Re: Temporal Units

[Rich Shepard <rshepard(at)appl-ecosys(dot)com>]

On Sun, 29 Apr 2007, Tom Lane wrote:

> What gives you the idea that type INTERVAL is Postgres-specific? It's in
> the SQL standard.

Tom,

  I know that and that was not to what I referred. Perhaps I mis-understood
Alexander's reference to the internal postgres interval support as different
from the SQL standard INTERVAL. If so, it's my mis-writing.

  Regulatory requirements are that monitoring is to be done 'once per
shift,' 'daily,' 'weekly,' 'twice each month,' or 'monthly.' That does not
mean the same day of the week or the same date each month. Even more vague
is the requirement for monitoring after a storm event that results in 0.5
inch of rain or more.

  I've not dealt with this type of temporal situation before. I've recorded
monitoring dates and have calculated intervals from them. But, now I want to
design the tables so as to make it easy to determine compliance with the
monitoring criteria. Perhaps I am in a deep rut and cannot see the obvious
solution just over the rim.

Rich

--
Richard B. Shepard, Ph.D.               |    The Environmental Permitting
Applied Ecosystem Services, Inc.        |          Accelerator(TM)
<http://www.appl-ecosys.com>     Voice: 503-667-4517      Fax: 503-667-8863