Commit | Line | Data |
---|---|---|
75c8c592 | 1 | /* Basic multilingual character support. |
35e623fb | 2 | Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN. |
75c8c592 | 3 | Licensed to the Free Software Foundation. |
4ed46869 | 4 | |
369314dc KH |
5 | This file is part of GNU Emacs. |
6 | ||
7 | GNU Emacs is free software; you can redistribute it and/or modify | |
8 | it under the terms of the GNU General Public License as published by | |
9 | the Free Software Foundation; either version 2, or (at your option) | |
10 | any later version. | |
4ed46869 | 11 | |
369314dc KH |
12 | GNU Emacs is distributed in the hope that it will be useful, |
13 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
14 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
15 | GNU General Public License for more details. | |
4ed46869 | 16 | |
369314dc KH |
17 | You should have received a copy of the GNU General Public License |
18 | along with GNU Emacs; see the file COPYING. If not, write to | |
19 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
20 | Boston, MA 02111-1307, USA. */ | |
4ed46869 KH |
21 | |
22 | /* At first, see the document in `charset.h' to understand the code in | |
23 | this file. */ | |
24 | ||
25 | #include <stdio.h> | |
26 | ||
27 | #ifdef emacs | |
28 | ||
29 | #include <sys/types.h> | |
30 | #include <config.h> | |
31 | #include "lisp.h" | |
32 | #include "buffer.h" | |
33 | #include "charset.h" | |
34 | #include "coding.h" | |
fc6b09bf | 35 | #include "disptab.h" |
4ed46869 KH |
36 | |
37 | #else /* not emacs */ | |
38 | ||
39 | #include "mulelib.h" | |
40 | ||
41 | #endif /* emacs */ | |
42 | ||
43 | Lisp_Object Qcharset, Qascii, Qcomposition; | |
44 | ||
45 | /* Declaration of special leading-codes. */ | |
46 | int leading_code_composition; /* for composite characters */ | |
47 | int leading_code_private_11; /* for private DIMENSION1 of 1-column */ | |
48 | int leading_code_private_12; /* for private DIMENSION1 of 2-column */ | |
49 | int leading_code_private_21; /* for private DIMENSION2 of 1-column */ | |
50 | int leading_code_private_22; /* for private DIMENSION2 of 2-column */ | |
51 | ||
52 | /* Declaration of special charsets. */ | |
53 | int charset_ascii; /* ASCII */ | |
54 | int charset_composition; /* for a composite character */ | |
55 | int charset_latin_iso8859_1; /* ISO8859-1 (Latin-1) */ | |
56 | int charset_jisx0208_1978; /* JISX0208.1978 (Japanese Kanji old set) */ | |
57 | int charset_jisx0208; /* JISX0208.1983 (Japanese Kanji) */ | |
58 | int charset_katakana_jisx0201; /* JISX0201.Kana (Japanese Katakana) */ | |
59 | int charset_latin_jisx0201; /* JISX0201.Roman (Japanese Roman) */ | |
60 | int charset_big5_1; /* Big5 Level 1 (Chinese Traditional) */ | |
61 | int charset_big5_2; /* Big5 Level 2 (Chinese Traditional) */ | |
62 | ||
b0e3cf2b KH |
63 | int min_composite_char; |
64 | ||
4ed46869 KH |
65 | Lisp_Object Qcharset_table; |
66 | ||
67 | /* A char-table containing information of each character set. */ | |
68 | Lisp_Object Vcharset_table; | |
69 | ||
70 | /* A vector of charset symbol indexed by charset-id. This is used | |
71 | only for returning charset symbol from C functions. */ | |
72 | Lisp_Object Vcharset_symbol_table; | |
73 | ||
74 | /* A list of charset symbols ever defined. */ | |
75 | Lisp_Object Vcharset_list; | |
76 | ||
537efd8d KH |
77 | /* Vector of translation table ever defined. |
78 | ID of a translation table is used to index this vector. */ | |
79 | Lisp_Object Vtranslation_table_vector; | |
b0e3cf2b | 80 | |
c1a08b4c KH |
81 | /* A char-table for characters which may invoke auto-filling. */ |
82 | Lisp_Object Vauto_fill_chars; | |
83 | ||
84 | Lisp_Object Qauto_fill_chars; | |
85 | ||
4ed46869 KH |
86 | /* Tables used by macros BYTES_BY_CHAR_HEAD and WIDTH_BY_CHAR_HEAD. */ |
87 | int bytes_by_char_head[256]; | |
88 | int width_by_char_head[256]; | |
89 | ||
90 | /* Mapping table from ISO2022's charset (specified by DIMENSION, | |
91 | CHARS, and FINAL-CHAR) to Emacs' charset. */ | |
92 | int iso_charset_table[2][2][128]; | |
93 | ||
513ee442 KH |
94 | /* Table of pointers to the structure `cmpchar_info' indexed by |
95 | CMPCHAR-ID. */ | |
96 | struct cmpchar_info **cmpchar_table; | |
97 | /* The current size of `cmpchar_table'. */ | |
98 | static int cmpchar_table_size; | |
99 | /* Number of the current composite characters. */ | |
100 | int n_cmpchars; | |
101 | ||
4ed46869 KH |
102 | /* Variables used locally in the macro FETCH_MULTIBYTE_CHAR. */ |
103 | unsigned char *_fetch_multibyte_char_p; | |
104 | int _fetch_multibyte_char_len; | |
105 | ||
35e623fb RS |
106 | /* Offset to add to a non-ASCII value when inserting it. */ |
107 | int nonascii_insert_offset; | |
108 | ||
4cf9710d RS |
109 | /* Translation table for converting non-ASCII unibyte characters |
110 | to multibyte codes, or nil. */ | |
b4e9dd77 | 111 | Lisp_Object Vnonascii_translation_table; |
4cf9710d | 112 | |
8a73a704 KH |
113 | /* List of all possible generic characters. */ |
114 | Lisp_Object Vgeneric_character_list; | |
115 | ||
046b1f03 RS |
116 | #define min(X, Y) ((X) < (Y) ? (X) : (Y)) |
117 | #define max(X, Y) ((X) > (Y) ? (X) : (Y)) | |
118 | \f | |
93bcb785 KH |
119 | void |
120 | invalid_character (c) | |
121 | int c; | |
122 | { | |
ba7434e5 | 123 | error ("Invalid character: 0%o, %d, 0x%x", c, c, c); |
93bcb785 KH |
124 | } |
125 | ||
126 | ||
4ed46869 KH |
127 | /* Set STR a pointer to the multi-byte form of the character C. If C |
128 | is not a composite character, the multi-byte form is set in WORKBUF | |
129 | and STR points WORKBUF. The caller should allocate at least 4-byte | |
130 | area at WORKBUF in advance. Returns the length of the multi-byte | |
bd4c6dd0 KH |
131 | form. If C is an invalid character to have a multi-byte form, |
132 | signal an error. | |
4ed46869 KH |
133 | |
134 | Use macro `CHAR_STRING (C, WORKBUF, STR)' instead of calling this | |
135 | function directly if C can be an ASCII character. */ | |
136 | ||
137 | int | |
138 | non_ascii_char_to_string (c, workbuf, str) | |
139 | int c; | |
140 | unsigned char *workbuf, **str; | |
141 | { | |
6dc0722d | 142 | int charset, c1, c2; |
4ed46869 KH |
143 | |
144 | if (COMPOSITE_CHAR_P (c)) | |
145 | { | |
146 | int cmpchar_id = COMPOSITE_CHAR_ID (c); | |
147 | ||
148 | if (cmpchar_id < n_cmpchars) | |
149 | { | |
150 | *str = cmpchar_table[cmpchar_id]->data; | |
151 | return cmpchar_table[cmpchar_id]->len; | |
152 | } | |
153 | else | |
154 | { | |
93bcb785 | 155 | invalid_character (c); |
4ed46869 KH |
156 | } |
157 | } | |
158 | ||
159 | SPLIT_NON_ASCII_CHAR (c, charset, c1, c2); | |
bd4c6dd0 KH |
160 | if (!charset |
161 | || ! CHARSET_DEFINED_P (charset) | |
162 | || c1 >= 0 && c1 < 32 | |
163 | || c2 >= 0 && c2 < 32) | |
93bcb785 | 164 | invalid_character (c); |
4ed46869 KH |
165 | |
166 | *str = workbuf; | |
167 | *workbuf++ = CHARSET_LEADING_CODE_BASE (charset); | |
168 | if (*workbuf = CHARSET_LEADING_CODE_EXT (charset)) | |
169 | workbuf++; | |
170 | *workbuf++ = c1 | 0x80; | |
6dc0722d | 171 | if (c2 >= 0) |
4ed46869 KH |
172 | *workbuf++ = c2 | 0x80; |
173 | ||
174 | return (workbuf - *str); | |
175 | } | |
176 | ||
177 | /* Return a non-ASCII character of which multi-byte form is at STR of | |
178 | length LEN. If ACTUAL_LEN is not NULL, the actual length of the | |
537efd8d KH |
179 | multibyte form is set to the address ACTUAL_LEN. |
180 | ||
181 | If exclude_tail_garbage is nonzero, ACTUAL_LEN excludes gabage | |
182 | bytes following the non-ASCII character. | |
4ed46869 KH |
183 | |
184 | Use macro `STRING_CHAR (STR, LEN)' instead of calling this function | |
185 | directly if STR can hold an ASCII character. */ | |
186 | ||
dfcf069d | 187 | int |
537efd8d | 188 | string_to_non_ascii_char (str, len, actual_len, exclude_tail_garbage) |
8867de67 | 189 | const unsigned char *str; |
537efd8d | 190 | int len, *actual_len, exclude_tail_garbage; |
4ed46869 KH |
191 | { |
192 | int charset; | |
193 | unsigned char c1, c2; | |
90d7b74e | 194 | register int c, bytes; |
4ed46869 | 195 | |
90d7b74e KH |
196 | c = *str; |
197 | bytes = 1; | |
198 | ||
199 | if (BASE_LEADING_CODE_P (c)) | |
4ed46869 | 200 | { |
90d7b74e KH |
201 | while (bytes < len && ! CHAR_HEAD_P (str[bytes])) bytes++; |
202 | ||
203 | if (c == LEADING_CODE_COMPOSITION) | |
204 | { | |
205 | int cmpchar_id = str_cmpchar_id (str, bytes); | |
206 | ||
207 | if (cmpchar_id >= 0) | |
5d76bc89 KH |
208 | { |
209 | c = MAKE_COMPOSITE_CHAR (cmpchar_id); | |
210 | if (exclude_tail_garbage) | |
211 | bytes = cmpchar_table[cmpchar_id]->len; | |
212 | } | |
90d7b74e KH |
213 | } |
214 | else | |
215 | { | |
216 | int charset = c, c1, c2 = 0; | |
537efd8d | 217 | int char_bytes = BYTES_BY_CHAR_HEAD (c); |
4ed46869 | 218 | |
90d7b74e KH |
219 | str++; |
220 | if (c >= LEADING_CODE_PRIVATE_11) | |
221 | charset = *str++; | |
537efd8d | 222 | if (char_bytes <= bytes && CHARSET_DEFINED_P (charset)) |
90d7b74e KH |
223 | { |
224 | c1 = *str++ & 0x7f; | |
225 | if (CHARSET_DIMENSION (charset) == 2) | |
226 | c2 = *str & 0x7F; | |
227 | c = MAKE_NON_ASCII_CHAR (charset, c1, c2); | |
537efd8d KH |
228 | if (exclude_tail_garbage) |
229 | bytes = char_bytes; | |
90d7b74e KH |
230 | } |
231 | } | |
232 | } | |
4ed46869 KH |
233 | |
234 | if (actual_len) | |
90d7b74e | 235 | *actual_len = bytes; |
4ed46869 KH |
236 | return c; |
237 | } | |
238 | ||
239 | /* Return the length of the multi-byte form at string STR of length LEN. */ | |
240 | int | |
241 | multibyte_form_length (str, len) | |
8867de67 | 242 | const unsigned char *str; |
4ed46869 KH |
243 | int len; |
244 | { | |
90d7b74e | 245 | int bytes = 1; |
4ed46869 | 246 | |
90d7b74e KH |
247 | if (BASE_LEADING_CODE_P (*str)) |
248 | while (bytes < len && ! CHAR_HEAD_P (str[bytes])) bytes++; | |
4ed46869 | 249 | |
90d7b74e | 250 | return bytes; |
4ed46869 KH |
251 | } |
252 | ||
253 | /* Check if string STR of length LEN contains valid multi-byte form of | |
254 | a character. If valid, charset and position codes of the character | |
255 | is set at *CHARSET, *C1, and *C2, and return 0. If not valid, | |
256 | return -1. This should be used only in the macro SPLIT_STRING | |
257 | which checks range of STR in advance. */ | |
258 | ||
dfcf069d | 259 | int |
4ed46869 | 260 | split_non_ascii_string (str, len, charset, c1, c2) |
8867de67 KH |
261 | register const unsigned char *str; |
262 | register unsigned char *c1, *c2; | |
4ed46869 KH |
263 | register int len, *charset; |
264 | { | |
265 | register unsigned int cs = *str++; | |
266 | ||
267 | if (cs == LEADING_CODE_COMPOSITION) | |
268 | { | |
269 | int cmpchar_id = str_cmpchar_id (str - 1, len); | |
270 | ||
271 | if (cmpchar_id < 0) | |
272 | return -1; | |
273 | *charset = cs, *c1 = cmpchar_id >> 7, *c2 = cmpchar_id & 0x7F; | |
274 | } | |
275 | else if ((cs < LEADING_CODE_PRIVATE_11 || (cs = *str++) >= 0xA0) | |
276 | && CHARSET_DEFINED_P (cs)) | |
277 | { | |
278 | *charset = cs; | |
279 | if (*str < 0xA0) | |
280 | return -1; | |
281 | *c1 = (*str++) & 0x7F; | |
282 | if (CHARSET_DIMENSION (cs) == 2) | |
283 | { | |
284 | if (*str < 0xA0) | |
285 | return -1; | |
286 | *c2 = (*str++) & 0x7F; | |
287 | } | |
288 | } | |
289 | else | |
290 | return -1; | |
291 | return 0; | |
292 | } | |
293 | ||
537efd8d | 294 | /* Translate character C by translation table TABLE. If C |
b4e9dd77 KH |
295 | is negative, translate a character specified by CHARSET, C1, and C2 |
296 | (C1 and C2 are code points of the character). If no translation is | |
297 | found in TABLE, return C. */ | |
dfcf069d | 298 | int |
b4e9dd77 | 299 | translate_char (table, c, charset, c1, c2) |
23d2a7f1 KH |
300 | Lisp_Object table; |
301 | int c, charset, c1, c2; | |
302 | { | |
303 | Lisp_Object ch; | |
304 | int alt_charset, alt_c1, alt_c2, dimension; | |
305 | ||
306 | if (c < 0) c = MAKE_CHAR (charset, c1, c2); | |
307 | if (!CHAR_TABLE_P (table) | |
308 | || (ch = Faref (table, make_number (c)), !INTEGERP (ch)) | |
309 | || XINT (ch) < 0) | |
310 | return c; | |
311 | ||
312 | SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); | |
313 | dimension = CHARSET_DIMENSION (alt_charset); | |
314 | if (dimension == 1 && alt_c1 > 0 || dimension == 2 && alt_c2 > 0) | |
315 | /* CH is not a generic character, just return it. */ | |
316 | return XFASTINT (ch); | |
317 | ||
318 | /* Since CH is a generic character, we must return a specific | |
319 | charater which has the same position codes as C from CH. */ | |
320 | if (charset < 0) | |
321 | SPLIT_CHAR (c, charset, c1, c2); | |
322 | if (dimension != CHARSET_DIMENSION (charset)) | |
323 | /* We can't make such a character because of dimension mismatch. */ | |
324 | return c; | |
23d2a7f1 KH |
325 | return MAKE_CHAR (alt_charset, c1, c2); |
326 | } | |
327 | ||
d2665018 | 328 | /* Convert the unibyte character C to multibyte based on |
b4e9dd77 | 329 | Vnonascii_translation_table or nonascii_insert_offset. If they can't |
d2665018 KH |
330 | convert C to a valid multibyte character, convert it based on |
331 | DEFAULT_NONASCII_INSERT_OFFSET which makes C a Latin-1 character. */ | |
35e623fb | 332 | |
dfcf069d | 333 | int |
35e623fb RS |
334 | unibyte_char_to_multibyte (c) |
335 | int c; | |
336 | { | |
bbf12bb3 | 337 | if (c < 0400) |
35e623fb | 338 | { |
d2665018 KH |
339 | int c_save = c; |
340 | ||
b4e9dd77 | 341 | if (! NILP (Vnonascii_translation_table)) |
bbf12bb3 KH |
342 | { |
343 | c = XINT (Faref (Vnonascii_translation_table, make_number (c))); | |
344 | if (c >= 0400 && ! VALID_MULTIBYTE_CHAR_P (c)) | |
345 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; | |
346 | } | |
347 | else if (c >= 0240 && nonascii_insert_offset > 0) | |
348 | { | |
349 | c += nonascii_insert_offset; | |
350 | if (c < 0400 || ! VALID_MULTIBYTE_CHAR_P (c)) | |
351 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; | |
352 | } | |
353 | else if (c >= 0240) | |
d2665018 | 354 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
35e623fb RS |
355 | } |
356 | return c; | |
357 | } | |
76d7b829 KH |
358 | |
359 | ||
360 | /* Convert the multibyte character C to unibyte 8-bit character based | |
361 | on Vnonascii_translation_table or nonascii_insert_offset. If | |
362 | REV_TBL is non-nil, it should be a reverse table of | |
363 | Vnonascii_translation_table, i.e. what given by: | |
364 | Fchar_table_extra_slot (Vnonascii_translation_table, make_number (0)) */ | |
365 | ||
366 | int | |
367 | multibyte_char_to_unibyte (c, rev_tbl) | |
368 | int c; | |
369 | Lisp_Object rev_tbl; | |
370 | { | |
371 | if (!SINGLE_BYTE_CHAR_P (c)) | |
372 | { | |
373 | int c_save = c; | |
374 | ||
375 | if (! CHAR_TABLE_P (rev_tbl) | |
376 | && CHAR_TABLE_P (Vnonascii_translation_table)) | |
377 | rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table, | |
378 | make_number (0)); | |
379 | if (CHAR_TABLE_P (rev_tbl)) | |
380 | { | |
381 | Lisp_Object temp; | |
382 | temp = Faref (rev_tbl, make_number (c)); | |
383 | if (INTEGERP (temp)) | |
384 | c = XINT (temp); | |
bbf12bb3 KH |
385 | if (c >= 256) |
386 | c = (c_save & 0177) + 0200; | |
387 | } | |
388 | else | |
389 | { | |
390 | if (nonascii_insert_offset > 0) | |
391 | c -= nonascii_insert_offset; | |
392 | if (c < 128 || c >= 256) | |
393 | c = (c_save & 0177) + 0200; | |
76d7b829 | 394 | } |
76d7b829 KH |
395 | } |
396 | ||
397 | return c; | |
398 | } | |
399 | ||
35e623fb | 400 | \f |
4ed46869 KH |
401 | /* Update the table Vcharset_table with the given arguments (see the |
402 | document of `define-charset' for the meaning of each argument). | |
403 | Several other table contents are also updated. The caller should | |
404 | check the validity of CHARSET-ID and the remaining arguments in | |
405 | advance. */ | |
406 | ||
407 | void | |
408 | update_charset_table (charset_id, dimension, chars, width, direction, | |
409 | iso_final_char, iso_graphic_plane, | |
410 | short_name, long_name, description) | |
411 | Lisp_Object charset_id, dimension, chars, width, direction; | |
412 | Lisp_Object iso_final_char, iso_graphic_plane; | |
413 | Lisp_Object short_name, long_name, description; | |
414 | { | |
415 | int charset = XINT (charset_id); | |
416 | int bytes; | |
417 | unsigned char leading_code_base, leading_code_ext; | |
418 | ||
6dc0722d KH |
419 | if (NILP (CHARSET_TABLE_ENTRY (charset))) |
420 | CHARSET_TABLE_ENTRY (charset) | |
421 | = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); | |
4ed46869 KH |
422 | |
423 | /* Get byte length of multibyte form, base leading-code, and | |
424 | extended leading-code of the charset. See the comment under the | |
425 | title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
426 | bytes = XINT (dimension); | |
427 | if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
428 | { | |
429 | /* Official charset, it doesn't have an extended leading-code. */ | |
430 | if (charset != CHARSET_ASCII) | |
431 | bytes += 1; /* For a base leading-code. */ | |
432 | leading_code_base = charset; | |
433 | leading_code_ext = 0; | |
434 | } | |
435 | else | |
436 | { | |
437 | /* Private charset. */ | |
438 | bytes += 2; /* For base and extended leading-codes. */ | |
439 | leading_code_base | |
440 | = (charset < LEADING_CODE_EXT_12 | |
441 | ? LEADING_CODE_PRIVATE_11 | |
442 | : (charset < LEADING_CODE_EXT_21 | |
443 | ? LEADING_CODE_PRIVATE_12 | |
444 | : (charset < LEADING_CODE_EXT_22 | |
445 | ? LEADING_CODE_PRIVATE_21 | |
446 | : LEADING_CODE_PRIVATE_22))); | |
447 | leading_code_ext = charset; | |
448 | } | |
449 | ||
450 | CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; | |
451 | CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
452 | CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
453 | CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
454 | CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
455 | CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
456 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
457 | = make_number (leading_code_base); | |
458 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
459 | = make_number (leading_code_ext); | |
460 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
461 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
462 | = iso_graphic_plane; | |
463 | CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
464 | CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
465 | CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
466 | CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
467 | ||
468 | { | |
469 | /* If we have already defined a charset which has the same | |
470 | DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
471 | DIRECTION, we must update the entry REVERSE-CHARSET of both | |
472 | charsets. If there's no such charset, the value of the entry | |
473 | is set to nil. */ | |
474 | int i; | |
475 | ||
513ee442 | 476 | for (i = 0; i <= MAX_CHARSET; i++) |
4ed46869 KH |
477 | if (!NILP (CHARSET_TABLE_ENTRY (i))) |
478 | { | |
479 | if (CHARSET_DIMENSION (i) == XINT (dimension) | |
480 | && CHARSET_CHARS (i) == XINT (chars) | |
481 | && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
482 | && CHARSET_DIRECTION (i) != XINT (direction)) | |
483 | { | |
484 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
485 | = make_number (i); | |
