emacs/init.org
2023-06-13 03:47:41 +02:00

1848 lines
55 KiB
Org Mode

#+TITLE: Emacs configuration file
#+AUTHOR: Lars Tveito
#+PROPERTY: header-args :tangle yes
#+STARTUP: overview
* About
This is an Emacs configuration file written in [[http://orgmode.org][Org mode]]. It is an attempt to
keep my =~/.emacs.d= tidy, but still be able to keep it all in one file. I
aim to briefly explain all my configurations as I go along!
I would not recommend using this configuration /as-is/, because it probably
contains a lot you don't really need. I do, however, hope people find some
golden nuggets that they can smuggle into their own configs.
If you really do want to try this config out, this is how I'd go about it:
Clone the repo.
#+begin_src sh :tangle no
git clone https://github.com/larstvei/dot-emacs
#+end_src
Backup your old =~/.emacs.d= (if necessary).
#+begin_src sh :tangle no
mv ~/.emacs.d ~/.emacs.d-bak
#+end_src
Backup your old =~/.emacs=-file (if necessary).
#+begin_src sh :tangle no
mv ~/.emacs ~/.emacs-bak
#+end_src
And finally
#+begin_src sh :tangle no
mv dot-emacs ~/.emacs.d
#+end_src
On first run it should install a bunch of packages (this might take a while),
and you might have to restart your Emacs the first time. If you experience
bugs, please let me know!
* Meta
All changes to the configuration should be done in =init.org=, *not* in =init.el=. Any changes in the =init.el= will be overwritten by saving =init.org=. The =init.el= in this repo should not be tracked by git, and is
replaced the first time Emacs is started (assuming it has been renamed to =~/.emacs.d=).
Emacs can't load =.org=-files directly, but =org-mode= provides functions to
extract the code blocks and write them to a file. There are multiple ways of
handling this; like suggested by [[http://emacs.stackexchange.com/questions/3143/can-i-use-org-mode-to-structure-my-emacs-or-other-el-configuration-file][this StackOverflow post]], one could just use =org-babel-load-file=, but I had problems with byte-compilation. Previously I
tracked both the =org.=- and =el.=-files, but the git commits got a little
messy. So here is a new approach.
When this configuration is loaded for the first time, the ~init.el~ is the
file that is loaded. It looks like this:
#+begin_src emacs-lisp :tangle no
;; This file replaces itself with the actual configuration at first run.
;; We can't tangle without org!
(require 'org)
;; Open the configuration
(find-file (concat user-emacs-directory "init.org"))
;; tangle it
(org-babel-tangle)
;; load it
(load-file (concat user-emacs-directory "init.el"))
;; finally byte-compile it
(byte-compile-file (concat user-emacs-directory "init.el"))
#+end_src
It tangles the org-file, so that this file is overwritten with the actual
configuration.
There is no reason to track the =init.el= that is generated; by running the
following command =git= will not bother tracking it:
#+begin_src sh :tangle no
git update-index --assume-unchanged init.el
#+end_src
If one wishes to make changes to the repo-version of =init.el= start tracking
again with:
#+begin_src sh :tangle no
git update-index --no-assume-unchanged init.el
#+end_src
I want lexical scoping for the init-file, which can be specified in the
header. The first line of the configuration is as follows:
#+begin_src emacs-lisp
;;; -*- lexical-binding: t -*-
#+end_src
The =init.el= should (after the first run) mirror the source blocks in the =init.org=. We can use =C-c C-v t= to run =org-babel-tangle=, which extracts
the code blocks from the current file into a source-specific file (in this
case a =.el=-file).
To avoid doing this each time a change is made we can add a function to the =after-save-hook= ensuring to always tangle and byte-compile the =org=-document after changes.
#+begin_src emacs-lisp
(defun tangle-init ()
"If the current buffer is init.org the code-blocks are
tangled, and the tangled file is compiled."
(when (equal (buffer-file-name)
(expand-file-name (concat user-emacs-directory "init.org")))
;; Avoid running hooks when tangling.
(let ((prog-mode-hook nil))
(org-babel-tangle)
(byte-compile-file (concat user-emacs-directory "init.el")))))
(add-hook 'after-save-hook 'tangle-init)
#+end_src
I'd like to keep a few settings private, so we load a =private.el= if it
exists after the init-file has loaded.
#+begin_src emacs-lisp
(add-hook
'after-init-hook
(lambda ()
(let ((private-file (concat user-emacs-directory "private.el")))
(when (file-exists-p private-file)
(load-file private-file))
(when custom-file
(load-file custom-file))
(server-start))))
#+end_src
A common optimization is to temporarily disable garbage collection during
initialization. Here, we set the ~gc-cons-threshold~ to a ridiculously large
number, and restore the default value after initialization.
#+begin_src emacs-lisp
(let ((old-gc-treshold gc-cons-threshold))
(setq gc-cons-threshold most-positive-fixnum)
(add-hook 'after-init-hook
(lambda () (setq gc-cons-threshold old-gc-treshold))))
#+end_src
* Packages
John Wiegley's extremely popular [[https://github.com/jwiegley/use-package][use-package]] was included in [[https://lists.gnu.org/archive/html/emacs-devel/2022-12/msg00261.html][Emacs 29]]. It
provides a powerful macro for isolating package configuration. After ignoring
this for a decade, I'll budge and give it a whirl.
#+begin_src emacs-lisp
(require 'use-package)
(setq use-package-always-ensure t)
#+end_src
Packages can be fetched from different mirrors, [[http://melpa.milkbox.net/#/][melpa]] is the largest archive
and is well maintained.
#+begin_src emacs-lisp
(setq package-archives
'(("GNU ELPA" . "https://elpa.gnu.org/packages/")
("MELPA Stable" . "https://stable.melpa.org/packages/")
("MELPA" . "https://melpa.org/packages/"))
package-archive-priorities
'(("GNU ELPA" . 10)
("MELPA" . 5)
("MELPA Stable" . 0)))
#+end_src
The configuration assumes that the packages listed below are installed. To
ensure we install missing packages if they are missing.
#+begin_src emacs-lisp
(let* ((package--builtins nil)
(packages
'(paredit ; minor mode for editing parentheses
pdf-tools ; Emacs support library for PDF files
proof-general ; A generic Emacs interface for proof assistants
racket-mode ; Major mode for Racket language
rustic ; Rust development environment
slime ; Superior Lisp Interaction Mode for Emacs
try ; Try out Emacs packages
vterm ; A terminal via libvterm
which-key ; Display available keybindings in popup
z3-mode))) ; z3/SMTLIBv2 interactive development
(when (memq window-system '(mac ns))
(push 'exec-path-from-shell packages)
(push 'reveal-in-osx-finder packages))
(let ((packages (seq-remove 'package-installed-p packages)))
(when packages
;; Install uninstalled packages
(package-refresh-contents)
(mapc 'package-install packages))))
#+end_src
* Mac OS X
I run this configuration mostly on Mac OS X, so we need a couple of settings
to make things work smoothly. In the package section =exec-path-from-shell=
is included (only if you're running OS X), this is to include
environment-variables from the shell. It makes using Emacs along with
external processes a lot simpler. I also prefer using the =Command=-key as
the =Meta=-key.
