Afkak: Twisted Python Kafka Client

Afkak is a Twisted-native Apache Kafka client library. It provides support for:

• Producing messages, with automatic batching and optional compression.
• Consuming messages, with group coordination and automatic commit.

Learn more in the documentation, download from PyPI, or review the contribution guidelines. Please report any issues on GitHub.

Status

Afkak supports these Pythons:

• CPython 2.7
• CPython 3.5, 3.6, 3.7, and 3.8
• PyPy and PyPy3 6.0+

We aim to support Kafka 1.1.x and later. Integration tests are run against these Kafka broker versions:

• 0.9.0.1
• 1.1.1

Testing against 2.0.0 is planned (see #45).

Newer broker releases will generally function, but not all Afkak features will work on older brokers. In particular, the coordinated consumer won’t work before Kafka 0.9.0.1. We don’t recommend deploying such old releases anyway, as they have serious bugs.

Usage

High level

Note: This code is not meant to be runnable. See producer_example and consumer_example for runnable example code.

from afkak.client import KafkaClient
from afkak.consumer import Consumer
from afkak.producer import Producer
from afkak.common import (OFFSET_EARLIEST, PRODUCER_ACK_ALL_REPLICAS,
    PRODUCER_ACK_LOCAL_WRITE)

kClient = KafkaClient("localhost:9092")

# To send messages
producer = Producer(kClient)
d1 = producer.send_messages("my-topic", msgs=[b"some message"])
d2 = producer.send_messages("my-topic", msgs=[b"takes a list", b"of messages"])
# To get confirmations/errors on the sends, add callbacks to the returned deferreds
d1.addCallbacks(handleResponses, handleErrors)

# To wait for acknowledgements
# PRODUCER_ACK_LOCAL_WRITE : server will wait till the data is written to
#                         a local log before sending response
# [ the default ]
# PRODUCER_ACK_ALL_REPLICAS : server will block until the message is committed
#                            by all in sync replicas before sending a response
producer = Producer(kClient,
                    req_acks=Producer.PRODUCER_ACK_LOCAL_WRITE,
                    ack_timeout=2000)

responseD = producer.send_messages("my-topic", msgs=[b"message"])

# Using twisted's @inlineCallbacks:
responses = yield responseD
if response:
    print(response[0].error)
    print(response[0].offset)

# To send messages in batch: You can use a producer with any of the
# partitioners for doing this. The following producer will collect
# messages in batch and send them to Kafka after 20 messages are
# collected or every 60 seconds (whichever comes first). You can
# also batch by number of bytes.
# Notes:
# * If the producer dies before the messages are sent, the caller would
# * not have had the callbacks called on the send_messages() returned
# * deferreds, and so can retry.
# * Calling producer.stop() before the messages are sent will
# errback() the deferred(s) returned from the send_messages call(s)
producer = Producer(kClient, batch_send=True,
                    batch_send_every_n=20,
                    batch_send_every_t=60)
responseD1 = producer.send_messages("my-topic", msgs=[b"message"])
responseD2 = producer.send_messages("my-topic", msgs=[b"message 2"])

# To consume messages
# define a function which takes a list of messages to process and
# possibly returns a deferred which fires when the processing is
# complete.
def processor_func(consumer, messages):
    #  Store_Messages_In_Database may return a deferred
    result = store_messages_in_database(messages)
    # record last processed message
    consumer.commit()
    return result

the_partition = 3  # Consume only from partition 3.
consumer = Consumer(kClient, "my-topic", the_partition, processor_func)
d = consumer.start(OFFSET_EARLIEST)  # Start reading at earliest message
# The deferred returned by consumer.start() will fire when an error
# occurs that can't handled by the consumer, or when consumer.stop()
# is called
yield d

consumer.stop()
kClient.close()

Keyed messages

from afkak.client import KafkaClient
from afkak.producer import Producer
from afkak.partitioner import HashedPartitioner, RoundRobinPartitioner

kafka = KafkaClient("localhost:9092")

# Use the HashedPartitioner so that the producer will use the optional key
# argument on send_messages()
producer = Producer(kafka, partitioner_class=HashedPartitioner)
producer.send_messages("my-topic", "key1", [b"some message"])
producer.send_messages("my-topic", "key2", [b"this method"])

Low level

from afkak.client import KafkaClient
kafka = KafkaClient("localhost:9092")
req = ProduceRequest(topic="my-topic", partition=1,
    messages=[KafkaProtocol.encode_message(b"some message")])
resps = afkak.send_produce_request(payloads=[req], fail_on_error=True)
kafka.close()

resps[0].topic      # b"my-topic"
resps[0].partition  # 1
resps[0].error      # 0 (hopefully)
resps[0].offset     # offset of the first message sent in this request

Install

Afkak releases are available on PyPI.

Because the Afkak dependencies Twisted and python-snappy have binary extension modules you will need to install the Python development headers for the interpreter you wish to use:

Then Afkak can be installed with pip as usual:

License

Copyright 2013, 2014, 2015 David Arthur under Apache License, v2.0. See LICENSE

Copyright 2014, 2015 Cyan, Inc. under Apache License, v2.0. See LICENSE

Copyright 2015, 2016, 2017, 2018, 2019 Ciena Corporation under Apache License, v2.0. See LICENSE

This project began as a port of the kafka-python library to Twisted.

See AUTHORS.md for the full contributor list.

Tests

In order to run Afkak's tests, you need to install the dependencies as specified in the install section.

The Afkak test suite uses Tox to execute the tests in all the supported Python versions. The preferred method to run the tests is to install Tox in a virtual environment before running the tests:

make venv

Testing Strategy

Afkak has two types of tests:

• Unit tests — unit tests are fast tests. They don't do I/O.

• Integration tests — tests that run against a real Kafka broker.

Run the unit tests

To run all unit tests in all the supported Python versions (requires all the versions to be installed in the system where the tests will run):

make toxu

Alternatively, you might want to run unit tests in a list of specific Python versions:

.env/bin/tox -e py27-unit-snappy,py35-unit-snappy

It is recommended for contributors to run unit tests in at least Python 2.7 and one Python 3 version before submitting a pull request.

Run the integration tests

The integration tests will actually start up real local ZooKeeper instance and Kafka brokers, and send messages in using the client.

The makefile knows how to download several versions of Kafka. This will run just the integration tests against Kafka 1.1.1:

KAFKA_VER=1.1.1 make toxi

Run all the tests against the default Kafka version

make toxa

Run the integration tests against all the Kafka versions the Makefile knows about

make toxik