UDC - TeraData NUMC mapping...

Greetings!
We are trying to extact data from TeraData using UDC.
Getting an error mapping a CHAR field in TeraData to NUMC in BW. Changing the TeraData field to Integer is truncating the leading zeros.
Any thougths on this mapping without losing leading zeros?
Thanks
Venkat

Hi,
well I am not sure about the information stored in the source field. But 0fiscper normally accepts values in the format YYYYMMM. So you normally need to convert it into the sap internal format with leading zeroes just for the month. --> January 2006 will be 2006001.
Siggi

Similar Messages

Map Datasource field NUMC to Cube Key Figure

Hi ,
How do we go about mapping a NUMC field from the datasource to a Key figure(DEC) in an InfoCube?

Hi Ravi,
Thanks for your answer, first i unmapped in transfer rules as per your sugession and executed the infopackage but the data has come when i was unmapped in update rules the data has not come.
My doubt is when i was unmapped in TR even though the data has come, so what is the use of unmapped here.
and how to unmapped char
please clarify my doubt
David
Edited by: david Rathod on Jul 22, 2010 5:40 PM

Correct mapping of NUMC

Hi,
After upgrade to SAP_GWFND 740 SP7 we are getting some errors due to mapping of NUMC to edm.string. Can anyone tell me the correct and recommended way of mapping NUMC to an EDM datatype?
From the best practices here OData Best Practices - SAP NetWeaver Gateway Foundation (SAP_GWFND) - SAP Library I can tell that mapping to decimal and int is not recommendend. In SAP note 1977590 it is described that mapping to string will give error whilst mapping to decimal might only give a warning. Also in this post NUMC to Edm.Decimal or Edm.IntXX per best practicesfrom SP8 - But how? it is described that a warning is the outcome. But I can't find anywhere describing the best way to map NUMC.
Can anyone guide me?
Best regards
Jan

Hi Ashwin Dutt R, Chandrashekhar Mahajan and others,
The problem is that when we try to use edm.string with max length 8 as shown below for mapping numc of length 8 we get the compile error below. Do you think it is a product error -> OSS?

Teradata and UDI (UDC Interface)

Hello BW and Teradata Gurus,
I am at a customer who intends to use Teradata as a DW along with BI. My understanding of their architecture is that they will use UDC Interface is to extract info from Terradata into BW. This will be as part of their reporting Interface, which they intend to acheive via the use of Vitual (Remote) InfoCubes.
LIke I mentioned they are planning on querying on information in Teradata based on query definition based on Remote Cubes, which could be part of a MultiProvider. To me, that stands out as a significant issue since I do believe that there are performance restrictions (data volume and number of concurrent users) when using remote cubes for querying information.
I do not have much exposure with either Teradata or UDC Interface. Would any of you have information regarding UDC Interface and how it interacts with Teradata. Any information you can provide regarding UDC Interface and SPECIFICALLY its impact on extraction and/or Query Performance especially with Teradata would be extremely helpful.
Thanks & Best Regards,
Rajesh

Hi,
The main reason to integrate with Teradata is a faster database. We load data from BW to Teradata and use a virtual cube based on a view in Teradata to report on the data (using TVAS). For us it was a relative quick win since we were planning on using the demand chain management solution from TD. Since we had to load a lot of data to Teradata anyway, why not use it for reporting.
There are some pros and cons to this method but overall we are happy with the solution. Teradata recently announced however that the TVAS method is no longer neccessary since you can run BW directly on a Teradata database. Check out the Teradata website for more information on this.
Hope some of this helps
http://www.teradata.com/t/assets/0/206/280/8356eed7-bfc1-403e-8d1c-e306fdba7a8a.pdf
http://www.teradata.com/t/assets/0/206/308/ebe4dcd4-52d7-4577-9efa-65afd6a6e63e.pdf
http://www.noticias.info/archivo/2005/200506/20050602/20050602_71599.shtm
http://www.groceryretailonline.com/article.mvc/Teradata-Completes-SAP-BW-35-UDC-Integration-0001?VNETCOOKIE=NO
http://www.encyclopedia.com/doc/1G1-145783747.html
Thanks & Regards,
Ravi.

ABAP Data type to XI data type mapping

Hi,
I am creating a message type in XI corresponding to a ABAP structure. There are couple of data types on the ABAP side (for example: RAW, QUAN, CURR CUKY, UNIT, NUMC etc) which i have no idea on what it should be mapped to on the XI side(xsd:string, xsd:token, xsd:int etc).
Is there is any document on mapping from ABAP data type to XML data type ? if yes please send it across.([email protected])
if no doc available, if some body could let me know what these ABAP data type would map on the XI side it would great.
Best Regards,
Kiran

Hi Kiran,
just some more background information to the data types CURR, CUKY, QUAN, UNIT:
1) Each field of data type CURR must be assigned to a field of data type CUKY (the financial amount must be assigned to a currency) and each field of data type QUAN must be assigned to a field of data type UNIT (each quantity must have a dimension). If you want to know this assignment for your specific ABAP structure, you can find it using transaction SE11 with your ABAP structure and go to the tab Currency/Quantity fields. Here you find the information under Reference table and Ref. field.
2) The default number of decimals for a field of data type CURR is 2, but this can be overwritten by the assigned CUKY field. You can check the actual number of decimals for a specific currency with transaction OY04. Here are all currency list that do not have 2 decimals.
3) The default number of decimals for a field of data type QUAN is 3, but this can be overwritten by the assigned UNIT field. You can check the actual number of decimals for a specific unit with transaction CUNI. Click on Units of measurement (pick the correct dimension first), double click on your unit and you find the info in the field Decimal places.
4) Depending on the data you receive your conversions can get quite tricky. Lets take as an example a field of type CURR. If you find the value 1000 in there, this would convert to 10.00 USD if the assigned CUKY field contains the value USD (US dollar). But the same value ca represent 1000 JPY (Japanese Yen) if the assigned CUKY field contains the value JPY. What i try to say is that the value in the field of type CURR alone does not give you the actual amount, you have to know the value of the CUKY field as well. Similar for QUAN and UNIT.
Regards,
Claus

Mapping problem Moving Complex Type to field

Dear all,
I am making a message using from RFC(ECC 6.0) to SOAP
Some fields of the structure of RFC are:
STRUCTURE_VALUE -> Complex Type
      END    - NUMC 20
      ID        - NUMC 3
      COST - NUMC 5
      RATE - NUMC 6
The structure of Data Type to SOAP is:
DATA_VALUE - STRING
I need to concatenate the first structure in the second structure but I have too many fields in the STRUCTURE_VALUE and I don't want to use the function CONCATENATE in the MAPPING EDITOR. I would like to move the complex type STRUCTURE_VALUE to structure DATA_VALUE.
Does anybody know if it's possible?
Best regards,
Fernando

Hi,
First of all what needs to be concatenated ? Each fields of the Source segment ?
If so write a user defined function, which reads all the fields and retruns one concatenated field as an output.
To know about this , you can go thru these- to understand the way how it works-and for e.g
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/be05e290-0201-0010-e997-b6e55f9548dd
Rgds,
Moorthy

Data Type NUMC from R3 to BW

Hi Expertts,
The datatype NUMC holding a value 00000 in R3 becomes
SPACE or NULL by the time it reaches to PSA. Is there
a way to prevent this from happening. I want the value
intact when and the same when it reaches the PSA up to
the cube.
Thanks for your reply.
Best regards,
Rose

hi Rose,
try to use 'domain' in r/3 without 'convers. routine' = ALHPA if it's your customized extraction.
you can double click the field used in r/3 and then double click 'domain'.
or check if remove ALPHA from conversion routine in bw infoobject used to map to that field.
hope this helps.

Mapping idoc MBGMCR03 , where are these fields?

Hello,
Trying to pass flat file dat to idoc MBGMCR03 to post MIGO. I can not map following fields in any segment. Any help will be appreciated.
sap/table     field
GODYNPRO     ACTION     CHAR     3     N     Type of action
GODYNPRO     REFDOC     CHAR     3     N     Type of action
GOITEM     BWART     CHAR     3     N     Movement Type
GODYNPRO     RESERVATION_NUMBER     NUMC     10     N     Reservation Number
Edited by: Emir Sirbegovic on May 6, 2010 9:08 PM
Edited by: Emir Sirbegovic on May 6, 2010 9:09 PM

E1BP2017_GM_ITEM_CREATE-MOVE_TYPE (movement type)
E1BP2017_GM_ITEM_CREATE-RESERV_NO (reservation no.) (RES_ITEM, RES_TYPE)
Depending on kind of posting (movement type) you have to fill E1BP2017_GM_CODE-GM_CODE.
Best Regards, Dirk

Issue in loading specific columns from a file to teradata table using IKM

Hi,
Can any one help to resolve the issue in loading specific columns from text file to teradata table.
i tried using IKM file teradata and columns are getting displaced.
my requirement suppose i have 5 columns in file and i have to load only 3columns to table using IKM.
same thing can be achived using LKM file to teradata but i want use IKM.
please suggest me on this
Regards
Vinod

Hi,
I believe that the problem you are having is that you have a delimited file, of which you want to pick columns from position 2,3,5. In this case, ODI will pick the first 3 columns of a delimited file regardless of position.
For example, if you a tab delimited file with c1,c2,c3,c4,c5 columns, and you want only columns c2,c3,c5 - when mapping these in an ODI interface, and executing, you will actually pick up the data from c1,c2,c3 as these are the first three columns in the file (reading from left to right). You can ignore "columns" on the right hand side of a file, but not the left. E.g delimited file with c1,c2,c3,c4,c5. Only pick columns c1,c2 will give you data for the first 2 columns
Create a temporary table to load all the data from the file, and use you temp table to extract the data you require. Or you could get the file created with the first three columns as the columns you require.
Cheers
Bos
Edited by: Bos on Jan 18, 2011 1:06 PM

Can a measure map to multiple fact tables

Hi Experts,
Can A Measure in an Universe map to multiple fact tables? For example, we have sales detail table by day and store and product, and have sales summary table by day and region and (product) department. We like to create a measure to map to both detail table and summary table. When I run query includes store and product, the query will automatically use the sales detail table; and query can automatically choose the summary table when it only includes region and department. Our database table could be in Teradata and Oracle. Can we achieve it and how? If a measure can't achieve this (dimension awareness), what the other option we can have and how to do it?
Thank you,
Richard

Hi Sion,
After I make the tables incompatible with the dimension objects on different levels, the measure can point to the right table if my query only include Store and Product or only include Region and Department. But in some scenario, the business could ask to include both Store and Region or Item and Department in same query. The current approach doesn't work on this further request. In our database, we have store table including Region with other higher level organization information, Product table including Department with other higher level product information, and we also have separated Region table and Department table. Could we make the objects more flexible u2013 query will be against aggregation sales table if dimension objects in query are only on higher level and will against detail sales table if some of dimension objects are on high level and some on low level?
Many thanks,
Richard

Char NUMC(5) - need to store value at database level without leading Zeros.

Hi All
We have 0Mat_Item like infoobjects in SAP BI, which have data type Char NUMC(5). Here if u load value '1', it would show us '1' at front end (BI or reporting level) but at database level stores as u201800001u2019.
My requirement is to store this value without leading zeros at database level. I wrote code in transformation u201Cshift u2026 left deleting leading u20180u2019 u201C & could load the data into New data table and at database level as I was getting result without leading zero. But once I activate the request, it gives SID generation error u201CValue '1' of characteristic 0ME_XCHRATE is not a number with 000009 spaces.u201D
How to solve this issue so that I could store u20181u2019 value instead of u201800001u2019 value in the database & could able to activate in DSO.
Thanks....

Hi.....
System may be storing the data in that way to maintain consistency.
Alpha conversion is used to store data consistently. It does this by storing numeric values prefixed with 0s Eg. If you have defined a material as length 6 (of type Numc) then material number 1 is stored as 000001 but displayed as 1; this removes inconsistencies between 01 vs. 001.
U can add the conversion routine for this Infoobject in either below way...
In the BW side...in the Datasource Level
In the datasource >> Go to fields tab >> there search the R/3 field which is mapped to that Infoobject in the Transformation >> Then in the Conv.Routine column select the Alpha Conversion Routine and In the Format column select External....
In the Transformation
Write a field level routine for that particular field abd call the Alpha conversion routine there...
In the R/3 side
In the Customer exit ....Go to SE37 >> EXIT_SAPLRSAP_001 >> Double click on ZXRSAU01 >> search the datasource......write the code there....
Hope below link helps you.
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/7108a690-0201-0010-4cbc-9bca94f9ad9c?quicklink=index&overridelayout=true
Regards,
Vinay.
Edited by: Vinay Joshi on Mar 10, 2011 4:54 AM

Mapping time field issue!

Hi,
i have a time field in R/3 which is of length 4 ( numc ).
and in BW side it was decided to be a key figure. so i had created a key figure of type time.
( as i heard key figure time type has length 8 )
is there any problem if i could map these two fields in transfer rules!
Thanks,
Ravi

Hi,
Assuming your R/3 field is hhmm
it will depend what you have chosen in the data type of your key figure:
- if TIMS you'll have to add two 00 at the end of your field.
Routine: CONCATENATE TRAN_STRUCTURE '00' INTO RESULT.
I don't think you will be able to calculate times in queries with this one.
if DEC: then it represent the number of seconds between 0 and your value.
You'll have then to write the code:
DATA: lt TYPE TIMS.
CONCATENATE TRAN_STRUCTURE '00' INTO lt.
MOVE lt TO RESULT. "ABAP will calculate the number of seconds accordingly
hope this helps...
Olivier.

Xml driver used with a teradata external database

hi everyone,
I would like to use the db_props feature within the xml driver to create the xml schema inside a teradata database. With some oracle examples, the JDBC URL looks like this :
jdbc:snps:xml?f=../demo/xml/xxx.xml&s=XXX&dod=true&iue=true&dp_driver=com.ncr.teradata.TeraDriver&dp_url=jdbc:teradata://localhost/TMODE=ANSI,CHARSET=UTF8&dp_user=xxx&dp_password=xxx&dp_schema=xxx
And this returns the following error :
java.sql.SQLException: Unsupported technology (Driver: com.teradata.jdbc.TeraDriver, Url: jdbc:teradata://localhost/TMODE=ANSI,CHARSET=UTF8)
Do you know how I can solve this ? Thanks

Hi,
For a customer Entity Framework integration, it is beyond the scope of our support.
This forum is related to the current versions of the ADO.NET Entity Framework and LINQ to Entities including object-relational mapping and entity data modeling.
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

TERADATA and UDConnect object

dear all,
i want to create a datasource which connects to TERADATA and query a table via UDConnect but i'm facing a slight problem here.
Under my extraction tab which requires me to input UDC SourceObject, i key in the TERADATA table e.g. TABLE1 and when i click on the proposal tab, i would expect to see a list of columns, but this is not happening. any idea?
anyway, some might suggest that TABLE1 is not in the system, well, a few pairs of eyes inthe office confirm that it is there.
any idea friends?

dear all,
i want to create a datasource which connects to TERADATA and query a table via UDConnect but i'm facing a slight problem here.
Under my extraction tab which requires me to input UDC SourceObject, i key in the TERADATA table e.g. TABLE1 and when i click on the proposal tab, i would expect to see a list of columns, but this is not happening. any idea?
anyway, some might suggest that TABLE1 is not in the system, well, a few pairs of eyes inthe office confirm that it is there.
any idea friends?

Info object as Data type as NUMC, master data doesnot store value as 0

Hello All,
I have an Info Object whose data type is NUMC.
It has only text Master Data.
Id and its description are the two fields of that Master data.
Now when below data is feeded in to that infoobject through transformation which has 1 to 1 mapping.
ID              Description
0                Submiited
1                 Approved
2                  cancelled
The data gets uploaded in text table but.
0 ID values turns out to be "       "    blank.
Which is wrong.
Result in Text Master table:
ID                Description
                    Submitted
1                  Approved
2                 Cancelled
if the ID is 0 it should be populated as 0 for "submitted"
No start routine our end routine is ther in the transformation.
Kinldy help me to solve this issue.
Points will be assigned

Hi,
When an InfoObject is created it'll have a blank record, in case of Numc it'll be as shown below:
ID Description
0   Blank
But this 0 gets displayed as blank as well.
When you post transaction/master data you can post it as
0 for Submitted
1 for...
2 for...
When you do reporting on this & if you only select text display only it'll display "Submitted", which is required. But if you display ID & text or only ID. It'll display # instead of 0.
See if you can replace this # with 0.
Mean while i'll be trying to get this solution. I'll post soon if i get the solution.
Regards,
Iliyas

UDC - TeraData NUMC mapping...

Similar Messages

Maybe you are looking for