Will this work?
select b.id
, trunc(b.today) tday
, b.code
, b.value
, sum(B.VALUE) over (partition by b.id, trunc(b.today), b.code order
by b.id, trunc(b.today), B.CODE range between unbounded preceding and
CURRENT ROW) val_total
from ( select id, trunc(today) today, code, sum(value) val_total
from sum_test
group by id, trunc(today), code
) a
, sum_test b
where a.id=b.id
and trunc(b.today)=a.today
and a.code=b.code
order by 2, 1
/
----- Original Message -----
To: "Multiple recipients of list ORACLE-L" <[EMAIL PROTECTED]>
Sent: Monday, February 03, 2003 10:09 AM
> I need some help with some SQL pleeeaasseee....
>
> I'm trying to do some summay work with some data using some of the
> Oracle9i Analytic functions.
>
> Here is the sample data:
>
>
> SQL> select * from sum_test;
>
> ID TODAY CODE VALUE
> ---------- --------- -------------------- ----------
> 1 02-FEB-03 A 100
> 1 02-FEB-03 A 200
> 1 02-FEB-03 B 50
> 1 03-FEB-03 A 50
> 1 03-FEB-03 B 25
> 1 05-FEB-03 A 725
> 1 05-FEB-03 C 125
> 1 02-FEB-03 A 100
> 1 03-FEB-03 A 100
> 2 02-FEB-03 A 100
> 2 03-FEB-03 A 100
> 2 04-FEB-03 B 100
>
> and here is how I want the output to look:
> Note that I want:
> 1. for the first day I want total by ID, tday (truncated date) and code. I
> want the daily total in the VAL_TOTAL column
> 2. For each subsiquent day, I want that days total in val_total ADDED
> to the previous total.
>
> ID TDAY CODE VALUE VAL_TOTAL
> ---------- --------- -------------------- ---------- ----------
> 1 02-FEB-03 A 100 400
> 1 02-FEB-03 A 200 400
> 1 02-FEB-03 A 100 400
> 1 02-FEB-03 B 50 50
> 2 02-FEB-03 A 100 100
> 1 03-FEB-03 A 50 550
> 1 03-FEB-03 A 100 550
> 1 03-FEB-03 B 25 75
> 2 03-FEB-03 A 100 750
> 2 04-FEB-03 B 100 175
> 1 05-FEB-03 A 725 1275
> 1 05-FEB-03 C 125 125
>
> Here is the query I'm trying:
>
> select b.id, trunc(b.today) tday, b.code, b.value,
> sum(a.val_total) over
> (partition by b.id, trunc(b.today), b.code order by b.id,
> trunc(b.today)
> range between unbounded preceding and unbounded following) val_total
> from
> (
> select id, trunc(today) today, code, sum(value) val_total
> from sum_test
> group by id, trunc(today), code
> ) a,
> sum_test b
> where a.id=b.id
> and trunc(b.today)=a.today
> and a.code=b.code
> order by 2, 1
> /
>
> Results...
>
> ID TDAY CODE VALUE VAL_TOTAL
> ---------- --------- -------------------- ---------- ----------
> 1 02-FEB-03 A 100 1200
> 1 02-FEB-03 A 200 1200
> 1 02-FEB-03 A 100 1200
> 1 02-FEB-03 B 50 50
> 2 02-FEB-03 A 100 100
> 1 03-FEB-03 A 50 300
> 1 03-FEB-03 A 100 300
> 1 03-FEB-03 B 25 25
> 2 03-FEB-03 A 100 100
> 2 04-FEB-03 B 100 100
> 1 05-FEB-03 A 725 725
> 1 05-FEB-03 C 125 125
> Obviously not correct. I know I can do this with a self join to the table,
> but I was hopeful that I could do it this way. Any ideas?
>
>
>
>
>
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.net
> --
> Author: Freeman Robert - IL
> INET: [EMAIL PROTECTED]
>
> Fat City Network Services -- 858-538-5051 http://www.fatcity.com
> San Diego, California -- Mailing list and web hosting services
> ---------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).
>
--
Please see the official ORACLE-L FAQ: http://www.orafaq.net
--
Author: Steve Perry
INET: [EMAIL PROTECTED]
Fat City Network Services -- 858-538-5051 http://www.fatcity.com
San Diego, California -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from). You may
also send the HELP command for other information (like subscribing).
