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; | |
bb63e573 KH |
194 | int c, bytes; |
195 | const unsigned char *begp = str; | |
4ed46869 | 196 | |
bb63e573 | 197 | c = *str++; |
90d7b74e KH |
198 | bytes = 1; |
199 | ||
200 | if (BASE_LEADING_CODE_P (c)) | |
bb63e573 KH |
201 | do { |
202 | while (bytes < len && ! CHAR_HEAD_P (begp[bytes])) bytes++; | |
90d7b74e KH |
203 | |
204 | if (c == LEADING_CODE_COMPOSITION) | |
205 | { | |
bb63e573 | 206 | int cmpchar_id = str_cmpchar_id (begp, bytes); |
90d7b74e KH |
207 | |
208 | if (cmpchar_id >= 0) | |
5d76bc89 KH |
209 | { |
210 | c = MAKE_COMPOSITE_CHAR (cmpchar_id); | |
bb63e573 | 211 | str += cmpchar_table[cmpchar_id]->len; |
5d76bc89 | 212 | } |
90d7b74e KH |
213 | } |
214 | else | |
215 | { | |
bb63e573 | 216 | const unsigned char *endp = begp + bytes; |
90d7b74e | 217 | int charset = c, c1, c2 = 0; |
4ed46869 | 218 | |
bb63e573 | 219 | if (str >= endp) break; |
a6c25326 | 220 | if (c >= LEADING_CODE_PRIVATE_11 && c <= LEADING_CODE_PRIVATE_22) |
90d7b74e | 221 | { |
bb63e573 KH |
222 | charset = *str++; |
223 | if (str < endp) | |
224 | c1 = *str++ & 0x7F; | |
225 | else | |
226 | c1 = charset, charset = c; | |
90d7b74e | 227 | } |
bb63e573 KH |
228 | else |
229 | c1 = *str++ & 0x7f; | |
230 | if (CHARSET_DEFINED_P (charset) | |
231 | && CHARSET_DIMENSION (charset) == 2 | |
232 | && str < endp) | |
233 | c2 = *str & 0x7F; | |
234 | c = MAKE_NON_ASCII_CHAR (charset, c1, c2); | |
90d7b74e | 235 | } |
bb63e573 | 236 | } while (0); |
4ed46869 KH |
237 | |
238 | if (actual_len) | |
bb63e573 | 239 | *actual_len = exclude_tail_garbage ? str - begp : bytes; |
4ed46869 KH |
240 | return c; |
241 | } | |
242 | ||
243 | /* Return the length of the multi-byte form at string STR of length LEN. */ | |
244 | int | |
245 | multibyte_form_length (str, len) | |
8867de67 | 246 | const unsigned char *str; |
4ed46869 KH |
247 | int len; |
248 | { | |
90d7b74e | 249 | int bytes = 1; |
4ed46869 | 250 | |
90d7b74e KH |
251 | if (BASE_LEADING_CODE_P (*str)) |
252 | while (bytes < len && ! CHAR_HEAD_P (str[bytes])) bytes++; | |
4ed46869 | 253 | |
90d7b74e | 254 | return bytes; |
4ed46869 KH |
255 | } |
256 | ||
257 | /* Check if string STR of length LEN contains valid multi-byte form of | |
258 | a character. If valid, charset and position codes of the character | |
259 | is set at *CHARSET, *C1, and *C2, and return 0. If not valid, | |
260 | return -1. This should be used only in the macro SPLIT_STRING | |
261 | which checks range of STR in advance. */ | |
262 | ||
dfcf069d | 263 | int |
4ed46869 | 264 | split_non_ascii_string (str, len, charset, c1, c2) |
8867de67 KH |
265 | register const unsigned char *str; |
266 | register unsigned char *c1, *c2; | |
4ed46869 KH |
267 | register int len, *charset; |
268 | { | |
269 | register unsigned int cs = *str++; | |
270 | ||
271 | if (cs == LEADING_CODE_COMPOSITION) | |
272 | { | |
273 | int cmpchar_id = str_cmpchar_id (str - 1, len); | |
274 | ||
275 | if (cmpchar_id < 0) | |
276 | return -1; | |
277 | *charset = cs, *c1 = cmpchar_id >> 7, *c2 = cmpchar_id & 0x7F; | |
278 | } | |
279 | else if ((cs < LEADING_CODE_PRIVATE_11 || (cs = *str++) >= 0xA0) | |
280 | && CHARSET_DEFINED_P (cs)) | |
281 | { | |
282 | *charset = cs; | |
283 | if (*str < 0xA0) | |
284 | return -1; | |
285 | *c1 = (*str++) & 0x7F; | |
286 | if (CHARSET_DIMENSION (cs) == 2) | |
287 | { | |
288 | if (*str < 0xA0) | |
289 | return -1; | |
290 | *c2 = (*str++) & 0x7F; | |
291 | } | |
292 | } | |
293 | else | |
294 | return -1; | |
295 | return 0; | |
296 | } | |
297 | ||
537efd8d | 298 | /* Translate character C by translation table TABLE. If C |
b4e9dd77 KH |
299 | is negative, translate a character specified by CHARSET, C1, and C2 |
300 | (C1 and C2 are code points of the character). If no translation is | |
301 | found in TABLE, return C. */ | |
dfcf069d | 302 | int |
b4e9dd77 | 303 | translate_char (table, c, charset, c1, c2) |
23d2a7f1 KH |
304 | Lisp_Object table; |
305 | int c, charset, c1, c2; | |
306 | { | |
307 | Lisp_Object ch; | |
308 | int alt_charset, alt_c1, alt_c2, dimension; | |
309 | ||
310 | if (c < 0) c = MAKE_CHAR (charset, c1, c2); | |
311 | if (!CHAR_TABLE_P (table) | |
312 | || (ch = Faref (table, make_number (c)), !INTEGERP (ch)) | |
313 | || XINT (ch) < 0) | |
314 | return c; | |
315 | ||
316 | SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); | |
317 | dimension = CHARSET_DIMENSION (alt_charset); | |
318 | if (dimension == 1 && alt_c1 > 0 || dimension == 2 && alt_c2 > 0) | |
319 | /* CH is not a generic character, just return it. */ | |
320 | return XFASTINT (ch); | |
321 | ||
322 | /* Since CH is a generic character, we must return a specific | |
323 | charater which has the same position codes as C from CH. */ | |
324 | if (charset < 0) | |
325 | SPLIT_CHAR (c, charset, c1, c2); | |
326 | if (dimension != CHARSET_DIMENSION (charset)) | |
327 | /* We can't make such a character because of dimension mismatch. */ | |
328 | return c; | |
23d2a7f1 KH |
329 | return MAKE_CHAR (alt_charset, c1, c2); |
330 | } | |
331 | ||
d2665018 | 332 | /* Convert the unibyte character C to multibyte based on |
b4e9dd77 | 333 | Vnonascii_translation_table or nonascii_insert_offset. If they can't |
d2665018 KH |
334 | convert C to a valid multibyte character, convert it based on |
335 | DEFAULT_NONASCII_INSERT_OFFSET which makes C a Latin-1 character. */ | |
35e623fb | 336 | |
dfcf069d | 337 | int |
35e623fb RS |
338 | unibyte_char_to_multibyte (c) |
339 | int c; | |
340 | { | |
bbf12bb3 | 341 | if (c < 0400) |
35e623fb | 342 | { |
d2665018 KH |
343 | int c_save = c; |
344 | ||
b4e9dd77 | 345 | if (! NILP (Vnonascii_translation_table)) |
bbf12bb3 KH |
346 | { |
347 | c = XINT (Faref (Vnonascii_translation_table, make_number (c))); | |
348 | if (c >= 0400 && ! VALID_MULTIBYTE_CHAR_P (c)) | |
349 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; | |
350 | } | |
351 | else if (c >= 0240 && nonascii_insert_offset > 0) | |
352 | { | |
353 | c += nonascii_insert_offset; | |
354 | if (c < 0400 || ! VALID_MULTIBYTE_CHAR_P (c)) | |
355 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; | |
356 | } | |
357 | else if (c >= 0240) | |
d2665018 | 358 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
35e623fb RS |
359 | } |
360 | return c; | |
361 | } | |
76d7b829 KH |
362 | |
363 | ||
364 | /* Convert the multibyte character C to unibyte 8-bit character based | |
365 | on Vnonascii_translation_table or nonascii_insert_offset. If | |
366 | REV_TBL is non-nil, it should be a reverse table of | |
367 | Vnonascii_translation_table, i.e. what given by: | |
368 | Fchar_table_extra_slot (Vnonascii_translation_table, make_number (0)) */ | |
369 | ||
370 | int | |
371 | multibyte_char_to_unibyte (c, rev_tbl) | |
372 | int c; | |
373 | Lisp_Object rev_tbl; | |
374 | { | |
375 | if (!SINGLE_BYTE_CHAR_P (c)) | |
376 | { | |
377 | int c_save = c; | |
378 | ||
379 | if (! CHAR_TABLE_P (rev_tbl) | |
380 | && CHAR_TABLE_P (Vnonascii_translation_table)) | |
381 | rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table, | |
382 | make_number (0)); | |
383 | if (CHAR_TABLE_P (rev_tbl)) | |
384 | { | |
385 | Lisp_Object temp; | |
386 | temp = Faref (rev_tbl, make_number (c)); | |
387 | if (INTEGERP (temp)) | |
388 | c = XINT (temp); | |
bbf12bb3 KH |
389 | if (c >= 256) |
390 | c = (c_save & 0177) + 0200; | |
391 | } | |
392 | else | |
393 | { | |
394 | if (nonascii_insert_offset > 0) | |
395 | c -= nonascii_insert_offset; | |
396 | if (c < 128 || c >= 256) | |
397 | c = (c_save & 0177) + 0200; | |
76d7b829 | 398 | } |
76d7b829 KH |
399 | } |
400 | ||
401 | return c; | |
402 | } | |
403 | ||
35e623fb | 404 | \f |
4ed46869 KH |
405 | /* Update the table Vcharset_table with the given arguments (see the |
406 | document of `define-charset' for the meaning of each argument). | |
407 | Several other table contents are also updated. The caller should | |
408 | check the validity of CHARSET-ID and the remaining arguments in | |
409 | advance. */ | |
410 | ||
411 | void | |
412 | update_charset_table (charset_id, dimension, chars, width, direction, | |
413 | iso_final_char, iso_graphic_plane, | |
414 | short_name, long_name, description) | |
415 | Lisp_Object charset_id, dimension, chars, width, direction; | |
416 | Lisp_Object iso_final_char, iso_graphic_plane; | |
417 | Lisp_Object short_name, long_name, description; | |
418 | { | |
419 | int charset = XINT (charset_id); | |
420 | int bytes; | |
421 | unsigned char leading_code_base, leading_code_ext; | |
422 | ||
6dc0722d KH |
423 | if (NILP (CHARSET_TABLE_ENTRY (charset))) |
424 | CHARSET_TABLE_ENTRY (charset) | |
425 | = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); | |
4ed46869 KH |
426 | |
