public IHTMLElementCollection GetElementCollection(IHTMLElementCollection elements)
{
if (elements == null) return null;
if (TagName == null) return elements;
return (IHTMLElementCollection) elements.tags(TagName);
}
public IHTMLElementCollection GetElementCollection(IHTMLElementCollection elements)
{
if (elements == null)
{
return(null);
}
if (TagName == null)
{
return(elements);
}
return((IHTMLElementCollection)elements.tags(TagName));
}
/// <summary>
/// Helper function to build up the list of form field elements
/// </summary>
private void verifyFieldElements(IHTMLDocument2 document)
{
if (fieldElements == null)
{
fieldElements = new ArrayList();
IHTMLElementCollection allElements = document.all;
IHTMLElementCollection inputElements = (IHTMLElementCollection)allElements.tags(ElementsSupport.InputTagName);
IHTMLElementCollection textareaElements = (IHTMLElementCollection)allElements.tags("textarea");
// Merge the two collections
for (int i = 0; i < inputElements.length; i++)
{
IHTMLElement node = (IHTMLElement)inputElements.item(i, null);
fieldElements.Add(node);
}
for (int i = 0; i < textareaElements.length; i++)
{
IHTMLElement node = (IHTMLElement)textareaElements.item(i, null);
fieldElements.Add(node);
}
}
}
/// <summary>
/// 利用mshtml进行分析
/// </summary>
/// <param name="html"></param>
/// <param name="tag"></param>
public string ParserHtml(string html, string tag)
{
// 首先html代码內容存入HTMLDocumentClass
IHTMLDocument2 document = new HTMLDocumentClass();
document.write(new object[] { html });
document.close();
// 属性body则是html中的body tag
// 而body本身就是一个IHTMLElement
// 所以可以用all这个属性将所有元素取出成为一个collection
IHTMLElementCollection body = (IHTMLElementCollection)document.body.all;
// 可以用tags这个方法过滤出我们所需要的tag
IHTMLElementCollection elements = (IHTMLElementCollection)body.tags("TABLE");
for (int i = 0; i < elements.length; i++)
{
IHTMLElement tr = (IHTMLElement)elements.item(i, null);
//result += element.innerHTML;
}
string result = "";
for (int i = 0; i < elements.length; i++)
{
// 使用item这个方法可以将集合中的元素取出
// 第一个参数代表的是顺序，但是在msdn中表示为name
// 第二个参数msdn中表示为index,但经过测试后,指的并不是顺序,所以目前无法确定它的用途
// 如果有知道的朋友，也请跟我说一下
IHTMLElement element = (IHTMLElement)elements.item(i, null);
if (string.IsNullOrEmpty(element.innerHTML))
{
continue;
}
result += element.innerHTML;
}
return(result);
}
private void Vpn_PageData()
{
log.Info($"Entered Vpn_PageData.");
log.Info($" Currently on page {current_page}/{total_pages}.");
#region write in sql table
/// 取得gvList
HTMLDocument d = (HTMLDocument)m.vpnweb.Document;
IHTMLElement gvDownLoad = d.getElementById(DOM_FOR_ACTUAL_DATA);
log.Info($"[{DOM_FOR_ACTUAL_DATA}] is read.");
/// 讀取
/// 20200503 我發現Html Agility Pack不能click
/// 只好第一層是ihtmlelement
IHTMLElementCollection trs_ = gvDownLoad.all;
IHTMLElementCollection trs = trs_.tags("tr");
DateTime current_time = DateTime.Now;
// 使用item這個方法可以將集合中的元素取出
// 第一個參數代表的是順序,但是在msdn中標示為name
// 第二個參數msdn中標示為index,但經過測試後,指的並不是順序,所以目前無法確定他的用途
// 如果有知道的朋友,也請跟我說一下
// current_line = 0 是標題行
/// 20200503 我發現Html Agility Pack不能click
/// 只好第一層是ihtmlelement
VPN_files.Clear();
for (current_line = 1; current_line < trs.length; current_line++)
{
IHTMLElement tr = trs.item(current_line, null);
HtmlDocument h_ = new HtmlDocument();
h_.LoadHtml(tr.innerHTML);
HtmlNodeCollection tds = h_.DocumentNode.SelectNodes("//td");
tbl_download result = _td_parcer(tds);
using (NHIDataContext dc = new NHIDataContext())
{
var q = from p in dc.tbl_download
where (p.f_name == result.f_name) && (p.SDATE == result.SDATE)
select p;
if (q.Count() == 0)
{
result.QDATE = current_time;
dc.tbl_download.InsertOnSubmit(result);
dc.SubmitChanges();
log.Info($" [{result.f_name}] added to SQL server");
}
}
}
#endregion write in sql table
#region download files
// making queue_files
foreach (KeyValuePair <int, string> v in VPN_files)
{
if (!Local_files.Contains(v.Value))
{
queue_files.Enqueue(v.Key);
log.Info($" {v.Key}: {v.Value} enqueued.");
}
}
if ((queue_files.Count == 0) && (current_page == total_pages))
{
// 這頁讀完, 且所有頁都讀完了.
