One of the fields in canonical document that i am routing to TN contains a chinese character. The processing rule calls another service(say serviceB) which updates/inserts the record in oracle DB. When serviceB get canonical document from bizdoc, this field having chinese character gets corrupted and I see question marks in the field.

The content part when viewed from TN console still shows a valid value in the field.

Appreciate if some one can give me a clue to fix this problem.

Thanks.
Jiten

i found the solution myself…

when converting bytes from bizdoc to string, set encoding type=UTF-8

Hi Jiten,
Just my thoughts on this…
Using a fixed encoding in the implementation may only be a temporary solution.
We have unicode and non-unicode SAP systems and it is always a problem with special characters when we use any specific type of charset.

For unicode sending systems “UTF-8” saves the day, but the same UTF-8 makes life difficult for data sent by non-unicode systems.

A generalized solution where our implementation should use the same encoding used by the sender system may be ideal. (As all the partners do not agree to use one encoding and its also not possible). But how to come up with such a solution? Any ideas on this?

Cheers
Guna
http://www.nibodo.com

是中国人吗？

Philos…

这不是王宇吗，好巧啊，你最近还好吗？