1 |
wakaba |
1.1 |
package What::HTML; |
2 |
|
|
use strict; |
3 |
wakaba |
1.9 |
our $VERSION=do{my @r=(q$Revision: 1.8 $=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
4 |
wakaba |
1.1 |
|
5 |
wakaba |
1.8 |
## This is an early version of an HTML parser. |
6 |
wakaba |
1.1 |
|
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 |
wakaba |
1.5 |
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 |
|
|
}; |
282 |
|
|
|
283 |
wakaba |
1.2 |
my $special_category = { |
284 |
|
|
address => 1, area => 1, base => 1, basefont => 1, bgsound => 1, |
285 |
|
|
blockquote => 1, body => 1, br => 1, center => 1, col => 1, colgroup => 1, |
286 |
|
|
dd => 1, dir => 1, div => 1, dl => 1, dt => 1, embed => 1, fieldset => 1, |
287 |
|
|
form => 1, frame => 1, frameset => 1, h1 => 1, h2 => 1, h3 => 1, |
288 |
|
|
h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, iframe => 1, image => 1, |
289 |
|
|
img => 1, input => 1, isindex => 1, li => 1, link => 1, listing => 1, |
290 |
|
|
menu => 1, meta => 1, noembed => 1, noframes => 1, noscript => 1, |
291 |
|
|
ol => 1, optgroup => 1, option => 1, p => 1, param => 1, plaintext => 1, |
292 |
|
|
pre => 1, script => 1, select => 1, spacer => 1, style => 1, tbody => 1, |
293 |
|
|
textarea => 1, tfoot => 1, thead => 1, title => 1, tr => 1, ul => 1, wbr => 1, |
294 |
|
|
}; |
295 |
|
|
my $scoping_category = { |
296 |
|
|
button => 1, caption => 1, html => 1, marquee => 1, object => 1, |
297 |
|
|
table => 1, td => 1, th => 1, |
298 |
|
|
}; |
299 |
|
|
my $formatting_category = { |
300 |
|
|
a => 1, b => 1, big => 1, em => 1, font => 1, i => 1, nobr => 1, |
301 |
|
|
s => 1, small => 1, strile => 1, strong => 1, tt => 1, u => 1, |
302 |
|
|
}; |
303 |
|
|
# $phrasing_category: all other elements |
304 |
|
|
|
305 |
wakaba |
1.8 |
sub parse_string ($$$;$) { |
306 |
|
|
my $self = shift->new; |
307 |
|
|
my $s = \$_[0]; |
308 |
|
|
$self->{document} = $_[1]; |
309 |
|
|
|
310 |
|
|
my $i; |
311 |
|
|
my $i = 0; |
312 |
|
|
$self->{set_next_input_character} = sub { |
313 |
|
|
my $self = shift; |
314 |
|
|
$self->{next_input_character} = -1 and return if $i >= length $$s; |
315 |
|
|
$self->{next_input_character} = ord substr $$s, $i++, 1; |
316 |
|
|
|
317 |
|
|
if ($self->{next_input_character} == 0x000D) { # CR |
318 |
|
|
if ($i >= length $$s) { |
319 |
|
|
# |
320 |
|
|
} else { |
321 |
|
|
my $next_char = ord substr $$s, $i++, 1; |
322 |
|
|
if ($next_char == 0x000A) { # LF |
323 |
|
|
# |
324 |
|
|
} else { |
325 |
|
|
push @{$self->{char}}, $next_char; |
326 |
|
|
} |
327 |
|
|
} |
328 |
|
|
$self->{next_input_character} = 0x000A; # LF # MUST |
329 |
|
|
} elsif ($self->{next_input_character} > 0x10FFFF) { |
330 |
|
|
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
331 |
|
|
} elsif ($self->{next_input_character} == 0x0000) { # NULL |
332 |
|
|
$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST |
333 |
|
|
} |
334 |
|
|
}; |
335 |
|
|
|
336 |
|
|
$self->{parse_error} = $_[2] || sub { |
337 |
|
|
warn "Parse error at character $i\n"; ## TODO: Report (line, column) pair |
338 |
|
|
}; |
339 |
|
|
|
340 |
|
|
$self->_initialize_tokenizer; |
341 |
|
|
$self->_initialize_tree_constructor; |
342 |
|
|
$self->_construct_tree; |
343 |
|
|
$self->_terminate_tree_constructor; |
344 |
|
|
|
345 |
|
|
return $self->{document}; |
346 |
|
|
} # parse_string |
347 |
|
|
|
348 |
wakaba |
1.1 |
sub new ($) { |
349 |
|
|
my $class = shift; |
350 |
|
|
my $self = bless {}, $class; |
351 |
|
|
$self->{set_next_input_character} = sub { |
352 |
|
|
$self->{next_input_character} = -1; |
353 |
|
|
}; |
354 |
|
|
$self->{parse_error} = sub { |
355 |
|
|
# |
356 |
|
|
}; |
357 |
|
|
return $self; |
358 |
|
|
} # new |
359 |
|
|
|
360 |
|
|
## Implementations MUST act as if state machine in the spec |
361 |
|
|
|
362 |
|
|
sub _initialize_tokenizer ($) { |
363 |
|
|
my $self = shift; |
364 |
|
|
$self->{state} = 'data'; # MUST |
365 |
|
|
$self->{content_model_flag} = 'PCDATA'; # be |
366 |
|
|
undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE |
367 |
|
|
undef $self->{current_attribute}; |
368 |
|
|
undef $self->{last_emitted_start_tag_name}; |
369 |
|
|
undef $self->{last_attribute_value_state}; |
370 |
|
|
$self->{char} = []; |
371 |
|
|
# $self->{next_input_character} |
372 |
|
|
!!!next-input-character; |
373 |
|
|
$self->{token} = []; |
374 |
|
|
} # _initialize_tokenizer |
375 |
|
|
|
376 |
|
|
## A token has: |
377 |
|
|
## ->{type} eq 'DOCTYPE', 'start tag', 'end tag', 'comment', |
378 |
|
|
## 'character', or 'end-of-file' |
379 |
|
|
## ->{name} (DOCTYPE, start tag (tagname), end tag (tagname)) |
380 |
|
|
## ISSUE: the spec need s/tagname/tag name/ |
381 |
|
|
## ->{error} == 1 or 0 (DOCTYPE) |
382 |
|
|
## ->{attributes} isa HASH (start tag, end tag) |
383 |
|
|
## ->{data} (comment, character) |
384 |
|
|
|
385 |
|
|
## Macros |
386 |
|
|
## Macros MUST be preceded by three EXCLAMATION MARKs. |
387 |
|
|
## emit ($token) |
388 |
|
|
## Emits the specified token. |
389 |
|
|
|
390 |
|
|
## Emitted token MUST immediately be handled by the tree construction state. |
391 |
|
|
|
392 |
|
|
## Before each step, UA MAY check to see if either one of the scripts in |
393 |
|
|
## "list of scripts that will execute as soon as possible" or the first |
394 |
|
|
## script in the "list of scripts that will execute asynchronously", |
395 |
|
|
## has completed loading. If one has, then it MUST be executed |
396 |
|
|
## and removed from the list. |
397 |
|
|
|
398 |
|
|
sub _get_next_token ($) { |
399 |
|
|
my $self = shift; |
400 |
|
|
if (@{$self->{token}}) { |
401 |
|
|
return shift @{$self->{token}}; |
402 |
|
|
} |
403 |
|
|
|
404 |
|
|
A: { |
405 |
|
|
if ($self->{state} eq 'data') { |
406 |
|
|
if ($self->{next_input_character} == 0x0026) { # & |
407 |
|
|
if ($self->{content_model_flag} eq 'PCDATA' or |
408 |
|
|
$self->{content_model_flag} eq 'RCDATA') { |
409 |
|
|
$self->{state} = 'entity data'; |
410 |
|
|
!!!next-input-character; |
411 |
|
|
redo A; |
412 |
|
|
} else { |
413 |
|
|
# |
414 |
|
|
} |
415 |
|
|
} elsif ($self->{next_input_character} == 0x003C) { # < |
416 |
|
|
if ($self->{content_model_flag} ne 'PLAINTEXT') { |
417 |
|
|
$self->{state} = 'tag open'; |
418 |
|
|
!!!next-input-character; |
419 |
|
|
redo A; |
420 |
|
|
} else { |
421 |
|
|
# |
422 |
|
|
} |
423 |
|
|
} elsif ($self->{next_input_character} == -1) { |
424 |
|
|
!!!emit ({type => 'end-of-file'}); |
425 |
|
|
last A; ## TODO: ok? |
426 |
|
|
} |
427 |
|
|
# Anything else |
428 |
|
|
my $token = {type => 'character', |
429 |
|
|
data => chr $self->{next_input_character}}; |
430 |
|
|
## Stay in the data state |
431 |
|
|
!!!next-input-character; |
432 |
|
|
|
433 |
|
|
!!!emit ($token); |
434 |
|
|
|
435 |
|
|
redo A; |
436 |
|
|
} elsif ($self->{state} eq 'entity data') { |
437 |
|
|
## (cannot happen in CDATA state) |
438 |
|
|
|
439 |
|
|
my $token = $self->_tokenize_attempt_to_consume_an_entity; |
440 |
|
|
|
441 |
|
|
$self->{state} = 'data'; |
442 |
|
|
# next-input-character is already done |
443 |
|
|
|
444 |
|
|
unless (defined $token) { |
445 |
|
|
!!!emit ({type => 'character', data => '&'}); |
446 |
|
|
} else { |
447 |
|
|
!!!emit ($token); |
448 |
|
|
} |
449 |
|
|
|
450 |
|
|
redo A; |
451 |
|
|
} elsif ($self->{state} eq 'tag open') { |
452 |
|
|
if ($self->{content_model_flag} eq 'RCDATA' or |
453 |
|
|
$self->{content_model_flag} eq 'CDATA') { |
454 |
|
|
if ($self->{next_input_character} == 0x002F) { # / |
455 |
|
|
!!!next-input-character; |
456 |
|
|
$self->{state} = 'close tag open'; |
457 |
|
|
redo A; |
458 |
|
|
} else { |
459 |
|
|
## reconsume |
460 |
|
|
$self->{state} = 'data'; |
461 |
|
|
|
462 |
wakaba |
1.7 |
!!!emit ({type => 'character', data => '<'}); |
463 |
wakaba |
1.1 |
|
464 |
|
|
redo A; |
465 |
|
|
} |
466 |
|
|
} elsif ($self->{content_model_flag} eq 'PCDATA') { |
467 |
|
|
if ($self->{next_input_character} == 0x0021) { # ! |
468 |
|
|
$self->{state} = 'markup declaration open'; |
469 |
|
|
!!!next-input-character; |
470 |
|
|
redo A; |
471 |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
472 |
|
|
$self->{state} = 'close tag open'; |
473 |
|
|
!!!next-input-character; |
474 |
|
|
redo A; |
475 |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
476 |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
477 |
|
|
$self->{current_token} |
478 |
|
|
= {type => 'start tag', |
479 |
|
|
tag_name => chr ($self->{next_input_character} + 0x0020)}; |
480 |
|
|
$self->{state} = 'tag name'; |
481 |
|
|
!!!next-input-character; |
482 |
|
|
redo A; |
483 |
|
|
} elsif (0x0061 <= $self->{next_input_character} and |
484 |
|
|
$self->{next_input_character} <= 0x007A) { # a..z |
485 |
|
|
$self->{current_token} = {type => 'start tag', |
486 |
|
|
tag_name => chr ($self->{next_input_character})}; |
487 |
|
|
$self->{state} = 'tag name'; |
488 |
|
|
!!!next-input-character; |
489 |
|
|
redo A; |
490 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
491 |
|
|
!!!parse-error; |
492 |
|
|
$self->{state} = 'data'; |
493 |
|
|
!!!next-input-character; |
494 |
|
|
|
495 |
wakaba |
1.3 |
!!!emit ({type => 'character', data => '<>'}); |
496 |
wakaba |
1.1 |
|
497 |
|
|
redo A; |
498 |
|
|
} elsif ($self->{next_input_character} == 0x003F) { # ? |
499 |
|
|
!!!parse-error; |
500 |
|
|
$self->{state} = 'bogus comment'; |
501 |
|
|
## $self->{next_input_character} is intentionally left as is |
502 |
|
|
redo A; |
503 |
|
|
} else { |
504 |
|
|
!!!parse-error; |
505 |
|
|
$self->{state} = 'data'; |
506 |
|
|
## reconsume |
507 |
|
|
|
508 |
|
|
!!!emit ({type => 'character', data => '<'}); |
509 |
|
|
|
510 |
|
|
redo A; |
511 |
|
|
} |
512 |
|
|
} else { |
513 |
|
|
die "$0: $self->{content_model_flag}: Unknown content model flag"; |
514 |
|
|
} |
515 |
|
|
} elsif ($self->{state} eq 'close tag open') { |
516 |
|
|
if ($self->{content_model_flag} eq 'RCDATA' or |
517 |
|
|
$self->{content_model_flag} eq 'CDATA') { |
518 |
|
|
my @next_char; |
519 |
|
|
TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) { |
520 |
|
|
push @next_char, $self->{next_input_character}; |
521 |
|
|
my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1); |
522 |
|
|
my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c; |
523 |
|
|
if ($self->{next_input_character} == $c or $self->{next_input_character} == $C) { |
524 |
|
|
!!!next-input-character; |
525 |
|
|
next TAGNAME; |
526 |
|
|
} else { |
527 |
|
|
!!!parse-error; |
528 |
|
|
$self->{next_input_character} = shift @next_char; # reconsume |
529 |
|
|
!!!back-next-input-character (@next_char); |
530 |
|
|
$self->{state} = 'data'; |
531 |
|
|
|
532 |
|
|
!!!emit ({type => 'character', data => '</'}); |
533 |
|
|
|
534 |
|
|
redo A; |
535 |
|
|
} |
536 |
|
|
} |
537 |
wakaba |
1.2 |
push @next_char, $self->{next_input_character}; |
538 |
wakaba |
1.1 |
|
539 |
wakaba |
1.2 |
unless ($self->{next_input_character} == 0x0009 or # HT |
540 |
|
|
$self->{next_input_character} == 0x000A or # LF |
541 |
|
|
$self->{next_input_character} == 0x000B or # VT |
542 |
|
|
$self->{next_input_character} == 0x000C or # FF |
543 |
|
|
$self->{next_input_character} == 0x0020 or # SP |
544 |
|
|
$self->{next_input_character} == 0x003E or # > |
545 |
|
|
$self->{next_input_character} == 0x002F or # / |
546 |
|
|
$self->{next_input_character} == 0x003C or # < |
547 |
wakaba |
1.1 |
$self->{next_input_character} == -1) { |
548 |
|
|
!!!parse-error; |
549 |
|
|
$self->{next_input_character} = shift @next_char; # reconsume |
550 |
|
|
!!!back-next-input-character (@next_char); |
551 |
|
|
$self->{state} = 'data'; |
552 |
|
|
|
553 |
|
|
!!!emit ({type => 'character', data => '</'}); |
554 |
|
|
|
555 |
|
|
redo A; |
556 |
|
|
} else { |
557 |
|
|
$self->{next_input_character} = shift @next_char; |
558 |
|
|
!!!back-next-input-character (@next_char); |
559 |
|
|
# and consume... |
560 |
|
|
} |
561 |
|
|
} |
562 |
|
|
|
563 |
|
|
if (0x0041 <= $self->{next_input_character} and |
564 |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
565 |
|
|
$self->{current_token} = {type => 'end tag', |
566 |
|
|
tag_name => chr ($self->{next_input_character} + 0x0020)}; |
567 |
|
|
$self->{state} = 'tag name'; |
568 |
|
|
!!!next-input-character; |
569 |
|
|
redo A; |
570 |
|
|
} elsif (0x0061 <= $self->{next_input_character} and |
571 |
|
|
$self->{next_input_character} <= 0x007A) { # a..z |
572 |
|
|
$self->{current_token} = {type => 'end tag', |
573 |
|
|
tag_name => chr ($self->{next_input_character})}; |
574 |
|
|
$self->{state} = 'tag name'; |
575 |
|
|
!!!next-input-character; |
576 |
|
|
redo A; |
577 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
578 |
|
|
!!!parse-error; |
579 |
|
|
$self->{state} = 'data'; |
580 |
|
|
!!!next-input-character; |
581 |
|
|
redo A; |
582 |
|
|
} elsif ($self->{next_input_character} == -1) { |
583 |
|
|
!!!parse-error; |
584 |
|
|
$self->{state} = 'data'; |
585 |
|
|
# reconsume |
586 |
|
|
|
587 |
|
|
!!!emit ({type => 'character', data => '</'}); |
588 |
|
|
|
589 |
|
|
redo A; |
590 |
|
|
} else { |
591 |
|
|
!!!parse-error; |
592 |
|
|
$self->{state} = 'bogus comment'; |
593 |
|
|
## $self->{next_input_character} is intentionally left as is |
594 |
|
|
redo A; |
595 |
|
|
} |
596 |
|
|
} elsif ($self->{state} eq 'tag name') { |
597 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
598 |
|
|
$self->{next_input_character} == 0x000A or # LF |
599 |
|
|
$self->{next_input_character} == 0x000B or # VT |
600 |
|
|
$self->{next_input_character} == 0x000C or # FF |
601 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
602 |
|
|
$self->{state} = 'before attribute name'; |
603 |
|
|
!!!next-input-character; |
604 |
|
|
redo A; |
605 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
606 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
607 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
608 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
609 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
610 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
611 |
wakaba |
1.1 |
!!!parse-error; |
612 |
|
|
} |
613 |
|
|
} else { |
614 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
615 |
|
|
} |
616 |
|
|
$self->{state} = 'data'; |
617 |
|
|
!!!next-input-character; |
618 |
|
|
|
619 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
620 |
|
|
undef $self->{current_token}; |
621 |
|
|
|
622 |
|
|
redo A; |
623 |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
624 |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
625 |
|
|
$self->{current_token}->{tag_name} .= chr ($self->{next_input_character} + 0x0020); |
626 |
|
|
# start tag or end tag |
627 |
|
|
## Stay in this state |
628 |
|
|
!!!next-input-character; |
629 |
|
|
redo A; |
630 |
|
|
} elsif ($self->{next_input_character} == 0x003C or # < |
631 |
|
|
$self->{next_input_character} == -1) { |
632 |
|
|
!!!parse-error; |
633 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
634 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
635 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
636 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
637 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
638 |
wakaba |
1.1 |
!!!parse-error; |
639 |
|
|
} |
640 |
|
|
} else { |
641 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
642 |
|
|
} |
643 |
|
|
$self->{state} = 'data'; |
644 |
|
|
# reconsume |
645 |
|
|
|
646 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
647 |
|
|
undef $self->{current_token}; |
648 |
|
|
|
649 |
|
|
redo A; |
650 |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
651 |
|
|
!!!next-input-character; |
652 |
|
|
if ($self->{next_input_character} == 0x003E and # > |
653 |
|
|
$self->{current_token}->{type} eq 'start tag' and |
654 |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
655 |
|
|
# permitted slash |
656 |
|
|
# |
657 |
|
|
} else { |
658 |
|
|
!!!parse-error; |
659 |
|
|
} |
660 |
|
|
$self->{state} = 'before attribute name'; |
661 |
|
|
# next-input-character is already done |
662 |
|
|
redo A; |
663 |
|
|
} else { |
664 |
|
|
$self->{current_token}->{tag_name} .= chr $self->{next_input_character}; |
665 |
|
|
# start tag or end tag |
666 |
|
|
## Stay in the state |
667 |
|
|
!!!next-input-character; |
668 |
|
|
redo A; |
669 |
|
|
} |
670 |
|
|
} elsif ($self->{state} eq 'before attribute name') { |
671 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
672 |
|
|
$self->{next_input_character} == 0x000A or # LF |
673 |
|
|
$self->{next_input_character} == 0x000B or # VT |
674 |
|
|
$self->{next_input_character} == 0x000C or # FF |
675 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
676 |
|
|
## Stay in the state |
677 |
|
|
!!!next-input-character; |
678 |
|
|
redo A; |
679 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
680 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
681 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
682 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
683 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
684 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
685 |
wakaba |
1.1 |
!!!parse-error; |
686 |
|
|
} |
687 |
|
|
} else { |
688 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
689 |
|
|
} |
690 |
|
|
$self->{state} = 'data'; |
691 |
|
|
!!!next-input-character; |
692 |
|
|
|
693 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
694 |
|
|
undef $self->{current_token}; |
695 |
|
|
|
696 |
|
|
redo A; |
697 |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
698 |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
699 |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), |
700 |
|
|
value => ''}; |
701 |
|
|
$self->{state} = 'attribute name'; |
702 |
|
|
!!!next-input-character; |
703 |
|
|
redo A; |
704 |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
705 |
|
|
!!!next-input-character; |
706 |
|
|
if ($self->{next_input_character} == 0x003E and # > |
707 |
|
|
$self->{current_token}->{type} eq 'start tag' and |
708 |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
709 |
|
|
# permitted slash |
710 |
|
|
# |
711 |
|
|
} else { |
712 |
|
|
!!!parse-error; |
713 |
|
|
} |
714 |
|
|
## Stay in the state |
715 |
|
|
# next-input-character is already done |
716 |
|
|
redo A; |
717 |
|
|
} elsif ($self->{next_input_character} == 0x003C or # < |
718 |
|
|
$self->{next_input_character} == -1) { |
719 |
|
|
!!!parse-error; |
720 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
721 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
722 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
723 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
724 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
725 |
wakaba |
1.1 |
!!!parse-error; |
726 |
|
|
} |
727 |
|
|
} else { |
728 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
729 |
|
|
} |
730 |
|
|
$self->{state} = 'data'; |
731 |
|
|
# reconsume |
732 |
|
|
|
733 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
734 |
|
|
undef $self->{current_token}; |
735 |
|
|
|
736 |
|
|
redo A; |
737 |
|
|
} else { |
738 |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character}), |
739 |
|
|
value => ''}; |
740 |
|
|
$self->{state} = 'attribute name'; |
741 |
|
|
!!!next-input-character; |
742 |
|
|
redo A; |
743 |
|
|
} |
744 |
|
|
} elsif ($self->{state} eq 'attribute name') { |
745 |
|
|
my $before_leave = sub { |
746 |
wakaba |
1.2 |
if (exists $self->{current_token}->{attributes} # start tag or end tag |
747 |
wakaba |
1.1 |
->{$self->{current_attribute}->{name}}) { # MUST |
748 |
|
|
!!!parse-error; |
749 |
|
|
## Discard $self->{current_attribute} # MUST |
750 |
|
|
} else { |
751 |
wakaba |
1.2 |
$self->{current_token}->{attributes}->{$self->{current_attribute}->{name}} |
752 |
wakaba |
1.1 |
= $self->{current_attribute}; |
753 |
|
|
} |
754 |
|
|
}; # $before_leave |
755 |
|
|
|
756 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
757 |
|
|
$self->{next_input_character} == 0x000A or # LF |
758 |
|
|
$self->{next_input_character} == 0x000B or # VT |
759 |
|
|
$self->{next_input_character} == 0x000C or # FF |
760 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
761 |
|
|
$before_leave->(); |
762 |
|
|
$self->{state} = 'after attribute name'; |
763 |
|
|
!!!next-input-character; |
764 |
|
|
redo A; |
765 |
|
|
} elsif ($self->{next_input_character} == 0x003D) { # = |
766 |
|
|
$before_leave->(); |
767 |
|
|
$self->{state} = 'before attribute value'; |
768 |
|
|
!!!next-input-character; |
769 |
|
|
redo A; |
770 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
771 |
|
|
$before_leave->(); |
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 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
777 |
wakaba |
1.1 |
!!!parse-error; |
778 |
|
|
} |
779 |
|
|
} else { |
780 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
781 |
|
|
} |
782 |
|
|
$self->{state} = 'data'; |
783 |
|
|
!!!next-input-character; |
784 |
|
|
|
785 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
786 |
|
|
undef $self->{current_token}; |
787 |
|
|
|
788 |
|
|
redo A; |
789 |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
790 |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
791 |
|
|
$self->{current_attribute}->{name} .= chr ($self->{next_input_character} + 0x0020); |
792 |
|
|
## Stay in the state |
793 |
|
|
!!!next-input-character; |
794 |
|
|
redo A; |
795 |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
796 |
|
|
$before_leave->(); |
797 |
|
|
!!!next-input-character; |
798 |
|
|
if ($self->{next_input_character} == 0x003E and # > |
799 |
|
|
$self->{current_token}->{type} eq 'start tag' and |
800 |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
801 |
|
|
# permitted slash |
802 |
|
|
# |
803 |
|
|
} else { |
804 |
|
|
!!!parse-error; |
805 |
|
|
} |
806 |
|
|
$self->{state} = 'before attribute name'; |
807 |
|
|
# next-input-character is already done |
808 |
|
|
redo A; |
809 |
|
|
} elsif ($self->{next_input_character} == 0x003C or # < |
810 |
|
|
$self->{next_input_character} == -1) { |
811 |
|
|
!!!parse-error; |
812 |
|
|
$before_leave->(); |
813 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
814 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
815 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
816 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
817 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
818 |
wakaba |
1.1 |
!!!parse-error; |
819 |
|
|
} |
820 |
|
|
} else { |
821 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
822 |
|
|
} |
823 |
|
|
$self->{state} = 'data'; |
824 |
|
|
# reconsume |
825 |
|
|
|
826 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
827 |
|
|
undef $self->{current_token}; |
828 |
|
|
|
829 |
|
|
redo A; |
830 |
|
|
} else { |
831 |
|
|
$self->{current_attribute}->{name} .= chr ($self->{next_input_character}); |
832 |
|
|
## Stay in the state |
833 |
|
|
!!!next-input-character; |
834 |
|
|
redo A; |
835 |
|
|
} |
836 |
|
|
} elsif ($self->{state} eq 'after attribute name') { |
837 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
838 |
|
|
$self->{next_input_character} == 0x000A or # LF |
839 |
|
|
$self->{next_input_character} == 0x000B or # VT |
840 |
|
|
$self->{next_input_character} == 0x000C or # FF |
841 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
842 |
|
|
## Stay in the state |
843 |
|
|
!!!next-input-character; |
844 |
|
|
redo A; |
845 |
|
|
} elsif ($self->{next_input_character} == 0x003D) { # = |
846 |
|
|
$self->{state} = 'before attribute value'; |
847 |
|
|
!!!next-input-character; |
848 |
|
|
redo A; |
849 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
850 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
851 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
852 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
853 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
854 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
855 |
wakaba |
1.1 |
!!!parse-error; |
856 |
|
|
} |
857 |
|
|
} else { |
858 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
859 |
|
|
} |
860 |
|
|
$self->{state} = 'data'; |
861 |
|
|
!!!next-input-character; |
862 |
|
|
|
863 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
864 |
|
|
undef $self->{current_token}; |
865 |
|
|
|
866 |
|
|
redo A; |
867 |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
868 |
|
|
$self->{next_input_character} <= 0x005A) { # A..Z |
869 |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020), |
870 |
|
|
value => ''}; |
871 |
|
|
$self->{state} = 'attribute name'; |
872 |
|
|
!!!next-input-character; |
873 |
|
|
redo A; |
874 |
|
|
} elsif ($self->{next_input_character} == 0x002F) { # / |
875 |
|
|
!!!next-input-character; |
876 |
|
|
if ($self->{next_input_character} == 0x003E and # > |
877 |
|
|
$self->{current_token}->{type} eq 'start tag' and |
878 |
|
|
$permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) { |
879 |
|
|
# permitted slash |
880 |
|
|
# |
881 |
|
|
} else { |
882 |
|
|
!!!parse-error; |
883 |
|
|
} |
884 |
|
|
$self->{state} = 'before attribute name'; |
885 |
|
|
# next-input-character is already done |
886 |
|
|
redo A; |
887 |
|
|
} elsif ($self->{next_input_character} == 0x003C or # < |
888 |
|
|
$self->{next_input_character} == -1) { |
889 |
|
|
!!!parse-error; |
890 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
891 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
892 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
893 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
894 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
895 |
wakaba |
1.1 |
!!!parse-error; |
896 |
|
|
} |
897 |
|
|
} else { |
898 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
899 |
|
|
} |
900 |
|
|
$self->{state} = 'data'; |
901 |
|
|
# reconsume |
902 |
|
|
|
903 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
904 |
|
|
undef $self->{current_token}; |
905 |
|
|
|
906 |
|
|
redo A; |
907 |
|
|
} else { |
908 |
|
|
$self->{current_attribute} = {name => chr ($self->{next_input_character}), |
909 |
|
|
value => ''}; |
910 |
|
|
$self->{state} = 'attribute name'; |
911 |
|
|
!!!next-input-character; |
912 |
|
|
redo A; |
913 |
|
|
} |
914 |
|
|
} elsif ($self->{state} eq 'before attribute value') { |
915 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
916 |
|
|
$self->{next_input_character} == 0x000A or # LF |
917 |
|
|
$self->{next_input_character} == 0x000B or # VT |
918 |
|
|
$self->{next_input_character} == 0x000C or # FF |
919 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
920 |
|
|
## Stay in the state |
921 |
|
|
!!!next-input-character; |
922 |
|
|
redo A; |
923 |
|
|
} elsif ($self->{next_input_character} == 0x0022) { # " |
924 |
|
|
$self->{state} = 'attribute value (double-quoted)'; |
925 |
|
|
!!!next-input-character; |
926 |
|
|
redo A; |
927 |
|
|
} elsif ($self->{next_input_character} == 0x0026) { # & |
928 |
|
|
$self->{state} = 'attribute value (unquoted)'; |
929 |
|
|
## reconsume |
930 |
|
|
redo A; |
931 |
|
|
} elsif ($self->{next_input_character} == 0x0027) { # ' |
932 |
|
|
$self->{state} = 'attribute value (single-quoted)'; |
933 |
|
|
!!!next-input-character; |
934 |
|
|
redo A; |
935 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
936 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
937 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
938 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
939 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
940 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
941 |
wakaba |
1.1 |
!!!parse-error; |
942 |
|
|
} |
943 |
|
|
} else { |
944 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
945 |
|
|
} |
946 |
|
|
$self->{state} = 'data'; |
947 |
|
|
!!!next-input-character; |
948 |
|
|
|
949 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
950 |
|
|
undef $self->{current_token}; |
951 |
|
|
|
952 |
|
|
redo A; |
953 |
|
|
} elsif ($self->{next_input_character} == 0x003C or # < |
954 |
|
|
$self->{next_input_character} == -1) { |
955 |
|
|
!!!parse-error; |
956 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
957 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
958 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
959 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
960 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
961 |
wakaba |
1.1 |
!!!parse-error; |
962 |
|
|
} |
963 |
|
|
} else { |
964 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
965 |
|
|
} |
966 |
|
|
$self->{state} = 'data'; |
967 |
|
|
## reconsume |
968 |
|
|
|
969 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
970 |
|
|
undef $self->{current_token}; |
971 |
|
|
|
972 |
|
|
redo A; |
973 |
|
|
} else { |
974 |
|
|
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
975 |
|
|
$self->{state} = 'attribute value (unquoted)'; |
976 |
|
|
!!!next-input-character; |
977 |
|
|
redo A; |
978 |
|
|
} |
979 |
|
|
} elsif ($self->{state} eq 'attribute value (double-quoted)') { |
980 |
|
|
if ($self->{next_input_character} == 0x0022) { # " |
981 |
|
|
$self->{state} = 'before attribute name'; |
982 |
|
|
!!!next-input-character; |
983 |
|
|
redo A; |
984 |
|
|
} elsif ($self->{next_input_character} == 0x0026) { # & |
985 |
|
|
$self->{last_attribute_value_state} = 'attribute value (double-quoted)'; |
986 |
|
|
$self->{state} = 'entity in attribute value'; |
987 |
|
|
!!!next-input-character; |
988 |
|
|
redo A; |
989 |
|
|
} elsif ($self->{next_input_character} == -1) { |
990 |
|
|
!!!parse-error; |
991 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
992 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
993 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
994 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
995 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
996 |
wakaba |
1.1 |
!!!parse-error; |
997 |
|
|
} |
998 |
|
|
} else { |
999 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
1000 |
|
|
} |
1001 |
|
|
$self->{state} = 'data'; |
1002 |
|
|
## reconsume |
1003 |
|
|
|
1004 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
1005 |
|
|
undef $self->{current_token}; |
1006 |
|
|
|
1007 |
|
|
redo A; |
1008 |
|
|
} else { |
1009 |
|
|
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
1010 |
|
|
## Stay in the state |
1011 |
|
|
!!!next-input-character; |
1012 |
|
|
redo A; |
1013 |
|
|
} |
1014 |
|
|
} elsif ($self->{state} eq 'attribute value (single-quoted)') { |
1015 |
|
|
if ($self->{next_input_character} == 0x0027) { # ' |
1016 |
|
|
$self->{state} = 'before attribute name'; |
1017 |
|
|
!!!next-input-character; |
1018 |
|
|
redo A; |
1019 |
|
|
} elsif ($self->{next_input_character} == 0x0026) { # & |
1020 |
|
|
$self->{last_attribute_value_state} = 'attribute value (single-quoted)'; |
1021 |
|
|
$self->{state} = 'entity in attribute value'; |
1022 |
|
|
!!!next-input-character; |
1023 |
|
|
redo A; |
1024 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1025 |
|
|
!!!parse-error; |
1026 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
1027 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
1028 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
1029 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
1030 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
1031 |
wakaba |
1.1 |
!!!parse-error; |
1032 |
|
|
} |
1033 |
|
|
} else { |
1034 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
1035 |
|
|
} |
1036 |
|
|
$self->{state} = 'data'; |
1037 |
|
|
## reconsume |
1038 |
|
|
|
1039 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
1040 |
|
|
undef $self->{current_token}; |
1041 |
|
|
|
1042 |
|
|
redo A; |
1043 |
|
|
} else { |
1044 |
|
|
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
1045 |
|
|
## Stay in the state |
1046 |
|
|
!!!next-input-character; |
1047 |
|
|
redo A; |
1048 |
|
|
} |
1049 |
|
|
} elsif ($self->{state} eq 'attribute value (unquoted)') { |
1050 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
1051 |
|
|
$self->{next_input_character} == 0x000A or # LF |
1052 |
|
|
$self->{next_input_character} == 0x000B or # HT |
1053 |
|
|
$self->{next_input_character} == 0x000C or # FF |
1054 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
1055 |
|
|
$self->{state} = 'before attribute name'; |
1056 |
|
|
!!!next-input-character; |
1057 |
|
|
redo A; |
1058 |
|
|
} elsif ($self->{next_input_character} == 0x0026) { # & |
1059 |
|
|
$self->{last_attribute_value_state} = 'attribute value (unquoted)'; |
1060 |
|
|
$self->{state} = 'entity in attribute value'; |
1061 |
|
|
!!!next-input-character; |
1062 |
|
|
redo A; |
1063 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
1064 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
1065 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
1066 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
1067 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
1068 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
1069 |
wakaba |
1.1 |
!!!parse-error; |
1070 |
|
|
} |
1071 |
|
|
} else { |
1072 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
1073 |
|
|
} |
1074 |
|
|
$self->{state} = 'data'; |
1075 |
|
|
!!!next-input-character; |
1076 |
|
|
|
1077 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
1078 |
|
|
undef $self->{current_token}; |
1079 |
|
|
|
1080 |
|
|
redo A; |
1081 |
|
|
} elsif ($self->{next_input_character} == 0x003C or # < |
1082 |
|
|
$self->{next_input_character} == -1) { |
1083 |
|
|
!!!parse-error; |
1084 |
|
|
if ($self->{current_token}->{type} eq 'start tag') { |
1085 |
|
|
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
1086 |
|
|
} elsif ($self->{current_token}->{type} eq 'end tag') { |
1087 |
|
|
$self->{content_model_flag} = 'PCDATA'; # MUST |
1088 |
wakaba |
1.2 |
if ($self->{current_token}->{attributes}) { |
1089 |
wakaba |
1.1 |
!!!parse-error; |
1090 |
|
|
} |
1091 |
|
|
} else { |
1092 |
|
|
die "$0: $self->{current_token}->{type}: Unknown token type"; |
1093 |
|
|
} |
1094 |
|
|
$self->{state} = 'data'; |
1095 |
|
|
## reconsume |
1096 |
|
|
|
1097 |
|
|
!!!emit ($self->{current_token}); # start tag or end tag |
1098 |
|
|
undef $self->{current_token}; |
1099 |
|
|
|
1100 |
|
|
redo A; |
1101 |
|
|
} else { |
1102 |
|
|
$self->{current_attribute}->{value} .= chr ($self->{next_input_character}); |
1103 |
|
|
## Stay in the state |
1104 |
|
|
!!!next-input-character; |
1105 |
|
|
redo A; |
1106 |
|
|
} |
1107 |
|
|
} elsif ($self->{state} eq 'entity in attribute value') { |
1108 |
|
|
my $token = $self->_tokenize_attempt_to_consume_an_entity; |
1109 |
|
|
|
1110 |
|
|
unless (defined $token) { |
1111 |
|
|
$self->{current_attribute}->{value} .= '&'; |
1112 |
|
|
} else { |
1113 |
|
|
$self->{current_attribute}->{value} .= $token->{data}; |
1114 |
|
|
## ISSUE: spec says "append the returned character token to the current attribute's value" |
1115 |
|
|
} |
1116 |
|
|
|
1117 |
|
|
$self->{state} = $self->{last_attribute_value_state}; |
1118 |
|
|
# next-input-character is already done |
1119 |
|
|
redo A; |
1120 |
|
|
} elsif ($self->{state} eq 'bogus comment') { |
1121 |
|
|
## (only happen if PCDATA state) |
1122 |
|
|
|
1123 |
|
|
my $token = {type => 'comment', data => ''}; |
1124 |
|
|
|
1125 |
|
|
BC: { |
1126 |
|
|
if ($self->{next_input_character} == 0x003E) { # > |
1127 |
|
|
$self->{state} = 'data'; |
1128 |
|
|
!!!next-input-character; |
1129 |
|
|
|
1130 |
|
|
!!!emit ($token); |
1131 |
|
|
|
1132 |
|
|
redo A; |
1133 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1134 |
|
|
$self->{state} = 'data'; |
1135 |
|
|
## reconsume |
1136 |
|
|
|
1137 |
|
|
!!!emit ($token); |
1138 |
|
|
|
1139 |
|
|
redo A; |
1140 |
|
|
} else { |
1141 |
|
|
$token->{data} .= chr ($self->{next_input_character}); |
1142 |
|
|
!!!next-input-character; |
1143 |
|
|
redo BC; |
1144 |
|
|
} |
1145 |
|
|
} # BC |
1146 |
|
|
} elsif ($self->{state} eq 'markup declaration open') { |
1147 |
|
|
## (only happen if PCDATA state) |
1148 |
|
|
|
1149 |
|
|
my @next_char; |
1150 |
|
|
push @next_char, $self->{next_input_character}; |
1151 |
|
|
|
1152 |
|
|
if ($self->{next_input_character} == 0x002D) { # - |
1153 |
|
|
!!!next-input-character; |
1154 |
|
|
push @next_char, $self->{next_input_character}; |
1155 |
|
|
if ($self->{next_input_character} == 0x002D) { # - |
1156 |
|
|
$self->{current_token} = {type => 'comment', data => ''}; |
1157 |
|
|
$self->{state} = 'comment'; |
1158 |
|
|
!!!next-input-character; |
1159 |
|
|
redo A; |
1160 |
|
|
} |
1161 |
|
|
} elsif ($self->{next_input_character} == 0x0044 or # D |
1162 |
|
|
$self->{next_input_character} == 0x0064) { # d |
1163 |
|
|
!!!next-input-character; |
1164 |
|
|
push @next_char, $self->{next_input_character}; |
1165 |
|
|
if ($self->{next_input_character} == 0x004F or # O |
1166 |
|
|
$self->{next_input_character} == 0x006F) { # o |
1167 |
|
|
!!!next-input-character; |
1168 |
|
|
push @next_char, $self->{next_input_character}; |
1169 |
|
|
if ($self->{next_input_character} == 0x0043 or # C |
1170 |
|
|
$self->{next_input_character} == 0x0063) { # c |
1171 |
|
|
!!!next-input-character; |
1172 |
|
|
push @next_char, $self->{next_input_character}; |
1173 |
|
|
if ($self->{next_input_character} == 0x0054 or # T |
1174 |
|
|
$self->{next_input_character} == 0x0074) { # t |
1175 |
|
|
!!!next-input-character; |
1176 |
|
|
push @next_char, $self->{next_input_character}; |
1177 |
|
|
if ($self->{next_input_character} == 0x0059 or # Y |
1178 |
|
|
$self->{next_input_character} == 0x0079) { # y |
1179 |
|
|
!!!next-input-character; |
1180 |
|
|
push @next_char, $self->{next_input_character}; |
1181 |
|
|
if ($self->{next_input_character} == 0x0050 or # P |
1182 |
|
|
$self->{next_input_character} == 0x0070) { # p |
1183 |
|
|
!!!next-input-character; |
1184 |
|
|
push @next_char, $self->{next_input_character}; |
1185 |
|
|
if ($self->{next_input_character} == 0x0045 or # E |
1186 |
|
|
$self->{next_input_character} == 0x0065) { # e |
1187 |
|
|
## ISSUE: What a stupid code this is! |
1188 |
|
|
$self->{state} = 'DOCTYPE'; |
1189 |
|
|
!!!next-input-character; |
1190 |
|
|
redo A; |
1191 |
|
|
} |
1192 |
|
|
} |
1193 |
|
|
} |
1194 |
|
|
} |
1195 |
|
|
} |
1196 |
|
|
} |
1197 |
|
|
} |
1198 |
|
|
|
1199 |
|
|
!!!parse-error; |
1200 |
|
|
$self->{next_input_character} = shift @next_char; |
1201 |
|
|
!!!back-next-input-character (@next_char); |
1202 |
|
|
$self->{state} = 'bogus comment'; |
1203 |
|
|
redo A; |
1204 |
|
|
|
1205 |
|
|
## ISSUE: typos in spec: chacacters, is is a parse error |
1206 |
|
|
## 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? |
1207 |
|
|
} elsif ($self->{state} eq 'comment') { |
1208 |
|
|
if ($self->{next_input_character} == 0x002D) { # - |
1209 |
|
|
$self->{state} = 'comment dash'; |
1210 |
|
|
!!!next-input-character; |
1211 |
|
|
redo A; |
1212 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1213 |
|
|
!!!parse-error; |
1214 |
|
|
$self->{state} = 'data'; |
1215 |
|
|
## reconsume |
1216 |
|
|
|
1217 |
|
|
!!!emit ($self->{current_token}); # comment |
1218 |
|
|
undef $self->{current_token}; |
1219 |
|
|
|
1220 |
|
|
redo A; |
1221 |
|
|
} else { |
1222 |
|
|
$self->{current_token}->{data} .= chr ($self->{next_input_character}); # comment |
1223 |
|
|
## Stay in the state |
1224 |
|
|
!!!next-input-character; |
1225 |
|
|
redo A; |
1226 |
|
|
} |
1227 |
|
|
} elsif ($self->{state} eq 'comment dash') { |
1228 |
|
|
if ($self->{next_input_character} == 0x002D) { # - |
1229 |
|
|
$self->{state} = 'comment end'; |
1230 |
|
|
!!!next-input-character; |
1231 |
|
|
redo A; |
1232 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1233 |
|
|
!!!parse-error; |
1234 |
|
|
$self->{state} = 'data'; |
1235 |
|
|
## reconsume |
1236 |
|
|
|
1237 |
|
|
!!!emit ($self->{current_token}); # comment |
1238 |
|
|
undef $self->{current_token}; |
1239 |
|
|
|
1240 |
|
|
redo A; |
1241 |
|
|
} else { |
1242 |
|
|
$self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment |
1243 |
|
|
$self->{state} = 'comment'; |
1244 |
|
|
!!!next-input-character; |
1245 |
|
|
redo A; |
1246 |
|
|
} |
1247 |
|
|
} elsif ($self->{state} eq 'comment end') { |
1248 |
|
|
if ($self->{next_input_character} == 0x003E) { # > |
1249 |
|
|
$self->{state} = 'data'; |
1250 |
|
|
!!!next-input-character; |
1251 |
|
|
|
1252 |
|
|
!!!emit ($self->{current_token}); # comment |
1253 |
|
|
undef $self->{current_token}; |
1254 |
|
|
|
1255 |
|
|
redo A; |
1256 |
|
|
} elsif ($self->{next_input_character} == 0x002D) { # - |
1257 |
|
|
!!!parse-error; |
1258 |
|
|
$self->{current_token}->{data} .= '-'; # comment |
1259 |
|
|
## Stay in the state |
1260 |
|
|
!!!next-input-character; |
1261 |
|
|
redo A; |
1262 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1263 |
|
|
!!!parse-error; |
1264 |
|
|
$self->{state} = 'data'; |
1265 |
|
|
## reconsume |
1266 |
|
|
|
1267 |
|
|
!!!emit ($self->{current_token}); # comment |
1268 |
|
|
undef $self->{current_token}; |
1269 |
|
|
|
1270 |
|
|
redo A; |
1271 |
|
|
} else { |
1272 |
|
|
!!!parse-error; |
1273 |
|
|
$self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment |
1274 |
|
|
$self->{state} = 'comment'; |
1275 |
|
|
!!!next-input-character; |
1276 |
|
|
redo A; |
1277 |
|
|
} |
1278 |
|
|
} elsif ($self->{state} eq 'DOCTYPE') { |
1279 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
1280 |
|
|
$self->{next_input_character} == 0x000A or # LF |
1281 |
|
|
$self->{next_input_character} == 0x000B or # VT |
1282 |
|
|
$self->{next_input_character} == 0x000C or # FF |
1283 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
1284 |
|
|
$self->{state} = 'before DOCTYPE name'; |
1285 |
|
|
!!!next-input-character; |
1286 |
|
|
redo A; |
1287 |
|
|
} else { |
1288 |
|
|
!!!parse-error; |
1289 |
|
|
$self->{state} = 'before DOCTYPE name'; |
1290 |
|
|
## reconsume |
1291 |
|
|
redo A; |
1292 |
|
|
} |
1293 |
|
|
} elsif ($self->{state} eq 'before DOCTYPE name') { |
1294 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
1295 |
|
|
$self->{next_input_character} == 0x000A or # LF |
1296 |
|
|
$self->{next_input_character} == 0x000B or # VT |
1297 |
|
|
$self->{next_input_character} == 0x000C or # FF |
1298 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
1299 |
|
|
## Stay in the state |
1300 |
|
|
!!!next-input-character; |
1301 |
|
|
redo A; |
1302 |
|
|
} elsif (0x0061 <= $self->{next_input_character} and |
1303 |
|
|
$self->{next_input_character} <= 0x007A) { # a..z |
1304 |
|
|
$self->{current_token} = {type => 'DOCTYPE', |
1305 |
|
|
name => chr ($self->{next_input_character} - 0x0020), |
1306 |
|
|
error => 1}; |
1307 |
|
|
$self->{state} = 'DOCTYPE name'; |
1308 |
|
|
!!!next-input-character; |
1309 |
|
|
redo A; |
1310 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
1311 |
|
|
!!!parse-error; |
1312 |
|
|
$self->{state} = 'data'; |
1313 |
|
|
!!!next-input-character; |
1314 |
|
|
|
1315 |
|
|
!!!emit ({type => 'DOCTYPE', name => '', error => 1}); |
1316 |
|
|
|
1317 |
|
|
redo A; |
1318 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1319 |
|
|
!!!parse-error; |
1320 |
|
|
$self->{state} = 'data'; |
1321 |
|
|
## reconsume |
1322 |
|
|
|
1323 |
|
|
!!!emit ({type => 'DOCTYPE', name => '', error => 1}); |
1324 |
|
|
|
1325 |
|
|
redo A; |
1326 |
|
|
} else { |
1327 |
|
|
$self->{current_token} = {type => 'DOCTYPE', |
1328 |
|
|
name => chr ($self->{next_input_character}), |
1329 |
|
|
error => 1}; |
1330 |
|
|
$self->{state} = 'DOCTYPE name'; |
1331 |
|
|
!!!next-input-character; |
1332 |
|
|
redo A; |
1333 |
|
|
} |
1334 |
|
|
} elsif ($self->{state} eq 'DOCTYPE name') { |
1335 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
1336 |
|
|
$self->{next_input_character} == 0x000A or # LF |
1337 |
|
|
$self->{next_input_character} == 0x000B or # VT |
1338 |
|
|
$self->{next_input_character} == 0x000C or # FF |
1339 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
1340 |
|
|
$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE |
1341 |
|
|
$self->{state} = 'after DOCTYPE name'; |
1342 |
|
|
!!!next-input-character; |
1343 |
|
|
redo A; |
1344 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
1345 |
|
|
$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE |
1346 |
|
|
$self->{state} = 'data'; |
1347 |
|
|
!!!next-input-character; |
1348 |
|
|
|
1349 |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
1350 |
|
|
undef $self->{current_token}; |
1351 |
|
|
|
1352 |
|
|
redo A; |
1353 |
|
|
} elsif (0x0061 <= $self->{next_input_character} and |
1354 |
|
|
$self->{next_input_character} <= 0x007A) { # a..z |
1355 |
|
|
$self->{current_token}->{name} .= chr ($self->{next_input_character} - 0x0020); # DOCTYPE |
1356 |
|
|
#$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); |
1357 |
|
|
## Stay in the state |
1358 |
|
|
!!!next-input-character; |
1359 |
|
|
redo A; |
1360 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1361 |
|
|
!!!parse-error; |
1362 |
|
|
$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE |
1363 |
|
|
$self->{state} = 'data'; |
1364 |
|
|
## reconsume |
1365 |
|
|
|
1366 |
|
|
!!!emit ($self->{current_token}); |
1367 |
|
|
undef $self->{current_token}; |
1368 |
|
|
|
1369 |
|
|
redo A; |
1370 |
|
|
} else { |
1371 |
wakaba |
1.3 |
$self->{current_token}->{name} |
1372 |
|
|
.= chr ($self->{next_input_character}); # DOCTYPE |
1373 |
wakaba |
1.1 |
#$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); |
1374 |
|
|
## Stay in the state |
1375 |
|
|
!!!next-input-character; |
1376 |
|
|
redo A; |
1377 |
|
|
} |
1378 |
|
|
} elsif ($self->{state} eq 'after DOCTYPE name') { |
1379 |
|
|
if ($self->{next_input_character} == 0x0009 or # HT |
1380 |
|
|
$self->{next_input_character} == 0x000A or # LF |
1381 |
|
|
$self->{next_input_character} == 0x000B or # VT |
1382 |
|
|
$self->{next_input_character} == 0x000C or # FF |
1383 |
|
|
$self->{next_input_character} == 0x0020) { # SP |
1384 |
|
|
## Stay in the state |
1385 |
|
|
!!!next-input-character; |
1386 |
|
|
redo A; |
1387 |
|
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
1388 |
|
|
$self->{state} = 'data'; |
1389 |
|
|
!!!next-input-character; |
1390 |
|
|
|
1391 |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
1392 |
|
|
undef $self->{current_token}; |
1393 |
|
|
|
1394 |
|
|
redo A; |
1395 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1396 |
|
|
!!!parse-error; |
1397 |
|
|
$self->{state} = 'data'; |
1398 |
|
|
## reconsume |
1399 |
|
|
|
1400 |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
1401 |
|
|
undef $self->{current_token}; |
1402 |
|
|
|
1403 |
|
|
redo A; |
1404 |
|
|
} else { |
1405 |
|
|
!!!parse-error; |
1406 |
|
|
$self->{current_token}->{error} = 1; # DOCTYPE |
1407 |
|
|
$self->{state} = 'bogus DOCTYPE'; |
1408 |
|
|
!!!next-input-character; |
1409 |
|
|
redo A; |
1410 |
|
|
} |
1411 |
|
|
} elsif ($self->{state} eq 'bogus DOCTYPE') { |
1412 |
|
|
if ($self->{next_input_character} == 0x003E) { # > |
1413 |
|
|
$self->{state} = 'data'; |
1414 |
|
|
!!!next-input-character; |
1415 |
|
|
|
1416 |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
1417 |
|
|
undef $self->{current_token}; |
1418 |
|
|
|
1419 |
|
|
redo A; |
1420 |
|
|
} elsif ($self->{next_input_character} == -1) { |
1421 |
|
|
!!!parse-error; |
1422 |
|
|
$self->{state} = 'data'; |
1423 |
|
|
## reconsume |
1424 |
|
|
|
1425 |
|
|
!!!emit ($self->{current_token}); # DOCTYPE |
1426 |
|
|
undef $self->{current_token}; |
1427 |
|
|
|
1428 |
|
|
redo A; |
1429 |
|
|
} else { |
1430 |
|
|
## Stay in the state |
1431 |
|
|
!!!next-input-character; |
1432 |
|
|
redo A; |
1433 |
|
|
} |
1434 |
|
|
} else { |
1435 |
|
|
die "$0: $self->{state}: Unknown state"; |
1436 |
|
|
} |
1437 |
|
|
} # A |
1438 |
|
|
|
1439 |
|
|
die "$0: _get_next_token: unexpected case"; |
1440 |
|
|
} # _get_next_token |
1441 |
|
|
|
1442 |
|
|
sub _tokenize_attempt_to_consume_an_entity ($) { |
1443 |
|
|
my $self = shift; |
1444 |
|
|
|
1445 |
|
|
if ($self->{next_input_character} == 0x0023) { # # |
1446 |
|
|
!!!next-input-character; |
1447 |
|
|
my $num; |
1448 |
|
|
if ($self->{next_input_character} == 0x0078 or # x |
1449 |
|
|
$self->{next_input_character} == 0x0058) { # X |
1450 |
|
|
X: { |
1451 |
|
|
my $x_char = $self->{next_input_character}; |
1452 |
|
|
!!!next-input-character; |
1453 |
|
|
if (0x0030 <= $self->{next_input_character} and |
1454 |
|
|
$self->{next_input_character} <= 0x0039) { # 0..9 |
1455 |
|
|
$num ||= 0; |
1456 |
|
|
$num *= 0x10; |
1457 |
|
|
$num += $self->{next_input_character} - 0x0030; |
1458 |
|
|
redo X; |
1459 |
|
|
} elsif (0x0061 <= $self->{next_input_character} and |
1460 |
|
|
$self->{next_input_character} <= 0x0066) { # a..f |
1461 |
|
|
## ISSUE: the spec says U+0078, which is apparently incorrect |
1462 |
|
|
$num ||= 0; |
1463 |
|
|
$num *= 0x10; |
1464 |
|
|
$num += $self->{next_input_character} - 0x0060 + 9; |
1465 |
|
|
redo X; |
1466 |
|
|
} elsif (0x0041 <= $self->{next_input_character} and |
1467 |
|
|
$self->{next_input_character} <= 0x0046) { # A..F |
1468 |
|
|
## ISSUE: the spec says U+0058, which is apparently incorrect |
1469 |
|
|
$num ||= 0; |
1470 |
|
|
$num *= 0x10; |
1471 |
|
|
$num += $self->{next_input_character} - 0x0040 + 9; |
1472 |
|
|
redo X; |
1473 |
|
|
} elsif (not defined $num) { # no hexadecimal digit |
1474 |
|
|
!!!parse-error; |
1475 |
|
|
$self->{next_input_character} = 0x0023; # # |
1476 |
|
|
!!!back-next-input-character ($x_char); |
1477 |
wakaba |
1.5 |
return undef; |
1478 |
wakaba |
1.1 |
} elsif ($self->{next_input_character} == 0x003B) { # ; |
1479 |
|
|
!!!next-input-character; |
1480 |
|
|
} else { |
1481 |
|
|
!!!parse-error; |
1482 |
|
|
} |
1483 |
|
|
|
1484 |
|
|
## TODO: check the definition for |a valid Unicode character|. |
1485 |
|
|
if ($num > 1114111 or $num == 0) { |
1486 |
|
|
$num = 0xFFFD; # REPLACEMENT CHARACTER |
1487 |
|
|
## ISSUE: Why this is not an error? |
1488 |
|
|
} |
1489 |
|
|
|
1490 |
wakaba |
1.5 |
return {type => 'character', data => chr $num}; |
1491 |
wakaba |
1.1 |
} # X |
1492 |
wakaba |
1.4 |
} elsif (0x0030 <= $self->{next_input_character} and |
1493 |
|
|
$self->{next_input_character} <= 0x0039) { # 0..9 |
1494 |
|
|
my $code = $self->{next_input_character} - 0x0030; |
1495 |
|
|
!!!next-input-character; |
1496 |
|
|
|
1497 |
|
|
while (0x0030 <= $self->{next_input_character} and |
1498 |
|
|
$self->{next_input_character} <= 0x0039) { # 0..9 |
1499 |
|
|
$code *= 10; |
1500 |
|
|
$code += $self->{next_input_character} - 0x0030; |
1501 |
|
|
|
1502 |
|
|
!!!next-input-character; |
1503 |
|
|
} |
1504 |
wakaba |
1.1 |
|
1505 |
wakaba |
1.4 |
if ($self->{next_input_character} == 0x003B) { # ; |
1506 |
|
|
!!!next-input-character; |
1507 |
|
|
} else { |
1508 |
|
|
!!!parse-error; |
1509 |
|
|
} |
1510 |
wakaba |
1.1 |
|
1511 |
wakaba |
1.4 |
## TODO: check the definition for |a valid Unicode character|. |
1512 |
|
|
if ($code > 1114111 or $code == 0) { |
1513 |
|
|
$code = 0xFFFD; # REPLACEMENT CHARACTER |
1514 |
|
|
## ISSUE: Why this is not an error? |
1515 |
|
|
} |
1516 |
|
|
|
1517 |
wakaba |
1.5 |
return {type => 'character', data => chr $code}; |
1518 |
wakaba |
1.4 |
} else { |
1519 |
|
|
!!!parse-error; |
1520 |
|
|
!!!back-next-input-character ($self->{next_input_character}); |
1521 |
|
|
$self->{next_input_character} = 0x0023; # # |
1522 |
wakaba |
1.5 |
return undef; |
1523 |
|
|
} |
1524 |
|
|
} elsif ((0x0041 <= $self->{next_input_character} and |
1525 |
|
|
$self->{next_input_character} <= 0x005A) or |
1526 |
|
|
(0x0061 <= $self->{next_input_character} and |
1527 |
|
|
$self->{next_input_character} <= 0x007A)) { |
1528 |
|
|
my $entity_name = chr $self->{next_input_character}; |
1529 |
|
|
!!!next-input-character; |
1530 |
|
|
|
1531 |
|
|
my $value = $entity_name; |
1532 |
|
|
my $match; |
1533 |
|
|
|
1534 |
|
|
while (length $entity_name < 10 and |
1535 |
|
|
## NOTE: Some number greater than the maximum length of entity name |
1536 |
|
|
((0x0041 <= $self->{next_input_character} and |
1537 |
|
|
$self->{next_input_character} <= 0x005A) or |
1538 |
|
|
(0x0061 <= $self->{next_input_character} and |
1539 |
|
|
$self->{next_input_character} <= 0x007A) or |
1540 |
|
|
(0x0030 <= $self->{next_input_character} and |
1541 |
|
|
$self->{next_input_character} <= 0x0039))) { |
1542 |
|
|
$entity_name .= chr $self->{next_input_character}; |
1543 |
|
|
if (defined $entity_char->{$entity_name}) { |
1544 |
|
|
$value = $entity_char->{$entity_name}; |
1545 |
|
|
$match = 1; |
1546 |
|
|
} else { |
1547 |
|
|
$value .= chr $self->{next_input_character}; |
1548 |
|
|
} |
1549 |
|
|
!!!next-input-character; |
1550 |
|
|
} |
1551 |
|
|
|
1552 |
|
|
if ($match) { |
1553 |
|
|
if ($self->{next_input_character} == 0x003B) { # ; |
1554 |
|
|
!!!next-input-character; |
1555 |
|
|
} else { |
1556 |
|
|
!!!parse-error; |
1557 |
|
|
} |
1558 |
|
|
|
1559 |
|
|
return {type => 'character', data => $value}; |
1560 |
|
|
} else { |
1561 |
|
|
!!!parse-error; |
1562 |
|
|
## NOTE: No characters are consumed in the spec. |
1563 |
|
|
!!!back-token ({type => 'character', data => $value}); |
1564 |
|
|
return undef; |
1565 |
wakaba |
1.1 |
} |
1566 |
wakaba |
1.5 |
} else { |
1567 |
|
|
## no characters are consumed |
1568 |
|
|
!!!parse-error; |
1569 |
|
|
return undef; |
1570 |
|
|
} |
1571 |
wakaba |
1.1 |
} # _tokenize_attempt_to_consume_an_entity |
1572 |
|
|
|
1573 |
wakaba |
1.2 |
sub _initialize_tree_constructor ($) { |
1574 |
|
|
my $self = shift; |
1575 |
wakaba |
1.8 |
## NOTE: $self->{document} MUST be specified before this method is called |
1576 |
wakaba |
1.2 |
$self->{document}->strict_error_checking (0); |
1577 |
|
|
## TODO: Turn mutation events off # MUST |
1578 |
|
|
## TODO: Turn loose Document option (manakai extension) on |
1579 |
wakaba |
1.8 |
## TODO: Mark the Document as an HTML document # MUST |
1580 |
wakaba |
1.2 |
} # _initialize_tree_constructor |
1581 |
|
|
|
1582 |
|
|
sub _terminate_tree_constructor ($) { |
1583 |
|
|
my $self = shift; |
1584 |
|
|
$self->{document}->strict_error_checking (1); |
1585 |
|
|
## TODO: Turn mutation events on |
1586 |
|
|
} # _terminate_tree_constructor |
1587 |
|
|
|
1588 |
|
|
## ISSUE: Should append_child (for example) in script executed in tree construction stage fire mutation events? |
1589 |
|
|
|
1590 |
|
|
sub _construct_tree ($) { |
1591 |
|
|
my ($self) = @_; |
1592 |
|
|
|
1593 |
|
|
## When an interactive UA render the $self->{document} available |
1594 |
|
|
## to the user, or when it begin accepting user input, are |
1595 |
|
|
## not defined. |
1596 |
|
|
|
1597 |
|
|
## Append a character: collect it and all subsequent consecutive |
1598 |
|
|
## characters and insert one Text node whose data is concatenation |
1599 |
|
|
## of all those characters. # MUST |
1600 |
|
|
|
1601 |
|
|
my $token; |
1602 |
|
|
!!!next-token; |
1603 |
|
|
|
1604 |
|
|
my $phase = 'initial'; # MUST |
1605 |
|
|
|
1606 |
|
|
my $open_elements = []; |
1607 |
|
|
my $active_formatting_elements = []; |
1608 |
|
|
my $head_element; |
1609 |
|
|
my $form_element; |
1610 |
|
|
my $insertion_mode = 'before head'; |
1611 |
|
|
|
1612 |
|
|
my $reconstruct_active_formatting_elements = sub { # MUST |
1613 |
wakaba |
1.7 |
my $insert = shift; |
1614 |
|
|
|
1615 |
wakaba |
1.2 |
## Step 1 |
1616 |
|
|
return unless @$active_formatting_elements; |
1617 |
|
|
|
1618 |
|
|
## Step 3 |
1619 |
|
|
my $i = -1; |
1620 |
|
|
my $entry = $active_formatting_elements->[$i]; |
1621 |
|
|
|
1622 |
|
|
## Step 2 |
1623 |
|
|
return if $entry->[0] eq '#marker'; |
1624 |
|
|
for (@$open_elements) { |
1625 |
|
|
if ($entry->[0] eq $_->[0]) { |
1626 |
|
|
return; |
1627 |
|
|
} |
1628 |
|
|
} |
1629 |
|
|
|
1630 |
|
|
S4: { |
1631 |
wakaba |
1.7 |
## Step 4 |
1632 |
wakaba |
1.2 |
last S4 if $active_formatting_elements->[0]->[0] eq $entry->[0]; |
1633 |
|
|
|
1634 |
|
|
## Step 5 |
1635 |
|
|
$i--; |
1636 |
|
|
$entry = $active_formatting_elements->[$i]; |
1637 |
|
|
|
1638 |
|
|
## Step 6 |
1639 |
|
|
if ($entry->[0] eq '#marker') { |
1640 |
|
|
# |
1641 |
|
|
} else { |
1642 |
|
|
my $in_open_elements; |
1643 |
|
|
OE: for (@$open_elements) { |
1644 |
|
|
if ($entry->[0] eq $_->[0]) { |
1645 |
wakaba |
1.7 |
$in_open_elements = 1; |
1646 |
|
|
last OE; |
1647 |
|
|
} |
1648 |
wakaba |
1.2 |
} |
1649 |
|
|
if ($in_open_elements) { |
1650 |
|
|
# |
1651 |
|
|
} else { |
1652 |
|
|
redo S4; |
1653 |
|
|
} |
1654 |
|
|
} |
1655 |
|
|
|
1656 |
|
|
## Step 7 |
1657 |
|
|
$i++; |
1658 |
|
|
$entry = $active_formatting_elements->[$i]; |
1659 |
|
|
} # S4 |
1660 |
|
|
|
1661 |
|
|
S7: { |
1662 |
|
|
## Step 8 |
1663 |
wakaba |
1.7 |
my $clone = [$entry->[0]->clone_node (0), $entry->[1]]; |
1664 |
wakaba |
1.2 |
|
1665 |
|
|
## Step 9 |
1666 |
wakaba |
1.7 |
$insert->($clone->[0]); |
1667 |
|
|
push @$open_elements, $clone; |
1668 |
wakaba |
1.2 |
|
1669 |
|
|
## Step 10 |
1670 |
|
|
$active_formatting_elements->[$i] = $open_elements->[-1]; |
1671 |
wakaba |
1.7 |
|
1672 |
|
|
## Step 11 |
1673 |
|
|
unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) { |
1674 |
wakaba |
1.2 |
## Step 7' |
1675 |
|
|
$i++; |
1676 |
|
|
$entry = $active_formatting_elements->[$i]; |
1677 |
|
|
|
1678 |
|
|
redo S7; |
1679 |
|
|
} |
1680 |
|
|
} # S7 |
1681 |
|
|
}; # $reconstruct_active_formatting_elements |
1682 |
|
|
|
1683 |
|
|
my $clear_up_to_marker = sub { |
1684 |
|
|
for (reverse 0..$#$active_formatting_elements) { |
1685 |
|
|
if ($active_formatting_elements->[$_]->[0] eq '#marker') { |
1686 |
|
|
splice @$active_formatting_elements, $_; |
1687 |
|
|
return; |
1688 |
|
|
} |
1689 |
|
|
} |
1690 |
|
|
}; # $clear_up_to_marker |
1691 |
|
|
|
1692 |
|
|
my $reset_insertion_mode = sub { |
1693 |
|
|
## Step 1 |
1694 |
|
|
my $last; |
1695 |
|
|
|
1696 |
|
|
## Step 2 |
1697 |
|
|
my $i = -1; |
1698 |
|
|
my $node = $open_elements->[$i]; |
1699 |
|
|
|
1700 |
|
|
## Step 3 |
1701 |
|
|
S3: { |
1702 |
|
|
$last = 1 if $open_elements->[0]->[0] eq $node->[0]; |
1703 |
|
|
## TODO: the element whose inner_html is set is neither td nor th, then $node = the element |
1704 |
|
|
|
1705 |
|
|
## Step 4..13 |
1706 |
|
|
my $new_mode = { |
1707 |
|
|
select => 'in select', |
1708 |
|
|
td => 'in cell', |
1709 |
|
|
th => 'in cell', |
1710 |
|
|
tr => 'in row', |
1711 |
|
|
tbody => 'in table body', |
1712 |
|
|
thead => 'in table head', |
1713 |
|
|
tfoot => 'in table foot', |
1714 |
|
|
caption => 'in caption', |
1715 |
|
|
colgroup => 'in column group', |
1716 |
|
|
table => 'in table', |
1717 |
|
|
head => 'in body', # not in head! |
1718 |
|
|
body => 'in body', |
1719 |
|
|
frameset => 'in frameset', |
1720 |
|
|
}->{$node->[1]}; |
1721 |
|
|
$insertion_mode = $new_mode and return if defined $new_mode; |
1722 |
|
|
|
1723 |
|
|
## Step 14 |
1724 |
|
|
if ($node->[1] eq 'html') { |
1725 |
|
|
unless (defined $head_element) { |
1726 |
|
|
$insertion_mode = 'before head'; |
1727 |
|
|
} else { |
1728 |
|
|
$insertion_mode = 'after head'; |
1729 |
|
|
} |
1730 |
|
|
return; |
1731 |
|
|
} |
1732 |
|
|
|
1733 |
|
|
## Step 15 |
1734 |
|
|
$insertion_mode = 'in body' and return if $last; |
1735 |
|
|
|
1736 |
|
|
## Step 16 |
1737 |
|
|
$i--; |
1738 |
|
|
$node = $open_elements->[$i]; |
1739 |
|
|
|
1740 |
|
|
## Step 17 |
1741 |
|
|
redo S3; |
1742 |
|
|
} # S3 |
1743 |
|
|
}; # $reset_insertion_mode |
1744 |
|
|
|
1745 |
|
|
my $style_start_tag = sub { |
1746 |
|
|
my $style_el; !!!create-element ($style_el, 'style'); |
1747 |
|
|
## $insertion_mode eq 'in head' and ... (always true) |
1748 |
|
|
(($insertion_mode eq 'in head' and defined $head_element) |
1749 |
|
|
? $head_element : $open_elements->[-1]->[0]) |
1750 |
|
|
->append_child ($style_el); |
1751 |
|
|
$self->{content_model_flag} = 'CDATA'; |
1752 |
|
|
|
1753 |
|
|
my $text = ''; |
1754 |
|
|
!!!next-token; |
1755 |
|
|
while ($token->{type} eq 'character') { |
1756 |
|
|
$text .= $token->{data}; |
1757 |
|
|
!!!next-token; |
1758 |
|
|
} # stop if non-character token or tokenizer stops tokenising |
1759 |
|
|
if (length $text) { |
1760 |
|
|
$style_el->manakai_append_text ($text); |
1761 |
|
|
} |
1762 |
|
|
|
1763 |
|
|
$self->{content_model_flag} = 'PCDATA'; |
1764 |
|
|
|
1765 |
|
|
if ($token->{type} eq 'end tag' and $token->{tag_name} eq 'style') { |
1766 |
|
|
## Ignore the token |
1767 |
|
|
} else { |
1768 |
|
|
!!!parse-error; |
1769 |
|
|
## ISSUE: And ignore? |
1770 |
|
|
} |
1771 |
|
|
!!!next-token; |
1772 |
|
|
}; # $style_start_tag |
1773 |
|
|
|
1774 |
|
|
my $script_start_tag = sub { |
1775 |
wakaba |
1.7 |
my $script_el; |
1776 |
|
|
!!!create-element ($script_el, 'script', $token->{attributes}); |
1777 |
wakaba |
1.2 |
## TODO: mark as "parser-inserted" |
1778 |
|
|
|
1779 |
|
|
$self->{content_model_flag} = 'CDATA'; |
1780 |
|
|
|
1781 |
|
|
my $text = ''; |
1782 |
|
|
!!!next-token; |
1783 |
|
|
while ($token->{type} eq 'character') { |
1784 |
|
|
$text .= $token->{data}; |
1785 |
|
|
!!!next-token; |
1786 |
|
|
} # stop if non-character token or tokenizer stops tokenising |
1787 |
|
|
if (length $text) { |
1788 |
|
|
$script_el->manakai_append_text ($text); |
1789 |
|
|
} |
1790 |
|
|
|
1791 |
|
|
$self->{content_model_flag} = 'PCDATA'; |
1792 |
wakaba |
1.7 |
|
1793 |
wakaba |
1.2 |
if ($token->{type} eq 'end tag' and |
1794 |
|
|
$token->{tag_name} eq 'script') { |
1795 |
|
|
## Ignore the token |
1796 |
|
|
} else { |
1797 |
|
|
!!!parse-error; |
1798 |
|
|
## ISSUE: And ignore? |
1799 |
|
|
## TODO: mark as "already executed" |
1800 |
|
|
} |
1801 |
|
|
|
1802 |
|
|
## TODO: inner_html mode then mark as "already executed" and skip |
1803 |
|
|
if (1) { |
1804 |
|
|
## TODO: $old_insertion_point = current insertion point |
1805 |
|
|
## TODO: insertion point = just before the next input character |
1806 |
|
|
|
1807 |
|
|
(($insertion_mode eq 'in head' and defined $head_element) |
1808 |
|
|
? $head_element : $open_elements->[-1]->[0])->append_child ($script_el); |
1809 |
|
|
|
1810 |
|
|
## TODO: insertion point = $old_insertion_point (might be "undefined") |
1811 |
|
|
|
1812 |
|
|
## TODO: if there is a script that will execute as soon as the parser resume, then... |
1813 |
|
|
} |
1814 |
|
|
|
1815 |
|
|
!!!next-token; |
1816 |
|
|
}; # $script_start_tag |
1817 |
|
|
|
1818 |
|
|
my $formatting_end_tag = sub { |
1819 |
|
|
my $tag_name = shift; |
1820 |
|
|
|
1821 |
|
|
FET: { |
1822 |
|
|
## Step 1 |
1823 |
|
|
my $formatting_element; |
1824 |
|
|
my $formatting_element_i_in_active; |
1825 |
|
|
AFE: for (reverse 0..$#$active_formatting_elements) { |
1826 |
|
|
if ($active_formatting_elements->[$_]->[1] eq $tag_name) { |
1827 |
|
|
$formatting_element = $active_formatting_elements->[$_]; |
1828 |
|
|
$formatting_element_i_in_active = $_; |
1829 |
|
|
last AFE; |
1830 |
|
|
} elsif ($active_formatting_elements->[$_]->[0] eq '#marker') { |
1831 |
|
|
last AFE; |
1832 |
|
|
} |
1833 |
|
|
} # AFE |
1834 |
|
|
unless (defined $formatting_element) { |
1835 |
|
|
!!!parse-error; |
1836 |
|
|
## Ignore the token |
1837 |
|
|
!!!next-token; |
1838 |
|
|
return; |
1839 |
|
|
} |
1840 |
|
|
## has an element in scope |
1841 |
|
|
my $in_scope = 1; |
1842 |
|
|
my $formatting_element_i_in_open; |
1843 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
1844 |
|
|
my $node = $open_elements->[$_]; |
1845 |
|
|
if ($node->[0] eq $formatting_element->[0]) { |
1846 |
|
|
if ($in_scope) { |
1847 |
|
|
$formatting_element_i_in_open = $_; |
1848 |
|
|
last INSCOPE; |
1849 |
|
|
} else { # in open elements but not in scope |
1850 |
|
|
!!!parse-error; |
1851 |
|
|
## Ignore the token |
1852 |
|
|
!!!next-token; |
1853 |
|
|
return; |
1854 |
|
|
} |
1855 |
|
|
} elsif ({ |
1856 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
1857 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
1858 |
|
|
}->{$node->[1]}) { |
1859 |
|
|
$in_scope = 0; |
1860 |
|
|
} |
1861 |
|
|
} # INSCOPE |
1862 |
|
|
unless (defined $formatting_element_i_in_open) { |
1863 |
|
|
!!!parse-error; |
1864 |
|
|
pop @$active_formatting_elements; # $formatting_element |
1865 |
|
|
!!!next-token; ## TODO: ok? |
1866 |
|
|
return; |
1867 |
|
|
} |
1868 |
|
|
if (not $open_elements->[-1]->[0] eq $formatting_element->[0]) { |
1869 |
|
|
!!!parse-error; |
1870 |
|
|
} |
1871 |
|
|
|
1872 |
|
|
## Step 2 |
1873 |
|
|
my $furthest_block; |
1874 |
|
|
my $furthest_block_i_in_open; |
1875 |
|
|
OE: for (reverse 0..$#$open_elements) { |
1876 |
|
|
my $node = $open_elements->[$_]; |
1877 |
|
|
if (not $formatting_category->{$node->[1]} and |
1878 |
|
|
#not $phrasing_category->{$node->[1]} and |
1879 |
|
|
($special_category->{$node->[1]} or |
1880 |
|
|
$scoping_category->{$node->[1]})) { |
1881 |
|
|
$furthest_block = $node; |
1882 |
|
|
$furthest_block_i_in_open = $_; |
1883 |
|
|
} elsif ($node->[0] eq $formatting_element->[0]) { |
1884 |
|
|
last OE; |
1885 |
|
|
} |
1886 |
|
|
} # OE |
1887 |
|
|
|
1888 |
|
|
## Step 3 |
1889 |
|
|
unless (defined $furthest_block) { # MUST |
1890 |
|
|
splice @$open_elements, $formatting_element_i_in_open; |
1891 |
|
|
splice @$active_formatting_elements, $formatting_element_i_in_active, 1; |
1892 |
|
|
!!!next-token; |
1893 |
|
|
return; |
1894 |
|
|
} |
1895 |
|
|
|
1896 |
|
|
## Step 4 |
1897 |
|
|
my $common_ancestor_node = $open_elements->[$formatting_element_i_in_open - 1]; |
1898 |
|
|
|
1899 |
|
|
## Step 5 |
1900 |
|
|
my $furthest_block_parent = $furthest_block->[0]->parent_node; |
1901 |
|
|
if (defined $furthest_block_parent) { |
1902 |
|
|
$furthest_block_parent->remove_child ($furthest_block->[0]); |
1903 |
|
|
} |
1904 |
|
|
|
1905 |
|
|
## Step 6 |
1906 |
|
|
my $bookmark_prev_el |
1907 |
|
|
= $active_formatting_elements->[$formatting_element_i_in_active - 1] |
1908 |
|
|
->[0]; |
1909 |
|
|
|
1910 |
|
|
## Step 7 |
1911 |
|
|
my $node = $furthest_block; |
1912 |
|
|
my $node_i_in_open = $furthest_block_i_in_open; |
1913 |
|
|
my $last_node = $furthest_block; |
1914 |
|
|
S7: { |
1915 |
|
|
## Step 1 |
1916 |
|
|
$node_i_in_open--; |
1917 |
|
|
$node = $open_elements->[$node_i_in_open]; |
1918 |
|
|
|
1919 |
|
|
## Step 2 |
1920 |
|
|
my $node_i_in_active; |
1921 |
|
|
S7S2: { |
1922 |
|
|
for (reverse 0..$#$active_formatting_elements) { |
1923 |
|
|
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
1924 |
|
|
$node_i_in_active = $_; |
1925 |
|
|
last S7S2; |
1926 |
|
|
} |
1927 |
|
|
} |
1928 |
|
|
splice @$open_elements, $node_i_in_open, 1; |
1929 |
|
|
redo S7; |
1930 |
|
|
} # S7S2 |
1931 |
|
|
|
1932 |
|
|
## Step 3 |
1933 |
|
|
last S7 if $node->[0] eq $formatting_element->[0]; |
1934 |
|
|
|
1935 |
|
|
## Step 4 |
1936 |
|
|
if ($last_node->[0] eq $furthest_block->[0]) { |
1937 |
|
|
$bookmark_prev_el = $node->[0]; |
1938 |
|
|
} |
1939 |
|
|
|
1940 |
|
|
## Step 5 |
1941 |
|
|
if ($node->[0]->has_child_nodes ()) { |
1942 |
|
|
my $clone = [$node->[0]->clone_node (0), $node->[1]]; |
1943 |
|
|
$active_formatting_elements->[$node_i_in_active] = $clone; |
1944 |
|
|
$open_elements->[$node_i_in_open] = $clone; |
1945 |
|
|
$node = $clone; |
1946 |
|
|
} |
1947 |
|
|
|
1948 |
|
|
## Step 6 |
1949 |
wakaba |
1.6 |
$node->[0]->append_child ($last_node->[0]); |
1950 |
wakaba |
1.2 |
|
1951 |
|
|
## Step 7 |
1952 |
|
|
$last_node = $node; |
1953 |
|
|
|
1954 |
|
|
## Step 8 |
1955 |
|
|
redo S7; |
1956 |
|
|
} # S7 |
1957 |
|
|
|
1958 |
|
|
## Step 8 |
1959 |
wakaba |
1.6 |
$common_ancestor_node->[0]->append_child ($last_node->[0]); |
1960 |
wakaba |
1.2 |
|
1961 |
|
|
## Step 9 |
1962 |
|
|
my $clone = [$formatting_element->[0]->clone_node (0), |
1963 |
|
|
$formatting_element->[1]]; |
1964 |
|
|
|
1965 |
|
|
## Step 10 |
1966 |
|
|
my @cn = @{$furthest_block->[0]->child_nodes}; |
1967 |
|
|
$clone->[0]->append_child ($_) for @cn; |
1968 |
|
|
|
1969 |
|
|
## Step 11 |
1970 |
|
|
$furthest_block->[0]->append_child ($clone->[0]); |
1971 |
|
|
|
1972 |
|
|
## Step 12 |
1973 |
|
|
my $i; |
1974 |
|
|
AFE: for (reverse 0..$#$active_formatting_elements) { |
1975 |
|
|
if ($active_formatting_elements->[$_]->[0] eq $formatting_element->[0]) { |
1976 |
|
|
splice @$active_formatting_elements, $_, 1; |
1977 |
|
|
$i-- and last AFE if defined $i; |
1978 |
|
|
} elsif ($active_formatting_elements->[$_]->[0] eq $bookmark_prev_el) { |
1979 |
|
|
$i = $_; |
1980 |
|
|
} |
1981 |
|
|
} # AFE |
1982 |
|
|
splice @$active_formatting_elements, $i + 1, 0, $clone; |
1983 |
|
|
|
1984 |
|
|
## Step 13 |
1985 |
|
|
undef $i; |
1986 |
|
|
OE: for (reverse 0..$#$open_elements) { |
1987 |
|
|
if ($open_elements->[$_]->[0] eq $formatting_element->[0]) { |
1988 |
|
|
splice @$open_elements, $_, 1; |
1989 |
|
|
$i-- and last OE if defined $i; |
1990 |
|
|
} elsif ($open_elements->[$_]->[0] eq $furthest_block->[0]) { |
1991 |
|
|
$i = $_; |
1992 |
|
|
} |
1993 |
|
|
} # OE |
1994 |
|
|
splice @$open_elements, $i + 1, 1, $clone; |
1995 |
|
|
|
1996 |
|
|
## Step 14 |
1997 |
|
|
redo FET; |
1998 |
|
|
} # FET |
1999 |
|
|
}; # $formatting_end_tag |
2000 |
|
|
|
2001 |
wakaba |
1.7 |
my $insert_to_current = sub { |
2002 |
|
|
$open_elements->[-1]->[0]->append_child (shift); |
2003 |
|
|
}; # $insert_to_current |
2004 |
|
|
|
2005 |
|
|
my $insert_to_foster = sub { |
2006 |
|
|
my $child = shift; |
2007 |
|
|
if ({ |
2008 |
|
|
table => 1, tbody => 1, tfoot => 1, |
2009 |
|
|
thead => 1, tr => 1, |
2010 |
|
|
}->{$open_elements->[-1]->[1]}) { |
2011 |
|
|
# MUST |
2012 |
|
|
my $foster_parent_element; |
2013 |
|
|
my $next_sibling; |
2014 |
|
|
OE: for (reverse 0..$#$open_elements) { |
2015 |
|
|
if ($open_elements->[$_]->[1] eq 'table') { |
2016 |
|
|
my $parent = $open_elements->[$_]->[0]->parent_node; |
2017 |
|
|
if (defined $parent and $parent->node_type == 1) { |
2018 |
|
|
$foster_parent_element = $parent; |
2019 |
|
|
$next_sibling = $open_elements->[$_]->[0]; |
2020 |
|
|
} else { |
2021 |
|
|
$foster_parent_element |
2022 |
|
|
= $open_elements->[$_ - 1]->[0]; |
2023 |
|
|
} |
2024 |
|
|
last OE; |
2025 |
|
|
} |
2026 |
|
|
} # OE |
2027 |
|
|
$foster_parent_element = $open_elements->[0]->[0] |
2028 |
|
|
unless defined $foster_parent_element; |
2029 |
|
|
$foster_parent_element->insert_before |
2030 |
|
|
($child, $next_sibling); |
2031 |
|
|
} else { |
2032 |
|
|
$open_elements->[-1]->[0]->append_child ($child); |
2033 |
|
|
} |
2034 |
|
|
}; # $insert_to_foster |
2035 |
|
|
|
2036 |
wakaba |
1.2 |
my $in_body = sub { |
2037 |
|
|
my $insert = shift; |
2038 |
|
|
if ($token->{type} eq 'start tag') { |
2039 |
|
|
if ($token->{tag_name} eq 'script') { |
2040 |
|
|
$script_start_tag->(); |
2041 |
|
|
return; |
2042 |
|
|
} elsif ($token->{tag_name} eq 'style') { |
2043 |
|
|
$style_start_tag->(); |
2044 |
|
|
return; |
2045 |
|
|
} elsif ({ |
2046 |
wakaba |
1.7 |
base => 1, link => 1, meta => 1, |
2047 |
wakaba |
1.2 |
}->{$token->{tag_name}}) { |
2048 |
wakaba |
1.8 |
!!!parse-error ($token->{tag_name}.' in body'); |
2049 |
wakaba |
1.2 |
## NOTE: This is an "as if in head" code clone |
2050 |
|
|
my $el; |
2051 |
|
|
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
2052 |
|
|
if (defined $head_element) { |
2053 |
|
|
$head_element->append_child ($el); |
2054 |
|
|
} else { |
2055 |
|
|
$insert->($el); |
2056 |
|
|
} |
2057 |
|
|
|
2058 |
|
|
!!!next-token; |
2059 |
|
|
return; |
2060 |
wakaba |
1.7 |
} elsif ($token->{tag_name} eq 'title') { |
2061 |
wakaba |
1.8 |
!!!parse-error ('title in body'); |
2062 |
wakaba |
1.7 |
## NOTE: There is an "as if in head" code clone |
2063 |
|
|
my $title_el; |
2064 |
|
|
!!!create-element ($title_el, 'title', $token->{attributes}); |
2065 |
|
|
(defined $head_element ? $head_element : $open_elements->[-1]->[0]) |
2066 |
|
|
->append_child ($title_el); |
2067 |
|
|
$self->{content_model_flag} = 'RCDATA'; |
2068 |
|
|
|
2069 |
|
|
my $text = ''; |
2070 |
|
|
!!!next-token; |
2071 |
|
|
while ($token->{type} eq 'character') { |
2072 |
|
|
$text .= $token->{data}; |
2073 |
|
|
!!!next-token; |
2074 |
|
|
} |
2075 |
|
|
if (length $text) { |
2076 |
|
|
$title_el->manakai_append_text ($text); |
2077 |
|
|
} |
2078 |
|
|
|
2079 |
|
|
$self->{content_model_flag} = 'PCDATA'; |
2080 |
|
|
|
2081 |
|
|
if ($token->{type} eq 'end tag' and |
2082 |
|
|
$token->{tag_name} eq 'title') { |
2083 |
|
|
## Ignore the token |
2084 |
|
|
} else { |
2085 |
|
|
!!!parse-error; |
2086 |
|
|
## ISSUE: And ignore? |
2087 |
|
|
} |
2088 |
|
|
!!!next-token; |
2089 |
|
|
return; |
2090 |
wakaba |
1.2 |
} elsif ($token->{tag_name} eq 'body') { |
2091 |
|
|
!!!parse-error; |
2092 |
|
|
|
2093 |
|
|
if (@$open_elements == 1 or |
2094 |
|
|
$open_elements->[1]->[1] ne 'body') { |
2095 |
|
|
## Ignore the token |
2096 |
|
|
} else { |
2097 |
|
|
my $body_el = $open_elements->[1]->[0]; |
2098 |
|
|
for my $attr_name (keys %{$token->{attributes}}) { |
2099 |
|
|
unless ($body_el->has_attribute_ns (undef, $attr_name)) { |
2100 |
|
|
$body_el->set_attribute_ns |
2101 |
|
|
(undef, [undef, $attr_name], |
2102 |
|
|
$token->{attributes}->{$attr_name}->{value}); |
2103 |
|
|
} |
2104 |
|
|
} |
2105 |
|
|
} |
2106 |
|
|
!!!next-token; |
2107 |
|
|
return; |
2108 |
|
|
} elsif ({ |
2109 |
|
|
address => 1, blockquote => 1, center => 1, dir => 1, |
2110 |
|
|
div => 1, dl => 1, fieldset => 1, listing => 1, |
2111 |
|
|
menu => 1, ol => 1, p => 1, ul => 1, |
2112 |
|
|
pre => 1, |
2113 |
|
|
}->{$token->{tag_name}}) { |
2114 |
|
|
## has a p element in scope |
2115 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2116 |
|
|
if ($_->[1] eq 'p') { |
2117 |
|
|
!!!back-token; |
2118 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2119 |
|
|
return; |
2120 |
|
|
} elsif ({ |
2121 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2122 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2123 |
|
|
}->{$_->[1]}) { |
2124 |
|
|
last INSCOPE; |
2125 |
|
|
} |
2126 |
|
|
} # INSCOPE |
2127 |
|
|
|
2128 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2129 |
|
|
if ($token->{tag_name} eq 'pre') { |
2130 |
|
|
!!!next-token; |
2131 |
|
|
if ($token->{type} eq 'character') { |
2132 |
|
|
$token->{data} =~ s/^\x0A//; |
2133 |
|
|
unless (length $token->{data}) { |
2134 |
|
|
!!!next-token; |
2135 |
|
|
} |
2136 |
|
|
} |
2137 |
|
|
} else { |
2138 |
|
|
!!!next-token; |
2139 |
|
|
} |
2140 |
|
|
return; |
2141 |
|
|
} elsif ($token->{tag_name} eq 'form') { |
2142 |
|
|
if (defined $form_element) { |
2143 |
|
|
!!!parse-error; |
2144 |
|
|
## Ignore the token |
2145 |
|
|
} else { |
2146 |
|
|
## has a p element in scope |
2147 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2148 |
|
|
if ($_->[1] eq 'p') { |
2149 |
|
|
!!!back-token; |
2150 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2151 |
|
|
return; |
2152 |
|
|
} elsif ({ |
2153 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2154 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2155 |
|
|
}->{$_->[1]}) { |
2156 |
|
|
last INSCOPE; |
2157 |
|
|
} |
2158 |
|
|
} # INSCOPE |
2159 |
|
|
|
2160 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2161 |
|
|
$form_element = $open_elements->[-1]->[0]; |
2162 |
|
|
!!!next-token; |
2163 |
|
|
return; |
2164 |
|
|
} |
2165 |
|
|
} elsif ($token->{tag_name} eq 'li') { |
2166 |
|
|
## has a p element in scope |
2167 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2168 |
|
|
if ($_->[1] eq 'p') { |
2169 |
|
|
!!!back-token; |
2170 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2171 |
|
|
return; |
2172 |
|
|
} elsif ({ |
2173 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2174 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2175 |
|
|
}->{$_->[1]}) { |
2176 |
|
|
last INSCOPE; |
2177 |
|
|
} |
2178 |
|
|
} # INSCOPE |
2179 |
|
|
|
2180 |
|
|
## Step 1 |
2181 |
|
|
my $i = -1; |
2182 |
|
|
my $node = $open_elements->[$i]; |
2183 |
|
|
LI: { |
2184 |
|
|
## Step 2 |
2185 |
|
|
if ($node->[1] eq 'li') { |
2186 |
|
|
splice @$open_elements, $i; |
2187 |
|
|
last LI; |
2188 |
|
|
} |
2189 |
|
|
|
2190 |
|
|
## Step 3 |
2191 |
|
|
if (not $formatting_category->{$node->[1]} and |
2192 |
|
|
#not $phrasing_category->{$node->[1]} and |
2193 |
|
|
($special_category->{$node->[1]} or |
2194 |
|
|
$scoping_category->{$node->[1]}) and |
2195 |
|
|
$node->[1] ne 'address' and $node->[1] ne 'div') { |
2196 |
|
|
last LI; |
2197 |
|
|
} |
2198 |
|
|
|
2199 |
|
|
## Step 4 |
2200 |
wakaba |
1.7 |
$i--; |
2201 |
wakaba |
1.2 |
$node = $open_elements->[$i]; |
2202 |
|
|
redo LI; |
2203 |
|
|
} # LI |
2204 |
|
|
|
2205 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2206 |
|
|
!!!next-token; |
2207 |
|
|
return; |
2208 |
|
|
} elsif ($token->{tag_name} eq 'dd' or $token->{tag_name} eq 'dt') { |
2209 |
|
|
## has a p element in scope |
2210 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2211 |
|
|
if ($_->[1] eq 'p') { |
2212 |
|
|
!!!back-token; |
2213 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2214 |
|
|
return; |
2215 |
|
|
} elsif ({ |
2216 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2217 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2218 |
|
|
}->{$_->[1]}) { |
2219 |
|
|
last INSCOPE; |
2220 |
|
|
} |
2221 |
|
|
} # INSCOPE |
2222 |
|
|
|
2223 |
|
|
## Step 1 |
2224 |
|
|
my $i = -1; |
2225 |
|
|
my $node = $open_elements->[$i]; |
2226 |
|
|
LI: { |
2227 |
|
|
## Step 2 |
2228 |
|
|
if ($node->[1] eq 'dt' or $node->[1] eq 'dd') { |
2229 |
|
|
splice @$open_elements, $i; |
2230 |
|
|
last LI; |
2231 |
|
|
} |
2232 |
|
|
|
2233 |
|
|
## Step 3 |
2234 |
|
|
if (not $formatting_category->{$node->[1]} and |
2235 |
|
|
#not $phrasing_category->{$node->[1]} and |
2236 |
|
|
($special_category->{$node->[1]} or |
2237 |
|
|
$scoping_category->{$node->[1]}) and |
2238 |
|
|
$node->[1] ne 'address' and $node->[1] ne 'div') { |
2239 |
|
|
last LI; |
2240 |
|
|
} |
2241 |
|
|
|
2242 |
|
|
## Step 4 |
2243 |
wakaba |
1.7 |
$i--; |
2244 |
wakaba |
1.2 |
$node = $open_elements->[$i]; |
2245 |
|
|
redo LI; |
2246 |
|
|
} # LI |
2247 |
|
|
|
2248 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2249 |
|
|
!!!next-token; |
2250 |
|
|
return; |
2251 |
|
|
} elsif ($token->{tag_name} eq 'plaintext') { |
2252 |
|
|
## has a p element in scope |
2253 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2254 |
|
|
if ($_->[1] eq 'p') { |
2255 |
|
|
!!!back-token; |
2256 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2257 |
|
|
return; |
2258 |
|
|
} elsif ({ |
2259 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2260 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2261 |
|
|
}->{$_->[1]}) { |
2262 |
|
|
last INSCOPE; |
2263 |
|
|
} |
2264 |
|
|
} # INSCOPE |
2265 |
|
|
|
2266 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2267 |
|
|
|
2268 |
|
|
$self->{content_model_flag} = 'PLAINTEXT'; |
2269 |
|
|
|
2270 |
|
|
!!!next-token; |
2271 |
|
|
return; |
2272 |
|
|
} elsif ({ |
2273 |
|
|
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
2274 |
|
|
}->{$token->{tag_name}}) { |
2275 |
|
|
## has a p element in scope |
2276 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
2277 |
|
|
my $node = $open_elements->[$_]; |
2278 |
|
|
if ($node->[1] eq 'p') { |
2279 |
|
|
!!!back-token; |
2280 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2281 |
|
|
return; |
2282 |
|
|
} elsif ({ |
2283 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2284 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2285 |
|
|
}->{$node->[1]}) { |
2286 |
|
|
last INSCOPE; |
2287 |
|
|
} |
2288 |
|
|
} # INSCOPE |
2289 |
|
|
|
2290 |
|
|
## has an element in scope |
2291 |
|
|
my $i; |
2292 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
2293 |
|
|
my $node = $open_elements->[$_]; |
2294 |
|
|
if ({ |
2295 |
|
|
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
2296 |
|
|
}->{$node->[1]}) { |
2297 |
|
|
$i = $_; |
2298 |
|
|
last INSCOPE; |
2299 |
|
|
} elsif ({ |
2300 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2301 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2302 |
|
|
}->{$node->[1]}) { |
2303 |
|
|
last INSCOPE; |
2304 |
|
|
} |
2305 |
|
|
} # INSCOPE |
2306 |
|
|
|
2307 |
|
|
if (defined $i) { |
2308 |
|
|
!!!parse-error; |
2309 |
|
|
splice @$open_elements, $i; |
2310 |
|
|
} |
2311 |
|
|
|
2312 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2313 |
|
|
|
2314 |
|
|
!!!next-token; |
2315 |
|
|
return; |
2316 |
|
|
} elsif ($token->{tag_name} eq 'a') { |
2317 |
|
|
AFE: for my $i (reverse 0..$#$active_formatting_elements) { |
2318 |
|
|
my $node = $active_formatting_elements->[$i]; |
2319 |
|
|
if ($node->[1] eq 'a') { |
2320 |
wakaba |
1.7 |
!!!parse-error ('a in a'); |
2321 |
wakaba |
1.2 |
|
2322 |
|
|
!!!back-token; |
2323 |
|
|
$token = {type => 'end tag', tag_name => 'a'}; |
2324 |
|
|
$formatting_end_tag->($token->{tag_name}); |
2325 |
|
|
|
2326 |
wakaba |
1.7 |
AFE2: for (reverse 0..$#$active_formatting_elements) { |
2327 |
|
|
if ($active_formatting_elements->[$_]->[0] eq $node->[0]) { |
2328 |
|
|
splice @$active_formatting_elements, $_, 1; |
2329 |
|
|
last AFE2; |
2330 |
|
|
} |
2331 |
|
|
} # AFE2 |
2332 |
wakaba |
1.2 |
OE: for (reverse 0..$#$open_elements) { |
2333 |
|
|
if ($open_elements->[$_]->[0] eq $node->[0]) { |
2334 |
wakaba |
1.7 |
splice @$open_elements, $_, 1; |
2335 |
wakaba |
1.2 |
last OE; |
2336 |
|
|
} |
2337 |
|
|
} # OE |
2338 |
|
|
last AFE; |
2339 |
|
|
} elsif ($node->[0] eq '#marker') { |
2340 |
|
|
last AFE; |
2341 |
|
|
} |
2342 |
|
|
} # AFE |
2343 |
|
|
|
2344 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2345 |
wakaba |
1.2 |
|
2346 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2347 |
|
|
push @$active_formatting_elements, $open_elements->[-1]; |
2348 |
|
|
|
2349 |
|
|
!!!next-token; |
2350 |
|
|
return; |
2351 |
|
|
} elsif ({ |
2352 |
|
|
b => 1, big => 1, em => 1, font => 1, i => 1, |
2353 |
|
|
nobr => 1, s => 1, small => 1, strile => 1, |
2354 |
|
|
strong => 1, tt => 1, u => 1, |
2355 |
|
|
}->{$token->{tag_name}}) { |
2356 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2357 |
wakaba |
1.2 |
|
2358 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2359 |
|
|
push @$active_formatting_elements, $open_elements->[-1]; |
2360 |
|
|
|
2361 |
|
|
!!!next-token; |
2362 |
|
|
return; |
2363 |
|
|
} elsif ($token->{tag_name} eq 'button') { |
2364 |
|
|
## has a button element in scope |
2365 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
2366 |
|
|
my $node = $open_elements->[$_]; |
2367 |
|
|
if ($node->[1] eq 'button') { |
2368 |
|
|
!!!parse-error; |
2369 |
|
|
!!!back-token; |
2370 |
|
|
$token = {type => 'end tag', tag_name => 'button'}; |
2371 |
|
|
return; |
2372 |
|
|
} elsif ({ |
2373 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2374 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2375 |
|
|
}->{$node->[1]}) { |
2376 |
|
|
last INSCOPE; |
2377 |
|
|
} |
2378 |
|
|
} # INSCOPE |
2379 |
|
|
|
2380 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2381 |
wakaba |
1.2 |
|
2382 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2383 |
|
|
push @$active_formatting_elements, ['#marker', '']; |
2384 |
|
|
|
2385 |
|
|
!!!next-token; |
2386 |
|
|
return; |
2387 |
|
|
} elsif ($token->{tag_name} eq 'marquee' or |
2388 |
|
|
$token->{tag_name} eq 'object') { |
