RT 4.0.22

1 files changed, 15 insertions, 17 deletions

diff --git a/rt/lib/RT/Attachment.pm b/rt/lib/RT/Attachment.pm
index 07fdea3b2..af1f82c15 100755
--- a/rt/lib/RT/Attachment.pm
+++ b/rt/lib/RT/Attachment.pm
@@ -128,19 +128,17 @@ sub Create {
     $Attachment->make_singlepart;
 
     # Get the subject
-    my $Subject = $Attachment->head->get( 'subject', 0 );
+    my $Subject = Encode::decode( 'UTF-8', $Attachment->head->get( 'subject' ) );
     $Subject = '' unless defined $Subject;
     chomp $Subject;
-    utf8::decode( $Subject ) unless utf8::is_utf8( $Subject );
 
     #Get the Message-ID
-    my $MessageId = $Attachment->head->get( 'Message-ID', 0 );
+    my $MessageId = Encode::decode( "UTF-8", $Attachment->head->get( 'Message-ID' ) );
     defined($MessageId) or $MessageId = '';
     chomp ($MessageId);
     $MessageId =~ s/^<(.*?)>$/$1/o;
 
     #Get the filename
-
     my $Filename = mime_recommended_filename($Attachment);
 
     # remove path part. 
@@ -148,8 +146,7 @@ sub Create {
 
     # MIME::Head doesn't support perl strings well and can return
     # octets which later will be double encoded in low-level code
-    my $head = $Attachment->head->as_string;
-    utf8::decode( $head ) unless utf8::is_utf8( $head );
+    my $head = Encode::decode( 'UTF-8', $Attachment->head->as_string );
 
     # If a message has no bodyhandle, that means that it has subparts (or appears to)
     # and we should act accordingly.  
@@ -289,7 +286,7 @@ before returning it.
 sub Content {
     my $self = shift;
     return $self->_DecodeLOB(
-        $self->ContentType,
+        $self->GetHeader('Content-Type'),  # Includes charset, unlike ->ContentType
         $self->ContentEncoding,
         $self->_Value('Content', decode_utf8 => 0),
     );
@@ -320,7 +317,6 @@ sub OriginalContent {
     }
 
     return $self->Content unless RT::I18N::IsTextualContentType($self->ContentType);
-    my $enc = $self->OriginalEncoding;
 
     my $content;
     if ( !$self->ContentEncoding || $self->ContentEncoding eq 'none' ) {
@@ -333,18 +329,20 @@ sub OriginalContent {
         return( $self->loc("Unknown ContentEncoding [_1]", $self->ContentEncoding));
     }
 
-    # Turn *off* the SvUTF8 bits here so decode_utf8 and from_to below can work.
-    local $@;
-    Encode::_utf8_off($content);
+    my $entity = MIME::Entity->new();
+    $entity->head->add("Content-Type", $self->GetHeader("Content-Type"));
+    $entity->bodyhandle( MIME::Body::Scalar->new( $content ) );
+    my $from = RT::I18N::_FindOrGuessCharset($entity);
+    $from = 'utf-8' if not $from or not Encode::find_encoding($from);
 
-    if (!$enc || $enc eq '' ||  $enc eq 'utf8' || $enc eq 'utf-8') {
-        # If we somehow fail to do the decode, at least push out the raw bits
-        eval { return( Encode::decode_utf8($content)) } || return ($content);
-    }
+    my $to = RT::I18N::_CanonicalizeCharset(
+        $self->OriginalEncoding || 'utf-8'
+    );
 
-    eval { Encode::from_to($content, 'utf8' => $enc) } if $enc;
+    local $@;
+    eval { Encode::from_to($content, $from => $to) };
     if ($@) {
-        $RT::Logger->error("Could not convert attachment from assumed utf8 to '$enc' :".$@);
+        $RT::Logger->error("Could not convert attachment from $from to $to: ".$@);
     }
     return $content;
 }