How to create a ClobDomain object with a default's different charset.

Question

Hi,
I'm developing a 10.1.3. ADF Faces application that need to upload a File to a CLOB column. The user will usually upload a text file CP1252 charset, but in samples like this:
// read data into a character array
char[] data = {'0','1','2','3','4','5','6','7','8','9'};
// write the array of character data to a CLOB
writer = ((CLOB)my_clob).getCharacterOutputStream();
writer.write(data);
writer.flush();
writer.close();
The writer object is always a defaults charset writer.
Does anybody knows how to do this?
Thanks,
Edited by: Samuel Fisch on 05/01/2010 02:06

Answer 1

Samuel,
My understanding is that unicode should be able to hold all possible characters codes. So you upload something in unicode, store it in the db and later on view it from the db you should see the contents of the clob (in the default encoding of your pc).
The question for me is if the db can store all character codes. this is not the case if you use an oracle db and have not set the db character code to unicode (i.e. utf-8). In this case all characters of the input stream which are not part of the db character set are mapped to <b>one</b> special character code which you can't read. When you read the clob back you see some strange characters.
This is not the problem of writing/reading the clob but a problem with the db character set.
If you load data to the clob from a system using windows-1252 character set and this data contains some special characters like 'öäü', the db character set is set to utf-8, you should be able to read the clob data on a system using iso-8859-1 character set (or any other which have 'äöü') and see the special characters. That's the reason the reader/writer uses the system default character set.
If you see a different behavior, describe where you see the error or wrong characters.
Timo