C#实现抓取和分析网页类实例

本文实例讲述了C#实现抓取和分析网页类。分享给大家供大家参考。具体分析如下：

这里介绍了抓取和分析网页的类。

其主要功能有：

1、提取网页的纯文本，去所有html标签和javascript代码
2、提取网页的链接，包括href和frame及iframe
3、提取网页的title等(其它的标签可依此类推，正则是一样的)
4、可以实现简单的表单提交及cookie保存

/*
* Author:Sunjoy at CCNU
* 如果您改进了这个类请发一份代码给我(ccnusjy 在gmail.com)
*/
using System;
using System.Data;
using System.Configuration;
using System.Net;
using System.IO;
using System.Text;
using System.Collections.Generic;
using System.Text.RegularExpressions;
using System.Threading;
using System.Web;
/// <summary>
/// 网页类
/// </summary>
public class WebPage
{
 #region 私有成员
 private Uri m_uri; //网址
 private List<Link> m_links; //此网页上的链接
 private string m_title;  //此网页的标题
 private string m_html;   //此网页的HTML代码
 private string m_outstr;  //此网页可输出的纯文本
 private bool m_good;   //此网页是否可用
 private int m_pagesize;  //此网页的大小
 private static Dictionary<string, CookieContainer> webcookies = new Dictionary<string, CookieContainer>();//存放所有网页的Cookie
 private string m_post; //此网页的登陆页需要的POST数据
 private string m_loginurl; //此网页的登陆页
 #endregion
 #region 私有方法
 /// <summary>
 /// 这私有方法从网页的HTML代码中分析出链接信息
 /// </summary>
 /// <returns>List<Link></returns>
 private List<Link> getLinks()
 {
  if (m_links.Count == 0)
  {
   Regex[] regex = new Regex[2];
   regex[0] = new Regex("(?m)<a[^><]+href=(\"|')?(?<url>([^>\"'\\s)])+)(\"|')?[^>]*>(?<text>(\\w|\\W)*?)</", RegexOptions.Multiline | RegexOptions.IgnoreCase);
   regex[1] = new Regex("<[i]*frame[^><]+src=(\"|')?(?<url>([^>\"'\\s)])+)(\"|')?[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase);
   for (int i = 0; i < 2; i++)
   {
    Match match = regex[i].Match(m_html);
    while (match.Success)
    {
     try
     {
      string url = new Uri(m_uri, match.Groups["url"].Value).AbsoluteUri;
      string text = "";
      if (i == 0) text = new Regex("(<[^>]+>)|(\\s)|( )|&|\"", RegexOptions.Multiline | RegexOptions.IgnoreCase).Replace(match.Groups["text"].Value, "");
      Link link = new Link(url, text);
      m_links.Add(link);
     }
     catch(Exception ex){Console.WriteLine(ex.Message); };
     match = match.NextMatch();
    }
   }
  }
  return m_links;
 }
 /// <summary>
 /// 此私有方法从一段HTML文本中提取出一定字数的纯文本
 /// </summary>
 /// <param name="instr">HTML代码</param>
 /// <param name="firstN">提取从头数多少个字</param>
 /// <param name="withLink">是否要链接里面的字</param>
 /// <returns>纯文本</returns>
 private string getFirstNchar(string instr, int firstN, bool withLink)
 {
  if (m_outstr == "")
  {
   m_outstr = instr.Clone() as string;
   m_outstr = new Regex(@"(?m)<script[^>]*>(\w|\W)*?</script[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).Replace(m_outstr, "");
   m_outstr = new Regex(@"(?m)<style[^>]*>(\w|\W)*?</style[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).Replace(m_outstr, "");
   m_outstr = new Regex(@"(?m)<select[^>]*>(\w|\W)*?</select[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).Replace(m_outstr, "");
   if (!withLink) m_outstr = new Regex(@"(?m)<a[^>]*>(\w|\W)*?</a[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase).Replace(m_outstr, "");
   Regex objReg = new System.Text.RegularExpressions.Regex("(<[^>]+?>)| ", RegexOptions.Multiline | RegexOptions.IgnoreCase);
   m_outstr = objReg.Replace(m_outstr, "");
