Hello, I have this wonderful Japanese keyboard with all the hiragana Symbols on it. I would like to use them in Linux. Is it possible to create a keyboard coding for this, simply by letting the Kana keys produce the respective Roman Characters (i.e. the press of "ka" syllable rendering the two Roman letters "k" and "a")? If yes, how can I do this so that it works for KDE 3.0? Best Regards, Thomas Piekenbrock -- GMX - Die Kommunikationsplattform im Internet. http://www.gmx.net
Thomas Piekenbrock <thomas.piekenbrock@gmx.de> は書きました:
I have this wonderful Japanese keyboard with all the hiragana Symbols on it. I would like to use them in Linux. Is it possible to create a keyboard coding for this,
Yes.
simply by letting the Kana keys produce the respective Roman Characters (i.e. the press of "ka" syllable rendering the two Roman letters "k" and "a")?
No, it's done differently.
If yes, how can I do this so that it works for KDE 3.0?
There's nothing special about KDE 3.0 in that respect, if you setup your Canna-clients like kinput2 to use the kana layout, it will work everywhere, including KDE 3.0. Here's how to make it work: Copy the attached ~/.canna-jisx6002.ctd to your home directory. You have to compile it to binary form with mfabian@gregory:~$ mkromdic .canna-jisx6002.ctd forcpp -7 < .canna-jisx6002.ctd | /lib/cpp |forcpp -8 | kpdic > .canna-jisx6002.cbp SIZE 603 KEYS 92 mfabian@gregory:~$ Now put (setq romkana-table ".canna-jisx6002.cbp") in your ~/.canna to load the binary version of the keyboard map. If you don't have a ~/.canna, create one or use the one I attach to this mail. Restart your canna clients like kinput2, XEmacs, Emacs, nvi-m17n, ... You must use the Japanese keyboard map of course, i.e. you should use setxkbmap jp jp106 and/or have Section "InputDevice" [...] Option "XkbLayout" "jp" Option "XkbModel" "jp106" [...] EndSection in /etc/X11/XF86Config. Now the keyboard layout should be correct for kana input. The following small deficiencies probably cannot be solved: - '\' and the 'Yen'-key both output 'ろ'. - '~' and "Overscore" both output 'を' 'Overscore' is the key left of the 'Yen' key, shifted. There is no real 'Overscore', therefore 'Overscore' always seems to output '~' on the Japanese keyboard layouts used by X11. On the JIS X 6002 keyboard, the 'Overscore' key is labelled with the repetition character '々', i.e. because of this problem you can't input this character directly. But that is no big problem, usually one doesn't need to input this character directly anyway. The conversion backend (Canna) usually does that for you when necessary. I.e. when you type 'いろいろ' and convert, you will get '色々' automatically. And if you really need to input 々 directly, you can still use the special symbol input as described in http://www.suse.de/~mfabian/suse-cjk/kinput2-canna-special-symbo.html - I was not sure on what key to put the 'ー' in my JIX X 6002 keymap. According to Figure 5-10 on page 248 in Ken Lundes book "CJK Information processing", this character should be on the same key as 'ろ' (shifted). Therefore I mapped '_' to 'ー'. -- Mike Fabian <mfabian@suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。
Mike Fabian wrote:
Here's how to make it work:
Copy the attached ~/.canna-jisx6002.ctd to your home directory. You have to compile it to binary form with
mfabian@gregory:~$ mkromdic .canna-jisx6002.ctd forcpp -7 < .canna-jisx6002.ctd | /lib/cpp |forcpp -8 | kpdic > .canna-jisx6002.cbp SIZE 603 KEYS 92 mfabian@gregory:~$
Now put
(setq romkana-table ".canna-jisx6002.cbp")
in your ~/.canna to load the binary version of the keyboard map.
If you don't have a ~/.canna, create one or use the one I attach to this mail.
Restart your canna clients like kinput2, XEmacs, Emacs, nvi-m17n, ...
You must use the Japanese keyboard map of course, i.e. you should use
setxkbmap jp jp106
and/or have
Section "InputDevice" [...] Option "XkbLayout" "jp" Option "XkbModel" "jp106" [...] EndSection
in /etc/X11/XF86Config.
Now the keyboard layout should be correct for kana input.
Thank you, it works. This now leads to one more question: As some punctuation characters, especially Zenkaku ones, are not accessible in Kana-Nyuuryoku mode, is there a simple way to toggle between Kana-Nyuuryoku and Romaji-Nyuuryoku?
The following small deficiencies probably cannot be solved:
- '\' and the 'Yen'-key both output '?'. - '~' and "Overscore" both output '?' 'Overscore' is the key left of the 'Yen' key, shifted. There is no real 'Overscore', therefore 'Overscore' always seems to output '~' on the Japanese keyboard layouts used by X11. On the JIS X 6002 keyboard, the 'Overscore' key is labelled with the repetition character '?', i.e. because of this problem you can't input this character directly. But that is no big problem, usually one doesn't need to input this character directly anyway. The conversion backend (Canna) usually does that for you when necessary. I.e. when you type '????' and convert, you will get '??' automatically. And if you really need to input ? directly, you can still use the special symbol input as described in
http://www.suse.de/~mfabian/suse-cjk/kinput2-canna-special-symbo.html
- I was not sure on what key to put the '?' in my JIX X 6002 keymap. According to Figure 5-10 on page 248 in Ken Lundes book "CJK Information processing", this character should be on the same key as '?' (shifted). Therefore I mapped '_' to '?'.
The mapping is ok for me. I assume that the line together with "ro" is not the Kanji for "ichi", but the Katakana symbol for vowel stretching. Is that correct? Best regards, Thomas Piekenbrock, Tokio
Thomas Piekenbrock <thomas.piekenbrock@gmx.de> writes:
Mike Fabian wrote:
Here's how to make it work:
[...]
Thank you, it works. This now leads to one more question: As some punctuation characters, especially Zenkaku ones, are not accessible in Kana-Nyuuryoku mode,
Which ones? The most common ones, i.e. 、。・ are easily accessible on the JIS X 6002 layout. For the others you can use the special symbol input of Canna.
is there a simple way to toggle between Kana-Nyuuryoku and Romaji-Nyuuryoku?
I don't think so. I don't know any way to change the value of (setq romkana-table "default.cbp") in ~/.canna later when the canna-client is already running. [...]
- I was not sure on what key to put the 'ー' in my JIX X 6002 keymap. According to Figure 5-10 on page 248 in Ken Lundes book "CJK Information processing", this character should be on the same key as 'ろ' (shifted). Therefore I mapped '_' to 'ー'.
The mapping is ok for me. I assume that the line together with "ro" is not the Kanji for "ichi", but the Katakana symbol for vowel stretching. Is that correct?
Yes, it is the vowel stretching character: Character `ー' UNIDATA information. --------------------------------- This is converted to U+30FC under the current environment. name KATAKANA-HIRAGANA PROLONGED SOUND MARK category (letter modifier) combining-class 0 => Spacing bidirectional-category L => Left-to-Right mirrored not-mirrored titlecase-mapping -1 -- Mike Fabian <mfabian@suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。
Mike Fabian wrote:
Thomas Piekenbrock <thomas.piekenbrock@gmx.de> writes:
Mike Fabian wrote:
Here's how to make it work:
[...]
Thank you, it works. This now leads to one more question: As some punctuation characters, especially Zenkaku ones, are not accessible in Kana-Nyuuryoku mode,
Which ones?
The most common ones, i.e. 、。・ are easily accessible on the JIS X 6002 layout. For the others you can use the special symbol input of Canna.
I mainly mean various kinds of brackets, question mark, and exclamation mark, and Zenkaku-Romaji and Zenkaku numbers. Would be convenient to switch to direct input when needed, but symbol input should be ok also. I am still somewhat illiterate about the usage of Canna. How can I access the special symbol input? Thomas
Thomas Piekenbrock <thomas.piekenbrock@gmx.de> writes:
Mike Fabian wrote:
Thomas Piekenbrock <thomas.piekenbrock@gmx.de> writes:
[...]
This now leads to one more question: As some punctuation characters, especially Zenkaku ones, are not accessible in Kana-Nyuuryoku mode,
Which ones?
The most common ones, i.e. 、。・ are easily accessible on the JIS X 6002 layout. For the others you can use the special symbol input of Canna.
I mainly mean various kinds of brackets, question mark, and exclamation mark, and Zenkaku-Romaji and Zenkaku numbers. Would be convenient to switch to direct input when needed, but symbol input should be ok also. I am still somewhat illiterate about the usage of Canna. How can I access the special symbol input?
http://www.suse.de/~mfabian/suse-cjk/kinput2-canna-menu.html http://www.suse.de/~mfabian/suse-cjk/kinput2-canna-special-symbo.html This explanation is for kinput2 with Canna, but it is the same with the Canna interfaces of XEmacs. The Canna interface of nvi-m17n and the Canna interface provided by tamago for Emacs are a bit different though. -- Mike Fabian <mfabian@suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。
participants (2)
-
Mike Fabian
-
Thomas Piekenbrock