;;; char-db-util.el --- Character Database utility
-;; Copyright (C) 1998,1999,2000,2001 MORIOKA Tomohiko.
+;; Copyright (C) 1998,1999,2000,2001,2002,2003 MORIOKA Tomohiko.
;; Author: MORIOKA Tomohiko <tomo@kanji.zinbun.kyoto-u.ac.jp>
;; Keywords: UTF-2000, ISO/IEC 10646, Unicode, UCS-4, MULE.
(let ((v (make-vector 215 nil))
(i 1))
(while (< i 215)
- (aset v i (int-char (+ #x2EFF i)))
+ (aset v i (decode-char 'ucs (+ #x2EFF i)))
(setq i (1+ i)))
- (if (< (charset-iso-final-char (car (split-char (aref v 34)))) ?0)
- (aset v 34 (make-char 'chinese-gb2312 #x62 #x3A)))
+ ;; (unless (charset-iso-final-char (car (split-char (aref v 34))))
+ ;; (aset v 34 (make-char 'chinese-gb2312 #x62 #x3A)))
v))
+(defvar char-db-ignored-attributes nil)
+
(defun char-attribute-name< (ka kb)
(cond
((find-charset ka)
((= (charset-dimension ka)
(charset-dimension kb))
(cond ((= (charset-chars ka)(charset-chars kb))
- (cond
- ((>= (charset-iso-final-char ka) ?@)
- (if (>= (charset-iso-final-char kb) ?@)
- (< (charset-iso-final-char ka)
- (charset-iso-final-char kb))
- t))
- ((>= (charset-iso-final-char ka) ?0)
- (cond
- ((>= (charset-iso-final-char kb) ?@)
- nil)
- ((>= (charset-iso-final-char kb) ?0)
- (< (charset-iso-final-char ka)
- (charset-iso-final-char kb)))
- (t)))
- (t (if (>= (charset-iso-final-char kb) ?0)
- nil
- (> (charset-id ka)(charset-id kb))))))
+ (if (charset-iso-final-char ka)
+ (cond
+ ((>= (charset-iso-final-char ka) ?@)
+ (if (and (charset-iso-final-char kb)
+ (>= (charset-iso-final-char kb) ?@))
+ (< (charset-iso-final-char ka)
+ (charset-iso-final-char kb))
+ t))
+ (t
+ (if (charset-iso-final-char kb)
+ (if (>= (charset-iso-final-char kb) ?@)
+ nil
+ (< (charset-iso-final-char ka)
+ (charset-iso-final-char kb)))
+ t)))
+ (if (charset-iso-final-char kb)
+ nil
+ (< (charset-id ka)(charset-id kb)))))
((<= (charset-chars ka)(charset-chars kb)))))
(t
(< (charset-dimension ka)
((symbolp kb)
nil)))
-(defun char-db-insert-char-spec (char &optional readable column)
- (unless column
- (setq column (current-column)))
- (let (char-spec ret al cal key)
+(defvar char-db-coded-charset-priority-list
+ (let ((rest default-coded-charset-priority-list)
+ dest)
+ (while rest
+ (when (symbolp (car rest))
+ (cond ((memq (car rest)
+ '(latin-viscii-lower
+ latin-viscii-upper
+ ipa
+ lao
+ ethiopic
+ arabic-digit
+ arabic-1-column
+ arabic-2-column)))
+ ((string-match "^ideograph-cbeta" (symbol-name (car rest))))
+ ((string-match "^china3-jef" (symbol-name (car rest))))
+ ((string-match "^chinese-big5" (symbol-name (car rest))))
+ ((string-match "^ideograph-gt-pj-" (symbol-name (car rest)))
+ (unless (memq 'ideograph-gt dest)
+ (setq dest (cons 'ideograph-gt dest))))
+ (t
+ (setq dest (cons (car rest) dest)))))
+ (setq rest (cdr rest)))
+ (append (sort dest #'char-attribute-name<)
+ '(chinese-big5-cdp ideograph-cbeta china3-jef
+ chinese-big5-eten chinese-big5))))
+
+(defun char-db-make-char-spec (char)
+ (let (ret char-spec)
(cond ((characterp char)
- (cond ((setq ret (get-char-attribute char 'ucs))
+ (cond ((and (setq ret (get-char-attribute char 'ucs))
+ (not (and (<= #xE000 ret)(<= ret #xF8FF))))
(setq char-spec (list (cons 'ucs ret)))
- (if (setq ret (get-char-attribute char 'name))
- (setq char-spec (cons (cons 'name ret) char-spec)))
+ (cond ((setq ret (get-char-attribute char 'name))
+ (setq char-spec (cons (cons 'name ret) char-spec))
+ )
+ ((setq ret (get-char-attribute char 'name*))
+ (setq char-spec (cons (cons 'name* ret) char-spec))
+ ))
)
- ((setq ret (split-char char))
+ ((setq ret
+ (catch 'tag
+ (let ((rest char-db-coded-charset-priority-list))
+ (while rest
+ (if (setq ret
+ (get-char-attribute char (car rest)))
+ (throw 'tag (cons (car rest) ret)))
+ (setq rest (cdr rest))))))
(setq char-spec (list ret))
(dolist (ccs (delq (car ret) (charset-list)))
- (if (and (>= (charset-iso-final-char ccs) ?0)
- (setq ret (get-char-attribute char ccs)))
+ (if (or (and (charset-iso-final-char ccs)
+ (setq ret (get-char-attribute char ccs)))
+ (eq ccs 'ideograph-daikanwa))
(setq char-spec (cons (cons ccs ret) char-spec))))
- (if (setq ret (get-char-attribute char 'name))
- (setq char-spec (cons (cons 'name ret) char-spec)))
- )))
+ (if (null char-spec)
+ (setq char-spec (split-char char)))
+ (cond ((setq ret (get-char-attribute char 'name))
+ (setq char-spec (cons (cons 'name ret) char-spec))
+ )
+ ((setq ret (get-char-attribute char 'name*))
+ (setq char-spec (cons (cons 'name* ret) char-spec))
+ ))
+ ))
+ char-spec)
((consp char)
- (setq char-spec char)
- (setq char nil)))
- (if (or char
- (setq char (condition-case nil
- (define-char char-spec)
- (error nil))))
- (progn
- (setq al nil
- cal nil)
- (while char-spec
- (setq key (car (car char-spec)))
- (if (find-charset key)
- (setq cal (cons key cal))
- (setq al (cons key al)))
- (setq char-spec (cdr char-spec)))
- (insert-char-attributes char
- readable
- (or al 'none) cal))
- (insert (prin1-to-string char-spec)))))
+ char))))
+
+(defun char-db-insert-char-spec (char &optional readable column)
+ (unless column
+ (setq column (current-column)))
+ (let (char-spec ret al cal key temp-char)
+ (setq char-spec (char-db-make-char-spec char))
+ (unless (or (characterp char) ; char
+ (condition-case nil
+ (setq char (find-char char-spec))
+ (error nil)))
+ ;; define temporary character
+ ;; Current implementation is dirty.
+ (setq temp-char (define-char (cons '(ideograph-daikanwa . 0)
+ char-spec)))
+ (remove-char-attribute temp-char 'ideograph-daikanwa)
+ (setq char temp-char))
+ (setq al nil
+ cal nil)
+ (while char-spec
+ (setq key (car (car char-spec)))
+ (unless (memq key char-db-ignored-attributes)
+ (if (find-charset key)
+ (if (get-char-attribute char key)
+ (setq cal (cons key cal)))
+ (setq al (cons key al))))
+ (setq char-spec (cdr char-spec)))
+ (unless cal
+ (setq char-spec (char-db-make-char-spec char))
+ (while char-spec
+ (setq key (car (car char-spec)))
+ (unless (memq key char-db-ignored-attributes)
+ (if (find-charset key)
+ (setq cal (cons key cal))
+ (setq al (cons key al))))
+ (setq char-spec (cdr char-spec)))
+ )
+ (unless (or cal
+ (memq 'ideographic-structure al))
+ (push 'ideographic-structure al))
+ (insert-char-attributes char
+ readable
+ (or al 'none) cal)
+ (when temp-char
+ ;; undefine temporary character
+ ;; Current implementation is dirty.
+ (setq char-spec (char-attribute-alist temp-char))
+ (while char-spec
+ (remove-char-attribute temp-char (car (car char-spec)))
+ (setq char-spec (cdr char-spec))))))
(defun char-db-insert-alist (alist &optional readable column)
(unless column
(cond ((eq name 'char)
(insert "(char . ")
(if (setq ret (condition-case nil
- (define-char value)
+ (find-char value)
(error nil)))
(progn
(setq al nil
(if (and (consp cell)
(consp (car cell))
(setq ret (condition-case nil
- (define-char cell)
+ (find-char cell)
(error nil)))
)
(progn
(setq column (current-column)))
(let ((line-breaking
(concat "\n" (make-string (1+ column) ?\ )))
+ (separator "")
name value)
(insert "(")
(while plist
(setq name (pop plist))
(setq value (pop plist))
(cond ((eq name :char)
+ (insert separator)
(insert ":char\t")
- (if (numberp value)
- (setq value (decode-char 'ucs value)))
+ (cond ((numberp value)
+ (setq value (decode-char 'ucs value)))
+ ;; ((consp value)
+ ;; (setq value (or (find-char value)
+ ;; value)))
+ )
(char-db-insert-char-spec value readable)
- (insert line-breaking))
- (t
- (insert (format "%s\t%S%s"
+ (insert line-breaking)
+ (setq separator ""))
+ ((eq name :radical)
+ (insert (format "%s%s\t%d ; %c%s"
+ separator
name value
- line-breaking))))
+ (aref ideographic-radicals value)
+ line-breaking))
+ (setq separator ""))
+ (t
+ (insert (format "%s%s\t%S" separator name value))
+ (setq separator line-breaking)))
))
(insert ")"))
(defun insert-char-attributes (char &optional readable
attributes ccs-attributes
column)
- (setq attributes
- (sort (if attributes
- (if (consp attributes)
- (copy-sequence attributes))
- (char-attribute-list))
- #'char-attribute-name<))
- (setq ccs-attributes
- (sort (if ccs-attributes
- (copy-sequence ccs-attributes)
- (charset-list))
- #'char-attribute-name<))
+ (let (atr-d ccs-d)
+ (setq attributes
+ (sort (if attributes
+ (if (consp attributes)
+ (progn
+ (dolist (name attributes)
+ (unless (memq name char-db-ignored-attributes)
+ (push name atr-d)))
+ atr-d))
+ (dolist (name (char-attribute-list))
+ (unless (memq name char-db-ignored-attributes)
+ (if (find-charset name)
+ (push name ccs-d)
+ (push name atr-d))))
+ atr-d)
+ #'char-attribute-name<))
+ (setq ccs-attributes
+ (sort (if ccs-attributes
+ (progn
+ (setq ccs-d nil)
+ (dolist (name ccs-attributes)
+ (unless (memq name char-db-ignored-attributes)
+ (push name ccs-d)))
+ ccs-d)
+ (or ccs-d
+ (progn
+ (dolist (name (charset-list))
+ (unless (memq name char-db-ignored-attributes)
+ (push name ccs-d)))
+ ccs-d)))
+ #'char-attribute-name<)))
(unless column
(setq column (current-column)))
(let (name value has-long-ccs-name rest
value line-breaking))
(setq attributes (delq 'name attributes))
)
+ (when (and (memq 'name* attributes)
+ (setq value (get-char-attribute char 'name*)))
+ (insert (format
+ (if (> (+ (current-column) (length value)) 48)
+ "(name* . %S)%s"
+ "(name* . %S)%s")
+ value line-breaking))
+ (setq attributes (delq 'name* attributes))
+ )
(when (and (memq 'script attributes)
(setq value (get-char-attribute char 'script)))
(insert (format "(script\t\t%s)%s"
line-breaking))
(setq attributes (delq '=>ucs attributes))
)
+ (when (and (memq '=>ucs* attributes)
+ (setq value (get-char-attribute char '=>ucs*)))
+ (insert (format "(=>ucs*\t\t. #x%04X)\t; %c%s"
+ value (decode-char 'ucs value)
+ line-breaking))
+ (setq attributes (delq '=>ucs* attributes))
+ )
+ (when (and (memq '=>ucs-gb attributes)
+ (setq value (get-char-attribute char '=>ucs-gb)))
+ (insert (format "(=>ucs-gb\t\t. #x%04X)\t; %c%s"
+ value (decode-char 'ucs value)
+ line-breaking))
+ (setq attributes (delq '=>ucs-gb attributes))
+ )
+ (when (and (memq '=>ucs-cns attributes)
+ (setq value (get-char-attribute char '=>ucs-cns)))
+ (insert (format "(=>ucs-cns\t\t. #x%04X)\t; %c%s"
+ value (decode-char 'ucs-cns value)
+ line-breaking))
+ (setq attributes (delq '=>ucs-cns attributes))
+ )
+ (when (and (memq '=>ucs-jis attributes)
+ (setq value (get-char-attribute char '=>ucs-jis)))
+ (insert (format "(=>ucs-jis\t\t. #x%04X)\t; %c%s"
+ value (decode-char 'ucs-jis value)
+ line-breaking))
+ (setq attributes (delq '=>ucs-jis attributes))
+ )
+ (when (and (memq '=>ucs-ks attributes)
+ (setq value (get-char-attribute char '=>ucs-ks)))
+ (insert (format "(=>ucs-ks\t\t. #x%04X)\t; %c%s"
+ value (decode-char 'ucs-ks value)
+ line-breaking))
+ (setq attributes (delq '=>ucs-ks attributes))
+ )
(when (and (memq '->ucs attributes)
(setq value (get-char-attribute char '->ucs)))
(insert (format (if char-db-convert-obsolete-format
(setq radical value)))
(setq attributes (delq 'shinjigen-1-radical attributes))
)
+ (when (and (memq 'ideographic- attributes)
+ (setq value (get-char-attribute char 'ideographic-)))
+ (insert "(ideographic- ")
+ (setq lbs (concat "\n" (make-string (current-column) ?\ ))
+ separator nil)
+ (while (consp value)
+ (setq cell (car value))
+ (if (integerp cell)
+ (setq cell (decode-char 'ucs cell)))
+ (cond ((characterp cell)
+ (if separator
+ (insert lbs))
+ (if readable
+ (insert (format "%S" cell))
+ (char-db-insert-char-spec cell readable))
+ (setq separator lbs))
+ ((consp cell)
+ (if separator
+ (insert lbs))
+ (if (consp (car cell))
+ (char-db-insert-char-spec cell readable)
+ (char-db-insert-char-reference cell readable))
+ (setq separator lbs))
+ (t
+ (if separator
+ (insert separator))
+ (insert (prin1-to-string cell))
+ (setq separator " ")))
+ (setq value (cdr value)))
+ (insert ")")
+ (insert line-breaking)
+ (setq attributes (delq 'ideographic- attributes)))
(when (and (memq 'total-strokes attributes)
(setq value (get-char-attribute char 'total-strokes)))
(insert (format "(total-strokes . %S)%s"
line-breaking))
(setq attributes (delq '->mojikyo attributes))
)
+ (when (and (memq 'hanyu-dazidian-vol attributes)
+ (setq value (get-char-attribute char 'hanyu-dazidian-vol)))
+ (insert (format "(hanyu-dazidian-vol . %d)%s"
+ value line-breaking))
+ (setq attributes (delq 'hanyu-dazidian-vol attributes))
+ )
+ (when (and (memq 'hanyu-dazidian-page attributes)
+ (setq value (get-char-attribute char 'hanyu-dazidian-page)))
+ (insert (format "(hanyu-dazidian-page . %d)%s"
+ value line-breaking))
+ (setq attributes (delq 'hanyu-dazidian-page attributes))
+ )
+ (when (and (memq 'hanyu-dazidian-char attributes)
+ (setq value (get-char-attribute char 'hanyu-dazidian-char)))
+ (insert (format "(hanyu-dazidian-char . %d)%s"
+ value line-breaking))
+ (setq attributes (delq 'hanyu-dazidian-char attributes))
+ )
+ (unless readable
+ (when (memq '->ucs-variants attributes)
+ (setq attributes (delq '->ucs-variants attributes))
+ )
+ (when (memq 'composition attributes)
+ (setq attributes (delq 'composition attributes))
+ ))
(setq rest ccs-attributes)
(while (and rest
(progn
(insert (format "(%-18s . #x%04X)%s"
name value
line-breaking)))
- ((memq name '(->lowercase
- ->uppercase ->titlecase
- ->fullwidth <-fullwidth
- ->vulgar-ideograph <-vulgar-ideograph
- ->ancient-ideograph <-ancient-ideograph
- ->original-ideograph <-original-ideograph
- ->simplified-ideograph <-simplified-ideograph
- ->same-ideograph
- ->ideographic-variants
- ->synonyms
- ->radical <-radical
- ->bopomofo <-bopomofo
- ->ideographic <-ideographic
- ideographic-structure))
+ ((or (eq name 'ideographic-structure)
+ (eq name 'ideographic-)
+ (string-match "^\\(->\\|<-\\)" (symbol-name name)))
(insert (format "(%-18s%s " name line-breaking))
(setq lbs (concat "\n" (make-string (current-column) ?\ ))
separator nil)
(cond ((characterp cell)
(if separator
(insert lbs))
- (char-db-insert-char-spec cell readable)
+ (if readable
+ (insert (format "%S" cell))
+ (char-db-insert-char-spec cell readable))
(setq separator lbs))
((consp cell)
(if separator
(setq value (cdr value)))
(insert ")")
(insert line-breaking))
- ((string-match "^->" (symbol-name name))
- (insert
- (format "(%-18s %s)%s"
- name
- (mapconcat (lambda (code)
- (cond ((symbolp code)
- (symbol-name code))
- ((integerp code)
- (format "#x%04X" code))
- (t
- (format "%s%S"
- line-breaking code))))
- value " ")
- line-breaking)))
+ ;; ((string-match "^->" (symbol-name name))
+ ;; (insert
+ ;; (format "(%-18s %s)%s"
+ ;; name
+ ;; (mapconcat (lambda (code)
+ ;; (cond ((symbolp code)
+ ;; (symbol-name code))
+ ;; ((integerp code)
+ ;; (format "#x%04X" code))
+ ;; (t
+ ;; (format "%s%S"
+ ;; line-breaking code))))
+ ;; value " ")
+ ;; line-breaking)))
((consp value)
(insert (format "(%-18s " name))
(setq lbs (concat "\n" (make-string (current-column) ?\ ))
(if (and (consp cell)
(consp (car cell))
(setq ret (condition-case nil
- (define-char cell)
+ (find-char cell)
(error nil))))
(progn
(setq rest cell
(setq value (get-char-attribute char name)))
(insert
(format
- (cond ((memq name '(ideograph-daikanwa ideograph-gt))
+ (cond ((memq name '(ideograph-daikanwa-2
+ ideograph-daikanwa
+ ideograph-gt
+ ideograph-gt-k
+ ideograph-cbeta))
(if has-long-ccs-name
"(%-26s . %05d)\t; %c%s"
"(%-18s . %05d)\t; %c%s"))
(if has-long-ccs-name
"(%-26s . %06d)\t; %c%s"
"(%-18s . %06d)\t; %c%s"))
- ((eq name 'ucs)
+ ((>= (charset-dimension name) 2) ; (eq name 'ucs)
(if has-long-ccs-name
"(%-26s . #x%04X)\t; %c%s"
"(%-18s . #x%04X)\t; %c%s"))
(tabify (point-min)(point-max))
))
-;;;###autoload
-(defun char-db-update-comment ()
- (interactive)
- (save-excursion
- (goto-char (point-min))
- (let (cdef table char)
- (while (re-search-forward "^[ \t]*\\(([^.()]+)\\)" nil t)
- (goto-char (match-beginning 1))
- (setq cdef (read (current-buffer)))
- (when (find-charset (car cdef))
- (goto-char (match-end 0))
- (setq char
- (if (and
- (not (eq (car cdef) 'ideograph-daikanwa))
- (or (memq (car cdef) '(ascii latin-viscii-upper
- latin-viscii-lower
- arabic-iso8859-6
- japanese-jisx0213-1
- japanese-jisx0213-2))
- (= (char-int (charset-iso-final-char (car cdef)))
- 0)))
- (apply (function make-char) cdef)
- (if (setq table (charset-mapping-table (car cdef)))
- (set-charset-mapping-table (car cdef) nil))
- (prog1
- (apply (function make-char) cdef)
- (if table
- (set-charset-mapping-table (car cdef) table)))))
- (when (not (or (< (char-int char) 32)
- (and (<= 128 (char-int char))
- (< (char-int char) 160))))
- (delete-region (point) (point-at-eol))
- (insert (format "\t; %c" char)))
- )))))
-
(defun insert-char-data-with-variant (char &optional printable
no-ucs-variant
script excluded-script)
(condition-case err
(progn
(insert-char-data-with-variant char 'printable)
+ (unless (char-attribute-alist char)
+ (insert (format ";; = %c\n"
+ (let* ((rest (split-char char))
+ (ccs (pop rest))
+ (code (pop rest)))
+ (while rest
+ (setq code (logior (lsh code 8)
+ (pop rest))))
+ (decode-char ccs code)))))
;; (char-db-update-comment)
(set-buffer-modified-p nil)
(view-mode the-buf (lambda (buf)