1) departmanet/job

select department_name, job_title, null, max(salary), sum(salary)
from employees e join jobs using (job_id) join departments using (department_id)
group by department_name, job_title

union all

select null, job_title, e.manager_id, max(salary), sum(salary)
from employees e join jobs using (job_id) join departments using (department_id)
group by job_title,e.manager_id

union all

select null, null, null, max(salary), sum(salary)
from employees e join jobs using (job_id) join departments using (department_id);


SELECT department_id, TO_CHAR(hire_date, 'yyyy'), SUM(salary)
FROM employees
WHERE department_id < 50
GROUP BY grouping sets((department_id,TO_CHAR(hire_date, 'yyyy')),(department_id));

R C GS
1) Cate criterii folosesc ptr gen grupuri ? (c1,c2,.....,cn)
toate criterrile posibile(2^n) => cube
n+1 criterii (forma triunghiluara) rollup (intotdeauna primul criteriu e folosit)
GS (G Sets) - definesti tu criterrile definite pt generarea coloanelor
Rollup simulat cu GS
Rollup(c1,c2,c3) => Grouping Sets((c1,c2,c3),(c1,c2),(c1),());

select department_name, job_title, e.manager_id, max(salary), sum(salary) ,
grouping(department_name),grouping(job_title),grouping(e.manager_id)
from employees e join jobs using (job_id) join departments using (department_id)
group by grouping sets((department_name,  job_title),(job_title,e.manager_id),());

select department_name,job_title,avg(salary), count(*)
from employees e full join departments d on(e.department_id = d.department_id)
join jobs j on (e.job_id = j.job_id)
group by cube(department_name,job_title);