Winforms Webbrowser Control and Character Encoding

Question

I am trying to use the Winforms WebBrowser control to display a webpage that can be in many different languages. In one example, I have a HTML file that has characters in Farsi (examples are not limited to this, however - the file can have Mongolian, Japanese, or any other language).

When I try to Navigate() to this file in the WebBrowser control, it displays in a bunch of garbled characters (eg. ÓÇÚÊ). If I display the same file in Firefox, however, it will correctly display with all the the expected characters (eg. ساعت). I can also load the raw HTML file in Notepad++ and it seems to automatically detect the character set/encoding that is being used.

I have read numerous threads that talk about setting the WebBrowser encoding like so:

webBrowser.Document.Encoding = "UTF-8"

however given a set of HTML files, I have no way of telling what language it is written in and therefore no way to know the encoding. I can also confirm there is no "meta" tag within the HTML source that specifies any encoding.

Is there some magic going on behind the scenes of Firefox and Notepad++ to detect the correct character set and if so, how? Can someone tell me how I can get the WebBrowser control to behave in a similar way?

Answer 1

As @Bizan pointed out, in the absence of meta tags to describe the encoding, modern browsers seem to use heuristics to determine the language encoding.

Since the WebBrowser control is based on dated Internet Explorer technology, it appears there is no in-built logic to do these heuristics automatically.

The solutions are:

• Implement your own heuristics method and then set the encoding manually.

• Use the WebView control (as mentioned by @PanagiotisKanavos) which appears to be use the latest Edge browser. I have tested the pages in Edge and they work correctly. The minimum requirement for WebView is Windows 10 however, which will rule out any use-cases where you need your software to run on earlier versions of Windows.