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. |
114f9c96 | 4 | Copyright (C) 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 |
ec62e0ac | 5 | Free Software Foundation, Inc. |
114f9c96 | 6 | Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 |
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> | |
d7306fe6 | 37 | #include <setjmp.h> |
0168c3d8 KH |
38 | #include "lisp.h" |
39 | #include "character.h" | |
40 | #include "buffer.h" | |
41 | #include "charset.h" | |
42 | #include "composite.h" | |
43 | #include "disptab.h" | |
44 | ||
45 | #else /* not emacs */ | |
46 | ||
47 | #include "mulelib.h" | |
48 | ||
49 | #endif /* emacs */ | |
50 | ||
51 | Lisp_Object Qcharacterp; | |
52 | ||
53 | /* Vector of translation table ever defined. | |
54 | ID of a translation table is used to index this vector. */ | |
55 | Lisp_Object Vtranslation_table_vector; | |
56 | ||
57 | /* A char-table for characters which may invoke auto-filling. */ | |
58 | Lisp_Object Vauto_fill_chars; | |
59 | ||
60 | Lisp_Object Qauto_fill_chars; | |
61 | ||
33f91981 | 62 | /* Char-table of information about which character to unify to which |
6f1b43a0 | 63 | Unicode character. Mainly used by the macro MAYBE_UNIFY_CHAR. */ |
0168c3d8 KH |
64 | Lisp_Object Vchar_unify_table; |
65 | ||
66 | /* A char-table. An element is non-nil iff the corresponding | |
67 | character has a printable glyph. */ | |
68 | Lisp_Object Vprintable_chars; | |
69 | ||
70 | /* A char-table. An elemnent is a column-width of the corresponding | |
71 | character. */ | |
72 | Lisp_Object Vchar_width_table; | |
73 | ||
74 | /* A char-table. An element is a symbol indicating the direction | |
75 | property of corresponding character. */ | |
76 | Lisp_Object Vchar_direction_table; | |
77 | ||
8973478b | 78 | /* Variable used locally in the macro FETCH_MULTIBYTE_CHAR. */ |
0168c3d8 | 79 | unsigned char *_fetch_multibyte_char_p; |
0168c3d8 | 80 | |
c57f3328 KH |
81 | /* Char table of scripts. */ |
82 | Lisp_Object Vchar_script_table; | |
83 | ||
c7e14352 KH |
84 | /* Alist of scripts vs representative characters. */ |
85 | Lisp_Object Vscript_representative_chars; | |
86 | ||
c57f3328 KH |
87 | static Lisp_Object Qchar_script_table; |
88 | ||
a3cbb631 | 89 | Lisp_Object Vunicode_category_table; |
0168c3d8 KH |
90 | \f |
91 | ||
2bde7652 KH |
92 | /* If character code C has modifier masks, reflect them to the |
93 | character code if possible. Return the resulting code. */ | |
94 | ||
95 | int | |
971de7fb | 96 | char_resolve_modifier_mask (int c) |
2bde7652 | 97 | { |
d0363d44 | 98 | /* A non-ASCII character can't reflect modifier bits to the code. */ |
2bde7652 KH |
99 | if (! ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) |
100 | return c; | |
101 | ||
102 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
2bde7652 KH |
103 | if (c & CHAR_SHIFT) |
104 | { | |
105 | /* Shift modifier is valid only with [A-Za-z]. */ | |
106 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
107 | c &= ~CHAR_SHIFT; | |
108 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
109 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
03365d0e KH |
110 | /* Shift modifier for control characters and SPC is ignored. */ |
111 | else if ((c & ~CHAR_MODIFIER_MASK) <= 0x20) | |
d0363d44 KH |
112 | c &= ~CHAR_SHIFT; |
113 | } | |
2bde7652 KH |
114 | if (c & CHAR_CTL) |
115 | { | |
116 | /* Simulate the code in lread.c. */ | |
117 | /* Allow `\C- ' and `\C-?'. */ | |
03365d0e KH |
118 | if ((c & 0377) == ' ') |
119 | c &= ~0177 & ~ CHAR_CTL; | |
120 | else if ((c & 0377) == '?') | |
121 | c = 0177 | (c & ~0177 & ~CHAR_CTL); | |
2bde7652 KH |
122 | /* ASCII control chars are made from letters (both cases), |
123 | as well as the non-letters within 0100...0137. */ | |
124 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
125 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
126 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
127 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
128 | } | |
e9c1637d | 129 | #if 0 /* This is outside the scope of this function. (bug#4751) */ |
03365d0e KH |
130 | if (c & CHAR_META) |
131 | { | |
132 | /* Move the meta bit to the right place for a string. */ | |
133 | c = (c & ~CHAR_META) | 0x80; | |
134 | } | |
e9c1637d | 135 | #endif |
2bde7652 KH |
136 | |
137 | return c; | |
138 | } | |
139 | ||
140 | ||
33f91981 KH |
141 | /* Store multibyte form of character C at P. If C has modifier bits, |
142 | handle them appropriately. */ | |
143 | ||
0168c3d8 | 144 | int |
971de7fb | 145 | char_string (unsigned int c, unsigned char *p) |
0168c3d8 KH |
146 | { |
147 | int bytes; | |
148 | ||
e3d8eb8c KH |
149 | if (c & CHAR_MODIFIER_MASK) |
150 | { | |
2bde7652 | 151 | c = (unsigned) char_resolve_modifier_mask ((int) c); |
e3d8eb8c KH |
152 | /* If C still has any modifier bits, just ignore it. */ |
153 | c &= ~CHAR_MODIFIER_MASK; | |
154 | } | |
155 | ||
0168c3d8 KH |
156 | MAYBE_UNIFY_CHAR (c); |
157 | ||
e3d8eb8c | 158 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
159 | { |
160 | bytes = CHAR_STRING (c, p); | |
161 | } | |
162 | else if (c <= MAX_4_BYTE_CHAR) | |
163 | { | |
164 | p[0] = (0xF0 | (c >> 18)); | |
165 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
166 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
167 | p[3] = (0x80 | (c & 0x3F)); | |
168 | bytes = 4; | |
169 | } | |
e3d8eb8c | 170 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
171 | { |
172 | p[0] = 0xF8; | |
173 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
174 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
175 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
176 | p[4] = (0x80 | (c & 0x3F)); | |
177 | bytes = 5; | |
178 | } | |
5aa91c9b | 179 | else if (c <= MAX_CHAR) |
e3d8eb8c KH |
180 | { |
181 | c = CHAR_TO_BYTE8 (c); | |
182 | bytes = BYTE8_STRING (c, p); | |
183 | } | |
5aa91c9b KH |
184 | else |
185 | error ("Invalid character: %d", c); | |
1889b238 | 186 | |
0168c3d8 KH |
187 | return bytes; |
188 | } | |
189 | ||
190 | ||
33f91981 KH |
191 | /* Return a character whose multibyte form is at P. Set LEN is not |
192 | NULL, it must be a pointer to integer. In that case, set *LEN to | |
193 | the byte length of the multibyte form. If ADVANCED is not NULL, is | |
194 | must be a pointer to unsigned char. In that case, set *ADVANCED to | |
195 | the ending address (i.e. the starting address of the next | |
196 | character) of the multibyte form. */ | |
197 | ||
0168c3d8 | 198 | int |
971de7fb | 199 | string_char (const unsigned char *p, const unsigned char **advanced, int *len) |
0168c3d8 | 200 | { |
1889b238 | 201 | int c; |
15843e6f | 202 | const unsigned char *saved_p = p; |
0168c3d8 KH |
203 | |
204 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
205 | { | |
206 | c = STRING_CHAR_ADVANCE (p); | |
207 | } | |
208 | else if (! (*p & 0x08)) | |
209 | { | |
210 | c = ((((p)[0] & 0xF) << 18) | |
211 | | (((p)[1] & 0x3F) << 12) | |
212 | | (((p)[2] & 0x3F) << 6) | |
213 | | ((p)[3] & 0x3F)); | |
214 | p += 4; | |
215 | } | |
216 | else | |
217 | { | |
218 | c = ((((p)[1] & 0x3F) << 18) | |
219 | | (((p)[2] & 0x3F) << 12) | |
220 | | (((p)[3] & 0x3F) << 6) | |
221 | | ((p)[4] & 0x3F)); | |
222 | p += 5; | |
223 | } | |
224 | ||
225 | MAYBE_UNIFY_CHAR (c); | |
226 | ||
227 | if (len) | |
228 | *len = p - saved_p; | |
229 | if (advanced) | |
230 | *advanced = p; | |
231 | return c; | |
232 | } | |
233 | ||
234 | ||
235 | /* Translate character C by translation table TABLE. If C is | |
236 | negative, translate a character specified by CHARSET and CODE. If | |
237 | no translation is found in TABLE, return the untranslated | |
10453be9 KH |
238 | character. If TABLE is a list, elements are char tables. In this |
239 | case, translace C by all tables. */ | |
0168c3d8 KH |
240 | |
241 | int | |
971de7fb | 242 | translate_char (Lisp_Object table, int c) |
0168c3d8 | 243 | { |
10453be9 KH |
244 | if (CHAR_TABLE_P (table)) |
245 | { | |
246 | Lisp_Object ch; | |
247 | ||
248 | ch = CHAR_TABLE_REF (table, c); | |
249 | if (CHARACTERP (ch)) | |
250 | c = XINT (ch); | |
251 | } | |
252 | else | |
253 | { | |
254 | for (; CONSP (table); table = XCDR (table)) | |
255 | c = translate_char (XCAR (table), c); | |
256 | } | |
257 | return c; | |
0168c3d8 KH |
258 | } |
259 | ||
2e5db15c KH |
260 | /* Convert ASCII or 8-bit character C to unibyte. If C is none of |
261 | them, return (C & 0xFF). | |
0168c3d8 KH |
262 | |
263 | The argument REV_TBL is now ignored. It will be removed in the | |
264 | future. */ | |
265 | ||
266 | int | |
971de7fb | 267 | multibyte_char_to_unibyte (int c, Lisp_Object rev_tbl) |
0168c3d8 | 268 | { |
2e5db15c KH |
269 | if (c < 0x80) |
270 | return c; | |
b672c5ae KH |
271 | if (CHAR_BYTE8_P (c)) |
272 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 273 | return (c & 0xFF); |
0168c3d8 KH |
274 | } |
275 | ||
935d5b02 KH |
276 | /* Like multibyte_char_to_unibyte, but return -1 if C is not supported |
277 | by charset_unibyte. */ | |
278 | ||
279 | int | |
971de7fb | 280 | multibyte_char_to_unibyte_safe (int c) |
935d5b02 | 281 | { |
2e5db15c KH |
282 | if (c < 0x80) |
283 | return c; | |
935d5b02 KH |
284 | if (CHAR_BYTE8_P (c)) |
285 | return CHAR_TO_BYTE8 (c); | |
2e5db15c | 286 | return -1; |
935d5b02 | 287 | } |
0168c3d8 KH |
288 | |
289 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
290 | doc: /* Return non-nil if OBJECT is a character. */) | |
5842a27b | 291 | (Lisp_Object object, Lisp_Object ignore) |
0168c3d8 KH |
292 | { |
293 | return (CHARACTERP (object) ? Qt : Qnil); | |
294 | } | |
295 | ||
296 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
297 | doc: /* Return the character of the maximum code. */) | |
5842a27b | 298 | (void) |
0168c3d8 KH |
299 | { |
300 | return make_number (MAX_CHAR); | |
301 | } | |
302 | ||
303 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
304 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
5556875b | 305 | doc: /* Convert the byte CH to multibyte character. */) |
5842a27b | 306 | (Lisp_Object ch) |
0168c3d8 KH |
307 | { |
308 | int c; | |
0168c3d8 KH |
309 | |
310 | CHECK_CHARACTER (ch); | |
311 | c = XFASTINT (ch); | |
2e5db15c KH |
312 | if (c >= 0x100) |
313 | error ("Not a unibyte character: %d", c); | |
4c0354d7 | 314 | MAKE_CHAR_MULTIBYTE (c); |
0168c3d8 KH |
315 | return make_number (c); |
316 | } | |
317 | ||
318 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
319 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
5556875b SM |
320 | doc: /* Convert the multibyte character CH to a byte. |
321 | If the multibyte character does not represent a byte, return -1. */) | |
5842a27b | 322 | (Lisp_Object ch) |
0168c3d8 | 323 | { |
5556875b | 324 | int cm; |
0168c3d8 KH |
325 | |
326 | CHECK_CHARACTER (ch); | |
5556875b SM |
327 | cm = XFASTINT (ch); |
328 | if (cm < 256) | |
329 | /* Can't distinguish a byte read from a unibyte buffer from | |
330 | a latin1 char, so let's let it slide. */ | |
331 | return ch; | |
332 | else | |
333 | { | |
2afc21f5 | 334 | int cu = CHAR_TO_BYTE_SAFE (cm); |
5556875b SM |
335 | return make_number (cu); |
336 | } | |
0168c3d8 KH |
337 | } |
338 | ||
339 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
340 | doc: /* Return 1 regardless of the argument CHAR. | |
add553ac JB |
341 | This is now an obsolete function. We keep it just for backward compatibility. |
342 | usage: (char-bytes CHAR) */) | |
5842a27b | 343 | (Lisp_Object ch) |
0168c3d8 KH |
344 | { |
345 | CHECK_CHARACTER (ch); | |
346 | return make_number (1); | |
347 | } | |
348 | ||
349 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
350 | doc: /* Return width of CHAR when displayed in the current buffer. | |
351 | The width is measured by how many columns it occupies on the screen. | |
add553ac JB |
352 | Tab is taken to occupy `tab-width' columns. |
353 | usage: (char-width CHAR) */) | |
5842a27b | 354 | (Lisp_Object ch) |
0168c3d8 KH |
355 | { |
356 | Lisp_Object disp; | |
357 | int c, width; | |
358 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
359 | ||
360 | CHECK_CHARACTER (ch); | |
361 | c = XINT (ch); | |
362 | ||
363 | /* Get the way the display table would display it. */ | |
364 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
365 | ||
366 | if (VECTORP (disp)) | |
367 | width = ASIZE (disp); | |
368 | else | |
369 | width = CHAR_WIDTH (c); | |
370 | ||
371 | return make_number (width); | |
372 | } | |
373 | ||
0168c3d8 KH |
374 | /* Return width of string STR of length LEN when displayed in the |
375 | current buffer. The width is measured by how many columns it | |
376 | occupies on the screen. If PRECISION > 0, return the width of | |
377 | longest substring that doesn't exceed PRECISION, and set number of | |
378 | characters and bytes of the substring in *NCHARS and *NBYTES | |
379 | respectively. */ | |
380 | ||
1889b238 | 381 | int |
63c125ab | 382 | c_string_width (const unsigned char *str, int len, int precision, int *nchars, int *nbytes) |
0168c3d8 KH |
383 | { |
384 | int i = 0, i_byte = 0; | |
385 | int width = 0; | |
386 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
387 | ||
388 | while (i_byte < len) | |
389 | { | |
390 | int bytes, thiswidth; | |
391 | Lisp_Object val; | |
62a6e103 | 392 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes); |
0168c3d8 KH |
393 | |
394 | if (dp) | |
395 | { | |
396 | val = DISP_CHAR_VECTOR (dp, c); | |
397 | if (VECTORP (val)) | |
398 | thiswidth = XVECTOR (val)->size; | |
399 | else | |
400 | thiswidth = CHAR_WIDTH (c); | |
401 | } | |
402 | else | |
403 | { | |
404 | thiswidth = CHAR_WIDTH (c); | |
405 | } | |
406 | ||
407 | if (precision > 0 | |
408 | && (width + thiswidth > precision)) | |
409 | { | |
410 | *nchars = i; | |
411 | *nbytes = i_byte; | |
412 | return width; | |
413 | } | |
414 | i++; | |
415 | i_byte += bytes; | |
416 | width += thiswidth; | |
417 | } | |
418 | ||
419 | if (precision > 0) | |
420 | { | |
421 | *nchars = i; | |
422 | *nbytes = i_byte; | |
423 | } | |
424 | ||
425 | return width; | |
426 | } | |
427 | ||
1889b238 KH |
428 | /* Return width of string STR of length LEN when displayed in the |
429 | current buffer. The width is measured by how many columns it | |
430 | occupies on the screen. */ | |
431 | ||
432 | int | |
eec47d6b | 433 | strwidth (const unsigned char *str, int len) |
1889b238 KH |
434 | { |
435 | return c_string_width (str, len, -1, NULL, NULL); | |
436 | } | |
437 | ||
0168c3d8 KH |
438 | /* Return width of Lisp string STRING when displayed in the current |
439 | buffer. The width is measured by how many columns it occupies on | |
440 | the screen while paying attention to compositions. If PRECISION > | |
441 | 0, return the width of longest substring that doesn't exceed | |
442 | PRECISION, and set number of characters and bytes of the substring | |
443 | in *NCHARS and *NBYTES respectively. */ | |
444 | ||
445 | int | |
971de7fb | 446 | lisp_string_width (Lisp_Object string, int precision, int *nchars, int *nbytes) |
0168c3d8 | 447 | { |
8f924df7 | 448 | int len = SCHARS (string); |
0aee65b9 KH |
449 | /* This set multibyte to 0 even if STRING is multibyte when it |
450 | contains only ascii and eight-bit-graphic, but that's | |
451 | intentional. */ | |
452 | int multibyte = len < SBYTES (string); | |
8f924df7 | 453 | unsigned char *str = SDATA (string); |
0168c3d8 KH |
454 | int i = 0, i_byte = 0; |
455 | int width = 0; | |
456 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
457 | ||
458 | while (i < len) | |
459 | { | |
460 | int chars, bytes, thiswidth; | |
461 | Lisp_Object val; | |
462 | int cmp_id; | |
f4bc0685 | 463 | EMACS_INT ignore, end; |
0168c3d8 KH |
464 | |
465 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
466 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
467 | >= 0)) | |
468 | { | |
469 | thiswidth = composition_table[cmp_id]->width; | |
470 | chars = end - i; | |
471 | bytes = string_char_to_byte (string, end) - i_byte; | |
472 | } | |
0168c3d8 KH |
473 | else |
474 | { | |
0aee65b9 | 475 | int c; |
0168c3d8 | 476 | |
0aee65b9 | 477 | if (multibyte) |
62a6e103 | 478 | c = STRING_CHAR_AND_LENGTH (str + i_byte, bytes); |
0aee65b9 KH |
479 | else |
480 | c = str[i_byte], bytes = 1; | |
0168c3d8 | 481 | chars = 1; |
0aee65b9 KH |
482 | if (dp) |
483 | { | |
484 | val = DISP_CHAR_VECTOR (dp, c); | |
485 | if (VECTORP (val)) | |
486 | thiswidth = XVECTOR (val)->size; | |
487 | else | |
488 | thiswidth = CHAR_WIDTH (c); | |
489 | } | |
490 | else | |
491 | { | |
492 | thiswidth = CHAR_WIDTH (c); | |
493 | } | |
0168c3d8 KH |
494 | } |
495 | ||
496 | if (precision > 0 | |
497 | && (width + thiswidth > precision)) | |
498 | { | |
499 | *nchars = i; | |
500 | *nbytes = i_byte; | |
501 | return width; | |
502 | } | |
503 | i += chars; | |
504 | i_byte += bytes; | |
505 | width += thiswidth; | |
506 | } | |
507 | ||
508 | if (precision > 0) | |
509 | { | |
510 | *nchars = i; | |
511 | *nbytes = i_byte; | |
512 | } | |
513 | ||
514 | return width; | |
515 | } | |
516 | ||
517 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
518 | doc: /* Return width of STRING when displayed in the current buffer. | |
519 | Width is measured by how many columns it occupies on the screen. | |
520 | When calculating width of a multibyte character in STRING, | |
521 | only the base leading-code is considered; the validity of | |
522 | the following bytes is not checked. Tabs in STRING are always | |
add553ac JB |
523 | taken to occupy `tab-width' columns. |
524 | usage: (string-width STRING) */) | |
5842a27b | 525 | (Lisp_Object str) |
0168c3d8 KH |
526 | { |
527 | Lisp_Object val; | |
528 | ||
529 | CHECK_STRING (str); | |
530 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
531 | return val; | |
532 | } | |
533 | ||
534 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
535 | doc: /* Return the direction of CHAR. | |
add553ac JB |
536 | The returned value is 0 for left-to-right and 1 for right-to-left. |
537 | usage: (char-direction CHAR) */) | |
5842a27b | 538 | (Lisp_Object ch) |
0168c3d8 KH |
539 | { |
540 | int c; | |
541 | ||
542 | CHECK_CHARACTER (ch); | |
543 | c = XINT (ch); | |
544 | return CHAR_TABLE_REF (Vchar_direction_table, c); | |
545 | } | |
546 | ||
0168c3d8 KH |
547 | /* Return the number of characters in the NBYTES bytes at PTR. |
548 | This works by looking at the contents and checking for multibyte | |
549 | sequences while assuming that there's no invalid sequence. | |
550 | However, if the current buffer has enable-multibyte-characters = | |
551 | nil, we treat each byte as a character. */ | |
552 | ||
13818c30 | 553 | EMACS_INT |
971de7fb | 554 | chars_in_text (const unsigned char *ptr, EMACS_INT nbytes) |
0168c3d8 KH |
555 | { |
556 | /* current_buffer is null at early stages of Emacs initialization. */ | |
557 | if (current_buffer == 0 | |
558 | || NILP (current_buffer->enable_multibyte_characters)) | |
559 | return nbytes; | |
560 | ||
561 | return multibyte_chars_in_text (ptr, nbytes); | |
562 | } | |
563 | ||
564 | /* Return the number of characters in the NBYTES bytes at PTR. | |
565 | This works by looking at the contents and checking for multibyte | |
566 | sequences while assuming that there's no invalid sequence. It | |
567 | ignores enable-multibyte-characters. */ | |
568 | ||
13818c30 | 569 | EMACS_INT |
971de7fb | 570 | multibyte_chars_in_text (const unsigned char *ptr, EMACS_INT nbytes) |
0168c3d8 | 571 | { |
8f924df7 | 572 | const unsigned char *endp = ptr + nbytes; |
0168c3d8 KH |
573 | int chars = 0; |
574 | ||
575 | while (ptr < endp) | |
576 | { | |
577 | int len = MULTIBYTE_LENGTH (ptr, endp); | |
578 | ||
579 | if (len == 0) | |
580 | abort (); | |
581 | ptr += len; | |
582 | chars++; | |
583 | } | |
584 | ||
585 | return chars; | |
586 | } | |
587 | ||
588 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
589 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
590 | respectively. On counting bytes, pay attention to that 8-bit | |
591 | characters not constructing a valid multibyte sequence are | |
592 | represented by 2-byte in a multibyte text. */ | |
593 | ||
594 | void | |
971de7fb | 595 | parse_str_as_multibyte (const unsigned char *str, int len, int *nchars, int *nbytes) |
0168c3d8 | 596 | { |
8f924df7 | 597 | const unsigned char *endp = str + len; |
0168c3d8 KH |
598 | int n, chars = 0, bytes = 0; |
599 | ||
600 | if (len >= MAX_MULTIBYTE_LENGTH) | |
601 | { | |
8f924df7 | 602 | const unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; |
0168c3d8 KH |
603 | while (str < adjusted_endp) |
604 | { | |
fc9a17bc KH |
605 | if (! CHAR_BYTE8_HEAD_P (*str) |
606 | && (n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
0168c3d8 KH |
607 | str += n, bytes += n; |
608 | else | |
609 | str++, bytes += 2; | |
610 | chars++; | |
611 | } | |
612 | } | |
613 | while (str < endp) | |
614 | { | |
fc9a17bc KH |
615 | if (! CHAR_BYTE8_HEAD_P (*str) |
616 | && (n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
0168c3d8 KH |
617 | str += n, bytes += n; |
618 | else | |
619 | str++, bytes += 2; | |
620 | chars++; | |
621 | } | |
622 | ||
623 | *nchars = chars; | |
624 | *nbytes = bytes; | |
625 | return; | |
626 | } | |
627 | ||
628 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
629 | It actually converts only such 8-bit characters that don't contruct | |
630 | a multibyte sequence to multibyte forms of Latin-1 characters. If | |
631 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
632 | text. It is assured that we can use LEN bytes at STR as a work | |
633 | area and that is enough. Return the number of bytes of the | |
634 | resulting text. */ | |
635 | ||
636 | int | |
971de7fb | 637 | str_as_multibyte (unsigned char *str, int len, int nbytes, int *nchars) |
0168c3d8 KH |
638 | { |
639 | unsigned char *p = str, *endp = str + nbytes; | |
640 | unsigned char *to; | |
641 | int chars = 0; | |
642 | int n; | |
643 | ||
644 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
645 | { | |
646 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
647 | while (p < adjusted_endp | |
fc9a17bc | 648 | && ! CHAR_BYTE8_HEAD_P (*p) |
0168c3d8 KH |
649 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) |
650 | p += n, chars++; | |
651 | } | |
fc9a17bc KH |
652 | while (p < endp |
653 | && ! CHAR_BYTE8_HEAD_P (*p) | |
654 | && (n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
0168c3d8 KH |
655 | p += n, chars++; |
656 | if (nchars) | |
657 | *nchars = chars; | |
658 | if (p == endp) | |
659 | return nbytes; | |
660 | ||
661 | to = p; | |
662 | nbytes = endp - p; | |
663 | endp = str + len; | |
72af86bd | 664 | memmove (endp - nbytes, p, nbytes); |
0168c3d8 KH |
665 | p = endp - nbytes; |
666 | ||
667 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
668 | { | |
669 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
670 | while (p < adjusted_endp) | |
671 | { | |
fc9a17bc KH |
672 | if (! CHAR_BYTE8_HEAD_P (*p) |
673 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
0168c3d8 KH |
674 | { |
675 | while (n--) | |
676 | *to++ = *p++; | |
677 | } | |
678 | else | |
679 | { | |
680 | int c = *p++; | |
681 | c = BYTE8_TO_CHAR (c); | |
682 | to += CHAR_STRING (c, to); | |
683 | } | |
684 | } | |
685 | chars++; | |
686 | } | |
687 | while (p < endp) | |
688 | { | |
fc9a17bc KH |
689 | if (! CHAR_BYTE8_HEAD_P (*p) |
690 | && (n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
0168c3d8 KH |
691 | { |
692 | while (n--) | |
693 | *to++ = *p++; | |
8f924df7 | 694 | } |
0168c3d8 KH |
695 | else |
696 | { | |
697 | int c = *p++; | |
698 | c = BYTE8_TO_CHAR (c); | |
699 | to += CHAR_STRING (c, to); | |
700 | } | |
701 | chars++; | |
702 | } | |
703 | if (nchars) | |
704 | *nchars = chars; | |
705 | return (to - str); | |
706 | } | |
707 | ||
708 | /* Parse unibyte string at STR of LEN bytes, and return the number of | |
709 | bytes it may ocupy when converted to multibyte string by | |
710 | `str_to_multibyte'. */ | |
711 | ||
712 | int | |
eec47d6b | 713 | parse_str_to_multibyte (const unsigned char *str, int len) |
0168c3d8 | 714 | { |
eec47d6b | 715 | const unsigned char *endp = str + len; |
0168c3d8 KH |
716 | int bytes; |
717 | ||
718 | for (bytes = 0; str < endp; str++) | |
719 | bytes += (*str < 0x80) ? 1 : 2; | |
720 | return bytes; | |
721 | } | |
722 | ||
723 | ||
724 | /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
725 | that contains the same single-byte characters. It actually | |
726 | converts all 8-bit characters to multibyte forms. It is assured | |
727 | that we can use LEN bytes at STR as a work area and that is | |
728 | enough. */ | |
729 | ||
730 | int | |
971de7fb | 731 | str_to_multibyte (unsigned char *str, int len, int bytes) |
0168c3d8 KH |
732 | { |
733 | unsigned char *p = str, *endp = str + bytes; | |
734 | unsigned char *to; | |
735 | ||
