| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Richard Huxton <dev(at)archonet(dot)com> |
| Cc: | pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: pg_dump/restore issue with schemas |
| Date: | 2003-10-03 15:20:13 |
| Message-ID: | 19762.1065194413@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
Richard Huxton <dev(at)archonet(dot)com> writes:
> table public.a
> function reports.f()
> The function refers to "a" without the full schema (i.e. not as "public.a")
> The function was originally defined with its name as "reports.f" while
> search_path = public...
> On dump/restore the search_path is set to reports, pg_catalog so of
> course you get a "no relation a" error
This is an SQL function right?
This seems to be an additional and fairly critical reason to disable
checking of SQL function bodies during a reload. I had already proposed
doing so:
http://archives.postgresql.org/pgsql-hackers/2003-09/msg00970.php
but that seemed to have been greeted by a collective yawn, so I was not
planning to do it for 7.4. Now I'm thinking we had better do it.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Richard Huxton | 2003-10-03 18:10:40 | Re: pg_dump/restore issue with schemas |
| Previous Message | Alexandr S | 2003-10-03 14:43:36 | bug reporting |