DataStage/QualityStage 8.0 : Support for CJK languages

[byk]

Please help me understand the support for CJK (Chinese-Japanese-Korean) languages.
If you have any ibm website link for the same, please let me know.

Thanks

[ArndW]

In order to understand how DataStage NLS uses CJK, you need to start off with understanding multibyte implementations at a system level; the rest is just an implementation question. There are some excellent reference works available (hardcover ones) that provide a very solid foundation in the basic concepts and methods.

[byk]

Thanks. What I a more interested is to know it from DataStage perspective. For e.g.

1. Any special settings (apart from NLS) in DataStage
2. Known issues where certain scripts are not supported etc.
3. Any special consideration for stages of QualityStage

[ArndW]

I can't think of anything offhand that isn't in the manuals and don't have recent QualityStage experience. We certainly had a number of teething issues with Japanese, Chinese and Korean implementations years ago but the installed bases and experience levels in those languages is quite widespread/high now so any inherent problems will have been dealt with.

[ray.wurlod]

There is a DataStage NLS manual. Read that. The CJK characters appear in each of the appropriate Chinese, Japanese and Korean character maps.

Note that server jobs use UV-UTF8 while parallel jobs use ICU (a "true" 16-bit implementation of Unicode).