#204 Wrong escaping of 3 byte Unicode in US-ASCII output

dom4j_1.6.x
open
nobody
None
1
2014-01-27
2014-01-27
Denis
No

XML with US-ASCII encoding:
<?xml version="1.0" encoding="US-ASCII"?>
<test>&#120139;</test>

outputs as:
<?xml version="1.0" encoding="US-ASCII"?>
<test>&#55349;&#56651;</test>

i.e. lower and upper surrogates are interpreted as separate characters and escaped.
is this expected behavior? it does not look correct.
DOM4j version used is 1.6.1

code is:

byte[] data = "<?xml version=\"1.0\" encoding=\"US-ASCII\"?><test>&#120139;</test>".getBytes();
SAXReader reader = new SAXReader();
ByteArrayInputStream in = new ByteArrayInputStream(data);
Document doc = reader.read(in);
System.out.println(doc.asXML());

Discussion