| 1 |
package Whatpm::HTML; |
| 2 |
use strict; |
| 3 |
our $VERSION=do{my @r=(q$Revision: 1.9 $=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
| 4 |
|
| 5 |
## This is an early version of an HTML parser. |
| 6 |
|
| 7 |
my $permitted_slash_tag_name = { |
| 8 |
base => 1, |
| 9 |
link => 1, |
| 10 |
meta => 1, |
| 11 |
hr => 1, |
| 12 |
br => 1, |
| 13 |
img=> 1, |
| 14 |
embed => 1, |
| 15 |
param => 1, |
| 16 |
area => 1, |
| 17 |
col => 1, |
| 18 |
input => 1, |
| 19 |
}; |
| 20 |
|
| 21 |
my $entity_char = { |
| 22 |
AElig => "\x{00C6}", |
| 23 |
Aacute => "\x{00C1}", |
| 24 |
Acirc => "\x{00C2}", |
| 25 |
Agrave => "\x{00C0}", |
| 26 |
Alpha => "\x{0391}", |
| 27 |
Aring => "\x{00C5}", |
| 28 |
Atilde => "\x{00C3}", |
| 29 |
Auml => "\x{00C4}", |
| 30 |
Beta => "\x{0392}", |
| 31 |
Ccedil => "\x{00C7}", |
| 32 |
Chi => "\x{03A7}", |
| 33 |
Dagger => "\x{2021}", |
| 34 |
Delta => "\x{0394}", |
| 35 |
ETH => "\x{00D0}", |
| 36 |
Eacute => "\x{00C9}", |
| 37 |
Ecirc => "\x{00CA}", |
| 38 |
Egrave => "\x{00C8}", |
| 39 |
Epsilon => "\x{0395}", |
| 40 |
Eta => "\x{0397}", |
| 41 |
Euml => "\x{00CB}", |
| 42 |
Gamma => "\x{0393}", |
| 43 |
Iacute => "\x{00CD}", |
| 44 |
Icirc => "\x{00CE}", |
| 45 |
Igrave => "\x{00CC}", |
| 46 |
Iota => "\x{0399}", |
| 47 |
Iuml => "\x{00CF}", |
| 48 |
Kappa => "\x{039A}", |
| 49 |
Lambda => "\x{039B}", |
| 50 |
Mu => "\x{039C}", |
| 51 |
Ntilde => "\x{00D1}", |
| 52 |
Nu => "\x{039D}", |
| 53 |
OElig => "\x{0152}", |
| 54 |
Oacute => "\x{00D3}", |
| 55 |
Ocirc => "\x{00D4}", |
| 56 |
Ograve => "\x{00D2}", |
| 57 |
Omega => "\x{03A9}", |
| 58 |
Omicron => "\x{039F}", |
| 59 |
Oslash => "\x{00D8}", |
| 60 |
Otilde => "\x{00D5}", |
| 61 |
Ouml => "\x{00D6}", |
| 62 |
Phi => "\x{03A6}", |
| 63 |
Pi => "\x{03A0}", |
| 64 |
Prime => "\x{2033}", |
| 65 |
Psi => "\x{03A8}", |
| 66 |
Rho => "\x{03A1}", |
| 67 |
Scaron => "\x{0160}", |
| 68 |
Sigma => "\x{03A3}", |
| 69 |
THORN => "\x{00DE}", |
| 70 |
Tau => "\x{03A4}", |
| 71 |
Theta => "\x{0398}", |
| 72 |
Uacute => "\x{00DA}", |
| 73 |
Ucirc => "\x{00DB}", |
| 74 |
Ugrave => "\x{00D9}", |
| 75 |
Upsilon => "\x{03A5}", |
| 76 |
Uuml => "\x{00DC}", |
| 77 |
Xi => "\x{039E}", |
| 78 |
Yacute => "\x{00DD}", |
| 79 |
Yuml => "\x{0178}", |
| 80 |
Zeta => "\x{0396}", |
| 81 |
aacute => "\x{00E1}", |
| 82 |
acirc => "\x{00E2}", |
| 83 |
acute => "\x{00B4}", |
| 84 |
aelig => "\x{00E6}", |
| 85 |
agrave => "\x{00E0}", |
| 86 |
alefsym => "\x{2135}", |
| 87 |
alpha => "\x{03B1}", |
| 88 |
amp => "\x{0026}", |
| 89 |
AMP => "\x{0026}", |
| 90 |
and => "\x{2227}", |
| 91 |
ang => "\x{2220}", |
| 92 |
apos => "\x{0027}", |
| 93 |
aring => "\x{00E5}", |
| 94 |
asymp => "\x{2248}", |
| 95 |
atilde => "\x{00E3}", |
| 96 |
auml => "\x{00E4}", |
| 97 |
bdquo => "\x{201E}", |
| 98 |
beta => "\x{03B2}", |
| 99 |
brvbar => "\x{00A6}", |
| 100 |
bull => "\x{2022}", |
| 101 |
cap => "\x{2229}", |
| 102 |
ccedil => "\x{00E7}", |
| 103 |
cedil => "\x{00B8}", |
| 104 |
cent => "\x{00A2}", |
| 105 |
chi => "\x{03C7}", |
| 106 |
circ => "\x{02C6}", |
| 107 |
clubs => "\x{2663}", |
| 108 |
cong => "\x{2245}", |
| 109 |
copy => "\x{00A9}", |
| 110 |
COPY => "\x{00A9}", |
| 111 |
crarr => "\x{21B5}", |
| 112 |
cup => "\x{222A}", |
| 113 |
curren => "\x{00A4}", |
| 114 |
dArr => "\x{21D3}", |
| 115 |
dagger => "\x{2020}", |
| 116 |
darr => "\x{2193}", |
| 117 |
deg => "\x{00B0}", |
| 118 |
delta => "\x{03B4}", |
| 119 |
diams => "\x{2666}", |
| 120 |
divide => "\x{00F7}", |
| 121 |
eacute => "\x{00E9}", |
| 122 |
ecirc => "\x{00EA}", |
| 123 |
egrave => "\x{00E8}", |
| 124 |
empty => "\x{2205}", |
| 125 |
emsp => "\x{2003}", |
| 126 |
ensp => "\x{2002}", |
| 127 |
epsilon => "\x{03B5}", |
| 128 |
equiv => "\x{2261}", |
| 129 |
eta => "\x{03B7}", |
| 130 |
eth => "\x{00F0}", |
| 131 |
euml => "\x{00EB}", |
| 132 |
euro => "\x{20AC}", |
| 133 |
exist => "\x{2203}", |
| 134 |
fnof => "\x{0192}", |
| 135 |
forall => "\x{2200}", |
| 136 |
frac12 => "\x{00BD}", |
| 137 |
frac14 => "\x{00BC}", |
| 138 |
frac34 => "\x{00BE}", |
| 139 |
frasl => "\x{2044}", |
| 140 |
gamma => "\x{03B3}", |
| 141 |
ge => "\x{2265}", |
| 142 |
gt => "\x{003E}", |
| 143 |
GT => "\x{003E}", |
| 144 |
hArr => "\x{21D4}", |
| 145 |
harr => "\x{2194}", |
| 146 |
hearts => "\x{2665}", |
| 147 |
hellip => "\x{2026}", |
| 148 |
iacute => "\x{00ED}", |
| 149 |
icirc => "\x{00EE}", |
| 150 |
iexcl => "\x{00A1}", |
| 151 |
igrave => "\x{00EC}", |
| 152 |
image => "\x{2111}", |
| 153 |
infin => "\x{221E}", |
| 154 |
int => "\x{222B}", |
| 155 |
iota => "\x{03B9}", |
| 156 |
iquest => "\x{00BF}", |
| 157 |
isin => "\x{2208}", |
| 158 |
iuml => "\x{00EF}", |
| 159 |
kappa => "\x{03BA}", |
| 160 |
lArr => "\x{21D0}", |
| 161 |
lambda => "\x{03BB}", |
| 162 |
lang => "\x{2329}", |
| 163 |
laquo => "\x{00AB}", |
| 164 |
larr => "\x{2190}", |
| 165 |
lceil => "\x{2308}", |
| 166 |
ldquo => "\x{201C}", |
| 167 |
le => "\x{2264}", |
| 168 |
lfloor => "\x{230A}", |
| 169 |
lowast => "\x{2217}", |
| 170 |
loz => "\x{25CA}", |
| 171 |
lrm => "\x{200E}", |
| 172 |
lsaquo => "\x{2039}", |
| 173 |
lsquo => "\x{2018}", |
| 174 |
lt => "\x{003C}", |
| 175 |
LT => "\x{003C}", |
| 176 |
macr => "\x{00AF}", |
| 177 |
mdash => "\x{2014}", |
| 178 |
micro => "\x{00B5}", |
| 179 |
middot => "\x{00B7}", |
| 180 |
minus => "\x{2212}", |
| 181 |
mu => "\x{03BC}", |
| 182 |
nabla => "\x{2207}", |
| 183 |
nbsp => "\x{00A0}", |
| 184 |
ndash => "\x{2013}", |
| 185 |
ne => "\x{2260}", |
| 186 |
ni => "\x{220B}", |
| 187 |
not => "\x{00AC}", |
| 188 |
notin => "\x{2209}", |
| 189 |
nsub => "\x{2284}", |
| 190 |
ntilde => "\x{00F1}", |
| 191 |
nu => "\x{03BD}", |
| 192 |
oacute => "\x{00F3}", |
| 193 |
ocirc => "\x{00F4}", |
| 194 |
oelig => "\x{0153}", |
| 195 |
ograve => "\x{00F2}", |
| 196 |
oline => "\x{203E}", |
| 197 |
omega => "\x{03C9}", |
| 198 |
omicron => "\x{03BF}", |
| 199 |
oplus => "\x{2295}", |
| 200 |
or => "\x{2228}", |
| 201 |
ordf => "\x{00AA}", |
| 202 |
ordm => "\x{00BA}", |
| 203 |
oslash => "\x{00F8}", |
| 204 |
otilde => "\x{00F5}", |
| 205 |
otimes => "\x{2297}", |
| 206 |
ouml => "\x{00F6}", |
| 207 |
para => "\x{00B6}", |
| 208 |
part => "\x{2202}", |
| 209 |
permil => "\x{2030}", |
| 210 |
perp => "\x{22A5}", |
| 211 |
phi => "\x{03C6}", |
| 212 |
pi => "\x{03C0}", |
| 213 |
piv => "\x{03D6}", |
| 214 |
plusmn => "\x{00B1}", |
| 215 |
pound => "\x{00A3}", |
| 216 |
prime => "\x{2032}", |
| 217 |
prod => "\x{220F}", |
| 218 |
prop => "\x{221D}", |
| 219 |
psi => "\x{03C8}", |
| 220 |
quot => "\x{0022}", |
| 221 |
QUOT => "\x{0022}", |
| 222 |
rArr => "\x{21D2}", |
| 223 |
radic => "\x{221A}", |
| 224 |
rang => "\x{232A}", |
| 225 |
raquo => "\x{00BB}", |
| 226 |
rarr => "\x{2192}", |
| 227 |
rceil => "\x{2309}", |
| 228 |
rdquo => "\x{201D}", |
| 229 |
real => "\x{211C}", |
| 230 |
reg => "\x{00AE}", |
| 231 |
REG => "\x{00AE}", |
| 232 |
rfloor => "\x{230B}", |
| 233 |
rho => "\x{03C1}", |
| 234 |
rlm => "\x{200F}", |
| 235 |
rsaquo => "\x{203A}", |
| 236 |
rsquo => "\x{2019}", |
| 237 |
sbquo => "\x{201A}", |
| 238 |
scaron => "\x{0161}", |
| 239 |
sdot => "\x{22C5}", |
| 240 |
sect => "\x{00A7}", |
| 241 |
shy => "\x{00AD}", |
| 242 |
sigma => "\x{03C3}", |
| 243 |
sigmaf => "\x{03C2}", |
| 244 |
sim => "\x{223C}", |
| 245 |
spades => "\x{2660}", |
| 246 |
sub => "\x{2282}", |
| 247 |
sube => "\x{2286}", |
| 248 |
sum => "\x{2211}", |
| 249 |
sup => "\x{2283}", |
| 250 |
sup1 => "\x{00B9}", |
| 251 |
sup2 => "\x{00B2}", |
| 252 |
sup3 => "\x{00B3}", |
| 253 |
supe => "\x{2287}", |
| 254 |
szlig => "\x{00DF}", |
| 255 |
tau => "\x{03C4}", |
| 256 |
there4 => "\x{2234}", |
| 257 |
theta => "\x{03B8}", |
| 258 |
thetasym => "\x{03D1}", |
| 259 |
thinsp => "\x{2009}", |
| 260 |
thorn => "\x{00FE}", |
| 261 |
tilde => "\x{02DC}", |
| 262 |
times => "\x{00D7}", |
| 263 |
trade => "\x{2122}", |
| 264 |
uArr => "\x{21D1}", |
| 265 |
uacute => "\x{00FA}", |
| 266 |
uarr => "\x{2191}", |
| 267 |
ucirc => "\x{00FB}", |
| 268 |
ugrave => "\x{00F9}", |
| 269 |
uml => "\x{00A8}", |
| 270 |
upsih => "\x{03D2}", |
| 271 |
upsilon => "\x{03C5}", |
| 272 |
uuml => "\x{00FC}", |
| 273 |
weierp => "\x{2118}", |
| 274 |
xi => "\x{03BE}", |
| 275 |
yacute => "\x{00FD}", |
| 276 |
yen => "\x{00A5}", |
| 277 |
yuml => "\x{00FF}", |
| 278 |
zeta => "\x{03B6}", |
| 279 |
zwj => "\x{200D}", |
| 280 |
zwnj => "\x{200C}", |
| 281 |
}; # $entity_char |
| 282 |
|
| 283 |
my $c1_entity_char = { |
| 284 |
0x80 => 0x20AC, |
| 285 |
0x81 => 0xFFFD, |
| 286 |
0x82 => 0x201A, |
| 287 |
0x83 => 0x0192, |
| 288 |
0x84 => 0x201E, |
| 289 |
0x85 => 0x2026, |
| 290 |
0x86 => 0x2020, |
| 291 |
0x87 => 0x2021, |
| 292 |
0x88 => 0x02C6, |
| 293 |
0x89 => 0x2030, |
| 294 |
0x8A => 0x0160, |
| 295 |
0x8B => 0x2039, |
| 296 |
0x8C => 0x0152, |
| 297 |
0x8D => 0xFFFD, |
| 298 |
0x8E => 0x017D, |
| 299 |
0x8F => 0xFFFD, |
| 300 |
0x90 => 0xFFFD, |
| 301 |
0x91 => 0x2018, |
| 302 |
0x92 => 0x2019, |
| 303 |
0x93 => 0x201C, |
| 304 |
0x94 => 0x201D, |
| 305 |
0x95 => 0x2022, |
| 306 |
0x96 => 0x2013, |
| 307 |
0x97 => 0x2014, |
| 308 |
0x98 => 0x02DC, |
| 309 |
0x99 => 0x2122, |
| 310 |
0x9A => 0x0161, |
| 311 |
0x9B => 0x203A, |
| 312 |
0x9C => 0x0153, |
| 313 |
0x9D => 0xFFFD, |
| 314 |
0x9E => 0x017E, |
| 315 |
0x9F => 0x0178, |
| 316 |
}; # $c1_entity_char |
| 317 |
|
| 318 |
my $special_category = { |
| 319 |
address => 1, area => 1, base => 1, basefont => 1, bgsound => 1, |
| 320 |
blockquote => 1, body => 1, br => 1, center => 1, col => 1, colgroup => 1, |
| 321 |
dd => 1, dir => 1, div => 1, dl => 1, dt => 1, embed => 1, fieldset => 1, |
| 322 |
form => 1, frame => 1, frameset => 1, h1 => 1, h2 => 1, h3 => 1, |
| 323 |
h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, iframe => 1, image => 1, |
| 324 |
img => 1, input => 1, isindex => 1, li => 1, link => 1, listing => 1, |
| 325 |
menu => 1, meta => 1, noembed => 1, noframes => 1, noscript => 1, |
| 326 |
ol => 1, optgroup => 1, option => 1, p => 1, param => 1, plaintext => 1, |
| 327 |
pre => 1, script => 1, select => 1, spacer => 1, style => 1, tbody => 1, |
| 328 |
textarea => 1, tfoot => 1, thead => 1, title => 1, tr => 1, ul => 1, wbr => 1, |
| 329 |
}; |
| 330 |
my $scoping_category = { |
| 331 |
button => 1, caption => 1, html => 1, marquee => 1, object => 1, |
| 332 |
table => 1, td => 1, th => 1, |
| 333 |
}; |
| 334 |
my $formatting_category = { |
| 335 |
a => 1, b => 1, big => 1, em => 1, font => 1, i => 1, nobr => 1, |
| 336 |
s => 1, small => 1, strile => 1, strong => 1, tt => 1, u => 1, |
| 337 |
}; |
| 338 |
# $phrasing_category: all other elements |
| 339 |
|
| 340 |
sub parse_string ($$$;$) { |
| 341 |
my $self = shift->new; |
| 342 |
my $s = \$_[0]; |
| 343 |
$self->{document} = $_[1]; |
| 344 |
|
| 345 |
## NOTE: |set_inner_html| copies most of this method's code |
| 346 |
|
| 347 |
my $i = 0; |
| 348 |
my $line = 1; |
| 349 |
my $column = 0; |
| 350 |
$self->{set_next_input_character} = sub { |
| 351 |
my $self = shift; |
| 352 |
$self->{next_input_character} = -1 and return if $i >= length $$s; |
| 353 |
$self->{next_input_character} = ord substr $$s, $i++, 1; |
| 354 |
$column++; |
| 355 |
|
| 356 |
if ($self->{next_input_character} == 0x000A) { # LF |
| 357 |
$line++; |
| 358 |
$column = 0; |
| 359 |
} elsif ($self->{next_input_character} == 0x000D) { # CR |
| 360 |
if ($i >= length $$s) { |
| 361 |
# |
| 362 |
} else { |
| 363 |
my $next_char = ord substr $$s, $i++, 1; |
| 364 |
if ($next_char == 0x000A) { # LF |
| 365 |
# |
| 366 |
} else { |
| 367 |
push @{$self->{char}}, $next_char; |
| 368 |
} |
| 369 |
} |
| 370 |
$self->{next_input_character} = 0x000A; # LF # MUST |
| 371 |
$line++; |
| 372 |
$column = 0; |
| 373 |
} elsif ($self->{next_input_character} > 0x10FFFF) { |
| 374 |
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 375 |
} elsif ($self->{next_input_character} == 0x0000) { # NULL |
| 376 |
!!!parse-error (type => 'NULL'); |
| 377 |
## TODO: test |
| 378 |
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 379 |
} |
| 380 |
}; |
| 381 |
|
| 382 |
my $onerror = $_[2] || sub { |
| 383 |
my (%opt) = @_; |
| 384 |
warn "Parse error ($opt{type}) at line $opt{line} column $opt{column}\n"; |
| 385 |
}; |
| 386 |
$self->{parse_error} = sub { |
| 387 |
$onerror->(@_, line => $line, column => $column); |
| 388 |
}; |
| 389 |
|
| 390 |
$self->_initialize_tokenizer; |
| 391 |
$self->_initialize_tree_constructor; |
| 392 |
$self->_construct_tree; |
| 393 |
$self->_terminate_tree_constructor; |
| 394 |
|
| 395 |
return $self->{document}; |
| 396 |
} # parse_string |
| 397 |
|
| 398 |
sub new ($) { |
| 399 |
my $class = shift; |
| 400 |
my $self = bless {}, $class; |
| 401 |
$self->{set_next_input_character} = sub { |
| 402 |
$self->{next_input_character} = -1; |
| 403 |
}; |
| 404 |
$self->{parse_error} = sub { |
| 405 |
# |
| 406 |
}; |
| 407 |
return $self; |
| 408 |
} # new |
| 409 |
|
| 410 |
## Implementations MUST act as if state machine in the spec |
| 411 |
|
| 412 |
sub _initialize_tokenizer ($) { |
| 413 |
my $self = shift; |
| 414 |
$self->{state} = 'data'; # MUST |
| 415 |
$self->{content_model_flag} = 'PCDATA'; # be |
| 416 |
undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE |
| 417 |
undef $self->{current_attribute}; |
| 418 |
undef $self->{last_emitted_start_tag_name}; |
| 419 |
undef $self->{last_attribute_value_state}; |
| 420 |
$self->{char} = []; |
| 421 |
# $self->{next_input_character} |
| 422 |
!!!next-input-character; |
| 423 |
$self->{token} = []; |
| 424 |
} # _initialize_tokenizer |
| 425 |
|
| 426 |
## A token has: |
| 427 |
## ->{type} eq 'DOCTYPE', 'start tag', 'end tag', 'comment', |
| 428 |
## 'character', or 'end-of-file' |
| 429 |
## ->{name} (DOCTYPE, start tag (tagname), end tag (tagname)) |
| 430 |
## ISSUE: the spec need s/tagname/tag name/ |
| 431 |
## ->{error} == 1 or 0 (DOCTYPE) |
| 432 |
## ->{attributes} isa HASH (start tag, end tag) |
| 433 |
## ->{data} (comment, character) |
| 434 |
|
| 435 |
## Macros |
| 436 |
## Macros MUST be preceded by three EXCLAMATION MARKs. |
| 437 |
## emit ($token) |
| 438 |
## Emits the specified token. |
| 439 |
|
| 440 |
## Emitted token MUST immediately be handled by the tree construction state. |
| 441 |
|
| 442 |
## Before each step, UA MAY check to see if either one of the scripts in |
| 443 |
## "list of scripts that will execute as soon as possible" or the first |
| 444 |
## script in the "list of scripts that will execute asynchronously", |
| 445 |
## has completed loading. If one has, then it MUST be executed |
| 446 |
## and removed from the list. |
| 447 |
|
| 448 |
## ISSUE: <http://html5.org/tools/web-apps-tracker?from=874&to=876> |
| 449 |
|
| 450 |
sub _get_next_token ($) { |
| 451 |
my $self = shift; |
| 452 |
if (@{$self->{token}}) { |
| 453 |
return shift @{$self->{token}}; |
| 454 |
} |
| 455 |
|
| 456 |
A: { |
| 457 |
if ($self->{state} eq 'data') { |
| 458 |
if ($self->{next_input_character} == 0x0026) { # & |
| 459 |
if ($self->{content_model_flag} eq 'PCDATA' or |
| 460 |
$self->{content_model_flag} eq 'RCDATA') { |
| 461 |
$self->{state} = 'entity data'; |
| 462 |
!!!next-input-character; |
| 463 |
redo A; |
| 464 |
} else { |
| 465 |
# |
| 466 |
} |
| 467 |
} elsif ($self->{next_input_character} == 0x003C) { # < |
| 468 |
if ($self->{content_model_flag} ne 'PLAINTEXT') { |
| 469 |
$self->{state} = 'tag open'; |
| 470 |
!!!next-input-character; |
| 471 |
redo A; |
| 472 |
} else { |
| 473 |
# |
| 474 |
} |
| 475 |
} elsif ($self->{next_input_character} == -1) { |
| 476 |
!!!emit ({type => 'end-of-file'}); |
| 477 |
last A; ## TODO: ok? |
| 478 |
} |
| 479 |
# Anything else |
| 480 |
my $token = {type => 'character', |
| 481 |
data => chr $self->{next_input_character}}; |
| 482 |
## Stay in the data state |
| 483 |
!!!next-input-character; |
| 484 |
|
| 485 |
!!!emit ($token); |
| 486 |
|
| 487 |
redo A; |
| 488 |
} elsif ($self->{state} eq 'entity data') { |
| 489 |
## (cannot happen in CDATA state) |
| 490 |
|
| 491 |
my $token = $self->_tokenize_attempt_to_consume_an_entity; |
| 492 |
|
| 493 |
$self->{state} = 'data'; |
| 494 |
# next-input-character is already done |
| 495 |
|
| 496 |
unless (defined $token) { |
| 497 |
!!!emit ({type => 'character', data => '&'}); |
| 498 |
} else { |
| 499 |
!!!emit ($token); |
| 500 |
} |
| 501 |
|
| 502 |
redo A; |
| 503 |
} elsif ($self->{state} eq 'tag open') { |
| 504 |
if ($self->{content_model_flag} eq 'RCDATA' or |
| 505 |
$self->{content_model_flag} eq 'CDATA') { |
| 506 |
if ($self->{next_input_character} == 0x002F) { # / |
| 507 |
!!!next-input-character; |
| 508 |
$self->{state} = 'close tag open'; |
| 509 |
redo A; |
| 510 |
} else { |
| 511 |
## reconsume |
| 512 |
$self->{state} = 'data'; |
| 513 |
|
| 514 |
!!!emit ({type => 'character', data => '<'}); |
| 515 |
|
| 516 |
redo A; |
| 517 |
} |
| 518 |
} elsif ($self->{content_model_flag} eq 'PCDATA') { |
| 519 |
if ($self->{next_input_character} == 0x0021) { # ! |
| 520 |
$self->{state} = 'markup declaration open'; |
| 521 |
!!!next-input-character; |
| 522 |
redo A; |
| 523 |
} elsif ($self->{next_input_character} == 0x002F) { # / |
| 524 |
$self->{state} = 'close tag open'; |
| 525 |
!!!next-input-character; |
| 526 |
redo A; |
| 527 |
} elsif (0x0041 <= $self->{next_input_character} and |
| 528 |
$self->{next_input_character} <= 0x005A) { # A..Z |
| 529 |
$self->{current_token} |
| 530 |
= {type => 'start tag', |
| 531 |
tag_name => chr ($self->{next_input_character} + 0x0020)}; |
| 532 |
$self->{state} = 'tag name'; |
| 533 |
!!!next-input-character; |
| 534 |
redo A; |
| 535 |
} elsif (0x0061 <= $self->{next_input_character} and |
| 536 |
$self->{next_input_character} <= 0x007A) { # a..z |
| 537 |
$self->{current_token} = {type => 'start tag', |
| 538 |
tag_name => chr ($self->{next_input_character})}; |
| 539 |
$self->{state} = 'tag name'; |
| 540 |
!!!next-input-character; |
| 541 |
redo A; |
| 542 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 543 |
!!!parse-error (type => 'empty start tag'); |
| 544 |
$self->{state} = 'data'; |
| 545 |
!!!next-input-character; |
| 546 |
|
| 547 |
!!!emit ({type => 'character', data => '<>'}); |
| 548 |
|
| 549 |
redo A; |
| 550 |
} elsif ($self->{next_input_character} == 0x003F) { # ? |
| 551 |
!!!parse-error (type => 'pio'); |
| 552 |
$self->{state} = 'bogus comment'; |
| 553 |
## $self->{next_input_character} is intentionally left as is |
| 554 |
redo A; |
| 555 |
} else { |
| 556 |
!!!parse-error (type => 'bare stago'); |
| 557 |
$self->{state} = 'data'; |
| 558 |
## reconsume |
| 559 |
|
| 560 |
!!!emit ({type => 'character', data => '<'}); |
| 561 |
|
| 562 |
redo A; |
| 563 |
} |
| 564 |
} else { |
| 565 |
die "$0: $self->{content_model_flag}: Unknown content model flag"; |
| 566 |
} |
| 567 |
} elsif ($self->{state} eq 'close tag open') { |
| 568 |
if ($self->{content_model_flag} eq 'RCDATA' or |
| 569 |
$self->{content_model_flag} eq 'CDATA') { |
| 570 |
my @next_char; |
| 571 |
TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) { |
| 572 |
push @next_char, $self->{next_input_character}; |
| 573 |
my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1); |
| 574 |
my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c; |
| 575 |
if ($self->{next_input_character} == $c or $self->{next_input_character} == $C) { |
| 576 |
!!!next-input-character; |
| 577 |
next TAGNAME; |
| 578 |
} else { |
| 579 |
!!!parse-error (type => 'unmatched end tag'); |
| 580 |
$self->{next_input_character} = shift @next_char; # reconsume |
| 581 |
!!!back-next-input-character (@next_char); |
| 582 |
$self->{state} = 'data'; |
| 583 |
|
| 584 |
!!!emit ({type => 'character', data => '</'}); |
| 585 |
|
| 586 |
redo A; |
| 587 |
} |
| 588 |
} |
| 589 |
push @next_char, $self->{next_input_character}; |
| 590 |
|
| 591 |
unless ($self->{next_input_character} == 0x0009 or # HT |
| 592 |
$self->{next_input_character} == 0x000A or # LF |
| 593 |
$self->{next_input_character} == 0x000B or # VT |
| 594 |
$self->{next_input_character} == 0x000C or # FF |
| 595 |
$self->{next_input_character} == 0x0020 or # SP |
| 596 |
$self->{next_input_character} == 0x003E or # > |
| 597 |
$self->{next_input_character} == 0x002F or # / |
| 598 |
$self->{next_input_character} == 0x003C or # < |
| 599 |
$self->{next_input_character} == -1) { |
| 600 |
!!!parse-error (type => 'unmatched end tag'); |
| 601 |
$self->{next_input_character} = shift @next_char; # reconsume |
| 602 |
!!!back-next-input-character (@next_char); |
| 603 |
$self->{state} = 'data'; |
| 604 |
|
| 605 |
!!!emit ({type => 'character', data => '</'}); |
| 606 |
|
| 607 |
redo A; |
| 608 |
} else { |
| 609 |
$self->{next_input_character} = shift @next_char; |
| 610 |
!!!back-next-input-character (@next_char); |
| 611 |
# and consume... |
| 612 |
} |
| 613 |
} |
| 614 |
|
| 615 |
if (0x0041 <= $self->{next_input_character} and |
| 616 |
$self->{next_input_character} <= 0x005A) { # A..Z |
| 617 |
$self->{current_token} = {type => 'end tag', |
| 618 |
tag_name => chr ($self->{next_input_character} + 0x0020)}; |
| 619 |
$self->{state} = 'tag name'; |
| 620 |
!!!next-input-character; |
| 621 |
redo A; |
| 622 |
} elsif (0x0061 <= $self->{next_input_character} and |
| 623 |
$self->{next_input_character} <= 0x007A) { # a..z |
| 624 |
$self->{current_token} = {type => 'end tag', |
| 625 |
tag_name => chr ($self->{next_input_character})}; |
| 626 |
$self->{state} = 'tag name'; |
| 627 |
!!!next-input-character; |
| 628 |
redo A; |
| 629 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 630 |
!!!parse-error (type => 'empty end tag'); |
| 631 |
$self->{state} = 'data'; |
| 632 |
!!!next-input-character; |
| 633 |
redo A; |
| 634 |
} elsif ($self->{next_input_character} == -1) { |
| 635 |
!!!parse-error (type => 'bare etago'); |
| 636 |
$self->{state} = 'data'; |
| 637 |
# reconsume |
| 638 |
|
| 639 |
!!!emit ({type => 'character', data => '</'}); |
| 640 |
|
| 641 |
redo A; |
| 642 |
} else { |
| 643 |
!!!parse-error (type => 'bogus end tag'); |
| 644 |
$self->{state} = 'bogus comment'; |
| 645 |
## $self->{next_input_character} is intentionally left as is |
| 646 |
redo A; |
| 647 |
} |
| 648 |
} elsif ($self->{state} eq 'tag name') { |
| 649 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 650 |
$self->{next_input_character} == 0x000A or # LF |
| 651 |
$self->{next_input_character} == 0x000B or # VT |
| 652 |
$self->{next_input_character} == 0x000C or # FF |
| 653 |
$self->{next_input_character} == 0x0020) { # SP |
| 654 |
$self->{state} = 'before attribute name'; |
| 655 |
!!!next-input-character; |
| 656 |
redo A; |
| 657 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 658 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 659 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 660 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 661 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 662 |
if ($self->{current_token}->{attributes}) { |
| 663 |
!!!parse-error (type => 'end tag attribute'); |
| 664 |
} |
| 665 |
} else { |
| 666 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 667 |
} |
| 668 |
$self->{state} = 'data'; |
| 669 |
!!!next-input-character; |
| 670 |
|
| 671 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 672 |
undef $self->{current_token}; |
| 673 |
|
| 674 |
redo A; |
| 675 |
} elsif (0x0041 <= $self->{next_input_character} and |
| 676 |
$self->{next_input_character} <= 0x005A) { # A..Z |
| 677 |
$self->{current_token}->{tag_name} .= chr ($self->{next_input_character} + 0x0020); |
| 678 |
# start tag or end tag |
| 679 |
## Stay in this state |
| 680 |
!!!next-input-character; |
| 681 |
redo A; |
| 682 |
} elsif ($self->{next_input_character} == 0x003C or # < |
| 683 |
$self->{next_input_character} == -1) { |
| 684 |
!!!parse-error (type => 'unclosed tag'); |
| 685 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 686 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 687 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 688 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 689 |
if ($self->{current_token}->{attributes}) { |
| 690 |
!!!parse-error (type => 'end tag attribute'); |
| 691 |
} |
| 692 |
} else { |
| 693 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 694 |
} |
| 695 |
$self->{state} = 'data'; |
| 696 |
# reconsume |
| 697 |
|
| 698 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 699 |
undef $self->{current_token}; |
| 700 |
|
| 701 |
redo A; |
| 702 |
} elsif ($self->{next_input_character} == 0x002F) { # / |
| 703 |
!!!next-input-character; |
| 704 |
if ($self->{next_input_character} == 0x003E and # > |
| 705 |
$self->{current_token}->{type} eq 'start tag' and |
| 706 |
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
| 707 |
# permitted slash |
| 708 |
# |
| 709 |
} else { |
| 710 |
!!!parse-error (type => 'nestc'); |
| 711 |
} |
| 712 |
$self->{state} = 'before attribute name'; |
| 713 |
# next-input-character is already done |
| 714 |
redo A; |
| 715 |
} else { |
| 716 |
$self->{current_token}->{tag_name} .= chr $self->{next_input_character}; |
| 717 |
# start tag or end tag |
| 718 |
## Stay in the state |
| 719 |
!!!next-input-character; |
| 720 |
redo A; |
| 721 |
} |
| 722 |
} elsif ($self->{state} eq 'before attribute name') { |
| 723 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 724 |
$self->{next_input_character} == 0x000A or # LF |
| 725 |
$self->{next_input_character} == 0x000B or # VT |
| 726 |
$self->{next_input_character} == 0x000C or # FF |
| 727 |
$self->{next_input_character} == 0x0020) { # SP |
| 728 |
## Stay in the state |
| 729 |
!!!next-input-character; |
| 730 |
redo A; |
| 731 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 732 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 733 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 734 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 735 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 736 |
if ($self->{current_token}->{attributes}) { |
| 737 |
!!!parse-error (type => 'end tag attribute'); |
| 738 |
} |
| 739 |
} else { |
| 740 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 741 |
} |
| 742 |
$self->{state} = 'data'; |
| 743 |
!!!next-input-character; |
| 744 |
|
| 745 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 746 |
undef $self->{current_token}; |
| 747 |
|
| 748 |
redo A; |
| 749 |
} elsif (0x0041 <= $self->{next_input_character} and |
| 750 |
$self->{next_input_character} <= 0x005A) { # A..Z |
| 751 |
$self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), |
| 752 |
value => ''}; |
| 753 |
$self->{state} = 'attribute name'; |
| 754 |
!!!next-input-character; |
| 755 |
redo A; |
| 756 |
} elsif ($self->{next_input_character} == 0x002F) { # / |
| 757 |
!!!next-input-character; |
| 758 |
if ($self->{next_input_character} == 0x003E and # > |
| 759 |
$self->{current_token}->{type} eq 'start tag' and |
| 760 |
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
| 761 |
# permitted slash |
| 762 |
# |
| 763 |
} else { |
| 764 |
!!!parse-error (type => 'nestc'); |
| 765 |
} |
| 766 |
## Stay in the state |
| 767 |
# next-input-character is already done |
| 768 |
redo A; |
| 769 |
} elsif ($self->{next_input_character} == 0x003C or # < |
| 770 |
$self->{next_input_character} == -1) { |
| 771 |
!!!parse-error (type => 'unclosed tag'); |
| 772 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 773 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 774 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 775 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 776 |
if ($self->{current_token}->{attributes}) { |
| 777 |
!!!parse-error (type => 'end tag attribute'); |
| 778 |
} |
| 779 |
} else { |
| 780 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 781 |
} |
| 782 |
$self->{state} = 'data'; |
| 783 |
# reconsume |
| 784 |
|
| 785 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 786 |
undef $self->{current_token}; |
| 787 |
|
| 788 |
redo A; |
| 789 |
} else { |
| 790 |
$self->{current_attribute} = {name => chr ($self->{next_input_character}), |
| 791 |
value => ''}; |
| 792 |
$self->{state} = 'attribute name'; |
| 793 |
!!!next-input-character; |
| 794 |
redo A; |
| 795 |
} |
| 796 |
} elsif ($self->{state} eq 'attribute name') { |
| 797 |
my $before_leave = sub { |
| 798 |
if (exists $self->{current_token}->{attributes} # start tag or end tag |
| 799 |
->{$self->{current_attribute}->{name}}) { # MUST |
| 800 |
!!!parse-error (type => 'dupulicate attribute'); |
| 801 |
## Discard $self->{current_attribute} # MUST |
| 802 |
} else { |
| 803 |
$self->{current_token}->{attributes}->{$self->{current_attribute}->{name}} |
| 804 |
= $self->{current_attribute}; |
| 805 |
} |
| 806 |
}; # $before_leave |
| 807 |
|
| 808 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 809 |
$self->{next_input_character} == 0x000A or # LF |
| 810 |
$self->{next_input_character} == 0x000B or # VT |
| 811 |
$self->{next_input_character} == 0x000C or # FF |
| 812 |
$self->{next_input_character} == 0x0020) { # SP |
| 813 |
$before_leave->(); |
| 814 |
$self->{state} = 'after attribute name'; |
| 815 |
!!!next-input-character; |
| 816 |
redo A; |
| 817 |
} elsif ($self->{next_input_character} == 0x003D) { # = |
| 818 |
$before_leave->(); |
| 819 |
$self->{state} = 'before attribute value'; |
| 820 |
!!!next-input-character; |
| 821 |
redo A; |
| 822 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 823 |
$before_leave->(); |
| 824 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 825 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 826 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 827 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 828 |
if ($self->{current_token}->{attributes}) { |
| 829 |
!!!parse-error (type => 'end tag attribute'); |
| 830 |
} |
| 831 |
} else { |
| 832 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 833 |
} |
| 834 |
$self->{state} = 'data'; |
| 835 |
!!!next-input-character; |
| 836 |
|
| 837 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 838 |
undef $self->{current_token}; |
| 839 |
|
| 840 |
redo A; |
| 841 |
} elsif (0x0041 <= $self->{next_input_character} and |
| 842 |
$self->{next_input_character} <= 0x005A) { # A..Z |
| 843 |
$self->{current_attribute}->{name} .= chr ($self->{next_input_character} + 0x0020); |
| 844 |
## Stay in the state |
| 845 |
!!!next-input-character; |
| 846 |
redo A; |
| 847 |
} elsif ($self->{next_input_character} == 0x002F) { # / |
| 848 |
$before_leave->(); |
| 849 |
!!!next-input-character; |
| 850 |
if ($self->{next_input_character} == 0x003E and # > |
| 851 |
$self->{current_token}->{type} eq 'start tag' and |
| 852 |
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
| 853 |
# permitted slash |
| 854 |
# |
| 855 |
} else { |
| 856 |
!!!parse-error (type => 'nestc'); |
| 857 |
} |
| 858 |
$self->{state} = 'before attribute name'; |
| 859 |
# next-input-character is already done |
| 860 |
redo A; |
| 861 |
} elsif ($self->{next_input_character} == 0x003C or # < |
| 862 |
$self->{next_input_character} == -1) { |
| 863 |
!!!parse-error (type => 'unclosed tag'); |
| 864 |
$before_leave->(); |
| 865 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 866 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 867 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 868 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 869 |
if ($self->{current_token}->{attributes}) { |
| 870 |
!!!parse-error (type => 'end tag attribute'); |
| 871 |
} |
| 872 |
} else { |
| 873 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 874 |
} |
| 875 |
$self->{state} = 'data'; |
| 876 |
# reconsume |
| 877 |
|
| 878 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 879 |
undef $self->{current_token}; |
| 880 |
|
| 881 |
redo A; |
| 882 |
} else { |
| 883 |
$self->{current_attribute}->{name} .= chr ($self->{next_input_character}); |
| 884 |
## Stay in the state |
| 885 |
!!!next-input-character; |
| 886 |
redo A; |
| 887 |
} |
| 888 |
} elsif ($self->{state} eq 'after attribute name') { |
| 889 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 890 |
$self->{next_input_character} == 0x000A or # LF |
| 891 |
$self->{next_input_character} == 0x000B or # VT |
| 892 |
$self->{next_input_character} == 0x000C or # FF |
| 893 |
$self->{next_input_character} == 0x0020) { # SP |
| 894 |
## Stay in the state |
| 895 |
!!!next-input-character; |
| 896 |
redo A; |
| 897 |
} elsif ($self->{next_input_character} == 0x003D) { # = |
| 898 |
$self->{state} = 'before attribute value'; |
| 899 |
!!!next-input-character; |
| 900 |
redo A; |
| 901 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 902 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 903 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 904 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 905 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 906 |
if ($self->{current_token}->{attributes}) { |
| 907 |
!!!parse-error (type => 'end tag attribute'); |
| 908 |
} |
| 909 |
} else { |
| 910 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 911 |
} |
| 912 |
$self->{state} = 'data'; |
| 913 |
!!!next-input-character; |
| 914 |
|
| 915 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 916 |
undef $self->{current_token}; |
| 917 |
|
| 918 |
redo A; |
| 919 |
} elsif (0x0041 <= $self->{next_input_character} and |
| 920 |
$self->{next_input_character} <= 0x005A) { # A..Z |
| 921 |
$self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), |
| 922 |
value => ''}; |
| 923 |
$self->{state} = 'attribute name'; |
| 924 |
!!!next-input-character; |
| 925 |
redo A; |
| 926 |
} elsif ($self->{next_input_character} == 0x002F) { # / |
| 927 |
!!!next-input-character; |
| 928 |
if ($self->{next_input_character} == 0x003E and # > |
| 929 |
$self->{current_token}->{type} eq 'start tag' and |
| 930 |
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
| 931 |
# permitted slash |
| 932 |
# |
| 933 |
} else { |
| 934 |
!!!parse-error (type => 'nestc'); |
| 935 |
} |
| 936 |
$self->{state} = 'before attribute name'; |
| 937 |
# next-input-character is already done |
| 938 |
redo A; |
| 939 |
} elsif ($self->{next_input_character} == 0x003C or # < |
| 940 |
$self->{next_input_character} == -1) { |
| 941 |
!!!parse-error (type => 'unclosed tag'); |
| 942 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 943 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 944 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 945 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 946 |
if ($self->{current_token}->{attributes}) { |
| 947 |
!!!parse-error (type => 'end tag attribute'); |
| 948 |
} |
| 949 |
} else { |
| 950 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 951 |
} |
| 952 |
$self->{state} = 'data'; |
| 953 |
# reconsume |
| 954 |
|
| 955 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 956 |
undef $self->{current_token}; |
| 957 |
|
| 958 |
redo A; |
| 959 |
} else { |
| 960 |
$self->{current_attribute} = {name => chr ($self->{next_input_character}), |
| 961 |
value => ''}; |
| 962 |
$self->{state} = 'attribute name'; |
| 963 |
!!!next-input-character; |
| 964 |
redo A; |
| 965 |
} |
| 966 |
} elsif ($self->{state} eq 'before attribute value') { |
| 967 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 968 |
$self->{next_input_character} == 0x000A or # LF |
| 969 |
$self->{next_input_character} == 0x000B or # VT |
| 970 |
$self->{next_input_character} == 0x000C or # FF |
| 971 |
$self->{next_input_character} == 0x0020) { # SP |
| 972 |
## Stay in the state |
| 973 |
!!!next-input-character; |
| 974 |
redo A; |
| 975 |
} elsif ($self->{next_input_character} == 0x0022) { # " |
| 976 |
$self->{state} = 'attribute value (double-quoted)'; |
| 977 |
!!!next-input-character; |
| 978 |
redo A; |
| 979 |
} elsif ($self->{next_input_character} == 0x0026) { # & |
| 980 |
$self->{state} = 'attribute value (unquoted)'; |
| 981 |
## reconsume |
| 982 |
redo A; |
| 983 |
} elsif ($self->{next_input_character} == 0x0027) { # ' |
| 984 |
$self->{state} = 'attribute value (single-quoted)'; |
| 985 |
!!!next-input-character; |
| 986 |
redo A; |
| 987 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 988 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 989 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 990 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 991 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 992 |
if ($self->{current_token}->{attributes}) { |
| 993 |
!!!parse-error (type => 'end tag attribute'); |
| 994 |
} |
| 995 |
} else { |
| 996 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 997 |
} |
| 998 |
$self->{state} = 'data'; |
| 999 |
!!!next-input-character; |
| 1000 |
|
| 1001 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 1002 |
undef $self->{current_token}; |
| 1003 |
|
| 1004 |
redo A; |
| 1005 |
} elsif ($self->{next_input_character} == 0x003C or # < |
| 1006 |
$self->{next_input_character} == -1) { |
| 1007 |
!!!parse-error (type => 'unclosed tag'); |
| 1008 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 1009 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 1010 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 1011 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 1012 |
if ($self->{current_token}->{attributes}) { |
| 1013 |
!!!parse-error (type => 'end tag attribute'); |
| 1014 |
} |
| 1015 |
} else { |
| 1016 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 1017 |
} |
| 1018 |
$self->{state} = 'data'; |
| 1019 |
## reconsume |
| 1020 |
|
| 1021 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 1022 |
undef $self->{current_token}; |
| 1023 |
|
| 1024 |
redo A; |
| 1025 |
} else { |
| 1026 |
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
| 1027 |
$self->{state} = 'attribute value (unquoted)'; |
| 1028 |
!!!next-input-character; |
| 1029 |
redo A; |
| 1030 |
} |
| 1031 |
} elsif ($self->{state} eq 'attribute value (double-quoted)') { |
| 1032 |
if ($self->{next_input_character} == 0x0022) { # " |
| 1033 |
$self->{state} = 'before attribute name'; |
| 1034 |
!!!next-input-character; |
| 1035 |
redo A; |
| 1036 |
} elsif ($self->{next_input_character} == 0x0026) { # & |
| 1037 |
$self->{last_attribute_value_state} = 'attribute value (double-quoted)'; |
| 1038 |
$self->{state} = 'entity in attribute value'; |
| 1039 |
!!!next-input-character; |
| 1040 |
redo A; |
| 1041 |
} elsif ($self->{next_input_character} == -1) { |
| 1042 |
!!!parse-error (type => 'unclosed attribute value'); |
| 1043 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 1044 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 1045 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 1046 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 1047 |
if ($self->{current_token}->{attributes}) { |
| 1048 |
!!!parse-error (type => 'end tag attribute'); |
| 1049 |
} |
| 1050 |
} else { |
| 1051 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 1052 |
} |
| 1053 |
$self->{state} = 'data'; |
| 1054 |
## reconsume |
| 1055 |
|
| 1056 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 1057 |
undef $self->{current_token}; |
| 1058 |
|
| 1059 |
redo A; |
| 1060 |
} else { |
| 1061 |
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
| 1062 |
## Stay in the state |
| 1063 |
!!!next-input-character; |
| 1064 |
redo A; |
| 1065 |
} |
| 1066 |
} elsif ($self->{state} eq 'attribute value (single-quoted)') { |
| 1067 |
if ($self->{next_input_character} == 0x0027) { # ' |
| 1068 |
$self->{state} = 'before attribute name'; |
| 1069 |
!!!next-input-character; |
| 1070 |
redo A; |
| 1071 |
} elsif ($self->{next_input_character} == 0x0026) { # & |
| 1072 |
$self->{last_attribute_value_state} = 'attribute value (single-quoted)'; |
| 1073 |
$self->{state} = 'entity in attribute value'; |
| 1074 |
!!!next-input-character; |
| 1075 |
redo A; |
| 1076 |
} elsif ($self->{next_input_character} == -1) { |
| 1077 |
!!!parse-error (type => 'unclosed attribute value'); |
| 1078 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 1079 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 1080 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 1081 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 1082 |
if ($self->{current_token}->{attributes}) { |
| 1083 |
!!!parse-error (type => 'end tag attribute'); |
| 1084 |
} |
| 1085 |
} else { |
| 1086 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 1087 |
} |
| 1088 |
$self->{state} = 'data'; |
| 1089 |
## reconsume |
| 1090 |
|
| 1091 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 1092 |
undef $self->{current_token}; |
| 1093 |
|
| 1094 |
redo A; |
| 1095 |
} else { |
| 1096 |
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
| 1097 |
## Stay in the state |
| 1098 |
!!!next-input-character; |
| 1099 |
redo A; |
| 1100 |
} |
| 1101 |
} elsif ($self->{state} eq 'attribute value (unquoted)') { |
| 1102 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 1103 |
$self->{next_input_character} == 0x000A or # LF |
| 1104 |
$self->{next_input_character} == 0x000B or # HT |
| 1105 |
$self->{next_input_character} == 0x000C or # FF |
| 1106 |
$self->{next_input_character} == 0x0020) { # SP |
| 1107 |
$self->{state} = 'before attribute name'; |
| 1108 |
!!!next-input-character; |
| 1109 |
redo A; |
| 1110 |
} elsif ($self->{next_input_character} == 0x0026) { # & |
| 1111 |
$self->{last_attribute_value_state} = 'attribute value (unquoted)'; |
| 1112 |
$self->{state} = 'entity in attribute value'; |
| 1113 |
!!!next-input-character; |
| 1114 |
redo A; |
| 1115 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1116 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 1117 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 1118 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 1119 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 1120 |
if ($self->{current_token}->{attributes}) { |
| 1121 |
!!!parse-error (type => 'end tag attribute'); |
| 1122 |
} |
| 1123 |
} else { |
| 1124 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 1125 |
} |
| 1126 |
$self->{state} = 'data'; |
| 1127 |
!!!next-input-character; |
| 1128 |
|
| 1129 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 1130 |
undef $self->{current_token}; |
| 1131 |
|
| 1132 |
redo A; |
| 1133 |
} elsif ($self->{next_input_character} == 0x003C or # < |
| 1134 |
$self->{next_input_character} == -1) { |
| 1135 |
!!!parse-error (type => 'unclosed tag'); |
| 1136 |
if ($self->{current_token}->{type} eq 'start tag') { |
| 1137 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 1138 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 1139 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
| 1140 |
if ($self->{current_token}->{attributes}) { |
| 1141 |
!!!parse-error (type => 'end tag attribute'); |
| 1142 |
} |
| 1143 |
} else { |
| 1144 |
die "$0: $self->{current_token}->{type}: Unknown token type"; |
| 1145 |
} |
| 1146 |
$self->{state} = 'data'; |
| 1147 |
## reconsume |
| 1148 |
|
| 1149 |
!!!emit ($self->{current_token}); # start tag or end tag |
| 1150 |
undef $self->{current_token}; |
| 1151 |
|
| 1152 |
redo A; |
| 1153 |
} else { |
| 1154 |
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
| 1155 |
## Stay in the state |
| 1156 |
!!!next-input-character; |
| 1157 |
redo A; |
| 1158 |
} |
| 1159 |
} elsif ($self->{state} eq 'entity in attribute value') { |
| 1160 |
my $token = $self->_tokenize_attempt_to_consume_an_entity; |
| 1161 |
|
| 1162 |
unless (defined $token) { |
| 1163 |
$self->{current_attribute}->{value} .= '&'; |
| 1164 |
} else { |
| 1165 |
$self->{current_attribute}->{value} .= $token->{data}; |
| 1166 |
## ISSUE: spec says "append the returned character token to the current attribute's value" |
| 1167 |
} |
| 1168 |
|
| 1169 |
$self->{state} = $self->{last_attribute_value_state}; |
| 1170 |
# next-input-character is already done |
| 1171 |
redo A; |
| 1172 |
} elsif ($self->{state} eq 'bogus comment') { |
| 1173 |
## (only happen if PCDATA state) |
| 1174 |
|
| 1175 |
my $token = {type => 'comment', data => ''}; |
| 1176 |
|
| 1177 |
BC: { |
| 1178 |
if ($self->{next_input_character} == 0x003E) { # > |
| 1179 |
$self->{state} = 'data'; |
| 1180 |
!!!next-input-character; |
| 1181 |
|
| 1182 |
!!!emit ($token); |
| 1183 |
|
| 1184 |
redo A; |
| 1185 |
} elsif ($self->{next_input_character} == -1) { |
| 1186 |
$self->{state} = 'data'; |
| 1187 |
## reconsume |
| 1188 |
|
| 1189 |
!!!emit ($token); |
| 1190 |
|
| 1191 |
redo A; |
| 1192 |
} else { |
| 1193 |
$token->{data} .= chr ($self->{next_input_character}); |
| 1194 |
!!!next-input-character; |
| 1195 |
redo BC; |
| 1196 |
} |
| 1197 |
} # BC |
| 1198 |
} elsif ($self->{state} eq 'markup declaration open') { |
| 1199 |
## (only happen if PCDATA state) |
| 1200 |
|
| 1201 |
my @next_char; |
| 1202 |
push @next_char, $self->{next_input_character}; |
| 1203 |
|
| 1204 |
if ($self->{next_input_character} == 0x002D) { # - |
| 1205 |
!!!next-input-character; |
| 1206 |
push @next_char, $self->{next_input_character}; |
| 1207 |
if ($self->{next_input_character} == 0x002D) { # - |
| 1208 |
$self->{current_token} = {type => 'comment', data => ''}; |
| 1209 |
$self->{state} = 'comment'; |
| 1210 |
!!!next-input-character; |
| 1211 |
redo A; |
| 1212 |
} |
| 1213 |
} elsif ($self->{next_input_character} == 0x0044 or # D |
