Re: OK, OK, I'm taking the DBI plunge. Now what?

[runrig]

Unless the first row of your text file contains the column names, you'll need to use the col_names attribute. See the DBD::CSV docs and examples.

[peppiv]

 use strict;
    use DBI;
    use CGI::Carp qw(fatalsToBrowser);
    use CGI qw (:standard);
    print "Content-type: text/html\n\n";
    
     my $dbh = DBI->connect(qq{DBI:CSV:f_dir=/usr/local/etc/httpd/cgi-
+bin/}, {RaiseError => 1} );

     $dbh->{'sep_char' => "|", 'col_names' => ["timestamp", "email", "
+name", "address", "city", "state", "zip"]};
     my $sth = $dbh->prepare( q{SELECT * FROM export_list WHERE state 
+= ? });

     my $rc = $sth->execute ('OH');

     print "Query will return $sth->{NUM_OF_FIELDS} fields.\n\n";

     while (my @result = $sth->fetchrow_array()) {
        print "@result\n";
        }
     
     die $sth->errstr if $sth->err;

     $dbh->disconnect;


I've used the col_names attribute and I don't receive any errors for that line. However I do get this error:
Attempt to fetch row from a Non-SELECT statement at /usr/local/etc/httpd/cgi-bin/DBI_start3.pl line 23.(*the fetchrow_array statement line).I'm researching this now. Question - If an error statement comes up for line 23, is it safe to assume that the lines of code before that are executing OK? Assuming you're not jumping around with subs.

[runrig]

You are not specifying which table you are setting the column names on. Look at the DBD::CSV docs. Your columns should be set like so:

$dbh->{'csv_tables'}->{'export_list'}={
  'sep_char' => "|",
  'col_names' => [qw(timestamp email name address city state zip)]};

Also, execute returns no useful information on a select statement, so there is no point in saving its return value. On a non-select statement (e.g. update, delete), it returns the number of rows affected.

And your die() statement is useless since you have RaiseError set. If there is a DBI error, DBI will execute a die() on its own before you ever get to that die statement.