Tom Lane wrote:
>
> Fernando Nasser <fnasser@redhat.com> writes:
> > The POSTQUEL extension of adding the tables for you (if I understood
> > right) is an aberration (if it is still supported it will ave to be
> > removed).
>
> No it won't. The implicit-RTE extension doesn't come into play until
> after you've failed to find a matching RTE. It cannot break queries
> that are valid according to spec --- it only affects queries that should
> flag an error according to spec.
>
> My question is about what it means to find a matching RTE and when two
> similarly-named RTEs should be rejected as posing a name conflict.
> Implicit RTEs are not relevant to the problem.
>
That was a side question, as I though this could get in the way.
I am glad it doesn't.
The rest I said is still valid and is unrelated to this.
BTW, I believe Oracle got the standard right this time.
What Joe Conway has been posting is exactly what I understood.
--
Fernando Nasser
Red Hat Canada Ltd. E-Mail: fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario M4P 2C9