m.Refresh_Table();
tb.ShowBalloonTip("結束", "完成所有頁面讀取", BalloonIcon.Info);
}
else if (queue_files.Count == 0)
{
log.Info($" Nothing enqueued on page {current_page}/{total_pages}");
Goto_next_page();
}
else
{
// 有東西才需要執行
tb.ShowBalloonTip("計時器開始", $"一共{queue_files.Count}個檔案要下載", BalloonIcon.Info);
// initialization
current_line = 0;
current_line = queue_files.Dequeue();
// execution
this._timer1 = new System.Timers.Timer
{
Interval = 6000
};
this._timer1.Elapsed += new System.Timers.ElapsedEventHandler(TimersTimer_Elapsed);
log.Info($" _timer1 started.");
this._timer1.Start();
}
#endregion download files
log.Info("Exited Vpn_PageData.");
}
private void TimersTimer_Elapsed(object sender, ElapsedEventArgs e)
{
log.Info($" Entered TimersTimer_Elapsed.");
m.Dispatcher.Invoke((Action)(() =>
{
HTMLDocument d = (HTMLDocument)m.vpnweb.Document;
// dispatcher的問題, 要叫用就要在這裡面
if (GOTO_NEXT_PAGE)
{
GOTO_NEXT_PAGE = false;
log.Info($" _timer1 stopped. this pages finished.");
this._timer1.Stop();
Goto_next_page();
log.Info($" Exited TimersTimer_Elapsed.");
return;
}
log.Info($" Now dealing with page {current_page}/{total_pages}.");
log.Info($" Now dealing with line {current_line}.");
IHTMLElement gvDownLoad = d.getElementById(DOM_FOR_ACTUAL_DATA);
IHTMLElementCollection trs_ = gvDownLoad.all;
IHTMLElementCollection trs = trs_.tags("tr");
IHTMLElement tr = trs.item(current_line, null);
IHTMLElement a = tr.children[4].children[0];
a.click();
log.Info($" button click: {a.innerHTML}");
System.Threading.ThreadStart th_begin = new System.Threading.ThreadStart(Work_todo);
System.Threading.Thread thr = new System.Threading.Thread(th_begin)
{
IsBackground = true,
Name = "PressS"
};
thr.Start();
// 判斷是否這一頁讀完了? 是否最後一頁了?
if ((queue_files.Count == 0) && (current_page == total_pages))
{
// 這頁讀完, 且所有頁都讀完了.
log.Info($" _timer1 stopped. all pages finished.");
m.Refresh_Table();
tb.ShowBalloonTip("結束", "完成所有頁面讀取", BalloonIcon.Info);
this._timer1.Stop();
}
else if (queue_files.Count == 0)
{
// 這頁讀完, 但還有下一頁.
GOTO_NEXT_PAGE = true;
}
else
{
current_line = queue_files.Dequeue();
log.Info($" go to next line: {current_line}.");
}
}));
log.Info($" Exited TimersTimer_Elapsed.");
return;
}
/// <summary>
/// FixHTML
/// </summary>
/// <param name="htmlToParse"></param>
/// <returns></returns>
string FixHTML(string htmlToParse)
{
try
{
//::......... Declare a new HTML document to use, and write our normal HTML
IHTMLDocument2 htmlDocument = new HTMLDocumentClass();
htmlDocument.write(htmlToParse);
htmlDocument.close();
//::......... With this we retrieve all of the HTML elements collection existing on out HTML block
IHTMLElementCollection allElements = (IHTMLElementCollection)htmlDocument.body.all;
IHTMLElementCollection allInputs = (IHTMLElementCollection)allElements.tags("img");
IHTMLElementCollection allStyles = (IHTMLElementCollection)htmlDocument.all.tags("style");
IHTMLElementCollection allhead = (IHTMLElementCollection)htmlDocument.all.tags("head");
string MyDoc = "<html>";
foreach (IHTMLElement HeadElement in allhead)
{
MyDoc += HeadElement.outerHTML;
break;
}
//::......... Change some properties
foreach (IHTMLElement element in allInputs)
{
string src = (string)element.getAttribute("src", 0);
int Pos = -1;
if ((Pos = src.IndexOf("_files/")) != -1)
{
string URL = GetSecurePDF();
string u = Request.Url.Authority;
if (u.ToLower() == "mytravelplans.com.sg")
{
URL += u;
}
else if (u.ToLower() == "mytravelplans.com.au")
{
URL += u;
}
else if (u.ToLower() == "mytravelplans.co.in")
{
URL += u;
}
else
{
URL += u;
}
URL += GetWebDirectory() + src.Substring(Pos + 1);
element.setAttribute("src", URL, 0);
}
}
//::......... Return the parent element content ( BODY > HTML )
MyDoc += htmlDocument.body.outerHTML;
MyDoc += "</html>";
return(MyDoc);
}
catch (Exception e)
{
logger.Error("Exception in HTML fix: " + e.Message);
}
return(htmlToParse);
}