cp ('t100'); |
| 2110 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'after head', |
| 2111 |
|
text => $token->{tag_name}, token => $token); |
| 2112 |
|
push @{$self->{open_elements}}, |
| 2113 |
|
[$self->{head_element}, $el_category->{head}]; |
| 2114 |
|
$self->{head_element_inserted} = 1; |
| 2115 |
} else { |
} else { |
| 2116 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t101'); |
| 2117 |
} |
} |
| 2118 |
} |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2119 |
|
pop @{$self->{open_elements}}; |
| 2120 |
if (defined $i) { |
pop @{$self->{open_elements}} # <head> |
| 2121 |
splice @{$self->{open_elements}}, $i; |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2122 |
} elsif ($token->{tag_name} eq 'p') { |
!!!nack ('t101.1'); |
| 2123 |
## As if <p>, then reprocess the current token |
!!!next-token; |
| 2124 |
my $el; |
next B; |
| 2125 |
!!!create-element ($el, 'p'); |
} elsif ($token->{tag_name} eq 'link') { |
| 2126 |
$insert->($el); |
## NOTE: There is a "as if in head" code clone. |
| 2127 |
} |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2128 |
$clear_up_to_marker->() |
!!!cp ('t102'); |
| 2129 |
if { |
!!!parse-error (type => 'after head', |
| 2130 |
button => 1, marquee => 1, object => 1, |
text => $token->{tag_name}, token => $token); |
| 2131 |
}->{$token->{tag_name}}; |
push @{$self->{open_elements}}, |
| 2132 |
!!!next-token; |
[$self->{head_element}, $el_category->{head}]; |
| 2133 |
return; |
$self->{head_element_inserted} = 1; |
| 2134 |
} elsif ($token->{tag_name} eq 'form') { |
} else { |
| 2135 |
## has an element in scope |
!!!cp ('t103'); |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$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; |
|
| 2136 |
} |
} |
| 2137 |
} # INSCOPE |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
|
|
|
|
if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) { |
|
| 2138 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2139 |
} else { |
pop @{$self->{open_elements}} # <head> |
| 2140 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2141 |
} |
!!!ack ('t103.1'); |
| 2142 |
|
!!!next-token; |
| 2143 |
|
next B; |
| 2144 |
|
} elsif ($token->{tag_name} eq 'command' or |
| 2145 |
|
$token->{tag_name} eq 'eventsource') { |
| 2146 |
|
if ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2147 |
|
## NOTE: If the insertion mode at the time of the emission |
| 2148 |
|
## of the token was "before head", $self->{insertion_mode} |
| 2149 |
|
## is already changed to |IN_HEAD_IM|. |
| 2150 |
|
|
| 2151 |
undef $self->{form_element}; |
## NOTE: There is a "as if in head" code clone. |
| 2152 |
!!!next-token; |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2153 |
return; |
pop @{$self->{open_elements}}; |
| 2154 |
} elsif ({ |
pop @{$self->{open_elements}} # <head> |
| 2155 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2156 |
}->{$token->{tag_name}}) { |
!!!ack ('t103.2'); |
| 2157 |
## has an element in scope |
!!!next-token; |
| 2158 |
my $i; |
next B; |
| 2159 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
} else { |
| 2160 |
my $node = $self->{open_elements}->[$_]; |
## NOTE: "in head noscript" or "after head" insertion mode |
| 2161 |
if ({ |
## - in these cases, these tags are treated as same as |
| 2162 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
## normal in-body tags. |
| 2163 |
}->{$node->[1]}) { |
!!!cp ('t103.3'); |
| 2164 |
## generate implied end tags |
# |
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$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; |
|
| 2165 |
} |
} |
| 2166 |
} # INSCOPE |
} elsif ($token->{tag_name} eq 'meta') { |
| 2167 |
|
## NOTE: There is a "as if in head" code clone. |
| 2168 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2169 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t104'); |
| 2170 |
} |
!!!parse-error (type => 'after head', |
| 2171 |
|
text => $token->{tag_name}, token => $token); |
| 2172 |
splice @{$self->{open_elements}}, $i if defined $i; |
push @{$self->{open_elements}}, |
| 2173 |
!!!next-token; |
[$self->{head_element}, $el_category->{head}]; |
| 2174 |
return; |
$self->{head_element_inserted} = 1; |
| 2175 |
} elsif ({ |
} else { |
| 2176 |
a => 1, |
!!!cp ('t105'); |
| 2177 |
b => 1, big => 1, em => 1, font => 1, i => 1, |
} |
| 2178 |
nobr => 1, s => 1, small => 1, strile => 1, |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2179 |
strong => 1, tt => 1, u => 1, |
my $meta_el = pop @{$self->{open_elements}}; |
|
}->{$token->{tag_name}}) { |
|
|
$formatting_end_tag->($token->{tag_name}); |
|
|
return; |
|
|
} elsif ($token->{tag_name} eq 'br') { |
|
|
!!!parse-error (type => 'unmatched end tag:br'); |
|
| 2180 |
|
|
| 2181 |
## As if <br> |
unless ($self->{confident}) { |
| 2182 |
$reconstruct_active_formatting_elements->($insert_to_current); |
if ($token->{attributes}->{charset}) { |
| 2183 |
|
!!!cp ('t106'); |
| 2184 |
my $el; |
## NOTE: Whether the encoding is supported or not is handled |
| 2185 |
!!!create-element ($el, 'br'); |
## in the {change_encoding} callback. |
| 2186 |
$insert->($el); |
$self->{change_encoding} |
| 2187 |
|
->($self, $token->{attributes}->{charset}->{value}, |
| 2188 |
## Ignore the token. |
$token); |
| 2189 |
!!!next-token; |
|
| 2190 |
return; |
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2191 |
} elsif ({ |
->set_user_data (manakai_has_reference => |
| 2192 |
caption => 1, col => 1, colgroup => 1, frame => 1, |
$token->{attributes}->{charset} |
| 2193 |
frameset => 1, head => 1, option => 1, optgroup => 1, |
->{has_reference}); |
| 2194 |
tbody => 1, td => 1, tfoot => 1, th => 1, |
} elsif ($token->{attributes}->{content}) { |
| 2195 |
thead => 1, tr => 1, |
if ($token->{attributes}->{content}->{value} |
| 2196 |
area => 1, basefont => 1, bgsound => 1, |
=~ /[Cc][Hh][Aa][Rr][Ss][Ee][Tt] |
| 2197 |
embed => 1, hr => 1, iframe => 1, image => 1, |
[\x09\x0A\x0C\x0D\x20]*= |
| 2198 |
img => 1, input => 1, isindex => 1, noembed => 1, |
[\x09\x0A\x0C\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 2199 |
noframes => 1, param => 1, select => 1, spacer => 1, |
([^"'\x09\x0A\x0C\x0D\x20] |
| 2200 |
table => 1, textarea => 1, wbr => 1, |
[^\x09\x0A\x0C\x0D\x20\x3B]*))/x) { |
| 2201 |
noscript => 0, ## TODO: if scripting is enabled |
!!!cp ('t107'); |
| 2202 |
}->{$token->{tag_name}}) { |
## NOTE: Whether the encoding is supported or not is handled |
| 2203 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
## in the {change_encoding} callback. |
| 2204 |
## Ignore the token |
$self->{change_encoding} |
| 2205 |
!!!next-token; |
->($self, defined $1 ? $1 : defined $2 ? $2 : $3, |
| 2206 |
return; |
$token); |
| 2207 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2208 |
## ISSUE: Issue on HTML5 new elements in spec |
->set_user_data (manakai_has_reference => |
| 2209 |
|
$token->{attributes}->{content} |
| 2210 |
} else { |
->{has_reference}); |
| 2211 |
## Step 1 |
} else { |
| 2212 |
my $node_i = -1; |
!!!cp ('t108'); |
| 2213 |
my $node = $self->{open_elements}->[$node_i]; |
} |
| 2214 |
|
} |
| 2215 |
|
} else { |
| 2216 |
|
if ($token->{attributes}->{charset}) { |
| 2217 |
|
!!!cp ('t109'); |
| 2218 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2219 |
|
->set_user_data (manakai_has_reference => |
| 2220 |
|
$token->{attributes}->{charset} |
| 2221 |
|
->{has_reference}); |
| 2222 |
|
} |
| 2223 |
|
if ($token->{attributes}->{content}) { |
| 2224 |
|
!!!cp ('t110'); |
| 2225 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2226 |
|
->set_user_data (manakai_has_reference => |
| 2227 |
|
$token->{attributes}->{content} |
| 2228 |
|
->{has_reference}); |
| 2229 |
|
} |
| 2230 |
|
} |
| 2231 |
|
|
| 2232 |
## Step 2 |
pop @{$self->{open_elements}} # <head> |
| 2233 |
S2: { |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2234 |
if ($node->[1] eq $token->{tag_name}) { |
!!!ack ('t110.1'); |
| 2235 |
## Step 1 |
!!!next-token; |
| 2236 |
## generate implied end tags |
next B; |
| 2237 |
if ({ |
} elsif ($token->{tag_name} eq 'title') { |
| 2238 |
dd => 1, dt => 1, li => 1, p => 1, |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2239 |
td => 1, th => 1, tr => 1, |
!!!cp ('t111'); |
| 2240 |
tbody => 1, tfoot=> 1, thead => 1, |
## As if </noscript> |
| 2241 |
}->{$self->{open_elements}->[-1]->[1]}) { |
pop @{$self->{open_elements}}; |
| 2242 |
!!!back-token; |
!!!parse-error (type => 'in noscript', text => 'title', |
| 2243 |
$token = {type => 'end tag', |
token => $token); |
| 2244 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
| 2245 |
return; |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2246 |
} |
## Reprocess in the "in head" insertion mode... |
| 2247 |
|
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2248 |
## Step 2 |
!!!cp ('t112'); |
| 2249 |
if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { |
!!!parse-error (type => 'after head', |
| 2250 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
text => $token->{tag_name}, token => $token); |
| 2251 |
} |
push @{$self->{open_elements}}, |
| 2252 |
|
[$self->{head_element}, $el_category->{head}]; |
| 2253 |
## Step 3 |
$self->{head_element_inserted} = 1; |
| 2254 |
splice @{$self->{open_elements}}, $node_i; |
} else { |
| 2255 |
|
!!!cp ('t113'); |
| 2256 |
|
} |
| 2257 |
|
|
| 2258 |
!!!next-token; |
## NOTE: There is a "as if in head" code clone. |
| 2259 |
last S2; |
$parse_rcdata->(RCDATA_CONTENT_MODEL); |
| 2260 |
|
## ISSUE: A spec bug [Bug 6038] |
| 2261 |
|
splice @{$self->{open_elements}}, -2, 1, () # <head> |
| 2262 |
|
if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2263 |
|
next B; |
| 2264 |
|
} elsif ($token->{tag_name} eq 'style' or |
| 2265 |
|
$token->{tag_name} eq 'noframes') { |
| 2266 |
|
## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
| 2267 |
|
## insertion mode IN_HEAD_IM) |
| 2268 |
|
## NOTE: There is a "as if in head" code clone. |
| 2269 |
|
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2270 |
|
!!!cp ('t114'); |
| 2271 |
|
!!!parse-error (type => 'after head', |
| 2272 |
|
text => $token->{tag_name}, token => $token); |
| 2273 |
|
push @{$self->{open_elements}}, |
| 2274 |
|
[$self->{head_element}, $el_category->{head}]; |
| 2275 |
|
$self->{head_element_inserted} = 1; |
| 2276 |
} else { |
} else { |
| 2277 |
## Step 3 |
!!!cp ('t115'); |
|
if (not $formatting_category->{$node->[1]} and |
|
|
#not $phrasing_category->{$node->[1]} and |
|
|
($special_category->{$node->[1]} or |
|
|
$scoping_category->{$node->[1]})) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
last S2; |
|
|
} |
|
| 2278 |
} |
} |
| 2279 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 2280 |
|
## ISSUE: A spec bug [Bug 6038] |
| 2281 |
|
splice @{$self->{open_elements}}, -2, 1, () # <head> |
| 2282 |
|
if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2283 |
|
next B; |
| 2284 |
|
} elsif ($token->{tag_name} eq 'noscript') { |
| 2285 |
|
if ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2286 |
|
!!!cp ('t116'); |
| 2287 |
|
## NOTE: and scripting is disalbed |
| 2288 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2289 |
|
$self->{insertion_mode} = IN_HEAD_NOSCRIPT_IM; |
| 2290 |
|
!!!nack ('t116.1'); |
| 2291 |
|
!!!next-token; |
| 2292 |
|
next B; |
| 2293 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2294 |
|
!!!cp ('t117'); |
| 2295 |
|
!!!parse-error (type => 'in noscript', text => 'noscript', |
| 2296 |
|
token => $token); |
| 2297 |
|
## Ignore the token |
| 2298 |
|
!!!nack ('t117.1'); |
| 2299 |
|
!!!next-token; |
| 2300 |
|
next B; |
| 2301 |
|
} else { |
| 2302 |
|
!!!cp ('t118'); |
| 2303 |
|
# |
| 2304 |
|
} |
| 2305 |
|
} elsif ($token->{tag_name} eq 'script') { |
| 2306 |
|
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2307 |
|
!!!cp ('t119'); |
| 2308 |
|
## As if </noscript> |
| 2309 |
|
pop @{$self->{open_elements}}; |
| 2310 |
|
!!!parse-error (type => 'in noscript', text => 'script', |
| 2311 |
|
token => $token); |
| 2312 |
|
|
| 2313 |
## Step 4 |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2314 |
$node_i--; |
## Reprocess in the "in head" insertion mode... |
| 2315 |
$node = $self->{open_elements}->[$node_i]; |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2316 |
|
!!!cp ('t120'); |
| 2317 |
## Step 5; |
!!!parse-error (type => 'after head', |
| 2318 |
redo S2; |
text => $token->{tag_name}, token => $token); |
| 2319 |
} # S2 |
push @{$self->{open_elements}}, |
| 2320 |
return; |
[$self->{head_element}, $el_category->{head}]; |
| 2321 |
} |
$self->{head_element_inserted} = 1; |
| 2322 |
} |
} else { |
| 2323 |
}; # $in_body |
!!!cp ('t121'); |
| 2324 |
|
} |
|
B: { |
|
|
if ($token->{type} eq 'DOCTYPE') { |
|
|
!!!parse-error (type => 'DOCTYPE in the middle'); |
|
|
## Ignore the token |
|
|
## Stay in the phase |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'end-of-file') { |
|
|
if ($token->{insertion_mode} ne 'trailing end') { |
|
|
## Generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 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]); |
|
|
} |
|
|
|
|
|
## ISSUE: There is an issue in the spec. |
|
|
} |
|
| 2325 |
|
|
| 2326 |
## Stop parsing |
## NOTE: There is a "as if in head" code clone. |
| 2327 |
last B; |
$script_start_tag->(); |
| 2328 |
} elsif ($token->{type} eq 'start tag' and |
## ISSUE: A spec bug [Bug 6038] |
| 2329 |
$token->{tag_name} eq 'html') { |
splice @{$self->{open_elements}}, -2, 1 # <head> |
| 2330 |
if ($self->{insertion_mode} eq 'trailing end') { |
if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2331 |
## Turn into the main phase |
next B; |
| 2332 |
!!!parse-error (type => 'after html:html'); |
} elsif ($token->{tag_name} eq 'body' or |
| 2333 |
$self->{insertion_mode} = $previous_insertion_mode; |
$token->{tag_name} eq 'frameset') { |
| 2334 |
} |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2335 |
|
!!!cp ('t122'); |
| 2336 |
|
## As if </noscript> |
| 2337 |
|
pop @{$self->{open_elements}}; |
| 2338 |
|
!!!parse-error (type => 'in noscript', |
| 2339 |
|
text => $token->{tag_name}, token => $token); |
| 2340 |
|
|
| 2341 |
|
## Reprocess in the "in head" insertion mode... |
| 2342 |
|
## As if </head> |
| 2343 |
|
pop @{$self->{open_elements}}; |
| 2344 |
|
|
| 2345 |
|
## Reprocess in the "after head" insertion mode... |
| 2346 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2347 |
|
!!!cp ('t124'); |
| 2348 |
|
pop @{$self->{open_elements}}; |
| 2349 |
|
|
| 2350 |
|
## Reprocess in the "after head" insertion mode... |
| 2351 |
|
} else { |
| 2352 |
|
!!!cp ('t125'); |
| 2353 |
|
} |
| 2354 |
|
|
| 2355 |
## ISSUE: "aa<html>" is not a parse error. |
## "after head" insertion mode |
| 2356 |
## ISSUE: "<html>" in fragment is not a parse error. |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2357 |
unless ($token->{first_start_tag}) { |
if ($token->{tag_name} eq 'body') { |
| 2358 |
!!!parse-error (type => 'not first start tag'); |
!!!cp ('t126'); |
| 2359 |
} |
$self->{insertion_mode} = IN_BODY_IM; |
| 2360 |
my $top_el = $self->{open_elements}->[0]->[0]; |
} elsif ($token->{tag_name} eq 'frameset') { |
| 2361 |
for my $attr_name (keys %{$token->{attributes}}) { |
!!!cp ('t127'); |
| 2362 |
unless ($top_el->has_attribute_ns (undef, $attr_name)) { |
$self->{insertion_mode} = IN_FRAMESET_IM; |
| 2363 |
$top_el->set_attribute_ns |
} else { |
| 2364 |
(undef, [undef, $attr_name], |
die "$0: tag name: $self->{tag_name}"; |
|
$token->{attributes}->{$attr_name}->{value}); |
|
|
} |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
if ($self->{insertion_mode} eq 'trailing end') { |
|
|
$self->{document}->append_child ($comment); |
|
|
} elsif ($self->{insertion_mode} eq 'after body') { |
|
|
$self->{open_elements}->[0]->[0]->append_child ($comment); |
|
|
} else { |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'before head') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
| 2365 |
} |
} |
| 2366 |
} |
!!!nack ('t127.1'); |
|
## As if <head> |
|
|
!!!create-element ($self->{head_element}, 'head'); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
my $attr = $token->{tag_name} eq 'head' ? $token->{attributes} : {}; |
|
|
!!!create-element ($self->{head_element}, 'head', $attr); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
if ($token->{tag_name} eq 'head') { |
|
| 2367 |
!!!next-token; |
!!!next-token; |
| 2368 |
#} elsif ({ |
next B; |
|
# base => 1, link => 1, meta => 1, |
|
|
# script => 1, style => 1, title => 1, |
|
|
# }->{$token->{tag_name}}) { |
|
|
# ## reprocess |
|
| 2369 |
} else { |
} else { |
| 2370 |
## reprocess |
!!!cp ('t128'); |
| 2371 |
|
# |
| 2372 |
} |
} |
| 2373 |
redo B; |
|
| 2374 |
} elsif ($token->{type} eq 'end tag') { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2375 |
if ({ |
!!!cp ('t129'); |
| 2376 |
head => 1, body => 1, html => 1, |
## As if </noscript> |
| 2377 |
p => 1, br => 1, |
pop @{$self->{open_elements}}; |
| 2378 |
}->{$token->{tag_name}}) { |
!!!parse-error (type => 'in noscript:/', |
| 2379 |
## As if <head> |
text => $token->{tag_name}, token => $token); |
| 2380 |
!!!create-element ($self->{head_element}, 'head'); |
|
| 2381 |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
## Reprocess in the "in head" insertion mode... |
| 2382 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
## As if </head> |
| 2383 |
$self->{insertion_mode} = 'in head'; |
pop @{$self->{open_elements}}; |
| 2384 |
## reprocess |
|
| 2385 |
redo B; |
## Reprocess in the "after head" insertion mode... |
| 2386 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2387 |
|
!!!cp ('t130'); |
| 2388 |
|
## As if </head> |
| 2389 |
|
pop @{$self->{open_elements}}; |
| 2390 |
|
|
| 2391 |
|
## Reprocess in the "after head" insertion mode... |
| 2392 |
} else { |
} else { |
| 2393 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t131'); |
|
## Ignore the token ## ISSUE: An issue in the spec. |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} else { |
|
|
die "$0: $token->{type}: Unknown type"; |
|
|
} |
|
|
} elsif ($self->{insertion_mode} eq 'in head' or |
|
|
$self->{insertion_mode} eq 'in head noscript' or |
|
|
$self->{insertion_mode} eq 'after head') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 2394 |
} |
} |
|
|
|
|
# |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ({base => ($self->{insertion_mode} eq 'in head' or |
|
|
$self->{insertion_mode} eq 'after head'), |
|
|
link => 1}->{$token->{tag_name}}) { |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'meta') { |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
|
| 2395 |
|
|
| 2396 |
unless ($self->{confident}) { |
## "after head" insertion mode |
| 2397 |
my $charset; |
## As if <body> |
| 2398 |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
!!!insert-element ('body',, $token); |
| 2399 |
$charset = $token->{attributes}->{charset}->{value}; |
$self->{insertion_mode} = IN_BODY_IM; |
| 2400 |
} |
## reprocess |
| 2401 |
if ($token->{attributes}->{'http-equiv'}) { |
!!!ack-later; |
| 2402 |
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
next B; |
| 2403 |
if ($token->{attributes}->{'http-equiv'}->{value} |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 2404 |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
if ($token->{tag_name} eq 'head') { |
| 2405 |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2406 |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
!!!cp ('t132'); |
| 2407 |
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
## As if <head> |
| 2408 |
} ## TODO: And if supported |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2409 |
} |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2410 |
## TODO: Change the encoding |
push @{$self->{open_elements}}, |
| 2411 |
} |
[$self->{head_element}, $el_category->{head}]; |
| 2412 |
|
|
| 2413 |
## TODO: Extracting |charset| from |meta|. |
## Reprocess in the "in head" insertion mode... |
| 2414 |
pop @{$self->{open_elements}} |
pop @{$self->{open_elements}}; |
| 2415 |
if $self->{insertion_mode} eq 'after head'; |
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2416 |
!!!next-token; |
!!!next-token; |
| 2417 |
redo B; |
next B; |
| 2418 |
} elsif ($token->{tag_name} eq 'title' and |
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2419 |
$self->{insertion_mode} eq 'in head') { |
!!!cp ('t133'); |
| 2420 |
## NOTE: There is a "as if in head" code clone. |
## As if </noscript> |
| 2421 |
if ($self->{insertion_mode} eq 'after head') { |
pop @{$self->{open_elements}}; |
| 2422 |
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
!!!parse-error (type => 'in noscript:/', |
| 2423 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
text => 'head', token => $token); |
| 2424 |
} |
|
| 2425 |
my $parent = defined $self->{head_element} ? $self->{head_element} |
## Reprocess in the "in head" insertion mode... |
| 2426 |
: $self->{open_elements}->[-1]->[0]; |
pop @{$self->{open_elements}}; |
| 2427 |
$parse_rcdata->(RCDATA_CONTENT_MODEL, |
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2428 |
sub { $parent->append_child ($_[0]) }); |
!!!next-token; |
| 2429 |
pop @{$self->{open_elements}} |
next B; |
| 2430 |
if $self->{insertion_mode} eq 'after head'; |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2431 |
redo B; |
!!!cp ('t134'); |
| 2432 |
} elsif ($token->{tag_name} eq 'style') { |
pop @{$self->{open_elements}}; |
| 2433 |
## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2434 |
## insertion mode 'in head') |
!!!next-token; |
| 2435 |
## NOTE: There is a "as if in head" code clone. |
next B; |
| 2436 |
if ($self->{insertion_mode} eq 'after head') { |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2437 |
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
!!!cp ('t134.1'); |
| 2438 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
!!!parse-error (type => 'unmatched end tag', text => 'head', |
| 2439 |
|
token => $token); |
| 2440 |
|
## Ignore the token |
| 2441 |
|
!!!next-token; |
| 2442 |
|
next B; |
| 2443 |
|
} else { |
| 2444 |
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2445 |
} |
} |
|
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert_to_current); |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
redo B; |
|
| 2446 |
} elsif ($token->{tag_name} eq 'noscript') { |
} elsif ($token->{tag_name} eq 'noscript') { |
| 2447 |
if ($self->{insertion_mode} eq 'in head') { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2448 |
## NOTE: and scripting is disalbed |
!!!cp ('t136'); |
| 2449 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
pop @{$self->{open_elements}}; |
| 2450 |
$self->{insertion_mode} = 'in head noscript'; |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2451 |
!!!next-token; |
!!!next-token; |
| 2452 |
redo B; |
next B; |
| 2453 |
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
} elsif ($self->{insertion_mode} == BEFORE_HEAD_IM or |
| 2454 |
!!!parse-error (type => 'in noscript:noscript'); |
$self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2455 |
## Ignore the token |
!!!cp ('t137'); |
| 2456 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2457 |
|
text => 'noscript', token => $token); |
| 2458 |
|
## Ignore the token ## ISSUE: An issue in the spec. |
| 2459 |
!!!next-token; |
!!!next-token; |
| 2460 |
redo B; |
next B; |
| 2461 |
} else { |
} else { |
| 2462 |
|
!!!cp ('t138'); |
| 2463 |
# |
# |
| 2464 |
} |
} |
| 2465 |
} elsif ($token->{tag_name} eq 'head' and |
} elsif ({ |
| 2466 |
$self->{insertion_mode} ne 'after head') { |
body => 1, html => 1, |
| 2467 |
!!!parse-error (type => 'in head:head'); # or in head noscript |
}->{$token->{tag_name}}) { |
| 2468 |
|
## TODO: This branch is entirely redundant. |
| 2469 |
|
if ($self->{insertion_mode} == BEFORE_HEAD_IM or |
| 2470 |
|
$self->{insertion_mode} == IN_HEAD_IM or |
| 2471 |
|
$self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2472 |
|
!!!cp ('t140'); |
| 2473 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2474 |
|
text => $token->{tag_name}, token => $token); |
| 2475 |
|
## Ignore the token |
| 2476 |
|
!!!next-token; |
| 2477 |
|
next B; |
| 2478 |
|
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2479 |
|
!!!cp ('t140.1'); |
| 2480 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2481 |
|
text => $token->{tag_name}, token => $token); |
| 2482 |
|
## Ignore the token |
| 2483 |
|
!!!next-token; |
| 2484 |
|
next B; |
| 2485 |
|
} else { |
| 2486 |
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2487 |
|
} |
| 2488 |
|
} elsif ($token->{tag_name} eq 'p') { |
| 2489 |
|
!!!cp ('t142'); |
| 2490 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2491 |
|
text => $token->{tag_name}, token => $token); |
| 2492 |
## Ignore the token |
## Ignore the token |
| 2493 |
!!!next-token; |
!!!next-token; |
| 2494 |
redo B; |
next B; |
| 2495 |
} elsif ($self->{insertion_mode} ne 'in head noscript' and |
} elsif ($token->{tag_name} eq 'br') { |
| 2496 |
$token->{tag_name} eq 'script') { |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2497 |
if ($self->{insertion_mode} eq 'after head') { |
!!!cp ('t142.2'); |
| 2498 |
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
## (before head) as if <head>, (in head) as if </head> |
| 2499 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2500 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2501 |
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2502 |
|
|
| 2503 |
|
## Reprocess in the "after head" insertion mode... |
| 2504 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2505 |
|
!!!cp ('t143.2'); |
| 2506 |
|
## As if </head> |
| 2507 |
|
pop @{$self->{open_elements}}; |
| 2508 |
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2509 |
|
|
| 2510 |
|
## Reprocess in the "after head" insertion mode... |
| 2511 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2512 |
|
!!!cp ('t143.3'); |
| 2513 |
|
## NOTE: Two parse errors for <head><noscript></br> |
| 2514 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2515 |
|
text => 'br', token => $token); |
| 2516 |
|
## As if </noscript> |
| 2517 |
|
pop @{$self->{open_elements}}; |
| 2518 |
|
$self->{insertion_mode} = IN_HEAD_IM; |
| 2519 |
|
|
| 2520 |
|
## Reprocess in the "in head" insertion mode... |
| 2521 |
|
## As if </head> |
| 2522 |
|
pop @{$self->{open_elements}}; |
| 2523 |
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2524 |
|
|
| 2525 |
|
## Reprocess in the "after head" insertion mode... |
| 2526 |
|
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2527 |
|
!!!cp ('t143.4'); |
| 2528 |
|
# |
| 2529 |
|
} else { |
| 2530 |
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2531 |
} |
} |
| 2532 |
## NOTE: There is a "as if in head" code clone. |
|
| 2533 |
$script_start_tag->($insert_to_current); |
## ISSUE: does not agree with IE7 - it doesn't ignore </br>. |
| 2534 |
pop @{$self->{open_elements}} |
!!!parse-error (type => 'unmatched end tag', |
| 2535 |
if $self->{insertion_mode} eq 'after head'; |
text => 'br', token => $token); |
| 2536 |
redo B; |
## Ignore the token |
|
} elsif ($self->{insertion_mode} eq 'after head' and |
|
|
$token->{tag_name} eq 'body') { |
|
|
!!!insert-element ('body', $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in body'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'after head' and |
|
|
$token->{tag_name} eq 'frameset') { |
|
|
!!!insert-element ('frameset', $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in frameset'; |
|
| 2537 |
!!!next-token; |
!!!next-token; |
| 2538 |
redo B; |
next B; |
| 2539 |
} else { |
} else { |
| 2540 |
# |
!!!cp ('t145'); |
| 2541 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2542 |
|
text => $token->{tag_name}, token => $token); |
| 2543 |
|
## Ignore the token |
| 2544 |
|
!!!next-token; |
| 2545 |
|
next B; |
| 2546 |
} |
} |
| 2547 |
} elsif ($token->{type} eq 'end tag') { |
|
| 2548 |
if ($self->{insertion_mode} eq 'in head' and |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2549 |
$token->{tag_name} eq 'head') { |
!!!cp ('t146'); |
| 2550 |
|
## As if </noscript> |
| 2551 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2552 |
$self->{insertion_mode} = 'after head'; |
!!!parse-error (type => 'in noscript:/', |
| 2553 |
!!!next-token; |
text => $token->{tag_name}, token => $token); |
| 2554 |
redo B; |
|
| 2555 |
} elsif ($self->{insertion_mode} eq 'in head noscript' and |
## Reprocess in the "in head" insertion mode... |
| 2556 |
$token->{tag_name} eq 'noscript') { |
## As if </head> |
| 2557 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2558 |
$self->{insertion_mode} = 'in head'; |
|
| 2559 |
!!!next-token; |
## Reprocess in the "after head" insertion mode... |
| 2560 |
redo B; |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2561 |
} elsif ($self->{insertion_mode} eq 'in head' and |
!!!cp ('t147'); |
| 2562 |
{ |
## As if </head> |
| 2563 |
body => 1, html => 1, |
pop @{$self->{open_elements}}; |
| 2564 |
p => 1, br => 1, |
|
| 2565 |
}->{$token->{tag_name}}) { |
## Reprocess in the "after head" insertion mode... |
| 2566 |
# |
} elsif ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2567 |
} elsif ($self->{insertion_mode} eq 'in head noscript' and |
## ISSUE: This case cannot be reached? |
| 2568 |
{ |
!!!cp ('t148'); |
| 2569 |
p => 1, br => 1, |
!!!parse-error (type => 'unmatched end tag', |
| 2570 |
}->{$token->{tag_name}}) { |
text => $token->{tag_name}, token => $token); |
| 2571 |
# |
## Ignore the token ## ISSUE: An issue in the spec. |
|
} elsif ($self->{insertion_mode} ne 'after head') { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
| 2572 |
!!!next-token; |
!!!next-token; |
| 2573 |
redo B; |
next B; |
| 2574 |
} else { |
} else { |
| 2575 |
# |
!!!cp ('t149'); |
| 2576 |
} |
} |
|
} else { |
|
|
# |
|
|
} |
|
| 2577 |
|
|
| 2578 |
## As if </head> or </noscript> or <body> |
## "after head" insertion mode |
| 2579 |
if ($self->{insertion_mode} eq 'in head') { |
## As if <body> |
| 2580 |
pop @{$self->{open_elements}}; |
!!!insert-element ('body',, $token); |
| 2581 |
$self->{insertion_mode} = 'after head'; |
$self->{insertion_mode} = IN_BODY_IM; |
| 2582 |
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
## reprocess |
| 2583 |
pop @{$self->{open_elements}}; |
next B; |
| 2584 |
!!!parse-error (type => 'in noscript:'.(defined $token->{tag_name} ? ($token->{type} eq 'end tag' ? '/' : '') . $token->{tag_name} : '#' . $token->{type})); |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 2585 |
$self->{insertion_mode} = 'in head'; |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2586 |
} else { # 'after head' |
!!!cp ('t149.1'); |
| 2587 |
!!!insert-element ('body'); |
|
| 2588 |
$self->{insertion_mode} = 'in body'; |
## NOTE: As if <head> |
| 2589 |
} |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2590 |
## reprocess |
$self->{open_elements}->[-1]->[0]->append_child |
| 2591 |
redo B; |
($self->{head_element}); |
| 2592 |
|
#push @{$self->{open_elements}}, |
| 2593 |
|
# [$self->{head_element}, $el_category->{head}]; |
| 2594 |
|
#$self->{insertion_mode} = IN_HEAD_IM; |
| 2595 |
|
## NOTE: Reprocess. |
| 2596 |
|
|
| 2597 |
|
## NOTE: As if </head> |
| 2598 |
|
#pop @{$self->{open_elements}}; |
| 2599 |
|
#$self->{insertion_mode} = IN_AFTER_HEAD_IM; |
| 2600 |
|
## NOTE: Reprocess. |
| 2601 |
|
|
| 2602 |
|
# |
| 2603 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2604 |
|
!!!cp ('t149.2'); |
| 2605 |
|
|
| 2606 |
|
## NOTE: As if </head> |
| 2607 |
|
pop @{$self->{open_elements}}; |
| 2608 |
|
#$self->{insertion_mode} = IN_AFTER_HEAD_IM; |
| 2609 |
|
## NOTE: Reprocess. |
| 2610 |
|
|
| 2611 |
|
# |
| 2612 |
|
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2613 |
|
!!!cp ('t149.3'); |
| 2614 |
|
|
| 2615 |
|
!!!parse-error (type => 'in noscript:#eof', token => $token); |
| 2616 |
|
|
| 2617 |
## ISSUE: An issue in the spec. |
## As if </noscript> |
| 2618 |
} elsif ($self->{insertion_mode} eq 'in body' or |
pop @{$self->{open_elements}}; |
| 2619 |
$self->{insertion_mode} eq 'in cell' or |
#$self->{insertion_mode} = IN_HEAD_IM; |
| 2620 |
$self->{insertion_mode} eq 'in caption') { |
## NOTE: Reprocess. |
| 2621 |
if ($token->{type} eq 'character') { |
|
| 2622 |
|
## NOTE: As if </head> |
| 2623 |
|
pop @{$self->{open_elements}}; |
| 2624 |
|
#$self->{insertion_mode} = IN_AFTER_HEAD_IM; |
| 2625 |
|
## NOTE: Reprocess. |
| 2626 |
|
|
| 2627 |
|
# |
| 2628 |
|
} else { |
| 2629 |
|
!!!cp ('t149.4'); |
| 2630 |
|
# |
| 2631 |
|
} |
| 2632 |
|
|
| 2633 |
|
## NOTE: As if <body> |
| 2634 |
|
!!!insert-element ('body',, $token); |
| 2635 |
|
$self->{insertion_mode} = IN_BODY_IM; |
| 2636 |
|
## NOTE: Reprocess. |
| 2637 |
|
next B; |
| 2638 |
|
} else { |
| 2639 |
|
die "$0: $token->{type}: Unknown token type"; |
| 2640 |
|
} |
| 2641 |
|
} elsif ($self->{insertion_mode} & BODY_IMS) { |
| 2642 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 2643 |
|
!!!cp ('t150'); |
| 2644 |
## NOTE: There is a code clone of "character in body". |
## NOTE: There is a code clone of "character in body". |
| 2645 |
$reconstruct_active_formatting_elements->($insert_to_current); |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2646 |
|
|
| 2647 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 2648 |
|
|
| 2649 |
!!!next-token; |
!!!next-token; |
| 2650 |
redo B; |
next B; |
| 2651 |
} elsif ($token->{type} eq 'start tag') { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 2652 |
if ({ |
if ({ |
| 2653 |
caption => 1, col => 1, colgroup => 1, tbody => 1, |
caption => 1, col => 1, colgroup => 1, tbody => 1, |
| 2654 |
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
| 2655 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 2656 |
if ($self->{insertion_mode} eq 'in cell') { |
if (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2657 |
## have an element in table scope |
## have an element in table scope |
| 2658 |
my $tn; |
for (reverse 0..$#{$self->{open_elements}}) { |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
| 2659 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2660 |
if ($node->[1] eq 'td' or $node->[1] eq 'th') { |
if ($node->[1] == TABLE_CELL_EL) { |
| 2661 |
$tn = $node->[1]; |
!!!cp ('t151'); |
| 2662 |
last INSCOPE; |
|
| 2663 |
} elsif ({ |
## Close the cell |
| 2664 |
table => 1, html => 1, |
!!!back-token; # <x> |
| 2665 |
}->{$node->[1]}) { |
$token = {type => END_TAG_TOKEN, |
| 2666 |
last INSCOPE; |
tag_name => $node->[0]->manakai_local_name, |
| 2667 |
} |
line => $token->{line}, |
| 2668 |
} # INSCOPE |
column => $token->{column}}; |
| 2669 |
unless (defined $tn) { |
next B; |
| 2670 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2671 |
## Ignore the token |
!!!cp ('t152'); |
| 2672 |
!!!next-token; |
## ISSUE: This case can never be reached, maybe. |
| 2673 |
redo B; |
last; |
| 2674 |
} |
} |
| 2675 |
|
} |
| 2676 |
|
|
| 2677 |
|
!!!cp ('t153'); |
| 2678 |
|
!!!parse-error (type => 'start tag not allowed', |
| 2679 |
|
text => $token->{tag_name}, token => $token); |
| 2680 |
|
## Ignore the token |
| 2681 |
|
!!!nack ('t153.1'); |
| 2682 |
|
!!!next-token; |
| 2683 |
|
next B; |
| 2684 |
|
} elsif (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2685 |
|
!!!parse-error (type => 'not closed', text => 'caption', |
| 2686 |
|
token => $token); |
| 2687 |
|
|
| 2688 |
## Close the cell |
## NOTE: As if </caption>. |
|
!!!back-token; # <?> |
|
|
$token = {type => 'end tag', tag_name => $tn}; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in caption') { |
|
|
!!!parse-error (type => 'not closed:caption'); |
|
|
|
|
|
## As if </caption> |
|
| 2689 |
## have a table element in table scope |
## have a table element in table scope |
| 2690 |
my $i; |
my $i; |
| 2691 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: { |
| 2692 |
my $node = $self->{open_elements}->[$_]; |
for (reverse 0..$#{$self->{open_elements}}) { |
| 2693 |
if ($node->[1] eq 'caption') { |
my $node = $self->{open_elements}->[$_]; |
| 2694 |
$i = $_; |
if ($node->[1] == CAPTION_EL) { |
| 2695 |
last INSCOPE; |
!!!cp ('t155'); |
| 2696 |
} elsif ({ |
$i = $_; |
| 2697 |
table => 1, html => 1, |
last INSCOPE; |
| 2698 |
}->{$node->[1]}) { |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2699 |
last INSCOPE; |
!!!cp ('t156'); |
| 2700 |
|
last; |
| 2701 |
|
} |
| 2702 |
} |
} |
| 2703 |
|
|
| 2704 |
|
!!!cp ('t157'); |
| 2705 |
|
!!!parse-error (type => 'start tag not allowed', |
| 2706 |
|
text => $token->{tag_name}, token => $token); |
| 2707 |
|
## Ignore the token |
| 2708 |
|
!!!nack ('t157.1'); |
| 2709 |
|
!!!next-token; |
| 2710 |
|
next B; |
| 2711 |
} # INSCOPE |
} # INSCOPE |
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:caption'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 2712 |
|
|
| 2713 |
## generate implied end tags |
## generate implied end tags |
| 2714 |
if ({ |
while ($self->{open_elements}->[-1]->[1] |
| 2715 |
dd => 1, dt => 1, li => 1, p => 1, |
& END_TAG_OPTIONAL_EL) { |
| 2716 |
td => 1, th => 1, tr => 1, |
!!!cp ('t158'); |
| 2717 |
tbody => 1, tfoot=> 1, thead => 1, |
pop @{$self->{open_elements}}; |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <?> |
|
|
$token = {type => 'end tag', tag_name => 'caption'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
| 2718 |
} |
} |
| 2719 |
|
|
| 2720 |
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2721 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t159'); |
| 2722 |
|
!!!parse-error (type => 'not closed', |
| 2723 |
|
text => $self->{open_elements}->[-1]->[0] |
| 2724 |
|
->manakai_local_name, |
| 2725 |
|
token => $token); |
| 2726 |
|
} else { |
| 2727 |
|
!!!cp ('t160'); |
| 2728 |
} |
} |
| 2729 |
|
|
| 2730 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 2731 |
|
|
| 2732 |
$clear_up_to_marker->(); |
$clear_up_to_marker->(); |
| 2733 |
|
|
| 2734 |
$self->{insertion_mode} = 'in table'; |
$self->{insertion_mode} = IN_TABLE_IM; |
| 2735 |
|
|
| 2736 |
## reprocess |
## reprocess |
| 2737 |
redo B; |
!!!ack-later; |
| 2738 |
|
next B; |
| 2739 |
} else { |
} else { |
| 2740 |
|
!!!cp ('t161'); |
| 2741 |
# |
# |
| 2742 |
} |
} |
| 2743 |
} else { |
} else { |
| 2744 |
|
!!!cp ('t162'); |
| 2745 |
# |
# |
| 2746 |
} |
} |
| 2747 |
} elsif ($token->{type} eq 'end tag') { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 2748 |
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
| 2749 |
if ($self->{insertion_mode} eq 'in cell') { |
if (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2750 |
## have an element in table scope |
## have an element in table scope |
| 2751 |
my $i; |
my $i; |
| 2752 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2753 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2754 |
if ($node->[1] eq $token->{tag_name}) { |
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 2755 |
|
!!!cp ('t163'); |
| 2756 |
$i = $_; |
$i = $_; |
| 2757 |
last INSCOPE; |
last INSCOPE; |
| 2758 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2759 |
table => 1, html => 1, |
!!!cp ('t164'); |
|
}->{$node->[1]}) { |
|
| 2760 |
last INSCOPE; |
last INSCOPE; |
| 2761 |
} |
} |
| 2762 |
} # INSCOPE |
} # INSCOPE |
| 2763 |
unless (defined $i) { |
unless (defined $i) { |
| 2764 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t165'); |
| 2765 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2766 |
|
text => $token->{tag_name}, |
| 2767 |
|
token => $token); |
| 2768 |
## Ignore the token |
## Ignore the token |
| 2769 |
!!!next-token; |
!!!next-token; |
| 2770 |
redo B; |
next B; |
| 2771 |
} |
} |
| 2772 |
|
|
| 2773 |
## generate implied end tags |
## generate implied end tags |
| 2774 |
if ({ |
while ($self->{open_elements}->[-1]->[1] |
| 2775 |
dd => 1, dt => 1, li => 1, p => 1, |
& END_TAG_OPTIONAL_EL) { |
| 2776 |
td => ($token->{tag_name} eq 'th'), |
!!!cp ('t166'); |
| 2777 |
th => ($token->{tag_name} eq 'td'), |
pop @{$self->{open_elements}}; |
|
tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
| 2778 |
} |
} |
| 2779 |
|
|
| 2780 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
| 2781 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
ne $token->{tag_name}) { |
| 2782 |
|
!!!cp ('t167'); |
| 2783 |
|
!!!parse-error (type => 'not closed', |
| 2784 |
|
text => $self->{open_elements}->[-1]->[0] |
| 2785 |
|
->manakai_local_name, |
| 2786 |
|
token => $token); |
| 2787 |
|
} else { |
| 2788 |
|
!!!cp ('t168'); |
| 2789 |
} |
} |
| 2790 |
|
|
| 2791 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 2792 |
|
|
| 2793 |
$clear_up_to_marker->(); |
$clear_up_to_marker->(); |
| 2794 |
|
|
| 2795 |
$self->{insertion_mode} = 'in row'; |
$self->{insertion_mode} = IN_ROW_IM; |
| 2796 |
|
|
| 2797 |
!!!next-token; |
!!!next-token; |
| 2798 |
redo B; |
next B; |
| 2799 |
} elsif ($self->{insertion_mode} eq 'in caption') { |
} elsif (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2800 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t169'); |
| 2801 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2802 |
|
text => $token->{tag_name}, token => $token); |
| 2803 |
## Ignore the token |
## Ignore the token |
| 2804 |
!!!next-token; |
!!!next-token; |
| 2805 |
redo B; |
next B; |
| 2806 |
} else { |
} else { |
| 2807 |
|
!!!cp ('t170'); |
| 2808 |
# |
# |
| 2809 |
} |
} |
| 2810 |
} elsif ($token->{tag_name} eq 'caption') { |
} elsif ($token->{tag_name} eq 'caption') { |
| 2811 |
if ($self->{insertion_mode} eq 'in caption') { |
if (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2812 |
## have a table element in table scope |
## have a table element in table scope |
| 2813 |
my $i; |
my $i; |
| 2814 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: { |
| 2815 |
my $node = $self->{open_elements}->[$_]; |
for (reverse 0..$#{$self->{open_elements}}) { |
| 2816 |
if ($node->[1] eq $token->{tag_name}) { |
my $node = $self->{open_elements}->[$_]; |
| 2817 |
$i = $_; |
if ($node->[1] == CAPTION_EL) { |
| 2818 |
last INSCOPE; |
!!!cp ('t171'); |
| 2819 |
} elsif ({ |
$i = $_; |
| 2820 |
table => 1, html => 1, |
last INSCOPE; |
| 2821 |
}->{$node->[1]}) { |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2822 |
last INSCOPE; |
!!!cp ('t172'); |
| 2823 |
|
last; |
| 2824 |
|
} |
| 2825 |
} |
} |
| 2826 |
|
|
| 2827 |
|
!!!cp ('t173'); |
| 2828 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2829 |
|
text => $token->{tag_name}, token => $token); |
| 2830 |
|
## Ignore the token |
| 2831 |
|
!!!next-token; |
| 2832 |
|
next B; |
| 2833 |
} # INSCOPE |
} # INSCOPE |
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 2834 |
|
|
| 2835 |
## generate implied end tags |
## generate implied end tags |
| 2836 |
if ({ |
while ($self->{open_elements}->[-1]->[1] |
| 2837 |
dd => 1, dt => 1, li => 1, p => 1, |
& END_TAG_OPTIONAL_EL) { |
| 2838 |
td => 1, th => 1, tr => 1, |
!!!cp ('t174'); |
| 2839 |
tbody => 1, tfoot=> 1, thead => 1, |
pop @{$self->{open_elements}}; |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
| 2840 |
} |
} |
| 2841 |
|
|
| 2842 |
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2843 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t175'); |
| 2844 |
|
!!!parse-error (type => 'not closed', |
| 2845 |
|
text => $self->{open_elements}->[-1]->[0] |
| 2846 |
|
->manakai_local_name, |
| 2847 |
|
token => $token); |
| 2848 |
|
} else { |
| 2849 |
|
!!!cp ('t176'); |
| 2850 |
} |
} |
| 2851 |
|
|
| 2852 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 2853 |
|
|
| 2854 |
$clear_up_to_marker->(); |
$clear_up_to_marker->(); |
| 2855 |
|
|
| 2856 |
$self->{insertion_mode} = 'in table'; |
$self->{insertion_mode} = IN_TABLE_IM; |
| 2857 |
|
|
| 2858 |
!!!next-token; |
!!!next-token; |
| 2859 |
redo B; |
next B; |
| 2860 |
} elsif ($self->{insertion_mode} eq 'in cell') { |
} elsif (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2861 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t177'); |
| 2862 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2863 |
|
text => $token->{tag_name}, token => $token); |
| 2864 |
## Ignore the token |
## Ignore the token |
| 2865 |
!!!next-token; |
!!!next-token; |
| 2866 |
redo B; |
next B; |
| 2867 |
} else { |
} else { |
| 2868 |
|
!!!cp ('t178'); |
| 2869 |
# |
# |
| 2870 |
} |
} |
| 2871 |
} elsif ({ |
} elsif ({ |
| 2872 |
table => 1, tbody => 1, tfoot => 1, |
table => 1, tbody => 1, tfoot => 1, |
| 2873 |
thead => 1, tr => 1, |
thead => 1, tr => 1, |
| 2874 |
}->{$token->{tag_name}} and |
}->{$token->{tag_name}} and |
| 2875 |
$self->{insertion_mode} eq 'in cell') { |
($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2876 |
## have an element in table scope |
## have an element in table scope |
| 2877 |
my $i; |
my $i; |
| 2878 |
my $tn; |
my $tn; |
| 2879 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: { |
| 2880 |
my $node = $self->{open_elements}->[$_]; |
for (reverse 0..$#{$self->{open_elements}}) { |
| 2881 |
if ($node->[1] eq $token->{tag_name}) { |
my $node = $self->{open_elements}->[$_]; |
| 2882 |
$i = $_; |
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 2883 |
last INSCOPE; |
!!!cp ('t179'); |
| 2884 |
} elsif ($node->[1] eq 'td' or $node->[1] eq 'th') { |
$i = $_; |
| 2885 |
$tn = $node->[1]; |
|
| 2886 |
## NOTE: There is exactly one |td| or |th| element |
## Close the cell |
| 2887 |
## in scope in the stack of open elements by definition. |
!!!back-token; # </x> |
| 2888 |
} elsif ({ |
$token = {type => END_TAG_TOKEN, tag_name => $tn, |
| 2889 |
table => 1, html => 1, |
line => $token->{line}, |
| 2890 |
}->{$node->[1]}) { |
column => $token->{column}}; |
| 2891 |
last INSCOPE; |
next B; |
| 2892 |
|
} elsif ($node->[1] == TABLE_CELL_EL) { |
| 2893 |
|
!!!cp ('t180'); |
| 2894 |
|
$tn = $node->[0]->manakai_local_name; |
| 2895 |
|
## NOTE: There is exactly one |td| or |th| element |
| 2896 |
|
## in scope in the stack of open elements by definition. |
| 2897 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2898 |
|
## ISSUE: Can this be reached? |
| 2899 |
|
!!!cp ('t181'); |
| 2900 |
|
last; |
| 2901 |
|
} |
| 2902 |
} |
} |
| 2903 |
} # INSCOPE |
|
| 2904 |
unless (defined $i) { |
!!!cp ('t182'); |
| 2905 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!parse-error (type => 'unmatched end tag', |
| 2906 |
|
text => $token->{tag_name}, token => $token); |
| 2907 |
## Ignore the token |
## Ignore the token |
| 2908 |
!!!next-token; |
!!!next-token; |
| 2909 |
redo B; |
next B; |
| 2910 |
} |
} # INSCOPE |
|
|
|
|
## Close the cell |
|
|
!!!back-token; # </?> |
|
|
$token = {type => 'end tag', tag_name => $tn}; |
|
|
redo B; |
|
| 2911 |
} elsif ($token->{tag_name} eq 'table' and |
} elsif ($token->{tag_name} eq 'table' and |
| 2912 |
$self->{insertion_mode} eq 'in caption') { |
($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2913 |
!!!parse-error (type => 'not closed:caption'); |
!!!parse-error (type => 'not closed', text => 'caption', |
| 2914 |
|
token => $token); |
| 2915 |
|
|
| 2916 |
## As if </caption> |
## As if </caption> |
| 2917 |
## have a table element in table scope |
## have a table element in table scope |
| 2918 |
my $i; |
my $i; |
| 2919 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2920 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2921 |
if ($node->[1] eq 'caption') { |
if ($node->[1] == CAPTION_EL) { |
| 2922 |
|
!!!cp ('t184'); |
| 2923 |
$i = $_; |
$i = $_; |
| 2924 |
last INSCOPE; |
last INSCOPE; |
| 2925 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 2926 |
table => 1, html => 1, |
!!!cp ('t185'); |
|
}->{$node->[1]}) { |
|
| 2927 |
last INSCOPE; |
last INSCOPE; |
| 2928 |
} |
} |
| 2929 |
} # INSCOPE |
} # INSCOPE |
| 2930 |
unless (defined $i) { |
unless (defined $i) { |
| 2931 |
!!!parse-error (type => 'unmatched end tag:caption'); |
!!!cp ('t186'); |
| 2932 |
|
## TODO: Wrong error type? |
| 2933 |
|
!!!parse-error (type => 'unmatched end tag', |
| 2934 |
|
text => 'caption', token => $token); |
| 2935 |
## Ignore the token |
## Ignore the token |
| 2936 |
!!!next-token; |
!!!next-token; |
| 2937 |
redo B; |
next B; |
| 2938 |
} |
} |
| 2939 |
|
|
| 2940 |
## generate implied end tags |
## generate implied end tags |
| 2941 |
if ({ |
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 2942 |
dd => 1, dt => 1, li => 1, p => 1, |
!!!cp ('t187'); |
| 2943 |
td => 1, th => 1, tr => 1, |
pop @{$self->{open_elements}}; |
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # </table> |
|
|
$token = {type => 'end tag', tag_name => 'caption'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
| 2944 |
} |
} |
| 2945 |
|
|
| 2946 |
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2947 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t188'); |
| 2948 |
|
!!!parse-error (type => 'not closed', |
| 2949 |
|
text => $self->{open_elements}->[-1]->[0] |
| 2950 |
|
->manakai_local_name, |
| 2951 |
|
token => $token); |
| 2952 |
|
} else { |
| 2953 |
|
!!!cp ('t189'); |
| 2954 |
} |
} |
| 2955 |
|
|
| 2956 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 2957 |
|
|
| 2958 |
$clear_up_to_marker->(); |
$clear_up_to_marker->(); |
| 2959 |
|
|
| 2960 |
$self->{insertion_mode} = 'in table'; |
$self->{insertion_mode} = IN_TABLE_IM; |
| 2961 |
|
|
| 2962 |
## reprocess |
## reprocess |
| 2963 |
redo B; |
next B; |
| 2964 |
} elsif ({ |
} elsif ({ |
| 2965 |
body => 1, col => 1, colgroup => 1, html => 1, |
body => 1, col => 1, colgroup => 1, html => 1, |
| 2966 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 2967 |
if ($self->{insertion_mode} eq 'in cell' or |
if ($self->{insertion_mode} & BODY_TABLE_IMS) { |
| 2968 |
$self->{insertion_mode} eq 'in caption') { |
!!!cp ('t190'); |
| 2969 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!parse-error (type => 'unmatched end tag', |
| 2970 |
|
text => $token->{tag_name}, token => $token); |
| 2971 |
## Ignore the token |
## Ignore the token |
| 2972 |
!!!next-token; |
!!!next-token; |
| 2973 |
redo B; |
next B; |
| 2974 |
} else { |
} else { |
| 2975 |
|
!!!cp ('t191'); |
| 2976 |
# |
# |
| 2977 |
} |
} |
| 2978 |
} elsif ({ |
} elsif ({ |
| 2979 |
tbody => 1, tfoot => 1, |
tbody => 1, tfoot => 1, |
| 2980 |
thead => 1, tr => 1, |
thead => 1, tr => 1, |
| 2981 |
}->{$token->{tag_name}} and |
}->{$token->{tag_name}} and |
| 2982 |
$self->{insertion_mode} eq 'in caption') { |
($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2983 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t192'); |
| 2984 |
## Ignore the token |
!!!parse-error (type => 'unmatched end tag', |
| 2985 |
!!!next-token; |
text => $token->{tag_name}, token => $token); |
| 2986 |
redo B; |
## Ignore the token |
| 2987 |
} else { |
!!!next-token; |
| 2988 |
# |
next B; |
| 2989 |
} |
} else { |
| 2990 |
} else { |
!!!cp ('t193'); |
| 2991 |
# |
# |
| 2992 |
|
} |
| 2993 |
|
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 2994 |
|
for my $entry (@{$self->{open_elements}}) { |
| 2995 |
|
unless ($entry->[1] & ALL_END_TAG_OPTIONAL_EL) { |
| 2996 |
|
!!!cp ('t75'); |
| 2997 |
|
!!!parse-error (type => 'in body:#eof', token => $token); |
| 2998 |
|
last; |
| 2999 |
} |
} |
| 3000 |
|
} |
| 3001 |
$in_body->($insert_to_current); |
|
| 3002 |
redo B; |
## Stop parsing. |
| 3003 |
} elsif ($self->{insertion_mode} eq 'in table') { |
last B; |
| 3004 |
if ($token->{type} eq 'character') { |
} else { |
| 3005 |
## NOTE: There are "character in table" code clones. |
die "$0: $token->{type}: Unknown token type"; |
| 3006 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
} |
| 3007 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
| 3008 |
|
$insert = $insert_to_current; |
| 3009 |
|
# |
| 3010 |
|
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
| 3011 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 3012 |
|
if (not $open_tables->[-1]->[1] and # tainted |
| 3013 |
|
$token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 3014 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3015 |
|
|
| 3016 |
unless (length $token->{data}) { |
unless (length $token->{data}) { |
| 3017 |
!!!next-token; |
!!!cp ('t194'); |
| 3018 |
redo B; |
!!!next-token; |
| 3019 |
} |
next B; |
| 3020 |
} |
} else { |
| 3021 |
|
!!!cp ('t195'); |
| 3022 |
|
} |
| 3023 |
|
} |
| 3024 |
|
|
| 3025 |
!!!parse-error (type => 'in table:#character'); |
!!!parse-error (type => 'in table:#text', token => $token); |
| 3026 |
|
|
| 3027 |
## As if in body, but insert into foster parent element |
## NOTE: As if in body, but insert into the foster parent element. |
| 3028 |
## ISSUE: Spec says that "whenever a node would be inserted |
$reconstruct_active_formatting_elements->($insert_to_foster); |
|
## into the current node" while characters might not be |
|
|
## result in a new Text node. |
|
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
|
| 3029 |
|
|
| 3030 |
if ({ |
if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 3031 |
table => 1, tbody => 1, tfoot => 1, |
# MUST |
| 3032 |
thead => 1, tr => 1, |
my $foster_parent_element; |
| 3033 |
}->{$self->{open_elements}->[-1]->[1]}) { |
my $next_sibling; |
| 3034 |
# MUST |
my $prev_sibling; |
| 3035 |
my $foster_parent_element; |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3036 |
my $next_sibling; |
if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 3037 |
my $prev_sibling; |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 3038 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
if (defined $parent and $parent->node_type == 1) { |
| 3039 |
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
$foster_parent_element = $parent; |
| 3040 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
!!!cp ('t196'); |
| 3041 |
if (defined $parent and $parent->node_type == 1) { |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 3042 |
$foster_parent_element = $parent; |
$prev_sibling = $next_sibling->previous_sibling; |
| 3043 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
# |
|
$prev_sibling = $next_sibling->previous_sibling; |
|
|
} else { |
|
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
|
|
$prev_sibling = $foster_parent_element->last_child; |
|
|
} |
|
|
last OE; |
|
|
} |
|
|
} # 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}); |
|
| 3044 |
} else { |
} else { |
| 3045 |
$foster_parent_element->insert_before |
!!!cp ('t197'); |
| 3046 |
($self->{document}->create_text_node ($token->{data}), |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 3047 |
$next_sibling); |
$prev_sibling = $foster_parent_element->last_child; |
| 3048 |
|
# |
| 3049 |
} |
} |
| 3050 |
} else { |
last OE; |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
| 3051 |
} |
} |
| 3052 |
|
} # OE |
| 3053 |
|
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
| 3054 |
|
$prev_sibling = $foster_parent_element->last_child |
| 3055 |
|
unless defined $foster_parent_element; |
| 3056 |
|
undef $prev_sibling unless $open_tables->[-1]->[2]; # ~node inserted |
| 3057 |
|
if (defined $prev_sibling and |
| 3058 |
|
$prev_sibling->node_type == 3) { |
| 3059 |
|
!!!cp ('t198'); |
| 3060 |
|
$prev_sibling->manakai_append_text ($token->{data}); |
| 3061 |
|
} else { |
| 3062 |
|
!!!cp ('t199'); |
| 3063 |
|
$foster_parent_element->insert_before |
| 3064 |
|
($self->{document}->create_text_node ($token->{data}), |
| 3065 |
|
$next_sibling); |
| 3066 |
|
} |
| 3067 |
|
$open_tables->[-1]->[1] = 1; # tainted |
| 3068 |
|
$open_tables->[-1]->[2] = 1; # ~node inserted |
| 3069 |
|
} else { |
| 3070 |
|
## NOTE: Fragment case or in a foster parent'ed element |
| 3071 |
|
## (e.g. |<table><span>a|). In fragment case, whether the |
| 3072 |
|
## character is appended to existing node or a new node is |
| 3073 |
|
## created is irrelevant, since the foster parent'ed nodes |
| 3074 |
|
## are discarded and fragment parsing does not invoke any |
| 3075 |
|
## script. |
| 3076 |
|
!!!cp ('t200'); |
| 3077 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text |
| 3078 |
|
($token->{data}); |
| 3079 |
|
} |
| 3080 |
|
|
| 3081 |
!!!next-token; |
!!!next-token; |
| 3082 |
redo B; |
next B; |
| 3083 |
} elsif ($token->{type} eq 'start tag') { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 3084 |
if ({ |
if ({ |
| 3085 |
caption => 1, |
tr => (($self->{insertion_mode} & IM_MASK) != IN_ROW_IM), |
| 3086 |
colgroup => 1, |
th => 1, td => 1, |
| 3087 |
tbody => 1, tfoot => 1, thead => 1, |
}->{$token->{tag_name}}) { |
| 3088 |
}->{$token->{tag_name}}) { |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_IM) { |
| 3089 |
## Clear back to table context |
## Clear back to table context |
| 3090 |
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
while (not ($self->{open_elements}->[-1]->[1] |
| 3091 |
$self->{open_elements}->[-1]->[1] ne 'html') { |
& TABLE_SCOPING_EL)) { |
| 3092 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t201'); |
| 3093 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
|
} |
|
|
|
|
|
push @$active_formatting_elements, ['#marker', ''] |
|
|
if $token->{tag_name} eq 'caption'; |
|
|
|
|
|
!!!insert-element ($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}}; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
col => 1, |
|
|
td => 1, th => 1, tr => 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 @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
!!!insert-element ($token->{tag_name} eq 'col' ? 'colgroup' : 'tbody'); |
|
|
$self->{insertion_mode} = $token->{tag_name} eq 'col' |
|
|
? 'in column group' : 'in table body'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($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]); |
|
|
|
|
|
## As if </table> |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'table') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:table'); |
|
|
## Ignore tokens </table><table> |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <table> |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'table') { |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, colgroup => 1, |
|
|
html => 1, tbody => 1, td => 1, tfoot => 1, th => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
| 3094 |
} |
} |
| 3095 |
} else { |
|
| 3096 |
# |
!!!insert-element ('tbody',, $token); |
| 3097 |
|
$self->{insertion_mode} = IN_TABLE_BODY_IM; |
| 3098 |
|
## reprocess in the "in table body" insertion mode... |
| 3099 |
} |
} |
| 3100 |
|
|
| 3101 |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { |
| 3102 |
$in_body->($insert_to_foster); |
unless ($token->{tag_name} eq 'tr') { |
| 3103 |
redo B; |
!!!cp ('t202'); |
| 3104 |
} elsif ($self->{insertion_mode} eq 'in column group') { |
!!!parse-error (type => 'missing start tag:tr', token => $token); |
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 3105 |
} |
} |
| 3106 |
|
|
| 3107 |
# |
## Clear back to table body context |
| 3108 |
} elsif ($token->{type} eq 'start tag') { |
while (not ($self->{open_elements}->[-1]->[1] |
| 3109 |
if ($token->{tag_name} eq 'col') { |
& TABLE_ROWS_SCOPING_EL)) { |
| 3110 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!cp ('t203'); |
| 3111 |
|
## ISSUE: Can this case be reached? |
| 3112 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
| 3113 |
} |
} |
| 3114 |
} elsif ($token->{type} eq 'end tag') { |
|
| 3115 |
if ($token->{tag_name} eq 'colgroup') { |
$self->{insertion_mode} = IN_ROW_IM; |
| 3116 |
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
if ($token->{tag_name} eq 'tr') { |
| 3117 |
!!!parse-error (type => 'unmatched end tag:colgroup'); |
!!!cp ('t204'); |
| 3118 |
## Ignore the token |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3119 |
!!!next-token; |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3120 |
redo B; |
!!!nack ('t204'); |
|
} else { |
|
|
pop @{$self->{open_elements}}; # colgroup |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} elsif ($token->{tag_name} eq 'col') { |
|
|
!!!parse-error (type => 'unmatched end tag:col'); |
|
|
## Ignore the token |
|
| 3121 |
!!!next-token; |
!!!next-token; |
| 3122 |
redo B; |
next B; |
| 3123 |
} else { |
} else { |
| 3124 |
# |
!!!cp ('t205'); |
| 3125 |
|
!!!insert-element ('tr',, $token); |
| 3126 |
|
## reprocess in the "in row" insertion mode |
| 3127 |
} |
} |
| 3128 |
} else { |
} else { |
| 3129 |
# |
!!!cp ('t206'); |
| 3130 |
} |
} |
| 3131 |
|
|
| 3132 |
## As if </colgroup> |
## Clear back to table row context |
| 3133 |
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
while (not ($self->{open_elements}->[-1]->[1] |
| 3134 |
!!!parse-error (type => 'unmatched end tag:colgroup'); |
& TABLE_ROW_SCOPING_EL)) { |
| 3135 |
## Ignore the token |
!!!cp ('t207'); |
| 3136 |
!!!next-token; |
pop @{$self->{open_elements}}; |
|
redo B; |
|
|
} else { |
|
|
pop @{$self->{open_elements}}; # colgroup |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} |
|
|
} elsif ($self->{insertion_mode} eq 'in table body') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: This is a "character in table" code clone. |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
| 3137 |
} |
} |
| 3138 |
} |
|
| 3139 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3140 |
!!!parse-error (type => 'in table:#character'); |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3141 |
|
$self->{insertion_mode} = IN_CELL_IM; |
|
## As if in body, but insert into foster parent element |
|
|
## ISSUE: Spec says that "whenever a node would be inserted |
|
|
## into the current node" while characters might not be |
|
|
## result in a new Text node. |
|
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
|
| 3142 |
|
|
| 3143 |
if ({ |
push @$active_formatting_elements, ['#marker', '']; |
| 3144 |
table => 1, tbody => 1, tfoot => 1, |
|
| 3145 |
thead => 1, tr => 1, |
!!!nack ('t207.1'); |
| 3146 |
}->{$self->{open_elements}->[-1]->[1]}) { |
!!!next-token; |
| 3147 |
# MUST |
next B; |
| 3148 |
my $foster_parent_element; |
} elsif ({ |
| 3149 |
my $next_sibling; |
caption => 1, col => 1, colgroup => 1, |
| 3150 |
my $prev_sibling; |
tbody => 1, tfoot => 1, thead => 1, |
| 3151 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
tr => 1, # $self->{insertion_mode} == IN_ROW_IM |
| 3152 |
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
}->{$token->{tag_name}}) { |
| 3153 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3154 |
if (defined $parent and $parent->node_type == 1) { |
## As if </tr> |
| 3155 |
$foster_parent_element = $parent; |
## have an element in table scope |
| 3156 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
my $i; |
| 3157 |
$prev_sibling = $next_sibling->previous_sibling; |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3158 |
} else { |
my $node = $self->{open_elements}->[$_]; |
| 3159 |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
if ($node->[1] == TABLE_ROW_EL) { |
| 3160 |
$prev_sibling = $foster_parent_element->last_child; |
!!!cp ('t208'); |
| 3161 |
} |
$i = $_; |
| 3162 |
last OE; |
last INSCOPE; |
| 3163 |
} |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3164 |
} # OE |
!!!cp ('t209'); |
| 3165 |
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
last INSCOPE; |
| 3166 |
$prev_sibling = $foster_parent_element->last_child |
} |
| 3167 |
unless defined $foster_parent_element; |
} # INSCOPE |
| 3168 |
if (defined $prev_sibling and |
unless (defined $i) { |
| 3169 |
$prev_sibling->node_type == 3) { |
!!!cp ('t210'); |
| 3170 |
$prev_sibling->manakai_append_text ($token->{data}); |
## TODO: This type is wrong. |
| 3171 |
} else { |
!!!parse-error (type => 'unmacthed end tag', |
| 3172 |
$foster_parent_element->insert_before |
text => $token->{tag_name}, token => $token); |
| 3173 |
($self->{document}->create_text_node ($token->{data}), |
## Ignore the token |
| 3174 |
$next_sibling); |
!!!nack ('t210.1'); |
| 3175 |
} |
!!!next-token; |
| 3176 |
} else { |
next B; |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
| 3177 |
} |
} |
| 3178 |
|
|
| 3179 |
!!!next-token; |
## Clear back to table row context |
| 3180 |
redo B; |
while (not ($self->{open_elements}->[-1]->[1] |
| 3181 |
} elsif ($token->{type} eq 'start tag') { |
& TABLE_ROW_SCOPING_EL)) { |
| 3182 |
if ({ |
!!!cp ('t211'); |
| 3183 |
tr => 1, |
## ISSUE: Can this case be reached? |
| 3184 |
th => 1, td => 1, |
pop @{$self->{open_elements}}; |
| 3185 |
}->{$token->{tag_name}}) { |
} |
| 3186 |
unless ($token->{tag_name} eq 'tr') { |
|
| 3187 |
!!!parse-error (type => 'missing start tag:tr'); |
pop @{$self->{open_elements}}; # tr |
| 3188 |
|
$self->{insertion_mode} = IN_TABLE_BODY_IM; |
| 3189 |
|
if ($token->{tag_name} eq 'tr') { |
| 3190 |
|
!!!cp ('t212'); |
| 3191 |
|
## reprocess |
| 3192 |
|
!!!ack-later; |
| 3193 |
|
next B; |
| 3194 |
|
} else { |
| 3195 |
|
!!!cp ('t213'); |
| 3196 |
|
## reprocess in the "in table body" insertion mode... |
| 3197 |
|
} |
| 3198 |
} |
} |
| 3199 |
|
|
| 3200 |
## Clear back to table body context |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { |
| 3201 |
while (not { |
## have an element in table scope |
| 3202 |
tbody => 1, tfoot => 1, thead => 1, html => 1, |
my $i; |
| 3203 |
}->{$self->{open_elements}->[-1]->[1]}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3204 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
my $node = $self->{open_elements}->[$_]; |
| 3205 |
pop @{$self->{open_elements}}; |
if ($node->[1] == TABLE_ROW_GROUP_EL) { |
| 3206 |
} |
!!!cp ('t214'); |
| 3207 |
|
$i = $_; |
| 3208 |
$self->{insertion_mode} = 'in row'; |
last INSCOPE; |
| 3209 |
if ($token->{tag_name} eq 'tr') { |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3210 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!cp ('t215'); |
| 3211 |
!!!next-token; |
last INSCOPE; |
| 3212 |
} else { |
} |
| 3213 |
!!!insert-element ('tr'); |
} # INSCOPE |
| 3214 |
## reprocess |
unless (defined $i) { |
| 3215 |
} |
!!!cp ('t216'); |
| 3216 |
redo B; |
## TODO: This erorr type is wrong. |
| 3217 |
} elsif ({ |
!!!parse-error (type => 'unmatched end tag', |
| 3218 |
caption => 1, col => 1, colgroup => 1, |
text => $token->{tag_name}, token => $token); |
| 3219 |
tbody => 1, tfoot => 1, thead => 1, |
## Ignore the token |
| 3220 |
}->{$token->{tag_name}}) { |
!!!nack ('t216.1'); |
| 3221 |
## have an element in table scope |
!!!next-token; |
| 3222 |
my $i; |
next B; |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ({ |
|
|
tbody => 1, thead => 1, tfoot => 1, |
|
|
}->{$node->[1]}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
| 3223 |
} |
} |
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 3224 |
|
|
| 3225 |
## Clear back to table body context |
## Clear back to table body context |
| 3226 |
while (not { |
while (not ($self->{open_elements}->[-1]->[1] |
| 3227 |
tbody => 1, tfoot => 1, thead => 1, html => 1, |
& TABLE_ROWS_SCOPING_EL)) { |
| 3228 |
}->{$self->{open_elements}->[-1]->[1]}) { |
!!!cp ('t217'); |
| 3229 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
## ISSUE: Can this state be reached? |
| 3230 |
|
pop @{$self->{open_elements}}; |
| 3231 |
|
} |
| 3232 |
|
|
| 3233 |
|
## As if <{current node}> |
| 3234 |
|
## have an element in table scope |
| 3235 |
|
## true by definition |
| 3236 |
|
|
