Question

我有以下5个表 -

   CREATE TABLE [dbo].[MSP_EpmProject](
    [ProjectUID] [uniqueidentifier] NOT NULL,
    [ProjectName] [nvarchar](255) NOT NULL,
    [ProjectAuthorName] [nvarchar](255) NULL,
 CONSTRAINT [PK_MSP_EpmProject] PRIMARY KEY CLUSTERED 
([ProjectUID] ASC)) 

CREATE TABLE [dbo].[Project_CI_Mapping](
    [ProjectName] [nvarchar](255) NOT NULL,
    [CI] [nvarchar](100) NOT NULL)

CREATE TABLE [dbo].[ca_owned_resource]( 
    [resource_name] [nvarchar](100) NOT NULL,
    [resource_description] [nvarchar](255) NULL,
    [resource_family] [int] NULL,
    [resource_class] [int] NOT NULL,
    [resource_status] [int] NULL,   
    [resource_tag] [nvarchar](64) NULL)

CREATE TABLE [dbo].[DimTeamProject](
    [ProjectNodeSK] [int] IDENTITY(1,1) NOT NULL,
    [ProjectNodeGUID] [uniqueidentifier] NOT NULL,
    [ProjectNodeName] [nvarchar](256) NULL,
PRIMARY KEY CLUSTERED 
([ProjectNodeSK] ASC))

CREATE TABLE [dbo].[DimIteration](
    [IterationSK] [int] IDENTITY(1,1) NOT NULL,
    [IterationName] [nvarchar](256) NULL,
    [IterationGUID] [nvarchar](256) NOT NULL,   
PRIMARY KEY CLUSTERED 
([IterationSK] ASC))

我有一个简单的查询，试图从所有表中获取列，但它返回给我重复的值。尝试INNER JOIN会返回重复值＆amp;尝试LEFT OUTER JOIN时，它为“DimIteration.IterationName”提供了NULL值。

查询是 -

select m.ProjectName,m.ProjectAuthorName "Project Manager", p.CI,c.resource_tag "Alt CI ID", i.IterationName 
from MSP_EpmProject m, Project_CI_Mapping p, ca_owned_resource c, DimTeamProject t, DimIteration i
where i.ProjectGUID = UPPER(CAST(t.ProjectNodeGUID AS NVARCHAR(256)))
and p.CI = c.resource_name
and m.ProjectName = p.ProjectName
order by m.ProjectName,m.ProjectAuthorName, p.CI,c.resource_tag, i.IterationName

可能的映射是 -

MSP_EpmProject.ProjectName =  Project_CI_Mapping.ProjectName 
Project_CI_Mapping.CI = ca_owned_resource.resource_name
ca_owned_resource.resource_tag = DimTeamProject.ProjectNodeName
DimIteration.ProjectGUID = UPPER(CAST(DimTeamProject.ProjectNodeGUID AS NVARCHAR(256)))

同样适用的解决方案是什么？

感谢。

Answer 1

如果不详细研究这个问题，摆脱重复的一种方法是在ORDER BY之前插入一个GROUP BY子句，如下所示：

select m.ProjectName,m.ProjectAuthorName "Project Manager", p.CI,c.resource_tag "Alt CI ID", i.IterationName 
from MSP_EpmProject m, Project_CI_Mapping p, ca_owned_resource c, DimTeamProject t, DimIteration i
where i.ProjectGUID = UPPER(CAST(t.ProjectNodeGUID AS NVARCHAR(256)))
and p.CI = c.resource_name
and m.ProjectName = p.ProjectName
GROUP BY m.ProjectName,m.ProjectAuthorName, p.CI,c.resource_tag, i.IterationName
order by m.ProjectName,m.ProjectAuthorName, p.CI,c.resource_tag, i.IterationName

另一种方法是在DISTINCT之后和您希望返回的第一列之前插入SELECT。

e.g。 SELECT DISTINCT m.ProjectName...

Answer 2

您的查询中有CROSS JOIN。如果你使用较新的ANSI-92语法重写它（无论如何我建议你做reasons explained here），你可以看到交叉连接的位置：

select  m.ProjectName,
        m.ProjectAuthorName "Project Manager", 
        p.CI,c.resource_tag "Alt CI ID", 
        i.IterationName 
from    MSP_EpmProject m
        INNER JOIN Project_CI_Mapping p
            ON m.ProjectName = p.ProjectName
        INNER JOIN ca_owned_resource c
            ON p.CI = c.resource_name
        CROSS JOIN DimTeamProject t
        INNER JOIN DimIteration i
            ON i.ProjectGUID = UPPER(CAST(t.ProjectNodeGUID AS NVARCHAR(256)))
order by m.ProjectName,m.ProjectAuthorName, p.CI,c.resource_tag, i.IterationName;

基本上没有任何事情可以事先将DimTeamProject与任何表格联系起来。基于你有这个

的事实

ca_owned_resource.resource_tag = DimTeamProject.ProjectNodeName

作为一种可能的关系，它根本不在您的查询中，我建议您的查询需要：

select m.ProjectName,m.ProjectAuthorName "Project Manager", p.CI,c.resource_tag "Alt CI ID", i.IterationName 
from MSP_EpmProject m, Project_CI_Mapping p, ca_owned_resource c, DimTeamProject t, DimIteration i
where i.ProjectGUID = UPPER(CAST(t.ProjectNodeGUID AS NVARCHAR(256)))
and p.CI = c.resource_name
and m.ProjectName = p.ProjectName
and c.resource_tag = t.ProjectNodeName -- NEW Clause
order by m.ProjectName,m.ProjectAuthorName, p.CI,c.resource_tag, i.IterationName

但是，正如我已经说过的，我建议使用ANSI 92显式连接，这样您的查询就会变成：

SELECT  m.ProjectName,
        m.ProjectAuthorName "Project Manager", 
        p.CI,c.resource_tag "Alt CI ID", 
        i.IterationName 
FROM    MSP_EpmProject m
        INNER JOIN Project_CI_Mapping p
            ON m.ProjectName = p.ProjectName
        INNER JOIN ca_owned_resource c
            ON p.CI = c.resource_name
        INNER JOIN DimTeamProject t
            ON t.ProjectNodeName = c.resource_tag
        INNER JOIN DimIteration i
            ON i.ProjectGUID = UPPER(CAST(t.ProjectNodeGUID AS NVARCHAR(256)))
ORDER BY m.ProjectName,m.ProjectAuthorName, p.CI,c.resource_tag, i.IterationName;

摆脱重复值 - Sql Server

2 个答案: