Double record make it to single record in Query for perticular period in BW

Hi
In my ODS some of the records got doubled from Mar 17th to May 31st but from 1st of june its correct , its Open orders, i dont want to reload data again to that ODS , because its Open orders anyways its going to be close after some time , automatically it will become "0".
So i want tocalculate in query itself to make it single
please some one suggest me how can i do for that perticular period divided by 2
for example :
Open Orders (its In ODS)
120
140
20
10
50
I need output should be
60
120
10
5
25
Note : BW 3.1 Version
Thanks,
GAl
12

hi,
You can create a a formula variable with replacement path on the Period(replace with key of Posting Date/Calday) and use it for comprision in the fomula like below
if posting date is in between date1 and date2
order value = order value/2.
else
ordervalue = ordervale.
In Bex Query designer you will have the below formula
(posting date formula variable >date1 and posting date formula variable <date2) * (Ordervalue/2) + ordervalue
But also note that when the order is closed the value of order value will net become zero since the value is doubled up in the DSO , when the delta records come into DSO the order will be closed but the value will still not become zero since the reverse image will give negative value of the actual order value.
For example if your order value is 100, in your DSO it will currently show the value is 200(doubled up). In the Delta the reverse value come as -100. So the final value will still be 100(200-100).
So the best solution will be to do selective deletion
Thanks,
Praveen
Edited by: Praveen kumar kamineni on Jun 11, 2010 11:06 AM

Similar Messages

How can I add recording time to a single recording not a series recording?

If I set-up a series recording it allows me to add minutes to the recording time why can't I add time to a single recording? For example if I want to record a single show tomorrow I find it on the guide and hit record. Now I want to add a few minutes on the end of the show because I know it always runs over it's advertised time why can't I add the time? It doesn't gives to extra options for a single recording. The only time I can add time to a single recording is if the show is currently on then it allows me to add time but, not before the show is on. So what good does that do? Comcast allowed me to set-up extra options to a single recording.

Actually you can. It's a bit clumsy but what you have to do is go to the DVR menu, select scheduled reordings, select the recording you want to extend, scroll through the menu and you will find an option to modify the start / end time.
If a forum member gives an answer you like, give them the Kudos they deserve. If a member gives you the answer to your question, mark the answer as Accepted Solution so others can see the solution to the problem.

A list of employees (a multi record block in a single record)

I'm making a tasks form. Each task is assigned to a number of employees.
I want to display my task details, and in the same form show a list of people who are assigned to it.
What control do I display the employees in? Can someone please guide me in the right direction?
+-----------------------------------------------------------+
| Task ID     Task Name                   Assigned to:      |
| [143     ] [A task to be solved     ] +--------------+ |
|                                         | John Smith   | |
| Description                             | Lucky Day    | |
| +------------------------------------+ | Ned Neiderla | |
| |This is the description of the task.| |              | |
| |                                    | |              | |
| |                                    | |              | |
| +------------------------------------+ +--------------+ |
+-----------------------------------------------------------+

Another question regarding the same thing.
I now made 2 pushbuttons to add people/groups to my ResponsibleContainer.
+-----------------------------------------------------------+
| Task ID     Task Name                   Assigned to:      |
| [143     ] [A task to be solved     ] +--------------+ |
|                                         | John Smith   | |
| Description                             | Lucky Day    | |
| +------------------------------------+ | Ned Neiderla | |
| |This is the description of the task.| |              | |
| |                                    | |              | |
| |                                    | |              | |
| +------------------------------------+ +--------------+ |
|                            add another: [User ] [Group ] |
+-----------------------------------------------------------+How should I achieve this? When-Button-Pressed, open a new form where I can select a User or a Group. But how do I return that User/Group value to the previous form, and update the list?
Thanks,
Botzy

Getting a list instead of a single record.

Hello Experts,
I’m trying to get a record out of a table with the highest amount of money. I’ve used the following query to get the maximum about of money sitting in the table and it gave me the results.
SELECT MAX(auction_item.item_highest_bid)
FROM `auction_item`
I also need two more fields from the same table. Item_id and seller_id. Now what happens is when I write the following SQL I get a list of records instead of a SINGLE record with the highest amount of money.
SELECT MAX(auction_item.item_highest_bid),
auction_item.item_id,
auction_item.seller_id
FROM `auction_item`
GROUP BY auction_item.item_id,
auction_item.seller_id
Can please somebody shows me what’s wrong with my query and how I can get SINGLE record instead of list.
Thanks a lot in advance.

There are several ways to get this information - subqueries, analytical sql. A lot of it depends on the data.
Is there a possibility that two or more records will have the same MAX(auction_item.item_highest_bid)? And if so, how is this to be handled? I'm making some assumptions about item_id and seller_id...so take with a grain of salt
As an aside...you need to read up on GROUP BY and aggregation in sql as this scenario is explained in documentation/books.
Without the benefit of a database to test this, here are a couple of ideas
SELECT MAX(auction_item.item_highest_bid),
auction_item.item_id,
auction_item.seller_id
FROM auction_item
WHERE auction_item.item_highest_bid = (SELECT MAX(i_auction_item.item_highest_bid)
FROM auction_item i_auction_item
WHERE i_auction_item.item_id = auction_item.item_id
AND i_auction_item.seller_id = auction_item.seller_id)
Or
SELECT i.item_highest_bid, i.item_id, i.seller_id
FROM (SELECT a.item_highest_bid, a.item_id, a.seller_id
FROM auction_item a
ORDER BY a.item_highest_bid DESC) i
WHERE i.rownum = 1
Or
SELECT i.item_highest_bid, i.item_id, i.seller_id
FROM (SELECT a.item_highest_bid, a.item_id, a.seller_id, ROW_NUMBER() OVER (ORDER BY a.item_highest_bid DESC) bid_rank
FROM auction_item a) i
WHERE i.bid_rank = 1

[AS] Trying to change Single Record number under data merge?

Any help would be great?

This is where I am trying to go with this:
tell application "Adobe InDesign CS3"
tell data merge 1 of document 1
tell data merge preferences
set record selection to all records
set records per page to single record
set record number to 3--or any number or i value I want to repeat with
end tell
end tell
tell document 1
tell data merge properties to merge records--and merge that one number
end tell
end tell
Setting the record number actually changes the field for single record but when I try to merge said number it does all the records. I am sure its an ordering thing but it is still causing some head scratches. You've all been really helpful with an area that doesnt seem to be detailed anywhere.
Any help would be great. Thanks.

How to make few columns of single record inactive in Table conrol?

Hi all,
Scenario is like this : i have one table control , it is displaying 10 records , in that first four columns are editable. Now my requirement is, whenever record having intial values( suppose one record having initial values among ten) except key field , we need to deactive that particular record , now table control should contain ten records , but nine records are in editable state and one record in non editable state .
How to make few columns of single record inactive in Table conrol?
Thanks in advance.

hi there...
select the column that u want to convert to non editable mode in the screen layout. then double click on it. go to the attributes tab and uncheck the input chk box. the check box will become disabled for input. u can do this for any number of columns.
if u want to do it dyanamically, loop at screen. check for the screen element's name. when itmatches the column u want to disable, simply set the input value as 0.
dont forget to modify sscreen after setting the attribute values.
i hope this helps.
do reward if helpful.

How to make a single record datablock?

Hi all,
I got a data-entry form (9i) that will input a record one at a time and commit before entering another record.
I try to set the block as single-record to yes but it gives error, and after checking with help, I understand that it can not work with database data block.
Is there any idea on how to make the data block to show only one record? The problem is that after entering some data on the record, sometime the data-entry clerk may push the arrow-key that make form shows the next record items, which will be empty, and this will confuse the clerk!
Thank you for your help.

Add Triggers KEY-UP(DOWN, NXTREC, PRVREC...) to the following effect:
NULL;
or
NEXT (PREVIOUS) _ITEM;

Data Merge - Sets of Single Records?

I am trying to use data merge to fill in fields, only problem is I need 2 sets of 2 records on one page. I could possibly reformat this to work on one page, but I would still need 2 sets for each record. Also, I know I could probably just change the data to have 2 of each record, but assume the data is unable to be changed. If there is a way to make a text box a slave of another where it could mimic the original record, maybe I could do it that way, I just don't know if that is possible. See image below.
I am using InDesign CS3. Any ideas?

I don't think there's any way to do this directly, but it could be done in two steps using a script to get your two-up layout.
First, set up a page to hold only one record and do the two instances of the names as a single record per page.
Next, go to http://forums.adobe.com/message/1308215 and copy the rotatepage.js script into a plain text editor like notepad, and save withthe .js extension in the Scripts panel subdolder of the scripts folder in your user profile or user library. In Windows XP the path is \Documents and Settings\<user name>\Application Data\Adobe\InDesign\Version 5.0\Scripts\Scripts Panel and it will be similar in other operating systems.
To run the script, double-click it in the scripts panel. I Think it should run as is, but if it fails, make a new subfolder in the scripts panel folder called "Version 3.0 Scripts" (without the quotes) and move it into that, then try again. Waht this script does is make a PDF of the open file, then create a new file with rotated pages and places the PDF rotated into the new file. The primary use was for doing things like calendars, but it will work to give you spreads that you can print (by checking the spreads box in the print dialog) as two-up.

Sender File Adapter cannot send single record per message ?

Hi,
I have scenario flat file to jdbc. but then why the sender file adapter didn't split the record to become single record per message eventhough i have set it in "Recordsets Per Message" = 1. ?
Document Name               MT_APINVOICE
Document Namespace          urn:file:jdbc:apivinvoice
Document Offset               8
Recordset Name               INPUT
Recordset Namespace
Recordset Structure                          RECORD,*
Recordset Sequence                          Ascending
Recordset per Message          1
Key Field Name
Key field Type               String
RECORD.fieldFixedLengths     10,5,10,10
RECORD.endSeparator     'nl'
RECORD.fieldNames          F1,F2,F3,F4
Please advise
Thank You and Best Regards
Fernand

>>but then how to make for example more then 1 records per message.
like 10 records per message. should i set RECORD,10 ?
That is right. Just try it out yourself.
@Shesagiri,
Number of record is decided by the parameter Recordset structure and number of recordset with in a message is decided by Recordset per Message.
Regards
Jaishankar

Select query for single record

hi all,
i am working for script which is letter.
i had to extract the fields as shown in the code. from two tables i had to extract single record. and pass this values to script.i developed the code. wen i execute the code its going to dump. can any body help me.
TABLES : VBAK, KNA1.
PARAMETERS : p_vbeln LIKE vbak-vbeln.
Data : name like kna1-name1.
Data: begin of wa_vbak ,
     vbeln(10),
       netwr(15),
       bstdk(8),
       kunnr(10),
       smenr(8),
       end of wa_vbak.
Data: begin of wa_final,
       vbeln(10),
       netwr(15),
       bstdk(8),
       kunnr(10),
       smenr(8),
       name1(35),
       end of wa_final.
       Select single vbeln
                     netwr
                     bstdk
                     kunnr
                     smenr from vbak
                     into wa_vbak where vbeln = p_vbeln.
    If sy-subrc = 0.
       Select single name1 from kna1 into name where kunnr = wa_vbak-kunnr.
    If sy-subrc = 0.
      Move wa_vbak-vbeln to wa_final-vbeln.
      Move wa_vbak-netwr to wa_final-netwr.
      Move wa_vbak-bstdk to wa_final-bstdk.
      Move wa_vbak-kunnr to wa_final-kunnr.
      Move wa_vbak-smenr to wa_final-smenr.
      Move name to wa_final-name1.
    endif.
    endif.
    write:/10 wa_final-vbeln, 20 wa_final-netwr.
thanks
venkat

Hai,
Make sure that field names(itab) are same as database tables.
And also INTO CORRESPONDING FIELDS OF TABLE .
The modified code is as below:
TABLES : VBAK, KNA1.
PARAMETERS : p_vbeln LIKE vbak-vbeln.
Data : name like kna1-name1.
Data: begin of wa_vbak ,
vbeln(10),
netwr(15),
bstdk(8),
kunnr(10),
smenr(8),
end of wa_vbak.
Data: begin of wa_final,
vbeln(10),
netwr(15),
bstdk(8),
kunnr(10),
smenr(8),
name1(35),
end of wa_final.
Data:
T_wa_vbak like standard table of wa_vbak.
Select single vbeln
netwr
bstdk
kunnr
smenr from vbak
into <b>CORRESPONDING FIELDS OF TABLE T_</b>wa_vbak where vbeln = p_vbeln.
If sy-subrc = 0.
Select single name1 from kna1 into name where kunnr = wa_vbak-kunnr.
If sy-subrc = 0.
Move wa_vbak-vbeln to wa_final-vbeln.
Move wa_vbak-netwr to wa_final-netwr.
Move wa_vbak-bstdk to wa_final-bstdk.
Move wa_vbak-kunnr to wa_final-kunnr.
Move wa_vbak-smenr to wa_final-smenr.
Move name to wa_final-name1.
endif.
endif.
write:/10 wa_final-vbeln, 20 wa_final-netwr.
Regards,
Rama chary.Pammi

