StockAlerter

StockAlerter is a simple helper for all, who invest their money on a stock exchange! It alerts you with an email, when the prices are extremely low/high and sends you a daily summary. All these with detailed calculations.

StockAlerter can operate multiple stocks for multiple users with different emails.

You just have to modify the config file with your transactions details!

Preparation

1. Clone the repo.
2. Install required packages (pip install -r requirements.txt), use an env with Python 3.7.
3. Create a config.json file (you can find config_sample.json in repo; see Config file chapter) with all required details.
4. Create two environmental variables with your email account details (see Sending an email chapter).
5. flask run

Details

Scheduler

The stocks are being checked in two ways:

• min_max_mail: every given interval (specified in config.json) StockAlerter downloads the stocks and compares them with mininum and maximum values for a given trade.
• daily_mail: from Monday to Friday at 6 pm. StockAlerter downloads all the stocks and sends a summary email to recipient.

Both ways are being called from schedulers (APScheduler has been used): Interval and Cron.

Both can be enabled/disabled in config.json.

See app/scheduling.py.

Downloading stock price

Since GPW (Giełda Papierów Wartościowych, main polish stock exchange) doesn't share its API for free (actually it's very expensive!), webscraping of bankier.pl has been used. Although it works fine, it has two main disadvantages:

• it's harder to implement and maintain than an API.
• it has a delay of approx. 20 minutes comparing to current GPW data.

For webscraping requests and BeautifulSoup have been used.

See app/stock_value.py.

Sending an email

Before an email is sent, it must be properly prepared. Since the email is the final product for the user, the email must contain every imporant information in readable form. This is where EmailPreparer comes in.

EmailPreparer handles all logic in terms of choosing right recipients and preparing email message. It's the last step before actual email shipment.

See: app/email_preparer.py.

For logic of sending mail EmailSender is responsible. It uses smtplib with TLS security protocol. A specific host and port could be specified in config.json.

All email account details, such as login and password, are only referenced from the code as environmental variables.

IMPORTANT: In order to make this project work, you have to set up these env variables:

STOCK_ALERTER_LOGIN - your email login from which the emails will be sent STOCK_ALERTER_PASSWORD - your email password from which the emails will be sent

See: app/email_sender.py.

History

The history is kept in history.csv file. It's being updated on every call. HistoryManager class is responsible for reading the file, searching for global minimums and global maximums and saving to the file.

See: app/history_manager.py.

Config file

The config.json file allows to manage things like:

• recipients (address)
• consent for a daily mail (daily_mail)
• transactions details (symbol, buy_price, buy_quantity)
• consent for a min/max mail within every transaction (min_max_mail)
• smtp details (host, port)
• data sources

You must have a proper config.json file for the app to run!

See a sample config file: config_sample.json.

Calculator

The Calculator calculates revenues, costs and profits taking under consideration commissions and taxes.

Assumptions:

• An investment tax rate: 19% (Polish podatek Belki).
• A commision rate: 0,39%.

Utils

Some of static utils function have been extracted to app/utils.py, e.g. calculate_time wrapper, which can be used as a decorator:

@calculate_time
def function():
    ....

Example output:

Execution of 'function' done in 1.87 seconds.

See: app/utils.py.

Technology stack

• python 3.7
• flask
• APScheduler (https://pypi.org/project/APScheduler/)
• BeautifulSoup