0

How to delete duplicate Records in SQL

24th Mar 2017, 5:38 AM
Akwin Lopez
Akwin Lopez - avatar
1 Réponse
+ 1
DELETE DUPLICATE ROWS FROM A TABLE IN SQL SERVER 2005 CREATE TABLE EMPLOYEE ( id INT IDENTITY(1,1) PRIMARY KEY NOT NULL, EMPNAME VARCHAR(100), SALARY DECIMAL ) SELECT * FROM EMPLOYEE INSERT INTO EMPLOYEE VALUES ( ‘Pawan’ , 100000 ) INSERT INTO EMPLOYEE VALUES ( ‘Pawan’ , 80000 ) INSERT INTO EMPLOYEE VALUES ( ‘Pawan’ , 40000 ) INSERT INTO EMPLOYEE VALUES ( ‘Rahul’ , 100000 ) INSERT INTO EMPLOYEE VALUES ( ‘Pawan’ , 100000 ) INSERT INTO EMPLOYEE VALUES ( ‘Gauri’ , 100000 ) INSERT INTO EMPLOYEE VALUES ( ‘Gauri’ , 100000 ) INSERT INTO EMPLOYEE VALUES ( ‘Rahul’ , 100000 ) INSERT INTO EMPLOYEE VALUES ( ‘Deepa’ , 100000 ) –Query to delete duplicate names in the above table DELETE from EMPLOYEE where id NOT IN (SELECT MAX(id) FROM EMPLOYEE GROUP BY EmpName) –Check with SELECT * FROM EMPLOYEE Note : If you do not have identity column then create one and remove that column after your deletion.
24th Mar 2017, 5:39 AM
Akwin Lopez
Akwin Lopez - avatar