本篇内容主要讲解“Spark-Streaming如何处理数据到MySQL中”,感兴趣的朋友不妨来看看。本文介绍的方法操作简单快捷,实用性强。下面就让小编来带大家学习“Spark-Streaming如何处理数据到mysql中”吧!
我们注重客户提出的每个要求,我们充分考虑每一个细节,我们积极的做好网站设计、网站建设服务,我们努力开拓更好的视野,通过不懈的努力,创新互联赢得了业内的良好声誉,这一切,也不断的激励着我们更好的服务客户。 主要业务:网站建设,网站制作,网站设计,小程序开发,网站开发,技术开发实力,DIV+CSS,PHP及ASP,ASP.Net,SQL数据库的技术开发工程师。
数据表如下:
create database test; use test; DROP TABLE IF EXISTS car_gps; CREATE TABLE IF NOT EXISTS car_gps( deployNum VARCHAR(30) COMMENT '调度编号', plateNum VARCHAR(10) COMMENT '车牌号', timeStr VARCHAR(20) COMMENT '时间戳', lng VARCHAR(20) COMMENT '经度', lat VARCHAR(20) COMMENT '纬度', dbtime TIMESTAMP DEFAULT CURRENT_TIMESTAMP COMMENT '数据入库时间', PRIMARY KEY(deployNum, plateNum, timeStr))
首先引入mysql的驱动
mysql mysql-connector-java 5.1.44
package com.hoult.Streaming.work import java.sql.{Connection, DriverManager, PreparedStatement} import java.util.Properties import com.hoult.structed.bean.BusInfo import org.apache.spark.sql.ForeachWriter class JdbcHelper extends ForeachWriter[BusInfo] { var conn: Connection = _ var statement: PreparedStatement = _ override def open(partitionId: Long, epochId: Long): Boolean = { if (conn == null) { conn = JdbcHelper.openConnection } true } override def process(value: BusInfo): Unit = { //把数据写入mysql表中 val arr: Array[String] = value.lglat.split("_") val sql = "insert into car_gps(deployNum,plateNum,timeStr,lng,lat) values(?,?,?,?,?)" statement = conn.prepareStatement(sql) statement.setString(1, value.deployNum) statement.setString(2, value.plateNum) statement.setString(3, value.timeStr) statement.setString(4, arr(0)) statement.setString(5, arr(1)) statement.executeUpdate() } override def close(errorOrNull: Throwable): Unit = { if (null != conn) conn.close() if (null != statement) statement.close() } } object JdbcHelper { var conn: Connection = _ val url = "jdbc:mysql://hadoop1:3306/test?useUnicode=true&characterEncoding=utf8" val username = "root" val password = "123456" def openConnection: Connection = { if (null == conn || conn.isClosed) { val p = new Properties Class.forName("com.mysql.jdbc.Driver") conn = DriverManager.getConnection(url, username, password) } conn } }
package com.hoult.Streaming.work import com.hoult.structed.bean.BusInfo import org.apache.spark.sql.{Column, DataFrame, Dataset, SparkSession} object KafkaToJdbc { def main(args: Array[String]): Unit = { System.setProperty("HADOOP_USER_NAME", "root") //1 获取sparksession val spark: SparkSession = SparkSession.builder() .master("local[*]") .appName(KafkaToJdbc.getClass.getName) .getOrCreate() spark.sparkContext.setLogLevel("WARN") import spark.implicits._ //2 定义读取kafka数据源 val kafkaDf: DataFrame = spark.readStream .format("kafka") .option("kafka.bootstrap.servers", "linux121:9092") .option("subscribe", "test_bus_info") .load() //3 处理数据 val kafkaValDf: DataFrame = kafkaDf.selectExpr("CAST(value AS STRING)") //转为ds val kafkaDs: Dataset[String] = kafkaValDf.as[String] //解析出经纬度数据,写入redis //封装为一个case class方便后续获取指定字段的数据 val busInfoDs: Dataset[BusInfo] = kafkaDs.map(BusInfo(_)).filter(_ != null) //将数据写入MySQL表 busInfoDs.writeStream .foreach(new JdbcHelper) .outputMode("append") .start() .awaitTermination() } }
kafka-topics.sh --zookeeper linux121:2181/myKafka --create --topic test_bus_info --partitions 2 --replication-factor 1 kafka-console-producer.sh --broker-list linux121:9092 --topic test_bus_info
到此,相信大家对“Spark-Streaming如何处理数据到mysql中”有了更深的了解,不妨来实际操作一番吧!这里是创新互联网站,更多相关内容可以进入相关频道进行查询,关注我们,继续学习!