| 150 |
return $self; |
return $self; |
| 151 |
} # new |
} # new |
| 152 |
|
|
| 153 |
|
sub CM_ENTITY () { 0b001 } # & markup in data |
| 154 |
|
sub CM_LIMITED_MARKUP () { 0b010 } # < markup in data (limited) |
| 155 |
|
sub CM_FULL_MARKUP () { 0b100 } # < markup in data (any) |
| 156 |
|
|
| 157 |
|
sub PLAINTEXT_CONTENT_MODEL () { 0 } |
| 158 |
|
sub CDATA_CONTENT_MODEL () { CM_LIMITED_MARKUP } |
| 159 |
|
sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP } |
| 160 |
|
sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP } |
| 161 |
|
|
| 162 |
## Implementations MUST act as if state machine in the spec |
## Implementations MUST act as if state machine in the spec |
| 163 |
|
|
| 164 |
sub _initialize_tokenizer ($) { |
sub _initialize_tokenizer ($) { |
| 165 |
my $self = shift; |
my $self = shift; |
| 166 |
$self->{state} = 'data'; # MUST |
$self->{state} = 'data'; # MUST |
| 167 |
$self->{content_model_flag} = 'PCDATA'; # be |
$self->{content_model} = PCDATA_CONTENT_MODEL; # be |
| 168 |
undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE |
undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE |
| 169 |
undef $self->{current_attribute}; |
undef $self->{current_attribute}; |
| 170 |
undef $self->{last_emitted_start_tag_name}; |
undef $self->{last_emitted_start_tag_name}; |
| 203 |
A: { |
A: { |
| 204 |
if ($self->{state} eq 'data') { |
if ($self->{state} eq 'data') { |
| 205 |
if ($self->{next_input_character} == 0x0026) { # & |
if ($self->{next_input_character} == 0x0026) { # & |
| 206 |
if ($self->{content_model_flag} eq 'PCDATA' or |
if ($self->{content_model} & CM_ENTITY) { # PCDATA | RCDATA |
|
$self->{content_model_flag} eq 'RCDATA') { |
|
| 207 |
$self->{state} = 'entity data'; |
$self->{state} = 'entity data'; |
| 208 |
!!!next-input-character; |
!!!next-input-character; |
| 209 |
redo A; |
redo A; |
| 211 |
# |
# |
| 212 |
} |
} |
| 213 |
} elsif ($self->{next_input_character} == 0x002D) { # - |
} elsif ($self->{next_input_character} == 0x002D) { # - |
| 214 |
if ($self->{content_model_flag} eq 'RCDATA' or |
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
$self->{content_model_flag} eq 'CDATA') { |
|
| 215 |
unless ($self->{escape}) { |
unless ($self->{escape}) { |
| 216 |
if ($self->{prev_input_character}->[0] == 0x002D and # - |
if ($self->{prev_input_character}->[0] == 0x002D and # - |
| 217 |
$self->{prev_input_character}->[1] == 0x0021 and # ! |
$self->{prev_input_character}->[1] == 0x0021 and # ! |
| 223 |
|
|
| 224 |
# |
# |
| 225 |
} elsif ($self->{next_input_character} == 0x003C) { # < |
} elsif ($self->{next_input_character} == 0x003C) { # < |
| 226 |
if ($self->{content_model_flag} eq 'PCDATA' or |
if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA |
| 227 |
(($self->{content_model_flag} eq 'CDATA' or |
(($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA |
|
$self->{content_model_flag} eq 'RCDATA') and |
|
| 228 |
not $self->{escape})) { |
not $self->{escape})) { |
| 229 |
$self->{state} = 'tag open'; |
$self->{state} = 'tag open'; |
| 230 |
!!!next-input-character; |
!!!next-input-character; |
| 234 |
} |
} |
| 235 |
} elsif ($self->{next_input_character} == 0x003E) { # > |
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 236 |
if ($self->{escape} and |
if ($self->{escape} and |
| 237 |
($self->{content_model_flag} eq 'RCDATA' or |
($self->{content_model} & CM_LIMITED_MARKUP)) { # RCDATA | CDATA |
|
$self->{content_model_flag} eq 'CDATA')) { |
|
| 238 |
if ($self->{prev_input_character}->[0] == 0x002D and # - |
if ($self->{prev_input_character}->[0] == 0x002D and # - |
| 239 |
$self->{prev_input_character}->[1] == 0x002D) { # - |
$self->{prev_input_character}->[1] == 0x002D) { # - |
| 240 |
delete $self->{escape}; |
delete $self->{escape}; |
| 271 |
|
|
| 272 |
redo A; |
redo A; |
| 273 |
} elsif ($self->{state} eq 'tag open') { |
} elsif ($self->{state} eq 'tag open') { |
| 274 |
if ($self->{content_model_flag} eq 'RCDATA' or |
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
$self->{content_model_flag} eq 'CDATA') { |
|
| 275 |
if ($self->{next_input_character} == 0x002F) { # / |
if ($self->{next_input_character} == 0x002F) { # / |
| 276 |
!!!next-input-character; |
!!!next-input-character; |
| 277 |
$self->{state} = 'close tag open'; |
$self->{state} = 'close tag open'; |
| 284 |
|
|
| 285 |
redo A; |
redo A; |
| 286 |
} |
} |
| 287 |
} elsif ($self->{content_model_flag} eq 'PCDATA') { |
} elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA |
| 288 |
if ($self->{next_input_character} == 0x0021) { # ! |
if ($self->{next_input_character} == 0x0021) { # ! |
| 289 |
$self->{state} = 'markup declaration open'; |
$self->{state} = 'markup declaration open'; |
| 290 |
!!!next-input-character; |
!!!next-input-character; |
| 331 |
redo A; |
redo A; |
| 332 |
} |
} |
| 333 |
} else { |
} else { |
| 334 |
die "$0: $self->{content_model_flag}: Unknown content model flag"; |
die "$0: $self->{content_model} in tag open"; |
| 335 |
} |
} |
| 336 |
} elsif ($self->{state} eq 'close tag open') { |
} elsif ($self->{state} eq 'close tag open') { |
| 337 |
if ($self->{content_model_flag} eq 'RCDATA' or |
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
|
$self->{content_model_flag} eq 'CDATA') { |
|
| 338 |
if (defined $self->{last_emitted_start_tag_name}) { |
if (defined $self->{last_emitted_start_tag_name}) { |
| 339 |
## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564> |
## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564> |
| 340 |
my @next_char; |
my @next_char; |
| 432 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 433 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 434 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 435 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 436 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 437 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 438 |
} |
} |
| 459 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 460 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 461 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 462 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 463 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 464 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 465 |
} |
} |
| 507 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 508 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 509 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 510 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 511 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 512 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 513 |
} |
} |
| 547 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 548 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 549 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 550 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 551 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 552 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 553 |
} |
} |
| 571 |
my $before_leave = sub { |
my $before_leave = sub { |
| 572 |
if (exists $self->{current_token}->{attributes} # start tag or end tag |
if (exists $self->{current_token}->{attributes} # start tag or end tag |
| 573 |
->{$self->{current_attribute}->{name}}) { # MUST |
->{$self->{current_attribute}->{name}}) { # MUST |
| 574 |
!!!parse-error (type => 'dupulicate attribute'); |
!!!parse-error (type => 'duplicate attribute:'.$self->{current_attribute}->{name}); |
| 575 |
## Discard $self->{current_attribute} # MUST |
## Discard $self->{current_attribute} # MUST |
| 576 |
} else { |
} else { |
| 577 |
$self->{current_token}->{attributes}->{$self->{current_attribute}->{name}} |
$self->{current_token}->{attributes}->{$self->{current_attribute}->{name}} |
| 600 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 601 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 602 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 603 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 604 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 605 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 606 |
} |
} |
| 641 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 642 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 643 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 644 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 645 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 646 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 647 |
} |
} |
| 679 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 680 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 681 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 682 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 683 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 684 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 685 |
} |
} |
| 720 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 721 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 722 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 723 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 724 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 725 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 726 |
} |
} |
| 767 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 768 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 769 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 770 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 771 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 772 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 773 |
} |
} |
| 787 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 788 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 789 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 790 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 791 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 792 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 793 |
} |
} |
| 823 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 824 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 825 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 826 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 827 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 828 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 829 |
} |
} |
| 859 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 860 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 861 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 862 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 863 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 864 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 865 |
} |
} |
| 898 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 899 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 900 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 901 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 902 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 903 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 904 |
} |
} |
| 918 |
= not defined $self->{last_emitted_start_tag_name}; |
= not defined $self->{last_emitted_start_tag_name}; |
| 919 |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
$self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name}; |
| 920 |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
} elsif ($self->{current_token}->{type} eq 'end tag') { |
| 921 |
$self->{content_model_flag} = 'PCDATA'; # MUST |
$self->{content_model} = PCDATA_CONTENT_MODEL; # MUST |
| 922 |
if ($self->{current_token}->{attributes}) { |
if ($self->{current_token}->{attributes}) { |
| 923 |
!!!parse-error (type => 'end tag attribute'); |
!!!parse-error (type => 'end tag attribute'); |
| 924 |
} |
} |
| 1647 |
redo X; |
redo X; |
| 1648 |
} elsif (not defined $code) { # no hexadecimal digit |
} elsif (not defined $code) { # no hexadecimal digit |
| 1649 |
!!!parse-error (type => 'bare hcro'); |
!!!parse-error (type => 'bare hcro'); |
| 1650 |
|
!!!back-next-input-character ($x_char, $self->{next_input_character}); |
| 1651 |
$self->{next_input_character} = 0x0023; # # |
$self->{next_input_character} = 0x0023; # # |
|
!!!back-next-input-character ($x_char); |
|
| 1652 |
return undef; |
return undef; |
| 1653 |
} elsif ($self->{next_input_character} == 0x003B) { # ; |
} elsif ($self->{next_input_character} == 0x003B) { # ; |
| 1654 |
!!!next-input-character; |
!!!next-input-character; |
| 1720 |
!!!next-input-character; |
!!!next-input-character; |
| 1721 |
|
|
| 1722 |
my $value = $entity_name; |
my $value = $entity_name; |
| 1723 |
my $match; |
my $match = 0; |
| 1724 |
require Whatpm::_NamedEntityList; |
require Whatpm::_NamedEntityList; |
| 1725 |
our $EntityChar; |
our $EntityChar; |
| 1726 |
|
|
| 1740 |
$match = 1; |
$match = 1; |
| 1741 |
!!!next-input-character; |
!!!next-input-character; |
| 1742 |
last; |
last; |
| 1743 |
} elsif (not $in_attr) { |
} else { |
| 1744 |
$value = $EntityChar->{$entity_name}; |
$value = $EntityChar->{$entity_name}; |
| 1745 |
$match = -1; |
$match = -1; |
| 1746 |
} else { |
!!!next-input-character; |
|
$value .= chr $self->{next_input_character}; |
|
| 1747 |
} |
} |
| 1748 |
} else { |
} else { |
| 1749 |
$value .= chr $self->{next_input_character}; |
$value .= chr $self->{next_input_character}; |
| 1750 |
|
$match *= 2; |
| 1751 |
|
!!!next-input-character; |
| 1752 |
} |
} |
|
!!!next-input-character; |
|
| 1753 |
} |
} |
| 1754 |
|
|
| 1755 |
if ($match > 0) { |
if ($match > 0) { |
| 1756 |
return {type => 'character', data => $value}; |
return {type => 'character', data => $value}; |
| 1757 |
} elsif ($match < 0) { |
} elsif ($match < 0) { |
| 1758 |
!!!parse-error (type => 'no refc'); |
!!!parse-error (type => 'no refc'); |
| 1759 |
return {type => 'character', data => $value}; |
if ($in_attr and $match < -1) { |
| 1760 |
|
return {type => 'character', data => '&'.$entity_name}; |
| 1761 |
|
} else { |
| 1762 |
|
return {type => 'character', data => $value}; |
| 1763 |
|
} |
| 1764 |
} else { |
} else { |
| 1765 |
!!!parse-error (type => 'bare ero'); |
!!!parse-error (type => 'bare ero'); |
| 1766 |
## NOTE: No characters are consumed in the spec. |
## NOTE: No characters are consumed in the spec. |
| 2067 |
th => 'in cell', |
th => 'in cell', |
| 2068 |
tr => 'in row', |
tr => 'in row', |
| 2069 |
tbody => 'in table body', |
tbody => 'in table body', |
| 2070 |
thead => 'in table head', |
thead => 'in table body', |
| 2071 |
tfoot => 'in table foot', |
tfoot => 'in table body', |
| 2072 |
caption => 'in caption', |
caption => 'in caption', |
| 2073 |
colgroup => 'in column group', |
colgroup => 'in column group', |
| 2074 |
table => 'in table', |
table => 'in table', |
| 2103 |
sub _tree_construction_main ($) { |
sub _tree_construction_main ($) { |
| 2104 |
my $self = shift; |
my $self = shift; |
| 2105 |
|
|
|
my $previous_insertion_mode; |
|
|
|
|
| 2106 |
my $active_formatting_elements = []; |
my $active_formatting_elements = []; |
| 2107 |
|
|
| 2108 |
my $reconstruct_active_formatting_elements = sub { # MUST |
my $reconstruct_active_formatting_elements = sub { # MUST |
| 2197 |
$insert->($el); # /context node/->append_child ($el) |
$insert->($el); # /context node/->append_child ($el) |
| 2198 |
|
|
| 2199 |
## Step 3 |
## Step 3 |
| 2200 |
$self->{content_model_flag} = $content_model_flag; # CDATA or RCDATA |
$self->{content_model} = $content_model_flag; # CDATA or RCDATA |
| 2201 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
| 2202 |
|
|
| 2203 |
## Step 4 |
## Step 4 |
| 2215 |
} |
} |
| 2216 |
|
|
| 2217 |
## Step 6 |
## Step 6 |
| 2218 |
$self->{content_model_flag} = 'PCDATA'; |
$self->{content_model} = PCDATA_CONTENT_MODEL; |
| 2219 |
|
|
| 2220 |
## Step 7 |
## Step 7 |
| 2221 |
if ($token->{type} eq 'end tag' and $token->{tag_name} eq $start_tag_name) { |
if ($token->{type} eq 'end tag' and $token->{tag_name} eq $start_tag_name) { |
| 2222 |
## Ignore the token |
## Ignore the token |
| 2223 |
|
} elsif ($content_model_flag == CDATA_CONTENT_MODEL) { |
| 2224 |
|
!!!parse-error (type => 'in CDATA:#'.$token->{type}); |
| 2225 |
|
} elsif ($content_model_flag == RCDATA_CONTENT_MODEL) { |
| 2226 |
|
!!!parse-error (type => 'in RCDATA:#'.$token->{type}); |
| 2227 |
} else { |
} else { |
| 2228 |
!!!parse-error (type => 'in '.$content_model_flag.':#'.$token->{type}); |
die "$0: $content_model_flag in parse_rcdata"; |
| 2229 |
} |
} |
| 2230 |
!!!next-token; |
!!!next-token; |
| 2231 |
}; # $parse_rcdata |
}; # $parse_rcdata |
| 2236 |
!!!create-element ($script_el, 'script', $token->{attributes}); |
!!!create-element ($script_el, 'script', $token->{attributes}); |
| 2237 |
## TODO: mark as "parser-inserted" |
## TODO: mark as "parser-inserted" |
| 2238 |
|
|
| 2239 |
$self->{content_model_flag} = 'CDATA'; |
$self->{content_model} = CDATA_CONTENT_MODEL; |
| 2240 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
| 2241 |
|
|
| 2242 |
my $text = ''; |
my $text = ''; |
| 2249 |
$script_el->manakai_append_text ($text); |
$script_el->manakai_append_text ($text); |
| 2250 |
} |
} |
| 2251 |
|
|
| 2252 |
$self->{content_model_flag} = 'PCDATA'; |
$self->{content_model} = PCDATA_CONTENT_MODEL; |
| 2253 |
|
|
| 2254 |
if ($token->{type} eq 'end tag' and |
if ($token->{type} eq 'end tag' and |
| 2255 |
$token->{tag_name} eq 'script') { |
$token->{tag_name} eq 'script') { |
| 2494 |
} |
} |
| 2495 |
}; # $insert_to_foster |
}; # $insert_to_foster |
| 2496 |
|
|
| 2497 |
|
my $insert; |
| 2498 |
|
|
| 2499 |
|
B: { |
| 2500 |
|
if ($token->{type} eq 'DOCTYPE') { |
| 2501 |
|
!!!parse-error (type => 'DOCTYPE in the middle'); |
| 2502 |
|
## Ignore the token |
| 2503 |
|
## Stay in the phase |
| 2504 |
|
!!!next-token; |
| 2505 |
|
redo B; |
| 2506 |
|
} elsif ($token->{type} eq 'end-of-file') { |
| 2507 |
|
if ($self->{insertion_mode} eq 'after html body' or |
| 2508 |
|
$self->{insertion_mode} eq 'after html frameset') { |
| 2509 |
|
# |
| 2510 |
|
} else { |
| 2511 |
|
## Generate implied end tags |
| 2512 |
|
if ({ |
| 2513 |
|
dd => 1, dt => 1, li => 1, p => 1, td => 1, th => 1, tr => 1, |
| 2514 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 2515 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 2516 |
|
!!!back-token; |
| 2517 |
|
$token = {type => 'end tag', tag_name => $self->{open_elements}->[-1]->[1]}; |
| 2518 |
|
redo B; |
| 2519 |
|
} |
| 2520 |
|
|
| 2521 |
|
if (@{$self->{open_elements}} > 2 or |
| 2522 |
|
(@{$self->{open_elements}} == 2 and $self->{open_elements}->[1]->[1] ne 'body')) { |
| 2523 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2524 |
|
} elsif (defined $self->{inner_html_node} and |
| 2525 |
|
@{$self->{open_elements}} > 1 and |
| 2526 |
|
$self->{open_elements}->[1]->[1] ne 'body') { |
| 2527 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 2528 |
|
} |
| 2529 |
|
|
| 2530 |
|
## ISSUE: There is an issue in the spec. |
| 2531 |
|
} |
| 2532 |
|
|
| 2533 |
|
## Stop parsing |
| 2534 |
|
last B; |
| 2535 |
|
} elsif ($token->{type} eq 'start tag' and |
| 2536 |
|
$token->{tag_name} eq 'html') { |
| 2537 |
|
if ($self->{insertion_mode} eq 'after html body') { |
| 2538 |
|
## Turn into the main phase |
| 2539 |
|
!!!parse-error (type => 'after html:html'); |
| 2540 |
|
$self->{insertion_mode} = 'after body'; |
| 2541 |
|
} elsif ($self->{insertion_mode} eq 'after html frameset') { |
| 2542 |
|
## Turn into the main phase |
| 2543 |
|
!!!parse-error (type => 'after html:html'); |
| 2544 |
|
$self->{insertion_mode} = 'after frameset'; |
| 2545 |
|
} |
| 2546 |
|
|
| 2547 |
|
## ISSUE: "aa<html>" is not a parse error. |
| 2548 |
|
## ISSUE: "<html>" in fragment is not a parse error. |
| 2549 |
|
unless ($token->{first_start_tag}) { |
| 2550 |
|
!!!parse-error (type => 'not first start tag'); |
| 2551 |
|
} |
| 2552 |
|
my $top_el = $self->{open_elements}->[0]->[0]; |
| 2553 |
|
for my $attr_name (keys %{$token->{attributes}}) { |
| 2554 |
|
unless ($top_el->has_attribute_ns (undef, $attr_name)) { |
| 2555 |
|
$top_el->set_attribute_ns |
| 2556 |
|
(undef, [undef, $attr_name], |
| 2557 |
|
$token->{attributes}->{$attr_name}->{value}); |
| 2558 |
|
} |
| 2559 |
|
} |
| 2560 |
|
!!!next-token; |
| 2561 |
|
redo B; |
| 2562 |
|
} elsif ($token->{type} eq 'comment') { |
| 2563 |
|
my $comment = $self->{document}->create_comment ($token->{data}); |
| 2564 |
|
if ($self->{insertion_mode} eq 'after html body' or |
| 2565 |
|
$self->{insertion_mode} eq 'after html frameset') { |
| 2566 |
|
$self->{document}->append_child ($comment); |
| 2567 |
|
} elsif ($self->{insertion_mode} eq 'after body') { |
| 2568 |
|
$self->{open_elements}->[0]->[0]->append_child ($comment); |
| 2569 |
|
} else { |
| 2570 |
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
| 2571 |
|
} |
| 2572 |
|
!!!next-token; |
| 2573 |
|
redo B; |
| 2574 |
|
} elsif ($self->{insertion_mode} eq 'in head' or |
| 2575 |
|
$self->{insertion_mode} eq 'in head noscript' or |
| 2576 |
|
$self->{insertion_mode} eq 'after head' or |
| 2577 |
|
$self->{insertion_mode} eq 'before head') { |
| 2578 |
|
if ($token->{type} eq 'character') { |
| 2579 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 2580 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 2581 |
|
unless (length $token->{data}) { |
| 2582 |
|
!!!next-token; |
| 2583 |
|
redo B; |
| 2584 |
|
} |
| 2585 |
|
} |
| 2586 |
|
|
| 2587 |
|
if ($self->{insertion_mode} eq 'before head') { |
| 2588 |
|
## As if <head> |
| 2589 |
|
!!!create-element ($self->{head_element}, 'head'); |
| 2590 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2591 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2592 |
|
|
| 2593 |
|
## Reprocess in the "in head" insertion mode... |
| 2594 |
|
pop @{$self->{open_elements}}; |
| 2595 |
|
|
| 2596 |
|
## Reprocess in the "after head" insertion mode... |
| 2597 |
|
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
| 2598 |
|
## As if </noscript> |
| 2599 |
|
pop @{$self->{open_elements}}; |
| 2600 |
|
!!!parse-error (type => 'in noscript:#character'); |
| 2601 |
|
|
| 2602 |
|
## Reprocess in the "in head" insertion mode... |
| 2603 |
|
## As if </head> |
| 2604 |
|
pop @{$self->{open_elements}}; |
| 2605 |
|
|
| 2606 |
|
## Reprocess in the "after head" insertion mode... |
| 2607 |
|
} elsif ($self->{insertion_mode} eq 'in head') { |
| 2608 |
|
pop @{$self->{open_elements}}; |
| 2609 |
|
|
| 2610 |
|
## Reprocess in the "after head" insertion mode... |
| 2611 |
|
} |
| 2612 |
|
|
| 2613 |
|
## "after head" insertion mode |
| 2614 |
|
## As if <body> |
| 2615 |
|
!!!insert-element ('body'); |
| 2616 |
|
$self->{insertion_mode} = 'in body'; |
| 2617 |
|
## reprocess |
| 2618 |
|
redo B; |
| 2619 |
|
} elsif ($token->{type} eq 'start tag') { |
| 2620 |
|
if ($token->{tag_name} eq 'head') { |
| 2621 |
|
if ($self->{insertion_mode} eq 'before head') { |
| 2622 |
|
!!!create-element ($self->{head_element}, $token->{tag_name}, $token->{attributes}); |
| 2623 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2624 |
|
push @{$self->{open_elements}}, [$self->{head_element}, $token->{tag_name}]; |
| 2625 |
|
$self->{insertion_mode} = 'in head'; |
| 2626 |
|
!!!next-token; |
| 2627 |
|
redo B; |
| 2628 |
|
} elsif ($self->{insertion_mode} ne 'after head') { |
| 2629 |
|
!!!parse-error (type => 'in head:head'); # or in head noscript |
| 2630 |
|
## Ignore the token |
| 2631 |
|
!!!next-token; |
| 2632 |
|
redo B; |
| 2633 |
|
} else { |
| 2634 |
|
# |
| 2635 |
|
} |
| 2636 |
|
} elsif ($self->{insertion_mode} eq 'before head') { |
| 2637 |
|
## As if <head> |
| 2638 |
|
!!!create-element ($self->{head_element}, 'head'); |
| 2639 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2640 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2641 |
|
|
| 2642 |
|
$self->{insertion_mode} = 'in head'; |
| 2643 |
|
## Reprocess in the "in head" insertion mode... |
| 2644 |
|
} |
| 2645 |
|
|
| 2646 |
|
if ($token->{tag_name} eq 'base') { |
| 2647 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2648 |
|
## As if </noscript> |
| 2649 |
|
pop @{$self->{open_elements}}; |
| 2650 |
|
!!!parse-error (type => 'in noscript:base'); |
| 2651 |
|
|
| 2652 |
|
$self->{insertion_mode} = 'in head'; |
| 2653 |
|
## Reprocess in the "in head" insertion mode... |
| 2654 |
|
} |
| 2655 |
|
|
| 2656 |
|
## NOTE: There is a "as if in head" code clone. |
| 2657 |
|
if ($self->{insertion_mode} eq 'after head') { |
| 2658 |
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 2659 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2660 |
|
} |
| 2661 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2662 |
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
| 2663 |
|
pop @{$self->{open_elements}} |
| 2664 |
|
if $self->{insertion_mode} eq 'after head'; |
| 2665 |
|
!!!next-token; |
| 2666 |
|
redo B; |
| 2667 |
|
} elsif ($token->{tag_name} eq 'link') { |
| 2668 |
|
## NOTE: There is a "as if in head" code clone. |
| 2669 |
|
if ($self->{insertion_mode} eq 'after head') { |
| 2670 |
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 2671 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2672 |
|
} |
| 2673 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2674 |
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
| 2675 |
|
pop @{$self->{open_elements}} |
| 2676 |
|
if $self->{insertion_mode} eq 'after head'; |
| 2677 |
|
!!!next-token; |
| 2678 |
|
redo B; |
| 2679 |
|
} elsif ($token->{tag_name} eq 'meta') { |
| 2680 |
|
## NOTE: There is a "as if in head" code clone. |
| 2681 |
|
if ($self->{insertion_mode} eq 'after head') { |
| 2682 |
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 2683 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2684 |
|
} |
| 2685 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2686 |
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
| 2687 |
|
|
| 2688 |
|
unless ($self->{confident}) { |
| 2689 |
|
my $charset; |
| 2690 |
|
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
| 2691 |
|
$charset = $token->{attributes}->{charset}->{value}; |
| 2692 |
|
} |
| 2693 |
|
if ($token->{attributes}->{'http-equiv'}) { |
| 2694 |
|
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
| 2695 |
|
if ($token->{attributes}->{'http-equiv'}->{value} |
| 2696 |
|
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
| 2697 |
|
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 2698 |
|
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
| 2699 |
|
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
| 2700 |
|
} ## TODO: And if supported |
| 2701 |
|
} |
| 2702 |
|
## TODO: Change the encoding |
| 2703 |
|
} |
| 2704 |
|
|
| 2705 |
|
## TODO: Extracting |charset| from |meta|. |
| 2706 |
|
pop @{$self->{open_elements}} |
| 2707 |
|
if $self->{insertion_mode} eq 'after head'; |
| 2708 |
|
!!!next-token; |
| 2709 |
|
redo B; |
| 2710 |
|
} elsif ($token->{tag_name} eq 'title') { |
| 2711 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2712 |
|
## As if </noscript> |
| 2713 |
|
pop @{$self->{open_elements}}; |
| 2714 |
|
!!!parse-error (type => 'in noscript:title'); |
| 2715 |
|
|
| 2716 |
|
$self->{insertion_mode} = 'in head'; |
| 2717 |
|
## Reprocess in the "in head" insertion mode... |
| 2718 |
|
} elsif ($self->{insertion_mode} eq 'after head') { |
| 2719 |
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 2720 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2721 |
|
} |
| 2722 |
|
|
| 2723 |
|
## NOTE: There is a "as if in head" code clone. |
| 2724 |
|
my $parent = defined $self->{head_element} ? $self->{head_element} |
| 2725 |
|
: $self->{open_elements}->[-1]->[0]; |
| 2726 |
|
$parse_rcdata->(RCDATA_CONTENT_MODEL, |
| 2727 |
|
sub { $parent->append_child ($_[0]) }); |
| 2728 |
|
pop @{$self->{open_elements}} |
| 2729 |
|
if $self->{insertion_mode} eq 'after head'; |
| 2730 |
|
redo B; |
| 2731 |
|
} elsif ($token->{tag_name} eq 'style') { |
| 2732 |
|
## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
| 2733 |
|
## insertion mode 'in head') |
| 2734 |
|
## NOTE: There is a "as if in head" code clone. |
| 2735 |
|
if ($self->{insertion_mode} eq 'after head') { |
| 2736 |
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 2737 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2738 |
|
} |
| 2739 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert_to_current); |
| 2740 |
|
pop @{$self->{open_elements}} |
| 2741 |
|
if $self->{insertion_mode} eq 'after head'; |
| 2742 |
|
redo B; |
| 2743 |
|
} elsif ($token->{tag_name} eq 'noscript') { |
| 2744 |
|
if ($self->{insertion_mode} eq 'in head') { |
| 2745 |
|
## NOTE: and scripting is disalbed |
| 2746 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2747 |
|
$self->{insertion_mode} = 'in head noscript'; |
| 2748 |
|
!!!next-token; |
| 2749 |
|
redo B; |
| 2750 |
|
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
| 2751 |
|
!!!parse-error (type => 'in noscript:noscript'); |
| 2752 |
|
## Ignore the token |
| 2753 |
|
!!!next-token; |
| 2754 |
|
redo B; |
| 2755 |
|
} else { |
| 2756 |
|
# |
| 2757 |
|
} |
| 2758 |
|
} elsif ($token->{tag_name} eq 'script') { |
| 2759 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2760 |
|
## As if </noscript> |
| 2761 |
|
pop @{$self->{open_elements}}; |
| 2762 |
|
!!!parse-error (type => 'in noscript:script'); |
| 2763 |
|
|
| 2764 |
|
$self->{insertion_mode} = 'in head'; |
| 2765 |
|
## Reprocess in the "in head" insertion mode... |
| 2766 |
|
} elsif ($self->{insertion_mode} eq 'after head') { |
| 2767 |
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
| 2768 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2769 |
|
} |
| 2770 |
|
|
| 2771 |
|
## NOTE: There is a "as if in head" code clone. |
| 2772 |
|
$script_start_tag->($insert_to_current); |
| 2773 |
|
pop @{$self->{open_elements}} |
| 2774 |
|
if $self->{insertion_mode} eq 'after head'; |
| 2775 |
|
redo B; |
| 2776 |
|
} elsif ($token->{tag_name} eq 'body' or |
| 2777 |
|
$token->{tag_name} eq 'frameset') { |
| 2778 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2779 |
|
## As if </noscript> |
| 2780 |
|
pop @{$self->{open_elements}}; |
| 2781 |
|
!!!parse-error (type => 'in noscript:'.$token->{tag_name}); |
| 2782 |
|
|
| 2783 |
|
## Reprocess in the "in head" insertion mode... |
| 2784 |
|
## As if </head> |
| 2785 |
|
pop @{$self->{open_elements}}; |
| 2786 |
|
|
| 2787 |
|
## Reprocess in the "after head" insertion mode... |
| 2788 |
|
} elsif ($self->{insertion_mode} eq 'in head') { |
| 2789 |
|
pop @{$self->{open_elements}}; |
| 2790 |
|
|
| 2791 |
|
## Reprocess in the "after head" insertion mode... |
| 2792 |
|
} |
| 2793 |
|
|
| 2794 |
|
## "after head" insertion mode |
| 2795 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2796 |
|
$self->{insertion_mode} = 'in '.$token->{tag_name}; |
| 2797 |
|
!!!next-token; |
| 2798 |
|
redo B; |
| 2799 |
|
} else { |
| 2800 |
|
# |
| 2801 |
|
} |
| 2802 |
|
|
| 2803 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2804 |
|
## As if </noscript> |
| 2805 |
|
pop @{$self->{open_elements}}; |
| 2806 |
|
!!!parse-error (type => 'in noscript:/'.$token->{tag_name}); |
| 2807 |
|
|
| 2808 |
|
## Reprocess in the "in head" insertion mode... |
| 2809 |
|
## As if </head> |
| 2810 |
|
pop @{$self->{open_elements}}; |
| 2811 |
|
|
| 2812 |
|
## Reprocess in the "after head" insertion mode... |
| 2813 |
|
} elsif ($self->{insertion_mode} eq 'in head') { |
| 2814 |
|
## As if </head> |
| 2815 |
|
pop @{$self->{open_elements}}; |
| 2816 |
|
|
| 2817 |
|
## Reprocess in the "after head" insertion mode... |
| 2818 |
|
} |
| 2819 |
|
|
| 2820 |
|
## "after head" insertion mode |
| 2821 |
|
## As if <body> |
| 2822 |
|
!!!insert-element ('body'); |
| 2823 |
|
$self->{insertion_mode} = 'in body'; |
| 2824 |
|
## reprocess |
| 2825 |
|
redo B; |
| 2826 |
|
} elsif ($token->{type} eq 'end tag') { |
| 2827 |
|
if ($token->{tag_name} eq 'head') { |
| 2828 |
|
if ($self->{insertion_mode} eq 'before head') { |
| 2829 |
|
## As if <head> |
| 2830 |
|
!!!create-element ($self->{head_element}, 'head'); |
| 2831 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2832 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2833 |
|
|
| 2834 |
|
## Reprocess in the "in head" insertion mode... |
| 2835 |
|
pop @{$self->{open_elements}}; |
| 2836 |
|
$self->{insertion_mode} = 'after head'; |
| 2837 |
|
!!!next-token; |
| 2838 |
|
redo B; |
| 2839 |
|
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
| 2840 |
|
## As if </noscript> |
| 2841 |
|
pop @{$self->{open_elements}}; |
| 2842 |
|
!!!parse-error (type => 'in noscript:script'); |
| 2843 |
|
|
| 2844 |
|
## Reprocess in the "in head" insertion mode... |
| 2845 |
|
pop @{$self->{open_elements}}; |
| 2846 |
|
$self->{insertion_mode} = 'after head'; |
| 2847 |
|
!!!next-token; |
| 2848 |
|
redo B; |
| 2849 |
|
} elsif ($self->{insertion_mode} eq 'in head') { |
| 2850 |
|
pop @{$self->{open_elements}}; |
| 2851 |
|
$self->{insertion_mode} = 'after head'; |
| 2852 |
|
!!!next-token; |
| 2853 |
|
redo B; |
| 2854 |
|
} else { |
| 2855 |
|
# |
| 2856 |
|
} |
| 2857 |
|
} elsif ($token->{tag_name} eq 'noscript') { |
| 2858 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2859 |
|
pop @{$self->{open_elements}}; |
| 2860 |
|
$self->{insertion_mode} = 'in head'; |
| 2861 |
|
!!!next-token; |
| 2862 |
|
redo B; |
| 2863 |
|
} elsif ($self->{insertion_mode} eq 'before head') { |
| 2864 |
|
!!!parse-error (type => 'unmatched end tag:noscript'); |
| 2865 |
|
## Ignore the token ## ISSUE: An issue in the spec. |
| 2866 |
|
!!!next-token; |
| 2867 |
|
redo B; |
| 2868 |
|
} else { |
| 2869 |
|
# |
| 2870 |
|
} |
| 2871 |
|
} elsif ({ |
| 2872 |
|
body => 1, html => 1, |
| 2873 |
|
}->{$token->{tag_name}}) { |
| 2874 |
|
if ($self->{insertion_mode} eq 'before head') { |
| 2875 |
|
## As if <head> |
| 2876 |
|
!!!create-element ($self->{head_element}, 'head'); |
| 2877 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2878 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2879 |
|
|
| 2880 |
|
$self->{insertion_mode} = 'in head'; |
| 2881 |
|
## Reprocess in the "in head" insertion mode... |
| 2882 |
|
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
| 2883 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2884 |
|
## Ignore the token |
| 2885 |
|
!!!next-token; |
| 2886 |
|
redo B; |
| 2887 |
|
} |
| 2888 |
|
|
| 2889 |
|
# |
| 2890 |
|
} elsif ({ |
| 2891 |
|
p => 1, br => 1, |
| 2892 |
|
}->{$token->{tag_name}}) { |
| 2893 |
|
if ($self->{insertion_mode} eq 'before head') { |
| 2894 |
|
## As if <head> |
| 2895 |
|
!!!create-element ($self->{head_element}, 'head'); |
| 2896 |
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
| 2897 |
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2898 |
|
|
| 2899 |
|
$self->{insertion_mode} = 'in head'; |
| 2900 |
|
## Reprocess in the "in head" insertion mode... |
| 2901 |
|
} |
| 2902 |
|
|
| 2903 |
|
# |
| 2904 |
|
} else { |
| 2905 |
|
if ($self->{insertion_mode} ne 'after head') { |
| 2906 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2907 |
|
## Ignore the token |
| 2908 |
|
!!!next-token; |
| 2909 |
|
redo B; |
| 2910 |
|
} else { |
| 2911 |
|
# |
| 2912 |
|
} |
| 2913 |
|
} |
| 2914 |
|
|
| 2915 |
|
if ($self->{insertion_mode} eq 'in head noscript') { |
| 2916 |
|
## As if </noscript> |
| 2917 |
|
pop @{$self->{open_elements}}; |
| 2918 |
|
!!!parse-error (type => 'in noscript:/'.$token->{tag_name}); |
| 2919 |
|
|
| 2920 |
|
## Reprocess in the "in head" insertion mode... |
| 2921 |
|
## As if </head> |
| 2922 |
|
pop @{$self->{open_elements}}; |
| 2923 |
|
|
| 2924 |
|
## Reprocess in the "after head" insertion mode... |
| 2925 |
|
} elsif ($self->{insertion_mode} eq 'in head') { |
| 2926 |
|
## As if </head> |
| 2927 |
|
pop @{$self->{open_elements}}; |
| 2928 |
|
|
| 2929 |
|
## Reprocess in the "after head" insertion mode... |
| 2930 |
|
} elsif ($self->{insertion_mode} eq 'before head') { |
| 2931 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2932 |
|
## Ignore the token ## ISSUE: An issue in the spec. |
| 2933 |
|
!!!next-token; |
| 2934 |
|
redo B; |
| 2935 |
|
} |
| 2936 |
|
|
| 2937 |
|
## "after head" insertion mode |
| 2938 |
|
## As if <body> |
| 2939 |
|
!!!insert-element ('body'); |
| 2940 |
|
$self->{insertion_mode} = 'in body'; |
| 2941 |
|
## reprocess |
| 2942 |
|
redo B; |
| 2943 |
|
} else { |
| 2944 |
|
die "$0: $token->{type}: Unknown token type"; |
| 2945 |
|
} |
| 2946 |
|
|
| 2947 |
|
## ISSUE: An issue in the spec. |
| 2948 |
|
} elsif ($self->{insertion_mode} eq 'in body' or |
| 2949 |
|
$self->{insertion_mode} eq 'in cell' or |
| 2950 |
|
$self->{insertion_mode} eq 'in caption') { |
| 2951 |
|
if ($token->{type} eq 'character') { |
| 2952 |
|
## NOTE: There is a code clone of "character in body". |
| 2953 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 2954 |
|
|
| 2955 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 2956 |
|
|
| 2957 |
|
!!!next-token; |
| 2958 |
|
redo B; |
| 2959 |
|
} elsif ($token->{type} eq 'start tag') { |
| 2960 |
|
if ({ |
| 2961 |
|
caption => 1, col => 1, colgroup => 1, tbody => 1, |
| 2962 |
|
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
| 2963 |
|
}->{$token->{tag_name}}) { |
| 2964 |
|
if ($self->{insertion_mode} eq 'in cell') { |
| 2965 |
|
## have an element in table scope |
| 2966 |
|
my $tn; |
| 2967 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2968 |
|
my $node = $self->{open_elements}->[$_]; |
| 2969 |
|
if ($node->[1] eq 'td' or $node->[1] eq 'th') { |
| 2970 |
|
$tn = $node->[1]; |
| 2971 |
|
last INSCOPE; |
| 2972 |
|
} elsif ({ |
| 2973 |
|
table => 1, html => 1, |
| 2974 |
|
}->{$node->[1]}) { |
| 2975 |
|
last INSCOPE; |
| 2976 |
|
} |
| 2977 |
|
} # INSCOPE |
| 2978 |
|
unless (defined $tn) { |
| 2979 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 2980 |
|
## Ignore the token |
| 2981 |
|
!!!next-token; |
| 2982 |
|
redo B; |
| 2983 |
|
} |
| 2984 |
|
|
| 2985 |
|
## Close the cell |
| 2986 |
|
!!!back-token; # <?> |
| 2987 |
|
$token = {type => 'end tag', tag_name => $tn}; |
| 2988 |
|
redo B; |
| 2989 |
|
} elsif ($self->{insertion_mode} eq 'in caption') { |
| 2990 |
|
!!!parse-error (type => 'not closed:caption'); |
| 2991 |
|
|
| 2992 |
|
## As if </caption> |
| 2993 |
|
## have a table element in table scope |
| 2994 |
|
my $i; |
| 2995 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 2996 |
|
my $node = $self->{open_elements}->[$_]; |
| 2997 |
|
if ($node->[1] eq 'caption') { |
| 2998 |
|
$i = $_; |
| 2999 |
|
last INSCOPE; |
| 3000 |
|
} elsif ({ |
| 3001 |
|
table => 1, html => 1, |
| 3002 |
|
}->{$node->[1]}) { |
| 3003 |
|
last INSCOPE; |
| 3004 |
|
} |
| 3005 |
|
} # INSCOPE |
| 3006 |
|
unless (defined $i) { |
| 3007 |
|
!!!parse-error (type => 'unmatched end tag:caption'); |
| 3008 |
|
## Ignore the token |
| 3009 |
|
!!!next-token; |
| 3010 |
|
redo B; |
| 3011 |
|
} |
| 3012 |
|
|
| 3013 |
|
## generate implied end tags |
| 3014 |
|
if ({ |
| 3015 |
|
dd => 1, dt => 1, li => 1, p => 1, |
| 3016 |
|
td => 1, th => 1, tr => 1, |
| 3017 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 3018 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3019 |
|
!!!back-token; # <?> |
| 3020 |
|
$token = {type => 'end tag', tag_name => 'caption'}; |
| 3021 |
|
!!!back-token; |
| 3022 |
|
$token = {type => 'end tag', |
| 3023 |
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3024 |
|
redo B; |
| 3025 |
|
} |
| 3026 |
|
|
| 3027 |
|
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
| 3028 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3029 |
|
} |
| 3030 |
|
|
| 3031 |
|
splice @{$self->{open_elements}}, $i; |
| 3032 |
|
|
| 3033 |
|
$clear_up_to_marker->(); |
| 3034 |
|
|
| 3035 |
|
$self->{insertion_mode} = 'in table'; |
| 3036 |
|
|
| 3037 |
|
## reprocess |
| 3038 |
|
redo B; |
| 3039 |
|
} else { |
| 3040 |
|
# |
| 3041 |
|
} |
| 3042 |
|
} else { |
| 3043 |
|
# |
| 3044 |
|
} |
| 3045 |
|
} elsif ($token->{type} eq 'end tag') { |
| 3046 |
|
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
| 3047 |
|
if ($self->{insertion_mode} eq 'in cell') { |
| 3048 |
|
## have an element in table scope |
| 3049 |
|
my $i; |
| 3050 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3051 |
|
my $node = $self->{open_elements}->[$_]; |
| 3052 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3053 |
|
$i = $_; |
| 3054 |
|
last INSCOPE; |
| 3055 |
|
} elsif ({ |
| 3056 |
|
table => 1, html => 1, |
| 3057 |
|
}->{$node->[1]}) { |
| 3058 |
|
last INSCOPE; |
| 3059 |
|
} |
| 3060 |
|
} # INSCOPE |
| 3061 |
|
unless (defined $i) { |
| 3062 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3063 |
|
## Ignore the token |
| 3064 |
|
!!!next-token; |
| 3065 |
|
redo B; |
| 3066 |
|
} |
| 3067 |
|
|
| 3068 |
|
## generate implied end tags |
| 3069 |
|
if ({ |
| 3070 |
|
dd => 1, dt => 1, li => 1, p => 1, |
| 3071 |
|
td => ($token->{tag_name} eq 'th'), |
| 3072 |
|
th => ($token->{tag_name} eq 'td'), |
| 3073 |
|
tr => 1, |
| 3074 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 3075 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3076 |
|
!!!back-token; |
| 3077 |
|
$token = {type => 'end tag', |
| 3078 |
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3079 |
|
redo B; |
| 3080 |
|
} |
| 3081 |
|
|
| 3082 |
|
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
| 3083 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3084 |
|
} |
| 3085 |
|
|
| 3086 |
|
splice @{$self->{open_elements}}, $i; |
| 3087 |
|
|
| 3088 |
|
$clear_up_to_marker->(); |
| 3089 |
|
|
| 3090 |
|
$self->{insertion_mode} = 'in row'; |
| 3091 |
|
|
| 3092 |
|
!!!next-token; |
| 3093 |
|
redo B; |
| 3094 |
|
} elsif ($self->{insertion_mode} eq 'in caption') { |
| 3095 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3096 |
|
## Ignore the token |
| 3097 |
|
!!!next-token; |
| 3098 |
|
redo B; |
| 3099 |
|
} else { |
| 3100 |
|
# |
| 3101 |
|
} |
| 3102 |
|
} elsif ($token->{tag_name} eq 'caption') { |
| 3103 |
|
if ($self->{insertion_mode} eq 'in caption') { |
| 3104 |
|
## have a table element in table scope |
| 3105 |
|
my $i; |
| 3106 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3107 |
|
my $node = $self->{open_elements}->[$_]; |
| 3108 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3109 |
|
$i = $_; |
| 3110 |
|
last INSCOPE; |
| 3111 |
|
} elsif ({ |
| 3112 |
|
table => 1, html => 1, |
| 3113 |
|
}->{$node->[1]}) { |
| 3114 |
|
last INSCOPE; |
| 3115 |
|
} |
| 3116 |
|
} # INSCOPE |
| 3117 |
|
unless (defined $i) { |
| 3118 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3119 |
|
## Ignore the token |
| 3120 |
|
!!!next-token; |
| 3121 |
|
redo B; |
| 3122 |
|
} |
| 3123 |
|
|
| 3124 |
|
## generate implied end tags |
| 3125 |
|
if ({ |
| 3126 |
|
dd => 1, dt => 1, li => 1, p => 1, |
| 3127 |
|
td => 1, th => 1, tr => 1, |
| 3128 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 3129 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3130 |
|
!!!back-token; |
| 3131 |
|
$token = {type => 'end tag', |
| 3132 |
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3133 |
|
redo B; |
| 3134 |
|
} |
| 3135 |
|
|
| 3136 |
|
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
| 3137 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3138 |
|
} |
| 3139 |
|
|
| 3140 |
|
splice @{$self->{open_elements}}, $i; |
| 3141 |
|
|
| 3142 |
|
$clear_up_to_marker->(); |
| 3143 |
|
|
| 3144 |
|
$self->{insertion_mode} = 'in table'; |
| 3145 |
|
|
| 3146 |
|
!!!next-token; |
| 3147 |
|
redo B; |
| 3148 |
|
} elsif ($self->{insertion_mode} eq 'in cell') { |
| 3149 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3150 |
|
## Ignore the token |
| 3151 |
|
!!!next-token; |
| 3152 |
|
redo B; |
| 3153 |
|
} else { |
| 3154 |
|
# |
| 3155 |
|
} |
| 3156 |
|
} elsif ({ |
| 3157 |
|
table => 1, tbody => 1, tfoot => 1, |
| 3158 |
|
thead => 1, tr => 1, |
| 3159 |
|
}->{$token->{tag_name}} and |
| 3160 |
|
$self->{insertion_mode} eq 'in cell') { |
| 3161 |
|
## have an element in table scope |
| 3162 |
|
my $i; |
| 3163 |
|
my $tn; |
| 3164 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3165 |
|
my $node = $self->{open_elements}->[$_]; |
| 3166 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3167 |
|
$i = $_; |
| 3168 |
|
last INSCOPE; |
| 3169 |
|
} elsif ($node->[1] eq 'td' or $node->[1] eq 'th') { |
| 3170 |
|
$tn = $node->[1]; |
| 3171 |
|
## NOTE: There is exactly one |td| or |th| element |
| 3172 |
|
## in scope in the stack of open elements by definition. |
| 3173 |
|
} elsif ({ |
| 3174 |
|
table => 1, html => 1, |
| 3175 |
|
}->{$node->[1]}) { |
| 3176 |
|
last INSCOPE; |
| 3177 |
|
} |
| 3178 |
|
} # INSCOPE |
| 3179 |
|
unless (defined $i) { |
| 3180 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3181 |
|
## Ignore the token |
| 3182 |
|
!!!next-token; |
| 3183 |
|
redo B; |
| 3184 |
|
} |
| 3185 |
|
|
| 3186 |
|
## Close the cell |
| 3187 |
|
!!!back-token; # </?> |
| 3188 |
|
$token = {type => 'end tag', tag_name => $tn}; |
| 3189 |
|
redo B; |
| 3190 |
|
} elsif ($token->{tag_name} eq 'table' and |
| 3191 |
|
$self->{insertion_mode} eq 'in caption') { |
| 3192 |
|
!!!parse-error (type => 'not closed:caption'); |
| 3193 |
|
|
| 3194 |
|
## As if </caption> |
| 3195 |
|
## have a table element in table scope |
| 3196 |
|
my $i; |
| 3197 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3198 |
|
my $node = $self->{open_elements}->[$_]; |
| 3199 |
|
if ($node->[1] eq 'caption') { |
| 3200 |
|
$i = $_; |
| 3201 |
|
last INSCOPE; |
| 3202 |
|
} elsif ({ |
| 3203 |
|
table => 1, html => 1, |
| 3204 |
|
}->{$node->[1]}) { |
| 3205 |
|
last INSCOPE; |
| 3206 |
|
} |
| 3207 |
|
} # INSCOPE |
| 3208 |
|
unless (defined $i) { |
| 3209 |
|
!!!parse-error (type => 'unmatched end tag:caption'); |
| 3210 |
|
## Ignore the token |
| 3211 |
|
!!!next-token; |
| 3212 |
|
redo B; |
| 3213 |
|
} |
| 3214 |
|
|
| 3215 |
|
## generate implied end tags |
| 3216 |
|
if ({ |
| 3217 |
|
dd => 1, dt => 1, li => 1, p => 1, |
| 3218 |
|
td => 1, th => 1, tr => 1, |
| 3219 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 3220 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3221 |
|
!!!back-token; # </table> |
| 3222 |
|
$token = {type => 'end tag', tag_name => 'caption'}; |
| 3223 |
|
!!!back-token; |
| 3224 |
|
$token = {type => 'end tag', |
| 3225 |
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3226 |
|
redo B; |
| 3227 |
|
} |
| 3228 |
|
|
| 3229 |
|
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
| 3230 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3231 |
|
} |
| 3232 |
|
|
| 3233 |
|
splice @{$self->{open_elements}}, $i; |
| 3234 |
|
|
| 3235 |
|
$clear_up_to_marker->(); |
| 3236 |
|
|
| 3237 |
|
$self->{insertion_mode} = 'in table'; |
| 3238 |
|
|
| 3239 |
|
## reprocess |
| 3240 |
|
redo B; |
| 3241 |
|
} elsif ({ |
| 3242 |
|
body => 1, col => 1, colgroup => 1, html => 1, |
| 3243 |
|
}->{$token->{tag_name}}) { |
| 3244 |
|
if ($self->{insertion_mode} eq 'in cell' or |
| 3245 |
|
$self->{insertion_mode} eq 'in caption') { |
| 3246 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3247 |
|
## Ignore the token |
| 3248 |
|
!!!next-token; |
| 3249 |
|
redo B; |
| 3250 |
|
} else { |
| 3251 |
|
# |
| 3252 |
|
} |
| 3253 |
|
} elsif ({ |
| 3254 |
|
tbody => 1, tfoot => 1, |
| 3255 |
|
thead => 1, tr => 1, |
| 3256 |
|
}->{$token->{tag_name}} and |
| 3257 |
|
$self->{insertion_mode} eq 'in caption') { |
| 3258 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3259 |
|
## Ignore the token |
| 3260 |
|
!!!next-token; |
| 3261 |
|
redo B; |
| 3262 |
|
} else { |
| 3263 |
|
# |
| 3264 |
|
} |
| 3265 |
|
} else { |
| 3266 |
|
die "$0: $token->{type}: Unknown token type"; |
| 3267 |
|
} |
| 3268 |
|
|
| 3269 |
|
$insert = $insert_to_current; |
| 3270 |
|
# |
| 3271 |
|
} elsif ($self->{insertion_mode} eq 'in row' or |
| 3272 |
|
$self->{insertion_mode} eq 'in table body' or |
| 3273 |
|
$self->{insertion_mode} eq 'in table') { |
| 3274 |
|
if ($token->{type} eq 'character') { |
| 3275 |
|
## NOTE: There are "character in table" code clones. |
| 3276 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3277 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3278 |
|
|
| 3279 |
|
unless (length $token->{data}) { |
| 3280 |
|
!!!next-token; |
| 3281 |
|
redo B; |
| 3282 |
|
} |
| 3283 |
|
} |
| 3284 |
|
|
| 3285 |
|
!!!parse-error (type => 'in table:#character'); |
| 3286 |
|
|
| 3287 |
|
## As if in body, but insert into foster parent element |
| 3288 |
|
## ISSUE: Spec says that "whenever a node would be inserted |
| 3289 |
|
## into the current node" while characters might not be |
| 3290 |
|
## result in a new Text node. |
| 3291 |
|
$reconstruct_active_formatting_elements->($insert_to_foster); |
| 3292 |
|
|
| 3293 |
|
if ({ |
| 3294 |
|
table => 1, tbody => 1, tfoot => 1, |
| 3295 |
|
thead => 1, tr => 1, |
| 3296 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3297 |
|
# MUST |
| 3298 |
|
my $foster_parent_element; |
| 3299 |
|
my $next_sibling; |
| 3300 |
|
my $prev_sibling; |
| 3301 |
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3302 |
|
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
| 3303 |
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
| 3304 |
|
if (defined $parent and $parent->node_type == 1) { |
| 3305 |
|
$foster_parent_element = $parent; |
| 3306 |
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
| 3307 |
|
$prev_sibling = $next_sibling->previous_sibling; |
| 3308 |
|
} else { |
| 3309 |
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
| 3310 |
|
$prev_sibling = $foster_parent_element->last_child; |
| 3311 |
|
} |
| 3312 |
|
last OE; |
| 3313 |
|
} |
| 3314 |
|
} # OE |
| 3315 |
|
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
| 3316 |
|
$prev_sibling = $foster_parent_element->last_child |
| 3317 |
|
unless defined $foster_parent_element; |
| 3318 |
|
if (defined $prev_sibling and |
| 3319 |
|
$prev_sibling->node_type == 3) { |
| 3320 |
|
$prev_sibling->manakai_append_text ($token->{data}); |
| 3321 |
|
} else { |
| 3322 |
|
$foster_parent_element->insert_before |
| 3323 |
|
($self->{document}->create_text_node ($token->{data}), |
| 3324 |
|
$next_sibling); |
| 3325 |
|
} |
| 3326 |
|
} else { |
| 3327 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3328 |
|
} |
| 3329 |
|
|
| 3330 |
|
!!!next-token; |
| 3331 |
|
redo B; |
| 3332 |
|
} elsif ($token->{type} eq 'start tag') { |
| 3333 |
|
if ({ |
| 3334 |
|
tr => ($self->{insertion_mode} ne 'in row'), |
| 3335 |
|
th => 1, td => 1, |
| 3336 |
|
}->{$token->{tag_name}}) { |
| 3337 |
|
if ($self->{insertion_mode} eq 'in table') { |
| 3338 |
|
## Clear back to table context |
| 3339 |
|
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
| 3340 |
|
$self->{open_elements}->[-1]->[1] ne 'html') { |
| 3341 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3342 |
|
pop @{$self->{open_elements}}; |
| 3343 |
|
} |
| 3344 |
|
|
| 3345 |
|
!!!insert-element ('tbody'); |
| 3346 |
|
$self->{insertion_mode} = 'in table body'; |
| 3347 |
|
## reprocess in the "in table body" insertion mode... |
| 3348 |
|
} |
| 3349 |
|
|
| 3350 |
|
if ($self->{insertion_mode} eq 'in table body') { |
| 3351 |
|
unless ($token->{tag_name} eq 'tr') { |
| 3352 |
|
!!!parse-error (type => 'missing start tag:tr'); |
| 3353 |
|
} |
| 3354 |
|
|
| 3355 |
|
## Clear back to table body context |
| 3356 |
|
while (not { |
| 3357 |
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3358 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3359 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3360 |
|
pop @{$self->{open_elements}}; |
| 3361 |
|
} |
| 3362 |
|
|
| 3363 |
|
$self->{insertion_mode} = 'in row'; |
| 3364 |
|
if ($token->{tag_name} eq 'tr') { |
| 3365 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3366 |
|
!!!next-token; |
| 3367 |
|
redo B; |
| 3368 |
|
} else { |
| 3369 |
|
!!!insert-element ('tr'); |
| 3370 |
|
## reprocess in the "in row" insertion mode |
| 3371 |
|
} |
| 3372 |
|
} |
| 3373 |
|
|
| 3374 |
|
## Clear back to table row context |
| 3375 |
|
while (not { |
| 3376 |
|
tr => 1, html => 1, |
| 3377 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3378 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3379 |
|
pop @{$self->{open_elements}}; |
| 3380 |
|
} |
| 3381 |
|
|
| 3382 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3383 |
|
$self->{insertion_mode} = 'in cell'; |
| 3384 |
|
|
| 3385 |
|
push @$active_formatting_elements, ['#marker', '']; |
| 3386 |
|
|
| 3387 |
|
!!!next-token; |
| 3388 |
|
redo B; |
| 3389 |
|
} elsif ({ |
| 3390 |
|
caption => 1, col => 1, colgroup => 1, |
| 3391 |
|
tbody => 1, tfoot => 1, thead => 1, |
| 3392 |
|
tr => 1, # $self->{insertion_mode} eq 'in row' |
| 3393 |
|
}->{$token->{tag_name}}) { |
| 3394 |
|
if ($self->{insertion_mode} eq 'in row') { |
| 3395 |
|
## As if </tr> |
| 3396 |
|
## have an element in table scope |
| 3397 |
|
my $i; |
| 3398 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3399 |
|
my $node = $self->{open_elements}->[$_]; |
| 3400 |
|
if ($node->[1] eq 'tr') { |
| 3401 |
|
$i = $_; |
| 3402 |
|
last INSCOPE; |
| 3403 |
|
} elsif ({ |
| 3404 |
|
table => 1, html => 1, |
| 3405 |
|
}->{$node->[1]}) { |
| 3406 |
|
last INSCOPE; |
| 3407 |
|
} |
| 3408 |
|
} # INSCOPE |
| 3409 |
|
unless (defined $i) { |
| 3410 |
|
!!!parse-error (type => 'unmacthed end tag:'.$token->{tag_name}); |
| 3411 |
|
## Ignore the token |
| 3412 |
|
!!!next-token; |
| 3413 |
|
redo B; |
| 3414 |
|
} |
| 3415 |
|
|
| 3416 |
|
## Clear back to table row context |
| 3417 |
|
while (not { |
| 3418 |
|
tr => 1, html => 1, |
| 3419 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3420 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3421 |
|
pop @{$self->{open_elements}}; |
| 3422 |
|
} |
| 3423 |
|
|
| 3424 |
|
pop @{$self->{open_elements}}; # tr |
| 3425 |
|
$self->{insertion_mode} = 'in table body'; |
| 3426 |
|
if ($token->{tag_name} eq 'tr') { |
| 3427 |
|
## reprocess |
| 3428 |
|
redo B; |
| 3429 |
|
} else { |
| 3430 |
|
## reprocess in the "in table body" insertion mode... |
| 3431 |
|
} |
| 3432 |
|
} |
| 3433 |
|
|
| 3434 |
|
if ($self->{insertion_mode} eq 'in table body') { |
| 3435 |
|
## have an element in table scope |
| 3436 |
|
my $i; |
| 3437 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3438 |
|
my $node = $self->{open_elements}->[$_]; |
| 3439 |
|
if ({ |
| 3440 |
|
tbody => 1, thead => 1, tfoot => 1, |
| 3441 |
|
}->{$node->[1]}) { |
| 3442 |
|
$i = $_; |
| 3443 |
|
last INSCOPE; |
| 3444 |
|
} elsif ({ |
| 3445 |
|
table => 1, html => 1, |
| 3446 |
|
}->{$node->[1]}) { |
| 3447 |
|
last INSCOPE; |
| 3448 |
|
} |
| 3449 |
|
} # INSCOPE |
| 3450 |
|
unless (defined $i) { |
| 3451 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3452 |
|
## Ignore the token |
| 3453 |
|
!!!next-token; |
| 3454 |
|
redo B; |
| 3455 |
|
} |
| 3456 |
|
|
| 3457 |
|
## Clear back to table body context |
| 3458 |
|
while (not { |
| 3459 |
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3460 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3461 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3462 |
|
pop @{$self->{open_elements}}; |
| 3463 |
|
} |
| 3464 |
|
|
| 3465 |
|
## As if <{current node}> |
| 3466 |
|
## have an element in table scope |
| 3467 |
|
## true by definition |
| 3468 |
|
|
| 3469 |
|
## Clear back to table body context |
| 3470 |
|
## nop by definition |
| 3471 |
|
|
| 3472 |
|
pop @{$self->{open_elements}}; |
| 3473 |
|
$self->{insertion_mode} = 'in table'; |
| 3474 |
|
## reprocess in "in table" insertion mode... |
| 3475 |
|
} |
| 3476 |
|
|
| 3477 |
|
if ($token->{tag_name} eq 'col') { |
| 3478 |
|
## Clear back to table context |
| 3479 |
|
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
| 3480 |
|
$self->{open_elements}->[-1]->[1] ne 'html') { |
| 3481 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3482 |
|
pop @{$self->{open_elements}}; |
| 3483 |
|
} |
| 3484 |
|
|
| 3485 |
|
!!!insert-element ('colgroup'); |
| 3486 |
|
$self->{insertion_mode} = 'in column group'; |
| 3487 |
|
## reprocess |
| 3488 |
|
redo B; |
| 3489 |
|
} elsif ({ |
| 3490 |
|
caption => 1, |
| 3491 |
|
colgroup => 1, |
| 3492 |
|
tbody => 1, tfoot => 1, thead => 1, |
| 3493 |
|
}->{$token->{tag_name}}) { |
| 3494 |
|
## Clear back to table context |
| 3495 |
|
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
| 3496 |
|
$self->{open_elements}->[-1]->[1] ne 'html') { |
| 3497 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3498 |
|
pop @{$self->{open_elements}}; |
| 3499 |
|
} |
| 3500 |
|
|
| 3501 |
|
push @$active_formatting_elements, ['#marker', ''] |
| 3502 |
|
if $token->{tag_name} eq 'caption'; |
| 3503 |
|
|
| 3504 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3505 |
|
$self->{insertion_mode} = { |
| 3506 |
|
caption => 'in caption', |
| 3507 |
|
colgroup => 'in column group', |
| 3508 |
|
tbody => 'in table body', |
| 3509 |
|
tfoot => 'in table body', |
| 3510 |
|
thead => 'in table body', |
| 3511 |
|
}->{$token->{tag_name}}; |
| 3512 |
|
!!!next-token; |
| 3513 |
|
redo B; |
| 3514 |
|
} else { |
| 3515 |
|
die "$0: in table: <>: $token->{tag_name}"; |
| 3516 |
|
} |
| 3517 |
|
} elsif ($token->{tag_name} eq 'table') { |
| 3518 |
|
## NOTE: There are code clones for this "table in table" |
| 3519 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3520 |
|
|
| 3521 |
|
## As if </table> |
| 3522 |
|
## have a table element in table scope |
| 3523 |
|
my $i; |
| 3524 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3525 |
|
my $node = $self->{open_elements}->[$_]; |
| 3526 |
|
if ($node->[1] eq 'table') { |
| 3527 |
|
$i = $_; |
| 3528 |
|
last INSCOPE; |
| 3529 |
|
} elsif ({ |
| 3530 |
|
table => 1, html => 1, |
| 3531 |
|
}->{$node->[1]}) { |
| 3532 |
|
last INSCOPE; |
| 3533 |
|
} |
| 3534 |
|
} # INSCOPE |
| 3535 |
|
unless (defined $i) { |
| 3536 |
|
!!!parse-error (type => 'unmatched end tag:table'); |
| 3537 |
|
## Ignore tokens </table><table> |
| 3538 |
|
!!!next-token; |
| 3539 |
|
redo B; |
| 3540 |
|
} |
| 3541 |
|
|
| 3542 |
|
## generate implied end tags |
| 3543 |
|
if ({ |
| 3544 |
|
dd => 1, dt => 1, li => 1, p => 1, |
| 3545 |
|
td => 1, th => 1, tr => 1, |
| 3546 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 3547 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3548 |
|
!!!back-token; # <table> |
| 3549 |
|
$token = {type => 'end tag', tag_name => 'table'}; |
| 3550 |
|
!!!back-token; |
| 3551 |
|
$token = {type => 'end tag', |
| 3552 |
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3553 |
|
redo B; |
| 3554 |
|
} |
| 3555 |
|
|
| 3556 |
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
| 3557 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3558 |
|
} |
| 3559 |
|
|
| 3560 |
|
splice @{$self->{open_elements}}, $i; |
| 3561 |
|
|
| 3562 |
|
$self->_reset_insertion_mode; |
| 3563 |
|
|
| 3564 |
|
## reprocess |
| 3565 |
|
redo B; |
| 3566 |
|
} else { |
| 3567 |
|
# |
| 3568 |
|
} |
| 3569 |
|
} elsif ($token->{type} eq 'end tag') { |
| 3570 |
|
if ($token->{tag_name} eq 'tr' and |
| 3571 |
|
$self->{insertion_mode} eq 'in row') { |
| 3572 |
|
## have an element in table scope |
| 3573 |
|
my $i; |
| 3574 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3575 |
|
my $node = $self->{open_elements}->[$_]; |
| 3576 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3577 |
|
$i = $_; |
| 3578 |
|
last INSCOPE; |
| 3579 |
|
} elsif ({ |
| 3580 |
|
table => 1, html => 1, |
| 3581 |
|
}->{$node->[1]}) { |
| 3582 |
|
last INSCOPE; |
| 3583 |
|
} |
| 3584 |
|
} # INSCOPE |
| 3585 |
|
unless (defined $i) { |
| 3586 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3587 |
|
## Ignore the token |
| 3588 |
|
!!!next-token; |
| 3589 |
|
redo B; |
| 3590 |
|
} |
| 3591 |
|
|
| 3592 |
|
## Clear back to table row context |
| 3593 |
|
while (not { |
| 3594 |
|
tr => 1, html => 1, |
| 3595 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3596 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3597 |
|
pop @{$self->{open_elements}}; |
| 3598 |
|
} |
| 3599 |
|
|
| 3600 |
|
pop @{$self->{open_elements}}; # tr |
| 3601 |
|
$self->{insertion_mode} = 'in table body'; |
| 3602 |
|
!!!next-token; |
| 3603 |
|
redo B; |
| 3604 |
|
} elsif ($token->{tag_name} eq 'table') { |
| 3605 |
|
if ($self->{insertion_mode} eq 'in row') { |
| 3606 |
|
## As if </tr> |
| 3607 |
|
## have an element in table scope |
| 3608 |
|
my $i; |
| 3609 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3610 |
|
my $node = $self->{open_elements}->[$_]; |
| 3611 |
|
if ($node->[1] eq 'tr') { |
| 3612 |
|
$i = $_; |
| 3613 |
|
last INSCOPE; |
| 3614 |
|
} elsif ({ |
| 3615 |
|
table => 1, html => 1, |
| 3616 |
|
}->{$node->[1]}) { |
| 3617 |
|
last INSCOPE; |
| 3618 |
|
} |
| 3619 |
|
} # INSCOPE |
| 3620 |
|
unless (defined $i) { |
| 3621 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{type}); |
| 3622 |
|
## Ignore the token |
| 3623 |
|
!!!next-token; |
| 3624 |
|
redo B; |
| 3625 |
|
} |
| 3626 |
|
|
| 3627 |
|
## Clear back to table row context |
| 3628 |
|
while (not { |
| 3629 |
|
tr => 1, html => 1, |
| 3630 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3631 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3632 |
|
pop @{$self->{open_elements}}; |
| 3633 |
|
} |
| 3634 |
|
|
| 3635 |
|
pop @{$self->{open_elements}}; # tr |
| 3636 |
|
$self->{insertion_mode} = 'in table body'; |
| 3637 |
|
## reprocess in the "in table body" insertion mode... |
| 3638 |
|
} |
| 3639 |
|
|
| 3640 |
|
if ($self->{insertion_mode} eq 'in table body') { |
| 3641 |
|
## have an element in table scope |
| 3642 |
|
my $i; |
| 3643 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3644 |
|
my $node = $self->{open_elements}->[$_]; |
| 3645 |
|
if ({ |
| 3646 |
|
tbody => 1, thead => 1, tfoot => 1, |
| 3647 |
|
}->{$node->[1]}) { |
| 3648 |
|
$i = $_; |
| 3649 |
|
last INSCOPE; |
| 3650 |
|
} elsif ({ |
| 3651 |
|
table => 1, html => 1, |
| 3652 |
|
}->{$node->[1]}) { |
| 3653 |
|
last INSCOPE; |
| 3654 |
|
} |
| 3655 |
|
} # INSCOPE |
| 3656 |
|
unless (defined $i) { |
| 3657 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3658 |
|
## Ignore the token |
| 3659 |
|
!!!next-token; |
| 3660 |
|
redo B; |
| 3661 |
|
} |
| 3662 |
|
|
| 3663 |
|
## Clear back to table body context |
| 3664 |
|
while (not { |
| 3665 |
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3666 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3667 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3668 |
|
pop @{$self->{open_elements}}; |
| 3669 |
|
} |
| 3670 |
|
|
| 3671 |
|
## As if <{current node}> |
| 3672 |
|
## have an element in table scope |
| 3673 |
|
## true by definition |
| 3674 |
|
|
| 3675 |
|
## Clear back to table body context |
| 3676 |
|
## nop by definition |
| 3677 |
|
|
| 3678 |
|
pop @{$self->{open_elements}}; |
| 3679 |
|
$self->{insertion_mode} = 'in table'; |
| 3680 |
|
## reprocess in the "in table" insertion mode... |
| 3681 |
|
} |
| 3682 |
|
|
| 3683 |
|
## have a table element in table scope |
| 3684 |
|
my $i; |
| 3685 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3686 |
|
my $node = $self->{open_elements}->[$_]; |
| 3687 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3688 |
|
$i = $_; |
| 3689 |
|
last INSCOPE; |
| 3690 |
|
} elsif ({ |
| 3691 |
|
table => 1, html => 1, |
| 3692 |
|
}->{$node->[1]}) { |
| 3693 |
|
last INSCOPE; |
| 3694 |
|
} |
| 3695 |
|
} # INSCOPE |
| 3696 |
|
unless (defined $i) { |
| 3697 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3698 |
|
## Ignore the token |
| 3699 |
|
!!!next-token; |
| 3700 |
|
redo B; |
| 3701 |
|
} |
| 3702 |
|
|
| 3703 |
|
## generate implied end tags |
| 3704 |
|
if ({ |
| 3705 |
|
dd => 1, dt => 1, li => 1, p => 1, |
| 3706 |
|
td => 1, th => 1, tr => 1, |
| 3707 |
|
tbody => 1, tfoot=> 1, thead => 1, |
| 3708 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3709 |
|
!!!back-token; |
| 3710 |
|
$token = {type => 'end tag', |
| 3711 |
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
| 3712 |
|
redo B; |
| 3713 |
|
} |
| 3714 |
|
|
| 3715 |
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
| 3716 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3717 |
|
} |
| 3718 |
|
|
| 3719 |
|
splice @{$self->{open_elements}}, $i; |
| 3720 |
|
|
| 3721 |
|
$self->_reset_insertion_mode; |
| 3722 |
|
|
| 3723 |
|
!!!next-token; |
| 3724 |
|
redo B; |
| 3725 |
|
} elsif ({ |
| 3726 |
|
tbody => 1, tfoot => 1, thead => 1, |
| 3727 |
|
}->{$token->{tag_name}} and |
| 3728 |
|
($self->{insertion_mode} eq 'in row' or |
| 3729 |
|
$self->{insertion_mode} eq 'in table body')) { |
| 3730 |
|
if ($self->{insertion_mode} eq 'in row') { |
| 3731 |
|
## have an element in table scope |
| 3732 |
|
my $i; |
| 3733 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3734 |
|
my $node = $self->{open_elements}->[$_]; |
| 3735 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3736 |
|
$i = $_; |
| 3737 |
|
last INSCOPE; |
| 3738 |
|
} elsif ({ |
| 3739 |
|
table => 1, html => 1, |
| 3740 |
|
}->{$node->[1]}) { |
| 3741 |
|
last INSCOPE; |
| 3742 |
|
} |
| 3743 |
|
} # INSCOPE |
| 3744 |
|
unless (defined $i) { |
| 3745 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3746 |
|
## Ignore the token |
| 3747 |
|
!!!next-token; |
| 3748 |
|
redo B; |
| 3749 |
|
} |
| 3750 |
|
|
| 3751 |
|
## As if </tr> |
| 3752 |
|
## have an element in table scope |
| 3753 |
|
my $i; |
| 3754 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3755 |
|
my $node = $self->{open_elements}->[$_]; |
| 3756 |
|
if ($node->[1] eq 'tr') { |
| 3757 |
|
$i = $_; |
| 3758 |
|
last INSCOPE; |
| 3759 |
|
} elsif ({ |
| 3760 |
|
table => 1, html => 1, |
| 3761 |
|
}->{$node->[1]}) { |
| 3762 |
|
last INSCOPE; |
| 3763 |
|
} |
| 3764 |
|
} # INSCOPE |
| 3765 |
|
unless (defined $i) { |
| 3766 |
|
!!!parse-error (type => 'unmatched end tag:tr'); |
| 3767 |
|
## Ignore the token |
| 3768 |
|
!!!next-token; |
| 3769 |
|
redo B; |
| 3770 |
|
} |
| 3771 |
|
|
| 3772 |
|
## Clear back to table row context |
| 3773 |
|
while (not { |
| 3774 |
|
tr => 1, html => 1, |
| 3775 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3776 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3777 |
|
pop @{$self->{open_elements}}; |
| 3778 |
|
} |
| 3779 |
|
|
| 3780 |
|
pop @{$self->{open_elements}}; # tr |
| 3781 |
|
$self->{insertion_mode} = 'in table body'; |
| 3782 |
|
## reprocess in the "in table body" insertion mode... |
| 3783 |
|
} |
| 3784 |
|
|
| 3785 |
|
## have an element in table scope |
| 3786 |
|
my $i; |
| 3787 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3788 |
|
my $node = $self->{open_elements}->[$_]; |
| 3789 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3790 |
|
$i = $_; |
| 3791 |
|
last INSCOPE; |
| 3792 |
|
} elsif ({ |
| 3793 |
|
table => 1, html => 1, |
| 3794 |
|
}->{$node->[1]}) { |
| 3795 |
|
last INSCOPE; |
| 3796 |
|
} |
| 3797 |
|
} # INSCOPE |
| 3798 |
|
unless (defined $i) { |
| 3799 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3800 |
|
## Ignore the token |
| 3801 |
|
!!!next-token; |
| 3802 |
|
redo B; |
| 3803 |
|
} |
| 3804 |
|
|
| 3805 |
|
## Clear back to table body context |
| 3806 |
|
while (not { |
| 3807 |
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
| 3808 |
|
}->{$self->{open_elements}->[-1]->[1]}) { |
| 3809 |
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3810 |
|
pop @{$self->{open_elements}}; |
| 3811 |
|
} |
| 3812 |
|
|
| 3813 |
|
pop @{$self->{open_elements}}; |
| 3814 |
|
$self->{insertion_mode} = 'in table'; |
| 3815 |
|
!!!next-token; |
| 3816 |
|
redo B; |
| 3817 |
|
} elsif ({ |
| 3818 |
|
body => 1, caption => 1, col => 1, colgroup => 1, |
| 3819 |
|
html => 1, td => 1, th => 1, |
| 3820 |
|
tr => 1, # $self->{insertion_mode} eq 'in row' |
| 3821 |
|
tbody => 1, tfoot => 1, thead => 1, # $self->{insertion_mode} eq 'in table' |
| 3822 |
|
}->{$token->{tag_name}}) { |
| 3823 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3824 |
|
## Ignore the token |
| 3825 |
|
!!!next-token; |
| 3826 |
|
redo B; |
| 3827 |
|
} else { |
| 3828 |
|
# |
| 3829 |
|
} |
| 3830 |
|
} else { |
| 3831 |
|
die "$0: $token->{type}: Unknown token type"; |
| 3832 |
|
} |
| 3833 |
|
|
| 3834 |
|
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
| 3835 |
|
|
| 3836 |
|
$insert = $insert_to_foster; |
| 3837 |
|
# |
| 3838 |
|
} elsif ($self->{insertion_mode} eq 'in column group') { |
| 3839 |
|
if ($token->{type} eq 'character') { |
| 3840 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3841 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3842 |
|
unless (length $token->{data}) { |
| 3843 |
|
!!!next-token; |
| 3844 |
|
redo B; |
| 3845 |
|
} |
| 3846 |
|
} |
| 3847 |
|
|
| 3848 |
|
# |
| 3849 |
|
} elsif ($token->{type} eq 'start tag') { |
| 3850 |
|
if ($token->{tag_name} eq 'col') { |
| 3851 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3852 |
|
pop @{$self->{open_elements}}; |
| 3853 |
|
!!!next-token; |
| 3854 |
|
redo B; |
| 3855 |
|
} else { |
| 3856 |
|
# |
| 3857 |
|
} |
| 3858 |
|
} elsif ($token->{type} eq 'end tag') { |
| 3859 |
|
if ($token->{tag_name} eq 'colgroup') { |
| 3860 |
|
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
| 3861 |
|
!!!parse-error (type => 'unmatched end tag:colgroup'); |
| 3862 |
|
## Ignore the token |
| 3863 |
|
!!!next-token; |
| 3864 |
|
redo B; |
| 3865 |
|
} else { |
| 3866 |
|
pop @{$self->{open_elements}}; # colgroup |
| 3867 |
|
$self->{insertion_mode} = 'in table'; |
| 3868 |
|
!!!next-token; |
| 3869 |
|
redo B; |
| 3870 |
|
} |
| 3871 |
|
} elsif ($token->{tag_name} eq 'col') { |
| 3872 |
|
!!!parse-error (type => 'unmatched end tag:col'); |
| 3873 |
|
## Ignore the token |
| 3874 |
|
!!!next-token; |
| 3875 |
|
redo B; |
| 3876 |
|
} else { |
| 3877 |
|
# |
| 3878 |
|
} |
| 3879 |
|
} else { |
| 3880 |
|
# |
| 3881 |
|
} |
| 3882 |
|
|
| 3883 |
|
## As if </colgroup> |
| 3884 |
|
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
| 3885 |
|
!!!parse-error (type => 'unmatched end tag:colgroup'); |
| 3886 |
|
## Ignore the token |
| 3887 |
|
!!!next-token; |
| 3888 |
|
redo B; |
| 3889 |
|
} else { |
| 3890 |
|
pop @{$self->{open_elements}}; # colgroup |
| 3891 |
|
$self->{insertion_mode} = 'in table'; |
| 3892 |
|
## reprocess |
| 3893 |
|
redo B; |
| 3894 |
|
} |
| 3895 |
|
} elsif ($self->{insertion_mode} eq 'in select') { |
| 3896 |
|
if ($token->{type} eq 'character') { |
| 3897 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 3898 |
|
!!!next-token; |
| 3899 |
|
redo B; |
| 3900 |
|
} elsif ($token->{type} eq 'start tag') { |
| 3901 |
|
if ($token->{tag_name} eq 'option') { |
| 3902 |
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 3903 |
|
## As if </option> |
| 3904 |
|
pop @{$self->{open_elements}}; |
| 3905 |
|
} |
| 3906 |
|
|
| 3907 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3908 |
|
!!!next-token; |
| 3909 |
|
redo B; |
| 3910 |
|
} elsif ($token->{tag_name} eq 'optgroup') { |
| 3911 |
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 3912 |
|
## As if </option> |
| 3913 |
|
pop @{$self->{open_elements}}; |
| 3914 |
|
} |
| 3915 |
|
|
| 3916 |
|
if ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
| 3917 |
|
## As if </optgroup> |
| 3918 |
|
pop @{$self->{open_elements}}; |
| 3919 |
|
} |
| 3920 |
|
|
| 3921 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 3922 |
|
!!!next-token; |
| 3923 |
|
redo B; |
| 3924 |
|
} elsif ($token->{tag_name} eq 'select') { |
| 3925 |
|
!!!parse-error (type => 'not closed:select'); |
| 3926 |
|
## As if </select> instead |
| 3927 |
|
## have an element in table scope |
| 3928 |
|
my $i; |
| 3929 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3930 |
|
my $node = $self->{open_elements}->[$_]; |
| 3931 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3932 |
|
$i = $_; |
| 3933 |
|
last INSCOPE; |
| 3934 |
|
} elsif ({ |
| 3935 |
|
table => 1, html => 1, |
| 3936 |
|
}->{$node->[1]}) { |
| 3937 |
|
last INSCOPE; |
| 3938 |
|
} |
| 3939 |
|
} # INSCOPE |
| 3940 |
|
unless (defined $i) { |
| 3941 |
|
!!!parse-error (type => 'unmatched end tag:select'); |
| 3942 |
|
## Ignore the token |
| 3943 |
|
!!!next-token; |
| 3944 |
|
redo B; |
| 3945 |
|
} |
| 3946 |
|
|
| 3947 |
|
splice @{$self->{open_elements}}, $i; |
| 3948 |
|
|
| 3949 |
|
$self->_reset_insertion_mode; |
| 3950 |
|
|
| 3951 |
|
!!!next-token; |
| 3952 |
|
redo B; |
| 3953 |
|
} else { |
| 3954 |
|
# |
| 3955 |
|
} |
| 3956 |
|
} elsif ($token->{type} eq 'end tag') { |
| 3957 |
|
if ($token->{tag_name} eq 'optgroup') { |
| 3958 |
|
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
| 3959 |
|
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
| 3960 |
|
## As if </option> |
| 3961 |
|
splice @{$self->{open_elements}}, -2; |
| 3962 |
|
} elsif ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
| 3963 |
|
pop @{$self->{open_elements}}; |
| 3964 |
|
} else { |
| 3965 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3966 |
|
## Ignore the token |
| 3967 |
|
} |
| 3968 |
|
!!!next-token; |
| 3969 |
|
redo B; |
| 3970 |
|
} elsif ($token->{tag_name} eq 'option') { |
| 3971 |
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 3972 |
|
pop @{$self->{open_elements}}; |
| 3973 |
|
} else { |
| 3974 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3975 |
|
## Ignore the token |
| 3976 |
|
} |
| 3977 |
|
!!!next-token; |
| 3978 |
|
redo B; |
| 3979 |
|
} elsif ($token->{tag_name} eq 'select') { |
| 3980 |
|
## have an element in table scope |
| 3981 |
|
my $i; |
| 3982 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 3983 |
|
my $node = $self->{open_elements}->[$_]; |
| 3984 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 3985 |
|
$i = $_; |
| 3986 |
|
last INSCOPE; |
| 3987 |
|
} elsif ({ |
| 3988 |
|
table => 1, html => 1, |
| 3989 |
|
}->{$node->[1]}) { |
| 3990 |
|
last INSCOPE; |
| 3991 |
|
} |
| 3992 |
|
} # INSCOPE |
| 3993 |
|
unless (defined $i) { |
| 3994 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 3995 |
|
## Ignore the token |
| 3996 |
|
!!!next-token; |
| 3997 |
|
redo B; |
| 3998 |
|
} |
| 3999 |
|
|
| 4000 |
|
splice @{$self->{open_elements}}, $i; |
| 4001 |
|
|
| 4002 |
|
$self->_reset_insertion_mode; |
| 4003 |
|
|
| 4004 |
|
!!!next-token; |
| 4005 |
|
redo B; |
| 4006 |
|
} elsif ({ |
| 4007 |
|
caption => 1, table => 1, tbody => 1, |
| 4008 |
|
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
| 4009 |
|
}->{$token->{tag_name}}) { |
| 4010 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4011 |
|
|
| 4012 |
|
## have an element in table scope |
| 4013 |
|
my $i; |
| 4014 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4015 |
|
my $node = $self->{open_elements}->[$_]; |
| 4016 |
|
if ($node->[1] eq $token->{tag_name}) { |
| 4017 |
|
$i = $_; |
| 4018 |
|
last INSCOPE; |
| 4019 |
|
} elsif ({ |
| 4020 |
|
table => 1, html => 1, |
| 4021 |
|
}->{$node->[1]}) { |
| 4022 |
|
last INSCOPE; |
| 4023 |
|
} |
| 4024 |
|
} # INSCOPE |
| 4025 |
|
unless (defined $i) { |
| 4026 |
|
## Ignore the token |
| 4027 |
|
!!!next-token; |
| 4028 |
|
redo B; |
| 4029 |
|
} |
| 4030 |
|
|
| 4031 |
|
## As if </select> |
| 4032 |
|
## have an element in table scope |
| 4033 |
|
undef $i; |
| 4034 |
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4035 |
|
my $node = $self->{open_elements}->[$_]; |
| 4036 |
|
if ($node->[1] eq 'select') { |
| 4037 |
|
$i = $_; |
| 4038 |
|
last INSCOPE; |
| 4039 |
|
} elsif ({ |
| 4040 |
|
table => 1, html => 1, |
| 4041 |
|
}->{$node->[1]}) { |
| 4042 |
|
last INSCOPE; |
| 4043 |
|
} |
| 4044 |
|
} # INSCOPE |
| 4045 |
|
unless (defined $i) { |
| 4046 |
|
!!!parse-error (type => 'unmatched end tag:select'); |
| 4047 |
|
## Ignore the </select> token |
| 4048 |
|
!!!next-token; ## TODO: ok? |
| 4049 |
|
redo B; |
| 4050 |
|
} |
| 4051 |
|
|
| 4052 |
|
splice @{$self->{open_elements}}, $i; |
| 4053 |
|
|
| 4054 |
|
$self->_reset_insertion_mode; |
| 4055 |
|
|
| 4056 |
|
## reprocess |
| 4057 |
|
redo B; |
| 4058 |
|
} else { |
| 4059 |
|
# |
| 4060 |
|
} |
| 4061 |
|
} else { |
| 4062 |
|
# |
| 4063 |
|
} |
| 4064 |
|
|
| 4065 |
|
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
| 4066 |
|
## Ignore the token |
| 4067 |
|
!!!next-token; |
| 4068 |
|
redo B; |
| 4069 |
|
} elsif ($self->{insertion_mode} eq 'after body' or |
| 4070 |
|
$self->{insertion_mode} eq 'after html body') { |
| 4071 |
|
if ($token->{type} eq 'character') { |
| 4072 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4073 |
|
my $data = $1; |
| 4074 |
|
## As if in body |
| 4075 |
|
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4076 |
|
|
| 4077 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 4078 |
|
|
| 4079 |
|
unless (length $token->{data}) { |
| 4080 |
|
!!!next-token; |
| 4081 |
|
redo B; |
| 4082 |
|
} |
| 4083 |
|
} |
| 4084 |
|
|
| 4085 |
|
if ($self->{insertion_mode} eq 'after html body') { |
| 4086 |
|
!!!parse-error (type => 'after html:#character'); |
| 4087 |
|
|
| 4088 |
|
## Reprocess in the "main" phase, "after body" insertion mode... |
| 4089 |
|
} |
| 4090 |
|
|
| 4091 |
|
## "after body" insertion mode |
| 4092 |
|
!!!parse-error (type => 'after body:#character'); |
| 4093 |
|
|
| 4094 |
|
$self->{insertion_mode} = 'in body'; |
| 4095 |
|
## reprocess |
| 4096 |
|
redo B; |
| 4097 |
|
} elsif ($token->{type} eq 'start tag') { |
| 4098 |
|
if ($self->{insertion_mode} eq 'after html body') { |
| 4099 |
|
!!!parse-error (type => 'after html:'.$token->{tag_name}); |
| 4100 |
|
|
| 4101 |
|
## Reprocess in the "main" phase, "after body" insertion mode... |
| 4102 |
|
} |
| 4103 |
|
|
| 4104 |
|
## "after body" insertion mode |
| 4105 |
|
!!!parse-error (type => 'after body:'.$token->{tag_name}); |
| 4106 |
|
|
| 4107 |
|
$self->{insertion_mode} = 'in body'; |
| 4108 |
|
## reprocess |
| 4109 |
|
redo B; |
| 4110 |
|
} elsif ($token->{type} eq 'end tag') { |
| 4111 |
|
if ($self->{insertion_mode} eq 'after html body') { |
| 4112 |
|
!!!parse-error (type => 'after html:/'.$token->{tag_name}); |
| 4113 |
|
|
| 4114 |
|
$self->{insertion_mode} = 'after body'; |
| 4115 |
|
## Reprocess in the "main" phase, "after body" insertion mode... |
| 4116 |
|
} |
| 4117 |
|
|
| 4118 |
|
## "after body" insertion mode |
| 4119 |
|
if ($token->{tag_name} eq 'html') { |
| 4120 |
|
if (defined $self->{inner_html_node}) { |
| 4121 |
|
!!!parse-error (type => 'unmatched end tag:html'); |
| 4122 |
|
## Ignore the token |
| 4123 |
|
!!!next-token; |
| 4124 |
|
redo B; |
| 4125 |
|
} else { |
| 4126 |
|
$self->{insertion_mode} = 'after html body'; |
| 4127 |
|
!!!next-token; |
| 4128 |
|
redo B; |
| 4129 |
|
} |
| 4130 |
|
} else { |
| 4131 |
|
!!!parse-error (type => 'after body:/'.$token->{tag_name}); |
| 4132 |
|
|
| 4133 |
|
$self->{insertion_mode} = 'in body'; |
| 4134 |
|
## reprocess |
| 4135 |
|
redo B; |
| 4136 |
|
} |
| 4137 |
|
} else { |
| 4138 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4139 |
|
} |
| 4140 |
|
} elsif ($self->{insertion_mode} eq 'in frameset' or |
| 4141 |
|
$self->{insertion_mode} eq 'after frameset' or |
| 4142 |
|
$self->{insertion_mode} eq 'after html frameset') { |
| 4143 |
|
if ($token->{type} eq 'character') { |
| 4144 |
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4145 |
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 4146 |
|
|
| 4147 |
|
unless (length $token->{data}) { |
| 4148 |
|
!!!next-token; |
| 4149 |
|
redo B; |
| 4150 |
|
} |
| 4151 |
|
} |
| 4152 |
|
|
| 4153 |
|
if ($token->{data} =~ s/^[^\x09\x0A\x0B\x0C\x20]+//) { |
| 4154 |
|
if ($self->{insertion_mode} eq 'in frameset') { |
| 4155 |
|
!!!parse-error (type => 'in frameset:#character'); |
| 4156 |
|
} elsif ($self->{insertion_mode} eq 'after frameset') { |
| 4157 |
|
!!!parse-error (type => 'after frameset:#character'); |
| 4158 |
|
} else { # "after html frameset" |
| 4159 |
|
!!!parse-error (type => 'after html:#character'); |
| 4160 |
|
|
| 4161 |
|
$self->{insertion_mode} = 'after frameset'; |
| 4162 |
|
## Reprocess in the "main" phase, "after frameset"... |
| 4163 |
|
!!!parse-error (type => 'after frameset:#character'); |
| 4164 |
|
} |
| 4165 |
|
|
| 4166 |
|
## Ignore the token. |
| 4167 |
|
if (length $token->{data}) { |
| 4168 |
|
## reprocess the rest of characters |
| 4169 |
|
} else { |
| 4170 |
|
!!!next-token; |
| 4171 |
|
} |
| 4172 |
|
redo B; |
| 4173 |
|
} |
| 4174 |
|
|
| 4175 |
|
die qq[$0: Character "$token->{data}"]; |
| 4176 |
|
} elsif ($token->{type} eq 'start tag') { |
| 4177 |
|
if ($self->{insertion_mode} eq 'after html frameset') { |
| 4178 |
|
!!!parse-error (type => 'after html:'.$token->{tag_name}); |
| 4179 |
|
|
| 4180 |
|
$self->{insertion_mode} = 'after frameset'; |
| 4181 |
|
## Process in the "main" phase, "after frameset" insertion mode... |
| 4182 |
|
} |
| 4183 |
|
|
| 4184 |
|
if ($token->{tag_name} eq 'frameset' and |
| 4185 |
|
$self->{insertion_mode} eq 'in frameset') { |
| 4186 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4187 |
|
!!!next-token; |
| 4188 |
|
redo B; |
| 4189 |
|
} elsif ($token->{tag_name} eq 'frame' and |
| 4190 |
|
$self->{insertion_mode} eq 'in frameset') { |
| 4191 |
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 4192 |
|
pop @{$self->{open_elements}}; |
| 4193 |
|
!!!next-token; |
| 4194 |
|
redo B; |
| 4195 |
|
} elsif ($token->{tag_name} eq 'noframes') { |
| 4196 |
|
## NOTE: As if in body. |
| 4197 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert_to_current); |
| 4198 |
|
redo B; |
| 4199 |
|
} else { |
| 4200 |
|
if ($self->{insertion_mode} eq 'in frameset') { |
| 4201 |
|
!!!parse-error (type => 'in frameset:'.$token->{tag_name}); |
| 4202 |
|
} else { |
| 4203 |
|
!!!parse-error (type => 'after frameset:'.$token->{tag_name}); |
| 4204 |
|
} |
| 4205 |
|
## Ignore the token |
| 4206 |
|
!!!next-token; |
| 4207 |
|
redo B; |
| 4208 |
|
} |
| 4209 |
|
} elsif ($token->{type} eq 'end tag') { |
| 4210 |
|
if ($self->{insertion_mode} eq 'after html frameset') { |
| 4211 |
|
!!!parse-error (type => 'after html:/'.$token->{tag_name}); |
| 4212 |
|
|
| 4213 |
|
$self->{insertion_mode} = 'after frameset'; |
| 4214 |
|
## Process in the "main" phase, "after frameset" insertion mode... |
| 4215 |
|
} |
| 4216 |
|
|
| 4217 |
|
if ($token->{tag_name} eq 'frameset' and |
| 4218 |
|
$self->{insertion_mode} eq 'in frameset') { |
| 4219 |
|
if ($self->{open_elements}->[-1]->[1] eq 'html' and |
| 4220 |
|
@{$self->{open_elements}} == 1) { |
| 4221 |
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 4222 |
|
## Ignore the token |
| 4223 |
|
!!!next-token; |
| 4224 |
|
} else { |
| 4225 |
|
pop @{$self->{open_elements}}; |
| 4226 |
|
!!!next-token; |
| 4227 |
|
} |
| 4228 |
|
|
| 4229 |
|
if (not defined $self->{inner_html_node} and |
| 4230 |
|
$self->{open_elements}->[-1]->[1] ne 'frameset') { |
| 4231 |
|
$self->{insertion_mode} = 'after frameset'; |
| 4232 |
|
} |
| 4233 |
|
redo B; |
| 4234 |
|
} elsif ($token->{tag_name} eq 'html' and |
| 4235 |
|
$self->{insertion_mode} eq 'after frameset') { |
| 4236 |
|
$self->{insertion_mode} = 'after html frameset'; |
| 4237 |
|
!!!next-token; |
| 4238 |
|
redo B; |
| 4239 |
|
} else { |
| 4240 |
|
if ($self->{insertion_mode} eq 'in frameset') { |
| 4241 |
|
!!!parse-error (type => 'in frameset:/'.$token->{tag_name}); |
| 4242 |
|
} else { |
| 4243 |
|
!!!parse-error (type => 'after frameset:/'.$token->{tag_name}); |
| 4244 |
|
} |
| 4245 |
|
## Ignore the token |
| 4246 |
|
!!!next-token; |
| 4247 |
|
redo B; |
| 4248 |
|
} |
| 4249 |
|
} else { |
| 4250 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4251 |
|
} |
| 4252 |
|
|
| 4253 |
|
## ISSUE: An issue in spec here |
| 4254 |
|
} else { |
| 4255 |
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
| 4256 |
|
} |
| 4257 |
|
|
| 4258 |
|
## "in body" insertion mode |
| 4259 |
my $in_body = sub { |
my $in_body = sub { |
|
my $insert = shift; |
|
| 4260 |
if ($token->{type} eq 'start tag') { |
if ($token->{type} eq 'start tag') { |
| 4261 |
if ($token->{tag_name} eq 'script') { |
if ($token->{tag_name} eq 'script') { |
| 4262 |
## NOTE: This is an "as if in head" code clone |
## NOTE: This is an "as if in head" code clone |
| 4264 |
return; |
return; |
| 4265 |
} elsif ($token->{tag_name} eq 'style') { |
} elsif ($token->{tag_name} eq 'style') { |
| 4266 |
## NOTE: This is an "as if in head" code clone |
## NOTE: This is an "as if in head" code clone |
| 4267 |
$parse_rcdata->('CDATA', $insert); |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
| 4268 |
return; |
return; |
| 4269 |
} elsif ({ |
} elsif ({ |
| 4270 |
base => 1, link => 1, |
base => 1, link => 1, |
| 4301 |
} elsif ($token->{tag_name} eq 'title') { |
} elsif ($token->{tag_name} eq 'title') { |
| 4302 |
!!!parse-error (type => 'in body:title'); |
!!!parse-error (type => 'in body:title'); |
| 4303 |
## NOTE: This is an "as if in head" code clone |
## NOTE: This is an "as if in head" code clone |
| 4304 |
$parse_rcdata->('RCDATA', sub { |
$parse_rcdata->(RCDATA_CONTENT_MODEL, sub { |
| 4305 |
if (defined $self->{head_element}) { |
if (defined $self->{head_element}) { |
| 4306 |
$self->{head_element}->append_child ($_[0]); |
$self->{head_element}->append_child ($_[0]); |
| 4307 |
} else { |
} else { |
| 4497 |
|
|
| 4498 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 4499 |
|
|
| 4500 |
$self->{content_model_flag} = 'PLAINTEXT'; |
$self->{content_model} = PLAINTEXT_CONTENT_MODEL; |
| 4501 |
|
|
| 4502 |
!!!next-token; |
!!!next-token; |
| 4503 |
return; |
return; |
| 4652 |
return; |
return; |
| 4653 |
} elsif ($token->{tag_name} eq 'xmp') { |
} elsif ($token->{tag_name} eq 'xmp') { |
| 4654 |
$reconstruct_active_formatting_elements->($insert_to_current); |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 4655 |
$parse_rcdata->('CDATA', $insert); |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
| 4656 |
return; |
return; |
| 4657 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
| 4658 |
## has a p element in scope |
## has a p element in scope |
| 4768 |
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
!!!create-element ($el, $token->{tag_name}, $token->{attributes}); |
| 4769 |
|
|
| 4770 |
## TODO: $self->{form_element} if defined |
## TODO: $self->{form_element} if defined |
| 4771 |
$self->{content_model_flag} = 'RCDATA'; |
$self->{content_model} = RCDATA_CONTENT_MODEL; |
| 4772 |
delete $self->{escape}; # MUST |
delete $self->{escape}; # MUST |
| 4773 |
|
|
| 4774 |
$insert->($el); |
$insert->($el); |
| 4789 |
$el->manakai_append_text ($text); |
$el->manakai_append_text ($text); |
| 4790 |
} |
} |
| 4791 |
|
|
| 4792 |
$self->{content_model_flag} = 'PCDATA'; |
$self->{content_model} = PCDATA_CONTENT_MODEL; |
| 4793 |
|
|
| 4794 |
if ($token->{type} eq 'end tag' and |
if ($token->{type} eq 'end tag' and |
| 4795 |
$token->{tag_name} eq $tag_name) { |
$token->{tag_name} eq $tag_name) { |
| 4805 |
noframes => 1, |
noframes => 1, |
| 4806 |
noscript => 0, ## TODO: 1 if scripting is enabled |
noscript => 0, ## TODO: 1 if scripting is enabled |
| 4807 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 4808 |
$parse_rcdata->('CDATA', $insert); |
## NOTE: There are two "as if in body" code clones. |
| 4809 |
|
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
| 4810 |
return; |
return; |
| 4811 |
} elsif ($token->{tag_name} eq 'select') { |
} elsif ($token->{tag_name} eq 'select') { |
| 4812 |
$reconstruct_active_formatting_elements->($insert_to_current); |
$reconstruct_active_formatting_elements->($insert_to_current); |
| 5100 |
} |
} |
| 5101 |
} |
} |
| 5102 |
}; # $in_body |
}; # $in_body |
| 5103 |
|
$in_body->($insert_to_current); |
| 5104 |
B: { |
redo B; |
|
if ($self->{insertion_mode} ne 'trailing end') { |
|
|
if ($token->{type} eq 'DOCTYPE') { |
|
|
!!!parse-error (type => 'in html:#DOCTYPE'); |
|
|
## Ignore the token |
|
|
## Stay in the phase |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag' and |
|
|
$token->{tag_name} eq 'html') { |
|
|
## ISSUE: "aa<html>" is not a parse error. |
|
|
## ISSUE: "<html>" in fragment is not a parse error. |
|
|
unless ($token->{first_start_tag}) { |
|
|
!!!parse-error (type => 'not first start tag'); |
|
|
} |
|
|
my $top_el = $self->{open_elements}->[0]->[0]; |
|
|
for my $attr_name (keys %{$token->{attributes}}) { |
|
|
unless ($top_el->has_attribute_ns (undef, $attr_name)) { |
|
|
$top_el->set_attribute_ns |
|
|
(undef, [undef, $attr_name], |
|
|
$token->{attributes}->{$attr_name}->{value}); |
|
|
} |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'end-of-file') { |
|
|
## Generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', tag_name => $self->{open_elements}->[-1]->[1]}; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if (@{$self->{open_elements}} > 2 or |
|
|
(@{$self->{open_elements}} == 2 and $self->{open_elements}->[1]->[1] ne 'body')) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} elsif (defined $self->{inner_html_node} and |
|
|
@{$self->{open_elements}} > 1 and |
|
|
$self->{open_elements}->[1]->[1] ne 'body') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
## Stop parsing |
|
|
last B; |
|
|
|
|
|
## ISSUE: There is an issue in the spec. |
|
|
} else { |
|
|
if ($self->{insertion_mode} eq 'before head') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
## As if <head> |
|
|
!!!create-element ($self->{head_element}, 'head'); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
my $attr = $token->{tag_name} eq 'head' ? $token->{attributes} : {}; |
|
|
!!!create-element ($self->{head_element}, 'head', $attr); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
if ($token->{tag_name} eq 'head') { |
|
|
!!!next-token; |
|
|
#} elsif ({ |
|
|
# base => 1, link => 1, meta => 1, |
|
|
# script => 1, style => 1, title => 1, |
|
|
# }->{$token->{tag_name}}) { |
|
|
# ## reprocess |
|
|
} else { |
|
|
## reprocess |
|
|
} |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ({ |
|
|
head => 1, body => 1, html => 1, |
|
|
p => 1, br => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## As if <head> |
|
|
!!!create-element ($self->{head_element}, 'head'); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($self->{head_element}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token ## ISSUE: An issue in the spec. |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} else { |
|
|
die "$0: $token->{type}: Unknown type"; |
|
|
} |
|
|
} elsif ($self->{insertion_mode} eq 'in head' or |
|
|
$self->{insertion_mode} eq 'in head noscript' or |
|
|
$self->{insertion_mode} eq 'after head') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
# |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ({base => ($self->{insertion_mode} eq 'in head' or |
|
|
$self->{insertion_mode} eq 'after head'), |
|
|
link => 1}->{$token->{tag_name}}) { |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'meta') { |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
|
|
|
|
|
unless ($self->{confident}) { |
|
|
my $charset; |
|
|
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
|
|
$charset = $token->{attributes}->{charset}->{value}; |
|
|
} |
|
|
if ($token->{attributes}->{'http-equiv'}) { |
|
|
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
|
|
if ($token->{attributes}->{'http-equiv'}->{value} |
|
|
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
|
|
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
|
|
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
|
|
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
|
|
} ## TODO: And if supported |
|
|
} |
|
|
## TODO: Change the encoding |
|
|
} |
|
|
|
|
|
## TODO: Extracting |charset| from |meta|. |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'title' and |
|
|
$self->{insertion_mode} eq 'in head') { |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
my $parent = defined $self->{head_element} ? $self->{head_element} |
|
|
: $self->{open_elements}->[-1]->[0]; |
|
|
$parse_rcdata->('RCDATA', sub { $parent->append_child ($_[0]) }); |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'style') { |
|
|
## NOTE: Or (scripting is enabled and tag_name eq 'noscript' and |
|
|
## insertion mode 'in head') |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
$parse_rcdata->('CDATA', $insert_to_current); |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'noscript') { |
|
|
if ($self->{insertion_mode} eq 'in head') { |
|
|
## NOTE: and scripting is disalbed |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in head noscript'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
|
|
!!!parse-error (type => 'in noscript:noscript'); |
|
|
## Ignore the token |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{tag_name} eq 'head' and |
|
|
$self->{insertion_mode} ne 'after head') { |
|
|
!!!parse-error (type => 'in head:head'); # or in head noscript |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} ne 'in head noscript' and |
|
|
$token->{tag_name} eq 'script') { |
|
|
if ($self->{insertion_mode} eq 'after head') { |
|
|
!!!parse-error (type => 'after head:'.$token->{tag_name}); |
|
|
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
|
|
} |
|
|
## NOTE: There is a "as if in head" code clone. |
|
|
$script_start_tag->($insert_to_current); |
|
|
pop @{$self->{open_elements}} |
|
|
if $self->{insertion_mode} eq 'after head'; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'after head' and |
|
|
$token->{tag_name} eq 'body') { |
|
|
!!!insert-element ('body', $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in body'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'after head' and |
|
|
$token->{tag_name} eq 'frameset') { |
|
|
!!!insert-element ('frameset', $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in frameset'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($self->{insertion_mode} eq 'in head' and |
|
|
$token->{tag_name} eq 'head') { |
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'after head'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in head noscript' and |
|
|
$token->{tag_name} eq 'noscript') { |
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in head' and |
|
|
{ |
|
|
body => 1, html => 1, |
|
|
p => 1, br => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
# |
|
|
} elsif ($self->{insertion_mode} eq 'in head noscript' and |
|
|
{ |
|
|
p => 1, br => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
# |
|
|
} elsif ($self->{insertion_mode} ne 'after head') { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
## As if </head> or </noscript> or <body> |
|
|
if ($self->{insertion_mode} eq 'in head') { |
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'after head'; |
|
|
} elsif ($self->{insertion_mode} eq 'in head noscript') { |
|
|
pop @{$self->{open_elements}}; |
|
|
!!!parse-error (type => 'in noscript:'.(defined $token->{tag_name} ? ($token->{type} eq 'end tag' ? '/' : '') . $token->{tag_name} : '#' . $token->{type})); |
|
|
$self->{insertion_mode} = 'in head'; |
|
|
} else { # 'after head' |
|
|
!!!insert-element ('body'); |
|
|
$self->{insertion_mode} = 'in body'; |
|
|
} |
|
|
## reprocess |
|
|
redo B; |
|
|
|
|
|
## ISSUE: An issue in the spec. |
|
|
} elsif ($self->{insertion_mode} eq 'in body') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: There is a code clone of "character in body". |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
## NOTE: There is a code clone of "comment in body". |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
$in_body->($insert_to_current); |
|
|
redo B; |
|
|
} |
|
|
} elsif ($self->{insertion_mode} eq 'in table') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: There are "character in table" code clones. |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:#character'); |
|
|
|
|
|
## 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 ({ |
|
|
table => 1, tbody => 1, tfoot => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
# MUST |
|
|
my $foster_parent_element; |
|
|
my $next_sibling; |
|
|
my $prev_sibling; |
|
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
|
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
|
|
if (defined $parent and $parent->node_type == 1) { |
|
|
$foster_parent_element = $parent; |
|
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
|
|
} else { |
|
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
|
|
$prev_sibling = $foster_parent_element->last_child; |
|
|
} |
|
|
last OE; |
|
|
} |
|
|
} # OE |
|
|
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
|
|
$prev_sibling = $foster_parent_element->last_child |
|
|
unless defined $foster_parent_element; |
|
|
if (defined $prev_sibling and |
|
|
$prev_sibling->node_type == 3) { |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
|
|
} else { |
|
|
$foster_parent_element->insert_before |
|
|
($self->{document}->create_text_node ($token->{data}), |
|
|
$next_sibling); |
|
|
} |
|
|
} else { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
} |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ({ |
|
|
caption => 1, |
|
|
colgroup => 1, |
|
|
tbody => 1, tfoot => 1, thead => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## Clear back to table context |
|
|
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
|
|
$self->{open_elements}->[-1]->[1] ne 'html') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
push @$active_formatting_elements, ['#marker', ''] |
|
|
if $token->{tag_name} eq 'caption'; |
|
|
|
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
$self->{insertion_mode} = { |
|
|
caption => 'in caption', |
|
|
colgroup => 'in column group', |
|
|
tbody => 'in table body', |
|
|
tfoot => 'in table body', |
|
|
thead => 'in table body', |
|
|
}->{$token->{tag_name}}; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
col => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## Clear back to table context |
|
|
while ($self->{open_elements}->[-1]->[1] ne 'table' and |
|
|
$self->{open_elements}->[-1]->[1] ne 'html') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
!!!insert-element ($token->{tag_name} eq 'col' ? 'colgroup' : 'tbody'); |
|
|
$self->{insertion_mode} = $token->{tag_name} eq 'col' |
|
|
? 'in column group' : 'in table body'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## NOTE: There are code clones for this "table in table" |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
|
|
|
## As if </table> |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'table') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:table'); |
|
|
## Ignore tokens </table><table> |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <table> |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'table') { |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, colgroup => 1, |
|
|
html => 1, tbody => 1, td => 1, tfoot => 1, th => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
|
|
$in_body->($insert_to_foster); |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in caption') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: This is a code clone of "character in body". |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
## NOTE: This is a code clone of "comment in body". |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ({ |
|
|
caption => 1, col => 1, colgroup => 1, tbody => 1, |
|
|
td => 1, tfoot => 1, th => 1, thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'not closed:caption'); |
|
|
|
|
|
## As if </caption> |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'caption') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:caption'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <?> |
|
|
$token = {type => 'end tag', tag_name => 'caption'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$clear_up_to_marker->(); |
|
|
|
|
|
$self->{insertion_mode} = 'in table'; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'caption') { |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$clear_up_to_marker->(); |
|
|
|
|
|
$self->{insertion_mode} = 'in table'; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
!!!parse-error (type => 'not closed:caption'); |
|
|
|
|
|
## As if </caption> |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'caption') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:caption'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # </table> |
|
|
$token = {type => 'end tag', tag_name => 'caption'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'caption') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$clear_up_to_marker->(); |
|
|
|
|
|
$self->{insertion_mode} = 'in table'; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, col => 1, colgroup => 1, |
|
|
html => 1, tbody => 1, td => 1, tfoot => 1, |
|
|
th => 1, thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
$in_body->($insert_to_current); |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in column group') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
# |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ($token->{tag_name} eq 'col') { |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'colgroup') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
|
|
!!!parse-error (type => 'unmatched end tag:colgroup'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
pop @{$self->{open_elements}}; # colgroup |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} elsif ($token->{tag_name} eq 'col') { |
|
|
!!!parse-error (type => 'unmatched end tag:col'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
## As if </colgroup> |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'html') { |
|
|
!!!parse-error (type => 'unmatched end tag:colgroup'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
pop @{$self->{open_elements}}; # colgroup |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} |
|
|
} elsif ($self->{insertion_mode} eq 'in table body') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: This is a "character in table" code clone. |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:#character'); |
|
|
|
|
|
## 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 ({ |
|
|
table => 1, tbody => 1, tfoot => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
# MUST |
|
|
my $foster_parent_element; |
|
|
my $next_sibling; |
|
|
my $prev_sibling; |
|
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
|
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
|
|
if (defined $parent and $parent->node_type == 1) { |
|
|
$foster_parent_element = $parent; |
|
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
|
|
} else { |
|
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
|
|
$prev_sibling = $foster_parent_element->last_child; |
|
|
} |
|
|
last OE; |
|
|
} |
|
|
} # OE |
|
|
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
|
|
$prev_sibling = $foster_parent_element->last_child |
|
|
unless defined $foster_parent_element; |
|
|
if (defined $prev_sibling and |
|
|
$prev_sibling->node_type == 3) { |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
|
|
} else { |
|
|
$foster_parent_element->insert_before |
|
|
($self->{document}->create_text_node ($token->{data}), |
|
|
$next_sibling); |
|
|
} |
|
|
} else { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
} |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
## Copied from 'in table' |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ({ |
|
|
tr => 1, |
|
|
th => 1, td => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
unless ($token->{tag_name} eq 'tr') { |
|
|
!!!parse-error (type => 'missing start tag:tr'); |
|
|
} |
|
|
|
|
|
## Clear back to table body context |
|
|
while (not { |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
$self->{insertion_mode} = 'in row'; |
|
|
if ($token->{tag_name} eq 'tr') { |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
!!!next-token; |
|
|
} else { |
|
|
!!!insert-element ('tr'); |
|
|
## reprocess |
|
|
} |
|
|
redo B; |
|
|
} elsif ({ |
|
|
caption => 1, col => 1, colgroup => 1, |
|
|
tbody => 1, tfoot => 1, thead => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ({ |
|
|
tbody => 1, thead => 1, tfoot => 1, |
|
|
}->{$node->[1]}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table body context |
|
|
while (not { |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
## As if <{current node}> |
|
|
## have an element in table scope |
|
|
## true by definition |
|
|
|
|
|
## Clear back to table body context |
|
|
## nop by definition |
|
|
|
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## NOTE: This is a code clone of "table in table" |
|
|
!!!parse-error (type => 'not closed:table'); |
|
|
|
|
|
## As if </table> |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'table') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:table'); |
|
|
## Ignore tokens </table><table> |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <table> |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ({ |
|
|
tbody => 1, tfoot => 1, thead => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table body context |
|
|
while (not { |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ({ |
|
|
tbody => 1, thead => 1, tfoot => 1, |
|
|
}->{$node->[1]}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table body context |
|
|
while (not { |
|
|
tbody => 1, tfoot => 1, thead => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
## As if <{current node}> |
|
|
## have an element in table scope |
|
|
## true by definition |
|
|
|
|
|
## Clear back to table body context |
|
|
## nop by definition |
|
|
|
|
|
pop @{$self->{open_elements}}; |
|
|
$self->{insertion_mode} = 'in table'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, colgroup => 1, |
|
|
html => 1, td => 1, th => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
## As if in table |
|
|
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
|
|
$in_body->($insert_to_foster); |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in row') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: This is a "character in table" code clone. |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:#character'); |
|
|
|
|
|
## 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 ({ |
|
|
table => 1, tbody => 1, tfoot => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
# MUST |
|
|
my $foster_parent_element; |
|
|
my $next_sibling; |
|
|
my $prev_sibling; |
|
|
OE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
if ($self->{open_elements}->[$_]->[1] eq 'table') { |
|
|
my $parent = $self->{open_elements}->[$_]->[0]->parent_node; |
|
|
if (defined $parent and $parent->node_type == 1) { |
|
|
$foster_parent_element = $parent; |
|
|
$next_sibling = $self->{open_elements}->[$_]->[0]; |
|
|
$prev_sibling = $next_sibling->previous_sibling; |
|
|
} else { |
|
|
$foster_parent_element = $self->{open_elements}->[$_ - 1]->[0]; |
|
|
$prev_sibling = $foster_parent_element->last_child; |
|
|
} |
|
|
last OE; |
|
|
} |
|
|
} # OE |
|
|
$foster_parent_element = $self->{open_elements}->[0]->[0] and |
|
|
$prev_sibling = $foster_parent_element->last_child |
|
|
unless defined $foster_parent_element; |
|
|
if (defined $prev_sibling and |
|
|
$prev_sibling->node_type == 3) { |
|
|