#+begin_src emacs-lisp
(when (memq window-system '(mac ns))
(setq ns-pop-up-frames nil
mac-option-modifier nil
mac-command-modifier 'meta
native-comp-async-report-warnings-errors nil)
(exec-path-from-shell-initialize)
(when (fboundp 'mac-auto-operator-composition-mode)
(mac-auto-operator-composition-mode 1))
(require 'ls-lisp)
(setq ls-lisp-use-insert-directory-program nil))
#+end_src
* Sane defaults
These are what /I/ consider to be saner defaults.
We can set variables to whatever value we'd like using =setq=.
#+begin_src emacs-lisp
(setq auto-revert-interval 1 ; Refresh buffers fast
default-input-method "TeX" ; Use TeX when toggling input method
echo-keystrokes 0.1 ; Show keystrokes asap
frame-inhibit-implied-resize 1 ; Don't resize frame implicitly
inhibit-startup-screen t ; No splash screen please
initial-scratch-message nil ; Clean scratch buffer
recentf-max-saved-items 10000 ; Show more recent files
ring-bell-function 'ignore ; Quiet
scroll-margin 1 ; Space between cursor and top/bottom
sentence-end-double-space nil ; No double space
custom-file ; Customizations in a separate file
(concat user-emacs-directory "custom.el"))
;; Some mac-bindings interfere with Emacs bindings.
(when (boundp 'mac-pass-command-to-system)
(setq mac-pass-command-to-system nil))
#+end_src
Some variables are buffer-local, so changing them using =setq= will only
change them in a single buffer. Using =setq-default= we change the
buffer-local variable's default value.
#+begin_src emacs-lisp
(setq-default tab-width 4 ; Smaller tabs
fill-column 79 ; Maximum line width
truncate-lines t ; Don't fold lines
indent-tabs-mode nil ; Use spaces instead of tabs
split-width-threshold 160 ; Split verticly by default
split-height-threshold nil ; Split verticly by default
frame-resize-pixelwise t ; Fine-grained frame resize
auto-fill-function 'do-auto-fill) ; Auto-fill-mode everywhere
#+end_src
The =load-path= specifies where Emacs should look for =.el=-files (or
Emacs lisp files). I have a directory called =site-lisp= where I keep all
extensions that have been installed manually (these are mostly my own
projects).
#+begin_src emacs-lisp
(let ((default-directory (concat user-emacs-directory "site-lisp/")))
(when (file-exists-p default-directory)
(setq load-path
(append
(let ((load-path (copy-sequence load-path)))
(normal-top-level-add-subdirs-to-load-path)) load-path))))
#+end_src
Answering /yes/ and /no/ to each question from Emacs can be tedious, a single /y/ or /n/ will suffice.
#+begin_src emacs-lisp
(fset 'yes-or-no-p 'y-or-n-p)
#+end_src
To avoid file system clutter we put all auto saved files in a single
directory.
#+begin_src emacs-lisp
(defvar emacs-autosave-directory
(concat user-emacs-directory "autosaves/")
"This variable dictates where to put auto saves. It is set to a
directory called autosaves located wherever your .emacs.d/ is
located.")
;; Sets all files to be backed up and auto saved in a single directory.
(setq backup-directory-alist
`((".*" . ,emacs-autosave-directory))
auto-save-file-name-transforms
`((".*" ,emacs-autosave-directory t)))
#+end_src
Set =utf-8= as preferred coding system.
#+begin_src emacs-lisp
(set-language-environment "UTF-8")
#+end_src
By default the =narrow-to-region= command is disabled and issues a
warning, because it might confuse new users. I find it useful sometimes,
and don't want to be warned.
#+begin_src emacs-lisp
(put 'narrow-to-region 'disabled nil)
#+end_src
Automaticly revert =doc-view=-buffers when the file changes on disk.
#+begin_src emacs-lisp
(add-hook 'doc-view-mode-hook 'auto-revert-mode)
#+end_src
* Modes
There are some modes that are enabled by default that I don't find
particularly useful. We create a list of these modes, and disable all of
these.
#+begin_src emacs-lisp
(dolist (mode
'(tool-bar-mode ; No toolbars, more room for text
scroll-bar-mode ; No scroll bars either
blink-cursor-mode)) ; The blinking cursor gets old
(funcall mode 0))
#+end_src
Let's apply the same technique for enabling modes that are disabled by
default.
#+begin_src emacs-lisp
(dolist (mode
'(abbrev-mode ; E.g. sopl -> System.out.println
column-number-mode ; Show column number in mode line
delete-selection-mode ; Replace selected text
dirtrack-mode ; directory tracking in *shell*
global-so-long-mode ; Mitigate performance for long lines
recentf-mode ; Recently opened files
show-paren-mode ; Highlight matching parentheses
which-key-mode)) ; Available keybindings in popup
(funcall mode 1))
#+end_src
* Visual
I am using a lot from [[https://github.com/rougier/nano-emacs][rougier's N Λ N O Emacs]], starting with the theme.
** Theme
For the light theme, I keep the light background toned down a touch.
#+begin_src emacs-lisp
;; N Λ N O theme
(use-package nano-theme
:init
(setq nano-light-background "#fafafa"
nano-light-highlight "#f5f7f8"))
#+end_src
The theme is set according to the system appearance (on macOS) if that is
available, defaulting to a light theme.
#+begin_src emacs-lisp
(defun load-nano-theme (variant)
(let ((theme (intern (concat "nano-" (symbol-name variant)))))
(load-theme theme t)))
(load-nano-theme (if (boundp 'ns-system-appearance) ns-system-appearance 'light))
#+end_src
Let's have Emacs change theme when the system appearance changes as well.
#+begin_src emacs-lisp
(when (boundp 'ns-system-appearance-change-functions)
(add-hook 'ns-system-appearance-change-functions 'load-nano-theme))
#+end_src
I want to be able to quickly switch between a light and a dark theme.
#+begin_src emacs-lisp
(defun cycle-themes ()
"Returns a function that lets you cycle your themes."
(let ((themes '(nano-light nano-dark)))
(lambda ()
(interactive)
;; Rotates the thme cycle and changes the current theme.
