Re: Inheriting table AMs for partitioned tables - Mailing list pgsql-hackers

From David Rowley
Subject Re: Inheriting table AMs for partitioned tables
Date
Msg-id CAKJS1f-8sm+DduF0jQhaXtB+knNLyhaBSZ-wsQEARr-SoAz+7g@mail.gmail.com
Whole thread Raw
In response to Re: Inheriting table AMs for partitioned tables  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
On Wed, 6 Mar 2019 at 07:19, Robert Haas <robertmhaas@gmail.com> wrote:
>
> On Tue, Mar 5, 2019 at 12:59 PM Andres Freund <andres@anarazel.de> wrote:
> > Based on this mail I'm currently planning to simply forbid specifying
> > USING for partitioned tables. Then we can argue about this later.
>
> +1.  I actually think that might be the right thing in the long-term,
> but it undeniably avoids committing to any particular decision in the
> short term, which seems good.

I've not really been following the storage am patch, but given that a
partition's TABLESPACE is inherited from its partitioned table, I'd
find it pretty surprising that USING wouldn't do the same.  They're
both storage options, so I think having them behave differently is
going to cause some confusion.

I think the patch I just submitted to [1] should make it pretty easy
to make this work the same as TABLESPACE does.

[1] https://www.postgresql.org/message-id/CAKJS1f_iyBpAuYBPQv_GGeME%3Dg9Rpr8yWjCaYV4E685yQ1uzkw%40mail.gmail.com

-- 
 David Rowley                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services


pgsql-hackers by date:

Previous
From: David Rowley
Date:
Subject: Re: pg_dump is broken for partition tablespaces
Next
From: Michael Meskes
Date:
Subject: Re: ECPG regression with DECLARE STATEMENT support