Update copyright years.
[emacs.git] / lisp / language / tibet-util.el
blob0f726269a6b2cb5341d4f119ce2c43a2dab376fa
1 ;;; tibet-util.el --- utilities for Tibetan -*- coding: iso-2022-7bit; -*-
3 ;; Copyright (C) 1997, 2001, 2002, 2003, 2004, 2005, 2006
4 ;; Free Software Foundation, Inc.
5 ;; Copyright (C) 1995, 1997, 1998, 2000
6 ;; National Institute of Advanced Industrial Science and Technology (AIST)
7 ;; Registration Number H14PRO021
9 ;; Keywords: multilingual, Tibetan
11 ;; This file is part of GNU Emacs.
13 ;; GNU Emacs is free software; you can redistribute it and/or modify
14 ;; it under the terms of the GNU General Public License as published by
15 ;; the Free Software Foundation; either version 2, or (at your option)
16 ;; any later version.
18 ;; GNU Emacs is distributed in the hope that it will be useful,
19 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
20 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
21 ;; GNU General Public License for more details.
23 ;; You should have received a copy of the GNU General Public License
24 ;; along with GNU Emacs; see the file COPYING. If not, write to the
25 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
26 ;; Boston, MA 02110-1301, USA.
28 ;; Author: Toru TOMABECHI, <Toru.Tomabechi@orient.unil.ch>
30 ;; Created: Feb. 17. 1997
32 ;;; History:
33 ;; 1997.03.13 Modification in treatment of text properties;
34 ;; Support for some special signs and punctuations.
35 ;; 1999.10.25 Modification for a new composition way by K.Handa.
37 ;;; Commentary:
39 ;;; Code:
41 (defconst tibetan-obsolete-glyphs
42 `(("\e$(7!=\e(B" . "\e$(8!=\e(B") ; 2 col <-> 1 col
43 ("\e$(7!?\e(B" . "\e$(8!?\e(B")
44 ("\e$(7!@\e(B" . "\e$(8!@\e(B")
45 ("\e$(7!A\e(B" . "\e$(8!A\e(B")
46 ("\e$(7"`\e(B" . "\e$(8"`\e(B")
47 ("\e$(7!;\e(B" . "\e$(8!;\e(B")
48 ("\e$(7!D\e(B" . "\e$(8!D\e(B")
49 ;; Yes these are dirty. But ...
50 ("\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]))
51 ("\e$(7!4!5!5\e(B" . ,(compose-string
52 "\e$(7#R#S#S#S\e(B" 0 4
53 [?\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]))
54 ("\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$(7#S\e(B]))
55 ("\e$(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]))
56 ("\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]))))
58 ;;;###autoload
59 (defun tibetan-char-p (ch)
60 "Check if char CH is Tibetan character.
61 Returns non-nil if CH is Tibetan. Otherwise, returns nil."
62 (memq (char-charset ch) '(tibetan tibetan-1-column)))
64 ;;; Functions for Tibetan <-> Tibetan-transcription.
66 ;;;###autoload
67 (defun tibetan-tibetan-to-transcription (str)
68 "Transcribe Tibetan string STR and return the corresponding Roman string."
69 (let (;; Accumulate transcriptions here in reverse order.
70 (trans nil)
71 (len (length str))
72 (i 0)
73 ch this-trans)
74 (while (< i len)
75 (let ((idx (string-match tibetan-precomposition-rule-regexp str i)))
76 (if (eq idx i)
77 ;; Ith character and the followings matches precomposable
78 ;; Tibetan sequence.
79 (setq i (match-end 0)
80 this-trans
81 (car (rassoc
82 (cdr (assoc (match-string 0 str)
83 tibetan-precomposition-rule-alist))
84 tibetan-precomposed-transcription-alist)))
85 (setq ch (substring str i (1+ i))
86 i (1+ i)
87 this-trans
88 (car (or (rassoc ch tibetan-consonant-transcription-alist)
89 (rassoc ch tibetan-vowel-transcription-alist)
90 (rassoc ch tibetan-subjoined-transcription-alist)))))
91 (setq trans (cons this-trans trans))))
92 (apply 'concat (nreverse trans))))
94 ;;;###autoload
95 (defun tibetan-transcription-to-tibetan (str)
96 "Convert Tibetan Roman string STR to Tibetan character string.
97 The returned string has no composition information."
98 (let (;; Case is significant.
99 (case-fold-search nil)
100 (idx 0)
101 ;; Accumulate Tibetan strings here in reverse order.
102 (t-str-list nil)
103 i subtrans)
104 (while (setq i (string-match tibetan-regexp str idx))
105 (if (< idx i)
106 ;; STR contains a pattern that doesn't match Tibetan
107 ;; transcription. Include the pattern as is.
108 (setq t-str-list (cons (substring str idx i) t-str-list)))
109 (setq subtrans (match-string 0 str)
110 idx (match-end 0))
111 (let ((t-char (cdr (assoc subtrans
112 tibetan-precomposed-transcription-alist))))
113 (if t-char
114 ;; SUBTRANS corresponds to a transcription for
115 ;; precomposable Tibetan sequence.
116 (setq t-char (car (rassoc t-char
117 tibetan-precomposition-rule-alist)))
118 (setq t-char
119 (cdr
120 (or (assoc subtrans tibetan-consonant-transcription-alist)
121 (assoc subtrans tibetan-vowel-transcription-alist)
122 (assoc subtrans tibetan-modifier-transcription-alist)
123 (assoc subtrans tibetan-subjoined-transcription-alist)))))
124 (setq t-str-list (cons t-char t-str-list))))
125 (if (< idx (length str))
126 (setq t-str-list (cons (substring str idx) t-str-list)))
127 (apply 'concat (nreverse t-str-list))))
130 ;;; Functions for composing/decomposing Tibetan sequence.
132 ;;; A Tibetan syllable is typically structured as follows:
134 ;;; [Prefix] C [C+] V [M] [Suffix [Post suffix]]
136 ;;; where C's are all vertically stacked, V appears below or above
137 ;;; consonant cluster and M is always put above the C[C+]V combination.
138 ;;; (Sanskrit visarga, though it is a vowel modifier, is considered
139 ;;; to be a punctuation.)
141 ;;; Here are examples of the words "bsgrubs" and "hfauM"
143 ;;; \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"R"U"_\e1\e(B
145 ;;; M
146 ;;; b s b s h
147 ;;; g fa
148 ;;; r u
149 ;;; u
151 ;;; Consonants `'' (\e$(7"A\e(B), `w' (\e$(7">\e(B), `y' (\e$(7"B\e(B), `r' (\e$(7"C\e(B) take special
152 ;;; forms when they are used as subjoined consonant. Consonant `r'
153 ;;; takes another special form when used as superjoined in such a case
154 ;;; as "rka", while it does not change its form when conjoined with
155 ;;; subjoined `'', `w' or `y' as in "rwa", "rya".
157 ;; Append a proper composition rule and glyph to COMPONENTS to compose
158 ;; CHAR with a composition that has COMPONENTS.
160 (defun tibetan-add-components (components char)
161 (let ((last (last components))
162 (stack-upper '(tc . bc))
163 (stack-under '(bc . tc))
164 rule comp-vowel tmp)
165 ;; Special treatment for 'a chung.
166 ;; If 'a follows a consonant, turn it into the subjoined form.
167 ;; * Disabled by Tomabechi 2000/06/09 *
168 ;; Because in Unicode, \e$(7"A\e(B may follow directly a consonant without
169 ;; any intervening vowel, as in \e4\e$(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$(7"A\e0"A\e1\e(B
170 ;;(if (and (= char ?\e$(7"A\e(B)
171 ;; (aref (char-category-set (car last)) ?0))
172 ;; (setq char ?\e$(7"R\e(B)) ;; modified for new font by Tomabechi 1999/12/10
174 ;; Composite vowel signs are decomposed before being added
175 ;; Added by Tomabechi 2000/06/08
176 (if (memq char '(?\e$(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))
177 (setq comp-vowel
178 (copy-sequence
179 (cddr (assoc (char-to-string char)
180 tibetan-composite-vowel-alist)))
181 char
182 (cadr (assoc (char-to-string char)
183 tibetan-composite-vowel-alist))))
184 (cond
185 ;; Compose upper vowel sign vertically over.
186 ((aref (char-category-set char) ?2)
187 (setq rule stack-upper))
189 ;; Compose lower vowel sign vertically under.
190 ((aref (char-category-set char) ?3)
191 (if (eq char ?\e$(7"Q\e(B) ;; `\e$(7"Q\e(B' should not visible when composed.
192 (setq rule nil)
193 (setq rule stack-under)))
194 ;; Transform ra-mgo (superscribed r) if followed by a subjoined
195 ;; consonant other than w, ', y, r.
196 ((and (= (car last) ?\e$(7"C\e(B)
197 (not (memq char '(?\e$(7#>\e(B ?\e$(7"R\e(B ?\e$(7#B\e(B ?\e$(7#C\e(B))))
198 (setcar last ?\e$(7!"\e(B) ;; modified for newfont by Tomabechi 1999/12/10
199 (setq rule stack-under))
200 ;; Transform initial base consonant if followed by a subjoined
201 ;; consonant but 'a.
