On 11/12/2012 02:24 PM, 高健 wrote: > Hi Craig: > Thank you for your reply. > I checked for constratint_exclusion , in my sample, on and partition > is same(I have no data on parent table). it really works for me. > I tried and found that constraint_exclusion can work in simple ways , > but not for complicated conditions such as "id=a or id=b". > And the union all of two simple query really produced a lower cost. For now you'll probably need to use that approach - UNION ALL of simpler queries. It's cumbersome and annoying, though.
It'd be really interesting to enhance the query planner to be smarter about this particular case, but the planner is way past my scary-code-voodoo level so I can't really help there; I'm more interested in usability issues in the tools for any development time I get. -- Craig Ringer -- Sent via pgsql-general mailing list (pgsql-general@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
