superstream-confluent-kafka-beta 2.4.0.3

Confluent's Python client for Apache Kafka

Superstream-Kafka

Configuration for Superstream SDK

To leverage the full capabilities of the Superstream SDK, it is essential to set the environment variables provided in the table below before initializing the SDK. Without setting-up the environment variables, the SDK will function as a standard Kafka SDK.

Environment Variable	Default	Required	Description
SUPERSTREAM_HOST	-	Yes	Specify the host URL of the Superstream service to connect to the appropriate Superstream environment.
SUPERSTREAM_TOKEN	-	No	This authentication token is required when the engine is configured to work with local authentication, to securely access the Superstream services.
SUPERSTREAM_TAGS	Empty string	No	Set this variable to tag the client. This value of this variable should be a valid JSON string.
SUPERSTREAM_DEBUG	False	No	Set this variable to true to enable Superstream logs. By default, there will not be any Superstream related logs.
SUPERSTREAM_RESPONSE_TIMEOUT	3000	No	Set this variable to specify the timeout in milliseconds for the Superstream service response.

[!IMPORTANT]
Ensure that these environment variables are properly configured in your system to fully utilize the enhanced features offered by Superstream SDK.

Confluent's Python Client for Apache Kafka^TM

confluent-kafka-python provides a high-level Producer, Consumer and AdminClient compatible with all Apache Kafka^TM brokers >= v0.8, Confluent Cloud and Confluent Platform. The client is:

• Reliable - It's a wrapper around librdkafka (provided automatically via binary wheels) which is widely deployed in a diverse set of production scenarios. It's tested using the same set of system tests as the Java client and more. It's supported by Confluent.

• Performant - Performance is a key design consideration. Maximum throughput is on par with the Java client for larger message sizes (where the overhead of the Python interpreter has less impact). Latency is on par with the Java client.

• Future proof - Confluent, founded by the creators of Kafka, is building a streaming platform with Apache Kafka at its core. It's high priority for us that client features keep pace with core Apache Kafka and components of the Confluent Platform.

Usage

For a step-by-step guide on using the client see Getting Started with Apache Kafka and Python.

Aditional examples can be found in the examples directory or the confluentinc/examples github repo, which include demonstration of:

• Exactly once data processing using the transactional API.
• Integration with asyncio.
• (De)serializing Protobuf, JSON, and Avro data with Confluent Schema Registry integration.
• Confluent Cloud configuration.

Also refer to the API documentation.

Finally, the tests are useful as a reference for example usage.

Basic Producer Example

from confluent_kafka import Producer

p = Producer({'bootstrap.servers': 'mybroker1,mybroker2'})

def delivery_report(err, msg):
    """ Called once for each message produced to indicate delivery result.
        Triggered by poll() or flush(). """
    if err is not None:
        print('Message delivery failed: {}'.format(err))
    else:
        print('Message delivered to {} [{}]'.format(msg.topic(), msg.partition()))

for data in some_data_source:
    # Trigger any available delivery report callbacks from previous produce() calls
    p.poll(0)

    # Asynchronously produce a message. The delivery report callback will
    # be triggered from the call to poll() above, or flush() below, when the
    # message has been successfully delivered or failed permanently.
    p.produce('mytopic', data.encode('utf-8'), callback=delivery_report)

# Wait for any outstanding messages to be delivered and delivery report
# callbacks to be triggered.
p.flush()

For a discussion on the poll based producer API, refer to the Integrating Apache Kafka With Python Asyncio Web Applications blog post.

Basic Consumer Example

from confluent_kafka import Consumer

c = Consumer({
    'bootstrap.servers': 'mybroker',
    'group.id': 'mygroup',
    'auto.offset.reset': 'earliest'
})

c.subscribe(['mytopic'])

while True:
    msg = c.poll(1.0)

    if msg is None:
        continue
    if msg.error():
        print("Consumer error: {}".format(msg.error()))
        continue

    print('Received message: {}'.format(msg.value().decode('utf-8')))

c.close()

Basic AdminClient Example

Create topics:

from confluent_kafka.admin import AdminClient, NewTopic

a = AdminClient({'bootstrap.servers': 'mybroker'})

new_topics = [NewTopic(topic, num_partitions=3, replication_factor=1) for topic in ["topic1", "topic2"]]
# Note: In a multi-cluster production scenario, it is more typical to use a replication_factor of 3 for durability.

# Call create_topics to asynchronously create topics. A dict
# of <topic,future> is returned.
fs = a.create_topics(new_topics)

# Wait for each operation to finish.
for topic, f in fs.items():
    try:
        f.result()  # The result itself is None
        print("Topic {} created".format(topic))
    except Exception as e:
        print("Failed to create topic {}: {}".format(topic, e))

Thread Safety

The Producer, Consumer and AdminClient are all thread safe.

Install

Install self-contained binary wheels

$ pip install confluent-kafka

NOTE: The pre-built Linux wheels do NOT contain SASL Kerberos/GSSAPI support. If you need SASL Kerberos/GSSAPI support you must install librdkafka and its dependencies using the repositories below and then build confluent-kafka using the instructions in the "Install from source" section below.

Install from source

For source install, see the Install from source section in INSTALL.md.

Broker Compatibility

The Python client (as well as the underlying C library librdkafka) supports all broker versions >= 0.8. But due to the nature of the Kafka protocol in broker versions 0.8 and 0.9 it is not safe for a client to assume what protocol version is actually supported by the broker, thus you will need to hint the Python client what protocol version it may use. This is done through two configuration settings:

• broker.version.fallback=YOUR_BROKER_VERSION (default 0.9.0.1)
• api.version.request=true|false (default true)

When using a Kafka 0.10 broker or later you don't need to do anything (api.version.request=true is the default). If you use Kafka broker 0.9 or 0.8 you must set api.version.request=false and set broker.version.fallback to your broker version, e.g broker.version.fallback=0.9.0.1.

More info here: https://github.com/edenhill/librdkafka/wiki/Broker-version-compatibility

SSL certificates

If you're connecting to a Kafka cluster through SSL you will need to configure the client with 'security.protocol': 'SSL' (or 'SASL_SSL' if SASL authentication is used).

The client will use CA certificates to verify the broker's certificate. The embedded OpenSSL library will look for CA certificates in /usr/lib/ssl/certs/ or /usr/lib/ssl/cacert.pem. CA certificates are typically provided by the Linux distribution's ca-certificates package which needs to be installed through apt, yum, et.al.

If your system stores CA certificates in another location you will need to configure the client with 'ssl.ca.location': '/path/to/cacert.pem'.

Alternatively, the CA certificates can be provided by the certifi Python package. To use certifi, add an import certifi line and configure the client's CA location with 'ssl.ca.location': certifi.where().

License

Apache License v2.0

KAFKA is a registered trademark of The Apache Software Foundation and has been licensed for use by confluent-kafka-python. confluent-kafka-python has no affiliation with and is not endorsed by The Apache Software Foundation.

Developer Notes

Instructions on building and testing confluent-kafka-python can be found here.

Confluent Cloud

For a step-by-step guide on using the Python client with Confluent Cloud see Getting Started with Apache Kafka and Python on Confluent Developer.