I have two tables – first table is “projects”:
| p_id | p_name | 
|---|---|
| 1 | test1 | 
| 2 | test2 | 
| 3 | test2 | 
| 4 | test3 | 
p_name is not unique
The second table is “employee”:
| e_id | jan | feb | mar | apr | may | 
|---|---|---|---|---|---|
| 1 | 2 | 2 | 3 | 3 | 4 | 
| 2 | 1 | 1 | 3 | 3 | 2 | 
| 3 | 1 | 3 | 2 | 3 | 3 | 
| 4 | 4 | 3 | 2 | 3 | 4 | 
I’d like to count how many p_name in employee depending on name
What I do:
select 
    p.id, p.name, 
    count(e1.jan) as e1, 
    count(e2.feb) as e2, 
    count(e3.mar) as e3  
from 
    projects p 
left join 
    employee e1 on (e1.jan = p.id) 
left join 
    employee e2 on (e2.feb = p.id) 
left join 
    employee e3 on (e3.mar = p.id) 
where 
    p.name = 'test2' 
group by 
    p.name;
But it doesn’t work. It doesn’t return the expected numbers.
Advertisement
Answer
I completely agree with the comment of @sTTu – the chosen design is not scalable and will run into problems as soon as a “new year” starts.
Purely out of curiousity I tried to come up with a solution nonetheless which would be the following:
(SQL-Server – but should also work in MySql)
SELECT name, SUM(jan) Jan, SUM(feb) feb, SUM(mar) mar FROM (
  SELECT id, name, 
    (SELECT count(*) FROM employee WHERE jan=id) jan,
    (SELECT count(*) FROM employee WHERE feb=id) feb,
    (SELECT count(*) FROM employee WHERE mar=id) mar
  from projects ) tbl
group by name;
Here is a working snippet: https://rextester.com/TBWZA82894
Result:
| name | Jan | feb | mar | 
|---|---|---|---|
| test1 | 2 | 1 | 0 | 
| test2 | 1 | 3 | 4 | 
| test3 | 1 | 0 | 0 | 
Or, in an even simpler way you can do it like this:
SELECT name, COUNT(CASE WHEN jan=id THEN 1 END) jan, COUNT(CASE WHEN feb=id THEN 1 END) feb, COUNT(CASE WHEN mar=id THEN 1 END) mar FROM employee INNER JOIN projects ON id>0 GROUP BY name
see the demo here: https://rextester.com/DEUUE59264