Creating a Repository

Estimated time: 10 minutes

Overview

Questions

• Where does Git store information?

Objectives

• Create a local Git repository.
• Describe the purpose of the .git directory.

Once Git is configured, we can start using it.

We will help Sarah with her new project, create a repository with all her recipes.

First, let’s create a new directory in the Desktop folder for our work and then change the current working directory to the newly created one:

BASH

$ cd ~/Desktop
$ mkdir recipes
$ cd recipes

Then we tell Git to make recipes a repository -- a place where Git can store versions of our files:

BASH

$ git init

It is important to note that git init will create a repository that can include subdirectories and their files—there is no need to create separate repositories nested within the recipes repository, whether subdirectories are present from the beginning or added later. Also, note that the creation of the recipes directory and its initialization as a repository are completely separate processes.

If we use ls to show the directory’s contents, it appears that nothing has changed:

BASH

$ ls

But if we add the -a flag to show everything, we can see that Git has created a hidden directory within recipes called .git:

BASH

$ ls -a

OUTPUT

.	..	.git

Git uses this special subdirectory to store all the information about the project, including the tracked files and sub-directories located within the project’s directory. If we ever delete the .git subdirectory, we will lose the project’s history.

Next, we will change the default branch to be called main. This might be the default branch depending on your settings and version of git. See the setup episode for more information on this change.

BASH

$ git checkout -b main

OUTPUT

Switched to a new branch 'main'

We can check that everything is set up correctly by asking Git to tell us the status of our project:

BASH

$ git status

OUTPUT

On branch main

No commits yet

nothing to commit (create/copy files and use "git add" to track)

If you are using a different version of git, the exact wording of the output might be slightly different.

Places to Create Git Repositories

Along with tracking information about recipes (the project we have already created), Sarah would also like to track information about cocktails. Despite Kai’s concerns, Sarah creates a cocktails project inside the recipes project with the following sequence of commands:

BASH

$ cd ~/Desktop    # return to Desktop directory
$ cd recipes      # go into recipes directory, which is already a Git repository
$ ls -a           # ensure the .git subdirectory is still present in the recipes directory
$ mkdir cocktails # make a sub-directory recipes/cocktails
$ cd cocktails    # go into cocktails subdirectory
$ git init        # make the cocktails subdirectory a Git repository
$ ls -a           # ensure the .git subdirectory is present indicating we have created a new Git repository

Is the git init command, run inside the cocktails subdirectory, required for tracking files stored in the cocktails subdirectory?

Show me the solution

No. Sarah does not need to make the cocktails subdirectory a Git repository because the recipes repository will track all files, sub-directories, and subdirectory files under the recipes directory. Thus, in order to track all information about cocktails, Sarah only needed to add the cocktails subdirectory to the recipes directory.

Additionally, Git repositories can interfere with each other if they are “nested”: the outer repository will try to version-control the inner repository. Therefore, it’s best to create each new Git repository in a separate directory. To be sure that there is no conflicting repository in the directory, check the output of git status. If it looks like the following, you are good to go to create a new repository as shown above:

BASH

$ git status

OUTPUT

fatal: Not a git repository (or any of the parent directories): .git

Correcting git init Mistakes

Kai explains to Sarah how a nested repository is redundant and may cause confusion down the road. Sarah would like to remove the nested repository. How can Sarah undo her last git init in the cocktails subdirectory?

Solution – USE WITH CAUTION!

Background

Removing files from a Git repository needs to be done with caution. But we have not learned yet how to tell Git to track a particular file; we will learn this in the next episode. Files that are not tracked by Git can easily be removed like any other “ordinary” files with

BASH

$ rm filename

Similarly a directory can be removed using rm -r dirname or rm -rf dirname. If the files or folder being removed in this fashion are tracked by Git, then their removal becomes another change that we will need to track, as we will see in the next episode.

Solution

Git keeps all of its files in the .git directory. To recover from this little mistake, Sarah can just remove the .git folder in the cocktails subdirectory by running the following command from inside the recipes directory:

BASH

$ rm -rf cocktails/.git

But be careful! Running this command in the wrong directory will remove the entire Git history of a project you might want to keep. Therefore, always check your current directory using the command pwd.

Key Points

• git init initializes a repository.
• Git stores all of its repository data in the .git directory.