C#完成抓取和剖析網頁類實例。本站提示廣大學習愛好者:(C#完成抓取和剖析網頁類實例)文章只能為提供參考,不一定能成為您想要的結果。以下是C#完成抓取和剖析網頁類實例正文
本文實例講述了C#完成抓取和剖析網頁類。分享給年夜家供年夜家參考。詳細剖析以下:
這裡引見了抓取和剖析網頁的類。
其重要功效有:
1、提取網頁的純文本,去一切html標簽和javascript代碼
2、提取網頁的鏈接,包含href和frame及iframe
3、提取網頁的title等(其它的標簽可依此類推,正則是一樣的)
4、可以完成簡略的表單提交及cookie保留
/* * Author:Sunjoy at CCNU * 假如您改良了這個類請發一份代碼給我(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> /// 網頁類 /// </summary> public class WebPage { #region 公有成員 private Uri m_uri; //網址 private List<Link> m_links; //此網頁上的鏈接 private string m_title; //此網頁的題目 private string m_html; //此網頁的HTML代碼 private string m_outstr; //此網頁可輸入的純文本 private bool m_good; //此網頁能否可用 private int m_pagesize; //此網頁的年夜小 private static Dictionary<string, CookieContainer> webcookies = new Dictionary<string, CookieContainer>();//寄存一切網頁的Cookie private string m_post; //此網頁的上岸頁須要的POST數據 private string m_loginurl; //此網頁的上岸頁 #endregion #region 公有辦法 /// <summary> /// 這公有辦法從網頁的HTML代碼平分析出鏈接信息 /// </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).WordStr(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> /// 此公有辦法從一段HTML文本中提掏出必定字數的純文本 /// </summary> /// <param name="instr">HTML代碼</param> /// <param name="firstN">提取從頭數若干個字</param> /// <param name="withLink">能否要鏈接外面的字</param> /// <returns>純文本</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 ).WordStr(m_outstr, ""); m_outstr = new Regex(@"(?m)<style[^>]*>(\w|\W)*?</style[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).WordStr(m_outstr, ""); m_outstr = new Regex(@"(?m)<select[^>]*>(\w|\W)*?</select[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ).WordStr(m_outstr, ""); if (!withLink) m_outstr = new Regex(@"(?m)<a[^>]*>(\w|\W)*?</a[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase).WordStr(m_outstr, ""); Regex objReg = new System.Text.RegularExpressions.Regex("(<[^>]+?>)| ", RegexOptions.Multiline | RegexOptions.IgnoreCase); m_outstr = objReg.WordStr(m_outstr, ""); Regex objReg2 = new System.Text.RegularExpressions.Regex("(\\s)+", RegexOptions.Multiline | RegexOptions.IgnoreCase); m_outstr = objReg2.WordStr(m_outstr, " "); } return m_outstr.Length > firstN ? m_outstr.Substring(0, firstN) : m_outstr; } /// <summary> /// 此公有辦法前往一個IP地址對應的無符號整數 /// </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 私有文法 /// <summary> /// 此私有辦法提取網頁中必定字數的純文本,包含鏈接文字 /// </summary> /// <param name="firstN">字數</param> /// <returns></returns> public string getContext(int firstN) { return getFirstNchar(m_html, firstN, true); } /// <summary> /// 此私有辦法提取網頁中必定字數的純文本,不包含鏈接文字 /// </summary> /// <param name="firstN"></param> /// <returns></returns> public string getContextWithOutLink(int firstN) { return getFirstNchar(m_html, firstN, false); } /// <summary> /// 此私有辦法從本網頁的鏈接中提取必定數目的鏈接,該鏈接的URL知足某正則式 /// </summary> /// <param name="pattern">正則式</param> /// <param name="count">前往的鏈接的個數</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> /// 此私有辦法從本網頁的鏈接中提取必定數目的鏈接,該鏈接的文字知足某正則式 /// </summary> /// <param name="pattern">正則式</param> /// <param name="count">前往的鏈接的個數</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> /// 此私有辦法取得一切鏈接中在必定IP規模的鏈接 /// </summary> /// <param name="_ip_start">肇端IP</param> /// <param name="_ip_end">終止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> /// 這私有辦法提取本網頁的純文本中知足某正則式的文字 /// </summary> /// <param name="pattern">正則式</param> /// <returns>前往文字</returns> 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 結構函數 private void Init(string _url) { try { m_uri = new Uri(_url); m_links = new List<Link>(); 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=(?<cding>[^=]+)?\"",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("(?<h>[^\x00-\xff]+)"); Match mc = re.Match(_url); if (mc.Success) { string han = mc.Groups["h"].Value; _url = _url.WordStr(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("(?<h>[^\x00-\xff]+)"); Match mc = re.Match(_url); if (mc.Success) { string han = mc.Groups["h"].Value; _url = _url.WordStr(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 上岸 string indata = _post; m_post = _post; m_loginurl = _loginurl; byte[] bytes = Encoding.Default.GetBytes(_post); CookieContainer myCookieContainer = new CookieContainer(); try { //新建一個CookieContainer來寄存Cookie聚集 HttpWebRequest myHttpWebRequest = (HttpWebRequest)WebRequest.Create(_loginurl); //新建一個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; //設置HttpWebRequest的CookieContainer為適才樹立的誰人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 上岸後再拜訪頁面 try { m_uri = new Uri(_url); m_links = new List<Link>(); 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 屬性 /// <summary> /// 經由過程此屬性可取得本網頁的網址,只讀 /// </summary> public string URL { get { return m_uri.AbsoluteUri; } } /// <summary> /// 經由過程此屬性可取得本網頁的題目,只讀 /// </summary> public string Title { get { if (m_title == "") { Regex reg = new Regex(@"(?m)<title[^>]*>(?<title>(?:\w|\W)*?)</title[^>]*>", RegexOptions.Multiline | RegexOptions.IgnoreCase ); Match mc = reg.Match(m_html); if (mc.Success) m_title= mc.Groups["title"].Value.Trim(); } return m_title; } } /// <summary> /// 此屬性取得本網頁的一切鏈接信息,只讀 /// </summary> public List<Link> Links { get { if (m_links.Count == 0) getLinks(); return m_links; } } /// <summary> /// 此屬性前往本網頁的全體純文本信息,只讀 /// </summary> public string Context { get { if (m_outstr == "") getContext(Int16.MaxValue); return m_outstr; } } /// <summary> /// 此屬性取得本網頁的年夜小 /// </summary> public int PageSize { get { return m_pagesize; } } /// <summary> /// 此屬性取得本網頁的一切站內鏈接 /// </summary> public List<Link> InsiteLinks { get { return getSpecialLinksByUrl("^http://"+m_uri.Host,Int16.MaxValue); } } /// <summary> /// 此屬性表現本網頁能否可用 /// </summary> public bool IsGood { get { return m_good; } } /// <summary> /// 此屬性表現網頁的地點的網站 /// </summary> public string Host { get { return m_uri.Host; } } /// <summary> /// 此網頁的上岸頁所需的POST數據 /// </summary> public string PostStr { get { return m_post; } } /// <summary> /// 此網頁的上岸頁 /// </summary> public string LoginURL { get { return m_loginurl; } } #endregion } /// <summary> /// 鏈接類 /// </summary> public class Link { public string url; //鏈接網址 public string text; //鏈接文字 public Link(string _url, string _text) { url = _url; text = _text; } }
願望本文所述對年夜家的C#法式設計有所贊助。