Automated Version Control

Overview

Teaching: 5 min
Exercises: 0 min

Questions

• What is version control and why should I use it?

Objectives

• Understand the benefits of an automated version control system.

• Understand the basics of how automated version control systems work.

We’ll start by exploring how version control can be used to keep track of what one person did and when. Even if you aren’t collaborating with other people, automated version control is much better than this situation:

“Piled Higher and Deeper” by Jorge Cham, http://www.phdcomics.com

We’ve all been in this situation before: it seems unnecessary to have multiple nearly-identical versions of the same document. Some word processors let us deal with this a little better, such as Microsoft Word’s Track Changes, Google Docs’ version history, or LibreOffice’s Recording and Displaying Changes.

Version control systems start with a base version of the document and then record changes you make each step of the way. You can think of it as a recording of your progress: you can rewind to start at the base document and play back each change you made, eventually arriving at your more recent version.

Once you think of changes as separate from the document itself, you can then think about “playing back” different sets of changes on the base document, ultimately resulting in different versions of that document. For example, two users can make independent sets of changes on the same document.

Unless multiple users make changes to the same section of the document - a conflict - you can incorporate two sets of changes into the same base document.

A version control system is a tool that keeps track of these changes for us, effectively creating different versions of our files. It allows us to decide which changes will be made to the next version (each record of these changes is called a commit), and keeps useful metadata about them. The complete history of commits for a particular project and their metadata make up a repository. Repositories can be kept in sync across different computers, facilitating collaboration among different people.

The Long History of Version Control Systems

Automated version control systems are nothing new. Tools like RCS, CVS, or Subversion have been around since the early 1980s and are used by many large companies. However, many of these are now considered legacy systems (i.e., outdated) due to various limitations in their capabilities. More modern systems, such as Git and Mercurial, are distributed, meaning that they do not need a centralized server to host the repository. These modern systems also include powerful merging tools that make it possible for multiple authors to work on the same files concurrently.

Paper Writing

• Imagine you drafted an excellent paragraph for a paper you are writing, but later ruin it. How would you retrieve the excellent version of your conclusion? Is it even possible?

• Imagine you have 5 co-authors. How would you manage the changes and comments they make to your paper? If you use LibreOffice Writer or Microsoft Word, what happens if you accept changes made using the Track Changes option? Do you have a history of those changes?

Solution

• Recovering the excellent version is only possible if you created a copy of the old version of the paper. The danger of losing good versions often leads to the problematic workflow illustrated in the PhD Comics cartoon at the top of this page.

• Collaborative writing with traditional word processors is cumbersome. Either every collaborator has to work on a document sequentially (slowing down the process of writing), or you have to send out a version to all collaborators and manually merge their comments into your document. The ‘track changes’ or ‘record changes’ option can highlight changes for you and simplifies merging, but as soon as you accept changes you will lose their history. You will then no longer know who suggested that change, why it was suggested, or when it was merged into the rest of the document. Even online word processors like Google Docs or Microsoft Office Online do not fully resolve these problems.

Key Points

• Version control is like an unlimited ‘undo’.

• Version control also allows many people to work in parallel.

Setting Up Git

Overview

Teaching: 5 min
Exercises: 0 min

Questions

• How do I get set up to use Git?

Objectives

• Configure git the first time it is used on a computer.

• Understand the meaning of the --global configuration flag.

When we use Git on a new computer for the first time, we need to configure a few things. Below are a few examples of configurations we will set as we get started with Git:

• our name and email address,
• what our preferred text editor is,
• and that we want to use these settings globally (i.e. for every project).

On a command line, Git commands are written as git verb options, where verb is what we actually want to do and options is additional optional information which may be needed for the verb. So here is how Dracula sets up his new laptop:

$ git config --global user.name "Vlad Dracula"
$ git config --global user.email "vlad@tran.sylvan.ia"

Please use your own name and email address instead of Dracula’s. This user name and email will be associated with your subsequent Git activity, which means that any changes pushed to GitHub, BitBucket, GitLab or another Git host server after this lesson will include this information.

For this lesson, we will be interacting with GitHub and so the email address used should be the same as the one used when setting up your GitHub account. If you are concerned about privacy, please review GitHub’s instructions for keeping your email address private.

Keeping your email private

If you elect to use a private email address with GitHub, then use that same email address for the user.email value, e.g. username@users.noreply.github.com replacing username with your GitHub one.

Line Endings

As with other keys, when you hit Enter or ↵ or on Macs, Return on your keyboard, your computer encodes this input as a character. Different operating systems use different character(s) to represent the end of a line. (You may also hear these referred to as newlines or line breaks.) Because Git uses these characters to compare files, it may cause unexpected issues when editing a file on different machines. Though it is beyond the scope of this lesson, you can read more about this issue in the Pro Git book.

You can change the way Git recognizes and encodes line endings using the core.autocrlf command to git config. The following settings are recommended:

On macOS and Linux:

$ git config --global core.autocrlf input

And on Windows:

$ git config --global core.autocrlf false

Dracula also has to set his favorite text editor, following this table:

It is possible to reconfigure the text editor for Git whenever you want to change it.

Exiting Vim

Note that Vim is the default editor for many programs. If you haven’t used Vim before and wish to exit a session without saving your changes, press Esc then type :q! and hit Enter or ↵ or on Macs, Return. If you want to save your changes and quit, press Esc then type :wq and hit Enter or ↵ or on Macs, Return.

Git (2.28+) allows configuration of the name of the branch created when you initialize any new repository. Dracula decides to use that feature to set it to main so it matches the cloud service he will eventually use.

$ git config --global init.defaultBranch main

Default Git branch naming

Source file changes are associated with a “branch.” For new learners in this lesson, it’s enough to know that branches exist, and this lesson uses one branch.
By default, Git will create a branch called master when you create a new repository with git init (as explained in the next Episode). This term evokes the racist practice of human slavery and the software development community has moved to adopt more inclusive language.

In 2020, most Git code hosting services transitioned to using main as the default branch. As an example, any new repository that is opened in GitHub and GitLab default to main. However, Git has not yet made the same change. As a result, local repositories must be manually configured have the same main branch name as most cloud services.

For versions of Git prior to 2.28, the change can be made on an individual repository level. The command for this is in the next episode. Note that if this value is unset in your local Git configuration, the init.defaultBranch value defaults to master.

The five commands we just ran above only need to be run once: the flag --global tells Git to use the settings for every project, in your user account, on this computer.

You can check your settings at any time:

$ git config --list

You can change your configuration as many times as you want: use the same commands to choose another editor or update your email address.

Proxy

In some networks you need to use a proxy. If this is the case, you may also need to tell Git about the proxy:

$ git config --global http.proxy proxy-url
$ git config --global https.proxy proxy-url

To disable the proxy, use

$ git config --global --unset http.proxy
$ git config --global --unset https.proxy

Git Help and Manual

Always remember that if you forget the subcommands or options of a git command, you can access the relevant list of options typing git <command> -h or access the corresponding Git manual by typing git <command> --help, e.g.:

$ git config -h
$ git config --help

While viewing the manual, remember the : is a prompt waiting for commands and you can press Q to exit the manual.

More generally, you can get the list of available git commands and further resources of the Git manual typing:

$ git help

Key Points

• Use git config with the --global option to configure a user name, email address, editor, and other preferences once per machine.

Creating a Repository

Overview

Teaching: 10 min
Exercises: 0 min

Questions

• Where does Git store information?

Objectives

• Create a local Git repository.

• Describe the purpose of the .git directory.

Once Git is configured, we can start using it.

We will continue with the story of Wolfman and Dracula who are investigating if it is possible to send a planetary lander to Mars.

Werewolf vs dracula by b-maze / Deviant Art. Mars by European Space Agency / CC-BY-SA 3.0 IGO. Pluto / Courtesy NASA/JPL-Caltech. Mummy © Gilad Fried / The Noun Project / CC BY 3.0. Moon © Luc Viatour / https://lucnix.be / CC BY-SA 3.0.

First, let’s create a new directory in the Desktop folder for our work and then change the current working directory to the newly created one:

$ cd ~/Desktop
$ mkdir planets
$ cd planets

Then we tell Git to make planets a repository – a place where Git can store versions of our files:

$ git init

It is important to note that git init will create a repository that can include subdirectories and their files—there is no need to create separate repositories nested within the planets repository, whether subdirectories are present from the beginning or added later. Also, note that the creation of the planets directory and its initialization as a repository are completely separate processes.

If we use ls to show the directory’s contents, it appears that nothing has changed:

$ ls

But if we add the -a flag to show everything, we can see that Git has created a hidden directory within planets called .git:

$ ls -a

.	..	.git

Git uses this special subdirectory to store all the information about the project, including the tracked files and sub-directories located within the project’s directory. If we ever delete the .git subdirectory, we will lose the project’s history.

Next, we will change the default branch to be called main. This might be the default branch depending on your settings and version of git. See the setup episode for more information on this change.

$ git checkout -b main

Switched to a new branch 'main'

We can check that everything is set up correctly by asking Git to tell us the status of our project:

$ git status

On branch main

No commits yet

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

If you are using a different version of git, the exact wording of the output might be slightly different.

Places to Create Git Repositories

Along with tracking information about planets (the project we have already created), Dracula would also like to track information about moons. Despite Wolfman’s concerns, Dracula creates a moons project inside his planets project with the following sequence of commands:

$ cd ~/Desktop   # return to Desktop directory
$ cd planets     # go into planets directory, which is already a Git repository
$ ls -a          # ensure the .git subdirectory is still present in the planets directory
$ mkdir moons    # make a subdirectory planets/moons
$ cd moons       # go into moons subdirectory
$ git init       # make the moons subdirectory a Git repository
$ ls -a          # ensure the .git subdirectory is present indicating we have created a new Git repository

Is the git init command, run inside the moons subdirectory, required for tracking files stored in the moons subdirectory?

Solution

No. Dracula does not need to make the moons subdirectory a Git repository because the planets repository can track any files, sub-directories, and subdirectory files under the planets directory. Thus, in order to track all information about moons, Dracula only needed to add the moons subdirectory to the planets directory.

Additionally, Git repositories can interfere with each other if they are “nested”: the outer repository will try to version-control the inner repository. Therefore, it’s best to create each new Git repository in a separate directory. To be sure that there is no conflicting repository in the directory, check the output of git status. If it looks like the following, you are good to go to create a new repository as shown above:

$ git status

fatal: Not a git repository (or any of the parent directories): .git

Correcting git init Mistakes

Wolfman explains to Dracula how a nested repository is redundant and may cause confusion down the road. Dracula would like to remove the nested repository. How can Dracula undo his last git init in the moons subdirectory?

Solution – USE WITH CAUTION!

Background

Removing files from a Git repository needs to be done with caution. But we have not learned yet how to tell Git to track a particular file; we will learn this in the next episode. Files that are not tracked by Git can easily be removed like any other “ordinary” files with

$ rm filename

Similarly a directory can be removed using rm -r dirname or rm -rf dirname. If the files or folder being removed in this fashion are tracked by Git, then their removal becomes another change that we will need to track, as we will see in the next episode.

Solution

Git keeps all of its files in the .git directory. To recover from this little mistake, Dracula can just remove the .git folder in the moons subdirectory by running the following command from inside the planets directory:

$ rm -rf moons/.git

But be careful! Running this command in the wrong directory will remove the entire Git history of a project you might want to keep. Therefore, always check your current directory using the command pwd.

Key Points

• git init initializes a repository.

• Git stores all of its repository data in the .git directory.

Tracking Changes

Overview

Teaching: 20 min
Exercises: 0 min

Questions

• How do I record changes in Git?

• How do I check the status of my version control repository?

• How do I record notes about what changes I made and why?

Objectives

• Go through the modify-add-commit cycle for one or more files.

• Explain where information is stored at each stage of that cycle.

• Distinguish between descriptive and non-descriptive commit messages.

First let’s make sure we’re still in the right directory. You should be in the planets directory.

$ cd ~/Desktop/planets

Let’s create a file called mars.txt that contains some notes about the Red Planet’s suitability as a base. We’ll use nano to edit the file; you can use whatever editor you like. In particular, this does not have to be the core.editor you set globally earlier. But remember, the bash command to create or edit a new file will depend on the editor you choose (it might not be nano). For a refresher on text editors, check out “Which Editor?” in The Unix Shell lesson.

$ nano mars.txt

Type the text below into the mars.txt file:

Cold and dry, but everything is my favorite color

Let’s first verify that the file was properly created by running the list command (ls):

$ ls

mars.txt

mars.txt contains a single line, which we can see by running:

$ cat mars.txt

Cold and dry, but everything is my favorite color

If we check the status of our project again, Git tells us that it’s noticed the new file:

$ git status

On branch main

No commits yet

Untracked files:
   (use "git add <file>..." to include in what will be committed)

	mars.txt

nothing added to commit but untracked files present (use "git add" to track)

The “untracked files” message means that there’s a file in the directory that Git isn’t keeping track of. We can tell Git to track a file using git add:

$ git add mars.txt

and then check that the right thing happened:

$ git status

On branch main

No commits yet

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

	new file:   mars.txt

Git now knows that it’s supposed to keep track of mars.txt, but it hasn’t recorded these changes as a commit yet. To get it to do that, we need to run one more command:

$ git commit -m "Start notes on Mars as a base"

[main (root-commit) f22b25e] Start notes on Mars as a base
 1 file changed, 1 insertion(+)
 create mode 100644 mars.txt

When we run git commit, Git takes everything we have told it to save by using git add and stores a copy permanently inside the special .git directory. This permanent copy is called a commit (or revision) and its short identifier is f22b25e. Your commit may have another identifier.

We use the -m flag (for “message”) to record a short, descriptive, and specific comment that will help us remember later on what we did and why. If we just run git commit without the -m option, Git will launch nano (or whatever other editor we configured as core.editor) so that we can write a longer message.

Good commit messages start with a brief (<50 characters) statement about the changes made in the commit. Generally, the message should complete the sentence “If applied, this commit will” . If you want to go into more detail, add a blank line between the summary line and your additional notes. Use this additional space to explain why you made changes and/or what their impact will be.

If we run git status now:

$ git status

On branch main
nothing to commit, working tree clean

it tells us everything is up to date. If we want to know what we’ve done recently, we can ask Git to show us the project’s history using git log:

$ git log

commit f22b25e3233b4645dabd0d81e651fe074bd8e73b
Author: Vlad Dracula <vlad@tran.sylvan.ia>
Date:   Thu Aug 22 09:51:46 2013 -0400

    Start notes on Mars as a base

git log lists all commits made to a repository in reverse chronological order. The listing for each commit includes the commit’s full identifier (which starts with the same characters as the short identifier printed by the git commit command earlier), the commit’s author, when it was created, and the log message Git was given when the commit was created.

Where Are My Changes?

If we run ls at this point, we will still see just one file called mars.txt. That’s because Git saves information about files’ history in the special .git directory mentioned earlier so that our filesystem doesn’t become cluttered (and so that we can’t accidentally edit or delete an old version).

Now suppose Dracula adds more information to the file. (Again, we’ll edit with nano and then cat the file to show its contents; you may use a different editor, and don’t need to cat.)

$ nano mars.txt
$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman

When we run git status now, it tells us that a file it already knows about has been modified:

$ git status

On branch main
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:   mars.txt

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

The last line is the key phrase: “no changes added to commit”. We have changed this file, but we haven’t told Git we will want to save those changes (which we do with git add) nor have we saved them (which we do with git commit). So let’s do that now. It is good practice to always review our changes before saving them. We do this using git diff. This shows us the differences between the current state of the file and the most recently saved version:

$ git diff

diff --git a/mars.txt b/mars.txt
index df0654a..315bf3a 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1 +1,2 @@
 Cold and dry, but everything is my favorite color
+The two moons may be a problem for Wolfman

The output is cryptic because it is actually a series of commands for tools like editors and patch telling them how to reconstruct one file given the other. If we break it down into pieces:

1. The first line tells us that Git is producing output similar to the Unix diff command comparing the old and new versions of the file.
2. The second line tells exactly which versions of the file Git is comparing; df0654a and 315bf3a are unique computer-generated labels for those versions.
3. The third and fourth lines once again show the name of the file being changed.
4. The remaining lines are the most interesting, they show us the actual differences and the lines on which they occur. In particular, the + marker in the first column shows where we added a line.

After reviewing our change, it’s time to commit it:

$ git commit -m "Add concerns about effects of Mars' moons on Wolfman"

On branch main
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:   mars.txt

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

Whoops: Git won’t commit because we didn’t use git add first. Let’s fix that:

$ git add mars.txt
$ git commit -m "Add concerns about effects of Mars' moons on Wolfman"

[main 34961b1] Add concerns about effects of Mars' moons on Wolfman
 1 file changed, 1 insertion(+)

Git insists that we add files to the set we want to commit before actually committing anything. This allows us to commit our changes in stages and capture changes in logical portions rather than only large batches. For example, suppose we’re adding a few citations to relevant research to our thesis. We might want to commit those additions, and the corresponding bibliography entries, but not commit some of our work drafting the conclusion (which we haven’t finished yet).

To allow for this, Git has a special staging area where it keeps track of things that have been added to the current changeset but not yet committed.

Staging Area

If you think of Git as taking snapshots of changes over the life of a project, git add specifies what will go in a snapshot (putting things in the staging area), and git commit then actually takes the snapshot, and makes a permanent record of it (as a commit). If you don’t have anything staged when you type git commit, Git will prompt you to use git commit -a or git commit --all, which is kind of like gathering everyone to take a group photo! However, it’s almost always better to explicitly add things to the staging area, because you might commit changes you forgot you made. (Going back to the group photo simile, you might get an extra with incomplete makeup walking on the stage for the picture because you used -a!) Try to stage things manually, or you might find yourself searching for “git undo commit” more than you would like!

Let’s watch as our changes to a file move from our editor to the staging area and into long-term storage. First, we’ll add another line to the file:

$ nano mars.txt
$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity

$ git diff

diff --git a/mars.txt b/mars.txt
index 315bf3a..b36abfd 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1,2 +1,3 @@
 Cold and dry, but everything is my favorite color
 The two moons may be a problem for Wolfman
+But the Mummy will appreciate the lack of humidity

So far, so good: we’ve added one line to the end of the file (shown with a + in the first column). Now let’s put that change in the staging area and see what git diff reports:

$ git add mars.txt
$ git diff

There is no output: as far as Git can tell, there’s no difference between what it’s been asked to save permanently and what’s currently in the directory. However, if we do this:

$ git diff --staged

diff --git a/mars.txt b/mars.txt
index 315bf3a..b36abfd 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1,2 +1,3 @@
 Cold and dry, but everything is my favorite color
 The two moons may be a problem for Wolfman
+But the Mummy will appreciate the lack of humidity

it shows us the difference between the last committed change and what’s in the staging area. Let’s save our changes:

$ git commit -m "Discuss concerns about Mars' climate for Mummy"

[main 005937f] Discuss concerns about Mars' climate for Mummy
 1 file changed, 1 insertion(+)

check our status:

$ git status

On branch main
nothing to commit, working tree clean

and look at the history of what we’ve done so far:

$ git log

commit 005937fbe2a98fb83f0ade869025dc2636b4dad5 (HEAD -> main)
Author: Vlad Dracula <vlad@tran.sylvan.ia>
Date:   Thu Aug 22 10:14:07 2013 -0400

    Discuss concerns about Mars' climate for Mummy

commit 34961b159c27df3b475cfe4415d94a6d1fcd064d
Author: Vlad Dracula <vlad@tran.sylvan.ia>
Date:   Thu Aug 22 10:07:21 2013 -0400

    Add concerns about effects of Mars' moons on Wolfman

commit f22b25e3233b4645dabd0d81e651fe074bd8e73b
Author: Vlad Dracula <vlad@tran.sylvan.ia>
Date:   Thu Aug 22 09:51:46 2013 -0400

    Start notes on Mars as a base

Word-based diffing

Sometimes, e.g. in the case of the text documents a line-wise diff is too coarse. That is where the --color-words option of git diff comes in very useful as it highlights the changed words using colors.

Paging the Log

When the output of git log is too long to fit in your screen, git uses a program to split it into pages of the size of your screen. When this “pager” is called, you will notice that the last line in your screen is a :, instead of your usual prompt.

• To get out of the pager, press Q.
• To move to the next page, press Spacebar.
• To search for some_word in all pages, press / and type some_word. Navigate through matches pressing N.

Limit Log Size

To avoid having git log cover your entire terminal screen, you can limit the number of commits that Git lists by using -N, where N is the number of commits that you want to view. For example, if you only want information from the last commit you can use:

$ git log -1

commit 005937fbe2a98fb83f0ade869025dc2636b4dad5 (HEAD -> main)
Author: Vlad Dracula <vlad@tran.sylvan.ia>
Date:   Thu Aug 22 10:14:07 2013 -0400

   Discuss concerns about Mars' climate for Mummy

You can also reduce the quantity of information using the --oneline option:

$ git log --oneline

005937f (HEAD -> main) Discuss concerns about Mars' climate for Mummy
34961b1 Add concerns about effects of Mars' moons on Wolfman
f22b25e Start notes on Mars as a base

You can also combine the --oneline option with others. One useful combination adds --graph to display the commit history as a text-based graph and to indicate which commits are associated with the current HEAD, the current branch main, or other Git references:

$ git log --oneline --graph

* 005937f (HEAD -> main) Discuss concerns about Mars' climate for Mummy
* 34961b1 Add concerns about effects of Mars' moons on Wolfman
* f22b25e Start notes on Mars as a base

Directories

Two important facts you should know about directories in Git.

1. Git does not track directories on their own, only files within them. Try it for yourself:

   $ mkdir spaceships
   $ git status
   $ git add spaceships
   $ git status
   

   Note, our newly created empty directory spaceships does not appear in the list of untracked files even if we explicitly add it (via git add) to our repository. This is the reason why you will sometimes see .gitkeep files in otherwise empty directories. Unlike .gitignore, these files are not special and their sole purpose is to populate a directory so that Git adds it to the repository. In fact, you can name such files anything you like.

2. If you create a directory in your Git repository and populate it with files, you can add all files in the directory at once by:

   git add <directory-with-files>
   

   Try it for yourself:

   $ touch spaceships/apollo-11 spaceships/sputnik-1
   $ git status
   $ git add spaceships
   $ git status
   

   Before moving on, we will commit these changes.

   $ git commit -m "Add some initial thoughts on spaceships"
   

To recap, when we want to add changes to our repository, we first need to add the changed files to the staging area (git add) and then commit the staged changes to the repository (git commit):

Choosing a Commit Message

Which of the following commit messages would be most appropriate for the last commit made to mars.txt?

1. “Changes”
2. “Added line ‘But the Mummy will appreciate the lack of humidity’ to mars.txt”
3. “Discuss effects of Mars’ climate on the Mummy”

Solution

Answer 1 is not descriptive enough, and the purpose of the commit is unclear; and answer 2 is redundant to using “git diff” to see what changed in this commit; but answer 3 is good: short, descriptive, and imperative.