| 3237 |
|
## Clear back to table body context |
| 3238 |
|
## nop by definition |
| 3239 |
|
|
| 3240 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3241 |
|
$self->{insertion_mode} = IN_TABLE_IM; |
| 3242 |
|
## reprocess in "in table" insertion mode... |
| 3243 |
|
} else { |
| 3244 |
|
!!!cp ('t218'); |
| 3245 |
} |
} |
| 3246 |
|
|
| 3247 |
## As if <{current node}> |
if ($token->{tag_name} eq 'col') { |
| 3248 |
## have an element in table scope |
## Clear back to table context |
| 3249 |
## true by definition |
while (not ($self->{open_elements}->[-1]->[1] |
| 3250 |
|
& TABLE_SCOPING_EL)) { |
| 3251 |
## Clear back to table body context |
!!!cp ('t219'); |
| 3252 |
## nop by definition |
## ISSUE: Can this state be reached? |
|
|
|
| 3253 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3254 |
$self->{insertion_mode} = 'in table'; |
} |
| 3255 |
## reprocess |
|
| 3256 |
redo B; |
!!!insert-element ('colgroup',, $token); |
| 3257 |
|
$self->{insertion_mode} = IN_COLUMN_GROUP_IM; |
| 3258 |
|
## reprocess |
| 3259 |
|
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3260 |
|
!!!ack-later; |
| 3261 |
|
next B; |
| 3262 |
|
} elsif ({ |
| 3263 |
|
caption => 1, |
| 3264 |
|
colgroup => 1, |
| 3265 |
|
tbody => 1, tfoot => 1, thead => 1, |
| 3266 |
|
}->{$token->{tag_name}}) { |
| 3267 |
|
## Clear back to table context |
| 3268 |
|
while (not ($self->{open_elements}->[-1]->[1] |
| 3269 |
|
& TABLE_SCOPING_EL)) { |
| 3270 |
|
!!!cp ('t220'); |
| 3271 |
|
## ISSUE: Can this state be reached? |
| 3272 |
|
pop @{$self->{open_elements}}; |
| 3273 |
|
} |
| 3274 |
|
|
| 3275 |
|
push @$active_formatting_elements, ['#marker', ''] |
| 3276 |
|
if $token->{tag_name} eq 'caption'; |
| 3277 |
|
|
| 3278 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3279 |
|
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3280 |
|
$self->{insertion_mode} = { |
| 3281 |
|
caption => IN_CAPTION_IM, |
| 3282 |
|
colgroup => IN_COLUMN_GROUP_IM, |
| 3283 |
|
tbody => IN_TABLE_BODY_IM, |
| 3284 |
|
tfoot => IN_TABLE_BODY_IM, |
| 3285 |
|
thead => IN_TABLE_BODY_IM, |
| 3286 |
|
}->{$token->{tag_name}}; |
| 3287 |
|
!!!next-token; |
| 3288 |
|
!!!nack ('t220.1'); |
| 3289 |
|
next B; |
| 3290 |
|
} else { |
| 3291 |
|
die "$0: in table: <>: $token->{tag_name}"; |
| 3292 |
|
} |
| 3293 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
| 3294 |
## NOTE: This is a code clone of "table in table" |
!!!parse-error (type => 'not closed', |
| 3295 |
!!!parse-error (type => 'not closed:table'); |
text => $self->{open_elements}->[-1]->[0] |
| 3296 |
|
->manakai_local_name, |
| 3297 |
|
token => $token); |
| 3298 |
|
|
| 3299 |
## As if </table> |
## As if </table> |
| 3300 |
## have a table element in table scope |
## have a table element in table scope |
| 3301 |
my $i; |
my $i; |
| 3302 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3303 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3304 |
if ($node->[1] eq 'table') { |
if ($node->[1] == TABLE_EL) { |
| 3305 |
|
!!!cp ('t221'); |
| 3306 |
$i = $_; |
$i = $_; |
| 3307 |
last INSCOPE; |
last INSCOPE; |
| 3308 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3309 |
table => 1, html => 1, |
!!!cp ('t222'); |
|
}->{$node->[1]}) { |
|
| 3310 |
last INSCOPE; |
last INSCOPE; |
| 3311 |
} |
} |
| 3312 |
} # INSCOPE |
} # INSCOPE |
| 3313 |
unless (defined $i) { |
unless (defined $i) { |
| 3314 |
!!!parse-error (type => 'unmatched end tag:table'); |
!!!cp ('t223'); |
| 3315 |
|
## TODO: The following is wrong, maybe. |
| 3316 |
|
!!!parse-error (type => 'unmatched end tag', text => 'table', |
| 3317 |
|
token => $token); |
| 3318 |
## Ignore tokens </table><table> |
## Ignore tokens </table><table> |
| 3319 |
|
!!!nack ('t223.1'); |
| 3320 |
!!!next-token; |
!!!next-token; |
| 3321 |
redo B; |
next B; |
| 3322 |
} |
} |
| 3323 |
|
|
| 3324 |
|
## TODO: Followings are removed from the latest spec. |
| 3325 |
## generate implied end tags |
## generate implied end tags |
| 3326 |
if ({ |
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 3327 |
dd => 1, dt => 1, li => 1, p => 1, |
!!!cp ('t224'); |
| 3328 |
td => 1, th => 1, tr => 1, |
pop @{$self->{open_elements}}; |
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <table> |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
| 3329 |
} |
} |
| 3330 |
|
|
| 3331 |
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
unless ($self->{open_elements}->[-1]->[1] == TABLE_EL) { |
| 3332 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!cp ('t225'); |
| 3333 |
|
## NOTE: |<table><tr><table>| |
| 3334 |
|
!!!parse-error (type => 'not closed', |
| 3335 |
|
text => $self->{open_elements}->[-1]->[0] |
| 3336 |
|
->manakai_local_name, |
| 3337 |
|
token => $token); |
| 3338 |
|
} else { |
| 3339 |
|
!!!cp ('t226'); |
| 3340 |
} |
} |
| 3341 |
|
|
| 3342 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 3343 |
|
pop @{$open_tables}; |
| 3344 |
|
|
| 3345 |
$self->_reset_insertion_mode; |
$self->_reset_insertion_mode; |
| 3346 |
|
|
| 3347 |
## reprocess |
## reprocess |
| 3348 |
redo B; |
!!!ack-later; |
| 3349 |
} else { |
next B; |
| 3350 |
# |
} elsif ($token->{tag_name} eq 'style') { |
| 3351 |
} |
if (not $open_tables->[-1]->[1]) { # tainted |
| 3352 |
} elsif ($token->{type} eq 'end tag') { |
!!!cp ('t227.8'); |
| 3353 |
if ({ |
## NOTE: This is a "as if in head" code clone. |
| 3354 |
tbody => 1, tfoot => 1, thead => 1, |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 3355 |
}->{$token->{tag_name}}) { |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3356 |
## have an element in table scope |
next B; |
| 3357 |
my $i; |
} else { |
| 3358 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
!!!cp ('t227.7'); |
| 3359 |
my $node = $self->{open_elements}->[$_]; |
# |
| 3360 |
if ($node->[1] eq $token->{tag_name}) { |
} |
| 3361 |
$i = $_; |
} elsif ($token->{tag_name} eq 'script') { |
| 3362 |
last INSCOPE; |
if (not $open_tables->[-1]->[1]) { # tainted |
| 3363 |
} elsif ({ |
!!!cp ('t227.6'); |
| 3364 |
table => 1, html => 1, |
## NOTE: This is a "as if in head" code clone. |
| 3365 |
}->{$node->[1]}) { |
$script_start_tag->(); |
| 3366 |
last INSCOPE; |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3367 |
} |
next B; |
| 3368 |
} # INSCOPE |
} else { |
| 3369 |
unless (defined $i) { |
!!!cp ('t227.5'); |
| 3370 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
# |
| 3371 |
## Ignore the token |
} |
| 3372 |
!!!next-token; |
} elsif ($token->{tag_name} eq 'input') { |
| 3373 |
redo B; |
if (not $open_tables->[-1]->[1]) { # tainted |
| 3374 |
} |
if ($token->{attributes}->{type}) { ## TODO: case |
| 3375 |
|
my $type = lc $token->{attributes}->{type}->{value}; |
| 3376 |
|
if ($type eq 'hidden') { |
| 3377 |
|
!!!cp ('t227.3'); |
| 3378 |
|
!!!parse-error (type => 'in table', |
| 3379 |
|
text => $token->{tag_name}, token => $token); |
| 3380 |
|
|
| 3381 |
## Clear back to table body context |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3382 |
while (not { |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
| 3383 |
|
|
| 3384 |
pop @{$self->{open_elements}}; |
## TODO: form element pointer |
|
$self->{insertion_mode} = 'in table'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ({ |
|
|
tbody => 1, thead => 1, tfoot => 1, |
|
|
}->{$node->[1]}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 3385 |
|
|
|
## Clear back to table body context |
|
|
while (not { |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
| 3386 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
|
} |
|
| 3387 |
|
|
| 3388 |
## As if <{current node}> |
!!!next-token; |
| 3389 |
## have an element in table scope |
!!!ack ('t227.2.1'); |
| 3390 |
## true by definition |
next B; |
| 3391 |
|
} else { |
| 3392 |
## Clear back to table body context |
!!!cp ('t227.2'); |
| 3393 |
## nop by definition |
# |
| 3394 |
|
} |
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, colgroup => 1, |
|
|
html => 1, td => 1, th => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
| 3395 |
} else { |
} else { |
| 3396 |
|
!!!cp ('t227.1'); |
| 3397 |
# |
# |
| 3398 |
} |
} |
| 3399 |
} else { |
} else { |
| 3400 |
|
!!!cp ('t227.4'); |
| 3401 |
# |
# |
| 3402 |
} |
} |
| 3403 |
|
} else { |
| 3404 |
## As if in table |
!!!cp ('t227'); |
| 3405 |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
# |
| 3406 |
$in_body->($insert_to_foster); |
} |
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in row') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: This is a "character in table" code clone. |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
| 3407 |
|
|
| 3408 |
!!!parse-error (type => 'in table:#character'); |
!!!parse-error (type => 'in table', text => $token->{tag_name}, |
| 3409 |
|
token => $token); |
| 3410 |
|
|
| 3411 |
## As if in body, but insert into foster parent element |
$insert = $insert_to_foster; |
| 3412 |
## ISSUE: Spec says that "whenever a node would be inserted |
# |
| 3413 |
## into the current node" while characters might not be |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3414 |
## result in a new Text node. |
if ($token->{tag_name} eq 'tr' and |
| 3415 |
$reconstruct_active_formatting_elements->($insert_to_foster); |
($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3416 |
|
## have an element in table scope |
|
if ({ |
|
|
table => 1, tbody => 1, tfoot => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
# MUST |
|
|
my $foster_parent_element; |
|
|
my $next_sibling; |
|
|
my $prev_sibling; |
|
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
|
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
|
|
if (defined $parent and $parent->node_type == 1) { |
|
|
$foster_parent_element = $parent; |
|
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
|
|
} else { |
|
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
|
|
$prev_sibling = $foster_parent_element->last_child; |
|
|
} |
|
|
last OE; |
|
|
} |
|
|
} # 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}); |
|
|
} else { |
|
|
$foster_parent_element->insert_before |
|
|
($self->{document}->create_text_node ($token->{data}), |
|
|
$next_sibling); |
|
|
} |
|
|
} else { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
} |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ($token->{tag_name} eq 'th' or |
|
|
$token->{tag_name} eq 'td') { |
|
|
## Clear back to table row context |
|
|
while (not { |
|
|
tr => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in cell'; |
|
|
|
|
|
push @$active_formatting_elements, ['#marker', '']; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
caption => 1, col => 1, colgroup => 1, |
|
|
tbody => 1, tfoot => 1, thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## As if </tr> |
|
|
## have an element in table scope |
|
| 3417 |
my $i; |
my $i; |
| 3418 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3419 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3420 |
if ($node->[1] eq 'tr') { |
if ($node->[1] == TABLE_ROW_EL) { |
| 3421 |
|
!!!cp ('t228'); |
| 3422 |
$i = $_; |
$i = $_; |
| 3423 |
last INSCOPE; |
last INSCOPE; |
| 3424 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3425 |
table => 1, html => 1, |
!!!cp ('t229'); |
|
}->{$node->[1]}) { |
|
| 3426 |
last INSCOPE; |
last INSCOPE; |
| 3427 |
} |
} |
| 3428 |
} # INSCOPE |
} # INSCOPE |
| 3429 |
unless (defined $i) { |
unless (defined $i) { |
| 3430 |
!!!parse-error (type => 'unmacthed end tag:'.$token->{tag_name}); |
!!!cp ('t230'); |
| 3431 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3432 |
|
text => $token->{tag_name}, token => $token); |
| 3433 |
## Ignore the token |
## Ignore the token |
| 3434 |
|
!!!nack ('t230.1'); |
| 3435 |
!!!next-token; |
!!!next-token; |
| 3436 |
redo B; |
next B; |
| 3437 |
|
} else { |
| 3438 |
|
!!!cp ('t232'); |
| 3439 |
} |
} |
| 3440 |
|
|
| 3441 |
## Clear back to table row context |
## Clear back to table row context |
| 3442 |
while (not { |
while (not ($self->{open_elements}->[-1]->[1] |
| 3443 |
tr => 1, html => 1, |
& TABLE_ROW_SCOPING_EL)) { |
| 3444 |
}->{$self->{open_elements}->[-1]->[1]}) { |
!!!cp ('t231'); |
| 3445 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
## ISSUE: Can this state be reached? |
| 3446 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3447 |
} |
} |
| 3448 |
|
|
| 3449 |
pop @{$self->{open_elements}}; # tr |
pop @{$self->{open_elements}}; # tr |
| 3450 |
$self->{insertion_mode} = 'in table body'; |
$self->{insertion_mode} = IN_TABLE_BODY_IM; |
| 3451 |
## reprocess |
!!!next-token; |
| 3452 |
redo B; |
!!!nack ('t231.1'); |
| 3453 |
|
next B; |
| 3454 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
| 3455 |
## NOTE: This is a code clone of "table in table" |
if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3456 |
!!!parse-error (type => 'not closed:table'); |
## As if </tr> |
| 3457 |
|
## have an element in table scope |
| 3458 |
## As if </table> |
my $i; |
| 3459 |
## have a table element in table scope |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3460 |
my $i; |
my $node = $self->{open_elements}->[$_]; |
| 3461 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
if ($node->[1] == TABLE_ROW_EL) { |
| 3462 |
my $node = $self->{open_elements}->[$_]; |
!!!cp ('t233'); |
| 3463 |
if ($node->[1] eq 'table') { |
$i = $_; |
| 3464 |
$i = $_; |
last INSCOPE; |
| 3465 |
last INSCOPE; |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3466 |
} elsif ({ |
!!!cp ('t234'); |
| 3467 |
table => 1, html => 1, |
last INSCOPE; |
| 3468 |
}->{$node->[1]}) { |
} |
| 3469 |
last INSCOPE; |
} # INSCOPE |
| 3470 |
|
unless (defined $i) { |
| 3471 |
|
!!!cp ('t235'); |
| 3472 |
|
## TODO: The following is wrong. |
| 3473 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3474 |
|
text => $token->{type}, token => $token); |
| 3475 |
|
## Ignore the token |
| 3476 |
|
!!!nack ('t236.1'); |
| 3477 |
|
!!!next-token; |
| 3478 |
|
next B; |
| 3479 |
} |
} |
| 3480 |
} # INSCOPE |
|
| 3481 |
unless (defined $i) { |
## Clear back to table row context |
| 3482 |
!!!parse-error (type => 'unmatched end tag:table'); |
while (not ($self->{open_elements}->[-1]->[1] |
| 3483 |
## Ignore tokens </table><table> |
& TABLE_ROW_SCOPING_EL)) { |
| 3484 |
!!!next-token; |
!!!cp ('t236'); |
| 3485 |
redo B; |
## ISSUE: Can this state be reached? |
| 3486 |
} |
pop @{$self->{open_elements}}; |
| 3487 |
|
} |
| 3488 |
## generate implied end tags |
|
| 3489 |
if ({ |
pop @{$self->{open_elements}}; # tr |
| 3490 |
dd => 1, dt => 1, li => 1, p => 1, |
$self->{insertion_mode} = IN_TABLE_BODY_IM; |
| 3491 |
td => 1, th => 1, tr => 1, |
## reprocess in the "in table body" insertion mode... |
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <table> |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
| 3492 |
} |
} |
| 3493 |
|
|
| 3494 |
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { |
| 3495 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
## have an element in table scope |
| 3496 |
|
my $i; |
| 3497 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3498 |
|
my $node = $self->{open_elements}->[$_]; |
| 3499 |
|
if ($node->[1] == TABLE_ROW_GROUP_EL) { |
| 3500 |
|
!!!cp ('t237'); |
| 3501 |
|
$i = $_; |
| 3502 |
|
last INSCOPE; |
| 3503 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3504 |
|
!!!cp ('t238'); |
| 3505 |
|
last INSCOPE; |
| 3506 |
|
} |
| 3507 |
|
} # INSCOPE |
| 3508 |
|
unless (defined $i) { |
| 3509 |
|
!!!cp ('t239'); |
| 3510 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3511 |
|
text => $token->{tag_name}, token => $token); |
| 3512 |
|
## Ignore the token |
| 3513 |
|
!!!nack ('t239.1'); |
| 3514 |
|
!!!next-token; |
| 3515 |
|
next B; |
| 3516 |
|
} |
| 3517 |
|
|
| 3518 |
|
## Clear back to table body context |
| 3519 |
|
while (not ($self->{open_elements}->[-1]->[1] |
| 3520 |
|
& TABLE_ROWS_SCOPING_EL)) { |
| 3521 |
|
!!!cp ('t240'); |
| 3522 |
|
pop @{$self->{open_elements}}; |
| 3523 |
|
} |
| 3524 |
|
|
| 3525 |
|
## As if <{current node}> |
| 3526 |
|
## have an element in table scope |
| 3527 |
|
## true by definition |
| 3528 |
|
|
| 3529 |
|
## Clear back to table body context |
| 3530 |
|
## nop by definition |
| 3531 |
|
|
| 3532 |
|
pop @{$self->{open_elements}}; |
| 3533 |
|
$self->{insertion_mode} = IN_TABLE_IM; |
| 3534 |
|
## reprocess in the "in table" insertion mode... |
| 3535 |
} |
} |
| 3536 |
|
|
| 3537 |
splice @{$self->{open_elements}}, $i; |
## NOTE: </table> in the "in table" insertion mode. |
| 3538 |
|
## When you edit the code fragment below, please ensure that |
| 3539 |
|
## the code for <table> in the "in table" insertion mode |
| 3540 |
|
## is synced with it. |
| 3541 |
|
|
| 3542 |
$self->_reset_insertion_mode; |
## have a table element in table scope |
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'tr') { |
|
|
## have an element in table scope |
|
| 3543 |
my $i; |
my $i; |
| 3544 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3545 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3546 |
if ($node->[1] eq $token->{tag_name}) { |
if ($node->[1] == TABLE_EL) { |
| 3547 |
|
!!!cp ('t241'); |
| 3548 |
$i = $_; |
$i = $_; |
| 3549 |
last INSCOPE; |
last INSCOPE; |
| 3550 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3551 |
table => 1, html => 1, |
!!!cp ('t242'); |
|
}->{$node->[1]}) { |
|
| 3552 |
last INSCOPE; |
last INSCOPE; |
| 3553 |
} |
} |
| 3554 |
} # INSCOPE |
} # INSCOPE |
| 3555 |
unless (defined $i) { |
unless (defined $i) { |
| 3556 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t243'); |
| 3557 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3558 |
|
text => $token->{tag_name}, token => $token); |
| 3559 |
## Ignore the token |
## Ignore the token |
| 3560 |
|
!!!nack ('t243.1'); |
| 3561 |
!!!next-token; |
!!!next-token; |
| 3562 |
redo B; |
next B; |
| 3563 |
} |
} |
| 3564 |
|
|
| 3565 |
## Clear back to table row context |
splice @{$self->{open_elements}}, $i; |
| 3566 |
while (not { |
pop @{$open_tables}; |
| 3567 |
tr => 1, html => 1, |
|
| 3568 |
}->{$self->{open_elements}->[-1]->[1]}) { |
$self->_reset_insertion_mode; |
| 3569 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
| 3570 |
pop @{$self->{open_elements}}; |
!!!next-token; |
| 3571 |
|
next B; |
| 3572 |
|
} elsif ({ |
| 3573 |
|
tbody => 1, tfoot => 1, thead => 1, |
| 3574 |
|
}->{$token->{tag_name}} and |
| 3575 |
|
$self->{insertion_mode} & ROW_IMS) { |
| 3576 |
|
if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3577 |
|
## have an element in table scope |
| 3578 |
|
my $i; |
| 3579 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3580 |
|
my $node = $self->{open_elements}->[$_]; |
| 3581 |
|
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 3582 |
|
!!!cp ('t247'); |
| 3583 |
|
$i = $_; |
| 3584 |
|
last INSCOPE; |
| 3585 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3586 |
|
!!!cp ('t248'); |
| 3587 |
|
last INSCOPE; |
| 3588 |
|
} |
| 3589 |
|
} # INSCOPE |
| 3590 |
|
unless (defined $i) { |
| 3591 |
|
!!!cp ('t249'); |
| 3592 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3593 |
|
text => $token->{tag_name}, token => $token); |
| 3594 |
|
## Ignore the token |
| 3595 |
|
!!!nack ('t249.1'); |
| 3596 |
|
!!!next-token; |
| 3597 |
|
next B; |
| 3598 |
|
} |
| 3599 |
|
|
| 3600 |
|
## As if </tr> |
| 3601 |
|
## have an element in table scope |
| 3602 |
|
my $i; |
| 3603 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3604 |
|
my $node = $self->{open_elements}->[$_]; |
| 3605 |
|
if ($node->[1] == TABLE_ROW_EL) { |
| 3606 |
|
!!!cp ('t250'); |
| 3607 |
|
$i = $_; |
| 3608 |
|
last INSCOPE; |
| 3609 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3610 |
|
!!!cp ('t251'); |
| 3611 |
|
last INSCOPE; |
| 3612 |
|
} |
| 3613 |
|
} # INSCOPE |
| 3614 |
|
unless (defined $i) { |
| 3615 |
|
!!!cp ('t252'); |
| 3616 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3617 |
|
text => 'tr', token => $token); |
| 3618 |
|
## Ignore the token |
| 3619 |
|
!!!nack ('t252.1'); |
| 3620 |
|
!!!next-token; |
| 3621 |
|
next B; |
| 3622 |
|
} |
| 3623 |
|
|
| 3624 |
|
## Clear back to table row context |
| 3625 |
|
while (not ($self->{open_elements}->[-1]->[1] |
| 3626 |
|
& TABLE_ROW_SCOPING_EL)) { |
| 3627 |
|
!!!cp ('t253'); |
| 3628 |
|
## ISSUE: Can this case be reached? |
| 3629 |
|
pop @{$self->{open_elements}}; |
| 3630 |
|
} |
| 3631 |
|
|
| 3632 |
|
pop @{$self->{open_elements}}; # tr |
| 3633 |
|
$self->{insertion_mode} = IN_TABLE_BODY_IM; |
| 3634 |
|
## reprocess in the "in table body" insertion mode... |
| 3635 |
} |
} |
| 3636 |
|
|
|
pop @{$self->{open_elements}}; # tr |
|
|
$self->{insertion_mode} = 'in table body'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## As if </tr> |
|
| 3637 |
## have an element in table scope |
## have an element in table scope |
| 3638 |
my $i; |
my $i; |
| 3639 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3640 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3641 |
if ($node->[1] eq 'tr') { |
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 3642 |
|
!!!cp ('t254'); |
| 3643 |
$i = $_; |
$i = $_; |
| 3644 |
last INSCOPE; |
last INSCOPE; |
| 3645 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3646 |
table => 1, html => 1, |
!!!cp ('t255'); |
|
}->{$node->[1]}) { |
|
| 3647 |
last INSCOPE; |
last INSCOPE; |
| 3648 |
} |
} |
| 3649 |
} # INSCOPE |
} # INSCOPE |
| 3650 |
unless (defined $i) { |
unless (defined $i) { |
| 3651 |
!!!parse-error (type => 'unmatched end tag:'.$token->{type}); |
!!!cp ('t256'); |
| 3652 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3653 |
|
text => $token->{tag_name}, token => $token); |
| 3654 |
## Ignore the token |
## Ignore the token |
| 3655 |
|
!!!nack ('t256.1'); |
| 3656 |
!!!next-token; |
!!!next-token; |
| 3657 |
redo B; |
next B; |
| 3658 |
} |
} |
| 3659 |
|
|
| 3660 |
## Clear back to table row context |
## Clear back to table body context |
| 3661 |
while (not { |
while (not ($self->{open_elements}->[-1]->[1] |
| 3662 |
tr => 1, html => 1, |
& TABLE_ROWS_SCOPING_EL)) { |
| 3663 |
}->{$self->{open_elements}->[-1]->[1]}) { |
!!!cp ('t257'); |
| 3664 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
## ISSUE: Can this case be reached? |
| 3665 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3666 |
} |
} |
| 3667 |
|
|
| 3668 |
pop @{$self->{open_elements}}; # tr |
