File Management System with AI Capabilities

Project Overview

This project is a sophisticated file management system that combines modern web technologies with AI capabilities. It features a React-based frontend and a FastAPI backend, providing a seamless experience for file management, PDF processing, and AI-powered document interaction.

Working Video

https://drive.google.com/file/d/1DVKwiVIc6HVCcfoEITgRKb1THtIgFS_5/view?usp=sharing

Architecture

Frontend

• Built with React and TypeScript
• Uses Tailwind CSS for styling
• Features a modern, responsive UI with a collapsible sidebar
• Implements real-time chat interface for document interaction

Backend

• FastAPI-based REST API
• SQLAlchemy for database management
• PDF processing capabilities using PyMuPDF
• Vector store integration for AI features
• CORS middleware for secure cross-origin requests

Key Features

1. File Management

# Backend file handling
@app.post("/api/files/upload")
async def upload_file(file: UploadFile):
    # File processing logic
    return {"filename": file.filename, "status": "success"}

2. PDF Processing

def extract_text_from_pdf(pdf_data: bytes) -> Dict[str, any]:
    """
    Extract text content from a PDF file with metadata and page information
    """
    pdf_document = fitz.open(stream=pdf_data, filetype="pdf")
    metadata = pdf_document.metadata
    page_count = len(pdf_document)
    # ... text extraction logic

3. Chat Interface

The system includes an AI-powered chat interface that allows users to interact with their documents:

<ChatInterface
  selectedFile={selectedFile}
  chatHistory={chatHistory}
  onSendMessage={handleSendMessage}
  isLoading={isChatLoading}
  userProfileImage={userProfileImage}
/>

Development Challenges and Solutions

1. PDF Processing

Challenge: Efficiently extracting and processing large PDF files while maintaining performance. Solution: Implemented streaming processing and caching mechanisms:

# PDF processing with error handling
try:
    pdf_document = fitz.open(stream=pdf_data, filetype="pdf")
    # Process pages individually
    for page_num in range(page_count):
        page = pdf_document[page_num]
        text = page.get_text()
        # Store in vector database
except Exception as e:
    handle_error(e)

2. Real-time Chat

Challenge: Managing real-time communication between frontend and backend. Solution: Implemented WebSocket connections and state management:

const handleSendMessage = async (message) => {
  try {
    setIsLoading(true);
    const response = await sendMessageToBackend(message);
    updateChatHistory(response);
  } catch (error) {
    handleError(error);
  } finally {
    setIsLoading(false);
  }
};

3. File Management

Challenge: Handling large file uploads and maintaining file integrity. Solution: Implemented chunked uploads and validation:

@app.post("/api/files/upload")
async def upload_file(file: UploadFile):
    try:
        # Validate file type and size
        if not is_valid_file(file):
            raise HTTPException(status_code=400, detail="Invalid file")
        
        # Process file in chunks
        contents = await file.read()
        # Store in database
        return {"status": "success"}
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))

Error Handling

The system implements comprehensive error handling at multiple levels:

1. Frontend Error Handling:

{error && (
  <div className="bg-red-50 border-l-4 border-red-400 p-4">
    <div className="flex">
      <div className="ml-3">
        <p className="text-sm text-red-700">{error}</p>
      </div>
    </div>
  </div>
)}

2. Backend Error Handling:

@app.exception_handler(Exception)
async def global_exception_handler(request: Request, exc: Exception):
    return JSONResponse(
        status_code=500,
        content={"detail": str(exc)}
    )

Setup and Installation

1. Backend Setup:

cd backend
python -m venv venv
source venv/bin/activate  # or `venv\Scripts\activate` on Windows
pip install -r requirements.txt

2. Frontend Setup:

cd frontend
npm install
npm run dev

Environment Variables

Create a .env file in the backend directory:

DATABASE_URL=sqlite:///./file_management.db
VECTOR_STORE_PATH=./vector_store

API Documentation

The API documentation is available at /docs when running the backend server. Key endpoints include:

• POST /api/files/upload - Upload new files
• GET /api/files - List all files
• POST /api/chat - Send chat messages
• GET /api/vector-store/search - Search through document content

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request