Skip to main content

A comprehensive solution to managing your dotfiles

Project description

dotgit

tests coverage publish-pypi downloads

A comprehensive and versatile dotfiles manager

dotgit will allows you to easily store all your dotfiles for any number of machines in a single git repository. Written in python with no external dependencies besides git, it works on both Linux and MacOS (should also work on other *nix environments)

Project goals

  • Share files between machines or keep separate versions, all in the same repo
  • Make use of an intuitive filelist
  • Grouping of files to make organization easy
  • Make git source-control convenient and easy to use

Why use dotgit?

  • You can very easily organize and categorize your dotfiles, making it easy to store different setups in the same repo (e.g. your workstation and your headless server dotfiles, both in the same repo)
  • dotgit was designed with its most important goal being to make managing multiple machine's dotfiles easy, by allowing to easily share and separate dotfiles between machines
  • Ease-of-use is baked into everything without hindering more advanced users. For instance, dotgit can automatically commit and push commits for you should you want it to, but you can just as easily make the commits yourself
  • dotgit has an automated test suite that tests its functionality with several versions of Python on Linux and MacOS to ensure cross-platform compatibility
  • Support for both symlinking or copying dotfiles to your home directory. Copying allows you to quickly bootstrap a machine without leaving your repo or dotgit on it
  • No external dependencies apart from git allowing you to install and use dotgit easily in any environment that supports Python

Usage example

An example filelist might look something like this:

laptop=tools,x
desktop=tools,x

.vimrc:tools
.vimrc:pi

.xinitrc:x

.bashrc

.foo:server

Firstly, there will be two .vimrc files. The first one will be shared between the hosts desktop and laptop. There will also be a separate .vimrc inside the dotgit repository that will only be used with the pi host.

The second thing to notice is that you can use categories to group dotfiles. In the example there is a tools and x category. This makes working with a group of dotfiles a breeze.

Since no host was specified with .bashrc it will reside inside the common category. This means that it will be shared among all hosts using this dotgit repository (unless a category is specifically used along with the dotgit commands).

Lastly the .foo will only be used when you explicitly use the category server. This way you can for instance keep your "server" dotfiles in your main dotfiles repo without cluttering your normal dotgit workflow.

You can also have a look at my personal dotfiles which I manage with dotgit here.

Installation

Arch Linux users can install the AUR package.

If you are not on Arch, the easiest way to install dotgit is using pip (substitute pip with pip3 if you're on a Debian-like system):

pip install dotgit

You can also get bash-completion to work by installing dotgit's bash-completion in your home folder:

curl https://raw.githubusercontent.com/kobus-v-schoor/dotgit/master/pkg/completion/bash.sh >> ~/.bash_completion

Getting started

  1. Choose a folder where you want to store your dotfiles, ~/.dotfiles is a good place to start
  2. Create your dotfiles folder and cd to it mkdir -p ~/.dotfiles; cd ~/.dotfiles
  3. Initialize your dotgit repo with dotgit init. You can also skip the first two steps and clone an empty repo that you created somewhere e.g. on Github and run the init step inside the cloned repo.
  4. Add your first dotfile echo .bashrc >> filelist
  5. Update your dotgit repo (you need to do this whenever you change the filelist) with dotgit update -v. Once you do this your dotfile has been moved from your home directory to your dotfiles repo, and a symlink was created in your home directory in place of the original file pointing to the file inside your dotfiles repo.
  6. Commit your changes to your dotgit repo by running dotgit commit

That's it! When you want to push your changes to your git remote you just do a git push as usual (dotgit will also offer to do this for you).

If you change your dotfiles (e.g. you edit your .bashrc file) you don't need to run an update operation again. You just cd into your dotfiles directory and run dotgit commit. dotgit will generate a commit for you with a fitting description and will offer to push your changes to your remote (should you have one).

To restore your dotfiles on another machine, just clone your repo, install dotgit and run dotgit restore inside your repo.

It is strongly recommended to look through dotgit's help by running dotgit --help. This will show you good ways to use dotgit and explain dotgit's options in detail.

Future goals

dotgit was written with a plugin architecture. It currently only has one plugin, namely the "plain" plugin, which just does symlinking. The following plugins are planned for future releases:

  • Encryption using GnuPG
  • Templating

Migrating from v1.x

After many years dotgit was finally completely rewritten in python. The first version was written in pure bash, and while this was appealing at first it quickly became a nightmare from a maintenance point-of-view. The new python rewrite comes with many advantages including:

  • Much better cross-platform compatibility, especially for MacOS. Using utilities like find became problematic between different environments
  • A fully automated test suite to test dotgit on both Linux and MacOS
  • Code that the author can understand after not seeing it for a week
  • Unified install method (pip) for all the platforms

Currently, two features are missing from the python rewrite:

  • Encryption support: this will be added in a future release
  • Directory support: after much consideration it was decided to rather to not re-implement this. It requires a lot of special treatment that breaks some of the logic that works very well for single files. Excluding it made the file-handling logic much more robust

Should you decide you'd like to stick to the old version of dotgit, you are welcome to do so. Installing the pip package will also make the original dotgit available as the command "dotgit.sh" (AUR package as well). Please note that I will not be able to support the old version anymore, and as such you're on your own if you decide to use the old version.

To make room for future improvements, the layout of the dotgit dotfiles repos had to change. Unfortunately this means that the new repos are not directly compatible with the old ones, although it is easy to migrate to the new version's format. To do so, do the following:

  • Firstly, backup your current dotfiles repo in case something goes wrong
  • Next, inside your dotfiles repo, move the dotfiles folder to its new location by running mv dotfiles tmp; mkdir dotfiles; mv tmp dotfiles/plain.
  • You can leave your filelist as-is, the filelist syntax hasn't changed. You will need to delete your "cryptlist" file as this signals to dotgit that this is an old repo. Once the new version supports encryption it will not make use of a separate "cryptlist" file anyway, so there is no reason to keep it. Note that the encrypted files in your repo will be deleted once you run the new dotgit since it won't be able to find them in the filelist.
  • With the new version of dotgit, run dotgit update -v. This will update the repo if necessary and will also fix the symlinks in your home folder.
  • Commit the changes to your repo using either git or dotgit commit
  • Familiarize yourself with the new dotgit syntax which has changed slightly to better follow conventions commonly found on the command-line by reading through the help using dotgit -h

Project details


Download files

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

Source Distribution

dotgit-2.1.0.tar.gz (27.4 kB view hashes)

Uploaded Source

Built Distribution

dotgit-2.1.0-py3-none-any.whl (32.9 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page