(let ((rotated (nconc (cdr themes) (list (car themes)))))
(load-theme (car (setq themes rotated)) t))
(message (concat "Switched to " (symbol-name (car themes)))))))
#+end_src
** Mode line
This is my setup for [[https://github.com/rougier/nano-modeline][N Λ N O Modeline]] after version 1.0.0:
#+begin_src emacs-lisp
;; N Λ N O modeline
(use-package nano-modeline
:hook ((prog-mode . nano-modeline-prog-mode)
(text-mode . nano-modeline-text-mode)
(org-mode . nano-modeline-org-mode)
(pdf-view-mode . nano-modeline-pdf-mode)
(mu4e-headers-mode . nano-modeline-mu4e-headers-mode)
(mu4e-view-mode . nano-modeline-mu4e-message-mode)
(elfeed-show-mode . nano-modeline-elfeed-entry-mode)
(elfeed-search-mode . nano-modeline-elfeed-search-mode)
(term-mode . nano-modeline-term-mode)
(xwidget-webkit-mode . nano-modeline-xwidget-mode)
(messages-buffer-mode . nano-modeline-message-mode)
(org-capture-mode . nano-modeline-org-capture-mode)
(org-agenda-mode . nano-modeline-org-agenda-mode))
;; Disable the default modeline and add a small margin around the frame
:init
(setq-default mode-line-format nil)
(add-to-list 'default-frame-alist '(internal-border-width . 24))
;; Set nano-modeline-text-mode as default
:config
(nano-modeline-text-mode 1))
#+end_src
** Font
Pick the first of the following fonts that is installed on the system.
#+begin_src emacs-lisp
(cond ((member "Source Code Pro" (font-family-list))
(set-face-attribute 'default nil :font "Source Code Pro-15"))
((member "Roboto Mono" (font-family-list))
(set-face-attribute 'default nil :font "Roboto Mono-14"))
((member "Fira Code" (font-family-list))
(set-face-attribute 'default nil :font "Fira Code-15"))
((member "Inconsolata" (font-family-list))
(set-face-attribute 'default nil :font "Inconsolata-14")))
#+end_src
New in Emacs 24.4 is the =prettify-symbols-mode=! It's neat.
#+begin_src emacs-lisp
(setq-default prettify-symbols-alist '(("lambda" . )
("delta" . )
("gamma" . )
("phi" . )
("psi" . )))
#+end_src
** Centering with Olivetti
[[https://github.com/rnkn/olivetti][Olivetti]] is a package that simply centers the text of a buffer. It is very
simple and beautiful. The default width is just a bit short.
#+begin_src emacs-lisp
;; Minor mode for a nice writing environment
(use-package olivetti
:defer t
:bind ("C-c o" . olivetti-mode)
:config
(setq-default olivetti-body-width (+ fill-column 3))
(remove-hook 'olivetti-mode-on-hook 'visual-line-mode))
#+end_src
** Focusing with focus
[[https://github.com/larstvei/Focus][Focus]] is my own package. It looks pretty nice, especially in combination
with Olivetti!
#+begin_src emacs-lisp
;; Dim color of text in surrounding sections
(use-package focus
:defer t
:bind ("C-c q" .
(lambda ()
(interactive)
(focus-mode 1)
(focus-read-only-mode 1))))
#+end_src
* Version control
Magit is the best.
#+begin_src emacs-lisp
;; A Git porcelain inside Emacs.
(use-package magit
:bind ("C-c m" . magit-status))
#+end_src
Have some visual indication where there are uncommitted changes.
#+begin_src emacs-lisp
;; Highlight uncommitted changes using VC
(use-package diff-hl
:config
(global-diff-hl-mode 1))
#+end_src
* Dashboard
#+begin_src emacs-lisp
;; A startup screen extracted from Spacemacs
(use-package dashboard
:config
(setq dashboard-banner-logo-title nil
dashboard-center-content t
dashboard-set-footer nil
dashboard-page-separator "\n\n\n"
dashboard-items '((projects . 15)
(recents . 15)
(bookmarks . 5)))
(dashboard-setup-startup-hook))
#+end_src
* EditorConfig
Using [[https://editorconfig.org/][EditorConfig]] is a must when collaborating with others. It is also a way
of having multiple tools that want to format your buffer to agree (e.g. both
the language's Emacs mode and some external formatter/prettifier).
#+begin_src emacs-lisp
;; EditorConfig Emacs Plugin
(use-package editorconfig
:config
(editorconfig-mode 1))
#+end_src
* Projectile
#+begin_src emacs-lisp
;; Manage and navigate projects in Emacs easily
(use-package projectile
:bind ("C-c p" . projectile-command-map))
#+end_src
* Ivy/Counsel
[[http://oremacs.com/swiper/][Ivy]] is a completion system, giving you completions and fuzzy search whenever
you interact with the minibuffer. I transitioned to Ivy from [[https://emacs-helm.github.io/helm/][Helm]], mainly due
to it being aesthetically noisy, and that I didn't fully take advantage of
all its features (which are numerous). Here are some customization's that
made the transition a bit easier.
#+begin_src emacs-lisp
;; Incremental Vertical completion
(use-package ivy
:bind ("C-x b" . ivy-switch-buffer)
:config
(setq ivy-wrap t ; Easier access to the last candidate
ivy-height 25 ; Give me more candidates to look at
ivy-use-virtual-buffers t ; C-x b displays recents and bookmarks
ivy-count-format "(%d/%d) " ; Display both the index and the count
ivy-on-del-error-function 'ignore ; Lets me hold in backspace
ivy-posframe-min-width 100 ; Keep ivy reasonably narrow
ivy-posframe-height ivy-height ; Maintain the height given by ivy
ivy-virtual-abbreviate 'abbreviate) ; Disambiguate same file in different dirs
(ivy-mode 1))
#+end_src
The completions are centered in a posframe (a frame at point).
#+begin_src emacs-lisp
;; Using posframe to show Ivy
(use-package ivy-posframe
:config
(ivy-posframe-mode 1))
#+end_src
Use counsel for =M-x=, yanking and finding files.
#+begin_src emacs-lisp
;; Various completion functions using Ivy
(use-package counsel
:bind
(("M-x" . counsel-M-x)
("M-y" . counsel-yank-pop)
("C-x C-f" . counsel-find-file)))
#+end_src
Use swiper for fancy search.
#+begin_src emacs-lisp
;; Isearch with an overview. Oh, man!
(use-package swiper
:bind ("C-c i" . swiper-isearch))
#+end_src
Have Ivy play with nice with Projectile.
#+begin_src emacs-lisp
;; Ivy integration for Projectile
(use-package counsel-projectile
:config
(counsel-projectile-mode 1))
#+end_src
Use smex to prioritize frequently used commands.