| 1214 |
$self->{next_input_character} == 0x0064) { # d |
| 1215 |
!!!next-input-character; |
| 1216 |
push @next_char, $self->{next_input_character}; |
| 1217 |
if ($self->{next_input_character} == 0x004F or # O |
| 1218 |
$self->{next_input_character} == 0x006F) { # o |
| 1219 |
!!!next-input-character; |
| 1220 |
push @next_char, $self->{next_input_character}; |
| 1221 |
if ($self->{next_input_character} == 0x0043 or # C |
| 1222 |
$self->{next_input_character} == 0x0063) { # c |
| 1223 |
!!!next-input-character; |
| 1224 |
push @next_char, $self->{next_input_character}; |
| 1225 |
if ($self->{next_input_character} == 0x0054 or # T |
| 1226 |
$self->{next_input_character} == 0x0074) { # t |
| 1227 |
!!!next-input-character; |
| 1228 |
push @next_char, $self->{next_input_character}; |
| 1229 |
if ($self->{next_input_character} == 0x0059 or # Y |
| 1230 |
$self->{next_input_character} == 0x0079) { # y |
| 1231 |
!!!next-input-character; |
| 1232 |
push @next_char, $self->{next_input_character}; |
| 1233 |
if ($self->{next_input_character} == 0x0050 or # P |
| 1234 |
$self->{next_input_character} == 0x0070) { # p |
| 1235 |
!!!next-input-character; |
| 1236 |
push @next_char, $self->{next_input_character}; |
| 1237 |
if ($self->{next_input_character} == 0x0045 or # E |
| 1238 |
$self->{next_input_character} == 0x0065) { # e |
| 1239 |
## ISSUE: What a stupid code this is! |
| 1240 |
$self->{state} = 'DOCTYPE'; |
| 1241 |
!!!next-input-character; |
| 1242 |
redo A; |
| 1243 |
} |
| 1244 |
} |
| 1245 |
} |
| 1246 |
} |
| 1247 |
} |
| 1248 |
} |
| 1249 |
} |
| 1250 |
|
| 1251 |
!!!parse-error (type => 'bogus comment open'); |
| 1252 |
$self->{next_input_character} = shift @next_char; |
| 1253 |
!!!back-next-input-character (@next_char); |
| 1254 |
$self->{state} = 'bogus comment'; |
| 1255 |
redo A; |
| 1256 |
|
| 1257 |
## ISSUE: typos in spec: chacacters, is is a parse error |
| 1258 |
## 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? |
| 1259 |
} elsif ($self->{state} eq 'comment') { |
| 1260 |
if ($self->{next_input_character} == 0x002D) { # - |
| 1261 |
$self->{state} = 'comment dash'; |
| 1262 |
!!!next-input-character; |
| 1263 |
redo A; |
| 1264 |
} elsif ($self->{next_input_character} == -1) { |
| 1265 |
!!!parse-error (type => 'unclosed comment'); |
| 1266 |
$self->{state} = 'data'; |
| 1267 |
## reconsume |
| 1268 |
|
| 1269 |
!!!emit ($self->{current_token}); # comment |
| 1270 |
undef $self->{current_token}; |
| 1271 |
|
| 1272 |
redo A; |
| 1273 |
} else { |
| 1274 |
$self->{current_token}->{data} .= chr ($self->{next_input_character}); # comment |
| 1275 |
## Stay in the state |
| 1276 |
!!!next-input-character; |
| 1277 |
redo A; |
| 1278 |
} |
| 1279 |
} elsif ($self->{state} eq 'comment dash') { |
| 1280 |
if ($self->{next_input_character} == 0x002D) { # - |
| 1281 |
$self->{state} = 'comment end'; |
| 1282 |
!!!next-input-character; |
| 1283 |
redo A; |
| 1284 |
} elsif ($self->{next_input_character} == -1) { |
| 1285 |
!!!parse-error (type => 'unclosed comment'); |
| 1286 |
$self->{state} = 'data'; |
| 1287 |
## reconsume |
| 1288 |
|
| 1289 |
!!!emit ($self->{current_token}); # comment |
| 1290 |
undef $self->{current_token}; |
| 1291 |
|
| 1292 |
redo A; |
| 1293 |
} else { |
| 1294 |
$self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment |
| 1295 |
$self->{state} = 'comment'; |
| 1296 |
!!!next-input-character; |
| 1297 |
redo A; |
| 1298 |
} |
| 1299 |
} elsif ($self->{state} eq 'comment end') { |
| 1300 |
if ($self->{next_input_character} == 0x003E) { # > |
| 1301 |
$self->{state} = 'data'; |
| 1302 |
!!!next-input-character; |
| 1303 |
|
| 1304 |
!!!emit ($self->{current_token}); # comment |
| 1305 |
undef $self->{current_token}; |
| 1306 |
|
| 1307 |
redo A; |
| 1308 |
} elsif ($self->{next_input_character} == 0x002D) { # - |
| 1309 |
!!!parse-error (type => 'dash in comment'); |
| 1310 |
$self->{current_token}->{data} .= '-'; # comment |
| 1311 |
## Stay in the state |
| 1312 |
!!!next-input-character; |
| 1313 |
redo A; |
| 1314 |
} elsif ($self->{next_input_character} == -1) { |
| 1315 |
!!!parse-error (type => 'unclosed comment'); |
| 1316 |
$self->{state} = 'data'; |
| 1317 |
## reconsume |
| 1318 |
|
| 1319 |
!!!emit ($self->{current_token}); # comment |
| 1320 |
undef $self->{current_token}; |
| 1321 |
|
| 1322 |
redo A; |
| 1323 |
} else { |
| 1324 |
!!!parse-error (type => 'dash in comment'); |
| 1325 |
$self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment |
| 1326 |
$self->{state} = 'comment'; |
| 1327 |
!!!next-input-character; |
| 1328 |
redo A; |
| 1329 |
} |
| 1330 |
} elsif ($self->{state} eq 'DOCTYPE') { |
| 1331 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 1332 |
$self->{next_input_character} == 0x000A or # LF |
| 1333 |
$self->{next_input_character} == 0x000B or # VT |
| 1334 |
$self->{next_input_character} == 0x000C or # FF |
| 1335 |
$self->{next_input_character} == 0x0020) { # SP |
| 1336 |
$self->{state} = 'before DOCTYPE name'; |
| 1337 |
!!!next-input-character; |
| 1338 |
redo A; |
| 1339 |
} else { |
| 1340 |
!!!parse-error (type => 'no space before DOCTYPE name'); |
| 1341 |
$self->{state} = 'before DOCTYPE name'; |
| 1342 |
## reconsume |
| 1343 |
redo A; |
| 1344 |
} |
| 1345 |
} elsif ($self->{state} eq 'before DOCTYPE name') { |
| 1346 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 1347 |
$self->{next_input_character} == 0x000A or # LF |
| 1348 |
$self->{next_input_character} == 0x000B or # VT |
| 1349 |
$self->{next_input_character} == 0x000C or # FF |
| 1350 |
$self->{next_input_character} == 0x0020) { # SP |
| 1351 |
## Stay in the state |
| 1352 |
!!!next-input-character; |
| 1353 |
redo A; |
| 1354 |
} elsif (0x0061 <= $self->{next_input_character} and |
| 1355 |
$self->{next_input_character} <= 0x007A) { # a..z |
| 1356 |
## ISSUE: "Set the token's name name to the" in the spec |
| 1357 |
$self->{current_token} = {type => 'DOCTYPE', |
| 1358 |
name => chr ($self->{next_input_character} - 0x0020), |
| 1359 |
error => 1}; |
| 1360 |
$self->{state} = 'DOCTYPE name'; |
| 1361 |
!!!next-input-character; |
| 1362 |
redo A; |
| 1363 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1364 |
!!!parse-error (type => 'no DOCTYPE name'); |
| 1365 |
$self->{state} = 'data'; |
| 1366 |
!!!next-input-character; |
| 1367 |
|
| 1368 |
!!!emit ({type => 'DOCTYPE', name => '', error => 1}); |
| 1369 |
|
| 1370 |
redo A; |
| 1371 |
} elsif ($self->{next_input_character} == -1) { |
| 1372 |
!!!parse-error (type => 'no DOCTYPE name'); |
| 1373 |
$self->{state} = 'data'; |
| 1374 |
## reconsume |
| 1375 |
|
| 1376 |
!!!emit ({type => 'DOCTYPE', name => '', error => 1}); |
| 1377 |
|
| 1378 |
redo A; |
| 1379 |
} else { |
| 1380 |
$self->{current_token} = {type => 'DOCTYPE', |
| 1381 |
name => chr ($self->{next_input_character}), |
| 1382 |
error => 1}; |
| 1383 |
## ISSUE: "Set the token's name name to the" in the spec |
| 1384 |
$self->{state} = 'DOCTYPE name'; |
| 1385 |
!!!next-input-character; |
| 1386 |
redo A; |
| 1387 |
} |
| 1388 |
} elsif ($self->{state} eq 'DOCTYPE name') { |
| 1389 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 1390 |
$self->{next_input_character} == 0x000A or # LF |
| 1391 |
$self->{next_input_character} == 0x000B or # VT |
| 1392 |
$self->{next_input_character} == 0x000C or # FF |
| 1393 |
$self->{next_input_character} == 0x0020) { # SP |
| 1394 |
$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE |
| 1395 |
$self->{state} = 'after DOCTYPE name'; |
| 1396 |
!!!next-input-character; |
| 1397 |
redo A; |
| 1398 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1399 |
$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE |
| 1400 |
$self->{state} = 'data'; |
| 1401 |
!!!next-input-character; |
| 1402 |
|
| 1403 |
!!!emit ($self->{current_token}); # DOCTYPE |
| 1404 |
undef $self->{current_token}; |
| 1405 |
|
| 1406 |
redo A; |
| 1407 |
} elsif (0x0061 <= $self->{next_input_character} and |
| 1408 |
$self->{next_input_character} <= 0x007A) { # a..z |
| 1409 |
$self->{current_token}->{name} .= chr ($self->{next_input_character} - 0x0020); # DOCTYPE |
| 1410 |
#$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); |
| 1411 |
## Stay in the state |
| 1412 |
!!!next-input-character; |
| 1413 |
redo A; |
| 1414 |
} elsif ($self->{next_input_character} == -1) { |
| 1415 |
!!!parse-error (type => 'unclosed DOCTYPE'); |
| 1416 |
$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE |
| 1417 |
$self->{state} = 'data'; |
| 1418 |
## reconsume |
| 1419 |
|
| 1420 |
!!!emit ($self->{current_token}); |
| 1421 |
undef $self->{current_token}; |
| 1422 |
|
| 1423 |
redo A; |
| 1424 |
} else { |
| 1425 |
$self->{current_token}->{name} |
| 1426 |
.= chr ($self->{next_input_character}); # DOCTYPE |
| 1427 |
#$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); |
| 1428 |
## Stay in the state |
| 1429 |
!!!next-input-character; |
| 1430 |
redo A; |
| 1431 |
} |
| 1432 |
} elsif ($self->{state} eq 'after DOCTYPE name') { |
| 1433 |
if ($self->{next_input_character} == 0x0009 or # HT |
| 1434 |
$self->{next_input_character} == 0x000A or # LF |
| 1435 |
$self->{next_input_character} == 0x000B or # VT |
| 1436 |
$self->{next_input_character} == 0x000C or # FF |
| 1437 |
$self->{next_input_character} == 0x0020) { # SP |
| 1438 |
## Stay in the state |
| 1439 |
!!!next-input-character; |
| 1440 |
redo A; |
| 1441 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1442 |
$self->{state} = 'data'; |
| 1443 |
!!!next-input-character; |
| 1444 |
|
| 1445 |
!!!emit ($self->{current_token}); # DOCTYPE |
| 1446 |
undef $self->{current_token}; |
| 1447 |
|
| 1448 |
redo A; |
| 1449 |
} elsif ($self->{next_input_character} == -1) { |
| 1450 |
!!!parse-error (type => 'unclosed DOCTYPE'); |
| 1451 |
$self->{state} = 'data'; |
| 1452 |
## reconsume |
| 1453 |
|
| 1454 |
!!!emit ($self->{current_token}); # DOCTYPE |
| 1455 |
undef $self->{current_token}; |
| 1456 |
|
| 1457 |
redo A; |
| 1458 |
} else { |
| 1459 |
!!!parse-error (type => 'string after DOCTYPE name'); |
| 1460 |
$self->{current_token}->{error} = 1; # DOCTYPE |
| 1461 |
$self->{state} = 'bogus DOCTYPE'; |
| 1462 |
!!!next-input-character; |
| 1463 |
redo A; |
| 1464 |
} |
| 1465 |
} elsif ($self->{state} eq 'bogus DOCTYPE') { |
| 1466 |
if ($self->{next_input_character} == 0x003E) { # > |
| 1467 |
$self->{state} = 'data'; |
| 1468 |
!!!next-input-character; |
| 1469 |
|
| 1470 |
!!!emit ($self->{current_token}); # DOCTYPE |
| 1471 |
undef $self->{current_token}; |
| 1472 |
|
| 1473 |
redo A; |
| 1474 |
} elsif ($self->{next_input_character} == -1) { |
| 1475 |
!!!parse-error (type => 'unclosed DOCTYPE'); |
| 1476 |
$self->{state} = 'data'; |
| 1477 |
## reconsume |
| 1478 |
|
| 1479 |
!!!emit ($self->{current_token}); # DOCTYPE |
| 1480 |
undef $self->{current_token}; |
| 1481 |
|
| 1482 |
redo A; |
| 1483 |
} else { |
| 1484 |
## Stay in the state |
| 1485 |
!!!next-input-character; |
| 1486 |
redo A; |
| 1487 |
} |
| 1488 |
} else { |
| 1489 |
die "$0: $self->{state}: Unknown state"; |
| 1490 |
} |
| 1491 |
} # A |
| 1492 |
|
| 1493 |
die "$0: _get_next_token: unexpected case"; |
| 1494 |
} # _get_next_token |
| 1495 |
|
| 1496 |
sub _tokenize_attempt_to_consume_an_entity ($) { |
| 1497 |
my $self = shift; |
| 1498 |
|
| 1499 |
if ($self->{next_input_character} == 0x0023) { # # |
| 1500 |
!!!next-input-character; |
| 1501 |
if ($self->{next_input_character} == 0x0078 or # x |
| 1502 |
$self->{next_input_character} == 0x0058) { # X |
| 1503 |
my $num; |
| 1504 |
X: { |
| 1505 |
my $x_char = $self->{next_input_character}; |
| 1506 |
!!!next-input-character; |
| 1507 |
if (0x0030 <= $self->{next_input_character} and |
| 1508 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
| 1509 |
$num ||= 0; |
| 1510 |
$num *= 0x10; |
| 1511 |
$num += $self->{next_input_character} - 0x0030; |
| 1512 |
redo X; |
| 1513 |
} elsif (0x0061 <= $self->{next_input_character} and |
| 1514 |
$self->{next_input_character} <= 0x0066) { # a..f |
| 1515 |
## ISSUE: the spec says U+0078, which is apparently incorrect |
| 1516 |
$num ||= 0; |
| 1517 |
$num *= 0x10; |
| 1518 |
$num += $self->{next_input_character} - 0x0060 + 9; |
| 1519 |
redo X; |
| 1520 |
} elsif (0x0041 <= $self->{next_input_character} and |
| 1521 |
$self->{next_input_character} <= 0x0046) { # A..F |
| 1522 |
## ISSUE: the spec says U+0058, which is apparently incorrect |
| 1523 |
$num ||= 0; |
| 1524 |
$num *= 0x10; |
| 1525 |
$num += $self->{next_input_character} - 0x0040 + 9; |
| 1526 |
redo X; |
| 1527 |
} elsif (not defined $num) { # no hexadecimal digit |
| 1528 |
!!!parse-error (type => 'bare hcro'); |
| 1529 |
$self->{next_input_character} = 0x0023; # # |
| 1530 |
!!!back-next-input-character ($x_char); |
| 1531 |
return undef; |
| 1532 |
} elsif ($self->{next_input_character} == 0x003B) { # ; |
| 1533 |
!!!next-input-character; |
| 1534 |
} else { |
| 1535 |
!!!parse-error (type => 'no refc'); |
| 1536 |
} |
| 1537 |
|
| 1538 |
## TODO: check the definition for |a valid Unicode character|. |
| 1539 |
## <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2006-December/thread.html#8189> |
| 1540 |
if ($num > 1114111 or $num == 0) { |
| 1541 |
$num = 0xFFFD; # REPLACEMENT CHARACTER |
| 1542 |
## ISSUE: Why this is not an error? |
| 1543 |
} elsif (0x80 <= $num and $num <= 0x9F) { |
| 1544 |
!!!parse-error (type => sprintf 'c1 entity:U+%04X', $num); |
| 1545 |
$num = $c1_entity_char->{$num}; |
| 1546 |
} |
| 1547 |
|
| 1548 |
return {type => 'character', data => chr $num}; |
| 1549 |
} # X |
| 1550 |
} elsif (0x0030 <= $self->{next_input_character} and |
| 1551 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
| 1552 |
my $code = $self->{next_input_character} - 0x0030; |
| 1553 |
!!!next-input-character; |
| 1554 |
|
| 1555 |
while (0x0030 <= $self->{next_input_character} and |
| 1556 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
| 1557 |
$code *= 10; |
| 1558 |
$code += $self->{next_input_character} - 0x0030; |
| 1559 |
|
| 1560 |
!!!next-input-character; |
| 1561 |
} |
| 1562 |
|
| 1563 |
if ($self->{next_input_character} == 0x003B) { # ; |
| 1564 |
!!!next-input-character; |
| 1565 |
} else { |
| 1566 |
!!!parse-error (type => 'no refc'); |
| 1567 |
} |
| 1568 |
|
| 1569 |
## TODO: check the definition for |a valid Unicode character|. |
| 1570 |
if ($code > 1114111 or $code == 0) { |
| 1571 |
$code = 0xFFFD; # REPLACEMENT CHARACTER |
| 1572 |
## ISSUE: Why this is not an error? |
| 1573 |
} elsif (0x80 <= $code and $code <= 0x9F) { |
| 1574 |
!!!parse-error (type => sprintf 'c1 entity:U+%04X', $code); |
| 1575 |
$code = $c1_entity_char->{$code}; |
| 1576 |
} |
| 1577 |
|
| 1578 |
return {type => 'character', data => chr $code}; |
| 1579 |
} else { |
| 1580 |
!!!parse-error (type => 'bare nero'); |
| 1581 |
!!!back-next-input-character ($self->{next_input_character}); |
| 1582 |
$self->{next_input_character} = 0x0023; # # |
| 1583 |
return undef; |
| 1584 |
} |
| 1585 |
} elsif ((0x0041 <= $self->{next_input_character} and |
| 1586 |
$self->{next_input_character} <= 0x005A) or |
| 1587 |
(0x0061 <= $self->{next_input_character} and |
| 1588 |
$self->{next_input_character} <= 0x007A)) { |
| 1589 |
my $entity_name = chr $self->{next_input_character}; |
| 1590 |
!!!next-input-character; |
| 1591 |
|
| 1592 |
my $value = $entity_name; |
| 1593 |
my $match; |
| 1594 |
|
| 1595 |
while (length $entity_name < 10 and |
| 1596 |
## NOTE: Some number greater than the maximum length of entity name |
| 1597 |
((0x0041 <= $self->{next_input_character} and |
| 1598 |
$self->{next_input_character} <= 0x005A) or |
| 1599 |
(0x0061 <= $self->{next_input_character} and |
| 1600 |
$self->{next_input_character} <= 0x007A) or |
| 1601 |
(0x0030 <= $self->{next_input_character} and |
| 1602 |
$self->{next_input_character} <= 0x0039))) { |
| 1603 |
$entity_name .= chr $self->{next_input_character}; |
| 1604 |
if (defined $entity_char->{$entity_name}) { |
| 1605 |
$value = $entity_char->{$entity_name}; |
| 1606 |
$match = 1; |
| 1607 |
} else { |
| 1608 |
$value .= chr $self->{next_input_character}; |
| 1609 |
} |
| 1610 |
!!!next-input-character; |
| 1611 |
} |
| 1612 |
|
| 1613 |
if ($match) { |
| 1614 |
if ($self->{next_input_character} == 0x003B) { # ; |
| 1615 |
!!!next-input-character; |
| 1616 |
} else { |
| 1617 |
!!!parse-error (type => 'refc'); |
| 1618 |
} |
| 1619 |
|
| 1620 |
return {type => 'character', data => $value}; |
| 1621 |
} else { |
| 1622 |
!!!parse-error (type => 'bare ero'); |
| 1623 |
## NOTE: No characters are consumed in the spec. |
| 1624 |
!!!back-token ({type => 'character', data => $value}); |
| 1625 |
return undef; |
| 1626 |
} |
| 1627 |
} else { |
| 1628 |
## no characters are consumed |
| 1629 |
!!!parse-error (type => 'bare ero'); |
| 1630 |
return undef; |
| 1631 |
} |
| 1632 |
} # _tokenize_attempt_to_consume_an_entity |
| 1633 |
|
| 1634 |
sub _initialize_tree_constructor ($) { |
| 1635 |
my $self = shift; |
| 1636 |
## NOTE: $self->{document} MUST be specified before this method is called |
| 1637 |
$self->{document}->strict_error_checking (0); |
| 1638 |
## TODO: Turn mutation events off # MUST |
| 1639 |
## TODO: Turn loose Document option (manakai extension) on |
| 1640 |
## TODO: Mark the Document as an HTML document # MUST |
| 1641 |
} # _initialize_tree_constructor |
| 1642 |
|
| 1643 |
sub _terminate_tree_constructor ($) { |
| 1644 |
my $self = shift; |
| 1645 |
$self->{document}->strict_error_checking (1); |
| 1646 |
## TODO: Turn mutation events on |
| 1647 |
} # _terminate_tree_constructor |
| 1648 |
|
| 1649 |
## ISSUE: Should append_child (for example) in script executed in tree construction stage fire mutation events? |
| 1650 |
|
| 1651 |
{ # tree construction stage |
| 1652 |
my $token; |
| 1653 |
|
| 1654 |
sub _construct_tree ($) { |
| 1655 |
my ($self) = @_; |
| 1656 |
|
| 1657 |
## When an interactive UA render the $self->{document} available |
| 1658 |
## to the user, or when it begin accepting user input, are |
| 1659 |
## not defined. |
| 1660 |
|
| 1661 |
## Append a character: collect it and all subsequent consecutive |
| 1662 |
## characters and insert one Text node whose data is concatenation |
| 1663 |
## of all those characters. # MUST |
| 1664 |
|
| 1665 |
!!!next-token; |
| 1666 |
|
| 1667 |
$self->{insertion_mode} = 'before head'; |
| 1668 |
undef $self->{form_element}; |
| 1669 |
undef $self->{head_element}; |
| 1670 |
$self->{open_elements} = []; |
| 1671 |
undef $self->{inner_html_node}; |
| 1672 |
|
| 1673 |
$self->_tree_construction_initial; # MUST |
| 1674 |
$self->_tree_construction_root_element; |
| 1675 |
$self->_tree_construction_main; |
| 1676 |
} # _construct_tree |
| 1677 |
|
| 1678 |
sub _tree_construction_initial ($) { |
| 1679 |
my $self = shift; |
| 1680 |
B: { |
| 1681 |
if ($token->{type} eq 'DOCTYPE') { |
| 1682 |
if ($token->{error}) { |
| 1683 |
## ISSUE: Spec currently left this case undefined. |
| 1684 |
!!!parse-error (type => 'bogus DOCTYPE'); |
| 1685 |
} |
| 1686 |
my $doctype = $self->{document}->create_document_type_definition |
| 1687 |
($token->{name}); |
| 1688 |
$self->{document}->append_child ($doctype); |
| 1689 |
#$phase = 'root element'; |
| 1690 |
!!!next-token; |
| 1691 |
#redo B; |
| 1692 |
return; |
| 1693 |
} elsif ({ |
| 1694 |
comment => 1, |
| 1695 |
'start tag' => 1, |
| 1696 |
'end tag' => 1, |
| 1697 |
'end-of-file' => 1, |
| 1698 |
}->{$token->{type}}) { |
| 1699 |
## ISSUE: Spec currently left this case undefined. |
| 1700 |
!!!parse-error (type => 'missing DOCTYPE'); |
| 1701 |
#$phase = 'root element'; |
| 1702 |
## reprocess |
| 1703 |
#redo B; |
| 1704 |
return; |
| 1705 |
} elsif ($token->{type} eq 'character') { |
| 1706 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 1707 |
$self->{document}->manakai_append_text ($1); |
| 1708 |
## ISSUE: DOM3 Core does not allow Document > Text |
| 1709 |
unless (length $token->{data}) { |
| 1710 |
## Stay in the phase |
| 1711 |
!!!next-token; |
| 1712 |
redo B; |
| 1713 |
} |
| 1714 |
} |
| 1715 |
## ISSUE: Spec currently left this case undefined. |
| 1716 |
!!!parse-error (type => 'missing DOCTYPE'); |
| 1717 |
#$phase = 'root element'; |
| 1718 |
## reprocess |
| 1719 |
#redo B; |
| 1720 |
return; |
| 1721 |
} else { |
| 1722 |
die "$0: $token->{type}: Unknown token"; |
| 1723 |
} |
| 1724 |
} # B |
| 1725 |
} # _tree_construction_initial |
| 1726 |
|
| 1727 |
sub _tree_construction_root_element ($) { |
| 1728 |
my $self = shift; |
| 1729 |
|
| 1730 |
B: { |
| 1731 |
if ($token->{type} eq 'DOCTYPE') { |
| 1732 |
!!!parse-error (type => 'in html:#DOCTYPE'); |
| 1733 |
## Ignore the token |
| 1734 |
## Stay in the phase |
| 1735 |
!!!next-token; |
| 1736 |
redo B; |
| 1737 |
} elsif ($token->{type} eq 'comment') { |
| 1738 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 1739 |
$self->{document}->append_child ($comment); |
| 1740 |
## Stay in the phase |
| 1741 |
!!!next-token; |
| 1742 |
redo B; |
| 1743 |
} elsif ($token->{type} eq 'character') { |
| 1744 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 1745 |
$self->{document}->manakai_append_text ($1); |
| 1746 |
## ISSUE: DOM3 Core does not allow Document > Text |
| 1747 |
unless (length $token->{data}) { |
| 1748 |
## Stay in the phase |
| 1749 |
!!!next-token; |
| 1750 |
redo B; |
| 1751 |
} |
| 1752 |
} |
| 1753 |
# |
| 1754 |
} elsif ({ |
| 1755 |
'start tag' => 1, |
| 1756 |
'end tag' => 1, |
| 1757 |
'end-of-file' => 1, |
| 1758 |
}->{$token->{type}}) { |
| 1759 |
## ISSUE: There is an issue in the spec |
| 1760 |
# |
| 1761 |
} else { |
| 1762 |
die "$0: $token->{type}: Unknown token"; |
| 1763 |
} |
| 1764 |
my $root_element; !!!create-element ($root_element, 'html'); |
| 1765 |
$self->{document}->append_child ($root_element); |
| 1766 |
push @{$self->{open_elements}}, [$root_element, 'html']; |
| 1767 |
#$phase = 'main'; |
| 1768 |
## reprocess |
| 1769 |
#redo B; |
| 1770 |
return; |
| 1771 |
} # B |
| 1772 |
} # _tree_construction_root_element |
| 1773 |
|
| 1774 |
sub _reset_insertion_mode ($) { |
| 1775 |
my $self = shift; |
| 1776 |
|
| 1777 |
## Step 1 |
| 1778 |
my $last; |
| 1779 |
|
| 1780 |
## Step 2 |
| 1781 |
my $i = -1; |
| 1782 |
my $node = $self->{open_elements}->[$i]; |
| 1783 |
|
| 1784 |
## Step 3 |
| 1785 |
S3: { |
| 1786 |
$last = 1 if $self->{open_elements}->[0]->[0] eq $node->[0]; |
| 1787 |
if (defined $self->{inner_html_node}) { |
| 1788 |
if ($self->{inner_html_node}->[1] eq 'td' or |
| 1789 |
$self->{inner_html_node}->[1] eq 'th') { |
| 1790 |
# |
| 1791 |
} else { |
| 1792 |
$node = $self->{inner_html_node}; |
| 1793 |
