/// <summary>
/// 置换连接
/// </summary>
private static string _ReplaceUrl(string strRe, string subMatch, string sFormartted, string sPageUrl)
{
Regex re = new Regex(strRe, RegexOptions.Multiline | RegexOptions.IgnorePatternWhitespace | RegexOptions.IgnoreCase);
MatchCollection mcs = re.Matches(sFormartted);
string sOriStr = "";
string sSubMatch = "";
string sReplaceStr = "";
foreach (Match mc in mcs)
{
sOriStr = mc.Value;
sSubMatch = mc.Groups[subMatch].Value;
sReplaceStr = sOriStr.Replace(sSubMatch, CRegex.GetUrl(sPageUrl, sSubMatch));
sFormartted = sFormartted.Replace(sOriStr, sReplaceStr);
}
return(sFormartted);
}
public static Dictionary <string, string> GetLinks(string sContent, string sUrl, ref Dictionary <string, string> lisDes)
{
Dictionary <string, string> lisA = new Dictionary <string, string>();
_GetLinks(sContent, sUrl, ref lisA);
string domain = CRegex.GetDomain(sUrl).ToLower();
//抓取脚本输出的链接
Regex re = new Regex(@"<script[^>]+src\s*=\s*(?:'(?<src>[^']+)'|""(?<src>[^""]+)""|(?<src>[^>\s]+))\s*[^>]*>", RegexOptions.Multiline | RegexOptions.IgnorePatternWhitespace | RegexOptions.IgnoreCase);
MatchCollection mcs = re.Matches(sContent);
//foreach (Match mc in mcs)
for (int i = mcs.Count - 1; i >= 0; i--)
{
Match mc = mcs[i];
string subUrl = CRegex.GetUrl(sUrl, mc.Groups["src"].Value);
if (domain.CompareTo(CRegex.GetDomain(subUrl).ToLower()) != 0)
{
//同一域的才提炼
continue;
}
string subContent = CSocket.GetHtmlByUrl(subUrl);
if (subContent.Length == 0)
{
continue;
}
_GetLinks(subContent, subUrl, ref lisA);
}
if (lisA.Count == 0)
{
return(GetLinksFromRss(sContent, sUrl, ref lisDes));
}
return(lisA);
}
private static void _GetLinks(string sContent, string sUrl, ref Dictionary <string, string> lisA)
{
const string sFilter =
@"首页|下载|中文|English|反馈|讨论区|投诉|建议|联系|关于|about|诚邀|工作|简介|新闻|掠影|风采
|登录|注销|注册|使用|体验|立即|收藏夹|收藏|添加|加入
|更多|more|专题|精选|热卖|热销|推荐|精彩
|加盟|联盟|友情|链接|相关
|订阅|阅读器|RSS
|免责|条款|声明|我的|我们|组织|概况|有限|免费|公司|法律|导航|广告|地图|隐私
|〖|〗|【|】|（|）|［|］|『|』|\.";
Regex re = new Regex(@"<a\s+[^>]*href\s*=\s*[^>]+>[\s\S]*?</a>", RegexOptions.IgnoreCase | RegexOptions.IgnorePatternWhitespace | RegexOptions.Multiline);
Regex re2 = new Regex(@"""|'", RegexOptions.IgnoreCase | RegexOptions.IgnorePatternWhitespace | RegexOptions.Multiline);
MatchCollection mcs = re.Matches(sContent);
//foreach (Match mc in mcs)
for (int i = mcs.Count - 1; i >= 0; i--)
{
Match mc = mcs[i];
string strHref = GetLink(mc.Value).Trim();
strHref = strHref.Replace("\\\"", "");//针对JS输出链接
strHref = strHref.Replace("\\\'", "");
string strTemp = RemoveByReg(strHref, @"^http.*/$");//屏蔽以“http”开头“/”结尾的链接地址
if (strTemp.Length < 2)
{
continue;
}
//过滤广告或无意义的链接
string strText = CString.ClearTag(GetTextByLink(mc.Value)).Trim();
strTemp = RemoveByReg(strText, sFilter);
if (CString.GetLength(strTemp) < 9)
{
continue;
}
if (re2.IsMatch(strText))
{
continue;
}
//换上绝对地址
strHref = CText.GetUrlByRelative(sUrl, strHref);
if (strHref.Length <= 18)//例如，http://www.163.com = 18
{
continue;
}
//计算#字符出现的位置，移除它后面的内容
//如果是域名地址，就跳过
int charIndex = strHref.IndexOf('#');
if (charIndex > -1)
{
strHref = strHref.Substring(0, charIndex);
}
strHref = strHref.Trim(new char[] { '/', '\\' });
string tmpDomainURL = CRegex.GetDomain(strHref);
