X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;ds=sidebyside;f=lisp%2Futf-2000%2Fchar-db-util.el;h=eaab82ea14ffe7f5a753a3f160bf288688348b86;hb=31b8e156c51ef5962cee6e715e34af022f63d43f;hp=75aa159d554c10d6d944dec94a7d5eda76b00fa9;hpb=048b9ddbb4157961e6db6a96ad4be2bab4b3b625;p=chise%2Fxemacs-chise.git diff --git a/lisp/utf-2000/char-db-util.el b/lisp/utf-2000/char-db-util.el index 75aa159..eaab82e 100644 --- a/lisp/utf-2000/char-db-util.el +++ b/lisp/utf-2000/char-db-util.el @@ -1,6 +1,6 @@ ;;; char-db-util.el --- Character Database utility -;; Copyright (C) 1998,1999,2000,2001,2002,2003,2004 MORIOKA Tomohiko. +;; Copyright (C) 1998,1999,2000,2001,2002,2003,2004,2005 MORIOKA Tomohiko. ;; Author: MORIOKA Tomohiko ;; Keywords: CHISE, Character Database, ISO/IEC 10646, Unicode, UCS-4, MULE. @@ -69,35 +69,41 @@ (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)) + jis-x0212 jis-x0213 cdp shinjigen misc unknown)) -(defvar char-db-ignored-attributes nil) +(defvar char-db-ignored-attributes '(ideographic-products)) (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) @@ -121,9 +127,9 @@ greek-iso8859-7 thai-tis620 =jis-x0208 - japanese-jisx0208 + =jis-x0208@1978 + =jis-x0208@1983 japanese-jisx0212 - japanese-jisx0208-1978 chinese-gb2312 chinese-cns11643-1 chinese-cns11643-2 @@ -132,7 +138,7 @@ chinese-cns11643-5 chinese-cns11643-6 chinese-cns11643-7 - =jis-x0208-1990 + =jis-x0208@1990 =jis-x0213-1-2000 =jis-x0213-2-2000 korean-ksc5601 @@ -159,9 +165,13 @@ ideograph-hanziku-10 ideograph-hanziku-11 ideograph-hanziku-12 + =gt-k + =ucs@iso + =ucs@unicode =big5 =big5-eten - =gt-k + =jis-x0208@1997 + =zinbun-oracle =jef-china3)) (defun char-db-make-char-spec (char) @@ -193,10 +203,12 @@ (dolist (ccs (delq (car ret) (charset-list))) (if (and (or (charset-iso-final-char ccs) (memq ccs - '(ideograph-daikanwa - =daikanwa-rev2 + '(=daikanwa + =daikanwa@rev2 ;; =gt-k ))) + (setq ccs (charset-name ccs)) + (null (assq ccs char-spec)) (setq ret (encode-char char ccs 'defined-only))) (setq char-spec (cons (cons ccs ret) char-spec)))) (if (null char-spec) @@ -212,10 +224,11 @@ ((consp char) char)))) -(defun char-db-insert-char-spec (char &optional readable column) +(defun char-db-insert-char-spec (char &optional readable column + 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 @@ -227,32 +240,10 @@ 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)) (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. @@ -267,7 +258,8 @@ (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 @@ -280,16 +272,19 @@ (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)) @@ -307,18 +302,21 @@ (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)) @@ -400,49 +398,69 @@ (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 char-db-insert-ccs-feature (name value line-breaking) + (insert + (format + (cond ((memq name '(=daikanwa + =daikanwa@rev1 =daikanwa@rev2 + =gt =gt-k =cbeta =zinbun-oracle)) + "(%-18s . %05d)\t; %c") + ((eq name 'mojikyo) + "(%-18s . %06d)\t; %c") + ((>= (charset-dimension name) 2) + "(%-18s . #x%04X)\t; %c") + (t + "(%-18s . #x%02X)\t; %c")) + 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))) + (if (and (= (charset-chars name) 94) + (= (charset-dimension name) 2)) + (insert (format " [%02d-%02d]" + (- (lsh value -8) 32) + (- (logand value 255) 32)))) + (insert line-breaking)) + +(defun insert-char-attributes (char &optional readable attributes column) (unless column (setq column (current-column))) - (let (name value has-long-ccs-name rest + (let (name value ; has-long-ccs-name + rest radical strokes (line-breaking (concat "\n" (make-string (1+ column) ?\ ))) lbs cell separator ret key al cal - dest-ccss) + dest-ccss + 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))) @@ -470,6 +488,14 @@ 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))) @@ -490,30 +516,21 @@ line-breaking)) (setq attributes (delq name attributes)) )) - (dolist (name '(=>ucs-gb =>ucs-cns =>ucs-jis =>ucs-ks =>ucs-big5)) + (dolist (name '(=>daikanwa)) (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)) + (insert + (if (integerp value) + (format "(%-18s . %05d)\t; %c%s" + name value (decode-char '=daikanwa value) + line-breaking) + (format "(%-18s %s)\t; %c%s" + name + (mapconcat (function prin1-to-string) + value " ") + (char-representative-of-daikanwa char) + 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) - line-breaking)) - (setq attributes (delq '->ucs attributes)) - ) (when (and (memq 'general-category attributes) (setq value (get-char-attribute char 'general-category))) (insert (format @@ -608,7 +625,19 @@ (setq attributes (delq 'ideographic-radical attributes)) ) (let (key) - (dolist (domain char-db-feature-domains) + (dolist (domain + (append + char-db-feature-domains + (let (dest domain) + (dolist (feature (char-attribute-list)) + (setq feature (symbol-name feature)) + (when (string-match + "\\(radical\\|strokes\\)@\\([^@*]+\\)\\(\\*\\|$\\)" + feature) + (setq domain (intern (match-string 2 feature))) + (unless (memq domain dest) + (setq dest (cons domain dest))))) + (sort dest #'string<)))) (setq key (intern (format "%s@%s" 'ideographic-radical domain))) (when (and (memq key attributes) (setq value (get-char-attribute char key))) @@ -835,28 +864,22 @@ (setq attributes (delq 'hanyu-dazidian-char attributes)) ) (unless readable - (when (memq '->ucs-unified attributes) - (setq attributes (delq '->ucs-unified attributes)) - ) - (when (memq 'composition attributes) - (setq attributes (delq 'composition attributes)) - )) - (setq rest ccs-attributes) - (while (and rest - (progn - (setq value (get-char-attribute char (car rest))) - (if value - (if (>= (length (symbol-name (car rest))) 19) - (progn - (setq has-long-ccs-name t) - nil) - t) - t))) - (setq rest (cdr rest))) + (dolist (ignored '(composition + ->denotational <-subsumptive ->ucs-unified + ->ideographic-component-forms)) + (setq attributes (delq ignored attributes)))) (while attributes (setq name (car attributes)) (if (setq value (get-char-attribute char name)) - (cond ((string-match "^=>ucs@" (symbol-name name)) + (cond ((setq ret (find-charset name)) + (setq name (charset-name ret)) + (if (and (not (memq name dest-ccss)) + (prog1 + (setq value (get-char-attribute char name)) + (setq dest-ccss (cons name dest-ccss)))) + (char-db-insert-ccs-feature name value line-breaking)) + ) + ((string-match "^=>ucs@" (symbol-name name)) (insert (format "(%-18s . #x%04X)\t; %c%s" name value (decode-char '=ucs value) line-breaking)) @@ -864,8 +887,33 @@ ((eq name 'jisx0208-1978/4X) (insert (format "(%-18s . #x%04X)%s" name value - line-breaking))) + line-breaking)) + ) + ((and (not readable) + (null (get-char-attribute + char + (intern (format "%s*sources" name)))) + (not (string-match "\\*sources$" (symbol-name name))) + (or (eq name '<-identical) + (string-match "^->halfwidth" (symbol-name name)) + (and + (string-match "^->fullwidth" (symbol-name name)) + (not + (and (consp value) + (characterp (car value)) + (encode-char + (car value) '=ucs 'defined-only)))) + (string-match "^->simplified" (symbol-name name)) + (string-match "^->vulgar" (symbol-name name)) + (string-match "^->wrong" (symbol-name name)) + (string-match "^->same" (symbol-name name)) + (string-match "^->original" (symbol-name name)) + (string-match "^->ancient" (symbol-name name)) + (string-match "^->Oracle-Bones" (symbol-name name)) + )) + ) ((or (eq name 'ideographic-structure) + (eq name 'ideographic-combination) (eq name 'ideographic-) (string-match "^\\(->\\|<-\\)" (symbol-name name))) (insert (format "(%-18s%s " name line-breaking)) @@ -875,12 +923,78 @@ (setq cell (car value)) (if (integerp cell) (setq cell (decode-char '=ucs cell))) - (cond ((characterp cell) + (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 + char + (intern (format "%s*sources" name)))) + (setq required-features nil) + (dolist (source sources) + (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 (insert (format "%S" cell)) - (char-db-insert-char-spec cell readable)) + (char-db-insert-char-spec cell readable + nil + required-features)) (setq separator lbs)) ((consp cell) (if separator @@ -952,9 +1066,15 @@ readable al cal) (setq separator lbs)) + (setq ret (prin1-to-string cell)) (if separator - (insert separator)) - (insert (prin1-to-string cell)) + (if (< (+ (current-column) + (length ret) + (length separator)) + 76) + (insert separator) + (insert lbs))) + (insert ret) (setq separator " ")) (setq value (cdr value))) (insert ")") @@ -965,56 +1085,15 @@ 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))) (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) @@ -1034,24 +1113,30 @@ no-ucs-unified script excluded-script) (insert-char-data char printable) - (let ((variants (or (char-variants char) - (let ((ucs (get-char-attribute char '->ucs))) - (if ucs - (delete char (char-variants (int-char ucs))))))) - variant vs) + (let ((variants (char-variants char)) + rest + variant vs ret) (setq variants (sort variants #'<)) - (while variants - (setq variant (car variants)) - (if (and (or (null script) - (null (setq vs (get-char-attribute variant 'script))) - (memq script vs)) - (or (null excluded-script) - (null (setq vs (get-char-attribute variant 'script))) - (not (memq excluded-script vs)))) - (or (and no-ucs-unified (get-char-attribute variant '=ucs)) - (insert-char-data variant printable))) - (setq variants (cdr variants)) - ))) + (setq rest variants) + (setq variants (cons char variants)) + (while rest + (setq variant (car rest)) + (unless (get-char-attribute variant '<-subsumptive) + (if (and (or (null script) + (null (setq vs (get-char-attribute variant 'script))) + (memq script vs)) + (or (null excluded-script) + (null (setq vs (get-char-attribute variant 'script))) + (not (memq excluded-script vs)))) + (unless (and no-ucs-unified (get-char-attribute variant '=ucs)) + (insert-char-data variant printable) + (if (setq ret (char-variants variant)) + (while ret + (or (memq (car ret) variants) + ;; (get-char-attribute (car ret) '<-subsumptive) + (setq rest (nconc rest (list (car ret))))) + (setq ret (cdr ret))))))) + (setq rest (cdr rest))))) (defun insert-char-range-data (min max &optional script excluded-script) (let ((code min) @@ -1065,9 +1150,10 @@ (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))))