Skip to content

CHANGELOG.md

Latest commit

 

History

History
189 lines (140 loc) · 8.43 KB

CHANGELOG.md

File metadata and controls

189 lines (140 loc) · 8.43 KB

Changelog

All notable changes to the Protegrity AI Developer Edition Python project will be documented in this file.

[1.1.1] - 2025-12-16

📦 Distribution

  • PyPI Package: Re-released as v1.1.1 to update package metadata and README on PyPI

Note: This is a documentation-only release. No code changes from v1.1.0.

[1.1.0] - 2025-12-15

🎉 Major New Features

Semantic Guardrails v1.1.0 SDK Support

  • Enhanced Risk Scoring: Updated SDK to support Semantic Guardrails v1.1.0 with improved risk assessment capabilities
  • Vertical-Specific Models: Added support for Finance and Healthcare industry-specific models
  • Multi-turn Conversation Support: Enhanced PII scanning and risk scoring across conversation history
  • Improved API Interface: Streamlined SDK interface for semantic guardrail operations

Data Discovery Enhancements

  • Harmonized Classifications: Support for categorized "harmonized" entity classifications
  • Entity Mapping Updates: Updated entity-to-data-element mapping to align with Data Discovery v1.1.1
  • Improved Accuracy: Enhanced classification accuracy and confidence scoring
  • Overlapping Labels: Fixed ordering logic for overlapping classification labels

Conda Package Support (NEW)

  • Conda Recipe: Added conda-recipe/ directory with complete build configuration
  • Cross-Platform Distribution: Support for conda package distribution across platforms
  • Meta.yaml Configuration: Comprehensive conda package metadata and dependencies

🏗️ Architecture & Structure Changes

Repository Structure Enhancements

  • Conda Recipe Directory: New conda-recipe/ with build scripts and metadata
  • Enhanced Test Structure: Improved test organization and expected outputs
  • Configuration Updates: Removed hardcoded endpoint URLs from mapping_config.json

SDK Interface Improvements

  • Cleaner APIs: Simplified method signatures for semantic guardrail operations
  • Better Error Handling: Enhanced error messages and exception handling
  • Type Hints: Improved type annotations for better IDE support

🔧 Enhanced Configuration & Service Features

Configuration Updates

  • Dynamic Endpoint Configuration: Removed hardcoded endpoint_url from mapping configuration
  • Flexible Mapping: Enhanced entity mapping configuration options
  • Environment-Based Config: Better support for environment-specific configurations

Testing Improvements

  • Updated Test Outputs: Refreshed expected test outputs to match Data Discovery 1.1.1 entity names and patterns
  • Semantic Guardrails Unit Tests: Updated unit tests for v1.1.0 compatibility
  • Better Test Coverage: Expanded test scenarios for new features

📚 Documentation & Developer Experience

README Enhancements

  • "Why This Matters" Section: Added context about the importance of data protection
  • Improved Examples: More comprehensive code examples and use cases
  • Better Prerequisites: Clearer setup instructions and dependency documentation

Developer Guidance

  • Conda Installation: New installation method via conda packages
  • API Documentation: Enhanced inline documentation and docstrings
  • Migration Notes: Guidance for upgrading from 1.0.0 to 1.1.0

🔄 Dependencies

  • Updated Requirements: Refreshed requirements.txt with compatible versions
  • Conda Dependencies: Added conda-specific dependency management
  • Python Version: Maintained Python 3.12.11+ requirement

🔐 Security

  • Dependency Updates: Updated to latest secure versions of dependencies
  • Vulnerability Fixes: Applied security patches as needed

⚠️ Breaking Changes

  • Configuration Schema: Removed endpoint_url from mapping_config.json - endpoints are now dynamically determined
  • Entity Mapping: Updated entity names and patterns to match Data Discovery 1.1.1 - may require configuration updates for custom mappings

📦 Distribution

  • PyPI Package: Available as protegrity-developer-python v1.1.0
  • Conda Package: New distribution channel via conda (coming soon)
  • Wheel Distribution: Pre-built wheel available for quick installation

[1.0.0] - 2025-09-30

🎉 Major New Features

New Application Protector Python Module (appython)

  • Data Protection & Unprotection: Complete functionality for protecting and unprotecting sensitive data elements
  • Session Management: Secure session handling for protection operations
  • Single Data Operations: Protect and unprotect individual data elements
  • Bulk Data Operations: Protect and unprotect multiple data elements in batch operations
  • Cloud-based Protection: Works without requiring local Protegrity AI Developer Edition installation

Enhanced Protegrity Developer Python Module

  • Find and Protect: New functionality for classifying and protecting PII in unstructured text
  • Find and Unprotect: New functionality for restoring original PII data from its protected form
  • Enhanced Data Discovery: Improved classification capabilities

🏗️ Architecture & Structure Changes

Repository Structure

  • New Module: Added src/appython/ directory with comprehensive protection capabilities
    • protector.py - Main protection/unprotection logic
    • service/ - Service layer components
    • utils/ - Utility functions
  • Enhanced Testing: Expanded test structure with dedicated paths for both modules
    • tests/unit/appython/ - Unit tests for Application Protector
    • tests/unit/find_and_secure/ - Enhanced tests for data discovery
    • Support for bulk operations testing
    • Mock testing capabilities

Configuration Files

  • Added: setup.cfg for enhanced build configuration
  • Enhanced: pyproject.toml with support for multiple modules

🔧 Prerequisites & Setup

New Authentication Requirements

  • API Key, Email, and Password: Required for Application Protector Python module
  • Developer Portal Registration: New registration process (in progress)
  • Alternative Signup: cURL-based signup method for immediate access

Updated System Requirements

  • Python Version: Upgraded minimum requirement from 3.9.23 to 3.12.11
  • Cross-platform Support: Maintained for Linux, Windows, and macOS

📦 Installation & Usage

Enhanced Installation Options

  • Fresh Installation: pip install .
  • Upgrade Installation: pip install --upgrade . for existing installations

New Usage Patterns

  • Environment Variables: Support for credential management via environment variables
  • Dual Module Support: Can use both modules independently or together
  • Session-based Operations: New session management for Application Protector

🔐 Security & Authentication

AI Developer Edition Portal Integration

  • Cloud Authentication: Integration with Protegrity AI Developer Edition Portal
  • API Key Management: Secure handling of API credentials
  • Registration Process: Streamlined developer account creation

📋 Documentation & Examples

Comprehensive Usage Examples

  • Application Protector Examples: Complete examples for single and bulk data operations
  • Enhanced Data Discovery Examples: Updated examples with new protect/unprotect capabilities
  • Configuration Guidance: Detailed setup instructions for both modules

Updated Documentation Links

  • Developer Portal: Enhanced documentation and API reference
  • Setup Instructions: Clear guidance for different use cases

🎯 Use Cases

Expanded Application Scenarios

  • GenAI Applications: Enhanced support for AI/ML applications
  • Structured Data Protection: New capabilities for protecting structured data elements
  • Unstructured Text Processing: Improved PII detection and protection in text
  • Hybrid Workflows: Support for both redaction and protection workflows

⚠️ Breaking Changes

  • Python Version: Minimum Python version increased from 3.9.23 to 3.12.11

🐛 Bug Fixes & Improvements

  • Enhanced Error Handling: Better error codes and handling for bulk operations
  • Improved Logging: Enhanced logging capabilities across both modules
  • Cross-platform Stability: Improved compatibility across different operating systems

[Previous Release] - README.md Baseline

Features (Baseline)

  • Basic Find and Redact functionality
  • Single module structure (protegrity_developer_python)
  • Python 3.9.23 support
  • Basic configuration options
  • Simple repository structure