Re: [Nmh-workers] utf-8 subject lines?

[Peter Maydell]

Paul Fox wrote:
>my, what a flurry of traffic.  just when i think i'm the only person
>left still using mh.  :-)
>
>sometimes i get mail with subject lines like this:
>    Subject: 
> =?utf-8?B?Q1JPTkQsIFNFTkRNQUlMOiBtYWtlIHRoZW0gd29yayB0b2dldGhlcg==?=
>
>of course, i can't read it.
>
>is this a config problem?  for me?  for the sender?

It's an RFC-2047-encoded header. nmh in CVS seems to cope with that
header fine.

>is it a bug?  in mh?  in the sender's mail client?

You could argue that the sender shouldn't be encoding subjects
(in this case "CROND, SENDMAIL: make them work together") that don't
actually have any non-ASCII in them, but really nmh has to cope.

>just for completeness, in the case of the message with the above
>subject, the content was also encoded:
>    Content-Type: text/plain; charset=utf-8
>    Content-Disposition: inline
>    Content-Transfer-Encoding: base64
>
>but mh handled that okay.  (but i sure hate it, since my reply
>scripts and forms _don't_ handle it, and i have to pull the
>output of "show" into my outgoing buffer, and edit it, in order
>to reply sensibly.  but that's a different problem.)

Yes. I think that better handling of MIME, especially in replies,
is really the most important issue at the moment.

>i seem to be running nmh-1.2.  if this is fixed in CVS, then i'm
>all for us putting out a 1.3.  i even volunteer to update the
>savannah frontpage again.  :-)

That would be nice :-)

-- PMM