[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Nmh-workers] Conflict between "mime" command and attach
From: |
Jon Steinhart |
Subject: |
Re: [Nmh-workers] Conflict between "mime" command and attach |
Date: |
Wed, 11 Dec 2013 07:19:36 -0800 |
Ralph Corderoy writes:
> Hi Ken,
>
> > The reason this is cropping up now is that we want to get to the point
> > where a MIME headers are always generated (I assume this is
> > non-controversial).
>
> Looks down. Mumbles "No". Scuffs foot.
No, I don't always want MIME headers. I'm old-fashioned and don't need
a kilobyte where 10 bytes will suffice. I like nmh because text-plain is
good enough for most of what I do.
> > Ok, Ralph did say that he wanted to look at the headers
> > post-MIMEification to adjust them
>
> No, sorry, when I said "edit" I was referring to a whatnow-entry to put
> me back in vi so I can read-only peruse the outcome of "mime". My
> intent is always to have "mime" do the work; if something's not right I
> go back to pre-"mime" and fix it because mhbuild could always do want I
> want AIUI.
Do the big MUAs let you do this? Can you look at post-MIME stuff in
thunderbird? Or do you just attach things and be happy with the results>
> There seem to be two issues. `#' is a poor magic character if mhbuild
> is to run all the time, clashing too much with unindented cpp(1) and
> sh(1) input. A simple way to attach files shouldn't stop mhbuild
> directives being used. (Personally, I think having `cd', `pwd', etc.,
> at the whatnow-prompt for "attach foo" is wrong; MH's raison d'etre was
> to not be its own little shell and whatnow shouldn't grow to be one.)
I agree (as the culprit) that having cd, pwd, etc. at whatnow is wrong.
But only because whatnow is wrong. IMHO one should finish composing a
draft and then use a send command or other command like tools like attach.
To some degree this is how it gets used in emacs I believe.
> How about if `#' was configurable and could be multiple characters? And
> that could further be overridden on a per-message basis by an
> nmh-header? The new default could be /^mhb\>/; something that's
> unlikely to be accidental. Any mhbs that don't parse stop whatnow's
> "send" working.
>
> mhb image/png \
> /home/foobar/junk/picture.png
> mhb forw +inbox 42 43 99
I can't support making a cryptic interface more cryptic.
> A new mhbuild directive that guesses the MIME type could provide a
> simple attach mechanism. whatnow's "attach" could append these instead
> of adding its header. Or still add its header and they're treated as if
> they were additional directives at the end of the body in the order
> they're encountered.
This is why I suggested having some optional arguments on the attach header
and attach command. And why I suggested a -check option to verify that
everything is OK before send. Nobody complains about whom -check!
Jon
Re: [Nmh-workers] Conflict between "mime" command and attach, David Levine, 2013/12/07
Re: [Nmh-workers] Conflict between "mime" command and attach, Ken Hornstein, 2013/12/12
Re: [Nmh-workers] Conflict between "mime" command and attach, Lyndon Nerenberg, 2013/12/12
Re: [Nmh-workers] Conflict between "mime" command and attach, Ken Hornstein, 2013/12/12
Re: [Nmh-workers] Conflict between "mime" command and attach, Jon Steinhart, 2013/12/12
Re: [Nmh-workers] Conflict between "mime" command and attach, Ralph Corderoy, 2013/12/15
Re: [Nmh-workers] Conflict between "mime" command and attach, Paul Fox, 2013/12/15