Hi Sandy_Bio_Perl,
I think that the problem isn't with the IS NOT NULL, but rather with the division on the column with the null value. Here's something I scratched together that works. Note the IFNULL in your SQL statement, as well as csv_null => 1 in the connection parameters:
#!/usr/bin/perl
use strict;
use warnings;
use DBI;
sub RunSqlSearch;
my $spreadsheet = 'newCsv.csv';
my $query
= "SELECT sid,genotype, dnabl, dnatw12 "
. "FROM $spreadsheet "
. "WHERE genotype = 'a' "
. "AND (IFNULL(dnatw12,0)/dnabl)<0.5 "
. "AND dnatw12 IS NOT NULL "
;
my ($queryResult) = RunSqlSearch($query);
print "Your query $query returned the following result:\n$queryResult\
+n";
sub RunSqlSearch($){
my $query = $_[0];
# error check #
if ($query eq ""){die "From runSqlQuery2 - No value entered for $
+query $!\n";}
# Connect to the database, (the directory containing our csv file(
+s))
my $dbh = DBI->connect ("dbi:CSV:", undef, undef, {
f_dir => ".",
f_encoding => "utf-8",
csv_eol => "\n",
csv_null => 1,
RaiseError => 1,
}) or die $DBI::errstr;
# Output using sql query #
my $sth = $dbh->prepare($query);
$sth->execute;
my @row;
my $queryResult="";
while (@row = $sth->fetchrow_array) {
$queryResult .= join("\t",@row) . "\n";
}
# output arguments #
if ($queryResult eq ""){$queryResult = "No result found";}
return ($queryResult);
}
Hope that points you in the right direction.
-bmt
UPDATE: Note that this will still blow up if dnabl is NULL. I played around a bit, but the IFNULL that worked in the numerator of the division failed to work in the denominator. Not sure why...
