At 04:35 PM 16/12/2001 -0800, Michael \(michka\) Kaplan wrote: >Actually, it should be possible to use UTF-16 and then treat a surrogate >pair the way you would treat any two code points that make a specific glyph? >The same trick could be used for keyboards as well, as far as I can see.
With keyboards under Windows, you must send a surrogate pair for WM_CHAR messages, but you need to send a UTF-32 codepoint for WM_UNICHAR messages. Marc Durdin Tavultesoft