2389 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2390 |
wakaba |
1.2 |
|
2391 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2392 |
|
|
push @$active_formatting_elements, ['#marker', '']; |
2393 |
|
|
|
2394 |
|
|
!!!next-token; |
2395 |
|
|
return; |
2396 |
|
|
} elsif ($token->{tag_name} eq 'xmp') { |
2397 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2398 |
wakaba |
1.2 |
|
2399 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2400 |
|
|
|
2401 |
|
|
$self->{content_model_flag} = 'CDATA'; |
2402 |
|
|
|
2403 |
|
|
!!!next-token; |
2404 |
|
|
return; |
2405 |
wakaba |
1.6 |
} elsif ($token->{tag_name} eq 'table') { |
2406 |
wakaba |
1.2 |
## has a p element in scope |
2407 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2408 |
|
|
if ($_->[1] eq 'p') { |
2409 |
|
|
!!!back-token; |
2410 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2411 |
|
|
return; |
2412 |
|
|
} elsif ({ |
2413 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2414 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2415 |
|
|
}->{$_->[1]}) { |
2416 |
|
|
last INSCOPE; |
2417 |
|
|
} |
2418 |
|
|
} # INSCOPE |
2419 |
|
|
|
2420 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2421 |
|
|
|
2422 |
|
|
$insertion_mode = 'in table'; |
2423 |
|
|
|
2424 |
|
|
!!!next-token; |
2425 |
|
|
return; |
2426 |
|
|
} elsif ({ |
2427 |
|
|
area => 1, basefont => 1, bgsound => 1, br => 1, |
2428 |
|
|
embed => 1, img => 1, param => 1, spacer => 1, wbr => 1, |
2429 |
|
|
image => 1, |
2430 |
|
|
}->{$token->{tag_name}}) { |
2431 |
|
|
if ($token->{tag_name} eq 'image') { |
2432 |
|
|
!!!parse-error; |
2433 |
|
|
$token->{tag_name} = 'img'; |
2434 |
|
|
} |
2435 |
|
|
|
2436 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2437 |
wakaba |
1.2 |
|
2438 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2439 |
|
|
pop @$open_elements; |
2440 |
|
|
|
2441 |
|
|
!!!next-token; |
2442 |
|
|
return; |
2443 |
|
|
} elsif ($token->{tag_name} eq 'hr') { |
2444 |
|
|
## has a p element in scope |
2445 |
|
|
INSCOPE: for (reverse @$open_elements) { |
2446 |
|
|
if ($_->[1] eq 'p') { |
2447 |
|
|
!!!back-token; |
2448 |
|
|
$token = {type => 'end tag', tag_name => 'p'}; |
2449 |
|
|
return; |
2450 |
|
|
} elsif ({ |
2451 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2452 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2453 |
|
|
}->{$_->[1]}) { |
2454 |
|
|
last INSCOPE; |
2455 |
|
|
} |
2456 |
|
|
} # INSCOPE |
2457 |
|
|
|
2458 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2459 |
|
|
pop @$open_elements; |
2460 |
|
|
|
2461 |
|
|
!!!next-token; |
2462 |
|
|
return; |
2463 |
|
|
} elsif ($token->{tag_name} eq 'input') { |
2464 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2465 |
wakaba |
1.2 |
|
2466 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2467 |
|
|
## TODO: associate with $form_element if defined |
2468 |
|
|
pop @$open_elements; |
2469 |
|
|
|
2470 |
|
|
!!!next-token; |
2471 |
|
|
return; |
2472 |
|
|
} elsif ($token->{tag_name} eq 'isindex') { |
2473 |
|
|
!!!parse-error; |
2474 |
|
|
|
2475 |
|
|
if (defined $form_element) { |
2476 |
|
|
## Ignore the token |
2477 |
|
|
!!!next-token; |
2478 |
|
|
return; |
2479 |
|
|
} else { |
2480 |
|
|
my $at = $token->{attributes}; |
2481 |
|
|
$at->{name} = {name => 'name', value => 'isindex'}; |
2482 |
|
|
my @tokens = ( |
2483 |
|
|
{type => 'start tag', tag_name => 'form'}, |
2484 |
|
|
{type => 'start tag', tag_name => 'hr'}, |
2485 |
|
|
{type => 'start tag', tag_name => 'p'}, |
2486 |
|
|
{type => 'start tag', tag_name => 'label'}, |
2487 |
|
|
{type => 'character', |
2488 |
wakaba |
1.7 |
data => 'This is a searchable index. Insert your search keywords here: '}, # SHOULD |
2489 |
wakaba |
1.2 |
## TODO: make this configurable |
2490 |
|
|
{type => 'start tag', tag_name => 'input', attributes => $at}, |
2491 |
|
|
#{type => 'character', data => ''}, # SHOULD |
2492 |
|
|
{type => 'end tag', tag_name => 'label'}, |
2493 |
|
|
{type => 'end tag', tag_name => 'p'}, |
2494 |
|
|
{type => 'start tag', tag_name => 'hr'}, |
2495 |
|
|
{type => 'end tag', tag_name => 'form'}, |
2496 |
|
|
); |
2497 |
|
|
$token = shift @tokens; |
2498 |
|
|
!!!back-token (@tokens); |
2499 |
|
|
return; |
2500 |
|
|
} |
2501 |
|
|
} elsif ({ |
2502 |
|
|
textarea => 1, |
2503 |
|
|
noembed => 1, |
2504 |
|
|
noframes => 1, |
2505 |
|
|
noscript => 0, ## TODO: 1 if scripting is enabled |
2506 |
|
|
}->{$token->{tag_name}}) { |
2507 |
|
|
my $tag_name = $token->{tag_name}; |
2508 |
|
|
my $el; |
2509 |
|
|
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
2510 |
|
|
|
2511 |
|
|
if ($token->{tag_name} eq 'textarea') { |
2512 |
|
|
## TODO: form_element if defined |
2513 |
|
|
$self->{content_model_flag} = 'RCDATA'; |
2514 |
|
|
} else { |
2515 |
|
|
$self->{content_model_flag} = 'CDATA'; |
2516 |
|
|
} |
2517 |
|
|
|
2518 |
|
|
$insert->($el); |
2519 |
|
|
|
2520 |
|
|
my $text = ''; |
2521 |
|
|
!!!next-token; |
2522 |
|
|
while ($token->{type} eq 'character') { |
2523 |
|
|
$text .= $token->{data}; |
2524 |
|
|
!!!next-token; |
2525 |
|
|
} |
2526 |
|
|
if (length $text) { |
2527 |
|
|
$el->manakai_append_text ($text); |
2528 |
|
|
} |
2529 |
|
|
|
2530 |
|
|
$self->{content_model_flag} = 'PCDATA'; |
2531 |
|
|
|
2532 |
|
|
if ($token->{type} eq 'end tag' and |
2533 |
|
|
$token->{tag_name} eq $tag_name) { |
2534 |
|
|
## Ignore the token |
2535 |
|
|
} else { |
2536 |
|
|
!!!parse-error; |
2537 |
|
|
## ISSUE: And ignore? |
2538 |
|
|
} |
2539 |
|
|
!!!next-token; |
2540 |
|
|
return; |
2541 |
wakaba |
1.7 |
} elsif ($token->{tag_name} eq 'select') { |
2542 |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
2543 |
wakaba |
1.2 |
|
2544 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2545 |
|
|
|
2546 |
|
|
$insertion_mode = 'in select'; |
2547 |
|
|
!!!next-token; |
2548 |
|
|
return; |
2549 |
|
|
} elsif ({ |
2550 |
|
|
caption => 1, col => 1, colgroup => 1, frame => 1, |
2551 |
|
|
frameset => 1, head => 1, option => 1, optgroup => 1, |
2552 |
|
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
2553 |
|
|
thead => 1, tr => 1, |
2554 |
|
|
}->{$token->{tag_name}}) { |
2555 |
|
|
!!!parse-error; |
2556 |
|
|
## Ignore the token |
2557 |
|
|
!!!next-token; |
2558 |
|
|
return; |
2559 |
|
|
|
2560 |
|
|
## ISSUE: An issue on HTML5 new elements in the spec. |
2561 |
|
|
} else { |
2562 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
2563 |
wakaba |
1.2 |
|
2564 |
|
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
2565 |
|
|
|
2566 |
|
|
!!!next-token; |
2567 |
|
|
return; |
2568 |
|
|
} |
2569 |
|
|
} elsif ($token->{type} eq 'end tag') { |
2570 |
|
|
if ($token->{tag_name} eq 'body') { |
2571 |
|
|
if (@$open_elements > 1 and $open_elements->[1]->[1] eq 'body') { |
2572 |
|
|
## ISSUE: There is an issue in the spec. |
2573 |
|
|
if ($open_elements->[-1]->[1] ne 'body') { |
2574 |
|
|
!!!parse-error; |
2575 |
|
|
} |
2576 |
|
|
$insertion_mode = 'after body'; |
2577 |
|
|
!!!next-token; |
2578 |
|
|
return; |
2579 |
|
|
} else { |
2580 |
|
|
!!!parse-error; |
2581 |
|
|
## Ignore the token |
2582 |
|
|
!!!next-token; |
2583 |
|
|
return; |
2584 |
|
|
} |
2585 |
|
|
} elsif ($token->{tag_name} eq 'html') { |
2586 |
|
|
if (@$open_elements > 1 and $open_elements->[1]->[1] eq 'body') { |
2587 |
|
|
## ISSUE: There is an issue in the spec. |
2588 |
|
|
if ($open_elements->[-1]->[1] ne 'body') { |
2589 |
|
|
!!!parse-error; |
2590 |
|
|
} |
2591 |
|
|
$insertion_mode = 'after body'; |
2592 |
|
|
## reprocess |
2593 |
|
|
return; |
2594 |
|
|
} else { |
2595 |
|
|
!!!parse-error; |
2596 |
|
|
## Ignore the token |
2597 |
|
|
!!!next-token; |
2598 |
|
|
return; |
2599 |
|
|
} |
2600 |
|
|
} elsif ({ |
2601 |
|
|
address => 1, blockquote => 1, center => 1, dir => 1, |
2602 |
|
|
div => 1, dl => 1, fieldset => 1, listing => 1, |
2603 |
|
|
menu => 1, ol => 1, pre => 1, ul => 1, |
2604 |
|
|
form => 1, |
2605 |
|
|
p => 1, |
2606 |
|
|
dd => 1, dt => 1, li => 1, |
2607 |
|
|
button => 1, marquee => 1, object => 1, |
2608 |
|
|
}->{$token->{tag_name}}) { |
2609 |
|
|
## has an element in scope |
2610 |
|
|
my $i; |
2611 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
2612 |
|
|
my $node = $open_elements->[$_]; |
2613 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
2614 |
|
|
## generate implied end tags |
2615 |
|
|
if ({ |
2616 |
|
|
dd => ($token->{tag_name} ne 'dd'), |
2617 |
|
|
dt => ($token->{tag_name} ne 'dt'), |
2618 |
|
|
li => ($token->{tag_name} ne 'li'), |
2619 |
|
|
p => ($token->{tag_name} ne 'p'), |
2620 |
|
|
td => 1, th => 1, tr => 1, |
2621 |
|
|
}->{$open_elements->[-1]->[1]}) { |
2622 |
|
|
!!!back-token; |
2623 |
|
|
$token = {type => 'end tag', |
2624 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
2625 |
|
|
return; |
2626 |
|
|
} |
2627 |
|
|
$i = $_; |
2628 |
|
|
last INSCOPE unless $token->{tag_name} eq 'p'; |
2629 |
|
|
} elsif ({ |
2630 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2631 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2632 |
|
|
}->{$node->[1]}) { |
2633 |
|
|
last INSCOPE; |
2634 |
|
|
} |
2635 |
|
|
} # INSCOPE |
2636 |
|
|
|
2637 |
|
|
if ($open_elements->[-1]->[1] ne $token->{tag_name}) { |
2638 |
|
|
!!!parse-error; |
2639 |
|
|
} |
2640 |
|
|
|
2641 |
|
|
splice @$open_elements, $i if defined $i; |
2642 |
|
|
undef $form_element if $token->{tag_name} eq 'form'; |
2643 |
|
|
$clear_up_to_marker->() |
2644 |
|
|
if { |
2645 |
|
|
button => 1, marquee => 1, object => 1, |
2646 |
|
|
}->{$token->{tag_name}}; |
2647 |
|
|
!!!next-token; |
2648 |
|
|
return; |
2649 |
|
|
} elsif ({ |
2650 |
|
|
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
2651 |
|
|
}->{$token->{tag_name}}) { |
2652 |
|
|
## has an element in scope |
2653 |
|
|
my $i; |
2654 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
2655 |
|
|
my $node = $open_elements->[$_]; |
2656 |
|
|
if ({ |
2657 |
|
|
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
2658 |
|
|
}->{$node->[1]}) { |
2659 |
|
|
## generate implied end tags |
2660 |
|
|
if ({ |
2661 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
2662 |
|
|
td => 1, th => 1, tr => 1, |
2663 |
|
|
}->{$open_elements->[-1]->[1]}) { |
2664 |
|
|
!!!back-token; |
2665 |
|
|
$token = {type => 'end tag', |
2666 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
2667 |
|
|
return; |
2668 |
|
|
} |
2669 |
|
|
$i = $_; |
2670 |
|
|
last INSCOPE; |
2671 |
|
|
} elsif ({ |
2672 |
|
|
table => 1, caption => 1, td => 1, th => 1, |
2673 |
|
|
button => 1, marquee => 1, object => 1, html => 1, |
2674 |
|
|
}->{$node->[1]}) { |
2675 |
|
|
last INSCOPE; |
2676 |
|
|
} |
2677 |
|
|
} # INSCOPE |
2678 |
|
|
|
2679 |
|
|
if ($open_elements->[-1]->[1] ne $token->{tag_name}) { |
2680 |
|
|
!!!parse-error; |
2681 |
|
|
} |
2682 |
|
|
|
2683 |
|
|
splice @$open_elements, $i if defined $i; |
2684 |
|
|
!!!next-token; |
2685 |
|
|
return; |
2686 |
|
|
} elsif ({ |
2687 |
|
|
a => 1, |
2688 |
|
|
b => 1, big => 1, em => 1, font => 1, i => 1, |
2689 |
|
|
nobr => 1, s => 1, small => 1, strile => 1, |
2690 |
|
|
strong => 1, tt => 1, u => 1, |
2691 |
|
|
}->{$token->{tag_name}}) { |
2692 |
|
|
$formatting_end_tag->($token->{tag_name}); |
2693 |
|
|
return; |
2694 |
|
|
} elsif ({ |
2695 |
|
|
caption => 1, col => 1, colgroup => 1, frame => 1, |
2696 |
|
|
frameset => 1, head => 1, option => 1, optgroup => 1, |
2697 |
|
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
2698 |
|
|
thead => 1, tr => 1, |
2699 |
|
|
area => 1, basefont => 1, bgsound => 1, br => 1, |
2700 |
|
|
embed => 1, hr => 1, iframe => 1, image => 1, |
2701 |
|
|
img => 1, input => 1, isindex=> 1, noembed => 1, |
2702 |
|
|
noframes => 1, param => 1, select => 1, spacer => 1, |
2703 |
|
|
table => 1, textarea => 1, wbr => 1, |
2704 |
|
|
noscript => 0, ## TODO: if scripting is enabled |
2705 |
|
|
}->{$token->{tag_name}}) { |
2706 |
|
|
!!!parse-error; |
2707 |
|
|
## Ignore the token |
2708 |
|
|
!!!next-token; |
2709 |
|
|
return; |
2710 |
|
|
|
2711 |
|
|
## ISSUE: Issue on HTML5 new elements in spec |
2712 |
|
|
|
2713 |
|
|
} else { |
2714 |
|
|
## Step 1 |
2715 |
|
|
my $node_i = -1; |
2716 |
|
|
my $node = $open_elements->[$node_i]; |
2717 |
|
|
|
2718 |
|
|
## Step 2 |
2719 |
|
|
S2: { |
2720 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
2721 |
|
|
## Step 1 |
2722 |
|
|
## generate implied end tags |
2723 |
|
|
if ({ |
2724 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
2725 |
|
|
td => 1, th => 1, tr => 1, |
2726 |
|
|
}->{$open_elements->[-1]->[1]}) { |
2727 |
|
|
!!!back-token; |
2728 |
|
|
$token = {type => 'end tag', |
2729 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
2730 |
|
|
return; |
2731 |
|
|
} |
2732 |
|
|
|
2733 |
|
|
## Step 2 |
2734 |
|
|
if ($token->{tag_name} ne $open_elements->[-1]->[1]) { |
2735 |
|
|
!!!parse-error; |
2736 |
|
|
} |
2737 |
|
|
|
2738 |
|
|
## Step 3 |
2739 |
|
|
splice @$open_elements, $node_i; |
2740 |
|
|
last S2; |
2741 |
|
|
} else { |
2742 |
|
|
## Step 3 |
2743 |
|
|
if (not $formatting_category->{$node->[1]} and |
2744 |
|
|
#not $phrasing_category->{$node->[1]} and |
2745 |
|
|
($special_category->{$node->[1]} or |
2746 |
|
|
$scoping_category->{$node->[1]})) { |
2747 |
|
|
!!!parse-error; |
2748 |
|
|
## Ignore the token |
2749 |
|
|
!!!next-token; |
2750 |
|
|
last S2; |
2751 |
|
|
} |
2752 |
|
|
} |
2753 |
|
|
|
2754 |
|
|
## Step 4 |
2755 |
|
|
$node_i--; |
2756 |
|
|
$node = $open_elements->[$node_i]; |
2757 |
|
|
|
2758 |
|
|
## Step 5; |
2759 |
|
|
redo S2; |
2760 |
|
|
} # S2 |
2761 |
|
|
} |
2762 |
|
|
} |
2763 |
|
|
}; # $in_body |
2764 |
|
|
|
2765 |
|
|
B: { |
2766 |
|
|
if ($phase eq 'initial') { |
2767 |
|
|
if ($token->{type} eq 'DOCTYPE') { |
2768 |
|
|
if ($token->{error}) { |
2769 |
|
|
## ISSUE: Spec currently left this case undefined. |
2770 |
wakaba |
1.6 |
!!!parse-error ('missing DOCTYPE'); |
2771 |
wakaba |
1.2 |
} |
2772 |
|
|
my $doctype = $self->{document}->create_document_type_definition |
2773 |
|
|
($token->{name}); |
2774 |
|
|
$self->{document}->append_child ($doctype); |
2775 |
|
|
$phase = 'root element'; |
2776 |
|
|
!!!next-token; |
2777 |
|
|
redo B; |
2778 |
|
|
} elsif ({ |
2779 |
|
|
comment => 1, |
2780 |
|
|
'start tag' => 1, |
2781 |
|
|
'end tag' => 1, |
2782 |
|
|
'end-of-file' => 1, |
2783 |
|
|
}->{$token->{type}}) { |
2784 |
|
|
## ISSUE: Spec currently left this case undefined. |
2785 |
wakaba |
1.6 |
!!!parse-error ('missing DOCTYPE'); |
2786 |
wakaba |
1.2 |
$phase = 'root element'; |
2787 |
|
|
## reprocess |
2788 |
|
|
redo B; |
2789 |
|
|
} elsif ($token->{type} eq 'character') { |
2790 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
2791 |
|
|
$self->{document}->manakai_append_text ($1); |
2792 |
|
|
## ISSUE: DOM3 Core does not allow Document > Text |
2793 |
|
|
unless (length $token->{data}) { |
2794 |
|
|
## Stay in the phase |
2795 |
|
|
!!!next-token; |
2796 |
|
|
redo B; |
2797 |
|
|
} |
2798 |
|
|
} |
2799 |
|
|
## ISSUE: Spec currently left this case undefined. |
2800 |
wakaba |
1.6 |
!!!parse-error ('missing DOCTYPE'); |
2801 |
wakaba |
1.2 |
$phase = 'root element'; |
2802 |
|
|
## reprocess |
2803 |
|
|
redo B; |
2804 |
|
|
} else { |
2805 |
|
|
die "$0: $token->{type}: Unknown token"; |
2806 |
|
|
} |
2807 |
|
|
} elsif ($phase eq 'root element') { |
2808 |
|
|
if ($token->{type} eq 'DOCTYPE') { |
2809 |
|
|
!!!parse-error; |
2810 |
|
|
## Ignore the token |
2811 |
|
|
## Stay in the phase |
2812 |
|
|
!!!next-token; |
2813 |
|
|
redo B; |
2814 |
|
|
} elsif ($token->{type} eq 'comment') { |
2815 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
2816 |
|
|
$self->{document}->append_child ($comment); |
2817 |
|
|
## Stay in the phase |
2818 |
|
|
!!!next-token; |
2819 |
|
|
redo B; |
2820 |
|
|
} elsif ($token->{type} eq 'character') { |
2821 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
2822 |
|
|
$self->{document}->manakai_append_text ($1); |
2823 |
|
|
## ISSUE: DOM3 Core does not allow Document > Text |
2824 |
|
|
unless (length $token->{data}) { |
2825 |
|
|
## Stay in the phase |
2826 |
|
|
!!!next-token; |
2827 |
|
|
redo B; |
2828 |
|
|
} |
2829 |
|
|
} |
2830 |
|
|
# |
2831 |
|
|
} elsif ({ |
2832 |
|
|
'start tag' => 1, |
2833 |
|
|
'end tag' => 1, |
2834 |
|
|
'end-of-file' => 1, |
2835 |
|
|
}->{$token->{type}}) { |
2836 |
|
|
## ISSUE: There is an issue in the spec |
2837 |
|
|
# |
2838 |
|
|
} else { |
2839 |
|
|
die "$0: $token->{type}: Unknown token"; |
2840 |
|
|
} |
2841 |
|
|
my $root_element; !!!create-element ($root_element, 'html'); |
2842 |
|
|
$self->{document}->append_child ($root_element); |
2843 |
|
|
$open_elements = [[$root_element, 'html']]; |
2844 |
|
|
$phase = 'main'; |
2845 |
|
|
## reprocess |
2846 |
|
|
redo B; |
2847 |
|
|
} elsif ($phase eq 'main') { |
2848 |
|
|
if ($token->{type} eq 'DOCTYPE') { |
2849 |
|
|
!!!parse-error; |
2850 |
|
|
## Ignore the token |
2851 |
|
|
## Stay in the phase |
2852 |
|
|
!!!next-token; |
2853 |
|
|
redo B; |
2854 |
|
|
} elsif ($token->{type} eq 'start tag' and |
2855 |
|
|
$token->{tag_name} eq 'html') { |
2856 |
|
|
## TODO: unless it is the first start tag token, parse-error |
2857 |
|
|
my $top_el = $open_elements->[0]->[0]; |
2858 |
|
|
for my $attr_name (keys %{$token->{attributes}}) { |
2859 |
|
|
unless ($top_el->has_attribute_ns (undef, $attr_name)) { |
2860 |
wakaba |
1.7 |
$top_el->set_attribute_ns |
2861 |
|
|
(undef, [undef, $attr_name], |
2862 |
|
|
$token->{attributes}->{$attr_name}->{value}); |
2863 |
wakaba |
1.2 |
} |
2864 |
|
|
} |
2865 |
|
|
!!!next-token; |
2866 |
|
|
redo B; |
2867 |
|
|
} elsif ($token->{type} eq 'end-of-file') { |
2868 |
|
|
## Generate implied end tags |
2869 |
|
|
if ({ |
2870 |
|
|
dd => 1, dt => 1, li => 1, p => 1, td => 1, th => 1, tr => 1, |
2871 |
|
|
}->{$open_elements->[-1]->[1]}) { |
2872 |
|
|
!!!back-token; |
2873 |
|
|
$token = {type => 'end tag', tag_name => $open_elements->[-1]->[1]}; |
2874 |
|
|
redo B; |
2875 |
|
|
} |
2876 |
|
|
|
2877 |
|
|
if (@$open_elements > 2 or |
2878 |
|
|
(@$open_elements == 2 and $open_elements->[1]->[1] ne 'body')) { |
2879 |
|
|
!!!parse-error; |
2880 |
|
|
} else { |
2881 |
|
|
## TODO: inner_html parser and @$open_elements > 1 and $open_elements->[1] ne 'body', then parse-error |
2882 |
|
|
} |
2883 |
|
|
|
2884 |
|
|
## Stop parsing |
2885 |
|
|
last B; |
2886 |
|
|
|
2887 |
|
|
## ISSUE: There is an issue in the spec. |
2888 |
|
|
} else { |
2889 |
|
|
if ($insertion_mode eq 'before head') { |
2890 |
|
|
if ($token->{type} eq 'character') { |
2891 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
2892 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
2893 |
|
|
unless (length $token->{data}) { |
2894 |
|
|
!!!next-token; |
2895 |
|
|
redo B; |
2896 |
|
|
} |
2897 |
|
|
} |
2898 |
|
|
## As if <head> |
2899 |
|
|
!!!create-element ($head_element, 'head'); |
2900 |
|
|
$open_elements->[-1]->[0]->append_child ($head_element); |
2901 |
|
|
push @$open_elements, [$head_element, 'head']; |
2902 |
|
|
$insertion_mode = 'in head'; |
2903 |
|
|
## reprocess |
2904 |
|
|
redo B; |
2905 |
|
|
} elsif ($token->{type} eq 'comment') { |
2906 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
2907 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
2908 |
|
|
!!!next-token; |
2909 |
|
|
redo B; |
2910 |
|
|
} elsif ($token->{type} eq 'start tag') { |
2911 |
|
|
my $attr = $token->{tag_name} eq 'head' ? $token->{attributes} : {}; |
2912 |
|
|
!!!create-element ($head_element, 'head', $attr); |
2913 |
|
|
$open_elements->[-1]->[0]->append_child ($head_element); |
2914 |
|
|
push @$open_elements, [$head_element, 'head']; |
2915 |
|
|
$insertion_mode = 'in head'; |
2916 |
|
|
if ($token->{tag_name} eq 'head') { |
2917 |
|
|
!!!next-token; |
2918 |
|
|
#} elsif ({ |
2919 |
|
|
# base => 1, link => 1, meta => 1, |
2920 |
|
|
# script => 1, style => 1, title => 1, |
2921 |
|
|
# }->{$token->{tag_name}}) { |
2922 |
|
|
# ## reprocess |
2923 |
|
|
} else { |
2924 |
|
|
## reprocess |
2925 |
|
|
} |
2926 |
|
|
redo B; |
2927 |
|
|
} elsif ($token->{type} eq 'end tag') { |
2928 |
|
|
if ($token->{tag_name} eq 'html') { |
2929 |
|
|
## As if <head> |
2930 |
|
|
!!!create-element ($head_element, 'head'); |
2931 |
|
|
$open_elements->[-1]->[0]->append_child ($head_element); |
2932 |
|
|
push @$open_elements, [$head_element, 'head']; |
2933 |
|
|
$insertion_mode = 'in head'; |
2934 |
|
|
## reprocess |
2935 |
|
|
redo B; |
2936 |
|
|
} else { |
2937 |
|
|
!!!parse-error; |
2938 |
|
|
## Ignore the token |
2939 |
wakaba |
1.6 |
!!!next-token; |
2940 |
wakaba |
1.2 |
redo B; |
2941 |
|
|
} |
2942 |
|
|
} else { |
2943 |
|
|
die "$0: $token->{type}: Unknown type"; |
2944 |
|
|
} |
2945 |
|
|
} elsif ($insertion_mode eq 'in head') { |
2946 |
|
|
if ($token->{type} eq 'character') { |
2947 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
2948 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
2949 |
|
|
unless (length $token->{data}) { |
2950 |
|
|
!!!next-token; |
2951 |
|
|
redo B; |
2952 |
|
|
} |
2953 |
|
|
} |
2954 |
|
|
|
2955 |
|
|
# |
2956 |
|
|
} elsif ($token->{type} eq 'comment') { |
2957 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
2958 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
2959 |
|
|
!!!next-token; |
2960 |
|
|
redo B; |
2961 |
|
|
} elsif ($token->{type} eq 'start tag') { |
2962 |
|
|
if ($token->{tag_name} eq 'title') { |
2963 |
wakaba |
1.7 |
## NOTE: There is an "as if in head" code clone |
2964 |
|
|
my $title_el; |
2965 |
|
|
!!!create-element ($title_el, 'title', $token->{attributes}); |
2966 |
wakaba |
1.2 |
(defined $head_element ? $head_element : $open_elements->[-1]->[0]) |
2967 |
|
|
->append_child ($title_el); |
2968 |
|
|
$self->{content_model_flag} = 'RCDATA'; |
2969 |
wakaba |
1.7 |
|
2970 |
wakaba |
1.2 |
my $text = ''; |
2971 |
|
|
!!!next-token; |
2972 |
|
|
while ($token->{type} eq 'character') { |
2973 |
|
|
$text .= $token->{data}; |
2974 |
|
|
!!!next-token; |
2975 |
|
|
} |
2976 |
|
|
if (length $text) { |
2977 |
|
|
$title_el->manakai_append_text ($text); |
2978 |
|
|
} |
2979 |
|
|
|
2980 |
|
|
$self->{content_model_flag} = 'PCDATA'; |
2981 |
|
|
|
2982 |
|
|
if ($token->{type} eq 'end tag' and |
2983 |
|
|
$token->{tag_name} eq 'title') { |
2984 |
|
|
## Ignore the token |
2985 |
|
|
} else { |
2986 |
|
|
!!!parse-error; |
2987 |
|
|
## ISSUE: And ignore? |
2988 |
|
|
} |
2989 |
|
|
!!!next-token; |
2990 |
|
|
redo B; |
2991 |
|
|
} elsif ($token->{tag_name} eq 'style') { |
2992 |
|
|
$style_start_tag->(); |
2993 |
|
|
redo B; |
2994 |
|
|
} elsif ($token->{tag_name} eq 'script') { |
2995 |
|
|
$script_start_tag->(); |
2996 |
|
|
redo B; |
2997 |
|
|
} elsif ({base => 1, link => 1, meta => 1}->{$token->{tag_name}}) { |
2998 |
|
|
## NOTE: There are "as if in head" code clones |
2999 |
|
|
my $el; |
3000 |
|
|
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
3001 |
|
|
(defined $head_element ? $head_element : $open_elements->[-1]->[0]) |
3002 |
|
|
->append_child ($el); |
3003 |
|
|
|
3004 |
|
|
!!!next-token; |
3005 |
|
|
redo B; |
3006 |
|
|
} elsif ($token->{tag_name} eq 'head') { |
3007 |
|
|
!!!parse-error; |
3008 |
|
|
## Ignore the token |
3009 |
|
|
!!!next-token; |
3010 |
|
|
redo B; |
3011 |
|
|
} else { |
3012 |
|
|
# |
3013 |
|
|
} |
3014 |
|
|
} elsif ($token->{type} eq 'end tag') { |
3015 |
|
|
if ($token->{tag_name} eq 'head') { |
3016 |
|
|
if ($open_elements->[-1]->[1] eq 'head') { |
3017 |
|
|
pop @$open_elements; |
3018 |
|
|
} else { |
3019 |
|
|
!!!parse-error; |
3020 |
|
|
} |
3021 |
|
|
$insertion_mode = 'after head'; |
3022 |
|
|
!!!next-token; |
3023 |
|
|
redo B; |
3024 |
|
|
} elsif ($token->{tag_name} eq 'html') { |
3025 |
|
|
# |
3026 |
|
|
} else { |
3027 |
|
|
!!!parse-error; |
3028 |
|
|
## Ignore the token |
3029 |
|
|
!!!next-token; |
3030 |
|
|
redo B; |
3031 |
|
|
} |
3032 |
|
|
} else { |
3033 |
|
|
# |
3034 |
|
|
} |
3035 |
|
|
|
3036 |
|
|
if ($open_elements->[-1]->[1] eq 'head') { |
3037 |
|
|
## As if </head> |
3038 |
|
|
pop @$open_elements; |
3039 |
|
|
} |
3040 |
|
|
$insertion_mode = 'after head'; |
3041 |
|
|
## reprocess |
3042 |
|
|
redo B; |
3043 |
|
|
|
3044 |
|
|
## ISSUE: An issue in the spec. |
3045 |
|
|
} elsif ($insertion_mode eq 'after head') { |
3046 |
|
|
if ($token->{type} eq 'character') { |
3047 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
3048 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
3049 |
|
|
unless (length $token->{data}) { |
3050 |
|
|
!!!next-token; |
3051 |
|
|
redo B; |
3052 |
|
|
} |
3053 |
|
|
} |
3054 |
|
|
|
3055 |
|
|
# |
3056 |
|
|
} elsif ($token->{type} eq 'comment') { |
3057 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3058 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3059 |
|
|
!!!next-token; |
3060 |
|
|
redo B; |
3061 |
|
|
} elsif ($token->{type} eq 'start tag') { |
3062 |
|
|
if ($token->{tag_name} eq 'body') { |
3063 |
|
|
!!!insert-element ('body', $token->{attributes}); |
3064 |
|
|
$insertion_mode = 'in body'; |
3065 |
|
|
!!!next-token; |
3066 |
|
|
redo B; |
3067 |
|
|
} elsif ($token->{tag_name} eq 'frameset') { |
3068 |
|
|
!!!insert-element ('frameset', $token->{attributes}); |
3069 |
|
|
$insertion_mode = 'in frameset'; |
3070 |
|
|
!!!next-token; |
3071 |
|
|
redo B; |
3072 |
|
|
} elsif ({ |
3073 |
|
|
base => 1, link => 1, meta => 1, |
3074 |
|
|
script=> 1, style => 1, title => 1, |
3075 |
|
|
}->{$token->{tag_name}}) { |
3076 |
|
|
!!!parse-error; |
3077 |
|
|
$insertion_mode = 'in head'; |
3078 |
|
|
## reprocess |
3079 |
|
|
redo B; |
3080 |
|
|
} else { |
3081 |
|
|
# |
3082 |
|
|
} |
3083 |
|
|
} else { |
3084 |
|
|
# |
3085 |
|
|
} |
3086 |
|
|
|
3087 |
|
|
## As if <body> |
3088 |
|
|
!!!insert-element ('body'); |
3089 |
|
|
$insertion_mode = 'in body'; |
3090 |
|
|
## reprocess |
3091 |
|
|
redo B; |
3092 |
|
|
} elsif ($insertion_mode eq 'in body') { |
3093 |
|
|
if ($token->{type} eq 'character') { |
3094 |
|
|
## NOTE: There is a code clone of "character in body". |
3095 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
3096 |
wakaba |
1.2 |
|
3097 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
3098 |
|
|
|
3099 |
|
|
!!!next-token; |
3100 |
|
|
redo B; |
3101 |
|
|
} elsif ($token->{type} eq 'comment') { |
3102 |
|
|
## NOTE: There is a code clone of "comment in body". |
3103 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3104 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3105 |
|
|
!!!next-token; |
3106 |
|
|
redo B; |
3107 |
|
|
} else { |
3108 |
wakaba |
1.7 |
$in_body->($insert_to_current); |
3109 |
wakaba |
1.2 |
redo B; |
3110 |
|
|
} |
3111 |
|
|
} elsif ($insertion_mode eq 'in table') { |
3112 |
|
|
if ($token->{type} eq 'character') { |
3113 |
wakaba |
1.7 |
## NOTE: There are "character in table" code clones. |
3114 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
3115 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
3116 |
|
|
|
3117 |
|
|
unless (length $token->{data}) { |
3118 |
|
|
!!!next-token; |
3119 |
|
|
redo B; |
3120 |
|
|
} |
3121 |
|
|
} |
3122 |
wakaba |
1.2 |
|
3123 |
wakaba |
1.7 |
## As if in body, but insert into foster parent element |
3124 |
|
|
## ISSUE: Spec says that "whenever a node would be inserted |
3125 |
|
|
## into the current node" while characters might not be |
3126 |
|
|
## result in a new Text node. |
3127 |
|
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
3128 |
|
|
|
3129 |
|
|
if ({ |
3130 |
|
|
table => 1, tbody => 1, tfoot => 1, |
3131 |
|
|
thead => 1, tr => 1, |
3132 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3133 |
|
|
# MUST |
3134 |
|
|
my $foster_parent_element; |
3135 |
|
|
my $next_sibling; |
3136 |
|
|
my $prev_sibling; |
3137 |
|
|
OE: for (reverse 0..$#$open_elements) { |
3138 |
|
|
if ($open_elements->[$_]->[1] eq 'table') { |
3139 |
|
|
my $parent = $open_elements->[$_]->[0]->parent_node; |
3140 |
|
|
if (defined $parent and $parent->node_type == 1) { |
3141 |
|
|
$foster_parent_element = $parent; |
3142 |
|
|
$next_sibling = $open_elements->[$_]->[0]; |
3143 |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
3144 |
|
|
} else { |
3145 |
|
|
$foster_parent_element = $open_elements->[$_ - 1]->[0]; |
3146 |
|
|
$prev_sibling = $foster_parent_element->last_child; |
3147 |
|
|
} |
3148 |
|
|
last OE; |
3149 |
|
|
} |
3150 |
|
|
} # OE |
3151 |
|
|
$foster_parent_element = $open_elements->[0]->[0] and |
3152 |
|
|
$prev_sibling = $foster_parent_element->last_child |
3153 |
|
|
unless defined $foster_parent_element; |
3154 |
|
|
if (defined $prev_sibling and |
3155 |
|
|
$prev_sibling->node_type == 3) { |
3156 |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
3157 |
|
|
} else { |
3158 |
|
|
$foster_parent_element->insert_before |
3159 |
|
|
($self->{document}->create_text_node ($token->{data}), |
3160 |
|
|
$next_sibling); |
3161 |
|
|
} |
3162 |
|
|
} else { |
3163 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
3164 |
|
|
} |
3165 |
|
|
|
3166 |
wakaba |
1.2 |
!!!next-token; |
3167 |
|
|
redo B; |
3168 |
|
|
} elsif ($token->{type} eq 'comment') { |
3169 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3170 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3171 |
|
|
!!!next-token; |
3172 |
|
|
redo B; |
3173 |
|
|
} elsif ($token->{type} eq 'start tag') { |
3174 |
|
|
if ({ |
3175 |
|
|
caption => 1, |
3176 |
|
|
colgroup => 1, |
3177 |
|
|
tbody => 1, tfoot => 1, thead => 1, |
3178 |
|
|
}->{$token->{tag_name}}) { |
3179 |
|
|
## Clear back to table context |
3180 |
|
|
while ($open_elements->[-1]->[1] ne 'table' and |
3181 |
|
|
$open_elements->[-1]->[1] ne 'html') { |
3182 |
|
|
!!!parse-error; |
3183 |
|
|
pop @$open_elements; |
3184 |
|
|
} |
3185 |
|
|
|
3186 |
|
|
push @$active_formatting_elements, ['#marker', ''] |
3187 |
|
|
if $token->{tag_name} eq 'caption'; |
3188 |
|
|
|
3189 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
3190 |
|
|
$insertion_mode = { |
3191 |
|
|
caption => 'in caption', |
3192 |
|
|
colgroup => 'in column group', |
3193 |
|
|
tbody => 'in table body', |
3194 |
|
|
tfoot => 'in table body', |
3195 |
|
|
thead => 'in table body', |
3196 |
|
|
}->{$token->{tag_name}}; |
3197 |
|
|
!!!next-token; |
3198 |
|
|
redo B; |
3199 |
|
|
} elsif ({ |
3200 |
|
|
col => 1, |
3201 |
|
|
td => 1, th => 1, tr => 1, |
3202 |
|
|
}->{$token->{tag_name}}) { |
3203 |
|
|
## Clear back to table context |
3204 |
|
|
while ($open_elements->[-1]->[1] ne 'table' and |
3205 |
|
|
$open_elements->[-1]->[1] ne 'html') { |
3206 |
|
|
!!!parse-error; |
3207 |
|
|
pop @$open_elements; |
3208 |
|
|
} |
3209 |
|
|
|
3210 |
|
|
!!!insert-element ($token->{tag_name} eq 'col' ? 'colgroup' : 'tbody'); |
3211 |
|
|
$insertion_mode = $token->{tag_name} eq 'col' |
3212 |
|
|
? 'in column group' : 'in table body'; |
3213 |
|
|
## reprocess |
3214 |
|
|
redo B; |
3215 |
|
|
} elsif ($token->{tag_name} eq 'table') { |
3216 |
|
|
## NOTE: There are code clones for this "table in table" |
3217 |
|
|
!!!parse-error; |
3218 |
|
|
|
3219 |
|
|
## As if </table> |
3220 |
|
|
## have a table element in table scope |
3221 |
|
|
my $i; |
3222 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3223 |
|
|
my $node = $open_elements->[$_]; |
3224 |
|
|
if ($node->[1] eq 'table') { |
3225 |
|
|
$i = $_; |
3226 |
|
|
last INSCOPE; |
3227 |
|
|
} elsif ({ |
3228 |
|
|
table => 1, html => 1, |
3229 |
|
|
}->{$node->[1]}) { |
3230 |
|
|
last INSCOPE; |
3231 |
|
|
} |
3232 |
|
|
} # INSCOPE |
3233 |
|
|
unless (defined $i) { |
3234 |
|
|
!!!parse-error; |
3235 |
|
|
## Ignore tokens </table><table> |
3236 |
|
|
!!!next-token; |
3237 |
|
|
redo B; |
3238 |
|
|
} |
3239 |
|
|
|
3240 |
|
|
## generate implied end tags |
3241 |
|
|
if ({ |
3242 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
3243 |
|
|
td => 1, th => 1, tr => 1, |
3244 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3245 |
|
|
!!!back-token; # <table> |
3246 |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
3247 |
|
|
!!!back-token; |
3248 |
|
|
$token = {type => 'end tag', |
3249 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
3250 |
|
|
redo B; |
3251 |
|
|
} |
3252 |
|
|
|
3253 |
|
|
if ($open_elements->[-1]->[1] ne 'table') { |
3254 |
|
|
!!!parse-error; |
3255 |
|
|
} |
3256 |
|
|
|
3257 |
|
|
splice @$open_elements, $i; |
3258 |
|
|
|
3259 |
|
|
$reset_insertion_mode->(); |
3260 |
|
|
|
3261 |
|
|
## reprocess |
3262 |
|
|
redo B; |
3263 |
|
|
} else { |
3264 |
|
|
# |
3265 |
|
|
} |
3266 |
|
|
} elsif ($token->{type} eq 'end tag') { |
3267 |
|
|
if ($token->{tag_name} eq 'table') { |
3268 |
|
|
## have a table element in table scope |
3269 |
|
|
my $i; |
3270 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3271 |
|
|
my $node = $open_elements->[$_]; |
3272 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
3273 |
|
|
$i = $_; |
3274 |
|
|
last INSCOPE; |
3275 |
|
|
} elsif ({ |
3276 |
|
|
table => 1, html => 1, |
3277 |
|
|
}->{$node->[1]}) { |
3278 |
|
|
last INSCOPE; |
3279 |
|
|
} |
3280 |
|
|
} # INSCOPE |
3281 |
|
|
unless (defined $i) { |
3282 |
|
|
!!!parse-error; |
3283 |
|
|
## Ignore the token |
3284 |
|
|
!!!next-token; |
3285 |
|
|
redo B; |
3286 |
|
|
} |
3287 |
|
|
|
3288 |
|
|
## generate implied end tags |
3289 |
|
|
if ({ |
3290 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
3291 |
|
|
td => 1, th => 1, tr => 1, |
3292 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3293 |
|
|
!!!back-token; |
3294 |
|
|
$token = {type => 'end tag', |
3295 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
3296 |
|
|
redo B; |
3297 |
|
|
} |
3298 |
|
|
|
3299 |
|
|
if ($open_elements->[-1]->[1] ne 'table') { |
3300 |
|
|
!!!parse-error; |
3301 |
|
|
} |
3302 |
|
|
|
3303 |
|
|
splice @$open_elements, $i; |
3304 |
|
|
|
3305 |
|
|
$reset_insertion_mode->(); |
3306 |
|
|
|
3307 |
|
|
!!!next-token; |
3308 |
|
|
redo B; |
3309 |
|
|
} elsif ({ |
3310 |
|
|
body => 1, caption => 1, col => 1, colgroup => 1, |
3311 |
|
|
html => 1, tbody => 1, td => 1, tfoot => 1, th => 1, |
3312 |
|
|
thead => 1, tr => 1, |
3313 |
|
|
}->{$token->{tag_name}}) { |
3314 |
|
|
!!!parse-error; |
3315 |
|
|
## Ignore the token |
3316 |
|
|
!!!next-token; |
3317 |
|
|
redo B; |
3318 |
|
|
} else { |
3319 |
|
|
# |
3320 |
|
|
} |
3321 |
|
|
} else { |
3322 |
|
|
# |
3323 |
|
|
} |
3324 |
|
|
|
3325 |
|
|
!!!parse-error; |
3326 |
wakaba |
1.7 |
$in_body->($insert_to_foster); |
3327 |
wakaba |
1.2 |
redo B; |
3328 |
|
|
} elsif ($insertion_mode eq 'in caption') { |
3329 |
wakaba |
1.6 |
if ($token->{type} eq 'character') { |
3330 |
|
|
## NOTE: This is a code clone of "character in body". |
3331 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
3332 |
wakaba |
1.6 |
|
3333 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
3334 |
|
|
|
3335 |
|
|
!!!next-token; |
3336 |
|
|
redo B; |
3337 |
|
|
} elsif ($token->{type} eq 'comment') { |
3338 |
|
|
## NOTE: This is a code clone of "comment in body". |
3339 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3340 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3341 |
|
|
!!!next-token; |
3342 |
|
|
redo B; |
3343 |
|
|
} elsif ($token->{type} eq 'start tag') { |
3344 |
wakaba |
1.2 |
if ({ |
3345 |
|
|
caption => 1, col => 1, colgroup => 1, tbody => 1, |
3346 |
|
|
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
3347 |
|
|
}->{$token->{tag_name}}) { |
3348 |
|
|
!!!parse-error; |
3349 |
|
|
|
3350 |
|
|
## As if </caption> |
3351 |
|
|
## have a table element in table scope |
3352 |
|
|
my $i; |
3353 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3354 |
|
|
my $node = $open_elements->[$_]; |
3355 |
|
|
if ($node->[1] eq 'caption') { |
3356 |
|
|
$i = $_; |
3357 |
|
|
last INSCOPE; |
3358 |
|
|
} elsif ({ |
3359 |
|
|
table => 1, html => 1, |
3360 |
|
|
}->{$node->[1]}) { |
3361 |
|
|
last INSCOPE; |
3362 |
|
|
} |
3363 |
|
|
} # INSCOPE |
3364 |
|
|
unless (defined $i) { |
3365 |
|
|
!!!parse-error; |
3366 |
|
|
## Ignore the token |
3367 |
|
|
!!!next-token; |
3368 |
|
|
redo B; |
3369 |
|
|
} |
3370 |
|
|
|
3371 |
|
|
## generate implied end tags |
3372 |
|
|
if ({ |
3373 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
3374 |
|
|
td => 1, th => 1, tr => 1, |
3375 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3376 |
|
|
!!!back-token; # <?> |
3377 |
|
|
$token = {type => 'end tag', tag_name => 'caption'}; |
3378 |
|
|
!!!back-token; |
3379 |
|
|
$token = {type => 'end tag', |
3380 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
3381 |
|
|
redo B; |
3382 |
|
|
} |
3383 |
|
|
|
3384 |
|
|
if ($open_elements->[-1]->[1] ne 'caption') { |
3385 |
|
|
!!!parse-error; |
3386 |
|
|
} |
3387 |
|
|
|
3388 |
|
|
splice @$open_elements, $i; |
3389 |
|
|
|
3390 |
|
|
$clear_up_to_marker->(); |
3391 |
|
|
|
3392 |
|
|
$insertion_mode = 'in table'; |
3393 |
|
|
|
3394 |
|
|
## reprocess |
3395 |
|
|
redo B; |
3396 |
|
|
} else { |
3397 |
|
|
# |
3398 |
|
|
} |
3399 |
|
|
} elsif ($token->{type} eq 'end tag') { |
3400 |
|
|
if ($token->{tag_name} eq 'caption') { |
3401 |
|
|
## have a table element in table scope |
3402 |
|
|
my $i; |
3403 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3404 |
|
|
my $node = $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; |
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 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3426 |
|
|
!!!back-token; |
3427 |
|
|
$token = {type => 'end tag', |
3428 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
3429 |
|
|
redo B; |
3430 |
|
|
} |
3431 |
|
|
|
3432 |
|
|
if ($open_elements->[-1]->[1] ne 'caption') { |
3433 |
|
|
!!!parse-error; |
3434 |
|
|
} |
3435 |
|
|
|
3436 |
|
|
splice @$open_elements, $i; |
3437 |
|
|
|
3438 |
|
|
$clear_up_to_marker->(); |
3439 |
|
|
|
3440 |
|
|
$insertion_mode = 'in table'; |
3441 |
|
|
|
3442 |
|
|
!!!next-token; |
3443 |
|
|
redo B; |
3444 |
|
|
} elsif ($token->{tag_name} eq 'table') { |
3445 |
|
|
!!!parse-error; |
3446 |
|
|
|
3447 |
|
|
## As if </caption> |
3448 |
|
|
## have a table element in table scope |
3449 |
|
|
my $i; |
3450 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3451 |
|
|
my $node = $open_elements->[$_]; |
3452 |
|
|
if ($node->[1] eq 'caption') { |
3453 |
|
|
$i = $_; |
3454 |
|
|
last INSCOPE; |
3455 |
|
|
} elsif ({ |
3456 |
|
|
table => 1, html => 1, |
3457 |
|
|
}->{$node->[1]}) { |
3458 |
|
|
last INSCOPE; |
3459 |
|
|
} |
3460 |
|
|
} # INSCOPE |
3461 |
|
|
unless (defined $i) { |
3462 |
|
|
!!!parse-error; |
3463 |
|
|
## Ignore the token |
3464 |
|
|
!!!next-token; |
3465 |
|
|
redo B; |
3466 |
|
|
} |
3467 |
|
|
|
3468 |
|
|
## generate implied end tags |
3469 |
|
|
if ({ |
3470 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
3471 |
|
|
td => 1, th => 1, tr => 1, |
3472 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3473 |
|
|
!!!back-token; # </table> |
3474 |
|
|
$token = {type => 'end tag', tag_name => 'caption'}; |
3475 |
|
|
!!!back-token; |
3476 |
|
|
$token = {type => 'end tag', |
3477 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
3478 |
|
|
redo B; |
3479 |
|
|
} |
3480 |
|
|
|
3481 |
|
|
if ($open_elements->[-1]->[1] ne 'caption') { |
3482 |
|
|
!!!parse-error; |
3483 |
|
|
} |
3484 |
|
|
|
3485 |
|
|
splice @$open_elements, $i; |
3486 |
|
|
|
3487 |
|
|
$clear_up_to_marker->(); |
3488 |
|
|
|
3489 |
|
|
$insertion_mode = 'in table'; |
3490 |
|
|
|
3491 |
|
|
## reprocess |
3492 |
|
|
redo B; |
3493 |
|
|
} elsif ({ |
3494 |
|
|
body => 1, col => 1, colgroup => 1, |
3495 |
|
|
html => 1, tbody => 1, td => 1, tfoot => 1, |
3496 |
|
|
th => 1, thead => 1, tr => 1, |
3497 |
|
|
}->{$token->{tag_name}}) { |
3498 |
|
|
!!!parse-error; |
3499 |
|
|
## Ignore the token |
3500 |
|
|
redo B; |
3501 |
|
|
} else { |
3502 |
|
|
# |
3503 |
|
|
} |
3504 |
|
|
} else { |
3505 |
|
|
# |
3506 |
|
|
} |
3507 |
|
|
|
3508 |
wakaba |
1.7 |
$in_body->($insert_to_current); |
3509 |
wakaba |
1.2 |
redo B; |
3510 |
|
|
} elsif ($insertion_mode eq 'in column group') { |
3511 |
|
|
if ($token->{type} eq 'character') { |
3512 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
3513 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
3514 |
|
|
unless (length $token->{data}) { |
3515 |
|
|
!!!next-token; |
3516 |
|
|
redo B; |
3517 |
|
|
} |
3518 |
|
|
} |
3519 |
|
|
|
3520 |
|
|
# |
3521 |
|
|
} elsif ($token->{type} eq 'comment') { |
3522 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3523 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3524 |
|
|
!!!next-token; |
3525 |
|
|
redo B; |
3526 |
|
|
} elsif ($token->{type} eq 'start tag') { |
3527 |
|
|
if ($token->{tag_name} eq 'col') { |
3528 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
3529 |
|
|
pop @$open_elements; |
3530 |
|
|
!!!next-token; |
3531 |
|
|
redo B; |
3532 |
|
|
} else { |
3533 |
|
|
# |
3534 |
|
|
} |
3535 |
|
|
} elsif ($token->{type} eq 'end tag') { |
3536 |
|
|
if ($token->{tag_name} eq 'colgroup') { |
3537 |
|
|
if ($open_elements->[-1]->[1] eq 'html') { |
3538 |
|
|
!!!parse-error; |
3539 |
|
|
## Ignore the token |
3540 |
|
|
!!!next-token; |
3541 |
|
|
redo B; |
3542 |
|
|
} else { |
3543 |
|
|
pop @$open_elements; # colgroup |
3544 |
|
|
$insertion_mode = 'in table'; |
3545 |
|
|
!!!next-token; |
3546 |
|
|
redo B; |
3547 |
|
|
} |
3548 |
|
|
} elsif ($token->{tag_name} eq 'col') { |
3549 |
|
|
!!!parse-error; |
3550 |
|
|
## Ignore the token |
3551 |
|
|
!!!next-token; |
3552 |
|
|
redo B; |
3553 |
|
|
} else { |
3554 |
|
|
# |
3555 |
|
|
} |
3556 |
|
|
} else { |
3557 |
|
|
# |
3558 |
|
|
} |
3559 |
|
|
|
3560 |
|
|
## As if </colgroup> |
3561 |
|
|
if ($open_elements->[-1]->[1] eq 'html') { |
3562 |
|
|
!!!parse-error; |
3563 |
|
|
## Ignore the token |
3564 |
|
|
!!!next-token; |
3565 |
|
|
redo B; |
3566 |
|
|
} else { |
3567 |
|
|
pop @$open_elements; # colgroup |
3568 |
|
|
$insertion_mode = 'in table'; |
3569 |
|
|
## reprocess |
3570 |
|
|
redo B; |
3571 |
|
|
} |
3572 |
|
|
} elsif ($insertion_mode eq 'in table body') { |
3573 |
|
|
if ($token->{type} eq 'character') { |
3574 |
wakaba |
1.7 |
## NOTE: This is a "character in table" code clone. |
3575 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
3576 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
3577 |
|
|
|
3578 |
|
|
unless (length $token->{data}) { |
3579 |
|
|
!!!next-token; |
3580 |
|
|
redo B; |
3581 |
|
|
} |
3582 |
|
|
} |
3583 |
wakaba |
1.2 |
|
3584 |
wakaba |
1.7 |
## As if in body, but insert into foster parent element |
3585 |
|
|
## ISSUE: Spec says that "whenever a node would be inserted |
3586 |
|
|
## into the current node" while characters might not be |
3587 |
|
|
## result in a new Text node. |
3588 |
|
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
3589 |
wakaba |
1.2 |
|
3590 |
wakaba |
1.7 |
if ({ |
3591 |
|
|
table => 1, tbody => 1, tfoot => 1, |
3592 |
|
|
thead => 1, tr => 1, |
3593 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3594 |
|
|
# MUST |
3595 |
|
|
my $foster_parent_element; |
3596 |
|
|
my $next_sibling; |
3597 |
|
|
my $prev_sibling; |
3598 |
|
|
OE: for (reverse 0..$#$open_elements) { |
3599 |
|
|
if ($open_elements->[$_]->[1] eq 'table') { |
3600 |
|
|
my $parent = $open_elements->[$_]->[0]->parent_node; |
3601 |
|
|
if (defined $parent and $parent->node_type == 1) { |
3602 |
|
|
$foster_parent_element = $parent; |
3603 |
|
|
$next_sibling = $open_elements->[$_]->[0]; |
3604 |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
3605 |
|
|
} else { |
3606 |
|
|
$foster_parent_element = $open_elements->[$_ - 1]->[0]; |
3607 |
|
|
$prev_sibling = $foster_parent_element->last_child; |
3608 |
|
|
} |
3609 |
|
|
last OE; |
3610 |
|
|
} |
3611 |
|
|
} # OE |
3612 |
|
|
$foster_parent_element = $open_elements->[0]->[0] and |
3613 |
|
|
$prev_sibling = $foster_parent_element->last_child |
3614 |
|
|
unless defined $foster_parent_element; |
3615 |
|
|
if (defined $prev_sibling and |
3616 |
|
|
$prev_sibling->node_type == 3) { |
3617 |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
3618 |
|
|
} else { |
3619 |
|
|
$foster_parent_element->insert_before |
3620 |
|
|
($self->{document}->create_text_node ($token->{data}), |
3621 |
|
|
$next_sibling); |
3622 |
|
|
} |
3623 |
|
|
} else { |
3624 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
3625 |
|
|
} |
3626 |
|
|
|
3627 |
wakaba |
1.2 |
!!!next-token; |
3628 |
|
|
redo B; |
3629 |
|
|
} elsif ($token->{type} eq 'comment') { |
3630 |
|
|
## Copied from 'in table' |
3631 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3632 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3633 |
|
|
!!!next-token; |
3634 |
|
|
redo B; |
3635 |
|
|
} elsif ($token->{type} eq 'start tag') { |
3636 |
|
|
if ({ |
3637 |
|
|
tr => 1, |
3638 |
|
|
th => 1, td => 1, |
3639 |
|
|
}->{$token->{tag_name}}) { |
3640 |
|
|
## Clear back to table body context |
3641 |
|
|
while (not { |
3642 |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
3643 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3644 |
|
|
!!!parse-error; |
3645 |
|
|
pop @$open_elements; |
3646 |
|
|
} |
3647 |
|
|
|
3648 |
|
|
$insertion_mode = 'in row'; |
3649 |
|
|
if ($token->{tag_name} eq 'tr') { |
3650 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
3651 |
|
|
!!!next-token; |
3652 |
|
|
} else { |
3653 |
|
|
!!!insert-element ('tr'); |
3654 |
|
|
## reprocess |
3655 |
|
|
} |
3656 |
|
|
redo B; |
3657 |
|
|
} elsif ({ |
3658 |
|
|
caption => 1, col => 1, colgroup => 1, |
3659 |
|
|
tbody => 1, tfoot => 1, thead => 1, |
3660 |
|
|
}->{$token->{tag_name}}) { |
3661 |
|
|
## have an element in table scope |
3662 |
|
|
my $i; |
3663 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3664 |
|
|
my $node = $open_elements->[$_]; |
3665 |
|
|
if ({ |
3666 |
|
|
tbody => 1, thead => 1, tfoot => 1, |
3667 |
|
|
}->{$node->[1]}) { |
3668 |
|
|
$i = $_; |
3669 |
|
|
last INSCOPE; |
3670 |
|
|
} elsif ({ |
3671 |
|
|
table => 1, html => 1, |
3672 |
|
|
}->{$node->[1]}) { |
3673 |
|
|
last INSCOPE; |
3674 |
|
|
} |
3675 |
|
|
} # INSCOPE |
3676 |
|
|
unless (defined $i) { |
3677 |
|
|
!!!parse-error; |
3678 |
|
|
## Ignore the token |
3679 |
|
|
!!!next-token; |
3680 |
|
|
redo B; |
3681 |
|
|
} |
3682 |
|
|
|
3683 |
|
|
## Clear back to table body context |
3684 |
|
|
while (not { |
3685 |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
3686 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3687 |
|
|
!!!parse-error; |
3688 |
|
|
pop @$open_elements; |
3689 |
|
|
} |
3690 |
|
|
|
3691 |
|
|
## As if <{current node}> |
3692 |
|
|
## have an element in table scope |
3693 |
|
|
## true by definition |
3694 |
|
|
|
3695 |
|
|
## Clear back to table body context |
3696 |
|
|
## nop by definition |
3697 |
|
|
|
3698 |
|
|
pop @$open_elements; |
3699 |
|
|
$insertion_mode = 'in table'; |
3700 |
|
|
## reprocess |
3701 |
|
|
redo B; |
3702 |
|
|
} elsif ($token->{tag_name} eq 'table') { |
3703 |
|
|
## NOTE: This is a code clone of "table in table" |
3704 |
|
|
!!!parse-error; |
3705 |
|
|
|
3706 |
|
|
## As if </table> |
3707 |
|
|
## have a table element in table scope |
3708 |
|
|
my $i; |
3709 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3710 |
|
|
my $node = $open_elements->[$_]; |
3711 |
|
|
if ($node->[1] eq 'table') { |
3712 |
|
|
$i = $_; |
3713 |
|
|
last INSCOPE; |
3714 |
|
|
} elsif ({ |
3715 |
|
|
table => 1, html => 1, |
3716 |
|
|
}->{$node->[1]}) { |
3717 |
|
|
last INSCOPE; |
3718 |
|
|
} |
3719 |
|
|
} # INSCOPE |
3720 |
|
|
unless (defined $i) { |
3721 |
|
|
!!!parse-error; |
3722 |
|
|
## Ignore tokens </table><table> |
3723 |
|
|
!!!next-token; |
3724 |
|
|
redo B; |
3725 |
|
|
} |
3726 |
|
|
|
3727 |
|
|
## generate implied end tags |
3728 |
|
|
if ({ |
3729 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
3730 |
|
|
td => 1, th => 1, tr => 1, |
3731 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3732 |
|
|
!!!back-token; # <table> |
3733 |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
3734 |
|
|
!!!back-token; |
3735 |
|
|
$token = {type => 'end tag', |
3736 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
3737 |
|
|
redo B; |
3738 |
|
|
} |
3739 |
|
|
|
3740 |
|
|
if ($open_elements->[-1]->[1] ne 'table') { |
3741 |
|
|
!!!parse-error; |
3742 |
|
|
} |
3743 |
|
|
|
3744 |
|
|
splice @$open_elements, $i; |
3745 |
|
|
|
3746 |
|
|
$reset_insertion_mode->(); |
3747 |
|
|
|
3748 |
|
|
## reprocess |
3749 |
|
|
redo B; |
3750 |
|
|
} else { |
3751 |
|
|
# |
3752 |
|
|
} |
3753 |
|
|
} elsif ($token->{type} eq 'end tag') { |
3754 |
|
|
if ({ |
3755 |
|
|
tbody => 1, tfoot => 1, thead => 1, |
3756 |
|
|
}->{$token->{tag_name}}) { |
3757 |
|
|
## have an element in table scope |
3758 |
|
|
my $i; |
3759 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3760 |
|
|
my $node = $open_elements->[$_]; |
3761 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
3762 |
|
|
$i = $_; |
3763 |
|
|
last INSCOPE; |
3764 |
|
|
} elsif ({ |
3765 |
|
|
table => 1, html => 1, |
3766 |
|
|
}->{$node->[1]}) { |
3767 |
|
|
last INSCOPE; |
3768 |
|
|
} |
3769 |
|
|
} # INSCOPE |
3770 |
|
|
unless (defined $i) { |
3771 |
|
|
!!!parse-error; |
3772 |
|
|
## Ignore the token |
3773 |
|
|
!!!next-token; |
3774 |
|
|
redo B; |
3775 |
|
|
} |
3776 |
|
|
|
3777 |
|
|
## Clear back to table body context |
3778 |
|
|
while (not { |
3779 |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
3780 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3781 |
|
|
!!!parse-error; |
3782 |
|
|
pop @$open_elements; |
3783 |
|
|
} |
3784 |
|
|
|
3785 |
|
|
pop @$open_elements; |
3786 |
|
|
$insertion_mode = 'in table'; |
3787 |
|
|
!!!next-token; |
3788 |
|
|
redo B; |
3789 |
|
|
} elsif ($token->{tag_name} eq 'table') { |
3790 |
|
|
## have an element in table scope |
3791 |
|
|
my $i; |
3792 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3793 |
|
|
my $node = $open_elements->[$_]; |
3794 |
|
|
if ({ |
3795 |
|
|
tbody => 1, thead => 1, tfoot => 1, |
3796 |
|
|
}->{$node->[1]}) { |
3797 |
|
|
$i = $_; |
3798 |
|
|
last INSCOPE; |
3799 |
|
|
} elsif ({ |
3800 |
|
|
table => 1, html => 1, |
3801 |
|
|
}->{$node->[1]}) { |
3802 |
|
|
last INSCOPE; |
3803 |
|
|
} |
3804 |
|
|
} # INSCOPE |
3805 |
|
|
unless (defined $i) { |
3806 |
|
|
!!!parse-error; |
3807 |
|
|
## Ignore the token |
3808 |
|
|
!!!next-token; |
3809 |
|
|
redo B; |
3810 |
|
|
} |
3811 |
|
|
|
3812 |
|
|
## Clear back to table body context |
3813 |
|
|
while (not { |
3814 |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
3815 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3816 |
|
|
!!!parse-error; |
3817 |
|
|
pop @$open_elements; |
3818 |
|
|
} |
3819 |
|
|
|
3820 |
|
|
## As if <{current node}> |
3821 |
|
|
## have an element in table scope |
3822 |
|
|
## true by definition |
3823 |
|
|
|
3824 |
|
|
## Clear back to table body context |
3825 |
|
|
## nop by definition |
3826 |
|
|
|
3827 |
|
|
pop @$open_elements; |
3828 |
|
|
$insertion_mode = 'in table'; |
3829 |
|
|
## reprocess |
3830 |
|
|
redo B; |
3831 |
|
|
} elsif ({ |
3832 |
|
|
body => 1, caption => 1, col => 1, colgroup => 1, |
3833 |
|
|
html => 1, td => 1, th => 1, tr => 1, |
3834 |
|
|
}->{$token->{tag_name}}) { |
3835 |
|
|
!!!parse-error; |
3836 |
|
|
## Ignore the token |
3837 |
|
|
!!!next-token; |
3838 |
|
|
redo B; |
3839 |
|
|
} else { |
3840 |
|
|
# |
3841 |
|
|
} |
3842 |
|
|
} else { |
3843 |
|
|
# |
3844 |
|
|
} |
3845 |
|
|
|
3846 |
|
|
## As if in table |
3847 |
|
|
!!!parse-error; |
3848 |
wakaba |
1.7 |
$in_body->($insert_to_foster); |
3849 |
wakaba |
1.2 |
redo B; |
3850 |
|
|
} elsif ($insertion_mode eq 'in row') { |
3851 |
|
|
if ($token->{type} eq 'character') { |
3852 |
wakaba |
1.7 |
## NOTE: This is a "character in table" code clone. |
3853 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
3854 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($1); |
3855 |
|
|
|
3856 |
|
|
unless (length $token->{data}) { |
3857 |
|
|
!!!next-token; |
3858 |
|
|
redo B; |
3859 |
|
|
} |
3860 |
|
|
} |
3861 |
wakaba |
1.2 |
|
3862 |
wakaba |
1.7 |
## As if in body, but insert into foster parent element |
3863 |
|
|
## ISSUE: Spec says that "whenever a node would be inserted |
3864 |
|
|
## into the current node" while characters might not be |
3865 |
|
|
## result in a new Text node. |
3866 |
|
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
3867 |
|
|
|
3868 |
|
|
if ({ |
3869 |
|
|
table => 1, tbody => 1, tfoot => 1, |
3870 |
|
|
thead => 1, tr => 1, |
3871 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3872 |
|
|
# MUST |
3873 |
|
|
my $foster_parent_element; |
3874 |
|
|
my $next_sibling; |
3875 |
|
|
my $prev_sibling; |
3876 |
|
|
OE: for (reverse 0..$#$open_elements) { |
3877 |
|
|
if ($open_elements->[$_]->[1] eq 'table') { |
3878 |
|
|
my $parent = $open_elements->[$_]->[0]->parent_node; |
3879 |
|
|
if (defined $parent and $parent->node_type == 1) { |
3880 |
|
|
$foster_parent_element = $parent; |
3881 |
|
|
$next_sibling = $open_elements->[$_]->[0]; |
3882 |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
3883 |
|
|
} else { |
3884 |
|
|
$foster_parent_element = $open_elements->[$_ - 1]->[0]; |
3885 |
|
|
$prev_sibling = $foster_parent_element->last_child; |
3886 |
|
|
} |
3887 |
|
|
last OE; |
3888 |
|
|
} |
3889 |
|
|
} # OE |
3890 |
|
|
$foster_parent_element = $open_elements->[0]->[0] and |
3891 |
|
|
$prev_sibling = $foster_parent_element->last_child |
3892 |
|
|
unless defined $foster_parent_element; |
3893 |
|
|
if (defined $prev_sibling and |
3894 |
|
|
$prev_sibling->node_type == 3) { |
3895 |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
3896 |
|
|
} else { |
3897 |
|
|
$foster_parent_element->insert_before |
3898 |
|
|
($self->{document}->create_text_node ($token->{data}), |
3899 |
|
|
$next_sibling); |
3900 |
|
|
} |
3901 |
|
|
} else { |
3902 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
3903 |
|
|
} |
3904 |
|
|
|
3905 |
wakaba |
1.2 |
!!!next-token; |
3906 |
|
|
redo B; |
3907 |
|
|
} elsif ($token->{type} eq 'comment') { |
3908 |
|
|
## Copied from 'in table' |
3909 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
3910 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
3911 |
|
|
!!!next-token; |
3912 |
|
|
redo B; |
3913 |
|
|
} elsif ($token->{type} eq 'start tag') { |
3914 |
|
|
if ($token->{tag_name} eq 'th' or |
3915 |
|
|
$token->{tag_name} eq 'td') { |
3916 |
|
|
## Clear back to table row context |
3917 |
|
|
while (not { |
3918 |
wakaba |
1.6 |
tr => 1, html => 1, |
3919 |
wakaba |
1.2 |
}->{$open_elements->[-1]->[1]}) { |
3920 |
|
|
!!!parse-error; |
3921 |
|
|
pop @$open_elements; |
3922 |
|
|
} |
3923 |
|
|
|
3924 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
3925 |
|
|
$insertion_mode = 'in cell'; |
3926 |
|
|
|
3927 |
|
|
push @$active_formatting_elements, ['#marker', '']; |
3928 |
|
|
|
3929 |
|
|
!!!next-token; |
3930 |
|
|
redo B; |
3931 |
|
|
} elsif ({ |
3932 |
|
|
caption => 1, col => 1, colgroup => 1, |
3933 |
|
|
tbody => 1, tfoot => 1, thead => 1, tr => 1, |
3934 |
|
|
}->{$token->{tag_name}}) { |
3935 |
|
|
## As if </tr> |
3936 |
|
|
## have an element in table scope |
3937 |
|
|
my $i; |
3938 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3939 |
|
|
my $node = $open_elements->[$_]; |
3940 |
|
|
if ($node->[1] eq 'tr') { |
3941 |
|
|
$i = $_; |
3942 |
|
|
last INSCOPE; |
3943 |
|
|
} elsif ({ |
3944 |
|
|
table => 1, html => 1, |
3945 |
|
|
}->{$node->[1]}) { |
3946 |
|
|
last INSCOPE; |
3947 |
|
|
} |
3948 |
|
|
} # INSCOPE |
3949 |
|
|
unless (defined $i) { |
3950 |
|
|
!!!parse-error; |
3951 |
|
|
## Ignore the token |
3952 |
|
|
!!!next-token; |
3953 |
|
|
redo B; |
3954 |
|
|
} |
3955 |
|
|
|
3956 |
|
|
## Clear back to table row context |
3957 |
|
|
while (not { |
3958 |
|
|
tr => 1, html => 1, |
3959 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3960 |
|
|
!!!parse-error; |
3961 |
|
|
pop @$open_elements; |
3962 |
|
|
} |
3963 |
|
|
|
3964 |
|
|
pop @$open_elements; # tr |
3965 |
|
|
$insertion_mode = 'in table body'; |
3966 |
|
|
## reprocess |
3967 |
|
|
redo B; |
3968 |
|
|
} elsif ($token->{tag_name} eq 'table') { |
3969 |
|
|
## NOTE: This is a code clone of "table in table" |
3970 |
|
|
!!!parse-error; |
3971 |
|
|
|
3972 |
|
|
## As if </table> |
3973 |
|
|
## have a table element in table scope |
3974 |
|
|
my $i; |
3975 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
3976 |
|
|
my $node = $open_elements->[$_]; |
3977 |
|
|
if ($node->[1] eq 'table') { |
3978 |
|
|
$i = $_; |
3979 |
|
|
last INSCOPE; |
3980 |
|
|
} elsif ({ |
3981 |
|
|
table => 1, html => 1, |
3982 |
|
|
}->{$node->[1]}) { |
3983 |
|
|
last INSCOPE; |
3984 |
|
|
} |
3985 |
|
|
} # INSCOPE |
3986 |
|
|
unless (defined $i) { |
3987 |
|
|
!!!parse-error; |
3988 |
|
|
## Ignore tokens </table><table> |
3989 |
|
|
!!!next-token; |
3990 |
|
|
redo B; |
3991 |
|
|
} |
3992 |
|
|
|
3993 |
|
|
## generate implied end tags |
3994 |
|
|
if ({ |
3995 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
3996 |
|
|
td => 1, th => 1, tr => 1, |
3997 |
|
|
}->{$open_elements->[-1]->[1]}) { |
3998 |
|
|
!!!back-token; # <table> |
3999 |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
4000 |
|
|
!!!back-token; |
4001 |
|
|
$token = {type => 'end tag', |
4002 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
4003 |
|
|
redo B; |
4004 |
|
|
} |
4005 |
|
|
|
4006 |
|
|
if ($open_elements->[-1]->[1] ne 'table') { |
4007 |
|
|
!!!parse-error; |
4008 |
|
|
} |
4009 |
|
|
|
4010 |
|
|
splice @$open_elements, $i; |
4011 |
|
|
|
4012 |
|
|
$reset_insertion_mode->(); |
4013 |
|
|
|
4014 |
|
|
## reprocess |
4015 |
|
|
redo B; |
4016 |
|
|
} else { |
4017 |
|
|
# |
4018 |
|
|
} |
4019 |
|
|
} elsif ($token->{type} eq 'end tag') { |
4020 |
|
|
if ($token->{tag_name} eq 'tr') { |
4021 |
|
|
## have an element in table scope |
4022 |
|
|
my $i; |
4023 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4024 |
|
|
my $node = $open_elements->[$_]; |
4025 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4026 |
|
|
$i = $_; |
4027 |
|
|
last INSCOPE; |
4028 |
|
|
} elsif ({ |
4029 |
|
|
table => 1, html => 1, |
4030 |
|
|
}->{$node->[1]}) { |
4031 |
|
|
last INSCOPE; |
4032 |
|
|
} |
4033 |
|
|
} # INSCOPE |
4034 |
|
|
unless (defined $i) { |
4035 |
|
|
!!!parse-error; |
4036 |
|
|
## Ignore the token |
4037 |
|
|
!!!next-token; |
4038 |
|
|
redo B; |
4039 |
|
|
} |
4040 |
|
|
|
4041 |
|
|
## Clear back to table row context |
4042 |
|
|
while (not { |
4043 |
|
|
tr => 1, html => 1, |
4044 |
|
|
}->{$open_elements->[-1]->[1]}) { |
4045 |
|
|
!!!parse-error; |
4046 |
|
|
pop @$open_elements; |
4047 |
|
|
} |
4048 |
|
|
|
4049 |
|
|
pop @$open_elements; # tr |
4050 |
|
|
$insertion_mode = 'in table body'; |
4051 |
|
|
!!!next-token; |
4052 |
|
|
redo B; |
4053 |
|
|
} elsif ($token->{tag_name} eq 'table') { |
4054 |
|
|
## As if </tr> |
4055 |
|
|
## have an element in table scope |
4056 |
|
|
my $i; |
4057 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4058 |
|
|
my $node = $open_elements->[$_]; |
4059 |
|
|
if ($node->[1] eq 'tr') { |
4060 |
|
|
$i = $_; |
4061 |
|
|
last INSCOPE; |
4062 |
|
|
} elsif ({ |
4063 |
|
|
table => 1, html => 1, |
4064 |
|
|
}->{$node->[1]}) { |
4065 |
|
|
last INSCOPE; |
4066 |
|
|
} |
4067 |
|
|
} # INSCOPE |
4068 |
|
|
unless (defined $i) { |
4069 |
|
|
!!!parse-error; |
4070 |
|
|
## Ignore the token |
4071 |
|
|
!!!next-token; |
4072 |
|
|
redo B; |
4073 |
|
|
} |
4074 |
|
|
|
4075 |
|
|
## Clear back to table row context |
4076 |
|
|
while (not { |
4077 |
|
|
tr => 1, html => 1, |
4078 |
|
|
}->{$open_elements->[-1]->[1]}) { |
4079 |
|
|
!!!parse-error; |
4080 |
|
|
pop @$open_elements; |
4081 |
|
|
} |
4082 |
|
|
|
4083 |
|
|
pop @$open_elements; # tr |
4084 |
|
|
$insertion_mode = 'in table body'; |
4085 |
|
|
## reprocess |
4086 |
|
|
redo B; |
4087 |
|
|
} elsif ({ |
4088 |
|
|
tbody => 1, tfoot => 1, thead => 1, |
4089 |
|
|
}->{$token->{tag_name}}) { |
4090 |
|
|
## have an element in table scope |
4091 |
|
|
my $i; |
4092 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4093 |
|
|
my $node = $open_elements->[$_]; |
4094 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4095 |
|
|
$i = $_; |
4096 |
|
|
last INSCOPE; |
4097 |
|
|
} elsif ({ |
4098 |
|
|
table => 1, html => 1, |
4099 |
|
|
}->{$node->[1]}) { |
4100 |
|
|
last INSCOPE; |
4101 |
|
|
} |
4102 |
|
|
} # INSCOPE |
4103 |
|
|
unless (defined $i) { |
4104 |
|
|
!!!parse-error; |
4105 |
|
|
## Ignore the token |
4106 |
|
|
!!!next-token; |
4107 |
|
|
redo B; |
4108 |
|
|
} |
4109 |
|
|
|
4110 |
|
|
## As if </tr> |
4111 |
|
|
## have an element in table scope |
4112 |
|
|
my $i; |
4113 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4114 |
|
|
my $node = $open_elements->[$_]; |
4115 |
|
|
if ($node->[1] eq 'tr') { |
4116 |
|
|
$i = $_; |
4117 |
|
|
last INSCOPE; |
4118 |
|
|
} elsif ({ |
4119 |
|
|
table => 1, html => 1, |
4120 |
|
|
}->{$node->[1]}) { |
4121 |
|
|
last INSCOPE; |
4122 |
|
|
} |
4123 |
|
|
} # INSCOPE |
4124 |
|
|
unless (defined $i) { |
4125 |
|
|
!!!parse-error; |
4126 |
|
|
## Ignore the token |
4127 |
|
|
!!!next-token; |
4128 |
|
|
redo B; |
4129 |
|
|
} |
4130 |
|
|
|
4131 |
|
|
## Clear back to table row context |
4132 |
|
|
while (not { |
4133 |
|
|
tr => 1, html => 1, |
4134 |
|
|
}->{$open_elements->[-1]->[1]}) { |
4135 |
|
|
!!!parse-error; |
4136 |
|
|
pop @$open_elements; |
4137 |
|
|
} |
4138 |
|
|
|
4139 |
|
|
pop @$open_elements; # tr |
4140 |
|
|
$insertion_mode = 'in table body'; |
4141 |
|
|
## reprocess |
4142 |
|
|
redo B; |
4143 |
|
|
} elsif ({ |
4144 |
|
|
body => 1, caption => 1, col => 1, |
4145 |
|
|
colgroup => 1, html => 1, td => 1, th => 1, |
4146 |
|
|
}->{$token->{tag_name}}) { |
4147 |
|
|
!!!parse-error; |
4148 |
|
|
## Ignore the token |
4149 |
|
|
!!!next-token; |
4150 |
|
|
redo B; |
4151 |
|
|
} else { |
4152 |
|
|
# |
4153 |
|
|
} |
4154 |
|
|
} else { |
4155 |
|
|
# |
4156 |
|
|
} |
4157 |
|
|
|
4158 |
|
|
## As if in table |
4159 |
|
|
!!!parse-error; |
4160 |
wakaba |
1.7 |
$in_body->($insert_to_foster); |
4161 |
wakaba |
1.2 |
redo B; |
4162 |
|
|
} elsif ($insertion_mode eq 'in cell') { |
4163 |
|
|
if ($token->{type} eq 'character') { |
4164 |
|
|
## NOTE: This is a code clone of "character in body". |
4165 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
4166 |
wakaba |
1.2 |
|
4167 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
4168 |
|
|
|
