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. |
bc001814 | 4 | Copyright (C) 2001, 2004 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. */ | |
4bc26a6c SM |
52 | EMACS_INT leading_code_private_11; /* for private DIMENSION1 of 1-column */ |
53 | EMACS_INT leading_code_private_12; /* for private DIMENSION1 of 2-column */ | |
54 | EMACS_INT leading_code_private_21; /* for private DIMENSION2 of 1-column */ | |
55 | EMACS_INT leading_code_private_22; /* for private DIMENSION2 of 2-column */ | |
4ed46869 | 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 | 100 | /* Offset to add to a non-ASCII value when inserting it. */ |
31ade731 | 101 | EMACS_INT nonascii_insert_offset; |
35e623fb | 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 | ||
d37478d0 EZ |
146 | /* 1 if CHARSET, C1, and C2 compose a valid character, else 0. |
147 | Note that this intentionally allows invalid components, such | |
148 | as 0xA0 0xA0, because there exist many files that contain | |
149 | such invalid byte sequences, especially in EUC-GB. */ | |
44c6492d | 150 | #define CHAR_COMPONENTS_VALID_P(charset, c1, c2) \ |
2e344af3 | 151 | ((charset) == CHARSET_ASCII \ |
63f4d579 | 152 | ? ((c1) >= 0 && (c1) <= 0x7F) \ |
2e344af3 KH |
153 | : ((charset) == CHARSET_8_BIT_CONTROL \ |
154 | ? ((c1) >= 0x80 && (c1) <= 0x9F) \ | |
155 | : ((charset) == CHARSET_8_BIT_GRAPHIC \ | |
156 | ? ((c1) >= 0x80 && (c1) <= 0xFF) \ | |
157 | : (CHARSET_DIMENSION (charset) == 1 \ | |
158 | ? ((c1) >= 0x20 && (c1) <= 0x7F) \ | |
159 | : ((c1) >= 0x20 && (c1) <= 0x7F \ | |
160 | && (c2) >= 0x20 && (c2) <= 0x7F))))) | |
93bcb785 | 161 | |
99529c2c KH |
162 | /* Store multi-byte form of the character C in STR. The caller should |
163 | allocate at least 4-byte area at STR in advance. Returns the | |
164 | length of the multi-byte form. If C is an invalid character code, | |
12bcae05 | 165 | return -1. */ |
4ed46869 KH |
166 | |
167 | int | |
12bcae05 | 168 | char_to_string_1 (c, str) |
4ed46869 | 169 | int c; |
99529c2c | 170 | unsigned char *str; |
4ed46869 | 171 | { |
99529c2c KH |
172 | unsigned char *p = str; |
173 | ||
6662e69b | 174 | if (c & CHAR_MODIFIER_MASK) /* This includes the case C is negative. */ |
8ac5a9cc | 175 | { |
6662e69b KH |
176 | /* Multibyte character can't have a modifier bit. */ |
177 | if (! SINGLE_BYTE_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) | |
12bcae05 | 178 | return -1; |
6662e69b KH |
179 | |
180 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
8ac5a9cc | 181 | if (c & CHAR_META) |
6662e69b KH |
182 | { |
183 | /* Move the meta bit to the right place for a string. */ | |
184 | c = (c & ~CHAR_META) | 0x80; | |
185 | } | |
186 | if (c & CHAR_SHIFT) | |
187 | { | |
188 | /* Shift modifier is valid only with [A-Za-z]. */ | |
189 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
190 | c &= ~CHAR_SHIFT; | |
191 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
192 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
193 | } | |
8ac5a9cc | 194 | if (c & CHAR_CTL) |
6662e69b KH |
195 | { |
196 | /* Simulate the code in lread.c. */ | |
197 | /* Allow `\C- ' and `\C-?'. */ | |
198 | if (c == (CHAR_CTL | ' ')) | |
199 | c = 0; | |
200 | else if (c == (CHAR_CTL | '?')) | |
201 | c = 127; | |
202 | /* ASCII control chars are made from letters (both cases), | |
203 | as well as the non-letters within 0100...0137. */ | |
204 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
205 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
206 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
207 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
208 | } | |
209 | ||
5ef834ea KH |
210 | /* If C still has any modifier bits, just ignore it. */ |
211 | c &= ~CHAR_MODIFIER_MASK; | |
2e344af3 | 212 | } |
177c0ea7 | 213 | |
2e344af3 KH |
214 | if (SINGLE_BYTE_CHAR_P (c)) |
215 | { | |
216 | if (ASCII_BYTE_P (c) || c >= 0xA0) | |
217 | *p++ = c; | |
218 | else | |
219 | { | |
220 | *p++ = LEADING_CODE_8_BIT_CONTROL; | |
221 | *p++ = c + 0x20; | |
222 | } | |
8ac5a9cc | 223 | } |
6ce974d4 | 224 | else if (CHAR_VALID_P (c, 0)) |
4ed46869 | 225 | { |
ac4137cc | 226 | int charset, c1, c2; |
4ed46869 | 227 | |
2e344af3 | 228 | SPLIT_CHAR (c, charset, c1, c2); |
99529c2c KH |
229 | |
230 | if (charset >= LEADING_CODE_EXT_11) | |
231 | *p++ = (charset < LEADING_CODE_EXT_12 | |
232 | ? LEADING_CODE_PRIVATE_11 | |
233 | : (charset < LEADING_CODE_EXT_21 | |
234 | ? LEADING_CODE_PRIVATE_12 | |
235 | : (charset < LEADING_CODE_EXT_22 | |
236 | ? LEADING_CODE_PRIVATE_21 | |
237 | : LEADING_CODE_PRIVATE_22))); | |
238 | *p++ = charset; | |
05ffc44b | 239 | if ((c1 > 0 && c1 < 32) || (c2 > 0 && c2 < 32)) |
12bcae05 | 240 | return -1; |
99529c2c | 241 | if (c1) |
4ed46869 | 242 | { |
99529c2c KH |
243 | *p++ = c1 | 0x80; |
244 | if (c2 > 0) | |
245 | *p++ = c2 | 0x80; | |
4ed46869 KH |
246 | } |
247 | } | |
2e344af3 | 248 | else |
12bcae05 | 249 | return -1; |
4ed46869 | 250 | |
2e344af3 | 251 | return (p - str); |
4ed46869 KH |
252 | } |
253 | ||
12bcae05 GM |
254 | |
255 | /* Store multi-byte form of the character C in STR. The caller should | |
256 | allocate at least 4-byte area at STR in advance. Returns the | |
257 | length of the multi-byte form. If C is an invalid character code, | |
258 | signal an error. | |
259 | ||
260 | Use macro `CHAR_STRING (C, STR)' instead of calling this function | |
261 | directly if C can be an ASCII character. */ | |
262 | ||
263 | int | |
264 | char_to_string (c, str) | |
265 | int c; | |
266 | unsigned char *str; | |
267 | { | |
268 | int len; | |
269 | len = char_to_string_1 (c, str); | |
270 | if (len == -1) | |
271 | invalid_character (c); | |
272 | return len; | |
273 | } | |
274 | ||
275 | ||
44c6492d KH |
276 | /* Return the non-ASCII character corresponding to multi-byte form at |
277 | STR of length LEN. If ACTUAL_LEN is not NULL, store the byte | |
278 | length of the multibyte form in *ACTUAL_LEN. | |
537efd8d | 279 | |
99529c2c KH |
280 | Use macros STRING_CHAR or STRING_CHAR_AND_LENGTH instead of calling |
281 | this function directly if you want ot handle ASCII characters as | |
282 | well. */ | |
4ed46869 | 283 | |
dfcf069d | 284 | int |
99529c2c | 285 | string_to_char (str, len, actual_len) |
8867de67 | 286 | const unsigned char *str; |
ac4137cc | 287 | int len, *actual_len; |
4ed46869 | 288 | { |
ac4137cc | 289 | int c, bytes, charset, c1, c2; |
4ed46869 | 290 | |
ac4137cc KH |
291 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, charset, c1, c2); |
292 | c = MAKE_CHAR (charset, c1, c2); | |
4ed46869 | 293 | if (actual_len) |
ac4137cc | 294 | *actual_len = bytes; |
4ed46869 KH |
295 | return c; |
296 | } | |
297 | ||
44c6492d KH |
298 | /* Return the length of the multi-byte form at string STR of length LEN. |
299 | Use the macro MULTIBYTE_FORM_LENGTH instead. */ | |
4ed46869 KH |
300 | int |
301 | multibyte_form_length (str, len) | |
8867de67 | 302 | const unsigned char *str; |
4ed46869 KH |
303 | int len; |
304 | { | |
ac4137cc | 305 | int bytes; |
4ed46869 | 306 | |
ac4137cc | 307 | PARSE_MULTIBYTE_SEQ (str, len, bytes); |
90d7b74e | 308 | return bytes; |
4ed46869 KH |
309 | } |
310 | ||
ac4137cc KH |
311 | /* Check multibyte form at string STR of length LEN and set variables |
312 | pointed by CHARSET, C1, and C2 to charset and position codes of the | |
313 | character at STR, and return 0. If there's no multibyte character, | |
4ed46869 KH |
314 | return -1. This should be used only in the macro SPLIT_STRING |
315 | which checks range of STR in advance. */ | |
316 | ||
dfcf069d | 317 | int |
99529c2c | 318 | split_string (str, len, charset, c1, c2) |
ac4137cc KH |
319 | const unsigned char *str; |
320 | unsigned char *c1, *c2; | |
321 | int len, *charset; | |
4ed46869 | 322 | { |
ac4137cc | 323 | register int bytes, cs, code1, code2 = -1; |
4ed46869 | 324 | |
ac4137cc KH |
325 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, cs, code1, code2); |
326 | if (cs == CHARSET_ASCII) | |
4ed46869 | 327 | return -1; |
ac4137cc KH |
328 | *charset = cs; |
329 | *c1 = code1; | |
330 | *c2 = code2; | |
5865af0d | 331 | return 0; |
ac4137cc KH |
332 | } |
333 | ||
44c6492d KH |
334 | /* Return 1 iff character C has valid printable glyph. |
335 | Use the macro CHAR_PRINTABLE_P instead. */ | |
ac4137cc KH |
336 | int |
337 | char_printable_p (c) | |
338 | int c; | |
339 | { | |
8ebae00c | 340 | int charset, c1, c2; |
ac4137cc | 341 | |
2e344af3 | 342 | if (ASCII_BYTE_P (c)) |
ac4137cc | 343 | return 1; |
2e344af3 KH |
344 | else if (SINGLE_BYTE_CHAR_P (c)) |
345 | return 0; | |
346 | else if (c >= MAX_CHAR) | |
99529c2c | 347 | return 0; |
177c0ea7 | 348 | |
2e344af3 | 349 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
350 | if (! CHARSET_DEFINED_P (charset)) |
351 | return 0; | |
352 | if (CHARSET_CHARS (charset) == 94 | |
