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

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;
/// 

/// 网页类
/// 
public class WebPage
{
 #region 私有成员
 private Uri m_uri; //网址
 private List 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 webcookies = new Dictionary();//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
 /// 

 /// 这私有方法从网页的HTML代码中分析出链接信息
 /// 
 /// List
 private List getlinks()
 {
  if (m_links.Count == 0)
  {
   Regex[] regex = new Regex[2];
   regex[0] = new Regex("(?m)<]+href=("|')?(?([^>"'\s)])+)("|')?[^>]*>(?(\w|\W)*?)<]+src=("|')?(?([^>"'\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;
 }
 /// 

 /// 此私有方法从一段HTML文本中提取出一定字数的纯文本
 /// 
 /// HTML代码
 /// 提取从头数多少个字
 /// 是否要链接里面的字
 /// 纯文本
 private string getFirstNchar(string instr, int firstN, bool withlink)
 {
  if (m_outstr == "")
  {
   m_outstr = instr.Clone() as string;
   m_outstr = new Regex(@"(?m)]*>(w|W)*?]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).Replace(m_outstr, "");
   m_outstr = new Regex(@"(?m)]*>(w|W)*?]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).Replace(m_outstr, "");
   m_outstr = new Regex(@"(?m)]*>(w|W)*?]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).Replace(m_outstr, "");
   if (!withlink) m_outstr = new Regex(@"(?m)]*>(w|W)*?]*>", 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;
 }
 /// 

 /// 此私有方法返回一个IP地址对应的无符号整数
 /// 
 /// IP地址
 /// 
 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
 /// 

 /// 此公有方法提取网页中一定字数的纯文本，包括链接文字
 /// 
 /// 字数
 /// 
 public string getContext(int firstN)
 {
  return getFirstNchar(m_html, firstN, true);
 }
 /// 

 /// 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
 /// 
 /// 
 /// 
 public string getContextWithOutlink(int firstN)
 {
  return getFirstNchar(m_html, firstN, false);
 }
 /// 

 /// 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
 /// 
 /// 正则式
 /// 返回的链接的个数
 /// List
 public List getSpeciallinksByUrl(string pattern,int count)
 {
  if(m_links.Count==0)getlinks();
  List Speciallinks = new List();
  List.Enumerator i;
  i = m_links.GetEnumerator();
  int cnt = 0;
  while (i.MoveNext() && cnt
 /// 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
 /// 
 /// 正则式
 /// 返回的链接的个数
 /// List
 public List getSpeciallinksByText(string pattern,int count)
 {
  if (m_links.Count == 0) getlinks();
  List Speciallinks = new List();
  List.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;
 }
 /// 

 /// 此公有方法获得所有链接中在一定IP范围的链接
 /// 
 /// 起始IP
 /// 终止IP
 /// 
 public List 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 Speciallinks = new List();
  List.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;
 }
 /// 

 /// 这公有方法提取本网页的纯文本中满足某正则式的文字
 /// 
 /// 正则式
 /// 返回文字
 public string getSpecialWords(string pattern)
 {
  if (m_outstr == "") getContext(Int16.MaxValue);
  Regex regex = new Regex(pattern, RegexOptions.Multiline | RegexOptions.IgnoreCase );
  Match mc=regex.Match(m_outstr);
  if (mc.Success)
   return mc.Groups[1].Value;
  return string.Empty;
 }
 #endregion
 #region 构造函数
 private void Init(string _url)
 {
  try
  {
   m_uri = new Uri(_url);
   m_links = new List();
   m_html = "";
   m_outstr = "";
   m_title = "";
   m_good = true;
   if (_url.EndsWith(".rar") || _url.EndsWith(".dat") || _url.EndsWith(".msi"))
   {
    m_good = false;
    return;
   }
   HttpWebRequest rqst = (HttpWebRequest)WebRequest.Create(m_uri);
   rqst.AllowAutoRedirect = true;
   rqst.MaximumAutomaticRedirections = 3;
   rqst.UserAgent = "Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)";
   rqst.KeepAlive = true;
   rqst.Timeout = 30000;
   lock (WebPage.webcookies)
   {
    if (WebPage.webcookies.ContainsKey(m_uri.Host))
     rqst.cookieContainer = WebPage.webcookies[m_uri.Host];
    else
    {
     cookieContainer cc = new cookieContainer();
     WebPage.webcookies[m_uri.Host] = cc;
     rqst.cookieContainer = cc;
    }
   }
   HttpWebResponse rsps = (HttpWebResponse)rqst.GetResponse();
   Stream sm = rsps.GetResponseStream();
   if (!rsps.ContentType.ToLower().StartsWith("text/") || rsps.ContentLength > 1 << 22)
   {
    rsps.Close();
    m_good = false;
    return;
   }
   Encoding cding = System.Text.Encoding.Default;
   string contenttype=rsps.ContentType.ToLower();
   int ix = contenttype.IndexOf("charset=");
   if (ix != -1)
   {
    try
    {
     cding = System.Text.Encoding.GetEncoding(rsps.ContentType.Substring(ix + "charset".Length + 1));
    }
    catch
    {
     cding = Encoding.Default;
    }
    m_html = new StreamReader(sm, cding).ReadToEnd();
   }
   else
   {
    m_html = new StreamReader(sm, cding).ReadToEnd();
    Regex regex = new Regex("charset=(?[^=]+)?"",RegexOptions.IgnoreCase);
    string strcding = regex.Match(m_html).Groups["cding"].Value;
    try
    {
     cding = Encoding.GetEncoding(strcding);
    }
    catch{
     cding = Encoding.Default;
    }
    byte[] bytes=Encoding.Default.GetBytes(m_html.ToCharArray());
    m_html = cding.GetString(bytes);
    if (m_html.Split('?').Length > 100)
    {
     m_html=Encoding.Default.GetString(bytes);
    }
   }
   
