Juliusz Chroboczek wrote on 2001-05-24 21:53 UTC:
>   http://www.cl.cam.ac.uk/~mgk25/ucs/scw-proposal.html
> 
> I am not very keen on adding yet another piece of state to terminal
> emulators.  Adding something like SCW would require all term{cap,info}
> entries for terminal initialisation to be updated to include SCW(0),
> and would be yet another source of puzzlement to users (``I swear the
> columns did align when I last tried that out!'').
> 
> My preference would be for a set of single shifts.

Reasonable argument. One could also add as a minor robust efficiency
improvement variants that remain active until the next LF or CANcel
control character.

> By default, the terminal emulator behaves according to some well defined
> set of widths (say, as in UTR 11).

Well, UTR 11 has lots of neutral/ambiguous characters, so it is not
suited alone as a reference for describing a "well defined set of
width". It just gives an overview of the problem, not a solution.

> There are three sequences, call them ``width
> override'':
> 
>   WO0: next character has width 0;
>   WO1: next character has width 1;
>   WO2: next character has width 2.

I do like the idea in principle to avoid long-term state.

I'm not sure exactly, what "next character has width 0" is good for. Is
this supposed to drop a character (the application can do this itself
already), or is it supposed to go back one position (or to the
previously written character) and overstrike it? Unicode is pretty clear
and unambiguous, what the combining characters suitable for overstriking
are (general category code Mn or Me in the Unicode database), so I see
no reason to force a normal character to become overstriking. Or do you
mean "next character has width 0" in the sense "either use the next
character as a combining character or ignore it, but under no
circumstances move the cursor", so as to get controllable cursor
position even if you don't know whether the terminal does or does not
implement combining characters?

Markus

-- 
Markus G. Kuhn, Computer Laboratory, University of Cambridge, UK
Email: mkuhn at acm.org,  WWW: <http://www.cl.cam.ac.uk/~mgk25/>

-
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/

Reply via email to