1 ;;; mule-util.el --- Utility functions for mulitilingual environment (mule)
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
6 ;; Keywords: mule, multilingual
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 2, or (at your option)
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; see the file COPYING. If not, write to the
22 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 ;; Boston, MA 02111-1307, USA.
27 ;;; String manipulations while paying attention to multibyte
31 (defun string-to-sequence (string type)
32 "Convert STRING to a sequence of TYPE which contains characters in STRING.
33 TYPE should be `list' or `vector'."
34 (or (eq type 'list) (eq type 'vector)
35 (error "Invalid type: %s" type))
36 (let* ((len (length string))
40 (setq ch (if enable-multibyte-characters
41 (sref string i) (aref string i)))
43 (setq i (+ i (char-bytes ch))))
50 (defsubst string-to-list (string)
51 "Return a list of characters in STRING."
52 (string-to-sequence string 'list))
55 (defsubst string-to-vector (string)
56 "Return a vector of characters in STRING."
57 (string-to-sequence string 'vector))
60 (defun store-substring (string idx obj)
61 "Embed OBJ (string or character) at index IDX of STRING."
62 (let* ((str (cond ((stringp obj) obj)
63 ((integerp obj) (char-to-string obj))
65 "Invalid argument (should be string or character): %s"
67 (string-len (length string))
70 (while (and (< i len) (< idx string-len))
71 (aset string idx (aref str i))
72 (setq idx (1+ idx) i (1+ i)))
76 (defun truncate-string-to-width (str end-column &optional start-column padding)
77 "Truncate string STR to end at column END-COLUMN.
78 The optional 2nd arg START-COLUMN, if non-nil, specifies
79 the starting column; that means to return the characters occupying
80 columns START-COLUMN ... END-COLUMN of STR.
82 The optional 3rd arg PADDING, if non-nil, specifies a padding character
83 to add at the end of the result if STR doesn't reach column END-COLUMN,
84 or if END-COLUMN comes in the middle of a character in STR.
85 PADDING is also added at the beginning of the result
86 if column START-COLUMN appears in the middle of a character in STR.
88 If PADDING is nil, no padding is added in these cases, so
89 the resulting string may be narrower than END-COLUMN."
91 (setq start-column 0))
92 (let ((len (length str))
95 (head-padding "") (tail-padding "")
96 ch last-column last-idx from-idx)
98 (while (< column start-column)
99 (setq ch (sref str idx)
100 column (+ column (char-width ch))
101 idx (+ idx (char-bytes ch))))
102 (args-out-of-range (setq idx len)))
103 (if (< column start-column)
104 (if padding (make-string end-column padding) "")
105 (if (and padding (> column start-column))
106 (setq head-padding (make-string (- column start-column) ?\ )))
108 (if (< end-column column)
111 (while (< column end-column)
112 (setq last-column column
115 column (+ column (char-width ch))
116 idx (+ idx (char-bytes ch))))
117 (args-out-of-range (setq idx len)))
118 (if (> column end-column)
119 (setq column last-column idx last-idx))
120 (if (and padding (< column end-column))
121 (setq tail-padding (make-string (- end-column column) padding))))
122 (setq str (substring str from-idx idx))
124 (concat head-padding str tail-padding)
127 ;;; For backward compatiblity ...
129 (defalias 'truncate-string 'truncate-string-to-width)
130 (make-obsolete 'truncate-string 'truncate-string-to-width)
132 ;;; Nested alist handler. Nested alist is alist whose elements are
133 ;;; also nested alist.
136 (defsubst nested-alist-p (obj)
137 "Return t if OBJ is a nesetd alist.
139 Nested alist is a list of the form (ENTRY . BRANCHES), where ENTRY is
140 any Lisp object, and BRANCHES is a list of cons cells of the form
141 (KEY-ELEMENT . NESTED-ALIST).
143 You can use a nested alist to store any Lisp object (ENTRY) for a key
144 sequence KEYSEQ, where KEYSEQ is a sequence of KEY-ELEMENT. KEYSEQ
145 can be a string, a vector, or a list."
146 (and obj (listp obj) (listp (cdr obj))))
149 (defun set-nested-alist (keyseq entry alist &optional len branches)
150 "Set ENTRY for KEYSEQ in a nested alist ALIST.
