0

我在一个小型 .NET 4.5、MVC4、EF5 项目上使用存储库模式。

我注意到，如果我更改从存储库传递的 IQueryable 对象的结果，则生成的 SQL 不太理想。

例如，在我的存储库中：

    public IQueryable<Entry> GetEntries()
    {
        // (db is an instance of the data context)
        return db.Entries.Where(e => e.UserId == WebSecurity.CurrentUserId);
    }

在我的控制器中，我限制返回的行数并对其进行排序：

    public ActionResult Index()
    {
        // (repo is an instance of the repository object)
        var entries = repo.GetEntries().Take(10).OrderByDescending(o => o.Created)
        return View(entries);
    }

这会生成以下 SQL：

DECLARE @p__linq__0 int = 1

SELECT 
[Project1].[UserId] AS [UserId], 
[Project1].[Id] AS [Id], 
[Project1].[Created] AS [Created], 
[Project1].[LastModified] AS [LastModified], 
[Project1].[BodyOriginal] AS [BodyOriginal], 
[Project1].[BodyFormatted] AS [BodyFormatted], 
[Project1].[FormatterVersion] AS [FormatterVersion], 
[Project1].[BodyDigest] AS [BodyDigest], 
[Project1].[FollowupId] AS [FollowupId], 
[Project1].[AddMethod] AS [AddMethod], 
[Project1].[Entry_Id] AS [Entry_Id]
FROM ( SELECT 
        [Extent1].[Id] AS [Id], 
        [Extent1].[UserId] AS [UserId], 
        [Extent1].[Created] AS [Created], 
        [Extent1].[LastModified] AS [LastModified], 
        [Extent1].[BodyOriginal] AS [BodyOriginal], 
        [Extent1].[BodyFormatted] AS [BodyFormatted], 
        [Extent1].[FormatterVersion] AS [FormatterVersion], 
        [Extent1].[BodyDigest] AS [BodyDigest], 
        [Extent1].[FollowupId] AS [FollowupId], 
        [Extent1].[AddMethod] AS [AddMethod], 
        [Extent1].[Entry_Id] AS [Entry_Id]
        FROM [dbo].[Entries] AS [Extent1]
        WHERE [Extent1].[UserId] = @p__linq__0
)  AS [Project1]
ORDER BY [Project1].[Created] DESC

如您所见，生成的 SQL 是高度冗余的。

如果我更改存储库方法以包含截断和排序的结果：

    public IQueryable<Entry> GetEntries()
    {
        // (db is an instance of the data context)
        return db.Entries.Where(e => e.UserId == WebSecurity.CurrentUserId).Take(10).OrderByDescending(o => o.Created);
    }

生成的 SQL 更好：

DECLARE @p__linq__0 int = 1

SELECT 
            [Extent1].[Id] AS [Id], 
            [Extent1].[UserId] AS [UserId], 
            [Extent1].[Created] AS [Created], 
            [Extent1].[LastModified] AS [LastModified], 
            [Extent1].[BodyOriginal] AS [BodyOriginal], 
            [Extent1].[BodyFormatted] AS [BodyFormatted], 
            [Extent1].[FormatterVersion] AS [FormatterVersion], 
            [Extent1].[BodyDigest] AS [BodyDigest], 
            [Extent1].[FollowupId] AS [FollowupId], 
            [Extent1].[AddMethod] AS [AddMethod], 
            [Extent1].[Entry_Id] AS [Entry_Id]
            FROM [dbo].[Entries] AS [Extent1]
            WHERE [Extent1].[UserId] = @p__linq__0
  ORDER BY [Extent1].[Created] DESC

我该如何克服这个问题，即利用存储库模式，同时仍然可以灵活地修改存储库的结果，而不会创建结构不佳的 SQL？

4

1 回答 1

1

这是存储库反模式，最好删除该“存储库”。如果你真的想使用 Repository，定义一个与 EF 无关的接口，并且没有方法返回 IQueryable。此外，是否使用存储库与 EF 如何生成 sql 无关。

于 2013-10-18T21:33:18.583 回答