欢迎您访问程序员文章站本站旨在为大家提供分享程序员计算机编程知识!
您现在的位置是: 首页  >  IT编程

asp.net开发中常见公共捕获异常方式总结(附源码)

程序员文章站 2023-12-18 16:07:52
本文实例总结了asp.net开发中常见公共捕获异常方式。分享给大家供大家参考,具体如下: 前言:在实际开发过程中,对于一个应用系统来说,应该有自己的一套成熟的异常处理框架...

本文实例总结了asp.net开发中常见公共捕获异常方式。分享给大家供大家参考,具体如下:

前言:在实际开发过程中,对于一个应用系统来说,应该有自己的一套成熟的异常处理框架,这样当异常发生时,也能得到统一的处理风格,将异常信息优雅地反馈给开发人员和用户。我们都知道,.net的异常处理是按照“异常链”的方式从底层向高层逐层抛出,如果不能尽可能地早判断异常发生的边界并捕获异常,clr会自动帮我们处理,但是这样系统的开销是非常大的,所以异常处理的一个重要原则是“早发现早抛出早处理”。但是本文总结的服务端公共捕获异常处理可以宽泛地看做是在表现层的操作,要捕获特定层的特定异常,不在讨论范围内。

1、basepage类处理方式

在页面的公共基类里重写onerror事件。在前面这篇《》里,楼猪已经贴了代码,就不再费事了。根据经验,很多人开发的时候几乎都这么写,而且对调试和维护还是很有帮助的。需要说明的是,每新添一个页面,其对应类都必须继承自basepage类异常处理才起作用。

2、global.asax处理方式

如1中所述,basepage类的异常处理要求每一个aspx类文件都继承它,适用性和性能显然会打折扣。而global.asax文件定义了asp.net应用程序中的所有应用程序对象共有的方法、属性和事件,我们可以不采用basepage的处理方式,在global.asax里实现application_error事件并处理也可以。下面模仿basepage类里的处理异常方法,实现如下:

/// <summary>
/// 出错处理:写日志,导航到公共出错页面
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
protected void application_error(object sender, eventargs e)
{
  if (server.getlasterror() == null) return;
  exception ex = server.getlasterror().getbaseexception();
  string error = this.dealexception(ex);
  dotnet.common.util.logger.writefilelog(error, httpcontext.current.request.physicalapplicationpath + "logfile");
  if (ex.innerexception != null)
  {
    error = this.dealexception(ex);
    dotnet.common.util.logger.writefilelog(error, httpcontext.current.request.physicalapplicationpath + "logfile");
  }
  this.server.clearerror();
  this.response.redirect("/error.aspx");
}
/// <summary>
/// 处理异常,用来将主要异常信息写入文本日志
/// </summary>
/// <param name="ex"></param>
/// <returns></returns>
private string dealexception(exception ex)
{
  this.application["stacktrace"] = ex.stacktrace;
  this.application["messageerror"] = ex.message;
  this.application["sourceerror"] = ex.source;
  this.application["targetsite"] = ex.targetsite.tostring();
  string error = string.format("url:{0}\n引发异常的方法:{1}\n错误信息:{2}\n错误堆栈:{3}\n",
    this.request.rawurl, ex.targetsite, ex.message, ex.stacktrace);
  return error;
}

上面方式的好处是,写一次代码,应用程序发生的大部分异常它都给你捕捉处理了。楼猪要在这里由衷地发一番感慨,感谢ms为我们提供了这么优秀的框架,太省事了吧。

3、ihttpmodule接口处理

1和2的处理方式大家都是非常熟悉的,楼猪在实际开发中基本上都是遵循上面两种写法,而且楼猪因为有了2中这种大小通吃的处理方式,甚至已经激动地感谢ms了。但是,在asp.net程序调用线程进行异步处理的时候,容易发生在后台线程或线程池里抛出的异常并不能被1或(和)2完全捕捉到,这就涉及到asp.net下未捕获异常的处理。也就是说楼猪以前做过的很多大小项目中对异常的处理是不完备的。这难道是nc楼猪没有先谢国家种下的恶果吗?感谢国家,感谢ms,感谢博客园,感谢无私的xdjm,感谢自己......

asp.net下未捕获异常的处理步骤如下:

(1)、创建一个实现ihttpmodule接口的类

using system;
using system.collections.generic;
using system.linq;
using system.web;
using system.web.ui;
using system.web.ui.webcontrols;
using system.text;
namespace dotnet.common.webform
{
  using dotnet.common.util;
  /// <summary>
  /// 通用未捕获异常处理 
  /// </summary>
  public class aspnetunhandledexceptionmodule : ihttpmodule
  {
    static object syncobj = new object();
    static bool isinit = false;
    public aspnetunhandledexceptionmodule()
    {
    }
    #region ihttpmodule methods
    public void init(httpapplication context)
    {
      lock (syncobj)
      {
        if (!isinit)
        {
          appdomain.currentdomain.unhandledexception += new unhandledexceptioneventhandler(onunhandledexception);
          isinit = true;
        }
      }
    }
    public void dispose()
    {
    }
    #endregion
    #region onunhandledexception
    void onunhandledexception(object o, unhandledexceptioneventargs e)
    {
      if (e.exceptionobject == null) return;
      exception ex = e.exceptionobject as exception;
      string error = string.format("引发异常的方法:{0}\n错误信息:{1}\n错误堆栈:{2}\n",
              ex.targetsite, ex.message, ex.stacktrace);
      logger.writefilelog(error, appdomain.currentdomain.basedirectory + "logfile");
    }
    #endregion
  }
}

(2)、web.config节点配置

<httpmodules>
   <add name="aspnetunhandledexceptionmodule" type="dotnet.common.webform.aspnetunhandledexceptionmodule, dotnet.common.webform"></add>
</httpmodules>

最后贴出测试代码:

protected void page_load(object sender, eventargs e)
{
  if (!ispostback)
  {
    system.threading.threadpool.queueuserworkitem(new system.threading.waitcallback(test), null);
  }
}
protected void test(object state)
{
  int[] numarr = new int[100];
  numarr[100] = 100; //异常
}

需要说明的是,通过线程或者线程池处理的程序,在发生异常时,每个线程都会有它自己独立的上下文,所以httpcontext对象应尽可能少地出现在异常处理阶段。

小结:不知道还有多少童鞋认为异常处理就是在代码里try...catch一下,抛出异常然后完事?如果有的话,呵呵,当年楼猪是拿“没有人天生就是十全十美的”这句话来安慰自己的。当然了,try...catch也不是不可以,只能说明我们对待异常的态度太草率了。为了显得我们的专业和全面,请参考其他异常处理专业性文章研读一番,相比异常处理的核心思想(异常处理的“大智慧”),这篇文章总结的(异常处理的“小技巧”)对初学者而言可能也是误导之作,请务必留意甄别。

完整实例代码代码点击此处本站下载

希望本文所述对大家asp.net程序设计有所帮助。

上一篇:

下一篇: