Fix bug #15064 with assertion violation due to mouse face.
[emacs.git] / leim / quail / hangul.el
blobb50917af335183f4d120db3c386c9ef643f2eb1f
1 ;;; hangul.el --- Korean Hangul input method
3 ;; Copyright (C) 2008-2013 Free Software Foundation, Inc.
5 ;; Author: Jihyun Cho <jihyun.jo@gmail.com>
6 ;; Keywords: multilingual, input method, Korean, Hangul
8 ;; This file is part of GNU Emacs.
10 ;; GNU Emacs is free software: you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation, either version 3 of the License, or
13 ;; (at your option) any later version.
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
23 ;;; Commentary:
25 ;; This file is to implement the following hangul automata:
26 ;; - Hangul 2-Bulsik input method
27 ;; - Hangul 3-Bulsik final input method
28 ;; - Hangul 3-Bulsik 390 input method
30 ;;; Code:
32 (require 'quail)
33 (require 'hanja-util)
35 ;; Hangul double Jamo table.
36 ;; The format is an alist of JAMO-TYPE vs. DOUBLE-JAMO-TABLE.
38 ;; JAMO-TYPE is a symbol `cho' for Choseong, `jung' for Jungseong, and
39 ;; `jong' for Jongseong.
41 ;; DOUBLE-JAMO-TABLE is an alist of Jamo index vs. the vector of Jamo
42 ;; indies that can be combined with the car part.
44 ;; Jamo index is a relative index in `hangul Compatibility Jamo' area
45 ;; of the Unicode (i.e. 1 for U+3131).
47 (defconst hangul-djamo-table
48 '((cho . ((1 . [1]) ; Choseong
49 (7 . [7])
50 (18 . [18])
51 (21 . [21])
52 (24 . [24])))
53 (jung . ((39 . [31 32 51]) ; Jungseong
54 (44 . [35 36 51])
55 (49 . [51])))
56 (jong . ((1 . [1 21]) ; Jongseong
57 (4 . [24 30])
58 (9 . [1 17 18 21 28 29 30])
59 (18 . [18 21])
60 (21 . [21])))))
62 ;; Hangul 2-Bulsik keymap.
63 ;; It converts an ASCII code A-Z, a-z, to the corresponding hangul
64 ;; Jamo index.
66 (defconst hangul2-keymap
67 [17 48 26 23 7 9 30 39 33 35 31 51 49 44 32 36 18 1 4 21 37 29 24 28 43 27])
69 ;; Hangul 3-Bulsik final keymap. 3-Bulsik use full keyboard layout.
70 ;; Therefore, we must map all printable ASCII codes (`!' to `~')
71 ;; to Hangul 3-Bulsik codes.
72 ;; Other parts are the same as `hangul2-keymap'.
73 (defconst hangul3-keymap
74 [2 183 24 15 14 8220 120 39 126 8221 43 44 41 46 74 119 30 22 18 78 83
75 68 73 85 79 52 110 44 62 46 33 10 7 63 27 12 5 11 69 48 55 49 50 51
76 34 45 56 57 29 16 6 13 54 3 28 20 53 26 40 58 60 61 59 42 23 79 71
77 86 72 66 84 96 109 115 93 116 122 113 118 121 21 67 4 70 99 74 9 1
78 101 17 37 92 47 8251])
80 ;; Hangul 3-Bulsik 390 keymap.
81 ;; The role is the same as `hangul3-keymap'.
82 (defconst hangul390-keymap
83 [24 34 35 36 37 38 120 40 41 42 43 44 45 46 73 119 30 22 18 77 82 67 72
84 84 78 58 110 50 61 51 63 64 7 33 11 10 27 2 47 39 56 52 53 54 49 48
85 57 62 29 68 6 59 55 16 28 20 60 26 91 92 93 94 95 96 23 78 70 85 71
86 65 83 90 109 115 87 116 122 113 118 121 21 66 4 69 99 73 9 1 101 17
87 123 124 125 126])
89 (defvar hangul-im-keymap
90 (let ((map (make-sparse-keymap)))
91 (define-key map "\d" 'hangul-delete-backward-char)
92 (define-key map [f9] 'hangul-to-hanja-conversion)
93 (define-key map [Hangul_Hanja] 'hangul-to-hanja-conversion)
94 map)
95 "Keymap for Hangul method. It is used by all Hangul input methods.")
97 ;; Current input character buffer. Store separated hangul character.
98 ;; The first and second are Choseong position.
99 ;; The third and forth are Jungseong position.
100 ;; The fifth and sixth are Jongseong position.
101 ;; The second, forth and sixth are double Jamo position.
102 (defvar hangul-queue
103 (make-vector 6 0))
105 (defsubst notzerop (number)
106 (not (zerop number)))
108 (defsubst alphabetp (char)
109 (or (and (>= char ?A) (<= char ?Z))
110 (and (>= char ?a) (<= char ?z))))
112 (defun hangul-character (cho jung jong)
113 "Convert CHO, JUNG, JONG to the precomposed `Hangul Syllables' character.
114 CHO, JUNG, JONG are relative indices in `Hangul Compatibility Jamo' of Unicode.
115 Return a zero-length string if the conversion fails."
117 (decode-char
118 'ucs
119 (if (and (/= cho 0) (/= jung 0))
120 (+ #xac00
121 (* 588
122 (- cho
123 (cond ((< cho 3) 1)
124 ((< cho 5) 2)
125 ((< cho 10) 4)
126 ((< cho 20) 11)
127 (t 12))))
128 (* 28 (- jung 31))
129 (- jong
130 (cond ((< jong 8) 0)
131 ((< jong 19) 1)
132 ((< jong 25) 2)
133 (t 3))))
134 (+ #x3130
135 (cond ((/= cho 0) cho)
136 ((/= jung 0) jung)
137 ((/= jong 0) jong)))))
138 ""))
140 (defun hangul-insert-character (&rest queues)
141 "Insert characters generated from QUEUES.
142 Each queue has the same form as `hangul-queue'.
143 Setup `quail-overlay' to the last character."
144 (if (and mark-active transient-mark-mode)
145 (progn
146 (delete-region (region-beginning) (region-end))
147 (deactivate-mark)))
148 (quail-delete-region)
149 (let ((first (car queues)))
150 (insert
151 (hangul-character
152 (+ (aref first 0) (hangul-djamo 'cho (aref first 0) (aref first 1)))
153 (+ (aref first 2) (hangul-djamo 'jung (aref first 2) (aref first 3)))
154 (+ (aref first 4) (hangul-djamo 'jong (aref first 4) (aref first 5))))))
155 (move-overlay quail-overlay (overlay-start quail-overlay) (point))
156 (dolist (queue (cdr queues))
157 (insert
158 (hangul-character
159 (+ (aref queue 0) (hangul-djamo 'cho (aref queue 0) (aref queue 1)))
160 (+ (aref queue 2) (hangul-djamo 'jung (aref queue 2) (aref queue 3)))
161 (+ (aref queue 4) (hangul-djamo 'jong (aref queue 4) (aref queue 5)))))
162 (move-overlay quail-overlay (1+ (overlay-start quail-overlay)) (point))))
164 (defun hangul-djamo (jamo char1 char2)
165 "Return the double Jamo index calculated from the arguments.
166 JAMO is a type of Hangul Jamo; `cho', `jung', or `jong'.
167 CHAR1 and CHAR2 are Hangul Jamo indices.
168 Return nil if CHAR1 and CHAR2 can not be combined."
169 (let* ((jamo (cdr (assoc jamo hangul-djamo-table)))
170 (char1 (cdr (assoc char1 jamo))))
171 (if char1
172 (let ((i (length char1)))
173 (or (catch 'found
174 (while (> i 0)
175 (if (= char2 (aref char1 (1- i)))
176 (throw 'found i))
177 (setf i (1- i))))
179 0)))
181 (defsubst hangul2-input-method-jaum (char)
182 "Store Hangul Jamo indice CHAR in `hangul-queue'.
