Adobe Helper

Adobe Helper is a Python library for converting PDF files to Word (DOCX) format using Adobe's online conversion services. It provides a clean, async API with automatic session management, rate limiting, and quota tracking.

⚠️ Current Status

This project is fully functional out-of-the-box! The library includes working API endpoint patterns discovered from Adobe's online service. Simply install and start converting PDFs to Word documents.

Recent Updates (2025-10-21)

✅ Embedded API Endpoints (NEW!)

• Working endpoint patterns now embedded in the codebase
• No manual endpoint discovery required for basic usage
• Library works out-of-the-box after pip install
• Automatic tenant ID substitution for each session

✅ Multi-Tenant Architecture

• Automatic tenant discovery during session initialization
• Dynamic endpoint switching per session
• Support for multiple regions and tenant IDs
• Each session discovers its own numeric tenant ID from Adobe's servers

✅ Logging Enhancement

• Examples now include proper logging configuration
• Real-time visibility into conversion progress
• Better debugging and troubleshooting support

Advanced users can still manually discover and configure custom endpoints if needed. See docs/discovery/API_DISCOVERY.md for details.

Features

✨ Easy to Use

• Simple async API with context manager support
• Automatic session management and rotation
• Built-in retry logic with exponential backoff
• Bypass local usage limits (mimics clearing browser data)

📊 Smart Management

• Optional usage tracking with daily limits
• Intelligent rate limiting with human-like delays
• Automatic session rotation for unlimited conversions
• Fresh session creation (like incognito mode)

🔒 Reliable

• Streaming upload/download for large files
• File integrity verification
• Comprehensive error handling
• Progress tracking support

🚀 Fast

• Async/await throughout
• HTTP/2 support via httpx
• Concurrent batch processing

Installation

Using uv (Recommended)

# Clone the repository
git clone https://github.com/karlorz/adobe-helper.git
cd adobe-helper

# Install with uv
uv sync --all-extras

Using pip

# Clone the repository
git clone https://github.com/karlorz/adobe-helper.git
cd adobe-helper

# Install in development mode
pip install -e .

Quick Start

Basic Usage

import asyncio
import logging
from pathlib import Path
from adobe import AdobePDFConverter

# Configure logging to see conversion progress
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
)

async def main():
    # Convert a PDF to Word (bypasses local limits by default)
    async with AdobePDFConverter(
        bypass_local_limits=True  # Mimics clearing browser data
    ) as converter:
        output_file = await converter.convert_pdf_to_word(
            Path("document.pdf")
        )
        print(f"Converted: {output_file}")

asyncio.run(main())

Batch Conversion

from adobe import AdobePDFConverter

async def batch_convert():
    pdf_files = [
        Path("doc1.pdf"),
        Path("doc2.pdf"),
        Path("doc3.pdf"),
    ]

    async with AdobePDFConverter() as converter:
        for pdf_file in pdf_files:
            try:
                output = await converter.convert_pdf_to_word(pdf_file)
                print(f"✓ {pdf_file.name} -> {output.name}")
            except Exception as e:
                print(f"✗ {pdf_file.name}: {e}")

Advanced Configuration

from adobe import AdobePDFConverter
from pathlib import Path

async def advanced_convert():
    # Custom configuration
    converter = AdobePDFConverter(
        session_dir=Path(".cache"),      # Custom cache directory
        use_session_rotation=True,       # Enable session rotation
        track_usage=True,                # Track daily quota
        enable_rate_limiting=True,       # Rate limiting
    )

    try:
        await converter.initialize()

        # Convert with custom output path
        output = await converter.convert_pdf_to_word(
            Path("input.pdf"),
            output_path=Path("output/converted.docx"),
        )

        # Check usage stats
        usage = converter.get_usage_summary()
        print(f"Daily usage: {usage['count']}/{usage['limit']}")

    finally:
        await converter.close()

Endpoint Discovery CLI

Use the bundled helper to capture endpoints and keep discovery files synced:

# Show available commands
python -m adobe.cli.api_discovery_helper --help

# Create or refresh the project discovery template
python -m adobe.cli.api_discovery_helper template

# Validate captured URLs and sync project ↔ user cache copies
python -m adobe.cli.api_discovery_helper update

# Installed entry point (after `pip install .`)
adobe-api-discovery checklist

See docs/discovery/API_DISCOVERY.md for the full walkthrough.

