/// <summary>
/// 得到网站的贴吧类数据
/// </summary>
private void GetWebTieBaInfo()
{
this.BeginInvoke(new MethodInvoker(delegate()
{
lbtieba.Text = "";
lbtieba.Visible = true;
}));
//相似链接
string Similar = "";
DataBaseServer.SQLitecommand cmd = new SQLitecommand();
//得到关键字列表
DataTable dtkey;
DataTable dtParts;
dtkey = cmd.GetTabel("select * from Keywords");
dtParts = cmd.GetTabel("SELECT * FROM partword");
//得到相似表
DataTable dtXs = new DataTable();
dtXs = cmd.GetTabel("Select * from WebAddress WHERE pid=5");
//相似表中的被抓取网址
string webInfo = "";
//要过滤链接中首页的正则
string strTopFormat = "https?://.+/";
List<string> strTop = new List<string>();
sb = new StringBuilder();
sb.Append("");
string filterStr = "";
#region 读取相似度表中的数据据,循环抓取
for (int xs = 0; xs < dtXs.Rows.Count; xs++)
{
this.BeginInvoke(new MethodInvoker(delegate()
{
lbtieba.ForeColor = Color.DarkBlue;
lbtieba.Text = "正在搜索:" + dtXs.Rows[xs]["name"].ToString();
}));
//读取相似表中要抓取的网址
webInfo = HtmlUtil.getHtml(dtXs.Rows[xs]["url"].ToString(), "");
//读取相似链接
Similar = dtXs.Rows[xs]["likeurl"].ToString();
//取出
//string[] strA = HtmlUtil.GetElementsByTagName(webInfo, "a");
List<string> strList = HtmlUtil.GetElementsByTagNameList(webInfo, "a");
string strURLformat = "https?://.[^\"]+";
TbReleaseInfo ri = new TbReleaseInfo();
string[] strA = GetLIstDate(strList.Distinct());
#region 逐个链接判断
//循环时判断是否要验证
bool isThere = false;
for (int i = 0; i < strA.Length; i++)
{
if (Program.ProClose == true) break;
Application.DoEvents();
Dictionary<string, int> events = new Dictionary<string, int>();
//创建数据对象
ModelReleaseInfo newsInfo = new ModelReleaseInfo();
try
{
//得到目标网址中的所有链接,如果未得到,那么就继续读取下一个
string tmp = strA[i];
newsInfo.Title = HtmlUtil.NoHTML(tmp);
//得到目标网址中的所有链接,如果未得到,那么就继续读取下一个
for (int j = 0; j < dtkey.Rows.Count; j++)
{
//Application.DoEvents();
string[] keys = dtkey.Rows[j][4].ToString().Split(new char[] { ' ' });
if (!events.ContainsKey(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString()))
{
events.Add(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString(), 1);
foreach (string k in keys)
{
if (!strA[i].ToLower().Contains(k.ToLower()))
{
events.Remove(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString());
break;
}
}
}
}
foreach (KeyValuePair<string, int> ev in events)
{
if (ev.Value == 1)
{
newsInfo.KeyWords += "," + ev.Key.Split(new char[] { '-' })[0] + "-" + int.Parse(ev.Key.Split(new char[] { '-' })[1]);
}
}
string[] _a = HtmlUtil.GetListByHtml(dtXs.Rows[xs]["url"].ToString(), strA[i], strURLformat);
if (_a.Length > 0)
{
strA[i] = _a[0];
}
else
{
strA[i] = "";
}
//处理含有单引号的链接
strA[i] = HtmlUtil.UrlCl(strA[i]);
//处理单引号的链接
if (strA[i].IndexOf("'") != -1)
{
strA[i] = HtmlUtil.GetstringByHtmlArray(strA[i], "https?://.[^\']+");
}
}
catch (Exception)
{
continue;
}
//得到相似值,大于0.70的认为相同,并开始抓取
if (HtmlUtil.getSimilarDegree(Similar, strA[i]) >= 0.70)
{
//判断这个链接是否已经在库中或者列表中,如果存在,此次就不再执行
strTop = HtmlUtil.GetListByHtmlArray(strA[i], strTopFormat);
if (strTop.Count != 0)
{
continue;//同新闻,如果将首页去掉
}
if (isThere)
{
continue;
}
else
{
if (UrlThereare(strA[i], this.dttiebainfo, dtTieBaInfo, true) != 0) {
//isThere = true;
continue;
}
}
Thread.Sleep(2000);
//得到此链接的源码
webInfo = HtmlUtil.getHtml(strA[i], "");
if (webInfo.Length == 0) { continue; }
try
{
//流水+1
newsInfo.Uid = this.dvAll.Rows.Count + 1;
//标题
//string[] strT = HtmlUtil.GetElementsByTagName(webInfo, "title");
//if (strT.Length == 0)
//{
// continue;
//}
//else
//{
// newsInfo.Title = HtmlUtil.NoHTML(HtmlUtil.GetElementsByTagName(webInfo, "title")[0]);
//}
//得到正文,以P标签来区分
string[] strContext = HtmlUtil.GetElementsByClass(webInfo, "tieba");
//string[] strContext = HtmlUtil.GetElementsByTagName(webInfo, "post_content");
newsInfo.Contexts = "";
for (int j = 0; j < strContext.Length; j++)
{
//循环累加正文信息
newsInfo.Contexts += HtmlUtil.NoHTML(strContext[j]);
}
//如果正文信息为空,那么将无法做关键字对照,此条数据舍弃
if (newsInfo.Contexts.Length == 0)
{
continue;
}
//网站链接
newsInfo.InfoSource = strA[i].Trim();
//关键字的设置
//newsInfo.KeyWords = "";
//Dictionary<string, int> events = new Dictionary<string, int>();
//for (int j = 0; j < dtkey.Rows.Count; j++)
//{
// Application.DoEvents();
// if (!events.ContainsKey(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString()))
// {
// events.Add(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString(), 1);
// }
// if (!newsInfo.Contexts.Contains(dtkey.Rows[j][4].ToString()))
// {
// if (events.ContainsKey(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString()) && events[dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString()] == 1)
// {
// events[dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString()] = 0;
// }
// }
//}
if (newsInfo.KeyWords == null || newsInfo.KeyWords.Length == 0)
{
for (int j = 0; j < dtkey.Rows.Count; j++)
{
Application.DoEvents();
string[] keys = dtkey.Rows[j][4].ToString().Split(new char[] { ' ' });
if (!events.ContainsKey(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString()))
{
events.Add(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString(), 1);
foreach (string k in keys)
{
if (!newsInfo.Contexts.ToLower().Contains(k.ToLower()))
{
events.Remove(dtkey.Rows[j][1].ToString() + "-" + dtkey.Rows[j][6].ToString());
break;
}
}
}
}
foreach (KeyValuePair<string, int> ev in events)
{
if (ev.Value == 1)
{
newsInfo.KeyWords += "," + ev.Key.Split(new char[] { '-' })[0] + "-" + int.Parse(ev.Key.Split(new char[] { '-' })[1]);
}
}
}
if (newsInfo.KeyWords.Length == 0) { continue; }
newsInfo.KeyWords = newsInfo.KeyWords.Substring(1);
//收集日期
newsInfo.CollectDate = string.Format(DateTime.Now.ToString("yyyy-MM-dd HH:mm:ss"));
//发布人和发布日期暂时无法取到,手工赋值为空
newsInfo.ReleaseDate = "";
newsInfo.ReleaseName = "";
//网页快照,这里为用户指定生成,如果未选择生成,那么为空
newsInfo.Snapshot = "";
newsInfo.Sheng = dtXs.Rows[xs]["sheng"] == null ? "" : dtXs.Rows[xs]["sheng"].ToString();
newsInfo.Shi = dtXs.Rows[xs]["shi"] == null ? "" : dtXs.Rows[xs]["shi"].ToString();
newsInfo.Xian = dtXs.Rows[xs]["xian"] == null ? "" : dtXs.Rows[xs]["xian"].ToString();
//网站名
newsInfo.WebName = dtXs.Rows[xs]["Name"].ToString();
//pid
newsInfo.Pid = 5;
//part正负判断
newsInfo.Part = GetParts(newsInfo.Contexts);
//reposts
newsInfo.Reposts = 0;
//comments
newsInfo.Comments = 0;
//新建数据行
DataRow dr = dttiebainfo.NewRow();
if (dvtieba.RowCount == 0)
{
dr[0] = 1;
}
else
{
dr[0] = int.Parse(dvtieba.Rows[dvtieba.RowCount - 1].Cells[0].Value.ToString()) + 1;
}
//dr[0] = newsInfo.Uid;
dr[1] = newsInfo.Title;
dr[2] = newsInfo.Contexts;
dr[3] = newsInfo.ReleaseDate;
dr[4] = newsInfo.InfoSource;
dr[5] = newsInfo.KeyWords.Substring(0, newsInfo.KeyWords.IndexOf("-"));
dr[6] = newsInfo.ReleaseName;
dr[7] = newsInfo.CollectDate;
dr[8] = newsInfo.Snapshot;
dr[9] = newsInfo.WebName;
dr[10] = newsInfo.Pid;
dr[11] = newsInfo.Part;
dr[12] = newsInfo.Reposts;
dr[13] = newsInfo.Comments;
//把行加到DT中
dttiebainfo.Rows.InsertAt(dr, 0);
//数据源刷新
if (dttiebainfo.Rows.Count >= 500)
{
dttiebainfo.Rows.RemoveAt(500);
}
this.BeginInvoke(new MethodInvoker(delegate()
{
dvtieba.Refresh();
}));
}
catch (Exception ex)
{
Comm.WriteErrorLog(ex.StackTrace);
}
////总表刷新
//dt.Rows.Add(dr);
//dvAll.Refresh();
//得到插入语句
try
{
if (isThere)
{
continue;
}
else
{
sb.Append(ri.GetInsString(newsInfo) + ";");
}
//每10次执行一次插入数据库
if (sb.ToString().Length != 0)
{
if (i % 10 == 0)
{
filterStr = sb.ToString();
filterStr = filterStr.Replace("[ ", "[");
filterStr = filterStr.Replace(" ]", "]");
//执行插入
cmd.ExecuteNonQuery(filterStr);
//清除插入字段串
sb.Clear();
filterStr = "";
}
}
}
catch (Exception ex)
{
Comm.WriteErrorLog(ex.StackTrace);
}
}
}
#endregion
}
#endregion
try
{
if (sb.ToString().Length != 0)
{
filterStr = sb.ToString();
filterStr = filterStr.Replace("[ ", "[");
filterStr = filterStr.Replace(" ]", "]");
//执行插入
cmd.ExecuteNonQuery(filterStr);
//清除插入字段串
sb.Clear();
filterStr = "";
}
}
catch (Exception ex)
{
Comm.WriteErrorLog(ex.StackTrace);
}
//执行完毕后,重新获取一次数据库的数据
dtTieBaInfo = tri.SelReleaseInfo(DateTime.Now.ToString("yyyy-MM-dd HH:mm:ss"), DateTime.Now.AddDays(-15).ToString("yyyy-MM-dd HH:mm:ss"), "5");
this.BeginInvoke(new MethodInvoker(delegate()
{
lbtieba.Text = "一轮搜索完毕！";
lbtieba.ForeColor = Color.Red;
}));
}
/// <summary>
/// 得到网站的新闻类数据
/// </summary>
private void GetWebNewsInfo()
{
lbWeb.Text = "";
lbWeb.Visible = true;
//相似链接
string Similar = "";
DataBaseServer.MySqlCmd cmd = new MySqlCmd();
//得到相似表
DataTable dtXs = new DataTable();
dtXs = cmd.GetTabel("Select * from WebAddress WHERE pid=0");
dtParts = cmd.GetTabel("SELECT * FROM partword");
DataTable dtkey = new DataTable();
