Understanding sum() over(order by) analytic function

Could you please explain Having_order_by column values computation for below query?
I understand that No_Partition column has been computed over entire result set
select level
,sum(level) over(order by level) Having_order_by
,sum(level) over() No_Partition
from dual
connect by level < 6

Hi,
ActiveSomeTimes wrote:
Could you please explain Having_order_by column values computation for below query?
I understand that No_Partition column has been computed over entire result set
select level
,sum(level) over(order by level) Having_order_by
,sum(level) over() No_Partition
from dual
connect by level < 6
When you have an ORDER BY clause, the function only operates on a window, that is, a subset of the result set, relative to the current row.
When you say "ORDER BY LEVEL", it will only operate on LEVELs less that or equal to the current LEVEL, so on
LEVEL = 1, the analytic fucntion will only look at LEVEL <= 1, that is, just 1; on
LEVEL = 2, the analytic fucntion will only look at LEVEL <= 2, that is, 1 and 2; on
LEVEL = 3, the analytic fucntion will only look at LEVEL <= 3, that is, 1, 2 and 3
LEVEL = 6, the analytic fucntion will only look at LEVEL <= 6, that is, 1, 2, 3, 4, 5 and 6
In the function call without the ORDER BY clause, the function looks at the entire result set, regrdless of what vlaue LEVEL has on the current row.

Similar Messages

Moving sum using date intervals - analytic functions help

let's say you have the following set of data:
DATE SALES
     09/02/2012     100
     09/02/2012     50
     09/02/2012     10
     09/02/2012     1000
     09/02/2012     20
     12/02/2012     1000
     12/02/2012     1100
     14/02/2012     1000
     14/02/2012     100
     15/02/2012     112500
     15/02/2012     13500
     15/02/2012     45000
     15/02/2012     1500
     19/02/2012     1500
     20/02/2012     400
     23/02/2012     2000
     27/02/2012     4320
     27/02/2012     300000
     01/03/2012     100
     04/03/2012     17280
     06/03/2012     100
     06/03/2012     100
     06/03/2012     4320
     08/03/2012     100
     13/03/2012     1000
for each day i need to know the sum of the sales in the present and preceding 5 days (calendar) [not five rows].
What qurey could i use???
Please help!

Hi.
Here's one way.
WITH data AS
     SELECT TO_DATE('09/02/2012','DD/MM/YYYY') d,     100 n FROM DUAL UNION ALL
     SELECT TO_DATE('09/02/2012','DD/MM/YYYY') d,     50 n FROM DUAL UNION ALL
     SELECT TO_DATE('09/02/2012','DD/MM/YYYY') d,     10 n FROM DUAL UNION ALL
     SELECT TO_DATE('09/02/2012','DD/MM/YYYY') d,     1000 n FROM DUAL UNION ALL
     SELECT TO_DATE('09/02/2012','DD/MM/YYYY') d,     20 n FROM DUAL UNION ALL
     SELECT TO_DATE('12/02/2012','DD/MM/YYYY') d,     1000 n FROM DUAL UNION ALL
     SELECT TO_DATE('12/02/2012','DD/MM/YYYY') d,     1100 n FROM DUAL UNION ALL
     SELECT TO_DATE('14/02/2012','DD/MM/YYYY') d,     1000 n FROM DUAL UNION ALL
     SELECT TO_DATE('14/02/2012','DD/MM/YYYY') d,     100 n FROM DUAL UNION ALL
     SELECT TO_DATE('15/02/2012','DD/MM/YYYY') d,     112500 n FROM DUAL UNION ALL
     SELECT TO_DATE('15/02/2012','DD/MM/YYYY') d,     13500 n FROM DUAL UNION ALL
     SELECT TO_DATE('15/02/2012','DD/MM/YYYY') d,     45000 n FROM DUAL UNION ALL
     SELECT TO_DATE('15/02/2012','DD/MM/YYYY') d,     1500 n FROM DUAL UNION ALL
     SELECT TO_DATE('19/02/2012','DD/MM/YYYY') d,     1500 n FROM DUAL UNION ALL
     SELECT TO_DATE('20/02/2012','DD/MM/YYYY') d,     400 n FROM DUAL UNION ALL
     SELECT TO_DATE('23/02/2012','DD/MM/YYYY') d,     2000 n FROM DUAL UNION ALL
     SELECT TO_DATE('27/02/2012','DD/MM/YYYY') d,     4320 n FROM DUAL UNION ALL
     SELECT TO_DATE('27/02/2012','DD/MM/YYYY') d,     300000 n FROM DUAL UNION ALL
     SELECT TO_DATE('01/03/2012','DD/MM/YYYY') d,     100 n FROM DUAL UNION ALL
     SELECT TO_DATE('04/03/2012','DD/MM/YYYY') d,     17280 n FROM DUAL UNION ALL
     SELECT TO_DATE('06/03/2012','DD/MM/YYYY') d,     100 n FROM DUAL UNION ALL
     SELECT TO_DATE('06/03/2012','DD/MM/YYYY') d,     100 n FROM DUAL UNION ALL
     SELECT TO_DATE('06/03/2012','DD/MM/YYYY') d,     4320 n FROM DUAL UNION ALL
     SELECT TO_DATE('08/03/2012','DD/MM/YYYY') d,     100 n FROM DUAL UNION ALL
     SELECT TO_DATE('13/03/2012','DD/MM/YYYY') d,     1000 n FROM DUAL
days AS
     SELECT TO_DATE('2012-02-01','YYYY-MM-DD')+(LEVEL-1) d
     FROM DUAL
     CONNECT BY LEVEL <= 60
totals_per_day AS
     SELECT dy.d,SUM(NVL(dt.n,0)) total_day
     FROM
          data dt,
          days dy
     WHERE
          dy.d = dt.d(+)
     GROUP BY dy.d
     ORDER BY 1
SELECT
     d,
     SUM(total_day) OVER
          ORDER BY d
         RANGE BETWEEN 5 PRECEDING AND CURRENT ROW
     ) AS five_day_total
FROM totals_per_day;
2012-02-01 00:00:00     0
2012-02-02 00:00:00     0
2012-02-03 00:00:00     0
2012-02-04 00:00:00     0
2012-02-05 00:00:00     0
2012-02-06 00:00:00     0
2012-02-07 00:00:00     0
2012-02-08 00:00:00     0
2012-02-09 00:00:00     1180
2012-02-10 00:00:00     1180
2012-02-11 00:00:00     1180
2012-02-12 00:00:00     3280
2012-02-13 00:00:00     3280
2012-02-14 00:00:00     4380
2012-02-15 00:00:00     175700
2012-02-16 00:00:00     175700
2012-02-17 00:00:00     175700
2012-02-18 00:00:00     173600
2012-02-19 00:00:00     175100
2012-02-20 00:00:00     174400
2012-02-21 00:00:00     1900
2012-02-22 00:00:00     1900
2012-02-23 00:00:00     3900
2012-02-24 00:00:00     3900
2012-02-25 00:00:00     2400
2012-02-26 00:00:00     2000
2012-02-27 00:00:00     306320
2012-02-28 00:00:00     306320
2012-02-29 00:00:00     304320
2012-03-01 00:00:00     304420
2012-03-02 00:00:00     304420
2012-03-03 00:00:00     304420
2012-03-04 00:00:00     17380
2012-03-05 00:00:00     17380
2012-03-06 00:00:00     21900
2012-03-07 00:00:00     21800
2012-03-08 00:00:00     21900
2012-03-09 00:00:00     21900
2012-03-10 00:00:00     4620
2012-03-11 00:00:00     4620
2012-03-12 00:00:00     100
2012-03-13 00:00:00     1100
2012-03-14 00:00:00     1000
2012-03-15 00:00:00     1000
2012-03-16 00:00:00     1000
2012-03-17 00:00:00     1000
2012-03-18 00:00:00     1000
2012-03-19 00:00:00     0
2012-03-20 00:00:00     0
2012-03-21 00:00:00     0
2012-03-22 00:00:00     0
2012-03-23 00:00:00     0
2012-03-24 00:00:00     0
2012-03-25 00:00:00     0
2012-03-26 00:00:00     0
2012-03-27 00:00:00     0
2012-03-28 00:00:00     0
2012-03-29 00:00:00     0
2012-03-30 00:00:00     0
2012-03-31 00:00:00     0Hope this helps.
Regards.

