config-keeper2 1.0.0a4

CLI tool for keeping your personal config files in a repository

config-keeper

User-friendly CLI for keeping your personal files or directories in a repository.

In a few words config-keeper does following:

• Collects information about what files on current machine it should keep
• Makes a temporary copy of all these files and pushes it to specified repository
• Pulls files from a repository later and puts them to the right places

Table of contents

• Use cases
• Key features
• Install
  • Using pip
• Usage
  • Quick start
  • CLI Reference

Use cases

1. You spend a lot of time writing launch/build tasks or making config files for a project you develop, but these stuff cannot be placed along with the project itself.
2. You have to switch between work computers from time to time while you working on same project and you have to send yourself archives with bunch of updated files.
3. You want to save some system-wide config (like .bashrc) to use it later or quickly restore if something goes wrong.

Finally, you want to automate these stuff.

Key features

• Create projects as logical groups of files to sync and a repository
• All configuration in a single YAML file - update it using CLI or by hands using validate command after
• Terminal auto-completion
• User-friendly error messages if something goes wrong

Install

Using pip

pip install --user config-keeper2

NOTE: if you are using latest versions of Ubuntu/Debian/Fedora, you may also need to use --break-system-packages flag. Refer to PEP 668 for more information.

Usage

config-keeper relies on a project as logical group of files to sync and a repository. Each project consists of

• Paths. These are any files or directories in your filesystem which you want to keep. Some of them might not exist yet, if you are want to pull them first from existing repository. But they must exist if you are going to push.
• Repository. Here your files and directories live. When you going to push or pull, this repository is used.
• Branch. The branch of repository. Nothing special. If you want, you can use the same repository across multiple projects with different branches.

Quick start

To begin, create a project:

config-keeper project create myproject
# Repository: [e-mail]:MyUser/personal-stuff.git
# Branch [main]: mybranch
# Project "myproject" saved.

Here assumed that repository exists and you have permissions to communicate with it. Branch may exist or not - if not, it will be created at first push.

Next, add path to the project:

config-keeper paths add \
    --project myproject \
    my_config.ini:~/configs/my_config.ini \
# Project "myproject" saved.

Here we said that ~/configs/my_config.ini file should live in our repository as my_config.ini.

Assuming that ~/configs/my_config.ini exists, let's push it:

config-keeper push myproject
# Going to push into following branches:
# - "mybranch" at [e-mail]:MyUser/personal-stuff.git (from "myproject")
# Proceed? [Y/n]:

If something goes wrong at this step (or at any other step) you will probably have a nice-readable error. If this is not your case please create an issue describing how to reproduce the bug and the desired behavior.

Now, if you visit repository, you should see that mybranch has recent commit with my_config.ini file.

Finally, imagine you want to restore my_config.ini on your local machine from a repository. All you need to do is

config-keeper pull myproject
# Following paths will most likely be replaced:
# - /home/<my_user>/configs/my_config.ini (from "myproject")
# Proceed? [Y/n]:

If you confirmed replacing, the ~/configs/my_config.ini file now should be the same as in repository one.

CLI Reference

To learn what commands are available, please refer to REFERENCE.

NOTE: you can always check the description of any command by using --help, e.g.

config-keeper push --help