如果所选列 postgresql sql 的列中的数据相同，则 Null 行值

Question

I have a table like this:

  cms_code  | product 
------------+---------
 55381XA384 | MS
 55381XA384 | MS
 55381XA384 | HSD
 55381XA384 | HSD
 55381XA382 | HSD
 55381XA382 | XP
 55381XA382 | MS
 55381XA382 | HSD
 55381XA382 | HSD

Now, I need a result like this:

  cms_code  | product 
------------+---------
 55381XA384 | MS
 NULL       | MS
 NULL       | HSD
 NULL       | HSD
 55381XA382 | HSD
 NULL       | XP
 NULL       | MS
 NULL       | HSD
 NULL       | HSD

Answer 1

you can use window functions to number the rows & set subsequent values to null , keeping the cms_code only for row number 1.

select 
  case 
    when row_number() over (partition by cms_code) = 1 then cms_code 
    else null 
  end cms_code
, product_code
from my_table
order by cms_code

Answer 2

Your results depend on the ordering of the rows. SQL tables represent unordered sets, so you cannot do what you want with the data you have.

Assuming you have a column that specifies the ordering, you could do:

select (case when row_number() over (partition by cms_code order by <ordering col< = 1
             then cms_code
        end) as cms_code, product_code
from t
order by cms_code, <ordering col>;

Without an ordering column, you can also do:

with cte as (
      select t.*, row_number() over (partition by cms_code order by cms_code) as seqnum
      from t
     )
select (case when seqnum = 1 then cms_code end) as cms_code,
       product_code
from t
order by cms_code, seqnum;

In both cases, you need the order by in the outer query to guarantee that the result set is ordered so the "first" row has the cms_code .

Answer 3

I'm assuming you want null for all the values of cms_code which is same after the first value. Correct me if I'm wrong Below query can be used for the same:

SELECT 
  CASE
    WHEN ROW_NUMBER() OVER(PARTITION BY cms_code order by product) = 1 
    THEN NULL 
    ELSE cms_code 
  END AS cms_code,product 
FROM table;