The command line
and git


SURE 2025

Department of Statistics & Data Science
Carnegie Mellon University

The command line

Why command line?

  • It’s simply a text-based way to communicate with your computer

  • Offers direct control over your computer

  • Enables complex tasks and automation that are not easily achievable through graphical interfaces

Basic commands

  • pwd: print working directory

  • ls: list files/folders (in your current directory)

  • cd: change directory (e.g. cd .. goes up one level)

  • mkdir: make directory (i.e. create a new folder)

  • touch: create an empty file (often used for quick testing)

  • cp: copy a file/folder to a specific location

  • mv: move (cut) a file/folder

  • rm: remove a file/folder

We’ll do it live!

Version control

About

  • Version control records changes to files over time (snapshots)

  • Why? So we can recall specific versions later

  • Rather than doing this…

The struggle

Why version control?

  • Collaboration

  • Reverting

  • Tracking changes/history

Git

  • Git is a distributed version control system

  • This means that everyone working on a project has a complete copy of the project’s history on their own computer

  • Distributed nature of Git: if one’s computer crashes, the project history is safe because everyone else has a copy

Git vs GitHub

Git and GitHub are NOT the same (just like R and RStudio…)

  • Git: the tool itself, does the actual work

  • GitHub: website that hosts Git repositories, built around Git (think of it as a place to store and share your Git projects)

  • Alternatives to GitHub: GitLab, Bitbucket (but GitHub is by far the most popular)

Basic Git workflow

Initializing a repository

We can create or turn an existing directory/folder into a Git repository

To turn a folder into a git repository, open your terminal, navigate to the desired folder (using cd) and then type:

git init

We can clone a git repository (using a web url)

git clone <repo-url>

Staging changes

  • Before saving a version/commit, you need to tell Git which changes you want to include. This is called staging.
git add <file>  # e.g., git add new_file.txt

  • If we replace <file> by ., then everything that was changed will be staged

  • To see what’s been staged

git status

Committing changes and viewing history

  • A commit saves your staged changes
git commit -m "Insert commit message"

  • Commit messages should be meaningful and descriptive

  • To see the history of commits (messages, author, date)

git log

The .gitignore file

  • .gitignore is a text file that tells Git which files/folders to ignore

  • Prevent them from being accidentally added to your repository and keeps your commit history clean

  • Ignore unnecessary files (e.g. large files, derivative files, system-specific files) that clutter your repository

  • Example .gitignore

**/*.DS_Store
**/*.pdf
data/

push, pull, and branch

  • push: upload your changes to a remote repository (like on GitHub)
git push
  • pull: dowload changes from a remote repository to your local computer
git pull

  • git pull before you start editing a file to avoid conflicts

  • Branches: see here

Resources

We’ll do it live!