如何在PySpark中转换数据框列中的JSON字符串？

from pyspark.sql import SparkSession
from pyspark.sql.functions import from_json
from pyspark.sql.types import StructType, StructField, StringType

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

# 创建示例数据框
data = [("John", '{"age": 30, "city": "New York"}'),
        ("Alice", '{"age": 25, "city": "San Francisco"}')]
df = spark.createDataFrame(data, ["name", "json_string"])

# 定义目标数据结构的模式
schema = StructType([
    StructField("age", StringType()),
    StructField("city", StringType())
])

# 使用from_json函数转换JSON字符串列
df = df.withColumn("json_struct", from_json(df.json_string, schema))

# 显示转换后的数据框
df.show(truncate=False)

+-----+-------------------+----------------+
|name |json_string        |json_struct     |
+-----+-------------------+----------------+
|John |{"age": 30, "city": "New York"}|{30, New York}  |
|Alice|{"age": 25, "city": "San Francisco"}|{25, San Francisco}|
+-----+-------------------+----------------+