427 | /* Get byte length of multibyte form, base leading-code, and | |
428 | extended leading-code of the charset. See the comment under the | |
429 | title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
430 | bytes = XINT (dimension); | |
431 | if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
432 | { | |
433 | /* Official charset, it doesn't have an extended leading-code. */ | |
434 | if (charset != CHARSET_ASCII) | |
435 | bytes += 1; /* For a base leading-code. */ | |
436 | leading_code_base = charset; | |
437 | leading_code_ext = 0; | |
438 | } | |
439 | else | |
440 | { | |
441 | /* Private charset. */ | |
442 | bytes += 2; /* For base and extended leading-codes. */ | |
443 | leading_code_base | |
444 | = (charset < LEADING_CODE_EXT_12 | |
445 | ? LEADING_CODE_PRIVATE_11 | |
446 | : (charset < LEADING_CODE_EXT_21 | |
447 | ? LEADING_CODE_PRIVATE_12 | |
448 | : (charset < LEADING_CODE_EXT_22 | |
449 | ? LEADING_CODE_PRIVATE_21 | |
450 | : LEADING_CODE_PRIVATE_22))); | |
451 | leading_code_ext = charset; | |
452 | } | |
453 | ||
6ef23ebb KH |
454 | if (BYTES_BY_CHAR_HEAD (leading_code_base) != bytes) |
455 | error ("Invalid dimension for the charset-ID %d", charset); | |
456 | ||
4ed46869 KH |
457 | CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; |
458 | CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
459 | CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
460 | CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
461 | CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
462 | CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
463 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
464 | = make_number (leading_code_base); | |
465 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
466 | = make_number (leading_code_ext); | |
467 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
468 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
469 | = iso_graphic_plane; | |
470 | CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
471 | CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
472 | CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
473 | CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
474 | ||
475 | { | |
476 | /* If we have already defined a charset which has the same | |
477 | DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
478 | DIRECTION, we must update the entry REVERSE-CHARSET of both | |
479 | charsets. If there's no such charset, the value of the entry | |
480 | is set to nil. */ | |
481 | int i; | |
482 | ||
513ee442 | 483 | for (i = 0; i <= MAX_CHARSET; i++) |
4ed46869 KH |
484 | if (!NILP (CHARSET_TABLE_ENTRY (i))) |
485 | { | |
486 | if (CHARSET_DIMENSION (i) == XINT (dimension) | |
487 | && CHARSET_CHARS (i) == XINT (chars) | |
488 | && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
489 | && CHARSET_DIRECTION (i) != XINT (direction)) | |
490 | { | |
491 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
492 | = make_number (i); | |
493 | CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
494 | break; | |
495 | } | |
496 | } | |
513ee442 | 497 | if (i > MAX_CHARSET) |
4ed46869 KH |
498 | /* No such a charset. */ |
499 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
500 | = make_number (-1); | |
501 | } | |
502 | ||
503 | if (charset != CHARSET_ASCII | |
504 | && charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
505 | { | |
4ed46869 KH |
506 | width_by_char_head[leading_code_base] = XINT (width); |
507 | ||
508 | /* Update table emacs_code_class. */ | |
509 | emacs_code_class[charset] = (bytes == 2 | |
510 | ? EMACS_leading_code_2 | |
511 | : (bytes == 3 | |
512 | ? EMACS_leading_code_3 | |
513 | : EMACS_leading_code_4)); | |
514 | } | |
515 | ||
516 | /* Update table iso_charset_table. */ | |
517 | if (ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0) | |
518 | ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset; | |
519 | } | |
520 | ||
521 | #ifdef emacs | |
522 | ||
523 | /* Return charset id of CHARSET_SYMBOL, or return -1 if CHARSET_SYMBOL | |
524 | is invalid. */ | |
525 | int | |
526 | get_charset_id (charset_symbol) | |
527 | Lisp_Object charset_symbol; | |
528 | { | |
529 | Lisp_Object val; | |
530 | int charset; | |
531 | ||
532 | return ((SYMBOLP (charset_symbol) | |
533 | && (val = Fget (charset_symbol, Qcharset), VECTORP (val)) | |
534 | && (charset = XINT (XVECTOR (val)->contents[CHARSET_ID_IDX]), | |
535 | CHARSET_VALID_P (charset))) | |
536 | ? charset : -1); | |
537 | } | |
538 | ||
539 | /* Return an identification number for a new private charset of | |
540 | DIMENSION and WIDTH. If there's no more room for the new charset, | |
541 | return 0. */ | |
542 | Lisp_Object | |
543 | get_new_private_charset_id (dimension, width) | |
544 | int dimension, width; | |
545 | { | |
546 | int charset, from, to; | |
547 | ||
548 | if (dimension == 1) | |
549 | { | |
550 | if (width == 1) | |
551 | from = LEADING_CODE_EXT_11, to = LEADING_CODE_EXT_12; | |
552 | else | |
553 | from = LEADING_CODE_EXT_12, to = LEADING_CODE_EXT_21; | |
554 | } | |
555 | else | |
556 | { | |
557 | if (width == 1) | |
558 | from = LEADING_CODE_EXT_21, to = LEADING_CODE_EXT_22; | |
559 | else | |
b0e3cf2b | 560 | from = LEADING_CODE_EXT_22, to = LEADING_CODE_EXT_MAX + 1; |
4ed46869 KH |
561 | } |
562 | ||
563 | for (charset = from; charset < to; charset++) | |
564 | if (!CHARSET_DEFINED_P (charset)) break; | |
565 | ||
566 | return make_number (charset < to ? charset : 0); | |
567 | } | |
568 | ||
569 | DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
570 | "Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR.\n\ | |
23d2a7f1 | 571 | If CHARSET-ID is nil, it is decided automatically, which means CHARSET is\n\ |
4ed46869 KH |
572 | treated as a private charset.\n\ |
573 | INFO-VECTOR is a vector of the format:\n\ | |
574 | [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE\n\ | |
575 | SHORT-NAME LONG-NAME DESCRIPTION]\n\ | |
576 | The meanings of each elements is as follows:\n\ | |
577 | DIMENSION (integer) is the number of bytes to represent a character: 1 or 2.\n\ | |
578 | CHARS (integer) is the number of characters in a dimension: 94 or 96.\n\ | |
579 | WIDTH (integer) is the number of columns a character in the charset\n\ | |
580 | occupies on the screen: one of 0, 1, and 2.\n\ | |
581 | \n\ | |
582 | DIRECTION (integer) is the rendering direction of characters in the\n\ | |
277576f6 KH |
583 | charset when rendering. If 0, render from left to right, else\n\ |
584 | render from right to left.\n\ | |
4ed46869 KH |
585 | \n\ |
586 | ISO-FINAL-CHAR (character) is the final character of the\n\ | |
587 | corresponding ISO 2022 charset.\n\ | |
588 | \n\ | |
589 | ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked\n\ | |
590 | while encoding to variants of ISO 2022 coding system, one of the\n\ | |
591 | following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).\n\ | |
592 | \n\ | |
593 | SHORT-NAME (string) is the short name to refer to the charset.\n\ | |
594 | \n\ | |
595 | LONG-NAME (string) is the long name to refer to the charset.\n\ | |
596 | \n\ | |
597 | DESCRIPTION (string) is the description string of the charset.") | |
598 | (charset_id, charset_symbol, info_vector) | |
599 | Lisp_Object charset_id, charset_symbol, info_vector; | |
600 | { | |
601 | Lisp_Object *vec; | |
602 | ||
603 | if (!NILP (charset_id)) | |
604 | CHECK_NUMBER (charset_id, 0); | |
605 | CHECK_SYMBOL (charset_symbol, 1); | |
606 | CHECK_VECTOR (info_vector, 2); | |
607 | ||
608 | if (! NILP (charset_id)) | |
609 | { | |
610 | if (! CHARSET_VALID_P (XINT (charset_id))) | |
611 | error ("Invalid CHARSET: %d", XINT (charset_id)); | |
612 | else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
613 | error ("Already defined charset: %d", XINT (charset_id)); | |
614 | } | |
615 | ||
616 | vec = XVECTOR (info_vector)->contents; | |
617 | if (XVECTOR (info_vector)->size != 9 | |
618 | || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
619 | || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
620 | || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
621 | || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
622 | || !INTEGERP (vec[4]) || !(XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~') | |
623 | || !INTEGERP (vec[5]) || !(XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
624 | || !STRINGP (vec[6]) | |
625 | || !STRINGP (vec[7]) | |
626 | || !STRINGP (vec[8])) | |
627 | error ("Invalid info-vector argument for defining charset %s", | |
628 | XSYMBOL (charset_symbol)->name->data); | |
629 | ||
630 | if (NILP (charset_id)) | |
631 | { | |
632 | charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
633 | if (XINT (charset_id) == 0) | |
