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. |
12bcae05 | 4 | Copyright (C) 2001 Free Software Foundation, Inc. |
4ed46869 | 5 | |
369314dc KH |
6 | This file is part of GNU Emacs. |
7 | ||
8 | GNU Emacs is free software; you can redistribute it and/or modify | |
9 | it under the terms of the GNU General Public License as published by | |
10 | the Free Software Foundation; either version 2, or (at your option) | |
11 | any later version. | |
4ed46869 | 12 | |
369314dc KH |
13 | GNU Emacs is distributed in the hope that it will be useful, |
14 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
15 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
16 | GNU General Public License for more details. | |
4ed46869 | 17 | |
369314dc KH |
18 | You should have received a copy of the GNU General Public License |
19 | along with GNU Emacs; see the file COPYING. If not, write to | |
20 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
21 | Boston, MA 02111-1307, USA. */ | |
4ed46869 KH |
22 | |
23 | /* At first, see the document in `charset.h' to understand the code in | |
24 | this file. */ | |
25 | ||
68c45bf0 PE |
26 | #ifdef emacs |
27 | #include <config.h> | |
28 | #endif | |
29 | ||
4ed46869 KH |
30 | #include <stdio.h> |
31 | ||
32 | #ifdef emacs | |
33 | ||
34 | #include <sys/types.h> | |
4ed46869 KH |
35 | #include "lisp.h" |
36 | #include "buffer.h" | |
37 | #include "charset.h" | |
3f62427c | 38 | #include "composite.h" |
4ed46869 | 39 | #include "coding.h" |
fc6b09bf | 40 | #include "disptab.h" |
4ed46869 KH |
41 | |
42 | #else /* not emacs */ | |
43 | ||
44 | #include "mulelib.h" | |
45 | ||
46 | #endif /* emacs */ | |
47 | ||
2e344af3 | 48 | Lisp_Object Qcharset, Qascii, Qeight_bit_control, Qeight_bit_graphic; |
0282eb69 | 49 | Lisp_Object Qunknown; |
4ed46869 KH |
50 | |
51 | /* Declaration of special leading-codes. */ | |
4ed46869 KH |
52 | int leading_code_private_11; /* for private DIMENSION1 of 1-column */ |
53 | int leading_code_private_12; /* for private DIMENSION1 of 2-column */ | |
54 | int leading_code_private_21; /* for private DIMENSION2 of 1-column */ | |
55 | int leading_code_private_22; /* for private DIMENSION2 of 2-column */ | |
56 | ||
2e344af3 KH |
57 | /* Declaration of special charsets. The values are set by |
58 | Fsetup_special_charsets. */ | |
4ed46869 KH |
59 | int charset_latin_iso8859_1; /* ISO8859-1 (Latin-1) */ |
60 | int charset_jisx0208_1978; /* JISX0208.1978 (Japanese Kanji old set) */ | |
61 | int charset_jisx0208; /* JISX0208.1983 (Japanese Kanji) */ | |
62 | int charset_katakana_jisx0201; /* JISX0201.Kana (Japanese Katakana) */ | |
63 | int charset_latin_jisx0201; /* JISX0201.Roman (Japanese Roman) */ | |
64 | int charset_big5_1; /* Big5 Level 1 (Chinese Traditional) */ | |
65 | int charset_big5_2; /* Big5 Level 2 (Chinese Traditional) */ | |
66 | ||
67 | Lisp_Object Qcharset_table; | |
68 | ||
69 | /* A char-table containing information of each character set. */ | |
70 | Lisp_Object Vcharset_table; | |
71 | ||
72 | /* A vector of charset symbol indexed by charset-id. This is used | |
73 | only for returning charset symbol from C functions. */ | |
74 | Lisp_Object Vcharset_symbol_table; | |
75 | ||
76 | /* A list of charset symbols ever defined. */ | |
77 | Lisp_Object Vcharset_list; | |
78 | ||
537efd8d KH |
79 | /* Vector of translation table ever defined. |
80 | ID of a translation table is used to index this vector. */ | |
81 | Lisp_Object Vtranslation_table_vector; | |
b0e3cf2b | 82 | |
c1a08b4c KH |
83 | /* A char-table for characters which may invoke auto-filling. */ |
84 | Lisp_Object Vauto_fill_chars; | |
85 | ||
86 | Lisp_Object Qauto_fill_chars; | |
87 | ||
4ed46869 KH |
88 | /* Tables used by macros BYTES_BY_CHAR_HEAD and WIDTH_BY_CHAR_HEAD. */ |
89 | int bytes_by_char_head[256]; | |
90 | int width_by_char_head[256]; | |
91 | ||
92 | /* Mapping table from ISO2022's charset (specified by DIMENSION, | |
93 | CHARS, and FINAL-CHAR) to Emacs' charset. */ | |
94 | int iso_charset_table[2][2][128]; | |
95 | ||
96 | /* Variables used locally in the macro FETCH_MULTIBYTE_CHAR. */ | |
97 | unsigned char *_fetch_multibyte_char_p; | |
98 | int _fetch_multibyte_char_len; | |
99 | ||
35e623fb RS |
100 | /* Offset to add to a non-ASCII value when inserting it. */ |
101 | int nonascii_insert_offset; | |
102 | ||
4cf9710d RS |
103 | /* Translation table for converting non-ASCII unibyte characters |
104 | to multibyte codes, or nil. */ | |
b4e9dd77 | 105 | Lisp_Object Vnonascii_translation_table; |
4cf9710d | 106 | |
8a73a704 KH |
107 | /* List of all possible generic characters. */ |
108 | Lisp_Object Vgeneric_character_list; | |
109 | ||
046b1f03 | 110 | \f |
93bcb785 KH |
111 | void |
112 | invalid_character (c) | |
113 | int c; | |
114 | { | |
ba7434e5 | 115 | error ("Invalid character: 0%o, %d, 0x%x", c, c, c); |
93bcb785 KH |
116 | } |
117 | ||
2e344af3 KH |
118 | /* Parse string STR of length LENGTH and fetch information of a |
119 | character at STR. Set BYTES to the byte length the character | |
120 | occupies, CHARSET, C1, C2 to proper values of the character. */ | |
121 | ||
122 | #define SPLIT_MULTIBYTE_SEQ(str, length, bytes, charset, c1, c2) \ | |
123 | do { \ | |
124 | (c1) = *(str); \ | |
125 | (bytes) = BYTES_BY_CHAR_HEAD (c1); \ | |
126 | if ((bytes) == 1) \ | |
127 | (charset) = ASCII_BYTE_P (c1) ? CHARSET_ASCII : CHARSET_8_BIT_GRAPHIC; \ | |
128 | else if ((bytes) == 2) \ | |
129 | { \ | |
130 | if ((c1) == LEADING_CODE_8_BIT_CONTROL) \ | |
131 | (charset) = CHARSET_8_BIT_CONTROL, (c1) = (str)[1] - 0x20; \ | |
132 | else \ | |
133 | (charset) = (c1), (c1) = (str)[1] & 0x7F; \ | |
134 | } \ | |
135 | else if ((bytes) == 3) \ | |
136 | { \ | |
137 | if ((c1) < LEADING_CODE_PRIVATE_11) \ | |
138 | (charset) = (c1), (c1) = (str)[1] & 0x7F, (c2) = (str)[2] & 0x7F; \ | |
139 | else \ | |
140 | (charset) = (str)[1], (c1) = (str)[2] & 0x7F; \ | |
141 | } \ | |
142 | else \ | |
143 | (charset) = (str)[1], (c1) = (str)[2] & 0x7F, (c2) = (str)[3] & 0x7F; \ | |
ac4137cc KH |
144 | } while (0) |
145 | ||
146 | /* 1 if CHARSET, C1, and C2 compose a valid character, else 0. */ | |
44c6492d | 147 | #define CHAR_COMPONENTS_VALID_P(charset, c1, c2) \ |
2e344af3 | 148 | ((charset) == CHARSET_ASCII \ |
63f4d579 | 149 | ? ((c1) >= 0 && (c1) <= 0x7F) \ |
2e344af3 KH |
150 | : ((charset) == CHARSET_8_BIT_CONTROL \ |
151 | ? ((c1) >= 0x80 && (c1) <= 0x9F) \ | |
152 | : ((charset) == CHARSET_8_BIT_GRAPHIC \ | |
153 | ? ((c1) >= 0x80 && (c1) <= 0xFF) \ | |
154 | : (CHARSET_DIMENSION (charset) == 1 \ | |
155 | ? ((c1) >= 0x20 && (c1) <= 0x7F) \ | |
156 | : ((c1) >= 0x20 && (c1) <= 0x7F \ | |
157 | && (c2) >= 0x20 && (c2) <= 0x7F))))) | |
93bcb785 | 158 | |
99529c2c KH |
159 | /* Store multi-byte form of the character C in STR. The caller should |
160 | allocate at least 4-byte area at STR in advance. Returns the | |
161 | length of the multi-byte form. If C is an invalid character code, | |
12bcae05 | 162 | return -1. */ |
4ed46869 KH |
163 | |
164 | int | |
12bcae05 | 165 | char_to_string_1 (c, str) |
4ed46869 | 166 | int c; |
99529c2c | 167 | unsigned char *str; |
4ed46869 | 168 | { |
99529c2c KH |
169 | unsigned char *p = str; |
170 | ||
6662e69b | 171 | if (c & CHAR_MODIFIER_MASK) /* This includes the case C is negative. */ |
8ac5a9cc | 172 | { |
6662e69b KH |
173 | /* Multibyte character can't have a modifier bit. */ |
174 | if (! SINGLE_BYTE_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) | |
12bcae05 | 175 | return -1; |
6662e69b KH |
176 | |
177 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
8ac5a9cc | 178 | if (c & CHAR_META) |
6662e69b KH |
179 | { |
180 | /* Move the meta bit to the right place for a string. */ | |
181 | c = (c & ~CHAR_META) | 0x80; | |
182 | } | |
183 | if (c & CHAR_SHIFT) | |
184 | { | |
185 | /* Shift modifier is valid only with [A-Za-z]. */ | |
186 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
187 | c &= ~CHAR_SHIFT; | |
188 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
189 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
190 | } | |
8ac5a9cc | 191 | if (c & CHAR_CTL) |
6662e69b KH |
192 | { |
193 | /* Simulate the code in lread.c. */ | |
194 | /* Allow `\C- ' and `\C-?'. */ | |
195 | if (c == (CHAR_CTL | ' ')) | |
196 | c = 0; | |
197 | else if (c == (CHAR_CTL | '?')) | |
198 | c = 127; | |
199 | /* ASCII control chars are made from letters (both cases), | |
200 | as well as the non-letters within 0100...0137. */ | |
201 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
202 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
203 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
204 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
205 | } | |
206 | ||
5ef834ea KH |
207 | /* If C still has any modifier bits, just ignore it. */ |
208 | c &= ~CHAR_MODIFIER_MASK; | |
2e344af3 | 209 | } |
12bcae05 | 210 | |
2e344af3 KH |
211 | if (SINGLE_BYTE_CHAR_P (c)) |
212 | { | |
213 | if (ASCII_BYTE_P (c) || c >= 0xA0) | |
214 | *p++ = c; | |
215 | else | |
216 | { | |
217 | *p++ = LEADING_CODE_8_BIT_CONTROL; | |
218 | *p++ = c + 0x20; | |
219 | } | |
8ac5a9cc | 220 | } |
6ce974d4 | 221 | else if (CHAR_VALID_P (c, 0)) |
4ed46869 | 222 | { |
ac4137cc | 223 | int charset, c1, c2; |
4ed46869 | 224 | |
2e344af3 | 225 | SPLIT_CHAR (c, charset, c1, c2); |
99529c2c KH |
226 | |
227 | if (charset >= LEADING_CODE_EXT_11) | |
228 | *p++ = (charset < LEADING_CODE_EXT_12 | |
229 | ? LEADING_CODE_PRIVATE_11 | |
230 | : (charset < LEADING_CODE_EXT_21 | |
231 | ? LEADING_CODE_PRIVATE_12 | |
232 | : (charset < LEADING_CODE_EXT_22 | |
233 | ? LEADING_CODE_PRIVATE_21 | |
234 | : LEADING_CODE_PRIVATE_22))); | |
235 | *p++ = charset; | |
236 | if (c1 > 0 && c1 < 32 || c2 > 0 && c2 < 32) | |
12bcae05 | 237 | return -1; |
99529c2c | 238 | if (c1) |
4ed46869 | 239 | { |
99529c2c KH |
240 | *p++ = c1 | 0x80; |
241 | if (c2 > 0) | |
242 | *p++ = c2 | 0x80; | |
4ed46869 KH |
243 | } |
244 | } | |
2e344af3 | 245 | else |
12bcae05 | 246 | return -1; |
4ed46869 | 247 | |
2e344af3 | 248 | return (p - str); |
4ed46869 KH |
249 | } |
250 | ||
12bcae05 GM |
251 | |
252 | /* Store multi-byte form of the character C in STR. The caller should | |
253 | allocate at least 4-byte area at STR in advance. Returns the | |
254 | length of the multi-byte form. If C is an invalid character code, | |
255 | signal an error. | |
256 | ||
257 | Use macro `CHAR_STRING (C, STR)' instead of calling this function | |
258 | directly if C can be an ASCII character. */ | |
259 | ||
260 | int | |
261 | char_to_string (c, str) | |
262 | int c; | |
263 | unsigned char *str; | |
264 | { | |
265 | int len; | |
266 | len = char_to_string_1 (c, str); | |
267 | if (len == -1) | |
268 | invalid_character (c); | |
269 | return len; | |
270 | } | |
271 | ||
272 | ||
44c6492d KH |
273 | /* Return the non-ASCII character corresponding to multi-byte form at |
274 | STR of length LEN. If ACTUAL_LEN is not NULL, store the byte | |
275 | length of the multibyte form in *ACTUAL_LEN. | |
537efd8d | 276 | |
99529c2c KH |
277 | Use macros STRING_CHAR or STRING_CHAR_AND_LENGTH instead of calling |
278 | this function directly if you want ot handle ASCII characters as | |
279 | well. */ | |
4ed46869 | 280 | |
dfcf069d | 281 | int |
99529c2c | 282 | string_to_char (str, len, actual_len) |
8867de67 | 283 | const unsigned char *str; |
ac4137cc | 284 | int len, *actual_len; |
4ed46869 | 285 | { |
ac4137cc | 286 | int c, bytes, charset, c1, c2; |
4ed46869 | 287 | |
ac4137cc KH |
288 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, charset, c1, c2); |
289 | c = MAKE_CHAR (charset, c1, c2); | |
4ed46869 | 290 | if (actual_len) |
ac4137cc | 291 | *actual_len = bytes; |
4ed46869 KH |
292 | return c; |
293 | } | |
294 | ||
44c6492d KH |
295 | /* Return the length of the multi-byte form at string STR of length LEN. |
296 | Use the macro MULTIBYTE_FORM_LENGTH instead. */ | |
4ed46869 KH |
297 | int |
298 | multibyte_form_length (str, len) | |
8867de67 | 299 | const unsigned char *str; |
4ed46869 KH |
300 | int len; |
301 | { | |
ac4137cc | 302 | int bytes; |
4ed46869 | 303 | |
ac4137cc | 304 | PARSE_MULTIBYTE_SEQ (str, len, bytes); |
90d7b74e | 305 | return bytes; |
4ed46869 KH |
306 | } |
307 | ||
ac4137cc KH |
308 | /* Check multibyte form at string STR of length LEN and set variables |
309 | pointed by CHARSET, C1, and C2 to charset and position codes of the | |
310 | character at STR, and return 0. If there's no multibyte character, | |
4ed46869 KH |
311 | return -1. This should be used only in the macro SPLIT_STRING |
312 | which checks range of STR in advance. */ | |
313 | ||
dfcf069d | 314 | int |
99529c2c | 315 | split_string (str, len, charset, c1, c2) |
ac4137cc KH |
316 | const unsigned char *str; |
317 | unsigned char *c1, *c2; | |
318 | int len, *charset; | |
4ed46869 | 319 | { |
ac4137cc | 320 | register int bytes, cs, code1, code2 = -1; |
4ed46869 | 321 | |
ac4137cc KH |
322 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, cs, code1, code2); |
323 | if (cs == CHARSET_ASCII) | |
4ed46869 | 324 | return -1; |
ac4137cc KH |
325 | *charset = cs; |
326 | *c1 = code1; | |
327 | *c2 = code2; | |
5865af0d | 328 | return 0; |
ac4137cc KH |
329 | } |
330 | ||
44c6492d KH |
331 | /* Return 1 iff character C has valid printable glyph. |
332 | Use the macro CHAR_PRINTABLE_P instead. */ | |
ac4137cc KH |
333 | int |
334 | char_printable_p (c) | |
335 | int c; | |
336 | { | |
8ebae00c | 337 | int charset, c1, c2; |
ac4137cc | 338 | |
2e344af3 | 339 | if (ASCII_BYTE_P (c)) |
ac4137cc | 340 | return 1; |
2e344af3 KH |
341 | else if (SINGLE_BYTE_CHAR_P (c)) |
342 | return 0; | |
343 | else if (c >= MAX_CHAR) | |
99529c2c | 344 | return 0; |
ac4137cc | 345 | |
2e344af3 | 346 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
347 | if (! CHARSET_DEFINED_P (charset)) |
348 | return 0; | |
349 | if (CHARSET_CHARS (charset) == 94 | |
350 | ? c1 <= 32 || c1 >= 127 | |
351 | : c1 < 32) | |
352 | return 0; | |
353 | if (CHARSET_DIMENSION (charset) == 2 | |
354 | && (CHARSET_CHARS (charset) == 94 | |
355 | ? c2 <= 32 || c2 >= 127 | |
356 | : c2 < 32)) | |
357 | return 0; | |
358 | return 1; | |
4ed46869 KH |
359 | } |
360 | ||
537efd8d | 361 | /* Translate character C by translation table TABLE. If C |
b4e9dd77 KH |
362 | is negative, translate a character specified by CHARSET, C1, and C2 |
363 | (C1 and C2 are code points of the character). If no translation is | |
364 | found in TABLE, return C. */ | |
dfcf069d | 365 | int |
b4e9dd77 | 366 | translate_char (table, c, charset, c1, c2) |
23d2a7f1 KH |
367 | Lisp_Object table; |
368 | int c, charset, c1, c2; | |
369 | { | |
370 | Lisp_Object ch; | |
371 | int alt_charset, alt_c1, alt_c2, dimension; | |
372 | ||
0ad3f83d | 373 | if (c < 0) c = MAKE_CHAR (charset, (c1 & 0x7F) , (c2 & 0x7F)); |
23d2a7f1 | 374 | if (!CHAR_TABLE_P (table) |
ac4137cc | 375 | || (ch = Faref (table, make_number (c)), !NATNUMP (ch))) |
23d2a7f1 KH |
376 | return c; |
377 | ||
378 | SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); | |
379 | dimension = CHARSET_DIMENSION (alt_charset); | |
380 | if (dimension == 1 && alt_c1 > 0 || dimension == 2 && alt_c2 > 0) | |
381 | /* CH is not a generic character, just return it. */ | |
382 | return XFASTINT (ch); | |
383 | ||
384 | /* Since CH is a generic character, we must return a specific | |
385 | charater which has the same position codes as C from CH. */ | |
386 | if (charset < 0) | |
387 | SPLIT_CHAR (c, charset, c1, c2); | |
388 | if (dimension != CHARSET_DIMENSION (charset)) | |
389 | /* We can't make such a character because of dimension mismatch. */ | |
390 | return c; | |
23d2a7f1 KH |
391 | return MAKE_CHAR (alt_charset, c1, c2); |
392 | } | |
393 | ||
d2665018 | 394 | /* Convert the unibyte character C to multibyte based on |
b4e9dd77 | 395 | Vnonascii_translation_table or nonascii_insert_offset. If they can't |
d2665018 KH |
396 | convert C to a valid multibyte character, convert it based on |
397 | DEFAULT_NONASCII_INSERT_OFFSET which makes C a Latin-1 character. */ | |
35e623fb | 398 | |
dfcf069d | 399 | int |
35e623fb RS |
400 | unibyte_char_to_multibyte (c) |
401 | int c; | |
402 | { | |
543b4f61 | 403 | if (c < 0400 && c >= 0200) |
35e623fb | 404 | { |
d2665018 KH |
405 | int c_save = c; |
406 | ||
b4e9dd77 | 407 | if (! NILP (Vnonascii_translation_table)) |
bbf12bb3 KH |
408 | { |
409 | c = XINT (Faref (Vnonascii_translation_table, make_number (c))); | |
ac4137cc | 410 | if (c >= 0400 && ! char_valid_p (c, 0)) |
bbf12bb3 KH |
411 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
412 | } | |
413 | else if (c >= 0240 && nonascii_insert_offset > 0) | |
414 | { | |
415 | c += nonascii_insert_offset; | |
ac4137cc | 416 | if (c < 0400 || ! char_valid_p (c, 0)) |
bbf12bb3 KH |
417 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
418 | } | |
419 | else if (c >= 0240) | |
d2665018 | 420 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
35e623fb RS |
421 | } |
422 | return c; | |
423 | } | |
76d7b829 KH |
424 | |
425 | ||
426 | /* Convert the multibyte character C to unibyte 8-bit character based | |
427 | on Vnonascii_translation_table or nonascii_insert_offset. If | |
428 | REV_TBL is non-nil, it should be a reverse table of | |
429 | Vnonascii_translation_table, i.e. what given by: | |
430 | Fchar_table_extra_slot (Vnonascii_translation_table, make_number (0)) */ | |
431 | ||
432 | int | |
433 | multibyte_char_to_unibyte (c, rev_tbl) | |
434 | int c; | |
435 | Lisp_Object rev_tbl; | |
436 | { | |
437 | if (!SINGLE_BYTE_CHAR_P (c)) | |
438 | { | |
439 | int c_save = c; | |
440 | ||
441 | if (! CHAR_TABLE_P (rev_tbl) | |
442 | && CHAR_TABLE_P (Vnonascii_translation_table)) | |
443 | rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table, | |
444 | make_number (0)); | |
445 | if (CHAR_TABLE_P (rev_tbl)) | |
446 | { | |
447 | Lisp_Object temp; | |
448 | temp = Faref (rev_tbl, make_number (c)); | |
449 | if (INTEGERP (temp)) | |
450 | c = XINT (temp); | |
bbf12bb3 KH |
451 | if (c >= 256) |
452 | c = (c_save & 0177) + 0200; | |
453 | } | |
454 | else | |
455 | { | |
456 | if (nonascii_insert_offset > 0) | |
457 | c -= nonascii_insert_offset; | |
458 | if (c < 128 || c >= 256) | |
459 | c = (c_save & 0177) + 0200; | |
76d7b829 | 460 | } |
76d7b829 KH |
461 | } |
462 | ||
463 | return c; | |
464 | } | |
465 | ||
35e623fb | 466 | \f |
4ed46869 KH |
467 | /* Update the table Vcharset_table with the given arguments (see the |
468 | document of `define-charset' for the meaning of each argument). | |
469 | Several other table contents are also updated. The caller should | |
470 | check the validity of CHARSET-ID and the remaining arguments in | |
471 | advance. */ | |
472 | ||
473 | void | |
474 | update_charset_table (charset_id, dimension, chars, width, direction, | |
475 | iso_final_char, iso_graphic_plane, | |
