Python Stream Processing

> pip install bytewax

> brew install waxctl

# dataflow.py
flow = Dataflow()
flow.input(
    "inp",
    KafkaInputConfig(brokers=["localhost:9092"], topic="input_topic")
)
flow.capture(
    KafkaOutputConfig(brokers=["localhost:9092"], topic="output_topic")
)

# dataflow.py
def deserialize():
  return x.decode()

def calc_avg():

cc = SystemClockConfig()
wc = TumblingWindowConfig(length=datetime.timedelta(seconds=5))

flow.map(deserialize)
flow.fold_window("5sec_avg", cc, wc, calc_avg)

> python dataflow.py

> waxctl dataflow deploy my-script.py \
  --name=cluster \
  --processes=5
  --workers=2