PDF Literature Reader

A web-based PDF literature reader with AI translation and chat capabilities.

Features

• Split-view PDF reading (original on left, translation on right)
• AI-powered translation
• Intelligent chat interface
• Docker Compose deployment ready
• RESTful API for PDF management and AI services

Project Structure

paper-assistant/
├── app/                 # FastAPI backend
│   ├── main.py         # Application entry point
│   ├── api/            # API routes
│   ├── core/           # Core configurations
│   ├── models/         # Data models
│   ├── schemas/        # Pydantic schemas
│   └── utils/          # Utility functions
├── frontend/           # Frontend application
├── docker-compose.yml  # Docker Compose configuration
├── Dockerfile          # Backend Dockerfile
└── requirements.txt    # Python dependencies

Quick Start

Using Virtual Environment (Development)

1. Set up the environment:

   make setup
   

2. Run the application:

   make run
   

3. Open your browser and navigate to http://localhost:8000

Using Docker Compose (Deployment)

1. Build and start the services:

   make docker
   

2. Access the application at http://localhost:8000

How to Use

1. Upload a PDF:

   • Click "Choose File" and select a PDF document
   • Click "Upload PDF"
   • The uploaded PDF will appear in the dropdown list

2. View a PDF:

   • Select a PDF from the dropdown list
   • The original PDF will display in the left panel
   • A simulated translation will appear in the right panel

3. Chat with AI:

   • Type your question in the chat input at the bottom
   • Click "Send" or press Enter
   • Receive a simulated AI response

API Endpoints

PDF Management

• POST /api/pdf/upload - Upload a PDF file
• GET /api/pdf/list - List all uploaded PDFs
• GET /api/pdf/view/{filename} - View a specific PDF

Chat and Translation

• POST /api/chat/message - Send a message to the AI assistant
• POST /api/chat/translate - Translate text to Chinese

Development

1. Install dependencies:

   pip install -r requirements.txt
   

2. Run the application:

   uvicorn app.main:app --reload
   

Testing

Run the API tests:

make test

Create a test PDF:

python create_test_pdf.py

Deployment with Docker Compose

docker-compose up -d

Stop the services:

make docker-stop

Architecture

This application follows a modular architecture with clear separation of concerns:

1. Frontend: Simple HTML/CSS/JavaScript interface for PDF viewing and chat
2. Backend: FastAPI application with separate modules for PDF management and chat services
3. Storage: Local file system storage for uploaded PDFs
4. Deployment: Docker Compose configuration for easy deployment

The application is designed to be easily extensible with additional AI services for translation and chat functionality.