SparkSQL学习案例:使用DataFrame和Dataset操作json数据

一、测试数据集(奥特曼.json)

 

 

二、源代码

 1 import org.apache.spark.sql.SparkSession
 2 
 3 //在Scala中,样例类在编译时会默认实现Product特质
 4 case class Ultraman(name: String, age: BigInt, address: Array[String])
 5 
 6 object DatasetAndDataFrameExample {
 7 
 8   def main(args: Array[String]): Unit = {
 9 
10     //实例化SparkSession
11     val spark = SparkSession
12       .builder()
13       .master("local[*]")
14       .appName("DatasetAndDataFrameExample")
15       .getOrCreate()
16 
17     //为避免影响输出结果,设置日志打印等级为"WARN"
18     spark.sparkContext.setLogLevel("WARN")
19 
20     import spark.implicits._
21 
22     val df1 = spark.read.json("/home/liuyin/IdeaProjects/Spark/src/main/scala/chap07_SparkSQL/奥特曼.json")
23     df1.show()
24     df1.filter($"address" === Array("M78")).filter($"age" > 10000).show()
25 
26     val ds1 = spark.read.json("/home/liuyin/IdeaProjects/Spark/src/main/scala/chap07_SparkSQL/奥特曼.json").as[Ultraman]
27     ds1.show()
28     ds1.filter(_.name == "迪迦").show()
29 
30     spark.stop()
31   }
32 }

 

 

三、输出结果