The triggers are being used to track changes to the tables. The developers
are concerned that using string references for the table names in this case
would create too much overhead as this is a frequent operation and is
performance critical.
Sounds like we have the choice of using string names or implementing a
custom migration script.
Thanks for all your help Thom.
Regards, Matt
-----Original Message-----
From: pgsql-general-owner@postgresql.org
[mailto:pgsql-general-owner@postgresql.org] On Behalf Of Thom Brown
Sent: 01 May 2012 12:43
To: Matthew Churcher
Cc: pgsql-general@postgresql.org
Subject: Re: [GENERAL] Having trouble with pg_dumpall -o
On 1 May 2012 12:37, Matthew Churcher <Matthew.Churcher@realvnc.com> wrote:
> OK, I think I've worked out what's going on. I've got my wires crossed
> between table column OIDS (deprecated) and the OID which uniquely
> identifies each table (?always enabled?).
>
> We're not using OID for each column, only to reference the tables
> themselves as that's how triggers are referring to them.
>
> It appears the -o flag is for migrating table column oids which we're
> not using.
>
> So... any ideas how we can migrate the OID of the table itself? Or are
> we doing something we shouldn't?
You shouldn't be relying on OIDs as they're not really for the end-user.
Instead you should refer to objects by name. How are you using them in
triggers?
--
Thom
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org) To make
changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general