DSXchange

Hi Martin,

I think you need to check your job. If the source and target has the same decimal type then it shouldn't be problem to process the data. Check whether precision and scale is correct in source and target. also check whether you are converting the source decimal type to sting in your tranformation and in the Aggreator Stage check whether you are using the right field for grouping.

Thanks
Rasi

I notice that you're using the local German convention for the character that identifies the decimal place ("2555,5"). This suggests to me that you may be using NLS.

If you are using NLS, make sure that the LANG setting, the Oracle language settings and the DataStage NLS locale settings are all compatible.

If you are not using NLS, you may have some success by converting the "," character to "." - for example Convert(",",".",Preis)


Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518

Ralf,

DataStage is an anglo-american Tool an when working within this tools you should see, that you keep everything anglo-american. Generally reading Databases with native driver or ODBC and the correct metadata imported and writing to database need no concern from you, but if you read from sequential files which have continental european signs like comma as decimal sign you have to first convert them into anglo-american. If you want to have continental european signs in currency, dates, timestamps and numbers you have to make the conversion.

The NLS-stuff within DataStage is something you have to concern when working with japanese, corean, russian or chinese. For central european purposes there are only a few things to look at and if you have understood this few things DataStage "frisst dir dann aus den Fingern".

Wolfgang

Gelegentlich das Problem zwischen die Tastatur und der Stuhl ist. [:D]