File Coverage

lib/UTF8/R2.pm

Criterion	Covered	Total	%
statement	351	399	87.9
branch	367	424	86.5
condition	40	68	58.8
subroutine	33	44	75.0
pod	7	24	29.1
total	798	959	83.2

package UTF8::R2;
######################################################################
#
# UTF8::R2 - makes UTF-8 scripting easy for enterprise use
#
# http://search.cpan.org/dist/UTF8-R2/
#
# Copyright (c) 2019, 2020, 2021, 2022 INABA Hitoshi  in a CPAN
######################################################################

use 5.00503;    # Universal Consensus 1998 for primetools
# use 5.008001; # Lancaster Consensus 2013 for toolchains

$VERSION = '0.26';
$VERSION = $VERSION;

use strict;
BEGIN { $INC{'warnings.pm'} = '' if $] < 5.006 } use warnings; local $^W=1;
use Symbol ();

my %utf8_codepoint = (

    # beautiful concept in young days, however disabled 5-6 octets for safety
    # https://www.ietf.org/rfc/rfc2279.txt
    'RFC2279' => qr{(?>@{[join('', qw(
        [\x00-\x7F\x80-\xBF\xC0-\xC1\xF5-\xFF]       |
        [\xC2-\xDF][\x80-\xBF]                       |
        [\xE0-\xEF][\x80-\xBF][\x80-\xBF]            |
        [\xF0-\xF4][\x80-\xBF][\x80-\xBF][\x80-\xBF] |
        [\x00-\xFF]
    ))]})}x,

    # https://tools.ietf.org/rfc/rfc3629.txt
    'RFC3629' => qr{(?>@{[join('', qw(
        [\x00-\x7F\x80-\xBF\xC0-\xC1\xF5-\xFF]       |
        [\xC2-\xDF][\x80-\xBF]                       |
        [\xE0-\xE0][\xA0-\xBF][\x80-\xBF]            |
        [\xE1-\xEC][\x80-\xBF][\x80-\xBF]            |
        [\xED-\xED][\x80-\x9F][\x80-\xBF]            |
        [\xEE-\xEF][\x80-\xBF][\x80-\xBF]            |
        [\xF0-\xF0][\x90-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF1-\xF3][\x80-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF4-\xF4][\x80-\x8F][\x80-\xBF][\x80-\xBF] |
        [\x00-\xFF]
    ))]})}x,

    # http://simonsapin.github.io/wtf-8/
    'WTF8' => qr{(?>@{[join('', qw(
        [\x00-\x7F\x80-\xBF\xC0-\xC1\xF5-\xFF]       |
        [\xC2-\xDF][\x80-\xBF]                       |
        [\xE0-\xE0][\xA0-\xBF][\x80-\xBF]            |
        [\xE1-\xEF][\x80-\xBF][\x80-\xBF]            |
        [\xF0-\xF0][\x90-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF1-\xF3][\x80-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF4-\xF4][\x80-\x8F][\x80-\xBF][\x80-\xBF] |
        [\x00-\xFF]
    ))]})}x,

    # optimized RFC3629 for ja_JP
    'RFC3629.ja_JP' => qr{(?>@{[join('', qw(
        [\x00-\x7F\x80-\xBF\xC0-\xC1\xF5-\xFF]       |
        [\xE1-\xEC][\x80-\xBF][\x80-\xBF]            |
        [\xC2-\xDF][\x80-\xBF]                       |
        [\xEE-\xEF][\x80-\xBF][\x80-\xBF]            |
        [\xF0-\xF0][\x90-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xE0-\xE0][\xA0-\xBF][\x80-\xBF]            |
        [\xED-\xED][\x80-\x9F][\x80-\xBF]            |
        [\xF1-\xF3][\x80-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF4-\xF4][\x80-\x8F][\x80-\xBF][\x80-\xBF] |
        [\x00-\xFF]
    ))]})}x,

    # optimized WTF-8 for ja_JP
    'WTF8.ja_JP' => qr{(?>@{[join('', qw(
        [\x00-\x7F\x80-\xBF\xC0-\xC1\xF5-\xFF]       |
        [\xE1-\xEF][\x80-\xBF][\x80-\xBF]            |
        [\xC2-\xDF][\x80-\xBF]                       |
        [\xE0-\xE0][\xA0-\xBF][\x80-\xBF]            |
        [\xF0-\xF0][\x90-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF1-\xF3][\x80-\xBF][\x80-\xBF][\x80-\xBF] |
        [\xF4-\xF4][\x80-\x8F][\x80-\xBF][\x80-\xBF] |
        [\x00-\xFF]
    ))]})}x,
);

# supports /./
my $x =
    ($^X =~ /jperl(\.exe)?\z/i) && (`$^X -v` =~ /SJIS version/) ?
    q{(?>[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC]|[\x00-\xFF])} : # debug tool using JPerl(SJIS version)
    $utf8_codepoint{'RFC3629'};

# supports [\b] \d \h \s \v \w
my $bare_backspace = '\x08';
my $bare_d = '0123456789';
my $bare_h = '\x09\x20';
my $bare_s = '\t\n\f\r\x20';
my $bare_v = '\x0A\x0B\x0C\x0D';
my $bare_w = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_';

#---------------------------------------------------------------------
# exports mb package
sub import {
    my $self = shift @_;

    # confirm version
    if (defined($_[0]) and ($_[0] =~ /\A [0-9] /xms)) {
        if ($_[0] ne $UTF8::R2::VERSION) {
            my($package,$filename,$line) = caller;
            die "$filename requires @{[__PACKAGE__]} $_[0], however @{[__FILE__]} am only $UTF8::R2::VERSION, stopped at $filename line $line.\n";
        }
        shift @_;
    }

    for (@_) {

        # export *mb
        if ($_ eq '*mb') {
            no strict qw(refs);

            # tie my %mb, __PACKAGE__; # makes: Parentheses missing around "my" list
            tie my %mb, 'UTF8::R2';
            *{caller().'::mb'} = \%mb;

