Re: BUG #4957: search_path and pg_dump -T switch
- From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
- To: "Maxim.Boguk" <maxim(dot)boguk(at)gmail(dot)com>
- Cc: pgsql-bugs(at)postgresql(dot)org
- Subject: Re: BUG #4957: search_path and pg_dump -T switch
- Date: Fri, 31 Jul 2009 09:59:43 -0400
- Message-id: <18556.1249048783@sss.pgh.pa.us> <text/plain>
"Maxim.Boguk" <maxim(dot)boguk(at)gmail(dot)com> writes:
> Now we have table test1 with one row outside of search_path and table test2
> with one row inside of search path.
> Now
> pg_dump -F p -D -T test1 -a test_db | grep test1
> will out:
> INSERT INTO test1 (id) VALUES (1);
> And
> pg_dump -F p -D -T test2 -a test_db | grep test2
> will out empty.
> First result is wrong.
No, I don't believe it is. The switch means "don't dump the table named
test1 as found in your search path". So in this case it doesn't do
anything. You could do "-T test.test1" or "-T *.test1" if you want to
suppress that table.
We could have pg_dump throw an error if -t or -T doesn't seem to refer
to any actual table, but I'm not sure that would make it more useful.
Particularly not with wild-card-pattern switches.
regards, tom lane
Home |
Main Index |
Thread Index