Writing GNU Emacs Extensions

Major Mode Skeleton

These are the steps involved in defining a major mode.

1. Choose a name. The name for our mode is quip.

2. Create a file called name.el to contain the code for the mode.

3. Define a variable called name-mode-hook. This will contain the user's hook functions to execute when entering the mode.

   (defvar quip-mode-hook nil
     "*List of functions to call when entering Quip mode.")

4. If appropriate, define a mode-specific keymap (see Keymaps later in this chapter). Put it in a variable called name-mode-map. Create a mode's keymap like this:

   (defvar name-mode-map nil
     "Keymap for name major mode.")
   (if name-mode-map
       nil
     (setq name-mode-map (make-keymap))
     (define-key name-mode-map  keysequence  command)
     ...)

   Instead of make-keymap, you could use make-sparse-keymap, which is better suited to keymaps that contain only a few keybindings.

5. If appropriate, define a mode-specific syntax table (see the section called Minor Adjustments in Chapter 7). Put it in a variable named name-mode-syntax-table.

6. If appropriate, define a mode-specific abbrev table. Put it in a variable named name-mode-abbrev-table.

7. Define a command named name-mode. This is the major mode command, and it takes no arguments (unlike a minor mode command, which takes one optional argument). When executed, it should cause the current buffer to enter name-mode by performing the following steps:

   1. It must call kill-all-local-variables, which removes the definitions for all buffer-local variables. This effectively turns off whatever modes,