[OpenIndiana-discuss] Brace expansion in bash & Locale, localedef, everything related..

[Hans J. Albertsson]

I found that the Oracle solaris IPS repo contains a localedef src 
package that installs in OI w/o any conflicts.

I also discovered that this whole business is in no way a clear and 
unmuddled water:  e g, what IS the proper sequence of characters to 
produce in response to a brace expansion from alphabet(1) to alphabet(last)?
In swedish most would assume it's a-z, minus the w, plus the three chars 
å, ä and ö. No more. But some swedish well-appointed dictionaries do 
actually list w separately.... And slowly swedes are starting to think 
of accented e-s and the german u-diaeresis as "first rank alphabet 
members".

And UTF-8 doesn't have the direct notion of a "base alphabet" , only as 
a derived subset of base sets.

As far as I can see. Unless I'm mistaken. Apply only as directed. Uzw.

I wonder if this problem is being considered elsewhere??

Is there some "i18n SIG" trying to do these kinds of things properly? I 
mean looking at rationalising odd things that came about just because 
ascii allowed them to happen?

On 2013-04-05 16:49, Hans J. Albertsson wrote:
> I realised suddenly that /usr/lib/localedef/src and everything 
> underneath is removed from Openindiana (and Illumos).
> When things like this happen, are they or the rationale behind the 
> change documented anywhere?
>
> I'd like to understand.
>
> _______________________________________________
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss at openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss