Well here is another rather sick way to do it if you want to keep the funky data structure you got. Note: I am not saying this is better or faster, just ugly and sort of fun

sub OpenFF {
    my ($file,$dlmt) = @_;
    $dlmt = $_[1] || "\t";
    my(%hash,%row);    
    open(READ,$file) or return;
    my @headers = split /$dlmt/, <READ>;
    s/^\s+|\s+$//g for @headers;
    chomp(@row{@headers} = split /$dlmt/, <READ>) 
    and %hash = map { $_=>{ $hash{$_} ? %{$hash{$_}} : (), $row{$heade
+rs[0]}=>$row{$_}} } @headers
    while !eof(READ);
    return %hash;
}
[download]