Remote and Fetch – Software Engineering

Remote git

While you can totally use a git repository for your own personal use, git is really built for collaboration with others.

So here is a shocker

A remote git repository does not actually have to be remote. A bit misleading, yes. A remote git repository is just a copy of the local repository somewhere else.

Exercise: Create a new repo `remote-git` and initialize it

Create a new directory remote-git and navigate into it. Do not do this inside an existing git repository.

Initialize the repository.

Solution: Create a new repo `remote-git` and initialize it

mkdir remote-git
cd remote-git
git init

This is distributed version control

So, a remote is just another git repository that is of the same project and has changes we may need or may need to share our changes with.

To manage remotes, we have the git remote command. Read more with man git-remote.

To add a remote, we use the git remote add <name> <uri> command.

The URI can be a path to a directory or a URL, like SSH or HTTPS.

Exercise: Add `learn-git` as a remote to the `remote-git` repository

Add a remote called learn-git to the remote-git repository.

The URI for the remote is <path-to-your-learn-git-directory>.

Use origin as the name of the remote. This is good practice for the authority repo.

Verify that the remote has been added.

Solution: Add `learn-git` as a remote to the `remote-git` repository

Here <path-to-your-learn-git-directory> is the path to the learn-git directory on your system. We use ../learn-git as path in this example.

git remote add origin ../learn-git

We can check the remotes by using the git remote -v command.

git remote -v

Output:

origin  ../learn-git (fetch)
origin  ../learn-git (push)

Good practices with remotes

Remote is a project repository

There is a single source of truth, which is the remote project repository. This should be named origin.

Remote is a fork

In this case the fork remote repository should be named origin and the project repository should be named upstream.

Now we will need to merge our changes from learn-git into remote-git.

Fetch

All state can be fetched from a remote repository. We use the git fetch command for this.

This operation does not affect currently checked out branches—just the history about them that is in origin/*.

(you should do that now)

git fetch

The output will look something like this. We got all the information about branches from learn-git.

Output:

remote: Enumerating objects: 28, done.
remote: Counting objects: 100% (28/28), done.
remote: Compressing objects: 100% (17/17), done.
remote: Total 28 (delta 5), reused 0 (delta 0), pack-reused 0 (from 0)
Unpacking objects: 100% (28/28), 2.12 KiB | 57.00 KiB/s, done.
From ../learn-git
 * [new branch]      feature             -> origin/feature
 * [new branch]      feature-delete-test -> origin/feature-delete-test
 * [new branch]      feature-rebase-main -> origin/feature-rebase-main
 * [new branch]      feature-two         -> origin/feature-two
 * [new branch]      main                -> origin/main

So now we are up to date, right?

git log

(try it now)

We can use git log origin/main to see the log of the main branch in the remote repository.

We can also use git branch -a to see all the branches, including the remote branches.

Question: Can you think of a way to get the changes from the remote?

We can do a merge.

git merge origin/main

But there is also git pull, which we will cover next.

← Previous Next →