From: Jarkko Hietaniemi <jhi@iki.fi>
Date: Mon, 8 Apr 2002 18:23:41 +0000 (+0000)
Subject: Upgrade to Encode 1.31, from Dan Kogai.
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=fdd579e2dc5d0da16b7e86246a01f0d838120df7;p=p5sagit%2Fp5-mst-13.2.git

Upgrade to Encode 1.31, from Dan Kogai.

p4raw-id: //depot/perl@15812
---

diff --git a/MANIFEST b/MANIFEST
index a907be9..ebb60de 100644
--- a/MANIFEST
+++ b/MANIFEST
@@ -232,6 +232,7 @@ ext/Encode/encoding.pm	Perl Pragmactic Module
 ext/Encode/lib/Encode/Alias.pm	        Encode extension
 ext/Encode/lib/Encode/CJKConstants.pm	Encode extension
 ext/Encode/lib/Encode/CN/HZ.pm		Encode extension
+ext/Encode/lib/Encode/Config.pm		Encode configuration module
 ext/Encode/lib/Encode/Encoder.pm	       OO Encoder
 ext/Encode/lib/Encode/Encoding.pm	Encode extension
 ext/Encode/lib/Encode/JP/2022_JP.pm	Encode extension
@@ -245,6 +246,7 @@ ext/Encode/lib/Encode/XS.pm		Encode extension
 ext/Encode/t/Aliases.t	Encode extension test
 ext/Encode/t/CN.t		Encode extension test
 ext/Encode/t/Encode.t		Encode extension test
+ext/Encode/t/Encoder.t		Encode::Encoder test
 ext/Encode/t/JP.t		Encode extension test
 ext/Encode/t/KR.t		Encode extension test
 ext/Encode/t/TW.t		Encode extension test
diff --git a/ext/Encode/Changes b/ext/Encode/Changes
index 145d8af..aaacc8b 100644
--- a/ext/Encode/Changes
+++ b/ext/Encode/Changes
@@ -1,15 +1,32 @@
 # Revision history for Perl extension Encode.
 #
-# $Id: Changes,v 1.30 2002/04/08 02:34:51 dankogai Exp $
+# $Id: Changes,v 1.31 2002/04/08 18:08:07 dankogai Exp dankogai $
 #
 
-1.30 $Date: 2002/04/08 02:34:51 $
+1.31 $Date: 2002/04/08 18:08:07 $
+! lib/Encode/Encoder.pm
++ t/Encoder.t
+  Encode::Encoder, once just a placeholder of an idea, is now much more 
+  practical.  See t/Encode.t to find how practical it can be.
++ lib/Encode/Config.pm
+! Encode.pm
+  my false laziness at Encode.pm is fixed.  Now %ExtModules are set
+  in Encode::Config and they are all literally, not programatically
+  set.  My false laziness was resulting many encodings missing from
+  %ExtModules.
+! lib/Encode/Unicode.pm
+! t/Unicode.t
+  BOM for 32LE was bogus as noted by Anton.  t/Unicode.t is fixed
+  so that it does not rely Encode::Unicode for BOM values
+  Message-Id: <FFEC33E9-4AFB-11D6-B415-00039301D480@dan.co.jp>
+
+1.30 2002/04/08 02:34:51
 + lib/Encode/Encoder.pm
   Object Oriented Encoder.  I reckon something like this is in need.
 ! Encode.pm
 ! t/Unicode.pm
 ! lib/Encode/Supported.pod
-  * autoloading but that prevented upper-case canonicals such as UTF-16
+  * autoloading bug that prevented upper-case canonicals such as UTF-16
     is fixed.  Now even UTF/UCS are autoloaded!
   * encodings() is now more intuitive.
   * t/Unicode.t fixed to explicitly use Unicode.pm -- BOM values are
@@ -172,7 +189,7 @@
   Typo fixes and improvements by jhi
   Message-Id: <200204010201.FAA03564@alpha.hut.fi>, et al.
 
-1.11  $Date: 2002/04/08 02:34:51 $
+1.11  $Date: 2002/04/08 18:08:07 $
 + t/encoding.t
 + t/jperl.t
 ! MANIFEST
diff --git a/ext/Encode/Encode.pm b/ext/Encode/Encode.pm
index caada11..a828d46 100644
--- a/ext/Encode/Encode.pm
+++ b/ext/Encode/Encode.pm
@@ -1,6 +1,6 @@
 package Encode;
 use strict;
-our $VERSION = do { my @r = (q$Revision: 1.30 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
+our $VERSION = do { my @r = (q$Revision: 1.31 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
 our $DEBUG = 0;
 
 require DynaLoader;
@@ -43,92 +43,7 @@ use Encode::Alias;
 
 # Make a %Encoding package variable to allow a certain amount of cheating
 our %Encoding;
-our %ExtModule;
-
-my @codepages = qw(
-		     37  424  437  500  737  775  850  852  855 
-		    856  857  860  861  862  863  864  865  866 
-		    869  874  875  932  936  949  950 1006 1026 
-		   1047 1250 1251 1252 1253 1254 1255 1256 1257
-		   1258
-		   );
-
-my @macintosh = qw(
-		   CentralEurRoman  Croatian  Cyrillic   Greek
-		   Iceland          Roman     Rumanian   Sami
-		   Thai             Turkish   Ukrainian
-		   );
-
-for my $k (2..11,13..16){
-    $ExtModule{"iso-8859-$k"} = 'Encode::Byte';
-}
-
-for my $k (@codepages){
-    $ExtModule{"cp$k"} = 'Encode::Byte';
-}
-
-for my $k (@macintosh)
-{
-    $ExtModule{"mac$k"} = 'Encode::Byte';
-}
-
-for my $k (qw(UCS-2BE UCS-2LE UTF-16 UTF-16BE UTF-16LE
-	      UTF-32 UTF-32BE UTF-32LE)){
-    $ExtModule{$k} = 'Encode::Unicode';
-}
-
-%ExtModule =
-    (%ExtModule,
-     'koi8-r'           => 'Encode::Byte',
-     'posix-bc'         => 'Encode::EBCDIC',
-     cp37               => 'Encode::EBCDIC',
-     cp1026             => 'Encode::EBCDIC',
-     cp1047             => 'Encode::EBCDIC',
-     cp500              => 'Encode::EBCDIC',
-     cp875              => 'Encode::EBCDIC',
-     dingbats           => 'Encode::Symbol',
-     macDingbats        => 'Encode::Symbol',
-     macSymbol          => 'Encode::Symbol',
-     symbol             => 'Encode::Symbol',
-     viscii             => 'Encode::Byte',
-);
-
-unless ($ON_EBCDIC) { # CJK added to autoload unless EBCDIC env
-%ExtModule =
-    (%ExtModule,
-
-     'cp936'		=> 'Encode::CN',
-     'euc-cn'           => 'Encode::CN',
-     'gb12345-raw'	=> 'Encode::CN',
-     'gb2312-raw'	=> 'Encode::CN',
-     'gbk'		=> 'Encode::CN',
-     'iso-ir-165'	=> 'Encode::CN',
-
-     '7bit-jis'         => 'Encode::JP',
-     'cp932'		=> 'Encode::JP',
-     'euc-jp'	        => 'Encode::JP',
-     'iso-2022-jp'	=> 'Encode::JP',
-     'iso-2022-jp-1'	=> 'Encode::JP',
-     'jis0201-raw'      => 'Encode::JP',
-     'jis0208-raw'      => 'Encode::JP',
-     'jis0212-raw'      => 'Encode::JP',
-     'macJapanese'      => 'Encode::JP',
-     'shiftjis'	        => 'Encode::JP',
-
-     'cp949'		=> 'Encode::KR',
-     'euc-kr'       	=> 'Encode::KR',
-     'ksc5601'		=> 'Encode::KR',
-     'macKorean'        => 'Encode::KR',
-
-     'big5'		=> 'Encode::TW',
-     'big5-hkscs'	=> 'Encode::TW',
-     'cp950'		=> 'Encode::TW',
-
-     'big5plus'     	=> 'Encode::HanExtra',
-     'euc-tw'   	=> 'Encode::HanExtra',
-     'gb18030'		=> 'Encode::HanExtra',
-    );
-}
+use Encode::Config;
 
