Message/MIME/Charset.pm


=head1 NAME

Message::MIME::Charset Perl module

=head1 DESCRIPTION

Perl module for MIME charset.

=cut

## NOTE: You should not require/use other module (even it
##       is part of Message::* Perl Modules) as far as possible,
##       to be able to use this module (M::M::Charset) from
##       other (non-Message::*) modules.

package Message::MIME::Charset;
use strict;
use vars qw(%CHARSET %MSNAME2IANANAME %REG $VERSION);
$VERSION=do{my @r=(q$Revision: 1.14 $=~/\d+/g);sprintf "%d."."%02d" x $#r,@r};

&_builtin_charset;
sub _builtin_charset () {

$CHARSET{'*DEFAULT'} = {
        preferred_name  => '',
        
        encoder => sub { $_[1] },
        decoder => sub { $_[1] },
        
        mime_text       => 1,   ## Suitability in use as MIME text/* charset
        #accept_cte     => [qw/7bit .../],
        cte_7bit_preferred      => 'quoted-printable',
};
$CHARSET{'*default'} = $CHARSET{'*DEFAULT'};

$CHARSET{'us-ascii'} = {
        preferred_name  => 'us-ascii',
        
        encoder => sub { $_[1] },
        decoder => sub { $_[1] },
        
        mime_text       => 1,
        cte_7bit_preferred      => 'quoted-printable',
};

$CHARSET{'iso-2022-int-1'} = {
        preferred_name  => 'iso-2022-int-1',
        
        encoder => sub { $_[1] },
        decoder => sub { $_[1] },
        
        mime_text       => 1,
};

$CHARSET{'unknown-8bit'} = {
        preferred_name  => 'unknown-8bit',
        
        encoder => sub { $_[1] },
        decoder => sub { $_[1] },
        
        mime_text       => 1,
        cte_7bit_preferred      => 'base64',
};

$CHARSET{'x-unknown'} = {
        preferred_name  => 'x-unknown',
        
        encoder => sub { $_[1] },
        decoder => sub { $_[1] },
        
        mime_text       => 0,
        cte_7bit_preferred      => 'base64',
};

$CHARSET{'*undef'} = {
        preferred_name  => '',
        
        #encoder        => sub { $_[1] },
        #decoder        => sub { $_[1] },
        
        mime_text       => 0,
        cte_7bit_preferred      => 'base64',
};

}       # /builtin_charset

my %_MINIMUMIZER = (
        'euc-jp'        => \&_name_euc_japan,
        'euc-jisx0213'  => \&_name_euc_japan,
        'euc-jisx0213-plane1'   => \&_name_euc_japan,
        'x-euc-jisx0213-packed' => \&_name_euc_japan,
        'x-iso-2022'    => \&_name_8bit_iso2022,
        'iso-2022-cn'   => \&_name_8bit_iso2022,
        'iso-2022-cn-ext'       => \&_name_8bit_iso2022,
        'iso-2022-int-1'        => \&_name_net_ascii_8bit,
        'iso-2022-jp'   => \&_name_8bit_iso2022,
        'iso-2022-jp-1' => \&_name_8bit_iso2022,
        'iso-2022-jp-2' => \&_name_8bit_iso2022,
        'iso-2022-jp-3' => \&_name_8bit_iso2022,
        'iso-2022-jp-3-plane1'  => \&_name_8bit_iso2022,
        'iso-2022-kr'   => \&_name_8bit_iso2022,
        'iso-8859-1'    => \&_name_8bit_iso2022,
        'iso-10646-j-1' => \&_name_utf16be,
        'iso-10646-ucs-2'       => \&_name_utf16be,
        'iso-10646-ucs-4'       => \&_name_utf32be,
        'iso-10646-ucs-basic'   => \&_name_utf16be,
        'iso-10646-unicode-latin1'      => \&_name_utf16be,
        jis_x0201       => \&_name_shift_jis,
        junet   => \&_name_8bit_iso2022,
        'x-junet8'      => \&_name_net_ascii_8bit,
        shift_jis       => \&_name_shift_jis,
        shift_jisx0213  => \&_name_shift_jis,
        'shift_jisx0213-plane1' => \&_name_shift_jis,
        'x-sjis'        => \&_name_shift_jis,
        'us-ascii'      => \&_name_net_ascii_8bit,
        'utf-8' => \&_name_net_ascii_8bit,
        'utf-16be'      => \&_name_utf16be,
        'utf-32be'      => \&_name_utf32be,
);

my %_IsMimeText;
for (qw(
        adobe-standard-encoding adobe-symbol-encoding
        big5    big5-eten       big5-hkscs
        cp950
        gbk     gb18030
        euc-jp  euc-jisx0213    euc-kr  euc-tw
        hp-roman8
        hz-gb-2312
        ibm437
        iso-2022-cn     iso-2022-cn-ext
        iso-2022-int-1
        iso-2022-jp     iso-2022-jp-1   iso-2022-jp-2   iso-2022-jp-3
        x-iso2022jp-cp932
        iso-2022-kr
        iso-8859-1      iso-8859-2      iso-8859-3
        iso-8859-4      iso-8859-5      iso-8859-6
        iso-8859-7      iso-8859-8      iso-8859-9
        iso-8859-10     iso-8859-12     iso-8859-13
        iso-8859-14     iso-8859-15     iso-8859-16
        jis_encoding
        koi8-r  koi8-u
        x-mac-arabic    x-mac-centralroman      x-mac-cyrillic  x-mac-greek
        x-mac-hebrew    x-mac-icelandic macintosh       x-mac-turkish
        x-mac-ukrainian x-mac-chinesesimp       x-mac-japanese  x-mac-korean
        shift_jis       shift_jisx0213  x-sjis
        tis-620
        unicode-1-1-utf-7       unicode-1-1-utf-8
        unicode-2-0-utf-7       unicode-2-0-utf-8
        utf-7   utf-8   utf-9
        viscii
        windows-1250    windows-1251    windows-1252    windows-1253
        windows-1254    windows-1255    windows-1256    windows-1257
        windows-1258    windows-31j     windows-949
)) { $_IsMimeText{$_} = 1 }

%MSNAME2IANANAME = (
        'iso-2022-jp'   => 'x-iso2022jp-cp932',
        'ks_c_5601-1987'        => 'windows-949',
);

sub make_charset ($%) {
  my $name = shift;
  return unless $name;  ## Note: charset "0" is not supported.
  my %definition = @_;
  
  $definition{preferred_name} ||= $name;
  if ($definition{preferred_name} ne $name
      && ref $CHARSET{$definition{preferred_name}}) {
  ## New charset is an alias of defined charset,
    $CHARSET{$name} = $CHARSET{$definition{preferred_name}};
    return;
  } elsif ($definition{alias_of} && ref $CHARSET{$definition{alias_of}}) {
  ## New charset is an alias of defined charset,
    $CHARSET{$name} = $CHARSET{$definition{alias_of}};
    return;
  }
  $CHARSET{$name} = \%definition;
  
  ## Set default values
  #$definition{encoder} ||= sub { $_[1] };
  #$definition{decoder} ||= sub { $_[1] };

  $definition{mime_text} = 0 unless defined $definition{mime_text};
  $definition{cte_7bit_preferred} = 'base64'
    unless defined $definition{cte_7bit_preferred};
}

sub encode ($$) {
  my ($charset, $s) = (lc shift, shift);
  my $c = ref $CHARSET{$charset}->{encoder}? $charset: '*undef';
  if (ref $CHARSET{$c}->{encoder}) {
    my ($t, %r) = &{$CHARSET{$c}->{encoder}} ($charset, $s);
    unless (defined $r{success}) {
      $r{success} = 1;
    }
    return ($t, %r);
  }
  ($s, success => 0);
}

sub decode ($$) {
  my ($charset, $s) = (lc shift, shift);
  my $c = ref $CHARSET{$charset}->{decoder}? $charset: '*undef';
  if (ref $CHARSET{$c}->{decoder}) {
    my ($t, %r) = &{$CHARSET{$c}->{decoder}} ($charset, $s);
    unless (defined $r{success}) {
      $r{success} = 1;
    }
    return ($t, %r);
  }
  ($s, success => 0);
}

sub name_normalize ($) {
  my $name = lc shift;
  if (ref $CHARSET{$name}->{preferred_name} eq 'CODE') {
    return &{ $CHARSET{$name}->{preferred_name} } ($name);
  } elsif ($CHARSET{$name}->{preferred_name}) {
    return $CHARSET{$name}->{preferred_name};
  } elsif (ref $CHARSET{'*undef'}->{preferred_name} eq 'CODE') {
    return &{ $CHARSET{'*undef'}->{preferred_name} } ($name);
  }
  $name;
}

sub name_minimumize ($$) {
  my ($charset, $s) = (lc shift, shift);
  if (ref $CHARSET{$charset}->{name_minimumizer} eq 'CODE') {
    return &{$CHARSET{$charset}->{name_minimumizer}} ($charset, $s);
  } elsif (ref $_MINIMUMIZER{$charset}) {
    return &{$_MINIMUMIZER{$charset}} ($charset, $s);
  } elsif (ref $CHARSET{'*undef'}->{name_minimumizer} eq 'CODE') {
    return &{$CHARSET{'*undef'}->{name_minimumizer}} ($charset, $s);
  }
  (charset => $charset);
}

sub msname2iananame ($) {
  my $mscharset = shift;
  $MSNAME2IANANAME{$mscharset} || $mscharset;
}

sub _name_7bit_iso2022 ($$) {shift;
  my $s = shift;
  if ($s =~ /[\x0E\x0F\x1B]/) {
    return (charset => 'iso-2022-jp')
      unless $s =~ /\x1B[^\x24\x28]
                   |\x1B\x24[^\x40B]
                   |\x1B\x28[^BJ]
                   |\x0E|\x0F/x;
    return (charset => 'iso-2022-jp-1')
      unless $s =~ /\x1B[^\x24\x28]
                   |\x1B\x24[^\x40B\x28]
                   |\x1B\x24\x28[^D]
                   |\x1B\x28[^BJ]
                   |\x0E|\x0F/x;
    return (charset => 'iso-2022-jp-3-plane1')
      unless $s =~ /\x1B[^\x24\x28]
                   |\x1B\x24[^\x28]     #[^B\x28]
                   |\x1B\x24\x28[^O]
                   |\x1B\x28[^B]
                   |\x0E|\x0F/x;
    return (charset => 'iso-2022-jp-3')
      unless $s =~ /\x1B[^\x24\x28]
                   |\x1B\x24[^\x28]     #[^B\x28]
                   |\x1B\x24\x28[^OP]
                   |\x1B\x28[^B]
                   |\x0E|\x0F/x;
    return (charset => 'iso-2022-kr')
      unless $s =~ /\x1B[^\x24]
                   |\x1B\x24[^\x29]
                   |\x1B\x24\x29[^C]/x;
    return (charset => 'iso-2022-jp-2')
      unless $s =~ /\x1B[^\x24\x28\x2E\x4E]
                   |\x1B\x24[^\x40AB\x28]
                   |\x1B\x24\x28[^CD]
                   |\x1B\x28[^BJ]
                   |\x1B\x2E[^AF]
                   |\x0E|\x0F/x;
    return (charset => 'iso-2022-cn')
      unless $s =~ /\x1B[^\x4E\x24]
                   |\x1B\x24[^\x29\x2A]
                   |\x1B\x24\x29[^AG]
                   |\x1B\x24\x2A[^H]/x;
    return (charset => 'iso-2022-cn-ext')
      unless $s =~ /\x1B[^\x4E\x4F\x24]
                   |\x1B\x24[^\x29\x2A]
                   |\x1B\x24\x29[^AEG]
                   |\x1B\x24\x2A[^HIJKLM]/x;
    return (charset => 'iso-2022-int-1')
      unless $s =~ /\x1B[^\x24\x28\x2D]
                   |\x1B\x24[^\x40AB\x28\x29]
                   |\x1B\x24\x28[^DGH]
                   |\x1B\x24\x29[^C]
                   |\x1B\x28[^BJ]
                   |\x1B\x2D[^AF]/x;
    return (charset => 'junet')
      unless $s =~ /\x1B[^\x24\x28\x2C]
                   |\x1B\x24[^\x28\x2C\x40-\x42]
                   |\x1B\x24[\x28\x2C][^\x20-\x7E]
                   |\x1B\x24[\x28\x2C][\x20-\x2F]+[^\x30-\x7E]
                   |\x1B[\x28\x2C][^\x20-\x7E]
                   |\x1B[\x28\x2C][\x20-\x2F]+[^\x30-\x7E]
                   |\x0E|\x0F/x;
    return (charset => 'x-iso-2022');
  } else {
    return (charset => 'us-ascii');
  }
}

sub _name_net_ascii_8bit ($) {
  my $name = shift; my $s = shift;
  return (charset => 'us-ascii') unless $s =~ /[\x1B\x0E\x0F\x80-\xFF]/;
  if ($s =~ /[\x80-\xFF]/) {
    if ($s =~ /[\xC0-\xFD][\x80-\xBF]*[\x80-\x8F]/) {
      if ($s =~ /\x1B/) {
        return (charset => 'x-junet8'); ## junet + UTF-8
      } else {
        return (charset => 'utf-8');
      }
    } elsif ($s =~ /\x1B/) {
      return (charset => 'x-iso-2022'); ## 8bit ISO 2022
    } else {
      return (charset => 'iso-8859-1');
    }
  } else {      ## 7bit ISO 2022
    return _name_7bit_iso2022 ($name, $s);
  }
}

sub _name_8bit_iso2022 ($$) {
  my $name = shift; my $s = shift;
  return (charset => 'us-ascii') unless $s =~ /[\x1B\x0E\x0F\x80-\xFF]/;
  if ($s =~ /[\x80-\xFF]/) {
    if ($s =~ /\x1B/) {
      return (charset => 'x-iso-2022'); ## 8bit ISO 2022
    } else {
      return (charset => 'iso-8859-1');
    }
  } else {      ## 7bit ISO 2022
    return _name_7bit_iso2022 ($name, $s);
  }
}

## Not completed.
## TODO: gb18030, cn-gb-12345
## TODO: _name_euc_gbf (cn-gb-12345, gb2312)
sub _name_euc_gb ($$) {
  my $name = shift; my $s = shift;
  if ($s =~ /[\x80-\xFF]/) {
    if ($s =~ /
                  (?:\G|[\x00-\x3F\x7F\x80\xFF])
                  (?:[\xA1-\xA9\xB0-\xFE][\xA1-\xFE]
                    |[\x40-\x7E])*
        (?:
          [\x81-\xA0\xAA-\xAF][\x40-\xFE]
         |[\xA1-\xFE][\x40-\xA0]
        )
      /x) {
      (charset => 'gbk');
    } elsif ($s =~ /
        (?:\xA2[\xA1-\xAA]
          |\xA6[\xE0-\xF5]
          |\xA8[\xBB-\xC0]
        )
          (?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]|\z))
      /x) {
      (charset => 'gbk');
    } elsif ($s =~ /
        (?:\xA3\xE7|\xA7[\xDD-\xF2]
          |\xA8[\xBB-\xC0]
          |[\xAA-\xAF\xF8-\xFE][\xA1-\xFE]
        )
          (?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]|\z))
      /x) {
      (charset => 'cn-gb-isoir165', 'charset-edition' => 1992);
    } elsif ($s =~ /\xEF\xF1    ## Typo bug of GB 2312
          (?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]|\z))
      /x) {
      (charset => 'gb2312');
    } else {
      (charset => 'gb2312', 'charset-edition' => 1980);
    }
  } elsif ($s =~ /[\x0E\x0F]/) {
    (charset => 'gb2312');      ## Actually, this is not "gb2312"
  } else {
    _name_7bit_iso2022 ($name, $s);
  }
}

sub _name_euc_japan ($$) {
  my $name = shift; my $s = shift;
  if ($s =~ /[\x80-\xFF]/) {
    if ($s =~ /\x8F[\xA1\xA3-\xA5\xA8\xAC-\xAF\xEE-\xFE][\xA1-\xFE]/) {
      if ($s =~ /\x8F[\xA2\xA6\xA7\xA9-\xAB\xB0-\xED][\xA1-\xFE]/) {
      ## JIS X 0213 plane 2 + JIS X 0212
        (charset => 'x-euc-jisx0213-packed');
      } else {
        (charset => 'euc-jisx0213');
      }
    } elsif ($s =~ m{(?<![\x8E\x8F])    ## Not G2/G3 character
                    (?: ## JIS X 0213:2000
                       [\xA9-\xAF\xF5-\xFE][\xA1-\xFE]
                      |\xA2[\xAF-\xB9\xC2-\xC9\xD1-\xDB\xE9-\xF1\xFA-\xFD]
                      |\xA3[\xA1-\xAF\xBA-\xC0\xDB-\xE0\xFB-\xFE]
                      |\xA4[\xF4-\xFE]|\xA5[\xF7-\xFE]
                      |\xA6[\xB9-\xC0\xD9-\xFE]|\xA7[\xC2-\xD0\xF2-\xFE]
                      |\xA8[\xC1-\xFE]|\xCF[\xD4-\xFE]|\xF4[\xA7-\xFE]
                    )
                    (?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]|\z))}x) {
      if ($s =~ /\x8F/) {       ## JIS X 0213 plane 1 + JIS X 0212
        (charset => 'x-euc-jisx0213-packed');
      } else {
        (charset => 'euc-jisx0213-plane1');
      }
    } else {
      (charset => 'euc-jp');
    }
  } elsif ($s =~ /\x0E|\x0F|\x1B[\x4E\x4F]/) {
    (charset => 'euc-jisx0213');        ## Actually, this is not euc-japan
  } else {
    _name_7bit_iso2022 ($name, $s);
  }
}

