7    Japanese Input with dxjim

This chapter describes the dxjim Japanese input server, which is the standard Japanese input server on the Tru64 UNIX operating system.

7.1    Outline of the Japanese Conversion Features

The Tru64 UNIX window environment allows you to input Japanese characters in text input fields. To input Japanese characters, type a Roman (Romaji) or Kana character string that represents the phones (YOMI) and then use the (HENKAN) key to convert it into a string of Kanji, Hiragana, or Katakana characters. Table 7-1 lists the Japanese input conversion features.

Table 7-1:  Japanese Input Conversion Features

7.2    Using dxjim

To use the dxjim input server, first start the application on the display where dxjim is running. Then, focus on the character input field (that is, bring it to the foreground). The dxjim title changes to represent the input style and you can now input Japanese characters.

The dxjim input server supports three input styles: over-the-spot (default), on-the-spot, and root-window.

Under the over-the-spot style, Japanese Conversion occurs in the application window and allows you to perform Japanese input without considering dxjim. However, because the character string currently being converted is displayed in the application window, an attempt to insert a character between existing characters might hide part of the existing string until the conversion is determined. Also, if you proceed with the process immediately after character input (for example, by pressing the [OK] button), the characters for which the conversion has been determined are not included. In this case, you have to determine the conversion by pressing [Ctrl/G] or [Ctrl/N] for the JVMS conversion key before proceeding with the process.

Under the on-the-spot style, Japanese Conversion occurs in the application window and allows you to perform Japanese input without considering dxjim. However, unlike the over-the-spot style, no part of an already typed character string is hidden when you attempt to insert a character between existing characters.

Under the root-window style, you must consider dxjim. That is, the root-window style displays in the dxjim window, any character string you typed in the application window. Because the entire character string currently being converted is displayed in the dxjim window, you must be aware of the dxjim window until the Japanese Conversion is determined.

For information on setting up the input styles, see Section 7.2.2.

7.2.1    Setting Up dxjim

When running, the dxjim input server displays either the character string (Kanji Conversion) or "ASCII" at the bottom left field of the dxjim window. While remains displayed there, the application can input Japanese characters using dxjim. When "ASCII" is displayed, Japanese input does not work. Immediately after startup of the application, dxjim is in the mode.

The key used to change the mode from "ASCII" to is called the Start key. The key used for reverse mode (from to "ASCII") is called the End key.

By default, the Start and End key is Shift-Space (that is, pressing the space bar while holding down the [Shift] key). To change the Start and End key settings, use the Start Key: Settings dialog box that displays when you select the Set Key menu from the Options menu. Changes to the Start and End keys made in the Start Key: Settings dialog box are effective for all applications.

The dxjim input server uses the user key definition library IMLIB. The dxjim input server uses the key definition of the user who has started the input server. For information on definitions of the conversion and user keys, and IMLIB, refer to the User Key Definition User's Guide.

To change the conversion key, select Key Definition from the Options menu. Using this method, you can change only the definition of the conversion key; you cannot change the manner of echoing. The conversion key change applies only to the widget you used for input of Japanese characters while the change was in effect.

If the application uses root-window input style, a preedit area is created in the dxjim window. The key used to determine the preedit for input to the application is called the Determination key. The Determination key default is the [Return] key.

To change the Determination key, use the Start key: Settings dialog box. The new Determination key applies to all applications that use the root-window input style.

7.2.2    Input Styles

The input styles supported by dxjim (on-the-spot, over-the-spot, and root-window) are listed as input styles for text window input in the X11R6.3 standard. (X11R6.3 also lists a few preedit modes, which are not supported by dxjim). The dxjim input server can also display the mode of the connected client.

You can use the XmText() and DXmCSText() widgets to set the input style using the VendorShell() widget's XmNpreeditType() resource. For this resource, specify one of the following values:

• OnTheSpot

• OverTheSpot

• Root

• OffTheSpot

The XmNpreeditType() resource allows you to specify two or more of these values, separated by commas.

The default input style is OverTheSpot. Under the OverTheSpot style, the character string currently being converted is initially overwritten on the existing text. After the conversion is determined, the string is inserted into the entire text.

7.2.3    Using dxjim Resources

Using resources, you can modify dxjim settings. The resources available for dxjim are listed in this section. The class name for dxjim is DXjim and the default resource file is /usr/i18n/lib/X11/app-defaults/DXjim.

• x

  Specifies the x coordinate of the top left corner of the dxjim window.

  	Setting:	Use an integer that represents the number of pixels counted from the left margin
  	Initial setting:	Bottom left corner of the screen

• y

  Specifies the y coordinate of the top left corner of the dxjim window.

  	Setting:	Use an integer that represents the number of pixels counted from the top margin
  	Initial setting:	Bottom left corner of the screen

• startupKanjiMode

  Enables or disables Japanese input immediately after startup of the text widget. If startupKanjiMode is on, you enter Japanese input mode immediately after startup of the text widget.

  	Setting:	on or off
  	Initial setting:	on

• startKey

  Sets the Start key. You can also set startKey in the Set Key dialog box by selecting Set Key from the Options menu.

  	Setting:	See Section 7.2.1
  	Initial setting:	Shift<Key>space

• endKey

  Sets the End key. You can also set endKey in the Set Key dialog box by selecting Set Key from the Options menu.

  	Setting:	See Section 7.2.1
  	Initial setting:	Shift<Key>space

• root_commitKey

  Sets the Determination key. You can also set root_commitKey in the Set Key dialog box by selecting Set Key from the Options menu.

  	Setting:	See Section 7.2.1
  	Initial setting:	<Key>Return

• kp_numlock

  If the kp_numlock setting is False, the keypad keys cannot be used to input preedit strings. The preedit is determined by pressing any keypad key during preedit. If the kp_numlock setting is True, the keypad keys, except the PF1 to PF4 and Enter, can be used to input preedit strings.

  	Setting:	True or False
  	Initial setting:	False

• icon_state

  If the icon_state setting is Normal, the dxjim window is restored by starting preedit with the dxjim window iconized. If the icon_state setting is Iconic, automatic restoration of the dxjim window never takes place. If the icon_state setting is Withdrawn, the dxjim window does not appear and the root preedit style is not available. To use an icon to start dxjim, set up the resource as follows:

           DXjim*initialState: IconicState
   
  

  	Setting:	Iconic, Withdrawn, or Normal
  	Initial setting:	Normal

• use_root_resources

  Determines whether to use the root preedit style resources (root_fontset, root_background, root_foreground, root_highlight, and root_highlight_fore). If the use_root_resources setting is False, the setting defined with the Xset-ICValues of the client will be used.

  	Setting:	True or False
  	Initial setting:	True

• root_background

  Specifies the background color that is used for preedit string areas of the root preedit style when the use_root_resource resource is True.

  	Type:	Pixel
  	Initial setting:	white

• root_foreground

  Specifies the foreground color that is used for preedit string areas of the root preedit style when the use_root_resource resource is True.

  	Setting:	Color setting. A named color is available.
  	Initial setting:	black

• root_highlight

  Specifies the background color of conversion candidate character strings that is used for preedit string areas of the root preedit style when the use_root_resource resource is True.

  	Setting:	Color setting. A named color is available.
  	Initial setting:	black

• root_hightlight_fore

  Specifies the foreground color of conversion candidate character strings that is used for preedit string areas of the root preedit style when the use_root_resource resource is True.

  	Setting:	Color setting. A named color can also be used.
  	Initial setting:	white

• root_fontset

  Specifies the font set that is used for preedit strings of the root preedit style when the use_root_resource resource is True.

  	Setting:	Color setting. A named color can also be used.
  	Initial setting:	See the following:

  -jdecw-gothic-medium-r-normal--*-120-*-*-m-*-jisx0208.1983-1,
  -jdecw-gothic-medium-r-normal--*-120-*-*-m-*-jisx0201.1976-0,
  -jdecw-kmenu-medium-r-normal--*-120-*-*-*-*-iso8859-1
  

• root_width

  Specifies the width of preedit areas of the root preedit style.

  	Setting:	The number of pixels (integer) that represents the width
  	Initial setting:	Determined dynamically

7.3    Japanese Input Conversion Keys

Tru64 UNIX provides definitions for the conversion keys used for text input. Japanese VMS (JVMS) is the default conversion key definition unless otherwise specified. See Section 7.3.1 for information on changing the conversion key definition.

The conversion key definitions are:

• JVMS (default)

  The standard Japanese VMS conversion key definition. The JVMS conversion keys are listed in Table 7-2.

• EVEJ

  The conversion key definition for the EVEJ editor and used in the EVEJ keypad mode of Japanese EVE. The EVEJ conversion keys are as listed in Table 7-3.

• KEYPAD

  The key definition that uses the VT-series terminal numeric keypad. The KEYPAD conversion keys are as listed in Table 7-4.

• TARO

  The conversion key definition that is similar to that of the word processing software Ichitaro. The TARO conversion keys are as listed in Table 7-5.

• CTRL

  The key definition that uses the Control key. The CTRL conversion keys are as listed in Table 7-6.

Table 7-2:  JVMS Conversion Key Definition

Table 7-3:  EVEJ Conversion Key Definition

Table 7-4:  KEYPAD Conversion Key Definition

Table 7-5:  TARO Conversion Key Definition

Table 7-6:  CTRL Conversion Key Definition

7.3.1    Changing the Conversion Key Definition

By default, dxjim uses the JVMS Conversion Key Definition.

To change the conversion key definition you use for applications, use the following steps. The steps use the example of changing the key definition to EVEJ.

1. Define the IM_PROFILE environment variable as follows:

   % setenv IM_PROFILE im_profile_evej  (When csh is in use)
    
   $ IM_PROFILE=im_profile_evej
   $ export IM_PROFILE                  (When sh or ksh is in use)
   

2. To use the JVMS, TARO, CTRL, or KEYPAD Conversion Key Definition, change im_profile_evej (shown in the example in step 1) to one of the following:

   • JVMS: im_profile_jvms

   • TARO: im_profile_taro

   • CTRL: im_profile_ctrl

   • KEYPAD: im_profile_keypad

3. To use the new key definition, restart the application.

If you include the definition in the user's .cshrc (or .profile if sh is in use), you will be able to use the selected key definition for subsequent sessions.

For details on key definitions, refer to the User Key Definition User's Guide.

7.4    Conversion Rules

This section defines the following conversion rules:

• Romaji-to-Kana Conversion (Section 7.4.1)

  Shows the correspondence between Roman and Kana characters that are used in typing the phones in Roman characters.

• Special-Character Conversion (Section 7.4.2)

  Shows the correspondence between symbols and special characters that are used in converting a character into a special character using Katakana Conversion.

• Symbol Conversion (Section 7.4.3)

  Shows the correspondence between the alphanumeric characters and symbols and the special characters that are used in converting a string of one to three alphanumeric characters or symbols into a special character.

• Code Input (Section 7.4.4)

  Describes how to convert Kanji codes into the Kanji characters using Symbol Conversion.

7.4.1    Romaji-to-Kana Conversion Correspondence Table

7.4.2    Special-Character Conversion Correspondence Table

7.4.3    Symbol Conversion Correspondence Table

• One-character conversion

  

• Two-Character Conversion

  
  
  
  
  
  
  
  
  

• Three-Character Conversion

  

7.4.4    Code Input

Type the Kanji code for symbol conversion to input a Kanji character using a Kanji code. Two types of Kanji codes are available:

• DEC Kanji

• Ku-Ten

If you use DEC Kanji code, attach an "X" to the beginning.

For example,

If you use a Ku-Ten code, attach a "J" to the beginning.

For example,

For the members of the DEC Kanji and Ku-Ten codes, see the book Kanji Code Table that is included in the Japanese version of the Tru64 UNIX Installation Kit.

7.5    Using the Keyboard Kana Mode

Tru64 UNIX supports Kana keyboards. After the Japanese Keyboard is setup on the system, a Kana keyboard is automatically available.

A Kana keyboard has two modes and you can toggle between the two. Kana Mode lets you type Kana characters. ASCII Mode lets you type alphabetic characters and digits. The default mode is ASCII.

To toggle between the ASCII and Kana Modes on the PCXAJ keyboard, press the [Ctrl] key located at the right.

Note

The PCXAJ-AA is equipped with two [Ctrl] keys.

To toggle between the ASCII and Kana Modes on any other Japanese keyboard, press the (KANA) key.

On a North American keyboard, the [Compose] or [Comp] key functions as the (KANA) key.

7.6    Supporting Japanese Keyboards Equipped with Conversion Keys

Tru64 UNIX supports the following Japanese keyboards, which are equipped with keys dedicated to Japanese input (for example, (HENKAN), (MU-HENKAN)):

• LK411-JJ/LK41W-JJ (JIS layout)

• LK411-AJ/LK41W-AJ (ANSI layout)

• PCXAJ-AA/LK97W-AA (JIS layout)

For details about setting up the keyboards, refer to the Release Notes.

Any of the keyboards equipped with the (HENKAN) key allow you to use the keystrokes listed in Table 7-7, as well as the keystrokes based on the conversion key definition (JVMS, EVEJ, or KEYPAD).

Note

For the TARO Conversion Key Definition, the keystrokes listed in Table 7-7 are not available.

Table 7-7:  Conversion Key Definition of a Japanese Keyboard Equipped with the (HENKAN) Keys

Function	Keystrokes [Footnote 7]
Kanji Conversion or Next Phrase Candidate
Kanji Candidate List
Hiragana Conversion or Phrase Hiragana Conversion
Full-Width Katakana Conversion or Full-Width Phrase Katakana Conversion	; is to press the key while holding down the [Shift] key
Full-Width Conversion
Half-Width Katakana Conversion or Half-Width Phrase Katakana Conversion	or
Half-Width Conversion
Symbol Conversion or Code Input
Move to Next Phrase
Move to Previous Phrase
Reduce Phrase
Enlarge Phrase
Previous Phrase Candidate	; is to press the key while holding down the [Shift] key
No Conversion
Cancel Conversion