476 | short_name, long_name, description) | |
477 | Lisp_Object charset_id, dimension, chars, width, direction; | |
478 | Lisp_Object iso_final_char, iso_graphic_plane; | |
479 | Lisp_Object short_name, long_name, description; | |
480 | { | |
481 | int charset = XINT (charset_id); | |
482 | int bytes; | |
483 | unsigned char leading_code_base, leading_code_ext; | |
484 | ||
6dc0722d KH |
485 | if (NILP (CHARSET_TABLE_ENTRY (charset))) |
486 | CHARSET_TABLE_ENTRY (charset) | |
487 | = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); | |
4ed46869 | 488 | |
d78bc582 KH |
489 | if (NILP (long_name)) |
490 | long_name = short_name; | |
491 | if (NILP (description)) | |
492 | description = long_name; | |
493 | ||
4ed46869 KH |
494 | /* Get byte length of multibyte form, base leading-code, and |
495 | extended leading-code of the charset. See the comment under the | |
496 | title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
497 | bytes = XINT (dimension); | |
498 | if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
499 | { | |
500 | /* Official charset, it doesn't have an extended leading-code. */ | |
2e344af3 | 501 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC) |
4ed46869 KH |
502 | bytes += 1; /* For a base leading-code. */ |
503 | leading_code_base = charset; | |
504 | leading_code_ext = 0; | |
505 | } | |
506 | else | |
507 | { | |
508 | /* Private charset. */ | |
509 | bytes += 2; /* For base and extended leading-codes. */ | |
510 | leading_code_base | |
511 | = (charset < LEADING_CODE_EXT_12 | |
512 | ? LEADING_CODE_PRIVATE_11 | |
513 | : (charset < LEADING_CODE_EXT_21 | |
514 | ? LEADING_CODE_PRIVATE_12 | |
515 | : (charset < LEADING_CODE_EXT_22 | |
516 | ? LEADING_CODE_PRIVATE_21 | |
517 | : LEADING_CODE_PRIVATE_22))); | |
518 | leading_code_ext = charset; | |
c83ef371 KH |
519 | if (BYTES_BY_CHAR_HEAD (leading_code_base) != bytes) |
520 | error ("Invalid dimension for the charset-ID %d", charset); | |
521 | } | |
6ef23ebb | 522 | |
4ed46869 KH |
523 | CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; |
524 | CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
525 | CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
526 | CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
527 | CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
528 | CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
529 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
530 | = make_number (leading_code_base); | |
531 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
532 | = make_number (leading_code_ext); | |
533 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
534 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
535 | = iso_graphic_plane; | |
536 | CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
537 | CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
538 | CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
539 | CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
540 | ||
541 | { | |
542 | /* If we have already defined a charset which has the same | |
543 | DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
544 | DIRECTION, we must update the entry REVERSE-CHARSET of both | |
545 | charsets. If there's no such charset, the value of the entry | |
546 | is set to nil. */ | |
547 | int i; | |
548 | ||
513ee442 | 549 | for (i = 0; i <= MAX_CHARSET; i++) |
4ed46869 KH |
550 | if (!NILP (CHARSET_TABLE_ENTRY (i))) |
551 | { | |
552 | if (CHARSET_DIMENSION (i) == XINT (dimension) | |
553 | && CHARSET_CHARS (i) == XINT (chars) | |
554 | && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
555 | && CHARSET_DIRECTION (i) != XINT (direction)) | |
556 | { | |
557 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
558 | = make_number (i); | |
559 | CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
560 | break; | |
561 | } | |
562 | } | |
513ee442 | 563 | if (i > MAX_CHARSET) |
4ed46869 KH |
564 | /* No such a charset. */ |
565 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
566 | = make_number (-1); | |
567 | } | |
568 | ||
c83ef371 | 569 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC |
4ed46869 KH |
570 | && charset < MIN_CHARSET_PRIVATE_DIMENSION1) |
571 | { | |
c83ef371 | 572 | bytes_by_char_head[leading_code_base] = bytes; |
4ed46869 KH |
573 | width_by_char_head[leading_code_base] = XINT (width); |
574 | ||
575 | /* Update table emacs_code_class. */ | |
576 | emacs_code_class[charset] = (bytes == 2 | |
577 | ? EMACS_leading_code_2 | |
578 | : (bytes == 3 | |
579 | ? EMACS_leading_code_3 | |
580 | : EMACS_leading_code_4)); | |
581 | } | |
582 | ||
583 | /* Update table iso_charset_table. */ | |
52e386c2 | 584 | if (XINT (iso_final_char) >= 0 |
2e344af3 | 585 | && ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0) |
4ed46869 KH |
586 | ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset; |
587 | } | |
588 | ||
589 | #ifdef emacs | |
590 | ||
591 | /* Return charset id of CHARSET_SYMBOL, or return -1 if CHARSET_SYMBOL | |
592 | is invalid. */ | |
593 | int | |
594 | get_charset_id (charset_symbol) | |
595 | Lisp_Object charset_symbol; | |
596 | { | |
597 | Lisp_Object val; | |
598 | int charset; | |
599 | ||
600 | return ((SYMBOLP (charset_symbol) | |
601 | && (val = Fget (charset_symbol, Qcharset), VECTORP (val)) | |
602 | && (charset = XINT (XVECTOR (val)->contents[CHARSET_ID_IDX]), | |
603 | CHARSET_VALID_P (charset))) | |
604 | ? charset : -1); | |
605 | } | |
606 | ||
607 | /* Return an identification number for a new private charset of | |
608 | DIMENSION and WIDTH. If there's no more room for the new charset, | |
609 | return 0. */ | |
610 | Lisp_Object | |
611 | get_new_private_charset_id (dimension, width) | |
612 | int dimension, width; | |
613 | { | |
614 | int charset, from, to; | |
615 | ||
616 | if (dimension == 1) | |
617 | { | |
3bb7b08b KH |
618 | from = LEADING_CODE_EXT_11; |
619 | to = LEADING_CODE_EXT_21; | |
4ed46869 KH |
620 | } |
621 | else | |
622 | { | |
3bb7b08b KH |
623 | from = LEADING_CODE_EXT_21; |
624 | to = LEADING_CODE_EXT_MAX + 1; | |
4ed46869 KH |
625 | } |
626 | ||
627 | for (charset = from; charset < to; charset++) | |
628 | if (!CHARSET_DEFINED_P (charset)) break; | |
629 | ||
630 | return make_number (charset < to ? charset : 0); | |
631 | } | |
632 | ||
633 | DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
fdb82f93 PJ |
634 | doc: /* Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR. |
635 | If CHARSET-ID is nil, it is decided automatically, which means CHARSET is | |
636 | treated as a private charset. | |
637 | INFO-VECTOR is a vector of the format: | |
638 | [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE | |
639 | SHORT-NAME LONG-NAME DESCRIPTION] | |
640 | The meanings of each elements is as follows: | |
641 | DIMENSION (integer) is the number of bytes to represent a character: 1 or 2. | |
642 | CHARS (integer) is the number of characters in a dimension: 94 or 96. | |
643 | WIDTH (integer) is the number of columns a character in the charset | |
644 | occupies on the screen: one of 0, 1, and 2. | |
645 | ||
646 | DIRECTION (integer) is the rendering direction of characters in the | |
647 | charset when rendering. If 0, render from left to right, else | |
648 | render from right to left. | |
649 | ||
650 | ISO-FINAL-CHAR (character) is the final character of the | |
651 | corresponding ISO 2022 charset. | |
652 | It may be -1 if the charset is internal use only. | |
653 | ||
654 | ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked | |
655 | while encoding to variants of ISO 2022 coding system, one of the | |
656 | following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR). | |
657 | It may be -1 if the charset is internal use only. | |
658 | ||
659 | SHORT-NAME (string) is the short name to refer to the charset. | |
660 | ||
661 | LONG-NAME (string) is the long name to refer to the charset. | |
662 | ||
663 | DESCRIPTION (string) is the description string of the charset. */) | |
664 | (charset_id, charset_symbol, info_vector) | |
4ed46869 KH |
665 | Lisp_Object charset_id, charset_symbol, info_vector; |
666 | { | |
667 | Lisp_Object *vec; | |
668 | ||
669 | if (!NILP (charset_id)) | |
670 | CHECK_NUMBER (charset_id, 0); | |
671 | CHECK_SYMBOL (charset_symbol, 1); | |
672 | CHECK_VECTOR (info_vector, 2); | |
673 | ||
674 | if (! NILP (charset_id)) | |
675 | { | |
676 | if (! CHARSET_VALID_P (XINT (charset_id))) | |
677 | error ("Invalid CHARSET: %d", XINT (charset_id)); | |
678 | else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
679 | error ("Already defined charset: %d", XINT (charset_id)); | |
680 | } | |
681 | ||
682 | vec = XVECTOR (info_vector)->contents; | |
683 | if (XVECTOR (info_vector)->size != 9 | |
684 | || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
685 | || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
686 | || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
687 | || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
2e344af3 KH |
688 | || !INTEGERP (vec[4]) |
689 | || !(XINT (vec[4]) == -1 || XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~') | |
690 | || !INTEGERP (vec[5]) | |
691 | || !(XINT (vec[5]) == -1 || XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
4ed46869 KH |
692 | || !STRINGP (vec[6]) |
693 | || !STRINGP (vec[7]) | |
694 | || !STRINGP (vec[8])) | |
695 | error ("Invalid info-vector argument for defining charset %s", | |
696 | XSYMBOL (charset_symbol)->name->data); | |
697 | ||
698 | if (NILP (charset_id)) | |
699 | { | |
700 | charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
701 | if (XINT (charset_id) == 0) | |
702 | error ("There's no room for a new private charset %s", | |
703 | XSYMBOL (charset_symbol)->name->data); | |
704 | } | |
705 | ||
706 | update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
707 | vec[4], vec[5], vec[6], vec[7], vec[8]); | |
6dc0722d | 708 | Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
4ed46869 KH |
709 | CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
710 | Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
711 | return Qnil; | |
712 | } | |
713 | ||
8a73a704 KH |
714 | DEFUN ("generic-character-list", Fgeneric_character_list, |
715 | Sgeneric_character_list, 0, 0, 0, | |
fdb82f93 PJ |
716 | doc: /* Return a list of all possible generic characters. |
717 | It includes a generic character for a charset not yet defined. */) | |
718 | () | |
8a73a704 KH |
719 | { |
720 | return Vgeneric_character_list; | |
721 | } | |
722 | ||
3fac5a51 KH |
723 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
724 | Sget_unused_iso_final_char, 2, 2, 0, | |
fdb82f93 PJ |
725 | doc: /* Return an unsed ISO's final char for a charset of DIMENISION and CHARS. |
726 | DIMENSION is the number of bytes to represent a character: 1 or 2. | |
727 | CHARS is the number of characters in a dimension: 94 or 96. | |
728 | ||
729 | This final char is for private use, thus the range is `0' (48) .. `?' (63). | |
730 | If there's no unused final char for the specified kind of charset, | |
731 | return nil. */) | |
732 | (dimension, chars) | |
3fac5a51 KH |
733 | Lisp_Object dimension, chars; |
734 | { | |
735 | int final_char; | |
736 | ||
737 | CHECK_NUMBER (dimension, 0); | |
738 | CHECK_NUMBER (chars, 1); | |
739 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
740 | error ("Invalid charset dimension %d, it should be 1 or 2", | |
741 | XINT (dimension)); | |
742 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
743 | error ("Invalid charset chars %d, it should be 94 or 96", | |
744 | XINT (chars)); | |
745 | for (final_char = '0'; final_char <= '?'; final_char++) | |
746 | { | |
747 | if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0) | |
748 | break; | |
749 | } | |
750 | return (final_char <= '?' ? make_number (final_char) : Qnil); | |
751 | } | |
752 | ||
4ed46869 KH |
753 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
754 | 4, 4, 0, | |
fdb82f93 PJ |
755 | doc: /* Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET. |
756 | CHARSET should be defined by `defined-charset' in advance. */) | |
757 | (dimension, chars, final_char, charset_symbol) | |
4ed46869 KH |
758 | Lisp_Object dimension, chars, final_char, charset_symbol; |
759 | { | |
760 | int charset; | |
761 | ||
762 | CHECK_NUMBER (dimension, 0); | |
763 | CHECK_NUMBER (chars, 1); | |
764 | CHECK_NUMBER (final_char, 2); | |
765 | CHECK_SYMBOL (charset_symbol, 3); | |
766 | ||
767 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
768 | error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
769 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
770 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
771 | if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
772 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
773 | if ((charset = get_charset_id (charset_symbol)) < 0) | |
774 | error ("Invalid charset %s", XSYMBOL (charset_symbol)->name->data); | |
775 | ||
776 | ISO_CHARSET_TABLE (dimension, chars, final_char) = charset; | |
777 | return Qnil; | |
778 | } | |
779 | ||
2e344af3 KH |
780 | /* Return information about charsets in the text at PTR of NBYTES |
781 | bytes, which are NCHARS characters. The value is: | |
f6302ac9 | 782 | |
cfe34140 | 783 | 0: Each character is represented by one byte. This is always |
f6302ac9 | 784 | true for unibyte text. |
2e344af3 KH |
785 | 1: No charsets other than ascii eight-bit-control, |
786 | eight-bit-graphic, and latin-1 are found. | |
787 | 2: Otherwise. | |
1d67c29b | 788 | |
2e344af3 KH |
789 | In addition, if CHARSETS is nonzero, for each found charset N, set |
790 | CHARSETS[N] to 1. For that, callers should allocate CHARSETS | |
791 | (MAX_CHARSET + 1 elements) in advance. It may lookup a translation | |
792 | table TABLE if supplied. For invalid charsets, set CHARSETS[1] to | |
793 | 1 (note that there's no charset whose ID is 1). */ | |
4ed46869 KH |
794 | |
795 | int | |
2e344af3 KH |
796 | find_charset_in_text (ptr, nchars, nbytes, charsets, table) |
797 | unsigned char *ptr; | |
798 | int nchars, nbytes, *charsets; | |
23d2a7f1 | 799 | Lisp_Object table; |
4ed46869 | 800 | { |
2e344af3 | 801 | if (nchars == nbytes) |
0282eb69 | 802 | { |
2e344af3 | 803 | if (charsets && nbytes > 0) |
0282eb69 | 804 | { |
2e344af3 KH |
805 | unsigned char *endp = ptr + nbytes; |
806 | int maskbits = 0; | |
807 | ||
808 | while (ptr < endp && maskbits != 7) | |
809 | { | |
810 | maskbits |= (*ptr < 0x80 ? 1 : *ptr < 0xA0 ? 2 : 4); | |
811 | ptr++; | |
812 | } | |
813 | ||
814 | if (maskbits & 1) | |
815 | charsets[CHARSET_ASCII] = 1; | |
816 | if (maskbits & 2) | |
817 | charsets[CHARSET_8_BIT_CONTROL] = 1; | |
818 | if (maskbits & 4) | |
819 | charsets[CHARSET_8_BIT_GRAPHIC] = 1; | |
0282eb69 | 820 | } |
2e344af3 | 821 | return 0; |
0282eb69 | 822 | } |
2e344af3 | 823 | else |
4ed46869 | 824 | { |
2e344af3 | 825 | int return_val = 1; |
99529c2c | 826 | int bytes, charset, c1, c2; |
05505664 | 827 | |
2e344af3 KH |
828 | if (! CHAR_TABLE_P (table)) |
829 | table = Qnil; | |
05505664 | 830 | |
2e344af3 | 831 | while (nchars-- > 0) |
23d2a7f1 | 832 | { |
2e344af3 KH |
833 | SPLIT_MULTIBYTE_SEQ (ptr, len, bytes, charset, c1, c2); |
834 | ptr += bytes; | |
4ed46869 | 835 | |
2e344af3 KH |
836 | if (!CHARSET_DEFINED_P (charset)) |
837 | charset = 1; | |
838 | else if (! NILP (table)) | |
839 | { | |
840 | int c = translate_char (table, -1, charset, c1, c2); | |
841 | if (c >= 0) | |
842 | charset = CHAR_CHARSET (c); | |
843 | } | |
844 | ||
845 | if (return_val == 1 | |
846 | && charset != CHARSET_ASCII | |
847 | && charset != CHARSET_8_BIT_CONTROL | |
848 | && charset != CHARSET_8_BIT_GRAPHIC | |
849 | && charset != charset_latin_iso8859_1) | |
850 | return_val = 2; | |
851 | ||
852 | if (charsets) | |
853 | charsets[charset] = 1; | |
854 | else if (return_val == 2) | |
855 | break; | |
4ed46869 | 856 | } |
2e344af3 | 857 | return return_val; |
4ed46869 | 858 | } |
4ed46869 KH |
859 | } |
860 | ||
861 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 862 | 2, 3, 0, |
fdb82f93 PJ |
863 | doc: /* Return a list of charsets in the region between BEG and END. |
864 | BEG and END are buffer positions. | |
865 | Optional arg TABLE if non-nil is a translation table to look up. | |
866 | ||
867 | If the region contains invalid multibyte characters, | |
868 | `unknown' is included in the returned list. | |
869 | ||
870 | If the current buffer is unibyte, the returned list may contain | |
871 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) | |
872 | (beg, end, table) | |
23d2a7f1 | 873 | Lisp_Object beg, end, table; |
4ed46869 | 874 | { |
028d516b | 875 | int charsets[MAX_CHARSET + 1]; |
6ae1f27e | 876 | int from, from_byte, to, stop, stop_byte, i; |
4ed46869 KH |
877 | Lisp_Object val; |
878 | ||
879 | validate_region (&beg, &end); | |
880 | from = XFASTINT (beg); | |
881 | stop = to = XFASTINT (end); | |
6ae1f27e | 882 | |
4ed46869 | 883 | if (from < GPT && GPT < to) |
6ae1f27e RS |
884 | { |
885 | stop = GPT; | |
886 | stop_byte = GPT_BYTE; | |
887 | } | |
888 | else | |
889 | stop_byte = CHAR_TO_BYTE (stop); | |
890 | ||
891 | from_byte = CHAR_TO_BYTE (from); | |
892 | ||
028d516b | 893 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
4ed46869 KH |
894 | while (1) |
895 | { | |
2e344af3 KH |
896 | find_charset_in_text (BYTE_POS_ADDR (from_byte), stop - from, |
897 | stop_byte - from_byte, charsets, table); | |
4ed46869 | 898 | if (stop < to) |
6ae1f27e RS |
899 | { |
900 | from = stop, from_byte = stop_byte; | |
901 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
902 | } | |
4ed46869 KH |
903 | else |
904 | break; | |
905 | } | |
6ae1f27e | 906 | |
4ed46869 | 907 | val = Qnil; |
2e344af3 | 908 | if (charsets[1]) |
0282eb69 | 909 | val = Fcons (Qunknown, val); |
2e344af3 KH |
910 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
911 | if (charsets[i]) | |
912 | val = Fcons (CHARSET_SYMBOL (i), val); | |
913 | if (charsets[0]) | |
914 | val = Fcons (Qascii, val); | |
4ed46869 KH |
915 | return val; |
916 | } | |
917 | ||
918 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 | 919 | 1, 2, 0, |
fdb82f93 PJ |
920 | doc: /* Return a list of charsets in STR. |
921 | Optional arg TABLE if non-nil is a translation table to look up. | |
922 | ||
923 | If the string contains invalid multibyte characters, | |
924 | `unknown' is included in the returned list. | |
925 | ||
926 | If STR is unibyte, the returned list may contain | |
927 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) | |
928 | (str, table) | |
23d2a7f1 | 929 | Lisp_Object str, table; |
4ed46869 | 930 | { |
a29e3b1b | 931 | int charsets[MAX_CHARSET + 1]; |
4ed46869 KH |
932 | int i; |
933 | Lisp_Object val; | |
934 | ||
935 | CHECK_STRING (str, 0); | |
87b089ad | 936 | |
a29e3b1b | 937 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
