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. |
0168c3d8 | 4 | Copyright (C) 2001 Free Software Foundation, Inc. |
8f924df7 | 5 | Copyright (C) 2003 |
0168c3d8 KH |
6 | National Institute of Advanced Industrial Science and Technology (AIST) |
7 | Registration Number H13PRO009 | |
8 | ||
9 | This file is part of GNU Emacs. | |
10 | ||
11 | GNU Emacs is free software; you can redistribute it and/or modify | |
12 | it under the terms of the GNU General Public License as published by | |
13 | the Free Software Foundation; either version 2, or (at your option) | |
14 | any later version. | |
15 | ||
16 | GNU Emacs is distributed in the hope that it will be useful, | |
17 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
18 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
19 | GNU General Public License for more details. | |
20 | ||
21 | You should have received a copy of the GNU General Public License | |
22 | along with GNU Emacs; see the file COPYING. If not, write to | |
23 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
24 | Boston, MA 02111-1307, USA. */ | |
25 | ||
26 | /* At first, see the document in `character.h' to understand the code | |
27 | in this file. */ | |
28 | ||
29 | #ifdef emacs | |
30 | #include <config.h> | |
31 | #endif | |
32 | ||
33 | #include <stdio.h> | |
34 | ||
35 | #ifdef emacs | |
36 | ||
37 | #include <sys/types.h> | |
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 KH |
62 | /* Char-table of information about which character to unify to which |
63 | Unicode character. */ | |
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 | ||
84 | static Lisp_Object Qchar_script_table; | |
85 | ||
b672c5ae KH |
86 | /* Mapping table from unibyte chars to multibyte chars. */ |
87 | int unibyte_to_multibyte_table[256]; | |
15843e6f | 88 | |
0168c3d8 KH |
89 | \f |
90 | ||
33f91981 KH |
91 | /* Store multibyte form of character C at P. If C has modifier bits, |
92 | handle them appropriately. */ | |
93 | ||
0168c3d8 | 94 | int |
e3d8eb8c | 95 | char_string (c, p) |
0168c3d8 | 96 | int c; |
1889b238 | 97 | unsigned char *p; |
0168c3d8 KH |
98 | { |
99 | int bytes; | |
100 | ||
e3d8eb8c KH |
101 | if (c & CHAR_MODIFIER_MASK) |
102 | { | |
cc6dfd2a KH |
103 | /* As an non-ASCII character can't have modifier bits, we just |
104 | ignore the bits. */ | |
105 | if (ASCII_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) | |
e3d8eb8c KH |
106 | { |
107 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
108 | if (c & CHAR_META) | |
109 | { | |
110 | /* Move the meta bit to the right place for a string. */ | |
111 | c = (c & ~CHAR_META) | 0x80; | |
112 | } | |
113 | if (c & CHAR_SHIFT) | |
114 | { | |
115 | /* Shift modifier is valid only with [A-Za-z]. */ | |
116 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
117 | c &= ~CHAR_SHIFT; | |
118 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
119 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
120 | } | |
121 | if (c & CHAR_CTL) | |
122 | { | |
123 | /* Simulate the code in lread.c. */ | |
124 | /* Allow `\C- ' and `\C-?'. */ | |
125 | if (c == (CHAR_CTL | ' ')) | |
126 | c = 0; | |
127 | else if (c == (CHAR_CTL | '?')) | |
128 | c = 127; | |
129 | /* ASCII control chars are made from letters (both cases), | |
130 | as well as the non-letters within 0100...0137. */ | |
131 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
132 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
133 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
134 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
135 | } | |
136 | } | |
137 | ||
138 | /* If C still has any modifier bits, just ignore it. */ | |
139 | c &= ~CHAR_MODIFIER_MASK; | |
140 | } | |
141 | ||
0168c3d8 KH |
142 | MAYBE_UNIFY_CHAR (c); |
143 | ||
e3d8eb8c | 144 | if (c <= MAX_3_BYTE_CHAR) |
0168c3d8 KH |
145 | { |
146 | bytes = CHAR_STRING (c, p); | |
147 | } | |
148 | else if (c <= MAX_4_BYTE_CHAR) | |
149 | { | |
