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, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
7 ;; National Institute of Advanced Industrial Science and Technology (AIST)
8 ;; Registration Number H14PRO021
10 ;; Keywords: multilingual, Tibetan
12 ;; This file is part of GNU Emacs.
14 ;; GNU Emacs is free software; you can redistribute it and/or modify
15 ;; it under the terms of the GNU General Public License as published by
16 ;; the Free Software Foundation; either version 2, or (at your option)
19 ;; GNU Emacs is distributed in the hope that it will be useful,
20 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
21 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 ;; GNU General Public License for more details.
24 ;; You should have received a copy of the GNU General Public License
25 ;; along with GNU Emacs; see the file COPYING. If not, write to the
26 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
27 ;; Boston, MA 02110-1301, USA.
29 ;; Author: Toru TOMABECHI, <Toru.Tomabechi@orient.unil.ch>
31 ;; Created: Feb. 17. 1997
34 ;; 1997.03.13 Modification in treatment of text properties;
35 ;; Support for some special signs and punctuations.
36 ;; 1999.10.25 Modification for a new composition way by K.Handa.
42 (defconst tibetan-obsolete-glyphs
43 `(("\e$(7!=\e(B" .
"\e$(7!=\e(B") ; 2 col <-> 1 col
44 ("\e$(7!?\e(B" .
"\e$(7!?\e(B")
45 ("\e$(7!@\e(B" .
"\e$(7!@\e(B")
46 ("\e$(7!A\e(B" .
"\e$(7!A\e(B")
47 ("\e$(7"`\e(B" . "\e$
(7"`\e(B")
48 ("\e$(7!;\e(B" .
"\e$(7!;\e(B")
49 ("\e$(7!D\e(B" .
"\e$(7!D\e(B")
50 ;; Yes these are dirty. But ...
51 ("\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]))
52 ("\e$(7!4!5!5\e(B" .
,(compose-string
53 "\e$(7#R#S#S#S\e(B" 0 4
54 [?
\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]))
55 ("\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]))
56 ("\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]))
57 ("\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]))))
60 (defun tibetan-char-p (ch)
61 "Check if char CH is Tibetan character.
62 Returns non-nil if CH is Tibetan. Otherwise, returns nil."
63 (memq (char-charset ch
) '(tibetan tibetan-1-column
)))
65 ;;; Functions for Tibetan <-> Tibetan-transcription.
68 (defun tibetan-tibetan-to-transcription (str)
69 "Transcribe Tibetan string STR and return the corresponding Roman string."
70 (let (;; Accumulate transcriptions here in reverse order.
76 (let ((idx (string-match tibetan-precomposition-rule-regexp str i
)))
78 ;; Ith character and the followings matches precomposable
83 (cdr (assoc (match-string 0 str
)
84 tibetan-precomposition-rule-alist
))
85 tibetan-precomposed-transcription-alist
)))
86 (setq ch
(substring str i
(1+ i
))
89 (car (or (rassoc ch tibetan-consonant-transcription-alist
)
90 (rassoc ch tibetan-vowel-transcription-alist
)
91 (rassoc ch tibetan-subjoined-transcription-alist
)))))
92 (setq trans
(cons this-trans trans
))))
93 (apply 'concat
(nreverse trans
))))
96 (defun tibetan-transcription-to-tibetan (str)
97 "Convert Tibetan Roman string STR to Tibetan character string.
98 The returned string has no composition information."
99 (let (;; Case is significant.
100 (case-fold-search nil
)
102 ;; Accumulate Tibetan strings here in reverse order.
105 (while (setq i
(string-match tibetan-regexp str idx
))
107 ;; STR contains a pattern that doesn't match Tibetan
108 ;; transcription. Include the pattern as is.
109 (setq t-str-list
(cons (substring str idx i
) t-str-list
)))
110 (setq subtrans
(match-string 0 str
)
112 (let ((t-char (cdr (assoc subtrans
113 tibetan-precomposed-transcription-alist
))))
115 ;; SUBTRANS corresponds to a transcription for
116 ;; precomposable Tibetan sequence.
