Datastage 11.5 RCP Oracle 12c Varchar field issue.

chulett · Post by **chulett** » Tue Jun 14, 2016 12:08 pm

So... as a total SWAG on my part... is there a way to include the extended "Unicode" property in the schema file?

Childrensmn · Post by **Childrensmn** » Tue Jun 14, 2016 12:43 pm

Hi Criag,

Thanks for the reply.

Yes in the schema file we can define it. Instead of "string" as the datatype if we define the datatype as "ustring", that will just act as extended "Unicode" for the varchar.

But if we have 200 staging tables, i will have to define 200 schema files.

And if i have to use the schema file, than i will have to go with stages like sequential file and i am not sure if sequential file stage will be handle the history load.

I am guessing it has to do more with the NLS, but just don't know what NLS setting i have to use.

JRodriguez · Post by **JRodriguez** » Tue Jun 14, 2016 4:55 pm

Check if the schema used at run time have all fields as ustring, if yes then add an environment variable to the DS project name it "NLS_LANG" and set it to "AMERICAN_AMERICA.WE8MSWIN1252", add it to the DS job

Let us know how it goes

Childrensmn · Post by **Childrensmn** » Wed Jun 15, 2016 10:06 am

Hi Julio,

Thanks for the suggestion, I did incorporate the same and it did work.

Added NLS_LANG to user defined environmental variables in DS Admin Client. Used the parameter in the job with the value set to "AMERICAN_AMERICA.WE8MSWIN1252" and ran the job. I see a perfect match between source and target data.