Flink复杂事件处理

Question

我有一个flink cep代码，可以从套接字读取数据并检测模式。假设模式（单词）为“警报”。如果单词alert出现五次或五次以上，则应创建一个警报。但我得到了一个输入不匹配错误。Flink版本为1.3.0。提前谢谢！！

package pattern;

import org.apache.flink.cep.CEP;
import org.apache.flink.cep.PatternStream;
import org.apache.flink.cep.pattern.Pattern;
import org.apache.flink.cep.pattern.conditions.IterativeCondition;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.util.Collector;

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

    public class cep {

         public static void main(String[] args) throws Exception {


             StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();

                DataStreamSource<String> dss = env.socketTextStream("localhost", 3005);

                dss.print();

                Pattern<String,String> pattern = Pattern.<String> begin("first")
                        .where(new IterativeCondition<String>() {
                            @Override
                            public boolean filter(String word, Context<String> context) throws Exception {
                                return word.equals("alert");
                            }
                        })
                        .times(5);


                PatternStream<String> patternstream = CEP.pattern(dss, pattern);

                DataStream<String> alerts = patternstream
                        .flatSelect((Map<String,List<String>> in, Collector<String> out) -> {

                            String first = in.get("first").get(0);

                            for (int i = 0; i < 6; i++ ) {

                                out.collect(first);

                            }


                        });

                alerts.print();

                env.execute();

            }

    }

Answer 1

所以我有了代码来工作。这是可行的解决方案，

    package pattern;

    import org.apache.flink.cep.CEP;
    import org.apache.flink.cep.PatternSelectFunction;
    import org.apache.flink.cep.PatternStream;
    import org.apache.flink.cep.pattern.Pattern;
    import org.apache.flink.cep.pattern.conditions.IterativeCondition;
    import org.apache.flink.streaming.api.datastream.DataStream;
    import org.apache.flink.streaming.api.datastream.DataStreamSource;
    import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
    import org.apache.flink.util.Collector;

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

    public class cep {

         public static void main(String[] args) throws Exception {


             StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();

                DataStreamSource<String> dss = env.socketTextStream("localhost", 3005);

                dss.print();

                Pattern<String,String> pattern = Pattern.<String> begin("first")
                        .where(new IterativeCondition<String>() {
                            @Override
                            public boolean filter(String word, Context<String> context) throws Exception {
                                return word.equals("alert");
                            }
                        })
                        .times(5);

                PatternStream<String> patternstream = CEP.pattern(dss, pattern);

                DataStream<String> alerts = patternstream
                        .select(new PatternSelectFunction<String, String>() {
                            @Override
                            public String select(Map<String, List<String>> in) throws Exception {

                                String first = in.get("first").get(0);

                                if(first.equals("alert")){

                                    return ("5 or more alerts");
                                }
                                else{

                                    return (" ");
                                }
                            }
                        });

                alerts.print();

                env.execute();

            }

    }

Answer 2

只是对最初的问题做了一些澄清。在1.3.0中，有一个bug使得使用lambdas作为参数来选择/扁平选择变得不可能。

它在1.3.1中已修复，因此您的第一个版本的代码将与1.3.1一起使用。

此外，我认为你误解了时间量词。它与精确的次数匹配。所以，在您的情况下，只有当事件精确匹配3次时，它才会返回，而不是3次或更多。