How to use analytic function with aggregate function

hello
can we use analytic function and aggrgate function in same qurey? i tried to find any example on Net but not get any example how both of these function works together. Any link or example plz share with me
Edited by: Oracle Studnet on Nov 15, 2009 10:29 PM

select
t1.region_name,
t2.division_name,
t3.month,
t3.amount mthly_sales,
max(t3.amount) over (partition by t1.region_name, t2.division_name)
max_mthly_sales
from
region t1,
division t2,
sales t3
where
t1.region_id=t3.region_id
and
t2.division_id=t3.division_id
and
t3.year=2004
Source:http://www.orafusion.com/art_anlytc.htm
Here max (aggregate) and over partition by (analytic) function is in same query. So it means we can use aggregate and analytic function in same query and more than one analytic function in same query also.
Hth
Girish Sharma

Case Statement in Analytic Function SUM(n) OVER(PARTITION BY x)

Hi Guys,
I have the following SQL that doesn't seem to consider the When clause I am using in the case staement inside the analytic function(SUM). Could somebody let me know why? and suggest the solution?
Select SUM(Case When (A.Flag = 'B' and B.Status != 'C') Then (NVL(A.Amount_Cr, 0) - (NVL(A.Amount_Dr,0))) Else 0 End) OVER (PARTITION BY A.Period_Year) Annual_amount
     , A.period_year
     , B.status
, A.Flag
from A, B, C
where A.period_year = 2006
and C.Account = '301010'
--and B.STATUS != 'C'
--and A.Flag = 'B'
and A.Col_x = B.Col_x
and A.Col_y = C.Col_y
When I use this SQL, I get
Annual_Amount Period_Year Status Flag
5721017.5 --------- 2006 ---------- C -------- B
5721017.5 --------- 2006 ---------- O -------- B
5721017.5 --------- 2006 ---------- NULL ----- A
And when I put the conditions in the where clause, I get
Annual_Amount Period_Year Status Flag
5721017.5 ---------- 2006 ---------- O -------- B

Here are some scripts,
create table testtable1 ( ColxID number(10), ColyID number(10) , Periodname varchar2(15), Flag varchar2(1), Periodyear number(15), debit number, credit number)
insert into testtable1 values(1, 1000, 'JAN-06', 'A', 2006, 7555523.71, 7647668)
insert into testtable1 values(2, 1001, 'FEB-06', 'B', 2006, 112710, 156047)
insert into testtable1 values(3, 1002, 'MAR-06', 'A', 2006, 200.57, 22376.43)
insert into testtable1 values(4, 1003, 'APR-06', 'B', 2006, 0, 53846)
insert into testtable1 values(5, 1004, 'MAY-06', 'A', 2006, 6349227.19, 6650278.03)
create table testtable2 ( ColxID number(10), Account number(10))
insert into testtable2 values(1, 300100)
insert into testtable2 values(2, 300200)
insert into testtable2 values(3, 300300)
insert into testtable2 values(4, 300400)
insert into testtable2 values(5, 300500)
create table apps.testtable3 ( ColyID number(10), Status varchar2(1))
insert into testtable3 values(1000, 'C')
insert into testtable3 values(1001, 'O')
insert into testtable3 values(1002, 'C')
My SQL:
select t1.periodyear
     , SUM(Case When (t1.Flag = 'B' and t3.Status != 'C') Then (NVL(t1.credit, 0) - (NVL(t1.debit,0))) Else 0 End) OVER (PARTITION BY t1.PeriodYear)
     Annual_amount
     , t1.flag
     , t3.status
     , t2.account
from testtable1 t1, testtable2 t2, testtable3 t3
where t1.colxid = t2.colxid
and t1.colyid = t3.colyid(+)
--and t1.Flag = 'B' and t3.Status != 'C'
Result:
PeriodYear ----- AnnualAmount ----- Flag ----- Status ----- Account
2006 ------------------ 43337 --------------- A ----------- C ---------- 300100
2006 ------------------ 43337 --------------- B ----------- O ---------- 300200
2006 ------------------ 43337 --------------- A ----------- C ---------- 300300
2006 ------------------ 43337 --------------- B ------------ ----------- 300400
2006 ------------------ 43337 --------------- A ------------ ----------- 300500
With condition "t1.Flag = 'B' and t3.Status != 'C'" in where clause instead of in Case statement, Result is (which is desired)
PeriodYear ----- AnnualAmount ----- Flag ----- Status ----- Account
2006 ------------------ 43337 --------------- B ----------- O ---------- 300200

Analytical function SUM() OVER (PARTITION BY ) in Crosstab

I am trying to resolve this from a very long time. I have an amount column that has to be grouped on Year, but all the other columns grouped by month. I am trying to achieve this using analytic function SUM(Case when (Condition1 and Condition2) then Sum(Amount) else 0 end) OVER ( PARTITION BY Account, Year), Where Account, Sub Account are the left axis columns. Now, column displays the values correctly, but at different rows. This is confusing.............
For Ex: For Account 00001, there are 3 sub accounts 1000,2000,3000. For Sub account 3000, conditions 1 and 2 are satisfied, so it should display the Amount in the row corresponding to Sub account 3000, and 0 for remaining Sub Accounts. And the Total amount of all the sub accounts, which will be the same as amount for SubAccount 3000 should be displayed in the row corresponding to Account 00001.
But I get blank rows for 1000 and 3000 Sub accounts and Amount displayed in 2000 Sub account, and blank for Account 00001 also.
When I created the same workbook in Tabular form, the same amount is displayed for all the SubAccounts of a single Account.
When I used this CASE statement in TOAD, I figured that this is due to the Analytic function. When I use a group by clause as shown below instead of partition by, I get the results I need.
SELECT (Case when (Condition1 and Condition2) then Sum(Amount) else 0 end), Account, Sub Account FROM tables WHERE conditions GROUP BY Year, Account, Sub Account
But I cannot use groupby for whole SQL of the workbook as I need the other columns with page item 'MONTH' not 'Year'.
Could somebody please help me with this?

Hi,
In your tabular form do you get the correct total display against all you subaccounts and account? If this correct then you can use case to ensure that the total is displayed only for the single account.
Once you have the correct totals working in a tabular form it is easier to re-produce what you want in a cross-tab.
Rod West

HTMLDB 1.6 and "order by" in analytic functions

In HTMLDB 1.6, oracle 10g, when i enter the string "order by" in the region source of a report of the type "sql query (pl/sql function body returning sql query", I get
1 error has occurred
* Your query can't include an "ORDER BY" clause when having column heading sorting enabled.
I understand the reason for this error, but unfortunately i need this for an analytic function:
row_number() over (partition by ... order by ...)
It seems that the check is performed by simply looking for the string "order by" in the "region source" (in fact the error fires even if that string is contained within a comment).
I know possible workarounds (eg creating a view and select'ing from it), i just wanted to let you know.
Regards
Alberto

Another one under the 'obvious route' category:
Seems that the ORDER BY check is apparentl for ORDER<space>BY... so simply adding extra whitespace between ORDER and BY bypasses the check (at least in 2.1.0.00.39).
To make it a bit more obious that a separation is intended, an empty comment, i.e. ORDER/*/BY*, works nicely
Edited by: mcstock on Nov 19, 2008 10:29 AM

Understanding row_number() and using it in an analytic function

Dear all;
I have been playing around with row_number and trying to understand how to use it and yet I still cant figure it out...
I have the following code below
create table Employee(
    ID                 VARCHAR2(4 BYTE)         NOT NULL,
   First_Name         VARCHAR2(10 BYTE),
   Last_Name          VARCHAR2(10 BYTE),
    Start_Date         DATE,
    End_Date           DATE,
     Salary             Number(8,2),
   City               VARCHAR2(10 BYTE),
    Description        VARCHAR2(15 BYTE)
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,       Description)
             values ('01','Jason',    'Martin', to_date('19960725','YYYYMMDD'), to_date('20060725','YYYYMMDD'), 1234.56, 'Toronto', 'Programmer');
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,       Description)
              values('02','Alison',   'Mathews', to_date('19760321','YYYYMMDD'), to_date('19860221','YYYYMMDD'), 6661.78, 'Vancouver','Tester')
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,       Description)
             values('03','James',    'Smith',   to_date('19781212','YYYYMMDD'), to_date('19900315','YYYYMMDD'), 6544.78, 'Vancouver','Tester')
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,       Description)
              values('04','Celia',    'Rice',    to_date('19821024','YYYYMMDD'), to_date('19990421','YYYYMMDD'), 2344.78, 'Vancouver','Manager')
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,       Description)
              values('05','Robert',   'Black',   to_date('19840115','YYYYMMDD'), to_date('19980808','YYYYMMDD'), 2334.78, 'Vancouver','Tester')
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,        Description)
              values('06','Linda',    'Green',   to_date('19870730','YYYYMMDD'), to_date('19960104','YYYYMMDD'), 4322.78,'New York', 'Tester')
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,        Description)
              values('07','David',    'Larry',   to_date('19901231','YYYYMMDD'), to_date('19980212','YYYYMMDD'), 7897.78,'New York', 'Manager')
insert into Employee(ID, First_Name, Last_Name, Start_Date,                     End_Date,                       Salary, City,        Description)
               values('08','James',    'Cat',     to_date('19960917','YYYYMMDD'), to_date('20020415','YYYYMMDD'), 1232.78,'Vancouver', 'Tester')I did a simple select statement
select * from Employee e
and it returns this below
ID   FIRST_NAME LAST_NAME START_DAT END_DATE      SALARY CITY       DESCRIPTION
01   Jason      Martin     25-JUL-96 25-JUL-06    1234.56 Toronto    Programmer
02   Alison     Mathews    21-MAR-76 21-FEB-86    6661.78 Vancouver Tester
03   James      Smith      12-DEC-78 15-MAR-90    6544.78 Vancouver Tester
04   Celia      Rice       24-OCT-82 21-APR-99    2344.78 Vancouver Manager
05   Robert     Black      15-JAN-84 08-AUG-98    2334.78 Vancouver Tester
06   Linda      Green      30-JUL-87 04-JAN-96    4322.78 New York   Tester
07   David      Larry      31-DEC-90 12-FEB-98    7897.78 New York   Manager
08   James      Cat        17-SEP-96 15-APR-02    1232.78 Vancouver TesterI wrote another select statement with row_number. see below
SELECT first_name, last_name, salary, city, description, id,
   ROW_NUMBER() OVER(PARTITION BY description ORDER BY city desc) "Test#"
   FROM employee
   and I get this result
First_name last_name   Salary         City             Description         ID         Test#
Celina          Rice         2344.78      Vancouver    Manager             04          1
David          Larry         7897.78      New York    Manager             07          2
Jason          Martin       1234.56      Toronto      Programmer        01          1
Alison         Mathews    6661.78      Vancouver   Tester               02          1
James         Cat           1232.78      Vancouver    Tester              08          2
Robert        Black         2334.78     Vancouver     Tester              05          3
James        Smith         6544.78     Vancouver     Tester              03          4
Linda         Green        4322.78      New York      Tester             06           5
I understand the partition by which means basically for each associated group a unique number wiill be assigned for that row, so in this case since tester is one group, manager is another group, and programmer is another group then tester gets its own unique number for each row, manager as well and etc.What is throwing me off is the order by and how this numbering are assigned. why is
1 assigned to Alison Mathews for the tester group and 2 assigned to James Cat and 3 assigned Robert Black
I apologize if this is a stupid question, i have tried reading about it online and looking at the oracle documentation but that still dont fully understand why.

user13328581 wrote:
understanding row_number() and using it in an analytic functionROW_NUMBER () IS an analytic fucntion. Are you trying to use the results of ROW_NUMBER in another analytic function? If so, you need a sub-query. Analuytic functions can't be nested within other analytic functions.
...I have the following code below
... I did a simple select statementThanks for posting all that! It's really helpful.
... and I get this result
First_name last_name   Salary         City             Description         ID         Test#
Celina          Rice         2344.78      Vancouver    Manager             04          1
David          Larry         7897.78      New York    Manager             07          2
Jason          Martin       1234.56      Toronto      Programmer        01          1
Alison         Mathews    6661.78      Vancouver   Tester               02          1
James         Cat           1232.78      Vancouver    Tester              08          2
Robert        Black         2334.78     Vancouver     Tester              05          3
James        Smith         6544.78     Vancouver     Tester              03          4
Linda         Green        4322.78      New York      Tester             06           5... What is throwing me off is the order by and how this numbering are assigned. why is
1 assigned to Alison Mathews for the tester group and 2 assigned to James Cat and 3 assigned Robert Black That's determined by the analytic ORDER BY clause. Yiou said "ORDER BY city desc", so a row where city='Vancouver' will get a lower namber than one where city='New York', since 'Vancouver' comes after 'New York' in alphabetic order.
If you have several rows that all have the same city, then you can be sure that ROW_NUMBER will assign them consecutive numbers, but it's arbitrary which one of them will be lowest and which highest. For example, you have 5 'Tester's: 4 from Vancouver and 1 from New York. There's no particular reason why the one with first_name='Alison' got assinge 1, and 'James' got #2. If you run the same query again, without changing the table at all, then 'Robert' might be #1. It's certain that the 4 Vancouver rows will be assigned numbers 1 through 4, but there's no way of telling which of those 4 rows will get which of those 4 numbers.
Similar to a query's ORDER BY clause, the analytic ORDER BY clause can have two or more expressions. The N-th one will only be considered if there was a tie for all (N-1) earlier ones. For example "ORDER BY city DESC, last_name, first_name" would mena 'Vancouver' comes before 'New York', but, if multiple rows all have city='Vancouver', last_name would determine the order: 'Black' would get a lower number than 'Cat'. If you had multiple rows with city='Vancouver' and last_name='Black', then the order would be determined by first_name.

