Select columns from different tables without repeating rows - sql

Question

For this example tables:
Table A:
Col1 Col2
-----------
2015 A
2015 B
2015 C
Table B:
Col1 Col2
------------
2015 X
2015 Y
2015 Z
I want a query that returns:
Col1 Col2
------------
A X
B Y
C Z
I have tried something like this:
SELECT TA.COL2, TB.COL2
FROM
(SELECT * FROM TABLE_A WHERE COL1=2015) TA,
(SELECT * FROM TABLE_B WHERE COL1=2015) TB,
But I'm getting duplicated results
Col1 Col2
-----------
A X
A Y
A Z
B X
B Y
B Z
C X
C Y
C Z

Answer 1

A way to do it is to use the row number:
SELECT TA.COL2, TB.COL2
FROM
(SELECT TABLE_A.COL2, ROWNUM AS R1 FROM TABLE_A WHERE COL1=2015) TA,
(SELECT TABLE_B.COL2, ROWNUM AS R2 FROM TABLE_B WHERE COL1=2015) TB,
WHERE T1.R1 = T2.R2

Answer 2

In a SELECT statement, include DISTINCT or DISTINCTROW keyword after the SELECT clause.
For More Details:
http://www.geeksengine.com/database/basic-select/eliminate-duplicate-rows.php

Answer 3

Not sure why you want this but here is my method for doing it.
SELECT
TA.COL2 AS Col1,
TB.COL2 AS Col2
FROM
(SELECT COL1, COL2, ROW_NUMBER() OVER (PARTITION BY COL1 ORDER BY COL2) AS Seq FROM TABLE_A) TA
JOIN
(SELECT COL1, COL2, ROW_NUMBER() OVER (PARTITION BY COL1 ORDER BY COL2) AS Seq FROM TABLE_B) TB
ON TA.COL1 = TB.COL1
AND TA.Seq = TB.Seq

Answer 4

A union all/group by approach may be what you want:
SELECT MAX(COL2_a) as COL2_a, MAX(COL2_B) as COL2_B
FROM ((SELECT COL2 as COL2_A, NULL as COL2_B,
ROW_NUMBER() OVER (ORDER BY COL2) as seqnum
FROM TABLE_A
WHERE COL1 = 2015
) UNION ALL
(SELECT NULL, COL2, ROW_NUMBER() OVER (ORDER BY COL2) as seqnum
FROM TABLE_B
WHERE COL1 = 2015
) UNION ALL
) t
GROUP BY seqnum;
Alternatively, use a FULL JOIN:
SELECT a.COL2 as COL2_a, b.COL2 as COL2_B
FROM (SELECT a.*,
ROW_NUMBER() OVER (ORDER BY COL2) as seqnum
FROM TABLE_A a
WHERE COL1 = 2015
) a FULL JOIN
(SELECT b.*,
ROW_NUMBER() OVER (ORDER BY COL2) as seqnum
FROM TABLE_B b
WHERE COL1 = 2015
) b
ON a.seqnum = b.seqnum;
Both these methods returns all values, if one table has more values than the other.