From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Alvaro Herrera <alvherre(at)commandprompt(dot)com> |
Cc: | Kynn Jones <kynnjo(at)gmail(dot)com>, pgsql-general General <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Why does pg_dump set default_with_oids to true? |
Date: | 2009-10-26 20:11:56 |
Message-ID: | 15045.1256587916@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
> Kynn Jones escribi:
>> The documentation deprecates setting this variable to true. Why does
>> pg_dump do it?
> To dump tables that were created WITH OIDS.
What is deprecated is the use of the WITH OIDS feature, not the variable
;-). But it is not pg_dump's charter to enforce that deprecation.
Maybe what you are asking is why it uses SET default_with_oids rather
than attaching WITH OIDS to the CREATE TABLE commands. That was done
as a compatibility measure so that the dump files would still be
readable by PG versions that didn't understand WITH OIDS --- they'd
reject the SET commands but still accept the CREATE TABLE commands.
By the same token the files are a shade more portable to other DBMSes
than they would be with WITH OIDS clauses. Neither of these arguments
is hugely compelling, perhaps, but avoiding the use of a variable isn't
very compelling either ...
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Dave Page | 2009-10-26 20:15:50 | Re: [ANNOUNCE] PGDay.EU 2009 - approaching fast! |
Previous Message | John R Pierce | 2009-10-26 20:09:20 | Re: Logging statements with errors in PostgreSQL 8.1 |