I was hoping that someone might be able to shed some light on a debate
I am having here with my colleagues. In short we have a stored
procedure (Oracle 10G R2) that is taking in an XML document (UTF-8
encoding) and is shredding it and then dumping the data into relational
tables (nvarchar2 attributes to support the unicode data). That is the
readers digest version of the chain of events. The shredding process
within the procedure can only accomodate clob and not nclob and it is
here where the problem lies. I think that if data comes in via the xml
doc and it has actualy UNICODE code points that once the clob recieves
it some funky conversion will take place and the code points will be
converted to something other than what they were intended to be. If
anyone can shed any light on this I would be greatful.
Regards, TFD.
Received on Fri Aug 04 2006 - 10:42:03 CDT
