Mercurial > emacs
annotate src/character.c @ 89122:92312a6e80d1
*** empty log message ***
author | Dave Love <fx@gnu.org> |
---|---|
date | Wed, 25 Sep 2002 00:00:33 +0000 |
parents | 8a6da305ac40 |
children | be75d5934738 |
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 | |
89020
b0277093a5f2
(_fetch_multibyte_char_len): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88990
diff
changeset
|
76 /* Variable used locally in the macro FETCH_MULTIBYTE_CHAR. */ |
88363 | 77 unsigned char *_fetch_multibyte_char_p; |
78 | |
88913
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
79 /* Char table of scripts. */ |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
80 Lisp_Object Vchar_script_table; |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
81 |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
82 static Lisp_Object Qchar_script_table; |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
83 |
89054
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
84 /* Mapping table from unibyte chars to multibyte chars. */ |
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
85 int unibyte_to_multibyte_table[256]; |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
86 |
88363 | 87 |
88 | |
89 int | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
90 char_string_with_unification (c, p) |
88363 | 91 int c; |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
92 unsigned char *p; |
88363 | 93 { |
94 int bytes; | |
95 | |
96 MAYBE_UNIFY_CHAR (c); | |
97 | |
98 if (c <= MAX_3_BYTE_CHAR || c > MAX_5_BYTE_CHAR) | |
99 { | |
100 bytes = CHAR_STRING (c, p); | |
101 } | |
102 else if (c <= MAX_4_BYTE_CHAR) | |
103 { | |
104 p[0] = (0xF0 | (c >> 18)); | |
105 p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
106 p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
107 p[3] = (0x80 | (c & 0x3F)); | |
108 bytes = 4; | |
109 } | |
110 else | |
111 { | |
112 p[0] = 0xF8; | |
113 p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
114 p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
115 p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
116 p[4] = (0x80 | (c & 0x3F)); | |
117 bytes = 5; | |
118 } | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
119 |
88363 | 120 return bytes; |
121 } | |
122 | |
123 | |
124 int | |
125 string_char_with_unification (p, advanced, len) | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
126 const unsigned char *p; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
127 const unsigned char **advanced; |
88363 | 128 int *len; |
129 { | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
130 int c; |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
131 const unsigned char *saved_p = p; |
88363 | 132 |
133 if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
134 { | |
135 c = STRING_CHAR_ADVANCE (p); | |
136 } | |
137 else if (! (*p & 0x08)) | |
138 { | |
139 c = ((((p)[0] & 0xF) << 18) | |
140 | (((p)[1] & 0x3F) << 12) | |
141 | (((p)[2] & 0x3F) << 6) | |
142 | ((p)[3] & 0x3F)); | |
143 p += 4; | |
144 } | |
145 else | |
146 { | |
147 c = ((((p)[1] & 0x3F) << 18) | |
148 | (((p)[2] & 0x3F) << 12) | |
149 | (((p)[3] & 0x3F) << 6) | |
150 | ((p)[4] & 0x3F)); | |
151 p += 5; | |
152 } | |
153 | |
154 MAYBE_UNIFY_CHAR (c); | |
155 | |
156 if (len) | |
157 *len = p - saved_p; | |
158 if (advanced) | |
159 *advanced = p; | |
160 return c; | |
161 } | |
162 | |
163 | |
164 /* Translate character C by translation table TABLE. If C is | |
165 negative, translate a character specified by CHARSET and CODE. If | |
166 no translation is found in TABLE, return the untranslated | |
167 character. */ | |
168 | |
169 int | |
170 translate_char (table, c) | |
171 Lisp_Object table; | |
172 int c; | |
173 { | |
174 Lisp_Object ch; | |
175 | |
176 if (! CHAR_TABLE_P (table)) | |
177 return c; | |
178 ch = CHAR_TABLE_REF (table, c); | |
179 if (! CHARACTERP (ch)) | |
180 return c; | |
181 return XINT (ch); | |
182 } | |
183 | |
184 /* Convert the multibyte character C to unibyte 8-bit character based | |
88983
2d504d707ce6
(unibyte_char_to_multibyte): Refer to
Kenichi Handa <handa@m17n.org>
parents:
88947
diff
changeset
|
185 on the current value of charset_unibyte. If dimension of |
2d504d707ce6
(unibyte_char_to_multibyte): Refer to
Kenichi Handa <handa@m17n.org>
parents:
88947
diff
changeset
|
186 charset_unibyte is more than one, return (C & 0xFF). |
88363 | 187 |
188 The argument REV_TBL is now ignored. It will be removed in the | |
189 future. */ | |
190 | |
191 int | |
192 multibyte_char_to_unibyte (c, rev_tbl) | |
193 int c; | |
194 Lisp_Object rev_tbl; | |
195 { | |
89054
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
196 struct charset *charset; |
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
197 unsigned c1; |
88363 | 198 |
89054
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
199 if (CHAR_BYTE8_P (c)) |
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
200 return CHAR_TO_BYTE8 (c); |
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
201 charset = CHARSET_FROM_ID (charset_unibyte); |
8a6da305ac40
(unibyte_to_multibyte_table): New variable.
Kenichi Handa <handa@m17n.org>
parents:
89029
diff
changeset
|
202 c1 = ENCODE_CHAR (charset, c); |
88363 | 203 return ((c1 != CHARSET_INVALID_CODE (charset)) ? c1 : c & 0xFF); |
204 } | |
205 | |
206 | |
207 DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
208 doc: /* Return non-nil if OBJECT is a character. */) | |
209 (object, ignore) | |
210 Lisp_Object object, ignore; | |
211 { | |
212 return (CHARACTERP (object) ? Qt : Qnil); | |
213 } | |
214 | |
215 DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
216 doc: /* Return the character of the maximum code. */) | |
217 () | |
218 { | |
219 return make_number (MAX_CHAR); | |
220 } | |
221 | |
222 DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
223 Sunibyte_char_to_multibyte, 1, 1, 0, | |
224 doc: /* Convert the unibyte character CH to multibyte character. | |
225 The multibyte character is a result of decoding CH by | |
88990
16317690f974
(Funibyte_char_to_multibyte): Doc fix.
Dave Love <fx@gnu.org>
parents:
88983
diff
changeset
|
226 the current unibyte charset (see `unibyte-charset'). */) |
88363 | 227 (ch) |
228 Lisp_Object ch; | |
229 { | |
230 int c; | |
231 struct charset *charset; | |
232 | |
233 CHECK_CHARACTER (ch); | |
234 c = XFASTINT (ch); | |
235 if (c >= 0400) | |
236 error ("Invalid unibyte character: %d", c); | |
88983
2d504d707ce6
(unibyte_char_to_multibyte): Refer to
Kenichi Handa <handa@m17n.org>
parents:
88947
diff
changeset
|
237 charset = CHARSET_FROM_ID (charset_unibyte); |
88363 | 238 c = DECODE_CHAR (charset, c); |
239 if (c < 0) | |
88947
894b8bce3208
(Funibyte_char_to_multibyte): If C can't be decoded
Kenichi Handa <handa@m17n.org>
parents:
88926
diff
changeset
|
240 c = BYTE8_TO_CHAR (XFASTINT (ch)); |
88363 | 241 return make_number (c); |
242 } | |
243 | |
244 DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
245 Smultibyte_char_to_unibyte, 1, 1, 0, | |
246 doc: /* Convert the multibyte character CH to unibyte character.\n\ | |
247 The unibyte character is a result of encoding CH by | |
248 the current primary charset (value of `charset-primary'). */) | |
249 (ch) | |
250 Lisp_Object ch; | |
251 { | |
252 int c; | |
253 | |
254 CHECK_CHARACTER (ch); | |
255 c = XFASTINT (ch); | |
88947
894b8bce3208
(Funibyte_char_to_multibyte): If C can't be decoded
Kenichi Handa <handa@m17n.org>
parents:
88926
diff
changeset
|
256 c = CHAR_TO_BYTE8 (c); |
894b8bce3208
(Funibyte_char_to_multibyte): If C can't be decoded
Kenichi Handa <handa@m17n.org>
parents:
88926
diff
changeset
|
257 return make_number (c); |
88363 | 258 } |
259 | |
260 DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
261 doc: /* Return 1 regardless of the argument CHAR. | |
262 This is now an obsolete function. We keep it just for backward compatibility. */) | |
263 (ch) | |
264 Lisp_Object ch; | |
265 { | |
266 CHECK_CHARACTER (ch); | |
267 return make_number (1); | |
268 } | |
269 | |
270 DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
271 doc: /* Return width of CHAR when displayed in the current buffer. | |
272 The width is measured by how many columns it occupies on the screen. | |
273 Tab is taken to occupy `tab-width' columns. */) | |
274 (ch) | |
275 Lisp_Object ch; | |
276 { | |
277 Lisp_Object disp; | |
278 int c, width; | |
279 struct Lisp_Char_Table *dp = buffer_display_table (); | |
280 | |
281 CHECK_CHARACTER (ch); | |
282 c = XINT (ch); | |
283 | |
284 /* Get the way the display table would display it. */ | |
285 disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
286 | |
287 if (VECTORP (disp)) | |
288 width = ASIZE (disp); | |
289 else | |
290 width = CHAR_WIDTH (c); | |
291 | |
292 return make_number (width); | |
293 } | |
294 | |
295 /* Return width of string STR of length LEN when displayed in the | |
296 current buffer. The width is measured by how many columns it | |
297 occupies on the screen. If PRECISION > 0, return the width of | |
298 longest substring that doesn't exceed PRECISION, and set number of | |
299 characters and bytes of the substring in *NCHARS and *NBYTES | |
300 respectively. */ | |
301 | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
302 int |
88363 | 303 c_string_width (str, len, precision, nchars, nbytes) |
304 unsigned char *str; | |
305 int precision, *nchars, *nbytes; | |
306 { | |
307 int i = 0, i_byte = 0; | |
308 int width = 0; | |
309 struct Lisp_Char_Table *dp = buffer_display_table (); | |
310 | |
311 while (i_byte < len) | |
312 { | |
313 int bytes, thiswidth; | |
314 Lisp_Object val; | |
315 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
316 | |
317 if (dp) | |
318 { | |
319 val = DISP_CHAR_VECTOR (dp, c); | |
320 if (VECTORP (val)) | |
321 thiswidth = XVECTOR (val)->size; | |
322 else | |
323 thiswidth = CHAR_WIDTH (c); | |
324 } | |
325 else | |
326 { | |
327 thiswidth = CHAR_WIDTH (c); | |
328 } | |
329 | |
330 if (precision > 0 | |
331 && (width + thiswidth > precision)) | |
332 { | |
333 *nchars = i; | |
334 *nbytes = i_byte; | |
335 return width; | |
336 } | |
337 i++; | |
338 i_byte += bytes; | |
339 width += thiswidth; | |
340 } | |
341 | |
342 if (precision > 0) | |
343 { | |
344 *nchars = i; | |
345 *nbytes = i_byte; | |
346 } | |
347 | |
348 return width; | |
349 } | |
350 | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
351 /* 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
|
352 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
|
353 occupies on the screen. */ |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
354 |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
355 int |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
356 strwidth (str, len) |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
357 unsigned char *str; |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
358 int len; |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
359 { |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
360 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
|
361 } |
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
362 |
88363 | 363 /* Return width of Lisp string STRING when displayed in the current |
364 buffer. The width is measured by how many columns it occupies on | |
365 the screen while paying attention to compositions. If PRECISION > | |
366 0, return the width of longest substring that doesn't exceed | |
367 PRECISION, and set number of characters and bytes of the substring | |
368 in *NCHARS and *NBYTES respectively. */ | |
369 | |
370 int | |
371 lisp_string_width (string, precision, nchars, nbytes) | |
372 Lisp_Object string; | |
373 int precision, *nchars, *nbytes; | |
374 { | |
375 int len = XSTRING (string)->size; | |
376 unsigned char *str = XSTRING (string)->data; | |
377 int i = 0, i_byte = 0; | |
378 int width = 0; | |
379 struct Lisp_Char_Table *dp = buffer_display_table (); | |
380 | |
381 while (i < len) | |
382 { | |
383 int chars, bytes, thiswidth; | |
384 Lisp_Object val; | |
385 int cmp_id; | |
386 int ignore, end; | |
387 | |
388 if (find_composition (i, -1, &ignore, &end, &val, string) | |
389 && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
390 >= 0)) | |
391 { | |
392 thiswidth = composition_table[cmp_id]->width; | |
