;;; char-db-util.el --- Character Database utility
-;; Copyright (C) 1998,1999,2000,2001,2002 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)))
- (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)
nil)))
(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 "^mojikyo-" (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 chinese-big5-eten chinese-big5))))
+ '(ascii
+ control-1
+ latin-iso8859-1
+ latin-iso8859-2
+ latin-iso8859-3
+ latin-iso8859-4
+ latin-iso8859-9
+ latin-jisx0201
+ cyrillic-iso8859-5
+ greek-iso8859-7
+ thai-tis620
+ =jis-x0208
+ japanese-jisx0208
+ japanese-jisx0212
+ japanese-jisx0208-1978
+ chinese-gb2312
+ =cns11643-1
+ chinese-cns11643-2
+ chinese-cns11643-3
+ chinese-cns11643-4
+ chinese-cns11643-5
+ chinese-cns11643-6
+ chinese-cns11643-7
+ =jis-x0208-1990
+ =jis-x0213-1-2000
+ =jis-x0213-2-2000
+ korean-ksc5601
+ chinese-isoir165
+ katakana-jisx0201
+ hebrew-iso8859-8
+ chinese-gb12345
+ latin-viscii
+ ethiopic-ucs
+ =gt
+ ideograph-daikanwa-2
+ ideograph-daikanwa
+ =cbeta
+ ideograph-hanziku-1
+ ideograph-hanziku-2
+ ideograph-hanziku-3
+ ideograph-hanziku-4
+ ideograph-hanziku-5
+ ideograph-hanziku-6
+ ideograph-hanziku-7
+ ideograph-hanziku-8
+ ideograph-hanziku-9
+ ideograph-hanziku-10
+ ideograph-hanziku-11
+ ideograph-hanziku-12
+ =big5
+ =big5-eten
+ =big5-cdp
+ =gt-k
+ =jef-china3))
(defun char-db-make-char-spec (char)
(let (ret char-spec)
(cond ((characterp char)
- (cond ((and (setq ret (get-char-attribute char 'ucs))
+ (cond ((and (setq ret (encode-char char '=ucs 'defined-only))
(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)))
+ (setq char-spec (list (cons '=ucs ret)))
+ (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
- (let ((default-coded-charset-priority-list
- char-db-coded-charset-priority-list))
- (split-char char)))
+ (catch 'tag
+ (let ((rest char-db-coded-charset-priority-list)
+ ccs)
+ (while rest
+ (setq ccs (charset-name
+ (find-charset (car rest))))
+ (if (setq ret
+ (encode-char char ccs
+ 'defined-only))
+ (throw 'tag (cons ccs ret)))
+ (setq rest (cdr rest))))))
(setq char-spec (list ret))
(dolist (ccs (delq (car ret) (charset-list)))
- (if (or (and (charset-iso-final-char ccs)
- (setq ret (get-char-attribute char ccs)))
- (eq ccs 'ideograph-daikanwa))
+ (if (and (or (charset-iso-final-char ccs)
+ (memq ccs
+ '(ideograph-daikanwa
+ =daikanwa-rev2
+ ;; =gt-k
+ )))
+ (setq ret (encode-char char ccs 'defined-only)))
(setq char-spec (cons (cons ccs ret) char-spec))))
(if (null char-spec)
(setq char-spec (split-char char)))
- (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))
+ ))
))
char-spec)
((consp char)
(setq column (current-column)))
(let (char-spec ret al cal key temp-char)
(setq char-spec (char-db-make-char-spec char))
- ;; (cond ((characterp char)
- ;; (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)))
- ;; )
- ;; ((setq ret
- ;; (let ((default-coded-charset-priority-list
- ;; char-db-coded-charset-priority-list))
- ;; (split-char char)))
- ;; (setq char-spec (list ret))
- ;; (dolist (ccs (delq (car ret) (charset-list)))
- ;; (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 (null char-spec)
- ;; (setq char-spec (split-char char)))
- ;; (if (setq ret (get-char-attribute char 'name))
- ;; (setq char-spec (cons (cons 'name ret) char-spec)))
- ;; )))
- ;; ((consp char)
- ;; (setq char-spec char)
- ;; (setq char nil)))
(unless (or (characterp char) ; char
(condition-case nil
(setq char (find-char 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)
+ (if (encode-char char key 'defined-only)
(setq cal (cons key cal)))
(setq al (cons key al))))
(setq char-spec (cdr char-spec)))
(insert separator)
(insert ":char\t")
(cond ((numberp value)
- (setq value (decode-char 'ucs value)))
+ (setq value (decode-char '=ucs value)))
;; ((consp value)
;; (setq value (or (find-char value)
;; value)))
(setq ret
(cond ((eq ccs 'arabic-iso8859-6)
(decode-char ccs code-point))
- ((and (memq ccs '(ideograph-gt-pj-1
- ideograph-gt-pj-2
- ideograph-gt-pj-3
- ideograph-gt-pj-4
- ideograph-gt-pj-5
- ideograph-gt-pj-6
- ideograph-gt-pj-7
- ideograph-gt-pj-8
- ideograph-gt-pj-9
- ideograph-gt-pj-10
- ideograph-gt-pj-11))
+ ((and (memq ccs '(=gt-pj-1
+ =gt-pj-2
+ =gt-pj-3
+ =gt-pj-4
+ =gt-pj-5
+ =gt-pj-6
+ =gt-pj-7
+ =gt-pj-8
+ =gt-pj-9
+ =gt-pj-10
+ =gt-pj-11))
(setq ret (decode-char ccs code-point))
- (setq ret (get-char-attribute ret 'ideograph-gt)))
- (decode-builtin-char 'ideograph-gt ret))
+ (setq ret (encode-char ret '=gt 'defined-only)))
+ (decode-builtin-char '=gt ret))
(t
(decode-builtin-char ccs code-point))))
(cond ((and (<= 0 (char-int ret))
(<= (char-int ret) #x1F))
- (decode-char 'ucs (+ #x2400 (char-int ret))))
+ (decode-char '=ucs (+ #x2400 (char-int ret))))
((= (char-int ret) #x7F)
?\u2421)
(t ret))))
(line-breaking
(concat "\n" (make-string (1+ column) ?\ )))
lbs cell separator ret
- key al cal)
+ key al cal
+ dest-ccss)
(insert "(")
(when (and (memq 'name attributes)
(setq value (get-char-attribute char 'name)))
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 'script 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* 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))
- )
+ (dolist (name '(=>ucs =>ucs*))
+ (when (and (memq name attributes)
+ (setq value (get-char-attribute char name)))
+ (insert (format "(%-18s . #x%04X)\t; %c%s"
+ name value (decode-char '=ucs value)
+ line-breaking))
+ (setq attributes (delq name attributes))))
+ (dolist (name '(=>ucs@gb =>ucs@cns =>ucs@jis =>ucs@ks =>ucs@big5))
+ (when (and (memq name attributes)
+ (setq value (get-char-attribute char name)))
+ (insert (format "(%-18s . #x%04X)\t; %c%s"
+ name value
+ (decode-char (intern
+ (concat "="
+ (substring
+ (symbol-name name) 2)))
+ value)
+ line-breaking))
+ (setq attributes (delq name attributes))
+ ))
+ (dolist (name '(=>ucs-gb =>ucs-cns =>ucs-jis =>ucs-ks =>ucs-big5))
+ (when (and (memq name attributes)
+ (setq value (get-char-attribute char name)))
+ (insert (format "(%-18s . #x%04X)\t; %c%s"
+ (intern
+ (concat "=>ucs@"
+ (substring (symbol-name name) 6)))
+ value
+ (decode-char (intern
+ (concat "=ucs@"
+ (substring
+ (symbol-name name) 6)))
+ value)
+ line-breaking))
+ (setq attributes (delq name attributes))))
(when (and (memq '->ucs attributes)
(setq value (get-char-attribute char '->ucs)))
(insert (format (if char-db-convert-obsolete-format
"(=>ucs\t\t. #x%04X)\t; %c%s"
"(->ucs\t\t. #x%04X)\t; %c%s")
- value (decode-char 'ucs value)
+ value (decode-char '=ucs value)
line-breaking))
(setq attributes (delq '->ucs attributes))
)
)
(setq radical nil
strokes nil)
+ (let (key)
+ (dolist (domain '(ucs daikanwa cns))
+ (setq key (intern (format "%s@%s" 'ideographic-radical domain)))
+ (when (and (memq key attributes)
+ (setq value (get-char-attribute char key)))
+ (setq radical value)
+ (insert (format "(%s . %S)\t; %c%s"
+ key
+ radical
+ (aref ideographic-radicals radical)
+ line-breaking))
+ (setq attributes (delq key attributes))
