Demystifying Git: From Init to Commit, Branches, and Remote Repos

Git is a distributed version‑control system whose design differs fundamentally from other VCSs; understanding its concepts—initialization, object storage, and workflow—makes using it intuitive.

Initializing a Repository

Running git init creates a .git directory containing HEAD, config, description, hooks, info, objects, and refs. The config file holds local settings (e.g., user.name, user.email).

# left side
mkdir git-demo
cd git-demo && git init
rm -rf .git/hooks/*.sample
# right side
watch -n 1 -d find .

Git Objects and SHA‑1

When git add is executed, Git creates a blob object that stores the file’s content and type, not its name. The object’s identifier is the SHA‑1 hash of blob <size>\0<content>. Identical file contents share the same blob.

# view object type and content
git cat-file -t 8d0e41
blob
git cat-file -p 8d0e41
hello git
git cat-file -s 8d0e41
10

Working Directory, Index, and Commit

The working directory holds actual files. git add copies file contents into the index (a binary .git/index file) and creates blob objects. git status compares the index with the working directory to show changes.

When git commit runs, Git writes a tree object that records the snapshot of the index (file names and blob hashes) and a commit object that references the tree, parent commit(s), author, and message. The HEAD file points to the current branch’s latest commit.

# commit example
git commit -m "1st commit"
git cat-file -t 6e4a700   # commit
git cat-file -p 6e4a700   # shows tree and metadata

Branches and HEAD

A branch is a named pointer to a commit (e.g., refs/heads/master). HEAD always points to the current branch; switching branches updates HEAD. Deleting a branch removes only the pointer, not the underlying commits.

# create and switch branch
git branch dev
git checkout dev
cat .git/HEAD   # now points to refs/heads/dev

Remote Repositories

Link a local repo to a remote with git remote add origin <url>. The remote’s configuration appears in .git/config. Pushing with git push -u origin master transfers new objects, creates remote refs (e.g., refs/remotes/origin/master), and updates the remote’s commit history.

# add remote and push
git remote add origin [email protected]:escapelife/git-demo.git
git push -u origin master

Recovering Lost Commits

Even after deleting a branch, its commits remain reachable via git reflog, which records all reference updates.

# find lost commit
git reflog
git checkout <commit-hash>

Summary of File Lifecycle

Files move from the working directory → index (staged) → commit (snapshot). Git tracks changes using blobs (content), trees (directory structure), and commits (history), while branches and HEAD manage which snapshot is active.