Re: jfbterm support
Hi. Thank you for your comment.
In article <[🔎] 20000118222639.A30465@transas.com>,
at Tue, 18 Jan 2000 22:26:39 +0300,
on Re: jfbterm support (Re: [Adam Di Carlo <adam@onshore.com>] Re: framebuffer),
Michael Sobolev <mss@transas.com> writes:
> On Wed, Jan 19, 2000 at 03:37:00AM +0900, Taketoshi Sano wrote:
> > Sorry, I have commit the patch from hattas, before reading the mails from Adam
> > and Michael. (with the modification to use the new feature of trim-mo in
> > pointerize 0.4, since it has been installed in frozen at master.debian.org.)
> The messages were posted after you comitted your changes. :) So you had no
> chance to read them by that time.
Thank you to make the fact assured. I don't intend to do smug.
If I had read the message from Adam before committing, I would
wait to commit them until the agreement will come up.
> I believe the bogl is not ready yet for multiple languages (as there is no
> working example of how to make it possible to display Russian (koi8-r) or Czech
> (iso-8859-2) or anything else.
I wish if we can use this bogl staff, this may grow up to an attractive
solution. But currently, the work to be done is too much for the release
of potato, I think.
> > currently bogl use .bdf (which is much bigger than
> > compressed .pcf)
> It does not matter since it converts them into C code. I do not think this
> representation is worse than what is used in .pcf.
hmm.
$ ./bdftobogl
usage: ./bdftobogl font.bdf > font.c
$ ./bdftobogl 8x16.bdf >8x16.c
$ ./bdftobogl 8x16rk.bdf >8x16rk.c
$ ./bdftobogl jiskan16.bdf >jiskan16.c
Segmentation fault
$ ls -ls 8x16* jiskan*
31 -rw-rw-r-- 2 sano sano 29916 Apr 27 1994 8x16.bdf
209 -rw-rw-r-- 1 sano sano 212198 Jan 19 23:33 8x16.c
27 -rw-rw-r-- 2 sano sano 25929 Apr 27 1994 8x16rk.bdf
182 -rw-rw-r-- 1 sano sano 184903 Jan 19 23:33 8x16rk.c
1108 -rw-rw-r-- 2 sano sano 1128104 Apr 27 1994 jiskan16.bdf
0 -rw-rw-r-- 1 sano sano 0 Jan 19 23:34 jiskan16.c
I think if we use this for boot-floppies, then we require the font
reduction technologies in this scheme,,,
> > it may lack of the scheme to switch the fonts in according to the encoding
> > of text. (this is required when we use mixed Kanji/Kana/Alphabet text)
> That's the main problem, IMHO.
Yeah.
> > # I tried to build "LINGUA=ja" boot-floppies today. it can boot up as well as,
> > # "LINGUA=C" floppies, but the messages in Japanese can not be read at all.
> > # I think the potato boot-floppies should be enable to display some readable
> > # messages at least.
> Hmm... And how it worked earlier. Do you use special floppies at debian@jp?
hamm-jp/slink-jp uses a special tuned "kon" (which is called "minikon") from
"kon/kon2" package to run the dbootstrap on that "kanji console".
But kon/kon2 uses VGA and is i386 specific, and not the multi language solution.
jfbterm is frame buffer solution, so it maybe multi architecture solution,
and it can be used the multi language solution for some CJK language.
> > if the name "_ja" is not adequate, then let's change that name
> > into something like "_mb" or "_multibyte".
> I believe this may be adequate. :) I mean "_ja", I would only change that code
> you comitted to check for FILE_$LINGUA.
Thanks.
--
Taketoshi Sano: <kgh12351@nifty.ne.jp>
Reply to: