Given various date ranges, count the occurrences of dates in PL/SQL

Question

i have a table of employees that have a start date and end date on a project. Some employees end date is null because they are still working on the project. So I want to find for each day in JAN, how many people on working on the project.
So I generate the dates JAN 1 - 31 with this:
Declare
  days DATE := To_Date('01-JAN-11','dd-mon-yy');
BEGIN
  FOR days IN 1..31
  LOOP
    Dbms_Output.Put_Line(TO_CHAR(days ,'dd-Mon-yy'));
    days  := days  + 1;
  END LOOP;
END; For employee X (ID 123), his start date is 1st and end date is 4th. So I use this query to list the dates he worked on (Jan1 jan2 jan3 jan4):
DECLARE
da Number;
days Date;     
sta emp.start%Type;
BEGIN
Select end-start+1
Into da
From emp
where emp_id = 123;
Select start
Into sta
From emp
Where emp_id = 123;
  days  := To_Date(sta,'dd-mon-yy');
FOR i IN 1..da
  LOOP
    Dbms_Output.Put_Line(TO_CHAR(days ,'dd-Mon-yy'));
    days  := days  + 1;
  END LOOP;
END;Now the problem I am having is trying to do a for loop to check the work days from each employee in the table. I guess I need to iterate over the employee IDs. The above code, I entered the empoyee ID myself to find his work days, but I am trying to loop over all the employees to find and count the days. Any help on this part?
Ultimately I am trying to get this:
Day Count
Jan 1 5
Jan 2 3
Jan 3 1
Jan 31 5

Answer 1

select  project_id,
        dt,
        nvl(count(emp_id),0)
  from  employee_projects,
        (select date '2010-12-31' + level dt from dual connect by date '2010-12-31' + level <= date '2011-01-31')
  where dt between start_date(+) and nvl(end_date(+),sysdate)
  group by project_id,
           dt
/SY.
Edited by: Solomon Yakobson on Jan 18, 2011 5:01 PM