Re: [Mutt] #2682: mutt should let me specify the Sender: header
#2682: mutt should let me specify the Sender: header
Old description:
> {{{
> mutt should let me compose messages and specify a Sender: field. Instead,
> mutt silently drops the Sender: field that I add in.
>
> For instance, RFC2822 gives us the sample where John's secretary Michael
> sends a message to Mary:
>
> ---
> ----
> From: John Doe <jdoe@xxxxxxxxxxxxxxx>
> Sender: Michael Jones <mjones@xxxxxxxxxxxxxxx>
> To: Mary Smith <mary@xxxxxxxxxxx>
> Subject: Saying Hello
> Date: Fri, 21 Nov 1997 09:55:06 -0600
> Message-ID: <1234@xxxxxxxxxxxxxxxxxxxxx>
>
> This is a message just to say hello.
> ----
>
> If I do try this in mutt, it lets me change from From: line to John, but
> drops my Sender: line. Thus, it looks to the recipient like the message
> actually came from John, with no clear indication that it really came
> from Michael.
>
> It appears the problem is that mutt knows about the Sender: header
> sufficiently well to treat it specially by incorporating it into the
> struct envelope, so that it is not treated as a user header, but does not
> know enough to write the header out when it writes out headers.
>
> Some might argue that it's not the job of the the MUA to specify the
> Sender: header and prefer the MTA do so. I disagree, because many MTAs
> don't, and I wish to customize my Sender: headers with explanatory text
> (e.g. a name) that would not be inserted by an MTA.
>
> >How-To-Repeat:
> In mutt, turn on header editing, compose a message, add a Sender: header,
> notice it disappears when you return to the composing menu, or if you go
> back to editing the message before ending it. Send the message. Observe
> the recipient does not have a Sender: header.
> >Fix:
> Attached is a patch to mutt_write_rfc822_header() (against 1.5.12) to
> write the Sender: header right after it writes the From: header.
>
> I don't think this needs to be conditionalized on !privacy, but I am not
> certain.
> }}}
New description:
mutt should let me compose messages and specify a Sender: field. Instead,
mutt silently drops the Sender: field that I add in.
For instance, RFC2822 gives us the sample where John's secretary Michael
sends a message to Mary:
{{{
From: John Doe <jdoe@xxxxxxxxxxxxxxx>
Sender: Michael Jones <mjones@xxxxxxxxxxxxxxx>
To: Mary Smith <mary@xxxxxxxxxxx>
Subject: Saying Hello
Date: Fri, 21 Nov 1997 09:55:06 -0600
Message-ID: <1234@xxxxxxxxxxxxxxxxxxxxx>
This is a message just to say hello.
}}}
If I do try this in mutt, it lets me change from From: line to John, but
drops my Sender: line. Thus, it looks to the recipient like the message
actually came from John, with no clear indication that it really came from
Michael.
It appears the problem is that mutt knows about the Sender: header
sufficiently well to treat it specially by incorporating it into the
struct envelope, so that it is not treated as a user header, but does not
know enough to write the header out when it writes out headers.
Some might argue that it's not the job of the the MUA to specify the
Sender: header and prefer the MTA do so. I disagree, because many MTAs
don't, and I wish to customize my Sender: headers with explanatory text
(e.g. a name) that would not be inserted by an MTA.
>How-To-Repeat:
In mutt, turn on header editing, compose a message, add a Sender: header,
notice it disappears when you return to the composing menu, or if you go
back to editing the message before ending it. Send the message. Observe
the recipient does not have a Sender: header.
>Fix:
Attached is a patch to mutt_write_rfc822_header() (against 1.5.12) to
write the Sender: header right after it writes the From: header.
I don't think this needs to be conditionalized on !privacy, but I am not
certain.
--
Ticket URL: <http://dev.mutt.org/trac/ticket/2682#comment:1>