Mercurial > emacs
annotate src/character.c @ 88885:c636c5e9d6ce
*** empty log message ***
author | Kenichi Handa <handa@m17n.org> |
---|---|
date | Wed, 24 Jul 2002 10:50:53 +0000 |
parents | a0ae1a5876c7 |
children | 76074dea1258 |
rev | line source |
---|---|
88363 | 1 /* Basic character support. |
2 Copyright (C) 1995, 1997, 1998, 2001 Electrotechnical Laboratory, JAPAN. | |
3 Licensed to the Free Software Foundation. | |
4 Copyright (C) 2001 Free Software Foundation, Inc. | |
5 Copyright (C) 2001, 2002 | |
6 National Institute of Advanced Industrial Science and Technology (AIST) | |
7 Registration Number H13PRO009 | |
8 | |
9 This file is part of GNU Emacs. | |
10 | |
11 GNU Emacs is free software; you can redistribute it and/or modify | |
12 it 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 GNU Emacs is distributed in the hope that it will be useful, | |
17 but WITHOUT ANY WARRANTY; without even the implied warranty of | |
18 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
19 GNU General Public License for more details. | |
20 | |
21 You should have received a copy of the GNU General Public License | |
22 along with GNU Emacs; see the file COPYING. If not, write to | |
23 the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
24 Boston, MA 02111-1307, USA. */ | |
25 | |
26 /* At first, see the document in `character.h' to understand the code | |
27 in this file. */ | |
28 | |
29 #ifdef emacs | |
30 #include <config.h> | |
31 #endif | |
32 | |
33 #include <stdio.h> | |
34 | |
35 #ifdef emacs | |
36 | |
37 #include <sys/types.h> | |
38 #include "lisp.h" | |
39 #include "character.h" | |
40 #include "buffer.h" | |
41 #include "charset.h" | |
42 #include "composite.h" | |
43 #include "disptab.h" | |
44 | |
45 #else /* not emacs */ | |
46 | |
47 #include "mulelib.h" | |
48 | |
49 #endif /* emacs */ | |
50 | |
51 Lisp_Object Qcharacterp; | |
52 | |
53 /* Vector of translation table ever defined. | |
54 ID of a translation table is used to index this vector. */ | |
55 Lisp_Object Vtranslation_table_vector; | |
56 | |
57 /* A char-table for characters which may invoke auto-filling. */ | |
58 Lisp_Object Vauto_fill_chars; | |
59 | |
60 Lisp_Object Qauto_fill_chars; | |
61 | |
62 Lisp_Object Vchar_unify_table; | |
63 | |
64 /* A char-table. An element is non-nil iff the corresponding | |
65 character has a printable glyph. */ | |
66 Lisp_Object Vprintable_chars; | |
67 | |
68 /* A char-table. An elemnent is a column-width of the corresponding | |
69 character. */ | |
70 Lisp_Object Vchar_width_table; | |
71 | |
72 /* A char-table. An element is a symbol indicating the direction | |
73 property of corresponding character. */ | |
74 Lisp_Object Vchar_direction_table; | |
75 | |
76 /* Variables used locally in the macro FETCH_MULTIBYTE_CHAR. */ | |
77 unsigned char *_fetch_multibyte_char_p; | |
78 int _fetch_multibyte_char_len; | |
79 | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
80 /* Alist of scripts vs character ranges. */ |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
81 Lisp_Object Vscript_alist; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
82 |
88363 | 83 |
84 | |
85 int | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
86 char_string_with_unification (c, p) |
88363 | 87 int c; |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
88 unsigned char *p; |
88363 | 89 { |
90 int bytes; | |
91 | |
92 MAYBE_UNIFY_CHAR (c); | |
93 | |
94 if (c <= MAX_3_BYTE_CHAR || c > MAX_5_BYTE_CHAR) | |
95 { | |
96 bytes = CHAR_STRING (c, p); | |
97 } | |
98 else if (c <= MAX_4_BYTE_CHAR) | |
99 { | |
100 p[0] = (0xF0 | (c >> 18)); | |
101 p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
102 p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
103 p[3] = (0x80 | (c & 0x3F)); | |
104 bytes = 4; | |
105 } | |
106 else | |
107 { | |