117 (setq t-char
(car (rassoc t-char
118 tibetan-precomposition-rule-alist
)))
121 (or (assoc subtrans tibetan-consonant-transcription-alist
)
122 (assoc subtrans tibetan-vowel-transcription-alist
)
123 (assoc subtrans tibetan-modifier-transcription-alist
)
124 (assoc subtrans tibetan-subjoined-transcription-alist
)))))
125 (setq t-str-list
(cons t-char t-str-list
))))
126 (if (< idx
(length str
))
127 (setq t-str-list
(cons (substring str idx
) t-str-list
)))
128 (apply 'concat
(nreverse t-str-list
))))
131 ;;; Functions for composing/decomposing Tibetan sequence.
133 ;;; A Tibetan syllable is typically structured as follows:
135 ;;; [Prefix] C [C+] V [M] [Suffix [Post suffix]]
137 ;;; where C's are all vertically stacked, V appears below or above
138 ;;; consonant cluster and M is always put above the C[C+]V combination.
139 ;;; (Sanskrit visarga, though it is a vowel modifier, is considered
140 ;;; to be a punctuation.)
142 ;;; Here are examples of the words "bsgrubs" and "hfauM"
144 ;;; \e$(7"7"G###C"U"7"G\e(B \e$(7"H"R"U"_\e(B
152 ;;; 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
153 ;;; forms when they are used as subjoined consonant. Consonant `r'
154 ;;; takes another special form when used as superjoined in such a case
155 ;;; as "rka", while it does not change its form when conjoined with
156 ;;; subjoined `'', `w' or `y' as in "rwa", "rya".
158 ;; Append a proper composition rule and glyph to COMPONENTS to compose
159 ;; CHAR with a composition that has COMPONENTS.
161 (defun tibetan-add-components (components char
)
162 (let ((last (last components
))
163 (stack-upper '(tc . bc
))
164 (stack-under '(bc . tc
))
166 ;; Special treatment for 'a chung.
167 ;; If 'a follows a consonant, turn it into the subjoined form.
168 ;; * Disabled by Tomabechi 2000/06/09 *
169 ;; Because in Unicode, \e$(7"A\e(B may follow directly a consonant without
170 ;; any intervening vowel, as in \e$(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$(7"A\e(B
171 ;;(if (and (= char ?\e$(7"A\e(B)
172 ;; (aref (char-category-set (car last)) ?0))
173 ;; (setq char ?\e$(7"R\e(B)) ;; modified for new font by Tomabechi 1999/12/10
175 ;; Composite vowel signs are decomposed before being added
176 ;; Added by Tomabechi 2000/06/08
177 (if (memq char
'(?
\e$
(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))
180 (cddr (assoc (char-to-string char)
181 tibetan-composite-vowel-alist)))
183 (cadr (assoc (char-to-string char)
184 tibetan-composite-vowel-alist))))
186 ;; Compose upper vowel sign vertically over.
187 ((aref (char-category-set char) ?2)
188 (setq rule stack-upper))
190 ;; Compose lower vowel sign vertically under.
191 ((aref (char-category-set char) ?3)
192 (if (or (eq char ?\e$(7"Q
\e(B) ;; `\e$(7"Q\e(B' and `\e$,1FP\e(B' should not visible when composed.
195 (setq rule stack-under
)))
196 ;; Transform ra-mgo (superscribed r) if followed by a subjoined
197 ;; consonant other than w, ', y, r.
198 ((and (= (car last
) ?
\e$
(7"C\e(B)
199 (not (memq char '(?\e$(7#>\e(B ?\e$(7"R
\e(B ?
\e$
(7#B
\e(B ?
\e$
(7#C
\e(B))))
200 (setcar last ?
\e$
(7!"\e(B) ;; modified for newfont by Tomabechi 1999/12/10
201 (setq rule stack-under))
202 ;; Transform initial base consonant if followed by a subjoined
205 (let ((laststr (char-to-string (car last))))
206 (if (and (/= char ?\e$(7"R
\e(B) ;; modified for new font by Tomabechi
207 (string-match "[\e$(7"!\e(B-\e$
(7"="?