150 | p[0] = (0xF0 | (c >> 18)); | |
151 | p[1] = (0x80 | ((c >> 12) & 0x3F)); | |
152 | p[2] = (0x80 | ((c >> 6) & 0x3F)); | |
153 | p[3] = (0x80 | (c & 0x3F)); | |
154 | bytes = 4; | |
155 | } | |
e3d8eb8c | 156 | else if (c <= MAX_5_BYTE_CHAR) |
0168c3d8 KH |
157 | { |
158 | p[0] = 0xF8; | |
159 | p[1] = (0x80 | ((c >> 18) & 0x0F)); | |
160 | p[2] = (0x80 | ((c >> 12) & 0x3F)); | |
161 | p[3] = (0x80 | ((c >> 6) & 0x3F)); | |
162 | p[4] = (0x80 | (c & 0x3F)); | |
163 | bytes = 5; | |
164 | } | |
e3d8eb8c KH |
165 | else |
166 | { | |
167 | c = CHAR_TO_BYTE8 (c); | |
168 | bytes = BYTE8_STRING (c, p); | |
169 | } | |
1889b238 | 170 | |
0168c3d8 KH |
171 | return bytes; |
172 | } | |
173 | ||
174 | ||
33f91981 KH |
175 | /* Return a character whose multibyte form is at P. Set LEN is not |
176 | NULL, it must be a pointer to integer. In that case, set *LEN to | |
177 | the byte length of the multibyte form. If ADVANCED is not NULL, is | |
178 | must be a pointer to unsigned char. In that case, set *ADVANCED to | |
179 | the ending address (i.e. the starting address of the next | |
180 | character) of the multibyte form. */ | |
181 | ||
0168c3d8 | 182 | int |
e3d8eb8c | 183 | string_char (p, advanced, len) |
15843e6f KH |
184 | const unsigned char *p; |
185 | const unsigned char **advanced; | |
0168c3d8 KH |
186 | int *len; |
187 | { | |
1889b238 | 188 | int c; |
15843e6f | 189 | const unsigned char *saved_p = p; |
0168c3d8 KH |
190 | |
191 | if (*p < 0x80 || ! (*p & 0x20) || ! (*p & 0x10)) | |
192 | { | |
193 | c = STRING_CHAR_ADVANCE (p); | |
194 | } | |
195 | else if (! (*p & 0x08)) | |
196 | { | |
197 | c = ((((p)[0] & 0xF) << 18) | |
198 | | (((p)[1] & 0x3F) << 12) | |
199 | | (((p)[2] & 0x3F) << 6) | |
200 | | ((p)[3] & 0x3F)); | |
201 | p += 4; | |
202 | } | |
203 | else | |
204 | { | |
205 | c = ((((p)[1] & 0x3F) << 18) | |
206 | | (((p)[2] & 0x3F) << 12) | |
207 | | (((p)[3] & 0x3F) << 6) | |
208 | | ((p)[4] & 0x3F)); | |
209 | p += 5; | |
210 | } | |
211 | ||
212 | MAYBE_UNIFY_CHAR (c); | |
213 | ||
214 | if (len) | |
215 | *len = p - saved_p; | |
216 | if (advanced) | |
217 | *advanced = p; | |
218 | return c; | |
219 | } | |
220 | ||
221 | ||
222 | /* Translate character C by translation table TABLE. If C is | |
223 | negative, translate a character specified by CHARSET and CODE. If | |
224 | no translation is found in TABLE, return the untranslated | |
10453be9 KH |
225 | character. If TABLE is a list, elements are char tables. In this |
226 | case, translace C by all tables. */ | |
0168c3d8 KH |
227 | |
228 | int | |
229 | translate_char (table, c) | |
230 | Lisp_Object table; | |
231 | int c; | |
232 | { | |
10453be9 KH |
233 | if (CHAR_TABLE_P (table)) |
234 | { | |
235 | Lisp_Object ch; | |
236 | ||
237 | ch = CHAR_TABLE_REF (table, c); | |
238 | if (CHARACTERP (ch)) | |
239 | c = XINT (ch); | |
240 | } | |
241 | else | |
242 | { | |
243 | for (; CONSP (table); table = XCDR (table)) | |
244 | c = translate_char (XCAR (table), c); | |
245 | } | |
246 | return c; | |
0168c3d8 KH |
247 | } |
248 | ||
0168c3d8 | 249 | /* Convert the multibyte character C to unibyte 8-bit character based |
ac86488b KH |
250 | on the current value of charset_unibyte. If dimension of |
251 | charset_unibyte is more than one, return (C & 0xFF). | |
0168c3d8 KH |
252 | |
253 | The argument REV_TBL is now ignored. It will be removed in the | |
254 | future. */ | |
255 | ||
256 | int | |
257 | multibyte_char_to_unibyte (c, rev_tbl) | |
258 | int c; | |
259 | Lisp_Object rev_tbl; | |
260 | { | |
b672c5ae KH |
261 | struct charset *charset; |
262 | unsigned c1; | |
0168c3d8 | 263 | |
b672c5ae KH |
264 | if (CHAR_BYTE8_P (c)) |
265 | return CHAR_TO_BYTE8 (c); | |
266 | charset = CHARSET_FROM_ID (charset_unibyte); | |
267 | c1 = ENCODE_CHAR (charset, c); | |
0168c3d8 KH |
268 | return ((c1 != CHARSET_INVALID_CODE (charset)) ? c1 : c & 0xFF); |
269 | } | |
270 | ||
271 | ||