Problem with SUM () analytic function

Dear all,
Please have a look at my problem.
SELECT CURR, DT, AMT, RATE,
SUM(AMT) OVER (PARTITION BY CURR ORDER BY DT) SUMOVER,
sum( amt * rate) over (PARTITION BY CURR ORDER BY DT) / SUM(AMT) OVER (PARTITION BY CURR ORDER BY DT) avgrt
FROM
select 'CHF' CURR, ADD_MONTHS(TO_DATE('01-DEC-07'), LEVEL -1) DT, 100 * LEVEL AMT, 1 + ( 5* LEVEL/100) RATE
FROM DUAL CONNECT BY LEVEL < 10
SQL> /
CUR DT               AMT       RATE    SUMOVER      AVGRT
CHF 01-DEC-07        100       1.05        100       1.05
CHF 01-JAN-08        200        1.1        300 1.08333333
CHF 01-FEB-08        300       1.15        600 1.11666667
CHF 01-MAR-08        400        1.2       1000       1.15
CHF 01-APR-08        500       1.25       1500 1.18333333
CHF 01-MAY-08        600        1.3       2100 1.21666667
CHF 01-JUN-08        700       1.35       2800       1.25
CHF 01-JUL-08        800        1.4       3600 1.28333333
CHF 01-AUG-08        900       1.45       4500 1.31666667
Table Revaluation
select 'CHF' CURR1, '31-DEC-07' DT , 1.08 RATE FROM DUAL UNION ALL
select 'CHF' CURR1, '31-MAR-08' DT , 1.22 RATE FROM DUAL UNION ALL
select 'CHF' CURR1, '30-JUN-08' DT , 1.38 RATE FROM DUAL
CUR DT              RATE
CHF 31-DEC-07       1.08
CHF 31-MAR-08       1.22
CHF 30-JUN-08       1.38.
Problem is with the calculation of average rate.
I want to consider the data in the revaluation table to be used in the calculation of
average rate.
So average rate for Jan-08 will be
(100 * 1.08(dec revaluation rate) + 200 * 1.1 ) / (300) = 1.093333333
for Feb-08
(100 * 1.08(dec revaluation rate) + 200 * 1.1 + 300 * 1.15) / (600) = 1.121666667
for mar-08
(100 * 1.08(dec revaluation rate) + 200 * 1.1 + 300 * 1.15 + 400 * 1.2) / (1000) = 1.153
for Apr-08
(1000 * 1.22(Apr revaluation rate) + 500 * 1.25) /1500 = 1.23
for May-08
(1000 * 1.22(Apr revaluation rate) + 500 * 1.25 + 600 * 1.30 ) /2100 = 1.25
and so on..
Kindly advice

Hi,
The main thing in this problem is that for every dt you want to compute the cumulative total from previous rows using the formula
SUM (amt * rate)
But rate can be either the rate from the revaluation table or the rate from the main table. For evaluating prior dates, you wnat to use the most recent rate.
I'm not sure if you can do this using analytic functions. Like Damorgan said, you should use a self-join.
The query below gives you the results you requested:
WITH
revaluation     AS
     SELECT 'CHF' curr1, TO_DATE ('31-DEC-07', 'DD-MON-RR') dt, 1.08 rate     FROM dual     UNION ALL
     SELECT 'CHF' curr1, TO_DATE ('31-MAR-08', 'DD-MON-RR') dt, 1.22 rate     FROM dual     UNION ALL
     SELECT 'CHF' curr1, TO_DATE ('30-JUN-08', 'DD-MON-RR') dt, 1.38 rate     FROM dual
original_data     AS
     select     'CHF'                              curr
     ,     ADD_MONTHS(TO_DATE('01-DEC-07'), LEVEL -1)     dt
     ,     100 * LEVEL                         amt
     ,     1 + ( 5* LEVEL/100)                    rate
     FROM     dual
     CONNECT BY     LEVEL < 10
two_rates     AS
     SELECT     od.*
          SELECT     MAX (dt)
          FROM     revaluation
          WHERE     curr1     = od.curr
          AND     dt     <= od.dt
          )          AS r_dt
          SELECT     AVG (rate) KEEP (DENSE_RANK LAST ORDER BY dt)
          FROM     revaluation
          WHERE     curr1     = od.curr
          AND     dt     <= od.dt
          )          AS r_rate
     FROM     original_data     od
SELECT     c.curr
,     c.dt
,     c.amt
,     c.rate
,     SUM (p.amt)          AS sumover
,     SUM     ( p.amt
          * CASE
               WHEN     p.dt <= c.r_dt
               THEN     c.r_rate
               ELSE     p.rate
            END
     / SUM (p.amt)          AS avgrt
FROM     two_rates     c
JOIN     original_data     p     ON     c.curr     = p.curr
                    AND     c.dt     >= p.dt
GROUP BY     c.curr,     c.dt,     c.amt,     c.rate
ORDER BY     c.curr,     c.dt
;

Order by in analytic functions

Hi All,
Please explain on how the order by clause in an analytic function vary the results ?.For eg.I get different results for SALSUM column if I order the below query by deptno/empno.
SELECT empno en,deptno dn,sal sal,SUM(sal) OVER (partition by deptno ORDER BY deptno) salsum FROM emp;
EN     DN     SAL     SALSUM
10     1     100     525
20     1     200     525
50     1     225     525
60     2     125     275
30     2     150     275
40     3     250     250
SELECT empno en,deptno dn,sal sal,SUM(sal) OVER (partition by deptno ORDER BY empno) salsum
FROM emp;
EN     DN     SAL     SALSUM
10     1     100     100
20     1     200     300
50     1     225     525
30     2     150     150
60     2     125     275
40     3     250     250

Hi,
SUM(sal) OVER (partition by deptno ORDER BY deptno)
In the above example. you compute sum(sal) department wise in the ascending order of the employee id
then the sum of amt dept 1 will be 525 (since it will sum up amounts acorrding to ascending order of dept no. which must be 1 since the data is partitioned by dept no. hence amt is same accross all the records i.e. 525 accross dept no 1)
and for dept 2 = 275
SUM(sal) OVER (partition by deptno ORDER BY empno)
for the second scenario ..
it will sum up the amounts partitioning by dept no. but considering the order of employee id in ascending order.
e.g for dept 1 there are 3 employess 10,20,50 arranged in ascending order
so the sumsal for emp no 10 = 100
for 10 & 20 = 100+200 = 300
& for 10 ,20, 50 = 100+200+225 = 525
Hope this calrifies..

