Giftly

Example application that tracks what is popular and recommends. Work in progress.

Use files from setup folder to compose environment. Use configuration repository to configure required paths. Find cassandra.sql to see how to setup cassandra and where to check for the data.

Quick start with docker

Build and deploy images to docker host

./docker-build.sh

Run services with using docker-compose

./docker-compose up

gift actions

Add gift

curl -i -X POST -H "Content-Type: application/json" -d '{"name":"start wars t-shirt medium orange"}' http://$GIFTLY_GATEWAY_HOST:8080/gifts/

curl http://$GIFTLY_GATEWAY_HOST:8080/gifts/names

Use elastic search directly:

curl "http://$ELASTIC_HOST:9200/giftly/_search?q=name:orange"

like actions

Start ingesting data:

./spark-submit --class org.koko.like.LikesIngest /home/ec2-user/giftly-data-stream-0.0.1-SNAPSHOT.jar $CASSANDRA_HOST $SPARK_HOST $KAFKA_BROKERS

Place some likes:

curl -i -X POST -H "Content-Type: application/json" -d '{"giftId":"gift1", "userId":"user1", "like_time":"2016-01-01 20:10:12", "liked":"1"}' http://$GIFTLY_GATEWAY_HOST:8080/likes/
curl -i -X POST -H "Content-Type: application/json" -d '{"giftId":"gift2", "userId":"user1", "like_time":"2016-01-01 20:10:42", "liked":"1"}' http://$GIFTLY_GATEWAY_HOST:8080/likes/
curl -i -X POST -H "Content-Type: application/json" -d '{"giftId":"gift1", "userId":"user2", "like_time":"2016-01-01 20:20:12", "liked":"1"}' http://$GIFTLY_GATEWAY_HOST:8080/likes/

Run daily batch:

./spark-submit --class org.koko.like.LikesDaily /home/ec2-user/giftly-data-stream-0.0.1-SNAPSHOT.jar "2016-01-01" $CASSANDRA_HOST $SPARK_HOST

Check current likes, produced top and daily stats in Cassandra.