TwitchChatStats

Not maintained anymore. See https://www.github.com/agapic/Twitch-Streamer

Uses Apache Spark Streaming to stream Twitch.tv chat messages real-time and transforms the data into various leaderboards; currently supports the top 50 words and top 50 active channels on Twitch.tv. The data is persisted in memory using Redis, and the results can be viewed over a web application built with Node/React.

More to come soon.

Here's a demo of the web app. While the GIF is custom made, the screenshots were taken immediately after the previous one.

TODO: create a script to automate the instructions below.

Quick bootstrapping to get running:

1. You can build the source code using sbt assembly while in the sbt folder, but a fully compiled JAR is provided for convenience.

2. Download Redis (https://redis.io/download). Ensure that you download the stable version. Run it using redis-server.

3. Download Node.js https://nodejs.org/en/download/)

4. Download Spark 2.0.0 (http://d3kbcqa49mib13.cloudfront.net/spark-2.0.0-bin-hadoop2.7.tgz). Add spark/bin to your PATH.

5. Register your Twitch.tv application for your client_id and client_secret here. (https://blog.twitch.tv/client-id-required-for-kraken-api-calls-afbb8e95f843#.7hu9s24ub)

6. Sign up for a Twitch.tv username to populate the username field. Then, to populate the password field, go to this website to automatically generate an oauth token (https://twitchapps.com/tmi/)

7. Run the spark application with spark-submit TwitchChat-assembly-1.0.jar

8. In the web directory, run npm install followed by npm start.

9. View the results at http://localhost:3002!