flink-cep实践

java 复制代码
package com.techwolf.hubble;

import com.alibaba.fastjson.JSONObject;
import com.techwolf.hubble.constant.Config;
import com.techwolf.hubble.model.TestEvent;
import org.apache.flink.api.common.eventtime.TimestampAssigner;
import org.apache.flink.api.common.eventtime.TimestampAssignerSupplier;
import org.apache.flink.api.common.eventtime.WatermarkStrategy;
import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.cep.CEP;
import org.apache.flink.cep.PatternFlatSelectFunction;
import org.apache.flink.cep.PatternFlatTimeoutFunction;
import org.apache.flink.cep.PatternStream;
import org.apache.flink.cep.pattern.Pattern;
import org.apache.flink.cep.pattern.conditions.SimpleCondition;
import org.apache.flink.streaming.api.TimeCharacteristic;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.datastream.SingleOutputStreamOperator;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.sink.PrintSinkFunction;
import org.apache.flink.streaming.api.windowing.time.Time;
import org.apache.flink.util.Collector;
import org.apache.flink.util.OutputTag;

import java.util.List;
import java.util.Map;


/**
 * Hello world!
 *
 */
public class App {

    public static void main(String[] args) throws Exception{
        //初始化环境
        StreamExecutionEnvironment env=StreamExecutionEnvironment.getExecutionEnvironment();

        env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
        //定义时间戳提取器作为输入流分配时间戳和水位线
        WatermarkStrategy<TestEvent> watermarkStrategy=WatermarkStrategy.<TestEvent>
                forMonotonousTimestamps().withTimestampAssigner(new EventTimeAssignerSupplier());

        DataStream<TestEvent> inputDataSteam=env.fromElements(
                new TestEvent("1","A",System.currentTimeMillis()-100*1000,"1"),
                new TestEvent("1","A",System.currentTimeMillis()-85*1000,"2"),
                new TestEvent("1","A",System.currentTimeMillis()-80*1000,"3"),
                new TestEvent("1","A",System.currentTimeMillis()-75*1000,"4"),
                new TestEvent("1","A",System.currentTimeMillis()-60*1000,"5"),
                new TestEvent("1","A",System.currentTimeMillis()-55*1000,"6"),
                new TestEvent("1","A",System.currentTimeMillis()-40*1000,"7"),
                new TestEvent("1","A",System.currentTimeMillis()-35*1000,"8"),
                new TestEvent("1","A",System.currentTimeMillis()-20*1000,"9"),
                new TestEvent("1","A",System.currentTimeMillis()-10*1000,"10"),
                new TestEvent("1","B",System.currentTimeMillis()-5*1000,"11")
        ).assignTimestampsAndWatermarks(watermarkStrategy);

        Pattern<TestEvent,TestEvent> pattern=Pattern.<TestEvent>begin("begin")
                .where(new SimpleCondition<TestEvent>() {
                    @Override
                    public boolean filter(TestEvent testEvent) throws Exception {
                        return testEvent.getAction().equals("A");
                    }
                }).
                followedBy("end")
                .where(new SimpleCondition<TestEvent>() {
                    @Override
                    public boolean filter(TestEvent testEvent) throws Exception {
                        return testEvent.getAction().equals("B");
                    }
                }).within(Time.seconds(10));


        PatternStream<TestEvent> patternStream=CEP.pattern(inputDataSteam.keyBy(TestEvent::getId),pattern);
        OutputTag<TestEvent> timeOutTag=new OutputTag<TestEvent>("timeOutTag"){};

        //处理匹配结果
        SingleOutputStreamOperator<TestEvent> twentySingleOutputStream=patternStream
                .flatSelect(timeOutTag,new EventTimeOut(),new FlatSelect())
                .uid("match_twenty_minutes_pattern");
        DataStream<String> result=twentySingleOutputStream.getSideOutput(timeOutTag).map(new MapFunction<TestEvent, String>() {
            @Override
            public String map(TestEvent testEvent) throws Exception {
                return JSONObject.toJSONString(testEvent);
            }
        });
        result.print();
        env.execute(Config.JOB_NAME);
    }

    public static class EventTimeOut implements PatternFlatTimeoutFunction<TestEvent,TestEvent> {
        private static final long serialVersionUID = -2471077777598713906L;
        @Override
        public void timeout(Map<String, List<TestEvent>> map, long l, Collector<TestEvent> collector) throws Exception {
            if (null != map.get("begin")) {
                for (TestEvent event : map.get("begin")) {
                    collector.collect(event);
                }
            }
        }
    }

    public static class FlatSelect implements PatternFlatSelectFunction<TestEvent,TestEvent> {
        private static final long serialVersionUID = 1753544074226581611L;
        @Override
        public void flatSelect(Map<String, List<TestEvent>> map, Collector<TestEvent> collector) throws Exception {
            if (null != map.get("begin")) {
                for (TestEvent event : map.get("begin")) {
                    collector.collect(event);
                }
            }
        }
    }

    public static class EventTimeAssignerSupplier implements TimestampAssignerSupplier<TestEvent> {
        private static final long serialVersionUID = -9040340771307752904L;

        @Override
        public TimestampAssigner<TestEvent> createTimestampAssigner(Context context) {
            return new EventTimeAssigner();
        }
    }

    public static class EventTimeAssigner implements TimestampAssigner<TestEvent> {
        @Override
        public long extractTimestamp(TestEvent event, long l) {
            return event.getEventTime();
        }
    }
}
相关推荐
soso196835 分钟前
DataWorks快速入门
大数据·数据仓库·信息可视化
The_Ticker41 分钟前
CFD平台如何接入实时行情源
java·大数据·数据库·人工智能·算法·区块链·软件工程
java1234_小锋1 小时前
Elasticsearch中的节点(比如共20个),其中的10个选了一个master,另外10个选了另一个master,怎么办?
大数据·elasticsearch·jenkins
Elastic 中国社区官方博客1 小时前
Elasticsearch 开放推理 API 增加了对 IBM watsonx.ai Slate 嵌入模型的支持
大数据·数据库·人工智能·elasticsearch·搜索引擎·ai·全文检索
我的运维人生1 小时前
Elasticsearch实战应用:构建高效搜索与分析平台
大数据·elasticsearch·jenkins·运维开发·技术共享
大数据编程之光1 小时前
Flink Standalone集群模式安装部署全攻略
java·大数据·开发语言·面试·flink
B站计算机毕业设计超人1 小时前
计算机毕业设计SparkStreaming+Kafka旅游推荐系统 旅游景点客流量预测 旅游可视化 旅游大数据 Hive数据仓库 机器学习 深度学习
大数据·数据仓库·hadoop·python·kafka·课程设计·数据可视化
在下不上天3 小时前
Flume日志采集系统的部署,实现flume负载均衡,flume故障恢复
大数据·开发语言·python
智慧化智能化数字化方案3 小时前
华为IPD流程管理体系L1至L5最佳实践-解读
大数据·华为
PersistJiao4 小时前
在 Spark RDD 中,sortBy 和 top 算子的各自适用场景
大数据·spark·top·sortby