| 354 |
return $self->parse_byte_stream ($charset_name, $input, @_[1..$#_]); |
return $self->parse_byte_stream ($charset_name, $input, @_[1..$#_]); |
| 355 |
} # parse_byte_string |
} # parse_byte_string |
| 356 |
|
|
| 357 |
sub parse_byte_stream ($$$$;$) { |
sub parse_byte_stream ($$$$;$$) { |
| 358 |
|
# my ($self, $charset_name, $byte_stream, $doc, $onerror, $get_wrapper) = @_; |
| 359 |
my $self = ref $_[0] ? shift : shift->new; |
my $self = ref $_[0] ? shift : shift->new; |
| 360 |
my $charset_name = shift; |
my $charset_name = shift; |
| 361 |
my $byte_stream = $_[0]; |
my $byte_stream = $_[0]; |
| 366 |
}; |
}; |
| 367 |
$self->{parse_error} = $onerror; # updated later by parse_char_string |
$self->{parse_error} = $onerror; # updated later by parse_char_string |
| 368 |
|
|
| 369 |
|
my $get_wrapper = $_[3] || sub ($) { |
| 370 |
|
return $_[0]; # $_[0] = byte stream handle, returned = arg to char handle |
| 371 |
|
}; |
| 372 |
|
|
| 373 |
## HTML5 encoding sniffing algorithm |
## HTML5 encoding sniffing algorithm |
| 374 |
require Message::Charset::Info; |
require Message::Charset::Info; |
| 375 |
my $charset; |
my $charset; |
| 377 |
my ($char_stream, $e_status); |
my ($char_stream, $e_status); |
| 378 |
|
|
| 379 |
SNIFFING: { |
SNIFFING: { |
| 380 |
|
## NOTE: By setting |allow_fallback| option true when the |
| 381 |
|
## |get_decode_handle| method is invoked, we ignore what the HTML5 |
| 382 |
|
## spec requires, i.e. unsupported encoding should be ignored. |
| 383 |
|
## TODO: We should not do this unless the parser is invoked |
| 384 |
|
## in the conformance checking mode, in which this behavior |
| 385 |
|
## would be useful. |
| 386 |
|
|
| 387 |
## Step 1 |
## Step 1 |
| 388 |
if (defined $charset_name) { |
if (defined $charset_name) { |
| 389 |
$charset = Message::Charset::Info->get_by_iana_name ($charset_name); |
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 390 |
|
## TODO: Is this ok? Transfer protocol's parameter should be |
| 391 |
|
## interpreted in its semantics? |
| 392 |
|
|
| 393 |
## ISSUE: Unsupported encoding is not ignored according to the spec. |
## ISSUE: Unsupported encoding is not ignored according to the spec. |
| 394 |
($char_stream, $e_status) = $charset->get_decode_handle |
($char_stream, $e_status) = $charset->get_decode_handle |
| 412 |
|
|
| 413 |
## Step 3 |
## Step 3 |
| 414 |
if ($byte_buffer =~ /^\xFE\xFF/) { |
if ($byte_buffer =~ /^\xFE\xFF/) { |
| 415 |
$charset = Message::Charset::Info->get_by_iana_name ('utf-16be'); |
$charset = Message::Charset::Info->get_by_html_name ('utf-16be'); |
| 416 |
($char_stream, $e_status) = $charset->get_decode_handle |
($char_stream, $e_status) = $charset->get_decode_handle |
| 417 |
($byte_stream, allow_error_reporting => 1, |
($byte_stream, allow_error_reporting => 1, |
| 418 |
allow_fallback => 1, byte_buffer => \$byte_buffer); |
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 419 |
$self->{confident} = 1; |
$self->{confident} = 1; |
| 420 |
last SNIFFING; |
last SNIFFING; |
| 421 |
} elsif ($byte_buffer =~ /^\xFF\xFE/) { |
} elsif ($byte_buffer =~ /^\xFF\xFE/) { |
| 422 |
$charset = Message::Charset::Info->get_by_iana_name ('utf-16le'); |
$charset = Message::Charset::Info->get_by_html_name ('utf-16le'); |
| 423 |
($char_stream, $e_status) = $charset->get_decode_handle |
($char_stream, $e_status) = $charset->get_decode_handle |
| 424 |
($byte_stream, allow_error_reporting => 1, |
($byte_stream, allow_error_reporting => 1, |
| 425 |
allow_fallback => 1, byte_buffer => \$byte_buffer); |
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 426 |
$self->{confident} = 1; |
$self->{confident} = 1; |
| 427 |
last SNIFFING; |
last SNIFFING; |
| 428 |
} elsif ($byte_buffer =~ /^\xEF\xBB\xBF/) { |
} elsif ($byte_buffer =~ /^\xEF\xBB\xBF/) { |
| 429 |
$charset = Message::Charset::Info->get_by_iana_name ('utf-8'); |
$charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 430 |
($char_stream, $e_status) = $charset->get_decode_handle |
($char_stream, $e_status) = $charset->get_decode_handle |
| 431 |
($byte_stream, allow_error_reporting => 1, |
($byte_stream, allow_error_reporting => 1, |
| 432 |
allow_fallback => 1, byte_buffer => \$byte_buffer); |
allow_fallback => 1, byte_buffer => \$byte_buffer); |
| 445 |
$charset_name = Whatpm::Charset::UniversalCharDet->detect_byte_string |
$charset_name = Whatpm::Charset::UniversalCharDet->detect_byte_string |
| 446 |
($byte_buffer); |
($byte_buffer); |
| 447 |
if (defined $charset_name) { |
if (defined $charset_name) { |
| 448 |
$charset = Message::Charset::Info->get_by_iana_name ($charset_name); |
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 449 |
|
|
| 450 |
## ISSUE: Unsupported encoding is not ignored according to the spec. |
## ISSUE: Unsupported encoding is not ignored according to the spec. |
| 451 |
require Whatpm::Charset::DecodeHandle; |
require Whatpm::Charset::DecodeHandle; |
| 468 |
|
|
| 469 |
## Step 7: default |
## Step 7: default |
| 470 |
## TODO: Make this configurable. |
## TODO: Make this configurable. |
| 471 |
$charset = Message::Charset::Info->get_by_iana_name ('windows-1252'); |
$charset = Message::Charset::Info->get_by_html_name ('windows-1252'); |
| 472 |
## NOTE: We choose |windows-1252| here, since |utf-8| should be |
## NOTE: We choose |windows-1252| here, since |utf-8| should be |
| 473 |
## detectable in the step 6. |
## detectable in the step 6. |
| 474 |
require Whatpm::Charset::DecodeHandle; |
require Whatpm::Charset::DecodeHandle; |
| 488 |
$self->{confident} = 0; |
$self->{confident} = 0; |
| 489 |
} # SNIFFING |
} # SNIFFING |
| 490 |
|
|
|
$self->{input_encoding} = $charset->get_iana_name; |
|
| 491 |
if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 492 |
|
$self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 493 |
!!!parse-error (type => 'chardecode:fallback', |
!!!parse-error (type => 'chardecode:fallback', |
| 494 |
text => $self->{input_encoding}, |
#text => $self->{input_encoding}, |
| 495 |
level => $self->{level}->{uncertain}, |
level => $self->{level}->{uncertain}, |
| 496 |
line => 1, column => 1, |
line => 1, column => 1, |
| 497 |
layer => 'encode'); |
layer => 'encode'); |
| 498 |
} elsif (not ($e_status & |
} elsif (not ($e_status & |
| 499 |
Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL())) { |
Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL())) { |
| 500 |
|
$self->{input_encoding} = $charset->get_iana_name; |
| 501 |
!!!parse-error (type => 'chardecode:no error', |
!!!parse-error (type => 'chardecode:no error', |
| 502 |
text => $self->{input_encoding}, |
text => $self->{input_encoding}, |
| 503 |
level => $self->{level}->{uncertain}, |
level => $self->{level}->{uncertain}, |
| 504 |
line => 1, column => 1, |
line => 1, column => 1, |
| 505 |
layer => 'encode'); |
layer => 'encode'); |
| 506 |
|
} else { |
| 507 |
|
$self->{input_encoding} = $charset->get_iana_name; |
| 508 |
} |
} |
| 509 |
|
|
| 510 |
$self->{change_encoding} = sub { |
$self->{change_encoding} = sub { |
| 512 |
$charset_name = shift; |
$charset_name = shift; |
| 513 |
my $token = shift; |
my $token = shift; |
| 514 |
|
|
| 515 |
$charset = Message::Charset::Info->get_by_iana_name ($charset_name); |
$charset = Message::Charset::Info->get_by_html_name ($charset_name); |
| 516 |
($char_stream, $e_status) = $charset->get_decode_handle |
($char_stream, $e_status) = $charset->get_decode_handle |
| 517 |
($byte_stream, allow_error_reporting => 1, allow_fallback => 1, |
($byte_stream, allow_error_reporting => 1, allow_fallback => 1, |
| 518 |
byte_buffer => \ $buffer->{buffer}); |
byte_buffer => \ $buffer->{buffer}); |
| 523 |
## Step 1 |
## Step 1 |
| 524 |
if ($charset->{category} & |
if ($charset->{category} & |
| 525 |
Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) { |
| 526 |
$charset = Message::Charset::Info->get_by_iana_name ('utf-8'); |
$charset = Message::Charset::Info->get_by_html_name ('utf-8'); |
| 527 |
($char_stream, $e_status) = $charset->get_decode_handle |
($char_stream, $e_status) = $charset->get_decode_handle |
| 528 |
($byte_stream, |
($byte_stream, |
| 529 |
byte_buffer => \ $buffer->{buffer}); |
byte_buffer => \ $buffer->{buffer}); |
| 567 |
${$opt{octets}} = "\x{FFFD}"; # relacement character |
${$opt{octets}} = "\x{FFFD}"; # relacement character |
| 568 |
} |
} |
| 569 |
}; |
}; |
| 570 |
$char_stream->onerror ($char_onerror); |
|
| 571 |
|
my $wrapped_char_stream = $get_wrapper->($char_stream); |
| 572 |
|
$wrapped_char_stream->onerror ($char_onerror); |
| 573 |
|
|
| 574 |
my @args = @_; shift @args; # $s |
my @args = @_; shift @args; # $s |
| 575 |
my $return; |
my $return; |
| 576 |
try { |
try { |
| 577 |
$return = $self->parse_char_stream ($char_stream, @args); |
$return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 578 |
} catch Whatpm::HTML::RestartParser with { |
} catch Whatpm::HTML::RestartParser with { |
| 579 |
## NOTE: Invoked after {change_encoding}. |
## NOTE: Invoked after {change_encoding}. |
| 580 |
|
|
|
$self->{input_encoding} = $charset->get_iana_name; |
|
| 581 |
if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) { |
| 582 |
|
$self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name? |
| 583 |
!!!parse-error (type => 'chardecode:fallback', |
!!!parse-error (type => 'chardecode:fallback', |
|
text => $self->{input_encoding}, |
|
| 584 |
level => $self->{level}->{uncertain}, |
level => $self->{level}->{uncertain}, |
| 585 |
|
#text => $self->{input_encoding}, |
| 586 |
line => 1, column => 1, |
line => 1, column => 1, |
| 587 |
layer => 'encode'); |
layer => 'encode'); |
| 588 |
} elsif (not ($e_status & |
} elsif (not ($e_status & |
| 589 |
Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL())) { |
Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL())) { |
| 590 |
|
$self->{input_encoding} = $charset->get_iana_name; |
| 591 |
!!!parse-error (type => 'chardecode:no error', |
!!!parse-error (type => 'chardecode:no error', |
| 592 |
text => $self->{input_encoding}, |
text => $self->{input_encoding}, |
| 593 |
level => $self->{level}->{uncertain}, |
level => $self->{level}->{uncertain}, |
| 594 |
line => 1, column => 1, |
line => 1, column => 1, |
| 595 |
layer => 'encode'); |
layer => 'encode'); |
| 596 |
|
} else { |
| 597 |
|
$self->{input_encoding} = $charset->get_iana_name; |
| 598 |
} |
} |
| 599 |
$self->{confident} = 1; |
$self->{confident} = 1; |
| 600 |
$char_stream->onerror ($char_onerror); |
|
| 601 |
$return = $self->parse_char_stream ($char_stream, @args); |
$wrapped_char_stream = $get_wrapper->($char_stream); |
| 602 |
|
$wrapped_char_stream->onerror ($char_onerror); |
| 603 |
|
|
| 604 |
|
$return = $self->parse_char_stream ($wrapped_char_stream, @args); |
| 605 |
}; |
}; |
| 606 |
return $return; |
return $return; |
| 607 |
} # parse_byte_stream |
} # parse_byte_stream |
| 615 |
## such as |parse_byte_string| in this module, must ensure that it does |
## such as |parse_byte_string| in this module, must ensure that it does |
| 616 |
## strip the BOM and never strip any ZWNBSP. |
## strip the BOM and never strip any ZWNBSP. |
| 617 |
|
|
| 618 |
sub parse_char_string ($$$;$) { |
sub parse_char_string ($$$;$$) { |
| 619 |
|
#my ($self, $s, $doc, $onerror, $get_wrapper) = @_; |
| 620 |
my $self = shift; |
my $self = shift; |
| 621 |
require utf8; |
require utf8; |
| 622 |
my $s = ref $_[0] ? $_[0] : \($_[0]); |
my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 623 |
open my $input, '<' . (utf8::is_utf8 ($$s) ? ':utf8' : ''), $s; |
open my $input, '<' . (utf8::is_utf8 ($$s) ? ':utf8' : ''), $s; |
| 624 |
|
if ($_[3]) { |
| 625 |
|
$input = $_[3]->($input); |
| 626 |
|
} |
| 627 |
return $self->parse_char_stream ($input, @_[1..$#_]); |
return $self->parse_char_stream ($input, @_[1..$#_]); |
| 628 |
} # parse_char_string |
} # parse_char_string |
| 629 |
*parse_string = \&parse_char_string; |
*parse_string = \&parse_char_string; ## NOTE: Alias for backward compatibility. |
| 630 |
|
|
| 631 |
sub parse_char_stream ($$$;$) { |
sub parse_char_stream ($$$;$) { |
| 632 |
my $self = ref $_[0] ? shift : shift->new; |
my $self = ref $_[0] ? shift : shift->new; |
| 736 |
my $class = shift; |
my $class = shift; |
| 737 |
my $self = bless { |
my $self = bless { |
| 738 |
level => {must => 'm', |
level => {must => 'm', |
| 739 |
|
should => 's', |
| 740 |
warn => 'w', |
warn => 'w', |
| 741 |
info => 'i', |
info => 'i', |
| 742 |
uncertain => 'u'}, |
uncertain => 'u'}, |
| 803 |
sub BOGUS_DOCTYPE_STATE () { 32 } |
sub BOGUS_DOCTYPE_STATE () { 32 } |
| 804 |
sub AFTER_ATTRIBUTE_VALUE_QUOTED_STATE () { 33 } |
sub AFTER_ATTRIBUTE_VALUE_QUOTED_STATE () { 33 } |
| 805 |
sub SELF_CLOSING_START_TAG_STATE () { 34 } |
sub SELF_CLOSING_START_TAG_STATE () { 34 } |
| 806 |
sub CDATA_BLOCK_STATE () { 35 } |
sub CDATA_SECTION_STATE () { 35 } |
| 807 |
|
sub MD_HYPHEN_STATE () { 36 } # "markup declaration open state" in the spec |
| 808 |
|
sub MD_DOCTYPE_STATE () { 37 } # "markup declaration open state" in the spec |
| 809 |
|
sub MD_CDATA_STATE () { 38 } # "markup declaration open state" in the spec |
| 810 |
|
sub CDATA_PCDATA_CLOSE_TAG_STATE () { 39 } # "close tag open state" in the spec |
| 811 |
|
sub CDATA_SECTION_MSE1_STATE () { 40 } # "CDATA section state" in the spec |
| 812 |
|
sub CDATA_SECTION_MSE2_STATE () { 41 } # "CDATA section state" in the spec |
| 813 |
|
sub PUBLIC_STATE () { 42 } # "after DOCTYPE name state" in the spec |
| 814 |
|
sub SYSTEM_STATE () { 43 } # "after DOCTYPE name state" in the spec |
| 815 |
|
sub ENTITY_STATE () { 44 } # "consume a character reference" in the spec |
| 816 |
|
|
| 817 |
sub DOCTYPE_TOKEN () { 1 } |
sub DOCTYPE_TOKEN () { 1 } |
| 818 |
sub COMMENT_TOKEN () { 2 } |
sub COMMENT_TOKEN () { 2 } |
| 865 |
sub _initialize_tokenizer ($) { |
sub _initialize_tokenizer ($) { |
| 866 |
my $self = shift; |
my $self = shift; |
| 867 |
$self->{state} = DATA_STATE; # MUST |
$self->{state} = DATA_STATE; # MUST |
| 868 |
|
#$self->{state_keyword}; # initialized when used |
| 869 |
$self->{content_model} = PCDATA_CONTENT_MODEL; # be |
$self->{content_model} = PCDATA_CONTENT_MODEL; # be |
| 870 |
undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE |
undef $self->{current_token}; |
| 871 |
undef $self->{current_attribute}; |
undef $self->{current_attribute}; |
| 872 |
undef $self->{last_emitted_start_tag_name}; |
undef $self->{last_emitted_start_tag_name}; |
| 873 |
undef $self->{last_attribute_value_state}; |
undef $self->{last_attribute_value_state}; |
| 941 |
if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA |
if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA |
| 942 |
not $self->{escape}) { |
not $self->{escape}) { |
| 943 |
!!!cp (1); |
!!!cp (1); |
| 944 |
$self->{state} = ENTITY_DATA_STATE; |
## NOTE: In the spec, the tokenizer is switched to the |
| 945 |
|
## "entity data state". In this implementation, the tokenizer |
| 946 |
|
## is switched to the |ENTITY_STATE|, which is an implementation |
| 947 |
|
## of the "consume a character reference" algorithm. |
| 948 |
|
#$self->{state} = ENTITY_DATA_STATE; |
| 949 |
|
$self->{entity_in_attr} = 0; |
| 950 |
|
$self->{entity_additional} = -1; |
| 951 |
|
$self->{state} = ENTITY_STATE; |
| 952 |
!!!next-input-character; |
!!!next-input-character; |
| 953 |
redo A; |
redo A; |
