On Sat, Jun 30, 2007 at 01:36:22AM -0400, Tom Lane wrote:
> Neil Conway <neilc@samurai.com> writes:
> > On Sat, 2007-30-06 at 00:26 -0400, Tom Lane wrote:
> >> There is exactly 0 chance of that happening, because it's always
> >> worked historically.
>
> > Agreed, but I think the patch should disallow ALTER VIEW ...
> > RENAME on a non-view, and ALTER SEQUENCE ... RENAME on a
> > non-sequence.
>
> No objection to that; it'd square with our treatment of TYPE and
> DOMAIN commands. What I'm wondering though is whether the whole
> patch has a reason to live at all, as compared to documenting
> someplace more prominent than now that ALTER TABLE works on views &
> sequences.
How could it be prominent short of documentation of the thing people
would expect, which is ALTER [SEQUENCE | VIEW] RENAME TO ... ? I
suppose we could document that they're actually done by ALTER TABLE,
but that just seems like a huge POLA violation, along with assuming
way too much knowledge of how sequences and views are implemented.
Cheers,
D
--
David Fetter <david@fetter.org> http://fetter.org/
phone: +1 415 235 3778 AIM: dfetter666
Skype: davidfetter
Remember to vote!
Consider donating to PostgreSQL: http://www.postgresql.org/about/donate