Commit | Line | Data |
---|---|---|
0168c3d8 | 1 | /* Basic character support. |
73b0cd50 | 2 | |
ab422c4d | 3 | Copyright (C) 2001-2013 Free Software Foundation, Inc. |
73b0cd50 GM |
4 | Copyright (C) 1995, 1997, 1998, 2001 Electrotechnical Laboratory, JAPAN. |
5 | Licensed to the Free Software Foundation. | |
6 | Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011 | |
7 | National Institute of Advanced Industrial Science and Technology (AIST) | |
8 | Registration Number H13PRO009 | |
0168c3d8 KH |
9 | |
10 | This file is part of GNU Emacs. | |
11 | ||
9ec0b715 | 12 | GNU Emacs is free software: you can redistribute it and/or modify |
0168c3d8 | 13 | it under the terms of the GNU General Public License as published by |
9ec0b715 GM |
14 | the Free Software Foundation, either version 3 of the License, or |
15 | (at your option) any later version. | |
0168c3d8 KH |
16 | |
17 | GNU Emacs is distributed in the hope that it will be useful, | |
18 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
19 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
20 | GNU General Public License for more details. | |
21 | ||
22 | You should have received a copy of the GNU General Public License | |
9ec0b715 | 23 | along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */ |
0168c3d8 KH |
24 | |
25 | /* At first, see the document in `character.h' to understand the code | |
26 | in this file. */ | |
27 | ||
28 | #ifdef emacs | |
29 | #include <config.h> | |
30 | #endif | |
31 | ||
f162bcc3 PE |
32 | #define CHARACTER_INLINE EXTERN_INLINE |
33 | ||
0168c3d8 KH |
34 | #include <stdio.h> |
35 | ||
36 | #ifdef emacs | |
37 | ||
38 | #include <sys/types.h> | |
2b4560a8 | 39 | #include <intprops.h> |
0168c3d8 KH |
40 | #include "lisp.h" |
41 | #include "character.h" | |
42 | #include "buffer.h" | |
43 | #include "charset.h" | |
44 | #include "composite.h" | |
45 | #include "disptab.h" | |
46 | ||
47 | #else /* not emacs */ | |
48 | ||
49 | #include "mulelib.h" | |
50 | ||
51 | #endif /* emacs */ | |
52 | ||
53 | Lisp_Object Qcharacterp; | |
54 | ||
955cbe7b | 55 | static Lisp_Object Qauto_fill_chars; |
0168c3d8 | 56 | |
33f91981 | 57 | /* Char-table of information about which character to unify to which |
6f1b43a0 | 58 | Unicode character. Mainly used by the macro MAYBE_UNIFY_CHAR. */ |
0168c3d8 KH |
59 | Lisp_Object Vchar_unify_table; |
60 | ||
c57f3328 KH |
61 | static Lisp_Object Qchar_script_table; |
62 | ||
0168c3d8 KH |
63 | \f |
64 | ||
2bde7652 KH |
65 | /* If character code C has modifier masks, reflect them to the |
66 | character code if possible. Return the resulting code. */ | |
67 | ||
d311d28c PE |
68 | EMACS_INT |
69 | char_resolve_modifier_mask (EMACS_INT c) | |
2bde7652 | 70 | { |
d0363d44 | 71 | /* A non-ASCII character can't reflect modifier bits to the code. */ |
2bde7652 KH |
72 | if (! ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) |
73 | return c; | |
74 | ||
75 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
2bde7652 KH |
76 | if (c & CHAR_SHIFT) |
77 | { | |
78 | /* Shift modifier is valid only with [A-Za-z]. */ | |
79 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
80 | c &= ~CHAR_SHIFT; | |
81 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
82 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
03365d0e KH |
83 | /* Shift modifier for control characters and SPC is ignored. */ |
84 | else if ((c & ~CHAR_MODIFIER_MASK) <= 0x20) | |
d0363d44 KH |
85 | c &= ~CHAR_SHIFT; |
86 | } | |
2bde7652 KH |
87 | if (c & CHAR_CTL) |
88 | { | |
89 | /* Simulate the code in lread.c. */ | |
90 | /* Allow `\C- ' and `\C-?'. */ | |
03365d0e KH |
91 | if ((c & 0377) == ' ') |
92 | c &= ~0177 & ~ CHAR_CTL; | |
93 | else if ((c & 0377) == '?') | |
94 | c = 0177 | (c & ~0177 & ~CHAR_CTL); | |
2bde7652 KH |
95 | /* ASCII control chars are made from letters (both cases), |
96 | as well as the non-letters within 0100...0137. */ | |
97 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
98 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
99 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
100 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
101 | } | |
e9c1637d | 102 | #if 0 /* This is outside the scope of this function. (bug#4751) */ |
03365d0e KH |
103 | if (c & CHAR_META) |
104 | { | |
105 | /* Move the meta bit to the right place for a string. */ | |
106 | c = (c & ~CHAR_META) | 0x80; | |
107 | } | |
e9c1637d | 108 | #endif |
2bde7652 KH |
109 | |
110 | return c; | |
111 | } | |
112 | ||
113 | ||
33f91981 KH |
114 | /* Store multibyte form of character C at P. If C has modifier bits, |
115 | handle them appropriately. */ | |
116 | ||
0168c3d8 | 117 | int |
971de7fb | 118 | char_string (unsigned int c, unsigned char *p) |
0168c3d8 KH |
119 | { |
120 | int bytes; | |
121 | ||
e3d8eb8c KH |
122 | if (c & CHAR_MODIFIER_MASK) |
123 | { | |
c5958d4c | 124 | c = char_resolve_modifier_mask (c); |
e3d8eb8c KH |
125 | /* If C still has any modifier bits, just ignore it. */ |
126 | c &= ~CHAR_MODIFIER_MASK; | |
