For some reason the regular 'cat' completion does not work, perhaps
because it is regularly aliased to 'bat' on my systems. This manually
fixes it with a spec alias for carapace
(https://carapace-sh.github.io/carapace-bin/spec/run.html#alias)
Should find a better implementation over time but for now this is a
simple workaround. I do _not_ know how it interacts with systems when no
'bat' command is found.
Rename tabs with `<leader><S-R>`. This will provide a (full-screen)
prompt in which you can exit with esc (not changing anything), provide a
static name for the tab, or hit enter with nothing on the prompt to
return to the default naming behavior.
Suggestions are given at the top of the completion menu. Additionally,
since that can get really annoying really quick there have been some
changes specifically for copilot.lua:
We start with it disabled. It can be enabled with `<leader>ap` and
disabled again with `<leader>aP` for Ai>coPilot. If the plugin at some
point exposes functionality to toggle itself on/off completely I would
gladly switch to that. The current toggle command only attaches/detaches
the buffer but still runs the client in the background. This is less
desirable for me, plus it means the completions are still done
automatically in blink.
Since I have two configurations for my dual-screen desktop setup, with
one having the two screens side-by-side and the other having the left
screen vertically rotated 90 degrees, I need two different kanshi
setups.
The vertical setup is called 'dockedvert' and is the default when the
two screens are detected. The horizontal setup is only called 'docked'
and can be switched to manually. Before, it would not correctly revert
the left screen from its vertical 90 degree rotation when going from the
vertical setup to the horizontal one.
So I applied the `transform` output directive but kanshi refused to
start. That is because to revert the transform into the default
horizontal position you do not use `0` but `normal` as the value.
With this fix everything works as expected.
Rename mail utility programs (yet again), to be prefixed with `mail-`
instead of `neomutt-`. Also did not substitute with other MUA like
`aerc-` since these utilities are mostly independent of specific
implementations.
The very specific `neomutt-filer` implementation script I completely
removed since for aerc the same can be achieved with a simple
configuration option.
Since we do not use neomutt anymore. Additionally, perhaps a 'mail-'
prefixed system is better for scripts etc which run independent of a
specific MUA.
Explanation here:
https://man.sr.ht/~rjarry/aerc/configurations/mailto.md
I built a custom script 'aerc-in-terminal' since I want to expand the
$TERMINAL env var and use that instead. `.desktop` files do not allow
expanding vars (since they don't run in a user shell), so this is a
compromise.
Mails get symbols in the flag column if they are sent by me, CC or BCC
me, or they are tagged as a 'list' (from a mailing list).
Added some additional info to status bar (current folder, recent and
unread counts), and extended the flag column slightly to accomodate more
flags.
Starting to replace neomutt in my workflow with aerc:
it is a little more lightweight, directly supports notmuch directories
and has a relatively sane configuration style (mostly just ini-like)
with 'go-templating' baked in.
In general, the configuration just feels less 'cobbled-together' than
before. I can make changes without worrying what other things are going
to break by doing so. I understand the complete configuration and
styling, instead of mostly relying on other people's formatting lines
for the styles.
I am still learning some of the configuration possibilities but it
already functions as a neomutt replacement.
This helps to show what is 'hidden' in the public repository and also
shows me at a glance all the dotter-supplied variables to keep an
overview of where we inject stuff.
Include the color config for mako from ~/.local/state/mako/style instead
of hardcoding it or templating it into the config file.
This means we have to ensure the directory and file exist before running
mako which we do with an `env.d` startup script.
Hitting `,m` in the compose menu creates a dual txt/html email group
when the highlighted file is written in (djot) markdown.
Delete the original message afterwards to create a fully functional html
email with plaintext fallback.
Completely taken from: https://tom.wemyss.net/posts/neomutt-markdown-email/
with many thanks. What an amazing use of pipelines in neomutt!
First we unbind all existing key mappings and then we slowly build up
our own again. May be somewhat unnecessary but I have been getting
_very_ confused with neomutt mappings just popping up all over the place
and never quite sure if they were from me or default settings or where
exactly they were configured.
This provides a clean slate to build upon.
Unmap the custom set Debug overlay mapping since I haven't had to use it
in absolute ages. Additionally we also unmap the default mapping for the
debug overlay - which frees up the key combination to be passed through
to TUIs.
As mentioned in
https://github.com/carapace-sh/carapace-bin/issues/2819#issuecomment-3092307945,
fzf-tab directory completes are empty if you try to complete on a
directory without any prior input.
I.e.:
`ls -hal ~/.config/a<tab>` shows output (dirs starting with a).
`ls -hal ~/.config/<tab>` is empty.
This fixes the issue by making the query string use the full inserted
value.
In its default pre- and post-hooks, `neomutt-syncmail` now checks if
`imapfilter` and `notmuch` exist before invoking them.
Similarly, it defaults to `pass open` opening a pass coffin if
`pass-coffin` exists.
Renamed `mutt2task` to `neomutt-2task`, and renamed `sync-mail` to
`neomutt-syncmail`.
Changed key maps accordingly. Also systemd service even though I don't
think I'll use it again.
We have differentiated views of new (italic) and unread (underline, and
a bit of highlight on the author) messages.
Additionally the indicator highlighting works better with the rest of
the colors, and we have a better paperclip icon icon for attachments.
E-Mails have 2-staged flag options for levels of 'seen':
Unseen/Seen and Unread/Read.
A message is unseen if it has arrived since the last time we had mutt
open. We do not have to have had the actual message open for it to be
marked seen (or 'old'). In other words, whichever messages we have seen
once on the index alone are not 'new' to us anymore, they have been
seen.
On the other hand, a message is only read if it is actually opened. So
even if we have seen it on the index page, or moved it to a different
mailbox or anything else, it can still be unread if we did not open it.
Additionally, I have it set up now so that the message only gets set to
'read' by neomutt after it has been open for 1 second.
This gives a tiny grace period if we are scrubbing through mails or
accidentally opened one we did not want.
This used to be default (and according to the docs still is?) but I had
to set it manually to still show all models to select from. Perhaps
because we explicitly set 'show_defaults' we also have to be explicit
about this.
