今日指数项目之FlinkCEP入门案例

Maynor

发布于 2022-05-07 10:32:10

2720

发布于 2022-05-07 10:32:10

文章被收录于专栏：最新最全的大数据技术体系最新最全的大数据技术体系

CEP案例

1.入门案例

需求：有一个业务系统，用户要使用该业务系统必须要先登陆过滤出来在2秒内连续登陆失败的用户

在test源码目录下创建测试类：cn.itcast.LoginFailDemo 开发步骤： 1.获取流处理执行环境 2.设置并行度,设置事件时间加载数据源,提取事件时间 4.定义匹配模式，设置时间长度 5.匹配模式（分组） 6.数据处理 7.打印 8.触发执行

数据源：

	Arrays.asList(
        new LoginUser (1, "192.168.0.1", "fail", 1558430842000L),		//2019-05-21 17:27:22
        new LoginUser (1, "192.168.0.2", "fail", 1558430843000L),		//2019-05-21 17:27:23
        new LoginUser (1, "192.168.0.3", "fail", 1558430844000L),		//2019-05-21 17:27:24
        new LoginUser (2, "192.168.10.10", "success", 1558430845000L)	//2019-05-21 17:27:25
)

参考代码

/**

 * 使用CEP实现三秒内登录失败两次的用户
   */
   public class LoginFailDemo {

   public static void main(String[] args) throws Exception {
       //1：初始化流式运行环境
       StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
       //2：设置并行度为1
       env.setParallelism(1);
       //3：指定数据按照事件时间进行处理
       env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
       //4：构建数据源
       DataStream<LoginUser > LoginUserStream = env.fromCollection(Arrays.asList(
               new LoginUser (1, "192.168.0.1", "fail", 1558430842000L),//2019-05-21 17:27:22
               new LoginUser (1, "192.168.0.2", "fail", 1558430843000L),//2019-05-21 17:27:23
               new LoginUser (1, "192.168.0.3", "fail", 1558430844000L),//2019-05-21 17:27:24
               new LoginUser (2, "192.168.10.10", "success", 1558430845000L)//2019-05-21 17:27:25
       )).assignTimestampsAndWatermarks(new BoundedOutOfOrdernessTimestampExtractor<LoginUser>(Time.seconds(0)) {
           @Override
           public long extractTimestamp(LoginUser element) {
               return element.getEventTime();
           }
       });

       //5.1：定义规则模型
       Pattern<LoginUser, LoginUser > LoginUserPattern = Pattern.<LoginUser >begin("begin")
               .where(new IterativeCondition<LoginUser>() {
                   @Override
                   public boolean filter(LoginUser loginUser, Context<LoginUser > context) throws Exception {
                       return loginUser.getEventType().equals("fail");
                   }
       
               })//匹配第一个事件，匹配的是登录失败
               .next("next") //匹配到第一个事件以后，紧跟着一个事件数据，next表示两个事件必须严格的临近
               .where(new IterativeCondition<LoginUser >() {
                   @Override
                   public boolean filter(LoginUser loginUser, Context<LoginUser> context) throws Exception {
                       return loginUser.getEventType().equals("fail");
                   }
               })//匹配第二个事件，匹配的是登录失败
               .within(Time.seconds(3));//定义结束状态，结束状态可以是时间触发也可以是满足了某个事件触发
       
       //5.2：将规则模型应用到数据流中
       PatternStream<LoginUser > patternDataStream = CEP.pattern(LoginUserStream.keyBy(LoginUser ::getUserId), LoginUserPattern);
       //5.3：获取到符合规则模型的数据
       /**
          * IN：传入的数据类型
          * OUT：返回值的数据类型
          *  (Long, String, String, Long):(用户id, 登录ip，登录状态，登录时间)
          */
       
       SingleOutputStreamOperator<Tuple4<Integer, String, String, Long>> loginFailDataStream = patternDataStream.select(new PatternSelectFunction<LoginUser, Tuple4<Integer, String, String, Long>>() {
           @Override
           public Tuple4<Integer, String, String, Long> select(Map<String, List<LoginUser>> pattern) throws Exception {
               //根据刚才的分析，符合规则的数据会存储到状态集合中，也就是state中，所以查找匹配的时候需要在state中获取数据
               LoginUser loginUser = pattern.getOrDefault("next", null).iterator().next();
       
               //返回匹配到的数据
               return Tuple4.of(loginUser.getUserId(), loginUser.getIp(), loginUser.getEventType(), loginUser.getEventTime());
           }
       });
       
       //打印出来符合条件的数据
       loginFailDataStream.print("连续两次登录失败的用户>>>");
       //执行任务
       env.execute();

   }

登陆对象：

   public int userId; //用户id
   public String ip;//用户Ip
   public String eventType; //状态
   public Long eventTime;//事件时间

