Re: Character set tagging considered harmful

["T.E.Dickey" <dickey@clark.net>]

> Sounds like a good idea.  Looking through the xterm code, I notice that it 
> stores each character of the screen in two bytes (excluding color and 
> attributes).  This can't be the UTF-8 code, that would require three bytes. 
> Since Vim does screen handling like xterm, I might end up doing the same. 

what you're considering UTF-8 is the input/output (which indeed can be 3 or
more bytes) - the corresponding UCS data are 2-byte.

(Markus has requested a second plane, which would increase the 2-byte codes
to 4 - but that's a different matter)

-- 
Thomas E. Dickey
dickey@clark.net
http://www.clark.net/pub/dickey
-
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/lists/