This the source code repo of vaccineupdates.ca site.

• sqlite is used as database. This is the sql for tweet table

CREATE TABLE tweet (
	id INTEGER NOT NULL, 
	tweet_id INTEGER, 
	tweet_text VARCHAR, 
	province VARCHAR, 
	age_groups VARCHAR, 
	cities VARCHAR, 
	"FSAs" VARCHAR, 
	created_at DATETIME DEFAULT (CURRENT_TIMESTAMP), 
	ingested_at DATETIME DEFAULT (CURRENT_TIMESTAMP), 
	PRIMARY KEY (id), 
	UNIQUE (tweet_id)
);

It's also possible easily to switch to other databases. Plese see SqlAlchemy supported engines how to configure other databases.

• litestream is used to back up sqlite db to s3. It's not necessary if you use other database.
• Please create a config.yml similar to below

# For sqlite
SQLALCHEMY_DATABASE_URI = 'sqlite:////paths-to-db.sqlite'
# For postgres
# SQLALCHEMY_DATABASE_URI = 'postgresql://local_user:mysecretpassword@localhost:5432/local_database'

TWITTER_CONSUMER_KEY = ''
TWITTER_CONSUMER_SECRET = ''

TWITTER_ACCESS_TOKEN = ''
TWITTER_ACCESS_TOKEN_SECRET = ''

and fill up necessary secrets to run the app.

• Build docker image: docker build -t vax-searcher .
• Run the batch sync docker run vax-searcher python stream_sync.py -d 3. -d 3 means last three days tweets will be synced and processed.
• Run the real-time sync docker run vax-searcher python batch_sync.py. This will use Twitter streaming API to sync Vaccine Hunters tweets in real-time.