Flink 流式计算 source与sink

Fink 处理过程可以简化为三步 (source transformations sink)

source表示数据来源

transformations表示执行flink的处理逻辑 (核心)

sink表示数据分布式处理完成之后的数据走向

source 获取数据的方式自带的api如下

公共pom

        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-java</artifactId>
            <version>1.11.1</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-streaming-java_2.12</artifactId>
            <version>1.11.1</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-clients_2.12</artifactId>
            <version>1.11.1</version>
        </dependency>

1.获取nc数据源

核心代码

    /**
     * 获取端口9999的数据 需要先启动nc (执行命令 nc -l -p 9999),否则显示连接拒绝
     *
     * @param env
     * @return
     */
    private static DataStream<String> getDataNC(StreamExecutionEnvironment env) {
        DataStreamSource<String> data = env.socketTextStream("localhost", 9999, "n");
        return data;
    }

2.获取kafka数据源

特定pom

        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-kafka_2.12</artifactId>
            <version>1.11.1</version>
        </dependency>

核心代码

    /**
     * 采集kafka的数据
     *
     * @param env
     * @return
     */
    private static DataStream<String> getDataKafka(StreamExecutionEnvironment env) {
        Properties props = new Properties();
        props.setProperty("bootstrap.servers", "ip:port");
        props.setProperty("group.id", "flink-group");
        FlinkKafkaConsumer<String> consumer = new FlinkKafkaConsumer<>("test", new SimpleStringSchema(), props);
        DataStreamSource<String> streamSource = env.addSource(consumer);
        return streamSource;
    }

source 自定义
 

import org.apache.flink.configuration.Configuration;
import org.apache.flink.streaming.api.functions.source.RichSourceFunction;

public class MySource extends RichSourceFunction<String> {//此处泛型需要主动显示指定,与sink不同

    /**
     * source获取数据的方法
     *
     * @param ctx
     * @throws Exception
     */
    @Override
    public void run(SourceContext<String> ctx) throws Exception {
        // 循环可以不停的读取静态数据
        while (true) {
            ctx.collect("1,2,3,4,5");
            System.out.println(this + "MySource获取数据");
            Thread.sleep(3000);
        }
    }
    /**
     * open方法在source第一次启动时调用，一般用于source的初始化操作,例如初始化数据库连接
     */
    @Override
    public void open(Configuration parameters) throws Exception {
        System.out.println("MySource初始化" + this);
        super.open(parameters);
    }
    /**
     * close方法在source退出时调用，一般用于source的资源回收操作,例如关闭数据库连接
     */
    @Override
    public void close() throws Exception {
        System.out.println("MySource回收资源" + this);
        super.close();
    }
    @Override
    public void cancel() {
        System.out.println("MySource取消" + this);
    }
}

使用source代码

final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(2);//设置flink分布式计算的迸发度
        DataStream<String> data = env.addSource(new MySource());

sink 获取数据的方式自带的api如下

windowCounts.addSink(new PrintSinkFunction<>());//打印到控制台

sink的经过flink分布式计算完成之后可以自定义自定义结果处理,例如数据需要保存到mysql

自定义sink需要继承类  org.apache.flink.streaming.api.functions.sink.RichSinkFunction

代码如下

import org.apache.flink.configuration.Configuration;
import org.apache.flink.streaming.api.functions.sink.RichSinkFunction;

/**
 * 自定义sink
 *
 * @param <T>
 */
public class MySink<T> extends RichSinkFunction<T> {//MySink可以继续使用泛型

    /**
     * open方法在sink第一次启动时调用，一般用于sink的初始化操作,例如初始化数据库连接
     */
    @Override
    public void open(Configuration parameters) throws Exception {
        System.out.println("初始化" + this);
        super.open(parameters);
    }

    /**
     * invoke方法是sink数据处理逻辑的方法，source端传来的数据都在invoke方法中进行处理
     * 其中invoke方法中第一个参数类型与RichSinkFunction<String>中的泛型对应。第二个参数
     * 为一些上下文信息
     */
    @Override
    public void invoke(T value, Context context) throws Exception {
        System.out.println(this + "输出:" + value);
    }

    /**
     * close方法在sink结束时调用，一般用于资源的回收操作,例如关闭数据库连接
     */
    @Override
    public void close() throws Exception {
        System.out.println("回收资源" + this);
        super.close();
    }
}

综合代码示例:

import org.apache.flink.api.common.functions.FlatMapFunction;
import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.api.common.functions.ReduceFunction;
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.api.java.functions.KeySelector;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.streaming.api.datastream.*;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.source.SourceFunction;
import org.apache.flink.streaming.api.windowing.time.Time;
import org.apache.flink.streaming.api.windowing.windows.TimeWindow;
import org.apache.flink.streaming.connectors.kafka.FlinkKafkaConsumer;
import org.apache.flink.util.Collector;

import java.util.Properties;


/**
 * Flink 单次统计
 */
public class SocketWindowWordCount {
    public static void main(String[] args) throws Exception {
        final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(1);//设置flink分布式计算的迸发度
        DataStream<String> data = env.addSource(new MySource());//添加source
        //单词切割
        SingleOutputStreamOperator<String> flatMap = data.flatMap(new FlatMapFunction<String, String>() {
            @Override
            public void flatMap(String value, Collector<String> out) throws Exception {
                for (String word : value.split(",")) {
                    out.collect(word);//单词切割
                }
            }
        });
        //单词记录为1
        SingleOutputStreamOperator<Tuple2<String, Long>> map = flatMap.map(new MapFunction<String, Tuple2<String, Long>>() {
            @Override
            public Tuple2<String, Long> map(String value) throws Exception {
                return new Tuple2<>(value, 1L);
            }
        });
        //分区
        KeyedStream<Tuple2<String, Long>, String> word = map.keyBy(new KeySelector<Tuple2<String, Long>, String>() {
            @Override
            public String getKey(Tuple2<String, Long> tp2) throws Exception {
                return tp2.f0;//根据Tuple2的f0 即 key分区
            }
        });
        //时间窗口长度为5秒,窗口移动为1秒进行计算
        WindowedStream<Tuple2<String, Long>, String, TimeWindow> windowedStream = word.timeWindow(Time.seconds(1), Time.seconds(1));
        //统计计算
        SingleOutputStreamOperator<Tuple2<String, Long>> windowCounts = windowedStream.reduce(new ReduceFunction<Tuple2<String, Long>>() {
            @Override
            public Tuple2<String, Long> reduce(Tuple2<String, Long> tp2_1, Tuple2<String, Long> tp2_2) throws Exception {
                return new Tuple2<>(tp2_1.f0, tp2_1.f1 + tp2_2.f1);
            }
        });
        windowCounts.addSink(new MySink<>());//打印到控制台
        env.execute("Socket Window WordCount"); //开始执行
    }

    /**
     * 获取端口9999的数据 需要先启动nc (执行命令 nc -l -p 9999),否则显示连接拒绝
     *
     * @param env
     * @return
     */
    private static DataStream<String> getDataNC(StreamExecutionEnvironment env) {
        DataStreamSource<String> data = env.socketTextStream("localhost", 9999, "n");
        return data;
    }

    /**
     * 采集kafka的数据
     *
     * @param env
     * @return
     */
    private static DataStream<String> getDataKafka(StreamExecutionEnvironment env) {
        Properties props = new Properties();
        props.setProperty("bootstrap.servers", "39.100.228.221:9092");
        props.setProperty("group.id", "flink-group");
        SourceFunction<String> consumer = new FlinkKafkaConsumer<>("test", new SimpleStringSchema(), props);
        DataStreamSource<String> streamSource = env.addSource(consumer);
        return streamSource;
    }
}

最后

以上就是奋斗皮皮虾为你收集整理的Flink 流式计算 source与sink的全部内容，希望文章能够帮你解决Flink 流式计算 source与sink所遇到的程序开发问题。

如果觉得靠谱客网站的内容还不错，欢迎将靠谱客网站推荐给程序员好友。