UTF-8 collations on FreeBSD?

Ivan Voras ivoras at freebsd.org
Sun Dec 16 10:41:25 PST 2007


On 16/12/2007, Andrey Chernov <ache at nagual.pp.ru> wrote:

> Multibyte support works, excepting to collation which defaults to
> single byte order.

Collation is what I most need.

> AFAIK, no one is taking this task yet.
> Speaking particulary about UTF-8, it will be better to have single sorting
> table for whole UTF-8 rather then for particular language.

I don't know how that would work because the position of a character
(the same character) in alphabets varies between languages.

Has anyone looked at possibly importing IBM's ICU, or parts of it? See
http://www.icu-project.org/. Its license is non-restrictive
(http://source.icu-project.org/repos/icu/icu/trunk/license.html) and
it covers probably everything and more than anyone needs?


More information about the freebsd-current mailing list