1 ;;; mule-util.el --- Utility functions for mulitilingual environment (mule)
3 ;; Copyright (C) 1995 Free Software Foundation, Inc.
4 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
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 Multibyte characters are conserned."
35 (or (eq type 'list) (eq type 'vector)
36 (error "Invalid type: %s" type))
37 (let* ((len (length string))
41 (setq ch (sref 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 width &optional start-column padding)
77 "Truncate string STR to fit in WIDTH columns.
78 Optional 1st arg START-COLUMN if non-nil specifies the starting column.
79 Optional 2nd arg PADDING if non-nil is a padding character to be padded at
80 the head and tail of the resulting string to fit in WIDTH if necessary.
81 If PADDING is nil, the resulting string may be narrower than WIDTH."
83 (setq start-column 0))
84 (let ((len (length str))
87 (head-padding "") (tail-padding "")
88 ch last-column last-idx from-idx)
90 (while (< column start-column)
91 (setq ch (sref str idx)
92 column (+ column (char-width ch))
93 idx (+ idx (char-bytes ch))))
94 (args-out-of-range (setq idx len)))
95 (if (< column start-column)
96 (if padding (make-string width padding) "")
97 (if (and padding (> column start-column))
98 (setq head-padding (make-string (- column start-column) ?\ )))
101 (while (< column width)
102 (setq last-column column
105 column (+ column (char-width ch))
106 idx (+ idx (char-bytes ch))))
107 (args-out-of-range (setq idx len)))
109 (setq column last-column idx last-idx))
110 (if (and padding (< column width))
111 (setq tail-padding (make-string (- width column) padding)))
112 (setq str (substring str from-idx idx))
114 (concat head-padding str tail-padding)
117 ;;; For backward compatiblity ...
119 (defalias 'truncate-string 'truncate-string-to-width)
120 (make-obsolete 'truncate-string 'truncate-string-to-width)
122 ;;; Nested alist handler. Nested alist is alist whose elements are
123 ;;; also nested alist.
126 (defsubst nested-alist-p (obj)
127 "Return t if OBJ is a nesetd alist.
129 Nested alist is a list of the form (ENTRY . BRANCHES), where ENTRY is
130 any Lisp object, and BRANCHES is a list of cons cells of the form
131 (KEY-ELEMENT . NESTED-ALIST).
133 You can use a nested alist to store any Lisp object (ENTRY) for a key
134 sequence KEYSEQ, where KEYSEQ is a sequence of KEY-ELEMENT. KEYSEQ
135 can be a string, a vector, or a list."
136 (and obj (listp obj) (listp (cdr obj))))
139 (defun set-nested-alist (keyseq entry alist &optional len branches)
140 "Set ENTRY for KEYSEQ in a nested alist ALIST.
141 Optional 4th arg LEN non-nil means the firlst LEN elements in KEYSEQ
143 Optional argument BRANCHES if non-nil is branches for a keyseq
145 See the documentation of `nested-alist-p' for more detail."
146 (or (nested-alist-p alist)
147 (error "Invalid arguement %s" alist))
148 (let ((islist (listp keyseq))
149 (len (or len (length keyseq)))
153 (if (null (nested-alist-p alist))
154 (error "Keyseq %s is too long for this nested alist" keyseq))
155 (setq key-elt (if islist (nth i keyseq) (aref keyseq i)))
156 (setq slot (assoc key-elt (cdr alist)))
159 (setq slot (cons key-elt (list t)))
160 (setcdr alist (cons slot (cdr alist)))))
161 (setq alist (cdr slot))
166 (error "Can't set branches for keyseq %s" keyseq)
167 (setcdr alist branches)))))
170 (defun lookup-nested-alist (keyseq alist &optional len start nil-for-too-long)
171 "Look up key sequence KEYSEQ in nested alist ALIST. Return the definition.
172 Optional 1st argument LEN specifies the length of KEYSEQ.
173 Optional 2nd argument START specifies index of the starting key.
174 The returned value is normally a nested alist of which
175 car part is the entry for KEYSEQ.
176 If ALIST is not deep enough for KEYSEQ, return number which is
177 how many key elements at the front of KEYSEQ it takes
178 to reach a leaf in ALIST.
179 Optional 3rd argument NIL-FOR-TOO-LONG non-nil means return nil
180 even if ALIST is not deep enough."
