Question

I have 2 simple tables that I would like to perform an INNER JOIN with, but the problem is that I’m getting duplicated (for the columns str1 and str2) results:

CREATE TABLE #A (Id INT, str1 nvarchar(50), str2 nvarchar(50))
insert into #A  values (1, 'a', 'b')
insert into #A  values (2, 'a', 'b')

CREATE TABLE #B (Id INT, str1 nvarchar(50), str2 nvarchar(50))
insert into #B values (7, 'a', 'b')
insert into #B  values (8, 'a', 'b')

select * from #A a
INNER JOIN #B b ON a.str1 = b.str1 AND a.str2 = b.str2  

It gave me 4 records when I really wanted 2.

What I got:
id | str1 | str2| id | str1 | str2
1 | a | b | 7 | a | b
2 | a | b | 7 | a | b
1 | a | b | 8 | a | b
2 | a | b | 8 | a | b

What I really wanted:
1 a | b | 7 | a | b
2 a | b | 8 | a | b

Can anyone help? I know this is achievable using a cursor and loop, but I’d like to avoid it and only use some type of JOIN if possible.

Answer 1

SELECT 
    a.id AS a_id, a.str1 AS a_str1, a.str2 AS a_str2, 
    b.id AS b_id, b.str1 AS b_str1, b.str2 AS b_str2
FROM 
    ( SELECT *
           , ROW_NUMBER() OVER (PARTITION BY str1, str2 ORDER BY id) AS rn
      FROM #A 
    ) a
  INNER JOIN 
    ( SELECT *
           , ROW_NUMBER() OVER (PARTITION BY str1, str2 ORDER BY id) AS rn
      FROM #B 
    ) b 
    ON  a.str1 = b.str1 
    AND a.str2 = b.str2 
    AND a.rn = b.rn ;

If you have more rows in one or the other tables for the same (str1, str2) combination, you can choose which ones will be returned by changing INNER join to either LEFT, RIGHT or FULL join.