            # supports mb package
            *{caller().'::mb::ORIG_PROGRAM_NAME'} = \$UTF8::R2::ORIG_PROGRAM_NAME;
            *{caller().'::mb::PERL'}              = \$UTF8::R2::PERL;
            *{caller().'::mb::chop'}              = \&UTF8::R2::chop;
            *{caller().'::mb::chr'}               = \&UTF8::R2::chr;
            *{caller().'::mb::do'}                = \&UTF8::R2::do;
            *{caller().'::mb::eval'}              = \&UTF8::R2::eval;
            *{caller().'::mb::getc'}              = \&UTF8::R2::getc;
            *{caller().'::mb::index'}             = \&UTF8::R2::index;
            *{caller().'::mb::index_byte'}        = \&UTF8::R2::index_byte;
            *{caller().'::mb::length'}            = \&UTF8::R2::length;
            *{caller().'::mb::ord'}               = \&UTF8::R2::ord;
            *{caller().'::mb::require'}           = \&UTF8::R2::require;
            *{caller().'::mb::reverse'}           = \&UTF8::R2::reverse;
            *{caller().'::mb::rindex'}            = \&UTF8::R2::rindex;
            *{caller().'::mb::rindex_byte'}       = \&UTF8::R2::rindex_byte;
            *{caller().'::mb::split'}             = \&UTF8::R2::split;
            *{caller().'::mb::substr'}            = \&UTF8::R2::substr;
            *{caller().'::mb::tr'}                = \&UTF8::R2::tr;
        }

        # export %mb
        elsif ($_ eq '%mb') {
            no strict qw(refs);

            # tie my %mb, __PACKAGE__; # makes: Parentheses missing around "my" list
            tie my %mb, 'UTF8::R2';
            *{caller().'::mb'} = \%mb;
        }

        # set script encoding
        elsif (defined $utf8_codepoint{$_}) {
            $x = $utf8_codepoint{$_};
        }
    }

    # $^X($EXECUTABLE_NAME) for execute MBCS Perl script
    $UTF8::R2::PERL = $^X;
    $UTF8::R2::PERL = $UTF8::R2::PERL; # to avoid: Name "UTF8::R2::PERL" used only once: possible typo at ...

    # original $0($PROGRAM_NAME)
    $UTF8::R2::ORIG_PROGRAM_NAME = $0;
    $UTF8::R2::ORIG_PROGRAM_NAME = $UTF8::R2::ORIG_PROGRAM_NAME; # to avoid: Name "UTF8::R2::ORIG_PROGRAM_NAME" used only once: possible typo at ...
}

#---------------------------------------------------------------------
# confess() for this module
sub confess {
    my $i = 0;
    my @confess = ();
    while (my($package,$filename,$line,$subroutine) = caller($i)) {
        push @confess, "[$i] $filename($line) $subroutine\n";
        $i++;
    }
    print STDERR "\n", @_, "\n";
    print STDERR CORE::reverse @confess;
    die;
}

#---------------------------------------------------------------------
# chop() for UTF-8 codepoint string
sub UTF8::R2::chop (@) {
    my $chop = '';
    for (@_ ? @_ : $_) {
        if (my @x = /\G$x/g) {
            $chop = pop @x;
            $_ = join '', @x;
        }
    }
    return $chop;
}

#---------------------------------------------------------------------
# chr() for UTF-8 codepoint string
sub UTF8::R2::chr (;$) {
    my $number = @_ ? $_[0] : $_;

# Negative values give the Unicode replacement character (chr(0xfffd)),
# except under the bytes pragma, where the low eight bits of the value
# (truncated to an integer) are used.

    my @octet = ();
    CORE::do {
        unshift @octet, ($number % 0x100);
        $number = int($number / 0x100);
    } while ($number > 0);
    return pack 'C*', @octet;
}

#---------------------------------------------------------------------
# mb::do() like do(), mb.pm compatible
sub UTF8::R2::do ($) {

    # run as Perl script
    return CORE::eval sprintf(<<'END', (caller)[0,2,1]);
package %s;
#line %s "%s"
CORE::do "$_[0]";
END
}

#---------------------------------------------------------------------
# mb::eval() like eval(), mb.pm compatible
sub UTF8::R2::eval (;$) {
    local $_ = @_ ? $_[0] : $_;

    # run as Perl script in caller package
    return CORE::eval sprintf(<<'END', (caller)[0,2,1], $_);
package %s;
#line %s "%s"
%s
END
}

#---------------------------------------------------------------------
# getc() for UTF-8 codepoint string
sub UTF8::R2::getc (;*) {
    my $fh = @_ ? Symbol::qualify_to_ref($_[0],caller()) : \*STDIN;
    my $getc = CORE::getc $fh;
    if ($getc =~ /\A [\x00-\x7F\x80-\xC1\xF5-\xFF] \z/xms) {
    }
    elsif ($getc =~ /\A [\xC2-\xDF] \z/xms) {
        $getc .= CORE::getc $fh;
    }
    elsif ($getc =~ /\A [\xE0-\xEF] \z/xms) {
        $getc .= CORE::getc $fh;
        $getc .= CORE::getc $fh;
    }
    elsif ($getc =~ /\A [\xF0-\xF4] \z/xms) {
        $getc .= CORE::getc $fh;
        $getc .= CORE::getc $fh;
        $getc .= CORE::getc $fh;
    }
    return $getc;
}

#---------------------------------------------------------------------
# index() for UTF-8 codepoint string
sub UTF8::R2::index ($$;$) {
    my $index = 0;
    if (@_ == 3) {
        $index = CORE::index $_[0], $_[1], CORE::length(UTF8::R2::substr($_[0], 0, $_[2]));
    }
    else {
        $index = CORE::index $_[0], $_[1];
    }
    if ($index == -1) {
        return -1;
    }
    else {
        return UTF8::R2::length(CORE::substr $_[0], 0, $index);
    }
}

#---------------------------------------------------------------------
# JPerl like index() for UTF-8 codepoint string
sub UTF8::R2::index_byte ($$;$) {
    if (@_ == 3) {
        return CORE::index $_[0], $_[1], CORE::length(UTF8::R2::substr($_[0], 0, $_[2]));
    }
    else {
        return CORE::index $_[0], $_[1];
    }
}

#---------------------------------------------------------------------
# universal lc() for UTF-8 codepoint string
sub UTF8::R2::lc (;$) {
    local $_ = @_ ? $_[0] : $_;
    #                          A a B b C c D d E e F f G g H h I i J j K k L l M m N n O o P p Q q R r S s T t U u V v W w X x Y y Z z
    return join '', map { {qw( A a B b C c D d E e F f G g H h I i J j K k L l M m N n O o P p Q q R r S s T t U u V v W w X x Y y Z z )}->{$_}||$_ } /\G$x/g;
    #                          A a B b C c D d E e F f G g H h I i J j K k L l M m N n O o P p Q q R r S s T t U u V v W w X x Y y Z z
}

#---------------------------------------------------------------------
# universal lcfirst() for UTF-8 codepoint string
sub UTF8::R2::lcfirst (;$) {
    local $_ = @_ ? $_[0] : $_;
    if (/\A($x)(.*)\z/s) {
        return UTF8::R2::lc($1) . $2;
    }
    else {
        return '';
    }
}

#---------------------------------------------------------------------
# length() for UTF-8 codepoint string
sub UTF8::R2::length (;$) {
    local $_ = @_ ? $_[0] : $_;
    return scalar(() = /\G$x/g);
}

#---------------------------------------------------------------------
# ord() for UTF-8 codepoint string
sub UTF8::R2::ord (;$) {
    local $_ = @_ ? $_[0] : $_;
    my $ord = 0;
    if (/\A($x)/) {
        for my $octet (unpack 'C*', $1) {
            $ord = $ord * 0x100 + $octet;
        }
    }
    return $ord;
}

#---------------------------------------------------------------------
# qr/ [A-Z] / for UTF-8 codepoint string
sub list_all_by_hyphen_utf8_like {
    my($a, $b) = @_;
    my @a = (undef, unpack 'C*', $a);
    my @b = (undef, unpack 'C*', $b);

    if (0) { }
    elsif (CORE::length($a) == 1) {
        if (0) { }
        elsif (CORE::length($b) == 1) {
            return (
$a[1]<=$b[1] ?  sprintf(join('', qw( [\x%02x-\x%02x]                                         )), $a[1],
                                                                                                 $b[1]) : (),
            );
        }
        elsif (CORE::length($b) == 2) {
            return (
                sprintf(join('', qw(       \x%02x  [\x80-\x%02x]                             )), $b[1], $b[2]),
0xC2 < $b[1] ?  sprintf(join('', qw( [\xC2-\x%02x] [\x80-\xBF  ]                             )), $b[1]-1     ) : (),
                sprintf(join('', qw( [\x%02x-\x7F]                                           )), $a[1]       ),
            );
        }
        elsif (CORE::length($b) == 3) {
            return (
                sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x]               )), $b[1], $b[2], $b[3]),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ]               )), $b[1], $b[2]-1     ) : (),
0xE0 < $b[1] ?  sprintf(join('', qw( [\xE0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ]               )), $b[1]-1            ) : (),
                sprintf(join('', qw( [\xC2-\xDF  ] [\x80-\xBF  ]                             )),                    ),
                sprintf(join('', qw( [\x%02x-\x7F]                                           )), $a[1]              ),
            );
        }
        elsif (CORE::length($b) == 4) {
            return (
                sprintf(join('', qw(       \x%02x        \x%02x        \x%02x  [\x80-\x%02x] )), $b[1], $b[2], $b[3], $b[4]),
0x80 < $b[3] ?  sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x] [\x80-\xBF  ] )), $b[1], $b[2], $b[3]-1     ) : (),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1], $b[2]-1            ) : (),
0xF0 < $b[1] ?  sprintf(join('', qw( [\xF0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1]-1                   ) : (),
                sprintf(join('', qw( [\xE0-\xEF  ] [\x80-\xBF  ] [\x80-\xBF  ]               )),                           ),
                sprintf(join('', qw( [\xC2-\xDF  ] [\x80-\xBF  ]                             )),                           ),
                sprintf(join('', qw( [\x%02x-\x7F]                                           )), $a[1]                     ),
            );
        }
    }
    elsif (CORE::length($a) == 2) {
        if (0) { }
        elsif (CORE::length($b) == 2) {
            my $lower_limit = join('|',
$a[1] < 0xDF ?  sprintf(join('', qw( [\x%02x-\xDF] [\x80-\xBF  ]                             )), $a[1]+1     ) : (),
                sprintf(join('', qw(  \x%02x       [\x%02x-\xBF]                             )), $a[1], $a[2]),
            );
            my $upper_limit = join('|',
                sprintf(join('', qw(       \x%02x  [\x80-\x%02x]                             )), $b[1], $b[2]),
0xC2 < $b[1] ?  sprintf(join('', qw( [\xC2-\x%02x] [\x80-\xBF  ]                             )), $b[1]-1     ) : (),
            );
            return qq{(?=$lower_limit)(?=$upper_limit)};
        }
        elsif (CORE::length($b) == 3) {
            return (
                sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x]               )), $b[1], $b[2], $b[3] ),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ]               )), $b[1], $b[2]-1      ) : (),
0xE0 < $b[1] ?  sprintf(join('', qw( [\xE0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ]               )), $b[1]-1             ) : (),
$a[1] < 0xDF ?  sprintf(join('', qw( [\x%02x-\xDF] [\x80-\xBF  ]                             )), $a[1]+1             ) : (),
                sprintf(join('', qw(  \x%02x       [\x%02x-\xBF]                             )), $a[1], $a[2]        ),
            );
        }
        elsif (CORE::length($b) == 4) {
            return (
                sprintf(join('', qw(       \x%02x        \x%02x        \x%02x  [\x80-\x%02x] )), $b[1], $b[2], $b[3], $b[4]),
0x80 < $b[3] ?  sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x] [\x80-\xBF  ] )), $b[1], $b[2], $b[3]-1     ) : (),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1], $b[2]-1            ) : (),
0xF0 < $b[1] ?  sprintf(join('', qw( [\xF0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1]-1                   ) : (),
                sprintf(join('', qw( [\xE0-\xEF  ] [\x80-\xBF  ] [\x80-\xBF  ]               )),                           ),
$a[1] < 0xDF ?  sprintf(join('', qw( [\x%02x-\xDF] [\x80-\xBF  ]                             )), $a[1]+1                   ) : (),
                sprintf(join('', qw(  \x%02x       [\x%02x-\xBF]                             )), $a[1], $a[2]              ),
            );
        }
    }
    elsif (CORE::length($a) == 3) {
        if (0) { }
        elsif (CORE::length($b) == 3) {
            my $lower_limit = join('|',
$a[1] < 0xEF ?  sprintf(join('', qw( [\x%02x-\xEF] [\x80-\xBF  ] [\x80-\xBF  ]               )), $a[1]+1            ) : (),
$a[2] < 0xBF ?  sprintf(join('', qw(  \x%02x       [\x%02x-\xBF] [\x80-\xBF  ]               )), $a[1], $a[2]+1     ) : (),
                sprintf(join('', qw(  \x%02x        \x%02x       [\x%02x-\xBF]               )), $a[1], $a[2], $a[3]),
            );
            my $upper_limit = join('|',
                sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x]               )), $b[1], $b[2], $b[3]),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ]               )), $b[1], $b[2]-1     ) : (),
0xE0 < $b[1] ?  sprintf(join('', qw( [\xE0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ]               )), $b[1]-1            ) : (),
            );
            return qq{(?=$lower_limit)(?=$upper_limit)};
        }
        elsif (CORE::length($b) == 4) {
            return (
                sprintf(join('', qw(       \x%02x        \x%02x        \x%02x  [\x80-\x%02x] )), $b[1], $b[2], $b[3], $b[4]),
0x80 < $b[3] ?  sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x] [\x80-\xBF  ] )), $b[1], $b[2], $b[3]-1     ) : (),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1], $b[2]-1            ) : (),
0xF0 < $b[1] ?  sprintf(join('', qw( [\xF0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1]-1                   ) : (),
$a[1] < 0xEF ?  sprintf(join('', qw( [\x%02x-\xEF] [\x80-\xBF  ] [\x80-\xBF  ]               )), $a[1]+1                   ) : (),
$a[2] < 0xBF ?  sprintf(join('', qw(  \x%02x       [\x%02x-\xBF] [\x80-\xBF  ]               )), $a[1], $a[2]+1            ) : (),
                sprintf(join('', qw(  \x%02x        \x%02x       [\x%02x-\xBF]               )), $a[1], $a[2], $a[3]       ),
            );
        }
    }
    elsif (CORE::length($a) == 4) {
        if (0) { }
        elsif (CORE::length($b) == 4) {
            my $lower_limit = join('|',
$a[1] < 0xF4 ?  sprintf(join('', qw( [\x%02x-\xF4] [\x80-\xBF  ] [\x80-\xBF  ] [\x80-\xBF  ] )), $a[1]+1                   ) : (),
$a[2] < 0xBF ?  sprintf(join('', qw(  \x%02x       [\x%02x-\xBF] [\x80-\xBF  ] [\x80-\xBF  ] )), $a[1], $a[2]+1            ) : (),
$a[3] < 0xBF ?  sprintf(join('', qw(  \x%02x        \x%02x       [\x%02x-\xBF] [\x80-\xBF  ] )), $a[1], $a[2], $a[3]+1     ) : (),
                sprintf(join('', qw(  \x%02x        \x%02x        \x%02x       [\x%02x-\xBF] )), $a[1], $a[2], $a[3], $a[4]),
            );
            my $upper_limit = join('|',
                sprintf(join('', qw(       \x%02x        \x%02x        \x%02x  [\x80-\x%02x] )), $b[1], $b[2], $b[3], $b[4]),
0x80 < $b[3] ?  sprintf(join('', qw(       \x%02x        \x%02x  [\x80-\x%02x] [\x80-\xBF  ] )), $b[1], $b[2], $b[3]-1     ) : (),
0x80 < $b[2] ?  sprintf(join('', qw(       \x%02x  [\x80-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1], $b[2]-1            ) : (),
0xF0 < $b[1] ?  sprintf(join('', qw( [\xF0-\x%02x] [\x80-\xBF  ] [\x80-\xBF  ] [\x80-\xBF  ] )), $b[1]-1                   ) : (),
            );
            return qq{(?=$lower_limit)(?=$upper_limit)};
        }
    }

    # over range of codepoint
    confess sprintf(qq{@{[__FILE__]}: codepoint class [$_[0]-$_[1]] is not 1 to 4 octets (%d-%d)}, CORE::length($a), CORE::length($b));
}

#---------------------------------------------------------------------
# qr// for UTF-8 codepoint string
sub UTF8::R2::qr ($) {

    my $modifiers = '';
    if (($modifiers) = $_[0] =~ /\A \( \? \^? (.*?) : /x) {
        $modifiers =~ s/-.*//;
    }

    my @after = ();
    while ($_[0] =~ s! \A (
        (?> \[ (?: \[:[^:]+?:\] | \\x\{[0123456789ABCDEFabcdef]+\} | \\c[\x00-\xFF] | (?>\\$x) | $x )+? \] ) |
                                  \\x\{[0123456789ABCDEFabcdef]+\} | \\c[\x00-\xFF] | (?>\\$x) | $x
    ) !!x) {
        my $before = $1;

        # [^...] or [...]
        if (my($negative,$class) = $before =~ /\A \[ (\^?) ((?>\\$x|$x)+?) \] \z/x) {
            my @classmate = $class =~ /\G (?: \[:.+?:\] | \\x\{[0123456789ABCDEFabcdef]+\} | (?>\\$x) | $x ) /xg;
            my @sbcs = ();
            my @xbcs = ();

            for (my $i=0; $i <= $#classmate; ) {
                my $classmate = $classmate[$i];

                # hyphen of [A-Z] or [^A-Z]
                if (($i < $#classmate) and ($classmate[$i+1] eq '-')) {
                    my $a = ($classmate[$i+0] =~ /\A \\x \{ ([0123456789ABCDEFabcdef]+) \} \z/x) ? UTF8::R2::chr(hex $1) : $classmate[$i+0];
                    my $b = ($classmate[$i+2] =~ /\A \\x \{ ([0123456789ABCDEFabcdef]+) \} \z/x) ? UTF8::R2::chr(hex $1) : $classmate[$i+2];
                    push @xbcs, list_all_by_hyphen_utf8_like($a, $b);
                    $i += 3;
                }

                # any "one"
                else {

                    # \x{UTF8hex}
                    if ($classmate =~ /\A \\x \{ ([0123456789ABCDEFabcdef]+) \} \z/x) {
                        push @xbcs, UTF8::R2::chr(hex $1);
                    }

                    # \any
                    elsif ($classmate eq '\D'         ) { push @xbcs, "(?:(?![$bare_d])$x)"  }
                    elsif ($classmate eq '\H'         ) { push @xbcs, "(?:(?![$bare_h])$x)"  }
#                   elsif ($classmate eq '\N'         ) { push @xbcs, "(?:(?!\\n)$x)"        } # \N in a character class must be a named character: \N{...} in regex
#                   elsif ($classmate eq '\R'         ) { push @xbcs, "(?>\\r\\n|[$bare_v])" } # Unrecognized escape \R in character class passed through in regex
                    elsif ($classmate eq '\S'         ) { push @xbcs, "(?:(?![$bare_s])$x)"  }
                    elsif ($classmate eq '\V'         ) { push @xbcs, "(?:(?![$bare_v])$x)"  }
                    elsif ($classmate eq '\W'         ) { push @xbcs, "(?:(?![$bare_w])$x)"  }
                    elsif ($classmate eq '\b'         ) { push @sbcs, $bare_backspace        }
                    elsif ($classmate eq '\d'         ) { push @sbcs, $bare_d                }
                    elsif ($classmate eq '\h'         ) { push @sbcs, $bare_h                }
                    elsif ($classmate eq '\s'         ) { push @sbcs, $bare_s                }
                    elsif ($classmate eq '\v'         ) { push @sbcs, $bare_v                }
                    elsif ($classmate eq '\w'         ) { push @sbcs, $bare_w                }

                    # [:POSIX:]
                    elsif ($classmate eq '[:alnum:]'  ) { push @sbcs, '\x30-\x39\x41-\x5A\x61-\x7A';                  }
                    elsif ($classmate eq '[:alpha:]'  ) { push @sbcs, '\x41-\x5A\x61-\x7A';                           }
                    elsif ($classmate eq '[:ascii:]'  ) { push @sbcs, '\x00-\x7F';                                    }
                    elsif ($classmate eq '[:blank:]'  ) { push @sbcs, '\x09\x20';                                     }
                    elsif ($classmate eq '[:cntrl:]'  ) { push @sbcs, '\x00-\x1F\x7F';                                }
                    elsif ($classmate eq '[:digit:]'  ) { push @sbcs, '\x30-\x39';                                    }
                    elsif ($classmate eq '[:graph:]'  ) { push @sbcs, '\x21-\x7F';                                    }
                    elsif ($classmate eq '[:lower:]'  ) { push @sbcs, '\x61-\x7A';                                    } # /i modifier requires 'a' to 'z' literally
                    elsif ($classmate eq '[:print:]'  ) { push @sbcs, '\x20-\x7F';                                    }
                    elsif ($classmate eq '[:punct:]'  ) { push @sbcs, '\x21-\x2F\x3A-\x3F\x40\x5B-\x5F\x60\x7B-\x7E'; }
                    elsif ($classmate eq '[:space:]'  ) { push @sbcs, '\s\x0B';                                       } # "\s" and vertical tab ("\cK")
                    elsif ($classmate eq '[:upper:]'  ) { push @sbcs, '\x41-\x5A';                                    } # /i modifier requires 'A' to 'Z' literally
                    elsif ($classmate eq '[:word:]'   ) { push @sbcs, '\x30-\x39\x41-\x5A\x5F\x61-\x7A';              }
                    elsif ($classmate eq '[:xdigit:]' ) { push @sbcs, '\x30-\x39\x41-\x46\x61-\x66';                  }

                    # [:^POSIX:]
                    elsif ($classmate eq '[:^alnum:]' ) { push @xbcs, "(?:(?![\\x30-\\x39\\x41-\\x5A\\x61-\\x7A])$x)";                      }
                    elsif ($classmate eq '[:^alpha:]' ) { push @xbcs, "(?:(?![\\x41-\\x5A\\x61-\\x7A])$x)";                                 }
                    elsif ($classmate eq '[:^ascii:]' ) { push @xbcs, "(?:(?![\\x00-\\x7F])$x)";                                            }
                    elsif ($classmate eq '[:^blank:]' ) { push @xbcs, "(?:(?![\\x09\\x20])$x)";                                             }
                    elsif ($classmate eq '[:^cntrl:]' ) { push @xbcs, "(?:(?![\\x00-\\x1F\\x7F])$x)";                                       }
                    elsif ($classmate eq '[:^digit:]' ) { push @xbcs, "(?:(?![\\x30-\\x39])$x)";                                            }
                    elsif ($classmate eq '[:^graph:]' ) { push @xbcs, "(?:(?![\\x21-\\x7F])$x)";                                            }
                    elsif ($classmate eq '[:^lower:]' ) { push @xbcs, "(?:(?![\\x61-\\x7A])$x)";                                            } # /i modifier requires 'a' to 'z' literally
                    elsif ($classmate eq '[:^print:]' ) { push @xbcs, "(?:(?![\\x20-\\x7F])$x)";                                            }
                    elsif ($classmate eq '[:^punct:]' ) { push @xbcs, "(?:(?![\\x21-\\x2F\\x3A-\\x3F\\x40\\x5B-\\x5F\\x60\\x7B-\\x7E])$x)"; }
                    elsif ($classmate eq '[:^space:]' ) { push @xbcs, "(?:(?![\\s\\x0B])$x)";                                               } # "\s" and vertical tab ("\cK")
                    elsif ($classmate eq '[:^upper:]' ) { push @xbcs, "(?:(?![\\x41-\\x5A])$x)";                                            } # /i modifier requires 'A' to 'Z' literally
                    elsif ($classmate eq '[:^word:]'  ) { push @xbcs, "(?:(?![\\x30-\\x39\\x41-\\x5A\\x5F\\x61-\\x7A])$x)";                 }
                    elsif ($classmate eq '[:^xdigit:]') { push @xbcs, "(?:(?![\\x30-\\x39\\x41-\\x46\\x61-\\x66])$x)";                      }

                    # other all
                    elsif (CORE::length($classmate)==1) { push @sbcs, $classmate }
                    else                                { push @xbcs, $classmate }
                    $i += 1;
                }
            }

            # [^...]
            if ($negative eq q[^]) {
                push @after,
                    ( @sbcs and  @xbcs) ? '(?:(?!' . join('|', @xbcs, '['.join('',@sbcs).']') . ")$x)" :
                    (!@sbcs and  @xbcs) ? '(?:(?!' . join('|', @xbcs                        ) . ")$x)" :
                    ( @sbcs and !@xbcs) ? '(?:(?!' .                  '['.join('',@sbcs).']'  . ")$x)" :
                    '';
            }

            # [...] on Perl 5.006
            elsif ($] =~ /\A5\.006/) {
                push @after,
                    ( @sbcs and  @xbcs) ? '(?:'    . join('|', @xbcs, '['.join('',@sbcs).']') .    ')' :
                    (!@sbcs and  @xbcs) ? '(?:'    . join('|', @xbcs                        ) .    ')' :
                    ( @sbcs and !@xbcs) ?                             '['.join('',@sbcs).']'           :
                    '';
            }

            # [...]
            else {
                push @after,
                    ( @sbcs and  @xbcs) ? '(?:(?=' . join('|', @xbcs, '['.join('',@sbcs).']') . ")$x)" :
                    (!@sbcs and  @xbcs) ? '(?:(?=' . join('|', @xbcs                        ) . ")$x)" :
                    ( @sbcs and !@xbcs) ?                             '['.join('',@sbcs).']'           :
                    '';
            }
        }

        # \any or /./
        elsif ($before eq '.' ) { push @after, ($modifiers =~ /s/) ? $x : "(?:(?!\\n)$x)"                    }
        elsif ($before eq '\B') { push @after, "(?:(?
        elsif ($before eq '\D') { push @after, "(?:(?![$bare_d])$x)"                                         }
        elsif ($before eq '\H') { push @after, "(?:(?![$bare_h])$x)"                                         }
        elsif ($before eq '\N') { push @after, "(?:(?!\\n)$x)"                                               }
        elsif ($before eq '\R') { push @after, "(?>\\r\\n|[$bare_v])"                                        }
        elsif ($before eq '\S') { push @after, "(?:(?![$bare_s])$x)"                                         }
        elsif ($before eq '\V') { push @after, "(?:(?![$bare_v])$x)"                                         }
        elsif ($before eq '\W') { push @after, "(?:(?![$bare_w])$x)"                                         }
        elsif ($before eq '\b') { push @after, "(?:(?
        elsif ($before eq '\d') { push @after, "[$bare_d]"                                                   }
        elsif ($before eq '\h') { push @after, "[$bare_h]"                                                   }
        elsif ($before eq '\s') { push @after, "[$bare_s]"                                                   }
        elsif ($before eq '\v') { push @after, "[$bare_v]"                                                   }
        elsif ($before eq '\w') { push @after, "[$bare_w]"                                                   }

        # quantifiers ? + * {n} {n,} {n,m}
        elsif ($before =~ /\A[?+*{]\z/) {
            if    (0)                                             { }
            elsif ($after[-1] =~ /\A \\c [\x00-\xFF]        \z/x) { } # \c) \c} \c] \cX
            elsif ($after[-1] =~ /\A \\  [\x00-\xFF]        \z/x) { } # \) \} \] \" \0 \1 \D \E \F \G \H \K \L \N \Q \R \S \U \V \W \\ \a \d \e \f \h \l \n \r \s \t \u \v \w
            elsif ($after[-1] =~ /\A     [\x00-\xFF]        \z/x) { } # (a) a{1} [a] a . \012 \x12 \o{12} \g{1}
            elsif ($after[-1] =~ /       [\x00-\xFF] [)}\]] \z/x) { } # (any) any{1} [any]
            else {                                                    # XBCS
                $after[-1] = '(?:' . $after[-1] . ')';
            }
            push @after, $before;
        }

        # \x{UTF8hex}
        elsif ($before =~ /\A \\x \{ ([0123456789ABCDEFabcdef]+) \} \z/x) {
            push @after, UTF8::R2::chr(hex $1);
        }

        # else
        else {
            push @after, $before;
        }
    }

    my $after = join '', @after;
    return qr/$after/;
}

#---------------------------------------------------------------------
# mb::require() like require(), mb.pm compatible
sub UTF8::R2::require (;$) {
    local $_ = @_ ? $_[0] : $_;

    # require perl version
    if (/^[0-9]/) {
        if ($] < $_) {
            confess "Perl $_ required--this is only version $], stopped";
        }
        else {
            undef $@;
            return 1;
        }
    }

    # require expr
    else {

        # find expr in @INC
        my $file = $_;
        if (($file =~ s{::}{/}g) or ($file !~ m{[\./\\]})) {
            $file .= '.pm';
        }
        if (exists $INC{$file}) {
            undef $@;
            return 1 if $INC{$file};
            confess "Compilation failed in require";
        }
        for my $prefix_file ($file, map { "$_/$file" } @INC) {
            if (-f $prefix_file) {
                $INC{$_} = $prefix_file;

                # run as Perl script
                # must use CORE::do to use , because CORE::eval cannot do it.
                local $@;
                my $result = CORE::eval sprintf(<<'END', (caller)[0,2,1]);
package %s;
#line %s "%s"
CORE::do "$prefix_file";
END

                # return result
                if ($@) {
                    $INC{$_} = undef;
                    confess $@;
                }
                elsif (not $result) {
                    delete $INC{$_};
                    confess "$_ did not return true value";
                }
                else {
                    return $result;
                }
            }
        }
        confess "Can't find $_ in \@INC";
    }
}

#---------------------------------------------------------------------
# reverse() for UTF-8 codepoint string
sub UTF8::R2::reverse (@) {

    # in list context,
    if (wantarray) {

        # returns a list value consisting of the elements of @_ in the opposite order
        return CORE::reverse @_;
    }

    # in scalar context,
    else {

        # returns a string value with all characters in the opposite order of
        return (join '',
            CORE::reverse(
                @_ ?
                join('',@_) =~ /\G$x/g : # concatenates the elements of @_
                /\G$x/g                  # $_ when without arguments
            )
        );
    }
}

#---------------------------------------------------------------------
# rindex() for UTF-8 codepoint string
sub UTF8::R2::rindex ($$;$) {
    my $rindex = 0;
    if (@_ == 3) {
        $rindex = CORE::rindex $_[0], $_[1], CORE::length(UTF8::R2::substr($_[0], 0, $_[2]));
    }
    else {
        $rindex = CORE::rindex $_[0], $_[1];
    }
    if ($rindex == -1) {
        return -1;
    }
    else {
        return UTF8::R2::length(CORE::substr $_[0], 0, $rindex);
    }
}

#---------------------------------------------------------------------
# JPerl like rindex() for UTF-8 codepoint string
sub UTF8::R2::rindex_byte ($$;$) {
    if (@_ == 3) {
        return CORE::rindex $_[0], $_[1], CORE::length(UTF8::R2::substr($_[0], 0, $_[2]));
    }
    else {
        return CORE::rindex $_[0], $_[1];
    }
}

#---------------------------------------------------------------------
# split() for UTF-8 codepoint string
sub UTF8::R2::split (;$$$) {
    if (defined($_[0]) and (($_[0] eq '') or ($_[0] =~ /\A \( \? \^? [-a-z]* : \) \z/x))) {
        my @x = (defined($_[1]) ? $_[1] : $_) =~ /\G$x/g;
        if (defined($_[2]) and ($_[2] > 0) and (scalar(@x) > $_[2])) {
            @x = (@x[0..$_[2]-1-1], join('', @x[$_[2]-1..$#x]));
        }
        if (wantarray) {
            return @x;
        }
        else {
            if ($] < 5.012) {
                warn "Use of implicit split to \@_ is deprecated" if $^W;
                @_ = @x; # unlike camel book and perldoc saying, can return only scalar(@_), cannot @_
            }
            return scalar @x;
        }
    }
    elsif (@_ == 3) {
        return CORE::split UTF8::R2::qr($_[0]), $_[1], $_[2];
    }
    elsif (@_ == 2) {
        return CORE::split UTF8::R2::qr($_[0]), $_[1];
    }
    elsif (@_ == 1) {
        return CORE::split UTF8::R2::qr($_[0]);
    }
    else {
        return CORE::split;
    }
}

#---------------------------------------------------------------------
# substr() for UTF-8 codepoint string
CORE::eval sprintf <<'END', ($] >= 5.014) ? ':lvalue' : '';
#                            vv--------------*******
sub UTF8::R2::substr ($$;$$) %s {
    my @x = $_[0] =~ /\G$x/g;

    # If the substring is beyond either end of the string, substr() returns the undefined
    # value and produces a warning. When used as an lvalue, specifying a substring that
    # is entirely outside the string raises an exception.
    # http://perldoc.perl.org/functions/substr.html

    # A return with no argument returns the scalar value undef in scalar context,
    # an empty list () in list context, and (naturally) nothing at all in void
    # context.

    if (($_[1] < (-1 * scalar(@x))) or (+1 * scalar(@x) < $_[1])) {
        return;
    }

    # substr($string,$offset,$length,$replacement)
    if (@_ == 4) {
        my $substr = join '', splice @x, $_[1], $_[2], $_[3];
        $_[0] = join '', @x;
        $substr; # "return $substr" doesn't work, don't write "return"
    }

    # substr($string,$offset,$length)
    elsif (@_ == 3) {
        local $SIG{__WARN__} = sub {}; # avoid: Use of uninitialized value in join or string at here
        my $octet_offset =
            ($_[1] < 0) ? -1 * CORE::length(join '', @x[$#x+$_[1]+1 .. $#x])     :
            ($_[1] > 0) ?      CORE::length(join '', @x[0           .. $_[1]-1]) :
            0;
        my $octet_length =
            ($_[2] < 0) ? -1 * CORE::length(join '', @x[$#x+$_[2]+1 .. $#x])           :
            ($_[2] > 0) ?      CORE::length(join '', @x[$_[1]       .. $_[1]+$_[2]-1]) :
            0;
        CORE::substr($_[0], $octet_offset, $octet_length);
    }

    # substr($string,$offset)
    else {
        my $octet_offset =
            ($_[1] < 0) ? -1 * CORE::length(join '', @x[$#x+$_[1]+1 .. $#x])     :
            ($_[1] > 0) ?      CORE::length(join '', @x[0           .. $_[1]-1]) :
            0;
        CORE::substr($_[0], $octet_offset);
    }
}
END

#---------------------------------------------------------------------
# tr/A-C/1-3/ for UTF-8 codepoint
sub list_all_ASCII_by_hyphen {
    my @hyphened = @_;
    my @list_all = ();
    for (my $i=0; $i <= $#hyphened; ) {
        if (
            ($i+1 < $#hyphened)      and
            ($hyphened[$i+1] eq '-') and
        1) {
            $hyphened[$i+0] = ($hyphened[$i+0] eq '\\-') ? '-' : $hyphened[$i+0];
            $hyphened[$i+2] = ($hyphened[$i+2] eq '\\-') ? '-' : $hyphened[$i+2];
            if (0) { }
            elsif ($hyphened[$i+0] !~ m/\A [\x00-\x7F] \z/xms) {
                confess sprintf(qq{@{[__FILE__]}: "$hyphened[$i+0]-$hyphened[$i+2]" in tr/// is not US-ASCII});
            }
            elsif ($hyphened[$i+2] !~ m/\A [\x00-\x7F] \z/xms) {
                confess sprintf(qq{@{[__FILE__]}: "$hyphened[$i+0]-$hyphened[$i+2]" in tr/// is not US-ASCII});
            }
            elsif ($hyphened[$i+0] gt $hyphened[$i+2]) {
                confess sprintf(qq{@{[__FILE__]}: "$hyphened[$i+0]-$hyphened[$i+2]" in tr/// is not "$hyphened[$i+0]" le "$hyphened[$i+2]"});
            }
            else {
                push @list_all, map { CORE::chr($_) } (CORE::ord($hyphened[$i+0]) .. CORE::ord($hyphened[$i+2]));
                $i += 3;
            }
        }
        else {
            if ($hyphened[$i] eq '\\-') {
                push @list_all, '-';
            }
            else {
                push @list_all, $hyphened[$i];
            }
            $i++;
        }
    }
    return @list_all;
}

#---------------------------------------------------------------------
# tr/// for UTF-8 codepoint string
sub UTF8::R2::tr ($$$;$) {
    my @x           = $_[0] =~ /\G($x)/xmsg;
    my @search      = list_all_ASCII_by_hyphen($_[1] =~ /\G(\\-|$x)/xmsg);
    my @replacement = list_all_ASCII_by_hyphen($_[2] =~ /\G(\\-|$x)/xmsg);
    my %modifier    = (defined $_[3]) ? (map { $_ => 1 } CORE::split //, $_[3]) : ();

    my %tr = ();
    for (my $i=0; $i <= $#search; $i++) {

        # tr/AAA/123/ works as tr/A/1/
        if (not exists $tr{$search[$i]}) {

            # tr/ABC/123/ makes %tr = ('A'=>'1','B'=>'2','C'=>'3',);
            if (defined($replacement[$i]) and ($replacement[$i] ne '')) {
                $tr{$search[$i]} = $replacement[$i];
            }

            # tr/ABC/12/d makes %tr = ('A'=>'1','B'=>'2','C'=>'',);
            elsif (exists $modifier{d}) {
                $tr{$search[$i]} = '';
            }

            # tr/ABC/12/ makes %tr = ('A'=>'1','B'=>'2','C'=>'2',);
            elsif (defined($replacement[-1]) and ($replacement[-1] ne '')) {
                $tr{$search[$i]} = $replacement[-1];
            }

            # tr/ABC// makes %tr = ('A'=>'A','B'=>'B','C'=>'C',);
            else {
                $tr{$search[$i]} = $search[$i];
            }
        }
    }

    my $tr = 0;
    my $replaced = '';

    # has /c modifier
    if (exists $modifier{c}) {

        # has /s modifier
        if (exists $modifier{s}) {
            my $last_transliterated = undef;
            while (defined(my $x = shift @x)) {

                # /c modifier works here
                if (exists $tr{$x}) {
                    $replaced .= $x;
                    $last_transliterated = undef;
                }
                else {

                    # /d modifier works here
                    if (exists $modifier{d}) {
                    }

                    elsif (defined $replacement[-1]) {

                        # /s modifier works here
                        if (defined($last_transliterated) and ($replacement[-1] eq $last_transliterated)) {
                        }

                        # tr/// works here
                        else {
                            $replaced .= ($last_transliterated = $replacement[-1]);
                        }
                    }
                    $tr++;
                }
            }
        }

        # has no /s modifier
        else {
            while (defined(my $x = shift @x)) {

                # /c modifier works here
                if (exists $tr{$x}) {
                    $replaced .= $x;
                }
                else {

                    # /d modifier works here
                    if (exists $modifier{d}) {
                    }

                    # tr/// works here
                    elsif (defined $replacement[-1]) {
                        $replaced .= $replacement[-1];
                    }
                    $tr++;
                }
            }
        }
    }

    # has no /c modifier
    else {

        # has /s modifier
        if (exists $modifier{s}) {
            my $last_transliterated = undef;
            while (defined(my $x = shift @x)) {
                if (exists $tr{$x}) {

                    # /d modifier works here
                    if ($tr{$x} eq '') {
                    }

                    # /s modifier works here
                    elsif (defined($last_transliterated) and ($tr{$x} eq $last_transliterated)) {
                    }

                    # tr/// works here
                    else {
                        $replaced .= ($last_transliterated = $tr{$x});
                    }
                    $tr++;
                }
                else {
                    $replaced .= $x;
                    $last_transliterated = undef;
                }
            }
        }

        # has no /s modifier
        else {
            while (defined(my $x = shift @x)) {
                if (exists $tr{$x}) {
                    $replaced .= $tr{$x};
                    $tr++;
                }
                else {
                    $replaced .= $x;
                }
            }
        }
    }

    # /r modifier works here
    if (exists $modifier{r}) {
        return $replaced;
    }

    # has no /r modifier
    else {
        $_[0] = $replaced;
        return $tr;
    }
}

#---------------------------------------------------------------------
# universal uc() for UTF-8 codepoint string
sub UTF8::R2::uc (;$) {
    local $_ = @_ ? $_[0] : $_;
    #                          a A b B c C d D e E f F g G h H i I j J k K l L m M n N o O p P q Q r R s S t T u U v V w W x X y Y z Z
    return join '', map { {qw( a A b B c C d D e E f F g G h H i I j J k K l L m M n N o O p P q Q r R s S t T u U v V w W x X y Y z Z )}->{$_}||$_ } /\G$x/g;
    #                          a A b B c C d D e E f F g G h H i I j J k K l L m M n N o O p P q Q r R s S t T u U v V w W x X y Y z Z
}

#---------------------------------------------------------------------
# universal ucfirst() for UTF-8 codepoint string
sub UTF8::R2::ucfirst (;$) {
    local $_ = @_ ? $_[0] : $_;
    if (/\A($x)(.*)\z/s) {
        return UTF8::R2::uc($1) . $2;
    }
    else {
        return '';
    }
}

# syntax sugar for UTF-8 codepoint regex
#
# tie my %mb, 'UTF8::R2';
# $result = $_ =~ $mb{qr/$utf8regex/imsxo}
# $result = $_ =~ m<\G$mb{qr/$utf8regex/imsxo}>gc
# $result = $_ =~ s<$mb{qr/before/imsxo}>egr

sub TIEHASH  { bless { }, $_[0] }
sub FETCH    { UTF8::R2::qr $_[1] }
sub STORE    { }
sub FIRSTKEY { }
sub NEXTKEY  { }
sub EXISTS   { }
sub DELETE   { }
sub CLEAR    { }
sub UNTIE    { }
sub DESTROY  { }
sub SCALAR   { }

1;

__END__