用于匹配和返回所有搜索字符串的SQL查询

我在一个列记录中有一个json文档，表如下所示。需要编写一个SQL查询，以在AAUP中显示字段a、b、k的所有值

结果应该是：

NAME1   age1    comment1
NAME2   age2    
NAME3            comment3

JSON数据：

{
    "reportfile": {
        "aaa": {
            "aaagroup": [{
                "a": "NAME1",
                "b": "age1",
                "k": "comment1"
            },
        {
                "a": "NAME2",
                "b": "age2"
            },
        {
                "a": "NAME3",
                "k": "comment3"
            }]
        },
        "dsa": {
            "dsagroup": [{
                "j": "Name"
            },
            {
                "j": "Title"
            }]
        }
    }
}

我对单个事件使用了以下查询：

数据：

查询：

select 
    substr(cc.BUS_NME, 1, strpos(cc.BUS_NME,'"')-1) as BUS_NME,
    substr(cc.AGE, 1, strpos(cc.AGE,'"')-1) as AGE
from 
    (substr(bb.aaa,strpos(bb.aaa,'"a":"')+5) as BUS_NME,
     substr(bb.aaa,strpos(bb.aaa,'"k":"')+5) as AGE 
from 
    (substr(aa.G, strpos(aa.G,'"aaagroup'),strpos(aa.G,'},')) as aaa                 
from 
    (select substr(record, strpos(record,'"aaagroup')) as G 
     from TABLE) aa) bb) cc

---

ush rani–如果我没有弄错您的问题，您将有一个这样的外部表，您可以在下面的查询中尝试从外部表获得所需的结果

外部表格示例：

CREATE EXTERNAL TABLE Ext_JSON_data(
reportfile string
  )
ROW FORMAT SERDE 
  'org.openx.data.jsonserde.JsonSerDe' 
WITH SERDEPROPERTIES (  
'serialization.format' = '1'
  )
LOCATION
  's3://bucket/folder/'

获取所需结果的查询：

WITH the_table AS (
SELECT CAST(social AS MAP(VARCHAR, JSON)) AS social_data
  FROM (
    VALUES
    (JSON '{"aaa": {"aaagroup": [{"a": "NAME1","b": "age1","k": "comment1"},{"a": "NAME2","b": "age2"},{"a": "NAME3","k": "comment3"}]},"dsa": {"dsagroup": [{"j": "Name"},{"j": "Title"}]}}')
) AS t (social)    
),
cte_first_level as
(  
SELECT 
    first_level_key
  ,CAST(first_level_value AS MAP(VARCHAR, JSON))As first_level_value
  FROM the_table
  CROSS JOIN UNNEST (social_data) AS t (first_level_key, first_level_value)
),
cte_second_level as
(
Select 
first_level_key
,SECOND_level_key
,SECOND_level_value
from 
cte_first_level  
CROSS JOIN UNNEST (first_level_value) AS t (SECOND_level_key, SECOND_level_value)
)
SELECT
first_level_key
,SECOND_level_key
,SECOND_level_value
,items
,items['a'] value_of_a
,items['b'] value_of_b
,items['k'] value_of_k
from 
cte_second_level
cross join unnest(cast(json_extract(SECOND_level_value, '$') AS ARRAY<MAP<VARCHAR, VARCHAR>>)) t (items)

查询输出：

---

请用您正在使用的RDBMS及其版本标记您的问题。JSON函数非常特定于供应商。它是一个拼花文件，加载到AWS Athena并创建一个表。

WITH the_table AS (
SELECT CAST(social AS MAP(VARCHAR, JSON)) AS social_data
  FROM (
    VALUES
    (JSON '{"aaa": {"aaagroup": [{"a": "NAME1","b": "age1","k": "comment1"},{"a": "NAME2","b": "age2"},{"a": "NAME3","k": "comment3"}]},"dsa": {"dsagroup": [{"j": "Name"},{"j": "Title"}]}}')
) AS t (social)    
),
cte_first_level as
(  
SELECT 
    first_level_key
  ,CAST(first_level_value AS MAP(VARCHAR, JSON))As first_level_value
  FROM the_table
  CROSS JOIN UNNEST (social_data) AS t (first_level_key, first_level_value)
),
cte_second_level as
(
Select 
first_level_key
,SECOND_level_key
,SECOND_level_value
from 
cte_first_level  
CROSS JOIN UNNEST (first_level_value) AS t (SECOND_level_key, SECOND_level_value)
)
SELECT
first_level_key
,SECOND_level_key
,SECOND_level_value
,items
,items['a'] value_of_a
,items['b'] value_of_b
,items['k'] value_of_k
from 
cte_second_level
cross join unnest(cast(json_extract(SECOND_level_value, '$') AS ARRAY<MAP<VARCHAR, VARCHAR>>)) t (items)