Recurrent single record Insert problems

Hi,
we have on production environment a Java based application that makes aprox 40.000 single record Inserts per hour into a table.
We ha traced the performance of this Insert and the medium time is 3ms, that is ok. Our Java architecture is based in Websphere Application Server and we access to Oracle 10g through a WAS datasource.
But we have detected that 3 or 4 times a day, during aprox 30 seconds, the Java service is not able to make any insertion in that table. And suddenly it makes all the "queued inserts" in only 1 second. That "pause" in the insertion cause problems of navigation because is the top layer there is a web application.
We are sure that is not a problem with the WAS or the Java code. We are sure that is a problem with the Oracle configuration, or some tunning action for this kind of applications that we don´t know. We first thought it could be a problem with a sequence field in the table. Also, a problem when occurs the change of the redo log. But we've checked it with our DBA and this is not the problem.
Has anybody any idea of what could be the origin of this extrange behaviour?
Thanks a lot in advance.
Jose.

If by one single disk you mean a single physical disk then spreading the online redo lobs onto multiple disks would be a good idea. Log Group A should be on a seperate disk from Log Group B and the members of each log group should also be on seperate disks.
Note that Oracle does not stop processing when it does a checkpoiint. Here is basically how it works. When online redo LogA fills a checkpoint is signaled and Oracle immediately starts writing to LogB. The ckeck point is processed by ckpt and dbwr.
If logB fills and a switch to LogC occurs before the ckeckpoint signaled by the switch from LogA to LogB has completed then you get the checkpoint not complete information message. Oracle will continue processing though it has more work to do having to combine two ckeckpoints. If the problem continues then you get 3 and 4 checkpoints backed up. This will bog Oracle down as the buffer cache has to be pretty full at this point. You cannot read data unless you have a clean buffer to read into.
The checkpoint not complete message normally means your online redo logs are too small and are filling too quickly. In the absence of Data Guard configured to use log shipping you should look to see how many log switches you are doing per day. If is is more the 24 - 48 then increase the size of the online redo logs.
Database writer performance issues are the next area you would want to ckeck after properly sizing the online redo logs and verifying the IO performance of log writer. Pretty much all you can do to help log writer IO is put enough physical disk under the logs to spread the IO out.
If your problem is caused by the activity level during peak periods then setting MTTR probably will not help since dbwr will be busy writing anyway.
You really should run a statspack or AWR report to help clarify the issue.
If you determine you are switching online redo logs too many times and increase the size take a set of statspack both before and after you make the change. This will help you see the effect on the database as a whole.
HTH -- Mark D Powell --

Reverse auto-population to a shell attribute form/ single record instance BP.

Hi All,
I was wondering if anyone had experience in reverse auto-populating data from a BP within a shell to that same shell's attribute form or a single record instance BP like Project Information.
My use case is that I have 7-10 BP's that gain more definition and have data within in them that changes from phase to phase. This leaves a need for one place with the most current information, which I thought Shell Attributes or Project Information.
That being said as the information changes in theses 7-10 BP's throughout the life of the project the Shell Attribute form and Project Information should be updated via Reverse auto-population.
Can anyone tell me if this scenario is possible, or if there is another way that i can achieve this?
Your help would be appreciated!!!

Yes, it is possible and fairly simple. Make sure that the DEs from the various BPs that will be passed are on the form of the shell or single instance record. And make sure that integration is defined for those DEs on the shell or SI record. This is a common mistake, as people forget to define the integration and that is a must for reverse auto-pop.
Then, on the BPs passing the data, open the form from which they will be passed and look at the properties. Define the reverse auto-pop normally, SI records are always available for this, and shells should be too.
Sean

Single Record in Form

All,
I have an application which has a form where users submit accounts for review. It is working great and populates a table in my schema, but of course we need to have these accounts reviewed by my team. I am attempting to design a button that will pull a single random account that was submitted into a report/form (not sure which to use) so that my team can review them for errors and have an 'accept' or 'deny' button.
What would be the best method for designing such a review. The button would need to execute PL/SQL that would update the 'user id' column of the person clicking the button and thus "assigning the account to them" and display that single record in this report/form so that they can easily review it for approval. I have already written the SQL that will pull a random record from the table. The Accept or Reject buttons would insert a Y or N into the accepted column.
We are using Oracle 11g and Apex 3.2. I am just not the best at PL/SQL and not sure how to get such a button to launch this form.
On a separate note I have used JQuery and have seen the posts on how to create a form via JQuery but not sure how to bring in a random loan with by executing SQL.

Any detail form can be made into this "random form" . Say you have detailed form based on the submitted records table. Now it would have a hidden field holding the PK value. If you were to link to this page from a any other page setting this primary key field, it would automatically populate the form with the record details.
So if the button that "randomly" pulls out a record were to redirect to this page and set the PK field with some value, it would be fetched and can be processed.
Suppose you don't set the PK field, but rather add an "onload before-header" PLSQL process in the same that fetches a random PK value(among submitted records) and set the PK feild automatically. Any redirection to the form would then pick up and show a random record.
That process could be as simple as
BEGIN
SELECT <PK column> INTO :<page PK Item>
FROM ( SELECT <PK column>
               FROM <table>
               WHERE <filter critera>
               ORDER BY dbms_random.value
WHERE rownum=1;
END;So just a create a normal detailed edit form and add this process to the onload(make sure it runs first).
Change the button to redirect to this page.

Single record insert performance problems

Hi,
we have on production environment a Java based application that makes aprox 40.000 single record Inserts per hour into a table.
We ha traced the performance of this Insert and the medium time is 3ms, that is ok. Our Java architecture is based in Websphere Application Server and we access to Oracle 10g through a WAS datasource.
But we have detected that 3 or 4 times a day, during aprox 30 seconds, the Java service is not able to make any insertion in that table. And suddenly it makes all the "queued inserts" in only 1 second. That "pause" in the insertion cause problems of navigation because is the top layer there is a web application.
We are sure that is not a problem with the WAS or the Java code. We are sure that is a problem with the Oracle configuration, or some tunning action for this kind of applications that we don´t know. We first thought it could be a problem with a sequence field in the table. Also, a problem when occurs the change of the redo log. But we've checked it with our DBA and this is not the problem.
Has anybody any idea of what could be the origin of this extrange behaviour?
Thanks a lot in advance.
Jose.

There are a couple of things you'd need to look at to diagnose this - As Joe says it's not really a JDBC issue from what we know.
I've seen issues with Oracle's automatic SGA resizing causing sporadic latency in OLTP systems. Another suspect would be log file sync wait events, which are associated with commits. Don't discount the impact of well meaning people using tools like TOAD to query the DB - they can sometimes cause more harm than good.
Right now I'd suggest you run AWR at 10 minute intervals and compare reports from when you had your problem with a time when you didn't.

Delete a single record of specific Master Data

Hello Experts,
We have this InfoObject '0FUND' which is time-dependent. For a specific Master Data, exp: 40002, we have a list of records, that vary in 'Valid from' and 'Valid To'. Example:
FMArea__Fund____ValidTo____ValidFrom____Attribute...
CM______40002__31.12.2001__01.01.1000_____MOD
CM______40002__31.12.2002__01.01.2002_____MOD
CM______40002__31.12.2003__01.01.2003_____MOD
This master data as no transactional data in any InfoProvider. When we try to delete a single record, the system delete all the records for 0FUND=40002.
Is there possible to make a selective deletion, in order to choose a single record?
At same time, we delete from the InfoCube all transactional data assign to a specific Master Data in a particular year. Exp (0FUND=40022; 0FISCYEAR=2004). When we try to delete this specific record:
FMArea__Fund____ValidTo____ValidFrom____Attribute...
CM_____40022___31.12.2004__01.01.2004_____MOD
The BW System doesn't allow deleting this specific record. Why this occur, its necessary delete all the transactional data assigned to 40022 Master Data??
Best Regards,
IB

Hi IB,
Yes. It is necessary to delete the transacton data where the master data is referred before deleting the master data. This because the master data value in the Cube/ODS looks up at the SID table for the master data.
Bye
Dinesh

Double record make it to single record in Query for perticular period in BW

Similar Messages

Maybe you are looking for