Committing Changes to Git

Which command(s) below would save the changes of myfile.txt to my local Git repository?

1. $ git commit -m "my recent changes"
   

2. $ git init myfile.txt
   $ git commit -m "my recent changes"
   

3. $ git add myfile.txt
   $ git commit -m "my recent changes"
   

4. $ git commit -m myfile.txt "my recent changes"
   

Solution

1. Would only create a commit if files have already been staged.
2. Would try to create a new repository.
3. Is correct: first add the file to the staging area, then commit.
4. Would try to commit a file “my recent changes” with the message myfile.txt.

Committing Multiple Files

The staging area can hold changes from any number of files that you want to commit as a single snapshot.

1. Add some text to mars.txt noting your decision to consider Venus as a base
2. Create a new file venus.txt with your initial thoughts about Venus as a base for you and your friends
3. Add changes from both files to the staging area, and commit those changes.

Solution

The output below from cat mars.txt reflects only content added during this exercise. Your output may vary.

First we make our changes to the mars.txt and venus.txt files:

$ nano mars.txt
$ cat mars.txt

Maybe I should start with a base on Venus.

$ nano venus.txt
$ cat venus.txt

Venus is a nice planet and I definitely should consider it as a base.

Now you can add both files to the staging area. We can do that in one line:

$ git add mars.txt venus.txt

Or with multiple commands:

$ git add mars.txt
$ git add venus.txt

Now the files are ready to commit. You can check that using git status. If you are ready to commit use:

$ git commit -m "Write plans to start a base on Venus"

[main cc127c2]
 Write plans to start a base on Venus
 2 files changed, 2 insertions(+)
 create mode 100644 venus.txt

bio Repository

• Create a new Git repository on your computer called bio.
• Write a three-line biography for yourself in a file called me.txt, commit your changes
• Modify one line, add a fourth line
• Display the differences between its updated state and its original state.

Solution

If needed, move out of the planets folder:

$ cd ..

Create a new folder called bio and ‘move’ into it:

$ mkdir bio
$ cd bio

Initialise git:

$ git init

Create your biography file me.txt using nano or another text editor. Once in place, add and commit it to the repository:

$ git add me.txt
$ git commit -m "Add biography file" 

Modify the file as described (modify one line, add a fourth line). To display the differences between its updated state and its original state, use git diff:

$ git diff me.txt

Key Points

• git status shows the status of a repository.

• Files can be stored in a project’s working directory (which users see), the staging area (where the next commit is being built up) and the local repository (where commits are permanently recorded).

• git add puts files in the staging area.

• git commit saves the staged content as a new commit in the local repository.

• Write a commit message that accurately describes your changes.

Exploring History

Overview

Teaching: 25 min
Exercises: 0 min

Questions

• How can I identify old versions of files?

• How do I review my changes?

• How can I recover old versions of files?

Objectives

• Explain what the HEAD of a repository is and how to use it.

• Identify and use Git commit numbers.

• Compare various versions of tracked files.

• Restore old versions of files.

As we saw in the previous episode, we can refer to commits by their identifiers. You can refer to the most recent commit of the working directory by using the identifier HEAD.

We’ve been adding one line at a time to mars.txt, so it’s easy to track our progress by looking, so let’s do that using our HEADs. Before we start, let’s make a change to mars.txt, adding yet another line.

$ nano mars.txt
$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
An ill-considered change

Now, let’s see what we get.

$ git diff HEAD mars.txt

diff --git a/mars.txt b/mars.txt
index b36abfd..0848c8d 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1,3 +1,4 @@
 Cold and dry, but everything is my favorite color
 The two moons may be a problem for Wolfman
 But the Mummy will appreciate the lack of humidity
+An ill-considered change.

which is the same as what you would get if you leave out HEAD (try it). The real goodness in all this is when you can refer to previous commits. We do that by adding ~1 (where “~” is “tilde”, pronounced [til-duh]) to refer to the commit one before HEAD.

$ git diff HEAD~1 mars.txt

If we want to see the differences between older commits we can use git diff again, but with the notation HEAD~1, HEAD~2, and so on, to refer to them:

$ git diff HEAD~3 mars.txt

diff --git a/mars.txt b/mars.txt
index df0654a..b36abfd 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1 +1,4 @@
 Cold and dry, but everything is my favorite color
+The two moons may be a problem for Wolfman
+But the Mummy will appreciate the lack of humidity
+An ill-considered change

We could also use git show which shows us what changes we made at an older commit as well as the commit message, rather than the differences between a commit and our working directory that we see by using git diff.

$ git show HEAD~3 mars.txt

commit f22b25e3233b4645dabd0d81e651fe074bd8e73b
Author: Vlad Dracula <vlad@tran.sylvan.ia>
Date:   Thu Aug 22 09:51:46 2013 -0400

    Start notes on Mars as a base

diff --git a/mars.txt b/mars.txt
new file mode 100644
index 0000000..df0654a
--- /dev/null
+++ b/mars.txt
@@ -0,0 +1 @@
+Cold and dry, but everything is my favorite color

In this way, we can build up a chain of commits. The most recent end of the chain is referred to as HEAD; we can refer to previous commits using the ~ notation, so HEAD~1 means “the previous commit”, while HEAD~123 goes back 123 commits from where we are now.

We can also refer to commits using those long strings of digits and letters that git log displays. These are unique IDs for the changes, and “unique” really does mean unique: every change to any set of files on any computer has a unique 40-character identifier. Our first commit was given the ID f22b25e3233b4645dabd0d81e651fe074bd8e73b, so let’s try this:

$ git diff f22b25e3233b4645dabd0d81e651fe074bd8e73b mars.txt

diff --git a/mars.txt b/mars.txt
index df0654a..93a3e13 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1 +1,4 @@
 Cold and dry, but everything is my favorite color
+The two moons may be a problem for Wolfman
+But the Mummy will appreciate the lack of humidity
+An ill-considered change

That’s the right answer, but typing out random 40-character strings is annoying, so Git lets us use just the first few characters (typically seven for normal size projects):

$ git diff f22b25e mars.txt

diff --git a/mars.txt b/mars.txt
index df0654a..93a3e13 100644
--- a/mars.txt
+++ b/mars.txt
@@ -1 +1,4 @@
 Cold and dry, but everything is my favorite color
+The two moons may be a problem for Wolfman
+But the Mummy will appreciate the lack of humidity
+An ill-considered change

All right! So we can save changes to files and see what we’ve changed. Now, how can we restore older versions of things? Let’s suppose we change our mind about the last update to mars.txt (the “ill-considered change”).

git status now tells us that the file has been changed, but those changes haven’t been staged:

$ git status

On branch main
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:   mars.txt

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

We can put things back the way they were by using git checkout:

$ git checkout HEAD mars.txt
$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity

As you might guess from its name, git checkout checks out (i.e., restores) an old version of a file. In this case, we’re telling Git that we want to recover the version of the file recorded in HEAD, which is the last saved commit. If we want to go back even further, we can use a commit identifier instead:

$ git checkout f22b25e mars.txt

$ cat mars.txt

Cold and dry, but everything is my favorite color

$ git status

On branch main
Changes to be committed:
  (use "git reset HEAD <file>..." to unstage)

    modified:   mars.txt

Notice that the changes are currently in the staging area. Again, we can put things back the way they were by using git checkout:

$ git checkout HEAD mars.txt

Don’t Lose Your HEAD

Above we used

$ git checkout f22b25e mars.txt

to revert mars.txt to its state after the commit f22b25e. But be careful! The command checkout has other important functionalities and Git will misunderstand your intentions if you are not accurate with the typing. For example, if you forget mars.txt in the previous command.

$ git checkout f22b25e

Note: checking out 'f22b25e'.

You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by performing another checkout.

If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -b with the checkout command again. Example:

 git checkout -b <new-branch-name>

HEAD is now at f22b25e Start notes on Mars as a base

The “detached HEAD” is like “look, but don’t touch” here, so you shouldn’t make any changes in this state. After investigating your repo’s past state, reattach your HEAD with git checkout main.

It’s important to remember that we must use the commit number that identifies the state of the repository before the change we’re trying to undo. A common mistake is to use the number of the commit in which we made the change we’re trying to discard. In the example below, we want to retrieve the state from before the most recent commit (HEAD~1), which is commit f22b25e:

So, to put it all together, here’s how Git works in cartoon form:

Simplifying the Common Case

If you read the output of git status carefully, you’ll see that it includes this hint:

(use "git checkout -- <file>..." to discard changes in working directory)

As it says, git checkout without a version identifier restores files to the state saved in HEAD. The double dash -- is needed to separate the names of the files being recovered from the command itself: without it, Git would try to use the name of the file as the commit identifier.

The fact that files can be reverted one by one tends to change the way people organize their work. If everything is in one large document, it’s hard (but not impossible) to undo changes to the introduction without also undoing changes made later to the conclusion. If the introduction and conclusion are stored in separate files, on the other hand, moving backward and forward in time becomes much easier.

Recovering Older Versions of a File

Jennifer has made changes to the Python script that she has been working on for weeks, and the modifications she made this morning “broke” the script and it no longer runs. She has spent ~ 1hr trying to fix it, with no luck…

Luckily, she has been keeping track of her project’s versions using Git! Which commands below will let her recover the last committed version of her Python script called data_cruncher.py?

1. $ git checkout HEAD

2. $ git checkout HEAD data_cruncher.py

3. $ git checkout HEAD~1 data_cruncher.py

4. $ git checkout <unique ID of last commit> data_cruncher.py

5. Both 2 and 4

Solution

The answer is (5)-Both 2 and 4.

The checkout command restores files from the repository, overwriting the files in your working directory. Answers 2 and 4 both restore the latest version in the repository of the file data_cruncher.py. Answer 2 uses HEAD to indicate the latest, whereas answer 4 uses the unique ID of the last commit, which is what HEAD means.

Answer 3 gets the version of data_cruncher.py from the commit before HEAD, which is NOT what we wanted.

Answer 1 can be dangerous! Without a filename, git checkout will restore all files in the current directory (and all directories below it) to their state at the commit specified. This command will restore data_cruncher.py to the latest commit version, but it will also restore any other files that are changed to that version, erasing any changes you may have made to those files! As discussed above, you are left in a detached HEAD state, and you don’t want to be there.

Reverting a Commit

Jennifer is collaborating with colleagues on her Python script. She realizes her last commit to the project’s repository contained an error, and wants to undo it. Jennifer wants to undo correctly so everyone in the project’s repository gets the correct change. The command git revert [erroneous commit ID] will create a new commit that reverses the erroneous commit.

The command git revert is different from git checkout [commit ID] because git checkout returns the files not yet committed within the local repository to a previous state, whereas git revert reverses changes committed to the local and project repositories.

Below are the right steps and explanations for Jennifer to use git revert, what is the missing command?

1. ________ # Look at the git history of the project to find the commit ID

2. Copy the ID (the first few characters of the ID, e.g. 0b1d055).

3. git revert [commit ID]

4. Type in the new commit message.

5. Save and close

Solution

The command git log lists project history with commit IDs.

The command git show HEAD shows changes made at the latest commit, and lists the commit ID; however, Jennifer should double-check it is the correct commit, and no one else has committed changes to the repository.

Understanding Workflow and History

What is the output of the last command in

$ cd planets
$ echo "Venus is beautiful and full of love" > venus.txt
$ git add venus.txt
$ echo "Venus is too hot to be suitable as a base" >> venus.txt
$ git commit -m "Comment on Venus as an unsuitable base"
$ git checkout HEAD venus.txt
$ cat venus.txt #this will print the contents of venus.txt to the screen

1. Venus is too hot to be suitable as a base
   

2. Venus is beautiful and full of love
   

3. Venus is beautiful and full of love
   Venus is too hot to be suitable as a base
   

4. Error because you have changed venus.txt without committing the changes
   

Solution

The answer is 2.