151 Optional 4th arg LEN non-nil means the firlst LEN elements in KEYSEQ
153 Optional argument BRANCHES if non-nil is branches for a keyseq
155 See the documentation of `nested-alist-p' for more detail."
156 (or (nested-alist-p alist)
157 (error "Invalid arguement %s" alist))
158 (let ((islist (listp keyseq))
159 (len (or len (length keyseq)))
163 (if (null (nested-alist-p alist))
164 (error "Keyseq %s is too long for this nested alist" keyseq))
165 (setq key-elt (if islist (nth i keyseq) (aref keyseq i)))
166 (setq slot (assoc key-elt (cdr alist)))
169 (setq slot (cons key-elt (list t)))
170 (setcdr alist (cons slot (cdr alist)))))
171 (setq alist (cdr slot))
176 (error "Can't set branches for keyseq %s" keyseq)
177 (setcdr alist branches)))))
180 (defun lookup-nested-alist (keyseq alist &optional len start nil-for-too-long)
181 "Look up key sequence KEYSEQ in nested alist ALIST. Return the definition.
182 Optional 1st argument LEN specifies the length of KEYSEQ.
183 Optional 2nd argument START specifies index of the starting key.
184 The returned value is normally a nested alist of which
185 car part is the entry for KEYSEQ.
186 If ALIST is not deep enough for KEYSEQ, return number which is
187 how many key elements at the front of KEYSEQ it takes
188 to reach a leaf in ALIST.
189 Optional 3rd argument NIL-FOR-TOO-LONG non-nil means return nil
190 even if ALIST is not deep enough."
191 (or (nested-alist-p alist)
192 (error "invalid arguement %s" alist))
194 (setq len (length keyseq)))
195 (let ((i (or start 0)))
196 (if (catch 'lookup-nested-alist-tag
199 (if (setq alist (cdr (assoc (nth i keyseq) (cdr alist))))
201 (throw 'lookup-nested-alist-tag t))))
203 (if (setq alist (cdr (assoc (aref keyseq i) (cdr alist))))
205 (throw 'lookup-nested-alist-tag t))))
206 ;; KEYSEQ is too long.
207 (if nil-for-too-long nil i)
211 ;; Coding system related functions.
214 (defun coding-system-eol-type-mnemonic (coding-system)
215 "Return mnemonic letter of eol-type of CODING-SYSTEM."
216 (let ((eol-type (coding-system-eol-type coding-system)))
217 (cond ((vectorp eol-type) eol-mnemonic-undecided)
218 ((eq eol-type 0) eol-mnemonic-unix)
219 ((eq eol-type 1) eol-mnemonic-unix)
220 ((eq eol-type 2) eol-mnemonic-unix)
224 (defun coding-system-post-read-conversion (coding-system)
225 "Return the value of CODING-SYSTEM's post-read-conversion property."
226 (coding-system-get coding-system 'post-read-conversion))
229 (defun coding-system-pre-write-conversion (coding-system)
230 "Return the value of CODING-SYSTEM's pre-write-conversion property."
231 (coding-system-get coding-system 'pre-write-conversion))
234 (defun coding-system-unification-table-for-decode (coding-system)
235 "Return the value of CODING-SYSTEM's unification-table-for-decode property."
236 (coding-system-get coding-system 'character-unification-table-for-decode))
239 (defun coding-system-unification-table-for-encode (coding-system)
240 "Return the value of CODING-SYSTEM's unification-table-for-encode property."
241 (coding-system-get coding-system 'character-unification-table-for-encode))
243 (defun coding-system-lessp (x y)
244 (cond ((eq x 'no-conversion) t)
245 ((eq y 'no-conversion) nil)
246 ((eq x 'emacs-mule) t)
247 ((eq y 'emacs-mule) nil)
248 ((eq x 'undecided) t)
249 ((eq y 'undecided) nil)
250 (t (let ((c1 (coding-system-mnemonic x))
251 (c2 (coding-system-mnemonic y)))
252 (or (< (downcase c1) (downcase c2))
253 (and (not (> (downcase c1) (downcase c2)))
257 (defun coding-system-list (&optional base-only)
258 "Return a list of all existing coding systems.
259 If optional arg BASE-ONLY is non-nil, only base coding systems are listed."
260 (let* ((codings (sort (copy-sequence coding-system-list)
261 'coding-system-lessp))
262 (tail (cons nil codings)))
263 ;; Remove subsidiary coding systems (eol variants) and alias
264 ;; coding systems (if necessary).
266 (let* ((coding (car (cdr tail)))
267 (aliases (coding-system-get coding 'alias-coding-systems)))
269 ;; CODING is an eol varinant if not in ALIASES.
270 (not (memq coding aliases))
271 ;; CODING is an alias if it is not car of ALISES.
272 (and base-only (not (eq coding (car aliases)))))
273 (setcdr tail (cdr (cdr tail)))
274 (setq tail (cdr tail)))))
278 (defun coding-system-equal (coding-system-1 coding-system-2)
279 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
280 Two coding systems are identical if two symbols are equal
281 or one is an alias of the other."
282 (or (eq coding-system-1 coding-system-2)
283 (and (equal (coding-system-spec coding-system-1)
284 (coding-system-spec coding-system-2))
285 (let ((eol-type-1 (coding-system-eol-type coding-system-1))
286 (eol-type-2 (coding-system-eol-type coding-system-2)))
287 (or (eq eol-type-1 eol-type-2)
288 (and (vectorp eol-type-1) (vectorp eol-type-2)))))))
291 (defun find-safe-coding-system (from to)
292 "Return a list of proper coding systems to encode a text between FROM and TO.
293 All coding systems in the list can safely encode any multibyte characters
296 If the region contains no multibyte charcters, the returned list
297 contains a single element `undecided'.
299 Kludgy feature: if FROM is a string, then that string is the target
300 for finding proper coding systems, and TO is ignored."
301 (let ((found (if (stringp from)
302 (find-charset-string from)
303 (find-charset-region from to)))
304 (l coding-system-list)
306 (if (and (= (length found) 1)
307 (eq 'ascii (car found)))
310 (setq coding (car l) l (cdr l))
311 (if (and (eq coding (coding-system-base coding))
312 (setq safe (coding-system-get coding 'safe-charsets))
315 (mapcar (function (lambda (x)
316 (if (not (memq x safe))
319 (setq codings (cons coding codings))))
323 ;;; Composite charcater manipulations.
326 (defun compose-region (start end)
327 "Compose all characters in the current region into one composite character.
328 When called from a program, expects two arguments,
329 positions (integers or markers) specifying the region."
332 (let ((str (buffer-substring start end)))
334 (delete-region start end)
335 (insert (compose-string str)))))
338 (defun decompose-region (start end)
339 "Decompose all composite characters in the current region.
340 Composite characters are broken up into individual components.
341 When called from a program, expects two arguments,
342 positions (integers or markers) specifying the region."
345 (narrow-to-region start end)
346 (goto-char (point-min))
347 (let ((enable-multibyte-characters nil)
348 ;; This matches the whole bytes of single composite character.
349 (re-cmpchar "\200[\240-\377]+")
351 (while (re-search-forward re-cmpchar nil t)
352 (setq str (buffer-substring (match-beginning 0) (match-end 0)))
353 (delete-region (match-beginning 0) (match-end 0))
354 (insert (decompose-composite-char (string-to-char str)))))))
357 (defun decompose-string (string)
358 "Decompose all composite characters in STRING."
359 (let* ((l (string-to-list string))
364 (setcar tail (if (cmpcharp ch) (decompose-composite-char ch)
365 (char-to-string ch)))
366 (setq tail (cdr tail)))
370 (defconst reference-point-alist
371 '((tl . 0) (tc . 1) (tr . 2)
372 (ml . 3) (mc . 4) (mr . 5)
373 (bl . 6) (bc . 7) (br . 8)
374 (top-left . 0) (top-center . 1) (top-right . 2)
375 (mid-left . 3) (mid-center . 4) (mid-right . 5)
376 (bottom-left . 6) (bottom-center . 7) (bottom-right . 8)
377 (0 . 0) (1 . 1) (2 . 2)
378 (3 . 3) (4 . 4) (5 . 5)
379 (6 . 6) (7 . 7) (8 . 8))
380 "Alist of reference point symbols vs reference point codes.
