Commit | Line | Data |
---|---|---|
75c8c592 | 1 | /* Basic multilingual character support. |
35e623fb | 2 | Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN. |
75c8c592 | 3 | Licensed to the Free Software Foundation. |
4ed46869 | 4 | |
369314dc KH |
5 | This file is part of GNU Emacs. |
6 | ||
7 | GNU Emacs is free software; you can redistribute it and/or modify | |
8 | it under the terms of the GNU General Public License as published by | |
9 | the Free Software Foundation; either version 2, or (at your option) | |
10 | any later version. | |
4ed46869 | 11 | |
369314dc KH |
12 | GNU Emacs is distributed in the hope that it will be useful, |
13 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
14 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
15 | GNU General Public License for more details. | |
4ed46869 | 16 | |
369314dc KH |
17 | You should have received a copy of the GNU General Public License |
18 | along with GNU Emacs; see the file COPYING. If not, write to | |
19 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
20 | Boston, MA 02111-1307, USA. */ | |
4ed46869 KH |
21 | |
22 | /* At first, see the document in `charset.h' to understand the code in | |
23 | this file. */ | |
24 | ||
68c45bf0 PE |
25 | #ifdef emacs |
26 | #include <config.h> | |
27 | #endif | |
28 | ||
4ed46869 KH |
29 | #include <stdio.h> |
30 | ||
31 | #ifdef emacs | |
32 | ||
33 | #include <sys/types.h> | |
4ed46869 KH |
34 | #include "lisp.h" |
35 | #include "buffer.h" | |
36 | #include "charset.h" | |
3f62427c | 37 | #include "composite.h" |
4ed46869 | 38 | #include "coding.h" |
fc6b09bf | 39 | #include "disptab.h" |
4ed46869 KH |
40 | |
41 | #else /* not emacs */ | |
42 | ||
43 | #include "mulelib.h" | |
44 | ||
45 | #endif /* emacs */ | |
46 | ||
2e344af3 | 47 | Lisp_Object Qcharset, Qascii, Qeight_bit_control, Qeight_bit_graphic; |
0282eb69 | 48 | Lisp_Object Qunknown; |
4ed46869 KH |
49 | |
50 | /* Declaration of special leading-codes. */ | |
4ed46869 KH |
51 | int leading_code_private_11; /* for private DIMENSION1 of 1-column */ |
52 | int leading_code_private_12; /* for private DIMENSION1 of 2-column */ | |
53 | int leading_code_private_21; /* for private DIMENSION2 of 1-column */ | |
54 | int leading_code_private_22; /* for private DIMENSION2 of 2-column */ | |
55 | ||
2e344af3 KH |
56 | /* Declaration of special charsets. The values are set by |
57 | Fsetup_special_charsets. */ | |
4ed46869 KH |
58 | int charset_latin_iso8859_1; /* ISO8859-1 (Latin-1) */ |
59 | int charset_jisx0208_1978; /* JISX0208.1978 (Japanese Kanji old set) */ | |
60 | int charset_jisx0208; /* JISX0208.1983 (Japanese Kanji) */ | |
61 | int charset_katakana_jisx0201; /* JISX0201.Kana (Japanese Katakana) */ | |
62 | int charset_latin_jisx0201; /* JISX0201.Roman (Japanese Roman) */ | |
63 | int charset_big5_1; /* Big5 Level 1 (Chinese Traditional) */ | |
64 | int charset_big5_2; /* Big5 Level 2 (Chinese Traditional) */ | |
65 | ||
66 | Lisp_Object Qcharset_table; | |
67 | ||
68 | /* A char-table containing information of each character set. */ | |
69 | Lisp_Object Vcharset_table; | |
70 | ||
71 | /* A vector of charset symbol indexed by charset-id. This is used | |
72 | only for returning charset symbol from C functions. */ | |
73 | Lisp_Object Vcharset_symbol_table; | |
74 | ||
75 | /* A list of charset symbols ever defined. */ | |
76 | Lisp_Object Vcharset_list; | |
77 | ||
537efd8d KH |
78 | /* Vector of translation table ever defined. |
79 | ID of a translation table is used to index this vector. */ | |
80 | Lisp_Object Vtranslation_table_vector; | |
b0e3cf2b | 81 | |
c1a08b4c KH |
82 | /* A char-table for characters which may invoke auto-filling. */ |
83 | Lisp_Object Vauto_fill_chars; | |
84 | ||
85 | Lisp_Object Qauto_fill_chars; | |
86 | ||
4ed46869 KH |
87 | /* Tables used by macros BYTES_BY_CHAR_HEAD and WIDTH_BY_CHAR_HEAD. */ |
88 | int bytes_by_char_head[256]; | |
89 | int width_by_char_head[256]; | |
90 | ||
91 | /* Mapping table from ISO2022's charset (specified by DIMENSION, | |
92 | CHARS, and FINAL-CHAR) to Emacs' charset. */ | |
93 | int iso_charset_table[2][2][128]; | |
94 | ||
95 | /* Variables used locally in the macro FETCH_MULTIBYTE_CHAR. */ | |
96 | unsigned char *_fetch_multibyte_char_p; | |
97 | int _fetch_multibyte_char_len; | |
98 | ||
35e623fb RS |
99 | /* Offset to add to a non-ASCII value when inserting it. */ |
100 | int nonascii_insert_offset; | |
101 | ||
4cf9710d RS |
102 | /* Translation table for converting non-ASCII unibyte characters |
103 | to multibyte codes, or nil. */ | |
b4e9dd77 | 104 | Lisp_Object Vnonascii_translation_table; |
4cf9710d | 105 | |
8a73a704 KH |
106 | /* List of all possible generic characters. */ |
107 | Lisp_Object Vgeneric_character_list; | |
108 | ||
046b1f03 RS |
109 | #define min(X, Y) ((X) < (Y) ? (X) : (Y)) |
110 | #define max(X, Y) ((X) > (Y) ? (X) : (Y)) | |
111 | \f | |
93bcb785 KH |
112 | void |
113 | invalid_character (c) | |
114 | int c; | |
115 | { | |
ba7434e5 | 116 | error ("Invalid character: 0%o, %d, 0x%x", c, c, c); |
93bcb785 KH |
117 | } |
118 | ||
2e344af3 KH |
119 | /* Parse string STR of length LENGTH and fetch information of a |
120 | character at STR. Set BYTES to the byte length the character | |
121 | occupies, CHARSET, C1, C2 to proper values of the character. */ | |
122 | ||
123 | #define SPLIT_MULTIBYTE_SEQ(str, length, bytes, charset, c1, c2) \ | |
124 | do { \ | |
125 | (c1) = *(str); \ | |
126 | (bytes) = BYTES_BY_CHAR_HEAD (c1); \ | |
127 | if ((bytes) == 1) \ | |
128 | (charset) = ASCII_BYTE_P (c1) ? CHARSET_ASCII : CHARSET_8_BIT_GRAPHIC; \ | |
129 | else if ((bytes) == 2) \ | |
130 | { \ | |
131 | if ((c1) == LEADING_CODE_8_BIT_CONTROL) \ | |
132 | (charset) = CHARSET_8_BIT_CONTROL, (c1) = (str)[1] - 0x20; \ | |
133 | else \ | |
134 | (charset) = (c1), (c1) = (str)[1] & 0x7F; \ | |
135 | } \ | |
136 | else if ((bytes) == 3) \ | |
137 | { \ | |
138 | if ((c1) < LEADING_CODE_PRIVATE_11) \ | |
139 | (charset) = (c1), (c1) = (str)[1] & 0x7F, (c2) = (str)[2] & 0x7F; \ | |
140 | else \ | |
141 | (charset) = (str)[1], (c1) = (str)[2] & 0x7F; \ | |
142 | } \ | |
143 | else \ | |
144 | (charset) = (str)[1], (c1) = (str)[2] & 0x7F, (c2) = (str)[3] & 0x7F; \ | |
ac4137cc KH |
145 | } while (0) |
146 | ||
147 | /* 1 if CHARSET, C1, and C2 compose a valid character, else 0. */ | |
44c6492d | 148 | #define CHAR_COMPONENTS_VALID_P(charset, c1, c2) \ |
2e344af3 | 149 | ((charset) == CHARSET_ASCII \ |
63f4d579 | 150 | ? ((c1) >= 0 && (c1) <= 0x7F) \ |
2e344af3 KH |
151 | : ((charset) == CHARSET_8_BIT_CONTROL \ |
152 | ? ((c1) >= 0x80 && (c1) <= 0x9F) \ | |
153 | : ((charset) == CHARSET_8_BIT_GRAPHIC \ | |
154 | ? ((c1) >= 0x80 && (c1) <= 0xFF) \ | |
155 | : (CHARSET_DIMENSION (charset) == 1 \ | |
156 | ? ((c1) >= 0x20 && (c1) <= 0x7F) \ | |
157 | : ((c1) >= 0x20 && (c1) <= 0x7F \ | |
158 | && (c2) >= 0x20 && (c2) <= 0x7F))))) | |
93bcb785 | 159 | |
99529c2c KH |
160 | /* Store multi-byte form of the character C in STR. The caller should |
161 | allocate at least 4-byte area at STR in advance. Returns the | |
162 | length of the multi-byte form. If C is an invalid character code, | |
163 | signal an error. | |
4ed46869 | 164 | |
99529c2c KH |
165 | Use macro `CHAR_STRING (C, STR)' instead of calling this function |
166 | directly if C can be an ASCII character. */ | |
4ed46869 KH |
167 | |
168 | int | |
99529c2c | 169 | char_to_string (c, str) |
4ed46869 | 170 | int c; |
99529c2c | 171 | unsigned char *str; |
4ed46869 | 172 | { |
99529c2c KH |
173 | unsigned char *p = str; |
174 | ||
6662e69b | 175 | if (c & CHAR_MODIFIER_MASK) /* This includes the case C is negative. */ |
8ac5a9cc | 176 | { |
6662e69b KH |
177 | /* Multibyte character can't have a modifier bit. */ |
178 | if (! SINGLE_BYTE_CHAR_P ((c & ~CHAR_MODIFIER_MASK))) | |
179 | invalid_character (c); | |
180 | ||
181 | /* For Meta, Shift, and Control modifiers, we need special care. */ | |
8ac5a9cc | 182 | if (c & CHAR_META) |
6662e69b KH |
183 | { |
184 | /* Move the meta bit to the right place for a string. */ | |
185 | c = (c & ~CHAR_META) | 0x80; | |
186 | } | |
187 | if (c & CHAR_SHIFT) | |
188 | { | |
189 | /* Shift modifier is valid only with [A-Za-z]. */ | |
190 | if ((c & 0377) >= 'A' && (c & 0377) <= 'Z') | |
191 | c &= ~CHAR_SHIFT; | |
192 | else if ((c & 0377) >= 'a' && (c & 0377) <= 'z') | |
193 | c = (c & ~CHAR_SHIFT) - ('a' - 'A'); | |
194 | } | |
8ac5a9cc | 195 | if (c & CHAR_CTL) |
6662e69b KH |
196 | { |
197 | /* Simulate the code in lread.c. */ | |
198 | /* Allow `\C- ' and `\C-?'. */ | |
199 | if (c == (CHAR_CTL | ' ')) | |
200 | c = 0; | |
201 | else if (c == (CHAR_CTL | '?')) | |
202 | c = 127; | |
203 | /* ASCII control chars are made from letters (both cases), | |
204 | as well as the non-letters within 0100...0137. */ | |
205 | else if ((c & 0137) >= 0101 && (c & 0137) <= 0132) | |
206 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
207 | else if ((c & 0177) >= 0100 && (c & 0177) <= 0137) | |
208 | c &= (037 | (~0177 & ~CHAR_CTL)); | |
209 | } | |
210 | ||
211 | /* If C still has any modifier bits, it is an invalid character. */ | |
212 | if (c & CHAR_MODIFIER_MASK) | |
213 | invalid_character (c); | |
2e344af3 KH |
214 | } |
215 | if (SINGLE_BYTE_CHAR_P (c)) | |
216 | { | |
217 | if (ASCII_BYTE_P (c) || c >= 0xA0) | |
218 | *p++ = c; | |
219 | else | |
220 | { | |
221 | *p++ = LEADING_CODE_8_BIT_CONTROL; | |
222 | *p++ = c + 0x20; | |
223 | } | |
8ac5a9cc | 224 | } |
6ce974d4 | 225 | else if (CHAR_VALID_P (c, 0)) |
4ed46869 | 226 | { |
ac4137cc | 227 | int charset, c1, c2; |
4ed46869 | 228 | |
2e344af3 | 229 | SPLIT_CHAR (c, charset, c1, c2); |
99529c2c KH |
230 | |
231 | if (charset >= LEADING_CODE_EXT_11) | |
232 | *p++ = (charset < LEADING_CODE_EXT_12 | |
233 | ? LEADING_CODE_PRIVATE_11 | |
234 | : (charset < LEADING_CODE_EXT_21 | |
235 | ? LEADING_CODE_PRIVATE_12 | |
236 | : (charset < LEADING_CODE_EXT_22 | |
237 | ? LEADING_CODE_PRIVATE_21 | |
238 | : LEADING_CODE_PRIVATE_22))); | |
239 | *p++ = charset; | |
240 | if (c1 > 0 && c1 < 32 || c2 > 0 && c2 < 32) | |
241 | invalid_character (c); | |
242 | if (c1) | |
4ed46869 | 243 | { |
99529c2c KH |
244 | *p++ = c1 | 0x80; |
245 | if (c2 > 0) | |
246 | *p++ = c2 | 0x80; | |
4ed46869 KH |
247 | } |
248 | } | |
2e344af3 KH |
249 | else |
250 | invalid_character (c); | |
4ed46869 | 251 | |
2e344af3 | 252 | return (p - str); |
4ed46869 KH |
253 | } |
254 | ||
44c6492d KH |
255 | /* Return the non-ASCII character corresponding to multi-byte form at |
256 | STR of length LEN. If ACTUAL_LEN is not NULL, store the byte | |
257 | length of the multibyte form in *ACTUAL_LEN. | |
537efd8d | 258 | |
99529c2c KH |
259 | Use macros STRING_CHAR or STRING_CHAR_AND_LENGTH instead of calling |
260 | this function directly if you want ot handle ASCII characters as | |
261 | well. */ | |
4ed46869 | 262 | |
dfcf069d | 263 | int |
99529c2c | 264 | string_to_char (str, len, actual_len) |
8867de67 | 265 | const unsigned char *str; |
ac4137cc | 266 | int len, *actual_len; |
4ed46869 | 267 | { |
ac4137cc | 268 | int c, bytes, charset, c1, c2; |
4ed46869 | 269 | |
ac4137cc KH |
270 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, charset, c1, c2); |
271 | c = MAKE_CHAR (charset, c1, c2); | |
4ed46869 | 272 | if (actual_len) |
ac4137cc | 273 | *actual_len = bytes; |
4ed46869 KH |
274 | return c; |
275 | } | |
276 | ||
44c6492d KH |
277 | /* Return the length of the multi-byte form at string STR of length LEN. |
278 | Use the macro MULTIBYTE_FORM_LENGTH instead. */ | |
4ed46869 KH |
279 | int |
280 | multibyte_form_length (str, len) | |
8867de67 | 281 | const unsigned char *str; |
4ed46869 KH |
282 | int len; |
283 | { | |
ac4137cc | 284 | int bytes; |
4ed46869 | 285 | |
ac4137cc | 286 | PARSE_MULTIBYTE_SEQ (str, len, bytes); |
90d7b74e | 287 | return bytes; |
4ed46869 KH |
288 | } |
289 | ||
ac4137cc KH |
290 | /* Check multibyte form at string STR of length LEN and set variables |
291 | pointed by CHARSET, C1, and C2 to charset and position codes of the | |
292 | character at STR, and return 0. If there's no multibyte character, | |
4ed46869 KH |
293 | return -1. This should be used only in the macro SPLIT_STRING |
294 | which checks range of STR in advance. */ | |
295 | ||
dfcf069d | 296 | int |
99529c2c | 297 | split_string (str, len, charset, c1, c2) |
ac4137cc KH |
298 | const unsigned char *str; |
299 | unsigned char *c1, *c2; | |
300 | int len, *charset; | |
4ed46869 | 301 | { |
ac4137cc | 302 | register int bytes, cs, code1, code2 = -1; |
4ed46869 | 303 | |
ac4137cc KH |
304 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, cs, code1, code2); |
305 | if (cs == CHARSET_ASCII) | |
4ed46869 | 306 | return -1; |
ac4137cc KH |
307 | *charset = cs; |
308 | *c1 = code1; | |
309 | *c2 = code2; | |
5865af0d | 310 | return 0; |
ac4137cc KH |
311 | } |
312 | ||
44c6492d KH |
313 | /* Return 1 iff character C has valid printable glyph. |
314 | Use the macro CHAR_PRINTABLE_P instead. */ | |
ac4137cc KH |
315 | int |
316 | char_printable_p (c) | |
317 | int c; | |
318 | { | |
8ebae00c | 319 | int charset, c1, c2; |
ac4137cc | 320 | |
2e344af3 | 321 | if (ASCII_BYTE_P (c)) |
ac4137cc | 322 | return 1; |
2e344af3 KH |
323 | else if (SINGLE_BYTE_CHAR_P (c)) |
324 | return 0; | |
325 | else if (c >= MAX_CHAR) | |
99529c2c | 326 | return 0; |
ac4137cc | 327 | |
2e344af3 | 328 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
329 | if (! CHARSET_DEFINED_P (charset)) |
330 | return 0; | |
331 | if (CHARSET_CHARS (charset) == 94 | |
332 | ? c1 <= 32 || c1 >= 127 | |
333 | : c1 < 32) | |
334 | return 0; | |
335 | if (CHARSET_DIMENSION (charset) == 2 | |
336 | && (CHARSET_CHARS (charset) == 94 | |
337 | ? c2 <= 32 || c2 >= 127 | |
338 | : c2 < 32)) | |
339 | return 0; | |
340 | return 1; | |
4ed46869 KH |
341 | } |
342 | ||
537efd8d | 343 | /* Translate character C by translation table TABLE. If C |
b4e9dd77 KH |
344 | is negative, translate a character specified by CHARSET, C1, and C2 |
345 | (C1 and C2 are code points of the character). If no translation is | |
346 | found in TABLE, return C. */ | |
dfcf069d | 347 | int |
b4e9dd77 | 348 | translate_char (table, c, charset, c1, c2) |
23d2a7f1 KH |
349 | Lisp_Object table; |
350 | int c, charset, c1, c2; | |
351 | { | |
352 | Lisp_Object ch; | |
353 | int alt_charset, alt_c1, alt_c2, dimension; | |
354 | ||
0ad3f83d | 355 | if (c < 0) c = MAKE_CHAR (charset, (c1 & 0x7F) , (c2 & 0x7F)); |
23d2a7f1 | 356 | if (!CHAR_TABLE_P (table) |
ac4137cc | 357 | || (ch = Faref (table, make_number (c)), !NATNUMP (ch))) |
23d2a7f1 KH |
358 | return c; |
359 | ||
360 | SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); | |
361 | dimension = CHARSET_DIMENSION (alt_charset); | |
362 | if (dimension == 1 && alt_c1 > 0 || dimension == 2 && alt_c2 > 0) | |
363 | /* CH is not a generic character, just return it. */ | |
364 | return XFASTINT (ch); | |
365 | ||
366 | /* Since CH is a generic character, we must return a specific | |
367 | charater which has the same position codes as C from CH. */ | |
368 | if (charset < 0) | |
369 | SPLIT_CHAR (c, charset, c1, c2); | |
370 | if (dimension != CHARSET_DIMENSION (charset)) | |
371 | /* We can't make such a character because of dimension mismatch. */ | |
372 | return c; | |
23d2a7f1 KH |
373 | return MAKE_CHAR (alt_charset, c1, c2); |
374 | } | |
375 | ||
d2665018 | 376 | /* Convert the unibyte character C to multibyte based on |
b4e9dd77 | 377 | Vnonascii_translation_table or nonascii_insert_offset. If they can't |
d2665018 KH |
378 | convert C to a valid multibyte character, convert it based on |
379 | DEFAULT_NONASCII_INSERT_OFFSET which makes C a Latin-1 character. */ | |
35e623fb | 380 | |
dfcf069d | 381 | int |
35e623fb RS |
382 | unibyte_char_to_multibyte (c) |
383 | int c; | |
384 | { | |
543b4f61 | 385 | if (c < 0400 && c >= 0200) |
35e623fb | 386 | { |
d2665018 KH |
387 | int c_save = c; |
388 | ||
b4e9dd77 | 389 | if (! NILP (Vnonascii_translation_table)) |
bbf12bb3 KH |
390 | { |
391 | c = XINT (Faref (Vnonascii_translation_table, make_number (c))); | |
ac4137cc | 392 | if (c >= 0400 && ! char_valid_p (c, 0)) |
bbf12bb3 KH |
393 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
394 | } | |
395 | else if (c >= 0240 && nonascii_insert_offset > 0) | |
396 | { | |
397 | c += nonascii_insert_offset; | |
ac4137cc | 398 | if (c < 0400 || ! char_valid_p (c, 0)) |
bbf12bb3 KH |
399 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
400 | } | |
401 | else if (c >= 0240) | |
d2665018 | 402 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
35e623fb RS |
403 | } |
404 | return c; | |
405 | } | |
76d7b829 KH |
406 | |
407 | ||
408 | /* Convert the multibyte character C to unibyte 8-bit character based | |
409 | on Vnonascii_translation_table or nonascii_insert_offset. If | |
410 | REV_TBL is non-nil, it should be a reverse table of | |
411 | Vnonascii_translation_table, i.e. what given by: | |
