mass-git-scripts

This repo contains a few scripts for managing large numbers of git repositories, either all those belonging to a user or organization on GitHub, or simply all repositories in a direct:ory.

The inspiration for this was the need to back up all of my personal repositories (over 100) with full history and all branches. I thought it would make a good side project to write a script to do this.

A few programs are required to run these scripts:

• jq: A lightweight and flexible command-line JSON processor
• parallel: A shell tool for executing jobs in parallel using one or more computers
• gh: GitHub CLI

You should be able to install them with your local package manager, for example on ubuntu:

sudo apt install jq parallel gh

To use, clone this repository, and run sudo make install to install the scripts in /usr/local/bin.

Then, for example, if you have your personal repos in ~/.src/github/username, and want to update all of them, you can run:

git-pull-all ~/.src/github/username

If you have reset your environment and want to clone all of your repos, you can run:

git-clone-all --owner user --dir ~/.src/github/user --limit 1000 --jobs 20

The examples below assume you have installed this repository, and the scripts within are available in your path.

Repositories that are already present in the target folder are skipped. The command output will report the number of skipped and cloned repositories.

The git-clone-all script clones all repositories belonging to a user or organization on GitHub, specified via the --owner flag.

The target directory in which the repositories will be cloned must be specified and exist. Pass it via the --dir argument.

You can set the clone depth with --depth; by default, the entire history is cloned.

To limit the number of repositories cloned, pass the --limit flag. The default --limit flag value is 25. If you wish to clone all repositories, consider setting the limit to a very high value, like --limit 1000.

The script clones repositories in parallel, with the number of concurrent jobs specified with the --jobs flag. The default number of jobs is 8.

Some examples:

git-clone-all --owner f3rno64 --dir ./f3rno64 --limit 200 --jobs 20
git-clone-all --owner google --dir ./google --limit 10000 --jobs 50 --depth 1
git-clone-all --owner microsoft --dir ./microsoft

The command output will report the number of new commits pulled if there are any.

When you have many repositories in a single directory, and they are frequently updated, git-pull-all can be used to update the active branch for each repo with the latest changes from the active remote.

This is very useful if you keep all repositories for an organization (such as your employer) in a single directory.

To use, just pass the directory containing the repositories as the first argument to the script:

git-pull-all ~/.src/github/f3rno64

To contribute, clone this repo and hack away! There is only one useful script in the manifest, lint, which uses shellcheck and markdownlint to lint the scripts and README.

Run it to ensure your changes are up to standard. It is also run as a pre-commit hook.

Make sure to install dependencies for development with pnpm i.

mass-git-scripts is a set of useful tools for anyone working with many git repositories organized in separate folders by their owner (user or organization).

Give it a try and see how it can improve your workflow.

Distributed under the MIT license. See LICENSE.md for more information.

1. Fork it
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create a new Pull Request