Commit | Line | Data |
---|---|---|
0168c3d8 KH |
1 | /* Basic character support. |
2 | Copyright (C) 1995, 1997, 1998, 2001 Electrotechnical Laboratory, JAPAN. | |
8f924df7 | 3 | Licensed to the Free Software Foundation. |
ec62e0ac GM |
4 | Copyright (C) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 |
5 | Free Software Foundation, Inc. | |
6 | Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008 | |
0168c3d8 KH |
7 | National Institute of Advanced Industrial Science and Technology (AIST) |
8 | Registration Number H13PRO009 | |
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> | |
37 | #include "lisp.h" | |
38 | #include "character.h" | |
39 | #include "buffer.h" | |
40 | #include "charset.h" | |
41 | #include "composite.h" | |
42 | #include "disptab.h" | |
43 | ||
44 | #else /* not emacs */ | |
45 | ||
46 | #include "mulelib.h" | |
47 | ||
48 | #endif /* emacs */ | |
49 | ||
50 | Lisp_Object Qcharacterp; | |
51 | ||
52 | /* Vector of translation table ever defined. | |
53 | ID of a translation table is used to index this vector. */ | |
54 | Lisp_Object Vtranslation_table_vector; | |
55 | ||
56 | /* A char-table for characters which may invoke auto-filling. */ | |
57 | Lisp_Object Vauto_fill_chars; | |
58 | ||
59 | Lisp_Object Qauto_fill_chars; | |
60 | ||
33f91981 KH |
61 | /* Char-table of information about which character to unify to which |
62 | Unicode character. */ | |
0168c3d8 KH |
63 | Lisp_Object Vchar_unify_table; |
64 | ||
65 | /* A char-table. An element is non-nil iff the corresponding | |
66 | character has a printable glyph. */ | |
67 | Lisp_Object Vprintable_chars; | |
68 | ||
69 | /* A char-table. An elemnent is a column-width of the corresponding | |
70 | character. */ | |
71 | Lisp_Object Vchar_width_table; | |
72 | ||
73 | /* A char-table. An element is a symbol indicating the direction | |
74 | property of corresponding character. */ | |
75 | Lisp_Object Vchar_direction_table; | |
76 | ||
8973478b | 77 | /* Variable used locally in the macro FETCH_MULTIBYTE_CHAR. */ |
0168c3d8 | 78 | unsigned char *_fetch_multibyte_char_p; |
0168c3d8 | 79 | |
c57f3328 KH |
80 | /* Char table of scripts. */ |
81 | Lisp_Object Vchar_script_table; | |
82 | ||
c7e14352 KH |
83 | /* Alist of scripts vs representative characters. */ |
84 | Lisp_Object Vscript_representative_chars; | |
85 | ||
c57f3328 KH |
86 | static Lisp_Object Qchar_script_table; |
87 | ||
a3cbb631 KH |
88 | Lisp_Object Vunicode_category_table; |
89 | ||
b672c5ae KH |
90 | /* Mapping table from unibyte chars to multibyte chars. */ |
91 | int unibyte_to_multibyte_table[256]; | |
15843e6f | 92 | |
90c9d035 KH |
93 | /* Nth element is 1 iff unibyte char N can be mapped to a multibyte |
94 | char. */ | |
95 | char unibyte_has_multibyte_table[256]; | |
96 | ||
0168c3d8 KH |
97 | \f |
98 | ||
2bde7652 KH |
99 | /* If character code C has modifier masks, reflect them to the |
100 | character code if possible. Return the resulting code. */ | |
101 | ||
102 | int | |
103 | char_resolve_modifier_mask (c) | |
104 | int c; | |
105 | { | |
d0363d44 | 106 | /* A non-ASCII character can't reflect modifier bits to the code. */ |
2bde7652 KH |
107 | if (! ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) |
108 | return c; | |
109 | ||
110 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
2bde7652 KH |
111 | if (c & CHAR_SHIFT) |
112 | { | |
113 | /* Shift modifier is valid only with [A-Za-z]. */ | |
114 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
115 | c &= ~CHAR_SHIFT; | |
116 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
117 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
03365d0e KH |
118 | /* Shift modifier for control characters and SPC is ignored. */ |
119 | else if ((c & ~CHAR_MODIFIER_MASK) <= 0x20) | |
d0363d44 KH |
120 | c &= ~CHAR_SHIFT; |
121 | } | |
2bde7652 KH |
122 | if (c & CHAR_CTL) |
123 | { | |
124 | /* Simulate the code in lread.c. */ | |
125 | /* Allow `\C- ' and `\C-?'. */ | |
03365d0e KH |
126 | if ((c & 0377) == ' ') |
127 | c &= ~0177 & ~ CHAR_CTL; | |
128 | else if ((c & 0377) == '?') | |
129 | c = 0177 | (c & ~0177 & ~CHAR_CTL); | |
2bde7652 KH |
130 | /* ASCII control chars are made from letters (both cases), |
131 | as well as the non-letters within 0100...0137. */ | |
132 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
133 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
134 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
135 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
136 | } | |
03365d0e KH |
137 | if (c & CHAR_META) |
138 | { | |
139 | /* Move the meta bit to the right place for a string. */ | |
140 | c = (c & ~CHAR_META) | 0x80; | |
141 | } | |
2bde7652 KH |
142 | |
143 | return c; | |
144 | } | |
145 | ||
146 | ||
33f91981 KH |
147 | /* Store multibyte form of character C at P. If C has modifier bits, |
148 | handle them appropriately. */ | |
149 | ||
0168c3d8 | 150 | int |
e3d8eb8c | 151 | char_string (c, p) |
5aa91c9b | 152 | unsigned c; |
1889b238 | 153 | unsigned char *p; |
