Branches

A branch is a reference to a commit, and can be moved to some future commit. This page gives an overview of the tools git has to work with branches.

Navigation

This section provides an overview of the basics of dealing with branches in git:

• git branch for listing available branches.

• git branch <branch_name> for creating a new branch.

• git checkout <branch_name> for switching between branches.

• How a branch reacts to the creation of a new commit while HEAD refers to that branch.

---

A newly created repository will only contain the master branch. The next cell shows the log of the repo after the first commit.

echo "some text" &> some_files
git add --all
git commit -m "first commit" &> /dev/null
git log --decorate

commit 526242f3943431da68cfe3f4b30d882b52b48b4b (HEAD -> master)
Author: fedorkobak <kobfedsur@gmail.com>
Date:   Thu Oct 30 23:17:07 2025 +0300

    first commit

The following cell shows awailable branches by using git branch.

git branch

* master

There is only the main branch. The following code creates an extra branch, new_branch, and lists available branches.

git branch new_branch
git branch

* master
  new_branch

Now, in git log, you can see that both master and new_branch refer to the same commit. However, the syntax HEAD -> master indicates that the master branch is currently being used.

git log

commit 526242f3943431da68cfe3f4b30d882b52b48b4b (HEAD -> master, new_branch)
Author: fedorkobak <kobfedsur@gmail.com>
Date:   Thu Oct 30 23:17:07 2025 +0300

    first commit

With git checkout new_branch you can set new_branch as the branch that HEAD points to.

git checkout new_branch
git log

Switched to branch 'new_branch'
commit 526242f3943431da68cfe3f4b30d882b52b48b4b (HEAD -> new_branch, master)
Author: fedorkobak <kobfedsur@gmail.com>
Date:   Thu Oct 30 23:17:07 2025 +0300

    first commit

But what does HEAD -> new_branch actually mean? During a git commit, HEAD points to the new commit, and the branch it refers to moves along with it. The following cell creates a new commit and displays the resulting git log.

echo "hello" > new_file
git add --all
git commit -m "add new file" &> /dev/null
git log --all --graph

* commit b62d5161b54f0ecb6e5ba71a92557838b851603c (HEAD -> new_branch)
| Author: fedorkobak <kobfedsur@gmail.com>
| Date:   Thu Oct 30 23:17:19 2025 +0300
| 
|     add new file
| 
* commit 526242f3943431da68cfe3f4b30d882b52b48b4b (master)
  Author: fedorkobak <kobfedsur@gmail.com>
  Date:   Thu Oct 30 23:17:07 2025 +0300
  
      first commit

As a result, HEAD and new_branch, which HEAD referred to, have moved to the new commit.

Merging

Merging is the case when you create the commit that takes changes two branches. Generally there are two possible cases:

Case Name	Description	History
1. Fast-Forward Merge	Occurs when the target branch (the one you are merging into) has no new commits since the feature branch diverged.	Git simply moves the target branch pointer forward to the new commit. No merge commit is created, and the history remains linear.
2. Three-Way Merge (or True Merge)	Occurs when the target branch has new commits (i.e., the branches have diverged). This matches the scenario you described in your first question.	Git compares the two branch tips and their common ancestor (merge base) to combine changes. A merge commit is created, which has two parent commits, preserving the history of both branches.

The commands you may need to merge branches:

• The git merge <target_branch> merges the current branch with the selected one.

• The command git checkout --ours/theirs <file> applies changes to the <file> from the current branch or merged one in the conflict case.

For more check the Merge page.