393 chars = end - i; | |
394 bytes = string_char_to_byte (string, end) - i_byte; | |
395 } | |
396 else if (dp) | |
397 { | |
398 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
399 | |
400 chars = 1; | |
401 val = DISP_CHAR_VECTOR (dp, c); | |
402 if (VECTORP (val)) | |
403 thiswidth = XVECTOR (val)->size; | |
404 else | |
405 thiswidth = CHAR_WIDTH (c); | |
406 } | |
407 else | |
408 { | |
409 int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
410 | |
411 chars = 1; | |
412 thiswidth = CHAR_WIDTH (c); | |
413 } | |
414 | |
415 if (precision > 0 | |
416 && (width + thiswidth > precision)) | |
417 { | |
418 *nchars = i; | |
419 *nbytes = i_byte; | |
420 return width; | |
421 } | |
422 i += chars; | |
423 i_byte += bytes; | |
424 width += thiswidth; | |
425 } | |
426 | |
427 if (precision > 0) | |
428 { | |
429 *nchars = i; | |
430 *nbytes = i_byte; | |
431 } | |
432 | |
433 return width; | |
434 } | |
435 | |
436 DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
437 doc: /* Return width of STRING when displayed in the current buffer. | |
438 Width is measured by how many columns it occupies on the screen. | |
439 When calculating width of a multibyte character in STRING, | |
440 only the base leading-code is considered; the validity of | |
441 the following bytes is not checked. Tabs in STRING are always | |
442 taken to occupy `tab-width' columns. */) | |
443 (str) | |
444 Lisp_Object str; | |
445 { | |
446 Lisp_Object val; | |
447 | |
448 CHECK_STRING (str); | |
449 XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
450 return val; | |
451 } | |
452 | |
453 DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
454 doc: /* Return the direction of CHAR. | |
455 The returned value is 0 for left-to-right and 1 for right-to-left. */) | |
456 (ch) | |
457 Lisp_Object ch; | |
458 { | |
459 int c; | |
460 | |
461 CHECK_CHARACTER (ch); | |
462 c = XINT (ch); | |
463 return CHAR_TABLE_REF (Vchar_direction_table, c); | |
464 } | |
465 | |
466 DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, | |
467 doc: /* Return number of characters between BEG and END. | |
468 This is now an obsolete function. We keep it just for backward compatibility. */) | |
469 (beg, end) | |
470 Lisp_Object beg, end; | |
471 { | |
472 int from, to; | |
473 | |
474 CHECK_NUMBER_COERCE_MARKER (beg); | |
475 CHECK_NUMBER_COERCE_MARKER (end); | |
476 | |
477 from = min (XFASTINT (beg), XFASTINT (end)); | |
478 to = max (XFASTINT (beg), XFASTINT (end)); | |
479 | |
480 return make_number (to - from); | |
481 } | |
482 | |
483 /* Return the number of characters in the NBYTES bytes at PTR. | |
484 This works by looking at the contents and checking for multibyte | |
485 sequences while assuming that there's no invalid sequence. | |
486 However, if the current buffer has enable-multibyte-characters = | |
487 nil, we treat each byte as a character. */ | |
488 | |
489 int | |
490 chars_in_text (ptr, nbytes) | |
491 unsigned char *ptr; | |
492 int nbytes; | |
493 { | |
494 /* current_buffer is null at early stages of Emacs initialization. */ | |
495 if (current_buffer == 0 | |
496 || NILP (current_buffer->enable_multibyte_characters)) | |
497 return nbytes; | |
498 | |
499 return multibyte_chars_in_text (ptr, nbytes); | |
500 } | |
501 | |
502 /* Return the number of characters in the NBYTES bytes at PTR. | |
503 This works by looking at the contents and checking for multibyte | |
504 sequences while assuming that there's no invalid sequence. It | |
505 ignores enable-multibyte-characters. */ | |
506 | |
507 int | |
508 multibyte_chars_in_text (ptr, nbytes) | |
509 unsigned char *ptr; | |
510 int nbytes; | |
511 { | |
512 unsigned char *endp = ptr + nbytes; | |
513 int chars = 0; | |
514 | |
515 while (ptr < endp) | |
516 { | |
517 int len = MULTIBYTE_LENGTH (ptr, endp); | |
518 | |
519 if (len == 0) | |
520 abort (); | |
521 ptr += len; | |
522 chars++; | |
523 } | |
524 | |
525 return chars; | |
526 } | |
527 | |
528 /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
529 characters and bytes in it, and store them in *NCHARS and *NBYTES | |
530 respectively. On counting bytes, pay attention to that 8-bit | |
531 characters not constructing a valid multibyte sequence are | |
532 represented by 2-byte in a multibyte text. */ | |
533 | |
534 void | |
535 parse_str_as_multibyte (str, len, nchars, nbytes) | |
536 unsigned char *str; | |
537 int len, *nchars, *nbytes; | |
538 { | |
539 unsigned char *endp = str + len; | |
540 int n, chars = 0, bytes = 0; | |
541 | |
542 if (len >= MAX_MULTIBYTE_LENGTH) | |
543 { | |
544 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
545 while (str < adjusted_endp) | |
546 { | |
547 if ((n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
548 str += n, bytes += n; | |
549 else | |
550 str++, bytes += 2; | |
551 chars++; | |
552 } | |
553 } | |
554 while (str < endp) | |
555 { | |
556 if ((n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
557 str += n, bytes += n; | |
558 else | |
559 str++, bytes += 2; | |
560 chars++; | |
561 } | |
562 | |
563 *nchars = chars; | |
564 *nbytes = bytes; | |
565 return; | |
566 } | |
567 | |
568 /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
569 It actually converts only such 8-bit characters that don't contruct | |
570 a multibyte sequence to multibyte forms of Latin-1 characters. If | |
571 NCHARS is nonzero, set *NCHARS to the number of characters in the | |
572 text. It is assured that we can use LEN bytes at STR as a work | |
573 area and that is enough. Return the number of bytes of the | |
574 resulting text. */ | |
575 | |
576 int | |
577 str_as_multibyte (str, len, nbytes, nchars) | |
578 unsigned char *str; | |
579 int len, nbytes, *nchars; | |
580 { | |
581 unsigned char *p = str, *endp = str + nbytes; | |
582 unsigned char *to; | |
583 int chars = 0; | |
584 int n; | |
585 | |
586 if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
587 { | |
588 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
589 while (p < adjusted_endp | |
590 && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
591 p += n, chars++; | |
592 } | |
593 while ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
594 p += n, chars++; | |
595 if (nchars) | |
596 *nchars = chars; | |
597 if (p == endp) | |
598 return nbytes; | |
599 | |
600 to = p; | |
601 nbytes = endp - p; | |
602 endp = str + len; | |
603 safe_bcopy ((char *) p, (char *) (endp - nbytes), nbytes); | |
604 p = endp - nbytes; | |
605 | |
606 if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
607 { | |
608 unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
609 while (p < adjusted_endp) | |
610 { | |
611 if ((n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
612 { | |
613 while (n--) | |
614 *to++ = *p++; | |
615 } | |
616 else | |
617 { | |
618 int c = *p++; | |
619 c = BYTE8_TO_CHAR (c); | |
620 to += CHAR_STRING (c, to); | |
621 } | |
622 } | |
623 chars++; | |
624 } | |
625 while (p < endp) | |
626 { | |
627 if ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
628 { | |
629 while (n--) | |
630 *to++ = *p++; | |
631 } | |
632 else | |
633 { | |
634 int c = *p++; | |
635 c = BYTE8_TO_CHAR (c); | |
636 to += CHAR_STRING (c, to); | |
637 } | |
638 chars++; | |
639 } | |
640 if (nchars) | |
641 *nchars = chars; | |
642 return (to - str); | |
643 } | |
644 | |
645 /* Parse unibyte string at STR of LEN bytes, and return the number of | |
646 bytes it may ocupy when converted to multibyte string by | |
647 `str_to_multibyte'. */ | |
648 | |
649 int | |
650 parse_str_to_multibyte (str, len) | |
651 unsigned char *str; | |
652 int len; | |
653 { | |
654 unsigned char *endp = str + len; | |
655 int bytes; | |
656 | |
657 for (bytes = 0; str < endp; str++) | |
658 bytes += (*str < 0x80) ? 1 : 2; | |
659 return bytes; | |
660 } | |
661 | |
662 | |
663 /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
664 that contains the same single-byte characters. It actually | |
665 converts all 8-bit characters to multibyte forms. It is assured | |
666 that we can use LEN bytes at STR as a work area and that is | |
667 enough. */ | |
668 | |
669 int | |
670 str_to_multibyte (str, len, bytes) | |
671 unsigned char *str; | |
672 int len, bytes; | |
673 { | |
674 unsigned char *p = str, *endp = str + bytes; | |
675 unsigned char *to; | |
676 | |
677 while (p < endp && *p < 0x80) p++; | |
678 if (p == endp) | |
679 return bytes; | |
680 to = p; | |
681 bytes = endp - p; | |
682 endp = str + len; | |
683 safe_bcopy ((char *) p, (char *) (endp - bytes), bytes); | |
684 p = endp - bytes; | |
685 while (p < endp) | |
686 { | |
687 int c = *p++; | |
688 | |
689 if (c >= 0x80) | |
690 c = BYTE8_TO_CHAR (c); | |
691 to += CHAR_STRING (c, to); | |
692 } | |
693 return (to - str); | |
694 } | |
695 | |
696 /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
697 actually converts characters in the range 0x80..0xFF to | |
698 unibyte. */ | |
699 | |
700 int | |
701 str_as_unibyte (str, bytes) | |
702 unsigned char *str; | |
703 int bytes; | |
704 { | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
705 const unsigned char *p = str, *endp = str + bytes; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
706 unsigned char *to; |
88363 | 707 int c, len; |
708 | |
709 while (p < endp) | |
710 { | |
711 c = *p; | |
712 len = BYTES_BY_CHAR_HEAD (c); | |
713 if (CHAR_BYTE8_HEAD_P (c)) | |
714 break; | |
715 p += len; | |
716 } | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
717 to = str + (p - str); |
88363 | 718 while (p < endp) |
719 { | |
720 c = *p; | |
721 len = BYTES_BY_CHAR_HEAD (c); | |
722 if (CHAR_BYTE8_HEAD_P (c)) | |
723 { | |
724 c = STRING_CHAR_ADVANCE (p); | |
725 *to++ = CHAR_TO_BYTE8 (c); | |
726 } | |
727 else | |
728 { | |
729 while (len--) *to++ = *p++; | |
730 } | |
731 } | |
732 return (to - str); | |
733 } | |
734 | |
735 int | |
736 string_count_byte8 (string) | |
737 Lisp_Object string; | |
738 { | |
739 int multibyte = STRING_MULTIBYTE (string); | |
740 int nbytes = STRING_BYTES (XSTRING (string)); | |
741 unsigned char *p = XSTRING (string)->data; | |
742 unsigned char *pend = p + nbytes; | |
743 int count = 0; | |
744 int c, len; | |
745 | |
746 if (multibyte) | |
747 while (p < pend) | |
748 { | |
749 c = *p; | |
750 len = BYTES_BY_CHAR_HEAD (c); | |
751 | |
752 if (CHAR_BYTE8_HEAD_P (c)) | |
753 count++; | |
754 p += len; | |
755 } | |
756 else | |
757 while (p < pend) | |
758 { | |
759 if (*p++ >= 0x80) | |
760 count++; | |
761 } | |
762 return count; | |
763 } | |
764 | |
765 | |
766 Lisp_Object | |
767 string_escape_byte8 (string) | |
768 Lisp_Object string; | |
769 { | |
770 int nchars = XSTRING (string)->size; | |
771 int nbytes = STRING_BYTES (XSTRING (string)); | |
772 int multibyte = STRING_MULTIBYTE (string); | |
