On Wed, 6 Sep 2000, Bruno Haible wrote:
> Thomas E. Dickey writes:
>
> > it's hard to say "differently" when the X/Open spec is especially vague
> > in this area - that spec is written with some implementations in mind,
> > but refraining from providing too many details: so unless I look at the
> > implementations, the spec isn't complete.
>
> I don't know what you are looking at; I'm looking at
> http://www.opengroup.org/onlinepubs/007908799/cursesix.html,
yes, I've read that (iirc, it's what I have on the cdrom).
btw, the cdrom contains a number of typographical errors which I found by
compiling the prototypes (around the beginning of 1999). I haven't dumped
the website to see if it's identical, but given the date at the top, it's
likely.
(In any case, there's a disclaimer somewhere that says that the website
and other machine-readable stuff aren't really the spec -- it's really on
paper ;-)
> in particular the "Curses Overview" page, and in particular the
> sections about Characters and Basic Character Operations are quite
> precise. OK, they don't specify which characters are non-spacing, but
we disagree on what's precise (there's no implementation details there,
just names of types).
> > ...but it's worth keeping in mind for small/embedded applications.
>
> libncurses already has 38 KB initialized data and 12 KB bss, so a few
> KB more for the main screen are not bad.
>
> > (and the other aspect - source compatibility would be broken in a number
> > of applications which assume they can OR attributes onto a character).
>
> You don't need to break source compatibility here. The 'chtype'
maybe (I'm only citing the caution in the X/Open spec).
--
T.E.Dickey <[EMAIL PROTECTED]>
http://dickey.his.com
ftp://dickey.his.com
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/lists/
