1

i want to make a console program to monitor a webpage's htmlsourcecode, because some of the page content are created by some javescript, so i have to use webbrowser control. like : View Generated Source (After AJAX/JavaScript) in C#

my code is below:

public class WebProcessor
{
    public string GeneratedSource;
    public string URL ;

    public DateTime beginTime;
    public DateTime endTime;

    public object GetGeneratedHTML(object url)
    {
        URL = url.ToString();
        try 
        {
            Thread[] t = new Thread[10];

            for (int i = 0; i < 10; i++) 
            {
                t[i] = new Thread(new ThreadStart(WebBrowserThread));
                t[i].SetApartmentState(ApartmentState.STA);
                t[i].Name = "Thread" + i.ToString();
                t[i].Start();
                //t[i].Join();          
            }

        } 
        catch (Exception ex) 
        {
            Console.WriteLine(ex.ToString());
        }

        return GeneratedSource;
    }

    private void WebBrowserThread()
    {
        WebBrowser wb = new WebBrowser();
        wb.ScriptErrorsSuppressed = true;

        wb.DocumentCompleted += 
            new WebBrowserDocumentCompletedEventHandler(
                wb_DocumentCompleted);

        while(true )
        {
            beginTime = DateTime.Now;

            wb.Navigate(URL);

            while (wb.ReadyState != WebBrowserReadyState.Complete)
            {
                Application.DoEvents();
            Thread.Sleep(new Random().Next(10,100));
            }
        }               

        //wb.Dispose();                 
    }

    private void wb_DocumentCompleted(object sender, 
        WebBrowserDocumentCompletedEventArgs e)
    {
        WebBrowser wb = (WebBrowser)sender;
        if (wb.ReadyState == WebBrowserReadyState.Complete)
        {
            GeneratedSource= wb.Document.Body.InnerHtml;

            endTime = DateTime.Now;
            Console.WriteLine("WebBrowser " + (endTime-beginTime).Milliseconds + Thread.CurrentThread.Name + wb.Document.Title);            
        }
    } 
}

when it run, after a while (20-50 times), it throw the exception like this

----------------------------------------------------------------------------
EXCEPTION code=ACCESS_VIOLATION
(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(nul
l)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(n
ull)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)
(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(null)(nul
l)(null)BACKTRACE: 33 stack frames:
 #0 0x083dba8db0 at MatchExactGetIDsOfNames in mshtml.dll
 #1 0x0879f9b837 at StrongNameErrorInfo in mscorwks.dll
 #2 0x0879f9b8e3 at StrongNameErrorInfo in mscorwks.dll
 #3 0x0879f9b93a at StrongNameErrorInfo in mscorwks.dll
 #4 0x0879f9b9e0 at StrongNameErrorInfo in mscorwks.dll
 #5 0x0879f9b677 at StrongNameErrorInfo in mscorwks.dll
 #6 0x0879f9b785 at StrongNameErrorInfo in mscorwks.dll
 #7 0x0879f192a8 at InstallCustomModule in mscorwks.dll
 #8 0x0879f19444 at InstallCustomModule in mscorwks.dll
 #9 0x0879f194ab at InstallCustomModule in mscorwks.dll
 #10 0x0879fa6491 at StrongNameErrorInfo in mscorwks.dll
 #11 0x0879f44bcf at DllGetClassObjectInternal in mscorwks.dll
 #12 0x089bbafa at  in
 #13 0x087b18cc10 at  in System.Windows.Forms.ni.dll
 #14 0x087b91f4c1 at  in System.Windows.Forms.ni.dll
 #15 0x08d00669 at  in
 #16 0x08792d6e46 at  in mscorlib.ni.dll
 #17 0x08792e02cf at  in mscorlib.ni.dll
 #18 0x08792d6dc4 at  in mscorlib.ni.dll
 #19 0x0879e71b4c at  in mscorwks.dll
 #20 0x0879e896ce at  in mscorwks.dll
 #21 0x0879e96ea9 at CoUninitializeEE in mscorwks.dll
 #22 0x0879e96edc at CoUninitializeEE in mscorwks.dll
 #23 0x0879e96efa at CoUninitializeEE in mscorwks.dll
 #24 0x0879f88357 at GetPrivateContextsPerfCounters in mscorwks.dll
 #25 0x0879e9cc8f at CoUninitializeEE in mscorwks.dll
 #26 0x0879e9cc2b at CoUninitializeEE in mscorwks.dll
 #27 0x0879e9cb51 at CoUninitializeEE in mscorwks.dll
 #28 0x0879e9ccdd at CoUninitializeEE in mscorwks.dll
 #29 0x0879f88128 at GetPrivateContextsPerfCounters in mscorwks.dll
 #30 0x0879f88202 at GetPrivateContextsPerfCounters in mscorwks.dll
 #31 0x0879f0e255 at InstallCustomModule in mscorwks.dll
 #32 0x087c80b729 at GetModuleFileNameA in KERNEL32.dll
----------------------------------------------------------------------------

i have try lots of methods to solve the problem, finally, i found that if i thread sleep more millseconds, it will run for a longer time, but the exception is still throw.

hope somebody give me the answer of how to slove the problem ... thanks very much !!!

Community
  • 1
  • 1

1 Answers1

0

Where do you get the exception ( I mean line number). I suppose it is on the call wb.ReadyState != WebBrowserReadyState.Complete as you are retrieving some data from a control on a Thread that is not the same Thread which created the control instance. If this is the case, then you have to call the property from the right thread.

I would change you while loop into this:

System.Windows.Forms.WebBrowserReadyState state = GetState(Program.BrowsingSystem.BrowserLink);
        while (state != System.Windows.Forms.WebBrowserReadyState.Complete)
        {
            Application.DoEvents();
            Thread.Sleep(new Random().Next(10, 100));
        }

associating it with the following called methods:

private System.Windows.Forms.WebBrowserReadyState GetState(System.Windows.Forms.WebBrowser instance)
    {
        if (instance.InvokeRequired)
        {
            var wbFunc = new Func<System.Windows.Forms.WebBrowser, System.Windows.Forms.WebBrowserReadyState>(RetrieveState);
            IAsyncResult FunFuncResult = wbFunc.BeginInvoke(instance, null, null);
            return wbFunc.EndInvoke(FunFuncResult);

        }
        else
        {
            return RetrieveState(instance);
        }
    }

    public System.Windows.Forms.WebBrowserReadyState RetrieveState(System.Windows.Forms.WebBrowser instance)
    {
        return instance.ReadyState;
    }

I cannot test it as I'm missing the rest of the app, but it should be close to final solution.

Pinoba
  • 66
  • 5