0

我在 sql 表列中有一个嵌套的 JSON，需要从“delinquincies”数组中解析出多个值。在下面的 JSON 中用 ** 突出显示

   {
   "creditReportDetails":[
      {
         "accountStatusSummary":{
            "accounts":[
               {
                  "amountOwed":0,
                  "amountPastDue":0,
                  "type":"Open-ended"
               },
               {
                  "amountOwed":1234,
                  "amountPastDue":0,
                  "type":"Revolving"
               },
               {
                  "amountOwed":123456,
                  "amountPastDue":0,
                  "type":"Installment"
               },
               {
                  "amountOwed":123456,
                  "amountPastDue":0,
                  "type":"Total"
               }
            ],
            **"delinquincies":[
               {
                  "badDebt":null,
                  "paysAccountasAgreed":34,
                  "paysPaid120Daysor4PaymentsPastDue":null,
                  "paysPaid3060DaysorMax2PaymentsPastDue":null,
                  "paysPaid6090DaysorMax3PaymentsPastDue":null,
                  "paysPaid90120DaysorMax4PaymentsPastDue":null,
                  "repossession":null,
                  "statusNotKnown":null,
                  "type":"current"
               },
               {
                  "badDebt":null,
                  "paysAccountasAgreed":null,
                  "paysPaid120Daysor4PaymentsPastDue":null,
                  "paysPaid3060DaysorMax2PaymentsPastDue":null,
                  "paysPaid6090DaysorMax3PaymentsPastDue":null,
                  "paysPaid90120DaysorMax4PaymentsPastDue":null,
                  "repossession":null,
                  "statusNotKnown":null,
                  "type":"history"
               }
            ]**
         },
         "collections":[

         ],
         "datePulled":"01/01/2020",
         "employmentHistory":[
            {
               "employer":"Some Company",
               "historyType":"Current"
            },
            {
               "employer":"Some Company",
               "historyType":"Former"
            }
         ],
         "personalInfoFromVendor":{
            "address":{
               "city":"Anywhere",
               "line1":"123 Anywhere",
               "line2":null,
               "postalZipCode":"12345",
               "territoryCode":"AB"
            },
            "dateOfBirth":"01/01/1900",
            "firstName":"ABCD",
            "lastName":"ABCD",
            "middleName":"C",
            "ssn":"1231456789",
            "suffix":null
         },
         "reportMessaging":{
            "duplicate":"No",
            "processingStatus":"Complete",
            "referenceNumber":"123456789"
         },
         "reportSummary":{
            "bankruptcies":"No",
            "collectionItems":"No",
            "expNatlRiskScore":"123",
            "publicRecords":"No"
         }
      }
   ],
   "personalCreditReportUrl":"https://abcdcreditreport.lmig.com/creditreport-details/CRS123456789012",
   "subjectId":"CRS123456789012",
   "totalCount":1,
   "CreditReportBaseResponse":{
      "lexisNexisReferenceNumber":"123456789012",
      "status":{
         "statusCode":"COMPLETE",
         "statusDescription":[
            "BALANCES ON REVOLVING ACCOUNTS",
            "BALANCE-TO-LIMIT RATIOS ON ACCOUNTS"
         ]
      },
      "subjectId":"CRS123456789012"
   }
}

我一直在玩各种 JSON_Value 的迭代，感觉好像我错过了一些简单的东西

Select id,
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].badDebt') as 'Current_Baddebt',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].paysAccountasAgreed') as 'Current_PaysAccountasAgreed',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].paysAccountasAgreed') as 'Current_PaysAccountasAgreed',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].paysPaid3060DaysorMax2PaymentsPastDue') as 'Current_Paid3060DaysorMax2PaymentsPastDue',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].payspaysPaid6090DaysorMax3PaymentsPastDue') as 'Current_Paid6090DaysorMax3PaymentsPastDue',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].paysPaid120Daysor4PaymentsPastDue') as 'Current_paysPaid120Daysor4PaymentsPastDue',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].repossession') as 'Current_repossession',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[0].repossession') as 'Current_statusNotKnown',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].badDebt') as 'Hisorical_Baddebt',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].paysAccountasAgreed') as 'Hisorical_PaysAccountasAgreed',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].paysAccountasAgreed') as 'Hisorical_PaysAccountasAgreed',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].paysPaid3060DaysorMax2PaymentsPastDue') as 'Hisorical_Paid3060DaysorMax2PaymentsPastDue',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].payspaysPaid6090DaysorMax3PaymentsPastDue') as 'Hisorical_Paid6090DaysorMax3PaymentsPastDue',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].paysPaid120Daysor4PaymentsPastDue') as 'Hisorical_paysPaid120Daysor4PaymentsPastDue',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].repossession') as 'Hisorical_repossession',
       JSON_VALUE (json_column, '$.creditReportDetails.accountStatusSummary.delinquincies[1].repossession') as 'Hisorical_statusNotKnown'
from Database.dbo.Credit

