Mercurial > hg > xemacs-beta

comparison lisp/mule/mule-util.el @ 165:5a88923fcbfe r20-3b9

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Import from CVS: tag r20-3b9
author cvs
date Mon, 13 Aug 2007 09:44:42 +0200
parents
children 85ec50267440
comparison
equal deleted inserted replaced
164:4e0740e5aab2 165:5a88923fcbfe
1 ;;; mule-util.el --- Utility functions for mulitilingual environment (mule)
2
3 ;; Copyright (C) 1995 Free Software Foundation, Inc.
4 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
5 ;; Copyright (C) 1997 MORIOKA Tomohiko
6
7 ;; Keywords: mule, multilingual
8
9 ;; This file is part of XEmacs.
10
11 ;; XEmacs is free software; you can redistribute it and/or modify it
12 ;; under the terms of the GNU General Public License as published by
13 ;; the Free Software Foundation; either version 2, or (at your option)
14 ;; any later version.
15
16 ;; XEmacs is distributed in the hope that it will be useful, but
17 ;; WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 ;; General Public License for more details.
20
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with XEmacs; see the file COPYING. If not, write to the Free
23 ;; Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
24 ;; 02111-1307, USA.
25
26 ;;; Code:
27
28 ;;; String manipulations while paying attention to multibyte
29 ;;; characters.
30
31 ;;;###autoload
32 (defsubst string-to-sequence (string type)
33 "Convert STRING to a sequence of TYPE which contains characters in STRING.
34 TYPE should be `list' or `vector'.
35 Multibyte characters are conserned."
36 (map type (function identity) string))
37
38 ;;;###autoload
39 (defsubst string-to-list (string)
40 "Return a list of characters in STRING."
41 (mapcar (function identity) string))
42
43 ;;;###autoload
44 (defsubst string-to-vector (string)
45 "Return a vector of characters in STRING."
46 (string-to-sequence string 'vector))
47
48 ;;;###autoload
49 (defun store-substring (string idx obj)
50 "Embed OBJ (string or character) at index IDX of STRING."
51 (let* ((str (cond ((stringp obj) obj)
52 ((characterp obj) (char-to-string obj))
53 (t (error
54 "Invalid argument (should be string or character): %s"
55 obj))))
56 (string-len (length string))
57 (len (length str))
58 (i 0))
59 (while (and (< i len) (< idx string-len))
60 (aset string idx (aref str i))
61 (setq idx (1+ idx) i (1+ i)))
62 string))
63
64 ;;;###autoload
65 (defun truncate-string-to-width (str width &optional start-column padding)
66 "Truncate string STR to fit in WIDTH columns.
67 Optional 1st arg START-COLUMN if non-nil specifies the starting column.
68 Optional 2nd arg PADDING if non-nil is a padding character to be padded at
69 the head and tail of the resulting string to fit in WIDTH if necessary.
70 If PADDING is nil, the resulting string may be narrower than WIDTH."
71 (or start-column
72 (setq start-column 0))
73 (let ((len (length str))
74 (idx 0)
75 (column 0)
76 (head-padding "") (tail-padding "")
77 ch last-column last-idx from-idx)
78 (condition-case nil
79 (while (< column start-column)
80 (setq ch (sref str idx)
81 column (+ column (char-width ch))
82 idx (+ idx (char-bytes ch))))
83 (args-out-of-range (setq idx len)))
84 (if (< column start-column)
85 (if padding (make-string width padding) "")
86 (if (and padding (> column start-column))
87 (setq head-padding (make-string (- column start-column) ?\ )))
88 (setq from-idx idx)
89 (condition-case nil
90 (while (< column width)
91 (setq last-column column
92 last-idx idx
93 ch (sref str idx)
94 column (+ column (char-width ch))
95 idx (+ idx (char-bytes ch))))
96 (args-out-of-range (setq idx len)))
97 (if (> column width)
98 (setq column last-column idx last-idx))
99 (if (and padding (< column width))
100 (setq tail-padding (make-string (- width column) padding)))
101 (setq str (substring str from-idx idx))
102 (if padding
103 (concat head-padding str tail-padding)
104 str))))
105
106 ;;; For backward compatiblity ...
107 ;;;###autoload
108 (defalias 'truncate-string 'truncate-string-to-width)
109 (make-obsolete 'truncate-string 'truncate-string-to-width)
110
111 ;;; Nested alist handler. Nested alist is alist whose elements are
112 ;;; also nested alist.
113
114 ;;;###autoload
115 (defsubst nested-alist-p (obj)
116 "Return t if OBJ is a nesetd alist.
117
118 Nested alist is a list of the form (ENTRY . BRANCHES), where ENTRY is
119 any Lisp object, and BRANCHES is a list of cons cells of the form
120 (KEY-ELEMENT . NESTED-ALIST).
121
122 You can use a nested alist to store any Lisp object (ENTRY) for a key
123 sequence KEYSEQ, where KEYSEQ is a sequence of KEY-ELEMENT. KEYSEQ
124 can be a string, a vector, or a list."
125 (and obj (listp obj) (listp (cdr obj))))
126
127 ;;;###autoload
128 (defun set-nested-alist (keyseq entry alist &optional len branches)
129 "Set ENTRY for KEYSEQ in a nested alist ALIST.
130 Optional 4th arg LEN non-nil means the firlst LEN elements in KEYSEQ
131 is considered.
132 Optional argument BRANCHES if non-nil is branches for a keyseq
133 longer than KEYSEQ.
134 See the documentation of `nested-alist-p' for more detail."
135 (or (nested-alist-p alist)
136 (error "Invalid arguement %s" alist))
137 (let ((islist (listp keyseq))
138 (len (or len (length keyseq)))
139 (i 0)
140 key-elt slot)
141 (while (< i len)
142 (if (null (nested-alist-p alist))
143 (error "Keyseq %s is too long for this nested alist" keyseq))
144 (setq key-elt (if islist (nth i keyseq) (aref keyseq i)))
145 (setq slot (assoc key-elt (cdr alist)))
146 (if (null slot)
147 (progn
148 (setq slot (cons key-elt (list t)))
149 (setcdr alist (cons slot (cdr alist)))))
150 (setq alist (cdr slot))
151 (setq i (1+ i)))
152 (setcar alist entry)
153 (if branches
154 (if (cdr alist)
155 (error "Can't set branches for keyseq %s" keyseq)
156 (setcdr alist branches)))))
157
158 ;;;###autoload
159 (defun lookup-nested-alist (keyseq alist &optional len start nil-for-too-long)
160 "Look up key sequence KEYSEQ in nested alist ALIST. Return the definition.
161 Optional 1st argument LEN specifies the length of KEYSEQ.
162 Optional 2nd argument START specifies index of the starting key.
163 The returned value is normally a nested alist of which
164 car part is the entry for KEYSEQ.
165 If ALIST is not deep enough for KEYSEQ, return number which is
166 how many key elements at the front of KEYSEQ it takes
167 to reach a leaf in ALIST.
168 Optional 3rd argument NIL-FOR-TOO-LONG non-nil means return nil
169 even if ALIST is not deep enough."
170 (or (nested-alist-p alist)
171 (error "invalid arguement %s" alist))
172 (or len
173 (setq len (length keyseq)))
174 (let ((i (or start 0)))
175 (if (catch 'lookup-nested-alist-tag
176 (if (listp keyseq)
177 (while (< i len)
178 (if (setq alist (cdr (assoc (nth i keyseq) (cdr alist))))
179 (setq i (1+ i))
180 (throw 'lookup-nested-alist-tag t))))
181 (while (< i len)
182 (if (setq alist (cdr (assoc (aref keyseq i) (cdr alist))))
183 (setq i (1+ i))
184 (throw 'lookup-nested-alist-tag t))))
185 ;; KEYSEQ is too long.
186 (if nil-for-too-long nil i)
187 alist)))
188
189 ;; Coding system related functions.
190
191 ;;;###autoload
192 (defun set-coding-system-alist (target-type regexp coding-system
193 &optional operation)
194 "Update `coding-system-alist' according to the arguments.
195 TARGET-TYPE specifies a type of the target: `file', `process', or `network'.
196 TARGET-TYPE tells which slots of coding-system-alist should be affected.
197 If `file', it affects slots for insert-file-contents and write-region.
198 If `process', it affects slots for call-process, call-process-region, and
199 start-process.