736 | while (p < endp && *p < 0x80) p++; | |
737 | if (p == endp) | |
738 | return bytes; | |
739 | to = p; | |
740 | bytes = endp - p; | |
741 | endp = str + len; | |
72af86bd | 742 | memmove (endp - bytes, p, bytes); |
0168c3d8 | 743 | p = endp - bytes; |
8f924df7 | 744 | while (p < endp) |
0168c3d8 KH |
745 | { |
746 | int c = *p++; | |
747 | ||
748 | if (c >= 0x80) | |
749 | c = BYTE8_TO_CHAR (c); | |
750 | to += CHAR_STRING (c, to); | |
751 | } | |
752 | return (to - str); | |
753 | } | |
754 | ||
755 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
756 | actually converts characters in the range 0x80..0xFF to | |
757 | unibyte. */ | |
758 | ||
759 | int | |
971de7fb | 760 | str_as_unibyte (unsigned char *str, int bytes) |
0168c3d8 | 761 | { |
15843e6f KH |
762 | const unsigned char *p = str, *endp = str + bytes; |
763 | unsigned char *to; | |
0168c3d8 KH |
764 | int c, len; |
765 | ||
766 | while (p < endp) | |
767 | { | |
768 | c = *p; | |
769 | len = BYTES_BY_CHAR_HEAD (c); | |
770 | if (CHAR_BYTE8_HEAD_P (c)) | |
771 | break; | |
772 | p += len; | |
773 | } | |
15843e6f | 774 | to = str + (p - str); |
8f924df7 | 775 | while (p < endp) |
0168c3d8 KH |
776 | { |
777 | c = *p; | |
778 | len = BYTES_BY_CHAR_HEAD (c); | |
779 | if (CHAR_BYTE8_HEAD_P (c)) | |
780 | { | |
781 | c = STRING_CHAR_ADVANCE (p); | |
782 | *to++ = CHAR_TO_BYTE8 (c); | |
783 | } | |
784 | else | |
785 | { | |
786 | while (len--) *to++ = *p++; | |
787 | } | |
788 | } | |
789 | return (to - str); | |
790 | } | |
791 | ||
4aa40bb8 KH |
792 | /* Convert eight-bit chars in SRC (in multibyte form) to the |
793 | corresponding byte and store in DST. CHARS is the number of | |
794 | characters in SRC. The value is the number of bytes stored in DST. | |
795 | Usually, the value is the same as CHARS, but is less than it if SRC | |
796 | contains a non-ASCII, non-eight-bit characater. If ACCEPT_LATIN_1 | |
797 | is nonzero, a Latin-1 character is accepted and converted to a byte | |
f27f70ec KH |
798 | of that character code. |
799 | Note: Currently the arg ACCEPT_LATIN_1 is not used. */ | |
4aa40bb8 KH |
800 | |
801 | EMACS_INT | |
971de7fb | 802 | str_to_unibyte (const unsigned char *src, unsigned char *dst, EMACS_INT chars, int accept_latin_1) |
4aa40bb8 KH |
803 | { |
804 | EMACS_INT i; | |
805 | ||
806 | for (i = 0; i < chars; i++) | |
807 | { | |
808 | int c = STRING_CHAR_ADVANCE (src); | |
809 | ||
810 | if (CHAR_BYTE8_P (c)) | |
811 | c = CHAR_TO_BYTE8 (c); | |
812 | else if (! ASCII_CHAR_P (c) | |
813 | && (! accept_latin_1 || c >= 0x100)) | |
814 | return i; | |
815 | *dst++ = c; | |
816 | } | |
817 | return i; | |
818 | } | |
819 | ||
820 | ||
0168c3d8 | 821 | int |
971de7fb | 822 | string_count_byte8 (Lisp_Object string) |
0168c3d8 KH |
823 | { |
824 | int multibyte = STRING_MULTIBYTE (string); | |
8f924df7 KH |
825 | int nbytes = SBYTES (string); |
826 | unsigned char *p = SDATA (string); | |
0168c3d8 KH |
827 | unsigned char *pend = p + nbytes; |
828 | int count = 0; | |
829 | int c, len; | |
830 | ||
831 | if (multibyte) | |
832 | while (p < pend) | |
833 | { | |
834 | c = *p; | |
835 | len = BYTES_BY_CHAR_HEAD (c); | |
836 | ||
837 | if (CHAR_BYTE8_HEAD_P (c)) | |
838 | count++; | |
839 | p += len; | |
840 | } | |
841 | else | |
842 | while (p < pend) | |
843 | { | |
844 | if (*p++ >= 0x80) | |
845 | count++; | |
846 | } | |
847 | return count; | |
848 | } | |
849 | ||
850 | ||
851 | Lisp_Object | |
971de7fb | 852 | string_escape_byte8 (Lisp_Object string) |
0168c3d8 | 853 | { |
8f924df7 KH |
854 | int nchars = SCHARS (string); |
855 | int nbytes = SBYTES (string); | |
0168c3d8 KH |
856 | int multibyte = STRING_MULTIBYTE (string); |
857 | int byte8_count; | |
15843e6f KH |
858 | const unsigned char *src, *src_end; |
859 | unsigned char *dst; | |
0168c3d8 KH |
860 | Lisp_Object val; |
861 | int c, len; | |
862 | ||
863 | if (multibyte && nchars == nbytes) | |
864 | return string; | |
865 | ||
866 | byte8_count = string_count_byte8 (string); | |
867 | ||
868 | if (byte8_count == 0) | |
869 | return string; | |
870 | ||
871 | if (multibyte) | |
872 | /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
7b40ebaf | 873 | val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
0168c3d8 KH |
874 | nbytes + byte8_count * 2); |
875 | else | |
876 | /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
877 | val = make_uninit_string (nbytes + byte8_count * 3); | |
878 | ||
8f924df7 | 879 | src = SDATA (string); |
0168c3d8 | 880 | src_end = src + nbytes; |
8f924df7 | 881 | dst = SDATA (val); |
0168c3d8 KH |
882 | if (multibyte) |
883 | while (src < src_end) | |
884 | { | |
885 | c = *src; | |
886 | len = BYTES_BY_CHAR_HEAD (c); | |
887 | ||
888 | if (CHAR_BYTE8_HEAD_P (c)) | |
889 | { | |
890 | c = STRING_CHAR_ADVANCE (src); | |
891 | c = CHAR_TO_BYTE8 (c); | |
1889b238 | 892 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
893 | dst += 4; |
894 | } | |
895 | else | |
896 | while (len--) *dst++ = *src++; | |
