将spark数据帧中的列聚合为json

from pyspark.sql import SparkSession
from pyspark.sql.functions import to_json

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

# 创建示例数据帧
data = [("Alice", 25, "Female"), ("Bob", 30, "Male"), ("Charlie", 35, "Male")]
df = spark.createDataFrame(data, ["Name", "Age", "Gender"])

# 将数据帧的列聚合为JSON
json_df = df.select(to_json(df.columns).alias("json"))

# 显示结果
json_df.show(truncate=False)

+-----------------------------+
|json                         |
+-----------------------------+
|{"Name":"Alice","Age":25,...}|
|{"Name":"Bob","Age":30,...}  |
|{"Name":"Charlie","Age":35...|
+-----------------------------+