| 618 |
sub parse_char_string ($$$;$$) { |
sub parse_char_string ($$$;$$) { |
| 619 |
#my ($self, $s, $doc, $onerror, $get_wrapper) = @_; |
#my ($self, $s, $doc, $onerror, $get_wrapper) = @_; |
| 620 |
my $self = shift; |
my $self = shift; |
|
require utf8; |
|
| 621 |
my $s = ref $_[0] ? $_[0] : \($_[0]); |
my $s = ref $_[0] ? $_[0] : \($_[0]); |
| 622 |
open my $input, '<' . (utf8::is_utf8 ($$s) ? ':utf8' : ''), $s; |
require Whatpm::Charset::DecodeHandle; |
| 623 |
|
my $input = Whatpm::Charset::DecodeHandle::CharString->new ($s); |
| 624 |
if ($_[3]) { |
if ($_[3]) { |
| 625 |
$input = $_[3]->($input); |
$input = $_[3]->($input); |
| 626 |
} |
} |
| 689 |
(0x007F <= $self->{next_char} and $self->{next_char} <= 0x009F) or |
(0x007F <= $self->{next_char} and $self->{next_char} <= 0x009F) or |
| 690 |
(0xD800 <= $self->{next_char} and $self->{next_char} <= 0xDFFF) or |
(0xD800 <= $self->{next_char} and $self->{next_char} <= 0xDFFF) or |
| 691 |
(0xFDD0 <= $self->{next_char} and $self->{next_char} <= 0xFDDF) or |
(0xFDD0 <= $self->{next_char} and $self->{next_char} <= 0xFDDF) or |
| 692 |
|
## ISSUE: U+FDE0-U+FDEF are not excluded |
| 693 |
{ |
{ |
| 694 |
0xFFFE => 1, 0xFFFF => 1, 0x1FFFE => 1, 0x1FFFF => 1, |
0xFFFE => 1, 0xFFFF => 1, 0x1FFFE => 1, 0x1FFFF => 1, |
| 695 |
0x2FFFE => 1, 0x2FFFF => 1, 0x3FFFE => 1, 0x3FFFF => 1, |
0x2FFFE => 1, 0x2FFFF => 1, 0x3FFFE => 1, 0x3FFFF => 1, |
| 714 |
$self->{prev_char} = [-1, -1, -1]; |
$self->{prev_char} = [-1, -1, -1]; |
| 715 |
$self->{next_char} = -1; |
$self->{next_char} = -1; |
| 716 |
|
|
| 717 |
|
$self->{read_until} = sub { |
| 718 |
|
#my ($scalar, $specials_range, $offset) = @_; |
| 719 |
|
my $specials_range = $_[1]; |
| 720 |
|
return 0 if defined $self->{next_next_char}; |
| 721 |
|
my $count = $input->manakai_read_until |
| 722 |
|
($_[0], |
| 723 |
|
qr/(?![$specials_range\x{FDD0}-\x{FDDF}\x{FFFE}\x{FFFF}\x{1FFFE}\x{1FFFF}\x{2FFFE}\x{2FFFF}\x{3FFFE}\x{3FFFF}\x{4FFFE}\x{4FFFF}\x{5FFFE}\x{5FFFF}\x{6FFFE}\x{6FFFF}\x{7FFFE}\x{7FFFF}\x{8FFFE}\x{8FFFF}\x{9FFFE}\x{9FFFF}\x{AFFFE}\x{AFFFF}\x{BFFFE}\x{BFFFF}\x{CFFFE}\x{CFFFF}\x{DFFFE}\x{DFFFF}\x{EFFFE}\x{EFFFF}\x{FFFFE}\x{FFFFF}])[\x20-\x7E\xA0-\x{D7FF}\x{E000}-\x{10FFFD}]/, |
| 724 |
|
$_[2]); |
| 725 |
|
if ($count) { |
| 726 |
|
$self->{column} += $count; |
| 727 |
|
$self->{column_prev} += $count; |
| 728 |
|
$self->{prev_char} = [-1, -1, -1]; |
| 729 |
|
$self->{next_char} = -1; |
| 730 |
|
} |
| 731 |
|
return $count; |
| 732 |
|
}; # $self->{read_until} |
| 733 |
|
|
| 734 |
my $onerror = $_[2] || sub { |
my $onerror = $_[2] || sub { |
| 735 |
my (%opt) = @_; |
my (%opt) = @_; |
| 736 |
my $line = $opt{token} ? $opt{token}->{line} : $opt{line}; |
my $line = $opt{token} ? $opt{token}->{line} : $opt{line}; |
| 1027 |
data => chr $self->{next_char}, |
data => chr $self->{next_char}, |
| 1028 |
line => $self->{line}, column => $self->{column}, |
line => $self->{line}, column => $self->{column}, |
| 1029 |
}; |
}; |
| 1030 |
|
$self->{read_until}->($token->{data}, q[-!<>&], length $token->{data}); |
| 1031 |
|
|
| 1032 |
## Stay in the data state |
## Stay in the data state |
| 1033 |
!!!next-input-character; |
!!!next-input-character; |
| 1034 |
|
|
| 7794 |
}; |
}; |
| 7795 |
$p->{prev_char} = [-1, -1, -1]; |
$p->{prev_char} = [-1, -1, -1]; |
| 7796 |
$p->{next_char} = -1; |
$p->{next_char} = -1; |
| 7797 |
|
|
| 7798 |
|
$p->{read_until} = sub { |
| 7799 |
|
## TODO: ... |
| 7800 |
|
return 0; |
| 7801 |
|
}; # $p->{read_until}; |
| 7802 |
|
|
| 7803 |
my $ponerror = $onerror || sub { |
my $ponerror = $onerror || sub { |
| 7804 |
my (%opt) = @_; |
my (%opt) = @_; |
| 7805 |
my $line = $opt{line}; |
my $line = $opt{line}; |