Scala:使用嵌套json结构转换和替换Spark DataFrame的值

import org.apache.spark.sql.functions._
import org.apache.spark.sql.types._

// 定义嵌套JSON结构的Schema
val schema = StructType(Seq(
  StructField("name", StringType),
  StructField("age", IntegerType),
  StructField("address", StructType(Seq(
    StructField("street", StringType),
    StructField("city", StringType),
    StructField("state", StringType)
  )))
))

// 创建示例DataFrame
val data = Seq(
  ("John", 25, Row("123 Main St", "New York", "NY")),
  ("Jane", 30, Row("456 Elm St", "San Francisco", "CA"))
)
val df = spark.createDataFrame(data).toDF("name", "age", "address")

// 定义转换和替换函数
val transformAddress = udf((address: Row) => {
  val newStreet = address.getAs[String]("street").toUpperCase
  val newCity = address.getAs[String]("city").toLowerCase
  val newState = address.getAs[String]("state").toUpperCase
  Row(newStreet, newCity, newState)
})

// 应用转换和替换函数
val transformedDF = df.withColumn("address", transformAddress(col("address")))

// 显示转换后的结果
transformedDF.show()