108 p[0] = 0xF8; | |
109 p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
110 p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
111 p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
112 p[4] = (0x80 | (c & 0x3F)); | |
113 bytes = 5; | |
114 } | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
115 |
88363 | 116 return bytes; |
117 } | |
118 | |
119 | |
120 int | |
121 string_char_with_unification (p, advanced, len) | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
122 const unsigned char *p; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
123 const unsigned char **advanced; |
88363 | 124 int *len; |
125 { | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
126 int c; |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
127 const unsigned char *saved_p = p; |
88363 | 128 |
129 if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
130 { | |
131 c = STRING_CHAR_ADVANCE (p); | |
132 } | |
133 else if (! (*p & 0x08)) | |
134 { | |
135 c = ((((p)[0] & 0xF) << 18) | |
136 | (((p)[1] & 0x3F) << 12) | |
137 | (((p)[2] & 0x3F) << 6) | |
138 | ((p)[3] & 0x3F)); | |
139 p += 4; | |
140 } | |
141 else | |
142 { | |
143 c = ((((p)[1] & 0x3F) << 18) | |
144 | (((p)[2] & 0x3F) << 12) | |
145 | (((p)[3] & 0x3F) << 6) | |
146 | ((p)[4] & 0x3F)); | |
147 p += 5; | |
148 } | |
149 | |
150 MAYBE_UNIFY_CHAR (c); | |
151 | |
152 if (len) | |
153 *len = p - saved_p; | |
154 if (advanced) | |
155 *advanced = p; | |
156 return c; | |
157 } | |
158 | |
159 | |
160 /* Translate character C by translation table TABLE. If C is | |
161 negative, translate a character specified by CHARSET and CODE. If | |
162 no translation is found in TABLE, return the untranslated | |
163 character. */ | |
164 | |
165 int | |
166 translate_char (table, c) | |
167 Lisp_Object table; | |
168 int c; | |
169 { | |
170 Lisp_Object ch; | |
171 | |
172 if (! CHAR_TABLE_P (table)) | |
173 return c; | |
174 ch = CHAR_TABLE_REF (table, c); | |
175 if (! CHARACTERP (ch)) | |
176 return c; | |
177 return XINT (ch); | |
178 } | |
179 | |
180 /* Convert the unibyte character C to the corresponding multibyte | |
181 character based on the current value of charset_primary. If C | |
182 can't be converted, return C. */ | |
183 | |
184 int | |
185 unibyte_char_to_multibyte (c) | |
186 int c; | |
187 { | |
188 struct charset *charset = CHARSET_FROM_ID (charset_primary); | |
189 int c1 = DECODE_CHAR (charset, c); | |
190 | |
191 return ((c1 >= 0) ? c1 : c); | |
192 } | |
193 | |
194 | |
195 /* Convert the multibyte character C to unibyte 8-bit character based | |
196 on the current value of charset_primary. If dimension of | |
197 charset_primary is more than one, return (C & 0xFF). | |
198 | |
199 The argument REV_TBL is now ignored. It will be removed in the | |
200 future. */ | |
201 | |
202 int | |
203 multibyte_char_to_unibyte (c, rev_tbl) | |
204 int c; | |
205 Lisp_Object rev_tbl; | |
206 { | |
207 struct charset *charset = CHARSET_FROM_ID (charset_primary); | |
208 unsigned c1 = ENCODE_CHAR (charset, c); | |
209 | |
210 return ((c1 != CHARSET_INVALID_CODE (charset)) ? c1 : c & 0xFF); | |
211 } | |
212 | |
213 | |
214 DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
215 doc: /* Return non-nil if OBJECT is a character. */) | |
216 (object, ignore) | |
217 Lisp_Object object, ignore; | |
218 { | |
219 return (CHARACTERP (object) ? Qt : Qnil); | |
220 } | |
221 | |
222 DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
223 doc: /* Return the character of the maximum code. */) | |
224 () | |
225 { | |
226 return make_number (MAX_CHAR); | |
227 } | |
228 | |
229 DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
230 Sunibyte_char_to_multibyte, 1, 1, 0, | |
231 doc: /* Convert the unibyte character CH to multibyte character. | |
232 The multibyte character is a result of decoding CH by | |
233 the current primary charset (value of `charset-primary'). */) | |
234 (ch) | |
235 Lisp_Object ch; | |
236 { | |
237 int c; | |
238 struct charset *charset; | |
239 | |
240 CHECK_CHARACTER (ch); | |
241 c = XFASTINT (ch); | |
242 if (c >= 0400) | |
243 error ("Invalid unibyte character: %d", c); | |
244 charset = CHARSET_FROM_ID (charset_primary); | |
245 c = DECODE_CHAR (charset, c); | |
246 if (c < 0) | |
247 error ("Can't convert to multibyte character: %d", XINT (ch)); | |
248 return make_number (c); | |
249 } | |
250 | |
251 DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
252 Smultibyte_char_to_unibyte, 1, 1, 0, | |
253 doc: /* Convert the multibyte character CH to unibyte character.\n\ | |
254 The unibyte character is a result of encoding CH by | |
255 the current primary charset (value of `charset-primary'). */) | |
256 (ch) | |
257 Lisp_Object ch; | |
258 { | |
259 int c; | |
260 unsigned code; | |
261 struct charset *charset; | |
262 | |
263 CHECK_CHARACTER (ch); | |
264 c = XFASTINT (ch); | |
265 charset = CHARSET_FROM_ID (charset_primary); | |
266 code = ENCODE_CHAR (charset, c); | |
267 if (code < CHARSET_MIN_CODE (charset) | |
268 || code > CHARSET_MAX_CODE (charset)) | |
269 error ("Can't convert to unibyte character: %d", XINT (ch)); | |
270 return make_number (code); | |
271 } | |
272 | |
273 DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
274 doc: /* Return 1 regardless of the argument CHAR. | |
275 This is now an obsolete function. We keep it just for backward compatibility. */) | |
276 (ch) | |
277 Lisp_Object ch; | |
278 { | |
279 CHECK_CHARACTER (ch); | |
280 return make_number (1); | |
281 } | |
282 | |
283 DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
284 doc: /* Return width of CHAR when displayed in the current buffer. | |
285 The width is measured by how many columns it occupies on the screen. | |
286 Tab is taken to occupy `tab-width' columns. */) | |
287 (ch) | |
288 Lisp_Object ch; | |
289 { | |
290 Lisp_Object disp; | |
291 int c, width; | |
292 struct Lisp_Char_Table *dp = buffer_display_table (); | |
293 | |
294 CHECK_CHARACTER (ch); | |
295 c = XINT (ch); | |
296 | |
297 /* Get the way the display table would display it. */ | |
298 disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
299 | |
300 if (VECTORP (disp)) | |
301 width = ASIZE (disp); | |
302 else | |
303 width = CHAR_WIDTH (c); | |
304 | |
305 return make_number (width); | |
306 } | |
307 | |
308 /* Return width of string STR of length LEN when displayed in the | |
309 current buffer. The width is measured by how many columns it | |
310 occupies on the screen. If PRECISION > 0, return the width of | |
311 longest substring that doesn't exceed PRECISION, and set number of | |
312 characters and bytes of the substring in *NCHARS and *NBYTES | |
313 respectively. */ | |
314 | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
315 int |
88363 | 316 c_string_width (str, len, precision, nchars, nbytes) |
317 unsigned char *str; | |
318 int precision, *nchars, *nbytes; | |
319 { | |
320 int i = 0, i_byte = 0; | |
321 int width = 0; | |
322 struct Lisp_Char_Table *dp = buffer_display_table (); | |
323 | |
324 while (i_byte < len) | |
325 { | |
326 int bytes, thiswidth; | |
327 Lisp_Object val; | |
328 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
329 | |
330 if (dp) | |
331 { | |
332 val = DISP_CHAR_VECTOR (dp, c); | |
333 if (VECTORP (val)) | |
334 thiswidth = XVECTOR (val)->size; | |
335 else | |
336 thiswidth = CHAR_WIDTH (c); | |
337 } | |
338 else | |
339 { | |
340 thiswidth = CHAR_WIDTH (c); | |
341 } | |
342 | |
343 if (precision > 0 | |
344 && (width + thiswidth > precision)) | |
345 { | |
346 *nchars = i; | |
347 *nbytes = i_byte; | |
348 return width; | |
349 } | |
350 i++; | |
351 i_byte += bytes; | |