634 | error ("There's no room for a new private charset %s", | |
635 | XSYMBOL (charset_symbol)->name->data); | |
636 | } | |
637 | ||
638 | update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
639 | vec[4], vec[5], vec[6], vec[7], vec[8]); | |
6dc0722d | 640 | Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
4ed46869 KH |
641 | CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
642 | Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
643 | return Qnil; | |
644 | } | |
645 | ||
8a73a704 KH |
646 | DEFUN ("generic-character-list", Fgeneric_character_list, |
647 | Sgeneric_character_list, 0, 0, 0, | |
648 | "Return a list of all possible generic characters.\n\ | |
649 | It includes a generic character for a charset not yet defined.") | |
650 | () | |
651 | { | |
652 | return Vgeneric_character_list; | |
653 | } | |
654 | ||
3fac5a51 KH |
655 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
656 | Sget_unused_iso_final_char, 2, 2, 0, | |
657 | "Return an unsed ISO's final char for a charset of DIMENISION and CHARS.\n\ | |
658 | DIMENSION is the number of bytes to represent a character: 1 or 2.\n\ | |
659 | CHARS is the number of characters in a dimension: 94 or 96.\n\ | |
660 | \n\ | |
661 | This final char is for private use, thus the range is `0' (48) .. `?' (63).\n\ | |
662 | If there's no unused final char for the specified kind of charset,\n\ | |
663 | return nil.") | |
664 | (dimension, chars) | |
665 | Lisp_Object dimension, chars; | |
666 | { | |
667 | int final_char; | |
668 | ||
669 | CHECK_NUMBER (dimension, 0); | |
670 | CHECK_NUMBER (chars, 1); | |
671 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
672 | error ("Invalid charset dimension %d, it should be 1 or 2", | |
673 | XINT (dimension)); | |
674 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
675 | error ("Invalid charset chars %d, it should be 94 or 96", | |
676 | XINT (chars)); | |
677 | for (final_char = '0'; final_char <= '?'; final_char++) | |
678 | { | |
679 | if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0) | |
680 | break; | |
681 | } | |
682 | return (final_char <= '?' ? make_number (final_char) : Qnil); | |
683 | } | |
684 | ||
4ed46869 KH |
685 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
686 | 4, 4, 0, | |
687 | "Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET.\n\ | |
688 | CHARSET should be defined by `defined-charset' in advance.") | |
689 | (dimension, chars, final_char, charset_symbol) | |
690 | Lisp_Object dimension, chars, final_char, charset_symbol; | |
691 | { | |
692 | int charset; | |
693 | ||
694 | CHECK_NUMBER (dimension, 0); | |
695 | CHECK_NUMBER (chars, 1); | |
696 | CHECK_NUMBER (final_char, 2); | |
697 | CHECK_SYMBOL (charset_symbol, 3); | |
698 | ||
699 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
700 | error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
701 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
702 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
703 | if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
704 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
705 | if ((charset = get_charset_id (charset_symbol)) < 0) | |
706 | error ("Invalid charset %s", XSYMBOL (charset_symbol)->name->data); | |
707 | ||
708 | ISO_CHARSET_TABLE (dimension, chars, final_char) = charset; | |
709 | return Qnil; | |
710 | } | |
711 | ||
712 | /* Return number of different charsets in STR of length LEN. In | |
713 | addition, for each found charset N, CHARSETS[N] is set 1. The | |
a29e3b1b | 714 | caller should allocate CHARSETS (MAX_CHARSET + 1 elements) in advance. |
1d67c29b KH |
715 | It may lookup a translation table TABLE if supplied. |
716 | ||
717 | If CMPCHARP is nonzero and some composite character is found, | |
718 | CHARSETS[128] is also set 1 and the returned number is incremented | |
719 | by 1. */ | |
4ed46869 KH |
720 | |
721 | int | |
1d67c29b | 722 | find_charset_in_str (str, len, charsets, table, cmpcharp) |
028d516b KH |
723 | unsigned char *str; |
724 | int len, *charsets; | |
23d2a7f1 | 725 | Lisp_Object table; |
1d67c29b | 726 | int cmpcharp; |
4ed46869 | 727 | { |
733eafd8 | 728 | register int num = 0, c; |
4ed46869 | 729 | |
23d2a7f1 KH |
730 | if (! CHAR_TABLE_P (table)) |
731 | table = Qnil; | |
732 | ||
4ed46869 KH |
733 | while (len > 0) |
734 | { | |
05505664 | 735 | int bytes, charset; |
733eafd8 | 736 | c = *str; |
23d2a7f1 | 737 | |
733eafd8 | 738 | if (c == LEADING_CODE_COMPOSITION) |
05505664 | 739 | { |
733eafd8 KH |
740 | int cmpchar_id = str_cmpchar_id (str, len); |
741 | GLYPH *glyph; | |
05505664 | 742 | |
1d67c29b | 743 | if (cmpchar_id >= 0) |
05505664 | 744 | { |
020da460 | 745 | struct cmpchar_info *cmp_p = cmpchar_table[cmpchar_id]; |
733eafd8 KH |
746 | int i; |
747 | ||
020da460 | 748 | for (i = 0; i < cmp_p->glyph_len; i++) |
733eafd8 | 749 | { |
020da460 | 750 | c = cmp_p->glyph[i]; |
733eafd8 KH |
751 | if (!NILP (table)) |
752 | { | |
b4e9dd77 | 753 | if ((c = translate_char (table, c, 0, 0, 0)) < 0) |
020da460 | 754 | c = cmp_p->glyph[i]; |
733eafd8 KH |
755 | } |
756 | if ((charset = CHAR_CHARSET (c)) < 0) | |
757 | charset = CHARSET_ASCII; | |
758 | if (!charsets[charset]) | |
759 | { | |
760 | charsets[charset] = 1; | |
761 | num += 1; | |
762 | } | |
763 | } | |
020da460 KH |
764 | str += cmp_p->len; |
765 | len -= cmp_p->len; | |
766 | if (cmpcharp && !charsets[CHARSET_COMPOSITION]) | |
767 | { | |
768 | charsets[CHARSET_COMPOSITION] = 1; | |
769 | num += 1; | |
770 | } | |
733eafd8 | 771 | continue; |
05505664 | 772 | } |
05505664 | 773 | |
733eafd8 KH |
774 | charset = CHARSET_ASCII; |
775 | bytes = 1; | |
776 | } | |
23d2a7f1 KH |
777 | else |
778 | { | |
733eafd8 KH |
779 | c = STRING_CHAR_AND_LENGTH (str, len, bytes); |
780 | if (! NILP (table)) | |
781 | { | |
b4e9dd77 | 782 | int c1 = translate_char (table, c, 0, 0, 0); |
733eafd8 KH |
783 | if (c1 >= 0) |
784 | c = c1; | |
785 | } | |
786 | charset = CHAR_CHARSET (c); | |
23d2a7f1 | 787 | } |
4ed46869 KH |
788 | |
789 | if (!charsets[charset]) | |
790 | { | |
791 | charsets[charset] = 1; | |
792 | num += 1; | |
793 | } | |
794 | str += bytes; | |
795 | len -= bytes; | |
796 | } | |
797 | return num; | |
798 | } | |
799 | ||
800 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 801 | 2, 3, 0, |
4ed46869 | 802 | "Return a list of charsets in the region between BEG and END.\n\ |
23d2a7f1 | 803 | BEG and END are buffer positions.\n\ |
020da460 KH |
804 | If the region contains any composite character,\n\ |
805 | `composition' is included in the returned list.\n\ | |
b4e9dd77 | 806 | Optional arg TABLE if non-nil is a translation table to look up.") |
23d2a7f1 KH |
807 | (beg, end, table) |
808 | Lisp_Object beg, end, table; | |
4ed46869 | 809 | { |
028d516b | 810 | int charsets[MAX_CHARSET + 1]; |
6ae1f27e | 811 | int from, from_byte, to, stop, stop_byte, i; |
4ed46869 KH |
812 | Lisp_Object val; |
813 | ||
814 | validate_region (&beg, &end); | |
815 | from = XFASTINT (beg); | |
816 | stop = to = XFASTINT (end); | |
6ae1f27e | 817 | |
2e45bf18 KH |
818 | if (NILP (current_buffer->enable_multibyte_characters)) |
819 | return (from == to | |
820 | ? Qnil | |
821 | : Fcons (Qascii, Qnil)); | |
822 | ||
4ed46869 | 823 | if (from < GPT && GPT < to) |
6ae1f27e RS |
824 | { |
825 | stop = GPT; | |
826 | stop_byte = GPT_BYTE; | |
827 | } | |
828 | else | |
829 | stop_byte = CHAR_TO_BYTE (stop); | |
830 | ||
831 | from_byte = CHAR_TO_BYTE (from); | |
832 | ||
028d516b | 833 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
4ed46869 KH |
834 | while (1) |
835 | { | |
6ae1f27e | 836 | find_charset_in_str (BYTE_POS_ADDR (from_byte), stop_byte - from_byte, |
020da460 | 837 | charsets, table, 1); |
4ed46869 | 838 | if (stop < to) |
6ae1f27e RS |
839 | { |
840 | from = stop, from_byte = stop_byte; | |
841 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
842 | } | |
4ed46869 KH |
843 | else |
844 | break; | |
845 | } | |
6ae1f27e | 846 | |
4ed46869 | 847 | val = Qnil; |
513ee442 | 848 | for (i = MAX_CHARSET; i >= 0; i--) |
4ed46869 KH |
849 | if (charsets[i]) |
850 | val = Fcons (CHARSET_SYMBOL (i), val); | |
851 | return val; | |
852 | } | |
853 | ||
854 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 KH |
855 | 1, 2, 0, |
856 | "Return a list of charsets in STR.\n\ | |
020da460 KH |
857 | If the string contains any composite characters,\n\ |
858 | `composition' is included in the returned list.\n\ | |
b4e9dd77 | 859 | Optional arg TABLE if non-nil is a translation table to look up.") |
23d2a7f1 KH |
860 | (str, table) |
861 | Lisp_Object str, table; | |
4ed46869 | 862 | { |
a29e3b1b | 863 | int charsets[MAX_CHARSET + 1]; |
4ed46869 KH |
864 | int i; |
865 | Lisp_Object val; | |
866 | ||
867 | CHECK_STRING (str, 0); | |
