Document.getInputEncoding()

[Dick Deneer]

I am using Xerces 2.7.1 and parsing a File to a DomTree using Dom Level 3 APi LSParser 
: document = parser.parseURI(fileName); 
Then I do document.getInputEncoding(). What I expect is to get the actual encoding of the File used by the parser. But I always get "UTF-8", also when in fact the file encoding is ISO-8859-1.
Is this normal behaviour or should it return the actual File encoding?

Dick Deneer