181 (or (nested-alist-p alist)
182 (error "invalid arguement %s" alist))
184 (setq len (length keyseq)))
185 (let ((i (or start 0)))
186 (if (catch 'lookup-nested-alist-tag
189 (if (setq alist (cdr (assoc (nth i keyseq) (cdr alist))))
191 (throw 'lookup-nested-alist-tag t))))
193 (if (setq alist (cdr (assoc (aref keyseq i) (cdr alist))))
195 (throw 'lookup-nested-alist-tag t))))
196 ;; KEYSEQ is too long.
197 (if nil-for-too-long nil i)
200 ;; Coding system related functions.
203 (defun set-coding-system-alist (target-type regexp coding-system
205 "Update `coding-system-alist' according to the arguments.
206 TARGET-TYPE specifies a type of the target: `file', `process', or `network'.
207 TARGET-TYPE tells which slots of coding-system-alist should be affected.
208 If `file', it affects slots for insert-file-contents and write-region.
209 If `process', it affects slots for call-process, call-process-region, and
211 If `network', it affects a slot for open-network-process.
212 REGEXP is a regular expression matching a target of I/O operation.
213 CODING-SYSTEM is a coding system to perform code conversion
214 on the I/O operation, or a cons of coding systems for decoding and
215 encoding respectively, or a function symbol which returns the cons.
216 Optional arg OPERATION if non-nil specifies directly one of slots above.
217 The valid value is: insert-file-contents, write-region,
218 call-process, call-process-region, start-process, or open-network-stream.
219 If OPERATION is specified, TARGET-TYPE is ignored.
220 See the documentation of `coding-system-alist' for more detail."
222 (error "Invalid regular expression: %s" regexp))
223 (or (memq target-type '(file process network))
224 (error "Invalid target type: %s" target-type))
225 (if (symbolp coding-system)
226 (if (not (fboundp coding-system))
228 (check-coding-system coding-system)
229 (setq coding-system (cons coding-system coding-system))))
230 (check-coding-system (car coding-system))
231 (check-coding-system (cdr coding-system)))
232 (let ((op-list (if operation (list operation)
233 (cond ((eq target-type 'file)
234 '(insert-file-contents write-region))
235 ((eq target-type 'process)
236 '(call-process call-process-region start-process))
237 (t ; i.e. (eq target-type network)
238 '(open-network-stream)))))
241 (setq slot (assq (car op-list) coding-system-alist))
243 (let ((chain (cdr slot)))
246 (if (string= regexp (car (car chain)))
248 (setcdr (car chain) coding-system)
250 (setq chain (cdr chain)))
252 (setcdr slot (cons (cons regexp coding-system) (cdr slot)))))
253 (setq coding-system-alist
254 (cons (cons (car op-list) (list (cons regexp coding-system)))
255 coding-system-alist)))
256 (setq op-list (cdr op-list)))))
259 (defun coding-system-list ()
260 "Return a list of all existing coding systems."
262 (mapatoms (lambda (x) (if (get x 'coding-system) (setq l (cons x l)))))
266 ;;; Composite charcater manipulations.
269 (defun compose-region (start end)
270 "Compose all characters in the current region into one composite character.
271 When called from a program, expects two arguments,
272 positions (integers or markers) specifying the region."
275 (let ((str (buffer-substring start end)))
277 (delete-region start end)
278 (insert (compose-string str)))))
281 (defun decompose-region (start end)
282 "Decompose all composite characters in the current region.
283 Composite characters are broken up into individual components.
284 When called from a program, expects two arguments,
285 positions (integers or markers) specifying the region."
288 (narrow-to-region start end)
289 (goto-char (point-min))
290 (let ((enable-multibyte-characters nil)
291 ;; This matches the whole bytes of single composite character.
292 (re-cmpchar "\200[\240-\377]+")
294 (while (re-search-forward re-cmpchar nil t)
295 (setq str (buffer-substring (match-beginning 0) (match-end 0)))
296 (delete-region (match-beginning 0) (match-end 0))
297 (insert (decompose-composite-char (string-to-char str)))))))
300 (defconst reference-point-alist
301 '((tl . 0) (tc . 1) (tr . 2)
302 (ml . 3) (mc . 4) (mr . 5)
303 (bl . 6) (bc . 7) (br . 8)
304 (top-left . 0) (top-center . 1) (top-right . 2)
305 (mid-left . 3) (mid-center . 4) (mid-right . 5)
306 (bottom-left . 6) (bottom-center . 7) (bottom-right . 8)
307 (0 . 0) (1 . 1) (2 . 2)
308 (3 . 3) (4 . 4) (5 . 5)
309 (6 . 6) (7 . 7) (8 . 8))
310 "Alist of reference point symbols vs reference point codes.
