|
Brian, I'm not sure I fully understood you (especially DISTINCT part - SQL keyword?) so make sure you test this on a dummy/copied file first. Alternatively, just run the SELECT part on its own to verify that projected rows are ones you want to preserve. WITH preserve AS ( SELECT SCACCODE,TRLRNUM,ASNNUM,ORDERTYP,ORDERNUM,MIN(FDATE) AS FDATE FROM SSW85 GROUP BY SCACCODE, TRLRNUM, ASNNUM, ORDERTYP, ORDERNUM) DELETE FROM SSW85 A, preserve P WHERE A.SCACODE = P.SCACODE AND A.TRLRNUM = P.TRLRNUM AND A.ASNNUM = P.ASNNUM AND A.ORDERTYP = P.ORDERTYP AND A.ORDERNUM = P.ORDERNUM AND A.FDATE <> P.FDATE Elvis -----Original Message----- Subject: SQL Ideas Please Hi All, (This is a two-part SQL question) Can someone tell me if there is an easy way to delete records from a table where there are only four data elements on which I wish to base the delete routine? For example, I have a table that has numerous records in it. I want to delete all records other than those that have a distinct SCACCODE, TRLRNUM, ASNNUM, ORDERTYP and ORDERNUM. However, I'm not sure how to express this in SQL, but in pseudo-code, I'd like to do this: DELETE ALL RECORDS IN SSW85 EXECPT THOSE RECORDS THAT HAVE A DISTINCT (SCACCODE, TRLRNUM, ASNNUM, ORDERTYP, ORDERNUM). My second SQL statement needs to pass over the records again and only keep records that have the minimum number for a specific column. For example, if I have these records: A B C D E F ABCD 1000 10039210 2 105049 206052971 ABCD 1000 10039210 2 105049 206052951 ABCD 1000 10039210 2 105049 206052991 ABCD 1000 10039210 2 105049 206053001 I only want to keep one of these records that has the minimum value in column F. So if I ran a min(COLUMN_F) on this, I would only get back the second record (since it is the lowest number for Column F). Pseudo-code wise, I'd need something like: DELETE ALL RECORDS IN SSW85 EXCEPT EACH DISTINCT (SCACCODE, TRLRNUM, ASNNUM, ORDERTYP, ORDERNUM) WITH THE LOWEST KDLOTSTR. Is there an easy way to do both of these statements (or combine them into one statement)? Thanks! Brian.
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.