OPEN out_cur FOR SELECT exception handling help

Question

Here is the stored procedure for reading a data based on the input, if no data then return null ref cursor and proper error message.
I am not sure, i am handling proper exception handling ? Please help me to complete this item.
Thanks.
PROCEDURE testing
module IN VARCHAR2,
module_id IN VARCHAR2,
out_cur OUT SYS_REFCURSOR,
out_error_no OUT NUMBER
BEGIN
out_error_no := 0;
CASE
WHEN module = 'a' AND module_id = 'b' THEN
BEGIN
OPEN out_cur FOR
SELECT id,
mime_type,
file_length,
file_name ,
uploadeddate,
created_user ,
status_name
FROM l_table_cnt
WHERE id = module_id;
EXCEPTION
WHEN OTHERS THEN
OPEN out_cur_file_cursor FOR
SELECT
NULL id,
NULL mime_type,
NULL file_length,
NULL file_name,
NULL uploadeddate,
NULL created_user,
NULL status_name
FROM dual
WHERE 1= 0;
out_error_no := 2;
RAISE_APPLICATION_ERROR(-20024,'No Document ');
END;

Answer 1

Venkadesh wrote:
The correct way is to just open the ref cursor and pass it back and then the receiving code that is going to use that cursor handles whether there is any code in it or not. can you please explain with simple exampleIs it really that difficult?
Ok...
Here's the procedure to return a ref cursor...
SQL> ed
Wrote file afiedt.buf
  1  create or replace procedure get_rc(p_deptno in number, p_rc out sys_refcursor) is
  2  begin
  3    open p_rc for 'select * from emp where deptno = :1' using p_deptno;
  4* end;
SQL> /
Procedure created.Now we have some application that wants to consume a ref cursor... in this case the application is SQL*Plus, but it could be Java or .NET etc.
It declares it's local reference to the ref cursor...
SQL> var r refcursor;then calls the procedure to get a ref cursor reference assigned...
SQL> exec get_rc(10, :r);
PL/SQL procedure successfully completed.Now, the application itself determines if there is any data when it comes to actually perform the fetches on it...
SQL> print r;
     EMPNO ENAME      JOB              MGR HIREDATE                    SAL       COMM     DEPTNO
      7782 CLARK      MANAGER         7839 09-JUN-1981 00:00:00       2450                    10
      7839 KING       PRESIDENT            17-NOV-1981 00:00:00       5000                    10
      7934 MILLER     CLERK           7782 23-JAN-1982 00:00:00       1300                    10in the above case it had data, so it displayed it.
So what if there isn't any data...
SQL> exec get_rc(90, :r);
PL/SQL procedure successfully completed.
SQL> print r;
no rows selected
SQL>SQL*Plus (the application that calls the procedure) is the one that has determined that there was not data when it came to fetch it (using the print statement in this case). And when it found there was no data it handled it itself (in this case printing the message "no rows returned").
The procedure doesn't have to do any overhead of determining if there is data going to be returned or not, because it's not it's responsibility and completely unnecessary. The calling application can easily determine if there is data or not when it starts to try and fetch it.