352 width += thiswidth; | |
353 } | |
354 | |
355 if (precision > 0) | |
356 { | |
357 *nchars = i; | |
358 *nbytes = i_byte; | |
359 } | |
360 | |
361 return width; | |
362 } | |
363 | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
364 /* Return width of string STR of length LEN when displayed in the |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
365 current buffer. The width is measured by how many columns it |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
366 occupies on the screen. */ |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
367 |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
368 int |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
369 strwidth (str, len) |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
370 unsigned char *str; |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
371 int len; |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
372 { |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
373 return c_string_width (str, len, -1, NULL, NULL); |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
374 } |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
375 |
88363 | 376 /* Return width of Lisp string STRING when displayed in the current |
377 buffer. The width is measured by how many columns it occupies on | |
378 the screen while paying attention to compositions. If PRECISION > | |
379 0, return the width of longest substring that doesn't exceed | |
380 PRECISION, and set number of characters and bytes of the substring | |
381 in *NCHARS and *NBYTES respectively. */ | |
382 | |
383 int | |
384 lisp_string_width (string, precision, nchars, nbytes) | |
385 Lisp_Object string; | |
386 int precision, *nchars, *nbytes; | |
387 { | |
388 int len = XSTRING (string)->size; | |
389 unsigned char *str = XSTRING (string)->data; | |
390 int i = 0, i_byte = 0; | |
391 int width = 0; | |
392 struct Lisp_Char_Table *dp = buffer_display_table (); | |
393 | |
394 while (i < len) | |
395 { | |
396 int chars, bytes, thiswidth; | |
397 Lisp_Object val; | |
398 int cmp_id; | |
399 int ignore, end; | |
400 | |
401 if (find_composition (i, -1, &ignore, &end, &val, string) | |
402 && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
403 >= 0)) | |
404 { | |
405 thiswidth = composition_table[cmp_id]->width; | |
406 chars = end - i; | |
407 bytes = string_char_to_byte (string, end) - i_byte; | |
408 } | |
409 else if (dp) | |
410 { | |
411 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
412 | |
413 chars = 1; | |
414 val = DISP_CHAR_VECTOR (dp, c); | |
415 if (VECTORP (val)) | |
416 thiswidth = XVECTOR (val)->size; | |
417 else | |
418 thiswidth = CHAR_WIDTH (c); | |
419 } | |
420 else | |
421 { | |
422 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
423 | |
424 chars = 1; | |
425 thiswidth = CHAR_WIDTH (c); | |
426 } | |
427 | |
428 if (precision > 0 | |
429 && (width + thiswidth > precision)) | |
430 { | |
431 *nchars = i; | |
432 *nbytes = i_byte; | |
433 return width; | |
434 } | |
435 i += chars; | |
436 i_byte += bytes; | |
437 width += thiswidth; | |
438 } | |
439 | |
440 if (precision > 0) | |
441 { | |
442 *nchars = i; | |
443 *nbytes = i_byte; | |
444 } | |
445 | |
446 return width; | |
447 } | |
448 | |
449 DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
450 doc: /* Return width of STRING when displayed in the current buffer. | |
451 Width is measured by how many columns it occupies on the screen. | |
452 When calculating width of a multibyte character in STRING, | |
453 only the base leading-code is considered; the validity of | |
454 the following bytes is not checked. Tabs in STRING are always | |
455 taken to occupy `tab-width' columns. */) | |
456 (str) | |
457 Lisp_Object str; | |
458 { | |
459 Lisp_Object val; | |
460 | |
461 CHECK_STRING (str); | |
462 XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