87b089ad RS |
868 | |
869 | if (! STRING_MULTIBYTE (str)) | |
2e45bf18 KH |
870 | return (XSTRING (str)->size == 0 |
871 | ? Qnil | |
872 | : Fcons (Qascii, Qnil)); | |
87b089ad | 873 | |
a29e3b1b | 874 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
fc932ac6 | 875 | find_charset_in_str (XSTRING (str)->data, STRING_BYTES (XSTRING (str)), |
020da460 | 876 | charsets, table, 1); |
4ed46869 | 877 | val = Qnil; |
513ee442 | 878 | for (i = MAX_CHARSET; i >= 0; i--) |
4ed46869 KH |
879 | if (charsets[i]) |
880 | val = Fcons (CHARSET_SYMBOL (i), val); | |
881 | return val; | |
882 | } | |
883 | \f | |
884 | DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
513ee442 | 885 | "") |
4ed46869 KH |
886 | (charset, code1, code2) |
887 | Lisp_Object charset, code1, code2; | |
888 | { | |
889 | CHECK_NUMBER (charset, 0); | |
890 | ||
891 | if (NILP (code1)) | |
892 | XSETFASTINT (code1, 0); | |
893 | else | |
894 | CHECK_NUMBER (code1, 1); | |
895 | if (NILP (code2)) | |
896 | XSETFASTINT (code2, 0); | |
897 | else | |
898 | CHECK_NUMBER (code2, 2); | |
899 | ||
900 | if (!CHARSET_DEFINED_P (XINT (charset))) | |
901 | error ("Invalid charset: %d", XINT (charset)); | |
902 | ||
903 | return make_number (MAKE_CHAR (XINT (charset), XINT (code1), XINT (code2))); | |
904 | } | |
905 | ||
906 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, | |
907 | "Return list of charset and one or two position-codes of CHAR.") | |
908 | (ch) | |
909 | Lisp_Object ch; | |
910 | { | |
911 | Lisp_Object val; | |
6dc0722d | 912 | int charset, c1, c2; |
4ed46869 KH |
913 | |
914 | CHECK_NUMBER (ch, 0); | |
915 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
6dc0722d | 916 | return (c2 >= 0 |
4ed46869 KH |
917 | ? Fcons (CHARSET_SYMBOL (charset), |
918 | Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
919 | : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
920 | } | |
921 | ||
922 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
923 | "Return charset of CHAR.") | |
924 | (ch) | |
925 | Lisp_Object ch; | |
926 | { | |
927 | CHECK_NUMBER (ch, 0); | |
928 | ||
929 | return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
930 | } | |
931 | ||
90d7b74e KH |
932 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
933 | "Return charset of a character in current buffer at position POS.\n\ | |
934 | If POS is nil, it defauls to the current point.") | |
935 | (pos) | |
936 | Lisp_Object pos; | |
937 | { | |
938 | register int pos_byte, c, charset; | |
939 | register unsigned char *p; | |
940 | ||
941 | if (NILP (pos)) | |
942 | pos_byte = PT_BYTE; | |
943 | else if (MARKERP (pos)) | |
944 | pos_byte = marker_byte_position (pos); | |
945 | else | |
946 | { | |
947 | CHECK_NUMBER (pos, 0); | |
948 | pos_byte = CHAR_TO_BYTE (XINT (pos)); | |
949 | } | |
950 | p = BYTE_POS_ADDR (pos_byte); | |
951 | c = STRING_CHAR (p, Z_BYTE - pos_byte); | |
952 | charset = CHAR_CHARSET (c); | |
953 | return CHARSET_SYMBOL (charset); | |
954 | } | |
955 | ||
4ed46869 | 956 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
2b71bb78 KH |
957 | "Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR.\n\ |
958 | \n\ | |
959 | ISO 2022's designation sequence (escape sequence) distinguishes charsets\n\ | |
960 | by their DIMENSION, CHARS, and FINAL-CHAR,\n\ | |
961 | where as Emacs distinguishes them by charset symbol.\n\ | |
962 | See the documentation of the function `charset-info' for the meanings of\n\ | |
963 | DIMENSION, CHARS, and FINAL-CHAR.") | |
4ed46869 KH |
964 | (dimension, chars, final_char) |
965 | Lisp_Object dimension, chars, final_char; | |
966 | { | |
967 | int charset; | |
968 | ||
969 | CHECK_NUMBER (dimension, 0); | |
970 | CHECK_NUMBER (chars, 1); | |
971 | CHECK_NUMBER (final_char, 2); | |
972 | ||
973 | if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
974 | return Qnil; | |
975 | return CHARSET_SYMBOL (charset); | |
976 | } | |
977 | ||
9d3d8cba KH |
978 | /* If GENERICP is nonzero, return nonzero iff C is a valid normal or |
979 | generic character. If GENERICP is zero, return nonzero iff C is a | |
980 | valid normal character. Do not call this function directly, | |
981 | instead use macro CHAR_VALID_P. */ | |
982 | int | |
983 | char_valid_p (c, genericp) | |
984 | int c, genericp; | |
985 | { | |
986 | int charset, c1, c2; | |
987 | ||
988 | if (c < 0) | |
989 | return 0; | |
990 | if (SINGLE_BYTE_CHAR_P (c)) | |
991 | return 1; | |
992 | SPLIT_NON_ASCII_CHAR (c, charset, c1, c2); | |
bb63e573 | 993 | if (!CHARSET_DEFINED_P (charset)) |
9d3d8cba KH |
994 | return 0; |
995 | return (c < MIN_CHAR_COMPOSITION | |
996 | ? ((c & CHAR_FIELD1_MASK) /* i.e. dimension of C is two. */ | |
997 | ? (genericp && c1 == 0 && c2 == 0 | |
998 | || c1 >= 32 && c2 >= 32) | |
999 | : (genericp && c1 == 0 | |
1000 | || c1 >= 32)) | |
1001 | : c < MIN_CHAR_COMPOSITION + n_cmpchars); | |
1002 | } | |
1003 | ||
1004 | DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0, | |
a9d02884 DL |
1005 | "Return t if OBJECT is a valid normal character.\n\ |
1006 | If optional arg GENERICP is non-nil, also return t if OBJECT is\n\ | |
9d3d8cba KH |
1007 | a valid generic character.") |
1008 | (object, genericp) | |
1009 | Lisp_Object object, genericp; | |
1010 | { | |
1011 | if (! NATNUMP (object)) | |
1012 | return Qnil; | |
1013 | return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil); | |
1014 | } | |
1015 | ||
d2665018 KH |
1016 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, |
1017 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
1018 | "Convert the unibyte character CH to multibyte character.\n\ | |
537efd8d | 1019 | The conversion is done based on `nonascii-translation-table' (which see)\n\ |
340b8d58 | 1020 | or `nonascii-insert-offset' (which see).") |
d2665018 KH |
1021 | (ch) |
1022 | Lisp_Object ch; | |
1023 | { | |
1024 | int c; | |
1025 | ||
1026 | CHECK_NUMBER (ch, 0); | |
1027 | c = XINT (ch); | |
1028 | if (c < 0 || c >= 0400) | |
1029 | error ("Invalid unibyte character: %d", c); | |
1030 | c = unibyte_char_to_multibyte (c); | |
1031 | if (c < 0) | |
1032 | error ("Can't convert to multibyte character: %d", XINT (ch)); | |
1033 | return make_number (c); | |
1034 | } | |
1035 | ||
1bcc1567 RS |
1036 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, |
1037 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
1038 | "Convert the multibyte character CH to unibyte character.\n\ | |
1039 | The conversion is done based on `nonascii-translation-table' (which see)\n\ | |
1040 | or `nonascii-insert-offset' (which see).") | |
1041 | (ch) | |
1042 | Lisp_Object ch; | |
1043 | { | |
1044 | int c; | |
1045 | ||
1046 | CHECK_NUMBER (ch, 0); | |
1047 | c = XINT (ch); | |
1048 | if (c < 0) | |
1049 | error ("Invalid multibyte character: %d", c); | |
1050 | c = multibyte_char_to_unibyte (c, Qnil); | |
1051 | if (c < 0) | |
1052 | error ("Can't convert to unibyte character: %d", XINT (ch)); | |
1053 | return make_number (c); | |
1054 | } | |
1055 | ||
4ed46869 | 1056 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, |
f78643ef | 1057 | "Return 1 regardless of the argument CHAR.\n\ |
60022cb7 | 1058 | This is now an obsolete function. We keep it just for backward compatibility.") |
4ed46869 KH |
1059 | (ch) |
1060 | Lisp_Object ch; | |
1061 | { | |
1062 | Lisp_Object val; | |
4ed46869 KH |
1063 | |
1064 | CHECK_NUMBER (ch, 0); | |
9b6a601f KH |
1065 | return make_number (1); |
1066 | } | |
1067 | ||
1068 | /* Return how many bytes C will occupy in a multibyte buffer. | |
1069 | Don't call this function directly, instead use macro CHAR_BYTES. */ | |
1070 | int | |
1071 | char_bytes (c) | |
1072 | int c; | |
1073 | { | |
1074 | int bytes; | |
1075 | ||
1076 | if (COMPOSITE_CHAR_P (c)) | |
4ed46869 | 1077 | { |
9b6a601f | 1078 | unsigned int id = COMPOSITE_CHAR_ID (c); |
4ed46869 KH |
1079 | |
1080 | bytes = (id < n_cmpchars ? cmpchar_table[id]->len : 1); | |
1081 | } | |
1082 | else | |
1083 | { | |
9b6a601f | 1084 | int charset = CHAR_CHARSET (c); |
4ed46869 KH |
1085 | |
1086 | bytes = CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1; | |
1087 | } | |
1088 | ||
60022cb7 | 1089 | return bytes; |
4ed46869 KH |
1090 | } |
1091 | ||
1092 | /* Return the width of character of which multi-byte form starts with | |
1093 | C. The width is measured by how many columns occupied on the | |
1094 | screen when displayed in the current buffer. */ | |
1095 | ||
1096 | #define ONE_BYTE_CHAR_WIDTH(c) \ | |
1097 | (c < 0x20 \ | |
1098 | ? (c == '\t' \ | |
53316e55 | 1099 | ? XFASTINT (current_buffer->tab_width) \ |
4ed46869 KH |
1100 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
1101 | : (c < 0x7f \ | |
1102 | ? 1 \ | |
1103 | : (c == 0x7F \ | |
1104 | ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
1105 | : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
1106 | && BASE_LEADING_CODE_P (c)) \ | |
1107 | ? WIDTH_BY_CHAR_HEAD (c) \ | |
b4e9dd77 | 1108 | : 4)))) |
4ed46869 KH |
1109 | |
1110 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
1111 | "Return width of CHAR when displayed in the current buffer.\n\ | |
1112 | The width is measured by how many columns it occupies on the screen.") | |
1113 | (ch) | |
1114 | Lisp_Object ch; | |
1115 | { | |
859f2b3c | 1116 | Lisp_Object val, disp; |
4ed46869 | 1117 | int c; |
51c4025f | 1118 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 KH |
1119 | |
1120 | CHECK_NUMBER (ch, 0); | |
1121 | ||
859f2b3c RS |
1122 | c = XINT (ch); |
1123 | ||
1124 | /* Get the way the display table would display it. */ | |
51c4025f | 1125 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; |
859f2b3c RS |
1126 | |
1127 | if (VECTORP (disp)) | |
1128 | XSETINT (val, XVECTOR (disp)->size); | |
1129 | else if (SINGLE_BYTE_CHAR_P (c)) | |
1130 | XSETINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
4ed46869 KH |
1131 | else if (COMPOSITE_CHAR_P (c)) |
1132 | { | |
1133 | int id = COMPOSITE_CHAR_ID (XFASTINT (ch)); | |
1134 | XSETFASTINT (val, (id < n_cmpchars ? cmpchar_table[id]->width : 0)); | |
1135 | } | |
1136 | else | |
1137 | { | |
1138 | int charset = CHAR_CHARSET (c); | |
1139 | ||
1140 | XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
1141 | } | |
1142 | return val; | |
1143 | } | |
1144 | ||
1145 | /* Return width of string STR of length LEN when displayed in the | |
1146 | current buffer. The width is measured by how many columns it | |
1147 | occupies on the screen. */ | |
859f2b3c | 1148 | |
4ed46869 KH |
1149 | int |
1150 | strwidth (str, len) | |
1151 | unsigned char *str; | |
1152 | int len; | |
1153 | { | |
1154 | unsigned char *endp = str + len; | |
1155 | int width = 0; | |
c4a4e28f | 1156 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1157 | |
859f2b3c RS |
1158 | while (str < endp) |
1159 | { | |
1160 | if (*str == LEADING_CODE_COMPOSITION) | |
1161 | { | |
1162 | int id = str_cmpchar_id (str, endp - str); | |
1163 | ||
1164 | if (id < 0) | |
1165 | { | |
1166 | width += 4; | |
1167 | str++; | |
1168 | } | |
1169 | else | |
1170 | { | |
1171 | width += cmpchar_table[id]->width; | |
1172 | str += cmpchar_table[id]->len; | |
1173 | } | |
1174 | } | |
1175 | else | |
1176 | { | |
1177 | Lisp_Object disp; | |
e515b0a9 KH |
1178 | int thislen; |
1179 | int c = STRING_CHAR_AND_LENGTH (str, endp - str, thislen); | |
859f2b3c RS |
1180 | |
1181 | /* Get the way the display table would display it. */ | |
acc35c36 RS |
1182 | if (dp) |
1183 | disp = DISP_CHAR_VECTOR (dp, c); | |
1184 | else | |
1185 | disp = Qnil; | |
859f2b3c RS |
1186 | |
1187 | if (VECTORP (disp)) | |
e515b0a9 | 1188 | width += XVECTOR (disp)->size; |
859f2b3c | 1189 | else |
e515b0a9 | 1190 | width += ONE_BYTE_CHAR_WIDTH (*str); |
859f2b3c | 1191 | |
e515b0a9 | 1192 | str += thislen; |
859f2b3c RS |
1193 | } |
1194 | } | |
4ed46869 KH |
1195 | return width; |
1196 | } | |
1197 | ||
1198 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
1199 | "Return width of STRING when displayed in the current buffer.\n\ | |
1200 | Width is measured by how many columns it occupies on the screen.\n\ | |
046b1f03 RS |
1201 | When calculating width of a multibyte character in STRING,\n\ |
1202 | only the base leading-code is considered; the validity of\n\ | |
1203 | the following bytes is not checked.") | |
4ed46869 KH |
1204 | (str) |
1205 | Lisp_Object str; | |
1206 | { | |
1207 | Lisp_Object val; | |
1208 | ||
1209 | CHECK_STRING (str, 0); | |
fc932ac6 RS |
1210 | XSETFASTINT (val, strwidth (XSTRING (str)->data, |
1211 | STRING_BYTES (XSTRING (str)))); | |
4ed46869 KH |
1212 | return val; |
1213 | } | |
1214 | ||
1215 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
1216 | "Return the direction of CHAR.\n\ | |
1217 | The returned value is 0 for left-to-right and 1 for right-to-left.") | |
1218 | (ch) | |
1219 | Lisp_Object ch; | |
1220 | { | |
1221 | int charset; | |
1222 | ||
1223 | CHECK_NUMBER (ch, 0); | |
1224 | charset = CHAR_CHARSET (XFASTINT (ch)); | |
1225 | if (!CHARSET_DEFINED_P (charset)) | |
93bcb785 | 1226 | invalid_character (XINT (ch)); |
4ed46869 KH |
1227 | return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); |
1228 | } | |
1229 | ||
af4fecb4 | 1230 | DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, |
6ae1f27e | 1231 | "Return number of characters between BEG and END.") |
046b1f03 RS |
1232 | (beg, end) |
1233 | Lisp_Object beg, end; | |
1234 | { | |
6ae1f27e | 1235 | int from, to; |
046b1f03 | 1236 | |
17e7ef1b RS |
1237 | CHECK_NUMBER_COERCE_MARKER (beg, 0); |
1238 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
1239 | ||
046b1f03 | 1240 | from = min (XFASTINT (beg), XFASTINT (end)); |
a8a35e61 | 1241 | to = max (XFASTINT (beg), XFASTINT (end)); |
046b1f03 | 1242 | |
a8c21066 | 1243 | return make_number (to - from); |
6ae1f27e | 1244 | } |
9036eb45 | 1245 | |
87b089ad RS |
1246 | /* Return the number of characters in the NBYTES bytes at PTR. |
1247 | This works by looking at the contents and checking for multibyte sequences. | |
1248 | However, if the current buffer has enable-multibyte-characters = nil, | |
1249 | we treat each byte as a character. */ | |
1250 | ||
6ae1f27e RS |
1251 | int |
1252 | chars_in_text (ptr, nbytes) | |
1253 | unsigned char *ptr; | |
1254 | int nbytes; | |
1255 | { | |
93bcb785 | 1256 | unsigned char *endp, c; |
6ae1f27e | 1257 | int chars; |
046b1f03 | 1258 | |
87b089ad RS |
1259 | /* current_buffer is null at early stages of Emacs initialization. */ |
1260 | if (current_buffer == 0 | |
1261 | || NILP (current_buffer->enable_multibyte_characters)) | |
6ae1f27e | 1262 | return nbytes; |
a8a35e61 | 1263 | |
6ae1f27e RS |
1264 | endp = ptr + nbytes; |
1265 | chars = 0; | |
046b1f03 | 1266 | |
6ae1f27e RS |
1267 | while (ptr < endp) |
1268 | { | |
93bcb785 KH |
1269 | c = *ptr++; |
1270 | ||
1271 | if (BASE_LEADING_CODE_P (c)) | |
1272 | while (ptr < endp && ! CHAR_HEAD_P (*ptr)) ptr++; | |
046b1f03 RS |
1273 | chars++; |
1274 | } | |
1275 | ||
6ae1f27e | 1276 | return chars; |
046b1f03 RS |
1277 | } |
1278 | ||
87b089ad RS |
1279 | /* Return the number of characters in the NBYTES bytes at PTR. |
1280 | This works by looking at the contents and checking for multibyte sequences. | |
1281 | It ignores enable-multibyte-characters. */ | |
1282 | ||
1283 | int | |
1284 | multibyte_chars_in_text (ptr, nbytes) | |
1285 | unsigned char *ptr; | |
1286 | int nbytes; | |
1287 | { | |
93bcb785 | 1288 | unsigned char *endp, c; |
87b089ad RS |
1289 | int chars; |
1290 | ||
1291 | endp = ptr + nbytes; | |
1292 | chars = 0; | |
1293 | ||
1294 | while (ptr < endp) | |
1295 | { | |
93bcb785 KH |
1296 | c = *ptr++; |
1297 | ||
1298 | if (BASE_LEADING_CODE_P (c)) | |
1299 | while (ptr < endp && ! CHAR_HEAD_P (*ptr)) ptr++; | |
87b089ad RS |
1300 | chars++; |
1301 | } | |
1302 | ||
1303 | return chars; | |
1304 | } | |
1305 | ||
1306 | DEFUN ("string", Fstring, Sstring, 1, MANY, 0, | |
4ed46869 | 1307 | "Concatenate all the argument characters and make the result a string.") |
53316e55 KH |
1308 | (n, args) |
1309 | int n; | |
4ed46869 KH |
1310 | Lisp_Object *args; |
1311 | { | |
53316e55 | 1312 | int i; |
4ed46869 | 1313 | unsigned char *buf |
bd4c6dd0 | 1314 | = (unsigned char *) alloca (MAX_LENGTH_OF_MULTI_BYTE_FORM * n); |
4ed46869 KH |
1315 | unsigned char *p = buf; |
1316 | Lisp_Object val; | |
1317 | ||
1318 | for (i = 0; i < n; i++) | |
1319 | { | |
1320 | int c, len; | |
1321 | unsigned char *str; | |
1322 | ||
1323 | if (!INTEGERP (args[i])) | |
b0e3cf2b | 1324 | CHECK_NUMBER (args[i], 0); |
4ed46869 KH |
1325 | c = XINT (args[i]); |
1326 | len = CHAR_STRING (c, p, str); | |
1327 | if (p != str) | |
1328 | /* C is a composite character. */ | |
1329 | bcopy (str, p, len); | |
1330 | p += len; | |
1331 | } | |
1332 | ||
020da460 KH |
1333 | /* Here, we can't use make_string_from_bytes because of byte |
1334 | combining problem. */ | |
1335 | val = make_string (buf, p - buf); | |
4ed46869 KH |
1336 | return val; |
1337 | } | |
1338 | ||
1339 | #endif /* emacs */ | |
1340 | \f | |
1341 | /*** Composite characters staffs ***/ | |
1342 | ||
1343 | /* Each composite character is identified by CMPCHAR-ID which is | |
1344 | assigned when Emacs needs the character code of the composite | |
1345 | character (e.g. when displaying it on the screen). See the | |
1346 | document "GENERAL NOTE on COMPOSITE CHARACTER" in `charset.h' how a | |
1347 | composite character is represented in Emacs. */ | |
1348 | ||
1349 | /* If `static' is defined, it means that it is defined to null string. */ | |
