Home/ Questions/Q 8458495
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-10T13:05:47+00:00

I have some what of a challenging delete statement that I need help with.

  • 0

I have some what of a challenging delete statement that I need help with. I’m trying to delete multiple duplicate records with multiple joins on the same table.

Here is an example of my data

versionid(PK) FileID(FK) Version DeleteDate DeleteIndicator

   1              1         1      12/01/2003        1
   2              1         1      12/02/2003        1 
   3              1         1         null           0
   4              2         2      01/02/2004        1
   5              2         2      01/03/2005        1
   6              2         2      01/03/2006        1

The data I need to delete is all duplicate data matched by the FileId and the Version where DeleteDate is less than 04/01/2011 and the DeleteIndicator is = 1. But, I need to keep the highest VersionID out of the duplicate data that has the same FileId and Version

After the delete I would have this remaining:

versionid(PK) FileID(FK) Version DeleteDate DeleteIndicator

  2              1         1      12/02/2003        1 
  3              1         1         null           0
  6              2         2      01/03/2006        1

I have a select w/ multiple joins that would give me the above records. I just don’t know how to turn that into a delete statement. Here is my select statement.

SELECT  t.VersionID ,
    t.FileID ,
    t.version ,
    COUNT(*) ,
    t.DeleteDate ,
    t.DeleteIndicator
FROM    tblFileVersions t
    JOIN ( SELECT   VersionID ,
                    FileID ,
                    MAX(VersionID) AS MaxVersion ,
                    DeleteDate ,
                    DeleteIndicator
           FROM     tblFileVersions
           GROUP BY VersionID ,
                    FileID ,
                    DeleteDate ,
                    DeleteIndicator
         ) x ON t.FileID = x.FileID
WHERE   t.DeleteDate < '2011/04/01'
    AND t.DeleteIndicator = 1
    AND t.VersionID < MaxVersion
GROUP BY t.VersionID ,
    t.FileID ,
    t.version ,
    t.DeleteDate ,
    t.DeleteIndicator ,
    Version
HAVING  COUNT(*) > 1

All I need to do is take what I have and make it a delete statement. Sorry if my text format is all screwed up this is the first time I have posted on here. Any help would be great thanks for your time.

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    Simpliest way to simple do that:

    DELETE FROM tblFileVersions 
WHERE VersionID NOT IN (SELECT  t.VersionID
FROM    tblFileVersions t
    JOIN ( SELECT   VersionID ,
                    FileID ,
                    MAX(VersionID) AS MaxVersion ,
                    DeleteDate ,
                    DeleteIndicator
           FROM     tblFileVersions
           GROUP BY VersionID ,
                    FileID ,
                    DeleteDate ,
                    DeleteIndicator
         ) x ON t.FileID = x.FileID
WHERE   t.DeleteDate < '2011/04/01'
    AND t.DeleteIndicator = 1
    AND t.VersionID < MaxVersion
GROUP BY t.VersionID ,
    t.FileID ,
    t.version ,
    t.DeleteDate ,
    t.DeleteIndicator ,
    Version
HAVING  COUNT(*) > 1)
    • 0