Cornell Men's Varsity Basketball Statistics Application

This repository contains all of the source code for an application for the use of the Cornell University men's basketball team coaching staff to more efficiently and intuitively parse through data to make better informed administrative decisions.

Quick Links

• Setup
  • Developers
  • Requirements
  • Installation
  • Users
• High-Level Implementation
• Low-Level Implementation Details

Setup

Developers

TODO: Add information on setting PROD=True or False

Requirements

1. If you are a developer and would like to make changes to the application, you will need Python 2.7 or higher (3.6+ is preferred). To install all the necessary Python libraries, you can navigate to the cs-5150-basketball-project directory and pip install -r requirements.txt.
2. You will also need npm.

Installation

To get a copy of the repository on your local device, begin by git cloneing the repository. Once this is complete, navigate into the app directory and run npm install. After the install process is complete, you can run npm start to launch the application. Changing code and refreshing the application window (or relaunching the application by again using npm start) will allow you to see your changes in action before you compile the application. In order to compile the application, navigate to the app directory and run electron-forge make. The compiled application can then be found in app/out/.

Users

End users simply need to download the pre-compiled app, one version of which can be found [here](todo:insert link).

High-Level Implementation

The application functions by parsing XML game data files provided by the Cornell coaching staff and JSON game data parsed from the ESPN website. The parsed data is then used to populate individual xml and json SQLite databases, which can be found in app/src/python/backend/. The database schema definition and initialization functions can be found in app/db.py, parsing functions can be found in app/parser.py (for XML files) and app/parse_json.py (for JSON files). Finally, the database population occurs in and app/populate_db.py.

app/xml_downloader.py fetches new XML data from a Google Drive account (the necessary credentials should be privately handed over to the coaching staff). Similarly, app/espn_scraper.py and app/espn_to_db.py contains code to fetch new JSON data.

Dynamically querying the database is handled in app/data_retriever.py, while static queries (like obtaining a set of team names or player names) occurs in app/auto_complete.py.

Information retrieved from the backend is passed on to the middle stack files for formatting and calculating advanced statistics. The advanced statistics formulas can be found in app/advanced_stat.py, and the rest of the middle stack operations take place in app/data_manager.py.

Finally, information is presented to the user in the frontend HTML/CSS in the various .html, .css, and .ts files present in app/src and app/src/assets (the latter of which contains CSS and JavaScript dependencies.)

Low-Level Implementation