if (strHref.Equals(tmpDomainURL, StringComparison.OrdinalIgnoreCase))
{
continue;
}
if (!lisA.ContainsKey(strHref) && !lisA.ContainsValue(strText))
{
lisA.Add(strHref, strText);
}
}
}
public static string GetContent(string sOriContent, string sOtherRemoveReg, string sPageUrl, DataTable dtAntiLink)
{
string sFormartted = sOriContent;
//去掉有危险的标记
sFormartted = Regex.Replace(sFormartted, @"<script[\s\S]*?</script>", "", RegexOptions.Multiline | RegexOptions.IgnorePatternWhitespace | RegexOptions.IgnoreCase);
sFormartted = Regex.Replace(sFormartted, @"<iframe[^>]*>[\s\S]*?</iframe>", "", RegexOptions.Multiline | RegexOptions.IgnorePatternWhitespace | RegexOptions.IgnoreCase);
Regex r = new Regex(@"<input[\s\S]+?>|<form[\s\S]+?>|</form[\s\S]*?>|<select[\s\S]+?>?</select>|<textarea[\s\S]*?>?</textarea>|<file[\s\S]*?>|<noscript>|</noscript>", RegexOptions.IgnoreCase);
sFormartted = r.Replace(sFormartted, "");
string[] sOtherReg = sOtherRemoveReg.Split(new string[] { "\r\n" }, StringSplitOptions.RemoveEmptyEntries);
foreach (string sRemoveReg in sOtherReg)
{
sFormartted = CRegex.Replace(sFormartted, sRemoveReg, "", 0);
}
//图片路径
//sFormartted = _ReplaceUrl("<img[^>]+src\\s*=\\s*(?:'(?<src>[^']+)'|\"(?<src>[^\"]+)\"|(?<src>[^>\\s]+))\\s*[^>]*>", "src", sFormartted,sPageUrl);
sFormartted = _ReplaceUrl("<img[\\s\\S]+?src\\s*=\\s*(?:'(?<src>[^']+)'|\"(?<src>[^\"]+)\"|(?<src>[^>\\s]+))\\s*[^>]*>", "src", sFormartted, sPageUrl);
//反防盗链
string domain = GetDomain(sPageUrl);
DataRow[] drs = dtAntiLink.Select("Domain='" + domain + "'");
if (drs.Length > 0)
{
foreach (DataRow dr in drs)
{
switch (Convert.ToInt32(dr["Type"]))
{
case 1: //置换
sFormartted = sFormartted.Replace(dr["imgUrl"].ToString(), "http://stat.580k.com/t.asp?url=");
break;
default: //附加
sFormartted = sFormartted.Replace(dr["imgUrl"].ToString(), "http://stat.580k.com/t.asp?url=" + dr["imgUrl"].ToString());
break;
}
}
}
//A链接
sFormartted = _ReplaceUrl(@"<a[^>]+href\s*=\s*(?:'(?<href>[^']+)'|""(?<href>[^""]+)""|(?<href>[^>\s]+))\s*[^>]*>", "href", sFormartted, sPageUrl);
//CSS
sFormartted = _ReplaceUrl(@"<link[^>]+href\s*=\s*(?:'(?<href>[^']+)'|""(?<href>[^""]+)""|(?<href>[^>\s]+))\s*[^>]*>", "href", sFormartted, sPageUrl);
//BACKGROUND
sFormartted = _ReplaceUrl(@"background\s*=\s*(?:'(?<img>[^']+)'|""(?<img>[^""]+)""|(?<img>[^>\s]+))", "img", sFormartted, sPageUrl);
//style方式的背景：background-image:url(...)
sFormartted = _ReplaceUrl(@"background-image\s*:\s*url\s*\x28(?<img>[^\x29]+)\x29", "img", sFormartted, sPageUrl);
//FLASH
sFormartted = _ReplaceUrl(@"<param\s[^>]+""movie""[^>]+value\s*=\s*""(?<flash>[^"">]+\x2eswf)""[^>]*>", "flash", sFormartted, sPageUrl);
//XSL
if (IsXml(sFormartted))
{
sFormartted = _ReplaceUrl(@"<\x3fxml-stylesheet\s+[^\x3f>]+href=\s*(?:'(?<href>[^']+)'|""(?<href>[^""]+)"")\s*[^\x3f>]*\x3f>", "href", sFormartted, sPageUrl);
}
//script
//sFormartted = _ReplaceUrl(@"<script[^>]+src\s*=\s*(?:'(?<src>[^']+)'|""(?<src>[^""]+)""|(?<src>[^>\s]+))\s*[^>]*>", "src", sFormartted,sPageUrl);
return(sFormartted);
}
/// <summary>
/// 网页内容
/// </summary>
/// <param name="sInput">输入内容</param>
public static string GetHtml(string sInput)
{
return(CRegex.Replace(sInput, @"(?<Head>[^<]+)<", "", "Head"));
}