203 (let ((laststr (char-to-string (car last))))
204 (if (and (/= char ?\e$(7"R\e(B) ;; modified for new font by Tomabechi
205 (string-match "[\e$(7"!\e(B-\e$(7"="?"@"D\e(B-\e$(7"J"K\e(B]" laststr))
206 (setcar last (string-to-char
207 (cdr (assoc (char-to-string (car last))
208 tibetan-base-to-subjoined-alist)))))
209 (setq rule stack-under))))
211 (if rule
212 (setcdr last (list rule char)))
213 ;; Added by Tomabechi 2000/06/08
214 (if comp-vowel
215 (nconc last comp-vowel))
218 ;;;###autoload
219 (defun tibetan-compose-string (str)
220 "Compose Tibetan string STR."
221 (let ((idx 0))
222 ;; `\e$(7"A\e(B' is included in the pattern for subjoined consonants
223 ;; because we treat it specially in tibetan-add-components.
224 ;; (This feature is removed by Tomabechi 2000/06/08)
225 (while (setq idx (string-match tibetan-composable-pattern str idx))
226 (let ((from idx)
227 (to (match-end 0))
228 components)
229 (if (eq (string-match tibetan-precomposition-rule-regexp str idx) idx)
230 (setq idx (match-end 0)
231 components
232 (list (string-to-char
233 (cdr
234 (assoc (match-string 0 str)
235 tibetan-precomposition-rule-alist)))))
236 (setq components (list (aref str idx))
237 idx (1+ idx)))
238 (while (< idx to)
239 (tibetan-add-components components (aref str idx))
240 (setq idx (1+ idx)))
241 (compose-string str from to components))))
242 str)
244 ;;;###autoload
245 (defun tibetan-compose-region (beg end)
246 "Compose Tibetan text the region BEG and END."
247 (interactive "r")
248 (let (str result chars)
249 (save-excursion
250 (save-restriction
251 (narrow-to-region beg end)
252 (goto-char (point-min))
253 ;; `\e$(7"A\e(B' is included in the pattern for subjoined consonants
254 ;; because we treat it specially in tibetan-add-components.
255 ;; (This feature is removed by Tomabechi 2000/06/08)
256 (while (re-search-forward tibetan-composable-pattern nil t)
257 (let ((from (match-beginning 0))
258 (to (match-end 0))
259 components)
260 (goto-char from)
261 (if (looking-at tibetan-precomposition-rule-regexp)
262 (progn
263 (setq components
264 (list (string-to-char
265 (cdr
266 (assoc (match-string 0)
267 tibetan-precomposition-rule-alist)))))
268 (goto-char (match-end 0)))
269 (setq components (list (char-after from)))
270 (forward-char 1))
271 (while (< (point) to)
272 (tibetan-add-components components (following-char))
273 (forward-char 1))
274 (compose-region from to components)))))))
276 (defvar tibetan-decompose-precomposition-alist
277 (mapcar (function (lambda (x) (cons (string-to-char (cdr x)) (car x))))
278 tibetan-precomposition-rule-alist))
280 ;;;###autoload
281 (defun tibetan-decompose-region (from to)
282 "Decompose Tibetan text in the region FROM and TO.
283 This is different from decompose-region because precomposed Tibetan characters
284 are decomposed into normal Tibetan character sequences."
285 (interactive "r")
286 (save-restriction
287 (narrow-to-region from to)
288 (decompose-region from to)
289 (goto-char from)
290 (while (not (eobp))
291 (let* ((char (following-char))
292 (slot (assq char tibetan-decompose-precomposition-alist)))
293 (if slot
294 (progn
295 (delete-char 1)
296 (insert (cdr slot)))
297 (forward-char 1))))))
300 ;;;###autoload
301 (defun tibetan-decompose-string (str)
302 "Decompose Tibetan string STR.
303 This is different from decompose-string because precomposed Tibetan characters
304 are decomposed into normal Tibetan character sequences."
305 (let ((new "")
306 (len (length str))
307 (idx 0)
308 char slot)
309 (while (< idx len)
310 (setq char (aref str idx)
311 slot (assq (aref str idx) tibetan-decompose-precomposition-alist)
312 new (concat new (if slot (cdr slot) (char-to-string char)))
313 idx (1+ idx)))
314 new))
316 ;;;###autoload
317 (defun tibetan-composition-function (from to pattern &optional string)
318 (if string
319 (tibetan-compose-string string)
320 (tibetan-compose-region from to))
321 (- to from))
324 ;;; This variable is used to avoid repeated decomposition.
