In Perl 6...

That reminds me of the danger of passing a var to a function that modifies that var. It's particularly easy to screw up when dealing with built-in globals variables such as $1.

sub func {
   'b' =~ /(.)/;
   print(@_, "\n");
}

'a' =~ /(.)/;  func($1);    # Prints "b"
'a' =~ /(.)/;  func("$1");  # Prints "a"
[download]

sub func {
   eval { die "bar\n" };
   print(@_);
}

eval { die "foo\n" };  func($@);    # Prints "bar"
eval { die "foo\n" };  func("$@");  # Prints "foo"
[download]

(If you want to mess with your mind, try removing the second 'a' =~ /(.)/; from the first snippet.)

        if (my ($a, $b) = ("$key$val" =~ /(..)(...)/)) {
            $hash2{$a} = $b;
        }
[download]

#!/usr/bin/perl
use Test::More "no_plan";
use Devel::Size "total_size";

my $key = "aa";
my $val = "a00";
my %hash1;
my %hash2;
while (length($key) == 2) {
        $hash1{$key} = $val;
        if (my ($a, $b) = ("$key$val" =~ /(..)(...)/)) {
            $hash2{$a} = $b;
        }

        ++$key;
        ++$val;
}

is(keys(%hash1), keys(%hash2), "same number of keys");
is_deeply(\%hash1, \%hash2, "is_deeply same");
is(total_size(\%hash1), total_size(\%hash2));
__END__
ok 1 - same number of keys
ok 2 - is_deeply same
ok 3
[download]

gam3 demonstrates a great method. ++. Perl Best Practices advocates avoiding using $1, $2, etc. whenever you possibly can, and that is a great way to do it. IMO it's sort of a "poor man's" named capture (one of the few regex features that Perl "lacks"), and you never have to refer to the numbered variables themselves except in some special cases like in the replacement clause of the substitution operator.

I am also particularly fond of a one-step syntax to get the results of a substitution on a variable without affecting the original variable. It's hard to explain but I'm sure you've either done it or wanted to do it at some point. Here's an example (which I apologize is less than ideal; I wasn't sure if any spaces would turn up in the match from the split):

use strict;
use warnings;
my @staff = `whoare -g somegroup`;
#chomping an array chomps all the elements
chomp @staff;

foreach my $emp (@staff) {
    my @fields = split /\s{2,}/, $emp;
    die "error! number of fields is " . scalar @fields . "! "
      unless ( scalar @fields == 3 );

    #dispense with the opening paren to get the group 
    ( my $def_grp = $fields[1] ) =~ s/\s*\(\s*//;
#etc.
}
[download]

use strict;
use warnings;

# First things first.  If I can't figure out who you are, you're outta
+ here.
my $editor;
if ( exists( $ENV{REMOTE_USER} ) ) {
    ($editor) = $ENV{REMOTE_USER} =~ m/(some regex)/i;
}
(defined $editor) || die "I can't determine who you are, so you can't 
+access this area.";

#use editor later throughout the program
[download]

While we're on the topic, I should note that it's a very good idea to use non-capturing parentheses(?:) whenever you want or need to use parentheses but don't need to remember the stuff that matched in there.

(?<taunt>My 5\.9\.5 perl has named captures. ;-\))

⠤⠤ ⠙⠊⠕⠞⠁⠇⠑⠧⠊

No kidding!? I figured it would be a feature in Perl6, but I didn't know it was already out. We're on 5.8.5 at our site and when I asked for an upgrade the guy who's in charge of the sitewide perl installs said he'd prefer to wait for 5.9 to become stable. He says there is not much of a payoff to move to a higher 5.8.x release.

This was last month...5.9 (with named captcha) isn't stable by any chance, is it?

---

I like computer programming because it's like Legos for the mind.