Possible to use stored procedure to fill excel template?

Question

Hi,
I want to fill a defined named range in an Excel file with a matrix of values from a stored procedure in Oracle. Is this possible?
Today from our ASP.NET client we ask an oracle database for resultset and then rearrange and work with the resultset and print the final resultset to defined named ranges in an Excel file with the OLEDB provider. We repeatingly execute a SQL statement like "INSERT INTO [DATA_RANGE1] VALUES ('var1','var2','var3');" and for each sql statement a new row is inserted in excel file in the defined named range "DATA_RANGE1". This works fine, but is VERY time consuming!
It would be nice writing all insert statements in one call to the excel file instead of repeatingly call the excel file.
If I try to use sql statements separated with ";", like "INSERT INTO [DATA_RANGE1] VALUES ('var1','var2','var3'); INSERT INTO [DATA_RANGE1] VALUES ('var1','var2','var3');" in one query from the ASP.NET client I get System.Data.Oledb.OledbException "Characters found after end of SQL statement".
Today the connectionstring look like
_connectionExcel = NewSystem.Data.OleDb.OleDbConnection ("Provider=Microsoft.ACE.OLEDB.12.0; Data Source="+ "'" & strFullPathAndName & "';Extended Properties=Excel 12.0 Macro;")
for Excel 2010 files and
_connectionExcel = NewSystem.Data.OleDb.OleDbConnection ("Provider=Microsoft.Jet.OLEDB.4.0;Data Source="+ "'" & strFullPathAndName & "';Extended Properties=Excel 8.0 Macro;")
for Excel 2003 files.
Is there another method to solve this?
I asked about this in an Excel forum and they advised me to use a stored procedure, but I am not sure about how to do this, and if it is possible?
BR, Siri

Answer 1

>
Is it possible to provide a very simple example how to implement such a lock?
>
Funny you should ask. I wanted to check out Tubby's suggestion also so did a very simple (i.e. non-useable) example.
-- function to get a handle - this could be in a package - the AUTONOMOUS_TRANSACTION is only needed if you care about the COMMIT that gets done.
create or replace function get_lock_handle (p_name IN varchar2)
  return VARCHAR2 is
--PRAGMA AUTONOMOUS_TRANSACTION;
  v_handle VARCHAR2(2000);
begin
    DBMS_LOCK.ALLOCATE_UNIQUE (P_NAME, v_handle);
    RETURN v_handle;
end;
-- actual procedure you want to serialize
create or replace procedure test_lock is
  v_handle VARCHAR2(2000);
  v_lock_result INTEGER;
   v_i INTEGER;
begin
  v_handle := get_lock_handle('test_lock'); -- here you could pass the name of the procedure as the name of the lock
  v_lock_result := dbms_lock.request(v_handle, timeout => 2); -- only wait 2 seconds to get the lock
  if (v_lock_result = 0) then
     v_i := 1;
     loop
        if (v_i = 10) then
             exit;
        end if;
        v_i := v_i + 1;    
    dbms_output.put_line ('procedure is executing'); 
     dbms_lock.sleep(5);
      end loop;
  else
    dbms_output.put_line('procedure is locked');
    return;
    end if;
    v_lock_result := DBMS_LOCK.RELEASE(v_handle);
   EXCEPTION
        WHEN OTHERS THEN
           v_lock_result := DBMS_LOCK.RELEASE(v_handle);
end;Now call EXEC TEST_LOCK from two different sessions.
The second session will time out after 2 seconds. Keep running it and once the first session releases the lock the second will execute.
Thanks for the question - and thanks Tubby for suggesting DBMS_LOCK - I hadn't used it for this purpose before.