Re: proper UTF-8 support under OSX

["FranÃois Revol" <revol@xxxxxxx>]

> > > No. fr_FR.UTF-8 doesn't necessarily exist. For instance, under
> > > Maemo Linux, one needs to use LC_CTYPE=fr_FR, which is UTF-8, but
> > > under most Linux-based platforms, fr_FR is ISO-8859-1, so that
> > > the
> > > charmap is
> >
> > This looks braindead to me, couldn't they just specify it to always
> > be
> > canonical ?
>
> Yes, this was a very bad choice IMHO. But the user has to deal with
> it.

Shame, indeed.

> > > incorrect after a SSH, unless one uses some tricks (I use a
> > > LC_CHARMAP variable in order to update the LC_CTYPE from my
> > > .zshenv if need be).
> >
> > Note however than charmap != encoding !!
>
> What do you mean here?
>
> The same notion is often designated under the following terms (which
> may have different meanings depending on the context): charset (e.g.
> in MIME), character set, codeset (e.g. in nl_langinfo), charmap (e.g.
> in locales), coding system (e.g. in Emacs), encoding (e.g. in XML).

No.
Charset = character set = the list of available glyphs, no matter how
they are coded.

Encoding = how they are represented.

Like, UTF-8 one of the many encodings or the Unicode charset.

FranÃois.