The command git add venus.txt places the current version of venus.txt into the staging area. The changes to the file from the second echo command are only applied to the working copy, not the version in the staging area.

So, when git commit -m "Comment on Venus as an unsuitable base" is executed, the version of venus.txt committed to the repository is the one from the staging area and has only one line.

At this time, the working copy still has the second line (and git status will show that the file is modified). However, git checkout HEAD venus.txt replaces the working copy with the most recently committed version of venus.txt.

So, cat venus.txt will output

 Venus is beautiful and full of love.

Checking Understanding of git diff

Consider this command: git diff HEAD~9 mars.txt. What do you predict this command will do if you execute it? What happens when you do execute it? Why?

Try another command, git diff [ID] mars.txt, where [ID] is replaced with the unique identifier for your most recent commit. What do you think will happen, and what does happen?

Getting Rid of Staged Changes

git checkout can be used to restore a previous commit when unstaged changes have been made, but will it also work for changes that have been staged but not committed? Make a change to mars.txt, add that change, and use git checkout to see if you can remove your change.

Explore and Summarize Histories

Exploring history is an important part of Git, and often it is a challenge to find the right commit ID, especially if the commit is from several months ago.

Imagine the planets project has more than 50 files. You would like to find a commit that modifies some specific text in mars.txt. When you type git log, a very long list appeared. How can you narrow down the search?

Recall that the git diff command allows us to explore one specific file, e.g., git diff mars.txt. We can apply a similar idea here.

$ git log mars.txt

Unfortunately some of these commit messages are very ambiguous, e.g., update files. How can you search through these files?

Both git diff and git log are very useful and they summarize a different part of the history for you. Is it possible to combine both? Let’s try the following:

$ git log --patch mars.txt

You should get a long list of output, and you should be able to see both commit messages and the difference between each commit.

Question: What does the following command do?

$ git log --patch HEAD~9 *.txt

Key Points

• git diff displays differences between commits.

• git checkout recovers old versions of files.

Ignoring Things

Overview

Teaching: 5 min
Exercises: 0 min

Questions

• How can I tell Git to ignore files I don’t want to track?

Objectives

• Configure Git to ignore specific files.

• Explain why ignoring files can be useful.

What if we have files that we do not want Git to track for us, like backup files created by our editor or intermediate files created during data analysis? Let’s create a few dummy files:

$ mkdir results
$ touch a.dat b.dat c.dat results/a.out results/b.out

and see what Git says:

$ git status

On branch main
Untracked files:
  (use "git add <file>..." to include in what will be committed)

	a.dat
	b.dat
	c.dat
	results/

nothing added to commit but untracked files present (use "git add" to track)

Putting these files under version control would be a waste of disk space. What’s worse, having them all listed could distract us from changes that actually matter, so let’s tell Git to ignore them.

We do this by creating a file in the root directory of our project called .gitignore:

$ nano .gitignore
$ cat .gitignore

*.dat
results/

These patterns tell Git to ignore any file whose name ends in .dat and everything in the results directory. (If any of these files were already being tracked, Git would continue to track them.)

Once we have created this file, the output of git status is much cleaner:

$ git status

On branch main
Untracked files:
  (use "git add <file>..." to include in what will be committed)

	.gitignore

nothing added to commit but untracked files present (use "git add" to track)

The only thing Git notices now is the newly-created .gitignore file. You might think we wouldn’t want to track it, but everyone we’re sharing our repository with will probably want to ignore the same things that we’re ignoring. Let’s add and commit .gitignore:

$ git add .gitignore
$ git commit -m "Ignore data files and the results folder."
$ git status

On branch main
nothing to commit, working tree clean

As a bonus, using .gitignore helps us avoid accidentally adding files to the repository that we don’t want to track:

$ git add a.dat

The following paths are ignored by one of your .gitignore files:
a.dat
Use -f if you really want to add them.

If we really want to override our ignore settings, we can use git add -f to force Git to add something. For example, git add -f a.dat. We can also always see the status of ignored files if we want:

$ git status --ignored

On branch main
Ignored files:
 (use "git add -f <file>..." to include in what will be committed)

        a.dat
        b.dat
        c.dat
        results/

nothing to commit, working tree clean

Ignoring Nested Files

Given a directory structure that looks like:

results/data
results/plots

How would you ignore only results/plots and not results/data?

Solution

If you only want to ignore the contents of results/plots, you can change your .gitignore to ignore only the /plots/ subfolder by adding the following line to your .gitignore:

results/plots/

This line will ensure only the contents of results/plots is ignored, and not the contents of results/data.

As with most programming issues, there are a few alternative ways that one may ensure this ignore rule is followed. The “Ignoring Nested Files: Variation” exercise has a slightly different directory structure that presents an alternative solution. Further, the discussion page has more detail on ignore rules.

Including Specific Files

How would you ignore all .dat files in your root directory except for final.dat? Hint: Find out what ! (the exclamation point operator) does

Solution

You would add the following two lines to your .gitignore:

*.dat           # ignore all data files
!final.dat      # except final.data

The exclamation point operator will include a previously excluded entry.

Note also that because you’ve previously committed .dat files in this lesson they will not be ignored with this new rule. Only future additions of .dat files added to the root directory will be ignored.

Ignoring Nested Files: Variation

Given a directory structure that looks similar to the earlier Nested Files exercise, but with a slightly different directory structure:

results/data
results/images
results/plots
results/analysis

How would you ignore all of the contents in the results folder, but not results/data?

Hint: think a bit about how you created an exception with the ! operator before.

Solution

If you want to ignore the contents of results/ but not those of results/data/, you can change your .gitignore to ignore the contents of results folder, but create an exception for the contents of the results/data subfolder. Your .gitignore would look like this:

results/*               # ignore everything in results folder
!results/data/          # do not ignore results/data/ contents

Ignoring all data Files in a Directory

Assuming you have an empty .gitignore file, and given a directory structure that looks like:

results/data/position/gps/a.dat
results/data/position/gps/b.dat
results/data/position/gps/c.dat
results/data/position/gps/info.txt
results/plots

What’s the shortest .gitignore rule you could write to ignore all .dat files in result/data/position/gps? Do not ignore the info.txt.

Solution

Appending results/data/position/gps/*.dat will match every file in results/data/position/gps that ends with .dat. The file results/data/position/gps/info.txt will not be ignored.

Ignoring all data Files in the repository

Let us assume you have many .dat files in different subdirectories of your repository. For example, you might have:

results/a.dat
data/experiment_1/b.dat
data/experiment_2/c.dat
data/experiment_2/variation_1/d.dat

How do you ignore all the .dat files, without explicitly listing the names of the corresponding folders?

Solution

In the .gitignore file, write:

**/*.dat               

This will ignore all the .dat files, regardless of their position in the directory tree. You can still include some specific exception with the exclamation point operator.

The Order of Rules

Given a .gitignore file with the following contents:

*.dat
!*.dat

What will be the result?

Solution

The ! modifier will negate an entry from a previously defined ignore pattern. Because the !*.dat entry negates all of the previous .dat files in the .gitignore, none of them will be ignored, and all .dat files will be tracked.

Log Files

You wrote a script that creates many intermediate log-files of the form log_01, log_02, log_03, etc. You want to keep them but you do not want to track them through git.

1. Write one .gitignore entry that excludes files of the form log_01, log_02, etc.

2. Test your “ignore pattern” by creating some dummy files of the form log_01, etc.

3. You find that the file log_01 is very important after all, add it to the tracked files without changing the .gitignore again.

4. Discuss with your neighbor what other types of files could reside in your directory that you do not want to track and thus would exclude via .gitignore.

Solution

1. append either log_* or log* as a new entry in your .gitignore
2. track log_01 using git add -f log_01

Key Points

• The .gitignore file tells Git what files to ignore.

Remotes in GitHub

Overview

Teaching: 30 min
Exercises: 0 min

Questions

• How do I share my changes with others on the web?

Objectives

• Explain what remote repositories are and why they are useful.

• Push to or pull from a remote repository.

Version control really comes into its own when we begin to collaborate with other people. We already have most of the machinery we need to do this; the only thing missing is to copy changes from one repository to another.

Systems like Git allow us to move work between any two repositories. In practice, though, it’s easiest to use one copy as a central hub, and to keep it on the web rather than on someone’s laptop. Most programmers use hosting services like GitHub, Bitbucket or GitLab to hold those main copies; we’ll explore the pros and cons of this in a later episode.

Let’s start by sharing the changes we’ve made to our current project with the world. To this end we are going to create a remote repository that will be linked to our local repository.

1. Create a remote repository

Log in to GitHub, then click on the icon in the top right corner to create a new repository called planets:

Name your repository “planets” and then click “Create Repository”.

Note: Since this repository will be connected to a local repository, it needs to be empty. Leave “Initialize this repository with a README” unchecked, and keep “None” as options for both “Add .gitignore” and “Add a license.” See the “GitHub License and README files” exercise below for a full explanation of why the repository needs to be empty.

As soon as the repository is created, GitHub displays a page with a URL and some information on how to configure your local repository:

This effectively does the following on GitHub’s servers:

$ mkdir planets
$ cd planets
$ git init

If you remember back to the earlier episode where we added and committed our earlier work on mars.txt, we had a diagram of the local repository which looked like this:

Now that we have two repositories, we need a diagram like this:

Note that our local repository still contains our earlier work on mars.txt, but the remote repository on GitHub appears empty as it doesn’t contain any files yet.

2. Connect local to remote repository

Now we connect the two repositories. We do this by making the GitHub repository a remote for the local repository. The home page of the repository on GitHub includes the URL string we need to identify it:

Click on the ‘SSH’ link to change the protocol from HTTPS to SSH.

HTTPS vs. SSH

We use SSH here because, while it requires some additional configuration, it is a security protocol widely used by many applications. The steps below describe SSH at a minimum level for GitHub. A supplemental episode to this lesson discusses advanced setup and concepts of SSH and key pairs, and other material supplemental to git related SSH.

Copy that URL from the browser, go into the local planets repository, and run this command:

$ git remote add origin git@github.com:vlad/planets.git

Make sure to use the URL for your repository rather than Vlad’s: the only difference should be your username instead of vlad.

origin is a local name used to refer to the remote repository. It could be called anything, but origin is a convention that is often used by default in git and GitHub, so it’s helpful to stick with this unless there’s a reason not to.

We can check that the command has worked by running git remote -v:

$ git remote -v

origin   git@github.com:vlad/planets.git (fetch)
origin   git@github.com:vlad/planets.git (push)

We’ll discuss remotes in more detail in the next episode, while talking about how they might be used for collaboration.

4. Push local changes to a remote

This command will push the changes from our local repository to the repository on GitHub:

$ git push origin main

Since Dracula set up a passphrase for his SSH key, git push will prompt him for it.
If you completed advanced settings for your authentication, it will not prompt for a passphrase.