2e344af3 KH |
938 | find_charset_in_text (XSTRING (str)->data, XSTRING (str)->size, |
939 | STRING_BYTES (XSTRING (str)), charsets, table); | |
940 | ||
4ed46869 | 941 | val = Qnil; |
2e344af3 | 942 | if (charsets[1]) |
0282eb69 | 943 | val = Fcons (Qunknown, val); |
2e344af3 KH |
944 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
945 | if (charsets[i]) | |
946 | val = Fcons (CHARSET_SYMBOL (i), val); | |
947 | if (charsets[0]) | |
948 | val = Fcons (Qascii, val); | |
4ed46869 KH |
949 | return val; |
950 | } | |
2e344af3 | 951 | |
4ed46869 KH |
952 | \f |
953 | DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
fdb82f93 PJ |
954 | doc: /* Return a character made from arguments. |
955 | Internal use only. */) | |
956 | (charset, code1, code2) | |
4ed46869 KH |
957 | Lisp_Object charset, code1, code2; |
958 | { | |
ac4137cc KH |
959 | int charset_id, c1, c2; |
960 | ||
4ed46869 | 961 | CHECK_NUMBER (charset, 0); |
ac4137cc KH |
962 | charset_id = XINT (charset); |
963 | if (!CHARSET_DEFINED_P (charset_id)) | |
964 | error ("Invalid charset ID: %d", XINT (charset)); | |
4ed46869 KH |
965 | |
966 | if (NILP (code1)) | |
ac4137cc | 967 | c1 = 0; |
4ed46869 | 968 | else |
ac4137cc KH |
969 | { |
970 | CHECK_NUMBER (code1, 1); | |
971 | c1 = XINT (code1); | |
972 | } | |
4ed46869 | 973 | if (NILP (code2)) |
ac4137cc | 974 | c2 = 0; |
4ed46869 | 975 | else |
ac4137cc KH |
976 | { |
977 | CHECK_NUMBER (code2, 2); | |
978 | c2 = XINT (code2); | |
979 | } | |
4ed46869 | 980 | |
2e344af3 KH |
981 | if (charset_id == CHARSET_ASCII) |
982 | { | |
983 | if (c1 < 0 || c1 > 0x7F) | |
984 | goto invalid_code_posints; | |
985 | return make_number (c1); | |
986 | } | |
987 | else if (charset_id == CHARSET_8_BIT_CONTROL) | |
988 | { | |
30736012 KH |
989 | if (NILP (code1)) |
990 | c1 = 0x80; | |
991 | else if (c1 < 0x80 || c1 > 0x9F) | |
2e344af3 KH |
992 | goto invalid_code_posints; |
993 | return make_number (c1); | |
994 | } | |
995 | else if (charset_id == CHARSET_8_BIT_GRAPHIC) | |
996 | { | |
30736012 KH |
997 | if (NILP (code1)) |
998 | c1 = 0xA0; | |
999 | else if (c1 < 0xA0 || c1 > 0xFF) | |
2e344af3 KH |
1000 | goto invalid_code_posints; |
1001 | return make_number (c1); | |
1002 | } | |
1003 | else if (c1 < 0 || c1 > 0xFF || c2 < 0 || c2 > 0xFF) | |
1004 | goto invalid_code_posints; | |
ac4137cc KH |
1005 | c1 &= 0x7F; |
1006 | c2 &= 0x7F; | |
1007 | if (c1 == 0 | |
1008 | ? c2 != 0 | |
1009 | : (c2 == 0 | |
87f67317 KR |
1010 | ? !CHAR_COMPONENTS_VALID_P (charset_id, c1, 0x20) |
1011 | : !CHAR_COMPONENTS_VALID_P (charset_id, c1, c2))) | |
2e344af3 | 1012 | goto invalid_code_posints; |
ac4137cc | 1013 | return make_number (MAKE_CHAR (charset_id, c1, c2)); |
2e344af3 KH |
1014 | |
1015 | invalid_code_posints: | |
1016 | error ("Invalid code points for charset ID %d: %d %d", charset_id, c1, c2); | |
4ed46869 KH |
1017 | } |
1018 | ||
1019 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, | |
fdb82f93 PJ |
1020 | doc: /* Return list of charset and one or two position-codes of CHAR. |
1021 | If CHAR is invalid as a character code, | |
1022 | return a list of symbol `unknown' and CHAR. */) | |
1023 | (ch) | |
4ed46869 KH |
1024 | Lisp_Object ch; |
1025 | { | |
0282eb69 | 1026 | int c, charset, c1, c2; |
4ed46869 KH |
1027 | |
1028 | CHECK_NUMBER (ch, 0); | |
0282eb69 KH |
1029 | c = XFASTINT (ch); |
1030 | if (!CHAR_VALID_P (c, 1)) | |
1031 | return Fcons (Qunknown, Fcons (ch, Qnil)); | |
4ed46869 | 1032 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
6dc0722d | 1033 | return (c2 >= 0 |
4ed46869 KH |
1034 | ? Fcons (CHARSET_SYMBOL (charset), |
1035 | Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
1036 | : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
1037 | } | |
1038 | ||
1039 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
fdb82f93 PJ |
1040 | doc: /* Return charset of CHAR. */) |
1041 | (ch) | |
4ed46869 KH |
1042 | Lisp_Object ch; |
1043 | { | |
1044 | CHECK_NUMBER (ch, 0); | |
1045 | ||
1046 | return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
1047 | } | |
1048 | ||
90d7b74e | 1049 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
fdb82f93 PJ |
1050 | doc: /* Return charset of a character in the current buffer at position POS. |
1051 | If POS is nil, it defauls to the current point. | |
1052 | If POS is out of range, the value is nil. */) | |
1053 | (pos) | |
90d7b74e KH |
1054 | Lisp_Object pos; |
1055 | { | |
2e344af3 KH |
1056 | Lisp_Object ch; |
1057 | int charset; | |
ac4137cc | 1058 | |
2e344af3 KH |
1059 | ch = Fchar_after (pos); |
1060 | if (! INTEGERP (ch)) | |
1061 | return ch; | |
1062 | charset = CHAR_CHARSET (XINT (ch)); | |
90d7b74e KH |
1063 | return CHARSET_SYMBOL (charset); |
1064 | } | |
1065 | ||
4ed46869 | 1066 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
fdb82f93 PJ |
1067 | doc: /* Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR. |
1068 | ||
1069 | ISO 2022's designation sequence (escape sequence) distinguishes charsets | |
1070 | by their DIMENSION, CHARS, and FINAL-CHAR, | |
1071 | where as Emacs distinguishes them by charset symbol. | |
1072 | See the documentation of the function `charset-info' for the meanings of | |
1073 | DIMENSION, CHARS, and FINAL-CHAR. */) | |
1074 | (dimension, chars, final_char) | |
4ed46869 KH |
1075 | Lisp_Object dimension, chars, final_char; |
1076 | { | |
1077 | int charset; | |
1078 | ||
1079 | CHECK_NUMBER (dimension, 0); | |
1080 | CHECK_NUMBER (chars, 1); | |
1081 | CHECK_NUMBER (final_char, 2); | |
1082 | ||
1083 | if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
1084 | return Qnil; | |
1085 | return CHARSET_SYMBOL (charset); | |
1086 | } | |
1087 | ||
9d3d8cba KH |
1088 | /* If GENERICP is nonzero, return nonzero iff C is a valid normal or |
1089 | generic character. If GENERICP is zero, return nonzero iff C is a | |
1090 | valid normal character. Do not call this function directly, | |
1091 | instead use macro CHAR_VALID_P. */ | |
1092 | int | |
1093 | char_valid_p (c, genericp) | |
1094 | int c, genericp; | |
1095 | { | |
1096 | int charset, c1, c2; | |
1097 | ||
0e235b7e | 1098 | if (c < 0 || c >= MAX_CHAR) |
9d3d8cba KH |
1099 | return 0; |
1100 | if (SINGLE_BYTE_CHAR_P (c)) | |
1101 | return 1; | |
2e344af3 | 1102 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
1103 | if (genericp) |
1104 | { | |
1105 | if (c1) | |
1106 | { | |
1107 | if (c2 <= 0) c2 = 0x20; | |
1108 | } | |
1109 | else | |
1110 | { | |
1111 | if (c2 <= 0) c1 = c2 = 0x20; | |
1112 | } | |
1113 | } | |
1114 | return (CHARSET_DEFINED_P (charset) | |
44c6492d | 1115 | && CHAR_COMPONENTS_VALID_P (charset, c1, c2)); |
9d3d8cba KH |
1116 | } |
1117 | ||
1118 | DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0, | |
fdb82f93 PJ |
1119 | doc: /* Return t if OBJECT is a valid normal character. |
1120 | If optional arg GENERICP is non-nil, also return t if OBJECT is | |
1121 | a valid generic character. */) | |
1122 | (object, genericp) | |
9d3d8cba KH |
1123 | Lisp_Object object, genericp; |
1124 | { | |
1125 | if (! NATNUMP (object)) | |
1126 | return Qnil; | |
1127 | return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil); | |
1128 | } | |
1129 | ||
d2665018 KH |
1130 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, |
1131 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
fdb82f93 PJ |
1132 | doc: /* Convert the unibyte character CH to multibyte character. |
1133 | The conversion is done based on `nonascii-translation-table' (which see) | |
1134 | or `nonascii-insert-offset' (which see). */) | |
1135 | (ch) | |
d2665018 KH |
1136 | Lisp_Object ch; |
1137 | { | |
1138 | int c; | |
1139 | ||
1140 | CHECK_NUMBER (ch, 0); | |
1141 | c = XINT (ch); | |
1142 | if (c < 0 || c >= 0400) | |
1143 | error ("Invalid unibyte character: %d", c); | |
1144 | c = unibyte_char_to_multibyte (c); | |
1145 | if (c < 0) | |
1146 | error ("Can't convert to multibyte character: %d", XINT (ch)); | |
1147 | return make_number (c); | |
1148 | } | |
1149 | ||
1bcc1567 RS |
1150 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, |
1151 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
fdb82f93 PJ |
1152 | doc: /* Convert the multibyte character CH to unibyte character. |
1153 | The conversion is done based on `nonascii-translation-table' (which see) | |
1154 | or `nonascii-insert-offset' (which see). */) | |
1155 | (ch) | |
1bcc1567 RS |
1156 | Lisp_Object ch; |
1157 | { | |
1158 | int c; | |
1159 | ||
1160 | CHECK_NUMBER (ch, 0); | |
1161 | c = XINT (ch); | |
ac4137cc | 1162 | if (! CHAR_VALID_P (c, 0)) |
1bcc1567 RS |
1163 | error ("Invalid multibyte character: %d", c); |
1164 | c = multibyte_char_to_unibyte (c, Qnil); | |
1165 | if (c < 0) | |
1166 | error ("Can't convert to unibyte character: %d", XINT (ch)); | |
1167 | return make_number (c); | |
1168 | } | |
1169 | ||
4ed46869 | 1170 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, |
fdb82f93 PJ |
1171 | doc: /* Return 1 regardless of the argument CHAR. |
1172 | This is now an obsolete function. We keep it just for backward compatibility. */) | |
1173 | (ch) | |
4ed46869 KH |
1174 | Lisp_Object ch; |
1175 | { | |
4ed46869 | 1176 | CHECK_NUMBER (ch, 0); |
9b6a601f KH |
1177 | return make_number (1); |
1178 | } | |
1179 | ||
1180 | /* Return how many bytes C will occupy in a multibyte buffer. | |
1181 | Don't call this function directly, instead use macro CHAR_BYTES. */ | |
1182 | int | |