773 int byte8_count; | |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
774 const unsigned char *src, *src_end; |
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
775 unsigned char *dst; |
88363 | 776 Lisp_Object val; |
777 int c, len; | |
778 | |
779 if (multibyte && nchars == nbytes) | |
780 return string; | |
781 | |
782 byte8_count = string_count_byte8 (string); | |
783 | |
784 if (byte8_count == 0) | |
785 return string; | |
786 | |
787 if (multibyte) | |
788 /* 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
|
789 val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
88363 | 790 nbytes + byte8_count * 2); |
791 else | |
792 /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
793 val = make_uninit_string (nbytes + byte8_count * 3); | |
794 | |
795 src = XSTRING (string)->data; | |
796 src_end = src + nbytes; | |
797 dst = XSTRING (val)->data; | |
798 if (multibyte) | |
799 while (src < src_end) | |
800 { | |
801 c = *src; | |
802 len = BYTES_BY_CHAR_HEAD (c); | |
803 | |
804 if (CHAR_BYTE8_HEAD_P (c)) | |
805 { | |
806 c = STRING_CHAR_ADVANCE (src); | |
807 c = CHAR_TO_BYTE8 (c); | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
808 sprintf ((char *) dst, "\\%03o", c); |
88363 | 809 dst += 4; |
810 } | |
811 else | |
812 while (len--) *dst++ = *src++; | |
813 } | |
814 else | |
815 while (src < src_end) | |
816 { | |
817 c = *src++; | |
818 if (c >= 0x80) | |
819 { | |
88427
1a3aec316071
(c_string_width): Add return type `int'.
Kenichi Handa <handa@m17n.org>
parents:
88363
diff
changeset
|
820 sprintf ((char *) dst, "\\%03o", c); |
88363 | 821 dst += 4; |
822 } | |
823 else | |
824 *dst++ = c; | |
825 } | |
826 return val; | |
827 } | |
828 | |
829 | |
830 DEFUN ("string", Fstring, Sstring, 1, MANY, 0, | |
831 doc: /* | |
88640 | 832 Concatenate all the argument characters and make the result a string. |
833 usage: (string &rest CHARACTERS) */) | |
88363 | 834 (n, args) |
835 int n; | |
836 Lisp_Object *args; | |
837 { | |
838 int i; | |
839 unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); | |
840 unsigned char *p = buf; | |
841 int c; | |
842 | |
843 for (i = 0; i < n; i++) | |
844 { | |
845 CHECK_CHARACTER (args[i]); | |
846 c = XINT (args[i]); | |
847 p += CHAR_STRING (c, p); | |
848 } | |
849 | |
850 return make_string_from_bytes ((char *) buf, n, p - buf); | |
851 } | |
852 | |
853 void | |
854 init_character_once () | |
855 { | |
856 } | |
857 | |
858 #ifdef emacs | |
859 | |
860 void | |
861 syms_of_character () | |
862 { | |
863 DEFSYM (Qcharacterp, "characterp"); | |
864 DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
865 | |
866 staticpro (&Vchar_unify_table); | |
867 Vchar_unify_table = Qnil; | |
868 | |
869 defsubr (&Smax_char); | |
870 defsubr (&Scharacterp); | |
871 defsubr (&Sunibyte_char_to_multibyte); | |
872 defsubr (&Smultibyte_char_to_unibyte); | |
873 defsubr (&Schar_bytes); | |
874 defsubr (&Schar_width); | |
875 defsubr (&Sstring_width); | |
876 defsubr (&Schar_direction); | |
877 defsubr (&Schars_in_region); | |
878 defsubr (&Sstring); | |
879 | |
880 DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
881 doc: /* | |
88926
f026b7e4fc61
(syms_of_character) <translation-table-vector>: Doc
Dave Love <fx@gnu.org>
parents:
88913
diff
changeset
|
882 Vector recording all translation tables ever defined. |
f026b7e4fc61
(syms_of_character) <translation-table-vector>: Doc
Dave Love <fx@gnu.org>
parents:
88913
diff
changeset
|
883 Each element is a pair (SYMBOL . TABLE) relating the table to the |
f026b7e4fc61
(syms_of_character) <translation-table-vector>: Doc
