[PATCH] (Draft2) Pay more attention to the POSIX locale on Unix

[Aidan Kehoe]

 Ar an séú lá is fiche de mí na Samhain, scríobh Ville Skyttä: 

 > > I have 956 distinct locale strings available locally, taken from
 > > Solaris, XFree86 (of which the locale strings are from the distinct
 > > systems it runs on), NetBSD and Linux, and these are the modifiers that
 > > are used:
 > > 
 > > cyrillic
 > > euro
 > > nynorsk
 > > ucs4
 > 
 > FWIW, dunno if this falls into the above category, but on my Fedora Core
 > 6 box, I see:
 > 
 > $ ls -1d /usr/share/locale/*@*
 > /usr/share/locale/ca_ES at valencian
 > /usr/share/locale/ca at valencia
 > /usr/share/locale/en at boldquot
 > /usr/share/locale/en at quot
 > /usr/share/locale/nds at NFE
 > /usr/share/locale/sr at ije
 > /usr/share/locale/sr at Latn
 > /usr/share/locale/uz at Latn

Ah, thanks for the pointers. W.r.t Valencian, all forms of Catalan fit in
Latin-1, so that modifier doesn’t change anything. It turns out that the
existing Serbian language environment defaults to the Roman alphabet--nnngh,
let’s not refight the most recent Balkan wars in XEmacs commits--and I don’t
know what FC means by sr at ije, nds at NFE, en at boldquot or en at quot. Uzbek in
Latin fits in ASCII, so our non-comprehension of it won’t be a problem, give
our fallback to ASCII.

-- 
Santa Maradona, priez pour moi!