Re: counting pairwise incidences

[jpl]

Or, as a minor variant of the code posted by moritz

use strict;
use warnings;

my %pairs;
while (<DATA>) {
    my @names = split;
    # my @names = sort split;    # if order shouldn't matter
    while ( @names > 1 ) {
        my $first = shift(@names);
        map { ++$pairs{ join '_', $first, $_ } } @names;
    }
}
while ( my ( $k, $v ) = each %pairs ) {
    print "$k: $v\n";
}

__DATA__
tomD gly4 phil
aesG tomD gly4 phil
aesG phil
aesG tomD gly4

[reubs85]

Thanks everyone for their help! The order doesn't matter so the variant of the code supplied by moritz and jpl works a treat. Highly impressed by the speed of answering btw; I just checked back on the site I never expected replies to be so quick... long live the brotherhood!

Cheers!

Reuben