381 A reference point symbol is to be used to specify a composition rule
382 while making a composite character by the function `compose-chars'
385 Meanings of reference point codes are as follows:
387 0----1----2 <-- ascent 0:tl or top-left
388 | | 1:tc or top-center
389 | | 2:tr or top-right
391 | 4 <--+---- center 4:mc or mid-center
392 | | 5:mr or mid-right
393 --- 3 5 <-- baseline 6:bl or bottom-left
394 | | 7:bc or bottom-center
395 6----7----8 <-- descent 8:br or bottom-right
397 Reference point symbols are to be used to specify composition rule of
398 the form \(GLOBAL-REF-POINT . NEW-REF-POINT), where GLOBAL-REF-POINT
399 is a reference point in the overall glyphs already composed, and
400 NEW-REF-POINT is a reference point in the new glyph to be added.
402 For instance, if GLOBAL-REF-POINT is 8 and NEW-REF-POINT is 1, the
403 overall glyph is updated as follows:
405 +-------+--+ <--- new ascent
409 --- | | | <--- baseline (doesn't change)
413 +----+-----+ <--- new descent
416 ;; Return a string for char CH to be embedded in multibyte form of
417 ;; composite character.
418 (defun compose-chars-component (ch)
420 (format "\240%c" (+ ch 128))
421 (let ((str (char-to-string ch)))
423 (substring str (if (= (aref str 1) ?\xFF) 2 1))
424 (aset str 0 (+ (aref str 0) ?\x20))
427 ;; Return a string for composition rule RULE to be embedded in
428 ;; multibyte form of composite character.
429 (defsubst compose-chars-rule (rule)
430 (char-to-string (+ ?\xA0
431 (* (cdr (assq (car rule) reference-point-alist)) 9)
432 (cdr (assq (cdr rule) reference-point-alist)))))
435 (defun compose-chars (first-component &rest args)
436 "Return one char string composed from the arguments.
437 Each argument is a character (including a composite chararacter)
438 or a composition rule.
439 A composition rule has the form \(GLOBAL-REF-POINT . NEW-REF-POINT).
440 See the documentation of `reference-point-alist' for more detail."
441 (if (= (length args) 0)
442 (char-to-string first-component)
443 (let* ((with-rule (consp (car args)))
444 (str (if with-rule (concat (vector leading-code-composition ?\xFF))
445 (char-to-string leading-code-composition))))
446 (setq str (concat str (compose-chars-component first-component)))
450 (if (not (consp (car args)))
451 (error "Invalid composition rule: %s" (car args)))
452 (setq str (concat str (compose-chars-rule (car args))
453 (compose-chars-component (car (cdr args))))
454 args (cdr (cdr args))))
455 (setq str (concat str (compose-chars-component (car args)))
460 (defun decompose-composite-char (char &optional type with-composition-rule)
461 "Convert composite character CHAR to a string containing components of CHAR.
462 Optional 1st arg TYPE specifies the type of sequence returned.
463 It should be `string' (default), `list', or `vector'.
464 Optional 2nd arg WITH-COMPOSITION-RULE non-nil means the returned
465 sequence contains embedded composition rules if any. In this case, the
466 order of elements in the sequence is the same as arguments for
467 `compose-chars' to create CHAR.
468 If TYPE is omitted or is `string', composition rules are omitted
469 even if WITH-COMPOSITION-RULE is t."
472 (let* ((len (composite-char-component-count char))
475 (setq with-composition-rule (and with-composition-rule
476 (not (eq type 'string))
477 (composite-char-composition-rule-p char)))
479 (setq l (cons (composite-char-component char i) l))
480 (if with-composition-rule
481 (let ((rule (- (composite-char-composition-rule char i) ?\xA0)))
482 (setq l (cons (cons (/ rule 9) (% rule 9)) l))))
484 (setq l (cons (composite-char-component char 0) l))
485 (cond ((eq type 'string)
486 (apply 'concat-chars l))
489 (t ; i.e. TYPE is vector
492 ;;; mule-util.el ends here