"@"D
\e(B-\e$
(7"J"K
\e(B]" laststr))
208 (setcar last (string-to-char
209 (cdr (assoc (char-to-string (car last))
210 tibetan-base-to-subjoined-alist)))))
211 (setq rule stack-under))))
214 (setcdr last (list rule char)))
215 ;; Added by Tomabechi 2000/06/08
217 (nconc last comp-vowel))
221 (defun tibetan-compose-string (str)
222 "Compose Tibetan string STR.
"
224 ;; `\e$(7"A
\e(B' is included in the pattern for subjoined consonants
225 ;; because we treat it specially in tibetan-add-components.
226 ;; (This feature is removed by Tomabechi 2000/06/08)
227 (while (setq idx
(string-match tibetan-composable-pattern str idx
))
231 (if (eq (string-match tibetan-precomposition-rule-regexp str idx
) idx
)
232 (setq idx
(match-end 0)
234 (list (string-to-char
236 (assoc (match-string 0 str
)
237 tibetan-precomposition-rule-alist
)))))
238 (setq components
(list (aref str idx
))
241 (tibetan-add-components components
(aref str idx
))
243 (compose-string str from to components
))))
247 (defun tibetan-compose-region (beg end
)
248 "Compose Tibetan text the region BEG and END."
250 (let (str result chars
)
253 (narrow-to-region beg end
)
254 (goto-char (point-min))
255 ;; `\e$(7"A\e(B' is included in the pattern for subjoined consonants
256 ;; because we treat it specially in tibetan-add-components.
257 ;; (This feature is removed by Tomabechi 2000/06/08)
258 (while (re-search-forward tibetan-composable-pattern nil t
)
259 (let ((from (match-beginning 0))
263 (if (looking-at tibetan-precomposition-rule-regexp
)
266 (list (string-to-char
268 (assoc (match-string 0)
269 tibetan-precomposition-rule-alist
)))))
270 (goto-char (match-end 0)))
271 (setq components
(list (char-after from
)))
273 (while (< (point) to
)
274 (tibetan-add-components components
(following-char))
276 (compose-region from to components
)))))))
278 (defvar tibetan-decompose-precomposition-alist
279 (mapcar (function (lambda (x) (cons (string-to-char (cdr x
)) (car x
))))
280 tibetan-precomposition-rule-alist
))
283 (defun tibetan-decompose-region (from to
)
284 "Decompose Tibetan text in the region FROM and TO.
285 This is different from decompose-region because precomposed Tibetan characters
286 are decomposed into normal Tibetan character sequences."
289 (narrow-to-region from to
)
290 (decompose-region from to
)
293 (let* ((char (following-char))
294 (slot (assq char tibetan-decompose-precomposition-alist
)))
299 (forward-char 1))))))
303 (defun tibetan-decompose-string (str)
304 "Decompose Tibetan string STR.
305 This is different from decompose-string because precomposed Tibetan characters
306 are decomposed into normal Tibetan character sequences."
312 (setq char
(aref str idx
)
313 slot
(assq (aref str idx
) tibetan-decompose-precomposition-alist
)
314 new
(concat new
(if slot
(cdr slot
) (char-to-string char
)))
319 (defun tibetan-composition-function (pos &optional string
)
322 ;; Not yet implemented.
324 (if (>= pos
(point-min))
327 (if (looking-at tibetan-composable-pattern
)
329 (tibetan-compose-region pos
(match-end 0))))))))
332 ;;; This variable is used to avoid repeated decomposition.
334 (setq-default tibetan-decomposed nil
)
337 (defun tibetan-decompose-buffer ()
338 "Decomposes Tibetan characters in the buffer into their components.
