SparkSQL -两个时间戳之间的差异(分钟)

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

// 创建一个包含时间戳的DataFrame
val df = spark.createDataFrame(Seq(
  ("2022-01-01 12:00:00"),
  ("2022-01-01 12:05:00"),
  ("2022-01-01 12:10:00")
)).toDF("timestamp")

// 将字符串类型的时间戳转换为Timestamp类型
val timestampFormat = "yyyy-MM-dd HH:mm:ss"
val dfWithTimestamp = df.withColumn("timestamp", to_timestamp(col("timestamp"), timestampFormat))

// 计算两个时间戳之间的差异（分钟）
val diffInMinutes = dfWithTimestamp.select(
  col("timestamp").alias("start_timestamp"),
  lead(col("timestamp"), 1).over(Window.orderBy("timestamp")).alias("end_timestamp")
).withColumn("diff_minutes", (col("end_timestamp").cast("long") - col("start_timestamp").cast("long")) / 60)

// 显示结果
diffInMinutes.show()