 /**
     * 构建登录对象
     */
    public static class LoginUser implements Serializable {
        public int userId; //用户id
        public String ip;//用户Ip
        public String eventType; //状态
        public Long eventTime;//事件时间

        public int getUserId() {
            return userId;
        }
    
        public void setUserId(int userId) {
            this.userId = userId;
        }
    
        public String getIp() {
            return ip;
        }
    
        public void setIp(String ip) {
            this.ip = ip;
        }
    
        public String getEventType() {
            return eventType;
        }
    
        public void setEventType(String eventType) {
            this.eventType = eventType;
        }
    
        public Long getEventTime() {
            return eventTime;
        }
    
        public void setEventTime(Long eventTime) {
            this.eventTime = eventTime;
        }
    
        public LoginEvent(int userId, String ip, String eventType, Long eventTime) {
            this.userId = userId;
            this.ip = ip;
            this.eventType = eventType;
            this.eventTime = eventTime;
        }
    
        @Override
        public String toString() {
            return "LoginEvent{" +
                    "userId=" + userId +
                    ", ip='" + ip + '\'' +
                    ", eventType='" + eventType + '\'' +
                    ", eventTime=" + eventTime +
                    '}';
        }
    }

}

2.监控市场价格

需求：物价局和工商局会监督市场上各种商品得销售价格，随着市场行情和商品供需得变化，商品价格会有一定程度得浮动，如果商品价格在指定得价格区间波动，政府部门是不会干预的额，如果商品价格在一定的时间范围内波动幅度超出了指定的区间范围，并且上行幅度过大，物价局会上报敏感数据信息，并规范市场价格。在此，我们假定如果商品售价在1分钟之内有连续两次超过预定商品价格阀值就发送告警信息。

测试数据

{"goodsId":100001,"goodsPrice":6,"goodsName":"apple","alias":"苹果","orderTime":1558430843000}
{"goodsId":100007,"goodsPrice":0.5,"goodsName":"mask","alias":"口罩","orderTime":1558430844000}
{"goodsId":100002,"goodsPrice":2,"goodsName":"rice","alias":"大米","orderTime":1558430845000}
{"goodsId":100003,"goodsPrice":2,"goodsName":"flour","alias":"面粉","orderTime":1558430846000}
{"goodsId":100004,"goodsPrice":12,"goodsName":"rice","alias":"大米","orderTime":1558430847000}
{"goodsId":100005,"goodsPrice":20,"goodsName":"apple","alias":"苹果","orderTime":1558430848000}
{"goodsId":100006,"goodsPrice":3,"goodsName":"banana","alias":"香蕉","orderTime":1558430849000}
{"goodsId":100007,"goodsPrice":10,"goodsName":"mask","alias":"口罩","orderTime":1558430850000}
{"goodsId":100001,"goodsPrice":16,"goodsName":"apple","alias":"苹果","orderTime":1558430852000}
{"goodsId":100007,"goodsPrice":15,"goodsName":"mask","alias":"口罩","orderTime":1558430853000}
{"goodsId":100002,"goodsPrice":12,"goodsName":"rice","alias":"大米","orderTime":1558430854000}
{"goodsId":100003,"goodsPrice":12,"goodsName":"flour","alias":"面粉","orderTime":1558430855000}
{"goodsId":100004,"goodsPrice":12,"goodsName":"rice","alias":"大米","orderTime":1558430856000}
{"goodsId":100005,"goodsPrice":20,"goodsName":"apple","alias":"苹果","orderTime":1558430857000}
{"goodsId":100006,"goodsPrice":13,"goodsName":"banana","alias":"香蕉","orderTime":1558430858000}
{"goodsId":100007,"goodsPrice":10,"goodsName":"mask","alias":"口罩","orderTime":1558430859000}

创建kafka topic

./kafka-topics.sh --create --topic cep --zookeeper node01:2181 --partitions 1 --replication-factor 1

生产数据

./kafka-console-producer.sh --broker-list node01:9092 --topic cep

redis保存限制价格

jedisCluster.hset(“product”,“apple”,“10”); jedisCluster.hset(“product”,“rice”,“6”); jedisCluster.hset(“product”,“flour”,“6”); jedisCluster.hset(“product”,“banana”,“8”); jedisCluster.hset(“product”,“mask”,“5”);

开发步骤在test源码目录下创建测试类：cn.itcast.CepMarkets 1.获取流处理执行环境 2.设置事件时间、并行度整合kafka 4.数据转换 5.process获取bean,设置status，并设置事件时间 6.定义匹配模式，设置时间长度 7.匹配模式（分组） 8.查询告警数据

2.1.代码开发

public class CepMarkets {

    public static void main(String[] args) throws Exception {
       
        //1.获取流处理执行环境
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(1);
        //2.设置事件时间
        env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
        //3.整合kafka
        Properties properties = new Properties();
        properties.setProperty("bootstrap.servers", "node01:9092"); //broker地址
        properties.setProperty("group.id", "cep"); //消费组
        properties.setProperty("enable.auto.commit", "true");
        properties.setProperty("auto.commit.interval.ms", "5000");
        FlinkKafkaConsumer011<String> kafkaConsumer = new FlinkKafkaConsumer011<>("cep", new SimpleStringSchema(), properties);
        kafkaConsumer.setStartFromEarliest();
        DataStreamSource<String> source = env.addSource(kafkaConsumer);
    