326 (setq-default tibetan-decomposed nil)
328 ;;;###autoload
329 (defun tibetan-decompose-buffer ()
330 "Decomposes Tibetan characters in the buffer into their components.
331 See also the documentation of the function `tibetan-decompose-region'."
332 (interactive)
333 (make-local-variable 'tibetan-decomposed)
334 (cond ((not tibetan-decomposed)
335 (tibetan-decompose-region (point-min) (point-max))
336 (setq tibetan-decomposed t))))
338 ;;;###autoload
339 (defun tibetan-compose-buffer ()
340 "Composes Tibetan character components in the buffer.
341 See also docstring of the function tibetan-compose-region."
342 (interactive)
343 (make-local-variable 'tibetan-decomposed)
344 (tibetan-compose-region (point-min) (point-max))
345 (setq tibetan-decomposed nil))
347 ;;;###autoload
348 (defun tibetan-post-read-conversion (len)
349 (save-excursion
350 (save-restriction
351 (let ((buffer-modified-p (buffer-modified-p)))
352 (narrow-to-region (point) (+ (point) len))
353 (tibetan-compose-region (point-min) (point-max))
354 (set-buffer-modified-p buffer-modified-p)
355 (make-local-variable 'tibetan-decomposed)
356 (setq tibetan-decomposed nil)
357 (- (point-max) (point-min))))))
360 ;;;###autoload
361 (defun tibetan-pre-write-conversion (from to)
362 (setq tibetan-decomposed-temp tibetan-decomposed)
363 (let ((old-buf (current-buffer)))
364 (set-buffer (generate-new-buffer " *temp*"))
365 (if (stringp from)
366 (insert from)
367 (insert-buffer-substring old-buf from to))
368 (if (not tibetan-decomposed-temp)
369 (tibetan-decompose-region (point-min) (point-max)))
370 ;; Should return nil as annotations.
371 nil))
375 ;;; Unicode-related definitions.
378 (defvar tibetan-canonicalize-for-unicode-alist
379 '(("\e$(7"Q\e(B" . "") ;; remove vowel a
380 ("\e$(7"T\e(B" . "\e$(7"R"S\e(B") ;; decompose vowels whose use is ``discouraged'' in Unicode 3.0
381 ("\e$(7"V\e(B" . "\e$(7"R"U\e(B")
382 ("\e$(7"W\e(B" . "\e$(7#C"a\e(B")
383 ("\e$(7"X\e(B" . "\e$(7#C"R"a\e(B")
384 ("\e$(7"Y\e(B" . "\e$(7#D"a\e(B")
385 ("\e$(7"Z\e(B" . "\e$(7#D"R"a\e(B")
386 ("\e$(7"b\e(B" . "\e$(7"R"a\e(B"))
387 "Rules for canonicalizing Tibetan vowels for Unicode.")
389 (defvar tibetan-canonicalize-for-unicode-regexp
390 "[\e$(7"Q"T"V"W"X"Y"Z"b\e(B]"
391 "Regexp for Tibetan vowels to be canonicalized in Unicode.")
393 (defun tibetan-canonicalize-for-unicode-region (from to)
394 (save-restriction
395 (narrow-to-region from to)
396 (goto-char from)
397 (while (re-search-forward tibetan-canonicalize-for-unicode-regexp nil t)
398 (let (
399 ;;(from (match-beginning 0))
400 ;;(to (match-end 0))
401 (canonical-form
402 (cdr (assoc (match-string 0)
403 tibetan-canonicalize-for-unicode-alist))))
404 ;;(goto-char from)
405 ;;(delete-region from to)
406 ;;(insert canonical-form)
407 (replace-match canonical-form)
408 ))))
410 (defvar tibetan-strict-unicode t
411 "*Flag to control Tibetan canonicalizing for Unicode.
413 If non-nil, the vowel a is removed and composite vowels are decomposed
414 before writing buffer in Unicode. See also
415 `tibetan-canonicalize-for-unicode-regexp' and
416 `tibetan-canonicalize-for-unicode-alist'.")
418 ;;;###autoload
419 (defun tibetan-pre-write-canonicalize-for-unicode (from to)
420 (let ((old-buf (current-buffer))
421 (strict-unicode tibetan-strict-unicode))
422 (set-buffer (generate-new-buffer " *temp*"))
423 (if (stringp from)
424 (insert from)
425 (insert-buffer-substring old-buf from to))
426 (if strict-unicode
427 (tibetan-canonicalize-for-unicode-region (point-min) (point-max)))
428 ;; Should return nil as annotations.
429 nil))
431 (provide 'tibet-util)
433 ;;; arch-tag: 7a7333e8-1584-446c-b39c-a02b9def265d
434 ;;; tibet-util.el ends here