463 return val; | |
464 } | |
465 | |
466 DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
467 doc: /* Return the direction of CHAR. | |
468 The returned value is 0 for left-to-right and 1 for right-to-left. */) | |
469 (ch) | |
470 Lisp_Object ch; | |
471 { | |
472 int c; | |
473 | |
474 CHECK_CHARACTER (ch); | |
475 c = XINT (ch); | |
476 return CHAR_TABLE_REF (Vchar_direction_table, c); | |
477 } | |
478 | |
479 DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, | |
480 doc: /* Return number of characters between BEG and END. | |
481 This is now an obsolete function. We keep it just for backward compatibility. */) | |
482 (beg, end) | |
483 Lisp_Object beg, end; | |
484 { | |
485 int from, to; | |
486 | |
487 CHECK_NUMBER_COERCE_MARKER (beg); | |
488 CHECK_NUMBER_COERCE_MARKER (end); | |
489 | |
490 from = min (XFASTINT (beg), XFASTINT (end)); | |
491 to = max (XFASTINT (beg), XFASTINT (end)); | |
492 | |
493 return make_number (to - from); | |
494 } | |
495 | |
496 /* Return the number of characters in the NBYTES bytes at PTR. | |
497 This works by looking at the contents and checking for multibyte | |
498 sequences while assuming that there's no invalid sequence. | |
499 However, if the current buffer has enable-multibyte-characters = | |
500 nil, we treat each byte as a character. */ | |
501 | |
502 int | |
503 chars_in_text (ptr, nbytes) | |
504 unsigned char *ptr; | |
505 int nbytes; | |
506 { | |
507 /* current_buffer is null at early stages of Emacs initialization. */ | |
508 if (current_buffer == 0 | |
509 || NILP (current_buffer->enable_multibyte_characters)) | |
510 return nbytes; | |
511 | |
512 return multibyte_chars_in_text (ptr, nbytes); | |
513 } | |
514 | |
515 /* Return the number of characters in the NBYTES bytes at PTR. | |
516 This works by looking at the contents and checking for multibyte | |
517 sequences while assuming that there's no invalid sequence. It | |
518 ignores enable-multibyte-characters. */ | |
519 | |
520 int | |
521 multibyte_chars_in_text (ptr, nbytes) | |
522 unsigned char *ptr; | |
523 int nbytes; | |
524 { | |
525 unsigned char *endp = ptr + nbytes; | |
526 int chars = 0; | |
527 | |
528 while (ptr < endp) | |
529 { | |
530 int len = MULTIBYTE_LENGTH (ptr, endp); | |
531 | |
532 if (len == 0) | |
533 abort (); | |
534 ptr += len; | |
535 chars++; | |
536 } | |
537 | |
538 return chars; | |
539 } | |
540 | |
541 /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
542 characters and bytes in it, and store them in *NCHARS and *NBYTES | |
543 respectively. On counting bytes, pay attention to that 8-bit | |
544 characters not constructing a valid multibyte sequence are | |
545 represented by 2-byte in a multibyte text. */ | |
546 | |
547 void | |
548 parse_str_as_multibyte (str, len, nchars, nbytes) | |
549 unsigned char *str; | |
550 int len, *nchars, *nbytes; | |
551 { | |
552 unsigned char *endp = str + len; | |
553 int n, chars = 0, bytes = 0; | |
554 | |
555 if (len >= MAX_MULTIBYTE_LENGTH) | |
556 { | |
557 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
558 while (str < adjusted_endp) | |
559 { | |
560 if ((n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
561 str += n, bytes += n; | |
562 else | |
563 str++, bytes += 2; | |
564 chars++; | |
565 } | |
566 } | |
567 while (str < endp) | |
568 { | |
569 if ((n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
570 str += n, bytes += n; | |
571 else | |
572 str++, bytes += 2; | |
573 chars++; | |
574 } | |
575 | |
576 *nchars = chars; | |
577 *nbytes = bytes; | |
578 return; | |
579 } | |
580 | |
581 /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
582 It actually converts only such 8-bit characters that don't contruct | |
583 a multibyte sequence to multibyte forms of Latin-1 characters. If | |