dtkey = cmd.GetTabel("select * from Keywords");
//相似表中的被抓取网址
string webInfo = "";
//要过滤链接中首页的正则
string strTopFormat = "http://.+/";
List<string> strTop = new List<string>();
sb = new StringBuilder();
sb.Append("");
string filterStr = "";
#region 读取相似度表中的数据据,循环抓取
for (int xs = 0; xs < dtXs.Rows.Count; xs++)
{
lbWeb.ForeColor = Color.DarkBlue;
lbWeb.Text = "正在搜索:" + dtXs.Rows[xs]["name"].ToString();
//读取相似表中要抓取的网址
webInfo = getHtml(dtXs.Rows[xs]["url"].ToString(), "");
//读取相似链接
Similar = dtXs.Rows[xs]["likeurl"].ToString();
//取出
//string[] strA = HtmlUtil.GetElementsByTagName(webInfo, "a");
List<string> strList = HtmlUtil.GetElementsByTagNameList(webInfo, "a");
string strURLformat = "http://.[^\"]+";
TbReleaseInfo ri = new TbReleaseInfo();
string[] strA = GetLIstDate(strList.Distinct());
#region 逐个链接判断
//循环时判断是否要验证
bool isThere = false;
for (int i = 0; i < strA.Length; i++)
{
if (Program.ProClose == true) break;
Application.DoEvents();
try
{
//得到目标网址中的所有链接,如果未得到,那么就继续读取下一个
strA[i] = HtmlUtil.GetListByHtml(dtXs.Rows[xs]["url"].ToString(), strA[i], strURLformat)[0];
//处理含有单引号的链接
strA[i] = UrlCl(strA[i]);
//处理单引号的链接
if (strA[i].IndexOf("'") != -1)
{
strA[i] = GetstringByHtmlArray(strA[i], "http://.[^\']+");
}
}
catch (Exception)
{
continue;
}
//得到相似值,大于0.70的认为相同,并开始抓取
if (HtmlUtil.getSimilarDegree(Similar, strA[i]) >= 0.60)
{
//判断这个链接是否已经在库中或者列表中,如果存在,此次就不再执行
strTop = HtmlUtil.GetListByHtmlArray(strA[i], strTopFormat);
if (strTop.Count != 0)
{
//if (strTop[0] == "http://blog.sohu.com/")
continue;//同新闻,如果将首页去掉
}
if (isThere)
{
continue;
}
else
{
//if (strA[i] == "http://news.ifeng.com/mainland/detail_2013_10/18/30459577_0.shtml'>[详细]</a>")
//{
// strA[i] = strA[i];
//}
if (UrlThereare(strA[i], this.dtnewsinfo, dtWebNewsInfo, true) != 0) { isThere = true; continue; }
}
//得到此链接的源码
webInfo = getHtml(strA[i], "");
if (webInfo.Length == 0) { continue; }
//创建数据对象
ModelReleaseInfo newsInfo = new ModelReleaseInfo();
try
{
//流水+1
newsInfo.Uid = this.dvAll.Rows.Count + 1;
//标题
string[] strT = HtmlUtil.GetElementsByTagName(webInfo, "title");
if (strT.Length == 0)
{
continue;
}
else
{
newsInfo.Title = HtmlUtil.NoHTML(HtmlUtil.GetElementsByTagName(webInfo, "title")[0]);
}
//得到正文,以P标签来区分
string[] strContext = HtmlUtil.GetElementsByTagName(webInfo, "p");
newsInfo.Contexts = "";
for (int j = 0; j < strContext.Length; j++)
{
//循环累加正文信息
newsInfo.Contexts += HtmlUtil.NoHTML(strContext[j]);
}
//如果正文信息为空,那么将无法做关键字对照,此条数据舍弃
if (newsInfo.Contexts.Length == 0)
{
continue;
}
//网站链接
newsInfo.InfoSource = strA[i].Trim();
//关键字的设置
newsInfo.KeyWords = "";
for (int j = 0; j < dtkey.Rows.Count; j++)
{
Application.DoEvents();
if (newsInfo.Contexts.IndexOf(dtkey.Rows[j][1].ToString()) > 0)
{ newsInfo.KeyWords += dtkey.Rows[j][1].ToString() + ","; }
else
{
}
}
if (newsInfo.KeyWords.Length == 0) { continue; }
