Problem with Polish characters in Paralle job

sivatallapaneni · Post by **sivatallapaneni** » Mon Sep 14, 2009 5:02 pm

I have a problem with one of the job that has CFF stage as source and oracle stage as target. The EBCDIC file has polish characters. NLS is installed on the server and Paralle maps are set to project default ASCL_ISO8859-2 character set which support polish characters.

CFF Stage and oracle stage has project default NLS map set. Did anybody run into the simila issue? what character set should I use if ASCL_ISO8859-2 charaxter set is not right.

Appreciate all the help.

chulett · Post by **chulett** » Mon Sep 14, 2009 5:45 pm

Since you didn't explicitly say, what character set does your target Oracle database use?

sivatallapaneni · Post by **sivatallapaneni** » Tue Sep 15, 2009 7:36 am

Database it self is set to UTF16, But the stage in the job has project default ASCL_ISO8859-2 NLS map.

saraswati · Post by **saraswati** » Tue Sep 15, 2009 8:11 am

can you set the nls setting of stages to the one you have in your database

sivatallapaneni · Post by **sivatallapaneni** » Tue Sep 15, 2009 8:27 am

I tried that too with no luck

ArndW · Post by **ArndW** » Tue Sep 15, 2009 8:33 am

make a job copy with just the CFF and writing to a sequential file. Does that correctly display the Polish characters? DON'T use "view data".l

sivatallapaneni · Post by **sivatallapaneni** » Tue Sep 22, 2009 7:16 pm

I tried this and polish characters are not coming right. I opened case with IBM and they said I need to check with my source system because I'm getting hex characters 9A and 9B. I know they are not going to give me a solution, We do have a work around, it is to read the file in server job using UTF8 and write it to another file and then use that file in parallel job. I dont like this solution because I might need to modify lot of jobs and we want to avoid using server pallet.

Appreciate if somebody can provide an alternative.