从数据上下文获取存储过程:linq 到 sql-尊龙凯时

问题描述

我在 sql server 中有一个名为 parsexml 的存储过程.我有一个使用 linq to sql 的存储库模式.我需要从存储库层中调用存储过程.与 gettable 方法不同，我们没有用于数据上下文的 getstoredprocedure 方法.这种情况下如何调用存储过程?

i have a stored procedure named parsexml in sql server. i have a repository pattern using linq to sql. i need to call the stored procedure from within the repository layer. unlike gettable method, we don’t have a getstoredprocedure method for data context. how can we call the stored procedure in such a scenario?

数据库代码

[global::system.data.linq.mapping.functionattribute(name="dbo.parsexml")]
    public isingleresult parsexml([global::system.data.linq.mapping.parameterattribute(name="inputxml", dbtype="xml")] system.xml.linq.xelement inputxml)
    {
        iexecuteresult result = this.executemethodcall(this, ((methodinfo)(methodinfo.getcurrentmethod())), inputxml);
        return ((isingleresult)(result.returnvalue));
    }

存储层

namespace repositorylayer
{
public interface ilijosbankrepository
{
    system.data.linq.datacontext context { get; set; }
    list getallaccountsforuser(int userid);
    void updatebankaccountusingstoredprocedure();
}
public class lijossimplebankrepository : ilijosbankrepository
{
    public system.data.linq.datacontext context
    {
        get;
        set;
    }
    public list getallaccountsforuser(int userid)
    {
        iqueryable queryresultentities = context.gettable().where(p => p.accountownerid == userid);
        return queryresultentities.tolist();
    }
    public virtual void updatebankaccountusingstoredprocedure()
    {
        //context.getstroedprocedures();
    }
}
}

参考:

1. 多个 unitofworks、isession 和存储库

推荐答案

让存储库的任何调用者知道特定方法调用是否导致从文件、sql 读取文本是 soc 的一个巨大突破语句、sprocs 甚至只是花园侏儒在文本终端上输入结果.

it's a pretty huge break of soc to have any callers of your repository be aware of whether or not a particular method call results in reading text from a file, sql statement, sprocs or even just garden gnomes typing results out on a text terminal.

为此，将 context 属性设为公开无济于事.使用存储库的全部意义在于让消费者免受持久性问题的困扰！

to that end, it doesn't help matters to have your context property be public. the whole point of using a repository is so that consumers are shielded from persistence concerns!

由于您似乎非常需要避免使用自定义类型的 context，您可以省去很多麻烦，只需发出一个直接的、老式的 sql 语句即可执行你的程序.

since you seem to have a strong need to avoid using a custom-typed context, you'd save yourself much trouble and just issue a straight-up, old-school sql statement that will execute your sproc.

考虑重构您的界面和逻辑，使其看起来更像这样:

consider refactoring your interface and logic to look more like this:

public interface ilijosbankrepository
{
    list getallaccountsforuser(int userid);
    void updatebankaccount(/* params go here */);
    /* ...other query methods, etc... */
}
public class lijosbankrepository : ilijosbankrepository
{
     private readonly datacontext context { get; set;}
     public lijosbankrepository(datacontext ctx) { ... }
     public void updatebankaccount(string inputxml)
     {
          context.executecommand("parsexml", inputxml);
     }
}