DSXchange

We are facing Japanese Character issue in DataStage during data load. We are getting EBCDIC file as a source containing Japanese (Hirakana) data and while we are loading the data into a flat file we are getting Japanese characters but in some other format (Katakana). We have tried all possible code pages (NLS code map) that are available in DataStage that support Japanese Character conversion. But we are not able to resolve this conversion problem.

Can someone help ?

The Hiragana and Katakana code points are distinct, and there are even differences in how many characters each has. Thus it is unlikely that a hiragana "Ha" is transformed to a katakana "Ha".

You need to get information from your EBCDIC Japanese source as to exactly which character set is being used. There are numerous possible source encodings and your problem is that you are just getting the wrong one, which can result in incorrect character conversions.

We are aware of at least fourteen different ways of encoding Japanese characters. As Arndo-san noted, you must ascertain precisely how these data are encoded.