如何在RDD中选择每个键时间最早的数据？

# 导入必要的库
from datetime import datetime
from pyspark import SparkContext

# 创建SparkContext
sc = SparkContext("local", "RDD Example")

# 创建一个包含键值对的RDD
data = [("key1", "data1", datetime(2022, 1, 1)),
        ("key2", "data2", datetime(2022, 2, 1)),
        ("key1", "data3", datetime(2022, 3, 1)),
        ("key2", "data4", datetime(2022, 4, 1))]

rdd = sc.parallelize(data)

# 使用groupByKey()将RDD中的数据按键分组
grouped_rdd = rdd.groupByKey()

# 使用reduceByKey()选择每个键时间最早的数据
earliest_data_rdd = grouped_rdd.reduceByKey(lambda x, y: x if x[2] < y[2] else y)

# 将结果返回到驱动程序并打印
result = earliest_data_rdd.collect()
for key, value in result:
    print("Key:", key)
    print("Earliest Data:", value)

# 关闭SparkContext
sc.stop()