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. |
76b6f707 | 4 | Copyright (C) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 |
ec62e0ac | 5 | Free Software Foundation, Inc. |
76b6f707 | 6 | Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009 |
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 | 61 | /* Char-table of information about which character to unify to which |
6f1b43a0 | 62 | Unicode character. Mainly used by the macro MAYBE_UNIFY_CHAR. */ |
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 | |
0168c3d8 KH |
93 | \f |
94 | ||
2bde7652 KH |
95 | /* If character code C has modifier masks, reflect them to the |
96 | character code if possible. Return the resulting code. */ | |
97 | ||
98 | int | |
99 | char_resolve_modifier_mask (c) | |
100 | int c; | |
101 | { | |
d0363d44 | 102 | /* A non-ASCII character can't reflect modifier bits to the code. */ |
2bde7652 KH |
103 | if (! ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) |
104 | return c; | |
105 | ||
106 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
2bde7652 KH |
107 | if (c & CHAR_SHIFT) |
108 | { | |
109 | /* Shift modifier is valid only with [A-Za-z]. */ | |
110 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
111 | c &= ~CHAR_SHIFT; | |
112 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
113 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
03365d0e KH |
114 | /* Shift modifier for control characters and SPC is ignored. */ |
115 | else if ((c & ~CHAR_MODIFIER_MASK) <= 0x20) | |
d0363d44 KH |
116 | c &= ~CHAR_SHIFT; |
117 | } | |
2bde7652 KH |
118 | if (c & CHAR_CTL) |
119 | { | |
120 | /* Simulate the code in lread.c. */ | |
121 | /* Allow `\C- ' and `\C-?'. */ | |
03365d0e KH |
122 | if ((c & 0377) == ' ') |
123 | c &= ~0177 & ~ CHAR_CTL; | |
124 | else if ((c & 0377) == '?') | |
125 | c = 0177 | (c & ~0177 & ~CHAR_CTL); | |
2bde7652 KH |
126 | /* ASCII control chars are made from letters (both cases), |
127 | as well as the non-letters within 0100...0137. */ | |
128 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
129 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
130 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
131 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
132 | } | |
03365d0e KH |
133 | if (c & CHAR_META) |
134 | { | |
135 | /* Move the meta bit to the right place for a string. */ | |
136 | c = (c & ~CHAR_META) | 0x80; | |
137 | } | |
2bde7652 KH |
138 | |
139 | return c; | |
140 | } | |
141 | ||
142 | ||
33f91981 KH |
143 | /* Store multibyte form of character C at P. If C has modifier bits, |
144 | handle them appropriately. */ | |
145 | ||
0168c3d8 | 146 | int |
e3d8eb8c | 147 | char_string (c, p) |
5aa91c9b | 148 | unsigned c; |
1889b238 | 149 | unsigned char *p; |
0168c3d8 KH |
150 | { |
151 | int bytes; | |
152 | ||
e3d8eb8c KH |
153 | if (c & CHAR_MODIFIER_MASK) |
154 | { | |
2bde7652 | 155 | c = (unsigned) char_resolve_modifier_mask ((int) c); |
e3d8eb8c KH |
156 | /* If C still has any modifier bits, just ignore it. */ |
157 | c &= ~CHAR_MODIFIER_MASK; | |
158 | } | |
159 | ||
0168c3d8 KH |
160 | MAYBE_UNIFY_CHAR (c); |
161 | ||
e3d8eb8c | 162 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
163 | { |
164 | bytes = CHAR_STRING (c, p); | |
165 | } | |
166 | else if (c <= MAX_4_BYTE_CHAR) | |
167 | { | |
168 | p[0] = (0xF0 | (c >> 18)); | |
169 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
170 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
171 | p[3] = (0x80 | (c & 0x3F)); | |
172 | bytes = 4; | |
173 | } | |
e3d8eb8c | 174 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
175 | { |
176 | p[0] = 0xF8; | |
177 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
178 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
179 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
180 | p[4] = (0x80 | (c & 0x3F)); | |
181 | bytes = 5; | |
182 | } | |
5aa91c9b | 183 | else if (c <= MAX_CHAR) |
e3d8eb8c KH |
184 | { |
185 | c = CHAR_TO_BYTE8 (c); | |
186 | bytes = BYTE8_STRING (c, p); | |
187 | } | |
5aa91c9b KH |
188 | else |
189 | error ("Invalid character: %d", c); | |
1889b238 | 190 | |
0168c3d8 KH |
191 | return bytes; |
192 | } | |
193 | ||
194 | ||
33f91981 KH |
195 | /* Return a character whose multibyte form is at P. Set LEN is not |
196 | NULL, it must be a pointer to integer. In that case, set *LEN to | |
197 | the byte length of the multibyte form. If ADVANCED is not NULL, is | |
198 | must be a pointer to unsigned char. In that case, set *ADVANCED to | |
199 | the ending address (i.e. the starting address of the next | |
200 | character) of the multibyte form. */ | |
201 | ||
0168c3d8 | 202 | int |
