4 This is the DTD of the XML-format kanji file combining information from
5 the KANJIDIC and KANJD212 files. It is intended to be largely self-
6 documenting, with each field being accompanied by an explanatory
9 The file covers the following kanji:
10 (a) the 6,355 kanji from JIS X 0208;
11 (b) the 5,801 kanji from JIS X 0212;
12 (c) the 3,625 kanji from JIS X 0213 as follows:
13 (i) the 2,741 kanji which are also in JIS X 0212 have
14 JIS X 0213 code-points (kuten) added to the existing entry;
15 (ii) the 884 "new" kanji have new entries.
17 At the end of the explanation for a number of fields there is a tag
18 with the format [N]. This indicates the leading letter(s) of the
19 equivalent field in the KANJIDIC and KANJD212 files.
21 The KANJIDIC documentation should also be read for additional
22 information about the information in the file.
23 --><!ELEMENT kanjidic2 (header , character*)
>
24 <!ELEMENT header (file_version , database_version , date_of_creation)
>
26 The single header element will contain identification information
27 about the version of the file
28 --><!ELEMENT file_version (#PCDATA)
>
30 This field denotes the version of kanjidic2 structure, as more
31 than one version may exist.
32 --><!ELEMENT database_version (#PCDATA)
>
34 The version of the file, in the format YYYY-NN, where NN will be
35 a number starting with 01 for the first version released in a
36 calendar year, then increasing for each version in that year.
37 --><!ELEMENT date_of_creation (#PCDATA)
>
39 The date the file was created in international format (YYYY-MM-DD).
40 --><!ELEMENT character (literal , codepoint , radical , misc , dic_number? , query_code? , reading_meaning? , nanori?)*
>
41 <!ELEMENT literal (#PCDATA)
>
43 The character itself in UTF8 coding.
44 --><!ELEMENT codepoint (cp_value)+
>
46 The codepoint element states the code of the character in the various
47 character set standards.
48 --><!ELEMENT cp_value (#PCDATA)
>
50 The cp_value contains the codepoint of the character in a particular
51 standard. The standard will be identified in the cp_type attribute.
52 --><!ATTLIST cp_value cp_type CDATA #REQUIRED
>
54 The cp_type attribute states the coding standard applying to the
55 element. The values assigned so far are:
56 jis208 - JIS X 0208-1997 - kuten coding (nn-nn)
57 jis212 - JIS X 0212-1990 - kuten coding (nn-nn)
58 jis213 - JIS X 0213-2000 - kuten coding (p-nn-nn)
59 ucs - Unicode 4.0 - hex coding (4 or 5 hexadecimal digits)
60 --><!ELEMENT radical (rad_value)+
>
61 <!ELEMENT rad_value (#PCDATA)
>
63 The radical number, in the range 1 to 214. The particular
64 classification type is stated in the rad_type attribute.
65 --><!ATTLIST rad_value rad_type CDATA #REQUIRED
>
67 The rad_type attribute states the type of radical classification.
68 classical - as recorded in the KangXi Zidian.
69 nelson - as used in the Nelson "Modern Japanese-English
70 Character Dictionary" (i.e. the Classic, not the New Nelson).
71 This will only be used where Nelson reclassified the kanji.
72 --><!ELEMENT misc (grade? , stroke_count+ , variant* , freq* , rad_name*)
>
73 <!ELEMENT grade (#PCDATA)
>
75 The Jouyou Kanji grade level. 1 through 6 indicate the grade in which
76 the kanji is taught in Japanese schools. 8 indicates it is one of the
77 remaining Jouyou Kanji to be learned in junior high school, and 9
78 indicates it is a Jinmeiyou (for use in names) kanji. [G]
79 --><!ELEMENT stroke_count (#PCDATA)
>
81 The stroke count of the kanji, including the radical. If more than
82 one, the first is considered the accepted count, while subsequent ones
83 are common miscounts. (See Appendix E. of the KANJIDIC documentation
84 for some of the rules applied when counting strokes in some of the
86 --><!ELEMENT variant (#PCDATA)
