Re: Timestamp input + copy

From: Steve Crawford <scrawford(at)pinpointresearch(dot)com>
To: "Kevin Bartz" <kbartz(at)loyaltymatrix(dot)com>, <pgsql-general(at)postgresql(dot)org>
Subject: Re: Timestamp input + copy
Date: 2004-07-28 23:15:42
Message-ID: 200407281615.42212.scrawford@pinpointresearch.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wednesday 28 July 2004 3:11 pm, Kevin Bartz wrote:
...
> I have a flat file with a column with dates formatted like this:
>
> 2004-04-15 18:04:26 PM
>
> It's a bit strange, I know, but I didn't create the file. My idea
> of Postgres's proper behavior would be to load this date as a
> military time (and ignore the "PM"). MS SQL Server behaves in this
> way. Postgres, however, won't even load the file:

Edit the file to eliminate the PM. I don't know how large the file is
but fixing the date should be pretty easy with vi, sed, perl, etc.
(choose the one that works for you).

What seems like PostgreSQL being annoying and pedantic is really
PostgreSQL protecting your a** - the importance of data integrity and
all that.

Some databases try to be "helpful" (or are just sloppy) so if you try
to insert a number bigger than that allowed by the field it just
truncates it to the largest number that will fit. I hope nobody uses
that database for financial data.

What should a database do if confronted with '2004-04-15 18:04:26 AM'?
I would much prefer the system to throw an error and let me evaluate
and fix it than silently "help" me by loading corrupt data.

Cheers and welcome to PostgreSQL,
Steve

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Daniel Verite 2004-07-28 23:16:06 Re: [DEFAULT] Daily digest v1.4551 (20 messages)
Previous Message Kevin Bartz 2004-07-28 23:06:56 Re: Timestamp input + copy