Path: dp-news.maxwell.syr.edu!spool.maxwell.syr.edu!drn.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.icl.net!newsfeed00.sul.t-online.de!t-online.de!inka.de!not-for-mail From: "Felix E. Klee" Newsgroups: comp.databases.theory Subject: Re: How to ensure data consistency? Date: Thu, 9 Sep 2004 01:32:34 +0200 Organization: (Posted via) INKA e.V. http://www.inka.de/ Lines: 27 Message-ID: <20040909013234.66839af9.felix.klee@inka.de> References: <3lmnj0tdqs6n98so05hslbs6b0rjqkkdk7@4ax.com> <7dhqj057v2nb3jvgig151psbhl1mg9hrsm@4ax.com> <3fsrj0lpe76pfvfjfe1ur2ubpb3hqu69fg@4ax.com> <20040907154957.697949a1.felix.klee@inka.de> <6dae7e65.0409080826.74822bd7@posting.google.com> NNTP-Posting-Host: puric.inka.de Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Trace: sapa.inka.de 1094686614 17326 193.197.184.17 (8 Sep 2004 23:36:54 GMT) X-Complaints-To: abuse@inka.de NNTP-Posting-Date: 8 Sep 2004 23:36:54 GMT X-Newsreader: Sylpheed version 0.9.9 (GTK+ 1.2.10; i486-slackware-linux-gnu) Xref: dp-news.maxwell.syr.edu comp.databases.theory:25875 On 8 Sep 2004 09:26:06 -0700 Lennart Jonsson wrote: > > I have a similar problem, although mine is about enforcing "complete > > disjoint specialization", whereas yours is about enforcing "incomplete > > disjoint specialization" (terminology is from the article that you > > pointed to). > > > > Not sure whether or not you found an answer to your question (I seem > to miss some posts in this thread). Any how, here is how I would model > it: > > [...] That's a very nice solution. Thanks. The underlying concept seems to be identical to that presented in a post by Joe Celko [1] to which Dan pointed in this thread. > As someone already mentioned it doesnt protect against a PERSORG > having 0 persons and 0 organisations. For that we would need defered > constraints or something similar. I wonder if PostgreSQL allows this. Felix [1] http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&selm=OEYKH2bbEHA.368%40TK2MSFTNGP10.phx.gbl&rnum=5