412 | Fchar_table_extra_slot (Vnonascii_translation_table, make_number (0)) */ | |
413 | ||
414 | int | |
415 | multibyte_char_to_unibyte (c, rev_tbl) | |
416 | int c; | |
417 | Lisp_Object rev_tbl; | |
418 | { | |
419 | if (!SINGLE_BYTE_CHAR_P (c)) | |
420 | { | |
421 | int c_save = c; | |
422 | ||
423 | if (! CHAR_TABLE_P (rev_tbl) | |
424 | && CHAR_TABLE_P (Vnonascii_translation_table)) | |
425 | rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table, | |
426 | make_number (0)); | |
427 | if (CHAR_TABLE_P (rev_tbl)) | |
428 | { | |
429 | Lisp_Object temp; | |
430 | temp = Faref (rev_tbl, make_number (c)); | |
431 | if (INTEGERP (temp)) | |
432 | c = XINT (temp); | |
bbf12bb3 KH |
433 | if (c >= 256) |
434 | c = (c_save & 0177) + 0200; | |
435 | } | |
436 | else | |
437 | { | |
438 | if (nonascii_insert_offset > 0) | |
439 | c -= nonascii_insert_offset; | |
440 | if (c < 128 || c >= 256) | |
441 | c = (c_save & 0177) + 0200; | |
76d7b829 | 442 | } |
76d7b829 KH |
443 | } |
444 | ||
445 | return c; | |
446 | } | |
447 | ||
35e623fb | 448 | \f |
4ed46869 KH |
449 | /* Update the table Vcharset_table with the given arguments (see the |
450 | document of `define-charset' for the meaning of each argument). | |
451 | Several other table contents are also updated. The caller should | |
452 | check the validity of CHARSET-ID and the remaining arguments in | |
453 | advance. */ | |
454 | ||
455 | void | |
456 | update_charset_table (charset_id, dimension, chars, width, direction, | |
457 | iso_final_char, iso_graphic_plane, | |
458 | short_name, long_name, description) | |
459 | Lisp_Object charset_id, dimension, chars, width, direction; | |
460 | Lisp_Object iso_final_char, iso_graphic_plane; | |
461 | Lisp_Object short_name, long_name, description; | |
462 | { | |
463 | int charset = XINT (charset_id); | |
464 | int bytes; | |
465 | unsigned char leading_code_base, leading_code_ext; | |
466 | ||
6dc0722d KH |
467 | if (NILP (CHARSET_TABLE_ENTRY (charset))) |
468 | CHARSET_TABLE_ENTRY (charset) | |
469 | = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); | |
4ed46869 | 470 | |
d78bc582 KH |
471 | if (NILP (long_name)) |
472 | long_name = short_name; | |
473 | if (NILP (description)) | |
474 | description = long_name; | |
475 | ||
4ed46869 KH |
476 | /* Get byte length of multibyte form, base leading-code, and |
477 | extended leading-code of the charset. See the comment under the | |
478 | title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
479 | bytes = XINT (dimension); | |
480 | if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
481 | { | |
482 | /* Official charset, it doesn't have an extended leading-code. */ | |
2e344af3 | 483 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC) |
4ed46869 KH |
484 | bytes += 1; /* For a base leading-code. */ |
485 | leading_code_base = charset; | |
486 | leading_code_ext = 0; | |
487 | } | |
488 | else | |
489 | { | |
490 | /* Private charset. */ | |
491 | bytes += 2; /* For base and extended leading-codes. */ | |
492 | leading_code_base | |
493 | = (charset < LEADING_CODE_EXT_12 | |
494 | ? LEADING_CODE_PRIVATE_11 | |
495 | : (charset < LEADING_CODE_EXT_21 | |
496 | ? LEADING_CODE_PRIVATE_12 | |
497 | : (charset < LEADING_CODE_EXT_22 | |
498 | ? LEADING_CODE_PRIVATE_21 | |
499 | : LEADING_CODE_PRIVATE_22))); | |
500 | leading_code_ext = charset; | |
c83ef371 KH |
501 | if (BYTES_BY_CHAR_HEAD (leading_code_base) != bytes) |
502 | error ("Invalid dimension for the charset-ID %d", charset); | |
503 | } | |
6ef23ebb | 504 | |
4ed46869 KH |
505 | CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; |
506 | CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
507 | CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
508 | CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
509 | CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
510 | CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
511 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
512 | = make_number (leading_code_base); | |
513 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
514 | = make_number (leading_code_ext); | |
515 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
516 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
517 | = iso_graphic_plane; | |
518 | CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
519 | CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
520 | CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
521 | CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
522 | ||
523 | { | |
524 | /* If we have already defined a charset which has the same | |
525 | DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
526 | DIRECTION, we must update the entry REVERSE-CHARSET of both | |
527 | charsets. If there's no such charset, the value of the entry | |
528 | is set to nil. */ | |
529 | int i; | |
530 | ||
513ee442 | 531 | for (i = 0; i <= MAX_CHARSET; i++) |
4ed46869 KH |
532 | if (!NILP (CHARSET_TABLE_ENTRY (i))) |
533 | { | |
534 | if (CHARSET_DIMENSION (i) == XINT (dimension) | |
535 | && CHARSET_CHARS (i) == XINT (chars) | |
536 | && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
537 | && CHARSET_DIRECTION (i) != XINT (direction)) | |
538 | { | |
539 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
540 | = make_number (i); | |
541 | CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
542 | break; | |
543 | } | |
544 | } | |
513ee442 | 545 | if (i > MAX_CHARSET) |
4ed46869 KH |
546 | /* No such a charset. */ |
547 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
548 | = make_number (-1); | |
549 | } | |
550 | ||
c83ef371 | 551 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC |
4ed46869 KH |
552 | && charset < MIN_CHARSET_PRIVATE_DIMENSION1) |
553 | { | |
c83ef371 | 554 | bytes_by_char_head[leading_code_base] = bytes; |
4ed46869 KH |
555 | width_by_char_head[leading_code_base] = XINT (width); |
556 | ||
557 | /* Update table emacs_code_class. */ | |
558 | emacs_code_class[charset] = (bytes == 2 | |
559 | ? EMACS_leading_code_2 | |
560 | : (bytes == 3 | |
561 | ? EMACS_leading_code_3 | |
562 | : EMACS_leading_code_4)); | |
563 | } | |
564 | ||
565 | /* Update table iso_charset_table. */ | |
52e386c2 | 566 | if (XINT (iso_final_char) >= 0 |
2e344af3 | 567 | && ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0) |
4ed46869 KH |
568 | ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset; |
569 | } | |
570 | ||
571 | #ifdef emacs | |
572 | ||
573 | /* Return charset id of CHARSET_SYMBOL, or return -1 if CHARSET_SYMBOL | |
574 | is invalid. */ | |
575 | int | |
576 | get_charset_id (charset_symbol) | |
577 | Lisp_Object charset_symbol; | |
578 | { | |
579 | Lisp_Object val; | |
580 | int charset; | |
581 | ||
582 | return ((SYMBOLP (charset_symbol) | |
583 | && (val = Fget (charset_symbol, Qcharset), VECTORP (val)) | |
584 | && (charset = XINT (XVECTOR (val)->contents[CHARSET_ID_IDX]), | |
585 | CHARSET_VALID_P (charset))) | |
586 | ? charset : -1); | |
587 | } | |
588 | ||
589 | /* Return an identification number for a new private charset of | |
590 | DIMENSION and WIDTH. If there's no more room for the new charset, | |
591 | return 0. */ | |
592 | Lisp_Object | |
593 | get_new_private_charset_id (dimension, width) | |
594 | int dimension, width; | |
595 | { | |
596 | int charset, from, to; | |
597 | ||
598 | if (dimension == 1) | |
599 | { | |
3bb7b08b KH |
600 | from = LEADING_CODE_EXT_11; |
601 | to = LEADING_CODE_EXT_21; | |
4ed46869 KH |
602 | } |
603 | else | |
604 | { | |
3bb7b08b KH |
605 | from = LEADING_CODE_EXT_21; |
606 | to = LEADING_CODE_EXT_MAX + 1; | |
4ed46869 KH |
607 | } |
608 | ||
609 | for (charset = from; charset < to; charset++) | |
610 | if (!CHARSET_DEFINED_P (charset)) break; | |
611 | ||
612 | return make_number (charset < to ? charset : 0); | |
613 | } | |
614 | ||
615 | DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
616 | "Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR.\n\ | |
23d2a7f1 | 617 | If CHARSET-ID is nil, it is decided automatically, which means CHARSET is\n\ |
4ed46869 KH |
618 | treated as a private charset.\n\ |
619 | INFO-VECTOR is a vector of the format:\n\ | |
620 | [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE\n\ | |
621 | SHORT-NAME LONG-NAME DESCRIPTION]\n\ | |
622 | The meanings of each elements is as follows:\n\ | |
623 | DIMENSION (integer) is the number of bytes to represent a character: 1 or 2.\n\ | |
624 | CHARS (integer) is the number of characters in a dimension: 94 or 96.\n\ | |
