#! perl
####################################################################
# Key        0        1       2       3               4
# Italian => Spanish, French, German, English|German, English|German
####################################################################

use strict;
use warnings;
use autodie;
use constant NAMES => qw(uno due tre quattro cinque sei sette otto nouve dieci undici dodici tredici);

my %hash;

open my $in, '<', './Test_Data_RandNumbers.txt';

while (<$in>)
{
    my ($ita, $spa, $ger) = split /[=\s,]+/; 

    if ($ita)
    {
        $hash{$ita}[0] = $spa;
        $hash{$ita}[2] = $ger if $ger;
    }
}

close   $in;
open my $in1, '<', './Test_Data_More_RandNumbers.txt';

while (<$in1>)
{
    # $num1 & $num2 may each be either English or German
    my ($ita, $fren, $num1, $num2) = split /[=\s,]+/;

    $hash{$ita}[1] = $fren;

    if ($num1)
    {
        $hash{$ita}[3] = $num1;
        $hash{$ita}[4] = $num2 if $num2;
    }
}

close   $in1;
open my $out , '>', './OUT_Test_Data_Ita_SpanFren_rest.txt';
open my $out1, '>', './OUT_Test_data_NO_match_SpanFren.txt';

for my $ita (sort { sort_italian() } keys %hash)
{
    if ($ita)
    {
        my    $fh = defined $hash{$ita}[0] &&
                    defined $hash{$ita}[1] ? $out : $out1;

        print $fh "$ita => ",
                   join(',', map { $_ // () } @{ $hash{$ita} }),
                  "\n";
    }
}

close $out;
close $out1;

{
    my %numbers;

    BEGIN
    {
        my $i    = 1;
        %numbers = map { $_ => $i++ } NAMES;
    }

    sub sort_italian
    {
        # Add error checking here!

        return $numbers{$a} <=> $numbers{$b};
    }
}

##</code><code>##

if ($ita)

##</code><code>##

if (defined $ita && $ita ne '')

##</code><code>##

join(',', map { $_ // () } @{ $hash{$ita} }),

##</code><code>##

Use of uninitialized value in join or string at ... line ...