>
88 A cross-reference code to another kanji, usually regarded as a variant.
89 The type of cross-reference is given in the var_type attribute.
90 --><!ATTLIST variant var_type CDATA #REQUIRED
>
92 The var_type attribute indicates the type of variant code. The current
94 jis208 - in JIS X 0208 - kuten coding
95 jis212 - in JIS X 0212 - kuten coding
96 jis213 - in JIS X 0213 - kuten coding
97 deroo - De Roo number - numeric
98 njecd - Halpern NJECD index number - numeric
99 s_h - The Kanji Dictionary (Spahn & Hadamitzky) - descriptor
100 nelson - "Classic" Nelson - numeric
101 oneill - Japanese Names (O'Neill) - numeric
102 --><!ELEMENT freq (#PCDATA)
>
104 A frequency-of-use ranking. The 2,500 most-used characters have a
105 ranking; those characters that lack this field are not ranked. The
106 frequency is a number from 1 to 2,500 that expresses the relative
107 frequency of occurrence of a character in modern Japanese. This is
108 based on a survey in newspapers, so it is biassed towards kanji
109 used in newspaper articles. The discrimination between the less
110 frequently used kanji is not strong.
111 --><!ELEMENT rad_name (#PCDATA)
>
113 When the kanji is itself a radical and has a name, this element
114 contains the name (in hiragana.) [T2]
115 --><!ELEMENT dic_number (dic_ref)+
>
117 This element contains the index numbers and similar unstructured
118 information such as page numbers in a number of published dictionaries,
119 and instructional books on kanji.
120 --><!ELEMENT dic_ref (#PCDATA)
>
122 Each dic_ref contains an index number. The particular dictionary,
123 etc. is defined by the dr_type attribute.
124 --><!ATTLIST dic_ref dr_type CDATA #REQUIRED
>
126 The dr_type defines the dictionary or reference book, etc. to which
127 dic_ref element applies. The initial allocation is:
128 nelson_c - "Modern Reader's Japanese-English Character Dictionary",
129 edited by Andrew Nelson (now published as the "Classic"
131 nelson_n - "The New Nelson Japanese-English Character Dictionary",
133 halpern_njecd - "New Japanese-English Character Dictionary",
134 edited by Jack Halpern.
135 halpern_kkld - "Kanji Learners Dictionary" (Kodansha) edited by
137 heisig - "Remembering The Kanji" by James Heisig.
138 gakken - "A New Dictionary of Kanji Usage" (Gakken)
139 oneill_names - "Japanese Names", by P.G. O'Neill.
140 oneill_kk - "Essential Kanji" by P.G. O'Neill.
141 moro - "Daikanwajiten" compiled by Morohashi. For some kanji two
142 additional attributes are used: m_vol: the volume of the
143 dictionary in which the kanji is found, and m_page: the page
144 number in the volume.
145 henshall - "A Guide To Remembering Japanese Characters" by
147 sh_kk - "Kanji and Kana" by Spahn and Hadamitzky.
148 sakade - "A Guide To Reading and Writing Japanese" edited by
150 henshall3 - "A Guide To Reading and Writing Japanese" 3rd
151 edition, edited by Henshall, Seeley and De Groot.
152 tutt_cards - Tuttle Kanji Cards, compiled by Alexander Kask.
153 crowley - "The Kanji Way to Japanese Language Power" by
155 kanji_in_context - "Kanji in Context" by Nishiguchi and Kono.
156 busy_people - "Japanese For Busy People" vols I-III, published
157 by the AJLT. The codes are the volume.chapter.
158 kodansha_compact - the "Kodansha Compact Kanji Guide".
159 --><!ATTLIST dic_ref m_vol CDATA #IMPLIED
>
161 See above under "moro".
162 --><!ATTLIST dic_ref m_page CDATA #IMPLIED
>
164 See above under "moro".
165 --><!ELEMENT query_code (q_code)+
>
167 These codes contain information relating to the glyph, and can be used
168 for finding a required kanji. The type of code is defined by the
170 --><!ELEMENT q_code (#PCDATA)
>
172 The q_code contains the actual query-code value, according to the
174 --><!ATTLIST q_code qc_type CDATA #REQUIRED
>
176 The q_code attribute defines the type of query code. The current values
178 skip - Halpern's SKIP (System of Kanji Indexing by Patterns)
179 code. The format is n-nn-nn. See the KANJIDIC documentation
180 for a description of the code and restrictions on the
181 commercial use of this data. [P]
183 sh_desc - the descriptor codes for The Kanji Dictionary (Tuttle
184 1996) by Spahn and Hadamitzky. They are in the form nxnn.n,
185 e.g. 3k11.2, where the kanji has 3 strokes in the
186 identifying radical, it is radical "k" in the SH
187 classification system, there are 11 other strokes, and it is
188 the 2nd kanji in the 3k11 sequence. (I am very grateful to
189 Mark Spahn for providing the list of these descriptor codes
190 for the kanji in this file.) [I]
191 four_corner - the "Four Corner" code for the kanji. This is a code
192 invented by Wang Chen in 1928. See the KANJIDIC documentation
193 for an overview of the Four Corner System. [Q]
195 deroo - the codes developed by the late Father Joseph De Roo, and
196 published in his book "2001 Kanji" (Bojinsha). Fr De Roo
197 gave his permission for these codes to be included. [DR]
198 misclass - a possible misclassification of the kanji according
199 to one of the code types. (See the "Z" codes in the KANJIDIC
200 documentation for more details.)
202 --><!ELEMENT reading_meaning (rmgroup* , nanori*)
>
204 The readings for the kanji in several languages, and the meanings, also
205 in several languages. The readings and meanings are grouped to enable
206 the handling of the situation where the meaning is differentiated by
208 --><!ELEMENT nanori (#PCDATA)
>
210 Japanese readings that are now only associated with names.
211 --><!ELEMENT rmgroup (reading* , meaning*)
>
212 <!ELEMENT reading (#PCDATA)
>
214 The reading element contains the reading or pronunciation
216 --><!ATTLIST reading r_type CDATA #REQUIRED
>
218 The r_type attribute defines the type of reading in the reading
219 element. The current values are:
220 pinyin - the modern PinYin romanization of the Chinese reading
221 of the kanji. The tones are represented by a concluding
223 korean_r - the romanized form of the Korean reading(s) of the
224 kanji. The readings are in the (Republic of Korea) Ministry
225 of Education style of romanization. [W]
226 korean_h - the Korean reading(s) of the kanji in hangul.
227 ja_on - the "on" Japanese reading of the kanji, in katakana. A
228 second attribute r_status, if present, will indicate with
229 a value of "jy" whether the reading is approved for a
231 ja_kun - the "kun" Japanese reading of the kanji, in hiragana.
232 Where relevant the okurigana is also included separated by a
233 ".". Readings associated with prefixes and suffixes are
234 marked with a "-". A second attribute r_status, if present,
235 will indicate with a value of "jy" whether the reading is
236 approved for a "Jouyou kanji".
237 --><!ATTLIST reading r_status CDATA #IMPLIED
>
239 See under ja_on and ja_kun above.
240 --><!ELEMENT meaning (#PCDATA)
>
242 The meaning associated with the kanji.
243 --><!ATTLIST meaning m_lang CDATA #IMPLIED
>
245 The m_lang attribute defines the target language of the meaning. It
246 will be coded using the two-letter language code from the ISO 639
247 standard. When absent, the value "en" (i.e. English) is implied. [{}]
projects / reactos.git / blob