Ensure that you have fonts that can represent all characters in ISO 8859-15 and Windows cp1252.
C1 code points in ISO 8859-15
Test passes if
A. The symbols describe the lines to their right.
B. The document encoding indicated by the browser chrome (if available), says ISO 8859-15.
Assertion: When an ISO 8859-15 encoded document contains code points in its C1 control range that correspond to graphic characters in the Windows 1252 encoding, these are not displayed as Windows 1252 characters, and the user agent continues to otherwise handle this document as ISO 8859-15.
Encoding information is sent in the HTTP header; the success of this test rests on the assumption that the HTTP header applies the expected character encoding.
The left side of the top line contains the following sequence of code points from the C1 range: 0x80 0x9A 0x9E 0x8E. These are not graphic characters in 8859-15, but in Windows cp1252 they represent the sequence of characters shown in the graphic below.
The characters to the right serve to distinguish ISO 8859-15 from Windows cp 1252, ISO 8859-1 and Windows cp1256 (Arabic).
If the XML fails to load due to a syntax error, that is also evidence that the assertion held true.