How do I clone a subdirectory only of a Git repository?

In Git, it is common to clone the entire repository. However, if you only need a specific subdirectory within the repository, Git offers a feature called sparse-checkout to achieve this.

Git Sparse Checkout Steps

1. Initialize the Repository: First, clone the repository without immediately checking out all files.

bash
git clone --no-checkout <repository-url>
cd <repository-name>

This gives you the full history of the repository without any files being checked out.

1. Enable Sparse Checkout:

bash
git config core.sparseCheckout true

This command activates the sparse-checkout feature in the repository.

1. Set the Subdirectories to Check Out:

Create a file named .git/info/sparse-checkout and specify the paths you want to check out within it.

bash
echo "subdirectory/" > .git/info/sparse-checkout

If you want to check out multiple subdirectories, add multiple lines. For example:

bash
echo "subdirectory1/" >> .git/info/sparse-checkout
echo "subdirectory2/" >> .git/info/sparse-checkout

1. Checkout the Required Files:

Now, you can check out the required subdirectory using the following command:

bash
git checkout <branch-name>

This will cause Git to check out the files based on the paths specified in the sparse-checkout file.

Example Scenario

Suppose you are working on a large project that includes a directory named docs, which you only need for writing documentation. You can use the sparse-checkout feature described above to clone and check out only this directory.

Note

Sparse-checkout is very useful when dealing with large repositories because it significantly reduces the number of local files, improving processing speed and saving disk space. However, it has certain limitations, such as requiring manual specification of each subdirectory you need, which can be somewhat tedious in some cases.

This is a detailed guide on how to use Git's sparse-checkout feature to clone only specific subdirectories of a repository.