Date Wise Report With various Users in Column Output

Question

Hi All,
Im Using Oracle 10g Release 2.
i Have a Table Transaction_Details as Like Below
Date Uploaded Pending UserName
14/02/2012 50 25 aa
14/02/2012 20 0 aa
14/02/2012 110 20 bb
14/02/2012 100 30 bb
14/02/2012 150 25 cc
15/02/2012 75 20 aa
15/02/2012 110 40 bb
16/02/2012 100 45 cc
16/02/2012 150 25 bb
i Want the Output Result as,
Date Uploaded    Pending      User Uploaded     Pending      User Uploaded     Pending      User
14/02/2012 70 25 aa 210 50 bb 150 25 cc
15/02/2012 75 20 aa 110 40 bb
16/02/2012 150 25 bb 100 45 cc
But,
I can Able to Generate Output using Group By function,
14/02/2012 70 25 aa
14/02/2012 210 50 bb
14/02/2012 150 25 cc
Kindly Share Ur Idea To Do This.
Thanks In Advance., Im Trying to get outout through a function,
Is this Possible in Sql or Plsql.

Answer 1

Looks like you want something like this...
SQL> ed
Wrote file afiedt.buf
  1  with t as (select to_date('14/02/2012','DD/MM/YYYY') as dt, 50 as up, 25 as pe, 'aa' as usr from dual union all
  2             select to_date('14/02/2012','DD/MM/YYYY'), 20, 0, 'aa' from dual union all
  3             select to_date('14/02/2012','DD/MM/YYYY'), 110, 20, 'bb' from dual union all
  4             select to_date('14/02/2012','DD/MM/YYYY'), 100, 30, 'bb' from dual union all
  5             select to_date('14/02/2012','DD/MM/YYYY'), 150, 25, 'cc' from dual union all
  6             select to_date('15/02/2012','DD/MM/YYYY'), 75, 20, 'aa' from dual union all
  7             select to_date('15/02/2012','DD/MM/YYYY'), 110, 40, 'bb' from dual union all
  8             select to_date('16/02/2012','DD/MM/YYYY'), 100, 45, 'cc' from dual union all
  9             select to_date('16/02/2012','DD/MM/YYYY'), 150, 25, 'bb' from dual)
10  --
11  -- end of test data
12  --
13  select dt
14        ,max(decode(rn,1,up)) as up_1
15        ,max(decode(rn,1,pe)) as pe_1
16        ,max(decode(rn,1,usr)) as usr_1
17        ,max(decode(rn,2,up)) as up_2
18        ,max(decode(rn,2,pe)) as pe_2
19        ,max(decode(rn,2,usr)) as usr_2
20        ,max(decode(rn,3,up)) as up_3
21        ,max(decode(rn,3,pe)) as pe_3
22        ,max(decode(rn,3,usr)) as usr_3
23  from (
24        select dt, up, pe, usr
25              ,row_number() over (partition by dt order by usr) as rn
26        from (
27              select dt, sum(up) as up, sum(pe) as pe, usr
28              from t
29              group by dt, usr
30             )
31       )
32* group by dt
SQL> /
DT                         UP_1       PE_1 US       UP_2       PE_2 US       UP_3       PE_3 US
14-FEB-2012 00:00:00         70         25 aa        210         50 bb        150         25 cc
15-FEB-2012 00:00:00         75         20 aa        110         40 bb
16-FEB-2012 00:00:00        150         25 bb        100         45 ccBear in mind that SQL projection is fixed, so you will need to include enough max(decode... statements to cater for the expected number of rows that are going to be pivoted for each date.
{thread:id=2309172}