Commit | Line | Data |
---|---|---|
0168c3d8 | 1 | /* Basic character support. |
73b0cd50 GM |
2 | |
3 | Copyright (C) 2001-2011 Free Software Foundation, Inc. | |
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 | ||
32 | #include <stdio.h> | |
33 | ||
34 | #ifdef emacs | |
35 | ||
36 | #include <sys/types.h> | |
d7306fe6 | 37 | #include <setjmp.h> |
2b4560a8 | 38 | #include <intprops.h> |
0168c3d8 KH |
39 | #include "lisp.h" |
40 | #include "character.h" | |
41 | #include "buffer.h" | |
42 | #include "charset.h" | |
43 | #include "composite.h" | |
44 | #include "disptab.h" | |
45 | ||
46 | #else /* not emacs */ | |
47 | ||
48 | #include "mulelib.h" | |
49 | ||
50 | #endif /* emacs */ | |
51 | ||
52 | Lisp_Object Qcharacterp; | |
53 | ||
955cbe7b | 54 | static Lisp_Object Qauto_fill_chars; |
0168c3d8 | 55 | |
33f91981 | 56 | /* Char-table of information about which character to unify to which |
6f1b43a0 | 57 | Unicode character. Mainly used by the macro MAYBE_UNIFY_CHAR. */ |
0168c3d8 KH |
58 | Lisp_Object Vchar_unify_table; |
59 | ||
8973478b | 60 | /* Variable used locally in the macro FETCH_MULTIBYTE_CHAR. */ |
0168c3d8 | 61 | unsigned char *_fetch_multibyte_char_p; |
0168c3d8 | 62 | |
c57f3328 KH |
63 | static Lisp_Object Qchar_script_table; |
64 | ||
0168c3d8 KH |
65 | \f |
66 | ||
2bde7652 KH |
67 | /* If character code C has modifier masks, reflect them to the |
68 | character code if possible. Return the resulting code. */ | |
69 | ||
70 | int | |
971de7fb | 71 | char_resolve_modifier_mask (int c) |
2bde7652 | 72 | { |
d0363d44 | 73 | /* A non-ASCII character can't reflect modifier bits to the code. */ |
2bde7652 KH |
74 | if (! ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) |
75 | return c; | |
76 | ||
77 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
2bde7652 KH |
78 | if (c & CHAR_SHIFT) |
79 | { | |
80 | /* Shift modifier is valid only with [A-Za-z]. */ | |
81 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
82 | c &= ~CHAR_SHIFT; | |
83 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
84 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
03365d0e KH |
85 | /* Shift modifier for control characters and SPC is ignored. */ |
86 | else if ((c & ~CHAR_MODIFIER_MASK) <= 0x20) | |
d0363d44 KH |
87 | c &= ~CHAR_SHIFT; |
88 | } | |
2bde7652 KH |
89 | if (c & CHAR_CTL) |
90 | { | |
91 | /* Simulate the code in lread.c. */ | |
92 | /* Allow `\C- ' and `\C-?'. */ | |
03365d0e KH |
93 | if ((c & 0377) == ' ') |
94 | c &= ~0177 & ~ CHAR_CTL; | |
95 | else if ((c & 0377) == '?') | |
96 | c = 0177 | (c & ~0177 & ~CHAR_CTL); | |
2bde7652 KH |
97 | /* ASCII control chars are made from letters (both cases), |
98 | as well as the non-letters within 0100...0137. */ | |
99 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
100 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
101 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
102 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
103 | } | |
e9c1637d | 104 | #if 0 /* This is outside the scope of this function. (bug#4751) */ |
03365d0e KH |
105 | if (c & CHAR_META) |
106 | { | |
107 | /* Move the meta bit to the right place for a string. */ | |
108 | c = (c & ~CHAR_META) | 0x80; | |
109 | } | |
e9c1637d | 110 | #endif |
2bde7652 KH |
111 | |
112 | return c; | |
113 | } | |
114 | ||
115 | ||
33f91981 KH |
116 | /* Store multibyte form of character C at P. If C has modifier bits, |
117 | handle them appropriately. */ | |
118 | ||
0168c3d8 | 119 | int |
971de7fb | 120 | char_string (unsigned int c, unsigned char *p) |
0168c3d8 KH |
121 | { |
122 | int bytes; | |
123 | ||
e3d8eb8c KH |
124 | if (c & CHAR_MODIFIER_MASK) |
125 | { | |
2bde7652 | 126 | c = (unsigned) char_resolve_modifier_mask ((int) c); |
e3d8eb8c KH |
127 | /* If C still has any modifier bits, just ignore it. */ |
128 | c &= ~CHAR_MODIFIER_MASK; | |
129 | } | |
130 | ||
0168c3d8 KH |
131 | MAYBE_UNIFY_CHAR (c); |
132 | ||
e3d8eb8c | 133 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
134 | { |
135 | bytes = CHAR_STRING (c, p); | |
136 | } | |
137 | else if (c <= MAX_4_BYTE_CHAR) | |
138 | { | |
139 | p[0] = (0xF0 | (c >> 18)); | |
140 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
141 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
142 | p[3] = (0x80 | (c & 0x3F)); | |
143 | bytes = 4; | |
144 | } | |
e3d8eb8c | 145 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
146 | { |
147 | p[0] = 0xF8; | |
148 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
149 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
150 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
151 | p[4] = (0x80 | (c & 0x3F)); | |
152 | bytes = 5; | |
153 | } | |
5aa91c9b | 154 | else if (c <= MAX_CHAR) |
e3d8eb8c KH |
155 | { |
156 | c = CHAR_TO_BYTE8 (c); | |
157 | bytes = BYTE8_STRING (c, p); | |
158 | } | |
5aa91c9b | 159 | else |
e6c3da20 | 160 | error ("Invalid character: %x", c); |
1889b238 | 161 | |
0168c3d8 KH |
162 | return bytes; |
163 | } | |
164 | ||
165 | ||
224a3131 | 166 | /* Return a character whose multibyte form is at P. If LEN is not |
33f91981 | 167 | NULL, it must be a pointer to integer. In that case, set *LEN to |
224a3131 | 168 | the byte length of the multibyte form. If ADVANCED is not NULL, it |
33f91981 | 169 | must be a pointer to unsigned char. In that case, set *ADVANCED to |
224a3131 | 170 | the ending address (i.e., the starting address of the next |
33f91981 KH |
171 | character) of the multibyte form. */ |
172 | ||
0168c3d8 | 173 | int |
971de7fb | 174 | string_char (const unsigned char *p, const unsigned char **advanced, int *len) |
0168c3d8 | 175 | { |
1889b238 | 176 | int c; |
15843e6f | 177 | const unsigned char *saved_p = p; |
0168c3d8 KH |
178 | |
179 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
180 | { | |
181 | c = STRING_CHAR_ADVANCE (p); | |
182 | } | |
183 | else if (! (*p & 0x08)) | |
184 | { | |
185 | c = ((((p)[0] & 0xF) << 18) | |
186 | | (((p)[1] & 0x3F) << 12) | |
187 | | (((p)[2] & 0x3F) << 6) | |
188 | | ((p)[3] & 0x3F)); | |
189 | p += 4; | |
190 | } | |
191 | else | |
192 | { | |
193 | c = ((((p)[1] & 0x3F) << 18) | |
194 | | (((p)[2] & 0x3F) << 12) | |
195 | | (((p)[3] & 0x3F) << 6) | |
196 | | ((p)[4] & 0x3F)); | |
197 | p += 5; | |
198 | } | |
199 | ||
200 | MAYBE_UNIFY_CHAR (c); | |
201 | ||
202 | if (len) | |
203 | *len = p - saved_p; | |
204 | if (advanced) | |
205 | *advanced = p; | |
206 | return c; | |
207 | } | |
208 | ||
209 | ||
224a3131 EZ |
210 | /* Translate character C by translation table TABLE. If no translation is |
211 | found in TABLE, return the untranslated character. If TABLE is a list, | |
212 | elements are char tables. In that case, recursively translate C by all the | |
213 | tables in the list. */ | |
0168c3d8 KH |
214 | |
215 | int | |
971de7fb | 216 | translate_char (Lisp_Object table, int c) |
0168c3d8 | 217 | { |
10453be9 KH |
218 | if (CHAR_TABLE_P (table)) |
219 | { | |
220 | Lisp_Object ch; | |
221 | ||
222 | ch = CHAR_TABLE_REF (table, c); | |
223 | if (CHARACTERP (ch)) | |
224 | c = XINT (ch); | |
225 | } | |
226 | else | |
227 | { | |
228 | for (; CONSP (table); table = XCDR (table)) | |
229 | c = translate_char (XCAR (table), c); | |
230 | } | |
231 | return c; | |
0168c3d8 KH |
232 | } |
233 | ||
2e5db15c | 234 | /* Convert ASCII or 8-bit character C to unibyte. If C is none of |
461c2ab9 | 235 | them, return (C & 0xFF). */ |
0168c3d8 KH |
236 | |
237 | int | |
461c2ab9 | 238 | multibyte_char_to_unibyte (int c) |
0168c3d8 | 239 | { |
2e5db15c KH |
240 | if (c < 0x80) |
241 | return c; | |
b672c5ae KH |
242 | if (CHAR_BYTE8_P (c)) |
243 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 244 | return (c & 0xFF); |
0168c3d8 KH |
245 | } |
246 | ||
935d5b02 KH |
247 | /* Like multibyte_char_to_unibyte, but return -1 if C is not supported |
248 | by charset_unibyte. */ | |
249 | ||
250 | int | |
971de7fb | 251 | multibyte_char_to_unibyte_safe (int c) |
935d5b02 | 252 | { |
2e5db15c KH |
253 | if (c < 0x80) |
254 | return c; | |
935d5b02 KH |
255 | if (CHAR_BYTE8_P (c)) |
256 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 257 | return -1; |
935d5b02 | 258 | } |
0168c3d8 KH |
259 | |
260 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
261 | doc: /* Return non-nil if OBJECT is a character. */) | |
5842a27b | 262 | (Lisp_Object object, Lisp_Object ignore) |
0168c3d8 KH |
263 | { |
264 | return (CHARACTERP (object) ? Qt : Qnil); | |
265 | } | |
266 | ||
267 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
268 | doc: /* Return the character of the maximum code. */) | |
5842a27b | 269 | (void) |
0168c3d8 KH |
270 | { |
271 | return make_number (MAX_CHAR); | |
272 | } | |
273 | ||
274 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
275 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
5556875b | 276 | doc: /* Convert the byte CH to multibyte character. */) |
5842a27b | 277 | (Lisp_Object ch) |
0168c3d8 KH |
278 | { |
279 | int c; | |
0168c3d8 KH |
280 | |
281 | CHECK_CHARACTER (ch); | |
282 | c = XFASTINT (ch); | |
2e5db15c KH |
283 | if (c >= 0x100) |
284 | error ("Not a unibyte character: %d", c); | |
4c0354d7 | 285 | MAKE_CHAR_MULTIBYTE (c); |
0168c3d8 KH |
286 | return make_number (c); |
287 | } | |
288 | ||
289 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
290 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
5556875b SM |
291 | doc: /* Convert the multibyte character CH to a byte. |
292 | If the multibyte character does not represent a byte, return -1. */) | |
5842a27b | 293 | (Lisp_Object ch) |
0168c3d8 | 294 | { |
5556875b | 295 | int cm; |
0168c3d8 KH |
296 | |
297 | CHECK_CHARACTER (ch); | |
5556875b SM |
298 | cm = XFASTINT (ch); |
299 | if (cm < 256) | |
300 | /* Can't distinguish a byte read from a unibyte buffer from | |
301 | a latin1 char, so let's let it slide. */ | |
302 | return ch; | |
303 | else | |
304 | { | |
2afc21f5 | 305 | int cu = CHAR_TO_BYTE_SAFE (cm); |
5556875b SM |
306 | return make_number (cu); |
307 | } | |
0168c3d8 KH |
308 | } |
309 | ||
a7ca3326 | 310 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, |
0168c3d8 KH |
311 | doc: /* Return width of CHAR when displayed in the current buffer. |
312 | The width is measured by how many columns it occupies on the screen. | |
add553ac JB |
313 | Tab is taken to occupy `tab-width' columns. |
314 | usage: (char-width CHAR) */) | |
5842a27b | 315 | (Lisp_Object ch) |
0168c3d8 KH |
316 | { |
317 | Lisp_Object disp; | |
318 | int c, width; | |
319 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
320 | ||
321 | CHECK_CHARACTER (ch); | |
322 | c = XINT (ch); | |
323 | ||
324 | /* Get the way the display table would display it. */ | |
325 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
326 | ||
327 | if (VECTORP (disp)) | |
328 | width = ASIZE (disp); | |
329 | else | |
330 | width = CHAR_WIDTH (c); | |
331 | ||
332 | return make_number (width); | |
333 | } | |
334 | ||
0168c3d8 KH |
335 | /* Return width of string STR of length LEN when displayed in the |
336 | current buffer. The width is measured by how many columns it | |
337 | occupies on the screen. If PRECISION > 0, return the width of | |
338 | longest substring that doesn't exceed PRECISION, and set number of | |
339 | characters and bytes of the substring in *NCHARS and *NBYTES | |
340 | respectively. */ | |
341 | ||
579c18d0 EZ |
342 | EMACS_INT |
343 | c_string_width (const unsigned char *str, EMACS_INT len, int precision, | |
344 | EMACS_INT *nchars, EMACS_INT *nbytes) | |
0168c3d8 | 345 | { |
579c18d0 EZ |
346 | EMACS_INT i = 0, i_byte = 0; |
347 | EMACS_INT width = 0; | |
0168c3d8 KH |
348 | struct Lisp_Char_Table *dp = buffer_display_table (); |
349 | ||
350 | while (i_byte < len) | |
351 | { | |
352 | int bytes, thiswidth; | |
353 | Lisp_Object val; | |
62a6e103 | 354 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes); |
0168c3d8 KH |
355 | |
356 | if (dp) | |
357 | { | |
358 | val = DISP_CHAR_VECTOR (dp, c); | |
359 | if (VECTORP (val)) | |
77b37c05 | 360 | thiswidth = ASIZE (val); |
0168c3d8 KH |
361 | else |
362 | thiswidth = CHAR_WIDTH (c); | |
363 | } | |
364 | else | |
365 | { | |
366 | thiswidth = CHAR_WIDTH (c); | |
367 | } | |
368 | ||
369 | if (precision > 0 | |
370 | && (width + thiswidth > precision)) | |
371 | { | |
372 | *nchars = i; | |
373 | *nbytes = i_byte; | |
374 | return width; | |
375 | } | |
376 | i++; | |
377 | i_byte += bytes; | |
378 | width += thiswidth; | |
379 | } | |
380 | ||
381 | if (precision > 0) | |
382 | { | |
383 | *nchars = i; | |
384 | *nbytes = i_byte; | |
385 | } | |
386 | ||
387 | return width; | |
388 | } | |
389 | ||
1889b238 KH |
390 | /* Return width of string STR of length LEN when displayed in the |
391 | current buffer. The width is measured by how many columns it | |
392 | occupies on the screen. */ | |
393 | ||
579c18d0 | 394 | EMACS_INT |
7469ef5d | 395 | strwidth (const char *str, EMACS_INT len) |
1889b238 | 396 | { |
7469ef5d | 397 | return c_string_width ((const unsigned char *) str, len, -1, NULL, NULL); |
1889b238 KH |
398 | } |
399 | ||
0168c3d8 KH |
400 | /* Return width of Lisp string STRING when displayed in the current |
401 | buffer. The width is measured by how many columns it occupies on | |
402 | the screen while paying attention to compositions. If PRECISION > | |
403 | 0, return the width of longest substring that doesn't exceed | |
404 | PRECISION, and set number of characters and bytes of the substring | |
405 | in *NCHARS and *NBYTES respectively. */ | |
406 | ||
579c18d0 | 407 | EMACS_INT |
2b4560a8 | 408 | lisp_string_width (Lisp_Object string, EMACS_INT precision, |
579c18d0 | 409 | EMACS_INT *nchars, EMACS_INT *nbytes) |
0168c3d8 | 410 | { |
579c18d0 | 411 | EMACS_INT len = SCHARS (string); |
0aee65b9 KH |
412 | /* This set multibyte to 0 even if STRING is multibyte when it |
413 | contains only ascii and eight-bit-graphic, but that's | |
414 | intentional. */ | |
415 | int multibyte = len < SBYTES (string); | |
8f924df7 | 416 | unsigned char *str = SDATA (string); |
579c18d0 EZ |
417 | EMACS_INT i = 0, i_byte = 0; |
418 | EMACS_INT width = 0; | |
0168c3d8 KH |
419 | struct Lisp_Char_Table *dp = buffer_display_table (); |
420 | ||
421 | while (i < len) | |
422 | { | |
2b4560a8 | 423 | EMACS_INT chars, bytes, thiswidth; |
0168c3d8 KH |
424 | Lisp_Object val; |
425 | int cmp_id; | |
f4bc0685 | 426 | EMACS_INT ignore, end; |
0168c3d8 KH |
427 | |
428 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
429 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
430 | >= 0)) | |
431 | { | |
432 | thiswidth = composition_table[cmp_id]->width; | |
433 | chars = end - i; | |
434 | bytes = string_char_to_byte (string, end) - i_byte; | |
435 | } | |
0168c3d8 KH |
436 | else |
437 | { | |
0aee65b9 | 438 | int c; |
0168c3d8 | 439 | |
0aee65b9 | 440 | if (multibyte) |
2b4560a8 PE |
441 | { |
442 | int cbytes; | |
443 | c = STRING_CHAR_AND_LENGTH (str + i_byte, cbytes); | |
444 | bytes = cbytes; | |
445 | } | |
0aee65b9 KH |
446 | else |
447 | c = str[i_byte], bytes = 1; | |
0168c3d8 | 448 | chars = 1; |
0aee65b9 KH |
449 | if (dp) |
450 | { | |
451 | val = DISP_CHAR_VECTOR (dp, c); | |
452 | if (VECTORP (val)) | |
77b37c05 | 453 | thiswidth = ASIZE (val); |
0aee65b9 KH |
454 | else |
455 | thiswidth = CHAR_WIDTH (c); | |
456 | } | |
457 | else | |
458 | { | |
459 | thiswidth = CHAR_WIDTH (c); | |
460 | } | |
0168c3d8 KH |
461 | } |
462 | ||
2b4560a8 PE |
463 | if (precision <= 0) |
464 | { | |
465 | #ifdef emacs | |
466 | if (INT_ADD_OVERFLOW (width, thiswidth)) | |
467 | string_overflow (); | |
468 | #endif | |
469 | } | |
470 | else if (precision - width < thiswidth) | |
0168c3d8 KH |
471 | { |
472 | *nchars = i; | |
473 | *nbytes = i_byte; | |
474 | return width; | |
475 | } | |
476 | i += chars; | |
477 | i_byte += bytes; | |
478 | width += thiswidth; | |
ef3ff036 | 479 | } |
0168c3d8 KH |
480 | |
