Query recent twitter account followings

This app query what a selected twitter account has been following recently, saves into a Postgres DB and send it to a discord bot. Interacts with twitter API and discord Webhook

├── databases
├── main.py                 
├── database.ini            # local DB config
├── .env                    # local env variables
└── test_data.json          # For development and API limits query a json 
                            # file containing recent followers used in 
                            # connect_to_endpoint_test in main()

Packages

• requests
• discord: to interact with discord api
• decouple: for local development
• psycopg2: for postgres

Local Development

Create an .env file containing these variables

• WEBHOOK = discord bot webhook
• MAX_RESULTS = limit the number of recent follows queried every time
• MESSAGE_TEMPLATE = Custom message to discord bot, like "Hello! This is the daily digest of followings"
• BEARER_TOKEN= Bearer token obtained on your own twitter app that allows interacting with the api. You need to create an app for this check https://developer.twitter.com -DATABASE_URL=heroku database URL

Initialising

In root folder, create an accounts.csv file containing 2 columns, column 1 is twitter ID and 2nd is twitter username. These are the accounts you want to retrieve followings from Example:

accounts.csv

123124129,elonmusk
121241242,ladygaga

You can create a test_data.json file to query dummy data. To use it, modify the line in main.py

json_response = connect_to_endpoint(url, params)
# change it to
json_response = connect_to_endpoint_test(url, params)

Data queried from recent followings will follow this structure:

{
    "data": [
        {
            "created_at": "2018-02-28T09:50:25.000Z",
            "id": "123456789",
            "name": "Elon Musk",
            "username": "elonmusk"
        },
        ...
    ]
}

database.ini file

To use local postgres database, create a database.ini config file:

[postgresql]
host=localhost
database=dbname
user=your-username
password=your-password

Running the script

Just type python main.py - will automatically connect to heroku DB URL set in environment variables file, create a account table and select data from accounts.csv file and insert it into the table

TO USE LOCAL DATABASE -> run python main.py dev with that extra argument

Data retrieved from twitter API for each user will be saved into a postgres table separately. E.g new_followers_ladygaga

Running on Heroku

You can set it as a cron scheduled job and run this script once or twice per day like a daily digest. Just add Heroku Scheduler as an addon and run python main.py for the script.