sub _name_shift_jis ($$) {
  my $name = shift; my $s = shift;
  if ($s =~ /[\x80-\xFF]/) {
    if ($s =~ /[\x0E\x0F\x1B]/) {
      (charset => 'x-sjis');
    } elsif ($s =~ /
                  (?:\G|[\x00-\x3F\x7F])
                  (?:[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC]
                    |[\x40-\x7E\xA1-\xDF])*
               [\xF0-\xFC][\x40-\x7E\x80-\xFC]
      /x) {
      (charset => 'shift_jisx0213');
    } elsif ($s =~ /
                  (?:\G|[\x00-\x3F\x7F])
                  (?:[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC]
                    |[\x40-\x7E\xA1-\xDF])*
              (?:
               [\x85-\x87\xEB-\xEF][\x40-\x7E\x80-\xFC]
              |\x81[\xAD-\xB7\xC0-\xC7\xCF-\xD9\xE9-\xEF\xF8-\xFB]
              |\x82[\x40-\x4E\x59-\x5F\x7A-\x80\x9B-\x9E\xF2-\xFC]
              |\x83[\x97-\x9E\xB7-\xBE\xD7-\xFC]
              |\x84[\x61-\x6F\x72-\x9E\xBF-\xFC]
              |\x88[\x40-\x9E]|\x98[\x73-\x9E]|\xEA[\xA5-\xFC]
              )
    /x) {
      (charset => 'shift_jisx0213-plane1');
    } else {
      (charset => 'shift_jis');
    }
  } elsif ($s =~ /[\x5C\x7E]/) {
    if ($s =~ /\x1B\x0E\x0F/) {
      (charset => 'x-sjis');    ## ISO 2022 with implied "ESC ( J"
        ## BUG: "ESC ( B foobar\aaa ESC ( J aiueo" also matchs this
    } else {
      (charset => 'jis_x0201');
    }
  } else {
    _name_7bit_iso2022 ($name, $s);
  }
}

sub _name_utf16be ($$) {
  shift; my $s = shift;
  if ($s =~ /[\xD8-\xDB][\x00-\xFF][\xDC-\xDF][\x00-\xFF]
             (?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
    (charset => 'utf-16be');
  } elsif ($s =~ /[\x01-\xFF][\x00-\xFF]
             (?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
    if ($s =~ /([^\x00\x03\x04\x23\x25\x30\xFE\xFF]
                     [\x00-\xFF]        # ^\x20\x22\x4E-\x9F\xF9\xFA
                  |\x03[^\x00-\x6F\xD0-\xFF]
                  #|\x20[^\x00-\x6F]
                  |\x25[^\x00-\x7F]
                  |\xFE[^\x30-\x4F]
                  |\xFF[^\x00-\xEF]
                  ## note 1 of RFC 1816 is ambitious, so block entire
                  ## is excepted
                    |\x30[\x00-\x3F]
                  )
             (?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
      (charset => 'iso-10646-ucs-2');
    } else {
      (charset => 'iso-10646-j-1');
    }
  } elsif ($s =~ /\x00[\x80-\xFF]
             (?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
    (charset => 'iso-10646-unicode-latin1');
  } else {
    (charset => 'iso-10646-ucs-basic');
  }
}

sub _name_utf32be ($$) {
  shift; my $s = shift;
  if ($s =~ /
    ([\x01-\x7F][\x00-\xFF]{3}
    |\x00[\x11-\xFF][\x00-\xFF][\x00-\xFF])
             (?=(?:[\x00-\xFF]{4})*\z)/sx) {
    (charset => 'iso-10646-ucs-4');
  } else {
    (charset => 'utf-32be');
  }
}

sub _utf8_on ($) {
  Encode::_utf8_on ($_[0]) if $Encode::VERSION;
}
sub _utf8_off ($) {
  Encode::_utf8_off ($_[0]) if $Encode::VERSION;
}

sub is_mime_text ($) {
  my $name = lc shift;
  if (ref $CHARSET{$name}->{mime_text} eq 'CODE') {
    return &{ $CHARSET{$name}->{mime_text} } ($name);
  } elsif (defined $CHARSET{$name}->{mime_text}) {
    return $CHARSET{$name}->{mime_text};
  } elsif (defined $_IsMimeText{$name}) {
    return $_IsMimeText{$name};
  } elsif (ref $CHARSET{'*undef'}->{mime_text} eq 'CODE') {
    return &{ $CHARSET{'*undef'}->{mime_text} } ($name);
  }
  0;
}

=head1 LICENSE

Copyright 2002 wakaba E<lt>w@suika.fam.cxE<gt>.

This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program; see the file COPYING.  If not, write to
the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA.

