Question

I have this table

CREATE TABLE T
(No1 INT,
 No2 INT
)

INSERT INTO T (No1, No2) VALUES (1, 2)

INSERT INTO T (No1, No2) VALUES (2, 1)

INSERT INTO T (No1, No2) VALUES (3, 4)

INSERT INTO T (No1, No2) VALUES (4, 3)

INSERT INTO T (No1, No2) VALUES (5, 6)

INSERT INTO T (No1, No2) VALUES (6, 5)

INSERT INTO T (No1, No2) VALUES (1, 6)

INSERT INTO T (No1, No2) VALUES (6, 1)

---

    No1 No2
    1   2 
    2   1
    3   4
    4   3
    5   6
    6   5
    1   6
    6   1

What I want is to eliminate rows that have the same value but in the other field. For me No1=1 No2=2 and No1=2 No2=1 is the same thing. At the end I want this as result.

No1 No2
1   2
3   4
5   6
1   6

I tried a lot of things and I did a lot of search but I find nothing. The only way I found is with a Cursor. As I play with millions of records, it is very slow. Is there a way to do this with a query?

Thank you.

Answer 1

select t1.*
from MyTable t1
left outer join MyTable t2 on t1.No1 = t2.No2 and t1.No2 = t2.No1
where t2.No2 is null
    or t1.No1 <= t2.No1

SQL Fiddle Example