Sink-printSink底层实现

上篇：自定义Source

二话不说，直接上代码

这个案例是一个无限流，执行环境默认的并行度是：4，SocketSink的并行度：1

package cn._51doit.flink.day01;

import org.apache.flink.api.common.functions.FlatMapFunction;
import org.apache.flink.api.common.functions.IterationRuntimeContext;
import org.apache.flink.api.common.functions.RuntimeContext;
import org.apache.flink.configuration.Configuration;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.datastream.SingleOutputStreamOperator;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.sink.RichSinkFunction;
import org.apache.flink.streaming.api.functions.source.RichSourceFunction;
import org.apache.flink.util.Collector;


/**
 * Sink-printSink底层实现
 *
 * print--》PrintSinkFunction---extends RichSinkFunction---@Override
 * 	public void invoke(IN record) {
 * 		writer.write(record);
 *        }
 *
 *
 */
public class PrintSinkDemo {
    public static void main(String[] args) throws Exception {
        //local模式默认的并行度是当前节点逻辑核的数量
        Configuration configuration = new Configuration();
        StreamExecutionEnvironment env = StreamExecutionEnvironment.createLocalEnvironmentWithWebUI(configuration);

        //DataStream的并行度
        int parallelism01 = env.getParallelism();
        System.out.println("执行环境默认的并行度是："+parallelism01);

        DataStreamSource<String> lines = env.socketTextStream("Master", 8888);

        int parallelism = lines.getParallelism();
        System.out.println("SocketSink的并行度："+parallelism);

        // lines.print();
        lines.addSink(new MyPrintSink()).name("my-print-sink");  //在执行计划的sink的算子取名为“my-print-sink”

        env.execute();
    }

    //定义内部类
    public static class MyPrintSink extends RichSinkFunction<String > {
        private int indexOfThisSubtask;

        //最终把数据输出的方法（如：mysql、jdbc）
        @Override
        public void invoke(String value, Context context) throws Exception {
            //：拿到索引编号[从0开始]
            RuntimeContext runtimeContext = getRuntimeContext();
            int indexOfThisSubtask = runtimeContext.getIndexOfThisSubtask();

            System.out.println(indexOfThisSubtask+"> "+value);
        }
    }

}

源码解析

在print类点击进去查看到它是继承RichSinkFunction方法，重写invoke方法，采用writer把数据输出

在open方法传进去看到几个参数的判断，看看并行度是否大于1，如果大于1，就当前的subtaskIndex + 1再拼上大于号输出

public void open(int subtaskIndex, int numParallelSubtasks) {
		// get the target stream
		stream = target == STD_OUT ? System.out : System.err;

		completedPrefix = sinkIdentifier;

		if (numParallelSubtasks > 1) {
			if (!completedPrefix.isEmpty()) {
				completedPrefix += ":";
			}
			completedPrefix += (subtaskIndex + 1);
		}

		if (!completedPrefix.isEmpty()) {
			completedPrefix += "> ";
		}