Enumerating objects: 16, done.
Counting objects: 100% (16/16), done.
Delta compression using up to 8 threads.
Compressing objects: 100% (11/11), done.
Writing objects: 100% (16/16), 1.45 KiB | 372.00 KiB/s, done.
Total 16 (delta 2), reused 0 (delta 0)
remote: Resolving deltas: 100% (2/2), done.
To https://github.com/vlad/planets.git
 * [new branch]      main -> main

Proxy

If the network you are connected to uses a proxy, there is a chance that your last command failed with “Could not resolve hostname” as the error message. To solve this issue, you need to tell Git about the proxy:

$ git config --global http.proxy http://user:password@proxy.url
$ git config --global https.proxy https://user:password@proxy.url

When you connect to another network that doesn’t use a proxy, you will need to tell Git to disable the proxy using:

$ git config --global --unset http.proxy
$ git config --global --unset https.proxy

Password Managers

If your operating system has a password manager configured, git push will try to use it when it needs your username and password. For example, this is the default behavior for Git Bash on Windows. If you want to type your username and password at the terminal instead of using a password manager, type:

$ unset SSH_ASKPASS

in the terminal, before you run git push. Despite the name, Git uses SSH_ASKPASS for all credential entry, so you may want to unset SSH_ASKPASS whether you are using Git via SSH or https.

You may also want to add unset SSH_ASKPASS at the end of your ~/.bashrc to make Git default to using the terminal for usernames and passwords.

Our local and remote repositories are now in this state:

The ‘-u’ Flag

You may see a -u option used with git push in some documentation. This option is synonymous with the --set-upstream option for the git branch command, and is used to associate the current branch with a remote branch so that the git pull command can be used without any arguments. To do this, simply use git push -u origin main once the remote has been set up.

We can pull changes from the remote repository to the local one as well:

$ git pull origin main

From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
Already up-to-date.

Pulling has no effect in this case because the two repositories are already synchronized. If someone else had pushed some changes to the repository on GitHub, though, this command would download them to our local repository.

GitHub GUI

Browse to your planets repository on GitHub. Under the Code tab, find and click on the text that says “XX commits” (where “XX” is some number). Hover over, and click on, the three buttons to the right of each commit. What information can you gather/explore from these buttons? How would you get that same information in the shell?

Solution

The left-most button (with the picture of a clipboard) copies the full identifier of the commit to the clipboard. In the shell, git log will show you the full commit identifier for each commit.

When you click on the middle button, you’ll see all of the changes that were made in that particular commit. Green shaded lines indicate additions and red ones removals. In the shell we can do the same thing with git diff. In particular, git diff ID1..ID2 where ID1 and ID2 are commit identifiers (e.g. git diff a3bf1e5..041e637) will show the differences between those two commits.

The right-most button lets you view all of the files in the repository at the time of that commit. To do this in the shell, we’d need to checkout the repository at that particular time. We can do this with git checkout ID where ID is the identifier of the commit we want to look at. If we do this, we need to remember to put the repository back to the right state afterwards!

Uploading files directly in GitHub browser

Github also allows you to skip the command line and upload files directly to your repository without having to leave the browser. There are two options. First you can click the “Upload files” button in the toolbar at the top of the file tree. Or, you can drag and drop files from your desktop onto the file tree. You can read more about this on this GitHub page

GitHub Timestamp

Create a remote repository on GitHub. Push the contents of your local repository to the remote. Make changes to your local repository and push these changes. Go to the repo you just created on GitHub and check the timestamps of the files. How does GitHub record times, and why?

Solution

GitHub displays timestamps in a human readable relative format (i.e. “22 hours ago” or “three weeks ago”). However, if you hover over the timestamp, you can see the exact time at which the last change to the file occurred.

Push vs. Commit

In this episode, we introduced the “git push” command. How is “git push” different from “git commit”?

Solution

When we push changes, we’re interacting with a remote repository to update it with the changes we’ve made locally (often this corresponds to sharing the changes we’ve made with others). Commit only updates your local repository.

GitHub License and README files

In this episode we learned about creating a remote repository on GitHub, but when you initialized your GitHub repo, you didn’t add a README.md or a license file. If you had, what do you think would have happened when you tried to link your local and remote repositories?

Solution

In this case, we’d see a merge conflict due to unrelated histories. When GitHub creates a README.md file, it performs a commit in the remote repository. When you try to pull the remote repository to your local repository, Git detects that they have histories that do not share a common origin and refuses to merge.

$ git pull origin main

warning: no common commits
remote: Enumerating objects: 3, done.
remote: Counting objects: 100% (3/3), done.
remote: Total 3 (delta 0), reused 0 (delta 0), pack-reused 0
Unpacking objects: 100% (3/3), done.
From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
 * [new branch]      main     -> origin/main
fatal: refusing to merge unrelated histories

You can force git to merge the two repositories with the option --allow-unrelated-histories. Be careful when you use this option and carefully examine the contents of local and remote repositories before merging.

$ git pull --allow-unrelated-histories origin main

From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
Merge made by the 'recursive' strategy.
README.md | 1 +
1 file changed, 1 insertion(+)
create mode 100644 README.md

Key Points

• A local Git repository can be connected to one or more remote repositories.

• Use the SSH protocol to connect to remote repositories.

• git push copies changes from a local repository to a remote repository.

• git pull copies changes from a remote repository to a local repository.

Collaborating

Overview

Teaching: 25 min
Exercises: 0 min

Questions

• How can I use version control to collaborate with other people?

Objectives

• Clone a remote repository.

• Collaborate by pushing to a common repository.

• Describe the basic collaborative workflow.

For the next step, get into pairs. One person will be the “Owner” and the other will be the “Collaborator”. The goal is that the Collaborator add changes into the Owner’s repository. We will switch roles at the end, so both persons will play Owner and Collaborator.

Practicing By Yourself

If you’re working through this lesson on your own, you can carry on by opening a second terminal window. This window will represent your partner, working on another computer. You won’t need to give anyone access on GitHub, because both ‘partners’ are you.

The Owner needs to give the Collaborator access. On GitHub, click the “Settings” button on the right, select “Collaborators”, click “Add people”, and then enter your partner’s username.

To accept access to the Owner’s repo, the Collaborator needs to go to https://github.com/notifications or check for email notification. Once there she can accept access to the Owner’s repo.

Next, the Collaborator needs to download a copy of the Owner’s repository to her machine. This is called “cloning a repo”.

The Collaborator doesn’t want to overwrite her own version of planets.git, so needs to clone the Owner’s repository to a different location than her own repository with the same name.

To clone the Owner’s repo into her Desktop folder, the Collaborator enters:

$ git clone git@github.com:vlad/planets.git ~/Desktop/vlad-planets

Replace ‘vlad’ with the Owner’s username.

If you choose to clone without the clone path (~/Desktop/vlad-planets) specified at the end, you will clone inside your own planets folder! Make sure to navigate to the Desktop folder first.

The Collaborator can now make a change in her clone of the Owner’s repository, exactly the same way as we’ve been doing before:

$ cd ~/Desktop/vlad-planets
$ nano pluto.txt
$ cat pluto.txt

It is so a planet!

$ git add pluto.txt
$ git commit -m "Add notes about Pluto"

 1 file changed, 1 insertion(+)
 create mode 100644 pluto.txt

Then push the change to the Owner’s repository on GitHub:

$ git push origin main

Enumerating objects: 4, done.
Counting objects: 4, done.
Delta compression using up to 4 threads.
Compressing objects: 100% (2/2), done.
Writing objects: 100% (3/3), 306 bytes, done.
Total 3 (delta 0), reused 0 (delta 0)
To https://github.com/vlad/planets.git
   9272da5..29aba7c  main -> main

Note that we didn’t have to create a remote called origin: Git uses this name by default when we clone a repository. (This is why origin was a sensible choice earlier when we were setting up remotes by hand.)

Take a look at the Owner’s repository on GitHub again, and you should be able to see the new commit made by the Collaborator. You may need to refresh your browser to see the new commit.

Some more about remotes

In this episode and the previous one, our local repository has had a single “remote”, called origin. A remote is a copy of the repository that is hosted somewhere else, that we can push to and pull from, and there’s no reason that you have to work with only one. For example, on some large projects you might have your own copy in your own GitHub account (you’d probably call this origin) and also the main “upstream” project repository (let’s call this upstream for the sake of examples). You would pull from upstream from time to time to get the latest updates that other people have committed.

Remember that the name you give to a remote only exists locally. It’s an alias that you choose - whether origin, or upstream, or fred - and not something intrinstic to the remote repository.

The git remote family of commands is used to set up and alter the remotes associated with a repository. Here are some of the most useful ones:

• git remote -v lists all the remotes that are configured (we already used this in the last episode)
• git remote add [name] [url] is used to add a new remote
• git remote remove [name] removes a remote. Note that it doesn’t affect the remote repository at all - it just removes the link to it from the local repo.
• git remote set-url [name] [newurl] changes the URL that is associated with the remote. This is useful if it has moved, e.g. to a different GitHub account, or from GitHub to a different hosting service. Or, if we made a typo when adding it!
• git remote rename [oldname] [newname] changes the local alias by which a remote is known - its name. For example, one could use this to change upstream to fred.

To download the Collaborator’s changes from GitHub, the Owner now enters:

$ git pull origin main

remote: Enumerating objects: 4, done.
remote: Counting objects: 100% (4/4), done.
remote: Compressing objects: 100% (2/2), done.
remote: Total 3 (delta 0), reused 3 (delta 0), pack-reused 0
Unpacking objects: 100% (3/3), done.
From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
   9272da5..29aba7c  main     -> origin/main
Updating 9272da5..29aba7c
Fast-forward
 pluto.txt | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 pluto.txt

Now the three repositories (Owner’s local, Collaborator’s local, and Owner’s on GitHub) are back in sync.

A Basic Collaborative Workflow

In practice, it is good to be sure that you have an updated version of the repository you are collaborating on, so you should git pull before making our changes. The basic collaborative workflow would be:

• update your local repo with git pull origin main,
• make your changes and stage them with git add,
• commit your changes with git commit -m, and
• upload the changes to GitHub with git push origin main

It is better to make many commits with smaller changes rather than of one commit with massive changes: small commits are easier to read and review.

Switch Roles and Repeat

Switch roles and repeat the whole process.

Review Changes

The Owner pushed commits to the repository without giving any information to the Collaborator. How can the Collaborator find out what has changed with command line? And on GitHub?

Solution

On the command line, the Collaborator can use git fetch origin main to get the remote changes into the local repository, but without merging them. Then by running git diff main origin/main the Collaborator will see the changes output in the terminal.

On GitHub, the Collaborator can go to the repository and click on “commits” to view the most recent commits pushed to the repository.

Comment Changes in GitHub

The Collaborator has some questions about one line change made by the Owner and has some suggestions to propose.

With GitHub, it is possible to comment on the diff of a commit. Over the line of code to comment, a blue comment icon appears to open a comment window.

The Collaborator posts her comments and suggestions using the GitHub interface.

Version History, Backup, and Version Control

Some backup software can keep a history of the versions of your files. They also allows you to recover specific versions. How is this functionality different from version control? What are some of the benefits of using version control, Git and GitHub?

Key Points

• git clone copies a remote repository to create a local repository with a remote called origin automatically set up.

Conflicts

Overview

Teaching: 15 min
Exercises: 0 min

Questions

• What do I do when my changes conflict with someone else’s?

Objectives

• Explain what conflicts are and when they can occur.

• Resolve conflicts resulting from a merge.

As soon as people can work in parallel, they’ll likely step on each other’s toes. This will even happen with a single person: if we are working on a piece of software on both our laptop and a server in the lab, we could make different changes to each copy. Version control helps us manage these conflicts by giving us tools to resolve overlapping changes.

