How is Default Char encoding for WebLogic determined ?

Question

How is the defualt char encoding for a Weblogic server determined?
We have two environments each running Weblogic 5.1 sp8 on Solaris 7. One is production,other
is for testing.
For testing german characters I use a TestGlobal.jsp to read parameters using
servlet API. From the same browser I access this URL
http://www.xyz.com/TestGlobal.jsp?userName=Tschüß.
In one I get the param Tschüß correct in the other I get Tschýý
I know for sure both the properties file are exactly the same.
I also know that I can work it using the new char encoding property for western
european alphabts. I want to know how does this work without setting the char
encoding prop. How is weblogic recognising Tschüß without this prop being set.
Does the underlying OS make the difference ?
THanks
Atish

Answer 1

In this specific case there are no differences for code points storing character data because used character set is the same.
But what is your Oracle 4 digits version ?
Are you sure that database character set and national character set are the same ?
In recent Oracle versions, database character set and national character set are different. For example:
SQL> select * from nls_database_parameters where parameter like '%SET%';
PARAMETER                      VALUE
NLS_CHARACTERSET               AL32UTF8
NLS_NCHAR_CHARACTERSET         AL16UTF16Edited by: P. Forstmann on 28 sept. 2011 18:51