| 1 |
On Fri, 30 Jul 2004 00:00:01 -0700 "Robin H. Johnson" |
| 2 |
<robbat2@g.o> wrote: |
| 3 |
| > Ok, since I've been asked about this several times on IRC... Some |
| 4 |
| > people have names which include non-[A-Za-z] characters. |
| 5 |
| UTF8 != binary. |
| 6 |
|
| 7 |
UFT-8 (sometimes) uses the upper bit on certain bytes. It is therefore, |
| 8 |
in the traditional sense, binary. You could, of course, just not mark it |
| 9 |
as such for CVS... |
| 10 |
|
| 11 |
| > Incidentally, when are we going to officially standardise on UTF-8 |
| 12 |
| > for ChangeLog and ebuilds? |
| 13 |
| There is absolutely nothing I'm aware of stopping you from using UTF |
| 14 |
| already in ChangeLogs and ebuilds, barring some common sense |
| 15 |
| limitations, like bash not accepting non-ASCII for the names of |
| 16 |
| variables and functions. |
| 17 |
|
| 18 |
No, but not everyone is using UTF-8, resulting in horridly mixed files. |
| 19 |
Hence the "officially standardise". Right now we have a mess. |
| 20 |
|
| 21 |
-- |
| 22 |
Ciaran McCreesh : Gentoo Developer (Sparc, MIPS, Vim, Fluxbox) |
| 23 |
Mail : ciaranm at gentoo.org |
| 24 |
Web : http://dev.gentoo.org/~ciaranm |
Archives