0168c3d8 KH |
154 | { |
155 | int bytes; | |
156 | ||
e3d8eb8c KH |
157 | if (c & CHAR_MODIFIER_MASK) |
158 | { | |
2bde7652 | 159 | c = (unsigned) char_resolve_modifier_mask ((int) c); |
e3d8eb8c KH |
160 | /* If C still has any modifier bits, just ignore it. */ |
161 | c &= ~CHAR_MODIFIER_MASK; | |
162 | } | |
163 | ||
0168c3d8 KH |
164 | MAYBE_UNIFY_CHAR (c); |
165 | ||
e3d8eb8c | 166 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
167 | { |
168 | bytes = CHAR_STRING (c, p); | |
169 | } | |
170 | else if (c <= MAX_4_BYTE_CHAR) | |
171 | { | |
172 | p[0] = (0xF0 | (c >> 18)); | |
173 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
174 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
175 | p[3] = (0x80 | (c & 0x3F)); | |
176 | bytes = 4; | |
177 | } | |
e3d8eb8c | 178 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
179 | { |
180 | p[0] = 0xF8; | |
181 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
182 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
183 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
184 | p[4] = (0x80 | (c & 0x3F)); | |
185 | bytes = 5; | |
186 | } | |
5aa91c9b | 187 | else if (c <= MAX_CHAR) |
e3d8eb8c KH |
188 | { |
189 | c = CHAR_TO_BYTE8 (c); | |
190 | bytes = BYTE8_STRING (c, p); | |
191 | } | |
5aa91c9b KH |
192 | else |
193 | error ("Invalid character: %d", c); | |
1889b238 | 194 | |
0168c3d8 KH |
195 | return bytes; |
196 | } | |
197 | ||
198 | ||
33f91981 KH |
199 | /* Return a character whose multibyte form is at P. Set LEN is not |
200 | NULL, it must be a pointer to integer. In that case, set *LEN to | |
201 | the byte length of the multibyte form. If ADVANCED is not NULL, is | |
202 | must be a pointer to unsigned char. In that case, set *ADVANCED to | |
203 | the ending address (i.e. the starting address of the next | |
204 | character) of the multibyte form. */ | |
205 | ||
0168c3d8 | 206 | int |
e3d8eb8c | 207 | string_char (p, advanced, len) |
15843e6f KH |
208 | const unsigned char *p; |
209 | const unsigned char **advanced; | |
0168c3d8 KH |
210 | int *len; |
211 | { | |
1889b238 | 212 | int c; |
15843e6f | 213 | const unsigned char *saved_p = p; |
0168c3d8 KH |
214 | |
215 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
216 | { | |
217 | c = STRING_CHAR_ADVANCE (p); | |
218 | } | |
219 | else if (! (*p & 0x08)) | |
220 | { | |
221 | c = ((((p)[0] & 0xF) << 18) | |
222 | | (((p)[1] & 0x3F) << 12) | |
223 | | (((p)[2] & 0x3F) << 6) | |
224 | | ((p)[3] & 0x3F)); | |
225 | p += 4; | |
226 | } | |
227 | else | |
228 | { | |
229 | c = ((((p)[1] & 0x3F) << 18) | |
230 | | (((p)[2] & 0x3F) << 12) | |
231 | | (((p)[3] & 0x3F) << 6) | |
232 | | ((p)[4] & 0x3F)); | |
233 | p += 5; | |
234 | } | |
235 | ||
236 | MAYBE_UNIFY_CHAR (c); | |
237 | ||
238 | if (len) | |
239 | *len = p - saved_p; | |
240 | if (advanced) | |
241 | *advanced = p; | |
242 | return c; | |
243 | } | |
244 | ||
245 | ||
246 | /* Translate character C by translation table TABLE. If C is | |
247 | negative, translate a character specified by CHARSET and CODE. If | |
248 | no translation is found in TABLE, return the untranslated | |
10453be9 KH |
249 | character. If TABLE is a list, elements are char tables. In this |
250 | case, translace C by all tables. */ | |
0168c3d8 KH |
251 | |
252 | int | |
253 | translate_char (table, c) | |
254 | Lisp_Object table; | |
255 | int c; | |
256 | { | |
10453be9 KH |
257 | if (CHAR_TABLE_P (table)) |
258 | { | |
259 | Lisp_Object ch; | |
260 | ||
261 | ch = CHAR_TABLE_REF (table, c); | |
262 | if (CHARACTERP (ch)) | |
263 | c = XINT (ch); | |
264 | } | |
265 | else | |
266 | { | |
267 | for (; CONSP (table); table = XCDR (table)) | |
268 | c = translate_char (XCAR (table), c); | |
269 | } | |
270 | return c; | |
0168c3d8 KH |
271 | } |
272 | ||
0168c3d8 | 273 | /* Convert the multibyte character C to unibyte 8-bit character based |
ac86488b KH |
274 | on the current value of charset_unibyte. If dimension of |
275 | charset_unibyte is more than one, return (C & 0xFF). | |
0168c3d8 KH |
276 | |
277 | The argument REV_TBL is now ignored. It will be removed in the | |
278 | future. */ | |
279 | ||
280 | int | |
281 | multibyte_char_to_unibyte (c, rev_tbl) | |
282 | int c; | |
283 | Lisp_Object rev_tbl; | |
284 | { | |
b672c5ae KH |
285 | struct charset *charset; |
286 | unsigned c1; | |
0168c3d8 | 287 | |
b672c5ae KH |
288 | if (CHAR_BYTE8_P (c)) |
289 | return CHAR_TO_BYTE8 (c); | |
290 | charset = CHARSET_FROM_ID (charset_unibyte); | |
291 | c1 = ENCODE_CHAR (charset, c); | |
0168c3d8 KH |
292 | return ((c1 != CHARSET_INVALID_CODE (charset)) ? c1 : c & 0xFF); |
293 | } | |
294 | ||
935d5b02 KH |
295 | /* Like multibyte_char_to_unibyte, but return -1 if C is not supported |
296 | by charset_unibyte. */ | |
297 | ||
298 | int | |
299 | multibyte_char_to_unibyte_safe (c) | |
300 | int c; | |
301 | { | |
302 | struct charset *charset; | |
303 | unsigned c1; | |
304 | ||
305 | if (CHAR_BYTE8_P (c)) | |
306 | return CHAR_TO_BYTE8 (c); | |
307 | charset = CHARSET_FROM_ID (charset_unibyte); | |
308 | c1 = ENCODE_CHAR (charset, c); | |
309 | return ((c1 != CHARSET_INVALID_CODE (charset)) ? c1 : -1); | |
310 | } | |
0168c3d8 KH |
311 | |
312 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
313 | doc: /* Return non-nil if OBJECT is a character. */) | |
314 | (object, ignore) | |
315 | Lisp_Object object, ignore; | |
316 | { | |
317 | return (CHARACTERP (object) ? Qt : Qnil); | |
318 | } | |
319 | ||
320 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
321 | doc: /* Return the character of the maximum code. */) | |
322 | () | |
323 | { | |
324 | return make_number (MAX_CHAR); | |
325 | } | |
326 | ||
327 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
328 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
5556875b | 329 | doc: /* Convert the byte CH to multibyte character. */) |
0168c3d8 KH |
330 | (ch) |
331 | Lisp_Object ch; | |
332 | { | |
333 | int c; | |
334 | struct charset *charset; | |
335 | ||
336 | CHECK_CHARACTER (ch); | |
337 | c = XFASTINT (ch); | |
338 | if (c >= 0400) | |
339 | error ("Invalid unibyte character: %d", c); | |
ac86488b | 340 | charset = CHARSET_FROM_ID (charset_unibyte); |
0168c3d8 KH |
341 | c = DECODE_CHAR (charset, c); |
342 | if (c < 0) | |
3c5a53bd | 343 | c = BYTE8_TO_CHAR (XFASTINT (ch)); |
0168c3d8 KH |
344 | return make_number (c); |
345 | } | |
346 | ||
347 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
348 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
5556875b SM |
349 | doc: /* Convert the multibyte character CH to a byte. |
350 | If the multibyte character does not represent a byte, return -1. */) | |
0168c3d8 KH |
351 | (ch) |
352 | Lisp_Object ch; | |
353 | { | |
5556875b | 354 | int cm; |
0168c3d8 KH |
355 | |
356 | CHECK_CHARACTER (ch); | |
5556875b SM |
357 | cm = XFASTINT (ch); |
358 | if (cm < 256) | |
359 | /* Can't distinguish a byte read from a unibyte buffer from | |
360 | a latin1 char, so let's let it slide. */ | |
361 | return ch; | |
362 | else | |
363 | { | |
2afc21f5 | 364 | int cu = CHAR_TO_BYTE_SAFE (cm); |
5556875b SM |
365 | return make_number (cu); |
366 | } | |
0168c3d8 KH |
367 | } |
368 | ||
369 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
370 | doc: /* Return 1 regardless of the argument CHAR. | |
add553ac JB |
371 | This is now an obsolete function. We keep it just for backward compatibility. |
372 | usage: (char-bytes CHAR) */) | |
0168c3d8 KH |
373 | (ch) |
374 | Lisp_Object ch; | |
375 | { | |
376 | CHECK_CHARACTER (ch); | |
377 | return make_number (1); | |
378 | } | |
379 | ||
380 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
381 | doc: /* Return width of CHAR when displayed in the current buffer. | |
382 | The width is measured by how many columns it occupies on the screen. | |
add553ac JB |
383 | Tab is taken to occupy `tab-width' columns. |
384 | usage: (char-width CHAR) */) | |
0168c3d8 KH |
385 | (ch) |
386 | Lisp_Object ch; | |
387 | { | |
388 | Lisp_Object disp; | |
389 | int c, width; | |
390 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
391 | ||
392 | CHECK_CHARACTER (ch); | |
393 | c = XINT (ch); | |
394 | ||
395 | /* Get the way the display table would display it. */ | |
396 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
397 | ||
398 | if (VECTORP (disp)) | |
399 | width = ASIZE (disp); | |
400 | else | |
401 | width = CHAR_WIDTH (c); | |
402 | ||
403 | return make_number (width); | |
404 | } | |
405 | ||
0168c3d8 KH |
406 | /* Return width of string STR of length LEN when displayed in the |
407 | current buffer. The width is measured by how many columns it | |
408 | occupies on the screen. If PRECISION > 0, return the width of | |
409 | longest substring that doesn't exceed PRECISION, and set number of | |
410 | characters and bytes of the substring in *NCHARS and *NBYTES | |
411 | respectively. */ | |
412 | ||
1889b238 | 413 | int |
0168c3d8 | 414 | c_string_width (str, len, precision, nchars, nbytes) |
8f924df7 | 415 | const unsigned char *str; |
0168c3d8 KH |
416 | int precision, *nchars, *nbytes; |
417 | { | |
418 | int i = 0, i_byte = 0; | |
419 | int width = 0; | |
420 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
421 | ||
422 | while (i_byte < len) | |
423 | { | |
424 | int bytes, thiswidth; | |
425 | Lisp_Object val; | |
426 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
427 | ||
428 | if (dp) | |
429 | { | |
430 | val = DISP_CHAR_VECTOR (dp, c); | |
431 | if (VECTORP (val)) | |
432 | thiswidth = XVECTOR (val)->size; | |
433 | else | |
434 | thiswidth = CHAR_WIDTH (c); | |
435 | } | |
436 | else | |
437 | { | |
438 | thiswidth = CHAR_WIDTH (c); | |
439 | } | |
440 | ||
441 | if (precision > 0 | |
442 | && (width + thiswidth > precision)) | |
443 | { | |
444 | *nchars = i; | |
445 | *nbytes = i_byte; | |
446 | return width; | |
447 | } | |
448 | i++; | |
449 | i_byte += bytes; | |
450 | width += thiswidth; | |
451 | } | |
452 | ||
453 | if (precision > 0) | |
454 | { | |
455 | *nchars = i; | |
456 | *nbytes = i_byte; | |
457 | } | |
458 | ||
459 | return width; | |
460 | } | |
461 | ||
1889b238 KH |
462 | /* Return width of string STR of length LEN when displayed in the |
463 | current buffer. The width is measured by how many columns it | |
464 | occupies on the screen. */ | |
465 | ||
466 | int | |
467 | strwidth (str, len) | |
468 | unsigned char *str; | |
469 | int len; | |
470 | { | |
471 | return c_string_width (str, len, -1, NULL, NULL); | |
472 | } | |
473 | ||
0168c3d8 KH |
474 | /* Return width of Lisp string STRING when displayed in the current |
475 | buffer. The width is measured by how many columns it occupies on | |
476 | the screen while paying attention to compositions. If PRECISION > | |
477 | 0, return the width of longest substring that doesn't exceed | |
478 | PRECISION, and set number of characters and bytes of the substring | |
479 | in *NCHARS and *NBYTES respectively. */ | |
480 | ||
481 | int | |
482 | lisp_string_width (string, precision, nchars, nbytes) | |
483 | Lisp_Object string; | |
484 | int precision, *nchars, *nbytes; | |
485 | { | |
8f924df7 | 486 | int len = SCHARS (string); |
0aee65b9 KH |
487 | /* This set multibyte to 0 even if STRING is multibyte when it |
488 | contains only ascii and eight-bit-graphic, but that's | |
489 | intentional. */ | |
490 | int multibyte = len < SBYTES (string); | |
8f924df7 | 491 | unsigned char *str = SDATA (string); |
0168c3d8 KH |
492 | int i = 0, i_byte = 0; |
493 | int width = 0; | |
494 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
495 | ||
496 | while (i < len) | |
497 | { | |
498 | int chars, bytes, thiswidth; | |
499 | Lisp_Object val; | |
500 | int cmp_id; | |
f4bc0685 | 501 | EMACS_INT ignore, end; |
0168c3d8 KH |
502 | |
503 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
504 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
505 | >= 0)) | |
506 | { | |
507 | thiswidth = composition_table[cmp_id]->width; | |
508 | chars = end - i; | |
509 | bytes = string_char_to_byte (string, end) - i_byte; | |
510 | } | |
0168c3d8 KH |
511 | else |
512 | { | |
0aee65b9 | 513 | int c; |
0168c3d8 | 514 | |
0aee65b9 KH |
515 | if (multibyte) |
516 | c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
517 | else | |
518 | c = str[i_byte], bytes = 1; | |
0168c3d8 | 519 | chars = 1; |
0aee65b9 KH |
520 | if (dp) |
521 | { | |
522 | val = DISP_CHAR_VECTOR (dp, c); | |
523 | if (VECTORP (val)) | |
524 | thiswidth = XVECTOR (val)->size; | |
525 | else | |
526 | thiswidth = CHAR_WIDTH (c); | |
527 | } | |
528 | else | |
529 | { | |
530 | thiswidth = CHAR_WIDTH (c); | |
531 | } | |
0168c3d8 KH |
532 | } |
533 | ||
534 | if (precision > 0 | |
535 | && (width + thiswidth > precision)) | |
536 | { | |
537 | *nchars = i; | |
538 | *nbytes = i_byte; | |
539 | return width; | |
540 | } | |
541 | i += chars; | |
542 | i_byte += bytes; | |
543 | width += thiswidth; | |
544 | } | |
545 | ||
546 | if (precision > 0) | |
547 | { | |
548 | *nchars = i; | |
549 | *nbytes = i_byte; | |
550 | } | |
551 | ||
552 | return width; | |
553 | } | |
554 | ||
555 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
556 | doc: /* Return width of STRING when displayed in the current buffer. | |
557 | Width is measured by how many columns it occupies on the screen. | |
558 | When calculating width of a multibyte character in STRING, | |
559 | only the base leading-code is considered; the validity of | |
560 | the following bytes is not checked. Tabs in STRING are always | |
add553ac JB |
561 | taken to occupy `tab-width' columns. |
562 | usage: (string-width STRING) */) | |
0168c3d8 KH |
563 | (str) |
564 | Lisp_Object str; | |
565 | { | |
566 | Lisp_Object val; | |
567 | ||
568 | CHECK_STRING (str); | |
569 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
570 | return val; | |
571 | } | |
572 | ||
573 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
574 | doc: /* Return the direction of CHAR. | |
add553ac JB |
575 | The returned value is 0 for left-to-right and 1 for right-to-left. |
576 | usage: (char-direction CHAR) */) | |
0168c3d8 KH |
577 | (ch) |
578 | Lisp_Object ch; | |
579 | { | |
580 | int c; | |
581 | ||
582 | CHECK_CHARACTER (ch); | |
583 | c = XINT (ch); | |
584 | return CHAR_TABLE_REF (Vchar_direction_table, c); | |
585 | } | |
586 | ||
0168c3d8 KH |
587 | /* Return the number of characters in the NBYTES bytes at PTR. |
588 | This works by looking at the contents and checking for multibyte | |
589 | sequences while assuming that there's no invalid sequence. | |
590 | However, if the current buffer has enable-multibyte-characters = | |
591 | nil, we treat each byte as a character. */ | |
592 | ||
13818c30 | 593 | EMACS_INT |
0168c3d8 | 594 | chars_in_text (ptr, nbytes) |
8f924df7 | 595 | const unsigned char *ptr; |
13818c30 | 596 | EMACS_INT nbytes; |
0168c3d8 KH |
597 | { |
598 | /* current_buffer is null at early stages of Emacs initialization. */ | |
599 | if (current_buffer == 0 | |
600 | || NILP (current_buffer->enable_multibyte_characters)) | |
601 | return nbytes; | |
602 | ||
603 | return multibyte_chars_in_text (ptr, nbytes); | |
604 | } | |
605 | ||
606 | /* Return the number of characters in the NBYTES bytes at PTR. | |
607 | This works by looking at the contents and checking for multibyte | |
608 | sequences while assuming that there's no invalid sequence. It | |
609 | ignores enable-multibyte-characters. */ | |
610 | ||
13818c30 | 611 | EMACS_INT |
0168c3d8 | 612 | multibyte_chars_in_text (ptr, nbytes) |
8f924df7 | 613 | const unsigned char *ptr; |
13818c30 | 614 | EMACS_INT nbytes; |
0168c3d8 | 615 | { |
8f924df7 | 616 | const unsigned char *endp = ptr + nbytes; |
0168c3d8 KH |
617 | int chars = 0; |
618 | ||
619 | while (ptr < endp) | |
620 | { | |
621 | int len = MULTIBYTE_LENGTH (ptr, endp); | |
622 | ||
623 | if (len == 0) | |
624 | abort (); | |
625 | ptr += len; | |
626 | chars++; | |
627 | } | |
628 | ||
629 | return chars; | |
630 | } | |
631 | ||
632 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
633 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
634 | respectively. On counting bytes, pay attention to that 8-bit | |
635 | characters not constructing a valid multibyte sequence are | |
636 | represented by 2-byte in a multibyte text. */ | |
637 | ||
638 | void | |
639 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
8f924df7 | 640 | const unsigned char *str; |
0168c3d8 KH |
641 | int len, *nchars, *nbytes; |
642 | { | |
8f924df7 | 643 | const unsigned char *endp = str + len; |
0168c3d8 KH |
644 | int n, chars = 0, bytes = 0; |
645 | ||
646 | if (len >= MAX_MULTIBYTE_LENGTH) | |
647 | { | |
8f924df7 | 648 | const unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; |
0168c3d8 KH |
649 | while (str < adjusted_endp) |
650 | { | |
651 | if ((n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
652 | str += n, bytes += n; | |
653 | else | |
654 | str++, bytes += 2; | |
655 | chars++; | |
656 | } | |
657 | } | |
658 | while (str < endp) | |
659 | { | |
660 | if ((n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
661 | str += n, bytes += n; | |
662 | else | |
663 | str++, bytes += 2; | |
664 | chars++; | |
665 | } | |
666 | ||
667 | *nchars = chars; | |
668 | *nbytes = bytes; | |
669 | return; | |
670 | } | |
671 | ||
672 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
673 | It actually converts only such 8-bit characters that don't contruct | |
674 | a multibyte sequence to multibyte forms of Latin-1 characters. If | |
675 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
676 | text. It is assured that we can use LEN bytes at STR as a work | |
677 | area and that is enough. Return the number of bytes of the | |
678 | resulting text. */ | |
679 | ||
680 | int | |
681 | str_as_multibyte (str, len, nbytes, nchars) | |
682 | unsigned char *str; | |
683 | int len, nbytes, *nchars; | |
684 | { | |
685 | unsigned char *p = str, *endp = str + nbytes; | |
686 | unsigned char *to; | |
687 | int chars = 0; | |
688 | int n; | |
689 | ||
690 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
691 | { | |
692 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
693 | while (p < adjusted_endp | |
694 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
695 | p += n, chars++; | |
696 | } | |
697 | while ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
698 | p += n, chars++; | |
699 | if (nchars) | |
700 | *nchars = chars; | |
701 | if (p == endp) | |
702 | return nbytes; | |
703 | ||
704 | to = p; | |
705 | nbytes = endp - p; | |
706 | endp = str + len; | |
707 | safe_bcopy ((char *) p, (char *) (endp - nbytes), nbytes); | |
708 | p = endp - nbytes; | |
709 | ||
710 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
711 | { | |
712 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
713 | while (p < adjusted_endp) | |
714 | { | |
715 | if ((n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
716 | { | |
717 | while (n--) | |
718 | *to++ = *p++; | |
719 | } | |
720 | else | |
721 | { | |
722 | int c = *p++; | |
723 | c = BYTE8_TO_CHAR (c); | |
724 | to += CHAR_STRING (c, to); | |
725 | } | |
726 | } | |
727 | chars++; | |
728 | } | |
729 | while (p < endp) | |
730 | { | |
731 | if ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
732 | { | |
733 | while (n--) | |
734 | *to++ = *p++; | |
8f924df7 | 735 | } |
0168c3d8 KH |
736 | else |
737 | { | |
738 | int c = *p++; | |
739 | c = BYTE8_TO_CHAR (c); | |
740 | to += CHAR_STRING (c, to); | |
741 | } | |
742 | chars++; | |
743 | } | |
744 | if (nchars) | |
745 | *nchars = chars; | |
746 | return (to - str); | |
747 | } | |
748 | ||
749 | /* Parse unibyte string at STR of LEN bytes, and return the number of | |
750 | bytes it may ocupy when converted to multibyte string by | |
751 | `str_to_multibyte'. */ | |
752 | ||
753 | int | |
754 | parse_str_to_multibyte (str, len) | |
755 | unsigned char *str; | |
756 | int len; | |
757 | { | |
758 | unsigned char *endp = str + len; | |
759 | int bytes; | |
760 | ||
761 | for (bytes = 0; str < endp; str++) | |
762 | bytes += (*str < 0x80) ? 1 : 2; | |
763 | return bytes; | |
764 | } | |
765 | ||
766 | ||
767 | /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
768 | that contains the same single-byte characters. It actually | |
769 | converts all 8-bit characters to multibyte forms. It is assured | |
770 | that we can use LEN bytes at STR as a work area and that is | |
771 | enough. */ | |
772 | ||
773 | int | |
774 | str_to_multibyte (str, len, bytes) | |
775 | unsigned char *str; | |
776 | int len, bytes; | |
777 | { | |
778 | unsigned char *p = str, *endp = str + bytes; | |
779 | unsigned char *to; | |
780 | ||
781 | while (p < endp && *p < 0x80) p++; | |
782 | if (p == endp) | |
783 | return bytes; | |
784 | to = p; | |
785 | bytes = endp - p; | |
786 | endp = str + len; | |
787 | safe_bcopy ((char *) p, (char *) (endp - bytes), bytes); | |
788 | p = endp - bytes; | |
8f924df7 | 789 | while (p < endp) |
0168c3d8 KH |
790 | { |
791 | int c = *p++; | |
792 | ||
793 | if (c >= 0x80) | |
794 | c = BYTE8_TO_CHAR (c); | |
795 | to += CHAR_STRING (c, to); | |
796 | } | |
797 | return (to - str); | |
798 | } | |
799 | ||
800 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
801 | actually converts characters in the range 0x80..0xFF to | |
802 | unibyte. */ | |
803 | ||
804 | int | |
805 | str_as_unibyte (str, bytes) | |
806 | unsigned char *str; | |
807 | int bytes; | |
808 | { | |
15843e6f KH |
809 | const unsigned char *p = str, *endp = str + bytes; |
810 | unsigned char *to; | |
0168c3d8 KH |
811 | int c, len; |
812 | ||
813 | while (p < endp) | |
814 | { | |
815 | c = *p; | |
816 | len = BYTES_BY_CHAR_HEAD (c); | |
817 | if (CHAR_BYTE8_HEAD_P (c)) | |
818 | break; | |
819 | p += len; | |
820 | } | |
15843e6f | 821 | to = str + (p - str); |
8f924df7 | 822 | while (p < endp) |
0168c3d8 KH |
823 | { |
824 | c = *p; | |
825 | len = BYTES_BY_CHAR_HEAD (c); | |
826 | if (CHAR_BYTE8_HEAD_P (c)) | |
827 | { | |
828 | c = STRING_CHAR_ADVANCE (p); | |
829 | *to++ = CHAR_TO_BYTE8 (c); | |
830 | } | |
831 | else | |
832 | { | |
833 | while (len--) *to++ = *p++; | |
834 | } | |
835 | } | |
836 | return (to - str); | |
837 | } | |
838 | ||
4aa40bb8 KH |
839 | /* Convert eight-bit chars in SRC (in multibyte form) to the |
840 | corresponding byte and store in DST. CHARS is the number of | |
841 | characters in SRC. The value is the number of bytes stored in DST. | |
842 | Usually, the value is the same as CHARS, but is less than it if SRC | |
843 | contains a non-ASCII, non-eight-bit characater. If ACCEPT_LATIN_1 | |
844 | is nonzero, a Latin-1 character is accepted and converted to a byte | |
f27f70ec KH |
845 | of that character code. |
846 | Note: Currently the arg ACCEPT_LATIN_1 is not used. */ | |
4aa40bb8 KH |
847 | |
848 | EMACS_INT | |
849 | str_to_unibyte (src, dst, chars, accept_latin_1) | |
850 | const unsigned char *src; | |
851 | unsigned char *dst; | |
852 | EMACS_INT chars; | |
853 | int accept_latin_1; | |
854 | { | |
855 | EMACS_INT i; | |
856 | ||
857 | for (i = 0; i < chars; i++) | |
858 | { | |
859 | int c = STRING_CHAR_ADVANCE (src); | |
860 | ||
861 | if (CHAR_BYTE8_P (c)) | |
862 | c = CHAR_TO_BYTE8 (c); | |
863 | else if (! ASCII_CHAR_P (c) | |
864 | && (! accept_latin_1 || c >= 0x100)) | |
865 | return i; | |
866 | *dst++ = c; | |
867 | } | |
868 | return i; | |
869 | } | |
870 | ||
871 | ||
0168c3d8 KH |
872 | int |
873 | string_count_byte8 (string) | |
874 | Lisp_Object string; | |
875 | { | |
876 | int multibyte = STRING_MULTIBYTE (string); | |
8f924df7 KH |
877 | int nbytes = SBYTES (string); |
878 | unsigned char *p = SDATA (string); | |
0168c3d8 KH |
879 | unsigned char *pend = p + nbytes; |
880 | int count = 0; | |
881 | int c, len; | |
882 | ||
883 | if (multibyte) | |
884 | while (p < pend) | |
885 | { | |
886 | c = *p; | |
887 | len = BYTES_BY_CHAR_HEAD (c); | |
888 | ||
889 | if (CHAR_BYTE8_HEAD_P (c)) | |
890 | count++; | |
891 | p += len; | |
892 | } | |
893 | else | |
894 | while (p < pend) | |
895 | { | |
896 | if (*p++ >= 0x80) | |
897 | count++; | |
898 | } | |
899 | return count; | |
900 | } | |
901 | ||
902 | ||
903 | Lisp_Object | |
904 | string_escape_byte8 (string) | |
905 | Lisp_Object string; | |
906 | { | |
8f924df7 KH |
907 | int nchars = SCHARS (string); |
908 | int nbytes = SBYTES (string); | |
0168c3d8 KH |
909 | int multibyte = STRING_MULTIBYTE (string); |
910 | int byte8_count; | |
15843e6f KH |
911 | const unsigned char *src, *src_end; |
912 | unsigned char *dst; | |
0168c3d8 KH |
913 | Lisp_Object val; |
914 | int c, len; | |
915 | ||
916 | if (multibyte && nchars == nbytes) | |
917 | return string; | |
918 | ||
919 | byte8_count = string_count_byte8 (string); | |
920 | ||
921 | if (byte8_count == 0) | |
922 | return string; | |
923 | ||
924 | if (multibyte) | |
925 | /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
7b40ebaf | 926 | val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
0168c3d8 KH |
927 | nbytes + byte8_count * 2); |
928 | else | |
929 | /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
930 | val = make_uninit_string (nbytes + byte8_count * 3); | |
931 | ||
8f924df7 | 932 | src = SDATA (string); |
0168c3d8 | 933 | src_end = src + nbytes; |
8f924df7 | 934 | dst = SDATA (val); |
0168c3d8 KH |
935 | if (multibyte) |
936 | while (src < src_end) | |
937 | { | |
938 | c = *src; | |
939 | len = BYTES_BY_CHAR_HEAD (c); | |
940 | ||
941 | if (CHAR_BYTE8_HEAD_P (c)) | |
942 | { | |
943 | c = STRING_CHAR_ADVANCE (src); | |
944 | c = CHAR_TO_BYTE8 (c); | |
1889b238 | 945 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
946 | dst += 4; |
947 | } | |
948 | else | |
949 | while (len--) *dst++ = *src++; | |
950 | } | |
951 | else | |
952 | while (src < src_end) | |
953 | { | |
954 | c = *src++; | |
955 | if (c >= 0x80) | |
956 | { | |
1889b238 | 957 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
958 | dst += 4; |
959 | } | |
960 | else | |
961 | *dst++ = c; | |
962 | } | |
963 | return val; | |
964 | } | |
965 | ||
966 | \f | |
8f924df7 | 967 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 968 | doc: /* |
d2e83296 DL |
969 | Concatenate all the argument characters and make the result a string. |
970 | usage: (string &rest CHARACTERS) */) | |
0168c3d8 KH |
971 | (n, args) |
972 | int n; | |
973 | Lisp_Object *args; | |
974 | { | |
975 | int i; | |
976 | unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); | |
977 | unsigned char *p = buf; | |
978 | int c; | |
979 | ||
980 | for (i = 0; i < n; i++) | |
981 | { | |
982 | CHECK_CHARACTER (args[i]); | |
983 | c = XINT (args[i]); | |
984 | p += CHAR_STRING (c, p); | |
985 | } | |
986 | ||
987 | return make_string_from_bytes ((char *) buf, n, p - buf); | |
988 | } | |
989 | ||
70b4969d | 990 | DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0, |
87d6f965 KH |
991 | doc: /* Concatenate all the argument bytes and make the result a unibyte string. |
992 | usage: (unibyte-string &rest BYTES) */) | |
70b4969d KH |
993 | (n, args) |
994 | int n; | |
995 | Lisp_Object *args; | |
996 | { | |
997 | int i; | |
998 | unsigned char *buf = (unsigned char *) alloca (n); | |
999 | unsigned char *p = buf; | |
1000 | unsigned c; | |
1001 | ||
1002 | for (i = 0; i < n; i++) | |
1003 | { | |
1004 | CHECK_NATNUM (args[i]); | |
1005 | c = XFASTINT (args[i]); | |
1006 | if (c >= 256) | |
1007 | args_out_of_range_3 (args[i], make_number (0), make_number (255)); | |
1008 | *p++ = c; | |
1009 | } | |
1010 | ||
1011 | return make_string_from_bytes ((char *) buf, n, p - buf); | |
1012 | } | |
1013 | ||
c73ae4ae | 1014 | DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers, |
d0363d44 KH |
1015 | Schar_resolve_modifiers, 1, 1, 0, |
1016 | doc: /* Resolve modifiers in the character CHAR. | |
1017 | The value is a character with modifiers resolved into the character | |
1018 | code. Unresolved modifiers are kept in the value. | |
c73ae4ae | 1019 | usage: (char-resolve-modifiers CHAR) */) |
d0363d44 KH |
1020 | (character) |
1021 | Lisp_Object character; | |
1022 | { | |
1023 | int c; | |
1024 | ||
1025 | CHECK_NUMBER (character); | |
1026 | c = XINT (character); | |
1027 | return make_number (char_resolve_modifier_mask (c)); | |
1028 | } | |
1029 | ||
ee107a89 KH |
1030 | DEFUN ("get-byte", Fget_byte, Sget_byte, 0, 2, 0, |
1031 | doc: /* Return a byte value of a character at point. | |
1032 | Optional 1st arg POSITION, if non-nil, is a position of a character to get | |
1033 | a byte value. | |
1034 | Optional 2nd arg STRING, if non-nil, is a string of which first | |
1035 | character is a target to get a byte value. In this case, POSITION, if | |
1036 | non-nil, is an index of a target character in the string. | |
1037 | ||
1038 | If the current buffer (or STRING) is multibyte, and the target | |
1039 | character is not ASCII nor 8-bit character, an error is signalled. */) | |
1040 | (position, string) | |
1041 | Lisp_Object position, string; | |
1042 | { | |
1043 | int c; | |
1044 | EMACS_INT pos; | |
1045 | unsigned char *p; | |
1046 | ||
1047 | if (NILP (string)) | |
1048 | { | |
1049 | if (NILP (position)) | |
1050 | { | |
1051 | p = PT_ADDR; | |
1052 | } | |
1053 | else | |
1054 | { | |
1055 | CHECK_NUMBER_COERCE_MARKER (position); | |
1056 | if (XINT (position) < BEGV || XINT (position) >= ZV) | |
1057 | args_out_of_range_3 (position, make_number (BEGV), make_number (ZV)); | |
1058 | pos = XFASTINT (position); | |
1059 | p = CHAR_POS_ADDR (pos); | |
1060 | } | |
d5998e03 KH |
1061 | if (NILP (current_buffer->enable_multibyte_characters)) |
1062 | return make_number (*p); | |
ee107a89 KH |
1063 | } |
1064 | else | |
1065 | { | |
1066 | CHECK_STRING (string); | |
1067 | if (NILP (position)) | |
1068 | { | |
1069 | p = SDATA (string); | |
1070 | } | |
1071 | else | |
1072 | { | |
1073 | CHECK_NATNUM (position); | |
1074 | if (XINT (position) >= SCHARS (string)) | |
1075 | args_out_of_range (string, position); | |
1076 | pos = XFASTINT (position); | |
1077 | p = SDATA (string) + string_char_to_byte (string, pos); | |
1078 | } | |
d5998e03 KH |
1079 | if (! STRING_MULTIBYTE (string)) |
1080 | return make_number (*p); | |
ee107a89 KH |
1081 | } |
