Question

我正在使用ms sql（2012版）中的“在多个列上进行旋转”。

我想从结构中获取值

Supplier Nr   SupplierName  Manager             Service           Type_of_Service     Planned_Turnover       Turnover          TransactionID            Month                 
 30              Honda       Name               CityService            varchar          100 000                95 000         009000011440114           201701
 50              Honda22     Name               CityService           varchar           200 000                195 000        0090001101144             201701
 30              Honda       Name               CityService            varchar          130 000                115 000        00900111114               201702
 50              Honda22     Name               CityService           varchar           230 000                205 000        001144                    201702

成所需的形式：

Supplier Nr   SupplierName  Manager             Service           Type_of_Service     Planned_Turnover 201701       Turnover 201701         TransactionID 201701  Planned_Turnover 201702       Turnover 201702         TransactionID 201702  
  30              Honda      Name               CityService            varchar              100 000                     95000               009000011440114         130 000                     115 000                     00900111114         
  50              Honda22    Name               CityService           varchar               200 000                     195000              0090001101144           230 000                     205 000                     001144

据我所知，如何基于一列来旋转表格 - 例如逐月获得营业额...但是如何获得更多价值（Planned_Turnover，Turnover，TransactionID彼此相邻？）

谢谢

Answer 1

这是一个使用临时表和动态SQL以及2个日期的数据透视表的示例。

create table #tempTable ([Supplier Nr] int, SupplierName varchar(30), Manager varchar(30), [Service] varchar(30), Type_of_Service varchar(30), Planned_Turnover varchar(30), Turnover varchar(30), TransactionID varchar(30), [Month] int);

insert into #tempTable ([Supplier Nr],SupplierName,Manager,[Service],Type_of_Service,Planned_Turnover,Turnover,TransactionID,[Month]) values
(30,'Honda','Name','CityService','varchar','100 000','95 000','009000011440114',201701),
(50,'Honda22','Name','CityService','varchar','200 000','195 000','0090001101144',201701),
(30,'Honda','Name','CityService','varchar','130 000','115 000','00900111114',201702),
(50,'Honda22','Name','CityService','varchar','230 000','205 000','001144',201702);

DECLARE @SQLString varchar(max);  

SET @SQLString = 'select * 
from (
  select [Supplier Nr], SupplierName, Manager, [Service], Type_of_Service, concat(''Planned_Turnover '',[Month]) as Title, Planned_Turnover as Value 
  from #tempTable where [Month] in (@Date1, @Date2)
  union all
  select [Supplier Nr], SupplierName, Manager, [Service], Type_of_Service, concat(''Turnover '',[Month]) as Title, Turnover as Value 
  from #tempTable where [Month] in (@Date1, @Date2)
  union all
  select [Supplier Nr], SupplierName, Manager, [Service], Type_of_Service, concat(''TransactionID '',[Month]) as Title, TransactionID as Value 
  from #tempTable where [Month] in (@Date1, @Date2)
) q
PIVOT (MAX(Value) FOR Title IN 
([Planned_Turnover @Date1],[Turnover @Date1],[TransactionID @Date1],[Planned_Turnover @Date2],[Turnover @Date2],[TransactionID @Date2])
) pvt';

declare @Date1 varchar(6) = '201701';
declare @Date2 varchar(6) = '201702';

SET @SQLString = replace(replace(@SQLString,'@Date1',@Date1),'@Date2',@Date2);

exec (@SQLString);

这仅限于2个月。如果它是几个月的范围，那么最好只计算枢轴字段的varchar，并在@SQLString中使用它。
示例代码：

declare @Date1 varchar(6) = '201712';
declare @Date2 varchar(6) = '201802';

declare @PivotFields varchar(max); 
with cte as (
   select convert(date,concat(@Date1,'01')) as dt
   union all
   select dateadd(month,1,dt) from cte where dt < convert(date,concat(@Date2,'01'))
)
select @PivotFields = concat(@PivotFields+', ',quotename(word+' '+FORMAT(dt,'yyyyMM')))
from cte cross join (values (1,'Planned_Turnover'),(2,'Turnover'),(3,'TransactionID')) words(wordid, word)
order by dt, wordid;

-- result: @PivotFields="[Planned_Turnover 201712], [Turnover 201712], [TransactionID 201712], [Planned_Turnover 201801], [Turnover 201801], [TransactionID 201801], [Planned_Turnover 201802], [Turnover 201802], [TransactionID 201802]"

从行到列获取值

1 个答案: