| 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); |
| 14 |
## "{U+FEFF}..." in UTF-16BE/UTF-16LE is three or four characters? |
## "{U+FEFF}..." in UTF-16BE/UTF-16LE is three or four characters? |
| 15 |
## "{U+FEFF}..." in GB18030? |
## "{U+FEFF}..." in GB18030? |
| 16 |
|
|
| 17 |
|
## TODO: Control charcters and noncharacters are not allowed (HTML5 revision 1263) |
| 18 |
|
## TODO: 1252 parse error (revision 1264) |
| 19 |
|
## TODO: 8859-11 = 874 (revision 1271) |
| 20 |
|
|
| 21 |
my $permitted_slash_tag_name = { |
my $permitted_slash_tag_name = { |
| 22 |
base => 1, |
base => 1, |
| 23 |
link => 1, |
link => 1, |
| 89 |
}; |
}; |
| 90 |
# $phrasing_category: all other elements |
# $phrasing_category: all other elements |
| 91 |
|
|
| 92 |
|
sub parse_byte_string ($$$$;$) { |
| 93 |
|
my $self = ref $_[0] ? shift : shift->new; |
| 94 |
|
my $charset = shift; |
| 95 |
|
my $bytes_s = ref $_[0] ? $_[0] : \($_[0]); |
| 96 |
|
my $s; |
| 97 |
|
|
| 98 |
|
if (defined $charset) { |
| 99 |
|
require Encode; ## TODO: decode(utf8) don't delete BOM |
| 100 |
|
$s = \ (Encode::decode ($charset, $$bytes_s)); |
| 101 |
|
$self->{input_encoding} = lc $charset; ## TODO: normalize name |
| 102 |
|
$self->{confident} = 1; |
| 103 |
|
} else { |
| 104 |
|
## TODO: Implement HTML5 detection algorithm |
| 105 |
|
require Whatpm::Charset::UniversalCharDet; |
| 106 |
|
$charset = Whatpm::Charset::UniversalCharDet->detect_byte_string |
| 107 |
|
(substr ($$bytes_s, 0, 1024)); |
| 108 |
|
$charset ||= 'windows-1252'; |
| 109 |
|
$s = \ (Encode::decode ($charset, $$bytes_s)); |
| 110 |
|
$self->{input_encoding} = $charset; |
| 111 |
|
$self->{confident} = 0; |
| 112 |
|
} |
| 113 |
|
|
| 114 |
|
$self->{change_encoding} = sub { |
| 115 |
|
my $self = shift; |
| 116 |
|
my $charset = lc shift; |
| 117 |
|
## TODO: if $charset is supported |
| 118 |
|
## TODO: normalize charset name |
| 119 |
|
|
| 120 |
|
## "Change the encoding" algorithm: |
| 121 |
|
|
| 122 |
|
## Step 1 |
| 123 |
|
if ($charset eq 'utf-16') { ## ISSUE: UTF-16BE -> UTF-8? UTF-16LE -> UTF-8? |
| 124 |
|
$charset = 'utf-8'; |
| 125 |
|
} |
| 126 |
|
|
| 127 |
|
## Step 2 |
| 128 |
|
if (defined $self->{input_encoding} and |
| 129 |
|
$self->{input_encoding} eq $charset) { |
| 130 |
|
$self->{confident} = 1; |
| 131 |
|
return; |
| 132 |
|
} |
| 133 |
|
|
| 134 |
|
!!!parse-error (type => 'charset label detected:'.$self->{input_encoding}. |
| 135 |
|
':'.$charset, level => 'w'); |
| 136 |
|
|
| 137 |
|
## Step 3 |
| 138 |
|
# if (can) { |
| 139 |
|
## change the encoding on the fly. |
| 140 |
|
#$self->{confident} = 1; |
| 141 |
|
#return; |
| 142 |
|
# } |
| 143 |
|
|
| 144 |
|
## Step 4 |
| 145 |
|
throw Whatpm::HTML::RestartParser (charset => $charset); |
| 146 |
|
}; # $self->{change_encoding} |
| 147 |
|
|
| 148 |
|
my @args = @_; shift @args; # $s |
| 149 |
|
my $return; |
| 150 |
|
try { |
| 151 |
|
$return = $self->parse_char_string ($s, @args); |
| 152 |
|
} catch Whatpm::HTML::RestartParser with { |
| 153 |
|
my $charset = shift->{charset}; |
| 154 |
|
$s = \ (Encode::decode ($charset, $$bytes_s)); |
| 155 |
|
$self->{input_encoding} = $charset; ## TODO: normalize |
| 156 |
|
$self->{confident} = 1; |
| 157 |
|
$return = $self->parse_char_string ($s, @args); |
| 158 |
|
}; |
| 159 |
|
return $return; |
| 160 |
|
} # parse_byte_string |
| 161 |
|
|
| 162 |
|
*parse_char_string = \&parse_string; |
| 163 |
|
|
| 164 |
sub parse_string ($$$;$) { |
sub parse_string ($$$;$) { |
| 165 |
my $self = shift->new; |
my $self = ref $_[0] ? shift : shift->new; |
| 166 |
my $s = \$_[0]; |
my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 167 |
$self->{document} = $_[1]; |
$self->{document} = $_[1]; |
| 168 |
|
@{$self->{document}->child_nodes} = (); |
| 169 |
|
|
| 170 |
## NOTE: |set_inner_html| copies most of this method's code |
## NOTE: |set_inner_html| copies most of this method's code |
| 171 |
|
|
| 172 |
|
$self->{confident} = 1 unless exists $self->{confident}; |
| 173 |
|
$self->{document}->input_encoding ($self->{input_encoding}) |
| 174 |
|
if defined $self->{input_encoding}; |
| 175 |
|
|
| 176 |
my $i = 0; |
my $i = 0; |
| 177 |
my $line = 1; |
my $line = 1; |
| 178 |
my $column = 0; |
my $column = 0; |
| 229 |
$self->{parse_error} = sub { |
$self->{parse_error} = sub { |
| 230 |
# |
# |
| 231 |
}; |
}; |
| 232 |
|
$self->{change_encoding} = sub { |
| 233 |
|
# if ($_[0] is a supported encoding) { |
| 234 |
|
# run "change the encoding" algorithm; |
| 235 |
|
# throw Whatpm::HTML::RestartParser (charset => $new_encoding); |
| 236 |
|
# } |
| 237 |
|
}; |
| 238 |
|
$self->{application_cache_selection} = sub { |
| 239 |
|
# |
| 240 |
|
}; |
| 241 |
return $self; |
return $self; |
| 242 |
} # new |
} # new |
| 243 |
|
|
| 344 |
## ->{system_identifier} (DOCTYPE_TOKEN) |
## ->{system_identifier} (DOCTYPE_TOKEN) |
| 345 |
## ->{correct} == 1 or 0 (DOCTYPE_TOKEN) |
## ->{correct} == 1 or 0 (DOCTYPE_TOKEN) |
| 346 |
## ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN) |
## ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN) |
| 347 |
|
## ->{name} |
| 348 |
|
## ->{value} |
| 349 |
|
## ->{has_reference} == 1 or 0 |
| 350 |
## ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN) |
## ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN) |
| 351 |
|
|
| 352 |
## Emitted token MUST immediately be handled by the tree construction state. |
## Emitted token MUST immediately be handled by the tree construction state. |
| 357 |
## has completed loading. If one has, then it MUST be executed |
## has completed loading. If one has, then it MUST be executed |
| 358 |
## and removed from the list. |
## and removed from the list. |
| 359 |
|
|
| 360 |
|
## NOTE: HTML5 "Writing HTML documents" section, applied to |
| 361 |
|
## documents and not to user agents and conformance checkers, |
| 362 |
|
## contains some requirements that are not detected by the |
| 363 |
|
## parsing algorithm: |
| 364 |
|
## - Some requirements on character encoding declarations. ## TODO |
| 365 |
|
## - "Elements MUST NOT contain content that their content model disallows." |
| 366 |
|
## ... Some are parse error, some are not (will be reported by c.c.). |
| 367 |
|
## - Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) ## TODO |
| 368 |
|
## - Text (in elements, attributes, and comments) SHOULD NOT contain |
| 369 |
|
## control characters other than space characters. ## TODO: (what is control character? C0, C1 and DEL? Unicode control character?) |
| 370 |
|
|
| 371 |
|
## TODO: HTML5 poses authors two SHOULD-level requirements that cannot |
| 372 |
|
## be detected by the HTML5 parsing algorithm: |
| 373 |
|
## - Text, |
| 374 |
|
|
| 375 |
sub _get_next_token ($) { |
sub _get_next_token ($) { |
| 376 |
my $self = shift; |
my $self = shift; |
| 377 |
if (@{$self->{token}}) { |
if (@{$self->{token}}) { |
| 1122 |
$self->{current_attribute}->{value} .= '&'; |
$self->{current_attribute}->{value} .= '&'; |
| 1123 |
} else { |
} else { |
| 1124 |
$self->{current_attribute}->{value} .= $token->{data}; |
$self->{current_attribute}->{value} .= $token->{data}; |
| 1125 |
|
$self->{current_attribute}->{has_reference} = $token->{has_reference}; |
| 1126 |
## 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" |
| 1127 |
} |
} |
| 1128 |
|
|
| 1559 |
$self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
$self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
| 1560 |
!!!next-input-character; |
!!!next-input-character; |
| 1561 |
redo A; |
redo A; |
| 1562 |
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1563 |
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
| 1564 |
|
|
| 1565 |
|
$self->{state} = DATA_STATE; |
| 1566 |
|
!!!next-input-character; |
| 1567 |
|
|
| 1568 |
|
delete $self->{current_token}->{correct}; |
| 1569 |
|
!!!emit ($self->{current_token}); # DOCTYPE |
| 1570 |
|
|
| 1571 |
|
redo A; |
| 1572 |
} elsif ($self->{next_input_character} == -1) { |
} elsif ($self->{next_input_character} == -1) { |
| 1573 |
!!!parse-error (type => 'unclosed PUBLIC literal'); |
!!!parse-error (type => 'unclosed PUBLIC literal'); |
| 1574 |
|
|
| 1591 |
$self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
$self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
| 1592 |
!!!next-input-character; |
!!!next-input-character; |
| 1593 |
redo A; |
redo A; |
| 1594 |
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1595 |
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
| 1596 |
|
|
| 1597 |
|
$self->{state} = DATA_STATE; |
| 1598 |
|
!!!next-input-character; |
| 1599 |
|
|
| 1600 |
|
delete $self->{current_token}->{correct}; |
| 1601 |
|
!!!emit ($self->{current_token}); # DOCTYPE |
| 1602 |
|
|
| 1603 |
|
redo A; |
| 1604 |
} elsif ($self->{next_input_character} == -1) { |
} elsif ($self->{next_input_character} == -1) { |
| 1605 |
!!!parse-error (type => 'unclosed PUBLIC literal'); |
!!!parse-error (type => 'unclosed PUBLIC literal'); |
| 1606 |
|
|
| 1707 |
$self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
$self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
| 1708 |
!!!next-input-character; |
!!!next-input-character; |
| 1709 |
redo A; |
redo A; |
| 1710 |
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1711 |
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
| 1712 |
|
|
| 1713 |
|
$self->{state} = DATA_STATE; |
| 1714 |
|
!!!next-input-character; |
| 1715 |
|
|
| 1716 |
|
delete $self->{current_token}->{correct}; |
| 1717 |
|
!!!emit ($self->{current_token}); # DOCTYPE |
| 1718 |
|
|
| 1719 |
|
redo A; |
| 1720 |
} elsif ($self->{next_input_character} == -1) { |
} elsif ($self->{next_input_character} == -1) { |
| 1721 |
!!!parse-error (type => 'unclosed SYSTEM literal'); |
!!!parse-error (type => 'unclosed SYSTEM literal'); |
| 1722 |
|
|
| 1739 |
$self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
$self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
| 1740 |
!!!next-input-character; |
!!!next-input-character; |
| 1741 |
redo A; |
redo A; |
| 1742 |
|
} elsif ($self->{next_input_character} == 0x003E) { # > |
| 1743 |
|
!!!parse-error (type => 'unclosed PUBLIC literal'); |
| 1744 |
|
|
| 1745 |
|
$self->{state} = DATA_STATE; |
| 1746 |
|
!!!next-input-character; |
| 1747 |
|
|
| 1748 |
|
delete $self->{current_token}->{correct}; |
| 1749 |
|
!!!emit ($self->{current_token}); # DOCTYPE |
| 1750 |
|
|
| 1751 |
|
redo A; |
| 1752 |
} elsif ($self->{next_input_character} == -1) { |
} elsif ($self->{next_input_character} == -1) { |
| 1753 |
!!!parse-error (type => 'unclosed SYSTEM literal'); |
!!!parse-error (type => 'unclosed SYSTEM literal'); |
| 1754 |
|
|
| 1889 |
$code = $c1_entity_char->{$code}; |
$code = $c1_entity_char->{$code}; |
| 1890 |
} |
} |
| 1891 |
|
|
| 1892 |
return {type => CHARACTER_TOKEN, data => chr $code}; |
return {type => CHARACTER_TOKEN, data => chr $code, |
| 1893 |
|
has_reference => 1}; |
| 1894 |
} # X |
} # X |
| 1895 |
} elsif (0x0030 <= $self->{next_input_character} and |
} elsif (0x0030 <= $self->{next_input_character} and |
| 1896 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
$self->{next_input_character} <= 0x0039) { # 0..9 |
| 1925 |
$code = $c1_entity_char->{$code}; |
$code = $c1_entity_char->{$code}; |
| 1926 |
} |
} |
| 1927 |
|
|
| 1928 |
return {type => CHARACTER_TOKEN, data => chr $code}; |
return {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1}; |
| 1929 |
} else { |
} else { |
| 1930 |
!!!parse-error (type => 'bare nero'); |
!!!parse-error (type => 'bare nero'); |
| 1931 |
!!!back-next-input-character ($self->{next_input_character}); |
!!!back-next-input-character ($self->{next_input_character}); |
| 1973 |
} |
} |
| 1974 |
|
|
| 1975 |
if ($match > 0) { |
if ($match > 0) { |
| 1976 |
return {type => CHARACTER_TOKEN, data => $value}; |
return {type => CHARACTER_TOKEN, data => $value, has_reference => 1}; |
| 1977 |
} elsif ($match < 0) { |
} elsif ($match < 0) { |
| 1978 |
!!!parse-error (type => 'no refc'); |
!!!parse-error (type => 'no refc'); |
| 1979 |
if ($in_attr and $match < -1) { |
if ($in_attr and $match < -1) { |
| 1980 |
return {type => CHARACTER_TOKEN, data => '&'.$entity_name}; |
return {type => CHARACTER_TOKEN, data => '&'.$entity_name}; |
| 1981 |
} else { |
} else { |
| 1982 |
return {type => CHARACTER_TOKEN, data => $value}; |
return {type => CHARACTER_TOKEN, data => $value, has_reference => 1}; |
| 1983 |
} |
} |
| 1984 |
} else { |
} else { |
| 1985 |
!!!parse-error (type => 'bare ero'); |
!!!parse-error (type => 'bare ero'); |
| 1986 |
## NOTE: No characters are consumed in the spec. |
## NOTE: "No characters are consumed" in the spec. |
| 1987 |
return {type => CHARACTER_TOKEN, data => '&'.$value}; |
return {type => CHARACTER_TOKEN, data => '&'.$value}; |
| 1988 |
} |
} |
| 1989 |
} else { |
} else { |
| 2231 |
redo B; |
redo B; |
| 2232 |
} |
} |
| 2233 |
} |
} |
| 2234 |
|
|
| 2235 |
|
$self->{application_cache_selection}->(undef); |
| 2236 |
|
|
| 2237 |
|
# |
| 2238 |
|
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 2239 |
|
if ($token->{tag_name} eq 'html' and |
| 2240 |
|
$token->{attributes}->{manifest}) { |
| 2241 |
|
$self->{application_cache_selection} |
| 2242 |
|
->($token->{attributes}->{manifest}->{value}); |
| 2243 |
|
## ISSUE: No relative reference resolution? |
| 2244 |
|
} else { |
| 2245 |
|
$self->{application_cache_selection}->(undef); |
| 2246 |
|
} |
| 2247 |
|
|
| 2248 |
|
## ISSUE: There is an issue in the spec |
| 2249 |
# |
# |
| 2250 |
} elsif ({ |
} elsif ({ |
|
START_TAG_TOKEN, 1, |
|
| 2251 |
END_TAG_TOKEN, 1, |
END_TAG_TOKEN, 1, |
| 2252 |
END_OF_FILE_TOKEN, 1, |
END_OF_FILE_TOKEN, 1, |
| 2253 |
}->{$token->{type}}) { |
}->{$token->{type}}) { |
| 2254 |
|
$self->{application_cache_selection}->(undef); |
| 2255 |
|
|
| 2256 |
## ISSUE: There is an issue in the spec |
## ISSUE: There is an issue in the spec |
| 2257 |
# |
# |
| 2258 |
} else { |
} else { |
| 2259 |
die "$0: $token->{type}: Unknown token type"; |
die "$0: $token->{type}: Unknown token type"; |
| 2260 |
} |
} |
| 2261 |
|
|
| 2262 |
my $root_element; !!!create-element ($root_element, 'html'); |
my $root_element; !!!create-element ($root_element, 'html'); |
| 2263 |
$self->{document}->append_child ($root_element); |
$self->{document}->append_child ($root_element); |
| 2264 |
push @{$self->{open_elements}}, [$root_element, 'html']; |
push @{$self->{open_elements}}, [$root_element, 'html']; |
| 2915 |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
push @{$self->{open_elements}}, [$self->{head_element}, 'head']; |
| 2916 |
} |
} |
| 2917 |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
!!!insert-element ($token->{tag_name}, $token->{attributes}); |
| 2918 |
pop @{$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. |
| 2919 |
|
|
| 2920 |
unless ($self->{confident}) { |
unless ($self->{confident}) { |
|
my $charset; |
|
| 2921 |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
| 2922 |
$charset = $token->{attributes}->{charset}->{value}; |
$self->{change_encoding} |
| 2923 |
} |
->($self, $token->{attributes}->{charset}->{value}); |
| 2924 |
if ($token->{attributes}->{'http-equiv'}) { |
|
| 2925 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2926 |
|
->set_user_data (manakai_has_reference => |
| 2927 |
|
$token->{attributes}->{charset} |
| 2928 |
|
->{has_reference}); |
| 2929 |
|
} elsif ($token->{attributes}->{content}) { |
| 2930 |
## 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. |
| 2931 |
if ($token->{attributes}->{'http-equiv'}->{value} |
if ($token->{attributes}->{content}->{value} |
| 2932 |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
=~ /\A[^;]*;[\x09-\x0D\x20]*[Cc][Hh][Aa][Rr][Ss][Ee][Tt] |
| 2933 |
|
[\x09-\x0D\x20]*= |
| 2934 |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 2935 |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
| 2936 |
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
$self->{change_encoding} |
| 2937 |
} ## TODO: And if supported |
->($self, defined $1 ? $1 : defined $2 ? $2 : $3); |
| 2938 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2939 |
|
->set_user_data (manakai_has_reference => |
| 2940 |
|
$token->{attributes}->{content} |
| 2941 |
|
->{has_reference}); |
| 2942 |
|
} |
| 2943 |
|
} |
| 2944 |
|
} else { |
| 2945 |
|
if ($token->{attributes}->{charset}) { |
| 2946 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 2947 |
|
->set_user_data (manakai_has_reference => |
| 2948 |
|
$token->{attributes}->{charset} |
| 2949 |
|
->{has_reference}); |
| 2950 |
|
} |
| 2951 |
|
if ($token->{attributes}->{content}) { |
| 2952 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 2953 |
|
->set_user_data (manakai_has_reference => |
| 2954 |
|
$token->{attributes}->{content} |
| 2955 |
|
->{has_reference}); |
| 2956 |
} |
} |
|
## TODO: Change the encoding |
|
| 2957 |
} |
} |
| 2958 |
|
|
|
## TODO: Extracting |charset| from |meta|. |
|
| 2959 |
pop @{$self->{open_elements}} |
pop @{$self->{open_elements}} |
| 2960 |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
if $self->{insertion_mode} == AFTER_HEAD_IM; |
| 2961 |
!!!next-token; |
!!!next-token; |
| 3525 |
$insert = $insert_to_current; |
$insert = $insert_to_current; |
| 3526 |
# |
# |
| 3527 |
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
