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 (let ((len (length string))
37 (cond ((eq type 'list)
38 (setq val (make-list len 0))
41 (setcar l (aref string i))
42 (setq l (cdr l) i (1+ i)))))
44 (setq val (make-vector len 0))
46 (aset val i (aref string i))
49 (error "Invalid type: %s" type)))
53 (defsubst string-to-list (string)
54 "Return a list of characters in STRING."
55 (string-to-sequence string 'list))
58 (defsubst string-to-vector (string)
59 "Return a vector of characters in STRING."
60 (string-to-sequence string 'vector))
63 (defun store-substring (string idx obj)
64 "Embed OBJ (string or character) at index IDX of STRING."
67 (let ((len1 (length obj))
68 (len2 (length string))
71 (aset string (+ idx i) (aref obj 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 (aref str idx)
100 column (+ column (char-width 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) padding)))
108 (if (< end-column column)
111 (while (< column end-column)
112 (setq last-column column
115 column (+ column (char-width 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 compatibility ...
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 nested 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 first 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 argument %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 argument %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 the string indicating end-of-line format of CODING-SYSTEM."
216 (let* ((eol-type (coding-system-eol-type coding-system))
217 (val (cond ((vectorp eol-type) eol-mnemonic-undecided)
218 ((eq eol-type 0) eol-mnemonic-unix)
219 ((eq eol-type 1) eol-mnemonic-dos)
220 ((eq eol-type 2) eol-mnemonic-mac)
224 (char-to-string val))))
227 (defun coding-system-post-read-conversion (coding-system)
228 "Return the value of CODING-SYSTEM's post-read-conversion property."
229 (coding-system-get coding-system 'post-read-conversion))
232 (defun coding-system-pre-write-conversion (coding-system)
233 "Return the value of CODING-SYSTEM's pre-write-conversion property."
234 (coding-system-get coding-system 'pre-write-conversion))
237 (defun coding-system-translation-table-for-decode (coding-system)
238 "Return the value of CODING-SYSTEM's translation-table-for-decode property."
239 (coding-system-get coding-system 'translation-table-for-decode))
242 (defun coding-system-translation-table-for-encode (coding-system)
243 "Return the value of CODING-SYSTEM's translation-table-for-encode property."
244 (coding-system-get coding-system 'translation-table-for-encode))
247 (defun coding-system-equal (coding-system-1 coding-system-2)
248 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
249 Two coding systems are identical if two symbols are equal
250 or one is an alias of the other."
251 (or (eq coding-system-1 coding-system-2)
252 (and (equal (coding-system-spec coding-system-1)
253 (coding-system-spec coding-system-2))
254 (let ((eol-type-1 (coding-system-eol-type coding-system-1))
255 (eol-type-2 (coding-system-eol-type coding-system-2)))
256 (or (eq eol-type-1 eol-type-2)
257 (and (vectorp eol-type-1) (vectorp eol-type-2)))))))
260 (defmacro detect-coding-with-priority (from to priority-list)
261 "Detect a coding system of the text between FROM and TO with PRIORITY-LIST.
262 PRIORITY-LIST is an alist of coding categories vs the corresponding
263 coding systems ordered by priority."
265 (let* ((prio-list ,priority-list)
266 (coding-category-list coding-category-list)
267 ,@(mapcar (function (lambda (x) (list x x)))
268 coding-category-list))
269 (mapcar (function (lambda (x) (set (car x) (cdr x))))
271 (set-coding-priority (mapcar (function (lambda (x) (car x)))
273 (detect-coding-region ,from ,to))
274 ;; We must restore the internal database.
275 (set-coding-priority coding-category-list)
276 (update-coding-systems-internal)))
279 (defun detect-coding-with-language-environment (from to lang-env)
280 "Detect a coding system of the text between FROM and TO with LANG-ENV.
281 The detection takes into account the coding system priorities for the
282 language environment LANG-ENV."
283 (let ((coding-priority (get-language-info lang-env 'coding-priority)))
285 (detect-coding-with-priority
287 (mapcar (function (lambda (x)
288 (cons (coding-system-get x 'coding-category) x)))
290 (detect-coding-region from to))))
293 ;;; Composite character manipulations.
296 (defun compose-region (start end)
297 "Compose all characters in the current region into one composite character.
298 When called from a program, expects two arguments,
299 positions (integers or markers) specifying the region."
302 (let ((str (buffer-substring start end)))
304 (insert (compose-string str))
305 (delete-char (- end start)))))
308 (defun decompose-region (start end)
309 "Decompose all composite characters in the current region.
310 Composite characters are broken up into individual components.
311 When called from a program, expects two arguments,
312 positions (integers or markers) specifying the region."
314 (let ((buf (current-buffer))
315 (cmpchar-head (char-to-string leading-code-composition)))
317 (insert-buffer-substring buf start end)
318 (set-buffer-multibyte nil)
319 (goto-char (point-min))
320 (while (search-forward cmpchar-head nil t)
321 (if (looking-at "[\240-\377][\240-\377][\240-\377][\240-\377]+")
322 (let* ((from (1- (point)))
324 (str (string-as-multibyte (buffer-substring from to))))
325 (if (cmpcharp (string-to-char str))
327 (delete-region from to)
328 (insert (string-as-unibyte (decompose-string str))))
330 (set-buffer-multibyte t)
331 (let ((tempbuf (current-buffer)))
335 (delete-region start end)
336 (insert-buffer-substring tempbuf))))))
339 (defun decompose-string (string)
340 "Decompose all composite characters in STRING."
341 (let ((len (length string))
347 (setq ch (aref string idx))
348 (if (>= ch min-composite-char)
351 (setq str-list (cons (substring string i idx) str-list)))
352 (setq str-list (cons (decompose-composite-char ch) str-list))
356 (copy-sequence string)
358 (setq str-list (cons (substring string i idx) str-list)))
359 (apply 'concat (nreverse str-list)))))
362 (defconst reference-point-alist
363 '((tl . 0) (tc . 1) (tr . 2)
364 (ml . 3) (mc . 4) (mr . 5)
365 (bl . 6) (bc . 7) (br . 8)
366 (top-left . 0) (top-center . 1) (top-right . 2)
367 (mid-left . 3) (mid-center . 4) (mid-right . 5)
368 (bottom-left . 6) (bottom-center . 7) (bottom-right . 8)
369 (0 . 0) (1 . 1) (2 . 2)
370 (3 . 3) (4 . 4) (5 . 5)
371 (6 . 6) (7 . 7) (8 . 8))
372 "Alist of reference point symbols vs reference point codes.
373 A reference point symbol is to be used to specify a composition rule
374 while making a composite character by the function `compose-chars'
377 Meanings of reference point codes are as follows:
379 0----1----2 <-- ascent 0:tl or top-left
380 | | 1:tc or top-center
381 | | 2:tr or top-right
383 | 4 <--+---- center 4:mc or mid-center
384 | | 5:mr or mid-right
385 --- 3 5 <-- baseline 6:bl or bottom-left
386 | | 7:bc or bottom-center
387 6----7----8 <-- descent 8:br or bottom-right
389 Reference point symbols are to be used to specify composition rule of
390 the form \(GLOBAL-REF-POINT . NEW-REF-POINT), where GLOBAL-REF-POINT
391 is a reference point in the overall glyphs already composed, and
392 NEW-REF-POINT is a reference point in the new glyph to be added.
394 For instance, if GLOBAL-REF-POINT is 8 and NEW-REF-POINT is 1, the
395 overall glyph is updated as follows:
397 +-------+--+ <--- new ascent
401 --- | | | <--- baseline (doesn't change)
405 +----+-----+ <--- new descent
408 ;; Return a string for char CH to be embedded in multibyte form of
409 ;; composite character.
