Attempt to cross a null member in function @X

Hi,
I am getting this error "Attempt to cross a null member in function @X" in the below code. How do i resolve this?
IF(Grade_All->@MEMBER(@CONCATENATE(Assumptions_,@NAME(@PARENT(@CURRMBR(Location)))))->"Seat Requirement Flag"==1)
"Production Seat"=((@sumrange("Manpower Required",@List("Agents_L1_Freshers","Agents_L1_Tenured","Agents_L2_L2.5_Billable","SME_Agents_L2_NonBillable"))) / (Grade_All->@MEMBER(@CONCATENATE(Assumptions_,@NAME(@PARENT(@CURRMBR(Location)))))->"Seat Utilization Ratio (Production)"));
ENDIF;)

I used a run time prompt in my FIX statement. Now i am not getting this error.
FIX([Location])
"Production Seat"(
IF(Grade_All->@MEMBER(@CONCATENATE(Assumptions_,@NAME(@PARENT(@CURRMBR(Location)))))->"Seat Requirement Flag"==1)
"Production Seat"=((@sumrange("Manpower Required",@List("Agents_L1_Freshers","Agents_L1_Tenured","Agents_L2_L2.5_Billable","SME_Agents_L2_NonBillable"))) / (Grade_All->@MEMBER(@CONCATENATE(Assumptions_,@NAME(@PARENT(@CURRMBR(Location)))))->"Seat Utilization Ratio (Production)"));
ENDIF;)
ENDFIX
Thanks for ur help.

Similar Messages

How to ignore nulls in analytic functions ( row_number() and count())

how to ignore nulls in analytic functions ( row_number() and count())

Iam attaching test data can any one help me please
thanks in advanceeeee
CREATE TABLE TEMP_table
ACCTNUM NUMBER,
l_DATE TIMESTAMP(3),
CODE VARCHAR2(35 BYTE),
VENDOR VARCHAR2(35 BYTE)
insert into temp_table values (1,sysdate+1/60,'bso','v1');
insert into temp_table values (1,sysdate+2/60,'bsof','v1');
insert into temp_table values (1,sysdate+3/60,'bsof','v2');
insert into temp_table values (1,sysdate+4/60,'','v1');
ian executing this my ;
SELECT acctnum,l_date,vendor,code_1,
       CASE
         WHEN code = 'bsof'
          AND COUNT (DISTINCT code) OVER (PARTITION BY acctnum, vendor) > 1
          AND row_number () OVER (PARTITION BY acctnum, vendor ORDER BY vendor, l_date) != 1
         THEN 'yes'
       ELSE 'no' END result
FROM (select acctnum,l_date,vendor, code code_1, case when code IN ('bso', 'bsof') then code
else null end code   from TEMP_TABLE
ORDER BY acctnum ,l_date);
my result :
1    3/23/2011 5:24:34.000 PM    v1    bso    no
1    3/23/2011 5:48:36.000 PM    v1    bsof    yes
1    3/23/2011 6:36:41.000 PM    v1    bsof    yes
1    3/24/2011 11:55:53.000 AM    v1        no
1    3/23/2011 6:12:38.000 PM    v2    bsof    no
I need to eliminate nulls in top query not in inner query (not using where condition in inner query)
[\code]

How to correct XML Output in Cross Tab Template for sum function?

I have designed a Cross Tab Template to summarize R12 Account Analysis data by Period by Party_name. The template is doing what I want it to do with the exception of amount. I have a function in the sum(accounted_net) field and it will only display 0.00 even though I know there are actual amount. Can someone help in looking at my template to see what I have done wrong?
Here is the sum funciton.
<?if:count(current-group()[CCID_SOURCE=$ABC])?><?sum(current-group()[CCID_SOURCE=$ABC]/ACCOUNTED_NET)?> <?end if?>
CCID_SOURCE is an element I created in XML to concatnate CCID and Party_name for grouping. $ABC is a variable that I defined for "CCID_SOURCE" to check if there is null value for a specific ccid_party. If it's null, it won't do the sum function, otherwise it will sum the accounted_net for the period_name, party_name.
Thank you for your help.
Stacey

Figured this out on our own

Calculated member - Mode function

Hello,
I need some help. I am quite new to MDX; I've created a cube and the measure needs to behave same as the MODE function in Excel (the following
link
describes what the function does). Apparently the only way of achieving this is by MDX (calling the MODE function directly from Excel is not an option for me). Practically the member should return a single value, that represents the price that appears
most times. There are 4 dimensions in my cube: dates, stores, products and client products. I need the following calculations:
1. MODE(price) for specific Date, Product and Client Product. There will be multiple lines returned for this set, as there are multiple stores.
2. MODE(price) for specific Store, Product and Client Product. There will be multiple lines returned for this set, as there are multiple dates.
I found the below code online, that I've adapted to work against my cube, currently the query returns correct results for the first part, Mode(price) for multiple stores. I can change it to do the same for date.
What I need is to create 2 calculated measures using the last select and I am unsure how this can be achieved. Any help will be much appreciated or any other approach in achieving this.
WITH
--Count how often each value appears
MEMBER [Measures].[ValueCount] AS
SUM( Union([Dim Stores].Store].CurrentMember.Level.Members,    {[Dim Stores].[Store].CurrentMember} AS CurrentStore),
IIF( ([Dim Stores].[Store].CurrentMember, [Measures].[Price]) =
(CurrentStore.Item(0).Item(0), [Measures].[Price]) , 1, null))
--Only get the items that appear the most
SET [MaxModes] AS
ORDER(FILTER(NONEMPTY([Dim Stores].[Store].Members, {[Measures].[Price]}),
[Measures].[ValueCount] = MAX(NONEMPTY([Dim Stores].[Store].[Store].Members, [Measures].[Price]),
[Measures].[ValueCount])), [Measures].[Price], ASC)
SELECT {[Measures].[Price]} on 0,
[MaxModes]
--Filter out the duplicates
HAVING [MaxModes].CurrentOrdinal = 0 OR [Measures].[Price] <>   ([Measures].[Price], [MaxModes].Item([MaxModes].CurrentOrdinal - 2)) ON 1
FROM [old_Prices_v2]
WHERE {[Dim Date].[Date].&[2013-06-23T00:00:00]}*{[Dim Client Products].[Client Product].&[13]}*{[Dim Products].[Product].&[551]}

I'm describing below what exactly is required:
The cube has 4 dimensions (Date, Store, Product, Client Product) and one measure (Price). I need a measure to behave same as the MODE function in
Excel (value that occurs most often). Practically the member should return a single value, that represents the price that appears most times.
While browsing the cube I need, for any combination of dimensions/hierarchies, to get the MODE of the price at its lowest grain (that is Date, Store,
Product, Client Product). No need to aggregate the measure in any way.
For example if someone looks at a specific combination of Product & Client Product, the calculated measure would show the price that appears most
times against that Product+Client Product, in any of the stores and in any of the dates.
Example:
Date
Store Product
Client Product
Price
05-Feb-2013
Store1 Prod1
ClProd1 50
05-Feb-2013
Store1 Prod1
ClProd1 60
06-Feb-2013
Store2 Prod1
ClProd1 60
06-Feb-2013
Store2 Prod1
ClProd1 70
05-Feb-2013
Store1 Prod1
ClProd2 80
05-Feb-2013
Store1 Prod1
ClProd2 50
06-Feb-2013
Store2 Prod1
ClProd2 50
06-Feb-2013
Store2 Prod1
ClProd2 70
Case1: If one looks at Prod1 & ClProd1 then the MODE(Price) is 60 as there are 4 rows to be aggregated and the value that repeats most is 60
Case2: If one looks at Prod1 & ClProd1 & Store1 then the MODE(Price) is either 50 or 60 as there are 2 rows to be aggregated and both 50 and
60 repeat 1 time. I this case any of them can be returned (act like TOP 1 in TSQL)
Case3: If one looks at Prod1 & Store1 then the MODE(Price) is 50 as there are 4 rows to be aggregated and the value that repeats most is 50
Case3: If one looks at Prod1 & 06-Feb-2013 then the MODE(Price) is 70 as there are 4 rows to be aggregated and the value that repeats most is 70.
Any guidance will be much appreciated!
Thank you in advance.
Mihai