584 NCHARS is nonzero, set *NCHARS to the number of characters in the | |
585 text. It is assured that we can use LEN bytes at STR as a work | |
586 area and that is enough. Return the number of bytes of the | |
587 resulting text. */ | |
588 | |
589 int | |
590 str_as_multibyte (str, len, nbytes, nchars) | |
591 unsigned char *str; | |
592 int len, nbytes, *nchars; | |
593 { | |
594 unsigned char *p = str, *endp = str + nbytes; | |
595 unsigned char *to; | |
596 int chars = 0; | |
597 int n; | |
598 | |
599 if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
600 { | |
601 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
602 while (p < adjusted_endp | |
603 && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
604 p += n, chars++; | |
605 } | |
606 while ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
607 p += n, chars++; | |
608 if (nchars) | |
609 *nchars = chars; | |
610 if (p == endp) | |
611 return nbytes; | |
612 | |
613 to = p; | |
614 nbytes = endp - p; | |
615 endp = str + len; | |
616 safe_bcopy ((char *) p, (char *) (endp - nbytes), nbytes); | |
617 p = endp - nbytes; | |
618 | |
619 if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
620 { | |
621 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
622 while (p < adjusted_endp) | |
623 { | |
624 if ((n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
625 { | |
626 while (n--) | |
627 *to++ = *p++; | |
628 } | |
629 else | |
630 { | |
631 int c = *p++; | |
632 c = BYTE8_TO_CHAR (c); | |
633 to += CHAR_STRING (c, to); | |
634 } | |
635 } | |
636 chars++; | |
637 } | |
638 while (p < endp) | |
639 { | |
640 if ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
641 { | |
642 while (n--) | |
643 *to++ = *p++; | |
644 } | |
645 else | |
646 { | |
647 int c = *p++; | |
648 c = BYTE8_TO_CHAR (c); | |
649 to += CHAR_STRING (c, to); | |
650 } | |
651 chars++; | |
652 } | |
653 if (nchars) | |
654 *nchars = chars; | |
655 return (to - str); | |
656 } | |
657 | |
658 /* Parse unibyte string at STR of LEN bytes, and return the number of | |
659 bytes it may ocupy when converted to multibyte string by | |
660 `str_to_multibyte'. */ | |
661 | |
662 int | |
663 parse_str_to_multibyte (str, len) | |
664 unsigned char *str; | |
665 int len; | |
666 { | |
667 unsigned char *endp = str + len; | |
668 int bytes; | |
669 | |
670 for (bytes = 0; str < endp; str++) | |
671 bytes += (*str < 0x80) ? 1 : 2; | |
672 return bytes; | |
673 } | |
674 | |
675 | |
676 /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
677 that contains the same single-byte characters. It actually | |
678 converts all 8-bit characters to multibyte forms. It is assured | |
679 that we can use LEN bytes at STR as a work area and that is | |
680 enough. */ | |
681 | |
682 int | |
683 str_to_multibyte (str, len, bytes) | |
684 unsigned char *str; | |
685 int len, bytes; | |
686 { | |
687 unsigned char *p = str, *endp = str + bytes; | |
688 unsigned char *to; | |
689 | |
690 while (p < endp && *p < 0x80) p++; | |
691 if (p == endp) | |
692 return bytes; | |
693 to = p; | |
694 bytes = endp - p; | |
695 endp = str + len; | |
696 safe_bcopy ((char *) p, (char *) (endp - bytes), bytes); | |
697 p = endp - bytes; | |
698 while (p < endp) | |
699 { | |
700 int c = *p++; | |
701 | |
702 if (c >= 0x80) | |
703 c = BYTE8_TO_CHAR (c); | |
704 to += CHAR_STRING (c, to); | |
705 } | |
706 return (to - str); | |
707 } | |
708 | |
709 /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
710 actually converts characters in the range 0x80..0xFF to | |
711 unibyte. */ | |
712 | |
713 int | |
714 str_as_unibyte (str, bytes) | |
715 unsigned char *str; | |
716 int bytes; | |
717 { | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
718 const unsigned char *p = str, *endp = str + bytes; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
