Git Series (2) — Command-Line Cheatsheet

In the last post, we discussed git internals to understand Git. In this post we will discuss Git commands, to use Git. I recommend reading the last post, to easily understand the terminologies used here.

I’m going to skip the commands to install git as it varies based on the operating system and y. Once you have our machine architecture. After installation following are the commonly used commands.

1. Check Git Version

git —-version check and confirm the installation

2. Configure Git

You can use the below command to set global configurations for Git.

git config --global key “value”.

By default, Git doesn’t allow you to commit until it knows who you are. So the first thing to do after installation is, to set the below properties. Preferably globally

git config --global user.name “My Name”
git config --global user.email “myemail@email.com”

NOTE. If your present working directory is inside a local git repository, you can omit the —-global option from the above command and set up a project-specific configuration.

3. Create a local repo

To create a git repo on your machine, you can either “clone” an already existing remote repository or you can create a fresh local repository.

git init — Converts the current folder into a local git repo.
git clone git@github:user1/repo1.git — Brings a repo from the remote server to the local machine.

4. Work with branches

The branch is just a pointer to a commit, usually at the head of the tree. Any commit you make is made on the branch you’re currently “checked out” to.

Here are a few commands to work with branches

git status — Show which branch you are on, (and some more info)
git branch— Show local branches
git branch -a — Show both local and remote branches

git branch branch-name — create a local branch
git checkout -b branch-name — create a new branch and switch to it.

git checkout branch-name —Switch to an existing branch
git switch branch-name —similar (not same) as above

git branch -d branch-name — delete a local branch
git push origin --delete branch-name — delete a remote branch. Might fail in the absence of required privileges.

git merge source-branch-name — merge the commits from the source branch into the current branch.

4: Add new content

Following are the commands to manage and add content to the repository

git status
Display the status of the working tree, staging area, and untracked files.

git add <pathspec>
Move the files matching the given pathspec, from the working tree to Staging Area.

git commit -m "<commit message>”
Move the code from the staging area to the repo-tree

Some examples for git add are

git add .            # all content of the current directory
git add ui/          # content from ui/ directory 
git add MyScript.js  # Only MyScript.js file

NOTE: pathspec is a lot more powerful than the above 3 simple examples. For the sake of the size of this post, I’ll cover it in detail in a separate post.

5. Remove / Undo content

git rm <pathspec>
Remove the file from the staging area & working tree.

git revert <commit-hash>
It doesn’t actually remove anything. When applied on a clean working tree, The revert creates a new commit in the repo tree. This commit has the code to invert changes from the given commit-hash

git commit --amend
Adds the currently staged changes to the previous commit

git reset --<option> <pointer>

common option values - soft | hard | mixed(default)
pointer values - commit-hash | HEAD

Commands update all 3 trees as follows

• Repo tree- The HEAD pointer of the current branch is moved to the given commit-hash. No change If there is no commit hash.
• staging area- effects of reset depend upon the used option
  hard — the changes are lost.
  mixed — the current content is moved to the working tree.
  soft — no change.
• working tree- effects of reset depends upon the used option
  hard — the changes are lost.
  mixed— the changes from staging are added back, and any file which was created after a given commit-hash stays in a tree but becomes untracked
  soft — no change.

5. Observe the content & repo

Following are the commands which do not affect data in any tree.
They are used only for observation or analysis

git status
Display the changes in the working tree, staging area, and untracked files.

git log
Show details from commit logs, there are many options to use this command the simplest is as above

git show <commit-hash>
Shows the changes done in a commit

git blame <file name>
Shows the author and time information for each line.

git diff
There are many variations for diff also. Here are some commonly used ones

• git diff — Difference between the working tree & repo-tree.
  for the file present in both working-tree & staged areas, shows the difference between working and staging
• git diff --cached — Difference between the staging area & repo-tree.
• git diff HEAD — Difference between (staging area+working-tree) & repo-tree. for the file present in both working-tree & staged areas, shows the difference between working and repo-tree
• git diff <commit-hash-1> <commit-hash-2> Difference between the two given commits

6. Working with remote

The repository usually stays on the local machine as well as remote servers. Here are a few commands to manage them

git remote -v — shows the remote repo associated with this local repo.
git remote add origin <URL> — set a remote repo for fetch & push

git fetch — download objects and refs from the remote repo. Does not merge them in the local repo tree
git pull — run git fetch followed by a git merge from origin to local branch.

git push — push object and refs from local to the remote repo.
git push origin HEAD push the current HEAD, which is the currently checked out branch, to the origin, which is the remote/ref.
git push -u origin HEAD This is good (but not limited) to newly created local-only branches. With the -u option git set the tracking information for the current local branch.

Before I explain the last command in the “working with remote” section, let's look in detail at some commands we already talked about above.

git checkout <remote-branch-name>
When we checkout first time, a local branch with the name same as remote-branch-name is created and its upstream is set to origin/remote-branch-name.

git push -u origin HEAD
When we push with -u, a remote branch with a name same as local-branch-name is created and upstream is set as origin/local-branch-name

In all other cases, our current branch does not have tracking information, which can be set with the below command

git branch --set-upstream-to=origin/<remote-branch> <local-branch> —sets a remote counterpart to run fetch & push on a given local-branch.

7. Stash

Git stash doesn’t fit anywhere in our 3 working trees, it’s a handy tool provided by git to stash away the changes that can be used later.

git stash
Stash changes available in the working tree into a directory away from 3 trees.

git stash pop
Apply stashed changes to the working tree and drop the stash

git stash apply
Apply the stashed change to the working tree.

git stash list
Show all the stashes

git stash drop
Drop the last stash

git stash clear
Drop all stashes

git stash show
show the changes in the last taken stash

NOTE — Once you stash, changes get added to the stash data structure at the top. so stash@{0} always points to the last stash. You can run git stash apply, drop, and show commands withstash@{x} parameter. where x is the index starting from 0 for the latest stash.

I intentionally did not cover rebase because I’ve had many experiences of developers using it incorrectly with force push and causing trouble for everyone else while merging. I would suggest avoiding rebase until you know what you are doing.

This is all as far as a quick cookbook is concerned. I barely scratched the surface here. but whatever is discussed above, is more than enough for everyday use.
However, every command has a plethora of options, Let me know in the comments if you want me to cover any of the above commands in detail. Thanks for reading.