需要帮助从 sql 内连接获取 SUM 数据

Posted 2023-05-08

【中文标题】需要帮助从 sql 内连接获取 SUM 数据

【英文标题】：Need help getting SUM data from sql inner joins

【发布时间】：2020-10-26 17:08:32

【问题描述】：

我有 3 个表，其中包含从 dbo.Associates 到（通过 associateId）dbo.Relocations 到（通过 relocationId）dbo.Expenses 的内部连接连接。以下是表格：

-- Microsoft SQL Server 2017

CREATE TABLE Associates(
    AssociateId varchar(150) not null PRIMARY KEY,
    FirstName varchar(150) not null,
    LastName varchar(150) not null
)

INSERT INTO Associates(AssociateId, FirstName, LastName)
VALUES
('1', 'Tom', 'Tom'),
('2', 'Sue', 'Sue')

CREATE TABLE Relocations(
    RelocationId int not null PRIMARY KEY,
    AssociateId varchar(150) FOREIGN KEY REFERENCES Associates(AssociateId),
    StartingLocation varchar(150),
    EndingLocation varchar(150),
    PlanningCenter varchar(150),
    CostCenter varchar(150),
    OpenDate datetime,
    CloseDate datetime,
    LimitedAmount decimal(18,2),
    RelocationTypeId varchar(150)
)

INSERT INTO Relocations(RelocationId, AssociateId, StartingLocation, 
EndingLocation, PlanningCenter, CostCenter,
OpenDate, CloseDate, RelocationTypeId)
VALUES
(1, '2', 'Jacksonville', 'Macon', 'Buffalo', 'A', '2020-10-20', '2020-10-22', 
'Full Relocation'),
(2, '2', 'Los Angelos', 'New York', 'Detroit', 'B', '2020-09-20', null, 'Full 
Relocation'),
(3, '1', 'Washington DC', 'Houston', 'Dalls', 'C', '2020-08-08', '2020-08- 
15', 'Full Relocation'),
(4, '1', 'Lakeland', 'Atlanta', 'Seattle', 'D', '2020-09-12', '2020-09-28', 
'Full Relocation'),
(5, '1', 'San Diego', 'Woodbury', 'Baltimore', 'E', '2020-10-02', '2020-10- 
17', 'Full Relocation')

CREATE TABLE Expenses(
    ExpenseId int not null PRIMARY KEY,
    RelocationId int FOREIGN KEY REFERENCES Relocations(RelocationId),
    ExepnseTypeId varchar(150),
    DateIncurred datetime,
    GLAccountId varchar(150),
    TaxableAmount decimal(18,2),
    VoucherNumber int,
    Amount decimal(18,2),
    ToBeRepaid decimal(18,2),
    ReportableAmount decimal(18,2),
    LastUpdated datetime
)

INSERT INTO Expenses(ExpenseId, RelocationId, ExepnseTypeId, DateIncurred, 
GLAccountId, TaxableAmount, VoucherNumber,
Amount, ToBeRepaid, ReportableAmount, LastUpdated)
VALUES
(1, 1, 'Item 1', '2020-01-15', 1, 30.30, 23, 500.00, 400.00, 600.00, '2020- 
01-15'),
(2, 1, 'Item 1', '2020-01-20', 2, null, 23, 300.00, 0.00, 300.00, '2020-01- 
20'),
(3, 2, 'Item 2', '2020-02-23', 1, null, 23, 0.00, 0.00, 45.45, '2020-02-23'),
(4, 2, 'Item 3', '2020-03-11', 1, 400.00, 33, 75.00, 80.00, 0.00, '2020-03- 
11'),
(5, 3, 'Item 4', '2020-04-12', 2, 123.23, 33, 2000.50, 76.76, 540.54, '2020- 
04-12'),
(6, 3, 'Item 4', '2020-04-23', 2, null, 33, 1500.00, 0.00, 300.21, '2020-04- 
23'),
(7, 4, 'Item 5', '2020-05-03', 1, 40.00, 43, 43.00, 0.00, 43.56, '2020-05- 
03'),
(8, 4, 'Item 6', '2020-10-15', 2, null, 43, 0.00, 50.55, 76.87, '2020-10- 
15'),
(9, 5, 'Item 7', '2020-10-13', 1, 55.00, 43, 0.00, 0.00, 0.00, '2020-10-13'),
(10, 5, 'Item 8', '2020-10-05', 2, null, 53, 400.00, 0.00, 0.00, '2020-10- 
05')
INSERT INTO Expenses(ExpenseId, RelocationId, ExepnseTypeId, DateIncurred, 
GLAccountId, TaxableAmount, VoucherNumber,
Amount, ToBeRepaid, ReportableAmount, LastUpdated)
VALUES
(11, 5, 'Item 1', '2020-05-17', 2, 25, 53, 150.00, 300.00, 700.00, '2020-05- 
17')

