Date format in flat file

[Former Member]

Hi all:

I want to create a flat file to load data into an IC.

i have cal month data in my flat file. i am having the format as:

200508

200509

200510

200511 and so on.

i used 0calmonth as my time characteristic in my Infosource and i have a flagged for conversion routine also.

but when i tried to preview in the Infopackage its not showing up. it says error: error 1 when loading data some times and also pern1 error also.

i think its the problem with the date format i am using in myflat file. can anybody please tell how to maniatin the format for calday/ calmonth/ calyear data in an flat file..

Thanks

kedar

[former_member619464 · Accepted solution]

Hi Kedar,

Format for 0calday in flat file (yearmonthday):

20051201

20061102

Format for 0calmonth in flat file (yearmonth):

200512

200611

I think the problem is not with ur flat file data. It may be due to the format in which it is loaded into ur cube. I mean to ask whether ur flat file is .csv file? If so, whether u ve chosen the settings correctly in the "external data" tab in ur infopackage?

Hope it helps....let me know

Regards,

RV.

[Former Member · Accepted solution]

Hi RV:

thanks for the reply.

i exactly did the same. 200508 and so on for a calmonth.

and ofcourse its a CSV file(saved file) and also i choose the csv in the external tab.

well, do u have any other alternatives to do the same.

thanks

kedar

[edwin_harpino · Accepted solution]

dear Kedar,

0calmonth in csv must have format MMYYYY,

and 0calday is YYYYMMDD.

and check your 0calmonth infoobject conversion routine,

it should PERI6, since your error message is pern1 ?

also try to check the infoobject sequences in 'transfer structure' (be careful can different from what you see in transfer rules infoobject mapping), should same as csv file. and try to save csv for just 1 or 2 records.

the other thing check if header for columns used in csv and 'ignore number of rows' in infopackage maintained properly. any infoobject with conversion routine pern1 used ?

separator/delimeter, decimal point, thousand separator etc check as well.

hope this helps.

[Former Member · Accepted solution]

Hi A.H.P:

thanks for the inputs. well, to be honest, i just changed the format and i got it done now. i mean for month::: 082005 and so on....

i did the same settings and got the data in IC and now i shd do the export DS settings from one one cibe and other.

but, in ur doc,(Achieve More Flexible Reports Using Time Distribution in Update Rules ) it says to to go to UR and change the time to time distribution and choose my ocalweek over there. but A.H.P, to my surprise, that time distribution radio button is grayed. i cant select anything.

i am stuck!!!!!. what to do now?

Thanks A.H.P

kedar

[edwin_harpino · Accepted solution]

dear Kedar,

sorry, you cant use 0calweek, it's not a 4 weeks in one month, but basically 0calweek is 52 weeks, has less granularity from 0calmonth.

please check in system if there is calweek2 something.

hope this helps.

[Former Member · Accepted solution]

Dear AHP,

I have a doubt about the conversion exits. Exactly what is conversin exits and how it wil work during data uploading. And exactly what happens when the data uploading from source system to PSA( if possible give some doucments about this ). I knew no transfer rules and updates rules are applied prior to this thing. Then how the data is manipulated when it reaches to PSA. I mean to say does any exits will work during this upload.

so,far i believe in one concept like that psa data should match with source system data ( is it not).

Could you pls give me some documents about conversion exits and how it will work....

Thanks & regards,

suresh

[Former Member · Accepted solution]

Hi suresh yadav,

Conversion exits are used mainly to transfer Source system representation of data to BW representation and Vice Versa..

eg:

say this is a source system field

EmployeeNumber: NUMC(7)

In R/3 Source system A: you may have it like 0000023

In Source system B:you may have it like 45 only for a 7 digit integer number

so when you load data to BW.. all the above different formats will be converted to one Single Internal BW representation like 0000023, 0000045.

this may be the case for any kind of data types that SAP supports between source systems and BW system.

hope i answered your question, and if you need more info on the conversion exits.. just give me ur mail id. i will send u a document on that..

Dont forget to assign points if u find it helpful..

bye..

kishore

[Former Member]

Kedar

I did flat file update long back try to do 082005, 092005 and so on.

I am not sure whether it helps or not

Thnaks

Sat