   Regex objReg2 = new System.Text.RegularExpressions.Regex("(\\s)+", RegexOptions.Multiline | RegexOptions.IgnoreCase);
   m_outstr = objReg2.Replace(m_outstr, " ");
  }
  return m_outstr.Length > firstN ? m_outstr.Substring(0, firstN) : m_outstr;
 }
 /// <summary>
 /// 此私有方法返回一个IP地址对应的无符号整数
 /// </summary>
 /// <param name="x">IP地址</param>
 /// <returns></returns>
 private uint getuintFromIP(IPAddress x)
 {
  Byte[] bt = x.GetAddressBytes();
  uint i = (uint)(bt[0] * 256 * 256 * 256);
  i += (uint)(bt[1] * 256 * 256);
  i += (uint)(bt[2] * 256);
  i += (uint)(bt[3]);
  return i;
 }
 #endregion
 #region 公有文法
 /// <summary>
 /// 此公有方法提取网页中一定字数的纯文本，包括链接文字
 /// </summary>
 /// <param name="firstN">字数</param>
 /// <returns></returns>
 public string getContext(int firstN)
 {
  return getFirstNchar(m_html, firstN, true);
 }
 /// <summary>
 /// 此公有方法提取网页中一定字数的纯文本，不包括链接文字
 /// </summary>
 /// <param name="firstN"></param>
 /// <returns></returns>
 public string getContextWithOutLink(int firstN)
 {
  return getFirstNchar(m_html, firstN, false);
 }
 /// <summary>
 /// 此公有方法从本网页的链接中提取一定数量的链接，该链接的URL满足某正则式
 /// </summary>
 /// <param name="pattern">正则式</param>
 /// <param name="count">返回的链接的个数</param>
 /// <returns>List<Link></returns>
 public List<Link> getSpecialLinksByUrl(string pattern,int count)
 {
  if(m_links.Count==0)getLinks();
  List<Link> SpecialLinks = new List<Link>();
  List<Link>.Enumerator i;
  i = m_links.GetEnumerator();
  int cnt = 0;
  while (i.MoveNext() && cnt<count)
  {
   if (new Regex(pattern, RegexOptions.Multiline | RegexOptions.IgnoreCase ).Match(i.Current.url).Success)
   {
    SpecialLinks.Add(i.Current);
    cnt++;
   }
  } 
  return SpecialLinks;
 }
 /// <summary>
 /// 此公有方法从本网页的链接中提取一定数量的链接，该链接的文字满足某正则式
 /// </summary>
 /// <param name="pattern">正则式</param>
 /// <param name="count">返回的链接的个数</param>
 /// <returns>List<Link></returns>
 public List<Link> getSpecialLinksByText(string pattern,int count)
 {
  if (m_links.Count == 0) getLinks();
  List<Link> SpecialLinks = new List<Link>();
  List<Link>.Enumerator i;
  i = m_links.GetEnumerator();
  int cnt = 0;
  while (i.MoveNext() && cnt < count)
  {
   if (new Regex(pattern, RegexOptions.Multiline | RegexOptions.IgnoreCase ).Match(i.Current.text).Success)
   {
    SpecialLinks.Add(i.Current);
    cnt++;
   }
  }
  return SpecialLinks;
 }
 /// <summary>
 /// 此公有方法获得所有链接中在一定IP范围的链接
 /// </summary>
 /// <param name="_ip_start">起始IP</param>
 /// <param name="_ip_end">终止IP</param>
 /// <returns></returns>
 public List<Link> getSpecialLinksByIP(string _ip_start, string _ip_end)
 {
  IPAddress ip_start = IPAddress.Parse(_ip_start);
  IPAddress ip_end = IPAddress.Parse(_ip_end);
  if (m_links.Count == 0) getLinks();
  List<Link> SpecialLinks = new List<Link>();
  List<Link>.Enumerator i;
  i = m_links.GetEnumerator();
  while (i.MoveNext())
  {
   IPAddress ip;
   try
   {
    ip = Dns.GetHostEntry(new Uri(i.Current.url).Host).AddressList[0];
   }
   catch { continue; }
   if(getuintFromIP(ip)>=getuintFromIP(ip_start) && getuintFromIP(ip)<=getuintFromIP(ip_end))
   {
    SpecialLinks.Add(i.Current);
   }
  }
  return SpecialLinks;
 }
 /// <summary>
 /// 这公有方法提取本网页的纯文本中满足某正则式的文字
 /// </summary>
 /// <param name="pattern">正