How do I extract the number of times a value appears in a hash?

[v]

I'm creating a hash that contains IP_Address:Size objects. The same IP_Address occurs many times, as is Size. i want to know the following:

• how many times each IP_Address appears
• sum of Size for any given IP_Address

Originally posted as a Categorized Question.

[BlaisePascal]

Is the "IP_Address:size" blocks the hash values, or the hash keys?

Assuming the the IP_Address:size blocks are the hash values (say, you were keying off of a connection ID, or something), then you could do something like:

my %counts;
my %sizes;
for (values %inhash) {
  my ($ip, $size) = split /:/;
  next unless defined $ip;
  $counts{$ip}++; 
  $sizes($ip)+= $size;
}

for $ip (keys $counts) {
  my ($count, $size) = ($counts{$ip},$sizes{$ip};
  print "IP Address $ip appeared $count times, for a total size of $si
+ze\n";
}

Does that help?

[ton]

Assuming that "IPaddr:Size" are the values:

map {if /^(.*):(\d*)$/ {$count{$1}++;$size{$1}+=$2;}} values(%hash);

This will place results in the %count and %size hashes, keyed by IP address. If "IPaddr:Size" are the keys, just replace the 'values' with 'keys' in the fragment above.

[dsb]

Pardon the building of the hash

#! /usr/bin/perl

%hash = (
    '1' => 'foo',
    '2' => 'bar',
    '3' => 'foo',
    '4' => 'foo',
);

%count;
foreach $key ( keys %hash ) {
    $val   = $hash{$key};
    $count{$val}++;
}

foreach $key1 ( keys %count ) {
    print $key1 . " appears " . $count{$key1} . " times\n";
}