Read from multiple Kafka topics to Dataflow

Shows how to create a Dataflow pipeline that reads from multiple Kafka topics and performs different business logic based on the topic name.

Explore further

For detailed documentation that includes this code sample, see the following:

• Read from Apache Kafka to Dataflow

Code sample

Java

To authenticate to Dataflow, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

import java.util.List;
import org.apache.beam.sdk.Pipeline;
import org.apache.beam.sdk.PipelineResult;
import org.apache.beam.sdk.io.TextIO;
import org.apache.beam.sdk.io.kafka.KafkaIO;
import org.apache.beam.sdk.options.Description;
import org.apache.beam.sdk.options.PipelineOptionsFactory;
import org.apache.beam.sdk.options.StreamingOptions;
import org.apache.beam.sdk.transforms.Filter;
import org.apache.beam.sdk.transforms.MapElements;
import org.apache.beam.sdk.values.TypeDescriptors;
import org.apache.kafka.common.serialization.LongDeserializer;
import org.apache.kafka.common.serialization.StringDeserializer;
import org.joda.time.Duration;
import org.joda.time.Instant;

public class KafkaReadTopics {

  public static Pipeline createPipeline(Options options) {
    String topic1 = options.getTopic1();
    String topic2 = options.getTopic2();

    // Build the pipeline.
    var pipeline = Pipeline.create(options);
    var allTopics = pipeline
        .apply(KafkaIO.<Long, String>read()
            .withTopics(List.of(topic1, topic2))
            .withBootstrapServers(options.getBootstrapServer())
            .withKeyDeserializer(LongDeserializer.class)
            .withValueDeserializer(StringDeserializer.class)
            .withMaxReadTime(Duration.standardSeconds(10))
            .withStartReadTime(Instant.EPOCH)
        );

    // Create separate pipeline branches for each topic.
    // The first branch filters on topic1.
    allTopics
        .apply(Filter.by(record -> record.getTopic().equals(topic1)))
        .apply(MapElements
            .into(TypeDescriptors.strings())
            .via(record -> record.getKV().getValue()))
        .apply(TextIO.write()
            .to(topic1)
            .withSuffix(".txt")
            .withNumShards(1)
        );

    // The second branch filters on topic2.
    allTopics
        .apply(Filter.by(record -> record.getTopic().equals(topic2)))
        .apply(MapElements
            .into(TypeDescriptors.strings())
            .via(record -> record.getKV().getValue()))
        .apply(TextIO.write()
            .to(topic2)
            .withSuffix(".txt")
            .withNumShards(1)
        );
    return pipeline;
  }
}

Python

To authenticate to Dataflow, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

import argparse

import apache_beam as beam

from apache_beam.io.kafka import ReadFromKafka
from apache_beam.io.textio import WriteToText
from apache_beam.options.pipeline_options import PipelineOptions


def read_from_kafka() -> None:
    # Parse the pipeline options passed into the application. Example:
    #   --bootstrap_server=$BOOTSTRAP_SERVER --output=$STORAGE_BUCKET --streaming
    # For more information, see
    # https://beam.apache.org/documentation/programming-guide/#configuring-pipeline-options
    class MyOptions(PipelineOptions):
        @staticmethod
        def _add_argparse_args(parser: argparse.ArgumentParser) -> None:
            parser.add_argument('--bootstrap_server')
            parser.add_argument('--output')

    options = MyOptions()
    with beam.Pipeline(options=options) as pipeline:
        # Read from two Kafka topics.
        all_topics = pipeline | ReadFromKafka(consumer_config={
                "bootstrap.servers": options.bootstrap_server
            },
            topics=["topic1", "topic2"],
            with_metadata=True,
            max_num_records=10,
            start_read_time=0
        )

        # Filter messages from one topic into one branch of the pipeline.
        (all_topics
            | beam.Filter(lambda message: message.topic == 'topic1')
            | beam.Map(lambda message: message.value.decode('utf-8'))
            | "Write topic1" >> WriteToText(
                file_path_prefix=options.output + '/topic1/output',
                file_name_suffix='.txt',
                num_shards=1))

        # Filter messages from the other topic.
        (all_topics
            | beam.Filter(lambda message: message.topic == 'topic2')
            | beam.Map(lambda message: message.value.decode('utf-8'))
            | "Write topic2" >> WriteToText(
                file_path_prefix=options.output + '/topic2/output',
                file_name_suffix='.txt',
                num_shards=1))

What's next

To search and filter code samples for other Google Cloud products, see the Google Cloud sample browser.