package Encode::Alias;
use strict;
+no warnings 'redefine';
use Encode;
-our $VERSION = do { my @r = (q$Revision: 1.0 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
-our $DEBUG = 0;
-require Exporter;
+our $VERSION = do { my @r = (q$Revision: 2.0 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
+sub DEBUG () { 0 }
-our @ISA = qw(Exporter);
+use base qw(Exporter);
# Public, encouraged API is exported by default
sub find_alias
{
my $class = shift;
- local $_ = shift;
- unless (exists $Alias{$_})
+ my $find = shift;
+ unless (exists $Alias{$find})
{
+ $Alias{$find} = undef; # Recursion guard
for (my $i=0; $i < @Alias; $i += 2)
{
my $alias = $Alias[$i];
my $val = $Alias[$i+1];
my $new;
- if (ref($alias) eq 'Regexp' && $_ =~ $alias)
+ if (ref($alias) eq 'Regexp' && $find =~ $alias)
{
- $DEBUG and warn "eval $val";
+ DEBUG and warn "eval $val";
$new = eval $val;
- # $@ and warn "$val, $@";
+ DEBUG and $@ and warn "$val, $@";
}
elsif (ref($alias) eq 'CODE')
{
- $DEBUG and warn "$alias", "->", "($val)";
- $new = $alias->($val);
+ DEBUG and warn "$alias", "->", "($find)";
+ $new = $alias->($find);
}
- elsif (lc($_) eq lc($alias))
+ elsif (lc($find) eq lc($alias))
{
$new = $val;
}
if (defined($new))
{
- next if $new eq $_; # avoid (direct) recursion on bugs
- $DEBUG and warn "$alias, $new";
+ next if $new eq $find; # avoid (direct) recursion on bugs
+ DEBUG and warn "$alias, $new";
my $enc = (ref($new)) ? $new : Encode::find_encoding($new);
if ($enc)
{
- $Alias{$_} = $enc;
+ $Alias{$find} = $enc;
last;
}
}
}
}
- if ($DEBUG){
+ if (DEBUG){
my $name;
- if (my $e = $Alias{$_}){
+ if (my $e = $Alias{$find}){
$name = $e->name;
}else{
$name = "";
}
- warn "find_alias($class, $_)->name = $name";
+ warn "find_alias($class, $find)->name = $name";
}
- return $Alias{$_};
+ return $Alias{$find};
}
sub define_alias
for my $k (@a){
if (ref($alias) eq 'Regexp' && $k =~ $alias)
{
- $DEBUG and warn "delete \$Alias\{$k\}";
+ DEBUG and warn "delete \$Alias\{$k\}";
delete $Alias{$k};
}
elsif (ref($alias) eq 'CODE')
{
- $DEBUG and warn "delete \$Alias\{$k\}";
+ DEBUG and warn "delete \$Alias\{$k\}";
delete $Alias{$alias->($name)};
}
}
}else{
- $DEBUG and warn "delete \$Alias\{$alias\}";
+ DEBUG and warn "delete \$Alias\{$alias\}";
delete $Alias{$alias};
}
}
sub init_aliases
{
undef_aliases();
- # 'C' => 'US-ascii' so you can feed default locale directly.
- define_alias('C' => 'US-ascii');
+
+ # Try all-lower-case version should all else fails
+ define_alias( qr/^(.*)$/ => '"\L$1"' );
+
+ # UTF/UCS stuff
+ define_alias( qr/^UTF-?7$/i => '"UTF-7"');
+ define_alias( qr/^UCS-?2-?LE$/i => '"UCS-2LE"' );
+ define_alias( qr/^UCS-?2-?(BE)?$/i => '"UCS-2BE"',
+ qr/^UCS-?4-?(BE|LE)?$/i => 'uc("UTF-32$1")',
+ qr/^iso-10646-1$/i => '"UCS-2BE"' );
+ define_alias( qr/^UTF(16|32)-?BE$/i => '"UTF-$1BE"',
+ qr/^UTF(16|32)-?LE$/i => '"UTF-$1LE"',
+ qr/^UTF(16|32)$/i => '"UTF-$1"',
+ );
+ # ASCII
+ define_alias(qr/^(?:US-?)ascii$/i => '"ascii"');
+ define_alias('C' => 'ascii');
+ define_alias(qr/\bISO[-_]?646[-_]?US$/i => '"ascii"');