897 | } | |
898 | else | |
899 | while (src < src_end) | |
900 | { | |
901 | c = *src++; | |
902 | if (c >= 0x80) | |
903 | { | |
1889b238 | 904 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
905 | dst += 4; |
906 | } | |
907 | else | |
908 | *dst++ = c; | |
909 | } | |
910 | return val; | |
911 | } | |
912 | ||
913 | \f | |
8f924df7 | 914 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 915 | doc: /* |
d2e83296 DL |
916 | Concatenate all the argument characters and make the result a string. |
917 | usage: (string &rest CHARACTERS) */) | |
5842a27b | 918 | (int n, Lisp_Object *args) |
0168c3d8 | 919 | { |
754790b6 CY |
920 | int i, c; |
921 | unsigned char *buf, *p; | |
922 | Lisp_Object str; | |
923 | USE_SAFE_ALLOCA; | |
924 | ||
925 | SAFE_ALLOCA (buf, unsigned char *, MAX_MULTIBYTE_LENGTH * n); | |
926 | p = buf; | |
0168c3d8 KH |
927 | |
928 | for (i = 0; i < n; i++) | |
929 | { | |
930 | CHECK_CHARACTER (args[i]); | |
931 | c = XINT (args[i]); | |
932 | p += CHAR_STRING (c, p); | |
933 | } | |
934 | ||
754790b6 CY |
935 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
936 | SAFE_FREE (); | |
937 | return str; | |
0168c3d8 KH |
938 | } |
939 | ||
70b4969d | 940 | DEFUN ("unibyte-string", Funibyte_string, Sunibyte_string, 0, MANY, 0, |
87d6f965 KH |
941 | doc: /* Concatenate all the argument bytes and make the result a unibyte string. |
942 | usage: (unibyte-string &rest BYTES) */) | |
5842a27b | 943 | (int n, Lisp_Object *args) |
70b4969d | 944 | { |
754790b6 CY |
945 | int i, c; |
946 | unsigned char *buf, *p; | |
947 | Lisp_Object str; | |
948 | USE_SAFE_ALLOCA; | |
949 | ||
950 | SAFE_ALLOCA (buf, unsigned char *, n); | |
951 | p = buf; | |
70b4969d KH |
952 | |
953 | for (i = 0; i < n; i++) | |
954 | { | |
955 | CHECK_NATNUM (args[i]); | |
956 | c = XFASTINT (args[i]); | |
957 | if (c >= 256) | |
958 | args_out_of_range_3 (args[i], make_number (0), make_number (255)); | |
959 | *p++ = c; | |
960 | } | |
961 | ||
754790b6 CY |
962 | str = make_string_from_bytes ((char *) buf, n, p - buf); |
963 | SAFE_FREE (); | |
964 | return str; | |
70b4969d KH |
965 | } |
966 | ||
c73ae4ae | 967 | DEFUN ("char-resolve-modifiers", Fchar_resolve_modifiers, |
d0363d44 KH |
968 | Schar_resolve_modifiers, 1, 1, 0, |
969 | doc: /* Resolve modifiers in the character CHAR. | |
970 | The value is a character with modifiers resolved into the character | |
971 | code. Unresolved modifiers are kept in the value. | |
c73ae4ae | 972 | usage: (char-resolve-modifiers CHAR) */) |
5842a27b | 973 | (Lisp_Object character) |
d0363d44 KH |
974 | { |
975 | int c; | |
976 | ||
977 | CHECK_NUMBER (character); | |
978 | c = XINT (character); | |
979 | return make_number (char_resolve_modifier_mask (c)); | |
980 | } | |
981 | ||
ee107a89 KH |
982 | DEFUN ("get-byte", Fget_byte, Sget_byte, 0, 2, 0, |
983 | doc: /* Return a byte value of a character at point. | |
984 | Optional 1st arg POSITION, if non-nil, is a position of a character to get | |
985 | a byte value. | |
986 | Optional 2nd arg STRING, if non-nil, is a string of which first | |
987 | character is a target to get a byte value. In this case, POSITION, if | |
988 | non-nil, is an index of a target character in the string. | |
989 | ||
990 | If the current buffer (or STRING) is multibyte, and the target | |
991 | character is not ASCII nor 8-bit character, an error is signalled. */) | |
5842a27b | 992 | (Lisp_Object position, Lisp_Object string) |
ee107a89 KH |
993 | { |
994 | int c; | |
995 | EMACS_INT pos; | |
996 | unsigned char *p; | |
997 | ||
998 | if (NILP (string)) | |
999 | { | |
1000 | if (NILP (position)) | |
1001 | { | |
1002 | p = PT_ADDR; | |
4d8e170e | 1003 | } |
ee107a89 KH |
1004 | else |
1005 | { | |
1006 | CHECK_NUMBER_COERCE_MARKER (position); | |
1007 | if (XINT (position) < BEGV || XINT (position) >= ZV) | |
1008 | args_out_of_range_3 (position, make_number (BEGV), make_number (ZV)); | |
1009 | pos = XFASTINT (position); | |
1010 | p = CHAR_POS_ADDR (pos); | |
1011 | } | |
d5998e03 KH |
1012 | if (NILP (current_buffer->enable_multibyte_characters)) |
1013 | return make_number (*p); | |
ee107a89 KH |
1014 | } |
1015 | else | |
1016 | { | |
1017 | CHECK_STRING (string); | |
1018 | if (NILP (position)) | |
1019 | { | |
1020 | p = SDATA (string); | |
1021 | } | |
1022 | else | |
1023 | { | |
1024 | CHECK_NATNUM (position); | |
1025 | if (XINT (position) >= SCHARS (string)) | |
1026 | args_out_of_range (string, position); | |
1027 | pos = XFASTINT (position); | |
1028 | p = SDATA (string) + string_char_to_byte (string, pos); | |
1029 | } | |
d5998e03 KH |
1030 | if (! STRING_MULTIBYTE (string)) |
1031 | return make_number (*p); | |
ee107a89 | 1032 | } |
62a6e103 | 1033 | c = STRING_CHAR (p); |
ee107a89 KH |
1034 | if (CHAR_BYTE8_P (c)) |
1035 | c = CHAR_TO_BYTE8 (c); | |
1036 | else if (! ASCII_CHAR_P (c)) | |
1037 | error ("Not an ASCII nor an 8-bit character: %d", c); | |
1038 | return make_number (c); | |