Need help on Null value check function

Hey guys,
I need to create a function that will accept two values and perform a comparison on them and based on whether or not they're different, return true or false.
Initially I had no idea about the problems of comparing when a value was null, hence the need for a function, as I need to compare many fields of rows of a table in order to find what has changed.
Now, I think I have the NULL check logic in place :
IF R1.X IS NULL THEN
     IF R2.X IS NULL THEN
          RETURN FALSE; -- both R1 and R2 are null so no diffs
     ELSE
          RETURN TRUE; -- R1 is null but R2 is not, so it's a diff
     END IF;
ELSE
     IF R2.X IS NULL THEN
          RETURN TRUE; -- R1 is not null but R2 is so a diff
     ELSE
          IF R1.X != R2.X THEN
               RETURN TRUE; -- both not null but different
          ELSE
               RETURN FALSE; -- both not null but equal
          END IF;
     END IF;
END IF;My problem is that I don't know how, or if i can, create function that can simply accept two column values without defining their datatypes in the function sig. Reason being that while I will always be comparing like-for-like datatypes, I won't always be comparing the "same" like-for-like datatypes; so as I loop through 100 table rows (differentiated by a date), I'll physically compare all the fields in the rows one by one, and in one instance i'm comapring two NUMBERs and then two VARCHARs and then two DATEs, etc.
What i didn't want was duplicate func's that are all the same apart from the sig's.
Is there an easy(ier) way to accomplish this? Is there a more "generic" value type I could use? basically i just want to have a function like :
CREATE OR REPLACE FUNCTION(val1 VALUE, val2 VALUE)
And not multiple like :
CREATE OR REPLACE FUNCTION(val1 NUMBER, val2 NUMBER)
CREATE OR REPLACE FUNCTION(val1 DATE, val2 DATE)
and so on.
Edited by: user11258962 on 28-Jun-2010 09:02

You can use function overloading. So, although three function will exist, they will all be called the same name, so your code will always call a single function
create or replace package pkg_compare as
function is_Alike (p1 in number
                    ,p2 in number) return boolean ;
function is_Alike (p1 in varchar2
                    ,p2 in varchar2) return boolean;
function is_Alike (p1 in date
                    ,p2 in date) return boolean;
end pkg_Compare;
create or replace package body pkg_compare as
function is_Alike (p1 in number
                    ,p2 in number) return boolean as
    begin
      return ( nvl(p1,0) = nvl(p2,-1) );
end is_Alike;
function is_Alike (p1 in varchar2
                    ,p2 in varchar2) return boolean as
    begin
      return ( nvl(p1,'X') = nvl(p2,'Z') );
end is_Alike;
function is_Alike (p1 in date
                    ,p2 in date) return boolean as
    begin
      return ( nvl(p1,sysdate) = nvl(p2,sysdate-1) );
end is_Alike;
end pkg_Compare;
/

