Mercurial > emacs
annotate src/character.c @ 88728:e18cd612d3c4
(vietnamese-tcvn, tcvn): New coding
system.
("Vietnamese"): Doc fix.
author | Dave Love <fx@gnu.org> |
---|---|
date | Fri, 07 Jun 2002 11:26:43 +0000 |
parents | f617b1bf301b |
children | 4570f1a4c391 |
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 | |
80 | |
81 | |
82 int | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
83 char_string_with_unification (c, p) |
88363 | 84 int c; |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
85 unsigned char *p; |
88363 | 86 { |
87 int bytes; | |
88 | |
89 MAYBE_UNIFY_CHAR (c); | |
90 | |
91 if (c <= MAX_3_BYTE_CHAR || c > MAX_5_BYTE_CHAR) | |
92 { | |
93 bytes = CHAR_STRING (c, p); | |
94 } | |
95 else if (c <= MAX_4_BYTE_CHAR) | |
96 { | |
97 p[0] = (0xF0 | (c >> 18)); | |
98 p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
99 p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
100 p[3] = (0x80 | (c & 0x3F)); | |
101 bytes = 4; | |
102 } | |
103 else | |
104 { | |
105 p[0] = 0xF8; | |
106 p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
107 p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
108 p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
109 p[4] = (0x80 | (c & 0x3F)); | |
110 bytes = 5; | |
111 } | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
112 |
88363 | 113 return bytes; |
114 } | |
115 | |
116 | |
117 int | |
118 string_char_with_unification (p, advanced, len) | |
119 unsigned char *p, **advanced; | |
120 int *len; | |
121 { | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
122 int c; |
88363 | 123 unsigned char *saved_p = p; |
124 | |
125 if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
126 { | |
127 c = STRING_CHAR_ADVANCE (p); | |
128 } | |
129 else if (! (*p & 0x08)) | |
130 { | |
131 c = ((((p)[0] & 0xF) << 18) | |
132 | (((p)[1] & 0x3F) << 12) | |
133 | (((p)[2] & 0x3F) << 6) | |
134 | ((p)[3] & 0x3F)); | |
135 p += 4; | |
136 } | |
137 else | |
138 { | |
139 c = ((((p)[1] & 0x3F) << 18) | |
140 | (((p)[2] & 0x3F) << 12) | |
141 | (((p)[3] & 0x3F) << 6) | |
142 | ((p)[4] & 0x3F)); | |
143 p += 5; | |
144 } | |
145 | |
146 MAYBE_UNIFY_CHAR (c); | |
147 | |
148 if (len) | |
149 *len = p - saved_p; | |
150 if (advanced) | |
151 *advanced = p; | |
152 return c; | |
153 } | |
154 | |
155 | |
156 /* Translate character C by translation table TABLE. If C is | |
157 negative, translate a character specified by CHARSET and CODE. If | |
158 no translation is found in TABLE, return the untranslated | |
159 character. */ | |
160 | |
161 int | |
162 translate_char (table, c) | |
163 Lisp_Object table; | |
164 int c; | |
165 { | |
166 Lisp_Object ch; | |
167 | |
168 if (! CHAR_TABLE_P (table)) | |
169 return c; | |
170 ch = CHAR_TABLE_REF (table, c); | |
171 if (! CHARACTERP (ch)) | |
172 return c; | |
173 return XINT (ch); | |
174 } | |
175 | |
176 /* Convert the unibyte character C to the corresponding multibyte | |
177 character based on the current value of charset_primary. If C | |
178 can't be converted, return C. */ | |
179 | |
180 int | |
181 unibyte_char_to_multibyte (c) | |
182 int c; | |
183 { | |
184 struct charset *charset = CHARSET_FROM_ID (charset_primary); | |
185 int c1 = DECODE_CHAR (charset, c); | |
186 | |
187 return ((c1 >= 0) ? c1 : c); | |
188 } | |
189 | |
190 | |
191 /* Convert the multibyte character C to unibyte 8-bit character based | |
192 on the current value of charset_primary. If dimension of | |
193 charset_primary is more than one, return (C & 0xFF). | |
194 | |
195 The argument REV_TBL is now ignored. It will be removed in the | |
196 future. */ | |
197 | |
