harold_77 test case

Hello!

It's
<?xml version='1.0'?>
<document xmlns:xi="http://www.w3.org/2001/XInclude">
   <xi:include parse="text" href="text/UnicodeBigUnmarked.xml"/>
</document>

where text/UnicodeBigUnmarked.xml document is probably USC2 encoded. The 
description says: "Can autodetect UCS2 big endian files with a without a 
byte order mark when parse="text"".

The question is how the encoding is supposed to be detected? I presume 
the resource mime type should be XMLish, but on .NET when loading 
text/UnicodeBigUnmarked.xml file from the file system I get 
"application/octet-stream" so the test fails.
Am I right this test presumes XMLish media type so the encoding is 
detected as per XML rules?

-- 
Oleg Tkachenko
http://blog.tkachenko.com
Multiconn Technologies, Israel