272 | DEFUN ("characterp", Fcharacterp, Scharacterp, 1, 2, 0, | |
273 | doc: /* Return non-nil if OBJECT is a character. */) | |
274 | (object, ignore) | |
275 | Lisp_Object object, ignore; | |
276 | { | |
277 | return (CHARACTERP (object) ? Qt : Qnil); | |
278 | } | |
279 | ||
280 | DEFUN ("max-char", Fmax_char, Smax_char, 0, 0, 0, | |
281 | doc: /* Return the character of the maximum code. */) | |
282 | () | |
283 | { | |
284 | return make_number (MAX_CHAR); | |
285 | } | |
286 | ||
287 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, | |
288 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
289 | doc: /* Convert the unibyte character CH to multibyte character. | |
290 | The multibyte character is a result of decoding CH by | |
ed1d5bc0 | 291 | the current unibyte charset (see `unibyte-charset'). */) |
0168c3d8 KH |
292 | (ch) |
293 | Lisp_Object ch; | |
294 | { | |
295 | int c; | |
296 | struct charset *charset; | |
297 | ||
298 | CHECK_CHARACTER (ch); | |
299 | c = XFASTINT (ch); | |
300 | if (c >= 0400) | |
301 | error ("Invalid unibyte character: %d", c); | |
ac86488b | 302 | charset = CHARSET_FROM_ID (charset_unibyte); |
0168c3d8 KH |
303 | c = DECODE_CHAR (charset, c); |
304 | if (c < 0) | |
3c5a53bd | 305 | c = BYTE8_TO_CHAR (XFASTINT (ch)); |
0168c3d8 KH |
306 | return make_number (c); |
307 | } | |
308 | ||
309 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, | |
310 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
311 | doc: /* Convert the multibyte character CH to unibyte character.\n\ | |
312 | The unibyte character is a result of encoding CH by | |
313 | the current primary charset (value of `charset-primary'). */) | |
314 | (ch) | |
315 | Lisp_Object ch; | |
316 | { | |
317 | int c; | |
0168c3d8 KH |
318 | |
319 | CHECK_CHARACTER (ch); | |
320 | c = XFASTINT (ch); | |
3c5a53bd KH |
321 | c = CHAR_TO_BYTE8 (c); |
322 | return make_number (c); | |
0168c3d8 KH |
323 | } |
324 | ||
325 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
326 | doc: /* Return 1 regardless of the argument CHAR. | |
327 | This is now an obsolete function. We keep it just for backward compatibility. */) | |
328 | (ch) | |
329 | Lisp_Object ch; | |
330 | { | |
331 | CHECK_CHARACTER (ch); | |
332 | return make_number (1); | |
333 | } | |
334 | ||
335 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
336 | doc: /* Return width of CHAR when displayed in the current buffer. | |
337 | The width is measured by how many columns it occupies on the screen. | |
338 | Tab is taken to occupy `tab-width' columns. */) | |
339 | (ch) | |
340 | Lisp_Object ch; | |
341 | { | |
342 | Lisp_Object disp; | |
343 | int c, width; | |
344 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
345 | ||
346 | CHECK_CHARACTER (ch); | |
347 | c = XINT (ch); | |
348 | ||
349 | /* Get the way the display table would display it. */ | |
350 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; | |
351 | ||
352 | if (VECTORP (disp)) | |
353 | width = ASIZE (disp); | |
354 | else | |
355 | width = CHAR_WIDTH (c); | |
356 | ||
357 | return make_number (width); | |
358 | } | |
359 | ||
0168c3d8 KH |
360 | /* Return width of string STR of length LEN when displayed in the |
361 | current buffer. The width is measured by how many columns it | |
362 | occupies on the screen. If PRECISION > 0, return the width of | |
363 | longest substring that doesn't exceed PRECISION, and set number of | |
364 | characters and bytes of the substring in *NCHARS and *NBYTES | |
365 | respectively. */ | |
366 | ||
1889b238 | 367 | int |
0168c3d8 | 368 | c_string_width (str, len, precision, nchars, nbytes) |
8f924df7 | 369 | const unsigned char *str; |
0168c3d8 KH |
370 | int precision, *nchars, *nbytes; |
371 | { | |
372 | int i = 0, i_byte = 0; | |
373 | int width = 0; | |
374 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
375 | ||
376 | while (i_byte < len) | |
377 | { | |
378 | int bytes, thiswidth; | |
379 | Lisp_Object val; | |
380 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
381 | ||
382 | if (dp) | |
383 | { | |
384 | val = DISP_CHAR_VECTOR (dp, c); | |
385 | if (VECTORP (val)) | |
386 | thiswidth = XVECTOR (val)->size; | |
387 | else | |
388 | thiswidth = CHAR_WIDTH (c); | |
389 | } | |
390 | else | |
391 | { | |
392 | thiswidth = CHAR_WIDTH (c); | |
393 | } | |
394 | ||
395 | if (precision > 0 | |
396 | && (width + thiswidth > precision)) | |
397 | { | |
398 | *nchars = i; | |
399 | *nbytes = i_byte; | |
400 | return width; | |
401 | } | |
402 | i++; | |
403 | i_byte += bytes; | |
404 | width += thiswidth; | |
405 | } | |
406 | ||
407 | if (precision > 0) | |
408 | { | |
409 | *nchars = i; | |
410 | *nbytes = i_byte; | |
411 | } | |
412 | ||
413 | return width; | |
414 | } | |
415 | ||
1889b238 KH |
416 | /* Return width of string STR of length LEN when displayed in the |
417 | current buffer. The width is measured by how many columns it | |
418 | occupies on the screen. */ | |
419 | ||
420 | int | |
421 | strwidth (str, len) | |
422 | unsigned char *str; | |
423 | int len; | |
424 | { | |
425 | return c_string_width (str, len, -1, NULL, NULL); | |
426 | } | |
427 | ||
0168c3d8 KH |
428 | /* Return width of Lisp string STRING when displayed in the current |
429 | buffer. The width is measured by how many columns it occupies on | |
430 | the screen while paying attention to compositions. If PRECISION > | |
431 | 0, return the width of longest substring that doesn't exceed | |
432 | PRECISION, and set number of characters and bytes of the substring | |
433 | in *NCHARS and *NBYTES respectively. */ | |
434 | ||
435 | int | |
436 | lisp_string_width (string, precision, nchars, nbytes) | |
437 | Lisp_Object string; | |
438 | int precision, *nchars, *nbytes; | |
439 | { | |
8f924df7 KH |
440 | int len = SCHARS (string); |
441 | unsigned char *str = SDATA (string); | |
0168c3d8 KH |
442 | int i = 0, i_byte = 0; |
443 | int width = 0; | |
444 | struct Lisp_Char_Table *dp = buffer_display_table (); | |
445 | ||
446 | while (i < len) | |
447 | { | |
448 | int chars, bytes, thiswidth; | |
449 | Lisp_Object val; | |
450 | int cmp_id; | |
f4bc0685 | 451 | EMACS_INT ignore, end; |
0168c3d8 KH |
452 | |
453 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
454 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
455 | >= 0)) | |
456 | { | |
457 | thiswidth = composition_table[cmp_id]->width; | |
458 | chars = end - i; | |
459 | bytes = string_char_to_byte (string, end) - i_byte; | |
460 | } | |
461 | else if (dp) | |
462 | { | |
463 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
464 | ||
465 | chars = 1; | |
466 | val = DISP_CHAR_VECTOR (dp, c); | |
467 | if (VECTORP (val)) | |
468 | thiswidth = XVECTOR (val)->size; | |
469 | else | |
470 | thiswidth = CHAR_WIDTH (c); | |
471 | } | |
472 | else | |
473 | { | |
474 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
475 | ||
476 | chars = 1; | |
477 | thiswidth = CHAR_WIDTH (c); | |
478 | } | |
479 | ||
480 | if (precision > 0 | |
481 | && (width + thiswidth > precision)) | |
482 | { | |
483 | *nchars = i; | |
484 | *nbytes = i_byte; | |
485 | return width; | |
486 | } | |
487 | i += chars; | |
488 | i_byte += bytes; | |
489 | width += thiswidth; | |
490 | } | |
491 | ||
492 | if (precision > 0) | |
493 | { | |
494 | *nchars = i; | |
495 | *nbytes = i_byte; | |
496 | } | |
497 | ||
498 | return width; | |
499 | } | |
500 | ||
501 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
502 | doc: /* Return width of STRING when displayed in the current buffer. | |
503 | Width is measured by how many columns it occupies on the screen. | |
504 | When calculating width of a multibyte character in STRING, | |
505 | only the base leading-code is considered; the validity of | |
506 | the following bytes is not checked. Tabs in STRING are always | |
507 | taken to occupy `tab-width' columns. */) | |
508 | (str) | |
509 | Lisp_Object str; | |
510 | { | |
511 | Lisp_Object val; | |
512 | ||
513 | CHECK_STRING (str); | |
514 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); | |
515 | return val; | |
516 | } | |
517 | ||
518 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
519 | doc: /* Return the direction of CHAR. | |
520 | The returned value is 0 for left-to-right and 1 for right-to-left. */) | |
521 | (ch) | |
522 | Lisp_Object ch; | |
523 | { | |
524 | int c; | |
525 | ||
526 | CHECK_CHARACTER (ch); | |
527 | c = XINT (ch); | |
528 | return CHAR_TABLE_REF (Vchar_direction_table, c); | |
529 | } | |
530 | ||
531 | DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, | |
532 | doc: /* Return number of characters between BEG and END. | |
533 | This is now an obsolete function. We keep it just for backward compatibility. */) | |
534 | (beg, end) | |
535 | Lisp_Object beg, end; | |
536 | { | |
537 | int from, to; | |
538 | ||
539 | CHECK_NUMBER_COERCE_MARKER (beg); | |
540 | CHECK_NUMBER_COERCE_MARKER (end); | |
541 | ||
542 | from = min (XFASTINT (beg), XFASTINT (end)); | |
543 | to = max (XFASTINT (beg), XFASTINT (end)); | |
544 | ||
545 | return make_number (to - from); | |
546 | } | |
547 | ||
548 | /* Return the number of characters in the NBYTES bytes at PTR. | |
549 | This works by looking at the contents and checking for multibyte | |
550 | sequences while assuming that there's no invalid sequence. | |
551 | However, if the current buffer has enable-multibyte-characters = | |
552 | nil, we treat each byte as a character. */ | |
553 | ||
554 | int | |
555 | chars_in_text (ptr, nbytes) | |
8f924df7 | 556 | const unsigned char *ptr; |
0168c3d8 KH |
557 | int nbytes; |
558 | { | |
559 | /* current_buffer is null at early stages of Emacs initialization. */ | |
560 | if (current_buffer == 0 | |
561 | || NILP (current_buffer->enable_multibyte_characters)) | |
562 | return nbytes; | |
563 | ||
564 | return multibyte_chars_in_text (ptr, nbytes); | |
565 | } | |
566 | ||
567 | /* Return the number of characters in the NBYTES bytes at PTR. | |
568 | This works by looking at the contents and checking for multibyte | |
569 | sequences while assuming that there's no invalid sequence. It | |
570 | ignores enable-multibyte-characters. */ | |
571 | ||
572 | int | |
573 | multibyte_chars_in_text (ptr, nbytes) | |
8f924df7 | 574 | const unsigned char *ptr; |
0168c3d8 KH |
575 | int nbytes; |
576 | { | |
8f924df7 | 577 | const unsigned char *endp = ptr + nbytes; |
0168c3d8 KH |
578 | int chars = 0; |
579 | ||
580 | while (ptr < endp) | |
581 | { | |
582 | int len = MULTIBYTE_LENGTH (ptr, endp); | |
583 | ||
584 | if (len == 0) | |
585 | abort (); | |
586 | ptr += len; | |
587 | chars++; | |
588 | } | |
589 | ||
590 | return chars; | |
591 | } | |
592 | ||
593 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, count | |
594 | characters and bytes in it, and store them in *NCHARS and *NBYTES | |
595 | respectively. On counting bytes, pay attention to that 8-bit | |
596 | characters not constructing a valid multibyte sequence are | |
597 | represented by 2-byte in a multibyte text. */ | |
598 | ||
599 | void | |
600 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
8f924df7 | 601 | const unsigned char *str; |
0168c3d8 KH |
602 | int len, *nchars, *nbytes; |
603 | { | |
8f924df7 | 604 | const unsigned char *endp = str + len; |
0168c3d8 KH |
605 | int n, chars = 0, bytes = 0; |
606 | ||
607 | if (len >= MAX_MULTIBYTE_LENGTH) | |
608 | { | |
8f924df7 | 609 | const unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; |
0168c3d8 KH |
610 | while (str < adjusted_endp) |
611 | { | |
612 | if ((n = MULTIBYTE_LENGTH_NO_CHECK (str)) > 0) | |
613 | str += n, bytes += n; | |
614 | else | |
615 | str++, bytes += 2; | |
616 | chars++; | |
617 | } | |
618 | } | |
619 | while (str < endp) | |
620 | { | |
621 | if ((n = MULTIBYTE_LENGTH (str, endp)) > 0) | |
622 | str += n, bytes += n; | |
623 | else | |
624 | str++, bytes += 2; | |
625 | chars++; | |
626 | } | |
627 | ||
628 | *nchars = chars; | |
629 | *nbytes = bytes; | |
630 | return; | |
631 | } | |
632 | ||
633 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
634 | It actually converts only such 8-bit characters that don't contruct | |
635 | a multibyte sequence to multibyte forms of Latin-1 characters. If | |
636 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
637 | text. It is assured that we can use LEN bytes at STR as a work | |
638 | area and that is enough. Return the number of bytes of the | |
639 | resulting text. */ | |
640 | ||
641 | int | |
642 | str_as_multibyte (str, len, nbytes, nchars) | |
643 | unsigned char *str; | |
644 | int len, nbytes, *nchars; | |
645 | { | |
646 | unsigned char *p = str, *endp = str + nbytes; | |
647 | unsigned char *to; | |
648 | int chars = 0; | |
649 | int n; | |
650 | ||
651 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
652 | { | |
653 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
654 | while (p < adjusted_endp | |
655 | && (n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
656 | p += n, chars++; | |
657 | } | |
658 | while ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
659 | p += n, chars++; | |
660 | if (nchars) | |
661 | *nchars = chars; | |
662 | if (p == endp) | |
663 | return nbytes; | |
664 | ||
665 | to = p; | |
666 | nbytes = endp - p; | |
667 | endp = str + len; | |
668 | safe_bcopy ((char *) p, (char *) (endp - nbytes), nbytes); | |
669 | p = endp - nbytes; | |
670 | ||
671 | if (nbytes >= MAX_MULTIBYTE_LENGTH) | |
672 | { | |
673 | unsigned char *adjusted_endp = endp - MAX_MULTIBYTE_LENGTH; | |
674 | while (p < adjusted_endp) | |
675 | { | |
676 | if ((n = MULTIBYTE_LENGTH_NO_CHECK (p)) > 0) | |
677 | { | |
678 | while (n--) | |
679 | *to++ = *p++; | |
680 | } | |
681 | else | |
682 | { | |
683 | int c = *p++; | |
684 | c = BYTE8_TO_CHAR (c); | |
685 | to += CHAR_STRING (c, to); | |
686 | } | |
687 | } | |
688 | chars++; | |
689 | } | |
690 | while (p < endp) | |
691 | { | |
692 | if ((n = MULTIBYTE_LENGTH (p, endp)) > 0) | |
693 | { | |
694 | while (n--) | |
695 | *to++ = *p++; | |
8f924df7 | 696 | } |
0168c3d8 KH |
697 | else |
698 | { | |
699 | int c = *p++; | |
700 | c = BYTE8_TO_CHAR (c); | |
701 | to += CHAR_STRING (c, to); | |
702 | } | |
703 | chars++; | |
704 | } | |
705 | if (nchars) | |
706 | *nchars = chars; | |
707 | return (to - str); | |
708 | } | |
709 | ||
710 | /* Parse unibyte string at STR of LEN bytes, and return the number of | |
711 | bytes it may ocupy when converted to multibyte string by | |
712 | `str_to_multibyte'. */ | |
713 | ||
714 | int | |
715 | parse_str_to_multibyte (str, len) | |
716 | unsigned char *str; | |
717 | int len; | |
718 | { | |
719 | unsigned char *endp = str + len; | |
720 | int bytes; | |
721 | ||
722 | for (bytes = 0; str < endp; str++) | |
723 | bytes += (*str < 0x80) ? 1 : 2; | |
724 | return bytes; | |
725 | } | |
726 | ||
727 | ||
728 | /* Convert unibyte text at STR of NBYTES bytes to a multibyte text | |
729 | that contains the same single-byte characters. It actually | |
730 | converts all 8-bit characters to multibyte forms. It is assured | |
731 | that we can use LEN bytes at STR as a work area and that is | |
732 | enough. */ | |
733 | ||
734 | int | |
735 | str_to_multibyte (str, len, bytes) | |
736 | unsigned char *str; | |
737 | int len, bytes; | |
738 | { | |
739 | unsigned char *p = str, *endp = str + bytes; | |
740 | unsigned char *to; | |
741 | ||
742 | while (p < endp && *p < 0x80) p++; | |
743 | if (p == endp) | |
744 | return bytes; | |
745 | to = p; | |
746 | bytes = endp - p; | |
747 | endp = str + len; | |
748 | safe_bcopy ((char *) p, (char *) (endp - bytes), bytes); | |
749 | p = endp - bytes; | |
8f924df7 | 750 | while (p < endp) |
0168c3d8 KH |
751 | { |
752 | int c = *p++; | |
753 | ||
754 | if (c >= 0x80) | |
755 | c = BYTE8_TO_CHAR (c); | |
756 | to += CHAR_STRING (c, to); | |
757 | } | |
758 | return (to - str); | |
759 | } | |
760 | ||
761 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It | |
762 | actually converts characters in the range 0x80..0xFF to | |
763 | unibyte. */ | |
764 | ||
765 | int | |
766 | str_as_unibyte (str, bytes) | |
767 | unsigned char *str; | |
768 | int bytes; | |
769 | { | |
15843e6f KH |
770 | const unsigned char *p = str, *endp = str + bytes; |
771 | unsigned char *to; | |
0168c3d8 KH |
772 | int c, len; |
773 | ||
774 | while (p < endp) | |
775 | { | |
776 | c = *p; | |
777 | len = BYTES_BY_CHAR_HEAD (c); | |
778 | if (CHAR_BYTE8_HEAD_P (c)) | |
779 | break; | |
780 | p += len; | |
781 | } | |
15843e6f | 782 | to = str + (p - str); |
8f924df7 | 783 | while (p < endp) |
0168c3d8 KH |
784 | { |
785 | c = *p; | |
786 | len = BYTES_BY_CHAR_HEAD (c); | |
787 | if (CHAR_BYTE8_HEAD_P (c)) | |
788 | { | |
789 | c = STRING_CHAR_ADVANCE (p); | |
790 | *to++ = CHAR_TO_BYTE8 (c); | |
791 | } | |
792 | else | |
793 | { | |
794 | while (len--) *to++ = *p++; | |
795 | } | |
796 | } | |
797 | return (to - str); | |
798 | } | |
799 | ||
800 | int | |
801 | string_count_byte8 (string) | |
802 | Lisp_Object string; | |
803 | { | |
804 | int multibyte = STRING_MULTIBYTE (string); | |
8f924df7 KH |
805 | int nbytes = SBYTES (string); |
806 | unsigned char *p = SDATA (string); | |
0168c3d8 KH |
807 | unsigned char *pend = p + nbytes; |
808 | int count = 0; | |
809 | int c, len; | |
810 | ||
811 | if (multibyte) | |
812 | while (p < pend) | |
813 | { | |
814 | c = *p; | |
815 | len = BYTES_BY_CHAR_HEAD (c); | |
816 | ||
817 | if (CHAR_BYTE8_HEAD_P (c)) | |
818 | count++; | |
819 | p += len; | |
820 | } | |
821 | else | |
822 | while (p < pend) | |
823 | { | |
824 | if (*p++ >= 0x80) | |
825 | count++; | |
826 | } | |
827 | return count; | |
828 | } | |
829 | ||
830 | ||
831 | Lisp_Object | |
832 | string_escape_byte8 (string) | |
833 | Lisp_Object string; | |
834 | { | |
8f924df7 KH |
835 | int nchars = SCHARS (string); |
836 | int nbytes = SBYTES (string); | |
0168c3d8 KH |
837 | int multibyte = STRING_MULTIBYTE (string); |
838 | int byte8_count; | |
15843e6f KH |
839 | const unsigned char *src, *src_end; |
840 | unsigned char *dst; | |
0168c3d8 KH |
841 | Lisp_Object val; |
842 | int c, len; | |
843 | ||
844 | if (multibyte && nchars == nbytes) | |
845 | return string; | |
846 | ||
847 | byte8_count = string_count_byte8 (string); | |
848 | ||
849 | if (byte8_count == 0) | |
850 | return string; | |
851 | ||
852 | if (multibyte) | |
853 | /* Convert 2-byte sequence of byte8 chars to 4-byte octal. */ | |
7b40ebaf | 854 | val = make_uninit_multibyte_string (nchars + byte8_count * 3, |
0168c3d8 KH |
855 | nbytes + byte8_count * 2); |
856 | else | |
857 | /* Convert 1-byte sequence of byte8 chars to 4-byte octal. */ | |
858 | val = make_uninit_string (nbytes + byte8_count * 3); | |
859 | ||
8f924df7 | 860 | src = SDATA (string); |
0168c3d8 | 861 | src_end = src + nbytes; |
8f924df7 | 862 | dst = SDATA (val); |
0168c3d8 KH |
863 | if (multibyte) |
864 | while (src < src_end) | |
865 | { | |
866 | c = *src; | |
867 | len = BYTES_BY_CHAR_HEAD (c); | |
868 | ||
869 | if (CHAR_BYTE8_HEAD_P (c)) | |
870 | { | |
871 | c = STRING_CHAR_ADVANCE (src); | |
872 | c = CHAR_TO_BYTE8 (c); | |
1889b238 | 873 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
874 | dst += 4; |
875 | } | |
876 | else | |
877 | while (len--) *dst++ = *src++; | |
878 | } | |
879 | else | |
880 | while (src < src_end) | |
881 | { | |
882 | c = *src++; | |
883 | if (c >= 0x80) | |
884 | { | |
1889b238 | 885 | sprintf ((char *) dst, "\\%03o", c); |
0168c3d8 KH |
886 | dst += 4; |
887 | } | |
888 | else | |
889 | *dst++ = c; | |
890 | } | |
891 | return val; | |
892 | } | |
893 | ||
894 | \f | |
8f924df7 | 895 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
0168c3d8 | 896 | doc: /* |
d2e83296 DL |
897 | Concatenate all the argument characters and make the result a string. |