1183 | char_bytes (c) | |
1184 | int c; | |
1185 | { | |
99529c2c | 1186 | int charset; |
9b6a601f | 1187 | |
2e344af3 KH |
1188 | if (ASCII_BYTE_P (c) || (c & ~((1 << CHARACTERBITS) -1))) |
1189 | return 1; | |
1190 | if (SINGLE_BYTE_CHAR_P (c) && c >= 0xA0) | |
8ac5a9cc KH |
1191 | return 1; |
1192 | ||
99529c2c KH |
1193 | charset = CHAR_CHARSET (c); |
1194 | return (CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1); | |
4ed46869 KH |
1195 | } |
1196 | ||
1197 | /* Return the width of character of which multi-byte form starts with | |
1198 | C. The width is measured by how many columns occupied on the | |
1199 | screen when displayed in the current buffer. */ | |
1200 | ||
1201 | #define ONE_BYTE_CHAR_WIDTH(c) \ | |
1202 | (c < 0x20 \ | |
1203 | ? (c == '\t' \ | |
53316e55 | 1204 | ? XFASTINT (current_buffer->tab_width) \ |
4ed46869 KH |
1205 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
1206 | : (c < 0x7f \ | |
1207 | ? 1 \ | |
1208 | : (c == 0x7F \ | |
1209 | ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
1210 | : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
1211 | && BASE_LEADING_CODE_P (c)) \ | |
1212 | ? WIDTH_BY_CHAR_HEAD (c) \ | |
b4e9dd77 | 1213 | : 4)))) |
4ed46869 KH |
1214 | |
1215 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
fdb82f93 PJ |
1216 | doc: /* Return width of CHAR when displayed in the current buffer. |
1217 | The width is measured by how many columns it occupies on the screen. | |
1218 | Tab is taken to occupy `tab-width' columns. */) | |
1219 | (ch) | |
1220 | Lisp_Object ch; | |
4ed46869 | 1221 | { |
859f2b3c | 1222 | Lisp_Object val, disp; |
4ed46869 | 1223 | int c; |
51c4025f | 1224 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 KH |
1225 | |
1226 | CHECK_NUMBER (ch, 0); | |
1227 | ||
859f2b3c RS |
1228 | c = XINT (ch); |
1229 | ||
1230 | /* Get the way the display table would display it. */ | |
51c4025f | 1231 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; |
859f2b3c RS |
1232 | |
1233 | if (VECTORP (disp)) | |
1234 | XSETINT (val, XVECTOR (disp)->size); | |
1235 | else if (SINGLE_BYTE_CHAR_P (c)) | |
1236 | XSETINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
4ed46869 KH |
1237 | else |
1238 | { | |
1239 | int charset = CHAR_CHARSET (c); | |
1240 | ||
1241 | XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
1242 | } | |
1243 | return val; | |
1244 | } | |
1245 | ||
1246 | /* Return width of string STR of length LEN when displayed in the | |
1247 | current buffer. The width is measured by how many columns it | |
1248 | occupies on the screen. */ | |
859f2b3c | 1249 | |
4ed46869 KH |
1250 | int |
1251 | strwidth (str, len) | |
1252 | unsigned char *str; | |
1253 | int len; | |
1254 | { | |
beeedaad KH |
1255 | return c_string_width (str, len, -1, NULL, NULL); |
1256 | } | |
1257 | ||
1258 | /* Return width of string STR of length LEN when displayed in the | |
1259 | current buffer. The width is measured by how many columns it | |
1260 | occupies on the screen. If PRECISION > 0, return the width of | |
1261 | longest substring that doesn't exceed PRECISION, and set number of | |
1262 | characters and bytes of the substring in *NCHARS and *NBYTES | |
1263 | respectively. */ | |
1264 | ||
1265 | int | |
1266 | c_string_width (str, len, precision, nchars, nbytes) | |
1267 | unsigned char *str; | |
1268 | int precision, *nchars, *nbytes; | |
1269 | { | |
1270 | int i = 0, i_byte = 0; | |
4ed46869 | 1271 | int width = 0; |
beeedaad | 1272 | int chars; |
c4a4e28f | 1273 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1274 | |
beeedaad | 1275 | while (i_byte < len) |
859f2b3c | 1276 | { |
beeedaad KH |
1277 | int bytes, thiswidth; |
1278 | Lisp_Object val; | |
859f2b3c | 1279 | |
99529c2c | 1280 | if (dp) |
beeedaad KH |
1281 | { |
1282 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
1283 | ||
1284 | chars = 1; | |
1285 | val = DISP_CHAR_VECTOR (dp, c); | |
1286 | if (VECTORP (val)) | |
1287 | thiswidth = XVECTOR (val)->size; | |
1288 | else | |
1289 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1290 | } | |
859f2b3c | 1291 | else |
beeedaad KH |
1292 | { |
1293 | chars = 1; | |
1294 | PARSE_MULTIBYTE_SEQ (str + i_byte, len - i_byte, bytes); | |
1295 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1296 | } | |
99529c2c | 1297 | |
beeedaad KH |
1298 | if (precision > 0 |
1299 | && (width + thiswidth > precision)) | |
1300 | { | |
1301 | *nchars = i; | |
1302 | *nbytes = i_byte; | |
1303 | return width; | |
1304 | } | |
1305 | i++; | |
1306 | i_byte += bytes; | |
1307 | width += thiswidth; | |
1308 | } | |
99529c2c | 1309 | |
beeedaad KH |
1310 | if (precision > 0) |
1311 | { | |
1312 | *nchars = i; | |
1313 | *nbytes = i_byte; | |
859f2b3c | 1314 | } |
beeedaad | 1315 | |
4ed46869 KH |
1316 | return width; |
1317 | } | |
1318 | ||
beeedaad KH |
1319 | /* Return width of Lisp string STRING when displayed in the current |
1320 | buffer. The width is measured by how many columns it occupies on | |
1321 | the screen while paying attention to compositions. If PRECISION > | |
1322 | 0, return the width of longest substring that doesn't exceed | |
1323 | PRECISION, and set number of characters and bytes of the substring | |
1324 | in *NCHARS and *NBYTES respectively. */ | |
1325 | ||
3f62427c | 1326 | int |
beeedaad KH |
1327 | lisp_string_width (string, precision, nchars, nbytes) |
1328 | Lisp_Object string; | |
1329 | int precision, *nchars, *nbytes; | |
3f62427c | 1330 | { |
beeedaad KH |
1331 | int len = XSTRING (string)->size; |
1332 | int len_byte = STRING_BYTES (XSTRING (string)); | |
1333 | unsigned char *str = XSTRING (string)->data; | |
1334 | int i = 0, i_byte = 0; | |
3f62427c | 1335 | int width = 0; |
beeedaad | 1336 | struct Lisp_Char_Table *dp = buffer_display_table (); |
3f62427c KH |
1337 | |
1338 | while (i < len) | |
1339 | { | |
beeedaad KH |
1340 | int chars, bytes, thiswidth; |
1341 | Lisp_Object val; | |
1342 | int cmp_id; | |
1343 | int ignore, end; | |
1344 | ||
1345 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
1346 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
1347 | >= 0)) | |
3f62427c | 1348 | { |
beeedaad KH |
1349 | thiswidth = composition_table[cmp_id]->width; |
1350 | chars = end - i; | |
1351 | bytes = string_char_to_byte (string, end) - i_byte; | |
1352 | } | |
1353 | else if (dp) | |
1354 | { | |
1355 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
1356 | ||
1357 | chars = 1; | |
1358 | val = DISP_CHAR_VECTOR (dp, c); | |
1359 | if (VECTORP (val)) | |
1360 | thiswidth = XVECTOR (val)->size; | |
1361 | else | |
1362 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
3f62427c KH |
1363 | } |
1364 | else | |
1365 | { | |
beeedaad KH |
1366 | chars = 1; |
1367 | PARSE_MULTIBYTE_SEQ (str + i_byte, len_byte - i_byte, bytes); | |
1368 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1369 | } | |
1370 | ||
1371 | if (precision > 0 | |
1372 | && (width + thiswidth > precision)) | |
1373 | { | |
1374 | *nchars = i; | |
1375 | *nbytes = i_byte; | |
1376 | return width; | |
3f62427c | 1377 | } |
beeedaad KH |
1378 | i += chars; |
1379 | i_byte += bytes; | |
1380 | width += thiswidth; | |
1381 | } | |
1382 | ||
1383 | if (precision > 0) | |
1384 | { | |
1385 | *nchars = i; | |
1386 | *nbytes = i_byte; | |
3f62427c | 1387 | } |
beeedaad | 1388 | |
3f62427c KH |
1389 | return width; |
1390 | } | |
1391 | ||
4ed46869 | 1392 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, |
fdb82f93 PJ |
1393 | doc: /* Return width of STRING when displayed in the current buffer. |
1394 | Width is measured by how many columns it occupies on the screen. | |
1395 | When calculating width of a multibyte character in STRING, | |
1396 | only the base leading-code is considered; the validity of | |
1397 | the following bytes is not checked. Tabs in STRING are always | |
1398 | taken to occupy `tab-width' columns. */) | |
1399 | (str) | |
4ed46869 KH |
1400 | Lisp_Object str; |
1401 | { | |
1402 | Lisp_Object val; | |
1403 | ||
1404 | CHECK_STRING (str, 0); | |
beeedaad | 1405 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); |
4ed46869 KH |
1406 | return val; |
1407 | } | |
1408 | ||
1409 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
fdb82f93 PJ |
1410 | doc: /* Return the direction of CHAR. |
1411 | The returned value is 0 for left-to-right and 1 for right-to-left. */) | |
1412 | (ch) | |
4ed46869 KH |
1413 | Lisp_Object ch; |
1414 | { | |
1415 | int charset; | |
1416 | ||
1417 | CHECK_NUMBER (ch, 0); | |
1418 | charset = CHAR_CHARSET (XFASTINT (ch)); | |
1419 | if (!CHARSET_DEFINED_P (charset)) | |
93bcb785 | 1420 | invalid_character (XINT (ch)); |
4ed46869 KH |
1421 | return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); |
1422 | } | |
1423 | ||
af4fecb4 | 1424 | DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, |
fdb82f93 PJ |
1425 | doc: /* Return number of characters between BEG and END. */) |
1426 | (beg, end) | |
046b1f03 RS |
1427 | Lisp_Object beg, end; |
1428 | { | |
6ae1f27e | 1429 | int from, to; |
046b1f03 | 1430 | |
17e7ef1b RS |
1431 | CHECK_NUMBER_COERCE_MARKER (beg, 0); |
1432 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
1433 | ||
046b1f03 | 1434 | from = min (XFASTINT (beg), XFASTINT (end)); |
a8a35e61 | 1435 | to = max (XFASTINT (beg), XFASTINT (end)); |
046b1f03 | 1436 | |
a8c21066 | 1437 | return make_number (to - from); |
6ae1f27e | 1438 | } |
9036eb45 | 1439 | |
87b089ad RS |