        //4.数据转换
        SingleOutputStreamOperator<Product> mapData = source.map(new MapFunction<String, Product>() {
            @Override
            public Product map(String value) throws Exception {
                JSONObject json = JSON.parseObject(value);
                Product product = new Product(
                        json.getLong("goodsId"),
                        json.getDouble("goodsPrice"),
                        json.getString("goodsName"),
                        json.getString("alias"),
                        json.getLong("orderTime"),
                        false
                );
                return product;
            }
        });
    
        //5.保留告警数据(设置时间)
        SingleOutputStreamOperator<Product> waterData = mapData.keyBy(Product::getGoodsId)
                .process(new KeyedProcessFunction<Long, Product, Product>() {
                    Map<String, String> map = null;
    
                    @Override
                    public void open(Configuration parameters) throws Exception {
                        JedisCluster jedisCluster = RedisUtil.getJedisCluster();
                        map = jedisCluster.hgetAll("product");
                    }
    
                    @Override
                    public void processElement(Product value, Context ctx, Collector<Product> out) throws Exception {
                        long priceAlert = Long.parseLong(map.get(value.getGoodsName()));
                        if (value.getGoodsPrice() > priceAlert) {
                            value.setStatus(true);
                        }
                        out.collect(value);
                    }
                })
                .assignTimestampsAndWatermarks(new BoundedOutOfOrdernessTimestampExtractor<Product>(Time.seconds(0)) {
                    @Override
                    public long extractTimestamp(Product element) {
                        return element.getOrderTime();
                    }
                })
                ;
         //6.定义匹配模式，设置时间长度
        Pattern<Product, Product> pattern = Pattern.<Product>begin("begin")
                .where(new SimpleCondition<Product>() {
                    @Override
                    public boolean filter(Product value) throws Exception {
                        return value.getStatus() == true;
                    }
                })
                .next("next")
                .where(new SimpleCondition<Product>() {
                    @Override
                    public boolean filter(Product value) throws Exception {
                        return value.getStatus() == true;
                    }
                })
                .within(Time.seconds(60));
    
        //7.匹配模式（分组）
        PatternStream<Product> cep = CEP.pattern(waterData.keyBy(Product::getGoodsId), pattern);
    
        //8.查询告警数据
        cep.select(new PatternSelectFunction<Product, Object>() {
            @Override
            public Object select(Map<String, List<Product>> pattern) throws Exception {
                List<Product> result = pattern.get("next");
                return result;
            }
        }).print("告警数据：");
    
        env.execute();
    }

}

2.2.Bean对象
属性：goodsId、goodsPrice、goodsName、alias、orderTime、status
public class Product {
    private Long goodsId;
    private Double goodsPrice;
    private String goodsName;
    private String alias;
    private Long orderTime;
    private Boolean status;

    public Product(Long goodsId, Double goodsPrice, String goodsName, String alias, Long orderTime, Boolean status) {
        this.goodsId = goodsId;
        this.goodsPrice = goodsPrice;
        this.goodsName = goodsName;
        this.alias = alias;
        this.orderTime = orderTime;
        this.status = status;
    }
    
    @Override
    public String toString() {
        return "Product{" +
                "goodsId=" + goodsId +
                ", goodsPrice=" + goodsPrice +
                ", goodsName='" + goodsName + '\'' +
                ", alias='" + alias + '\'' +
                ", orderTime=" + orderTime +
                ", status=" + status +
                '}';
    }
    
    public Long getGoodsId() {
        return goodsId;
    }
    
    public void setGoodsId(Long goodsId) {
        this.goodsId = goodsId;
    }
    
    public Double getGoodsPrice() {
        return goodsPrice;
    }
    
    public void setGoodsPrice(Double goodsPrice) {
        this.goodsPrice = goodsPrice;
    }
    
    public String getGoodsName() {
        return goodsName;
    }
    
    public void setGoodsName(String goodsName) {
        this.goodsName = goodsName;
    }
    
    public String getAlias() {
        return alias;
    }
    
    public void setAlias(String alias) {
        this.alias = alias;
    }
    
    public Long getOrderTime() {
        return orderTime;
    }
    
    public void setOrderTime(Long orderTime) {
        this.orderTime = orderTime;
    }
    
    public Boolean getStatus() {
        return status;
    }
    
    public void setStatus(Boolean status) {
        this.status = status;
    }

}

本文参与腾讯云自媒体分享计划，分享自作者个人站点/博客。

原始发表：2022-05-06 ，如有侵权请联系 cloudcommunity@tencent.com 删除

kafka

腾讯云测试服务

本文分享自作者个人站点/博客前往查看

如有侵权，请联系 cloudcommunity@tencent.com 删除。

本文参与腾讯云自媒体分享计划，欢迎热爱写作的你一起参与！

kafka

腾讯云测试服务

登录后参与评论

0 条评论

热度