tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: UTF-8 capable fmt(1)



On Sat, Jan 09, 2016 at 02:57:27PM +0100, Kamil Rytarowski wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
> 
> On 09.01.2016 02:46, Thomas Klausner wrote:
> > On Sat, Jan 09, 2016 at 02:36:07AM +0100, Kamil Rytarowski wrote:
> >> 1. I have an impression that it's basically moving the focus
> >> from ASCII to narrowed-down UTF-8, while there are plenty of
> >> other locales and character systems.
> > 
> > It uses standard headers like locale.h, wchar.h and wctype.h, and
> > uses mbtowc(). What is the problem with these functions?
> > 
> 
> Here is a short explanation by libunistring developers:
> 
> http://www.gnu.org/software/libunistring/manual/libunistring.html#The-wchar_005ft-mess

...and of course, they are mixing stupid things with not understanding
things. Narrowing i18n support down to Unicode is still not a valid
approach and ignorant. We should investigate how to make wchar_t locale
independent. That's not the same as making wchar_t UCS4. Doing that is
part of the necessary libc work.

Joerg


Home | Main Index | Thread Index | Old Index