486 | CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
487 | break; | |
488 | } | |
489 | } | |
513ee442 | 490 | if (i > MAX_CHARSET) |
4ed46869 KH |
491 | /* No such a charset. */ |
492 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
493 | = make_number (-1); | |
494 | } | |
495 | ||
496 | if (charset != CHARSET_ASCII | |
497 | && charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
498 | { | |
499 | /* Update tables bytes_by_char_head and width_by_char_head. */ | |
500 | bytes_by_char_head[leading_code_base] = bytes; | |
501 | width_by_char_head[leading_code_base] = XINT (width); | |
502 | ||
503 | /* Update table emacs_code_class. */ | |
504 | emacs_code_class[charset] = (bytes == 2 | |
505 | ? EMACS_leading_code_2 | |
506 | : (bytes == 3 | |
507 | ? EMACS_leading_code_3 | |
508 | : EMACS_leading_code_4)); | |
509 | } | |
510 | ||
511 | /* Update table iso_charset_table. */ | |
512 | if (ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0) | |
513 | ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset; | |
514 | } | |
515 | ||
516 | #ifdef emacs | |
517 | ||
518 | /* Return charset id of CHARSET_SYMBOL, or return -1 if CHARSET_SYMBOL | |
519 | is invalid. */ | |
520 | int | |
521 | get_charset_id (charset_symbol) | |
522 | Lisp_Object charset_symbol; | |
523 | { | |
524 | Lisp_Object val; | |
525 | int charset; | |
526 | ||
527 | return ((SYMBOLP (charset_symbol) | |
528 | && (val = Fget (charset_symbol, Qcharset), VECTORP (val)) | |
529 | && (charset = XINT (XVECTOR (val)->contents[CHARSET_ID_IDX]), | |
530 | CHARSET_VALID_P (charset))) | |
531 | ? charset : -1); | |
532 | } | |
533 | ||
534 | /* Return an identification number for a new private charset of | |
535 | DIMENSION and WIDTH. If there's no more room for the new charset, | |
536 | return 0. */ | |
537 | Lisp_Object | |
538 | get_new_private_charset_id (dimension, width) | |
539 | int dimension, width; | |
540 | { | |
541 | int charset, from, to; | |
542 | ||
543 | if (dimension == 1) | |
544 | { | |
545 | if (width == 1) | |
546 | from = LEADING_CODE_EXT_11, to = LEADING_CODE_EXT_12; | |
547 | else | |
548 | from = LEADING_CODE_EXT_12, to = LEADING_CODE_EXT_21; | |
549 | } | |
550 | else | |
551 | { | |
552 | if (width == 1) | |
553 | from = LEADING_CODE_EXT_21, to = LEADING_CODE_EXT_22; | |
554 | else | |
b0e3cf2b | 555 | from = LEADING_CODE_EXT_22, to = LEADING_CODE_EXT_MAX + 1; |
4ed46869 KH |
556 | } |
557 | ||
558 | for (charset = from; charset < to; charset++) | |
559 | if (!CHARSET_DEFINED_P (charset)) break; | |
560 | ||
561 | return make_number (charset < to ? charset : 0); | |
562 | } | |
563 | ||
564 | DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
565 | "Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR.\n\ | |
23d2a7f1 | 566 | If CHARSET-ID is nil, it is decided automatically, which means CHARSET is\n\ |
4ed46869 KH |
567 | treated as a private charset.\n\ |
568 | INFO-VECTOR is a vector of the format:\n\ | |
569 | [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE\n\ | |
570 | SHORT-NAME LONG-NAME DESCRIPTION]\n\ | |
571 | The meanings of each elements is as follows:\n\ | |
572 | DIMENSION (integer) is the number of bytes to represent a character: 1 or 2.\n\ | |
573 | CHARS (integer) is the number of characters in a dimension: 94 or 96.\n\ | |
574 | WIDTH (integer) is the number of columns a character in the charset\n\ | |
575 | occupies on the screen: one of 0, 1, and 2.\n\ | |
576 | \n\ | |
577 | DIRECTION (integer) is the rendering direction of characters in the\n\ | |
277576f6 KH |
578 | charset when rendering. If 0, render from left to right, else\n\ |
579 | render from right to left.\n\ | |
4ed46869 KH |
580 | \n\ |
581 | ISO-FINAL-CHAR (character) is the final character of the\n\ | |
582 | corresponding ISO 2022 charset.\n\ | |
583 | \n\ | |
584 | ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked\n\ | |
585 | while encoding to variants of ISO 2022 coding system, one of the\n\ | |
586 | following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).\n\ | |
587 | \n\ | |
588 | SHORT-NAME (string) is the short name to refer to the charset.\n\ | |
589 | \n\ | |
590 | LONG-NAME (string) is the long name to refer to the charset.\n\ | |
591 | \n\ | |
592 | DESCRIPTION (string) is the description string of the charset.") | |
593 | (charset_id, charset_symbol, info_vector) | |
594 | Lisp_Object charset_id, charset_symbol, info_vector; | |
595 | { | |
596 | Lisp_Object *vec; | |
597 | ||
598 | if (!NILP (charset_id)) | |
599 | CHECK_NUMBER (charset_id, 0); | |
600 | CHECK_SYMBOL (charset_symbol, 1); | |
601 | CHECK_VECTOR (info_vector, 2); | |
602 | ||
603 | if (! NILP (charset_id)) | |
604 | { | |
605 | if (! CHARSET_VALID_P (XINT (charset_id))) | |
606 | error ("Invalid CHARSET: %d", XINT (charset_id)); | |
607 | else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
608 | error ("Already defined charset: %d", XINT (charset_id)); | |
609 | } | |
610 | ||
611 | vec = XVECTOR (info_vector)->contents; | |
612 | if (XVECTOR (info_vector)->size != 9 | |
613 | || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
614 | || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
615 | || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
616 | || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
617 | || !INTEGERP (vec[4]) || !(XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~') | |
618 | || !INTEGERP (vec[5]) || !(XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
619 | || !STRINGP (vec[6]) | |
620 | || !STRINGP (vec[7]) | |
621 | || !STRINGP (vec[8])) | |
622 | error ("Invalid info-vector argument for defining charset %s", | |
623 | XSYMBOL (charset_symbol)->name->data); | |
624 | ||
625 | if (NILP (charset_id)) | |
626 | { | |
627 | charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
628 | if (XINT (charset_id) == 0) | |
629 | error ("There's no room for a new private charset %s", | |
630 | XSYMBOL (charset_symbol)->name->data); | |
631 | } | |
632 | ||
633 | update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
634 | vec[4], vec[5], vec[6], vec[7], vec[8]); | |
6dc0722d | 635 | Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
4ed46869 KH |
636 | CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
637 | Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
638 | return Qnil; | |
639 | } | |
640 | ||
8a73a704 KH |
641 | DEFUN ("generic-character-list", Fgeneric_character_list, |
642 | Sgeneric_character_list, 0, 0, 0, | |
643 | "Return a list of all possible generic characters.\n\ | |
644 | It includes a generic character for a charset not yet defined.") | |
645 | () | |
646 | { | |
647 | return Vgeneric_character_list; | |
648 | } | |
649 | ||
3fac5a51 KH |
650 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
651 | Sget_unused_iso_final_char, 2, 2, 0, | |
652 | "Return an unsed ISO's final char for a charset of DIMENISION and CHARS.\n\ | |
653 | DIMENSION is the number of bytes to represent a character: 1 or 2.\n\ | |
654 | CHARS is the number of characters in a dimension: 94 or 96.\n\ | |
655 | \n\ | |
656 | This final char is for private use, thus the range is `0' (48) .. `?' (63).\n\ | |
657 | If there's no unused final char for the specified kind of charset,\n\ | |
658 | return nil.") | |
659 | (dimension, chars) | |
660 | Lisp_Object dimension, chars; | |
661 | { | |
662 | int final_char; | |
663 | ||
664 | CHECK_NUMBER (dimension, 0); | |
665 | CHECK_NUMBER (chars, 1); | |
666 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
667 | error ("Invalid charset dimension %d, it should be 1 or 2", | |
668 | XINT (dimension)); | |
669 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
670 | error ("Invalid charset chars %d, it should be 94 or 96", | |
671 | XINT (chars)); | |
672 | for (final_char = '0'; final_char <= '?'; final_char++) | |
673 | { | |
674 | if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0) | |
675 | break; | |
676 | } | |
677 | return (final_char <= '?' ? make_number (final_char) : Qnil); | |
678 | } | |
679 | ||
4ed46869 KH |
680 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
681 | 4, 4, 0, | |
682 | "Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET.\n\ | |
683 | CHARSET should be defined by `defined-charset' in advance.") | |
684 | (dimension, chars, final_char, charset_symbol) | |
685 | Lisp_Object dimension, chars, final_char, charset_symbol; | |
686 | { | |
687 | int charset; | |
688 | ||
689 | CHECK_NUMBER (dimension, 0); | |
690 | CHECK_NUMBER (chars, 1); | |
691 | CHECK_NUMBER (final_char, 2); | |
692 | CHECK_SYMBOL (charset_symbol, 3); | |
693 | ||
694 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
695 | error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
696 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
697 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
698 | if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
699 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
700 | if ((charset = get_charset_id (charset_symbol)) < 0) | |
701 | error ("Invalid charset %s", XSYMBOL (charset_symbol)->name->data); | |
702 | ||
703 | ISO_CHARSET_TABLE (dimension, chars, final_char) = charset; | |
704 | return Qnil; | |
705 | } | |
706 | ||
707 | /* Return number of different charsets in STR of length LEN. In | |
708 | addition, for each found charset N, CHARSETS[N] is set 1. The | |
a29e3b1b | 709 | caller should allocate CHARSETS (MAX_CHARSET + 1 elements) in advance. |
1d67c29b KH |
710 | It may lookup a translation table TABLE if supplied. |
711 | ||
712 | If CMPCHARP is nonzero and some composite character is found, | |
