Robert Haas wrote:
> On Wed, Oct 18, 2017 at 11:27 AM, Alvaro Herrera
> <alvherre@alvh.no-ip.org> wrote:
> > Maybe there are combinations of different persistence values that can be
> > allowed to differ (an unlogged partition is probably OK with a permanent
> > parent), but I don't think the current check is good enough.
>
> This is also a sort of long-standing historical problem, I think.
Sure.
Actually, the code I'm calling attention to is ATExecAttachPartition()
which was specifically written for partitioning. Looks like it was copied
verbatim from ATExecAddInherit, but there's no shared code there AFAICS.
I'm okay with prohibiting the case of different persistence values as
you suggest. And I do suggest to back-patch that prohibition to pg10.
Let me add that I'm not looking to blame anyone for what I report here.
I'm very excited about the partitioning stuff and I'm happy of what was
done for pg10. I'm now working on more partitioning-related changes
which means I review the existing code as I go along, so I just report
things that look wrong to me as I discover them, just with an interest
in seeing them fixed, or documented, or at least discussed and
explicitly agreed upon.
--
Álvaro Herrera https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers