Change Quarters to Months

Question

Hello all,
Alright so I'm trying to change the quarters, to a set of twelve months.
I have data that looks like...
Date--------col2------col3---amount
2009Q1------22------33------100
2009Q2------55------66------200
I'm doing a substr and case to turn 2009Q1 into JAN-09 and so on for the other quarters. What I need to do is also create rows for the other months in the quarter, not just the first one.
So, using the example above, I need the data to look like this...
Date--------col2------col3---amount
JAN-09------22------33------100
FEB-09------22------33------0
MAR-09------22------33------0
APR-09------55------66------200
MAY-09------55------66------0
JUN-09------55------66------0
Q1 has been transformed into Jan/Feb/Mar and the amount only stays in the first month. The following 2 months should get a 0 for the amount.
Any help is appreciated!
Thanks
-Chris

Answer 1

Hi,
Solution with model (10g).
REgards salim.
SELECT to_char(dt,'mon-yyyy') dt,col2,col3,amount
  FROM t
  model
  return updated rows
  partition by (q qq)
  dimension by (1 rn)
  measures(col2,col3,amount,to_date( substr(q,1,4)||'-01-01','yyyy-mm-dd') dt, case when substr(q,-1)=1 then 1
                                    when substr(q,-1)=2 then 4
                                    when substr(q,-1)=3 then 7 
                                    else 10 end  start_month ,
                                    case when substr(q,-1)=1 then 3
                                    when substr(q,-1)=2 then 6
                                    when substr(q,-1)=3 then 9 
                                    else 12 end  end_month  )
  (dt[ for rn from start_month[1] to end_month[1] increment 1]=add_months(dt[1],cv(rn)-1),
  col2[ for rn from start_month[1] to end_month[1] increment 1]=col2[1] ,
  col3[ for rn from start_month[1] to end_month[1] increment 1]=col3[1],
  amount[ for rn from start_month[1] to end_month[1] increment 1]=case when cv(rn)=start_month[1] then amount[1]
                                                                       else 0 end )
  order by qq,rn
alter session set nls_language=english;
WITH t AS
     (SELECT '2009Q1' q, 22 col2, 33 col3, 100 amount
        FROM DUAL
      UNION ALL
      SELECT '2009Q2', 55, 66, 200
        FROM DUAL
        UNION ALL
      SELECT '2009Q3', 66, 88, 300
        FROM DUAL
        UNION ALL
      SELECT '2009Q4', 77, 99, 400
        FROM DUAL
        UNION ALL
      SELECT '2010Q1', 99, 11, 500
        FROM DUAL)
SELECT to_char(dt,'mon-yyyy') dt,col2,col3,amount
  FROM t
  model
  return updated rows
  partition by (q qq)
  dimension by (1 rn)
  measures(col2,col3,amount,to_date( substr(q,1,4)||'-01-01','yyyy-mm-dd') dt, case when substr(q,-1)=1 then 1
                                    when substr(q,-1)=2 then 4
                                    when substr(q,-1)=3 then 7 
                                    else 10 end  start_month ,
                                    case when substr(q,-1)=1 then 3
                                    when substr(q,-1)=2 then 6
                                    when substr(q,-1)=3 then 9 
                                    else 12 end  end_month  )
  (dt[ for rn from start_month[1] to end_month[1] increment 1]=add_months(dt[1],cv(rn)-1),
  col2[ for rn from start_month[1] to end_month[1] increment 1]=col2[1] ,
  col3[ for rn from start_month[1] to end_month[1] increment 1]=col3[1],
  amount[ for rn from start_month[1] to end_month[1] increment 1]=case when cv(rn)=start_month[1] then amount[1]
                                                                       else 0 end )
  order by qq,rn;
DT             COL2       COL3     AMOUNT
jan-2009         22         33        100
feb-2009         22         33          0
mar-2009         22         33          0
apr-2009         55         66        200
may-2009         55         66          0
jun-2009         55         66          0
jul-2009         66         88        300
aug-2009         66         88          0
sep-2009         66         88          0
oct-2009         77         99        400
nov-2009         77         99          0
dec-2009         77         99          0
jan-2010         99         11        500
feb-2010         99         11          0
mar-2010         99         11          0
15 rows selected.