Samir Mehrotra wrote:
> Q. As i am working to provide
> Internationalization support to an already developed software
> having millions of lines
> of code and developed using Oracle 8 at the backend, the idea
> is to make minimal changes to the database as it would effect a large
> number of Screens, reports, procedures etc.. Is it really
> possible to support multibyte character sets such
> Korean/Japanese/Chinese
> etc without changing the width specification of the
> columns..if yes then how?
You don't necessarily have to change the size of your database fields.
English messages translated in Chinese and Japanese (also Korean, I think)
tend to be much shorter (in number of characters) than the original,
probably because ideographs and syllabic letters "convey more information"
than alphabetic letters.
Using CJK character sets (GB, Big-5, JIS, etc.) this was balanced by the
fact that each CJK character needed 2 bytes vs. 1 byte of English, so the
amount of storage needed was about the same.
Now UTF-8 uses 3 bytes to represent each CJK character, so the situation is
not as easy as it used to be.
However, there is not only one way to translate a phrase from a language to
another: if your translator is aware of the size limits for each field,
he/she could be clever enough to fit in the available space by carefully
choosing words.
_ Marco
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:59 EDT