pop @{$self->{open_elements}}; |
| 3669 |
$self->{insertion_mode} = 'in table body'; |
$self->{insertion_mode} = IN_TABLE_IM; |
| 3670 |
## reprocess |
!!!nack ('t257.1'); |
| 3671 |
redo B; |
!!!next-token; |
| 3672 |
|
next B; |
| 3673 |
} elsif ({ |
} elsif ({ |
| 3674 |
tbody => 1, tfoot => 1, thead => 1, |
body => 1, caption => 1, col => 1, colgroup => 1, |
| 3675 |
|
html => 1, td => 1, th => 1, |
| 3676 |
|
tr => 1, # $self->{insertion_mode} == IN_ROW_IM |
| 3677 |
|
tbody => 1, tfoot => 1, thead => 1, # $self->{insertion_mode} == IN_TABLE_IM |
| 3678 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 3679 |
## have an element in table scope |
!!!cp ('t258'); |
| 3680 |
my $i; |
!!!parse-error (type => 'unmatched end tag', |
| 3681 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
text => $token->{tag_name}, token => $token); |
| 3682 |
my $node = $self->{open_elements}->[$_]; |
## Ignore the token |
| 3683 |
if ($node->[1] eq $token->{tag_name}) { |
!!!nack ('t258.1'); |
| 3684 |
$i = $_; |
!!!next-token; |
| 3685 |
last INSCOPE; |
next B; |
| 3686 |
} elsif ({ |
} else { |
| 3687 |
table => 1, html => 1, |
!!!cp ('t259'); |
| 3688 |
}->{$node->[1]}) { |
!!!parse-error (type => 'in table:/', |
| 3689 |
last INSCOPE; |
text => $token->{tag_name}, token => $token); |
| 3690 |
} |
|
| 3691 |
} # INSCOPE |
$insert = $insert_to_foster; |
| 3692 |
unless (defined $i) { |
# |
| 3693 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
} |
| 3694 |
## Ignore the token |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 3695 |
|
unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 3696 |
|
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 3697 |
|
!!!parse-error (type => 'in body:#eof', token => $token); |
| 3698 |
|
!!!cp ('t259.1'); |
| 3699 |
|
# |
| 3700 |
|
} else { |
| 3701 |
|
!!!cp ('t259.2'); |
| 3702 |
|
# |
| 3703 |
|
} |
| 3704 |
|
|
| 3705 |
|
## Stop parsing |
| 3706 |
|
last B; |
| 3707 |
|
} else { |
| 3708 |
|
die "$0: $token->{type}: Unknown token type"; |
| 3709 |
|
} |
| 3710 |
|
} elsif (($self->{insertion_mode} & IM_MASK) == IN_COLUMN_GROUP_IM) { |
| 3711 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 3712 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 3713 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3714 |
|
unless (length $token->{data}) { |
| 3715 |
|
!!!cp ('t260'); |
| 3716 |
!!!next-token; |
!!!next-token; |
| 3717 |
redo B; |
next B; |
| 3718 |
} |
} |
| 3719 |
|
} |
| 3720 |
## As if </tr> |
|
| 3721 |
## have an element in table scope |
!!!cp ('t261'); |
| 3722 |
my $i; |
# |
| 3723 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 3724 |
my $node = $self->{open_elements}->[$_]; |
if ($token->{tag_name} eq 'col') { |
| 3725 |
if ($node->[1] eq 'tr') { |
!!!cp ('t262'); |
| 3726 |
$i = $_; |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3727 |
last INSCOPE; |
pop @{$self->{open_elements}}; |
| 3728 |
} elsif ({ |
!!!ack ('t262.1'); |
| 3729 |
table => 1, html => 1, |
!!!next-token; |
| 3730 |
}->{$node->[1]}) { |
next B; |
| 3731 |
last INSCOPE; |
} else { |
| 3732 |
} |
!!!cp ('t263'); |
| 3733 |
} # INSCOPE |
# |
| 3734 |
unless (defined $i) { |
} |
| 3735 |
!!!parse-error (type => 'unmatched end tag:tr'); |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3736 |
|
if ($token->{tag_name} eq 'colgroup') { |
| 3737 |
|
if ($self->{open_elements}->[-1]->[1] == HTML_EL) { |
| 3738 |
|
!!!cp ('t264'); |
| 3739 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3740 |
|
text => 'colgroup', token => $token); |
| 3741 |
## Ignore the token |
## Ignore the token |
| 3742 |
!!!next-token; |
!!!next-token; |
| 3743 |
redo B; |
next B; |
| 3744 |
} |
} else { |
| 3745 |
|
!!!cp ('t265'); |
| 3746 |
## Clear back to table row context |
pop @{$self->{open_elements}}; # colgroup |
| 3747 |
while (not { |
$self->{insertion_mode} = IN_TABLE_IM; |
| 3748 |
tr => 1, html => 1, |
!!!next-token; |
| 3749 |
}->{$self->{open_elements}->[-1]->[1]}) { |
next B; |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
| 3750 |
} |
} |
| 3751 |
|
} elsif ($token->{tag_name} eq 'col') { |
| 3752 |
pop @{$self->{open_elements}}; # tr |
!!!cp ('t266'); |
| 3753 |
$self->{insertion_mode} = 'in table body'; |
!!!parse-error (type => 'unmatched end tag', |
| 3754 |
## reprocess |
text => 'col', token => $token); |
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, |
|
|
colgroup => 1, html => 1, td => 1, th => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
| 3755 |
## Ignore the token |
## Ignore the token |
| 3756 |
!!!next-token; |
!!!next-token; |
| 3757 |
redo B; |
next B; |
| 3758 |
} else { |
} else { |
| 3759 |
# |
!!!cp ('t267'); |
| 3760 |
|
# |
| 3761 |
} |
} |
| 3762 |
} else { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 3763 |
# |
if ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 3764 |
} |
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 3765 |
|
!!!cp ('t270.2'); |
| 3766 |
|
## Stop parsing. |
| 3767 |
|
last B; |
| 3768 |
|
} else { |
| 3769 |
|
## NOTE: As if </colgroup>. |
| 3770 |
|
!!!cp ('t270.1'); |
| 3771 |
|
pop @{$self->{open_elements}}; # colgroup |
| 3772 |
|
$self->{insertion_mode} = IN_TABLE_IM; |
| 3773 |
|
## Reprocess. |
| 3774 |
|
next B; |
| 3775 |
|
} |
| 3776 |
|
} else { |
| 3777 |
|
die "$0: $token->{type}: Unknown token type"; |
| 3778 |
|
} |
| 3779 |
|
|
| 3780 |
## As if in table |
## As if </colgroup> |
| 3781 |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
if ($self->{open_elements}->[-1]->[1] == HTML_EL) { |
| 3782 |
$in_body->($insert_to_foster); |
!!!cp ('t269'); |
| 3783 |
redo B; |
## TODO: Wrong error type? |
| 3784 |
} elsif ($self->{insertion_mode} eq 'in select') { |
!!!parse-error (type => 'unmatched end tag', |
| 3785 |
if ($token->{type} eq 'character') { |
text => 'colgroup', token => $token); |
| 3786 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
## Ignore the token |
| 3787 |
|
!!!nack ('t269.1'); |
| 3788 |
!!!next-token; |
!!!next-token; |
| 3789 |
redo B; |
next B; |
| 3790 |
} elsif ($token->{type} eq 'start tag') { |
} else { |
| 3791 |
if ($token->{tag_name} eq 'option') { |
!!!cp ('t270'); |
| 3792 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
pop @{$self->{open_elements}}; # colgroup |
| 3793 |
## As if </option> |
$self->{insertion_mode} = IN_TABLE_IM; |
| 3794 |
pop @{$self->{open_elements}}; |
!!!ack-later; |
| 3795 |
} |
## reprocess |
| 3796 |
|
next B; |
| 3797 |
|
} |
| 3798 |
|
} elsif ($self->{insertion_mode} & SELECT_IMS) { |
| 3799 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 3800 |
|
!!!cp ('t271'); |
| 3801 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3802 |
|
!!!next-token; |
| 3803 |
|
next B; |
| 3804 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 3805 |
|
if ($token->{tag_name} eq 'option') { |
| 3806 |
|
if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { |
| 3807 |
|
!!!cp ('t272'); |
| 3808 |
|
## As if </option> |
| 3809 |
|
pop @{$self->{open_elements}}; |
| 3810 |
|
} else { |
| 3811 |
|
!!!cp ('t273'); |
| 3812 |
|
} |
| 3813 |
|
|
| 3814 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3815 |
!!!next-token; |
!!!nack ('t273.1'); |
| 3816 |
redo B; |
!!!next-token; |
| 3817 |
} elsif ($token->{tag_name} eq 'optgroup') { |
next B; |
| 3818 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
} elsif ($token->{tag_name} eq 'optgroup') { |
| 3819 |
## As if </option> |
if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { |
| 3820 |
pop @{$self->{open_elements}}; |
!!!cp ('t274'); |
| 3821 |
} |
## As if </option> |
| 3822 |
|
pop @{$self->{open_elements}}; |
| 3823 |
|
} else { |
| 3824 |
|
!!!cp ('t275'); |
| 3825 |
|
} |
| 3826 |
|
|
| 3827 |
if ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
if ($self->{open_elements}->[-1]->[1] == OPTGROUP_EL) { |
| 3828 |
## As if </optgroup> |
!!!cp ('t276'); |
| 3829 |
pop @{$self->{open_elements}}; |
## As if </optgroup> |
| 3830 |
} |
pop @{$self->{open_elements}}; |
| 3831 |
|
} else { |
| 3832 |
|
!!!cp ('t277'); |
| 3833 |
|
} |
| 3834 |
|
|
| 3835 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3836 |
!!!next-token; |
!!!nack ('t277.1'); |
| 3837 |
redo B; |
!!!next-token; |
| 3838 |
} elsif ($token->{tag_name} eq 'select') { |
next B; |
| 3839 |
!!!parse-error (type => 'not closed:select'); |
} elsif ({ |
| 3840 |
## As if </select> instead |
select => 1, input => 1, textarea => 1, keygen => 1, |
| 3841 |
## have an element in table scope |
}->{$token->{tag_name}} or |
| 3842 |
my $i; |
(($self->{insertion_mode} & IM_MASK) |
| 3843 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
== IN_SELECT_IN_TABLE_IM and |
| 3844 |
my $node = $self->{open_elements}->[$_]; |
{ |
| 3845 |
if ($node->[1] eq $token->{tag_name}) { |
caption => 1, table => 1, |
| 3846 |
$i = $_; |
tbody => 1, tfoot => 1, thead => 1, |
| 3847 |
last INSCOPE; |
tr => 1, td => 1, th => 1, |
| 3848 |
} elsif ({ |
}->{$token->{tag_name}})) { |
| 3849 |
table => 1, html => 1, |
## TODO: The type below is not good - <select> is replaced by </select> |
| 3850 |
}->{$node->[1]}) { |
!!!parse-error (type => 'not closed', text => 'select', |
| 3851 |
last INSCOPE; |
token => $token); |
| 3852 |
} |
## NOTE: As if the token were </select> (<select> case) or |
| 3853 |
} # INSCOPE |
## as if there were </select> (otherwise). |
| 3854 |
unless (defined $i) { |
## have an element in table scope |
| 3855 |
!!!parse-error (type => 'unmatched end tag:select'); |
my $i; |
| 3856 |
## Ignore the token |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3857 |
!!!next-token; |
my $node = $self->{open_elements}->[$_]; |
| 3858 |
redo B; |
if ($node->[1] == SELECT_EL) { |
| 3859 |
} |
!!!cp ('t278'); |
| 3860 |
|
$i = $_; |
| 3861 |
|
last INSCOPE; |
| 3862 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3863 |
|
!!!cp ('t279'); |
| 3864 |
|
last INSCOPE; |
| 3865 |
|
} |
| 3866 |
|
} # INSCOPE |
| 3867 |
|
unless (defined $i) { |
| 3868 |
|
!!!cp ('t280'); |
| 3869 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3870 |
|
text => 'select', token => $token); |
| 3871 |
|
## Ignore the token |
| 3872 |
|
!!!nack ('t280.1'); |
| 3873 |
|
!!!next-token; |
| 3874 |
|
next B; |
| 3875 |
|
} |
| 3876 |
|
|
| 3877 |
splice @{$self->{open_elements}}, $i; |
!!!cp ('t281'); |
| 3878 |
|
splice @{$self->{open_elements}}, $i; |
| 3879 |
|
|
| 3880 |
$self->_reset_insertion_mode; |
$self->_reset_insertion_mode; |
| 3881 |
|
|
| 3882 |
!!!next-token; |
if ($token->{tag_name} eq 'select') { |
| 3883 |
redo B; |
!!!nack ('t281.2'); |
| 3884 |
} else { |
!!!next-token; |
| 3885 |
# |
next B; |
| 3886 |
|
} else { |
| 3887 |
|
!!!cp ('t281.1'); |
| 3888 |
|
!!!ack-later; |
| 3889 |
|
## Reprocess the token. |
| 3890 |
|
next B; |
| 3891 |
|
} |
| 3892 |
|
} else { |
| 3893 |
|
!!!cp ('t282'); |
| 3894 |
|
!!!parse-error (type => 'in select', |
| 3895 |
|
text => $token->{tag_name}, token => $token); |
| 3896 |
|
## Ignore the token |
| 3897 |
|
!!!nack ('t282.1'); |
| 3898 |
|
!!!next-token; |
| 3899 |
|
next B; |
| 3900 |
|
} |
| 3901 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3902 |
|
if ($token->{tag_name} eq 'optgroup') { |
| 3903 |
|
if ($self->{open_elements}->[-1]->[1] == OPTION_EL and |
| 3904 |
|
$self->{open_elements}->[-2]->[1] == OPTGROUP_EL) { |
| 3905 |
|
!!!cp ('t283'); |
| 3906 |
|
## As if </option> |
| 3907 |
|
splice @{$self->{open_elements}}, -2; |
| 3908 |
|
} elsif ($self->{open_elements}->[-1]->[1] == OPTGROUP_EL) { |
| 3909 |
|
!!!cp ('t284'); |
| 3910 |
|
pop @{$self->{open_elements}}; |
| 3911 |
|
} else { |
| 3912 |
|
!!!cp ('t285'); |
| 3913 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3914 |
|
text => $token->{tag_name}, token => $token); |
| 3915 |
|
## Ignore the token |
| 3916 |
|
} |
| 3917 |
|
!!!nack ('t285.1'); |
| 3918 |
|
!!!next-token; |
| 3919 |
|
next B; |
| 3920 |
|
} elsif ($token->{tag_name} eq 'option') { |
| 3921 |
|
if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { |
| 3922 |
|
!!!cp ('t286'); |
| 3923 |
|
pop @{$self->{open_elements}}; |
| 3924 |
|
} else { |
| 3925 |
|
!!!cp ('t287'); |
| 3926 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3927 |
|
text => $token->{tag_name}, token => $token); |
| 3928 |
|
## Ignore the token |
| 3929 |
|
} |
| 3930 |
|
!!!nack ('t287.1'); |
| 3931 |
|
!!!next-token; |
| 3932 |
|
next B; |
| 3933 |
|
} elsif ($token->{tag_name} eq 'select') { |
| 3934 |
|
## have an element in table scope |
| 3935 |
|
my $i; |
| 3936 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3937 |
|
my $node = $self->{open_elements}->[$_]; |
| 3938 |
|
if ($node->[1] == SELECT_EL) { |
| 3939 |
|
!!!cp ('t288'); |
| 3940 |
|
$i = $_; |
| 3941 |
|
last INSCOPE; |
| 3942 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3943 |
|
!!!cp ('t289'); |
| 3944 |
|
last INSCOPE; |
| 3945 |
} |
} |
| 3946 |
} elsif ($token->{type} eq 'end tag') { |
} # INSCOPE |
| 3947 |
if ($token->{tag_name} eq 'optgroup') { |
unless (defined $i) { |
| 3948 |
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
!!!cp ('t290'); |
| 3949 |
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
!!!parse-error (type => 'unmatched end tag', |
| 3950 |
## As if </option> |
text => $token->{tag_name}, token => $token); |
| 3951 |
splice @{$self->{open_elements}}, -2; |
## Ignore the token |
| 3952 |
} elsif ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
!!!nack ('t290.1'); |
| 3953 |
pop @{$self->{open_elements}}; |
!!!next-token; |
| 3954 |
} else { |
next B; |
| 3955 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
} |
|
## Ignore the token |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'option') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
|
|
pop @{$self->{open_elements}}; |
|
|
} else { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'select') { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 3956 |
|
|
| 3957 |
splice @{$self->{open_elements}}, $i; |
!!!cp ('t291'); |
| 3958 |
|
splice @{$self->{open_elements}}, $i; |
| 3959 |
|
|
| 3960 |
$self->_reset_insertion_mode; |
$self->_reset_insertion_mode; |
| 3961 |
|
|
| 3962 |
!!!next-token; |
!!!nack ('t291.1'); |
| 3963 |
redo B; |
!!!next-token; |
| 3964 |
} elsif ({ |
next B; |
| 3965 |
caption => 1, table => 1, tbody => 1, |
} elsif (($self->{insertion_mode} & IM_MASK) |
| 3966 |
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
== IN_SELECT_IN_TABLE_IM and |
| 3967 |
}->{$token->{tag_name}}) { |
{ |
| 3968 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
caption => 1, table => 1, tbody => 1, |
| 3969 |
|
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
| 3970 |
## have an element in table scope |
}->{$token->{tag_name}}) { |
| 3971 |
my $i; |
## TODO: The following is wrong? |
| 3972 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
!!!parse-error (type => 'unmatched end tag', |
| 3973 |
my $node = $self->{open_elements}->[$_]; |
text => $token->{tag_name}, token => $token); |
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
| 3974 |
|
|
| 3975 |
## As if </select> |
## have an element in table scope |
| 3976 |
## have an element in table scope |
my $i; |
| 3977 |
undef $i; |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3978 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
my $node = $self->{open_elements}->[$_]; |
| 3979 |
my $node = $self->{open_elements}->[$_]; |
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 3980 |
if ($node->[1] eq 'select') { |
!!!cp ('t292'); |
| 3981 |
$i = $_; |
$i = $_; |
| 3982 |
last INSCOPE; |
last INSCOPE; |
| 3983 |
} elsif ({ |
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3984 |
table => 1, html => 1, |
!!!cp ('t293'); |
| 3985 |
}->{$node->[1]}) { |
last INSCOPE; |
| 3986 |
last INSCOPE; |
} |
| 3987 |
} |
} # INSCOPE |
| 3988 |
} # INSCOPE |
unless (defined $i) { |
| 3989 |
unless (defined $i) { |
!!!cp ('t294'); |
| 3990 |
!!!parse-error (type => 'unmatched end tag:select'); |
## Ignore the token |
| 3991 |
## Ignore the </select> token |
!!!nack ('t294.1'); |
| 3992 |
!!!next-token; ## TODO: ok? |
!!!next-token; |
| 3993 |
redo B; |
next B; |
| 3994 |
} |
} |
| 3995 |
|
|
| 3996 |
splice @{$self->{open_elements}}, $i; |
## As if </select> |
| 3997 |
|
## have an element in table scope |
| 3998 |
$self->_reset_insertion_mode; |
undef $i; |
| 3999 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4000 |
## reprocess |
my $node = $self->{open_elements}->[$_]; |
| 4001 |
redo B; |
if ($node->[1] == SELECT_EL) { |
| 4002 |
} else { |
!!!cp ('t295'); |
| 4003 |
# |
$i = $_; |
| 4004 |
|
last INSCOPE; |
| 4005 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 4006 |
|
## ISSUE: Can this state be reached? |
| 4007 |
|
!!!cp ('t296'); |
| 4008 |
|
last INSCOPE; |
| 4009 |
} |
} |
| 4010 |
} else { |
} # INSCOPE |
| 4011 |
# |
unless (defined $i) { |
| 4012 |
|
!!!cp ('t297'); |
| 4013 |
|
## TODO: The following error type is correct? |
| 4014 |
|
!!!parse-error (type => 'unmatched end tag', |
| 4015 |
|
text => 'select', token => $token); |
| 4016 |
|
## Ignore the </select> token |
| 4017 |
|
!!!nack ('t297.1'); |
| 4018 |
|
!!!next-token; ## TODO: ok? |
| 4019 |
|
next B; |
| 4020 |
} |
} |
| 4021 |
|
|
| 4022 |
|
!!!cp ('t298'); |
| 4023 |
|
splice @{$self->{open_elements}}, $i; |
| 4024 |
|
|
| 4025 |
|
$self->_reset_insertion_mode; |
| 4026 |
|
|
| 4027 |
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
!!!ack-later; |
| 4028 |
|
## reprocess |
| 4029 |
|
next B; |
| 4030 |
|
} else { |
| 4031 |
|
!!!cp ('t299'); |
| 4032 |
|
!!!parse-error (type => 'in select:/', |
| 4033 |
|
text => $token->{tag_name}, token => $token); |
| 4034 |
## Ignore the token |
## Ignore the token |
| 4035 |
|
!!!nack ('t299.3'); |
| 4036 |
!!!next-token; |
!!!next-token; |
| 4037 |
redo B; |
next B; |
| 4038 |
} elsif ($self->{insertion_mode} eq 'after body') { |
} |
| 4039 |
if ($token->{type} eq 'character') { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 4040 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4041 |
my $data = $1; |
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 4042 |
## As if in body |
!!!cp ('t299.1'); |
| 4043 |
$reconstruct_active_formatting_elements->($insert_to_current); |
!!!parse-error (type => 'in body:#eof', token => $token); |
| 4044 |
|
} else { |
| 4045 |
|
!!!cp ('t299.2'); |
| 4046 |
|
} |
| 4047 |
|
|
| 4048 |
|
## Stop parsing. |
| 4049 |
|
last B; |
| 4050 |
|
} else { |
| 4051 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4052 |
|
} |
| 4053 |
|
} elsif ($self->{insertion_mode} & BODY_AFTER_IMS) { |
| 4054 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 4055 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 4056 |
|
my $data = $1; |
| 4057 |
|
## As if in body |
| 4058 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4059 |
|
|
| 4060 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 4061 |
|
|
| 4062 |
|
unless (length $token->{data}) { |
| 4063 |
|
!!!cp ('t300'); |
| 4064 |
|
!!!next-token; |
| 4065 |
|
next B; |
| 4066 |
|
} |
| 4067 |
|
} |
| 4068 |
|
|
| 4069 |
|
if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { |
| 4070 |
|
!!!cp ('t301'); |
| 4071 |
|
!!!parse-error (type => 'after html:#text', token => $token); |
| 4072 |
|
# |
| 4073 |
|
} else { |
| 4074 |
|
!!!cp ('t302'); |
| 4075 |
|
## "after body" insertion mode |
| 4076 |
|
!!!parse-error (type => 'after body:#text', token => $token); |
| 4077 |
|
# |
| 4078 |
|
} |
| 4079 |
|
|
| 4080 |
unless (length $token->{data}) { |
$self->{insertion_mode} = IN_BODY_IM; |
| 4081 |
!!!next-token; |
## reprocess |
| 4082 |
redo B; |
next B; |
| 4083 |
} |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 4084 |
} |
if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { |
| 4085 |
|
!!!cp ('t303'); |
| 4086 |
# |
!!!parse-error (type => 'after html', |
| 4087 |
!!!parse-error (type => 'after body:#character'); |
text => $token->{tag_name}, token => $token); |
| 4088 |
} elsif ($token->{type} eq 'start tag') { |
# |
| 4089 |
!!!parse-error (type => 'after body:'.$token->{tag_name}); |
} else { |
| 4090 |
# |
!!!cp ('t304'); |
| 4091 |
} elsif ($token->{type} eq 'end tag') { |
## "after body" insertion mode |
| 4092 |
if ($token->{tag_name} eq 'html') { |
!!!parse-error (type => 'after body', |
| 4093 |
if (defined $self->{inner_html_node}) { |
text => $token->{tag_name}, token => $token); |
| 4094 |
!!!parse-error (type => 'unmatched end tag:html'); |
# |
| 4095 |
## Ignore the token |
} |
| 4096 |
!!!next-token; |
|
| 4097 |
redo B; |
$self->{insertion_mode} = IN_BODY_IM; |
| 4098 |
} else { |
!!!ack-later; |
| 4099 |
$previous_insertion_mode = $self->{insertion_mode}; |
## reprocess |
| 4100 |
$self->{insertion_mode} = 'trailing end'; |
next B; |
| 4101 |
!!!next-token; |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 4102 |
redo B; |
if ($self->{insertion_mode} == AFTER_HTML_BODY_IM) { |
| 4103 |
} |
!!!cp ('t305'); |
| 4104 |
} else { |
!!!parse-error (type => 'after html:/', |
| 4105 |
!!!parse-error (type => 'after body:/'.$token->{tag_name}); |
text => $token->{tag_name}, token => $token); |
| 4106 |
} |
|
| 4107 |
|
$self->{insertion_mode} = IN_BODY_IM; |
| 4108 |
|
## Reprocess. |
| 4109 |
|
next B; |
| 4110 |
|
} else { |
| 4111 |
|
!!!cp ('t306'); |
| 4112 |
|
} |
| 4113 |
|
|
| 4114 |
|
## "after body" insertion mode |
| 4115 |
|
if ($token->{tag_name} eq 'html') { |
| 4116 |
|
if (defined $self->{inner_html_node}) { |
| 4117 |
|
!!!cp ('t307'); |
| 4118 |
|
!!!parse-error (type => 'unmatched end tag', |
| 4119 |
|
text => 'html', token => $token); |
| 4120 |
|
## Ignore the token |
| 4121 |
|
!!!next-token; |
| 4122 |
|
next B; |
| 4123 |
} else { |
} else { |
| 4124 |
die "$0: $token->{type}: Unknown token type"; |
!!!cp ('t308'); |
| 4125 |
|
$self->{insertion_mode} = AFTER_HTML_BODY_IM; |
| 4126 |
|
!!!next-token; |
| 4127 |
|
next B; |
| 4128 |
} |
} |
| 4129 |
|
} else { |
| 4130 |
|
!!!cp ('t309'); |
| 4131 |
|
!!!parse-error (type => 'after body:/', |
| 4132 |
|
text => $token->{tag_name}, token => $token); |
| 4133 |
|
|
| 4134 |
$self->{insertion_mode} = 'in body'; |
$self->{insertion_mode} = IN_BODY_IM; |
| 4135 |
## reprocess |
## reprocess |
| 4136 |
redo B; |
next B; |
| 4137 |
} elsif ($self->{insertion_mode} eq 'in frameset') { |
} |
| 4138 |
