Flink Trigger代码实战

 说明，就是数据大于2 或者 时间到凌晨的时候就触发执行。

针对下面的业务做的简单的案例：

实际的意思是正常的情况下正常触发，在到达凌晨的时候直接触发，然后缓存清零

package application;

import com.alibaba.fastjson.JSONObject;
import operator.DayProcessOperator;
import operator.DayTrigger;
import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.api.java.utils.ParameterTool;
import org.apache.flink.streaming.api.TimeCharacteristic;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.windowing.assigners.TumblingProcessingTimeWindows;
import org.apache.flink.streaming.api.windowing.time.Time;
import org.apache.flink.streaming.connectors.kafka.FlinkKafkaConsumer010;

import java.util.Properties;

/**
 * todo 数据分析
 * todo http://docs-aliyun.cn-hangzhou.oss.aliyun-inc.com/assets/pic/49939/cn_zh/1487929553566/man-screen-shot.jpg
 *
 * todo 指标维度： 1，活跃用户  2，新增用户 3，登陆会员 4，新注册用户 5，启动次数
 *
 * todo 思路1： 每秒活跃用户数量都要存储到hbase，使用窗口函数，窗口大小为1s
 * todo 思路2：每天凌晨状态清零，开始重新累加计算，所以需要
 *
 */
public class Data_analysis_demo {
    public static void main(String[] args) {

        final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(4);
        //todo 获取kafka的配置属性
        args = new String[]{"--input-topic", "topn_test", "--bootstrap.servers", "node2.hadoop:9092,node3.hadoop:9092",
                "--zookeeper.connect", "node1.hadoop:2181,node2.hadoop:2181,node3.hadoop:2181", "--group.id", "cc2"};


        ParameterTool parameterTool = ParameterTool.fromArgs(args);
        env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
        Properties sendPros = parameterTool.getProperties();
        Properties pros = parameterTool.getProperties();

        //todo 指定输入数据为kafka topic
        DataStream<String> kafkaDstream = env.addSource(new FlinkKafkaConsumer010<String>(
                pros.getProperty("input-topic"),
                new SimpleStringSchema(),
                pros).setStartFromLatest()

        );

        //todo 先转成JSON

        DataStream<JSONObject> str2JsonDstream = kafkaDstream.map(new MapFunction<String, JSONObject>() {
            @Override
            public JSONObject map(String input) throws Exception {
                JSONObject inputJson = null;
                try {
                    inputJson = JSONObject.parseObject(input);
                } catch (Exception e) {
                    e.printStackTrace();
                }
                return inputJson;
            }
        });

        //todo 使用window算子
        str2JsonDstream.keyBy(value -> value.getString("appKey"))
                .window(TumblingProcessingTimeWindows.of(Time.milliseconds(2000)))
                .trigger(new DayTrigger())
                .process(new DayProcessOperator())
                .print();
        try {
            env.execute("开始执行....................");
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}

package operator;


import org.apache.flink.api.common.functions.ReduceFunction;
import org.apache.flink.api.common.state.ReducingStateDescriptor;
import org.apache.flink.api.common.state.ValueState;
import org.apache.flink.api.common.state.ValueStateDescriptor;
import org.apache.flink.api.common.typeutils.base.LongSerializer;
import org.apache.flink.streaming.api.windowing.triggers.Trigger;
import org.apache.flink.streaming.api.windowing.triggers.TriggerResult;
import org.apache.flink.streaming.api.windowing.windows.Window;


public class DayTrigger<JSONObject> extends Trigger<com.alibaba.fastjson.JSONObject, Window> {
    private ReducingStateDescriptor<Long> countStateDescriptor =
            new ReducingStateDescriptor("counter", new Sum(), LongSerializer.INSTANCE);

    ValueStateDescriptor<Integer> stateDescriptor = new ValueStateDescriptor<>("total", Integer.class);

    //每个元素都会触发
    @Override
    public TriggerResult onElement(com.alibaba.fastjson.JSONObject element, long timestamp, Window window, TriggerContext ctx) throws Exception {
        System.out.println("先打印timestamp = " + timestamp+",本地时间："+System.currentTimeMillis());
        ValueState<Integer> sumState = ctx.getPartitionedState(stateDescriptor);
        if (null == sumState.value()) {
            sumState.update(0);
        }

        sumState.update(1 + sumState.value());
        if (sumState.value() >= 2) {
            //这里可以选择手动处理状态
            //  默认的trigger发送是TriggerResult.FIRE 不会清除窗口数据
            System.out.println("触发.....数据条数为：" + (1 + sumState.value()));
            return TriggerResult.FIRE;
        }
        Long todayZeroPointTimestamps = getTodayZeroPointTimestamps();
        System.out.println("todayZeroPointTimestamps = " + todayZeroPointTimestamps);
        if (timestamp >= todayZeroPointTimestamps) {
            return TriggerResult.FIRE_AND_PURGE;
        }
        if (timestamp >= window.maxTimestamp()) {
            return TriggerResult.FIRE_AND_PURGE;

        } else {
            return TriggerResult.CONTINUE;
        }


//        ReducingState<Long> countState = ctx.getPartitionedState(countStateDescriptor);
//        countState.add(1L);
//
        System.out.println("打印countState.get() = " + countState.get());
//        if (countState.get() >= 3) {
//            System.out.println("触发了............. " );
//             //这里是触发并且清除
//            return TriggerResult.FIRE_AND_PURGE;
//        }
//        return TriggerResult.CONTINUE;
    }

    @Override
    public TriggerResult onProcessingTime(long time, Window window, TriggerContext triggerContext) throws Exception {
        return TriggerResult.CONTINUE;
    }

    @Override
    public TriggerResult onEventTime(long time, Window window, TriggerContext triggerContext) throws Exception {
        return TriggerResult.CONTINUE;
    }

    @Override
    public void clear(Window window, TriggerContext ctx) throws Exception {
//        System.out.println("清理窗口状态  窗口内保存值为" + ctx.getPartitionedState(stateDescriptor).value());
        ctx.getPartitionedState(stateDescriptor).clear();

    }

    class Sum implements ReduceFunction<Long> {

        @Override
        public Long reduce(Long value1, Long value2) throws Exception {
            return value1 + value2;
        }
    }

    public static Long getTodayZeroPointTimestamps() {

        long now = System.currentTimeMillis();
        long daySecond = 60 * 60 * 24 * 1000;
        long dayTime = now - (now + 8 * 3600 * 1000) % daySecond + 1 * daySecond;
        return dayTime;

    }

    public static void main(String[] args) {
        Long todayZeroPointTimestamps = getTodayZeroPointTimestamps();
        System.out.println("todayZeroPointTimestamps = " + todayZeroPointTimestamps);

    }

}

 

最后

以上就是结实胡萝卜最近收集整理的关于Flink Trigger代码实战的全部内容，更多相关Flink内容请搜索靠谱客的其他文章。