How to select maximum value of two identical columns in same table? - sql

Question

id1 id2 name
1 1 2 a
2 3 4 b
3 5 6 c
4 7 8 d
5 9 10 e
select id1, id2, name
from Emp3
where id2 in (select MAX(id2) from Emp3)
How can I print only the maximum number?

Answer 1

USe
select id2
from Emp3
where id2 in (select MAX(id2) from Emp3)
This will print only 10
if you want the maximum among those 2 columns then use
SELECT
CASE
WHEN MAX(id1) >= MAX(id2) THEN MAX(id1)
WHEN MAX(id2) >= MAX(id1) THEN MAX(id2)
END AS MaxValue
FROM Emp3

Answer 2

Use TOP and ORDER BY to get the result
SELECT TOP 1 ID2 FROM Emp3 ORDER BY ID2 DESC

Answer 3

select max(id) from (
select max(id1) as id from Emp3
union
select max(id2) as id from Emp3
)

Answer 4

DECLARE #T TABLE (ID1 INT ,ID2 INT ,NAME VARCHAR(80))
INSERT INTO #T VALUES (1,2,'NME1')
INSERT INTO #T VALUES (3,4,'NME2')
INSERT INTO #T VALUES (5,6,'NME3')
INSERT INTO #T VALUES (7,8,'NME4')
INSERT INTO #T VALUES (9,10,'NME5')
INSERT INTO #T VALUES (11,12,'NME6')
SELECT * FROM #T
SELECT MAX(ID1) ID FROM
(
SELECT ID1 FROM #T T1
UNION
SELECT ID2 FROM #T T2
)TT

Answer 5

If both id columns are indexed, use derived table with UNION ALL:
select max(id)
from
(
select max(id1) as id from Emp3
union all
select max(id2) from Emp3
)
If they are not indexed, use CASE:
select max(case when id1 > id2 then id1
when id2 > id1 then id2
else coalesce(id1,id2) end)
from Emp3
This answer uses COALESCE to handle NULLs (if there are any...) The table will be read only once. (With the UNION solution the table will be read twice, and you don't want to do that without any index!)
Older answers:
I guess you want the row with the highest id1/id2 value?
Return a row when no other row with higher id value exists:
select id1, id2, name
from Emp3 e1
where not exists (select 1 from Emp3 e2
where e2.id1 > e1.id1
or e2.id1 > e1.id2
or e2.id2 > e1.id1
or e2.id2 > e1.id2)
Will return both rows if there's a tie. (Two or more rows with same highest value.)
Alternative solution, use TOP, combined with ORDER BY with CASE to find each rows larger id value:
select TOP 1 id1, id2, name
from Emp3
order by case when id1 > id2 then id1 else id2 end desc
Alternative 3, a sub-query with UNION ALL to find largest id:
select TOP 1 id1, id2, name
from Emp3
where (select max(case when id1 > id2 then id1 else id2 end) from Emp3) in (id1,id2)