/// <summary>
/// 和GetTxtFromHtml功能一样，不过保留换行符号
/// </summary>
/// <param name="sHtml"></param>
/// <returns></returns>
public static string GetTxtFromHtml2(string sHtml)
{
string del = @"<head[^>]*>[\s\S]*?</head>";
string content = CText.RemoveByReg(sHtml, del);
del = @"(<script[^>]*>[\s\S]*?</script>)|(<IFRAME[^>]*>[\s\S]*?</IFRAME>)|(<style[^>]*>[\s\S]*?</style>|<title[^>]*>[\s\S]*?</title>|<meta[^>]*>|<option[^>]*>[\s\S]*?</option>)";
content = CText.RemoveByReg(content, del);
del = @"( )|([\t]+)";//del = @"( )|([\n\t]+)";
content = CText.RemoveByReg(content, del);
string re = @"(<table(\s+[^>]*)*>)|(<td(\s+[^>]*)*>)|(<tr(\s+[^>]*)*>)|(<p(\s+[^>]*)*>)|(<div(\s+[^>]*)*>)|(<ul(\s+[^>]*)*>)|(<li(\s+[^>]*)*>)|</table>|</td>|</tr>|</p>|<br>|</div>|</li>|</ul>|<p />|<br />";
content = CText.ReplaceByReg(content, "", re);
//content = CText.ReplaceByReg(content, "", @"[\f\n\r\v]+");
content = CText.RemoveByReg(content, @"<a(\s+[^>]*)*>[\s\S]*?</a>");
content = CText.RemoveByReg(content, "<[^>]+>");//去除各种HTML标记，获得纯内容
content = content.Trim();
return(content);
}
/// <summary>
/// 获取连接信息
/// </summary>
/// <param name="sContent"></param>
/// <param name="sUrl"></param>
/// <param name="lisA"></param>
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);
}
}
}