=head1 CHANGE

See F<ChangeLog>.
$Date: 2002/07/22 07:48:01 $

=cut

1;
1	wakaba	1.1
2			=head1 NAME
3
4			Message::MIME::Charset Perl module
5
6			=head1 DESCRIPTION
7
8			Perl module for MIME charset.
9
10			=cut
11
12	wakaba	1.10	## NOTE: You should not require/use other module (even it
13			## is part of Message::* Perl Modules) as far as possible,
14			## to be able to use this module (M::M::Charset) from
15			## other (non-Message::*) modules.
16
17	wakaba	1.1	package Message::MIME::Charset;
18			use strict;
19	wakaba	1.12	use vars qw(%CHARSET %MSNAME2IANANAME %REG $VERSION);
20	wakaba	1.15	$VERSION=do{my @r=(q$Revision: 1.14 $=~/\d+/g);sprintf "%d."."%02d" x $#r,@r};
21	wakaba	1.1
22	wakaba	1.9	&_builtin_charset;
23			sub _builtin_charset () {
24	wakaba	1.4
25			$CHARSET{'*DEFAULT'} = {
26			preferred_name => '',
27
28			encoder => sub { $_[1] },
29			decoder => sub { $_[1] },
30
31			mime_text => 1, ## Suitability in use as MIME text/* charset
32			#accept_cte => [qw/7bit .../],
33			cte_7bit_preferred => 'quoted-printable',
34			};
35			$CHARSET{'default'} = $CHARSET{'DEFAULT'};
36
37			$CHARSET{'us-ascii'} = {
38			preferred_name => 'us-ascii',
39
40			encoder => sub { $_[1] },
41			decoder => sub { $_[1] },
42	wakaba	1.5
43			mime_text => 1,
44	wakaba	1.13	cte_7bit_preferred => 'quoted-printable',
45	wakaba	1.4	};
46
47			$CHARSET{'iso-2022-int-1'} = {
48			preferred_name => 'iso-2022-int-1',
49
50			encoder => sub { $_[1] },
51			decoder => sub { $_[1] },
52	wakaba	1.5
53			mime_text => 1,
54	wakaba	1.4	};
55
56			$CHARSET{'unknown-8bit'} = {
57			preferred_name => 'unknown-8bit',
58
59			encoder => sub { $_[1] },
60			decoder => sub { $_[1] },
61
62	wakaba	1.13	mime_text => 1,
63			cte_7bit_preferred => 'base64',
64			};
65
66			$CHARSET{'x-unknown'} = {
67			preferred_name => 'x-unknown',
68
69			encoder => sub { $_[1] },
70			decoder => sub { $_[1] },
71
72			mime_text => 0,
73			cte_7bit_preferred => 'base64',
74			};
75
76			$CHARSET{'*undef'} = {
77			preferred_name => '',
78
79			#encoder => sub { $_[1] },
80			#decoder => sub { $_[1] },
81
82	wakaba	1.4	mime_text => 0,
83	wakaba	1.6	cte_7bit_preferred => 'base64',
84	wakaba	1.4	};
85
86	wakaba	1.9	} # /builtin_charset
87
88			my %_MINIMUMIZER = (
89			'euc-jp' => \&_name_euc_japan,
90			'euc-jisx0213' => \&_name_euc_japan,
91			'euc-jisx0213-plane1' => \&_name_euc_japan,
92			'x-euc-jisx0213-packed' => \&_name_euc_japan,
93	wakaba	1.11	'x-iso-2022' => \&_name_8bit_iso2022,
94			'iso-2022-cn' => \&_name_8bit_iso2022,
95			'iso-2022-cn-ext' => \&_name_8bit_iso2022,
96	wakaba	1.9	'iso-2022-int-1' => \&_name_net_ascii_8bit,
97	wakaba	1.11	'iso-2022-jp' => \&_name_8bit_iso2022,
98			'iso-2022-jp-1' => \&_name_8bit_iso2022,
99			'iso-2022-jp-2' => \&_name_8bit_iso2022,
100			'iso-2022-jp-3' => \&_name_8bit_iso2022,
101			'iso-2022-jp-3-plane1' => \&_name_8bit_iso2022,
102			'iso-2022-kr' => \&_name_8bit_iso2022,
103			'iso-8859-1' => \&_name_8bit_iso2022,
104	wakaba	1.9	'iso-10646-j-1' => \&_name_utf16be,
105			'iso-10646-ucs-2' => \&_name_utf16be,
106			'iso-10646-ucs-4' => \&_name_utf32be,
107			'iso-10646-ucs-basic' => \&_name_utf16be,
108			'iso-10646-unicode-latin1' => \&_name_utf16be,
109			jis_x0201 => \&_name_shift_jis,
110	wakaba	1.11	junet => \&_name_8bit_iso2022,
111	wakaba	1.9	'x-junet8' => \&_name_net_ascii_8bit,
112			shift_jis => \&_name_shift_jis,
113			shift_jisx0213 => \&_name_shift_jis,
114			'shift_jisx0213-plane1' => \&_name_shift_jis,
115			'x-sjis' => \&_name_shift_jis,
116			'us-ascii' => \&_name_net_ascii_8bit,
117			'utf-8' => \&_name_net_ascii_8bit,
118			'utf-16be' => \&_name_utf16be,
119			'utf-32be' => \&_name_utf32be,
120			);
121
122	wakaba	1.14	my %_IsMimeText;
123			for (qw(
124			adobe-standard-encoding adobe-symbol-encoding
125			big5 big5-eten big5-hkscs
126			cp950
127			gbk gb18030
128			euc-jp euc-jisx0213 euc-kr euc-tw
129			hp-roman8
130			hz-gb-2312
131			ibm437
132			iso-2022-cn iso-2022-cn-ext
133			iso-2022-int-1
134			iso-2022-jp iso-2022-jp-1 iso-2022-jp-2 iso-2022-jp-3
135	wakaba	1.15	x-iso2022jp-cp932
136	wakaba	1.14	iso-2022-kr
137			iso-8859-1 iso-8859-2 iso-8859-3
138			iso-8859-4 iso-8859-5 iso-8859-6
139			iso-8859-7 iso-8859-8 iso-8859-9
140			iso-8859-10 iso-8859-12 iso-8859-13
141			iso-8859-14 iso-8859-15 iso-8859-16
142			jis_encoding
143			koi8-r koi8-u
144			x-mac-arabic x-mac-centralroman x-mac-cyrillic x-mac-greek
145			x-mac-hebrew x-mac-icelandic macintosh x-mac-turkish
146			x-mac-ukrainian x-mac-chinesesimp x-mac-japanese x-mac-korean
147			shift_jis shift_jisx0213 x-sjis
148			tis-620
