Question

I use EntityFramework 4 + generated POCOs with Lazy loading disabled.
Let s say there are SQL tables named Table1, Table2, Table3 and Table4 and assume they contain some data.
Let s assume the simplified POCO representation of these tables looks like this:

public class Table1
{
    public int ID;
    public DateTime TableDate;
    public int Table2ID;
    public Table2 Table2;
    public ICollection<Table3> Table3s;
}

public class Table2
{
    public int ID;
    public string SomeString;
    public int Table4ID;
    public Table4 Table4;
}

public class Table3
{
    public int ID;
    public int Table1ID;
    public Table1 Table1;
    public decimal SomeDecimal;
}

public decimal Table4
{
    public int ID;
    public string SomeName;
}

如果执行以下法典:

Database DB = new Database(); // object context
var result = DB.Table1
    .Where(x => x.TableDate >= DateTime.MinValue);

EF将发表以下文件:

exec sp_executesql N SELECT 
[Extent1].[ID] AS [ID], 
[Extent1].[TableDate] AS [TableDate], 
[Extent1].[Table2ID] As [Table2ID]
FROM [dbo].[Table1] AS [Extent1]
WHERE ([Extent1].[TableDate] >= @p__linq__0) ,N @p__linq__0 datetime2(7) ,@p__linq__0= 0001-01-01 00:00:00

and the query would return the expected data.
However, if the following code would be executed:

Database DB = new Database(); // object context
var result = DB.Table1
    .Include("Table2")
    .Include("Table2.Table4")
    .Include("Table3")
    .Where(x => x.TableDate >= DateTime.MinValue);

EF将发表以下文件:

exec sp_executesql N SELECT 
[Project1].[ID2] AS [ID], 
[Project1].[ID] AS [ID1], 
[Project1].[TableDate] AS [TableDate], 
[Project1].[ID1] AS [ID2], 
[Project1].[SomeString] AS [SomeString], 
[Project1].[Table4ID] AS [Table4ID], 
[Project1].[ID3] AS [ID3], 
[Project1].[SomeName] AS [SomeName], 
[Project1].[ID4] AS [ID4], 
[Project1].[SomeDecimal] AS [SomeDecimal], 
[Project1].[Table1ID] AS [Table1ID]
FROM ( SELECT 
[Extent1].[ID] AS [ID], 
[Extent1].[TableDate] AS [TableDate], 
[Extent2].[ID] AS [ID1], 
[Extent2].[SomeString] AS [SomeString], 
[Extent2].[Table4ID] AS [Table4ID], 
[Extent3].[ID] AS [ID2], 
[Extent4].[ID] AS [ID3], 
[Extent4].[SomeName] AS [SomeName], 
[Extent5].[ID] AS [ID4], 
[Extent5].[SomeDecimal] AS [SomeDecimal], 
[Extent5].[Table1ID] AS [Table1ID], 
CASE WHEN ([Extent5].[ID] IS NULL) THEN CAST(NULL AS int) ELSE 1 END AS [C1]
FROM     [dbo].[Table1] AS [Extent1]
INNER JOIN [dbo].[Table2] AS [Extent2] ON [Extent1].[Table2ID] = [Extent2].[ID]
LEFT OUTER JOIN [dbo].[Table2] AS [Extent3] ON [Extent1].[Table2ID] = [Extent3].[ID]
LEFT OUTER JOIN [dbo].[Table4] AS [Extent4] ON [Extent3].[Table4ID] = [Extent4].[ID]
LEFT OUTER JOIN [dbo].[Table3] AS [Extent5] ON [Extent1].[ID] = [Extent5].[Table1ID]
WHERE ([Extent1].[TableDate] >= @p__linq__0)
)  AS [Project1]
ORDER BY [Project1].[ID2] ASC, [Project1].[ID] ASC, [Project1].[ID1] ASC, [Project1].[ID3] ASC, [Project1].[C1] ASC ,N @p__linq__0 datetime2(7) ,@p__linq__0= 0001-01-01 00:00:00

and the query would return nothing.

为什么会发生这种情况?

下面是关于编制上述表格的声明:

CREATE TABLE [dbo].[Table1](
[ID] [int] IDENTITY(1,1) NOT NULL,
[Table2ID] [int] NOT NULL,
[TableDate] [date] NOT NULL,
 CONSTRAINT [PK_Table1] PRIMARY KEY CLUSTERED 
(
[ID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]

ALTER TABLE [dbo].[Table1]  WITH NOCHECK ADD  CONSTRAINT [FK_Table1_Table2] FOREIGN KEY([Table2ID])
REFERENCES [dbo].[Table2] ([ID])

ALTER TABLE [dbo].[Table1] CHECK CONSTRAINT [FK_Table1_Table2]

CREATE TABLE [dbo].[Table2](
[ID] [int] NOT NULL,
[SomeString] [nvarchar](50) NOT NULL,
[Table4ID] [int] NULL,
 CONSTRAINT [PK_Table2] PRIMARY KEY CLUSTERED 
(
[ID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]

ALTER TABLE [dbo].[Table2]  WITH NOCHECK ADD  CONSTRAINT [FK_Table2_Table4] FOREIGN KEY([Table4ID])
REFERENCES [dbo].[Table4] ([ID])
ON UPDATE CASCADE

ALTER TABLE [dbo].[Table2] CHECK CONSTRAINT [FK_Table2_Table4]

CREATE TABLE [dbo].[Table3](
[ID] [int] IDENTITY(1,1) NOT NULL,
[SomeDecimal] [decimal](18, 4) NOT NULL,
[Table1ID] [int] NOT NULL,
 CONSTRAINT [PK_Table3] PRIMARY KEY CLUSTERED 
(
[ID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]

ALTER TABLE [dbo].[Table3]  WITH NOCHECK ADD  CONSTRAINT [FK_Table3_Table1] FOREIGN KEY([Table1ID])
REFERENCES [dbo].[Table1] ([ID])
ON DELETE CASCADE

ALTER TABLE [dbo].[Table3] CHECK CONSTRAINT [FK_Table3_Table1]

CREATE TABLE [dbo].[Table4](
[ID] [int] NOT NULL,
[SomeName] [nvarchar](50) NOT NULL,
 CONSTRAINT [PK_Table4] PRIMARY KEY CLUSTERED 
(
[ID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]

EDIT 2

这种查询也不会归还任何记录,可以作为最低实例:

Database DB = new Database();
var result = DB.Table1
    .Include("Table2")
    .Where(x => x.TableDate >= DateTime.MinValue);

热点:

exec sp_executesql N SELECT 
[Extent1].[ID] AS [ID], 
[Extent1].[Table2ID] AS [Table2ID], 
[Extent1].[TableDate] AS [TableDate], 
[Extent2].[ID] AS [ID1], 
[Extent2].[SomeString] AS [SomeString], 
[Extent2].[Table4ID] AS [Table4ID], 
FROM  [dbo].[Table1] AS [Extent1]
INNER JOIN [dbo].[Table2] AS [Extent2] ON [Extent1].[Table2ID] = [Extent2].[ID]
WHERE ([Extent1].[TableDate] >= @p__linq__0) ,N @p__linq__0 datetime2(7) ,@p__linq__0= 0001-01-01 00:00:00

此外,这里摘录于:

<EntityContainer>
      <AssociationSet Name="FK_Table1_Table2" Association="MyModel.Store.FK_Table1_Table2">
        <End Role="Table2" EntitySet="Table2" />
        <End Role="Table1" EntitySet="Table1" />
      </AssociationSet>
</EntityContainer>

<!-- ... -->

<EntityType Name="Table2">
  <Key>
    <PropertyRef Name="ID" />
  </Key>
  <Property Name="ID" Type="int" Nullable="false" />
  <Property Name="SomeString" Type="nvarchar" Nullable="false" MaxLength="50" />
  <Property Name="Table4ID" Type="int" />
</EntityType>

<!-- ... -->

<EntityType Name="Table1">
  <Key>
    <PropertyRef Name="ID" />
  </Key>
  <Property Name="ID" Type="int" Nullable="false" StoreGeneratedPattern="Identity" />
  <Property Name="TableDate" Type="date" Nullable="false" />
  <Property Name="Table2ID" Type="int" Nullable="false" />
</EntityType>

<!-- ... -->

<Association Name="FK_Table1_Table2">
  <End Role="Table2" Type="MyModel.Store.Table2" Multiplicity="1" />
  <End Role="Table1" Type="MyModel.Store.Table1" Multiplicity="*" />
  <ReferentialConstraint>
    <Principal Role="Table2">
      <PropertyRef Name="ID" />
    </Principal>
    <Dependent Role="Table1">
      <PropertyRef Name="Table2ID" />
    </Dependent>
   </ReferentialConstraint>
</Association>

Answer 1

It appears it was a problem with actual data in SQL server being inconsistent.
As stated there,

当两个表至少有一个对应点时,国家创新协会的关键词句返回。如果在“表1”中出现与“表2”不匹配的“表1”,这些行文将被列入名单。

除非“表2”与“表1”相匹配,否则,没有理由失败。然而,由于科索沃警察部队的制约因素得到了执行,但这个问题值得另一个问题,这种情况已经结束。

Answer 2

我的猜测是,“包括”使你无法按你所期望的顺序进行评估,因此,在你被点名时,无法提供表格。如果你进行武力评价,那么你会说什么?

友情链接