统一Spark中具有不同列数的两个表

import org.apache.spark.sql.SparkSession

// 创建SparkSession
val spark = SparkSession.builder()
  .appName("Merge Tables")
  .master("local")
  .getOrCreate()

// 创建表1
val table1 = spark.createDataFrame(Seq(
  (1, "John", 25),
  (2, "Alice", 30),
  (3, "Bob", 35)
)).toDF("id", "name", "age")

// 创建表2
val table2 = spark.createDataFrame(Seq(
  (4, "Tom"),
  (5, "Emily"),
  (6, "Mike")
)).toDF("id", "name")

// 添加缺失的列
val table1WithAge = table1.withColumn("salary", lit(5000))

// 删除多余的列
val table2WithoutAge = table2.drop("age")

// 对齐列顺序
val table2Aligned = table2WithoutAge.select("id", "name", lit(null).as("age"), lit(null).as("salary"))

// 合并两个表
val mergedTable = table1WithAge.union(table2Aligned)

// 显示合并后的表
mergedTable.show()