} elsif ($self->{insertion_mode} & TABLE_IMS) { |
| 3528 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
|
## NOTE: There are "character in table" code clones. |
|
| 3529 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 3530 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($1); |
| 3531 |
|
|
| 3582 |
|
|
| 3583 |
!!!next-token; |
!!!next-token; |
| 3584 |
redo B; |
redo B; |
| 3585 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 3586 |
if ({ |
if ({ |
| 3587 |
tr => ($self->{insertion_mode} != IN_ROW_IM), |
tr => ($self->{insertion_mode} != IN_ROW_IM), |
| 3588 |
th => 1, td => 1, |
th => 1, td => 1, |
| 3768 |
die "$0: in table: <>: $token->{tag_name}"; |
die "$0: in table: <>: $token->{tag_name}"; |
| 3769 |
} |
} |
| 3770 |
} elsif ($token->{tag_name} eq 'table') { |
} elsif ($token->{tag_name} eq 'table') { |
|
## NOTE: There are code clones for this "table in table" |
|
| 3771 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 3772 |
|
|
| 3773 |
## As if </table> |
## As if </table> |
| 3815 |
|
|
| 3816 |
## reprocess |
## reprocess |
| 3817 |
redo B; |
redo B; |
| 3818 |
} else { |
} else { |
| 3819 |
# |
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
| 3820 |
} |
|
| 3821 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
$insert = $insert_to_foster; |
| 3822 |
|
# |
| 3823 |
|
} |
| 3824 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 3825 |
if ($token->{tag_name} eq 'tr' and |
if ($token->{tag_name} eq 'tr' and |
| 3826 |
$self->{insertion_mode} == IN_ROW_IM) { |
$self->{insertion_mode} == IN_ROW_IM) { |
| 3827 |
## have an element in table scope |
## have an element in table scope |
| 4078 |
## Ignore the token |
## Ignore the token |
| 4079 |
!!!next-token; |
!!!next-token; |
| 4080 |
redo B; |
redo B; |
| 4081 |
} else { |
} else { |
| 4082 |
# |
!!!parse-error (type => 'in table:/'.$token->{tag_name}); |
|
} |
|
|
} else { |
|
|
die "$0: $token->{type}: Unknown token type"; |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in table:'.$token->{tag_name}); |
|
| 4083 |
|
|
| 4084 |
$insert = $insert_to_foster; |
$insert = $insert_to_foster; |
| 4085 |
# |
# |
| 4086 |
|
} |
| 4087 |
|
} else { |
| 4088 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4089 |
|
} |
| 4090 |
} elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) { |
} elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) { |
| 4091 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 4092 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4145 |
redo B; |
redo B; |
| 4146 |
} |
} |
| 4147 |
} elsif ($self->{insertion_mode} == IN_SELECT_IM) { |
} elsif ($self->{insertion_mode} == IN_SELECT_IM) { |
| 4148 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 4149 |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
$self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data}); |
| 4150 |
!!!next-token; |
!!!next-token; |
| 4151 |
redo B; |
redo B; |
| 4152 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
| 4153 |
if ($token->{tag_name} eq 'option') { |
if ($token->{tag_name} eq 'option') { |
| 4154 |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
if ($self->{open_elements}->[-1]->[1] eq 'option') { |
| 4155 |
## As if </option> |
## As if </option> |
| 4202 |
|
|
| 4203 |
!!!next-token; |
!!!next-token; |
| 4204 |
redo B; |
redo B; |
| 4205 |
} else { |
} else { |
| 4206 |
# |
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
| 4207 |
} |
## Ignore the token |
| 4208 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
!!!next-token; |
| 4209 |
|
redo B; |
| 4210 |
|
} |
| 4211 |
|
} elsif ($token->{type} == END_TAG_TOKEN) { |
| 4212 |
if ($token->{tag_name} eq 'optgroup') { |
if ($token->{tag_name} eq 'optgroup') { |
| 4213 |
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
if ($self->{open_elements}->[-1]->[1] eq 'option' and |
| 4214 |
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
$self->{open_elements}->[-2]->[1] eq 'optgroup') { |
| 4310 |
|
|
| 4311 |
## reprocess |
## reprocess |
| 4312 |
redo B; |
redo B; |
| 4313 |
} else { |
} else { |
| 4314 |
# |
!!!parse-error (type => 'in select:/'.$token->{tag_name}); |
|
} |
|
|
} else { |
|
|
# |
|
|
} |
|
|
|
|
|
!!!parse-error (type => 'in select:'.$token->{tag_name}); |
|
| 4315 |
## Ignore the token |
## Ignore the token |
| 4316 |
!!!next-token; |
!!!next-token; |
| 4317 |
redo B; |
redo B; |
| 4318 |
|
} |
| 4319 |
|
} else { |
| 4320 |
|
die "$0: $token->{type}: Unknown token type"; |
| 4321 |
|
} |
| 4322 |
} elsif ($self->{insertion_mode} & BODY_AFTER_IMS) { |
} elsif ($self->{insertion_mode} & BODY_AFTER_IMS) { |
| 4323 |
if ($token->{type} == CHARACTER_TOKEN) { |
if ($token->{type} == CHARACTER_TOKEN) { |
| 4324 |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { |
| 4526 |
} elsif ($token->{tag_name} eq 'meta') { |
} elsif ($token->{tag_name} eq 'meta') { |
| 4527 |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
## NOTE: This is an "as if in head" code clone, only "-t" differs |
| 4528 |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
!!!insert-element-t ($token->{tag_name}, $token->{attributes}); |
| 4529 |
pop @{$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. |
| 4530 |
|
|
| 4531 |
unless ($self->{confident}) { |
unless ($self->{confident}) { |
|
my $charset; |
|
| 4532 |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
if ($token->{attributes}->{charset}) { ## TODO: And if supported |
| 4533 |
$charset = $token->{attributes}->{charset}->{value}; |
$self->{change_encoding} |
| 4534 |
} |
->($self, $token->{attributes}->{charset}->{value}); |
| 4535 |
if ($token->{attributes}->{'http-equiv'}) { |
|
| 4536 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 4537 |
|
->set_user_data (manakai_has_reference => |
| 4538 |
|
$token->{attributes}->{charset} |
| 4539 |
|
->{has_reference}); |
| 4540 |
|
} elsif ($token->{attributes}->{content}) { |
| 4541 |
## 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. |
| 4542 |
if ($token->{attributes}->{'http-equiv'}->{value} |
if ($token->{attributes}->{content}->{value} |
| 4543 |
=~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*= |
=~ /\A[^;]*;[\x09-\x0D\x20]*[Cc][Hh][Aa][Rr][Ss][Ee][Tt] |
| 4544 |
|
[\x09-\x0D\x20]*= |
| 4545 |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
[\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'| |
| 4546 |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) { |
| 4547 |
$charset = defined $1 ? $1 : defined $2 ? $2 : $3; |
$self->{change_encoding} |
| 4548 |
} ## TODO: And if supported |
->($self, defined $1 ? $1 : defined $2 ? $2 : $3); |
| 4549 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 4550 |
|
->set_user_data (manakai_has_reference => |
| 4551 |
|
$token->{attributes}->{content} |
| 4552 |
|
->{has_reference}); |
| 4553 |
|
} |
| 4554 |
|
} |
| 4555 |
|
} else { |
| 4556 |
|
if ($token->{attributes}->{charset}) { |
