SQL - 解析没有列标题的 JSON 数组答案

【问题标题】：SQL - Parse JSON Array without column headersSQL - 解析没有列标题的 JSON 数组
【发布时间】：2020-04-03 19:50:56
【问题描述】：

我有以下 json 负载：

"rows": [["202003290528", "/home", "/home", "(not set)", "/home", "desktop", "Charlotte NC", "1", "1", "1", "0.0", "1", "1", "0.0", "1"], ["202003291516", "/home", "/home", "(not set)", "/home", "mobile", "Chicago IL", "1", "1", "1", "0.0", "1", "0", "0.0", "1"], ["202003291930", "/home", "/home", "(not set)", "/home", "mobile", "Boston MA-Manchester NH", "1", "1", "1", "0.0", "1", "0", "0.0", "1"], ["202003291942", "/home", "/home", "(not set)", "/home", "mobile", "Des Moines-Ames IA", "1", "1", "1", "0.0", "1", "0", "0.0", "1"]]

如何使用 SQL Server 2016 解析它以填充 SQL 表：

dateHourMinute  pagePath    landingPagePath secondPagePath  exitPagePath    deviceCategory  metro   sessions    pageviews   uniquePageviews timeOnPage  users   newUsers    sessionDuration bounces

【问题讨论】：

标签： sql json tsql sql-server-2016

【解决方案1】：

您需要将OPENJSON() 与显式架构一起使用。只需在WITH 子句中使用适当的列名即可：

声明：

DECLARE @json nvarchar(max) = N'{
   "rows":[
      [
         "202003290528",
         "/home",
         "/home",
         "(not set)",
         "/home",
         "desktop",
         "Charlotte NC",
         "1",
         "1",
         "1",
         "0.0",
         "1",
         "1",
         "0.0",
         "1"
      ],
      [
         "202003291516",
         "/home",
         "/home",
         "(not set)",
         "/home",
         "mobile",
         "Chicago IL",
         "1",
         "1",
         "1",
         "0.0",
         "1",
         "0",
         "0.0",
         "1"
      ],
      [
         "202003291930",
         "/home",
         "/home",
         "(not set)",
         "/home",
         "mobile",
         "Boston MA-Manchester NH",
         "1",
         "1",
         "1",
         "0.0",
         "1",
         "0",
         "0.0",
         "1"
      ],
      [
         "202003291942",
         "/home",
         "/home",
         "(not set)",
         "/home",
         "mobile",
         "Des Moines-Ames IA",
         "1",
         "1",
         "1",
         "0.0",
         "1",
         "0",
         "0.0",
         "1"
      ]
   ]
}'

声明：

SELECT *
FROM OPENJSON(@json, '$.rows') WITH (
   dateHourMinute varchar(100) '$[0]',
   pagePath varchar(100) '$[1]',
   landingPagePath varchar(100) '$[2]',
   secondPagePath varchar(100) '$[3]',
   exitPagePath varchar(100) '$[4]',
   deviceCategory varchar(100) '$[5]',
   metro varchar(100) '$[6]',
   sessions varchar(100) '$[7]',
   pageviews varchar(100) '$[8]',
   uniquePageviews varchar(100) '$[9]',
   timeOnPage varchar(100) '$[10]',
   users varchar(100) '$[11]',
   newUsers varchar(100) '$[12]',
   sessionDuration varchar(100) '$[13]',
   bounces varchar(100) '$[14]'
)

【讨论】：