Arroyo is a library to build streaming applications that consume from and produce to Kafka.
It relies on the confluent_kafka python library, which itself relies on librdkafka.
Arroyo provides mainly three functionalities:
A set of abstractions inspired to common messaging applications patterns.
Some abstractions to simplify offset management and rebalancing.
An in memory broker abstraction to simplify writing unit tests.