e3d8eb8c | 203 | string_char (p, advanced, len) |
15843e6f KH |
204 | const unsigned char *p; |
205 | const unsigned char **advanced; | |
0168c3d8 KH |
206 | int *len; |
207 | { | |
1889b238 | 208 | int c; |
15843e6f | 209 | const unsigned char *saved_p = p; |
0168c3d8 KH |
210 | |
211 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
212 | { | |
213 | c = STRING_CHAR_ADVANCE (p); | |
214 | } | |
215 | else if (! (*p & 0x08)) | |
216 | { | |
217 | c = ((((p)[0] & 0xF) << 18) | |
218 | | (((p)[1] & 0x3F) << 12) | |
219 | | (((p)[2] & 0x3F) << 6) | |
220 | | ((p)[3] & 0x3F)); | |
221 | p += 4; | |
222 | } | |
223 | else | |
224 | { | |
225 | c = ((((p)[1] & 0x3F) << 18) | |
226 | | (((p)[2] & 0x3F) << 12) | |
227 | | (((p)[3] & 0x3F) << 6) | |
228 | | ((p)[4] & 0x3F)); | |
229 | p += 5; | |
230 | } | |
231 | ||
232 | MAYBE_UNIFY_CHAR (c); | |
233 | ||
234 | if (len) | |
235 | *len = p - saved_p; | |
236 | if (advanced) | |
237 | *advanced = p; | |
238 | return c; | |
239 | } | |
240 | ||
241 | ||
242 | /* Translate character C by translation table TABLE. If C is | |
243 | negative, translate a character specified by CHARSET and CODE. If | |
244 | no translation is found in TABLE, return the untranslated | |
10453be9 KH |
245 | character. If TABLE is a list, elements are char tables. In this |
246 | case, translace C by all tables. */ | |
0168c3d8 KH |
247 | |
248 | int | |
249 | translate_char (table, c) | |
250 | Lisp_Object table; | |
251 | int c; | |
252 | { | |
10453be9 KH |
253 | if (CHAR_TABLE_P (table)) |
254 | { | |
255 | Lisp_Object ch; | |
256 | ||
257 | ch = CHAR_TABLE_REF (table, c); | |
258 | if (CHARACTERP (ch)) | |
259 | c = XINT (ch); | |
260 | } | |
261 | else | |
262 | { | |
263 | for (; CONSP (table); table = XCDR (table)) | |
264 | c = translate_char (XCAR (table), c); | |
265 | } | |
266 | return c; | |
0168c3d8 KH |
267 | } |
268 | ||
2e5db15c KH |
269 | /* Convert ASCII or 8-bit character C to unibyte. If C is none of |
270 | them, return (C & 0xFF). | |
0168c3d8 KH |
271 | |
272 | The argument REV_TBL is now ignored. It will be removed in the | |
273 | future. */ | |
274 | ||
275 | int | |
276 | multibyte_char_to_unibyte (c, rev_tbl) | |
277 | int c; | |
278 | Lisp_Object rev_tbl; | |
279 | { | |
2e5db15c KH |
280 | if (c < 0x80) |
281 | return c; | |
b672c5ae KH |
282 | if (CHAR_BYTE8_P (c)) |
283 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 284 | return (c & 0xFF); |
0168c3d8 KH |
285 | } |
286 | ||
935d5b02 KH |
287 | /* Like multibyte_char_to_unibyte, but return -1 if C is not supported |
288 | by charset_unibyte. */ | |
289 | ||
290 | int | |
291 | multibyte_char_to_unibyte_safe (c) | |
292 | int c; | |
293 | { | |
2e5db15c KH |
294 | if (c < 0x80) |
295 | return c; | |
935d5b02 KH |
296 | if (CHAR_BYTE8_P (c)) |
297 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 298 | return -1; |
935d5b02 | 299 | } |
0168c3d8 KH |
300 | |
301 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
302 | doc: /* Return non-nil if OBJECT is a character. */) | |
303 | (object, ignore) | |
304 | Lisp_Object object, ignore; | |
305 | { | |
306 | return (CHARACTERP (object) ? Qt : Qnil); | |
307 | } | |
308 | ||
309 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
310 | doc: /* Return the character of the maximum code. */) | |
311 | () | |
312 | { | |
313 | return make_number (MAX_CHAR); | |
314 | } | |
315 | ||
316 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
317 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
5556875b | 318 | doc: /* Convert the byte CH to multibyte character. */) |
0168c3d8 KH |
319 | (ch) |
320 | Lisp_Object ch; | |
321 | { | |
322 | int c; | |
0168c3d8 KH |
323 | |
324 | CHECK_CHARACTER (ch); | |
325 | c = XFASTINT (ch); | |
2e5db15c KH |
326 | if (c >= 0x100) |
327 | error ("Not a unibyte character: %d", c); | |
328 | if (c >= 0x80) | |
329 | c = BYTE8_TO_CHAR (c); | |
0168c3d8 KH |
330 | return make_number (c); |
331 | } | |
332 | ||
333 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
334 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
5556875b SM |
335 | doc: /* Convert the multibyte character CH to a byte. |
336 | If the multibyte character does not represent a byte, return -1. */) | |
0168c3d8 KH |
337 | (ch) |
338 | Lisp_Object ch; | |
339 | { | |
5556875b | 340 | int cm; |
0168c3d8 KH |
341 | |
342 | CHECK_CHARACTER (ch); | |
5556875b SM |
343 | cm = XFASTINT (ch); |
344 | if (cm < 256) | |
345 | /* Can't distinguish a byte read from a unibyte buffer from | |
346 | a latin1 char, so let's let it slide. */ | |
347 | return ch; | |
348 | else | |
349 | { | |
2afc21f5 | 350 | int cu = CHAR_TO_BYTE_SAFE (cm); |
5556875b SM |
351 | return make_number (cu); |
352 | } | |
0168c3d8 KH |
353 | } |
354 | ||
355 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
356 | doc: /* Return 1 regardless of the argument CHAR. | |
add553ac JB |
357 | This is now an obsolete function. We keep it just for backward compatibility. |
358 | usage: (char-bytes CHAR) */) | |
0168c3d8 KH |
359 | (ch) |
360 | Lisp_Object ch; | |
361 | { | |
362 | CHECK_CHARACTER (ch); | |
363 | return make_number (1); | |
364 | } | |
365 | ||
366 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
367 | doc: /* Return width of CHAR when displayed in the current buffer. | |
368 | The width is measured by how many columns it occupies on the screen. | |
add553ac JB |
369 | Tab is taken to occupy `tab-width' columns. |
370 | usage: (char-width CHAR) */) | |
0168c3d8 KH |
371 | (ch) |
372 | Lisp_Object ch; | |
373 | { | |
374 | Lisp_Object disp; | |
375 | int c, width; | |
376 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
377 | ||
378 | CHECK_CHARACTER (ch); | |
379 | c = XINT (ch); | |
380 | ||
381 | /* Get the way the display table would display it. */ | |
382 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
383 | ||
384 | if (VECTORP (disp)) | |
385 | width = ASIZE (disp); | |
386 | else | |
387 | width = CHAR_WIDTH (c); | |
388 | ||
389 | return make_number (width); | |
390 | } | |
391 | ||
0168c3d8 KH |
392 | /* Return width of string STR of length LEN when displayed in the |
393 | current buffer. The width is measured by how many columns it | |
394 | occupies on the screen. If PRECISION > 0, return the width of | |
395 | longest substring that doesn't exceed PRECISION, and set number of | |
396 | characters and bytes of the substring in *NCHARS and *NBYTES | |
397 | respectively. */ | |
398 | ||
1889b238 | 399 | int |
63c125ab | 400 | c_string_width (const unsigned char *str, int len, int precision, int *nchars, int *nbytes) |
0168c3d8 KH |
401 | { |
402 | int i = 0, i_byte = 0; | |
403 | int width = 0; | |
404 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
405 | ||
406 | while (i_byte < len) | |
407 | { | |
408 | int bytes, thiswidth; | |
409 | Lisp_Object val; | |
410 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
411 | ||
412 | if (dp) | |
413 | { | |
414 | val = DISP_CHAR_VECTOR (dp, c); | |
415 | if (VECTORP (val)) | |
416 | thiswidth = XVECTOR (val)->size; | |
417 | else | |
418 | thiswidth = CHAR_WIDTH (c); | |
419 | } | |
420 | else | |
421 | { | |
422 | thiswidth = CHAR_WIDTH (c); | |
423 | } | |
424 | ||
425 | if (precision > 0 | |
426 | && (width + thiswidth > precision)) | |
427 | { | |
428 | *nchars = i; | |
429 | *nbytes = i_byte; | |
430 | return width; | |
431 | } | |
432 | i++; | |
433 | i_byte += bytes; | |
434 | width += thiswidth; | |
435 | } | |
436 | ||
437 | if (precision > 0) | |
438 | { | |
439 | *nchars = i; | |
440 | *nbytes = i_byte; | |
441 | } | |
442 | ||
443 | return width; | |
444 | } | |
445 | ||
1889b238 KH |
446 | /* Return width of string STR of length LEN when displayed in the |
447 | current buffer. The width is measured by how many columns it | |
448 | occupies on the screen. */ | |
449 | ||
450 | int | |
451 | strwidth (str, len) | |
452 | unsigned char *str; | |
453 | int len; | |
454 | { | |
455 | return c_string_width (str, len, -1, NULL, NULL); | |
456 | } | |
457 | ||
0168c3d8 KH |
458 | /* Return width of Lisp string STRING when displayed in the current |
459 | buffer. The width is measured by how many columns it occupies on | |
460 | the screen while paying attention to compositions. If PRECISION > | |
461 | 0, return the width of longest substring that doesn't exceed | |
462 | PRECISION, and set number of characters and bytes of the substring | |
463 | in *NCHARS and *NBYTES respectively. */ | |
464 | ||
465 | int | |
466 | lisp_string_width (string, precision, nchars, nbytes) | |
467 | Lisp_Object string; | |
468 | int precision, *nchars, *nbytes; | |
469 | { | |
8f924df7 | 470 | int len = SCHARS (string); |
0aee65b9 KH |
471 | /* This set multibyte to 0 even if STRING is multibyte when it |
472 | contains only ascii and eight-bit-graphic, but that's | |
473 | intentional. */ | |
474 | int multibyte = len < SBYTES (string); | |
8f924df7 | 475 | unsigned char *str = SDATA (string); |
0168c3d8 KH |
476 | int i = 0, i_byte = 0; |
477 | int width = 0; | |
478 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
479 | ||
480 | while (i < len) | |
481 | { | |
482 | int chars, bytes, thiswidth; | |
483 | Lisp_Object val; | |
484 | int cmp_id; | |
f4bc0685 | 485 | EMACS_INT ignore, end; |
0168c3d8 KH |
486 | |
487 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
488 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
489 | >= 0)) | |
490 | { | |
491 | thiswidth = composition_table[cmp_id]->width; | |
492 | chars = end - i; | |
493 | bytes = string_char_to_byte (string, end) - i_byte; | |
494 | } | |
0168c3d8 KH |
495 | else |
496 | { | |
0aee65b9 | 497 | int c; |
0168c3d8 | 498 | |
0aee65b9 KH |
499 | if (multibyte) |
500 | c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
501 | else | |
502 | c = str[i_byte], bytes = 1; | |
0168c3d8 | 503 | chars = 1; |
0aee65b9 KH |
504 | if (dp) |
505 | { | |
506 | val = DISP_CHAR_VECTOR (dp, c); | |
507 | if (VECTORP (val)) | |
508 | thiswidth = XVECTOR (val)->size; | |
509 | else | |
510 | thiswidth = CHAR_WIDTH (c); | |
511 | } | |
512 | else | |
513 | { | |
514 | thiswidth = CHAR_WIDTH (c); | |
515 | } | |
0168c3d8 KH |
516 | } |
517 | ||
518 | if (precision > 0 | |
519 | && (width + thiswidth > precision)) | |
520 | { | |
521 | *nchars = i; | |
522 | *nbytes = i_byte; | |
523 | return width; | |
524 | } | |
525 | i += chars; | |
526 | i_byte += bytes; | |
527 | width += thiswidth; | |
528 | } | |
529 | ||
530 | if (precision > 0) | |
531 | { | |
532 | *nchars = i; | |
533 | *nbytes = i_byte; | |
534 | } | |
535 | ||
536 | return width; | |
537 | } | |
538 | ||
539 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
540 | doc: /* Return width of STRING when displayed in the current buffer. | |
541 | Width is measured by how many columns it occupies on the screen. | |
542 | When calculating width of a multibyte character in STRING, | |
543 | only the base leading-code is considered; the validity of | |
544 | the following bytes is not checked. Tabs in STRING are always | |
add553ac JB |
545 | taken to occupy `tab-width' columns. |
546 | usage: (string-width STRING) */) | |
0168c3d8 KH |
547 | (str) |
548 | Lisp_Object str; | |
549 | { | |
550 | Lisp_Object val; | |
551 | ||
552 | CHECK_STRING (str); | |
553 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
554 | return val; | |
555 | } | |
556 | ||
557 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
558 | doc: /* Return the direction of CHAR. | |
add553ac JB |
559 | The returned value is 0 for left-to-right and 1 for right-to-left. |
560 | usage: (char-direction CHAR) */) | |
0168c3d8 KH |
561 | (ch) |
562 | Lisp_Object ch; | |
563 | { | |
564 | int c; | |
565 | ||
566 | CHECK_CHARACTER (ch); | |
567 | c = XINT (ch); | |
568 | return CHAR_TABLE_REF (Vchar_direction_table, c); | |
569 | } | |
570 | ||
0168c3d8 KH |
571 | /* Return the number of characters in the NBYTES bytes at PTR. |
572 | This works by looking at the contents and checking for multibyte | |
573 | sequences while assuming that there's no invalid sequence. | |
574 | However, if the current buffer has enable-multibyte-characters = | |
575 | nil, we treat each byte as a character. */ | |
576 | ||
13818c30 | 577 | EMACS_INT |
0168c3d8 | 578 | chars_in_text (ptr, nbytes) |
8f924df7 | 579 | const unsigned char *ptr; |
13818c30 | 580 | EMACS_INT nbytes; |
0168c3d8 KH |
581 | { |
582 | /* current_buffer is null at early stages of Emacs initialization. */ | |
583 | if (current_buffer == 0 | |
584 | || NILP (current_buffer->enable_multibyte_characters)) | |
585 | return nbytes; | |
586 | ||
587 | return multibyte_chars_in_text (ptr, nbytes); | |
588 | } | |
589 | ||
590 | /* Return the number of characters in the NBYTES bytes at PTR. | |
591 | This works by looking at the contents and checking for multibyte | |
592 | sequences while assuming that there's no invalid sequence. It | |
593 | ignores enable-multibyte-characters. */ | |
594 | ||
13818c30 | 595 | EMACS_INT |
0168c3d8 | 596 | multibyte_chars_in_text (ptr, nbytes) |
8f924df7 | 597 | const unsigned char *ptr; |
13818c30 | 598 | EMACS_INT nbytes; |
0168c3d8 | 599 | { |
8f924df7 | 600 | const unsigned char *endp = ptr + nbytes; |
0168c3d8 KH |
601 | int chars = 0; |
602 | ||
603 | while (ptr < endp) | |
604 | { | |
605 | int len = MULTIBYTE_LENGTH (ptr, endp); | |
606 | ||
607 | if (len == 0) | |
608 | abort (); | |
609 | ptr += len; | |
610 | chars++; | |
611 | } | |
612 | ||
613 | return chars; | |
614 | } | |
615 | ||
616 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
617 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
618 | respectively. On counting bytes, pay attention to that 8-bit | |
619 | characters not constructing a valid multibyte sequence are | |
620 | represented by 2-byte in a multibyte text. */ | |
621 | ||
622 | void | |
623 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
8f924df7 | 624 | const unsigned char *str; |
0168c3d8 KH |
625 | int len, *nchars, *nbytes; |
626 | { | |
8f924df7 | 627 | const unsigned char *endp = str + len; |
0168c3d8 KH |
