| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | martellilaurent(at)gmail(dot)com |
| Cc: | pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: BUG #14652: pg_dump: VIEW dumped as TABLE |
| Date: | 2017-05-12 23:27:55 |
| Message-ID: | 18549.1494631675@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
martellilaurent(at)gmail(dot)com writes:
> Some views of are dumped as TABLE. And this is really annoying because it
> has "REPLICA IDENTITY NOTHING" which is not understood if I try to restore
> in 9.1.
There has never been any promise that you could load output from
pg_dump version X into server versions before X without manual
adjustments. Furthermore, this case doesn't even pose any particular
need for manual adjustments: you need only ignore the error.
I see no bug here.
> But if I dump the whole schema, I get a TABLE with RULE:
FWIW, that typically happens when needed to break a circular dependency.
In this case, since the GROUP BY is obviously inadequate to make the
SELECT legal on its own, I speculate that the query is legal only because
invoice_invoice.id is a primary key, so that the rule has to be emitted
after that pkey constraint is created.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2017-05-13 14:02:18 | Re: BUG #14651: Uninitialized page fix corrupted TOAST table |
| Previous Message | martellilaurent | 2017-05-12 22:19:48 | BUG #14652: pg_dump: VIEW dumped as TABLE |