Reputation: 306
I need to write query on employee table to fetch the employee with employee ID & how many days he is present absent & half-day for given date range.
Employee
AID EmpID Status Date
1 10 Present 17-03-2015
2 10 Absent 18-03-2015
3 10 HalfDay 19-03-2015
4 10 Present 20-03-2015
5 11 Present 21-03-2015
6 11 Absent 22-03-2015
7 11 HalfDay 23-03-2015
Expected Output will be :
EmpID Present Absent HalfDay
10 2 1 1
11 1 1 1
Can you please help me with the Sql query ?
Here Is the query I tried
SELECT EMP.EMPID,
(CASE WHEN EMP.STATUS = 'Present' THEN COUNT(STATUS) ELSE 0 END) Pres,
(CASE WHEN EMP.STATUS = 'Absent' THEN COUNT(STATUS) ELSE 0 END) ABSENT,
(CASE WHEN emp.status = 'HalfDay' THEN Count(status) ELSE 0 END) HalfDay
FROM EMPLOYEE EMP GROUP BY emp.empid
Upvotes: 0
Views: 1280
Reputation: 146299
The COUNT() function tests if the value is NOT NULL. Therefore it will always increment for both sides of a CASE statement like this:
COUNT(CASE Status WHEN 'Present' THEN 1 ELSE 0) AS Present
So we need to use SUM() ...
select empid,
sum(case when status='Present' then 1 else 0 end) present_tot,
sum(case when status='Absent' then 1 else 0 end) absent_tot,
sum(case when status='HalfDay' then 1 else 0 end) halfday_tot
from employee
group by empid
order by empid
/
... or use COUNT() with a NULL else clause. Both produce the same output, perhaps this one is clearer:
SQL> select empid,
2 count(case when status='Present' then 1 end) present_tot,
3 count(case when status='Absent' then 1 end) absent_tot,
4 count(case when status='HalfDay' then 1 end) halfday_tot
5 from employee
6 group by empid
7 order by empid
8 /
EMPID PRESENT_TOT ABSENT_TOT HALFDAY_TOT
---------- ----------- ---------- -----------
10 2 1 1
11 1 1 1
SQL>
Note that we need to use ORDER BY to guarantee the order of the result set. Oracle introduced a hashing optimization for aggregations in 10g which meant GROUP BY rarely returns a predictable sort order.
Upvotes: 4
Reputation: 679
Replace 0 with null because it would be also come in count and added the where clause for date range, check the example below:
select empID,
count(case when status='Present' then 1 else null end) Present_Days,
count(case when status='Absent' then 1 else null end) Absent_Days,
count(case when status='HalfDay' then 1 else null end) HalfDays
from Employee
where date >= to_date('17mar2015') and date <= to_date('23mar2015')
group by empID
Upvotes: 0