898 | usage: (string &rest CHARACTERS) */) | |
0168c3d8 KH |
899 | (n, args) |
900 | int n; | |
901 | Lisp_Object *args; | |
902 | { | |
903 | int i; | |
904 | unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); | |
905 | unsigned char *p = buf; | |
906 | int c; | |
907 | ||
908 | for (i = 0; i < n; i++) | |
909 | { | |
910 | CHECK_CHARACTER (args[i]); | |
911 | c = XINT (args[i]); | |
912 | p += CHAR_STRING (c, p); | |
913 | } | |
914 | ||
915 | return make_string_from_bytes ((char *) buf, n, p - buf); | |
916 | } | |
917 | ||
918 | void | |
919 | init_character_once () | |
920 | { | |
921 | } | |
922 | ||
923 | #ifdef emacs | |
924 | ||
925 | void | |
926 | syms_of_character () | |
927 | { | |
928 | DEFSYM (Qcharacterp, "characterp"); | |
929 | DEFSYM (Qauto_fill_chars, "auto-fill-chars"); | |
930 | ||
931 | staticpro (&Vchar_unify_table); | |
932 | Vchar_unify_table = Qnil; | |
933 | ||
934 | defsubr (&Smax_char); | |
935 | defsubr (&Scharacterp); | |
936 | defsubr (&Sunibyte_char_to_multibyte); | |
937 | defsubr (&Smultibyte_char_to_unibyte); | |
938 | defsubr (&Schar_bytes); | |
939 | defsubr (&Schar_width); | |
940 | defsubr (&Sstring_width); | |
941 | defsubr (&Schar_direction); | |
942 | defsubr (&Schars_in_region); | |
943 | defsubr (&Sstring); | |
944 | ||
945 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, | |
946 | doc: /* | |
68978cf0 DL |
947 | Vector recording all translation tables ever defined. |
948 | Each element is a pair (SYMBOL . TABLE) relating the table to the | |
949 | symbol naming it. The ID of a translation table is an index into this vector. */); | |
0168c3d8 KH |
950 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
951 | ||
952 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
953 | doc: /* | |
954 | A char-table for characters which invoke auto-filling. | |
955 | Such characters have value t in this table. */); | |
956 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
6cc0e1ca DL |
957 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); |
958 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
0168c3d8 KH |
959 | |
960 | DEFVAR_LISP ("char-width-table", &Vchar_width_table, | |
961 | doc: /* | |
962 | A char-table for width (columns) of each character. */); | |
963 | Vchar_width_table = Fmake_char_table (Qnil, make_number (1)); | |
be8b50bc KH |
964 | char_table_set_range (Vchar_width_table, 0x80, 0x9F, make_number (4)); |
965 | char_table_set_range (Vchar_width_table, MAX_5_BYTE_CHAR + 1, MAX_CHAR, | |
966 | make_number (4)); | |
0168c3d8 KH |
967 | |
968 | DEFVAR_LISP ("char-direction-table", &Vchar_direction_table, | |
969 | doc: /* A char-table for direction of each character. */); | |
970 | Vchar_direction_table = Fmake_char_table (Qnil, make_number (1)); | |
971 | ||
972 | DEFVAR_LISP ("printable-chars", &Vprintable_chars, | |
973 | doc: /* A char-table for each printable character. */); | |
db6d4189 | 974 | Vprintable_chars = Fmake_char_table (Qnil, Qnil); |
67dde660 KH |
975 | Fset_char_table_range (Vprintable_chars, |
976 | Fcons (make_number (32), make_number (126)), Qt); | |
977 | Fset_char_table_range (Vprintable_chars, | |
978 | Fcons (make_number (160), | |
979 | make_number (MAX_5_BYTE_CHAR)), Qt); | |
15843e6f | 980 | |
c57f3328 KH |
981 | DEFVAR_LISP ("char-script-table", &Vchar_script_table, |
982 | doc: /* Char table of script symbols. | |
983 | It has one extra slot whose value is a list of script symbols. */); | |
984 | ||
985 | /* Intern this now in case it isn't already done. | |
986 | Setting this variable twice is harmless. | |
987 | But don't staticpro it here--that is done in alloc.c. */ | |
988 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
989 | DEFSYM (Qchar_script_table, "char-script-table"); | |
990 | Fput (Qchar_script_table, Qchar_table_extra_slots, make_number (1)); | |
991 | Vchar_script_table = Fmake_char_table (Qchar_script_table, Qnil); | |
0168c3d8 KH |
992 | } |
993 | ||
994 | #endif /* emacs */ |