1082 | c = STRING_CHAR (p, 0); | |
1083 | if (CHAR_BYTE8_P (c)) | |
1084 | c = CHAR_TO_BYTE8 (c); | |
1085 | else if (! ASCII_CHAR_P (c)) | |
1086 | error ("Not an ASCII nor an 8-bit character: %d", c); | |
1087 | return make_number (c); | |
1088 | } | |
1089 | ||
1090 | ||
0168c3d8 KH |
1091 | void |
1092 | init_character_once () | |
1093 | { | |
1094 | } | |
1095 | ||
1096 | #ifdef emacs | |
1097 | ||
1098 | void | |
1099 | syms_of_character () | |
1100 | { | |
1101 | DEFSYM (Qcharacterp, "characterp"); | |
1102 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
1103 | ||
1104 | staticpro (&Vchar_unify_table); | |
1105 | Vchar_unify_table = Qnil; | |
1106 | ||
1107 | defsubr (&Smax_char); | |
1108 | defsubr (&Scharacterp); | |
1109 | defsubr (&Sunibyte_char_to_multibyte); | |
1110 | defsubr (&Smultibyte_char_to_unibyte); | |
1111 | defsubr (&Schar_bytes); | |
1112 | defsubr (&Schar_width); | |
1113 | defsubr (&Sstring_width); | |
1114 | defsubr (&Schar_direction); | |
0168c3d8 | 1115 | defsubr (&Sstring); |
70b4969d | 1116 | defsubr (&Sunibyte_string); |
d0363d44 | 1117 | defsubr (&Schar_resolve_modifiers); |
ee107a89 | 1118 | defsubr (&Sget_byte); |
0168c3d8 KH |
1119 | |
1120 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
1121 | doc: /* | |
68978cf0 DL |
1122 | Vector recording all translation tables ever defined. |
1123 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
1124 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
1125 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
1126 | ||
1127 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
1128 | doc: /* | |
1129 | A char-table for characters which invoke auto-filling. | |
1130 | Such characters have value t in this table. */); | |
1131 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
1132 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
1133 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 KH |
1134 | |
1135 | DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
1136 | doc: /* | |
1137 | A char-table for width (columns) of each character. */); | |
1138 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
1139 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
1140 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
1141 | make_number (4)); | |
0168c3d8 KH |
1142 | |
1143 | DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
1144 | doc: /* A char-table for direction of each character. */); | |
1145 | Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
1146 | ||
1147 | DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
1148 | doc: /* A char-table for each printable character. */); | |
db6d4189 | 1149 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
1150 | Fset_char_table_range (Vprintable_chars, |
1151 | Fcons (make_number (32), make_number (126)), Qt); | |
1152 | Fset_char_table_range (Vprintable_chars, | |
1153 | Fcons (make_number (160), | |
1154 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 1155 | |
c57f3328 KH |
1156 | DEFVAR_LISP ("char-script-table", &Vchar_script_table, |
1157 | doc: /* Char table of script symbols. | |
1158 | It has one extra slot whose value is a list of script symbols. */); | |
1159 | ||
1160 | /* Intern this now in case it isn't already done. | |
1161 | Setting this variable twice is harmless. | |
1162 | But don't staticpro it here--that is done in alloc.c. */ | |
1163 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1164 | DEFSYM (Qchar_script_table, "char-script-table"); | |
1165 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
1166 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
c7e14352 KH |
1167 | |
1168 | DEFVAR_LISP ("script-representative-chars", &Vscript_representative_chars, | |
c3bb7671 KH |
1169 | doc: /* Alist of scripts vs the representative characters. |
1170 | Each element is a cons (SCRIPT . CHARS), where SCRIPT is a script name symbol, | |
1171 | CHARS is a list or a vector of characters. | |
472a4dc9 | 1172 | If it is a list, all characters in the list are necessary for supporting SCRIPT. |
c3bb7671 KH |
1173 | If it is a vector, one of the characters in the vector is necessary. |
1174 | This variable is used to find a font for a specific script. */); | |
c7e14352 | 1175 | Vscript_representative_chars = Qnil; |
a3cbb631 KH |
1176 | |
1177 | DEFVAR_LISP ("unicode-category-table", &Vunicode_category_table, | |
1178 | doc: /* Char table of Unicode's "General Category". | |
472a4dc9 JB |
1179 | All Unicode characters have one of the following values (symbol): |
1180 | Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po, | |
a3cbb631 KH |
1181 | Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn |
1182 | See The Unicode Standard for the meaning of those values. */); | |
1183 | /* The correct char-table is setup in characters.el. */ | |
1184 | Vunicode_category_table = Qnil; | |
0168c3d8 KH |
1185 | } |
1186 | ||
1187 | #endif /* emacs */ | |
fbaf0946 MB |
1188 | |
1189 | /* arch-tag: b6665960-3c3d-4184-85cd-af4318197999 | |
1190 | (do not change this comment) */ |