In the output metadata contained in a xena file there is an element "opendocument:opendocument". This element seems to be metadata related to the normalised version of the input file (i.e. the open version). However under this there is an attribute of "opendocument:type" that is set to the Type of the input file rather than the type of the normalised file. This attribute name is ambiguous and confusing. For example normalising a doc file will result in a xena file containing a base64 encoded odt file. However the opendocument:type attribute will be "Microsoft Word" in this case which is the type of the input file rather than the normalised file.
If changes are mad update publised documentation for the wrapper.