411 (defun compose-chars-component (ch)
413 (format "\240%c" (+ ch 128))
414 (let ((str (string-as-unibyte (char-to-string ch))))
416 (if (= (aref str 1) ?\xFF)
417 (error "Can't compose a rule-based composition character")
418 (substring str (if (= (aref str 1) ?\xFF) 2 1)))
419 (aset str 0 (+ (aref str 0) ?\x20))
422 ;; Return a string for composition rule RULE to be embedded in
423 ;; multibyte form of composite character.
424 (defsubst compose-chars-rule (rule)
425 (char-to-string (+ ?\xA0
426 (* (cdr (assq (car rule) reference-point-alist)) 9)
427 (cdr (assq (cdr rule) reference-point-alist)))))
430 (defun compose-chars (first-component &rest args)
431 "Return one char string composed from the arguments.
432 For relative composition, each argument should be a non-composition character
433 or a relative-composition character.
434 For rule-based composition, Nth (where N is odd) argument should be
435 a non-composition character or a rule-based-composition character,
436 and Mth (where M is even) argument should be a composition rule.
437 A composition rule has the form \(GLOBAL-REF-POINT . NEW-REF-POINT).
438 See the documentation of `reference-point-alist' for more detail."
439 (if (= (length args) 0)
440 (char-to-string first-component)
441 (let* ((with-rule (consp (car args)))
442 (str (if (cmpcharp first-component)
443 (string-as-unibyte (char-to-string first-component))
445 (concat (vector leading-code-composition ?\xFF)
446 (compose-chars-component first-component))
447 (concat (char-to-string leading-code-composition)
448 (compose-chars-component first-component))))))
449 (if (and (cmpcharp first-component)
450 (eq with-rule (/= (aref str 1) ?\xFF)))
451 (error "%s-compostion-character is not allowed in %s composition: %c"
452 (if with-rule "relative" "rule-based")
453 (if with-rule "rule-based" "relative")
457 (setq str (concat str (compose-chars-rule (car args)))
459 (if (cmpcharp (car args))
460 (let ((cmp-str (string-as-unibyte (char-to-string (car args)))))
461 (if (eq with-rule (/= (aref cmp-str 1) ?\xFF))
462 (error "%s-compostion-character is not allowed in %s composition: %c"
463 (if with-rule "relative" "rule-based")
464 (if with-rule "rule-based" "relative")
466 (setq str (concat str (substring cmp-str
467 (if with-rule 2 1)))))
468 (setq str (concat str (compose-chars-component (car args)))))
469 (setq args (cdr args)))
470 (string-as-multibyte str))))
473 (defun decompose-composite-char (char &optional type with-composition-rule)
474 "Convert composite character CHAR to a sequence of the components.
475 Optional 1st arg TYPE specifies the type of sequence returned.
476 It should be `string' (default), `list', or `vector'.
477 Optional 2nd arg WITH-COMPOSITION-RULE non-nil means the returned
478 sequence contains embedded composition rules if any. In this case, the
479 order of elements in the sequence is the same as arguments for
480 `compose-chars' to create CHAR.
481 If TYPE is omitted or is `string', composition rules are omitted
482 even if WITH-COMPOSITION-RULE is t."
485 (let* ((len (composite-char-component-count char))
488 (setq with-composition-rule (and with-composition-rule
489 (not (eq type 'string))
490 (composite-char-composition-rule-p char)))
492 (setq l (cons (composite-char-component char i) l))
493 (if with-composition-rule
494 (let ((rule (- (composite-char-composition-rule char i) ?\xA0)))
495 (setq l (cons (cons (/ rule 9) (% rule 9)) l))))
497 (setq l (cons (composite-char-component char 0) l))
498 (cond ((eq type 'string)
502 (t ; i.e. TYPE is vector
505 ;;; mule-util.el ends here