| 1 |
package Whatpm::HTML; |
package Whatpm::HTML; |
| 2 |
use strict; |
use strict; |
| 3 |
our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r}; |
| 4 |
|
use Error qw(:try); |
| 5 |
|
|
| 6 |
## ISSUE: |
## ISSUE: |
| 7 |
## var doc = implementation.createDocument (null, null, null); |
## var doc = implementation.createDocument (null, null, null); |
| 85 |
}; |
}; |
| 86 |
# $phrasing_category: all other elements |
# $phrasing_category: all other elements |
| 87 |
|
|
| 88 |
|
sub parse_byte_string ($$$$;$) { |
| 89 |
|
my $self = ref $_[0] ? shift : shift->new; |
| 90 |
|
my $charset = shift; |
| 91 |
|
my $bytes_s = ref $_[0] ? $_[0] : \($_[0]); |
| 92 |
|
my $s; |
| 93 |
|
|
| 94 |
|
if (defined $charset) { |
| 95 |
|
require Encode; ## TODO: decode(utf8) don't delete BOM |
| 96 |
|
$s = \ (Encode::decode ($charset, $$bytes_s)); |
| 97 |
|
$self->{input_encoding} = lc $charset; ## TODO: normalize name |
| 98 |
|
$self->{confident} = 1; |
| 99 |
|
} else { |
| 100 |
|
## TODO: Implement HTML5 detection algorithm |
| 101 |
|
require Whatpm::Charset::UniversalCharDet; |
| 102 |
|
$charset = Whatpm::Charset::UniversalCharDet->detect_byte_string |
| 103 |
|
(substr ($$bytes_s, 0, 1024)); |
| 104 |
|
$charset ||= 'windows-1252'; |
| 105 |
|
$s = \ (Encode::decode ($charset, $$bytes_s)); |
| 106 |
|
$self->{input_encoding} = $charset; |
| 107 |
|
$self->{confident} = 0; |
| 108 |
|
} |
| 109 |
|
|
| 110 |
|
$self->{change_encoding} = sub { |
| 111 |
|
my $self = shift; |
| 112 |
|
my $charset = lc shift; |
| 113 |
|
## TODO: if $charset is supported |
| 114 |
|
## TODO: normalize charset name |
| 115 |
|
|
| 116 |
|
## "Change the encoding" algorithm: |
| 117 |
|
|
| 118 |
|
## Step 1 |
| 119 |
|
if ($charset eq 'utf-16') { ## ISSUE: UTF-16BE -> UTF-8? UTF-16LE -> UTF-8? |
| 120 |
|
$charset = 'utf-8'; |
| 121 |
|
} |
| 122 |
|
|
| 123 |
|
## Step 2 |
| 124 |
|
if (defined $self->{input_encoding} and |
| 125 |
|
$self->{input_encoding} eq $charset) { |
| 126 |
|
$self->{confident} = 1; |
| 127 |
|
return; |
| 128 |
|
} |
| 129 |
|
|
| 130 |
|
!!!parse-error (type => 'charset label detected:'.$self->{input_encoding}. |
| 131 |
|
':'.$charset, level => 'w'); |
| 132 |
|
|
| 133 |
|
## Step 3 |
| 134 |
|
# if (can) { |
| 135 |
|
## change the encoding on the fly. |
| 136 |
|
#$self->{confident} = 1; |
| 137 |
|
#return; |
| 138 |
|
# } |
| 139 |
|
|
| 140 |
|
## Step 4 |
| 141 |
|
throw Whatpm::HTML::RestartParser (charset => $charset); |
| 142 |
|
}; # $self->{change_encoding} |
| 143 |
|
|
| 144 |
|
my @args = @_; shift @args; # $s |
| 145 |
|
my $return; |
| 146 |
|
try { |
| 147 |
|
$return = $self->parse_char_string ($s, @args); |
| 148 |
|
} catch Whatpm::HTML::RestartParser with { |
| 149 |
|
my $charset = shift->{charset}; |
| 150 |
|
$s = \ (Encode::decode ($charset, $$bytes_s)); |
| 151 |
|
$self->{input_encoding} = $charset; ## TODO: normalize |
| 152 |
|
$self->{confident} = 1; |
| 153 |
|
$return = $self->parse_char_string ($s, @args); |
| 154 |
|
}; |
| 155 |
|
return $return; |
| 156 |
|
} # parse_byte_string |
| 157 |
|
|
| 158 |
|
*parse_char_string = \&parse_string; |
| 159 |
|
|
| 160 |
sub parse_string ($$$;$) { |
sub parse_string ($$$;$) { |
| 161 |
my $self = shift->new; |
my $self = ref $_[0] ? shift : shift->new; |
| 162 |
my $s = \$_[0]; |
my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 163 |
$self->{document} = $_[1]; |
$self->{document} = $_[1]; |
| 164 |
|
@{$self->{document}->child_nodes} = (); |
| 165 |
|
|
| 166 |
## NOTE: |set_inner_html| copies most of this method's code |
## NOTE: |set_inner_html| copies most of this method's code |
| 167 |
|
|
| 168 |
|
$self->{confident} = 1 unless exists $self->{confident}; |
| 169 |
|
$self->{document}->input_encoding ($self->{input_encoding}) |
| 170 |
|
if defined $self->{input_encoding}; |
| 171 |
|
|
| 172 |
my $i = 0; |
my $i = 0; |
| 173 |
my $line = 1; |
my $line = 1; |
| 174 |
my $column = 0; |
my $column = 0; |
| 225 |
$self->{parse_error} = sub { |
$self->{parse_error} = sub { |
| 226 |
# |
# |
| 227 |
}; |
}; |
| 228 |
|
$self->{change_encoding} = sub { |
| 229 |
|
# if ($_[0] is a supported encoding) { |
| 230 |
|
# run "change the encoding" algorithm; |
| 231 |
|
# throw Whatpm::HTML::RestartParser (charset => $new_encoding); |
| 232 |
|
# } |
| 233 |
|
}; |
| 234 |
|
$self->{application_cache_selection} = sub { |
| 235 |
|
# |
| 236 |
|
}; |
| 237 |
return $self; |
return $self; |
| 238 |
} # new |
} # new |
| 239 |
|
|
| 340 |
## ->{system_identifier} (DOCTYPE_TOKEN) |
## ->{system_identifier} (DOCTYPE_TOKEN) |
| 341 |
## ->{correct} == 1 or 0 (DOCTYPE_TOKEN) |
## ->{correct} == 1 or 0 (DOCTYPE_TOKEN) |
| 342 |
## ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN) |
## ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN) |
| 343 |
|
## ->{name} |
| 344 |
|
## ->{value} |
| 345 |
|
## ->{has_reference} == 1 or 0 |
| 346 |
## ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN) |
## ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN) |
| 347 |
|
|
| 348 |
## Emitted token MUST immediately be handled by the tree construction state. |
## Emitted token MUST immediately be handled by the tree construction state. |
| 353 |
## has completed loading. If one has, then it MUST be executed |
## has completed loading. If one has, then it MUST be executed |
| 354 |
## and removed from the list. |
## and removed from the list. |
| 355 |
|
|
| 356 |
|
## NOTE: HTML5 "Writing HTML documents" section, applied to |
| 357 |
|
## documents and not to user agents and conformance checkers, |
| 358 |
|
## contains some requirements that are not detected by the |
| 359 |
|
## parsing algorithm: |
| 360 |
|
## - Some requirements on character encoding declarations. ## TODO |
| 361 |
|
## - "Elements MUST NOT contain content that their content model disallows." |
| 362 |
|
## ... Some are parse error, some are not (will be reported by c.c.). |
| 363 |
|
## - Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) ## TODO |
| 364 |
|
## - Text (in elements, attributes, and comments) SHOULD NOT contain |
| 365 |
|
## control characters other than space characters. ## TODO: (what is control character? C0, C1 and DEL? Unicode control character?) |
| 366 |
|
|
| 367 |
|
## TODO: HTML5 poses authors two SHOULD-level requirements that cannot |
| 368 |
|
## be detected by the HTML5 parsing algorithm: |
| 369 |
|
## - Text, |
| 370 |
|
|
| 371 |
sub _get_next_token ($) { |
sub _get_next_token ($) { |
| 372 |
my $self = shift; |
my $self = shift; |
| 373 |
if (@{$self->{token}}) { |
if (@{$self->{token}}) { |
| 1118 |
$self->{current_attribute}->{value} .= '&'; |
$self->{current_attribute}->{value} .= '&'; |
| 1119 |
} else { |
} else { |
| 1120 |
$self->{current_attribute}->{value} .= $token->{data}; |
$self->{current_attribute}->{value} .= $token->{data}; |
| 1121 |
|
$self->{current_attribute}->{has_reference} = $token->{has_reference}; |
| 1122 |
## ISSUE: spec says "append the returned character token to the current attribute's value" |
## ISSUE: spec says "append the returned character token to the current attribute's value" |
| 1123 |
} |
} |
| 1124 |
|
|
| 1845 |
$code = $c1_entity_char->{$code}; |
$code = $c1_entity_char->{$code}; |
| 1846 |
} |
} |
| 1847 |
|
|
| 1848 |
return {type => CHARACTER_TOKEN, data => chr $code}; |
return {type => CHARACTER_TOKEN, data => chr $code, |
| 1849 |
|
has_reference => 1}; |
| 1850 |
} # X |
} # X |
| 1851 |
} elsif (0x0030 <= $self->{next_input_character} and |
} elsif (0x0030 <= $self->{next_input_character} and |
| 1852 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
| 1881 |
$code = $c1_entity_char->{$code}; |
$code = $c1_entity_char->{$code}; |
| 1882 |
} |
} |
| 1883 |
|
|
| 1884 |
return {type => CHARACTER_TOKEN, data => chr $code}; |
return {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1}; |
| 1885 |
} else { |
} else { |
| 1886 |
!!!parse-error (type => 'bare nero'); |
!!!parse-error (type => 'bare nero'); |
| 1887 |
!!!back-next-input-character ($self->{next_input_character}); |
!!!back-next-input-character ($self->{next_input_character}); |
| 1929 |
} |
} |
| 1930 |
|
|
| 1931 |
if ($match > 0) { |
if ($match > 0) { |
| 1932 |
return {type => CHARACTER_TOKEN, data => $value}; |
return {type => CHARACTER_TOKEN, data => $value, has_reference => 1}; |
| 1933 |
} elsif ($match < 0) { |
} elsif ($match < 0) { |
| 1934 |
!!!parse-error (type => 'no refc'); |
!!!parse-error (type => 'no refc'); |
| 1935 |
if ($in_attr and $match < -1) { |
if ($in_attr and $match < -1) { |
| 1936 |
return {type => CHARACTER_TOKEN, data => '&'.$entity_name}; |
return {type => CHARACTER_TOKEN, data => '&'.$entity_name}; |
| 1937 |
} else { |
} else { |
| 1938 |
return {type => CHARACTER_TOKEN, data => $value}; |
return {type => CHARACTER_TOKEN, data => $value, has_reference => 1}; |
| 1939 |
} |
} |
| 1940 |
} else { |
} else { |
| 1941 |
!!!parse-error (type => 'bare ero'); |
!!!parse-error (type => 'bare ero'); |
| 1942 |
## NOTE: No characters are consumed in the spec. |
## NOTE: "No characters are consumed" in the spec. |
| 1943 |
return {type => CHARACTER_TOKEN, data => '&'.$value}; |
return {type => CHARACTER_TOKEN, data => '&'.$value}; |
| 1944 |
} |
} |
| 1945 |
} else { |
} else { |
| 2187 |
redo B; |
redo B; |
| 2188 |
} |
} |
| 2189 |
} |
} |
| 2190 |
|
|
| 2191 |
|
$self->{application_cache_selection}->(undef); |
| 2192 |
|
|
| 2193 |
|
# |
| 2194 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 2195 |
|
if ($token->{tag_name} eq 'html' and |
| 2196 |
|
$token->{attributes}->{manifest}) { |
| 2197 |
|
$self->{application_cache_selection} |
| 2198 |
|
->($token->{attributes}->{manifest}->{value}); |
| 2199 |
|
## ISSUE: No relative reference resolution? |
| 2200 |
|
} else { |
| 2201 |
|
$self->{application_cache_selection}->(undef); |
| 2202 |
|
} |
| 2203 |
|
|
| 2204 |
|
## ISSUE: There is an issue in the spec |
| 2205 |
# |
# |
| 2206 |
} elsif ({ |
} elsif ({ |
|
START_TAG_TOKEN, 1, |
|
| 2207 |
END_TAG_TOKEN, 1, |
END_TAG_TOKEN, 1, |
| 2208 |
END_OF_FILE_TOKEN, 1, |
END_OF_FILE_TOKEN, 1, |
| 2209 |
}->{$token->{type}}) { |
}->{$token->{type}}) { |
| 2210 |
|
$self->{application_cache_selection}->(undef); |
| 2211 |
|
|
| 2212 |
## ISSUE: There is an issue in the spec |
## ISSUE: There is an issue in the spec |
| 2213 |
# |
# |
| 2214 |
} else { |
} else { |
| 2215 |
die "$0: $token->{type}: Unknown token type"; |
die "$0: $token->{type}: Unknown token type"; |
| 2216 |
} |
} |
| 2217 |
|
|
| 2218 |
my $root_element; !!!create-element ($root_element, 'html'); |
my $root_element; !!!create-element ($root_element, 'html'); |
| 2219 |
$self->{document}->append_child ($root_element); |
$self->{document}->append_child ($root_element); |
| 2220 |
push @{$self->{open_elements}}, [$root_element, 'html']; |
push @{$self->{open_elements}}, [$root_element, 'html']; |
| 2871 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2872 |
} |
} |
| 2873 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2874 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
my $meta_el = pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
| 2875 |
|
|
| 2876 |
unless ($self->{confident}) { |
unless ($self->{confident}) { |
|
my $charset; |
|
| 2877 |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
| 2878 |
$charset = $token->{attributes}->{charset}->{value}; |
$self->{change_encoding} |
| 2879 |
} |
->($self, $token->{attributes}->{charset}->{value}); |
| 2880 |
if ($token->{attributes}->{'http-equiv'}) { |
|
| 2881 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2882 |
|
->set_user_data (manakai_has_reference => |
| 2883 |
|
$token->{attributes}->{charset} |
| 2884 |
|
->{has_reference}); |
| 2885 |
|
} elsif ($token->{attributes}->{content}) { |
| 2886 |
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
| 2887 |
if ($token->{attributes}->{'http-equiv'}->{value} |
if ($token->{attributes}->{content}->{value} |
| 2888 |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
| 2889 |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 2890 |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
| 2891 |
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
$self->{change_encoding} |
| 2892 |
} ## TODO: And if supported |
->($self, defined $1 ? $1 : defined $2 ? $2 : $3); |
| 2893 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2894 |
|
->set_user_data (manakai_has_reference => |
| 2895 |
|
$token->{attributes}->{content} |
| 2896 |
|
->{has_reference}); |
| 2897 |
|
} |
| 2898 |
|
} |
| 2899 |
|
} else { |
| 2900 |
|
if ($token->{attributes}->{charset}) { |
| 2901 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2902 |
|
->set_user_data (manakai_has_reference => |
| 2903 |
|
$token->{attributes}->{charset} |
| 2904 |
|
->{has_reference}); |
| 2905 |
|
} |
| 2906 |
|
if ($token->{attributes}->{content}) { |
| 2907 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2908 |
|
->set_user_data (manakai_has_reference => |
| 2909 |
|
$token->{attributes}->{content} |
| 2910 |
|
->{has_reference}); |
| 2911 |
} |
} |
|
## TODO: Change the encoding |
|
| 2912 |
} |
} |
| 2913 |
|
|
|
## TODO: Extracting |charset| from |meta|. |
|
| 2914 |
pop @{$self->{open_elements}} |
pop @{$self->{open_elements}} |
| 2915 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2916 |
!!!next-token; |
!!!next-token; |
| 3480 |
$insert = $insert_to_current; |
$insert = $insert_to_current; |
| 3481 |
# |
# |
| 3482 |
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
| 3483 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
|
## NOTE: There are "character in table" code clones. |
|
| 3484 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3485 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3486 |
|
|
| 3537 |
|
|
| 3538 |
!!!next-token; |
!!!next-token; |
| 3539 |
redo B; |
redo B; |
| 3540 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 3541 |
if ({ |
if ({ |
| 3542 |
tr => ($self->{insertion_mode} != IN_ROW_IM), |
tr => ($self->{insertion_mode} != IN_ROW_IM), |
| 3543 |
th => 1, td => 1, |
th => 1, td => 1, |
| 3723 |
die "$0: in table: <>: $token->{tag_name}"; |
die "$0: in table: <>: $token->{tag_name}"; |
| 3724 |
} |
} |
| 3725 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
|
## NOTE: There are code clones for this "table in table" |
|
| 3726 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3727 |
|
|
| 3728 |
## As if </table> |
## As if </table> |
| 3770 |
|
|
| 3771 |
## reprocess |
## reprocess |
| 3772 |
redo B; |
redo B; |
| 3773 |
} else { |
} else { |
| 3774 |
# |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
| 3775 |
} |
|
| 3776 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
$insert = $insert_to_foster; |
| 3777 |
|
# |
| 3778 |
|
} |
| 3779 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3780 |
if ($token->{tag_name} eq 'tr' and |
if ($token->{tag_name} eq 'tr' and |
| 3781 |
$self->{insertion_mode} == IN_ROW_IM) { |
$self->{insertion_mode} == IN_ROW_IM) { |
| 3782 |
## have an element in table scope |
## have an element in table scope |
| 4033 |
## Ignore the token |
## Ignore the token |
| 4034 |
!!!next-token; |
!!!next-token; |
| 4035 |
redo B; |
redo B; |
| 4036 |
} else { |
} else { |
| 4037 |
# |
!!!parse-error (type => 'in table:/'.$token->{tag_name}); |
|
} |
|
|
} else { |
|
|
die "$0: $token->{type}: Unknown token type"; |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
|
| 4038 |
|
|
| 4039 |
$insert = $insert_to_foster; |
$insert = $insert_to_foster; |
| 4040 |
# |
# |
| 4041 |
|
} |
| 4042 |
|
} else { |
| 4043 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4044 |
|
} |
| 4045 |
} elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) { |
} elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) { |
| 4046 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 4047 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4100 |
redo B; |
redo B; |
| 4101 |
} |
} |
| 4102 |
} elsif ($self->{insertion_mode} == IN_SELECT_IM) { |
} elsif ($self->{insertion_mode} == IN_SELECT_IM) { |
| 4103 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 4104 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4105 |
!!!next-token; |
!!!next-token; |
| 4106 |
redo B; |
redo B; |
| 4107 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 4108 |
if ($token->{tag_name} eq 'option') { |
if ($token->{tag_name} eq 'option') { |
| 4109 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 4110 |
## As if </option> |
## As if </option> |
| 4157 |
|
|
| 4158 |
!!!next-token; |
!!!next-token; |
| 4159 |
redo B; |
redo B; |
| 4160 |
} else { |
} else { |
| 4161 |
# |
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
| 4162 |
} |
## Ignore the token |
| 4163 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
!!!next-token; |
| 4164 |
|
redo B; |
| 4165 |
|
} |
| 4166 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 4167 |
if ($token->{tag_name} eq 'optgroup') { |
if ($token->{tag_name} eq 'optgroup') { |
| 4168 |
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
| 4169 |
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
| 4265 |
|
|
| 4266 |
## reprocess |
## reprocess |
| 4267 |
redo B; |
redo B; |
| 4268 |
} else { |
} else { |
| 4269 |
# |
!!!parse-error (type => 'in select:/'.$token->{tag_name}); |
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
|
| 4270 |
## Ignore the token |
## Ignore the token |
| 4271 |
!!!next-token; |
!!!next-token; |
| 4272 |
redo B; |
redo B; |
| 4273 |
|
} |
| 4274 |
|
} else { |
| 4275 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4276 |
|
} |
| 4277 |
} elsif ($self->{insertion_mode} & BODY_AFTER_IMS) { |
} elsif ($self->{insertion_mode} & BODY_AFTER_IMS) { |
| 4278 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 4279 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4481 |
} elsif ($token->{tag_name} eq 'meta') { |
} elsif ($token->{tag_name} eq 'meta') { |
| 4482 |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
| 4483 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 4484 |
pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
my $meta_el = pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec. |
| 4485 |
|
|
| 4486 |
unless ($self->{confident}) { |
unless ($self->{confident}) { |
|
my $charset; |
|
| 4487 |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
| 4488 |
$charset = $token->{attributes}->{charset}->{value}; |
$self->{change_encoding} |
| 4489 |
} |
->($self, $token->{attributes}->{charset}->{value}); |
| 4490 |
if ($token->{attributes}->{'http-equiv'}) { |
|
| 4491 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 4492 |
|
->set_user_data (manakai_has_reference => |
| 4493 |
|
$token->{attributes}->{charset} |
| 4494 |
|
->{has_reference}); |
| 4495 |
|
} elsif ($token->{attributes}->{content}) { |
| 4496 |
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition. |
| 4497 |
if ($token->{attributes}->{'http-equiv'}->{value} |
if ($token->{attributes}->{content}->{value} |
| 4498 |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
| 4499 |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 4500 |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
| 4501 |
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
$self->{change_encoding} |
| 4502 |
} ## TODO: And if supported |
->($self, defined $1 ? $1 : defined $2 ? $2 : $3); |
| 4503 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 4504 |
|
->set_user_data (manakai_has_reference => |
| 4505 |
|
$token->{attributes}->{content} |
| 4506 |
|
->{has_reference}); |
| 4507 |
|
} |
| 4508 |
|
} |
| 4509 |
|
} else { |
| 4510 |
|
if ($token->{attributes}->{charset}) { |
| 4511 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 4512 |
|
->set_user_data (manakai_has_reference => |
| 4513 |
|
$token->{attributes}->{charset} |
| 4514 |
|
->{has_reference}); |
| 4515 |
|
} |
| 4516 |
|
if ($token->{attributes}->{content}) { |
| 4517 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 4518 |
|
->set_user_data (manakai_has_reference => |
| 4519 |
|
$token->{attributes}->{content} |
| 4520 |
|
->{has_reference}); |
| 4521 |
} |
} |
|
## TODO: Change the encoding |
|
| 4522 |
} |
} |
| 4523 |
|
|
| 4524 |
!!!next-token; |
!!!next-token; |
| 4825 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4826 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 4827 |
if ($node->[1] eq 'nobr') { |
if ($node->[1] eq 'nobr') { |
| 4828 |
!!!parse-error (type => 'not closed:nobr'); |
!!!parse-error (type => 'in nobr:nobr'); |
| 4829 |
!!!back-token; |
!!!back-token; |
| 4830 |
$token = {type => END_TAG_TOKEN, tag_name => 'nobr'}; |
$token = {type => END_TAG_TOKEN, tag_name => 'nobr'}; |
| 4831 |
redo B; |
redo B; |
| 5030 |
noframes => 1, |
noframes => 1, |
| 5031 |
noscript => 0, ## TODO: 1 if scripting is enabled |
noscript => 0, ## TODO: 1 if scripting is enabled |
| 5032 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5033 |
## NOTE: There are two "as if in body" code clones. |
## NOTE: There is an "as if in body" code clone. |
| 5034 |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
| 5035 |
redo B; |
redo B; |
| 5036 |
} elsif ($token->{tag_name} eq 'select') { |
} elsif ($token->{tag_name} eq 'select') { |
| 5186 |
if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) { |
if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) { |
| 5187 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 5188 |
} else { |
} else { |
| 5189 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 5190 |
} |
} |
| 5191 |
|
|
| 5192 |
undef $self->{form_element}; |
undef $self->{form_element}; |
| 5224 |
} # INSCOPE |
} # INSCOPE |
| 5225 |
|
|
| 5226 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
| 5227 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 5228 |
} |
} |
| 5229 |
|
|
| 5230 |
splice @{$self->{open_elements}}, $i if defined $i; |
splice @{$self->{open_elements}}, $i if defined $i; |
| 5293 |
|
|
| 5294 |
## Step 2 |
## Step 2 |
| 5295 |
if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { |
if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { |
| 5296 |
|
## NOTE: <x><y></x> |
| 5297 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 5298 |
} |
} |
| 5299 |
|
|
| 5344 |
my $s = \$_[0]; |
my $s = \$_[0]; |
| 5345 |
my $onerror = $_[1]; |
my $onerror = $_[1]; |
| 5346 |
|
|
| 5347 |
|
## ISSUE: Should {confident} be true? |
| 5348 |
|
|
| 5349 |
my $nt = $node->node_type; |
my $nt = $node->node_type; |
| 5350 |
if ($nt == 9) { |
if ($nt == 9) { |
| 5351 |
# MUST |
# MUST |
| 5498 |
|
|
| 5499 |
} # tree construction stage |
} # tree construction stage |
| 5500 |
|
|
| 5501 |
sub get_inner_html ($$$) { |
package Whatpm::HTML::RestartParser; |
| 5502 |
my (undef, $node, $on_error) = @_; |
push our @ISA, 'Error'; |
|
|
|
|
## Step 1 |
|
|
my $s = ''; |
|
|
|
|
|
my $in_cdata; |
|
|
my $parent = $node; |
|
|
while (defined $parent) { |
|
|
if ($parent->node_type == 1 and |
|
|
$parent->namespace_uri eq 'http://www.w3.org/1999/xhtml' and |
|
|
{ |
|
|
style => 1, script => 1, xmp => 1, iframe => 1, |
|
|
noembed => 1, noframes => 1, noscript => 1, |
|
|
}->{$parent->local_name}) { ## TODO: case thingy |
|
|
$in_cdata = 1; |
|
|
} |
|
|
$parent = $parent->parent_node; |
|
|
} |
|
|
|
|
|
## Step 2 |
|
|
my @node = @{$node->child_nodes}; |
|
|
C: while (@node) { |
|
|
my $child = shift @node; |
|
|
unless (ref $child) { |
|
|
if ($child eq 'cdata-out') { |
|
|
$in_cdata = 0; |
|
|
} else { |
|
|
$s .= $child; # end tag |
|
|
} |
|
|
next C; |
|
|
} |
|
|
|
|
|
my $nt = $child->node_type; |
|
|
if ($nt == 1) { # Element |
|
|
my $tag_name = $child->tag_name; ## TODO: manakai_tag_name |
|
|
$s .= '<' . $tag_name; |
|
|
## NOTE: Non-HTML case: |
|
|
## <http://permalink.gmane.org/gmane.org.w3c.whatwg.discuss/11191> |
|
|
|
|
|
my @attrs = @{$child->attributes}; # sort order MUST be stable |
|
|
for my $attr (@attrs) { # order is implementation dependent |
|
|
my $attr_name = $attr->name; ## TODO: manakai_name |
|
|
$s .= ' ' . $attr_name . '="'; |
|
|
my $attr_value = $attr->value; |
|
|
## escape |
|
|
$attr_value =~ s/&/&/g; |
|
|
$attr_value =~ s/</</g; |
|
|
$attr_value =~ s/>/>/g; |
|
|
$attr_value =~ s/"/"/g; |
|
|
$s .= $attr_value . '"'; |
|
|
} |
|
|
$s .= '>'; |
|
|
|
|
|
next C if { |
|
|
area => 1, base => 1, basefont => 1, bgsound => 1, |
|
|
br => 1, col => 1, embed => 1, frame => 1, hr => 1, |
|
|
img => 1, input => 1, link => 1, meta => 1, param => 1, |
|
|
spacer => 1, wbr => 1, |
|
|
}->{$tag_name}; |
|
|
|
|
|
$s .= "\x0A" if $tag_name eq 'pre' or $tag_name eq 'textarea'; |
|
|
|
|
|
if (not $in_cdata and { |
|
|
style => 1, script => 1, xmp => 1, iframe => 1, |
|
|
noembed => 1, noframes => 1, noscript => 1, |
|
|
plaintext => 1, |
|
|
}->{$tag_name}) { |
|
|
unshift @node, 'cdata-out'; |
|
|
$in_cdata = 1; |
|
|
} |
|
|
|
|
|
unshift @node, @{$child->child_nodes}, '</' . $tag_name . '>'; |
|
|
} elsif ($nt == 3 or $nt == 4) { |
|
|
if ($in_cdata) { |
|
|
$s .= $child->data; |
|
|
} else { |
|
|
my $value = $child->data; |
|
|
$value =~ s/&/&/g; |
|
|
$value =~ s/</</g; |
|
|
$value =~ s/>/>/g; |
|
|
$value =~ s/"/"/g; |
|
|
$s .= $value; |
|
|
} |
|
|
} elsif ($nt == 8) { |
|
|
$s .= '<!--' . $child->data . '-->'; |
|
|
} elsif ($nt == 10) { |
|
|
$s .= '<!DOCTYPE ' . $child->name . '>'; |
|
|
} elsif ($nt == 5) { # entrefs |
|
|
push @node, @{$child->child_nodes}; |
|
|
} else { |
|
|
$on_error->($child) if defined $on_error; |
|
|
} |
|
|
## ISSUE: This code does not support PIs. |
|
|
} # C |
|
|
|
|
|
## Step 3 |
|
|
return \$s; |
|
|
} # get_inner_html |
|
| 5503 |
|
|
| 5504 |
1; |
1; |
| 5505 |
# $Date$ |
# $Date$ |