JupyterCheckpoints

A JupyterLab extension for managing multiple checkpoints, allowing users to save and restore multiple historical versions of files.

Features

• Maintains up to 5 checkpoints for each file (configurable)
• Restore to any checkpoint directly from the JupyterLab interface
• Supports notebooks (.ipynb) and regular text files
• Provides a user-friendly interface showing checkpoint creation time and relative time (e.g., "5 minutes ago")
• Automatically manages checkpoints, removing older versions

Installation

Using pip

pip install jupyter-checkpoints

From Source

1. Clone the repository

git clone https://github.com/yourusername/jupyter-checkpoints.git
cd jupyter-checkpoints

2. Install the Python package

pip install -e .

3. Install the frontend extension

jupyter labextension develop --overwrite .

Usage

After installation, JupyterCheckpoints will be automatically activated. A "Restore Checkpoint" button will be added to the toolbar of each open notebook and file editor.

Restoring to a Checkpoint

1. Open any file or notebook
2. Click the "Restore Checkpoint" button (undo icon) in the toolbar
3. Select the checkpoint to restore from the dropdown menu
4. Confirm the restore operation

Configuration

You can set the maximum number of checkpoints to keep for each file in your Jupyter configuration file:

c.AsyncMultiCheckpoints.max_checkpoints = 10  # Default is 5

Development

Prerequisites

• JupyterLab >= 4.0.0
• Python >= 3.12
• Node.js

Setting Up Development Environment

1. Clone the repository

git clone https://github.com/yourusername/jupyter-checkpoints.git
cd jupyter-checkpoints

2. Create and activate a virtual environment

python -m venv .venv
source .venv/bin/activate  # Linux/Mac
# or
.venv\Scripts\activate  # Windows

3. Install development dependencies

pip install -e .

4. Install JavaScript dependencies and build

cd js
npm install
npm run build

5. Install the extension in development mode

jupyter labextension develop --overwrite .

Development Workflow

• After modifying Python code, restart the Jupyter server
• After modifying JavaScript code:

  cd js
  npm run build
  

  Then refresh the JupyterLab page

License

MIT