Finance tracker

App to keep track of your finances. The app is an experiment created from a simple web framework built from scratch without the react library.

Ideal to be used in a local network in the family. No log in is needed.

Features

1. Fast filtering of your transactions using Datatables.
2. Group the transactions by types and category and show intuitive respective pie charts.
3. Aggregate transaction by month.
4. Insert new transactions and update existing transactions.
5. Insert transaction by importing json files.

Future Features

• User Authentication: Add login/signup
• End-to-End Encryption:
  • All transaction data (amounts, dates, categories, types, notes) is encrypted in the browser before it’s sent to the server.
  • The server stores only ciphertext and necessary metadata (e.g. IVs, blind‐index tags) and never has access to decryption keys.
  • Each user’s keys are isolated and managed so that only they can decrypt their own records.
  • Privacy-preserving queries (e.g. category/type filters) use encrypted indexes so raw values are never exposed to the backend.
• Enhanced Search & Filtering: Implement privacy-preserving search (blind-index) for encrypted fields.

Setup

Install the packages by

make setup

Run the app

Back end

Start PostgresQL DB

cd backend
docker compose up postgres

Start Fast api

cd backend
uvicorn fin_pool.main:app --host 0.0.0.0 --port 8000 --reload

Front-end

Run parcel by

cd frontend
npm run start-parcel

In the Config.ts, you can set the url of the backend server if your server is not in localhost.

To initialize tha database with dummy data

1. Start the postgres DB
2. Start the Fast API
3. Navigate to the Fast API docs
4. Run the upload file end point with the file dummy_db_json file provided.

Note:

1. This must be done through the FastAPI documentation interface, as the feature is currently not enabled from the frontend.
2. Only json files are supported.