将不同格式的日期从字符串格式解析为日期格式pyspark when子句

from pyspark.sql import SparkSession
from pyspark.sql.functions import when, to_date

# 创建SparkSession
spark = SparkSession.builder.getOrCreate()

# 创建示例数据集
data = [("2022-01-01",), ("01/01/2022",), ("2022-01-01 12:00:00",)]

# 创建DataFrame
df = spark.createDataFrame(data, ["date_string"])

# 使用when子句解析日期格式
df = df.withColumn("date", when(df.date_string.contains("-"), to_date(df.date_string, "yyyy-MM-dd"))
                   .when(df.date_string.contains("/"), to_date(df.date_string, "MM/dd/yyyy"))
                   .otherwise(to_date(df.date_string, "yyyy-MM-dd HH:mm:ss")))

# 显示结果
df.show()