example custom reducer script for Hive

collect.pl

#!/usr/bin/perl

undef $currentId;
@vals=();

while (<STDIN>) {
    chomp();
    processRow(split(/\t/));
}

output();

sub output() {
    print $currentId . "\t" . join(",", @vals) . "\n";
}

sub processRow() {
    my ($id, @ll) = @_;

    if (! defined($currentId)) {
        $currentId = $id;
        push(@vals, @ll);
        return;
    }

    if ($currentId ne $id) {
        output();
        $currentId = $id;
        @vals=(@ll);
        return;
    }

    push(@vals, @ll);
}