From mboxrd@z Thu Jan  1 00:00:00 1970
From: okamoto@granite.cias.osakafu-u.ac.jp
Message-Id: <200008160247.WAA05212@cse.psu.edu>
Subject: [9fans] ktrans for Release 3 Plan 9
Date: Wed, 16 Aug 2000 11:49:09 +0900
To: 9fans@cse.psu.edu
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Topicbox-Message-UUID: fb4e754c-eac8-11e9-9e20-41e7f4b1d025

I put "ktrans for Release 3 Plan 9" on our web site=20
(http://basalt.cias.osakafu-u.ac.jp/plan9/p9index.html). =20
Ktrans was included in the Release 2 Plan 9, which enables us to=20
input Japanese and Greek and Russian using English key board. =20
This is a "derived work" Licensed distribution of Plan 9.  Above page=20
is written in UTF-8, and you can search the part read as English, and=20
take it, even if you'd find many of strange language there. =E2=98=BA

Below is my README file for convenience to others.

This version of ktrans works with pipefile by Rob Pike, which replaces /d=
ev/kbd of
the Release 2 Plan 9 by Dennis Ritchie.  By this new idea, we can input J=
apanese
anytime onto any windows.  I included his post to 9fans mailing-list here=
 (pipefile
file).

This is a distribution as a derived work license of Plan 9, and I include=
d the
Plan 9 License term here.  Please refer to it on the details of this lice=
nse.


Main features are as follows:

1) have a "local" dictionary file for translation from kana to kanji whic=
h can=20
      easily editted by user.  The default file name is $home/lib/ktrans-=
jisho.
      If you want to use other dictionary file, set the KTJISHO environme=
nt=20
      variable to point to that file.  This dictionary complize many line=
s of
      text strings. =20

2) capital romaji input for a word such as verbs or adjectives with okuri=
gana,=20
      which follows the idea of SKK system by Masahiko Sato of Kyoto Univ.
      (masahiko@kuis.kyoto-u.ac.jp).  If you want to get kanji string (ru=
nes)
      "=E5=8B=95=E3=81=8B=E3=81=99", which is a verb, you may input "ugoK=
asu" from the keyboard. =20
      Note here the Kasu's k is a capital (important).  You will see a hi=
ragana
      runes "=E3=81=86=E3=81=94=E3=81=8B=E3=81=99", and then, the kanji r=
unes of "=E5=8B=95=E3=81=8B=E3=81=99", when you hit=20
      "Shift+Space".=20
        If you are satisfied by that translation, continue to input next =
word.  =20
      If you are not pleasure with that candidate, hit 'Shift+Space' once=
 more=20
      to see more candiate for that hiragana inputs.  When no more other =
candidates
      are registered in your dictionary, you will see the initial hiragan=
a input.

3) for Japanese "joshi", a postpositioned short word after noun, you can =
use another
      method which I developed for this work.  If you want kanji string (=
runes)
      "=E7=A7=81=E3=81=AF", then, try to hit "watashiHA" from the keyboar=
d.  Note that the sound
      of "wa(ha)" is express as capitalized "HA".  You will see a hiragan=
a string
      of "=E3=82=8F=E3=81=9F=E3=81=97=E3=81=AF", and then, "=E7=A7=81=E3=81=
=AF" after `Shift+Space'.

4) a control sequence of 'ctl-l' is introduced to leave input hiragana ru=
nes unchanged.
      This is occasionally neccessary.

5) simple leaning mechanism has been implemented on the on-memory hashing=
 dictinary
      where most recently used kanji runes (candidate) moves to the top o=
f the
      list of candidates.  This is valid only during the session you call=
ed ktrans.
      This is done so intentionally, because present learning method is .=
.well...=20
      naive. ^_^  I know this, however, I believe you can solve it by mak=
ing a good=20
      dictionary best fitted to your purpose by yourself.

6)  'ctl-x' re-reads the new kana-kanji translation dictionary when you e=
dited your
      kana-kanji translation to update by acme, and want to reflect it in=
to your
      current on memory translation dictionary.
      The kana-kanji translation dictionary is, usually, read only once a=
t the=20
      beginning of the ktrans session. I believe you will be suffered not=
hing=20
      from this simplification, because you can edit the dictionary anyti=
me by
      co-working with acme, and use this ctl-x.