To see how we can resolve conflicts, we must first create one. The file mars.txt currently looks like this in both partners’ copies of our planets repository:

$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity

Let’s add a line to the collaborator’s copy only:

$ nano mars.txt
$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
This line added to Wolfman's copy

and then push the change to GitHub:

$ git add mars.txt
$ git commit -m "Add a line in our home copy"

[main 5ae9631] Add a line in our home copy
 1 file changed, 1 insertion(+)

$ git push origin main

Enumerating objects: 5, done.
Counting objects: 100% (5/5), done.
Delta compression using up to 8 threads
Compressing objects: 100% (3/3), done.
Writing objects: 100% (3/3), 331 bytes | 331.00 KiB/s, done.
Total 3 (delta 2), reused 0 (delta 0)
remote: Resolving deltas: 100% (2/2), completed with 2 local objects.
To https://github.com/vlad/planets.git
   29aba7c..dabb4c8  main -> main

Now let’s have the owner make a different change to their copy without updating from GitHub:

$ nano mars.txt
$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
We added a different line in the other copy

We can commit the change locally:

$ git add mars.txt
$ git commit -m "Add a line in my copy"

[main 07ebc69] Add a line in my copy
 1 file changed, 1 insertion(+)

but Git won’t let us push it to GitHub:

$ git push origin main

To https://github.com/vlad/planets.git
 ! [rejected]        main -> main (fetch first)
error: failed to push some refs to 'https://github.com/vlad/planets.git'
hint: Updates were rejected because the remote contains work that you do
hint: not have locally. This is usually caused by another repository pushing
hint: to the same ref. You may want to first integrate the remote changes
hint: (e.g., 'git pull ...') before pushing again.
hint: See the 'Note about fast-forwards' in 'git push --help' for details.

Git rejects the push because it detects that the remote repository has new updates that have not been incorporated into the local branch. What we have to do is pull the changes from GitHub, merge them into the copy we’re currently working in, and then push that. Let’s start by pulling:

$ git pull origin main

remote: Enumerating objects: 5, done.
remote: Counting objects: 100% (5/5), done.
remote: Compressing objects: 100% (1/1), done.
remote: Total 3 (delta 2), reused 3 (delta 2), pack-reused 0
Unpacking objects: 100% (3/3), done.
From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
   2d8a740..3eb9f7c  main       -> origin/main
hint: You have divergent branches and need to specify how to reconcile them.
hint: You can do so by running one of the following commands sometime before
hint: your next pull:
hint: 
hint:   git config pull.rebase false  # merge
hint:   git config pull.rebase true   # rebase
hint:   git config pull.ff only       # fast-forward only
hint: 
hint: You can replace "git config" with "git config --global" to set a default
hint: preference for all repositories. You can also pass --rebase, --no-rebase,
hint: or --ff-only on the command line to override the configured default per
hint: invocation.
fatal: Need to specify how to reconcile divergent branches.

If this is the first time you run this command on the machine you are using, you might be facing this error message. As the error message suggests, it can be fixed by using

git config --global pull.rebase false

Once this issue in our configuration is fixed, we can try again to pull:

$ git pull origin main

remote: Enumerating objects: 5, done.
remote: Counting objects: 100% (5/5), done.
remote: Compressing objects: 100% (1/1), done.
remote: Total 3 (delta 2), reused 3 (delta 2), pack-reused 0
Unpacking objects: 100% (3/3), done.
From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
   2d8a740..3eb9f7c  main       -> origin/main
Auto-merging mars.txt
CONFLICT (content): Merge conflict in mars.txt
Automatic merge failed; fix conflicts and then commit the result.

The git pull command updates the local repository to include those changes already included in the remote repository. After the changes from remote branch have been fetched, Git detects that changes made to the local copy overlap with those made to the remote repository, and therefore refuses to merge the two versions to stop us from trampling on our previous work. The conflict is marked in in the affected file:

$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
<<<<<<< HEAD
We added a different line in the other copy
=======
This line added to Wolfman's copy
>>>>>>> dabb4c8c450e8475aee9b14b4383acc99f42af1d

Our change is preceded by <<<<<<< HEAD. Git has then inserted ======= as a separator between the conflicting changes and marked the end of the content downloaded from GitHub with >>>>>>>. (The string of letters and digits after that marker identifies the commit we’ve just downloaded.)

It is now up to us to edit this file to remove these markers and reconcile the changes. We can do anything we want: keep the change made in the local repository, keep the change made in the remote repository, write something new to replace both, or get rid of the change entirely. Let’s replace both so that the file looks like this:

$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
We removed the conflict on this line

To finish merging, we add mars.txt to the changes being made by the merge and then commit:

$ git add mars.txt
$ git status

On branch main
All conflicts fixed but you are still merging.
  (use "git commit" to conclude merge)

Changes to be committed:

	modified:   mars.txt

$ git commit -m "Merge changes from GitHub"

[main 2abf2b1] Merge changes from GitHub

Now we can push our changes to GitHub:

$ git push origin main

Enumerating objects: 10, done.
Counting objects: 100% (10/10), done.
Delta compression using up to 8 threads
Compressing objects: 100% (6/6), done.
Writing objects: 100% (6/6), 645 bytes | 645.00 KiB/s, done.
Total 6 (delta 4), reused 0 (delta 0)
remote: Resolving deltas: 100% (4/4), completed with 2 local objects.
To https://github.com/vlad/planets.git
   dabb4c8..2abf2b1  main -> main

Git keeps track of what we’ve merged with what, so we don’t have to fix things by hand again when the collaborator who made the first change pulls again:

$ git pull origin main

remote: Enumerating objects: 10, done.
remote: Counting objects: 100% (10/10), done.
remote: Compressing objects: 100% (2/2), done.
remote: Total 6 (delta 4), reused 6 (delta 4), pack-reused 0
Unpacking objects: 100% (6/6), done.
From https://github.com/vlad/planets
 * branch            main     -> FETCH_HEAD
    dabb4c8..2abf2b1  main     -> origin/main
Updating dabb4c8..2abf2b1
Fast-forward
 mars.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

We get the merged file:

$ cat mars.txt

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
We removed the conflict on this line

We don’t need to merge again because Git knows someone has already done that.

Git’s ability to resolve conflicts is very useful, but conflict resolution costs time and effort, and can introduce errors if conflicts are not resolved correctly. If you find yourself resolving a lot of conflicts in a project, consider these technical approaches to reducing them:

• Pull from upstream more frequently, especially before starting new work
• Use topic branches to segregate work, merging to main when complete
• Make smaller more atomic commits
• Where logically appropriate, break large files into smaller ones so that it is less likely that two authors will alter the same file simultaneously

Conflicts can also be minimized with project management strategies:

• Clarify who is responsible for what areas with your collaborators
• Discuss what order tasks should be carried out in with your collaborators so that tasks expected to change the same lines won’t be worked on simultaneously
• If the conflicts are stylistic churn (e.g. tabs vs. spaces), establish a project convention that is governing and use code style tools (e.g. htmltidy, perltidy, rubocop, etc.) to enforce, if necessary

A more informative conflict display style

Git can display conflicts in a more informative way if needed, for example

Cold and dry, but everything is my favorite color
The two moons may be a problem for Wolfman
But the Mummy will appreciate the lack of humidity
<<<<<<< HEAD
We added a different line in the other copy
||||||| 2d8a740
=======
This line added to Wolfman's copy

In this case, an additional section is added that starts with ||||||| 2d8a740. This represents the state of the file before the two set of conflicting changes were made, i.e. at commit 2d8a740, and in this case it is empty because the line causing the conflict did not exist back then. This style can be set as default with

git config --global merge.conflictstyle diff3

Solving Conflicts that You Create

Clone the repository created by your instructor. Add a new file to it, and modify an existing file (your instructor will tell you which one). When asked by your instructor, pull her changes from the repository to create a conflict, then resolve it.

Conflicts on Non-textual files

What does Git do when there is a conflict in an image or some other non-textual file that is stored in version control?

Solution

Let’s try it. Suppose Dracula takes a picture of Martian surface and calls it mars.jpg.

If you do not have an image file of Mars available, you can create a dummy binary file like this:

$ head -c 1024 /dev/urandom > mars.jpg
$ ls -lh mars.jpg

-rw-r--r-- 1 vlad 57095 1.0K Mar  8 20:24 mars.jpg

ls shows us that this created a 1-kilobyte file. It is full of random bytes read from the special file, /dev/urandom.

Now, suppose Dracula adds mars.jpg to his repository:

$ git add mars.jpg
$ git commit -m "Add picture of Martian surface"

[main 8e4115c] Add picture of Martian surface
 1 file changed, 0 insertions(+), 0 deletions(-)
 create mode 100644 mars.jpg

Suppose that Wolfman has added a similar picture in the meantime. His is a picture of the Martian sky, but it is also called mars.jpg. When Dracula tries to push, he gets a familiar message:

$ git push origin main

To https://github.com/vlad/planets.git
 ! [rejected]        main -> main (fetch first)
error: failed to push some refs to 'https://github.com/vlad/planets.git'
hint: Updates were rejected because the remote contains work that you do
hint: not have locally. This is usually caused by another repository pushing
hint: to the same ref. You may want to first integrate the remote changes
hint: (e.g., 'git pull ...') before pushing again.
hint: See the 'Note about fast-forwards' in 'git push --help' for details.

We’ve learned that we must pull first and resolve any conflicts:

$ git pull origin main

When there is a conflict on an image or other binary file, git prints a message like this:

$ git pull origin main
remote: Counting objects: 3, done.
remote: Compressing objects: 100% (3/3), done.
remote: Total 3 (delta 0), reused 0 (delta 0)
Unpacking objects: 100% (3/3), done.
From https://github.com/vlad/planets.git
 * branch            main     -> FETCH_HEAD
   6a67967..439dc8c  main     -> origin/main
warning: Cannot merge binary files: mars.jpg (HEAD vs. 439dc8c08869c342438f6dc4a2b615b05b93c76e)
Auto-merging mars.jpg
CONFLICT (add/add): Merge conflict in mars.jpg
Automatic merge failed; fix conflicts and then commit the result.

The conflict message here is mostly the same as it was for mars.txt, but there is one key additional line:

warning: Cannot merge binary files: mars.jpg (HEAD vs. 439dc8c08869c342438f6dc4a2b615b05b93c76e)

Git cannot automatically insert conflict markers into an image as it does for text files. So, instead of editing the image file, we must check out the version we want to keep. Then we can add and commit this version.

On the key line above, Git has conveniently given us commit identifiers for the two versions of mars.jpg. Our version is HEAD, and Wolfman’s version is 439dc8c0.... If we want to use our version, we can use git checkout:

$ git checkout HEAD mars.jpg
$ git add mars.jpg
$ git commit -m "Use image of surface instead of sky"

[main 21032c3] Use image of surface instead of sky

If instead we want to use Wolfman’s version, we can use git checkout with Wolfman’s commit identifier, 439dc8c0:

$ git checkout 439dc8c0 mars.jpg
$ git add mars.jpg
$ git commit -m "Use image of sky instead of surface"

[main da21b34] Use image of sky instead of surface

We can also keep both images. The catch is that we cannot keep them under the same name. But, we can check out each version in succession and rename it, then add the renamed versions. First, check out each image and rename it:

$ git checkout HEAD mars.jpg
$ git mv mars.jpg mars-surface.jpg
$ git checkout 439dc8c0 mars.jpg
$ mv mars.jpg mars-sky.jpg

Then, remove the old mars.jpg and add the two new files:

$ git rm mars.jpg
$ git add mars-surface.jpg
$ git add mars-sky.jpg
$ git commit -m "Use two images: surface and sky"

[main 94ae08c] Use two images: surface and sky
 2 files changed, 0 insertions(+), 0 deletions(-)
 create mode 100644 mars-sky.jpg
 rename mars.jpg => mars-surface.jpg (100%)

Now both images of Mars are checked into the repository, and mars.jpg no longer exists.

A Typical Work Session

You sit down at your computer to work on a shared project that is tracked in a remote Git repository. During your work session, you take the following actions, but not in this order:

• Make changes by appending the number 100 to a text file numbers.txt
• Update remote repository to match the local repository
• Celebrate your success with some fancy beverage(s)
• Update local repository to match the remote repository
• Stage changes to be committed
• Commit changes to the local repository

In what order should you perform these actions to minimize the chances of conflicts? Put the commands above in order in the action column of the table below. When you have the order right, see if you can write the corresponding commands in the command column. A few steps are populated to get you started.

order	action . . . . . . . . . .	command . . . . . . . . . .
1
2		echo 100 >> numbers.txt
3
4
5
6	Celebrate!	AFK

Solution

order	action . . . . . .	command . . . . . . . . . . . . . . . . . . .
1	Update local	git pull origin main
2	Make changes	echo 100 >> numbers.txt
3	Stage changes	git add numbers.txt
4	Commit changes	git commit -m "Add 100 to numbers.txt"
5	Update remote	git push origin main
6	Celebrate!	AFK

Key Points

• Conflicts occur when two or more people change the same lines of the same file.

• The version control system does not allow people to overwrite each other’s changes blindly, but highlights conflicts so that they can be resolved.

Branching

Overview

Teaching: 25 min
Exercises: 15 min

Questions

• How can I keep track of different lines of work efficiently?

Objectives

• Be able to create and merge branches.

• Know the difference between a branch and a tag.

So far, every time we have made a change to our repository and commited it, we always built on top of the work we had done previously. When collaborating with other people, though, we have worked independently, starting from a common commit, and then we have merged our contributions together by pulling from a remote repository into a local repository.

We can look at the graph of commits in our repository with the command

git log  --oneline --decorate --graph --all

*   a53b0af (HEAD -> main, origin/main, origin/HEAD) Merge changes from GitHub
|\  
| * 3eb9f7c Add a line in our home copy
* | 5f992dd Add a line in my copy
|/  
* 2d8a740 Add notes about Pluto
* 54765f6 Ignore data files and the result folder.
* 47122ce Add some initial thoughts on spaceships
* 66ca75a Discuss concerns about Mars' climate for Mummy
* 0aeaae3 Add concerns about effects of Mars' moons on Wolfman
* 118561d Start notes on Mars as a base

We notice that the commits do not sit on a single line, but there is a branching point and a merge.

Saving keystrokes and avoiding mistakes: git aliases

The command above

git log  --oneline --decorate --graph --all

tends to be very useful, but it takes very long to type, and making a mistake while typing it is very likely. For this reason, git allows aliases to be configured, with, for example

git config --global alias.graph "log --oneline --decorate --graph --all"

After configuring the alias in this way, we can use our custom-defined command git graph in place of the long one, with the same effect.

All this time, though, we have never explicitly created any branch nor explicitly merged one, and have kept working on the main branch.

This might be perfectly fine for simple development workflows, but for software development tasks we might benefit from a more sophisticated approach.

In particular:

• We need to keep reference to a version (commit) of the code that we trust (e.g., that works as expected, give expected results)
• We want to work on different features, and the develoment of each feature might take several commits
• We want to be able to make experimental changes in our code without breaking the working version or the work on other features

Git facilitates these kind of workflows with a powerful branching mechanism, that allows the use to isolate different tracks of work which can later be merged together. Each repository can have a number of branches, which can be matched to branches in other repositories (when pushing and pulling).

To view and manage branches we can use the git branch command:

$ git branch

* main 

By default, it prints out a list of branches available in the repository we are in, marking with a star the branch we are working on at the moment.

What is a branch?

When we look at a commit graph, we identify a branch as (from the Oxford Learner’s Dictionaries):

A part of a tree that grows out from the main stem […].

We might also improperly refer to a branch as a group of commits that creates a single narrative. But in the context of Git, a branch is just a human-readable name that points to a commit - keeping to the botanic analogy, a leaf usually at the tip of the branch. If we make more commits on a branch, the pointed commit will be updated.

In particular, this means that deleting a branch does not mean deleting any work (although, commits that cannot be reached from any branch might be automatically deleted at some point).

Creating branches and working with them

At some point, Dracula has the idea to get a gravity assist from Venus to get to Mars faster. It is not at all clear if this would work or not, so he decides to save all the work based on this idea in another branch.

Using the git branch command with some arguments

git branch gravity-assist main

we create another branch called gravity-assist that will “graft” onto the last commit on the main branch. We can verify that we created a new branch with the git branch command again, without arguments this time:

git branch

  gravity-assist
* main

We notice that we are still on the main branch. To switch to the new branch, we use git checkout:

git checkout gravity-assist

Switched to branch 'gravity-assist'

Optionally, we can verify again the branch we are on with git branch. Whatever work we do now is not going to affect the main branch, and we are free to experiment as much as we like. We add a file called venus.txt with the following content:

The orbital period of Venus is 224 days.

We can now add our new file to the staging area and commit:

git add venus.txt
git commit -m "Start notes about Venus"

It turns out we did not properly round the orbital period of Venus, which is actually 224.7 days. Let us correct that, and make another commit

The orbital period of Venus is 225 days.

git add venus.txt
git commit -m "Fix orbital period of Venus"

We want to publish the branch we are working on to GitHub, so we can discuss the work and share it with collaborators (and also feel safer in case anything bad happens to our laptop):

git push origin gravity-assist

We can check on GitHub that a new branch was created. While we wait for feedback, we decide to continue our work on the main branch:

git checkout main

We make sure that we are on the main branch again with git branch:

git branch

  gravity-assist
* main

We keep working on mars.txt, adding the following line:

Landing on Mars (successfully) has historically proven difficult.

Then we add the changes and commit them:

git add mars.txt
git commit -m "Add notes on difficulty of the endeavour"

We can check our commit graph again with the git log command (using the arrow up key to get it from the bash history):

git log  --oneline --decorate --graph --all

* 0746194 (HEAD -> main) Add notes on difficulty of the endeavour
| * 8da5718 (gravity-assist) Fix orbital period of Venus
| * ed6944a Start notes about Venus
|/  
*   a53b0af (tag: 09-conflict-resolution, origin/main, origin/HEAD) Merge changes from GitHub
|\  
| * 3eb9f7c (tag: 09-conflict-principal) Add a line in our home copy
* | 5f992dd (tag: 09-conflict-collaborator) Add a line in my copy
|/  
* 2d8a740 (tag: 08-collab) Add notes about Pluto
* 54765f6 (tag: 06-ignore) Ignore data files and the result folder.
* 47122ce (tag: exercises-04-changes) Add some initial thoughts on spaceships
* 66ca75a (tag: 04-changes) Discuss concerns about Mars' climate for Mummy
* 0aeaae3 Add concerns about effects of Mars' moons on Wolfman
* 118561d Start notes on Mars as a base

Merging branches

After discussing with our collaborators, we decide that the gravity assist is worth pursuing, and we want to merge the work on the gravity-assist branch into main. We make sure once again that we are on the main branch:

git branch

  gravity-assist
* main

We are then ready to merge gravity-assist into main:

git merge gravity-assist

This will open a text editor for a commit message. We can leave the default message as it is, save and exit the editor. We can check again the commit graph:

git log  --oneline --decorate --graph --all

*   691f9d2 (HEAD -> main) Merge branch 'gravity-assist'
|\  
| * 8da5718 (gravity-assist) Fix orbital period of Venus
| * ed6944a Start notes about Venus
* | 0746194 Add notes on difficulty of the endeavour
|/  
*   a53b0af (tag: 09-conflict-resolution, origin/main, origin/HEAD) Merge changes from GitHub
|\  
| * 3eb9f7c (tag: 09-conflict-principal) Add a line in our home copy
* | 5f992dd (tag: 09-conflict-collaborator) Add a line in my copy
|/  
* 2d8a740 (tag: 08-collab) Add notes about Pluto
* 54765f6 (tag: 06-ignore) Ignore data files and the result folder.
* 47122ce (tag: exercises-04-changes) Add some initial thoughts on spaceships
* 66ca75a (tag: 04-changes) Discuss concerns about Mars' climate for Mummy
* 0aeaae3 Add concerns about effects of Mars' moons on Wolfman
* 118561d Start notes on Mars as a base

We can then push our changes to main to the remote repository:

git push origin main

Deleting branches safely

After a branch has been merged it is good practice to delete it. We can delete gravity-assist with the git branch -d command:

git branch -d gravity-assist 

Deleted branch gravity-assist (was 8da5718).

We can check the commit graph again:

git log  --oneline --decorate --graph --all

*   691f9d2 (HEAD -> main) Merge branch 'gravity-assist'
|\  
| * 8da5718 Fix orbital period of Venus
| * ed6944a Start notes about Venus
* | 0746194 Add notes on difficulty of the endeavour
|/  
*   a53b0af (tag: 09-conflict-resolution, origin/main, origin/HEAD) Merge changes from GitHub
|\  
| * 3eb9f7c (tag: 09-conflict-principal) Add a line in our home copy
* | 5f992dd (tag: 09-conflict-collaborator) Add a line in my copy
|/  
* 2d8a740 (tag: 08-collab) Add notes about Pluto
* 54765f6 (tag: 06-ignore) Ignore data files and the result folder.
* 47122ce (tag: exercises-04-changes) Add some initial thoughts on spaceships
* 66ca75a (tag: 04-changes) Discuss concerns about Mars' climate for Mummy
* 0aeaae3 Add concerns about effects of Mars' moons on Wolfman
* 118561d Start notes on Mars as a base

We see that the only the “pointer” to the gravity-assist branch is gone, but not the commits, which have been already safely merged into main.

Git will not let you delete a branch which has not been merged in an other existing branch, unless you insist by using git branch -D. Even then, the commits are not lost, but they might be very hard to find as there is no branch pointing to them.

Tags

Tags are similar to branches, in the sense that they are a “pointer” to a commit, but contrarily to branches they do not change over time. They are a mechanism to elect some commits to a more important status.

Other commands to create branches

Creating a branch can be also done with git checkout and with git switch, using the right options. Can you find the right options from the man page?

Hint: use git help checkout and git help switch, and type /create to search for the word “create”.

Solution

The command

git checkout -b new-branch

does create a new branch and checks it out, it is a shortcut for git branch new-branch ; git checkout new-branch. The command

git switch -c new-branch

does the same. The git switch command is possibly easier to remember, but it might not be available in older versions of Git.

Delete remote branches

Using the planets repository as a test playground, create a branch called test-branch, and push it to origin. Verify that the branch is there. To delete it, use

git push origin -d test-branch

Verify that the branch is not there anymore.

Using Tags

Using the planets repository as a test playground, create a tag using the command

git tag -a v0.1 -m "A test tag" 

You can verify that a new tag was created wit the git tag command (without arguments). Check on the web page of your remote repository on GitHub (or wherever origin is): are any tags listed? (The number of tags is usually shown at the top of the page)

Push the tag to the remote repository with

git push origin v0.1

Can you see it on the remote repository now?

Deleting Local Tags

Deleting branches is done with git branch -d. Can you figure out how to delete a tag?

Solution

To delete a tag locally:

git tag -d <tag-name> 

Deleting Remote Tags

Create a tag called v0.2 with message “test tag please ignore”, and push it to the remote repository.
To delete a branch named test-branch on the remote repository called origin One has to do git push origin -d test-branch.
Can you figure out how to delete a tag on a remote repository?

Solution

The web interface on GitHub or GitLab provide the options to to delete a tag in the menus. Alternatively, one can use exactly the same command as for branches: ~~~ git push origin -d test-tag

Hard: recover deleted branches

1. Create a new branch, add some work to it, and commit it. Do not merge it with any other branch.
2. Force-delete it with git branch -D.
3. How can you retrieve the lost work, by looking at the terminal?
4. Can the git reflog command be useful?