311 Meanings of reference point codes are as follows:
313 0----1----2 <-- ascent 0:tl or top-left
314 | | 1:tc or top-center
315 | | 2:tr or top-right
317 | 4 <--+---- center 4:mc or mid-center
318 | | 5:mr or mid-right
319 --- 3 5 <-- baseline 6:bl or bottom-left
320 | | 7:bc or bottom-center
321 6----7----8 <-- descent 8:br or bottom-right
323 Reference point symbols are to be used to specify composition rule of
324 the form \(GLOBAL-REF-POINT . NEW-REF-POINT), where GLOBAL-REF-POINT
325 is a reference point in the overall glyphs already composed, and
326 NEW-REF-POINT is a reference point in the new glyph to be added.
328 For instance, if GLOBAL-REF-POINT is 8 and NEW-REF-POINT is 1, the
329 overall glyph is updated as follows:
331 +-------+--+ <--- new ascent
335 --- | | | <--- baseline (doesn't change)
339 +----+-----+ <--- new descent
342 ;; Return a string for char CH to be embedded in multibyte form of
343 ;; composite character.
344 (defun compose-chars-component (ch)
346 (format "\240%c" (+ ch 128))
347 (let ((str (char-to-string ch)))
349 (if (/= (aref str 1) ?\xFF)
350 (error "Char %c can't be composed" ch)
352 (aset str 0 (+ (aref str 0) ?\x20))
355 ;; Return a string for composition rule RULE to be embedded in
356 ;; multibyte form of composite character.
357 (defsubst compose-chars-rule (rule)
358 (char-to-string (+ ?\xA0
359 (* (cdr (assq (car rule) reference-point-alist)) 9)
360 (cdr (assq (cdr rule) reference-point-alist)))))
363 (defun compose-chars (first-component &rest args)
364 "Return one char string composed from the arguments.
365 Each argument is a character (including a composite chararacter)
366 or a composition rule.
367 A composition rule has the form \(GLOBAL-REF-POINT . NEW-REF-POINT).
368 See the documentation of `reference-point-alist' for more detail."
369 (if (= (length args) 0)
370 (char-to-string first-component)
371 (let* ((with-rule (consp (car args)))
372 (str (if with-rule (concat (vector leading-code-composition ?\xFF))
373 (char-to-string leading-code-composition))))
374 (setq str (concat str (compose-chars-component first-component)))
378 (if (not (consp (car args)))
379 (error "Invalid composition rule: %s" (car args)))
380 (setq str (concat str (compose-chars-rule (car args))
381 (compose-chars-component (car (cdr args))))
382 args (cdr (cdr args))))
383 (setq str (concat str (compose-chars-component (car args)))
388 (defun decompose-composite-char (char &optional type with-composition-rule)
389 "Convert composite character CHAR to a string containing components of CHAR.
390 Optional 1st arg TYPE specifies the type of sequence returned.
391 It should be `string' (default), `list', or `vector'.
392 Optional 2nd arg WITH-COMPOSITION-RULE non-nil means the returned
393 sequence contains embedded composition rules if any. In this case, the
394 order of elements in the sequence is the same as arguments for
395 `compose-chars' to create CHAR.
396 If TYPE is omitted or is `string', composition rules are omitted
397 even if WITH-COMPOSITION-RULE is t."
400 (let* ((len (composite-char-component-count char))
403 (setq with-composition-rule (and with-composition-rule
404 (not (eq type 'string))
405 (composite-char-composition-rule-p char)))
407 (setq l (cons (composite-char-component char i) l))
408 (if with-composition-rule
409 (let ((rule (- (composite-char-composition-rule char i) ?\xA0)))
410 (setq l (cons (cons (/ rule 9) (% rule 9)) l))))
412 (setq l (cons (composite-char-component char 0) l))
413 (cond ((eq type 'string)
414 (apply 'concat-chars l))
417 (t ; i.e. TYPE is vector
420 ;;; mule-util.el ends here