From: | Sam Mason <sam(at)samason(dot)me(dot)uk> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Best database model for canvassing (and analysing) opinion |
Date: | 2009-08-18 12:49:11 |
Message-ID: | 20090818124911.GR5407@samason.me.uk |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Tue, Aug 18, 2009 at 05:24:52AM -0700, John R Pierce wrote:
> Sam Mason wrote:
> > SELECT course, date, COUNT(*)
> > FROM application_preferred_date
> > GROUP BY course, date;
>
> the problem as stated is more complex than that. A student could, in
> theory, pick several different courses on the same dates on the
> assumption that he doesn't care what course on which date.
Yup
> so if the
> optimizer/solver has counted this student for course1 on date1, he can't
> be counted for course2 on that same date, only on an alternate date he
> may have also specified.
Huh, fun optimisation problem. There must be standard solutions to it
though and I'm pretty sure it's something I'd like to solve outside the
database. Or am I missing something else?
I was just answering call for "advice/tips/pointers/suggestions for a
database design newbie".
--
Sam http://samason.me.uk/
From | Date | Subject | |
---|---|---|---|
Next Message | Pavel Stehule | 2009-08-18 12:54:57 | Re: design, plpgsql and sql injection in dynamically generated sql |
Previous Message | Pavel Stehule | 2009-08-18 12:48:19 | Re: There are procedures in Postgres 8.3? |