719 unsigned char *to; |
88363 | 720 int c, len; |
721 | |
722 while (p < endp) | |
723 { | |
724 c = *p; | |
725 len = BYTES_BY_CHAR_HEAD (c); | |
726 if (CHAR_BYTE8_HEAD_P (c)) | |
727 break; | |
728 p += len; | |
729 } | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
730 to = str + (p - str); |
88363 | 731 while (p < endp) |
732 { | |
733 c = *p; | |
734 len = BYTES_BY_CHAR_HEAD (c); | |
735 if (CHAR_BYTE8_HEAD_P (c)) | |
736 { | |
737 c = STRING_CHAR_ADVANCE (p); | |
738 *to++ = CHAR_TO_BYTE8 (c); | |
739 } | |
740 else | |
741 { | |
742 while (len--) *to++ = *p++; | |
743 } | |
744 } | |
745 return (to - str); | |
746 } | |
747 | |
748 int | |
749 string_count_byte8 (string) | |
750 Lisp_Object string; | |
751 { | |
752 int multibyte = STRING_MULTIBYTE (string); | |
753 int nbytes = STRING_BYTES (XSTRING (string)); | |
754 unsigned char *p = XSTRING (string)->data; | |
755 unsigned char *pend = p + nbytes; | |
756 int count = 0; | |
757 int c, len; | |
758 | |
759 if (multibyte) | |
760 while (p < pend) | |
761 { | |
762 c = *p; | |
763 len = BYTES_BY_CHAR_HEAD (c); | |
764 | |
765 if (CHAR_BYTE8_HEAD_P (c)) | |
766 count++; | |
767 p += len; | |
768 } | |
769 else | |
770 while (p < pend) | |
771 { | |
772 if (*p++ >= 0x80) | |
773 count++; | |
774 } | |
775 return count; | |
776 } | |
777 | |
778 | |
779 Lisp_Object | |
780 string_escape_byte8 (string) | |
781 Lisp_Object string; | |
782 { | |
783 int nchars = XSTRING (string)->size; | |
784 int nbytes = STRING_BYTES (XSTRING (string)); | |
785 int multibyte = STRING_MULTIBYTE (string); | |
786 int byte8_count; | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
787 const unsigned char *src, *src_end; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
788 unsigned char *dst; |
88363 | 789 Lisp_Object val; |
790 int c, len; | |
791 | |
792 if (multibyte && nchars == nbytes) | |
793 return string; | |
794 | |
795 byte8_count = string_count_byte8 (string); | |
796 | |
797 if (byte8_count == 0) | |
798 return string; | |
799 | |
800 if (multibyte) | |
801 /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
88526
6bac4ae1c326
(string_escape_byte8): Make multibyte string with correct size.
Kenichi Handa <handa@m17n.org>
parents:
88427
diff
changeset
|
802 val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
88363 | 803 nbytes + byte8_count * 2); |
804 else | |
805 /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
806 val = make_uninit_string (nbytes + byte8_count * 3); | |
807 | |
808 src = XSTRING (string)->data; | |
809 src_end = src + nbytes; | |
810 dst = XSTRING (val)->data; | |
811 if (multibyte) | |
812 while (src < src_end) | |
813 { | |
814 c = *src; | |
815 len = BYTES_BY_CHAR_HEAD (c); | |
816 | |
817 if (CHAR_BYTE8_HEAD_P (c)) | |
818 { | |
819 c = STRING_CHAR_ADVANCE (src); | |
820 c = CHAR_TO_BYTE8 (c); | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
821 sprintf ((char *) dst, "\\%03o", c); |
88363 | 822 dst += 4; |
823 } | |
824 else | |
825 while (len--) *dst++ = *src++; | |
826 } | |
827 else | |
828 while (src < src_end) | |
829 { | |
830 c = *src++; | |
831 if (c >= 0x80) | |
832 { | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
833 sprintf ((char *) dst, "\\%03o", c); |
88363 | 834 dst += 4; |
835 } | |
836 else | |
837 *dst++ = c; | |
838 } | |
839 return val; | |
840 } | |
841 | |
842 | |
843 DEFUN ("string", Fstring, Sstring, 1, MANY, 0, | |
844 doc: /* | |
88640 | 845 Concatenate all the argument characters and make the result a string. |
846 usage: (string &rest CHARACTERS) */) | |
88363 | 847 (n, args) |
848 int n; | |
849 Lisp_Object *args; | |
850 { | |
851 int i; | |
852 unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); | |
853 unsigned char *p = buf; | |
854 int c; | |
855 | |
856 for (i = 0; i < n; i++) | |
857 { | |
858 CHECK_CHARACTER (args[i]); | |