481 | if (precision > 0) | |
482 | { | |
483 | *nchars = i; | |
484 | *nbytes = i_byte; | |
485 | } | |
486 | ||
487 | return width; | |
488 | } | |
489 | ||
490 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
491 | doc: /* Return width of STRING when displayed in the current buffer. | |
492 | Width is measured by how many columns it occupies on the screen. | |
493 | When calculating width of a multibyte character in STRING, | |
494 | only the base leading-code is considered; the validity of | |
495 | the following bytes is not checked. Tabs in STRING are always | |
add553ac JB |
496 | taken to occupy `tab-width' columns. |
497 | usage: (string-width STRING) */) | |
5842a27b | 498 | (Lisp_Object str) |
0168c3d8 KH |
499 | { |
500 | Lisp_Object val; | |
501 | ||
502 | CHECK_STRING (str); | |
503 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
504 | return val; | |
505 | } | |
506 | ||
0168c3d8 KH |
507 | /* Return the number of characters in the NBYTES bytes at PTR. |
508 | This works by looking at the contents and checking for multibyte | |
509 | sequences while assuming that there's no invalid sequence. | |
510 | However, if the current buffer has enable-multibyte-characters = | |
511 | nil, we treat each byte as a character. */ | |
512 | ||
13818c30 | 513 | EMACS_INT |
971de7fb | 514 | chars_in_text (const unsigned char *ptr, EMACS_INT nbytes) |
0168c3d8 KH |
515 | { |
516 | /* current_buffer is null at early stages of Emacs initialization. */ | |
517 | if (current_buffer == 0 | |
4b4deea2 | 518 | || NILP (BVAR (current_buffer, enable_multibyte_characters))) |
0168c3d8 KH |
519 | return nbytes; |
520 | ||
521 | return multibyte_chars_in_text (ptr, nbytes); | |
522 | } | |
523 | ||
524 | /* Return the number of characters in the NBYTES bytes at PTR. | |
525 | This works by looking at the contents and checking for multibyte | |
526 | sequences while assuming that there's no invalid sequence. It | |
527 | ignores enable-multibyte-characters. */ | |
528 | ||
13818c30 | 529 | EMACS_INT |
971de7fb | 530 | multibyte_chars_in_text (const unsigned char *ptr, EMACS_INT nbytes) |
0168c3d8 | 531 | { |
8f924df7 | 532 | const unsigned char *endp = ptr + nbytes; |
579c18d0 | 533 | EMACS_INT chars = 0; |
0168c3d8 KH |
534 | |
535 | while (ptr < endp) | |
536 | { | |
579c18d0 | 537 | EMACS_INT len = MULTIBYTE_LENGTH (ptr, endp); |
0168c3d8 KH |
538 | |
539 | if (len == 0) | |
540 | abort (); | |
541 | ptr += len; | |
542 | chars++; | |
543 | } | |
544 | ||
545 | return chars; | |
546 | } | |
547 | ||
548 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
549 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
550 | respectively. On counting bytes, pay attention to that 8-bit | |
551 | characters not constructing a valid multibyte sequence are | |
552 | represented by 2-byte in a multibyte text. */ | |
553 | ||
554 | void | |
14162469 EZ |
555 | parse_str_as_multibyte (const unsigned char *str, EMACS_INT len, |
556 | EMACS_INT *nchars, EMACS_INT *nbytes) | |
0168c3d8 | 557 | { |
8f924df7 | 558 | const unsigned char *endp = str + len; |
14162469 | 559 | EMACS_INT n, 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. | |
590 | It actually converts only such 8-bit characters that don't contruct | |
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 | ||
14162469 EZ |
597 | EMACS_INT |
598 | str_as_multibyte (unsigned char *str, EMACS_INT len, EMACS_INT nbytes, | |
599 | EMACS_INT *nchars) | |
0168c3d8 KH |
600 | { |
601 | unsigned char *p = str, *endp = str + nbytes; | |
602 | unsigned char *to; | |
14162469 | 603 | EMACS_INT 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 | |
671 | bytes it may ocupy when converted to multibyte string by | |
672 | `str_to_multibyte'. */ | |
673 | ||
14162469 | 674 | EMACS_INT |
de883a70 | 675 | count_size_as_multibyte (const unsigned char *str, EMACS_INT len) |
0168c3d8 | 676 | { |
eec47d6b | 677 | const unsigned char *endp = str + len; |
14162469 | 678 | EMACS_INT 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 | ||
14162469 EZ |
697 | EMACS_INT |
698 | str_to_multibyte (unsigned char *str, EMACS_INT len, EMACS_INT 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 | ||
14162469 EZ |
726 | EMACS_INT |
727 | str_as_unibyte (unsigned char *str, EMACS_INT 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 | |
8307f923 | 763 | contains a non-ASCII, non-eight-bit character. If ACCEPT_LATIN_1 |
4aa40bb8 | 764 | is nonzero, a Latin-1 character is accepted and converted to a byte |
f27f70ec KH |
765 | of that character code. |
766 | Note: Currently the arg ACCEPT_LATIN_1 is not used. */ | |
4aa40bb8 KH |
767 | |
768 | EMACS_INT | |
971de7fb | 769 | str_to_unibyte (const unsigned char *src, unsigned char *dst, EMACS_INT chars, int accept_latin_1) |
4aa40bb8 KH |
770 | { |
771 | EMACS_INT i; | |
772 | ||
773 | for (i = 0; i < chars; i++) | |
774 | { | |
775 | int c = STRING_CHAR_ADVANCE (src); | |
776 | ||
777 | if (CHAR_BYTE8_P (c)) | |
778 | c = CHAR_TO_BYTE8 (c); | |
779 | else if (! ASCII_CHAR_P (c) | |
780 | && (! accept_latin_1 || c >= 0x100)) | |
781 | return i; | |
782 | *dst++ = c; | |
783 | } | |
784 | return i; | |
785 | } | |
786 | ||
787 | ||
d0891610 | 788 | static EMACS_INT |
971de7fb | 789 | string_count_byte8 (Lisp_Object string) |
0168c3d8 KH |
790 | { |
791 | int multibyte = STRING_MULTIBYTE (string); | |
14162469 | 792 | EMACS_INT nbytes = SBYTES (string); |
8f924df7 | 793 | unsigned char *p = SDATA (string); |
0168c3d8 | 794 | unsigned char *pend = p + nbytes; |
14162469 | 795 | EMACS_INT 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 | { |
14162469 EZ |
821 | EMACS_INT nchars = SCHARS (string); |
822 | EMACS_INT nbytes = SBYTES (string); | |
0168c3d8 | 823 | int multibyte = STRING_MULTIBYTE (string); |
14162469 | 824 | EMACS_INT 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 | |
d1f3d2af | 841 | || (STRING_BYTES_MAX - 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 | { |
d1f3d2af | 850 | if ((STRING_BYTES_MAX - nchars) / 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); | |
1889b238 | 870 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
871 | dst += 4; |
872 | } | |
873 | else | |
874 | while (len--) *dst++ = *src++; | |
875 | } | |
876 | else | |
877 | while (src < src_end) | |
878 | { | |
879 | c = *src++; | |
880 | if (c >= 0x80) | |
881 | { | |
1889b238 | 882 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
883 | dst += 4; |
884 | } | |
885 | else | |
886 | *dst++ = c; | |
887 | } | |
888 | return val; | |
889 | } | |
890 | ||
891 | \f | |
a7ca3326 | 892 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 893 | doc: /* |
d2e83296 DL |
894 | Concatenate all the argument characters and make the result a string. |
895 | usage: (string &rest CHARACTERS) */) | |
c5101a77 | 896 | (size_t n, Lisp_Object *args) |
0168c3d8 | 897 | { |
c5101a77 PE |
898 | size_t i; |
899 | int c; | |
754790b6 CY |
900 | unsigned char *buf, *p; |
901 | Lisp_Object str; | |
902 | USE_SAFE_ALLOCA; | |
903 | ||
904 | SAFE_ALLOCA (buf, unsigned char *, MAX_MULTIBYTE_LENGTH * n); | |
905 | p = buf; | |
0168c3d8 KH |
906 | |
907 | for (i = 0; i < n; i++) | |
908 | { | |
909 | CHECK_CHARACTER (args[i]); | |
910 | c = XINT (args[i]); | |
911 | p += CHAR_STRING (c, p); | |
912 | } | |
913 | ||
754790b6 CY |
914 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
915 | SAFE_FREE (); | |
916 | return str; | |
0168c3d8 KH |
917 | } |
918 | ||
70b4969d | 919 | DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0, |
87d6f965 KH |
920 | doc: /* Concatenate all the argument bytes and make the result a unibyte string. |
921 | usage: (unibyte-string &rest BYTES) */) | |
c5101a77 | 922 | (size_t n, Lisp_Object *args) |
70b4969d | 923 | { |
c5101a77 PE |
924 | size_t i; |
925 | int c; | |
754790b6 CY |
926 | unsigned char *buf, *p; |
927 | Lisp_Object str; | |
928 | USE_SAFE_ALLOCA; | |
929 | ||
930 | SAFE_ALLOCA (buf, unsigned char *, n); | |
931 | p = buf; | |
70b4969d KH |
932 | |
933 | for (i = 0; i < n; i++) | |
934 | { | |
935 | CHECK_NATNUM (args[i]); | |
936 | c = XFASTINT (args[i]); | |
937 | if (c >= 256) | |
938 | args_out_of_range_3 (args[i], make_number (0), make_number (255)); | |
939 | *p++ = c; | |
940 | } | |
941 | ||
754790b6 CY |
942 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
943 | SAFE_FREE (); | |
944 | return str; | |
70b4969d KH |
945 | } |
946 | ||
c73ae4ae | 947 | DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers, |
d0363d44 KH |
948 | Schar_resolve_modifiers, 1, 1, 0, |
949 | doc: /* Resolve modifiers in the character CHAR. | |
950 | The value is a character with modifiers resolved into the character | |
951 | code. Unresolved modifiers are kept in the value. | |
c73ae4ae | 952 | usage: (char-resolve-modifiers CHAR) */) |
5842a27b | 953 | (Lisp_Object character) |
d0363d44 KH |
954 | { |
955 | int c; | |
956 | ||
957 | CHECK_NUMBER (character); | |
958 | c = XINT (character); | |
959 | return make_number (char_resolve_modifier_mask (c)); | |
960 | } | |
961 | ||
ee107a89 KH |
962 | DEFUN ("get-byte", Fget_byte, Sget_byte, 0, 2, 0, |
963 | doc: /* Return a byte value of a character at point. | |
964 | Optional 1st arg POSITION, if non-nil, is a position of a character to get | |
965 | a byte value. | |
966 | Optional 2nd arg STRING, if non-nil, is a string of which first | |
967 | character is a target to get a byte value. In this case, POSITION, if | |
968 | non-nil, is an index of a target character in the string. | |
969 | ||
970 | If the current buffer (or STRING) is multibyte, and the target | |
971 | character is not ASCII nor 8-bit character, an error is signalled. */) | |
5842a27b | 972 | (Lisp_Object position, Lisp_Object string) |
ee107a89 KH |
973 | { |
974 | int c; | |
975 | EMACS_INT pos; | |
976 | unsigned char *p; | |
977 | ||
978 | if (NILP (string)) | |
979 | { | |
980 | if (NILP (position)) | |
981 | { | |
982 | p = PT_ADDR; | |
4d8e170e | 983 | } |
ee107a89 KH |
984 | else |
985 | { | |
986 | CHECK_NUMBER_COERCE_MARKER (position); | |
987 | if (XINT (position) < BEGV || XINT (position) >= ZV) | |
988 | args_out_of_range_3 (position, make_number (BEGV), make_number (ZV)); | |
989 | pos = XFASTINT (position); | |
990 | p = CHAR_POS_ADDR (pos); | |
991 | } | |
4b4deea2 | 992 | if (NILP (BVAR (current_buffer, enable_multibyte_characters))) |
d5998e03 | 993 | return make_number (*p); |
ee107a89 KH |
994 | } |
995 | else | |
996 | { | |
997 | CHECK_STRING (string); | |
998 | if (NILP (position)) | |
999 | { | |
1000 | p = SDATA (string); | |
1001 | } | |
1002 | else | |
1003 | { | |
1004 | CHECK_NATNUM (position); | |
1005 | if (XINT (position) >= SCHARS (string)) | |
1006 | args_out_of_range (string, position); | |
1007 | pos = XFASTINT (position); | |
1008 | p = SDATA (string) + string_char_to_byte (string, pos); | |
1009 | } | |
d5998e03 KH |
1010 | if (! STRING_MULTIBYTE (string)) |
1011 | return make_number (*p); | |
ee107a89 | 1012 | } |
62a6e103 | 1013 | c = STRING_CHAR (p); |
ee107a89 KH |
1014 | if (CHAR_BYTE8_P (c)) |
1015 | c = CHAR_TO_BYTE8 (c); | |
1016 | else if (! ASCII_CHAR_P (c)) | |
1017 | error ("Not an ASCII nor an 8-bit character: %d", c); | |
1018 | return make_number (c); | |
1019 | } | |
1020 | ||
1021 | ||
0168c3d8 | 1022 | void |
971de7fb | 1023 | init_character_once (void) |
0168c3d8 KH |
1024 | { |
1025 | } | |
1026 | ||
1027 | #ifdef emacs | |
1028 | ||
1029 | void | |
971de7fb | 1030 | syms_of_character (void) |
0168c3d8 KH |
1031 | { |
1032 | DEFSYM (Qcharacterp, "characterp"); | |
1033 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
1034 | ||
1035 | staticpro (&Vchar_unify_table); | |
1036 | Vchar_unify_table = Qnil; | |
1037 | ||
1038 | defsubr (&Smax_char); | |
1039 | defsubr (&Scharacterp); | |
1040 | defsubr (&Sunibyte_char_to_multibyte); | |
1041 | defsubr (&Smultibyte_char_to_unibyte); | |
0168c3d8 KH |
1042 | defsubr (&Schar_width); |
