Asp.net获得远程网页源代码

利用.NET框架提供的 WebClient类和 WebRequest类,我们可以很轻易地得到给定URL地址的源代码,很简单,以下是C#的完整的例子.

GetPageHtml.aspx

1 <%@ Page language="c#" validateRequest="false" Codebehind="GetPageHtml.aspx.cs"
2 AutoEventWireup="false" Inherits="eMeng.Exam.GetPageHtml" %>
3     <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
4     <HTML>
5
6         <HEAD>
7             <title>
8                 得到网页源代码
9             </title>
10             <meta name="GENERATOR" Content="Microsoft Visual Studio 7.0">
11             <meta name="CODE_LANGUAGE" Content="C#">
12             <meta name="vs_defaultClientScript" content="JavaScript">
13             <meta name="vs_targetSchema" content="http://schemas.microsoft.com/intellisense/ie5">
14         </HEAD>
15
16         <body MS_POSITIONING="GridLayout">
17             <form id="aspNetBuffer" method="post" runat="server">
18                 <div align="center" style="FONT-WEIGHT: bold">
19                     得到任意网页源代码
20                 </div>
21                 <asp:TextBox id="UrlText" runat="server" Width="400px">
22                     http://dotnet.aspx.cc/content.aspx
23                 </asp:TextBox>
24                 <asp:Button id="WebClientButton" Runat="server" Text="用WebClient得到">
25                 </asp:Button>
26                 <asp:Button id="WebRequestButton" runat="server" Text="用WebRequest得到">
27                 </asp:Button>
28                 <br>
29                 <asp:TextBox id="ContentHtml" runat="server" Width="100%" Height="360px"
30                 TextMode="MultiLine">
31                 </asp:TextBox>
32             </form>
33         </body>
34
35     </HTML>

GetPageHtml.aspx.cs

using System;
using System.Collections;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Web;
using System.Web.SessionState;
using System.Web.UI;
using System.Web.UI.WebControls;
using System.Web.UI.HtmlControls;
using System.IO;
using System.Net;
using System.Text;
using System.Text.RegularExpressions;
namespace eMeng.Exam
{
/// <summary>
/// GetPageHtml 的摘要说明。
/// </summary>
public class GetPageHtml : System.Web.UI.Page
{
protected System.Web.UI.WebControls.Button WebClientButton;
protected System.Web.UI.WebControls.Button WebRequestButton;
protected System.Web.UI.WebControls.TextBox ContentHtml;
protected System.Web.UI.WebControls.TextBox UrlText;
protected System.Web.UI.WebControls.Button GetText;
private string PageUrl = "";

private void Page_Load(object sender, System.EventArgs e)
{}

#region Web Form Designer generated code
override protected void OnInit(EventArgs e)
{
InitializeComponent();
base.OnInit(e);
}

/// <summary>
/// 设计器支持所需的方法 - 不要使用代码编辑器修改
/// 此方法的内容。
/// </summary>
private void InitializeComponent()
{
    this.WebClientButton.Click += new System.EventHandler(this.WebClientButton_Click);
    this.WebRequestButton.Click += new System.EventHandler(this.WebRequestButton_Click);
    this.GetText.Click += new System.EventHandler(this.GetText_Click);
    this.Load += new System.EventHandler(this.Page_Load);

}
#endregion

private void WebClientButton_Click(object sender, System.EventArgs e)
{
PageUrl = UrlText.Text;
WebClient wc = new WebClient();
wc.Credentials = CredentialCache.DefaultCredentials;

///方法一：
Byte[] pageData = wc.DownloadData(PageUrl);
ContentHtml.Text = Encoding.Default.GetString(pageData);

/// 方法二：
/// ***************代码开始**********
/// Stream resStream = wc.OpenRead(PageUrl);
/// StreamReader sr = new StreamReader(resStream,System.Text.Encoding.Default);
/// ContentHtml.Text = sr.ReadToEnd();
/// resStream.Close();
/// **************代码结束********
///
wc.Dispose();
}

private void WebRequestButton_Click(object sender, System.EventArgs e)
{
PageUrl = UrlText.Text;
WebRequest  request = WebRequest.Create(PageUrl);
WebResponse response = request.GetResponse();
Stream resStream = response.GetResponseStream();
StreamReader sr = new StreamReader(resStream, System.Text.Encoding.Default);
ContentHtml.Text = sr.ReadToEnd();
resStream.Close();
sr.Close();
}

private void GetText_Click(object sender, System.EventArgs e)
{
    PageUrl = UrlText.Text;
    WebRequest  request = WebRequest.Create(PageUrl);
    WebResponse response = request.GetResponse();
    Stream resStream = response.GetResponseStream();
    StreamReader sr = new StreamReader(resStream, System.Text.Encoding.Default);
    ContentHtml.Text = sr.ReadToEnd();
    resStream.Close();
    sr.Close();
    ContentHtml.Text = Regex.Replace(ContentHtml.Text,"<[^>]*>", "");
    //替换空格
    ContentHtml.Text = Regex.Replace(ContentHtml.Text,"\\s+", " ");
}
}
}