859 c = XINT (args[i]); | |
860 p += CHAR_STRING (c, p); | |
861 } | |
862 | |
863 return make_string_from_bytes ((char *) buf, n, p - buf); | |
864 } | |
865 | |
866 void | |
867 init_character_once () | |
868 { | |
869 } | |
870 | |
871 #ifdef emacs | |
872 | |
873 void | |
874 syms_of_character () | |
875 { | |
876 DEFSYM (Qcharacterp, "characterp"); | |
877 DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
878 | |
879 staticpro (&Vchar_unify_table); | |
880 Vchar_unify_table = Qnil; | |
881 | |
882 defsubr (&Smax_char); | |
883 defsubr (&Scharacterp); | |
884 defsubr (&Sunibyte_char_to_multibyte); | |
885 defsubr (&Smultibyte_char_to_unibyte); | |
886 defsubr (&Schar_bytes); | |
887 defsubr (&Schar_width); | |
888 defsubr (&Sstring_width); | |
889 defsubr (&Schar_direction); | |
890 defsubr (&Schars_in_region); | |
891 defsubr (&Sstring); | |
892 | |
893 DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
894 doc: /* | |
895 Vector of cons cell of a symbol and translation table ever defined. | |
896 An ID of a translation table is an index of this vector. */); | |
897 Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); | |
898 | |
899 DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
900 doc: /* | |
901 A char-table for characters which invoke auto-filling. | |
902 Such characters have value t in this table. */); | |
903 Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
88853
503c60238368
(syms_of_character): Fix CHAR_TABLE_SET call.
Dave Love <fx@gnu.org>
parents:
88743
diff
changeset
|
904 CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
503c60238368
(syms_of_character): Fix CHAR_TABLE_SET call.
Dave Love <fx@gnu.org>
parents:
88743
diff
changeset
|
905 CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); |
88363 | 906 |
907 DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
908 doc: /* | |
909 A char-table for width (columns) of each character. */); | |
910 Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
88743
4570f1a4c391
(syms_of_character): Setup Vchar_width_table for
Kenichi Handa <handa@m17n.org>
parents:
88716
diff
changeset
|
911 char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
4570f1a4c391
(syms_of_character): Setup Vchar_width_table for
Kenichi Handa <handa@m17n.org>
parents:
88716
diff
changeset
|
912 char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, |
4570f1a4c391
(syms_of_character): Setup Vchar_width_table for
Kenichi Handa <handa@m17n.org>
parents:
88716
diff
changeset
|
913 make_number (4)); |
88363 | 914 |
915 DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
916 doc: /* A char-table for direction of each character. */); | |
917 Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
918 | |
919 DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
920 doc: /* A char-table for each printable character. */); | |
88716
f617b1bf301b
(syms_of_character): Set the default value of
Kenichi Handa <handa@m17n.org>
parents:
88640
diff
changeset
|
921 Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
922 |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
923 DEFVAR_LISP ("script-alist", &Vscript_alist, |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
924 doc: /* Alist of scripts vs the corresponding character ranges. |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
925 Each element has this form: |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
926 ( SCRIPT (FROM-1 . TO-1) (FROM-2 . TO-2) ...) |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
927 SCRIPT is a symbol representing a script name. |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
928 FROM-n and TO-n specifies ranges of characters that belongs to SCRIPT. */); |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
929 Vscript_alist = Qnil; |
88363 | 930 } |
931 | |
932 #endif /* emacs */ |