625 | WIDTH (integer) is the number of columns a character in the charset\n\ | |
626 | occupies on the screen: one of 0, 1, and 2.\n\ | |
627 | \n\ | |
628 | DIRECTION (integer) is the rendering direction of characters in the\n\ | |
277576f6 KH |
629 | charset when rendering. If 0, render from left to right, else\n\ |
630 | render from right to left.\n\ | |
4ed46869 KH |
631 | \n\ |
632 | ISO-FINAL-CHAR (character) is the final character of the\n\ | |
633 | corresponding ISO 2022 charset.\n\ | |
2e344af3 | 634 | It may be -1 if the charset is internal use only.\n\ |
4ed46869 KH |
635 | \n\ |
636 | ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked\n\ | |
637 | while encoding to variants of ISO 2022 coding system, one of the\n\ | |
638 | following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).\n\ | |
2e344af3 | 639 | It may be -1 if the charset is internal use only.\n\ |
4ed46869 KH |
640 | \n\ |
641 | SHORT-NAME (string) is the short name to refer to the charset.\n\ | |
642 | \n\ | |
643 | LONG-NAME (string) is the long name to refer to the charset.\n\ | |
644 | \n\ | |
645 | DESCRIPTION (string) is the description string of the charset.") | |
646 | (charset_id, charset_symbol, info_vector) | |
647 | Lisp_Object charset_id, charset_symbol, info_vector; | |
648 | { | |
649 | Lisp_Object *vec; | |
650 | ||
651 | if (!NILP (charset_id)) | |
652 | CHECK_NUMBER (charset_id, 0); | |
653 | CHECK_SYMBOL (charset_symbol, 1); | |
654 | CHECK_VECTOR (info_vector, 2); | |
655 | ||
656 | if (! NILP (charset_id)) | |
657 | { | |
658 | if (! CHARSET_VALID_P (XINT (charset_id))) | |
659 | error ("Invalid CHARSET: %d", XINT (charset_id)); | |
660 | else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
661 | error ("Already defined charset: %d", XINT (charset_id)); | |
662 | } | |
663 | ||
664 | vec = XVECTOR (info_vector)->contents; | |
665 | if (XVECTOR (info_vector)->size != 9 | |
666 | || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
667 | || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
668 | || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
669 | || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
2e344af3 KH |
670 | || !INTEGERP (vec[4]) |
671 | || !(XINT (vec[4]) == -1 || XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~') | |
672 | || !INTEGERP (vec[5]) | |
673 | || !(XINT (vec[5]) == -1 || XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
4ed46869 KH |
674 | || !STRINGP (vec[6]) |
675 | || !STRINGP (vec[7]) | |
676 | || !STRINGP (vec[8])) | |
677 | error ("Invalid info-vector argument for defining charset %s", | |
678 | XSYMBOL (charset_symbol)->name->data); | |
679 | ||
680 | if (NILP (charset_id)) | |
681 | { | |
682 | charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
683 | if (XINT (charset_id) == 0) | |
684 | error ("There's no room for a new private charset %s", | |
685 | XSYMBOL (charset_symbol)->name->data); | |
686 | } | |
687 | ||
688 | update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
689 | vec[4], vec[5], vec[6], vec[7], vec[8]); | |
6dc0722d | 690 | Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
4ed46869 KH |
691 | CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
692 | Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
693 | return Qnil; | |
694 | } | |
695 | ||
8a73a704 KH |
696 | DEFUN ("generic-character-list", Fgeneric_character_list, |
697 | Sgeneric_character_list, 0, 0, 0, | |
698 | "Return a list of all possible generic characters.\n\ | |
699 | It includes a generic character for a charset not yet defined.") | |
700 | () | |
701 | { | |
702 | return Vgeneric_character_list; | |
703 | } | |
704 | ||
3fac5a51 KH |
705 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
706 | Sget_unused_iso_final_char, 2, 2, 0, | |
707 | "Return an unsed ISO's final char for a charset of DIMENISION and CHARS.\n\ | |
708 | DIMENSION is the number of bytes to represent a character: 1 or 2.\n\ | |
709 | CHARS is the number of characters in a dimension: 94 or 96.\n\ | |
710 | \n\ | |
711 | This final char is for private use, thus the range is `0' (48) .. `?' (63).\n\ | |
712 | If there's no unused final char for the specified kind of charset,\n\ | |
713 | return nil.") | |
714 | (dimension, chars) | |
715 | Lisp_Object dimension, chars; | |
716 | { | |
717 | int final_char; | |
718 | ||
719 | CHECK_NUMBER (dimension, 0); | |
720 | CHECK_NUMBER (chars, 1); | |
721 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
722 | error ("Invalid charset dimension %d, it should be 1 or 2", | |
723 | XINT (dimension)); | |
724 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
725 | error ("Invalid charset chars %d, it should be 94 or 96", | |
726 | XINT (chars)); | |
727 | for (final_char = '0'; final_char <= '?'; final_char++) | |
728 | { | |
729 | if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0) | |
730 | break; | |
731 | } | |
732 | return (final_char <= '?' ? make_number (final_char) : Qnil); | |
733 | } | |
734 | ||
4ed46869 KH |
735 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
736 | 4, 4, 0, | |
737 | "Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET.\n\ | |
738 | CHARSET should be defined by `defined-charset' in advance.") | |
739 | (dimension, chars, final_char, charset_symbol) | |
740 | Lisp_Object dimension, chars, final_char, charset_symbol; | |
741 | { | |
742 | int charset; | |
743 | ||
744 | CHECK_NUMBER (dimension, 0); | |
745 | CHECK_NUMBER (chars, 1); | |
746 | CHECK_NUMBER (final_char, 2); | |
747 | CHECK_SYMBOL (charset_symbol, 3); | |
748 | ||
749 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
750 | error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
751 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
752 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
753 | if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
754 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
755 | if ((charset = get_charset_id (charset_symbol)) < 0) | |
756 | error ("Invalid charset %s", XSYMBOL (charset_symbol)->name->data); | |
757 | ||
758 | ISO_CHARSET_TABLE (dimension, chars, final_char) = charset; | |
759 | return Qnil; | |
760 | } | |
761 | ||
2e344af3 KH |
762 | /* Return information about charsets in the text at PTR of NBYTES |
763 | bytes, which are NCHARS characters. The value is: | |
f6302ac9 | 764 | |
cfe34140 | 765 | 0: Each character is represented by one byte. This is always |
f6302ac9 | 766 | true for unibyte text. |
2e344af3 KH |
767 | 1: No charsets other than ascii eight-bit-control, |
768 | eight-bit-graphic, and latin-1 are found. | |
769 | 2: Otherwise. | |
1d67c29b | 770 | |
2e344af3 KH |
771 | In addition, if CHARSETS is nonzero, for each found charset N, set |
772 | CHARSETS[N] to 1. For that, callers should allocate CHARSETS | |
773 | (MAX_CHARSET + 1 elements) in advance. It may lookup a translation | |
774 | table TABLE if supplied. For invalid charsets, set CHARSETS[1] to | |
775 | 1 (note that there's no charset whose ID is 1). */ | |
4ed46869 KH |
776 | |
777 | int | |
2e344af3 KH |
778 | find_charset_in_text (ptr, nchars, nbytes, charsets, table) |
779 | unsigned char *ptr; | |
780 | int nchars, nbytes, *charsets; | |
23d2a7f1 | 781 | Lisp_Object table; |
4ed46869 | 782 | { |
2e344af3 | 783 | if (nchars == nbytes) |
0282eb69 | 784 | { |
2e344af3 | 785 | if (charsets && nbytes > 0) |
0282eb69 | 786 | { |
2e344af3 KH |
787 | unsigned char *endp = ptr + nbytes; |
788 | int maskbits = 0; | |
789 | ||
790 | while (ptr < endp && maskbits != 7) | |
791 | { | |
792 | maskbits |= (*ptr < 0x80 ? 1 : *ptr < 0xA0 ? 2 : 4); | |
793 | ptr++; | |
794 | } | |
795 | ||
796 | if (maskbits & 1) | |
797 | charsets[CHARSET_ASCII] = 1; | |
798 | if (maskbits & 2) | |
799 | charsets[CHARSET_8_BIT_CONTROL] = 1; | |
800 | if (maskbits & 4) | |
801 | charsets[CHARSET_8_BIT_GRAPHIC] = 1; | |
0282eb69 | 802 | } |
2e344af3 | 803 | return 0; |
0282eb69 | 804 | } |
2e344af3 | 805 | else |
4ed46869 | 806 | { |
2e344af3 | 807 | int return_val = 1; |
99529c2c | 808 | int bytes, charset, c1, c2; |
05505664 | 809 | |
2e344af3 KH |
810 | if (! CHAR_TABLE_P (table)) |
811 | table = Qnil; | |
05505664 | 812 | |
2e344af3 | 813 | while (nchars-- > 0) |
23d2a7f1 | 814 | { |
2e344af3 KH |
815 | SPLIT_MULTIBYTE_SEQ (ptr, len, bytes, charset, c1, c2); |
816 | ptr += bytes; | |
4ed46869 | 817 | |
2e344af3 KH |
818 | if (!CHARSET_DEFINED_P (charset)) |
819 | charset = 1; | |
820 | else if (! NILP (table)) | |
821 | { | |
822 | int c = translate_char (table, -1, charset, c1, c2); | |
823 | if (c >= 0) | |
824 | charset = CHAR_CHARSET (c); | |
825 | } | |
826 | ||
827 | if (return_val == 1 | |
828 | && charset != CHARSET_ASCII | |
829 | && charset != CHARSET_8_BIT_CONTROL | |
830 | && charset != CHARSET_8_BIT_GRAPHIC | |
831 | && charset != charset_latin_iso8859_1) | |
832 | return_val = 2; | |
833 | ||
834 | if (charsets) | |
835 | charsets[charset] = 1; | |
836 | else if (return_val == 2) | |
837 | break; | |
4ed46869 | 838 | } |
2e344af3 | 839 | return return_val; |
4ed46869 | 840 | } |
4ed46869 KH |
841 | } |
842 | ||
843 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 844 | 2, 3, 0, |
4ed46869 | 845 | "Return a list of charsets in the region between BEG and END.\n\ |
23d2a7f1 | 846 | BEG and END are buffer positions.\n\ |
0282eb69 KH |
847 | Optional arg TABLE if non-nil is a translation table to look up.\n\ |
848 | \n\ | |
365dd325 | 849 | If the region contains invalid multibyte characters,\n\ |
38f02ede | 850 | `unknown' is included in the returned list.\n\ |
0282eb69 | 851 | \n\ |
2e344af3 KH |
852 | If the current buffer is unibyte, the returned list may contain\n\ |
853 | only `ascii', `eight-bit-control', and `eight-bit-graphic'.") | |
23d2a7f1 KH |
854 | (beg, end, table) |
855 | Lisp_Object beg, end, table; | |
4ed46869 | 856 | { |
028d516b | 857 | int charsets[MAX_CHARSET + 1]; |
6ae1f27e | 858 | int from, from_byte, to, stop, stop_byte, i; |
4ed46869 KH |
859 | Lisp_Object val; |
860 | ||
861 | validate_region (&beg, &end); | |
862 | from = XFASTINT (beg); | |
863 | stop = to = XFASTINT (end); | |
6ae1f27e | 864 | |
4ed46869 | 865 | if (from < GPT && GPT < to) |
6ae1f27e RS |
866 | { |
867 | stop = GPT; | |
868 | stop_byte = GPT_BYTE; | |
869 | } | |
870 | else | |
871 | stop_byte = CHAR_TO_BYTE (stop); | |
872 | ||
873 | from_byte = CHAR_TO_BYTE (from); | |
874 | ||
028d516b | 875 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
4ed46869 KH |
876 | while (1) |
877 | { | |
2e344af3 KH |
878 | find_charset_in_text (BYTE_POS_ADDR (from_byte), stop - from, |
879 | stop_byte - from_byte, charsets, table); | |
4ed46869 | 880 | if (stop < to) |
6ae1f27e RS |
881 | { |
882 | from = stop, from_byte = stop_byte; | |
883 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
884 | } | |
4ed46869 KH |
885 | else |
886 | break; | |
887 | } | |
6ae1f27e | 888 | |
4ed46869 | 889 | val = Qnil; |
2e344af3 | 890 | if (charsets[1]) |
0282eb69 | 891 | val = Fcons (Qunknown, val); |
2e344af3 KH |
892 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
893 | if (charsets[i]) | |
894 | val = Fcons (CHARSET_SYMBOL (i), val); | |
895 | if (charsets[0]) | |
896 | val = Fcons (Qascii, val); | |
4ed46869 KH |
897 | return val; |
898 | } | |
899 | ||
900 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 KH |
901 | 1, 2, 0, |
902 | "Return a list of charsets in STR.\n\ | |
0282eb69 KH |
903 | Optional arg TABLE if non-nil is a translation table to look up.\n\ |
904 | \n\ | |
365dd325 | 905 | If the string contains invalid multibyte characters,\n\ |
0282eb69 KH |
906 | `unknown' is included in the returned list.\n\ |
907 | \n\ | |
2e344af3 KH |
908 | If STR is unibyte, the returned list may contain\n\ |
909 | only `ascii', `eight-bit-control', and `eight-bit-graphic'.") | |
23d2a7f1 KH |
910 | (str, table) |
911 | Lisp_Object str, table; | |
4ed46869 | 912 | { |
a29e3b1b | 913 | int charsets[MAX_CHARSET + 1]; |
4ed46869 KH |
914 | int i; |
915 | Lisp_Object val; | |
916 | ||
917 | CHECK_STRING (str, 0); | |
87b089ad | 918 | |
a29e3b1b | 919 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
2e344af3 KH |
920 | find_charset_in_text (XSTRING (str)->data, XSTRING (str)->size, |
921 | STRING_BYTES (XSTRING (str)), charsets, table); | |
922 | ||
4ed46869 | 923 | val = Qnil; |
2e344af3 | 924 | if (charsets[1]) |
0282eb69 | 925 | val = Fcons (Qunknown, val); |
2e344af3 KH |
926 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
927 | if (charsets[i]) | |
928 | val = Fcons (CHARSET_SYMBOL (i), val); | |
929 | if (charsets[0]) | |
930 | val = Fcons (Qascii, val); | |
4ed46869 KH |
931 | return val; |
932 | } | |
2e344af3 | 933 | |
4ed46869 KH |
934 | \f |
935 | DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
513ee442 | 936 | "") |
4ed46869 KH |
937 | (charset, code1, code2) |
938 | Lisp_Object charset, code1, code2; | |
939 | { | |
ac4137cc KH |
940 | int charset_id, c1, c2; |
941 | ||
4ed46869 | 942 | CHECK_NUMBER (charset, 0); |
ac4137cc KH |
943 | charset_id = XINT (charset); |
944 | if (!CHARSET_DEFINED_P (charset_id)) | |
945 | error ("Invalid charset ID: %d", XINT (charset)); | |
4ed46869 KH |
946 | |
947 | if (NILP (code1)) | |
ac4137cc | 948 | c1 = 0; |
4ed46869 | 949 | else |
ac4137cc KH |
950 | { |
951 | CHECK_NUMBER (code1, 1); | |
952 | c1 = XINT (code1); | |
953 | } | |
4ed46869 | 954 | if (NILP (code2)) |
ac4137cc | 955 | c2 = 0; |
4ed46869 | 956 | else |
ac4137cc KH |
957 | { |
958 | CHECK_NUMBER (code2, 2); | |
959 | c2 = XINT (code2); | |
960 | } | |
4ed46869 | 961 | |
2e344af3 KH |
962 | if (charset_id == CHARSET_ASCII) |
963 | { | |
964 | if (c1 < 0 || c1 > 0x7F) | |
965 | goto invalid_code_posints; | |
966 | return make_number (c1); | |
967 | } | |
968 | else if (charset_id == CHARSET_8_BIT_CONTROL) | |
969 | { | |
30736012 KH |
970 | if (NILP (code1)) |
971 | c1 = 0x80; | |
972 | else if (c1 < 0x80 || c1 > 0x9F) | |
2e344af3 KH |
973 | goto invalid_code_posints; |
974 | return make_number (c1); | |
975 | } | |
976 | else if (charset_id == CHARSET_8_BIT_GRAPHIC) | |
977 | { | |
30736012 KH |
978 | if (NILP (code1)) |
979 | c1 = 0xA0; | |
980 | else if (c1 < 0xA0 || c1 > 0xFF) | |
2e344af3 KH |
981 | goto invalid_code_posints; |
982 | return make_number (c1); | |
983 | } | |
984 | else if (c1 < 0 || c1 > 0xFF || c2 < 0 || c2 > 0xFF) | |
985 | goto invalid_code_posints; | |
ac4137cc KH |
986 | c1 &= 0x7F; |
987 | c2 &= 0x7F; | |
988 | if (c1 == 0 | |
989 | ? c2 != 0 | |
990 | : (c2 == 0 | |
87f67317 KR |
991 | ? !CHAR_COMPONENTS_VALID_P (charset_id, c1, 0x20) |
992 | : !CHAR_COMPONENTS_VALID_P (charset_id, c1, c2))) | |
2e344af3 | 993 | goto invalid_code_posints; |
ac4137cc | 994 | return make_number (MAKE_CHAR (charset_id, c1, c2)); |
2e344af3 KH |
995 | |
996 | invalid_code_posints: | |
997 | error ("Invalid code points for charset ID %d: %d %d", charset_id, c1, c2); | |
4ed46869 KH |
998 | } |
999 | ||
1000 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, | |
0282eb69 KH |
1001 | "Return list of charset and one or two position-codes of CHAR.\n\ |
1002 | If CHAR is invalid as a character code,\n\ | |
1003 | return a list of symbol `unknown' and CHAR.") | |
4ed46869 KH |
1004 | (ch) |
1005 | Lisp_Object ch; | |
1006 | { | |
0282eb69 | 1007 | int c, charset, c1, c2; |
4ed46869 KH |
1008 | |
1009 | CHECK_NUMBER (ch, 0); | |
0282eb69 KH |
1010 | c = XFASTINT (ch); |
1011 | if (!CHAR_VALID_P (c, 1)) | |
1012 | return Fcons (Qunknown, Fcons (ch, Qnil)); | |
4ed46869 | 1013 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
6dc0722d | 1014 | return (c2 >= 0 |
4ed46869 KH |
1015 | ? Fcons (CHARSET_SYMBOL (charset), |
1016 | Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
1017 | : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
1018 | } | |
1019 | ||
1020 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
1021 | "Return charset of CHAR.") | |
1022 | (ch) | |
1023 | Lisp_Object ch; | |
1024 | { | |
1025 | CHECK_NUMBER (ch, 0); | |
1026 | ||
1027 | return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
1028 | } | |
1029 | ||
90d7b74e | 1030 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
ac4137cc | 1031 | "Return charset of a character in the current buffer at position POS.\n\ |
e6e114f2 KH |
1032 | If POS is nil, it defauls to the current point.\n\ |
1033 | If POS is out of range, the value is nil.") | |
90d7b74e KH |
1034 | (pos) |
1035 | Lisp_Object pos; | |
1036 | { | |
2e344af3 KH |
1037 | Lisp_Object ch; |
1038 | int charset; | |
ac4137cc | 1039 | |
2e344af3 KH |
1040 | ch = Fchar_after (pos); |
1041 | if (! INTEGERP (ch)) | |
1042 | return ch; | |
1043 | charset = CHAR_CHARSET (XINT (ch)); | |
90d7b74e KH |
1044 | return CHARSET_SYMBOL (charset); |
1045 | } | |
1046 | ||
4ed46869 | 1047 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
2b71bb78 KH |
1048 | "Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR.\n\ |
1049 | \n\ | |