1350 | #ifndef static | |
1351 | /* The following function is copied from lread.c. */ | |
1352 | static int | |
1353 | hash_string (ptr, len) | |
1354 | unsigned char *ptr; | |
1355 | int len; | |
1356 | { | |
1357 | register unsigned char *p = ptr; | |
1358 | register unsigned char *end = p + len; | |
1359 | register unsigned char c; | |
1360 | register int hash = 0; | |
1361 | ||
1362 | while (p != end) | |
1363 | { | |
1364 | c = *p++; | |
1365 | if (c >= 0140) c -= 40; | |
1366 | hash = ((hash<<3) + (hash>>28) + c); | |
1367 | } | |
1368 | return hash & 07777777777; | |
1369 | } | |
1370 | #endif | |
1371 | ||
4ed46869 KH |
1372 | #define CMPCHAR_HASH_TABLE_SIZE 0xFFF |
1373 | ||
1374 | static int *cmpchar_hash_table[CMPCHAR_HASH_TABLE_SIZE]; | |
1375 | ||
1376 | /* Each element of `cmpchar_hash_table' is a pointer to an array of | |
1377 | integer, where the 1st element is the size of the array, the 2nd | |
1378 | element is how many elements are actually used in the array, and | |
1379 | the remaining elements are CMPCHAR-IDs of composite characters of | |
1380 | the same hash value. */ | |
1381 | #define CMPCHAR_HASH_SIZE(table) table[0] | |
1382 | #define CMPCHAR_HASH_USED(table) table[1] | |
1383 | #define CMPCHAR_HASH_CMPCHAR_ID(table, i) table[i] | |
1384 | ||
1385 | /* Return CMPCHAR-ID of the composite character in STR of the length | |
1386 | LEN. If the composite character has not yet been registered, | |
1387 | register it in `cmpchar_table' and assign new CMPCHAR-ID. This | |
1388 | is the sole function for assigning CMPCHAR-ID. */ | |
1389 | int | |
1390 | str_cmpchar_id (str, len) | |
8867de67 | 1391 | const unsigned char *str; |
4ed46869 KH |
1392 | int len; |
1393 | { | |
1394 | int hash_idx, *hashp; | |
1395 | unsigned char *buf; | |
1396 | int embedded_rule; /* 1 if composition rule is embedded. */ | |
1397 | int chars; /* number of components. */ | |
1398 | int i; | |
1399 | struct cmpchar_info *cmpcharp; | |
1400 | ||
4ed46869 KH |
1401 | /* The second byte 0xFF means compostion rule is embedded. */ |
1402 | embedded_rule = (str[1] == 0xFF); | |
1403 | ||
1404 | /* At first, get the actual length of the composite character. */ | |
1405 | { | |
8867de67 | 1406 | const unsigned char *p, *endp = str + 1, *lastp = str + len; |
4ed46869 KH |
1407 | int bytes; |
1408 | ||
6ae1f27e | 1409 | while (endp < lastp && ! CHAR_HEAD_P (*endp)) endp++; |
93bcb785 KH |
1410 | if (endp - str < 5) |
1411 | /* Any composite char have at least 5-byte length. */ | |
1412 | return -1; | |
1413 | ||
4ed46869 | 1414 | chars = 0; |
93bcb785 | 1415 | p = str + 1; |
4ed46869 KH |
1416 | while (p < endp) |
1417 | { | |
93bcb785 | 1418 | if (embedded_rule) p++; |
4ed46869 | 1419 | /* No need of checking if *P is 0xA0 because |
93bcb785 KH |
1420 | BYTES_BY_CHAR_HEAD (0x80) surely returns 2. */ |
1421 | p += BYTES_BY_CHAR_HEAD (*p - 0x20); | |
4ed46869 KH |
1422 | chars++; |
1423 | } | |
93bcb785 KH |
1424 | if (p > endp || chars < 2 || chars > MAX_COMPONENT_COUNT) |
1425 | /* Invalid components. */ | |
4ed46869 | 1426 | return -1; |
93bcb785 | 1427 | len = p - str; |
4ed46869 KH |
1428 | } |
1429 | hash_idx = hash_string (str, len) % CMPCHAR_HASH_TABLE_SIZE; | |
1430 | hashp = cmpchar_hash_table[hash_idx]; | |
1431 | ||
1432 | /* Then, look into the hash table. */ | |
1433 | if (hashp != NULL) | |
1434 | /* Find the correct one among composite characters of the same | |
1435 | hash value. */ | |
1436 | for (i = 2; i < CMPCHAR_HASH_USED (hashp); i++) | |
1437 | { | |
1438 | cmpcharp = cmpchar_table[CMPCHAR_HASH_CMPCHAR_ID (hashp, i)]; | |
1439 | if (len == cmpcharp->len | |
1440 | && ! bcmp (str, cmpcharp->data, len)) | |
1441 | return CMPCHAR_HASH_CMPCHAR_ID (hashp, i); | |
1442 | } | |
1443 | ||
1444 | /* We have to register the composite character in cmpchar_table. */ | |
513ee442 KH |
1445 | if (n_cmpchars > (CHAR_FIELD2_MASK | CHAR_FIELD3_MASK)) |
1446 | /* No, we have no more room for a new composite character. */ | |
1447 | return -1; | |
1448 | ||
4ed46869 KH |
1449 | /* Make the entry in hash table. */ |
1450 | if (hashp == NULL) | |
1451 | { | |
1452 | /* Make a table for 8 composite characters initially. */ | |
1453 | hashp = (cmpchar_hash_table[hash_idx] | |
1454 | = (int *) xmalloc (sizeof (int) * (2 + 8))); | |
1455 | CMPCHAR_HASH_SIZE (hashp) = 10; | |
1456 | CMPCHAR_HASH_USED (hashp) = 2; | |
1457 | } | |
1458 | else if (CMPCHAR_HASH_USED (hashp) >= CMPCHAR_HASH_SIZE (hashp)) | |
1459 | { | |
1460 | CMPCHAR_HASH_SIZE (hashp) += 8; | |
1461 | hashp = (cmpchar_hash_table[hash_idx] | |
1462 | = (int *) xrealloc (hashp, | |
1463 | sizeof (int) * CMPCHAR_HASH_SIZE (hashp))); | |
1464 | } | |
1465 | CMPCHAR_HASH_CMPCHAR_ID (hashp, CMPCHAR_HASH_USED (hashp)) = n_cmpchars; | |
1466 | CMPCHAR_HASH_USED (hashp)++; | |
1467 | ||
1468 | /* Set information of the composite character in cmpchar_table. */ | |
1469 | if (cmpchar_table_size == 0) | |
1470 | { | |
1471 | /* This is the first composite character to be registered. */ | |
1472 | cmpchar_table_size = 256; | |
1473 | cmpchar_table | |
1474 | = (struct cmpchar_info **) xmalloc (sizeof (cmpchar_table[0]) | |
1475 | * cmpchar_table_size); | |
1476 | } | |
1477 | else if (cmpchar_table_size <= n_cmpchars) | |
1478 | { | |
1479 | cmpchar_table_size += 256; | |
1480 | cmpchar_table | |
1481 | = (struct cmpchar_info **) xrealloc (cmpchar_table, | |
1482 | sizeof (cmpchar_table[0]) | |
1483 | * cmpchar_table_size); | |
1484 | } | |
1485 | ||
1486 | cmpcharp = (struct cmpchar_info *) xmalloc (sizeof (struct cmpchar_info)); | |
1487 | ||
1488 | cmpcharp->len = len; | |
1489 | cmpcharp->data = (unsigned char *) xmalloc (len + 1); | |
1490 | bcopy (str, cmpcharp->data, len); | |
1491 | cmpcharp->data[len] = 0; | |
1492 | cmpcharp->glyph_len = chars; | |
1493 | cmpcharp->glyph = (GLYPH *) xmalloc (sizeof (GLYPH) * chars); | |
1494 | if (embedded_rule) | |
1495 | { | |
1496 | cmpcharp->cmp_rule = (unsigned char *) xmalloc (chars); | |
1497 | cmpcharp->col_offset = (float *) xmalloc (sizeof (float) * chars); | |
1498 | } | |
1499 | else | |
1500 | { | |
1501 | cmpcharp->cmp_rule = NULL; | |
1502 | cmpcharp->col_offset = NULL; | |
1503 | } | |
1504 | ||
1505 | /* Setup GLYPH data and composition rules (if any) so as not to make | |
1506 | them every time on displaying. */ | |
1507 | { | |
1508 | unsigned char *bufp; | |
1509 | int width; | |
1510 | float leftmost = 0.0, rightmost = 1.0; | |
1511 | ||
1512 | if (embedded_rule) | |
1513 | /* At first, col_offset[N] is set to relative to col_offset[0]. */ | |
1514 | cmpcharp->col_offset[0] = 0; | |
1515 | ||
1516 | for (i = 0, bufp = cmpcharp->data + 1; i < chars; i++) | |
1517 | { | |
1518 | if (embedded_rule) | |
1519 | cmpcharp->cmp_rule[i] = *bufp++; | |
1520 | ||
1521 | if (*bufp == 0xA0) /* This is an ASCII character. */ | |
1522 | { | |
1523 | cmpcharp->glyph[i] = FAST_MAKE_GLYPH ((*++bufp & 0x7F), 0); | |
1524 | width = 1; | |
1525 | bufp++; | |
1526 | } | |
1527 | else /* Multibyte character. */ | |
1528 | { | |
1529 | /* Make `bufp' point normal multi-byte form temporally. */ | |
1530 | *bufp -= 0x20; | |
1531 | cmpcharp->glyph[i] | |
537efd8d | 1532 | = FAST_MAKE_GLYPH (string_to_non_ascii_char (bufp, 4, 0, 0), 0); |
4ed46869 KH |
1533 | width = WIDTH_BY_CHAR_HEAD (*bufp); |
1534 | *bufp += 0x20; | |
1535 | bufp += BYTES_BY_CHAR_HEAD (*bufp - 0x20); | |
1536 | } | |
1537 | ||
1538 | if (embedded_rule && i > 0) | |
1539 | { | |
1540 | /* Reference points (global_ref and new_ref) are | |
1541 | encoded as below: | |
1542 | ||
1543 | 0--1--2 -- ascent | |
1544 | | | | |
1545 | | | | |
1546 | | 4 -+--- center | |
1547 | -- 3 5 -- baseline | |
1548 | | | | |
1549 | 6--7--8 -- descent | |
1550 | ||
1551 | Now, we calculate the column offset of the new glyph | |
1552 | from the left edge of the first glyph. This can avoid | |
1553 | the same calculation everytime displaying this | |
1554 | composite character. */ | |
1555 | ||
1556 | /* Reference points of global glyph and new glyph. */ | |
1557 | int global_ref = (cmpcharp->cmp_rule[i] - 0xA0) / 9; | |
1558 | int new_ref = (cmpcharp->cmp_rule[i] - 0xA0) % 9; | |
1559 | /* Column offset relative to the first glyph. */ | |
1560 | float left = (leftmost | |
1561 | + (global_ref % 3) * (rightmost - leftmost) / 2.0 | |
1562 | - (new_ref % 3) * width / 2.0); | |
1563 | ||
1564 | cmpcharp->col_offset[i] = left; | |
1565 | if (left < leftmost) | |
1566 | leftmost = left; | |
1567 | if (left + width > rightmost) | |
1568 | rightmost = left + width; | |
1569 | } | |
1570 | else | |
1571 | { | |
1572 | if (width > rightmost) | |
1573 | rightmost = width; | |
1574 | } | |
1575 | } | |
1576 | if (embedded_rule) | |
1577 | { | |
1578 | /* Now col_offset[N] are relative to the left edge of the | |
