实体框架HierarchyId变通方法

Question

EF 5.0

我正在研究一个原型来一起测试hierarchyid和实体框架.我有以下架构:

Create Table dbo.Employee
(
   EmployeeId int identity not null,
   Name nvarchar(100) not null,
   Node hierarchyid not null,
   NodePath as Node.ToString() persisted,
   Level AS Node.GetLevel() persisted,
   ManagerNode as Node.GetAncestor(1) persisted,
   ManagerNodePath as Node.GetAncestor(1).ToString() persisted
);

Alter Table dbo.Employee
    Add Constraint EmployeePK Primary Key NonClustered (EmployeeId);

Go

--Enforce Hierarchy
Alter Table dbo.Employee
    Add Constraint EmployeeManagerNodeNodeFK Foreign Key (ManagerNode) References Employee(Node);
Go

Create Unique Clustered Index EmployeeDepthFirstIndex on dbo.Employee(Node);

Go

Create NonClustered Index EmployeeBreathFirstIndex on dbo.Employee(Level, Node);

Go

从我的阅读中,EF目前不支持hierarchyid数据类型,但是有些人提出了一些变通方法,例如创建我上面已经完成的计算列(Node.ToString()).

有没有办法设置EF,以便它识别父/子关系,这样我可以有效地拥有一个下属集合？例如

Employee.Subordinates

我唯一能想到的是创建一个带有FK的ManagerId列,但后来我有效地将层次结构存储在两个地方.

谢谢你的帮助!

Answer 1

EF6现在是开源的,因此很容易添加HierarcyID支持.我也加了它.您可以从codeplex下载修改后的源代码和编译/签名的dll:http: //entityframework.codeplex.com/SourceControl/network/forks/zgabi/efhierarchyidrc1(有时候fork名称更改)或者来自NuGet:https:// www .nu​​get.org/packages/EntityFrameworkWithHierarchyId / Currenty EF6处于RC1状态,但我会将修改合并到EF6的每个后续版本中.

我有以下型号:

public class Employee
{
    public int EmployeeId { get; set; }
    [Required, MaxLength(100)]
    public string Name { get; set; }
    [Required]
    public HierarchyId Node { get; set; }

    public IQueryable<Employee> GetSubordinates(MyContext context)
    {
        return context.Employees.Where(o => Node == o.Node.GetAncestor(1));
    }
}

public class MyContextInitializer : CreateDatabaseIfNotExists<MyContext>
{
    protected override void Seed(MyContext context)
    {
        context.Database.ExecuteSqlCommand(
            "ALTER TABLE [dbo].[Employees] ADD [ManagerNode] AS ([Node].[GetAncestor]((1))) PERSISTED");
        context.Database.ExecuteSqlCommand(
            "ALTER TABLE [dbo].[Employees] ADD CONSTRAINT [UK_EmployeeNode] UNIQUE NONCLUSTERED (Node)");
        context.Database.ExecuteSqlCommand(
            "ALTER TABLE [dbo].[Employees]  WITH CHECK ADD CONSTRAINT [EmployeeManagerNodeNodeFK] " +
            "FOREIGN KEY([ManagerNode]) REFERENCES [dbo].[Employees] ([Node])");
        context.Employees.Add(new Employee { Name = "Root", Node = new HierarchyId("/") });
        context.Employees.Add(new Employee { Name = "Emp1", Node = new HierarchyId("/1/") });
        context.Employees.Add(new Employee { Name = "Emp2", Node = new HierarchyId("/2/") });
        context.Employees.Add(new Employee { Name = "Emp3", Node = new HierarchyId("/1/1/") });
        context.Employees.Add(new Employee { Name = "Emp4", Node = new HierarchyId("/1/1/1/") });
        context.Employees.Add(new Employee { Name = "Emp5", Node = new HierarchyId("/2/1/") });
        context.Employees.Add(new Employee { Name = "Emp6", Node = new HierarchyId("/1/2/") });
    }
}

public class MyContext : DbContext
{
    public DbSet<Employee> Employees { get; set; }
}

生成的数据库:

CREATE TABLE [dbo].[Employees](
    [EmployeeId] [int] IDENTITY(1,1) NOT NULL,
    [Name] [nvarchar](100) NOT NULL,
    [Node] [hierarchyid] NOT NULL,
    [ManagerNode]  AS ([Node].[GetAncestor]((1))) PERSISTED,
 CONSTRAINT [PK_dbo.Employees] PRIMARY KEY CLUSTERED 
(
    [EmployeeId] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY],
 CONSTRAINT [UK_EmployeeNode] UNIQUE NONCLUSTERED 
(
    [Node] 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].[Employees]  WITH CHECK ADD  CONSTRAINT [EmployeeManagerNodeNodeFK] FOREIGN KEY([ManagerNode])
REFERENCES [dbo].[Employees] ([Node])

获取Emp1员工子节点的示例:

    using (var c = new MyContext())
    {
        var firstItem = c.Employees.Single(o => o.Node == new HierarchyId("/1/"));

        foreach (var table1 in firstItem.GetSubordinates(c))
        {
            Console.WriteLine(table1.EmployeeId + " " + table1.Name);
        }
    }

结果:

4 Emp3
7 Emp6

Answer 2

使用varbinary（892）代替architectureid。EF识别varbinary返回的字节数组。您可以将字节数组转换为SqlHierarchyid类型，并使用hyrarchy pod函数。通过这种解决方法，您甚至可以在其他数据库中使用architectureid函数。请参阅http://www.casavillar.com.br/blog，以获取更多详细信息以及指向nugget和github的链接，您将在其中找到包括MySql在内的示例