Re: UTF-8 versus utf8

["Alexander Voropay" <a.voropay@xxxxxxxxxxxx>]

Markus Kuhn <mgk25@xxxxxxxxx> wrote:


>However, glibc normalizes encoding names internally.
>It might be nicer if glibc used instead of a normalization routine an
>internal table of known encoding names, and just matched with the
>normalization routine against the stored names, and if it recognizes
>a name such as "UTF-8", it should output that as the cannonical form,
>not the normalized "utf8". That would avoid lots of headaches.


 Yes. The same problem with russinan "KOI8-R" charset name.
Glibc uses very simplified "name mangling", so some charset aliases doesn't
fit to this scheme : CP-866 .vs. IBM-866 as example.
Fortunately, RedHat produces correct names for some charset as special case
(after glibc install)
Just do in your distributive :
$ mv  /usr/share/locale/en_US.utf8   /usr/share/locale/en_US.UTF-8
$ locale -a

 We should ask Ulrich Drepper to change this.

-- 
-=AV=-

-
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/