This is an application to monitor youtube channel data and collect information on how likes, dislikes, comments and favorites change overtime for each video in a channel. Keep in mind that google youtube api only allows 10000 requests per day. So monitoring large channels are not possible.

The app is based on an excelent flask-base boilerplate project.

The application requires to be run on python 3.6. Install the dependancies via pipenv install. Install redis via your platform way. Set the required environment variables (you can store them in config.env file in ./flask-base folder):

DATABASE_URL=postgresql://postgres:mysecretpassword@localhost:5432/youtube
SQLALCHEMY_DATABASE_URL=postgresql://postgres:mysecretpassword@localhost:5432/youtube
DEV_DATABASE_URL=postgresql://{user}:{password}@{host}:{port}/{database}
SECRET_KEY={seed for random functions}
APP_NAME=Youtube-data-collector
ADMIN_EMAIL=admin@admin.adm
ADMIN_PASSWORD=password
YOUTUBE_API_KEY={your youtube api key}

if you don't set the DATABASE_URL the default sqlite database will be used in flask-base/data-dev.sqlite

after all this is done, initialize the database:

$ cd flask-base
$ python manage.py recreate_db
$ python manage.py setup_dev

Start monitoring some youtube channel:

$ python manage.py start_scraping --channel-id=UCwUizOU8pPWXdXNniXypQEQ

Run the app using $ honcho start -f Local this will run flask in gunicorn server, start 1 redis worker and start periodically monitoring youtube channels.

Write tests and implement better exception handling.