--ID should be primary key

​

--get duplicate records using RANK

SELECT E.ID, 

    E.firstname, 

    E.lastname, 

    E.country, 

    T.rank

FROM [SampleDB].[dbo].[Employee] E

  INNER JOIN

(

 SELECT *, 

        RANK() OVER(PARTITION BY firstname, 

                                 lastname, 

                                 country

        ORDER BY id) rank

 FROM [SampleDB].[dbo].[Employee]

) T ON E.ID = t.ID;

​

--delete duplications

DELETE E

    FROM [SampleDB].[dbo].[Employee] E

         INNER JOIN

    (

        SELECT *, 

               RANK() OVER(PARTITION BY firstname, 

                                        lastname, 

                                        country

               ORDER BY id) rank

        FROM [SampleDB].[dbo].[Employee]

    ) T ON E.ID = t.ID

    WHERE rank > 1;

SELECT DISTINCT col1,col2... FROM table_name where Condition;

WITH cte AS (

    SELECT 

        contact_id, 

        first_name, 

        last_name, 

        email, 

        ROW_NUMBER() OVER (

            PARTITION BY 

                first_name, 

                last_name, 

                email

            ORDER BY 

                first_name, 

                last_name, 

                email

        ) row_num

     FROM 

        sales.contacts

)

DELETE FROM cte

WHERE row_num > 1;

/*

Code language: SQL (Structured Query Language) (sql)

​

In this statement:

​

First, the CTE uses the ROW_NUMBER() function to find the duplicate rows 

specified by values in the first_name, last_name, and email columns.

​

Then, the DELETE statement deletes all the duplicate rows but keeps only one 

occurrence of each duplicate group.*/

DELETE FROM Customers WHERE ROWID(SELECT MAX (rowid) FROM Customers C WHERE CustomerNumber = C.CustomerNumber);

-- Oracle

DELETE FROM films

WHERE rowid NOT IN (

    SELECT min(rowid)

    FROM films

    GROUP BY title, uk_release_date

);

WITH cte AS (

    SELECT 

        contact_id, 

        first_name, 

        last_name, 

        email, 

        ROW_NUMBER() OVER (

            PARTITION BY 

                first_name, 

                last_name, 

                email

            ORDER BY 

                first_name, 

                last_name, 

                email

        ) row_num

     FROM 

        sales.contacts

)

DELETE FROM cte

WHERE row_num > 1;

DELETE FROM [SampleDB].[dbo].[Employee]

    WHERE ID NOT IN

    (

        SELECT MAX(ID) AS MaxRecordID

        FROM [SampleDB].[dbo].[Employee]

        GROUP BY [FirstName], 

                 [LastName], 

                 [Country]

    );

WITH cte AS (

    SELECT 

        contact_id, 

        first_name, 

        last_name, 

        email, 

        ROW_NUMBER() OVER (

            PARTITION BY 

                first_name, 

                last_name, 

                email

            ORDER BY 

                first_name, 

                last_name, 

                email

        ) row_num

     FROM 

        sales.contacts

)

DELETE FROM cte

WHERE row_num > 1;

Code language: SQL (Structured Query Language) (sql)

SELECT DISTINCT column1, column2, ...

FROM table_name;

SET ROWCOUNT 1

DELETE [dbo].[table] where ....

SET ROWCOUNT 0

SELECT s.siteName, s.siteIP, h.date

FROM sites s INNER JOIN

     (select h.*, row_number() over (partition by siteName order by date desc) as seqnum

      from history h

     ) h

    ON s.siteName = h.siteName and seqnum = 1

ORDER BY s.siteName, h.date