Re: View updating in practice?

From: Mikito Harakiri <mikharakiri_at_yahoo.com>
Date: 8 Nov 2002 16:17:37 -0800
Message-ID: <bdf69bdf.0211081617.1ee8b8f1@posting.google.com>

"Jens Lechtenbörger <lechtej_at_uni-muenster.de> wrote in message news:<m2of9087s0.fsf_at_pcwi1068.uni-muenster.de>...
> 3. Does anybody use SQL:1999 INSTEAD triggers to implement view
> updates?

Checkout "inverse view" thread on comp.databses.theory:

http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&threadm=PL%25K6.2267%24j65.183466%40www.newsranger.com&rnum=1&prev=/groups%3Fas_q%3Dinverse%2520view%26safe%3Dimages%26ie%3DISO-8859-1%26as_ugroup%3Dcomp.databases.theory%26lr%3D%26as_drrb%3Db%26as_mind%3D12%26as_minm%3D5%26as_miny%3D1981%26as_maxd%3D28%26as_maxm%3D12%26as_maxy%3D2002%26hl%3Den

The original poster shows that if 'view inverse' is known, then writing INSTEAD OF trigger is essentially a mechaical task.

BTW, it looks like the poster rediscovered 'view inverse' and the idea of missing information -- 'view complement' -- without knowing about Bancilhon&Spyratos Paper.

Thank you for the posting, Jan: quality contributions to this group are not that common:-( I'm also enjoying yours "On Computation of Relational View Complements". As far as your unpublished paper, next time don't call it "... considered harmful" -- the CS community allowed only one paper with such a title:-) Received on Fri Nov 08 2002 - 18:17:37 CST