Git Course

Getting started

Installing git

Linux

If you have admin user rights in your computer, use sudo apt-get install git. Otherwise download and run script.sh from this repository.

Mac

Download and install from Git website.

Configuring your profile

First check that git is properly installed by typing git in your terminal. Then set your name and e-mail address:

$ git config --global user.name "Homer Simpson"
$ git config --global user.email "homer@springfieldpower.com"

I also recommend adding an alias that makes git log much prettier:

$ git config alias.hist "log --pretty=format:\"%h %ad | [%an] %s%d\" --graph --date=short"

And then use git hist in place of git log.

First repo

Init the repository

Make a new directory for your project and then let's git init the repository!

$ mkdir project
$ cd project
$ git init
Initialized empty Git repository in /home/user/project/.git/

You'll get a confirmation that an empty repository has been created.

Adding files to the project

To add a file to your project, you first need to add them in the stage. The stage is a temporary space where files are added before commiting them. You can freely add and remove files to the stage before finally commiting them to the repository. Once the staged files are commited, their current versions remain in your repository and you'll be able to recover them in the future, but the commit can no longer be modified. Let's start by creating an empty file.

$ touch file.txt

Now check the status of the repository with

$ git status
On branch master

Initial commit

nothing to commit (create/copy files and use "git add" to track)

You will see that file.txt appears in the list of Untracked files. This means that file.txt does not belong to the repository and git is not tracking the changes made on this file. To add it to the project, we first need to add them to the staging area using git add.

$ git add file.txt
$ git status
On branch master

Initial commit

Changes to be committed:
  (use "git rm --cached <file>..." to unstage)

	new file:   file.txt

Voila! See that now file.txt is in the Changes to be commited? That means that we are only one step away from permanently storing file.txt in its current (empty) version! Let's make a commit to add file.txt to the project:

$ git commit -m "Add file.txt"
[master (root-commit) 7995707] Add file.txt.
 1 file changed, 0 insertions(+), 0 deletions(-)
 create mode 100644 file.txt

Let us briefly explain the commit command, it's actually pretty simple. Once you have the desired files in the staging area (remember that we add them using git add and check what is in the staging area with git status) use git commit -m "message", where message should be a short description of the updates performed in this commit.

Now we have added file.txt to the repository and git will track the changes made to this file. If you immediately type git status git will report nothing to commit, working directory clean. This means that all the tracked files are in their latest versions. Let's now update file.txt and see how git reacts...

$ echo "Hello!" > file.txt
$ git status
On branch master
Changes not staged for commit:
  (use "git add <file>..." to update what will be committed)
  (use "git checkout -- <file>..." to discard changes in working directory)

	modified:   file.txt

no changes added to commit (use "git add" and/or "git commit -a")

Note that file.txt is now in the list of file that have changes not staged for commit, that means that the file differs from its latest version (the last commit). We use git diff to see which changes have been made to any file since the last commit:

$ git diff file.txt
diff --git a/file.txt b/file.txt
index e69de29..10ddd6d 100644
--- a/file.txt
+++ b/file.txt
@@ -0,0 +1 @@
+Hello!

Version history

Now that our project has one commit, we can check the commit history with git log:

$ git log
commit 7995707d1498fe32541818d3f4e19c580472d893
Author: Homer Simpson <homer@springfieldpower.com>
Date:   Wed Aug 10 17:55:53 2016 +0200

    Add file.txt.

Indeed, our project has only one commit, the one in which we added file.txt. Let's now add another files and save the previous changes to file.txt made in the previous section (remember we echo'ed "Hello!" to file.txt). So first, we will create the file index.html and then add the updated file.txt together with index.html.

$ touch index.html
$ git add index.html file.txt
$ git status
On branch master
Changes to be committed:
  (use "git reset HEAD <file>..." to unstage)

	modified:   file.txt
	new file:   index.html

Now we commit the changes:

$ git commit -m "Add empty index.html. Update file.txt."
[master 5f3852c] Add empty index.html. Update file.txt
 2 files changed, 1 insertion(+)
 create mode 100644 index.html

So now our history should have two commits, right? Let's check:

$ git log
commit 5f3852cfe844488f101ab9cb15690aadbdb1a642
Author: Homer Simpson <homer@springfieldpower.com>
Date:   Wed Aug 10 19:07:06 2016 +0200

    Add empty index.html. Update file.txt

commit 7995707d1498fe32541818d3f4e19c580472d893
Author: Homer Simpson <homer@springfieldpower.com>
Date:   Wed Aug 10 17:55:53 2016 +0200

    Add file.txt.

Note that each commit has a different hash. The hash is the bunch of numbers and letters that appear after the word commit, i.e. 5f3852cfe844488f101ab9cb15690aadbdb1a642 and 7995707d1498fe32541818d3f4e19c580472d893. This is the only name that identifies the commit, but we don't need to use the whole hash, using the first 7 characters is enough. Also, imagine that we have a project with thousands of commits... The output of git log would be very long and quite unreadable! So, let's try the alias git hist we set at the beginning of this tutorial:

$ git hist
* 5f3852c 2016-08-10 | [Homer Simpson] Add empty index.html. Update file.txt (HEAD, master)
* 7995707 2016-08-10 | [Homer Simpson] Add file.txt.

Aha! This is much more compact and readable. Note that the first commit has (HEAD, master)?

ezorita / gitcourse