# Allow variants of iso-8859-1 etc.
define_alias( qr/\biso[-_]?(\d+)[-_](\d+)$/i => '"iso-$1-$2"' );
define_alias( qr/^(.+)\@euro$/i => '"$1"' );
define_alias( qr/\b(?:iso[-_]?)?latin[-_]?(\d+)$/i
- => '"iso-8859-$Encode::Alias::Latin2iso[$1]"' );
+ => 'defined $Encode::Alias::Latin2iso[$1] ? "iso-8859-$Encode::Alias::Latin2iso[$1]" : undef' );
define_alias( qr/\bwin(latin[12]|cyrillic|baltic|greek|turkish|
hebrew|arabic|baltic|vietnamese)$/ix =>
# At least AIX has IBM-NNN (surprisingly...) instead of cpNNN.
# And Microsoft has their own naming (again, surprisingly).
# And windows-* is registered in IANA!
- define_alias( qr/\b(?:ibm|ms|windows)[-_]?(\d\d\d\d?)$/i => '"cp$1"');
+ define_alias( qr/\b(?:cp|ibm|ms|windows)[-_ ]?(\d{2,4})$/i => '"cp$1"');
# Sometimes seen with a leading zero.
- define_alias( qr/\bcp037\b/i => '"cp37"');
-
- # Ououououou.
- define_alias( qr/\bmacRomanian$/i => '"macRumanian"');
-
-# Standardize on the dashed versions.
- # define_alias( qr/\butf8$/i => 'utf-8' );
- define_alias( qr/\bkoi8r$/i => 'koi8-r' );
- define_alias( qr/\bkoi8u$/i => 'koi8-u' );
+ # define_alias( qr/\bcp037\b/i => '"cp37"');
+
+ # Mac Mappings
+ # predefined in *.ucm; unneeded
+ # define_alias( qr/\bmacIcelandic$/i => '"macIceland"');
+ define_alias( qr/^mac_(.*)$/i => '"mac$1"');
+ # Ououououou. gone. They are differente!
+ # define_alias( qr/\bmacRomanian$/i => '"macRumanian"');
+
+ # Standardize on the dashed versions.
+ # define_alias( qr/\butf8$/i => '"utf-8"' );
+ define_alias( qr/\bkoi8[\s-_]*([ru])$/i => '"koi8-$1"' );
unless ($Encode::ON_EBCDIC){
# for Encode::CN
# CP936 doesn't have vendor-addon for GBK, so they're identical.
define_alias( qr/^gbk$/i => '"cp936"');
# This fixes gb2312 vs. euc-cn confusion, practically
- define_alias( qr/\bGB[-_ ]?2312(?:\D.*$|$)/i => '"euc-cn"' );
+ define_alias( qr/\bGB[-_ ]?2312(?!-?raw)/i => '"euc-cn"' );
# for Encode::JP
define_alias( qr/\bjis$/i => '"7bit-jis"' );
define_alias( qr/\beuc.*jp$/i => '"euc-jp"' );
define_alias( qr/(?:x-)?windows-949$/i => '"cp949"' );
define_alias( qr/\bks_c_5601-1987$/i => '"cp949"' );
# for Encode::TW
- define_alias( qr/\bbig-?5$/i => '"big5"' );
- define_alias( qr/\bbig5-hk(?:scs)?$/i => '"big5-hkscs"' );
+ define_alias( qr/\bbig-?5$/i => '"big5-eten"' );
+ define_alias( qr/\bbig5-?et(?:en)?$/i => '"big5-eten"' );
+ define_alias( qr/\btca[-_]?big5$/i => '"big5-eten"' );
+ define_alias( qr/\bbig5-?hk(?:scs)?$/i => '"big5-hkscs"' );
+ define_alias( qr/\bhk(?:scs)?[-_]?big5$/i => '"big5-hkscs"' );
}
-
+ # utf8 is blessed :)
+ define_alias( qr/^UTF-8$/i => '"utf8"',);
# At last, Map white space and _ to '-'
define_alias( qr/^(\S+)[\s_]+(.*)$/i => '"$1-$2"' );
}
=head1 NAME
-Encode::Alias - alias defintions to encodings
+Encode::Alias - alias definitions to encodings
=head1 SYNOPSIS
=head1 DESCRIPTION
-Allows newName to be used as am alias for ENCODING. ENCODING may be
-either the name of an encoding or and encoding object (as described in L<Encode>).
+Allows newName to be used as an alias for ENCODING. ENCODING may be
+either the name of an encoding or an encoding object (as described
+in L<Encode>).
Currently I<newName> can be specified in the following ways:
define_alias( qr/^iso8859-(\d+)$/i => '"iso-8859-$1"' );
-In this case if I<ENCODING> is not a reference it is C<eval>-ed to
-allow C<$1> etc. to be subsituted. The example is one way to names as
-used in X11 font names to alias the MIME names for the iso-8859-*
-family. Note the double quote inside the single quote.
+In this case, if I<ENCODING> is not a reference, it is C<eval>-ed
+in order to allow C<$1> etc. to be substituted. The example is one
+way to alias names as used in X11 fonts to the MIME names for the
+iso-8859-* family. Note the double quotes inside the single quotes.
-If you are using regex here, you have to do so or it won't work in
-this case. Also not regex is tricky even for the experienced. Use it
-with caution.
+(or, you don't have to do this yourself because this example is predefined)
+
+If you are using a regex here, you have to use the quotes as shown or
+it won't work. Also note that regex handling is tricky even for the
+experienced. Use this feature with caution.
=item As a code reference, e.g.:
- define_alias( sub { return /^iso8859-(\d+)$/i ? "iso-8859-$1" : undef } , '');
+ define_alias( sub {shift =~ /^iso8859-(\d+)$/i ? "iso-8859-$1" : undef } );
-In this case C<$_> will be set to the name that is being looked up and
-I<ENCODING> is passed to the sub as its first argument. The example
-is another way to names as used in X11 font names to alias the MIME
-names for the iso-8859-* family.
+The same effect as the example above in a different way. The coderef
+takes the alias name as an argument and returns a canonical name on
+success or undef if not. Note the second argument is not required.
+Use this with even more caution than the regex version.
=back
-=head2 Alias overloading
+=head3 Changes in code reference aliasing
+
+As of Encode 1.87, the older form
+
+ define_alias( sub { return /^iso8859-(\d+)$/i ? "iso-8859-$1" : undef } );
-You can override predefined aliases by simply applying define_alias().
-New alias is always evaluated first and when neccessary define_alias()
-flushes internal cache to make new definition available.
+no longer works.
- # redirect SHIFT_JIS to MS/IBM Code Page 932, which is a
+Encode up to 1.86 internally used "local $_" to implement ths older
+form. But consider the code below;
+
+ use Encode;
+ $_ = "eeeee" ;
+ while (/(e)/g) {
+ my $utf = decode('aliased-encoding-name', $1);
+ print "position:",pos,"\n";
+ }
+
+Prior to Encode 1.86 this fails because of "local $_".
+
+=head2 Alias overloading
+
+You can override predefined aliases by simply applying define_alias().
+The new alias is always evaluated first, and when neccessary,
+define_alias() flushes the internal cache to make the new definition
+available.
+
+ # redirect SHIFT_JIS to MS/IBM Code Page 932, which is a
# superset of SHIFT_JIS
define_alias( qr/shift.*jis$/i => '"cp932"' );
define_alias( qr/sjis$/i => '"cp932"' );
-If you want to zap all predefined aliases, you can
+If you want to zap all predefined aliases, you can use
Encode::Alias->undef_aliases;
Encode::Alias->init_aliases;
-gets factory setting back.
-
+gets the factory settings back.
=head1 SEE ALSO