Using rank - urgent

Question

Consider oracle standard emp table, I wrote the below query
select rownum,a.* from (
select empno, deptno, sal, rank() over (partition by deptno order by sal desc nulls last) rank,
dense_rank() over (partition by deptno order by sal desc nulls last) dept_rank
from emp) a
the output is:
ROWNUM EMPNO DEPTNO SAL RANK DEPT_RANK
1 7839 10 5000 1 1
2 7782 10 2950 2 2
3 7934 10 1300 3 3
4 7788 20 3000 1 1
5 7902 20 3000 1 1
6 7566 20 2975 3 2
7 7876 20 1100 4 3
8 7369 20 800 5 4
9 7698 30 2850 1 1
10 7499 30 1600 2 2
11 7844 30 1500 3 3
12 7654 30 1250 4 4
13 7521 30 1250 4 4
14 7900 30 950 6 5
I want to display 1,2 as rank if salaries are same

Answer 1

Hallo ,
in this case you need analytical function row_number():
select rownum,a.* from (
select empno, deptno, sal, rank() over (partition by deptno order by sal desc nulls last) rank,
dense_rank() over (partition by deptno order by sal desc nulls last) dept_rank,
row_number() over (partition by deptno order by sal desc nulls last) row_num_rank
from emp) aRegards
Dmytro
In this example Row_number() shows 1 and 2
for equal salaries randomily, because there is no other sort criteria in order by , other as salary.
You might change your order by - clause,
for example order by sal, empno. Then the rows with equal salary are numerated ascendant by empno
Message was edited by:
Dmytro Dekhtyaryuk