eneral
Sent by: cc:
On Sep 29, 2005, at 1:23 PM, Keary Suska wrote:
You should always finish() every "select" statement handle, or both
DBI and
libpq will leak:
$sth->finish;
after the closing bracket of the while() loop.
No, you don't need to call finish() if you fall off the end of a
while $sth->fet
on 9/29/05 6:38 AM, [EMAIL PROTECTED] purportedly said:
> Any ideas please? This seems well weird to me, but...
Assuming the code snippet is representative, a couple things:
1. $dbh->prepare($stmt)->execute () || die;
This is mostly a point of protocol, but I can't resist pointing it out.
Altho
On Sep 29, 2005, at 8:38 AM, Bob Parkinson wrote:
Upgraded PG to 8.0.3, DBI to 1.48 and DBD::Pg to 1.43 this morning,
but made no difference. FreeBSD 5.4 (they'll get it right
RSN :- )), perl 5.8.2
I've had bad mem leaks on FreeBSD with perl < 5.8.6. Try upgrading
that as well.
--
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> Any ideas please? This seems well weird to me, but...
I could not duplicate this. Can you provide a self-contained test case?
It's not useful if we don't know what's actually in the table and what
the table structure looks like.
- --
Greg Sabino
Hi,
I'd really like to aviod another list (ie. interfaces) if at all possible, so...
Got a script (trimmed version below) that starts to have exponential memory use
after a number of iterations around the while(($sth->fetch())) { } struct.
Table is quite small (520 odd rows) and the row siz
