| Subject: | [PATCH] decode_utf8 and non-PVs |
$ perl -lMblib -e ' use Encode; print decode_utf8("*main::foo")'
*main::foo
$ perl -lMblib -e ' use Encode; print decode_utf8(*foo)'
The second one-liner prints nothing.
Under a different perl version I get gibberish:
$ ./perl -lIlib -e ' use Encode; print decode_utf8(*foo)'
Wide character in print at -e line 1.
*clone_encoding8�@X���� FB_CONSTS@��X�(��X�8a�X����X�B��AX��c�X�p��X�B��AX����X�HY@X����X�HY@X�P��X�HY@X�|
The input should be forcibly stringified before using SvCUR and SvEND.
See the attached patch.
(This also makes it handle copy-on-write correctly, which was how I
encountered this. I am currently working on a debug mode for perl
that changes COW buffer violations into crashes.)
| Subject: | patch.txt |
diff -rup Encode-2.55-uZLdOK-orig/Encode.pm Encode-2.55-uZLdOK/Encode.pm
--- Encode-2.55-uZLdOK-orig/Encode.pm 2013-09-14 00:52:18.000000000 -0700
+++ Encode-2.55-uZLdOK/Encode.pm 2013-12-21 17:53:01.000000000 -0800
@@ -209,7 +209,7 @@ my $utf8enc;
sub decode_utf8($;$) {
my ( $octets, $check ) = @_;
return undef unless defined $octets;
- $octets .= '' if ref $octets;
+ $octets .= '';
$check ||= 0;
$utf8enc ||= find_encoding('utf8');
diff -rup Encode-2.55-uZLdOK-orig/t/Encode.t Encode-2.55-uZLdOK/t/Encode.t
--- Encode-2.55-uZLdOK-orig/t/Encode.t 2013-12-21 17:51:53.000000000 -0800
+++ Encode-2.55-uZLdOK/t/Encode.t 2013-12-21 17:52:34.000000000 -0800
@@ -25,7 +25,7 @@ my @character_set = ('0'..'9', 'A'..'Z',
my @source = qw(ascii iso8859-1 cp1250);
my @destiny = qw(cp1047 cp37 posix-bc);
my @ebcdic_sets = qw(cp1047 cp37 posix-bc);
-plan test => 38+$n*@encodings + 2*@source*@destiny*@character_set + 2*@ebcdic_sets*256 + 6 + 4;
+plan test => 38+$n*@encodings + 2*@source*@destiny*@character_set + 2*@ebcdic_sets*256 + 6 + 5;
my $str = join('',map(chr($_),0x20..0x7E));
my $cpy = $str;
ok(length($str),from_to($cpy,'iso8859-1','Unicode'),"Length Wrong");
@@ -150,6 +150,9 @@ package main;
ok(decode(latin1 => Encode::Dummy->new("foobar")), "foobar");
ok(encode(utf8 => Encode::Dummy->new("foobar")), "foobar");
+# decode_utf8 with non-string arguments
+ok(decode_utf8(*1), "*main::1");
+
# hash keys
my $key = (keys %{{ "whatever\x{100}" => '' }})[0];
my $kopy = $key;