Re: [vile] Utf-8 word movement broken.

[Ramil Farkhshatov]

Thomas Dickey <address@hidden> wrote:

> On Fri, 8 Apr 2011, Ramil Farkhshatov wrote:
>
> > Thomas Dickey <address@hidden> wrote:
> >
> >> On Mon, Jan 24, 2011 at 09:15:04PM +0300, Ramil Farkhshatov wrote:
> >>> Hello.
> >>>
> >>> Recently, I've noticed, that 'next-punctuated-word' in utf-8 cyrillic
> >>> text can move cursor inside a word.  'next-word' acts fine.  I couldn't
> >>> find exact letters that act like punctuation, but the issue can be
> >>> reproduced on these lines (utf-8):
> >>> абвгодеёжзийклмнопрстуфхцчшщьъюэюя
> >>> АБВГОДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЬЪЮЭЮЯ
> >>> ЗЕЛЁНЫЙ
> >>> зелёный
> >>>
> >>> It seems that it has been broken by 9.8c patch.
> >>>
> >>> System info
> >>> OS/Distro: Arch Linux
> >>> Locale: ru_RU.UTF-8
> >>> vile-9.8[cd]
> >>
> >> Your message indicates that you're using 9.8d
> >> However, with 9.8d (or current code, which shouldn't
> >> be different in this case), I cannot reproduce this.
> >> That's with Debian 6.0 and 5.0.
> >>
> >> Using
> >>
> >> :set sc
> >> :show-wide-printable
> >>
> >> it seems that vile is seeing all of the characters in
> >> the same as alphabetic (either upper or lower).
> >
> > On my system 'show-wide-printable' does not show cyrillic letters at
> > all.  But after ':set sc' modeline shows utf-8 char codes correctly.
> > (file-encoding=locale, if this information is useful)
>
> perhaps this is (yet another) problem with my workaround for the "narrow" 
> locale.  What does "locale -a" say on your system?

% locale -a
C
en_GB.utf8
en_US
en_US.iso88591
en_US.utf8
POSIX
ru_RU.utf8