Skip to content

Latest commit

 

History

History
64 lines (53 loc) · 2.92 KB

README.markdown

File metadata and controls

64 lines (53 loc) · 2.92 KB
Raw

dot-emacs

This emacs setup was bootstrapped with techomancy's emacs-starter-kit. It has been updated to use version 2 of the starter kit.

The goal of this project is to facilitate the sharing of a common emacs setup within a company (or anyone really). ELPA packages are the preferred way to bring in dependencies but in many cases this is not possible so git submodules or just committed files are used. See the section on Structure for more details.

Installation

You'll need Emacs 24, which comes with package.el. It's not hard to compile from source, but precompiled versions are readily available for Debian-based systems, Mac OS X, and Windows.

Clone this project as ~/.emacs.d, checkout the git submodules, and launch emacs!

Structure and Sharing

The layout of your .emacs.d will look something like this:

.emacs.d
|-- init.el
|-- initializers.available
|   |-- some-framework.el
|    -- some-language.el
|   initializers.enabled
|    -- some-language.el -> ../initializers.available/some-language.el
|-- my-host-name.el (optional)
|-- my-user-name
|   |-- custom.el
|   |-- my-super-config.el (optional)
|    -- random-function.el (optional)
|-- vendor
    |-- some-git-submodule
     -- some-shared-file.el

The init.el file is where everything begins and is shared by everyone. As such, this file should remain as general as possible and not include user specific configuration at all.

The main way to share various customizations and modes is via "initializers". The initializers.available contains all the actual initializers. An initializer may install ELPA packages via the provided ensure-packages-installed function, or append to the load-path code from the vendor path. Either way, the initializers should try to use autoload instead of require or load (unless it is bringing in an autoload file). The best way to get an idea of what an initializer does is just to look at a few. To enable an initializer you need to add a corresponding symlink to it in initializers.enabled. Upon startup all the initializers found in initializers.enabled will be loaded.

Your user-specific-config should be held in either ~/.emacs.d/your-user-name.el or in files located in ~/.emacs.d/your-user-name. Additionally, the Starter Kit also allows host-specific configuration by loading ~/.emacs.d/your-hostname.el. Emacs' "custom" library is configured to save its settings to ~/.emcas.d/your-user-name/custom.el. The advantage of this is that it won't clutter any of your regular configuration files. Having all of your personal customizations namespaced by your login and hostnames allows for easy sharing and keeping in sync with the main repository.