Re: Question about vim-6.0

[Bruno Haible <haible@xxxxxxx>]

Hello Khoa,

Thanks for the details. Most of your expectations are correct.

> vim-6.0r.rt.tar.gz,
> vim-6.0r.extra.tar.gz, vim-6.0r.lang.tar.gz on the
> site
> ftp://ftp.vim.org/pub/vim. I compiled it with these
> commands:
> ./configure --with-features=big

You also need to pass --enable-multibyte. This option was implicit in
--with-features=big in earlier versions (vim 6.0g) but now you have to
give it explicitly.

> I loaded my unicode font with the command:
> consolechars -f myunicodefont
> I used the more command to view the contents of the
> text unicode file. I was very good. I saw all unicode
> character. But when I used the command:
> vim mytestunicode.c
> These were something wrong. Some characters with two
> (2) bytes displayed well, but some unicode characters
> not displayed true. I don't know the reason why.

The fonts that "consolechars" can load into the VGA card can have at
most 512 characters. This is too small for all of Unicode. There are
two ways to get over this limitation:

   1) Compile VESA framebuffer support into your Linux kernel, and
      at boot time, put the graphics card into a graphics mode, e.g.
      say "vga = ask" in /etc/lilo.conf and type 0305 when Linux asks
      for the VGA mode.

      Then install an UTF-8 capable terminal emulator which runs on
      the framebuffer. There is one in Debian CVS:

      cvs -d :pserver:anonymous@xxxxxxxxxxxxxx:/cvs/debian-boot login
      cvs -d :pserver:anonymous@xxxxxxxxxxxxxx:/cvs/debian-boot co boot-floppies/utilities/bogl

      (Bug reports about this one to Edmund GRIMLEY EVANS <edmundo@xxxxxxxx>.)

   2) Hook an xterm instead of a getty on your console. You might call
      this cheating, but it works fine for me :-)

      See http://mail.nl.linux.org/linux-utf8/2000-07/msg00053.html

      For this, you need xterm-146 or newer.

Bruno
-
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/lists/