/*-----------------------------------------------------------------------------------*/
//获取百度的当前页码
public int GetCurPage(Winista.Text.HtmlParser.Util.NodeList nodes)
{
nodes = nodes.ExtractAllNodesThatMatch(new HasAttributeFilter("id", "page"), true);
NodeFilter filter = new AndFilter(new HasAttributeFilter("class", "pc"), new HasSiblingFilter(new HasAttributeFilter("class", "fk fk_cur")));
nodes = nodes.ExtractAllNodesThatMatch(filter, true);
return(DataConverter.CLng(nodes.AsString()));
}
public DataTable GetWXBySogou(string key, int count, DateTime time)
{
string baseurl = "http://weixin.sogou.com/weixin?type=2&query={0}&fr=sgsearch&ie=utf8&_ast=1433216256&_asf=null&w=01059900&cid=null&page={1}";
if (string.IsNullOrEmpty(key))
{
return(null);
}
DataTable dt = GetStruct(key); DateTime cdate = DateTime.Now;
for (int p = 0; p * 10 < count; p++)
{
string url = string.Format(baseurl, HttpUtility.UrlEncode(key), p + 1);
string html = ieHelp.GetHtmlFromSite(url);
HtmlPage page = htmlHelp.GetPage(html);
//int cpage = GetCurPage(page.Body); if (cpage <= p) { break; }
Winista.Text.HtmlParser.Util.NodeList nodes = page.Body.ExtractAllNodesThatMatch(new HasAttributeFilter("class", "wx-rb wx-rb3"), true);
if (nodes.Count <= 0)
{
break;
}
//将其序列化为模型并存入相应类中
for (int i = 0; i < nodes.Count; i++)
{
Winista.Text.HtmlParser.Util.NodeList cnodes = nodes[i].Children;
DataRow dr = dt.NewRow();
NodeFilter f_title = new AndFilter(new HasParentFilter(new TagNameFilter("h4")), new TagNameFilter("a"));
ATag a = (ATag)cnodes.ExtractAllNodesThatMatch(f_title, true)[0];
dr["Title"] = a.StringText;
dr["Link"] = a.Link;
f_title = new AndFilter(new HasAttributeFilter("id", "weixin_account"), new TagNameFilter("a"));
ATag author_a = (ATag)cnodes.ExtractAllNodesThatMatch(f_title, true)[0];
dr["Author"] = author_a.GetAttribute("title");
f_title = new HasAttributeFilter("class", "s-p");
Div div = (Div)cnodes.ExtractAllNodesThatMatch(f_title, true)[0];
string unixtime = div.GetAttribute("t");
dr["Cdate"] = GetDateTime(unixtime);
dr["Day"] = GetDateTime(unixtime).Day;
dr["Source"] = "微信";
dt.Rows.Add(dr);
}
if (cdate < time)
{
break;
}
}
return(dt);
}
/// <summary>
/// 利用百度搜索,抓取百度,新浪,网易的博客数据
/// </summary>
public DataTable GetBlogByBaidu(string key, int count, DateTime time)
{
string baseurl = "http://www.baidu.com/s?rtt=2&tn=baiduwb&pn={0}&wd={1}";
if (string.IsNullOrEmpty(key))
{
return(null);
}
int pageSize = 10;
DataTable dt = GetStruct(key);
DateTime cdate = DateTime.Now;
for (int p = 0; p * pageSize < count; p++)//最多采100页即1000条数据
{
string url = string.Format(baseurl, p * pageSize, HttpUtility.UrlEncode(key));
//string html = htmlHelp.GetHtmlFromSite(url);
string html = htmlHelp.GetHtmlFromSite(url);
HtmlPage page = htmlHelp.GetPage(html);
//int cpage = GetCurPage(page.Body); if (cpage <= p) { break; }
Winista.Text.HtmlParser.Util.NodeList nodes = htmlHelp.GetTagList(html, "li");
nodes = nodes.ExtractAllNodesThatMatch(new HasAttributeFilter("id"));
for (int i = 0; i < nodes.Count; i++)
{
Winista.Text.HtmlParser.Util.NodeList cnodes = nodes[i].Children;
DataRow dr = dt.NewRow();
NodeFilter linkFilter = new AndFilter(new TagNameFilter("a"), new HasAttributeFilter("class", "weibo_all"));
ATag a = (ATag)cnodes.ExtractAllNodesThatMatch(linkFilter, true)[0];
dr["Title"] = "关键词:" + key;
dr["Link"] = a.Link;
dr["Author"] = cnodes.ExtractAllNodesThatMatch(new HasAttributeFilter("name", "weibo_rootnick"), true)[0].ToPlainTextString();