127 | } | |
128 | ||
e3d8eb8c | 129 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
130 | { |
131 | bytes = CHAR_STRING (c, p); | |
132 | } | |
133 | else if (c <= MAX_4_BYTE_CHAR) | |
134 | { | |
135 | p[0] = (0xF0 | (c >> 18)); | |
136 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
137 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
138 | p[3] = (0x80 | (c & 0x3F)); | |
139 | bytes = 4; | |
140 | } | |
e3d8eb8c | 141 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
142 | { |
143 | p[0] = 0xF8; | |
144 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
145 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
146 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
147 | p[4] = (0x80 | (c & 0x3F)); | |
148 | bytes = 5; | |
149 | } | |
5aa91c9b | 150 | else if (c <= MAX_CHAR) |
e3d8eb8c KH |
151 | { |
152 | c = CHAR_TO_BYTE8 (c); | |
153 | bytes = BYTE8_STRING (c, p); | |
154 | } | |
5aa91c9b | 155 | else |
e6c3da20 | 156 | error ("Invalid character: %x", c); |
1889b238 | 157 | |
0168c3d8 KH |
158 | return bytes; |
159 | } | |
160 | ||
161 | ||
224a3131 | 162 | /* Return a character whose multibyte form is at P. If LEN is not |
33f91981 | 163 | NULL, it must be a pointer to integer. In that case, set *LEN to |
224a3131 | 164 | the byte length of the multibyte form. If ADVANCED is not NULL, it |
33f91981 | 165 | must be a pointer to unsigned char. In that case, set *ADVANCED to |
224a3131 | 166 | the ending address (i.e., the starting address of the next |
33f91981 KH |
167 | character) of the multibyte form. */ |
168 | ||
0168c3d8 | 169 | int |
971de7fb | 170 | string_char (const unsigned char *p, const unsigned char **advanced, int *len) |
0168c3d8 | 171 | { |
1889b238 | 172 | int c; |
15843e6f | 173 | const unsigned char *saved_p = p; |
0168c3d8 KH |
174 | |
175 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
176 | { | |
177 | c = STRING_CHAR_ADVANCE (p); | |
178 | } | |
179 | else if (! (*p & 0x08)) | |
180 | { | |
181 | c = ((((p)[0] & 0xF) << 18) | |
182 | | (((p)[1] & 0x3F) << 12) | |
183 | | (((p)[2] & 0x3F) << 6) | |
184 | | ((p)[3] & 0x3F)); | |
185 | p += 4; | |
186 | } | |
187 | else | |
188 | { | |
189 | c = ((((p)[1] & 0x3F) << 18) | |
190 | | (((p)[2] & 0x3F) << 12) | |
191 | | (((p)[3] & 0x3F) << 6) | |
192 | | ((p)[4] & 0x3F)); | |
193 | p += 5; | |
194 | } | |
195 | ||
0168c3d8 KH |
196 | if (len) |
197 | *len = p - saved_p; | |
198 | if (advanced) | |
199 | *advanced = p; | |
200 | return c; | |
201 | } | |
202 | ||
203 | ||
224a3131 EZ |
204 | /* Translate character C by translation table TABLE. If no translation is |
205 | found in TABLE, return the untranslated character. If TABLE is a list, | |
206 | elements are char tables. In that case, recursively translate C by all the | |
207 | tables in the list. */ | |
0168c3d8 KH |
208 | |
209 | int | |
971de7fb | 210 | translate_char (Lisp_Object table, int c) |
0168c3d8 | 211 | { |
10453be9 KH |
212 | if (CHAR_TABLE_P (table)) |
213 | { | |
214 | Lisp_Object ch; | |
215 | ||
216 | ch = CHAR_TABLE_REF (table, c); | |
217 | if (CHARACTERP (ch)) | |
218 | c = XINT (ch); | |
219 | } | |
220 | else | |
221 | { | |
222 | for (; CONSP (table); table = XCDR (table)) | |
223 | c = translate_char (XCAR (table), c); | |
224 | } | |
225 | return c; | |
0168c3d8 KH |
226 | } |
227 | ||
2e5db15c | 228 | /* Convert ASCII or 8-bit character C to unibyte. If C is none of |
461c2ab9 | 229 | them, return (C & 0xFF). */ |
0168c3d8 KH |
230 | |
231 | int | |
461c2ab9 | 232 | multibyte_char_to_unibyte (int c) |
0168c3d8 | 233 | { |
2e5db15c KH |
234 | if (c < 0x80) |
235 | return c; | |
b672c5ae KH |
236 | if (CHAR_BYTE8_P (c)) |
237 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 238 | return (c & 0xFF); |
0168c3d8 KH |
239 | } |
240 | ||
935d5b02 KH |
241 | /* Like multibyte_char_to_unibyte, but return -1 if C is not supported |
242 | by charset_unibyte. */ | |
243 | ||
244 | int | |
971de7fb | 245 | multibyte_char_to_unibyte_safe (int c) |
935d5b02 | 246 | { |
2e5db15c KH |
247 | if (c < 0x80) |
248 | return c; | |
935d5b02 KH |
249 | if (CHAR_BYTE8_P (c)) |
250 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 251 | return -1; |
935d5b02 | 252 | } |
0168c3d8 KH |
253 | |
254 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
bc985c87 | 255 | doc: /* Return non-nil if OBJECT is a character. |
4abcdac8 CY |
256 | In Emacs Lisp, characters are represented by character codes, which |
257 | are non-negative integers. The function `max-char' returns the | |
258 | maximum character code. | |
bc985c87 | 259 | usage: (characterp OBJECT) */) |
5842a27b | 260 | (Lisp_Object object, Lisp_Object ignore) |
0168c3d8 KH |
261 | { |
262 | return (CHARACTERP (object) ? Qt : Qnil); | |
263 | } | |
264 | ||
265 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
266 | doc: /* Return the character of the maximum code. */) | |
5842a27b | 267 | (void) |
0168c3d8 KH |
268 | { |
269 | return make_number (MAX_CHAR); | |
270 | } | |
271 | ||
272 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
273 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
5556875b | 274 | doc: /* Convert the byte CH to multibyte character. */) |
5842a27b | 275 | (Lisp_Object ch) |
0168c3d8 KH |
276 | { |
277 | int c; | |
0168c3d8 KH |
278 | |
279 | CHECK_CHARACTER (ch); | |
280 | c = XFASTINT (ch); | |
2e5db15c KH |
281 | if (c >= 0x100) |
282 | error ("Not a unibyte character: %d", c); | |
4c0354d7 | 283 | MAKE_CHAR_MULTIBYTE (c); |
0168c3d8 KH |
284 | return make_number (c); |
285 | } | |
286 | ||
287 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
288 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
5556875b SM |
289 | doc: /* Convert the multibyte character CH to a byte. |
290 | If the multibyte character does not represent a byte, return -1. */) | |
5842a27b | 291 | (Lisp_Object ch) |
0168c3d8 | 292 | { |
5556875b | 293 | int cm; |
0168c3d8 KH |
294 | |
295 | CHECK_CHARACTER (ch); | |
5556875b SM |
296 | cm = XFASTINT (ch); |
297 | if (cm < 256) | |
298 | /* Can't distinguish a byte read from a unibyte buffer from | |
299 | a latin1 char, so let's let it slide. */ | |
300 | return ch; | |
301 | else | |
302 | { | |
2afc21f5 | 303 | int cu = CHAR_TO_BYTE_SAFE (cm); |
5556875b SM |
304 | return make_number (cu); |
305 | } | |
0168c3d8 KH |
306 | } |
307 | ||
25ed9e61 KH |
308 | |
309 | /* Return width (columns) of C considering the buffer display table DP. */ | |
310 | ||
6e6c82a4 | 311 | static ptrdiff_t |
25ed9e61 KH |
312 | char_width (int c, struct Lisp_Char_Table *dp) |
313 | { | |
6e6c82a4 | 314 | ptrdiff_t width = CHAR_WIDTH (c); |
25ed9e61 KH |
315 | |
316 | if (dp) | |
317 | { | |
318 | Lisp_Object disp = DISP_CHAR_VECTOR (dp, c), ch; | |
319 | int i; | |
320 | ||
321 | if (VECTORP (disp)) | |
322 | for (i = 0, width = 0; i < ASIZE (disp); i++) | |
323 | { | |
324 | ch = AREF (disp, i); | |
325 | if (CHARACTERP (ch)) | |
df0b2940 PE |
326 | { |
327 | int w = CHAR_WIDTH (XFASTINT (ch)); | |
328 | if (INT_ADD_OVERFLOW (width, w)) | |
329 | string_overflow (); | |
330 | width += w; | |
331 | } | |
25ed9e61 KH |
332 | } |
333 | } | |
334 | return width; | |
335 | } | |
336 | ||
337 | ||
a7ca3326 | 338 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, |
0168c3d8 KH |
339 | doc: /* Return width of CHAR when displayed in the current buffer. |
340 | The width is measured by how many columns it occupies on the screen. | |
add553ac JB |
341 | Tab is taken to occupy `tab-width' columns. |
342 | usage: (char-width CHAR) */) | |
5842a27b | 343 | (Lisp_Object ch) |
0168c3d8 | 344 | { |
df0b2940 | 345 | int c; |
6e6c82a4 | 346 | ptrdiff_t width; |
0168c3d8 KH |
347 | |
348 | CHECK_CHARACTER (ch); | |
349 | c = XINT (ch); | |
25ed9e61 | 350 | width = char_width (c, buffer_display_table ()); |
0168c3d8 KH |
351 | return make_number (width); |
352 | } | |
353 | ||
0168c3d8 KH |
354 | /* Return width of string STR of length LEN when displayed in the |
355 | current buffer. The width is measured by how many columns it | |
356 | occupies on the screen. If PRECISION > 0, return the width of | |
357 | longest substring that doesn't exceed PRECISION, and set number of | |
358 | characters and bytes of the substring in *NCHARS and *NBYTES | |
359 | respectively. */ | |
360 | ||
d311d28c PE |
361 | ptrdiff_t |
362 | c_string_width (const unsigned char *str, ptrdiff_t len, int precision, | |
363 | ptrdiff_t *nchars, ptrdiff_t *nbytes) | |
0168c3d8 | 364 | { |
d311d28c PE |
365 | ptrdiff_t i = 0, i_byte = 0; |
366 | ptrdiff_t width = 0; | |
0168c3d8 KH |
367 | struct Lisp_Char_Table *dp = buffer_display_table (); |
368 | ||
369 | while (i_byte < len) | |
370 | { | |
25ed9e61 | 371 | int bytes; |
62a6e103 | 372 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes); |
6e6c82a4 | 373 | ptrdiff_t thiswidth = char_width (c, dp); |
0168c3d8 | 374 | |
df0b2940 | 375 | if (precision <= 0) |
0168c3d8 | 376 | { |
df0b2940 PE |
377 | if (INT_ADD_OVERFLOW (width, thiswidth)) |
378 | string_overflow (); | |
0168c3d8 | 379 | } |
df0b2940 | 380 | else if (precision - width < thiswidth) |
0168c3d8 KH |
381 | { |
382 | *nchars = i; | |
383 | *nbytes = i_byte; | |
384 | return width; | |
385 | } | |
386 | i++; | |
387 | i_byte += bytes; | |
388 | width += thiswidth; | |
389 | } | |
390 | ||
391 | if (precision > 0) | |
392 | { | |
393 | *nchars = i; | |
394 | *nbytes = i_byte; | |
395 | } | |
396 | ||
397 | return width; | |
398 | } | |
399 | ||
1889b238 KH |
400 | /* Return width of string STR of length LEN when displayed in the |
401 | current buffer. The width is measured by how many columns it | |
402 | occupies on the screen. */ | |
403 | ||
d311d28c PE |
404 | ptrdiff_t |
405 | strwidth (const char *str, ptrdiff_t len) | |
1889b238 | 406 | { |
7469ef5d | 407 | return c_string_width ((const unsigned char *) str, len, -1, NULL, NULL); |
1889b238 KH |
408 | } |
409 | ||
0168c3d8 KH |
410 | /* Return width of Lisp string STRING when displayed in the current |
411 | buffer. The width is measured by how many columns it occupies on | |
412 | the screen while paying attention to compositions. If PRECISION > | |
413 | 0, return the width of longest substring that doesn't exceed | |
414 | PRECISION, and set number of characters and bytes of the substring | |
415 | in *NCHARS and *NBYTES respectively. */ | |
416 | ||
d311d28c PE |
417 | ptrdiff_t |
418 | lisp_string_width (Lisp_Object string, ptrdiff_t precision, | |
419 | ptrdiff_t *nchars, ptrdiff_t *nbytes) | |
0168c3d8 | 420 | { |
d311d28c | 421 | ptrdiff_t len = SCHARS (string); |
0aee65b9 KH |
422 | /* This set multibyte to 0 even if STRING is multibyte when it |
423 | contains only ascii and eight-bit-graphic, but that's | |
424 | intentional. */ | |
d5172d4f | 425 | bool multibyte = len < SBYTES (string); |
8f924df7 | 426 | unsigned char *str = SDATA (string); |
d311d28c PE |
427 | ptrdiff_t i = 0, i_byte = 0; |
428 | ptrdiff_t width = 0; | |
0168c3d8 KH |
429 | struct Lisp_Char_Table *dp = buffer_display_table (); |
430 | ||
431 | while (i < len) | |
432 | { | |
d311d28c | 433 | ptrdiff_t chars, bytes, thiswidth; |
0168c3d8 | 434 | Lisp_Object val; |
ebfa62c0 | 435 | ptrdiff_t cmp_id; |
d311d28c | 436 | ptrdiff_t ignore, end; |
0168c3d8 KH |
437 | |
438 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
439 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
440 | >= 0)) | |
441 | { | |
442 | thiswidth = composition_table[cmp_id]->width; | |
443 | chars = end - i; | |
444 | bytes = string_char_to_byte (string, end) - i_byte; | |
445 | } | |
0168c3d8 KH |
446 | else |
447 | { | |
0aee65b9 | 448 | int c; |
0168c3d8 | 449 | |
0aee65b9 | 450 | if (multibyte) |
2b4560a8 PE |
451 | { |
452 | int cbytes; | |
453 | c = STRING_CHAR_AND_LENGTH (str + i_byte, cbytes); | |
454 | bytes = cbytes; | |
455 | } | |
0aee65b9 KH |
456 | else |
457 | c = str[i_byte], bytes = 1; | |
0168c3d8 | 458 | chars = 1; |
25ed9e61 | 459 | thiswidth = char_width (c, dp); |
0168c3d8 KH |
460 | } |
461 | ||
2b4560a8 PE |
462 | if (precision <= 0) |
463 | { | |
464 | #ifdef emacs | |
465 | if (INT_ADD_OVERFLOW (width, thiswidth)) | |
466 | string_overflow (); | |
467 | #endif | |
468 | } | |
469 | else if (precision - width < thiswidth) | |
0168c3d8 KH |
470 | { |
471 | *nchars = i; | |
472 | *nbytes = i_byte; | |
473 | return width; | |
474 | } | |
475 | i += chars; | |
476 | i_byte += bytes; | |
477 | width += thiswidth; | |
ef3ff036 | 478 | } |
0168c3d8 KH |
479 | |
480 | if (precision > 0) | |
481 | { | |
482 | *nchars = i; | |
483 | *nbytes = i_byte; | |
484 | } | |
485 | ||
486 | return width; | |
487 | } | |
488 | ||
489 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
490 | doc: /* Return width of STRING when displayed in the current buffer. | |
491 | Width is measured by how many columns it occupies on the screen. | |
492 | When calculating width of a multibyte character in STRING, | |
493 | only the base leading-code is considered; the validity of | |
494 | the following bytes is not checked. Tabs in STRING are always | |
add553ac JB |
495 | taken to occupy `tab-width' columns. |
496 | usage: (string-width STRING) */) | |
5842a27b | 497 | (Lisp_Object str) |
0168c3d8 KH |
498 | { |
499 | Lisp_Object val; | |
500 | ||
501 | CHECK_STRING (str); | |
502 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
503 | return val; | |
504 | } | |
505 | ||
0168c3d8 KH |
506 | /* Return the number of characters in the NBYTES bytes at PTR. |
507 | This works by looking at the contents and checking for multibyte | |
508 | sequences while assuming that there's no invalid sequence. | |
509 | However, if the current buffer has enable-multibyte-characters = | |
510 | nil, we treat each byte as a character. */ | |
511 | ||
d311d28c PE |
512 | ptrdiff_t |
513 | chars_in_text (const unsigned char *ptr, ptrdiff_t nbytes) | |
0168c3d8 KH |
514 | { |
515 | /* current_buffer is null at early stages of Emacs initialization. */ | |
516 | if (current_buffer == 0 | |
4b4deea2 | 517 | || NILP (BVAR (current_buffer, enable_multibyte_characters))) |
0168c3d8 KH |
518 | return nbytes; |
519 | ||
520 | return multibyte_chars_in_text (ptr, nbytes); | |
521 | } | |
522 | ||
523 | /* Return the number of characters in the NBYTES bytes at PTR. | |
524 | This works by looking at the contents and checking for multibyte | |
525 | sequences while assuming that there's no invalid sequence. It | |
526 | ignores enable-multibyte-characters. */ | |
527 | ||
d311d28c PE |
528 | ptrdiff_t |
529 | multibyte_chars_in_text (const unsigned char *ptr, ptrdiff_t nbytes) | |
0168c3d8 | 530 | { |
8f924df7 | 531 | const unsigned char *endp = ptr + nbytes; |
d311d28c | 532 | ptrdiff_t chars = 0; |
0168c3d8 KH |
533 | |
534 | while (ptr < endp) | |
535 | { | |
d311d28c | 536 | int len = MULTIBYTE_LENGTH (ptr, endp); |
0168c3d8 KH |
537 | |
538 | if (len == 0) | |
1088b922 | 539 | emacs_abort (); |
0168c3d8 KH |
540 | ptr += len; |
541 | chars++; | |
542 | } | |
543 | ||
544 | return chars; | |
545 | } | |
546 | ||
547 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
