Actual difference

Question

Hi Gurus
I'm in the process of learning collection and unable to understand some points. I have the following code:
Code
create or replace
PROCEDURE process_all_rows
IS
   TYPE employees_aat
   IS TABLE OF employees%ROWTYPE;
      --INDEX BY PLS_INTEGER;
   l_employees employees_aat;
BEGIN
   SELECT *
   BULK COLLECT INTO l_employees
      FROM employees;
   FOR indx IN 1 .. l_employees.COUNT
   LOOP
     --  analyze_compensation
     -- (l_employees(indx));
     dbms_output.put_line(l_employees(indx).EMPLOYEE_ID);
     null;
   END LOOP;
END process_all_rows;
I think I'm using  nested table collection type and if you noticed  that I also remarks "Index by PLS Integer"  in my program and if  I uncomment it then it will become Associative array  collection type, Rite?
I executed this program with and without INDEX BY PLS_INTEGER and it gives same records, My question is that in which scenario I have to use INDEX BY PLS_INTEGER clause or in which scenario I don't need to use this clause. Please guide. Thanks.
If possible please refer me some good codes for oracle collection to understand it more properly. Thanks

Answer 1

Associative Array (Or call it Indexed collection) are sparse. You can do this
declare
  type tbl is table of varchar2(10) index by pls_integer;
  var tbl;
begin
  var(100) := 'karthick';
  var(200) := 'oracle';
end;
Such a thing is not possible in nested PL/SQL collection type. If you use Nested collection type you need to use .extend and create 200 elements in your collection to store that 2 values. Basically when ever you want to bulk collect a result from SELECT you can use Nested PL/SQL collection and for manual assignment you can use Associative Array.