Solved

Delete duplicate records in SQL database using stored procedure

Posted on 2007-12-04
12
1,703 Views
Last Modified: 2008-09-11
I have a SQL server 2000 database that has over a million records in it.  I need to create a stored procedure that i can run that we check for which records have duplicates and then delete all of the duplicates except for one.  The criteria for the duplicates is based on five different columns.  I am hoping somebody can give me some guideance on how to do this.  
0
Comment
Question by:Mike_Stevens
12 Comments
 
LVL 17

Expert Comment

by:Daniel Reynolds
ID: 20405242
Can you provide the structure of the table? Is there a unique key?
0
 

Author Comment

by:Mike_Stevens
ID: 20405276

tblStatus
Column: RecordNumber -  Unique Key
Column: IDNum
Column: Code
Column: Date
Column: Time
Column: State

All of the columns except RecordNumber need to be used in criteria.  Sample below

RecordNumber  IDNum  Code  Date         Time   State
------------------  ---------  ------  -------        ------   --------
100                   12345     A     12122007  1630    CA
101                   12345     A     12122007  1630    CA
102                   12345     A     12122007  1630    CA
103                   12345     A     12122007  1630    CA
104                   12345     A     12122007  1630    CA

Using the sample data above, procedure needs to delete all but one of the records.

0
 
LVL 3

Expert Comment

by:randomjames
ID: 20405356
I would create a new table with the unique index on those fields, export the data from your old table into this new one (which will only take one of each duplicate because of the index), then clear out your old table and move the records from your new table back to the old table.
0
Optimizing Cloud Backup for Low Bandwidth

With cloud storage prices going down a growing number of SMBs start to use it for backup storage. Unfortunately, business data volume rarely fits the average Internet speed. This article provides an overview of main Internet speed challenges and reveals backup best practices.

 
LVL 17

Accepted Solution

by:
Daniel Reynolds earned 500 total points
ID: 20405880
Here is some code you can run iteratively until the rows affected = 0

SET ROWCOUNT 1
DELETE  dbo.tblStatus
--SELECT top 1 * 
FROM dbo.tblStatus
WHERE Exists (Select Count(*), IDNum, Code, [Date], [Time], [State]
                              from dbo.tblStatus
                             Group By IDNum, Code, [Date], [Time], [State]
                             Having Count(*) > 1)

Open in new window

0
 
LVL 17

Expert Comment

by:Daniel Reynolds
ID: 20405903
And here is a link with additional information.
http://support.microsoft.com/kb/139444
and one that shows how to do it with a little bit more programming, but all at once.
http://www.sql-server-performance.com/articles/dev/dv_delete_duplicates_p1.aspx

0
 
LVL 11

Expert Comment

by:yuching
ID: 20408127
Code provided by xDJR1875: will delete all record which have duplicate data but Mike_Stevens intended to leave a row for the duplicate data.

You can try as below, it will only left the max record number for the duplicates data.

DELETE  FROM tblStatus
WHERE  NOT EXISTS (
        SELECT MaxRecordNumber, IDNum, Code, Date, Time, State
        FROM
        (
              SELECT Max(RecordNumber) AS MaxRecordNumber, IDNum, Code, Date, Time, State
              FROM  dbo.tblStatus
              GROUP BY IDNum, Code, Date, Time, State
          ) A
         WHERE A.MaxRecordNumber = tblStatus.RecordNumber AND
                     A.IDNum = tblStatus.IDNum AND A.Code = tblStatus.Code AND
                     A.Date = tblStatus.Date AND A.Time = tblStatus.Time AND
                     A.State = tblStatus.State
)
0
 
LVL 17

Expert Comment

by:Daniel Reynolds
ID: 20408152
Actually, with the SET ROWCOUNT = 1
The code does not delete all duplicate records.
0
 
LVL 11

Expert Comment

by:yuching
ID: 20408626
Sorry, my mistake, but for the example of data, the duplicate rows can be more than 2, set rowcount =1  will only delete one of the records which having duplicate.
0
 
LVL 17

Expert Comment

by:Daniel Reynolds
ID: 20409336
You are correct that is why I suggest Mike run it iteratively until rows affected = 0.

It is not the most elegant of solutions, but maybe one of  the simplest.
0
 

Expert Comment

by:yordan_georgiev
ID: 21502162
DELETE
    FROM    MyTable
    WHERE    ID NOT IN
    (
    SELECT    MAX(ID)
        FROM        MyTable
        GROUP BY    DuplicatevalueColumn1, DuplicateValueColumn2,
                DuplicateValueColumn2)

Reference:Pinal Dave
http://blog.sqlauthority.com/2007/03/01/sql-server-delete-duplicate-records-rows/
0

Featured Post

NAS Cloud Backup Strategies

This article explains backup scenarios when using network storage. We review the so-called “3-2-1 strategy” and summarize the methods you can use to send NAS data to the cloud

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Suggested Solutions

Everyone has problem when going to load data into Data warehouse (EDW). They all need to confirm that data quality is good but they don't no how to proceed. Microsoft has provided new task within SSIS 2008 called "Data Profiler Task". It solve th…
Ever wondered why sometimes your SQL Server is slow or unresponsive with connections spiking up but by the time you go in, all is well? The following article will show you how to install and configure a SQL job that will send you email alerts includ…
This video shows, step by step, how to configure Oracle Heterogeneous Services via the Generic Gateway Agent in order to make a connection from an Oracle session and access a remote SQL Server database table.
This video shows how to set up a shell script to accept a positional parameter when called, pass that to a SQL script, accept the output from the statement back and then manipulate it in the Shell.

786 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question