private static void smethod_0(string string_0, string string_1, ref Dictionary <string, string> dictionary_0)
{
Regex regex = new Regex(@"<a\s+[^>]*href\s*=\s*[^>]+>[\s\S]*?</a>", RegexOptions.IgnorePatternWhitespace | RegexOptions.Multiline | RegexOptions.IgnoreCase);
Regex regex2 = new Regex("\"|'", RegexOptions.IgnorePatternWhitespace | RegexOptions.Multiline | RegexOptions.IgnoreCase);
MatchCollection matchs = regex.Matches(string_0);
for (int i = matchs.Count - 1; i >= 0; i--)
{
Match match = matchs[i];
string sContent = GetLink(match.Value).Trim().Replace("\\\"", "").Replace(@"\'", "");
if (RemoveByReg(sContent, "^http.*/$").Length >= 2)
{
string str = CString.ClearTag(GetTextByLink(match.Value)).Trim();
if ((CString.GetLength(RemoveByReg(str, "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\r\n|登录|注销|注册|使用|体验|立即|收藏夹|收藏|添加|加入\r\n|更多|more|专题|精选|热卖|热销|推荐|精彩\r\n|加盟|联盟|友情|链接|相关\r\n|订阅|阅读器|RSS\r\n|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\r\n|〖|〗|【|】|（|）|［|］|『|』|\\.")) >= 9) && !regex2.IsMatch(str))
{
sContent = GetUrlByRelative(string_1, sContent);
if (sContent.Length > 0x12)
{
int index = sContent.IndexOf('#');
if (index > -1)
{
sContent = sContent.Substring(0, index);
}
sContent = sContent.Trim(new char[] { '/', '\\' });
string domain = CRegex.GetDomain(sContent);
if (!sContent.Equals(domain, StringComparison.OrdinalIgnoreCase) && !(dictionary_0.ContainsKey(sContent) || dictionary_0.ContainsValue(str)))
{
dictionary_0.Add(sContent, str);
}
}
}
}
}
}
public static Dictionary <string, string> GetLinks(string sContent, string sUrl, ref Dictionary <string, string> lisDes)
{
Dictionary <string, string> dictionary = new Dictionary <string, string>();
smethod_0(sContent, sUrl, ref dictionary);
string str = CRegex.GetDomain(sUrl).ToLower();
MatchCollection matchs = new Regex("<script[^>]+src\\s*=\\s*(?:'(?<src>[^']+)'|\"(?<src>[^\"]+)\"|(?<src>[^>\\s]+))\\s*[^>]*>", RegexOptions.IgnorePatternWhitespace | RegexOptions.Multiline | RegexOptions.IgnoreCase).Matches(sContent);
for (int i = matchs.Count - 1; i >= 0; i--)
{
Match match = matchs[i];
string url = CRegex.GetUrl(sUrl, match.Groups["src"].Value);
if (str.CompareTo(CRegex.GetDomain(url).ToLower()) == 0)
{
string htmlByUrl = CSocket.GetHtmlByUrl(url);
if (htmlByUrl.Length != 0)
{
smethod_0(htmlByUrl, url, ref dictionary);
}
}
}
if (dictionary.Count == 0)
{
return(GetLinksFromRss(sContent, sUrl, ref lisDes));
}
return(dictionary);
}
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);
}
}
}