Tim Anderson's ITWriting

November 19, 2003

Unicode in HTML Editor for dot net

I've finally uploaded an update to the C# HtmlEditor that loads unicode strings correctly.

You can find it here.

It turns out that when you load a stream into mshtml, it has no way of knowing what the content is except by "sniffing" - having a look. In order to persuade it that it really is a unicode string, you need to add the byte-order mark. Doing this seems to have fixed the problem of unicode characters being rendered as query symbols.

While I'm on the subject, I've also had a quick look at System.Forms.WebBrowser, the new web browser control in Whidbey. Excellent news that this is now properly integrated and supported. On the other hand, I'd still like to see more of Mshtml's functionality nicely wrapped and exposed in the Framework (which is what my htmleditor tries to do). Either that, or a true native HTML control. It's apparent from the interest in the HtmlEditor that many people want a simple way of loading, editing and retrieving HTML content. I also consider that mshtml, despite its faults (reformatting your code!) is one of the best things about the Windows platform. It's pretty hard to replicate in Java, for example.

Of course it's early days for Whidbey. But from I see right now, they'll still be a need for 3rd party efforts like HtmlEditor after Whidbey is released next year.