$prev_sibling->manakai_append_text ($token->{data}); |
|
|
} else { |
|
|
$foster_parent_element->insert_before |
|
|
($self->{document}->create_text_node ($token->{data}), |
|
|
$next_sibling); |
|
|
} |
|
|
} else { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
} |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
## Copied from 'in table' |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ($token->{tag_name} eq 'th' or |
|
|
$token->{tag_name} eq 'td') { |
|
|
## Clear back to table row context |
|
|
while (not { |
|
|
tr => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
$self->{insertion_mode} = 'in cell'; |
|
|
|
|
|
push @$active_formatting_elements, ['#marker', '']; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
caption => 1, col => 1, colgroup => 1, |
|
|
tbody => 1, tfoot => 1, thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## As if </tr> |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'tr') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmacthed end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table row context |
|
|
while (not { |
|
|
tr => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
pop @{$self->{open_elements}}; # tr |
|
|
$self->{insertion_mode} = 'in table body'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## NOTE: This is a code clone of "table in table" |
|
|
!!!parse-error (type => 'not closed:table'); |
|
|
|
|
|
## As if </table> |
|
|
## have a table element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'table') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:table'); |
|
|
## Ignore tokens </table><table> |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => 1, th => 1, tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; # <table> |
|
|
$token = {type => 'end tag', tag_name => 'table'}; |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne 'table') { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'tr') { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table row context |
|
|
while (not { |
|
|
tr => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
pop @{$self->{open_elements}}; # tr |
|
|
$self->{insertion_mode} = 'in table body'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'table') { |
|
|
## 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] eq 'tr') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{type}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table row context |
|
|
while (not { |
|
|
tr => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
pop @{$self->{open_elements}}; # tr |
|
|
$self->{insertion_mode} = 'in table body'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ({ |
|
|
tbody => 1, tfoot => 1, thead => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## 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] eq 'tr') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:tr'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Clear back to table row context |
|
|
while (not { |
|
|
tr => 1, html => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
pop @{$self->{open_elements}}; # tr |
|
|
$self->{insertion_mode} = 'in table body'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, |
|
|
colgroup => 1, html => 1, td => 1, th => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
## As if in table |
|
|
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
|
|
$in_body->($insert_to_foster); |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in cell') { |
|
|
if ($token->{type} eq 'character') { |
|
|
## NOTE: This is a code clone of "character in body". |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
## NOTE: This is a code clone of "comment in body". |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ({ |
|
|
caption => 1, col => 1, colgroup => 1, |
|
|
tbody => 1, td => 1, tfoot => 1, th => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## have an element in table scope |
|
|
my $tn; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'td' or $node->[1] eq 'th') { |
|
|
$tn = $node->[1]; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $tn) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Close the cell |
|
|
!!!back-token; # <?> |
|
|
$token = {type => 'end tag', tag_name => $tn}; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'td' or $token->{tag_name} eq 'th') { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## generate implied end tags |
|
|
if ({ |
|
|
dd => 1, dt => 1, li => 1, p => 1, |
|
|
td => ($token->{tag_name} eq 'th'), |
|
|
th => ($token->{tag_name} eq 'td'), |
|
|
tr => 1, |
|
|
tbody => 1, tfoot=> 1, thead => 1, |
|
|
}->{$self->{open_elements}->[-1]->[1]}) { |
|
|
!!!back-token; |
|
|
$token = {type => 'end tag', |
|
|
tag_name => $self->{open_elements}->[-1]->[1]}; # MUST |
|
|
redo B; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
|
|
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$clear_up_to_marker->(); |
|
|
|
|
|
$self->{insertion_mode} = 'in row'; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
body => 1, caption => 1, col => 1, |
|
|
colgroup => 1, html => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
table => 1, tbody => 1, tfoot => 1, |
|
|
thead => 1, tr => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
my $tn; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ($node->[1] eq 'td' or $node->[1] eq 'th') { |
|
|
$tn = $node->[1]; |
|
|
## NOTE: There is exactly one |td| or |th| element |
|
|
## in scope in the stack of open elements by definition. |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## Close the cell |
|
|
!!!back-token; # </?> |
|
|
$token = {type => 'end tag', tag_name => $tn}; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
$in_body->($insert_to_current); |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in select') { |
|
|
if ($token->{type} eq 'character') { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ($token->{tag_name} eq 'option') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
|
|
## As if </option> |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'optgroup') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
|
|
## As if </option> |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
if ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
|
|
## As if </optgroup> |
|
|
pop @{$self->{open_elements}}; |
|
|
} |
|
|
|
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'select') { |
|
|
!!!parse-error (type => 'not closed:select'); |
|
|
## As if </select> instead |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:select'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'optgroup') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
|
|
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
|
|
## As if </option> |
|
|
splice @{$self->{open_elements}}, -2; |
|
|
} elsif ($self->{open_elements}->[-1]->[1] eq 'optgroup') { |
|
|
pop @{$self->{open_elements}}; |
|
|
} else { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'option') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
|
|
pop @{$self->{open_elements}}; |
|
|
} else { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
} |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'select') { |
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ({ |
|
|
caption => 1, table => 1, tbody => 1, |
|
|
tfoot => 1, thead => 1, tr => 1, td => 1, th => 1, |
|
|
}->{$token->{tag_name}}) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
|
|
|
## have an element in table scope |
|
|
my $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq $token->{tag_name}) { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
|
|
|
## As if </select> |
|
|
## have an element in table scope |
|
|
undef $i; |
|
|
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
|
|
my $node = $self->{open_elements}->[$_]; |
|
|
if ($node->[1] eq 'select') { |
|
|
$i = $_; |
|
|
last INSCOPE; |
|
|
} elsif ({ |
|
|
table => 1, html => 1, |
|
|
}->{$node->[1]}) { |
|
|
last INSCOPE; |
|
|
} |
|
|
} # INSCOPE |
|
|
unless (defined $i) { |
|
|
!!!parse-error (type => 'unmatched end tag:select'); |
|
|
## Ignore the </select> token |
|
|
!!!next-token; ## TODO: ok? |
|
|
redo B; |
|
|
} |
|
|
|
|
|
splice @{$self->{open_elements}}, $i; |
|
|
|
|
|
$self->_reset_insertion_mode; |
|
|
|
|
|
## reprocess |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'after body') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
my $data = $1; |
|
|
## As if in body |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
# |
|
|
!!!parse-error (type => 'after body:#'.$token->{type}); |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[0]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
!!!parse-error (type => 'after body:'.$token->{tag_name}); |
|
|
# |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'html') { |
|
|
if (defined $self->{inner_html_node}) { |
|
|
!!!parse-error (type => 'unmatched end tag:html'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
$previous_insertion_mode = $self->{insertion_mode}; |
|
|
$self->{insertion_mode} = 'trailing end'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} else { |
|
|
!!!parse-error (type => 'after body:/'.$token->{tag_name}); |
|
|
} |
|
|
} else { |
|
|
!!!parse-error (type => 'after body:#'.$token->{type}); |
|
|
} |
|
|
|
|
|
$self->{insertion_mode} = 'in body'; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'in frameset') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
# |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ($token->{tag_name} eq 'frameset') { |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'frame') { |
|
|
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
|
|
pop @{$self->{open_elements}}; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{tag_name} eq 'noframes') { |
|
|
$in_body->($insert_to_current); |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'frameset') { |
|
|
if ($self->{open_elements}->[-1]->[1] eq 'html' and |
|
|
@{$self->{open_elements}} == 1) { |
|
|
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
} else { |
|
|
pop @{$self->{open_elements}}; |
|
|
!!!next-token; |
|
|
} |
|
|
|
|
|
## if not inner_html and |
|
|
if ($self->{open_elements}->[-1]->[1] ne 'frameset') { |
|
|
$self->{insertion_mode} = 'after frameset'; |
|
|
} |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
if (defined $token->{tag_name}) { |
|
|
!!!parse-error (type => 'in frameset:'.($token->{type} eq 'end tag' ? '/' : '').$token->{tag_name}); |
|
|
} else { |
|
|
!!!parse-error (type => 'in frameset:#'.$token->{type}); |
|
|
} |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($self->{insertion_mode} eq 'after frameset') { |
|
|
if ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
if ($token->{data} =~ s/^[^\x09\x0A\x0B\x0C\x20]+//) { |
|
|
!!!parse-error (type => 'after frameset:#character'); |
|
|
|
|
|
## Ignore the token. |
|
|
if (length $token->{data}) { |
|
|
## reprocess the rest of characters |
|
|
} else { |
|
|
!!!next-token; |
|
|
} |
|
|
redo B; |
|
|
} |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{open_elements}->[-1]->[0]->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag') { |
|
|
if ($token->{tag_name} eq 'noframes') { |
|
|
$in_body->($insert_to_current); |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} elsif ($token->{type} eq 'end tag') { |
|
|
if ($token->{tag_name} eq 'html') { |
|
|
$previous_insertion_mode = $self->{insertion_mode}; |
|
|
$self->{insertion_mode} = 'trailing end'; |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} else { |
|
|
# |
|
|
} |
|
|
} else { |
|
|
die "$0: $token->{type}: Unknown token type"; |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'after frameset:'.($token->{tag_name} eq 'end tag' ? '/' : '').$token->{tag_name}); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
|
|
|
## ISSUE: An issue in spec there |
|
|
} else { |
|
|
die "$0: $self->{insertion_mode}: Unknown insertion mode"; |
|
|
} |
|
|
} |
|
|
} elsif ($self->{insertion_mode} eq 'trailing end') { |
|
|
## states in the main stage is preserved yet # MUST |
|
|
|
|
|
if ($token->{type} eq 'DOCTYPE') { |
|
|
!!!parse-error (type => 'after html:#DOCTYPE'); |
|
|
## Ignore the token |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'comment') { |
|
|
my $comment = $self->{document}->create_comment ($token->{data}); |
|
|
$self->{document}->append_child ($comment); |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'character') { |
|
|
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
|
|
my $data = $1; |
|
|
## As if in the main phase. |
|
|
## NOTE: The insertion mode in the main phase |
|
|
## just before the phase has been changed to the trailing |
|
|
## end phase is either "after body" or "after frameset". |
|
|
$reconstruct_active_formatting_elements->($insert_to_current); |
|
|
|
|
|
$self->{open_elements}->[-1]->[0]->manakai_append_text ($data); |
|
|
|
|
|
unless (length $token->{data}) { |
|
|
!!!next-token; |
|
|
redo B; |
|
|
} |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'after html:#character'); |
|
|
$self->{insertion_mode} = $previous_insertion_mode; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'start tag' or |
|
|
$token->{type} eq 'end tag') { |
|
|
!!!parse-error (type => 'after html:'.($token->{type} eq 'end tag' ? '/' : '').$token->{tag_name}); |
|
|
$self->{insertion_mode} = $previous_insertion_mode; |
|
|
## reprocess |
|
|
redo B; |
|
|
} elsif ($token->{type} eq 'end-of-file') { |
|
|
## Stop parsing |
|
|
last B; |
|
|
} else { |
|
|
die "$0: $token->{type}: Unknown token"; |
|
|
} |
|
|
} |
|
| 5105 |
} # B |
} # B |
| 5106 |
|
|
| 5107 |
|
## NOTE: The "trailing end" phase in HTML5 is split into |
| 5108 |
|
## two insertion modes: "after html body" and "after html frameset". |
| 5109 |
|
## NOTE: States in the main stage is preserved while |
| 5110 |
|
## the parser stays in the trailing end phase. # MUST |
| 5111 |
|
|
| 5112 |
## Stop parsing # MUST |
## Stop parsing # MUST |
| 5113 |
|
|
| 5114 |
## TODO: script stuffs |
## TODO: script stuffs |
| 5193 |
|
|
| 5194 |
## Step 2 |
## Step 2 |
| 5195 |
my $node_ln = $node->local_name; |
my $node_ln = $node->local_name; |
| 5196 |
$p->{content_model_flag} = { |
$p->{content_model} = { |
| 5197 |
title => 'RCDATA', |
title => RCDATA_CONTENT_MODEL, |
| 5198 |
textarea => 'RCDATA', |
textarea => RCDATA_CONTENT_MODEL, |
| 5199 |
style => 'CDATA', |
style => CDATA_CONTENT_MODEL, |
| 5200 |
script => 'CDATA', |
script => CDATA_CONTENT_MODEL, |
| 5201 |
xmp => 'CDATA', |
xmp => CDATA_CONTENT_MODEL, |
| 5202 |
iframe => 'CDATA', |
iframe => CDATA_CONTENT_MODEL, |
| 5203 |
noembed => 'CDATA', |
noembed => CDATA_CONTENT_MODEL, |
| 5204 |
noframes => 'CDATA', |
noframes => CDATA_CONTENT_MODEL, |
| 5205 |
noscript => 'CDATA', |
noscript => CDATA_CONTENT_MODEL, |
| 5206 |
plaintext => 'PLAINTEXT', |
plaintext => PLAINTEXT_CONTENT_MODEL, |
| 5207 |
}->{$node_ln} || 'PCDATA'; |
}->{$node_ln}; |
| 5208 |
## ISSUE: What is "the name of the element"? local name? |
$p->{content_model} = PCDATA_CONTENT_MODEL |
| 5209 |
|
unless defined $p->{content_model}; |
| 5210 |
|
## ISSUE: What is "the name of the element"? local name? |
| 5211 |
|
|
| 5212 |
$p->{inner_html_node} = [$node, $node_ln]; |
$p->{inner_html_node} = [$node, $node_ln]; |
| 5213 |
|
|