Replace row value with empty string if duplicate

问题

is it possible to replace row value with empty string if duplicate value found?

For example

SELECT ProductCode, Color FROM Product

--------------------
ProductCode | Color
--------------------
   00A0B    |  Red
   00A0B    |  Blue
   00A0C    |  Red
   00A0C    |  Black
   00A0C    |  White
--------------------

--------------------
ProductCode | Color
--------------------
   00A0B    |  Red
            |  Blue
   00A0C    |  Red
            |  Black
            |  White
--------------------

I'm using SQL Server 2012.

回答1:

Often, this type of transformation is better done at the application layer, because the result-set is not "SQL-ish". That is, the ordering is important for understanding the rows.

But, you can do this as:

select (case when row_number() over (partition by ProductCode order by (select NULL)) = 1
             then ProductCode
        end) as ProductCode
       Color
from Product
order by ProductCode;

回答2:

Use ROW_NUMBER and CASE:

WITH Cte AS(
    SELECT *,
        Rn = ROW_NUMBER() OVER(PARTITION BY ProductCode ORDER BY (SELECT NULL))
    FROM Product
)
SELECT
    ProductCode = CASE WHEN Rn = 1 THEN c.ProductCode ELSE '' END,
    Color 
FROM Cte c
ORDER BY c.ProductCode

回答3:

ROW_NUMBER() helps to find duplicate record : 

 SELECT Productcode=( CASE
                           WHEN rn > 1 THEN ''
                           ELSE Productcode
                         END ),
           color
    FROM   (SELECT Row_number()
                     OVER (
                       partition BY productcode
                       ORDER BY productcode) AS rn,
                   *
            FROM   table)a

来源：https://stackoverflow.com/questions/33138663/replace-row-value-with-empty-string-if-duplicate

标签

sql

sql-server

sql-server-2012