198 int | |
199 multibyte_char_to_unibyte (c, rev_tbl) | |
200 int c; | |
201 Lisp_Object rev_tbl; | |
202 { | |
203 struct charset *charset = CHARSET_FROM_ID (charset_primary); | |
204 unsigned c1 = ENCODE_CHAR (charset, c); | |
205 | |
206 return ((c1 != CHARSET_INVALID_CODE (charset)) ? c1 : c & 0xFF); | |
207 } | |
208 | |
209 | |
210 DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
211 doc: /* Return non-nil if OBJECT is a character. */) | |
212 (object, ignore) | |
213 Lisp_Object object, ignore; | |
214 { | |
215 return (CHARACTERP (object) ? Qt : Qnil); | |
216 } | |
217 | |
218 DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
219 doc: /* Return the character of the maximum code. */) | |
220 () | |
221 { | |
222 return make_number (MAX_CHAR); | |
223 } | |
224 | |
225 DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
226 Sunibyte_char_to_multibyte, 1, 1, 0, | |
227 doc: /* Convert the unibyte character CH to multibyte character. | |
228 The multibyte character is a result of decoding CH by | |
229 the current primary charset (value of `charset-primary'). */) | |
230 (ch) | |
231 Lisp_Object ch; | |
232 { | |
233 int c; | |
234 struct charset *charset; | |
235 | |
236 CHECK_CHARACTER (ch); | |
237 c = XFASTINT (ch); | |
238 if (c >= 0400) | |
239 error ("Invalid unibyte character: %d", c); | |
240 charset = CHARSET_FROM_ID (charset_primary); | |
241 c = DECODE_CHAR (charset, c); | |
242 if (c < 0) | |
243 error ("Can't convert to multibyte character: %d", XINT (ch)); | |
244 return make_number (c); | |
245 } | |
246 | |
247 DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
248 Smultibyte_char_to_unibyte, 1, 1, 0, | |
249 doc: /* Convert the multibyte character CH to unibyte character.\n\ | |
250 The unibyte character is a result of encoding CH by | |
251 the current primary charset (value of `charset-primary'). */) | |
252 (ch) | |
253 Lisp_Object ch; | |
254 { | |
255 int c; | |
256 unsigned code; | |
257 struct charset *charset; | |
258 | |
259 CHECK_CHARACTER (ch); | |
260 c = XFASTINT (ch); | |
261 charset = CHARSET_FROM_ID (charset_primary); | |
262 code = ENCODE_CHAR (charset, c); | |
263 if (code < CHARSET_MIN_CODE (charset) | |
264 || code > CHARSET_MAX_CODE (charset)) | |
265 error ("Can't convert to unibyte character: %d", XINT (ch)); | |
266 return make_number (code); | |
267 } | |
268 | |
269 DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
270 doc: /* Return 1 regardless of the argument CHAR. | |
271 This is now an obsolete function. We keep it just for backward compatibility. */) | |
272 (ch) | |
273 Lisp_Object ch; | |
274 { | |
275 CHECK_CHARACTER (ch); | |
276 return make_number (1); | |
277 } | |
278 | |
279 DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
280 doc: /* Return width of CHAR when displayed in the current buffer. | |
281 The width is measured by how many columns it occupies on the screen. | |
282 Tab is taken to occupy `tab-width' columns. */) | |
283 (ch) | |
284 Lisp_Object ch; | |
285 { | |
286 Lisp_Object disp; | |
287 int c, width; | |
288 struct Lisp_Char_Table *dp = buffer_display_table (); | |
289 | |
290 CHECK_CHARACTER (ch); | |
291 c = XINT (ch); | |
292 | |
293 /* Get the way the display table would display it. */ | |
294 disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
295 | |
296 if (VECTORP (disp)) | |
297 width = ASIZE (disp); | |
298 else | |
299 width = CHAR_WIDTH (c); | |
300 | |
301 return make_number (width); | |
302 } | |
303 | |
304 /* Return width of string STR of length LEN when displayed in the | |
305 current buffer. The width is measured by how many columns it | |
306 occupies on the screen. If PRECISION > 0, return the width of | |
307 longest substring that doesn't exceed PRECISION, and set number of | |
308 characters and bytes of the substring in *NCHARS and *NBYTES | |
309 respectively. */ | |
310 | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
311 int |
88363 | 312 c_string_width (str, len, precision, nchars, nbytes) |
313 unsigned char *str; | |
314 int precision, *nchars, *nbytes; | |
315 { | |
316 int i = 0, i_byte = 0; | |
317 int width = 0; | |
318 struct Lisp_Char_Table *dp = buffer_display_table (); | |
319 | |
320 while (i_byte < len) | |
321 { | |
322 int bytes, thiswidth; | |
323 Lisp_Object val; | |
324 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
325 | |
326 if (dp) | |
327 { | |
328 val = DISP_CHAR_VECTOR (dp, c); | |
329 if (VECTORP (val)) | |
330 thiswidth = XVECTOR (val)->size; | |
331 else | |
332 thiswidth = CHAR_WIDTH (c); | |
333 } | |
334 else | |
335 { | |
336 thiswidth = CHAR_WIDTH (c); | |
337 } | |
338 | |
339 if (precision > 0 | |
340 && (width + thiswidth > precision)) | |
341 { | |
342 *nchars = i; | |
343 *nbytes = i_byte; | |
344 return width; | |
345 } | |
346 i++; | |
347 i_byte += bytes; | |
348 width += thiswidth; | |
349 } | |
350 | |
351 if (precision > 0) | |
352 { | |
353 *nchars = i; | |
354 *nbytes = i_byte; | |
355 } | |
356 | |
357 return width; | |
358 } | |
359 | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
360 /* 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
|
361 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
|
362 occupies on the screen. */ |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
363 |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
364 int |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
365 strwidth (str, len) |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
366 unsigned char *str; |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
367 int len; |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
368 { |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
369 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
|
370 } |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
371 |
88363 | 372 /* Return width of Lisp string STRING when displayed in the current |
373 buffer. The width is measured by how many columns it occupies on | |
374 the screen while paying attention to compositions. If PRECISION > | |
375 0, return the width of longest substring that doesn't exceed | |
376 PRECISION, and set number of characters and bytes of the substring | |
377 in *NCHARS and *NBYTES respectively. */ | |
378 | |
379 int | |
380 lisp_string_width (string, precision, nchars, nbytes) | |
381 Lisp_Object string; | |
382 int precision, *nchars, *nbytes; | |
383 { | |
384 int len = XSTRING (string)->size; | |
385 unsigned char *str = XSTRING (string)->data; | |
386 int i = 0, i_byte = 0; | |
387 int width = 0; | |
388 struct Lisp_Char_Table *dp = buffer_display_table (); | |
389 | |
390 while (i < len) | |
391 { | |
392 int chars, bytes, thiswidth; | |
393 Lisp_Object val; | |
394 int cmp_id; | |
395 int ignore, end; | |
396 | |
397 if (find_composition (i, -1, &ignore, &end, &val, string) | |
398 && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
399 >= 0)) | |
400 { | |
401 thiswidth = composition_table[cmp_id]->width; | |
402 chars = end - i; | |
403 bytes = string_char_to_byte (string, end) - i_byte; | |
404 } | |
405 else if (dp) | |
406 { | |
407 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
408 | |
409 chars = 1; | |
410 val = DISP_CHAR_VECTOR (dp, c); | |
411 if (VECTORP (val)) | |
412 thiswidth = XVECTOR (val)->size; | |
413 else | |
414 thiswidth = CHAR_WIDTH (c); | |
415 } | |
416 else | |
417 { | |
418 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
419 | |
420 chars = 1; | |
421 thiswidth = CHAR_WIDTH (c); | |
422 } | |
423 | |
424 if (precision > 0 | |
425 && (width + thiswidth > precision)) | |
426 { | |
427 *nchars = i; | |
428 *nbytes = i_byte; | |
429 return width; | |
430 } | |
431 i += chars; | |
432 i_byte += bytes; | |
433 width += thiswidth; | |
434 } | |
435 | |
436 if (precision > 0) | |
437 { | |
438 *nchars = i; | |
439 *nbytes = i_byte; | |
440 } | |
441 | |
442 return width; | |
443 } | |
444 | |
445 DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
446 doc: /* Return width of STRING when displayed in the current buffer. | |
447 Width is measured by how many columns it occupies on the screen. | |
448 When calculating width of a multibyte character in STRING, | |
449 only the base leading-code is considered; the validity of | |
450 the following bytes is not checked. Tabs in STRING are always | |
451 taken to occupy `tab-width' columns. */) | |
452 (str) | |
453 Lisp_Object str; | |
454 { | |
455 Lisp_Object val; | |
456 | |
457 CHECK_STRING (str); | |
458 XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
459 return val; | |
460 } | |
461 | |
462 DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
463 doc: /* Return the direction of CHAR. | |
464 The returned value is 0 for left-to-right and 1 for right-to-left. */) | |
465 (ch) | |
466 Lisp_Object ch; | |
467 { | |
468 int c; | |
469 | |
470 CHECK_CHARACTER (ch); | |
471 c = XINT (ch); | |
472 return CHAR_TABLE_REF (Vchar_direction_table, c); | |
473 } | |
474 | |
475 DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, | |
476 doc: /* Return number of characters between BEG and END. | |
477 This is now an obsolete function. We keep it just for backward compatibility. */) | |
478 (beg, end) | |
479 Lisp_Object beg, end; | |
480 { | |
481 int from, to; | |
482 | |
483 CHECK_NUMBER_COERCE_MARKER (beg); | |
484 CHECK_NUMBER_COERCE_MARKER (end); | |
485 | |
486 from = min (XFASTINT (beg), XFASTINT (end)); | |
487 to = max (XFASTINT (beg), XFASTINT (end)); | |
488 | |
489 return make_number (to - from); | |
490 } | |
491 | |
492 /* Return the number of characters in the NBYTES bytes at PTR. | |
493 This works by looking at the contents and checking for multibyte | |
494 sequences while assuming that there's no invalid sequence. | |
495 However, if the current buffer has enable-multibyte-characters = | |
496 nil, we treat each byte as a character. */ | |
497 | |
498 int | |
499 chars_in_text (ptr, nbytes) | |
500 unsigned char *ptr; | |
501 int nbytes; | |
502 { | |
503 /* current_buffer is null at early stages of Emacs initialization. */ | |
504 if (current_buffer == 0 | |
505 || NILP (current_buffer->enable_multibyte_characters)) | |
506 return nbytes; | |
507 | |
508 return multibyte_chars_in_text (ptr, nbytes); | |
509 } | |
510 | |
511 /* Return the number of characters in the NBYTES bytes at PTR. | |
512 This works by looking at the contents and checking for multibyte | |
513 sequences while assuming that there's no invalid sequence. It | |
514 ignores enable-multibyte-characters. */ | |
515 | |
516 int | |
517 multibyte_chars_in_text (ptr, nbytes) | |
518 unsigned char *ptr; | |
519 int nbytes; | |
520 { | |
521 unsigned char *endp = ptr + nbytes; | |
522 int chars = 0; | |
523 | |
524 while (ptr < endp) | |
525 { | |
526 int len = MULTIBYTE_LENGTH (ptr, endp); | |
527 | |
528 if (len == 0) | |
529 abort (); | |
530 ptr += len; | |