7) mode change to kana-input is triggered by 'ctl-n', ctl-k for katakana =
mode,=20
      ctl-g for Greek mode, and ctl-r for Russian mode.

8) as the starting $home/lib/ktrans-jisho, you may re-format the SKK-JISY=
O.S=20
      (66.9KB) of SKK system, which can be reached from ftp.kuis.kyoto-u.=
ac.jp. =20
      The next three lines shows the short sed filter to transform from S=
KK type=20
      dictionary to Plan 9.  Before this, you should change kanji code of=
 the SKK
      dictionary from ujis(euc) to UTF-8 by tcs utility, of course.
            s/\// /g
            s/  /	/g
            s/ $//g
      The header items are sorted in a strange order in the original SKK=20
      dictionary.  This implementation does not care the order, therefore=
,=20
      you can change it by yourself.

9) SKK jisho, such as SKK-JISYO.S, is composed of two parts, okuri-ari an=
d=20
      okuri-nashi entries.  This is greatly depend on the Japanese gramme=
r,
      and okuri-ari may represent verb/adjective etc., i.e., not noun.
        These two parts work differently in the original SKK system, howe=
ver,
      I did not employed that method, rather, a simple approarch as descr=
ibed
      in (2) and (3).  Here, we have no difference between these two part=
s,=20
      and the reason why I leaved the two part structure remained is just=
=20
      to make easier to read for editting.  Of course, you can change it
      without any side-effects.
   =20
10) This implementation of Japanese input method is to convert every one =
word=20
      by one key triggering essentially.  This may cause some cumbersome =
feeling=20
      to Nihongo users who are accustomed to, say, Windows. I know this. =
=20
      However, I intended to keep the codes compact as possible as a firs=
t step=20
      to develope Nihongo input system on Plan 9.  Furthermore, I've neve=
r seen=20
      the latters worked perfectly.  I think the conversion failed essent=
ially=20
      when we see more than, say, five/six candidates for one input hirag=
ana
      runes.

11) The trabslation trigger key has been chnaged from ^t to Shift+Space, =
because
      we experienced butting of the trigger key sequence sometime.  For t=
his reason,
      you have to edit key binding table such as kbtabshift[0x39] in /sys=
/src/9/pc/kbd.c
      or keymapshift[0x79] in /sys/src/9/ss/screen.c to assign Shift+Spac=
e to '^\'.
      In my patch to translate.c, I use '^\' as the trigger key.  Therefo=
re, you can
      trigger translation from kana to kanji by Shit+Space or just '^\'.

12) a usage example: if you want to make the Japanese text as below:

           =E7=A7=81=E3=81=AF=E6=AF=8E=E6=97=A535=E5=88=86=E4=BB=A5=E4=B8=
=8A=E6=AD=A9=E3=81=84=E3=81=A6=E3=80=81 =E6=9B=B4=E3=81=AB10=E5=88=86=E9=9B=
=BB=E8=BB=8A=E3=81=AB=E4=B9=97=E3=81=A3=E3=81=A6=E5=AD=A6=E6=A0=A1=E3=81=AB=
=E9=80=9A=E3=81=84=E3=81=BE=E3=81=99=E3=80=82
           =E5=81=A5=E5=BA=B7=E3=81=AE=E7=B6=AD=E6=8C=81=E3=81=AB=E3=82=82=
=E5=BD=B9=E3=81=A0=E3=81=A3=E3=81=A6=E3=81=84=E3=81=BE=E3=81=99=E3=81=8C=E3=
=80=81 =E3=81=AA=E3=81=8B=E3=81=AA=E3=81=8B=E3=81=9F=E3=81=AE=E3=81=97=E3=
=81=84=E3=82=82=E3=81=AE=E3=81=A7=E3=81=99=E3=80=82

      your keyboard hitting stream should be:

      watashiHA[^t]mainichi[^t]35[^l]fun[^t]ijou[^t]aruIte, [^t]saraNI[^t=
]
      10[^l]fun[^t]denshaNI[^t]noTte[^t]gakkouNI[^t]kayoImasu.[^t]
      kenkouNO[^t]ijiNImo[^t]yakuDAtteimasuga, [^t]nakanaka[^l]tanoshiI[^=
t]
      monodesu.[^l]

    where [^t], [^l] indicates 'Shift+Space' and 'ctl-l' respectively.


		       				Kenji  Okamoto  August 14, 2000