-;;; tibet-util.el --- Support for inputting Tibetan characters
+;;; tibet-util.el --- utilities for Tibetan -*- coding: iso-2022-7bit; -*-
;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
;; Licensed to the Free Software Foundation.
;; Created: Feb. 17. 1997
-;; History:
+;;; History:
;; 1997.03.13 Modification in treatment of text properties;
;; Support for some special signs and punctuations.
;; 1999.10.25 Modification for a new composition way by K.Handa.
+;;; Commentary:
+
;;; Code:
+(defconst tibetan-obsolete-glyphs
+ `(("\e$(7!=\e(B" . "\e$(7!=\e(B") ; 2 col <-> 1 col
+ ("\e$(7!?\e(B" . "\e$(7!?\e(B")
+ ("\e$(7!@\e(B" . "\e$(7!@\e(B")
+ ("\e$(7!A\e(B" . "\e$(7!A\e(B")
+ ("\e$(7"`\e(B" . "\e$(7"`\e(B")
+ ("\e$(7!;\e(B" . "\e$(7!;\e(B")
+ ("\e$(7!D\e(B" . "\e$(7!D\e(B")
+ ;; Yes these are dirty. But ...
+ ("\e$(7!>\e(B \e$(7!>\e(B" . ,(compose-string "\e$(7!>\e(B \e$(7!>\e(B" 0 3 [?\e$(7!>\e(B (Br . Bl) ? (Br . Bl) ?\e$(7!>\e(B]))
+ ("\e$(7!4!5!5\e(B" . ,(compose-string
+ "\e$(7#R#S#S#S\e(B" 0 4
+ [?\e$(7#R\e(B (Br . Bl) ?\e$(7#S\e(B (Br . Bl) ?\e$(7#S\e(B (Br . Bl) ?\e$(7#S\e(B]))
+ ("\e$(7!4!5\e(B" . ,(compose-string "\e$(7#R#S#S\e(B" 0 3 [?\e$(7#R\e(B (Br . Bl) ?\e$(7#S\e(B (Br . Bl) ?\e$(7#S\e(B]))
+ ("\e$(7!6\e(B" . ,(compose-string "\e$(7#R#S!I\e(B" 0 3 [?\e$(7#R\e(B (Br . Bl) ?\e$(7#S\e(B (br . tr) ?\e$(7!I\e(B]))
+ ("\e$(7!4\e(B" . ,(compose-string "\e$(7#R#S\e(B" 0 2 [?\e$(7#R\e(B (Br . Bl) ?\e$(7#S\e(B]))))
+
;;;###autoload
(defun tibetan-char-p (ch)
"Check if char CH is Tibetan character.
(i 0)
ch this-trans)
(while (< i len)
- (let ((idx (string-match tibetan-precomposition-rule-alist str i)))
+ (let ((idx (string-match tibetan-precomposition-rule-regexp str i)))
(if (eq idx i)
;; Ith character and the followings matches precomposable
;; Tibetan sequence.
;;; (Sanskrit visarga, though it is a vowel modifier, is considered
;;; to be a punctuation.)
;;;
-;;; Here are examples of the words "bsgrubs" and "h'uM"
+;;; Here are examples of the words "bsgrubs" and "hfauM"
;;;
-;;; \e4\e$(7"7\e0"7\e1\e4%qx!"U\e0"G###C"U\e1\e4"7\e0"7\e1\e4"G\e0"G\e1\e(B \e4\e$(7"Hx!"Rx!"Ur'"_\e0"H"A"U"_\e1\e(B
+;;; \e$(7"7"G###C"U"7"G\e(B \e$(7"H"R"U"_\e(B
;;;
;;; M
;;; b s b s h
-;;; g '
+;;; g fa
;;; r u
;;; u
;;;
;; If 'a follows a consonant, turn it into the subjoined form.
;; * Disabled by Tomabechi 2000/06/09 *
;; Because in Unicode, \e$(7"A\e(B may follow directly a consonant without
- ;; any intervening vowel, as in \e4\e$(7"9\e0"9\e1\e4""\e0"""Q\e1\e4"A\e0"A\e1!;\e(B=\e4\e$(7"9\e0"9\e1\e(B \e4\e$(7""\e0""\e1\e(B \e4\e$(7"A\e0"A\e1\e(B not \e4\e$(7"9\e0"9\e1\e(B \e4\e$(7""\e0""\e1\e(B \e$(7"Q\e(B \e4\e$(7"A\e0"A\e1\e(B
+ ;; any intervening vowel, as in \e$(7"9"""Q"A!;\e(B=\e$(7"9\e(B \e$(7""\e(B \e$(7"A\e(B not \e$(7"9\e(B \e$(7""\e(B \e$(7"Q\e(B \e$(7"A\e(B
;;(if (and (= char ?\e$(7"A\e(B)
;; (aref (char-category-set (car last)) ?0))
;; (setq char ?\e$(7"R\e(B)) ;; modified for new font by Tomabechi 1999/12/10
;; Added by Tomabechi 2000/06/08
(if (memq char '(?\e$(7"T\e(B ?\e$(7"V\e(B ?\e$(7"W\e(B ?\e$(7"X\e(B ?\e$(7"Y\e(B ?\e$(7"Z\e(B ?\e$(7"b\e(B))
(setq comp-vowel
- (cddr (assoc (char-to-string char)
- tibetan-composite-vowel-alist))
+ (copy-sequence
+ (cddr (assoc (char-to-string char)
+ tibetan-composite-vowel-alist)))
char
(cadr (assoc (char-to-string char)
tibetan-composite-vowel-alist))))
;; Compose lower vowel sign vertically under.