1043 | defsubr (&Sstring_width); | |
0168c3d8 | 1044 | defsubr (&Sstring); |
70b4969d | 1045 | defsubr (&Sunibyte_string); |
d0363d44 | 1046 | defsubr (&Schar_resolve_modifiers); |
ee107a89 | 1047 | defsubr (&Sget_byte); |
0168c3d8 | 1048 | |
29208e82 | 1049 | DEFVAR_LISP ("translation-table-vector", Vtranslation_table_vector, |
0168c3d8 | 1050 | doc: /* |
68978cf0 DL |
1051 | Vector recording all translation tables ever defined. |
1052 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
1053 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
1054 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
1055 | ||
29208e82 | 1056 | DEFVAR_LISP ("auto-fill-chars", Vauto_fill_chars, |
0168c3d8 KH |
1057 | doc: /* |
1058 | A char-table for characters which invoke auto-filling. | |
1059 | Such characters have value t in this table. */); | |
1060 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
1061 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
1062 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 | 1063 | |
29208e82 | 1064 | DEFVAR_LISP ("char-width-table", Vchar_width_table, |
0168c3d8 KH |
1065 | doc: /* |
1066 | A char-table for width (columns) of each character. */); | |
1067 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
1068 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
1069 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
1070 | make_number (4)); | |
0168c3d8 | 1071 | |
29208e82 | 1072 | DEFVAR_LISP ("printable-chars", Vprintable_chars, |
0168c3d8 | 1073 | doc: /* A char-table for each printable character. */); |
db6d4189 | 1074 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
1075 | Fset_char_table_range (Vprintable_chars, |
1076 | Fcons (make_number (32), make_number (126)), Qt); | |
1077 | Fset_char_table_range (Vprintable_chars, | |
1078 | Fcons (make_number (160), | |
1079 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 1080 | |
29208e82 | 1081 | DEFVAR_LISP ("char-script-table", Vchar_script_table, |
c57f3328 KH |
1082 | doc: /* Char table of script symbols. |
1083 | It has one extra slot whose value is a list of script symbols. */); | |
1084 | ||
1085 | /* Intern this now in case it isn't already done. | |
1086 | Setting this variable twice is harmless. | |
1087 | But don't staticpro it here--that is done in alloc.c. */ | |
d67b4f80 | 1088 | Qchar_table_extra_slots = intern_c_string ("char-table-extra-slots"); |
c57f3328 KH |
1089 | DEFSYM (Qchar_script_table, "char-script-table"); |
1090 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
1091 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
c7e14352 | 1092 | |
29208e82 | 1093 | DEFVAR_LISP ("script-representative-chars", Vscript_representative_chars, |
c3bb7671 | 1094 | doc: /* Alist of scripts vs the representative characters. |
f4427a54 | 1095 | Each element is a cons (SCRIPT . CHARS). |
4d8e170e | 1096 | SCRIPT is a symbol representing a script or a subgroup of a script. |
c3bb7671 | 1097 | CHARS is a list or a vector of characters. |
472a4dc9 | 1098 | If it is a list, all characters in the list are necessary for supporting SCRIPT. |
c3bb7671 KH |
1099 | If it is a vector, one of the characters in the vector is necessary. |
1100 | This variable is used to find a font for a specific script. */); | |
c7e14352 | 1101 | Vscript_representative_chars = Qnil; |
a3cbb631 | 1102 | |
29208e82 | 1103 | DEFVAR_LISP ("unicode-category-table", Vunicode_category_table, |
a3cbb631 | 1104 | doc: /* Char table of Unicode's "General Category". |
472a4dc9 JB |
1105 | All Unicode characters have one of the following values (symbol): |
1106 | Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po, | |
a3cbb631 KH |
1107 | Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn |
1108 | See The Unicode Standard for the meaning of those values. */); | |
1109 | /* The correct char-table is setup in characters.el. */ | |
1110 | Vunicode_category_table = Qnil; | |
0168c3d8 KH |
1111 | } |
1112 | ||
1113 | #endif /* emacs */ |