.vim

This is my .vim directory, containing all my configurations for vim and the associated plugins I use.

Installation

Clone the repository:

git clone git://github.com/sds/.vim.git ~/.vim

Add a symlink to init.vim...

ln -s ~/.vim/init.vim ~/.vimrc

Alternatively, you can source it at the beginning of your current .vimrc if you want to keep your current settings along with the ones defined in this repository:

source ~/.vim/init.vim

Finally, run the update script to get all the plugins:

cd ~/.vim && ./update

You can run the update script at any time to fetch the latest changes from each of the plugin repositories. If you have a lot of plugins and want to speed this up, you could manually specify a revision of the plugin that you want, so that it only updates when you change the revision.

Configuration

Configuration files are broken up into various categories in the config directory, including settings for individual plugins (if you're familiar with the directory layout for vim, config replaces the plugin directory).

The after, autoload, ftdetect, ftplugin, and indent directories act in the way vim normally uses them. To summarize:

• after: Initialization to run "after" vim has finished loading. Useful for defining shortcuts that override default key mappings.

• autoload: Contains scripts which are "lazy-loaded". For example, when calling pathogen#infect(), vim looks for the infect function in autoload/pathogen.vim.

• ftdetect: Any scripts that are used to determine the type of a file.

• ftplugin: Any plugins that are only run for a specific filetype. For example, if vim determines a file is of type html, it will run the commands found in ftplugin/html.vim.

• indent: The same as ftplugin, except specifically for setting up indent settings for specific filetypes.

Managing Plugins

The plugins.list file contains the locations of the repositories for the plugins. Plugins are stored in the bundles directory, which is read by Pathogen to load all the plugins.

To remove a plugin, simply remove it from plugins.list and run the update script.

Why not use submodules?

When I was first exposed to using submodules for managing plugins with Pathogen, I found them clunky and annoying. For example, the documentation tag files would dirty the working tree, even though I didn't care that the tree of the submodule was dirty.

While better alternatives may exist (I believe you can ignore dirty trees in submodules from the containing repository with a recent version of git), I find the approach of explicitly listing your plugins in a file and running a script simpler and cleaner.

License

WTFPL