On Sat, Nov 28, 2015 at 12:45:24AM -0600, Felipe Gasper wrote:
> Not quite … that’s dealing with just customizing the messages, not
> specifying what the encoding of those messages is to be.
Right it is missing the "charset" attribute that needs to be folded
into the "Content-Type: text/plain; ..." header of "part 1" of the
multipart/report.
As that is not a header of the outer message, there does not seem
to be an obvious way to map the documented syntax to a specification
of the charset to be used in a MIME part.
The simplest solution might be to restrict non-ASCII content in
the message template to be UTF-8 (no charsets other than "us-ascii"
and "utf-8"). Then just scan for non-ascii characters, and if any
are found, declare the content to be utf-8.
--
Viktor.
