使用Scala/Spark列出目录中的文件(包括文件信息)

import org.apache.spark.sql.SparkSession
import java.io.File

val spark = SparkSession.builder()
  .appName("ListFiles")
  .master("local")
  .getOrCreate()

def listFiles(directory: String): Array[File] = {
  val dir = new File(directory)
  if (dir.exists && dir.isDirectory) {
    dir.listFiles.filter(_.isFile)
  } else {
    Array[File]()
  }
}

val directory = "/path/to/directory"
val files = listFiles(directory)

files.foreach { file =>
  println("文件名：" + file.getName)
  println("路径：" + file.getAbsolutePath)
  println("大小：" + file.length)
  println("最后修改时间：" + file.lastModified)
  println("-----")
}