| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | fluxent(at)yahoo(dot)com (Bill Seitz) |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: massaging data from multiple timezones |
| Date: | 2004-01-19 04:48:01 |
| Message-ID: | 20487.1074487681@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
fluxent(at)yahoo(dot)com (Bill Seitz) writes:
> The raw data comes in as UTC, and the current code converts it to
> server-local time and inserts it into the database (the event_time
> field is type 'timestamp').
> But almost all the reporting should be "thinking" in the *source's*
> timezone. Rather than having to alter 50 different programs that touch
> the data, it seems more sensible to me to have that initial data table
> store the data in source-local time.
Why don't you store it in UTC and read out in whatever timezone you
want? That is what the Postgres timestamptz type is designed to do...
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2004-01-19 04:51:20 | Re: UNICODE-encoded database does not accept umlaut-characters. |
| Previous Message | Tom Lane | 2004-01-19 04:40:20 | Re: Adding long text to a database text field generates errors |