Reputation: 187
Database structure
Table 'applicants'
id org_id team_id
1 1 1
Table 'teams'
id name
1 Test
Table 'teams_members'
id team_id user_id
1 1 1
2 1 2
Table 'users_playeraccounts'
id user_id summoner_id rank_solo
1 1 1 5
2 1 2 8
3 2 3 7
select sum(rank_solo) as rank_sum,
max(rank_solo) as highest_rank,
count(tt.id) as members,
t.name,
o.team_id
from applicants o
join teams t on o.team_id = t.id
join teams_members tt on t.id = tt.team_id
join users_playeraccounts p on tt.user_id = p.user_id
where org_id = :org
group by team_id
This offcourse gives me a result like
rank_sum highest_rank members name team_id
20 8 3 Test 1
Is there a way for me to get both the count of members with their playeraccounts aka
If 1 user has 2 it'll be 2
And also a way for me to keep it as 1 so it literally just counts the rows found in teams_members neglecting the entries in users_playeraccounts?
I want to receive both 2 and 3 as a result of my query.
Upvotes: 0
Views: 24
Reputation: 5668
You want to count the distinct number of entries in tt.id
, so you can do that like this:
SELECT ... COUNT(DISTINCT tt.id) AS distinct_members ...
Rather than giving you a count of every row that has a non-null tt.id
, you'll get a count of the number of unique values.
Upvotes: 1