Testiness.

[Rob Landley]

On Thursday 01 September 2005 09:21, Robin Farine wrote:
> On Thursday September 1 2005 02:35, Wolfgang Denk wrote:
> > In message <200508311921.16444.rob at landley.net> you wrote:
> > > Now add in the fact that sort (which we were talking about) has the -f
> > > option to ignore case.  (It's in the SUSv3 spec.)  I did not supply the
> > > -f option. The _bug_ that I'm seeing is that with UTF8, the -f option
> > > is forced on.
> >
> > It's even worse than that: if you use something  like  '[a-z]*'  then
> > it's on for A-Y, but not for Z :-(
>
> I followed this thread with amusement and now I just wonder: since
> busybox is meant to be used in size constrained systems, isn't it
> somewhat incompatible with "internationalization" and all the implied
> bells and whistles?

Yes.  We haven't implemented it and have no immediate plans to.  I was vaguely 
thinking that support for the UTF8 character set might be useful (since 99% 
of it is just ascii and the rest is high-ascii escape sequences).  I think 
we're collectively boggling at the huge gratuitous behavior changes the 
internationalization people have made without warning, and working out ways 
of avoiding having to deal with this brain damage if at all possible.

> Everyone here is trying to spare a few bytes in each applet, so is it
> worth trying to support anything but LC_ALL=C? For the full monty thing,
> we already have the GNU stuff ...

I'd like to replace the GNU stuff in a full working system.  That's why I'm 
looking at someday supporting UTF8, at least as a config option.

In theory, most stuff shouldn't require any changes at all...

> Robin

Rob