Proper (complete) Html encoding

[Logan1337]

Hello. I need to take a string in UTF-8 with extended characters (e.g trademark, curly quotes, etc) and encode it for html, with either the html named entities or xml numbered (unicode) entities.

I've tried HttpUtility.HtmlEncode(), but from what I can gather, this function is really (at least originally) only intended to prevent cross-site scripting, and therefore only encodes some characters, leaving the others unchanged.

For example, when I encode the string "<trademark><registered>" I get back "<trademark><ampersand>#174;". (sorry for the retarded representation, I ironically couldn't figure out how to represent characters in posts.) In other words, only the registered symbol is properly encoded, while the trademark symbol remains.

Is there a proper method or way to encode a string such that all extended characters are properly encoded?

[Logan1337]

Ok guys, this is the best I've come up with, and to be quite honest I'm hardly satisfied. But it works, and right now that's what I need. If anyone has a better solution PLEASE post it. I've searched the web and found diddly.

Iterate over each character in the given string and use Convert.ToInt32(c) to get its ("hopefully" unicode) integer value. For all characters over say 127 (as I believe that's where standard ASCII ends), do this:

"&#" + intValue.ToString() + ";";

Believe it or not, it actually works. Believe me, no one was more surprised than I, especially after searching the web for a solution for more than an hour and finding nothing of any particular value.

Of course I will also apply HttpUtility.HtmlEncode() beforehand, as that will take care of the "lesser" control characters like & and <,>, quotes etc.

I've tested it with:

This is a "test" of my make-shift html encode & decode algorithm for FeedBeast™. © 2007

...and it works, so hurray.