Sync up with r21-4-15-chise-0_21-31-daikanwa-rev.
[chise/xemacs-chise.git-] / lisp / utf-2000 / char-db-util.el
1 ;;; char-db-util.el --- Character Database utility
2
3 ;; Copyright (C) 1998,1999,2000,2001,2002,2003,2004 MORIOKA Tomohiko.
4
5 ;; Author: MORIOKA Tomohiko <tomo@kanji.zinbun.kyoto-u.ac.jp>
6 ;; Keywords: CHISE, Character Database, ISO/IEC 10646, Unicode, UCS-4, MULE.
7
8 ;; This file is part of XEmacs CHISE.
9
10 ;; XEmacs CHISE is free software; you can redistribute it and/or
11 ;; modify it under the terms of the GNU General Public License as
12 ;; published by the Free Software Foundation; either version 2, or (at
13 ;; your option) any later version.
14
15 ;; XEmacs CHISE is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
18 ;; General Public License for more details.
19
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with XEmacs CHISE; see the file COPYING.  If not, write to
22 ;; the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 ;; Boston, MA 02111-1307, USA.
24
25 ;;; Code:
26
27 (require 'alist)
28
29 (defconst unidata-normative-category-alist
30   '(("Lu" letter        uppercase)
31     ("Ll" letter        lowercase)
32     ("Lt" letter        titlecase)
33     ("Mn" mark          non-spacing)
34     ("Mc" mark          spacing-combining)
35     ("Me" mark          enclosing)
36     ("Nd" number        decimal-digit)
37     ("Nl" number        letter)
38     ("No" number        other)
39     ("Zs" separator     space)
40     ("Zl" separator     line)
41     ("Zp" separator     paragraph)
42     ("Cc" other         control)
43     ("Cf" other         format)
44     ("Cs" other         surrogate)
45     ("Co" other         private-use)
46     ("Cn" other         not-assigned)))
47
48 (defconst unidata-informative-category-alist
49   '(("Lm" letter        modifier)
50     ("Lo" letter        other)
51     ("Pc" punctuation   connector)
52     ("Pd" punctuation   dash)
53     ("Ps" punctuation   open)
54     ("Pe" punctuation   close)
55     ("Pi" punctuation   initial-quote)
56     ("Pf" punctuation   final-quote)
57     ("Po" punctuation   other)
58     ("Sm" symbol        math)
59     ("Sc" symbol        currency)
60     ("Sk" symbol        modifier)
61     ("So" symbol        other)
62     ))
63
64 (defconst ideographic-radicals
65   (let ((v (make-vector 215 nil))
66         (i 1))
67     (while (< i 215)
68       (aset v i (decode-char '=ucs (+ #x2EFF i)))
69       (setq i (1+ i)))
70     v))
71
72 (defvar char-db-file-coding-system 'utf-8-mcs-er)
73
74 (defvar char-db-feature-domains
75   '(ucs daikanwa cns gt jis jis/alt jis/a jis/b
76         jis-x0212 jis-x0213 cdp shinjigen misc unknown))
77
78 (defvar char-db-ignored-attributes nil)
79
80 (defun char-attribute-name< (ka kb)
81   (cond
82    ((eq '->denotational kb)
83     t)
84    ((eq '->subsumptive kb)
85     (not (eq '->denotational ka)))
86    ((eq '->denotational ka)
87     nil)
88    ((eq '->subsumptive ka)
89     nil)
90    ((find-charset ka)
91     (if (find-charset kb)
92         (if (<= (charset-id ka) 1)
93             (if (<= (charset-id kb) 1)
94                 (cond
95                  ((= (charset-dimension ka)
96                      (charset-dimension kb))
97                   (> (charset-id ka)(charset-id kb)))
98                  (t
99                   (> (charset-dimension ka)
100                      (charset-dimension kb))
101                   ))
102               t)
103           (if (<= (charset-id kb) 1)
104               nil
105             (< (charset-id ka)(charset-id kb))))
106       nil))
107    ((find-charset kb)
108     t)
109    ((symbolp ka)
110     (cond ((symbolp kb)
111            (string< (symbol-name ka)
112                     (symbol-name kb)))
113           (t)))
114    ((symbolp kb)
115     nil)))
116
117 (defvar char-db-coded-charset-priority-list
118   '(ascii
119     control-1
120     latin-iso8859-1
121     latin-iso8859-2
122     latin-iso8859-3
123     latin-iso8859-4
124     latin-iso8859-9
125     latin-jisx0201
126     cyrillic-iso8859-5
127     greek-iso8859-7
128     thai-tis620
129     =jis-x0208
130     japanese-jisx0208
131     japanese-jisx0212
132     japanese-jisx0208-1978
133     chinese-gb2312
134     chinese-cns11643-1
135     chinese-cns11643-2
136     chinese-cns11643-3
137     chinese-cns11643-4
138     chinese-cns11643-5
139     chinese-cns11643-6
140     chinese-cns11643-7
141     =jis-x0208-1990
142     =jis-x0213-1-2000
143     =jis-x0213-2-2000
144     korean-ksc5601
145     chinese-isoir165
146     katakana-jisx0201
147     hebrew-iso8859-8
148     chinese-gb12345
149     latin-viscii
150     ethiopic-ucs
151     =big5-cdp
152     =gt
153     ideograph-daikanwa-2
154     ideograph-daikanwa
155     =cbeta
156     ideograph-hanziku-1
157     ideograph-hanziku-2
158     ideograph-hanziku-3
159     ideograph-hanziku-4
160     ideograph-hanziku-5
161     ideograph-hanziku-6
162     ideograph-hanziku-7
163     ideograph-hanziku-8
164     ideograph-hanziku-9
165     ideograph-hanziku-10
166     ideograph-hanziku-11
167     ideograph-hanziku-12
168     =big5
169     =big5-eten
170     =gt-k
171     =jef-china3))
172
173 (defun char-db-make-char-spec (char)
174   (let (ret char-spec)
175     (cond ((characterp char)
176            (cond ((and (setq ret (encode-char char '=ucs 'defined-only))
177                        (not (and (<= #xE000 ret)(<= ret #xF8FF))))
178                   (setq char-spec (list (cons '=ucs ret)))
179                   (cond ((setq ret (get-char-attribute char 'name))
180                          (setq char-spec (cons (cons 'name ret) char-spec))
181                          )
182                         ((setq ret (get-char-attribute char 'name*))
183                          (setq char-spec (cons (cons 'name* ret) char-spec))
184                          ))
185                   )
186                  ((setq ret
187                         (catch 'tag
188                           (let ((rest char-db-coded-charset-priority-list)
189                                 ccs)
190                             (while rest
191                               (setq ccs (charset-name
192                                          (find-charset (car rest))))
193                               (if (setq ret
194                                         (encode-char char ccs
195                                                      'defined-only))
196                                   (throw 'tag (cons ccs ret)))
197                               (setq rest (cdr rest))))))
198                   (setq char-spec (list ret))
199                   (dolist (ccs (delq (car ret) (charset-list)))
200                     (if (and (or (charset-iso-final-char ccs)
201                                  (memq ccs
202                                        '(=daikanwa
203                                          =daikanwa@rev2
204                                          ;; =gt-k
205                                          )))
206                              (setq ret (encode-char char ccs 'defined-only)))
207                         (setq char-spec (cons (cons ccs ret) char-spec))))
208                   (if (null char-spec)
209                       (setq char-spec (split-char char)))
210                   (cond ((setq ret (get-char-attribute char 'name))
211                          (setq char-spec (cons (cons 'name ret) char-spec))
212                          )
213                         ((setq ret (get-char-attribute char 'name*))
214                          (setq char-spec (cons (cons 'name* ret) char-spec))
215                          ))
216                   ))
217            char-spec)
218           ((consp char)
219            char))))
220     
221 (defun char-db-insert-char-spec (char &optional readable column
222                                       required-features)
223   (unless column
224     (setq column (current-column)))
225   (let (char-spec temp-char)
226     (setq char-spec (char-db-make-char-spec char))
227     (unless (or (characterp char) ; char
228                 (condition-case nil
229                     (setq char (find-char char-spec))
230                   (error nil)))
231       ;; define temporary character
232       ;;   Current implementation is dirty.
233       (setq temp-char (define-char (cons '(ideograph-daikanwa . 0)
234                                          char-spec)))
235       (remove-char-attribute temp-char 'ideograph-daikanwa)
236       (setq char temp-char))
237     (insert-char-attributes char
238                             readable
239                             (union (mapcar #'car char-spec)
240                                    required-features))
241     (when temp-char
242       ;; undefine temporary character
243       ;;   Current implementation is dirty.
244       (setq char-spec (char-attribute-alist temp-char))
245       (while char-spec
246         (remove-char-attribute temp-char (car (car char-spec)))
247         (setq char-spec (cdr char-spec))))))
248
249 (defun char-db-insert-alist (alist &optional readable column)
250   (unless column
251     (setq column (current-column)))
252   (let ((line-breaking
253          (concat "\n" (make-string (1+ column) ?\ )))
254         name value
255         ret al ; cal
256         key
257         lbs cell rest separator)
258     (insert "(")
259     (while alist
260       (setq name (car (car alist))
261             value (cdr (car alist)))
262       (cond ((eq name 'char)
263              (insert "(char . ")
264              (if (setq ret (condition-case nil
265                                (find-char value)
266                              (error nil)))
267                  (progn
268                    (setq al nil
269                          ;; cal nil
270                          )
271                    (while value
272                      (setq key (car (car value)))
273                      ;; (if (find-charset key)
274                      ;;     (setq cal (cons key cal))
275                      (setq al (cons key al))
276                      ;; )
277                      (setq value (cdr value)))
278                    (insert-char-attributes ret
279                                            readable
280                                            (or al 'none) ; cal
281                                            ))
282                (insert (prin1-to-string value)))
283              (insert ")")
284              (insert line-breaking))
285             ((consp value)
286              (insert (format "(%-18s " name))
287              (setq lbs (concat "\n" (make-string (current-column) ?\ )))
288              (while (consp value)
289                (setq cell (car value))
290                (if (and (consp cell)
291                         (consp (car cell))
292                         (setq ret (condition-case nil
293                                       (find-char cell)
294                                     (error nil)))
295                         )
296                    (progn
297                      (setq rest cell
298                            al nil
299                            ;; cal nil
300                            )
301                      (while rest
302                        (setq key (car (car rest)))
303                        ;; (if (find-charset key)
304                        ;;     (setq cal (cons key cal))
305                        (setq al (cons key al))
306                        ;; )
307                        (setq rest (cdr rest)))
308                      (if separator
309                          (insert lbs))
310                      (insert-char-attributes ret
311                                              readable
312                                              al ; cal
313                                              )
314                      (setq separator lbs))
315                  (if separator
316                      (insert separator))
317                  (insert (prin1-to-string cell))
318                  (setq separator " "))
319                (setq value (cdr value)))
320              (insert ")")
321              (insert line-breaking))
322             (t
323              (insert (format "(%-18s . %S)%s"
324                              name value
325                              line-breaking))))
326       (setq alist (cdr alist))))
327   (insert ")"))
328
329 (defun char-db-insert-char-reference (plist &optional readable column)
330   (unless column
331     (setq column (current-column)))
332   (let ((line-breaking
333          (concat "\n" (make-string (1+ column) ?\ )))
334         (separator "")
335         name value)
336     (insert "(")
337     (while plist
338       (setq name (pop plist))
339       (setq value (pop plist))
340       (cond ((eq name :char)
341              (insert separator)
342              (insert ":char\t")
343              (cond ((numberp value)
344                     (setq value (decode-char '=ucs value)))
345                    ;; ((consp value)
346                    ;;  (setq value (or (find-char value)
347                    ;;                  value)))
348                    )
349              (char-db-insert-char-spec value readable)
350              (insert line-breaking)
351              (setq separator ""))
352             ((eq name :radical)
353              (insert (format "%s%s\t%d ; %c%s"
354                              separator
355                              name value
356                              (aref ideographic-radicals value)
357                              line-breaking))
358              (setq separator ""))
359             (t
360              (insert (format "%s%s\t%S" separator name value))
361              (setq separator line-breaking)))
362       ))
363   (insert ")"))
364
365 (defun char-db-decode-isolated-char (ccs code-point)
366   (let (ret)
367     (setq ret
368           (cond ((eq ccs 'arabic-iso8859-6)
369                  (decode-char ccs code-point))
370                 ((and (memq ccs '(=gt-pj-1
371                                   =gt-pj-2
372                                   =gt-pj-3
373                                   =gt-pj-4
374                                   =gt-pj-5
375                                   =gt-pj-6
376                                   =gt-pj-7
377                                   =gt-pj-8
378                                   =gt-pj-9
379                                   =gt-pj-10
380                                   =gt-pj-11))
381                       (setq ret (decode-char ccs code-point))
382                       (setq ret (encode-char ret '=gt 'defined-only)))
383                  (decode-builtin-char '=gt ret))
384                 (t
385                  (decode-builtin-char ccs code-point))))
386     (cond ((and (<= 0 (char-int ret))
387                 (<= (char-int ret) #x1F))
388            (decode-char '=ucs (+ #x2400 (char-int ret))))
389           ((= (char-int ret) #x7F)
390            ?\u2421)
391           (t ret))))
392
393 (defvar char-db-convert-obsolete-format t)
394
395 (defun char-db-insert-ccs-feature (name value line-breaking)
396   (insert
397    (format
398     (cond ((memq name '(=daikanwa
399                         =daikanwa@rev1 =daikanwa@rev2
400                         =gt =gt-k =cbeta))
401            "(%-18s . %05d)\t; %c")
402           ((eq name 'mojikyo)
403            "(%-18s . %06d)\t; %c")
404           ((>= (charset-dimension name) 2)
405            "(%-18s . #x%04X)\t; %c")
406           (t
407            "(%-18s . #x%02X)\t; %c"))
408     name
409     (if (= (charset-iso-graphic-plane name) 1)
410         (logior value
411                 (cond ((= (charset-dimension name) 1)
412                        #x80)
413                       ((= (charset-dimension name) 2)
414                        #x8080)
415                       ((= (charset-dimension name) 3)
416                        #x808080)
417                       (t 0)))
418       value)
419     (char-db-decode-isolated-char name value)))
420   (if (and (= (charset-chars name) 94)
421            (= (charset-dimension name) 2))
422       (insert (format " [%02d-%02d]"
423                       (- (lsh value -8) 32)
424                       (- (logand value 255) 32))))
425   (insert line-breaking))
426
427 (defun insert-char-attributes (char &optional readable attributes column)
428   (unless column
429     (setq column (current-column)))
430   (let (name value ; has-long-ccs-name
431         rest
432         radical strokes
433         (line-breaking
434          (concat "\n" (make-string (1+ column) ?\ )))
435         lbs cell separator ret
436         key al cal
437         dest-ccss
438         sources required-features
439         ccss)
440     (let (atr-d)
441       (setq attributes
442             (sort (if attributes
443                       (if (consp attributes)
444                           (progn
445                             (dolist (name attributes)
446                               (unless (memq name char-db-ignored-attributes)
447                                 (if (find-charset name)
448                                     (push name ccss))
449                                 (push name atr-d)))
450                             atr-d))
451                     (dolist (name (char-attribute-list))
452                       (unless (memq name char-db-ignored-attributes)
453                         (if (find-charset name)
454                             (push name ccss))
455                         (push name atr-d)))
456                     atr-d)
457                   #'char-attribute-name<)))
458     (insert "(")
459     (when (and (memq 'name attributes)
460                (setq value (get-char-attribute char 'name)))
461       (insert (format
462                (if (> (+ (current-column) (length value)) 48)
463                    "(name . %S)%s"
464                  "(name               . %S)%s")
465                value line-breaking))
466       (setq attributes (delq 'name attributes))
467       )
468     (when (and (memq 'name* attributes)
469                (setq value (get-char-attribute char 'name*)))
470       (insert (format
471                (if (> (+ (current-column) (length value)) 48)
472                    "(name* . %S)%s"
473                  "(name*              . %S)%s")
474                value line-breaking))
475       (setq attributes (delq 'name* attributes))
476       )
477     (when (and (memq 'script attributes)
478                (setq value (get-char-attribute char 'script)))
479       (insert (format "(script\t\t%s)%s"
480                       (mapconcat (function prin1-to-string)
481                                  value " ")
482                       line-breaking))
483       (setq attributes (delq 'script attributes))
484       )
485     ;; (when (and (memq '<-denotational attributes)
486     ;;            (setq value (get-char-attribute char '<-denotational))
487     ;;            (null (cdr value))
488     ;;            (setq value (encode-char (car value) 'ucs 'defined-only)))
489     ;;   (insert (format "(%-18s . #x%04X)\t; %c%s"
490     ;;                   '=>ucs value (decode-char 'ucs value)
491     ;;                   line-breaking))
492     ;;   (setq attributes (delq '<-denotational attributes)))
493     (dolist (name '(=>ucs =>ucs*))
494       (when (and (memq name attributes)
495                  (setq value (get-char-attribute char name)))
496         (insert (format "(%-18s . #x%04X)\t; %c%s"
497                         name value (decode-char '=ucs value)
498                         line-breaking))
499         (setq attributes (delq name attributes))))
500     (dolist (name '(=>ucs@gb =>ucs@cns =>ucs@jis =>ucs@ks =>ucs@big5))
501       (when (and (memq name attributes)
502                  (setq value (get-char-attribute char name)))
503         (insert (format "(%-18s . #x%04X)\t; %c%s"
504                         name value
505                         (decode-char (intern
506                                       (concat "="
507                                               (substring
508                                                (symbol-name name) 2)))
509                                      value)
510                         line-breaking))
511         (setq attributes (delq name attributes))
512         ))
513     (dolist (name '(=>daikanwa))
514       (when (and (memq name attributes)
515                  (setq value (get-char-attribute char name)))
516         (insert
517          (if (integerp value)
518              (format "(%-18s . %05d)\t; %c%s"
519                      name value (decode-char '=daikanwa value)
520                      line-breaking)
521            (format "(%-18s %s)\t; %c%s"
522                    name
523                    (mapconcat (function prin1-to-string)
524                               value " ")
525                    (char-representative-of-daikanwa char)
526                    line-breaking)))
527         (setq attributes (delq name attributes))))
528     (when (and (memq 'general-category attributes)
529                (setq value (get-char-attribute char 'general-category)))
530       (insert (format
531                "(general-category\t%s) ; %s%s"
532                (mapconcat (lambda (cell)
533                             (format "%S" cell))
534                           value " ")
535                (cond ((rassoc value unidata-normative-category-alist)
536                       "Normative Category")
537                      ((rassoc value unidata-informative-category-alist)
538                       "Informative Category")
539                      (t
540                       "Unknown Category"))
541                line-breaking))
542       (setq attributes (delq 'general-category attributes))
543       )
544     (when (and (memq 'bidi-category attributes)
545                (setq value (get-char-attribute char 'bidi-category)))
546       (insert (format "(bidi-category\t. %S)%s"
547                       value
548                       line-breaking))
549       (setq attributes (delq 'bidi-category attributes))
550       )
551     (unless (or (not (memq 'mirrored attributes))
552                 (eq (setq value (get-char-attribute char 'mirrored 'empty))
553                     'empty))
554       (insert (format "(mirrored\t\t. %S)%s"
555                       value
556                       line-breaking))
557       (setq attributes (delq 'mirrored attributes))
558       )
559     (cond
560      ((and (memq 'decimal-digit-value attributes)
561            (setq value (get-char-attribute char 'decimal-digit-value)))
562       (insert (format "(decimal-digit-value . %S)%s"
563                       value
564                       line-breaking))
565       (setq attributes (delq 'decimal-digit-value attributes))
566       (when (and (memq 'digit-value attributes)
567                  (setq value (get-char-attribute char 'digit-value)))
568         (insert (format "(digit-value\t . %S)%s"
569                         value
570                         line-breaking))
571         (setq attributes (delq 'digit-value attributes))
572         )
573       (when (and (memq 'numeric-value attributes)
574                  (setq value (get-char-attribute char 'numeric-value)))
575         (insert (format "(numeric-value\t . %S)%s"
576                         value
577                         line-breaking))
578         (setq attributes (delq 'numeric-value attributes))
579         )
580       )
581      (t
582       (when (and (memq 'digit-value attributes)
583                  (setq value (get-char-attribute char 'digit-value)))
584         (insert (format "(digit-value\t. %S)%s"
585                         value
586                         line-breaking))
587         (setq attributes (delq 'digit-value attributes))
588         )
589       (when (and (memq 'numeric-value attributes)
590                  (setq value (get-char-attribute char 'numeric-value)))
591         (insert (format "(numeric-value\t. %S)%s"
592                         value
593                         line-breaking))
594         (setq attributes (delq 'numeric-value attributes))
595         )))
596     (when (and (memq 'iso-10646-comment attributes)
597                (setq value (get-char-attribute char 'iso-10646-comment)))
598       (insert (format "(iso-10646-comment\t. %S)%s"
599                       value
600                       line-breaking))
601       (setq attributes (delq 'iso-10646-comment attributes))
602       )
603     (when (and (memq 'morohashi-daikanwa attributes)
604                (setq value (get-char-attribute char 'morohashi-daikanwa)))
605       (insert (format "(morohashi-daikanwa\t%s)%s"
606                       (mapconcat (function prin1-to-string) value " ")
607                       line-breaking))
608       (setq attributes (delq 'morohashi-daikanwa attributes))
609       )
610     (setq radical nil
611           strokes nil)
612     (when (and (memq 'ideographic-radical attributes)
613                (setq value (get-char-attribute char 'ideographic-radical)))
614       (setq radical value)
615       (insert (format "(ideographic-radical . %S)\t; %c%s"
616                       radical
617                       (aref ideographic-radicals radical)
618                       line-breaking))
619       (setq attributes (delq 'ideographic-radical attributes))
620       )
621     (let (key)
622       (dolist (domain char-db-feature-domains)
623         (setq key (intern (format "%s@%s" 'ideographic-radical domain)))
624         (when (and (memq key attributes)
625                    (setq value (get-char-attribute char key)))
626           (setq radical value)
627           (insert (format "(%s . %S)\t; %c%s"
628                           key
629                           radical
630                           (aref ideographic-radicals radical)
631                           line-breaking))
632           (setq attributes (delq key attributes))
633           )
634         (setq key (intern (format "%s@%s" 'ideographic-strokes domain)))
635         (when (and (memq key attributes)
636                    (setq value (get-char-attribute char key)))
637           (setq strokes value)
638           (insert (format "(%s . %S)%s"
639                           key
640                           strokes
641                           line-breaking))
642           (setq attributes (delq key attributes))
643           )
644         (setq key (intern (format "%s@%s" 'total-strokes domain)))
645         (when (and (memq key attributes)
646                    (setq value (get-char-attribute char key)))
647           (insert (format "(%s       . %S)%s"
648                           key
649                           value
650                           line-breaking))
651           (setq attributes (delq key attributes))
652           )
653         (dolist (feature '(ideographic-radical
654                            ideographic-strokes
655                            total-strokes))
656           (setq key (intern (format "%s@%s*sources" feature domain)))
657           (when (and (memq key attributes)
658                      (setq value (get-char-attribute char key)))
659             (insert (format "(%s%s" key line-breaking))
660             (dolist (cell value)
661               (insert (format " %s" cell)))
662             (insert ")")
663             (insert line-breaking)
664             (setq attributes (delq key attributes))
665             ))
666         ))
667     (when (and (memq 'ideographic-strokes attributes)
668                (setq value (get-char-attribute char 'ideographic-strokes)))
669       (setq strokes value)
670       (insert (format "(ideographic-strokes . %S)%s"
671                       strokes
672                       line-breaking))
673       (setq attributes (delq 'ideographic-strokes attributes))
674       )
675     (when (and (memq 'kangxi-radical attributes)
676                (setq value (get-char-attribute char 'kangxi-radical)))
677       (unless (eq value radical)
678         (insert (format "(kangxi-radical\t . %S)\t; %c%s"
679                         value
680                         (aref ideographic-radicals value)
681                         line-breaking))
682         (or radical
683             (setq radical value)))
684       (setq attributes (delq 'kangxi-radical attributes))
685       )
686     (when (and (memq 'kangxi-strokes attributes)
687                (setq value (get-char-attribute char 'kangxi-strokes)))
688       (unless (eq value strokes)
689         (insert (format "(kangxi-strokes\t . %S)%s"
690                         value
691                         line-breaking))
692         (or strokes
693             (setq strokes value)))
694       (setq attributes (delq 'kangxi-strokes attributes))
695       )
696     (when (and (memq 'japanese-radical attributes)
697                (setq value (get-char-attribute char 'japanese-radical)))
698       (unless (eq value radical)
699         (insert (format "(japanese-radical\t . %S)\t; %c%s"
700                         value
701                         (aref ideographic-radicals value)
702                         line-breaking))
703         (or radical
704             (setq radical value)))
705       (setq attributes (delq 'japanese-radical attributes))
706       )
707     (when (and (memq 'japanese-strokes attributes)
708                (setq value (get-char-attribute char 'japanese-strokes)))
709       (unless (eq value strokes)
710         (insert (format "(japanese-strokes\t . %S)%s"
711                         value
712                         line-breaking))
713         (or strokes
714             (setq strokes value)))
715       (setq attributes (delq 'japanese-strokes attributes))
716       )
717     (when (and (memq 'cns-radical attributes)
718                (setq value (get-char-attribute char 'cns-radical)))
719       (insert (format "(cns-radical\t . %S)\t; %c%s"
720                       value
721                       (aref ideographic-radicals value)
722                       line-breaking))
723       (setq attributes (delq 'cns-radical attributes))
724       )
725     (when (and (memq 'cns-strokes attributes)
726                (setq value (get-char-attribute char 'cns-strokes)))
727       (unless (eq value strokes)
728         (insert (format "(cns-strokes\t . %S)%s"
729                         value
730                         line-breaking))
731         (or strokes
732             (setq strokes value)))
733       (setq attributes (delq 'cns-strokes attributes))
734       )
735     (when (and (memq 'shinjigen-1-radical attributes)
736                (setq value (get-char-attribute char 'shinjigen-1-radical)))
737       (unless (eq value radical)
738         (insert (format "(shinjigen-1-radical . %S)\t; %c%s"
739                         value
740                         (aref ideographic-radicals value)
741                         line-breaking))
742         (or radical
743             (setq radical value)))
744       (setq attributes (delq 'shinjigen-1-radical attributes))
745       )
746     (when (and (memq 'ideographic- attributes)
747                (setq value (get-char-attribute char 'ideographic-)))
748       (insert "(ideographic-       ")
749       (setq lbs (concat "\n" (make-string (current-column) ?\ ))
750             separator nil)
751       (while (consp value)
752         (setq cell (car value))
753         (if (integerp cell)
754             (setq cell (decode-char '=ucs cell)))
755         (cond ((characterp cell)
756                (if separator
757                    (insert lbs))
758                (if readable
759                    (insert (format "%S" cell))
760                  (char-db-insert-char-spec cell readable))
761                (setq separator lbs))
762               ((consp cell)
763                (if separator
764                    (insert lbs))
765                (if (consp (car cell))
766                    (char-db-insert-char-spec cell readable)
767                  (char-db-insert-char-reference cell readable))
768                (setq separator lbs))
769               (t
770                (if separator
771                    (insert separator))
772                (insert (prin1-to-string cell))
773                (setq separator " ")))
774         (setq value (cdr value)))
775       (insert ")")
776       (insert line-breaking)
777       (setq attributes (delq 'ideographic- attributes)))
778     (when (and (memq 'total-strokes attributes)
779                (setq value (get-char-attribute char 'total-strokes)))
780       (insert (format "(total-strokes       . %S)%s"
781                       value
782                       line-breaking))
783       (setq attributes (delq 'total-strokes attributes))
784       )
785     (when (and (memq '->ideograph attributes)
786                (setq value (get-char-attribute char '->ideograph)))
787       (insert (format "(->ideograph\t%s)%s"
788                       (mapconcat (lambda (code)
789                                    (cond ((symbolp code)
790                                           (symbol-name code))
791                                          ((integerp code)
792                                           (format "#x%04X" code))
793                                          (t
794                                           (format "%s %S"
795                                                   line-breaking code))))
796                                  value " ")
797                       line-breaking))
798       (setq attributes (delq '->ideograph attributes))
799       )
800     (when (and (memq '->decomposition attributes)
801                (setq value (get-char-attribute char '->decomposition)))
802       (insert (format "(->decomposition\t%s)%s"
803                       (mapconcat (lambda (code)
804                                    (cond ((symbolp code)
805                                           (symbol-name code))
806                                          ((characterp code)
807                                           (if readable
808                                               (format "%S" code)
809                                             (format "#x%04X"
810                                                     (char-int code))
811                                             ))
812                                          ((integerp code)
813                                           (format "#x%04X" code))
814                                          (t
815                                           (format "%s%S" line-breaking code))))
816                                  value " ")
817                       line-breaking))
818       (setq attributes (delq '->decomposition attributes))
819       )
820     (if (equal (get-char-attribute char '->titlecase)
821                (get-char-attribute char '->uppercase))
822         (setq attributes (delq '->titlecase attributes)))
823     (when (and (memq '->mojikyo attributes)
824                (setq value (get-char-attribute char '->mojikyo)))
825       (insert (format "(->mojikyo\t\t. %06d)\t; %c%s"
826                       value (decode-char 'mojikyo value)
827                       line-breaking))
828       (setq attributes (delq '->mojikyo attributes))
829       )
830     (when (and (memq 'hanyu-dazidian-vol attributes)
831                (setq value (get-char-attribute char 'hanyu-dazidian-vol)))
832       (insert (format "(hanyu-dazidian-vol  . %d)%s"
833                       value line-breaking))
834       (setq attributes (delq 'hanyu-dazidian-vol attributes))
835       )
836     (when (and (memq 'hanyu-dazidian-page attributes)
837                (setq value (get-char-attribute char 'hanyu-dazidian-page)))
838       (insert (format "(hanyu-dazidian-page . %d)%s"
839                       value line-breaking))
840       (setq attributes (delq 'hanyu-dazidian-page attributes))
841       )
842     (when (and (memq 'hanyu-dazidian-char attributes)
843                (setq value (get-char-attribute char 'hanyu-dazidian-char)))
844       (insert (format "(hanyu-dazidian-char . %d)%s"
845                       value line-breaking))
846       (setq attributes (delq 'hanyu-dazidian-char attributes))
847       )
848     (unless readable
849       (dolist (ignored '(composition
850                          ->denotational <-subsumptive ->ucs-unified
851                          ->ideographic-component-forms))
852         (setq attributes (delq ignored attributes))))
853     (while attributes
854       (setq name (car attributes))
855       (if (setq value (get-char-attribute char name))
856           (cond ((setq ret (find-charset name))
857                  (setq name (charset-name ret))
858                  (if (and (not (memq name dest-ccss))
859                           (prog1
860                               (setq value (get-char-attribute char name))
861                             (setq dest-ccss (cons name dest-ccss))))
862                      (char-db-insert-ccs-feature name value line-breaking)
863                      )
864                  )
865                 ((string-match "^=>ucs@" (symbol-name name))
866                  (insert (format "(%-18s . #x%04X)\t; %c%s"
867                                  name value (decode-char '=ucs value)
868                                  line-breaking))
869                  )
870                 ((eq name 'jisx0208-1978/4X)
871                  (insert (format "(%-18s . #x%04X)%s"
872                                  name value
873                                  line-breaking))
874                  )
875                 ((and (not readable)
876                       (null (get-char-attribute
877                              char
878                              (intern (format "%s*sources" name))))
879                       (not (string-match "\\*sources$" (symbol-name name)))
880                       (or (eq name '<-identical)
881                           (string-match "^->simplified" (symbol-name name))
882                           ;; (string-match "^<-same" (symbol-name name))
883                           (string-match "^->same" (symbol-name name))
884                           ;; (string-match "^->ideographic-same" (symbol-name name))
885                           (string-match "^->vulgar" (symbol-name name))
886                           (string-match "^->wrong" (symbol-name name))
887                           (string-match "^->original" (symbol-name name))
888                           (string-match "^->ancient" (symbol-name name))
889                           ))
890                  )
891                 ((or (eq name 'ideographic-structure)
892                      (eq name 'ideographic-)
893                      (string-match "^\\(->\\|<-\\)" (symbol-name name)))
894                  (insert (format "(%-18s%s " name line-breaking))
895                  (setq lbs (concat "\n" (make-string (current-column) ?\ ))
896                        separator nil)
897                  (while (consp value)
898                    (setq cell (car value))
899                    (if (integerp cell)
900                        (setq cell (decode-char '=ucs cell)))
901                    (cond ((eq name '->subsumptive)
902                           (when (or (not
903                                      (some (lambda (atr)
904                                              (get-char-attribute cell atr))
905                                            char-db-ignored-attributes))
906                                     (some (lambda (ccs)
907                                             (encode-char cell ccs
908                                                          'defined-only))
909                                           ccss))
910                             (if separator
911                                 (insert lbs))
912                             (let ((char-db-ignored-attributes
913                                    (cons '<-subsumptive
914                                          char-db-ignored-attributes)))
915                               (insert-char-attributes cell readable))
916                             (setq separator lbs))
917                           )
918                          ((characterp cell)
919                           (setq sources
920                                 (get-char-attribute
921                                  char
922                                  (intern (format "%s*sources" name))))
923                           (setq required-features nil)
924                           (dolist (source sources)
925                             (cond
926                              ((memq source '(JP JP/Jouyou
927                                                 shinjigen-1))
928                               (setq required-features
929                                     (union required-features
930                                            '(=jis-x0208
931                                              =jis-x0208-1990
932                                              =jis-x0213-1-2000
933                                              =jis-x0213-2-2000
934                                              =jis-x0212
935                                              =jis-x0208-1983
936                                              =jis-x0208-1978))))
937                              ((eq source 'CN)
938                               (setq required-features
939                                     (union required-features
940                                            '(=gb2312
941                                              =gb12345
942                                              =iso-ir165)))))
943                             (cond
944                              ((find-charset
945                                (setq ret (intern (format "=%s" source))))
946                               (setq required-features
947                                     (cons ret required-features)))
948                              (t (setq required-features
949                                       (cons source required-features)))))
950                           (cond ((string-match "@JP" (symbol-name name))
951                                  (setq required-features
952                                        (union required-features
953                                               '(=jis-x0208
954                                                 =jis-x0208-1990
955                                                 =jis-x0213-1-2000
956                                                 =jis-x0213-2-2000
957                                                 =jis-x0212
958                                                 =jis-x0208-1983
959                                                 =jis-x0208-1978))))
960                                 ((string-match "@CN" (symbol-name name))
961                                  (setq required-features
962                                        (union required-features
963                                               '(=gb2312
964                                                 =gb12345
965                                                 =iso-ir165)))))
966                           (if separator
967                               (insert lbs))
968                           (if readable
969                               (insert (format "%S" cell))
970                             (char-db-insert-char-spec cell readable
971                                                       nil
972                                                       required-features))
973                           (setq separator lbs))
974                          ((consp cell)
975                           (if separator
976                               (insert lbs))
977                           (if (consp (car cell))
978                               (char-db-insert-char-spec cell readable)
979                             (char-db-insert-char-reference cell readable))
980                           (setq separator lbs))
981                          (t
982                           (if separator
983                               (insert separator))
984                           (insert (prin1-to-string cell))
985                           (setq separator " ")))
986                    (setq value (cdr value)))
987                  (insert ")")
988                  (insert line-breaking))
989                 ((memq name '(ideograph=
990                               original-ideograph-of
991                               ancient-ideograph-of
992                               vulgar-ideograph-of
993                               wrong-ideograph-of
994                               ;; simplified-ideograph-of
995                               ideographic-variants
996                               ;; ideographic-different-form-of
997                               ))
998                  (insert (format "(%-18s%s " name line-breaking))
999                  (setq lbs (concat "\n" (make-string (current-column) ?\ ))
1000                        separator nil)
1001                  (while (consp value)
1002                    (setq cell (car value))
1003                    (if (and (consp cell)
1004                             (consp (car cell)))
1005                        (progn
1006                          (if separator
1007                              (insert lbs))
1008                          (char-db-insert-alist cell readable)
1009                          (setq separator lbs))
1010                      (if separator
1011                          (insert separator))
1012                      (insert (prin1-to-string cell))
1013                      (setq separator " "))
1014                    (setq value (cdr value)))
1015                  (insert ")")
1016                  (insert line-breaking))
1017                 ((consp value)
1018                  (insert (format "(%-18s " name))
1019                  (setq lbs (concat "\n" (make-string (current-column) ?\ ))
1020                        separator nil)
1021                  (while (consp value)
1022                    (setq cell (car value))
1023                    (if (and (consp cell)
1024                             (consp (car cell))
1025                             (setq ret (condition-case nil
1026                                           (find-char cell)
1027                                         (error nil))))
1028                        (progn
1029                          (setq rest cell
1030                                al nil
1031                                cal nil)
1032                          (while rest
1033                            (setq key (car (car rest)))
1034                            (if (find-charset key)
1035                                (setq cal (cons key cal))
1036                              (setq al (cons key al)))
1037                            (setq rest (cdr rest)))
1038                          (if separator
1039                              (insert lbs))
1040                          (insert-char-attributes ret
1041                                                  readable
1042                                                  al cal)
1043                          (setq separator lbs))
1044                      (setq ret (prin1-to-string cell))
1045                      (if separator
1046                          (if (< (+ (current-column)
1047                                    (length ret)
1048                                    (length separator))
1049                                 76)
1050                              (insert separator)
1051                            (insert lbs)))
1052                      (insert ret)
1053                      (setq separator " "))
1054                    (setq value (cdr value)))
1055                  (insert ")")
1056                  (insert line-breaking))
1057                 (t
1058                  (insert (format "(%-18s . %S)%s"
1059                                  name value
1060                                  line-breaking)))
1061                 ))
1062       (setq attributes (cdr attributes)))
1063     (insert ")")))
1064
1065 (defun insert-char-data (char &optional readable
1066                               attributes)
1067   (save-restriction
1068     (narrow-to-region (point)(point))
1069     (insert "(define-char
1070   '")
1071     (insert-char-attributes char readable attributes)
1072     (insert ")\n")
1073     (goto-char (point-min))
1074     (while (re-search-forward "[ \t]+$" nil t)
1075       (replace-match ""))
1076     ;; from tabify.
1077     (goto-char (point-min))
1078     (while (re-search-forward "[ \t][ \t][ \t]*" nil t)
1079       (let ((column (current-column))
1080             (indent-tabs-mode t))
1081         (delete-region (match-beginning 0) (point))
1082         (indent-to column)))
1083     (goto-char (point-max))
1084     ;; (tabify (point-min)(point-max))
1085     ))
1086
1087 (defun insert-char-data-with-variant (char &optional printable
1088                                            no-ucs-unified
1089                                            script excluded-script)
1090   (insert-char-data char printable)
1091   (let ((variants (or (char-variants char)
1092                       (let ((ucs (get-char-attribute char '->ucs)))
1093                         (if ucs
1094                             (delete char (char-variants (int-char ucs)))))))
1095         variant vs)
1096     (setq variants (sort variants #'<))
1097     (while variants
1098       (setq variant (car variants))
1099       (if (and (or (null script)
1100                    (null (setq vs (get-char-attribute variant 'script)))
1101                    (memq script vs))
1102                (or (null excluded-script)
1103                    (null (setq vs (get-char-attribute variant 'script)))
1104                    (not (memq excluded-script vs))))
1105           (or (and no-ucs-unified (get-char-attribute variant '=ucs))
1106               (insert-char-data variant printable)))
1107       (setq variants (cdr variants))
1108       )))
1109
1110 (defun insert-char-range-data (min max &optional script excluded-script)
1111   (let ((code min)
1112         char)
1113     (while (<= code max)
1114       (setq char (decode-char '=ucs code))
1115       (if (encode-char char '=ucs 'defined-only)
1116           (insert-char-data-with-variant char nil 'no-ucs-unified
1117                                          script excluded-script))
1118       (setq code (1+ code)))))
1119
1120 (defun write-char-range-data-to-file (min max file
1121                                           &optional script excluded-script)
1122   (let ((coding-system-for-write char-db-file-coding-system))
1123     (with-temp-buffer
1124       (insert (format ";; -*- coding: %s -*-\n"
1125                       char-db-file-coding-system))
1126       (insert-char-range-data min max script excluded-script)
1127       (write-region (point-min)(point-max) file))))
1128
1129 (defvar what-character-original-window-configuration)
1130
1131 ;;;###autoload
1132 (defun what-char-definition (char)
1133   (interactive (list (char-after)))
1134   (let ((buf (get-buffer-create "*Character Description*"))
1135         (the-buf (current-buffer))
1136         (win-conf (current-window-configuration)))
1137     (pop-to-buffer buf)
1138     (make-local-variable 'what-character-original-window-configuration)
1139     (setq what-character-original-window-configuration win-conf)
1140     (setq buffer-read-only nil)
1141     (erase-buffer)
1142     (condition-case err
1143         (progn
1144           (insert-char-data-with-variant char 'printable)
1145           (unless (char-attribute-alist char)
1146             (insert (format ";; = %c\n"
1147                             (let* ((rest (split-char char))
1148                                    (ccs (pop rest))
1149                                    (code (pop rest)))
1150                               (while rest
1151                                 (setq code (logior (lsh code 8)
1152                                                    (pop rest))))
1153                               (decode-char ccs code)))))
1154           ;; (char-db-update-comment)
1155           (set-buffer-modified-p nil)
1156           (view-mode the-buf (lambda (buf)
1157                                (set-window-configuration
1158                                 what-character-original-window-configuration)
1159                                ))
1160           (goto-char (point-min)))
1161       (error (progn
1162                (set-window-configuration
1163                 what-character-original-window-configuration)
1164                (signal (car err) (cdr err)))))))
1165
1166 (provide 'char-db-util)
1167
1168 ;;; char-db-util.el ends here