Flink - souce算子

水善利万物而不争，处众人之所恶，故几于道

目录

  1. 从Java的集合中读取数据

  2. 从本地文件中读取数据

  3. 从HDFS中读取数据

  4. 从Socket中读取数据

  5. 从Kafka中读取数据

  6. 自定义Source

---

官方文档 - Flink1.13

---

1. 从Java的集合中读取数据

fromCollection(waterSensors)

public static void main(String[] args) {
    Configuration conf = new Configuration();
    conf.setInteger("rest.port",1000);
    StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(conf);
    env.setParallelism(1);

    List<WaterSensor> waterSensors = Arrays.asList(
            new WaterSensor("ws_001", 1577844001L, 45),
            new WaterSensor("ws_002", 1577844015L, 43),
            new WaterSensor("ws_003", 1577844020L, 42));
    
    env
            .fromCollection(waterSensors)
            .print();

    try {
        env.execute();
    } catch (Exception e) {
        e.printStackTrace();
    }
}

运行结果：

2. 从本地文件中读取数据

readTextFile(“input/words.txt”)，支持相对路径和绝对路径

public static void main(String[] args) {
    Configuration conf = new Configuration();
    conf.setInteger("rest.port",1000);
    StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(conf);
    env.setParallelism(1);

    env.readTextFile("input/words.txt").print();

    try {
        env.execute();
    } catch (Exception e) {
        e.printStackTrace();
    }

}

运行结果：

3. 从HDFS中读取数据

readTextFile(“hdfs://hadoop101:8020/flink/data/words.txt”)

要先在pom文件中添加hadoop-client依赖：

<dependency>
    <groupId>org.apache.hadoopgroupId>
    <artifactId>hadoop-clientartifactId>
    <version>3.1.3version>
dependency>

public static void main(String[] args) {
    Configuration conf = new Configuration();
    conf.setInteger("rest.port",1000);
    StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(conf);
    env.setParallelism(1);

    env.readTextFile("hdfs://hadoop101:8020/flink/data/words.txt").print();
    
    try {
        env.execute();
    } catch (Exception e) {
        e.printStackTrace();
    }
}

运行结果：

4. 从Socket中读取数据

socketTextStream(“hadoop101”,9999)，这个输入源不支持多个并行度。

public static void main(String[] args) {
    Configuration conf = new Configuration();
    conf.setInteger("rest.port",1000);
    StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(conf);
    env.setParallelism(1);

    //从端口中读数据，  windows中 nc -lp 9999     Linux nc -lk 9999
    env.socketTextStream("hadoop101",9999).print();

    try {
        env.execute();
    } catch (Exception e) {
        e.printStackTrace();
    }
}

运行结果：

5. 从Kafka中读取数据

addSource(new FlinkKafkaConsumer<>(“flink_source_kafka”,new SimpleStringSchema(),properties))

第一个参数是topic，

第二个参数是序列化器，序列化器就是在Kafka和flink之间转换数据 - 官方注释：The de-/serializer used to convert between Kafka’s byte messages and Flink’s objects.（反-序列化程序用于在Kafka的字节消息和Flink的对象之间进行转换。）

第三个参数是Kafka的配置。

public static void main(String[] args) {
    Configuration conf = new Configuration();
    conf.setInteger("rest.port",1000);
    StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(conf);
    env.setParallelism(1);

    Properties properties = new Properties();
    // 设置集群地址
    properties.setProperty("bootstrap.servers", "hadoop101:9092,hadoop102:9092,hadoop103:9092");
    // 设置所属消费者组
    properties.setProperty("group.id", "flink_consumer_group");
    env.addSource(new FlinkKafkaConsumer<>("flink_source_kafka",new SimpleStringSchema(),properties)).print();

    try {
        env.execute();
    } catch (Exception e) {
        e.printStackTrace();
    }
}

运行结果：

6. 自定义Source

addSource(new XXXX())

  大多数情况下，前面的数据源已经能够满足需要，但是难免会存在特殊情况的场合，所以flink也提供了能自定义数据源的方式.

public class Flink06_myDefDataSource {
    public static void main(String[] args) {
        Configuration conf = new Configuration();
        conf.setInteger("rest.port",1000);
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(conf);
        env.setParallelism(1);

        env.addSource(new RandomWatersensor()).print();

        try {
            env.execute();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}

  自定义数据源需要定义一个类，然后实现SourceFunction接口，然后实现其中的两个方法，run和cancel，run方法包含具体读数据的逻辑，当调用cancel方法的时候应该可以让run方法中的读数据逻辑停止

public class RandomWatersensor implements SourceFunction<WaterSensor> {
    private Boolean running = true;

    @Override
    public void run(SourceContext<WaterSensor> sourceContext) throws Exception {
        Random random = new Random();
        while (running){
            sourceContext.collect(new WaterSensor(
                    "sensor" + random.nextInt(50),
                    Calendar.getInstance().getTimeInMillis(),
                    random.nextInt(100)
            ));
            Thread.sleep(1000);
        }
    }

    /**
     * 大多数的source在run方法内部都会有一个while循环,
     * 当调用这个方法的时候, 应该可以让run方法中的while循环结束
     */
    @Override
    public void cancel() {
        running = false;
    }

}

运行结果：

---

demo2 - 自定义从socket中读取数据

public class Flink04_Source_Custom {
    public static void main(String[] args) throws Exception {


        // 1. 创建执行环境
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env
          .addSource(new MySource("hadoop102", 9999))
          .print();

        env.execute();
    }

    public static class MySource implements SourceFunction<WaterSensor> {
        private String host;
        private int port;
        private volatile boolean isRunning = true;
        private Socket socket;

        public MySource(String host, int port) {
            this.host = host;
            this.port = port;
        }


        @Override
        public void run(SourceContext<WaterSensor> ctx) throws Exception {
            // 实现一个从socket读取数据的source
            socket = new Socket(host, port);
            BufferedReader reader = new BufferedReader(new InputStreamReader(socket.getInputStream(), StandardCharsets.UTF_8));
            String line = null;
            while (isRunning && (line = reader.readLine()) != null) {
                String[] split = line.split(",");
                ctx.collect(new WaterSensor(split[0], Long.valueOf(split[1]), Integer.valueOf(split[2])));
            }
        }

        /**
         * 大多数的source在run方法内部都会有一个while循环,
         * 当调用这个方法的时候, 应该可以让run方法中的while循环结束
         */

        @Override
        public void cancel() {
            isRunning = false;
            try {
                socket.close();
            } catch (IOException e) {
                e.printStackTrace();
            }
        }
    }
}
/*
sensor_1,1607527992000,20
sensor_1,1607527993000,40
sensor_1,1607527994000,50
 */