Hi Ashok,
MJD also found that bug. I've fixed it but not uploaded the new version
yet because I'm fixing a couple of other things. You can apply this
patch yourself if you need it soon:
Changing line 1189 of SQL/Statement.pm like so fixes it:
- $rows = $g->calc;
+ $rows = $g->calc || [];
Sorry for the problems.
--
Jeff
Ashok Vemuri wrote:
Hi,
I am having a bit of trouble using aggregate functions with DBI::CSV
(version 0.22) and could use some help figuring out why.
I have the following code snippet that queries a CSV file named
"masterdata", located at c:/test/output folder.
============================
use DBI;
my $dbh = DBI->connect("DBI:CSV:f_dir=c:/test/output")
or die "Cannot connect: " . $DBI::errstr();
my $sth = $dbh->prepare("SELECT callingPartyUnicodeLoginUserID as user,
count(duration) as calls, sum(duration) as duration FROM cmasterdata WHERE
dateTimeOrigination > 1201979253 AND callingPartyUnicodeLoginUserID not like
'' group by callingPartyUnicodeLoginUserID order by calls desc")
or die "Cannot prepare: " . $dbh->errstr();
eval {
$sth->execute();
}
if ($@) {
print "error: [EMAIL PROTECTED]";
}
=============================
The execute() statement above fails with the following error message when
there are no matching records in the CSV file.
DBD::CSV::st execute failed: Can't use an undefined value as an ARRAY
reference at C:/MYINST~1/objects/perl/ActivePerl-
5.8.8.822-MSWin32-x86-280952/site/lib/SQL/Statement.pm line 1197, <GEN0>
line 242.
[for Statement "SELECT callingPartyUnicodeLoginUserID as
users,count(duration)as calls,sum(duration) as duration FROM masterdata
WHERE dateTimeOrigination > 1201991649 AND callingPartyUnicodeLoginUserID
not like '' group by callingPartyUnicodeLoginUserID order by calls desc"]
However, the code runs fine (on the same CSV file) if I remove aggregate
functions from the query. For example, the following query is fine:
SELECT callingPartyUnicodeLoginUserID as user, dateTimeOrigination as
datetime FROM cmasterdata WHERE dateTimeOrigination > 1201979253 AND
callingPartyUnicodeLoginUserID not like ''
Is there a known issue with DBD::CSV where trying aggregate functions when
no matching records gives an error, or am I just doing something wrong here?
TIA,
Ashok
