Martijn Tonies wrote:

>>> And when decoding this to a string in the client app, the decode routine
>>> recognizes multi-byte chars and decodes them accordingly?
>
> And this decode routine decides how many bytes to use, it determines this
> by the byte values, so the first byte value of the multi-byte character
> signals the decode routine that it is actually 2 bytes?

Actually, I have no idea. We just send the binary buffer to wxString and
tell it to treat that as UTF8 (using wxConvUTF8), or whatever is the
character set and it Just Works (tm). I believe wxString uses iconv
library for these conversions, you can see the supported character sets
and encodings here:

http://www.gnu.org/software/libiconv/


--
Milan Babuskov
http://www.flamerobin.org
http://www.guacosoft.com