1579 | first component. Make them relative to the left edge of | |
1580 | overall glyph. */ | |
1581 | for (i = 0; i < chars; i++) | |
1582 | cmpcharp->col_offset[i] -= leftmost; | |
1583 | /* Make rightmost holds width of overall glyph. */ | |
1584 | rightmost -= leftmost; | |
1585 | } | |
1586 | ||
1587 | cmpcharp->width = rightmost; | |
1588 | if (cmpcharp->width < rightmost) | |
1589 | /* To get a ceiling integer value. */ | |
1590 | cmpcharp->width++; | |
1591 | } | |
1592 | ||
1593 | cmpchar_table[n_cmpchars] = cmpcharp; | |
1594 | ||
1595 | return n_cmpchars++; | |
1596 | } | |
1597 | ||
1598 | /* Return the Nth element of the composite character C. */ | |
1599 | int | |
1600 | cmpchar_component (c, n) | |
1601 | unsigned int c, n; | |
1602 | { | |
1603 | int id = COMPOSITE_CHAR_ID (c); | |
1604 | ||
1605 | if (id >= n_cmpchars /* C is not a valid composite character. */ | |
1606 | || n >= cmpchar_table[id]->glyph_len) /* No such component. */ | |
1607 | return -1; | |
1608 | /* No face data is stored in glyph code. */ | |
1609 | return ((int) (cmpchar_table[id]->glyph[n])); | |
1610 | } | |
1611 | ||
1612 | DEFUN ("cmpcharp", Fcmpcharp, Scmpcharp, 1, 1, 0, | |
1613 | "T if CHAR is a composite character.") | |
1614 | (ch) | |
1615 | Lisp_Object ch; | |
1616 | { | |
1617 | CHECK_NUMBER (ch, 0); | |
1618 | return (COMPOSITE_CHAR_P (XINT (ch)) ? Qt : Qnil); | |
1619 | } | |
1620 | ||
1621 | DEFUN ("composite-char-component", Fcmpchar_component, Scmpchar_component, | |
1622 | 2, 2, 0, | |
1623 | "Return the IDXth component character of composite character CHARACTER.") | |
1624 | (character, idx) | |
1625 | Lisp_Object character, idx; | |
1626 | { | |
1627 | int c; | |
1628 | ||
1629 | CHECK_NUMBER (character, 0); | |
1630 | CHECK_NUMBER (idx, 1); | |
1631 | ||
1632 | if ((c = cmpchar_component (XINT (character), XINT (idx))) < 0) | |
1633 | args_out_of_range (character, idx); | |
1634 | ||
1635 | return make_number (c); | |
1636 | } | |
1637 | ||
1638 | DEFUN ("composite-char-composition-rule", Fcmpchar_cmp_rule, Scmpchar_cmp_rule, | |
1639 | 2, 2, 0, | |
55001746 KH |
1640 | "Return the Nth composition rule embedded in composite character CHARACTER.\n\ |
1641 | The returned rule is for composing the Nth component\n\ | |
1642 | on the (N-1)th component. If N is 0, the returned value is always 255.") | |
1643 | (character, n) | |
1644 | Lisp_Object character, n; | |
4ed46869 KH |
1645 | { |
1646 | int id, i; | |
1647 | ||
1648 | CHECK_NUMBER (character, 0); | |
55001746 | 1649 | CHECK_NUMBER (n, 1); |
4ed46869 KH |
1650 | |
1651 | id = COMPOSITE_CHAR_ID (XINT (character)); | |
1652 | if (id < 0 || id >= n_cmpchars) | |
1653 | error ("Invalid composite character: %d", XINT (character)); | |
55001746 | 1654 | i = XINT (n); |
4ed46869 | 1655 | if (i > cmpchar_table[id]->glyph_len) |
55001746 | 1656 | args_out_of_range (character, n); |
4ed46869 KH |
1657 | |
1658 | return make_number (cmpchar_table[id]->cmp_rule[i]); | |
1659 | } | |
1660 | ||
1661 | DEFUN ("composite-char-composition-rule-p", Fcmpchar_cmp_rule_p, | |
1662 | Scmpchar_cmp_rule_p, 1, 1, 0, | |
1663 | "Return non-nil if composite character CHARACTER contains a embedded rule.") | |
1664 | (character) | |
1665 | Lisp_Object character; | |
1666 | { | |
1667 | int id; | |
1668 | ||
1669 | CHECK_NUMBER (character, 0); | |
1670 | id = COMPOSITE_CHAR_ID (XINT (character)); | |
1671 | if (id < 0 || id >= n_cmpchars) | |
1672 | error ("Invalid composite character: %d", XINT (character)); | |
1673 | ||
1674 | return (cmpchar_table[id]->cmp_rule ? Qt : Qnil); | |
1675 | } | |
1676 | ||
1677 | DEFUN ("composite-char-component-count", Fcmpchar_cmp_count, | |
1678 | Scmpchar_cmp_count, 1, 1, 0, | |
1679 | "Return number of compoents of composite character CHARACTER.") | |
1680 | (character) | |
1681 | Lisp_Object character; | |
1682 | { | |
1683 | int id; | |
1684 | ||
1685 | CHECK_NUMBER (character, 0); | |
1686 | id = COMPOSITE_CHAR_ID (XINT (character)); | |
1687 | if (id < 0 || id >= n_cmpchars) | |
1688 | error ("Invalid composite character: %d", XINT (character)); | |
1689 | ||
1690 | return (make_number (cmpchar_table[id]->glyph_len)); | |
1691 | } | |
1692 | ||
1693 | DEFUN ("compose-string", Fcompose_string, Scompose_string, | |
1694 | 1, 1, 0, | |
1695 | "Return one char string composed from all characters in STRING.") | |
1696 | (str) | |
1697 | Lisp_Object str; | |
1698 | { | |
1699 | unsigned char buf[MAX_LENGTH_OF_MULTI_BYTE_FORM], *p, *pend, *ptemp; | |
1700 | int len, i; | |
1701 | ||
1702 | CHECK_STRING (str, 0); | |
1703 | ||
1704 | buf[0] = LEADING_CODE_COMPOSITION; | |
1705 | p = XSTRING (str)->data; | |
fc932ac6 | 1706 | pend = p + STRING_BYTES (XSTRING (str)); |
4ed46869 KH |
1707 | i = 1; |
1708 | while (p < pend) | |
1709 | { | |
1710 | if (*p < 0x20 || *p == 127) /* control code */ | |
1711 | error ("Invalid component character: %d", *p); | |
1712 | else if (*p < 0x80) /* ASCII */ | |
1713 | { | |
1714 | if (i + 2 >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1715 | error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1716 | /* Prepend an ASCII charset indicator 0xA0, set MSB of the | |
1717 | code itself. */ | |
1718 | buf[i++] = 0xA0; | |
1719 | buf[i++] = *p++ + 0x80; | |
1720 | } | |
1721 | else if (*p == LEADING_CODE_COMPOSITION) /* composite char */ | |
1722 | { | |
1723 | /* Already composed. Eliminate the heading | |
1724 | LEADING_CODE_COMPOSITION, keep the remaining bytes | |
1725 | unchanged. */ | |
1726 | p++; | |
1727 | ptemp = p; | |
6ae1f27e | 1728 | while (! CHAR_HEAD_P (*p)) p++; |
4ed46869 KH |
1729 | if (i + (p - ptemp) >= MAX_LENGTH_OF_MULTI_BYTE_FORM) |
1730 | error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1731 | bcopy (ptemp, buf + i, p - ptemp); | |
1732 | i += p - ptemp; | |
1733 | } | |
1734 | else /* multibyte char */ | |
1735 | { | |
1736 | /* Add 0x20 to the base leading-code, keep the remaining | |
1737 | bytes unchanged. */ | |
1738 | len = BYTES_BY_CHAR_HEAD (*p); | |
1739 | if (i + len >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1740 | error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1741 | bcopy (p, buf + i, len); | |
1742 | buf[i] += 0x20; | |
1743 | p += len, i += len; | |
1744 | } | |
1745 | } | |
1746 | ||
1747 | if (i < 5) | |
1748 | /* STR contains only one character, which can't be composed. */ | |
1749 | error ("Too short string to be composed: %s", XSTRING (str)->data); | |
1750 | ||
27802600 | 1751 | return make_string_from_bytes (buf, 1, i); |
4ed46869 KH |
1752 | } |
1753 | ||
1754 | \f | |
dfcf069d | 1755 | int |
4ed46869 KH |
1756 | charset_id_internal (charset_name) |
1757 | char *charset_name; | |
1758 | { | |
76d7b829 | 1759 | Lisp_Object val; |
4ed46869 | 1760 | |
76d7b829 | 1761 | val= Fget (intern (charset_name), Qcharset); |
4ed46869 KH |
1762 | if (!VECTORP (val)) |
1763 | error ("Charset %s is not defined", charset_name); | |
1764 | ||
1765 | return (XINT (XVECTOR (val)->contents[0])); | |
1766 | } | |
1767 | ||
1768 | DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
1769 | Ssetup_special_charsets, 0, 0, 0, "Internal use only.") | |
1770 | () | |
1771 | { | |
1772 | charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1773 | charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1774 | charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1775 | charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1776 | charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1777 | charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1778 | charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1779 | return Qnil; | |
1780 | } | |
1781 | ||
dfcf069d | 1782 | void |
4ed46869 KH |
1783 | init_charset_once () |
1784 | { | |
1785 | int i, j, k; | |
1786 | ||
1787 | staticpro (&Vcharset_table); | |
1788 | staticpro (&Vcharset_symbol_table); | |
8a73a704 | 1789 | staticpro (&Vgeneric_character_list); |
4ed46869 KH |
1790 | |
1791 | /* This has to be done here, before we call Fmake_char_table. */ | |
1792 | Qcharset_table = intern ("charset-table"); | |
1793 | staticpro (&Qcharset_table); | |
1794 | ||
1795 | /* Intern this now in case it isn't already done. | |
1796 | Setting this variable twice is harmless. | |
1797 | But don't staticpro it here--that is done in alloc.c. */ | |
1798 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1799 | ||
1800 | /* Now we are ready to set up this property, so we can | |
1801 | create the charset table. */ | |
1802 | Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1803 | Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1804 | ||
513ee442 | 1805 | Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), Qnil); |
4ed46869 KH |
1806 | |