713 | CHARSETS[128] is also set 1 and the returned number is incremented | |
714 | by 1. */ | |
4ed46869 KH |
715 | |
716 | int | |
1d67c29b | 717 | find_charset_in_str (str, len, charsets, table, cmpcharp) |
028d516b KH |
718 | unsigned char *str; |
719 | int len, *charsets; | |
23d2a7f1 | 720 | Lisp_Object table; |
1d67c29b | 721 | int cmpcharp; |
4ed46869 | 722 | { |
733eafd8 | 723 | register int num = 0, c; |
4ed46869 | 724 | |
23d2a7f1 KH |
725 | if (! CHAR_TABLE_P (table)) |
726 | table = Qnil; | |
727 | ||
4ed46869 KH |
728 | while (len > 0) |
729 | { | |
05505664 | 730 | int bytes, charset; |
733eafd8 | 731 | c = *str; |
23d2a7f1 | 732 | |
733eafd8 | 733 | if (c == LEADING_CODE_COMPOSITION) |
05505664 | 734 | { |
733eafd8 KH |
735 | int cmpchar_id = str_cmpchar_id (str, len); |
736 | GLYPH *glyph; | |
05505664 | 737 | |
1d67c29b | 738 | if (cmpchar_id >= 0) |
05505664 | 739 | { |
020da460 | 740 | struct cmpchar_info *cmp_p = cmpchar_table[cmpchar_id]; |
733eafd8 KH |
741 | int i; |
742 | ||
020da460 | 743 | for (i = 0; i < cmp_p->glyph_len; i++) |
733eafd8 | 744 | { |
020da460 | 745 | c = cmp_p->glyph[i]; |
733eafd8 KH |
746 | if (!NILP (table)) |
747 | { | |
b4e9dd77 | 748 | if ((c = translate_char (table, c, 0, 0, 0)) < 0) |
020da460 | 749 | c = cmp_p->glyph[i]; |
733eafd8 KH |
750 | } |
751 | if ((charset = CHAR_CHARSET (c)) < 0) | |
752 | charset = CHARSET_ASCII; | |
753 | if (!charsets[charset]) | |
754 | { | |
755 | charsets[charset] = 1; | |
756 | num += 1; | |
757 | } | |
758 | } | |
020da460 KH |
759 | str += cmp_p->len; |
760 | len -= cmp_p->len; | |
761 | if (cmpcharp && !charsets[CHARSET_COMPOSITION]) | |
762 | { | |
763 | charsets[CHARSET_COMPOSITION] = 1; | |
764 | num += 1; | |
765 | } | |
733eafd8 | 766 | continue; |
05505664 | 767 | } |
05505664 | 768 | |
733eafd8 KH |
769 | charset = CHARSET_ASCII; |
770 | bytes = 1; | |
771 | } | |
23d2a7f1 KH |
772 | else |
773 | { | |
733eafd8 KH |
774 | c = STRING_CHAR_AND_LENGTH (str, len, bytes); |
775 | if (! NILP (table)) | |
776 | { | |
b4e9dd77 | 777 | int c1 = translate_char (table, c, 0, 0, 0); |
733eafd8 KH |
778 | if (c1 >= 0) |
779 | c = c1; | |
780 | } | |
781 | charset = CHAR_CHARSET (c); | |
23d2a7f1 | 782 | } |
4ed46869 KH |
783 | |
784 | if (!charsets[charset]) | |
785 | { | |
786 | charsets[charset] = 1; | |
787 | num += 1; | |
788 | } | |
789 | str += bytes; | |
790 | len -= bytes; | |
791 | } | |
792 | return num; | |
793 | } | |
794 | ||
795 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 796 | 2, 3, 0, |
4ed46869 | 797 | "Return a list of charsets in the region between BEG and END.\n\ |
23d2a7f1 | 798 | BEG and END are buffer positions.\n\ |
020da460 KH |
799 | If the region contains any composite character,\n\ |
800 | `composition' is included in the returned list.\n\ | |
b4e9dd77 | 801 | Optional arg TABLE if non-nil is a translation table to look up.") |
23d2a7f1 KH |
802 | (beg, end, table) |
803 | Lisp_Object beg, end, table; | |
4ed46869 | 804 | { |
028d516b | 805 | int charsets[MAX_CHARSET + 1]; |
6ae1f27e | 806 | int from, from_byte, to, stop, stop_byte, i; |
4ed46869 KH |
807 | Lisp_Object val; |
808 | ||
809 | validate_region (&beg, &end); | |
810 | from = XFASTINT (beg); | |
811 | stop = to = XFASTINT (end); | |
6ae1f27e | 812 | |
2e45bf18 KH |
813 | if (NILP (current_buffer->enable_multibyte_characters)) |
814 | return (from == to | |
815 | ? Qnil | |
816 | : Fcons (Qascii, Qnil)); | |
817 | ||
4ed46869 | 818 | if (from < GPT && GPT < to) |
6ae1f27e RS |
819 | { |
820 | stop = GPT; | |
821 | stop_byte = GPT_BYTE; | |
822 | } | |
823 | else | |
824 | stop_byte = CHAR_TO_BYTE (stop); | |
825 | ||
826 | from_byte = CHAR_TO_BYTE (from); | |
827 | ||
028d516b | 828 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
4ed46869 KH |
829 | while (1) |
830 | { | |
6ae1f27e | 831 | find_charset_in_str (BYTE_POS_ADDR (from_byte), stop_byte - from_byte, |
020da460 | 832 | charsets, table, 1); |
4ed46869 | 833 | if (stop < to) |
6ae1f27e RS |
834 | { |
835 | from = stop, from_byte = stop_byte; | |
836 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
837 | } | |
4ed46869 KH |
838 | else |
839 | break; | |
840 | } | |
6ae1f27e | 841 | |
4ed46869 | 842 | val = Qnil; |
513ee442 | 843 | for (i = MAX_CHARSET; i >= 0; i--) |
4ed46869 KH |
844 | if (charsets[i]) |
845 | val = Fcons (CHARSET_SYMBOL (i), val); | |
846 | return val; | |
847 | } | |
848 | ||
849 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 KH |
850 | 1, 2, 0, |
851 | "Return a list of charsets in STR.\n\ | |
020da460 KH |
852 | If the string contains any composite characters,\n\ |
853 | `composition' is included in the returned list.\n\ | |
b4e9dd77 | 854 | Optional arg TABLE if non-nil is a translation table to look up.") |
23d2a7f1 KH |
855 | (str, table) |
856 | Lisp_Object str, table; | |
4ed46869 | 857 | { |
a29e3b1b | 858 | int charsets[MAX_CHARSET + 1]; |
4ed46869 KH |
859 | int i; |
860 | Lisp_Object val; | |
861 | ||
862 | CHECK_STRING (str, 0); | |
87b089ad RS |
863 | |
864 | if (! STRING_MULTIBYTE (str)) | |
2e45bf18 KH |
865 | return (XSTRING (str)->size == 0 |
866 | ? Qnil | |
867 | : Fcons (Qascii, Qnil)); | |
87b089ad | 868 | |
a29e3b1b | 869 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
fc932ac6 | 870 | find_charset_in_str (XSTRING (str)->data, STRING_BYTES (XSTRING (str)), |
020da460 | 871 | charsets, table, 1); |
4ed46869 | 872 | val = Qnil; |
513ee442 | 873 | for (i = MAX_CHARSET; i >= 0; i--) |
4ed46869 KH |
874 | if (charsets[i]) |
875 | val = Fcons (CHARSET_SYMBOL (i), val); | |
876 | return val; | |
877 | } | |
878 | \f | |
879 | DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
513ee442 | 880 | "") |
4ed46869 KH |
881 | (charset, code1, code2) |
882 | Lisp_Object charset, code1, code2; | |
883 | { | |
884 | CHECK_NUMBER (charset, 0); | |
885 | ||
886 | if (NILP (code1)) | |
887 | XSETFASTINT (code1, 0); | |
888 | else | |
889 | CHECK_NUMBER (code1, 1); | |
890 | if (NILP (code2)) | |
891 | XSETFASTINT (code2, 0); | |
892 | else | |
893 | CHECK_NUMBER (code2, 2); | |
894 | ||
895 | if (!CHARSET_DEFINED_P (XINT (charset))) | |
896 | error ("Invalid charset: %d", XINT (charset)); | |
897 | ||
898 | return make_number (MAKE_CHAR (XINT (charset), XINT (code1), XINT (code2))); | |
899 | } | |
900 | ||
901 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, | |
902 | "Return list of charset and one or two position-codes of CHAR.") | |
903 | (ch) | |
904 | Lisp_Object ch; | |
905 | { | |
906 | Lisp_Object val; | |
6dc0722d | 907 | int charset, c1, c2; |
4ed46869 KH |
908 | |
909 | CHECK_NUMBER (ch, 0); | |
910 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
6dc0722d | 911 | return (c2 >= 0 |
4ed46869 KH |
912 | ? Fcons (CHARSET_SYMBOL (charset), |
913 | Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
914 | : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
915 | } | |
916 | ||
917 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
918 | "Return charset of CHAR.") | |
919 | (ch) | |
920 | Lisp_Object ch; | |
921 | { | |
922 | CHECK_NUMBER (ch, 0); | |
923 | ||
924 | return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
925 | } | |
926 | ||
90d7b74e KH |
927 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
928 | "Return charset of a character in current buffer at position POS.\n\ | |
929 | If POS is nil, it defauls to the current point.") | |
930 | (pos) | |
931 | Lisp_Object pos; | |
932 | { | |
933 | register int pos_byte, c, charset; | |
934 | register unsigned char *p; | |
935 | ||
936 | if (NILP (pos)) | |
937 | pos_byte = PT_BYTE; | |
938 | else if (MARKERP (pos)) | |
939 | pos_byte = marker_byte_position (pos); | |
940 | else | |
941 | { | |
942 | CHECK_NUMBER (pos, 0); | |
943 | pos_byte = CHAR_TO_BYTE (XINT (pos)); | |
944 | } | |
945 | p = BYTE_POS_ADDR (pos_byte); | |
946 | c = STRING_CHAR (p, Z_BYTE - pos_byte); | |
947 | charset = CHAR_CHARSET (c); | |
948 | return CHARSET_SYMBOL (charset); | |
949 | } | |
950 | ||
4ed46869 | 951 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
2b71bb78 KH |
952 | "Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR.\n\ |
953 | \n\ | |
954 | ISO 2022's designation sequence (escape sequence) distinguishes charsets\n\ | |
955 | by their DIMENSION, CHARS, and FINAL-CHAR,\n\ | |
956 | where as Emacs distinguishes them by charset symbol.\n\ | |
957 | See the documentation of the function `charset-info' for the meanings of\n\ | |
958 | DIMENSION, CHARS, and FINAL-CHAR.") | |
4ed46869 KH |
959 | (dimension, chars, final_char) |
960 | Lisp_Object dimension, chars, final_char; | |
961 | { | |
962 | int charset; | |
963 | ||
964 | CHECK_NUMBER (dimension, 0); | |
965 | CHECK_NUMBER (chars, 1); | |
966 | CHECK_NUMBER (final_char, 2); | |
967 | ||
968 | if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
969 | return Qnil; | |
970 | return CHARSET_SYMBOL (charset); | |
971 | } | |
972 | ||
9d3d8cba KH |
973 | /* If GENERICP is nonzero, return nonzero iff C is a valid normal or |
974 | generic character. If GENERICP is zero, return nonzero iff C is a | |
975 | valid normal character. Do not call this function directly, | |
976 | instead use macro CHAR_VALID_P. */ | |
977 | int | |
978 | char_valid_p (c, genericp) | |
979 | int c, genericp; | |
980 | { | |
981 | int charset, c1, c2; | |
982 | ||
983 | if (c < 0) | |
984 | return 0; | |
985 | if (SINGLE_BYTE_CHAR_P (c)) | |
986 | return 1; | |
987 | SPLIT_NON_ASCII_CHAR (c, charset, c1, c2); | |
988 | if (!CHARSET_VALID_P (charset)) | |
989 | return 0; | |
990 | return (c < MIN_CHAR_COMPOSITION | |
991 | ? ((c & CHAR_FIELD1_MASK) /* i.e. dimension of C is two. */ | |