531 chars++; | |
532 } | |
533 | |
534 return chars; | |
535 } | |
536 | |
537 /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
538 characters and bytes in it, and store them in *NCHARS and *NBYTES | |
539 respectively. On counting bytes, pay attention to that 8-bit | |
540 characters not constructing a valid multibyte sequence are | |
541 represented by 2-byte in a multibyte text. */ | |
542 | |
543 void | |
544 parse_str_as_multibyte (str, len, nchars, nbytes) | |
545 unsigned char *str; | |
546 int len, *nchars, *nbytes; | |
547 { | |
548 unsigned char *endp = str + len; | |
549 int n, chars = 0, bytes = 0; | |
550 | |
551 if (len >= MAX_MULTIBYTE_LENGTH) | |
552 { | |
553 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
554 while (str < adjusted_endp) | |
555 { | |
556 if ((n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
557 str += n, bytes += n; | |
558 else | |
559 str++, bytes += 2; | |
560 chars++; | |
561 } | |
562 } | |
563 while (str < endp) | |
564 { | |
565 if ((n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
566 str += n, bytes += n; | |
567 else | |
568 str++, bytes += 2; | |
569 chars++; | |
570 } | |
571 | |
572 *nchars = chars; | |
573 *nbytes = bytes; | |
574 return; | |
575 } | |
576 | |
577 /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
578 It actually converts only such 8-bit characters that don't contruct | |
579 a multibyte sequence to multibyte forms of Latin-1 characters. If | |
580 NCHARS is nonzero, set *NCHARS to the number of characters in the | |
581 text. It is assured that we can use LEN bytes at STR as a work | |
582 area and that is enough. Return the number of bytes of the | |
583 resulting text. */ | |
584 | |
585 int | |
586 str_as_multibyte (str, len, nbytes, nchars) | |
587 unsigned char *str; | |
588 int len, nbytes, *nchars; | |
589 { | |
590 unsigned char *p = str, *endp = str + nbytes; | |
591 unsigned char *to; | |
592 int chars = 0; | |
593 int n; | |
594 | |
595 if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
596 { | |
597 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
598 while (p < adjusted_endp | |
599 && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
600 p += n, chars++; | |
601 } | |
602 while ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
603 p += n, chars++; | |
604 if (nchars) | |
605 *nchars = chars; | |
606 if (p == endp) | |
607 return nbytes; | |
608 | |
609 to = p; | |
610 nbytes = endp - p; | |
611 endp = str + len; | |
612 safe_bcopy ((char *) p, (char *) (endp - nbytes), nbytes); | |
613 p = endp - nbytes; | |
614 | |
615 if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
616 { | |
617 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
618 while (p < adjusted_endp) | |
619 { | |
620 if ((n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
621 { | |
622 while (n--) | |
623 *to++ = *p++; | |
624 } | |
625 else | |
626 { | |
627 int c = *p++; | |
628 c = BYTE8_TO_CHAR (c); | |
629 to += CHAR_STRING (c, to); | |
630 } | |
631 } | |
632 chars++; | |
633 } | |
634 while (p < endp) | |
635 { | |
636 if ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
637 { | |
638 while (n--) | |
639 *to++ = *p++; | |
640 } | |
641 else | |
642 { | |
643 int c = *p++; | |
644 c = BYTE8_TO_CHAR (c); | |
645 to += CHAR_STRING (c, to); | |
646 } | |
647 chars++; | |
648 } | |
649 if (nchars) | |
650 *nchars = chars; | |
651 return (to - str); | |
652 } | |
653 | |
654 /* Parse unibyte string at STR of LEN bytes, and return the number of | |
655 bytes it may ocupy when converted to multibyte string by | |
656 `str_to_multibyte'. */ | |
657 | |
658 int | |
