Re: [nmh-workers] Can't forward MIME-encoded message

[DuaneTime]

Apparently, it depends on whether you define "MIME Message" as "Conforms to RFC 
2045 (Or whatever...  op. cit. ref. "Steeenking Badges"), or "Contains Readable 
Mime Data"...  I guess if I were *generating* a "Mime Message", I would use the 
former definition, whereas if I were trying to *detect* a "Mime Message", I 
would use the latter...

I'm not at all sure how I'd define "Bug", in this situation...  However, since 
(apparently?) the presence of *either* the "Mime-Version" *or* the 
"Content-Transfer-Encoding" header will Trigger "mhbuild", shouldn't a 
"Content-type" header with "Boundary" flag (or maybe, just a "Content-type" of 
"mixed, or "alternative" trigger it likewise?  I mean, that header pretty well 
guarantees that the message is MIME-formatted, even if it doesn't exactly 
conform to the RFC...

Sent with ProtonMail Secure Email.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday, May 9, 2019 10:42 AM, Ken Hornstein <address@hidden> wrote:

> > The only mandatory header for "forwarding" (yeah, right) a
> > pre-existing message is "Content-type", since it has the (pre-generated)
> > "Boundary" flag... So I just edit everything else out.
>
> This is the problem, here, because that is NOT true.
>
> A MIME message needs a MIME-Version header, full stop, no exceptions;
> this is unambiguous (see RFC 2045, §4). This is what we trigger on to
> make mhbuild do the right thing. If we don't see a MIME-Version header,
> we go on the assumption that the existing draft is NOT a MIME message,
> and we MIME-ify the resulting draft, which I suspect ends up generating
> a new Content-Type header for your message which does not contain a
> boundary marker (because the message isn't a MIME message according to
> the rules).
>
> So, I have to say that I don't see this as a nmh bug ... if you want
> to feed a MIME message into mhbuild you should include a MIME-Version
> header; it's not meaningless at all. If you are trying to send out a
> message that does not contain a MIME-Version header, then that is not
> valid according to the standards.
>
> --Ken