Reputation: 861
CUSTOMER(ID,CASE_ID,NAME,STATE)
1,100,Alex,NY
2,100,Alex,VA
3,100,Alex,CT
4,100,Tom,PA
5,102,Peter,MO
6,103,Dave,TN
.
.
.
How to write a query to pull 2nd smallest (min) id (if present) for every group of case_id
Upvotes: 1
Views: 131
Reputation: 116458
You can use a windowing function:
with cte as (
select ID, CASE_ID, ROW_NUMBER() over (partition by CASE_ID order by ID) rn
from CUSTOMER
)
select ID, CASE_ID
from cte
where rn = 2
Or you can use an exists clause to remove the first row (i.e. get the minimum value where there is a row with a lower value):
select MIN(ID) ID, CASE_ID
from CUSTOMER c
where exists (select 1 from CUSTOMER c2 where c2.ID < c.ID and c2.CASE_ID = c.CASE_ID)
group by CASE_ID
Or, written another way:
select MIN(ID) ID, CASE_ID
from CUSTOMER c
where c.ID > (select MIN(ID) from CUSTOMER c2 where c2.CASE_ID = c.CASE_ID)
group by CASE_ID
Upvotes: 1
Reputation: 18629
Please try:
SELECT
ID,
CASE_ID
FROM
(
SELECT
*,
ROW_NUMBER() OVER(PARTITION BY CASE_ID ORDER BY ID) Rn
FROM CUSTOMER
)x
WHERE Rn=2
Upvotes: 3