659 parse_str_to_multibyte (str, len) | |
660 unsigned char *str; | |
661 int len; | |
662 { | |
663 unsigned char *endp = str + len; | |
664 int bytes; | |
665 | |
666 for (bytes = 0; str < endp; str++) | |
667 bytes += (*str < 0x80) ? 1 : 2; | |
668 return bytes; | |
669 } | |
670 | |
671 | |
672 /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
673 that contains the same single-byte characters. It actually | |
674 converts all 8-bit characters to multibyte forms. It is assured | |
675 that we can use LEN bytes at STR as a work area and that is | |
676 enough. */ | |
677 | |
678 int | |
679 str_to_multibyte (str, len, bytes) | |
680 unsigned char *str; | |
681 int len, bytes; | |
682 { | |
683 unsigned char *p = str, *endp = str + bytes; | |
684 unsigned char *to; | |
685 | |
686 while (p < endp && *p < 0x80) p++; | |
687 if (p == endp) | |
688 return bytes; | |
689 to = p; | |
690 bytes = endp - p; | |
691 endp = str + len; | |
692 safe_bcopy ((char *) p, (char *) (endp - bytes), bytes); | |
693 p = endp - bytes; | |
694 while (p < endp) | |
695 { | |
696 int c = *p++; | |
697 | |
698 if (c >= 0x80) | |
699 c = BYTE8_TO_CHAR (c); | |
700 to += CHAR_STRING (c, to); | |
701 } | |
702 return (to - str); | |
703 } | |
704 | |
705 /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
706 actually converts characters in the range 0x80..0xFF to | |
707 unibyte. */ | |
708 | |
709 int | |
710 str_as_unibyte (str, bytes) | |
711 unsigned char *str; | |
712 int bytes; | |
713 { | |
714 unsigned char *p = str, *endp = str + bytes; | |
715 unsigned char *to = str; | |
716 int c, len; | |
717 | |
718 while (p < endp) | |
719 { | |
720 c = *p; | |
721 len = BYTES_BY_CHAR_HEAD (c); | |
722 if (CHAR_BYTE8_HEAD_P (c)) | |
723 break; | |
724 p += len; | |
725 } | |
726 to = p; | |
727 while (p < endp) | |
728 { | |
729 c = *p; | |
730 len = BYTES_BY_CHAR_HEAD (c); | |
731 if (CHAR_BYTE8_HEAD_P (c)) | |
732 { | |
733 c = STRING_CHAR_ADVANCE (p); | |
734 *to++ = CHAR_TO_BYTE8 (c); | |
735 } | |
736 else | |
737 { | |
738 while (len--) *to++ = *p++; | |
739 } | |
740 } | |
741 return (to - str); | |
742 } | |
743 | |
744 int | |
745 string_count_byte8 (string) | |
746 Lisp_Object string; | |
747 { | |
748 int multibyte = STRING_MULTIBYTE (string); | |
749 int nbytes = STRING_BYTES (XSTRING (string)); | |
750 unsigned char *p = XSTRING (string)->data; | |
751 unsigned char *pend = p + nbytes; | |
752 int count = 0; | |
753 int c, len; | |
754 | |
755 if (multibyte) | |
756 while (p < pend) | |
757 { | |
758 c = *p; | |
759 len = BYTES_BY_CHAR_HEAD (c); | |
760 | |
761 if (CHAR_BYTE8_HEAD_P (c)) | |
762 count++; | |
763 p += len; | |
764 } | |
765 else | |
766 while (p < pend) | |
767 { | |
768 if (*p++ >= 0x80) | |
769 count++; | |
770 } | |
771 return count; | |
772 } | |
773 | |
774 | |
775 Lisp_Object | |
776 string_escape_byte8 (string) | |
777 Lisp_Object string; | |
778 { | |
779 int nchars = XSTRING (string)->size; | |
780 int nbytes = STRING_BYTES (XSTRING (string)); | |
781 int multibyte = STRING_MULTIBYTE (string); | |
782 int byte8_count; | |
783 unsigned char *src, *src_end, *dst; | |
784 Lisp_Object val; | |
785 int c, len; | |
786 | |
787 if (multibyte && nchars == nbytes) | |
788 return string; | |
789 | |
790 byte8_count = string_count_byte8 (string); | |
791 | |
792 if (byte8_count == 0) | |
793 return string; | |
794 | |
795 if (multibyte) | |
796 /* 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
|
797 val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
88363 | 798 nbytes + byte8_count * 2); |
799 else | |
800 /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
