Recently I've been making an effort to better understand what's happening under the hood of some of the packages I use, and I got curious about how bind-key manages to ensure my bindings are never overridden by minor modes (which can be very annoying).
To understand how bind-key does it you need to learn two things. First, you need to understand the order of priority which Emacs assigns to different keymaps, which is well explained at Mastering Emacs. Second, you should read the bind-key source code.
The technique bind-key uses is to create a minor mode that you bind keys in,
and insert those bindings into the variable
keymaps in that variable take priority over all other minor modes.
The code is actually very simple:
;; First, create the keymap (defvar bosskey-mode-map (make-keymap) "Keymap for bosskey-mode") ;; Next, create the minor mode, switch it on by default, make it global, ;; and assign the keymap to it. (define-minor-mode bosskey-mode "Minor mode for my personal keybindings." :init-value t :global t :keymap bosskey-mode-map) ;; Next, add the keymap to `emulation-mode-map-alists' (add-to-list 'emulation-mode-map-alists `((bosskey-mode . ,bosskey-mode-map))) ;; Finally, bind your keys! (define-key bosskey-mode-map (kbd "C-<return>") 'execute-extended-command)
Now, some people would argue that it's much simpler to just use the bind-key package, but I've found it very helpful to learn this stuff for times when I can't use a package and need to create my own solutions.