After some additional debugging, I've found that the encoding problems occur because the CXF BC strips off the SOAP envelope (which specifies the encoding in the XML header) and converts the SOAP message body to a UTF-8 encoded XML message complete with an XML header that indicates UTF-8. Later inside the Camel service unit a call to convertBodyTo(String.class) converts the body to the default charset which on the Windows box that I'm using is Windows-1252. However, the UTF-8 header from the CXF BC conversion process is still intact. Thus, when the message arrives at a place where XML parsing is performed (such as the JMS:Consumer that pulls the message off the second queue), the parser fails because of the mismatch between the encoding header and the actual encoding of the XML.
So now my problem is to determine if it is possible to configure the CXF BC to respect the encoding of the message that it's consuming instead of converting it to UTF-8. -- View this message in context: http://servicemix.396122.n5.nabble.com/Character-Encoding-Problems-Using-CXF-Consumer-with-JMS-tp461424p510174.html Sent from the ServiceMix - User mailing list archive at Nabble.com.
