Flink流批一体计算(23):Flink SQL之多流kafka写入多个mysql sink

目录

1. 准备工作

生成数据

创建数据表

2. 创建数据表

创建数据源表

创建数据目标表

3. 计算

WITH子句


1. 准备工作

生成数据

source kafka json 数据格式 :

topic  case_kafka_mysql:

{"ts": "20201011","id": 8,"price_amt":211}

topic  flink_test_2:

{"id": 8,"coupon_price_amt":100}

注意:针对双流中的每条记录都发触发

topic: case_kafka_mysql

docker exec -it 192d1369463a bash

bash-5.1# cd /opt/kafka_2.12-2.5.0/bin

bash-5.1# ./kafka-console-producer.sh --broker-list localhost:9092 --topic case_kafka_mysql

>{"ts": "20201011","id": 8,"price_amt":211}

topic: flink_test_2

docker exec -it 192d1369463a bash

bash-5.1# cd /opt/kafka_2.12-2.5.0/bin

bash-5.1# ./kafka-console-producer.sh --broker-list localhost:9092 --topic flink_test_2

>{"id": 8,"coupon_price_amt":100}

创建数据表

mysql 建表语句

CREATE TABLE `sync_test_2` (

  `id` bigint(11) NOT NULL AUTO_INCREMENT,

  `ts` varchar(64) DEFAULT NULL,

  `total_gmv` bigint(11) DEFAULT NULL,

  PRIMARY KEY (`id`),

  UNIQUE KEY `uidx` (`ts`) USING BTREE

) ENGINE=InnoDB AUTO_INCREMENT=5 DEFAULT CHARSET=utf8mb4;



CREATE TABLE `sync_test_22` (

  `id` bigint(11) NOT NULL AUTO_INCREMENT,

  `ts` varchar(64) DEFAULT NULL,

  `coupon_ratio` double DEFAULT NULL,

  PRIMARY KEY (`id`),

  UNIQUE KEY `uidx` (`ts`) USING BTREE

) ENGINE=InnoDB AUTO_INCREMENT=5 DEFAULT CHARSET=utf8mb4;

2. 创建数据表

创建数据源表
create table flink_test_2_1 (
  id BIGINT,
  ts VARCHAR,
  price_amt BIGINT,
  proctime AS PROCTIME ()
)
 with (
   'connector' = 'kafka',
   'topic' = 'case_kafka_mysql',
   'properties.bootstrap.servers' = '127.0.0.1:9092',
   'properties.group.id' = 'flink_gp_test2-1',
   'scan.startup.mode' = 'earliest-offset',
   'format' = 'json',
   'json.fail-on-missing-field' = 'false',
   'json.ignore-parse-errors' = 'true',
   'properties.zookeeper.connect' = '127.0.0.1:2181/kafka'
 );

create table flink_test_2_2 (
  id BIGINT,
  coupon_price_amt BIGINT,
  proctime AS PROCTIME ()
)
 with (
   'connector' = 'kafka',
   'topic' = 'flink_test_2',
   'properties.bootstrap.servers' = '127.0.0.1:9092',
   'properties.group.id' = 'flink_gp_test2-2',
   'scan.startup.mode' = 'earliest-offset',
   'format' = 'json',
   'json.fail-on-missing-field' = 'false',
   'json.ignore-parse-errors' = 'true',
   'properties.zookeeper.connect' = '127.0.0.1:2181/kafka'
 );

关键配置的说明:

json.fail-on-missing-field:在json缺失字段时是否报错

json.ignore-parse-errors:在解析json失败时是否报错

一般无法保证json格式,所以以上两个配置是比较重要的。

创建数据目标表
CREATE TABLE sync_test_2 (
                   ts string,
                   total_gmv bigint,
                   PRIMARY KEY (ts) NOT ENFORCED

 ) WITH (
   'connector' = 'jdbc',
   'url' = 'jdbc:mysql://127.0.0.1:3306/db01?characterEncoding=UTF-8',
   'table-name' = 'sync_test_2',
   'username' = 'root',
   'password' = 'Admin'
 );

CREATE TABLE sync_test_22 (
                   ts string,
                   coupon_ration bigint,
                   PRIMARY KEY (ts) NOT ENFORCED
 ) WITH (
   'connector' = 'jdbc',
   'url' = 'jdbc:mysql://127.0.0.1:3306/db01?characterEncoding=UTF-8',
   'table-name' = 'sync_test_2',
   'username' = 'root',
   'password' = 'Admin'
 );

