/Letterboxd-friend-ranker

Program that computes, ranks a given user and their friends based on Letterboxd ratings

Primary LanguagePythonMIT LicenseMIT

Letterboxd Friend Ranker

SEE HOW MUCH YOU HAVE IN COMMON WITH YOUR FRIENDS' FILM TASTES

Letterboxd Friend Ranker takes a given user, scrapes all their corresponding friends' ratings data, and proceeds to compute and rank the user's friends based on a commonality score. Each score is computed as such:

Remarks:

I intend to implement more functionality; such as: clean charts illustrating difference between friends, predictions,further statistics, GUI interface with Tkinter, and more. Additionally, I will be working on releasing a PyPI library for scraping Letterboxd data and a Java package. Update (8/8/19): Moved away from Selenium in favor of Beautiful Soup; significantly increases speed of data scraping.

Getting Started

It is recommended to use a virtualenv to install necessary dependencies. The following illustrates how to do so:

Installing

Pre-setup:

You will need to have virtualenv and pip (the Python package manager) installed

apt-get install python3-pip

Then, install pip3

pip3 install virtualenv

Step (i)

Create and activate a virtualenv

virtualenv some-name

Step (ii)

Install the necessary dependencies

pip3 install -r requirements.txt

Running and utilizing

To run the program:

python3 -u main.py

License

This project is licensed under the MIT License - see the LICENSE file for details