使用spark/scala将JSON文件连接到数据帧中

import org.apache.spark.sql.{SparkSession, DataFrame}
import org.apache.spark.sql.functions._

val spark = SparkSession.builder()
  .appName("JSON to DataFrame")
  .getOrCreate()

val jsonFile = "path/to/json/file.json"
val jsonDataFrame = spark.read.json(jsonFile)

jsonDataFrame.printSchema()
jsonDataFrame.show()

val filteredDataFrame = jsonDataFrame.filter(col("age") > 25)
val aggregatedDataFrame = jsonDataFrame.groupBy("gender").agg(avg("age"))
val sortedDataFrame = jsonDataFrame.orderBy(desc("age"))

val outputPath = "path/to/output/file"
jsonDataFrame.write.parquet(outputPath)