At Tue, 01 Jul 2014 16:30:41 +0900, Etsuro Fujita <fujita.ets...@lab.ntt.co.jp> 
wrote in <53b263a1.3060...@lab.ntt.co.jp>
> I've got the point.
> As I said upthread, I'll work on calculating attr_needed for child
> rels, and I hope that that will eliminate your concern.

Inheritance tree is expanded far after where attr_needed for the
parent built, in set_base_rel_sizes() in make_one_rel(). I'm
afraid that building all attr_needed for whole inheritance tree
altogether is a bit suffering.

I have wanted the point of inheritance expansion earlier for
another patch. Do you think that rearranging there? Or generate
them individually in crete_foreign_plan()?

Anyway, I'm lookin forward to your next patch. So no answer


Kyotaro Horiguchi
NTT Open Source Software Center

Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:

Reply via email to