Skip to main content

Automate deployment of dotfiles to local paths or remote hosts

Project description

dotlink is a simple script to automate deploying an arbitrary set of “dotfiles” from a repository to either a local path or even a remote host using ssh/scp.

dotlink does not manage the dotfiles themselves, but uses a simple text file mapping dotfile names in the repository to their ultimate location relative to the target path. This allows you to have files without the dot prefix, for instance, or avoid replicating deep directory structures to track a single file.

[![Build Status](https://travis-ci.org/jreese/dotlink.png?branch=master)](https://travis-ci.org/jreese/dotlink)

setup

dotlink requires Python, either version 2.7 or versions 3.2 and newer. No external libraries are required.

To install dotlink system-wide or in a virtualenv:

# pip install dotlink

Or if you want to hack on dotlink a bit, clone the repo and run:

# python setup.py develop

usage

You’ll first need to create a mapping file in the root of your dotfile repository named .dotfiles (or just dotfiles), following this format:

  • Each line contains the local repository filename, optionally followed by a colon and the target filename if different than the local filename.
  • External dotfile mappings can be included using #include “path/to/mapping”, and the requested map will be loaded as well. The requested path can either be an explicit filename, or a directory containing a mapping named dotfiles or .dotfiles.
  • Comments are denoted by lines beginning with the # symbol, and are ignored.

An example mapping might look like:

# the dotfiles to care about aliases: .alias .bashrc .profile: .profile htop: .config/htop/htoprc vim: .vim vimrc: .vimrc

# some external dotfile paths to include #include “repo2/dotfiles” #include “/full/patt/to/repo3/”

Once the mapping is in place, run dotlink, and tell it where your source repository is, as well as where you want it to deploy to:

$ dotlink [path/to/repository] [[[user@]host:]path/to/target]

The source path is optional; dotlink will assume it’s your current directory if it finds a .dotfiles mapping file unless you specify otherwise. The target path is also optional, and assumed to be your local home directory.

Sources can include local paths or remote git repository URLs. Targets can also include local paths, or remote paths accessible via ssh by providing a hostname as well as username if different than your current login.

advanced

### embedding

If you’d like to embed dotlink within your dotfile repository, dotlink/dotlink.py is a self-contained script, specifically to allow for this use case. Simply copy dotfile.py into your repository; it has no external dependencies.

### remote sources

Sources can also point at remote locations, such as git repositories, and dotlink will clone the remote data into a temporary path and then copy the contents into the appropriate paths. dotlink will guess this if your source path begins with “git://” or “git@”, but you can also use –git to force this for non-standard URLs. The following command will clone my public dotfile repo, and install it to your home directory:

$ dotlink git://github.com/jreese/dotfiles

This is essentially equivalent to the following sequence of commands:

$ git clone git://github.com/jreese/dotfiles $ dotlink dotfiles $ rm -rf dotfiles

You can even combine remote sources with remote targets, to clone the dotfile repository locally, and then copy the dotfiles to the remote host via scp:

$ dotlink git://github.com/jreese/dotfiles jreese@devserver:

todo

Some planned features and changes are:

  • Add support for remote sources, like ssh/scp, as well as git repos or tarballs
  • Generate mapping file from repository contents ?

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
Dotlink-0.6.0.tar.gz (7.3 kB) Copy SHA256 hash SHA256 Source None Dec 4, 2013

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page