On 2/28/06, Jovan Kostovski <chom_devel@xxxxxxxxx> wrote:
Hi,
I have the following table running on postgres 8.04:
CREATE TABLE ValueLog
(
idValueLog int4 NOT NULL DEFAULT nextval ('public.idValueLog_seq'::text),
PPx integer,
iValue integer DEFAULT -1,
fValue real DEFAULT -1,
t_tag timestamp without time zone,
t_arr timestamp without time zone,
cot integer,
ack boolean,
blk boolean,
qd_iv boolean,
qd_nt boolean,
qd_sb boolean,
qd_bl boolean,
qd_ov boolean,
CONSTR AINT idValueLog PRIMARY KEY (idValueLog)
)
WITHOUT OIDS;
ALTER TABLE ValueLog OWNER TO $admin_user;"
I've created the following SQL statement to select the records newer then a given date:
select almvalue.almname , ppoint.ppdimstr, valuelog.ack, valuelog.blk, to_char(valuelog.t_arr,'DD.MM.YYYY HH24:MI:SS') from ppoint, valuelog, almvalue where ppoint.ppx = valuelog.ppx and almvalue.almvalue = valuelog.ivalue and valuelog.ppx = $db_ppx and to_char(valuelog.t_arr,'YYYYMMDD') >= '$ts
$ts is a string, the date value input by the user and formated YYYYMMDD
I've developed the web site on my Ubuntu box, but when I uploaded the site on the server Fedora Core 3, I had problems with the date filtering. Running psql on the server I've noticed that there were more data in the select * from value log results, field t_arr something like MM-DD-YYYY-dsfsdfd (something else which I don't know what it is)
I think that the local time settings on the server makes me the troubles.
What's the solution?
I've even checked the result of:
select to_char( valuelog.t_arr,'DD.MM.YYYY HH24:MI:SS') from valuelog;
but instead of data I got empty data set. Why?
What's the best way to do filtering (selecting) data by time?
TIA, Jovan
Yahoo! Mail
Bring photos to life! New PhotoMail makes sharing a breeze.
--
Atte. Rodrigo A. Llanos N.
Saludos.