1. 6

I’ve used mutt for a while, thinking of having a play with this. Anyone used it before?

  1.  

  2. 5

    For what it’s worth, NeoMutt is what you get when you install the mutt package on recent versions of Debian:

     $ mutt -v | head -n2
     NeoMutt 20170113 (1.7.2)
     Copyright (C) 1996-2016 Michael R. Elkins and others.
    

    Regardless of that, I have normally used all of the mutt-related documentation around the web with no obvious differences. I think the Sidebar is the most notable patch applied by NeoMutt, and it’s worth it (although it might be merged into a recent stable version of Mutt as well).

    1. 4

      No, but I’d like to have a working CLI mail client, even just to check and mark as read.

      I use multiple accounts (university and Gmail), how do neomutt works with this kind of workload?

      Do I have to setup a MTA?

      Since most email are (sadly) html, how could I read these?

      1. 2

        I normally use Mutt with multiple accounts. There are several ways you can set that up, and a lot of them are decently documented. I use something very similar to this.

        It’s not necessary to run your own MTA; I use Mutt’s built-in SMTP support with multiple GMail accounts with no issues.

        Most of the people and services I communicate with send a plaintext version of the email alongside the HTML version, so I have a lot less HTML trouble than expected. However, for HTML-only email, the best option I could find is to save the HTML file to the disk and use a web browser to read it. It should be fairly easy to create a macro for that in Mutt.

        1. 4

          I have a .mailcap file setup with the below. It does a pretty good job letting me view what I need from HTML emails. text/html; lynx -dump %s; copiousoutput; nametemplate=%s.html

          1. 1

            Thank you for your the inputs, I’ll check out (neo) mutt as soon as possible.

        2. 2

          I’ve converted my mail flow to use mblaze. Along with fsf and a few helper scripts, it’s perfect for my use. Used in conjunction with offlineimap + msmtp, and $EDITOR.

          Never looking back.

          1. 1

            What is fsf in this context? For that matter, what is mblaze?

            1. 2

              I think mblaze is this and I think fsf would be some sort of fuzzyfinder?

              1. 2

                Maybe fzf? I could see that being a nice workflow.

                1. 2

                  Correct. I just hooked it up to my script and forgot about it. It does nicely with the –preview command. There’s probably more I could do with that, but for now it solves pretty much all of my mail consumption related issues

                  1. 1

                    I’d be interested in a more in-depth explanation of your setup, if you’d like to share it…

                    1. 2

                      Aside from a regular offlineimap setup and an also-regular msmtp setup, I have two scripts to help me with mblaze: one called “mymail” and one called “mshowwithawk” (which is a mouthful, but I never invoke it by hand so I don’t care.

                      mshowwithawk is:

                      #!/bin/bash
                      
                      mshow $(echo $1 | awk '{print $2}')
                      

                      and mymail is:

                      #!/bin/bash
                      mlist -s ~/Mail/$1/$2 | msort -d | mseq -S | mscan 2> /dev/null | fzf --preview="mshowwithawk {}"
                      

                      Usage is: mymail <any ~/Mail/ subdir that contains maildirs> . Reason for this script is I have two accounts, and I often switch between my work and personal email, so I often call it like “mymail otremblay INBOX” or “mymail work INBOX”. Next improvements are gonna be defaulting to INBOX and allowing for the -s flag to be passed or not from the mymail script (because sometimes, I need to see old mail too.)

                      The output is a list of selectable items, with a preview of the currently selected item on the right. Yes, right in the terminal. The list is populated by mail prefixed with an ID I can then use with mshow if I need better output (say, in case the email provided a worthless (but still present) text/plain thing. I use elinks to dump out text/html messages (configurable in mblaze).

                      I use mblaze’s “minc” to pass messages from the maildir’s new to cur, and mflag -S to flag everything as read once I’m done.

                      I like the workflow because it is just a construction of a collection of small specialized programs working together. I mean, if needed, I can still just invoke mlist by itself and grep through email headers, if I so desire. Or pump the whole output elsewhere to any other unix-standard utility if I want to. Heck, it would be trivial to include spamassassin header parsing, or any other kind of header parsing. I’m also a sucker for CLI interfaces, mostly on account of it being the easiest way I know to compose software with one another out of small blocks. I feel like I should probably start a blog about my crap, but I’m afraid that said crap would be too trivial for people to enjoy.

            2. 1

              mblaze is indeed pretty nice (similar to mu. I use it to automate some tasks with n my email workflow (archiving, marking as done, digging up the full thread when a mail arrives, …) it helps me a ton. But when it comes to actually read and reply to mails, it doesn’t cut it, so I use mutt for that.