if ($token->{type} eq 'character') { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 4139 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
!!!cp ('t309.2'); |
| 4140 |
|
## Stop parsing |
| 4141 |
|
last B; |
| 4142 |
|
} else { |
| 4143 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4144 |
|
} |
| 4145 |
|
} elsif ($self->{insertion_mode} & FRAME_IMS) { |
| 4146 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 4147 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 4148 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 4149 |
|
|
| 4150 |
unless (length $token->{data}) { |
unless (length $token->{data}) { |
| 4151 |
|
!!!cp ('t310'); |
| 4152 |
!!!next-token; |
!!!next-token; |
| 4153 |
redo B; |
next B; |
| 4154 |
} |
} |
| 4155 |
} |
} |
| 4156 |
|
|
| 4157 |
!!!parse-error (type => 'in frameset:#character'); |
if ($token->{data} =~ s/^[^\x09\x0A\x0C\x20]+//) { |
| 4158 |
## Ignore the token |
if ($self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4159 |
!!!next-token; |
!!!cp ('t311'); |
| 4160 |
redo B; |
!!!parse-error (type => 'in frameset:#text', token => $token); |
| 4161 |
} elsif ($token->{type} eq 'start tag') { |
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 4162 |
if ($token->{tag_name} eq 'frameset') { |
!!!cp ('t312'); |
| 4163 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!parse-error (type => 'after frameset:#text', token => $token); |
| 4164 |
|
} else { # "after after frameset" |
| 4165 |
|
!!!cp ('t313'); |
| 4166 |
|
!!!parse-error (type => 'after html:#text', token => $token); |
| 4167 |
|
} |
| 4168 |
|
|
| 4169 |
|
## Ignore the token. |
| 4170 |
|
if (length $token->{data}) { |
| 4171 |
|
!!!cp ('t314'); |
| 4172 |
|
## reprocess the rest of characters |
| 4173 |
|
} else { |
| 4174 |
|
!!!cp ('t315'); |
| 4175 |
|
!!!next-token; |
| 4176 |
|
} |
| 4177 |
|
next B; |
| 4178 |
|
} |
| 4179 |
|
|
| 4180 |
|
die qq[$0: Character "$token->{data}"]; |
| 4181 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 4182 |
|
if ($token->{tag_name} eq 'frameset' and |
| 4183 |
|
$self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4184 |
|
!!!cp ('t318'); |
| 4185 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 4186 |
|
!!!nack ('t318.1'); |
| 4187 |
!!!next-token; |
!!!next-token; |
| 4188 |
redo B; |
next B; |
| 4189 |
} elsif ($token->{tag_name} eq 'frame') { |
} elsif ($token->{tag_name} eq 'frame' and |
| 4190 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4191 |
|
!!!cp ('t319'); |
| 4192 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 4193 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 4194 |
|
!!!ack ('t319.1'); |
| 4195 |
!!!next-token; |
!!!next-token; |
| 4196 |
redo B; |
next B; |
| 4197 |
} elsif ($token->{tag_name} eq 'noframes') { |
} elsif ($token->{tag_name} eq 'noframes') { |
| 4198 |
$in_body->($insert_to_current); |
!!!cp ('t320'); |
| 4199 |
redo B; |
## NOTE: As if in head. |
| 4200 |
} else { |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 4201 |
!!!parse-error (type => 'in frameset:'.$token->{tag_name}); |
next B; |
| 4202 |
|
|
| 4203 |
|
## NOTE: |<!DOCTYPE HTML><frameset></frameset></html><noframes></noframes>| |
| 4204 |
|
## has no parse error. |
| 4205 |
|
} else { |
| 4206 |
|
if ($self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4207 |
|
!!!cp ('t321'); |
| 4208 |
|
!!!parse-error (type => 'in frameset', |
| 4209 |
|
text => $token->{tag_name}, token => $token); |
| 4210 |
|
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 4211 |
|
!!!cp ('t322'); |
| 4212 |
|
!!!parse-error (type => 'after frameset', |
| 4213 |
|
text => $token->{tag_name}, token => $token); |
| 4214 |
|
} else { # "after after frameset" |
| 4215 |
|
!!!cp ('t322.2'); |
| 4216 |
|
!!!parse-error (type => 'after after frameset', |
| 4217 |
|
text => $token->{tag_name}, token => $token); |
| 4218 |
|
} |
| 4219 |
## Ignore the token |
## Ignore the token |
| 4220 |
|
!!!nack ('t322.1'); |
| 4221 |
!!!next-token; |
!!!next-token; |
| 4222 |
redo B; |
next B; |
| 4223 |
} |
} |
| 4224 |
} elsif ($token->{type} eq 'end tag') { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 4225 |
if ($token->{tag_name} eq 'frameset') { |
if ($token->{tag_name} eq 'frameset' and |
| 4226 |
if ($self->{open_elements}->[-1]->[1] eq 'html' and |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4227 |
|
if ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4228 |
@{$self->{open_elements}} == 1) { |
@{$self->{open_elements}} == 1) { |
| 4229 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
!!!cp ('t325'); |
| 4230 |
|
!!!parse-error (type => 'unmatched end tag', |
| 4231 |
|
text => $token->{tag_name}, token => $token); |
| 4232 |
## Ignore the token |
## Ignore the token |
| 4233 |
!!!next-token; |
!!!next-token; |
| 4234 |
} else { |
} else { |
| 4235 |
|
!!!cp ('t326'); |
| 4236 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 4237 |
!!!next-token; |
!!!next-token; |
| 4238 |
} |
} |
| 4239 |
|
|
| 4240 |
if (not defined $self->{inner_html_node} and |
if (not defined $self->{inner_html_node} and |
| 4241 |
$self->{open_elements}->[-1]->[1] ne 'frameset') { |
not ($self->{open_elements}->[-1]->[1] == FRAMESET_EL)) { |
| 4242 |
$self->{insertion_mode} = 'after frameset'; |
!!!cp ('t327'); |
| 4243 |
|
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
| 4244 |
|
} else { |
| 4245 |
|
!!!cp ('t328'); |
| 4246 |
} |
} |
| 4247 |
redo B; |
next B; |
| 4248 |
|
} elsif ($token->{tag_name} eq 'html' and |
| 4249 |
|
$self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 4250 |
|
!!!cp ('t329'); |
| 4251 |
|
$self->{insertion_mode} = AFTER_HTML_FRAMESET_IM; |
| 4252 |
|
!!!next-token; |
| 4253 |
|
next B; |
| 4254 |
} else { |
} else { |
| 4255 |
!!!parse-error (type => 'in frameset:/'.$token->{tag_name}); |
if ($self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4256 |
|
!!!cp ('t330'); |
| 4257 |
|
!!!parse-error (type => 'in frameset:/', |
| 4258 |
|
text => $token->{tag_name}, token => $token); |
| 4259 |
|
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 4260 |
|
!!!cp ('t330.1'); |
| 4261 |
|
!!!parse-error (type => 'after frameset:/', |
| 4262 |
|
text => $token->{tag_name}, token => $token); |
| 4263 |
|
} else { # "after after html" |
| 4264 |
|
!!!cp ('t331'); |
| 4265 |
|
!!!parse-error (type => 'after after frameset:/', |
| 4266 |
|
text => $token->{tag_name}, token => $token); |
| 4267 |
|
} |
| 4268 |
## Ignore the token |
## Ignore the token |
| 4269 |
!!!next-token; |
!!!next-token; |
| 4270 |
redo B; |
next B; |
| 4271 |
} |
} |
| 4272 |
|
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 4273 |
|
unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4274 |
|
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 4275 |
|
!!!cp ('t331.1'); |
| 4276 |
|
!!!parse-error (type => 'in body:#eof', token => $token); |
| 4277 |
|
} else { |
| 4278 |
|
!!!cp ('t331.2'); |
| 4279 |
|
} |
| 4280 |
|
|
| 4281 |
|
## Stop parsing |
| 4282 |
|
last B; |
| 4283 |
} else { |
} else { |
| 4284 |
die "$0: $token->{type}: Unknown token type"; |
die "$0: $token->{type}: Unknown token type"; |
| 4285 |
} |
} |
| 4286 |
} elsif ($self->{insertion_mode} eq 'after frameset') { |
} else { |
| 4287 |
if ($token->{type} eq 'character') { |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 4288 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
} |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
| 4289 |
|
|
| 4290 |
unless (length $token->{data}) { |
## "in body" insertion mode |
| 4291 |
!!!next-token; |
if ($token->{type} == START_TAG_TOKEN) { |
| 4292 |
redo B; |
if ($token->{tag_name} eq 'script') { |
| 4293 |
|
!!!cp ('t332'); |
| 4294 |
|
## NOTE: This is an "as if in head" code clone |
| 4295 |
|
$script_start_tag->(); |
| 4296 |
|
next B; |
| 4297 |
|
} elsif ($token->{tag_name} eq 'style') { |
| 4298 |
|
!!!cp ('t333'); |
| 4299 |
|
## NOTE: This is an "as if in head" code clone |
| 4300 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 4301 |
|
next B; |
| 4302 |
|
} elsif ({ |
| 4303 |
|
base => 1, command => 1, eventsource => 1, link => 1, |
| 4304 |
|
}->{$token->{tag_name}}) { |
| 4305 |
|
!!!cp ('t334'); |
| 4306 |
|
## NOTE: This is an "as if in head" code clone, only "-t" differs |
| 4307 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4308 |
|
pop @{$self->{open_elements}}; |
| 4309 |
|
!!!ack ('t334.1'); |
| 4310 |
|
!!!next-token; |
| 4311 |
|
next B; |
| 4312 |
|
} elsif ($token->{tag_name} eq 'meta') { |
| 4313 |
|
## NOTE: This is an "as if in head" code clone, only "-t" differs |
| 4314 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4315 |
|
my $meta_el = pop @{$self->{open_elements}}; |
| 4316 |
|
|
| 4317 |
|
unless ($self->{confident}) { |
| 4318 |
|
if ($token->{attributes}->{charset}) { |
| 4319 |
|
!!!cp ('t335'); |
| 4320 |
|
## NOTE: Whether the encoding is supported or not is handled |
| 4321 |
|
## in the {change_encoding} callback. |
| 4322 |
|
$self->{change_encoding} |
| 4323 |
|
->($self, $token->{attributes}->{charset}->{value}, $token); |
| 4324 |
|
|
| 4325 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 4326 |
|
->set_user_data (manakai_has_reference => |
| 4327 |
|
$token->{attributes}->{charset} |
| 4328 |
|
->{has_reference}); |
| 4329 |
|
} elsif ($token->{attributes}->{content}) { |
| 4330 |
|
if ($token->{attributes}->{content}->{value} |
| 4331 |
|
=~ /[Cc][Hh][Aa][Rr][Ss][Ee][Tt] |
| 4332 |
|
[\x09\x0A\x0C\x0D\x20]*= |
| 4333 |
|
[\x09\x0A\x0C\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 4334 |
|
([^"'\x09\x0A\x0C\x0D\x20][^\x09\x0A\x0C\x0D\x20\x3B]*)) |
| 4335 |
|
/x) { |
| 4336 |
|
!!!cp ('t336'); |
| 4337 |
|
## NOTE: Whether the encoding is supported or not is handled |
| 4338 |
|
## in the {change_encoding} callback. |
| 4339 |
|
$self->{change_encoding} |
| 4340 |
|
->($self, defined $1 ? $1 : defined $2 ? $2 : $3, $token); |
| 4341 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 4342 |
|
->set_user_data (manakai_has_reference => |
| 4343 |
|
$token->{attributes}->{content} |
| 4344 |
|
->{has_reference}); |
| 4345 |
|
} |
| 4346 |
|
} |
| 4347 |
|
} else { |
| 4348 |
|
if ($token->{attributes}->{charset}) { |
| 4349 |
|
!!!cp ('t337'); |
| 4350 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 4351 |
|
->set_user_data (manakai_has_reference => |
| 4352 |
|
$token->{attributes}->{charset} |
| 4353 |
|
->{has_reference}); |
| 4354 |
|
} |
| 4355 |
|
if ($token->{attributes}->{content}) { |
| 4356 |
|
!!!cp ('t338'); |
| 4357 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 4358 |
|
->set_user_data (manakai_has_reference => |
| 4359 |
|
$token->{attributes}->{content} |
| 4360 |
|
->{has_reference}); |
| 4361 |
|
} |
| 4362 |
|
} |
| 4363 |
|
|
| 4364 |
|
!!!ack ('t338.1'); |
| 4365 |
|
!!!next-token; |
| 4366 |
|
next B; |
| 4367 |
|
} elsif ($token->{tag_name} eq 'title') { |
| 4368 |
|
!!!cp ('t341'); |
| 4369 |
|
## NOTE: This is an "as if in head" code clone |
| 4370 |
|
$parse_rcdata->(RCDATA_CONTENT_MODEL); |
| 4371 |
|
next B; |
| 4372 |
|
} elsif ($token->{tag_name} eq 'body') { |
| 4373 |
|
!!!parse-error (type => 'in body', text => 'body', token => $token); |
| 4374 |
|
|
| 4375 |
|
if (@{$self->{open_elements}} == 1 or |
| 4376 |
|
not ($self->{open_elements}->[1]->[1] == BODY_EL)) { |
| 4377 |
|
!!!cp ('t342'); |
| 4378 |
|
## Ignore the token |
| 4379 |
|
} else { |
| 4380 |
|
my $body_el = $self->{open_elements}->[1]->[0]; |
| 4381 |
|
for my $attr_name (keys %{$token->{attributes}}) { |
| 4382 |
|
unless ($body_el->has_attribute_ns (undef, $attr_name)) { |
| 4383 |
|
!!!cp ('t343'); |
| 4384 |
|
$body_el->set_attribute_ns |
| 4385 |
|
(undef, [undef, $attr_name], |
| 4386 |
|
$token->{attributes}->{$attr_name}->{value}); |
| 4387 |
|
} |
| 4388 |
|
} |
| 4389 |
|
} |
| 4390 |
|
!!!nack ('t343.1'); |
| 4391 |
|
!!!next-token; |
| 4392 |
|
next B; |
| 4393 |
|
} elsif ({ |
| 4394 |
|
## NOTE: Start tags for non-phrasing flow content elements |
| 4395 |
|
|
| 4396 |
|
## NOTE: The normal one |
| 4397 |
|
address => 1, article => 1, aside => 1, blockquote => 1, |
| 4398 |
|
center => 1, datagrid => 1, details => 1, dialog => 1, |
| 4399 |
|
dir => 1, div => 1, dl => 1, fieldset => 1, figure => 1, |
| 4400 |
|
footer => 1, h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, |
| 4401 |
|
h6 => 1, header => 1, menu => 1, nav => 1, ol => 1, p => 1, |
| 4402 |
|
section => 1, ul => 1, |
| 4403 |
|
## NOTE: As normal, but drops leading newline |
| 4404 |
|
pre => 1, listing => 1, |
| 4405 |
|
## NOTE: As normal, but interacts with the form element pointer |
| 4406 |
|
form => 1, |
| 4407 |
|
|
| 4408 |
|
table => 1, |
| 4409 |
|
hr => 1, |
| 4410 |
|
}->{$token->{tag_name}}) { |
| 4411 |
|
if ($token->{tag_name} eq 'form' and defined $self->{form_element}) { |
| 4412 |
|
!!!cp ('t350'); |
| 4413 |
|
!!!parse-error (type => 'in form:form', token => $token); |
| 4414 |
|
## Ignore the token |
| 4415 |
|
!!!nack ('t350.1'); |
| 4416 |
|
!!!next-token; |
| 4417 |
|
next B; |
| 4418 |
|
} |
| 4419 |
|
|
| 4420 |
|
if ($token->{tag_name} ne 'table' or # The Hixie Quirk |
| 4421 |
|
$self->{document}->manakai_compat_mode ne 'quirks') { |
| 4422 |
|
## has a p element in scope |
| 4423 |
|
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4424 |
|
if ($_->[1] == P_EL) { |
| 4425 |
|
!!!cp ('t344'); |
| 4426 |
|
!!!back-token; # <form> |
| 4427 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4428 |
|
line => $token->{line}, column => $token->{column}}; |
| 4429 |
|
next B; |
| 4430 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 4431 |
|
!!!cp ('t345'); |
| 4432 |
|
last INSCOPE; |
| 4433 |
|
} |
| 4434 |
|
} # INSCOPE |
| 4435 |
|
} |
| 4436 |
|
|
| 4437 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4438 |
|
if ($token->{tag_name} eq 'pre' or $token->{tag_name} eq 'listing') { |
| 4439 |
|
!!!nack ('t346.1'); |
| 4440 |
|
!!!next-token; |
| 4441 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 4442 |
|
$token->{data} =~ s/^\x0A//; |
| 4443 |
|
unless (length $token->{data}) { |
| 4444 |
|
!!!cp ('t346'); |
| 4445 |
|
!!!next-token; |
| 4446 |
|
} else { |
| 4447 |
|
!!!cp ('t349'); |
| 4448 |
|
} |
| 4449 |
|
} else { |
| 4450 |
|
!!!cp ('t348'); |
| 4451 |
|
} |
| 4452 |
|
} elsif ($token->{tag_name} eq 'form') { |
| 4453 |
|
!!!cp ('t347.1'); |
| 4454 |
|
$self->{form_element} = $self->{open_elements}->[-1]->[0]; |
| 4455 |
|
|
| 4456 |
|
!!!nack ('t347.2'); |
| 4457 |
|
!!!next-token; |
| 4458 |
|
} elsif ($token->{tag_name} eq 'table') { |
| 4459 |
|
!!!cp ('t382'); |
| 4460 |
|
push @{$open_tables}, [$self->{open_elements}->[-1]->[0]]; |
| 4461 |
|
|
| 4462 |
|
$self->{insertion_mode} = IN_TABLE_IM; |
| 4463 |
|
|
| 4464 |
|
!!!nack ('t382.1'); |
| 4465 |
|
!!!next-token; |
| 4466 |
|
} elsif ($token->{tag_name} eq 'hr') { |
| 4467 |
|
!!!cp ('t386'); |
| 4468 |
|
pop @{$self->{open_elements}}; |
| 4469 |
|
|
| 4470 |
|
!!!nack ('t386.1'); |
| 4471 |
|
!!!next-token; |
| 4472 |
|
} else { |
| 4473 |
|
!!!nack ('t347.1'); |
| 4474 |
|
!!!next-token; |
| 4475 |
|
} |
| 4476 |
|
next B; |
| 4477 |
|
} elsif ($token->{tag_name} eq 'li') { |
| 4478 |
|
## NOTE: As normal, but imply </li> when there's another <li> ... |
| 4479 |
|
|
| 4480 |
|
## NOTE: Special, Scope (<li><foo><li> == <li><foo><li/></foo></li>) |
| 4481 |
|
## Interpreted as <li><foo/></li><li/> (non-conforming) |
| 4482 |
|
## blockquote (O9.27), center (O), dd (Fx3, O, S3.1.2, IE7), |
| 4483 |
|
## dt (Fx, O, S, IE), dl (O), fieldset (O, S, IE), form (Fx, O, S), |
| 4484 |
|
## hn (O), pre (O), applet (O, S), button (O, S), marquee (Fx, O, S), |
| 4485 |
|
## object (Fx) |
| 4486 |
|
## Generate non-tree (non-conforming) |
| 4487 |
|
## basefont (IE7 (where basefont is non-void)), center (IE), |
| 4488 |
|
## form (IE), hn (IE) |
| 4489 |
|
## address, div, p (<li><foo><li> == <li><foo/></li><li/>) |
| 4490 |
|
## Interpreted as <li><foo><li/></foo></li> (non-conforming) |
| 4491 |
|
## div (Fx, S) |
| 4492 |
|
|
| 4493 |
|
my $non_optional; |
| 4494 |
|
my $i = -1; |
| 4495 |
|
|
| 4496 |
|
## 1. |
| 4497 |
|
for my $node (reverse @{$self->{open_elements}}) { |
| 4498 |
|
if ($node->[1] == LI_EL) { |
| 4499 |
|
## 2. (a) As if </li> |
| 4500 |
|
{ |
| 4501 |
|
## If no </li> - not applied |
| 4502 |
|
# |
| 4503 |
|
|
| 4504 |
|
## Otherwise |
| 4505 |
|
|
| 4506 |
|
## 1. generate implied end tags, except for </li> |
| 4507 |
|
# |
| 4508 |
|
|
| 4509 |
|
## 2. If current node != "li", parse error |
| 4510 |
|
if ($non_optional) { |
| 4511 |
|
!!!parse-error (type => 'not closed', |
| 4512 |
|
text => $non_optional->[0]->manakai_local_name, |
| 4513 |
|
token => $token); |
| 4514 |
|
!!!cp ('t355'); |
| 4515 |
|
} else { |
| 4516 |
|
!!!cp ('t356'); |
| 4517 |
} |
} |
| 4518 |
|
|
| 4519 |
|
## 3. Pop |
| 4520 |
|
splice @{$self->{open_elements}}, $i; |
| 4521 |
} |
} |
| 4522 |
|
|
| 4523 |
if ($token->{data} =~ s/^[^\x09\x0A\x0B\x0C\x20]+//) { |
last; ## 2. (b) goto 5. |
| 4524 |
!!!parse-error (type => 'after frameset:#character'); |
} elsif ( |
| 4525 |
|
## NOTE: not "formatting" and not "phrasing" |
| 4526 |
|
($node->[1] & SPECIAL_EL or |
| 4527 |
|
$node->[1] & SCOPING_EL) and |
| 4528 |
|
## NOTE: "li", "dt", and "dd" are in |SPECIAL_EL|. |
| 4529 |
|
(not $node->[1] & ADDRESS_DIV_P_EL) |
| 4530 |
|
) { |
| 4531 |
|
## 3. |
| 4532 |
|
!!!cp ('t357'); |
| 4533 |
|
last; ## goto 5. |
| 4534 |
|
} elsif ($node->[1] & END_TAG_OPTIONAL_EL) { |
| 4535 |
|
!!!cp ('t358'); |
| 4536 |
|
# |
| 4537 |
|
} else { |
| 4538 |
|
!!!cp ('t359'); |
| 4539 |
|
$non_optional ||= $node; |
| 4540 |
|
# |
| 4541 |
|
} |
| 4542 |
|
## 4. |
| 4543 |
|
## goto 2. |
| 4544 |
|
$i--; |
| 4545 |
|
} |
| 4546 |
|
|
| 4547 |
|
## 5. (a) has a |p| element in scope |
| 4548 |
|
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4549 |
|
if ($_->[1] == P_EL) { |
| 4550 |
|
!!!cp ('t353'); |
| 4551 |
|
|
| 4552 |
|
## NOTE: |<p><li>|, for example. |
| 4553 |
|
|
| 4554 |
|
!!!back-token; # <x> |
| 4555 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4556 |
|
line => $token->{line}, column => $token->{column}}; |
| 4557 |
|
next B; |
| 4558 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 4559 |
|
!!!cp ('t354'); |
| 4560 |
|
last INSCOPE; |
| 4561 |
|
} |
| 4562 |
|
} # INSCOPE |
| 4563 |
|
|
| 4564 |
|
## 5. (b) insert |
| 4565 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4566 |
|
!!!nack ('t359.1'); |
| 4567 |
|
!!!next-token; |
| 4568 |
|
next B; |
| 4569 |
|
} elsif ($token->{tag_name} eq 'dt' or |
| 4570 |
|
$token->{tag_name} eq 'dd') { |
| 4571 |
|
## NOTE: As normal, but imply </dt> or </dd> when ... |
| 4572 |
|
|
| 4573 |
|
my $non_optional; |
| 4574 |
|
my $i = -1; |
| 4575 |
|
|
| 4576 |
|
## 1. |
| 4577 |
|
for my $node (reverse @{$self->{open_elements}}) { |
| 4578 |
|
if ($node->[1] == DTDD_EL) { |
| 4579 |
|
## 2. (a) As if </li> |
| 4580 |
|
{ |
| 4581 |
|
## If no </li> - not applied |
| 4582 |
|
# |
| 4583 |
|
|
| 4584 |
|
## Otherwise |
| 4585 |
|
|
| 4586 |
## Ignore the token. |
## 1. generate implied end tags, except for </dt> or </dd> |
| 4587 |
if (length $token->{data}) { |
# |
| 4588 |
## reprocess the rest of characters |
|
| 4589 |
|
## 2. If current node != "dt"|"dd", parse error |
| 4590 |
|
if ($non_optional) { |
| 4591 |
|
!!!parse-error (type => 'not closed', |
| 4592 |
|
text => $non_optional->[0]->manakai_local_name, |
| 4593 |
|
token => $token); |
| 4594 |
|
!!!cp ('t355.1'); |
| 4595 |
} else { |
} else { |
| 4596 |
!!!next-token; |
!!!cp ('t356.1'); |
| 4597 |
} |
} |
| 4598 |
redo B; |
|
| 4599 |
|
## 3. Pop |
| 4600 |
|
splice @{$self->{open_elements}}, $i; |
| 4601 |
} |
} |
| 4602 |
|
|
| 4603 |
die qq[$0: Character "$token->{data}"]; |
last; ## 2. (b) goto 5. |
| 4604 |
} elsif ($token->{type} eq 'start tag') { |
} elsif ( |
| 4605 |
if ($token->{tag_name} eq 'noframes') { |
## NOTE: not "formatting" and not "phrasing" |
| 4606 |
$in_body->($insert_to_current); |
($node->[1] & SPECIAL_EL or |
| 4607 |
redo B; |
$node->[1] & SCOPING_EL) and |
| 4608 |
|
## NOTE: "li", "dt", and "dd" are in |SPECIAL_EL|. |
| 4609 |
|
|
| 4610 |
|
(not $node->[1] & ADDRESS_DIV_P_EL) |
| 4611 |
|
) { |
| 4612 |
|
## 3. |
| 4613 |
|
!!!cp ('t357.1'); |
| 4614 |
|
last; ## goto 5. |
| 4615 |
|
} elsif ($node->[1] & END_TAG_OPTIONAL_EL) { |
| 4616 |
|
!!!cp ('t358.1'); |
| 4617 |
|
# |
| 4618 |
|
} else { |
| 4619 |
|
!!!cp ('t359.1'); |
| 4620 |
|
$non_optional ||= $node; |
| 4621 |
|
# |
| 4622 |
|
} |
| 4623 |
|
## 4. |
| 4624 |
|
## goto 2. |
| 4625 |
|
$i--; |
| 4626 |
|
} |
| 4627 |
|
|
| 4628 |
|
## 5. (a) has a |p| element in scope |
| 4629 |
|
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4630 |
|
if ($_->[1] == P_EL) { |
| 4631 |
|
!!!cp ('t353.1'); |
| 4632 |
|
!!!back-token; # <x> |
| 4633 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4634 |
|
line => $token->{line}, column => $token->{column}}; |
| 4635 |
|
next B; |
| 4636 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 4637 |
|
!!!cp ('t354.1'); |
| 4638 |
|
last INSCOPE; |
| 4639 |
|
} |
| 4640 |
|
} # INSCOPE |
| 4641 |
|
|
| 4642 |
|
## 5. (b) insert |
| 4643 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4644 |
|
!!!nack ('t359.2'); |
| 4645 |
|
!!!next-token; |
| 4646 |
|
next B; |
| 4647 |
|
} elsif ($token->{tag_name} eq 'plaintext') { |
| 4648 |
|
## NOTE: As normal, but effectively ends parsing |
| 4649 |
|
|
| 4650 |
|
## has a p element in scope |
| 4651 |
|
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4652 |
|
if ($_->[1] == P_EL) { |
| 4653 |
|
!!!cp ('t367'); |
| 4654 |
|
!!!back-token; # <plaintext> |
| 4655 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4656 |
|
line => $token->{line}, column => $token->{column}}; |
| 4657 |
|
next B; |
| 4658 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 4659 |
|
!!!cp ('t368'); |
| 4660 |
|
last INSCOPE; |
| 4661 |
|
} |
| 4662 |
|
} # INSCOPE |
| 4663 |
|
|
| 4664 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4665 |
|
|
| 4666 |
|
$self->{content_model} = PLAINTEXT_CONTENT_MODEL; |
| 4667 |
|
|
| 4668 |
|
!!!nack ('t368.1'); |
| 4669 |
|
!!!next-token; |
| 4670 |
|
next B; |
| 4671 |
|
} elsif ($token->{tag_name} eq 'a') { |
| 4672 |
|
AFE: for my $i (reverse 0..$#$active_formatting_elements) { |
| 4673 |
|
my $node = $active_formatting_elements->[$i]; |
| 4674 |
|
if ($node->[1] == A_EL) { |
| 4675 |
|
!!!cp ('t371'); |
| 4676 |
|
!!!parse-error (type => 'in a:a', token => $token); |
| 4677 |
|
|
| 4678 |
|
!!!back-token; # <a> |
| 4679 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'a', |
| 4680 |
|
line => $token->{line}, column => $token->{column}}; |
| 4681 |
|
$formatting_end_tag->($token); |
| 4682 |
|
|
| 4683 |
|
AFE2: for (reverse 0..$#$active_formatting_elements) { |
| 4684 |
|
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
| 4685 |
|
!!!cp ('t372'); |
| 4686 |
|
splice @$active_formatting_elements, $_, 1; |
| 4687 |
|
last AFE2; |
| 4688 |
|
} |
| 4689 |
|
} # AFE2 |
| 4690 |
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4691 |
|
if ($self->{open_elements}->[$_]->[0] eq $node->[0]) { |
| 4692 |
|
!!!cp ('t373'); |
| 4693 |
|
splice @{$self->{open_elements}}, $_, 1; |
| 4694 |
|
last OE; |
| 4695 |
|
} |
| 4696 |
|
} # OE |
| 4697 |
|
last AFE; |
| 4698 |
|
} elsif ($node->[0] eq '#marker') { |
| 4699 |
|
!!!cp ('t374'); |
| 4700 |
|
last AFE; |
| 4701 |
|
} |
| 4702 |
|
} # AFE |
| 4703 |
|
|
| 4704 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4705 |
|
|
| 4706 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4707 |
|
push @$active_formatting_elements, $self->{open_elements}->[-1]; |
| 4708 |
|
|
| 4709 |
|
!!!nack ('t374.1'); |
| 4710 |
|
!!!next-token; |
| 4711 |
|
next B; |
| 4712 |
|
} elsif ($token->{tag_name} eq 'nobr') { |
| 4713 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4714 |
|
|
| 4715 |
|
## has a |nobr| element in scope |
| 4716 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4717 |
|
my $node = $self->{open_elements}->[$_]; |
| 4718 |
|
if ($node->[1] == NOBR_EL) { |
| 4719 |
|
!!!cp ('t376'); |
| 4720 |
|
!!!parse-error (type => 'in nobr:nobr', token => $token); |
| 4721 |
|
!!!back-token; # <nobr> |
| 4722 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'nobr', |
| 4723 |
|
line => $token->{line}, column => $token->{column}}; |
| 4724 |
|
next B; |
| 4725 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 4726 |
|
!!!cp ('t377'); |
| 4727 |
|
last INSCOPE; |
| 4728 |
|
} |
| 4729 |
|
} # INSCOPE |
| 4730 |
|
|
| 4731 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4732 |
|
push @$active_formatting_elements, $self->{open_elements}->[-1]; |
| 4733 |
|
|
| 4734 |
|
!!!nack ('t377.1'); |
| 4735 |
|
!!!next-token; |
| 4736 |
|
next B; |
| 4737 |
|
} elsif ($token->{tag_name} eq 'button') { |
| 4738 |
|
## has a button element in scope |
| 4739 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4740 |
|
my $node = $self->{open_elements}->[$_]; |
| 4741 |
|
if ($node->[1] == BUTTON_EL) { |
| 4742 |
|
!!!cp ('t378'); |
| 4743 |
|
!!!parse-error (type => 'in button:button', token => $token); |
| 4744 |
|
!!!back-token; # <button> |
| 4745 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'button', |
| 4746 |
|
line => $token->{line}, column => $token->{column}}; |
| 4747 |
|
next B; |
| 4748 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 4749 |
|
!!!cp ('t379'); |
| 4750 |
|
last INSCOPE; |
| 4751 |
|
} |
| 4752 |
|
} # INSCOPE |
| 4753 |
|
|
| 4754 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4755 |
|
|
| 4756 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4757 |
|
|
| 4758 |
|
## TODO: associate with $self->{form_element} if defined |
| 4759 |
|
|
| 4760 |
|
push @$active_formatting_elements, ['#marker', '']; |
| 4761 |
|
|
| 4762 |
|
!!!nack ('t379.1'); |
| 4763 |
|
!!!next-token; |
| 4764 |
|
next B; |
| 4765 |
|
} elsif ({ |
| 4766 |
|
xmp => 1, |
| 4767 |
|
iframe => 1, |
| 4768 |
|
noembed => 1, |
| 4769 |
|
noframes => 1, ## NOTE: This is an "as if in head" code clone. |
| 4770 |
|
noscript => 0, ## TODO: 1 if scripting is enabled |
| 4771 |
|
}->{$token->{tag_name}}) { |
| 4772 |
|
if ($token->{tag_name} eq 'xmp') { |
| 4773 |
|
!!!cp ('t381'); |
| 4774 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4775 |
} else { |
} else { |
| 4776 |
!!!parse-error (type => 'after frameset:'.$token->{tag_name}); |
!!!cp ('t399'); |
| 4777 |
|
} |
| 4778 |
|
## NOTE: There is an "as if in body" code clone. |
| 4779 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 4780 |
|
next B; |
| 4781 |
|
} elsif ($token->{tag_name} eq 'isindex') { |
| 4782 |
|
!!!parse-error (type => 'isindex', token => $token); |
| 4783 |
|
|
| 4784 |
|
if (defined $self->{form_element}) { |
| 4785 |
|
!!!cp ('t389'); |
| 4786 |
## Ignore the token |
## Ignore the token |
| 4787 |
|
!!!nack ('t389'); ## NOTE: Not acknowledged. |
| 4788 |
|
!!!next-token; |
| 4789 |
|
next B; |
| 4790 |
|
} else { |
| 4791 |
|
!!!ack ('t391.1'); |
| 4792 |
|
|
| 4793 |
|
my $at = $token->{attributes}; |
| 4794 |
|
my $form_attrs; |
| 4795 |
|
$form_attrs->{action} = $at->{action} if $at->{action}; |
| 4796 |
|
my $prompt_attr = $at->{prompt}; |
| 4797 |
|
$at->{name} = {name => 'name', value => 'isindex'}; |
| 4798 |
|
delete $at->{action}; |
| 4799 |
|
delete $at->{prompt}; |
| 4800 |
|
my @tokens = ( |
| 4801 |
|
{type => START_TAG_TOKEN, tag_name => 'form', |
| 4802 |
|
attributes => $form_attrs, |
| 4803 |
|
line => $token->{line}, column => $token->{column}}, |
| 4804 |
|
{type => START_TAG_TOKEN, tag_name => 'hr', |
| 4805 |
|
line => $token->{line}, column => $token->{column}}, |
| 4806 |
|
{type => START_TAG_TOKEN, tag_name => 'label', |
| 4807 |
|
line => $token->{line}, column => $token->{column}}, |
| 4808 |
|
); |
| 4809 |
|
if ($prompt_attr) { |
| 4810 |
|
!!!cp ('t390'); |
| 4811 |
|
push @tokens, {type => CHARACTER_TOKEN, data => $prompt_attr->{value}, |
| 4812 |
|
#line => $token->{line}, column => $token->{column}, |
| 4813 |
|
}; |
| 4814 |
|
} else { |
| 4815 |
|
!!!cp ('t391'); |
| 4816 |
|
push @tokens, {type => CHARACTER_TOKEN, |
| 4817 |
|
data => 'This is a searchable index. Insert your search keywords here: ', |
| 4818 |
|
#line => $token->{line}, column => $token->{column}, |
| 4819 |
|
}; # SHOULD |
| 4820 |
|
## TODO: make this configurable |
| 4821 |
|
} |
| 4822 |
|
push @tokens, |
| 4823 |
|
{type => START_TAG_TOKEN, tag_name => 'input', attributes => $at, |
| 4824 |
|
line => $token->{line}, column => $token->{column}}, |
| 4825 |
|
#{type => CHARACTER_TOKEN, data => ''}, # SHOULD |
| 4826 |
|
{type => END_TAG_TOKEN, tag_name => 'label', |
| 4827 |
|
line => $token->{line}, column => $token->{column}}, |
| 4828 |
|
{type => START_TAG_TOKEN, tag_name => 'hr', |
| 4829 |
|
line => $token->{line}, column => $token->{column}}, |
| 4830 |
|
{type => END_TAG_TOKEN, tag_name => 'form', |
| 4831 |
|
line => $token->{line}, column => $token->{column}}; |
| 4832 |
|
!!!back-token (@tokens); |
| 4833 |
!!!next-token; |
!!!next-token; |
| 4834 |
redo B; |
next B; |
| 4835 |
} |
} |
| 4836 |
} elsif ($token->{type} eq 'end tag') { |
} elsif ($token->{tag_name} eq 'textarea') { |
| 4837 |
if ($token->{tag_name} eq 'html') { |
## Step 1 |
| 4838 |
$previous_insertion_mode = $self->{insertion_mode}; |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4839 |
$self->{insertion_mode} = 'trailing end'; |
|
| 4840 |
|
## Step 2 |
| 4841 |
|
## TODO: $self->{form_element} if defined |
| 4842 |
|
|
| 4843 |
|
## Step 3 |
| 4844 |
|
$self->{ignore_newline} = 1; |
| 4845 |
|
|
| 4846 |
|
## Step 4 |
| 4847 |
|
## ISSUE: This step is wrong. (r2302 enbugged) |
| 4848 |
|
|
| 4849 |
|
## Step 5 |
| 4850 |
|
$self->{content_model} = RCDATA_CONTENT_MODEL; |
| 4851 |
|
delete $self->{escape}; # MUST |
| 4852 |
|
|
| 4853 |
|
## Step 6-7 |
| 4854 |
|
$self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
| 4855 |
|
|
| 4856 |
|
!!!nack ('t392.1'); |
| 4857 |
|
!!!next-token; |
| 4858 |
|
next B; |
| 4859 |
|
} elsif ($token->{tag_name} eq 'optgroup' or |
| 4860 |
|
$token->{tag_name} eq 'option') { |
| 4861 |
|
## has an |option| element in scope |
| 4862 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4863 |
|
my $node = $self->{open_elements}->[$_]; |
| 4864 |
|
if ($node->[1] == OPTION_EL) { |
| 4865 |
|
!!!cp ('t397.1'); |
| 4866 |
|
## NOTE: As if </option> |
| 4867 |
|
!!!back-token; # <option> or <optgroup> |
| 4868 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'option', |
| 4869 |
|
line => $token->{line}, column => $token->{column}}; |
| 4870 |
|
next B; |
| 4871 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 4872 |
|
!!!cp ('t397.2'); |
| 4873 |
|
last INSCOPE; |
| 4874 |
|
} |
| 4875 |
|
} # INSCOPE |
| 4876 |
|
|
| 4877 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4878 |
|
|
| 4879 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4880 |
|
|
| 4881 |
|
!!!nack ('t397.3'); |
| 4882 |
|
!!!next-token; |
| 4883 |
|
redo B; |
| 4884 |
|
} elsif ($token->{tag_name} eq 'rt' or |
| 4885 |
|
$token->{tag_name} eq 'rp') { |
| 4886 |
|
## has a |ruby| element in scope |
| 4887 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4888 |
|
my $node = $self->{open_elements}->[$_]; |
| 4889 |
|
if ($node->[1] == RUBY_EL) { |
| 4890 |
|
!!!cp ('t398.1'); |
| 4891 |
|
## generate implied end tags |
| 4892 |
|
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 4893 |
|
!!!cp ('t398.2'); |
| 4894 |
|
pop @{$self->{open_elements}}; |
| 4895 |
|
} |
| 4896 |
|
unless ($self->{open_elements}->[-1]->[1] == RUBY_EL) { |
| 4897 |
|
!!!cp ('t398.3'); |
| 4898 |
|
!!!parse-error (type => 'not closed', |
| 4899 |
|
text => $self->{open_elements}->[-1]->[0] |
| 4900 |
|
->manakai_local_name, |
| 4901 |
|
token => $token); |
| 4902 |
|
pop @{$self->{open_elements}} |
| 4903 |
|
while not $self->{open_elements}->[-1]->[1] == RUBY_EL; |
| 4904 |
|
} |
| 4905 |
|
last INSCOPE; |
| 4906 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 4907 |
|
!!!cp ('t398.4'); |
| 4908 |
|
last INSCOPE; |
| 4909 |
|
} |
| 4910 |
|
} # INSCOPE |
| 4911 |
|
|
| 4912 |
|
## TODO: <non-ruby><rt> is not allowed. |
| 4913 |
|
|
| 4914 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4915 |
|
|
| 4916 |
|
!!!nack ('t398.5'); |
| 4917 |
|
!!!next-token; |
| 4918 |
|
redo B; |
| 4919 |
|
} elsif ($token->{tag_name} eq 'math' or |
| 4920 |
|
$token->{tag_name} eq 'svg') { |
| 4921 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4922 |
|
|
| 4923 |
|
## "Adjust MathML attributes" ('math' only) - done in insert-element-f |
| 4924 |
|
|
| 4925 |
|
## "adjust SVG attributes" ('svg' only) - done in insert-element-f |
| 4926 |
|
|
| 4927 |
|
## "adjust foreign attributes" - done in insert-element-f |
| 4928 |
|
|
| 4929 |
|
!!!insert-element-f ($token->{tag_name} eq 'math' ? $MML_NS : $SVG_NS, $token->{tag_name}, $token->{attributes}, $token); |
| 4930 |
|
|
| 4931 |
|
if ($self->{self_closing}) { |
| 4932 |
|
pop @{$self->{open_elements}}; |
| 4933 |
|
!!!ack ('t398.6'); |
| 4934 |
|
} else { |
| 4935 |
|
!!!cp ('t398.7'); |
| 4936 |
|
$self->{insertion_mode} |= IN_FOREIGN_CONTENT_IM; |
| 4937 |
|
## NOTE: |<body><math><mi><svg>| -> "in foreign content" insertion |
| 4938 |
|
## mode, "in body" (not "in foreign content") secondary insertion |
| 4939 |
|
## mode, maybe. |
| 4940 |
|
} |
| 4941 |
|
|
| 4942 |
|
!!!next-token; |
| 4943 |
|
next B; |
| 4944 |
|
} elsif ({ |
| 4945 |
|
caption => 1, col => 1, colgroup => 1, frame => 1, |
| 4946 |
|
frameset => 1, head => 1, |
| 4947 |
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
| 4948 |
|
thead => 1, tr => 1, |
| 4949 |
|
}->{$token->{tag_name}}) { |
| 4950 |
|
!!!cp ('t401'); |
| 4951 |
|
!!!parse-error (type => 'in body', |
| 4952 |
|
text => $token->{tag_name}, token => $token); |
| 4953 |
|
## Ignore the token |
| 4954 |
|
!!!nack ('t401.1'); ## NOTE: |<col/>| or |<frame/>| here is an error. |
| 4955 |
|
!!!next-token; |
| 4956 |
|
next B; |
| 4957 |
|
} elsif ($token->{tag_name} eq 'param' or |
| 4958 |
|
$token->{tag_name} eq 'source') { |
| 4959 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4960 |
|
pop @{$self->{open_elements}}; |
| 4961 |
|
|
| 4962 |
|
!!!ack ('t398.5'); |
| 4963 |
|
!!!next-token; |
| 4964 |
|
redo B; |
| 4965 |
|
} else { |
| 4966 |
|
if ($token->{tag_name} eq 'image') { |
| 4967 |
|
!!!cp ('t384'); |
| 4968 |
|
!!!parse-error (type => 'image', token => $token); |
| 4969 |
|
$token->{tag_name} = 'img'; |
| 4970 |
|
} else { |
| 4971 |
|
!!!cp ('t385'); |
| 4972 |
|
} |
| 4973 |
|
|
| 4974 |
|
## NOTE: There is an "as if <br>" code clone. |
| 4975 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4976 |
|
|
| 4977 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4978 |
|
|
| 4979 |
|
if ({ |
| 4980 |
|
applet => 1, marquee => 1, object => 1, |
| 4981 |
|
}->{$token->{tag_name}}) { |
| 4982 |
|
!!!cp ('t380'); |
| 4983 |
|
push @$active_formatting_elements, ['#marker', '']; |
| 4984 |
|
!!!nack ('t380.1'); |
| 4985 |
|
} elsif ({ |
| 4986 |
|
b => 1, big => 1, em => 1, font => 1, i => 1, |
| 4987 |
|
s => 1, small => 1, strike => 1, |
| 4988 |
|
strong => 1, tt => 1, u => 1, |
| 4989 |
|
}->{$token->{tag_name}}) { |
| 4990 |
|
!!!cp ('t375'); |
| 4991 |
|
push @$active_formatting_elements, $self->{open_elements}->[-1]; |
| 4992 |
|
!!!nack ('t375.1'); |
| 4993 |
|
} elsif ($token->{tag_name} eq 'input') { |
| 4994 |
|
!!!cp ('t388'); |
| 4995 |
|
## TODO: associate with $self->{form_element} if defined |
| 4996 |
|
pop @{$self->{open_elements}}; |
| 4997 |
|
!!!ack ('t388.2'); |
| 4998 |
|
} elsif ({ |
| 4999 |
|
area => 1, basefont => 1, bgsound => 1, br => 1, |
| 5000 |
|
embed => 1, img => 1, spacer => 1, wbr => 1, |
| 5001 |
|
}->{$token->{tag_name}}) { |
| 5002 |
|
!!!cp ('t388.1'); |
| 5003 |
|
pop @{$self->{open_elements}}; |
| 5004 |
|
!!!ack ('t388.3'); |
| 5005 |
|
} elsif ($token->{tag_name} eq 'select') { |
| 5006 |
|
## TODO: associate with $self->{form_element} if defined |
| 5007 |
|
|
| 5008 |
|
if ($self->{insertion_mode} & TABLE_IMS or |
| 5009 |
|
$self->{insertion_mode} & BODY_TABLE_IMS or |
| 5010 |
|
($self->{insertion_mode} & IM_MASK) == IN_COLUMN_GROUP_IM) { |
| 5011 |
|
!!!cp ('t400.1'); |
| 5012 |
|
$self->{insertion_mode} = IN_SELECT_IN_TABLE_IM; |
| 5013 |
|
} else { |
| 5014 |
|
!!!cp ('t400.2'); |
| 5015 |
|
$self->{insertion_mode} = IN_SELECT_IM; |
| 5016 |
|
} |
| 5017 |
|
!!!nack ('t400.3'); |
| 5018 |
|
} else { |
| 5019 |
|
!!!nack ('t402'); |
| 5020 |
|
} |
| 5021 |
|
|
| 5022 |
|
!!!next-token; |
| 5023 |
|
next B; |
| 5024 |
|
} |
| 5025 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 5026 |
|
if ($token->{tag_name} eq 'body') { |
| 5027 |
|
## has a |body| element in scope |
| 5028 |
|
my $i; |
| 5029 |
|
INSCOPE: { |
| 5030 |
|
for (reverse @{$self->{open_elements}}) { |
| 5031 |
|
if ($_->[1] == BODY_EL) { |
| 5032 |
|
!!!cp ('t405'); |
| 5033 |
|
$i = $_; |
| 5034 |
|
last INSCOPE; |
| 5035 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 5036 |
|
!!!cp ('t405.1'); |
| 5037 |
|
last; |
| 5038 |
|
} |
| 5039 |
|
} |
| 5040 |
|
|
| 5041 |
|
## NOTE: |<marquee></body>|, |<svg><foreignobject></body>| |
| 5042 |
|
|
| 5043 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5044 |
|
text => $token->{tag_name}, token => $token); |
| 5045 |
|
## NOTE: Ignore the token. |
| 5046 |
!!!next-token; |
!!!next-token; |
| 5047 |
redo B; |
next B; |
| 5048 |
|
} # INSCOPE |
| 5049 |
|
|
| 5050 |
|
for (@{$self->{open_elements}}) { |
| 5051 |
|
unless ($_->[1] & ALL_END_TAG_OPTIONAL_EL) { |
| 5052 |
|
!!!cp ('t403'); |
| 5053 |
|
!!!parse-error (type => 'not closed', |
| 5054 |
|
text => $_->[0]->manakai_local_name, |
| 5055 |
|
token => $token); |
| 5056 |
|
last; |
| 5057 |
|
} else { |
| 5058 |
|
!!!cp ('t404'); |
| 5059 |
|
} |
| 5060 |
|
} |
| 5061 |
|
|
| 5062 |
|
$self->{insertion_mode} = AFTER_BODY_IM; |
| 5063 |
|
!!!next-token; |
| 5064 |
|
next B; |
| 5065 |
|
} elsif ($token->{tag_name} eq 'html') { |
| 5066 |
|
## TODO: Update this code. It seems that the code below is not |
| 5067 |
|
## up-to-date, though it has same effect as speced. |
| 5068 |
|
if (@{$self->{open_elements}} > 1 and |
| 5069 |
|
$self->{open_elements}->[1]->[1] == BODY_EL) { |
| 5070 |
|
unless ($self->{open_elements}->[-1]->[1] == BODY_EL) { |
| 5071 |
|
!!!cp ('t406'); |
| 5072 |
|
!!!parse-error (type => 'not closed', |
| 5073 |
|
text => $self->{open_elements}->[1]->[0] |
| 5074 |
|
->manakai_local_name, |
| 5075 |
|
token => $token); |
| 5076 |
|
} else { |
| 5077 |
|
!!!cp ('t407'); |
| 5078 |
|
} |
| 5079 |
|
$self->{insertion_mode} = AFTER_BODY_IM; |
| 5080 |
|
## reprocess |
| 5081 |
|
next B; |
| 5082 |
} else { |
} else { |
| 5083 |
!!!parse-error (type => 'after frameset:/'.$token->{tag_name}); |
!!!cp ('t408'); |
| 5084 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5085 |
|
text => $token->{tag_name}, token => $token); |
| 5086 |
## Ignore the token |
## Ignore the token |
| 5087 |
!!!next-token; |
!!!next-token; |
| 5088 |
redo B; |
next B; |
| 5089 |
} |
} |
| 5090 |
} else { |
} elsif ({ |
| 5091 |
die "$0: $token->{type}: Unknown token type"; |
## NOTE: End tags for non-phrasing flow content elements |
|
} |
|
| 5092 |
|
|
| 5093 |
## ISSUE: An issue in spec here |
## NOTE: The normal ones |
| 5094 |
} elsif ($self->{insertion_mode} eq 'trailing end') { |
address => 1, article => 1, aside => 1, blockquote => 1, |
| 5095 |
## states in the main stage is preserved yet # MUST |
center => 1, datagrid => 1, details => 1, dialog => 1, |
| 5096 |
|
dir => 1, div => 1, dl => 1, fieldset => 1, figure => 1, |
| 5097 |
if ($token->{type} eq 'character') { |
footer => 1, header => 1, listing => 1, menu => 1, nav => 1, |
| 5098 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
ol => 1, pre => 1, section => 1, ul => 1, |
| 5099 |
my $data = $1; |
|
| 5100 |
## As if in the main phase. |
## NOTE: As normal, but ... optional tags |
| 5101 |
## NOTE: The insertion mode in the main phase |
dd => 1, dt => 1, li => 1, |
| 5102 |
## just before the phase has been changed to the trailing |
|
| 5103 |
## end phase is either "after body" or "after frameset". |
applet => 1, button => 1, marquee => 1, object => 1, |
| 5104 |
$reconstruct_active_formatting_elements->($insert_to_current); |
}->{$token->{tag_name}}) { |
| 5105 |
|
## NOTE: Code for <li> start tags includes "as if </li>" code. |
| 5106 |
|
## Code for <dt> or <dd> start tags includes "as if </dt> or |
| 5107 |
|
## </dd>" code. |
| 5108 |
|
|
| 5109 |
|
## has an element in scope |
| 5110 |
|
my $i; |
| 5111 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5112 |
|
my $node = $self->{open_elements}->[$_]; |
| 5113 |
|
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
| 5114 |
|
!!!cp ('t410'); |
| 5115 |
|
$i = $_; |
| 5116 |
|
last INSCOPE; |
| 5117 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 5118 |
|
!!!cp ('t411'); |
| 5119 |
|
last INSCOPE; |
| 5120 |
|
} |
| 5121 |
|
} # INSCOPE |
| 5122 |
|
|
| 5123 |
|
unless (defined $i) { # has an element in scope |
| 5124 |
|
!!!cp ('t413'); |
| 5125 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5126 |
|
text => $token->{tag_name}, token => $token); |
| 5127 |
|
## NOTE: Ignore the token. |
| 5128 |
|
} else { |
| 5129 |
|
## Step 1. generate implied end tags |
| 5130 |
|
while ({ |
| 5131 |
|
## END_TAG_OPTIONAL_EL |
| 5132 |
|
dd => ($token->{tag_name} ne 'dd'), |
| 5133 |
|
dt => ($token->{tag_name} ne 'dt'), |
| 5134 |
|
li => ($token->{tag_name} ne 'li'), |
| 5135 |
|
option => 1, |
| 5136 |
|
optgroup => 1, |
| 5137 |
|
p => 1, |
| 5138 |
|
rt => 1, |
| 5139 |
|
rp => 1, |
| 5140 |
|
}->{$self->{open_elements}->[-1]->[0]->manakai_local_name}) { |
| 5141 |
|
!!!cp ('t409'); |
| 5142 |
|
pop @{$self->{open_elements}}; |
| 5143 |
|
} |
| 5144 |
|
|
| 5145 |
|
## Step 2. |
| 5146 |
|
if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
| 5147 |
|
ne $token->{tag_name}) { |
| 5148 |
|
!!!cp ('t412'); |
| 5149 |
|
!!!parse-error (type => 'not closed', |
| 5150 |
|
text => $self->{open_elements}->[-1]->[0] |
| 5151 |
|
->manakai_local_name, |
| 5152 |
|
token => $token); |
| 5153 |
|
} else { |
| 5154 |
|
!!!cp ('t414'); |
| 5155 |
|
} |
| 5156 |
|
|
| 5157 |
|
## Step 3. |
| 5158 |
|
splice @{$self->{open_elements}}, $i; |
| 5159 |
|
|
| 5160 |
|
## Step 4. |
| 5161 |
|
$clear_up_to_marker->() |
| 5162 |
|
if { |
| 5163 |
|
applet => 1, button => 1, marquee => 1, object => 1, |
| 5164 |
|
}->{$token->{tag_name}}; |
| 5165 |
|
} |
| 5166 |
|
!!!next-token; |
| 5167 |
|
next B; |
| 5168 |
|
} elsif ($token->{tag_name} eq 'form') { |
| 5169 |
|
## NOTE: As normal, but interacts with the form element pointer |
| 5170 |
|
|
| 5171 |
|
undef $self->{form_element}; |
| 5172 |
|
|
| 5173 |
|
## has an element in scope |
| 5174 |
|
my $i; |
| 5175 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5176 |
|
my $node = $self->{open_elements}->[$_]; |
| 5177 |
|
if ($node->[1] == FORM_EL) { |
| 5178 |
|
!!!cp ('t418'); |
| 5179 |
|
$i = $_; |
| 5180 |
|
last INSCOPE; |
| 5181 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 5182 |
|
!!!cp ('t419'); |
| 5183 |
|
last INSCOPE; |
| 5184 |
|
} |
| 5185 |
|
} # INSCOPE |
| 5186 |
|
|
| 5187 |
|
unless (defined $i) { # has an element in scope |
| 5188 |
|
!!!cp ('t421'); |
| 5189 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5190 |
|
text => $token->{tag_name}, token => $token); |
| 5191 |
|
## NOTE: Ignore the token. |
| 5192 |
|
} else { |
| 5193 |
|
## Step 1. generate implied end tags |
| 5194 |
|
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 5195 |
|
!!!cp ('t417'); |
| 5196 |
|
pop @{$self->{open_elements}}; |
| 5197 |
|
} |
| 5198 |
|
|
| 5199 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($data); |
## Step 2. |
| 5200 |
|
if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
| 5201 |
|
ne $token->{tag_name}) { |
| 5202 |
|
!!!cp ('t417.1'); |
| 5203 |
|
!!!parse-error (type => 'not closed', |
| 5204 |
|
text => $self->{open_elements}->[-1]->[0] |
| 5205 |
|
->manakai_local_name, |
| 5206 |
|
token => $token); |
| 5207 |
|
} else { |
| 5208 |
|
!!!cp ('t420'); |
| 5209 |
|
} |
| 5210 |
|
|
| 5211 |
unless (length $token->{data}) { |
## Step 3. |
| 5212 |
!!!next-token; |
splice @{$self->{open_elements}}, $i; |
| 5213 |
redo B; |
} |
| 5214 |
|
|
| 5215 |
|
!!!next-token; |
| 5216 |
|
next B; |
| 5217 |
|
} elsif ({ |
| 5218 |
|
## NOTE: As normal, except acts as a closer for any ... |
| 5219 |
|
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
| 5220 |
|
}->{$token->{tag_name}}) { |
| 5221 |
|
## has an element in scope |
| 5222 |
|
my $i; |
| 5223 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5224 |
|
my $node = $self->{open_elements}->[$_]; |
