On 10 Aug 2002, Rob Brown-Bayliss wrote:
> On Sat, 2002-08-10 at 08:13, Elaine Lindelef wrote:
>
> > Add another column of type SERIAL and use it to store a plain
> > incremented integer, and use it like you used to use OIDs.
>
> Except that this brings me back to the reason for not useing a sequence
> as a primary key:
>
> What happens when site2 has already placed a row with value 1234 in this
> column and site1's sequence is up to 1233, I insert the row, and now I
> have two rows with 1234 as the ID row...
>
> At least OID's will be unique as when inserting rows from another site
> they still use the OID from the local machine.
I believe you are misunderstanding the suggestion.
what happens if for your table without OID you define a column called oid with
a sequence attached? How is that different to using the OID column from a
table with OID?
I haven't paid attention to this thread until now but are you confusing primary
key with OID?
And in a separate message you ask how will PostgreSQL identify data if OIDs are
removed from the system? Well doesn't the data identify the data?
--
Nigel J. Andrews
Director
---
Logictree Systems Limited
Computer Consultants