I am on Oracle 11.2 and Red Hat Linux
I have a csv file which contains non ASCII character e.g. some Swedish characters.
I am loading in database using sql loader.
My NLS_LANG=AMERICAN_AMERICA.AL32UTF8
I have also tried with NLS_LANG=AMERICAN_AMERICA.UTF8
My LANG=en_US.UTF-8
I have tried to unset LANG as well.
I sqlldr control file is
load data append into table material_master
fields terminated by "," optionally enclosed by '"' TRAILING NULLCOLS
(
PLANT,
MATERIAL_NUMBER,
MATERIAL_DESC,
PROFIT_CENTER_NAME
)
When I look in sql developer what was loaded in this table, non ASCII characters show as square.
My database was created with character set of AL32UTF8.
I suspect something has happened in Oracle 11.2 with Linux as operating system.
On linux box, if I cat the csv file being loaded, it does show the Swedish characters correctly, it is jut when information gets loaded in database, it is creating problems.
Outside of sqlldr, I have loaded the file using a Java program, still same problem.
In the past when we were on Oracle 10.2.0.4 on HP UNIX, we never encountered such issue either with sqlldr or java. We were always able to load non ASCII Characters without any problem.