} |
| 1794 |
} |
| 1795 |
|
| 1796 |
## Step 4..13 |
| 1797 |
my $new_mode = { |
| 1798 |
select => 'in select', |
| 1799 |
td => 'in cell', |
| 1800 |
th => 'in cell', |
| 1801 |
tr => 'in row', |
| 1802 |
tbody => 'in table body', |
| 1803 |
thead => 'in table head', |
| 1804 |
tfoot => 'in table foot', |
| 1805 |
caption => 'in caption', |
| 1806 |
colgroup => 'in column group', |
| 1807 |
table => 'in table', |
| 1808 |
head => 'in body', # not in head! |
| 1809 |
body => 'in body', |
| 1810 |
frameset => 'in frameset', |
| 1811 |
}->{$node->[1]}; |
| 1812 |
$self->{insertion_mode} = $new_mode and return if defined $new_mode; |
| 1813 |
|
| 1814 |
## Step 14 |
| 1815 |
if ($node->[1] eq 'html') { |
| 1816 |
unless (defined $self->{head_element}) { |
| 1817 |
$self->{insertion_mode} = 'before head'; |
| 1818 |
} else { |
| 1819 |
$self->{insertion_mode} = 'after head'; |
| 1820 |
} |
| 1821 |
return; |
| 1822 |
} |
| 1823 |
|
| 1824 |
## Step 15 |
| 1825 |
$self->{insertion_mode} = 'in body' and return if $last; |
| 1826 |
|
| 1827 |
## Step 16 |
| 1828 |
$i--; |
| 1829 |
$node = $self->{open_elements}->[$i]; |
| 1830 |
|
| 1831 |
## Step 17 |
| 1832 |
redo S3; |
| 1833 |
} # S3 |
| 1834 |
} # _reset_insertion_mode |
| 1835 |
|
| 1836 |
sub _tree_construction_main ($) { |
| 1837 |
my $self = shift; |
| 1838 |
|
| 1839 |
my $phase = 'main'; |
| 1840 |
|
| 1841 |
my $active_formatting_elements = []; |
| 1842 |
|
| 1843 |
my $reconstruct_active_formatting_elements = sub { # MUST |
| 1844 |
my $insert = shift; |
| 1845 |
|
| 1846 |
## Step 1 |
| 1847 |
return unless @$active_formatting_elements; |
| 1848 |
|
| 1849 |
## Step 3 |
| 1850 |
my $i = -1; |
| 1851 |
my $entry = $active_formatting_elements->[$i]; |
| 1852 |
|
| 1853 |
## Step 2 |
| 1854 |
return if $entry->[0] eq '#marker'; |
| 1855 |
for (@{$self->{open_elements}}) { |
| 1856 |
if ($entry->[0] eq $_->[0]) { |
| 1857 |
return; |
| 1858 |
} |
| 1859 |
} |
| 1860 |
|
| 1861 |
S4: { |
| 1862 |
## Step 4 |
| 1863 |
last S4 if $active_formatting_elements->[0]->[0] eq $entry->[0]; |
| 1864 |
|
| 1865 |
## Step 5 |
| 1866 |
$i--; |
| 1867 |
$entry = $active_formatting_elements->[$i]; |
| 1868 |
|
| 1869 |
## Step 6 |
| 1870 |
if ($entry->[0] eq '#marker') { |
| 1871 |
# |
| 1872 |
} else { |
| 1873 |
my $in_open_elements; |
| 1874 |
OE: for (@{$self->{open_elements}}) { |
| 1875 |
if ($entry->[0] eq $_->[0]) { |
| 1876 |
$in_open_elements = 1; |
| 1877 |
last OE; |
| 1878 |
} |
| 1879 |
} |
| 1880 |
if ($in_open_elements) { |
| 1881 |
# |
| 1882 |
} else { |
| 1883 |
redo S4; |
| 1884 |
} |
| 1885 |
} |
| 1886 |
|
| 1887 |
## Step 7 |
| 1888 |
$i++; |
| 1889 |
$entry = $active_formatting_elements->[$i]; |
| 1890 |
} # S4 |
| 1891 |
|
| 1892 |
S7: { |
| 1893 |
## Step 8 |
| 1894 |
my $clone = [$entry->[0]->clone_node (0), $entry->[1]]; |
| 1895 |
|
| 1896 |
## Step 9 |
| 1897 |
$insert->($clone->[0]); |
| 1898 |
push @{$self->{open_elements}}, $clone; |
| 1899 |
|
| 1900 |
## Step 10 |
| 1901 |
$active_formatting_elements->[$i] = $self->{open_elements}->[-1]; |
| 1902 |
|
| 1903 |
## Step 11 |
| 1904 |
unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { |
| 1905 |
## Step 7' |
| 1906 |
$i++; |
| 1907 |
$entry = $active_formatting_elements->[$i]; |
| 1908 |
|
| 1909 |
redo S7; |
| 1910 |
} |
| 1911 |
} # S7 |
| 1912 |
}; # $reconstruct_active_formatting_elements |
| 1913 |
|
| 1914 |
my $clear_up_to_marker = sub { |
| 1915 |
for (reverse 0..$#$active_formatting_elements) { |
| 1916 |
if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 1917 |
splice @$active_formatting_elements, $_; |
| 1918 |
return; |
| 1919 |
} |
| 1920 |
} |
| 1921 |
}; # $clear_up_to_marker |
| 1922 |
|
| 1923 |
my $style_start_tag = sub { |
| 1924 |
my $style_el; !!!create-element ($style_el, 'style', $token->{attributes}); |
| 1925 |
## $self->{insertion_mode} eq 'in head' and ... (always true) |
| 1926 |
(($self->{insertion_mode} eq 'in head' and defined $self->{head_element}) |
| 1927 |
? $self->{head_element} : $self->{open_elements}->[-1]->[0]) |
| 1928 |
->append_child ($style_el); |
| 1929 |
$self->{content_model_flag} = 'CDATA'; |
| 1930 |
|
| 1931 |
my $text = ''; |
| 1932 |
!!!next-token; |
| 1933 |
while ($token->{type} eq 'character') { |
| 1934 |
$text .= $token->{data}; |
| 1935 |
!!!next-token; |
| 1936 |
} # stop if non-character token or tokenizer stops tokenising |
| 1937 |
if (length $text) { |
| 1938 |
$style_el->manakai_append_text ($text); |
| 1939 |
} |
| 1940 |
|
| 1941 |
$self->{content_model_flag} = 'PCDATA'; |
| 1942 |
|
| 1943 |
if ($token->{type} eq 'end tag' and $token->{tag_name} eq 'style') { |
| 1944 |
## Ignore the token |
| 1945 |
} else { |
| 1946 |
!!!parse-error (type => 'in CDATA:#'.$token->{type}); |
| 1947 |
## ISSUE: And ignore? |
| 1948 |
} |
| 1949 |
!!!next-token; |
| 1950 |
}; # $style_start_tag |
| 1951 |
|
| 1952 |
my $script_start_tag = sub { |
| 1953 |
my $script_el; |
| 1954 |
!!!create-element ($script_el, 'script', $token->{attributes}); |
| 1955 |
## TODO: mark as "parser-inserted" |
| 1956 |
|
| 1957 |
$self->{content_model_flag} = 'CDATA'; |
| 1958 |
|
| 1959 |
my $text = ''; |
| 1960 |
!!!next-token; |
| 1961 |
while ($token->{type} eq 'character') { |
| 1962 |
$text .= $token->{data}; |
| 1963 |
!!!next-token; |
| 1964 |
} # stop if non-character token or tokenizer stops tokenising |
| 1965 |
if (length $text) { |
| 1966 |
$script_el->manakai_append_text ($text); |
| 1967 |
} |
| 1968 |
|
| 1969 |
$self->{content_model_flag} = 'PCDATA'; |
| 1970 |
|
| 1971 |
if ($token->{type} eq 'end tag' and |
| 1972 |
$token->{tag_name} eq 'script') { |
| 1973 |
## Ignore the token |
| 1974 |
} else { |
| 1975 |
!!!parse-error (type => 'in CDATA:#'.$token->{type}); |
| 1976 |
## ISSUE: And ignore? |
| 1977 |
## TODO: mark as "already executed" |
| 1978 |
} |
| 1979 |
|
| 1980 |
if (defined $self->{inner_html_node}) { |
| 1981 |
## TODO: mark as "already executed" |
| 1982 |
} else { |
| 1983 |
## TODO: $old_insertion_point = current insertion point |
| 1984 |
## TODO: insertion point = just before the next input character |
| 1985 |
|
| 1986 |
(($self->{insertion_mode} eq 'in head' and defined $self->{head_element}) |
| 1987 |
? $self->{head_element} : $self->{open_elements}->[-1]->[0])->append_child ($script_el); |
| 1988 |
|
| 1989 |
## TODO: insertion point = $old_insertion_point (might be "undefined") |
| 1990 |
|
| 1991 |
## TODO: if there is a script that will execute as soon as the parser resume, then... |
| 1992 |
} |
| 1993 |
|
| 1994 |
!!!next-token; |
| 1995 |
}; # $script_start_tag |
| 1996 |
|
| 1997 |
my $formatting_end_tag = sub { |
| 1998 |
my $tag_name = shift; |
| 1999 |
|
| 2000 |
FET: { |
| 2001 |
## Step 1 |
| 2002 |
my $formatting_element; |
| 2003 |
my $formatting_element_i_in_active; |
| 2004 |
AFE: for (reverse 0..$#$active_formatting_elements) { |
| 2005 |
if ($active_formatting_elements->[$_]->[1] eq $tag_name) { |
| 2006 |
$formatting_element = $active_formatting_elements->[$_]; |
| 2007 |
$formatting_element_i_in_active = $_; |
| 2008 |
last AFE; |
| 2009 |
} elsif ($active_formatting_elements->[$_]->[0] eq '#marker') { |
| 2010 |
last AFE; |
| 2011 |
} |
| 2012 |
} # AFE |
| 2013 |
unless (defined $formatting_element) { |
| 2014 |
!!!parse-error (type => 'unmatched end tag:'.$tag_name); |
| 2015 |
## Ignore the token |
| 2016 |
!!!next-token; |
| 2017 |
return; |
| 2018 |
} |
| 2019 |
## has an element in scope |
| 2020 |
my $in_scope = 1; |
| 2021 |
my $formatting_element_i_in_open; |
| 2022 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2023 |
my $node = $self->{open_elements}->[$_]; |
| 2024 |
if ($node->[0] eq $formatting_element->[0]) { |
| 2025 |
if ($in_scope) { |
| 2026 |
$formatting_element_i_in_open = $_; |
| 2027 |
last INSCOPE; |
| 2028 |
} else { # in open elements but not in scope |
| 2029 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2030 |
## Ignore the token |
| 2031 |
!!!next-token; |
| 2032 |
return; |
| 2033 |
} |
| 2034 |
} elsif ({ |
| 2035 |
table => 1, caption => 1, td => 1, th => 1, |
| 2036 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2037 |
}->{$node->[1]}) { |
| 2038 |
$in_scope = 0; |
| 2039 |
} |
| 2040 |
} # INSCOPE |
| 2041 |
unless (defined $formatting_element_i_in_open) { |
| 2042 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2043 |
pop @$active_formatting_elements; # $formatting_element |
| 2044 |
!!!next-token; ## TODO: ok? |
| 2045 |
return; |
| 2046 |
} |
| 2047 |
if (not $self->{open_elements}->[-1]->[0] eq $formatting_element->[0]) { |
| 2048 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2049 |
} |
| 2050 |
|
| 2051 |
## Step 2 |
| 2052 |
my $furthest_block; |
| 2053 |
my $furthest_block_i_in_open; |
| 2054 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2055 |
my $node = $self->{open_elements}->[$_]; |
| 2056 |
if (not $formatting_category->{$node->[1]} and |
| 2057 |
#not $phrasing_category->{$node->[1]} and |
| 2058 |
($special_category->{$node->[1]} or |
| 2059 |
$scoping_category->{$node->[1]})) { |
| 2060 |
$furthest_block = $node; |
| 2061 |
$furthest_block_i_in_open = $_; |
| 2062 |
} elsif ($node->[0] eq $formatting_element->[0]) { |
| 2063 |
last OE; |
| 2064 |
} |
| 2065 |
} # OE |
| 2066 |
|
| 2067 |
## Step 3 |
| 2068 |
unless (defined $furthest_block) { # MUST |
| 2069 |
splice @{$self->{open_elements}}, $formatting_element_i_in_open; |
| 2070 |
splice @$active_formatting_elements, $formatting_element_i_in_active, 1; |
| 2071 |
!!!next-token; |
| 2072 |
return; |
| 2073 |
} |
| 2074 |
|
| 2075 |
## Step 4 |
| 2076 |
my $common_ancestor_node = $self->{open_elements}->[$formatting_element_i_in_open - 1]; |
| 2077 |
|
| 2078 |
## Step 5 |
| 2079 |
my $furthest_block_parent = $furthest_block->[0]->parent_node; |
| 2080 |
if (defined $furthest_block_parent) { |
| 2081 |
$furthest_block_parent->remove_child ($furthest_block->[0]); |
| 2082 |
} |
| 2083 |
|
| 2084 |
## Step 6 |
| 2085 |
my $bookmark_prev_el |
| 2086 |
= $active_formatting_elements->[$formatting_element_i_in_active - 1] |
| 2087 |
->[0]; |
| 2088 |
|
| 2089 |
## Step 7 |
| 2090 |
my $node = $furthest_block; |
| 2091 |
my $node_i_in_open = $furthest_block_i_in_open; |
| 2092 |
my $last_node = $furthest_block; |
| 2093 |
S7: { |
| 2094 |
## Step 1 |
| 2095 |
$node_i_in_open--; |
| 2096 |
$node = $self->{open_elements}->[$node_i_in_open]; |
| 2097 |
|
| 2098 |
## Step 2 |
| 2099 |
my $node_i_in_active; |
| 2100 |
S7S2: { |
| 2101 |
for (reverse 0..$#$active_formatting_elements) { |
| 2102 |
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
| 2103 |
$node_i_in_active = $_; |
| 2104 |
last S7S2; |
| 2105 |
} |
| 2106 |
} |
| 2107 |
splice @{$self->{open_elements}}, $node_i_in_open, 1; |
| 2108 |
redo S7; |
| 2109 |
} # S7S2 |
| 2110 |
|
| 2111 |
## Step 3 |
| 2112 |
last S7 if $node->[0] eq $formatting_element->[0]; |
| 2113 |
|
| 2114 |
## Step 4 |
| 2115 |
if ($last_node->[0] eq $furthest_block->[0]) { |
| 2116 |
$bookmark_prev_el = $node->[0]; |
| 2117 |
} |
| 2118 |
|
| 2119 |
## Step 5 |
| 2120 |
if ($node->[0]->has_child_nodes ()) { |
| 2121 |
my $clone = [$node->[0]->clone_node (0), $node->[1]]; |
| 2122 |
$active_formatting_elements->[$node_i_in_active] = $clone; |
| 2123 |
$self->{open_elements}->[$node_i_in_open] = $clone; |
| 2124 |
$node = $clone; |
| 2125 |
} |
| 2126 |
|
| 2127 |
## Step 6 |
| 2128 |
$node->[0]->append_child ($last_node->[0]); |
| 2129 |
|
| 2130 |
## Step 7 |
| 2131 |
$last_node = $node; |
| 2132 |
|
| 2133 |
## Step 8 |
| 2134 |
redo S7; |
| 2135 |
} # S7 |
| 2136 |
|
| 2137 |
## Step 8 |
| 2138 |
$common_ancestor_node->[0]->append_child ($last_node->[0]); |
| 2139 |
|
| 2140 |
## Step 9 |
| 2141 |
my $clone = [$formatting_element->[0]->clone_node (0), |
| 2142 |
$formatting_element->[1]]; |
| 2143 |
|
| 2144 |
## Step 10 |
| 2145 |
my @cn = @{$furthest_block->[0]->child_nodes}; |
| 2146 |
$clone->[0]->append_child ($_) for @cn; |
| 2147 |
|
| 2148 |
## Step 11 |
| 2149 |
$furthest_block->[0]->append_child ($clone->[0]); |
| 2150 |
|
| 2151 |
## Step 12 |
| 2152 |
my $i; |
| 2153 |
AFE: for (reverse 0..$#$active_formatting_elements) { |
| 2154 |
if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { |
| 2155 |
splice @$active_formatting_elements, $_, 1; |
| 2156 |
$i-- and last AFE if defined $i; |
| 2157 |
} elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { |
| 2158 |
$i = $_; |
| 2159 |
} |
| 2160 |
} # AFE |
| 2161 |
splice @$active_formatting_elements, $i + 1, 0, $clone; |
| 2162 |
|
| 2163 |
## Step 13 |
| 2164 |
undef $i; |
| 2165 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2166 |
if ($self->{open_elements}->[$_]->[0] eq $formatting_element->[0]) { |
| 2167 |
splice @{$self->{open_elements}}, $_, 1; |
| 2168 |
$i-- and last OE if defined $i; |
| 2169 |
} elsif ($self->{open_elements}->[$_]->[0] eq $furthest_block->[0]) { |
| 2170 |
$i = $_; |
| 2171 |
} |
| 2172 |
} # OE |
| 2173 |
splice @{$self->{open_elements}}, $i + 1, 1, $clone; |
| 2174 |
|
| 2175 |
## Step 14 |
| 2176 |
redo FET; |
| 2177 |
} # FET |
| 2178 |
}; # $formatting_end_tag |
| 2179 |
|
| 2180 |
my $insert_to_current = sub { |
| 2181 |
$self->{open_elements}->[-1]->[0]->append_child (shift); |
| 2182 |
}; # $insert_to_current |
| 2183 |
|
| 2184 |
my $insert_to_foster = sub { |
| 2185 |
my $child = shift; |
| 2186 |
if ({ |
| 2187 |
table => 1, tbody => 1, tfoot => 1, |
| 2188 |
thead => 1, tr => 1, |
| 2189 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 2190 |
# MUST |
| 2191 |
my $foster_parent_element; |
| 2192 |
my $next_sibling; |
| 2193 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2194 |
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
| 2195 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 2196 |
if (defined $parent and $parent->node_type == 1) { |
| 2197 |
$foster_parent_element = $parent; |
| 2198 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 2199 |
} else { |
| 2200 |
$foster_parent_element |
| 2201 |
= $self->{open_elements}->[$_ - 1]->[0]; |
| 2202 |
} |
| 2203 |
last OE; |
| 2204 |
} |
| 2205 |
} # OE |
| 2206 |
$foster_parent_element = $self->{open_elements}->[0]->[0] |
| 2207 |
unless defined $foster_parent_element; |
| 2208 |
$foster_parent_element->insert_before |
| 2209 |
($child, $next_sibling); |
| 2210 |
} else { |
| 2211 |
$self->{open_elements}->[-1]->[0]->append_child ($child); |
| 2212 |
} |
| 2213 |
}; # $insert_to_foster |
| 2214 |
|
| 2215 |
my $in_body = sub { |
| 2216 |
my $insert = shift; |
| 2217 |
if ($token->{type} eq 'start tag') { |
| 2218 |
if ($token->{tag_name} eq 'script') { |
| 2219 |
$script_start_tag->(); |
| 2220 |
return; |
| 2221 |
} elsif ($token->{tag_name} eq 'style') { |
| 2222 |
$style_start_tag->(); |
| 2223 |
return; |
| 2224 |
} elsif ({ |
| 2225 |
base => 1, link => 1, meta => 1, |
| 2226 |
}->{$token->{tag_name}}) { |
| 2227 |
!!!parse-error (type => 'in body:'.$token->{tag_name}); |
| 2228 |
## NOTE: This is an "as if in head" code clone |
| 2229 |
my $el; |
| 2230 |
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
| 2231 |
if (defined $self->{head_element}) { |
| 2232 |
$self->{head_element}->append_child ($el); |
| 2233 |
} else { |
| 2234 |
$insert->($el); |
| 2235 |
} |
| 2236 |
|
| 2237 |
!!!next-token; |
| 2238 |
return; |
| 2239 |
} elsif ($token->{tag_name} eq 'title') { |
| 2240 |
!!!parse-error (type => 'in body:title'); |
| 2241 |
## NOTE: There is an "as if in head" code clone |
| 2242 |
my $title_el; |
| 2243 |
!!!create-element ($title_el, 'title', $token->{attributes}); |
| 2244 |
(defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) |
| 2245 |
->append_child ($title_el); |
| 2246 |
$self->{content_model_flag} = 'RCDATA'; |
| 2247 |
|
| 2248 |
my $text = ''; |
| 2249 |
!!!next-token; |
| 2250 |
while ($token->{type} eq 'character') { |
| 2251 |
$text .= $token->{data}; |
| 2252 |
!!!next-token; |
| 2253 |
} |
| 2254 |
if (length $text) { |
| 2255 |
$title_el->manakai_append_text ($text); |
| 2256 |
} |
| 2257 |
|
| 2258 |
$self->{content_model_flag} = 'PCDATA'; |
| 2259 |
|
| 2260 |
if ($token->{type} eq 'end tag' and |
| 2261 |
$token->{tag_name} eq 'title') { |
| 2262 |
## Ignore the token |
| 2263 |
} else { |
| 2264 |
!!!parse-error (type => 'in RCDATA:#'.$token->{type}); |
| 2265 |
## ISSUE: And ignore? |
| 2266 |
} |
| 2267 |
!!!next-token; |
| 2268 |
return; |
| 2269 |
} elsif ($token->{tag_name} eq 'body') { |
| 2270 |
!!!parse-error (type => 'in body:body'); |
| 2271 |
|
| 2272 |
if (@{$self->{open_elements}} == 1 or |
| 2273 |
$self->{open_elements}->[1]->[1] ne 'body') { |
| 2274 |
## Ignore the token |
| 2275 |
} else { |
| 2276 |
my $body_el = $self->{open_elements}->[1]->[0]; |
| 2277 |
for my $attr_name (keys %{$token->{attributes}}) { |
| 2278 |
unless ($body_el->has_attribute_ns (undef, $attr_name)) { |
| 2279 |
$body_el->set_attribute_ns |
| 2280 |
(undef, [undef, $attr_name], |
| 2281 |
$token->{attributes}->{$attr_name}->{value}); |
| 2282 |
} |
| 2283 |
} |
| 2284 |
} |
| 2285 |
!!!next-token; |
| 2286 |
return; |
| 2287 |
} elsif ({ |
| 2288 |
address => 1, blockquote => 1, center => 1, dir => 1, |
| 2289 |
div => 1, dl => 1, fieldset => 1, listing => 1, |
| 2290 |
menu => 1, ol => 1, p => 1, ul => 1, |
| 2291 |
pre => 1, |
| 2292 |
}->{$token->{tag_name}}) { |
| 2293 |
## has a p element in scope |
| 2294 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2295 |
if ($_->[1] eq 'p') { |
| 2296 |
!!!back-token; |
| 2297 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2298 |
return; |
| 2299 |
} elsif ({ |
| 2300 |
table => 1, caption => 1, td => 1, th => 1, |
| 2301 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2302 |
}->{$_->[1]}) { |
| 2303 |
last INSCOPE; |
| 2304 |
} |
| 2305 |
} # INSCOPE |
| 2306 |
|
| 2307 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2308 |
if ($token->{tag_name} eq 'pre') { |
| 2309 |
!!!next-token; |
| 2310 |
if ($token->{type} eq 'character') { |
| 2311 |
$token->{data} =~ s/^\x0A//; |
| 2312 |
unless (length $token->{data}) { |
| 2313 |
!!!next-token; |
| 2314 |
} |
| 2315 |
} |
| 2316 |
} else { |
| 2317 |
!!!next-token; |
| 2318 |
} |
| 2319 |
return; |
| 2320 |
} elsif ($token->{tag_name} eq 'form') { |
| 2321 |
if (defined $self->{form_element}) { |
| 2322 |
!!!parse-error (type => 'in form:form'); |
| 2323 |
## Ignore the token |
| 2324 |
!!!next-token; |
| 2325 |
return; |
| 2326 |
} else { |
| 2327 |
## has a p element in scope |
| 2328 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2329 |
if ($_->[1] eq 'p') { |
| 2330 |
!!!back-token; |
| 2331 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2332 |
return; |
| 2333 |
} elsif ({ |
| 2334 |
table => 1, caption => 1, td => 1, th => 1, |
| 2335 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2336 |
}->{$_->[1]}) { |
| 2337 |
last INSCOPE; |
| 2338 |
} |
| 2339 |
} # INSCOPE |
| 2340 |
|
| 2341 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2342 |
$self->{form_element} = $self->{open_elements}->[-1]->[0]; |
| 2343 |
!!!next-token; |
| 2344 |
return; |
| 2345 |
} |
| 2346 |
} elsif ($token->{tag_name} eq 'li') { |
| 2347 |
## has a p element in scope |
| 2348 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2349 |
if ($_->[1] eq 'p') { |
| 2350 |
!!!back-token; |
| 2351 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2352 |
return; |
| 2353 |
} elsif ({ |
| 2354 |
table => 1, caption => 1, td => 1, th => 1, |
| 2355 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2356 |
}->{$_->[1]}) { |
| 2357 |
last INSCOPE; |
| 2358 |
} |
| 2359 |
} # INSCOPE |
| 2360 |
|
| 2361 |
## Step 1 |
| 2362 |
my $i = -1; |
| 2363 |
my $node = $self->{open_elements}->[$i]; |
| 2364 |
LI: { |
| 2365 |
## Step 2 |
| 2366 |
if ($node->[1] eq 'li') { |
| 2367 |
if ($i != -1) { |
| 2368 |
!!!parse-error (type => 'end tag missing:'. |
| 2369 |
$self->{open_elements}->[-1]->[1]); |
| 2370 |
## TODO: test |
| 2371 |
} |
| 2372 |
splice @{$self->{open_elements}}, $i; |
| 2373 |
last LI; |
| 2374 |
} |
| 2375 |
|
| 2376 |
## Step 3 |
| 2377 |
if (not $formatting_category->{$node->[1]} and |
| 2378 |
#not $phrasing_category->{$node->[1]} and |
| 2379 |
($special_category->{$node->[1]} or |
| 2380 |
$scoping_category->{$node->[1]}) and |
| 2381 |
$node->[1] ne 'address' and $node->[1] ne 'div') { |
| 2382 |
last LI; |
| 2383 |
} |
| 2384 |
|
| 2385 |
## Step 4 |
| 2386 |
$i--; |
| 2387 |
$node = $self->{open_elements}->[$i]; |
| 2388 |
redo LI; |
| 2389 |
} # LI |
| 2390 |
|
| 2391 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2392 |
!!!next-token; |
| 2393 |
return; |
| 2394 |
} elsif ($token->{tag_name} eq 'dd' or $token->{tag_name} eq 'dt') { |
| 2395 |
## has a p element in scope |
| 2396 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2397 |
if ($_->[1] eq 'p') { |
| 2398 |
!!!back-token; |
| 2399 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2400 |
return; |
| 2401 |
} elsif ({ |
| 2402 |
table => 1, caption => 1, td => 1, th => 1, |
| 2403 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2404 |
}->{$_->[1]}) { |
| 2405 |
last INSCOPE; |
| 2406 |
} |
| 2407 |
} # INSCOPE |
| 2408 |
|
| 2409 |
## Step 1 |
| 2410 |
my $i = -1; |
| 2411 |
my $node = $self->{open_elements}->[$i]; |
| 2412 |
LI: { |
| 2413 |
## Step 2 |
| 2414 |
if ($node->[1] eq 'dt' or $node->[1] eq 'dd') { |
| 2415 |
if ($i != -1) { |
| 2416 |
!!!parse-error (type => 'end tag missing:'. |
| 2417 |
$self->{open_elements}->[-1]->[1]); |
| 2418 |
## TODO: test |
| 2419 |
} |
| 2420 |
splice @{$self->{open_elements}}, $i; |
| 2421 |
last LI; |
| 2422 |
} |
| 2423 |
|
| 2424 |
## Step 3 |
| 2425 |
if (not $formatting_category->{$node->[1]} and |
| 2426 |
#not $phrasing_category->{$node->[1]} and |
| 2427 |
($special_category->{$node->[1]} or |
| 2428 |
$scoping_category->{$node->[1]}) and |
| 2429 |
$node->[1] ne 'address' and $node->[1] ne 'div') { |
| 2430 |
last LI; |
| 2431 |
} |
| 2432 |
|
| 2433 |
## Step 4 |
| 2434 |
$i--; |
| 2435 |
$node = $self->{open_elements}->[$i]; |
| 2436 |
redo LI; |
| 2437 |
} # LI |
| 2438 |
|
| 2439 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2440 |
!!!next-token; |
| 2441 |
return; |
| 2442 |
} elsif ($token->{tag_name} eq 'plaintext') { |
| 2443 |
## has a p element in scope |
| 2444 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2445 |
if ($_->[1] eq 'p') { |
| 2446 |
!!!back-token; |
| 2447 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2448 |
return; |
| 2449 |
} elsif ({ |
| 2450 |
table => 1, caption => 1, td => 1, th => 1, |
| 2451 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2452 |
}->{$_->[1]}) { |
| 2453 |
last INSCOPE; |
| 2454 |
} |
| 2455 |
} # INSCOPE |
| 2456 |
|
| 2457 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2458 |
|
| 2459 |
$self->{content_model_flag} = 'PLAINTEXT'; |
| 2460 |
|
| 2461 |
!!!next-token; |
| 2462 |
return; |
| 2463 |
} elsif ({ |
| 2464 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
| 2465 |
}->{$token->{tag_name}}) { |
| 2466 |
## has a p element in scope |
| 2467 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2468 |
my $node = $self->{open_elements}->[$_]; |
| 2469 |
if ($node->[1] eq 'p') { |
| 2470 |
!!!back-token; |
| 2471 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2472 |
return; |
| 2473 |
} elsif ({ |
| 2474 |
table => 1, caption => 1, td => 1, th => 1, |
| 2475 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2476 |
}->{$node->[1]}) { |
| 2477 |
last INSCOPE; |
| 2478 |
} |
| 2479 |
} # INSCOPE |
| 2480 |
|
| 2481 |
## has an element in scope |
| 2482 |
my $i; |
| 2483 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2484 |
my $node = $self->{open_elements}->[$_]; |
| 2485 |
if ({ |
| 2486 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
| 2487 |
}->{$node->[1]}) { |
| 2488 |
$i = $_; |
| 2489 |
last INSCOPE; |
| 2490 |
} elsif ({ |
| 2491 |
table => 1, caption => 1, td => 1, th => 1, |
| 2492 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2493 |
}->{$node->[1]}) { |
| 2494 |
last INSCOPE; |
| 2495 |
} |
| 2496 |
} # INSCOPE |
| 2497 |
|
| 2498 |
if (defined $i) { |
| 2499 |
!!!parse-error (type => 'in hn:hn'); |
| 2500 |
splice @{$self->{open_elements}}, $i; |
| 2501 |
} |
| 2502 |
|
| 2503 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2504 |
|
| 2505 |
!!!next-token; |
| 2506 |
return; |
| 2507 |
} elsif ($token->{tag_name} eq 'a') { |
| 2508 |
AFE: for my $i (reverse 0..$#$active_formatting_elements) { |
| 2509 |
my $node = $active_formatting_elements->[$i]; |
| 2510 |
if ($node->[1] eq 'a') { |
| 2511 |
!!!parse-error (type => 'in a:a'); |
| 2512 |
|
| 2513 |
!!!back-token; |
| 2514 |
$token = {type => 'end tag', tag_name => 'a'}; |
| 2515 |
$formatting_end_tag->($token->{tag_name}); |
| 2516 |
|
| 2517 |
AFE2: for (reverse 0..$#$active_formatting_elements) { |
| 2518 |
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
| 2519 |
splice @$active_formatting_elements, $_, 1; |
| 2520 |
last AFE2; |
| 2521 |
} |
| 2522 |
} # AFE2 |
| 2523 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2524 |
if ($self->{open_elements}->[$_]->[0] eq $node->[0]) { |
| 2525 |
splice @{$self->{open_elements}}, $_, 1; |
| 2526 |
last OE; |
| 2527 |
} |
| 2528 |
} # OE |
| 2529 |
last AFE; |
| 2530 |
} elsif ($node->[0] eq '#marker') { |
| 2531 |
last AFE; |
| 2532 |
} |
| 2533 |
} # AFE |
| 2534 |
|
| 2535 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2536 |
|
| 2537 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2538 |
push @$active_formatting_elements, $self->{open_elements}->[-1]; |
| 2539 |
|
| 2540 |
!!!next-token; |
| 2541 |
return; |
| 2542 |
} elsif ({ |
| 2543 |
b => 1, big => 1, em => 1, font => 1, i => 1, |
| 2544 |
nobr => 1, s => 1, small => 1, strile => 1, |
| 2545 |
strong => 1, tt => 1, u => 1, |
| 2546 |
}->{$token->{tag_name}}) { |
| 2547 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2548 |
|
| 2549 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2550 |
push @$active_formatting_elements, $self->{open_elements}->[-1]; |
| 2551 |
|
| 2552 |
!!!next-token; |
| 2553 |
return; |
| 2554 |
} elsif ($token->{tag_name} eq 'button') { |
| 2555 |
## has a button element in scope |
| 2556 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2557 |
my $node = $self->{open_elements}->[$_]; |
| 2558 |
if ($node->[1] eq 'button') { |
| 2559 |
!!!parse-error (type => 'in button:button'); |
| 2560 |
!!!back-token; |
| 2561 |
$token = {type => 'end tag', tag_name => 'button'}; |
| 2562 |
return; |
| 2563 |
} elsif ({ |
| 2564 |
table => 1, caption => 1, td => 1, th => 1, |
| 2565 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2566 |
}->{$node->[1]}) { |
| 2567 |
last INSCOPE; |
| 2568 |
} |
| 2569 |
} # INSCOPE |
| 2570 |
|
| 2571 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2572 |
|
| 2573 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2574 |
push @$active_formatting_elements, ['#marker', '']; |
| 2575 |
|
| 2576 |
!!!next-token; |
| 2577 |
return; |
| 2578 |
} elsif ($token->{tag_name} eq 'marquee' or |
| 2579 |
$token->{tag_name} eq 'object') { |
| 2580 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2581 |
|
| 2582 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2583 |
push @$active_formatting_elements, ['#marker', '']; |
| 2584 |
|
| 2585 |
!!!next-token; |
| 2586 |
return; |
| 2587 |
} elsif ($token->{tag_name} eq 'xmp') { |
| 2588 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2589 |
|
| 2590 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2591 |
|
| 2592 |
$self->{content_model_flag} = 'CDATA'; |
| 2593 |
|
| 2594 |
!!!next-token; |
| 2595 |
return; |
| 2596 |
} elsif ($token->{tag_name} eq 'table') { |
| 2597 |
## has a p element in scope |
| 2598 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2599 |
if ($_->[1] eq 'p') { |
| 2600 |
!!!back-token; |
| 2601 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2602 |
return; |
| 2603 |
} elsif ({ |
| 2604 |
table => 1, caption => 1, td => 1, th => 1, |
| 2605 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2606 |
}->{$_->[1]}) { |
| 2607 |
last INSCOPE; |
| 2608 |
} |
| 2609 |
} # INSCOPE |
| 2610 |
|
| 2611 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2612 |
|
| 2613 |
$self->{insertion_mode} = 'in table'; |
| 2614 |
|
| 2615 |
!!!next-token; |
| 2616 |
return; |
| 2617 |
} elsif ({ |
| 2618 |
area => 1, basefont => 1, bgsound => 1, br => 1, |
| 2619 |
embed => 1, img => 1, param => 1, spacer => 1, wbr => 1, |
| 2620 |
image => 1, |
| 2621 |
}->{$token->{tag_name}}) { |
| 2622 |
if ($token->{tag_name} eq 'image') { |
| 2623 |
!!!parse-error (type => 'image'); |
| 2624 |
$token->{tag_name} = 'img'; |
| 2625 |
} |
| 2626 |
|
| 2627 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2628 |
|
| 2629 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2630 |
pop @{$self->{open_elements}}; |
| 2631 |
|
| 2632 |
!!!next-token; |
| 2633 |
return; |
| 2634 |
} elsif ($token->{tag_name} eq 'hr') { |
| 2635 |
## has a p element in scope |
| 2636 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 2637 |
if ($_->[1] eq 'p') { |
| 2638 |
!!!back-token; |
| 2639 |
$token = {type => 'end tag', tag_name => 'p'}; |
| 2640 |
return; |
| 2641 |
} elsif ({ |
| 2642 |
table => 1, caption => 1, td => 1, th => 1, |
| 2643 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2644 |
}->{$_->[1]}) { |
| 2645 |
last INSCOPE; |
| 2646 |
} |
| 2647 |
} # INSCOPE |
| 2648 |
|
| 2649 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2650 |
pop @{$self->{open_elements}}; |
| 2651 |
|
| 2652 |
!!!next-token; |
| 2653 |
return; |
| 2654 |
} elsif ($token->{tag_name} eq 'input') { |
| 2655 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2656 |
|
| 2657 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2658 |
## TODO: associate with $self->{form_element} if defined |
| 2659 |
pop @{$self->{open_elements}}; |
| 2660 |
|
| 2661 |
!!!next-token; |
| 2662 |
return; |
| 2663 |
} elsif ($token->{tag_name} eq 'isindex') { |
| 2664 |
!!!parse-error (type => 'isindex'); |
| 2665 |
|
| 2666 |
if (defined $self->{form_element}) { |
| 2667 |
## Ignore the token |
| 2668 |
!!!next-token; |
| 2669 |
return; |
| 2670 |
} else { |
| 2671 |
my $at = $token->{attributes}; |
| 2672 |
$at->{name} = {name => 'name', value => 'isindex'}; |
| 2673 |
my @tokens = ( |
| 2674 |
{type => 'start tag', tag_name => 'form'}, |
| 2675 |
{type => 'start tag', tag_name => 'hr'}, |
| 2676 |
{type => 'start tag', tag_name => 'p'}, |
| 2677 |
{type => 'start tag', tag_name => 'label'}, |
| 2678 |
{type => 'character', |
| 2679 |
data => 'This is a searchable index. Insert your search keywords here: '}, # SHOULD |
| 2680 |
## TODO: make this configurable |
| 2681 |
{type => 'start tag', tag_name => 'input', attributes => $at}, |
| 2682 |
#{type => 'character', data => ''}, # SHOULD |
| 2683 |
{type => 'end tag', tag_name => 'label'}, |
| 2684 |
{type => 'end tag', tag_name => 'p'}, |
| 2685 |
{type => 'start tag', tag_name => 'hr'}, |
| 2686 |
{type => 'end tag', tag_name => 'form'}, |
| 2687 |
); |
| 2688 |
$token = shift @tokens; |
| 2689 |
!!!back-token (@tokens); |
| 2690 |
return; |
| 2691 |
} |
| 2692 |
} elsif ({ |
| 2693 |
textarea => 1, |
| 2694 |
iframe => 1, |
| 2695 |
noembed => 1, |
| 2696 |
noframes => 1, |
| 2697 |
noscript => 0, ## TODO: 1 if scripting is enabled |
| 2698 |
}->{$token->{tag_name}}) { |
| 2699 |
my $tag_name = $token->{tag_name}; |
| 2700 |
my $el; |
| 2701 |
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
| 2702 |
|
| 2703 |
if ($token->{tag_name} eq 'textarea') { |
| 2704 |
## TODO: $self->{form_element} if defined |
| 2705 |
$self->{content_model_flag} = 'RCDATA'; |
| 2706 |
} else { |
| 2707 |
$self->{content_model_flag} = 'CDATA'; |
| 2708 |
} |
| 2709 |
|
| 2710 |
$insert->($el); |
| 2711 |
|
| 2712 |
my $text = ''; |
| 2713 |
if ($token->{tag_name} eq 'textarea') { |
| 2714 |
!!!next-token; |
| 2715 |
if ($token->{type} eq 'character') { |
| 2716 |
$token->{data} =~ s/^\x0A//; |
| 2717 |
unless (length $token->{data}) { |
| 2718 |
!!!next-token; |
| 2719 |
} |
| 2720 |
} |
| 2721 |
} else { |
| 2722 |
!!!next-token; |
| 2723 |
} |
| 2724 |
while ($token->{type} eq 'character') { |
| 2725 |
$text .= $token->{data}; |
| 2726 |
!!!next-token; |
| 2727 |
} |
| 2728 |
if (length $text) { |
| 2729 |
$el->manakai_append_text ($text); |
| 2730 |
} |
| 2731 |
|
| 2732 |
$self->{content_model_flag} = 'PCDATA'; |
| 2733 |
|
| 2734 |
if ($token->{type} eq 'end tag' and |
| 2735 |
$token->{tag_name} eq $tag_name) { |
| 2736 |
## Ignore the token |
| 2737 |
} else { |
| 2738 |
if ($token->{tag_name} eq 'textarea') { |
| 2739 |
!!!parse-error (type => 'in RCDATA:#'.$token->{type}); |
| 2740 |
} else { |
| 2741 |
!!!parse-error (type => 'in CDATA:#'.$token->{type}); |
| 2742 |
} |
| 2743 |
## ISSUE: And ignore? |
| 2744 |
} |
| 2745 |
!!!next-token; |
| 2746 |
return; |
| 2747 |
} elsif ($token->{tag_name} eq 'select') { |
| 2748 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2749 |
|
| 2750 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2751 |
|
| 2752 |
$self->{insertion_mode} = 'in select'; |
| 2753 |
!!!next-token; |
| 2754 |
return; |
| 2755 |
} elsif ({ |
| 2756 |
caption => 1, col => 1, colgroup => 1, frame => 1, |
| 2757 |
frameset => 1, head => 1, option => 1, optgroup => 1, |
| 2758 |
tbody => 1, td => 1, tfoot => 1, th => 1, |
| 2759 |
thead => 1, tr => 1, |
| 2760 |
}->{$token->{tag_name}}) { |
| 2761 |
!!!parse-error (type => 'in body:'.$token->{tag_name}); |
| 2762 |
## Ignore the token |
| 2763 |
!!!next-token; |
| 2764 |
return; |
| 2765 |
|
| 2766 |
## ISSUE: An issue on HTML5 new elements in the spec. |
| 2767 |
} else { |
| 2768 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2769 |
|
| 2770 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 2771 |
|
| 2772 |
!!!next-token; |
| 2773 |
return; |
| 2774 |
} |
| 2775 |
} elsif ($token->{type} eq 'end tag') { |
| 2776 |
if ($token->{tag_name} eq 'body') { |
| 2777 |
if (@{$self->{open_elements}} > 1 and $self->{open_elements}->[1]->[1] eq 'body') { |
| 2778 |
## ISSUE: There is an issue in the spec. |
| 2779 |
if ($self->{open_elements}->[-1]->[1] ne 'body') { |
| 2780 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2781 |
} |
| 2782 |
$self->{insertion_mode} = 'after body'; |
| 2783 |
!!!next-token; |
| 2784 |
return; |
| 2785 |
} else { |
| 2786 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2787 |
## Ignore the token |
| 2788 |
!!!next-token; |
| 2789 |
return; |
| 2790 |
} |
| 2791 |
} elsif ($token->{tag_name} eq 'html') { |
| 2792 |
if (@{$self->{open_elements}} > 1 and $self->{open_elements}->[1]->[1] eq 'body') { |
| 2793 |
## ISSUE: There is an issue in the spec. |
| 2794 |
if ($self->{open_elements}->[-1]->[1] ne 'body') { |
| 2795 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[1]->[1]); |
| 2796 |
} |
| 2797 |
$self->{insertion_mode} = 'after body'; |
| 2798 |
## reprocess |
| 2799 |
return; |
| 2800 |
} else { |
| 2801 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2802 |
## Ignore the token |
| 2803 |
!!!next-token; |
| 2804 |
return; |
| 2805 |
} |
| 2806 |
} elsif ({ |
| 2807 |
address => 1, blockquote => 1, center => 1, dir => 1, |
| 2808 |
div => 1, dl => 1, fieldset => 1, listing => 1, |
| 2809 |
menu => 1, ol => 1, pre => 1, ul => 1, |
| 2810 |
form => 1, |
| 2811 |
p => 1, |
| 2812 |
dd => 1, dt => 1, li => 1, |
| 2813 |
button => 1, marquee => 1, object => 1, |
| 2814 |
}->{$token->{tag_name}}) { |
| 2815 |
## has an element in scope |
| 2816 |
my $i; |
| 2817 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2818 |
my $node = $self->{open_elements}->[$_]; |
| 2819 |
if ($node->[1] eq $token->{tag_name}) { |
| 2820 |
## generate implied end tags |
| 2821 |
if ({ |
| 2822 |
dd => ($token->{tag_name} ne 'dd'), |
| 2823 |
dt => ($token->{tag_name} ne 'dt'), |
| 2824 |
li => ($token->{tag_name} ne 'li'), |
| 2825 |
p => ($token->{tag_name} ne 'p'), |
| 2826 |
td => 1, th => 1, tr => 1, |
| 2827 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 2828 |
!!!back-token; |
| 2829 |
$token = {type => 'end tag', |
| 2830 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 2831 |
return; |
| 2832 |
} |
| 2833 |
$i = $_; |
| 2834 |
last INSCOPE unless $token->{tag_name} eq 'p'; |
| 2835 |
} elsif ({ |
| 2836 |
table => 1, caption => 1, td => 1, th => 1, |
| 2837 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2838 |
}->{$node->[1]}) { |
| 2839 |
last INSCOPE; |
| 2840 |
} |
| 2841 |
} # INSCOPE |
| 2842 |
|
| 2843 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
| 2844 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2845 |
} |
| 2846 |
|
| 2847 |
splice @{$self->{open_elements}}, $i if defined $i; |
| 2848 |
undef $self->{form_element} if $token->{tag_name} eq 'form'; |
| 2849 |
$clear_up_to_marker->() |
| 2850 |
if { |
| 2851 |
button => 1, marquee => 1, object => 1, |
| 2852 |
}->{$token->{tag_name}}; |
| 2853 |
!!!next-token; |
| 2854 |
return; |
| 2855 |
} elsif ({ |
| 2856 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
| 2857 |
}->{$token->{tag_name}}) { |
| 2858 |
## has an element in scope |
| 2859 |
my $i; |
| 2860 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2861 |
my $node = $self->{open_elements}->[$_]; |
| 2862 |
if ({ |
| 2863 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
| 2864 |
}->{$node->[1]}) { |
| 2865 |
## generate implied end tags |
| 2866 |
if ({ |
| 2867 |
dd => 1, dt => 1, li => 1, p => 1, |
| 2868 |
td => 1, th => 1, tr => 1, |
| 2869 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 2870 |
!!!back-token; |
| 2871 |
$token = {type => 'end tag', |
| 2872 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 2873 |
return; |
| 2874 |
} |
| 2875 |
$i = $_; |
| 2876 |
last INSCOPE; |
| 2877 |
} elsif ({ |
| 2878 |
table => 1, caption => 1, td => 1, th => 1, |
| 2879 |
button => 1, marquee => 1, object => 1, html => 1, |
| 2880 |
}->{$node->[1]}) { |
| 2881 |
last INSCOPE; |
| 2882 |
} |
| 2883 |
} # INSCOPE |
| 2884 |
|
| 2885 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
| 2886 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2887 |
} |
| 2888 |
|
| 2889 |
splice @{$self->{open_elements}}, $i if defined $i; |
| 2890 |
!!!next-token; |
| 2891 |
return; |
| 2892 |
} elsif ({ |
| 2893 |
a => 1, |
| 2894 |
b => 1, big => 1, em => 1, font => 1, i => 1, |
| 2895 |
nobr => 1, s => 1, small => 1, strile => 1, |
| 2896 |
strong => 1, tt => 1, u => 1, |
| 2897 |
}->{$token->{tag_name}}) { |
| 2898 |
$formatting_end_tag->($token->{tag_name}); |
| 2899 |
## TODO: <http://html5.org/tools/web-apps-tracker?from=883&to=884> |
| 2900 |
return; |
| 2901 |
} elsif ({ |
| 2902 |
caption => 1, col => 1, colgroup => 1, frame => 1, |
| 2903 |
frameset => 1, head => 1, option => 1, optgroup => 1, |
| 2904 |
tbody => 1, td => 1, tfoot => 1, th => 1, |
| 2905 |
thead => 1, tr => 1, |
| 2906 |
area => 1, basefont => 1, bgsound => 1, br => 1, |
| 2907 |
embed => 1, hr => 1, iframe => 1, image => 1, |
| 2908 |
img => 1, input => 1, isindex => 1, noembed => 1, |
| 2909 |
noframes => 1, param => 1, select => 1, spacer => 1, |
| 2910 |
table => 1, textarea => 1, wbr => 1, |
| 2911 |
noscript => 0, ## TODO: if scripting is enabled |
| 2912 |
}->{$token->{tag_name}}) { |
| 2913 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2914 |
## Ignore the token |
| 2915 |
!!!next-token; |
| 2916 |
return; |
| 2917 |
|
| 2918 |
## ISSUE: Issue on HTML5 new elements in spec |
| 2919 |
|
| 2920 |
} else { |
| 2921 |
## Step 1 |
| 2922 |
my $node_i = -1; |
| 2923 |
my $node = $self->{open_elements}->[$node_i]; |
| 2924 |
|
| 2925 |
## Step 2 |
| 2926 |
S2: { |
| 2927 |
if ($node->[1] eq $token->{tag_name}) { |
| 2928 |
## Step 1 |
| 2929 |
## generate implied end tags |
| 2930 |
if ({ |
| 2931 |
dd => 1, dt => 1, li => 1, p => 1, |
| 2932 |
td => 1, th => 1, tr => 1, |
| 2933 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 2934 |
!!!back-token; |
| 2935 |
$token = {type => 'end tag', |
| 2936 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 2937 |
return; |
| 2938 |
} |
| 2939 |
|
| 2940 |
## Step 2 |
| 2941 |
if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { |
| 2942 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2943 |
} |
| 2944 |
|
| 2945 |
## Step 3 |
| 2946 |
splice @{$self->{open_elements}}, $node_i; |
| 2947 |
|
| 2948 |
!!!next-token; |
| 2949 |
last S2; |
| 2950 |
} else { |
| 2951 |
## Step 3 |
| 2952 |
if (not $formatting_category->{$node->[1]} and |
| 2953 |
#not $phrasing_category->{$node->[1]} and |
| 2954 |
($special_category->{$node->[1]} or |
| 2955 |
$scoping_category->{$node->[1]})) { |
| 2956 |
!!!parse-error (type => 'not closed:'.$node->[1]); |
| 2957 |
## Ignore the token |
| 2958 |
!!!next-token; |
| 2959 |
last S2; |
| 2960 |
} |
| 2961 |
} |
| 2962 |
|
| 2963 |
## Step 4 |
| 2964 |
$node_i--; |
| 2965 |
$node = $self->{open_elements}->[$node_i]; |
| 2966 |
|
| 2967 |
## Step 5; |
| 2968 |
redo S2; |
| 2969 |
} # S2 |
| 2970 |
return; |
| 2971 |
} |
| 2972 |
} |
| 2973 |
}; # $in_body |
| 2974 |
|
| 2975 |
B: { |
| 2976 |
if ($phase eq 'main') { |
| 2977 |
if ($token->{type} eq 'DOCTYPE') { |
| 2978 |
!!!parse-error (type => 'in html:#DOCTYPE'); |
| 2979 |
## Ignore the token |
| 2980 |
## Stay in the phase |
| 2981 |
!!!next-token; |
| 2982 |
redo B; |
| 2983 |
} elsif ($token->{type} eq 'start tag' and |
| 2984 |
$token->{tag_name} eq 'html') { |
| 2985 |
## TODO: unless it is the first start tag token, parse-error |
| 2986 |
my $top_el = $self->{open_elements}->[0]->[0]; |
| 2987 |
for my $attr_name (keys %{$token->{attributes}}) { |
| 2988 |
unless ($top_el->has_attribute_ns (undef, $attr_name)) { |
| 2989 |
$top_el->set_attribute_ns |
| 2990 |
(undef, [undef, $attr_name], |
| 2991 |
$token->{attributes}->{$attr_name}->{value}); |
| 2992 |
} |
| 2993 |
} |
| 2994 |
!!!next-token; |
| 2995 |
redo B; |
| 2996 |
} elsif ($token->{type} eq 'end-of-file') { |
| 2997 |
## Generate implied end tags |
| 2998 |
if ({ |
| 2999 |
dd => 1, dt => 1, li => 1, p => 1, td => 1, th => 1, tr => 1, |
| 3000 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3001 |
!!!back-token; |
| 3002 |
$token = {type => 'end tag', tag_name => $self->{open_elements}->[-1]->[1]}; |
| 3003 |
redo B; |
| 3004 |
} |
| 3005 |
|
| 3006 |
if (@{$self->{open_elements}} > 2 or |
| 3007 |
(@{$self->{open_elements}} == 2 and $self->{open_elements}->[1]->[1] ne 'body')) { |
| 3008 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3009 |
} elsif (defined $self->{inner_html_node} and |
| 3010 |
@{$self->{open_elements}} > 1 and |
| 3011 |
$self->{open_elements}->[1]->[1] ne 'body') { |
| 3012 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3013 |
} |
| 3014 |
|
| 3015 |
## Stop parsing |
| 3016 |
last B; |
| 3017 |
|
| 3018 |
## ISSUE: There is an issue in the spec. |
| 3019 |
} else { |
| 3020 |
if ($self->{insertion_mode} eq 'before head') { |
| 3021 |
if ($token->{type} eq 'character') { |
| 3022 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3023 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3024 |
unless (length $token->{data}) { |
| 3025 |
!!!next-token; |
| 3026 |
redo B; |
| 3027 |
} |
| 3028 |
} |
| 3029 |
## As if <head> |
| 3030 |
!!!create-element ($self->{head_element}, 'head'); |
| 3031 |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 3032 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 3033 |
$self->{insertion_mode} = 'in head'; |
| 3034 |
## reprocess |
| 3035 |
redo B; |
| 3036 |
} elsif ($token->{type} eq 'comment') { |
| 3037 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3038 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3039 |
!!!next-token; |
| 3040 |
redo B; |
| 3041 |
} elsif ($token->{type} eq 'start tag') { |
| 3042 |
my $attr = $token->{tag_name} eq 'head' ? $token->{attributes} : {}; |
| 3043 |
!!!create-element ($self->{head_element}, 'head', $attr); |
| 3044 |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 3045 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 3046 |
$self->{insertion_mode} = 'in head'; |
| 3047 |
if ($token->{tag_name} eq 'head') { |
| 3048 |
!!!next-token; |
| 3049 |
#} elsif ({ |
| 3050 |
# base => 1, link => 1, meta => 1, |
| 3051 |
# script => 1, style => 1, title => 1, |
| 3052 |
# }->{$token->{tag_name}}) { |
| 3053 |
# ## reprocess |
| 3054 |
} else { |
| 3055 |
## reprocess |
| 3056 |
} |
| 3057 |
redo B; |
| 3058 |
} elsif ($token->{type} eq 'end tag') { |
| 3059 |
if ($token->{tag_name} eq 'html') { |
| 3060 |
## As if <head> |
| 3061 |
!!!create-element ($self->{head_element}, 'head'); |
| 3062 |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 3063 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 3064 |
$self->{insertion_mode} = 'in head'; |
| 3065 |
## reprocess |
| 3066 |
redo B; |
| 3067 |
} else { |
| 3068 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3069 |
## Ignore the token |
| 3070 |
!!!next-token; |
| 3071 |
redo B; |
| 3072 |
} |
| 3073 |
} else { |
| 3074 |
die "$0: $token->{type}: Unknown type"; |
| 3075 |
} |
| 3076 |
} elsif ($self->{insertion_mode} eq 'in head') { |
| 3077 |
if ($token->{type} eq 'character') { |
| 3078 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3079 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3080 |
unless (length $token->{data}) { |
| 3081 |
!!!next-token; |
| 3082 |
redo B; |
| 3083 |
} |
| 3084 |
} |
| 3085 |
|
| 3086 |
# |
| 3087 |
} elsif ($token->{type} eq 'comment') { |
| 3088 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3089 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3090 |
!!!next-token; |
| 3091 |
redo B; |
| 3092 |
} elsif ($token->{type} eq 'start tag') { |
| 3093 |
if ($token->{tag_name} eq 'title') { |
| 3094 |
## NOTE: There is an "as if in head" code clone |
| 3095 |
my $title_el; |
| 3096 |
!!!create-element ($title_el, 'title', $token->{attributes}); |
| 3097 |
(defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) |
| 3098 |
->append_child ($title_el); |
| 3099 |
$self->{content_model_flag} = 'RCDATA'; |
| 3100 |
|
| 3101 |
my $text = ''; |
| 3102 |
!!!next-token; |
| 3103 |
while ($token->{type} eq 'character') { |
| 3104 |
$text .= $token->{data}; |
| 3105 |
!!!next-token; |
| 3106 |
} |
| 3107 |
if (length $text) { |
| 3108 |
$title_el->manakai_append_text ($text); |
| 3109 |
} |
| 3110 |
|
| 3111 |
$self->{content_model_flag} = 'PCDATA'; |
| 3112 |
|
| 3113 |
if ($token->{type} eq 'end tag' and |
| 3114 |
$token->{tag_name} eq 'title') { |
| 3115 |
## Ignore the token |
| 3116 |
} else { |
| 3117 |
!!!parse-error (type => 'in RCDATA:#'.$token->{type}); |
| 3118 |
## ISSUE: And ignore? |
| 3119 |
} |
| 3120 |
!!!next-token; |
| 3121 |
redo B; |
| 3122 |
} elsif ($token->{tag_name} eq 'style') { |
| 3123 |
$style_start_tag->(); |
| 3124 |
redo B; |
| 3125 |
} elsif ($token->{tag_name} eq 'script') { |
| 3126 |
$script_start_tag->(); |
| 3127 |
redo B; |
| 3128 |
} elsif ({base => 1, link => 1, meta => 1}->{$token->{tag_name}}) { |
| 3129 |
## NOTE: There are "as if in head" code clones |
| 3130 |
my $el; |
| 3131 |
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
| 3132 |
(defined $self->{head_element} ? $self->{head_element} : $self->{open_elements}->[-1]->[0]) |
| 3133 |
->append_child ($el); |
| 3134 |
|
| 3135 |
!!!next-token; |
| 3136 |
redo B; |
| 3137 |
} elsif ($token->{tag_name} eq 'head') { |
| 3138 |
!!!parse-error (type => 'in head:head'); |
| 3139 |
## Ignore the token |
| 3140 |
!!!next-token; |
| 3141 |
redo B; |
| 3142 |
} else { |
| 3143 |
# |
| 3144 |
} |
| 3145 |
} elsif ($token->{type} eq 'end tag') { |
| 3146 |
if ($token->{tag_name} eq 'head') { |
| 3147 |
if ($self->{open_elements}->[-1]->[1] eq 'head') { |
| 3148 |
pop @{$self->{open_elements}}; |
| 3149 |
} else { |
| 3150 |
!!!parse-error (type => 'unmatched end tag:head'); |
| 3151 |
} |
| 3152 |
$self->{insertion_mode} = 'after head'; |
| 3153 |
!!!next-token; |
| 3154 |
redo B; |
| 3155 |
} elsif ($token->{tag_name} eq 'html') { |
| 3156 |
# |
| 3157 |
} else { |
| 3158 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3159 |
## Ignore the token |
| 3160 |
!!!next-token; |
| 3161 |
redo B; |
| 3162 |
} |
| 3163 |
} else { |
| 3164 |
# |
| 3165 |
} |
| 3166 |
|
| 3167 |
if ($self->{open_elements}->[-1]->[1] eq 'head') { |
| 3168 |
## As if </head> |
| 3169 |
pop @{$self->{open_elements}}; |
| 3170 |
} |
| 3171 |
$self->{insertion_mode} = 'after head'; |
| 3172 |
## reprocess |
| 3173 |
redo B; |
| 3174 |
|
| 3175 |
## ISSUE: An issue in the spec. |
| 3176 |
} elsif ($self->{insertion_mode} eq 'after head') { |
| 3177 |
if ($token->{type} eq 'character') { |
| 3178 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3179 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3180 |
unless (length $token->{data}) { |
| 3181 |
!!!next-token; |
| 3182 |
redo B; |
| 3183 |
} |
| 3184 |
} |
| 3185 |
|
| 3186 |
# |
| 3187 |
} elsif ($token->{type} eq 'comment') { |
| 3188 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3189 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3190 |
!!!next-token; |
| 3191 |
redo B; |
| 3192 |
} elsif ($token->{type} eq 'start tag') { |
| 3193 |
if ($token->{tag_name} eq 'body') { |
| 3194 |
!!!insert-element ('body', $token->{attributes}); |
| 3195 |
$self->{insertion_mode} = 'in body'; |
| 3196 |
!!!next-token; |
| 3197 |
redo B; |
| 3198 |
} elsif ($token->{tag_name} eq 'frameset') { |
| 3199 |
!!!insert-element ('frameset', $token->{attributes}); |
| 3200 |
$self->{insertion_mode} = 'in frameset'; |
| 3201 |
!!!next-token; |
| 3202 |
redo B; |
| 3203 |
} elsif ({ |
| 3204 |
base => 1, link => 1, meta => 1, |
| 3205 |
script => 1, style => 1, title => 1, |
| 3206 |
}->{$token->{tag_name}}) { |
| 3207 |
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 3208 |
$self->{insertion_mode} = 'in head'; |
| 3209 |
## reprocess |
| 3210 |
redo B; |
| 3211 |
} else { |
| 3212 |
# |
| 3213 |
} |
| 3214 |
} else { |
| 3215 |
# |
| 3216 |
} |
| 3217 |
|
| 3218 |
## As if <body> |
| 3219 |
!!!insert-element ('body'); |
| 3220 |
$self->{insertion_mode} = 'in body'; |
| 3221 |
## reprocess |
| 3222 |
redo B; |
| 3223 |
} elsif ($self->{insertion_mode} eq 'in body') { |
| 3224 |
if ($token->{type} eq 'character') { |
| 3225 |
## NOTE: There is a code clone of "character in body". |
| 3226 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 3227 |
|
| 3228 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3229 |
|
| 3230 |
!!!next-token; |
| 3231 |
redo B; |
| 3232 |
} elsif ($token->{type} eq 'comment') { |
| 3233 |
## NOTE: There is a code clone of "comment in body". |
| 3234 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3235 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3236 |
!!!next-token; |
| 3237 |
redo B; |
| 3238 |
} else { |
| 3239 |
$in_body->($insert_to_current); |
| 3240 |
redo B; |
| 3241 |
} |
| 3242 |
} elsif ($self->{insertion_mode} eq 'in table') { |
| 3243 |
if ($token->{type} eq 'character') { |
| 3244 |
## NOTE: There are "character in table" code clones. |
| 3245 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3246 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3247 |
|
| 3248 |
unless (length $token->{data}) { |
| 3249 |
!!!next-token; |
| 3250 |
redo B; |
| 3251 |
} |
| 3252 |
} |
| 3253 |
|
| 3254 |
!!!parse-error (type => 'in table:#character'); |
| 3255 |
|
| 3256 |
## As if in body, but insert into foster parent element |
| 3257 |
## ISSUE: Spec says that "whenever a node would be inserted |
| 3258 |
## into the current node" while characters might not be |
| 3259 |
## result in a new Text node. |
| 3260 |
$reconstruct_active_formatting_elements->($insert_to_foster); |
| 3261 |
|
| 3262 |
if ({ |
| 3263 |
table => 1, tbody => 1, tfoot => 1, |
| 3264 |
thead => 1, tr => 1, |
| 3265 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3266 |
# MUST |
| 3267 |
my $foster_parent_element; |
| 3268 |
my $next_sibling; |
| 3269 |
my $prev_sibling; |
| 3270 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3271 |
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
| 3272 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 3273 |
if (defined $parent and $parent->node_type == 1) { |
| 3274 |
$foster_parent_element = $parent; |
| 3275 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 3276 |
$prev_sibling = $next_sibling->previous_sibling; |
| 3277 |
} else { |
| 3278 |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 3279 |
$prev_sibling = $foster_parent_element->last_child; |
| 3280 |
} |
| 3281 |
last OE; |
| 3282 |
} |
| 3283 |
} # OE |
| 3284 |
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
| 3285 |
$prev_sibling = $foster_parent_element->last_child |
| 3286 |
unless defined $foster_parent_element; |
| 3287 |
if (defined $prev_sibling and |
| 3288 |
$prev_sibling->node_type == 3) { |
| 3289 |
$prev_sibling->manakai_append_text ($token->{data}); |
| 3290 |
} else { |
| 3291 |
$foster_parent_element->insert_before |
| 3292 |
($self->{document}->create_text_node ($token->{data}), |
| 3293 |
$next_sibling); |
| 3294 |
} |
| 3295 |
} else { |
| 3296 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3297 |
} |
| 3298 |
|
| 3299 |
!!!next-token; |
| 3300 |
redo B; |
| 3301 |
} elsif ($token->{type} eq 'comment') { |
| 3302 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3303 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3304 |
!!!next-token; |
| 3305 |
redo B; |
| 3306 |
} elsif ($token->{type} eq 'start tag') { |
| 3307 |
if ({ |
| 3308 |
caption => 1, |
| 3309 |
colgroup => 1, |
| 3310 |
tbody => 1, tfoot => 1, thead => 1, |
| 3311 |
}->{$token->{tag_name}}) { |
| 3312 |
## Clear back to table context |
| 3313 |
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
| 3314 |
$self->{open_elements}->[-1]->[1] ne 'html') { |
| 3315 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3316 |
pop @{$self->{open_elements}}; |
| 3317 |
} |
| 3318 |
|
| 3319 |
push @$active_formatting_elements, ['#marker', ''] |
| 3320 |
if $token->{tag_name} eq 'caption'; |
| 3321 |
|
| 3322 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3323 |
$self->{insertion_mode} = { |
| 3324 |
caption => 'in caption', |
| 3325 |
colgroup => 'in column group', |
| 3326 |
tbody => 'in table body', |
| 3327 |
tfoot => 'in table body', |
| 3328 |
thead => 'in table body', |
| 3329 |
}->{$token->{tag_name}}; |
| 3330 |
!!!next-token; |
| 3331 |
redo B; |
| 3332 |
} elsif ({ |
| 3333 |
col => 1, |
| 3334 |
td => 1, th => 1, tr => 1, |
| 3335 |
}->{$token->{tag_name}}) { |
| 3336 |
## Clear back to table context |
| 3337 |
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
| 3338 |
$self->{open_elements}->[-1]->[1] ne 'html') { |
| 3339 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3340 |
pop @{$self->{open_elements}}; |
| 3341 |
} |
| 3342 |
|
| 3343 |
!!!insert-element ($token->{tag_name} eq 'col' ? 'colgroup' : 'tbody'); |
| 3344 |
$self->{insertion_mode} = $token->{tag_name} eq 'col' |
| 3345 |
? 'in column group' : 'in table body'; |
| 3346 |
## reprocess |
| 3347 |
redo B; |
| 3348 |
} elsif ($token->{tag_name} eq 'table') { |
| 3349 |
## NOTE: There are code clones for this "table in table" |
| 3350 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3351 |
|
| 3352 |
## As if </table> |
| 3353 |
## have a table element in table scope |
| 3354 |
my $i; |
| 3355 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3356 |
my $node = $self->{open_elements}->[$_]; |
| 3357 |
if ($node->[1] eq 'table') { |
| 3358 |
$i = $_; |
| 3359 |
last INSCOPE; |
| 3360 |
} elsif ({ |
| 3361 |
table => 1, html => 1, |
| 3362 |
}->{$node->[1]}) { |
| 3363 |
last INSCOPE; |
| 3364 |
} |
| 3365 |
} # INSCOPE |
| 3366 |
unless (defined $i) { |
| 3367 |
!!!parse-error (type => 'unmatched end tag:table'); |
| 3368 |
## Ignore tokens </table><table> |
| 3369 |
!!!next-token; |
| 3370 |
redo B; |
| 3371 |
} |
| 3372 |
|
| 3373 |
## generate implied end tags |
| 3374 |
if ({ |
| 3375 |
dd => 1, dt => 1, li => 1, p => 1, |
| 3376 |
td => 1, th => 1, tr => 1, |
| 3377 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3378 |
!!!back-token; # <table> |
| 3379 |
$token = {type => 'end tag', tag_name => 'table'}; |
| 3380 |
!!!back-token; |
| 3381 |
$token = {type => 'end tag', |
| 3382 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3383 |
redo B; |
| 3384 |
} |
| 3385 |
|
| 3386 |
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
| 3387 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3388 |
} |
| 3389 |
|
| 3390 |
splice @{$self->{open_elements}}, $i; |
| 3391 |
|
| 3392 |
$self->_reset_insertion_mode; |
| 3393 |
|
| 3394 |
## reprocess |
| 3395 |
redo B; |
| 3396 |
} else { |
| 3397 |
# |
| 3398 |
} |
| 3399 |
} elsif ($token->{type} eq 'end tag') { |
| 3400 |
if ($token->{tag_name} eq 'table') { |
| 3401 |
## have a table element in table scope |
| 3402 |
my $i; |
| 3403 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3404 |
my $node = $self->{open_elements}->[$_]; |
| 3405 |
if ($node->[1] eq $token->{tag_name}) { |
| 3406 |
$i = $_; |
| 3407 |
last INSCOPE; |
| 3408 |
} elsif ({ |
| 3409 |
table => 1, html => 1, |
| 3410 |
}->{$node->[1]}) { |
| 3411 |
last INSCOPE; |
| 3412 |
} |
| 3413 |
} # INSCOPE |
| 3414 |
unless (defined $i) { |
| 3415 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3416 |
## Ignore the token |
| 3417 |
!!!next-token; |
| 3418 |
redo B; |
| 3419 |
} |
| 3420 |
|
| 3421 |
## generate implied end tags |
| 3422 |
if ({ |
| 3423 |
dd => 1, dt => 1, li => 1, p => 1, |
| 3424 |
td => 1, th => 1, tr => 1, |
| 3425 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3426 |
!!!back-token; |
| 3427 |
$token = {type => 'end tag', |
| 3428 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3429 |
redo B; |
| 3430 |
} |
| 3431 |
|
| 3432 |
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
| 3433 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3434 |
} |
| 3435 |
|
| 3436 |
splice @{$self->{open_elements}}, $i; |
| 3437 |
|
| 3438 |
$self->_reset_insertion_mode; |
| 3439 |
|
| 3440 |
!!!next-token; |
| 3441 |
redo B; |
| 3442 |
} elsif ({ |
| 3443 |
body => 1, caption => 1, col => 1, colgroup => 1, |
| 3444 |
html => 1, tbody => 1, td => 1, tfoot => 1, th => 1, |
| 3445 |
thead => 1, tr => 1, |
| 3446 |
}->{$token->{tag_name}}) { |
| 3447 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3448 |
## Ignore the token |
| 3449 |
!!!next-token; |
| 3450 |
redo B; |
| 3451 |
} else { |
| 3452 |
# |
| 3453 |
} |
| 3454 |
} else { |
| 3455 |
# |
| 3456 |
} |
| 3457 |
|
| 3458 |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
| 3459 |
$in_body->($insert_to_foster); |
| 3460 |
redo B; |
| 3461 |
} elsif ($self->{insertion_mode} eq 'in caption') { |
| 3462 |
if ($token->{type} eq 'character') { |
| 3463 |
## NOTE: This is a code clone of "character in body". |
| 3464 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 3465 |
|
| 3466 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3467 |
|
| 3468 |
!!!next-token; |
| 3469 |
redo B; |
| 3470 |
} elsif ($token->{type} eq 'comment') { |
| 3471 |
## NOTE: This is a code clone of "comment in body". |
| 3472 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3473 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3474 |
!!!next-token; |
| 3475 |
redo B; |
| 3476 |
} elsif ($token->{type} eq 'start tag') { |
| 3477 |
if ({ |
| 3478 |
caption => 1, col => 1, colgroup => 1, tbody => 1, |
| 3479 |
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
| 3480 |
}->{$token->{tag_name}}) { |
| 3481 |
!!!parse-error (type => 'not closed:caption'); |
| 3482 |
|
| 3483 |
## As if </caption> |
| 3484 |
## have a table element in table scope |
| 3485 |
my $i; |
| 3486 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3487 |
my $node = $self->{open_elements}->[$_]; |
| 3488 |
if ($node->[1] eq 'caption') { |
| 3489 |
$i = $_; |
| 3490 |
last INSCOPE; |
| 3491 |
} elsif ({ |
| 3492 |
table => 1, html => 1, |
| 3493 |
}->{$node->[1]}) { |
| 3494 |
last INSCOPE; |
| 3495 |
} |
| 3496 |
} # INSCOPE |
| 3497 |
unless (defined $i) { |
| 3498 |
!!!parse-error (type => 'unmatched end tag:caption'); |
| 3499 |
## Ignore the token |
| 3500 |
!!!next-token; |
| 3501 |
redo B; |
| 3502 |
} |
| 3503 |
|
| 3504 |
## generate implied end tags |
| 3505 |
if ({ |
| 3506 |
dd => 1, dt => 1, li => 1, p => 1, |
| 3507 |
td => 1, th => 1, tr => 1, |
| 3508 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3509 |
!!!back-token; # <?> |
| 3510 |
$token = {type => 'end tag', tag_name => 'caption'}; |
| 3511 |
!!!back-token; |
| 3512 |
$token = {type => 'end tag', |
| 3513 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3514 |
redo B; |
| 3515 |
} |
| 3516 |
|
| 3517 |
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
| 3518 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3519 |
} |
| 3520 |
|
| 3521 |
splice @{$self->{open_elements}}, $i; |
| 3522 |
|
| 3523 |
$clear_up_to_marker->(); |
| 3524 |
|
| 3525 |
$self->{insertion_mode} = 'in table'; |
| 3526 |
|
| 3527 |
## reprocess |
| 3528 |
redo B; |
| 3529 |
} else { |
| 3530 |
# |
| 3531 |
} |
| 3532 |
} elsif ($token->{type} eq 'end tag') { |
| 3533 |
if ($token->{tag_name} eq 'caption') { |
| 3534 |
## have a table element in table scope |
| 3535 |
my $i; |
| 3536 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3537 |
my $node = $self->{open_elements}->[$_]; |
| 3538 |
if ($node->[1] eq $token->{tag_name}) { |
| 3539 |
$i = $_; |
| 3540 |
last INSCOPE; |
| 3541 |
} elsif ({ |
| 3542 |
table => 1, html => 1, |
| 3543 |
}->{$node->[1]}) { |
| 3544 |
last INSCOPE; |
| 3545 |
} |
| 3546 |
} # INSCOPE |
| 3547 |
unless (defined $i) { |
| 3548 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3549 |
## Ignore the token |
| 3550 |
!!!next-token; |
| 3551 |
redo B; |
| 3552 |
} |
| 3553 |
|
| 3554 |
## generate implied end tags |
| 3555 |
if ({ |
| 3556 |
dd => 1, dt => 1, li => 1, p => 1, |
| 3557 |
td => 1, th => 1, tr => 1, |
| 3558 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3559 |
!!!back-token; |
| 3560 |
$token = {type => 'end tag', |
| 3561 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3562 |
redo B; |
| 3563 |
} |
| 3564 |
|
| 3565 |
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
| 3566 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3567 |
} |
| 3568 |
|
| 3569 |
splice @{$self->{open_elements}}, $i; |
| 3570 |
|
| 3571 |
$clear_up_to_marker->(); |
| 3572 |
|
| 3573 |
$self->{insertion_mode} = 'in table'; |
| 3574 |
|
| 3575 |
!!!next-token; |
| 3576 |
redo B; |
| 3577 |
} elsif ($token->{tag_name} eq 'table') { |
| 3578 |
!!!parse-error (type => 'not closed:caption'); |
| 3579 |
|
| 3580 |
## As if </caption> |
| 3581 |
## have a table element in table scope |
| 3582 |
my $i; |
| 3583 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3584 |
my $node = $self->{open_elements}->[$_]; |
| 3585 |
if ($node->[1] eq 'caption') { |
| 3586 |
$i = $_; |
| 3587 |
last INSCOPE; |
| 3588 |
} elsif ({ |
| 3589 |
table => 1, html => 1, |
| 3590 |
}->{$node->[1]}) { |
| 3591 |
last INSCOPE; |
| 3592 |
} |
| 3593 |
} # INSCOPE |
| 3594 |
unless (defined $i) { |
| 3595 |
!!!parse-error (type => 'unmatched end tag:caption'); |
| 3596 |
## Ignore the token |
| 3597 |
!!!next-token; |
| 3598 |
redo B; |
| 3599 |
} |
| 3600 |
|
| 3601 |
## generate implied end tags |
| 3602 |
if ({ |
| 3603 |
dd => 1, dt => 1, li => 1, p => 1, |
| 3604 |
td => 1, th => 1, tr => 1, |
| 3605 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3606 |
!!!back-token; # </table> |
| 3607 |
$token = {type => 'end tag', tag_name => 'caption'}; |
| 3608 |
!!!back-token; |
| 3609 |
$token = {type => 'end tag', |
| 3610 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3611 |
redo B; |
| 3612 |
} |
| 3613 |
|
| 3614 |
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
| 3615 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3616 |
} |
| 3617 |
|
| 3618 |
splice @{$self->{open_elements}}, $i; |
| 3619 |
|
| 3620 |
$clear_up_to_marker->(); |
| 3621 |
|
| 3622 |
$self->{insertion_mode} = 'in table'; |
| 3623 |
|
| 3624 |
## reprocess |
| 3625 |
redo B; |
| 3626 |
} elsif ({ |
| 3627 |
body => 1, col => 1, colgroup => 1, |
| 3628 |
html => 1, tbody => 1, td => 1, tfoot => 1, |
| 3629 |
th => 1, thead => 1, tr => 1, |
| 3630 |
}->{$token->{tag_name}}) { |
| 3631 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3632 |
## Ignore the token |
| 3633 |
redo B; |
| 3634 |
} else { |
| 3635 |
# |
| 3636 |
} |
| 3637 |
} else { |
| 3638 |
# |
| 3639 |
} |
| 3640 |
|
| 3641 |
$in_body->($insert_to_current); |
| 3642 |
redo B; |
| 3643 |
} elsif ($self->{insertion_mode} eq 'in column group') { |
| 3644 |
if ($token->{type} eq 'character') { |
| 3645 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3646 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3647 |
unless (length $token->{data}) { |
| 3648 |
!!!next-token; |
| 3649 |
redo B; |
| 3650 |
} |
| 3651 |
} |
| 3652 |
|
| 3653 |
# |
| 3654 |
} elsif ($token->{type} eq 'comment') { |
| 3655 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3656 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3657 |
!!!next-token; |
| 3658 |
redo B; |
| 3659 |
} elsif ($token->{type} eq 'start tag') { |
| 3660 |
if ($token->{tag_name} eq 'col') { |
| 3661 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3662 |
pop @{$self->{open_elements}}; |
| 3663 |
!!!next-token; |
| 3664 |
redo B; |
| 3665 |
} else { |
| 3666 |
# |
| 3667 |
} |
| 3668 |
} elsif ($token->{type} eq 'end tag') { |
| 3669 |
if ($token->{tag_name} eq 'colgroup') { |
| 3670 |
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
| 3671 |
!!!parse-error (type => 'unmatched end tag:colgroup'); |
| 3672 |
## Ignore the token |
| 3673 |
!!!next-token; |
| 3674 |
redo B; |
| 3675 |
} else { |
| 3676 |
pop @{$self->{open_elements}}; # colgroup |
| 3677 |
$self->{insertion_mode} = 'in table'; |
| 3678 |
!!!next-token; |
| 3679 |
redo B; |
| 3680 |
} |
| 3681 |
} elsif ($token->{tag_name} eq 'col') { |
| 3682 |
!!!parse-error (type => 'unmatched end tag:col'); |
| 3683 |
## Ignore the token |
| 3684 |
!!!next-token; |
| 3685 |
redo B; |
| 3686 |
} else { |
| 3687 |
# |
| 3688 |
} |
| 3689 |
} else { |
| 3690 |
# |
| 3691 |
} |
| 3692 |
|
| 3693 |
## As if </colgroup> |
| 3694 |
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
| 3695 |
!!!parse-error (type => 'unmatched end tag:colgroup'); |
| 3696 |
## Ignore the token |
| 3697 |
!!!next-token; |
| 3698 |
redo B; |
| 3699 |
} else { |
| 3700 |
pop @{$self->{open_elements}}; # colgroup |
| 3701 |
$self->{insertion_mode} = 'in table'; |
| 3702 |
## reprocess |
| 3703 |
redo B; |
| 3704 |
} |
| 3705 |
} elsif ($self->{insertion_mode} eq 'in table body') { |
| 3706 |
if ($token->{type} eq 'character') { |
| 3707 |
## NOTE: This is a "character in table" code clone. |
| 3708 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3709 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3710 |
|
| 3711 |
unless (length $token->{data}) { |
| 3712 |
!!!next-token; |
| 3713 |
redo B; |
| 3714 |
} |
| 3715 |
} |
| 3716 |
|
| 3717 |
!!!parse-error (type => 'in table:#character'); |
| 3718 |
|
| 3719 |
## As if in body, but insert into foster parent element |
| 3720 |
## ISSUE: Spec says that "whenever a node would be inserted |
| 3721 |
## into the current node" while characters might not be |
| 3722 |
## result in a new Text node. |
| 3723 |
$reconstruct_active_formatting_elements->($insert_to_foster); |
| 3724 |
|
| 3725 |
if ({ |
| 3726 |
table => 1, tbody => 1, tfoot => 1, |
| 3727 |
thead => 1, tr => 1, |
| 3728 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3729 |
# MUST |
| 3730 |
my $foster_parent_element; |
| 3731 |
my $next_sibling; |
| 3732 |
my $prev_sibling; |
| 3733 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3734 |
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
| 3735 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 3736 |
if (defined $parent and $parent->node_type == 1) { |
| 3737 |
$foster_parent_element = $parent; |
| 3738 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 3739 |
$prev_sibling = $next_sibling->previous_sibling; |
| 3740 |
} else { |
| 3741 |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 3742 |
$prev_sibling = $foster_parent_element->last_child; |
| 3743 |
} |
| 3744 |
last OE; |
| 3745 |
} |
| 3746 |
} # OE |
| 3747 |
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
| 3748 |
$prev_sibling = $foster_parent_element->last_child |
| 3749 |
unless defined $foster_parent_element; |
| 3750 |
if (defined $prev_sibling and |
| 3751 |
$prev_sibling->node_type == 3) { |
| 3752 |
$prev_sibling->manakai_append_text ($token->{data}); |
| 3753 |
} else { |
| 3754 |
$foster_parent_element->insert_before |
| 3755 |
($self->{document}->create_text_node ($token->{data}), |
| 3756 |
$next_sibling); |
| 3757 |
} |
| 3758 |
} else { |
| 3759 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3760 |
} |
| 3761 |
|
| 3762 |
!!!next-token; |
| 3763 |
redo B; |
| 3764 |
} elsif ($token->{type} eq 'comment') { |
| 3765 |
## Copied from 'in table' |
| 3766 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 3767 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 3768 |
!!!next-token; |
| 3769 |
redo B; |
| 3770 |
} elsif ($token->{type} eq 'start tag') { |
| 3771 |
if ({ |
| 3772 |
tr => 1, |
| 3773 |
th => 1, td => 1, |
| 3774 |
}->{$token->{tag_name}}) { |
| 3775 |
unless ($token->{tag_name} eq 'tr') { |
| 3776 |
!!!parse-error (type => 'missing start tag:tr'); |
| 3777 |
} |
| 3778 |
|
| 3779 |
## Clear back to table body context |
| 3780 |
while (not { |
| 3781 |
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3782 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3783 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3784 |
pop @{$self->{open_elements}}; |
| 3785 |
} |
| 3786 |
|
| 3787 |
$self->{insertion_mode} = 'in row'; |
| 3788 |
if ($token->{tag_name} eq 'tr') { |
| 3789 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3790 |
!!!next-token; |
| 3791 |
} else { |
| 3792 |
!!!insert-element ('tr'); |
| 3793 |
## reprocess |
| 3794 |
} |
| 3795 |
redo B; |
| 3796 |
} elsif ({ |
| 3797 |
caption => 1, col => 1, colgroup => 1, |
| 3798 |
tbody => 1, tfoot => 1, thead => 1, |
| 3799 |
}->{$token->{tag_name}}) { |
| 3800 |
## have an element in table scope |
| 3801 |
my $i; |
| 3802 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3803 |
my $node = $self->{open_elements}->[$_]; |
| 3804 |
if ({ |
| 3805 |
tbody => 1, thead => 1, tfoot => 1, |
| 3806 |
}->{$node->[1]}) { |
| 3807 |
$i = $_; |
| 3808 |
last INSCOPE; |
| 3809 |
} elsif ({ |
| 3810 |
table => 1, html => 1, |
| 3811 |
}->{$node->[1]}) { |
| 3812 |
last INSCOPE; |
| 3813 |
} |
| 3814 |
} # INSCOPE |
| 3815 |
unless (defined $i) { |
| 3816 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3817 |
## Ignore the token |
| 3818 |
!!!next-token; |
| 3819 |
redo B; |
| 3820 |
} |
| 3821 |
|
| 3822 |
## Clear back to table body context |
| 3823 |
while (not { |
| 3824 |
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3825 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3826 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3827 |
pop @{$self->{open_elements}}; |
| 3828 |
} |
| 3829 |
|
| 3830 |
## As if <{current node}> |
| 3831 |
## have an element in table scope |
| 3832 |
## true by definition |
| 3833 |
|
| 3834 |
## Clear back to table body context |
| 3835 |
## nop by definition |
| 3836 |
|
| 3837 |
pop @{$self->{open_elements}}; |
| 3838 |
$self->{insertion_mode} = 'in table'; |
| 3839 |
## reprocess |
| 3840 |
redo B; |
| 3841 |
} elsif ($token->{tag_name} eq 'table') { |
| 3842 |
## NOTE: This is a code clone of "table in table" |
| 3843 |
!!!parse-error (type => 'not closed:table'); |
| 3844 |
|
| 3845 |
## As if </table> |
| 3846 |
## have a table element in table scope |
| 3847 |
my $i; |
| 3848 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3849 |
my $node = $self->{open_elements}->[$_]; |
| 3850 |
if ($node->[1] eq 'table') { |
| 3851 |
$i = $_; |
| 3852 |
last INSCOPE; |
| 3853 |
} elsif ({ |
| 3854 |
table => 1, html => 1, |
| 3855 |
}->{$node->[1]}) { |
| 3856 |
last INSCOPE; |
| 3857 |
} |
| 3858 |
} # INSCOPE |
| 3859 |
unless (defined $i) { |
| 3860 |
!!!parse-error (type => 'unmatched end tag:table'); |
| 3861 |
## Ignore tokens </table><table> |
| 3862 |
!!!next-token; |
| 3863 |
redo B; |
| 3864 |
} |
| 3865 |
|
| 3866 |
## generate implied end tags |
| 3867 |
if ({ |
| 3868 |
dd => 1, dt => 1, li => 1, p => 1, |
| 3869 |
td => 1, th => 1, tr => 1, |
| 3870 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3871 |
!!!back-token; # <table> |
| 3872 |
$token = {type => 'end tag', tag_name => 'table'}; |
| 3873 |
!!!back-token; |
| 3874 |
$token = {type => 'end tag', |
| 3875 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3876 |
redo B; |
| 3877 |
} |
| 3878 |
|
| 3879 |
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
| 3880 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3881 |
} |
| 3882 |
|
| 3883 |
splice @{$self->{open_elements}}, $i; |
| 3884 |
|
| 3885 |
$self->_reset_insertion_mode; |
| 3886 |
|
| 3887 |
## reprocess |
| 3888 |
redo B; |
| 3889 |
} else { |
| 3890 |
# |
| 3891 |
} |
| 3892 |
} elsif ($token->{type} eq 'end tag') { |
| 3893 |
if ({ |
| 3894 |
tbody => 1, tfoot => 1, thead => 1, |
| 3895 |
}->{$token->{tag_name}}) { |
| 3896 |
## have an element in table scope |
| 3897 |
my $i; |
| 3898 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3899 |
my $node = $self->{open_elements}->[$_]; |
| 3900 |
if ($node->[1] eq $token->{tag_name}) { |
| 3901 |
$i = $_; |
| 3902 |
last INSCOPE; |
| 3903 |
} elsif ({ |
| 3904 |
table => 1, html => 1, |
| 3905 |
}->{$node->[1]}) { |
| 3906 |
last INSCOPE; |
| 3907 |
} |
| 3908 |
} # INSCOPE |
| 3909 |
unless (defined $i) { |
| 3910 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3911 |
## Ignore the token |
| 3912 |
!!!next-token; |
| 3913 |
redo B; |
| 3914 |
} |
| 3915 |
|
| 3916 |
## Clear back to table body context |
| 3917 |
while (not { |
| 3918 |
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3919 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3920 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3921 |
pop @{$self->{open_elements}}; |
| 3922 |
} |
| 3923 |
|
| 3924 |
pop @{$self->{open_elements}}; |
| 3925 |
$self->{insertion_mode} = 'in table'; |
| 3926 |
!!!next-token; |
| 3927 |
redo B; |
| 3928 |
} elsif ($token->{tag_name} eq 'table') { |
| 3929 |
## have an element in table scope |
| 3930 |
my $i; |
| 3931 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3932 |
my $node = $self->{open_elements}->[$_]; |
| 3933 |
if ({ |
| 3934 |
tbody => 1, thead => 1, tfoot => 1, |
| 3935 |
}->{$node->[1]}) { |
| 3936 |
$i = $_; |
| 3937 |
last INSCOPE; |
| 3938 |
} elsif ({ |
| 3939 |
table => 1, html => 1, |
| 3940 |
}->{$node->[1]}) { |
| 3941 |
last INSCOPE; |
| 3942 |
} |
| 3943 |
} # INSCOPE |
| 3944 |
unless (defined $i) { |
| 3945 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3946 |
## Ignore the token |
| 3947 |
!!!next-token; |
| 3948 |
redo B; |
| 3949 |
} |
| 3950 |
|
| 3951 |
## Clear back to table body context |
| 3952 |
while (not { |
| 3953 |
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3954 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3955 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3956 |
pop @{$self->{open_elements}}; |
| 3957 |
} |
| 3958 |
|
| 3959 |
## As if <{current node}> |
| 3960 |
## have an element in table scope |
| 3961 |
## true by definition |
| 3962 |
|
| 3963 |
## Clear back to table body context |
| 3964 |
## nop by definition |
| 3965 |
|
| 3966 |
pop @{$self->{open_elements}}; |
| 3967 |
$self->{insertion_mode} = 'in table'; |
| 3968 |
## reprocess |
| 3969 |
redo B; |
| 3970 |
} elsif ({ |
| 3971 |
body => 1, caption => 1, col => 1, colgroup => 1, |
| 3972 |
html => 1, td => 1, th => 1, tr => 1, |
| 3973 |
}->{$token->{tag_name}}) { |
| 3974 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3975 |
## Ignore the token |
| 3976 |
!!!next-token; |
| 3977 |
redo B; |
| 3978 |
} else { |
| 3979 |
# |
| 3980 |
} |
| 3981 |
} else { |
| 3982 |
# |
| 3983 |
} |
| 3984 |
|
| 3985 |
## As if in table |
| 3986 |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
| 3987 |
$in_body->($insert_to_foster); |
| 3988 |
redo B; |
| 3989 |
} elsif ($self->{insertion_mode} eq 'in row') { |
| 3990 |
if ($token->{type} eq 'character') { |
| 3991 |
## NOTE: This is a "character in table" code clone. |
| 3992 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3993 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3994 |
|
| 3995 |
unless (length $token->{data}) { |
| 3996 |
!!!next-token; |
| 3997 |
redo B; |
| 3998 |
} |
| 3999 |
} |
| 4000 |
|
| 4001 |
!!!parse-error (type => 'in table:#character'); |
| 4002 |
|
| 4003 |
## As if in body, but insert into foster parent element |
| 4004 |
## ISSUE: Spec says that "whenever a node would be inserted |
| 4005 |
## into the current node" while characters might not be |
| 4006 |
## result in a new Text node. |
| 4007 |
$reconstruct_active_formatting_elements->($insert_to_foster); |
| 4008 |
|
| 4009 |
if ({ |
| 4010 |
table => 1, tbody => 1, tfoot => 1, |
| 4011 |
thead => 1, tr => 1, |
| 4012 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4013 |
# MUST |
| 4014 |
my $foster_parent_element; |
| 4015 |
my $next_sibling; |
| 4016 |
my $prev_sibling; |
| 4017 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4018 |
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
| 4019 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 4020 |
if (defined $parent and $parent->node_type == 1) { |
| 4021 |
$foster_parent_element = $parent; |
| 4022 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 4023 |
$prev_sibling = $next_sibling->previous_sibling; |
| 4024 |
} else { |
| 4025 |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 4026 |
$prev_sibling = $foster_parent_element->last_child; |
| 4027 |
} |
| 4028 |
last OE; |
| 4029 |
} |
| 4030 |
} # OE |
| 4031 |
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
| 4032 |
$prev_sibling = $foster_parent_element->last_child |
| 4033 |
unless defined $foster_parent_element; |
| 4034 |
if (defined $prev_sibling and |
| 4035 |
$prev_sibling->node_type == 3) { |
| 4036 |
$prev_sibling->manakai_append_text ($token->{data}); |
| 4037 |
} else { |
| 4038 |
$foster_parent_element->insert_before |
| 4039 |
($self->{document}->create_text_node ($token->{data}), |
| 4040 |
$next_sibling); |
| 4041 |
} |
| 4042 |
} else { |
| 4043 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4044 |
} |
| 4045 |
|
| 4046 |
!!!next-token; |
| 4047 |
redo B; |
| 4048 |
} elsif ($token->{type} eq 'comment') { |
| 4049 |
## Copied from 'in table' |
| 4050 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4051 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 4052 |
!!!next-token; |
| 4053 |
redo B; |
| 4054 |
} elsif ($token->{type} eq 'start tag') { |
| 4055 |
if ($token->{tag_name} eq 'th' or |
| 4056 |
$token->{tag_name} eq 'td') { |
| 4057 |
## Clear back to table row context |
| 4058 |
while (not { |
| 4059 |
tr => 1, html => 1, |
| 4060 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4061 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4062 |
pop @{$self->{open_elements}}; |
| 4063 |
} |
| 4064 |
|
| 4065 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4066 |
$self->{insertion_mode} = 'in cell'; |
| 4067 |
|
| 4068 |
push @$active_formatting_elements, ['#marker', '']; |
| 4069 |
|
| 4070 |
!!!next-token; |
| 4071 |
redo B; |
| 4072 |
} elsif ({ |
| 4073 |
caption => 1, col => 1, colgroup => 1, |
| 4074 |
tbody => 1, tfoot => 1, thead => 1, tr => 1, |
| 4075 |
}->{$token->{tag_name}}) { |
| 4076 |
## As if </tr> |
| 4077 |
## have an element in table scope |
| 4078 |
my $i; |
| 4079 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4080 |
my $node = $self->{open_elements}->[$_]; |
| 4081 |
if ($node->[1] eq 'tr') { |
| 4082 |
$i = $_; |
| 4083 |
last INSCOPE; |
| 4084 |
} elsif ({ |
| 4085 |
table => 1, html => 1, |
| 4086 |
}->{$node->[1]}) { |
| 4087 |
last INSCOPE; |
| 4088 |
} |
| 4089 |
} # INSCOPE |
| 4090 |
unless (defined $i) { |
| 4091 |
!!!parse-error (type => 'unmacthed end tag:'.$token->{tag_name}); |
| 4092 |
## Ignore the token |
| 4093 |
!!!next-token; |
| 4094 |
redo B; |
| 4095 |
} |
| 4096 |
|
| 4097 |
## Clear back to table row context |
| 4098 |
while (not { |
| 4099 |
tr => 1, html => 1, |
| 4100 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4101 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4102 |
pop @{$self->{open_elements}}; |
| 4103 |
} |
| 4104 |
|
| 4105 |
pop @{$self->{open_elements}}; # tr |
| 4106 |
$self->{insertion_mode} = 'in table body'; |
| 4107 |
## reprocess |
| 4108 |
redo B; |
| 4109 |
} elsif ($token->{tag_name} eq 'table') { |
| 4110 |
## NOTE: This is a code clone of "table in table" |
| 4111 |
!!!parse-error (type => 'not closed:table'); |
| 4112 |
|
| 4113 |
## As if </table> |
| 4114 |
## have a table element in table scope |
| 4115 |
my $i; |
| 4116 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4117 |
my $node = $self->{open_elements}->[$_]; |
| 4118 |
if ($node->[1] eq 'table') { |
| 4119 |
$i = $_; |
| 4120 |
last INSCOPE; |
| 4121 |
} elsif ({ |
| 4122 |
table => 1, html => 1, |
| 4123 |
}->{$node->[1]}) { |
| 4124 |
last INSCOPE; |
| 4125 |
} |
| 4126 |
} # INSCOPE |
| 4127 |
unless (defined $i) { |
| 4128 |
!!!parse-error (type => 'unmatched end tag:table'); |
| 4129 |
## Ignore tokens </table><table> |
| 4130 |
!!!next-token; |
| 4131 |
redo B; |
| 4132 |
} |
| 4133 |
|
| 4134 |
## generate implied end tags |
| 4135 |
if ({ |
| 4136 |
dd => 1, dt => 1, li => 1, p => 1, |
| 4137 |
td => 1, th => 1, tr => 1, |
| 4138 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4139 |
!!!back-token; # <table> |
| 4140 |
$token = {type => 'end tag', tag_name => 'table'}; |
| 4141 |
!!!back-token; |
| 4142 |
$token = {type => 'end tag', |
| 4143 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 4144 |
redo B; |
| 4145 |
} |
| 4146 |
|
| 4147 |
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
| 4148 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4149 |
} |
| 4150 |
|
| 4151 |
splice @{$self->{open_elements}}, $i; |
| 4152 |
|
| 4153 |
$self->_reset_insertion_mode; |
| 4154 |
|
| 4155 |
## reprocess |
| 4156 |
redo B; |
| 4157 |
} else { |
| 4158 |
# |
| 4159 |
} |
| 4160 |
} elsif ($token->{type} eq 'end tag') { |
| 4161 |
if ($token->{tag_name} eq 'tr') { |
| 4162 |
## have an element in table scope |
| 4163 |
my $i; |
| 4164 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4165 |
my $node = $self->{open_elements}->[$_]; |
| 4166 |
if ($node->[1] eq $token->{tag_name}) { |
| 4167 |
$i = $_; |
| 4168 |
last INSCOPE; |
| 4169 |
} elsif ({ |
| 4170 |
table => 1, html => 1, |
| 4171 |
}->{$node->[1]}) { |
| 4172 |
last INSCOPE; |
| 4173 |
} |
| 4174 |
} # INSCOPE |
| 4175 |
unless (defined $i) { |
| 4176 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4177 |
## Ignore the token |
| 4178 |
!!!next-token; |
| 4179 |
redo B; |
| 4180 |
} |
| 4181 |
|
| 4182 |
## Clear back to table row context |
| 4183 |
while (not { |
| 4184 |
tr => 1, html => 1, |
| 4185 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4186 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4187 |
pop @{$self->{open_elements}}; |
| 4188 |
} |
| 4189 |
|
| 4190 |
pop @{$self->{open_elements}}; # tr |
| 4191 |
$self->{insertion_mode} = 'in table body'; |
| 4192 |
!!!next-token; |
| 4193 |
redo B; |
| 4194 |
} elsif ($token->{tag_name} eq 'table') { |
| 4195 |
## As if </tr> |
| 4196 |
## have an element in table scope |
| 4197 |
my $i; |
| 4198 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4199 |
my $node = $self->{open_elements}->[$_]; |
| 4200 |
if ($node->[1] eq 'tr') { |
| 4201 |
$i = $_; |
| 4202 |
last INSCOPE; |
| 4203 |
} elsif ({ |
| 4204 |
table => 1, html => 1, |
| 4205 |
}->{$node->[1]}) { |
| 4206 |
last INSCOPE; |
| 4207 |
} |
| 4208 |
} # INSCOPE |
| 4209 |
unless (defined $i) { |
| 4210 |
!!!parse-error (type => 'unmatched end tag:'.$token->{type}); |
| 4211 |
## Ignore the token |
| 4212 |
!!!next-token; |
| 4213 |
redo B; |
| 4214 |
} |
| 4215 |
|
| 4216 |
## Clear back to table row context |
| 4217 |
while (not { |
| 4218 |
tr => 1, html => 1, |
| 4219 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4220 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4221 |
pop @{$self->{open_elements}}; |
| 4222 |
} |
| 4223 |
|
| 4224 |
pop @{$self->{open_elements}}; # tr |
| 4225 |
$self->{insertion_mode} = 'in table body'; |
| 4226 |
## reprocess |
| 4227 |
redo B; |
| 4228 |
} elsif ({ |
| 4229 |
tbody => 1, tfoot => 1, thead => 1, |
| 4230 |
}->{$token->{tag_name}}) { |
| 4231 |
## have an element in table scope |
| 4232 |
my $i; |
| 4233 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4234 |
my $node = $self->{open_elements}->[$_]; |
| 4235 |
if ($node->[1] eq $token->{tag_name}) { |
| 4236 |
$i = $_; |
| 4237 |
last INSCOPE; |
| 4238 |
} elsif ({ |
| 4239 |
table => 1, html => 1, |
| 4240 |
}->{$node->[1]}) { |
| 4241 |
last INSCOPE; |
| 4242 |
} |
| 4243 |
} # INSCOPE |
| 4244 |
unless (defined $i) { |
| 4245 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4246 |
## Ignore the token |
| 4247 |
!!!next-token; |
| 4248 |
redo B; |
| 4249 |
} |
| 4250 |
|
| 4251 |
## As if </tr> |
| 4252 |
## have an element in table scope |
| 4253 |
my $i; |
| 4254 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4255 |
my $node = $self->{open_elements}->[$_]; |
| 4256 |
if ($node->[1] eq 'tr') { |
| 4257 |
$i = $_; |
| 4258 |
last INSCOPE; |
| 4259 |
} elsif ({ |
| 4260 |
table => 1, html => 1, |
| 4261 |
}->{$node->[1]}) { |
| 4262 |
last INSCOPE; |
| 4263 |
} |
| 4264 |
} # INSCOPE |
| 4265 |
unless (defined $i) { |
| 4266 |
!!!parse-error (type => 'unmatched end tag:tr'); |
| 4267 |
## Ignore the token |
| 4268 |
!!!next-token; |
| 4269 |
redo B; |
| 4270 |
} |
| 4271 |
|
| 4272 |
## Clear back to table row context |
| 4273 |
while (not { |
| 4274 |
tr => 1, html => 1, |
| 4275 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4276 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4277 |
pop @{$self->{open_elements}}; |
| 4278 |
} |
| 4279 |
|
| 4280 |
pop @{$self->{open_elements}}; # tr |
| 4281 |
$self->{insertion_mode} = 'in table body'; |
| 4282 |
## reprocess |
| 4283 |
redo B; |
| 4284 |
} elsif ({ |
| 4285 |
body => 1, caption => 1, col => 1, |
| 4286 |
colgroup => 1, html => 1, td => 1, th => 1, |
| 4287 |
}->{$token->{tag_name}}) { |
| 4288 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4289 |
## Ignore the token |
| 4290 |
!!!next-token; |
| 4291 |
redo B; |
| 4292 |
} else { |
| 4293 |
# |
| 4294 |
} |
| 4295 |
} else { |
| 4296 |
# |
| 4297 |
} |
| 4298 |
|
| 4299 |
## As if in table |
| 4300 |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
| 4301 |
$in_body->($insert_to_foster); |
| 4302 |
redo B; |
| 4303 |
} elsif ($self->{insertion_mode} eq 'in cell') { |
| 4304 |
if ($token->{type} eq 'character') { |
| 4305 |
## NOTE: This is a code clone of "character in body". |
| 4306 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4307 |
|
| 4308 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4309 |
|
| 4310 |
!!!next-token; |
| 4311 |
redo B; |
| 4312 |
} elsif ($token->{type} eq 'comment') { |
| 4313 |
## NOTE: This is a code clone of "comment in body". |
| 4314 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4315 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 4316 |
!!!next-token; |
| 4317 |
redo B; |
| 4318 |
} elsif ($token->{type} eq 'start tag') { |
| 4319 |
if ({ |
| 4320 |
caption => 1, col => 1, colgroup => 1, |
| 4321 |
tbody => 1, td => 1, tfoot => 1, th => 1, |
| 4322 |
thead => 1, tr => 1, |
| 4323 |
}->{$token->{tag_name}}) { |
| 4324 |
## have an element in table scope |
| 4325 |
my $tn; |
| 4326 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4327 |
my $node = $self->{open_elements}->[$_]; |
| 4328 |
if ($node->[1] eq 'td' or $node->[1] eq 'th') { |
| 4329 |
$tn = $node->[1]; |
| 4330 |
last INSCOPE; |
| 4331 |
} elsif ({ |
| 4332 |
table => 1, html => 1, |
| 4333 |
}->{$node->[1]}) { |
| 4334 |
last INSCOPE; |
| 4335 |
} |
| 4336 |
} # INSCOPE |
| 4337 |
unless (defined $tn) { |
| 4338 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4339 |
## Ignore the token |
| 4340 |
!!!next-token; |
| 4341 |
redo B; |
| 4342 |
} |
| 4343 |
|
| 4344 |
## Close the cell |
| 4345 |
!!!back-token; # <?> |
| 4346 |
$token = {type => 'end tag', tag_name => $tn}; |
| 4347 |
redo B; |
| 4348 |
} else { |
| 4349 |
# |
| 4350 |
} |
| 4351 |
} elsif ($token->{type} eq 'end tag') { |
| 4352 |
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
| 4353 |
## have an element in table scope |
| 4354 |
my $i; |
| 4355 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4356 |
my $node = $self->{open_elements}->[$_]; |
| 4357 |
if ($node->[1] eq $token->{tag_name}) { |
| 4358 |
$i = $_; |
| 4359 |
last INSCOPE; |
| 4360 |
} elsif ({ |
| 4361 |
table => 1, html => 1, |
| 4362 |
}->{$node->[1]}) { |
| 4363 |
last INSCOPE; |
| 4364 |
} |
| 4365 |
} # INSCOPE |
| 4366 |
unless (defined $i) { |
| 4367 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4368 |
## Ignore the token |
| 4369 |
!!!next-token; |
| 4370 |
redo B; |
| 4371 |
} |
| 4372 |
|
| 4373 |
## generate implied end tags |
| 4374 |
if ({ |
| 4375 |
dd => 1, dt => 1, li => 1, p => 1, |
| 4376 |
td => ($token->{tag_name} eq 'th'), |
| 4377 |
th => ($token->{tag_name} eq 'td'), |
| 4378 |
tr => 1, |
| 4379 |
}->{$self->{open_elements}->[-1]->[1]}) { |
| 4380 |
!!!back-token; |
| 4381 |
$token = {type => 'end tag', |
| 4382 |
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 4383 |
redo B; |
| 4384 |
} |
| 4385 |
|
| 4386 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
| 4387 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 4388 |
} |
| 4389 |
|
| 4390 |
splice @{$self->{open_elements}}, $i; |
| 4391 |
|
| 4392 |
$clear_up_to_marker->(); |
| 4393 |
|
| 4394 |
$self->{insertion_mode} = 'in row'; |
| 4395 |
|
| 4396 |
!!!next-token; |
| 4397 |
redo B; |
| 4398 |
} elsif ({ |
| 4399 |
body => 1, caption => 1, col => 1, |
| 4400 |
colgroup => 1, html => 1, |
| 4401 |
}->{$token->{tag_name}}) { |
| 4402 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4403 |
## Ignore the token |
| 4404 |
!!!next-token; |
| 4405 |
redo B; |
| 4406 |
} elsif ({ |
| 4407 |
table => 1, tbody => 1, tfoot => 1, |
| 4408 |
thead => 1, tr => 1, |
| 4409 |
}->{$token->{tag_name}}) { |
| 4410 |
## have an element in table scope |
| 4411 |
my $i; |
| 4412 |
my $tn; |
| 4413 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4414 |
my $node = $self->{open_elements}->[$_]; |
| 4415 |
if ($node->[1] eq $token->{tag_name}) { |
| 4416 |
$i = $_; |
| 4417 |
last INSCOPE; |
| 4418 |
} elsif ($node->[1] eq 'td' or $node->[1] eq 'th') { |
| 4419 |
$tn = $node->[1]; |
| 4420 |
## NOTE: There is exactly one |td| or |th| element |
| 4421 |
## in scope in the stack of open elements by definition. |
| 4422 |
} elsif ({ |
| 4423 |
table => 1, html => 1, |
| 4424 |
}->{$node->[1]}) { |
| 4425 |
last INSCOPE; |
| 4426 |
} |
| 4427 |
} # INSCOPE |
| 4428 |
unless (defined $i) { |
| 4429 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4430 |
## Ignore the token |
| 4431 |
!!!next-token; |
| 4432 |
redo B; |
| 4433 |
} |
| 4434 |
|
| 4435 |
## Close the cell |
| 4436 |
!!!back-token; # </?> |
| 4437 |
$token = {type => 'end tag', tag_name => $tn}; |
| 4438 |
redo B; |
| 4439 |
} else { |
| 4440 |
# |
| 4441 |
} |
| 4442 |
} else { |
| 4443 |
# |
| 4444 |
} |
| 4445 |
|
| 4446 |
$in_body->($insert_to_current); |
| 4447 |
redo B; |
| 4448 |
} elsif ($self->{insertion_mode} eq 'in select') { |
| 4449 |
if ($token->{type} eq 'character') { |
| 4450 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4451 |
!!!next-token; |
| 4452 |
redo B; |
| 4453 |
} elsif ($token->{type} eq 'comment') { |
| 4454 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4455 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 4456 |
!!!next-token; |
| 4457 |
redo B; |
| 4458 |
} elsif ($token->{type} eq 'start tag') { |
| 4459 |
if ($token->{tag_name} eq 'option') { |
| 4460 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 4461 |
## As if </option> |
| 4462 |
pop @{$self->{open_elements}}; |
| 4463 |
} |
| 4464 |
|
| 4465 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4466 |
!!!next-token; |
| 4467 |
redo B; |
| 4468 |
} elsif ($token->{tag_name} eq 'optgroup') { |
| 4469 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 4470 |
## As if </option> |
| 4471 |
pop @{$self->{open_elements}}; |
| 4472 |
} |
| 4473 |
|
| 4474 |
if ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
| 4475 |
## As if </optgroup> |
| 4476 |
pop @{$self->{open_elements}}; |
| 4477 |
} |
| 4478 |
|
| 4479 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4480 |
!!!next-token; |
| 4481 |
redo B; |
| 4482 |
} elsif ($token->{tag_name} eq 'select') { |
| 4483 |
!!!parse-error (type => 'not closed:select'); |
| 4484 |
## As if </select> instead |
| 4485 |
## have an element in table scope |
| 4486 |
my $i; |
| 4487 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4488 |
my $node = $self->{open_elements}->[$_]; |
| 4489 |
if ($node->[1] eq $token->{tag_name}) { |
| 4490 |
$i = $_; |
| 4491 |
last INSCOPE; |
| 4492 |
} elsif ({ |
| 4493 |
table => 1, html => 1, |
| 4494 |
}->{$node->[1]}) { |
| 4495 |
last INSCOPE; |
| 4496 |
} |
| 4497 |
} # INSCOPE |
| 4498 |
unless (defined $i) { |
| 4499 |
!!!parse-error (type => 'unmatched end tag:select'); |
| 4500 |
## Ignore the token |
| 4501 |
!!!next-token; |
| 4502 |
redo B; |
| 4503 |
} |
| 4504 |
|
| 4505 |
splice @{$self->{open_elements}}, $i; |
| 4506 |
|
| 4507 |
$self->_reset_insertion_mode; |
| 4508 |
|
| 4509 |
!!!next-token; |
| 4510 |
redo B; |
| 4511 |
} else { |
| 4512 |
# |
| 4513 |
} |
| 4514 |
} elsif ($token->{type} eq 'end tag') { |
| 4515 |
if ($token->{tag_name} eq 'optgroup') { |
| 4516 |
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
| 4517 |
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
| 4518 |
## As if </option> |
| 4519 |
splice @{$self->{open_elements}}, -2; |
| 4520 |
} elsif ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
| 4521 |
pop @{$self->{open_elements}}; |
| 4522 |
} else { |
| 4523 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4524 |
## Ignore the token |
| 4525 |
} |
| 4526 |
!!!next-token; |
| 4527 |
redo B; |
| 4528 |
} elsif ($token->{tag_name} eq 'option') { |
| 4529 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 4530 |
pop @{$self->{open_elements}}; |
| 4531 |
} else { |
| 4532 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4533 |
## Ignore the token |
| 4534 |
} |
| 4535 |
!!!next-token; |
| 4536 |
redo B; |
| 4537 |
} elsif ($token->{tag_name} eq 'select') { |
| 4538 |
## have an element in table scope |
| 4539 |
my $i; |
| 4540 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4541 |
my $node = $self->{open_elements}->[$_]; |
| 4542 |
if ($node->[1] eq $token->{tag_name}) { |
| 4543 |
$i = $_; |
| 4544 |
last INSCOPE; |
| 4545 |
} elsif ({ |
| 4546 |
table => 1, html => 1, |
| 4547 |
}->{$node->[1]}) { |
| 4548 |
last INSCOPE; |
| 4549 |
} |
| 4550 |
} # INSCOPE |
| 4551 |
unless (defined $i) { |
| 4552 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4553 |
## Ignore the token |
| 4554 |
!!!next-token; |
| 4555 |
redo B; |
| 4556 |
} |
| 4557 |
|
| 4558 |
splice @{$self->{open_elements}}, $i; |
| 4559 |
|
| 4560 |
$self->_reset_insertion_mode; |
| 4561 |
|
| 4562 |
!!!next-token; |
| 4563 |
redo B; |
| 4564 |
} elsif ({ |
| 4565 |
caption => 1, table => 1, tbody => 1, |
| 4566 |
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
| 4567 |
}->{$token->{tag_name}}) { |
| 4568 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4569 |
|
| 4570 |
## have an element in table scope |
| 4571 |
my $i; |
| 4572 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4573 |
my $node = $self->{open_elements}->[$_]; |
| 4574 |
if ($node->[1] eq $token->{tag_name}) { |
| 4575 |
$i = $_; |
| 4576 |
last INSCOPE; |
| 4577 |
} elsif ({ |
| 4578 |
table => 1, html => 1, |
| 4579 |
}->{$node->[1]}) { |
| 4580 |
last INSCOPE; |
| 4581 |
} |
| 4582 |
} # INSCOPE |
| 4583 |
unless (defined $i) { |
| 4584 |
## Ignore the token |
| 4585 |
!!!next-token; |
| 4586 |
redo B; |
| 4587 |
} |
| 4588 |
|
| 4589 |
## As if </select> |
| 4590 |
## have an element in table scope |
| 4591 |
undef $i; |
| 4592 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4593 |
my $node = $self->{open_elements}->[$_]; |
| 4594 |
if ($node->[1] eq 'select') { |
| 4595 |
$i = $_; |
| 4596 |
last INSCOPE; |
| 4597 |
} elsif ({ |
| 4598 |
table => 1, html => 1, |
| 4599 |
}->{$node->[1]}) { |
| 4600 |
last INSCOPE; |
| 4601 |
} |
| 4602 |
} # INSCOPE |
| 4603 |
unless (defined $i) { |
| 4604 |
!!!parse-error (type => 'unmatched end tag:select'); |
| 4605 |
## Ignore the </select> token |
| 4606 |
!!!next-token; ## TODO: ok? |
| 4607 |
redo B; |
| 4608 |
} |
| 4609 |
|
| 4610 |
splice @{$self->{open_elements}}, $i; |
| 4611 |
|
| 4612 |
$self->_reset_insertion_mode; |
| 4613 |
|
| 4614 |
## reprocess |
| 4615 |
redo B; |
| 4616 |
} else { |
| 4617 |
# |
| 4618 |
} |
| 4619 |
} else { |
| 4620 |
# |
| 4621 |
} |
| 4622 |
|
| 4623 |
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
| 4624 |
## Ignore the token |
| 4625 |
!!!next-token; |
| 4626 |
redo B; |
| 4627 |
} elsif ($self->{insertion_mode} eq 'after body') { |
| 4628 |
if ($token->{type} eq 'character') { |
| 4629 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4630 |
## As if in body |
| 4631 |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4632 |
|
| 4633 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4634 |
|
| 4635 |
unless (length $token->{data}) { |
| 4636 |
!!!next-token; |
| 4637 |
redo B; |
| 4638 |
} |
| 4639 |
} |
| 4640 |
|
| 4641 |
# |
| 4642 |
!!!parse-error (type => 'after body:#'.$token->{type}); |
| 4643 |
} elsif ($token->{type} eq 'comment') { |
| 4644 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4645 |
$self->{open_elements}->[0]->[0]->append_child ($comment); |
| 4646 |
!!!next-token; |
| 4647 |
redo B; |
| 4648 |
} elsif ($token->{type} eq 'start tag') { |
| 4649 |
!!!parse-error (type => 'after body:'.$token->{tag_name}); |
| 4650 |
# |
| 4651 |
} elsif ($token->{type} eq 'end tag') { |
| 4652 |
if ($token->{tag_name} eq 'html') { |
| 4653 |
if (defined $self->{inner_html_node}) { |
| 4654 |
!!!parse-error (type => 'unmatched end tag:html'); |
| 4655 |
## Ignore the token |
| 4656 |
!!!next-token; |
| 4657 |
redo B; |
| 4658 |
} else { |
| 4659 |
$phase = 'trailing end'; |
| 4660 |
!!!next-token; |
| 4661 |
redo B; |
| 4662 |
} |
| 4663 |
} else { |
| 4664 |
!!!parse-error (type => 'after body:/'.$token->{tag_name}); |
| 4665 |
} |
| 4666 |
} else { |
| 4667 |
!!!parse-error (type => 'after body:#'.$token->{type}); |
| 4668 |
} |
| 4669 |
|
| 4670 |
$self->{insertion_mode} = 'in body'; |
| 4671 |
## reprocess |
| 4672 |
redo B; |
| 4673 |
} elsif ($self->{insertion_mode} eq 'in frameset') { |
| 4674 |
if ($token->{type} eq 'character') { |
| 4675 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4676 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4677 |
|
| 4678 |
unless (length $token->{data}) { |
| 4679 |
!!!next-token; |
| 4680 |
redo B; |
| 4681 |
} |
| 4682 |
} |
| 4683 |
|
| 4684 |
# |
| 4685 |
} elsif ($token->{type} eq 'comment') { |
| 4686 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4687 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 4688 |
!!!next-token; |
| 4689 |
redo B; |
| 4690 |
} elsif ($token->{type} eq 'start tag') { |
| 4691 |
if ($token->{tag_name} eq 'frameset') { |
| 4692 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4693 |
!!!next-token; |
| 4694 |
redo B; |
| 4695 |
} elsif ($token->{tag_name} eq 'frame') { |
| 4696 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4697 |
pop @{$self->{open_elements}}; |
| 4698 |
!!!next-token; |
| 4699 |
redo B; |
| 4700 |
} elsif ($token->{tag_name} eq 'noframes') { |
| 4701 |
$in_body->($insert_to_current); |
| 4702 |
redo B; |
| 4703 |
} else { |
| 4704 |
# |
| 4705 |
} |
| 4706 |
} elsif ($token->{type} eq 'end tag') { |
| 4707 |
if ($token->{tag_name} eq 'frameset') { |
| 4708 |
if ($self->{open_elements}->[-1]->[1] eq 'html' and |
| 4709 |
@{$self->{open_elements}} == 1) { |
| 4710 |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4711 |
## Ignore the token |
| 4712 |
!!!next-token; |
| 4713 |
} else { |
| 4714 |
pop @{$self->{open_elements}}; |
| 4715 |
!!!next-token; |
| 4716 |
} |
| 4717 |
|
| 4718 |
## if not inner_html and |
| 4719 |
if ($self->{open_elements}->[-1]->[1] ne 'frameset') { |
| 4720 |
$self->{insertion_mode} = 'after frameset'; |
| 4721 |
} |
| 4722 |
redo B; |
| 4723 |
} else { |
| 4724 |
# |
| 4725 |
} |
| 4726 |
} else { |
| 4727 |
# |
| 4728 |
} |
| 4729 |
|
| 4730 |
if (defined $token->{tag_name}) { |
| 4731 |
!!!parse-error (type => 'in frameset:'.$token->{tag_name}); |
| 4732 |
} else { |
| 4733 |
!!!parse-error (type => 'in frameset:#'.$token->{type}); |
| 4734 |
} |
| 4735 |
## Ignore the token |
| 4736 |
!!!next-token; |
| 4737 |
redo B; |
| 4738 |
} elsif ($self->{insertion_mode} eq 'after frameset') { |
| 4739 |
if ($token->{type} eq 'character') { |
| 4740 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4741 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4742 |
|
| 4743 |
unless (length $token->{data}) { |
| 4744 |
!!!next-token; |
| 4745 |
redo B; |
| 4746 |
} |
| 4747 |
} |
| 4748 |
|
| 4749 |
# |
| 4750 |
} elsif ($token->{type} eq 'comment') { |
| 4751 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4752 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 4753 |
!!!next-token; |
| 4754 |
redo B; |
| 4755 |
} elsif ($token->{type} eq 'start tag') { |
| 4756 |
if ($token->{tag_name} eq 'noframes') { |
| 4757 |
$in_body->($insert_to_current); |
| 4758 |
redo B; |
| 4759 |
} else { |
| 4760 |
# |
| 4761 |
} |
| 4762 |
} elsif ($token->{type} eq 'end tag') { |
| 4763 |
if ($token->{tag_name} eq 'html') { |
| 4764 |
$phase = 'trailing end'; |
| 4765 |
!!!next-token; |
| 4766 |
redo B; |
| 4767 |
} else { |
| 4768 |
# |
| 4769 |
} |
| 4770 |
} else { |
| 4771 |
# |
| 4772 |
} |
| 4773 |
|
| 4774 |
if (defined $token->{tag_name}) { |
| 4775 |
!!!parse-error (type => 'after frameset:'.$token->{tag_name}); |
| 4776 |
} else { |
| 4777 |
!!!parse-error (type => 'after frameset:#'.$token->{type}); |
| 4778 |
} |
| 4779 |
## Ignore the token |
| 4780 |
!!!next-token; |
| 4781 |
redo B; |
| 4782 |
|
| 4783 |
## ISSUE: An issue in spec there |
| 4784 |
} else { |
| 4785 |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 4786 |
} |
| 4787 |
} |
| 4788 |
} elsif ($phase eq 'trailing end') { |
| 4789 |
## states in the main stage is preserved yet # MUST |
| 4790 |
|
| 4791 |
if ($token->{type} eq 'DOCTYPE') { |
| 4792 |
!!!parse-error (type => 'after html:#DOCTYPE'); |
| 4793 |
## Ignore the token |
| 4794 |
!!!next-token; |
| 4795 |
redo B; |
| 4796 |
} elsif ($token->{type} eq 'comment') { |
| 4797 |
my $comment = $self->{document}->create_comment ($token->{data}); |
| 4798 |
$self->{document}->append_child ($comment); |
| 4799 |
!!!next-token; |
| 4800 |
redo B; |
| 4801 |
} elsif ($token->{type} eq 'character') { |
| 4802 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4803 |
my $data = $1; |
| 4804 |
## As if in the main phase. |
| 4805 |
## NOTE: The insertion mode in the main phase |
| 4806 |
## just before the phase has been changed to the trailing |
| 4807 |
## end phase is either "after body" or "after frameset". |
| 4808 |
$reconstruct_active_formatting_elements->($insert_to_current) |
| 4809 |
if $phase eq 'main'; |
| 4810 |
|
| 4811 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($data); |
| 4812 |
|
| 4813 |
unless (length $token->{data}) { |
| 4814 |
!!!next-token; |
| 4815 |
redo B; |
| 4816 |
} |
| 4817 |
} |
| 4818 |
|
| 4819 |
!!!parse-error (type => 'after html:#character'); |
| 4820 |
$phase = 'main'; |
| 4821 |
## reprocess |
| 4822 |
redo B; |
| 4823 |
} elsif ($token->{type} eq 'start tag' or |
| 4824 |
$token->{type} eq 'end tag') { |
| 4825 |
!!!parse-error (type => 'after html:'.$token->{tag_name}); |
| 4826 |
$phase = 'main'; |
| 4827 |
## reprocess |
| 4828 |
redo B; |
| 4829 |
} elsif ($token->{type} eq 'end-of-file') { |
| 4830 |
## Stop parsing |
| 4831 |
last B; |
| 4832 |
} else { |
| 4833 |
die "$0: $token->{type}: Unknown token"; |
| 4834 |
} |
| 4835 |
} |
| 4836 |
} # B |
| 4837 |
|
| 4838 |
## Stop parsing # MUST |
| 4839 |
|
| 4840 |
## TODO: script stuffs |
| 4841 |
} # _tree_construct_main |
| 4842 |
|
| 4843 |
sub set_inner_html ($$$) { |
| 4844 |
my $class = shift; |
| 4845 |
my $node = shift; |
| 4846 |
my $s = \$_[0]; |
| 4847 |
my $onerror = $_[1]; |
| 4848 |
|
| 4849 |
my $nt = $node->node_type; |
| 4850 |
if ($nt == 9) { |
| 4851 |
# MUST |
| 4852 |
|
| 4853 |
## Step 1 # MUST |
| 4854 |
## TODO: If the document has an active parser, ... |
| 4855 |
## ISSUE: There is an issue in the spec. |
| 4856 |
|
| 4857 |
## Step 2 # MUST |
| 4858 |
my @cn = @{$node->child_nodes}; |
| 4859 |
for (@cn) { |
| 4860 |
$node->remove_child ($_); |
| 4861 |
} |
| 4862 |
|
| 4863 |
## Step 3, 4, 5 # MUST |
| 4864 |
$class->parse_string ($$s => $node, $onerror); |
| 4865 |
} elsif ($nt == 1) { |
| 4866 |
## TODO: If non-html element |
| 4867 |
|
| 4868 |
## NOTE: Most of this code is copied from |parse_string| |
| 4869 |
|
| 4870 |
## Step 1 # MUST |
| 4871 |
my $doc = $node->owner_document->implementation->create_document; |
| 4872 |
## TODO: Mark as HTML document |
| 4873 |
my $p = $class->new; |
| 4874 |
$p->{document} = $doc; |
| 4875 |
|
| 4876 |
## Step 9 # MUST |
| 4877 |
my $i = 0; |
| 4878 |
my $line = 1; |
| 4879 |
my $column = 0; |
| 4880 |
$p->{set_next_input_character} = sub { |
| 4881 |
my $self = shift; |
| 4882 |
$self->{next_input_character} = -1 and return if $i >= length $$s; |
| 4883 |
$self->{next_input_character} = ord substr $$s, $i++, 1; |
| 4884 |
$column++; |
| 4885 |
|
| 4886 |
if ($self->{next_input_character} == 0x000A) { # LF |
| 4887 |
$line++; |
| 4888 |
$column = 0; |
| 4889 |
} elsif ($self->{next_input_character} == 0x000D) { # CR |
| 4890 |
if ($i >= length $$s) { |
| 4891 |
# |
| 4892 |
} else { |
| 4893 |
my $next_char = ord substr $$s, $i++, 1; |
| 4894 |
if ($next_char == 0x000A) { # LF |
| 4895 |
# |
| 4896 |
} else { |
| 4897 |
push @{$self->{char}}, $next_char; |
| 4898 |
} |
| 4899 |
} |
| 4900 |
$self->{next_input_character} = 0x000A; # LF # MUST |
| 4901 |
$line++; |
| 4902 |
$column = 0; |
| 4903 |
} elsif ($self->{next_input_character} > 0x10FFFF) { |
| 4904 |
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 4905 |
} elsif ($self->{next_input_character} == 0x0000) { # NULL |
| 4906 |
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
| 4907 |
} |
| 4908 |
}; |
| 4909 |
|
| 4910 |
my $ponerror = $onerror || sub { |
| 4911 |
my (%opt) = @_; |
| 4912 |
warn "Parse error ($opt{type}) at line $opt{line} column $opt{column}\n"; |
| 4913 |
}; |
| 4914 |
$p->{parse_error} = sub { |
| 4915 |
$ponerror->(@_, line => $line, column => $column); |
| 4916 |
}; |
| 4917 |
|
| 4918 |
$p->_initialize_tokenizer; |
| 4919 |
$p->_initialize_tree_constructor; |
| 4920 |
|
| 4921 |
## Step 2 |
| 4922 |
my $node_ln = $node->local_name; |
| 4923 |
$p->{content_model_flag} = { |
| 4924 |
title => 'RCDATA', |
| 4925 |
textarea => 'RCDATA', |
| 4926 |
style => 'CDATA', |
| 4927 |
script => 'CDATA', |
| 4928 |
xmp => 'CDATA', |
| 4929 |
iframe => 'CDATA', |
| 4930 |
noembed => 'CDATA', |
| 4931 |
noframes => 'CDATA', |
| 4932 |
noscript => 'CDATA', |
| 4933 |
plaintext => 'PLAINTEXT', |
| 4934 |
}->{$node_ln} || 'PCDATA'; |
| 4935 |
## ISSUE: What is "the name of the element"? local name? |
| 4936 |
|
| 4937 |
$p->{inner_html_node} = [$node, $node_ln]; |
| 4938 |
|
| 4939 |
## Step 4 |
| 4940 |
my $root = $doc->create_element_ns |
| 4941 |
('http://www.w3.org/1999/xhtml', [undef, 'html']); |
| 4942 |
|
| 4943 |
## Step 5 # MUST |
| 4944 |
$doc->append_child ($root); |
| 4945 |
|
| 4946 |
## Step 6 # MUST |
| 4947 |
push @{$p->{open_elements}}, [$root, 'html']; |
| 4948 |
|
| 4949 |
undef $p->{head_element}; |
| 4950 |
|
| 4951 |
## Step 7 # MUST |
| 4952 |
$p->_reset_insertion_mode; |
| 4953 |
|
| 4954 |
## Step 8 # MUST |
| 4955 |
my $anode = $node; |
| 4956 |
AN: while (defined $anode) { |
| 4957 |
if ($anode->node_type == 1) { |
| 4958 |
my $nsuri = $anode->namespace_uri; |
| 4959 |
if (defined $nsuri and $nsuri eq 'http://www.w3.org/1999/xhtml') { |
| 4960 |
if ($anode->local_name eq 'form') { ## TODO: case? |
| 4961 |
$p->{form_element} = $anode; |
| 4962 |
last AN; |
| 4963 |
} |
| 4964 |
} |
| 4965 |
} |
| 4966 |
$anode = $anode->parent_node; |
| 4967 |
} # AN |
| 4968 |
|
| 4969 |
## Step 3 # MUST |
| 4970 |
## Step 10 # MUST |
| 4971 |
{ |
| 4972 |
my $self = $p; |
| 4973 |
!!!next-token; |
| 4974 |
} |
| 4975 |
$p->_tree_construction_main; |
| 4976 |
|
| 4977 |
## Step 11 # MUST |
| 4978 |
my @cn = @{$node->child_nodes}; |
| 4979 |
for (@cn) { |
| 4980 |
$node->remove_child ($_); |
| 4981 |
} |
| 4982 |
## ISSUE: mutation events? read-only? |
| 4983 |
|
| 4984 |
## Step 12 # MUST |
| 4985 |
@cn = @{$root->child_nodes}; |
| 4986 |
for (@cn) { |
| 4987 |
$node->append_child ($_); |
| 4988 |
} |
| 4989 |
## ISSUE: adopt_node? mutation events? |
| 4990 |
|
| 4991 |
$p->_terminate_tree_constructor; |
| 4992 |
} else { |
| 4993 |
die "$0: |set_inner_html| is not defined for node of type $nt"; |
| 4994 |
} |
| 4995 |
} # set_inner_html |
| 4996 |
|
| 4997 |
} # tree construction stage |
| 4998 |
|
| 4999 |
sub get_inner_html ($$$) { |
| 5000 |
my (undef, $node, $on_error) = @_; |
| 5001 |
|
| 5002 |
## Step 1 |
| 5003 |
my $s = ''; |
| 5004 |
|
| 5005 |
my $in_cdata; |
| 5006 |
my $parent = $node; |
| 5007 |
while (defined $parent) { |
| 5008 |
if ($parent->node_type == 1 and |
| 5009 |
$parent->namespace_uri eq 'http://www.w3.org/1999/xhtml' and |
| 5010 |
{ |
| 5011 |
style => 1, script => 1, xmp => 1, iframe => 1, |
| 5012 |
noembed => 1, noframes => 1, noscript => 1, |
| 5013 |
}->{$parent->local_name}) { ## TODO: case thingy |
| 5014 |
$in_cdata = 1; |
| 5015 |
} |
| 5016 |
$parent = $parent->parent_node; |
| 5017 |
} |
| 5018 |
|
| 5019 |
## Step 2 |
| 5020 |
my @node = @{$node->child_nodes}; |
| 5021 |
C: while (@node) { |
| 5022 |
my $child = shift @node; |
| 5023 |
unless (ref $child) { |
| 5024 |
if ($child eq 'cdata-out') { |
| 5025 |
$in_cdata = 0; |
| 5026 |
} else { |
| 5027 |
$s .= $child; # end tag |
| 5028 |
} |
| 5029 |
next C; |
| 5030 |
} |
| 5031 |
|
| 5032 |
my $nt = $child->node_type; |
| 5033 |
if ($nt == 1) { # Element |
| 5034 |
my $tag_name = lc $child->tag_name; ## ISSUE: Definition of "lowercase" |
| 5035 |
$s .= '<' . $tag_name; |
| 5036 |
|
| 5037 |
## ISSUE: Non-html elements |
| 5038 |
|
| 5039 |
my @attrs = @{$child->attributes}; # sort order MUST be stable |
| 5040 |
for my $attr (@attrs) { # order is implementation dependent |
| 5041 |
my $attr_name = lc $attr->name; ## ISSUE: Definition of "lowercase" |
| 5042 |
$s .= ' ' . $attr_name . '="'; |
| 5043 |
my $attr_value = $attr->value; |
| 5044 |
## escape |
| 5045 |
$attr_value =~ s/&/&/g; |
| 5046 |
$attr_value =~ s/</</g; |
| 5047 |
$attr_value =~ s/>/>/g; |
| 5048 |
$attr_value =~ s/"/"/g; |
| 5049 |
$s .= $attr_value . '"'; |
| 5050 |
} |
| 5051 |
$s .= '>'; |
| 5052 |
|
| 5053 |
next C if { |
| 5054 |
area => 1, base => 1, basefont => 1, bgsound => 1, |
| 5055 |
br => 1, col => 1, embed => 1, frame => 1, hr => 1, |
| 5056 |
img => 1, input => 1, link => 1, meta => 1, param => 1, |
| 5057 |
spacer => 1, wbr => 1, |
| 5058 |
}->{$tag_name}; |
| 5059 |
|
| 5060 |
if (not $in_cdata and { |
| 5061 |
style => 1, script => 1, xmp => 1, iframe => 1, |
| 5062 |
noembed => 1, noframes => 1, noscript => 1, |
| 5063 |
}->{$tag_name}) { |
| 5064 |
unshift @node, 'cdata-out'; |
| 5065 |
$in_cdata = 1; |
| 5066 |
} |
| 5067 |
|
| 5068 |
unshift @node, @{$child->child_nodes}, '</' . $tag_name . '>'; |
| 5069 |
} elsif ($nt == 3 or $nt == 4) { |
| 5070 |
if ($in_cdata) { |
| 5071 |
$s .= $child->data; |
| 5072 |
} else { |
| 5073 |
my $value = $child->data; |
| 5074 |
$value =~ s/&/&/g; |
| 5075 |
$value =~ s/</</g; |
| 5076 |
$value =~ s/>/>/g; |
| 5077 |
$value =~ s/"/"/g; |
| 5078 |
$s .= $value; |
| 5079 |
} |
| 5080 |
} elsif ($nt == 8) { |
| 5081 |
$s .= '<!--' . $child->data . '-->'; |
| 5082 |
} elsif ($nt == 10) { |
| 5083 |
$s .= '<!DOCTYPE ' . $child->name . '>'; |
| 5084 |
} elsif ($nt == 5) { # entrefs |
| 5085 |
push @node, @{$child->child_nodes}; |
| 5086 |
} else { |
| 5087 |
$on_error->($child) if defined $on_error; |
| 5088 |
} |
| 5089 |
## ISSUE: This code does not support PIs. |
| 5090 |
} # C |
| 5091 |
|
| 5092 |
## Step 3 |
| 5093 |
return \$s; |
| 5094 |
} # get_inner_html |
| 5095 |
|
| 5096 |
1; |
| 5097 |
# $Date: 2007/06/23 02:41:51 $ |