[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mutt-users
Subject:    Re: How to make mutt execute commands from the command line?
From:       Nicolas Rachinsky <mutt-users-1 () ml ! turing-complete ! org>
Date:       2006-11-09 14:37:22
Message-ID: 20061109143722.GA40691 () mid ! pc5 ! i ! 0x5 ! de
[Download RAW message or body]

* "M. Fioretti" <mfioretti@mclink.it> [2006-11-09 06:23 +0100]:
> >> 1) Mutt beeps when starting and remains open after copying the messages.
> >>    This is a sure sign that the '<quit><enter>' part is wrong, but I
> >
> >Are you sure <copy-message> does not ask for confirmation for
> >appending to or creating the target folder?
> 
> yes. What I get is a mutt index of the original mailbox in which all
> the messages are already tagged, and if I hit q mutt closes without
> questions and I do find the new Maildir mailbox ready.

Where does it end if you leave out the '<quit><enter>' part? Do you
unset confirmcreate/confirmappend?

> >>    as it has copied everything to the new maildir... (remember this
> >>    should eventually run unattended inside a bash loop)
> >
> >I'm not that sure, that running mutt unattended, is the best idea.
> 
> why, exactly? Consider that I would do this only once offline, to
> change format and name of many _archived_ mailboxes in a way that will
> be surely usable by mutt when, in the future, they are opened. Is mutt
> known for losing data when used in this way? Or maybe the doubts come
> from something else?

IIRC mutt does always clear the terminal and similar things. So
there is output in all cases, making it difficult to detect errors,
the error messages may hide in a bunch of control codes for the
terminal.  And the exit code isn't usable to detect (all) errors,
too.

Nicolas

-- 
http://www.rachinsky.de/nicolas
[prev in list] [next in list] [prev in thread] [next in thread]