Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!hplabs!pyramid!infmx!davek
From: davek@informix.com (David Kosenko)
Newsgroups: comp.databases
Subject: Re: SQL Duplicate Row Deletion ???
Message-ID: <1991Apr9.154457.8318@informix.com>
Date: 9 Apr 91 15:44:57 GMT
References: <1991Apr1.163615.56@cim-vax.honeywell.com> <670651675.103469@paladin.owego.ny.us> <1991Apr7.054945.8610@egsner.cirr.com>
Sender: news@informix.com (Usenet News)
Distribution: na
Organization: Informix Software, Inc.
Lines: 24

In article <1991Apr7.054945.8610@egsner.cirr.com> clifton
	@egsner.cirr.com (Clifton Bean) writes:
>
>I use Informix.  Their tables all contain a "hidden" index on the row_id of
>the records within the table.  In Informix, I would code the following:

	Actually, we do not have any "hidden" indexes, though it is very
commonly misunderstood that we do.  Rowid is best described as an l-value,
or the location of a particular record in a database table.  With SE databases
(using a Unix file for the data rows), the rowid represents the offset into
the file where the row is found.  With Turbo/OnLine, it provides a page number/
slot table number  combination, which is used in a similar fashion to locate
the row in question.  In no case is there any true index built on this rowid
value.  You can use the rowid in a WHERE clause, in which case you are asking 
for the row at that particular location.

Dave Kosenko
Informix Client Services

-- 
Disclaimer: These opinions subject to change without notice.
**************************************************************************
The heart and the mind on a parallel course, never the two shall meet.
						-E. Saliers