This server handles regulatory document processing with the following core components:

• PDF document ingestion
• Q&A pair generation
• JSON data storage in data_json/

server/
├── regulation_extractor.py     # Document parsing logic
├── qa_generator.py             # Q&A generation algorithm
├── rag_server_reranker.py      # Main API server implementation
├── data_json/                  # Generated Q&A JSON files
│   └── sample.json             # Example output format
├── .gitignore                  # Ignored files (logs, env files)
└── README.md                   # This file

1. Setup

# Install dependencies
pip install -r requirements.txt

# Start Main server
python3 rag_server_reranker.py

# Start Document server
python3 regulation_extractor.py

2. API Endpoints

• POST /upload-pdf - Upload regulatory documents
• GET /qa/:document_id - Retrieve Q&A pairs

Sample JSON structure (auto-generated in data_json/):

{
  "document_id": "reg_2025_01",
  "content": [
    {
      "question": "What is the purpose of this regulation?",
      "answer": "To establish employee management guidelines..."
    }
  ]
}

• PDF files should be placed in server/data/
• Q&A results are saved to server/data_json/

Required packages (install via pip):

• PyPDF2
• langchain
• fastapi
• uvicorn (for development server)