#+begin_src emacs-lisp
;; M-x interface with Ido-style fuzzy matching
(use-package smex)
#+end_src
* PDF Tools
[[https://github.com/politza/pdf-tools][PDF Tools]] makes a huge improvement on the built-in [[http://www.gnu.org/software/emacs/manual/html_node/emacs/Document-View.html][doc-view-mode]]; the only
drawback is the =pdf-tools-install= (which has to be executed before the
package can be used) takes a couple of /seconds/ to execute. Instead of
running it at init-time, we'll run it whenever a PDF is opened. Note that
it's only slow on the first run!
#+begin_src emacs-lisp
(pdf-loader-install)
#+end_src
#+begin_src emacs-lisp
(add-hook 'pdf-view-mode-hook
(lambda () (setq header-line-format nil)))
#+end_src
* Completion
[[https://github.com/auto-complete/auto-complete][Auto-Complete]] has been a part of my config for years, but I want to try out
[[http://company-mode.github.io/][company-mode]]. If I code in an environment with good completion, I've made an
habit of trying to /guess/ function-names, and looking at the completions for
the right one. So I want a pretty aggressive completion system, hence the no
delay settings and short prefix length.
#+begin_src emacs-lisp
;; Modular text completion framework
(use-package company
:bind (;; Note that a bug (probably) makes the quotes necessary here
:map company-active-map
("C-d" . 'company-show-doc-buffer)
("C-n" . 'company-select-next)
("C-p" . 'company-select-previous))
:config
(setq company-idle-delay 0
company-echo-delay 0
company-dabbrev-downcase nil
company-minimum-prefix-length 2
company-selection-wrap-around t
company-transformers '(company-sort-by-occurrence
company-sort-by-backend-importance))
(global-company-mode 1))
#+end_src
* Spelling
** Flyspell
Flyspell offers on-the-fly spell checking.
When working with several languages, we should be able to cycle through the
languages we most frequently use. Every buffer should have a separate cycle
of languages, so that cycling in one buffer does not change the state in a
different buffer (this problem occurs if you only have one global cycle). We
can implement this by using a [[http://www.gnu.org/software/emacs/manual/html_node/elisp/Closures.html][closure]].
#+begin_src emacs-lisp
(defun cycle-languages ()
"Changes the ispell dictionary to the first element in
ISPELL-LANGUAGES, and returns an interactive function that cycles
the languages in ISPELL-LANGUAGES when invoked."
(let ((ispell-languages (list "american" "norsk")))
(lambda ()
(interactive)
;; Rotates the languages cycle and changes the ispell dictionary.
(let ((rotated (nconc (cdr ispell-languages) (list (car ispell-languages)))))
(ispell-change-dictionary (car (setq ispell-languages rotated)))))))
#+end_src
We enable =flyspell-mode= for all text-modes, and use =flyspell-prog-mode=
for spell checking comments and strings in all programming modes. We bind =C-c l= to a function returned from =cycle-languages=, giving a language
switcher for every buffer where flyspell is enabled.
#+begin_src emacs-lisp
(use-package flyspell
:defer t
:if (executable-find "aspell")
:hook ((text-mode . flyspell-mode)
(prog-mode . flyspell-prog-mode)
(flyspell-mode . (lambda ()
(local-set-key
(kbd "C-c l")
(cycle-languages)))))
:config
(ispell-change-dictionary "american" t))
#+end_src
** Define word
This super neat package looks up the word at point. I use it a lot!
#+begin_src emacs-lisp
;; display the definition of word at point
(use-package define-word
:defer t
:bind ("C-c D" . define-word-at-point))
#+end_src
* Org
I use Org mode extensively. Some of these configurations may be unfortunate,
but it is a bit impractical to change, as I have years worth of org-files and
want to avoid having to reformat a lot of files.
One example is =org-adapt-indentation=, which changed default value in
version 9.5 of Org mode. Another is that I for some unknown reason decided to
content within source content not be indented by two spaces (which is the
default).
Note that I disable some safety features, so please don't copy and paste
mindlessly (see the documentation for =org-confirm-babel-evaluate= and =org-export-allow-bind-keywords=).
#+begin_src emacs-lisp
;; Outline-based notes management and organizer
(use-package org
:defer t
:config
(setq org-adapt-indentation t
org-src-fontify-natively t
org-confirm-babel-evaluate nil
org-export-allow-bind-keywords t
org-edit-src-content-indentation 0))
#+end_src
** LaTeX export
For LaTeX export, I default to using XeLaTeX for compilation, and the [[https://www.ctan.org/pkg/minted][minted]]
package for syntax highlighting source blocks. I have PDFs open directly in
Emacs ([[PDF Tools]]). In addition, I have support for a couple of custom LaTeX
classes.
#+begin_src emacs-lisp
;; LaTeX Back-End for Org Export Engine
(use-package ox-latex
:ensure nil
:after org
:config
;; Use Minted and XeLaTeX
(setq org-latex-src-block-backend 'minted
org-latex-packages-alist '(("newfloat" "minted"))
org-latex-pdf-process
'("latexmk -pdflatex='xelatex -shell-escape -interaction nonstopmode' -pdf -f %f"))
(add-to-list 'org-file-apps '("\\.pdf\\'" . emacs))
(add-to-list 'org-latex-classes
'("ifimaster"
"\\documentclass{ifimaster}
[DEFAULT-PACKAGES]
[PACKAGES]
[EXTRA]
\\usepackage{babel,csquotes,ifimasterforside,url,varioref}"
("\\chapter{%s}" . "\\chapter*{%s}")
("\\section{%s}" . "\\section*{%s}")
("\\subsection{%s}" . "\\subsection*{%s}")
("\\subsubsection{%s}" . "\\subsubsection*{%s}")
("\\paragraph{%s}" . "\\paragraph*{%s}")
("\\subparagraph{%s}" . "\\subparagraph*{%s}")))
(add-to-list 'org-latex-classes
'("easychair" "\\documentclass{easychair}"
("\\section{%s}" . "\\section*{%s}")
("\\subsection{%s}" . "\\subsection*{%s}")
("\\subsubsection{%s}" . "\\subsubsection*{%s}")
("\\paragraph{%s}" . "\\paragraph*{%s}")
("\\subparagraph{%s}" . "\\subparagraph*{%s}"))))
#+end_src
** Babel
Add a few languages for org-babel.
#+begin_src emacs-lisp
;; Working with Code Blocks in Org
(use-package ob
:ensure nil
:after org
:config
(org-babel-do-load-languages
'org-babel-load-languages
'((emacs-lisp . t)
(python . t)
(clojure . t)
(chatgpt-shell . t))))
#+end_src
Default to use whatever interpreter is set by =python-shell-interpreter=.
#+begin_src emacs-lisp
;; Babel Functions for Python
(use-package ob-python
:ensure nil
:after (ob python)
:config
(setq org-babel-python-command python-shell-interpreter))
#+end_src
** Tempo
Since version 9.2 of Org mode, typing =<s= to get a source block (and
similar variants) has been tucked away in the Org Tempo library, hoping that
users rather use =C-c C-,=. Hopefully I'll stop typing =<s= at some point,
and adapt the much saner =C-c C-,=.
#+begin_src emacs-lisp
;; Template expansion for Org structures
(use-package org-tempo
:ensure nil
:after org)
#+end_src
** Bullets
Touch up the headings a bit, with some fancy UTF-8 characters.
#+begin_src emacs-lisp
;; Show bullets in org-mode as UTF-8 characters
(use-package org-bullets
:after org
:hook (org-mode . (lambda () (org-bullets-mode 1))))
#+end_src
** Email with org mode
The package org-msg allows me to compose emails with Org mode. That means I
easily can add headings, tables, source code, etc. It is really neat.
#+begin_src emacs-lisp
;; Org mode to send and reply to email in HTML
(use-package org-msg
:after (org mu4e)
:config
(add-to-list 'mu4e-compose-pre-hook 'org-msg-mode)
(setq org-msg-enforce-css (concat user-emacs-directory "email-style.css")
org-msg-options "html-postamble:nil toc:nil num:nil author:nil email:nil"
org-msg-default-alternatives '((new . (text html))
(reply-to-html . (text html))
(reply-to-text . (text)))
org-msg-signature "
,,#+begin_signature
,,#+begin_export html
- Lars
,,#+end_export
,,#+end_signature\n"))
#+end_src
** GitHub flavored markdown
I guess I have to include my (semi-abandoned) mode [[https://github.com/larstvei/ox-gfm][ox-gfm]] for exporting org
mode to GitHub Flavored Markdown.
#+begin_src emacs-lisp
;; Export Github Flavored Markdown from Org
(use-package ox-gfm
:after (org))
#+end_src
* Markdown
#+begin_src emacs-lisp
;; Emacs Major mode for Markdown-formatted files
(use-package markdown-mode
:defer t)
#+end_src
* Direnv
I use [[https://direnv.net][direnv]] in combination with [[https://nixos.org][nix]] to allow for programs to only be
available in certain directories. The [[https://github.com/wbolster/emacs-direnv][emacs-direnv]] makes Emacs play nice with
direnv, so that it for instance can detect a language server that is only
available within some project. The =direnv-always-show-summary= is set to =nil= to avoid having long messages pop up in the messages buffer whenever I
enter a directory that interacts with direnv.
#+begin_src emacs-lisp
;; direnv integration
(use-package direnv
:config
(setq direnv-always-show-summary nil)
(direnv-mode 1))
#+end_src
* Email
I've used Emacs for email in the past, where I've always had the need for a
more standard email client in addition. I'm going to give it another go.
#+begin_src emacs-lisp
(defvar load-mail-setup
(and (file-exists-p "~/Maildir")
(executable-find "mbsync")
(executable-find "msmtp")
(executable-find "mu")))
#+end_src
I use [[http://www.djcbsoftware.nl/code/mu/mu4e.html][mu4e]] (which is a part of [[http://www.djcbsoftware.nl/code/mu/][mu]]) along with [[https://isync.sourceforge.io/][mbsync]].
#+begin_src emacs-lisp
(when load-mail-setup
(with-eval-after-load 'mu4e
(setq
mail-user-agent 'mu4e-user-agent
user-full-name "Lars Tveito" ; Your full name
user-mail-address "larstvei@ifi.uio.no" ; And email-address
sendmail-program (executable-find "msmtp")
send-mail-function 'smtpmail-send-it
message-sendmail-f-is-evil t
message-sendmail-extra-arguments '("--read-envelope-from")
message-send-mail-function 'message-send-mail-with-sendmail
message-kill-buffer-on-exit t
mu4e-get-mail-command (concat (executable-find "mbsync") " -a")
mu4e-change-filenames-when-moving t
mu4e-user-mail-address-list '("larstvei@ifi.uio.no")
mu4e-maildir-shortcuts '(("/Inbox" . ?i) ("/Sent Items" . ?s))
mu4e-sent-folder "/Sent Items"
mu4e-trash-folder "/Deleted Items"
mu4e-trash-folder "/Drafts"
mu4e-use-fancy-chars t))
(autoload 'mu4e "mu4e" nil t))
#+end_src
* ChatGPT
I have a line like this:
#+begin_example
machine api.openai.com password OPEN-AI-KEY
#+end_example
in my ~.authinfo~ file. Then the ~chatgpt-shell-openai-key~ can by set by:
#+begin_src emacs-lisp
;; Interaction mode for ChatGPT
(use-package chatgpt-shell
:defer t
:config
(setq chatgpt-shell-openai-key
(auth-source-pick-first-password
:host "api.openai.com")))
#+end_src
#+begin_src emacs-lisp
;; Org babel functions for ChatGPT evaluation
(use-package ob-chatgpt-shell
:after ob)
#+end_src
* Multiple cursors
I use this /all the time/. Perhaps more than I should?
#+begin_src emacs-lisp
;; Multiple cursors for Emacs
(use-package multiple-cursors
:defer t
:bind (("C-c e" . mc/edit-lines)
("C-c a" . mc/mark-all-like-this)
("C-c n" . mc/mark-next-like-this)))
#+end_src
* Expand region
This is neat, and I use it way less than I should.
#+begin_src emacs-lisp
;; Increase selected region by semantic units
(use-package expand-region
:defer t
:bind ("C-=" . er/expand-region))
#+end_src
* Interactive functions
<<sec:defuns>> =just-one-space= removes all whitespace around a point - giving it a negative
argument it removes newlines as well. We wrap a interactive function around
it to be able to bind it to a key. In Emacs 24.4 =cycle-spacing= was
introduced, and it works like =just-one-space=, but when run in succession it
cycles between one, zero and the original number of spaces.
#+begin_src emacs-lisp
(defun cycle-spacing-delete-newlines ()
"Removes whitespace before and after the point."
(interactive)
(if (version< emacs-version "24.4")
(just-one-space -1)
(cycle-spacing -1)))
#+end_src
Often I want to find other occurrences of a word I'm at, or more specifically
the symbol (or tag) I'm at. The =isearch-forward-symbol-at-point= in Emacs
24.4 works well for this, but I don't want to be bothered with the =isearch=
interface. Rather jump quickly between occurrences of a symbol, or if non is
found, don't do anything.
#+begin_src emacs-lisp
(defun jump-to-symbol-internal (&optional backwardp)
"Jumps to the next symbol near the point if such a symbol
exists. If BACKWARDP is non-nil it jumps backward."
(let* ((point (point))
(bounds (find-tag-default-bounds))
(beg (car bounds)) (end (cdr bounds))
(str (isearch-symbol-regexp (find-tag-default)))
(search (if backwardp 'search-backward-regexp
'search-forward-regexp)))
(goto-char (if backwardp beg end))
(funcall search str nil t)
(cond ((<= beg (point) end) (goto-char point))
(backwardp (forward-char (- point beg)))
(t (backward-char (- end point))))))
(defun jump-to-previous-like-this ()
"Jumps to the previous occurrence of the symbol at point."
