Not sure why you want it as arrays -- this is ideal for a hash:

my @array = qw( a a b c d d );
my %counts;
$counts{$_}++ for @array;

# if you really want arrays back at the end:
my @keys = keys %counts;
my @vals = values %counts;
[download]

Indeed, however I wonder if order counts for the OP, i.e.: should

qw/a a b c d d a/
[download]

map to

user wants?
(a => 2, b => 1, c => 1, d => 2, a => 1):
# or
(a => 3, b => 1, c => 1, d => 2);
[download]

I guess the latter, but still preserving the input order, in which case the common technique for uniq'ing a list can be easily adapted to do it:

my %saw;
my @sum=map {$_ => $saw{$_}} grep !$saw{$_}++, @array;
[download]

Now that I think of it, this can be even adapted for the former task, provided that grep's selection is improved to delete the key is it's not the same as the previous one. I'd probably chose a completely different strategy, though.

my @array = qw( a a b c d d );

my %counts;
my @keys = grep { !($counts{$_}++) } @array;

print("$_: $counts{$_}\n")
   foreach @keys;
[download]

use strict;
use warnings;
my %counts;
foreach my $key ( @myArray ){
  $counts{$key}++;
}
[download]

sort lines.txt | uniq -c
[download]

Now that's good. Thank you!