Question

我正在构建一个相当简单的CMS。我需要拦截Web应用程序中大多数.aspx页面的请求，以便完全控制输出。在大多数情况下，输出将从缓存中提取，并且只是纯HTML。

但是，我还需要使用几个页面来使用asp：controls on。我认为绕过一些特定请求的最好方法是继承System.Web.UI.PageHandlerFactory，并在需要时调用MyBase实现（如果我错了，请纠正我）。但是如何将所有其他请求转移到我的自定义处理程序？

Answer 1

当我写一个简单的CMS时，我很难使用PageHandlerFactory让它做我想做的事情。最后我换成了IHttpModule。

我的模块首先会检查请求路径中是否有.aspx文件。如果页面上有用户控件或由于某种原因不适合CMS，我只会这样做。因此，如果文件存在，它将返回模块。之后，它将查看请求的路径并将其压缩为“导航标记”。因此〜/ aboutus / default.aspx将成为page.aspx？nt = aboutusdefault。 page.aspx将从CMS加载适当的内容。当然，重定向发生在服务器端，因此用户/蜘蛛永远不会知道发生了什么不同。

using System;
using System.Data;
using System.Collections.Generic;
using System.Configuration;
using System.Reflection;
using System.Text.RegularExpressions;
using System.Web;

namespace MyCMS.Handlers {
    /// <summary>
    /// Checks to see if we should display a virutal page to replace the current request.
    /// Code adapted from:
    /// Rewrite.NET -- A URL Rewriting Engine for .NET
    /// By Robert Chartier
    /// http://www.15seconds.com/issue/030522.htm
    /// </summary>
    public class VirtualPageModule : IHttpModule {
        /// <summary>
        /// Init is required from the IHttpModule interface
        /// </summary>
        /// <param name="Appl"></param>
        public void Init(System.Web.HttpApplication Appl) {
            // make sure to wire up to BeginRequest
            Appl.BeginRequest += new System.EventHandler(Rewrite_BeginRequest);
        }

        /// <summary>
        /// Dispose is required from the IHttpModule interface
        /// </summary>
        public void Dispose() {
            // make sure you clean up after yourself
        }

        /// <summary>
        /// To handle the starting of the incoming request
        /// </summary>
        /// <param name="sender"></param>
        /// <param name="args"></param>
        public void Rewrite_BeginRequest(object sender, System.EventArgs args) {
            // Cast the sender to an HttpApplication object
            HttpApplication httpApp = (HttpApplication)sender;

            // See if the requested file already exists
            if (System.IO.File.Exists(httpApp.Request.PhysicalPath)) {
                // Do nothing, process the request as usual
                return;
            }

            string requestPath = VirtualPathUtility.ToAppRelative(httpApp.Request.Path);

            // Organic navigation tag (~/aboutus/default.aspx = nt "aboutusdefault")
            Regex regex = new Regex("[~/\\!@#$%^&*()+=-]");
            requestPath = regex.Replace(requestPath, string.Empty).Replace(".aspx", string.Empty);
            string pageName = "~/page.aspx";
            string destinationUrl = VirtualPathUtility.ToAbsolute(pageName) + "?nt=" + requestPath;
            SendToNewUrl(destinationUrl, httpApp);
        }

        public void SendToNewUrl(string url, HttpApplication httpApp) {
            applyTrailingSlashHack(httpApp);
            httpApp.Context.RewritePath(
                url,
                false // RebaseClientPath must be false for ~/ to continue working in subdirectories.
            );
        }

        /// <summary>
        /// Applies the trailing slash hack. To circumvent an ASP.NET bug related to dynamically
        /// generated virtual directories ending in a trailing slash (/).
        /// As described by BuddyDvd:
        /// http://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?FeedbackID=105061
        /// </summary>
        /// <param name="httpApp">The HttpApplication.</param>
        /// <remarks>
        /// Execute this function before calling RewritePath.
        /// </remarks>
        private void applyTrailingSlashHack(HttpApplication httpApp) {
            if (httpApp.Request.Url.AbsoluteUri.EndsWith("/") && !httpApp.Request.Url.AbsolutePath.Equals("/")) {
                Type requestType = httpApp.Context.Request.GetType();
                object clientFilePath = requestType.InvokeMember("ClientFilePath", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.GetProperty, null, httpApp.Context.Request, null);
                string virtualPathString = (string)clientFilePath.GetType().InvokeMember("_virtualPath", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.GetField, null, clientFilePath, null);
                clientFilePath.GetType().InvokeMember("_virtualPath", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.SetField, null, clientFilePath, new object[] { virtualPathString });
                requestType.InvokeMember("_clientFilePath", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.SetField, null, HttpContext.Current.Request, new object[] { clientFilePath });
                object clientBaseDir = requestType.InvokeMember("ClientBaseDir", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.GetProperty, null, httpApp.Context.Request, null);
                clientBaseDir.GetType().InvokeMember("_virtualPath", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.SetField, null, clientBaseDir, new object[] { virtualPathString });
                requestType.InvokeMember("_clientBaseDir", BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.SetField, null, HttpContext.Current.Request, new object[] { clientBaseDir });
            }
        }
    }
}

Answer 2

你的意思是你要注射控制吗？如果是这种情况，您可能需要考虑必需的基类而不是Page类。 Page实现了IHttpHandler，因此您可以创建派生类，然后更改页面以从派生类派生。您将对页面拥有更多控制权，并能够挂钩并呈现它。

.aspx的自定义PageHandlerFactory

2 个答案: