json - 从配置单元表中的 json 字符串中提取值

Question

我有如下数据

+----------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
|asset_id  |chg_log                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
+----------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
|4455628986|[{'oldValues': [], 'newValues': ['COMPO_MAIL'], 'fieldName': 'Communication Type', 'fieldType': 'TEXT', 'lookupInfo': {'fieldType': 'TEXT'}, 'additional': {'GROUP_QUALIFIER': [{'qualifiers': [{'lookupKey': 'Communication_Type', 'value': 'COMPO_MAIL'}]}]}}, {'oldValues': [], 'lookupInfo': {'isClientLevel': False, 'fieldType': 'DATE'}, 'fieldName': 'Delivery Due Date', 'fieldType': 'DATE', 'newValues': ['1601771520000']}, {'oldValues': [], 'lookupInfo': {'lookupType': 'CUST_ID', 'fieldType': 'CUST_ID'}, 'fieldName': 'Customer Id', 'fieldType': 'CUST_ID', 'newValues': ['10486']}, {'oldValues': [], 'lookupInfo': {'isClientLevel': False, 'fieldType': 'DROPDOWN'}, 'fieldName': 'Process_Status', 'fieldType': 'PICKLIST', 'newValues': ['Request Review']}]  |
+----------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

来自 列的 Json 字符串chg_log 我想提取 fieldName - Process_Status 的值，它是 newValues 作为请求审查。预期结果如下。

+----------+---------------+
|asset_id  |Process_Status |
+----------+---------------+
|4455628986|Request Review |
+----------+---------------+

json 每次的顺序都不相同。有时 Process_Status 首先出现在 json 字符串中，然后是 Communication Type 等等......
我尝试使用函数 json_extract 但无法获得它。
如何在 spark Sql 或 Pyspark 或 Hive 中实现这一点。有人能帮我吗？？
提前致谢

Answer 1

您的专栏chg_log似乎是一个字符串化的 Python 字典，而不是一个有效的 JSON 字符串。
在 Pyspark 中，您可以使用 UDF 将 dict 转换为 json，然后使用 from_json 将其转换为结构数组。最后过滤数组以找到字段 Process_Status :

import ast
from pyspark.sql import functions as F

dict_to_json = F.udf(lambda x: json.dumps(ast.literal_eval(x)))

df = df.withColumn("chg_log", dict_to_json(F.col("chg_log")))

df1 = df.withColumn(
    "chg_log",
    F.from_json("chg_log", F.schema_of_json(df.select("chg_log").head()[0]))
).withColumn(
    "chg_log",
    F.expr("filter(chg_log, x -> x.fieldName = 'Process_Status')")[0]
).select(
    F.col("asset_id"), F.col("chg_log.newValues").alias("Process_Status")
)

df1.show()
# +----------+----------------+
# |  asset_id|  Process_Status|
# +----------+----------------+
# |4455628986|[Request Review]|
# +----------+----------------+

直接在 UDF 中进行查找的另一种方法:

parse_status = F.udf(
    lambda x: next(i["newValues"] for i in ast.literal_eval(x) if i["fieldName"] == "Process_Status"),
    ArrayType(StringType())
)

df1 = df.select(F.col("asset_id"), parse_status(F.col("chg_log")).alias("Process_Status"))