 sub encodings
 {
diff --git a/ext/Encode/MANIFEST b/ext/Encode/MANIFEST
index 849baea..c3deba8 100644
--- a/ext/Encode/MANIFEST
+++ b/ext/Encode/MANIFEST
@@ -34,6 +34,7 @@ encoding.pm	Perl Pragmactic Module
 lib/Encode/Alias.pm	        Encode extension
 lib/Encode/CJKConstants.pm	Encode extension
 lib/Encode/CN/HZ.pm		Encode extension
+lib/Encode/Config.pm	        Encode configuration module
 lib/Encode/Encoder.pm	       OO Encoder
 lib/Encode/Encoding.pm	Encode extension
 lib/Encode/JP/2022_JP.pm	Encode extension
@@ -47,6 +48,7 @@ lib/Encode/XS.pm		Encode extension
 t/Aliases.t	Encode extension test
 t/CN.t		Encode extension test
 t/Encode.t		Encode extension test
+t/Encoder.t	Encode::Encoder test
 t/JP.t		Encode extension test
 t/KR.t		Encode extension test
 t/TW.t		Encode extension test
diff --git a/ext/Encode/lib/Encode/Config.pm b/ext/Encode/lib/Encode/Config.pm
new file mode 100644
index 0000000..48095f6
--- /dev/null
+++ b/ext/Encode/lib/Encode/Config.pm
@@ -0,0 +1,147 @@
+#
+# Demand-load module list
+#
+package Encode::Config;
+our $VERSION = do { my @r = (q$Revision: 1.1 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
+
+use strict;
+require Exporter;
+
+our %ExtModule = 
+    (
+      # Encode::Byte
+      #iso-8859-1 is on Encode.pm itself
+     'iso-8859-2'             => 'Encode::Byte',
+     'iso-8859-3'             => 'Encode::Byte',
+     'iso-8859-4'             => 'Encode::Byte',
+     'iso-8859-5'             => 'Encode::Byte',
+     'iso-8859-6'             => 'Encode::Byte',
+     'iso-8859-7'             => 'Encode::Byte',
+     'iso-8859-8'             => 'Encode::Byte',
+     'iso-8859-9'             => 'Encode::Byte',
+     'iso-8859-10'            => 'Encode::Byte',
+     'iso-8859-11'            => 'Encode::Byte',
+     'iso-8859-13'            => 'Encode::Byte',
+     'iso-8859-14'            => 'Encode::Byte',
+     'iso-8859-15'            => 'Encode::Byte',
+     'iso-8859-16'            => 'Encode::Byte',
+     'koi8-f'                 => 'Encode::Byte',
+     'koi8-r'                 => 'Encode::Byte',
+     'koi8-u'                 => 'Encode::Byte',
+     'viscii'                 => 'Encode::Byte',
+     'cp424'                  => 'Encode::Byte',
+     'cp437'                  => 'Encode::Byte',
+     'cp737'                  => 'Encode::Byte',
+     'cp775'                  => 'Encode::Byte',
+     'cp850'                  => 'Encode::Byte',
+     'cp852'                  => 'Encode::Byte',
+     'cp855'                  => 'Encode::Byte',
+     'cp856'                  => 'Encode::Byte',
+     'cp857'                  => 'Encode::Byte',
+     'cp860'                  => 'Encode::Byte',
+     'cp861'                  => 'Encode::Byte',
+     'cp862'                  => 'Encode::Byte',
+     'cp863'                  => 'Encode::Byte',
+     'cp864'                  => 'Encode::Byte',
+     'cp865'                  => 'Encode::Byte',
+     'cp866'                  => 'Encode::Byte',
+     'cp869'                  => 'Encode::Byte',
+     'cp874'                  => 'Encode::Byte',
+     'cp1006'                 => 'Encode::Byte',
+     'cp1250'                 => 'Encode::Byte',
+     'cp1251'                 => 'Encode::Byte',
+     'cp1252'                 => 'Encode::Byte',
+     'cp1253'                 => 'Encode::Byte',
+     'cp1254'                 => 'Encode::Byte',
+     'cp1255'                 => 'Encode::Byte',
+     'cp1256'                 => 'Encode::Byte',
+     'cp1257'                 => 'Encode::Byte',
+     'cp1258'                 => 'Encode::Byte',
+     'AdobeStandardEncoding'  => 'Encode::Byte',
+     'MacArabic'              => 'Encode::Byte',
+     'MacCentralEurRoman'     => 'Encode::Byte',
+     'MacCroatian'            => 'Encode::Byte',
+     'MacCyrillic'            => 'Encode::Byte',
+     'MacFarsi'               => 'Encode::Byte',
+     'MacGreek'               => 'Encode::Byte',
+     'MacHebrew'              => 'Encode::Byte',
+     'MacIcelandic'           => 'Encode::Byte',
+     'MacRoman'               => 'Encode::Byte',
+     'MacRomanian'            => 'Encode::Byte',
+     'MacRumanian'            => 'Encode::Byte',
+     'MacSami'                => 'Encode::Byte',
+     'MacThai'                => 'Encode::Byte',
+     'MacTurkish'             => 'Encode::Byte',
+     'MacUkrainian'           => 'Encode::Byte',
+     'nextstep'               => 'Encode::Byte',
+     'hp-roman8'              => 'Encode::Byte',
+     'gsm0338'                => 'Encode::Byte',
+     # Encode::EBCDIC
+     'cp37'                   => 'Encode::EBCDIC',
+     'cp500'                  => 'Encode::EBCDIC',
+     'cp875'                  => 'Encode::EBCDIC',
+     'cp1026'                 => 'Encode::EBCDIC',
+     'cp1047'                 => 'Encode::EBCDIC',
+     'posix-bc'               => 'Encode::EBCDIC',
+     # Encode::Symbol
+     'dingbats'               => 'Encode::Symbol',
+     'symbol'                 => 'Encode::Symbol',
+     'AdobeSymbol'            => 'Encode::Symbol',
+     'AdobeZdingbat'          => 'Encode::Symbol',
+     'MacDingbats'            => 'Encode::Symbol',
+     'MacSymbol'              => 'Encode::Symbol',
+     # Encode::Unicode
+     'UCS-2BE'                => 'Encode::Unicode',
+     'UCS-2LE'                => 'Encode::Unicode',
+     'UTF-16'                 => 'Encode::Unicode',
+     'UTF-16BE'               => 'Encode::Unicode',
+     'UTF-16LE'               => 'Encode::Unicode',
+     'UTF-32'                 => 'Encode::Unicode',
+     'UTF-32BE'               => 'Encode::Unicode',
+     'UTF-32LE'               => 'Encode::Unicode',
+    );
+
+unless (ord("A") == 193){
+    %ExtModule =
+	(
+	 %ExtModule,
+	 'euc-cn'             => 'Encode::CN',
+	 'gb12345-raw'        => 'Encode::CN',
+	 'gb2312-raw'         => 'Encode::CN',
+	 'iso-ir-165'         => 'Encode::CN',
+	 'cp936'              => 'Encode::CN',
+	 'MacChineseSimp'     => 'Encode::CN',
+
+	 '7bit-jis'           => 'Encode::JP',
+	 'euc-jp'             => 'Encode::JP',
+	 'iso-2022-jp'        => 'Encode::JP',
+	 'iso-2022-jp-1'      => 'Encode::JP',
+	 'jis0201-raw'        => 'Encode::JP',
+	 'jis0208-raw'        => 'Encode::JP',
+	 'jis0212-raw'        => 'Encode::JP',
+	 'cp932'              => 'Encode::JP',
+	 'MacJapanese'        => 'Encode::JP',
+	 'shiftjis'           => 'Encode::JP',
+
+
+	 'euc-kr'             => 'Encode::KR',
+	 'iso-2022-kr'        => 'Encode::KR',
+	 'johab'              => 'Encode::KR',
+	 'ksc5601-raw'        => 'Encode::KR',
+	 'cp949'              => 'Encode::KR',
+	 'MacKorean'          => 'Encode::KR',
+
+	 'big5'               => 'Encode::TW',
+	 'big5-hkscs'         => 'Encode::TW',
+	 'cp950'              => 'Encode::TW',
+	 'MacChineseTrad'     => 'Encode::TW',
+
+	 'big5plus'           => 'Encode::HanExtra',
+	 'euc-tw'             => 'Encode::HanExtra',
+	 'gb18030'            => 'Encode::HanExtra',
+	);
+}
+
+*Encode::ExtModule = \%ExtModule;
+
+1;
diff --git a/ext/Encode/lib/Encode/Encoder.pm b/ext/Encode/lib/Encode/Encoder.pm
index 1fc65ea..68cc65c 100644
--- a/ext/Encode/lib/Encode/Encoder.pm
+++ b/ext/Encode/lib/Encode/Encoder.pm
@@ -1,16 +1,14 @@
 #
-# $Id: Encoder.pm,v 0.1 2002/04/08 02:35:10 dankogai Exp $
+# $Id: Encoder.pm,v 0.2 2002/04/08 18:08:07 dankogai Exp dankogai $
 #
-package Encoder;
+package Encode::Encoder;
 use strict;
-our $VERSION = do { my @r = (q$Revision: 0.1 $ =~ /\d+/g); sprintf "%d."."%02d"  x $#r, @r };
+use warnings;
+our $VERSION = do { my @r = (q$Revision: 0.2 $ =~ /\d+/g); sprintf "%d."."%02d"  x $#r, @r };
 
 require Exporter;
 our @ISA = qw(Exporter);
-# Public, encouraged API is exported by default
-our @EXPORT = qw (
-  encoder
-);
+our @EXPORT = qw ( encoder );
 
 our $AUTOLOAD;
 our $DEBUG = 0;
@@ -19,22 +17,30 @@ use Carp;
 
 sub new{
     my ($class, $data, $encname) = @_;
-    $encname ||= 'utf8';
-    my $obj = find_encoding($encname) 
-	or croak __PACKAGE__, ": unknown encoding: $encname";
+    unless($encname){
+	$encname = Encode::is_utf8($data) ? 'utf8' : '';
+    }else{
+	my $obj = find_encoding($encname) 
+	    or croak __PACKAGE__, ": unknown encoding: $encname";
+	$encname = $obj->name;
+    }
     my $self = {
 		data     => $data,
-		encoding => $obj->name,
+		encoding => $encname,
 	       };
     bless $self => $class;
 }
 
-sub encoder{ shift->new(@_) }
+sub encoder{ __PACKAGE__->new(@_) }
 
 sub data{
     my ($self, $data) = shift;
-    defined $data and $self->{data} = $data;
-    return $self;
+    if (defined $data){
+	$self->{data} = $data;
+	return $data;
+    }else{
+	return $self->{data};
+    }
 }
 
 sub encoding{
@@ -43,8 +49,24 @@ sub encoding{
 	my $obj = find_encoding($encname) 
 	    or confess __PACKAGE__, ": unknown encoding: $encname";
 	$self->{encoding} = $obj->name;
+	return $self;
+    }else{
+	return $self->{encoding}
     }
-    $self;
+}
+
+sub bytes {
+    my ($self, $encname) = @_;
+    $encname ||= $self->{encoding};
+    my $obj = find_encoding($encname) 
+	    or confess __PACKAGE__, ": unknown encoding: $encname";
+    $self->{data} = $obj->decode($self->{data}, 1);
+    $self->{encoding} = '' ;
+    return $self;
+}
+
+sub DESTROY{ # defined so it won't autoload.
+    $DEBUG and warn shift;
 }
 
 sub AUTOLOAD {
@@ -56,7 +78,11 @@ sub AUTOLOAD {
     my $obj = find_encoding($myname) 
 	    or confess __PACKAGE__, ": unknown encoding: $myname";
     $DEBUG and warn $self->{encoding}, " => ", $obj->name;
-    from_to($self->{data}, $self->{encoding}, $obj->name, 1);
+    if ($self->{encoding}){
+	from_to($self->{data}, $self->{encoding}, $obj->name, 1);
+    }else{
+	$self->{data} = $obj->encode($self->{data}, 1);
+    }
     $self->{encoding} = $obj->name;
     return $self;
 }
@@ -82,11 +108,13 @@ Encode::Encoder -- Object Oriented Encoder
   # shortcut
   encoder($data)->iso_8859_1;
   # you can stack them!
-  encoder($data)->iso_8859_1->base64; # provided base64() is defined
+  encoder($data)->iso_8859_1->base64;  # provided base64() is defined
+  # you can use it as a decoder as well
+  encoder($base64)->bytes('base64')->latin1;
   # stringified
-  print encoder($utf8)->latin1     # prints the string in latin1
+  print encoder($data)->utf8->latin1;  # prints the string in latin1
   # numified
-  encoder("\x{abcd}\x{ef}g") == 6; # true. bytes::length($data)
+  encoder("\x{abcd}\x{ef}g")->utf8 == 6; # true. bytes::length($data)
 
 =head1 ABSTRACT
 
@@ -100,6 +128,7 @@ converted to Latin1 then Base64, you can simply say
 instead of
 
   my $latin1 = encode("latin1", $utf8);
+  my $base64 = encode_base64($utf8);
 
 or lazier and convolted
 
@@ -122,7 +151,56 @@ When there is no method, it takes the method name as the name of
 encoding and encode instance I<data> with I<encoding>.  If successful,
 instance I<encoding> is set accordingly.
 
-=item * 
+=item *
+
+You can retrieve the result via -E<gt>data but usually you don't have to 
+because the stringify operator ("") is overridden to do exactly that.
+
+=back
+
+=head2 Predefined Methods
+
+This module predefines the methods below;
+
+=over 4
+
+=item $e = Encode::Encoder-E<gt>new([$data, $encoding]);
+
+returns the encoder object.  Its data is initialized with $data if
+there, and its encoding is set to $encoding if there.
+
+When $encoding is omitted, it defaults to utf8 if $data is already in
+utf8 or "" (empty string) otherwise.
+
+=item encoder()
+
+is an alias of Encode::Encoder-E<gt>new().  This one is exported for
+convenience.
+
+=item $e-E<gt>data([$data])
+
+when $data is present, sets instance data to $data and returns the
+object itself. otherwise the current instance data is returned.
+
+=item $e-E<gt>encoding([$encoding])
+
+when $encoding is present, sets instance encoding to $encoding and
+returns the object itself. otherwise the current instance encoding is
+returned.
+
+=item $e-E<gt>bytes([$encoding])
+
+decodes instance data from $encoding, or instance encoding if omitted.
+when the conversion is successful, the enstance encoding will be set
+to "" . 
+
+The name I<bytes> was deliberately picked to avoid namespace tainting
+-- this module may be used as a base class so method names that appear
+in Encode::Encoding are avoided.
+
+=back
+
+=head2 Example: base64 transcoder
 
 This module is desined to work with L<Encode::Encoding>.
 To make the Base64 transcorder example above really work, you should
@@ -130,15 +208,15 @@ write a module like this.
 
   package Encode::Base64;
   use base 'Encode::Encoding';
-  __PACKAGE->Define('base64');
+  __PACKAGE__->Define('base64');
   use MIME::Base64;
   sub encode{ 
-    my ($obj, $data) = @_; 
-    return encode_base64($data);
+      my ($obj, $data) = @_; 
+      return encode_base64($data);
   }
   sub decode{
-    my ($obj, $data) = @_; 
-    return decode_base64($data);
+      my ($obj, $data) = @_; 
+      return decode_base64($data);
   }
   1;
   __END__
@@ -147,7 +225,11 @@ And your caller module should be like this;
 
   use Encode::Encoder;
   use Encode::Base64;
-  # and be creative.
+
+  # now you can really do the following
+
+  encoder($data)->iso_8859_1->base64;
+  encoder($base64)->bytes('base64')->latin1;
 
 =head2 operator overloading
 
@@ -159,34 +241,6 @@ Numify returns the number of bytes therein.
 
 They come in handy when you want to print or find the size of data.
 
-=back
-
-=head2 Predefined Methods
-
-This module predefines the methods below;
-
-=over 4
-
-=item $e = Encode::Encoder-E<gt>new([$data, $encoding]);
-
-returns the encoder object.  Its data is initialized with $data if
-there, and its encoding is set to $encoding if there.
-
-=item encoder()
-
-is an alias of Encode::Encoder-E<gt>new().  This one is exported for
-convenience.
-
-=item $e-E<gt>data($data)
-
-sets instance data to $data.
-
-=item $e-E<gt>encoding($encoding)
-
-sets instance encoding to $encoding
-
-=back
-
 =head1 SEE ALSO
 
 L<Encode>
diff --git a/ext/Encode/lib/Encode/Unicode.pm b/ext/Encode/lib/Encode/Unicode.pm
index c90e8b6..03fab5c 100644
--- a/ext/Encode/lib/Encode/Unicode.pm
+++ b/ext/Encode/lib/Encode/Unicode.pm
@@ -3,7 +3,7 @@ package Encode::Unicode;
 use strict;
 use warnings;
 
-our $VERSION = do { my @r = (q$Revision: 1.26 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
+our $VERSION = do { my @r = (q$Revision: 1.28 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
 
 #
 # Aux. subs & constants
@@ -12,7 +12,7 @@ our $VERSION = do { my @r = (q$Revision: 1.26 $ =~ /\d+/g); sprintf "%d."."%02d"
 sub FBCHAR(){ 0xFFFd }
 sub BOM_BE(){ 0xFeFF }
 sub BOM16LE(){ 0xFFFe }
-sub BOM32LE(){ 0xFeFF0000 }
+sub BOM32LE(){ 0xFFFe0000 }
 
 sub valid_ucs2($){
     if ($_[0] < 0xD800){
@@ -344,8 +344,8 @@ called Byte Order Mark (BOM) is prepended to the head of string.
 
             16         32 bits/char
 -------------------------
-BE	0xFeFF 0x0000FeFF
-LE      0xFFeF 0xFeFF0000
+BE      0xFeFF 0x0000FeFF
+LE      0xFFeF 0xFFFe0000
 -------------------------
 
 =back
@@ -377,6 +377,7 @@ line, not file, is prepended with BOMs.
 C<UCS-2> is an exception.  Unlike others this is an alias of UCS-2BE.
 UCS-2 is already registered by IANA and others that way.
 
+=back
 
 =head1 The Surrogate Pair
 
@@ -418,6 +419,10 @@ does not prohibit them for uses.
 
 =head1 SEE ALSO
 
-L<Encode>, L<http://www.unicode.org/glossary/>
+L<Encode>, L<http://www.unicode.org/glossary/>,
 
-=back
+RFC 2781 L<http://rfc.net/rfc2781.html>,
+
+L<http://www.unicode.org/unicode/faq/utf_bom.html>
+
+=cut
diff --git a/ext/Encode/t/Encoder.t b/ext/Encode/t/Encoder.t
new file mode 100644
index 0000000..e970ecc
--- /dev/null
+++ b/ext/Encode/t/Encoder.t
@@ -0,0 +1,54 @@
+#
+# $Id: Encoder.t,v 1.1 2002/04/08 18:07:31 dankogai Exp $
+#
+
+BEGIN {
+    require Config; import Config;
+    if ($Config{'extensions'} !~ /\bEncode\b/) {
+      print "1..0 # Skip: Encode was not built\n";
+      exit 0;
+    }
+# should work without perlio
+#     unless (find PerlIO::Layer 'perlio') {
+# 	print "1..0 # Skip: PerlIO was not built\n";
+# 	exit 0;
+#     }
+# should work on EBCDIC
+#    if (ord("A") == 193) {
+# 	print "1..0 # Skip: EBCDIC\n";
+# 	exit 0;
+#    }
+    $| = 1;
+}
+
+use strict;
+#use Test::More 'no_plan';
+use Test::More tests => 512;
+use Encode::Encoder;
+use MIME::Base64;
+package Encode::Base64;
+use base 'Encode::Encoding';
+__PACKAGE__->Define('base64');
+use MIME::Base64;
+sub encode{
+    my ($obj, $data) = @_;
+    return encode_base64($data);
+}
+sub decode{
+    my ($obj, $data) = @_;
+    return decode_base64($data);
+}
+
+package main;
+
+my $data = '';
+for my $i (0..255){
+    no warnings;
+    $data .= chr($i);
+    my $base64 = encode_base64($data);
+    is(encoder($data)->base64, $base64, "encode");
+    is(encoder($base64)->bytes('base64'), $data, "decode");
+}
+
+1;
+__END__
diff --git a/ext/Encode/t/Unicode.t b/ext/Encode/t/Unicode.t
index ea52f15..df4b8df 100644
--- a/ext/Encode/t/Unicode.t
+++ b/ext/Encode/t/Unicode.t
@@ -1,5 +1,5 @@
 #
-# $Id: Unicode.t,v 1.4 2002/04/08 02:35:48 dankogai Exp dankogai $
+# $Id: Unicode.t,v 1.5 2002/04/08 14:17:19 dankogai Exp $
 #
 # This script is written entirely in ASCII, even though quoted literals
 # do include non-BMP unicode characters -- Are you happy, jhi?
@@ -28,7 +28,6 @@ use strict;
 #use Test::More 'no_plan';
 use Test::More tests => 22;
 use Encode qw(encode decode);
-use Encode::Unicode; # to load BOM defs
 
 #
 # see
@@ -54,10 +53,10 @@ my $n_32be =
 my $n_32le = 
     pack("C*", map {hex($_)} qw<4D 00 00 00 61 00 00 00 cd ab 01 00>);
 
-my $n_16bb = pack('n', Encode::Unicode::BOM_BE)  . $n_16be;
-my $n_16lb = pack('n', Encode::Unicode::BOM16LE) . $n_16le;
-my $n_32bb = pack('N', Encode::Unicode::BOM_BE ) . $n_32be;
-my $n_32lb = pack('N', Encode::Unicode::BOM32LE) . $n_32le;
+my $n_16bb = pack('n', 0xFeFF) . $n_16be;
+my $n_16lb = pack('v', 0xFeFF) . $n_16le;
+my $n_32bb = pack('N', 0xFeFF) . $n_32be;
+my $n_32lb = pack('V', 0xFeFF) . $n_32le;
 
 is($n_16be, encode('UTF-16BE', $nasty),  qq{encode UTF-16BE});
 is($n_16le, encode('UTF-16LE', $nasty),  qq{encode UTF-16LE});
@@ -91,7 +90,3 @@ ok($@=~/^UCS-2LE:/, "encode UCS-2LE: exception");
 
 1;
 __END__
-
-use Devel::Peek;
-my $foo = decode('UTF-16BE', $n_16be);
-Dump $n_16be; Dump $foo;