Extract2MD v2.0.0 - Major Release

🚀 Full Redesign & Complete API Overhaul

Release Date: 24-05-2025
Version: 2.0.0 (Breaking Changes)
Migration Support: Legacy API maintained for transition period

📋 Release Overview

Extract2MD v2.0.0 represents a complete reimagining of the library with a focus on developer experience, intuitive usage patterns, and modern architecture. This major release introduces a revolutionary scenario-based API that replaces the complex instance-based approach with clear, purpose-driven methods.

Core Philosophy: Instead of configuring complex options, developers now choose from 5 distinct conversion scenarios that match their specific use cases.

⚠️ Breaking Changes

API Complete Redesign

Old: Instance-based API with complex configuration options
New: Static methods with scenario-based approach
Impact: All existing integrations require updates
Migration: Legacy API available as LegacyExtract2MDConverter during transition

Configuration Changes

Old: Loose configuration object with numerous optional parameters
New: Structured configuration with validation and default merging
Impact: Configuration structure has changed significantly
Migration: Use ConfigValidator for seamless config handling

Import/Export Changes

Old: Single converter class export
New: Modular exports with main converter and utilities
Impact: Import statements need updating
Migration: Update imports and follow new module structure

✨ New Features

🎯 Scenario-Based API

Five distinct conversion methods designed for specific use cases:

1. Quick Only - `Extract2MDConverter.quickOnly()`

Purpose: Fast PDF.js-based text extraction
Best For: Clean PDFs with selectable text
Performance: Fastest option, minimal processing
Use Case: Documentation, reports, digital-native PDFs

2. High Accuracy OCR Only - `Extract2MDConverter.highAccuracyOCROnly()`

Purpose: Tesseract OCR with canvas rendering
Best For: Scanned documents, images, complex layouts
Performance: Slower but highly accurate
Use Case: Scanned books, historical documents, printed materials

3. Quick + LLM - `Extract2MDConverter.quickPlusLLM()`

Purpose: Fast extraction enhanced with AI processing
Best For: PDFs needing structure improvement
Performance: Moderate, WebGPU accelerated
Use Case: Business documents, formatted reports

4. High Accuracy + LLM - `Extract2MDConverter.highAccuracyPlusLLM()`

Purpose: OCR processing with AI enhancement
Best For: Complex documents requiring both OCR and AI
Performance: Comprehensive, highest quality
Use Case: Academic papers, technical documents

5. Combined + LLM - `Extract2MDConverter.combinedPlusLLM()`

Purpose: All extraction methods with AI post-processing
Best For: Maximum accuracy and formatting
Performance: Most thorough, longest processing time
Use Case: Critical documents, archival processing

🧩 Modular Architecture

Complete internal refactoring into specialized modules:

Extract2MDConverter.js - Main converter with scenario methods
WebLLMEngine.js - Encapsulated LLM integration
ConfigValidator.js - Configuration validation and defaults
OutputParser.js - LLM output cleaning and formatting
SystemPrompts.js - Centralized prompt management

📚 Comprehensive Documentation Suite

New Documentation Files:

MIGRATION.md - Step-by-step migration guide with code examples
DEPLOYMENT.md - Complete deployment guide for all environments
config.example.json - Full configuration example
Updated README.md - Rewritten for new API

Interactive Examples:

demo.html - Live interactive demo showcasing all 5 scenarios
usage-examples.js - Updated code examples for new API
SSL certificates - Demo server setup for local testing

⚙️ Enhanced Configuration System

Structured Configuration Object with clear hierarchy
Built-in Validation with ConfigValidator utility
JSON Configuration Support for external config files
Default Value Merging for simplified setup
Type Safety with comprehensive TypeScript definitions

🧪 Robust Testing Framework

New comprehensive test suite:

scenarios.test.js - Tests for all 5 scenario methods
simple.test.js - Basic structure validation
newline-optimization.test.js - Markdown formatting tests
simple-newline.test.js - Standalone newline processing tests
validate-deployment.js - Deployment readiness validation

🔧 Technical Improvements

Build System Enhancements

Dual Bundle Generation: UMD and ESM formats
Optimized Distribution: Essential workers and definitions copied to dist
Updated Entry Points: Proper main, module, and types configuration
Enhanced Packaging: Improved file inclusion/exclusion

TypeScript Integration

Complete Type Definitions in src/types/index.d.ts
Scenario Method Types with proper return types and parameters
Configuration Interfaces for type-safe config handling
Legacy Compatibility Types for migration support

Performance Optimizations

WebGPU Capability Detection for LLM scenarios
Modular Loading reduces initial bundle size
Optimized Canvas Rendering for OCR processing
Streaming LLM Support for better user experience

Developer Experience

Clear Error Messages with improved error handling
Progress Tracking across all conversion scenarios
Intuitive Method Names that clearly indicate functionality
Consistent Return Formats across all scenarios

🛤️ Migration Guide

Immediate Steps

Install v2.0.0: npm install extract2md@2.0.0
Use Legacy API: Replace Extract2MDConverter with LegacyExtract2MDConverter
Test Functionality: Ensure existing code works with legacy API
Plan Migration: Review MIGRATION.md for upgrade path

Recommended Migration Process

Identify Usage Patterns: Determine which scenarios match your current usage
Update Configuration: Migrate to new structured config format
Replace Method Calls: Switch to appropriate scenario-based methods
Update Error Handling: Adapt to new error formats
Test Thoroughly: Validate output quality and performance

Timeline

v2.0.0 - v2.x.x: Legacy API available alongside new API
v3.0.0: Legacy API will be removed (future major release)
Recommended: Migrate within 1 months for best support

📦 Installation & Deployment

NPM Installation

npm install extract2md@2.0.0

Import Examples

// New API (recommended)
import { Extract2MDConverter } from 'extract2md';

// Legacy API (for migration)
import { LegacyExtract2MDConverter } from 'extract2md';

// Utilities
import { ConfigValidator, OutputParser } from 'extract2md';

Deployment Options

Node.js Applications: Full feature support
Web Applications: Browser-compatible with WebWorkers
CDN Distribution: Direct browser usage
Static Sites: Pre-built bundle integration

🌟 What's New in Detail

WebLLM Engine Integration

Standalone Engine Class for better modularity
Streaming Support for real-time processing feedback
Model Loading Management with error handling
WebGPU Optimization for enhanced performance

Output Processing Pipeline

Thinking Tag Removal from LLM outputs
Markdown Normalization for consistent formatting
Newline Optimization for better readability
Post-processing Hooks for custom transformations

Configuration Validation

Schema-based Validation with clear error messages
Default Value Injection for missing configuration
Type Coercion for flexible config input
JSON File Support for external configuration

Enhanced Error Handling

Scenario-specific Errors with context information
Validation Errors with field-level details
Processing Errors with progress context
Recovery Suggestions for common issues

🔮 Looking Forward

Planned Enhancements

Additional Scenarios based on user feedback
Performance Optimizations for large document processing
Enhanced LLM Models support and configuration
Advanced Output Formats beyond Markdown

Community & Support

Migration Support: Comprehensive documentation and examples
Community Feedback: Open to suggestions for new scenarios
Regular Updates: Incremental improvements and bug fixes
Long-term Support: Commitment to stable API evolution

📞 Support & Resources

Migration Guide: MIGRATION.md - Complete migration instructions
Deployment Guide: DEPLOYMENT.md - Production deployment best practices
Interactive Demo: examples/demo.html - Try all scenarios
Configuration Example: config.example.json - Complete config reference
Type Definitions: Full TypeScript support included

🙏 Acknowledgments

This major release represents months of development focused on creating the most intuitive and powerful PDF-to-Markdown conversion experience. Thank you to all contributors and early adopters who provided feedbac...

Releases: hashangit/Extract2MD

v2.0.0