Git is a way of keeping track of differences in your "thing".
Git can be used to track differences through time in any kind of file. However, it works best for text files.
What have I used git to track:
- Code: All my code is in git, and it makes life so much easier.
- Analysis scripts: Normally the first thing I do to a dataset is
git init. Then, I keep track of what I run using git, adding only the scripts. - Presentations: I'll normally use LaTeX & Brewer to write my presentations, and git keeps me from making stupid mistakes. GitHub is a good way to share them, too.
- Thesis: I wrote my thesis in latex, tracking the changes with git. I even got comments as git pull requests from Justin...
- Papers: Ditto to the last two points
- Documentation: Most code that has documentation keeps it in the same git repo as the actual code.
So, there are many git commands.
mkdir git_repo
cd git_repo
git init # makes a repo
Now you have an actual repo, you better do some actual work. To "commit" or
save it, use git commit, which is the time resolution of git. You make a
commit when you're happy with whatever you're adding, or at least when it is
finished for the day/week
git add . # adds everything in the current folder to the git repo.
git commit # makes the commit
You can use git add --patch to modify the commit on the fly, which is
REALLY helpful!