Interactive Form PoCs with Pydantic

This repository contains multiple proof-of-concept interactive forms using Pydantic for data validation, implemented in different UI frameworks: Tkinter, MagicGui + Qt, FastUI, and Jupyter Notebook

---

Folder Overview

1. tkinter/

• Purpose: A desktop GUI version of the AIND Data Transfer Lite workflow using Tkinter
• Features:
  • Select a metadata directory
  • Select one or more modality directories (add more dynamically)
  • Optional settings: Dry Run and S3 Bucket
  • Validate inputs against a fake JobSettings Pydantic model
  • Output box displays validation results, and content can be copied to clipboard

---

2. magicgui/

• Purpose: A Qt-based interactive form PoC using MagicGUI
• Features:
  • Dynamically generate form from a Pydantic model
  • Add/delete multiple modality directories
  • Optional Dry Run and S3 Bucket fields
  • Validate and submit configuration
  • Output is read-only but selectable; can copy to clipboard

---

3. fastui/

• Purpose: A FastAPI + HTML version of a dynamic form
• Features:
  • Dynamically generates a form from a Pydantic model on a web page
  • POST form submission validates data and shows success or error messages
  • Simulate model updates by adding new fields dynamically

---

4. jupyter_notebook/

• Purpose: Interactive form inside a Jupyter Notebook using ipywidgets
• Features:
  • Dynamically generates fields from a Pydantic model
  • Submit button validates the inputs
  • Simulate model updates by adding new fields dynamically
  • Output is displayed below the form

---

Requirements

Each subproject has its own requirements.txt. Make sure to install dependencies in a dedicated virtual environment for each if needed