((aref (char-category-set char) ?3)
- (if (eq char ?\e$(7"Q\e(B) ;; `\e$(7"Q\e(B' should not visible when composed.
+ (if (or (eq char ?\e$(7"Q\e(B) ;; `\e$(7"Q\e(B' and `\e$,1FP\e(B' should not visible when composed.
+ (eq char #xF70))
(setq rule nil)
(setq rule stack-under)))
;; Transform ra-mgo (superscribed r) if followed by a subjoined
(defun tibetan-decompose-region (from to)
"Decompose Tibetan text in the region FROM and TO.
This is different from decompose-region because precomposed Tibetan characters
-are decomposed into normal Tiebtan character sequences."
+are decomposed into normal Tibetan character sequences."
(interactive "r")
(save-restriction
(narrow-to-region from to)
(defun tibetan-decompose-string (str)
"Decompose Tibetan string STR.
This is different from decompose-string because precomposed Tibetan characters
-are decomposed into normal Tiebtan character sequences."
+are decomposed into normal Tibetan character sequences."
(let ((new "")
(len (length str))
(idx 0)
new))
;;;###autoload
-(defun tibetan-composition-function (from to pattern &optional string)
+(defun tibetan-composition-function (pos &optional string)
+ (setq pos (1- pos))
(if string
- (tibetan-compose-string string)
- (tibetan-compose-region from to))
- (- to from))
+ ;; Not yet implemented.
+ nil
+ (if (>= pos (point-min))
+ (save-excursion
+ (goto-char pos)
+ (if (looking-at tibetan-composable-pattern)
+ (prog1 (match-end 0)
+ (tibetan-compose-region pos (match-end 0))))))))
;;;
;;; This variable is used to avoid repeated decomposition.
;; Should return nil as annotations.
nil))
+\f
+;;;
+;;; Unicode-related definitions.
+;;;
+
+(defvar tibetan-canonicalize-for-unicode-alist
+ '(("\e$(7"Q\e(B" . "") ;; remove vowel a
+ ("\e$(7"T\e(B" . "\e$(7"R"S\e(B") ;; decompose vowels whose use is ``discouraged'' in Unicode 3.0
+ ("\e$(7"V\e(B" . "\e$(7"R"U\e(B")
+ ("\e$(7"W\e(B" . "\e$(7#C"a\e(B")
+ ("\e$(7"X\e(B" . "\e$(7#C"R"a\e(B")
+ ("\e$(7"Y\e(B" . "\e$(7#D"a\e(B")
+ ("\e$(7"Z\e(B" . "\e$(7#D"R"a\e(B")
+ ("\e$(7"b\e(B" . "\e$(7"R"a\e(B"))
+ "Rules for canonicalizing Tibetan vowels for Unicode.")
+
+(defvar tibetan-canonicalize-for-unicode-regexp
+ "[\e$(7"Q"T"V"W"X"Y"Z"b\e(B]"
+ "Regexp for Tibetan vowels to be canonicalized in Unicode.")
+
+(defun tibetan-canonicalize-for-unicode-region (from to)
+ (save-restriction
+ (narrow-to-region from to)
+ (goto-char from)
+ (while (re-search-forward tibetan-canonicalize-for-unicode-regexp nil t)
+ (let (
+ ;;(from (match-beginning 0))
+ ;;(to (match-end 0))
+ (canonical-form
+ (cdr (assoc (match-string 0)
+ tibetan-canonicalize-for-unicode-alist))))
+ ;;(goto-char from)
+ ;;(delete-region from to)
+ ;;(insert canonical-form)
+ (replace-match canonical-form)
+ ))))
+
+(defvar tibetan-strict-unicode t
+ "*Flag to control Tibetan canonicalizing for Unicode.
+
+If non-nil, the vowel a is removed and composite vowels are decomposed
+before writing buffer in Unicode. See also
+`tibetan-canonicalize-for-unicode-regexp' and
+`tibetan-canonicalize-for-unicode-alist'.")
+
+;;;###autoload
+(defun tibetan-pre-write-canonicalize-for-unicode (from to)
+ (let ((old-buf (current-buffer))
+ (strict-unicode tibetan-strict-unicode))
+ (set-buffer (generate-new-buffer " *temp*"))
+ (if (stringp from)
+ (insert from)
+ (insert-buffer-substring old-buf from to))
+ (if strict-unicode
+ (tibetan-canonicalize-for-unicode-region (point-min) (point-max)))
+ ;; Should return nil as annotations.
+ nil))
+
(provide 'tibet-util)
-;;; language/tibet-util.el ends here.
+;;; arch-tag: 7a7333e8-1584-446c-b39c-a02b9def265d
+;;; tibet-util.el ends here
HCoop / bpt / emacs.git / blobdiff