期望 Null 值和 Current_paysAccountasAgreed 34

4

1 回答 1

0

以下是如何处理违规行为。

由于嵌套的 JSON 结构，我们需要CROSS APPLY OPENJSON(...)多次使用子句。

SQL

-- DDL and sample data population, start
DECLARE @tbl TABLE (ID INT IDENTITY PRIMARY KEY, json_column NVARCHAR(MAX));
INSERT INTO @tbl (json_column)
VALUES
(N'{
   "creditReportDetails":[
      {
         "accountStatusSummary":{
            "accounts":[
               {
                  "amountOwed":0,
                  "amountPastDue":0,
                  "type":"Open-ended"
               },
               {
                  "amountOwed":1234,
                  "amountPastDue":0,
                  "type":"Revolving"
               },
               {
                  "amountOwed":123456,
                  "amountPastDue":0,
                  "type":"Installment"
               },
               {
                  "amountOwed":123456,
                  "amountPastDue":0,
                  "type":"Total"
               }
            ],
            "delinquincies":[
               {
                  "badDebt":null,
                  "paysAccountasAgreed":34,
                  "paysPaid120Daysor4PaymentsPastDue":null,
                  "paysPaid3060DaysorMax2PaymentsPastDue":null,
                  "paysPaid6090DaysorMax3PaymentsPastDue":null,
                  "paysPaid90120DaysorMax4PaymentsPastDue":null,
                  "repossession":null,
                  "statusNotKnown":null,
                  "type":"current"
               },
               {
                  "badDebt":null,
                  "paysAccountasAgreed":null,
                  "paysPaid120Daysor4PaymentsPastDue":null,
                  "paysPaid3060DaysorMax2PaymentsPastDue":null,
                  "paysPaid6090DaysorMax3PaymentsPastDue":null,
                  "paysPaid90120DaysorMax4PaymentsPastDue":null,
                  "repossession":null,
                  "statusNotKnown":null,
                  "type":"history"
               }
            ]
         },
         "collections":[

         ],
         "datePulled":"01/01/2020",
         "employmentHistory":[
            {
               "employer":"Some Company",
               "historyType":"Current"
            },
            {
               "employer":"Some Company",
               "historyType":"Former"
            }
         ],
         "personalInfoFromVendor":{
            "address":{
               "city":"Anywhere",
               "line1":"123 Anywhere",
               "line2":null,
               "postalZipCode":"12345",
               "territoryCode":"AB"
            },
            "dateOfBirth":"01/01/1900",
            "firstName":"ABCD",
            "lastName":"ABCD",
            "middleName":"C",
            "ssn":"1231456789",
            "suffix":null
         },
         "reportMessaging":{
            "duplicate":"No",
            "processingStatus":"Complete",
            "referenceNumber":"123456789"
         },
         "reportSummary":{
            "bankruptcies":"No",
            "collectionItems":"No",
            "expNatlRiskScore":"123",
            "publicRecords":"No"
         }
      }
   ],
   "personalCreditReportUrl":"https://abcdcreditreport.lmig.com/creditreport-details/CRS123456789012",
   "subjectId":"CRS123456789012",
   "totalCount":1,
   "CreditReportBaseResponse":{
      "lexisNexisReferenceNumber":"123456789012",
      "status":{
         "statusCode":"COMPLETE",
         "statusDescription":[
            "BALANCES ON REVOLVING ACCOUNTS",
            "BALANCE-TO-LIMIT RATIOS ON ACCOUNTS"
         ]
      },
      "subjectId":"CRS123456789012"
   }
}');
-- DDL and sample data population, end

SELECT ID
    , JSON_VALUE (delinq.[Value], '$.badDebt') as [badDebt]
    , JSON_VALUE (delinq.[Value], '$.paysAccountasAgreed') as [paysAccountasAgreed]
    -- put here everything else in between
    , JSON_VALUE (delinq.[Value], '$.type') as [type]
FROM @tbl AS tbl
    CROSS APPLY OPENJSON(tbl. json_column, 'strict $.creditReportDetails') AS cr
    CROSS APPLY OPENJSON(cr.value, 'strict $.accountStatusSummary.delinquincies') AS delinq;

输出

+----+---------+---------------------+---------+
| ID | badDebt | paysAccountasAgreed |  type   |
+----+---------+---------------------+---------+
|  1 | NULL    | 34                  | current |
|  1 | NULL    | NULL                | history |
+----+---------+---------------------+---------+
于 2020-06-03T17:44:18.797 回答