(interactive)
(jump-to-symbol-internal t))
(defun jump-to-next-like-this ()
"Jumps to the next occurrence of the symbol at point."
(interactive)
(jump-to-symbol-internal))
#+end_src
I sometimes regret killing the =*scratch*=-buffer, and have realized I never
want to actually kill it. I just want to get it out of the way, and clean it
up. The function below does just this for the =*scratch*=-buffer, and works
like =kill-this-buffer= for any other buffer. It removes all buffer content
and buries the buffer (this means making it the least likely candidate for =other-buffer=).
#+begin_src emacs-lisp
(defun kill-this-buffer-unless-scratch ()
"Works like `kill-this-buffer' unless the current buffer is the
,*scratch* buffer. In witch case the buffer content is deleted and
the buffer is buried."
(interactive)
(if (not (string= (buffer-name) "*scratch*"))
(kill-this-buffer)
(delete-region (point-min) (point-max))
(switch-to-buffer (other-buffer))
(bury-buffer "*scratch*")))
#+end_src
To duplicate either selected text or a line we define this interactive
function.
#+begin_src emacs-lisp
(defun duplicate-thing (comment)
"Duplicates the current line, or the region if active. If an argument is
given, the duplicated region will be commented out."
(interactive "P")
(save-excursion
(let ((start (if (region-active-p) (region-beginning) (line-beginning-position)))
(end (if (region-active-p) (region-end) (line-end-position)))
(fill-column most-positive-fixnum))
(goto-char end)
(unless (region-active-p)
(newline))
(insert (buffer-substring start end))
(when comment (comment-region start end)))))
#+end_src
To tidy up a buffer we define this function borrowed from [[https://github.com/simenheg][simenheg]].
#+begin_src emacs-lisp
(defun tidy ()
"Ident, untabify and unwhitespacify current buffer, or region if active."
(interactive)
(let ((beg (if (region-active-p) (region-beginning) (point-min)))
(end (if (region-active-p) (region-end) (point-max))))
(indent-region beg end)
(whitespace-cleanup)
(untabify beg (if (< end (point-max)) end (point-max)))))
#+end_src
Org mode does currently not support synctex (which enables you to jump from a
point in your TeX-file to the corresponding point in the pdf), and it [[http://comments.gmane.org/gmane.emacs.orgmode/69454][seems
like a tricky problem]].
Calling this function from an org-buffer jumps to the corresponding section
in the exported pdf (given that the pdf-file exists), using pdf-tools.
#+begin_src emacs-lisp
(defun org-sync-pdf ()
(interactive)
(let ((headline (nth 4 (org-heading-components)))
(pdf (concat (file-name-base (buffer-name)) ".pdf")))
(when (file-exists-p pdf)
(find-file-other-window pdf)
(pdf-links-action-perform
(cl-find headline (pdf-info-outline pdf)
:key (lambda (alist) (cdr (assoc 'title alist)))
:test 'string-equal)))))
#+end_src
The opposite of fill paragraph (from [[https://www.emacswiki.org/emacs/UnfillParagraph][EmacsWiki]]),
#+begin_src emacs-lisp
(defun unfill-paragraph ()
(interactive)
(let ((fill-column most-positive-fixnum))
(fill-paragraph nil (region-active-p))))
#+end_src
* Advice
An advice can be given to a function to make it behave differently. This
advice makes =eval-last-sexp= (bound to =C-x C-e=) replace the sexp with the
value.
#+begin_src emacs-lisp
(defadvice eval-last-sexp (around replace-sexp (arg) activate)
"Replace sexp when called with a prefix argument."
(if arg
(let ((pos (point)))
ad-do-it
(goto-char pos)
(backward-kill-sexp)
(forward-sexp))
ad-do-it))
#+end_src
When interactively changing the theme (using =M-x load-theme=), the current
custom theme is not disabled. This often gives weird-looking results; we can
advice =load-theme= to always disable themes currently enabled themes.
#+begin_src emacs-lisp
(defadvice load-theme
(before disable-before-load (theme &optional no-confirm no-enable) activate)
(mapc 'disable-theme custom-enabled-themes))
#+end_src
* global-scale-mode
These functions provide something close to ~text-scale-mode~, but for every
buffer, including the minibuffer and mode line.
#+begin_src emacs-lisp
(let* ((default (face-attribute 'default :height))
(size default))
(defun global-scale-default ()
(interactive)
(global-scale-internal (setq size default)))
(defun global-scale-up ()
(interactive)
(global-scale-internal (setq size (+ size 20))))
(defun global-scale-down ()
(interactive)
(global-scale-internal (setq size (- size 20))))
(defun global-scale-internal (arg)
(set-face-attribute 'default (selected-frame) :height arg)
(set-temporary-overlay-map
(let ((map (make-sparse-keymap)))
(define-key map (kbd "C-=") 'global-scale-up)
(define-key map (kbd "C-+") 'global-scale-up)
(define-key map (kbd "C--") 'global-scale-down)
(define-key map (kbd "C-0") 'global-scale-default) map))))
#+end_src
* Mode specific
** Eglot
I am using [[https://joaotavora.github.io/eglot/][eglot]], which is built in from [[https://git.savannah.gnu.org/cgit/emacs.git/tree/etc/NEWS?h=emacs-29#n3273][emacs 29.1]]. Some performance issues
led me to set =eglot-events-buffer-size= to 0.
#+begin_src emacs-lisp
(setq eglot-events-buffer-size 0)
(add-hook 'eglot-managed-mode-hook (lambda () (eglot-inlay-hints-mode -1)))
#+end_src
** Compilation
I often run ~latexmk -pdf -pvc~ in a compilation buffer, which recompiles
the latex-file whenever it is changed. This often results in annoyingly
large compilation buffers; the following snippet limits the buffer size in
accordance with ~comint-buffer-maximum-size~, which defaults to 1024 lines.
#+begin_src emacs-lisp
(add-hook 'compilation-filter-hook 'comint-truncate-buffer)
#+end_src
** vterm
Inspired by [[https://github.com/torenord/.emacs.d][torenord]], I maintain quick access to shell buffers with bindings ~M-1~ to ~M-9~. In addition, the ~M-z~ toggles between the last visited
shell, and the last visited non-shell buffer. The following functions
facilitate this, and are bound in the [[Key bindings]] section.
