Unit 2.4 Character Sets: Using eudcedit.exe

1. Starting the programme:
Click "Start" -> "Run" and enter "eudcedit.exe".

2. Selecting an empty code point.
Select an empty code-point and click "OK". By default, private characters are linked to all fonts.

3. Copying (parts of) sample characters
Click on "Window" -> "Reference".

3.1 Setting up the reference font
Make sure the correct font is selected. You should select an Unicode font, like SimSun or Arial Unicode MS.

3.2 Finding the code-point for the required character
We recommend to use the database available at Unicode.org.
Find the link to the "Unihan Database" and then go to the "Unihan Search Page" or use the direct link to the search page: http://www.unicode.org/charts/unihansearch.html.

The "Definition" field contains an english rendering of the character's meaning. You can also change the category to "Mandarin" and search by Pinyin pronouncition with or without tone-numbers.
Thus, if you look for the character ? you enter "zhe4" into the search field.

This search will list you all unicode characters that are pronounced "zhe" in the fourth tone. The code-point is given below the character, for it is "U+9019", thus the code you have to enter in eudcedit is "9019".
If you want to know more about the character, click on the image in the result list and you get the full unicode entry.

There also is the possibility to browse the database by radical and strokes in the "Unihan Radical-stroke Index" available at http://www.unicode.org/charts/unihanrsindex.html. However, browsing the database may turn out to be quite slow unless you narrow down the range of strokes.

EXCURSUS: The Unihan data for a character

The information provided by the unihan database is extremely useful, but as entering the data is very time consuming and this is a still ongoing project, many entries are still empty right now. However, be sure to re-check it once in a while.

The most interesting data are these:

  • Glyphs
    Displays the standard glyph as a) an image, and b) as your browser can display it. Besides showing you where you are, this can be useful to check your browsers standard font.
  • Encoding Forms
    Here all four Unicode encoding forms for that character are given.
  • Mappings to Major Standards - Chinese
    The most important entries in this table are GB2312 and BigFive.
  • Dictionary Information
    This important table lists the number of the character in the Kangxi zidian 康熙字典, the Morohashi 大漢和辭典, and the Hanyu da zidian 漢語大辭典.
  • Other Indices
    Here, among others, you find the number of the character in Karlgren (Grammata Serica Recensa), Nelson (Japanes-English Character Dictionary), and Mathews (Chinese-English Dictionary).
  • Radical-stroke Counts
    These include the data from Unicode, the Kangxi zidian, and Morohashi.
  • Phonetic Data
    In this table the different pronounciations are listed, including the Tang and Japanese On- and Kun readings.
  • Other Dictionary Data
    This includes the character's definition (i.e. an english rendering of its meaning), the total numvbers of its strokes, and the Cangjie code.
  • Variants
    A list of character variants.
  • Chinese Compounds
    Terms, where this character appears in.
  • Other Data
    A list of other data. At this state, some information that does not appear in the tables above may do so here.

3.3 Constructing the character
Now you have found the code point of the character you want to insert, enter it in the "Code" filed in the "Reference" window. The selected character will be displayed in a new window. Use the mouse to draw a box around it by holding the left mouse-button. Whe the complete character or the part you want to use is properly selected right click the selection and choose "Copy" (or press <Strg>+<C>). Then activate the empty character window and paste it in (right-click "paste", or <Strg>+<V>). To resize it use the little squares and drag them.
NOTE: Do not rezise the new character more than two times as its form may not be recognizable afterwards.

Repeat this step for all parts of characters you need.
- You can only open one reference-window at a time.
- If parts of characters overlap use the "freeform Selection" tool instead of the "Rectangular" one.
- Allow yourself some training time. You will soon get used to the somewhat peculiar behaviours of the Editor.
- You will have to apply some polioshing, as the characters get squeezed and the size of strokes may not fit: Use the "Pencil" for black and the "Eraser" (2x2 pixel only!) for white.

3.4 Saving the new character
After you finished editing the character, save it. If you are not sure that a free space was selected, use the "Save as..." dialog from the menu bar.

4. Entering the character into a text
The characters for private use are accessible via the "Character Map".

To insert a character into a text:

  1. Set the cursor to the correct position within your text.
  2. Click on "Start" -> "Programs" -> "Accessories" -> "System Tools" -> "Character Map"
  3. The Character map will display all your private characters.
  4. Select the character you wish to insert and click on select. You can continue selecting other characters, if necessary.
  5. Click "Copy".
  6. Go back to your text and press <Strg>+<V> to insert the character from the clipboard into the text.

These "self-made" Chinese chartacters will work only on the machine, they were created, unless you save them into a separate font-file and distribute it together with your document. A manual on this topic will follow.

Last update: 27 Apr, 2004 (MA)