200 If `network', it affects a slot for open-network-process.
201 REGEXP is a regular expression matching a target of I/O operation.
202 CODING-SYSTEM is a coding system to perform code conversion
203 on the I/O operation, or a cons of coding systems for decoding and
204 encoding respectively, or a function symbol which returns the cons.
205 Optional arg OPERATION if non-nil specifies directly one of slots above.
206 The valid value is: insert-file-contents, write-region,
207 call-process, call-process-region, start-process, or open-network-stream.
208 If OPERATION is specified, TARGET-TYPE is ignored.
209 See the documentation of `coding-system-alist' for more detail."
210 (or (stringp regexp)
211 (error "Invalid regular expression: %s" regexp))
212 (or (memq target-type '(file process network))
213 (error "Invalid target type: %s" target-type))
214 (if (symbolp coding-system)
215 (if (not (fboundp coding-system))
216 (progn
217 (check-coding-system coding-system)
218 (setq coding-system (cons coding-system coding-system))))
219 (check-coding-system (car coding-system))
220 (check-coding-system (cdr coding-system)))
221 (let ((op-list (if operation (list operation)
222 (cond ((eq target-type 'file)
223 '(insert-file-contents write-region))
224 ((eq target-type 'process)
225 '(call-process call-process-region start-process))
226 (t ; i.e. (eq target-type network)
227 '(open-network-stream)))))
228 slot)
229 (while op-list
230 (setq slot (assq (car op-list) coding-system-alist))
231 (if slot
232 (let ((chain (cdr slot)))
233 (if (catch 'tag
234 (while chain
235 (if (string= regexp (car (car chain)))
236 (progn
237 (setcdr (car chain) coding-system)
238 (throw 'tag nil)))
239 (setq chain (cdr chain)))
240 t)
241 (setcdr slot (cons (cons regexp coding-system) (cdr slot)))))
242 (setq coding-system-alist
243 (cons (cons (car op-list) (list (cons regexp coding-system)))
244 coding-system-alist)))
245 (setq op-list (cdr op-list)))))
246
247
248 ;;; Composite charcater manipulations.
249
250 ;;;###autoload
251 (defun compose-region (start end &optional buffer)
252 "Compose characters in the current region into one composite character.
253 From a Lisp program, pass two arguments, START to END.
254 The composite character replaces the composed characters.
255 BUFFER defaults to the current buffer if omitted."
256 (interactive "r")
257 (let ((ch (make-composite-char (buffer-substring start end buffer))))
258 (delete-region start end buffer)
259 (insert-char ch nil nil buffer)))
260
261 ;;;###autoload
262 (defun decompose-region (start end &optional buffer)
263 "Decompose any composite characters in the current region.
264 From a Lisp program, pass two arguments, START to END.
265 This converts each composite character into one or more characters,
266 the individual characters out of which the composite character was formed.
267 Non-composite characters are left as-is. BUFFER defaults to the current
268 buffer if omitted."
269 (interactive "r")
270 (save-excursion
271 (set-buffer buffer)
272 (save-restriction
273 (narrow-to-region start end)
274 (goto-char (point-min))
275 (let ((compcharset (get-charset 'composite)))
276 (while (< (point) (point-max))
277 (let ((ch (char-after (point))))
278 (if (eq compcharset (char-charset ch))
279 (progn
280 (delete-char 1)
281 (insert (composite-char-string ch))))))))))
282
283 ;;;###autoload
284 (defconst reference-point-alist
285 '((tl . 0) (tc . 1) (tr . 2)
286 (ml . 3) (mc . 4) (mr . 5)
287 (bl . 6) (bc . 7) (br . 8)
288 (top-left . 0) (top-center . 1) (top-right . 2)
289 (mid-left . 3) (mid-center . 4) (mid-right . 5)
290 (bottom-left . 6) (bottom-center . 7) (bottom-right . 8)
291 (0 . 0) (1 . 1) (2 . 2)
292 (3 . 3) (4 . 4) (5 . 5)
293 (6 . 6) (7 . 7) (8 . 8))
294 "Alist of reference point symbols vs reference point codes.