339 See also the documentation of the function `tibetan-decompose-region'."
341 (make-local-variable 'tibetan-decomposed
)
342 (cond ((not tibetan-decomposed
)
343 (tibetan-decompose-region (point-min) (point-max))
344 (setq tibetan-decomposed t
))))
347 (defun tibetan-compose-buffer ()
348 "Composes Tibetan character components in the buffer.
349 See also docstring of the function tibetan-compose-region."
351 (make-local-variable 'tibetan-decomposed
)
352 (tibetan-compose-region (point-min) (point-max))
353 (setq tibetan-decomposed nil
))
356 (defun tibetan-post-read-conversion (len)
359 (let ((buffer-modified-p (buffer-modified-p)))
360 (narrow-to-region (point) (+ (point) len
))
361 (tibetan-compose-region (point-min) (point-max))
362 (set-buffer-modified-p buffer-modified-p
)
363 (make-local-variable 'tibetan-decomposed
)
364 (setq tibetan-decomposed nil
)
365 (- (point-max) (point-min))))))
369 (defun tibetan-pre-write-conversion (from to
)
370 (setq tibetan-decomposed-temp tibetan-decomposed
)
371 (let ((old-buf (current-buffer)))
372 (set-buffer (generate-new-buffer " *temp*"))
375 (insert-buffer-substring old-buf from to
))
376 (if (not tibetan-decomposed-temp
)
377 (tibetan-decompose-region (point-min) (point-max)))
378 ;; Should return nil as annotations.
383 ;;; Unicode-related definitions.
386 (defvar tibetan-canonicalize-for-unicode-alist
387 '(("\e$(7"Q
\e(B" . "") ;; remove vowel a
388 ("\e$
(7"T\e(B" .
"\e$(7"R
"S\e(B") ;; decompose vowels whose use is ``discouraged'' in Unicode 3.0
389 ("\e$(7"V
\e(B" . "\e$
(7"R"U
\e(B")
390 ("\e$
(7"W\e(B" .
"\e$(7#C"a
\e(B")
391 ("\e$
(7"X\e(B" .
"\e$(7#C"R
"a\e(B")
392 ("\e$(7"Y
\e(B" . "\e$
(7#D
"a\e(B")
393 ("\e$(7"Z
\e(B" . "\e$
(7#D
"R"a
\e(B")
394 ("\e$
(7"b\e(B" .
"\e$(7"R
"a\e(B"))
395 "Rules for canonicalizing Tibetan vowels for Unicode.")
397 (defvar tibetan-canonicalize-for-unicode-regexp
398 "[\e$(7"Q
"T"V
"W"X
"Y"Z
"b\e(B]"
399 "Regexp for Tibetan vowels to be canonicalized in Unicode.")
401 (defun tibetan-canonicalize-for-unicode-region (from to
)
403 (narrow-to-region from to
)
405 (while (re-search-forward tibetan-canonicalize-for-unicode-regexp nil t
)
407 ;;(from (match-beginning 0))
410 (cdr (assoc (match-string 0)
411 tibetan-canonicalize-for-unicode-alist
))))
413 ;;(delete-region from to)
414 ;;(insert canonical-form)
415 (replace-match canonical-form
)
418 (defvar tibetan-strict-unicode t
419 "*Flag to control Tibetan canonicalizing for Unicode.
421 If non-nil, the vowel a is removed and composite vowels are decomposed
422 before writing buffer in Unicode. See also
423 `tibetan-canonicalize-for-unicode-regexp' and
424 `tibetan-canonicalize-for-unicode-alist'.")
427 (defun tibetan-pre-write-canonicalize-for-unicode (from to
)
428 (let ((old-buf (current-buffer))
429 (strict-unicode tibetan-strict-unicode
))
430 (set-buffer (generate-new-buffer " *temp*"))
433 (insert-buffer-substring old-buf from to
))
435 (tibetan-canonicalize-for-unicode-region (point-min) (point-max)))
436 ;; Should return nil as annotations.
439 (provide 'tibet-util
)
441 ;;; arch-tag: 7a7333e8-1584-446c-b39c-a02b9def265d
442 ;;; tibet-util.el ends here