Infinispan Pokemon

This demo uses the query and REST capabilities of Infinispan Server to help you battle Pokemons!

Data is ingested as JSON, and internally stored efficiently as Protobuf. Both query and retrieval happens using JSON taking advantage of the mapping between Protobuf and JSON present in Infinispan

Requirements

• Download the dataset archive.zip from https://www.kaggle.com/rounakbanik/pokemon
• Make sure Python 3 installed python --version

Running

• Start Infinispan Server:

  docker run -it --name infinispan-server -p 11222:11222 -e "USER=user" -e "PASS=user" infinispan/server:13.0

• Prepare data

  python3 prepare-data.py

• Creating an indexed cache

  curl -u user:user --digest -H "Content-Type: application/json" -d '{"distributed-cache":{"mode":"SYNC","encoding":{"key":{"media-type":"application/x-protostream"},"value":{"media-type":"application/x-protostream"}},"indexing":{"indexed-entities":["Pokemon"]}}}' http://127.0.0.1:11222/rest/v2/caches/pokemon

• Register the protobuf schema

  curl -u user:user --digest --data-binary @./pokemon.proto http://127.0.0.1:11222/rest/v2/schemas/pokemon.proto

• Ingest data

  ./ingest-data.sh

Querying

Example queries:

• Get Pokemon by key (name)

  http://localhost:11222/rest/v2/caches/pokemon/Whismur

• Get all Pokemons:

  from Pokemon

• Count Pokemons by generation:

  select count(p.name) from Pokemon group by generation

• Do a full text search on the name

  from Pokemon where name:'pikachu'

• Select top 5 Pokemons that can better withstand fire:

  from Pokemon order by against_fire