阅读 135

FlinkSQL写入Kafka/ES/MySQL示例-JAVA

FlinkSQL写入Kafka/ES/MySQL示例-JAVA

一、背景说明

Flink的API做了4层的封装，上两层TableAPI、SQL语法相对简单便于编写，面对小需求可以快速上手解决，本文参考官网及部分线上教程编写source端、sink端代码，分别读取socket、kafka及文本作为source，并将流数据输出写入Kafka、ES及MySQL，方便后续查看使用。

二、代码部分

说明：这里使用connect及DDL两种写法，connect满足Flink1.10及以前版本使用，目前官方文档均是以DDL写法作为介绍，建议1.10以后的版本使用DDL写法操作，通用性更强。

1.读取（Source）端写法

1.1 基础环境建立，方便演示并行度为1且不设置CK

//建立Stream环境，设置并行度为1StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment().setParallelism(1);//建立Table环境StreamTableEnvironment tableEnv = StreamTableEnvironment.create(env);

1.2 读取Socket端口数据，并使用TableAPI及SQL两种方式查询

//读取服务器9999端口数据，并转换为对应JavaBeanSingleOutputStreamOperator<WaterSensor> mapDS = env.socketTextStream("hadoop102", 9999)
		.map(value -> {
                        String[] split = value.split(",");			return new WaterSensor(split[0]
					, Long.parseLong(split[1])
					, Integer.parseInt(split[2]));});//创建表：将流转换成动态表。Table table = tableEnv.fromDataStream(mapDS);//对动态表进行查询,TableAPI方式Table selectResult = table.where($("id").isEqual("ws_001")).select($("id"), $("ts"), $("vc"));//对动态表镜像查询，SQL方式-未注册表Table selectResult = tableEnv.sqlQuery("select * from " + table);

1.3 读取文本（FileSystem）数据，并使用TableAPI进行查询

//Flink1.10写法使用connect方式，读取txt文件并建立临时表tableEnv.connect(new FileSystem().path("input/sensor.txt"))
		.withFormat(new Csv().fieldDelimiter(',').lineDelimiter("\n"))
		.withSchema(new Schema().field("id", DataTypes.STRING())
					.field("ts", DataTypes.BIGINT())
				        .field("vc",DataTypes.INT()))
		.createTemporaryTable("sensor");//转换成表对象，对表进行查询。SQL写法参考Socket段写法Table table = tableEnv.from("sensor");
Table selectResult = table.groupBy($("id")).aggregate($("id").count().as("id_count"))select($("id"), $("id_count"));

1.4 消费Kafka数据，并使用TableAPI进行查询，分别用conncet及DDL写法

//Flink1.10写法使用connect方式，消费kafka对应主题并建立临时表tableEnv.connect(new Kafka().version("universal")
                            .topic("sensor")
			    .startFromLatest()
			    .property(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG,"hadoop102:9092")
			    .property(ConsumerConfig.GROUP_ID_CONFIG,"BD"))//消费者组
        .withSchema(new Schema().field("id", DataTypes.STRING())
		                .field("ts", DataTypes.BIGINT())
		                .field("vc",DataTypes.INT()))
		                .withFormat(new Csv())
		                .createTemporaryTable("sensor");//Flink1.10以后使用DDL写法tableEnv.executeSql("CREATE TABLE sensor (" +		"  `id` STRING," +		"  `ts` BIGINT," +		"  `vc` INT" +		") WITH (" +		"  'connector' = 'kafka'," +		"  'topic' = 'sensor'," +		"  'properties.bootstrap.servers' = 'hadoop102:9092'," +		"  'properties.group.id' = 'BD'," +		"  'scan.startup.mode' = 'latest-offset'," +		"  'format' = 'csv'" +		")");//转换成表对象，对表进行查询。SQL写法参考Socket段写法Table table = tableEnv.from("sensor");
Table selectResult = table.groupBy($("id")).aggregate($("id").count().as("id_count"))
		          .select($("id"), $("id_count"));

2.写入（Sink）端部分写法

2.1 写入文本文件

//创建表：创建输出表，connect写法tableEnv.connect(new FileSystem().path("out/sensor.txt"))
		.withFormat(new Csv())
		.withSchema(new Schema().field("id", DataTypes.STRING())
				.field("ts", DataTypes.BIGINT())
				.field("vc",DataTypes.INT()))
		.createTemporaryTable("sensor");//将数据写入到输出表中即实现sink写入，selectResult则是上面source侧查询出来的结果表selectResult.executeInsert("sensor");

2.2 写入Kafka

//connect写法tableEnv.connect(new Kafka().version("universal")
		.topic("sensor")
		.sinkPartitionerRoundRobin() //轮询写入
		.property(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG,"hadoop102:9092"))
		.withSchema(new Schema().field("id", DataTypes.STRING())
				        .field("ts", DataTypes.BIGINT())
				        .field("vc",DataTypes.INT()))
		.withFormat(new Json())
		.createTemporaryTable("sensor");//DDL写法tableEnv.executeSql("CREATE TABLE sensor (" +		"  `id` STRING," +		"  `ts` BIGINT," +		"  `vc` INT" +		") WITH (" +		"  'connector' = 'kafka'," +		"  'topic' = 'sensor'," +		"  'properties.bootstrap.servers' = 'hadoop102:9092'," +		"  'format' = 'json'" +		")");//将数据写入到输出表中即实现sink写入，selectResult则是上面source侧查询出来的结果表selectResult.executeInsert("sensor");

2.3 写入MySQL（JDBC方式，这里手动导入了mysql-connector-java-5.1.9.jar）

//DDLtableEnv.executeSql("CREATE TABLE sink_sensor (" +		"  id STRING," +		"  ts BIGINT," +		"  vc INT," +		"  PRIMARY KEY (id) NOT ENFORCED" +		") WITH (" +		"  'connector' = 'jdbc'," +		"  'url' = 'jdbc:mysql://hadoop102:3306/test?useSSL=false'," +		"  'table-name' = 'sink_test'," +		"  'username' = 'root'," +		"  'password' = '123456'" +		")");//将数据写入到输出表中即实现sink写入，selectResult则是上面source侧查询出来的结果表selectResult.executeInsert("sensor");

2.4 写入ES

//connect写法tableEnv.connect(new Elasticsearch()
		.index("sensor")
		.documentType("_doc")
		.version("7")
		.host("localhost",9200,"http")                //设置为1，每行数据都写入是方便客户端输出展示，生产勿使用
		.bulkFlushMaxActions(1)) 
		.withSchema(new Schema()
				.field("id", DataTypes.STRING())
				.field("ts", DataTypes.BIGINT())
				.field("vc",DataTypes.INT()))
		.withFormat(new Json())
		.inAppendMode()
		.createTemporaryTable("sensor");//DDL写法tableEnv.executeSql("CREATE TABLE sensor (" +					"  id STRING," +					"  ts BIGINT," +					"  vc INT," +					"  PRIMARY KEY (id) NOT ENFORCED" +					") WITH (" +					"  'connector' = 'elasticsearch-7'," +					"  'hosts' = 'http://localhost:9200'," +					"  'index' = 'users'," +					"  'sink.bulk-flush.max-actions' = '1')";)//将数据写入到输出表中即实现sink写入，selectResult则是上面source侧查询出来的结果表selectResult.executeInsert("sensor");

三、补充说明

依赖部分pom.xml

 <properties>
        <java.version>1.8</java.version>
        <maven.compiler.source>${java.version}</maven.compiler.source>
        <maven.compiler.target>${java.version}</maven.compiler.target>
        <flink.version>1.12.0</flink.version>
        <scala.version>2.12</scala.version>
        <hadoop.version>3.1.3</hadoop.version>
        <maven.compiler.source>8</maven.compiler.source>
        <maven.compiler.target>8</maven.compiler.target>
    </properties>

    <dependencies>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-java</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-clients_${scala.version}</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.elasticsearch</groupId>
            <artifactId>elasticsearch</artifactId>
            <version>7.8.0</version>
        </dependency>
        <!-- elasticsearch 的客户端 -->
        <dependency>
            <groupId>org.elasticsearch.client</groupId>
            <artifactId>elasticsearch-rest-high-level-client</artifactId>
            <version>7.8.0</version>
        </dependency>
        <!-- elasticsearch 依赖 2.x 的 log4j -->
        <dependency>
            <groupId>org.apache.logging.log4j</groupId>
            <artifactId>log4j-api</artifactId>
            <version>2.8.2</version>
        </dependency>
        <dependency>
            <groupId>org.apache.logging.log4j</groupId>
            <artifactId>log4j-core</artifactId>
            <version>2.8.2</version>
        </dependency>
        <dependency>
            <groupId>com.fasterxml.jackson.core</groupId>
            <artifactId>jackson-databind</artifactId>
            <version>2.9.9</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-elasticsearch7_${scala.version}</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.projectlombok</groupId>
            <artifactId>lombok</artifactId>
            <version>1.18.16</version>
            <scope>provided</scope>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-table-api-java-bridge_${scala.version}</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-table-planner-blink_${scala.version}</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-csv</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-connector-kafka_${scala.version}</artifactId>
            <version>${flink.version}</version>
        </dependency>
        <dependency>
            <groupId>org.apache.flink</groupId>
            <artifactId>flink-json</artifactId>
            <version>${flink.version}</version>
        </dependency>
    </dependencies></project>

学习交流，有任何问题还请随时评论指出交流。

来源https://www.cnblogs.com/rango-lhl/p/14900235.html

文章分类
后端
文章标签
flinksql
kafka
es
mysql
示例
java
版权声明：本站是系统测试站点，无实际运营。本文内容由互联网用户自发贡献，该文观点仅代表作者本人。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容， 请发送邮件至 XXXXXXo@163.com 举报，一经查实，本站将立刻删除。
相关推荐