Question

我有一张表如：

Create Table SalesTable
( StuffID int identity not null,
  County geography not null,
  SaleAmount decimal(12,8) not null,
  SaleTime datetime not null )

它记录了每次销售的数量，时间以及销售所在县的地理位置。

我想运行这样的查询：

Select sum(SaleAmount), County from SalesTable group by County

但如果我尝试这样做，我会得到：

The type "geography" is not comparable. It cannot be used in the GROUP BY clause.

但我想知道每个县发生了多少销售。令人讨厌的是，如果我有缩写的县（SDC，LAC，SIC等），那么我可以将它们分组，因为它只是一个varchar。但是我出于其他原因使用了geography数据类型。

Answer 1

我建议采用略有不同的结构：

create table dbo.County (
  CountyID int identity not null
    constraint [PK_County] primary key clustered (CountyID),
  Name varchar(200) not null,
  Abbreviation varchar(10) not null,
  geo geography not null
);
Create Table SalesTable
( 
  StuffID int identity not null,
  CountyID int not null
     constraint FK_Sales_County foreign key (CountyID)
     references dbo.County (CountyID),
  SaleAmount decimal(12,8) not null,
  SaleTime datetime not null 
);

从那里，您的聚合看起来像：

Select c.Abbreviation, sum(SaleAmount) 
from SalesTable as s
join dbo.County as c
   on s.CountyID = c.CountyID
group by c.Abbreviation;

如果您确实需要聚合中的地理列，那么您将成为子查询或公共表格表达式：

with s as (
    Select c.CountyID, c.Abbreviation, 
        sum(s.SaleAmount) as [TotalSalesAmount]
    from SalesTable as s
    join dbo.County as c
        on s.CountyID = c.CountyID
    group by c.Abbreviation
)
select s.Abbreviation, s.geo, s.TotalSalesAmount
from s
join dbo.County as c
    on s.CountyID = s.CountyID;

Answer 2

有一个函数可以使用地理类型作为char

试试这个

Select sum(SaleAmount), County.STAsText() from SalesTable 
group by County.STAsText()

按地理（或几何）分组的聚合/计数行

2 个答案: