language character set confusion

Have personal oracle 8i and am running it on win98.

If I create a table in my database that has a attribute with the nvarchar2() datatype and I through toad or sql*plus try to insert into this table I receive a "ORA-12704: character set mismatch". I don't understand why.

If I create a table with an attribute of type varchar2() I have no problems at all with inserting things into it. I don't understand why.

I've been trying to figure out where the actual cause of the mismatch is and I just get more and more confused. The following is my contents of the the 4 views that I've understood are of importance:

v$nls_parameters nls_database_parameters nls_instance_parameters nls_session_parameters

select * from v$nls_parameters NLS_LANGUAGE SWEDISH NLS_TERRITORY SWEDEN NLS_CHARACTERSET WE8ISO8859P1 select * from nls_database_parameters NLS_LANGUAGE AMERICAN NLS_TERRITORY AMERICA NLS_CHARACTERSET WE8ISO8859P1 NLS_NCHAR_CHARACTERSET WE8ISO8859P1 select * from nls_instance_parameters NLS_LANGUAGE AMERICAN NLS_TERRITORY AMERICA select * from nls_session_parameters NLS_LANGUAGE SWEDISH NLS_TERRITORY SWEDEN in registry HKEY_LOCAL_MACHINE\Software\oracle\home0\nls_lang = "SWEDISH_SWEDEN.WE8ISO8859P1" Things don't look totally consistent with the American and Swedish mix of things.

I'm not entierly sure what the point of these 4 different views plus the registry entry is. Isn't it enough with one view telling me the charset for the database, why 4 views giving me different information..?

If someone knows, what would a consistent non character mismatch setting look like (if I want to use the swedish character set). And how do I make those changes.

--
Posted via dBforums
http://dbforums.com