Sum Analytic Function

Hi,
I have a query in SQL that generates percentage totals. I am having trouble replicating this code in BMM layer of the repository. I have created a new logical column, the sql query is below:
SELECT id, seq, asset_cost ,
CASE
WHEN asset_cost > 0
THEN ROUND(RATIO_TO_REPORT (
CASE
WHEN asset_cost > 0
THEN SUM (asset_cost)
END) OVER (partition BY id)*100)
END total
FROM test
GROUP BY id, seq asset_cost
Can anyone help with replicating the above expression in the logical layer column. ]
*** how can i use the Ratio_to_report function in obiee
The above link shows a workaround
Are there any alternatives to 'RATIO_TO_REPORT' in OBIEE functions?
Thanks
Edited by: sliderrules on 16-May-2012 04:23

Hi,
I have just been through the Oracle documentation to understand that 'RATIO_TO_REPORT' would compute the ratio of a value to sum of values. For your requirement, what you could do is
1. Bring in the measure 'asset_cost' into the BMM with aggregation rule as sum. (I think you could include a condition here itself as asset_cost >0)
2. Create another measure with the 'Derived from another logical column as source' option chosen and the function as
EVALUATE('RATIO_TO_REPORT(%1) OVER (PARTITION BY %2)' AS DOUBLE, asset_cost,id)
The above function does the following steps:
EVALUATE will send the analytic function to the database.
SUM(asset_cost) would be the first parameter
id would be the second parameter.
I might not be pretty good with the syntax here, but hope you could get it while implementing.
Hope this helps.
Thank you,
Dhar

SUM analytical function

Hi,
I am using the SUM analytical function to accumulate some data from one record to the other record (data per month):
TPS_MOI_CODE     PRD_PRD_CODE     PDV_PDV_CODE     RTTCAVCANV
200510     01     9302     -8050
200511     01     9302     -15500
200512     01     9302     -16150
200601     01     9302     -16150
200602     01     9302     -16150
200603     01     9302     -16150
The result is correct. However, I also want to restart the sum from January, i.e every months contain the sum of all the previous month, and it must restart in January.
How do I do that ?
Thanks in advance for your answers.

You should extract a year and use it as the partition in over() clause, for example:
SQL> select * from t;
     DATE#        QTY
    200510          1
    200511          2
    200512          3
    200601          4
    200602          5
    200603          6
6 rows selected.
SQL> desc t;
Name                                      Null?    Type
DATE#                                              NUMBER
QTY                                                NUMBER
SQL> select date#, sum(qty) over(partition by substr(date#,1,4) order by date#) cum_sum
2 from t;
     DATE#    CUM_SUM
    200510          1
    200511          3
    200512          6
    200601          4
    200602          9
    200603         15
6 rows selected.Rgds.

Analytical function sum() ...for Till-date reporting

Hi,
I need help in forming an SQL with analytical function.
Here is my scenario:
create table a (name varchar2(10), qty_sold number,on_date date);
insert into a values ('abc',10,'10-JAN-2007 00:01:00');
insert into a values ('abc',01,'10-JUL-2007 00:01:00');
insert into a values ('abc',05,'10-JUL-2007 08:11:00');
insert into a values ('abc',17,'10-JUL-2007 09:11:00');
insert into a values ('def',10,'10-JAN-2006 08:01:00');
insert into a values ('def',01,'10-JUN-2006 10:01:00');
insert into a values ('def',05,'10-JUL-2006 08:10:00');
insert into a values ('pqr',17,'10-JUL-2006 09:11:00');
Now I want to have a sql which displays the following:
NAME--TOTAL_QTY_SOLD_IN_LAST_10_DAYS, TOTAL_QTY_SOLD_IN_LAST_20_DAYS...etc
I know we can do it using sum(qty_sold) over (order on_date range interval '10' days and preceding) .... but I get too many rows for each "NAME" ....for each of the date in the database table a ... I want just one row for each "Name"...and sum() should be till SYSDATE ....
Any help is highly appreciated.
Thanks.

SQL> select name
2       , sum(case when sysdate - on_date <= 10 then qty_sold end) total_qty_last_10_days
3       , sum(case when sysdate - on_date <= 100 then qty_sold end) total_qty_last_100_days
4       , sum(case when sysdate - on_date <= 500 then qty_sold end) total_qty_last_500_days
5    from a
6   group by name
7 /
NAME          TOTAL_QTY_LAST_10_DAYS   TOTAL_QTY_LAST_100_DAYS   TOTAL_QTY_LAST_500_DAYS
abc                                                         23                        33
def                                                                                    6
pqr                                                                                   17
3 rijen zijn geselecteerd.Regards,
Rob.

Help using oracle syntax "SUM(col1) over (order by col2)" using ODI

Hi all
I want to load data from oracle to ESSBASE using ODI, and I know oracle have such syntax sum(col1) over (order by col2,col3) which can get the accumulation data, e.g
Oracle data table
col1, col2, value
A 2009-1 10
A 2009-2 10
A 2009-3 10
And the essbase need
col1 col2 value
A 2009-1 10
A 2009-2 20
A 2009-3 30
However after i try this in ODI, error occur:
org.apache.bsf.BSFException: exception from Jython:
Traceback (innermost last):
File "<string>", line 32, in ?
java.sql.SQLException: ORA-00979: not a GROUP BY expression
and the original generated SQl by ODI :
select 'HSP_InputValue' "HSP_Rates",MAP_KMDZ_TABLE.BUD_DYKM "Account",MAP_MONTH.ESS_MONTH "Period",MAP_YEAR.ESS_YEAR "Year",'Actual' "Scenario",'Draft' "Version",TEMP_LIRUN.CURRENCY "Currency",MAP_COMPANYCODE.ESS_COMPCODE "Entity",substr(MAP_KMDZ_TABLE.BUD_BUSINESSOBJECT,1,80) "BusinessObject",'Route_NoRoute' "Route",MAP_TRANSPORT.ESS_TRANSPORT "Transport",substr(MAP_KMDZ_TABLE.BUD_BUSINESSACTIVITY,1,80) "BusinessActivity",substr(MAP_KMDZ_TABLE.BUD_CHANNEL,1,80) "Source",'NoCounterparty' "Counterparty",sum(TEMP_LIRUN.DATAVALUE) over (order by MAP_KMDZ_TABLE.BUD_DYKM,MAP_YEAR.ESS_YEAR,MAP_MONTH.ESS_MONTH,TEMP_LIRUN.CURRENCY,MAP_COMPANYCODE.ESS_COMPCODE,MAP_TRANSPORT.ESS_TRANSPORT,MAP_KMDZ_TABLE.BUD_BUSINESSACTIVITY,MAP_KMDZ_TABLE.BUD_BUSINESSOBJECT,MAP_KMDZ_TABLE.BUD_CHANNEL) "Data" from ETL_DEV.TEMP_LIRUN TEMP_LIRUN, ETL_DEV.MAP_KMDZ_TABLE MAP_KMDZ_TABLE, ETL_DEV.MAP_MONTH MAP_MONTH, ETL_DEV.MAP_YEAR MAP_YEAR, ETL_DEV.MAP_COMPANYCODE MAP_COMPANYCODE, ETL_DEV.MAP_TRANSPORT MAP_TRANSPORT where (1=1) And (TEMP_LIRUN.COSTELMNT=MAP_KMDZ_TABLE.SAP_ZZKM)
AND (TEMP_LIRUN.FISCYEAR=MAP_YEAR.SAP_YEAR)
AND (TEMP_LIRUN.FISCPER3=MAP_MONTH.SAP_MONTH)
AND (TEMP_LIRUN.COMP_CODE=MAP_COMPANYCODE.SAP_COMPCODE)
AND (TEMP_LIRUN.WWHC=MAP_TRANSPORT.SAP_WWHC) Group By MAP_KMDZ_TABLE.BUD_DYKM,
MAP_MONTH.ESS_MONTH,
MAP_YEAR.ESS_YEAR,
TEMP_LIRUN.CURRENCY,
MAP_COMPANYCODE.ESS_COMPCODE,
substr(MAP_KMDZ_TABLE.BUD_BUSINESSOBJECT,1,80),
MAP_TRANSPORT.ESS_TRANSPORT,
substr(MAP_KMDZ_TABLE.BUD_BUSINESSACTIVITY,1,80),
substr(MAP_KMDZ_TABLE.BUD_CHANNEL,1,80)
I know ODI think sum.. over must append group by , however it's not! How to solve this problem.
Thank All for your attention
SOS!
Ethan

Hi Ethan,
In my exeprnc I faced a similar kind of situation.
Two work arounds.
1. Write one procedure and execute the same using ODI procedure.
2. Customize a Km and use that KM in your interface.
I guess in your query Group by function is not needed. (if this is the case you can achive this by a smple customization step in KM)
for example : your current KM will generate a query like this:-
select x,y, sum(x) over (order by y) as sumx FROM TestTable group by x, y
and you need a query like this
select x,y, sum(x) over (order by y) as sumx FROM TestTable
go to your KM (duplicate the KM which you are using and rename _withoutGroup )
remove the group by function from select query
(remove the API function <%=snpRef.getGrpBy()%> from insert into i$ table step)
please let me know if you need more help on this
regards,
Rathish

COUNT(DISTINCT) WITH ORDER BY in an analytic function

-- I create a table with three fields: Name, Amount, and a Trans_Date.
CREATE TABLE TEST
NAME VARCHAR2(19) NULL,
AMOUNT VARCHAR2(8) NULL,
TRANS_DATE DATE NULL
-- I insert a few rows into my table:
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '110', TO_DATE('06/01/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '20', TO_DATE('06/01/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '110', TO_DATE('06/02/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '21', TO_DATE('06/03/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '68', TO_DATE('06/04/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '110', TO_DATE('06/05/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Anna', '20', TO_DATE('06/06/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Bill', '43', TO_DATE('06/01/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Bill', '77', TO_DATE('06/02/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Bill', '221', TO_DATE('06/03/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Bill', '43', TO_DATE('06/04/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
INSERT INTO TEST ( TEST.NAME, TEST.AMOUNT, TEST.TRANS_DATE ) VALUES ( 'Bill', '73', TO_DATE('06/05/2005 08:00:00 PM', 'MM/DD/YYYY HH12:MI:SS PM') );
commit;
/* I want to retrieve all the distinct count of amount for every row in an analytic function with COUNT(DISTINCT AMOUNT) sorted by name and ordered by trans_date where I get only calculate for the last four trans_date for each row (i.e., for the row "Anna 110 6/5/2005 8:00:00.000 PM," I only want to look at the previous dates from 6/2/2005 to 6/5/2005 and get the distinct count of how many amounts there are different for Anna). Note, I cannot use the DISTINCT keyword in this query because it doesn't work with the ORDER BY */
select NAME, AMOUNT, TRANS_DATE, COUNT(/*DISTINCT*/ AMOUNT) over ( partition by NAME
order by TRANS_DATE range between numtodsinterval(3,'day') preceding and current row ) as COUNT_AMOUNT
from TEST t;
This is the results I get if I just count all the AMOUNT without using distinct:
NAME     AMOUNT     TRANS_DATE     COUNT_AMOUNT
Anna 110 6/1/2005 8:00:00.000 PM     2
Anna 20 6/1/2005 8:00:00.000 PM     2
Anna 110     6/2/2005 8:00:00.000 PM     3
Anna 21     6/3/2005 8:00:00.000 PM     4
Anna 68     6/4/2005 8:00:00.000 PM     5
Anna 110     6/5/2005 8:00:00.000 PM     4
Anna 20     6/6/2005 8:00:00.000 PM     4
Bill 43     6/1/2005 8:00:00.000 PM     1
Bill 77     6/2/2005 8:00:00.000 PM     2
Bill 221     6/3/2005 8:00:00.000 PM     3
Bill 43     6/4/2005 8:00:00.000 PM     4
Bill 73     6/5/2005 8:00:00.000 PM     4
The COUNT_DISTINCT_AMOUNT is the desired output:
NAME     AMOUNT     TRANS_DATE     COUNT_DISTINCT_AMOUNT
Anna     110     6/1/2005 8:00:00.000 PM     1
Anna     20     6/1/2005 8:00:00.000 PM     2
Anna     110     6/2/2005 8:00:00.000 PM     2
Anna     21     6/3/2005 8:00:00.000 PM     3
Anna     68     6/4/2005 8:00:00.000 PM     4
Anna     110     6/5/2005 8:00:00.000 PM     3
Anna     20     6/6/2005 8:00:00.000 PM     4
Bill     43     6/1/2005 8:00:00.000 PM     1
Bill     77     6/2/2005 8:00:00.000 PM     2
Bill     221     6/3/2005 8:00:00.000 PM     3
Bill     43     6/4/2005 8:00:00.000 PM     3
Bill     73     6/5/2005 8:00:00.000 PM     4
Thanks in advance.

