It seems like you're doing a lot of work for just looking for non-printable characters.  (Actually, they're ALL ascii characters, even the non-printable ones).  Also, I'm not sure why you're comparing with 121; the printable characters go all the way to 127 126 (= "~").  (Thanks ambrus, I knew that, but somehow my thinking was "off by one").

Instead of sorting, how about just applying the test to each character in the line?  Here's a subroutine which will return nonzero if the line contains non-printable characters, and zero otherwise:

sub line_contains_non_printable {
    my ($line) = @_;
    my @split = split //, $line;
    foreach (@split) {
        my $ord = ord($_);
        return 1 if ($ord < 32 || $ord > 126);
    }
    return 0;
}
[download]

I'm sure there's at least several good modules for doing this as well (that I don't know of offhand), which other monks may be able to recommend.

~ is character 126; character 127 is a control character.

Wow, that seems a really brute force way to detect a non-ASCII character. If all you really need to do is detect it, I'd apply a regex to each line ... maybe like in this (untested) example:

  while (<$file_handle>) {
    if (/[^[:ascii:]]/) { 
      # do something
    }
  }
[download]

Hi rsriram,

Here is one way to do it.

$str = 'a[bdy]dfjaPÑsdafÜ';

while ($str =~ /[^!-~\s]/g)
{
    print "Non ASCII Characater:\t$&\n";
}

Another way:

while ($str =~ /[^[:print:]]/g) 
{
    print "Non Printable Characater:\t$&\n";

}
[download]

If you would have super searched, you would have got detecting non-ascii chars in a string, Getting rid of non-standard ASCII characters and more similar nodes.

There is a simpler way - count them with tr///,

$_ = 'a[bdy]dfjaPÃ?sdafÃ?';

print tr/\x0a\x20-\x7d//c, "unwanted characters in the string\n";
__END__
2 unwanted characters in the string
[download]