Eli Zaretskii <eliz@gnu.org> writes:

> I think you are right.  But we could create such an encoding, see
> etc/charsets/ and the coding-system definitions to go with them.

We could, but unfortunately, I'm not able to find any quality source for
the charset.  The closest I've been able to find is the file from IBM
(attached), but it doesn't map to Unicode code points, of course:

...
90 LI610000        i Dotless Small
91 SD130000        Grave Accent
92 SD110000        Acute Accent

glibc doesn't seem to have this, and I can't find it on the Unicode web
site, either.

So we'd have to maintain this by hand (and the easiest way is probably
to copy the table from Wikipedia and massage it).

But...  it seems like an awful lot of work for something like this, so I
think I'll bow out.  If somebody else wants to implement this, that's
totally OK, though.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no