| 4557 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'charset') |
| 4558 |
|
->set_user_data (manakai_has_reference => |
| 4559 |
|
$token->{attributes}->{charset} |
| 4560 |
|
->{has_reference}); |
| 4561 |
|
} |
| 4562 |
|
if ($token->{attributes}->{content}) { |
| 4563 |
|
$meta_el->[0]->get_attribute_node_ns (undef, 'content') |
| 4564 |
|
->set_user_data (manakai_has_reference => |
| 4565 |
|
$token->{attributes}->{content} |
| 4566 |
|
->{has_reference}); |
| 4567 |
} |
} |
|
## TODO: Change the encoding |
|
| 4568 |
} |
} |
| 4569 |
|
|
| 4570 |
!!!next-token; |
!!!next-token; |
| 4871 |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
INSCOPE: for (reverse 0..$#{$self->{open_elements}}) { |
| 4872 |
my $node = $self->{open_elements}->[$_]; |
my $node = $self->{open_elements}->[$_]; |
| 4873 |
if ($node->[1] eq 'nobr') { |
if ($node->[1] eq 'nobr') { |
| 4874 |
!!!parse-error (type => 'not closed:nobr'); |
!!!parse-error (type => 'in nobr:nobr'); |
| 4875 |
!!!back-token; |
!!!back-token; |
| 4876 |
$token = {type => END_TAG_TOKEN, tag_name => 'nobr'}; |
$token = {type => END_TAG_TOKEN, tag_name => 'nobr'}; |
| 4877 |
redo B; |
redo B; |
| 5076 |
noframes => 1, |
noframes => 1, |
| 5077 |
noscript => 0, ## TODO: 1 if scripting is enabled |
noscript => 0, ## TODO: 1 if scripting is enabled |
| 5078 |
}->{$token->{tag_name}}) { |
}->{$token->{tag_name}}) { |
| 5079 |
## NOTE: There are two "as if in body" code clones. |
## NOTE: There is an "as if in body" code clone. |
| 5080 |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
$parse_rcdata->(CDATA_CONTENT_MODEL, $insert); |
| 5081 |
redo B; |
redo B; |
| 5082 |
} elsif ($token->{tag_name} eq 'select') { |
} elsif ($token->{tag_name} eq 'select') { |
| 5232 |
if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) { |
if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) { |
| 5233 |
pop @{$self->{open_elements}}; |
pop @{$self->{open_elements}}; |
| 5234 |
} else { |
} else { |
| 5235 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 5236 |
} |
} |
| 5237 |
|
|
| 5238 |
undef $self->{form_element}; |
undef $self->{form_element}; |
| 5270 |
} # INSCOPE |
} # INSCOPE |
| 5271 |
|
|
| 5272 |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) { |
| 5273 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'unmatched end tag:'.$token->{tag_name}); |
| 5274 |
} |
} |
| 5275 |
|
|
| 5276 |
splice @{$self->{open_elements}}, $i if defined $i; |
splice @{$self->{open_elements}}, $i if defined $i; |
| 5339 |
|
|
| 5340 |
## Step 2 |
## Step 2 |
| 5341 |
if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { |
if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) { |
| 5342 |
|
## NOTE: <x><y></x> |
| 5343 |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
!!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]); |
| 5344 |
} |
} |
| 5345 |
|
|
| 5390 |
my $s = \$_[0]; |
my $s = \$_[0]; |
| 5391 |
my $onerror = $_[1]; |
my $onerror = $_[1]; |
| 5392 |
|
|
| 5393 |
|
## ISSUE: Should {confident} be true? |
| 5394 |
|
|
| 5395 |
my $nt = $node->node_type; |
my $nt = $node->node_type; |
| 5396 |
if ($nt == 9) { |
if ($nt == 9) { |
| 5397 |
# MUST |
# MUST |
| 5544 |
|
|
| 5545 |
} # tree construction stage |
} # tree construction stage |
| 5546 |
|
|
| 5547 |
sub get_inner_html ($$$) { |
package Whatpm::HTML::RestartParser; |
| 5548 |
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 |
|
| 5549 |
|
|
| 5550 |
1; |
1; |
| 5551 |
# $Date$ |
# $Date$ |