1440 | /* Return the number of characters in the NBYTES bytes at PTR. |
1441 | This works by looking at the contents and checking for multibyte sequences. | |
1442 | However, if the current buffer has enable-multibyte-characters = nil, | |
1443 | we treat each byte as a character. */ | |
1444 | ||
6ae1f27e RS |
1445 | int |
1446 | chars_in_text (ptr, nbytes) | |
1447 | unsigned char *ptr; | |
1448 | int nbytes; | |
1449 | { | |
87b089ad RS |
1450 | /* current_buffer is null at early stages of Emacs initialization. */ |
1451 | if (current_buffer == 0 | |
1452 | || NILP (current_buffer->enable_multibyte_characters)) | |
6ae1f27e | 1453 | return nbytes; |
a8a35e61 | 1454 | |
ac4137cc | 1455 | return multibyte_chars_in_text (ptr, nbytes); |
046b1f03 RS |
1456 | } |
1457 | ||
87b089ad RS |
1458 | /* Return the number of characters in the NBYTES bytes at PTR. |
1459 | This works by looking at the contents and checking for multibyte sequences. | |
1460 | It ignores enable-multibyte-characters. */ | |
1461 | ||
1462 | int | |
1463 | multibyte_chars_in_text (ptr, nbytes) | |
1464 | unsigned char *ptr; | |
1465 | int nbytes; | |
1466 | { | |
ac4137cc KH |
1467 | unsigned char *endp; |
1468 | int chars, bytes; | |
87b089ad RS |
1469 | |
1470 | endp = ptr + nbytes; | |
1471 | chars = 0; | |
1472 | ||
1473 | while (ptr < endp) | |
1474 | { | |
2e344af3 KH |
1475 | PARSE_MULTIBYTE_SEQ (ptr, endp - ptr, bytes); |
1476 | ptr += bytes; | |
1477 | chars++; | |
1478 | } | |
1479 | ||
1480 | return chars; | |
1481 | } | |
1482 | ||
d8e4f486 | 1483 | /* Parse unibyte text at STR of LEN bytes as multibyte text, and |
2e344af3 | 1484 | count the numbers of characters and bytes in it. On counting |
d8e4f486 DL |
1485 | bytes, pay attention to the fact that 8-bit characters in the range |
1486 | 0x80..0x9F are represented by 2 bytes in multibyte text. */ | |
2e344af3 KH |
1487 | void |
1488 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
1489 | unsigned char *str; | |
1490 | int len, *nchars, *nbytes; | |
1491 | { | |
1492 | unsigned char *endp = str + len; | |
1493 | int n, chars = 0, bytes = 0; | |
1494 | ||
1495 | while (str < endp) | |
1496 | { | |
1497 | if (UNIBYTE_STR_AS_MULTIBYTE_P (str, endp - str, n)) | |
1498 | str += n, bytes += n; | |
1499 | else | |
1500 | str++, bytes += 2; | |
1501 | chars++; | |
1502 | } | |
1503 | *nchars = chars; | |
1504 | *nbytes = bytes; | |
1505 | return; | |
1506 | } | |
1507 | ||
d8e4f486 | 1508 | /* Arrange unibyte text at STR of NBYTES bytes as multibyte text. |
2e344af3 KH |
1509 | It actually converts only 8-bit characters in the range 0x80..0x9F |
1510 | that don't contruct multibyte characters to multibyte forms. If | |
1511 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
1512 | text. It is assured that we can use LEN bytes at STR as a work | |
1513 | area and that is enough. Return the number of bytes of the | |
1514 | resulting text. */ | |
1515 | ||
1516 | int | |
1517 | str_as_multibyte (str, len, nbytes, nchars) | |
1518 | unsigned char *str; | |
1519 | int len, nbytes, *nchars; | |
1520 | { | |
1521 | unsigned char *p = str, *endp = str + nbytes; | |
1522 | unsigned char *to; | |
1523 | int chars = 0; | |
1524 | int n; | |
1525 | ||
1526 | while (p < endp && UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
1527 | p += n, chars++; | |
1528 | if (nchars) | |
1529 | *nchars = chars; | |
1530 | if (p == endp) | |
1531 | return nbytes; | |
1532 | ||
1533 | to = p; | |
1534 | nbytes = endp - p; | |
1535 | endp = str + len; | |
1536 | safe_bcopy (p, endp - nbytes, nbytes); | |
1537 | p = endp - nbytes; | |
1538 | while (p < endp) | |
1539 | { | |
1540 | if (UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
ac4137cc | 1541 | { |
2e344af3 KH |
1542 | while (n--) |
1543 | *to++ = *p++; | |
1544 | } | |
ac4137cc KH |
1545 | else |
1546 | { | |
2e344af3 KH |
1547 | *to++ = LEADING_CODE_8_BIT_CONTROL; |
1548 | *to++ = *p++ + 0x20; | |
ac4137cc | 1549 | } |
87b089ad RS |
1550 | chars++; |
1551 | } | |
2e344af3 KH |
1552 | if (nchars) |
1553 | *nchars = chars; | |
1554 | return (to - str); | |
1555 | } | |
87b089ad | 1556 | |
740f080d KH |
1557 | /* Parse unibyte string at STR of LEN bytes, and return the number of |
1558 | bytes it may ocupy when converted to multibyte string by | |
1559 | `str_to_multibyte'. */ | |
1560 | ||
1561 | int | |
1562 | parse_str_to_multibyte (str, len) | |
1563 | unsigned char *str; | |
1564 | int len; | |
1565 | { | |
1566 | unsigned char *endp = str + len; | |
1567 | int bytes; | |
1568 | ||
1569 | for (bytes = 0; str < endp; str++) | |
1570 | bytes += (*str < 0x80 || *str >= 0xA0) ? 1 : 2; | |
1571 | return bytes; | |
1572 | } | |
1573 | ||
d8e4f486 | 1574 | /* Convert unibyte text at STR of NBYTES bytes to multibyte text |
2e344af3 KH |
1575 | that contains the same single-byte characters. It actually |
1576 | converts all 8-bit characters to multibyte forms. It is assured | |
1577 | that we can use LEN bytes at STR as a work area and that is | |
1578 | enough. */ | |
1579 | ||
1580 | int | |
1581 | str_to_multibyte (str, len, bytes) | |
1582 | unsigned char *str; | |
1583 | int len, bytes; | |
1584 | { | |
1585 | unsigned char *p = str, *endp = str + bytes; | |
1586 | unsigned char *to; | |
2e344af3 KH |
1587 | |
1588 | while (p < endp && (*p < 0x80 || *p >= 0xA0)) p++; | |
1589 | if (p == endp) | |
1590 | return bytes; | |
1591 | to = p; | |
1592 | bytes = endp - p; | |
1593 | endp = str + len; | |
1594 | safe_bcopy (p, endp - bytes, bytes); | |
1595 | p = endp - bytes; | |
1596 | while (p < endp) | |
1597 | { | |
1598 | if (*p < 0x80 || *p >= 0xA0) | |
1599 | *to++ = *p++; | |
1600 | else | |
1601 | *to++ = LEADING_CODE_8_BIT_CONTROL, *to++ = *p++ + 0x20; | |
1602 | } | |
1603 | return (to - str); | |
87b089ad RS |
1604 | } |
1605 | ||
2e344af3 KH |
1606 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It |
1607 | actually converts only 8-bit characters in the range 0x80..0x9F to | |
1608 | unibyte forms. */ | |
1609 | ||
1610 | int | |
1611 | str_as_unibyte (str, bytes) | |
1612 | unsigned char *str; | |
1613 | int bytes; | |
1614 | { | |
1615 | unsigned char *p = str, *endp = str + bytes; | |
1616 | unsigned char *to = str; | |
1617 | ||
1618 | while (p < endp && *p != LEADING_CODE_8_BIT_CONTROL) p++; | |
1619 | to = p; | |
1620 | while (p < endp) | |
1621 | { | |
1622 | if (*p == LEADING_CODE_8_BIT_CONTROL) | |
1623 | *to++ = *(p + 1) - 0x20, p += 2; | |
1624 | else | |
1625 | *to++ = *p++; | |
1626 | } | |
1627 | return (to - str); | |
1628 | } | |
1629 | ||
1630 | \f | |
87b089ad | 1631 | DEFUN ("string", Fstring, Sstring, 1, MANY, 0, |
fdb82f93 PJ |
1632 | doc: /* Concatenate all the argument characters and make the result a string. */) |
1633 | (n, args) | |
53316e55 | 1634 | int n; |
4ed46869 KH |
1635 | Lisp_Object *args; |
1636 | { | |
53316e55 | 1637 | int i; |
99529c2c | 1638 | unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); |
4ed46869 | 1639 | unsigned char *p = buf; |
2e344af3 | 1640 | int c; |
5729c92f | 1641 | int multibyte = 0; |
4ed46869 KH |
1642 | |
1643 | for (i = 0; i < n; i++) | |
1644 | { | |
d0c037d8 | 1645 | CHECK_NUMBER (args[i], 0); |
5729c92f KH |
1646 | if (!multibyte && !SINGLE_BYTE_CHAR_P (XFASTINT (args[i]))) |
1647 | multibyte = 1; | |
1648 | } | |
1649 | ||
1650 | for (i = 0; i < n; i++) | |
1651 | { | |
4ed46869 | 1652 | c = XINT (args[i]); |
5729c92f KH |
1653 | if (multibyte) |
1654 | p += CHAR_STRING (c, p); | |
1655 | else | |
555e35d2 | 1656 | *p++ = c; |
4ed46869 KH |
1657 | } |
1658 | ||
2e344af3 | 1659 | return make_string_from_bytes (buf, n, p - buf); |
4ed46869 KH |
1660 | } |
1661 | ||
1662 | #endif /* emacs */ | |
1663 | \f | |
dfcf069d | 1664 | int |
4ed46869 KH |
1665 | charset_id_internal (charset_name) |
1666 | char *charset_name; | |
1667 | { | |
76d7b829 | 1668 | Lisp_Object val; |
4ed46869 | 1669 | |
76d7b829 | 1670 | val= Fget (intern (charset_name), Qcharset); |
4ed46869 KH |
1671 | if (!VECTORP (val)) |
1672 | error ("Charset %s is not defined", charset_name); | |
1673 | ||
1674 | return (XINT (XVECTOR (val)->contents[0])); | |
1675 | } | |
1676 | ||
1677 | DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
fdb82f93 PJ |
1678 | Ssetup_special_charsets, 0, 0, 0, doc: /* Internal use only. */) |
1679 | () | |
4ed46869 KH |
1680 | { |
1681 | charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1682 | charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1683 | charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1684 | charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1685 | charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1686 | charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1687 | charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1688 | return Qnil; | |
1689 | } | |
1690 | ||
dfcf069d | 1691 | void |
4ed46869 KH |
1692 | init_charset_once () |
1693 | { | |
1694 | int i, j, k; | |
1695 | ||
1696 | staticpro (&Vcharset_table); | |
1697 | staticpro (&Vcharset_symbol_table); | |
8a73a704 | 1698 | staticpro (&Vgeneric_character_list); |
4ed46869 KH |
1699 | |
1700 | /* This has to be done here, before we call Fmake_char_table. */ | |
1701 | Qcharset_table = intern ("charset-table"); | |
1702 | staticpro (&Qcharset_table); | |
1703 | ||
1704 | /* Intern this now in case it isn't already done. | |
1705 | Setting this variable twice is harmless. | |
