Re: Need to delete duplicate rows
From: David Owen <dowen_at_midsomer.org>
Date: 2000/06/24
Message-ID: <m2g0q3plh3.fsf_at_kashka.midsomer.org>#1/1
Date: 2000/06/24
Message-ID: <m2g0q3plh3.fsf_at_kashka.midsomer.org>#1/1
>>>>> "William" == William Boyle <woboyle_at_ieee.org> writes:
William> This is fine for Oracle, but rowid is not valid for other William> DBMS. Some have an equivalent construct, but the SQL standard William> most definitely does not support the notion of a row William> identifier. In fact, it is anathema to relational theory all William> together. In a properly specified relation/table - one that William> has a proper primary key - duplicates tuples/rows are not William> possible as a relation is properly a set.
Anathema they may be, but it is possible to have them surely? Its just that in a properly constructed relation/table you could not distinguish between them.
-- David Owen Midsomer Consultants Inc. dowen_at_midsomer.org Maintainer of the Sybase FAQ: http://www.isug.com/Sybase_FAQReceived on Sat Jun 24 2000 - 00:00:00 CEST