4169 |
|
|
!!!next-token; |
4170 |
|
|
redo B; |
4171 |
|
|
} elsif ($token->{type} eq 'comment') { |
4172 |
|
|
## NOTE: This is a code clone of "comment in body". |
4173 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
4174 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
4175 |
|
|
!!!next-token; |
4176 |
|
|
redo B; |
4177 |
|
|
} elsif ($token->{type} eq 'start tag') { |
4178 |
|
|
if ({ |
4179 |
|
|
caption => 1, col => 1, colgroup => 1, |
4180 |
|
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
4181 |
|
|
thead => 1, tr => 1, |
4182 |
|
|
}->{$token->{tag_name}}) { |
4183 |
|
|
## have an element in table scope |
4184 |
|
|
my $tn; |
4185 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4186 |
|
|
my $node = $open_elements->[$_]; |
4187 |
|
|
if ($node->[1] eq 'td' or $node->[1] eq 'th') { |
4188 |
|
|
$tn = $node->[1]; |
4189 |
|
|
last INSCOPE; |
4190 |
|
|
} elsif ({ |
4191 |
|
|
table => 1, html => 1, |
4192 |
|
|
}->{$node->[1]}) { |
4193 |
|
|
last INSCOPE; |
4194 |
|
|
} |
4195 |
|
|
} # INSCOPE |
4196 |
|
|
unless (defined $tn) { |
4197 |
|
|
!!!parse-error; |
4198 |
|
|
## Ignore the token |
4199 |
|
|
!!!next-token; |
4200 |
|
|
redo B; |
4201 |
|
|
} |
4202 |
|
|
|
4203 |
|
|
## Close the cell |
4204 |
|
|
!!!back-token; # <?> |
4205 |
|
|
$token = {type => 'end tag', tag_name => $tn}; |
4206 |
|
|
redo B; |
4207 |
|
|
} else { |
4208 |
|
|
# |
4209 |
|
|
} |
4210 |
|
|
} elsif ($token->{type} eq 'end tag') { |
4211 |
wakaba |
1.6 |
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
4212 |
wakaba |
1.2 |
## have an element in table scope |
4213 |
|
|
my $i; |
4214 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4215 |
|
|
my $node = $open_elements->[$_]; |
4216 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4217 |
|
|
$i = $_; |
4218 |
|
|
last INSCOPE; |
4219 |
|
|
} elsif ({ |
4220 |
|
|
table => 1, html => 1, |
4221 |
|
|
}->{$node->[1]}) { |
4222 |
|
|
last INSCOPE; |
4223 |
|
|
} |
4224 |
|
|
} # INSCOPE |
4225 |
|
|
unless (defined $i) { |
4226 |
|
|
!!!parse-error; |
4227 |
|
|
## Ignore the token |
4228 |
|
|
!!!next-token; |
4229 |
|
|
redo B; |
4230 |
|
|
} |
4231 |
|
|
|
4232 |
|
|
## generate implied end tags |
4233 |
|
|
if ({ |
4234 |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
4235 |
|
|
td => ($token->{tag_name} eq 'th'), |
4236 |
|
|
th => ($token->{tag_name} eq 'td'), |
4237 |
|
|
tr => 1, |
4238 |
|
|
}->{$open_elements->[-1]->[1]}) { |
4239 |
|
|
!!!back-token; |
4240 |
|
|
$token = {type => 'end tag', |
4241 |
|
|
tag_name => $open_elements->[-1]->[1]}; # MUST |
4242 |
|
|
redo B; |
4243 |
|
|
} |
4244 |
|
|
|
4245 |
|
|
if ($open_elements->[-1]->[1] ne $token->{tag_name}) { |
4246 |
|
|
!!!parse-error; |
4247 |
|
|
} |
4248 |
|
|
|
4249 |
|
|
splice @$open_elements, $i; |
4250 |
|
|
|
4251 |
|
|
$clear_up_to_marker->(); |
4252 |
|
|
|
4253 |
|
|
$insertion_mode = 'in row'; |
4254 |
|
|
|
4255 |
|
|
!!!next-token; |
4256 |
|
|
redo B; |
4257 |
|
|
} elsif ({ |
4258 |
|
|
body => 1, caption => 1, col => 1, |
4259 |
|
|
colgroup => 1, html => 1, |
4260 |
|
|
}->{$token->{tag_name}}) { |
4261 |
|
|
!!!parse-error; |
4262 |
|
|
## Ignore the token |
4263 |
|
|
!!!next-token; |
4264 |
|
|
redo B; |
4265 |
|
|
} elsif ({ |
4266 |
|
|
table => 1, tbody => 1, tfoot => 1, |
4267 |
|
|
thead => 1, tr => 1, |
4268 |
|
|
}->{$token->{tag_name}}) { |
4269 |
|
|
## have an element in table scope |
4270 |
|
|
my $i; |
4271 |
|
|
my $tn; |
4272 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4273 |
|
|
my $node = $open_elements->[$_]; |
4274 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4275 |
|
|
$i = $_; |
4276 |
|
|
last INSCOPE; |
4277 |
|
|
} elsif ($node->[1] eq 'td' or $node->[1] eq 'th') { |
4278 |
|
|
$tn = $node->[1]; |
4279 |
|
|
## NOTE: There is exactly one |td| or |th| element |
4280 |
|
|
## in scope in the stack of open elements by definition. |
4281 |
|
|
} elsif ({ |
4282 |
|
|
table => 1, html => 1, |
4283 |
|
|
}->{$node->[1]}) { |
4284 |
|
|
last INSCOPE; |
4285 |
|
|
} |
4286 |
|
|
} # INSCOPE |
4287 |
|
|
unless (defined $i) { |
4288 |
|
|
!!!parse-error; |
4289 |
|
|
## Ignore the token |
4290 |
|
|
!!!next-token; |
4291 |
|
|
redo B; |
4292 |
|
|
} |
4293 |
|
|
|
4294 |
|
|
## Close the cell |
4295 |
|
|
!!!back-token; # </?> |
4296 |
|
|
$token = {type => 'end tag', tag_name => $tn}; |
4297 |
|
|
redo B; |
4298 |
|
|
} else { |
4299 |
|
|
# |
4300 |
|
|
} |
4301 |
|
|
} else { |
4302 |
|
|
# |
4303 |
|
|
} |
4304 |
|
|
|
4305 |
wakaba |
1.7 |
$in_body->($insert_to_current); |
4306 |
wakaba |
1.2 |
redo B; |
4307 |
|
|
} elsif ($insertion_mode eq 'in select') { |
4308 |
|
|
if ($token->{type} eq 'character') { |
4309 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
4310 |
|
|
!!!next-token; |
4311 |
|
|
redo B; |
4312 |
|
|
} elsif ($token->{type} eq 'comment') { |
4313 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
4314 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
4315 |
|
|
!!!next-token; |
4316 |
|
|
redo B; |
4317 |
|
|
} elsif ($token->{type} eq 'start tag') { |
4318 |
|
|
if ($token->{tag_name} eq 'option') { |
4319 |
|
|
if ($open_elements->[-1]->[1] eq 'option') { |
4320 |
|
|
## As if </option> |
4321 |
|
|
pop @$open_elements; |
4322 |
|
|
} |
4323 |
|
|
|
4324 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
4325 |
|
|
!!!next-token; |
4326 |
|
|
redo B; |
4327 |
|
|
} elsif ($token->{tag_name} eq 'optgroup') { |
4328 |
|
|
if ($open_elements->[-1]->[1] eq 'option') { |
4329 |
|
|
## As if </option> |
4330 |
|
|
pop @$open_elements; |
4331 |
|
|
} |
4332 |
|
|
|
4333 |
|
|
if ($open_elements->[-1]->[1] eq 'optgroup') { |
4334 |
|
|
## As if </optgroup> |
4335 |
|
|
pop @$open_elements; |
4336 |
|
|
} |
4337 |
|
|
|
4338 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
4339 |
|
|
!!!next-token; |
4340 |
|
|
redo B; |
4341 |
|
|
} elsif ($token->{tag_name} eq 'select') { |
4342 |
|
|
!!!parse-error; |
4343 |
|
|
## As if </select> instead |
4344 |
|
|
## have an element in table scope |
4345 |
|
|
my $i; |
4346 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4347 |
|
|
my $node = $open_elements->[$_]; |
4348 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4349 |
|
|
$i = $_; |
4350 |
|
|
last INSCOPE; |
4351 |
|
|
} elsif ({ |
4352 |
|
|
table => 1, html => 1, |
4353 |
|
|
}->{$node->[1]}) { |
4354 |
|
|
last INSCOPE; |
4355 |
|
|
} |
4356 |
|
|
} # INSCOPE |
4357 |
|
|
unless (defined $i) { |
4358 |
|
|
!!!parse-error; |
4359 |
|
|
## Ignore the token |
4360 |
|
|
!!!next-token; |
4361 |
|
|
redo B; |
4362 |
|
|
} |
4363 |
|
|
|
4364 |
|
|
splice @$open_elements, $i; |
4365 |
|
|
|
4366 |
|
|
$reset_insertion_mode->(); |
4367 |
|
|
|
4368 |
|
|
!!!next-token; |
4369 |
|
|
redo B; |
4370 |
|
|
} else { |
4371 |
|
|
# |
4372 |
|
|
} |
4373 |
|
|
} elsif ($token->{type} eq 'end tag') { |
4374 |
|
|
if ($token->{tag_name} eq 'optgroup') { |
4375 |
|
|
if ($open_elements->[-1]->[1] eq 'option' and |
4376 |
|
|
$open_elements->[-2]->[1] eq 'optgroup') { |
4377 |
|
|
## As if </option> |
4378 |
|
|
splice @$open_elements, -2; |
4379 |
|
|
} elsif ($open_elements->[-1]->[1] eq 'optgroup') { |
4380 |
|
|
pop @$open_elements; |
4381 |
|
|
} else { |
4382 |
|
|
!!!parse-error; |
4383 |
|
|
## Ignore the token |
4384 |
|
|
} |
4385 |
|
|
!!!next-token; |
4386 |
|
|
redo B; |
4387 |
|
|
} elsif ($token->{tag_name} eq 'option') { |
4388 |
|
|
if ($open_elements->[-1]->[1] eq 'option') { |
4389 |
|
|
pop @$open_elements; |
4390 |
|
|
} else { |
4391 |
|
|
!!!parse-error; |
4392 |
|
|
## Ignore the token |
4393 |
|
|
} |
4394 |
|
|
!!!next-token; |
4395 |
|
|
redo B; |
4396 |
|
|
} elsif ($token->{tag_name} eq 'select') { |
4397 |
|
|
## have an element in table scope |
4398 |
|
|
my $i; |
4399 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4400 |
|
|
my $node = $open_elements->[$_]; |
4401 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4402 |
|
|
$i = $_; |
4403 |
|
|
last INSCOPE; |
4404 |
|
|
} elsif ({ |
4405 |
|
|
table => 1, html => 1, |
4406 |
|
|
}->{$node->[1]}) { |
4407 |
|
|
last INSCOPE; |
4408 |
|
|
} |
4409 |
|
|
} # INSCOPE |
4410 |
|
|
unless (defined $i) { |
4411 |
|
|
!!!parse-error; |
4412 |
|
|
## Ignore the token |
4413 |
|
|
!!!next-token; |
4414 |
|
|
redo B; |
4415 |
|
|
} |
4416 |
|
|
|
4417 |
|
|
splice @$open_elements, $i; |
4418 |
|
|
|
4419 |
|
|
$reset_insertion_mode->(); |
4420 |
|
|
|
4421 |
|
|
!!!next-token; |
4422 |
|
|
redo B; |
4423 |
|
|
} elsif ({ |
4424 |
|
|
caption => 1, table => 1, tbody => 1, |
4425 |
|
|
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
4426 |
|
|
}->{$token->{tag_name}}) { |
4427 |
|
|
!!!parse-error; |
4428 |
|
|
|
4429 |
|
|
## have an element in table scope |
4430 |
|
|
my $i; |
4431 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4432 |
|
|
my $node = $open_elements->[$_]; |
4433 |
|
|
if ($node->[1] eq $token->{tag_name}) { |
4434 |
|
|
$i = $_; |
4435 |
|
|
last INSCOPE; |
4436 |
|
|
} elsif ({ |
4437 |
|
|
table => 1, html => 1, |
4438 |
|
|
}->{$node->[1]}) { |
4439 |
|
|
last INSCOPE; |
4440 |
|
|
} |
4441 |
|
|
} # INSCOPE |
4442 |
|
|
unless (defined $i) { |
4443 |
|
|
## Ignore the token |
4444 |
|
|
!!!next-token; |
4445 |
|
|
redo B; |
4446 |
|
|
} |
4447 |
|
|
|
4448 |
|
|
## As if </select> |
4449 |
|
|
## have an element in table scope |
4450 |
|
|
undef $i; |
4451 |
|
|
INSCOPE: for (reverse 0..$#$open_elements) { |
4452 |
|
|
my $node = $open_elements->[$_]; |
4453 |
|
|
if ($node->[1] eq 'select') { |
4454 |
|
|
$i = $_; |
4455 |
|
|
last INSCOPE; |
4456 |
|
|
} elsif ({ |
4457 |
|
|
table => 1, html => 1, |
4458 |
|
|
}->{$node->[1]}) { |
4459 |
|
|
last INSCOPE; |
4460 |
|
|
} |
4461 |
|
|
} # INSCOPE |
4462 |
|
|
unless (defined $i) { |
4463 |
|
|
!!!parse-error; |
4464 |
|
|
## Ignore the </select> token |
4465 |
|
|
!!!next-token; ## TODO: ok? |
4466 |
|
|
redo B; |
4467 |
|
|
} |
4468 |
|
|
|
4469 |
|
|
splice @$open_elements, $i; |
4470 |
|
|
|
4471 |
|
|
$reset_insertion_mode->(); |
4472 |
|
|
|
4473 |
|
|
## reprocess |
4474 |
|
|
redo B; |
4475 |
|
|
} else { |
4476 |
|
|
# |
4477 |
|
|
} |
4478 |
|
|
} else { |
4479 |
|
|
# |
4480 |
|
|
} |
4481 |
|
|
|
4482 |
|
|
!!!parse-error; |
4483 |
|
|
## Ignore the token |
4484 |
wakaba |
1.7 |
!!!next-token; |
4485 |
wakaba |
1.2 |
redo B; |
4486 |
|
|
} elsif ($insertion_mode eq 'after body') { |
4487 |
|
|
if ($token->{type} eq 'character') { |
4488 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
4489 |
|
|
## As if in body |
4490 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current); |
4491 |
wakaba |
1.2 |
|
4492 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
4493 |
|
|
|
4494 |
|
|
unless (length $token->{data}) { |
4495 |
|
|
!!!next-token; |
4496 |
|
|
redo B; |
4497 |
|
|
} |
4498 |
|
|
} |
4499 |
|
|
|
4500 |
|
|
# |
4501 |
|
|
} elsif ($token->{type} eq 'comment') { |
4502 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
4503 |
|
|
$open_elements->[0]->[0]->append_child ($comment); |
4504 |
|
|
!!!next-token; |
4505 |
|
|
redo B; |
4506 |
|
|
} elsif ($token->{type} eq 'end tag') { |
4507 |
wakaba |
1.6 |
if ($token->{tag_name} eq 'html') { |
4508 |
wakaba |
1.2 |
## TODO: if inner_html, parse-error, ignore the token; otherwise, |
4509 |
|
|
|
4510 |
|
|
$phase = 'trailing end'; |
4511 |
|
|
!!!next-token; |
4512 |
|
|
redo B; |
4513 |
|
|
} else { |
4514 |
|
|
# |
4515 |
|
|
} |
4516 |
|
|
} else { |
4517 |
|
|
# |
4518 |
|
|
} |
4519 |
|
|
|
4520 |
wakaba |
1.8 |
!!!parse-error ('data after body'); |
4521 |
wakaba |
1.2 |
$insertion_mode = 'in body'; |
4522 |
|
|
## reprocess |
4523 |
|
|
redo B; |
4524 |
|
|
} elsif ($insertion_mode eq 'in frameset') { |
4525 |
|
|
if ($token->{type} eq 'character') { |
4526 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
4527 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
4528 |
|
|
|
4529 |
|
|
unless (length $token->{data}) { |
4530 |
|
|
!!!next-token; |
4531 |
|
|
redo B; |
4532 |
|
|
} |
4533 |
|
|
} |
4534 |
|
|
|
4535 |
|
|
# |
4536 |
|
|
} elsif ($token->{type} eq 'comment') { |
4537 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
4538 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
4539 |
|
|
!!!next-token; |
4540 |
|
|
redo B; |
4541 |
|
|
} elsif ($token->{type} eq 'start tag') { |
4542 |
|
|
if ($token->{tag_name} eq 'frameset') { |
4543 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
4544 |
|
|
!!!next-token; |
4545 |
|
|
redo B; |
4546 |
|
|
} elsif ($token->{tag_name} eq 'frame') { |
4547 |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
4548 |
|
|
pop @$open_elements; |
4549 |
|
|
!!!next-token; |
4550 |
|
|
redo B; |
4551 |
|
|
} elsif ($token->{tag_name} eq 'noframes') { |
4552 |
wakaba |
1.7 |
$in_body->($insert_to_current); |
4553 |
wakaba |
1.2 |
redo B; |
4554 |
|
|
} else { |
4555 |
|
|
# |
4556 |
|
|
} |
4557 |
|
|
} elsif ($token->{type} eq 'end tag') { |
4558 |
|
|
if ($token->{tag_name} eq 'frameset') { |
4559 |
|
|
if ($open_elements->[-1]->[1] eq 'html' and |
4560 |
|
|
@$open_elements == 1) { |
4561 |
|
|
!!!parse-error; |
4562 |
|
|
## Ignore the token |
4563 |
|
|
!!!next-token; |
4564 |
|
|
} else { |
4565 |
|
|
pop @$open_elements; |
4566 |
|
|
!!!next-token; |
4567 |
|
|
} |
4568 |
|
|
|
4569 |
|
|
## if not inner_html and |
4570 |
|
|
if ($open_elements->[-1]->[1] ne 'frameset') { |
4571 |
|
|
$insertion_mode = 'after frameset'; |
4572 |
|
|
} |
4573 |
|
|
redo B; |
4574 |
|
|
} else { |
4575 |
|
|
# |
4576 |
|
|
} |
4577 |
|
|
} else { |
4578 |
|
|
# |
4579 |
|
|
} |
4580 |
|
|
|
4581 |
|
|
!!!parse-error; |
4582 |
|
|
## Ignore the token |
4583 |
|
|
!!!next-token; |
4584 |
|
|
redo B; |
4585 |
|
|
} elsif ($insertion_mode eq 'after frameset') { |
4586 |
|
|
if ($token->{type} eq 'character') { |
4587 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
4588 |
|
|
$open_elements->[-1]->[0]->manakai_append_text ($token->{data}); |
4589 |
|
|
|
4590 |
|
|
unless (length $token->{data}) { |
4591 |
|
|
!!!next-token; |
4592 |
|
|
redo B; |
4593 |
|
|
} |
4594 |
|
|
} |
4595 |
|
|
|
4596 |
|
|
# |
4597 |
|
|
} elsif ($token->{type} eq 'comment') { |
4598 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
4599 |
|
|
$open_elements->[-1]->[0]->append_child ($comment); |
4600 |
|
|
!!!next-token; |
4601 |
|
|
redo B; |
4602 |
|
|
} elsif ($token->{type} eq 'start tag') { |
4603 |
|
|
if ($token->{tag_name} eq 'noframes') { |
4604 |
wakaba |
1.7 |
$in_body->($insert_to_current); |
4605 |
wakaba |
1.2 |
redo B; |
4606 |
|
|
} else { |
4607 |
|
|
# |
4608 |
|
|
} |
4609 |
|
|
} elsif ($token->{type} eq 'end tag') { |
4610 |
|
|
if ($token->{tag_name} eq 'html') { |
4611 |
|
|
$phase = 'trailing end'; |
4612 |
|
|
!!!next-token; |
4613 |
|
|
redo B; |
4614 |
|
|
} else { |
4615 |
|
|
# |
4616 |
|
|
} |
4617 |
|
|
} else { |
4618 |
|
|
# |
4619 |
|
|
} |
4620 |
|
|
|
4621 |
|
|
!!!parse-error; |
4622 |
|
|
## Ignore the token |
4623 |
|
|
!!!next-token; |
4624 |
|
|
redo B; |
4625 |
|
|
|
4626 |
|
|
## ISSUE: An issue in spec there |
4627 |
|
|
} else { |
4628 |
|
|
die "$0: $insertion_mode: Unknown insertion mode"; |
4629 |
|
|
} |
4630 |
|
|
} |
4631 |
|
|
} elsif ($phase eq 'trailing end') { |
4632 |
|
|
## states in the main stage is preserved yet # MUST |
4633 |
|
|
|
4634 |
|
|
if ($token->{type} eq 'DOCTYPE') { |
4635 |
|
|
!!!parse-error; |
4636 |
|
|
## Ignore the token |
4637 |
|
|
!!!next-token; |
4638 |
|
|
redo B; |
4639 |
|
|
} elsif ($token->{type} eq 'comment') { |
4640 |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
4641 |
|
|
$self->{document}->append_child ($comment); |
4642 |
|
|
!!!next-token; |
4643 |
|
|
redo B; |
4644 |
|
|
} elsif ($token->{type} eq 'character') { |
4645 |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
4646 |
wakaba |
1.7 |
my $data = $1; |
4647 |
wakaba |
1.2 |
## As if in the main phase. |
4648 |
|
|
## NOTE: The insertion mode in the main phase |
4649 |
|
|
## just before the phase has been changed to the trailing |
4650 |
|
|
## end phase is either "after body" or "after frameset". |
4651 |
wakaba |
1.7 |
$reconstruct_active_formatting_elements->($insert_to_current) |
4652 |
wakaba |
1.2 |
if $phase eq 'main'; |
4653 |
|
|
|
4654 |
wakaba |
1.7 |
$open_elements->[-1]->[0]->manakai_append_text ($data); |
4655 |
wakaba |
1.2 |
|
4656 |
|
|
unless (length $token->{data}) { |
4657 |
|
|
!!!next-token; |
4658 |
|
|
redo B; |
4659 |
|
|
} |
4660 |
|
|
} |
4661 |
|
|
|
4662 |
|
|
!!!parse-error; |
4663 |
|
|
$phase = 'main'; |
4664 |
|
|
## reprocess |
4665 |
|
|
redo B; |
4666 |
|
|
} elsif ($token->{type} eq 'start tag' or |
4667 |
|
|
$token->{type} eq 'end tag') { |
4668 |
|
|
!!!parse-error; |
4669 |
|
|
$phase = 'main'; |
4670 |
|
|
## reprocess |
4671 |
|
|
redo B; |
4672 |
|
|
} elsif ($token->{type} eq 'end-of-file') { |
4673 |
|
|
## Stop parsing |
4674 |
|
|
last B; |
4675 |
|
|
} else { |
4676 |
|
|
die "$0: $token->{type}: Unknown token"; |
4677 |
|
|
} |
4678 |
|
|
} |
4679 |
|
|
} # B |
4680 |
|
|
|
4681 |
|
|
## Stop parsing # MUST |
4682 |
|
|
|
4683 |
|
|
## TODO: script stuffs |
4684 |
|
|
} # _construct_tree |
4685 |
|
|
|
4686 |
wakaba |
1.8 |
sub get_inner_html ($$$) { |
4687 |
wakaba |
1.2 |
my ($class, $node, $on_error) = @_; |
4688 |
|
|
|
4689 |
|
|
## Step 1 |
4690 |
|
|
my $s = ''; |
4691 |
|
|
|
4692 |
|
|
my $in_cdata; |
4693 |
|
|
my $parent = $node; |
4694 |
|
|
while (defined $parent) { |
4695 |
|
|
if ($parent->node_type == 1 and |
4696 |
|
|
$parent->namespace_uri eq 'http://www.w3.org/1999/xhtml' and |
4697 |
|
|
{ |
4698 |
|
|
style => 1, script => 1, xmp => 1, iframe => 1, |
4699 |
|
|
noembed => 1, noframes => 1, noscript => 1, |
4700 |
|
|
}->{$parent->local_name}) { ## TODO: case thingy |
4701 |
|
|
$in_cdata = 1; |
4702 |
|
|
} |
4703 |
|
|
$parent = $parent->parent_node; |
4704 |
|
|
} |
4705 |
|
|
|
4706 |
|
|
## Step 2 |
4707 |
|
|
my @node = @{$node->child_nodes}; |
4708 |
|
|
C: while (@node) { |
4709 |
|
|
my $child = shift @node; |
4710 |
|
|
unless (ref $child) { |
4711 |
|
|
if ($child eq 'cdata-out') { |
4712 |
|
|
$in_cdata = 0; |
4713 |
|
|
} else { |
4714 |
|
|
$s .= $child; # end tag |
4715 |
|
|
} |
4716 |
|
|
next C; |
4717 |
|
|
} |
4718 |
|
|
|
4719 |
|
|
my $nt = $child->node_type; |
4720 |
|
|
if ($nt == 1) { # Element |
4721 |
|
|
my $tag_name = lc $child->tag_name; ## ISSUE: Definition of "lowercase" |
4722 |
|
|
$s .= '<' . $tag_name; |
4723 |
|
|
|
4724 |
|
|
## ISSUE: Non-html elements |
4725 |
|
|
|
4726 |
|
|
my @attrs = @{$child->attributes}; # sort order MUST be stable |
4727 |
|
|
for my $attr (@attrs) { # order is implementation dependent |
4728 |
|
|
my $attr_name = lc $attr->name; ## ISSUE: Definition of "lowercase" |
4729 |
|
|
$s .= ' ' . $attr_name . '="'; |
4730 |
|
|
my $attr_value = $attr->value; |
4731 |
|
|
## escape |
4732 |
|
|
$attr_value =~ s/&/&/g; |
4733 |
|
|
$attr_value =~ s/</</g; |
4734 |
|
|
$attr_value =~ s/>/>/g; |
4735 |
|
|
$attr_value =~ s/"/"/g; |
4736 |
|
|
$s .= $attr_value . '"'; |
4737 |
|
|
} |
4738 |
|
|
$s .= '>'; |
4739 |
|
|
|
4740 |
|
|
next C if { |
4741 |
|
|
area => 1, base => 1, basefont => 1, bgsound => 1, |
4742 |
|
|
br => 1, col => 1, embed => 1, frame => 1, hr => 1, |
4743 |
|
|
img => 1, input => 1, link => 1, meta => 1, param => 1, |
4744 |
|
|
spacer => 1, wbr => 1, |
4745 |
|
|
}->{$tag_name}; |
4746 |
|
|
|
4747 |
|
|
if (not $in_cdata and { |
4748 |
|
|
style => 1, script => 1, xmp => 1, iframe => 1, |
4749 |
|
|
noembed => 1, noframes => 1, noscript => 1, |
4750 |
|
|
}->{$tag_name}) { |
4751 |
|
|
unshift @node, 'cdata-out'; |
4752 |
|
|
$in_cdata = 1; |
4753 |
|
|
} |
4754 |
|
|
|
4755 |
|
|
unshift @node, @{$child->child_nodes}, '</' . $tag_name . '>'; |
4756 |
|
|
} elsif ($nt == 3 or $nt == 4) { |
4757 |
|
|
if ($in_cdata) { |
4758 |
|
|
$s .= $child->data; |
4759 |
|
|
} else { |
4760 |
|
|
my $value = $child->data; |
4761 |
|
|
$value =~ s/&/&/g; |
4762 |
|
|
$value =~ s/</</g; |
4763 |
|
|
$value =~ s/>/>/g; |
4764 |
|
|
$value =~ s/"/"/g; |
4765 |
|
|
$s .= $value; |
4766 |
|
|
} |
4767 |
|
|
} elsif ($nt == 8) { |
4768 |
|
|
$s .= '<!--' . $child->data . '-->'; |
4769 |
|
|
} elsif ($nt == 10) { |
4770 |
|
|
$s .= '<!DOCTYPE ' . $child->name . '>'; |
4771 |
|
|
} elsif ($nt == 5) { # entrefs |
4772 |
|
|
push @node, @{$child->child_nodes}; |
4773 |
|
|
} else { |
4774 |
wakaba |
1.8 |
$on_error->($child) if defined $on_error; |
4775 |
wakaba |
1.2 |
} |
4776 |
wakaba |
1.8 |
## ISSUE: This code does not support PIs. |
4777 |
wakaba |
1.2 |
} # C |
4778 |
|
|
|
4779 |
|
|
## Step 3 |
4780 |
|
|
return \$s; |
4781 |
wakaba |
1.8 |
} # get_inner_html |
4782 |
wakaba |
1.2 |
|
4783 |
wakaba |
1.1 |
1; |
4784 |
wakaba |
1.9 |
# $Date: 2007/05/01 07:46:42 $ |