如何在 SQL Server 2005 中的变量中选择多个行值？答案

【问题标题】：How to select multiple row values in to a variable in SQL Server 2005?如何在 SQL Server 2005 中的变量中选择多个行值？
【发布时间】：2019-04-30 14:36:53
【问题描述】：

我写了一个定义变量的脚本。

其中一个变量是这样定义的。

SET @Yesterday = (SELECT SUM (ri.Sales_Quantity) 
                  FROM ReportTransactions AS RT     
                  LEFT JOIN ReportItems AS ri ON rt.Report_Transaction_ID = ri.Report_Transaction_ID    
                  LEFT JOIN MMGroups AS mmg ON ri.MMGroup2_ID = MMG.ID  
                  LEFT JOIN Locations AS L ON L.Location_ID = RT.Store_ID 
                  WHERE rt.Transaction_Date > DATEADD(DAY, -1, GETDATE()) 
                    AND (NULL IS NULL OR rt.Store_ID = NULL)   
                    AND (rt.Training_Mode IS NULL OR rt.Training_Mode = 0)   
                    AND ri.Item_Voided = 0   
                    AND ri.Void_Type = 0
                    AND rt.Transaction_Type IN (0, 1, 2, 3, 4, 5, 6)   
                    AND ri.Item_Type IN (1, 2, 3, 4, 6, 7, 54, 55, 81, 84, 11, 12, 13, 22, 25, 27, 82) 
                  GROUP BY  
                      RT.Store_ID);

然而，这会返回多个值，因此无法定义变量并引发错误。

我的报告需要所有这些行，那么如何定义它们并检索它们，最佳做法是什么？

【问题讨论】：

将它们存储在临时表中，而不是变量中。

标签： sql sql-server-2005

【解决方案1】：

您需要表格 temporary 或 base 因为您需要多行或多列：

SELECT RT.Store_ID, SUM (ri.Sales_Quantity) AS Sum_Qty INTO #table
FROM ReportTransactions AS RT LEFT JOIN 
     ReportItems AS ri 
     ON rt.Report_Transaction_ID = ri.Report_Transaction_ID LEFT JOIN 
     MMGroups AS mmg 
     ON ri.MMGroup2_ID = MMG.ID LEFT JOIN 
     Locations AS L 
     ON L.Location_ID = RT.Store_ID 
WHERE . . .     
GROUP BY RT.Store_ID;

然而，基表需要INSERT INTO 语句和SELECT 语句：

INSERT INTO table (col1, col2)
     SELECT . . . ;

【讨论】：

【解决方案2】：

也许你可以试试表变量。

DECLARE @Yesterday  Table
(
Sum_Qty  int
)
insert into @Yesterday
SELECT SUM (ri.Sales_Quantity) 
                  FROM ReportTransactions AS RT     
                  LEFT JOIN ReportItems AS ri ON rt.Report_Transaction_ID = ri.Report_Transaction_ID    
                  LEFT JOIN MMGroups AS mmg ON ri.MMGroup2_ID = MMG.ID  
                  LEFT JOIN Locations AS L ON L.Location_ID = RT.Store_ID 
                  WHERE rt.Transaction_Date > DATEADD(DAY, -1, GETDATE()) 
                    AND (NULL IS NULL OR rt.Store_ID = NULL)   
                    AND (rt.Training_Mode IS NULL OR rt.Training_Mode = 0)   
                    AND ri.Item_Voided = 0   
                    AND ri.Void_Type = 0
                    AND rt.Transaction_Type IN (0, 1, 2, 3, 4, 5, 6)   
                    AND ri.Item_Type IN (1, 2, 3, 4, 6, 7, 54, 55, 81, 84, 11, 12, 13, 22, 25, 27, 82) 
                  GROUP BY  
                      RT.Store_ID

【讨论】：