syft-objects 0.1.0

Distributed file discovery and addressing system with permission control

🔐 Syft Objects

Distributed file discovery and addressing with granular permission control

✨ What is Syft Objects?

Syft Objects is a distributed file discovery system that lets you share files with explicit mock vs private content control and granular permission management. Perfect for data science, research, and any scenario where you need to share different versions of files with different people.

from syft_objects import syobj

# Create an object with explicit mock vs private content
obj = syobj(
    name="Customer Analysis",
    mock_contents="Sample: 100 customers, avg age 42",
    private_contents="Full: 47,293 customers, avg age 41.7, LTV $1,247",
    discovery_read=["public"],           # Who knows it exists
    mock_read=["[e-mail]"],      # Who sees demo
    private_read=["[e-mail]"]   # Who sees real data
)

# Beautiful HTML display in Jupyter
obj

🚀 Key Features

🎯 Explicit Content Control

• Mock content: What everyone sees (demo version)
• Private content: What authorized users see (real data)
• No magic: You control exactly what goes where

🔒 Granular Permissions

• Discovery: Who knows the object exists
• Mock: Who can read/write demo content
• Private: Who can read/write real data

📱 Beautiful Jupyter Display

• Rich HTML widgets with permission badges
• File path display for all three file types
• Clean, organized layout
• No external dependencies

🌐 Distributed by Design

• Works across many locations (not centralized)
• SyftBox integration for proper distributed storage
• Lightweight for search and discovery

📦 Installation

pip install syft-objects

For SyftBox integration:

pip install syft-objects[syftbox]

🎯 Quick Start

Basic Usage

from syft_objects import syobj

# Simple content sharing
obj = syobj(
    name="AI Model Results",
    mock_contents="Model achieved good performance",
    private_contents="Accuracy: 94.5%, F1: 0.92, Cost: $127"
)

File-Based Sharing

# Use existing files
obj = syobj(
    name="Customer Database",
    mock_file="sample_100_rows.csv",      # Demo file
    private_file="full_50k_rows.csv",     # Real dataset
    mock_read=["[e-mail]"],
    private_read=["[e-mail]"]
)

Restricted Discovery

# Control who even knows it exists
obj = syobj(
    name="Confidential Report",
    mock_contents="Q4 summary available",
    private_contents="Revenue: $2.5M, detailed breakdown...",
    discovery_read=["[e-mail]"],  # Only director knows it exists
    mock_read=["[e-mail]"],        # Manager sees summary
    private_read=["[e-mail]"]          # CEO sees details
)

📚 Documentation

• Quick Start Guide - Get up and running in 5 minutes
• Tutorial Notebook - Interactive Jupyter tutorial
• API Reference - Complete function documentation
• SyftBox Integration - Distributed storage setup

🎨 Design Philosophy

Ultra-Clean API

• One function: syobj() - that's it
• Clear parameters: mock_contents vs private_contents
• Explicit permissions: discovery_read, mock_read, private_read

No Surprises

• You specify exactly what goes in mock vs private
• No auto-generation or magic content creation
• Full control over permissions and access

Jupyter-First

• Beautiful HTML widgets out of the box
• File paths displayed clearly
• Permission badges for easy understanding

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

📄 License

Apache License 2.0 - see LICENSE file for details.

🔗 Related Projects

• SyftBox - Distributed storage backend
• Syft - Privacy-preserving machine learning
• OpenMined - Privacy-preserving AI ecosystem

---

Built with ❤️ by OpenMined