On Thu, 25 Jan 2001, Markus Kuhn wrote:
> Robert Brady wrote on 2001-01-25 15:49 UTC:
> > We are just talking about how the terminal interprets certain control
> > codes. One of these is rather confusingly named "backspace".
>
> Unfortunately, ISO 6429 clearly was not written with biwidth output
> in mind and doesn't provide a useful/unambiguous definition here:
>
> ???
Indeed.
>From an implementation POV, I find backspace-for-columns to be better than
for characters, because what should
0041 0300 0008 0061
do, if the backspace only removes the 0300? Print Aa? Well, ok. So what
should
0041 0300 0300 0008 0008 0061 do? Print Aa as well?
And what about
0041 0300 0300 0300 0300 0300
0008 0008 0008 0008 0008 0061 do? Print Aa?
But that would mean we'd have to keep _all_ the combining characters
stored. (xterm currently only keeps 2 of them), or at least keep track of
how many there were. This would be a major hassle.
So we cannot have ^H honour wcwidth() for combining characters. So I
don't feel theres much point honouring it for ==2 characters, either, that
would just be inconsistent.
--
Robert
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/lists/
