[EMAIL PROTECTED] wrote:
What are the consequences of using a database that does not uses UTF-8
encoding ? Or a container that's not configured to use UTF-8 encoding ....
For the database it will start making a difference if you start using
characters outside of the subset of UTF-8 that is in common with the
encodings in your database character set. You'll get question marks (?)
or other garbling of the text data. (I'd suggest NLS_CHARACTERSET
(database character set) to UTF8 for Oracle; you probably also want to
set it to use CHAR length semantics). Both are best done at CREATE
DATABASE time if you can.
If you can't set it database wide you may be able to change the schema
to use all NVARCHAR2 types and then define the NLS_NCHAR_CHARACTERSET
(national character set) to UTF8. That seems pretty painful to me, and
I really can't recommend that route, but it may be possible.
If the application container isn't setup properly for UTF-8 encoding
you're also likely to see text garbled at various points where we do
expect UTF-8 and we encode/decode specifically for that.
