From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Ted Petrosky" <tedpet5(at)yahoo(dot)com> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: BUG #2867: FULL PATH name problem |
Date: | 2006-12-28 01:17:50 |
Message-ID: | 28157.1167268670@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
"Ted Petrosky" <tedpet5(at)yahoo(dot)com> writes:
> I have been working with WebObjects and the EOModeler. When I tell EOModeler
> to examine a table, it comes up with this for the primary key.
> ALTER TABLE public.article ADD CONSTRAINT public.article_PK PRIMARY KEY
> (entityid);
> I must change it to be:
> ALTER TABLE public.article ADD CONSTRAINT article_PK PRIMARY KEY
> (entityid);
> because postgresql doesn't like the 'public.' after the CONSTRAINT. I guess
> the question is, should it?
No, I think that'd be a bad idea. The SQL spec does call for a
schema-qualified constraint name here, but that's because they have a
different model of the constraint namespace than we do, to wit, unique
per schema vs. our unique per table. Unless you want to buy into the
SQL naming model, you shouldn't be trying to specify a schema name here.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tigran Mkrtchyan | 2006-12-28 12:12:22 | transactions getting slon in councurrent environment |
Previous Message | Ted Petrosky | 2006-12-28 00:15:18 | BUG #2867: FULL PATH name problem |