使用 ADLS Gen2 中的属性通过 Apache Flink® 扩充 Apache Kafka® 中的事件

2024-09-20

重要

AKS 上的 Azure HDInsight 已于 2025 年 1 月 31 日停用。了解更多信息，请参阅此公告。

需要将工作负荷迁移到 Microsoft Fabric 或等效的 Azure 产品，以避免工作负荷突然终止。

重要

此功能目前以预览版提供。 Microsoft Azure 预览版的附加使用条款包括适用于 beta 版、预览版或尚未正式发布的 Azure 功能的更多法律条款。有关此特定预览版的信息，请参阅 AKS 上的 Azure HDInsight 预览信息。有关问题或功能建议，请在 AskHDInsight 上提交请求，并提供详细信息。请关注我们以获取关于 Azure HDInsight 社区的更多更新。

在本文中，您可以学习如何使用 Flink 流，将 Kafka 的流与 ADLS Gen2 上的表进行联接，从而丰富实时事件。我们使用 Flink 流式处理 API 将 HDInsight Kafka 中的事件与 ADLS Gen2 中的属性联接。进一步，我们将带有已联接属性的事件写入另一个 Kafka 主题。

先决条件

基于 AKS 的 HDInsight 上的 Flink 群集
在 HDInsight 上的 Kafka 群集
- 确保按照使用 Kafka on HDInsight 中的描述对网络设置进行处理，以确保 AKS 上的 HDInsight 和 HDInsight 群集处于同一 VNet 中。
对于本次演示，我们使用 Windows VM 作为 Maven 项目的开发环境，该项目位于 AKS 上 HDInsight 所在的同一 VNet 中。

Kafka 主题准备

我们正在创建一个名为 user_events的主题。

目的是使用 Flink 从 Kafka 主题读取实时事件流。每个事件都有以下字段：
```
user_id,
item_id, 
type, 
timestamp, 
```

Kafka 3.2.0

/usr/hdp/current/kafka-broker/bin/kafka-topics.sh --create --replication-factor 2 --partitions 3 --topic user_events --bootstrap-server wn0-contsk:9092
/usr/hdp/current/kafka-broker/bin/kafka-topics.sh --create --replication-factor 2 --partitions 3 --topic user_events_output --bootstrap-server wn0-contsk:9092

在 ADLS Gen2 上准备文件

我们正在存储中创建名为 item attributes 的文件

目的是从 ADLS Gen2 上的文件中读取一批 item attributes 数据。每个项具有以下字段：
```
item_id, 
brand, 
category, 
timestamp, 
```

开发 Apache Flink 任务

在此步骤中，我们将执行以下活动

通过将 ADLS Gen2 文件中的 user_events 合并进来，从 Kafka 扩充 item attributes 主题。
我们将此步骤的结果作为包含事件的丰富化用户活动推送到 Kafka 主题中。

开发 Maven 项目

pom.xml

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>

    <groupId>contoso.example</groupId>
    <artifactId>FlinkKafkaJoinGen2</artifactId>
    <version>1.0-SNAPSHOT</version>

    <properties>
        <maven.compiler.source>1.8</maven.compiler.source>
        <maven.compiler.target>1.8</maven.compiler.target>
        <flink.version>1.17.0</flink.version>
        <java.version>1.8</java.version>
        <scala.binary.version>2.12</scala.binary.version>
        <kafka.version>3.2.0</kafka.version>
    </properties>
    <dependencies>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-java</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <!-- https://mvnrepository.com/artifact/org.apache.flink/flink-streaming-java -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-streaming-java</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <!-- https://mvnrepository.com/artifact/org.apache.flink/flink-clients -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-clients</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <!-- https://mvnrepository.com/artifact/org.apache.flink/flink-connector-files -->
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-files</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-kafka</artifactId>
            <version>${flink.version}</version>
        </dependency>
    </dependencies>
    <build>
        <plugins>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-assembly-plugin</artifactId>
                <version>3.0.0</version>
                <configuration>
                    <appendAssemblyId>false</appendAssemblyId>
                    <descriptorRefs>
                        <descriptorRef>jar-with-dependencies</descriptorRef>
                    </descriptorRefs>
                </configuration>
                <executions>
                    <execution>
                        <id>make-assembly</id>
                        <phase>package</phase>
                        <goals>
                            <goal>single</goal>
                        </goals>
                    </execution>
                </executions>
            </plugin>
        </plugins>
    </build>
</project>

将 Kafka 主题与 ADLS Gen2 文件结合

KafkaJoinGen2Demo.java

package contoso.example;

import org.apache.flink.api.common.eventtime.WatermarkStrategy;
import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.api.common.functions.RichMapFunction;
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.api.java.tuple.Tuple4;
import org.apache.flink.api.java.tuple.Tuple7;
import org.apache.flink.configuration.Configuration;
import org.apache.flink.connector.kafka.sink.KafkaRecordSerializationSchema;
import org.apache.flink.connector.kafka.sink.KafkaSink;
import org.apache.flink.connector.kafka.source.KafkaSource;
import org.apache.flink.connector.kafka.source.enumerator.initializer.OffsetsInitializer;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;

import java.io.BufferedReader;
import java.io.FileReader;
import java.util.HashMap;
import java.util.Map;

public class KafkaJoinGen2Demo {
    public static void main(String[] args) throws Exception {
        // 1. Set up the stream execution environment
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();

        // Kafka source configuration, update with your broker IPs
        String brokers = "<broker-ip>:9092,<broker-ip>:9092,<broker-ip>:9092";
        String inputTopic = "user_events";
        String outputTopic = "user_events_output";
        String groupId = "my_group";

        // 2. Register the cached file, update your container name and storage name
        env.registerCachedFile("abfs://<container-name>@<storagename>.dfs.core.windows.net/flink/data/item.txt", "file1");

        // 3. Read a stream of real-time user behavior event from a Kafka topic
        KafkaSource<String> kafkaSource = KafkaSource.<String>builder()
                .setBootstrapServers(brokers)
                .setTopics(inputTopic)
                .setGroupId(groupId)
                .setStartingOffsets(OffsetsInitializer.earliest())
                .setValueOnlyDeserializer(new SimpleStringSchema())
                .build();

        DataStream<String> kafkaData = env.fromSource(kafkaSource, WatermarkStrategy.noWatermarks(), "Kafka Source");

        // Parse Kafka source data
      DataStream<Tuple4<String, String, String, String>> userEvents = kafkaData.map(new MapFunction<String, Tuple4<String, String, String, String>>() {
          @Override
          public Tuple4<String, String, String, String> map(String value) throws Exception {
              // Parse the line into a Tuple4
              String[] parts = value.split(",");
              if (parts.length < 4) {
                  // Log and skip malformed record
                  System.out.println("Malformed record: " + value);
                  return null;
              }
              return new Tuple4<>(parts[0], parts[1], parts[2], parts[3]);
           }
       });

        // 4. Enrich the user activity events by joining the items' attributes from a file
        DataStream<Tuple7<String,String,String,String,String,String,String>> enrichedData = userEvents.map(new MyJoinFunction());

        // 5. Output the enriched user activity events to a Kafka topic
        KafkaSink<String> sink = KafkaSink.<String>builder()
                .setBootstrapServers(brokers)
                .setRecordSerializer(KafkaRecordSerializationSchema.builder()
                        .setTopic(outputTopic)
                        .setValueSerializationSchema(new SimpleStringSchema())
                        .build()
                )
                .build();

        enrichedData.map(value -> value.toString()).sinkTo(sink);

        // 6. Execute the Flink job
        env.execute("Kafka Join Batch gen2 file, sink to another Kafka Topic");
    }

    private static class MyJoinFunction extends RichMapFunction<Tuple4<String,String,String,String>, Tuple7<String,String,String,String,String,String,String>> {
        private Map<String, Tuple4<String, String, String, String>> itemAttributes;

        @Override
        public void open(Configuration parameters) throws Exception {
            super.open(parameters);

            // Read the cached file and parse its contents into a map
            itemAttributes = new HashMap<>();
            try (BufferedReader reader = new BufferedReader(new FileReader(getRuntimeContext().getDistributedCache().getFile("file1")))) {
                String line;
                while ((line = reader.readLine()) != null) {
                    String[] parts = line.split(",");
                    itemAttributes.put(parts[0], new Tuple4<>(parts[0], parts[1], parts[2], parts[3]));
                }
            }
        }

        @Override
        public Tuple7<String,String,String,String,String,String,String> map(Tuple4<String,String,String,String> value) throws Exception {
            Tuple4<String, String, String, String> broadcastValue = itemAttributes.get(value.f1);
            if (broadcastValue != null) {
                return Tuple7.of(value.f0,value.f1,value.f2,value.f3,broadcastValue.f1,broadcastValue.f2,broadcastValue.f3);
            } else {
                return null;
            }
        }
    }
}