Reputation: 23
Given below table called table
+---+---+
| x | y |
+---+---+
| 1 | 2 |
| 1 | 5 |
| 5 | 2 |
| 5 | 1 |
+---+---+
I want to have sql query for the following results
+----+-------------+
| id | count_total |
+----+-------------+
| 1 | 3 |
| 2 | 2 |
| 5 | 3 |
+----+-------------+
Note: I was able to count separately the rows per id but I could not get the sum for the same id. so I want to combine or get sum of below queries in a single query.
SELECT x, count(*) as total_x FROM table GROUP BY x
SELECT y, count(*) as total_y FROM table GROUP BY y
Upvotes: 2
Views: 6781
Reputation: 94894
Probably the easiest way is to select all x and all y and then aggregate over them.
select id, count(*) as count_total
from (select x as id from mytable union all select y from mytable) ids
group by id
order by id;
Upvotes: 0
Reputation: 1
declare @table table ( x int, y int ) insert into @table select 1,2 union all select 1,5 union all select 5,2 union all select 5,1 select x,SUM(a) from ( select x,COUNT() as a from @table group by x union all select y,COUNT() as a from @table group by y ) a group by x
Upvotes: 0
Reputation: 1787
Try:
SELECT
A.ID, SUM(A.COUNTS) AS COUNT_TOTAL
FROM
(
SELECT X AS ID, COUNT(*) AS COUNTS FROM TABLE1 GROUP BY X
UNION ALL
SELECT Y AS ID, COUNT(*) AS COUNTS FROM TABLE1 GROUP BY Y
) A
GROUP BY A.ID
ORDER BY A.ID;
Upvotes: 4
Reputation: 23756
You can use union
to bring them together, I did not try it actually but it should work fine. if it did not work please leave a comment and I would be happy to help and edit my answer.
select u.x, sum(u.total)
from
(
(SELECT x as x, count(*) as total FROM table GROUP BY x)
union all
(SELECT y as x, count(*) as total FROM table GROUP BY y)
) as u
group by u.x
Upvotes: 0