183 It is a Hangul 2-Bulsik Jaum.
184 This function processes a Hangul 2-Bulsik Jaum.
185 The Hangul 2-Bulsik is composed of a Jaum and a Moum.
186 The Jaum can be located in a Choseong position and a Jongseong position.
187 Unless the function inserts CHAR to `hangul-queue',
188 commit current `hangul-queue' and then set a new `hangul-queue',
189 and insert CHAR to new `hangul-queue'."
190 (if (cond ((zerop (aref hangul-queue 0))
191 (aset hangul-queue 0 char))
192 ((and (zerop (aref hangul-queue 1))
193 (zerop (aref hangul-queue 2))
194 (notzerop (hangul-djamo 'cho (aref hangul-queue 0) char)))
195 (aset hangul-queue 1 char))
196 ((and (zerop (aref hangul-queue 4))
197 (notzerop (aref hangul-queue 2))
198 (/= char 8)
199 (/= char 19)
200 (/= char 25)
201 (numberp
202 (hangul-character
203 (+ (aref hangul-queue 0)
204 (hangul-djamo
205 'cho
206 (aref hangul-queue 0)
207 (aref hangul-queue 1)))
208 (+ (aref hangul-queue 2)
209 (hangul-djamo
210 'jung
211 (aref hangul-queue 2)
212 (aref hangul-queue 3)))
213 char)))
214 (aset hangul-queue 4 char))
215 ((and (zerop (aref hangul-queue 5))
216 (notzerop (hangul-djamo 'jong (aref hangul-queue 4) char))
217 (numberp
218 (hangul-character
219 (+ (aref hangul-queue 0)
220 (hangul-djamo
221 'cho
222 (aref hangul-queue 0)
223 (aref hangul-queue 1)))
224 (+ (aref hangul-queue 2)
225 (hangul-djamo
226 'jung
227 (aref hangul-queue 2)
228 (aref hangul-queue 3)))
229 (+ (aref hangul-queue 4)
230 (hangul-djamo
231 'jong
232 (aref hangul-queue 4)
233 char)))))
234 (aset hangul-queue 5 char)))
235 (hangul-insert-character hangul-queue)
236 (hangul-insert-character hangul-queue
237 (setq hangul-queue (vector char 0 0 0 0 0)))))
239 (defsubst hangul2-input-method-moum (char)
240 "Store Hangul Jamo indice CHAR in `hangul-queue'.
241 It is a Hangul 2-Bulsik Moum.
242 This function processes a Hangul 2-Bulsik Moum.
243 The Moum can be located in a Jungseong position.
244 Other parts are the same as a `hangul2-input-method-jaum'."
245 (if (cond ((zerop (aref hangul-queue 2))
246 (aset hangul-queue 2 char))
247 ((and (zerop (aref hangul-queue 3))
248 (zerop (aref hangul-queue 4))
249 (notzerop (hangul-djamo 'jung (aref hangul-queue 2) char)))
250 (aset hangul-queue 3 char)))
251 (hangul-insert-character hangul-queue)
252 (let ((next-char (vector 0 0 char 0 0 0)))
253 (cond ((notzerop (aref hangul-queue 5))
254 (aset next-char 0 (aref hangul-queue 5))
255 (aset hangul-queue 5 0))
256 ((notzerop (aref hangul-queue 4))
257 (aset next-char 0 (aref hangul-queue 4))
258 (aset hangul-queue 4 0)))
259 (hangul-insert-character hangul-queue
260 (setq hangul-queue next-char)))))
262 (defsubst hangul3-input-method-cho (char)
263 "Store Hangul Jamo indice CHAR in `hangul-queue'.
