Make 1 row turn into X rows, based on value X in column?

Question

Hello everyone, got a question for you! I've got a table as follows:
ITEM_NUM,IN_STOCK
1001A,2
1001B,1
1001C,2
1001D,3
I'd like to get the following output:
1 1001A
2 1001A
3 1001B
4 1001C
5 1001C
6 1001D
7 1001D
8 1001D
So each row gets repeated for as many times as it has for the in_stock value. And down the left side, we get rownum.
I can do this with PL/SQL code (a cursor to get both columns, and for each row, run a loop to place these into a collection). But if there's a way just to skip the loop and write a query, I'd rather do this!
Thanks!
-Thomas H

Answer 1

Well you can pretty much use any row source here but I favour pipelined function for overall simplicity / clarity and performance.
Personal Oracle Database 10g Release 10.1.0.2.0 - Production
With the Partitioning, OLAP and Data Mining options
SQL> CREATE TABLE t (item_num VARCHAR2 (5), in_stock NUMBER);
Table created.
SQL> INSERT INTO t VALUES ('1001A',2);
1 row created.
SQL> INSERT INTO t VALUES ('1001B',1);
1 row created.
SQL> INSERT INTO t VALUES ('1001C',2);
1 row created.
SQL> INSERT INTO t VALUES ('1001D',3);
1 row created.
SQL> CREATE OR REPLACE FUNCTION many (
  2    p_rows IN NUMBER)
  3    RETURN number_table PIPELINED
  4  IS
  5  BEGIN
  6    FOR l_row IN 1..p_rows LOOP
  7      PIPE ROW (l_row);
  8    END LOOP;
  9    RETURN;
10  END;
11  /
Function created.
SQL> SELECT ROWNUM, item_num
  2  FROM   t, TABLE (many (in_stock));
    ROWNUM ITEM_
         1 1001A
         2 1001A
         3 1001B
         4 1001C
         5 1001C
         6 1001D
         7 1001D
         8 1001D
8 rows selected.
SQL>