548 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
549 | respectively. On counting bytes, pay attention to that 8-bit | |
550 | characters not constructing a valid multibyte sequence are | |
551 | represented by 2-byte in a multibyte text. */ | |
552 | ||
553 | void | |
d311d28c PE |
554 | parse_str_as_multibyte (const unsigned char *str, ptrdiff_t len, |
555 | ptrdiff_t *nchars, ptrdiff_t *nbytes) | |
0168c3d8 | 556 | { |
8f924df7 | 557 | const unsigned char *endp = str + len; |
d311d28c PE |
558 | int n; |
559 | ptrdiff_t chars = 0, bytes = 0; | |
0168c3d8 KH |
560 | |
561 | if (len >= MAX_MULTIBYTE_LENGTH) | |
562 | { | |
8f924df7 | 563 | const unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; |
0168c3d8 KH |
564 | while (str < adjusted_endp) |
565 | { | |
fc9a17bc KH |
566 | if (! CHAR_BYTE8_HEAD_P (*str) |
567 | && (n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
0168c3d8 KH |
568 | str += n, bytes += n; |
569 | else | |
570 | str++, bytes += 2; | |
571 | chars++; | |
572 | } | |
573 | } | |
574 | while (str < endp) | |
575 | { | |
fc9a17bc KH |
576 | if (! CHAR_BYTE8_HEAD_P (*str) |
577 | && (n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
0168c3d8 KH |
578 | str += n, bytes += n; |
579 | else | |
580 | str++, bytes += 2; | |
581 | chars++; | |
582 | } | |
583 | ||
584 | *nchars = chars; | |
585 | *nbytes = bytes; | |
586 | return; | |
587 | } | |
588 | ||
589 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
da6062e6 | 590 | It actually converts only such 8-bit characters that don't construct |
0168c3d8 KH |
591 | a multibyte sequence to multibyte forms of Latin-1 characters. If |
592 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
593 | text. It is assured that we can use LEN bytes at STR as a work | |
594 | area and that is enough. Return the number of bytes of the | |
595 | resulting text. */ | |
596 | ||
d311d28c PE |
597 | ptrdiff_t |
598 | str_as_multibyte (unsigned char *str, ptrdiff_t len, ptrdiff_t nbytes, | |
599 | ptrdiff_t *nchars) | |
0168c3d8 KH |
600 | { |
601 | unsigned char *p = str, *endp = str + nbytes; | |
602 | unsigned char *to; | |
d311d28c | 603 | ptrdiff_t chars = 0; |
0168c3d8 KH |
604 | int n; |
605 | ||
606 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
607 | { | |
608 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
609 | while (p < adjusted_endp | |
fc9a17bc | 610 | && ! CHAR_BYTE8_HEAD_P (*p) |
0168c3d8 KH |
611 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) |
612 | p += n, chars++; | |
613 | } | |
fc9a17bc KH |
614 | while (p < endp |
615 | && ! CHAR_BYTE8_HEAD_P (*p) | |
616 | && (n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
0168c3d8 KH |
617 | p += n, chars++; |
618 | if (nchars) | |
619 | *nchars = chars; | |
620 | if (p == endp) | |
621 | return nbytes; | |
622 | ||
623 | to = p; | |
624 | nbytes = endp - p; | |
625 | endp = str + len; | |
72af86bd | 626 | memmove (endp - nbytes, p, nbytes); |
0168c3d8 KH |
627 | p = endp - nbytes; |
628 | ||
629 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
630 | { | |
631 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
632 | while (p < adjusted_endp) | |
633 | { | |
fc9a17bc KH |
634 | if (! CHAR_BYTE8_HEAD_P (*p) |
635 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
0168c3d8 KH |
636 | { |
637 | while (n--) | |
638 | *to++ = *p++; | |
639 | } | |
640 | else | |
641 | { | |
642 | int c = *p++; | |
643 | c = BYTE8_TO_CHAR (c); | |
644 | to += CHAR_STRING (c, to); | |
645 | } | |
646 | } | |
647 | chars++; | |
648 | } | |
649 | while (p < endp) | |
650 | { | |
fc9a17bc KH |
651 | if (! CHAR_BYTE8_HEAD_P (*p) |
652 | && (n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
0168c3d8 KH |
653 | { |
654 | while (n--) | |
655 | *to++ = *p++; | |
8f924df7 | 656 | } |
0168c3d8 KH |
657 | else |
658 | { | |
659 | int c = *p++; | |
660 | c = BYTE8_TO_CHAR (c); | |
661 | to += CHAR_STRING (c, to); | |
662 | } | |
663 | chars++; | |
664 | } | |
665 | if (nchars) | |
666 | *nchars = chars; | |
667 | return (to - str); | |
668 | } | |
669 | ||
670 | /* Parse unibyte string at STR of LEN bytes, and return the number of | |
e1dbe924 | 671 | bytes it may occupy when converted to multibyte string by |
0168c3d8 KH |
672 | `str_to_multibyte'. */ |
673 | ||
d311d28c PE |
674 | ptrdiff_t |
675 | count_size_as_multibyte (const unsigned char *str, ptrdiff_t len) | |
0168c3d8 | 676 | { |
eec47d6b | 677 | const unsigned char *endp = str + len; |
d311d28c | 678 | ptrdiff_t bytes; |
0168c3d8 KH |
679 | |
680 | for (bytes = 0; str < endp; str++) | |
de883a70 PE |
681 | { |
682 | int n = *str < 0x80 ? 1 : 2; | |
683 | if (INT_ADD_OVERFLOW (bytes, n)) | |
684 | string_overflow (); | |
685 | bytes += n; | |
686 | } | |
0168c3d8 KH |
687 | return bytes; |
688 | } | |
689 | ||
690 | ||
ef3ff036 | 691 | /* Convert unibyte text at STR of BYTES bytes to a multibyte text |
0168c3d8 KH |
692 | that contains the same single-byte characters. It actually |
693 | converts all 8-bit characters to multibyte forms. It is assured | |
694 | that we can use LEN bytes at STR as a work area and that is | |
695 | enough. */ | |
696 | ||
d311d28c PE |
697 | ptrdiff_t |
698 | str_to_multibyte (unsigned char *str, ptrdiff_t len, ptrdiff_t bytes) | |
0168c3d8 KH |
699 | { |
700 | unsigned char *p = str, *endp = str + bytes; | |
701 | unsigned char *to; | |
702 | ||
703 | while (p < endp && *p < 0x80) p++; | |
704 | if (p == endp) | |
705 | return bytes; | |
706 | to = p; | |
707 | bytes = endp - p; | |
708 | endp = str + len; | |
72af86bd | 709 | memmove (endp - bytes, p, bytes); |
0168c3d8 | 710 | p = endp - bytes; |
8f924df7 | 711 | while (p < endp) |
0168c3d8 KH |
712 | { |
713 | int c = *p++; | |
714 | ||
715 | if (c >= 0x80) | |
716 | c = BYTE8_TO_CHAR (c); | |
717 | to += CHAR_STRING (c, to); | |
718 | } | |
719 | return (to - str); | |
720 | } | |
721 | ||
722 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
723 | actually converts characters in the range 0x80..0xFF to | |
724 | unibyte. */ | |
725 | ||
d311d28c PE |
726 | ptrdiff_t |
727 | str_as_unibyte (unsigned char *str, ptrdiff_t bytes) | |
0168c3d8 | 728 | { |
15843e6f KH |
729 | const unsigned char *p = str, *endp = str + bytes; |
730 | unsigned char *to; | |
0168c3d8 KH |
731 | int c, len; |
732 | ||
733 | while (p < endp) | |
734 | { | |
735 | c = *p; | |
736 | len = BYTES_BY_CHAR_HEAD (c); | |
737 | if (CHAR_BYTE8_HEAD_P (c)) | |
738 | break; | |
739 | p += len; | |
740 | } | |
15843e6f | 741 | to = str + (p - str); |
8f924df7 | 742 | while (p < endp) |
0168c3d8 KH |
743 | { |
744 | c = *p; | |
745 | len = BYTES_BY_CHAR_HEAD (c); | |
746 | if (CHAR_BYTE8_HEAD_P (c)) | |
747 | { | |
748 | c = STRING_CHAR_ADVANCE (p); | |
749 | *to++ = CHAR_TO_BYTE8 (c); | |
750 | } | |
751 | else | |
752 | { | |
753 | while (len--) *to++ = *p++; | |
754 | } | |
755 | } | |
756 | return (to - str); | |
757 | } | |
758 | ||
4aa40bb8 KH |
759 | /* Convert eight-bit chars in SRC (in multibyte form) to the |
760 | corresponding byte and store in DST. CHARS is the number of | |
761 | characters in SRC. The value is the number of bytes stored in DST. | |
762 | Usually, the value is the same as CHARS, but is less than it if SRC | |
d5172d4f | 763 | contains a non-ASCII, non-eight-bit character. */ |
4aa40bb8 | 764 | |
d311d28c | 765 | ptrdiff_t |
d5172d4f | 766 | str_to_unibyte (const unsigned char *src, unsigned char *dst, ptrdiff_t chars) |
4aa40bb8 | 767 | { |
d311d28c | 768 | ptrdiff_t i; |
4aa40bb8 KH |
769 | |
770 | for (i = 0; i < chars; i++) | |
771 | { | |
772 | int c = STRING_CHAR_ADVANCE (src); | |
773 | ||
774 | if (CHAR_BYTE8_P (c)) | |
775 | c = CHAR_TO_BYTE8 (c); | |
d5172d4f | 776 | else if (! ASCII_CHAR_P (c)) |
4aa40bb8 KH |
777 | return i; |
778 | *dst++ = c; | |
779 | } | |
780 | return i; | |
781 | } | |
782 | ||
783 | ||
d311d28c | 784 | static ptrdiff_t |
971de7fb | 785 | string_count_byte8 (Lisp_Object string) |
0168c3d8 | 786 | { |
d5172d4f | 787 | bool multibyte = STRING_MULTIBYTE (string); |
d311d28c | 788 | ptrdiff_t nbytes = SBYTES (string); |
8f924df7 | 789 | unsigned char *p = SDATA (string); |
0168c3d8 | 790 | unsigned char *pend = p + nbytes; |
d311d28c | 791 | ptrdiff_t count = 0; |
0168c3d8 KH |
792 | int c, len; |
793 | ||
794 | if (multibyte) | |
795 | while (p < pend) | |
796 | { | |
797 | c = *p; | |
798 | len = BYTES_BY_CHAR_HEAD (c); | |
799 | ||
800 | if (CHAR_BYTE8_HEAD_P (c)) | |
801 | count++; | |
802 | p += len; | |
803 | } | |
804 | else | |
805 | while (p < pend) | |
806 | { | |
807 | if (*p++ >= 0x80) | |
808 | count++; | |
809 | } | |
810 | return count; | |
811 | } | |
812 | ||
813 | ||
814 | Lisp_Object | |
971de7fb | 815 | string_escape_byte8 (Lisp_Object string) |
0168c3d8 | 816 | { |
d311d28c PE |
817 | ptrdiff_t nchars = SCHARS (string); |
818 | ptrdiff_t nbytes = SBYTES (string); | |
d5172d4f | 819 | bool multibyte = STRING_MULTIBYTE (string); |
d311d28c | 820 | ptrdiff_t byte8_count; |
15843e6f KH |
821 | const unsigned char *src, *src_end; |
822 | unsigned char *dst; | |
0168c3d8 KH |
823 | Lisp_Object val; |
824 | int c, len; | |
825 | ||
826 | if (multibyte && nchars == nbytes) | |
827 | return string; | |
828 | ||
829 | byte8_count = string_count_byte8 (string); | |
830 | ||
831 | if (byte8_count == 0) | |
832 | return string; | |
833 | ||
834 | if (multibyte) | |
14162469 EZ |
835 | { |
836 | if ((MOST_POSITIVE_FIXNUM - nchars) / 3 < byte8_count | |
c9d624c6 | 837 | || (STRING_BYTES_BOUND - nbytes) / 2 < byte8_count) |
cb93f9be | 838 | string_overflow (); |
14162469 EZ |
839 | |
840 | /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
841 | val = make_uninit_multibyte_string (nchars + byte8_count * 3, | |
842 | nbytes + byte8_count * 2); | |
843 | } | |
0168c3d8 | 844 | else |
14162469 | 845 | { |
c9d624c6 | 846 | if ((STRING_BYTES_BOUND - nbytes) / 3 < byte8_count) |
cb93f9be | 847 | string_overflow (); |
ef3ff036 | 848 | |
14162469 EZ |
849 | /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ |
850 | val = make_uninit_string (nbytes + byte8_count * 3); | |
851 | } | |
0168c3d8 | 852 | |
8f924df7 | 853 | src = SDATA (string); |
0168c3d8 | 854 | src_end = src + nbytes; |
8f924df7 | 855 | dst = SDATA (val); |
0168c3d8 KH |
856 | if (multibyte) |
857 | while (src < src_end) | |
858 | { | |
859 | c = *src; | |
860 | len = BYTES_BY_CHAR_HEAD (c); | |
861 | ||
862 | if (CHAR_BYTE8_HEAD_P (c)) | |
863 | { | |
864 | c = STRING_CHAR_ADVANCE (src); | |
865 | c = CHAR_TO_BYTE8 (c); | |
99027bdd | 866 | dst += sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
867 | } |
868 | else | |
869 | while (len--) *dst++ = *src++; | |
870 | } | |
871 | else | |
872 | while (src < src_end) | |
873 | { | |
874 | c = *src++; | |
875 | if (c >= 0x80) | |
99027bdd | 876 | dst += sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
877 | else |
878 | *dst++ = c; | |
879 | } | |
880 | return val; | |
881 | } | |
882 | ||
883 | \f | |
a7ca3326 | 884 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 885 | doc: /* |
d2e83296 DL |
886 | Concatenate all the argument characters and make the result a string. |
887 | usage: (string &rest CHARACTERS) */) | |
f66c7cf8 | 888 | (ptrdiff_t n, Lisp_Object *args) |
0168c3d8 | 889 | { |
f66c7cf8 | 890 | ptrdiff_t i; |
c5101a77 | 891 | int c; |
754790b6 CY |
892 | unsigned char *buf, *p; |
893 | Lisp_Object str; | |
894 | USE_SAFE_ALLOCA; | |
895 | ||
0065d054 | 896 | SAFE_NALLOCA (buf, MAX_MULTIBYTE_LENGTH, n); |
754790b6 | 897 | p = buf; |
0168c3d8 KH |
898 | |
899 | for (i = 0; i < n; i++) | |
900 | { | |
901 | CHECK_CHARACTER (args[i]); | |
902 | c = XINT (args[i]); | |
903 | p += CHAR_STRING (c, p); | |
904 | } | |
905 | ||
754790b6 CY |
906 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
907 | SAFE_FREE (); | |
908 | return str; | |
0168c3d8 KH |
909 | } |
910 | ||
70b4969d | 911 | DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0, |
87d6f965 KH |
912 | doc: /* Concatenate all the argument bytes and make the result a unibyte string. |
913 | usage: (unibyte-string &rest BYTES) */) | |
f66c7cf8 | 914 | (ptrdiff_t n, Lisp_Object *args) |
70b4969d | 915 | { |
f66c7cf8 | 916 | ptrdiff_t i; |
754790b6 CY |
917 | Lisp_Object str; |
918 | USE_SAFE_ALLOCA; | |
98c6f1e3 PE |
919 | unsigned char *buf = SAFE_ALLOCA (n); |
920 | unsigned char *p = buf; | |
70b4969d KH |
921 | |
922 | for (i = 0; i < n; i++) | |
923 | { | |
af5a5a98 | 924 | CHECK_RANGED_INTEGER (args[i], 0, 255); |
a14e1568 | 925 | *p++ = XINT (args[i]); |
70b4969d KH |
926 | } |
927 | ||
754790b6 CY |
928 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
929 | SAFE_FREE (); | |
930 | return str; | |
70b4969d KH |
931 | } |
932 | ||
c73ae4ae | 933 | DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers, |
d0363d44 KH |
934 | Schar_resolve_modifiers, 1, 1, 0, |
935 | doc: /* Resolve modifiers in the character CHAR. | |
936 | The value is a character with modifiers resolved into the character | |
937 | code. Unresolved modifiers are kept in the value. | |
c73ae4ae | 938 | usage: (char-resolve-modifiers CHAR) */) |
5842a27b | 939 | (Lisp_Object character) |
d0363d44 | 940 | { |
d311d28c | 941 | EMACS_INT c; |
d0363d44 KH |
942 | |
943 | CHECK_NUMBER (character); | |
944 | c = XINT (character); | |
945 | return make_number (char_resolve_modifier_mask (c)); | |
946 | } | |
947 | ||
ee107a89 KH |
948 | DEFUN ("get-byte", Fget_byte, Sget_byte, 0, 2, 0, |
949 | doc: /* Return a byte value of a character at point. | |
950 | Optional 1st arg POSITION, if non-nil, is a position of a character to get | |
951 | a byte value. | |
952 | Optional 2nd arg STRING, if non-nil, is a string of which first | |
953 | character is a target to get a byte value. In this case, POSITION, if | |
954 | non-nil, is an index of a target character in the string. | |
955 | ||
956 | If the current buffer (or STRING) is multibyte, and the target | |
8350f087 | 957 | character is not ASCII nor 8-bit character, an error is signaled. */) |
5842a27b | 958 | (Lisp_Object position, Lisp_Object string) |
ee107a89 KH |
959 | { |
960 | int c; | |
d311d28c | 961 | ptrdiff_t pos; |
ee107a89 KH |
962 | unsigned char *p; |
963 | ||
964 | if (NILP (string)) | |
965 | { | |
966 | if (NILP (position)) | |
967 | { | |
968 | p = PT_ADDR; | |
4d8e170e | 969 | } |
ee107a89 KH |
970 | else |
971 | { | |
972 | CHECK_NUMBER_COERCE_MARKER (position); | |
973 | if (XINT (position) < BEGV || XINT (position) >= ZV) | |
974 | args_out_of_range_3 (position, make_number (BEGV), make_number (ZV)); | |
975 | pos = XFASTINT (position); | |
976 | p = CHAR_POS_ADDR (pos); | |
977 | } | |
4b4deea2 | 978 | if (NILP (BVAR (current_buffer, enable_multibyte_characters))) |
d5998e03 | 979 | return make_number (*p); |
ee107a89 KH |
980 | } |
981 | else | |
982 | { | |
983 | CHECK_STRING (string); | |
984 | if (NILP (position)) | |
985 | { | |
986 | p = SDATA (string); | |
987 | } | |
988 | else | |
989 | { | |
990 | CHECK_NATNUM (position); | |
991 | if (XINT (position) >= SCHARS (string)) | |
992 | args_out_of_range (string, position); | |
993 | pos = XFASTINT (position); | |
