"David F. Skoll" <dfs@roaringpenguin.com> writes:
> On Wed, 21 Jul 2004, Tom Lane wrote:
>> pg_dump -t s1.t1 -t s2.t2 -- Dump s1.t1 and s2.t2
> That's a good idea, but then it's questionable whether we need the -n
> switch at all.
Sure we do --- for backwards compatibility if nothing else.
> It might be simpler to extend the -t switch to accept:
> pg-dump -t 's1.*'
That would not be the same thing --- that would mean to dump *only tables*
from s1, rather than objects of all types. Anyway, I think it's a bit
late in this cycle to be proposing to implement wild-card matching.
Maybe for next time someone can do that, but for 7.5 I think we should
limit ourselves to cleaning up any design flaws of the already-submitted
patch.
regards, tom lane