0001-use-perl5.8-s-unicode-way.patch

From 699d735d94897a2b281b6924535bff573c47e5d6 Mon Sep 17 00:00:00 2001
From: Tokuhiro Matsuno <[email protected]>
Date: Tue, 17 Feb 2009 12:36:22 +0900
Subject: [PATCH] use perl5.8's unicode way

---
 lib/FormValidator/Simple/Plugin/Japanese.pm |   78 +++++++++------------------
 1 files changed, 25 insertions(+), 53 deletions(-)

diff --git a/lib/FormValidator/Simple/Plugin/Japanese.pm b/lib/FormValidator/Simple/Plugin/Japanese.pm
index ea6da51..1105f32 100644
--- a/lib/FormValidator/Simple/Plugin/Japanese.pm
+++ b/lib/FormValidator/Simple/Plugin/Japanese.pm
@@ -4,76 +4,48 @@ use base qw/
     FormValidator::Simple::Plugin::Number::Phone::JP
 /;
 use FormValidator::Simple::Exception;
+use FormValidator::Simple::Validator;
 use FormValidator::Simple::Constants;
-use Unicode::RecursiveDowngrade;
 use Mail::Address::MobileJp;
-use Jcode;
+use Encode ();
 
 our $VERSION = '0.05';
 
-# plugin specific method
-sub __japanese_check_charset {
-    my ($self, $charset) = @_;
-    my @charsets = (
-        [qw/UTF-8     utf8     utf8/],
-        [qw/EUC-JP    euc      euc /],
-        [qw/Shift_JIS shiftjis sjis/],
-    );
-    foreach my $set ( @charsets ) {
-        foreach my $name ( @$set ) {
-            if( $charset eq $name ) {
-                return $set->[2];
-            }
-        }
-    }
-    FormValidator::Simple::Exception->throw(
-        qq/wrong charset "$charset"./
-    );
-}
-
-# plugin specific method
-sub __japanese_encode2euc {
-    my ($self, $value) = @_;
-    my $charset = $self->options->{charset} || 'utf8';
-    $charset    = $self->__japanese_check_charset($charset);
-
-    my $rd = Unicode::RecursiveDowngrade->new;
-    $rd->filter( sub { Jcode->new($value, $charset)->euc } );
-    return $rd->downgrade($value);
-}
-
-# plugin specific method
-sub __japanese_delete_sp {
-    my ($self, $value) = @_;
-    $value          = $self->__japanese_encode2euc($value);
-    $value          =~ s/ //g;
-    my $ascii       = '[\x00-\x7F]';
-    my $two_bytes   = '[\x8E\xA1-\xFE][\xA1-\xFE]';
-    my $three_bytes = '\x8F[\xA1-\xFE][\xA1-\xFE]';
-    $value =~ s/\G((?:$ascii|$two_bytes|$three_bytes)*?)(?:\xA1\xA1)/$1/g;
-    return $value;
-}
+our ($DECODE, $DELETE_SP);
+BEGIN {
+    $DECODE = sub {
+        my ($self, $str) = @_;
+        return $str if Encode::is_utf8($str);
+
+        my $charset = $self->options->{charset} || 'utf8';
+        return Encode::decode($charset, $str);
+    };
+    $DELETE_SP = sub {
+        local $_ = shift;
+        s/\s//g;
+        $_;
+    };
+};
 
 sub HIRAGANA {
     my ($self, $params, $args) = @_;
-    my $value = $self->__japanese_delete_sp( $params->[0] );
-    return $value =~ /^(?:\xA4[\x00-\xFF]|\xA1\xBC)+$/ ? TRUE : FALSE;
+    my $v = $DELETE_SP->($DECODE->($self, $params->[0]));
+    return $v =~ /^\p{InHiragana}+$/ ? TRUE : FALSE;
 }
 
 sub KATAKANA {
     my ($self, $params, $args) = @_;
-    my $value = $self->__japanese_delete_sp( $params->[0] );
-    return $value =~ /^(?:\xA5[\x00-\xFF]|\xA1\xBC)+$/ ? TRUE : FALSE;
+    my $v = $DELETE_SP->($DECODE->($self, $params->[0]));
+    return $v =~ /^\p{InKatakana}+$/ ? TRUE : FALSE;
 }
 
+# this method is only for backward compatibility.
+# you can count correct multi-byte string length in perl5.8's unicode way.
+# please use LENGTH instead of this.
 sub JLENGTH {
     my ($self, $params, $args) = @_;
-    my $min = $args->[0] || 0;
-    my $max = $args->[1] || 0;
-    my $value = $self->__japanese_encode2euc( $params->[0] );
-    my $length = Jcode->new($value, 'euc')->jlength;
-    $min += 0; $max += 0; $max ||= $min;
-    return ($min <= $length and $length <= $max) ? TRUE : FALSE;
+    my $v = $DELETE_SP->($DECODE->($self, $params->[0]));
+    FormValidator::Simple::Validator::LENGTH( $self, [$v], $args );
 }
 
 sub ZIP_JP {
-- 
1.6.1.2