Branch

In Git, a branch is a sequence of commits that represents the evolution of the project (or a specific feature) over time. There's always at least one branch in a repository, usually called the main branch. You can create new branches to work on new features or bug fixes and merge them back into the main branch when they're ready. Branches let you work on multiple features at the same time without interfering with each other.

Branches are lightweight and easy to create and manage in Git. They allow you to experiment with new ideas, work on different features in parallel, and isolate changes from the main branch until they're ready to be integrated. Branches are a powerful tool for organizing and managing your project's development workflow.

When you create a new branch, it starts from the latest commit on the current branch.

Remote tracking branches

When your repository is connected to a remote repository (e.g. when you cloned it from GitHub), there will be a special read-only kind of branches created for you whenever you fetch changes from the remote: remote tracking branches.

For example, say your repo is connected to a remote repository named origin. On that remote repo, there's a branch called cake. When you run git fetch origin, Git will download the cake branch to your local repository and name it origin/cake. Then, you can create a local branch based on it, like git switch -c cake origin/cake, or you can merge it into your current branch with git merge origin/cake.

Examples

git switch -c new-feature

git switch main

git merge other-feature

git branch -d other-feature

# Currently, on a branch named `feature`:
git branch --set-upstream-to=origin/feature
# Shorter version:
git branch -u origin/feature

# Now, you can run `git pull` without specifying
# the remote branch. This will fetch changes
# from `origin/feature` and merge them into
# the `feature` branch:
git pull