File Coverage

blib/lib/Text/Demoroniser.pm

Criterion	Covered	Total	%
statement	29	29	100.0
branch	4	4	100.0
condition			n/a
subroutine	7	7	100.0
pod	2	2	100.0
total	42	42	100.0

package Text::Demoroniser;

use strict;
use warnings;

use vars qw( $VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS );
use Encode::ZapCP1252;

BEGIN {
    require Exporter;
    $VERSION = '0.07';
    @ISA = qw( Exporter );
    @EXPORT = qw();
    %EXPORT_TAGS = (
        'all' => [ qw( demoroniser demoroniser_utf8 ) ]
    );
    @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
}

my %character = (   #   ASCII   UTF8
    "\xE2\x80\x9A" => [ ',',    "\x201A" ],     # 82 - SINGLE LOW-9 QUOTATION MARK
    "\xE2\x80\x9E" => [ ',,',   "\x201E" ],     # 84 - DOUBLE LOW-9 QUOTATION MARK
    "\xE2\x80\xA6" => [ '...',  "\x2026" ],     # 85 - HORIZONTAL ELLIPSIS
    "\xCB\x86"     => [ '^',    "\x02C6" ],     # 88 - MODIFIER LETTER CIRCUMFLEX ACCENT
    "\xE2\x80\x98" => [ '`',    "\x2018" ],     # 91 - LEFT SINGLE QUOTATION MARK
    "\xE2\x80\x99" => [ q{'},   "\x2019" ],     # 92 - RIGHT SINGLE QUOTATION MARK
    "\xE2\x80\x9C" => [ '"',    "\x201C" ],     # 93 - LEFT DOUBLE QUOTATION MARK
    "\xE2\x80\x9D" => [ '"',    "\x201D" ],     # 94 - RIGHT DOUBLE QUOTATION MARK
    "\xE2\x80\xA2" => [ '*',    "\x2022" ],     # 95 - BULLET
    "\xE2\x80\x93" => [ '-',    "\x2013" ],     # 96 - EN DASH
    "\xE2\x80\x94" => [ '-',    "\x2014" ],     # 97 - EM DASH

    "\xE2\x80\xB9" => [ '<',    "\x2039" ],     # 8B - SINGLE LEFT-POINTING ANGLE
                                                #      QUOTATION MARK
    "\xE2\x80\xBA" => [ '>',    "\x203A" ],     # 9B - SINGLE RIGHT-POINTING ANGLE
                                                #      QUOTATION MARK
);

my $characters_re = '(' . join( '|', keys %character ) . ')';

sub demoroniser {
    my $str = shift;
    return  unless(defined $str);

    $str =~ s/$characters_re/$character{$1}[0]/g;

    zap_cp1252($str);

    return $str;
}

sub demoroniser_utf8 {
    my $str = shift;
    return  unless(defined $str);

    $str =~ s/$characters_re/$character{$1}[1]/g;

    fix_cp1252($str);

    return $str;
}

1;

__END__