353 | ? c1 <= 32 || c1 >= 127 | |
354 | : c1 < 32) | |
355 | return 0; | |
356 | if (CHARSET_DIMENSION (charset) == 2 | |
357 | && (CHARSET_CHARS (charset) == 94 | |
358 | ? c2 <= 32 || c2 >= 127 | |
359 | : c2 < 32)) | |
360 | return 0; | |
361 | return 1; | |
4ed46869 KH |
362 | } |
363 | ||
537efd8d | 364 | /* Translate character C by translation table TABLE. If C |
b4e9dd77 KH |
365 | is negative, translate a character specified by CHARSET, C1, and C2 |
366 | (C1 and C2 are code points of the character). If no translation is | |
367 | found in TABLE, return C. */ | |
dfcf069d | 368 | int |
b4e9dd77 | 369 | translate_char (table, c, charset, c1, c2) |
23d2a7f1 KH |
370 | Lisp_Object table; |
371 | int c, charset, c1, c2; | |
372 | { | |
373 | Lisp_Object ch; | |
374 | int alt_charset, alt_c1, alt_c2, dimension; | |
375 | ||
0ad3f83d | 376 | if (c < 0) c = MAKE_CHAR (charset, (c1 & 0x7F) , (c2 & 0x7F)); |
23d2a7f1 | 377 | if (!CHAR_TABLE_P (table) |
ac4137cc | 378 | || (ch = Faref (table, make_number (c)), !NATNUMP (ch))) |
23d2a7f1 KH |
379 | return c; |
380 | ||
381 | SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); | |
382 | dimension = CHARSET_DIMENSION (alt_charset); | |
05ffc44b | 383 | if ((dimension == 1 && alt_c1 > 0) || (dimension == 2 && alt_c2 > 0)) |
23d2a7f1 KH |
384 | /* CH is not a generic character, just return it. */ |
385 | return XFASTINT (ch); | |
386 | ||
387 | /* Since CH is a generic character, we must return a specific | |
388 | charater which has the same position codes as C from CH. */ | |
389 | if (charset < 0) | |
390 | SPLIT_CHAR (c, charset, c1, c2); | |
391 | if (dimension != CHARSET_DIMENSION (charset)) | |
392 | /* We can't make such a character because of dimension mismatch. */ | |
393 | return c; | |
23d2a7f1 KH |
394 | return MAKE_CHAR (alt_charset, c1, c2); |
395 | } | |
396 | ||
d2665018 | 397 | /* Convert the unibyte character C to multibyte based on |
b4e9dd77 | 398 | Vnonascii_translation_table or nonascii_insert_offset. If they can't |
d2665018 KH |
399 | convert C to a valid multibyte character, convert it based on |
400 | DEFAULT_NONASCII_INSERT_OFFSET which makes C a Latin-1 character. */ | |
35e623fb | 401 | |
dfcf069d | 402 | int |
35e623fb RS |
403 | unibyte_char_to_multibyte (c) |
404 | int c; | |
405 | { | |
543b4f61 | 406 | if (c < 0400 && c >= 0200) |
35e623fb | 407 | { |
d2665018 KH |
408 | int c_save = c; |
409 | ||
b4e9dd77 | 410 | if (! NILP (Vnonascii_translation_table)) |
bbf12bb3 KH |
411 | { |
412 | c = XINT (Faref (Vnonascii_translation_table, make_number (c))); | |
ac4137cc | 413 | if (c >= 0400 && ! char_valid_p (c, 0)) |
bbf12bb3 KH |
414 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
415 | } | |
416 | else if (c >= 0240 && nonascii_insert_offset > 0) | |
417 | { | |
418 | c += nonascii_insert_offset; | |
ac4137cc | 419 | if (c < 0400 || ! char_valid_p (c, 0)) |
bbf12bb3 KH |
420 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
421 | } | |
422 | else if (c >= 0240) | |
d2665018 | 423 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
35e623fb RS |
424 | } |
425 | return c; | |
426 | } | |
76d7b829 KH |
427 | |
428 | ||
429 | /* Convert the multibyte character C to unibyte 8-bit character based | |
430 | on Vnonascii_translation_table or nonascii_insert_offset. If | |
431 | REV_TBL is non-nil, it should be a reverse table of | |
432 | Vnonascii_translation_table, i.e. what given by: | |
433 | Fchar_table_extra_slot (Vnonascii_translation_table, make_number (0)) */ | |
434 | ||
435 | int | |
436 | multibyte_char_to_unibyte (c, rev_tbl) | |
437 | int c; | |
438 | Lisp_Object rev_tbl; | |
439 | { | |
440 | if (!SINGLE_BYTE_CHAR_P (c)) | |
441 | { | |
442 | int c_save = c; | |
443 | ||
444 | if (! CHAR_TABLE_P (rev_tbl) | |
445 | && CHAR_TABLE_P (Vnonascii_translation_table)) | |
446 | rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table, | |
447 | make_number (0)); | |
448 | if (CHAR_TABLE_P (rev_tbl)) | |
449 | { | |
450 | Lisp_Object temp; | |
451 | temp = Faref (rev_tbl, make_number (c)); | |
452 | if (INTEGERP (temp)) | |
453 | c = XINT (temp); | |
bbf12bb3 KH |
454 | if (c >= 256) |
455 | c = (c_save & 0177) + 0200; | |
456 | } | |
457 | else | |
458 | { | |
459 | if (nonascii_insert_offset > 0) | |
460 | c -= nonascii_insert_offset; | |
461 | if (c < 128 || c >= 256) | |
462 | c = (c_save & 0177) + 0200; | |
76d7b829 | 463 | } |
76d7b829 KH |
464 | } |
465 | ||
466 | return c; | |
467 | } | |
468 | ||
35e623fb | 469 | \f |
4ed46869 KH |
470 | /* Update the table Vcharset_table with the given arguments (see the |
471 | document of `define-charset' for the meaning of each argument). | |
472 | Several other table contents are also updated. The caller should | |
473 | check the validity of CHARSET-ID and the remaining arguments in | |
474 | advance. */ | |
475 | ||
476 | void | |
477 | update_charset_table (charset_id, dimension, chars, width, direction, | |
478 | iso_final_char, iso_graphic_plane, | |
479 | short_name, long_name, description) | |
480 | Lisp_Object charset_id, dimension, chars, width, direction; | |
481 | Lisp_Object iso_final_char, iso_graphic_plane; | |
482 | Lisp_Object short_name, long_name, description; | |
483 | { | |
484 | int charset = XINT (charset_id); | |
485 | int bytes; | |
486 | unsigned char leading_code_base, leading_code_ext; | |
487 | ||
6dc0722d KH |
488 | if (NILP (CHARSET_TABLE_ENTRY (charset))) |
489 | CHARSET_TABLE_ENTRY (charset) | |
490 | = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); | |
4ed46869 | 491 | |
d78bc582 KH |
492 | if (NILP (long_name)) |
493 | long_name = short_name; | |
494 | if (NILP (description)) | |
495 | description = long_name; | |
496 | ||
4ed46869 KH |
497 | /* Get byte length of multibyte form, base leading-code, and |
498 | extended leading-code of the charset. See the comment under the | |
499 | title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
500 | bytes = XINT (dimension); | |
501 | if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
502 | { | |
503 | /* Official charset, it doesn't have an extended leading-code. */ | |
2e344af3 | 504 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC) |
4ed46869 KH |
505 | bytes += 1; /* For a base leading-code. */ |
506 | leading_code_base = charset; | |
507 | leading_code_ext = 0; | |
508 | } | |
509 | else | |
510 | { | |
511 | /* Private charset. */ | |
512 | bytes += 2; /* For base and extended leading-codes. */ | |
513 | leading_code_base | |
514 | = (charset < LEADING_CODE_EXT_12 | |
515 | ? LEADING_CODE_PRIVATE_11 | |
516 | : (charset < LEADING_CODE_EXT_21 | |
517 | ? LEADING_CODE_PRIVATE_12 | |
518 | : (charset < LEADING_CODE_EXT_22 | |
519 | ? LEADING_CODE_PRIVATE_21 | |
520 | : LEADING_CODE_PRIVATE_22))); | |
521 | leading_code_ext = charset; | |
c83ef371 KH |
522 | if (BYTES_BY_CHAR_HEAD (leading_code_base) != bytes) |
523 | error ("Invalid dimension for the charset-ID %d", charset); | |
524 | } | |
6ef23ebb | 525 | |
4ed46869 KH |
526 | CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; |
527 | CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
528 | CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
529 | CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
530 | CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
531 | CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
532 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
533 | = make_number (leading_code_base); | |
534 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
535 | = make_number (leading_code_ext); | |
536 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
537 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
538 | = iso_graphic_plane; | |
539 | CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
540 | CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
541 | CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
542 | CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
543 | ||
544 | { | |
545 | /* If we have already defined a charset which has the same | |
546 | DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
547 | DIRECTION, we must update the entry REVERSE-CHARSET of both | |
548 | charsets. If there's no such charset, the value of the entry | |
549 | is set to nil. */ | |
550 | int i; | |
551 | ||
513ee442 | 552 | for (i = 0; i <= MAX_CHARSET; i++) |
4ed46869 KH |
553 | if (!NILP (CHARSET_TABLE_ENTRY (i))) |
554 | { | |
555 | if (CHARSET_DIMENSION (i) == XINT (dimension) | |
556 | && CHARSET_CHARS (i) == XINT (chars) | |
557 | && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
558 | && CHARSET_DIRECTION (i) != XINT (direction)) | |
559 | { | |
560 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
561 | = make_number (i); | |
562 | CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
563 | break; | |
564 | } | |
565 | } | |
513ee442 | 566 | if (i > MAX_CHARSET) |
4ed46869 KH |
567 | /* No such a charset. */ |
568 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
569 | = make_number (-1); | |
570 | } | |
571 | ||
c83ef371 | 572 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC |
4ed46869 KH |
573 | && charset < MIN_CHARSET_PRIVATE_DIMENSION1) |
574 | { | |
c83ef371 | 575 | bytes_by_char_head[leading_code_base] = bytes; |
4ed46869 KH |
576 | width_by_char_head[leading_code_base] = XINT (width); |
577 | ||
578 | /* Update table emacs_code_class. */ | |
579 | emacs_code_class[charset] = (bytes == 2 | |
580 | ? EMACS_leading_code_2 | |
581 | : (bytes == 3 | |
582 | ? EMACS_leading_code_3 | |
583 | : EMACS_leading_code_4)); | |
584 | } | |
585 | ||
586 | /* Update table iso_charset_table. */ | |
52e386c2 | 587 | if (XINT (iso_final_char) >= 0 |
2e344af3 | 588 | && ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0) |
4ed46869 KH |
589 | ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset; |
590 | } | |
591 | ||
592 | #ifdef emacs | |
593 | ||
594 | /* Return charset id of CHARSET_SYMBOL, or return -1 if CHARSET_SYMBOL | |
595 | is invalid. */ | |
596 | int | |
597 | get_charset_id (charset_symbol) | |
598 | Lisp_Object charset_symbol; | |
599 | { | |
600 | Lisp_Object val; | |
601 | int charset; | |
602 | ||
1a45ff10 EZ |
603 | /* This originally used a ?: operator, but reportedly the HP-UX |
604 | compiler version HP92453-01 A.10.32.22 miscompiles that. */ | |
605 | if (SYMBOLP (charset_symbol) | |
606 | && VECTORP (val = Fget (charset_symbol, Qcharset)) | |
607 | && CHARSET_VALID_P (charset = | |
608 | XINT (XVECTOR (val)->contents[CHARSET_ID_IDX]))) | |
609 | return charset; | |
610 | else | |
611 | return -1; | |
4ed46869 KH |
612 | } |
613 | ||
614 | /* Return an identification number for a new private charset of | |
615 | DIMENSION and WIDTH. If there's no more room for the new charset, | |
616 | return 0. */ | |
617 | Lisp_Object | |
618 | get_new_private_charset_id (dimension, width) | |
619 | int dimension, width; | |
620 | { | |
621 | int charset, from, to; | |
622 | ||
623 | if (dimension == 1) | |
624 | { | |
3bb7b08b KH |
625 | from = LEADING_CODE_EXT_11; |
626 | to = LEADING_CODE_EXT_21; | |
4ed46869 KH |
627 | } |
628 | else | |
629 | { | |
3bb7b08b KH |
630 | from = LEADING_CODE_EXT_21; |
631 | to = LEADING_CODE_EXT_MAX + 1; | |
4ed46869 KH |
632 | } |
633 | ||
634 | for (charset = from; charset < to; charset++) | |
635 | if (!CHARSET_DEFINED_P (charset)) break; | |
636 | ||
637 | return make_number (charset < to ? charset : 0); | |
638 | } | |
639 | ||
640 | DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
fdb82f93 PJ |
641 | doc: /* Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR. |
642 | If CHARSET-ID is nil, it is decided automatically, which means CHARSET is | |
643 | treated as a private charset. | |
644 | INFO-VECTOR is a vector of the format: | |
645 | [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE | |
646 | SHORT-NAME LONG-NAME DESCRIPTION] | |
647 | The meanings of each elements is as follows: | |
648 | DIMENSION (integer) is the number of bytes to represent a character: 1 or 2. | |
649 | CHARS (integer) is the number of characters in a dimension: 94 or 96. | |
650 | WIDTH (integer) is the number of columns a character in the charset | |
651 | occupies on the screen: one of 0, 1, and 2. | |
652 | ||
653 | DIRECTION (integer) is the rendering direction of characters in the | |
654 | charset when rendering. If 0, render from left to right, else | |
655 | render from right to left. | |
656 | ||
657 | ISO-FINAL-CHAR (character) is the final character of the | |
658 | corresponding ISO 2022 charset. | |
659 | It may be -1 if the charset is internal use only. | |
660 | ||
661 | ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked | |
662 | while encoding to variants of ISO 2022 coding system, one of the | |
663 | following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR). | |
664 | It may be -1 if the charset is internal use only. | |
665 | ||
666 | SHORT-NAME (string) is the short name to refer to the charset. | |
667 | ||
668 | LONG-NAME (string) is the long name to refer to the charset. | |
669 | ||
670 | DESCRIPTION (string) is the description string of the charset. */) | |
671 | (charset_id, charset_symbol, info_vector) | |
4ed46869 KH |
672 | Lisp_Object charset_id, charset_symbol, info_vector; |
673 | { | |
674 | Lisp_Object *vec; | |
675 | ||
676 | if (!NILP (charset_id)) | |
b7826503 PJ |
677 | CHECK_NUMBER (charset_id); |
678 | CHECK_SYMBOL (charset_symbol); | |
679 | CHECK_VECTOR (info_vector); | |
4ed46869 KH |
680 | |
681 | if (! NILP (charset_id)) | |
682 | { | |
683 | if (! CHARSET_VALID_P (XINT (charset_id))) | |
684 | error ("Invalid CHARSET: %d", XINT (charset_id)); | |
685 | else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
686 | error ("Already defined charset: %d", XINT (charset_id)); | |
687 | } | |
688 | ||
689 | vec = XVECTOR (info_vector)->contents; | |
690 | if (XVECTOR (info_vector)->size != 9 | |
691 | || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
692 | || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
693 | || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
694 | || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
2e344af3 | 695 | || !INTEGERP (vec[4]) |
05ffc44b | 696 | || !(XINT (vec[4]) == -1 || (XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~')) |
2e344af3 KH |
697 | || !INTEGERP (vec[5]) |
698 | || !(XINT (vec[5]) == -1 || XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
4ed46869 KH |
699 | || !STRINGP (vec[6]) |
700 | || !STRINGP (vec[7]) | |
701 | || !STRINGP (vec[8])) | |
702 | error ("Invalid info-vector argument for defining charset %s", | |
d5db4077 | 703 | SDATA (SYMBOL_NAME (charset_symbol))); |
4ed46869 KH |
704 | |
705 | if (NILP (charset_id)) | |
706 | { | |
707 | charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
708 | if (XINT (charset_id) == 0) | |
709 | error ("There's no room for a new private charset %s", | |
d5db4077 | 710 | SDATA (SYMBOL_NAME (charset_symbol))); |
4ed46869 KH |
711 | } |
712 | ||
713 | update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
714 | vec[4], vec[5], vec[6], vec[7], vec[8]); | |
6dc0722d | 715 | Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
4ed46869 KH |
716 | CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
717 | Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
538482ab | 718 | Fupdate_coding_systems_internal (); |
4ed46869 KH |
719 | return Qnil; |
720 | } | |
721 | ||
8a73a704 KH |
722 | DEFUN ("generic-character-list", Fgeneric_character_list, |
723 | Sgeneric_character_list, 0, 0, 0, | |
fdb82f93 PJ |
724 | doc: /* Return a list of all possible generic characters. |
725 | It includes a generic character for a charset not yet defined. */) | |
726 | () | |
8a73a704 KH |
727 | { |
728 | return Vgeneric_character_list; | |
729 | } | |
730 | ||
3fac5a51 KH |
731 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
732 | Sget_unused_iso_final_char, 2, 2, 0, | |
bc001814 | 733 | doc: /* Return an unused ISO's final char for a charset of DIMENSION and CHARS. |
fdb82f93 PJ |
734 | DIMENSION is the number of bytes to represent a character: 1 or 2. |
735 | CHARS is the number of characters in a dimension: 94 or 96. | |
736 | ||
737 | This final char is for private use, thus the range is `0' (48) .. `?' (63). | |
738 | If there's no unused final char for the specified kind of charset, | |
739 | return nil. */) | |
740 | (dimension, chars) | |
3fac5a51 KH |
741 | Lisp_Object dimension, chars; |
742 | { | |
743 | int final_char; | |
744 | ||
b7826503 PJ |
745 | CHECK_NUMBER (dimension); |
746 | CHECK_NUMBER (chars); | |
3fac5a51 KH |
747 | if (XINT (dimension) != 1 && XINT (dimension) != 2) |
748 | error ("Invalid charset dimension %d, it should be 1 or 2", | |
749 | XINT (dimension)); | |
750 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
751 | error ("Invalid charset chars %d, it should be 94 or 96", | |
752 | XINT (chars)); | |
753 | for (final_char = '0'; final_char <= '?'; final_char++) | |
754 | { | |
755 | if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0) | |
756 | break; | |
757 | } | |
758 | return (final_char <= '?' ? make_number (final_char) : Qnil); | |
759 | } | |
760 | ||
4ed46869 KH |
761 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
762 | 4, 4, 0, | |
74ac5074 KH |
763 | doc: /* Declare an equivalent charset for ISO-2022 decoding. |
764 | ||
765 | On decoding by an ISO-2022 base coding system, when a charset | |
766 | specified by DIMENSION, CHARS, and FINAL-CHAR is designated, behave as | |
767 | if CHARSET is designated instead. */) | |
768 | (dimension, chars, final_char, charset) | |
769 | Lisp_Object dimension, chars, final_char, charset; | |
