Parsing fails to use BOM markers
When a HTML file is saved in a format other than unicode it is usual for a multi byte BOM marker to be added to the beginning of the file. For UTF-8 this is 3 bytes of EF BB BF heading the file.
Unfortunately XMLC doesn’t pre-check the file and its processed as Unicode
resulting in  prefixing the document output along with
other chars > 128 being incorrectly prefixed. This is an issue when you have
multiple files saved in Unicode or UTF format and during a build it is not
possible to check each file and apply an encoding option before its parsed.
It's also applicable when wanting to use XMLC in a dynamic environment with
deferred parsing when you don’t know what changes will be made.
As an example, create a text file in Notepad. Enter a small amount of HTML and include a pound sign by using the key sequence HoldAlt-Num0-Num1-Num6- Num3 to generate the character 0xA3. Save the file with SaveAs and change the encoding type to UTF-8 before selecting ok. As an example of the output, run the 'dynamic' example using the created file as the input.