=head1 A Hash of lists

A common requirement is to read in a data file where each
line is a record of multiple fields and create a structure
in memory to hold that data, split down so that it can be
processed. Here's an example - a hash of lists

=cut

# Data is build into this teaching example rather than
# bring a separate file (keeps it all together for training)

$info = <<ABOUT;
Cambridge Flat City
Norwich Boring City
Norwich Alabamish City
Norwich Fuuny Virtual
Lakenheath Military Village
Thetford Forest Town
Norwich New York Burb
London Big City
ABOUT

# Read through each line of the data
# Take off the first field as the key
# Make the key unique (just in case two were identical)
# Add the list into the hash based on the (modified) key

foreach $line(split(/\n/,$info)) {
        my ($place,@fields) = split(/\t/,$line);
        $place .= "_" while ($records{$place});
        $records{$place} = \@fields;
        }

# Demonstration to show how we can pass the whole
# structure into a sub, just by passing a single address

sub shoutabout {
        $inparam = $_[0];
        for $k (keys(%$inparam)) {
                $$inparam{$k}[0] .= " on Friday";
        }
}
shoutabout(\%records);

print ${$records{"Norwich"}}[0],"\n";
print $records{"Norwich"}->[0],"\n";
print $records{"Norwich"}[0],"\n";

for $pk (grep(/^Norwich_*$/,keys %records)) {
        $pkp = $pk;
        $pkp =~ s/_+//;
        print "$pkp ...>> ",$records{$pk}[0],"\n";
        }

__END__

Sample output

wizard:csr2 graham$ perl rain
Boring on Friday
Boring on Friday
Boring on Friday
Norwich ...>> Fuuny on Friday
Norwich ...>> Alabamish on Friday
Norwich ...>> Boring on Friday
Norwich ...>> New York on Friday
wizard:csr2 graham$