Hosted at: https://stockwatch.cloud/

Introduction

This COE714 project focuses on the creation of a web scraper website that would offer customers real-time notifications on price variations for certain stocks set by the user. This service allows consumers to specify personalized price change thresholds and receive timely notifications through email, allowing them to make educated investing decisions.

This project's major purpose is to create and construct a comprehensive website capable of monitoring and analyzing stock pricing changes for specific stocks set by users. Users may use this tool to remain up to date on the pricing dynamics of Stocks they are interested in acquiring. The scraper will gather data in a systematic manner, process it, and inform users via email when prices fall below or surpass predefined limits. The Project's Scope The following important components are included in the project:

1. Implementation of a Web Scraper

2. Design of the User Interface (UI)

3. Notification System through Email

4. Threshold Configuration Functionality

Features

1. User Friendly UI:

Allows easy interaction, setting preferences, and customizing email notifications. Also added feature of “Night Mode” for a darker theme.

2. User Customizable Thresholds:

Enables users to set personalized pricing alerts.

3. Automated Email Alerts:

Sends timely notifications based on user-defined price thresholds.

4. User Registration/Login (Custom Dashboard):

Users can create an account by registering or logging in to view their custom/specific stock portfolios.

Techstack Used

Frontend: React, HTML, CSS, MaterialUI

Backend: Python

Key Python Libraries: APScheduler, FastAPI, PyMongo, uvicorn

Database: MongoDB

API: YFinance API was primarily used for stock scraping

Hosting: Oracle Cloud VM, Docker, Nginx, Certbot for SSL, GitHub Actions for CI/CD

Status

Rudimentary functioning, under optimization.✅

Installation/Setup

Pip libraries to install (use pip3 install -r requirements.txt to install them for you): -pymongo -fastapi -yfinance -uvicorn

To start fresh on a new machine without affecting your host machine's python setup: (Steps 1 to 5 are only required for the initial setup)

Experimental: setupRepo.sh (unix based) and setupRepo.cmd (windows) automates this setup (1-5) for you.

1. In the backend directory cd backend, Create a virtual environment python3 -m venv venv.
2. Activate the virtual environment (if unix based: source ./venv/bin/activate, windows: venv\Scripts\activate) from within the backend folder.
3. Install required python libraries pip3 install -r requirements.txt
4. Install node modules for React within the frontend folder cd ../frontend, then npm i.
5. Create an .env file in the root of the project with the Email Account credentials.

EMAIL="stockwatch714@gmail.com"
PASSWORD="gmailAppPasswordFromDiscord"
SYMMETRIC_KEY="Byte String Generated by Fernet.generate_key()"

To Run:

Experimental: rundev.sh (unix based) and rundev.cmd (windows) automates this for you.

7. Start flask backend server uvicorn stockWatchAPI:app --reload from the backend directory
8. On another terminal, assuming starting from the repo's root directory, cd frontend and then Start React front-end npm run start.
9. When done working with backend, deactivate virtual env when done with deactivate or exit the terminal.

Overall, on an existing setup:

• Steps 2 and 6 to start the backend server. Step 7 to start the react dev frontend.

Changelog/Notes

Nov 7, 2023

• Added containerization support. You can run the whole solution using docker compose -f ./compose.yml up from the root project directory. To rebuild containers docker compose -f ./compose.yml build. To specify a https server endpoint other than 127.0.0.1 for the container, before building, run export RASP=stockwatch.cloud where stockwatch-cloud is the desired endpoint. For Windows, replace export with set.

The following are meant to be run from project/repo root:

• rundev.sh/.cmd - Run the development server/react front end in one command from the root project directory.
• setupRepo.sh/.cmd - Sets up the repository when freshly cloned, adds main repo upstream, creates dummy .env file if not present (replace with actual password after creation), and installs dependencies for front/back end.