newsInfo.KeyWords = newsInfo.KeyWords.Substring(0, newsInfo.KeyWords.Length - 1);
//收集日期
newsInfo.CollectDate = string.Format(DateTime.Now.ToString("yyyy-MM-dd HH:mm:ss"));
//发布人和发布日期暂时无法取到,手工赋值为空
newsInfo.ReleaseDate = "";
newsInfo.ReleaseName = "";
//网页快照,这里为用户指定生成,如果未选择生成,那么为空
newsInfo.Snapshot = "";
//网站名
newsInfo.WebName = dtXs.Rows[xs]["Name"].ToString();
//pid
newsInfo.Pid = 0;
//part正负判断
newsInfo.Part = GetParts(newsInfo.Contexts);
//reposts
newsInfo.Reposts = 0;
//comments
newsInfo.Comments = 0;
//新建数据行
DataRow dr = dtnewsinfo.NewRow();
if (dvWeb.RowCount == 0)
{
dr[0] = 1;
}
else
{
dr[0] = int.Parse(dvWeb.Rows[dvWeb.RowCount - 1].Cells[0].Value.ToString()) + 1;
}
//dr[0] = newsInfo.Uid;
dr[1] = newsInfo.Title;
dr[2] = newsInfo.Contexts;
dr[3] = newsInfo.ReleaseDate;
dr[4] = newsInfo.InfoSource;
dr[5] = newsInfo.KeyWords;
dr[6] = newsInfo.ReleaseName;
dr[7] = newsInfo.CollectDate;
dr[8] = newsInfo.Snapshot;
dr[9] = newsInfo.WebName;
dr[10] = newsInfo.Pid;
dr[11] = newsInfo.Part;
dr[12] = newsInfo.Reposts;
dr[13] = newsInfo.Comments;
//把行加到DT中
dtnewsinfo.Rows.InsertAt(dr, 0);
//数据源刷新
if (dtnewsinfo.Rows.Count >= 500)
{
dtnewsinfo.Rows.RemoveAt(500);
}
dvWeb.Refresh();
}
catch (Exception ex)
{
StreamWriter sw = File.AppendText("log.txt");
sw.WriteLine(DateTime.Now.ToLongDateString());
sw.WriteLine("begin");
sw.WriteLine(ex.Message);
sw.WriteLine(sb.ToString());
sw.WriteLine("end");
sw.WriteLine("");
sw.Close();
}
////总表刷新
//dt.Rows.Add(dr);
//dvAll.Refresh();
//得到插入语句
try
{
if (isThere)
{
continue;
}
else
{
sb.Append(ri.GetInsString(newsInfo) + ";");
}
//每10次执行一次插入数据库
if (sb.ToString().Length != 0)
{
if (i % 10 == 0)
{
filterStr = sb.ToString();
filterStr = filterStr.Replace("[ ", "[");
filterStr = filterStr.Replace(" ]", "]");
//执行插入
cmd.ExecuteNonQuery(filterStr);
//清除插入字段串
sb.Clear();
filterStr = "";
}
}
}
catch (Exception ex)
{
StreamWriter sw = File.AppendText("log.txt");
sw.WriteLine(DateTime.Now.ToLongDateString());
sw.WriteLine("begin");
sw.WriteLine(ex.Message);
sw.WriteLine(sb.ToString());
sw.WriteLine("end");
sw.WriteLine("");
sw.Close();
}
}
}
#endregion
}
#endregion
try
{
if (sb.ToString().Length != 0)
{
filterStr = sb.ToString();
filterStr = filterStr.Replace("[ ", "[");
filterStr = filterStr.Replace(" ]", "]");
//执行插入
cmd.ExecuteNonQuery(filterStr);
//清除插入字段串
sb.Clear();
filterStr = "";
}
}
catch (Exception ex)
{
StreamWriter sw = File.AppendText("log.txt");
sw.WriteLine(DateTime.Now.ToLongDateString());
sw.WriteLine("begin");
sw.WriteLine(ex.Message);
sw.WriteLine(sb.ToString());
sw.WriteLine("end");
sw.WriteLine("");
sw.Close();
}
//执行完毕后,重新获取一次数据库的数据
dtWebNewsInfo = tri.SelReleaseInfo(DateTime.Now.ToString("yyyy-MM-dd HH:mm:ss"), DateTime.Now.AddDays(-15).ToString("yyyy-MM-dd HH:mm:ss"), "0 AND webName<>'百度'");
//MessageBox.Show("ok");
lbWeb.Text = "一轮搜索完毕！";
lbWeb.ForeColor = Color.Red;
}
