$file-> defectdata looks like a function call and is likely to return something different each time it's called. Maybe what you want to do is:

foreach (@files) {
    $file = "$klarfdir" . "\\" . "$_";
    $file = Klarf::Parser->new($file);

    my %result = %{ $file->defectdata () };
    foreach ( keys %result ) {
        print( $result{$_}->{XINDEX} );
    }
}
[download]

That worked.
Thank you.
I don't understand why it returns somthing different though.
~~David~~

It's a function call. Looks like it's for a parser. I'd guess each time it's called it returns a result from parsing the next "chunk". You need to read to docs for the module.

---

DWIM is Perl's answer to Gödel

Even if the contents of the hash are the same each time you call the sub, it's still a different hash. To take the most trivial example:

sub get_hashref {
  my %hash;
  return \%hash;
}
[download]

it creates a new my %hash each time it's called, since the old one went out of scope and no longer exists¹. If you want to prove this to yourself, create a trivial sub like that and call it twice, storing both hashrefs. Add some data to one hashref, then look in the other - it will still be empty.

(There are ways to have it return references to the same hash every time it's called, such as

my %hash;
sub get_hashref2 {
  return \%hash;
}
[download]

but defectdata() apparently doesn't do this.)

¹ Not entirely true, in that the previous hash still exists and can be accessed by the previously-returned hashref, but it has become anonymous and is no longer my %hash. Read up on "closures" for more about how this works (and how it can be used to do some pretty powerful stuff).