| From: | "David E(dot) Wheeler" <david(at)kineticode(dot)com> |
|---|---|
| To: | len(at)pdx(dot)edu |
| Cc: | "Jeff Davis" <pgsql(at)j-davis(dot)com>, "Postgresql PDX_Users" <pdxpug(at)postgresql(dot)org>, "David Maier" <maier(at)cs(dot)pdx(dot)edu>, "Mark Wong" <markwkm(at)gmail(dot)com> |
| Subject: | Re: 9/18 Visual Planner Meeting Wrapup |
| Date: | 2008-11-10 23:56:20 |
| Message-ID: | 81322F2A-A6B6-4413-A0D8-FC4CF880E99E@kineticode.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pdxpug |
On Nov 10, 2008, at 3:52 PM, Len Shapiro wrote:
> Perhaps now it is clearer why I am not happy with an "imaginary"
> schema. It does no good if there is another table K such that L join
> K is an FK join but K is not in the database. Because then the
> optimizer cannot check that the target join attribute in K is unique
> and K cannot be sampled.
Yes, having the context from which the question is asked helps a lot.
I much better understand what it was you were trying to ascertain.
Best,
David
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Selena Deckelmann | 2008-11-19 23:12:31 | Fwd: PostgreSQL meetings on Calagator? |
| Previous Message | Len Shapiro | 2008-11-10 23:52:15 | Re: 9/18 Visual Planner Meeting Wrapup |