Lesson Plan

Before we start

• Have you installed (i) R, (ii) RStudio, and of course (iii) git on your machine?
• Have you installed a bash compatible shell? (Windows only)
  • Git for windows Comes pre-installed with Git, often providing all the necessary functionality for the course as the baseline requirement.
  • Rtools shell comes pre-installed with Rtools
  • Windows Subsystem for Linux (WSL) is an Ubuntu bash shell for Windows. Comes pre-installed with htop, apt-get, and make.
• Are they all updated to their most recent version?
• Have you already created an account on GitHub and have your password ready?
  • Does your account have a profile picture? Your full name? 👀
• Make sure that some further software is installed
  • Windows
    • Install Rtools
    • Install Chocolatey
  • Mac
    • Install Homebrew
    • Configure C++ toolchain
  • Linux
    • None :)

Introductory Resources

Note

Do not worry, not all of the resources need to be read. The resources have a large overlap, but some people might like one resource more than another (we certainly did observe this among us). Important is, that you get a first overview of what git, GitHub, and the shell are.

• Happy Git with R is a beginner-friendly website that introduces git with R and RStudio
  • In addition, we recommend appendix A which links R, git, and the shell
• GitHub provides a nice CheatSheet
• GitLab also provides a nice CheatSheet
• TryBash is an interactive course and cheatsheet for basic bash commands
• Git Gud is a Python-based gamified introduction to git
• There's also this playlist on YouTube
• Code and Data for Social Science is a great text that can guide readers into important aspects of coding and data for social sciences
  • Some sections may be more advanced, but the chapter on Version Control will be especially useful for us
• The Holy Grail, i.e. Grant McDermott's Lecture Notes has everything you need and more on Git, the shell, and anything else your mind might think of.
• My git is now irreversibly messed up
• Additional Reading
  • Shell for Novices
  • Data Science at the Command Line

Shell

Note

A CheatSheet for bash can be found in the bash-cheatsheet folder.

• opening the shell on Mac, Linux, Windows
  • Use bash or zsh preferably
  • Windows may prefer Windows Subsystem for Linux (WSL)
    • Is this better than using the shells from git or Rtools? Yes - for certain applications. WSL supports e.g. the native Linux package managers and provides an overall more seamless portal to Linux.
    • WSL is like having your own Ubuntu terminal in RStudio without having to install Linux
    • However, for working with git and performing simple operations, the shells from git or Rtools are plenty sufficient
• Introductory definitions
  • What do all the weird symbols mean? (~ or $ or *)
• current working directory
• navigating directories
• creating folders
• deleting folders and files
• moving folders and files
• htop and process viewers
• installing software using brew (Mac), apt (Ubuntu), Chocolatey/Winget (Windows)

Git

• What is a git folder?
• how to initialise a new git folder?
• gitignore
• making changes
• staging / unstaging changes
• committing
• branches/forks
• going back in history (various ways)
• keeping an overview with gitk

GitHub

• logging in
• making a new repository
• private / public repos
• becoming a contributer (add them to this repo so that we can keep having all course notes in this repository)
• adding remote
• SSH
• adding a README
  • Markdown basics
• pushing / pulling
• pull requests
• Github Issues
• merge conflicts
• continuous integration (??) --> standard when developing R and Julia packages but more complicated

Git, Github, and RProjects

• How to connect Git/Github with RStudio Projects
• Cloning an existing GitHub repository
• Uploading an existing RProject to GitHub