From: | John R Pierce <pierce(at)hogranch(dot)com> |
---|---|
To: | pgsql-jdbc(at)postgresql(dot)org |
Subject: | Re: Upgrade to Scientific Linux 6.5 x86_64 breaks query |
Date: | 2014-07-23 19:33:09 |
Message-ID: | 53D00DF5.3030806@hogranch.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-jdbc |
On 7/23/2014 12:13 PM, CG wrote:
> The parenthesis don't work, but if I explicitly cast that second
> parameter to interval "CAST( $2 AS interval)" I can force it to work.
> That's not ideal.
>
> I don't think it is the lack of an operator. Timestamptz should not be
> comparable directly to an interval, right? That's like saying "2 + 2
> >= Orange" or "today is greater than a minute". I think something is
> eating that first "now() -" so that what is being prepared looks like
> "trans_date between $2 and now()".
>
> The question is what is causing valid SQL, prepared properly on 32bit
> linux to fail to prepare properly on 64bit linux? Based on that log,
> it looks like client-side prepare is falling down.
I think I'd try some experiments with select pg_typeof($1), passing it
your 'interval' value and see what it thinks it is.
test=# select pg_typeof(interval '1 day');
pg_typeof
-----------
interval
(1 row)
test=# select pg_typeof('1 day');
pg_typeof
-----------
unknown
(1 row)
--
john r pierce 37N 122W
somewhere on the middle of the left coast
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2014-07-23 19:39:35 | Re: Upgrade to Scientific Linux 6.5 x86_64 breaks query |
Previous Message | John Neal | 2014-07-23 19:25:36 | JBDC LDAP support for connection parameters |