Perl: lecture d'un repertoire et de fichiers

#!/usr/bin/perl -w
use strict;
use warnings;
use diagnostics;
use CGI qw/:standard/; 
use CGI::Carp 'fatalsToBrowser';

our $dir = '/chemin/Fichiers/';
our $meanOfLenghts1;
our $meanOfLenghts2;
our $mimetype;
our $numberOfFilesListed;
our $sum2;
our $sum1;
our $file;
our $i;
our @elements;
our @elements1;
our @elements2;
our $fh,
our @length;
#our $length;
our @listeFichiersMem;
our $debut;
our $fin;
our $line;
our $result;
our $value;

my $cgi = CGI -> new();

# read all entries in the directory:
opendir DIR, $dir or die "Cannot open $dir $!";
@listeFichiersMem = grep /\.txt$/, readdir DIR;
closedir DIR;
#printf $cgi -> header("Content-type: $mimetype; charset=utf-8; Content-Disposition: attachment;");

foreach $file (@listeFichiersMem) 
{
		printf "Hallo test";
    $i = 1;
		$numberOfFilesListed++;
		open $fh, "<", $file or die $!;
	  print "$file" . "\n";

		while ($line = $fh) #read each line from FILE.
    {
          chomp ($line);
          while ((($debut, $fin) = $result =~ /debut\s+([0-9]+).+?fin\s+([0-9]+)/g) and $i < 3)
          {
                        $length[$i] = $fin - $debut; # Calculation of the lenght of the first segment, then the lenght of the second segment
                        #push(@elements[$i], $length[$i]); #Push the 2 computed lenghts into a table to compute the mean of lenght for the 2 segments
                        $elements[$i] -> push($length[$i]);
                        $i++;
          }
    }
		close $fh;
}

foreach $value (@elements1)
{
        $sum1 += $value;
				printf $sum1;
}

foreach $value (@elements2)
{
        $sum2 += $value;
}

$meanOfLenghts1 = $sum1/$numberOfFilesListed;
$meanOfLenghts2 = $sum2/$numberOfFilesListed;

printf ("%d %d", $meanOfLenghts1, $meanOfLenghts2);