Reputation: 145
I have table with dept,user and so on, I need to find the number of count of user that belongs to different combinations of the dept.
Lets consider I've a table like this:
dept user
1 33
1 33
1 45
2 11
2 12
3 33
3 15
Then I've to find the uniq user and dept combination: something like this: select distinct dept,user from x;
Which will give me result like :
Dept user
1 33
1 45
2 11
2 12
3 33
3 15
which actually removes the duplicates of the combination: And here's the thing which i need to do :
My output should look like this:
dep_1_1 dep_1_2 dep_1_3 dep_2_2 dep_2_1 dep_2_3 Dep_3_1 Dep_3_2 Dep_3_3
2 0 1 2 0 0 1 0 2
So, Basically I need to find the count of common users between all the combinations of departments
Thanks for the help
Upvotes: 0
Views: 50
Reputation: 60462
You can get a row for each department combination using a self-join of your Distinct Select:
with cte as
(
select distinct dept,user from x
)
select t1.dept, t2.dept, count(*)
from cte a st1 join cte as t2
on t1.user = t2.user -- same user
and t1.dept < t2.dept -- different department
group by t1.dept, t2.dept
order by t1.dept, t2.dept
Upvotes: 1