801 val = make_uninit_string (nbytes + byte8_count * 3); | |
802 | |
803 src = XSTRING (string)->data; | |
804 src_end = src + nbytes; | |
805 dst = XSTRING (val)->data; | |
806 if (multibyte) | |
807 while (src < src_end) | |
808 { | |
809 c = *src; | |
810 len = BYTES_BY_CHAR_HEAD (c); | |
811 | |
812 if (CHAR_BYTE8_HEAD_P (c)) | |
813 { | |
814 c = STRING_CHAR_ADVANCE (src); | |
815 c = CHAR_TO_BYTE8 (c); | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
816 sprintf ((char *) dst, "\\%03o", c); |
88363 | 817 dst += 4; |
818 } | |
819 else | |
820 while (len--) *dst++ = *src++; | |
821 } | |
822 else | |
823 while (src < src_end) | |
824 { | |
825 c = *src++; | |
826 if (c >= 0x80) | |
827 { | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
828 sprintf ((char *) dst, "\\%03o", c); |
88363 | 829 dst += 4; |
830 } | |
831 else | |
832 *dst++ = c; | |
833 } | |
834 return val; | |
835 } | |
836 | |
837 | |
838 DEFUN ("string", Fstring, Sstring, 1, MANY, 0, | |
839 doc: /* | |
88640 | 840 Concatenate all the argument characters and make the result a string. |
841 usage: (string &rest CHARACTERS) */) | |
88363 | 842 (n, args) |
843 int n; | |
844 Lisp_Object *args; | |
845 { | |
846 int i; | |
847 unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); | |
848 unsigned char *p = buf; | |
849 int c; | |
850 | |
851 for (i = 0; i < n; i++) | |
852 { | |
853 CHECK_CHARACTER (args[i]); | |
854 c = XINT (args[i]); | |
855 p += CHAR_STRING (c, p); | |
856 } | |
857 | |
858 return make_string_from_bytes ((char *) buf, n, p - buf); | |
859 } | |
860 | |
861 void | |
862 init_character_once () | |
863 { | |
864 } | |
865 | |
866 #ifdef emacs | |
867 | |
868 void | |
869 syms_of_character () | |
870 { | |
871 DEFSYM (Qcharacterp, "characterp"); | |
872 DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
873 | |
874 staticpro (&Vchar_unify_table); | |
875 Vchar_unify_table = Qnil; | |
876 | |
877 defsubr (&Smax_char); | |
878 defsubr (&Scharacterp); | |
879 defsubr (&Sunibyte_char_to_multibyte); | |
880 defsubr (&Smultibyte_char_to_unibyte); | |
881 defsubr (&Schar_bytes); | |
882 defsubr (&Schar_width); | |
883 defsubr (&Sstring_width); | |
884 defsubr (&Schar_direction); | |
885 defsubr (&Schars_in_region); | |
886 defsubr (&Sstring); | |
887 | |
888 DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
889 doc: /* | |
890 Vector of cons cell of a symbol and translation table ever defined. | |
891 An ID of a translation table is an index of this vector. */); | |
892 Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); | |
893 | |
894 DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
895 doc: /* | |
896 A char-table for characters which invoke auto-filling. | |
897 Such characters have value t in this table. */); | |
898 Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
899 CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt); | |
900 CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt); | |
901 | |
902 DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
903 doc: /* | |
904 A char-table for width (columns) of each character. */); | |
905 Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
906 | |
907 DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
908 doc: /* A char-table for direction of each character. */); | |
909 Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
910 | |
911 DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
912 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
|
913 Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
88363 | 914 } |
915 | |
916 #endif /* emacs */ |