From: | "Dann Corbit" <DCorbit(at)connx(dot)com> |
---|---|
To: | "Pgsql Group" <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: About OIDs |
Date: | 2003-03-28 01:47:40 |
Message-ID: | D90A5A6C612A39408103E6ECDD77B829408AB6@voyager.corporate.connx.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
> -----Original Message-----
> From: Martijn van Oosterhout [mailto:kleptog(at)svana(dot)org]
> Sent: Thursday, March 27, 2003 5:39 PM
> To: Dann Corbit
> Cc: Pgsql Group
> Subject: Re: [GENERAL] About OIDs
>
>
> On Thu, Mar 27, 2003 at 01:44:06PM -0800, Dann Corbit wrote:
> > They are very handy for our ODBC driver. It ensures that there is
> > always some sort of unique identifier for any table, even if the
> > designer did not create a primary key. (Sure, that's a
> naughty no-no,
> > but it is astonishing how often it happens). At any rate, some
> > complex joins can be sped up enormously by the existence of Odes.
>
> OIDs are not necessarily unique. How do you handle that?
I do assume that they are unique within a single table's Oid set.
However, I create a unique index on the Oid column, so even when the
counter does wrap, I get a failure and an error message (yes, it has
happened).
When the failure does occur, I am forced to drop the database and
rebuild it. I will have a better solution once the 7.x official win32
port becomes available. God willing and the crick don't rise.
From | Date | Subject | |
---|---|---|---|
Next Message | Dann Corbit | 2003-03-28 01:58:24 | Re: Slow query needs a kick in the pants. |
Previous Message | Martijn van Oosterhout | 2003-03-28 01:43:24 | Re: Python Error |