Re: Remove duplicates from a field possible?

From: Michel Cadot <micadot{at}altern{dot}org>
Date: Tue, 7 Mar 2006 06:44:57 +0100
Message-ID: <440d1d59$0$23316$636a55ce@news.free.fr>

"Mladen Gogala" <gogala_at_sbcglobal.net> a écrit dans le message de news: pan.2006.03.06.21.24.40.976859_at_sbcglobal.net...
| On Mon, 06 Mar 2006 17:56:53 +0100, Michel Cadot wrote:
|
|
| Allow me to propose a modified kyteian solution. Modification consists
| of using the EXCEPTIONS table, which will significantly speed up the
| whole thing in cases where the base table is large and there are
| relatively few duplicate rows. Column naming (mad_max) was my own and
| not Tom's. I call it a kyteian solution, because Tom Kyte wrote about
| using the analytic functions for duplicate rows removal in his book
| "Effective Oracle by Design". If the underlying table has 100M rows
| and there are 20,000 duplicates, my procedure will be the fastest
| way of doing it.
|

<exemple snipped>
|
| --
| http://www.mgogala.com
|

Exception table is my favourite tool to remove duplicates but here we don't have duplicate between rows but duplicate inside a field in a row. You can't use this way to eliminate the duplicates.

Regards
Michel Cadot Received on Mon Mar 06 2006 - 23:44:57 CST