Parsing JavaScript Web Page in C# with AngleSharp

Question

The webpage use javascript to build its html so I need html parser with js support.
I found anglesharp but I can't make it working.

using AngleSharp;
using System;
using System.Collections.Generic;
using System.Linq;
using System.Net;
using System.Text;
using System.Threading.Tasks;

namespace AngleSharpScraping
{
    class Program
    {
        static void Main(string[] args)
        {
            GetMkvToolNix();
            Console.ReadKey();
        }

        static async void GetMkvToolNix()
        {
            // Create a new configuration with javascript interpreter.
            var config = new Configuration().WithJavaScript();

            // Parsing process.
            var document = await BrowsingContext.New(config).OpenAsync(Url.Create("http://www.fosshub.com/MKVToolNix.html"));
            var link = document.QuerySelector("body > div.container.page-content > div > div.col-sm-9 > article > div.main-dl-box > p:nth-child(2) > a.dwl-link.xlink").GetAttribute("data");

            Console.WriteLine(link);
        }
    }
}

Answer 1

AngleSharp is a text parser. If you want to scrape dynamic web pages with JS, you'll need a headless browser.

This answer provides a couple of options (at least one free and open source: WebKit.NET).

Answer 2

AngleSharp alone only provides an HTML and CSS parser. However, AngleSharp may be extended with JavaScript capabilities. Right now the package you've used (AngleSharp.Scripting.JavaScript) is experimental and more or less a proof of concept.

The JavaScript files on the page are still too complex for the experimental support. It is my effort to enable support for such scenarios as soon as possible, but right now I would say that WebKit.NET is probably your best shot for solving your problem.

Another possible solution might be to use the C# driver for Selenium.

Unrelated to the whole JavaScript topic: If you want to load external resources you need to provide a proper (http) requester. The easiest way to do that is by using the default one:

var config = new Configuration().WithDefaultLoader();
var document = await BrowsingContext.New(config).OpenAsync("http://www.fosshub.com/MKVToolNix.html");
// ...

In this setting external documents are loaded, but other resources (e.g., images, scripts, ...) are not loaded.