Here's a "traditional" sort of counter, with a '.' for every percentage point and numbers at every 10% mark:

#!/usr/bin/perl -w
use strict;
$|++;

open Fh, "file" or die "file: $!\n";
my $s = -s Fh;

my ($last, $total) = 0;
while (<Fh>){
    $total += length;
    my $p = int($total * 100 / $s);
    if ($p != $last){
        $last = $p;
        print STDERR $p % 10 ? "." : "$p%";
    }
}
[download]

-- 
Human history becomes more and more a race between education and catastrophe. -- HG Wells

Thx, you helped me a lot. It works now.

Use -s to determine the size of the file. Sum line lengths into a progress counter as the lines are processed. For each progress update divide the progress counter by the file size, multiply to 100, and there's your percentage.

• count lines before the actual processing, i.e. process the file twice. wc -l might help.
• stat the file to get its size and then count bytes processed.

UPDATE: As pointed out by GrandFather in a very similar topic today (Estimate line count in text file), you can use tell to detect your position inside the file. That together with detecting the size of the file gives quite a nice solution.

#in the loop
# update $percentage, e.g. 100 * $summed_size / $file_size; 
...
(printf STDERR "Progress: %3d%%\r", $percentage),$lastpercentage=int($
+percentage) 
     if int($percentage) != $lastpercentage;
...
[download]

See also Smart::Comments (which is a Categorized Damian Module).