Solution

After you saved the last commit message. you should have got a commit hash in the output message. You can re-create a branch named rescue-branch pointing to that commit with

git branch rescue-branch-name <commit-id-recovered-from-terminal>

A more robust way to recover commit hashes is the command git reflog, which shows the history of the movements of HEAD. Nothing is really lost when using Git. Even if you delete branches!

Copyright notice

This episode is based on/inspired by the episode on branching in the course Introduction to Version Control with Git developed by CodeRefinery

Key Points

• use git branch to view, create and delete branches.

• use git tag to view, create and delete tags.

Open Science

Overview

Teaching: 5 min
Exercises: 5 min

Questions

• How can version control help me make my work more open?

Objectives

• Explain how a version control system can be leveraged as an electronic lab notebook for computational work.

The opposite of “open” isn’t “closed”. The opposite of “open” is “broken”.

— John Wilbanks

Free sharing of information might be the ideal in science, but the reality is often more complicated. Normal practice today looks something like this:

• A scientist collects some data and stores it on a machine that is occasionally backed up by her department.
• She then writes or modifies a few small programs (which also reside on her machine) to analyze that data.
• Once she has some results, she writes them up and submits her paper. She might include her data – a growing number of journals require this – but she probably doesn’t include her code.
• Time passes.
• The journal sends her reviews written anonymously by a handful of other people in her field. She revises her paper to satisfy them, during which time she might also modify the scripts she wrote earlier, and resubmits.
• More time passes.
• The paper is eventually published. It might include a link to an online copy of her data, but the paper itself will be behind a paywall: only people who have personal or institutional access will be able to read it.

For a growing number of scientists, though, the process looks like this:

• The data that the scientist collects is stored in an open access repository like figshare or Zenodo, possibly as soon as it’s collected, and given its own Digital Object Identifier (DOI). Or the data was already published and is stored in Dryad.
• The scientist creates a new repository on GitHub to hold her work.
• As she does her analysis, she pushes changes to her scripts (and possibly some output files) to that repository. She also uses the repository for her paper; that repository is then the hub for collaboration with her colleagues.
• When she’s happy with the state of her paper, she posts a version to arXiv or some other preprint server to invite feedback from peers.
• Based on that feedback, she may post several revisions before finally submitting her paper to a journal.
• The published paper includes links to her preprint and to her code and data repositories, which makes it much easier for other scientists to use her work as starting point for their own research.

This open model accelerates discovery: the more open work is, the more widely it is cited and re-used. However, people who want to work this way need to make some decisions about what exactly “open” means and how to do it. You can find more on the different aspects of Open Science in this book.

This is one of the (many) reasons we teach version control. When used diligently, it answers the “how” question by acting as a shareable electronic lab notebook for computational work:

• The conceptual stages of your work are documented, including who did what and when. Every step is stamped with an identifier (the commit ID) that is for most intents and purposes unique.
• You can tie documentation of rationale, ideas, and other intellectual work directly to the changes that spring from them.
• You can refer to what you used in your research to obtain your computational results in a way that is unique and recoverable.
• With a version control system such as Git, the entire history of the repository is easy to archive for perpetuity.

Making Code Citable

Anything that is hosted in a version control repository (data, code, papers, etc.) can be turned into a citable object. You’ll learn how to do this in the later episode on Citation.

How Reproducible Is My Work?

Ask one of your labmates to reproduce a result you recently obtained using only what they can find in your papers or on the web. Try to do the same for one of their results, then try to do it for a result from a lab you work with.

How to Find an Appropriate Data Repository?

Surf the internet for a couple of minutes and check out the data repositories mentioned above: Figshare, Zenodo, Dryad. Depending on your field of research, you might find community-recognized repositories that are well-known in your field. You might also find useful these data repositories recommended by Nature. Discuss with your neighbor which data repository you might want to approach for your current project and explain why.

How to Track Large Data or Image Files using Git?

Large data or image files such as .md5 or .psd file types can be tracked within a github repository using the Git Large File Storage open source extension tool. This tool automatically uploads large file contents to a remote server and replaces the file with a text pointer within the github repository.

Try downloading and installing the Git Large File Storage extension tool, then add tracking of a large file to your github repository. Ask a colleague to clone your repository and describe what they see when they access that large file.

Key Points

• Open scientific work is more useful and more highly cited than closed.

Licensing

Overview

Teaching: 5 min
Exercises: 0 min

Questions

• What licensing information should I include with my work?

Objectives

• Explain why adding licensing information to a repository is important.

• Choose a proper license.

• Explain differences in licensing and social expectations.

When a repository with source code, a manuscript or other creative works becomes public, it should include a file LICENSE or LICENSE.txt in the base directory of the repository that clearly states under which license the content is being made available. This is because creative works are automatically eligible for intellectual property (and thus copyright) protection. Reusing creative works without a license is dangerous, because the copyright holders could sue you for copyright infringement.

A license solves this problem by granting rights to others (the licensees) that they would otherwise not have. What rights are being granted under which conditions differs, often only slightly, from one license to another. In practice, a few licenses are by far the most popular, and choosealicense.com will help you find a common license that suits your needs. Important considerations include:

• Whether you want to address patent rights.
• Whether you require people distributing derivative works to also distribute their source code.
• Whether the content you are licensing is source code.
• Whether you want to license the code at all.

Choosing a license that is in common use makes life easier for contributors and users, because they are more likely to already be familiar with the license and don’t have to wade through a bunch of jargon to decide if they’re ok with it. The Open Source Initiative and Free Software Foundation both maintain lists of licenses which are good choices.

This article provides an excellent overview of licensing and licensing options from the perspective of scientists who also write code.

At the end of the day what matters is that there is a clear statement as to what the license is. Also, the license is best chosen from the get-go, even if for a repository that is not public. Pushing off the decision only makes it more complicated later, because each time a new collaborator starts contributing, they, too, hold copyright and will thus need to be asked for approval once a license is chosen.

Can I Use Open License?

Find out whether you are allowed to apply an open license to your software. Can you do this unilaterally, or do you need permission from someone in your institution? If so, who?

What licenses have I already accepted?

Many of the software tools we use on a daily basis (including in this workshop) are released as open-source software. Pick a project on GitHub from the list below, or one of your own choosing. Find its license (usually in a file called LICENSE or COPYING) and talk about how it restricts your use of the software. Is it one of the licenses discussed in this session? How is it different?

• Git, the source-code management tool
• CPython, the standard implementation of the Python language
• Jupyter, the project behind the web-based Python notebooks we’ll be using
• EtherPad, a real-time collaborative editor

Key Points

• The LICENSE, LICENSE.md, or LICENSE.txt file is often used in a repository to indicate how the contents of the repo may be used by others.

• People who incorporate General Public License (GPL’d) software into their own software must make their software also open under the GPL license; most other open licenses do not require this.

• The Creative Commons family of licenses allow people to mix and match requirements and restrictions on attribution, creation of derivative works, further sharing, and commercialization.

• People who are not lawyers should not try to write licenses from scratch.

Citation

Overview

Teaching: 2 min
Exercises: 0 min

Questions

• How can I make my work easier to cite?

Objectives

• Make your work easy to cite

You may want to include a file called CITATION or CITATION.txt that describes how to reference your project; the one for Software Carpentry states:

To reference Software Carpentry in publications, please cite both of the following:

Greg Wilson: "Software Carpentry: Getting Scientists to Write Better
Code by Making Them More Productive".  Computing in Science &
Engineering, Nov-Dec 2006.

Greg Wilson: "Software Carpentry: Lessons Learned". arXiv:1307.5448,
July 2013.

@article{wilson-software-carpentry-2006,
    author =  {Greg Wilson},
    title =   {Software Carpentry: Getting Scientists to Write Better Code by Making Them More Productive},
    journal = {Computing in Science \& Engineering},
    month =   {November--December},
    year =    {2006},
}

@online{wilson-software-carpentry-2013,
  author      = {Greg Wilson},
  title       = {Software Carpentry: Lessons Learned},
  version     = {1},
  date        = {2013-07-20},
  eprinttype  = {arxiv},
  eprint      = {1307.5448}
}

More detailed advice, and other ways to make your code citable can be found at the Software Sustainability Institute blog and in:

Smith AM, Katz DS, Niemeyer KE, FORCE11 Software Citation Working Group. (2016) Software citation
principles. [PeerJ Computer Science 2:e86](https://peerj.com/articles/cs-86/)
https://doi.org/10.7717/peerj-cs.8

There is also an @software{... BibTeX entry type in case no “umbrella” citation like a paper or book exists for the project you want to make citable.

Key Points

• Add a CITATION file to a repository to explain how you want your work cited.

Hosting

Overview

Teaching: 10 min
Exercises: 0 min

Questions

• Where should I host my version control repositories?

Objectives

• Explain different options for hosting scientific work.

The second big question for groups that want to open up their work is where to host their code and data. One option is for the lab, the department, or the university to provide a server, manage accounts and backups, and so on. The main benefit of this is that it clarifies who owns what, which is particularly important if any of the material is sensitive (i.e., relates to experiments involving human subjects or may be used in a patent application). The main drawbacks are the cost of providing the service and its longevity: a scientist who has spent ten years collecting data would like to be sure that data will still be available ten years from now, but that’s well beyond the lifespan of most of the grants that fund academic infrastructure.

Another option is to purchase a domain and pay an Internet service provider (ISP) to host it. This gives the individual or group more control, and sidesteps problems that can arise when moving from one institution to another, but requires more time and effort to set up than either the option above or the option below.

The third option is to use a public hosting service like GitHub, GitLab,or BitBucket. Each of these services provides a web interface that enables people to create, view, and edit their code repositories. These services also provide communication and project management tools including issue tracking, wiki pages, email notifications, and code reviews. These services benefit from economies of scale and network effects: it’s easier to run one large service well than to run many smaller services to the same standard. It’s also easier for people to collaborate. Using a popular service can help connect your project with communities already using the same service.

As an example, Software Carpentry is on GitHub where you can find the source for this page. Anyone with a GitHub account can suggest changes to this text.

GitHub repositories can also be assigned DOIs, by connecting its releases to Zenodo. For example, 10.5281/zenodo.57467 is the DOI that has been “minted” for this introduction to Git.

Using large, well-established services can also help you quickly take advantage of powerful tools. One such tool, continuous integration (CI), can automatically run software builds and tests whenever code is committed or pull requests are submitted. Direct integration of CI with an online hosting service means this information is present in any pull request, and helps maintain code integrity and quality standards. While CI is still available in self-hosted situations, there is much less setup and maintenance involved with using an online service. Furthermore, such tools are often provided free of charge to open source projects, and are also available for private repositories for a fee.

Institutional Barriers

Sharing is the ideal for science, but many institutions place restrictions on sharing, for example to protect potentially patentable intellectual property. If you encounter such restrictions, it can be productive to inquire about the underlying motivations and either to request an exception for a specific project or domain, or to push more broadly for institutional reform to support more open science.

Can My Work Be Public?

Find out whether you are allowed to host your work openly in a public repository. Can you do this unilaterally, or do you need permission from someone in your institution? If so, who?

Where Can I Share My Work?

Does your institution have a repository or repositories that you can use to share your papers, data and software? How do institutional repositories differ from services like arXiV, figshare, GitHub or GitLab?

Key Points

• Projects can be hosted on university servers, on personal domains, or on a public hosting service.

• Rules regarding intellectual property and storage of sensitive information apply no matter where code and data are hosted.