992 | ? (genericp && c1 == 0 && c2 == 0 | |
993 | || c1 >= 32 && c2 >= 32) | |
994 | : (genericp && c1 == 0 | |
995 | || c1 >= 32)) | |
996 | : c < MIN_CHAR_COMPOSITION + n_cmpchars); | |
997 | } | |
998 | ||
999 | DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0, | |
a9d02884 DL |
1000 | "Return t if OBJECT is a valid normal character.\n\ |
1001 | If optional arg GENERICP is non-nil, also return t if OBJECT is\n\ | |
9d3d8cba KH |
1002 | a valid generic character.") |
1003 | (object, genericp) | |
1004 | Lisp_Object object, genericp; | |
1005 | { | |
1006 | if (! NATNUMP (object)) | |
1007 | return Qnil; | |
1008 | return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil); | |
1009 | } | |
1010 | ||
d2665018 KH |
1011 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, |
1012 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
1013 | "Convert the unibyte character CH to multibyte character.\n\ | |
537efd8d | 1014 | The conversion is done based on `nonascii-translation-table' (which see)\n\ |
340b8d58 | 1015 | or `nonascii-insert-offset' (which see).") |
d2665018 KH |
1016 | (ch) |
1017 | Lisp_Object ch; | |
1018 | { | |
1019 | int c; | |
1020 | ||
1021 | CHECK_NUMBER (ch, 0); | |
1022 | c = XINT (ch); | |
1023 | if (c < 0 || c >= 0400) | |
1024 | error ("Invalid unibyte character: %d", c); | |
1025 | c = unibyte_char_to_multibyte (c); | |
1026 | if (c < 0) | |
1027 | error ("Can't convert to multibyte character: %d", XINT (ch)); | |
1028 | return make_number (c); | |
1029 | } | |
1030 | ||
1bcc1567 RS |
1031 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, |
1032 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
1033 | "Convert the multibyte character CH to unibyte character.\n\ | |
1034 | The conversion is done based on `nonascii-translation-table' (which see)\n\ | |
1035 | or `nonascii-insert-offset' (which see).") | |
1036 | (ch) | |
1037 | Lisp_Object ch; | |
1038 | { | |
1039 | int c; | |
1040 | ||
1041 | CHECK_NUMBER (ch, 0); | |
1042 | c = XINT (ch); | |
1043 | if (c < 0) | |
1044 | error ("Invalid multibyte character: %d", c); | |
1045 | c = multibyte_char_to_unibyte (c, Qnil); | |
1046 | if (c < 0) | |
1047 | error ("Can't convert to unibyte character: %d", XINT (ch)); | |
1048 | return make_number (c); | |
1049 | } | |
1050 | ||
4ed46869 | 1051 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, |
f78643ef | 1052 | "Return 1 regardless of the argument CHAR.\n\ |
9b6a601f | 1053 | This is now an obsolte function. We keep is just for backward compatibility.") |
4ed46869 KH |
1054 | (ch) |
1055 | Lisp_Object ch; | |
1056 | { | |
1057 | Lisp_Object val; | |
4ed46869 KH |
1058 | |
1059 | CHECK_NUMBER (ch, 0); | |
9b6a601f KH |
1060 | return make_number (1); |
1061 | } | |
1062 | ||
1063 | /* Return how many bytes C will occupy in a multibyte buffer. | |
1064 | Don't call this function directly, instead use macro CHAR_BYTES. */ | |
1065 | int | |
1066 | char_bytes (c) | |
1067 | int c; | |
1068 | { | |
1069 | int bytes; | |
1070 | ||
1071 | if (COMPOSITE_CHAR_P (c)) | |
4ed46869 | 1072 | { |
9b6a601f | 1073 | unsigned int id = COMPOSITE_CHAR_ID (c); |
4ed46869 KH |
1074 | |
1075 | bytes = (id < n_cmpchars ? cmpchar_table[id]->len : 1); | |
1076 | } | |
1077 | else | |
1078 | { | |
9b6a601f | 1079 | int charset = CHAR_CHARSET (c); |
4ed46869 KH |
1080 | |
1081 | bytes = CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1; | |
1082 | } | |
1083 | ||
9b6a601f | 1084 | return make_number (bytes); |
4ed46869 KH |
1085 | } |
1086 | ||
1087 | /* Return the width of character of which multi-byte form starts with | |
1088 | C. The width is measured by how many columns occupied on the | |
1089 | screen when displayed in the current buffer. */ | |
1090 | ||
1091 | #define ONE_BYTE_CHAR_WIDTH(c) \ | |
1092 | (c < 0x20 \ | |
1093 | ? (c == '\t' \ | |
53316e55 | 1094 | ? XFASTINT (current_buffer->tab_width) \ |
4ed46869 KH |
1095 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
1096 | : (c < 0x7f \ | |
1097 | ? 1 \ | |
1098 | : (c == 0x7F \ | |
1099 | ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
1100 | : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
1101 | && BASE_LEADING_CODE_P (c)) \ | |
1102 | ? WIDTH_BY_CHAR_HEAD (c) \ | |
b4e9dd77 | 1103 | : 4)))) |
4ed46869 KH |
1104 | |
1105 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
1106 | "Return width of CHAR when displayed in the current buffer.\n\ | |
1107 | The width is measured by how many columns it occupies on the screen.") | |
1108 | (ch) | |
1109 | Lisp_Object ch; | |
1110 | { | |
859f2b3c | 1111 | Lisp_Object val, disp; |
4ed46869 | 1112 | int c; |
51c4025f | 1113 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 KH |
1114 | |
1115 | CHECK_NUMBER (ch, 0); | |
1116 | ||
859f2b3c RS |
1117 | c = XINT (ch); |
1118 | ||
1119 | /* Get the way the display table would display it. */ | |
51c4025f | 1120 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; |
859f2b3c RS |
1121 | |
1122 | if (VECTORP (disp)) | |
1123 | XSETINT (val, XVECTOR (disp)->size); | |
1124 | else if (SINGLE_BYTE_CHAR_P (c)) | |
1125 | XSETINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
4ed46869 KH |
1126 | else if (COMPOSITE_CHAR_P (c)) |
1127 | { | |
1128 | int id = COMPOSITE_CHAR_ID (XFASTINT (ch)); | |
1129 | XSETFASTINT (val, (id < n_cmpchars ? cmpchar_table[id]->width : 0)); | |
1130 | } | |
1131 | else | |
1132 | { | |
1133 | int charset = CHAR_CHARSET (c); | |
1134 | ||
1135 | XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
1136 | } | |
1137 | return val; | |
1138 | } | |
1139 | ||
1140 | /* Return width of string STR of length LEN when displayed in the | |
1141 | current buffer. The width is measured by how many columns it | |
1142 | occupies on the screen. */ | |
859f2b3c | 1143 | |
4ed46869 KH |
1144 | int |
1145 | strwidth (str, len) | |
1146 | unsigned char *str; | |
1147 | int len; | |
1148 | { | |
1149 | unsigned char *endp = str + len; | |
1150 | int width = 0; | |
c4a4e28f | 1151 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1152 | |
859f2b3c RS |
1153 | while (str < endp) |
1154 | { | |
1155 | if (*str == LEADING_CODE_COMPOSITION) | |
1156 | { | |
1157 | int id = str_cmpchar_id (str, endp - str); | |
1158 | ||
1159 | if (id < 0) | |
1160 | { | |
1161 | width += 4; | |
1162 | str++; | |
1163 | } | |
1164 | else | |
1165 | { | |
1166 | width += cmpchar_table[id]->width; | |
1167 | str += cmpchar_table[id]->len; | |
1168 | } | |
1169 | } | |
1170 | else | |
1171 | { | |
1172 | Lisp_Object disp; | |
e515b0a9 KH |
1173 | int thislen; |
1174 | int c = STRING_CHAR_AND_LENGTH (str, endp - str, thislen); | |
859f2b3c RS |
1175 | |
1176 | /* Get the way the display table would display it. */ | |
acc35c36 RS |
1177 | if (dp) |
1178 | disp = DISP_CHAR_VECTOR (dp, c); | |
1179 | else | |
1180 | disp = Qnil; | |
859f2b3c RS |
1181 | |
1182 | if (VECTORP (disp)) | |
e515b0a9 | 1183 | width += XVECTOR (disp)->size; |
859f2b3c | 1184 | else |
e515b0a9 | 1185 | width += ONE_BYTE_CHAR_WIDTH (*str); |
859f2b3c | 1186 | |
e515b0a9 | 1187 | str += thislen; |
859f2b3c RS |
1188 | } |
1189 | } | |
4ed46869 KH |
1190 | return width; |
1191 | } | |
1192 | ||
1193 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
1194 | "Return width of STRING when displayed in the current buffer.\n\ | |
1195 | Width is measured by how many columns it occupies on the screen.\n\ | |
046b1f03 RS |
1196 | When calculating width of a multibyte character in STRING,\n\ |
1197 | only the base leading-code is considered; the validity of\n\ | |
1198 | the following bytes is not checked.") | |
4ed46869 KH |
1199 | (str) |
1200 | Lisp_Object str; | |
1201 | { | |
1202 | Lisp_Object val; | |
1203 | ||
1204 | CHECK_STRING (str, 0); | |
fc932ac6 RS |
1205 | XSETFASTINT (val, strwidth (XSTRING (str)->data, |
1206 | STRING_BYTES (XSTRING (str)))); | |
4ed46869 KH |
1207 | return val; |
1208 | } | |
1209 | ||
1210 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
1211 | "Return the direction of CHAR.\n\ | |
1212 | The returned value is 0 for left-to-right and 1 for right-to-left.") | |
1213 | (ch) | |
1214 | Lisp_Object ch; | |
1215 | { | |
1216 | int charset; | |
1217 | ||
1218 | CHECK_NUMBER (ch, 0); | |
1219 | charset = CHAR_CHARSET (XFASTINT (ch)); | |
1220 | if (!CHARSET_DEFINED_P (charset)) | |
93bcb785 | 1221 | invalid_character (XINT (ch)); |
4ed46869 KH |
1222 | return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); |
1223 | } | |
1224 | ||
af4fecb4 | 1225 | DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, |
6ae1f27e | 1226 | "Return number of characters between BEG and END.") |
046b1f03 RS |
1227 | (beg, end) |
1228 | Lisp_Object beg, end; | |
1229 | { | |
6ae1f27e | 1230 | int from, to; |
046b1f03 | 1231 | |
17e7ef1b RS |
1232 | CHECK_NUMBER_COERCE_MARKER (beg, 0); |
1233 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
1234 | ||
046b1f03 | 1235 | from = min (XFASTINT (beg), XFASTINT (end)); |
a8a35e61 | 1236 | to = max (XFASTINT (beg), XFASTINT (end)); |
046b1f03 | 1237 | |
a8c21066 | 1238 | return make_number (to - from); |
6ae1f27e | 1239 | } |
9036eb45 | 1240 | |
87b089ad RS |
1241 | /* Return the number of characters in the NBYTES bytes at PTR. |
1242 | This works by looking at the contents and checking for multibyte sequences. | |
1243 | However, if the current buffer has enable-multibyte-characters = nil, | |
1244 | we treat each byte as a character. */ | |
1245 | ||
6ae1f27e RS |
1246 | int |
1247 | chars_in_text (ptr, nbytes) | |
1248 | unsigned char *ptr; | |
1249 | int nbytes; | |
1250 | { | |
93bcb785 | 1251 | unsigned char *endp, c; |
6ae1f27e | 1252 | int chars; |