#+begin_src emacs-lisp
(let ((last-vterm ""))
(defun toggle-vterm ()
(interactive)
(cond ((string-match-p "^\\vterm<[1-9][0-9]*>$" (buffer-name))
(goto-non-vterm-buffer))
((get-buffer last-vterm) (switch-to-buffer last-vterm))
(t (vterm (setq last-vterm "vterm<1>")))))
(defun goto-non-vterm-buffer ()
(let* ((r "^\\vterm<[1-9][0-9]*>$")
(vterm-buffer-p (lambda (b) (string-match-p r (buffer-name b))))
(non-vterms (cl-remove-if vterm-buffer-p (buffer-list))))
(when non-vterms
(switch-to-buffer (car non-vterms)))))
(defun switch-vterm (n)
(let ((buffer-name (format "vterm<%d>" n)))
(setq last-vterm buffer-name)
(cond ((get-buffer buffer-name)
(switch-to-buffer buffer-name))
(t (vterm buffer-name)
(rename-buffer buffer-name))))))
#+end_src
Don't query whether or not the ~shell~-buffer should be killed, just kill
it.
#+begin_src emacs-lisp
(defadvice vterm (after kill-with-no-query nil activate)
(set-process-query-on-exit-flag (get-buffer-process ad-return-value) nil))
#+end_src
** Lisp
I use =Paredit= when editing lisp code, we enable this for all lisp-modes.
#+begin_src emacs-lisp
(dolist (mode '(cider-repl-mode
clojure-mode
ielm-mode
racket-mode
racket-repl-mode
slime-repl-mode
lisp-mode
emacs-lisp-mode
lisp-interaction-mode
scheme-mode))
;; add paredit-mode to all mode-hooks
(add-hook (intern (concat (symbol-name mode) "-hook")) 'paredit-mode))
#+end_src
Paredit version 25 seems to interfere with REPL-modes. This is the proposed
fix:
#+begin_src emacs-lisp
(with-eval-after-load 'paredit
(define-key paredit-mode-map (kbd "RET") nil))
#+end_src
*** Emacs Lisp
In =emacs-lisp-mode= we can enable =eldoc-mode= to display information
about a function or a variable in the echo area.
#+begin_src emacs-lisp
(add-hook 'emacs-lisp-mode-hook 'turn-on-eldoc-mode)
(add-hook 'lisp-interaction-mode-hook 'turn-on-eldoc-mode)
#+end_src
*** Clojure
A very simple setup for Clojure. Cider works pretty much out of the box!
#+begin_src emacs-lisp
;; Clojure Interactive Development Environment
(use-package cider
:defer t
:bind (:map cider-repl-mode-map ("C-l" . cider-repl-clear-buffer)))
#+end_src
#+begin_src emacs-lisp
;; Commands for refactoring Clojure code
(use-package clj-refactor
:defer t)
#+end_src
*** Common lisp
I use [[http://www.common-lisp.net/project/slime/][Slime]] along with =lisp-mode= to edit Common Lisp code. Slime provides
code evaluation and other great features, a must have for a Common Lisp
developer. [[http://www.quicklisp.org/beta/][Quicklisp]] is a library manager for Common Lisp, and you can
install Slime following the instructions from the site along with this
snippet.
#+begin_src emacs-lisp
(defun activate-slime-helper ()
(when (file-exists-p "~/.quicklisp/slime-helper.el")
(load (expand-file-name "~/.quicklisp/slime-helper.el"))
(define-key slime-repl-mode-map (kbd "C-l")
'slime-repl-clear-buffer))
(remove-hook 'common-lisp-mode-hook #'activate-slime-helper))
(add-hook 'common-lisp-mode-hook #'activate-slime-helper)
#+end_src
We can specify what Common Lisp program Slime should use (I use SBCL).
#+begin_src emacs-lisp
(setq inferior-lisp-program "sbcl")
#+end_src
More sensible =loop= indentation, borrowed from [[https://github.com/simenheg][simenheg]].
#+begin_src emacs-lisp
(setq lisp-loop-forms-indentation 6
lisp-simple-loop-indentation 2
lisp-loop-keyword-indentation 6)
#+end_src
** Python
#+begin_src emacs-lisp
(setq python-shell-interpreter "python3.10")
(add-hook 'python-mode-hook
(lambda () (setq forward-sexp-function nil)))
#+end_src
** C
The =c-mode-common-hook= is a general hook that work on all C-like languages
(C, C++, Java, etc...). I like being able to quickly compile using =C-c C-c=
(instead of =M-x compile=), a habit from =latex-mode=.
#+begin_src emacs-lisp
(defun c-setup ()
(local-set-key (kbd "C-c C-c") 'compile))
(add-hook 'c-mode-hook 'c-setup)
#+end_src
** Java
Some statements in Java appear often, and become tedious to write out. We
can use abbrevs to speed this up.
#+begin_src emacs-lisp
(define-abbrev-table 'java-mode-abbrev-table
'(("psv" "public static void main(String[] args) {" nil 0)
("sopl" "System.out.println" nil 0)
("sop" "System.out.printf" nil 0)))
#+end_src
To be able to use the abbrev table defined above, =abbrev-mode= must be
activated.
#+begin_src emacs-lisp
(add-hook 'java-mode-hook 'eglot-ensure)
#+end_src
** Assembler
When writing assembler code I use =#= for comments. By defining =comment-start= we can add comments using =M-;= like in other programming
modes. Also in assembler should one be able to compile using =C-c C-c=.
#+begin_src emacs-lisp
(defun asm-setup ()
(setq comment-start "#")
(local-set-key (kbd "C-c C-c") 'compile))
(add-hook 'asm-mode-hook 'asm-setup)
#+end_src
** LaTeX
#+begin_src emacs-lisp
;; Integrated environment for *TeX*
(use-package auctex
:defer t)
#+end_src
** Erlang
Erlang mode works out of the box.
#+begin_src emacs-lisp
;; Erlang major mode
(use-package erlang
:defer t)
#+end_src
** Nix
#+begin_src emacs-lisp
;; Major mode for editing .nix files
(use-package nix-mode
:defer t
:hook (nix-mode . eglot-ensure))
#+end_src
** Haskell =haskell-doc-mode= is similar to =eldoc=, it displays documentation in the
echo area. Haskell has several indentation modes - I prefer using =haskell-indent=.
#+begin_src emacs-lisp
;; A Haskell editing mode
(use-package haskell-mode
:defer t
:hook ((haskell-mode . interactive-haskell-mode)
(haskell-mode . turn-on-haskell-doc-mode)
(haskell-mode . turn-on-haskell-indent)))
#+end_src
** Maude
Use =---= for comments in Maude.
#+begin_src emacs-lisp
;; Emacs mode for the programming language Maude
(use-package maude-mode
:defer t
:hook (maude-mode . (lambda () (setq-local comment-start "---")))
:config
(add-to-list 'maude-command-options "-no-wrap"))
#+end_src
** Minizinc
Provide a default =compile-command=.
#+begin_src emacs-lisp
(defun minizinc-setup-compile-command ()
(let ((command (concat "minizinc " (buffer-file-name) " "))
(f (concat (file-name-base (buffer-file-name)) ".dzn")))
(local-set-key (kbd "C-c C-c") 'recompile)
(setq-local compile-command (concat command (if (file-exists-p f) f "")))))
#+end_src
Use =minizinc-mode=, and hook up the =minizinc-setup-compile-command= above.
#+begin_src emacs-lisp
;; Major mode for MiniZinc code
(use-package minizinc-mode
:defer t
:mode "\\.mzn\\'"
:hook (minizinc-mode . minizinc-setup-compile-command))
#+end_src
** Coq
#+begin_src emacs-lisp
;; A collection of extensions PG's Coq mode
(use-package company-coq
:hook (coq-mode . company-coq-mode))
#+end_src
** Rust
#+begin_src emacs-lisp
(setq rustic-lsp-client 'eglot)
#+end_src
** Go
#+begin_src emacs-lisp
;; Major mode for the Go programming language
(use-package go-mode
:defer t
:mode "\\.go\\'"
:hook (go-mode . eglot-ensure))
#+end_src
** Lua
#+begin_src emacs-lisp
;; a major-mode for editing Lua scripts
(use-package lua-mode
:defer t)
#+end_src
** Webdev
My webdev setup isn't much, but with eglot, I don't find myself missing
much. It depends on [[https://tree-sitter.github.io/tree-sitter/][Tree-sitter]], which was added in [[https://git.savannah.gnu.org/cgit/emacs.git/tree/etc/NEWS?h=emacs-29#n36][emacs 29.1]].
The following is what I use for plain Javascript:
#+begin_src emacs-lisp
(add-to-list 'auto-mode-alist '("\\.jsx?\\'" . js-ts-mode))
(add-hook 'js-ts-mode-hook 'eglot-ensure)
#+end_src
Similarly for Typescript:
#+begin_src emacs-lisp
(add-to-list 'auto-mode-alist '("\\.tsx?\\'" . tsx-ts-mode))
(add-hook 'tsx-ts-mode-hook 'eglot-ensure)
(setq typescript-ts-mode-indent-offset 4)
#+end_src
I am using [[https://svelte.dev][Svelte]] for some projects, where I find [[https://web-mode.org][web-mode]] along with the
[[https://github.com/sveltejs/language-tools][Svelte Language Server]] to work well.
#+begin_src emacs-lisp
(add-to-list 'auto-mode-alist '("\\.svelte\\'" . web-mode))
(add-hook 'web-mode-hook 'eglot-ensure)
(with-eval-after-load "web-mode"
(add-to-list 'web-mode-engines-alist '("svelte" . "\\.svelte\\'")))
(with-eval-after-load "eglot"
(add-to-list 'eglot-server-programs
'(web-mode . ("svelteserver" "--stdio"))))
#+end_src
* Key bindings
Inspired by [[http://stackoverflow.com/questions/683425/globally-override-key-binding-in-emacs][this StackOverflow post]] I keep a =custom-bindings-map= that holds
all my custom bindings. This map can be activated by toggling a simple =minor-mode= that does nothing more than activating the map. This inhibits
other =major-modes= to override these bindings. I keep this at the end of the
init-file to make sure that all functions are actually defined.
#+begin_src emacs-lisp
(defvar custom-bindings-map (make-keymap)
"A keymap for custom bindings.")
#+end_src
** Bindings for [[https://github.com/magnars/multiple-cursors.el][multiple-cursors]]
#+begin_src emacs-lisp
(define-key custom-bindings-map (kbd "C-c e") 'mc/edit-lines)
(define-key custom-bindings-map (kbd "C-c a") 'mc/mark-all-like-this)
(define-key custom-bindings-map (kbd "C-c n") 'mc/mark-next-like-this)
#+end_src
** Bindings for mu4e
#+begin_src emacs-lisp
(define-key custom-bindings-map (kbd "C-x m") 'mu4e)
#+end_src
** Bindings for built-ins
#+begin_src emacs-lisp
(define-key custom-bindings-map (kbd "M-u") 'upcase-dwim)
(define-key custom-bindings-map (kbd "M-c") 'capitalize-dwim)
(define-key custom-bindings-map (kbd "M-l") 'downcase-dwim)
(define-key custom-bindings-map (kbd "M-]") 'other-frame)
(define-key custom-bindings-map (kbd "C-j") 'newline-and-indent)
(define-key custom-bindings-map (kbd "C-c s") 'ispell-word)
(define-key comint-mode-map (kbd "C-l") 'comint-clear-buffer)
#+end_src
** Bindings for functions defined [[sec:defuns][above]].
#+begin_src emacs-lisp
(define-key global-map (kbd "M-p") 'jump-to-previous-like-this)
(define-key global-map (kbd "M-n") 'jump-to-next-like-this)
(define-key custom-bindings-map (kbd "M-,") 'jump-to-previous-like-this)
(define-key custom-bindings-map (kbd "M-.") 'jump-to-next-like-this)
(define-key custom-bindings-map (kbd "C-c .") (cycle-themes))
(define-key custom-bindings-map (kbd "C-x k") 'kill-this-buffer-unless-scratch)
(define-key custom-bindings-map (kbd "C-c C-0") 'global-scale-default)
(define-key custom-bindings-map (kbd "C-c C-=") 'global-scale-up)
(define-key custom-bindings-map (kbd "C-c C-+") 'global-scale-up)
(define-key custom-bindings-map (kbd "C-c C--") 'global-scale-down)
(define-key custom-bindings-map (kbd "C-c j") 'cycle-spacing-delete-newlines)
(define-key custom-bindings-map (kbd "C-c d") 'duplicate-thing)
(define-key custom-bindings-map (kbd "<C-tab>") 'tidy)
(define-key custom-bindings-map (kbd "C-z") 'toggle-vterm)
(dolist (n (number-sequence 1 9))
(let ((key (concat "M-" (int-to-string n))))
(with-eval-after-load 'vterm
(define-key vterm-mode-map (kbd key) nil))
(define-key custom-bindings-map (kbd key)
(lambda () (interactive) (switch-vterm n)))))
(with-eval-after-load 'org
(define-key org-mode-map (kbd "C-'") 'org-sync-pdf))
#+end_src
Lastly we need to activate the map by creating and activating the =minor-mode=.
#+begin_src emacs-lisp
(define-minor-mode custom-bindings-mode
"A mode that activates custom-bindings."
t nil custom-bindings-map)
#+end_src
* License
My Emacs configurations written in Org mode.
Copyright (c) 2013 - 2023 Lars Tveito
This program is free software: you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation, either version 3 of the License, or (at your option) any later
version.
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
details.
You should have received a copy of the GNU General Public License along with
this program. If not, see <http://www.gnu.org/licenses/>.