1050 | ISO 2022's designation sequence (escape sequence) distinguishes charsets\n\ | |
1051 | by their DIMENSION, CHARS, and FINAL-CHAR,\n\ | |
1052 | where as Emacs distinguishes them by charset symbol.\n\ | |
1053 | See the documentation of the function `charset-info' for the meanings of\n\ | |
1054 | DIMENSION, CHARS, and FINAL-CHAR.") | |
4ed46869 KH |
1055 | (dimension, chars, final_char) |
1056 | Lisp_Object dimension, chars, final_char; | |
1057 | { | |
1058 | int charset; | |
1059 | ||
1060 | CHECK_NUMBER (dimension, 0); | |
1061 | CHECK_NUMBER (chars, 1); | |
1062 | CHECK_NUMBER (final_char, 2); | |
1063 | ||
1064 | if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
1065 | return Qnil; | |
1066 | return CHARSET_SYMBOL (charset); | |
1067 | } | |
1068 | ||
9d3d8cba KH |
1069 | /* If GENERICP is nonzero, return nonzero iff C is a valid normal or |
1070 | generic character. If GENERICP is zero, return nonzero iff C is a | |
1071 | valid normal character. Do not call this function directly, | |
1072 | instead use macro CHAR_VALID_P. */ | |
1073 | int | |
1074 | char_valid_p (c, genericp) | |
1075 | int c, genericp; | |
1076 | { | |
1077 | int charset, c1, c2; | |
1078 | ||
0e235b7e | 1079 | if (c < 0 || c >= MAX_CHAR) |
9d3d8cba KH |
1080 | return 0; |
1081 | if (SINGLE_BYTE_CHAR_P (c)) | |
1082 | return 1; | |
2e344af3 | 1083 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
1084 | if (genericp) |
1085 | { | |
1086 | if (c1) | |
1087 | { | |
1088 | if (c2 <= 0) c2 = 0x20; | |
1089 | } | |
1090 | else | |
1091 | { | |
1092 | if (c2 <= 0) c1 = c2 = 0x20; | |
1093 | } | |
1094 | } | |
1095 | return (CHARSET_DEFINED_P (charset) | |
44c6492d | 1096 | && CHAR_COMPONENTS_VALID_P (charset, c1, c2)); |
9d3d8cba KH |
1097 | } |
1098 | ||
1099 | DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0, | |
a9d02884 DL |
1100 | "Return t if OBJECT is a valid normal character.\n\ |
1101 | If optional arg GENERICP is non-nil, also return t if OBJECT is\n\ | |
9d3d8cba KH |
1102 | a valid generic character.") |
1103 | (object, genericp) | |
1104 | Lisp_Object object, genericp; | |
1105 | { | |
1106 | if (! NATNUMP (object)) | |
1107 | return Qnil; | |
1108 | return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil); | |
1109 | } | |
1110 | ||
d2665018 KH |
1111 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, |
1112 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
1113 | "Convert the unibyte character CH to multibyte character.\n\ | |
537efd8d | 1114 | The conversion is done based on `nonascii-translation-table' (which see)\n\ |
340b8d58 | 1115 | or `nonascii-insert-offset' (which see).") |
d2665018 KH |
1116 | (ch) |
1117 | Lisp_Object ch; | |
1118 | { | |
1119 | int c; | |
1120 | ||
1121 | CHECK_NUMBER (ch, 0); | |
1122 | c = XINT (ch); | |
1123 | if (c < 0 || c >= 0400) | |
1124 | error ("Invalid unibyte character: %d", c); | |
1125 | c = unibyte_char_to_multibyte (c); | |
1126 | if (c < 0) | |
1127 | error ("Can't convert to multibyte character: %d", XINT (ch)); | |
1128 | return make_number (c); | |
1129 | } | |
1130 | ||
1bcc1567 RS |
1131 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, |
1132 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
1133 | "Convert the multibyte character CH to unibyte character.\n\ | |
1134 | The conversion is done based on `nonascii-translation-table' (which see)\n\ | |
1135 | or `nonascii-insert-offset' (which see).") | |
1136 | (ch) | |
1137 | Lisp_Object ch; | |
1138 | { | |
1139 | int c; | |
1140 | ||
1141 | CHECK_NUMBER (ch, 0); | |
1142 | c = XINT (ch); | |
ac4137cc | 1143 | if (! CHAR_VALID_P (c, 0)) |
1bcc1567 RS |
1144 | error ("Invalid multibyte character: %d", c); |
1145 | c = multibyte_char_to_unibyte (c, Qnil); | |
1146 | if (c < 0) | |
1147 | error ("Can't convert to unibyte character: %d", XINT (ch)); | |
1148 | return make_number (c); | |
1149 | } | |
1150 | ||
4ed46869 | 1151 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, |
f78643ef | 1152 | "Return 1 regardless of the argument CHAR.\n\ |
60022cb7 | 1153 | This is now an obsolete function. We keep it just for backward compatibility.") |
4ed46869 KH |
1154 | (ch) |
1155 | Lisp_Object ch; | |
1156 | { | |
4ed46869 | 1157 | CHECK_NUMBER (ch, 0); |
9b6a601f KH |
1158 | return make_number (1); |
1159 | } | |
1160 | ||
1161 | /* Return how many bytes C will occupy in a multibyte buffer. | |
1162 | Don't call this function directly, instead use macro CHAR_BYTES. */ | |
1163 | int | |
1164 | char_bytes (c) | |
1165 | int c; | |
1166 | { | |
99529c2c | 1167 | int charset; |
9b6a601f | 1168 | |
2e344af3 KH |
1169 | if (ASCII_BYTE_P (c) || (c & ~((1 << CHARACTERBITS) -1))) |
1170 | return 1; | |
1171 | if (SINGLE_BYTE_CHAR_P (c) && c >= 0xA0) | |
8ac5a9cc KH |
1172 | return 1; |
1173 | ||
99529c2c KH |
1174 | charset = CHAR_CHARSET (c); |
1175 | return (CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1); | |
4ed46869 KH |
1176 | } |
1177 | ||
1178 | /* Return the width of character of which multi-byte form starts with | |
1179 | C. The width is measured by how many columns occupied on the | |
1180 | screen when displayed in the current buffer. */ | |
1181 | ||
1182 | #define ONE_BYTE_CHAR_WIDTH(c) \ | |
1183 | (c < 0x20 \ | |
1184 | ? (c == '\t' \ | |
53316e55 | 1185 | ? XFASTINT (current_buffer->tab_width) \ |
4ed46869 KH |
1186 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
1187 | : (c < 0x7f \ | |
1188 | ? 1 \ | |
1189 | : (c == 0x7F \ | |
1190 | ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
1191 | : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
1192 | && BASE_LEADING_CODE_P (c)) \ | |
1193 | ? WIDTH_BY_CHAR_HEAD (c) \ | |
b4e9dd77 | 1194 | : 4)))) |
4ed46869 KH |
1195 | |
1196 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
1197 | "Return width of CHAR when displayed in the current buffer.\n\ | |
4245f3d2 KH |
1198 | The width is measured by how many columns it occupies on the screen.\n\ |
1199 | Tab is taken to occupy `tab-width' columns.") | |
4ed46869 KH |
1200 | (ch) |
1201 | Lisp_Object ch; | |
1202 | { | |
859f2b3c | 1203 | Lisp_Object val, disp; |
4ed46869 | 1204 | int c; |
51c4025f | 1205 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 KH |
1206 | |
1207 | CHECK_NUMBER (ch, 0); | |
1208 | ||
859f2b3c RS |
1209 | c = XINT (ch); |
1210 | ||
1211 | /* Get the way the display table would display it. */ | |
51c4025f | 1212 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; |
859f2b3c RS |
1213 | |
1214 | if (VECTORP (disp)) | |
1215 | XSETINT (val, XVECTOR (disp)->size); | |
1216 | else if (SINGLE_BYTE_CHAR_P (c)) | |
1217 | XSETINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
4ed46869 KH |
1218 | else |
1219 | { | |
1220 | int charset = CHAR_CHARSET (c); | |
1221 | ||
1222 | XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
1223 | } | |
1224 | return val; | |
1225 | } | |
1226 | ||
1227 | /* Return width of string STR of length LEN when displayed in the | |
1228 | current buffer. The width is measured by how many columns it | |
1229 | occupies on the screen. */ | |
859f2b3c | 1230 | |
4ed46869 KH |
1231 | int |
1232 | strwidth (str, len) | |
1233 | unsigned char *str; | |
1234 | int len; | |
1235 | { | |
beeedaad KH |
1236 | return c_string_width (str, len, -1, NULL, NULL); |
1237 | } | |
1238 | ||
1239 | /* Return width of string STR of length LEN when displayed in the | |
1240 | current buffer. The width is measured by how many columns it | |
1241 | occupies on the screen. If PRECISION > 0, return the width of | |
1242 | longest substring that doesn't exceed PRECISION, and set number of | |
1243 | characters and bytes of the substring in *NCHARS and *NBYTES | |
1244 | respectively. */ | |
1245 | ||
1246 | int | |
1247 | c_string_width (str, len, precision, nchars, nbytes) | |
1248 | unsigned char *str; | |
1249 | int precision, *nchars, *nbytes; | |
1250 | { | |
1251 | int i = 0, i_byte = 0; | |
4ed46869 | 1252 | int width = 0; |
beeedaad | 1253 | int chars; |
c4a4e28f | 1254 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1255 | |
beeedaad | 1256 | while (i_byte < len) |
859f2b3c | 1257 | { |
beeedaad KH |
1258 | int bytes, thiswidth; |
1259 | Lisp_Object val; | |
859f2b3c | 1260 | |
99529c2c | 1261 | if (dp) |
beeedaad KH |
1262 | { |
1263 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
1264 | ||
1265 | chars = 1; | |
1266 | val = DISP_CHAR_VECTOR (dp, c); | |
1267 | if (VECTORP (val)) | |
1268 | thiswidth = XVECTOR (val)->size; | |
1269 | else | |
1270 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1271 | } | |
859f2b3c | 1272 | else |
beeedaad KH |
1273 | { |
1274 | chars = 1; | |
1275 | PARSE_MULTIBYTE_SEQ (str + i_byte, len - i_byte, bytes); | |
1276 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1277 | } | |
99529c2c | 1278 | |
beeedaad KH |
1279 | if (precision > 0 |
1280 | && (width + thiswidth > precision)) | |
1281 | { | |
1282 | *nchars = i; | |
1283 | *nbytes = i_byte; | |
1284 | return width; | |
1285 | } | |
1286 | i++; | |
1287 | i_byte += bytes; | |
1288 | width += thiswidth; | |
1289 | } | |
99529c2c | 1290 | |
beeedaad KH |
1291 | if (precision > 0) |
1292 | { | |
1293 | *nchars = i; | |
1294 | *nbytes = i_byte; | |
859f2b3c | 1295 | } |
beeedaad | 1296 | |
4ed46869 KH |
1297 | return width; |
1298 | } | |
1299 | ||
beeedaad KH |
1300 | /* Return width of Lisp string STRING when displayed in the current |
1301 | buffer. The width is measured by how many columns it occupies on | |
1302 | the screen while paying attention to compositions. If PRECISION > | |
1303 | 0, return the width of longest substring that doesn't exceed | |
1304 | PRECISION, and set number of characters and bytes of the substring | |
1305 | in *NCHARS and *NBYTES respectively. */ | |
1306 | ||
3f62427c | 1307 | int |
beeedaad KH |
1308 | lisp_string_width (string, precision, nchars, nbytes) |
1309 | Lisp_Object string; | |
1310 | int precision, *nchars, *nbytes; | |
3f62427c | 1311 | { |
beeedaad KH |
1312 | int len = XSTRING (string)->size; |
1313 | int len_byte = STRING_BYTES (XSTRING (string)); | |
1314 | unsigned char *str = XSTRING (string)->data; | |
1315 | int i = 0, i_byte = 0; | |
3f62427c | 1316 | int width = 0; |
beeedaad | 1317 | struct Lisp_Char_Table *dp = buffer_display_table (); |
3f62427c KH |
1318 | |
1319 | while (i < len) | |
1320 | { | |
beeedaad KH |
1321 | int chars, bytes, thiswidth; |
1322 | Lisp_Object val; | |
1323 | int cmp_id; | |
1324 | int ignore, end; | |
1325 | ||
1326 | if (find_composition (i, -1, &ignore, &end, &val, string) | |
1327 | && ((cmp_id = get_composition_id (i, i_byte, end - i, val, string)) | |
1328 | >= 0)) | |
3f62427c | 1329 | { |
beeedaad KH |
1330 | thiswidth = composition_table[cmp_id]->width; |
1331 | chars = end - i; | |
1332 | bytes = string_char_to_byte (string, end) - i_byte; | |
1333 | } | |
1334 | else if (dp) | |
1335 | { | |
1336 | int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes); | |
1337 | ||
1338 | chars = 1; | |
1339 | val = DISP_CHAR_VECTOR (dp, c); | |
1340 | if (VECTORP (val)) | |
1341 | thiswidth = XVECTOR (val)->size; | |
1342 | else | |
1343 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
3f62427c KH |
1344 | } |
1345 | else | |
1346 | { | |
beeedaad KH |
1347 | chars = 1; |
1348 | PARSE_MULTIBYTE_SEQ (str + i_byte, len_byte - i_byte, bytes); | |
1349 | thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]); | |
1350 | } | |
1351 | ||
1352 | if (precision > 0 | |
1353 | && (width + thiswidth > precision)) | |
1354 | { | |
1355 | *nchars = i; | |
1356 | *nbytes = i_byte; | |
1357 | return width; | |
3f62427c | 1358 | } |
beeedaad KH |
1359 | i += chars; |
1360 | i_byte += bytes; | |
1361 | width += thiswidth; | |
1362 | } | |
1363 | ||
1364 | if (precision > 0) | |
1365 | { | |
1366 | *nchars = i; | |
1367 | *nbytes = i_byte; | |
3f62427c | 1368 | } |
beeedaad | 1369 | |
3f62427c KH |
1370 | return width; |
1371 | } | |
1372 | ||
4ed46869 KH |
1373 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, |
1374 | "Return width of STRING when displayed in the current buffer.\n\ | |
1375 | Width is measured by how many columns it occupies on the screen.\n\ | |
046b1f03 RS |
1376 | When calculating width of a multibyte character in STRING,\n\ |
1377 | only the base leading-code is considered; the validity of\n\ | |
5af47d5f GM |
1378 | the following bytes is not checked. Tabs in STRING are always\n\ |
1379 | taken to occupy `tab-width' columns.") | |
4ed46869 KH |
1380 | (str) |
1381 | Lisp_Object str; | |
1382 | { | |
1383 | Lisp_Object val; | |
1384 | ||
1385 | CHECK_STRING (str, 0); | |
beeedaad | 1386 | XSETFASTINT (val, lisp_string_width (str, -1, NULL, NULL)); |
4ed46869 KH |
1387 | return val; |
1388 | } | |
1389 | ||
1390 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
1391 | "Return the direction of CHAR.\n\ | |
1392 | The returned value is 0 for left-to-right and 1 for right-to-left.") | |
1393 | (ch) | |
1394 | Lisp_Object ch; | |
1395 | { | |
1396 | int charset; | |
1397 | ||
1398 | CHECK_NUMBER (ch, 0); | |
1399 | charset = CHAR_CHARSET (XFASTINT (ch)); | |
1400 | if (!CHARSET_DEFINED_P (charset)) | |
93bcb785 | 1401 | invalid_character (XINT (ch)); |
4ed46869 KH |
1402 | return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); |
1403 | } | |
1404 | ||
af4fecb4 | 1405 | DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, |
6ae1f27e | 1406 | "Return number of characters between BEG and END.") |
046b1f03 RS |
1407 | (beg, end) |
1408 | Lisp_Object beg, end; | |
1409 | { | |
6ae1f27e | 1410 | int from, to; |
046b1f03 | 1411 | |
17e7ef1b RS |
1412 | CHECK_NUMBER_COERCE_MARKER (beg, 0); |
1413 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
1414 | ||
046b1f03 | 1415 | from = min (XFASTINT (beg), XFASTINT (end)); |
a8a35e61 | 1416 | to = max (XFASTINT (beg), XFASTINT (end)); |
046b1f03 | 1417 | |
a8c21066 | 1418 | return make_number (to - from); |
6ae1f27e | 1419 | } |
9036eb45 | 1420 | |
87b089ad RS |
1421 | /* Return the number of characters in the NBYTES bytes at PTR. |
1422 | This works by looking at the contents and checking for multibyte sequences. | |
1423 | However, if the current buffer has enable-multibyte-characters = nil, | |
1424 | we treat each byte as a character. */ | |
1425 | ||
6ae1f27e RS |
1426 | int |
1427 | chars_in_text (ptr, nbytes) | |
1428 | unsigned char *ptr; | |
1429 | int nbytes; | |
1430 | { | |
87b089ad RS |
1431 | /* current_buffer is null at early stages of Emacs initialization. */ |
1432 | if (current_buffer == 0 | |
1433 | || NILP (current_buffer->enable_multibyte_characters)) | |
6ae1f27e | 1434 | return nbytes; |
a8a35e61 | 1435 | |
ac4137cc | 1436 | return multibyte_chars_in_text (ptr, nbytes); |
046b1f03 RS |
1437 | } |
1438 | ||
87b089ad RS |
1439 | /* Return the number of characters in the NBYTES bytes at PTR. |
1440 | This works by looking at the contents and checking for multibyte sequences. | |
1441 | It ignores enable-multibyte-characters. */ | |
1442 | ||
1443 | int | |
1444 | multibyte_chars_in_text (ptr, nbytes) | |
1445 | unsigned char *ptr; | |
1446 | int nbytes; | |
1447 | { | |
ac4137cc KH |
1448 | unsigned char *endp; |
1449 | int chars, bytes; | |
87b089ad RS |
1450 | |
1451 | endp = ptr + nbytes; | |
1452 | chars = 0; | |
1453 | ||
1454 | while (ptr < endp) | |
1455 | { | |
2e344af3 KH |
1456 | PARSE_MULTIBYTE_SEQ (ptr, endp - ptr, bytes); |
1457 | ptr += bytes; | |
1458 | chars++; | |
1459 | } | |
1460 | ||
1461 | return chars; | |
1462 | } | |
1463 | ||
d8e4f486 | 1464 | /* Parse unibyte text at STR of LEN bytes as multibyte text, and |
2e344af3 | 1465 | count the numbers of characters and bytes in it. On counting |
d8e4f486 DL |
1466 | bytes, pay attention to the fact that 8-bit characters in the range |
1467 | 0x80..0x9F are represented by 2 bytes in multibyte text. */ | |
2e344af3 KH |
1468 | void |
1469 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
1470 | unsigned char *str; | |
1471 | int len, *nchars, *nbytes; | |
1472 | { | |
1473 | unsigned char *endp = str + len; | |
1474 | int n, chars = 0, bytes = 0; | |
1475 | ||
1476 | while (str < endp) | |
1477 | { | |
1478 | if (UNIBYTE_STR_AS_MULTIBYTE_P (str, endp - str, n)) | |
1479 | str += n, bytes += n; | |
1480 | else | |
1481 | str++, bytes += 2; | |
1482 | chars++; | |
1483 | } | |
1484 | *nchars = chars; | |
1485 | *nbytes = bytes; | |
1486 | return; | |
1487 | } | |
1488 | ||
d8e4f486 | 1489 | /* Arrange unibyte text at STR of NBYTES bytes as multibyte text. |
2e344af3 KH |
1490 | It actually converts only 8-bit characters in the range 0x80..0x9F |
1491 | that don't contruct multibyte characters to multibyte forms. If | |
1492 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
1493 | text. It is assured that we can use LEN bytes at STR as a work | |
1494 | area and that is enough. Return the number of bytes of the | |
1495 | resulting text. */ | |
1496 | ||
1497 | int | |
1498 | str_as_multibyte (str, len, nbytes, nchars) | |
