Commit | Line | Data |
---|---|---|
0168c3d8 | 1 | /* Basic character support. |
73b0cd50 | 2 | |
acaf905b | 3 | Copyright (C) 2001-2012 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 | ||
0168c3d8 KH |
129 | MAYBE_UNIFY_CHAR (c); |
130 | ||
e3d8eb8c | 131 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
132 | { |
133 | bytes = CHAR_STRING (c, p); | |
134 | } | |
135 | else if (c <= MAX_4_BYTE_CHAR) | |
136 | { | |
137 | p[0] = (0xF0 | (c >> 18)); | |
138 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
139 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
140 | p[3] = (0x80 | (c & 0x3F)); | |
141 | bytes = 4; | |
142 | } | |
e3d8eb8c | 143 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
144 | { |
145 | p[0] = 0xF8; | |
146 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
147 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
148 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
149 | p[4] = (0x80 | (c & 0x3F)); | |
150 | bytes = 5; | |
151 | } | |
5aa91c9b | 152 | else if (c <= MAX_CHAR) |
e3d8eb8c KH |
153 | { |
154 | c = CHAR_TO_BYTE8 (c); | |
155 | bytes = BYTE8_STRING (c, p); | |
156 | } | |
5aa91c9b | 157 | else |
e6c3da20 | 158 | error ("Invalid character: %x", c); |
1889b238 | 159 | |
0168c3d8 KH |
160 | return bytes; |
161 | } | |
162 | ||
163 | ||
224a3131 | 164 | /* Return a character whose multibyte form is at P. If LEN is not |
33f91981 | 165 | NULL, it must be a pointer to integer. In that case, set *LEN to |
224a3131 | 166 | the byte length of the multibyte form. If ADVANCED is not NULL, it |
33f91981 | 167 | must be a pointer to unsigned char. In that case, set *ADVANCED to |
224a3131 | 168 | the ending address (i.e., the starting address of the next |
33f91981 KH |
169 | character) of the multibyte form. */ |
170 | ||
0168c3d8 | 171 | int |
971de7fb | 172 | string_char (const unsigned char *p, const unsigned char **advanced, int *len) |
0168c3d8 | 173 | { |
1889b238 | 174 | int c; |
15843e6f | 175 | const unsigned char *saved_p = p; |
0168c3d8 KH |
176 | |
177 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
178 | { | |
179 | c = STRING_CHAR_ADVANCE (p); | |
180 | } | |
181 | else if (! (*p & 0x08)) | |
182 | { | |
183 | c = ((((p)[0] & 0xF) << 18) | |
184 | | (((p)[1] & 0x3F) << 12) | |
185 | | (((p)[2] & 0x3F) << 6) | |
186 | | ((p)[3] & 0x3F)); | |
187 | p += 4; | |
188 | } | |
189 | else | |
190 | { | |
191 | c = ((((p)[1] & 0x3F) << 18) | |
192 | | (((p)[2] & 0x3F) << 12) | |
193 | | (((p)[3] & 0x3F) << 6) | |
194 | | ((p)[4] & 0x3F)); | |
195 | p += 5; | |
196 | } | |
197 | ||
198 | MAYBE_UNIFY_CHAR (c); | |
199 | ||
200 | if (len) | |
201 | *len = p - saved_p; | |
202 | if (advanced) | |
203 | *advanced = p; | |
204 | return c; | |
205 | } | |
206 | ||
207 | ||
224a3131 EZ |
208 | /* Translate character C by translation table TABLE. If no translation is |
209 | found in TABLE, return the untranslated character. If TABLE is a list, | |
210 | elements are char tables. In that case, recursively translate C by all the | |
211 | tables in the list. */ | |
0168c3d8 KH |
212 | |
213 | int | |
971de7fb | 214 | translate_char (Lisp_Object table, int c) |
0168c3d8 | 215 | { |
10453be9 KH |
216 | if (CHAR_TABLE_P (table)) |
217 | { | |
218 | Lisp_Object ch; | |
219 | ||
220 | ch = CHAR_TABLE_REF (table, c); | |
221 | if (CHARACTERP (ch)) | |
222 | c = XINT (ch); | |
223 | } | |
224 | else | |
225 | { | |
226 | for (; CONSP (table); table = XCDR (table)) | |
227 | c = translate_char (XCAR (table), c); | |
228 | } | |
229 | return c; | |
0168c3d8 KH |
230 | } |
231 | ||
2e5db15c | 232 | /* Convert ASCII or 8-bit character C to unibyte. If C is none of |
461c2ab9 | 233 | them, return (C & 0xFF). */ |
0168c3d8 KH |
234 | |
235 | int | |
461c2ab9 | 236 | multibyte_char_to_unibyte (int c) |
0168c3d8 | 237 | { |
2e5db15c KH |
238 | if (c < 0x80) |
239 | return c; | |
b672c5ae KH |
240 | if (CHAR_BYTE8_P (c)) |
241 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 242 | return (c & 0xFF); |
0168c3d8 KH |
243 | } |
244 | ||
935d5b02 KH |
245 | /* Like multibyte_char_to_unibyte, but return -1 if C is not supported |
246 | by charset_unibyte. */ | |
247 | ||
248 | int | |
971de7fb | 249 | multibyte_char_to_unibyte_safe (int c) |
935d5b02 | 250 | { |
2e5db15c KH |
251 | if (c < 0x80) |
252 | return c; | |
935d5b02 KH |
253 | if (CHAR_BYTE8_P (c)) |
254 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 255 | return -1; |
935d5b02 | 256 | } |
0168c3d8 KH |
257 | |
258 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
bc985c87 | 259 | doc: /* Return non-nil if OBJECT is a character. |
4abcdac8 CY |
260 | In Emacs Lisp, characters are represented by character codes, which |
261 | are non-negative integers. The function `max-char' returns the | |
262 | maximum character code. | |
bc985c87 | 263 | usage: (characterp OBJECT) */) |
5842a27b | 264 | (Lisp_Object object, Lisp_Object ignore) |
0168c3d8 KH |
265 | { |
266 | return (CHARACTERP (object) ? Qt : Qnil); | |
267 | } | |
268 | ||
269 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
270 | doc: /* Return the character of the maximum code. */) | |
5842a27b | 271 | (void) |
0168c3d8 KH |
272 | { |
273 | return make_number (MAX_CHAR); | |
274 | } | |
275 | ||
276 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
277 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
5556875b | 278 | doc: /* Convert the byte CH to multibyte character. */) |
5842a27b | 279 | (Lisp_Object ch) |
0168c3d8 KH |
280 | { |
281 | int c; | |
0168c3d8 KH |
282 | |
283 | CHECK_CHARACTER (ch); | |
284 | c = XFASTINT (ch); | |
2e5db15c KH |
285 | if (c >= 0x100) |
286 | error ("Not a unibyte character: %d", c); | |
4c0354d7 | 287 | MAKE_CHAR_MULTIBYTE (c); |
0168c3d8 KH |
288 | return make_number (c); |
289 | } | |
290 | ||
291 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
292 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
5556875b SM |
293 | doc: /* Convert the multibyte character CH to a byte. |
294 | If the multibyte character does not represent a byte, return -1. */) | |
5842a27b | 295 | (Lisp_Object ch) |
0168c3d8 | 296 | { |
5556875b | 297 | int cm; |
0168c3d8 KH |
298 | |
299 | CHECK_CHARACTER (ch); | |
5556875b SM |
300 | cm = XFASTINT (ch); |
301 | if (cm < 256) | |
302 | /* Can't distinguish a byte read from a unibyte buffer from | |
303 | a latin1 char, so let's let it slide. */ | |
304 | return ch; | |
305 | else | |
306 | { | |
2afc21f5 | 307 | int cu = CHAR_TO_BYTE_SAFE (cm); |
5556875b SM |
308 | return make_number (cu); |
309 | } | |
0168c3d8 KH |
310 | } |
311 | ||
25ed9e61 KH |
312 | |
313 | /* Return width (columns) of C considering the buffer display table DP. */ | |
314 | ||
6e6c82a4 | 315 | static ptrdiff_t |
25ed9e61 KH |
316 | char_width (int c, struct Lisp_Char_Table *dp) |
317 | { | |
6e6c82a4 | 318 | ptrdiff_t width = CHAR_WIDTH (c); |
25ed9e61 KH |
319 | |
320 | if (dp) | |
321 | { | |
322 | Lisp_Object disp = DISP_CHAR_VECTOR (dp, c), ch; | |
323 | int i; | |
324 | ||
325 | if (VECTORP (disp)) | |
326 | for (i = 0, width = 0; i < ASIZE (disp); i++) | |
327 | { | |
328 | ch = AREF (disp, i); | |
329 | if (CHARACTERP (ch)) | |
df0b2940 PE |
330 | { |
331 | int w = CHAR_WIDTH (XFASTINT (ch)); | |
332 | if (INT_ADD_OVERFLOW (width, w)) | |
333 | string_overflow (); | |
334 | width += w; | |
335 | } | |
25ed9e61 KH |
336 | } |
337 | } | |
338 | return width; | |
339 | } | |
340 | ||
341 | ||
a7ca3326 | 342 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, |
0168c3d8 KH |
343 | doc: /* Return width of CHAR when displayed in the current buffer. |
344 | The width is measured by how many columns it occupies on the screen. | |
add553ac JB |
345 | Tab is taken to occupy `tab-width' columns. |
346 | usage: (char-width CHAR) */) | |
5842a27b | 347 | (Lisp_Object ch) |
0168c3d8 | 348 | { |
df0b2940 | 349 | int c; |
6e6c82a4 | 350 | ptrdiff_t width; |
0168c3d8 KH |
351 | |
352 | CHECK_CHARACTER (ch); | |
353 | c = XINT (ch); | |
25ed9e61 | 354 | width = char_width (c, buffer_display_table ()); |
0168c3d8 KH |
355 | return make_number (width); |
356 | } | |
357 | ||
0168c3d8 KH |
358 | /* Return width of string STR of length LEN when displayed in the |
359 | current buffer. The width is measured by how many columns it | |
360 | occupies on the screen. If PRECISION > 0, return the width of | |
361 | longest substring that doesn't exceed PRECISION, and set number of | |
362 | characters and bytes of the substring in *NCHARS and *NBYTES | |
363 | respectively. */ | |
364 | ||
d311d28c PE |
365 | ptrdiff_t |
366 | c_string_width (const unsigned char *str, ptrdiff_t len, int precision, | |
367 | ptrdiff_t *nchars, ptrdiff_t *nbytes) | |
0168c3d8 | 368 | { |
d311d28c PE |
369 | ptrdiff_t i = 0, i_byte = 0; |
370 | ptrdiff_t width = 0; | |
0168c3d8 KH |
371 | struct Lisp_Char_Table *dp = buffer_display_table (); |
372 | ||
373 | while (i_byte < len) | |
374 | { | |
25ed9e61 | 375 | int bytes; |
62a6e103 | 376 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes); |
6e6c82a4 | 377 | ptrdiff_t thiswidth = char_width (c, dp); |
0168c3d8 | 378 | |
df0b2940 | 379 | if (precision <= 0) |
0168c3d8 | 380 | { |
df0b2940 PE |
381 | if (INT_ADD_OVERFLOW (width, thiswidth)) |
382 | string_overflow (); | |
0168c3d8 | 383 | } |
df0b2940 | 384 | else if (precision - width < thiswidth) |
0168c3d8 KH |
385 | { |
386 | *nchars = i; | |
387 | *nbytes = i_byte; | |
388 | return width; | |
389 | } | |
390 | i++; | |
391 | i_byte += bytes; | |
392 | width += thiswidth; | |
393 | } | |
394 | ||
395 | if (precision > 0) | |
396 | { | |
397 | *nchars = i; | |
398 | *nbytes = i_byte; | |
399 | } | |
400 | ||
401 | return width; | |
402 | } | |
403 | ||
1889b238 KH |
404 | /* Return width of string STR of length LEN when displayed in the |
405 | current buffer. The width is measured by how many columns it | |
406 | occupies on the screen. */ | |
407 | ||
d311d28c PE |
408 | ptrdiff_t |
409 | strwidth (const char *str, ptrdiff_t len) | |
1889b238 | 410 | { |
7469ef5d | 411 | return c_string_width ((const unsigned char *) str, len, -1, NULL, NULL); |
1889b238 KH |
412 | } |
413 | ||
0168c3d8 KH |
414 | /* Return width of Lisp string STRING when displayed in the current |
415 | buffer. The width is measured by how many columns it occupies on | |
416 | the screen while paying attention to compositions. If PRECISION > | |
417 | 0, return the width of longest substring that doesn't exceed | |
418 | PRECISION, and set number of characters and bytes of the substring | |
419 | in *NCHARS and *NBYTES respectively. */ | |
420 | ||
d311d28c PE |
421 | ptrdiff_t |
422 | lisp_string_width (Lisp_Object string, ptrdiff_t precision, | |
423 | ptrdiff_t *nchars, ptrdiff_t *nbytes) | |
0168c3d8 | 424 | { |
d311d28c | 425 | ptrdiff_t len = SCHARS (string); |
0aee65b9 KH |
426 | /* This set multibyte to 0 even if STRING is multibyte when it |
427 | contains only ascii and eight-bit-graphic, but that's | |
428 | intentional. */ | |
d5172d4f | 429 | bool multibyte = len < SBYTES (string); |
8f924df7 | 430 | unsigned char *str = SDATA (string); |
d311d28c PE |
431 | ptrdiff_t i = 0, i_byte = 0; |
432 | ptrdiff_t width = 0; | |
0168c3d8 KH |
433 | struct Lisp_Char_Table *dp = buffer_display_table (); |
434 | ||
435 | while (i < len) | |
436 | { | |
d311d28c | 437 | ptrdiff_t chars, bytes, thiswidth; |
0168c3d8 | 438 | Lisp_Object val; |
ebfa62c0 | 439 | ptrdiff_t cmp_id; |
d311d28c | 440 | ptrdiff_t ignore, end; |
0168c3d8 KH |
441 | |
442 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
443 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
444 | >= 0)) | |
445 | { | |
446 | thiswidth = composition_table[cmp_id]->width; | |
447 | chars = end - i; | |
448 | bytes = string_char_to_byte (string, end) - i_byte; | |
449 | } | |
0168c3d8 KH |
450 | else |
451 | { | |
0aee65b9 | 452 | int c; |
0168c3d8 | 453 | |
0aee65b9 | 454 | if (multibyte) |
2b4560a8 PE |
455 | { |
456 | int cbytes; | |
457 | c = STRING_CHAR_AND_LENGTH (str + i_byte, cbytes); | |
458 | bytes = cbytes; | |
459 | } | |
0aee65b9 KH |
460 | else |
461 | c = str[i_byte], bytes = 1; | |
0168c3d8 | 462 | chars = 1; |
25ed9e61 | 463 | thiswidth = char_width (c, dp); |
0168c3d8 KH |
464 | } |
465 | ||
2b4560a8 PE |
466 | if (precision <= 0) |
467 | { | |
468 | #ifdef emacs | |
469 | if (INT_ADD_OVERFLOW (width, thiswidth)) | |
470 | string_overflow (); | |
471 | #endif | |
472 | } | |
473 | else if (precision - width < thiswidth) | |
0168c3d8 KH |
474 | { |
475 | *nchars = i; | |
476 | *nbytes = i_byte; | |
477 | return width; | |
478 | } | |
479 | i += chars; | |
480 | i_byte += bytes; | |
481 | width += thiswidth; | |
ef3ff036 | 482 | } |
0168c3d8 KH |
483 | |
484 | if (precision > 0) | |
485 | { | |
486 | *nchars = i; | |
487 | *nbytes = i_byte; | |
488 | } | |
489 | ||
490 | return width; | |
491 | } | |
492 | ||
493 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
494 | doc: /* Return width of STRING when displayed in the current buffer. | |
495 | Width is measured by how many columns it occupies on the screen. | |
496 | When calculating width of a multibyte character in STRING, | |
497 | only the base leading-code is considered; the validity of | |
498 | the following bytes is not checked. Tabs in STRING are always | |
add553ac JB |
499 | taken to occupy `tab-width' columns. |
500 | usage: (string-width STRING) */) | |
5842a27b | 501 | (Lisp_Object str) |
0168c3d8 KH |
502 | { |
503 | Lisp_Object val; | |
504 | ||
505 | CHECK_STRING (str); | |
506 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
507 | return val; | |
508 | } | |
509 | ||
0168c3d8 KH |
510 | /* Return the number of characters in the NBYTES bytes at PTR. |
511 | This works by looking at the contents and checking for multibyte | |
512 | sequences while assuming that there's no invalid sequence. | |
513 | However, if the current buffer has enable-multibyte-characters = | |
514 | nil, we treat each byte as a character. */ | |
515 | ||
d311d28c PE |
516 | ptrdiff_t |
517 | chars_in_text (const unsigned char *ptr, ptrdiff_t nbytes) | |
0168c3d8 KH |
518 | { |
519 | /* current_buffer is null at early stages of Emacs initialization. */ | |
520 | if (current_buffer == 0 | |
4b4deea2 | 521 | || NILP (BVAR (current_buffer, enable_multibyte_characters))) |
0168c3d8 KH |
522 | return nbytes; |
523 | ||
524 | return multibyte_chars_in_text (ptr, nbytes); | |
525 | } | |
526 | ||
527 | /* Return the number of characters in the NBYTES bytes at PTR. | |
528 | This works by looking at the contents and checking for multibyte | |
529 | sequences while assuming that there's no invalid sequence. It | |
530 | ignores enable-multibyte-characters. */ | |
531 | ||
d311d28c PE |
532 | ptrdiff_t |
533 | multibyte_chars_in_text (const unsigned char *ptr, ptrdiff_t nbytes) | |
0168c3d8 | 534 | { |
8f924df7 | 535 | const unsigned char *endp = ptr + nbytes; |
d311d28c | 536 | ptrdiff_t chars = 0; |
0168c3d8 KH |
537 | |
538 | while (ptr < endp) | |
539 | { | |
d311d28c | 540 | int len = MULTIBYTE_LENGTH (ptr, endp); |
0168c3d8 KH |
541 | |
542 | if (len == 0) | |
1088b922 | 543 | emacs_abort (); |
0168c3d8 KH |
544 | ptr += len; |
545 | chars++; | |
546 | } | |
547 | ||
548 | return chars; | |
549 | } | |
550 | ||
551 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
552 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
553 | respectively. On counting bytes, pay attention to that 8-bit | |
554 | characters not constructing a valid multibyte sequence are | |
555 | represented by 2-byte in a multibyte text. */ | |
556 | ||
557 | void | |
d311d28c PE |
558 | parse_str_as_multibyte (const unsigned char *str, ptrdiff_t len, |
559 | ptrdiff_t *nchars, ptrdiff_t *nbytes) | |
0168c3d8 | 560 | { |
8f924df7 | 561 | const unsigned char *endp = str + len; |
d311d28c PE |
562 | int n; |
563 | ptrdiff_t chars = 0, bytes = 0; | |
0168c3d8 KH |
564 | |
565 | if (len >= MAX_MULTIBYTE_LENGTH) | |
566 | { | |
8f924df7 | 567 | const unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; |
0168c3d8 KH |
568 | while (str < adjusted_endp) |
569 | { | |
fc9a17bc KH |
570 | if (! CHAR_BYTE8_HEAD_P (*str) |
571 | && (n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
0168c3d8 KH |
572 | str += n, bytes += n; |
573 | else | |
574 | str++, bytes += 2; | |
575 | chars++; | |
576 | } | |
577 | } | |
578 | while (str < endp) | |
579 | { | |
fc9a17bc KH |
580 | if (! CHAR_BYTE8_HEAD_P (*str) |
581 | && (n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
0168c3d8 KH |
582 | str += n, bytes += n; |
583 | else | |
584 | str++, bytes += 2; | |
585 | chars++; | |
586 | } | |
587 | ||
588 | *nchars = chars; | |
589 | *nbytes = bytes; | |
590 | return; | |
591 | } | |
592 | ||
593 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
da6062e6 | 594 | It actually converts only such 8-bit characters that don't construct |
0168c3d8 KH |
595 | a multibyte sequence to multibyte forms of Latin-1 characters. If |
596 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
597 | text. It is assured that we can use LEN bytes at STR as a work | |
598 | area and that is enough. Return the number of bytes of the | |
599 | resulting text. */ | |
600 | ||
d311d28c PE |
601 | ptrdiff_t |
602 | str_as_multibyte (unsigned char *str, ptrdiff_t len, ptrdiff_t nbytes, | |
603 | ptrdiff_t *nchars) | |
0168c3d8 KH |
604 | { |
605 | unsigned char *p = str, *endp = str + nbytes; | |
606 | unsigned char *to; | |
d311d28c | 607 | ptrdiff_t chars = 0; |
0168c3d8 KH |
608 | int n; |
609 | ||
610 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
611 | { | |
612 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
613 | while (p < adjusted_endp | |
fc9a17bc | 614 | && ! CHAR_BYTE8_HEAD_P (*p) |
0168c3d8 KH |
615 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) |
616 | p += n, chars++; | |
617 | } | |
fc9a17bc KH |
618 | while (p < endp |
619 | && ! CHAR_BYTE8_HEAD_P (*p) | |
620 | && (n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
0168c3d8 KH |
621 | p += n, chars++; |
622 | if (nchars) | |
623 | *nchars = chars; | |
624 | if (p == endp) | |
625 | return nbytes; | |
626 | ||
627 | to = p; | |
628 | nbytes = endp - p; | |
629 | endp = str + len; | |
72af86bd | 630 | memmove (endp - nbytes, p, nbytes); |
0168c3d8 KH |
631 | p = endp - nbytes; |
632 | ||
633 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
634 | { | |
635 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
636 | while (p < adjusted_endp) | |
637 | { | |
fc9a17bc KH |
638 | if (! CHAR_BYTE8_HEAD_P (*p) |
639 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
0168c3d8 KH |
640 | { |
641 | while (n--) | |
642 | *to++ = *p++; | |
643 | } | |
644 | else | |
645 | { | |
646 | int c = *p++; | |
647 | c = BYTE8_TO_CHAR (c); | |
648 | to += CHAR_STRING (c, to); | |
649 | } | |
650 | } | |
651 | chars++; | |
652 | } | |
653 | while (p < endp) | |
654 | { | |
fc9a17bc KH |
655 | if (! CHAR_BYTE8_HEAD_P (*p) |
656 | && (n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
0168c3d8 KH |
657 | { |
658 | while (n--) | |
659 | *to++ = *p++; | |
8f924df7 | 660 | } |
0168c3d8 KH |
661 | else |
662 | { | |
663 | int c = *p++; | |
664 | c = BYTE8_TO_CHAR (c); | |
665 | to += CHAR_STRING (c, to); | |
666 | } | |
667 | chars++; | |
668 | } | |
669 | if (nchars) | |
670 | *nchars = chars; | |
671 | return (to - str); | |
672 | } | |
673 | ||
674 | /* Parse unibyte string at STR of LEN bytes, and return the number of | |
e1dbe924 | 675 | bytes it may occupy when converted to multibyte string by |
0168c3d8 KH |
676 | `str_to_multibyte'. */ |
677 | ||
d311d28c PE |
678 | ptrdiff_t |
679 | count_size_as_multibyte (const unsigned char *str, ptrdiff_t len) | |
0168c3d8 | 680 | { |
eec47d6b | 681 | const unsigned char *endp = str + len; |
d311d28c | 682 | ptrdiff_t bytes; |
0168c3d8 KH |
683 | |
684 | for (bytes = 0; str < endp; str++) | |
de883a70 PE |
685 | { |
686 | int n = *str < 0x80 ? 1 : 2; | |
687 | if (INT_ADD_OVERFLOW (bytes, n)) | |
688 | string_overflow (); | |
689 | bytes += n; | |
690 | } | |
0168c3d8 KH |
691 | return bytes; |
692 | } | |
693 | ||
694 | ||
ef3ff036 | 695 | /* Convert unibyte text at STR of BYTES bytes to a multibyte text |
0168c3d8 KH |
696 | that contains the same single-byte characters. It actually |
697 | converts all 8-bit characters to multibyte forms. It is assured | |
698 | that we can use LEN bytes at STR as a work area and that is | |
699 | enough. */ | |
700 | ||
d311d28c PE |
701 | ptrdiff_t |
702 | str_to_multibyte (unsigned char *str, ptrdiff_t len, ptrdiff_t bytes) | |
0168c3d8 KH |
703 | { |
704 | unsigned char *p = str, *endp = str + bytes; | |
705 | unsigned char *to; | |
706 | ||
707 | while (p < endp && *p < 0x80) p++; | |
708 | if (p == endp) | |
709 | return bytes; | |
710 | to = p; | |
711 | bytes = endp - p; | |
712 | endp = str + len; | |
72af86bd | 713 | memmove (endp - bytes, p, bytes); |
0168c3d8 | 714 | p = endp - bytes; |
8f924df7 | 715 | while (p < endp) |
0168c3d8 KH |
716 | { |
717 | int c = *p++; | |
718 | ||
719 | if (c >= 0x80) | |
720 | c = BYTE8_TO_CHAR (c); | |
721 | to += CHAR_STRING (c, to); | |
722 | } | |
723 | return (to - str); | |
724 | } | |
725 | ||
726 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
727 | actually converts characters in the range 0x80..0xFF to | |
728 | unibyte. */ | |
729 | ||
d311d28c PE |
730 | ptrdiff_t |
731 | str_as_unibyte (unsigned char *str, ptrdiff_t bytes) | |
0168c3d8 | 732 | { |
15843e6f KH |
733 | const unsigned char *p = str, *endp = str + bytes; |
734 | unsigned char *to; | |
0168c3d8 KH |
735 | int c, len; |
736 | ||
737 | while (p < endp) | |
738 | { | |
739 | c = *p; | |
740 | len = BYTES_BY_CHAR_HEAD (c); | |
741 | if (CHAR_BYTE8_HEAD_P (c)) | |
742 | break; | |
743 | p += len; | |
744 | } | |
15843e6f | 745 | to = str + (p - str); |
8f924df7 | 746 | while (p < endp) |
0168c3d8 KH |
747 | { |
748 | c = *p; | |
749 | len = BYTES_BY_CHAR_HEAD (c); | |
750 | if (CHAR_BYTE8_HEAD_P (c)) | |
751 | { | |
752 | c = STRING_CHAR_ADVANCE (p); | |
753 | *to++ = CHAR_TO_BYTE8 (c); | |
754 | } | |
755 | else | |
756 | { | |
757 | while (len--) *to++ = *p++; | |
758 | } | |
759 | } | |
760 | return (to - str); | |
761 | } | |
762 | ||
4aa40bb8 KH |
763 | /* Convert eight-bit chars in SRC (in multibyte form) to the |
764 | corresponding byte and store in DST. CHARS is the number of | |
765 | characters in SRC. The value is the number of bytes stored in DST. | |
766 | Usually, the value is the same as CHARS, but is less than it if SRC | |
d5172d4f | 767 | contains a non-ASCII, non-eight-bit character. */ |
4aa40bb8 | 768 | |
d311d28c | 769 | ptrdiff_t |
d5172d4f | 770 | str_to_unibyte (const unsigned char *src, unsigned char *dst, ptrdiff_t chars) |
4aa40bb8 | 771 | { |
d311d28c | 772 | ptrdiff_t i; |
4aa40bb8 KH |
773 | |
774 | for (i = 0; i < chars; i++) | |
775 | { | |
776 | int c = STRING_CHAR_ADVANCE (src); | |
777 | ||
778 | if (CHAR_BYTE8_P (c)) | |
779 | c = CHAR_TO_BYTE8 (c); | |
d5172d4f | 780 | else if (! ASCII_CHAR_P (c)) |
4aa40bb8 KH |
781 | return i; |
782 | *dst++ = c; | |
783 | } | |
784 | return i; | |
785 | } | |
786 | ||
787 | ||
d311d28c | 788 | static ptrdiff_t |
971de7fb | 789 | string_count_byte8 (Lisp_Object string) |
0168c3d8 | 790 | { |
d5172d4f | 791 | bool multibyte = STRING_MULTIBYTE (string); |
d311d28c | 792 | ptrdiff_t nbytes = SBYTES (string); |
8f924df7 | 793 | unsigned char *p = SDATA (string); |
0168c3d8 | 794 | unsigned char *pend = p + nbytes; |
d311d28c | 795 | ptrdiff_t count = 0; |
0168c3d8 KH |
796 | int c, len; |
797 | ||
798 | if (multibyte) | |
799 | while (p < pend) | |
800 | { | |
801 | c = *p; | |
802 | len = BYTES_BY_CHAR_HEAD (c); | |
803 | ||
804 | if (CHAR_BYTE8_HEAD_P (c)) | |
805 | count++; | |
806 | p += len; | |
807 | } | |
808 | else | |
809 | while (p < pend) | |
810 | { | |
811 | if (*p++ >= 0x80) | |
812 | count++; | |
813 | } | |
814 | return count; | |
815 | } | |
816 | ||
817 | ||
818 | Lisp_Object | |
971de7fb | 819 | string_escape_byte8 (Lisp_Object string) |
0168c3d8 | 820 | { |
d311d28c PE |
821 | ptrdiff_t nchars = SCHARS (string); |
822 | ptrdiff_t nbytes = SBYTES (string); | |
d5172d4f | 823 | bool multibyte = STRING_MULTIBYTE (string); |
d311d28c | 824 | ptrdiff_t byte8_count; |
15843e6f KH |
825 | const unsigned char *src, *src_end; |
826 | unsigned char *dst; | |
0168c3d8 KH |
827 | Lisp_Object val; |
828 | int c, len; | |
829 | ||
830 | if (multibyte && nchars == nbytes) | |
831 | return string; | |
832 | ||
833 | byte8_count = string_count_byte8 (string); | |
834 | ||
835 | if (byte8_count == 0) | |
836 | return string; | |
837 | ||
838 | if (multibyte) | |
14162469 EZ |
839 | { |
840 | if ((MOST_POSITIVE_FIXNUM - nchars) / 3 < byte8_count | |
c9d624c6 | 841 | || (STRING_BYTES_BOUND - nbytes) / 2 < byte8_count) |
cb93f9be | 842 | string_overflow (); |
14162469 EZ |
843 | |
844 | /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
845 | val = make_uninit_multibyte_string (nchars + byte8_count * 3, | |
846 | nbytes + byte8_count * 2); | |
847 | } | |
0168c3d8 | 848 | else |
14162469 | 849 | { |
c9d624c6 | 850 | if ((STRING_BYTES_BOUND - nbytes) / 3 < byte8_count) |
cb93f9be | 851 | string_overflow (); |
ef3ff036 | 852 | |
14162469 EZ |
853 | /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ |
854 | val = make_uninit_string (nbytes + byte8_count * 3); | |
855 | } | |
0168c3d8 | 856 | |
8f924df7 | 857 | src = SDATA (string); |
0168c3d8 | 858 | src_end = src + nbytes; |
8f924df7 | 859 | dst = SDATA (val); |
0168c3d8 KH |
860 | if (multibyte) |
861 | while (src < src_end) | |
862 | { | |
863 | c = *src; | |
864 | len = BYTES_BY_CHAR_HEAD (c); | |
865 | ||
866 | if (CHAR_BYTE8_HEAD_P (c)) | |
867 | { | |
868 | c = STRING_CHAR_ADVANCE (src); | |
869 | c = CHAR_TO_BYTE8 (c); | |
99027bdd | 870 | dst += sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
871 | } |
872 | else | |
873 | while (len--) *dst++ = *src++; | |
874 | } | |
875 | else | |
876 | while (src < src_end) | |
877 | { | |
878 | c = *src++; | |
879 | if (c >= 0x80) | |
99027bdd | 880 | dst += sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
881 | else |
882 | *dst++ = c; | |
883 | } | |
884 | return val; | |
885 | } | |
886 | ||
887 | \f | |
a7ca3326 | 888 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 889 | doc: /* |
d2e83296 DL |
890 | Concatenate all the argument characters and make the result a string. |
891 | usage: (string &rest CHARACTERS) */) | |
f66c7cf8 | 892 | (ptrdiff_t n, Lisp_Object *args) |
0168c3d8 | 893 | { |
f66c7cf8 | 894 | ptrdiff_t i; |