628 | int n, chars = 0, bytes = 0; |
629 | ||
630 | if (len >= MAX_MULTIBYTE_LENGTH) | |
631 | { | |
8f924df7 | 632 | const unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; |
0168c3d8 KH |
633 | while (str < adjusted_endp) |
634 | { | |
635 | if ((n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
636 | str += n, bytes += n; | |
637 | else | |
638 | str++, bytes += 2; | |
639 | chars++; | |
640 | } | |
641 | } | |
642 | while (str < endp) | |
643 | { | |
644 | if ((n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
645 | str += n, bytes += n; | |
646 | else | |
647 | str++, bytes += 2; | |
648 | chars++; | |
649 | } | |
650 | ||
651 | *nchars = chars; | |
652 | *nbytes = bytes; | |
653 | return; | |
654 | } | |
655 | ||
656 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
657 | It actually converts only such 8-bit characters that don't contruct | |
658 | a multibyte sequence to multibyte forms of Latin-1 characters. If | |
659 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
660 | text. It is assured that we can use LEN bytes at STR as a work | |
661 | area and that is enough. Return the number of bytes of the | |
662 | resulting text. */ | |
663 | ||
664 | int | |
665 | str_as_multibyte (str, len, nbytes, nchars) | |
666 | unsigned char *str; | |
667 | int len, nbytes, *nchars; | |
668 | { | |
669 | unsigned char *p = str, *endp = str + nbytes; | |
670 | unsigned char *to; | |
671 | int chars = 0; | |
672 | int n; | |
673 | ||
674 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
675 | { | |
676 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
677 | while (p < adjusted_endp | |
678 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
679 | p += n, chars++; | |
680 | } | |
681 | while ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
682 | p += n, chars++; | |
683 | if (nchars) | |
684 | *nchars = chars; | |
685 | if (p == endp) | |
686 | return nbytes; | |
687 | ||
688 | to = p; | |
689 | nbytes = endp - p; | |
690 | endp = str + len; | |
691 | safe_bcopy ((char *) p, (char *) (endp - nbytes), nbytes); | |
692 | p = endp - nbytes; | |
693 | ||
694 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
695 | { | |
696 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
697 | while (p < adjusted_endp) | |
698 | { | |
699 | if ((n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
700 | { | |
701 | while (n--) | |
702 | *to++ = *p++; | |
703 | } | |
704 | else | |
705 | { | |
706 | int c = *p++; | |
707 | c = BYTE8_TO_CHAR (c); | |
708 | to += CHAR_STRING (c, to); | |
709 | } | |
710 | } | |
711 | chars++; | |
712 | } | |
713 | while (p < endp) | |
714 | { | |
715 | if ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
716 | { | |
717 | while (n--) | |
718 | *to++ = *p++; | |
8f924df7 | 719 | } |
0168c3d8 KH |
720 | else |
721 | { | |
722 | int c = *p++; | |
723 | c = BYTE8_TO_CHAR (c); | |
724 | to += CHAR_STRING (c, to); | |
725 | } | |
726 | chars++; | |
727 | } | |
728 | if (nchars) | |
729 | *nchars = chars; | |
730 | return (to - str); | |
731 | } | |
732 | ||
733 | /* Parse unibyte string at STR of LEN bytes, and return the number of | |
734 | bytes it may ocupy when converted to multibyte string by | |
735 | `str_to_multibyte'. */ | |
736 | ||
737 | int | |
738 | parse_str_to_multibyte (str, len) | |
739 | unsigned char *str; | |
740 | int len; | |
741 | { | |
742 | unsigned char *endp = str + len; | |
743 | int bytes; | |
744 | ||
745 | for (bytes = 0; str < endp; str++) | |
746 | bytes += (*str < 0x80) ? 1 : 2; | |
747 | return bytes; | |
748 | } | |
749 | ||
750 | ||
751 | /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
752 | that contains the same single-byte characters. It actually | |
753 | converts all 8-bit characters to multibyte forms. It is assured | |
754 | that we can use LEN bytes at STR as a work area and that is | |
755 | enough. */ | |
756 | ||
757 | int | |
758 | str_to_multibyte (str, len, bytes) | |
759 | unsigned char *str; | |
760 | int len, bytes; | |
761 | { | |
762 | unsigned char *p = str, *endp = str + bytes; | |
763 | unsigned char *to; | |
764 | ||
765 | while (p < endp && *p < 0x80) p++; | |
766 | if (p == endp) | |
767 | return bytes; | |
768 | to = p; | |
769 | bytes = endp - p; | |
770 | endp = str + len; | |
771 | safe_bcopy ((char *) p, (char *) (endp - bytes), bytes); | |
772 | p = endp - bytes; | |
8f924df7 | 773 | while (p < endp) |
0168c3d8 KH |
774 | { |
775 | int c = *p++; | |
776 | ||
777 | if (c >= 0x80) | |
778 | c = BYTE8_TO_CHAR (c); | |
779 | to += CHAR_STRING (c, to); | |
780 | } | |
781 | return (to - str); | |
782 | } | |
783 | ||
784 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
785 | actually converts characters in the range 0x80..0xFF to | |
786 | unibyte. */ | |
787 | ||
788 | int | |
789 | str_as_unibyte (str, bytes) | |
790 | unsigned char *str; | |
791 | int bytes; | |
792 | { | |
15843e6f KH |
793 | const unsigned char *p = str, *endp = str + bytes; |
794 | unsigned char *to; | |
0168c3d8 KH |
795 | int c, len; |
796 | ||
797 | while (p < endp) | |
798 | { | |
799 | c = *p; | |
800 | len = BYTES_BY_CHAR_HEAD (c); | |
801 | if (CHAR_BYTE8_HEAD_P (c)) | |
802 | break; | |
803 | p += len; | |
804 | } | |
15843e6f | 805 | to = str + (p - str); |
8f924df7 | 806 | while (p < endp) |
0168c3d8 KH |
807 | { |
808 | c = *p; | |
809 | len = BYTES_BY_CHAR_HEAD (c); | |
810 | if (CHAR_BYTE8_HEAD_P (c)) | |
811 | { | |
812 | c = STRING_CHAR_ADVANCE (p); | |
813 | *to++ = CHAR_TO_BYTE8 (c); | |
814 | } | |
815 | else | |
816 | { | |
817 | while (len--) *to++ = *p++; | |
818 | } | |
819 | } | |
820 | return (to - str); | |
821 | } | |
822 | ||
4aa40bb8 KH |
823 | /* Convert eight-bit chars in SRC (in multibyte form) to the |
824 | corresponding byte and store in DST. CHARS is the number of | |
825 | characters in SRC. The value is the number of bytes stored in DST. | |
826 | Usually, the value is the same as CHARS, but is less than it if SRC | |
827 | contains a non-ASCII, non-eight-bit characater. If ACCEPT_LATIN_1 | |
828 | is nonzero, a Latin-1 character is accepted and converted to a byte | |
f27f70ec KH |
829 | of that character code. |
830 | Note: Currently the arg ACCEPT_LATIN_1 is not used. */ | |
4aa40bb8 KH |
831 | |
832 | EMACS_INT | |
833 | str_to_unibyte (src, dst, chars, accept_latin_1) | |
834 | const unsigned char *src; | |
835 | unsigned char *dst; | |
836 | EMACS_INT chars; | |
837 | int accept_latin_1; | |
838 | { | |
839 | EMACS_INT i; | |
840 | ||
841 | for (i = 0; i < chars; i++) | |
842 | { | |
843 | int c = STRING_CHAR_ADVANCE (src); | |
844 | ||
845 | if (CHAR_BYTE8_P (c)) | |
846 | c = CHAR_TO_BYTE8 (c); | |
847 | else if (! ASCII_CHAR_P (c) | |
848 | && (! accept_latin_1 || c >= 0x100)) | |
849 | return i; | |
850 | *dst++ = c; | |
851 | } | |
852 | return i; | |
853 | } | |
854 | ||
855 | ||
0168c3d8 KH |
856 | int |
857 | string_count_byte8 (string) | |
858 | Lisp_Object string; | |
859 | { | |
860 | int multibyte = STRING_MULTIBYTE (string); | |
8f924df7 KH |
861 | int nbytes = SBYTES (string); |
862 | unsigned char *p = SDATA (string); | |
0168c3d8 KH |
863 | unsigned char *pend = p + nbytes; |
864 | int count = 0; | |
865 | int c, len; | |
866 | ||
867 | if (multibyte) | |
868 | while (p < pend) | |
869 | { | |
870 | c = *p; | |
871 | len = BYTES_BY_CHAR_HEAD (c); | |
872 | ||
873 | if (CHAR_BYTE8_HEAD_P (c)) | |
874 | count++; | |
875 | p += len; | |
876 | } | |
877 | else | |
878 | while (p < pend) | |
879 | { | |
880 | if (*p++ >= 0x80) | |
881 | count++; | |
882 | } | |
883 | return count; | |
884 | } | |
885 | ||
886 | ||
887 | Lisp_Object | |
888 | string_escape_byte8 (string) | |
889 | Lisp_Object string; | |
890 | { | |
8f924df7 KH |
891 | int nchars = SCHARS (string); |
892 | int nbytes = SBYTES (string); | |
0168c3d8 KH |
893 | int multibyte = STRING_MULTIBYTE (string); |
894 | int byte8_count; | |
15843e6f KH |
895 | const unsigned char *src, *src_end; |
896 | unsigned char *dst; | |
0168c3d8 KH |
897 | Lisp_Object val; |
898 | int c, len; | |
899 | ||
900 | if (multibyte && nchars == nbytes) | |
901 | return string; | |
902 | ||
903 | byte8_count = string_count_byte8 (string); | |
904 | ||
905 | if (byte8_count == 0) | |
906 | return string; | |
907 | ||
908 | if (multibyte) | |
909 | /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
7b40ebaf | 910 | val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
0168c3d8 KH |
911 | nbytes + byte8_count * 2); |
912 | else | |
913 | /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
914 | val = make_uninit_string (nbytes + byte8_count * 3); | |
915 | ||
8f924df7 | 916 | src = SDATA (string); |
0168c3d8 | 917 | src_end = src + nbytes; |
8f924df7 | 918 | dst = SDATA (val); |
0168c3d8 KH |
919 | if (multibyte) |
920 | while (src < src_end) | |
921 | { | |
922 | c = *src; | |
923 | len = BYTES_BY_CHAR_HEAD (c); | |
924 | ||
925 | if (CHAR_BYTE8_HEAD_P (c)) | |
926 | { | |
927 | c = STRING_CHAR_ADVANCE (src); | |
928 | c = CHAR_TO_BYTE8 (c); | |
1889b238 | 929 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
930 | dst += 4; |
931 | } | |
932 | else | |
933 | while (len--) *dst++ = *src++; | |
934 | } | |
935 | else | |
936 | while (src < src_end) | |
937 | { | |
938 | c = *src++; | |
939 | if (c >= 0x80) | |
940 | { | |
1889b238 | 941 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
