Responding to a MIME attachment

Responding to a MIME attachment

[Hrvoje Niksic]

I don't know how to (easily) respond to a message that is included as
a message/rfc822 attachment.  It used to be possible to press C-d,
find the message and respond to it, but that no longer works.
Instead, I get "No articles in group nnml:private-11330".

Any comments?

Re: Responding to a MIME attachment

[Shenghuo ZHU]

>>>>> "Hrvoje" == Hrvoje Niksic <hniksic@iskon.hr> writes:

Hrvoje> I don't know how to (easily) respond to a message that is
Hrvoje> included as a message/rfc822 attachment.  It used to be
Hrvoje> possible to press C-d, find the message and respond to it, but
Hrvoje> that no longer works.  Instead, I get "No articles in group
Hrvoje> nnml:private-11330".

Hrvoje> Any comments?

I guess the message is not message/rfc822 attachment but old format
forward and the part is generated by mm-uu. To verify my conjecture,
type (C-u g).  However, digest does not support forward format.

Simply creating a forward type in digest would help, but not in all
situations. Some mail servers (MTA's) may encode a message with
base64, which causes the simple solution failed, since Gnus can not
find the boundary in an encoded message, I guess.

I think we need a reply function on a mm part.

-- 
Shenghuo

Re: Responding to a MIME attachment

[Hrvoje Niksic]

Shenghuo ZHU <zsh@cs.rochester.edu> writes:

> I guess the message is not message/rfc822 attachment but old format
> forward and the part is generated by mm-uu. To verify my conjecture,
> type (C-u g).

Yes, that's the case.  The problem with the mm-uu stuff is that it
makes things _look_ as if you know what's going on, except you don't,
and things fail mysteriously.

> I think we need a reply function on a mm part.

That might be a bit of a clutter, because you'd then also need a
wide-reply function, a send-forward function, an include-into-digest
function, etc.  What we need is a way to insert the part into a
summary, so you can do with it whatever you'd normally do with summary
buffers.

One more reason why C-d should use the mm stuff.  :-)

Re: Responding to a MIME attachment

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@iskon.hr> writes:

> What we need is a way to insert the part into a summary, so you can
> do with it whatever you'd normally do with summary buffers.

That's a good idea.  Sounds like work, though.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Responding to a MIME attachment

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Hrvoje Niksic <hniksic@iskon.hr> writes:
> 
> > What we need is a way to insert the part into a summary, so you can
> > do with it whatever you'd normally do with summary buffers.
> 
> That's a good idea.  Sounds like work, though.

Yes.  But it would be a major win for many things.  I like the
"virtual articles" created by `X u' and stuff.  Maybe we could reuse
that, user-interface-wise?

Re: Responding to a MIME attachment

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@iskon.hr> writes:

> > > What we need is a way to insert the part into a summary, so you can
> > > do with it whatever you'd normally do with summary buffers.
> > 
> > That's a good idea.  Sounds like work, though.
> 
> Yes.  But it would be a major win for many things.  I like the
> "virtual articles" created by `X u' and stuff.  Maybe we could reuse
> that, user-interface-wise?

I've been thinking over this a bit, and I feel that the pseudo-article
thing isn't all that useful.  You have these articles that respond to
some commands, but not to others, and they are kinda awkward.  I'm
more in favor of the nndoc approach -- split the thing into parts,
fudge up sensible headers for stuff that doesn't have it, and present
the user with a new summary buffer...

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen