comma-mode

A modal editing minor-mode designed with Emacs in mind.

Overview

Comma-mode is minor mode for Emacs that defines a single key map to facilitate fast modal editing. Most of the modal commands our derived from key bindings that come with a standard Emacs installation. For example, C-f is bound to f in the comma-mode key map to move forward a character. This allows to use the same mnemonics that one has learned from vanilla Emacs to navigate a buffer but without always having to hold down the control key.

Setup

Place comma-mode.el in the lisp folder in your .emacs.d directory. Then add the following somewhere in your init.el file:

(require 'comma-mode)

By default comma-mode does its best to not interfere with your emacs setup as you need to manually activate it. The first thing that you will want to do is assign a key to activate comma-mode while in a Emacs buffer. For example, I use the following:

(global-set-key (kbd "C-,") 'comma-mode)

This sets the control plus the comma key to the global key map to activate comma-mode (hence the name). Of course, when you visit a buffer comma-mode will not automatically activate. This is to facilitate learning comma-mode incrementally but soon you may want to have comma-mode activate automatically when you visit certain buffers. For example, I have the following in my init.el file for when I visit a buffer that is a text, programming language, or org file:

(dolist (hook '(emacs-lisp-mode-hook
                org-mode-hook
                prog-mode-hook
                text-mode-hook))
  (add-hook hook 'comma-mode))

Comma-mode being a minor mode also allows adding custom command to various keystrokes. This will be discussed at the end of the manual.

Manual

The rest of this document will serve as a manual for comma-mode and is written in a manner to allow the reader to follow along with Emacs. If you haven't already, install comma-mode with the instructions above and open any text file, such as this README, in Emacs or open up the scratch buffer.

Entering and Exiting Comma-Mode

Enter comma-mode with C-, if you matched the setup instructions above. You can also activate comma-mode with the command M-x comma-mode but that's pretty cumbersome so you will most likely will want a key binding.

To exit comma-mode, you can either use:

• Hit C-, again. The same key to enter is the same for exiting.
• i will exit and place the point in front of the current character.
• I is like i but places the point after the current charcter.
• C-g or ESC, ESC, ESC as with vanilla emacs to quit something.

Basic Motions

Simple navigating around a buffer matches Emacs Control + Key movements:

• f move forward one character
• b move backward one character
• n move one line down
• p move one line up

Intermediate Motions

• l move right one word
• h move left one word
• j move down one paragraph
• k move up one paragraph
• a move to the beginning of the current line
• e move to the end of the current line

Programming Motions

Comma-mode was designed with programming in mind so many motions are made to quickly navigate files of programming language text.

• M-l move right one sub-word
• M-h move left one sub-word
• L move right one expression
• H move left one expression
• v jump inside into next expression (useful for lisp)
• V jump outside of current expression
• K jump to the beginning of a function (beginning-of-defun)
• J jump to the end of a function (end-of-defun)

More interesting is that comma-mode allows you to jump directly to the any of the following !@#$%^&*()-=_+[]{}\;:'"<>,./? glyphs by pressing that key. In other words, pressing [ will jump to the next [ in the buffer. Alternatively, you can hold the Meta key while hitting any of those keys to go backwards to the most recent glyph of that key.

Advanced Motions

Quick jumps to any glyph can be done forwards and backwards. This mimics C-s and C-r in Emacs but is for a single character as opposed to searching for an entire word.

• s + any key: jump forward to the next occurrence of that glyph
• r + any key: jump backward to the previous occurrence of that glyph

Finally, we can combine any of the previous jump motions, s or r commands and the jump direct to punctuation commands, with the more motion command:

• m repeat the last jump motion in that direction
• M repeat the last jump motion in the reverse direction

Note that you can repeatedly press m to continually repeat that last jump motion. Furthermore, you can press M to immediately reverse that jump motion to the previous occurrence. This is useful if you overshoot your target location from repeated presses of m. Remember, that this command repeats the previous direction so if you were to search backward for the letter e with the jump motion of r e then repeatingly pressing m will continue to search backwards for the previous e while pressing M will reverse and search forward for the next occurrence of letter e.

Similiar to VIM's % command, comma-mode uses Y, as in Yo-Yo, to jump between matching pairs of brackets [], curley brackes {}, carets <>, and parentheses ().

Other Useful Commands

Comma-mode provides keys for other commonly used Emacs commands:

• y yank
• w kill-region
• d delete-char
• M-j join-line
• ~ swap case of the character at the point
• o insert newline below the point
• O insert newline above the point
• z repeat the last command as in C-x z
• X execute-extended-command
• SPC set-mark-command
• u universal-argument -- Note that repeating u doesn't work like repeating C-u. Also, you can follow u with any numbers as the argument, such as C-u 13.

Using SPC as the set-mark-command deserves special mention since it be combined with other motions to execute specific actions. For example, you can enter the keystrokes SPC . f BSPC with comma-mode active to delete a sentence. In other words, what we have done is:

1. called set-mark-command with SPC
2. jump forward to the next period with .
3. move foward one character past the period with f
4. deleted the current region with BSPC

As for another example, suppose we are working on a function in some programming language and we want to move the whole function to a different part of the file (or another file altogether). We can achieve this with the keystroke sequence K SPC J w. That is jump to the beginning of the function, set the mark, jump to then end of the function, and kill the current region. Now we can where we want to place the function and paste it with y.

This is the closest thing to VIM's text objects that comma-mode has to offer. While technically one more keystroke and has to be done from the beginning of a particular position, I find this method more natural as it encompasses motions that I'm constantly using to navigate and edit text without having to think of the mnemonic to perform a given action.

Leader Commands

Comma-mode does not offer a leader key as in VIM. Instead it uses key combinations with x and c prefix followed by another key to execute various commands. Here are some commonly used commands that ship with comma-mode:

• x s save-buffer
• x f ido-find-file
• x u upcase-region
• x : eval-expression
• x 4 o ido-display-buffer
• x 4 f ido-find-file-other-window
• c g grep
• c c comment-region
• c u uncomment-region

Note that x commands are derived from the default Emacs control chord counterparts. There are several others and you can view the full list at the bottom of comma-mode.el. You can easily define you own key combinations by placing them in your init.el file. For example, you would add the following after (require 'comma-mode) in your init.el file to have the key sequence c s execute the magit-status command (assuming you have magit installed):

(define-key comma-mode-map (kbd "c s") 'magit-status)

This concludes the manual for comma-mode. I hope it brings you as much joy as it has for me!