So I needed a hash of ISO-639 2-letter language codes and their English names. A quick google search turned up this list, and instead of search for a plain text list I decided to use it directly:

use 5.012;
use Mojo::UserAgent;
use warnings;
binmode STDOUT, ':encoding(UTF-8)';

my $ua = Mojo::UserAgent->new();
my $r = $ua->get('http://www.loc.gov/standards/iso639-2/php/code_list.
+php');
for my $row ($r->res->dom('tr')->each) {
    my ($three, $two, $english_name, $french_name) =
            map $_->text, $row->find('td')->each;
    say "    $two => '$english_name'," if length($two) == 2;
}
__END__
Output:
    aa => 'Afar',
    ab => 'Abkhazian',
    af => 'Afrikaans',
    ak => 'Akan',
    sq => 'Albanian',
...
[download]

Only after that I found that use Locales; Locales->new->code2language("qu") does too what I needed.

Thus is the life of the Perl developer: we reinvent stuff because it's just so easy to do.