1039 | } | |
1040 | ||
1041 | ||
0168c3d8 | 1042 | void |
971de7fb | 1043 | init_character_once (void) |
0168c3d8 KH |
1044 | { |
1045 | } | |
1046 | ||
1047 | #ifdef emacs | |
1048 | ||
1049 | void | |
971de7fb | 1050 | syms_of_character (void) |
0168c3d8 KH |
1051 | { |
1052 | DEFSYM (Qcharacterp, "characterp"); | |
1053 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
1054 | ||
1055 | staticpro (&Vchar_unify_table); | |
1056 | Vchar_unify_table = Qnil; | |
1057 | ||
1058 | defsubr (&Smax_char); | |
1059 | defsubr (&Scharacterp); | |
1060 | defsubr (&Sunibyte_char_to_multibyte); | |
1061 | defsubr (&Smultibyte_char_to_unibyte); | |
1062 | defsubr (&Schar_bytes); | |
1063 | defsubr (&Schar_width); | |
1064 | defsubr (&Sstring_width); | |
1065 | defsubr (&Schar_direction); | |
0168c3d8 | 1066 | defsubr (&Sstring); |
70b4969d | 1067 | defsubr (&Sunibyte_string); |
d0363d44 | 1068 | defsubr (&Schar_resolve_modifiers); |
ee107a89 | 1069 | defsubr (&Sget_byte); |
0168c3d8 KH |
1070 | |
1071 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
1072 | doc: /* | |
68978cf0 DL |
1073 | Vector recording all translation tables ever defined. |
1074 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
1075 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
1076 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
1077 | ||
1078 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
1079 | doc: /* | |
1080 | A char-table for characters which invoke auto-filling. | |
1081 | Such characters have value t in this table. */); | |
1082 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
1083 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
1084 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 KH |
1085 | |
1086 | DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
1087 | doc: /* | |
1088 | A char-table for width (columns) of each character. */); | |
1089 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
1090 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
1091 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
1092 | make_number (4)); | |
0168c3d8 KH |
1093 | |
1094 | DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
1095 | doc: /* A char-table for direction of each character. */); | |
1096 | Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
1097 | ||
1098 | DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
1099 | doc: /* A char-table for each printable character. */); | |
db6d4189 | 1100 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
1101 | Fset_char_table_range (Vprintable_chars, |
1102 | Fcons (make_number (32), make_number (126)), Qt); | |
1103 | Fset_char_table_range (Vprintable_chars, | |
1104 | Fcons (make_number (160), | |
1105 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 1106 | |
c57f3328 KH |
1107 | DEFVAR_LISP ("char-script-table", &Vchar_script_table, |
1108 | doc: /* Char table of script symbols. | |
1109 | It has one extra slot whose value is a list of script symbols. */); | |
1110 | ||
1111 | /* Intern this now in case it isn't already done. | |
1112 | Setting this variable twice is harmless. | |
1113 | But don't staticpro it here--that is done in alloc.c. */ | |
d67b4f80 | 1114 | Qchar_table_extra_slots = intern_c_string ("char-table-extra-slots"); |
c57f3328 KH |
1115 | DEFSYM (Qchar_script_table, "char-script-table"); |
1116 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
1117 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
c7e14352 KH |
1118 | |
1119 | DEFVAR_LISP ("script-representative-chars", &Vscript_representative_chars, | |
c3bb7671 | 1120 | doc: /* Alist of scripts vs the representative characters. |
f4427a54 | 1121 | Each element is a cons (SCRIPT . CHARS). |
4d8e170e | 1122 | SCRIPT is a symbol representing a script or a subgroup of a script. |
c3bb7671 | 1123 | CHARS is a list or a vector of characters. |
472a4dc9 | 1124 | If it is a list, all characters in the list are necessary for supporting SCRIPT. |
c3bb7671 KH |
1125 | If it is a vector, one of the characters in the vector is necessary. |
1126 | This variable is used to find a font for a specific script. */); | |
c7e14352 | 1127 | Vscript_representative_chars = Qnil; |
a3cbb631 KH |
1128 | |
1129 | DEFVAR_LISP ("unicode-category-table", &Vunicode_category_table, | |
1130 | doc: /* Char table of Unicode's "General Category". | |
472a4dc9 JB |
1131 | All Unicode characters have one of the following values (symbol): |
1132 | Lu, Ll, Lt, Lm, Lo, Mn, Mc, Me, Nd, Nl, No, Pc, Pd, Ps, Pe, Pi, Pf, Po, | |
a3cbb631 KH |
1133 | Sm, Sc, Sk, So, Zs, Zl, Zp, Cc, Cf, Cs, Co, Cn |
1134 | See The Unicode Standard for the meaning of those values. */); | |
1135 | /* The correct char-table is setup in characters.el. */ | |
1136 | Vunicode_category_table = Qnil; | |
0168c3d8 KH |
1137 | } |
1138 | ||
1139 | #endif /* emacs */ | |
fbaf0946 MB |
1140 | |
1141 | /* arch-tag: b6665960-3c3d-4184-85cd-af4318197999 | |
1142 | (do not change this comment) */ |