| 3 |
our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
| 4 |
use Error qw(:try); |
use Error qw(:try); |
| 5 |
|
|
| 6 |
|
use Whatpm::HTML::Tokenizer; |
| 7 |
|
|
| 8 |
## NOTE: This module don't check all HTML5 parse errors; character |
## NOTE: This module don't check all HTML5 parse errors; character |
| 9 |
## encoding related parse errors are expected to be handled by relevant |
## encoding related parse errors are expected to be handled by relevant |
| 10 |
## modules. |
## modules. |
| 23 |
|
|
| 24 |
require IO::Handle; |
require IO::Handle; |
| 25 |
|
|
| 26 |
|
## Namespace URLs |
| 27 |
|
|
| 28 |
my $HTML_NS = q<http://www.w3.org/1999/xhtml>; |
my $HTML_NS = q<http://www.w3.org/1999/xhtml>; |
| 29 |
my $MML_NS = q<http://www.w3.org/1998/Math/MathML>; |
my $MML_NS = q<http://www.w3.org/1998/Math/MathML>; |
| 30 |
my $SVG_NS = q<http://www.w3.org/2000/svg>; |
my $SVG_NS = q<http://www.w3.org/2000/svg>; |
| 32 |
my $XML_NS = q<http://www.w3.org/XML/1998/namespace>; |
my $XML_NS = q<http://www.w3.org/XML/1998/namespace>; |
| 33 |
my $XMLNS_NS = q<http://www.w3.org/2000/xmlns/>; |
my $XMLNS_NS = q<http://www.w3.org/2000/xmlns/>; |
| 34 |
|
|
| 35 |
sub A_EL () { 0b1 } |
## Element categories |
|
sub ADDRESS_EL () { 0b10 } |
|
|
sub BODY_EL () { 0b100 } |
|
|
sub BUTTON_EL () { 0b1000 } |
|
|
sub CAPTION_EL () { 0b10000 } |
|
|
sub DD_EL () { 0b100000 } |
|
|
sub DIV_EL () { 0b1000000 } |
|
|
sub DT_EL () { 0b10000000 } |
|
|
sub FORM_EL () { 0b100000000 } |
|
|
sub FORMATTING_EL () { 0b1000000000 } |
|
|
sub FRAMESET_EL () { 0b10000000000 } |
|
|
sub HEADING_EL () { 0b100000000000 } |
|
|
sub HTML_EL () { 0b1000000000000 } |
|
|
sub LI_EL () { 0b10000000000000 } |
|
|
sub NOBR_EL () { 0b100000000000000 } |
|
|
sub OPTION_EL () { 0b1000000000000000 } |
|
|
sub OPTGROUP_EL () { 0b10000000000000000 } |
|
|
sub P_EL () { 0b100000000000000000 } |
|
|
sub SELECT_EL () { 0b1000000000000000000 } |
|
|
sub TABLE_EL () { 0b10000000000000000000 } |
|
|
sub TABLE_CELL_EL () { 0b100000000000000000000 } |
|
|
sub TABLE_ROW_EL () { 0b1000000000000000000000 } |
|
|
sub TABLE_ROW_GROUP_EL () { 0b10000000000000000000000 } |
|
|
sub MISC_SCOPING_EL () { 0b100000000000000000000000 } |
|
|
sub MISC_SPECIAL_EL () { 0b1000000000000000000000000 } |
|
|
sub FOREIGN_EL () { 0b10000000000000000000000000 } |
|
|
sub FOREIGN_FLOW_CONTENT_EL () { 0b100000000000000000000000000 } |
|
|
sub MML_AXML_EL () { 0b1000000000000000000000000000 } |
|
|
sub RUBY_EL () { 0b10000000000000000000000000000 } |
|
|
sub RUBY_COMPONENT_EL () { 0b100000000000000000000000000000 } |
|
|
|
|
|
sub TABLE_ROWS_EL () { |
|
|
TABLE_EL | |
|
|
TABLE_ROW_EL | |
|
|
TABLE_ROW_GROUP_EL |
|
|
} |
|
| 36 |
|
|
| 37 |
## NOTE: Used in "generate implied end tags" algorithm. |
## Bits 12-15 |
| 38 |
## NOTE: There is a code where a modified version of END_TAG_OPTIONAL_EL |
sub SPECIAL_EL () { 0b1_000000000000000 } |
| 39 |
## is used in "generate implied end tags" implementation (search for the |
sub SCOPING_EL () { 0b1_00000000000000 } |
| 40 |
## function mae). |
sub FORMATTING_EL () { 0b1_0000000000000 } |
| 41 |
sub END_TAG_OPTIONAL_EL () { |
sub PHRASING_EL () { 0b1_000000000000 } |
| 42 |
DD_EL | |
|
| 43 |
DT_EL | |
## Bits 10-11 |
| 44 |
LI_EL | |
#sub FOREIGN_EL () { 0b1_00000000000 } # see Whatpm::HTML::Tokenizer |
| 45 |
P_EL | |
sub FOREIGN_FLOW_CONTENT_EL () { 0b1_0000000000 } |
| 46 |
RUBY_COMPONENT_EL |
|
| 47 |
} |
## Bits 6-9 |
| 48 |
|
sub TABLE_SCOPING_EL () { 0b1_000000000 } |
| 49 |
|
sub TABLE_ROWS_SCOPING_EL () { 0b1_00000000 } |
| 50 |
|
sub TABLE_ROW_SCOPING_EL () { 0b1_0000000 } |
| 51 |
|
sub TABLE_ROWS_EL () { 0b1_000000 } |
| 52 |
|
|
| 53 |
|
## Bit 5 |
| 54 |
|
sub ADDRESS_DIV_P_EL () { 0b1_00000 } |
| 55 |
|
|
| 56 |
## NOTE: Used in </body> and EOF algorithms. |
## NOTE: Used in </body> and EOF algorithms. |
| 57 |
sub ALL_END_TAG_OPTIONAL_EL () { |
## Bit 4 |
| 58 |
DD_EL | |
sub ALL_END_TAG_OPTIONAL_EL () { 0b1_0000 } |
|
DT_EL | |
|
|
LI_EL | |
|
|
P_EL | |
|
|
|
|
|
BODY_EL | |
|
|
HTML_EL | |
|
|
TABLE_CELL_EL | |
|
|
TABLE_ROW_EL | |
|
|
TABLE_ROW_GROUP_EL |
|
|
} |
|
| 59 |
|
|
| 60 |
sub SCOPING_EL () { |
## NOTE: Used in "generate implied end tags" algorithm. |
| 61 |
BUTTON_EL | |
## NOTE: There is a code where a modified version of |
| 62 |
CAPTION_EL | |
## END_TAG_OPTIONAL_EL is used in "generate implied end tags" |
| 63 |
HTML_EL | |
## implementation (search for the algorithm name). |
| 64 |
TABLE_EL | |
## Bit 3 |
| 65 |
TABLE_CELL_EL | |
sub END_TAG_OPTIONAL_EL () { 0b1_000 } |
| 66 |
MISC_SCOPING_EL |
|
| 67 |
|
## Bits 0-2 |
| 68 |
|
|
| 69 |
|
sub MISC_SPECIAL_EL () { SPECIAL_EL | 0b000 } |
| 70 |
|
sub FORM_EL () { SPECIAL_EL | 0b001 } |
| 71 |
|
sub FRAMESET_EL () { SPECIAL_EL | 0b010 } |
| 72 |
|
sub HEADING_EL () { SPECIAL_EL | 0b011 } |
| 73 |
|
sub SELECT_EL () { SPECIAL_EL | 0b100 } |
| 74 |
|
sub SCRIPT_EL () { SPECIAL_EL | 0b101 } |
| 75 |
|
|
| 76 |
|
sub ADDRESS_DIV_EL () { SPECIAL_EL | ADDRESS_DIV_P_EL | 0b001 } |
| 77 |
|
sub BODY_EL () { SPECIAL_EL | ALL_END_TAG_OPTIONAL_EL | 0b001 } |
| 78 |
|
|
| 79 |
|
sub DTDD_EL () { |
| 80 |
|
SPECIAL_EL | |
| 81 |
|
END_TAG_OPTIONAL_EL | |
| 82 |
|
ALL_END_TAG_OPTIONAL_EL | |
| 83 |
|
0b010 |
| 84 |
} |
} |
| 85 |
|
sub LI_EL () { |
| 86 |
sub TABLE_SCOPING_EL () { |
SPECIAL_EL | |
| 87 |
HTML_EL | |
END_TAG_OPTIONAL_EL | |
| 88 |
TABLE_EL |
ALL_END_TAG_OPTIONAL_EL | |
| 89 |
|
0b100 |
| 90 |
} |
} |
| 91 |
|
sub P_EL () { |
| 92 |
sub TABLE_ROWS_SCOPING_EL () { |
SPECIAL_EL | |
| 93 |
HTML_EL | |
ADDRESS_DIV_P_EL | |
| 94 |
TABLE_ROW_GROUP_EL |
END_TAG_OPTIONAL_EL | |
| 95 |
|
ALL_END_TAG_OPTIONAL_EL | |
| 96 |
|
0b001 |
| 97 |
} |
} |
| 98 |
|
|
| 99 |
sub TABLE_ROW_SCOPING_EL () { |
sub TABLE_ROW_EL () { |
| 100 |
HTML_EL | |
SPECIAL_EL | |
| 101 |
TABLE_ROW_EL |
TABLE_ROWS_EL | |
| 102 |
|
TABLE_ROW_SCOPING_EL | |
| 103 |
|
ALL_END_TAG_OPTIONAL_EL | |
| 104 |
|
0b001 |
| 105 |
|
} |
| 106 |
|
sub TABLE_ROW_GROUP_EL () { |
| 107 |
|
SPECIAL_EL | |
| 108 |
|
TABLE_ROWS_EL | |
| 109 |
|
TABLE_ROWS_SCOPING_EL | |
| 110 |
|
ALL_END_TAG_OPTIONAL_EL | |
| 111 |
|
0b001 |
| 112 |
} |
} |
| 113 |
|
|
| 114 |
sub SPECIAL_EL () { |
sub MISC_SCOPING_EL () { SCOPING_EL | 0b000 } |
| 115 |
ADDRESS_EL | |
sub BUTTON_EL () { SCOPING_EL | 0b001 } |
| 116 |
BODY_EL | |
sub CAPTION_EL () { SCOPING_EL | 0b010 } |
| 117 |
DIV_EL | |
sub HTML_EL () { |
| 118 |
|
SCOPING_EL | |
| 119 |
DD_EL | |
TABLE_SCOPING_EL | |
| 120 |
DT_EL | |
TABLE_ROWS_SCOPING_EL | |
| 121 |
LI_EL | |
TABLE_ROW_SCOPING_EL | |
| 122 |
P_EL | |
ALL_END_TAG_OPTIONAL_EL | |
| 123 |
|
0b001 |
| 124 |
FORM_EL | |
} |
| 125 |
FRAMESET_EL | |
sub TABLE_EL () { |
| 126 |
HEADING_EL | |
SCOPING_EL | |
| 127 |
OPTION_EL | |
TABLE_ROWS_EL | |
| 128 |
OPTGROUP_EL | |
TABLE_SCOPING_EL | |
| 129 |
SELECT_EL | |
0b001 |
|
TABLE_ROW_EL | |
|
|
TABLE_ROW_GROUP_EL | |
|
|
MISC_SPECIAL_EL |
|
| 130 |
} |
} |
| 131 |
|
sub TABLE_CELL_EL () { |
| 132 |
|
SCOPING_EL | |
| 133 |
|
TABLE_ROW_SCOPING_EL | |
| 134 |
|
ALL_END_TAG_OPTIONAL_EL | |
| 135 |
|
0b001 |
| 136 |
|
} |
| 137 |
|
|
| 138 |
|
sub MISC_FORMATTING_EL () { FORMATTING_EL | 0b000 } |
| 139 |
|
sub A_EL () { FORMATTING_EL | 0b001 } |
| 140 |
|
sub NOBR_EL () { FORMATTING_EL | 0b010 } |
| 141 |
|
|
| 142 |
|
sub RUBY_EL () { PHRASING_EL | 0b001 } |
| 143 |
|
|
| 144 |
|
## ISSUE: ALL_END_TAG_OPTIONAL_EL? |
| 145 |
|
sub OPTGROUP_EL () { PHRASING_EL | END_TAG_OPTIONAL_EL | 0b001 } |
| 146 |
|
sub OPTION_EL () { PHRASING_EL | END_TAG_OPTIONAL_EL | 0b010 } |
| 147 |
|
sub RUBY_COMPONENT_EL () { PHRASING_EL | END_TAG_OPTIONAL_EL | 0b100 } |
| 148 |
|
|
| 149 |
|
sub MML_AXML_EL () { PHRASING_EL | FOREIGN_EL | 0b001 } |
| 150 |
|
|
| 151 |
my $el_category = { |
my $el_category = { |
| 152 |
a => A_EL | FORMATTING_EL, |
a => A_EL, |
| 153 |
address => ADDRESS_EL, |
address => ADDRESS_DIV_EL, |
| 154 |
applet => MISC_SCOPING_EL, |
applet => MISC_SCOPING_EL, |
| 155 |
area => MISC_SPECIAL_EL, |
area => MISC_SPECIAL_EL, |
| 156 |
|
article => MISC_SPECIAL_EL, |
| 157 |
|
aside => MISC_SPECIAL_EL, |
| 158 |
b => FORMATTING_EL, |
b => FORMATTING_EL, |
| 159 |
base => MISC_SPECIAL_EL, |
base => MISC_SPECIAL_EL, |
| 160 |
basefont => MISC_SPECIAL_EL, |
basefont => MISC_SPECIAL_EL, |
| 168 |
center => MISC_SPECIAL_EL, |
center => MISC_SPECIAL_EL, |
| 169 |
col => MISC_SPECIAL_EL, |
col => MISC_SPECIAL_EL, |
| 170 |
colgroup => MISC_SPECIAL_EL, |
colgroup => MISC_SPECIAL_EL, |
| 171 |
dd => DD_EL, |
command => MISC_SPECIAL_EL, |
| 172 |
|
datagrid => MISC_SPECIAL_EL, |
| 173 |
|
dd => DTDD_EL, |
| 174 |
|
details => MISC_SPECIAL_EL, |
| 175 |
|
dialog => MISC_SPECIAL_EL, |
| 176 |
dir => MISC_SPECIAL_EL, |
dir => MISC_SPECIAL_EL, |
| 177 |
div => DIV_EL, |
div => ADDRESS_DIV_EL, |
| 178 |
dl => MISC_SPECIAL_EL, |
dl => MISC_SPECIAL_EL, |
| 179 |
dt => DT_EL, |
dt => DTDD_EL, |
| 180 |
em => FORMATTING_EL, |
em => FORMATTING_EL, |
| 181 |
embed => MISC_SPECIAL_EL, |
embed => MISC_SPECIAL_EL, |
| 182 |
fieldset => MISC_SPECIAL_EL, |
fieldset => MISC_SPECIAL_EL, |
| 183 |
|
figure => MISC_SPECIAL_EL, |
| 184 |
font => FORMATTING_EL, |
font => FORMATTING_EL, |
| 185 |
|
footer => MISC_SPECIAL_EL, |
| 186 |
form => FORM_EL, |
form => FORM_EL, |
| 187 |
frame => MISC_SPECIAL_EL, |
frame => MISC_SPECIAL_EL, |
| 188 |
frameset => FRAMESET_EL, |
frameset => FRAMESET_EL, |
| 193 |
h5 => HEADING_EL, |
h5 => HEADING_EL, |
| 194 |
h6 => HEADING_EL, |
h6 => HEADING_EL, |
| 195 |
head => MISC_SPECIAL_EL, |
head => MISC_SPECIAL_EL, |
| 196 |
|
header => MISC_SPECIAL_EL, |
| 197 |
|
hgroup => MISC_SPECIAL_EL, |
| 198 |
hr => MISC_SPECIAL_EL, |
hr => MISC_SPECIAL_EL, |
| 199 |
html => HTML_EL, |
html => HTML_EL, |
| 200 |
i => FORMATTING_EL, |
i => FORMATTING_EL, |
| 201 |
iframe => MISC_SPECIAL_EL, |
iframe => MISC_SPECIAL_EL, |
| 202 |
img => MISC_SPECIAL_EL, |
img => MISC_SPECIAL_EL, |
| 203 |
|
#image => MISC_SPECIAL_EL, ## NOTE: Commented out in the spec. |
| 204 |
input => MISC_SPECIAL_EL, |
input => MISC_SPECIAL_EL, |
| 205 |
isindex => MISC_SPECIAL_EL, |
isindex => MISC_SPECIAL_EL, |
| 206 |
|
## XXX keygen? (Whether a void element is in Special or not does not |
| 207 |
|
## affect to the processing, however.) |
| 208 |
li => LI_EL, |
li => LI_EL, |
| 209 |
link => MISC_SPECIAL_EL, |
link => MISC_SPECIAL_EL, |
| 210 |
listing => MISC_SPECIAL_EL, |
listing => MISC_SPECIAL_EL, |
| 211 |
marquee => MISC_SCOPING_EL, |
marquee => MISC_SCOPING_EL, |
| 212 |
menu => MISC_SPECIAL_EL, |
menu => MISC_SPECIAL_EL, |
| 213 |
meta => MISC_SPECIAL_EL, |
meta => MISC_SPECIAL_EL, |
| 214 |
nobr => NOBR_EL | FORMATTING_EL, |
nav => MISC_SPECIAL_EL, |
| 215 |
|
nobr => NOBR_EL, |
| 216 |
noembed => MISC_SPECIAL_EL, |
noembed => MISC_SPECIAL_EL, |
| 217 |
noframes => MISC_SPECIAL_EL, |
noframes => MISC_SPECIAL_EL, |
| 218 |
noscript => MISC_SPECIAL_EL, |
noscript => MISC_SPECIAL_EL, |
| 230 |
s => FORMATTING_EL, |
s => FORMATTING_EL, |
| 231 |
script => MISC_SPECIAL_EL, |
script => MISC_SPECIAL_EL, |
| 232 |
select => SELECT_EL, |
select => SELECT_EL, |
| 233 |
|
section => MISC_SPECIAL_EL, |
| 234 |
small => FORMATTING_EL, |
small => FORMATTING_EL, |
| 235 |
spacer => MISC_SPECIAL_EL, |
spacer => MISC_SPECIAL_EL, |
| 236 |
strike => FORMATTING_EL, |
strike => FORMATTING_EL, |
| 249 |
u => FORMATTING_EL, |
u => FORMATTING_EL, |
| 250 |
ul => MISC_SPECIAL_EL, |
ul => MISC_SPECIAL_EL, |
| 251 |
wbr => MISC_SPECIAL_EL, |
wbr => MISC_SPECIAL_EL, |
| 252 |
|
xmp => MISC_SPECIAL_EL, |
| 253 |
}; |
}; |
| 254 |
|
|
| 255 |
my $el_category_f = { |
my $el_category_f = { |
| 256 |
$MML_NS => { |
$MML_NS => { |
| 257 |
'annotation-xml' => MML_AXML_EL, |
'annotation-xml' => MML_AXML_EL, |
| 258 |
mi => FOREIGN_FLOW_CONTENT_EL, |
mi => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 259 |
mo => FOREIGN_FLOW_CONTENT_EL, |
mo => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 260 |
mn => FOREIGN_FLOW_CONTENT_EL, |
mn => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 261 |
ms => FOREIGN_FLOW_CONTENT_EL, |
ms => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 262 |
mtext => FOREIGN_FLOW_CONTENT_EL, |
mtext => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 263 |
}, |
}, |
| 264 |
$SVG_NS => { |
$SVG_NS => { |
| 265 |
foreignObject => FOREIGN_FLOW_CONTENT_EL, |
foreignObject => SCOPING_EL | FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 266 |
desc => FOREIGN_FLOW_CONTENT_EL, |
desc => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 267 |
title => FOREIGN_FLOW_CONTENT_EL, |
title => FOREIGN_EL | FOREIGN_FLOW_CONTENT_EL, |
| 268 |
}, |
}, |
| 269 |
## NOTE: In addition, FOREIGN_EL is set to non-HTML elements. |
## NOTE: In addition, FOREIGN_EL is set to non-HTML elements. |
| 270 |
}; |
}; |
| 351 |
|
|
| 352 |
## ISSUE: xmlns:xlink="non-xlink-ns" is not an error. |
## ISSUE: xmlns:xlink="non-xlink-ns" is not an error. |
| 353 |
|
|
| 354 |
my $charref_map = { |
## TODO: Invoke the reset algorithm when a resettable element is |
| 355 |
0x0D => 0x000A, |
## created (cf. HTML5 revision 2259). |
|
0x80 => 0x20AC, |
|
|
0x81 => 0xFFFD, |
|
|
0x82 => 0x201A, |
|
|
0x83 => 0x0192, |
|
|
0x84 => 0x201E, |
|
|
0x85 => 0x2026, |
|
|
0x86 => 0x2020, |
|
|
0x87 => 0x2021, |
|
|
0x88 => 0x02C6, |
|
|
0x89 => 0x2030, |
|
|
0x8A => 0x0160, |
|
|
0x8B => 0x2039, |
|
|
0x8C => 0x0152, |
|
|
0x8D => 0xFFFD, |
|
|
0x8E => 0x017D, |
|
|
0x8F => 0xFFFD, |
|
|
0x90 => 0xFFFD, |
|
|
0x91 => 0x2018, |
|
|
0x92 => 0x2019, |
|
|
0x93 => 0x201C, |
|
|
0x94 => 0x201D, |
|
|
0x95 => 0x2022, |
|
|
0x96 => 0x2013, |
|
|
0x97 => 0x2014, |
|
|
0x98 => 0x02DC, |
|
|
0x99 => 0x2122, |
|
|
0x9A => 0x0161, |
|
|
0x9B => 0x203A, |
|
|
0x9C => 0x0153, |
|
|
0x9D => 0xFFFD, |
|
|
0x9E => 0x017E, |
|
|
0x9F => 0x0178, |
|
|
}; # $charref_map |
|
|
$charref_map->{$_} = 0xFFFD |
|
|
for 0x0000..0x0008, 0x000B, 0x000E..0x001F, 0x007F, |
|
|
0xD800..0xDFFF, 0xFDD0..0xFDDF, ## ISSUE: 0xFDEF |
|
|
0xFFFE, 0xFFFF, 0x1FFFE, 0x1FFFF, 0x2FFFE, 0x2FFFF, 0x3FFFE, 0x3FFFF, |
|
|
0x4FFFE, 0x4FFFF, 0x5FFFE, 0x5FFFF, 0x6FFFE, 0x6FFFF, 0x7FFFE, |
|
|
0x7FFFF, 0x8FFFE, 0x8FFFF, 0x9FFFE, 0x9FFFF, 0xAFFFE, 0xAFFFF, |
|
|
0xBFFFE, 0xBFFFF, 0xCFFFE, 0xCFFFF, 0xDFFFE, 0xDFFFF, 0xEFFFE, |
|
|
0xEFFFF, 0xFFFFE, 0xFFFFF, 0x10FFFE, 0x10FFFF; |
|
| 356 |
|
|
| 357 |
sub parse_byte_string ($$$$;$) { |
sub parse_byte_string ($$$$;$) { |
| 358 |
my $self = shift; |
my $self = shift; |
| 457 |
if (defined $charset_name) { |
if (defined $charset_name) { |
| 458 |
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 459 |
|
|
|
## ISSUE: Unsupported encoding is not ignored according to the spec. |
|
| 460 |
require Whatpm::Charset::DecodeHandle; |
require Whatpm::Charset::DecodeHandle; |
| 461 |
$buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
$buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new |
| 462 |
($byte_stream); |
($byte_stream); |
| 528 |
|
|
| 529 |
if ($char_stream) { # if supported |
if ($char_stream) { # if supported |
| 530 |
## "Change the encoding" algorithm: |
## "Change the encoding" algorithm: |
|
|
|
|
## Step 1 |
|
|
if ($charset->{category} & |
|
|
Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
|
|
$charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
|
|
($char_stream, $e_status) = $charset->get_decode_handle |
|
|
($byte_stream, |
|
|
byte_buffer => \ $buffer->{buffer}); |
|
|
} |
|
|
$charset_name = $charset->get_iana_name; |
|
| 531 |
|
|
| 532 |
## Step 2 |
## Step 1 |
| 533 |
if (defined $self->{input_encoding} and |
if (defined $self->{input_encoding} and |
| 534 |
$self->{input_encoding} eq $charset_name) { |
$self->{input_encoding} eq $charset_name) { |
| 535 |
!!!parse-error (type => 'charset label:matching', |
!!!parse-error (type => 'charset label:matching', |
| 539 |
return; |
return; |
| 540 |
} |
} |
| 541 |
|
|
| 542 |
|
## Step 2 (HTML5 revision 3205) |
| 543 |
|
if (defined $self->{input_encoding} and |
| 544 |
|
Message::Charset::Info->get_by_html_name ($self->{input_encoding}) |
| 545 |
|
->{category} & Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 546 |
|
$self->{confident} = 1; |
| 547 |
|
return; |
| 548 |
|
} |
| 549 |
|
|
| 550 |
|
## Step 3 |
| 551 |
|
if ($charset->{category} & |
| 552 |
|
Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 553 |
|
$charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 554 |
|
($char_stream, $e_status) = $charset->get_decode_handle |
| 555 |
|
($byte_stream, |
| 556 |
|
byte_buffer => \ $buffer->{buffer}); |
| 557 |
|
} |
| 558 |
|
$charset_name = $charset->get_iana_name; |
| 559 |
|
|
| 560 |
!!!parse-error (type => 'charset label detected', |
!!!parse-error (type => 'charset label detected', |
| 561 |
text => $self->{input_encoding}, |
text => $self->{input_encoding}, |
| 562 |
value => $charset_name, |
value => $charset_name, |
| 563 |
level => $self->{level}->{warn}, |
level => $self->{level}->{warn}, |
| 564 |
token => $token); |
token => $token); |
| 565 |
|
|
| 566 |
## Step 3 |
## Step 4 |
| 567 |
# if (can) { |
# if (can) { |
| 568 |
## change the encoding on the fly. |
## change the encoding on the fly. |
| 569 |
#$self->{confident} = 1; |
#$self->{confident} = 1; |
| 570 |
#return; |
#return; |
| 571 |
# } |
# } |
| 572 |
|
|
| 573 |
## Step 4 |
## Step 5 |
| 574 |
throw Whatpm::HTML::RestartParser (); |
throw Whatpm::HTML::RestartParser (); |
| 575 |
} |
} |
| 576 |
}; # $self->{change_encoding} |
}; # $self->{change_encoding} |
| 650 |
|
|
| 651 |
## NOTE: |set_inner_html| copies most of this method's code |
## NOTE: |set_inner_html| copies most of this method's code |
| 652 |
|
|
| 653 |
|
## Confidence: irrelevant. |
| 654 |
$self->{confident} = 1 unless exists $self->{confident}; |
$self->{confident} = 1 unless exists $self->{confident}; |
| 655 |
|
|
| 656 |
$self->{document}->input_encoding ($self->{input_encoding}) |
$self->{document}->input_encoding ($self->{input_encoding}) |
| 657 |
if defined $self->{input_encoding}; |
if defined $self->{input_encoding}; |
| 658 |
## TODO: |{input_encoding}| is needless? |
## TODO: |{input_encoding}| is needless? |
| 813 |
return $self; |
return $self; |
| 814 |
} # new |
} # new |
| 815 |
|
|
| 816 |
sub CM_ENTITY () { 0b001 } # & markup in data |
## Insertion modes |
|
sub CM_LIMITED_MARKUP () { 0b010 } # < markup in data (limited) |
|
|
sub CM_FULL_MARKUP () { 0b100 } # < markup in data (any) |
|
|
|
|
|
sub PLAINTEXT_CONTENT_MODEL () { 0 } |
|
|
sub CDATA_CONTENT_MODEL () { CM_LIMITED_MARKUP } |
|
|
sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP } |
|
|
sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP } |
|
|
|
|
|
sub DATA_STATE () { 0 } |
|
|
#sub ENTITY_DATA_STATE () { 1 } |
|
|
sub TAG_OPEN_STATE () { 2 } |
|
|
sub CLOSE_TAG_OPEN_STATE () { 3 } |
|
|
sub TAG_NAME_STATE () { 4 } |
|
|
sub BEFORE_ATTRIBUTE_NAME_STATE () { 5 } |
|
|
sub ATTRIBUTE_NAME_STATE () { 6 } |
|
|
sub AFTER_ATTRIBUTE_NAME_STATE () { 7 } |
|
|
sub BEFORE_ATTRIBUTE_VALUE_STATE () { 8 } |
|
|
sub ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE () { 9 } |
|
|
sub ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE () { 10 } |
|
|
sub ATTRIBUTE_VALUE_UNQUOTED_STATE () { 11 } |
|
|
#sub ENTITY_IN_ATTRIBUTE_VALUE_STATE () { 12 } |
|
|
sub MARKUP_DECLARATION_OPEN_STATE () { 13 } |
|
|
sub COMMENT_START_STATE () { 14 } |
|
|
sub COMMENT_START_DASH_STATE () { 15 } |
|
|
sub COMMENT_STATE () { 16 } |
|
|
sub COMMENT_END_STATE () { 17 } |
|
|
sub COMMENT_END_DASH_STATE () { 18 } |
|
|
sub BOGUS_COMMENT_STATE () { 19 } |
|
|
sub DOCTYPE_STATE () { 20 } |
|
|
sub BEFORE_DOCTYPE_NAME_STATE () { 21 } |
|
|
sub DOCTYPE_NAME_STATE () { 22 } |
|
|
sub AFTER_DOCTYPE_NAME_STATE () { 23 } |
|
|
sub BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE () { 24 } |
|
|
sub DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE () { 25 } |
|
|
sub DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE () { 26 } |
|
|
sub AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE () { 27 } |
|
|
sub BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 28 } |
|
|
sub DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE () { 29 } |
|
|
sub DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE () { 30 } |
|
|
sub AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 31 } |
|
|
sub BOGUS_DOCTYPE_STATE () { 32 } |
|
|
sub AFTER_ATTRIBUTE_VALUE_QUOTED_STATE () { 33 } |
|
|
sub SELF_CLOSING_START_TAG_STATE () { 34 } |
|
|
sub CDATA_SECTION_STATE () { 35 } |
|
|
sub MD_HYPHEN_STATE () { 36 } # "markup declaration open state" in the spec |
|
|
sub MD_DOCTYPE_STATE () { 37 } # "markup declaration open state" in the spec |
|
|
sub MD_CDATA_STATE () { 38 } # "markup declaration open state" in the spec |
|
|
sub CDATA_RCDATA_CLOSE_TAG_STATE () { 39 } # "close tag open state" in the spec |
|
|
sub CDATA_SECTION_MSE1_STATE () { 40 } # "CDATA section state" in the spec |
|
|
sub CDATA_SECTION_MSE2_STATE () { 41 } # "CDATA section state" in the spec |
|
|
sub PUBLIC_STATE () { 42 } # "after DOCTYPE name state" in the spec |
|
|
sub SYSTEM_STATE () { 43 } # "after DOCTYPE name state" in the spec |
|
|
## NOTE: "Entity data state", "entity in attribute value state", and |
|
|
## "consume a character reference" algorithm are jointly implemented |
|
|
## using the following six states: |
|
|
sub ENTITY_STATE () { 44 } |
|
|
sub ENTITY_HASH_STATE () { 45 } |
|
|
sub NCR_NUM_STATE () { 46 } |
|
|
sub HEXREF_X_STATE () { 47 } |
|
|
sub HEXREF_HEX_STATE () { 48 } |
|
|
sub ENTITY_NAME_STATE () { 49 } |
|
|
sub PCDATA_STATE () { 50 } # "data state" in the spec |
|
|
|
|
|
sub DOCTYPE_TOKEN () { 1 } |
|
|
sub COMMENT_TOKEN () { 2 } |
|
|
sub START_TAG_TOKEN () { 3 } |
|
|
sub END_TAG_TOKEN () { 4 } |
|
|
sub END_OF_FILE_TOKEN () { 5 } |
|
|
sub CHARACTER_TOKEN () { 6 } |
|
| 817 |
|
|
| 818 |
sub AFTER_HTML_IMS () { 0b100 } |
sub AFTER_HTML_IMS () { 0b100 } |
| 819 |
sub HEAD_IMS () { 0b1000 } |
sub HEAD_IMS () { 0b1000 } |
| 824 |
sub BODY_AFTER_IMS () { 0b100000000 } |
sub BODY_AFTER_IMS () { 0b100000000 } |
| 825 |
sub FRAME_IMS () { 0b1000000000 } |
sub FRAME_IMS () { 0b1000000000 } |
| 826 |
sub SELECT_IMS () { 0b10000000000 } |
sub SELECT_IMS () { 0b10000000000 } |
| 827 |
sub IN_FOREIGN_CONTENT_IM () { 0b100000000000 } |
#sub IN_FOREIGN_CONTENT_IM () { 0b100000000000 } # see Whatpm::HTML::Tokenizer |
| 828 |
## NOTE: "in foreign content" insertion mode is special; it is combined |
## NOTE: "in foreign content" insertion mode is special; it is combined |
| 829 |
## with the secondary insertion mode. In this parser, they are stored |
## with the secondary insertion mode. In this parser, they are stored |
| 830 |
## together in the bit-or'ed form. |
## together in the bit-or'ed form. |
| 831 |
|
sub IN_CDATA_RCDATA_IM () { 0b1000000000000 } |
| 832 |
|
## NOTE: "in CDATA/RCDATA" insertion mode is also special; it is |
| 833 |
|
## combined with the original insertion mode. In thie parser, |
| 834 |
|
## they are stored together in the bit-or'ed form. |
| 835 |
|
|
| 836 |
|
sub IM_MASK () { 0b11111111111 } |
| 837 |
|
|
| 838 |
## NOTE: "initial" and "before html" insertion modes have no constants. |
## NOTE: "initial" and "before html" insertion modes have no constants. |
| 839 |
|
|
| 860 |
sub IN_SELECT_IN_TABLE_IM () { SELECT_IMS | 0b10 } |
sub IN_SELECT_IN_TABLE_IM () { SELECT_IMS | 0b10 } |
| 861 |
sub IN_COLUMN_GROUP_IM () { 0b10 } |
sub IN_COLUMN_GROUP_IM () { 0b10 } |
| 862 |
|
|
|
## Implementations MUST act as if state machine in the spec |
|
|
|
|
|
sub _initialize_tokenizer ($) { |
|
|
my $self = shift; |
|
|
$self->{state} = DATA_STATE; # MUST |
|
|
#$self->{s_kwd}; # state keyword - initialized when used |
|
|
#$self->{entity__value}; # initialized when used |
|
|
#$self->{entity__match}; # initialized when used |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # be |
|
|
undef $self->{ct}; # current token |
|
|
undef $self->{ca}; # current attribute |
|
|
undef $self->{last_stag_name}; # last emitted start tag name |
|
|
#$self->{prev_state}; # initialized when used |
|
|
delete $self->{self_closing}; |
|
|
$self->{char_buffer} = ''; |
|
|
$self->{char_buffer_pos} = 0; |
|
|
$self->{nc} = -1; # next input character |
|
|
#$self->{next_nc} |
|
|
!!!next-input-character; |
|
|
$self->{token} = []; |
|
|
# $self->{escape} |
|
|
} # _initialize_tokenizer |
|
|
|
|
|
## A token has: |
|
|
## ->{type} == DOCTYPE_TOKEN, START_TAG_TOKEN, END_TAG_TOKEN, COMMENT_TOKEN, |
|
|
## CHARACTER_TOKEN, or END_OF_FILE_TOKEN |
|
|
## ->{name} (DOCTYPE_TOKEN) |
|
|
## ->{tag_name} (START_TAG_TOKEN, END_TAG_TOKEN) |
|
|
## ->{pubid} (DOCTYPE_TOKEN) |
|
|
## ->{sysid} (DOCTYPE_TOKEN) |
|
|
## ->{quirks} == 1 or 0 (DOCTYPE_TOKEN): "force-quirks" flag |
|
|
## ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN) |
|
|
## ->{name} |
|
|
## ->{value} |
|
|
## ->{has_reference} == 1 or 0 |
|
|
## ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN) |
|
|
## NOTE: The "self-closing flag" is hold as |$self->{self_closing}|. |
|
|
## |->{self_closing}| is used to save the value of |$self->{self_closing}| |
|
|
## while the token is pushed back to the stack. |
|
|
|
|
|
## Emitted token MUST immediately be handled by the tree construction state. |
|
|
|
|
|
## Before each step, UA MAY check to see if either one of the scripts in |
|
|
## "list of scripts that will execute as soon as possible" or the first |
|
|
## script in the "list of scripts that will execute asynchronously", |
|
|
## has completed loading. If one has, then it MUST be executed |
|
|
## and removed from the list. |
|
|
|
|
|
## TODO: Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) |
|
|
## (This requirement was dropped from HTML5 spec, unfortunately.) |
|
|
|
|
|
my $is_space = { |
|
|
0x0009 => 1, # CHARACTER TABULATION (HT) |
|
|
0x000A => 1, # LINE FEED (LF) |
|
|
#0x000B => 0, # LINE TABULATION (VT) |
|
|
0x000C => 1, # FORM FEED (FF) |
|
|
#0x000D => 1, # CARRIAGE RETURN (CR) |
|
|
0x0020 => 1, # SPACE (SP) |
|
|
}; |
|
|
|
|
|
sub _get_next_token ($) { |
|
|
my $self = shift; |
|
|
|
|
|
if ($self->{self_closing}) { |
|
|
!!!parse-error (type => 'nestc', token => $self->{ct}); |
|
|
## NOTE: The |self_closing| flag is only set by start tag token. |
|
|
## In addition, when a start tag token is emitted, it is always set to |
|
|
## |ct|. |
|
|
delete $self->{self_closing}; |
|
|
} |
|
|
|
|
|
if (@{$self->{token}}) { |
|
|
$self->{self_closing} = $self->{token}->[0]->{self_closing}; |
|
|
return shift @{$self->{token}}; |
|
|
} |
|
|
|
|
|
A: { |
|
|
if ($self->{state} == PCDATA_STATE) { |
|
|
## NOTE: Same as |DATA_STATE|, but only for |PCDATA| content model. |
|
|
|
|
|
if ($self->{nc} == 0x0026) { # & |
|
|
!!!cp (0.1); |
|
|
## NOTE: In the spec, the tokenizer is switched to the |
|
|
## "entity data state". In this implementation, the tokenizer |
|
|
## is switched to the |ENTITY_STATE|, which is an implementation |
|
|
## of the "consume a character reference" algorithm. |
|
|
$self->{entity_add} = -1; |
|
|
$self->{prev_state} = DATA_STATE; |
|
|
$self->{state} = ENTITY_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003C) { # < |
|
|
!!!cp (0.2); |
|
|
$self->{state} = TAG_OPEN_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (0.3); |
|
|
!!!emit ({type => END_OF_FILE_TOKEN, |
|
|
line => $self->{line}, column => $self->{column}}); |
|
|
last A; ## TODO: ok? |
|
|
} else { |
|
|
!!!cp (0.4); |
|
|
# |
|
|
} |
|
|
|
|
|
# Anything else |
|
|
my $token = {type => CHARACTER_TOKEN, |
|
|
data => chr $self->{nc}, |
|
|
line => $self->{line}, column => $self->{column}, |
|
|
}; |
|
|
$self->{read_until}->($token->{data}, q[<&], length $token->{data}); |
|
|
|
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
!!!emit ($token); |
|
|
redo A; |
|
|
} elsif ($self->{state} == DATA_STATE) { |
|
|
$self->{s_kwd} = '' unless defined $self->{s_kwd}; |
|
|
if ($self->{nc} == 0x0026) { # & |
|
|
$self->{s_kwd} = ''; |
|
|
if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA |
|
|
not $self->{escape}) { |
|
|
!!!cp (1); |
|
|
## NOTE: In the spec, the tokenizer is switched to the |
|
|
## "entity data state". In this implementation, the tokenizer |
|
|
## is switched to the |ENTITY_STATE|, which is an implementation |
|
|
## of the "consume a character reference" algorithm. |
|
|
$self->{entity_add} = -1; |
|
|
$self->{prev_state} = DATA_STATE; |
|
|
$self->{state} = ENTITY_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (2); |
|
|
# |
|
|
} |
|
|
} elsif ($self->{nc} == 0x002D) { # - |
|
|
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
|
$self->{s_kwd} .= '-'; |
|
|
|
|
|
if ($self->{s_kwd} eq '<!--') { |
|
|
!!!cp (3); |
|
|
$self->{escape} = 1; # unless $self->{escape}; |
|
|
$self->{s_kwd} = '--'; |
|
|
# |
|
|
} elsif ($self->{s_kwd} eq '---') { |
|
|
!!!cp (4); |
|
|
$self->{s_kwd} = '--'; |
|
|
# |
|
|
} else { |
|
|
!!!cp (5); |
|
|
# |
|
|
} |
|
|
} |
|
|
|
|
|
# |
|
|
} elsif ($self->{nc} == 0x0021) { # ! |
|
|
if (length $self->{s_kwd}) { |
|
|
!!!cp (5.1); |
|
|
$self->{s_kwd} .= '!'; |
|
|
# |
|
|
} else { |
|
|
!!!cp (5.2); |
|
|
#$self->{s_kwd} = ''; |
|
|
# |
|
|
} |
|
|
# |
|
|
} elsif ($self->{nc} == 0x003C) { # < |
|
|
if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA |
|
|
(($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA |
|
|
not $self->{escape})) { |
|
|
!!!cp (6); |
|
|
$self->{state} = TAG_OPEN_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (7); |
|
|
$self->{s_kwd} = ''; |
|
|
# |
|
|
} |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{escape} and |
|
|
($self->{content_model} & CM_LIMITED_MARKUP)) { # RCDATA | CDATA |
|
|
if ($self->{s_kwd} eq '--') { |
|
|
!!!cp (8); |
|
|
delete $self->{escape}; |
|
|
} else { |
|
|
!!!cp (9); |
|
|
} |
|
|
} else { |
|
|
!!!cp (10); |
|
|
} |
|
|
|
|
|
$self->{s_kwd} = ''; |
|
|
# |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (11); |
|
|
$self->{s_kwd} = ''; |
|
|
!!!emit ({type => END_OF_FILE_TOKEN, |
|
|
line => $self->{line}, column => $self->{column}}); |
|
|
last A; ## TODO: ok? |
|
|
} else { |
|
|
!!!cp (12); |
|
|
$self->{s_kwd} = ''; |
|
|
# |
|
|
} |
|
|
|
|
|
# Anything else |
|
|
my $token = {type => CHARACTER_TOKEN, |
|
|
data => chr $self->{nc}, |
|
|
line => $self->{line}, column => $self->{column}, |
|
|
}; |
|
|
if ($self->{read_until}->($token->{data}, q[-!<>&], |
|
|
length $token->{data})) { |
|
|
$self->{s_kwd} = ''; |
|
|
} |
|
|
|
|
|
## Stay in the data state. |
|
|
if ($self->{content_model} == PCDATA_CONTENT_MODEL) { |
|
|
!!!cp (13); |
|
|
$self->{state} = PCDATA_STATE; |
|
|
} else { |
|
|
!!!cp (14); |
|
|
## Stay in the state. |
|
|
} |
|
|
!!!next-input-character; |
|
|
!!!emit ($token); |
|
|
redo A; |
|
|
} elsif ($self->{state} == TAG_OPEN_STATE) { |
|
|
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
|
if ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (15); |
|
|
!!!next-input-character; |
|
|
$self->{state} = CLOSE_TAG_OPEN_STATE; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0021) { # ! |
|
|
!!!cp (15.1); |
|
|
$self->{s_kwd} = '<' unless $self->{escape}; |
|
|
# |
|
|
} else { |
|
|
!!!cp (16); |
|
|
# |
|
|
} |
|
|
|
|
|
## reconsume |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '<', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}, |
|
|
}); |
|
|
redo A; |
|
|
} elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA |
|
|
if ($self->{nc} == 0x0021) { # ! |
|
|
!!!cp (17); |
|
|
$self->{state} = MARKUP_DECLARATION_OPEN_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (18); |
|
|
$self->{state} = CLOSE_TAG_OPEN_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) { # A..Z |
|
|
!!!cp (19); |
|
|
$self->{ct} |
|
|
= {type => START_TAG_TOKEN, |
|
|
tag_name => chr ($self->{nc} + 0x0020), |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}}; |
|
|
$self->{state} = TAG_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif (0x0061 <= $self->{nc} and |
|
|
$self->{nc} <= 0x007A) { # a..z |
|
|
!!!cp (20); |
|
|
$self->{ct} = {type => START_TAG_TOKEN, |
|
|
tag_name => chr ($self->{nc}), |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}}; |
|
|
$self->{state} = TAG_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (21); |
|
|
!!!parse-error (type => 'empty start tag', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '<>', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}, |
|
|
}); |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003F) { # ? |
|
|
!!!cp (22); |
|
|
!!!parse-error (type => 'pio', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}); |
|
|
$self->{state} = BOGUS_COMMENT_STATE; |
|
|
$self->{ct} = {type => COMMENT_TOKEN, data => '', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}, |
|
|
}; |
|
|
## $self->{nc} is intentionally left as is |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (23); |
|
|
!!!parse-error (type => 'bare stago', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '<', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}, |
|
|
}); |
|
|
|
|
|
redo A; |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{content_model} in tag open"; |
|
|
} |
|
|
} elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) { |
|
|
## NOTE: The "close tag open state" in the spec is implemented as |
|
|
## |CLOSE_TAG_OPEN_STATE| and |CDATA_RCDATA_CLOSE_TAG_STATE|. |
|
|
|
|
|
my ($l, $c) = ($self->{line_prev}, $self->{column_prev} - 1); # "<"of"</" |
|
|
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
|
if (defined $self->{last_stag_name}) { |
|
|
$self->{state} = CDATA_RCDATA_CLOSE_TAG_STATE; |
|
|
$self->{s_kwd} = ''; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} else { |
|
|
## No start tag token has ever been emitted |
|
|
## NOTE: See <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>. |
|
|
!!!cp (28); |
|
|
$self->{state} = DATA_STATE; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '</', |
|
|
line => $l, column => $c, |
|
|
}); |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
|
|
|
if (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) { # A..Z |
|
|
!!!cp (29); |
|
|
$self->{ct} |
|
|
= {type => END_TAG_TOKEN, |
|
|
tag_name => chr ($self->{nc} + 0x0020), |
|
|
line => $l, column => $c}; |
|
|
$self->{state} = TAG_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif (0x0061 <= $self->{nc} and |
|
|
$self->{nc} <= 0x007A) { # a..z |
|
|
!!!cp (30); |
|
|
$self->{ct} = {type => END_TAG_TOKEN, |
|
|
tag_name => chr ($self->{nc}), |
|
|
line => $l, column => $c}; |
|
|
$self->{state} = TAG_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (31); |
|
|
!!!parse-error (type => 'empty end tag', |
|
|
line => $self->{line_prev}, ## "<" in "</>" |
|
|
column => $self->{column_prev} - 1); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (32); |
|
|
!!!parse-error (type => 'bare etago'); |
|
|
$self->{state} = DATA_STATE; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '</', |
|
|
line => $l, column => $c, |
|
|
}); |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (33); |
|
|
!!!parse-error (type => 'bogus end tag'); |
|
|
$self->{state} = BOGUS_COMMENT_STATE; |
|
|
$self->{ct} = {type => COMMENT_TOKEN, data => '', |
|
|
line => $self->{line_prev}, # "<" of "</" |
|
|
column => $self->{column_prev} - 1, |
|
|
}; |
|
|
## NOTE: $self->{nc} is intentionally left as is. |
|
|
## Although the "anything else" case of the spec not explicitly |
|
|
## states that the next input character is to be reconsumed, |
|
|
## it will be included to the |data| of the comment token |
|
|
## generated from the bogus end tag, as defined in the |
|
|
## "bogus comment state" entry. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == CDATA_RCDATA_CLOSE_TAG_STATE) { |
|
|
my $ch = substr $self->{last_stag_name}, length $self->{s_kwd}, 1; |
|
|
if (length $ch) { |
|
|
my $CH = $ch; |
|
|
$ch =~ tr/a-z/A-Z/; |
|
|
my $nch = chr $self->{nc}; |
|
|
if ($nch eq $ch or $nch eq $CH) { |
|
|
!!!cp (24); |
|
|
## Stay in the state. |
|
|
$self->{s_kwd} .= $nch; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (25); |
|
|
$self->{state} = DATA_STATE; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, |
|
|
data => '</' . $self->{s_kwd}, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}, |
|
|
}); |
|
|
redo A; |
|
|
} |
|
|
} else { # after "<{tag-name}" |
|
|
unless ($is_space->{$self->{nc}} or |
|
|
{ |
|
|
0x003E => 1, # > |
|
|
0x002F => 1, # / |
|
|
-1 => 1, # EOF |
|
|
}->{$self->{nc}}) { |
|
|
!!!cp (26); |
|
|
## Reconsume. |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!emit ({type => CHARACTER_TOKEN, |
|
|
data => '</' . $self->{s_kwd}, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (27); |
|
|
$self->{ct} |
|
|
= {type => END_TAG_TOKEN, |
|
|
tag_name => $self->{last_stag_name}, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}}; |
|
|
$self->{state} = TAG_NAME_STATE; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
} elsif ($self->{state} == TAG_NAME_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (34); |
|
|
$self->{state} = BEFORE_ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (35); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
#if ($self->{ct}->{attributes}) { |
|
|
# ## NOTE: This should never be reached. |
|
|
# !!! cp (36); |
|
|
# !!! parse-error (type => 'end tag attribute'); |
|
|
#} else { |
|
|
!!!cp (37); |
|
|
#} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) { # A..Z |
|
|
!!!cp (38); |
|
|
$self->{ct}->{tag_name} .= chr ($self->{nc} + 0x0020); |
|
|
# start tag or end tag |
|
|
## Stay in this state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (39); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
#if ($self->{ct}->{attributes}) { |
|
|
# ## NOTE: This state should never be reached. |
|
|
# !!! cp (40); |
|
|
# !!! parse-error (type => 'end tag attribute'); |
|
|
#} else { |
|
|
!!!cp (41); |
|
|
#} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (42); |
|
|
$self->{state} = SELF_CLOSING_START_TAG_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (44); |
|
|
$self->{ct}->{tag_name} .= chr $self->{nc}; |
|
|
# start tag or end tag |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BEFORE_ATTRIBUTE_NAME_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (45); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (46); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (47); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
!!!cp (48); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) { # A..Z |
|
|
!!!cp (49); |
|
|
$self->{ca} |
|
|
= {name => chr ($self->{nc} + 0x0020), |
|
|
value => '', |
|
|
line => $self->{line}, column => $self->{column}}; |
|
|
$self->{state} = ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (50); |
|
|
$self->{state} = SELF_CLOSING_START_TAG_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (52); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (53); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
!!!cp (54); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
if ({ |
|
|
0x0022 => 1, # " |
|
|
0x0027 => 1, # ' |
|
|
0x003D => 1, # = |
|
|
}->{$self->{nc}}) { |
|
|
!!!cp (55); |
|
|
!!!parse-error (type => 'bad attribute name'); |
|
|
} else { |
|
|
!!!cp (56); |
|
|
} |
|
|
$self->{ca} |
|
|
= {name => chr ($self->{nc}), |
|
|
value => '', |
|
|
line => $self->{line}, column => $self->{column}}; |
|
|
$self->{state} = ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ATTRIBUTE_NAME_STATE) { |
|
|
my $before_leave = sub { |
|
|
if (exists $self->{ct}->{attributes} # start tag or end tag |
|
|
->{$self->{ca}->{name}}) { # MUST |
|
|
!!!cp (57); |
|
|
!!!parse-error (type => 'duplicate attribute', text => $self->{ca}->{name}, line => $self->{ca}->{line}, column => $self->{ca}->{column}); |
|
|
## Discard $self->{ca} # MUST |
|
|
} else { |
|
|
!!!cp (58); |
|
|
$self->{ct}->{attributes}->{$self->{ca}->{name}} |
|
|
= $self->{ca}; |
|
|
} |
|
|
}; # $before_leave |
|
|
|
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (59); |
|
|
$before_leave->(); |
|
|
$self->{state} = AFTER_ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003D) { # = |
|
|
!!!cp (60); |
|
|
$before_leave->(); |
|
|
$self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
$before_leave->(); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (61); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
!!!cp (62); |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) { # A..Z |
|
|
!!!cp (63); |
|
|
$self->{ca}->{name} .= chr ($self->{nc} + 0x0020); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (64); |
|
|
$before_leave->(); |
|
|
$self->{state} = SELF_CLOSING_START_TAG_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
$before_leave->(); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (66); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (67); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (68); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
if ($self->{nc} == 0x0022 or # " |
|
|
$self->{nc} == 0x0027) { # ' |
|
|
!!!cp (69); |
|
|
!!!parse-error (type => 'bad attribute name'); |
|
|
} else { |
|
|
!!!cp (70); |
|
|
} |
|
|
$self->{ca}->{name} .= chr ($self->{nc}); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == AFTER_ATTRIBUTE_NAME_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (71); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003D) { # = |
|
|
!!!cp (72); |
|
|
$self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (73); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (74); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (75); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) { # A..Z |
|
|
!!!cp (76); |
|
|
$self->{ca} |
|
|
= {name => chr ($self->{nc} + 0x0020), |
|
|
value => '', |
|
|
line => $self->{line}, column => $self->{column}}; |
|
|
$self->{state} = ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (77); |
|
|
$self->{state} = SELF_CLOSING_START_TAG_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (79); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (80); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (81); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
# reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
if ($self->{nc} == 0x0022 or # " |
|
|
$self->{nc} == 0x0027) { # ' |
|
|
!!!cp (78); |
|
|
!!!parse-error (type => 'bad attribute name'); |
|
|
} else { |
|
|
!!!cp (82); |
|
|
} |
|
|
$self->{ca} |
|
|
= {name => chr ($self->{nc}), |
|
|
value => '', |
|
|
line => $self->{line}, column => $self->{column}}; |
|
|
$self->{state} = ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BEFORE_ATTRIBUTE_VALUE_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (83); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0022) { # " |
|
|
!!!cp (84); |
|
|
$self->{state} = ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0026) { # & |
|
|
!!!cp (85); |
|
|
$self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE; |
|
|
## reconsume |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0027) { # ' |
|
|
!!!cp (86); |
|
|
$self->{state} = ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!parse-error (type => 'empty unquoted attribute value'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (87); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (88); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (89); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (90); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (91); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (92); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
if ($self->{nc} == 0x003D) { # = |
|
|
!!!cp (93); |
|
|
!!!parse-error (type => 'bad attribute value'); |
|
|
} else { |
|
|
!!!cp (94); |
|
|
} |
|
|
$self->{ca}->{value} .= chr ($self->{nc}); |
|
|
$self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE) { |
|
|
if ($self->{nc} == 0x0022) { # " |
|
|
!!!cp (95); |
|
|
$self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0026) { # & |
|
|
!!!cp (96); |
|
|
## NOTE: In the spec, the tokenizer is switched to the |
|
|
## "entity in attribute value state". In this implementation, the |
|
|
## tokenizer is switched to the |ENTITY_STATE|, which is an |
|
|
## implementation of the "consume a character reference" algorithm. |
|
|
$self->{prev_state} = $self->{state}; |
|
|
$self->{entity_add} = 0x0022; # " |
|
|
$self->{state} = ENTITY_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed attribute value'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (97); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (98); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (99); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (100); |
|
|
$self->{ca}->{value} .= chr ($self->{nc}); |
|
|
$self->{read_until}->($self->{ca}->{value}, |
|
|
q["&], |
|
|
length $self->{ca}->{value}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE) { |
|
|
if ($self->{nc} == 0x0027) { # ' |
|
|
!!!cp (101); |
|
|
$self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0026) { # & |
|
|
!!!cp (102); |
|
|
## NOTE: In the spec, the tokenizer is switched to the |
|
|
## "entity in attribute value state". In this implementation, the |
|
|
## tokenizer is switched to the |ENTITY_STATE|, which is an |
|
|
## implementation of the "consume a character reference" algorithm. |
|
|
$self->{entity_add} = 0x0027; # ' |
|
|
$self->{prev_state} = $self->{state}; |
|
|
$self->{state} = ENTITY_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed attribute value'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (103); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (104); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (105); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (106); |
|
|
$self->{ca}->{value} .= chr ($self->{nc}); |
|
|
$self->{read_until}->($self->{ca}->{value}, |
|
|
q['&], |
|
|
length $self->{ca}->{value}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ATTRIBUTE_VALUE_UNQUOTED_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (107); |
|
|
$self->{state} = BEFORE_ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0026) { # & |
|
|
!!!cp (108); |
|
|
## NOTE: In the spec, the tokenizer is switched to the |
|
|
## "entity in attribute value state". In this implementation, the |
|
|
## tokenizer is switched to the |ENTITY_STATE|, which is an |
|
|
## implementation of the "consume a character reference" algorithm. |
|
|
$self->{entity_add} = -1; |
|
|
$self->{prev_state} = $self->{state}; |
|
|
$self->{state} = ENTITY_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (109); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (110); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (111); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (112); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (113); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (114); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
if ({ |
|
|
0x0022 => 1, # " |
|
|
0x0027 => 1, # ' |
|
|
0x003D => 1, # = |
|
|
}->{$self->{nc}}) { |
|
|
!!!cp (115); |
|
|
!!!parse-error (type => 'bad attribute value'); |
|
|
} else { |
|
|
!!!cp (116); |
|
|
} |
|
|
$self->{ca}->{value} .= chr ($self->{nc}); |
|
|
$self->{read_until}->($self->{ca}->{value}, |
|
|
q["'=& >], |
|
|
length $self->{ca}->{value}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == AFTER_ATTRIBUTE_VALUE_QUOTED_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (118); |
|
|
$self->{state} = BEFORE_ATTRIBUTE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (119); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (120); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (121); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002F) { # / |
|
|
!!!cp (122); |
|
|
$self->{state} = SELF_CLOSING_START_TAG_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (122.3); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (122.1); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (122.2); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
## Reconsume. |
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp ('124.1'); |
|
|
!!!parse-error (type => 'no space between attributes'); |
|
|
$self->{state} = BEFORE_ATTRIBUTE_NAME_STATE; |
|
|
## reconsume |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == SELF_CLOSING_START_TAG_STATE) { |
|
|
if ($self->{nc} == 0x003E) { # > |
|
|
if ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
!!!cp ('124.2'); |
|
|
!!!parse-error (type => 'nestc', token => $self->{ct}); |
|
|
## TODO: Different type than slash in start tag |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp ('124.4'); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
!!!cp ('124.5'); |
|
|
} |
|
|
## TODO: Test |<title></title/>| |
|
|
} else { |
|
|
!!!cp ('124.3'); |
|
|
$self->{self_closing} = 1; |
|
|
} |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!parse-error (type => 'unclosed tag'); |
|
|
if ($self->{ct}->{type} == START_TAG_TOKEN) { |
|
|
!!!cp (124.7); |
|
|
$self->{last_stag_name} = $self->{ct}->{tag_name}; |
|
|
} elsif ($self->{ct}->{type} == END_TAG_TOKEN) { |
|
|
if ($self->{ct}->{attributes}) { |
|
|
!!!cp (124.5); |
|
|
!!!parse-error (type => 'end tag attribute'); |
|
|
} else { |
|
|
## NOTE: This state should never be reached. |
|
|
!!!cp (124.6); |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{ct}->{type}: Unknown token type"; |
|
|
} |
|
|
$self->{state} = DATA_STATE; |
|
|
## Reconsume. |
|
|
!!!emit ($self->{ct}); # start tag or end tag |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp ('124.4'); |
|
|
!!!parse-error (type => 'nestc'); |
|
|
## TODO: This error type is wrong. |
|
|
$self->{state} = BEFORE_ATTRIBUTE_NAME_STATE; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BOGUS_COMMENT_STATE) { |
|
|
## (only happen if PCDATA state) |
|
|
|
|
|
## NOTE: Unlike spec's "bogus comment state", this implementation |
|
|
## consumes characters one-by-one basis. |
|
|
|
|
|
if ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (124); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (125); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (126); |
|
|
$self->{ct}->{data} .= chr ($self->{nc}); # comment |
|
|
$self->{read_until}->($self->{ct}->{data}, |
|
|
q[>], |
|
|
length $self->{ct}->{data}); |
|
|
|
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) { |
|
|
## (only happen if PCDATA state) |
|
|
|
|
|
if ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (133); |
|
|
$self->{state} = MD_HYPHEN_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0044 or # D |
|
|
$self->{nc} == 0x0064) { # d |
|
|
## ASCII case-insensitive. |
|
|
!!!cp (130); |
|
|
$self->{state} = MD_DOCTYPE_STATE; |
|
|
$self->{s_kwd} = chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and |
|
|
$self->{open_elements}->[-1]->[1] & FOREIGN_EL and |
|
|
$self->{nc} == 0x005B) { # [ |
|
|
!!!cp (135.4); |
|
|
$self->{state} = MD_CDATA_STATE; |
|
|
$self->{s_kwd} = '['; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (136); |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'bogus comment', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1); |
|
|
## Reconsume. |
|
|
$self->{state} = BOGUS_COMMENT_STATE; |
|
|
$self->{ct} = {type => COMMENT_TOKEN, data => '', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1, |
|
|
}; |
|
|
redo A; |
|
|
} elsif ($self->{state} == MD_HYPHEN_STATE) { |
|
|
if ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (127); |
|
|
$self->{ct} = {type => COMMENT_TOKEN, data => '', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 2, |
|
|
}; |
|
|
$self->{state} = COMMENT_START_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (128); |
|
|
!!!parse-error (type => 'bogus comment', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 2); |
|
|
$self->{state} = BOGUS_COMMENT_STATE; |
|
|
## Reconsume. |
|
|
$self->{ct} = {type => COMMENT_TOKEN, |
|
|
data => '-', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 2, |
|
|
}; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == MD_DOCTYPE_STATE) { |
|
|
## ASCII case-insensitive. |
|
|
if ($self->{nc} == [ |
|
|
undef, |
|
|
0x004F, # O |
|
|
0x0043, # C |
|
|
0x0054, # T |
|
|
0x0059, # Y |
|
|
0x0050, # P |
|
|
]->[length $self->{s_kwd}] or |
|
|
$self->{nc} == [ |
|
|
undef, |
|
|
0x006F, # o |
|
|
0x0063, # c |
|
|
0x0074, # t |
|
|
0x0079, # y |
|
|
0x0070, # p |
|
|
]->[length $self->{s_kwd}]) { |
|
|
!!!cp (131); |
|
|
## Stay in the state. |
|
|
$self->{s_kwd} .= chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ((length $self->{s_kwd}) == 6 and |
|
|
($self->{nc} == 0x0045 or # E |
|
|
$self->{nc} == 0x0065)) { # e |
|
|
!!!cp (129); |
|
|
$self->{state} = DOCTYPE_STATE; |
|
|
$self->{ct} = {type => DOCTYPE_TOKEN, |
|
|
quirks => 1, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 7, |
|
|
}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (132); |
|
|
!!!parse-error (type => 'bogus comment', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}); |
|
|
$self->{state} = BOGUS_COMMENT_STATE; |
|
|
## Reconsume. |
|
|
$self->{ct} = {type => COMMENT_TOKEN, |
|
|
data => $self->{s_kwd}, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}, |
|
|
}; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == MD_CDATA_STATE) { |
|
|
if ($self->{nc} == { |
|
|
'[' => 0x0043, # C |
|
|
'[C' => 0x0044, # D |
|
|
'[CD' => 0x0041, # A |
|
|
'[CDA' => 0x0054, # T |
|
|
'[CDAT' => 0x0041, # A |
|
|
}->{$self->{s_kwd}}) { |
|
|
!!!cp (135.1); |
|
|
## Stay in the state. |
|
|
$self->{s_kwd} .= chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{s_kwd} eq '[CDATA' and |
|
|
$self->{nc} == 0x005B) { # [ |
|
|
!!!cp (135.2); |
|
|
$self->{ct} = {type => CHARACTER_TOKEN, |
|
|
data => '', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 7}; |
|
|
$self->{state} = CDATA_SECTION_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (135.3); |
|
|
!!!parse-error (type => 'bogus comment', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}); |
|
|
$self->{state} = BOGUS_COMMENT_STATE; |
|
|
## Reconsume. |
|
|
$self->{ct} = {type => COMMENT_TOKEN, |
|
|
data => $self->{s_kwd}, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1 - length $self->{s_kwd}, |
|
|
}; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == COMMENT_START_STATE) { |
|
|
if ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (137); |
|
|
$self->{state} = COMMENT_START_DASH_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (138); |
|
|
!!!parse-error (type => 'bogus comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (139); |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (140); |
|
|
$self->{ct}->{data} # comment |
|
|
.= chr ($self->{nc}); |
|
|
$self->{state} = COMMENT_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == COMMENT_START_DASH_STATE) { |
|
|
if ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (141); |
|
|
$self->{state} = COMMENT_END_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (142); |
|
|
!!!parse-error (type => 'bogus comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (143); |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (144); |
|
|
$self->{ct}->{data} # comment |
|
|
.= '-' . chr ($self->{nc}); |
|
|
$self->{state} = COMMENT_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == COMMENT_STATE) { |
|
|
if ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (145); |
|
|
$self->{state} = COMMENT_END_DASH_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (146); |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (147); |
|
|
$self->{ct}->{data} .= chr ($self->{nc}); # comment |
|
|
$self->{read_until}->($self->{ct}->{data}, |
|
|
q[-], |
|
|
length $self->{ct}->{data}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == COMMENT_END_DASH_STATE) { |
|
|
if ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (148); |
|
|
$self->{state} = COMMENT_END_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (149); |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (150); |
|
|
$self->{ct}->{data} .= '-' . chr ($self->{nc}); # comment |
|
|
$self->{state} = COMMENT_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == COMMENT_END_STATE) { |
|
|
if ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (151); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x002D) { # - |
|
|
!!!cp (152); |
|
|
!!!parse-error (type => 'dash in comment', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}); |
|
|
$self->{ct}->{data} .= '-'; # comment |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (153); |
|
|
!!!parse-error (type => 'unclosed comment'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # comment |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (154); |
|
|
!!!parse-error (type => 'dash in comment', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}); |
|
|
$self->{ct}->{data} .= '--' . chr ($self->{nc}); # comment |
|
|
$self->{state} = COMMENT_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == DOCTYPE_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (155); |
|
|
$self->{state} = BEFORE_DOCTYPE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (156); |
|
|
!!!parse-error (type => 'no space before DOCTYPE name'); |
|
|
$self->{state} = BEFORE_DOCTYPE_NAME_STATE; |
|
|
## reconsume |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BEFORE_DOCTYPE_NAME_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (157); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (158); |
|
|
!!!parse-error (type => 'no DOCTYPE name'); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE (quirks) |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (159); |
|
|
!!!parse-error (type => 'no DOCTYPE name'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE (quirks) |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (160); |
|
|
$self->{ct}->{name} = chr $self->{nc}; |
|
|
delete $self->{ct}->{quirks}; |
|
|
## ISSUE: "Set the token's name name to the" in the spec |
|
|
$self->{state} = DOCTYPE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == DOCTYPE_NAME_STATE) { |
|
|
## ISSUE: Redundant "First," in the spec. |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (161); |
|
|
$self->{state} = AFTER_DOCTYPE_NAME_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (162); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (163); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (164); |
|
|
$self->{ct}->{name} |
|
|
.= chr ($self->{nc}); # DOCTYPE |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == AFTER_DOCTYPE_NAME_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (165); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (166); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (167); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0050 or # P |
|
|
$self->{nc} == 0x0070) { # p |
|
|
$self->{state} = PUBLIC_STATE; |
|
|
$self->{s_kwd} = chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0053 or # S |
|
|
$self->{nc} == 0x0073) { # s |
|
|
$self->{state} = SYSTEM_STATE; |
|
|
$self->{s_kwd} = chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (180); |
|
|
!!!parse-error (type => 'string after DOCTYPE name'); |
|
|
$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == PUBLIC_STATE) { |
|
|
## ASCII case-insensitive |
|
|
if ($self->{nc} == [ |
|
|
undef, |
|
|
0x0055, # U |
|
|
0x0042, # B |
|
|
0x004C, # L |
|
|
0x0049, # I |
|
|
]->[length $self->{s_kwd}] or |
|
|
$self->{nc} == [ |
|
|
undef, |
|
|
0x0075, # u |
|
|
0x0062, # b |
|
|
0x006C, # l |
|
|
0x0069, # i |
|
|
]->[length $self->{s_kwd}]) { |
|
|
!!!cp (175); |
|
|
## Stay in the state. |
|
|
$self->{s_kwd} .= chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ((length $self->{s_kwd}) == 5 and |
|
|
($self->{nc} == 0x0043 or # C |
|
|
$self->{nc} == 0x0063)) { # c |
|
|
!!!cp (168); |
|
|
$self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (169); |
|
|
!!!parse-error (type => 'string after DOCTYPE name', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} + 1 - length $self->{s_kwd}); |
|
|
$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == SYSTEM_STATE) { |
|
|
## ASCII case-insensitive |
|
|
if ($self->{nc} == [ |
|
|
undef, |
|
|
0x0059, # Y |
|
|
0x0053, # S |
|
|
0x0054, # T |
|
|
0x0045, # E |
|
|
]->[length $self->{s_kwd}] or |
|
|
$self->{nc} == [ |
|
|
undef, |
|
|
0x0079, # y |
|
|
0x0073, # s |
|
|
0x0074, # t |
|
|
0x0065, # e |
|
|
]->[length $self->{s_kwd}]) { |
|
|
!!!cp (170); |
|
|
## Stay in the state. |
|
|
$self->{s_kwd} .= chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ((length $self->{s_kwd}) == 5 and |
|
|
($self->{nc} == 0x004D or # M |
|
|
$self->{nc} == 0x006D)) { # m |
|
|
!!!cp (171); |
|
|
$self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (172); |
|
|
!!!parse-error (type => 'string after DOCTYPE name', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} + 1 - length $self->{s_kwd}); |
|
|
$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (181); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} eq 0x0022) { # " |
|
|
!!!cp (182); |
|
|
$self->{ct}->{pubid} = ''; # DOCTYPE |
|
|
$self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} eq 0x0027) { # ' |
|
|
!!!cp (183); |
|
|
$self->{ct}->{pubid} = ''; # DOCTYPE |
|
|
$self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} eq 0x003E) { # > |
|
|
!!!cp (184); |
|
|
!!!parse-error (type => 'no PUBLIC literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (185); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (186); |
|
|
!!!parse-error (type => 'string after PUBLIC'); |
|
|
$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE) { |
|
|
if ($self->{nc} == 0x0022) { # " |
|
|
!!!cp (187); |
|
|
$self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (188); |
|
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (189); |
|
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (190); |
|
|
$self->{ct}->{pubid} # DOCTYPE |
|
|
.= chr $self->{nc}; |
|
|
$self->{read_until}->($self->{ct}->{pubid}, q[">], |
|
|
length $self->{ct}->{pubid}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE) { |
|
|
if ($self->{nc} == 0x0027) { # ' |
|
|
!!!cp (191); |
|
|
$self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (192); |
|
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (193); |
|
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (194); |
|
|
$self->{ct}->{pubid} # DOCTYPE |
|
|
.= chr $self->{nc}; |
|
|
$self->{read_until}->($self->{ct}->{pubid}, q['>], |
|
|
length $self->{ct}->{pubid}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (195); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0022) { # " |
|
|
!!!cp (196); |
|
|
$self->{ct}->{sysid} = ''; # DOCTYPE |
|
|
$self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0027) { # ' |
|
|
!!!cp (197); |
|
|
$self->{ct}->{sysid} = ''; # DOCTYPE |
|
|
$self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (198); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (199); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (200); |
|
|
!!!parse-error (type => 'string after PUBLIC literal'); |
|
|
$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (201); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0022) { # " |
|
|
!!!cp (202); |
|
|
$self->{ct}->{sysid} = ''; # DOCTYPE |
|
|
$self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x0027) { # ' |
|
|
!!!cp (203); |
|
|
$self->{ct}->{sysid} = ''; # DOCTYPE |
|
|
$self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (204); |
|
|
!!!parse-error (type => 'no SYSTEM literal'); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (205); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (206); |
|
|
!!!parse-error (type => 'string after SYSTEM'); |
|
|
$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE) { |
|
|
if ($self->{nc} == 0x0022) { # " |
|
|
!!!cp (207); |
|
|
$self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (208); |
|
|
!!!parse-error (type => 'unclosed SYSTEM literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (209); |
|
|
!!!parse-error (type => 'unclosed SYSTEM literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (210); |
|
|
$self->{ct}->{sysid} # DOCTYPE |
|
|
.= chr $self->{nc}; |
|
|
$self->{read_until}->($self->{ct}->{sysid}, q[">], |
|
|
length $self->{ct}->{sysid}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE) { |
|
|
if ($self->{nc} == 0x0027) { # ' |
|
|
!!!cp (211); |
|
|
$self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (212); |
|
|
!!!parse-error (type => 'unclosed SYSTEM literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (213); |
|
|
!!!parse-error (type => 'unclosed SYSTEM literal'); |
|
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (214); |
|
|
$self->{ct}->{sysid} # DOCTYPE |
|
|
.= chr $self->{nc}; |
|
|
$self->{read_until}->($self->{ct}->{sysid}, q['>], |
|
|
length $self->{ct}->{sysid}); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE) { |
|
|
if ($is_space->{$self->{nc}}) { |
|
|
!!!cp (215); |
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (216); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (217); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
$self->{ct}->{quirks} = 1; |
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (218); |
|
|
!!!parse-error (type => 'string after SYSTEM literal'); |
|
|
#$self->{ct}->{quirks} = 1; |
|
|
|
|
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == BOGUS_DOCTYPE_STATE) { |
|
|
if ($self->{nc} == 0x003E) { # > |
|
|
!!!cp (219); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
!!!cp (220); |
|
|
!!!parse-error (type => 'unclosed DOCTYPE'); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{ct}); # DOCTYPE |
|
|
|
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (221); |
|
|
my $s = ''; |
|
|
$self->{read_until}->($s, q[>], 0); |
|
|
|
|
|
## Stay in the state |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == CDATA_SECTION_STATE) { |
|
|
## NOTE: "CDATA section state" in the state is jointly implemented |
|
|
## by three states, |CDATA_SECTION_STATE|, |CDATA_SECTION_MSE1_STATE|, |
|
|
## and |CDATA_SECTION_MSE2_STATE|. |
|
|
|
|
|
if ($self->{nc} == 0x005D) { # ] |
|
|
!!!cp (221.1); |
|
|
$self->{state} = CDATA_SECTION_MSE1_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == -1) { |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
if (length $self->{ct}->{data}) { # character |
|
|
!!!cp (221.2); |
|
|
!!!emit ($self->{ct}); # character |
|
|
} else { |
|
|
!!!cp (221.3); |
|
|
## No token to emit. $self->{ct} is discarded. |
|
|
} |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (221.4); |
|
|
$self->{ct}->{data} .= chr $self->{nc}; |
|
|
$self->{read_until}->($self->{ct}->{data}, |
|
|
q<]>, |
|
|
length $self->{ct}->{data}); |
|
|
|
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
|
|
|
## ISSUE: "text tokens" in spec. |
|
|
} elsif ($self->{state} == CDATA_SECTION_MSE1_STATE) { |
|
|
if ($self->{nc} == 0x005D) { # ] |
|
|
!!!cp (221.5); |
|
|
$self->{state} = CDATA_SECTION_MSE2_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (221.6); |
|
|
$self->{ct}->{data} .= ']'; |
|
|
$self->{state} = CDATA_SECTION_STATE; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == CDATA_SECTION_MSE2_STATE) { |
|
|
if ($self->{nc} == 0x003E) { # > |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!next-input-character; |
|
|
if (length $self->{ct}->{data}) { # character |
|
|
!!!cp (221.7); |
|
|
!!!emit ($self->{ct}); # character |
|
|
} else { |
|
|
!!!cp (221.8); |
|
|
## No token to emit. $self->{ct} is discarded. |
|
|
} |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x005D) { # ] |
|
|
!!!cp (221.9); # character |
|
|
$self->{ct}->{data} .= ']'; ## Add first "]" of "]]]". |
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (221.11); |
|
|
$self->{ct}->{data} .= ']]'; # character |
|
|
$self->{state} = CDATA_SECTION_STATE; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ENTITY_STATE) { |
|
|
if ($is_space->{$self->{nc}} or |
|
|
{ |
|
|
0x003C => 1, 0x0026 => 1, -1 => 1, # <, & |
|
|
$self->{entity_add} => 1, |
|
|
}->{$self->{nc}}) { |
|
|
!!!cp (1001); |
|
|
## Don't consume |
|
|
## No error |
|
|
## Return nothing. |
|
|
# |
|
|
} elsif ($self->{nc} == 0x0023) { # # |
|
|
!!!cp (999); |
|
|
$self->{state} = ENTITY_HASH_STATE; |
|
|
$self->{s_kwd} = '#'; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ((0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x005A) or # A..Z |
|
|
(0x0061 <= $self->{nc} and |
|
|
$self->{nc} <= 0x007A)) { # a..z |
|
|
!!!cp (998); |
|
|
require Whatpm::_NamedEntityList; |
|
|
$self->{state} = ENTITY_NAME_STATE; |
|
|
$self->{s_kwd} = chr $self->{nc}; |
|
|
$self->{entity__value} = $self->{s_kwd}; |
|
|
$self->{entity__match} = 0; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (1027); |
|
|
!!!parse-error (type => 'bare ero'); |
|
|
## Return nothing. |
|
|
# |
|
|
} |
|
|
|
|
|
## NOTE: No character is consumed by the "consume a character |
|
|
## reference" algorithm. In other word, there is an "&" character |
|
|
## that does not introduce a character reference, which would be |
|
|
## appended to the parent element or the attribute value in later |
|
|
## process of the tokenizer. |
|
|
|
|
|
if ($self->{prev_state} == DATA_STATE) { |
|
|
!!!cp (997); |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '&', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev}, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (996); |
|
|
$self->{ca}->{value} .= '&'; |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ENTITY_HASH_STATE) { |
|
|
if ($self->{nc} == 0x0078 or # x |
|
|
$self->{nc} == 0x0058) { # X |
|
|
!!!cp (995); |
|
|
$self->{state} = HEXREF_X_STATE; |
|
|
$self->{s_kwd} .= chr $self->{nc}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif (0x0030 <= $self->{nc} and |
|
|
$self->{nc} <= 0x0039) { # 0..9 |
|
|
!!!cp (994); |
|
|
$self->{state} = NCR_NUM_STATE; |
|
|
$self->{s_kwd} = $self->{nc} - 0x0030; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'bare nero', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1); |
|
|
|
|
|
## NOTE: According to the spec algorithm, nothing is returned, |
|
|
## and then "&#" is appended to the parent element or the attribute |
|
|
## value in the later processing. |
|
|
|
|
|
if ($self->{prev_state} == DATA_STATE) { |
|
|
!!!cp (1019); |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, |
|
|
data => '&#', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 1, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (993); |
|
|
$self->{ca}->{value} .= '&#'; |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
} elsif ($self->{state} == NCR_NUM_STATE) { |
|
|
if (0x0030 <= $self->{nc} and |
|
|
$self->{nc} <= 0x0039) { # 0..9 |
|
|
!!!cp (1012); |
|
|
$self->{s_kwd} *= 10; |
|
|
$self->{s_kwd} += $self->{nc} - 0x0030; |
|
|
|
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003B) { # ; |
|
|
!!!cp (1013); |
|
|
!!!next-input-character; |
|
|
# |
|
|
} else { |
|
|
!!!cp (1014); |
|
|
!!!parse-error (type => 'no refc'); |
|
|
## Reconsume. |
|
|
# |
|
|
} |
|
|
|
|
|
my $code = $self->{s_kwd}; |
|
|
my $l = $self->{line_prev}; |
|
|
my $c = $self->{column_prev}; |
|
|
if ($charref_map->{$code}) { |
|
|
!!!cp (1015); |
|
|
!!!parse-error (type => 'invalid character reference', |
|
|
text => (sprintf 'U+%04X', $code), |
|
|
line => $l, column => $c); |
|
|
$code = $charref_map->{$code}; |
|
|
} elsif ($code > 0x10FFFF) { |
|
|
!!!cp (1016); |
|
|
!!!parse-error (type => 'invalid character reference', |
|
|
text => (sprintf 'U-%08X', $code), |
|
|
line => $l, column => $c); |
|
|
$code = 0xFFFD; |
|
|
} |
|
|
|
|
|
if ($self->{prev_state} == DATA_STATE) { |
|
|
!!!cp (992); |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => chr $code, |
|
|
line => $l, column => $c, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (991); |
|
|
$self->{ca}->{value} .= chr $code; |
|
|
$self->{ca}->{has_reference} = 1; |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == HEXREF_X_STATE) { |
|
|
if ((0x0030 <= $self->{nc} and $self->{nc} <= 0x0039) or |
|
|
(0x0041 <= $self->{nc} and $self->{nc} <= 0x0046) or |
|
|
(0x0061 <= $self->{nc} and $self->{nc} <= 0x0066)) { |
|
|
# 0..9, A..F, a..f |
|
|
!!!cp (990); |
|
|
$self->{state} = HEXREF_HEX_STATE; |
|
|
$self->{s_kwd} = 0; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} else { |
|
|
!!!parse-error (type => 'bare hcro', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - 2); |
|
|
|
|
|
## NOTE: According to the spec algorithm, nothing is returned, |
|
|
## and then "&#" followed by "X" or "x" is appended to the parent |
|
|
## element or the attribute value in the later processing. |
|
|
|
|
|
if ($self->{prev_state} == DATA_STATE) { |
|
|
!!!cp (1005); |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, |
|
|
data => '&' . $self->{s_kwd}, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - length $self->{s_kwd}, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (989); |
|
|
$self->{ca}->{value} .= '&' . $self->{s_kwd}; |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
} elsif ($self->{state} == HEXREF_HEX_STATE) { |
|
|
if (0x0030 <= $self->{nc} and $self->{nc} <= 0x0039) { |
|
|
# 0..9 |
|
|
!!!cp (1002); |
|
|
$self->{s_kwd} *= 0x10; |
|
|
$self->{s_kwd} += $self->{nc} - 0x0030; |
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif (0x0061 <= $self->{nc} and |
|
|
$self->{nc} <= 0x0066) { # a..f |
|
|
!!!cp (1003); |
|
|
$self->{s_kwd} *= 0x10; |
|
|
$self->{s_kwd} += $self->{nc} - 0x0060 + 9; |
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif (0x0041 <= $self->{nc} and |
|
|
$self->{nc} <= 0x0046) { # A..F |
|
|
!!!cp (1004); |
|
|
$self->{s_kwd} *= 0x10; |
|
|
$self->{s_kwd} += $self->{nc} - 0x0040 + 9; |
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} elsif ($self->{nc} == 0x003B) { # ; |
|
|
!!!cp (1006); |
|
|
!!!next-input-character; |
|
|
# |
|
|
} else { |
|
|
!!!cp (1007); |
|
|
!!!parse-error (type => 'no refc', |
|
|
line => $self->{line}, |
|
|
column => $self->{column}); |
|
|
## Reconsume. |
|
|
# |
|
|
} |
|
|
|
|
|
my $code = $self->{s_kwd}; |
|
|
my $l = $self->{line_prev}; |
|
|
my $c = $self->{column_prev}; |
|
|
if ($charref_map->{$code}) { |
|
|
!!!cp (1008); |
|
|
!!!parse-error (type => 'invalid character reference', |
|
|
text => (sprintf 'U+%04X', $code), |
|
|
line => $l, column => $c); |
|
|
$code = $charref_map->{$code}; |
|
|
} elsif ($code > 0x10FFFF) { |
|
|
!!!cp (1009); |
|
|
!!!parse-error (type => 'invalid character reference', |
|
|
text => (sprintf 'U-%08X', $code), |
|
|
line => $l, column => $c); |
|
|
$code = 0xFFFD; |
|
|
} |
|
|
|
|
|
if ($self->{prev_state} == DATA_STATE) { |
|
|
!!!cp (988); |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => chr $code, |
|
|
line => $l, column => $c, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (987); |
|
|
$self->{ca}->{value} .= chr $code; |
|
|
$self->{ca}->{has_reference} = 1; |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} elsif ($self->{state} == ENTITY_NAME_STATE) { |
|
|
if (length $self->{s_kwd} < 30 and |
|
|
## NOTE: Some number greater than the maximum length of entity name |
|
|
((0x0041 <= $self->{nc} and # a |
|
|
$self->{nc} <= 0x005A) or # x |
|
|
(0x0061 <= $self->{nc} and # a |
|
|
$self->{nc} <= 0x007A) or # z |
|
|
(0x0030 <= $self->{nc} and # 0 |
|
|
$self->{nc} <= 0x0039) or # 9 |
|
|
$self->{nc} == 0x003B)) { # ; |
|
|
our $EntityChar; |
|
|
$self->{s_kwd} .= chr $self->{nc}; |
|
|
if (defined $EntityChar->{$self->{s_kwd}}) { |
|
|
if ($self->{nc} == 0x003B) { # ; |
|
|
!!!cp (1020); |
|
|
$self->{entity__value} = $EntityChar->{$self->{s_kwd}}; |
|
|
$self->{entity__match} = 1; |
|
|
!!!next-input-character; |
|
|
# |
|
|
} else { |
|
|
!!!cp (1021); |
|
|
$self->{entity__value} = $EntityChar->{$self->{s_kwd}}; |
|
|
$self->{entity__match} = -1; |
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} else { |
|
|
!!!cp (1022); |
|
|
$self->{entity__value} .= chr $self->{nc}; |
|
|
$self->{entity__match} *= 2; |
|
|
## Stay in the state. |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} |
|
|
} |
|
|
|
|
|
my $data; |
|
|
my $has_ref; |
|
|
if ($self->{entity__match} > 0) { |
|
|
!!!cp (1023); |
|
|
$data = $self->{entity__value}; |
|
|
$has_ref = 1; |
|
|
# |
|
|
} elsif ($self->{entity__match} < 0) { |
|
|
!!!parse-error (type => 'no refc'); |
|
|
if ($self->{prev_state} != DATA_STATE and # in attribute |
|
|
$self->{entity__match} < -1) { |
|
|
!!!cp (1024); |
|
|
$data = '&' . $self->{s_kwd}; |
|
|
# |
|
|
} else { |
|
|
!!!cp (1025); |
|
|
$data = $self->{entity__value}; |
|
|
$has_ref = 1; |
|
|
# |
|
|
} |
|
|
} else { |
|
|
!!!cp (1026); |
|
|
!!!parse-error (type => 'bare ero', |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} - length $self->{s_kwd}); |
|
|
$data = '&' . $self->{s_kwd}; |
|
|
# |
|
|
} |
|
|
|
|
|
## NOTE: In these cases, when a character reference is found, |
|
|
## it is consumed and a character token is returned, or, otherwise, |
|
|
## nothing is consumed and returned, according to the spec algorithm. |
|
|
## In this implementation, anything that has been examined by the |
|
|
## tokenizer is appended to the parent element or the attribute value |
|
|
## as string, either literal string when no character reference or |
|
|
## entity-replaced string otherwise, in this stage, since any characters |
|
|
## that would not be consumed are appended in the data state or in an |
|
|
## appropriate attribute value state anyway. |
|
|
|
|
|
if ($self->{prev_state} == DATA_STATE) { |
|
|
!!!cp (986); |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
!!!emit ({type => CHARACTER_TOKEN, |
|
|
data => $data, |
|
|
line => $self->{line_prev}, |
|
|
column => $self->{column_prev} + 1 - length $self->{s_kwd}, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (985); |
|
|
$self->{ca}->{value} .= $data; |
|
|
$self->{ca}->{has_reference} = 1 if $has_ref; |
|
|
$self->{state} = $self->{prev_state}; |
|
|
## Reconsume. |
|
|
redo A; |
|
|
} |
|
|
} else { |
|
|
die "$0: $self->{state}: Unknown state"; |
|
|
} |
|
|
} # A |
|
|
|
|
|
die "$0: _get_next_token: unexpected case"; |
|
|
} # _get_next_token |
|
|
|
|
| 863 |
sub _initialize_tree_constructor ($) { |
sub _initialize_tree_constructor ($) { |
| 864 |
my $self = shift; |
my $self = shift; |
| 865 |
## NOTE: $self->{document} MUST be specified before this method is called |
## NOTE: $self->{document} MUST be specified before this method is called |
| 869 |
$self->{document}->manakai_is_html (1); # MUST |
$self->{document}->manakai_is_html (1); # MUST |
| 870 |
$self->{document}->set_user_data (manakai_source_line => 1); |
$self->{document}->set_user_data (manakai_source_line => 1); |
| 871 |
$self->{document}->set_user_data (manakai_source_column => 1); |
$self->{document}->set_user_data (manakai_source_column => 1); |
| 872 |
|
|
| 873 |
|
$self->{frameset_ok} = 1; |
| 874 |
} # _initialize_tree_constructor |
} # _initialize_tree_constructor |
| 875 |
|
|
| 876 |
sub _terminate_tree_constructor ($) { |
sub _terminate_tree_constructor ($) { |
| 890 |
## When an interactive UA render the $self->{document} available |
## When an interactive UA render the $self->{document} available |
| 891 |
## to the user, or when it begin accepting user input, are |
## to the user, or when it begin accepting user input, are |
| 892 |
## not defined. |
## not defined. |
|
|
|
|
## Append a character: collect it and all subsequent consecutive |
|
|
## characters and insert one Text node whose data is concatenation |
|
|
## of all those characters. # MUST |
|
| 893 |
|
|
| 894 |
!!!next-token; |
!!!next-token; |
| 895 |
|
|
| 896 |
undef $self->{form_element}; |
undef $self->{form_element}; |
| 897 |
undef $self->{head_element}; |
undef $self->{head_element}; |
| 898 |
|
undef $self->{head_element_inserted}; |
| 899 |
$self->{open_elements} = []; |
$self->{open_elements} = []; |
| 900 |
undef $self->{inner_html_node}; |
undef $self->{inner_html_node}; |
| 901 |
|
undef $self->{ignore_newline}; |
| 902 |
|
|
| 903 |
## NOTE: The "initial" insertion mode. |
## NOTE: The "initial" insertion mode. |
| 904 |
$self->_tree_construction_initial; # MUST |
$self->_tree_construction_initial; # MUST |
| 918 |
|
|
| 919 |
INITIAL: { |
INITIAL: { |
| 920 |
if ($token->{type} == DOCTYPE_TOKEN) { |
if ($token->{type} == DOCTYPE_TOKEN) { |
| 921 |
## NOTE: Conformance checkers MAY, instead of reporting "not HTML5" |
## NOTE: Conformance checkers MAY, instead of reporting "not |
| 922 |
## error, switch to a conformance checking mode for another |
## HTML5" error, switch to a conformance checking mode for |
| 923 |
## language. |
## another language. (We don't support such mode switchings; it |
| 924 |
|
## is nonsense to do anything different from what browsers do.) |
| 925 |
my $doctype_name = $token->{name}; |
my $doctype_name = $token->{name}; |
| 926 |
$doctype_name = '' unless defined $doctype_name; |
$doctype_name = '' unless defined $doctype_name; |
| 927 |
$doctype_name =~ tr/a-z/A-Z/; # ASCII case-insensitive |
my $doctype = $self->{document}->create_document_type_definition |
| 928 |
if (not defined $token->{name} or # <!DOCTYPE> |
($doctype_name); |
| 929 |
defined $token->{sysid}) { |
|
| 930 |
|
$doctype_name =~ tr/A-Z/a-z/; # ASCII case-insensitive |
| 931 |
|
if ($doctype_name ne 'html') { |
| 932 |
!!!cp ('t1'); |
!!!cp ('t1'); |
| 933 |
!!!parse-error (type => 'not HTML5', token => $token); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 934 |
} elsif ($doctype_name ne 'HTML') { |
} elsif (defined $token->{pubid}) { |
| 935 |
!!!cp ('t2'); |
!!!cp ('t2'); |
| 936 |
|
## XXX Obsolete permitted DOCTYPEs |
| 937 |
!!!parse-error (type => 'not HTML5', token => $token); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 938 |
} elsif (defined $token->{pubid}) { |
} elsif (defined $token->{sysid}) { |
| 939 |
if ($token->{pubid} eq 'XSLT-compat') { |
if ($token->{sysid} eq 'about:legacy-compat') { |
| 940 |
!!!cp ('t1.2'); |
!!!cp ('t1.2'); ## <!DOCTYPE HTML SYSTEM "about:legacy-compat"> |
| 941 |
!!!parse-error (type => 'XSLT-compat', token => $token, |
!!!parse-error (type => 'XSLT-compat', token => $token, |
| 942 |
level => $self->{level}->{should}); |
level => $self->{level}->{should}); |
| 943 |
} else { |
} else { |
| 944 |
!!!parse-error (type => 'not HTML5', token => $token); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 945 |
} |
} |
| 946 |
} else { |
} else { ## <!DOCTYPE HTML> |
| 947 |
!!!cp ('t3'); |
!!!cp ('t3'); |
| 948 |
# |
# |
| 949 |
} |
} |
| 950 |
|
|
|
my $doctype = $self->{document}->create_document_type_definition |
|
|
($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)? |
|
| 951 |
## NOTE: Default value for both |public_id| and |system_id| attributes |
## NOTE: Default value for both |public_id| and |system_id| attributes |
| 952 |
## are empty strings, so that we don't set any value in missing cases. |
## are empty strings, so that we don't set any value in missing cases. |
| 953 |
$doctype->public_id ($token->{pubid}) if defined $token->{pubid}; |
$doctype->public_id ($token->{pubid}) if defined $token->{pubid}; |
| 954 |
$doctype->system_id ($token->{sysid}) if defined $token->{sysid}; |
$doctype->system_id ($token->{sysid}) if defined $token->{sysid}; |
| 955 |
|
|
| 956 |
## NOTE: Other DocumentType attributes are null or empty lists. |
## NOTE: Other DocumentType attributes are null or empty lists. |
| 957 |
## ISSUE: internalSubset = null?? |
## In Firefox3, |internalSubset| attribute is set to the empty |
| 958 |
|
## string, while |null| is an allowed value for the attribute |
| 959 |
|
## according to DOM3 Core. |
| 960 |
$self->{document}->append_child ($doctype); |
$self->{document}->append_child ($doctype); |
| 961 |
|
|
| 962 |
if ($token->{quirks} or $doctype_name ne 'HTML') { |
if ($token->{quirks} or $doctype_name ne 'html') { |
| 963 |
!!!cp ('t4'); |
!!!cp ('t4'); |
| 964 |
$self->{document}->manakai_compat_mode ('quirks'); |
$self->{document}->manakai_compat_mode ('quirks'); |
| 965 |
} elsif (defined $token->{pubid}) { |
} elsif (defined $token->{pubid}) { |
| 1209 |
## NOTE: Reprocess the token. |
## NOTE: Reprocess the token. |
| 1210 |
!!!ack-later; |
!!!ack-later; |
| 1211 |
return; ## Go to the "before head" insertion mode. |
return; ## Go to the "before head" insertion mode. |
|
|
|
|
## ISSUE: There is an issue in the spec |
|
| 1212 |
} # B |
} # B |
| 1213 |
|
|
| 1214 |
die "$0: _tree_construction_root_element: This should never be reached"; |
die "$0: _tree_construction_root_element: This should never be reached"; |
| 1244 |
## SVG elements. Currently the HTML syntax supports only MathML and |
## SVG elements. Currently the HTML syntax supports only MathML and |
| 1245 |
## SVG elements as foreigners. |
## SVG elements as foreigners. |
| 1246 |
$new_mode = IN_BODY_IM | IN_FOREIGN_CONTENT_IM; |
$new_mode = IN_BODY_IM | IN_FOREIGN_CONTENT_IM; |
| 1247 |
} elsif ($node->[1] & TABLE_CELL_EL) { |
} elsif ($node->[1] == TABLE_CELL_EL) { |
| 1248 |
if ($last) { |
if ($last) { |
| 1249 |
!!!cp ('t28.2'); |
!!!cp ('t28.2'); |
| 1250 |
# |
# |
| 1273 |
$self->{insertion_mode} = $new_mode and return if defined $new_mode; |
$self->{insertion_mode} = $new_mode and return if defined $new_mode; |
| 1274 |
|
|
| 1275 |
## Step 15 |
## Step 15 |
| 1276 |
if ($node->[1] & HTML_EL) { |
if ($node->[1] == HTML_EL) { |
| 1277 |
unless (defined $self->{head_element}) { |
unless (defined $self->{head_element}) { |
| 1278 |
!!!cp ('t29'); |
!!!cp ('t29'); |
| 1279 |
$self->{insertion_mode} = BEFORE_HEAD_IM; |
$self->{insertion_mode} = BEFORE_HEAD_IM; |
| 1405 |
|
|
| 1406 |
## Step 1 |
## Step 1 |
| 1407 |
my $start_tag_name = $token->{tag_name}; |
my $start_tag_name = $token->{tag_name}; |
| 1408 |
my $el; |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
|
!!!create-element ($el, $HTML_NS, $start_tag_name, $token->{attributes}, $token); |
|
| 1409 |
|
|
| 1410 |
## Step 2 |
## Step 2 |
|
$insert->($el); |
|
|
|
|
|
## Step 3 |
|
| 1411 |
$self->{content_model} = $content_model_flag; # CDATA or RCDATA |
$self->{content_model} = $content_model_flag; # CDATA or RCDATA |
| 1412 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
| 1413 |
|
|
| 1414 |
## Step 4 |
## Step 3, 4 |
| 1415 |
my $text = ''; |
$self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
|
!!!nack ('t40.1'); |
|
|
!!!next-token; |
|
|
while ($token->{type} == CHARACTER_TOKEN) { # or until stop tokenizing |
|
|
!!!cp ('t40'); |
|
|
$text .= $token->{data}; |
|
|
!!!next-token; |
|
|
} |
|
|
|
|
|
## Step 5 |
|
|
if (length $text) { |
|
|
!!!cp ('t41'); |
|
|
my $text = $self->{document}->create_text_node ($text); |
|
|
$el->append_child ($text); |
|
|
} |
|
|
|
|
|
## Step 6 |
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; |
|
| 1416 |
|
|
| 1417 |
## Step 7 |
!!!nack ('t40.1'); |
|
if ($token->{type} == END_TAG_TOKEN and |
|
|
$token->{tag_name} eq $start_tag_name) { |
|
|
!!!cp ('t42'); |
|
|
## Ignore the token |
|
|
} else { |
|
|
## NOTE: An end-of-file token. |
|
|
if ($content_model_flag == CDATA_CONTENT_MODEL) { |
|
|
!!!cp ('t43'); |
|
|
!!!parse-error (type => 'in CDATA:#eof', token => $token); |
|
|
} elsif ($content_model_flag == RCDATA_CONTENT_MODEL) { |
|
|
!!!cp ('t44'); |
|
|
!!!parse-error (type => 'in RCDATA:#eof', token => $token); |
|
|
} else { |
|
|
die "$0: $content_model_flag in parse_rcdata"; |
|
|
} |
|
|
} |
|
| 1418 |
!!!next-token; |
!!!next-token; |
| 1419 |
}; # $parse_rcdata |
}; # $parse_rcdata |
| 1420 |
|
|
| 1421 |
my $script_start_tag = sub () { |
my $script_start_tag = sub () { |
| 1422 |
|
## Step 1 |
| 1423 |
my $script_el; |
my $script_el; |
| 1424 |
!!!create-element ($script_el, $HTML_NS, 'script', $token->{attributes}, $token); |
!!!create-element ($script_el, $HTML_NS, 'script', $token->{attributes}, $token); |
| 1425 |
|
|
| 1426 |
|
## Step 2 |
| 1427 |
## TODO: mark as "parser-inserted" |
## TODO: mark as "parser-inserted" |
| 1428 |
|
|
| 1429 |
|
## Step 3 |
| 1430 |
|
## TODO: Mark as "already executed", if ... |
| 1431 |
|
|
| 1432 |
|
## Step 4 (HTML5 revision 2702) |
| 1433 |
|
$insert->($script_el); |
| 1434 |
|
push @{$self->{open_elements}}, [$script_el, $el_category->{script}]; |
| 1435 |
|
|
| 1436 |
|
## Step 5 |
| 1437 |
$self->{content_model} = CDATA_CONTENT_MODEL; |
$self->{content_model} = CDATA_CONTENT_MODEL; |
| 1438 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
|
|
|
|
my $text = ''; |
|
|
!!!nack ('t45.1'); |
|
|
!!!next-token; |
|
|
while ($token->{type} == CHARACTER_TOKEN) { |
|
|
!!!cp ('t45'); |
|
|
$text .= $token->{data}; |
|
|
!!!next-token; |
|
|
} # stop if non-character token or tokenizer stops tokenising |
|
|
if (length $text) { |
|
|
!!!cp ('t46'); |
|
|
$script_el->manakai_append_text ($text); |
|
|
} |
|
|
|
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; |
|
| 1439 |
|
|
| 1440 |
if ($token->{type} == END_TAG_TOKEN and |
## Step 6-7 |
| 1441 |
$token->{tag_name} eq 'script') { |
$self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
|
!!!cp ('t47'); |
|
|
## Ignore the token |
|
|
} else { |
|
|
!!!cp ('t48'); |
|
|
!!!parse-error (type => 'in CDATA:#eof', token => $token); |
|
|
## ISSUE: And ignore? |
|
|
## TODO: mark as "already executed" |
|
|
} |
|
|
|
|
|
if (defined $self->{inner_html_node}) { |
|
|
!!!cp ('t49'); |
|
|
## TODO: mark as "already executed" |
|
|
} else { |
|
|
!!!cp ('t50'); |
|
|
## TODO: $old_insertion_point = current insertion point |
|
|
## TODO: insertion point = just before the next input character |
|
| 1442 |
|
|
| 1443 |
$insert->($script_el); |
!!!nack ('t40.2'); |
|
|
|
|
## TODO: insertion point = $old_insertion_point (might be "undefined") |
|
|
|
|
|
## TODO: if there is a script that will execute as soon as the parser resume, then... |
|
|
} |
|
|
|
|
| 1444 |
!!!next-token; |
!!!next-token; |
| 1445 |
}; # $script_start_tag |
}; # $script_start_tag |
| 1446 |
|
|
| 1447 |
## NOTE: $open_tables->[-1]->[0] is the "current table" element node. |
## NOTE: $open_tables->[-1]->[0] is the "current table" element node. |
| 1448 |
## NOTE: $open_tables->[-1]->[1] is the "tainted" flag. |
## NOTE: $open_tables->[-1]->[1] is the "tainted" flag (OBSOLETE; unused). |
| 1449 |
|
## NOTE: $open_tables->[-1]->[2] is set false when non-Text node inserted. |
| 1450 |
my $open_tables = [[$self->{open_elements}->[0]->[0]]]; |
my $open_tables = [[$self->{open_elements}->[0]->[0]]]; |
| 1451 |
|
|
| 1452 |
my $formatting_end_tag = sub { |
my $formatting_end_tag = sub { |
| 1531 |
!!!cp ('t59'); |
!!!cp ('t59'); |
| 1532 |
$furthest_block = $node; |
$furthest_block = $node; |
| 1533 |
$furthest_block_i_in_open = $_; |
$furthest_block_i_in_open = $_; |
| 1534 |
|
## NOTE: The topmost (eldest) node. |
| 1535 |
} elsif ($node->[0] eq $formatting_element->[0]) { |
} elsif ($node->[0] eq $formatting_element->[0]) { |
| 1536 |
!!!cp ('t60'); |
!!!cp ('t60'); |
| 1537 |
last OE; |
last OE; |
| 1615 |
|
|
| 1616 |
## Step 8 |
## Step 8 |
| 1617 |
if ($common_ancestor_node->[1] & TABLE_ROWS_EL) { |
if ($common_ancestor_node->[1] & TABLE_ROWS_EL) { |
| 1618 |
|
## Foster parenting. |
| 1619 |
my $foster_parent_element; |
my $foster_parent_element; |
| 1620 |
my $next_sibling; |
my $next_sibling; |
| 1621 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1622 |
if ($self->{open_elements}->[$_]->[1] & TABLE_EL) { |
if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1623 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
!!!cp ('t65.2'); |
| 1624 |
if (defined $parent and $parent->node_type == 1) { |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1625 |
!!!cp ('t65.1'); |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1626 |
$foster_parent_element = $parent; |
undef $next_sibling |
| 1627 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
unless $next_sibling->parent_node eq $foster_parent_element; |
| 1628 |
} else { |
last OE; |
| 1629 |
!!!cp ('t65.2'); |
} |
| 1630 |
$foster_parent_element |
} # OE |
| 1631 |
= $self->{open_elements}->[$_ - 1]->[0]; |
$foster_parent_element ||= $self->{open_elements}->[0]->[0]; |
| 1632 |
} |
|
|
last OE; |
|
|
} |
|
|
} # OE |
|
|
$foster_parent_element = $self->{open_elements}->[0]->[0] |
|
|
unless defined $foster_parent_element; |
|
| 1633 |
$foster_parent_element->insert_before ($last_node->[0], $next_sibling); |
$foster_parent_element->insert_before ($last_node->[0], $next_sibling); |
| 1634 |
$open_tables->[-1]->[1] = 1; # tainted |
$open_tables->[-1]->[1] = 1; # tainted |
| 1635 |
} else { |
} else { |
| 1674 |
$i = $_; |
$i = $_; |
| 1675 |
} |
} |
| 1676 |
} # OE |
} # OE |
| 1677 |
splice @{$self->{open_elements}}, $i + 1, 1, $clone; |
splice @{$self->{open_elements}}, $i + 1, 0, $clone; |
| 1678 |
|
|
| 1679 |
## Step 14 |
## Step 14 |
| 1680 |
redo FET; |
redo FET; |
| 1685 |
$self->{open_elements}->[-1]->[0]->append_child ($_[0]); |
$self->{open_elements}->[-1]->[0]->append_child ($_[0]); |
| 1686 |
}; # $insert_to_current |
}; # $insert_to_current |
| 1687 |
|
|
| 1688 |
|
## Foster parenting. Note that there are three "foster parenting" |
| 1689 |
|
## code in the parser: for elements (this one), for texts, and for |
| 1690 |
|
## elements in the AAA code. |
| 1691 |
my $insert_to_foster = sub { |
my $insert_to_foster = sub { |
| 1692 |
my $child = shift; |
my $child = shift; |
| 1693 |
if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1695 |
my $foster_parent_element; |
my $foster_parent_element; |
| 1696 |
my $next_sibling; |
my $next_sibling; |
| 1697 |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1698 |
if ($self->{open_elements}->[$_]->[1] & TABLE_EL) { |
if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1699 |
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
!!!cp ('t71'); |
| 1700 |
if (defined $parent and $parent->node_type == 1) { |
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1701 |
!!!cp ('t70'); |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1702 |
$foster_parent_element = $parent; |
undef $next_sibling |
| 1703 |
$next_sibling = $self->{open_elements}->[$_]->[0]; |
unless $next_sibling->parent_node eq $foster_parent_element; |
| 1704 |
} else { |
last OE; |
| 1705 |
!!!cp ('t71'); |
} |
| 1706 |
$foster_parent_element |
} # OE |
| 1707 |
= $self->{open_elements}->[$_ - 1]->[0]; |
$foster_parent_element ||= $self->{open_elements}->[0]->[0]; |
| 1708 |
} |
|
| 1709 |
last OE; |
$foster_parent_element->insert_before ($child, $next_sibling); |
|
} |
|
|
} # OE |
|
|
$foster_parent_element = $self->{open_elements}->[0]->[0] |
|
|
unless defined $foster_parent_element; |
|
|
$foster_parent_element->insert_before |
|
|
($child, $next_sibling); |
|
| 1710 |
$open_tables->[-1]->[1] = 1; # tainted |
$open_tables->[-1]->[1] = 1; # tainted |
| 1711 |
} else { |
} else { |
| 1712 |
!!!cp ('t72'); |
!!!cp ('t72'); |
| 1714 |
} |
} |
| 1715 |
}; # $insert_to_foster |
}; # $insert_to_foster |
| 1716 |
|
|
| 1717 |
|
## NOTE: Insert a character (MUST): When a character is inserted, if |
| 1718 |
|
## the last node that was inserted by the parser is a Text node and |
| 1719 |
|
## the character has to be inserted after that node, then the |
| 1720 |
|
## character is appended to the Text node. However, if any other |
| 1721 |
|
## node is inserted by the parser, then a new Text node is created |
| 1722 |
|
## and the character is appended as that Text node. If I'm not |
| 1723 |
|
## wrong, for a parser with scripting disabled, there are only two |
| 1724 |
|
## cases where this occurs. One is the case where an element node |
| 1725 |
|
## is inserted to the |head| element. This is covered by using the |
| 1726 |
|
## |$self->{head_element_inserted}| flag. Another is the case where |
| 1727 |
|
## an element or comment is inserted into the |table| subtree while |
| 1728 |
|
## foster parenting happens. This is covered by using the [2] flag |
| 1729 |
|
## of the |$open_tables| structure. All other cases are handled |
| 1730 |
|
## simply by calling |manakai_append_text| method. |
| 1731 |
|
|
| 1732 |
|
## TODO: |<body><script>document.write("a<br>"); |
| 1733 |
|
## document.body.removeChild (document.body.lastChild); |
| 1734 |
|
## document.write ("b")</script>| |
| 1735 |
|
|
| 1736 |
B: while (1) { |
B: while (1) { |
| 1737 |
|
|
| 1738 |
|
## The "in table text" insertion mode. |
| 1739 |
|
if ($self->{insertion_mode} & TABLE_IMS and |
| 1740 |
|
not $self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and |
| 1741 |
|
not $self->{insertion_mode} & IN_CDATA_RCDATA_IM) { |
| 1742 |
|
C: { |
| 1743 |
|
my $s; |
| 1744 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 1745 |
|
!!!cp ('t194'); |
| 1746 |
|
$self->{pending_chars} ||= []; |
| 1747 |
|
push @{$self->{pending_chars}}, $token; |
| 1748 |
|
!!!next-token; |
| 1749 |
|
next B; |
| 1750 |
|
} else { |
| 1751 |
|
if ($self->{pending_chars}) { |
| 1752 |
|
$s = join '', map { $_->{data} } @{$self->{pending_chars}}; |
| 1753 |
|
delete $self->{pending_chars}; |
| 1754 |
|
if ($s =~ /[^\x09\x0A\x0C\x0D\x20]/) { |
| 1755 |
|
!!!cp ('t195'); |
| 1756 |
|
# |
| 1757 |
|
} else { |
| 1758 |
|
!!!cp ('t195.1'); |
| 1759 |
|
#$self->{open_elements}->[-1]->[0]->manakai_append_text ($s); |
| 1760 |
|
$self->{open_elements}->[-1]->[0]->append_child |
| 1761 |
|
($self->{document}->create_text_node ($s)); |
| 1762 |
|
last C; |
| 1763 |
|
} |
| 1764 |
|
} else { |
| 1765 |
|
!!!cp ('t195.2'); |
| 1766 |
|
last C; |
| 1767 |
|
} |
| 1768 |
|
} |
| 1769 |
|
|
| 1770 |
|
## Foster parenting. |
| 1771 |
|
!!!parse-error (type => 'in table:#text', token => $token); |
| 1772 |
|
|
| 1773 |
|
## NOTE: As if in body, but insert into the foster parent element. |
| 1774 |
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
| 1775 |
|
|
| 1776 |
|
if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
| 1777 |
|
# MUST |
| 1778 |
|
my $foster_parent_element; |
| 1779 |
|
my $next_sibling; |
| 1780 |
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 1781 |
|
if ($self->{open_elements}->[$_]->[1] == TABLE_EL) { |
| 1782 |
|
!!!cp ('t197'); |
| 1783 |
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 1784 |
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 1785 |
|
undef $next_sibling |
| 1786 |
|
unless $next_sibling->parent_node eq $foster_parent_element; |
| 1787 |
|
last OE; |
| 1788 |
|
} |
| 1789 |
|
} # OE |
| 1790 |
|
$foster_parent_element ||= $self->{open_elements}->[0]->[0]; |
| 1791 |
|
|
| 1792 |
|
!!!cp ('t199'); |
| 1793 |
|
$foster_parent_element->insert_before |
| 1794 |
|
($self->{document}->create_text_node ($s), $next_sibling); |
| 1795 |
|
|
| 1796 |
|
$open_tables->[-1]->[1] = 1; # tainted |
| 1797 |
|
$open_tables->[-1]->[2] = 1; # ~node inserted |
| 1798 |
|
} else { |
| 1799 |
|
## NOTE: Fragment case or in a foster parent'ed element |
| 1800 |
|
## (e.g. |<table><span>a|). In fragment case, whether the |
| 1801 |
|
## character is appended to existing node or a new node is |
| 1802 |
|
## created is irrelevant, since the foster parent'ed nodes |
| 1803 |
|
## are discarded and fragment parsing does not invoke any |
| 1804 |
|
## script. |
| 1805 |
|
!!!cp ('t200'); |
| 1806 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($s); |
| 1807 |
|
} |
| 1808 |
|
} # C |
| 1809 |
|
} # TABLE_IMS |
| 1810 |
|
|
| 1811 |
if ($token->{type} == DOCTYPE_TOKEN) { |
if ($token->{type} == DOCTYPE_TOKEN) { |
| 1812 |
!!!cp ('t73'); |
!!!cp ('t73'); |
| 1813 |
!!!parse-error (type => 'in html:#DOCTYPE', token => $token); |
!!!parse-error (type => 'in html:#DOCTYPE', token => $token); |
| 1854 |
} else { |
} else { |
| 1855 |
!!!cp ('t87'); |
!!!cp ('t87'); |
| 1856 |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 1857 |
|
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 1858 |
} |
} |
| 1859 |
!!!next-token; |
!!!next-token; |
| 1860 |
next B; |
next B; |
| 1861 |
|
} elsif ($self->{insertion_mode} & IN_CDATA_RCDATA_IM) { |
| 1862 |
|
if ($token->{type} == CHARACTER_TOKEN) { |
| 1863 |
|
$token->{data} =~ s/^\x0A// if $self->{ignore_newline}; |
| 1864 |
|
delete $self->{ignore_newline}; |
| 1865 |
|
|
| 1866 |
|
if (length $token->{data}) { |
| 1867 |
|
!!!cp ('t43'); |
| 1868 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text |
| 1869 |
|
($token->{data}); |
| 1870 |
|
} else { |
| 1871 |
|
!!!cp ('t43.1'); |
| 1872 |
|
} |
| 1873 |
|
!!!next-token; |
| 1874 |
|
next B; |
| 1875 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 1876 |
|
delete $self->{ignore_newline}; |
| 1877 |
|
|
| 1878 |
|
if ($token->{tag_name} eq 'script') { |
| 1879 |
|
!!!cp ('t50'); |
| 1880 |
|
|
| 1881 |
|
## Para 1-2 |
| 1882 |
|
my $script = pop @{$self->{open_elements}}; |
| 1883 |
|
|
| 1884 |
|
## Para 3 |
| 1885 |
|
$self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1886 |
|
|
| 1887 |
|
## Para 4 |
| 1888 |
|
## TODO: $old_insertion_point = $current_insertion_point; |
| 1889 |
|
## TODO: $current_insertion_point = just before $self->{nc}; |
| 1890 |
|
|
| 1891 |
|
## Para 5 |
| 1892 |
|
## TODO: Run the $script->[0]. |
| 1893 |
|
|
| 1894 |
|
## Para 6 |
| 1895 |
|
## TODO: $current_insertion_point = $old_insertion_point; |
| 1896 |
|
|
| 1897 |
|
## Para 7 |
| 1898 |
|
## TODO: if ($pending_external_script) { |
| 1899 |
|
## TODO: ... |
| 1900 |
|
## TODO: } |
| 1901 |
|
|
| 1902 |
|
!!!next-token; |
| 1903 |
|
next B; |
| 1904 |
|
} else { |
| 1905 |
|
!!!cp ('t42'); |
| 1906 |
|
|
| 1907 |
|
pop @{$self->{open_elements}}; |
| 1908 |
|
|
| 1909 |
|
$self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1910 |
|
!!!next-token; |
| 1911 |
|
next B; |
| 1912 |
|
} |
| 1913 |
|
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 1914 |
|
delete $self->{ignore_newline}; |
| 1915 |
|
|
| 1916 |
|
!!!cp ('t44'); |
| 1917 |
|
!!!parse-error (type => 'not closed', |
| 1918 |
|
text => $self->{open_elements}->[-1]->[0] |
| 1919 |
|
->manakai_local_name, |
| 1920 |
|
token => $token); |
| 1921 |
|
|
| 1922 |
|
#if ($self->{open_elements}->[-1]->[1] == SCRIPT_EL) { |
| 1923 |
|
# ## TODO: Mark as "already executed" |
| 1924 |
|
#} |
| 1925 |
|
|
| 1926 |
|
pop @{$self->{open_elements}}; |
| 1927 |
|
|
| 1928 |
|
$self->{insertion_mode} &= ~ IN_CDATA_RCDATA_IM; |
| 1929 |
|
## Reprocess. |
| 1930 |
|
next B; |
| 1931 |
|
} else { |
| 1932 |
|
die "$0: $token->{type}: In CDATA/RCDATA: Unknown token type"; |
| 1933 |
|
} |
| 1934 |
} elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM) { |
} elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM) { |
| 1935 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 1936 |
!!!cp ('t87.1'); |
!!!cp ('t87.1'); |
| 1937 |
|
|
| 1938 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 1939 |
|
|
| 1940 |
|
if ($token->{data} =~ /[^\x09\x0A\x0C\x0D\x20]/) { |
| 1941 |
|
delete $self->{frameset_ok}; |
| 1942 |
|
} |
| 1943 |
|
|
| 1944 |
!!!next-token; |
!!!next-token; |
| 1945 |
next B; |
next B; |
| 1946 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 1948 |
$self->{open_elements}->[-1]->[1] & FOREIGN_FLOW_CONTENT_EL) or |
$self->{open_elements}->[-1]->[1] & FOREIGN_FLOW_CONTENT_EL) or |
| 1949 |
not ($self->{open_elements}->[-1]->[1] & FOREIGN_EL) or |
not ($self->{open_elements}->[-1]->[1] & FOREIGN_EL) or |
| 1950 |
($token->{tag_name} eq 'svg' and |
($token->{tag_name} eq 'svg' and |
| 1951 |
$self->{open_elements}->[-1]->[1] & MML_AXML_EL)) { |
$self->{open_elements}->[-1]->[1] == MML_AXML_EL)) { |
| 1952 |
## NOTE: "using the rules for secondary insertion mode"then"continue" |
## NOTE: "using the rules for secondary insertion mode"then"continue" |
| 1953 |
!!!cp ('t87.2'); |
!!!cp ('t87.2'); |
| 1954 |
# |
# |
| 1955 |
} elsif ({ |
} elsif ({ |
| 1956 |
b => 1, big => 1, blockquote => 1, body => 1, br => 1, |
b => 1, big => 1, blockquote => 1, body => 1, br => 1, |
| 1957 |
center => 1, code => 1, dd => 1, div => 1, dl => 1, dt => 1, |
center => 1, code => 1, dd => 1, div => 1, dl => 1, dt => 1, |
| 1958 |
em => 1, embed => 1, font => 1, h1 => 1, h2 => 1, h3 => 1, |
em => 1, embed => 1, h1 => 1, h2 => 1, h3 => 1, |
| 1959 |
h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, i => 1, |
h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, i => 1, |
| 1960 |
img => 1, li => 1, listing => 1, menu => 1, meta => 1, |
img => 1, li => 1, listing => 1, menu => 1, meta => 1, |
| 1961 |
nobr => 1, ol => 1, p => 1, pre => 1, ruby => 1, s => 1, |
nobr => 1, ol => 1, p => 1, pre => 1, ruby => 1, s => 1, |
| 1962 |
small => 1, span => 1, strong => 1, strike => 1, sub => 1, |
small => 1, span => 1, strong => 1, strike => 1, sub => 1, |
| 1963 |
sup => 1, table => 1, tt => 1, u => 1, ul => 1, var => 1, |
sup => 1, table => 1, tt => 1, u => 1, ul => 1, var => 1, |
| 1964 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}} or |
| 1965 |
|
($token->{tag_name} eq 'font' and |
| 1966 |
|
($token->{attributes}->{color} or |
| 1967 |
|
$token->{attributes}->{face} or |
| 1968 |
|
$token->{attributes}->{size}))) { |
| 1969 |
!!!cp ('t87.2'); |
!!!cp ('t87.2'); |
| 1970 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 1971 |
text => $self->{open_elements}->[-1]->[0] |
text => $self->{open_elements}->[-1]->[0] |
| 2041 |
} |
} |
| 2042 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 2043 |
## NOTE: "using the rules for secondary insertion mode" then "continue" |
## NOTE: "using the rules for secondary insertion mode" then "continue" |
| 2044 |
!!!cp ('t87.5'); |
if ($token->{tag_name} eq 'script') { |
| 2045 |
# |
!!!cp ('t87.41'); |
| 2046 |
|
# |
| 2047 |
|
## XXXscript: Execute script here. |
| 2048 |
|
} else { |
| 2049 |
|
!!!cp ('t87.5'); |
| 2050 |
|
# |
| 2051 |
|
} |
| 2052 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 2053 |
!!!cp ('t87.6'); |
!!!cp ('t87.6'); |
| 2054 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 2059 |
pop @{$self->{open_elements}} |
pop @{$self->{open_elements}} |
| 2060 |
while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; |
while $self->{open_elements}->[-1]->[1] & FOREIGN_EL; |
| 2061 |
|
|
| 2062 |
|
## NOTE: |<span><svg>| ... two parse errors, |<svg>| ... a parse error. |
| 2063 |
|
|
| 2064 |
$self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
$self->{insertion_mode} &= ~ IN_FOREIGN_CONTENT_IM; |
| 2065 |
## Reprocess. |
## Reprocess. |
| 2066 |
next B; |
next B; |
| 2073 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 2074 |
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 2075 |
unless ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
unless ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2076 |
!!!cp ('t88.2'); |
if ($self->{head_element_inserted}) { |
| 2077 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
!!!cp ('t88.3'); |
| 2078 |
# |
$self->{open_elements}->[-1]->[0]->append_child |
| 2079 |
|
($self->{document}->create_text_node ($1)); |
| 2080 |
|
delete $self->{head_element_inserted}; |
| 2081 |
|
## NOTE: |</head> <link> | |
| 2082 |
|
# |
| 2083 |
|
} else { |
| 2084 |
|
!!!cp ('t88.2'); |
| 2085 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 2086 |
|
## NOTE: |</head>  | |
| 2087 |
|
# |
| 2088 |
|
} |
| 2089 |
} else { |
} else { |
| 2090 |
!!!cp ('t88.1'); |
!!!cp ('t88.1'); |
| 2091 |
## Ignore the token. |
## Ignore the token. |
| 2135 |
## As if <body> |
## As if <body> |
| 2136 |
!!!insert-element ('body',, $token); |
!!!insert-element ('body',, $token); |
| 2137 |
$self->{insertion_mode} = IN_BODY_IM; |
$self->{insertion_mode} = IN_BODY_IM; |
| 2138 |
## reprocess |
## The "frameset-ok" flag is left unchanged in this case. |
| 2139 |
|
## Reporcess the token. |
| 2140 |
next B; |
next B; |
| 2141 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 2142 |
if ($token->{tag_name} eq 'head') { |
if ($token->{tag_name} eq 'head') { |
| 2182 |
!!!cp ('t97'); |
!!!cp ('t97'); |
| 2183 |
} |
} |
| 2184 |
|
|
| 2185 |
if ($token->{tag_name} eq 'base') { |
if ($token->{tag_name} eq 'base') { |
| 2186 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2187 |
!!!cp ('t98'); |
!!!cp ('t98'); |
| 2188 |
## As if </noscript> |
## As if </noscript> |
| 2189 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2190 |
!!!parse-error (type => 'in noscript', text => 'base', |
!!!parse-error (type => 'in noscript', text => 'base', |
| 2191 |
token => $token); |
token => $token); |
| 2192 |
|
|
| 2193 |
$self->{insertion_mode} = IN_HEAD_IM; |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2194 |
## Reprocess in the "in head" insertion mode... |
## Reprocess in the "in head" insertion mode... |
| 2195 |
} else { |
} else { |
| 2196 |
!!!cp ('t99'); |
!!!cp ('t99'); |
| 2197 |
} |
} |
| 2198 |
|
|
| 2199 |
## NOTE: There is a "as if in head" code clone. |
## NOTE: There is a "as if in head" code clone. |
| 2200 |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2201 |
!!!cp ('t100'); |
!!!cp ('t100'); |
| 2202 |
!!!parse-error (type => 'after head', |
!!!parse-error (type => 'after head', |
| 2203 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 2204 |
push @{$self->{open_elements}}, |
push @{$self->{open_elements}}, |
| 2205 |
[$self->{head_element}, $el_category->{head}]; |
[$self->{head_element}, $el_category->{head}]; |
| 2206 |
} else { |
$self->{head_element_inserted} = 1; |
| 2207 |
!!!cp ('t101'); |
} else { |
| 2208 |
} |
!!!cp ('t101'); |
| 2209 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
} |
| 2210 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2211 |
pop @{$self->{open_elements}} # <head> |
pop @{$self->{open_elements}}; |
| 2212 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
pop @{$self->{open_elements}} # <head> |
| 2213 |
!!!nack ('t101.1'); |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2214 |
!!!next-token; |
!!!nack ('t101.1'); |
| 2215 |
next B; |
!!!next-token; |
| 2216 |
} elsif ($token->{tag_name} eq 'link') { |
next B; |
| 2217 |
## NOTE: There is a "as if in head" code clone. |
} elsif ($token->{tag_name} eq 'link') { |
| 2218 |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
## NOTE: There is a "as if in head" code clone. |
| 2219 |
!!!cp ('t102'); |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2220 |
!!!parse-error (type => 'after head', |
!!!cp ('t102'); |
| 2221 |
text => $token->{tag_name}, token => $token); |
!!!parse-error (type => 'after head', |
| 2222 |
push @{$self->{open_elements}}, |
text => $token->{tag_name}, token => $token); |
| 2223 |
[$self->{head_element}, $el_category->{head}]; |
push @{$self->{open_elements}}, |
| 2224 |
} else { |
[$self->{head_element}, $el_category->{head}]; |
| 2225 |
!!!cp ('t103'); |
$self->{head_element_inserted} = 1; |
| 2226 |
} |
} else { |
| 2227 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
!!!cp ('t103'); |
| 2228 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
} |
| 2229 |
pop @{$self->{open_elements}} # <head> |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2230 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
pop @{$self->{open_elements}}; |
| 2231 |
!!!ack ('t103.1'); |
pop @{$self->{open_elements}} # <head> |
| 2232 |
!!!next-token; |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2233 |
next B; |
!!!ack ('t103.1'); |
| 2234 |
} elsif ($token->{tag_name} eq 'meta') { |
!!!next-token; |
| 2235 |
## NOTE: There is a "as if in head" code clone. |
next B; |
| 2236 |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
} elsif ($token->{tag_name} eq 'command') { |
| 2237 |
!!!cp ('t104'); |
if ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2238 |
!!!parse-error (type => 'after head', |
## NOTE: If the insertion mode at the time of the emission |
| 2239 |
text => $token->{tag_name}, token => $token); |
## of the token was "before head", $self->{insertion_mode} |
| 2240 |
push @{$self->{open_elements}}, |
## is already changed to |IN_HEAD_IM|. |
| 2241 |
[$self->{head_element}, $el_category->{head}]; |
|
| 2242 |
} else { |
## NOTE: There is a "as if in head" code clone. |
| 2243 |
!!!cp ('t105'); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2244 |
} |
pop @{$self->{open_elements}}; |
| 2245 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
pop @{$self->{open_elements}} # <head> |
| 2246 |
my $meta_el = pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2247 |
|
!!!ack ('t103.2'); |
| 2248 |
|
!!!next-token; |
| 2249 |
|
next B; |
| 2250 |
|
} else { |
| 2251 |
|
## NOTE: "in head noscript" or "after head" insertion mode |
| 2252 |
|
## - in these cases, these tags are treated as same as |
| 2253 |
|
## normal in-body tags. |
| 2254 |
|
!!!cp ('t103.3'); |
| 2255 |
|
# |
| 2256 |
|
} |
| 2257 |
|
} elsif ($token->{tag_name} eq 'meta') { |
| 2258 |
|
## NOTE: There is a "as if in head" code clone. |
| 2259 |
|
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2260 |
|
!!!cp ('t104'); |
| 2261 |
|
!!!parse-error (type => 'after head', |
| 2262 |
|
text => $token->{tag_name}, token => $token); |
| 2263 |
|
push @{$self->{open_elements}}, |
| 2264 |
|
[$self->{head_element}, $el_category->{head}]; |
| 2265 |
|
$self->{head_element_inserted} = 1; |
| 2266 |
|
} else { |
| 2267 |
|
!!!cp ('t105'); |
| 2268 |
|
} |
| 2269 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2270 |
|
my $meta_el = pop @{$self->{open_elements}}; |
| 2271 |
|
|
| 2272 |
unless ($self->{confident}) { |
unless ($self->{confident}) { |
| 2273 |
if ($token->{attributes}->{charset}) { |
if ($token->{attributes}->{charset}) { |
| 2325 |
!!!ack ('t110.1'); |
!!!ack ('t110.1'); |
| 2326 |
!!!next-token; |
!!!next-token; |
| 2327 |
next B; |
next B; |
| 2328 |
} elsif ($token->{tag_name} eq 'title') { |
} elsif ($token->{tag_name} eq 'title') { |
| 2329 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2330 |
!!!cp ('t111'); |
!!!cp ('t111'); |
| 2331 |
## As if </noscript> |
## As if </noscript> |
| 2332 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2333 |
!!!parse-error (type => 'in noscript', text => 'title', |
!!!parse-error (type => 'in noscript', text => 'title', |
| 2334 |
token => $token); |
token => $token); |
| 2335 |
|
|
| 2336 |
$self->{insertion_mode} = IN_HEAD_IM; |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2337 |
## Reprocess in the "in head" insertion mode... |
## Reprocess in the "in head" insertion mode... |
| 2338 |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2339 |
!!!cp ('t112'); |
!!!cp ('t112'); |
| 2340 |
!!!parse-error (type => 'after head', |
!!!parse-error (type => 'after head', |
| 2341 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 2342 |
push @{$self->{open_elements}}, |
push @{$self->{open_elements}}, |
| 2343 |
[$self->{head_element}, $el_category->{head}]; |
[$self->{head_element}, $el_category->{head}]; |
| 2344 |
} else { |
$self->{head_element_inserted} = 1; |
| 2345 |
!!!cp ('t113'); |
} else { |
| 2346 |
} |
!!!cp ('t113'); |
| 2347 |
|
} |
| 2348 |
|
|
| 2349 |
## NOTE: There is a "as if in head" code clone. |
## NOTE: There is a "as if in head" code clone. |
| 2350 |
my $parent = defined $self->{head_element} ? $self->{head_element} |
$parse_rcdata->(RCDATA_CONTENT_MODEL); |
| 2351 |
: $self->{open_elements}->[-1]->[0]; |
|
| 2352 |
$parse_rcdata->(RCDATA_CONTENT_MODEL); |
## NOTE: At this point the stack of open elements contain |
| 2353 |
pop @{$self->{open_elements}} # <head> |
## the |head| element (index == -2) and the |script| element |
| 2354 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
## (index == -1). In the "after head" insertion mode the |
| 2355 |
next B; |
## |head| element is inserted only for the purpose of |
| 2356 |
} elsif ($token->{tag_name} eq 'style' or |
## providing the context for the |script| element, and |
| 2357 |
$token->{tag_name} eq 'noframes') { |
## therefore we can now and have to remove the element from |
| 2358 |
## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
## the stack. |
| 2359 |
## insertion mode IN_HEAD_IM) |
splice @{$self->{open_elements}}, -2, 1, () # <head> |
| 2360 |
## NOTE: There is a "as if in head" code clone. |
if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2361 |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
next B; |
| 2362 |
!!!cp ('t114'); |
} elsif ($token->{tag_name} eq 'style' or |
| 2363 |
!!!parse-error (type => 'after head', |
$token->{tag_name} eq 'noframes') { |
| 2364 |
text => $token->{tag_name}, token => $token); |
## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
| 2365 |
push @{$self->{open_elements}}, |
## insertion mode IN_HEAD_IM) |
| 2366 |
[$self->{head_element}, $el_category->{head}]; |
## NOTE: There is a "as if in head" code clone. |
| 2367 |
} else { |
if ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2368 |
!!!cp ('t115'); |
!!!cp ('t114'); |
| 2369 |
} |
!!!parse-error (type => 'after head', |
| 2370 |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
text => $token->{tag_name}, token => $token); |
| 2371 |
pop @{$self->{open_elements}} # <head> |
push @{$self->{open_elements}}, |
| 2372 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
[$self->{head_element}, $el_category->{head}]; |
| 2373 |
next B; |
$self->{head_element_inserted} = 1; |
| 2374 |
} elsif ($token->{tag_name} eq 'noscript') { |
} else { |
| 2375 |
|
!!!cp ('t115'); |
| 2376 |
|
} |
| 2377 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 2378 |
|
## ISSUE: A spec bug [Bug 6038] |
| 2379 |
|
splice @{$self->{open_elements}}, -2, 1, () # <head> |
| 2380 |
|
if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2381 |
|
next B; |
| 2382 |
|
} elsif ($token->{tag_name} eq 'noscript') { |
| 2383 |
if ($self->{insertion_mode} == IN_HEAD_IM) { |
if ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2384 |
!!!cp ('t116'); |
!!!cp ('t116'); |
| 2385 |
## NOTE: and scripting is disalbed |
## NOTE: and scripting is disalbed |
| 2400 |
!!!cp ('t118'); |
!!!cp ('t118'); |
| 2401 |
# |
# |
| 2402 |
} |
} |
| 2403 |
} elsif ($token->{tag_name} eq 'script') { |
} elsif ($token->{tag_name} eq 'script') { |
| 2404 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2405 |
!!!cp ('t119'); |
!!!cp ('t119'); |
| 2406 |
## As if </noscript> |
## As if </noscript> |
| 2407 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2408 |
!!!parse-error (type => 'in noscript', text => 'script', |
!!!parse-error (type => 'in noscript', text => 'script', |
| 2409 |
token => $token); |
token => $token); |
| 2410 |
|
|
| 2411 |
$self->{insertion_mode} = IN_HEAD_IM; |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2412 |
## Reprocess in the "in head" insertion mode... |
## Reprocess in the "in head" insertion mode... |
| 2413 |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2414 |
!!!cp ('t120'); |
!!!cp ('t120'); |
| 2415 |
!!!parse-error (type => 'after head', |
!!!parse-error (type => 'after head', |
| 2416 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 2417 |
push @{$self->{open_elements}}, |
push @{$self->{open_elements}}, |
| 2418 |
[$self->{head_element}, $el_category->{head}]; |
[$self->{head_element}, $el_category->{head}]; |
| 2419 |
} else { |
$self->{head_element_inserted} = 1; |
| 2420 |
!!!cp ('t121'); |
} else { |
| 2421 |
} |
!!!cp ('t121'); |
| 2422 |
|
} |
| 2423 |
|
|
| 2424 |
## NOTE: There is a "as if in head" code clone. |
## NOTE: There is a "as if in head" code clone. |
| 2425 |
$script_start_tag->(); |
$script_start_tag->(); |
| 2426 |
pop @{$self->{open_elements}} # <head> |
## ISSUE: A spec bug [Bug 6038] |
| 2427 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
splice @{$self->{open_elements}}, -2, 1 # <head> |
| 2428 |
next B; |
if ($self->{insertion_mode} & IM_MASK) == AFTER_HEAD_IM; |
| 2429 |
} elsif ($token->{tag_name} eq 'body' or |
next B; |
| 2430 |
$token->{tag_name} eq 'frameset') { |
} elsif ($token->{tag_name} eq 'body' or |
| 2431 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
$token->{tag_name} eq 'frameset') { |
| 2432 |
!!!cp ('t122'); |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2433 |
## As if </noscript> |
!!!cp ('t122'); |
| 2434 |
pop @{$self->{open_elements}}; |
## As if </noscript> |
| 2435 |
!!!parse-error (type => 'in noscript', |
pop @{$self->{open_elements}}; |
| 2436 |
text => $token->{tag_name}, token => $token); |
!!!parse-error (type => 'in noscript', |
| 2437 |
|
text => $token->{tag_name}, token => $token); |
| 2438 |
## Reprocess in the "in head" insertion mode... |
|
| 2439 |
## As if </head> |
## Reprocess in the "in head" insertion mode... |
| 2440 |
pop @{$self->{open_elements}}; |
## As if </head> |
| 2441 |
|
pop @{$self->{open_elements}}; |
| 2442 |
## Reprocess in the "after head" insertion mode... |
|
| 2443 |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
## Reprocess in the "after head" insertion mode... |
| 2444 |
!!!cp ('t124'); |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2445 |
pop @{$self->{open_elements}}; |
!!!cp ('t124'); |
| 2446 |
|
pop @{$self->{open_elements}}; |
| 2447 |
## Reprocess in the "after head" insertion mode... |
|
| 2448 |
} else { |
## Reprocess in the "after head" insertion mode... |
| 2449 |
!!!cp ('t125'); |
} else { |
| 2450 |
} |
!!!cp ('t125'); |
| 2451 |
|
} |
| 2452 |
|
|
| 2453 |
## "after head" insertion mode |
## "after head" insertion mode |
| 2454 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 2455 |
if ($token->{tag_name} eq 'body') { |
if ($token->{tag_name} eq 'body') { |
| 2456 |
!!!cp ('t126'); |
!!!cp ('t126'); |
| 2457 |
$self->{insertion_mode} = IN_BODY_IM; |
delete $self->{frameset_ok}; |
| 2458 |
} elsif ($token->{tag_name} eq 'frameset') { |
$self->{insertion_mode} = IN_BODY_IM; |
| 2459 |
!!!cp ('t127'); |
} elsif ($token->{tag_name} eq 'frameset') { |
| 2460 |
$self->{insertion_mode} = IN_FRAMESET_IM; |
!!!cp ('t127'); |
| 2461 |
} else { |
$self->{insertion_mode} = IN_FRAMESET_IM; |
| 2462 |
die "$0: tag name: $self->{tag_name}"; |
} else { |
| 2463 |
} |
die "$0: tag name: $self->{tag_name}"; |
| 2464 |
!!!nack ('t127.1'); |
} |
| 2465 |
!!!next-token; |
!!!nack ('t127.1'); |
| 2466 |
next B; |
!!!next-token; |
| 2467 |
} else { |
next B; |
| 2468 |
!!!cp ('t128'); |
} else { |
| 2469 |
# |
!!!cp ('t128'); |
| 2470 |
} |
# |
| 2471 |
|
} |
| 2472 |
|
|
| 2473 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2474 |
!!!cp ('t129'); |
!!!cp ('t129'); |
| 2492 |
!!!cp ('t131'); |
!!!cp ('t131'); |
| 2493 |
} |
} |
| 2494 |
|
|
| 2495 |
## "after head" insertion mode |
## "after head" insertion mode |
| 2496 |
## As if <body> |
## As if <body> |
| 2497 |
!!!insert-element ('body',, $token); |
!!!insert-element ('body',, $token); |
| 2498 |
$self->{insertion_mode} = IN_BODY_IM; |
$self->{insertion_mode} = IN_BODY_IM; |
| 2499 |
## reprocess |
## The "frameset-ok" flag is not changed in this case. |
| 2500 |
!!!ack-later; |
## Reprocess the token. |
| 2501 |
next B; |
!!!ack-later; |
| 2502 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
next B; |
| 2503 |
if ($token->{tag_name} eq 'head') { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 2504 |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
## "Before head", "in head", and "after head" insertion modes |
| 2505 |
!!!cp ('t132'); |
## ignore most of end tags. Exceptions are "body", "html", |
| 2506 |
## As if <head> |
## and "br" end tags. "Before head" and "in head" insertion |
| 2507 |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
## modes also recognize "head" end tag. "In head noscript" |
| 2508 |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
## insertion modes ignore end tags except for "noscript" and |
| 2509 |
push @{$self->{open_elements}}, |
## "br". |
|
[$self->{head_element}, $el_category->{head}]; |
|
| 2510 |
|
|
| 2511 |
## Reprocess in the "in head" insertion mode... |
if ($token->{tag_name} eq 'head') { |
| 2512 |
pop @{$self->{open_elements}}; |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2513 |
$self->{insertion_mode} = AFTER_HEAD_IM; |
!!!cp ('t132'); |
| 2514 |
!!!next-token; |
## As if <head> |
| 2515 |
next B; |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2516 |
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2517 |
!!!cp ('t133'); |
push @{$self->{open_elements}}, |
| 2518 |
## As if </noscript> |
[$self->{head_element}, $el_category->{head}]; |
| 2519 |
pop @{$self->{open_elements}}; |
|
| 2520 |
!!!parse-error (type => 'in noscript:/', |
## Reprocess in the "in head" insertion mode... |
| 2521 |
text => 'head', token => $token); |
pop @{$self->{open_elements}}; |
| 2522 |
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2523 |
## Reprocess in the "in head" insertion mode... |
!!!next-token; |
| 2524 |
pop @{$self->{open_elements}}; |
next B; |
| 2525 |
$self->{insertion_mode} = AFTER_HEAD_IM; |
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2526 |
!!!next-token; |
!!!cp ('t133'); |
| 2527 |
next B; |
# |
| 2528 |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2529 |
!!!cp ('t134'); |
!!!cp ('t134'); |
| 2530 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2531 |
$self->{insertion_mode} = AFTER_HEAD_IM; |
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2532 |
!!!next-token; |
!!!next-token; |
| 2533 |
next B; |
next B; |
| 2534 |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2535 |
!!!cp ('t134.1'); |
!!!cp ('t134.1'); |
| 2536 |
!!!parse-error (type => 'unmatched end tag', text => 'head', |
# |
| 2537 |
token => $token); |
} else { |
| 2538 |
## Ignore the token |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2539 |
!!!next-token; |
} |
| 2540 |
next B; |
} elsif ($token->{tag_name} eq 'noscript') { |
| 2541 |
} else { |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2542 |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
!!!cp ('t136'); |
| 2543 |
} |
pop @{$self->{open_elements}}; |
| 2544 |
} elsif ($token->{tag_name} eq 'noscript') { |
$self->{insertion_mode} = IN_HEAD_IM; |
| 2545 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
!!!next-token; |
| 2546 |
!!!cp ('t136'); |
next B; |
| 2547 |
pop @{$self->{open_elements}}; |
} else { |
| 2548 |
$self->{insertion_mode} = IN_HEAD_IM; |
!!!cp ('t138'); |
| 2549 |
!!!next-token; |
# |
| 2550 |
next B; |
} |
| 2551 |
} elsif ($self->{insertion_mode} == BEFORE_HEAD_IM or |
} elsif ({ |
| 2552 |
$self->{insertion_mode} == AFTER_HEAD_IM) { |
body => ($self->{insertion_mode} != IN_HEAD_NOSCRIPT_IM), |
| 2553 |
!!!cp ('t137'); |
html => ($self->{insertion_mode} != IN_HEAD_NOSCRIPT_IM), |
| 2554 |
!!!parse-error (type => 'unmatched end tag', |
br => 1, |
| 2555 |
text => 'noscript', token => $token); |
}->{$token->{tag_name}}) { |
| 2556 |
## Ignore the token ## ISSUE: An issue in the spec. |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2557 |
!!!next-token; |
!!!cp ('t142.2'); |
| 2558 |
next B; |
## (before head) as if <head>, (in head) as if </head> |
| 2559 |
} else { |
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
| 2560 |
!!!cp ('t138'); |
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2561 |
# |
$self->{insertion_mode} = AFTER_HEAD_IM; |
|
} |
|
|
} elsif ({ |
|
|
body => 1, html => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
if ($self->{insertion_mode} == BEFORE_HEAD_IM or |
|
|
$self->{insertion_mode} == IN_HEAD_IM or |
|
|
$self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
|
|
!!!cp ('t140'); |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
next B; |
|
|
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
|
|
!!!cp ('t140.1'); |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
next B; |
|
|
} else { |
|
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
|
|
} |
|
|
} elsif ($token->{tag_name} eq 'p') { |
|
|
!!!cp ('t142'); |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
next B; |
|
|
} elsif ($token->{tag_name} eq 'br') { |
|
|
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
|
|
!!!cp ('t142.2'); |
|
|
## (before head) as if <head>, (in head) as if </head> |
|
|
!!!create-element ($self->{head_element}, $HTML_NS, 'head',, $token); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
|
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
|
| 2562 |
|
|
| 2563 |
## Reprocess in the "after head" insertion mode... |
## Reprocess in the "after head" insertion mode... |
| 2564 |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
| 2565 |
!!!cp ('t143.2'); |
!!!cp ('t143.2'); |
| 2566 |
## As if </head> |
## As if </head> |
| 2567 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2568 |
$self->{insertion_mode} = AFTER_HEAD_IM; |
$self->{insertion_mode} = AFTER_HEAD_IM; |
| 2569 |
|
|
| 2570 |
## Reprocess in the "after head" insertion mode... |
## Reprocess in the "after head" insertion mode... |
| 2571 |
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
} elsif ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
| 2572 |
!!!cp ('t143.3'); |
!!!cp ('t143.3'); |
| 2573 |
## ISSUE: Two parse errors for <head><noscript></br> |
## NOTE: Two parse errors for <head><noscript></br> |
| 2574 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 2575 |
text => 'br', token => $token); |
text => $token->{tag_name}, token => $token); |
| 2576 |
## As if </noscript> |
## As if </noscript> |
| 2577 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2578 |
$self->{insertion_mode} = IN_HEAD_IM; |
$self->{insertion_mode} = IN_HEAD_IM; |
|
|
|
|
## Reprocess in the "in head" insertion mode... |
|
|
## As if </head> |
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = AFTER_HEAD_IM; |
|
|
|
|
|
## Reprocess in the "after head" insertion mode... |
|
|
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
|
|
!!!cp ('t143.4'); |
|
|
# |
|
|
} else { |
|
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
|
|
} |
|
|
|
|
|
## ISSUE: does not agree with IE7 - it doesn't ignore </br>. |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => 'br', token => $token); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
next B; |
|
|
} else { |
|
|
!!!cp ('t145'); |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
next B; |
|
|
} |
|
| 2579 |
|
|
| 2580 |
if ($self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM) { |
## Reprocess in the "in head" insertion mode... |
| 2581 |
!!!cp ('t146'); |
## As if </head> |
| 2582 |
## As if </noscript> |
pop @{$self->{open_elements}}; |
| 2583 |
pop @{$self->{open_elements}}; |
$self->{insertion_mode} = AFTER_HEAD_IM; |
|
!!!parse-error (type => 'in noscript:/', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
|
|
|
## Reprocess in the "in head" insertion mode... |
|
|
## As if </head> |
|
|
pop @{$self->{open_elements}}; |
|
| 2584 |
|
|
| 2585 |
## Reprocess in the "after head" insertion mode... |
## Reprocess in the "after head" insertion mode... |
| 2586 |
} elsif ($self->{insertion_mode} == IN_HEAD_IM) { |
} elsif ($self->{insertion_mode} == AFTER_HEAD_IM) { |
| 2587 |
!!!cp ('t147'); |
!!!cp ('t143.4'); |
| 2588 |
## As if </head> |
# |
| 2589 |
pop @{$self->{open_elements}}; |
} else { |
| 2590 |
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 2591 |
|
} |
| 2592 |
|
|
| 2593 |
## Reprocess in the "after head" insertion mode... |
## "after head" insertion mode |
| 2594 |
} elsif ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
## As if <body> |
| 2595 |
## ISSUE: This case cannot be reached? |
!!!insert-element ('body',, $token); |
| 2596 |
!!!cp ('t148'); |
$self->{insertion_mode} = IN_BODY_IM; |
| 2597 |
!!!parse-error (type => 'unmatched end tag', |
## The "frameset-ok" flag is left unchanged in this case. |
| 2598 |
text => $token->{tag_name}, token => $token); |
## Reprocess the token. |
| 2599 |
## Ignore the token ## ISSUE: An issue in the spec. |
next B; |
| 2600 |
!!!next-token; |
} |
|
next B; |
|
|
} else { |
|
|
!!!cp ('t149'); |
|
|
} |
|
| 2601 |
|
|
| 2602 |
## "after head" insertion mode |
## End tags are ignored by default. |
| 2603 |
## As if <body> |
!!!cp ('t145'); |
| 2604 |
!!!insert-element ('body',, $token); |
!!!parse-error (type => 'unmatched end tag', |
| 2605 |
$self->{insertion_mode} = IN_BODY_IM; |
text => $token->{tag_name}, token => $token); |
| 2606 |
## reprocess |
## Ignore the token. |
| 2607 |
next B; |
!!!next-token; |
| 2608 |
|
next B; |
| 2609 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 2610 |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
if ($self->{insertion_mode} == BEFORE_HEAD_IM) { |
| 2611 |
!!!cp ('t149.1'); |
!!!cp ('t149.1'); |
| 2658 |
## NOTE: As if <body> |
## NOTE: As if <body> |
| 2659 |
!!!insert-element ('body',, $token); |
!!!insert-element ('body',, $token); |
| 2660 |
$self->{insertion_mode} = IN_BODY_IM; |
$self->{insertion_mode} = IN_BODY_IM; |
| 2661 |
## NOTE: Reprocess. |
## The "frameset-ok" flag is left unchanged in this case. |
| 2662 |
|
## Reprocess the token. |
| 2663 |
next B; |
next B; |
| 2664 |
} else { |
} else { |
| 2665 |
die "$0: $token->{type}: Unknown token type"; |
die "$0: $token->{type}: Unknown token type"; |
| 2666 |
} |
} |
|
|
|
|
## ISSUE: An issue in the spec. |
|
| 2667 |
} elsif ($self->{insertion_mode} & BODY_IMS) { |
} elsif ($self->{insertion_mode} & BODY_IMS) { |
| 2668 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 2669 |
!!!cp ('t150'); |
!!!cp ('t150'); |
| 2670 |
## NOTE: There is a code clone of "character in body". |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2671 |
$reconstruct_active_formatting_elements->($insert_to_current); |
|
| 2672 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
| 2673 |
|
|
| 2674 |
!!!next-token; |
if ($token->{data} =~ /[^\x09\x0A\x0C\x0D\x20]/) { |
| 2675 |
next B; |
delete $self->{frameset_ok}; |
| 2676 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} |
| 2677 |
|
|
| 2678 |
|
!!!next-token; |
| 2679 |
|
next B; |
| 2680 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 2681 |
if ({ |
if ({ |
| 2682 |
caption => 1, col => 1, colgroup => 1, tbody => 1, |
caption => 1, col => 1, colgroup => 1, tbody => 1, |
| 2683 |
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
| 2684 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 2685 |
if ($self->{insertion_mode} == IN_CELL_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2686 |
## have an element in table scope |
## have an element in table scope |
| 2687 |
for (reverse 0..$#{$self->{open_elements}}) { |
for (reverse 0..$#{$self->{open_elements}}) { |
| 2688 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2689 |
if ($node->[1] & TABLE_CELL_EL) { |
if ($node->[1] == TABLE_CELL_EL) { |
| 2690 |
!!!cp ('t151'); |
!!!cp ('t151'); |
| 2691 |
|
|
| 2692 |
## Close the cell |
## Close the cell |
| 2710 |
!!!nack ('t153.1'); |
!!!nack ('t153.1'); |
| 2711 |
!!!next-token; |
!!!next-token; |
| 2712 |
next B; |
next B; |
| 2713 |
} elsif ($self->{insertion_mode} == IN_CAPTION_IM) { |
} elsif (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2714 |
!!!parse-error (type => 'not closed', text => 'caption', |
!!!parse-error (type => 'not closed', text => 'caption', |
| 2715 |
token => $token); |
token => $token); |
| 2716 |
|
|
| 2720 |
INSCOPE: { |
INSCOPE: { |
| 2721 |
for (reverse 0..$#{$self->{open_elements}}) { |
for (reverse 0..$#{$self->{open_elements}}) { |
| 2722 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2723 |
if ($node->[1] & CAPTION_EL) { |
if ($node->[1] == CAPTION_EL) { |
| 2724 |
!!!cp ('t155'); |
!!!cp ('t155'); |
| 2725 |
$i = $_; |
$i = $_; |
| 2726 |
last INSCOPE; |
last INSCOPE; |
| 2746 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2747 |
} |
} |
| 2748 |
|
|
| 2749 |
unless ($self->{open_elements}->[-1]->[1] & CAPTION_EL) { |
unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2750 |
!!!cp ('t159'); |
!!!cp ('t159'); |
| 2751 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 2752 |
text => $self->{open_elements}->[-1]->[0] |
text => $self->{open_elements}->[-1]->[0] |
| 2775 |
} |
} |
| 2776 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 2777 |
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
| 2778 |
if ($self->{insertion_mode} == IN_CELL_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2779 |
## have an element in table scope |
## have an element in table scope |
| 2780 |
my $i; |
my $i; |
| 2781 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2825 |
|
|
| 2826 |
!!!next-token; |
!!!next-token; |
| 2827 |
next B; |
next B; |
| 2828 |
} elsif ($self->{insertion_mode} == IN_CAPTION_IM) { |
} elsif (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2829 |
!!!cp ('t169'); |
!!!cp ('t169'); |
| 2830 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 2831 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 2837 |
# |
# |
| 2838 |
} |
} |
| 2839 |
} elsif ($token->{tag_name} eq 'caption') { |
} elsif ($token->{tag_name} eq 'caption') { |
| 2840 |
if ($self->{insertion_mode} == IN_CAPTION_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2841 |
## have a table element in table scope |
## have a table element in table scope |
| 2842 |
my $i; |
my $i; |
| 2843 |
INSCOPE: { |
INSCOPE: { |
| 2844 |
for (reverse 0..$#{$self->{open_elements}}) { |
for (reverse 0..$#{$self->{open_elements}}) { |
| 2845 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2846 |
if ($node->[1] & CAPTION_EL) { |
if ($node->[1] == CAPTION_EL) { |
| 2847 |
!!!cp ('t171'); |
!!!cp ('t171'); |
| 2848 |
$i = $_; |
$i = $_; |
| 2849 |
last INSCOPE; |
last INSCOPE; |
| 2868 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2869 |
} |
} |
| 2870 |
|
|
| 2871 |
unless ($self->{open_elements}->[-1]->[1] & CAPTION_EL) { |
unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2872 |
!!!cp ('t175'); |
!!!cp ('t175'); |
| 2873 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 2874 |
text => $self->{open_elements}->[-1]->[0] |
text => $self->{open_elements}->[-1]->[0] |
| 2886 |
|
|
| 2887 |
!!!next-token; |
!!!next-token; |
| 2888 |
next B; |
next B; |
| 2889 |
} elsif ($self->{insertion_mode} == IN_CELL_IM) { |
} elsif (($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2890 |
!!!cp ('t177'); |
!!!cp ('t177'); |
| 2891 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 2892 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 2901 |
table => 1, tbody => 1, tfoot => 1, |
table => 1, tbody => 1, tfoot => 1, |
| 2902 |
thead => 1, tr => 1, |
thead => 1, tr => 1, |
| 2903 |
}->{$token->{tag_name}} and |
}->{$token->{tag_name}} and |
| 2904 |
$self->{insertion_mode} == IN_CELL_IM) { |
($self->{insertion_mode} & IM_MASK) == IN_CELL_IM) { |
| 2905 |
## have an element in table scope |
## have an element in table scope |
| 2906 |
my $i; |
my $i; |
| 2907 |
my $tn; |
my $tn; |
| 2918 |
line => $token->{line}, |
line => $token->{line}, |
| 2919 |
column => $token->{column}}; |
column => $token->{column}}; |
| 2920 |
next B; |
next B; |
| 2921 |
} elsif ($node->[1] & TABLE_CELL_EL) { |
} elsif ($node->[1] == TABLE_CELL_EL) { |
| 2922 |
!!!cp ('t180'); |
!!!cp ('t180'); |
| 2923 |
$tn = $node->[0]->manakai_local_name; |
$tn = $node->[0]->manakai_local_name; |
| 2924 |
## NOTE: There is exactly one |td| or |th| element |
## NOTE: There is exactly one |td| or |th| element |
| 2938 |
next B; |
next B; |
| 2939 |
} # INSCOPE |
} # INSCOPE |
| 2940 |
} elsif ($token->{tag_name} eq 'table' and |
} elsif ($token->{tag_name} eq 'table' and |
| 2941 |
$self->{insertion_mode} == IN_CAPTION_IM) { |
($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 2942 |
!!!parse-error (type => 'not closed', text => 'caption', |
!!!parse-error (type => 'not closed', text => 'caption', |
| 2943 |
token => $token); |
token => $token); |
| 2944 |
|
|
| 2947 |
my $i; |
my $i; |
| 2948 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2949 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 2950 |
if ($node->[1] & CAPTION_EL) { |
if ($node->[1] == CAPTION_EL) { |
| 2951 |
!!!cp ('t184'); |
!!!cp ('t184'); |
| 2952 |
$i = $_; |
$i = $_; |
| 2953 |
last INSCOPE; |
last INSCOPE; |
| 2958 |
} # INSCOPE |
} # INSCOPE |
| 2959 |
unless (defined $i) { |
unless (defined $i) { |
| 2960 |
!!!cp ('t186'); |
!!!cp ('t186'); |
| 2961 |
|
## TODO: Wrong error type? |
| 2962 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 2963 |
text => 'caption', token => $token); |
text => 'caption', token => $token); |
| 2964 |
## Ignore the token |
## Ignore the token |
| 2972 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 2973 |
} |
} |
| 2974 |
|
|
| 2975 |
unless ($self->{open_elements}->[-1]->[1] & CAPTION_EL) { |
unless ($self->{open_elements}->[-1]->[1] == CAPTION_EL) { |
| 2976 |
!!!cp ('t188'); |
!!!cp ('t188'); |
| 2977 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 2978 |
text => $self->{open_elements}->[-1]->[0] |
text => $self->{open_elements}->[-1]->[0] |
| 3004 |
!!!cp ('t191'); |
!!!cp ('t191'); |
| 3005 |
# |
# |
| 3006 |
} |
} |
| 3007 |
} elsif ({ |
} elsif ({ |
| 3008 |
tbody => 1, tfoot => 1, |
tbody => 1, tfoot => 1, |
| 3009 |
thead => 1, tr => 1, |
thead => 1, tr => 1, |
| 3010 |
}->{$token->{tag_name}} and |
}->{$token->{tag_name}} and |
| 3011 |
$self->{insertion_mode} == IN_CAPTION_IM) { |
($self->{insertion_mode} & IM_MASK) == IN_CAPTION_IM) { |
| 3012 |
!!!cp ('t192'); |
!!!cp ('t192'); |
| 3013 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 3014 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 3015 |
## Ignore the token |
## Ignore the token |
| 3016 |
!!!next-token; |
!!!next-token; |
| 3017 |
next B; |
next B; |
| 3018 |
} else { |
} else { |
| 3019 |
!!!cp ('t193'); |
!!!cp ('t193'); |
| 3020 |
# |
# |
| 3021 |
} |
} |
| 3022 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 3023 |
for my $entry (@{$self->{open_elements}}) { |
for my $entry (@{$self->{open_elements}}) { |
| 3024 |
unless ($entry->[1] & ALL_END_TAG_OPTIONAL_EL) { |
unless ($entry->[1] & ALL_END_TAG_OPTIONAL_EL) { |
| 3037 |
$insert = $insert_to_current; |
$insert = $insert_to_current; |
| 3038 |
# |
# |
| 3039 |
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
| 3040 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == START_TAG_TOKEN) { |
|
if (not $open_tables->[-1]->[1] and # tainted |
|
|
$token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!cp ('t194'); |
|
|
!!!next-token; |
|
|
next B; |
|
|
} else { |
|
|
!!!cp ('t195'); |
|
|
} |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:#text', token => $token); |
|
|
|
|
|
## As if in body, but insert into foster parent element |
|
|
## ISSUE: Spec says that "whenever a node would be inserted |
|
|
## into the current node" while characters might not be |
|
|
## result in a new Text node. |
|
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] & TABLE_ROWS_EL) { |
|
|
# MUST |
|
|
my $foster_parent_element; |
|
|
my $next_sibling; |
|
|
my $prev_sibling; |
|
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
if ($self->{open_elements}->[$_]->[1] & TABLE_EL) { |
|
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
|
|
if (defined $parent and $parent->node_type == 1) { |
|
|
!!!cp ('t196'); |
|
|
$foster_parent_element = $parent; |
|
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
|
|
} else { |
|
|
!!!cp ('t197'); |
|
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
|
|
$prev_sibling = $foster_parent_element->last_child; |
|
|
} |
|
|
last OE; |
|
|
} |
|
|
} # OE |
|
|
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
|
|
$prev_sibling = $foster_parent_element->last_child |
|
|
unless defined $foster_parent_element; |
|
|
if (defined $prev_sibling and |
|
|
$prev_sibling->node_type == 3) { |
|
|
!!!cp ('t198'); |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
|
|
} else { |
|
|
!!!cp ('t199'); |
|
|
$foster_parent_element->insert_before |
|
|
($self->{document}->create_text_node ($token->{data}), |
|
|
$next_sibling); |
|
|
} |
|
|
$open_tables->[-1]->[1] = 1; # tainted |
|
|
} else { |
|
|
!!!cp ('t200'); |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
} |
|
|
|
|
|
!!!next-token; |
|
|
next B; |
|
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
|
| 3041 |
if ({ |
if ({ |
| 3042 |
tr => ($self->{insertion_mode} != IN_ROW_IM), |
tr => (($self->{insertion_mode} & IM_MASK) != IN_ROW_IM), |
| 3043 |
th => 1, td => 1, |
th => 1, td => 1, |
| 3044 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 3045 |
if ($self->{insertion_mode} == IN_TABLE_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_IM) { |
| 3046 |
## Clear back to table context |
## Clear back to table context |
| 3047 |
while (not ($self->{open_elements}->[-1]->[1] |
while (not ($self->{open_elements}->[-1]->[1] |
| 3048 |
& TABLE_SCOPING_EL)) { |
& TABLE_SCOPING_EL)) { |
| 3055 |
## reprocess in the "in table body" insertion mode... |
## reprocess in the "in table body" insertion mode... |
| 3056 |
} |
} |
| 3057 |
|
|
| 3058 |
if ($self->{insertion_mode} == IN_TABLE_BODY_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { |
| 3059 |
unless ($token->{tag_name} eq 'tr') { |
unless ($token->{tag_name} eq 'tr') { |
| 3060 |
!!!cp ('t202'); |
!!!cp ('t202'); |
| 3061 |
!!!parse-error (type => 'missing start tag:tr', token => $token); |
!!!parse-error (type => 'missing start tag:tr', token => $token); |
| 3069 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3070 |
} |
} |
| 3071 |
|
|
| 3072 |
$self->{insertion_mode} = IN_ROW_IM; |
$self->{insertion_mode} = IN_ROW_IM; |
| 3073 |
if ($token->{tag_name} eq 'tr') { |
if ($token->{tag_name} eq 'tr') { |
| 3074 |
!!!cp ('t204'); |
!!!cp ('t204'); |
| 3075 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3076 |
!!!nack ('t204'); |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3077 |
!!!next-token; |
!!!nack ('t204'); |
| 3078 |
next B; |
!!!next-token; |
| 3079 |
} else { |
next B; |
| 3080 |
!!!cp ('t205'); |
} else { |
| 3081 |
!!!insert-element ('tr',, $token); |
!!!cp ('t205'); |
| 3082 |
## reprocess in the "in row" insertion mode |
!!!insert-element ('tr',, $token); |
| 3083 |
} |
## reprocess in the "in row" insertion mode |
| 3084 |
} else { |
} |
| 3085 |
!!!cp ('t206'); |
} else { |
| 3086 |
} |
!!!cp ('t206'); |
| 3087 |
|
} |
| 3088 |
|
|
| 3089 |
## Clear back to table row context |
## Clear back to table row context |
| 3090 |
while (not ($self->{open_elements}->[-1]->[1] |
while (not ($self->{open_elements}->[-1]->[1] |
| 3093 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3094 |
} |
} |
| 3095 |
|
|
| 3096 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3097 |
$self->{insertion_mode} = IN_CELL_IM; |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3098 |
|
$self->{insertion_mode} = IN_CELL_IM; |
| 3099 |
|
|
| 3100 |
push @$active_formatting_elements, ['#marker', '']; |
push @$active_formatting_elements, ['#marker', '']; |
| 3101 |
|
|
| 3102 |
!!!nack ('t207.1'); |
!!!nack ('t207.1'); |
| 3103 |
|
!!!next-token; |
| 3104 |
|
next B; |
| 3105 |
|
} elsif ({ |
| 3106 |
|
caption => 1, col => 1, colgroup => 1, |
| 3107 |
|
tbody => 1, tfoot => 1, thead => 1, |
| 3108 |
|
tr => 1, # $self->{insertion_mode} == IN_ROW_IM |
| 3109 |
|
}->{$token->{tag_name}}) { |
| 3110 |
|
if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3111 |
|
## As if </tr> |
| 3112 |
|
## have an element in table scope |
| 3113 |
|
my $i; |
| 3114 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3115 |
|
my $node = $self->{open_elements}->[$_]; |
| 3116 |
|
if ($node->[1] == TABLE_ROW_EL) { |
| 3117 |
|
!!!cp ('t208'); |
| 3118 |
|
$i = $_; |
| 3119 |
|
last INSCOPE; |
| 3120 |
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
| 3121 |
|
!!!cp ('t209'); |
| 3122 |
|
last INSCOPE; |
| 3123 |
|
} |
| 3124 |
|
} # INSCOPE |
| 3125 |
|
unless (defined $i) { |
| 3126 |
|
!!!cp ('t210'); |
| 3127 |
|
## TODO: This type is wrong. |
| 3128 |
|
!!!parse-error (type => 'unmacthed end tag', |
| 3129 |
|
text => $token->{tag_name}, token => $token); |
| 3130 |
|
## Ignore the token |
| 3131 |
|
!!!nack ('t210.1'); |
| 3132 |
!!!next-token; |
!!!next-token; |
| 3133 |
next B; |
next B; |
| 3134 |
} elsif ({ |
} |
|
caption => 1, col => 1, colgroup => 1, |
|
|
tbody => 1, tfoot => 1, thead => 1, |
|
|
tr => 1, # $self->{insertion_mode} == IN_ROW_IM |
|
|
}->{$token->{tag_name}}) { |
|
|
if ($self->{insertion_mode} == IN_ROW_IM) { |
|
|
## As if </tr> |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] & TABLE_ROW_EL) { |
|
|
!!!cp ('t208'); |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ($node->[1] & TABLE_SCOPING_EL) { |
|
|
!!!cp ('t209'); |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!cp ('t210'); |
|
|
## TODO: This type is wrong. |
|
|
!!!parse-error (type => 'unmacthed end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
|
!!!nack ('t210.1'); |
|
|
!!!next-token; |
|
|
next B; |
|
|
} |
|
| 3135 |
|
|
| 3136 |
## Clear back to table row context |
## Clear back to table row context |
| 3137 |
while (not ($self->{open_elements}->[-1]->[1] |
while (not ($self->{open_elements}->[-1]->[1] |
| 3154 |
} |
} |
| 3155 |
} |
} |
| 3156 |
|
|
| 3157 |
if ($self->{insertion_mode} == IN_TABLE_BODY_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { |
| 3158 |
## have an element in table scope |
## have an element in table scope |
| 3159 |
my $i; |
my $i; |
| 3160 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3161 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3162 |
if ($node->[1] & TABLE_ROW_GROUP_EL) { |
if ($node->[1] == TABLE_ROW_GROUP_EL) { |
| 3163 |
!!!cp ('t214'); |
!!!cp ('t214'); |
| 3164 |
$i = $_; |
$i = $_; |
| 3165 |
last INSCOPE; |
last INSCOPE; |
| 3201 |
!!!cp ('t218'); |
!!!cp ('t218'); |
| 3202 |
} |
} |
| 3203 |
|
|
| 3204 |
if ($token->{tag_name} eq 'col') { |
if ($token->{tag_name} eq 'col') { |
| 3205 |
## Clear back to table context |
## Clear back to table context |
| 3206 |
while (not ($self->{open_elements}->[-1]->[1] |
while (not ($self->{open_elements}->[-1]->[1] |
| 3207 |
& TABLE_SCOPING_EL)) { |
& TABLE_SCOPING_EL)) { |
| 3208 |
!!!cp ('t219'); |
!!!cp ('t219'); |
| 3209 |
## ISSUE: Can this state be reached? |
## ISSUE: Can this state be reached? |
| 3210 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3211 |
} |
} |
| 3212 |
|
|
| 3213 |
!!!insert-element ('colgroup',, $token); |
!!!insert-element ('colgroup',, $token); |
| 3214 |
$self->{insertion_mode} = IN_COLUMN_GROUP_IM; |
$self->{insertion_mode} = IN_COLUMN_GROUP_IM; |
| 3215 |
## reprocess |
## reprocess |
| 3216 |
!!!ack-later; |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3217 |
next B; |
!!!ack-later; |
| 3218 |
} elsif ({ |
next B; |
| 3219 |
caption => 1, |
} elsif ({ |
| 3220 |
colgroup => 1, |
caption => 1, |
| 3221 |
tbody => 1, tfoot => 1, thead => 1, |
colgroup => 1, |
| 3222 |
}->{$token->{tag_name}}) { |
tbody => 1, tfoot => 1, thead => 1, |
| 3223 |
## Clear back to table context |
}->{$token->{tag_name}}) { |
| 3224 |
|
## Clear back to table context |
| 3225 |
while (not ($self->{open_elements}->[-1]->[1] |
while (not ($self->{open_elements}->[-1]->[1] |
| 3226 |
& TABLE_SCOPING_EL)) { |
& TABLE_SCOPING_EL)) { |
| 3227 |
!!!cp ('t220'); |
!!!cp ('t220'); |
| 3229 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3230 |
} |
} |
| 3231 |
|
|
| 3232 |
push @$active_formatting_elements, ['#marker', ''] |
push @$active_formatting_elements, ['#marker', ''] |
| 3233 |
if $token->{tag_name} eq 'caption'; |
if $token->{tag_name} eq 'caption'; |
| 3234 |
|
|
| 3235 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3236 |
$self->{insertion_mode} = { |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3237 |
caption => IN_CAPTION_IM, |
$self->{insertion_mode} = { |
| 3238 |
colgroup => IN_COLUMN_GROUP_IM, |
caption => IN_CAPTION_IM, |
| 3239 |
tbody => IN_TABLE_BODY_IM, |
colgroup => IN_COLUMN_GROUP_IM, |
| 3240 |
tfoot => IN_TABLE_BODY_IM, |
tbody => IN_TABLE_BODY_IM, |
| 3241 |
thead => IN_TABLE_BODY_IM, |
tfoot => IN_TABLE_BODY_IM, |
| 3242 |
}->{$token->{tag_name}}; |
thead => IN_TABLE_BODY_IM, |
| 3243 |
!!!next-token; |
}->{$token->{tag_name}}; |
| 3244 |
!!!nack ('t220.1'); |
!!!next-token; |
| 3245 |
next B; |
!!!nack ('t220.1'); |
| 3246 |
} else { |
next B; |
| 3247 |
die "$0: in table: <>: $token->{tag_name}"; |
} else { |
| 3248 |
} |
die "$0: in table: <>: $token->{tag_name}"; |
| 3249 |
|
} |
| 3250 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
| 3251 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 3252 |
text => $self->{open_elements}->[-1]->[0] |
text => $self->{open_elements}->[-1]->[0] |
| 3258 |
my $i; |
my $i; |
| 3259 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3260 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3261 |
if ($node->[1] & TABLE_EL) { |
if ($node->[1] == TABLE_EL) { |
| 3262 |
!!!cp ('t221'); |
!!!cp ('t221'); |
| 3263 |
$i = $_; |
$i = $_; |
| 3264 |
last INSCOPE; |
last INSCOPE; |
| 3285 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3286 |
} |
} |
| 3287 |
|
|
| 3288 |
unless ($self->{open_elements}->[-1]->[1] & TABLE_EL) { |
unless ($self->{open_elements}->[-1]->[1] == TABLE_EL) { |
| 3289 |
!!!cp ('t225'); |
!!!cp ('t225'); |
| 3290 |
## NOTE: |<table><tr><table>| |
## NOTE: |<table><tr><table>| |
| 3291 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 3305 |
!!!ack-later; |
!!!ack-later; |
| 3306 |
next B; |
next B; |
| 3307 |
} elsif ($token->{tag_name} eq 'style') { |
} elsif ($token->{tag_name} eq 'style') { |
| 3308 |
if (not $open_tables->[-1]->[1]) { # tainted |
!!!cp ('t227.8'); |
| 3309 |
!!!cp ('t227.8'); |
## NOTE: This is a "as if in head" code clone. |
| 3310 |
## NOTE: This is a "as if in head" code clone. |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 3311 |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3312 |
next B; |
next B; |
|
} else { |
|
|
!!!cp ('t227.7'); |
|
|
# |
|
|
} |
|
| 3313 |
} elsif ($token->{tag_name} eq 'script') { |
} elsif ($token->{tag_name} eq 'script') { |
| 3314 |
if (not $open_tables->[-1]->[1]) { # tainted |
!!!cp ('t227.6'); |
| 3315 |
!!!cp ('t227.6'); |
## NOTE: This is a "as if in head" code clone. |
| 3316 |
## NOTE: This is a "as if in head" code clone. |
$script_start_tag->(); |
| 3317 |
$script_start_tag->(); |
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3318 |
next B; |
next B; |
|
} else { |
|
|
!!!cp ('t227.5'); |
|
|
# |
|
|
} |
|
| 3319 |
} elsif ($token->{tag_name} eq 'input') { |
} elsif ($token->{tag_name} eq 'input') { |
| 3320 |
if (not $open_tables->[-1]->[1]) { # tainted |
if ($token->{attributes}->{type}) { |
| 3321 |
if ($token->{attributes}->{type}) { ## TODO: case |
my $type = $token->{attributes}->{type}->{value}; |
| 3322 |
my $type = lc $token->{attributes}->{type}->{value}; |
$type =~ tr/A-Z/a-z/; ## ASCII case-insensitive. |
| 3323 |
if ($type eq 'hidden') { |
if ($type eq 'hidden') { |
| 3324 |
!!!cp ('t227.3'); |
!!!cp ('t227.3'); |
| 3325 |
!!!parse-error (type => 'in table', |
!!!parse-error (type => 'in table', |
| 3326 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 3327 |
|
|
| 3328 |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element ($token->{tag_name}, $token->{attributes}, $token); |
| 3329 |
|
$open_tables->[-1]->[2] = 0 if @$open_tables; # ~node inserted |
| 3330 |
|
|
| 3331 |
## TODO: form element pointer |
## TODO: form element pointer |
| 3332 |
|
|
| 3333 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3334 |
|
|
| 3335 |
!!!next-token; |
!!!next-token; |
| 3336 |
!!!ack ('t227.2.1'); |
!!!ack ('t227.2.1'); |
| 3337 |
next B; |
next B; |
|
} else { |
|
|
!!!cp ('t227.2'); |
|
|
# |
|
|
} |
|
| 3338 |
} else { |
} else { |
| 3339 |
!!!cp ('t227.1'); |
!!!cp ('t227.1'); |
| 3340 |
# |
# |
| 3354 |
$insert = $insert_to_foster; |
$insert = $insert_to_foster; |
| 3355 |
# |
# |
| 3356 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3357 |
if ($token->{tag_name} eq 'tr' and |
if ($token->{tag_name} eq 'tr' and |
| 3358 |
$self->{insertion_mode} == IN_ROW_IM) { |
($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3359 |
## have an element in table scope |
## have an element in table scope |
| 3360 |
my $i; |
my $i; |
| 3361 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3362 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3363 |
if ($node->[1] & TABLE_ROW_EL) { |
if ($node->[1] == TABLE_ROW_EL) { |
| 3364 |
!!!cp ('t228'); |
!!!cp ('t228'); |
| 3365 |
$i = $_; |
$i = $_; |
| 3366 |
last INSCOPE; |
last INSCOPE; |
| 3395 |
!!!nack ('t231.1'); |
!!!nack ('t231.1'); |
| 3396 |
next B; |
next B; |
| 3397 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
| 3398 |
if ($self->{insertion_mode} == IN_ROW_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3399 |
## As if </tr> |
## As if </tr> |
| 3400 |
## have an element in table scope |
## have an element in table scope |
| 3401 |
my $i; |
my $i; |
| 3402 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3403 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3404 |
if ($node->[1] & TABLE_ROW_EL) { |
if ($node->[1] == TABLE_ROW_EL) { |
| 3405 |
!!!cp ('t233'); |
!!!cp ('t233'); |
| 3406 |
$i = $_; |
$i = $_; |
| 3407 |
last INSCOPE; |
last INSCOPE; |
| 3434 |
## reprocess in the "in table body" insertion mode... |
## reprocess in the "in table body" insertion mode... |
| 3435 |
} |
} |
| 3436 |
|
|
| 3437 |
if ($self->{insertion_mode} == IN_TABLE_BODY_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_TABLE_BODY_IM) { |
| 3438 |
## have an element in table scope |
## have an element in table scope |
| 3439 |
my $i; |
my $i; |
| 3440 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3441 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3442 |
if ($node->[1] & TABLE_ROW_GROUP_EL) { |
if ($node->[1] == TABLE_ROW_GROUP_EL) { |
| 3443 |
!!!cp ('t237'); |
!!!cp ('t237'); |
| 3444 |
$i = $_; |
$i = $_; |
| 3445 |
last INSCOPE; |
last INSCOPE; |
| 3486 |
my $i; |
my $i; |
| 3487 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3488 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3489 |
if ($node->[1] & TABLE_EL) { |
if ($node->[1] == TABLE_EL) { |
| 3490 |
!!!cp ('t241'); |
!!!cp ('t241'); |
| 3491 |
$i = $_; |
$i = $_; |
| 3492 |
last INSCOPE; |
last INSCOPE; |
| 3516 |
tbody => 1, tfoot => 1, thead => 1, |
tbody => 1, tfoot => 1, thead => 1, |
| 3517 |
}->{$token->{tag_name}} and |
}->{$token->{tag_name}} and |
| 3518 |
$self->{insertion_mode} & ROW_IMS) { |
$self->{insertion_mode} & ROW_IMS) { |
| 3519 |
if ($self->{insertion_mode} == IN_ROW_IM) { |
if (($self->{insertion_mode} & IM_MASK) == IN_ROW_IM) { |
| 3520 |
## have an element in table scope |
## have an element in table scope |
| 3521 |
my $i; |
my $i; |
| 3522 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3545 |
my $i; |
my $i; |
| 3546 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3547 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3548 |
if ($node->[1] & TABLE_ROW_EL) { |
if ($node->[1] == TABLE_ROW_EL) { |
| 3549 |
!!!cp ('t250'); |
!!!cp ('t250'); |
| 3550 |
$i = $_; |
$i = $_; |
| 3551 |
last INSCOPE; |
last INSCOPE; |
| 3635 |
# |
# |
| 3636 |
} |
} |
| 3637 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 3638 |
unless ($self->{open_elements}->[-1]->[1] & HTML_EL and |
unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 3639 |
@{$self->{open_elements}} == 1) { # redundant, maybe |
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 3640 |
!!!parse-error (type => 'in body:#eof', token => $token); |
!!!parse-error (type => 'in body:#eof', token => $token); |
| 3641 |
!!!cp ('t259.1'); |
!!!cp ('t259.1'); |
| 3650 |
} else { |
} else { |
| 3651 |
die "$0: $token->{type}: Unknown token type"; |
die "$0: $token->{type}: Unknown token type"; |
| 3652 |
} |
} |
| 3653 |
} elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) { |
} elsif (($self->{insertion_mode} & IM_MASK) == IN_COLUMN_GROUP_IM) { |
| 3654 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 3655 |
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0C\x20]+)//) { |
| 3656 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3677 |
} |
} |
| 3678 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3679 |
if ($token->{tag_name} eq 'colgroup') { |
if ($token->{tag_name} eq 'colgroup') { |
| 3680 |
if ($self->{open_elements}->[-1]->[1] & HTML_EL) { |
if ($self->{open_elements}->[-1]->[1] == HTML_EL) { |
| 3681 |
!!!cp ('t264'); |
!!!cp ('t264'); |
| 3682 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 3683 |
text => 'colgroup', token => $token); |
text => 'colgroup', token => $token); |
| 3703 |
# |
# |
| 3704 |
} |
} |
| 3705 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 3706 |
if ($self->{open_elements}->[-1]->[1] & HTML_EL and |
if ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 3707 |
@{$self->{open_elements}} == 1) { # redundant, maybe |
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 3708 |
!!!cp ('t270.2'); |
!!!cp ('t270.2'); |
| 3709 |
## Stop parsing. |
## Stop parsing. |
| 3721 |
} |
} |
| 3722 |
|
|
| 3723 |
## As if </colgroup> |
## As if </colgroup> |
| 3724 |
if ($self->{open_elements}->[-1]->[1] & HTML_EL) { |
if ($self->{open_elements}->[-1]->[1] == HTML_EL) { |
| 3725 |
!!!cp ('t269'); |
!!!cp ('t269'); |
| 3726 |
## TODO: Wrong error type? |
## TODO: Wrong error type? |
| 3727 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 3746 |
next B; |
next B; |
| 3747 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 3748 |
if ($token->{tag_name} eq 'option') { |
if ($token->{tag_name} eq 'option') { |
| 3749 |
if ($self->{open_elements}->[-1]->[1] & OPTION_EL) { |
if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { |
| 3750 |
!!!cp ('t272'); |
!!!cp ('t272'); |
| 3751 |
## As if </option> |
## As if </option> |
| 3752 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3759 |
!!!next-token; |
!!!next-token; |
| 3760 |
next B; |
next B; |
| 3761 |
} elsif ($token->{tag_name} eq 'optgroup') { |
} elsif ($token->{tag_name} eq 'optgroup') { |
| 3762 |
if ($self->{open_elements}->[-1]->[1] & OPTION_EL) { |
if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { |
| 3763 |
!!!cp ('t274'); |
!!!cp ('t274'); |
| 3764 |
## As if </option> |
## As if </option> |
| 3765 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3767 |
!!!cp ('t275'); |
!!!cp ('t275'); |
| 3768 |
} |
} |
| 3769 |
|
|
| 3770 |
if ($self->{open_elements}->[-1]->[1] & OPTGROUP_EL) { |
if ($self->{open_elements}->[-1]->[1] == OPTGROUP_EL) { |
| 3771 |
!!!cp ('t276'); |
!!!cp ('t276'); |
| 3772 |
## As if </optgroup> |
## As if </optgroup> |
| 3773 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3780 |
!!!next-token; |
!!!next-token; |
| 3781 |
next B; |
next B; |
| 3782 |
} elsif ({ |
} elsif ({ |
| 3783 |
select => 1, input => 1, textarea => 1, |
select => 1, input => 1, textarea => 1, keygen => 1, |
| 3784 |
}->{$token->{tag_name}} or |
}->{$token->{tag_name}} or |
| 3785 |
($self->{insertion_mode} == IN_SELECT_IN_TABLE_IM and |
(($self->{insertion_mode} & IM_MASK) |
| 3786 |
|
== IN_SELECT_IN_TABLE_IM and |
| 3787 |
{ |
{ |
| 3788 |
caption => 1, table => 1, |
caption => 1, table => 1, |
| 3789 |
tbody => 1, tfoot => 1, thead => 1, |
tbody => 1, tfoot => 1, thead => 1, |
| 3790 |
tr => 1, td => 1, th => 1, |
tr => 1, td => 1, th => 1, |
| 3791 |
}->{$token->{tag_name}})) { |
}->{$token->{tag_name}})) { |
| 3792 |
## TODO: The type below is not good - <select> is replaced by </select> |
|
| 3793 |
!!!parse-error (type => 'not closed', text => 'select', |
## 1. Parse error. |
| 3794 |
token => $token); |
if ($token->{tag_name} eq 'select') { |
| 3795 |
## NOTE: As if the token were </select> (<select> case) or |
!!!parse-error (type => 'select in select', ## XXX: documentation |
| 3796 |
## as if there were </select> (otherwise). |
token => $token); |
| 3797 |
## have an element in table scope |
} else { |
| 3798 |
|
!!!parse-error (type => 'not closed', text => 'select', |
| 3799 |
|
token => $token); |
| 3800 |
|
} |
| 3801 |
|
|
| 3802 |
|
## 2./<select>-1. Unless "have an element in table scope" (select): |
| 3803 |
my $i; |
my $i; |
| 3804 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3805 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3806 |
if ($node->[1] & SELECT_EL) { |
if ($node->[1] == SELECT_EL) { |
| 3807 |
!!!cp ('t278'); |
!!!cp ('t278'); |
| 3808 |
$i = $_; |
$i = $_; |
| 3809 |
last INSCOPE; |
last INSCOPE; |
| 3814 |
} # INSCOPE |
} # INSCOPE |
| 3815 |
unless (defined $i) { |
unless (defined $i) { |
| 3816 |
!!!cp ('t280'); |
!!!cp ('t280'); |
| 3817 |
!!!parse-error (type => 'unmatched end tag', |
if ($token->{tag_name} eq 'select') { |
| 3818 |
text => 'select', token => $token); |
## NOTE: This error would be raised when |
| 3819 |
## Ignore the token |
## |select.innerHTML = '<select>'| is executed; in this |
| 3820 |
|
## case two errors, "select in select" and "unmatched |
| 3821 |
|
## end tags" are reported to the user, the latter might |
| 3822 |
|
## be confusing but this is what the spec requires. |
| 3823 |
|
!!!parse-error (type => 'unmatched end tag', |
| 3824 |
|
text => 'select', |
| 3825 |
|
token => $token); |
| 3826 |
|
} |
| 3827 |
|
## Ignore the token. |
| 3828 |
!!!nack ('t280.1'); |
!!!nack ('t280.1'); |
| 3829 |
!!!next-token; |
!!!next-token; |
| 3830 |
next B; |
next B; |
| 3831 |
} |
} |
| 3832 |
|
|
| 3833 |
|
## 3. Otherwise, as if there were <select>: |
| 3834 |
|
|
| 3835 |
!!!cp ('t281'); |
!!!cp ('t281'); |
| 3836 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 3847 |
## Reprocess the token. |
## Reprocess the token. |
| 3848 |
next B; |
next B; |
| 3849 |
} |
} |
| 3850 |
|
} elsif ($token->{tag_name} eq 'script') { |
| 3851 |
|
!!!cp ('t281.3'); |
| 3852 |
|
## NOTE: This is an "as if in head" code clone |
| 3853 |
|
$script_start_tag->(); |
| 3854 |
|
next B; |
| 3855 |
} else { |
} else { |
| 3856 |
!!!cp ('t282'); |
!!!cp ('t282'); |
| 3857 |
!!!parse-error (type => 'in select', |
!!!parse-error (type => 'in select', |
| 3863 |
} |
} |
| 3864 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3865 |
if ($token->{tag_name} eq 'optgroup') { |
if ($token->{tag_name} eq 'optgroup') { |
| 3866 |
if ($self->{open_elements}->[-1]->[1] & OPTION_EL and |
if ($self->{open_elements}->[-1]->[1] == OPTION_EL and |
| 3867 |
$self->{open_elements}->[-2]->[1] & OPTGROUP_EL) { |
$self->{open_elements}->[-2]->[1] == OPTGROUP_EL) { |
| 3868 |
!!!cp ('t283'); |
!!!cp ('t283'); |
| 3869 |
## As if </option> |
## As if </option> |
| 3870 |
splice @{$self->{open_elements}}, -2; |
splice @{$self->{open_elements}}, -2; |
| 3871 |
} elsif ($self->{open_elements}->[-1]->[1] & OPTGROUP_EL) { |
} elsif ($self->{open_elements}->[-1]->[1] == OPTGROUP_EL) { |
| 3872 |
!!!cp ('t284'); |
!!!cp ('t284'); |
| 3873 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3874 |
} else { |
} else { |
| 3881 |
!!!next-token; |
!!!next-token; |
| 3882 |
next B; |
next B; |
| 3883 |
} elsif ($token->{tag_name} eq 'option') { |
} elsif ($token->{tag_name} eq 'option') { |
| 3884 |
if ($self->{open_elements}->[-1]->[1] & OPTION_EL) { |
if ($self->{open_elements}->[-1]->[1] == OPTION_EL) { |
| 3885 |
!!!cp ('t286'); |
!!!cp ('t286'); |
| 3886 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 3887 |
} else { |
} else { |
| 3898 |
my $i; |
my $i; |
| 3899 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3900 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3901 |
if ($node->[1] & SELECT_EL) { |
if ($node->[1] == SELECT_EL) { |
| 3902 |
!!!cp ('t288'); |
!!!cp ('t288'); |
| 3903 |
$i = $_; |
$i = $_; |
| 3904 |
last INSCOPE; |
last INSCOPE; |
| 3925 |
!!!nack ('t291.1'); |
!!!nack ('t291.1'); |
| 3926 |
!!!next-token; |
!!!next-token; |
| 3927 |
next B; |
next B; |
| 3928 |
} elsif ($self->{insertion_mode} == IN_SELECT_IN_TABLE_IM and |
} elsif (($self->{insertion_mode} & IM_MASK) |
| 3929 |
|
== IN_SELECT_IN_TABLE_IM and |
| 3930 |
{ |
{ |
| 3931 |
caption => 1, table => 1, tbody => 1, |
caption => 1, table => 1, tbody => 1, |
| 3932 |
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
| 3961 |
undef $i; |
undef $i; |
| 3962 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3963 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 3964 |
if ($node->[1] & SELECT_EL) { |
if ($node->[1] == SELECT_EL) { |
| 3965 |
!!!cp ('t295'); |
!!!cp ('t295'); |
| 3966 |
$i = $_; |
$i = $_; |
| 3967 |
last INSCOPE; |
last INSCOPE; |
| 4000 |
next B; |
next B; |
| 4001 |
} |
} |
| 4002 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 4003 |
unless ($self->{open_elements}->[-1]->[1] & HTML_EL and |
unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4004 |
@{$self->{open_elements}} == 1) { # redundant, maybe |
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 4005 |
!!!cp ('t299.1'); |
!!!cp ('t299.1'); |
| 4006 |
!!!parse-error (type => 'in body:#eof', token => $token); |
!!!parse-error (type => 'in body:#eof', token => $token); |
| 4187 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 4188 |
if ($token->{tag_name} eq 'frameset' and |
if ($token->{tag_name} eq 'frameset' and |
| 4189 |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
| 4190 |
if ($self->{open_elements}->[-1]->[1] & HTML_EL and |
if ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4191 |
@{$self->{open_elements}} == 1) { |
@{$self->{open_elements}} == 1) { |
| 4192 |
!!!cp ('t325'); |
!!!cp ('t325'); |
| 4193 |
!!!parse-error (type => 'unmatched end tag', |
!!!parse-error (type => 'unmatched end tag', |
| 4201 |
} |
} |
| 4202 |
|
|
| 4203 |
if (not defined $self->{inner_html_node} and |
if (not defined $self->{inner_html_node} and |
| 4204 |
not ($self->{open_elements}->[-1]->[1] & FRAMESET_EL)) { |
not ($self->{open_elements}->[-1]->[1] == FRAMESET_EL)) { |
| 4205 |
!!!cp ('t327'); |
!!!cp ('t327'); |
| 4206 |
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
| 4207 |
} else { |
} else { |
| 4233 |
next B; |
next B; |
| 4234 |
} |
} |
| 4235 |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
} elsif ($token->{type} == END_OF_FILE_TOKEN) { |
| 4236 |
unless ($self->{open_elements}->[-1]->[1] & HTML_EL and |
unless ($self->{open_elements}->[-1]->[1] == HTML_EL and |
| 4237 |
@{$self->{open_elements}} == 1) { # redundant, maybe |
@{$self->{open_elements}} == 1) { # redundant, maybe |
| 4238 |
!!!cp ('t331.1'); |
!!!cp ('t331.1'); |
| 4239 |
!!!parse-error (type => 'in body:#eof', token => $token); |
!!!parse-error (type => 'in body:#eof', token => $token); |
| 4246 |
} else { |
} else { |
| 4247 |
die "$0: $token->{type}: Unknown token type"; |
die "$0: $token->{type}: Unknown token type"; |
| 4248 |
} |
} |
|
|
|
|
## ISSUE: An issue in spec here |
|
| 4249 |
} else { |
} else { |
| 4250 |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 4251 |
} |
} |
| 4263 |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 4264 |
next B; |
next B; |
| 4265 |
} elsif ({ |
} elsif ({ |
| 4266 |
base => 1, link => 1, |
base => 1, command => 1, link => 1, |
| 4267 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 4268 |
!!!cp ('t334'); |
!!!cp ('t334'); |
| 4269 |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
| 4270 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4271 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
pop @{$self->{open_elements}}; |
| 4272 |
!!!ack ('t334.1'); |
!!!ack ('t334.1'); |
| 4273 |
!!!next-token; |
!!!next-token; |
| 4274 |
next B; |
next B; |
| 4275 |
} elsif ($token->{tag_name} eq 'meta') { |
} elsif ($token->{tag_name} eq 'meta') { |
| 4276 |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
| 4277 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4278 |
my $meta_el = pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
my $meta_el = pop @{$self->{open_elements}}; |
| 4279 |
|
|
| 4280 |
unless ($self->{confident}) { |
unless ($self->{confident}) { |
| 4281 |
if ($token->{attributes}->{charset}) { |
if ($token->{attributes}->{charset}) { |
| 4336 |
!!!parse-error (type => 'in body', text => 'body', token => $token); |
!!!parse-error (type => 'in body', text => 'body', token => $token); |
| 4337 |
|
|
| 4338 |
if (@{$self->{open_elements}} == 1 or |
if (@{$self->{open_elements}} == 1 or |
| 4339 |
not ($self->{open_elements}->[1]->[1] & BODY_EL)) { |
not ($self->{open_elements}->[1]->[1] == BODY_EL)) { |
| 4340 |
!!!cp ('t342'); |
!!!cp ('t342'); |
| 4341 |
## Ignore the token |
## Ignore the token |
| 4342 |
} else { |
} else { |
| 4353 |
!!!nack ('t343.1'); |
!!!nack ('t343.1'); |
| 4354 |
!!!next-token; |
!!!next-token; |
| 4355 |
next B; |
next B; |
| 4356 |
|
} elsif ($token->{tag_name} eq 'frameset') { |
| 4357 |
|
!!!parse-error (type => 'in body', text => $token->{tag_name}, |
| 4358 |
|
token => $token); |
| 4359 |
|
|
| 4360 |
|
if (@{$self->{open_elements}} == 1 or |
| 4361 |
|
not ($self->{open_elements}->[1]->[1] == BODY_EL)) { |
| 4362 |
|
!!!cp ('t343.2'); |
| 4363 |
|
## Ignore the token. |
| 4364 |
|
} elsif (not $self->{frameset_ok}) { |
| 4365 |
|
!!!cp ('t343.3'); |
| 4366 |
|
## Ignore the token. |
| 4367 |
|
} else { |
| 4368 |
|
!!!cp ('t343.4'); |
| 4369 |
|
|
| 4370 |
|
## 1. Remove the second element. |
| 4371 |
|
my $body = $self->{open_elements}->[1]->[0]; |
| 4372 |
|
my $body_parent = $body->parent_node; |
| 4373 |
|
$body_parent->remove_child ($body) if $body_parent; |
| 4374 |
|
|
| 4375 |
|
## 2. Pop nodes. |
| 4376 |
|
splice @{$self->{open_elements}}, 1; |
| 4377 |
|
|
| 4378 |
|
## 3. Insert. |
| 4379 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4380 |
|
|
| 4381 |
|
## 4. Switch. |
| 4382 |
|
$self->{insertion_mode} = IN_FRAMESET_IM; |
| 4383 |
|
} |
| 4384 |
|
|
| 4385 |
|
!!!nack ('t343.5'); |
| 4386 |
|
!!!next-token; |
| 4387 |
|
next B; |
| 4388 |
} elsif ({ |
} elsif ({ |
| 4389 |
address => 1, blockquote => 1, center => 1, dir => 1, |
## NOTE: Start tags for non-phrasing flow content elements |
| 4390 |
div => 1, dl => 1, fieldset => 1, |
|
| 4391 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
## NOTE: The normal one |
| 4392 |
menu => 1, ol => 1, p => 1, ul => 1, |
address => 1, article => 1, aside => 1, blockquote => 1, |
| 4393 |
|
center => 1, datagrid => 1, details => 1, dialog => 1, |
| 4394 |
|
dir => 1, div => 1, dl => 1, fieldset => 1, figure => 1, |
| 4395 |
|
footer => 1, h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, |
| 4396 |
|
h6 => 1, header => 1, hgroup => 1, |
| 4397 |
|
menu => 1, nav => 1, ol => 1, p => 1, |
| 4398 |
|
section => 1, ul => 1, |
| 4399 |
|
## NOTE: As normal, but drops leading newline |
| 4400 |
pre => 1, listing => 1, |
pre => 1, listing => 1, |
| 4401 |
|
## NOTE: As normal, but interacts with the form element pointer |
| 4402 |
form => 1, |
form => 1, |
| 4403 |
|
|
| 4404 |
table => 1, |
table => 1, |
| 4405 |
hr => 1, |
hr => 1, |
| 4406 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 4407 |
|
|
| 4408 |
|
## 1. When there is an opening |form| element: |
| 4409 |
if ($token->{tag_name} eq 'form' and defined $self->{form_element}) { |
if ($token->{tag_name} eq 'form' and defined $self->{form_element}) { |
| 4410 |
!!!cp ('t350'); |
!!!cp ('t350'); |
| 4411 |
!!!parse-error (type => 'in form:form', token => $token); |
!!!parse-error (type => 'in form:form', token => $token); |
| 4415 |
next B; |
next B; |
| 4416 |
} |
} |
| 4417 |
|
|
| 4418 |
## has a p element in scope |
## 2. Close the |p| element, if any. |
| 4419 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
if ($token->{tag_name} ne 'table' or # The Hixie Quirk |
| 4420 |
if ($_->[1] & P_EL) { |
$self->{document}->manakai_compat_mode ne 'quirks') { |
| 4421 |
!!!cp ('t344'); |
## has a p element in scope |
| 4422 |
!!!back-token; # <form> |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4423 |
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
if ($_->[1] == P_EL) { |
| 4424 |
line => $token->{line}, column => $token->{column}}; |
!!!cp ('t344'); |
| 4425 |
next B; |
!!!back-token; # <form> |
| 4426 |
} elsif ($_->[1] & SCOPING_EL) { |
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4427 |
!!!cp ('t345'); |
line => $token->{line}, column => $token->{column}}; |
| 4428 |
last INSCOPE; |
next B; |
| 4429 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 4430 |
|
!!!cp ('t345'); |
| 4431 |
|
last INSCOPE; |
| 4432 |
|
} |
| 4433 |
|
} # INSCOPE |
| 4434 |
|
} |
| 4435 |
|
|
| 4436 |
|
## 3. Close the opening <hn> element, if any. |
| 4437 |
|
if ({h1 => 1, h2 => 1, h3 => 1, |
| 4438 |
|
h4 => 1, h5 => 1, h6 => 1}->{$token->{tag_name}}) { |
| 4439 |
|
if ($self->{open_elements}->[-1]->[1] == HEADING_EL) { |
| 4440 |
|
!!!parse-error (type => 'not closed', |
| 4441 |
|
text => $self->{open_elements}->[-1]->[0]->manakai_local_name, |
| 4442 |
|
token => $token); |
| 4443 |
|
pop @{$self->{open_elements}}; |
| 4444 |
} |
} |
| 4445 |
} # INSCOPE |
} |
| 4446 |
|
|
| 4447 |
|
## 4. Insertion. |
| 4448 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4449 |
if ($token->{tag_name} eq 'pre' or $token->{tag_name} eq 'listing') { |
if ($token->{tag_name} eq 'pre' or $token->{tag_name} eq 'listing') { |
| 4450 |
!!!nack ('t346.1'); |
!!!nack ('t346.1'); |
| 4460 |
} else { |
} else { |
| 4461 |
!!!cp ('t348'); |
!!!cp ('t348'); |
| 4462 |
} |
} |
| 4463 |
|
|
| 4464 |
|
delete $self->{frameset_ok}; |
| 4465 |
} elsif ($token->{tag_name} eq 'form') { |
} elsif ($token->{tag_name} eq 'form') { |
| 4466 |
!!!cp ('t347.1'); |
!!!cp ('t347.1'); |
| 4467 |
$self->{form_element} = $self->{open_elements}->[-1]->[0]; |
$self->{form_element} = $self->{open_elements}->[-1]->[0]; |
| 4471 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
| 4472 |
!!!cp ('t382'); |
!!!cp ('t382'); |
| 4473 |
push @{$open_tables}, [$self->{open_elements}->[-1]->[0]]; |
push @{$open_tables}, [$self->{open_elements}->[-1]->[0]]; |
| 4474 |
|
|
| 4475 |
|
delete $self->{frameset_ok}; |
| 4476 |
|
|
| 4477 |
$self->{insertion_mode} = IN_TABLE_IM; |
$self->{insertion_mode} = IN_TABLE_IM; |
| 4478 |
|
|
| 4481 |
} elsif ($token->{tag_name} eq 'hr') { |
} elsif ($token->{tag_name} eq 'hr') { |
| 4482 |
!!!cp ('t386'); |
!!!cp ('t386'); |
| 4483 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 4484 |
|
|
| 4485 |
!!!nack ('t386.1'); |
!!!ack ('t386.1'); |
| 4486 |
|
|
| 4487 |
|
delete $self->{frameset_ok}; |
| 4488 |
|
|
| 4489 |
!!!next-token; |
!!!next-token; |
| 4490 |
} else { |
} else { |
| 4491 |
!!!nack ('t347.1'); |
!!!nack ('t347.1'); |
| 4492 |
!!!next-token; |
!!!next-token; |
| 4493 |
} |
} |
| 4494 |
next B; |
next B; |
| 4495 |
} elsif ({li => 1, dt => 1, dd => 1}->{$token->{tag_name}}) { |
} elsif ($token->{tag_name} eq 'li') { |
| 4496 |
## has a p element in scope |
## NOTE: As normal, but imply </li> when there's another <li> ... |
| 4497 |
|
|
| 4498 |
|
## NOTE: Special, Scope (<li><foo><li> == <li><foo><li/></foo></li>):: |
| 4499 |
|
## Interpreted as <li><foo/></li><li/> (non-conforming): |
| 4500 |
|
## blockquote (O9.27), center (O), dd (Fx3, O, S3.1.2, IE7), |
| 4501 |
|
## dt (Fx, O, S, IE), dl (O), fieldset (O, S, IE), form (Fx, O, S), |
| 4502 |
|
## hn (O), pre (O), applet (O, S), button (O, S), marquee (Fx, O, S), |
| 4503 |
|
## object (Fx) |
| 4504 |
|
## Generate non-tree (non-conforming): |
| 4505 |
|
## basefont (IE7 (where basefont is non-void)), center (IE), |
| 4506 |
|
## form (IE), hn (IE) |
| 4507 |
|
## address, div, p (<li><foo><li> == <li><foo/></li><li/>):: |
| 4508 |
|
## Interpreted as <li><foo><li/></foo></li> (non-conforming): |
| 4509 |
|
## div (Fx, S) |
| 4510 |
|
|
| 4511 |
|
## 1. Frameset-ng |
| 4512 |
|
delete $self->{frameset_ok}; |
| 4513 |
|
|
| 4514 |
|
my $non_optional; |
| 4515 |
|
my $i = -1; |
| 4516 |
|
|
| 4517 |
|
## 2. |
| 4518 |
|
for my $node (reverse @{$self->{open_elements}}) { |
| 4519 |
|
if ($node->[1] == LI_EL) { |
| 4520 |
|
## 3. (a) As if </li> |
| 4521 |
|
{ |
| 4522 |
|
## If no </li> - not applied |
| 4523 |
|
# |
| 4524 |
|
|
| 4525 |
|
## Otherwise |
| 4526 |
|
|
| 4527 |
|
## 1. generate implied end tags, except for </li> |
| 4528 |
|
# |
| 4529 |
|
|
| 4530 |
|
## 2. If current node != "li", parse error |
| 4531 |
|
if ($non_optional) { |
| 4532 |
|
!!!parse-error (type => 'not closed', |
| 4533 |
|
text => $non_optional->[0]->manakai_local_name, |
| 4534 |
|
token => $token); |
| 4535 |
|
!!!cp ('t355'); |
| 4536 |
|
} else { |
| 4537 |
|
!!!cp ('t356'); |
| 4538 |
|
} |
| 4539 |
|
|
| 4540 |
|
## 3. Pop |
| 4541 |
|
splice @{$self->{open_elements}}, $i; |
| 4542 |
|
} |
| 4543 |
|
|
| 4544 |
|
last; ## 3. (b) goto 5. |
| 4545 |
|
} elsif ( |
| 4546 |
|
## NOTE: not "formatting" and not "phrasing" |
| 4547 |
|
($node->[1] & SPECIAL_EL or |
| 4548 |
|
$node->[1] & SCOPING_EL) and |
| 4549 |
|
## NOTE: "li", "dt", and "dd" are in |SPECIAL_EL|. |
| 4550 |
|
(not $node->[1] & ADDRESS_DIV_P_EL) |
| 4551 |
|
) { |
| 4552 |
|
## 4. |
| 4553 |
|
!!!cp ('t357'); |
| 4554 |
|
last; ## goto 6. |
| 4555 |
|
} elsif ($node->[1] & END_TAG_OPTIONAL_EL) { |
| 4556 |
|
!!!cp ('t358'); |
| 4557 |
|
# |
| 4558 |
|
} else { |
| 4559 |
|
!!!cp ('t359'); |
| 4560 |
|
$non_optional ||= $node; |
| 4561 |
|
# |
| 4562 |
|
} |
| 4563 |
|
## 5. |
| 4564 |
|
## goto 3. |
| 4565 |
|
$i--; |
| 4566 |
|
} |
| 4567 |
|
|
| 4568 |
|
## 6. (a) has a |p| element in scope |
| 4569 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4570 |
if ($_->[1] & P_EL) { |
if ($_->[1] == P_EL) { |
| 4571 |
!!!cp ('t353'); |
!!!cp ('t353'); |
| 4572 |
|
|
| 4573 |
|
## NOTE: |<p><li>|, for example. |
| 4574 |
|
|
| 4575 |
!!!back-token; # <x> |
!!!back-token; # <x> |
| 4576 |
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4577 |
line => $token->{line}, column => $token->{column}}; |
line => $token->{line}, column => $token->{column}}; |
| 4581 |
last INSCOPE; |
last INSCOPE; |
| 4582 |
} |
} |
| 4583 |
} # INSCOPE |
} # INSCOPE |
| 4584 |
|
|
| 4585 |
## Step 1 |
## 6. (b) insert |
| 4586 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4587 |
|
!!!nack ('t359.1'); |
| 4588 |
|
!!!next-token; |
| 4589 |
|
next B; |
| 4590 |
|
} elsif ($token->{tag_name} eq 'dt' or |
| 4591 |
|
$token->{tag_name} eq 'dd') { |
| 4592 |
|
## NOTE: As normal, but imply </dt> or </dd> when ... |
| 4593 |
|
|
| 4594 |
|
## 1. Frameset-ng |
| 4595 |
|
delete $self->{frameset_ok}; |
| 4596 |
|
|
| 4597 |
|
my $non_optional; |
| 4598 |
my $i = -1; |
my $i = -1; |
| 4599 |
my $node = $self->{open_elements}->[$i]; |
|
| 4600 |
my $li_or_dtdd = {li => {li => 1}, |
## 2. |
| 4601 |
dt => {dt => 1, dd => 1}, |
for my $node (reverse @{$self->{open_elements}}) { |
| 4602 |
dd => {dt => 1, dd => 1}}->{$token->{tag_name}}; |
if ($node->[1] == DTDD_EL) { |
| 4603 |
LI: { |
## 3. (a) As if </li> |
| 4604 |
## Step 2 |
{ |
| 4605 |
if ($li_or_dtdd->{$node->[0]->manakai_local_name}) { |
## If no </li> - not applied |
| 4606 |
if ($i != -1) { |
# |
| 4607 |
!!!cp ('t355'); |
|
| 4608 |
!!!parse-error (type => 'not closed', |
## Otherwise |
| 4609 |
text => $self->{open_elements}->[-1]->[0] |
|
| 4610 |
->manakai_local_name, |
## 1. generate implied end tags, except for </dt> or </dd> |
| 4611 |
token => $token); |
# |
| 4612 |
} else { |
|
| 4613 |
!!!cp ('t356'); |
## 2. If current node != "dt"|"dd", parse error |
| 4614 |
|
if ($non_optional) { |
| 4615 |
|
!!!parse-error (type => 'not closed', |
| 4616 |
|
text => $non_optional->[0]->manakai_local_name, |
| 4617 |
|
token => $token); |
| 4618 |
|
!!!cp ('t355.1'); |
| 4619 |
|
} else { |
| 4620 |
|
!!!cp ('t356.1'); |
| 4621 |
|
} |
| 4622 |
|
|
| 4623 |
|
## 3. Pop |
| 4624 |
|
splice @{$self->{open_elements}}, $i; |
| 4625 |
} |
} |
| 4626 |
splice @{$self->{open_elements}}, $i; |
|
| 4627 |
last LI; |
last; ## 3. (b) goto 5. |
| 4628 |
|
} elsif ( |
| 4629 |
|
## NOTE: not "formatting" and not "phrasing" |
| 4630 |
|
($node->[1] & SPECIAL_EL or |
| 4631 |
|
$node->[1] & SCOPING_EL) and |
| 4632 |
|
## NOTE: "li", "dt", and "dd" are in |SPECIAL_EL|. |
| 4633 |
|
|
| 4634 |
|
(not $node->[1] & ADDRESS_DIV_P_EL) |
| 4635 |
|
) { |
| 4636 |
|
## 4. |
| 4637 |
|
!!!cp ('t357.1'); |
| 4638 |
|
last; ## goto 5. |
| 4639 |
|
} elsif ($node->[1] & END_TAG_OPTIONAL_EL) { |
| 4640 |
|
!!!cp ('t358.1'); |
| 4641 |
|
# |
| 4642 |
} else { |
} else { |
| 4643 |
!!!cp ('t357'); |
!!!cp ('t359.1'); |
| 4644 |
} |
$non_optional ||= $node; |
| 4645 |
|
# |
|
## Step 3 |
|
|
if (not ($node->[1] & FORMATTING_EL) and |
|
|
#not $phrasing_category->{$node->[1]} and |
|
|
($node->[1] & SPECIAL_EL or |
|
|
$node->[1] & SCOPING_EL) and |
|
|
not ($node->[1] & ADDRESS_EL) and |
|
|
not ($node->[1] & DIV_EL)) { |
|
|
!!!cp ('t358'); |
|
|
last LI; |
|
| 4646 |
} |
} |
| 4647 |
|
## 5. |
| 4648 |
!!!cp ('t359'); |
## goto 3. |
|
## Step 4 |
|
| 4649 |
$i--; |
$i--; |
| 4650 |
$node = $self->{open_elements}->[$i]; |
} |
| 4651 |
redo LI; |
|
| 4652 |
} # LI |
## 6. (a) has a |p| element in scope |
| 4653 |
|
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4654 |
|
if ($_->[1] == P_EL) { |
| 4655 |
|
!!!cp ('t353.1'); |
| 4656 |
|
!!!back-token; # <x> |
| 4657 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4658 |
|
line => $token->{line}, column => $token->{column}}; |
| 4659 |
|
next B; |
| 4660 |
|
} elsif ($_->[1] & SCOPING_EL) { |
| 4661 |
|
!!!cp ('t354.1'); |
| 4662 |
|
last INSCOPE; |
| 4663 |
|
} |
| 4664 |
|
} # INSCOPE |
| 4665 |
|
|
| 4666 |
|
## 6. (b) insert |
| 4667 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4668 |
!!!nack ('t359.1'); |
!!!nack ('t359.2'); |
| 4669 |
!!!next-token; |
!!!next-token; |
| 4670 |
next B; |
next B; |
| 4671 |
} elsif ($token->{tag_name} eq 'plaintext') { |
} elsif ($token->{tag_name} eq 'plaintext') { |
| 4672 |
|
## NOTE: As normal, but effectively ends parsing |
| 4673 |
|
|
| 4674 |
## has a p element in scope |
## has a p element in scope |
| 4675 |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
INSCOPE: for (reverse @{$self->{open_elements}}) { |
| 4676 |
if ($_->[1] & P_EL) { |
if ($_->[1] == P_EL) { |
| 4677 |
!!!cp ('t367'); |
!!!cp ('t367'); |
| 4678 |
!!!back-token; # <plaintext> |
!!!back-token; # <plaintext> |
| 4679 |
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
$token = {type => END_TAG_TOKEN, tag_name => 'p', |
| 4695 |
} elsif ($token->{tag_name} eq 'a') { |
} elsif ($token->{tag_name} eq 'a') { |
| 4696 |
AFE: for my $i (reverse 0..$#$active_formatting_elements) { |
AFE: for my $i (reverse 0..$#$active_formatting_elements) { |
| 4697 |
my $node = $active_formatting_elements->[$i]; |
my $node = $active_formatting_elements->[$i]; |
| 4698 |
if ($node->[1] & A_EL) { |
if ($node->[1] == A_EL) { |
| 4699 |
!!!cp ('t371'); |
!!!cp ('t371'); |
| 4700 |
!!!parse-error (type => 'in a:a', token => $token); |
!!!parse-error (type => 'in a:a', token => $token); |
| 4701 |
|
|
| 4739 |
## has a |nobr| element in scope |
## has a |nobr| element in scope |
| 4740 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4741 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 4742 |
if ($node->[1] & NOBR_EL) { |
if ($node->[1] == NOBR_EL) { |
| 4743 |
!!!cp ('t376'); |
!!!cp ('t376'); |
| 4744 |
!!!parse-error (type => 'in nobr:nobr', token => $token); |
!!!parse-error (type => 'in nobr:nobr', token => $token); |
| 4745 |
!!!back-token; # <nobr> |
!!!back-token; # <nobr> |
| 4762 |
## has a button element in scope |
## has a button element in scope |
| 4763 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4764 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 4765 |
if ($node->[1] & BUTTON_EL) { |
if ($node->[1] == BUTTON_EL) { |
| 4766 |
!!!cp ('t378'); |
!!!cp ('t378'); |
| 4767 |
!!!parse-error (type => 'in button:button', token => $token); |
!!!parse-error (type => 'in button:button', token => $token); |
| 4768 |
!!!back-token; # <button> |
!!!back-token; # <button> |
| 4783 |
|
|
| 4784 |
push @$active_formatting_elements, ['#marker', '']; |
push @$active_formatting_elements, ['#marker', '']; |
| 4785 |
|
|
| 4786 |
|
delete $self->{frameset_ok}; |
| 4787 |
|
|
| 4788 |
!!!nack ('t379.1'); |
!!!nack ('t379.1'); |
| 4789 |
!!!next-token; |
!!!next-token; |
| 4790 |
next B; |
next B; |
| 4798 |
if ($token->{tag_name} eq 'xmp') { |
if ($token->{tag_name} eq 'xmp') { |
| 4799 |
!!!cp ('t381'); |
!!!cp ('t381'); |
| 4800 |
$reconstruct_active_formatting_elements->($insert_to_current); |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4801 |
|
|
| 4802 |
|
delete $self->{frameset_ok}; |
| 4803 |
|
} elsif ($token->{tag_name} eq 'iframe') { |
| 4804 |
|
!!!cp ('t381.1'); |
| 4805 |
|
delete $self->{frameset_ok}; |
| 4806 |
} else { |
} else { |
| 4807 |
!!!cp ('t399'); |
!!!cp ('t399'); |
| 4808 |
} |
} |
| 4834 |
line => $token->{line}, column => $token->{column}}, |
line => $token->{line}, column => $token->{column}}, |
| 4835 |
{type => START_TAG_TOKEN, tag_name => 'hr', |
{type => START_TAG_TOKEN, tag_name => 'hr', |
| 4836 |
line => $token->{line}, column => $token->{column}}, |
line => $token->{line}, column => $token->{column}}, |
|
{type => START_TAG_TOKEN, tag_name => 'p', |
|
|
line => $token->{line}, column => $token->{column}}, |
|
| 4837 |
{type => START_TAG_TOKEN, tag_name => 'label', |
{type => START_TAG_TOKEN, tag_name => 'label', |
| 4838 |
line => $token->{line}, column => $token->{column}}, |
line => $token->{line}, column => $token->{column}}, |
| 4839 |
); |
); |
| 4856 |
#{type => CHARACTER_TOKEN, data => ''}, # SHOULD |
#{type => CHARACTER_TOKEN, data => ''}, # SHOULD |
| 4857 |
{type => END_TAG_TOKEN, tag_name => 'label', |
{type => END_TAG_TOKEN, tag_name => 'label', |
| 4858 |
line => $token->{line}, column => $token->{column}}, |
line => $token->{line}, column => $token->{column}}, |
|
{type => END_TAG_TOKEN, tag_name => 'p', |
|
|
line => $token->{line}, column => $token->{column}}, |
|
| 4859 |
{type => START_TAG_TOKEN, tag_name => 'hr', |
{type => START_TAG_TOKEN, tag_name => 'hr', |
| 4860 |
line => $token->{line}, column => $token->{column}}, |
line => $token->{line}, column => $token->{column}}, |
| 4861 |
{type => END_TAG_TOKEN, tag_name => 'form', |
{type => END_TAG_TOKEN, tag_name => 'form', |
| 4865 |
next B; |
next B; |
| 4866 |
} |
} |
| 4867 |
} elsif ($token->{tag_name} eq 'textarea') { |
} elsif ($token->{tag_name} eq 'textarea') { |
| 4868 |
my $tag_name = $token->{tag_name}; |
## 1. Insert |
| 4869 |
my $el; |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
|
!!!create-element ($el, $HTML_NS, $token->{tag_name}, $token->{attributes}, $token); |
|
| 4870 |
|
|
| 4871 |
|
## Step 2 # XXX |
| 4872 |
## TODO: $self->{form_element} if defined |
## TODO: $self->{form_element} if defined |
| 4873 |
|
|
| 4874 |
|
## 2. Drop U+000A LINE FEED |
| 4875 |
|
$self->{ignore_newline} = 1; |
| 4876 |
|
|
| 4877 |
|
## 3. RCDATA |
| 4878 |
$self->{content_model} = RCDATA_CONTENT_MODEL; |
$self->{content_model} = RCDATA_CONTENT_MODEL; |
| 4879 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
| 4880 |
|
|
| 4881 |
$insert->($el); |
## 4., 6. Insertion mode |
| 4882 |
|
$self->{insertion_mode} |= IN_CDATA_RCDATA_IM; |
| 4883 |
my $text = ''; |
|
| 4884 |
|
## 5. Frameset-ng. |
| 4885 |
|
delete $self->{frameset_ok}; |
| 4886 |
|
|
| 4887 |
!!!nack ('t392.1'); |
!!!nack ('t392.1'); |
| 4888 |
!!!next-token; |
!!!next-token; |
| 4889 |
if ($token->{type} == CHARACTER_TOKEN) { |
next B; |
| 4890 |
$token->{data} =~ s/^\x0A//; |
} elsif ($token->{tag_name} eq 'optgroup' or |
| 4891 |
unless (length $token->{data}) { |
$token->{tag_name} eq 'option') { |
| 4892 |
!!!cp ('t392'); |
## has an |option| element in scope |
| 4893 |
!!!next-token; |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4894 |
} else { |
my $node = $self->{open_elements}->[$_]; |
| 4895 |
!!!cp ('t393'); |
if ($node->[1] == OPTION_EL) { |
| 4896 |
|
!!!cp ('t397.1'); |
| 4897 |
|
## NOTE: As if </option> |
| 4898 |
|
!!!back-token; # <option> or <optgroup> |
| 4899 |
|
$token = {type => END_TAG_TOKEN, tag_name => 'option', |
| 4900 |
|
line => $token->{line}, column => $token->{column}}; |
| 4901 |
|
next B; |
| 4902 |
|
} elsif ($node->[1] & SCOPING_EL) { |
| 4903 |
|
!!!cp ('t397.2'); |
| 4904 |
|
last INSCOPE; |
| 4905 |
} |
} |
| 4906 |
} else { |
} # INSCOPE |
| 4907 |
!!!cp ('t394'); |
|
| 4908 |
} |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4909 |
while ($token->{type} == CHARACTER_TOKEN) { |
|
| 4910 |
!!!cp ('t395'); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4911 |
$text .= $token->{data}; |
|
| 4912 |
!!!next-token; |
!!!nack ('t397.3'); |
|
} |
|
|
if (length $text) { |
|
|
!!!cp ('t396'); |
|
|
$el->manakai_append_text ($text); |
|
|
} |
|
|
|
|
|
$self->{content_model} = PCDATA_CONTENT_MODEL; |
|
|
|
|
|
if ($token->{type} == END_TAG_TOKEN and |
|
|
$token->{tag_name} eq $tag_name) { |
|
|
!!!cp ('t397'); |
|
|
## Ignore the token |
|
|
} else { |
|
|
!!!cp ('t398'); |
|
|
!!!parse-error (type => 'in RCDATA:#eof', token => $token); |
|
|
} |
|
| 4913 |
!!!next-token; |
!!!next-token; |
| 4914 |
next B; |
redo B; |
| 4915 |
} elsif ($token->{tag_name} eq 'rt' or |
} elsif ($token->{tag_name} eq 'rt' or |
| 4916 |
$token->{tag_name} eq 'rp') { |
$token->{tag_name} eq 'rp') { |
| 4917 |
## has a |ruby| element in scope |
## has a |ruby| element in scope |
| 4918 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4919 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 4920 |
if ($node->[1] & RUBY_EL) { |
if ($node->[1] == RUBY_EL) { |
| 4921 |
!!!cp ('t398.1'); |
!!!cp ('t398.1'); |
| 4922 |
## generate implied end tags |
## generate implied end tags |
| 4923 |
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 4924 |
!!!cp ('t398.2'); |
!!!cp ('t398.2'); |
| 4925 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 4926 |
} |
} |
| 4927 |
unless ($self->{open_elements}->[-1]->[1] & RUBY_EL) { |
unless ($self->{open_elements}->[-1]->[1] == RUBY_EL) { |
| 4928 |
!!!cp ('t398.3'); |
!!!cp ('t398.3'); |
| 4929 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 4930 |
text => $self->{open_elements}->[-1]->[0] |
text => $self->{open_elements}->[-1]->[0] |
| 4931 |
->manakai_local_name, |
->manakai_local_name, |
| 4932 |
token => $token); |
token => $token); |
| 4933 |
pop @{$self->{open_elements}} |
pop @{$self->{open_elements}} |
| 4934 |
while not $self->{open_elements}->[-1]->[1] & RUBY_EL; |
while not $self->{open_elements}->[-1]->[1] == RUBY_EL; |
| 4935 |
} |
} |
| 4936 |
last INSCOPE; |
last INSCOPE; |
| 4937 |
} elsif ($node->[1] & SCOPING_EL) { |
} elsif ($node->[1] & SCOPING_EL) { |
| 4939 |
last INSCOPE; |
last INSCOPE; |
| 4940 |
} |
} |
| 4941 |
} # INSCOPE |
} # INSCOPE |
| 4942 |
|
|
| 4943 |
|
## TODO: <non-ruby><rt> is not allowed. |
| 4944 |
|
|
| 4945 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4946 |
|
|
| 4961 |
|
|
| 4962 |
if ($self->{self_closing}) { |
if ($self->{self_closing}) { |
| 4963 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 4964 |
!!!ack ('t398.1'); |
!!!ack ('t398.6'); |
| 4965 |
} else { |
} else { |
| 4966 |
!!!cp ('t398.2'); |
!!!cp ('t398.7'); |
| 4967 |
$self->{insertion_mode} |= IN_FOREIGN_CONTENT_IM; |
$self->{insertion_mode} |= IN_FOREIGN_CONTENT_IM; |
| 4968 |
## NOTE: |<body><math><mi><svg>| -> "in foreign content" insertion |
## NOTE: |<body><math><mi><svg>| -> "in foreign content" insertion |
| 4969 |
## mode, "in body" (not "in foreign content") secondary insertion |
## mode, "in body" (not "in foreign content") secondary insertion |
| 4974 |
next B; |
next B; |
| 4975 |
} elsif ({ |
} elsif ({ |
| 4976 |
caption => 1, col => 1, colgroup => 1, frame => 1, |
caption => 1, col => 1, colgroup => 1, frame => 1, |
| 4977 |
frameset => 1, head => 1, option => 1, optgroup => 1, |
head => 1, |
| 4978 |
tbody => 1, td => 1, tfoot => 1, th => 1, |
tbody => 1, td => 1, tfoot => 1, th => 1, |
| 4979 |
thead => 1, tr => 1, |
thead => 1, tr => 1, |
| 4980 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 4985 |
!!!nack ('t401.1'); ## NOTE: |<col/>| or |<frame/>| here is an error. |
!!!nack ('t401.1'); ## NOTE: |<col/>| or |<frame/>| here is an error. |
| 4986 |
!!!next-token; |
!!!next-token; |
| 4987 |
next B; |
next B; |
| 4988 |
|
} elsif ($token->{tag_name} eq 'param' or |
| 4989 |
## ISSUE: An issue on HTML5 new elements in the spec. |
$token->{tag_name} eq 'source') { |
| 4990 |
|
!!!insert-element-t ($token->{tag_name}, $token->{attributes}, $token); |
| 4991 |
|
pop @{$self->{open_elements}}; |
| 4992 |
|
|
| 4993 |
|
!!!ack ('t398.5'); |
| 4994 |
|
!!!next-token; |
| 4995 |
|
redo B; |
| 4996 |
} else { |
} else { |
| 4997 |
if ($token->{tag_name} eq 'image') { |
if ($token->{tag_name} eq 'image') { |
| 4998 |
!!!cp ('t384'); |
!!!cp ('t384'); |
| 5011 |
applet => 1, marquee => 1, object => 1, |
applet => 1, marquee => 1, object => 1, |
| 5012 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5013 |
!!!cp ('t380'); |
!!!cp ('t380'); |
| 5014 |
|
|
| 5015 |
push @$active_formatting_elements, ['#marker', '']; |
push @$active_formatting_elements, ['#marker', '']; |
| 5016 |
|
|
| 5017 |
|
delete $self->{frameset_ok}; |
| 5018 |
|
|
| 5019 |
!!!nack ('t380.1'); |
!!!nack ('t380.1'); |
| 5020 |
} elsif ({ |
} elsif ({ |
| 5021 |
b => 1, big => 1, em => 1, font => 1, i => 1, |
b => 1, big => 1, em => 1, font => 1, i => 1, |
| 5022 |
s => 1, small => 1, strile => 1, |
s => 1, small => 1, strike => 1, |
| 5023 |
strong => 1, tt => 1, u => 1, |
strong => 1, tt => 1, u => 1, |
| 5024 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5025 |
!!!cp ('t375'); |
!!!cp ('t375'); |
| 5032 |
!!!ack ('t388.2'); |
!!!ack ('t388.2'); |
| 5033 |
} elsif ({ |
} elsif ({ |
| 5034 |
area => 1, basefont => 1, bgsound => 1, br => 1, |
area => 1, basefont => 1, bgsound => 1, br => 1, |
| 5035 |
embed => 1, img => 1, param => 1, spacer => 1, wbr => 1, |
embed => 1, img => 1, spacer => 1, wbr => 1, |
| 5036 |
#image => 1, |
keygen => 1, |
| 5037 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5038 |
!!!cp ('t388.1'); |
!!!cp ('t388.1'); |
| 5039 |
|
|
| 5040 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 5041 |
|
|
| 5042 |
|
delete $self->{frameset_ok}; |
| 5043 |
|
|
| 5044 |
!!!ack ('t388.3'); |
!!!ack ('t388.3'); |
| 5045 |
} elsif ($token->{tag_name} eq 'select') { |
} elsif ($token->{tag_name} eq 'select') { |
| 5046 |
## TODO: associate with $self->{form_element} if defined |
## TODO: associate with $self->{form_element} if defined |
| 5047 |
|
|
| 5048 |
|
delete $self->{frameset_ok}; |
| 5049 |
|
|
| 5050 |
if ($self->{insertion_mode} & TABLE_IMS or |
if ($self->{insertion_mode} & TABLE_IMS or |
| 5051 |
$self->{insertion_mode} & BODY_TABLE_IMS or |
$self->{insertion_mode} & BODY_TABLE_IMS or |
| 5052 |
$self->{insertion_mode} == IN_COLUMN_GROUP_IM) { |
($self->{insertion_mode} & IM_MASK) == IN_COLUMN_GROUP_IM) { |
| 5053 |
!!!cp ('t400.1'); |
!!!cp ('t400.1'); |
| 5054 |
$self->{insertion_mode} = IN_SELECT_IN_TABLE_IM; |
$self->{insertion_mode} = IN_SELECT_IN_TABLE_IM; |
| 5055 |
} else { |
} else { |
| 5065 |
next B; |
next B; |
| 5066 |
} |
} |
| 5067 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 5068 |
if ($token->{tag_name} eq 'body') { |
if ($token->{tag_name} eq 'body' or $token->{tag_name} eq 'html') { |
| 5069 |
## has a |body| element in scope |
|
| 5070 |
|
## 1. If not "have an element in scope": |
| 5071 |
|
## "has a |body| element in scope" |
| 5072 |
my $i; |
my $i; |
| 5073 |
INSCOPE: { |
INSCOPE: { |
| 5074 |
for (reverse @{$self->{open_elements}}) { |
for (reverse @{$self->{open_elements}}) { |
| 5075 |
if ($_->[1] & BODY_EL) { |
if ($_->[1] == BODY_EL) { |
| 5076 |
!!!cp ('t405'); |
!!!cp ('t405'); |
| 5077 |
$i = $_; |
$i = $_; |
| 5078 |
last INSCOPE; |
last INSCOPE; |
| 5082 |
} |
} |
| 5083 |
} |
} |
| 5084 |
|
|
| 5085 |
!!!parse-error (type => 'start tag not allowed', |
## NOTE: |<marquee></body>|, |<svg><foreignobject></body>|, |
| 5086 |
|
## and fragment cases. |
| 5087 |
|
|
| 5088 |
|
!!!parse-error (type => 'unmatched end tag', |
| 5089 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 5090 |
## NOTE: Ignore the token. |
## Ignore the token. (</body> or </html>) |
| 5091 |
!!!next-token; |
!!!next-token; |
| 5092 |
next B; |
next B; |
| 5093 |
} # INSCOPE |
} # INSCOPE |
| 5094 |
|
|
| 5095 |
|
## 2. If unclosed elements: |
| 5096 |
for (@{$self->{open_elements}}) { |
for (@{$self->{open_elements}}) { |
| 5097 |
unless ($_->[1] & ALL_END_TAG_OPTIONAL_EL) { |
unless ($_->[1] & ALL_END_TAG_OPTIONAL_EL || |
| 5098 |
|
$_->[1] == OPTGROUP_EL || |
| 5099 |
|
$_->[1] == OPTION_EL || |
| 5100 |
|
$_->[1] == RUBY_COMPONENT_EL) { |
| 5101 |
!!!cp ('t403'); |
!!!cp ('t403'); |
| 5102 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 5103 |
text => $_->[0]->manakai_local_name, |
text => $_->[0]->manakai_local_name, |
| 5108 |
} |
} |
| 5109 |
} |
} |
| 5110 |
|
|
| 5111 |
|
## 3. Switch the insertion mode. |
| 5112 |
$self->{insertion_mode} = AFTER_BODY_IM; |
$self->{insertion_mode} = AFTER_BODY_IM; |
| 5113 |
!!!next-token; |
if ($token->{tag_name} eq 'body') { |
|
next B; |
|
|
} elsif ($token->{tag_name} eq 'html') { |
|
|
## TODO: Update this code. It seems that the code below is not |
|
|
## up-to-date, though it has same effect as speced. |
|
|
if (@{$self->{open_elements}} > 1 and |
|
|
$self->{open_elements}->[1]->[1] & BODY_EL) { |
|
|
## ISSUE: There is an issue in the spec. |
|
|
unless ($self->{open_elements}->[-1]->[1] & BODY_EL) { |
|
|
!!!cp ('t406'); |
|
|
!!!parse-error (type => 'not closed', |
|
|
text => $self->{open_elements}->[1]->[0] |
|
|
->manakai_local_name, |
|
|
token => $token); |
|
|
} else { |
|
|
!!!cp ('t407'); |
|
|
} |
|
|
$self->{insertion_mode} = AFTER_BODY_IM; |
|
|
## reprocess |
|
|
next B; |
|
|
} else { |
|
|
!!!cp ('t408'); |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
| 5114 |
!!!next-token; |
!!!next-token; |
| 5115 |
next B; |
} else { # html |
| 5116 |
|
## Reprocess. |
| 5117 |
} |
} |
| 5118 |
|
next B; |
| 5119 |
} elsif ({ |
} elsif ({ |
| 5120 |
address => 1, blockquote => 1, center => 1, dir => 1, |
## NOTE: End tags for non-phrasing flow content elements |
| 5121 |
div => 1, dl => 1, fieldset => 1, listing => 1, |
|
| 5122 |
menu => 1, ol => 1, pre => 1, ul => 1, |
## NOTE: The normal ones |
| 5123 |
|
address => 1, article => 1, aside => 1, blockquote => 1, |
| 5124 |
|
center => 1, datagrid => 1, details => 1, dialog => 1, |
| 5125 |
|
dir => 1, div => 1, dl => 1, fieldset => 1, figure => 1, |
| 5126 |
|
footer => 1, header => 1, hgroup => 1, |
| 5127 |
|
listing => 1, menu => 1, nav => 1, |
| 5128 |
|
ol => 1, pre => 1, section => 1, ul => 1, |
| 5129 |
|
|
| 5130 |
|
## NOTE: As normal, but ... optional tags |
| 5131 |
dd => 1, dt => 1, li => 1, |
dd => 1, dt => 1, li => 1, |
| 5132 |
|
|
| 5133 |
applet => 1, button => 1, marquee => 1, object => 1, |
applet => 1, button => 1, marquee => 1, object => 1, |
| 5134 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5135 |
|
## NOTE: Code for <li> start tags includes "as if </li>" code. |
| 5136 |
|
## Code for <dt> or <dd> start tags includes "as if </dt> or |
| 5137 |
|
## </dd>" code. |
| 5138 |
|
|
| 5139 |
## has an element in scope |
## has an element in scope |
| 5140 |
my $i; |
my $i; |
| 5141 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5162 |
dd => ($token->{tag_name} ne 'dd'), |
dd => ($token->{tag_name} ne 'dd'), |
| 5163 |
dt => ($token->{tag_name} ne 'dt'), |
dt => ($token->{tag_name} ne 'dt'), |
| 5164 |
li => ($token->{tag_name} ne 'li'), |
li => ($token->{tag_name} ne 'li'), |
| 5165 |
|
option => 1, |
| 5166 |
|
optgroup => 1, |
| 5167 |
p => 1, |
p => 1, |
| 5168 |
rt => 1, |
rt => 1, |
| 5169 |
rp => 1, |
rp => 1, |
| 5196 |
!!!next-token; |
!!!next-token; |
| 5197 |
next B; |
next B; |
| 5198 |
} elsif ($token->{tag_name} eq 'form') { |
} elsif ($token->{tag_name} eq 'form') { |
| 5199 |
|
## NOTE: As normal, but interacts with the form element pointer |
| 5200 |
|
|
| 5201 |
undef $self->{form_element}; |
undef $self->{form_element}; |
| 5202 |
|
|
| 5203 |
## has an element in scope |
## has an element in scope |
| 5204 |
my $i; |
my $i; |
| 5205 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5206 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 5207 |
if ($node->[1] & FORM_EL) { |
if ($node->[1] == FORM_EL) { |
| 5208 |
!!!cp ('t418'); |
!!!cp ('t418'); |
| 5209 |
$i = $_; |
$i = $_; |
| 5210 |
last INSCOPE; |
last INSCOPE; |
| 5245 |
!!!next-token; |
!!!next-token; |
| 5246 |
next B; |
next B; |
| 5247 |
} elsif ({ |
} elsif ({ |
| 5248 |
|
## NOTE: As normal, except acts as a closer for any ... |
| 5249 |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
h1 => 1, h2 => 1, h3 => 1, h4 => 1, h5 => 1, h6 => 1, |
| 5250 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5251 |
## has an element in scope |
## has an element in scope |
| 5252 |
my $i; |
my $i; |
| 5253 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5254 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 5255 |
if ($node->[1] & HEADING_EL) { |
if ($node->[1] == HEADING_EL) { |
| 5256 |
!!!cp ('t423'); |
!!!cp ('t423'); |
| 5257 |
$i = $_; |
$i = $_; |
| 5258 |
last INSCOPE; |
last INSCOPE; |
| 5291 |
!!!next-token; |
!!!next-token; |
| 5292 |
next B; |
next B; |
| 5293 |
} elsif ($token->{tag_name} eq 'p') { |
} elsif ($token->{tag_name} eq 'p') { |
| 5294 |
|
## NOTE: As normal, except </p> implies <p> and ... |
| 5295 |
|
|
| 5296 |
## has an element in scope |
## has an element in scope |
| 5297 |
|
my $non_optional; |
| 5298 |
my $i; |
my $i; |
| 5299 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 5300 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 5301 |
if ($node->[1] & P_EL) { |
if ($node->[1] == P_EL) { |
| 5302 |
!!!cp ('t410.1'); |
!!!cp ('t410.1'); |
| 5303 |
$i = $_; |
$i = $_; |
| 5304 |
last INSCOPE; |
last INSCOPE; |
| 5305 |
} elsif ($node->[1] & SCOPING_EL) { |
} elsif ($node->[1] & SCOPING_EL) { |
| 5306 |
!!!cp ('t411.1'); |
!!!cp ('t411.1'); |
| 5307 |
last INSCOPE; |
last INSCOPE; |
| 5308 |
|
} elsif ($node->[1] & END_TAG_OPTIONAL_EL) { |
| 5309 |
|
## NOTE: |END_TAG_OPTIONAL_EL| includes "p" |
| 5310 |
|
!!!cp ('t411.2'); |
| 5311 |
|
# |
| 5312 |
|
} else { |
| 5313 |
|
!!!cp ('t411.3'); |
| 5314 |
|
$non_optional ||= $node; |
| 5315 |
|
# |
| 5316 |
} |
} |
| 5317 |
} # INSCOPE |
} # INSCOPE |
| 5318 |
|
|
| 5319 |
if (defined $i) { |
if (defined $i) { |
| 5320 |
if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
## 1. Generate implied end tags |
| 5321 |
ne $token->{tag_name}) { |
# |
| 5322 |
|
|
| 5323 |
|
## 2. If current node != "p", parse error |
| 5324 |
|
if ($non_optional) { |
| 5325 |
!!!cp ('t412.1'); |
!!!cp ('t412.1'); |
| 5326 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 5327 |
text => $self->{open_elements}->[-1]->[0] |
text => $non_optional->[0]->manakai_local_name, |
|
->manakai_local_name, |
|
| 5328 |
token => $token); |
token => $token); |
| 5329 |
} else { |
} else { |
| 5330 |
!!!cp ('t414.1'); |
!!!cp ('t414.1'); |
| 5331 |
} |
} |
| 5332 |
|
|
| 5333 |
|
## 3. Pop |
| 5334 |
splice @{$self->{open_elements}}, $i; |
splice @{$self->{open_elements}}, $i; |
| 5335 |
} else { |
} else { |
| 5336 |
!!!cp ('t413.1'); |
!!!cp ('t413.1'); |
| 5350 |
} elsif ({ |
} elsif ({ |
| 5351 |
a => 1, |
a => 1, |
| 5352 |
b => 1, big => 1, em => 1, font => 1, i => 1, |
b => 1, big => 1, em => 1, font => 1, i => 1, |
| 5353 |
nobr => 1, s => 1, small => 1, strile => 1, |
nobr => 1, s => 1, small => 1, strike => 1, |
| 5354 |
strong => 1, tt => 1, u => 1, |
strong => 1, tt => 1, u => 1, |
| 5355 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5356 |
!!!cp ('t427'); |
!!!cp ('t427'); |
| 5371 |
## Ignore the token. |
## Ignore the token. |
| 5372 |
!!!next-token; |
!!!next-token; |
| 5373 |
next B; |
next B; |
|
} elsif ({ |
|
|
caption => 1, col => 1, colgroup => 1, frame => 1, |
|
|
frameset => 1, head => 1, option => 1, optgroup => 1, |
|
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
|
|
thead => 1, tr => 1, |
|
|
area => 1, basefont => 1, bgsound => 1, |
|
|
embed => 1, hr => 1, iframe => 1, image => 1, |
|
|
img => 1, input => 1, isindex => 1, noembed => 1, |
|
|
noframes => 1, param => 1, select => 1, spacer => 1, |
|
|
table => 1, textarea => 1, wbr => 1, |
|
|
noscript => 0, ## TODO: if scripting is enabled |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!cp ('t429'); |
|
|
!!!parse-error (type => 'unmatched end tag', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
next B; |
|
|
|
|
|
## ISSUE: Issue on HTML5 new elements in spec |
|
|
|
|
| 5374 |
} else { |
} else { |
| 5375 |
|
if ($token->{tag_name} eq 'sarcasm') { |
| 5376 |
|
sleep 0.001; # take a deep breath |
| 5377 |
|
} |
| 5378 |
|
|
| 5379 |
## Step 1 |
## Step 1 |
| 5380 |
my $node_i = -1; |
my $node_i = -1; |
| 5381 |
my $node = $self->{open_elements}->[$node_i]; |
my $node = $self->{open_elements}->[$node_i]; |
| 5382 |
|
|
| 5383 |
## Step 2 |
## Step 2 |
| 5384 |
S2: { |
S2: { |
| 5385 |
if ($node->[0]->manakai_local_name eq $token->{tag_name}) { |
my $node_tag_name = $node->[0]->manakai_local_name; |
| 5386 |
|
$node_tag_name =~ tr/A-Z/a-z/; # for SVG camelCase tag names |
| 5387 |
|
if ($node_tag_name eq $token->{tag_name}) { |
| 5388 |
## Step 1 |
## Step 1 |
| 5389 |
## generate implied end tags |
## generate implied end tags |
| 5390 |
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
while ($self->{open_elements}->[-1]->[1] & END_TAG_OPTIONAL_EL) { |
| 5397 |
} |
} |
| 5398 |
|
|
| 5399 |
## Step 2 |
## Step 2 |
| 5400 |
if ($self->{open_elements}->[-1]->[0]->manakai_local_name |
my $current_tag_name |
| 5401 |
ne $token->{tag_name}) { |
= $self->{open_elements}->[-1]->[0]->manakai_local_name; |
| 5402 |
|
$current_tag_name =~ tr/A-Z/a-z/; |
| 5403 |
|
if ($current_tag_name ne $token->{tag_name}) { |
| 5404 |
!!!cp ('t431'); |
!!!cp ('t431'); |
| 5405 |
## NOTE: <x><y></x> |
## NOTE: <x><y></x> |
| 5406 |
!!!parse-error (type => 'not closed', |
!!!parse-error (type => 'not closed', |
| 5428 |
## Ignore the token |
## Ignore the token |
| 5429 |
!!!next-token; |
!!!next-token; |
| 5430 |
last S2; |
last S2; |
|
} |
|
| 5431 |
|
|
| 5432 |
|
## NOTE: |<span><dd></span>a|: In Safari 3.1.2 and Opera |
| 5433 |
|
## 9.27, "a" is a child of <dd> (conforming). In |
| 5434 |
|
## Firefox 3.0.2, "a" is a child of <body>. In WinIE 7, |
| 5435 |
|
## "a" is a child of both <body> and <dd>. |
| 5436 |
|
} |
| 5437 |
|
|
| 5438 |
!!!cp ('t434'); |
!!!cp ('t434'); |
| 5439 |
} |
} |
| 5440 |
|
|
| 5475 |
## TODO: script stuffs |
## TODO: script stuffs |
| 5476 |
} # _tree_construct_main |
} # _tree_construct_main |
| 5477 |
|
|
| 5478 |
|
## XXX: How this method is organized is somewhat out of date, although |
| 5479 |
|
## it still does what the current spec documents. |
| 5480 |
sub set_inner_html ($$$$;$) { |
sub set_inner_html ($$$$;$) { |
| 5481 |
my $class = shift; |
my $class = shift; |
| 5482 |
my $node = shift; |
my $node = shift; # /context/ |
| 5483 |
#my $s = \$_[0]; |
#my $s = \$_[0]; |
| 5484 |
my $onerror = $_[1]; |
my $onerror = $_[1]; |
| 5485 |
my $get_wrapper = $_[2] || sub ($) { return $_[0] }; |
my $get_wrapper = $_[2] || sub ($) { return $_[0] }; |
| 5486 |
|
|
|
## ISSUE: Should {confident} be true? |
|
|
|
|
| 5487 |
my $nt = $node->node_type; |
my $nt = $node->node_type; |
| 5488 |
if ($nt == 9) { |
if ($nt == 9) { # Document (invoke the algorithm with no /context/ element) |
| 5489 |
# MUST |
# MUST |
| 5490 |
|
|
| 5491 |
## Step 1 # MUST |
## Step 1 # MUST |
| 5500 |
|
|
| 5501 |
## Step 3, 4, 5 # MUST |
## Step 3, 4, 5 # MUST |
| 5502 |
$class->parse_char_string ($_[0] => $node, $onerror, $get_wrapper); |
$class->parse_char_string ($_[0] => $node, $onerror, $get_wrapper); |
| 5503 |
} elsif ($nt == 1) { |
} elsif ($nt == 1) { # Element (invoke the algorithm with /context/ element) |
| 5504 |
## TODO: If non-html element |
## TODO: If non-html element |
| 5505 |
|
|
| 5506 |
## NOTE: Most of this code is copied from |parse_string| |
## NOTE: Most of this code is copied from |parse_string| |
| 5507 |
|
|
| 5508 |
## TODO: Support for $get_wrapper |
## TODO: Support for $get_wrapper |
| 5509 |
|
|
| 5510 |
## Step 1 # MUST |
## F1. Create an HTML document. |
| 5511 |
my $this_doc = $node->owner_document; |
my $this_doc = $node->owner_document; |
| 5512 |
my $doc = $this_doc->implementation->create_document; |
my $doc = $this_doc->implementation->create_document; |
| 5513 |
$doc->manakai_is_html (1); |
$doc->manakai_is_html (1); |
| 5514 |
|
|
| 5515 |
|
## F2. Propagate quirkness flag |
| 5516 |
|
my $node_doc = $node->owner_document; |
| 5517 |
|
$doc->manakai_compat_mode ($node_doc->manakai_compat_mode); |
| 5518 |
|
|
| 5519 |
|
## F3. Create an HTML parser |
| 5520 |
my $p = $class->new; |
my $p = $class->new; |
| 5521 |
$p->{document} = $doc; |
$p->{document} = $doc; |
| 5522 |
|
|
| 5644 |
$p->_initialize_tokenizer; |
$p->_initialize_tokenizer; |
| 5645 |
$p->_initialize_tree_constructor; |
$p->_initialize_tree_constructor; |
| 5646 |
|
|
| 5647 |
## Step 2 |
## F4. If /context/ is not undef... |
| 5648 |
|
|
| 5649 |
|
## F4.1. content model flag |
| 5650 |
my $node_ln = $node->manakai_local_name; |
my $node_ln = $node->manakai_local_name; |
| 5651 |
$p->{content_model} = { |
$p->{content_model} = { |
| 5652 |
title => RCDATA_CONTENT_MODEL, |
title => RCDATA_CONTENT_MODEL, |
| 5662 |
}->{$node_ln}; |
}->{$node_ln}; |
| 5663 |
$p->{content_model} = PCDATA_CONTENT_MODEL |
$p->{content_model} = PCDATA_CONTENT_MODEL |
| 5664 |
unless defined $p->{content_model}; |
unless defined $p->{content_model}; |
|
## ISSUE: What is "the name of the element"? local name? |
|
| 5665 |
|
|
| 5666 |
$p->{inner_html_node} = [$node, $el_category->{$node_ln}]; |
$p->{inner_html_node} = [$node, $el_category->{$node_ln}]; |
| 5667 |
## TODO: Foreign element OK? |
## TODO: Foreign element OK? |
| 5668 |
|
|
| 5669 |
## Step 3 |
## F4.2. Root |html| element |
| 5670 |
my $root = $doc->create_element_ns |
my $root = $doc->create_element_ns |
| 5671 |
('http://www.w3.org/1999/xhtml', [undef, 'html']); |
('http://www.w3.org/1999/xhtml', [undef, 'html']); |
| 5672 |
|
|
| 5673 |
## Step 4 # MUST |
## F4.3. |
| 5674 |
$doc->append_child ($root); |
$doc->append_child ($root); |
| 5675 |
|
|
| 5676 |
## Step 5 # MUST |
## F4.4. |
| 5677 |
push @{$p->{open_elements}}, [$root, $el_category->{html}]; |
push @{$p->{open_elements}}, [$root, $el_category->{html}]; |
| 5678 |
|
|
| 5679 |
undef $p->{head_element}; |
undef $p->{head_element}; |
| 5680 |
|
undef $p->{head_element_inserted}; |
| 5681 |
|
|
| 5682 |
## Step 6 # MUST |
## F4.5. |
| 5683 |
$p->_reset_insertion_mode; |
$p->_reset_insertion_mode; |
| 5684 |
|
|
| 5685 |
## Step 7 # MUST |
## F4.6. |
| 5686 |
my $anode = $node; |
my $anode = $node; |
| 5687 |
AN: while (defined $anode) { |
AN: while (defined $anode) { |
| 5688 |
if ($anode->node_type == 1) { |
if ($anode->node_type == 1) { |
| 5697 |
} |
} |
| 5698 |
$anode = $anode->parent_node; |
$anode = $anode->parent_node; |
| 5699 |
} # AN |
} # AN |
| 5700 |
|
|
| 5701 |
## Step 9 # MUST |
## F.5. Set the input stream. |
| 5702 |
|
$p->{confident} = 1; ## Confident: irrelevant. |
| 5703 |
|
|
| 5704 |
|
## F.6. Start the parser. |
| 5705 |
{ |
{ |
| 5706 |
my $self = $p; |
my $self = $p; |
| 5707 |
!!!next-token; |
!!!next-token; |
| 5708 |
} |
} |
| 5709 |
$p->_tree_construction_main; |
$p->_tree_construction_main; |
| 5710 |
|
|
| 5711 |
## Step 10 # MUST |
## F.7. |
| 5712 |
my @cn = @{$node->child_nodes}; |
my @cn = @{$node->child_nodes}; |
| 5713 |
for (@cn) { |
for (@cn) { |
| 5714 |
$node->remove_child ($_); |
$node->remove_child ($_); |