   m_pagesize = m_html.Length;
   m_uri = rsps.ResponseUri;
   rsps.Close();
  }
  catch (Exception ex)
  {
   Console.WriteLine(ex.Message+m_uri.ToString());
   m_good = false;
  }
 }
 public WebPage(string _url)
 {
  string uurl = "";
  try
  {
   uurl = Uri.UnescapeDataString(_url);
   _url = uurl;
  }
  catch { };
  Regex re = new Regex("(?[^x00-xff]+)");
  Match mc = re.Match(_url);
  if (mc.Success)
  {
   string han = mc.Groups["h"].Value;
   _url = _url.Replace(han, System.Web.HttpUtility.UrlEncode(han, Encoding.GetEncoding("GB2312")));
  }
  Init(_url);
 }
 public WebPage(string _url, string _loginurl, string _post)
 {
  string uurl = "";
  try
  {
   uurl = Uri.UnescapeDataString(_url);
   _url = uurl;
  }
  catch { };
  Regex re = new Regex("(?[^x00-xff]+)");
  Match mc = re.Match(_url);
  if (mc.Success)
  {
   string han = mc.Groups["h"].Value;
   _url = _url.Replace(han, System.Web.HttpUtility.UrlEncode(han, Encoding.GetEncoding("GB2312")));
  }
  if (_loginurl.Trim() == "" || _post.Trim() == "" || WebPage.webcookies.ContainsKey(new Uri(_url).Host))
  {
   Init(_url);
  }
  else
  {
   #region 登陆
   string indata = _post;
   m_post = _post;
   m_loginurl = _loginurl;
   byte[] bytes = Encoding.Default.GetBytes(_post);
   cookieContainer mycookieContainer = new cookieContainer();
   try
   {
    //新建一个cookieContainer来存放cookie集合 
    HttpWebRequest myHttpWebRequest = (HttpWebRequest)WebRequest.Create(_loginurl);
    //新建一个HttpWebRequest 
    myHttpWebRequest.ContentType = "application/x-www-form-urlencoded";
    myHttpWebRequest.AllowAutoRedirect = false;
    myHttpWebRequest.UserAgent = "Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)";
    myHttpWebRequest.Timeout = 60000;
    myHttpWebRequest.KeepAlive = true;
    myHttpWebRequest.ContentLength = bytes.Length;
    myHttpWebRequest.Method = "POST";
    myHttpWebRequest.cookieContainer = mycookieContainer;
    //设置HttpWebRequest的cookieContainer为刚才建立的那个mycookieContainer 
    Stream myRequestStream = myHttpWebRequest.GetRequestStream();
    myRequestStream.Write(bytes, 0, bytes.Length);
    myRequestStream.Close();
    HttpWebResponse myHttpWebResponse = (HttpWebResponse)myHttpWebRequest.GetResponse();
    foreach (cookie ck in myHttpWebResponse.cookies)
    {
     mycookieContainer.Add(ck);
    }
    myHttpWebResponse.Close();
   }
   catch
   {
    Init(_url);
    return;
   }
   #endregion
   #region 登陆后再访问页面
   try
   {
    m_uri = new Uri(_url);
    m_links = new List();
    m_html = "";
    m_outstr = "";
    m_title = "";
    m_good = true;
    if (_url.EndsWith(".rar") || _url.EndsWith(".dat") || _url.EndsWith(".msi"))
    {
     m_good = false;
     return;
    }
    HttpWebRequest rqst = (HttpWebRequest)WebRequest.Create(m_uri);
    rqst.AllowAutoRedirect = true;
    rqst.MaximumAutomaticRedirections = 3;
    rqst.UserAgent = "Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)";
    rqst.KeepAlive = true;
    rqst.Timeout = 30000;
    rqst.cookieContainer = mycookieContainer;
    lock (WebPage.webcookies)
    {
     WebPage.webcookies[m_uri.Host] = mycookieContainer;
    }
    HttpWebResponse rsps = (HttpWebResponse)rqst.GetResponse();
    Stream sm = rsps.GetResponseStream();
    if (!rsps.ContentType.ToLower().StartsWith("text/") || rsps.ContentLength > 1 << 22)
    {
     rsps.Close();
     m_good = false;
     return;
    }
    Encoding cding = System.Text.Encoding.Default;
    int ix = rsps.ContentType.ToLower().IndexOf("charset=");
    if (ix != -1)
    {
     try
     {
      cding = System.Text.Encoding.GetEncoding(rsps.ContentType.Substring(ix + "charset".Length + 1));
     }
     catch
     {
      cding = Encoding.Default;
     }
    }
    m_html = new StreamReader(sm, cding).ReadToEnd();
    m_pagesize = m_html.Length;
    m_uri = rsps.ResponseUri;
    rsps.Close();
   }
   catch (Exception ex)
   {
    Console.WriteLine(ex.Message+m_uri.ToString());
    m_good = false;
   }
   #endregion
  }
 }
 #endregion
 #region 属性
 /// 

 /// 通过此属性可获得本网页的网址，只读
 /// 
 public string URL
 {
  get
  {
   return m_uri.AbsoluteUri;
  }
 }
 /// 

 /// 通过此属性可获得本网页的标题，只读
 /// 
 public string Title
 {
  get
  {
   if (m_title == "")
   {
    Regex reg = new Regex(@"(?m)]*>(?