UPDATE
Doh. I shouldhave read your bugreport more carefully. Obviously both of these are incorrect in the respect that you mention. But i'm pretty sure that I can resolve that. Ill let you know.

Oh and thanks, I've been meaning to get the Dumper test cases and run them through BFDump, but keep forgetting. So now that I've been reminded the download is running... ;-)

use Data::BFDump qw(BFDump);

my @dogs = ( 'Fido', 'Wags' );
my %kennel = (
  First => \$dogs[0],
  Second =>  \$dogs[1],
);
$dogs[2] = \%kennel;
my $mutts = \%kennel;

print "BFDump(\\\@dogs, \\\%kennel, \$mutts);\n";
print BFDump(\@dogs, \%kennel, $mutts),"\n\n";
print "BFDump(\\\%kennel, \\\@dogs, \$mutts);\n";
print BFDump(\%kennel, \@dogs, $mutts),"\n";

__END__
BFDump(\@dogs, \%kennel, $mutts);
do {
    my $RT_ARRAY = [
                     'Fido',
                     'Wags',
                     {}
                   ];
    my $RT_HASH  = {
                     First  => \do { my $v = 'Fido' },
                     Second => \do { my $v = 'Wags' }
                   };
    $RT_ARRAY->[2] = $RT_HASH;
    ( $RT_ARRAY, $RT_HASH, $RT_HASH )
}

BFDump(\%kennel, \@dogs, $mutts);
do {
    my $RT_HASH  = {
                     First  => \do { my $v = 'Fido' },
                     Second => \do { my $v = 'Wags' }
                   };
    my $RT_ARRAY = [
                     'Fido',
                     'Wags',
                     $RT_HASH
                   ];
    ( $RT_HASH, $RT_ARRAY, $RT_HASH )
}
[download]

Yves / DeMerphq
---
Writing a good benchmark isnt as easy as it might look.