I am trying to import supplier records into my IFS8 system using xml files generated by an external system. The xml files are supposed to be using utf-8 encoding. However, because some of the records are generated from Korean and Ukrainian sources, not all the characters are utf-8 despite the xml referencing <?xml version="1.0" encoding="UTF-8"?>. As a result, when the connector reader gets the xml file, it reports:

com.sun.org.apache.xerces.internal.impl.io.MalformedByteSequenceException: Invalid byte 1 of 1-byte UTF-8 sequence.

And the file is not processed any further.

Sometimes the control characters will clobber the xml tags, which nullifies the entire file structure.

Aside from forcing the sender to reformat the data, is there anything I can do in IFS to compensate for the bad characters when the connector reader tries to pick it up?

Here is an example of one of the bad xml files opened in notepad++.