conectorShare.md

connector 共用

NOTE:新版chunjun支持connector和flinkSql自带的connector共用。

在chunjun中使用flinkSql的connector

• 本地调试
  • 在chunjun-local-test模块下将对应的flink connector的GAV拷贝到pom.xml文件中
  • 将chunjun中的connector的GAV拷贝到pom.xml文件中(部分flink connector已经通过chunjun connector引入,如果是则上一步省略)
  • 在LocalTest类中指定参数运行即可 sql:

      -- {"id":100,"name":"lb james阿道夫","money":293.899778,"dateone":"2020-07-30 10:08:22","age":"33","datethree":"2020-07-30 10:08:22.123","datesix":"2020-07-30 10:08:22.123456","datenigth":"2020-07-30 10:08:22.123456789","dtdate":"2020-07-30","dttime":"10:08:22"}
      CREATE TABLE source_ods_fact_user_ippv 
      (
          id INT
          , name STRING
          , money decimal
          , dateone timestamp
          , age bigint
          , datethree timestamp
          , datesix timestamp(6)
          , datenigth timestamp(9)
          , dtdate date
          , dttime time
      ) WITH (
          'connector' = 'kafka'
          ,'topic' = 'da'
          ,'properties.bootstrap.servers' = 'localhost:9092'
          ,'properties.group.id' = 'luna_g'
          ,'scan.startup.mode' = 'earliest-offset'
          ,'format' = 'json'
          ,'json.timestamp-format.standard' = 'SQL'
      );
      
      CREATE TABLE result_total_pvuv_min
      (
          id INT
          , name STRING
          , money decimal
          , dateone timestamp
          , age bigint
          , datethree timestamp
          , datesix timestamp(6)
          , datenigth timestamp(9)
          , dtdate date
          , dttime time
      ) WITH (
          'connector' = 'stream-x'
      );
      INSERT INTO result_total_pvuv_min
      SELECT *
      from source_ods_fact_user_ippv;

    pom:

    <dependency>
        <groupId>org.apache.flink</groupId>
        <artifactId>flink-connector-kafka_2.12</artifactId>
        <version>1.12.2</version>
        <exclusions>
            <exclusion>
            <artifactId>flink-core</artifactId>
            <groupId>org.apache.flink</groupId>
            </exclusion>
        </exclusions>
    </dependency>
    <dependency>
        <groupId>com.dtstack.chunjun</groupId>
        <artifactId>chunjun-connector-stream</artifactId>
        <version>1.12-SNAPSHOT</version>
    </dependency>
    

• 服务器上运行:

  • 1.将需要的connector的jar包，和其他(包括chunjun-core-feat_1.12_pluginMerge.jar，如果kafka中用到了json也需要将json的format jar包)拷贝到chunjun/lib目录下,构建jobGraph使用

  • 2.将需要的connector的jar包，和其他(包括chunjun-core-feat_1.12_pluginMerge.jar，如果kafka中用到了json也需要将json的format jar包)拷贝到flink/lib下

  • 3.在在chunjun/lib下目录下执行命令：

    • local模式

    java -cp  "./*" com.dtstack.chunjun.client.Launcher \
    -mode local \
    -jobType sql \
    -jobName flink1.12_SPI \
    -job /yourjobpath/sqlFile.sql \
    -chunjunDistDir /chunjun/chunjun-dist

    

    • yarnPer模式

    java -cp  "./*" com.dtstack.chunjun.client.Launcher \
    -mode yarnPer \
    -jobType sql \
    -jobName flink1.12_SPI \
    -job /yourjobpath/sqlFile.sql \
    -chunjunDistDir /chunjun/chunjun-dist \
    -flinkConfDir /flink/conf \
    -hadoopConfDir /hadoop/etc \
    -flinkLibDir /flink/lib \
    -queue default

    

在flinkSql中使用chunjun的connector

• 本地调试
  • 在自己项目中将对应的flink connector的GAV拷贝到pom.xml文件中
  • 将chunjun中的core和connector的GAV拷贝到pom.xml文件中(需要先deploy项目)
  • 运行自己的任务 sql:

      -- {"id":100,"name":"lb james阿道夫","money":293.899778,"dateone":"2020-07-30 10:08:22","age":"33","datethree":"2020-07-30 10:08:22.123","datesix":"2020-07-30 10:08:22.123456","datenigth":"2020-07-30 10:08:22.123456789","dtdate":"2020-07-30","dttime":"10:08:22"}
      CREATE TABLE source_ods_fact_user_ippv 
      (
          id INT
          , name STRING
          , money decimal
          , dateone timestamp
          , age bigint
          , datethree timestamp
          , datesix timestamp(6)
          , datenigth timestamp(9)
          , dtdate date
          , dttime time
      ) WITH (
          'connector' = 'kafka'
          ,'topic' = 'da'
          ,'properties.bootstrap.servers' = 'localhost:9092'
          ,'properties.group.id' = 'luna_g'
          ,'scan.startup.mode' = 'earliest-offset'
          ,'format' = 'json'
          ,'json.timestamp-format.standard' = 'SQL'
      );
      
      CREATE TABLE result_total_pvuv_min
      (
          id INT
          , name STRING
          , money decimal
          , dateone timestamp
          , age bigint
          , datethree timestamp
          , datesix timestamp(6)
          , datenigth timestamp(9)
          , dtdate date
          , dttime time
      ) WITH (
          'connector' = 'stream-x'
      );
      INSERT INTO result_total_pvuv_min
      SELECT *
      from source_ods_fact_user_ippv;

  pom:

      <dependency>
          <groupId>org.apache.flink</groupId>
          <artifactId>flink-sql-connector-kafka_2.12</artifactId>
          <version>1.12.2</version>
      </dependency>
      <dependency>
          <groupId>com.dtstack.chunjun</groupId>
          <artifactId>chunjun-core</artifactId>
          <version>1.12-SNAPSHOT</version>
      </dependency>
      <dependency>
          <groupId>com.dtstack.chunjun</groupId>
          <artifactId>chunjun-connector-stream</artifactId>
          <version>1.12-SNAPSHOT</version>
      </dependency>
  

• 服务器上运行:
  • 将flink需要的connector和chunjun的chunjun-core-feat_1.12_pluginMerge.jar包和对应connector的jar引入到自己项目的pom中，将项目打成fat包，提交任务即可。