264 It is a Hangul 3-Bulsik Choseong.
265 This function processes a Hangul 3-Bulsik Choseong.
266 The Hangul 3-Bulsik is composed of a Choseong, a Jungseong and a Jongseong.
267 The Choseong can be located in a Choseong position.
268 Other parts are the same as a `hangul2-input-method-jaum'."
269 (if (cond ((and (zerop (aref hangul-queue 0))
270 (zerop (aref hangul-queue 4)))
271 (aset hangul-queue 0 char))
272 ((and (zerop (aref hangul-queue 1))
273 (zerop (aref hangul-queue 2))
274 (notzerop (hangul-djamo 'cho (aref hangul-queue 0) char)))
275 (aset hangul-queue 1 char)))
276 (hangul-insert-character hangul-queue)
277 (hangul-insert-character hangul-queue
278 (setq hangul-queue (vector char 0 0 0 0 0)))))
280 (defsubst hangul3-input-method-jung (char)
281 "Store Hangul Jamo indice CHAR in `hangul-queue'.
282 It is a Hangul 3-Bulsik Jungseong.
283 This function processes a Hangul 3-Bulsik Jungseong.
284 The Jungseong can be located in a Jungseong position.
285 Other parts are the same as a `hangul3-input-method-cho'."
286 (if (cond ((and (zerop (aref hangul-queue 2))
287 (zerop (aref hangul-queue 4)))
288 (aset hangul-queue 2 char))
289 ((and (zerop (aref hangul-queue 3))
290 (notzerop (hangul-djamo 'jung (aref hangul-queue 2) char)))
291 (aset hangul-queue 3 char)))
292 (hangul-insert-character hangul-queue)
293 (hangul-insert-character hangul-queue
294 (setq hangul-queue (vector 0 0 char 0 0 0)))))
296 (defsubst hangul3-input-method-jong (char)
297 "Store Hangul Jamo indice CHAR in `hangul-queue'.
298 It is a Hangul 3-Bulsik Jongseong.
299 This function processes a Hangul 3-Bulsik Jongseong.
300 The Jongseong can be located in a Jongseong position.
301 Other parts are the same as a `hangul3-input-method-cho'."
302 (if (cond ((and (zerop (aref hangul-queue 4))
303 (notzerop (aref hangul-queue 0))
304 (notzerop (aref hangul-queue 2))
305 (numberp
306 (hangul-character
307 (+ (aref hangul-queue 0)
308 (hangul-djamo
309 'cho
310 (aref hangul-queue 0)
311 (aref hangul-queue 1)))
312 (+ (aref hangul-queue 2)
313 (hangul-djamo
314 'jung
315 (aref hangul-queue 2)
316 (aref hangul-queue 3)))
317 char)))
318 (aset hangul-queue 4 char))
319 ((and (zerop (aref hangul-queue 5))
320 (notzerop (hangul-djamo 'jong (aref hangul-queue 4) char))
321 (numberp
322 (hangul-character
323 (+ (aref hangul-queue 0)
324 (hangul-djamo
325 'cho
326 (aref hangul-queue 0)
327 (aref hangul-queue 1)))
328 (+ (aref hangul-queue 2)
329 (hangul-djamo
330 'jung
331 (aref hangul-queue 2)
332 (aref hangul-queue 3)))
333 (+ (aref hangul-queue 4)
334 (hangul-djamo
335 'jong
336 (aref hangul-queue 4)
337 char)))))
338 (aset hangul-queue 5 char)))
339 (hangul-insert-character hangul-queue)
340 (if (zerop (apply '+ (append hangul-queue nil)))
341 (hangul-insert-character (setq hangul-queue (vector 0 0 0 0 char 0)))
342 (hangul-insert-character hangul-queue
343 (setq hangul-queue (vector 0 0 0 0 char 0))))))
345 (defun hangul-delete-backward-char ()
346 "Delete the previous hangul character by Jaso units."
347 (interactive)
348 (let ((i 5))
349 (while (and (> i 0) (zerop (aref hangul-queue i)))
350 (setq i (1- i)))
351 (aset hangul-queue i 0))
352 (if (notzerop (apply '+ (append hangul-queue nil)))
353 (hangul-insert-character hangul-queue)
354 (delete-backward-char 1)))
356 (defun hangul-to-hanja-conversion ()
357 "Convert the previous hangul character to the corresponding hanja character.
