how to fix this function in pl-sql - oracle

Question

I wrote a function in pl-sql to check if all employees have their salary between the correct range of maximum and minimum salary. But it's giving me error like this :
Error(9,3): PL/SQL: Statement ignored.
Error(9,16): PLS-00306: wrong number or types of arguments in call to '>'
CREATE OR REPLACE FUNCTION MIN_MAX_SAL RETURN NUMBER AS
cursor emp_cur is select salary from employees ;
emp_sal emp_cur%rowtype;
min_sal jobs.min_salary%type;
max_sal jobs.max_salary%type;
BEGIN
select min_salary , max_salary into min_sal , max_sal from jobs;
for emp_sal in emp_cur loop
if ((emp_sal > max_sal) or (emp_sal < min_sal)) then
return 0;
end if;
end loop;
RETURN 1;
END MIN_MAX_SAL;
what is the wrong ??

Answer 1

emp_sal is a row type, even though it only contains one column. You need to specify the column within that which you're comparing:
if ((emp_sal.salary > max_sal) or (emp_sal.salary < min_sal)) then
You don't need to declare emp_sal with this form of cursor loop.
If there is more than one record in jobs, which seems likely, then you first select will throw ORA-02112, too many rows selected. You probably mean to compare each employee with their job's salary range. Assuming you're intentionally doing this in PL/SQL as an exercise and want to use cursors - since it could easily be done in a single SQL statement - you perhaps want to use nested loops, with one loop to find job information and then an inner loop checking all employees with that job, something like:
create or replace function min_max_sal return number as
cursor job_cur is
select job_id, min_salary, max_salary from jobs;
cursor emp_cur(p_job_id jobs.job_id%type) is
select emp_id, salary from employees where job_id = p_job_id;
begin
for job_rec in job_cur loop
for emp_rec in emp_cur(job_rec.job_id) loop
if (emp_rec.salary > job_rec.max_salary)
or (emp_rec.salary < job_rec.min_salary) then
return 0;
end if;
end loop;
end loop;
return 1;
end min_max_sal;
/
Of course you can just join to do this in one hit:
select count(*)
from employees e
join jobs j on j.job_id = e.job_id
where (e.salary > j.max_salary)
or (e.salary < j.min_salary);
or:
where not e.salary between j.min_salary and j.max_salary
You could still use that within a PL/SQL Block if you wanted to, selectng that value into a variable and then returning based on whether the count was zero or not.