DocumentQAChatBot

DocumentQAChatBot is a web-based Question-Answering chatbot system built using Python 3.12, FastAPI backend, NiceGUI frontend, and Agno >=1.5.0 for AI agent integration. The system allows users to have normal chat conversations with an AI model and also perform PDF-based Q&A by uploading PDF documents. All chat sessions are handled in memory per user session; no persistent storage is used.

---

Project Features

Frontend

• Built on NiceGUI 2.18.0.
• Login Page: Simulated demo login credentials for authentication.
• Dashboard Page:
  • New Chat Session Button: Generates a unique session ID for each user. Shows:
    • Chat Box
    • PDF Upload Box
    • Status Box
  • Show All Chat Session History: Placeholder implementation ("can be done later").
• PDF Upload: Upload PDF files (max 5MB, non-empty, PDF-only) to enable Q&A based on document content.
• Chat Box: Allows normal or document-based conversation.
• Status Box: Shows system and upload statuses.

Backend

• Built on FastAPI and Python 3.12.
• APIs:
  1. /api/v1/login/authenticate_user_login → Authenticate demo users, return USER-ID.
  2. /api/v1/chat/ask → End-user asks a question; returns streaming response token-by-token from Agno agent.
  3. /api/v1/pdf/upload_pdf → Handles PDF file upload, parsing, and adds parsed content to the user session for Q&A.
  4. /api/v1/chat/chat_session_history → Returns chat history for a given USER-ID + SESSION-ID.
  5. /api/v1/chat/reset_chat_session_history → Resets chat session history for a user session.
• Agno Agent: Handles AI chat with session and in-memory history.
• PDF Parsing: Uses PyMuPDF (fitz) to extract text content per page.

---

Project Structure

│
├── backend/
│ ├── api/
│ │ ├── v1/
│ │ │ ├── routes/ # system_health, login, chat, pdf APIs
│ │ │ ├── schemas/ # login, chat, pdf schemas
│ │ │ ├── services/ # pdf_parser_service.py
│ │ │ └── agents/ # chatbot_agents.py (Agno integration)
│ └── utils/ # helper functions
│
├── frontend/
│ ├── main_ui.py # main NiceGUI frontend
│ └── components/ # chat_box, upload_box, status_box
│
├── data/ # uploaded PDFs
├── tests/ # unit & integration tests
├── main.py # FastAPI entry point
├── requirements.txt # Python dependencies
├── Dockerfile_backend_production_machine
├── Dockerfile_frontent_production_machine
├── docker-compose_production_machine.yaml
└── .env.example # Environment sample variables

---

Setup Instructions

Prerequisites

• Python 3.12
• Docker & Docker Compose (for containerized setup)

Steps to Follows to execute project into the local machine:

1. Make this [PythonProject] directory from terminal
2. Goto this directory [PythonProject] from terminal
3. Execute this command: git clone https://github.com/ChiggyJain/PythonDocumentQAChatBot.git
4. Goto this directory [PythonDocumentQAChatBot] from terminal
5. Copy the [].env.example] file content into [.env] file and also replace OPENI_API_KEY value.
6. Execute this command from terminal one-by-one: a. docker compose -f docker-compose_production_machine.yaml down -v b. docker compose -f docker-compose_production_machine.yaml build --no-cache c. docker compose -f docker-compose_production_machine.yaml up
7. This url [http://0.0.0.0:8000/docs] is for accessing backend-system APIs documentation and hitting the various apis request.
8. Open this url [http://localhost:8080] in any browser for accessing the frontend-system.
9. Follows these steps to execute test-cases of backend-system docker container services: a. Execute this command from your terminal: docker container ps -a b. Copy the container-id from the terminal whose name is: documentqa_backend_container c. Execute this command from the terminal: i. docker exec -it Put-Here-Docker-Container-Services-ID bash d. Exeucte this command for unit testing: i. PYTHONPATH=$(pwd) pytest tests/unit --disable-warnings -v -s --maxfail=0 -W ignore::DeprecationWarning e. Exeucte this command for integration testing: i. PYTHONPATH=$(pwd) pytest tests/integration --disable-warnings -v -s --maxfail=0 -W ignore::DeprecationWarning