The helper stores discovered endpoints in ~/.adobe-helper by default, but will fall back to ./.adobe-helper (or the system temp directory) automatically when the home directory is not writable—useful for containerized or sandboxed environments.

Architecture

Core Components

adobe/
├── client.py              # Main AdobePDFConverter class
├── auth.py                # Session management
├── session_cycling.py     # Anonymous session rotation
├── cookie_manager.py      # Cookie persistence
├── upload.py              # File upload handler
├── conversion.py          # Conversion workflow manager
├── download.py            # File download handler
├── rate_limiter.py        # Rate limiting with backoff
├── usage_tracker.py       # Free tier quota tracking
├── models.py              # Pydantic data models
├── exceptions.py          # Custom exceptions
├── constants.py           # Configuration constants
├── urls.py                # API endpoints
└── utils.py               # Helper functions

Data Flow

PDF File → Upload → Conversion Job → Poll Status → Download DOCX
           ↓         ↓                 ↓             ↓
        Validate  Create Job      Wait/Poll    Stream Download
        Retry     Track Status    Adaptive      Verify
                                  Polling       Integrity

Examples

See the examples/adobe/ directory for complete examples:

• basic_usage.py - Simple conversion with bypass enabled
• batch_convert.py - Sequential and concurrent batch processing
• advanced_usage.py - Advanced configuration and error handling

Legacy bypass/reset scripts now live under archive/docs/ for reference.

Bypassing Usage Limits

By default, the library now bypasses local usage tracking and relies on Adobe's server-side limits with automatic session rotation:

# Automatic session rotation (recommended for batch processing)
async with AdobePDFConverter(
    bypass_local_limits=True,  # Default: True
    use_session_rotation=True,  # Auto-rotate sessions
) as converter:
    for pdf in pdf_files:
        await converter.convert_pdf_to_word(pdf)

For more details, see BYPASS_LIMITS.md.

Quick reset: Call AdobePDFConverter.reset_session_data() (or use AdobePDFConverter.create_with_fresh_session()) to clear all local state; the legacy helper script now resides in archive/docs/.

API Discovery Required

⚠️ Important: Before this library can perform actual conversions, you need to discover Adobe's API endpoints using Chrome DevTools.

See docs/discovery/API_DISCOVERY.md for detailed instructions.

Discovered endpoint files are cached automatically: any discovered_endpoints.json found in docs/discovery/ or archive/discovery/ is copied into ~/.adobe-helper/ on first run, and a template is generated if missing.

Development

Setup Development Environment

# Install UV (if not already installed)
curl -LsSf https://astral.sh/uv/install.sh | sh

# Clone and setup
git clone https://github.com/karlorz/adobe-helper.git
cd adobe-helper
uv sync --all-extras --dev

Run Tests

# Run all tests
uv run pytest

# Run with coverage
uv run pytest --cov=adobe --cov-report=html

# Run specific test file
uv run pytest tests/test_models.py -v

Code Quality

# Format code
uv run black adobe/ tests/

# Lint code
uv run ruff check adobe/ tests/

# Type checking
uv run mypy adobe/

Project Status

✅ Completed (Phases 1-10)

• Project setup and architecture
• Data models with Pydantic validation
• Custom exception hierarchy
• Session management and rotation
• Cookie management
• Rate limiting with adaptive backoff
• Usage tracking
• File upload handler
• Conversion workflow manager
• File download handler
• Main client class
• Example scripts with logging
• Unit tests (30 tests, 100% pass rate)
• Documentation
• Multi-tenant architecture with automatic discovery ✨ NEW
• Dynamic endpoint switching per session ✨ NEW

🔄 Remaining

• API endpoint discovery (critical - see docs/discovery/API_DISCOVERY.md)
• Integration tests with real API
• CLI tool (optional)
• Browser automation fallback (optional)

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests
5. Run code quality checks
6. Submit a pull request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Disclaimer

This library is for legitimate use only. Please respect Adobe's Terms of Service and rate limits. The library includes built-in rate limiting and quota tracking to prevent abuse.

Acknowledgments

• Inspired by Adobe's online PDF conversion services
• Built with httpx, pydantic, and modern Python async patterns
• Developed using uv for fast dependency management

Support

• 📫 Issues: GitHub Issues
• 📖 Documentation: See examples/ and AGENTS.md
• 💬 Discussions: GitHub Discussions