Internals of GIT [WIP]

Description

Git's internal workings is quite complex, involving a range of data structures and algorithms to manage the repository's history, branches, commits, and more. Here's a high-level overview of some key aspects of Git's internal workings.

Object Model

The object model in Git is a core component that underpins how Git stores and manages data. In Git, everything is stored as an object, and these objects are content-addressable, meaning they are referenced by a hash of their content. The primary types of objects in Git are blobs, trees, commits, and tags. Here's a detailed look at each of these objects and how they work together.

Types of Git Objects

1. Blob (Binary Large Object)

  • Purpose: Represents the content of a file.

  • Content: Stores the raw file data (i.e., the contents of a file).

  • Identification: Identified by a SHA-1 hash of the file content.

  • Metadata: Does not store any metadata such as the file name or permissions.

Example of creating a blob:

echo "Hello, Git!" > hello.txt
git add hello.txt
# Creates a blob for hello.txt

2. Tree

  • Purpose: Represents a directory and its contents.

  • Content: Contains references (hashes) to blobs and other trees, along with the associated metadata (file names, types, and modes).

  • Identification: Identified by a SHA-1 hash of its contents.

Example of creating a tree:

git write-tree

3. Commit

  • Purpose: Represents a snapshot of the repository at a point in time.

  • Content: Contains metadata (author, committer, message), a reference to a tree object (the state of the file system), and references to parent commits.

  • Identification: Identified by a SHA-1 hash of its contents.

Example of creating a commit:

git commit -m "Initial commit"

4. Tag

  • Purpose: Marks a specific commit as significant, often used to mark release points.

  • Content: Contains metadata (tagger, message) and a reference to a commit object.

  • Identification: Identified by a SHA-1 hash of its contents.

Example of creating an annotated tag:

git tag -a v1.0 -m "Version 1.0"

How Objects are Stored

  • Content-Addressable Storage:

    • Git uses SHA-1 hashes to uniquely identify each object. The SHA-1 hash is derived from the content of the object.

    • The content of each object is stored in the .git/objects directory in a subdirectory named after the first two characters of the hash, with the remaining characters as the filename.

  • Loose Objects and Packed Objects:

    • Loose Objects: Individual files stored in the .git/objects directory. These are typically created during initial commits.

    • Packed Objects: Over time, loose objects are packed into packfiles to save space and improve performance. Packfiles are stored in the .git/objects/pack directory.

Object Creation and Relationship

  • Blob Creation:

    • When we add a file to Git (git add), Git creates a blob object containing the file's content.

  • Tree Creation:

    • When you commit changes (git commit), Git creates a tree object representing the directory structure, containing references to blob objects for files and other tree objects for subdirectories.

  • Commit Creation:

    • A commit object is created referencing the tree object and containing metadata about the commit (author, message, parent commits).

  • Tag Creation:

    • A tag object is created to reference a specific commit, often annotated with additional information (tagger, message).

Object Model Visualization

Here's a simple visualization of how these objects might be linked together.

Inspecting Git Objects

-- List all objects
git rev-list --all --objects

-- View object type and size
git cat-file -t <object-hash>
git cat-file -s <object-hash>

-- View object content
git cat-file -p <object-hash>

Reference - https://towardsdatascience.com/how-git-truly-works-cd9c375966f6

PreviousGITNextCommands

Last updated