Updating in duplicate records

  • IN MY BELOW sample RECORDS I have to update f_authorized=0 for the record which f_date_Stamp column contains maximum value.
    All the columns having duplicate values except F_DATE_STAMP column.

    F_PRODUCT    F_FORMAT    F_SUBFORMAT    F_LANGUAGE    F_PLANT    F_AUTHORIZED    F_DATE_STAMP
    000309      MTR        WPR1       EN      TEST      -1       2013-07-11 14:18:24.000
    000309      MTR        WPR1       EN      TEST      -1       2013-07-10 14:18:22.000

    sample output

    F_PRODUCT    F_FORMAT    F_SUBFORMAT    F_LANGUAGE    F_PLANT    F_AUTHORIZED    F_DATE_STAMP
    000309      MTR        WPR1       EN      TEST      0       2013-07-11 14:18:24.000
    000309      MTR        WPR1      EN      TEST      -1       2013-07-10 14:18:22.000
  • If you provide the sample data in a consumable format (ie, one which someone can cut & paste directly into SSMS), you'll soon get a coded solution (probably using a CTE and the ROW_NUMBER() function).

    The absence of evidence is not evidence of absence
    - Martin Rees
    The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
    - Phil Parkin

  • Can try something like ROW_NUMBER.

    ROW_NUMBER() OVER(PARTITION BY DuplicatedFields ORDER BY F_DATE_STAMP DESC) AS Row#

    Or you can join the data ontop of each other with MAX().

    Then learn if it's possible to do an UPDATE with a JOIN.

  • jkramprakash - Wednesday, September 12, 2018 6:21 AM

    IN MY BELOW sample RECORDS I have to update f_authorized=0 for the record which f_date_Stamp column contains maximum value.
    All the columns having duplicate values except F_DATE_STAMP column.

    F_PRODUCT    F_FORMAT    F_SUBFORMAT    F_LANGUAGE    F_PLANT    F_AUTHORIZED    F_DATE_STAMP
    000309      MTR        WPR1       EN      TEST      -1      2013-07-11 14:18:24.000
    000309      MTR        WPR1       EN      TEST      -1      2013-07-10 14:18:22.000

    sample output

    F_PRODUCT    F_FORMAT    F_SUBFORMAT    F_LANGUAGE    F_PLANT    F_AUTHORIZED    F_DATE_STAMP
    000309      MTR        WPR1       EN      TEST      0       2013-07-11 14:18:24.000
    000309      MTR        WPR1      EN      TEST      -1       2013-07-10 14:18:22.000

    And how did you select the two rows that you are showing?   That SELECT statement can have a ROW_NUMBER() function added to it that uses OVER (ORDER BY F_DATE_STAMP DESC) as a CTE, and the UPDATE references the CTE and applies a WHERE clause that specifies the column that is the row number = 1.

    Steve (aka sgmunson) 🙂 🙂 🙂
    Rent Servers for Income (picks and shovels strategy)

  • Here's one way to do it

    -- Use a common table expression to find the max values for records and records that are duplicated
    -- This won't work if there are duplicate timestamps for the max record.
    WITH CTE AS (SELECT F_PRODUCT, F_FORMAT, F_SUBFORMAT, F_LANGUAGE, F_PLANT, F_AUTHORIZED, MAX(F_DATE_STAMP) AS STAMP
                             FROM yourTable
                             GROUP BY F_PRODUCT, F_FORMAT, F_SUBFORMAT, F_LANGUAGE, F_PLANT, F_AUTHORIZED
                             HAVING COUNT(F_PRODUCT) > 1)
    UPDATE T SET F_AUTHORIZED = 0
    FROM CTE INNER JOIN yourTable T ON
      CTE.F_PRODUCT = T.F_PRODUCT
      AND CTE.F_FORMAT = T.F_FORMAT
      AND CTE.F_SUBFORMAT = T.F_SUBFORMAT
      AND CTE.F_LANGUAGE = T.F_LANGUAGE
      AND CTE.F_PLANT = T.F_PLANT
      AND CTE.STAMP = T.F_DATE_STAMP

  • souLTower - Wednesday, September 12, 2018 10:01 AM

    Here's one way to do it

    -- Use a common table expression to find the max values for records and records that are duplicated
    -- This won't work if there are duplicate timestamps for the max record.
    WITH CTE AS (SELECT F_PRODUCT, F_FORMAT, F_SUBFORMAT, F_LANGUAGE, F_PLANT, F_AUTHORIZED, MAX(F_DATE_STAMP) AS STAMP
                             FROM yourTable
                             GROUP BY F_PRODUCT, F_FORMAT, F_SUBFORMAT, F_LANGUAGE, F_PLANT, F_AUTHORIZED
                             HAVING COUNT(F_PRODUCT) > 1)
    UPDATE T SET F_AUTHORIZED = 0
    FROM CTE INNER JOIN yourTable T ON
      CTE.F_PRODUCT = T.F_PRODUCT
      AND CTE.F_FORMAT = T.F_FORMAT
      AND CTE.F_SUBFORMAT = T.F_SUBFORMAT
      AND CTE.F_LANGUAGE = T.F_LANGUAGE
      AND CTE.F_PLANT = T.F_PLANT
      AND CTE.STAMP = T.F_DATE_STAMP

    This requires you to read the table twice: once to figure out the max values, and once to update the table.  The method that Steve alluded to only requires one read of the table.

    Drew

    J. Drew Allen
    Business Intelligence Analyst
    Philadelphia, PA

  • I agree with you J. Drew Allen.  It's not the MOST efficient way to do it but it demonstrates the mechanics of how to do it.  The approach you mention is more advanced and I felt that posting it hides the details.  I was trying to use a teaching approach.

  • souLTower - Thursday, September 13, 2018 6:12 AM

    I agree with you J. Drew Allen.  It's not the MOST efficient way to do it but it demonstrates the mechanics of how to do it.  The approach you mention is more advanced and I felt that posting it hides the details.  I was trying to use a teaching approach.

    Thank you.This table contains ten thousand records only.so i will use this method.

Viewing 8 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic. Login to reply