358 When a Korean input method is off, convert the following hangul character."
359 (interactive)
360 (let ((echo-keystrokes 0)
361 delete-func
362 hanja-character)
363 (if (and (overlayp quail-overlay) (overlay-start quail-overlay))
364 (progn
365 (setq hanja-character (hangul-to-hanja-char (preceding-char)))
366 (setq delete-func (lambda () (delete-backward-char 1))))
367 (setq hanja-character (hangul-to-hanja-char (following-char)))
368 (setq delete-func (lambda () (delete-char 1))))
369 (when hanja-character
370 (funcall delete-func)
371 (insert hanja-character)
372 (setq hangul-queue (make-vector 6 0))
373 (if (and (overlayp quail-overlay) (overlay-start quail-overlay))
374 (move-overlay quail-overlay (point) (point))))))
376 ;; Support function for `hangul2-input-method'. Actually, this
377 ;; function handles the Hangul 2-Bulsik. KEY is an entered key code
378 ;; used for looking up `hangul2-keymap'."
379 (defun hangul2-input-method-internal (key)
380 (let ((char (+ (aref hangul2-keymap (1- (% key 32)))
381 (cond ((or (= key ?O) (= key ?P)) 2)
382 ((or (= key ?E) (= key ?Q) (= key ?R)
383 (= key ?T) (= key ?W)) 1)
384 (t 0)))))
385 (if (< char 31)
386 (hangul2-input-method-jaum char)
387 (hangul2-input-method-moum char))))
389 (defun hangul2-input-method (key)
390 "2-Bulsik input method."
391 (if (or buffer-read-only (not (alphabetp key)))
392 (list key)
393 (quail-setup-overlays nil)
394 (let ((input-method-function nil)
395 (echo-keystrokes 0)
396 (help-char nil))
397 (setq hangul-queue (make-vector 6 0))
398 (hangul2-input-method-internal key)
399 (unwind-protect
400 (catch 'exit-input-loop
401 (while t
402 (let* ((seq (read-key-sequence nil))
403 (cmd (lookup-key hangul-im-keymap seq))
404 key)
405 (cond ((and (stringp seq)
406 (= 1 (length seq))
407 (setq key (aref seq 0))
408 (alphabetp key))
409 (hangul2-input-method-internal key))
410 ((commandp cmd)
411 (call-interactively cmd))
413 (setq unread-command-events (listify-key-sequence seq))
414 (throw 'exit-input-loop nil))))))
415 (quail-delete-overlays)))))
417 ;; Support function for `hangul3-input-method'. Actually, this
418 ;; function handles the Hangul 3-Bulsik final. KEY is an entered key
419 ;; code used for looking up `hangul3-keymap'."
420 (defun hangul3-input-method-internal (key)
421 (let ((char (aref hangul3-keymap (- key 33))))
422 (cond ((and (> char 92) (< char 123))
423 (hangul3-input-method-cho (- char 92)))
424 ((and (> char 65) (< char 87))
425 (hangul3-input-method-jung (- char 35)))
426 ((< char 31)
427 (hangul3-input-method-jong char))
429 (setq hangul-queue (make-vector 6 0))
430 (insert (decode-char 'ucs char))
431 (move-overlay quail-overlay (point) (point))))))
433 (defun hangul3-input-method (key)
434 "3-Bulsik final input method."