c5101a77 | 895 | int c; |
754790b6 CY |
896 | unsigned char *buf, *p; |
897 | Lisp_Object str; | |
898 | USE_SAFE_ALLOCA; | |
899 | ||
0065d054 | 900 | SAFE_NALLOCA (buf, MAX_MULTIBYTE_LENGTH, n); |
754790b6 | 901 | p = buf; |
0168c3d8 KH |
902 | |
903 | for (i = 0; i < n; i++) | |
904 | { | |
905 | CHECK_CHARACTER (args[i]); | |
906 | c = XINT (args[i]); | |
907 | p += CHAR_STRING (c, p); | |
908 | } | |
909 | ||
754790b6 CY |
910 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
911 | SAFE_FREE (); | |
912 | return str; | |
0168c3d8 KH |
913 | } |
914 | ||
70b4969d | 915 | DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0, |
87d6f965 KH |
916 | doc: /* Concatenate all the argument bytes and make the result a unibyte string. |
917 | usage: (unibyte-string &rest BYTES) */) | |
f66c7cf8 | 918 | (ptrdiff_t n, Lisp_Object *args) |
70b4969d | 919 | { |
f66c7cf8 | 920 | ptrdiff_t i; |
754790b6 CY |
921 | Lisp_Object str; |
922 | USE_SAFE_ALLOCA; | |
98c6f1e3 PE |
923 | unsigned char *buf = SAFE_ALLOCA (n); |
924 | unsigned char *p = buf; | |
70b4969d KH |
925 | |
926 | for (i = 0; i < n; i++) | |
927 | { | |
af5a5a98 | 928 | CHECK_RANGED_INTEGER (args[i], 0, 255); |
a14e1568 | 929 | *p++ = XINT (args[i]); |
70b4969d KH |
930 | } |
931 | ||
754790b6 CY |
932 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
933 | SAFE_FREE (); | |
934 | return str; | |
70b4969d KH |
935 | } |
936 | ||
c73ae4ae | 937 | DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers, |
d0363d44 KH |
938 | Schar_resolve_modifiers, 1, 1, 0, |
939 | doc: /* Resolve modifiers in the character CHAR. | |
940 | The value is a character with modifiers resolved into the character | |
941 | code. Unresolved modifiers are kept in the value. | |
c73ae4ae | 942 | usage: (char-resolve-modifiers CHAR) */) |
5842a27b | 943 | (Lisp_Object character) |
d0363d44 | 944 | { |
d311d28c | 945 | EMACS_INT c; |
d0363d44 KH |
946 | |
947 | CHECK_NUMBER (character); | |
948 | c = XINT (character); | |
949 | return make_number (char_resolve_modifier_mask (c)); | |
950 | } | |
951 | ||
ee107a89 KH |
952 | DEFUN ("get-byte", Fget_byte, Sget_byte, 0, 2, 0, |
953 | doc: /* Return a byte value of a character at point. | |
954 | Optional 1st arg POSITION, if non-nil, is a position of a character to get | |
955 | a byte value. | |
956 | Optional 2nd arg STRING, if non-nil, is a string of which first | |
957 | character is a target to get a byte value. In this case, POSITION, if | |
958 | non-nil, is an index of a target character in the string. | |
959 | ||
960 | If the current buffer (or STRING) is multibyte, and the target | |
8350f087 | 961 | character is not ASCII nor 8-bit character, an error is signaled. */) |
5842a27b | 962 | (Lisp_Object position, Lisp_Object string) |
ee107a89 KH |
963 | { |
964 | int c; | |
d311d28c | 965 | ptrdiff_t pos; |
ee107a89 KH |
966 | unsigned char *p; |
967 | ||
968 | if (NILP (string)) | |
969 | { | |
970 | if (NILP (position)) | |
971 | { | |
972 | p = PT_ADDR; | |
4d8e170e | 973 | } |
ee107a89 KH |
974 | else |
975 | { | |
976 | CHECK_NUMBER_COERCE_MARKER (position); | |
977 | if (XINT (position) < BEGV || XINT (position) >= ZV) | |
978 | args_out_of_range_3 (position, make_number (BEGV), make_number (ZV)); | |
979 | pos = XFASTINT (position); | |
980 | p = CHAR_POS_ADDR (pos); | |
981 | } | |
4b4deea2 | 982 | if (NILP (BVAR (current_buffer, enable_multibyte_characters))) |
d5998e03 | 983 | return make_number (*p); |
ee107a89 KH |
984 | } |
985 | else | |
986 | { | |
987 | CHECK_STRING (string); | |
988 | if (NILP (position)) | |
989 | { | |
990 | p = SDATA (string); | |
991 | } | |
992 | else | |
993 | { | |
994 | CHECK_NATNUM (position); | |
995 | if (XINT (position) >= SCHARS (string)) | |
996 | args_out_of_range (string, position); | |
997 | pos = XFASTINT (position); | |
998 | p = SDATA (string) + string_char_to_byte (string, pos); | |
999 | } | |
d5998e03 KH |
1000 | if (! STRING_MULTIBYTE (string)) |
1001 | return make_number (*p); | |
ee107a89 | 1002 | } |
62a6e103 | 1003 | c = STRING_CHAR (p); |
ee107a89 KH |
1004 | if (CHAR_BYTE8_P (c)) |
1005 | c = CHAR_TO_BYTE8 (c); | |
1006 | else if (! ASCII_CHAR_P (c)) | |
1007 | error ("Not an ASCII nor an 8-bit character: %d", c); | |
1008 | return make_number (c); | |
1009 | } | |
1010 | ||
0168c3d8 KH |
1011 | #ifdef emacs |
1012 | ||
1013 | void | |
971de7fb | 1014 | syms_of_character (void) |
0168c3d8 KH |
1015 | { |
1016 | DEFSYM (Qcharacterp, "characterp"); | |
1017 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
1018 | ||
1019 | staticpro (&Vchar_unify_table); | |
1020 | Vchar_unify_table = Qnil; | |
1021 | ||
1022 | defsubr (&Smax_char); | |
1023 | defsubr (&Scharacterp); | |
1024 | defsubr (&Sunibyte_char_to_multibyte); | |
1025 | defsubr (&Smultibyte_char_to_unibyte); | |
0168c3d8 KH |
1026 | defsubr (&Schar_width); |
1027 | defsubr (&Sstring_width); | |
0168c3d8 | 1028 | defsubr (&Sstring); |
70b4969d | 1029 | defsubr (&Sunibyte_string); |
d0363d44 | 1030 | defsubr (&Schar_resolve_modifiers); |
ee107a89 | 1031 | defsubr (&Sget_byte); |
0168c3d8 | 1032 | |
29208e82 | 1033 | DEFVAR_LISP ("translation-table-vector", Vtranslation_table_vector, |
0168c3d8 | 1034 | doc: /* |
68978cf0 DL |
1035 | Vector recording all translation tables ever defined. |
1036 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
1037 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
1038 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
1039 | ||
29208e82 | 1040 | DEFVAR_LISP ("auto-fill-chars", Vauto_fill_chars, |
0168c3d8 KH |
1041 | doc: /* |
1042 | A char-table for characters which invoke auto-filling. | |
1043 | Such characters have value t in this table. */); | |
1044 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
1045 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
1046 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 | 1047 | |
29208e82 | 1048 | DEFVAR_LISP ("char-width-table", Vchar_width_table, |
0168c3d8 KH |
1049 | doc: /* |
1050 | A char-table for width (columns) of each character. */); | |
1051 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
1052 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
1053 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
1054 | make_number (4)); | |
0168c3d8 | 1055 | |
29208e82 | 1056 | DEFVAR_LISP ("printable-chars", Vprintable_chars, |
0168c3d8 | 1057 | doc: /* A char-table for each printable character. */); |
db6d4189 | 1058 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
1059 | Fset_char_table_range (Vprintable_chars, |
1060 | Fcons (make_number (32), make_number (126)), Qt); | |
1061 | Fset_char_table_range (Vprintable_chars, | |
1062 | Fcons (make_number (160), | |
1063 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 1064 | |
29208e82 | 1065 | DEFVAR_LISP ("char-script-table", Vchar_script_table, |
c57f3328 KH |
1066 | doc: /* Char table of script symbols. |
1067 | It has one extra slot whose value is a list of script symbols. */); | |
1068 | ||
1069 | /* Intern this now in case it isn't already done. | |
1070 | Setting this variable twice is harmless. | |
1071 | But don't staticpro it here--that is done in alloc.c. */ | |
d67b4f80 | 1072 | Qchar_table_extra_slots = intern_c_string ("char-table-extra-slots"); |
c57f3328 KH |
1073 | DEFSYM (Qchar_script_table, "char-script-table"); |
1074 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
1075 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
c7e14352 | 1076 | |
29208e82 | 1077 | DEFVAR_LISP ("script-representative-chars", Vscript_representative_chars, |
c3bb7671 | 1078 | doc: /* Alist of scripts vs the representative characters. |
f4427a54 | 1079 | Each element is a cons (SCRIPT . CHARS). |
4d8e170e | 1080 | SCRIPT is a symbol representing a script or a subgroup of a script. |
c3bb7671 | 1081 | CHARS is a list or a vector of characters. |
472a4dc9 | 1082 | If it is a list, all characters in the list are necessary for supporting SCRIPT. |
c3bb7671 KH |
1083 | If it is a vector, one of the characters in the vector is necessary. |
1084 | This variable is used to find a font for a specific script. */); | |
c7e14352 | 1085 | Vscript_representative_chars = Qnil; |
a3cbb631 | 1086 | |
29208e82 | 1087 | DEFVAR_LISP ("unicode-category-table", Vunicode_category_table, |
a3cbb631 | 1088 | doc: /* Char table of Unicode's "General Category". |
472a4dc9 JB |
1089 | All Unicode characters have one of the following values (symbol): |
1090 | Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po, | |
a3cbb631 KH |
1091 | Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn |
1092 | See The Unicode Standard for the meaning of those values. */); | |
1093 | /* The correct char-table is setup in characters.el. */ | |
1094 | Vunicode_category_table = Qnil; | |
0168c3d8 KH |
1095 | } |
1096 | ||
1097 | #endif /* emacs */ |