最终我试图生成一个匹配的输出：

 OpenDate    CloseDate    Account   LastName   FirstName   Amount   Item1   Item2   Item3   Item4    Item5   Item6   Item7   Item8  
------------ ------------ --------- ---------- ----------- -------- ------- ------- ------- -------- ------- ------- ------- ------- 
10/20/2020   10/22/2020         1   Sue        Sue            500   500     null    null    null     null    null    null    null   
10/20/2020   10/22/2020         2   Sue        Sue            300   300     null    null    null     null    null    null    null  
9/20/2020    NULL               1   Sue        Sue             75   null    null    75      null     null    null    null    null   
8/8/2020     8/15/2020          2   Tom        Tom         3500.5   null    null    null    3500.5   null    null    null    null   
9/12/2020    9/28/2020          1   Tom        Tom             43   null    null    null    null     43      null    null    null   
9/12/2020    9/28/2020          2   Tom        Tom              0   null    null    null    null     null    null    null    null   
10/2/2020    10/17/2020         1   Tom        Tom              0   null    null    null    null     null    null    null    null   
10/2/2020    10/17/2020         2   Tom        Tom            550   150     null    null    null     null    null    null    400

在下面的查询中，我可以获得每个员工搬迁的总费用总额，也可以通过 GLAccount（费用表中的字段）获得，如下所示：

select 
    OpenDate,
    CloseDate,
    GLAccountId as Account,
    LastName,
    FirstName,
    SUM(t3.Amount) as Amount
from dbo.Associates t1
inner join dbo.Relocations t2
on t1.AssociateId = t2.AssociateId
inner join dbo.Expenses t3
on t2.RelocationId = t3.RelocationId
group by 
OpenDate,
CloseDate,
GLAccountId,
LastName,
FirstName

上述查询为我提供了所需输出中的前 6 个字段。但我不知道如何获取剩余的 Item 字段

【问题讨论】：

您正在寻找数据透视查询。但是由于行中的元素没有固定的名称（假设您有超过 8 个项目），您将需要动态生成它。这是你的问题有点回答：***.com/a/10404455/460557

【参考方案1】：

对于固定的项目列表，可以使用条件聚合：

select 
    r.OpenDate,
    r.CloseDate,
    e.GLAccountId as Account,
    a.LastName,
    a.FirstName,
    sum(e.Amount) as Amount,
    sum(case when e.ExpenseTypeId = 'Item 1' then e.amount end) item1,
    sum(case when e.ExpenseTypeId = 'Item 2' then e.amount end) item2,
    sum(case when e.ExpenseTypeId = 'Item 3' then e.amount end) item3,
    ...
    sum(case when e.ExpenseTypeId = 'Item 8' then e.amount end) item8
from dbo.Associates a
inner join dbo.Relocations r on r.AssociateId = a.AssociateId
inner join dbo.Expenses    e on e.RelocationId = r.RelocationId
group by r.OpenDate, r.CloseDate, e.GLAccountId, a.LastName, a.FirstName

请注意，我使用了有意义的表别名，并使用它们所属的表来限定所有列名。这大大提高了查询的可读性，从而提高了它的可维护性。

【讨论】：

一点小事，ExpenseTypeId 是 varchar，内容不只匹配数字 :)