如何在Pyspark 2.2.0中计算不包括周末的日期之间的差异

from pyspark.sql import SparkSession
from pyspark.sql.functions import udf
from pyspark.sql.types import DateType
from datetime import datetime, timedelta
import calendar

spark = SparkSession.builder.appName("DateDifference").getOrCreate()

def is_weekend(date):
    # 将字符串日期转换为datetime对象
    date_obj = datetime.strptime(date, "%Y-%m-%d")
    # 获取日期对应的星期几（0代表周一，6代表周日）
    weekday = date_obj.weekday()
    # 判断是否为周末（周六或周日）
    if weekday >= 5:
        return True
    else:
        return False

# 将自定义函数注册为Spark UDF
is_weekend_udf = udf(is_weekend, DateType())

start_date = "2022-01-01"
end_date = "2022-01-31"

date_range_df = spark.range(0, (datetime.strptime(end_date, "%Y-%m-%d") - datetime.strptime(start_date, "%Y-%m-%d")).days + 1, 1)\
    .selectExpr("date_add('{}', cast(id as int)) as date".format(start_date))

filtered_df = date_range_df.filter(~is_weekend_udf(date_range_df.date))

date_count = filtered_df.count()

from pyspark.sql import SparkSession
from pyspark.sql.functions import udf
from pyspark.sql.types import DateType
from datetime import datetime, timedelta
import calendar

spark = SparkSession.builder.appName("DateDifference").getOrCreate()

def is_weekend(date):
    date_obj = datetime.strptime(date, "%Y-%m-%d")
    weekday = date_obj.weekday()
    if weekday >= 5:
        return True
    else:
        return False

is_weekend_udf = udf(is_weekend, DateType())

start_date = "2022-01-01"
end_date = "2022-01-31"

date_range_df = spark.range(0, (datetime.strptime(end_date, "%Y-%m-%d") - datetime.strptime(start_date, "%Y-%m-%d")).days + 1, 1)\
    .selectExpr("date_add('{}', cast(id as int)) as date".format(start_date))

filtered_df = date_range_df.filter(~is_weekend_udf(date_range_df.date))

date_count = filtered_df.count()

print("不包括周末的日期数量为：", date_count)