dselman / tweet-saver

Saves Tweets to a database.

Tweet Saver

This application monitors a Twitter stream for tweets which include keywords. The tweets containing the keywords are then filtered using some exclusion keywords and then persisted into a Postgres database for subsequent offline analysis.

There is also a frontend web-application which may be used to monitor the filtered tweets, displaying them in real-time on a web-page via a websocket connection.

Database Schema

The file init.sql defines the database schema used to persist the Tweets.

Environment Variables

• TWITTER_CONSUMER_KEY : Twitter API credentials
• TWITTER_CONSUMER_SECRET : Twitter API credentials
• TWITTER_ACCESS_TOKEN_KEY : Twitter API credentials
• TWITTER_ACCESS_TOKEN_SECRET : Twitter API credentials
• TRACK : Keywords to track, passed to Twitter statuses/filter API
• TRACK_RT : When true, include retweets
• EXCLUDE : A comma seperated list of words. If any of these words appear in a Tweet it is excluded
• DELETE_ALL : WARNING! when true all existing database data is deleted
• LOGDB : When true Tweets are stored in the database
• LOG_FILE : When true Tweets are stored in a file
• PORT : Port number used for the HTTP server
• NODE_ENV : set the "production" to use an SSL database connection
• PGUSER : Postgres user name
• PGPASSWORD : Postgres password
• PGHOST : Postgres host
• PGPORT : Postgres port
• PGDATABASE : Postgres database name
• DATABASE_URL : Postgres database URL (used in production)
• PUBLIC_URL : Public URL for the frontend application