tryParseLocalDate 字符串轉化為日期 minusMonths 減一個月,3月15調用結果為2月15;3月31的調用結果為2月28(如果是閏年則是29) plusMonths 加一個月,對返回日期的處理同上 dayOfWeek 返回星期幾,返回值為Property對象,需要調用get方法獲取int值 withMinimumValue 對某個屬性加工,把這個屬性變成范圍內最小值,返回修改之后的時間,例如:tryParseLocalDate("2020-6-30").dayOfWeek().withMinimumValue() 返回"2020-6-29",因為29號是星期一,即一周的最小值;該方法同樣適用於獲取月份第一天 withMaximumValue 對某個屬性加工,把這個屬性變成范圍內最大值,返回修改之后的時間,例如:tryParseLocalDate("2020-6-30").dayOfWeek().withMaximumValue() 返回"2020-7-05",因為7月5號是星期日,即一周的最大值;該方法同樣適用於獲取月份最后一天 someDate.dayOfMonth().withMaximumValue() 返回someDate對應月份的最后一天
Spark 測試類
package com.xx.xxx
import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.functions.{substring, when}
object AgeGroupUtils {
// def get5YearsAgeGroup(age: Integer): String = {
// val ageGroup = age match {
// case age if (age == null || age < 15) => SparkConfig.UNSPECIFIED
// case age if (age <= 25 && age >= 15) => SparkConfig.AGE_15TO25
// case age if (age <= 35 && age >= 26) => SparkConfig.AGE_26TO35
// case age if (age <= 45 && age >= 36) => SparkConfig.AGE_36TO45
// case age if (age <= 55 && age >= 46) => SparkConfig.AGE_46TO55
// case age if (age <= 100 && age >= 56) => SparkConfig.AGE_56TO100
// case age if (age > 100) => SparkConfig.UNSPECIFIED
// case _ => SparkConfig.UNSPECIFIED
// }
// ageGroup
}
def main(args: Array[String]): Unit = {
val spark = SparkSession.builder
.master("local")
.appName("appName")
.getOrCreate()
import spark.implicits._
var data3 = Seq(
(null, "11", "22", "null", "1"),
(null, "11", "22", "2020-01-02", "2"),
("666","10010", "19219", "5666", "3")
).toDF("DRCN_BirthYear", "DRCN_AgeFrom", "DRCN_AgeTo", "DRCN_RegDate", "id")
.select(
$"id",
$"DRCN_RegDate",
substring($"DRCN_RegDate",0,4) as "tmp",
substring($"DRCN_RegDate",0,4)-($"DRCN_AgeFrom"+$"DRCN_AgeTo")/2 as "Y",
when($"DRCN_BirthYear".isNotNull or $"DRCN_AgeFrom".isNull or $"DRCN_AgeTo".isNull,$"DRCN_BirthYear")
.otherwise((substring($"DRCN_RegDate",0,4)-($"DRCN_AgeFrom"+$"DRCN_AgeTo")/2).cast("int")) as "BirthYear"
)
data3.show()
var m: Integer = null;
// println(get5YearsAgeGroup(m))
// println(get5YearsAgeGroup(33))
// println(get5YearsAgeGroup(66))
// println(get5YearsAgeGroup(166))
}
}
