C# parsing web site with ajax loaded content

Question

If I recive a web site with this function I get the whole page, but without the ajax loaded values.

htmlDoc.LoadHtml(new WebClient().DownloadString(url));

Is it possible to load the web site like in gChrome with all values?

Answer 1

You can use a WebBrowser control to get and render the page. Unfortunately, the control uses Internet Explorer and you have to change a registry value in order to force it to use the latest version and even then the implementation is very brittle.

Another option is to take a standalone browser engine like WebKit and make it work in .NET. I found a page explaining how to do this, but it's pretty dated: http://webkitdotnet.sourceforge.net/basics.php

I worked on a little demo app to get the content and this is what I came up with:

    class Program
    {
        static void Main(string[] args)
        {
            GetRenderedWebPage("https://siderite.dev", TimeSpan.FromSeconds(5), output =>
            {
                Console.Write(output);
                File.WriteAllText("output.txt", output);
            });
            Console.ReadKey();
        }

        private static void GetRenderedWebPage(string url, TimeSpan waitAfterPageLoad, Action<string> callBack)
        {
            const string cEndLine= "All output received";

            var sb = new StringBuilder();
            var p = new PhantomJS();
            p.OutputReceived += (sender, e) =>
            {
                if (e.Data==cEndLine)
                {
                    callBack(sb.ToString());
                } else
                {
                    sb.AppendLine(e.Data);
                }
            };
            p.RunScript(@"
var page = require('webpage').create();
page.viewportSize = { width: 1920, height: 1080 };
page.onLoadFinished = function(status) {
    if (status=='success') {
        setTimeout(function() {
            console.log(page.content);
            console.log('" + cEndLine + @"');
            phantom.exit();
        }," + waitAfterPageLoad.TotalMilliseconds + @");
    }
};
var url = '" + url + @"';
page.open(url);", new string[0]);
        }
    }

This uses the PhantomJS "headless" browser by way of the wrapper NReco.PhantomJS which you can get through "reference NuGet package" directly from Visual Studio. I am sure it can be done better, but this is what I did today. You might want to take a look at the PhantomJS callbacks so you can properly debug what is going on. My example will wait forever if the URL doesn't work, for example. Here is a useful link: https://newspaint.wordpress.com/2013/04/25/getting-to-the-bottom-of-why-a-phantomjs-page-load-fails/