295 Meanings of reference point codes are as follows:
296
297 0----1----2 <-- ascent 0:tl or top-left
298 | | 1:tc or top-center
299 | | 2:tr or top-right
300 | | 3:ml or mid-left
301 | 4 <--+---- center 4:mc or mid-center
302 | | 5:mr or mid-right
303 --- 3 5 <-- baseline 6:bl or bottom-left
304 | | 7:bc or bottom-center
305 6----7----8 <-- descent 8:br or bottom-right
306
307 Reference point symbols are to be used to specify composition rule of
308 the form \(GLOBAL-REF-POINT . NEW-REF-POINT), where GLOBAL-REF-POINT
309 is a reference point in the overall glyphs already composed, and
310 NEW-REF-POINT is a reference point in the new glyph to be added.
311
312 For instance, if GLOBAL-REF-POINT is 8 and NEW-REF-POINT is 1, the
313 overall glyph is updated as follows:
314
315 +-------+--+ <--- new ascent
316 | | |
317 | global| |
318 | glyph | |
319 --- | | | <--- baseline (doesn't change)
320 +----+--+--+
321 | | new |
322 | |glyph|
323 +----+-----+ <--- new descent
324 ")
325
326 ;; Return a string for char CH to be embedded in multibyte form of
327 ;; composite character.
328 (defun compose-chars-component (ch)
329 (if (< ch 128)
330 (format "\240%c" (+ ch 128))
331 (let ((str (char-to-string ch)))
332 (if (cmpcharp ch)
333 (if (/= (aref str 1) ?\xFF)
334 (error "Char %c can't be composed" ch)
335 (substring str 2))
336 (aset str 0 (+ (aref str 0) ?\x20))
337 str))))
338
339 ;; Return a string for composition rule RULE to be embedded in
340 ;; multibyte form of composite character.
341 (defsubst compose-chars-rule (rule)
342 (char-to-string (+ ?\xA0
343 (* (cdr (assq (car rule) reference-point-alist)) 9)
344 (cdr (assq (cdr rule) reference-point-alist)))))
345
346 ;;;###autoload
347 (defun compose-chars (first-component &rest args)
348 "Return one char string composed from the arguments.
349 Each argument is a character (including a composite chararacter)
350 or a composition rule.
351 A composition rule has the form \(GLOBAL-REF-POINT . NEW-REF-POINT).
352 See the documentation of `reference-point-alist' for more detail."
353 (if (= (length args) 0)
354 (char-to-string first-component)
355 (let* ((with-rule (consp (car args)))
356 (str (if with-rule (concat (vector leading-code-composition ?\xFF))
357 (char-to-string leading-code-composition))))
358 (setq str (concat str (compose-chars-component first-component)))
359 (while args
360 (if with-rule
361 (progn
362 (if (not (consp (car args)))
363 (error "Invalid composition rule: %s" (car args)))
364 (setq str (concat str (compose-chars-rule (car args))
365 (compose-chars-component (car (cdr args))))
366 args (cdr (cdr args))))
367 (setq str (concat str (compose-chars-component (car args)))
368 args (cdr args))))
369 str)))
370
371 ;;;###autoload
372 (defun decompose-composite-char (char &optional type with-composition-rule)
373 "Convert composite character CHAR to a string containing components of CHAR.
374 Optional 1st arg TYPE specifies the type of sequence returned.
375 It should be `string' (default), `list', or `vector'.
376 Optional 2nd arg WITH-COMPOSITION-RULE non-nil means the returned
377 sequence contains embedded composition rules if any. In this case, the
378 order of elements in the sequence is the same as arguments for
379 `compose-chars' to create CHAR.
380 If TYPE is omitted or is `string', composition rules are omitted
381 even if WITH-COMPOSITION-RULE is t."
382 (or type
383 (setq type 'string))
384 (let* ((len (composite-char-component-count char))
385 (i (1- len))
386 l)
387 (setq with-composition-rule (and with-composition-rule
388 (not (eq type 'string))
389 (composite-char-composition-rule-p char)))
390 (while (> i 0)
391 (setq l (cons (composite-char-component char i) l))
392 (if with-composition-rule
393 (let ((rule (- (composite-char-composition-rule char i) ?\xA0)))
394 (setq l (cons (cons (/ rule 9) (% rule 9)) l))))
395 (setq i (1- i)))
396 (setq l (cons (composite-char-component char 0) l))
397 (cond ((eq type 'string)
398 (apply 'concat-chars l))
399 ((eq type 'list)
400 l)
401 (t ; i.e. TYPE is vector
402 (vconcat l)))))
403
404 ;;; mule-util.el ends here