1807 | /* Setup tables. */ | |
1808 | for (i = 0; i < 2; i++) | |
1809 | for (j = 0; j < 2; j++) | |
1810 | for (k = 0; k < 128; k++) | |
1811 | iso_charset_table [i][j][k] = -1; | |
1812 | ||
1813 | bzero (cmpchar_hash_table, sizeof cmpchar_hash_table); | |
1814 | cmpchar_table_size = n_cmpchars = 0; | |
1815 | ||
6ef23ebb | 1816 | for (i = 0; i < 128; i++) |
4ed46869 | 1817 | BYTES_BY_CHAR_HEAD (i) = 1; |
6ef23ebb KH |
1818 | for (i = MIN_CHARSET_OFFICIAL_DIMENSION1; |
1819 | i <= MAX_CHARSET_OFFICIAL_DIMENSION1; i++) | |
1820 | BYTES_BY_CHAR_HEAD (i) = 2; | |
1821 | for (i = MIN_CHARSET_OFFICIAL_DIMENSION2; | |
1822 | i <= MAX_CHARSET_OFFICIAL_DIMENSION2; i++) | |
1823 | BYTES_BY_CHAR_HEAD (i) = 3; | |
4ed46869 KH |
1824 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_11) = 3; |
1825 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_12) = 3; | |
1826 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_21) = 4; | |
1827 | BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_22) = 4; | |
6ef23ebb | 1828 | /* The followings don't reflect the actual bytes, but just to tell |
4ed46869 KH |
1829 | that it is a start of a multibyte character. */ |
1830 | BYTES_BY_CHAR_HEAD (LEADING_CODE_COMPOSITION) = 2; | |
6ef23ebb KH |
1831 | BYTES_BY_CHAR_HEAD (0x9E) = 2; |
1832 | BYTES_BY_CHAR_HEAD (0x9F) = 2; | |
4ed46869 KH |
1833 | |
1834 | for (i = 0; i < 128; i++) | |
1835 | WIDTH_BY_CHAR_HEAD (i) = 1; | |
1836 | for (; i < 256; i++) | |
1837 | WIDTH_BY_CHAR_HEAD (i) = 4; | |
1838 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_11) = 1; | |
1839 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_12) = 2; | |
1840 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_21) = 1; | |
1841 | WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_22) = 2; | |
8a73a704 KH |
1842 | |
1843 | { | |
76d7b829 | 1844 | Lisp_Object val; |
8a73a704 | 1845 | |
76d7b829 | 1846 | val = Qnil; |
8a73a704 KH |
1847 | for (i = 0x81; i < 0x90; i++) |
1848 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1849 | for (; i < 0x9A; i++) | |
1850 | val = Fcons (make_number ((i - 0x8F) << 14), val); | |
1851 | for (i = 0xA0; i < 0xF0; i++) | |
1852 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1853 | for (; i < 0xFF; i++) | |
1854 | val = Fcons (make_number ((i - 0xE0) << 14), val); | |
1855 | val = Fcons (make_number (GENERIC_COMPOSITION_CHAR), val); | |
1856 | Vgeneric_character_list = Fnreverse (val); | |
1857 | } | |
bbf12bb3 KH |
1858 | |
1859 | nonascii_insert_offset = 0; | |
1860 | Vnonascii_translation_table = Qnil; | |
4ed46869 KH |
1861 | } |
1862 | ||
1863 | #ifdef emacs | |
1864 | ||
dfcf069d | 1865 | void |
4ed46869 KH |
1866 | syms_of_charset () |
1867 | { | |
1868 | Qascii = intern ("ascii"); | |
1869 | staticpro (&Qascii); | |
1870 | ||
1871 | Qcharset = intern ("charset"); | |
1872 | staticpro (&Qcharset); | |
1873 | ||
1874 | /* Define ASCII charset now. */ | |
1875 | update_charset_table (make_number (CHARSET_ASCII), | |
1876 | make_number (1), make_number (94), | |
1877 | make_number (1), | |
1878 | make_number (0), | |
1879 | make_number ('B'), | |
1880 | make_number (0), | |
1881 | build_string ("ASCII"), | |
1882 | build_string ("ASCII"), | |
1883 | build_string ("ASCII (ISO646 IRV)")); | |
1884 | CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1885 | Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1886 | ||
1887 | Qcomposition = intern ("composition"); | |
1888 | staticpro (&Qcomposition); | |
1889 | CHARSET_SYMBOL (CHARSET_COMPOSITION) = Qcomposition; | |
1890 | ||
c1a08b4c KH |
1891 | Qauto_fill_chars = intern ("auto-fill-chars"); |
1892 | staticpro (&Qauto_fill_chars); | |
1893 | Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0)); | |
1894 | ||
4ed46869 | 1895 | defsubr (&Sdefine_charset); |
8a73a704 | 1896 | defsubr (&Sgeneric_character_list); |
3fac5a51 | 1897 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
1898 | defsubr (&Sdeclare_equiv_charset); |
1899 | defsubr (&Sfind_charset_region); | |
1900 | defsubr (&Sfind_charset_string); | |
1901 | defsubr (&Smake_char_internal); | |
1902 | defsubr (&Ssplit_char); | |
1903 | defsubr (&Schar_charset); | |
90d7b74e | 1904 | defsubr (&Scharset_after); |
4ed46869 | 1905 | defsubr (&Siso_charset); |
9d3d8cba | 1906 | defsubr (&Schar_valid_p); |
d2665018 | 1907 | defsubr (&Sunibyte_char_to_multibyte); |
1bcc1567 | 1908 | defsubr (&Smultibyte_char_to_unibyte); |
4ed46869 KH |
1909 | defsubr (&Schar_bytes); |
1910 | defsubr (&Schar_width); | |
1911 | defsubr (&Sstring_width); | |
1912 | defsubr (&Schar_direction); | |
af4fecb4 | 1913 | defsubr (&Schars_in_region); |
87b089ad | 1914 | defsubr (&Sstring); |
4ed46869 KH |
1915 | defsubr (&Scmpcharp); |
1916 | defsubr (&Scmpchar_component); | |
1917 | defsubr (&Scmpchar_cmp_rule); | |
1918 | defsubr (&Scmpchar_cmp_rule_p); | |
1919 | defsubr (&Scmpchar_cmp_count); | |
1920 | defsubr (&Scompose_string); | |
1921 | defsubr (&Ssetup_special_charsets); | |
1922 | ||
1923 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
1924 | "List of charsets ever defined."); | |
1925 | Vcharset_list = Fcons (Qascii, Qnil); | |
1926 | ||
537efd8d | 1927 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, |
b4e9dd77 KH |
1928 | "Vector of cons cell of a symbol and translation table ever defined.\n\ |
1929 | An ID of a translation table is an index of this vector."); | |
537efd8d | 1930 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
b0e3cf2b | 1931 | |
4ed46869 KH |
1932 | DEFVAR_INT ("leading-code-composition", &leading_code_composition, |
1933 | "Leading-code of composite characters."); | |
1934 | leading_code_composition = LEADING_CODE_COMPOSITION; | |
1935 | ||
1936 | DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, | |
1937 | "Leading-code of private TYPE9N charset of column-width 1."); | |
1938 | leading_code_private_11 = LEADING_CODE_PRIVATE_11; | |
1939 | ||
1940 | DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
1941 | "Leading-code of private TYPE9N charset of column-width 2."); | |
1942 | leading_code_private_12 = LEADING_CODE_PRIVATE_12; | |
1943 | ||
1944 | DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
1945 | "Leading-code of private TYPE9Nx9N charset of column-width 1."); | |
1946 | leading_code_private_21 = LEADING_CODE_PRIVATE_21; | |
1947 | ||
1948 | DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
1949 | "Leading-code of private TYPE9Nx9N charset of column-width 2."); | |
1950 | leading_code_private_22 = LEADING_CODE_PRIVATE_22; | |
35e623fb RS |
1951 | |
1952 | DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset, | |
d2665018 | 1953 | "Offset for converting non-ASCII unibyte codes 0240...0377 to multibyte.\n\ |
4cf9710d RS |
1954 | This is used for converting unibyte text to multibyte,\n\ |
1955 | and for inserting character codes specified by number.\n\n\ | |
3e8ceaac RS |
1956 | This serves to convert a Latin-1 or similar 8-bit character code\n\ |
1957 | to the corresponding Emacs multibyte character code.\n\ | |
1958 | Typically the value should be (- (make-char CHARSET 0) 128),\n\ | |
1959 | for your choice of character set.\n\ | |
537efd8d | 1960 | If `nonascii-translation-table' is non-nil, it overrides this variable."); |
35e623fb | 1961 | nonascii_insert_offset = 0; |
b0e3cf2b | 1962 | |
b4e9dd77 | 1963 | DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table, |
537efd8d | 1964 | "Translation table to convert non-ASCII unibyte codes to multibyte.\n\ |
4cf9710d RS |
1965 | This is used for converting unibyte text to multibyte,\n\ |
1966 | and for inserting character codes specified by number.\n\n\ | |
1967 | Conversion is performed only when multibyte characters are enabled,\n\ | |
1968 | and it serves to convert a Latin-1 or similar 8-bit character code\n\ | |
1969 | to the corresponding Emacs character code.\n\n\ | |
da4d65af | 1970 | If this is nil, `nonascii-insert-offset' is used instead.\n\ |
b4e9dd77 KH |
1971 | See also the docstring of `make-translation-table'."); |
1972 | Vnonascii_translation_table = Qnil; | |
4cf9710d | 1973 | |
b0e3cf2b KH |
1974 | DEFVAR_INT ("min-composite-char", &min_composite_char, |
1975 | "Minimum character code of a composite character."); | |
1976 | min_composite_char = MIN_CHAR_COMPOSITION; | |
c1a08b4c KH |
1977 | |
1978 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, | |
1979 | "A char-table for characters which invoke auto-filling.\n\ | |
1980 | Such characters has value t in this table."); | |
1981 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); | |
60022cb7 AS |
1982 | CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt); |
1983 | CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt); | |
4ed46869 KH |
1984 | } |
1985 | ||
1986 | #endif /* emacs */ |