Handling NULL in Oracle Function

I try to handle NULL with IS NULL in side the Function. But I some how can't do it. Control always goes to NO_DATA_FOUND Exception block.
In BEGIN block, if I try to handle with IS NULL condition but it doesn't work. But if I use DBMS_OUTPUT.PUTLINE and display the variable with NVL, it tells me it is NULL.
My code is like;
<code>
FUNCTION F_GetProductKey (p_CompanyCod varchar2, p_ProductCod varchar2) return number;
as
vi_ProductKey Number;
BEGIN
DBMS_OUTPUT.PUT_LINE( NVL(p_ProductCod,'abc') <----- This displays 'abc'
IF p_ProductCod IS NULL THEN
SELECT PRODUCT_KEY in vi_ProductKey from DIM_PRODUCT WHERE Product_Cod=-2;
ELSE <---- Control Always Goes to ELSE even p_ProductCod is NULL & then it goes to NO_DATA_FOUND Exception
SELECT PRODUCT_KEY in vi_ProductKey from DIM_PRODUCT WHERE Product_Cod=p_ProductCod AND COMPANY_COD = p_CompanyCod ;
END IF;
return vi_Product_Key;
EXCEPTION
WHEN NO_DATA_FOUND THEN
return -20060;
WHEN OTHERS THEN
return -20001;
END;
</code>
I've a hard coded value in table for Product_Cod=-2, so I should never get No_Data_Found exception if ProductCod is NULL. I checked, for CompanyCod and it is NOT NULL when I call function.
I Googled to look at what I'm doing wrong. I find that NULL values can be handled by ISNULL but some how it doesn't work in my case.
Is there any other way to handle NULL inside function??

I've a hard coded value in table for Product_Cod=-2, so I should
never get No_Data_Found exception if ProductCod is NULL.Are you sure? It seems to me the behaviour you describe could equally well be explained by the control going down the IF ... IS NULL branch and then not finding the PRODUCT_COD in question.
Cheers, APC
Blog : http://radiofreetooting.blogspot.com/
P.S. Having read your subsequent message, specifically this
Product_cod is varchar2. It is actually comparing to
'UNK' not -2 in side the code.I am even more certain that the problem lies in your data. Larry Ellison did not become a multi-billionnaire on the basis of code which behaved in an unpredictable fashion.
Message was edited by:
APC

Member /static function and procedure

hi guys,
i'm trying to figure out diffrerences between the following;
1. member function and static function
2.*member* procdure and static procedure.
i wanna know when to use them when creating an object type.thanks.

hope this enlighten you
http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14261/objects.htm#CHDEFBEA

Hard coded member names vs. member set function

Hello,I'm having a difficult time troubleshooting the following dynamic calc. This formula works perfectly - it gives me the total units sold in the prior 13 weeks:@SUMRANGE("POS Units", @CURRMBRRANGE(Time, LEV, 0, -13, -1))However I'd like to make the member more generic, so instead of "POS Units" I thought I could simply substitute @CURRMBR(Measures), as in:@SUMRANGE(@CURRMBR(Measures), @CURRMBRRANGE(Time, LEV, 0, -13, -1))But when I check the syntax in the outline formula editor, it tells me "Invalid Parameter Type".I have the same problem with the 0 as a parameter of @CURRMBRRANGE. I tried substituting the function @CURLEV(Time) and got the same error.-dan

Well, there are a couple of things. First, if the total member (to store the 13 time periods total) is in the Accounts dimension, then you cannot have it store sums for all possible measures.The SUMRANGE function wants a member to sum - aside from the syntax problem - when SUMRANGE is calculating the Total Units Sold, what do you think THe CURRMBR is from Measures? Its the Total Units Sold! You would be calculating the sum of Total Units Sold over time and putting the result into Total Units Sold. The reason for this is that two members from the same dimension do not intersect each other.Now, if you put a member in a different dimension, and you want to sum any possible measure, use the @SUM command rather than @SUMRANGE. It will implicitly choose the current member from the other dimensions as it calculates.Finally, despite what the documentation says, some functions don't work when used in combination. Also, some function give syntax validation errors, but do work when the calc is run. THere is a "Server Formula Validate" option in the outline editor that sometimes properly evaluates formulas.Hope that helps.Regards,Jade-------------------------------Jade ColeSenior Business Intelligence ConsultantClarity [email protected]

Microsoft VBScript runtime error '800a005e' Invalid use of Null: 'clng' /mc/functions/rpt_downline.asp, line 187

while I am in one of my sites I can access most pages except one and I get this message.

"samspram" <[email protected]> wrote in
message
news:fifhoq$cr7$[email protected]..
> Hi There
> Thanks for the reply.
> It is a MySQL DB, the fields I am referencing are TEXT
datatypes and I
> have
> tried referencing them both left and right.
> The data stored in the fields are comma separated
strings e.g. 1, 2, 3, 4,
> which I am loading into Session variables at login with
the following
> code:-
> Session("allowedsubmenus") =
> Left(rsLogin.Fields.Item("u_allowed_sub_menus").Value,
> (Len(rsLogin.Fields.Item("u_allowed_sub_menus").Value) -
1))
> Session("allowedtopmenus") =
> Left(rsLogin.Fields.Item("u_allowed_top_menus").Value,
> (Len(rsLogin.Fields.Item("u_allowed_top_menus").Value) -
1))
> Session("allowedempmenus") =
> Left(rsLogin.Fields.Item("u_allowed_emp_menus").Value,
> (Len(rsLogin.Fields.Item("u_allowed_emp_menus").Value) -
1))
> Session("allowedcoys") =
> Left(rsLogin.Fields.Item("u_allowed_companies").Value,
> (Len(rsLogin.Fields.Item("u_allowed_companies").Value) -
1))
>
> If I load the data into a variable before performing the
Left() function
> on
> the field then it goes past the lines OK but when I try
and use the
> Session
> variable it then throws the Invalid Use of Null error
again.
> i.e.
> Dim varNum
> varNum =
(Len(rsLogin.Fields.Item("u_allowed_top_menus").Value) - 1)
> Session("allowedsubmenus") =
> Left(rsLogin.Fields.Item("u_allowed_sub_menus").Value,
varNum)
> Code will execute past the loading of sessions in this
way but when I try
> to
> use the session later i.e. as with the Split() function
I get the same
> error
> again.
>
> Regards
> Brendan
You are referencing the Recordset Column Value multiple
times.
try putting it into a variable first
varValue = rsLogin.Fields.Item("u_allowed_sub_menus").Value
then proceed with your operations using that variable
Session("allowedsubmenus") = Left(varValue , (Len(varValue )
- 1))

Expected type [NUMBER] found [STRING] (["member"]) in function[operator@div

Hi,
I am trying to recalculate a figure based on an annual percentage. The script is
FIX(yr2011, Plan, CC_4363,draft)
"AC_&&&&&& Salary" ="AC_&&&&&&Salary"* ("pay increase%"/100)
endfix
endexclude
I receive the error "expected type [NUMBER] found [STRING] (["AC_&&&&&&"]) in function[operator@div"
I looked at this forum and someone suggested using the @match function. I tried
"AC_&&&&&& Salary" ="AC_&&&&&& Salary"* (@match(Accounts," pay increase%")/100);
This validated but the salary figure disappeared.
The reason I am not loading at monthly (level zero) is due to a business requirement. I am having to spread the figures down the months from the total year but if the total year is wrong or missing the montly data obviously disappears too.
Thanks,
Nathan

A few questions:
1) I see an ENDEXCLUDE, but no starting EXCLUDE. How does that relate to the code?
2) Do you really have member names with six ampersands in them? Is that even an allowed character? I guess it is but it looks odd as & is the character used to define the usage of an Essbase Substitution Variable. Okay, that was a comment, not a question, I guess.
3) Do you have a Pay Increase% in every month? Or is it annual?
3) Why don't you stick the salary value to be allocated into another Account, e.g., "Annual Salary" and place it in a single month. Then your code could look like:
FIX(yr2011, Plan, CC_4363,draft, "Jan":"Dec")
"AC_&&&&&& Salary" ="Annual Salary"->"BegBalance" * ("pay increase%"->"BegBalance" / 100)
endfix
I created a BegBalance member which you may not have -- chose Dec or Jan as your home for Annual Salary and Pay Increase and go from there. I also forced the calc to happen at the month level -- I am assuming you have months in your app, but maybe not. Salt to taste.
Regards,
Cameron Lackpour

Can we change the display order of member set functions like @Idescendant?

Generally, for example, if we have dimension with member hierarchy as below:
Parent
Child1
Child11
Child2
Child21
then, when we use function @Idescendant(Parent) to define the row during form design, the form will display the members in the following order:
Child11
Child1
Child21
Child2
Parent
This is not what I want, I need the order to be the same as I defined them.
I know I can set the members one by one to display any order as I wish, but this is not doable when I have hundreds of members.
is there any good way to solve this problem? Any answer is appreciated.
thanks
tony

Hi,
The answer is unfortunately no, there isn't any easy way of doing it. There are ways of doing it and automating it but I can't say it is easy. The solution I found once required me to update form definition tables through a stored procedure in Planning and recycle services after every update.
Cheers,
Alp

Member Name Function

Can it b e ran in a report outside of the grid to get a "prompted" value?

Yes, it can be done. Please use the correct Member Function syntax.

Attempt to access a null bean.

Hi,
When i run my program it says,
"Attempted a bean operation on a null object".
<jsp :useBean id="myCar"class="com.wrox.cars.CarBean" />
I have a <jsp:getProperty name="myCar" property="make" />
<jsp:setProperty name="myCar" property="make" value="Ferrari" />
Now I have a <jsp:getProperty name="myCar" property="make" /
Any suggestions.
Thanks.

Hi Annie,
I am totaly new to jsp.
all i have is two files.
1)package com.wrox.cars;
import java.io.Serializable;
public class CarBean implements Serializable
private String make="Ford";
public CarBean()
public String getMake()
return make;
public void setMake(String make)
this.make=make;
then 2)
<html>
<head>
<title>Using a JAVA BEAN</title>
</head>
<body>
<h2>Using a JAVA BEAN</h2>
<jsp :useBean id="myCar"class="com.wrox.cars.CarBean" />
I have a <jsp:getProperty name="myCar" property="make" />
<jsp:setProperty name="myCar" property="make" value="Ferrari" />
Now I have a <jsp:getProperty name="myCar" property="make" />
</body>
</html>
I did complie the java file before running this.

Safari freezes when attempting to go back using the wipe function of the magic mouse.

Whenever I wipe a little bit to the side on my magic mous the opened tab starts freezing when attempting to go back.
Restart didn't help and there was no system error Message.
Thank you for supporting!
By the way, I'm running Maverick already.

From the Safari menu bar, select
Safari ▹ Preferences... ▹ Privacy ▹ Remove All Website Data
and confirm. Test.

Function to retrieve ancestor member

Hi All,
I'm having issues retrieving an ancestor in a calc script on 9.2.0.1.
I am calc'ing a Level0 entity (in the fix statement) and Im trying to say that it equals its "grand" parent. For example:
ParEnt
- Child01
-Child001
Child001=ParEnt
I need a function because this calc gets repeated for various Level0 entities.
I thought @ANCEST or @ANCESTORS or @PARENT would work but I dont seem to be able to get it right and keep getting:
"Error 1200370 Error calculating formula: attempt to cross a null member in function [@X]"
I have tried:
@ANCEST ("Entities",5)
@ANCEST (@CURRMBR("Entities"),5)
Also tried -2 instead of 5 to indicate a Level number. It works with -1 (Level1) or 2 (generation 2) but not with the generations or levels that I need (5 and 2 respectively).
There are definitely at least 2 levels for every combination.
Ive also tried @PARENT(@PARENT("Entities")) but this only returns the value of the parent member, not the parent of the parent.
Any ideas what the correct function should be?
One more thing...All the above functions DO validate but I get the error when the execution of the calc script.
Thanks for your help.
Seb

Hi Seb,
Not sure if i am understanding this correctly, but from your post you say you want Child001=ParEnt.
If for one account member:
FIX(@RELATIVE("Entities",0))
ACCOUNT = @PARENTVAL("Entities",ACCOUNT);
ENDFIX
Or, if for a number of accounts use a static single member in the formula, i.e. SCENARIO
FIX(@RELATIVE("Entities",0), @DESCENDANTS(ACCOUNT))
SCENARIO = @PARENTVAL("Entities",SCENARIO);
ENDFIX
You can also change PARENTVAL to ANCESTVAL, see admin guide for the syntax
J

Attempt to cross a null member in function @X

Similar Messages

Maybe you are looking for