+ )
+ (setq key (intern (format "%s@%s" 'ideographic-strokes domain)))
+ (when (and (memq key attributes)
+ (setq value (get-char-attribute char key)))
+ (setq strokes value)
+ (insert (format "(%s . %S)%s"
+ key
+ strokes
+ line-breaking))
+ (setq attributes (delq key attributes))
+ )
+ (setq key (intern (format "%s@%s*sources"
+ 'ideographic-radical domain)))
+ (when (and (memq key attributes)
+ (setq value (get-char-attribute char key)))
+ (insert (format "(%s%s" key line-breaking))
+ (dolist (cell value)
+ (insert (format " %s" cell)))
+ (insert ")")
+ (insert line-breaking)
+ (setq attributes (delq key attributes))
+ )
+ ))
(when (and (memq 'ideographic-radical attributes)
(setq value (get-char-attribute char 'ideographic-radical)))
(setq radical value)
(while (consp value)
(setq cell (car value))
(if (integerp cell)
- (setq cell (decode-char 'ucs cell)))
+ (setq cell (decode-char '=ucs cell)))
(cond ((characterp cell)
(if separator
(insert lbs))
(while attributes
(setq name (car attributes))
(if (setq value (get-char-attribute char name))
- (cond ((eq name 'jisx0208-1978/4X)
+ (cond ((string-match "^=>ucs@" (symbol-name name))
+ (insert (format "(%-18s . #x%04X)\t; %c%s"
+ name value (decode-char '=ucs value)
+ line-breaking))
+ )
+ ((eq name 'jisx0208-1978/4X)
(insert (format "(%-18s . #x%04X)%s"
name value
line-breaking)))
(while (consp value)
(setq cell (car value))
(if (integerp cell)
- (setq cell (decode-char 'ucs cell)))
+ (setq cell (decode-char '=ucs cell)))
(cond ((characterp cell)
(if separator
(insert lbs))
))
(setq attributes (cdr attributes)))
(while ccs-attributes
- (setq name (car ccs-attributes))
- (if (and (eq name (charset-name name))
- (setq value (get-char-attribute char name)))
+ (setq name (charset-name (car ccs-attributes)))
+ (if (and (not (memq name dest-ccss))
+ (prog1
+ (setq value (get-char-attribute char name))
+ (setq dest-ccss (cons name dest-ccss))))
(insert
(format
- (cond ((memq name '(ideograph-daikanwa-2
- ideograph-daikanwa
- ideograph-gt
- ideograph-cbeta))
+ (cond ((memq name '(=daikanwa
+ =daikanwa-rev1 =daikanwa-rev2
+ =gt =gt-k =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)
(if has-long-ccs-name
"(%-26s . #x%04X)\t; %c%s"
"(%-18s . #x%04X)\t; %c%s"))
(goto-char (point-min))
(while (re-search-forward "[ \t]+$" nil t)
(replace-match ""))
+ ;; from tabify.
+ (goto-char (point-min))
+ (while (re-search-forward "[ \t][ \t][ \t]*" nil t)
+ (let ((column (current-column))
+ (indent-tabs-mode t))
+ (delete-region (match-beginning 0) (point))
+ (indent-to column)))
(goto-char (point-max))
- (tabify (point-min)(point-max))
+ ;; (tabify (point-min)(point-max))
))
(defun insert-char-data-with-variant (char &optional printable
(or (null excluded-script)
(null (setq vs (get-char-attribute variant 'script)))
(not (memq excluded-script vs))))
- (or (and no-ucs-variant (get-char-attribute variant 'ucs))
+ (or (and no-ucs-variant (get-char-attribute variant '=ucs))
(insert-char-data variant printable)))
(setq variants (cdr variants))
)))
(let ((code min)
char)
(while (<= code max)
- (setq char (decode-char 'ucs code))
- (if (get-char-attribute char 'ucs)
+ (setq char (decode-char '=ucs code))
+ (if (encode-char char '=ucs 'defined-only)
(insert-char-data-with-variant char nil 'no-ucs-variant
script excluded-script))
- (setq code (1+ code))
- )))
+ (setq code (1+ code)))))
(defun write-char-range-data-to-file (min max file
&optional script excluded-script)