4ed46869 | 770 | { |
74ac5074 | 771 | int charset_id; |
4ed46869 | 772 | |
b7826503 PJ |
773 | CHECK_NUMBER (dimension); |
774 | CHECK_NUMBER (chars); | |
775 | CHECK_NUMBER (final_char); | |
74ac5074 | 776 | CHECK_SYMBOL (charset); |
4ed46869 KH |
777 | |
778 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
779 | error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
780 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
781 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
782 | if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
783 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
74ac5074 KH |
784 | if ((charset_id = get_charset_id (charset)) < 0) |
785 | error ("Invalid charset %s", SDATA (SYMBOL_NAME (charset))); | |
4ed46869 | 786 | |
74ac5074 | 787 | ISO_CHARSET_TABLE (dimension, chars, final_char) = charset_id; |
4ed46869 KH |
788 | return Qnil; |
789 | } | |
790 | ||
2e344af3 KH |
791 | /* Return information about charsets in the text at PTR of NBYTES |
792 | bytes, which are NCHARS characters. The value is: | |
f6302ac9 | 793 | |
cfe34140 | 794 | 0: Each character is represented by one byte. This is always |
f6302ac9 | 795 | true for unibyte text. |
2e344af3 KH |
796 | 1: No charsets other than ascii eight-bit-control, |
797 | eight-bit-graphic, and latin-1 are found. | |
798 | 2: Otherwise. | |
1d67c29b | 799 | |
2e344af3 KH |
800 | In addition, if CHARSETS is nonzero, for each found charset N, set |
801 | CHARSETS[N] to 1. For that, callers should allocate CHARSETS | |
802 | (MAX_CHARSET + 1 elements) in advance. It may lookup a translation | |
803 | table TABLE if supplied. For invalid charsets, set CHARSETS[1] to | |
804 | 1 (note that there's no charset whose ID is 1). */ | |
4ed46869 KH |
805 | |
806 | int | |
2e344af3 | 807 | find_charset_in_text (ptr, nchars, nbytes, charsets, table) |
fb4f7f50 | 808 | const unsigned char *ptr; |
2e344af3 | 809 | int nchars, nbytes, *charsets; |
23d2a7f1 | 810 | Lisp_Object table; |
4ed46869 | 811 | { |
2e344af3 | 812 | if (nchars == nbytes) |
0282eb69 | 813 | { |
2e344af3 | 814 | if (charsets && nbytes > 0) |
0282eb69 | 815 | { |
fb4f7f50 | 816 | const unsigned char *endp = ptr + nbytes; |
2e344af3 KH |
817 | int maskbits = 0; |
818 | ||
819 | while (ptr < endp && maskbits != 7) | |
820 | { | |
821 | maskbits |= (*ptr < 0x80 ? 1 : *ptr < 0xA0 ? 2 : 4); | |
822 | ptr++; | |
177c0ea7 | 823 | } |
2e344af3 KH |
824 | |
825 | if (maskbits & 1) | |
826 | charsets[CHARSET_ASCII] = 1; | |
827 | if (maskbits & 2) | |
828 | charsets[CHARSET_8_BIT_CONTROL] = 1; | |
829 | if (maskbits & 4) | |
830 | charsets[CHARSET_8_BIT_GRAPHIC] = 1; | |
0282eb69 | 831 | } |
2e344af3 | 832 | return 0; |
0282eb69 | 833 | } |
2e344af3 | 834 | else |
4ed46869 | 835 | { |
2e344af3 | 836 | int return_val = 1; |
99529c2c | 837 | int bytes, charset, c1, c2; |
05505664 | 838 | |
2e344af3 KH |
839 | if (! CHAR_TABLE_P (table)) |
840 | table = Qnil; | |
05505664 | 841 | |
2e344af3 | 842 | while (nchars-- > 0) |
23d2a7f1 | 843 | { |
2e344af3 KH |
844 | SPLIT_MULTIBYTE_SEQ (ptr, len, bytes, charset, c1, c2); |
845 | ptr += bytes; | |
4ed46869 | 846 | |
2e344af3 KH |
847 | if (!CHARSET_DEFINED_P (charset)) |
848 | charset = 1; | |
849 | else if (! NILP (table)) | |
850 | { | |
851 | int c = translate_char (table, -1, charset, c1, c2); | |
852 | if (c >= 0) | |
853 | charset = CHAR_CHARSET (c); | |
854 | } | |
855 | ||
856 | if (return_val == 1 | |
857 | && charset != CHARSET_ASCII | |
858 | && charset != CHARSET_8_BIT_CONTROL | |
859 | && charset != CHARSET_8_BIT_GRAPHIC | |
860 | && charset != charset_latin_iso8859_1) | |
861 | return_val = 2; | |
862 | ||
863 | if (charsets) | |
864 | charsets[charset] = 1; | |
865 | else if (return_val == 2) | |
866 | break; | |
4ed46869 | 867 | } |
2e344af3 | 868 | return return_val; |
4ed46869 | 869 | } |
4ed46869 KH |
870 | } |
871 | ||
872 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 873 | 2, 3, 0, |
fdb82f93 PJ |
874 | doc: /* Return a list of charsets in the region between BEG and END. |
875 | BEG and END are buffer positions. | |
876 | Optional arg TABLE if non-nil is a translation table to look up. | |
877 | ||
878 | If the region contains invalid multibyte characters, | |
879 | `unknown' is included in the returned list. | |
880 | ||
881 | If the current buffer is unibyte, the returned list may contain | |
882 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) | |
883 | (beg, end, table) | |
23d2a7f1 | 884 | Lisp_Object beg, end, table; |
4ed46869 | 885 | { |
028d516b | 886 | int charsets[MAX_CHARSET + 1]; |
6ae1f27e | 887 | int from, from_byte, to, stop, stop_byte, i; |
4ed46869 KH |
888 | Lisp_Object val; |
889 | ||
890 | validate_region (&beg, &end); | |
891 | from = XFASTINT (beg); | |
892 | stop = to = XFASTINT (end); | |
6ae1f27e | 893 | |
4ed46869 | 894 | if (from < GPT && GPT < to) |
6ae1f27e RS |
895 | { |
896 | stop = GPT; | |
897 | stop_byte = GPT_BYTE; | |
898 | } | |
899 | else | |
900 | stop_byte = CHAR_TO_BYTE (stop); | |
901 | ||
902 | from_byte = CHAR_TO_BYTE (from); | |
903 | ||
028d516b | 904 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
4ed46869 KH |
905 | while (1) |
906 | { | |
2e344af3 KH |
907 | find_charset_in_text (BYTE_POS_ADDR (from_byte), stop - from, |
908 | stop_byte - from_byte, charsets, table); | |
4ed46869 | 909 | if (stop < to) |
6ae1f27e RS |
910 | { |
911 | from = stop, from_byte = stop_byte; | |
912 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
913 | } | |
4ed46869 KH |
914 | else |
915 | break; | |
916 | } | |
6ae1f27e | 917 | |
4ed46869 | 918 | val = Qnil; |
2e344af3 | 919 | if (charsets[1]) |
0282eb69 | 920 | val = Fcons (Qunknown, val); |
2e344af3 KH |
921 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
922 | if (charsets[i]) | |
923 | val = Fcons (CHARSET_SYMBOL (i), val); | |
924 | if (charsets[0]) | |
925 | val = Fcons (Qascii, val); | |
4ed46869 KH |
926 | return val; |
927 | } | |
928 | ||
929 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 | 930 | 1, 2, 0, |
fdb82f93 PJ |
931 | doc: /* Return a list of charsets in STR. |
932 | Optional arg TABLE if non-nil is a translation table to look up. | |
933 | ||
934 | If the string contains invalid multibyte characters, | |
935 | `unknown' is included in the returned list. | |
936 | ||
937 | If STR is unibyte, the returned list may contain | |
938 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) | |
939 | (str, table) | |
23d2a7f1 | 940 | Lisp_Object str, table; |
4ed46869 | 941 | { |
a29e3b1b | 942 | int charsets[MAX_CHARSET + 1]; |
4ed46869 KH |
943 | int i; |
944 | Lisp_Object val; | |
945 | ||
b7826503 | 946 | CHECK_STRING (str); |
87b089ad | 947 | |
a29e3b1b | 948 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
d5db4077 KR |
949 | find_charset_in_text (SDATA (str), SCHARS (str), |
950 | SBYTES (str), charsets, table); | |
2e344af3 | 951 | |
4ed46869 | 952 | val = Qnil; |
2e344af3 | 953 | if (charsets[1]) |
0282eb69 | 954 | val = Fcons (Qunknown, val); |
2e344af3 KH |
955 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
956 | if (charsets[i]) | |
957 | val = Fcons (CHARSET_SYMBOL (i), val); | |
958 | if (charsets[0]) | |
959 | val = Fcons (Qascii, val); | |
4ed46869 KH |
960 | return val; |
961 | } | |
2e344af3 | 962 | |
4ed46869 KH |
963 | \f |
964 | DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
fdb82f93 PJ |
965 | doc: /* Return a character made from arguments. |
966 | Internal use only. */) | |
967 | (charset, code1, code2) | |
4ed46869 KH |
968 | Lisp_Object charset, code1, code2; |
969 | { | |
ac4137cc KH |
970 | int charset_id, c1, c2; |
971 | ||
b7826503 | 972 | CHECK_NUMBER (charset); |
ac4137cc KH |
973 | charset_id = XINT (charset); |
974 | if (!CHARSET_DEFINED_P (charset_id)) | |
975 | error ("Invalid charset ID: %d", XINT (charset)); | |
4ed46869 KH |
976 | |
977 | if (NILP (code1)) | |
ac4137cc | 978 | c1 = 0; |
4ed46869 | 979 | else |
ac4137cc | 980 | { |
b7826503 | 981 | CHECK_NUMBER (code1); |
ac4137cc KH |
982 | c1 = XINT (code1); |
983 | } | |
4ed46869 | 984 | if (NILP (code2)) |
ac4137cc | 985 | c2 = 0; |
4ed46869 | 986 | else |
ac4137cc | 987 | { |
b7826503 | 988 | CHECK_NUMBER (code2); |
ac4137cc KH |
989 | c2 = XINT (code2); |
990 | } | |
4ed46869 | 991 | |
2e344af3 KH |
992 | if (charset_id == CHARSET_ASCII) |
993 | { | |
994 | if (c1 < 0 || c1 > 0x7F) | |
995 | goto invalid_code_posints; | |
996 | return make_number (c1); | |
997 | } | |
998 | else if (charset_id == CHARSET_8_BIT_CONTROL) | |
999 | { | |
30736012 KH |
1000 | if (NILP (code1)) |
1001 | c1 = 0x80; | |
1002 | else if (c1 < 0x80 || c1 > 0x9F) | |
2e344af3 KH |
1003 | goto invalid_code_posints; |
1004 | return make_number (c1); | |
1005 | } | |
1006 | else if (charset_id == CHARSET_8_BIT_GRAPHIC) | |
1007 | { | |
30736012 KH |
1008 | if (NILP (code1)) |
1009 | c1 = 0xA0; | |
1010 | else if (c1 < 0xA0 || c1 > 0xFF) | |
2e344af3 KH |
1011 | goto invalid_code_posints; |
1012 | return make_number (c1); | |
1013 | } | |
1014 | else if (c1 < 0 || c1 > 0xFF || c2 < 0 || c2 > 0xFF) | |
1015 | goto invalid_code_posints; | |
ac4137cc KH |
1016 | c1 &= 0x7F; |
1017 | c2 &= 0x7F; | |
1018 | if (c1 == 0 | |
1019 | ? c2 != 0 | |
1020 | : (c2 == 0 | |
87f67317 KR |
1021 | ? !CHAR_COMPONENTS_VALID_P (charset_id, c1, 0x20) |
1022 | : !CHAR_COMPONENTS_VALID_P (charset_id, c1, c2))) | |
2e344af3 | 1023 | goto invalid_code_posints; |
ac4137cc | 1024 | return make_number (MAKE_CHAR (charset_id, c1, c2)); |
2e344af3 KH |
1025 | |
1026 | invalid_code_posints: | |
1027 | error ("Invalid code points for charset ID %d: %d %d", charset_id, c1, c2); | |
4ed46869 KH |
1028 | } |
1029 | ||
1030 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, | |
bc001814 JB |
1031 | doc: /* Return list of charset and one or two position-codes of CH. |
1032 | If CH is invalid as a character code, | |
1033 | return a list of symbol `unknown' and CH. */) | |
fdb82f93 | 1034 | (ch) |
4ed46869 KH |
1035 | Lisp_Object ch; |
1036 | { | |
0282eb69 | 1037 | int c, charset, c1, c2; |
4ed46869 | 1038 | |
b7826503 | 1039 | CHECK_NUMBER (ch); |
0282eb69 KH |
1040 | c = XFASTINT (ch); |
1041 | if (!CHAR_VALID_P (c, 1)) | |
1042 | return Fcons (Qunknown, Fcons (ch, Qnil)); | |
4ed46869 | 1043 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
6dc0722d | 1044 | return (c2 >= 0 |
4ed46869 KH |
1045 | ? Fcons (CHARSET_SYMBOL (charset), |
1046 | Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
1047 | : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
1048 | } | |
1049 | ||
1050 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
bc001814 | 1051 | doc: /* Return charset of CH. */) |
fdb82f93 | 1052 | (ch) |
4ed46869 KH |
1053 | Lisp_Object ch; |
1054 | { | |
b7826503 | 1055 | CHECK_NUMBER (ch); |
4ed46869 KH |
1056 | |
1057 | return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
1058 | } | |
1059 | ||
90d7b74e | 1060 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
fdb82f93 PJ |
1061 | doc: /* Return charset of a character in the current buffer at position POS. |
1062 | If POS is nil, it defauls to the current point. | |
1063 | If POS is out of range, the value is nil. */) | |
1064 | (pos) | |
90d7b74e KH |
1065 | Lisp_Object pos; |
1066 | { | |
2e344af3 KH |
1067 | Lisp_Object ch; |
1068 | int charset; | |
ac4137cc | 1069 | |
2e344af3 KH |
1070 | ch = Fchar_after (pos); |
1071 | if (! INTEGERP (ch)) | |
1072 | return ch; | |
1073 | charset = CHAR_CHARSET (XINT (ch)); | |
90d7b74e KH |
1074 | return CHARSET_SYMBOL (charset); |
1075 | } | |
1076 | ||
4ed46869 | 1077 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
fdb82f93 PJ |
1078 | doc: /* Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR. |
1079 | ||
1080 | ISO 2022's designation sequence (escape sequence) distinguishes charsets | |
1081 | by their DIMENSION, CHARS, and FINAL-CHAR, | |
1082 | where as Emacs distinguishes them by charset symbol. | |
1083 | See the documentation of the function `charset-info' for the meanings of | |
1084 | DIMENSION, CHARS, and FINAL-CHAR. */) | |
1085 | (dimension, chars, final_char) | |
4ed46869 KH |
1086 | Lisp_Object dimension, chars, final_char; |
1087 | { | |
1088 | int charset; | |
1089 | ||
b7826503 PJ |
1090 | CHECK_NUMBER (dimension); |
1091 | CHECK_NUMBER (chars); | |
1092 | CHECK_NUMBER (final_char); | |
4ed46869 KH |
1093 | |
1094 | if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
1095 | return Qnil; | |
1096 | return CHARSET_SYMBOL (charset); | |
1097 | } | |
1098 | ||
9d3d8cba KH |
1099 | /* If GENERICP is nonzero, return nonzero iff C is a valid normal or |
1100 | generic character. If GENERICP is zero, return nonzero iff C is a | |
1101 | valid normal character. Do not call this function directly, | |
1102 | instead use macro CHAR_VALID_P. */ | |
1103 | int | |
1104 | char_valid_p (c, genericp) | |
1105 | int c, genericp; | |
1106 | { | |
1107 | int charset, c1, c2; | |
1108 | ||
0e235b7e | 1109 | if (c < 0 || c >= MAX_CHAR) |
9d3d8cba KH |
1110 | return 0; |
1111 | if (SINGLE_BYTE_CHAR_P (c)) | |
1112 | return 1; | |
2e344af3 | 1113 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
1114 | if (genericp) |
1115 | { | |
1116 | if (c1) | |
1117 | { | |
1118 | if (c2 <= 0) c2 = 0x20; | |
1119 | } | |
1120 | else | |
1121 | { | |
1122 | if (c2 <= 0) c1 = c2 = 0x20; | |
1123 | } | |
1124 | } | |
1125 | return (CHARSET_DEFINED_P (charset) | |
44c6492d | 1126 | && CHAR_COMPONENTS_VALID_P (charset, c1, c2)); |
9d3d8cba KH |
1127 | } |
1128 | ||
1129 | DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0, | |
fdb82f93 PJ |
1130 | doc: /* Return t if OBJECT is a valid normal character. |
1131 | If optional arg GENERICP is non-nil, also return t if OBJECT is | |
1132 | a valid generic character. */) | |
1133 | (object, genericp) | |
9d3d8cba KH |
1134 | Lisp_Object object, genericp; |
1135 | { | |
1136 | if (! NATNUMP (object)) | |
1137 | return Qnil; | |
1138 | return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil); | |
1139 | } | |
1140 | ||
d2665018 KH |
1141 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, |
1142 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
fdb82f93 PJ |
1143 | doc: /* Convert the unibyte character CH to multibyte character. |
1144 | The conversion is done based on `nonascii-translation-table' (which see) | |
1145 | or `nonascii-insert-offset' (which see). */) | |
1146 | (ch) | |
d2665018 KH |
1147 | Lisp_Object ch; |
1148 | { | |
1149 | int c; | |
1150 | ||
b7826503 | 1151 | CHECK_NUMBER (ch); |
d2665018 KH |
1152 | c = XINT (ch); |
1153 | if (c < 0 || c >= 0400) | |
1154 | error ("Invalid unibyte character: %d", c); | |
1155 | c = unibyte_char_to_multibyte (c); | |
1156 | if (c < 0) | |
1157 | error ("Can't convert to multibyte character: %d", XINT (ch)); | |
1158 | return make_number (c); | |
1159 | } | |
1160 | ||
1bcc1567 RS |
1161 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, |
1162 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
fdb82f93 PJ |
1163 | doc: /* Convert the multibyte character CH to unibyte character. |
1164 | The conversion is done based on `nonascii-translation-table' (which see) | |
1165 | or `nonascii-insert-offset' (which see). */) | |
1166 | (ch) | |
1bcc1567 RS |
1167 | Lisp_Object ch; |
1168 | { | |
1169 | int c; | |
1170 | ||
b7826503 | 1171 | CHECK_NUMBER (ch); |
1bcc1567 | 1172 | c = XINT (ch); |
ac4137cc | 1173 | if (! CHAR_VALID_P (c, 0)) |
1bcc1567 RS |
1174 | error ("Invalid multibyte character: %d", c); |
1175 | c = multibyte_char_to_unibyte (c, Qnil); | |
1176 | if (c < 0) | |
1177 | error ("Can't convert to unibyte character: %d", XINT (ch)); | |
1178 | return make_number (c); | |
1179 | } | |
1180 | ||
4ed46869 | 1181 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, |
bc001814 | 1182 | doc: /* Return 1 regardless of the argument CH. */) |
fdb82f93 | 1183 | (ch) |
4ed46869 KH |
1184 | Lisp_Object ch; |
1185 | { | |
b7826503 | 1186 | CHECK_NUMBER (ch); |
9b6a601f KH |
1187 | return make_number (1); |
1188 | } | |
1189 | ||
1190 | /* Return how many bytes C will occupy in a multibyte buffer. | |
1191 | Don't call this function directly, instead use macro CHAR_BYTES. */ | |
1192 | int | |
1193 | char_bytes (c) | |
1194 | int c; | |
1195 | { | |
99529c2c | 1196 | int charset; |
9b6a601f | 1197 | |
2e344af3 KH |
1198 | if (ASCII_BYTE_P (c) || (c & ~((1 << CHARACTERBITS) -1))) |
1199 | return 1; | |
1200 | if (SINGLE_BYTE_CHAR_P (c) && c >= 0xA0) | |
8ac5a9cc KH |
1201 | return 1; |
1202 | ||
99529c2c KH |
1203 | charset = CHAR_CHARSET (c); |
1204 | return (CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1); | |
4ed46869 KH |
1205 | } |
1206 | ||
1207 | /* Return the width of character of which multi-byte form starts with | |
1208 | C. The width is measured by how many columns occupied on the | |
1209 | screen when displayed in the current buffer. */ | |
1210 | ||
1211 | #define ONE_BYTE_CHAR_WIDTH(c) \ | |
1212 | (c < 0x20 \ | |
1213 | ? (c == '\t' \ | |
53316e55 | 1214 | ? XFASTINT (current_buffer->tab_width) \ |
4ed46869 KH |
1215 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
1216 | : (c < 0x7f \ | |
1217 | ? 1 \ | |
1218 | : (c == 0x7F \ | |
1219 | ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
1220 | : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
1221 | && BASE_LEADING_CODE_P (c)) \ | |
1222 | ? WIDTH_BY_CHAR_HEAD (c) \ | |
b4e9dd77 | 1223 | : 4)))) |
4ed46869 KH |
1224 | |
1225 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
bc001814 | 1226 | doc: /* Return width of CH when displayed in the current buffer. |
fdb82f93 PJ |
1227 | The width is measured by how many columns it occupies on the screen. |
1228 | Tab is taken to occupy `tab-width' columns. */) | |
1229 | (ch) | |
1230 | Lisp_Object ch; | |
4ed46869 | 1231 | { |
859f2b3c | 1232 | Lisp_Object val, disp; |
4ed46869 | 1233 | int c; |
51c4025f | 1234 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1235 | |
b7826503 | 1236 | CHECK_NUMBER (ch); |
4ed46869 | 1237 | |
859f2b3c RS |
1238 | c = XINT (ch); |
1239 | ||
1240 | /* Get the way the display table would display it. */ | |
51c4025f | 1241 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; |
859f2b3c RS |
1242 | |
1243 | if (VECTORP (disp)) | |
1244 | XSETINT (val, XVECTOR (disp)->size); | |
1245 | else if (SINGLE_BYTE_CHAR_P (c)) | |
1246 | XSETINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
4ed46869 KH |
1247 | else |
1248 | { | |
1249 | int charset = CHAR_CHARSET (c); | |
1250 | ||
1251 | XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
1252 | } | |
1253 | return val; | |
1254 | } | |
1255 | ||
1256 | /* Return width of string STR of length LEN when displayed in the | |
1257 | current buffer. The width is measured by how many columns it | |
1258 | occupies on the screen. */ | |
859f2b3c | 1259 | |
4ed46869 KH |
1260 | int |
1261 | strwidth (str, len) | |
1262 | unsigned char *str; | |
1263 | int len; | |
1264 | { | |
beeedaad KH |
1265 | return c_string_width (str, len, -1, NULL, NULL); |
1266 | } | |
1267 | ||
1268 | /* Return width of string STR of length LEN when displayed in the | |
1269 | current buffer. The width is measured by how many columns it | |
1270 | occupies on the screen. If PRECISION > 0, return the width of | |
1271 | longest substring that doesn't exceed PRECISION, and set number of | |
1272 | characters and bytes of the substring in *NCHARS and *NBYTES | |
1273 | respectively. */ | |
1274 | ||
1275 | int | |
1276 | c_string_width (str, len, precision, nchars, nbytes) | |
fb4f7f50 | 1277 | const unsigned char *str; |
020ea113 | 1278 | int len, precision, *nchars, *nbytes; |
beeedaad KH |
1279 | { |
1280 | int i = 0, i_byte = 0; | |
4ed46869 | 1281 | int width = 0; |
beeedaad | 1282 | int chars; |
c4a4e28f | 1283 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1284 | |
beeedaad | 1285 | while (i_byte < len) |
859f2b3c | 1286 | { |
beeedaad KH |
1287 | int bytes, thiswidth; |
1288 | Lisp_Object val; | |
859f2b3c | 1289 | |
99529c2c | 1290 | if (dp) |
beeedaad KH |
1291 | { |
1292 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
1293 | ||
1294 | chars = 1; | |
1295 | val = DISP_CHAR_VECTOR (dp, c); | |
1296 | if (VECTORP (val)) | |
1297 | thiswidth = XVECTOR (val)->size; | |
1298 | else | |
1299 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1300 | } | |
859f2b3c | 1301 | else |
beeedaad KH |
1302 | { |
1303 | chars = 1; | |
1304 | PARSE_MULTIBYTE_SEQ (str + i_byte, len - i_byte, bytes); | |
1305 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1306 | } | |
99529c2c | 1307 | |
beeedaad KH |
1308 | if (precision > 0 |
1309 | && (width + thiswidth > precision)) | |
1310 | { | |
1311 | *nchars = i; | |
1312 | *nbytes = i_byte; | |
1313 | return width; | |
1314 | } | |
1315 | i++; | |
1316 | i_byte += bytes; | |
1317 | width += thiswidth; | |
1318 | } | |
99529c2c | 1319 | |
beeedaad KH |
1320 | if (precision > 0) |
1321 | { | |
1322 | *nchars = i; | |
1323 | *nbytes = i_byte; | |
859f2b3c | 1324 | } |
beeedaad | 1325 | |
4ed46869 KH |
1326 | return width; |
1327 | } | |
1328 | ||
beeedaad KH |
1329 | /* Return width of Lisp string STRING when displayed in the current |
1330 | buffer. The width is measured by how many columns it occupies on | |
1331 | the screen while paying attention to compositions. If PRECISION > | |
1332 | 0, return the width of longest substring that doesn't exceed | |
1333 | PRECISION, and set number of characters and bytes of the substring | |
1334 | in *NCHARS and *NBYTES respectively. */ | |
1335 | ||
3f62427c | 1336 | int |
beeedaad KH |
1337 | lisp_string_width (string, precision, nchars, nbytes) |
1338 | Lisp_Object string; | |
1339 | int precision, *nchars, *nbytes; | |
3f62427c | 1340 | { |
d5db4077 KR |
1341 | int len = SCHARS (string); |
1342 | int len_byte = SBYTES (string); | |
fb4f7f50 | 1343 | const unsigned char *str = SDATA (string); |
beeedaad | 1344 | int i = 0, i_byte = 0; |
3f62427c | 1345 | int width = 0; |
beeedaad | 1346 | struct Lisp_Char_Table *dp = buffer_display_table (); |
3f62427c KH |
1347 | |
1348 | while (i < len) | |
1349 | { | |
beeedaad KH |
1350 | int chars, bytes, thiswidth; |
1351 | Lisp_Object val; | |
1352 | int cmp_id; | |
1353 | int ignore, end; | |
1354 | ||
1355 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
1356 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
1357 | >= 0)) | |
3f62427c | 1358 | { |
beeedaad KH |
1359 | thiswidth = composition_table[cmp_id]->width; |
1360 | chars = end - i; | |
1361 | bytes = string_char_to_byte (string, end) - i_byte; | |
1362 | } | |
1363 | else if (dp) | |
1364 | { | |
1365 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
1366 | ||
1367 | chars = 1; | |
1368 | val = DISP_CHAR_VECTOR (dp, c); | |
1369 | if (VECTORP (val)) | |
1370 | thiswidth = XVECTOR (val)->size; | |
1371 | else | |
1372 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
3f62427c KH |
1373 | } |
1374 | else | |
1375 | { | |
beeedaad KH |
1376 | chars = 1; |
1377 | PARSE_MULTIBYTE_SEQ (str + i_byte, len_byte - i_byte, bytes); | |
1378 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1379 | } | |
1380 | ||
1381 | if (precision > 0 | |
1382 | && (width + thiswidth > precision)) | |
1383 | { | |
1384 | *nchars = i; | |
1385 | *nbytes = i_byte; | |
1386 | return width; | |
3f62427c | 1387 | } |
beeedaad KH |
1388 | i += chars; |
1389 | i_byte += bytes; | |
1390 | width += thiswidth; | |
1391 | } | |
1392 | ||
1393 | if (precision > 0) | |
1394 | { | |
1395 | *nchars = i; | |
1396 | *nbytes = i_byte; | |
3f62427c | 1397 | } |
beeedaad | 1398 | |
3f62427c KH |
1399 | return width; |
1400 | } | |
1401 | ||
4ed46869 | 1402 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, |
fdb82f93 PJ |
1403 | doc: /* Return width of STRING when displayed in the current buffer. |
1404 | Width is measured by how many columns it occupies on the screen. | |
1405 | When calculating width of a multibyte character in STRING, | |
1406 | only the base leading-code is considered; the validity of | |
1407 | the following bytes is not checked. Tabs in STRING are always | |
1408 | taken to occupy `tab-width' columns. */) | |
bc001814 JB |
1409 | (string) |
1410 | Lisp_Object string; | |
4ed46869 KH |
1411 | { |
1412 | Lisp_Object val; | |
1413 | ||
bc001814 JB |
1414 | CHECK_STRING (string); |
1415 | XSETFASTINT (val, lisp_string_width (string, -1, NULL, NULL)); | |
4ed46869 KH |
1416 | return val; |
1417 | } | |
1418 | ||
1419 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
bc001814 | 1420 | doc: /* Return the direction of CH. |
fdb82f93 PJ |
1421 | The returned value is 0 for left-to-right and 1 for right-to-left. */) |
1422 | (ch) | |
4ed46869 KH |
1423 | Lisp_Object ch; |
1424 | { | |
1425 | int charset; | |
1426 | ||
b7826503 | 1427 | CHECK_NUMBER (ch); |
4ed46869 KH |
1428 | charset = CHAR_CHARSET (XFASTINT (ch)); |
1429 | if (!CHARSET_DEFINED_P (charset)) | |
93bcb785 | 1430 | invalid_character (XINT (ch)); |
4ed46869 KH |
1431 | return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); |
1432 | } | |
1433 | ||
87b089ad RS |
1434 | /* Return the number of characters in the NBYTES bytes at PTR. |
1435 | This works by looking at the contents and checking for multibyte sequences. | |
1436 | However, if the current buffer has enable-multibyte-characters = nil, | |
1437 | we treat each byte as a character. */ | |
1438 | ||
6ae1f27e RS |
1439 | int |
1440 | chars_in_text (ptr, nbytes) | |
fb4f7f50 | 1441 | const unsigned char *ptr; |
6ae1f27e RS |
1442 | int nbytes; |
1443 | { | |
87b089ad RS |
1444 | /* current_buffer is null at early stages of Emacs initialization. */ |
1445 | if (current_buffer == 0 | |
1446 | || NILP (current_buffer->enable_multibyte_characters)) | |
6ae1f27e | 1447 | return nbytes; |
a8a35e61 | 1448 | |
ac4137cc | 1449 | return multibyte_chars_in_text (ptr, nbytes); |
046b1f03 RS |
1450 | } |
1451 | ||
87b089ad RS |
1452 | /* Return the number of characters in the NBYTES bytes at PTR. |
1453 | This works by looking at the contents and checking for multibyte sequences. | |
1454 | It ignores enable-multibyte-characters. */ | |
1455 | ||
1456 | int | |
1457 | multibyte_chars_in_text (ptr, nbytes) | |
fb4f7f50 | 1458 | const unsigned char *ptr; |
87b089ad RS |
1459 | int nbytes; |
1460 | { | |
fb4f7f50 | 1461 | const unsigned char *endp; |
ac4137cc | 1462 | int chars, bytes; |
87b089ad RS |
1463 | |
1464 | endp = ptr + nbytes; | |
1465 | chars = 0; | |
1466 | ||
1467 | while (ptr < endp) | |
1468 | { | |
2e344af3 KH |
1469 | PARSE_MULTIBYTE_SEQ (ptr, endp - ptr, bytes); |
1470 | ptr += bytes; | |
1471 | chars++; | |
1472 | } | |
1473 | ||
1474 | return chars; | |
1475 | } | |
1476 | ||
d8e4f486 | 1477 | /* Parse unibyte text at STR of LEN bytes as multibyte text, and |
2e344af3 | 1478 | count the numbers of characters and bytes in it. On counting |
d8e4f486 DL |
1479 | bytes, pay attention to the fact that 8-bit characters in the range |
1480 | 0x80..0x9F are represented by 2 bytes in multibyte text. */ | |
2e344af3 KH |
1481 | void |
1482 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
fb4f7f50 | 1483 | const unsigned char *str; |
2e344af3 KH |
1484 | int len, *nchars, *nbytes; |
1485 | { | |
fb4f7f50 | 1486 | const unsigned char *endp = str + len; |
2e344af3 KH |
1487 | int n, chars = 0, bytes = 0; |
1488 | ||
1489 | while (str < endp) | |
1490 | { | |
1491 | if (UNIBYTE_STR_AS_MULTIBYTE_P (str, endp - str, n)) | |
1492 | str += n, bytes += n; | |
1493 | else | |
1494 | str++, bytes += 2; | |
1495 | chars++; | |
1496 | } | |
1497 | *nchars = chars; | |
1498 | *nbytes = bytes; | |
1499 | return; | |
1500 | } | |
1501 | ||
d8e4f486 | 1502 | /* Arrange unibyte text at STR of NBYTES bytes as multibyte text. |
2e344af3 KH |
1503 | It actually converts only 8-bit characters in the range 0x80..0x9F |
1504 | that don't contruct multibyte characters to multibyte forms. If | |
1505 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
1506 | text. It is assured that we can use LEN bytes at STR as a work | |
1507 | area and that is enough. Return the number of bytes of the | |
1508 | resulting text. */ | |
1509 | ||
1510 | int | |
1511 | str_as_multibyte (str, len, nbytes, nchars) | |
1512 | unsigned char *str; | |
1513 | int len, nbytes, *nchars; | |
1514 | { | |
1515 | unsigned char *p = str, *endp = str + nbytes; | |
1516 | unsigned char *to; | |
1517 | int chars = 0; | |
1518 | int n; | |
1519 | ||
1520 | while (p < endp && UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
1521 | p += n, chars++; | |
1522 | if (nchars) | |
1523 | *nchars = chars; | |
1524 | if (p == endp) | |
1525 | return nbytes; | |
1526 | ||
1527 | to = p; | |
1528 | nbytes = endp - p; | |
1529 | endp = str + len; | |
1530 | safe_bcopy (p, endp - nbytes, nbytes); | |
1531 | p = endp - nbytes; | |
1532 | while (p < endp) | |
1533 | { | |
1534 | if (UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
ac4137cc | 1535 | { |
2e344af3 KH |
1536 | while (n--) |
1537 | *to++ = *p++; | |
177c0ea7 | 1538 | } |
ac4137cc KH |
1539 | else |
1540 | { | |
2e344af3 KH |
1541 | *to++ = LEADING_CODE_8_BIT_CONTROL; |
1542 | *to++ = *p++ + 0x20; | |
ac4137cc | 1543 | } |
87b089ad RS |
1544 | chars++; |
1545 | } | |
2e344af3 KH |
1546 | if (nchars) |
1547 | *nchars = chars; | |
1548 | return (to - str); | |
1549 | } | |
87b089ad | 1550 | |
740f080d KH |
1551 | /* Parse unibyte string at STR of LEN bytes, and return the number of |
1552 | bytes it may ocupy when converted to multibyte string by | |
1553 | `str_to_multibyte'. */ | |
1554 | ||
1555 | int | |
1556 | parse_str_to_multibyte (str, len) | |
1557 | unsigned char *str; | |
1558 | int len; | |
1559 | { | |
1560 | unsigned char *endp = str + len; | |
1561 | int bytes; | |
1562 | ||
1563 | for (bytes = 0; str < endp; str++) | |
1564 | bytes += (*str < 0x80 || *str >= 0xA0) ? 1 : 2; | |
1565 | return bytes; | |
1566 | } | |
1567 | ||
d8e4f486 | 1568 | /* Convert unibyte text at STR of NBYTES bytes to multibyte text |
2e344af3 KH |
1569 | that contains the same single-byte characters. It actually |
1570 | converts all 8-bit characters to multibyte forms. It is assured | |
1571 | that we can use LEN bytes at STR as a work area and that is | |
1572 | enough. */ | |
1573 | ||
1574 | int | |
1575 | str_to_multibyte (str, len, bytes) | |
1576 | unsigned char *str; | |
1577 | int len, bytes; | |
1578 | { | |
1579 | unsigned char *p = str, *endp = str + bytes; | |
1580 | unsigned char *to; | |
2e344af3 KH |
1581 | |
1582 | while (p < endp && (*p < 0x80 || *p >= 0xA0)) p++; | |
1583 | if (p == endp) | |
1584 | return bytes; | |
1585 | to = p; | |
1586 | bytes = endp - p; | |
1587 | endp = str + len; | |
1588 | safe_bcopy (p, endp - bytes, bytes); | |
1589 | p = endp - bytes; | |
177c0ea7 | 1590 | while (p < endp) |
2e344af3 KH |
1591 | { |
1592 | if (*p < 0x80 || *p >= 0xA0) | |
1593 | *to++ = *p++; | |
1594 | else | |
1595 | *to++ = LEADING_CODE_8_BIT_CONTROL, *to++ = *p++ + 0x20; | |
1596 | } | |
1597 | return (to - str); | |
87b089ad RS |
1598 | } |
1599 | ||
2e344af3 KH |
1600 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It |
1601 | actually converts only 8-bit characters in the range 0x80..0x9F to | |
1602 | unibyte forms. */ | |
1603 | ||
1604 | int | |
1605 | str_as_unibyte (str, bytes) | |
1606 | unsigned char *str; | |
1607 | int bytes; | |
1608 | { | |
1609 | unsigned char *p = str, *endp = str + bytes; | |
1610 | unsigned char *to = str; | |
1611 | ||
1612 | while (p < endp && *p != LEADING_CODE_8_BIT_CONTROL) p++; | |
1613 | to = p; | |
177c0ea7 | 1614 | while (p < endp) |
2e344af3 KH |
1615 | { |
1616 | if (*p == LEADING_CODE_8_BIT_CONTROL) | |
1617 | *to++ = *(p + 1) - 0x20, p += 2; | |
1618 | else | |
1619 | *to++ = *p++; | |
1620 | } | |
1621 | return (to - str); | |
1622 | } | |
1623 | ||
1624 | \f | |
4bc26a6c | 1625 | DEFUN ("string", Fstring, Sstring, 0, MANY, 0, |
5f5d497e MB |
1626 | doc: /* Concatenate all the argument characters and make the result a string. |
1627 | usage: (string &rest CHARACTERS) */) | |
fdb82f93 | 1628 | (n, args) |
53316e55 | 1629 | int n; |
4ed46869 KH |
1630 | Lisp_Object *args; |
1631 | { | |
4894eddd KS |
1632 | int i, bufsize; |
1633 | unsigned char *buf, *p; | |
2e344af3 | 1634 | int c; |
5729c92f | 1635 | int multibyte = 0; |
4894eddd KS |
1636 | Lisp_Object ret; |
1637 | USE_SAFE_ALLOCA; | |
1638 | ||
1639 | bufsize = MAX_MULTIBYTE_LENGTH * n; | |
1640 | SAFE_ALLOCA (buf, unsigned char *, bufsize); | |
1641 | p = buf; | |
4ed46869 KH |
1642 | |
1643 | for (i = 0; i < n; i++) | |
1644 | { | |
b7826503 | 1645 | CHECK_NUMBER (args[i]); |
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 | ||
4894eddd | 1659 | ret = make_string_from_bytes (buf, n, p - buf); |
233f3db6 | 1660 | SAFE_FREE (); |
4894eddd KS |
1661 | |
1662 | return ret; | |
4ed46869 KH |
1663 | } |
1664 | ||
1665 | #endif /* emacs */ | |
1666 | \f | |
dfcf069d | 1667 | int |
4ed46869 KH |
1668 | charset_id_internal (charset_name) |
1669 | char *charset_name; | |
1670 | { | |
76d7b829 | 1671 | Lisp_Object val; |
4ed46869 | 1672 | |
76d7b829 | 1673 | val= Fget (intern (charset_name), Qcharset); |
4ed46869 KH |
1674 | if (!VECTORP (val)) |
1675 | error ("Charset %s is not defined", charset_name); | |
1676 | ||
1677 | return (XINT (XVECTOR (val)->contents[0])); | |
1678 | } | |
1679 | ||
1680 | DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
fdb82f93 PJ |
1681 | Ssetup_special_charsets, 0, 0, 0, doc: /* Internal use only. */) |
1682 | () | |
4ed46869 KH |
1683 | { |
1684 | charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1685 | charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1686 | charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1687 | charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1688 | charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1689 | charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1690 | charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1691 | return Qnil; | |
1692 | } | |
1693 | ||
dfcf069d | 1694 | void |
4ed46869 KH |
1695 | init_charset_once () |
1696 | { | |
1697 | int i, j, k; | |
1698 | ||
1699 | staticpro (&Vcharset_table); | |
1700 | staticpro (&Vcharset_symbol_table); | |
8a73a704 | 1701 | staticpro (&Vgeneric_character_list); |
4ed46869 KH |
1702 | |
1703 | /* This has to be done here, before we call Fmake_char_table. */ | |
1704 | Qcharset_table = intern ("charset-table"); | |
1705 | staticpro (&Qcharset_table); | |
1706 | ||
1707 | /* Intern this now in case it isn't already done. | |
1708 | Setting this variable twice is harmless. | |
1709 | But don't staticpro it here--that is done in alloc.c. */ | |
1710 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1711 | ||
1712 | /* Now we are ready to set up this property, so we can | |
1713 | create the charset table. */ | |
1714 | Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1715 | Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1716 | ||
0282eb69 KH |
1717 | Qunknown = intern ("unknown"); |
1718 | staticpro (&Qunknown); | |
1719 | Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), | |
1720 | Qunknown); | |
4ed46869 KH |
1721 | |
1722 | /* Setup tables. */ | |
1723 | for (i = 0; i < 2; i++) | |
1724 | for (j = 0; j < 2; j++) | |
1725 | for (k = 0; k < 128; k++) | |
1726 | iso_charset_table [i][j][k] = -1; | |
1727 | ||
60383934 | 1728 | for (i = 0; i < 256; i++) |
2e344af3 | 1729 | bytes_by_char_head[i] = 1; |
2e344af3 KH |
1730 | bytes_by_char_head[LEADING_CODE_PRIVATE_11] = 3; |
1731 | bytes_by_char_head[LEADING_CODE_PRIVATE_12] = 3; | |
1732 | bytes_by_char_head[LEADING_CODE_PRIVATE_21] = 4; | |
1733 | bytes_by_char_head[LEADING_CODE_PRIVATE_22] = 4; | |
4ed46869 KH |
1734 | |
1735 | for (i = 0; i < 128; i++) | |
2e344af3 | 1736 | width_by_char_head[i] = 1; |
4ed46869 | 1737 | for (; i < 256; i++) |
2e344af3 KH |
1738 | width_by_char_head[i] = 4; |
1739 | width_by_char_head[LEADING_CODE_PRIVATE_11] = 1; | |
1740 | width_by_char_head[LEADING_CODE_PRIVATE_12] = 2; | |
1741 | width_by_char_head[LEADING_CODE_PRIVATE_21] = 1; | |
1742 | width_by_char_head[LEADING_CODE_PRIVATE_22] = 2; | |
8a73a704 KH |
1743 | |
1744 | { | |
76d7b829 | 1745 | Lisp_Object val; |
8a73a704 | 1746 | |
76d7b829 | 1747 | val = Qnil; |
8a73a704 KH |
1748 | for (i = 0x81; i < 0x90; i++) |
1749 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1750 | for (; i < 0x9A; i++) | |
1751 | val = Fcons (make_number ((i - 0x8F) << 14), val); | |
1752 | for (i = 0xA0; i < 0xF0; i++) | |
1753 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1754 | for (; i < 0xFF; i++) | |
1755 | val = Fcons (make_number ((i - 0xE0) << 14), val); | |
8a73a704 KH |
1756 | Vgeneric_character_list = Fnreverse (val); |
1757 | } | |
bbf12bb3 KH |
1758 | |
1759 | nonascii_insert_offset = 0; | |
1760 | Vnonascii_translation_table = Qnil; | |
4ed46869 KH |
1761 | } |
1762 | ||
1763 | #ifdef emacs | |
1764 | ||
dfcf069d | 1765 | void |
4ed46869 KH |
1766 | syms_of_charset () |
1767 | { | |
2e344af3 KH |
1768 | Qcharset = intern ("charset"); |
1769 | staticpro (&Qcharset); | |
1770 | ||
4ed46869 KH |
1771 | Qascii = intern ("ascii"); |
1772 | staticpro (&Qascii); | |
1773 | ||
2e344af3 KH |
1774 | Qeight_bit_control = intern ("eight-bit-control"); |
1775 | staticpro (&Qeight_bit_control); | |
1776 | ||
1777 | Qeight_bit_graphic = intern ("eight-bit-graphic"); | |
1778 | staticpro (&Qeight_bit_graphic); | |
4ed46869 | 1779 | |
2e344af3 KH |
1780 | /* Define special charsets ascii, eight-bit-control, and |
1781 | eight-bit-graphic. */ | |
4ed46869 KH |
1782 | update_charset_table (make_number (CHARSET_ASCII), |
1783 | make_number (1), make_number (94), | |
1784 | make_number (1), | |
1785 | make_number (0), | |
1786 | make_number ('B'), | |
1787 | make_number (0), | |
1788 | build_string ("ASCII"), | |
d78bc582 | 1789 | Qnil, /* same as above */ |
4ed46869 KH |
1790 | build_string ("ASCII (ISO646 IRV)")); |
1791 | CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1792 | Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1793 | ||
2e344af3 KH |
1794 | update_charset_table (make_number (CHARSET_8_BIT_CONTROL), |
1795 | make_number (1), make_number (96), | |
797a084a | 1796 | make_number (4), |
2e344af3 KH |
1797 | make_number (0), |
1798 | make_number (-1), | |
1799 | make_number (-1), | |
1800 | build_string ("8-bit control code (0x80..0x9F)"), | |
d78bc582 KH |
1801 | Qnil, /* same as above */ |
1802 | Qnil); /* same as above */ | |
2e344af3 KH |
1803 | CHARSET_SYMBOL (CHARSET_8_BIT_CONTROL) = Qeight_bit_control; |
1804 | Fput (Qeight_bit_control, Qcharset, | |
1805 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_CONTROL)); | |
1806 | ||
1807 | update_charset_table (make_number (CHARSET_8_BIT_GRAPHIC), | |
1808 | make_number (1), make_number (96), | |
797a084a | 1809 | make_number (4), |
2e344af3 KH |
1810 | make_number (0), |
1811 | make_number (-1), | |
1812 | make_number (-1), | |
2e344af3 | 1813 | build_string ("8-bit graphic char (0xA0..0xFF)"), |
d78bc582 KH |
1814 | Qnil, /* same as above */ |
1815 | Qnil); /* same as above */ | |
2e344af3 KH |
1816 | CHARSET_SYMBOL (CHARSET_8_BIT_GRAPHIC) = Qeight_bit_graphic; |
1817 | Fput (Qeight_bit_graphic, Qcharset, | |
1818 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_GRAPHIC)); | |
1819 | ||
c1a08b4c KH |
1820 | Qauto_fill_chars = intern ("auto-fill-chars"); |
1821 | staticpro (&Qauto_fill_chars); | |
1822 | Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0)); | |
1823 | ||
4ed46869 | 1824 | defsubr (&Sdefine_charset); |
8a73a704 | 1825 | defsubr (&Sgeneric_character_list); |
3fac5a51 | 1826 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
1827 | defsubr (&Sdeclare_equiv_charset); |
1828 | defsubr (&Sfind_charset_region); | |
1829 | defsubr (&Sfind_charset_string); | |
1830 | defsubr (&Smake_char_internal); | |
1831 | defsubr (&Ssplit_char); | |
1832 | defsubr (&Schar_charset); | |
90d7b74e | 1833 | defsubr (&Scharset_after); |
4ed46869 | 1834 | defsubr (&Siso_charset); |
9d3d8cba | 1835 | defsubr (&Schar_valid_p); |
d2665018 | 1836 | defsubr (&Sunibyte_char_to_multibyte); |
1bcc1567 | 1837 | defsubr (&Smultibyte_char_to_unibyte); |
4ed46869 KH |
1838 | defsubr (&Schar_bytes); |
1839 | defsubr (&Schar_width); | |
1840 | defsubr (&Sstring_width); | |
1841 | defsubr (&Schar_direction); | |
87b089ad | 1842 | defsubr (&Sstring); |
4ed46869 KH |
1843 | defsubr (&Ssetup_special_charsets); |
1844 | ||
1845 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
fdb82f93 | 1846 | doc: /* List of charsets ever defined. */); |
2e344af3 KH |
1847 | Vcharset_list = Fcons (Qascii, Fcons (Qeight_bit_control, |
1848 | Fcons (Qeight_bit_graphic, Qnil))); | |
4ed46869 | 1849 | |
537efd8d | 1850 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, |
fdb82f93 PJ |
1851 | doc: /* Vector of cons cell of a symbol and translation table ever defined. |
1852 | An ID of a translation table is an index of this vector. */); | |
537efd8d | 1853 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
b0e3cf2b | 1854 | |
4ed46869 | 1855 | DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, |
fdb82f93 | 1856 | doc: /* Leading-code of private TYPE9N charset of column-width 1. */); |
4ed46869 KH |
1857 | leading_code_private_11 = LEADING_CODE_PRIVATE_11; |
1858 | ||
1859 | DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
fdb82f93 | 1860 | doc: /* Leading-code of private TYPE9N charset of column-width 2. */); |
4ed46869 KH |
1861 | leading_code_private_12 = LEADING_CODE_PRIVATE_12; |
1862 | ||
1863 | DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
fdb82f93 | 1864 | doc: /* Leading-code of private TYPE9Nx9N charset of column-width 1. */); |
4ed46869 KH |
1865 | leading_code_private_21 = LEADING_CODE_PRIVATE_21; |
1866 | ||
1867 | DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
fdb82f93 | 1868 | doc: /* Leading-code of private TYPE9Nx9N charset of column-width 2. */); |
4ed46869 | 1869 | leading_code_private_22 = LEADING_CODE_PRIVATE_22; |
35e623fb RS |
1870 | |
1871 | DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset, | |
fdb82f93 PJ |
1872 | doc: /* Offset for converting non-ASCII unibyte codes 0240...0377 to multibyte. |
1873 | This is used for converting unibyte text to multibyte, | |
1874 | and for inserting character codes specified by number. | |
1875 | ||
1876 | This serves to convert a Latin-1 or similar 8-bit character code | |
1877 | to the corresponding Emacs multibyte character code. | |
1878 | Typically the value should be (- (make-char CHARSET 0) 128), | |
1879 | for your choice of character set. | |
1880 | If `nonascii-translation-table' is non-nil, it overrides this variable. */); | |
35e623fb | 1881 | nonascii_insert_offset = 0; |
b0e3cf2b | 1882 | |
b4e9dd77 | 1883 | DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table, |
fdb82f93 PJ |
1884 | doc: /* Translation table to convert non-ASCII unibyte codes to multibyte. |
1885 | This is used for converting unibyte text to multibyte, | |
1886 | and for inserting character codes specified by number. | |
1887 | ||
1888 | Conversion is performed only when multibyte characters are enabled, | |
1889 | and it serves to convert a Latin-1 or similar 8-bit character code | |
1890 | to the corresponding Emacs character code. | |
1891 | ||
1892 | If this is nil, `nonascii-insert-offset' is used instead. | |
1893 | See also the docstring of `make-translation-table'. */); | |
b4e9dd77 | 1894 | Vnonascii_translation_table = Qnil; |
4cf9710d | 1895 | |
c1a08b4c | 1896 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, |
fdb82f93 PJ |
1897 | doc: /* A char-table for characters which invoke auto-filling. |
1898 | Such characters have value t in this table. */); | |
c1a08b4c | 1899 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); |
60022cb7 AS |
1900 | CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt); |
1901 | CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt); | |
4ed46869 KH |
1902 | } |
1903 | ||
1904 | #endif /* emacs */ | |
ab5796a9 MB |
1905 | |
1906 | /* arch-tag: 66a89b8d-4c28-47d3-9ca1-56f78440d69f | |
1907 | (do not change this comment) */ |