CountZero!

Thanks for that awesome snippet. I'm going to give it a try for sure. It will be useful in many cases for me and I'll let you know about my success with it. Thanks for the level of detail you provided! :-)

I just wanted to clarify a bit from my post(s). I currently have a process that connects to a DB, does a SELECT *, and then outputs a CSV. I'm using Text::CSV_XS to join my arrary_ref together, by hand. I may have given the impression I'm using DBD::CSV, but I'm not. Before outputting the CSV, because I have a DBI statement handler, I have access to:

    my $type_ref =  $stmt->{TYPE};
    my $precision_ref =  $stmt->{PRECISION};
    my $scale_ref =  $stmt->{SCALE};
    my $colnames = $stmt->{NAME};
    my $nullable = $stmt->{NULLABLE};
[download]

I should have mentioned that there are no foreign keys on these tables and they are essentially flat.

Instead of dumping the CSV, like I am now, I'd like to be able to generate the CREATE TABLE DDL from the statement handler I already have.

I'll continue to play around.

Thanks for your insights.

Kurt