What Are Branches in Git?

In the previous chapters, we saw how Git records your project's history as a series of snapshots, or commits, typically forming a linear sequence. Each commit builds upon the one before it. However, software development rarely proceeds in such a strictly linear fashion. You often need to work on different things simultaneously. Imagine you're developing a major new feature, but suddenly an urgent bug needs fixing in the code already released to users. You need a way to pause your feature work, fix the bug on the stable codebase, and then return to your feature development, without the two lines of work interfering with each other.

This is where Git branches come into play. Think of a branch not as a complete copy of your project files, but as a lightweight movable pointer to a specific commit. When you start working, Git automatically creates a default branch, typically named main (or sometimes master in older projects). This main branch represents the primary line of development, often tracking the stable or production-ready code.

Every time you make a commit, this branch pointer automatically moves forward to the newest commit you just created. So, a simple history might look like this: the main branch pointer points to the latest commit, which itself points back to its parent commit, and so on, forming the project's history.

A simple Git history. Commits (C1, C2, C3) form a sequence. The main branch points to the latest commit (C3), and HEAD indicates that main is the currently active branch.

Creating a new branch means creating a new pointer that starts at the same commit your current branch points to. Let's say you create a new branch called new-feature while you are on the main branch which points to commit C3. Initially, both main and new-feature will point to C3.

After creating the new-feature branch. Both main and new-feature pointers reference the same commit, C3. HEAD still points to main.

Now, if you switch to the new-feature branch (we'll cover how to do this shortly) and make a new commit, say C4, something interesting happens. The new-feature pointer moves forward to point to C4, but the main branch pointer stays exactly where it was, still pointing at C3. Your HEAD pointer now indicates you're working on the new-feature branch.

After switching to new-feature and making commit C4. The new-feature branch pointer advances to C4, while main remains at C3. HEAD now points to new-feature.

This creates two independent lines of development diverging from commit C3. You can continue working on your feature on the new-feature branch without affecting the main branch. If needed, you could switch back to main, create another branch (e.g., bug-fix), make commits there, and it wouldn't affect main or new-feature until you explicitly decide to combine them (which is called merging, covered later).

The main takeaway is that Git branches are incredibly lightweight. Creating a new branch doesn't involve copying files or directories; it just creates a small file containing the 40-character SHA-1 hash of the commit it points to. This efficiency makes branching a frequent and fundamental part of the Git workflow, unlike in some older version control systems where branching was a more "expensive" operation.

Using branches provides several significant advantages:

1. Isolation: Work on new features or bug fixes without disrupting the stable codebase (main).
2. Experimentation: Try out new ideas or refactor code on a separate branch. If it doesn't work out, you can simply discard the branch without affecting the main project.
3. Parallel Development: Multiple developers (or even one developer working on multiple tasks) can work on different features or fixes concurrently.
4. Clean History: Keep the main branch clean, containing only stable, tested code, while development happens on other branches.

Understanding this pointer concept is fundamental to effectively using Git's branching capabilities, which we will explore further in the upcoming sections.