I want to delete duplicate rows in SQL. My query can delete duplicate rows for RN > 1
, but I want to delete max(RN)
instead:
WITH CTE AS(
SELECT [col1], [col2], [col3]
RN = ROW_NUMBER()OVER(PARTITION BY [col1], [col2] ORDER BY [col1], [col2])
FROM dbo.Table1
)
DELETE FROM CTE WHERE RN > 1
Just change the partition to the reverse of the current order, then delete just records with the first row number:
WITH CTE AS (
SELECT [col1], [col2], [col3]
RN = ROW_NUMBER() OVER (PARTITION BY [col1], [col2] ORDER BY [col1] DESC, [col2] DESC)
FROM dbo.Table1
)
DELETE FROM CTE WHERE RN = 1;