ATag datea = (ATag)cnodes.ExtractAllNodesThatMatch(new HasParentFilter(new HasAttributeFilter("class", "m")), true)[0];
cdate = ConverBDDate(datea.StringText);
dr["CDate"] = cdate;
dr["Day"] = cdate.Day;
dr["Source"] = "微博";
//dr["Content"] = getacBll.GetArticleFromWeb(htmlHelp.GetHtmlFromSite(a.Link),a.Link);
dt.Rows.Add(dr);
}
if (cdate < time)
{
break;
}
}
return(dt);
}
/*
* pn:页数,值为10的等数差列,是过滤掉前多少个贴子
* lm:指定时间内百度收录,值1为最近24小时,7为7天
* rn:搜索结果显示条数,取值范围10-100之间,缺少为10(无用)
*/
/// <summary>
/// 抓取百度新闻关键词数据
/// </summary>
/// <param name="key">需要抓取的关键词</param>
/// <param name="count">最多抓取多少条数据</param>
/// <param name="time">抓取该时间段之后的数据</param>
/// <returns></returns>
public DataTable GetBaiduNews(string key, int count, DateTime time)
{
if (string.IsNullOrEmpty(key))
{
return(null);
}
int pageSize = 10;
string baseurl = "http://news.baidu.com/ns?bs=%B7%F6%C0%CF%C8%CB&sr=0&cl=2&tn=news&ct=0&clk=sortbytime&rn=10&pn={0}&word={1}";
DataTable dt = GetStruct(key); DateTime cdate = DateTime.Now;
//1,根据关键词从百度上获取相关信息
for (int p = 0; p * pageSize < count; p++)
{
//样本
//<div class="result" id="1"><h3 class="c-title"><a href="http://news.qudong.com/article/329733.shtml" target="_blank" data-click="{ 'f0':'77A717EA', 'f1':'9F63F1E4', 'f2':'4CA6DD6E', 'f3':'54E5343F', 't':'1462862118' }">1万亿美金的累计收入!继苹果之后<em>微软</em>也做到了</a></h3><div class="c-summary c-row c-gap-top-small"><div class="c-span6"><a class="c_photo" href="http://news.qudong.com/article/329733.shtml" target="_blank"><img class="c-img c-img6" alt="" src="http://t12.baidu.com/it/u=2227277846,1917450397&fm=82&s=B78AA7E278021ED6942CE89C0300509B&w=121&h=81&img.JPEG"></a></div><div class="c-span18 c-span-last"><p class="c-author">驱动中国 1小时前</p>1万亿美金的收入这在寻常企业眼中就是一个天文数字,然而对于科技大鳄苹果<em>微软</em>来说,已然成为可能,根据<em>微软</em>在华盛顿州的纳税记录可以发现,该公司的历史累计收入在上个... <span class="c-info"><a class="c-cache" href="http://cache.baidu.com/c?m=9d78d513d9d430d84f9e94697c1cc0116f4381132ba1d40209d6843898732f325321a3e52878564291d27d141cb2150bafb12172404067e1c694dd5dddccc375709574743647d71f45ce18afc04324c037902da8f55fb8e4&p=ce7ec816d9c111a05beb8f624c0d&newp=8749c54ad5c51bec17aac7710f5292695912c10e38dc8a563093&user=baidu&fm=sc&query=%CE%A2%C8%ED%D6%D0%B9%FA&qid=cdb5a37a00017e68&p1=1" target="_blank" data-click="{'fm':'sc'}">百度快照</a></span></div></div></div>
string url = string.Format(baseurl, p * pageSize, HttpUtility.UrlEncode(key));
string html = htmlHelp.GetHtmlFromSite(url);
HtmlPage page = htmlHelp.GetPage(html);
int cpage = GetCurPage(page.Body); if (cpage <= p)
{
break;
}
Winista.Text.HtmlParser.Util.NodeList nodes = htmlHelp.GetTagList(html, "div");//以前版本为li
nodes = nodes.ExtractAllNodesThatMatch(new HasAttributeFilter("class", "result"));
for (int i = 0; i < nodes.Count; i++)
{
DataRow dr = dt.NewRow();
Div div = (Div)nodes[i].Children[1];
ATag a = (ATag)nodes[i].Children[0].Children.ExtractAllNodesThatMatch(new TagNameFilter("a"))[0];
dr["Title"] = a.StringText;
dr["Link"] = a.Link;
string author = ""; cdate = DateTime.Now;
GetAuthorAndDate(ref author, ref cdate, div.Children.ExtractAllNodesThatMatch(new HasAttributeFilter("class", "c-author"), true).AsString());
dr["Author"] = author;
dr["CDate"] = DataConverter.CDate(cdate);
dr["Day"] = DataConverter.CDate(cdate).Day;
dr["Source"] = "新闻";
//dr["Content"] = getacBll.GetArticleFromWeb(htmlHelp.GetHtmlFromSite(a.Link),a.Link);
dt.Rows.Add(dr);
}
if (cdate < time)
{
break;
}
}
return(dt);
}