149			unicode-1-1-utf-7 unicode-1-1-utf-8
150			unicode-2-0-utf-7 unicode-2-0-utf-8
151			utf-7 utf-8 utf-9
152			viscii
153			windows-1250 windows-1251 windows-1252 windows-1253
154			windows-1254 windows-1255 windows-1256 windows-1257
155			windows-1258 windows-31j windows-949
156			)) { $_IsMimeText{$_} = 1 }
157
158	wakaba	1.12	%MSNAME2IANANAME = (
159			'iso-2022-jp' => 'x-iso2022jp-cp932',
160			'ks_c_5601-1987' => 'windows-949',
161			);
162
163	wakaba	1.4	sub make_charset ($%) {
164			my $name = shift;
165			return unless $name; ## Note: charset "0" is not supported.
166			my %definition = @_;
167	wakaba	1.6
168			$definition{preferred_name} \|\|= $name;
169	wakaba	1.4	if ($definition{preferred_name} ne $name
170			&& ref $CHARSET{$definition{preferred_name}}) {
171			## New charset is an alias of defined charset,
172			$CHARSET{$name} = $CHARSET{$definition{preferred_name}};
173			return;
174			} elsif ($definition{alias_of} && ref $CHARSET{$definition{alias_of}}) {
175			## New charset is an alias of defined charset,
176			$CHARSET{$name} = $CHARSET{$definition{alias_of}};
177			return;
178			}
179			$CHARSET{$name} = \%definition;
180
181			## Set default values
182	wakaba	1.6	#$definition{encoder} \|\|= sub { $_[1] };
183			#$definition{decoder} \|\|= sub { $_[1] };
184	wakaba	1.4
185			$definition{mime_text} = 0 unless defined $definition{mime_text};
186			$definition{cte_7bit_preferred} = 'base64'
187			unless defined $definition{cte_7bit_preferred};
188			}
189	wakaba	1.1
190			sub encode ($$) {
191			my ($charset, $s) = (lc shift, shift);
192	wakaba	1.13	my $c = ref $CHARSET{$charset}->{encoder}? $charset: '*undef';
193			if (ref $CHARSET{$c}->{encoder}) {
194			my ($t, %r) = &{$CHARSET{$c}->{encoder}} ($charset, $s);
195			unless (defined $r{success}) {
196			$r{success} = 1;
197			}
198			return ($t, %r);
199	wakaba	1.1	}
200	wakaba	1.5	($s, success => 0);
201	wakaba	1.1	}
202
203			sub decode ($$) {
204			my ($charset, $s) = (lc shift, shift);
205	wakaba	1.13	my $c = ref $CHARSET{$charset}->{decoder}? $charset: '*undef';
206			if (ref $CHARSET{$c}->{decoder}) {
207			my ($t, %r) = &{$CHARSET{$c}->{decoder}} ($charset, $s);
208			unless (defined $r{success}) {
209			$r{success} = 1;
210			}
211			return ($t, %r);
212	wakaba	1.1	}
213	wakaba	1.13	($s, success => 0);
214	wakaba	1.1	}
215
216			sub name_normalize ($) {
217			my $name = lc shift;
218	wakaba	1.14	if (ref $CHARSET{$name}->{preferred_name} eq 'CODE') {
219			return &{ $CHARSET{$name}->{preferred_name} } ($name);
220			} elsif ($CHARSET{$name}->{preferred_name}) {
221			return $CHARSET{$name}->{preferred_name};
222			} elsif (ref $CHARSET{'*undef'}->{preferred_name} eq 'CODE') {
223			return &{ $CHARSET{'*undef'}->{preferred_name} } ($name);
224			}
225			$name;
226	wakaba	1.1	}
227
228	wakaba	1.6	sub name_minimumize ($$) {
229			my ($charset, $s) = (lc shift, shift);
230	wakaba	1.9	if (ref $CHARSET{$charset}->{name_minimumizer} eq 'CODE') {
231	wakaba	1.6	return &{$CHARSET{$charset}->{name_minimumizer}} ($charset, $s);
232	wakaba	1.9	} elsif (ref $_MINIMUMIZER{$charset}) {
233			return &{$_MINIMUMIZER{$charset}} ($charset, $s);
234	wakaba	1.13	} elsif (ref $CHARSET{'*undef'}->{name_minimumizer} eq 'CODE') {
235			return &{$CHARSET{'*undef'}->{name_minimumizer}} ($charset, $s);
236	wakaba	1.9	}
237			(charset => $charset);
238			}
239
240	wakaba	1.12	sub msname2iananame ($) {
241			my $mscharset = shift;
242			$MSNAME2IANANAME{$mscharset} \|\| $mscharset;
243			}
244
245	wakaba	1.9	sub _name_7bit_iso2022 ($$) {shift;
246			my $s = shift;
247			if ($s =~ /[\x0E\x0F\x1B]/) {
248			return (charset => 'iso-2022-jp')
249			unless $s =~ /\x1B[^\x24\x28]
250			\|\x1B\x24[^\x40B]
251			\|\x1B\x28[^BJ]
252			\|\x0E\|\x0F/x;
253			return (charset => 'iso-2022-jp-1')
254			unless $s =~ /\x1B[^\x24\x28]
255			\|\x1B\x24[^\x40B\x28]
256			\|\x1B\x24\x28[^D]
257			\|\x1B\x28[^BJ]
258			\|\x0E\|\x0F/x;
259			return (charset => 'iso-2022-jp-3-plane1')
260			unless $s =~ /\x1B[^\x24\x28]
261			\|\x1B\x24[^\x28] #[^B\x28]
262			\|\x1B\x24\x28[^O]
263			\|\x1B\x28[^B]
264			\|\x0E\|\x0F/x;
265			return (charset => 'iso-2022-jp-3')
266			unless $s =~ /\x1B[^\x24\x28]
267			\|\x1B\x24[^\x28] #[^B\x28]
268			\|\x1B\x24\x28[^OP]
269			\|\x1B\x28[^B]
270			\|\x0E\|\x0F/x;
271			return (charset => 'iso-2022-kr')
272			unless $s =~ /\x1B[^\x24]
273			\|\x1B\x24[^\x29]
274			\|\x1B\x24\x29[^C]/x;
275			return (charset => 'iso-2022-jp-2')
276			unless $s =~ /\x1B[^\x24\x28\x2E\x4E]
277			\|\x1B\x24[^\x40AB\x28]
278			\|\x1B\x24\x28[^CD]
279			\|\x1B\x28[^BJ]
280			\|\x1B\x2E[^AF]
281			\|\x0E\|\x0F/x;
282			return (charset => 'iso-2022-cn')
283			unless $s =~ /\x1B[^\x4E\x24]
284			\|\x1B\x24[^\x29\x2A]
285			\|\x1B\x24\x29[^AG]
286			\|\x1B\x24\x2A[^H]/x;
287			return (charset => 'iso-2022-cn-ext')
288			unless $s =~ /\x1B[^\x4E\x4F\x24]
289			\|\x1B\x24[^\x29\x2A]
290			\|\x1B\x24\x29[^AEG]
291			\|\x1B\x24\x2A[^HIJKLM]/x;
292			return (charset => 'iso-2022-int-1')
293			unless $s =~ /\x1B[^\x24\x28\x2D]
294			\|\x1B\x24[^\x40AB\x28\x29]
295			\|\x1B\x24\x28[^DGH]
296			\|\x1B\x24\x29[^C]
297			\|\x1B\x28[^BJ]
298			\|\x1B\x2D[^AF]/x;
299			return (charset => 'junet')
300			unless $s =~ /\x1B[^\x24\x28\x2C]
301			\|\x1B\x24[^\x28\x2C\x40-\x42]
302			\|\x1B\x24[\x28\x2C][^\x20-\x7E]
303			\|\x1B\x24[\x28\x2C][\x20-\x2F]+[^\x30-\x7E]
304			\|\x1B[\x28\x2C][^\x20-\x7E]
305			\|\x1B[\x28\x2C][\x20-\x2F]+[^\x30-\x7E]
306			\|\x0E\|\x0F/x;
307			return (charset => 'x-iso-2022');
308			} else {
309			return (charset => 'us-ascii');
310	wakaba	1.6	}
311			}
312
313	wakaba	1.9	sub _name_net_ascii_8bit ($) {
314			my $name = shift; my $s = shift;
315	wakaba	1.8	return (charset => 'us-ascii') unless $s =~ /[\x1B\x0E\x0F\x80-\xFF]/;
316			if ($s =~ /[\x80-\xFF]/) {
317			if ($s =~ /[\xC0-\xFD][\x80-\xBF]*[\x80-\x8F]/) {
318			if ($s =~ /\x1B/) {
319	wakaba	1.9	return (charset => 'x-junet8'); ## junet + UTF-8
320	wakaba	1.8	} else {
321			return (charset => 'utf-8');
322			}
323			} elsif ($s =~ /\x1B/) {
324	wakaba	1.9	return (charset => 'x-iso-2022'); ## 8bit ISO 2022
325			} else {
326			return (charset => 'iso-8859-1');
327			}
328			} else { ## 7bit ISO 2022
329			return _name_7bit_iso2022 ($name, $s);
330			}
331			}
332
333	wakaba	1.11	sub _name_8bit_iso2022 ($$) {
334	wakaba	1.9	my $name = shift; my $s = shift;
335			return (charset => 'us-ascii') unless $s =~ /[\x1B\x0E\x0F\x80-\xFF]/;
336			if ($s =~ /[\x80-\xFF]/) {
337			if ($s =~ /\x1B/) {
338			return (charset => 'x-iso-2022'); ## 8bit ISO 2022
339	wakaba	1.8	} else {
340			return (charset => 'iso-8859-1');
341			}
342	wakaba	1.9	} else { ## 7bit ISO 2022
343			return _name_7bit_iso2022 ($name, $s);
344			}
345			}
346
347			## Not completed.
348			## TODO: gb18030, cn-gb-12345
349			## TODO: _name_euc_gbf (cn-gb-12345, gb2312)
350			sub _name_euc_gb ($$) {
351			my $name = shift; my $s = shift;
352			if ($s =~ /[\x80-\xFF]/) {
353			if ($s =~ /
354			(?:\G\|[\x00-\x3F\x7F\x80\xFF])
355			(?:[\xA1-\xA9\xB0-\xFE][\xA1-\xFE]
356			\|[\x40-\x7E])*
357			(?:
358			[\x81-\xA0\xAA-\xAF][\x40-\xFE]
359			\|[\xA1-\xFE][\x40-\xA0]
360			)
361			/x) {
362			(charset => 'gbk');
363			} elsif ($s =~ /
364			(?:\xA2[\xA1-\xAA]
365			\|\xA6[\xE0-\xF5]
366			\|\xA8[\xBB-\xC0]
367			)
368			(?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]\|\z))
369			/x) {
370			(charset => 'gbk');
371			} elsif ($s =~ /
372			(?:\xA3\xE7\|\xA7[\xDD-\xF2]
373			\|\xA8[\xBB-\xC0]
374			\|[\xAA-\xAF\xF8-\xFE][\xA1-\xFE]
375			)
376			(?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]\|\z))
377			/x) {
378			(charset => 'cn-gb-isoir165', 'charset-edition' => 1992);
379			} elsif ($s =~ /\xEF\xF1 ## Typo bug of GB 2312
380			(?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]\|\z))
381			/x) {
382			(charset => 'gb2312');
383			} else {
384			(charset => 'gb2312', 'charset-edition' => 1980);
385			}
386			} elsif ($s =~ /[\x0E\x0F]/) {
387			(charset => 'gb2312'); ## Actually, this is not "gb2312"
388			} else {
389	wakaba	1.11	_name_7bit_iso2022 ($name, $s);
390	wakaba	1.9	}
391			}
392
393			sub _name_euc_japan ($$) {
394			my $name = shift; my $s = shift;
395			if ($s =~ /[\x80-\xFF]/) {
396			if ($s =~ /\x8F[\xA1\xA3-\xA5\xA8\xAC-\xAF\xEE-\xFE][\xA1-\xFE]/) {
397			if ($s =~ /\x8F[\xA2\xA6\xA7\xA9-\xAB\xB0-\xED][\xA1-\xFE]/) {
398			## JIS X 0213 plane 2 + JIS X 0212
399			(charset => 'x-euc-jisx0213-packed');
400			} else {
401			(charset => 'euc-jisx0213');
402			}
403			} elsif ($s =~ m{(?<![\x8E\x8F]) ## Not G2/G3 character
404			(?: ## JIS X 0213:2000
405			[\xA9-\xAF\xF5-\xFE][\xA1-\xFE]
406			\|\xA2[\xAF-\xB9\xC2-\xC9\xD1-\xDB\xE9-\xF1\xFA-\xFD]
407			\|\xA3[\xA1-\xAF\xBA-\xC0\xDB-\xE0\xFB-\xFE]
408			\|\xA4[\xF4-\xFE]\|\xA5[\xF7-\xFE]
409			\|\xA6[\xB9-\xC0\xD9-\xFE]\|\xA7[\xC2-\xD0\xF2-\xFE]
410			\|\xA8[\xC1-\xFE]\|\xCF[\xD4-\xFE]\|\xF4[\xA7-\xFE]
411			)
412			(?=(?:[\xA1-\xFE][\xA1-\xFE])*(?:[\x00-\xA0\xFF]\|\z))}x) {
413			if ($s =~ /\x8F/) { ## JIS X 0213 plane 1 + JIS X 0212
414			(charset => 'x-euc-jisx0213-packed');
415			} else {
416			(charset => 'euc-jisx0213-plane1');
417			}
418			} else {
419			(charset => 'euc-jp');
420			}
421			} elsif ($s =~ /\x0E\|\x0F\|\x1B[\x4E\x4F]/) {
422			(charset => 'euc-jisx0213'); ## Actually, this is not euc-japan
423			} else {
424	wakaba	1.11	_name_7bit_iso2022 ($name, $s);
425	wakaba	1.9	}
426			}
427
428			sub _name_shift_jis ($$) {
429			my $name = shift; my $s = shift;
430			if ($s =~ /[\x80-\xFF]/) {
431			if ($s =~ /[\x0E\x0F\x1B]/) {
432			(charset => 'x-sjis');
433			} elsif ($s =~ /
434			(?:\G\|[\x00-\x3F\x7F])
435			(?:[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC]
436			\|[\x40-\x7E\xA1-\xDF])*
437			[\xF0-\xFC][\x40-\x7E\x80-\xFC]
438			/x) {
439			(charset => 'shift_jisx0213');
440			} elsif ($s =~ /
441			(?:\G\|[\x00-\x3F\x7F])
442			(?:[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC]
443			\|[\x40-\x7E\xA1-\xDF])*
444			(?:
445			[\x85-\x87\xEB-\xEF][\x40-\x7E\x80-\xFC]
446			\|\x81[\xAD-\xB7\xC0-\xC7\xCF-\xD9\xE9-\xEF\xF8-\xFB]
447			\|\x82[\x40-\x4E\x59-\x5F\x7A-\x80\x9B-\x9E\xF2-\xFC]
448			\|\x83[\x97-\x9E\xB7-\xBE\xD7-\xFC]
449			\|\x84[\x61-\x6F\x72-\x9E\xBF-\xFC]
450			\|\x88[\x40-\x9E]\|\x98[\x73-\x9E]\|\xEA[\xA5-\xFC]
451			)
452			/x) {
453			(charset => 'shift_jisx0213-plane1');
454			} else {
455			(charset => 'shift_jis');
456			}
457			} elsif ($s =~ /[\x5C\x7E]/) {
458			if ($s =~ /\x1B\x0E\x0F/) {
459			(charset => 'x-sjis'); ## ISO 2022 with implied "ESC ( J"
460			## BUG: "ESC ( B foobar\aaa ESC ( J aiueo" also matchs this
461			} else {
462			(charset => 'jis_x0201');
463			}
464			} else {
465	wakaba	1.11	_name_7bit_iso2022 ($name, $s);
466	wakaba	1.9	}
467			}
468
469			sub _name_utf16be ($$) {
470			shift; my $s = shift;
471			if ($s =~ /[\xD8-\xDB][\x00-\xFF][\xDC-\xDF][\x00-\xFF]
472			(?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
473			(charset => 'utf-16be');
474			} elsif ($s =~ /[\x01-\xFF][\x00-\xFF]
475			(?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
476			if ($s =~ /([^\x00\x03\x04\x23\x25\x30\xFE\xFF]
477			[\x00-\xFF] # ^\x20\x22\x4E-\x9F\xF9\xFA
478			\|\x03[^\x00-\x6F\xD0-\xFF]
479			#\|\x20[^\x00-\x6F]
480			\|\x25[^\x00-\x7F]
481			\|\xFE[^\x30-\x4F]
482			\|\xFF[^\x00-\xEF]
483			## note 1 of RFC 1816 is ambitious, so block entire
484			## is excepted
485			\|\x30[\x00-\x3F]
486			)
487			(?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
488			(charset => 'iso-10646-ucs-2');
489			} else {
490			(charset => 'iso-10646-j-1');
491			}
492			} elsif ($s =~ /\x00[\x80-\xFF]
493			(?=(?:[\x00-\xFF][\x00-\xFF])*\z)/sx) {
494			(charset => 'iso-10646-unicode-latin1');
495			} else {
496			(charset => 'iso-10646-ucs-basic');
497			}
498			}
499
500			sub _name_utf32be ($$) {
501			shift; my $s = shift;
502			if ($s =~ /
503			([\x01-\x7F][\x00-\xFF]{3}
504			\|\x00[\x11-\xFF][\x00-\xFF][\x00-\xFF])
505			(?=(?:[\x00-\xFF]{4})*\z)/sx) {
506			(charset => 'iso-10646-ucs-4');
507			} else {
508			(charset => 'utf-32be');
509	wakaba	1.8	}
510			}
511
512	wakaba	1.13	sub _utf8_on ($) {
513			Encode::_utf8_on ($_[0]) if $Encode::VERSION;
514			}
515			sub _utf8_off ($) {
516			Encode::_utf8_off ($_[0]) if $Encode::VERSION;
517			}
518
519	wakaba	1.14	sub is_mime_text ($) {
520			my $name = lc shift;
521			if (ref $CHARSET{$name}->{mime_text} eq 'CODE') {
522			return &{ $CHARSET{$name}->{mime_text} } ($name);
523			} elsif (defined $CHARSET{$name}->{mime_text}) {
524			return $CHARSET{$name}->{mime_text};
525			} elsif (defined $_IsMimeText{$name}) {
526			return $_IsMimeText{$name};
527			} elsif (ref $CHARSET{'*undef'}->{mime_text} eq 'CODE') {
528			return &{ $CHARSET{'*undef'}->{mime_text} } ($name);
529			}
530			0;
531			}
532
533	wakaba	1.1	=head1 LICENSE
534
535			Copyright 2002 wakaba E<lt>w@suika.fam.cxE<gt>.
536
537			This program is free software; you can redistribute it and/or modify
538			it under the terms of the GNU General Public License as published by
539			the Free Software Foundation; either version 2 of the License, or
540			(at your option) any later version.
541
542			This program is distributed in the hope that it will be useful,
543			but WITHOUT ANY WARRANTY; without even the implied warranty of
544			MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
545			GNU General Public License for more details.
546
547			You should have received a copy of the GNU General Public License
548			along with this program; see the file COPYING. If not, write to
549			the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
550			Boston, MA 02111-1307, USA.
551
552			=head1 CHANGE
553
554			See F<ChangeLog>.
555	wakaba	1.15	$Date: 2002/07/22 07:48:01 $
556	wakaba	1.1
557			=cut
558
559			1;