Flink 1.14.2 CEP 没有输出

背景

最近几天把Flink项目的版本从1.12升级到了最新的1.14.2，然后发现项目里的CEP事件都没有输出了，即使将Stream打印到控制台，也是啥也没有。

问题原因

Flink在1.12版本之后，PatternStream默认使用Event Time。如果业务使用的事Processing Time，必须要明确配置。

解决办法

样例代码，下面的代码是不会有任何输出的。

package spendreport;

import org.apache.flink.cep.CEP;
import org.apache.flink.cep.PatternSelectFunction;
import org.apache.flink.cep.PatternStream;
import org.apache.flink.cep.nfa.aftermatch.AfterMatchSkipStrategy;
import org.apache.flink.cep.pattern.Pattern;
import org.apache.flink.cep.pattern.conditions.SimpleCondition;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.walkthrough.common.entity.Transaction;
import org.apache.flink.walkthrough.common.source.TransactionSource;


/**
 * @author jixiang.ma@mail.nwpu.edu.cn
 * @date 2022/1/7 19:00
 * @copyright © 2021 ruanjian.nwpu all rights reserved.
 */
public class CEPtest {
    public static void main(String[] args) throws Exception {
        final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment().setParallelism(1);

        DataStream<Transaction> dataStream = env.addSource(new TransactionSource());

        Pattern<Transaction, ?> pattern = Pattern.<Transaction>begin("begin", AfterMatchSkipStrategy.noSkip()).where(
                new SimpleCondition<Transaction>() {
                    @Override
                    public boolean filter(Transaction transaction) throws Exception {
                        return transaction.getAmount() > 300;
                    }
                }).timesOrMore(1);

        PatternStream<Transaction> patternStream = CEP.pattern(dataStream, pattern);
        DataStream<Transaction> d = patternStream.select(
                (PatternSelectFunction<Transaction, Transaction>) map -> map.get("begin").get(0))
                .name("bbbb");

        d.print();

        env.execute();
    }
}

 

而下面的代码就正常了

package spendreport;

import org.apache.flink.cep.CEP;
import org.apache.flink.cep.PatternSelectFunction;
import org.apache.flink.cep.PatternStream;
import org.apache.flink.cep.nfa.aftermatch.AfterMatchSkipStrategy;
import org.apache.flink.cep.pattern.Pattern;
import org.apache.flink.cep.pattern.conditions.SimpleCondition;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.walkthrough.common.entity.Transaction;
import org.apache.flink.walkthrough.common.source.TransactionSource;


/**
 * @author jixiang.ma@mail.nwpu.edu.cn
 * @date 2022/1/7 19:00
 * @copyright © 2021 ruanjian.nwpu all rights reserved.
 */
public class CEPtest {
    public static void main(String[] args) throws Exception {
        final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment().setParallelism(1);

        DataStream<Transaction> dataStream = env.addSource(new TransactionSource());

        Pattern<Transaction, ?> pattern = Pattern.<Transaction>begin("begin", AfterMatchSkipStrategy.noSkip()).where(
                new SimpleCondition<Transaction>() {
                    @Override
                    public boolean filter(Transaction transaction) throws Exception {
                        return transaction.getAmount() > 300;
                    }
                }).timesOrMore(1);

        PatternStream<Transaction> patternStream = CEP.pattern(dataStream, pattern).inProcessingTime();
        DataStream<Transaction> d = patternStream.select(
                (PatternSelectFunction<Transaction, Transaction>) map -> map.get("begin").get(0))
                .name("bbbb");

        d.print();

        env.execute();
    }
}

 只需要在在后面加上inProcessingTime()或者inEventTime()即可。

PatternStream<Transaction> patternStream = CEP.pattern(dataStream, pattern).inProcessingTime();

总结

当代码能跑起来的时候，能别动就别动！

备注：示例中的数据源是Flink内置数据源，需要加载Maven pom：

        <dependency>
			<groupId>org.apache.flink</groupId>
			<artifactId>flink-walkthrough-common_${scala.binary.version}</artifactId>
			<version>${flink.version}</version>
		</dependency>