RE: SQL Delete question

["Elvis Budimlic" <ebudimlic@xxxxxxxxxxxxxxxxxxxxxxxxx>]

While it may not be star performer, something like this ought to do it:

DELETE FROM file a 
WHERE RRN(a) NOT IN 
( 
  SELECT keeper 
  FROM 
    ( 
      SELECT MIN(RRN(b)) as keeper, 
             Account_Number, Order_Date, Sequence_Number
      FROM file b                                          
      GROUP BY Account_Number, Order_Date, Sequence_Number
    ) t1
)

HTH

Elvis

-----Original Message-----
Subject: SQL Delete question

Greetings!

I have a file that, among other fields, has Account_Number, Date, Sequence
number.

In the file there are multiple occurrences of those three fields.  Example

Account_Number     Date    Sequence_number
            1             20061206           01
            1             20061206           01
            1             20061206           01
            1             20061206           02
            1             20061206           02
            1             20061206           02
            1             20061206           03
            1             20061206           03
            1             20061206           03
            1             20061206           04
            1             20061206           04
            1             20061206           04

What I am looking for is a way to delete all but one occurrence of the
repeated data to get a file that looks like;

Account_Number     Date    Sequence_number
            1             20061206           01
            1             20061206           02
            1             20061206           03
            1             20061206           04


Is this something I can do with SQL?  If so, what would the SQL statement
look like?

Jim