Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: De-duplication of large table

Re: De-duplication of large table

From: Steve McDaniels <steve.mcdaniels_at_sierra.com>
Date: Mon, 22 Nov 1999 11:15:17 -0800
Message-ID: <81c4os$mmp$1@plo.sierra.com>


Isn't there an option when using ALTER TABLE something like this?

create an exceptions table (see UTLEXCPT.SQL script)

alter table my_table
add constraint pk_my_table
primary key (my_field)
using index
exceptions into EXCEPTIONS;

This will produce a table which contains the ROWIDS of all rows violating the constraint.
which is hopefully, many times smaller that your main table.

use this to remove the dup records you don't want.

(I'm looking at the Oracle 8 Server SQL Reference, page 4-219, page 4-223)

Dave <none_at_nowhere.nothere.oops> wrote in message news:811m34$na5$1_at_lure.pipex.net...
> I have a 43,000,000 row table with about 225,000
> duplicate records.
> What is the easiest way to remove the duplicates.
> I want to remove one of every duplicate record
> from the table.
>
> Dave
>
>
>
Received on Mon Nov 22 1999 - 13:15:17 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US