Not really an obfu?

This code isn't a JAPH nor is it a properly obscure obfu I think. But it's definitely in "play" mode, so if you are in "play" mode please read on -- and see if you can help me improve it or learn something new.

The problem:

trim quotes from string. Do not molest any quotes that might be embedded inside the string -- only balanced quotes at ^ and $ may be removed.

What I came up with cannot be "production code" because the only way to do this in production code that I think would be countenanced, might be

my $quoted_str = q{"Eat Potatoes Daily"};
   $quoted_str =~ s< ^" ([^"]+) "$ >  < $1 >x
print "Unquoted: $quoted_str\n";
[download]

But in play mode I wanted something more opaque ;-) and also thought to myself, "I don't want to pay the (run-time) penalty for that back-reference ($1) in the replacement expression." So I came up with this ugly little beast:

my $quoted_str = q{"Eat Potatoes Daily"};
  # Remove surrounding quotes, an obfu way:
    for (my $n = 1, my $ip = '^', my $tp = ''; 3 > $n; $tp = '$')
    {
    $ip = '' if $n += $quoted_str =~ s<$ip"$tp><>;
    }
print "Unquoted: $quoted_str\n";
[download]

At one point I had in mind that the flip-flop operator " .. " might help me, used in a scalar context, but I couldn't work that out. Anybody see how to use it to toggle between matching at the beginning of the string and at the end? I couldn't.

Any other alternative solutions are definitely appreciated. And if editors feel that I've posted this in the wrong area of the Monastery, please edit Title / move at will.

    Soren/Intrepid

-- 
use PerlMonk::Tye qw(:wisely);