942 | dst += 4; |
943 | } | |
944 | else | |
945 | *dst++ = c; | |
946 | } | |
947 | return val; | |
948 | } | |
949 | ||
950 | \f | |
8f924df7 | 951 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 952 | doc: /* |
d2e83296 DL |
953 | Concatenate all the argument characters and make the result a string. |
954 | usage: (string &rest CHARACTERS) */) | |
0168c3d8 KH |
955 | (n, args) |
956 | int n; | |
957 | Lisp_Object *args; | |
958 | { | |
959 | int i; | |
960 | unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); | |
961 | unsigned char *p = buf; | |
962 | int c; | |
963 | ||
964 | for (i = 0; i < n; i++) | |
965 | { | |
966 | CHECK_CHARACTER (args[i]); | |
967 | c = XINT (args[i]); | |
968 | p += CHAR_STRING (c, p); | |
969 | } | |
970 | ||
971 | return make_string_from_bytes ((char *) buf, n, p - buf); | |
972 | } | |
973 | ||
70b4969d | 974 | DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0, |
87d6f965 KH |
975 | doc: /* Concatenate all the argument bytes and make the result a unibyte string. |
976 | usage: (unibyte-string &rest BYTES) */) | |
70b4969d KH |
977 | (n, args) |
978 | int n; | |
979 | Lisp_Object *args; | |
980 | { | |
981 | int i; | |
982 | unsigned char *buf = (unsigned char *) alloca (n); | |
983 | unsigned char *p = buf; | |
984 | unsigned c; | |
985 | ||
986 | for (i = 0; i < n; i++) | |
987 | { | |
988 | CHECK_NATNUM (args[i]); | |
989 | c = XFASTINT (args[i]); | |
990 | if (c >= 256) | |
991 | args_out_of_range_3 (args[i], make_number (0), make_number (255)); | |
992 | *p++ = c; | |
993 | } | |
994 | ||
995 | return make_string_from_bytes ((char *) buf, n, p - buf); | |
996 | } | |
997 | ||
c73ae4ae | 998 | DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers, |
d0363d44 KH |
999 | Schar_resolve_modifiers, 1, 1, 0, |
1000 | doc: /* Resolve modifiers in the character CHAR. | |
1001 | The value is a character with modifiers resolved into the character | |
1002 | code. Unresolved modifiers are kept in the value. | |
c73ae4ae | 1003 | usage: (char-resolve-modifiers CHAR) */) |
d0363d44 KH |
1004 | (character) |
1005 | Lisp_Object character; | |
1006 | { | |
1007 | int c; | |
1008 | ||
1009 | CHECK_NUMBER (character); | |
1010 | c = XINT (character); | |
1011 | return make_number (char_resolve_modifier_mask (c)); | |
1012 | } | |
1013 | ||
ee107a89 KH |
1014 | DEFUN ("get-byte", Fget_byte, Sget_byte, 0, 2, 0, |
1015 | doc: /* Return a byte value of a character at point. | |
1016 | Optional 1st arg POSITION, if non-nil, is a position of a character to get | |
1017 | a byte value. | |
1018 | Optional 2nd arg STRING, if non-nil, is a string of which first | |
1019 | character is a target to get a byte value. In this case, POSITION, if | |
1020 | non-nil, is an index of a target character in the string. | |
1021 | ||
1022 | If the current buffer (or STRING) is multibyte, and the target | |
1023 | character is not ASCII nor 8-bit character, an error is signalled. */) | |
1024 | (position, string) | |
1025 | Lisp_Object position, string; | |
1026 | { | |
1027 | int c; | |
1028 | EMACS_INT pos; | |
1029 | unsigned char *p; | |
1030 | ||
1031 | if (NILP (string)) | |
1032 | { | |
1033 | if (NILP (position)) | |
1034 | { | |
1035 | p = PT_ADDR; | |
4d8e170e | 1036 | } |
ee107a89 KH |
1037 | else |
1038 | { | |
1039 | CHECK_NUMBER_COERCE_MARKER (position); | |
1040 | if (XINT (position) < BEGV || XINT (position) >= ZV) | |
1041 | args_out_of_range_3 (position, make_number (BEGV), make_number (ZV)); | |
1042 | pos = XFASTINT (position); | |
1043 | p = CHAR_POS_ADDR (pos); | |
1044 | } | |
d5998e03 KH |
1045 | if (NILP (current_buffer->enable_multibyte_characters)) |
1046 | return make_number (*p); | |
ee107a89 KH |
1047 | } |
1048 | else | |
1049 | { | |
1050 | CHECK_STRING (string); | |
1051 | if (NILP (position)) | |
1052 | { | |
1053 | p = SDATA (string); | |
1054 | } | |
1055 | else | |
1056 | { | |
1057 | CHECK_NATNUM (position); | |
1058 | if (XINT (position) >= SCHARS (string)) | |
1059 | args_out_of_range (string, position); | |
1060 | pos = XFASTINT (position); | |
1061 | p = SDATA (string) + string_char_to_byte (string, pos); | |
1062 | } | |
d5998e03 KH |
1063 | if (! STRING_MULTIBYTE (string)) |
1064 | return make_number (*p); | |
ee107a89 KH |
1065 | } |
1066 | c = STRING_CHAR (p, 0); | |
1067 | if (CHAR_BYTE8_P (c)) | |
1068 | c = CHAR_TO_BYTE8 (c); | |
1069 | else if (! ASCII_CHAR_P (c)) | |
1070 | error ("Not an ASCII nor an 8-bit character: %d", c); | |
1071 | return make_number (c); | |
1072 | } | |
1073 | ||
1074 | ||
0168c3d8 KH |
1075 | void |
1076 | init_character_once () | |
1077 | { | |
1078 | } | |
1079 | ||
1080 | #ifdef emacs | |
1081 | ||
1082 | void | |
