Markus Kuhn wrote:
>
> Robert de Bath wrote on 2001-06-11 18:46 UTC:
> >
> > There's also an issue of '0' not being the same as 'default'
> > conflicting with VT100 compatibility ... but I need to check
> > some things first.
>
> ECMA-48 implies nowhere that the default of every parameter for
> every control function has to be 0, otherwise quote the exact
> section. The default is specified for every function. I see no
> conflict with VT100 compatibility here and it is absolutely
> trivial to write a parameter parser properly such that 0 and
> default are distinguishable.
The VT100 was not written to comply with the latest edition of ECMA-48,
but ANSI X3.64-1977, in which an empty parameter string and one with a
value of zero were identical. Therefore distinguishing between the two
must only be done for the new sequence, although changing a parser to
distinguish the two cases and then treat them as the same for most
sequences is not difficult -- which has nothing to do with writing a
parser "properly".
> I am slightly concerned though that CSI ... w was apparently
> already used by DEC, so I guess it will have to be CSI ... !w
> then.
That's true, but it wasn't my major concern. Your paper starts with the
line
"This proposal adds a new control sequence to the ISO 6429 = ECMA-48
standard"
which implies that this is not a private sequence. Therefore, you cannot
use a final byte of 07/00 to 07/14. Read section 5.4(d).
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/linux-utf8/
