Download a file through the WebBrowser control

一曲冷凌霜 提交于 2019-11-27 08:46:17

Add a SaveFileDialog control to your form, then add the following code on your WebBrowser's Navigating event:

private void webBrowser1_Navigating(object sender, WebBrowserNavigatingEventArgs e)
{
    if (e.Url.Segments[e.Url.Segments.Length - 1].EndsWith(".pdf"))
    {
        e.Cancel = true;
        string filepath = null;

        saveFileDialog1.FileName = e.Url.Segments[e.Url.Segments.Length - 1];
        if (saveFileDialog1.ShowDialog() == DialogResult.OK)
        {
            filepath = saveFileDialog1.FileName;
            WebClient client = new WebClient();
            client.DownloadFileCompleted += new AsyncCompletedEventHandler(client_DownloadFileCompleted);
            client.DownloadFileAsync(e.Url, filepath);
        }
    }
}

//Callback function

void client_DownloadFileCompleted(object sender, AsyncCompletedEventArgs e)
{
    MessageBox.Show("File downloaded");
}

Source: http://social.msdn.microsoft.com/Forums/en-US/csharpgeneral/thread/d338a2c8-96df-4cb0-b8be-c5fbdd7c9202

The solution I ended up using:

I did everything else as-needed to get the URL where it needed to go. Knowing that all of the login information, required settings, viewstates, etc. were stored in the cookies, I was finally able to grab the file using a hybrid of the web control to navigate then the WebClient object to actually snag the file bytes.

    public byte[] GetPDF(string keyValue)
    {
        DoLogin();

        // Ask the source to generate the PDF. The PDF doesn't 
        // exist on the server until you have visited this page 
        // at least ONCE. The PDF exists for five minutes after 
        // the visit, so you have to snag it pretty quick.
        LoadUrl(string.Format(
            "https://www.theMagicSource.com/getimage.do?&key={0}&imageoutputformat=PDF", 
            keyValue));

        // Now that we're logged in (not shown here), and 
        // (hopefully) at the right location, snag the cookies. 
        // We can use them to download the PDF directly.
        string cookies = GetCookies();

        byte[] fileBytes = null;
        try
        {
            // We are fully logged in, and by now, the PDF should
            // be generated. GO GET IT! 
            WebClient wc = new WebClient();
            wc.Headers.Add("Cookie: " + cookies);

            string tmpFile = Path.GetTempFileName();

            wc.DownloadFile(string.Format(
                "https://www.theMagicSource.com/document?id={0}_final.PDF", 
                keyValue), tmpFile);

            fileBytes = File.ReadAllBytes(tmpFile);
            File.Delete(tmpFile);
        }
        catch (Exception ex)
        {
            // If we can't get the PDF here, then just ignore the error and return null.
            throw new WebScrapePDFException(
                "Could not find the specified file.", ex);
        }

        return fileBytes;
    }

    private void LoadUrl(string url)
    {
        InternalBrowser.Navigate(url);

        // Let the browser control do what it needs to do to start 
        // processing the page.
        Thread.Sleep(100);

        // If EITHER we can't continue OR
        // the web browser has not been idle for 10 consecutive seconds yet, 
        // then wait some more.
        // ...
        // ... Some stuff here to make sure the page is fully loaded and ready. 
        // ... Removed to reduce complexity, but you get the idea.
        // ...
    }

    private string GetCookies()
    {
        if (InternalBrowser.InvokeRequired)
        {
            return (string)InternalBrowser.Invoke(new Func<string>(() => GetCookies()));
        }
        else
        {
            return InternalBrowser.Document.Cookie;
        }
    }  
    bool documentCompleted = false;
    string getInnerText(string url)
    {
        documentCompleted = false;
        web.Navigate(url);

        while (!documentCompleted)         
            Application.DoEvents();


        return web.Document.Body.InnerText;
    }
    private void web_DocumentCompleted(object sender, WebBrowserDocumentCompletedEventArgs e)
    {
        documentCompleted = true;
    }
易学教程内所有资源均来自网络或用户发布的内容,如有违反法律规定的内容欢迎反馈
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!