Selecting missing date range with previous records value

Question

Hello,
SQL> select * from v$version;
BANNER
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
PL/SQL Release 11.2.0.1.0 - Production
CORE    11.2.0.1.0      Production
TNS for 32-bit Windows: Version 11.2.0.1.0 - Production
NLSRTL Version 11.2.0.1.0 - Production
SET DEFINE OFF;
create table t(NBR_OF_S number, NBR_OF_C number, S_DATE date);
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (34, 40, TO_DATE('05/01/2012 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (27, 29, TO_DATE('03/01/2012 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (21, 23, TO_DATE('12/01/2011 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (19, 20, TO_DATE('10/01/2011 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (18, 19, TO_DATE('09/01/2011 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (17, 17, TO_DATE('08/01/2011 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (16, 16, TO_DATE('06/01/2011 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (9, 9, TO_DATE('12/01/2010 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (5, 5, TO_DATE('11/01/2010 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
Insert into  T
   (NBR_OF_S, NBR_OF_C, S_DATE)
Values
   (2, 2, TO_DATE('01/01/2010 00:00:00', 'MM/DD/YYYY HH24:MI:SS'));
COMMIT;
SQL> select * from t order by 3 desc;
  NBR_OF_S   NBR_OF_C S_DATE
        34         40 01-MAY-12
        27         29 01-MAR-12
        21         23 01-DEC-11
        19         20 01-OCT-11
        18         19 01-SEP-11
        17         17 01-AUG-11
        16         16 01-JUN-11
         9          9 01-DEC-10
         5          5 01-NOV-10
         2          2 01-JAN-10
10 rows selected.I want the output like as follows, all those missing date i need to carry on the last one's number
  NBR_OF_S   NBR_OF_C S_DATE
        34         40 01-MAY-12
        27         29 01-APR-12
        27         29 01-MAR-12
        21         23 01-FEB-12
        21         23 01-JAN-12
        21         23 01-DEC-11
        19         20 01-NOV-11
        19         20 01-OCT-11
        18         19 01-SEP-11
        17         17 01-AUG-11
        16         16 01-JUL-11
        16         16 01-JUN-11
         9          9 01-MAY-11
         9          9 01-APR-11
         9          9 01-MAR-11
         9          9 01-FEB-11
         9          9 01-JAN-11
         9          9 01-DEC-10
         5          5 01-NOV-10
         2          2 01-OCT-10
         2          2 01-SEP-10
         2          2 01-AUG-10
         2          2 01-JUL-10
         2          2 01-JUN-10
         2          2 01-MAY-10
         2          2 01-APR-10
         2          2 01-MAR-10
         2          2 01-FEB-10
         2          2 01-JAN-10Any help would be greatly appreciate.
Note: The date value I have created for this sample is monthly, based on the condition the data value I may need to generate weekly also. That's Monthly or weekly either one
Thanks,

Answer 1

And what is AMT? You didn't provide such column in your original post. Anyway, MODEL solution based on original post:
select  nbr_of_s,
        nbr_of_c,
        s_date
  from  t
  model
    partition by(rowid)
    dimension by(1 d)
    measures(
             nbr_of_s,
             nbr_of_c,
             s_date,
             months_between(lag(s_date) over(order by s_date desc),s_date) cnt
    rules(
          nbr_of_s[for d from 1 to cnt[1] increment 1] = nbr_of_s[1],
          nbr_of_c[for d from 1 to cnt[1] increment 1] = nbr_of_c[1],
            s_date[for d from 1 to cnt[1] increment 1] = add_months(s_date[1],cv(d) - 1)
  order by s_date desc
  NBR_OF_S   NBR_OF_C S_DATE   
        34         40 01-MAY-12
        27         29 01-APR-12
        27         29 01-MAR-12
        21         23 01-FEB-12
        21         23 01-JAN-12
        21         23 01-DEC-11
        19         20 01-NOV-11
        19         20 01-OCT-11
  NBR_OF_S   NBR_OF_C S_DATE   
        18         19 01-SEP-11
        17         17 01-AUG-11
        16         16 01-JUL-11
        16         16 01-JUN-11
         9          9 01-MAY-11
         9          9 01-APR-11
         9          9 01-MAR-11
         9          9 01-FEB-11
  NBR_OF_S   NBR_OF_C S_DATE   
         9          9 01-JAN-11
         9          9 01-DEC-10
         5          5 01-NOV-10
         2          2 01-OCT-10
         2          2 01-SEP-10
         2          2 01-AUG-10
         2          2 01-JUL-10
         2          2 01-JUN-10
  NBR_OF_S   NBR_OF_C S_DATE   
         2          2 01-MAY-10
         2          2 01-APR-10
         2          2 01-MAR-10
         2          2 01-FEB-10
         2          2 01-JAN-10
29 rows selected.
SQL> SY.
Edited by: Solomon Yakobson on Jun 11, 2012 1:34 PM