Setting up and using version control for a project
In the last episode of Code Club we started a new project. It was a very exciting day! The project we’re going to be working on over the next Code Club episodes seeks to determine to what degree do inter- and intra-genomic variation limit the interpretation of amplicon sequence variants (ASVs). As I mentioned, this is a newish approach to analyzing 16S rRNA gene sequence data that is all the rage, but not enough people are questioning the assumptions that are baked into the method. We’re going to test the assumptions that ASVs represent a biologically coherent entities. We also got familiar with our command line interface using a unix style environment and set up a basic organization structure for our project. Along the way we’ll learn different elements of what it takes to make an analysis reproducible and how to use a variety of tools that will help out. Even if you don’t find the problem we are studying interesting, you will hopefully find the approach generalizable to a variety of problems that do interest you. For today’s episode we’ll add version control to our project and learn to use git and GitHub to track the history of our project. git is one of those things that can be very confusing, but it doesn’t need to be. We’ll take it slow and we’ll be using git in nearly every episode that follows, so we’ll get a lot of practice!
Please take the time to watch today’s episode, follow along on your own computer, and attempt the exercises. Don’t worry if you aren’t sure how to solve the exercises, at the end I will provide solutions. The reference notes and links that follow are a supplement to the material in the video.
If you’re like me, you use the undo function in Microsoft Word or Excel countless times a day. Sometimes, I need to undo a large number of things to get back to a point before I screwed things up. One problem with the undo function is that if I quit the program, the thread of changes is broken and I can’t undo something from a previous session in the program. If we use it right, version control is a lot like the undo function. The big difference is that version control won’t lose the thread of changes when we quit an application. It will also keep track of my changes across many files in a project. Even better, it works well with text files like those we’ll be using for our project. Here’s a brief list of things that version control will help us with:
- Documenting the history to a project - When, why, and how did I make a change?
- Back up system - I can delete a file or even my entire project directory and get it back
- Facilitates making your work public - I can use version control to share my code with others
- Enables collaboration - Other people can make suggestions to improve my project
To achieve these benefits, we’ll use the most popular version control tool called
git. We will be running
git on our computer to maintain what we call our local repository or “local”. Note that even if you doing a project on a computer that’s not in the same room as the computer you’re typing on, I’ll call it your local repository. We’ll also make use of GitHub to host our repository as a remote repository or “remote”. Here are the
git commands we’ll use for today’s episode and somethings you should do to get set up in GitHub.
Watching the video, you’ll see me work through a handful of
git commands. About 95% of what I do with
git uses 5 commands. So if you can master those, you’ll be in good shape. If you’re on a Windows computer, be sure you have the the Ubuntu Linux BASH shell for Windows 10 installed. If you’re using Mac OS X, we’ll be using the Terminal app so much that you probably want to keep it in your Dock. Here are the commands we’ll use.
git config- set up your local git environment (typically copy and paste commands from elsewhere). Be sure to customize values in quotes for you
git config --global user.name "Your Name" git config --global user.email "email@example.com" git config --global github.user "your_github_account_name" git config --global core.autocrlf input #for mac os x git config --global core.autocrlf true #for windows git config --global core.editor "nano -w" git config --global --list
git init- create a git repository (look for
.git/when you use
git status- what files are not being tracked? which have been changed? which have been staged?
git add- stage files that are not being tracked and stage changes for committing
git commit- commit changes to your repository
git pull- get a copy of your remote repository (i.e. GitHub) and merge it with your local repository
git push- send your local repository to the remote (i.e. GitHub)
GitHub is the most popular website for hosting
git repositories and we will be using it to host our project’s repository. If you don’t have an account already, go ahead and set up a GitHub account. You will get unlimited public repositories for free. If you want to keep your repositories private, you will need to create a paid subscription. However, if you are an educator or student, you can apply for an academic discount. This will give you many benefits including unlimited private repositories. If you would prefer to keep your repository private while you are developing your ideas, then I would hope that once you are done with your project you will be willing to convert your private repository into a public repository.
You can connect to GitHub from the command line interface using https or SSH. In the video we’ll use https, which requires us to enter a password. We’ll use a setting that only requires us to enter our password once an hour. You can learn how to use your computer’s password manager. Alternatively, you can use a SSH key to connect to your repositories. I find this to be the most convenient, but it is a bit more involved to set up. GitHub has some great instructions to get you going.
1. As an exercise in the last episode we added
README.md files to each of our directories. Commit these files to your repository and push the changes to your GitHub account.
2. I would like to use a MIT license for this project because we will have a fair amount of code, it is a simple license, and it will allow anyone to use the code and develop it further as long as they provide attribution to us by including the license. You can find the text of the license at https://opensource.org/licenses/MIT. Copy the text of the license into your
LICENSE.md file and make no changes to the file. Save the text and commit the change to your repository.
3. Edit the
LICENSE.md file to indicate the year and your name as the copyright holder (if we’re doing this in parallel, I’m not sure who the true copyright holder is!). Go ahead and commit the change to your repository. Push the changes to your GitHub account.