435 (if (or buffer-read-only (< key 33) (>= key 127))
436 (list key)
437 (quail-setup-overlays nil)
438 (let ((input-method-function nil)
439 (echo-keystrokes 0)
440 (help-char nil))
441 (setq hangul-queue (make-vector 6 0))
442 (hangul3-input-method-internal key)
443 (unwind-protect
444 (catch 'exit-input-loop
445 (while t
446 (let* ((seq (read-key-sequence nil))
447 (cmd (lookup-key hangul-im-keymap seq))
448 key)
449 (cond ((and (stringp seq)
450 (= 1 (length seq))
451 (setq key (aref seq 0))
452 (and (>= key 33) (< key 127)))
453 (hangul3-input-method-internal key))
454 ((commandp cmd)
455 (call-interactively cmd))
457 (setq unread-command-events (listify-key-sequence seq))
458 (throw 'exit-input-loop nil))))))
459 (quail-delete-overlays)))))
461 ;; Support function for `hangul390-input-method'. Actually, this
462 ;; function handles the Hangul 3-Bulsik 390. KEY is an entered key
463 ;; code used for looking up `hangul390-keymap'."
464 (defun hangul390-input-method-internal (key)
465 (let ((char (aref hangul390-keymap (- key 33))))
466 (cond ((or (and (> char 86) (< char 91))
467 (and (> char 96) (< char 123)))
468 (hangul3-input-method-cho (- char (if (< char 97) 86 92))))
469 ((and (> char 64) (< char 86))
470 (hangul3-input-method-jung (- char 34)))
471 ((< char 31)
472 (hangul3-input-method-jong char))
474 (setq hangul-queue (make-vector 6 0))
475 (insert (decode-char 'ucs char))
476 (move-overlay quail-overlay (point) (point))))))
478 (defun hangul390-input-method (key)
479 "3-Bulsik 390 input method."
480 (if (or buffer-read-only (< key 33) (>= key 127))
481 (list key)
482 (quail-setup-overlays nil)
483 (let ((input-method-function nil)
484 (echo-keystrokes 0)
485 (help-char nil))
486 (setq hangul-queue (make-vector 6 0))
487 (hangul390-input-method-internal key)
488 (unwind-protect
489 (catch 'exit-input-loop
490 (while t
491 (let* ((seq (read-key-sequence nil))
492 (cmd (lookup-key hangul-im-keymap seq))
493 key)
494 (cond ((and (stringp seq)
495 (= 1 (length seq))
496 (setq key (aref seq 0))
497 (and (>= key 33) (< key 127)))
498 (hangul390-input-method-internal key))
499 ((commandp cmd)
500 (call-interactively cmd))
502 (setq unread-command-events (listify-key-sequence seq))
503 (throw 'exit-input-loop nil))))))
504 (quail-delete-overlays)))))
506 ;; Text shown by describe-input-method. Set to a proper text by
507 ;; hangul-input-method-activate.
508 (defvar hangul-input-method-help-text nil)
509 (make-variable-buffer-local 'hangul-input-method-help-text)
511 (defun hangul-input-method-activate (input-method func help-text &rest args)
512 "Activate Hangul input method INPUT-METHOD.
513 FUNC is a function to handle input key.
514 HELP-TEXT is a text set in `hangul-input-method-help-text'."
515 (setq deactivate-current-input-method-function 'hangul-input-method-deactivate
516 describe-current-input-method-function 'hangul-input-method-help
517 hangul-input-method-help-text help-text)
518 (quail-delete-overlays)
519 (if (eq (selected-window) (minibuffer-window))
520 (add-hook 'minibuffer-exit-hook 'quail-exit-from-minibuffer))
521 (set (make-local-variable 'input-method-function) func))
523 (defun hangul-input-method-deactivate ()
524 "Deactivate the current Hangul input method."
525 (interactive)
526 (unwind-protect
527 (progn
528 (quail-hide-guidance)
529 (quail-delete-overlays)
530 (setq describe-current-input-method-function nil))
531 (kill-local-variable 'input-method-function)))
533 (define-obsolete-function-alias
534 'hangul-input-method-inactivate
535 'hangul-input-method-deactivate "24.3")
537 (defun hangul-input-method-help ()
538 "Describe the current Hangul input method."
539 (interactive)
540 (with-output-to-temp-buffer "*Help*"
541 (princ hangul-input-method-help-text)))
543 (provide 'hangul)
545 ;;; hangul.el ends here