1706 | But don't staticpro it here--that is done in alloc.c. */ | |
1707 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1708 | ||
1709 | /* Now we are ready to set up this property, so we can | |
1710 | create the charset table. */ | |
1711 | Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1712 | Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1713 | ||
0282eb69 KH |
1714 | Qunknown = intern ("unknown"); |
1715 | staticpro (&Qunknown); | |
1716 | Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), | |
1717 | Qunknown); | |
4ed46869 KH |
1718 | |
1719 | /* Setup tables. */ | |
1720 | for (i = 0; i < 2; i++) | |
1721 | for (j = 0; j < 2; j++) | |
1722 | for (k = 0; k < 128; k++) | |
1723 | iso_charset_table [i][j][k] = -1; | |
1724 | ||
60383934 | 1725 | for (i = 0; i < 256; i++) |
2e344af3 | 1726 | bytes_by_char_head[i] = 1; |
2e344af3 KH |
1727 | bytes_by_char_head[LEADING_CODE_PRIVATE_11] = 3; |
1728 | bytes_by_char_head[LEADING_CODE_PRIVATE_12] = 3; | |
1729 | bytes_by_char_head[LEADING_CODE_PRIVATE_21] = 4; | |
1730 | bytes_by_char_head[LEADING_CODE_PRIVATE_22] = 4; | |
4ed46869 KH |
1731 | |
1732 | for (i = 0; i < 128; i++) | |
2e344af3 | 1733 | width_by_char_head[i] = 1; |
4ed46869 | 1734 | for (; i < 256; i++) |
2e344af3 KH |
1735 | width_by_char_head[i] = 4; |
1736 | width_by_char_head[LEADING_CODE_PRIVATE_11] = 1; | |
1737 | width_by_char_head[LEADING_CODE_PRIVATE_12] = 2; | |
1738 | width_by_char_head[LEADING_CODE_PRIVATE_21] = 1; | |
1739 | width_by_char_head[LEADING_CODE_PRIVATE_22] = 2; | |
8a73a704 KH |
1740 | |
1741 | { | |
76d7b829 | 1742 | Lisp_Object val; |
8a73a704 | 1743 | |
76d7b829 | 1744 | val = Qnil; |
8a73a704 KH |
1745 | for (i = 0x81; i < 0x90; i++) |
1746 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1747 | for (; i < 0x9A; i++) | |
1748 | val = Fcons (make_number ((i - 0x8F) << 14), val); | |
1749 | for (i = 0xA0; i < 0xF0; i++) | |
1750 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1751 | for (; i < 0xFF; i++) | |
1752 | val = Fcons (make_number ((i - 0xE0) << 14), val); | |
8a73a704 KH |
1753 | Vgeneric_character_list = Fnreverse (val); |
1754 | } | |
bbf12bb3 KH |
1755 | |
1756 | nonascii_insert_offset = 0; | |
1757 | Vnonascii_translation_table = Qnil; | |
4ed46869 KH |
1758 | } |
1759 | ||
1760 | #ifdef emacs | |
1761 | ||
dfcf069d | 1762 | void |
4ed46869 KH |
1763 | syms_of_charset () |
1764 | { | |
2e344af3 KH |
1765 | Qcharset = intern ("charset"); |
1766 | staticpro (&Qcharset); | |
1767 | ||
4ed46869 KH |
1768 | Qascii = intern ("ascii"); |
1769 | staticpro (&Qascii); | |
1770 | ||
2e344af3 KH |
1771 | Qeight_bit_control = intern ("eight-bit-control"); |
1772 | staticpro (&Qeight_bit_control); | |
1773 | ||
1774 | Qeight_bit_graphic = intern ("eight-bit-graphic"); | |
1775 | staticpro (&Qeight_bit_graphic); | |
4ed46869 | 1776 | |
2e344af3 KH |
1777 | /* Define special charsets ascii, eight-bit-control, and |
1778 | eight-bit-graphic. */ | |
4ed46869 KH |
1779 | update_charset_table (make_number (CHARSET_ASCII), |
1780 | make_number (1), make_number (94), | |
1781 | make_number (1), | |
1782 | make_number (0), | |
1783 | make_number ('B'), | |
1784 | make_number (0), | |
1785 | build_string ("ASCII"), | |
d78bc582 | 1786 | Qnil, /* same as above */ |
4ed46869 KH |
1787 | build_string ("ASCII (ISO646 IRV)")); |
1788 | CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1789 | Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1790 | ||
2e344af3 KH |
1791 | update_charset_table (make_number (CHARSET_8_BIT_CONTROL), |
1792 | make_number (1), make_number (96), | |
797a084a | 1793 | make_number (4), |
2e344af3 KH |
1794 | make_number (0), |
1795 | make_number (-1), | |
1796 | make_number (-1), | |
1797 | build_string ("8-bit control code (0x80..0x9F)"), | |
d78bc582 KH |
1798 | Qnil, /* same as above */ |
1799 | Qnil); /* same as above */ | |
2e344af3 KH |
1800 | CHARSET_SYMBOL (CHARSET_8_BIT_CONTROL) = Qeight_bit_control; |
1801 | Fput (Qeight_bit_control, Qcharset, | |
1802 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_CONTROL)); | |
1803 | ||
1804 | update_charset_table (make_number (CHARSET_8_BIT_GRAPHIC), | |
1805 | make_number (1), make_number (96), | |
797a084a | 1806 | make_number (4), |
2e344af3 KH |
1807 | make_number (0), |
1808 | make_number (-1), | |
1809 | make_number (-1), | |
2e344af3 | 1810 | build_string ("8-bit graphic char (0xA0..0xFF)"), |
d78bc582 KH |
1811 | Qnil, /* same as above */ |
1812 | Qnil); /* same as above */ | |
2e344af3 KH |
1813 | CHARSET_SYMBOL (CHARSET_8_BIT_GRAPHIC) = Qeight_bit_graphic; |
1814 | Fput (Qeight_bit_graphic, Qcharset, | |
1815 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_GRAPHIC)); | |
1816 | ||
c1a08b4c KH |
1817 | Qauto_fill_chars = intern ("auto-fill-chars"); |
1818 | staticpro (&Qauto_fill_chars); | |
1819 | Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0)); | |
1820 | ||
4ed46869 | 1821 | defsubr (&Sdefine_charset); |
8a73a704 | 1822 | defsubr (&Sgeneric_character_list); |
3fac5a51 | 1823 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
1824 | defsubr (&Sdeclare_equiv_charset); |
1825 | defsubr (&Sfind_charset_region); | |
1826 | defsubr (&Sfind_charset_string); | |
1827 | defsubr (&Smake_char_internal); | |
1828 | defsubr (&Ssplit_char); | |
1829 | defsubr (&Schar_charset); | |
90d7b74e | 1830 | defsubr (&Scharset_after); |
4ed46869 | 1831 | defsubr (&Siso_charset); |
9d3d8cba | 1832 | defsubr (&Schar_valid_p); |
d2665018 | 1833 | defsubr (&Sunibyte_char_to_multibyte); |
1bcc1567 | 1834 | defsubr (&Smultibyte_char_to_unibyte); |
4ed46869 KH |
1835 | defsubr (&Schar_bytes); |
1836 | defsubr (&Schar_width); | |
1837 | defsubr (&Sstring_width); | |
1838 | defsubr (&Schar_direction); | |
af4fecb4 | 1839 | defsubr (&Schars_in_region); |
87b089ad | 1840 | defsubr (&Sstring); |
4ed46869 KH |
1841 | defsubr (&Ssetup_special_charsets); |
1842 | ||
1843 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
fdb82f93 | 1844 | doc: /* List of charsets ever defined. */); |
2e344af3 KH |
1845 | Vcharset_list = Fcons (Qascii, Fcons (Qeight_bit_control, |
1846 | Fcons (Qeight_bit_graphic, Qnil))); | |
4ed46869 | 1847 | |
537efd8d | 1848 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, |
fdb82f93 PJ |
1849 | doc: /* Vector of cons cell of a symbol and translation table ever defined. |
1850 | An ID of a translation table is an index of this vector. */); | |
537efd8d | 1851 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
b0e3cf2b | 1852 | |
4ed46869 | 1853 | DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, |
fdb82f93 | 1854 | doc: /* Leading-code of private TYPE9N charset of column-width 1. */); |
4ed46869 KH |
1855 | leading_code_private_11 = LEADING_CODE_PRIVATE_11; |
1856 | ||
1857 | DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
fdb82f93 | 1858 | doc: /* Leading-code of private TYPE9N charset of column-width 2. */); |
4ed46869 KH |
1859 | leading_code_private_12 = LEADING_CODE_PRIVATE_12; |
1860 | ||
1861 | DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
fdb82f93 | 1862 | doc: /* Leading-code of private TYPE9Nx9N charset of column-width 1. */); |
4ed46869 KH |
1863 | leading_code_private_21 = LEADING_CODE_PRIVATE_21; |
1864 | ||
1865 | DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
fdb82f93 | 1866 | doc: /* Leading-code of private TYPE9Nx9N charset of column-width 2. */); |
4ed46869 | 1867 | leading_code_private_22 = LEADING_CODE_PRIVATE_22; |
35e623fb RS |
1868 | |
1869 | DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset, | |
fdb82f93 PJ |
1870 | doc: /* Offset for converting non-ASCII unibyte codes 0240...0377 to multibyte. |
1871 | This is used for converting unibyte text to multibyte, | |
1872 | and for inserting character codes specified by number. | |
1873 | ||
1874 | This serves to convert a Latin-1 or similar 8-bit character code | |
1875 | to the corresponding Emacs multibyte character code. | |
1876 | Typically the value should be (- (make-char CHARSET 0) 128), | |
1877 | for your choice of character set. | |
1878 | If `nonascii-translation-table' is non-nil, it overrides this variable. */); | |
35e623fb | 1879 | nonascii_insert_offset = 0; |
b0e3cf2b | 1880 | |
b4e9dd77 | 1881 | DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table, |
fdb82f93 PJ |
1882 | doc: /* Translation table to convert non-ASCII unibyte codes to multibyte. |
1883 | This is used for converting unibyte text to multibyte, | |
1884 | and for inserting character codes specified by number. | |
1885 | ||
1886 | Conversion is performed only when multibyte characters are enabled, | |
1887 | and it serves to convert a Latin-1 or similar 8-bit character code | |
1888 | to the corresponding Emacs character code. | |
1889 | ||
1890 | If this is nil, `nonascii-insert-offset' is used instead. | |
1891 | See also the docstring of `make-translation-table'. */); | |
b4e9dd77 | 1892 | Vnonascii_translation_table = Qnil; |
4cf9710d | 1893 | |
c1a08b4c | 1894 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, |
fdb82f93 PJ |
1895 | doc: /* A char-table for characters which invoke auto-filling. |
1896 | Such characters have value t in this table. */); | |
c1a08b4c | 1897 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); |
60022cb7 AS |
1898 | CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt); |
1899 | CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt); | |
4ed46869 KH |
1900 | } |
1901 | ||
1902 | #endif /* emacs */ |