Comment 3 for bug 191199

To handle the accented characters, would it make sense to set the encoding of
the xml output to latin1? Something like this:
  <?xml version="1.0" encoding="iso8859-1" standalone="no"?>

Alternatively, the data could be reencoded to UTF-8 (which is the default
encoding for XML files).

The first option is probably the easiest one, and is probably correct if
bugzilla is assuming things work in latin1.

The encoding could easily be part of the template, so that bugzillas in other
countries which have bugs using a different encoding could be exported correctly.

I think a compromise like this would be necessary, given that the encoding of
the text returned by the bugzilla user isn't known (or isn't recorded).