| 954 |
} else { |
} else { |
| 1019 |
|
|
| 1020 |
redo A; |
redo A; |
| 1021 |
} elsif ($self->{state} == ENTITY_DATA_STATE) { |
} elsif ($self->{state} == ENTITY_DATA_STATE) { |
|
## (cannot happen in CDATA state) |
|
|
|
|
| 1022 |
my ($l, $c) = ($self->{line_prev}, $self->{column_prev}); |
my ($l, $c) = ($self->{line_prev}, $self->{column_prev}); |
| 1023 |
|
|
| 1024 |
my $token = $self->_tokenize_attempt_to_consume_an_entity (0, -1); |
my $token = $self->{entity_return}; |
| 1025 |
|
|
| 1026 |
$self->{state} = DATA_STATE; |
$self->{state} = DATA_STATE; |
| 1027 |
# next-input-character is already done |
# next-input-character is already done |
| 1133 |
die "$0: $self->{content_model} in tag open"; |
die "$0: $self->{content_model} in tag open"; |
| 1134 |
} |
} |
| 1135 |
} elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) { |
} elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) { |
| 1136 |
|
## NOTE: The "close tag open state" in the spec is implemented as |
| 1137 |
|
## |CLOSE_TAG_OPEN_STATE| and |CDATA_PCDATA_CLOSE_TAG_STATE|. |
| 1138 |
|
|
| 1139 |
my ($l, $c) = ($self->{line_prev}, $self->{column_prev} - 1); # "<"of"</" |
my ($l, $c) = ($self->{line_prev}, $self->{column_prev} - 1); # "<"of"</" |
| 1140 |
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA |
| 1141 |
if (defined $self->{last_emitted_start_tag_name}) { |
if (defined $self->{last_emitted_start_tag_name}) { |
| 1142 |
|
$self->{state} = CDATA_PCDATA_CLOSE_TAG_STATE; |
| 1143 |
## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564> |
$self->{state_keyword} = ''; |
| 1144 |
my @next_char; |
## Reconsume. |
| 1145 |
TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) { |
redo A; |
|
push @next_char, $self->{next_char}; |
|
|
my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1); |
|
|
my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c; |
|
|
if ($self->{next_char} == $c or $self->{next_char} == $C) { |
|
|
!!!cp (24); |
|
|
!!!next-input-character; |
|
|
next TAGNAME; |
|
|
} else { |
|
|
!!!cp (25); |
|
|
$self->{next_char} = shift @next_char; # reconsume |
|
|
!!!back-next-input-character (@next_char); |
|
|
$self->{state} = DATA_STATE; |
|
|
|
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '</', |
|
|
line => $l, column => $c, |
|
|
}); |
|
|
|
|
|
redo A; |
|
|
} |
|
|
} |
|
|
push @next_char, $self->{next_char}; |
|
|
|
|
|
unless ($self->{next_char} == 0x0009 or # HT |
|
|
$self->{next_char} == 0x000A or # LF |
|
|
$self->{next_char} == 0x000B or # VT |
|
|
$self->{next_char} == 0x000C or # FF |
|
|
$self->{next_char} == 0x0020 or # SP |
|
|
$self->{next_char} == 0x003E or # > |
|
|
$self->{next_char} == 0x002F or # / |
|
|
$self->{next_char} == -1) { |
|
|
!!!cp (26); |
|
|
$self->{next_char} = shift @next_char; # reconsume |
|
|
!!!back-next-input-character (@next_char); |
|
|
$self->{state} = DATA_STATE; |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => '</', |
|
|
line => $l, column => $c, |
|
|
}); |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (27); |
|
|
$self->{next_char} = shift @next_char; |
|
|
!!!back-next-input-character (@next_char); |
|
|
# and consume... |
|
|
} |
|
| 1146 |
} else { |
} else { |
| 1147 |
## No start tag token has ever been emitted |
## No start tag token has ever been emitted |
| 1148 |
|
## NOTE: See <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>. |
| 1149 |
!!!cp (28); |
!!!cp (28); |
|
# next-input-character is already done |
|
| 1150 |
$self->{state} = DATA_STATE; |
$self->{state} = DATA_STATE; |
| 1151 |
|
## Reconsume. |
| 1152 |
!!!emit ({type => CHARACTER_TOKEN, data => '</', |
!!!emit ({type => CHARACTER_TOKEN, data => '</', |
| 1153 |
line => $l, column => $c, |
line => $l, column => $c, |
| 1154 |
}); |
}); |
| 1155 |
redo A; |
redo A; |
| 1156 |
} |
} |
| 1157 |
} |
} |
| 1158 |
|
|
| 1159 |
if (0x0041 <= $self->{next_char} and |
if (0x0041 <= $self->{next_char} and |
| 1160 |
$self->{next_char} <= 0x005A) { # A..Z |
$self->{next_char} <= 0x005A) { # A..Z |
| 1161 |
!!!cp (29); |
!!!cp (29); |
| 1202 |
line => $self->{line_prev}, # "<" of "</" |
line => $self->{line_prev}, # "<" of "</" |
| 1203 |
column => $self->{column_prev} - 1, |
column => $self->{column_prev} - 1, |
| 1204 |
}; |
}; |
| 1205 |
## $self->{next_char} is intentionally left as is |
## NOTE: $self->{next_char} is intentionally left as is. |
| 1206 |
redo A; |
## Although the "anything else" case of the spec not explicitly |
| 1207 |
|
## states that the next input character is to be reconsumed, |
| 1208 |
|
## it will be included to the |data| of the comment token |
| 1209 |
|
## generated from the bogus end tag, as defined in the |
| 1210 |
|
## "bogus comment state" entry. |
| 1211 |
|
redo A; |
| 1212 |
|
} |
| 1213 |
|
} elsif ($self->{state} == CDATA_PCDATA_CLOSE_TAG_STATE) { |
| 1214 |
|
my $ch = substr $self->{last_emitted_start_tag_name}, length $self->{state_keyword}, 1; |
| 1215 |
|
if (length $ch) { |
| 1216 |
|
my $CH = $ch; |
| 1217 |
|
$ch =~ tr/a-z/A-Z/; |
| 1218 |
|
my $nch = chr $self->{next_char}; |
| 1219 |
|
if ($nch eq $ch or $nch eq $CH) { |
| 1220 |
|
!!!cp (24); |
| 1221 |
|
## Stay in the state. |
| 1222 |
|
$self->{state_keyword} .= $nch; |
| 1223 |
|
!!!next-input-character; |
| 1224 |
|
redo A; |
| 1225 |
|
} else { |
| 1226 |
|
!!!cp (25); |
| 1227 |
|
$self->{state} = DATA_STATE; |
| 1228 |
|
## Reconsume. |
| 1229 |
|
!!!emit ({type => CHARACTER_TOKEN, |
| 1230 |
|
data => '</' . $self->{state_keyword}, |
| 1231 |
|
line => $self->{line_prev}, |
| 1232 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}, |
| 1233 |
|
}); |
| 1234 |
|
redo A; |
| 1235 |
|
} |
| 1236 |
|
} else { # after "<{tag-name}" |
| 1237 |
|
unless ({ |
| 1238 |
|
0x0009 => 1, # HT |
| 1239 |
|
0x000A => 1, # LF |
| 1240 |
|
0x000B => 1, # VT |
| 1241 |
|
0x000C => 1, # FF |
| 1242 |
|
0x0020 => 1, # SP |
| 1243 |
|
0x003E => 1, # > |
| 1244 |
|
0x002F => 1, # / |
| 1245 |
|
-1 => 1, # EOF |
| 1246 |
|
}->{$self->{next_char}}) { |
| 1247 |
|
!!!cp (26); |
| 1248 |
|
## Reconsume. |
| 1249 |
|
$self->{state} = DATA_STATE; |
| 1250 |
|
!!!emit ({type => CHARACTER_TOKEN, |
| 1251 |
|
data => '</' . $self->{state_keyword}, |
| 1252 |
|
line => $self->{line_prev}, |
| 1253 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}, |
| 1254 |
|
}); |
| 1255 |
|
redo A; |
| 1256 |
|
} else { |
| 1257 |
|
!!!cp (27); |
| 1258 |
|
$self->{current_token} |
| 1259 |
|
= {type => END_TAG_TOKEN, |
| 1260 |
|
tag_name => $self->{last_emitted_start_tag_name}, |
| 1261 |
|
line => $self->{line_prev}, |
| 1262 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}}; |
| 1263 |
|
$self->{state} = TAG_NAME_STATE; |
| 1264 |
|
## Reconsume. |
| 1265 |
|
redo A; |
| 1266 |
|
} |
| 1267 |
} |
} |
| 1268 |
} elsif ($self->{state} == TAG_NAME_STATE) { |
} elsif ($self->{state} == TAG_NAME_STATE) { |
| 1269 |
if ($self->{next_char} == 0x0009 or # HT |
if ($self->{next_char} == 0x0009 or # HT |
| 1711 |
} elsif ($self->{next_char} == 0x0026) { # & |
} elsif ($self->{next_char} == 0x0026) { # & |
| 1712 |
!!!cp (96); |
!!!cp (96); |
| 1713 |
$self->{last_attribute_value_state} = $self->{state}; |
$self->{last_attribute_value_state} = $self->{state}; |
| 1714 |
$self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE; |
## NOTE: In the spec, the tokenizer is switched to the |
| 1715 |
|
## "entity in attribute value state". In this implementation, the |
| 1716 |
|
## tokenizer is switched to the |ENTITY_STATE|, which is an |
| 1717 |
|
## implementation of the "consume a character reference" algorithm. |
| 1718 |
|
#$self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE; |
| 1719 |
|
$self->{entity_in_attr} = 1; |
| 1720 |
|
$self->{entity_additional} = 0x0022; # " |
| 1721 |
|
$self->{state} = ENTITY_STATE; |
| 1722 |
!!!next-input-character; |
!!!next-input-character; |
| 1723 |
redo A; |
redo A; |
| 1724 |
} elsif ($self->{next_char} == -1) { |
} elsif ($self->{next_char} == -1) { |
| 1760 |
} elsif ($self->{next_char} == 0x0026) { # & |
} elsif ($self->{next_char} == 0x0026) { # & |
| 1761 |
!!!cp (102); |
!!!cp (102); |
| 1762 |
$self->{last_attribute_value_state} = $self->{state}; |
$self->{last_attribute_value_state} = $self->{state}; |
| 1763 |
$self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE; |
## NOTE: In the spec, the tokenizer is switched to the |
| 1764 |
|
## "entity in attribute value state". In this implementation, the |
| 1765 |
|
## tokenizer is switched to the |ENTITY_STATE|, which is an |
| 1766 |
|
## implementation of the "consume a character reference" algorithm. |
| 1767 |
|
#$self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE; |
| 1768 |
|
$self->{entity_in_attr} = 1; |
| 1769 |
|
$self->{entity_additional} = 0x0027; # ' |
| 1770 |
|
$self->{state} = ENTITY_STATE; |
| 1771 |
!!!next-input-character; |
!!!next-input-character; |
| 1772 |
redo A; |
redo A; |
| 1773 |
} elsif ($self->{next_char} == -1) { |
} elsif ($self->{next_char} == -1) { |
| 1813 |
} elsif ($self->{next_char} == 0x0026) { # & |
} elsif ($self->{next_char} == 0x0026) { # & |
| 1814 |
!!!cp (108); |
!!!cp (108); |
| 1815 |
$self->{last_attribute_value_state} = $self->{state}; |
$self->{last_attribute_value_state} = $self->{state}; |
| 1816 |
$self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE; |
## NOTE: In the spec, the tokenizer is switched to the |
| 1817 |
|
## "entity in attribute value state". In this implementation, the |
| 1818 |
|
## tokenizer is switched to the |ENTITY_STATE|, which is an |
| 1819 |
|
## implementation of the "consume a character reference" algorithm. |
| 1820 |
|
#$self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE; |
| 1821 |
|
$self->{entity_in_attr} = 1; |
| 1822 |
|
$self->{entity_additional} = -1; |
| 1823 |
|
$self->{state} = ENTITY_STATE; |
| 1824 |
!!!next-input-character; |
!!!next-input-character; |
| 1825 |
redo A; |
redo A; |
| 1826 |
} elsif ($self->{next_char} == 0x003E) { # > |
} elsif ($self->{next_char} == 0x003E) { # > |
| 1885 |
redo A; |
redo A; |
| 1886 |
} |
} |
| 1887 |
} elsif ($self->{state} == ENTITY_IN_ATTRIBUTE_VALUE_STATE) { |
} elsif ($self->{state} == ENTITY_IN_ATTRIBUTE_VALUE_STATE) { |
| 1888 |
my $token = $self->_tokenize_attempt_to_consume_an_entity |
my $token = $self->{entity_return}; |
|
(1, |
|
|
$self->{last_attribute_value_state} |
|
|
== ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE ? 0x0022 : # " |
|
|
$self->{last_attribute_value_state} |
|
|
== ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE ? 0x0027 : # ' |
|
|
-1); |
|
| 1889 |
|
|
| 1890 |
unless (defined $token) { |
unless (defined $token) { |
| 1891 |
!!!cp (117); |
!!!cp (117); |
| 2019 |
} |
} |
| 2020 |
} elsif ($self->{state} == BOGUS_COMMENT_STATE) { |
} elsif ($self->{state} == BOGUS_COMMENT_STATE) { |
| 2021 |
## (only happen if PCDATA state) |
## (only happen if PCDATA state) |
|
|
|
|
## NOTE: Set by the previous state |
|
|
#my $token = {type => COMMENT_TOKEN, data => ''}; |
|
| 2022 |
|
|
| 2023 |
BC: { |
## NOTE: Unlike spec's "bogus comment state", this implementation |
| 2024 |
if ($self->{next_char} == 0x003E) { # > |
## consumes characters one-by-one basis. |
| 2025 |
!!!cp (124); |
|
| 2026 |
$self->{state} = DATA_STATE; |
if ($self->{next_char} == 0x003E) { # > |
| 2027 |
!!!next-input-character; |
!!!cp (124); |
| 2028 |
|
$self->{state} = DATA_STATE; |
| 2029 |
!!!emit ($self->{current_token}); # comment |
!!!next-input-character; |
|
|
|
|
redo A; |
|
|
} elsif ($self->{next_char} == -1) { |
|
|
!!!cp (125); |
|
|
$self->{state} = DATA_STATE; |
|
|
## reconsume |
|
|
|
|
|
!!!emit ($self->{current_token}); # comment |
|
| 2030 |
|
|
| 2031 |
redo A; |
!!!emit ($self->{current_token}); # comment |
| 2032 |
} else { |
redo A; |
| 2033 |
!!!cp (126); |
} elsif ($self->{next_char} == -1) { |
| 2034 |
$self->{current_token}->{data} .= chr ($self->{next_char}); # comment |
!!!cp (125); |
| 2035 |
!!!next-input-character; |
$self->{state} = DATA_STATE; |
| 2036 |
redo BC; |
## reconsume |
|
} |
|
|
} # BC |
|
| 2037 |
|
|
| 2038 |
die "$0: _get_next_token: unexpected case [BC]"; |
!!!emit ($self->{current_token}); # comment |
| 2039 |
|
redo A; |
| 2040 |
|
} else { |
| 2041 |
|
!!!cp (126); |
| 2042 |
|
$self->{current_token}->{data} .= chr ($self->{next_char}); # comment |
| 2043 |
|
## Stay in the state. |
| 2044 |
|
!!!next-input-character; |
| 2045 |
|
redo A; |
| 2046 |
|
} |
| 2047 |
} elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) { |
} elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) { |
| 2048 |
## (only happen if PCDATA state) |
## (only happen if PCDATA state) |
|
|
|
|
my ($l, $c) = ($self->{line_prev}, $self->{column_prev} - 1); |
|
|
|
|
|
my @next_char; |
|
|
push @next_char, $self->{next_char}; |
|
| 2049 |
|
|
| 2050 |
if ($self->{next_char} == 0x002D) { # - |
if ($self->{next_char} == 0x002D) { # - |
| 2051 |
|
!!!cp (133); |
| 2052 |
|
$self->{state} = MD_HYPHEN_STATE; |
| 2053 |
!!!next-input-character; |
!!!next-input-character; |
| 2054 |
push @next_char, $self->{next_char}; |
redo A; |
|
if ($self->{next_char} == 0x002D) { # - |
|
|
!!!cp (127); |
|
|
$self->{current_token} = {type => COMMENT_TOKEN, data => '', |
|
|
line => $l, column => $c, |
|
|
}; |
|
|
$self->{state} = COMMENT_START_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (128); |
|
|
} |
|
| 2055 |
} elsif ($self->{next_char} == 0x0044 or # D |
} elsif ($self->{next_char} == 0x0044 or # D |
| 2056 |
$self->{next_char} == 0x0064) { # d |
$self->{next_char} == 0x0064) { # d |
| 2057 |
|
## ASCII case-insensitive. |
| 2058 |
|
!!!cp (130); |
| 2059 |
|
$self->{state} = MD_DOCTYPE_STATE; |
| 2060 |
|
$self->{state_keyword} = chr $self->{next_char}; |
| 2061 |
!!!next-input-character; |
!!!next-input-character; |
| 2062 |
push @next_char, $self->{next_char}; |
redo A; |
|
if ($self->{next_char} == 0x004F or # O |
|
|
$self->{next_char} == 0x006F) { # o |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0043 or # C |
|
|
$self->{next_char} == 0x0063) { # c |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0054 or # T |
|
|
$self->{next_char} == 0x0074) { # t |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0059 or # Y |
|
|
$self->{next_char} == 0x0079) { # y |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0050 or # P |
|
|
$self->{next_char} == 0x0070) { # p |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0045 or # E |
|
|
$self->{next_char} == 0x0065) { # e |
|
|
!!!cp (129); |
|
|
## TODO: What a stupid code this is! |
|
|
$self->{state} = DOCTYPE_STATE; |
|
|
$self->{current_token} = {type => DOCTYPE_TOKEN, |
|
|
quirks => 1, |
|
|
line => $l, column => $c, |
|
|
}; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (130); |
|
|
} |
|
|
} else { |
|
|
!!!cp (131); |
|
|
} |
|
|
} else { |
|
|
!!!cp (132); |
|
|
} |
|
|
} else { |
|
|
!!!cp (133); |
|
|
} |
|
|
} else { |
|
|
!!!cp (134); |
|
|
} |
|
|
} else { |
|
|
!!!cp (135); |
|
|
} |
|
| 2063 |
} elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and |
} elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and |
| 2064 |
$self->{open_elements}->[-1]->[1] & FOREIGN_EL and |
$self->{open_elements}->[-1]->[1] & FOREIGN_EL and |
| 2065 |
$self->{next_char} == 0x005B) { # [ |
$self->{next_char} == 0x005B) { # [ |
| 2066 |
|
!!!cp (135.4); |
| 2067 |
|
$self->{state} = MD_CDATA_STATE; |
| 2068 |
|
$self->{state_keyword} = '['; |
| 2069 |
!!!next-input-character; |
!!!next-input-character; |
| 2070 |
push @next_char, $self->{next_char}; |
redo A; |
|
if ($self->{next_char} == 0x0043) { # C |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0044) { # D |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0041) { # A |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0054) { # T |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x0041) { # A |
|
|
!!!next-input-character; |
|
|
push @next_char, $self->{next_char}; |
|
|
if ($self->{next_char} == 0x005B) { # [ |
|
|
!!!cp (135.1); |
|
|
$self->{state} = CDATA_BLOCK_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (135.2); |
|
|
} |
|
|
} else { |
|
|
!!!cp (135.3); |
|
|
} |
|
|
} else { |
|
|
!!!cp (135.4); |
|
|
} |
|
|
} else { |
|
|
!!!cp (135.5); |
|
|
} |
|
|
} else { |
|
|
!!!cp (135.6); |
|
|
} |
|
|
} else { |
|
|
!!!cp (135.7); |
|
|
} |
|
| 2071 |
} else { |
} else { |
| 2072 |
!!!cp (136); |
!!!cp (136); |
| 2073 |
} |
} |
| 2074 |
|
|
| 2075 |
!!!parse-error (type => 'bogus comment'); |
!!!parse-error (type => 'bogus comment', |
| 2076 |
$self->{next_char} = shift @next_char; |
line => $self->{line_prev}, |
| 2077 |
!!!back-next-input-character (@next_char); |
column => $self->{column_prev} - 1); |
| 2078 |
|
## Reconsume. |
| 2079 |
$self->{state} = BOGUS_COMMENT_STATE; |
$self->{state} = BOGUS_COMMENT_STATE; |
| 2080 |
$self->{current_token} = {type => COMMENT_TOKEN, data => '', |
$self->{current_token} = {type => COMMENT_TOKEN, data => '', |
| 2081 |
line => $l, column => $c, |
line => $self->{line_prev}, |
| 2082 |
|
column => $self->{column_prev} - 1, |
| 2083 |
}; |
}; |
| 2084 |
redo A; |
redo A; |
| 2085 |
|
} elsif ($self->{state} == MD_HYPHEN_STATE) { |
| 2086 |
## ISSUE: typos in spec: chacacters, is is a parse error |
if ($self->{next_char} == 0x002D) { # - |
| 2087 |
## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it? |
!!!cp (127); |
| 2088 |
|
$self->{current_token} = {type => COMMENT_TOKEN, data => '', |
| 2089 |
|
line => $self->{line_prev}, |
| 2090 |
|
column => $self->{column_prev} - 2, |
| 2091 |
|
}; |
| 2092 |
|
$self->{state} = COMMENT_START_STATE; |
| 2093 |
|
!!!next-input-character; |
| 2094 |
|
redo A; |
| 2095 |
|
} else { |
| 2096 |
|
!!!cp (128); |
| 2097 |
|
!!!parse-error (type => 'bogus comment', |
| 2098 |
|
line => $self->{line_prev}, |
| 2099 |
|
column => $self->{column_prev} - 2); |
| 2100 |
|
$self->{state} = BOGUS_COMMENT_STATE; |
| 2101 |
|
## Reconsume. |
| 2102 |
|
$self->{current_token} = {type => COMMENT_TOKEN, |
| 2103 |
|
data => '-', |
| 2104 |
|
line => $self->{line_prev}, |
| 2105 |
|
column => $self->{column_prev} - 2, |
| 2106 |
|
}; |
| 2107 |
|
redo A; |
| 2108 |
|
} |
| 2109 |
|
} elsif ($self->{state} == MD_DOCTYPE_STATE) { |
| 2110 |
|
## ASCII case-insensitive. |
| 2111 |
|
if ($self->{next_char} == [ |
| 2112 |
|
undef, |
| 2113 |
|
0x004F, # O |
| 2114 |
|
0x0043, # C |
| 2115 |
|
0x0054, # T |
| 2116 |
|
0x0059, # Y |
| 2117 |
|
0x0050, # P |
| 2118 |
|
]->[length $self->{state_keyword}] or |
| 2119 |
|
$self->{next_char} == [ |
| 2120 |
|
undef, |
| 2121 |
|
0x006F, # o |
| 2122 |
|
0x0063, # c |
| 2123 |
|
0x0074, # t |
| 2124 |
|
0x0079, # y |
| 2125 |
|
0x0070, # p |
| 2126 |
|
]->[length $self->{state_keyword}]) { |
| 2127 |
|
!!!cp (131); |
| 2128 |
|
## Stay in the state. |
| 2129 |
|
$self->{state_keyword} .= chr $self->{next_char}; |
| 2130 |
|
!!!next-input-character; |
| 2131 |
|
redo A; |
| 2132 |
|
} elsif ((length $self->{state_keyword}) == 6 and |
| 2133 |
|
($self->{next_char} == 0x0045 or # E |
| 2134 |
|
$self->{next_char} == 0x0065)) { # e |
| 2135 |
|
!!!cp (129); |
| 2136 |
|
$self->{state} = DOCTYPE_STATE; |
| 2137 |
|
$self->{current_token} = {type => DOCTYPE_TOKEN, |
| 2138 |
|
quirks => 1, |
| 2139 |
|
line => $self->{line_prev}, |
| 2140 |
|
column => $self->{column_prev} - 7, |
| 2141 |
|
}; |
| 2142 |
|
!!!next-input-character; |
| 2143 |
|
redo A; |
| 2144 |
|
} else { |
| 2145 |
|
!!!cp (132); |
| 2146 |
|
!!!parse-error (type => 'bogus comment', |
| 2147 |
|
line => $self->{line_prev}, |
| 2148 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}); |
| 2149 |
|
$self->{state} = BOGUS_COMMENT_STATE; |
| 2150 |
|
## Reconsume. |
| 2151 |
|
$self->{current_token} = {type => COMMENT_TOKEN, |
| 2152 |
|
data => $self->{state_keyword}, |
| 2153 |
|
line => $self->{line_prev}, |
| 2154 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}, |
| 2155 |
|
}; |
| 2156 |
|
redo A; |
| 2157 |
|
} |
| 2158 |
|
} elsif ($self->{state} == MD_CDATA_STATE) { |
| 2159 |
|
if ($self->{next_char} == { |
| 2160 |
|
'[' => 0x0043, # C |
| 2161 |
|
'[C' => 0x0044, # D |
| 2162 |
|
'[CD' => 0x0041, # A |
| 2163 |
|
'[CDA' => 0x0054, # T |
| 2164 |
|
'[CDAT' => 0x0041, # A |
| 2165 |
|
}->{$self->{state_keyword}}) { |
| 2166 |
|
!!!cp (135.1); |
| 2167 |
|
## Stay in the state. |
| 2168 |
|
$self->{state_keyword} .= chr $self->{next_char}; |
| 2169 |
|
!!!next-input-character; |
| 2170 |
|
redo A; |
| 2171 |
|
} elsif ($self->{state_keyword} eq '[CDATA' and |
| 2172 |
|
$self->{next_char} == 0x005B) { # [ |
| 2173 |
|
!!!cp (135.2); |
| 2174 |
|
$self->{current_token} = {type => CHARACTER_TOKEN, |
| 2175 |
|
data => '', |
| 2176 |
|
line => $self->{line_prev}, |
| 2177 |
|
column => $self->{column_prev} - 7}; |
| 2178 |
|
$self->{state} = CDATA_SECTION_STATE; |
| 2179 |
|
!!!next-input-character; |
| 2180 |
|
redo A; |
| 2181 |
|
} else { |
| 2182 |
|
!!!cp (135.3); |
| 2183 |
|
!!!parse-error (type => 'bogus comment', |
| 2184 |
|
line => $self->{line_prev}, |
| 2185 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}); |
| 2186 |
|
$self->{state} = BOGUS_COMMENT_STATE; |
| 2187 |
|
## Reconsume. |
| 2188 |
|
$self->{current_token} = {type => COMMENT_TOKEN, |
| 2189 |
|
data => $self->{state_keyword}, |
| 2190 |
|
line => $self->{line_prev}, |
| 2191 |
|
column => $self->{column_prev} - 1 - length $self->{state_keyword}, |
| 2192 |
|
}; |
| 2193 |
|
redo A; |
| 2194 |
|
} |
| 2195 |
} elsif ($self->{state} == COMMENT_START_STATE) { |
} elsif ($self->{state} == COMMENT_START_STATE) { |
| 2196 |
if ($self->{next_char} == 0x002D) { # - |
if ($self->{next_char} == 0x002D) { # - |
| 2197 |
!!!cp (137); |
!!!cp (137); |
| 2458 |
redo A; |
redo A; |
| 2459 |
} elsif ($self->{next_char} == 0x0050 or # P |
} elsif ($self->{next_char} == 0x0050 or # P |
| 2460 |
$self->{next_char} == 0x0070) { # p |
$self->{next_char} == 0x0070) { # p |
| 2461 |
|
$self->{state} = PUBLIC_STATE; |
| 2462 |
|
$self->{state_keyword} = chr $self->{next_char}; |
| 2463 |
!!!next-input-character; |
!!!next-input-character; |
| 2464 |
if ($self->{next_char} == 0x0055 or # U |
redo A; |
|
$self->{next_char} == 0x0075) { # u |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x0042 or # B |
|
|
$self->{next_char} == 0x0062) { # b |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x004C or # L |
|
|
$self->{next_char} == 0x006C) { # l |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x0049 or # I |
|
|
$self->{next_char} == 0x0069) { # i |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x0043 or # C |
|
|
$self->{next_char} == 0x0063) { # c |
|
|
!!!cp (168); |
|
|
$self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (169); |
|
|
} |
|
|
} else { |
|
|
!!!cp (170); |
|
|
} |
|
|
} else { |
|
|
!!!cp (171); |
|
|
} |
|
|
} else { |
|
|
!!!cp (172); |
|
|
} |
|
|
} else { |
|
|
!!!cp (173); |
|
|
} |
|
|
|
|
|
# |
|
| 2465 |
} elsif ($self->{next_char} == 0x0053 or # S |
} elsif ($self->{next_char} == 0x0053 or # S |
| 2466 |
$self->{next_char} == 0x0073) { # s |
$self->{next_char} == 0x0073) { # s |
| 2467 |
|
$self->{state} = SYSTEM_STATE; |
| 2468 |
|
$self->{state_keyword} = chr $self->{next_char}; |
| 2469 |
!!!next-input-character; |
!!!next-input-character; |
| 2470 |
if ($self->{next_char} == 0x0059 or # Y |
redo A; |
|
$self->{next_char} == 0x0079) { # y |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x0053 or # S |
|
|
$self->{next_char} == 0x0073) { # s |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x0054 or # T |
|
|
$self->{next_char} == 0x0074) { # t |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x0045 or # E |
|
|
$self->{next_char} == 0x0065) { # e |
|
|
!!!next-input-character; |
|
|
if ($self->{next_char} == 0x004D or # M |
|
|
$self->{next_char} == 0x006D) { # m |
|
|
!!!cp (174); |
|
|
$self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
|
|
!!!next-input-character; |
|
|
redo A; |
|
|
} else { |
|
|
!!!cp (175); |
|
|
} |
|
|
} else { |
|
|
!!!cp (176); |
|
|
} |
|
|
} else { |
|
|
!!!cp (177); |
|
|
} |
|
|
} else { |
|
|
!!!cp (178); |
|
|
} |
|
|
} else { |
|
|
!!!cp (179); |
|
|
} |
|
|
|
|
|
# |
|
| 2471 |
} else { |
} else { |
| 2472 |
!!!cp (180); |
!!!cp (180); |
| 2473 |
|
!!!parse-error (type => 'string after DOCTYPE name'); |
| 2474 |
|
$self->{current_token}->{quirks} = 1; |
| 2475 |
|
|
| 2476 |
|
$self->{state} = BOGUS_DOCTYPE_STATE; |
| 2477 |
!!!next-input-character; |
!!!next-input-character; |
| 2478 |
# |
redo A; |
| 2479 |
} |
} |
| 2480 |
|
} elsif ($self->{state} == PUBLIC_STATE) { |
| 2481 |
|
## ASCII case-insensitive |
| 2482 |
|
if ($self->{next_char} == [ |
| 2483 |
|
undef, |
| 2484 |
|
0x0055, # U |
| 2485 |
|
0x0042, # B |
| 2486 |
|
0x004C, # L |
| 2487 |
|
0x0049, # I |
| 2488 |
|
]->[length $self->{state_keyword}] or |
| 2489 |
|
$self->{next_char} == [ |
| 2490 |
|
undef, |
| 2491 |
|
0x0075, # u |
| 2492 |
|
0x0062, # b |
| 2493 |
|
0x006C, # l |
| 2494 |
|
0x0069, # i |
| 2495 |
|
]->[length $self->{state_keyword}]) { |
| 2496 |
|
!!!cp (175); |
| 2497 |
|
## Stay in the state. |
| 2498 |
|
$self->{state_keyword} .= chr $self->{next_char}; |
| 2499 |
|
!!!next-input-character; |
| 2500 |
|
redo A; |
| 2501 |
|
} elsif ((length $self->{state_keyword}) == 5 and |
| 2502 |
|
($self->{next_char} == 0x0043 or # C |
| 2503 |
|
$self->{next_char} == 0x0063)) { # c |
| 2504 |
|
!!!cp (168); |
| 2505 |
|
$self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE; |
| 2506 |
|
!!!next-input-character; |
| 2507 |
|
redo A; |
| 2508 |
|
} else { |
| 2509 |
|
!!!cp (169); |
| 2510 |
|
!!!parse-error (type => 'string after DOCTYPE name', |
| 2511 |
|
line => $self->{line_prev}, |
| 2512 |
|
column => $self->{column_prev} + 1 - length $self->{state_keyword}); |
| 2513 |
|
$self->{current_token}->{quirks} = 1; |
| 2514 |
|
|
| 2515 |
!!!parse-error (type => 'string after DOCTYPE name'); |
$self->{state} = BOGUS_DOCTYPE_STATE; |
| 2516 |
$self->{current_token}->{quirks} = 1; |
## Reconsume. |
| 2517 |
|
redo A; |
| 2518 |
|
} |
| 2519 |
|
} elsif ($self->{state} == SYSTEM_STATE) { |
| 2520 |
|
## ASCII case-insensitive |
| 2521 |
|
if ($self->{next_char} == [ |
| 2522 |
|
undef, |
| 2523 |
|
0x0059, # Y |
| 2524 |
|
0x0053, # S |
| 2525 |
|
0x0054, # T |
| 2526 |
|
0x0045, # E |
| 2527 |
|
]->[length $self->{state_keyword}] or |
| 2528 |
|
$self->{next_char} == [ |
| 2529 |
|
undef, |
| 2530 |
|
0x0079, # y |
| 2531 |
|
0x0073, # s |
| 2532 |
|
0x0074, # t |
| 2533 |
|
0x0065, # e |
| 2534 |
|
]->[length $self->{state_keyword}]) { |
| 2535 |
|
!!!cp (170); |
| 2536 |
|
## Stay in the state. |
| 2537 |
|
$self->{state_keyword} .= chr $self->{next_char}; |
| 2538 |
|
!!!next-input-character; |
| 2539 |
|
redo A; |
| 2540 |
|
} elsif ((length $self->{state_keyword}) == 5 and |
| 2541 |
|
($self->{next_char} == 0x004D or # M |
| 2542 |
|
$self->{next_char} == 0x006D)) { # m |
| 2543 |
|
!!!cp (171); |
| 2544 |
|
$self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE; |
| 2545 |
|
!!!next-input-character; |
| 2546 |
|
redo A; |
| 2547 |
|
} else { |
| 2548 |
|
!!!cp (172); |
| 2549 |
|
!!!parse-error (type => 'string after DOCTYPE name', |
| 2550 |
|
line => $self->{line_prev}, |
| 2551 |
|
column => $self->{column_prev} + 1 - length $self->{state_keyword}); |
| 2552 |
|
$self->{current_token}->{quirks} = 1; |
| 2553 |
|
|
| 2554 |
$self->{state} = BOGUS_DOCTYPE_STATE; |
$self->{state} = BOGUS_DOCTYPE_STATE; |
| 2555 |
# next-input-character is already done |
## Reconsume. |
| 2556 |
redo A; |
redo A; |
| 2557 |
|
} |
| 2558 |
} elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) { |
} elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) { |
| 2559 |
if ({ |
if ({ |
| 2560 |
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, |
0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1, |
| 2911 |
!!!next-input-character; |
!!!next-input-character; |
| 2912 |
redo A; |
redo A; |
| 2913 |
} |
} |
| 2914 |
} elsif ($self->{state} == CDATA_BLOCK_STATE) { |
} elsif ($self->{state} == CDATA_SECTION_STATE) { |
| 2915 |
my $s = ''; |
## NOTE: "CDATA section state" in the state is jointly implemented |
| 2916 |
|
## by three states, |CDATA_SECTION_STATE|, |CDATA_SECTION_MSE1_STATE|, |
| 2917 |
|
## and |CDATA_SECTION_MSE2_STATE|. |
| 2918 |
|
|
| 2919 |
my ($l, $c) = ($self->{line}, $self->{column}); |
if ($self->{next_char} == 0x005D) { # ] |
| 2920 |
|
!!!cp (221.1); |
| 2921 |
|
$self->{state} = CDATA_SECTION_MSE1_STATE; |
| 2922 |
|
!!!next-input-character; |
| 2923 |
|
redo A; |
| 2924 |
|
} elsif ($self->{next_char} == -1) { |
| 2925 |
|
$self->{state} = DATA_STATE; |
| 2926 |
|
!!!next-input-character; |
| 2927 |
|
if (length $self->{current_token}->{data}) { # character |
| 2928 |
|
!!!cp (221.2); |
| 2929 |
|
!!!emit ($self->{current_token}); # character |
| 2930 |
|
} else { |
| 2931 |
|
!!!cp (221.3); |
| 2932 |
|
## No token to emit. $self->{current_token} is discarded. |
| 2933 |
|
} |
| 2934 |
|
redo A; |
| 2935 |
|
} else { |
| 2936 |
|
!!!cp (221.4); |
| 2937 |
|
$self->{current_token}->{data} .= chr $self->{next_char}; |
| 2938 |
|
## Stay in the state. |
| 2939 |
|
!!!next-input-character; |
| 2940 |
|
redo A; |
| 2941 |
|
} |
| 2942 |
|
|
| 2943 |
CS: while ($self->{next_char} != -1) { |
## ISSUE: "text tokens" in spec. |
| 2944 |
if ($self->{next_char} == 0x005D) { # ] |
} elsif ($self->{state} == CDATA_SECTION_MSE1_STATE) { |
| 2945 |
!!!next-input-character; |
if ($self->{next_char} == 0x005D) { # ] |
| 2946 |
if ($self->{next_char} == 0x005D) { # ] |
!!!cp (221.5); |
| 2947 |
!!!next-input-character; |
$self->{state} = CDATA_SECTION_MSE2_STATE; |
| 2948 |
MDC: { |
!!!next-input-character; |
| 2949 |
if ($self->{next_char} == 0x003E) { # > |
redo A; |
| 2950 |
!!!cp (221.1); |
} else { |
| 2951 |
!!!next-input-character; |
!!!cp (221.6); |
| 2952 |
last CS; |
$self->{current_token}->{data} .= ']'; |
| 2953 |
} elsif ($self->{next_char} == 0x005D) { # ] |
$self->{state} = CDATA_SECTION_STATE; |
| 2954 |
!!!cp (221.2); |
## Reconsume. |
| 2955 |
$s .= ']'; |
redo A; |
| 2956 |
!!!next-input-character; |
} |
| 2957 |
redo MDC; |
} elsif ($self->{state} == CDATA_SECTION_MSE2_STATE) { |
| 2958 |
} else { |
if ($self->{next_char} == 0x003E) { # > |
| 2959 |
!!!cp (221.3); |
$self->{state} = DATA_STATE; |
| 2960 |
$s .= ']]'; |
!!!next-input-character; |
| 2961 |
# |
if (length $self->{current_token}->{data}) { # character |
| 2962 |
} |
!!!cp (221.7); |
| 2963 |
} # MDC |
!!!emit ($self->{current_token}); # character |
|
} else { |
|
|
!!!cp (221.4); |
|
|
$s .= ']'; |
|
|
# |
|
|
} |
|
| 2964 |
} else { |
} else { |
| 2965 |
!!!cp (221.5); |
!!!cp (221.8); |
| 2966 |
# |
## No token to emit. $self->{current_token} is discarded. |
| 2967 |
} |
} |
| 2968 |
$s .= chr $self->{next_char}; |
redo A; |
| 2969 |
|
} elsif ($self->{next_char} == 0x005D) { # ] |
| 2970 |
|
!!!cp (221.9); # character |
| 2971 |
|
$self->{current_token}->{data} .= ']'; ## Add first "]" of "]]]". |
| 2972 |
|
## Stay in the state. |
| 2973 |
!!!next-input-character; |
!!!next-input-character; |
| 2974 |
} # CS |
redo A; |
|
|
|
|
$self->{state} = DATA_STATE; |
|
|
## next-input-character done or EOF, which is reconsumed. |
|
|
|
|
|
if (length $s) { |
|
|
!!!cp (221.6); |
|
|
!!!emit ({type => CHARACTER_TOKEN, data => $s, |
|
|
line => $l, column => $c}); |
|
| 2975 |
} else { |
} else { |
| 2976 |
!!!cp (221.7); |
!!!cp (221.11); |
| 2977 |
|
$self->{current_token}->{data} .= ']]'; # character |
| 2978 |
|
$self->{state} = CDATA_SECTION_STATE; |
| 2979 |
|
## Reconsume. |
| 2980 |
|
redo A; |
| 2981 |
} |
} |
| 2982 |
|
|
| 2983 |
redo A; |
} elsif ($self->{state} == ENTITY_STATE) { |
| 2984 |
|
my $in_attr = $self->{entity_in_attr}; |
| 2985 |
## ISSUE: "text tokens" in spec. |
my $additional = $self->{entity_additional}; |
|
## TODO: Streaming support |
|
|
} else { |
|
|
die "$0: $self->{state}: Unknown state"; |
|
|
} |
|
|
} # A |
|
|
|
|
|
die "$0: _get_next_token: unexpected case"; |
|
|
} # _get_next_token |
|
|
|
|
|
sub _tokenize_attempt_to_consume_an_entity ($$$) { |
|
|
my ($self, $in_attr, $additional) = @_; |
|
| 2986 |
|
|
| 2987 |
my ($l, $c) = ($self->{line_prev}, $self->{column_prev}); |
my ($l, $c) = ($self->{line_prev}, $self->{column_prev}); |
| 2988 |
|
|
| 2994 |
!!!cp (1001); |
!!!cp (1001); |
| 2995 |
## Don't consume |
## Don't consume |
| 2996 |
## No error |
## No error |
| 2997 |
return undef; |
$self->{entity_return} = undef; |
| 2998 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 2999 |
|
redo A; |
| 3000 |
} elsif ($self->{next_char} == 0x0023) { # # |
} elsif ($self->{next_char} == 0x0023) { # # |
| 3001 |
!!!next-input-character; |
!!!next-input-character; |
| 3002 |
if ($self->{next_char} == 0x0078 or # x |
if ($self->{next_char} == 0x0078 or # x |
| 3031 |
!!!parse-error (type => 'bare hcro', line => $l, column => $c); |
!!!parse-error (type => 'bare hcro', line => $l, column => $c); |
| 3032 |
!!!back-next-input-character ($x_char, $self->{next_char}); |
!!!back-next-input-character ($x_char, $self->{next_char}); |
| 3033 |
$self->{next_char} = 0x0023; # # |
$self->{next_char} = 0x0023; # # |
| 3034 |
return undef; |
$self->{entity_return} = undef; |
| 3035 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3036 |
|
redo A; |
| 3037 |
} elsif ($self->{next_char} == 0x003B) { # ; |
} elsif ($self->{next_char} == 0x003B) { # ; |
| 3038 |
!!!cp (1006); |
!!!cp (1006); |
| 3039 |
!!!next-input-character; |
!!!next-input-character; |
| 3064 |
$code = $c1_entity_char->{$code}; |
$code = $c1_entity_char->{$code}; |
| 3065 |
} |
} |
| 3066 |
|
|
| 3067 |
return {type => CHARACTER_TOKEN, data => chr $code, |
$self->{entity_return} = {type => CHARACTER_TOKEN, data => chr $code, |
| 3068 |
has_reference => 1, |
has_reference => 1, |
| 3069 |
line => $l, column => $c, |
line => $l, column => $c, |
| 3070 |
}; |
}; |
| 3071 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3072 |
|
redo A; |
| 3073 |
} # X |
} # X |
| 3074 |
} elsif (0x0030 <= $self->{next_char} and |
} elsif (0x0030 <= $self->{next_char} and |
| 3075 |
$self->{next_char} <= 0x0039) { # 0..9 |
$self->{next_char} <= 0x0039) { # 0..9 |
| 3118 |
$code = $c1_entity_char->{$code}; |
$code = $c1_entity_char->{$code}; |
| 3119 |
} |
} |
| 3120 |
|
|
| 3121 |
return {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1, |
$self->{entity_return} = {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1, |
| 3122 |
line => $l, column => $c, |
line => $l, column => $c, |
| 3123 |
}; |
}; |
| 3124 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3125 |
|
redo A; |
| 3126 |
} else { |
} else { |
| 3127 |
!!!cp (1019); |
!!!cp (1019); |
| 3128 |
!!!parse-error (type => 'bare nero', line => $l, column => $c); |
!!!parse-error (type => 'bare nero', line => $l, column => $c); |
| 3129 |
!!!back-next-input-character ($self->{next_char}); |
!!!back-next-input-character ($self->{next_char}); |
| 3130 |
$self->{next_char} = 0x0023; # # |
$self->{next_char} = 0x0023; # # |
| 3131 |
return undef; |
$self->{entity_return} = undef; |
| 3132 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3133 |
|
redo A; |
| 3134 |
} |
} |
| 3135 |
} elsif ((0x0041 <= $self->{next_char} and |
} elsif ((0x0041 <= $self->{next_char} and |
| 3136 |
$self->{next_char} <= 0x005A) or |
$self->{next_char} <= 0x005A) or |
| 3177 |
|
|
| 3178 |
if ($match > 0) { |
if ($match > 0) { |
| 3179 |
!!!cp (1023); |
!!!cp (1023); |
| 3180 |
return {type => CHARACTER_TOKEN, data => $value, has_reference => 1, |
$self->{entity_return} = {type => CHARACTER_TOKEN, data => $value, has_reference => 1, |
| 3181 |
line => $l, column => $c, |
line => $l, column => $c, |
| 3182 |
}; |
}; |
| 3183 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3184 |
|
redo A; |
| 3185 |
} elsif ($match < 0) { |
} elsif ($match < 0) { |
| 3186 |
!!!parse-error (type => 'no refc', line => $l, column => $c); |
!!!parse-error (type => 'no refc', line => $l, column => $c); |
| 3187 |
if ($in_attr and $match < -1) { |
if ($in_attr and $match < -1) { |
| 3188 |
!!!cp (1024); |
!!!cp (1024); |
| 3189 |
return {type => CHARACTER_TOKEN, data => '&'.$entity_name, |
$self->{entity_return} = {type => CHARACTER_TOKEN, data => '&'.$entity_name, |
| 3190 |
line => $l, column => $c, |
line => $l, column => $c, |
| 3191 |
}; |
}; |
| 3192 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3193 |
|
redo A; |
| 3194 |
} else { |
} else { |
| 3195 |
!!!cp (1025); |
!!!cp (1025); |
| 3196 |
return {type => CHARACTER_TOKEN, data => $value, has_reference => 1, |
$self->{entity_return} = {type => CHARACTER_TOKEN, data => $value, has_reference => 1, |
| 3197 |
line => $l, column => $c, |
line => $l, column => $c, |
| 3198 |
}; |
}; |
| 3199 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3200 |
|
redo A; |
| 3201 |
} |
} |
| 3202 |
} else { |
} else { |
| 3203 |
!!!cp (1026); |
!!!cp (1026); |
| 3204 |
!!!parse-error (type => 'bare ero', line => $l, column => $c); |
!!!parse-error (type => 'bare ero', line => $l, column => $c); |
| 3205 |
## NOTE: "No characters are consumed" in the spec. |
## NOTE: "No characters are consumed" in the spec. |
| 3206 |
return {type => CHARACTER_TOKEN, data => '&'.$value, |
$self->{entity_return} = {type => CHARACTER_TOKEN, data => '&'.$value, |
| 3207 |
line => $l, column => $c, |
line => $l, column => $c, |
| 3208 |
}; |
}; |
| 3209 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3210 |
|
redo A; |
| 3211 |
} |
} |
| 3212 |
} else { |
} else { |
| 3213 |
!!!cp (1027); |
!!!cp (1027); |
| 3214 |
## no characters are consumed |
## no characters are consumed |
| 3215 |
!!!parse-error (type => 'bare ero', line => $l, column => $c); |
!!!parse-error (type => 'bare ero', line => $l, column => $c); |
| 3216 |
return undef; |
$self->{entity_return} = undef; |
| 3217 |
|
$self->{state} = $self->{entity_in_attr} ? ENTITY_IN_ATTRIBUTE_VALUE_STATE : ENTITY_DATA_STATE; |
| 3218 |
|
redo A; |
| 3219 |
} |
} |
| 3220 |
} # _tokenize_attempt_to_consume_an_entity |
|
| 3221 |
|
} else { |
| 3222 |
|
die "$0: $self->{state}: Unknown state"; |
| 3223 |
|
} |
| 3224 |
|
} # A |
| 3225 |
|
|
| 3226 |
|
die "$0: _get_next_token: unexpected case"; |
| 3227 |
|
} # _get_next_token |
| 3228 |
|
|
| 3229 |
sub _initialize_tree_constructor ($) { |
sub _initialize_tree_constructor ($) { |
| 3230 |
my $self = shift; |
my $self = shift; |
| 3289 |
## language. |
## language. |
| 3290 |
my $doctype_name = $token->{name}; |
my $doctype_name = $token->{name}; |
| 3291 |
$doctype_name = '' unless defined $doctype_name; |
$doctype_name = '' unless defined $doctype_name; |
| 3292 |
$doctype_name =~ tr/a-z/A-Z/; |
$doctype_name =~ tr/a-z/A-Z/; # ASCII case-insensitive |
| 3293 |
if (not defined $token->{name} or # <!DOCTYPE> |
if (not defined $token->{name} or # <!DOCTYPE> |
|
defined $token->{public_identifier} or |
|
| 3294 |
defined $token->{system_identifier}) { |
defined $token->{system_identifier}) { |
| 3295 |
!!!cp ('t1'); |
!!!cp ('t1'); |
| 3296 |
!!!parse-error (type => 'not HTML5', token => $token); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 3297 |
} elsif ($doctype_name ne 'HTML') { |
} elsif ($doctype_name ne 'HTML') { |
| 3298 |
!!!cp ('t2'); |
!!!cp ('t2'); |
|
## ISSUE: ASCII case-insensitive? (in fact it does not matter) |
|
| 3299 |
!!!parse-error (type => 'not HTML5', token => $token); |
!!!parse-error (type => 'not HTML5', token => $token); |
| 3300 |
|
} elsif (defined $token->{public_identifier}) { |
| 3301 |
|
if ($token->{public_identifier} eq 'XSLT-compat') { |
| 3302 |
|
!!!cp ('t1.2'); |
| 3303 |
|
!!!parse-error (type => 'XSLT-compat', token => $token, |
| 3304 |
|
level => $self->{level}->{should}); |
| 3305 |
|
} else { |
| 3306 |
|
!!!parse-error (type => 'not HTML5', token => $token); |
| 3307 |
|
} |
| 3308 |
} else { |
} else { |
| 3309 |
!!!cp ('t3'); |
!!!cp ('t3'); |
| 3310 |
|
# |
| 3311 |
} |
} |
| 3312 |
|
|
| 3313 |
my $doctype = $self->{document}->create_document_type_definition |
my $doctype = $self->{document}->create_document_type_definition |
| 6443 |
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 6444 |
!!!cp ('t312'); |
!!!cp ('t312'); |
| 6445 |
!!!parse-error (type => 'after frameset:#text', token => $token); |
!!!parse-error (type => 'after frameset:#text', token => $token); |
| 6446 |
} else { # "after html frameset" |
} else { # "after after frameset" |
| 6447 |
!!!cp ('t313'); |
!!!cp ('t313'); |
| 6448 |
!!!parse-error (type => 'after html:#text', token => $token); |
!!!parse-error (type => 'after html:#text', token => $token); |
|
|
|
|
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
|
|
## Reprocess in the "after frameset" insertion mode. |
|
|
!!!parse-error (type => 'after frameset:#text', token => $token); |
|
| 6449 |
} |
} |
| 6450 |
|
|
| 6451 |
## Ignore the token. |
## Ignore the token. |
| 6461 |
|
|
| 6462 |
die qq[$0: Character "$token->{data}"]; |
die qq[$0: Character "$token->{data}"]; |
| 6463 |
} elsif ($token->{type} == START_TAG_TOKEN) { |
} elsif ($token->{type} == START_TAG_TOKEN) { |
|
if ($self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) { |
|
|
!!!cp ('t316'); |
|
|
!!!parse-error (type => 'after html', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
|
|
|
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
|
|
## Process in the "after frameset" insertion mode. |
|
|
} else { |
|
|
!!!cp ('t317'); |
|
|
} |
|
|
|
|
| 6464 |
if ($token->{tag_name} eq 'frameset' and |
if ($token->{tag_name} eq 'frameset' and |
| 6465 |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
| 6466 |
!!!cp ('t318'); |
!!!cp ('t318'); |
| 6481 |
## NOTE: As if in head. |
## NOTE: As if in head. |
| 6482 |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
$parse_rcdata->(CDATA_CONTENT_MODEL); |
| 6483 |
next B; |
next B; |
| 6484 |
|
|
| 6485 |
|
## NOTE: |<!DOCTYPE HTML><frameset></frameset></html><noframes></noframes>| |
| 6486 |
|
## has no parse error. |
| 6487 |
} else { |
} else { |
| 6488 |
if ($self->{insertion_mode} == IN_FRAMESET_IM) { |
if ($self->{insertion_mode} == IN_FRAMESET_IM) { |
| 6489 |
!!!cp ('t321'); |
!!!cp ('t321'); |
| 6490 |
!!!parse-error (type => 'in frameset', |
!!!parse-error (type => 'in frameset', |
| 6491 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 6492 |
} else { |
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 6493 |
!!!cp ('t322'); |
!!!cp ('t322'); |
| 6494 |
!!!parse-error (type => 'after frameset', |
!!!parse-error (type => 'after frameset', |
| 6495 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 6496 |
|
} else { # "after after frameset" |
| 6497 |
|
!!!cp ('t322.2'); |
| 6498 |
|
!!!parse-error (type => 'after after frameset', |
| 6499 |
|
text => $token->{tag_name}, token => $token); |
| 6500 |
} |
} |
| 6501 |
## Ignore the token |
## Ignore the token |
| 6502 |
!!!nack ('t322.1'); |
!!!nack ('t322.1'); |
| 6504 |
next B; |
next B; |
| 6505 |
} |
} |
| 6506 |
} elsif ($token->{type} == END_TAG_TOKEN) { |
} elsif ($token->{type} == END_TAG_TOKEN) { |
|
if ($self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) { |
|
|
!!!cp ('t323'); |
|
|
!!!parse-error (type => 'after html:/', |
|
|
text => $token->{tag_name}, token => $token); |
|
|
|
|
|
$self->{insertion_mode} = AFTER_FRAMESET_IM; |
|
|
## Process in the "after frameset" insertion mode. |
|
|
} else { |
|
|
!!!cp ('t324'); |
|
|
} |
|
|
|
|
| 6507 |
if ($token->{tag_name} eq 'frameset' and |
if ($token->{tag_name} eq 'frameset' and |
| 6508 |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
$self->{insertion_mode} == IN_FRAMESET_IM) { |
| 6509 |
if ($self->{open_elements}->[-1]->[1] & HTML_EL and |
if ($self->{open_elements}->[-1]->[1] & HTML_EL and |
| 6538 |
!!!cp ('t330'); |
!!!cp ('t330'); |
| 6539 |
!!!parse-error (type => 'in frameset:/', |
!!!parse-error (type => 'in frameset:/', |
| 6540 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 6541 |
} else { |
} elsif ($self->{insertion_mode} == AFTER_FRAMESET_IM) { |
| 6542 |
!!!cp ('t331'); |
!!!cp ('t330.1'); |
| 6543 |
!!!parse-error (type => 'after frameset:/', |
!!!parse-error (type => 'after frameset:/', |
| 6544 |
text => $token->{tag_name}, token => $token); |
text => $token->{tag_name}, token => $token); |
| 6545 |
|
} else { # "after after html" |
| 6546 |
|
!!!cp ('t331'); |
| 6547 |
|
!!!parse-error (type => 'after after frameset:/', |
| 6548 |
|
text => $token->{tag_name}, token => $token); |
| 6549 |
} |
} |
| 6550 |
## Ignore the token |
## Ignore the token |
| 6551 |
!!!next-token; |
!!!next-token; |
| 7578 |
## TODO: script stuffs |
## TODO: script stuffs |
| 7579 |
} # _tree_construct_main |
} # _tree_construct_main |
| 7580 |
|
|
| 7581 |
sub set_inner_html ($$$) { |
sub set_inner_html ($$$;$) { |
| 7582 |
my $class = shift; |
my $class = shift; |
| 7583 |
my $node = shift; |
my $node = shift; |
| 7584 |
my $s = \$_[0]; |
my $s = \$_[0]; |
| 7585 |
my $onerror = $_[1]; |
my $onerror = $_[1]; |
| 7586 |
|
my $get_wrapper = $_[2] || sub ($) { return $_[0] }; |
| 7587 |
|
|
| 7588 |
## ISSUE: Should {confident} be true? |
## ISSUE: Should {confident} be true? |
| 7589 |
|
|
| 7602 |
} |
} |
| 7603 |
|
|
| 7604 |
## Step 3, 4, 5 # MUST |
## Step 3, 4, 5 # MUST |
| 7605 |
$class->parse_string ($$s => $node, $onerror); |
$class->parse_char_string ($$s => $node, $onerror, $get_wrapper); |
| 7606 |
} elsif ($nt == 1) { |
} elsif ($nt == 1) { |
| 7607 |
## TODO: If non-html element |
## TODO: If non-html element |
| 7608 |
|
|
| 7609 |
## NOTE: Most of this code is copied from |parse_string| |
## NOTE: Most of this code is copied from |parse_string| |
| 7610 |
|
|
| 7611 |
|
## TODO: Support for $get_wrapper |
| 7612 |
|
|
| 7613 |
## Step 1 # MUST |
## Step 1 # MUST |
| 7614 |
my $this_doc = $node->owner_document; |
my $this_doc = $node->owner_document; |
| 7615 |
my $doc = $this_doc->implementation->create_document; |
my $doc = $this_doc->implementation->create_document; |