046b1f03 | 1253 | |
87b089ad RS |
1254 | /* current_buffer is null at early stages of Emacs initialization. */ |
1255 | if (current_buffer == 0 | |
1256 | || NILP (current_buffer->enable_multibyte_characters)) | |
6ae1f27e | 1257 | return nbytes; |
a8a35e61 | 1258 | |
6ae1f27e RS |
1259 | endp = ptr + nbytes; |
1260 | chars = 0; | |
046b1f03 | 1261 | |
6ae1f27e RS |
1262 | while (ptr < endp) |
1263 | { | |
93bcb785 KH |
1264 | c = *ptr++; |
1265 | ||
1266 | if (BASE_LEADING_CODE_P (c)) | |
1267 | while (ptr < endp && ! CHAR_HEAD_P (*ptr)) ptr++; | |
046b1f03 RS |
1268 | chars++; |
1269 | } | |
1270 | ||
6ae1f27e | 1271 | return chars; |
046b1f03 RS |
1272 | } |
1273 | ||
87b089ad RS |
1274 | /* Return the number of characters in the NBYTES bytes at PTR. |
1275 | This works by looking at the contents and checking for multibyte sequences. | |
1276 | It ignores enable-multibyte-characters. */ | |
1277 | ||
1278 | int | |
1279 | multibyte_chars_in_text (ptr, nbytes) | |
1280 | unsigned char *ptr; | |
1281 | int nbytes; | |
1282 | { | |
93bcb785 | 1283 | unsigned char *endp, c; |
87b089ad RS |
1284 | int chars; |
1285 | ||
1286 | endp = ptr + nbytes; | |
1287 | chars = 0; | |
1288 | ||
1289 | while (ptr < endp) | |
1290 | { | |
93bcb785 KH |
1291 | c = *ptr++; |
1292 | ||
1293 | if (BASE_LEADING_CODE_P (c)) | |
1294 | while (ptr < endp && ! CHAR_HEAD_P (*ptr)) ptr++; | |
87b089ad RS |
1295 | chars++; |
1296 | } | |
1297 | ||
1298 | return chars; | |
1299 | } | |
1300 | ||
1301 | DEFUN ("string", Fstring, Sstring, 1, MANY, 0, | |
4ed46869 | 1302 | "Concatenate all the argument characters and make the result a string.") |
53316e55 KH |
1303 | (n, args) |
1304 | int n; | |
4ed46869 KH |
1305 | Lisp_Object *args; |
1306 | { | |
53316e55 | 1307 | int i; |
4ed46869 | 1308 | unsigned char *buf |
bd4c6dd0 | 1309 | = (unsigned char *) alloca (MAX_LENGTH_OF_MULTI_BYTE_FORM * n); |
4ed46869 KH |
1310 | unsigned char *p = buf; |
1311 | Lisp_Object val; | |
1312 | ||
1313 | for (i = 0; i < n; i++) | |
1314 | { | |
1315 | int c, len; | |
1316 | unsigned char *str; | |
1317 | ||
1318 | if (!INTEGERP (args[i])) | |
b0e3cf2b | 1319 | CHECK_NUMBER (args[i], 0); |
4ed46869 KH |
1320 | c = XINT (args[i]); |
1321 | len = CHAR_STRING (c, p, str); | |
1322 | if (p != str) | |
1323 | /* C is a composite character. */ | |
1324 | bcopy (str, p, len); | |
1325 | p += len; | |
1326 | } | |
1327 | ||
020da460 KH |
1328 | /* Here, we can't use make_string_from_bytes because of byte |
1329 | combining problem. */ | |
1330 | val = make_string (buf, p - buf); | |
4ed46869 KH |
1331 | return val; |
1332 | } | |
1333 | ||
1334 | #endif /* emacs */ | |
1335 | \f | |
1336 | /*** Composite characters staffs ***/ | |
1337 | ||
1338 | /* Each composite character is identified by CMPCHAR-ID which is | |
1339 | assigned when Emacs needs the character code of the composite | |
1340 | character (e.g. when displaying it on the screen). See the | |
1341 | document "GENERAL NOTE on COMPOSITE CHARACTER" in `charset.h' how a | |
1342 | composite character is represented in Emacs. */ | |
1343 | ||
1344 | /* If `static' is defined, it means that it is defined to null string. */ | |
1345 | #ifndef static | |
1346 | /* The following function is copied from lread.c. */ | |
1347 | static int | |
1348 | hash_string (ptr, len) | |
1349 | unsigned char *ptr; | |
1350 | int len; | |
1351 | { | |
1352 | register unsigned char *p = ptr; | |
1353 | register unsigned char *end = p + len; | |
1354 | register unsigned char c; | |
1355 | register int hash = 0; | |
1356 | ||
1357 | while (p != end) | |
1358 | { | |
1359 | c = *p++; | |
1360 | if (c >= 0140) c -= 40; | |
1361 | hash = ((hash<<3) + (hash>>28) + c); | |
1362 | } | |
1363 | return hash & 07777777777; | |
1364 | } | |
1365 | #endif | |
1366 | ||
4ed46869 KH |
1367 | #define CMPCHAR_HASH_TABLE_SIZE 0xFFF |
1368 | ||
1369 | static int *cmpchar_hash_table[CMPCHAR_HASH_TABLE_SIZE]; | |
1370 | ||
1371 | /* Each element of `cmpchar_hash_table' is a pointer to an array of | |
1372 | integer, where the 1st element is the size of the array, the 2nd | |
1373 | element is how many elements are actually used in the array, and | |
1374 | the remaining elements are CMPCHAR-IDs of composite characters of | |
1375 | the same hash value. */ | |
1376 | #define CMPCHAR_HASH_SIZE(table) table[0] | |
1377 | #define CMPCHAR_HASH_USED(table) table[1] | |
1378 | #define CMPCHAR_HASH_CMPCHAR_ID(table, i) table[i] | |
1379 | ||
1380 | /* Return CMPCHAR-ID of the composite character in STR of the length | |
1381 | LEN. If the composite character has not yet been registered, | |
1382 | register it in `cmpchar_table' and assign new CMPCHAR-ID. This | |
1383 | is the sole function for assigning CMPCHAR-ID. */ | |
1384 | int | |
1385 | str_cmpchar_id (str, len) | |
8867de67 | 1386 | const unsigned char *str; |
4ed46869 KH |
1387 | int len; |
1388 | { | |
1389 | int hash_idx, *hashp; | |
1390 | unsigned char *buf; | |
1391 | int embedded_rule; /* 1 if composition rule is embedded. */ | |
1392 | int chars; /* number of components. */ | |
1393 | int i; | |
1394 | struct cmpchar_info *cmpcharp; | |
1395 | ||
4ed46869 KH |
1396 | /* The second byte 0xFF means compostion rule is embedded. */ |
1397 | embedded_rule = (str[1] == 0xFF); | |
1398 | ||
1399 | /* At first, get the actual length of the composite character. */ | |
1400 | { | |
8867de67 | 1401 | const unsigned char *p, *endp = str + 1, *lastp = str + len; |
4ed46869 KH |
1402 | int bytes; |
1403 | ||
6ae1f27e | 1404 | while (endp < lastp && ! CHAR_HEAD_P (*endp)) endp++; |
93bcb785 KH |
1405 | if (endp - str < 5) |
1406 | /* Any composite char have at least 5-byte length. */ | |
1407 | return -1; | |
1408 | ||
4ed46869 | 1409 | chars = 0; |
93bcb785 | 1410 | p = str + 1; |
4ed46869 KH |
1411 | while (p < endp) |
1412 | { | |
93bcb785 | 1413 | if (embedded_rule) p++; |
4ed46869 | 1414 | /* No need of checking if *P is 0xA0 because |
93bcb785 KH |
1415 | BYTES_BY_CHAR_HEAD (0x80) surely returns 2. */ |
1416 | p += BYTES_BY_CHAR_HEAD (*p - 0x20); | |
4ed46869 KH |
1417 | chars++; |
1418 | } | |
93bcb785 KH |
1419 | if (p > endp || chars < 2 || chars > MAX_COMPONENT_COUNT) |
1420 | /* Invalid components. */ | |
4ed46869 | 1421 | return -1; |
93bcb785 | 1422 | len = p - str; |
4ed46869 KH |
1423 | } |
1424 | hash_idx = hash_string (str, len) % CMPCHAR_HASH_TABLE_SIZE; | |
1425 | hashp = cmpchar_hash_table[hash_idx]; | |
1426 | ||
1427 | /* Then, look into the hash table. */ | |
1428 | if (hashp != NULL) | |
1429 | /* Find the correct one among composite characters of the same | |
1430 | hash value. */ | |
1431 | for (i = 2; i < CMPCHAR_HASH_USED (hashp); i++) | |
1432 | { | |
1433 | cmpcharp = cmpchar_table[CMPCHAR_HASH_CMPCHAR_ID (hashp, i)]; | |
1434 | if (len == cmpcharp->len | |
1435 | && ! bcmp (str, cmpcharp->data, len)) | |
1436 | return CMPCHAR_HASH_CMPCHAR_ID (hashp, i); | |
1437 | } | |
1438 | ||
1439 | /* We have to register the composite character in cmpchar_table. */ | |
513ee442 KH |
1440 | if (n_cmpchars > (CHAR_FIELD2_MASK | CHAR_FIELD3_MASK)) |
1441 | /* No, we have no more room for a new composite character. */ | |
1442 | return -1; | |
1443 | ||
4ed46869 KH |
1444 | /* Make the entry in hash table. */ |
1445 | if (hashp == NULL) | |
1446 | { | |
1447 | /* Make a table for 8 composite characters initially. */ | |
1448 | hashp = (cmpchar_hash_table[hash_idx] | |
1449 | = (int *) xmalloc (sizeof (int) * (2 + 8))); | |
1450 | CMPCHAR_HASH_SIZE (hashp) = 10; | |
1451 | CMPCHAR_HASH_USED (hashp) = 2; | |
1452 | } | |
1453 | else if (CMPCHAR_HASH_USED (hashp) >= CMPCHAR_HASH_SIZE (hashp)) | |
1454 | { | |
1455 | CMPCHAR_HASH_SIZE (hashp) += 8; | |
1456 | hashp = (cmpchar_hash_table[hash_idx] | |
1457 | = (int *) xrealloc (hashp, | |
1458 | sizeof (int) * CMPCHAR_HASH_SIZE (hashp))); | |
1459 | } | |
1460 | CMPCHAR_HASH_CMPCHAR_ID (hashp, CMPCHAR_HASH_USED (hashp)) = n_cmpchars; | |
1461 | CMPCHAR_HASH_USED (hashp)++; | |
1462 | ||
1463 | /* Set information of the composite character in cmpchar_table. */ | |
1464 | if (cmpchar_table_size == 0) | |
1465 | { | |
1466 | /* This is the first composite character to be registered. */ | |
1467 | cmpchar_table_size = 256; | |
1468 | cmpchar_table | |
1469 | = (struct cmpchar_info **) xmalloc (sizeof (cmpchar_table[0]) | |
1470 | * cmpchar_table_size); | |
1471 | } | |
1472 | else if (cmpchar_table_size <= n_cmpchars) | |
1473 | { | |
1474 | cmpchar_table_size += 256; | |
1475 | cmpchar_table | |
1476 | = (struct cmpchar_info **) xrealloc (cmpchar_table, | |
1477 | sizeof (cmpchar_table[0]) | |
1478 | * cmpchar_table_size); | |
1479 | } | |
1480 | ||
1481 | cmpcharp = (struct cmpchar_info *) xmalloc (sizeof (struct cmpchar_info)); | |
1482 | ||
1483 | cmpcharp->len = len; | |
1484 | cmpcharp->data = (unsigned char *) xmalloc (len + 1); | |
1485 | bcopy (str, cmpcharp->data, len); | |
1486 | cmpcharp->data[len] = 0; | |
1487 | cmpcharp->glyph_len = chars; | |
1488 | cmpcharp->glyph = (GLYPH *) xmalloc (sizeof (GLYPH) * chars); | |
1489 | if (embedded_rule) | |
1490 | { | |
1491 | cmpcharp->cmp_rule = (unsigned char *) xmalloc (chars); | |
1492 | cmpcharp->col_offset = (float *) xmalloc (sizeof (float) * chars); | |
1493 | } | |
1494 | else | |
1495 | { | |
1496 | cmpcharp->cmp_rule = NULL; | |
1497 | cmpcharp->col_offset = NULL; | |
1498 | } | |
1499 | ||
1500 | /* Setup GLYPH data and composition rules (if any) so as not to make | |
1501 | them every time on displaying. */ | |
1502 | { | |
1503 | unsigned char *bufp; | |
1504 | int width; | |
1505 | float leftmost = 0.0, rightmost = 1.0; | |
1506 | ||
1507 | if (embedded_rule) | |
1508 | /* At first, col_offset[N] is set to relative to col_offset[0]. */ | |
1509 | cmpcharp->col_offset[0] = 0; | |
1510 | ||
1511 | for (i = 0, bufp = cmpcharp->data + 1; i < chars; i++) | |
1512 | { | |
1513 | if (embedded_rule) | |
1514 | cmpcharp->cmp_rule[i] = *bufp++; | |
1515 | ||
1516 | if (*bufp == 0xA0) /* This is an ASCII character. */ | |
1517 | { | |
1518 | cmpcharp->glyph[i] = FAST_MAKE_GLYPH ((*++bufp & 0x7F), 0); | |
1519 | width = 1; | |
1520 | bufp++; | |
1521 | } | |
1522 | else /* Multibyte character. */ | |
1523 | { | |
1524 | /* Make `bufp' point normal multi-byte form temporally. */ | |
1525 | *bufp -= 0x20; | |
1526 | cmpcharp->glyph[i] | |
537efd8d | 1527 | = FAST_MAKE_GLYPH (string_to_non_ascii_char (bufp, 4, 0, 0), 0); |
4ed46869 KH |
1528 | width = WIDTH_BY_CHAR_HEAD (*bufp); |
1529 | *bufp += 0x20; | |
1530 | bufp += BYTES_BY_CHAR_HEAD (*bufp - 0x20); | |
1531 | } | |
1532 | ||
1533 | if (embedded_rule && i > 0) | |
1534 | { | |
1535 | /* Reference points (global_ref and new_ref) are | |
1536 | encoded as below: | |
1537 | ||
1538 | 0--1--2 -- ascent | |
1539 | | | | |
1540 | | | | |
1541 | | 4 -+--- center | |
1542 | -- 3 5 -- baseline | |
1543 | | | | |
1544 | 6--7--8 -- descent | |
1545 | ||
1546 | Now, we calculate the column offset of the new glyph | |
1547 | from the left edge of the first glyph. This can avoid | |
1548 | the same calculation everytime displaying this | |
1549 | composite character. */ | |
1550 | ||
1551 | /* Reference points of global glyph and new glyph. */ | |
1552 | int global_ref = (cmpcharp->cmp_rule[i] - 0xA0) / 9; | |
1553 | int new_ref = (cmpcharp->cmp_rule[i] - 0xA0) % 9; | |
1554 | /* Column offset relative to the first glyph. */ | |
1555 | float left = (leftmost | |
1556 | + (global_ref % 3) * (rightmost - leftmost) / 2.0 | |
1557 | - (new_ref % 3) * width / 2.0); | |
1558 | ||
1559 | cmpcharp->col_offset[i] = left; | |
1560 | if (left < leftmost) | |
1561 | leftmost = left; | |
1562 | if (left + width > rightmost) | |
1563 | rightmost = left + width; | |
1564 | } | |
1565 | else | |
1566 | { | |
1567 | if (width > rightmost) | |
1568 | rightmost = width; | |
1569 | } | |
1570 | } | |
1571 | if (embedded_rule) | |
1572 | { | |
1573 | /* Now col_offset[N] are relative to the left edge of the | |
1574 | first component. Make them relative to the left edge of | |
1575 | overall glyph. */ | |
1576 | for (i = 0; i < chars; i++) | |
1577 | cmpcharp->col_offset[i] -= leftmost; | |
1578 | /* Make rightmost holds width of overall glyph. */ | |
1579 | rightmost -= leftmost; | |
1580 | } | |
1581 | ||
1582 | cmpcharp->width = rightmost; | |
1583 | if (cmpcharp->width < rightmost) | |
1584 | /* To get a ceiling integer value. */ | |
1585 | cmpcharp->width++; | |
1586 | } | |
1587 | ||
1588 | cmpchar_table[n_cmpchars] = cmpcharp; | |
1589 | ||
1590 | return n_cmpchars++; | |
1591 | } | |
1592 | ||
1593 | /* Return the Nth element of the composite character C. */ | |
1594 | int | |
1595 | cmpchar_component (c, n) | |
1596 | unsigned int c, n; | |
1597 | { | |
1598 | int id = COMPOSITE_CHAR_ID (c); | |
1599 | ||
1600 | if (id >= n_cmpchars /* C is not a valid composite character. */ | |
1601 | || n >= cmpchar_table[id]->glyph_len) /* No such component. */ | |
1602 | return -1; | |
1603 | /* No face data is stored in glyph code. */ | |
1604 | return ((int) (cmpchar_table[id]->glyph[n])); | |
1605 | } | |
1606 | ||
1607 | DEFUN ("cmpcharp", Fcmpcharp, Scmpcharp, 1, 1, 0, | |
1608 | "T if CHAR is a composite character.") | |
1609 | (ch) | |
1610 | Lisp_Object ch; | |
1611 | { | |
1612 | CHECK_NUMBER (ch, 0); | |
1613 | return (COMPOSITE_CHAR_P (XINT (ch)) ? Qt : Qnil); | |
1614 | } | |
1615 | ||
1616 | DEFUN ("composite-char-component", Fcmpchar_component, Scmpchar_component, | |
1617 | 2, 2, 0, | |
1618 | "Return the IDXth component character of composite character CHARACTER.") | |
1619 | (character, idx) | |
1620 | Lisp_Object character, idx; | |
1621 | { | |
1622 | int c; | |
1623 | ||
1624 | CHECK_NUMBER (character, 0); | |
1625 | CHECK_NUMBER (idx, 1); | |
1626 | ||
1627 | if ((c = cmpchar_component (XINT (character), XINT (idx))) < 0) | |
1628 | args_out_of_range (character, idx); | |
1629 | ||
1630 | return make_number (c); | |
1631 | } | |
1632 | ||
1633 | DEFUN ("composite-char-composition-rule", Fcmpchar_cmp_rule, Scmpchar_cmp_rule, | |
1634 | 2, 2, 0, | |
55001746 KH |
1635 | "Return the Nth composition rule embedded in composite character CHARACTER.\n\ |
1636 | The returned rule is for composing the Nth component\n\ | |
1637 | on the (N-1)th component. If N is 0, the returned value is always 255.") | |
1638 | (character, n) | |
1639 | Lisp_Object character, n; | |
4ed46869 KH |
1640 | { |
1641 | int id, i; | |
1642 | ||
1643 | CHECK_NUMBER (character, 0); | |
55001746 | 1644 | CHECK_NUMBER (n, 1); |
4ed46869 KH |
1645 | |
1646 | id = COMPOSITE_CHAR_ID (XINT (character)); | |
1647 | if (id < 0 || id >= n_cmpchars) | |
1648 | error ("Invalid composite character: %d", XINT (character)); | |
55001746 | 1649 | i = XINT (n); |
4ed46869 | 1650 | if (i > cmpchar_table[id]->glyph_len) |
55001746 | 1651 | args_out_of_range (character, n); |
4ed46869 KH |
1652 | |
1653 | return make_number (cmpchar_table[id]->cmp_rule[i]); | |
1654 | } | |
1655 | ||
1656 | DEFUN ("composite-char-composition-rule-p", Fcmpchar_cmp_rule_p, | |
1657 | Scmpchar_cmp_rule_p, 1, 1, 0, | |
1658 | "Return non-nil if composite character CHARACTER contains a embedded rule.") | |
1659 | (character) | |
1660 | Lisp_Object character; | |
1661 | { | |
1662 | int id; | |
1663 | ||
1664 | CHECK_NUMBER (character, 0); | |
1665 | id = COMPOSITE_CHAR_ID (XINT (character)); | |
1666 | if (id < 0 || id >= n_cmpchars) | |
1667 | error ("Invalid composite character: %d", XINT (character)); | |
1668 | ||
1669 | return (cmpchar_table[id]->cmp_rule ? Qt : Qnil); | |
1670 | } | |
1671 | ||
1672 | DEFUN ("composite-char-component-count", Fcmpchar_cmp_count, | |
1673 | Scmpchar_cmp_count, 1, 1, 0, | |
1674 | "Return number of compoents of composite character CHARACTER.") | |
1675 | (character) | |
1676 | Lisp_Object character; | |
1677 | { | |
1678 | int id; | |
1679 | ||
1680 | CHECK_NUMBER (character, 0); | |
1681 | id = COMPOSITE_CHAR_ID (XINT (character)); | |
1682 | if (id < 0 || id >= n_cmpchars) | |
1683 | error ("Invalid composite character: %d", XINT (character)); | |
1684 | ||
1685 | return (make_number (cmpchar_table[id]->glyph_len)); | |
1686 | } | |
1687 | ||
1688 | DEFUN ("compose-string", Fcompose_string, Scompose_string, | |
1689 | 1, 1, 0, | |
1690 | "Return one char string composed from all characters in STRING.") | |
1691 | (str) | |
1692 | Lisp_Object str; | |
1693 | { | |
1694 | unsigned char buf[MAX_LENGTH_OF_MULTI_BYTE_FORM], *p, *pend, *ptemp; | |
1695 | int len, i; | |
1696 | ||
1697 | CHECK_STRING (str, 0); | |
1698 | ||
1699 | buf[0] = LEADING_CODE_COMPOSITION; | |
1700 | p = XSTRING (str)->data; | |
fc932ac6 | 1701 | pend = p + STRING_BYTES (XSTRING (str)); |
4ed46869 KH |
1702 | i = 1; |
1703 | while (p < pend) | |
1704 | { | |
1705 | if (*p < 0x20 || *p == 127) /* control code */ | |
1706 | error ("Invalid component character: %d", *p); | |
1707 | else if (*p < 0x80) /* ASCII */ | |
1708 | { | |
1709 | if (i + 2 >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1710 | error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1711 | /* Prepend an ASCII charset indicator 0xA0, set MSB of the | |
1712 | code itself. */ | |
1713 | buf[i++] = 0xA0; | |
1714 | buf[i++] = *p++ + 0x80; | |
1715 | } | |
1716 | else if (*p == LEADING_CODE_COMPOSITION) /* composite char */ | |
1717 | { | |
1718 | /* Already composed. Eliminate the heading | |
1719 | LEADING_CODE_COMPOSITION, keep the remaining bytes | |
1720 | unchanged. */ | |
1721 | p++; | |
1722 | ptemp = p; | |
6ae1f27e | 1723 | while (! CHAR_HEAD_P (*p)) p++; |
4ed46869 KH |
1724 | if (i + (p - ptemp) >= MAX_LENGTH_OF_MULTI_BYTE_FORM) |
1725 | error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1726 | bcopy (ptemp, buf + i, p - ptemp); | |
1727 | i += p - ptemp; | |
1728 | } | |
1729 | else /* multibyte char */ | |
1730 | { | |
1731 | /* Add 0x20 to the base leading-code, keep the remaining | |
1732 | bytes unchanged. */ | |
1733 | len = BYTES_BY_CHAR_HEAD (*p); | |
1734 | if (i + len >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1735 | error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1736 | bcopy (p, buf + i, len); | |
1737 | buf[i] += 0x20; | |
1738 | p += len, i += len; | |
1739 | } | |
1740 | } | |
1741 | ||
1742 | if (i < 5) | |
1743 | /* STR contains only one character, which can't be composed. */ | |
1744 | error ("Too short string to be composed: %s", XSTRING (str)->data); | |
1745 | ||
27802600 | 1746 | return make_string_from_bytes (buf, 1, i); |
4ed46869 KH |
1747 | } |
1748 | ||
1749 | \f | |
dfcf069d | 1750 | int |
4ed46869 KH |
1751 | charset_id_internal (charset_name) |
1752 | char *charset_name; | |
1753 | { | |
76d7b829 | 1754 | Lisp_Object val; |
4ed46869 | 1755 | |
76d7b829 | 1756 | val= Fget (intern (charset_name), Qcharset); |
4ed46869 KH |
1757 | if (!VECTORP (val)) |
1758 | error ("Charset %s is not defined", charset_name); | |
1759 | ||
1760 | return (XINT (XVECTOR (val)->contents[0])); | |
1761 | } | |
1762 | ||
1763 | DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
1764 | Ssetup_special_charsets, 0, 0, 0, "Internal use only.") | |
1765 | () | |
1766 | { | |
1767 | charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1768 | charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1769 | charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1770 | charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1771 | charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1772 | charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1773 | charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1774 | return Qnil; | |
1775 | } | |
1776 | ||
dfcf069d | 1777 | void |
4ed46869 KH |
1778 | init_charset_once () |
1779 | { | |
1780 | int i, j, k; | |
1781 | ||
1782 | staticpro (&Vcharset_table); | |
1783 | staticpro (&Vcharset_symbol_table); | |
8a73a704 | 1784 | staticpro (&Vgeneric_character_list); |
4ed46869 KH |
1785 | |
1786 | /* This has to be done here, before we call Fmake_char_table. */ | |
1787 | Qcharset_table = intern ("charset-table"); | |
1788 | staticpro (&Qcharset_table); | |
1789 | ||
1790 | /* Intern this now in case it isn't already done. | |
1791 | Setting this variable twice is harmless. | |
1792 | But don't staticpro it here--that is done in alloc.c. */ | |
1793 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1794 | ||
1795 | /* Now we are ready to set up this property, so we can | |
1796 | create the charset table. */ | |
1797 | Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1798 | Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1799 | ||
513ee442 | 1800 | Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), Qnil); |
4ed46869 KH |
1801 | |
1802 | /* Setup tables. */ | |
1803 | for (i = 0; i < 2; i++) | |
1804 | for (j = 0; j < 2; j++) | |
1805 | for (k = 0; k < 128; k++) | |
1806 | iso_charset_table [i][j][k] = -1; | |
1807 | ||
1808 | bzero (cmpchar_hash_table, sizeof cmpchar_hash_table); | |
1809 | cmpchar_table_size = n_cmpchars = 0; | |
1810 | ||
1811 | for (i = 0; i < 256; i++) | |
1812 | BYTES_BY_CHAR_HEAD (i) = 1; | |
1813 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_11) = 3; | |
1814 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_12) = 3; | |
1815 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_21) = 4; | |
1816 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_22) = 4; | |
1817 | /* The following doesn't reflect the actual bytes, but just to tell | |
1818 | that it is a start of a multibyte character. */ | |
1819 | BYTES_BY_CHAR_HEAD (LEADING_CODE_COMPOSITION) = 2; | |
1820 | ||
1821 | for (i = 0; i < 128; i++) | |
1822 | WIDTH_BY_CHAR_HEAD (i) = 1; | |
1823 | for (; i < 256; i++) | |
1824 | WIDTH_BY_CHAR_HEAD (i) = 4; | |
1825 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_11) = 1; | |
1826 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_12) = 2; | |
1827 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_21) = 1; | |
1828 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_22) = 2; | |
8a73a704 KH |
1829 | |
1830 | { | |
76d7b829 | 1831 | Lisp_Object val; |
8a73a704 | 1832 | |
76d7b829 | 1833 | val = Qnil; |
8a73a704 KH |
1834 | for (i = 0x81; i < 0x90; i++) |
1835 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1836 | for (; i < 0x9A; i++) | |
1837 | val = Fcons (make_number ((i - 0x8F) << 14), val); | |
1838 | for (i = 0xA0; i < 0xF0; i++) | |
1839 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1840 | for (; i < 0xFF; i++) | |
1841 | val = Fcons (make_number ((i - 0xE0) << 14), val); | |
1842 | val = Fcons (make_number (GENERIC_COMPOSITION_CHAR), val); | |
1843 | Vgeneric_character_list = Fnreverse (val); | |
1844 | } | |
bbf12bb3 KH |
1845 | |
1846 | nonascii_insert_offset = 0; | |
1847 | Vnonascii_translation_table = Qnil; | |
4ed46869 KH |
1848 | } |
1849 | ||
1850 | #ifdef emacs | |
1851 | ||
dfcf069d | 1852 | void |
4ed46869 KH |
1853 | syms_of_charset () |
1854 | { | |
1855 | Qascii = intern ("ascii"); | |
1856 | staticpro (&Qascii); | |
1857 | ||
1858 | Qcharset = intern ("charset"); | |
1859 | staticpro (&Qcharset); | |
1860 | ||
1861 | /* Define ASCII charset now. */ | |
1862 | update_charset_table (make_number (CHARSET_ASCII), | |
1863 | make_number (1), make_number (94), | |
1864 | make_number (1), | |
1865 | make_number (0), | |
1866 | make_number ('B'), | |
1867 | make_number (0), | |
1868 | build_string ("ASCII"), | |
1869 | build_string ("ASCII"), | |
1870 | build_string ("ASCII (ISO646 IRV)")); | |
1871 | CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1872 | Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1873 | ||
1874 | Qcomposition = intern ("composition"); | |
1875 | staticpro (&Qcomposition); | |
1876 | CHARSET_SYMBOL (CHARSET_COMPOSITION) = Qcomposition; | |
1877 | ||
c1a08b4c KH |
1878 | Qauto_fill_chars = intern ("auto-fill-chars"); |
1879 | staticpro (&Qauto_fill_chars); | |
1880 | Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0)); | |
1881 | ||
4ed46869 | 1882 | defsubr (&Sdefine_charset); |
8a73a704 | 1883 | defsubr (&Sgeneric_character_list); |
3fac5a51 | 1884 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
1885 | defsubr (&Sdeclare_equiv_charset); |
1886 | defsubr (&Sfind_charset_region); | |
1887 | defsubr (&Sfind_charset_string); | |
1888 | defsubr (&Smake_char_internal); | |
1889 | defsubr (&Ssplit_char); | |
1890 | defsubr (&Schar_charset); | |
90d7b74e | 1891 | defsubr (&Scharset_after); |
4ed46869 | 1892 | defsubr (&Siso_charset); |
9d3d8cba | 1893 | defsubr (&Schar_valid_p); |
d2665018 | 1894 | defsubr (&Sunibyte_char_to_multibyte); |
1bcc1567 | 1895 | defsubr (&Smultibyte_char_to_unibyte); |
4ed46869 KH |
1896 | defsubr (&Schar_bytes); |
1897 | defsubr (&Schar_width); | |
1898 | defsubr (&Sstring_width); | |
1899 | defsubr (&Schar_direction); | |
af4fecb4 | 1900 | defsubr (&Schars_in_region); |
87b089ad | 1901 | defsubr (&Sstring); |
4ed46869 KH |
1902 | defsubr (&Scmpcharp); |
1903 | defsubr (&Scmpchar_component); | |
1904 | defsubr (&Scmpchar_cmp_rule); | |
1905 | defsubr (&Scmpchar_cmp_rule_p); | |
1906 | defsubr (&Scmpchar_cmp_count); | |
1907 | defsubr (&Scompose_string); | |
1908 | defsubr (&Ssetup_special_charsets); | |
1909 | ||
1910 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
1911 | "List of charsets ever defined."); | |
1912 | Vcharset_list = Fcons (Qascii, Qnil); | |
1913 | ||
537efd8d | 1914 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, |
b4e9dd77 KH |
1915 | "Vector of cons cell of a symbol and translation table ever defined.\n\ |
1916 | An ID of a translation table is an index of this vector."); | |
537efd8d | 1917 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
b0e3cf2b | 1918 | |
4ed46869 KH |
1919 | DEFVAR_INT ("leading-code-composition", &leading_code_composition, |
1920 | "Leading-code of composite characters."); | |
1921 | leading_code_composition = LEADING_CODE_COMPOSITION; | |
1922 | ||
1923 | DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, | |
1924 | "Leading-code of private TYPE9N charset of column-width 1."); | |
1925 | leading_code_private_11 = LEADING_CODE_PRIVATE_11; | |
1926 | ||
1927 | DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
1928 | "Leading-code of private TYPE9N charset of column-width 2."); | |
1929 | leading_code_private_12 = LEADING_CODE_PRIVATE_12; | |
1930 | ||
1931 | DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
1932 | "Leading-code of private TYPE9Nx9N charset of column-width 1."); | |
1933 | leading_code_private_21 = LEADING_CODE_PRIVATE_21; | |
1934 | ||
1935 | DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
1936 | "Leading-code of private TYPE9Nx9N charset of column-width 2."); | |
1937 | leading_code_private_22 = LEADING_CODE_PRIVATE_22; | |
35e623fb RS |
1938 | |
1939 | DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset, | |
d2665018 | 1940 | "Offset for converting non-ASCII unibyte codes 0240...0377 to multibyte.\n\ |
4cf9710d RS |
1941 | This is used for converting unibyte text to multibyte,\n\ |
1942 | and for inserting character codes specified by number.\n\n\ | |
3e8ceaac RS |
1943 | This serves to convert a Latin-1 or similar 8-bit character code\n\ |
1944 | to the corresponding Emacs multibyte character code.\n\ | |
1945 | Typically the value should be (- (make-char CHARSET 0) 128),\n\ | |
1946 | for your choice of character set.\n\ | |
537efd8d | 1947 | If `nonascii-translation-table' is non-nil, it overrides this variable."); |
35e623fb | 1948 | nonascii_insert_offset = 0; |
b0e3cf2b | 1949 | |
b4e9dd77 | 1950 | DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table, |
537efd8d | 1951 | "Translation table to convert non-ASCII unibyte codes to multibyte.\n\ |
4cf9710d RS |
1952 | This is used for converting unibyte text to multibyte,\n\ |
1953 | and for inserting character codes specified by number.\n\n\ | |
1954 | Conversion is performed only when multibyte characters are enabled,\n\ | |
1955 | and it serves to convert a Latin-1 or similar 8-bit character code\n\ | |
1956 | to the corresponding Emacs character code.\n\n\ | |
da4d65af | 1957 | If this is nil, `nonascii-insert-offset' is used instead.\n\ |
b4e9dd77 KH |
1958 | See also the docstring of `make-translation-table'."); |
1959 | Vnonascii_translation_table = Qnil; | |
4cf9710d | 1960 | |
b0e3cf2b KH |
1961 | DEFVAR_INT ("min-composite-char", &min_composite_char, |
1962 | "Minimum character code of a composite character."); | |
1963 | min_composite_char = MIN_CHAR_COMPOSITION; | |
c1a08b4c KH |
1964 | |
1965 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
1966 | "A char-table for characters which invoke auto-filling.\n\ | |
1967 | Such characters has value t in this table."); | |
1968 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
1969 | CHAR_TABLE_SET (Vauto_fill_chars, ' ', Qt); | |
1970 | CHAR_TABLE_SET (Vauto_fill_chars, '\n', Qt); | |
4ed46869 KH |
1971 | } |
1972 | ||
1973 | #endif /* emacs */ |