/// <summary>
/// 获取页面的类型
/// </summary>
/// <param name="sUrl">URL地址</param>
/// <param name="sHtml">页面内容</param>
/// <returns></returns>
public static PageType GetPageType(string sUrl, ref string sHtml)
{
PageType pt = PageType.HTML;
//看有没有RSS FEED
string regRss = @"<link\s+[^>]*((type=""application/rss\+xml"")|(type=application/rss\+xml))[^>]*>";
Regex r = new Regex(regRss, RegexOptions.IgnoreCase);
Match m = r.Match(sHtml);
if (m.Captures.Count != 0)
{//有，则转向从RSS FEED中抓取
string regHref = @"href=\s*(?:'(?<href>[^']+)'|""(?<href>[^""]+)""|(?<href>[^>\s]+))";
r = new Regex(regHref, RegexOptions.IgnoreCase);
m = r.Match(m.Captures[0].Value);
if (m.Captures.Count > 0)
{
//有可能是相对路径，加上绝对路径
string rssFile = CRegex.GetUrl(sUrl, m.Groups["href"].Value);
sHtml = GetHtmlByUrl(rssFile);
pt = PageType.RSS;
}
}
else
{//看这个地址本身是不是一个Rss feed
r = new Regex(@"<rss\s+[^>]*>", RegexOptions.IgnoreCase);
m = r.Match(sHtml);
if (m.Captures.Count > 0)
{
pt = PageType.RSS;
}
}
return(pt);
}
private static string[] DealWithFrame(string strReg, string url, string content)
{
ArrayList alFrame = new ArrayList();
Regex r = new Regex(strReg, RegexOptions.IgnoreCase);
Match m = r.Match(content);
while (m.Success)
{
alFrame.Add(CRegex.GetUrl(url, m.Groups["src"].Value));
m = m.NextMatch();
}
return((string[])alFrame.ToArray(System.Type.GetType("System.String")));
}
/// <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);
}
/// <summary>
/// 获取连接信息
/// </summary>
/// <param name="sContent">网页内容</param>
/// <param name="sUrl">网页url</param>
/// <param name="lisDes">连接列表</param>
/// <returns></returns>
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);
}