you can try to write your own udag.
here is a fake example, just to show how it "could" work. I am here using only 1,2,4,8,16,32 as potential values.
create or replace type CountDistinctType as object
   bitor_number number,
   static function ODCIAggregateInitialize(sctx IN OUT CountDistinctType)
     return number,
   member function ODCIAggregateIterate(self IN OUT CountDistinctType,
     value IN number) return number,
   member function ODCIAggregateTerminate(self IN CountDistinctType,
     returnValue OUT number, flags IN number) return number,
    member function ODCIAggregateMerge(self IN OUT CountDistinctType,
      ctx2 IN CountDistinctType) return number
create or replace type body CountDistinctType is
static function ODCIAggregateInitialize(sctx IN OUT CountDistinctType)
return number is
begin
   sctx := CountDistinctType('');
   return ODCIConst.Success;
end;
member function ODCIAggregateIterate(self IN OUT CountDistinctType, value IN number)
return number is
begin
    if (self.bitor_number is null) then
      self.bitor_number := value;
    else
      self.bitor_number := self.bitor_number+value-bitand(self.bitor_number,value);
    end if;
    return ODCIConst.Success;
end;
member function ODCIAggregateTerminate(self IN CountDistinctType, returnValue OUT
number, flags IN number) return number is
begin
    returnValue := 0;
    for i in 0..log(2,self.bitor_number) loop
      if (bitand(power(2,i),self.bitor_number)!=0) then
        returnValue := returnValue+1;
      end if;
    end loop;
    return ODCIConst.Success;
end;
member function ODCIAggregateMerge(self IN OUT CountDistinctType, ctx2 IN
CountDistinctType) return number is
begin
    return ODCIConst.Success;
end;
end;
CREATE or REPLACE FUNCTION CountDistinct (n number) RETURN number
PARALLEL_ENABLE AGGREGATE USING CountDistinctType;
drop table t;
create table t as select rownum r, power(2,trunc(dbms_random.value(0,6))) p from all_objects;
SQL> select r,p,countdistinct(p) over (order by r) d from t where rownum<10 order by r;
         R          P          D
         1          4          1
         2          1          2
         3          8          3
         4         32          4
         5          1          4
         6         16          5
         7         16          5
         8          4          5
         9          4          5buy some good book if you want to start at writting your own "distinct" algorythm.
Message was edited by:
Laurent Schneider
a simpler but memory killer algorithm would use a plsql table in an udag and do the count(distinct) over that table to return the value

Running Sum without analytic function

Hi
I have data like below
Create table Test (Name Varchar(30),M Int, Y Int, Val Int);
Insert into Test Values ('A',1,2011,2);
Insert into Test Values ('A',2,2011,2);
Insert into Test Values ('A',3,2011,2);
Insert into Test Values ('A',4,2011,2);
Insert into Test Values ('A',5,2011,2);
Insert into Test Values ('A',6,2011,2);
Insert into Test Values ('A',7,2011,2);
Insert into Test Values ('A',8,2011,2);
Insert into Test Values ('A',9,2011,2);
Insert into Test Values ('A',10,2011,2);
Insert into Test Values ('A',11,2011,2);
Insert into Test Values ('A',12,2011,2);
Insert into Test Values ('A',1,2012,2);
Insert into Test Values ('A',2,2012,2);
Insert into Test Values ('A',3,2012,2);
Insert into Test Values ('A',4,2012,2);
Insert into Test Values ('A',5,2012,2);
Insert into Test Values ('A',6,2012,2);
Insert into Test Values ('A',7,2012,2);
Now based on above data I need to calculate running sum for past 18 Months. Condition is I can not use analytic function or Oracle specific SQL functions (for portability).
I tries following SQL but it dint work
select Name,rnk, SUM(val) from (
SELECT a.Name,a.m,a.Y,b.val, count(*) rnk
from Test a, Test b
where (a.Name=b.Name and (a.M <= b.M and a.Y<= b.Y))
group by a.Name,a.Y,a.m
order by a.Name,a.Y,a.m
) abc
group By Name,rnk
Order by Name,rnk
Can some one give suggastion.

Hi,
I don't see what your query or your desired results have to do with the last 18 months. Is the task here to show for a given month (July, 2012, for example) the total of the 18 months ending in that month (February, 2011 through July, 2012 in this case) for the same name? If so:
SELECT       c.name, c.y, c.m
,       SUM (p.val)     AS running_total
FROM       test     c
JOIN       test     p ON     ( ((12 * c.y) + c.m)
               - ((12 * p.y) + p.m)
               ) BETWEEN 0 AND 17
GROUP BY c.name, c.y, c.m
ORDER BY c.name, c.y, c.m
;Output:
NAME                Y          M RUNNING_TOTAL
A                2011          1             2
A                2011          2             4
A                2011          3             6
A                2011          4             8
A                2011          5            10
A                2011          6            12
A                2011          7            14
A                2011          8            16
A                2011          9            18
A                2011         10            20
A                2011         11            22
A                2011         12            24
A                2012          1            26
A                2012          2            28
A                2012          3            30
A                2012          4            32
A                2012          5            34
A                2012          6            36
A                2012          7            36

Understanding sum() over(order by) analytic function

Similar Messages

Maybe you are looking for