Arabic Character issues in Sequential File

K.A.R.PAULSON · Post by **K.A.R.PAULSON** » Mon Jun 19, 2006 4:13 am

Hi All

We have a problem in Arabic Characters while processing the data from Oracle 9i table to Sequential File, but the same Job was success full when we were processed between Oracle to Oracle tables. What will be the problem? If the NLS setting installation then please let me know where I have to look into it and how to trouble shoot this issue.

The NLS setting we are using ISO8859-6, MS1256.

We have a serious issue on this any one help it will be great.

Thanks and Regards

Regi Paulson

ray.wurlod · Post by **ray.wurlod** » Mon Jun 19, 2006 4:16 am

Can you be a little more specific about the nature of the problem? Can you also advise the NLS parameter values from uvconfig, from the project (as set in Administrator) and from the job itself, in particular in the Sequential File stage?

chulett · Post by **chulett** » Mon Jun 19, 2006 6:12 am

You problem may also be the NLS setting of the tool/platform used to view the sequential file. I just went through this recently, the file was fine but my pc's NLS setting was 'wrong' and so it looked... bad.

ray.wurlod · Post by **ray.wurlod** » Mon Jun 19, 2006 4:52 pm

The "blah-CS" maps are for client/server communication. Can you try setting the map in the Sequential File stage to an appropriate UNIX map for Arabic? (If your server were a Windows server, the appropriate map would be MS1256-WIN2K; however, the Microsoft code pages are not appropriate for UNIX servers. Research the appropriate map name to use in the NLS Guide. You may need to build/load this map using the Administrator client.)

K.A.R.PAULSON · Post by **K.A.R.PAULSON** » Mon Jun 19, 2006 10:38 pm

Hi

I can not get you, can you please explain me in details.
For your kind information
DataStage Server install in Unix Box
Client DataStage install in Windows 2000 Professional

Thanks and Regards

Regi Paulson

ArndW · Post by **ArndW** » Tue Jun 20, 2006 12:35 am

So your DataStage is on a UNIX machine, and you are writing to a text file on the UNIX machine and also viewing it there, yet you specified

NLS Setting MS1256-WIN2K-CS

for the file output?

K.A.R.PAULSON · Post by **K.A.R.PAULSON** » Tue Jun 20, 2006 1:37 am

Exactly you are correct.

Thanks and regards

Regi Paulson

ray.wurlod · Post by **ray.wurlod** » Tue Jun 20, 2006 3:46 am

The point I was trying to make is that MS1256-WIN2K-CS is (clearly) a map intended to be used on a Windows machine. You need to find out what the appropriate map name to use on a UNIX machine is.

Amos.Rosmarin · Post by **Amos.Rosmarin** » Tue Jun 20, 2006 4:29 am

Hi,

If you work on unix, use ISO8859-6 as your default map.
I assume your client machine supports Arabic.

It suspect the DS job is ok but the view-data is corrupted. I suggest you create a demo job that wrties all the letters into a sequencial file and then look at it using the

Code: Select all

od -cx  file_name

command . this will show you the hex value and you can see what you realy have in the file.

I often have the same problem with Hebrew (1255)

Good luck
Amos

K.A.R.PAULSON · Post by **K.A.R.PAULSON** » Wed Jun 21, 2006 5:45 am

Hi All,

Thank you very much to helping me with your view and ideas, finally I had resolved the issues. I had resolved this issue with enabling MS1256-WIN2K NSL setting in DS which is comfortable in Unix NLS settings for windows version.

Once again thank you very much for all of you.

Thanks and Regards

Regi Paulson

DSXchange

Arabic Character issues in Sequential File

Arabic Character issues in Sequential File

Arabic Character issues - Resloved