the correct way is to use exists on each element.

But if you wanna check multiple keys in one run, why not using keys to get them?

update

you could try this to apply hash-slices:

  DB<230>  sub tst{
             my $href=shift;
             my %needed;
             @needed{qw/name unit quantity/}=();
             delete @needed{ keys %$href };
             print "Elements missing; ", keys %needed if %needed;
           }

  DB<231> tst({name=>1,bla=>2,unit=>3,quantity=>4})

  DB<232> tst({name=>1,bla=>2,unit=>3})
Elements missing; quantity

  DB<233> tst({bla=>2,unit=>3})
Elements missing; quantityname
[download]

but this is certainly easier understood and maintained:

  DB<234>  sub tst{
             my $href=shift;
             my @missing = grep { ! exists $href->{$_} } qw/name unit 
+quantity/;
             print "Elements missing: @missing" if @missing;
           }

  DB<235> tst({name=>1,bla=>2,unit=>3,quantity=>4})

  DB<236> tst({name=>1,bla=>2,unit=>3})
Elements missing: quantity
[download]