Dave Love <fx@gnu.org>
parents:
88913
diff
changeset
|
884 symbol naming it. The ID of a translation table is an index into this vector. */); |
88363 | 885 Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
886 | |
887 DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
888 doc: /* | |
889 A char-table for characters which invoke auto-filling. | |
890 Such characters have value t in this table. */); | |
891 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
|
892 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
|
893 CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); |
88363 | 894 |
895 DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
896 doc: /* | |
897 A char-table for width (columns) of each character. */); | |
898 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
|
899 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
|
900 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
|
901 make_number (4)); |
88363 | 902 |
903 DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
904 doc: /* A char-table for direction of each character. */); | |
905 Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
906 | |
907 DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
908 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
|
909 Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
89029
ad8f56efa7c6
(syms_of_character): Setup Vprintable_chars.
Kenichi Handa <handa@m17n.org>
parents:
89020
diff
changeset
|
910 Fset_char_table_range (Vprintable_chars, |
ad8f56efa7c6
(syms_of_character): Setup Vprintable_chars.
Kenichi Handa <handa@m17n.org>
parents:
89020
diff
changeset
|
911 Fcons (make_number (32), make_number (126)), Qt); |
ad8f56efa7c6
(syms_of_character): Setup Vprintable_chars.
Kenichi Handa <handa@m17n.org>
parents:
89020
diff
changeset
|
912 Fset_char_table_range (Vprintable_chars, |
ad8f56efa7c6
(syms_of_character): Setup Vprintable_chars.
Kenichi Handa <handa@m17n.org>
parents:
89020
diff
changeset
|
913 Fcons (make_number (160), |
ad8f56efa7c6
(syms_of_character): Setup Vprintable_chars.
Kenichi Handa <handa@m17n.org>
parents:
89020
diff
changeset
|
914 make_number (MAX_5_BYTE_CHAR)), Qt); |
88878
a0ae1a5876c7
(Vscript_alist): New variable.
Kenichi Handa <handa@m17n.org>
parents:
88853
diff
changeset
|
915 |
88913
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
916 DEFVAR_LISP ("char-script-table", &Vchar_script_table, |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
917 doc: /* Char table of script symbols. |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
918 It has one extra slot whose value is a list of script symbols. */); |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
919 |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
920 /* Intern this now in case it isn't already done. |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
921 Setting this variable twice is harmless. |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
922 But don't staticpro it here--that is done in alloc.c. */ |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
923 Qchar_table_extra_slots = intern ("char-table-extra-slots"); |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
924 DEFSYM (Qchar_script_table, "char-script-table"); |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
925 Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); |
76074dea1258
(Vscript_alist): This variable deleted.
Kenichi Handa <handa@m17n.org>
parents:
88878
diff
changeset
|
926 Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); |
88363 | 927 } |
928 | |
929 #endif /* emacs */ |