3. 计算

一个作业中写入一个Sink或多个Sink

说明 写入多个Sink语句时,需要以BEGIN STATEMENT SET;开头,以END;结尾。

BEGIN STATEMENT SET;      --写入多个Sink时,必填。
INSERT INTO sync_test_2
SELECT
  ts,
  SUM(price_amt - coupon_price_amt) AS total_gmv
FROM
  (
    SELECT
      a.ts as ts,
      a.price_amt as price_amt,
      b.coupon_price_amt as coupon_price_amt
    FROM
      flink_test_2_1 as a
      LEFT JOIN flink_test_2_2 b on b.id = a.id
  )
GROUP BY ts;

INSERT INTO sync_test_22
SELECT
  ts,
  sum(coupon_price_amt)/sum(amount) AS coupon_ration
FROM
  (
    SELECT
      a.ts as ts,
      a.price_amt as price_amt,
      b.coupon_price_amt as coupon_price_amt
    FROM
      flink_test_2_1 as a
      LEFT JOIN flink_test_2_2 b on b.id = a.id
  )
GROUP BY ts;;
END;      --写入多个Sink时,必填。

WITH子句

WITH提供了一种编写辅助语句以用于更大的查询的方法。这些语句通常被称为公共表表达式(CTE),可以被视为定义仅针对一个查询存在的临时视图。

改写上述查询:

BEGIN STATEMENT SET;      --写入多个Sink时,必填。
with orders_with_coupon AS (
    SELECT
      a.ts as ts,
      a.price_amt as price_amt,
      b.coupon_price_amt as coupon_price_amt
    FROM
      flink_test_2_1 as a
      LEFT JOIN flink_test_2_2 b on b.id = a.id
)

INSERT INTO sync_test_2
SELECT
  ts,
  SUM(price_amt - coupon_price_amt) AS total_gmv
FROM orders_with_coupon
GROUP BY ts;

INSERT INTO sync_test_22
SELECT
  ts,
  coupon_price_amt/price_amt AS coupon_ration
FROM orders_with_coupon
GROUP BY ts;;
END;      --写入多个Sink时,必填。

  • 1
    点赞
  • 3
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
好的,您可以按照以下步骤操作: 1. 在 Flink 中添加 KafkaMySQL 的依赖: ```xml <dependency> <groupId>org.apache.flink</groupId> <artifactId>flink-connector-kafka_${scala.binary.version}</artifactId> <version>${flink.version}</version> </dependency> <dependency> <groupId>org.apache.flink</groupId> <artifactId>flink-connector-jdbc_${scala.binary.version}</artifactId> <version>${flink.version}</version> </dependency> <dependency> <groupId>mysql</groupId> <artifactId>mysql-connector-java</artifactId> <version>8.0.23</version> </dependency> ``` 2. 创建 Flink SQL 的执行环境: ```java StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment(); EnvironmentSettings settings = EnvironmentSettings.newInstance() .useBlinkPlanner() .inStreamingMode() .build(); StreamTableEnvironment tableEnv = StreamTableEnvironment.create(env, settings); ``` 3. 注册 Kafka 数据源和 MySQL 数据汇: ```java tableEnv.executeSql("CREATE TABLE kafka_source (\n" + " id INT,\n" + " name STRING,\n" + " age INT,\n" + " PRIMARY KEY (id) NOT ENFORCED\n" + ") WITH (\n" + " 'connector' = 'kafka',\n" + " 'topic' = 'test',\n" + " 'properties.bootstrap.servers' = 'localhost:9092',\n" + " 'properties.group.id' = 'testGroup',\n" + " 'format' = 'json',\n" + " 'scan.startup.mode' = 'earliest-offset'\n" + ")"); tableEnv.executeSql("CREATE TABLE mysql_sink (\n" + " id INT,\n" + " name STRING,\n" + " age INT,\n" + " PRIMARY KEY (id)\n" + ") WITH (\n" + " 'connector' = 'jdbc',\n" + " 'url' = 'jdbc:mysql://localhost:3306/test',\n" + " 'table-name' = 'user',\n" + " 'driver' = 'com.mysql.cj.jdbc.Driver',\n" + " 'username' = 'root',\n" + " 'password' = 'root'\n" + ")"); ``` 4. 使用 Flink SQL 读取 Kafka 数据源并将数据写入 MySQL 数据汇: ```java tableEnv.executeSql("INSERT INTO mysql_sink SELECT * FROM kafka_source"); env.execute(); ``` 这样就可以使用 Flink SQLKafka 中读取数据,并将数据写入 MySQL 数据库中了。

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值