994 | p = SDATA (string) + string_char_to_byte (string, pos); | |
995 | } | |
d5998e03 KH |
996 | if (! STRING_MULTIBYTE (string)) |
997 | return make_number (*p); | |
ee107a89 | 998 | } |
62a6e103 | 999 | c = STRING_CHAR (p); |
ee107a89 KH |
1000 | if (CHAR_BYTE8_P (c)) |
1001 | c = CHAR_TO_BYTE8 (c); | |
1002 | else if (! ASCII_CHAR_P (c)) | |
1003 | error ("Not an ASCII nor an 8-bit character: %d", c); | |
1004 | return make_number (c); | |
1005 | } | |
1006 | ||
0168c3d8 KH |
1007 | #ifdef emacs |
1008 | ||
1009 | void | |
971de7fb | 1010 | syms_of_character (void) |
0168c3d8 KH |
1011 | { |
1012 | DEFSYM (Qcharacterp, "characterp"); | |
1013 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
1014 | ||
1015 | staticpro (&Vchar_unify_table); | |
1016 | Vchar_unify_table = Qnil; | |
1017 | ||
1018 | defsubr (&Smax_char); | |
1019 | defsubr (&Scharacterp); | |
1020 | defsubr (&Sunibyte_char_to_multibyte); | |
1021 | defsubr (&Smultibyte_char_to_unibyte); | |
0168c3d8 KH |
1022 | defsubr (&Schar_width); |
1023 | defsubr (&Sstring_width); | |
0168c3d8 | 1024 | defsubr (&Sstring); |
70b4969d | 1025 | defsubr (&Sunibyte_string); |
d0363d44 | 1026 | defsubr (&Schar_resolve_modifiers); |
ee107a89 | 1027 | defsubr (&Sget_byte); |
0168c3d8 | 1028 | |
29208e82 | 1029 | DEFVAR_LISP ("translation-table-vector", Vtranslation_table_vector, |
0168c3d8 | 1030 | doc: /* |
68978cf0 DL |
1031 | Vector recording all translation tables ever defined. |
1032 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
1033 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
1034 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
1035 | ||
29208e82 | 1036 | DEFVAR_LISP ("auto-fill-chars", Vauto_fill_chars, |
0168c3d8 KH |
1037 | doc: /* |
1038 | A char-table for characters which invoke auto-filling. | |
1039 | Such characters have value t in this table. */); | |
1040 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
1041 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
1042 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 | 1043 | |
29208e82 | 1044 | DEFVAR_LISP ("char-width-table", Vchar_width_table, |
0168c3d8 KH |
1045 | doc: /* |
1046 | A char-table for width (columns) of each character. */); | |
1047 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
1048 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
1049 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
1050 | make_number (4)); | |
0168c3d8 | 1051 | |
29208e82 | 1052 | DEFVAR_LISP ("printable-chars", Vprintable_chars, |
0168c3d8 | 1053 | doc: /* A char-table for each printable character. */); |
db6d4189 | 1054 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
1055 | Fset_char_table_range (Vprintable_chars, |
1056 | Fcons (make_number (32), make_number (126)), Qt); | |
1057 | Fset_char_table_range (Vprintable_chars, | |
1058 | Fcons (make_number (160), | |
1059 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 1060 | |
29208e82 | 1061 | DEFVAR_LISP ("char-script-table", Vchar_script_table, |
c57f3328 KH |
1062 | doc: /* Char table of script symbols. |
1063 | It has one extra slot whose value is a list of script symbols. */); | |
1064 | ||
1065 | /* Intern this now in case it isn't already done. | |
1066 | Setting this variable twice is harmless. | |
1067 | But don't staticpro it here--that is done in alloc.c. */ | |
d67b4f80 | 1068 | Qchar_table_extra_slots = intern_c_string ("char-table-extra-slots"); |
c57f3328 KH |
1069 | DEFSYM (Qchar_script_table, "char-script-table"); |
1070 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
1071 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
c7e14352 | 1072 | |
29208e82 | 1073 | DEFVAR_LISP ("script-representative-chars", Vscript_representative_chars, |
c3bb7671 | 1074 | doc: /* Alist of scripts vs the representative characters. |
f4427a54 | 1075 | Each element is a cons (SCRIPT . CHARS). |
4d8e170e | 1076 | SCRIPT is a symbol representing a script or a subgroup of a script. |
c3bb7671 | 1077 | CHARS is a list or a vector of characters. |
472a4dc9 | 1078 | If it is a list, all characters in the list are necessary for supporting SCRIPT. |
c3bb7671 KH |
1079 | If it is a vector, one of the characters in the vector is necessary. |
1080 | This variable is used to find a font for a specific script. */); | |
c7e14352 | 1081 | Vscript_representative_chars = Qnil; |
a3cbb631 | 1082 | |
29208e82 | 1083 | DEFVAR_LISP ("unicode-category-table", Vunicode_category_table, |
a3cbb631 | 1084 | doc: /* Char table of Unicode's "General Category". |
472a4dc9 JB |
1085 | All Unicode characters have one of the following values (symbol): |
1086 | Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po, | |
a3cbb631 KH |
1087 | Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn |
1088 | See The Unicode Standard for the meaning of those values. */); | |
1089 | /* The correct char-table is setup in characters.el. */ | |
1090 | Vunicode_category_table = Qnil; | |
0168c3d8 KH |
1091 | } |
1092 | ||
1093 | #endif /* emacs */ |