2

我是一名新的 Web 开发人员，我正在开发一个简单的培训管理系统，为用户提供测验。该系统使用 ASP.NET 和 MS SQLServer 开发。现在，我需要提出一个查询，显示每个部门在所有提供的安全测验中的参与百分比（这意味着 IsSent = true 的测验）。

我有以下数据库设计：

Employee Table: Username,  Name,    DivisionCode
Divisions Table: SapCode, Divison
Quiz Table: QuizID, Title, Description, IsSent
UserQuiz: UserQuizID, QuizID, DateTimeComplete, Username

（DivisionCode 是 SapCode 的外键。IsSent 是一个标志，用于指示已提供并发送给用户参与的测验。）

例如，如果我有四个部门；A、B、C 和 D，我已经向这些部门的员工发送了大约 23 个测验。由于每个部门在每个测验中都有特定的参与百分比。我的查询应该显示参与每个部门所有提供的测验的总体百分比。因此，不应针对每个测验对结果进行分解。

那么该怎么做呢？

这是数据库的架构及其数据：

USE [Test]
GO
/****** Object:  Table [dbo].[Divisions]    Script Date: 08/03/2012 19:36:09 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
SET ANSI_PADDING ON
GO
CREATE TABLE [dbo].[Divisions](
 [SapCode] [nvarchar](50) NOT NULL,
 [Division] [varchar](50) NOT NULL,
 CONSTRAINT [PK_Divisions] PRIMARY KEY CLUSTERED 
(
 [SapCode] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET ANSI_PADDING OFF
GO
INSERT [dbo].[Divisions] ([SapCode], [Division]) VALUES (N'1', N'A')
INSERT [dbo].[Divisions] ([SapCode], [Division]) VALUES (N'2', N'B')
INSERT [dbo].[Divisions] ([SapCode], [Division]) VALUES (N'3', N'C')
/****** Object:  Table [dbo].[Quiz]    Script Date: 08/03/2012 19:36:09 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
SET ANSI_PADDING ON
GO
CREATE TABLE [dbo].[Quiz](
 [QuizID] [int] IDENTITY(1,1) NOT NULL,
 [Title] [varchar](50) NOT NULL,
 [Description] [varchar](50) NOT NULL,
 [IsSent] [bit] NOT NULL,
 CONSTRAINT [PK_Quiz] PRIMARY KEY CLUSTERED 
(
 [QuizID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET ANSI_PADDING OFF
GO
SET IDENTITY_INSERT [dbo].[Quiz] ON
INSERT [dbo].[Quiz] ([QuizID], [Title], [Description], [IsSent]) VALUES (1, N'Quiz I', N'Test', 1)
INSERT [dbo].[Quiz] ([QuizID], [Title], [Description], [IsSent]) VALUES (2, N'Quiz II', N'Test test', 1)
INSERT [dbo].[Quiz] ([QuizID], [Title], [Description], [IsSent]) VALUES (3, N'Quiz III', N'TEST TEST', 0)
SET IDENTITY_INSERT [dbo].[Quiz] OFF
/****** Object:  Table [dbo].[Employee]    Script Date: 08/03/2012 19:36:09 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
SET ANSI_PADDING ON
GO
CREATE TABLE [dbo].[Employee](
 [Usename] [nvarchar](50) NOT NULL,
 [Name] [varchar](50) NOT NULL,
 [DivisionCode] [nvarchar](50) NOT NULL,
 CONSTRAINT [PK_Employee] PRIMARY KEY CLUSTERED 
(
 [Usename] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET ANSI_PADDING OFF
GO
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'John12', N'John A', N'1')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'John13', N'John B', N'1')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'John15', N'John C', N'3')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'Maria12', N'Maria A', N'3')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'Rony14', N'Rony A', N'2')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'Tid52', N'Tid A', N'3')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'Tim12', N'Tim A', N'1')
INSERT [dbo].[Employee] ([Usename], [Name], [DivisionCode]) VALUES (N'Tim15', N'Tim B', N'2')
/****** Object:  Table [dbo].[UserQuiz]    Script Date: 08/03/2012 19:36:09 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[UserQuiz](
 [UserQuizID] [int] IDENTITY(1,1) NOT NULL,
 [QuizID] [int] NOT NULL,
 [DateTimeComplete] [datetime] NOT NULL,
 [Score] [float] NOT NULL,
 [Username] [nvarchar](50) NOT NULL,
 CONSTRAINT [PK_UserQuiz] PRIMARY KEY CLUSTERED 
(
 [UserQuizID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET IDENTITY_INSERT [dbo].[UserQuiz] ON
INSERT [dbo].[UserQuiz] ([UserQuizID], [QuizID], [DateTimeComplete], [Score], [Username]) VALUES (1, 1, CAST(0x0000A07900000000 AS DateTime), 100, N'John12')
INSERT [dbo].[UserQuiz] ([UserQuizID], [QuizID], [DateTimeComplete], [Score], [Username]) VALUES (2, 1, CAST(0x0000A07900000000 AS DateTime), 50, N'Tim12')
INSERT [dbo].[UserQuiz] ([UserQuizID], [QuizID], [DateTimeComplete], [Score], [Username]) VALUES (3, 1, CAST(0x0000A07B00000000 AS DateTime), 100, N'Rony14')
INSERT [dbo].[UserQuiz] ([UserQuizID], [QuizID], [DateTimeComplete], [Score], [Username]) VALUES (4, 1, CAST(0x0000A07900000000 AS DateTime), 0, N'Tim15')
INSERT [dbo].[UserQuiz] ([UserQuizID], [QuizID], [DateTimeComplete], [Score], [Username]) VALUES (5, 1, CAST(0x0000A07900000000 AS DateTime), 100, N'Tid52')
SET IDENTITY_INSERT [dbo].[UserQuiz] OFF
/****** Object:  ForeignKey [FK_Employee_Divisions]    Script Date: 08/03/2012 19:36:09 ******/
ALTER TABLE [dbo].[Employee]  WITH CHECK ADD  CONSTRAINT [FK_Employee_Divisions] FOREIGN KEY([DivisionCode])
REFERENCES [dbo].[Divisions] ([SapCode])
GO
ALTER TABLE [dbo].[Employee] CHECK CONSTRAINT [FK_Employee_Divisions]
GO
/****** Object:  ForeignKey [FK_UserQuiz_Employee]    Script Date: 08/03/2012 19:36:09 ******/
ALTER TABLE [dbo].[UserQuiz]  WITH CHECK ADD  CONSTRAINT [FK_UserQuiz_Employee] FOREIGN KEY([Username])
REFERENCES [dbo].[Employee] ([Usename])
GO
ALTER TABLE [dbo].[UserQuiz] CHECK CONSTRAINT [FK_UserQuiz_Employee]
GO
/****** Object:  ForeignKey [FK_UserQuiz_Quiz]    Script Date: 08/03/2012 19:36:09 ******/
ALTER TABLE [dbo].[UserQuiz]  WITH CHECK ADD  CONSTRAINT [FK_UserQuiz_Quiz] FOREIGN KEY([QuizID])
REFERENCES [dbo].[Quiz] ([QuizID])
GO
ALTER TABLE [dbo].[UserQuiz] CHECK CONSTRAINT [FK_UserQuiz_Quiz]
GO

更新： 结果应该是这样的：

部门* ** * ** * ** * ***总体百分比

A * ** * ** * ** * ** * ** * 80

B * ** * ** * ** * ** * ** * 60

C * ** * ** * ** * ** * ** * 50

总百分比是每个测验中每个部门的所有完成百分比的总和。例如，Quiz#1 = 40%、Quiz#2 = 60% 和 Quiz#3 = 90% 中的部分 A 的完成百分比，此查询应显示的总百分比是所有其中除以发送的测验总数（即 3）。我希望我很清楚，这个概念很明显。

更新#2：

我仍在为此苦苦挣扎，但我想出了一个查询，该查询显示了每个部门每个测验中的员工总数和参与者总数。这是查询：

SELECT     COUNT(dbo.Employee.Usename) AS [Total Number of Employees], COUNT(dbo.UserQuiz.Username) AS [Total Number of Participants], dbo.Divisions.Division, 
                      dbo.Quiz.Title
FROM         dbo.Divisions INNER JOIN
                      dbo.Employee ON dbo.Divisions.SapCode = dbo.Employee.DivisionCode INNER JOIN
                      dbo.UserQuiz ON dbo.Employee.Usename = dbo.UserQuiz.Username INNER JOIN
                      dbo.Quiz ON dbo.UserQuiz.QuizID = dbo.Quiz.QuizID
WHERE     (dbo.Quiz.IsSent = 1)
GROUP BY dbo.Divisions.Division, dbo.Quiz.Title

我应该怎么做才能改进它并得到我需要的东西？

4

1 回答 1

0

The following query may do what you want. I calculates the number of employees in each division, multiplies this by the number of quizzes. It then divides the number actually taken by this total:

select de.Division, de.numemployees, de.numemployees*q.numquizzes as qe_combos,
       numquizzestaken/1.0*de.numemployees*q.numquizzes as proportion
from (select d.Division, count(*) as numemployees
      from Employee e join
           Division d
           on e.DivisionCode = d.SapCode
      group by d.Division
     ) de left outer join
     (select d.Division, count(*) as numquizzestaken
      from Employee e join
           Division d
           on e.DivisionCode = d.SapCode join
           UserQuiz uq
           on e.UserName = uq.UserName join
           Quiz q
           on uq.QuidId = q.QuizId and
              q.issent = true
      group by de.Division
     ) dq  
     on de.division = dq.division cross join
     (select count(*) as numquizes
      from Quiz q
      where isSent = true
     ) q
于 2012-08-14T13:48:36.277 回答