#!/usr/bin/perl

use warnings;
use strict;

my %keys;

$keys{'C10'}->{'C10-1'}=1;
$keys{'C10'}->{'C10-2'}=1;
$keys{'C10'}->{'C10-20'}=1;
$keys{'C10'}->{'C10-3'}=1;

$keys{'C200'}->{'C200-1'}=1;
$keys{'C200'}->{'C200-20'}=1;
$keys{'C200'}->{'C200-3'}=1;
$keys{'C200'}->{'C200-21'}=1;

$keys{'C30'}->{'C30-1'}=1;
$keys{'C30'}->{'C30-2'}=1;
$keys{'C30'}->{'C30-30'}=1;
$keys{'C30'}->{'C30-4'}=1; 

sub sorted { 
    return map { $_->[0] }
        sort { 
            $a->[1] <=> $b->[1] ||
            $a->[2] <=> $b->[2]
        } map { [ $_, split("-", substr($_,1)) ] } @_;
}

foreach my $key (sorted(keys %keys)) {
    print "'$key' => ", $keys{$key}, "\n";
    foreach my $subkey (sorted(keys %{$keys{$key}})) {
        print "   '$subkey' => $keys{$key}->{$subkey}\n";
    }
}
[download]

Arjen

This works perfect although I don't understand yet 100% how the Schwartzian transform works. Their is one small problem that I have with the code is that it should look for all leading characters to be split because now it substr only one but it could be more. (CR30-1 for example) So I tried to modify the code but no success.

sub sorted { 
    return map { $_->[0] }
        sort { 
            $a->[1] <=> $b->[1] ||
            $a->[2] <=> $b->[2]
        } map { [ $_, split("-", s/^[A-Z]+//) ] } @_;
}

foreach my $key (sorted(keys %keys)) {
    print "'$key' => ", $keys{$key}, "\n";
    foreach my $subkey (sorted(keys %{$keys{$key}})) {
        print "   '$subkey' => $keys{$key}->{$subkey}\n";
    }
}
[download]

The Schwartzian Transform is explained here by the person after who this sorting technique is named.

A modified version of the sorted routine which also takes the leading letters into account:

sub sorted { 
    return map { $_->[0] }
    sort { 
        $a->[1] cmp $b->[1] ||
        $a->[2] <=> $b->[2] ||
        $a->[3] <=> $b->[3]
    } map { [ $_, (/([A-Z]+)(\d+)(?:-(\d+)){1,2}/) ] } @_;
}
[download]

The split function is replaced by a regex which returns a list of letters and the numbers.

Arjen