如何使用udf更新包含数组的spark数据框列

from pyspark.sql import SparkSession
from pyspark.sql.functions import udf
from pyspark.sql.types import ArrayType, StringType

spark = SparkSession.builder.getOrCreate()

def uppercase_array(arr):
    return [x.upper() for x in arr]

# 注册UDF
uppercase_udf = udf(uppercase_array, ArrayType(StringType()))

data = [("John", ["apple", "banana", "orange"]),
        ("Alice", ["grape", "kiwi", "mango"]),
        ("Bob", ["pear", "pineapple", "watermelon"])]

df = spark.createDataFrame(data, ["name", "fruits"])
df.show()

+-----+-------------------+
| name|             fruits|
+-----+-------------------+
| John|[apple, banana, or...|
|Alice| [grape, kiwi, mango]|
|  Bob|[pear, pineapple, ...|
+-----+-------------------+

df = df.withColumn("fruits_uppercase", uppercase_udf(df["fruits"]))
df.show()

+-----+-------------------+-----------------+
| name|             fruits| fruits_uppercase|
+-----+-------------------+-----------------+
| John|[apple, banana, or...|[APPLE, BANANA, ...|
|Alice| [grape, kiwi, mango]|[GRAPE, KIWI, MA...|
|  Bob|[pear, pineapple, ...|[PEAR, PINEAPPLE...|
+-----+-------------------+-----------------+