From: JRStern@gte.net (JRStern)
Newsgroups: comp.databases.theory
Subject: Re: another simple data modelling question
Message-ID: <3b0476e9.32763421@news.gte.net>
References: <8QWM6.6168$6j3.557704@www.newsranger.com>
X-Newsreader: Forte Free Agent 1.1/32.230
Lines: 17
X-Trace: /w3Za7eKnmz2iaNm4gqsSc/RsKxrSRLPEA45JagRmlt5vuu9aNmia36Kb8GgqyitSpU0QKRRXuUy!H4N96XFmBLwuch3Uh8mDdMEr1U7DLeYuRzsW4R0nT36+qEuvNXF1m/w5wXATw/w=
X-Complaints-To: abuse@gte.net
X-Abuse-Info: Please be sure to forward a copy of ALL headers
X-Abuse-Info: Otherwise we will be unable to process your complaint properly
NNTP-Posting-Date: Fri, 18 May 2001 01:09:36 GMT
Distribution: world
Date: Fri, 18 May 2001 01:09:36 GMT


On Thu, 17 May 2001 20:50:12 GMT, crappy <nospam@newsranger.com>
wrote:
>now my problem is this: how do you ensure that person/characteristic pairs are
>valid with respect to the characteristic/personality pairs, *as well as* the
>implied personality of each person (via the child-parent relationship between
>person and personality)?
>
>in other words, with the given setup, it is possible to give a person a
>characteristic (in person_characteristic) that isn't 'allowed' for that person's
>personality.

Apparently you have some such constraints, of characteristics allowed
for a personality.  Put them in a fourth table, put a trigger on the
person_characteristic table to check it.

J.