Re: [Nano-devel] nano trouble with Vietnamese UI translation

[David Lawrence Ramsey]

Phan Vinh Thinh wrote:

>hi,
>Please check this screen-shot:
>http://iatp.vspu.ac.ru/phan/transproj/nano-vi-err.png (ver 1.3.5, but
>the same with 1.3.6)

That's odd.  Since nano has UTF-8 support starting with 1.3.6, it should
at least be working with that (given your list of terminals below).

>There are some "anomalous" characters. The title should be "B m
>mi" (New buffer), but displayed... Same with menu at the bottom.
>People said that this must be a bug.
> Some info:
>You can take translated .po from here:
>http://iatp.vspu.ac.ru/phan/transproj/nano-1.3.5.po.gz
>mlterm, uxterm, urxvt, konsole, bash in UTF-8 mode give the same
>error. Other translations (bash, tar, vim, util-linux,...) *work
>fine*.
>My distro Debian GNU/Linux sid. Kernel 2.6.10.
>Please, let me know if you need more information & CC me.

Thanks for the report.  I'll need more information, though:

1. Are you in a UTF-8 locale when the problems occur?  The UTF-8 support
in 1.3.6 is only enabled if the locale contains the string "UTF8" or
"UTF-8", case insensitive.

2. What curses library is 1.3.6 linked against?  If ncurses, it should
be linked against the wide version of ncurses (ncursesw).  If slang, it
should be linked against one of the official slang 2.0 prereleases. 
Debian's slang1a-utf8 package won't work.

3. How are you getting 1.3.6?  Are you building it from source, or are
you pulling the 1.3.6 package from Debian experimental?  If it's the
former, see if the configure script gives you a warning about
insufficient wide character support and tells you that nano won't be
able to support UTF-8.  If so, then you need to make sure that you have
the proper curses library as listed above, and the following functions
in your C library: iswalnum(), iswspace() or iswblank(), mblen(),
mbtowc(), wctomb(), and wcwidth().