Financial Planner

This project contains a simple Streamlit web application for tracking personal finances.
It lets you manage accounts and transactions, upload PDF statements and view a summary dashboard. The app stores data in Supabase and uses Microsoft OAuth for authentication.
PDF extraction is powered by Google's Generative AI model.

Installation

Install the dependencies listed in requirements.txt:

pip install -r requirements.txt

Create a .env file and set the required environment variables:

• CLIENT_ID, TENANT_ID, CLIENT_SECRET – Azure AD credentials
• SUPABASE_URL, SUPABASE_KEY – Supabase connection info
• GOOGLE_API_KEY, GOOGLE_PROJECT_ID, GEMINI_MODEL – configuration for the PDF extractor

Running

Launch the development server with Streamlit:

streamlit run app.py

A Dockerfile is included for containerized execution.

Directory structure

.
├── app.py               # Main Streamlit entry point
├── auth.py              # Microsoft OAuth helpers
├── database.py          # Supabase CRUD operations
├── dashboard.py         # Dashboard UI helpers
├── extractor.py         # PDF parsing and AI logic
├── utils.py             # Misc utilities
├── components/          # Reusable Streamlit widgets
│   ├── auth_widgets.py
│   ├── charts.py
│   └── sidebar.py
├── pages/               # Streamlit page modules
│   ├── 1_accounts.py
│   ├── 2_transactions.py
│   ├── 3_statements.py
│   └── 4_dashboard.py
├── requirements.txt     # Python dependencies
├── Dockerfile           # Container setup
└── test.py              # Basic test harness

Testing

Run the simple tests with:

python test.py