On Wed, 2010-08-11 at 11:03 -0400, Tom Lane wrote:
> Simon Riggs <simon@2ndQuadrant.com> writes:
> > On Wed, 2010-08-11 at 13:25 +0300, Heikki Linnakangas wrote:
> >> I concur that Boxuan's suggested "difficult" approach seems like the
> >> right one.
>
> > Right, but you've completely ignored my proposal: lets do this in two
> > pieces. Get what we have now ready to commit, then add support for
> > partitioning later, as a second project.
>
> Do we really think this is anywhere near committable now?
>
> If it's committable in every other respect, I could see just having it
> throw a NOT_IMPLEMENTED error when the target table has children.
> I thought we were still a very long way from that though.
Well, if we go off chasing this particular goose then we will set
ourselves back at least one commitfest. I'd rather work towards having a
fully committable patch without inheritance sooner than an even bigger
patch arriving later in the cycle, which could make things difficult for
us.
I cite recent big patch experience as admissible evidence, m'lord.
-- Simon Riggs www.2ndQuadrant.comPostgreSQL Development, 24x7 Support, Training and Services