1083 | syms_of_character () | |
1084 | { | |
1085 | DEFSYM (Qcharacterp, "characterp"); | |
1086 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
1087 | ||
1088 | staticpro (&Vchar_unify_table); | |
1089 | Vchar_unify_table = Qnil; | |
1090 | ||
1091 | defsubr (&Smax_char); | |
1092 | defsubr (&Scharacterp); | |
1093 | defsubr (&Sunibyte_char_to_multibyte); | |
1094 | defsubr (&Smultibyte_char_to_unibyte); | |
1095 | defsubr (&Schar_bytes); | |
1096 | defsubr (&Schar_width); | |
1097 | defsubr (&Sstring_width); | |
1098 | defsubr (&Schar_direction); | |
0168c3d8 | 1099 | defsubr (&Sstring); |
70b4969d | 1100 | defsubr (&Sunibyte_string); |
d0363d44 | 1101 | defsubr (&Schar_resolve_modifiers); |
ee107a89 | 1102 | defsubr (&Sget_byte); |
0168c3d8 KH |
1103 | |
1104 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
1105 | doc: /* | |
68978cf0 DL |
1106 | Vector recording all translation tables ever defined. |
1107 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
1108 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
1109 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
1110 | ||
1111 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
1112 | doc: /* | |
1113 | A char-table for characters which invoke auto-filling. | |
1114 | Such characters have value t in this table. */); | |
1115 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
1116 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
1117 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 KH |
1118 | |
1119 | DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
1120 | doc: /* | |
1121 | A char-table for width (columns) of each character. */); | |
1122 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
1123 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
1124 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
1125 | make_number (4)); | |
0168c3d8 KH |
1126 | |
1127 | DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
1128 | doc: /* A char-table for direction of each character. */); | |
1129 | Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
1130 | ||
1131 | DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
1132 | doc: /* A char-table for each printable character. */); | |
db6d4189 | 1133 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
1134 | Fset_char_table_range (Vprintable_chars, |
1135 | Fcons (make_number (32), make_number (126)), Qt); | |
1136 | Fset_char_table_range (Vprintable_chars, | |
1137 | Fcons (make_number (160), | |
1138 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 1139 | |
c57f3328 KH |
1140 | DEFVAR_LISP ("char-script-table", &Vchar_script_table, |
1141 | doc: /* Char table of script symbols. | |
1142 | It has one extra slot whose value is a list of script symbols. */); | |
1143 | ||
1144 | /* Intern this now in case it isn't already done. | |
1145 | Setting this variable twice is harmless. | |
1146 | But don't staticpro it here--that is done in alloc.c. */ | |
1147 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1148 | DEFSYM (Qchar_script_table, "char-script-table"); | |
1149 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
1150 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
c7e14352 KH |
1151 | |
1152 | DEFVAR_LISP ("script-representative-chars", &Vscript_representative_chars, | |
c3bb7671 | 1153 | doc: /* Alist of scripts vs the representative characters. |
f4427a54 | 1154 | Each element is a cons (SCRIPT . CHARS). |
4d8e170e | 1155 | SCRIPT is a symbol representing a script or a subgroup of a script. |
c3bb7671 | 1156 | CHARS is a list or a vector of characters. |
472a4dc9 | 1157 | If it is a list, all characters in the list are necessary for supporting SCRIPT. |
c3bb7671 KH |
1158 | If it is a vector, one of the characters in the vector is necessary. |
1159 | This variable is used to find a font for a specific script. */); | |
c7e14352 | 1160 | Vscript_representative_chars = Qnil; |
a3cbb631 KH |
1161 | |
1162 | DEFVAR_LISP ("unicode-category-table", &Vunicode_category_table, | |
1163 | doc: /* Char table of Unicode's "General Category". | |
472a4dc9 JB |
1164 | All Unicode characters have one of the following values (symbol): |
1165 | Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po, | |
a3cbb631 KH |
1166 | Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn |
1167 | See The Unicode Standard for the meaning of those values. */); | |
1168 | /* The correct char-table is setup in characters.el. */ | |
1169 | Vunicode_category_table = Qnil; | |
0168c3d8 KH |
1170 | } |
1171 | ||
1172 | #endif /* emacs */ | |
fbaf0946 MB |
1173 | |
1174 | /* arch-tag: b6665960-3c3d-4184-85cd-af4318197999 | |
1175 | (do not change this comment) */ |