| 5225 |
|
if ($node->[1] == HEADING_EL) { |
| 5226 |
|
!!!cp ('t423'); |
| 5227 |
|
$i = $_; |
| 5228 |
|
last INSCOPE; |
| 5229 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 5230 |
|
!!!cp ('t424'); |
| 5231 |
|
last INSCOPE; |
| 5232 |
|
} |
| 5233 |
|
} # INSCOPE |
| 5234 |
|
|
| 5235 |
|
unless (defined $i) { # has an element in scope |
| 5236 |
|
!!!cp ('t425.1'); |
| 5237 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5238 |
|
text => $token->{tag_name}, token => $token); |
| 5239 |
|
## NOTE: Ignore the token. |
| 5240 |
|
} else { |
| 5241 |
|
## Step 1. generate implied end tags |
| 5242 |
|
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 5243 |
|
!!!cp ('t422'); |
| 5244 |
|
pop @{$self->{open_elements}}; |
| 5245 |
} |
} |
| 5246 |
|
|
| 5247 |
|
## Step 2. |
| 5248 |
|
if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
| 5249 |
|
ne $token->{tag_name}) { |
| 5250 |
|
!!!cp ('t425'); |
| 5251 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5252 |
|
text => $token->{tag_name}, token => $token); |
| 5253 |
|
} else { |
| 5254 |
|
!!!cp ('t426'); |
| 5255 |
|
} |
| 5256 |
|
|
| 5257 |
|
## Step 3. |
| 5258 |
|
splice @{$self->{open_elements}}, $i; |
| 5259 |
} |
} |
| 5260 |
|
|
| 5261 |
|
!!!next-token; |
| 5262 |
|
next B; |
| 5263 |
|
} elsif ($token->{tag_name} eq 'p') { |
| 5264 |
|
## NOTE: As normal, except </p> implies <p> and ... |
| 5265 |
|
|
| 5266 |
!!!parse-error (type => 'after html:#character'); |
## has an element in scope |
| 5267 |
$self->{insertion_mode} = $previous_insertion_mode; |
my $non_optional; |
| 5268 |
## reprocess |
my $i; |
| 5269 |
redo B; |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5270 |
} elsif ($token->{type} eq 'start tag') { |
my $node = $self->{open_elements}->[$_]; |
| 5271 |
!!!parse-error (type => 'after html:'.$token->{tag_name}); |
if ($node->[1] == P_EL) { |
| 5272 |
$self->{insertion_mode} = $previous_insertion_mode; |
!!!cp ('t410.1'); |
| 5273 |
## reprocess |
$i = $_; |
| 5274 |
redo B; |
last INSCOPE; |
| 5275 |
} elsif ($token->{type} eq 'end tag') { |
} elsif ($node->[1] & SCOPING_EL) { |
| 5276 |
!!!parse-error (type => 'after html:/'.$token->{tag_name}); |
!!!cp ('t411.1'); |
| 5277 |
$self->{insertion_mode} = $previous_insertion_mode; |
last INSCOPE; |
| 5278 |
## reprocess |
} elsif ($node->[1] & END_TAG_OPTIONAL_EL) { |
| 5279 |
redo B; |
## NOTE: |END_TAG_OPTIONAL_EL| includes "p" |
| 5280 |
|
!!!cp ('t411.2'); |
| 5281 |
|
# |
| 5282 |
|
} else { |
| 5283 |
|
!!!cp ('t411.3'); |
| 5284 |
|
$non_optional ||= $node; |
| 5285 |
|
# |
| 5286 |
|
} |
| 5287 |
|
} # INSCOPE |
| 5288 |
|
|
| 5289 |
|
if (defined $i) { |
| 5290 |
|
## 1. Generate implied end tags |
| 5291 |
|
# |
| 5292 |
|
|
| 5293 |
|
## 2. If current node != "p", parse error |
| 5294 |
|
if ($non_optional) { |
| 5295 |
|
!!!cp ('t412.1'); |
| 5296 |
|
!!!parse-error (type => 'not closed', |
| 5297 |
|
text => $non_optional->[0]->manakai_local_name, |
| 5298 |
|
token => $token); |
| 5299 |
|
} else { |
| 5300 |
|
!!!cp ('t414.1'); |
| 5301 |
|
} |
| 5302 |
|
|
| 5303 |
|
## 3. Pop |
| 5304 |
|
splice @{$self->{open_elements}}, $i; |
| 5305 |
|
} else { |
| 5306 |
|
!!!cp ('t413.1'); |
| 5307 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5308 |
|
text => $token->{tag_name}, token => $token); |
| 5309 |
|
|
| 5310 |
|
!!!cp ('t415.1'); |
| 5311 |
|
## As if <p>, then reprocess the current token |
| 5312 |
|
my $el; |
| 5313 |
|
!!!create-element ($el, $HTML_NS, 'p',, $token); |
| 5314 |
|
$insert->($el); |
| 5315 |
|
## NOTE: Not inserted into |$self->{open_elements}|. |
| 5316 |
|
} |
| 5317 |
|
|
| 5318 |
|
!!!next-token; |
| 5319 |
|
next B; |
| 5320 |
|
} elsif ({ |
| 5321 |
|
a => 1, |
| 5322 |
|
b => 1, big => 1, em => 1, font => 1, i => 1, |
| 5323 |
|
nobr => 1, s => 1, small => 1, strike => 1, |
| 5324 |
|
strong => 1, tt => 1, u => 1, |
| 5325 |
|
}->{$token->{tag_name}}) { |
| 5326 |
|
!!!cp ('t427'); |
| 5327 |
|
$formatting_end_tag->($token); |
| 5328 |
|
next B; |
| 5329 |
|
} elsif ($token->{tag_name} eq 'br') { |
| 5330 |
|
!!!cp ('t428'); |
| 5331 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5332 |
|
text => 'br', token => $token); |
| 5333 |
|
|
| 5334 |
|
## As if <br> |
| 5335 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 5336 |
|
|
| 5337 |
|
my $el; |
| 5338 |
|
!!!create-element ($el, $HTML_NS, 'br',, $token); |
| 5339 |
|
$insert->($el); |
| 5340 |
|
|
| 5341 |
|
## Ignore the token. |
| 5342 |
|
!!!next-token; |
| 5343 |
|
next B; |
| 5344 |
} else { |
} else { |
| 5345 |
die "$0: $token->{type}: Unknown token"; |
if ($token->{tag_name} eq 'sarcasm') { |
| 5346 |
|
sleep 0.001; # take a deep breath |
| 5347 |
|
} |
| 5348 |
|
|
| 5349 |
|
## Step 1 |
| 5350 |
|
my $node_i = -1; |
| 5351 |
|
my $node = $self->{open_elements}->[$node_i]; |
| 5352 |
|
|
| 5353 |
|
## Step 2 |
| 5354 |
|
S2: { |
| 5355 |
|
my $node_tag_name = $node->[0]->manakai_local_name; |
| 5356 |
|
$node_tag_name =~ tr/A-Z/a-z/; # for SVG camelCase tag names |
| 5357 |
|
if ($node_tag_name eq $token->{tag_name}) { |
| 5358 |
|
## Step 1 |
| 5359 |
|
## generate implied end tags |
| 5360 |
|
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 5361 |
|
!!!cp ('t430'); |
| 5362 |
|
## NOTE: |<ruby><rt></ruby>|. |
| 5363 |
|
## ISSUE: <ruby><rt></rt> will also take this code path, |
| 5364 |
|
## which seems wrong. |
| 5365 |
|
pop @{$self->{open_elements}}; |
| 5366 |
|
$node_i++; |
| 5367 |
|
} |
| 5368 |
|
|
| 5369 |
|
## Step 2 |
| 5370 |
|
my $current_tag_name |
| 5371 |
|
= $self->{open_elements}->[-1]->[0]->manakai_local_name; |
| 5372 |
|
$current_tag_name =~ tr/A-Z/a-z/; |
| 5373 |
|
if ($current_tag_name ne $token->{tag_name}) { |
| 5374 |
|
!!!cp ('t431'); |
| 5375 |
|
## NOTE: <x><y></x> |
| 5376 |
|
!!!parse-error (type => 'not closed', |
| 5377 |
|
text => $self->{open_elements}->[-1]->[0] |
| 5378 |
|
->manakai_local_name, |
| 5379 |
|
token => $token); |
| 5380 |
|
} else { |
| 5381 |
|
!!!cp ('t432'); |
| 5382 |
|
} |
| 5383 |
|
|
| 5384 |
|
## Step 3 |
| 5385 |
|
splice @{$self->{open_elements}}, $node_i if $node_i < 0; |
| 5386 |
|
|
| 5387 |
|
!!!next-token; |
| 5388 |
|
last S2; |
| 5389 |
|
} else { |
| 5390 |
|
## Step 3 |
| 5391 |
|
if (not ($node->[1] & FORMATTING_EL) and |
| 5392 |
|
#not $phrasing_category->{$node->[1]} and |
| 5393 |
|
($node->[1] & SPECIAL_EL or |
| 5394 |
|
$node->[1] & SCOPING_EL)) { |
| 5395 |
|
!!!cp ('t433'); |
| 5396 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5397 |
|
text => $token->{tag_name}, token => $token); |
| 5398 |
|
## Ignore the token |
| 5399 |
|
!!!next-token; |
| 5400 |
|
last S2; |
| 5401 |
|
|
| 5402 |
|
## NOTE: |<span><dd></span>a|: In Safari 3.1.2 and Opera |
| 5403 |
|
## 9.27, "a" is a child of <dd> (conforming). In |
| 5404 |
|
## Firefox 3.0.2, "a" is a child of <body>. In WinIE 7, |
| 5405 |
|
## "a" is a child of both <body> and <dd>. |
| 5406 |
|
} |
| 5407 |
|
|
| 5408 |
|
!!!cp ('t434'); |
| 5409 |
|
} |
| 5410 |
|
|
| 5411 |
|
## Step 4 |
| 5412 |
|
$node_i--; |
| 5413 |
|
$node = $self->{open_elements}->[$node_i]; |
| 5414 |
|
|
| 5415 |
|
## Step 5; |
| 5416 |
|
redo S2; |
| 5417 |
|
} # S2 |
| 5418 |
|
next B; |
| 5419 |
} |
} |
| 5420 |
} else { |
} |
| 5421 |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
next B; |
| 5422 |
|
} continue { # B |
| 5423 |
|
if ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM) { |
| 5424 |
|
## NOTE: The code below is executed in cases where it does not have |
| 5425 |
|
## to be, but it it is harmless even in those cases. |
| 5426 |
|
## has an element in scope |
| 5427 |
|
INSCOPE: { |
| 5428 |
|
for (reverse 0..$#{$self->{open_elements}}) { |
| 5429 |
|
my $node = $self->{open_elements}->[$_]; |
| 5430 |
|
if ($node->[1] & FOREIGN_EL) { |
| 5431 |
|
last INSCOPE; |
| 5432 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 5433 |
|
last; |
| 5434 |
|
} |
| 5435 |
|
} |
| 5436 |
|
|
| 5437 |
|
## NOTE: No foreign element in scope. |
| 5438 |
|
$self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
| 5439 |
|
} # INSCOPE |
| 5440 |
} |
} |
| 5441 |
} # B |
} # B |
| 5442 |
|
|
| 5445 |
## TODO: script stuffs |
## TODO: script stuffs |
| 5446 |
} # _tree_construct_main |
} # _tree_construct_main |
| 5447 |
|
|
| 5448 |
sub set_inner_html ($$$) { |
sub set_inner_html ($$$$;$) { |
| 5449 |
my $class = shift; |
my $class = shift; |
| 5450 |
my $node = shift; |
my $node = shift; |
| 5451 |
my $s = \$_[0]; |
#my $s = \$_[0]; |
| 5452 |
my $onerror = $_[1]; |
my $onerror = $_[1]; |
| 5453 |
|
my $get_wrapper = $_[2] || sub ($) { return $_[0] }; |
| 5454 |
|
|
| 5455 |
|
## ISSUE: Should {confident} be true? |
| 5456 |
|
|
| 5457 |
my $nt = $node->node_type; |
my $nt = $node->node_type; |
| 5458 |
if ($nt == 9) { |
if ($nt == 9) { |
| 5469 |
} |
} |
| 5470 |
|
|
| 5471 |
## Step 3, 4, 5 # MUST |
## Step 3, 4, 5 # MUST |
| 5472 |
$class->parse_string ($$s => $node, $onerror); |
$class->parse_char_string ($_[0] => $node, $onerror, $get_wrapper); |
| 5473 |
} elsif ($nt == 1) { |
} elsif ($nt == 1) { |
| 5474 |
## TODO: If non-html element |
## TODO: If non-html element |
| 5475 |
|
|
| 5476 |
## NOTE: Most of this code is copied from |parse_string| |
## NOTE: Most of this code is copied from |parse_string| |
| 5477 |
|
|
| 5478 |
|
## TODO: Support for $get_wrapper |
| 5479 |
|
|
| 5480 |
## Step 1 # MUST |
## Step 1 # MUST |
| 5481 |
my $this_doc = $node->owner_document; |
my $this_doc = $node->owner_document; |
| 5482 |
my $doc = $this_doc->implementation->create_document; |
my $doc = $this_doc->implementation->create_document; |
| 5484 |
my $p = $class->new; |
my $p = $class->new; |
| 5485 |
$p->{document} = $doc; |
$p->{document} = $doc; |
| 5486 |
|
|
| 5487 |
## Step 9 # MUST |
## Step 8 # MUST |
| 5488 |
my $i = 0; |
my $i = 0; |
| 5489 |
my $line = 1; |
$p->{line_prev} = $p->{line} = 1; |
| 5490 |
my $column = 0; |
$p->{column_prev} = $p->{column} = 0; |
| 5491 |
$p->{set_next_input_character} = sub { |
require Whatpm::Charset::DecodeHandle; |
| 5492 |
|
my $input = Whatpm::Charset::DecodeHandle::CharString->new (\($_[0])); |
| 5493 |
|
$input = $get_wrapper->($input); |
| 5494 |
|
$p->{set_nc} = sub { |
| 5495 |
my $self = shift; |
my $self = shift; |
| 5496 |
|
|
| 5497 |
pop @{$self->{prev_input_character}}; |
my $char = ''; |
| 5498 |
unshift @{$self->{prev_input_character}}, $self->{next_input_character}; |
if (defined $self->{next_nc}) { |
| 5499 |
|
$char = $self->{next_nc}; |
| 5500 |
|
delete $self->{next_nc}; |
| 5501 |
|
$self->{nc} = ord $char; |
| 5502 |
|
} else { |
| 5503 |
|
$self->{char_buffer} = ''; |
| 5504 |
|
$self->{char_buffer_pos} = 0; |
| 5505 |
|
|
| 5506 |
|
my $count = $input->manakai_read_until |
| 5507 |
|
($self->{char_buffer}, qr/[^\x00\x0A\x0D]/, |
| 5508 |
|
$self->{char_buffer_pos}); |
| 5509 |
|
if ($count) { |
| 5510 |
|
$self->{line_prev} = $self->{line}; |
| 5511 |
|
$self->{column_prev} = $self->{column}; |
| 5512 |
|
$self->{column}++; |
| 5513 |
|
$self->{nc} |
| 5514 |
|
= ord substr ($self->{char_buffer}, |
| 5515 |
|
$self->{char_buffer_pos}++, 1); |
| 5516 |
|
return; |
| 5517 |
|
} |
| 5518 |
|
|
| 5519 |
|
if ($input->read ($char, 1)) { |
| 5520 |
|
$self->{nc} = ord $char; |
| 5521 |
|
} else { |
| 5522 |
|
$self->{nc} = -1; |
| 5523 |
|
return; |
| 5524 |
|
} |
| 5525 |
|
} |
| 5526 |
|
|
| 5527 |
|
($p->{line_prev}, $p->{column_prev}) = ($p->{line}, $p->{column}); |
| 5528 |
|
$p->{column}++; |
| 5529 |
|
|
| 5530 |
$self->{next_input_character} = -1 and return if $i >= length $$s; |
if ($self->{nc} == 0x000A) { # LF |
| 5531 |
$self->{next_input_character} = ord substr $$s, $i++, 1; |
$p->{line}++; |
| 5532 |
$column++; |
$p->{column} = 0; |
| 5533 |
|
!!!cp ('i1'); |
| 5534 |
if ($self->{next_input_character} == 0x000A) { # LF |
} elsif ($self->{nc} == 0x000D) { # CR |
| 5535 |
$line++; |
## TODO: support for abort/streaming |
| 5536 |
$column = 0; |
my $next = ''; |
| 5537 |
} elsif ($self->{next_input_character} == 0x000D) { # CR |
if ($input->read ($next, 1) and $next ne "\x0A") { |
| 5538 |
$i++ if substr ($$s, $i, 1) eq "\x0A"; |
$self->{next_nc} = $next; |
| 5539 |
$self->{next_input_character} = 0x000A; # LF # MUST |
} |
| 5540 |
$line++; |
$self->{nc} = 0x000A; # LF # MUST |
| 5541 |
$column = 0; |
$p->{line}++; |
| 5542 |
} elsif ($self->{next_input_character} > 0x10FFFF) { |
$p->{column} = 0; |
| 5543 |
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
!!!cp ('i2'); |
| 5544 |
} elsif ($self->{next_input_character} == 0x0000) { # NULL |
} elsif ($self->{nc} == 0x0000) { # NULL |
| 5545 |
|
!!!cp ('i4'); |
| 5546 |
!!!parse-error (type => 'NULL'); |
!!!parse-error (type => 'NULL'); |
| 5547 |
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
$self->{nc} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 5548 |
} |
} |
| 5549 |
}; |
}; |
| 5550 |
$p->{prev_input_character} = [-1, -1, -1]; |
|
| 5551 |
$p->{next_input_character} = -1; |
$p->{read_until} = sub { |
| 5552 |
|
#my ($scalar, $specials_range, $offset) = @_; |
| 5553 |
|
return 0 if defined $p->{next_nc}; |
| 5554 |
|
|
| 5555 |
|
my $pattern = qr/[^$_[1]\x00\x0A\x0D]/; |
| 5556 |
|
my $offset = $_[2] || 0; |
| 5557 |
|
|
| 5558 |
|
if ($p->{char_buffer_pos} < length $p->{char_buffer}) { |
| 5559 |
|
pos ($p->{char_buffer}) = $p->{char_buffer_pos}; |
| 5560 |
|
if ($p->{char_buffer} =~ /\G(?>$pattern)+/) { |
| 5561 |
|
substr ($_[0], $offset) |
| 5562 |
|
= substr ($p->{char_buffer}, $-[0], $+[0] - $-[0]); |
| 5563 |
|
my $count = $+[0] - $-[0]; |
| 5564 |
|
if ($count) { |
| 5565 |
|
$p->{column} += $count; |
| 5566 |
|
$p->{char_buffer_pos} += $count; |
| 5567 |
|
$p->{line_prev} = $p->{line}; |
| 5568 |
|
$p->{column_prev} = $p->{column} - 1; |
| 5569 |
|
$p->{nc} = -1; |
| 5570 |
|
} |
| 5571 |
|
return $count; |
| 5572 |
|
} else { |
| 5573 |
|
return 0; |
| 5574 |
|
} |
| 5575 |
|
} else { |
| 5576 |
|
my $count = $input->manakai_read_until ($_[0], $pattern, $_[2]); |
| 5577 |
|
if ($count) { |
| 5578 |
|
$p->{column} += $count; |
| 5579 |
|
$p->{column_prev} += $count; |
| 5580 |
|
$p->{nc} = -1; |
| 5581 |
|
} |
| 5582 |
|
return $count; |
| 5583 |
|
} |
| 5584 |
|
}; # $p->{read_until} |
| 5585 |
|
|
| 5586 |
my $ponerror = $onerror || sub { |
my $ponerror = $onerror || sub { |
| 5587 |
my (%opt) = @_; |
my (%opt) = @_; |
| 5588 |
warn "Parse error ($opt{type}) at line $opt{line} column $opt{column}\n"; |
my $line = $opt{line}; |
| 5589 |
|
my $column = $opt{column}; |
| 5590 |
|
if (defined $opt{token} and defined $opt{token}->{line}) { |
| 5591 |
|
$line = $opt{token}->{line}; |
| 5592 |
|
$column = $opt{token}->{column}; |
| 5593 |
|
} |
| 5594 |
|
warn "Parse error ($opt{type}) at line $line column $column\n"; |
| 5595 |
}; |
}; |
| 5596 |
$p->{parse_error} = sub { |
$p->{parse_error} = sub { |
| 5597 |
$ponerror->(@_, line => $line, column => $column); |
$ponerror->(line => $p->{line}, column => $p->{column}, @_); |
| 5598 |
}; |
}; |
| 5599 |
|
|
| 5600 |
|
my $char_onerror = sub { |
| 5601 |
|
my (undef, $type, %opt) = @_; |
| 5602 |
|
$ponerror->(layer => 'encode', |
| 5603 |
|
line => $p->{line}, column => $p->{column} + 1, |
| 5604 |
|
%opt, type => $type); |
| 5605 |
|
}; # $char_onerror |
| 5606 |
|
$input->onerror ($char_onerror); |
| 5607 |
|
|
| 5608 |
$p->_initialize_tokenizer; |
$p->_initialize_tokenizer; |
| 5609 |
$p->_initialize_tree_constructor; |
$p->_initialize_tree_constructor; |
| 5610 |
|
|
| 5611 |
## Step 2 |
## Step 2 |
| 5612 |
my $node_ln = $node->local_name; |
my $node_ln = $node->manakai_local_name; |
| 5613 |
$p->{content_model} = { |
$p->{content_model} = { |
| 5614 |
title => RCDATA_CONTENT_MODEL, |
title => RCDATA_CONTENT_MODEL, |
| 5615 |
textarea => RCDATA_CONTENT_MODEL, |
textarea => RCDATA_CONTENT_MODEL, |
| 5624 |
}->{$node_ln}; |
}->{$node_ln}; |
| 5625 |
$p->{content_model} = PCDATA_CONTENT_MODEL |
$p->{content_model} = PCDATA_CONTENT_MODEL |
| 5626 |
unless defined $p->{content_model}; |
unless defined $p->{content_model}; |
|
## ISSUE: What is "the name of the element"? local name? |
|
| 5627 |
|
|
| 5628 |
$p->{inner_html_node} = [$node, $node_ln]; |
$p->{inner_html_node} = [$node, $el_category->{$node_ln}]; |
| 5629 |
|
## TODO: Foreign element OK? |
| 5630 |
|
|
| 5631 |
## Step 4 |
## Step 3 |
| 5632 |
my $root = $doc->create_element_ns |
my $root = $doc->create_element_ns |
| 5633 |
('http://www.w3.org/1999/xhtml', [undef, 'html']); |
('http://www.w3.org/1999/xhtml', [undef, 'html']); |
| 5634 |
|
|
| 5635 |
## Step 5 # MUST |
## Step 4 # MUST |
| 5636 |
$doc->append_child ($root); |
$doc->append_child ($root); |
| 5637 |
|
|
| 5638 |
## Step 6 # MUST |
## Step 5 # MUST |
| 5639 |
push @{$p->{open_elements}}, [$root, 'html']; |
push @{$p->{open_elements}}, [$root, $el_category->{html}]; |
| 5640 |
|
|
| 5641 |
undef $p->{head_element}; |
undef $p->{head_element}; |
| 5642 |
|
undef $p->{head_element_inserted}; |
| 5643 |
|
|
| 5644 |
## Step 7 # MUST |
## Step 6 # MUST |
| 5645 |
$p->_reset_insertion_mode; |
$p->_reset_insertion_mode; |
| 5646 |
|
|
| 5647 |
## Step 8 # MUST |
## Step 7 # MUST |
| 5648 |
my $anode = $node; |
my $anode = $node; |
| 5649 |
AN: while (defined $anode) { |
AN: while (defined $anode) { |
| 5650 |
if ($anode->node_type == 1) { |
if ($anode->node_type == 1) { |
| 5651 |
my $nsuri = $anode->namespace_uri; |
my $nsuri = $anode->namespace_uri; |
| 5652 |
if (defined $nsuri and $nsuri eq 'http://www.w3.org/1999/xhtml') { |
if (defined $nsuri and $nsuri eq 'http://www.w3.org/1999/xhtml') { |
| 5653 |
if ($anode->local_name eq 'form') { ## TODO: case? |
if ($anode->manakai_local_name eq 'form') { |
| 5654 |
|
!!!cp ('i5'); |
| 5655 |
$p->{form_element} = $anode; |
$p->{form_element} = $anode; |
| 5656 |
last AN; |
last AN; |
| 5657 |
} |
} |
| 5660 |
$anode = $anode->parent_node; |
$anode = $anode->parent_node; |
| 5661 |
} # AN |
} # AN |
| 5662 |
|
|
| 5663 |
## Step 3 # MUST |
## Step 9 # MUST |
|
## Step 10 # MUST |
|
| 5664 |
{ |
{ |
| 5665 |
my $self = $p; |
my $self = $p; |
| 5666 |
!!!next-token; |
!!!next-token; |
| 5667 |
} |
} |
| 5668 |
$p->_tree_construction_main; |
$p->_tree_construction_main; |
| 5669 |
|
|
| 5670 |
## Step 11 # MUST |
## Step 10 # MUST |
| 5671 |
my @cn = @{$node->child_nodes}; |
my @cn = @{$node->child_nodes}; |
| 5672 |
for (@cn) { |
for (@cn) { |
| 5673 |
$node->remove_child ($_); |
$node->remove_child ($_); |
| 5674 |
} |
} |
| 5675 |
## ISSUE: mutation events? read-only? |
## ISSUE: mutation events? read-only? |
| 5676 |
|
|
| 5677 |
## Step 12 # MUST |
## Step 11 # MUST |
| 5678 |
@cn = @{$root->child_nodes}; |
@cn = @{$root->child_nodes}; |
| 5679 |
for (@cn) { |
for (@cn) { |
| 5680 |
$this_doc->adopt_node ($_); |
$this_doc->adopt_node ($_); |
| 5683 |
## ISSUE: mutation events? |
## ISSUE: mutation events? |
| 5684 |
|
|
| 5685 |
$p->_terminate_tree_constructor; |
$p->_terminate_tree_constructor; |
| 5686 |
|
|
| 5687 |
|
delete $p->{parse_error}; # delete loop |
| 5688 |
} else { |
} else { |
| 5689 |
die "$0: |set_inner_html| is not defined for node of type $nt"; |
die "$0: |set_inner_html| is not defined for node of type $nt"; |
| 5690 |
} |
} |
| 5692 |
|
|
| 5693 |
} # tree construction stage |
} # tree construction stage |
| 5694 |
|
|
| 5695 |
sub get_inner_html ($$$) { |
package Whatpm::HTML::RestartParser; |
| 5696 |
my (undef, $node, $on_error) = @_; |
push our @ISA, 'Error'; |
|
|
|
|
## Step 1 |
|
|
my $s = ''; |
|
|
|
|
|
my $in_cdata; |
|
|
my $parent = $node; |
|
|
while (defined $parent) { |
|
|
if ($parent->node_type == 1 and |
|
|
$parent->namespace_uri eq 'http://www.w3.org/1999/xhtml' and |
|
|
{ |
|
|
style => 1, script => 1, xmp => 1, iframe => 1, |
|
|
noembed => 1, noframes => 1, noscript => 1, |
|
|
}->{$parent->local_name}) { ## TODO: case thingy |
|
|
$in_cdata = 1; |
|
|
} |
|
|
$parent = $parent->parent_node; |
|
|
} |
|
|
|
|
|
## Step 2 |
|
|
my @node = @{$node->child_nodes}; |
|
|
C: while (@node) { |
|
|
my $child = shift @node; |
|
|
unless (ref $child) { |
|
|
if ($child eq 'cdata-out') { |
|
|
$in_cdata = 0; |
|
|
} else { |
|
|
$s .= $child; # end tag |
|
|
} |
|
|
next C; |
|
|
} |
|
|
|
|
|
my $nt = $child->node_type; |
|
|
if ($nt == 1) { # Element |
|
|
my $tag_name = $child->tag_name; ## TODO: manakai_tag_name |
|
|
$s .= '<' . $tag_name; |
|
|
## NOTE: Non-HTML case: |
|
|
## <http://permalink.gmane.org/gmane.org.w3c.whatwg.discuss/11191> |
|
|
|
|
|
my @attrs = @{$child->attributes}; # sort order MUST be stable |
|
|
for my $attr (@attrs) { # order is implementation dependent |
|
|
my $attr_name = $attr->name; ## TODO: manakai_name |
|
|
$s .= ' ' . $attr_name . '="'; |
|
|
my $attr_value = $attr->value; |
|
|
## escape |
|
|
$attr_value =~ s/&/&/g; |
|
|
$attr_value =~ s/</</g; |
|
|
$attr_value =~ s/>/>/g; |
|
|
$attr_value =~ s/"/"/g; |
|
|
$s .= $attr_value . '"'; |
|
|
} |
|
|
$s .= '>'; |
|
|
|
|
|
next C if { |
|
|
area => 1, base => 1, basefont => 1, bgsound => 1, |
|
|
br => 1, col => 1, embed => 1, frame => 1, hr => 1, |
|
|
img => 1, input => 1, link => 1, meta => 1, param => 1, |
|
|
spacer => 1, wbr => 1, |
|
|
}->{$tag_name}; |
|
|
|
|
|
$s .= "\x0A" if $tag_name eq 'pre' or $tag_name eq 'textarea'; |
|
|
|
|
|
if (not $in_cdata and { |
|
|
style => 1, script => 1, xmp => 1, iframe => 1, |
|
|
noembed => 1, noframes => 1, noscript => 1, |
|
|
plaintext => 1, |
|
|
}->{$tag_name}) { |
|
|
unshift @node, 'cdata-out'; |
|
|
$in_cdata = 1; |
|
|
} |
|
|
|
|
|
unshift @node, @{$child->child_nodes}, '</' . $tag_name . '>'; |
|
|
} elsif ($nt == 3 or $nt == 4) { |
|
|
if ($in_cdata) { |
|
|
$s .= $child->data; |
|
|
} else { |
|
|
my $value = $child->data; |
|
|
$value =~ s/&/&/g; |
|
|
$value =~ s/</</g; |
|
|
$value =~ s/>/>/g; |
|
|
$value =~ s/"/"/g; |
|
|
$s .= $value; |
|
|
} |
|
|
} elsif ($nt == 8) { |
|
|
$s .= '<!--' . $child->data . '-->'; |
|
|
} elsif ($nt == 10) { |
|
|
$s .= '<!DOCTYPE ' . $child->name . '>'; |
|
|
} elsif ($nt == 5) { # entrefs |
|
|
push @node, @{$child->child_nodes}; |
|
|
} else { |
|
|
$on_error->($child) if defined $on_error; |
|
|
} |
|
|
## ISSUE: This code does not support PIs. |
|
|
} # C |
|
|
|
|
|
## Step 3 |
|
|
return \$s; |
|
|
} # get_inner_html |
|
| 5697 |
|
|
| 5698 |
1; |
1; |
| 5699 |
# $Date$ |
# $Date$ |
Parent Directory
| 
Revision Log
| 
Patch