X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=lisp%2Futf-2000%2Fchar-db-util.el;h=104f52d8b6d41e4ad6a201863a0f895d7fd53baa;hb=9279e5e6f882d49e7c3143c33129f078408ccb68;hp=59890a80736ba965d40a6d14418144c693aab645;hpb=ac5df5187661b22586d6b97bf3a4c10914c8b1c5;p=chise%2Fxemacs-chise.git.1 diff --git a/lisp/utf-2000/char-db-util.el b/lisp/utf-2000/char-db-util.el index 59890a8..104f52d 100644 --- a/lisp/utf-2000/char-db-util.el +++ b/lisp/utf-2000/char-db-util.el @@ -69,6 +69,8 @@ (setq i (1+ i))) v)) +(defvar char-db-file-coding-system 'utf-8-mcs-er) + (defvar char-db-feature-domains '(ucs daikanwa cns gt jis jis/alt jis/a jis/b jis-x0213 misc unknown)) @@ -77,27 +79,31 @@ (defun char-attribute-name< (ka kb) (cond + ((eq '->denotational kb) + t) + ((eq '->subsumptive kb) + (not (eq '->denotational ka))) + ((eq '->denotational ka) + nil) + ((eq '->subsumptive ka) + nil) ((find-charset ka) - (cond - ((find-charset kb) - (if (<= (charset-id ka) 0) - (if (<= (charset-id kb) 0) - (cond - ((= (charset-dimension ka) - (charset-dimension kb)) - (> (charset-id ka)(charset-id kb))) - (t - (> (charset-dimension ka) - (charset-dimension kb)) - )) - t) - (if (<= (charset-id kb) 0) - nil - (< (charset-id ka)(charset-id kb))))) - ((symbolp kb) - nil) - (t - t))) + (if (find-charset kb) + (if (<= (charset-id ka) 1) + (if (<= (charset-id kb) 1) + (cond + ((= (charset-dimension ka) + (charset-dimension kb)) + (> (charset-id ka)(charset-id kb))) + (t + (> (charset-dimension ka) + (charset-dimension kb)) + )) + t) + (if (<= (charset-id kb) 1) + nil + (< (charset-id ka)(charset-id kb)))) + nil)) ((find-charset kb) t) ((symbolp ka) @@ -193,7 +199,7 @@ (dolist (ccs (delq (car ret) (charset-list))) (if (and (or (charset-iso-final-char ccs) (memq ccs - '(ideograph-daikanwa + '(=daikanwa =daikanwa-rev2 ;; =gt-k ))) @@ -216,7 +222,7 @@ required-features) (unless column (setq column (current-column))) - (let (char-spec al cal key temp-char) + (let (char-spec temp-char) (setq char-spec (char-db-make-char-spec char)) (unless (or (characterp char) ; char (condition-case nil @@ -228,37 +234,10 @@ 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 (encode-char char key 'defined-only) - (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)) - (dolist (feature required-features) - (if (find-charset feature) - (if (encode-char char feature 'defined-only) - (setq cal (adjoin feature cal))) - (setq al (adjoin feature al)))) (insert-char-attributes char readable - (or al 'none) cal) + (union (mapcar #'car char-spec) + required-features)) (when temp-char ;; undefine temporary character ;; Current implementation is dirty. @@ -273,7 +252,8 @@ (let ((line-breaking (concat "\n" (make-string (1+ column) ?\ ))) name value - ret al cal key + ret al ; cal + key lbs cell rest separator) (insert "(") (while alist @@ -286,16 +266,19 @@ (error nil))) (progn (setq al nil - cal nil) + ;; cal nil + ) (while value (setq key (car (car value))) - (if (find-charset key) - (setq cal (cons key cal)) - (setq al (cons key al))) + ;; (if (find-charset key) + ;; (setq cal (cons key cal)) + (setq al (cons key al)) + ;; ) (setq value (cdr value))) (insert-char-attributes ret readable - (or al 'none) cal)) + (or al 'none) ; cal + )) (insert (prin1-to-string value))) (insert ")") (insert line-breaking)) @@ -313,18 +296,21 @@ (progn (setq rest cell al nil - cal nil) + ;; cal nil + ) (while rest (setq key (car (car rest))) - (if (find-charset key) - (setq cal (cons key cal)) - (setq al (cons key al))) + ;; (if (find-charset key) + ;; (setq cal (cons key cal)) + (setq al (cons key al)) + ;; ) (setq rest (cdr rest))) (if separator (insert lbs)) (insert-char-attributes ret readable - al cal) + al ; cal + ) (setq separator lbs)) (if separator (insert separator)) @@ -406,40 +392,7 @@ (defvar char-db-convert-obsolete-format t) -(defun insert-char-attributes (char &optional readable - attributes ccs-attributes - column) - (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<))) +(defun insert-char-attributes (char &optional readable attributes column) (unless column (setq column (current-column))) (let (name value has-long-ccs-name rest @@ -449,7 +402,26 @@ lbs cell separator ret key al cal dest-ccss - sources required-features) + sources required-features + ccss) + (let (atr-d) + (setq attributes + (sort (if attributes + (if (consp attributes) + (progn + (dolist (name attributes) + (unless (memq name char-db-ignored-attributes) + (if (find-charset name) + (push name ccss)) + (push name atr-d))) + atr-d)) + (dolist (name (char-attribute-list)) + (unless (memq name char-db-ignored-attributes) + (if (find-charset name) + (push name ccss)) + (push name atr-d))) + atr-d) + #'char-attribute-name<))) (insert "(") (when (and (memq 'name attributes) (setq value (get-char-attribute char 'name))) @@ -477,6 +449,14 @@ line-breaking)) (setq attributes (delq 'script attributes)) ) + ;; (when (and (memq '<-denotational attributes) + ;; (setq value (get-char-attribute char '<-denotational)) + ;; (null (cdr value)) + ;; (setq value (encode-char (car value) 'ucs 'defined-only))) + ;; (insert (format "(%-18s . #x%04X)\t; %c%s" + ;; '=>ucs value (decode-char 'ucs value) + ;; line-breaking)) + ;; (setq attributes (delq '<-denotational attributes))) (dolist (name '(=>ucs =>ucs*)) (when (and (memq name attributes) (setq value (get-char-attribute char name))) @@ -531,7 +511,7 @@ line-breaking) (format "(%-18s %s)\t; %c%s" name - (mapconcat #'prin1-to-string + (mapconcat (function prin1-to-string) value " ") (char-representative-of-daikanwa char) line-breaking))) @@ -857,31 +837,67 @@ (setq attributes (delq 'hanyu-dazidian-char attributes)) ) (unless readable - (when (memq '->ucs-unified attributes) - (setq attributes (delq '->ucs-unified attributes)) - ) - (when (memq '->denotational attributes) - (setq attributes (delq '->denotational attributes)) - ) - (when (memq 'composition attributes) - (setq attributes (delq 'composition attributes)) - )) - (setq rest ccs-attributes) - (while (and rest - (progn - (setq value (get-char-attribute char (car rest))) - (if value - (if (>= (length (symbol-name (car rest))) 19) - (progn - (setq has-long-ccs-name t) - nil) - t) - t))) - (setq rest (cdr rest))) + (dolist (ignored '(composition + ->denotational <-subsumptive ->ucs-unified + ->ideographic-component-forms + <-same)) + (setq attributes (delq ignored attributes)))) + ;; (setq rest ccs-attributes) + ;; (while (and rest + ;; (progn + ;; (setq value (get-char-attribute char (car rest))) + ;; (if value + ;; (if (>= (length (symbol-name (car rest))) 19) + ;; (progn + ;; (setq has-long-ccs-name t) + ;; nil) + ;; t) + ;; t))) + ;; (setq rest (cdr rest))) (while attributes (setq name (car attributes)) (if (setq value (get-char-attribute char name)) - (cond ((string-match "^=>ucs@" (symbol-name name)) + (cond ((setq ret (find-charset name)) + (setq name (charset-name ret)) + (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 '(=daikanwa + =daikanwa-rev1 =daikanwa-rev2 + =gt =gt-k =cbeta)) + (if has-long-ccs-name + "(%-26s . %05d)\t; %c%s" + "(%-18s . %05d)\t; %c%s")) + ((eq name 'mojikyo) + (if has-long-ccs-name + "(%-26s . %06d)\t; %c%s" + "(%-18s . %06d)\t; %c%s")) + ((>= (charset-dimension name) 2) + (if has-long-ccs-name + "(%-26s . #x%04X)\t; %c%s" + "(%-18s . #x%04X)\t; %c%s")) + (t + (if has-long-ccs-name + "(%-26s . #x%02X)\t; %c%s" + "(%-18s . #x%02X)\t; %c%s"))) + name + (if (= (charset-iso-graphic-plane name) 1) + (logior value + (cond ((= (charset-dimension name) 1) + #x80) + ((= (charset-dimension name) 2) + #x8080) + ((= (charset-dimension name) 3) + #x808080) + (t 0))) + value) + (char-db-decode-isolated-char name value) + line-breaking))) + ) + ((string-match "^=>ucs@" (symbol-name name)) (insert (format "(%-18s . #x%04X)\t; %c%s" name value (decode-char '=ucs value) line-breaking)) @@ -889,7 +905,14 @@ ((eq name 'jisx0208-1978/4X) (insert (format "(%-18s . #x%04X)%s" name value - line-breaking))) + line-breaking)) + ) + ((and (not readable) + (or (eq name '<-identical) + (string-match "^->simplified" (symbol-name name)) + (string-match "^->vulgar" (symbol-name name)) + )) + ) ((or (eq name 'ideographic-structure) (eq name 'ideographic-) (string-match "^\\(->\\|<-\\)" (symbol-name name))) @@ -900,14 +923,23 @@ (setq cell (car value)) (if (integerp cell) (setq cell (decode-char '=ucs cell))) - (cond ((eq name '->unified) - (if separator - (insert lbs)) - (let ((char-db-ignored-attributes - (cons '<-unified - char-db-ignored-attributes))) - (insert-char-attributes cell readable)) - (setq separator lbs)) + (cond ((eq name '->subsumptive) + (when (or (not + (some (lambda (atr) + (get-char-attribute cell atr)) + char-db-ignored-attributes)) + (some (lambda (ccs) + (encode-char cell ccs + 'defined-only)) + ccss)) + (if separator + (insert lbs)) + (let ((char-db-ignored-attributes + (cons '<-subsumptive + char-db-ignored-attributes))) + (insert-char-attributes cell readable)) + (setq separator lbs)) + ) ((characterp cell) (setq sources (get-char-attribute @@ -915,24 +947,47 @@ (intern (format "%s*sources" name)))) (setq required-features nil) (dolist (source sources) - (setq required-features - (cons - (if (find-charset - (setq ret (intern - (format "=%s" source)))) - ret - source) - required-features))) - (when (string-match "@JP" (symbol-name name)) - (setq required-features - (union required-features - '(=jis-x0208 - =jis-x0208-1990 - =jis-x0213-1-2000 - =jis-x0213-2-2000 - =jis-x0212 - =jis-x0208-1983 - =jis-x0208-1978)))) + (cond + ((memq source '(JP JP/Jouyou + shinjigen-1)) + (setq required-features + (union required-features + '(=jis-x0208 + =jis-x0208-1990 + =jis-x0213-1-2000 + =jis-x0213-2-2000 + =jis-x0212 + =jis-x0208-1983 + =jis-x0208-1978)))) + ((eq source 'CN) + (setq required-features + (union required-features + '(=gb2312 + =gb12345 + =iso-ir165))))) + (cond + ((find-charset + (setq ret (intern (format "=%s" source)))) + (setq required-features + (cons ret required-features))) + (t (setq required-features + (cons source required-features))))) + (cond ((string-match "@JP" (symbol-name name)) + (setq required-features + (union required-features + '(=jis-x0208 + =jis-x0208-1990 + =jis-x0213-1-2000 + =jis-x0213-2-2000 + =jis-x0212 + =jis-x0208-1983 + =jis-x0208-1978)))) + ((string-match "@CN" (symbol-name name)) + (setq required-features + (union required-features + '(=gb2312 + =gb12345 + =iso-ir165))))) (if separator (insert lbs)) (if readable @@ -1030,56 +1085,55 @@ line-breaking))) )) (setq attributes (cdr attributes))) - (while ccs-attributes - (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 '(=daikanwa - =daikanwa-rev1 =daikanwa-rev2 - =gt =gt-k =cbeta)) - (if has-long-ccs-name - "(%-26s . %05d)\t; %c%s" - "(%-18s . %05d)\t; %c%s")) - ((eq name 'mojikyo) - (if has-long-ccs-name - "(%-26s . %06d)\t; %c%s" - "(%-18s . %06d)\t; %c%s")) - ((>= (charset-dimension name) 2) - (if has-long-ccs-name - "(%-26s . #x%04X)\t; %c%s" - "(%-18s . #x%04X)\t; %c%s")) - (t - (if has-long-ccs-name - "(%-26s . #x%02X)\t; %c%s" - "(%-18s . #x%02X)\t; %c%s"))) - name - (if (= (charset-iso-graphic-plane name) 1) - (logior value - (cond ((= (charset-dimension name) 1) - #x80) - ((= (charset-dimension name) 2) - #x8080) - ((= (charset-dimension name) 3) - #x808080) - (t 0))) - value) - (char-db-decode-isolated-char name value) - line-breaking))) - (setq ccs-attributes (cdr ccs-attributes))) + ;; (while ccs-attributes + ;; (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 '(=daikanwa + ;; =daikanwa-rev1 =daikanwa-rev2 + ;; =gt =gt-k =cbeta)) + ;; (if has-long-ccs-name + ;; "(%-26s . %05d)\t; %c%s" + ;; "(%-18s . %05d)\t; %c%s")) + ;; ((eq name 'mojikyo) + ;; (if has-long-ccs-name + ;; "(%-26s . %06d)\t; %c%s" + ;; "(%-18s . %06d)\t; %c%s")) + ;; ((>= (charset-dimension name) 2) + ;; (if has-long-ccs-name + ;; "(%-26s . #x%04X)\t; %c%s" + ;; "(%-18s . #x%04X)\t; %c%s")) + ;; (t + ;; (if has-long-ccs-name + ;; "(%-26s . #x%02X)\t; %c%s" + ;; "(%-18s . #x%02X)\t; %c%s"))) + ;; name + ;; (if (= (charset-iso-graphic-plane name) 1) + ;; (logior value + ;; (cond ((= (charset-dimension name) 1) + ;; #x80) + ;; ((= (charset-dimension name) 2) + ;; #x8080) + ;; ((= (charset-dimension name) 3) + ;; #x808080) + ;; (t 0))) + ;; value) + ;; (char-db-decode-isolated-char name value) + ;; line-breaking))) + ;; (setq ccs-attributes (cdr ccs-attributes))) (insert ")"))) (defun insert-char-data (char &optional readable - attributes ccs-attributes) + attributes) (save-restriction (narrow-to-region (point)(point)) (insert "(define-char '") - (insert-char-attributes char readable - attributes ccs-attributes) + (insert-char-attributes char readable attributes) (insert ")\n") (goto-char (point-min)) (while (re-search-forward "[ \t]+$" nil t) @@ -1130,9 +1184,10 @@ (defun write-char-range-data-to-file (min max file &optional script excluded-script) - (let ((coding-system-for-write 'utf-8-mcs)) + (let ((coding-system-for-write char-db-file-coding-system)) (with-temp-buffer - (insert ";; -*- coding: utf-8-mcs -*-\n") + (insert (format ";; -*- coding: %s -*-\n" + char-db-file-coding-system)) (insert-char-range-data min max script excluded-script) (write-region (point-min)(point-max) file))))