Re: Entering '<1' and 'BDL'

On Tue, 2010-04-20 at 11:38 -0600, Ewan, Michael wrote:
> As you noted, NULL will fit the bill for your NAN values, but make sure the table allows NULL's in that column and that your code handles NULL's either by skipping them (a self join can often help) or by converting them to 0 if you want them to count in the statistics.

I missed most of this discussion but, why aren't you using numeric which
actually supports NaN?

Joshua D. Drake

>
>
> > -----Original Message-----
> > From: pdxpug-owner(at)postgresql(dot)org [mailto:pdxpug-owner(at)postgresql(dot)org]
> > On Behalf Of Rich Shepard
> > Sent: Monday, April 19, 2010 4:12 PM
> > To: pdxpug(at)postgresql(dot)org
> > Subject: [pdxpug] Entering '<1' and 'BDL'
> >
> > For a scientific database holding observed and measured values there
> > are
> > two non-numeric (NAN) legitimate entries: <1 and 'below detection
> > limits.'
> > Is anyone aware of a standard way of expressing these?
> >
> > The problem is that the column needs to be of a numeric type because
> > the
> > numbers will be manipulated mathematically and summarized
> > statistically. So
> > a NAN such as '<1' or 'BDL' doesn't cut it.
> >
> > Any thoughts?
> >
> > Rich
> >
> > --
> > Richard B. Shepard, Ph.D. | Integrity
> > Credibility
> > Applied Ecosystem Services, Inc. | Innovation
> > <http://www.appl-ecosys.com> Voice: 503-667-4517 Fax: 503-667-
> > 8863
> >
> > --
> > Sent via pdxpug mailing list (pdxpug(at)postgresql(dot)org)
> > To make changes to your subscription:
> > http://www.postgresql.org/mailpref/pdxpug
>

--
PostgreSQL.org Major Contributor
Command Prompt, Inc: http://www.commandprompt.com/ - 503.667.4564
Consulting, Training, Support, Custom Development, Engineering