Introduction

JSON Streams is a low-code event streaming platform, where messages are exchanged in JSON format. With the platform you can:

• Manage state in an event-driven way.
• Transform event streams.
• Validate event streams.
• Join event streams.
• Merge multiple event streams into one.

The languages you need to work with it are the MongoDB Operator language and JSLT. The backbone of an event stream is a MongoDB aggregation pipeline. This is a streaming implementation of the aggregation language. State is managed with aggregates that have reducers written in JSLT, which is a JSON transformation language. Reducers can also be written as pipelines. A reducer receives a command and the current state of an aggregate instance. It returns the new state. The difference is saved and published as an event.

The platform comes with a Docker image. Multiple instances will form a cluster and divide the work amongst them. It needs Kafka and MongoDB to run. It can auto-scale based on message lag constraints.