/pirula-time

Measures the mean duration for Pirula's videos

Primary LanguageHTMLMIT LicenseMIT

pirula-time

Measures the average duration of Pirula's videos

This is intended to be run on AWS.

The scraper is a Node.js 6.10 AWS Lambda serverless function.

The webapp is a Python 3.4 Flask app.

The scraper is called hourly, gathers youtube data from the Google APIs, and stores it on DynamoDB; which is then read by the webapp.

The subjective assessment of each relevant value is determined by the yaml files in the webapp/subjectives directory.

Contributions/Contribuições

You're welcome to edit the yaml files if you have any funny jokes.

Você pode editar os arquivos yaml se tiver alguma piada legal.

The files can be edited directly on the browser, no need to clone or anything.

Os arquivos podem ser editados usando o navegador, sem precisar clonar nada.