mcalavera81/spring-boot-reactive-event-driven-kafka-mongo

Description

Stock prices processing components:

• service.ApiStockService: Stateful service. On each call, we get a batch of pseudo-random-generated stock prices. Each call also moves forward the current time.

• message.StockTickSupplierConfig: Using a mock API service.ApiStockService, it generates an infinite sequence of stock prices with a stock-update.frequency.seconds frequency and pushes these prices into the testtock Kafka topic.

• message.StockTickConsumerConfig: Consumes the stock prices from the testtock Kafka topic and stores them in a raw format in the stockTick Mongodb collection using service.StockTickService

• message.CandleStickConsumerConfig: Consumes the stock prices from the testtock Kafka topic and calls service.CandleStickService to update the day-based and hour-based CandleSticks

API endpoints:

• Company endpoints:
  • GET /api/v1/companies: list all companies with stock prices. It is defined statically in initializer.MasterDataLoader, instead of being defined by the incoming stock prices. It is a coroutine-based reactive endpoint.
• StockTick endpoints:
  • GET /api/v1/ticks/changelog: Server pushes new stock prices being stored in MongoDB using Server-sent events. Only for debugging purposes.
  • GET /api/v1/ticks: retrieves all stock prices. Just for debugging in non-prod envs. due to the scale of data.
  • POST /api/v1/ticks/delete-all: deletes all stock prices from database. Only for debugging purposes.
• CandleStick endpoints:
  • GET /api/v1/candles/changelog: Server pushes candle updates being stored in MongoDB using Server-sent events. Only for debugging purposes.
  • GET /api/v1/candles: retrieves all candles, optionally filtered by query params ticker and type. Valid values for ticker: AMZN,AAPL,GOOG,META. Valid values for type: DAY,ONE_HOUR.
  • POST /api/v1/candles/delete-all: deletes all candles from database. Only for debugging purposes.

Run

Requirements

• Install docker and docker-compose
• JVM

Steps

1. Run docker-compose up
2. mvn spring-boot:run
3. curl http://localhost:8080/api/v1/candles/?ticker=META&type=ONE_HOUR

[
  {
    "id": "63849bcad4c41217e8e109e5",
    "type": "ONE_HOUR",
    "ticker": "META",
    "timestamp": "2022-11-28T15:00:00",
    "high": 180,
    "low": 105
  },
  ...
]

TODO

• Introduce tests. Refactor app code as needed for testability.
• Split code into modules: API, Stock API Consumer, Prices Processor(s)
• Implement Weekly Candlesticks.
• Better input handling and validation.
• Add Open and Close prices for Candlesticks.