1499 | unsigned char *str; | |
1500 | int len, nbytes, *nchars; | |
1501 | { | |
1502 | unsigned char *p = str, *endp = str + nbytes; | |
1503 | unsigned char *to; | |
1504 | int chars = 0; | |
1505 | int n; | |
1506 | ||
1507 | while (p < endp && UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
1508 | p += n, chars++; | |
1509 | if (nchars) | |
1510 | *nchars = chars; | |
1511 | if (p == endp) | |
1512 | return nbytes; | |
1513 | ||
1514 | to = p; | |
1515 | nbytes = endp - p; | |
1516 | endp = str + len; | |
1517 | safe_bcopy (p, endp - nbytes, nbytes); | |
1518 | p = endp - nbytes; | |
1519 | while (p < endp) | |
1520 | { | |
1521 | if (UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
ac4137cc | 1522 | { |
2e344af3 KH |
1523 | while (n--) |
1524 | *to++ = *p++; | |
1525 | } | |
ac4137cc KH |
1526 | else |
1527 | { | |
2e344af3 KH |
1528 | *to++ = LEADING_CODE_8_BIT_CONTROL; |
1529 | *to++ = *p++ + 0x20; | |
ac4137cc | 1530 | } |
87b089ad RS |
1531 | chars++; |
1532 | } | |
2e344af3 KH |
1533 | if (nchars) |
1534 | *nchars = chars; | |
1535 | return (to - str); | |
1536 | } | |
87b089ad | 1537 | |
d8e4f486 | 1538 | /* Convert unibyte text at STR of NBYTES bytes to multibyte text |
2e344af3 KH |
1539 | that contains the same single-byte characters. It actually |
1540 | converts all 8-bit characters to multibyte forms. It is assured | |
1541 | that we can use LEN bytes at STR as a work area and that is | |
1542 | enough. */ | |
1543 | ||
1544 | int | |
1545 | str_to_multibyte (str, len, bytes) | |
1546 | unsigned char *str; | |
1547 | int len, bytes; | |
1548 | { | |
1549 | unsigned char *p = str, *endp = str + bytes; | |
1550 | unsigned char *to; | |
2e344af3 KH |
1551 | |
1552 | while (p < endp && (*p < 0x80 || *p >= 0xA0)) p++; | |
1553 | if (p == endp) | |
1554 | return bytes; | |
1555 | to = p; | |
1556 | bytes = endp - p; | |
1557 | endp = str + len; | |
1558 | safe_bcopy (p, endp - bytes, bytes); | |
1559 | p = endp - bytes; | |
1560 | while (p < endp) | |
1561 | { | |
1562 | if (*p < 0x80 || *p >= 0xA0) | |
1563 | *to++ = *p++; | |
1564 | else | |
1565 | *to++ = LEADING_CODE_8_BIT_CONTROL, *to++ = *p++ + 0x20; | |
1566 | } | |
1567 | return (to - str); | |
87b089ad RS |
1568 | } |
1569 | ||
2e344af3 KH |
1570 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It |
1571 | actually converts only 8-bit characters in the range 0x80..0x9F to | |
1572 | unibyte forms. */ | |
1573 | ||
1574 | int | |
1575 | str_as_unibyte (str, bytes) | |
1576 | unsigned char *str; | |
1577 | int bytes; | |
1578 | { | |
1579 | unsigned char *p = str, *endp = str + bytes; | |
1580 | unsigned char *to = str; | |
1581 | ||
1582 | while (p < endp && *p != LEADING_CODE_8_BIT_CONTROL) p++; | |
1583 | to = p; | |
1584 | while (p < endp) | |
1585 | { | |
1586 | if (*p == LEADING_CODE_8_BIT_CONTROL) | |
1587 | *to++ = *(p + 1) - 0x20, p += 2; | |
1588 | else | |
1589 | *to++ = *p++; | |
1590 | } | |
1591 | return (to - str); | |
1592 | } | |
1593 | ||
1594 | \f | |
87b089ad | 1595 | DEFUN ("string", Fstring, Sstring, 1, MANY, 0, |
4ed46869 | 1596 | "Concatenate all the argument characters and make the result a string.") |
53316e55 KH |
1597 | (n, args) |
1598 | int n; | |
4ed46869 KH |
1599 | Lisp_Object *args; |
1600 | { | |
53316e55 | 1601 | int i; |
99529c2c | 1602 | unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); |
4ed46869 | 1603 | unsigned char *p = buf; |
2e344af3 | 1604 | int c; |
4ed46869 KH |
1605 | |
1606 | for (i = 0; i < n; i++) | |
1607 | { | |
d0c037d8 | 1608 | CHECK_NUMBER (args[i], 0); |
4ed46869 | 1609 | c = XINT (args[i]); |
99529c2c | 1610 | p += CHAR_STRING (c, p); |
4ed46869 KH |
1611 | } |
1612 | ||
2e344af3 | 1613 | return make_string_from_bytes (buf, n, p - buf); |
4ed46869 KH |
1614 | } |
1615 | ||
1616 | #endif /* emacs */ | |
1617 | \f | |
dfcf069d | 1618 | int |
4ed46869 KH |
1619 | charset_id_internal (charset_name) |
1620 | char *charset_name; | |
1621 | { | |
76d7b829 | 1622 | Lisp_Object val; |
4ed46869 | 1623 | |
76d7b829 | 1624 | val= Fget (intern (charset_name), Qcharset); |
4ed46869 KH |
1625 | if (!VECTORP (val)) |
1626 | error ("Charset %s is not defined", charset_name); | |
1627 | ||
1628 | return (XINT (XVECTOR (val)->contents[0])); | |
1629 | } | |
1630 | ||
1631 | DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
1632 | Ssetup_special_charsets, 0, 0, 0, "Internal use only.") | |
1633 | () | |
1634 | { | |
1635 | charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1636 | charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1637 | charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1638 | charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1639 | charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1640 | charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1641 | charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1642 | return Qnil; | |
1643 | } | |
1644 | ||
dfcf069d | 1645 | void |
4ed46869 KH |
1646 | init_charset_once () |
1647 | { | |
1648 | int i, j, k; | |
1649 | ||
1650 | staticpro (&Vcharset_table); | |
1651 | staticpro (&Vcharset_symbol_table); | |
8a73a704 | 1652 | staticpro (&Vgeneric_character_list); |
4ed46869 KH |
1653 | |
1654 | /* This has to be done here, before we call Fmake_char_table. */ | |
1655 | Qcharset_table = intern ("charset-table"); | |
1656 | staticpro (&Qcharset_table); | |
1657 | ||
1658 | /* Intern this now in case it isn't already done. | |
1659 | Setting this variable twice is harmless. | |
1660 | But don't staticpro it here--that is done in alloc.c. */ | |
1661 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1662 | ||
1663 | /* Now we are ready to set up this property, so we can | |
1664 | create the charset table. */ | |
1665 | Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1666 | Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1667 | ||
0282eb69 KH |
1668 | Qunknown = intern ("unknown"); |
1669 | staticpro (&Qunknown); | |
1670 | Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), | |
1671 | Qunknown); | |
4ed46869 KH |
1672 | |
1673 | /* Setup tables. */ | |
1674 | for (i = 0; i < 2; i++) | |
1675 | for (j = 0; j < 2; j++) | |
1676 | for (k = 0; k < 128; k++) | |
1677 | iso_charset_table [i][j][k] = -1; | |
1678 | ||
60383934 | 1679 | for (i = 0; i < 256; i++) |
2e344af3 | 1680 | bytes_by_char_head[i] = 1; |
2e344af3 KH |
1681 | bytes_by_char_head[LEADING_CODE_PRIVATE_11] = 3; |
1682 | bytes_by_char_head[LEADING_CODE_PRIVATE_12] = 3; | |
1683 | bytes_by_char_head[LEADING_CODE_PRIVATE_21] = 4; | |
1684 | bytes_by_char_head[LEADING_CODE_PRIVATE_22] = 4; | |
4ed46869 KH |
1685 | |
1686 | for (i = 0; i < 128; i++) | |
2e344af3 | 1687 | width_by_char_head[i] = 1; |
4ed46869 | 1688 | for (; i < 256; i++) |
2e344af3 KH |
1689 | width_by_char_head[i] = 4; |
1690 | width_by_char_head[LEADING_CODE_PRIVATE_11] = 1; | |
1691 | width_by_char_head[LEADING_CODE_PRIVATE_12] = 2; | |
1692 | width_by_char_head[LEADING_CODE_PRIVATE_21] = 1; | |
1693 | width_by_char_head[LEADING_CODE_PRIVATE_22] = 2; | |
8a73a704 KH |
1694 | |
1695 | { | |
76d7b829 | 1696 | Lisp_Object val; |
8a73a704 | 1697 | |
76d7b829 | 1698 | val = Qnil; |
8a73a704 KH |
1699 | for (i = 0x81; i < 0x90; i++) |
1700 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1701 | for (; i < 0x9A; i++) | |
1702 | val = Fcons (make_number ((i - 0x8F) << 14), val); | |
1703 | for (i = 0xA0; i < 0xF0; i++) | |
1704 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1705 | for (; i < 0xFF; i++) | |
1706 | val = Fcons (make_number ((i - 0xE0) << 14), val); | |
8a73a704 KH |
1707 | Vgeneric_character_list = Fnreverse (val); |
1708 | } | |
bbf12bb3 KH |
1709 | |
1710 | nonascii_insert_offset = 0; | |
1711 | Vnonascii_translation_table = Qnil; | |
4ed46869 KH |
1712 | } |
1713 | ||
1714 | #ifdef emacs | |
1715 | ||
dfcf069d | 1716 | void |
4ed46869 KH |
1717 | syms_of_charset () |
1718 | { | |
2e344af3 KH |
1719 | Qcharset = intern ("charset"); |
1720 | staticpro (&Qcharset); | |
1721 | ||
4ed46869 KH |
1722 | Qascii = intern ("ascii"); |
1723 | staticpro (&Qascii); | |
1724 | ||
2e344af3 KH |
1725 | Qeight_bit_control = intern ("eight-bit-control"); |
1726 | staticpro (&Qeight_bit_control); | |
1727 | ||
1728 | Qeight_bit_graphic = intern ("eight-bit-graphic"); | |
1729 | staticpro (&Qeight_bit_graphic); | |
4ed46869 | 1730 | |
2e344af3 KH |
1731 | /* Define special charsets ascii, eight-bit-control, and |
1732 | eight-bit-graphic. */ | |
4ed46869 KH |
1733 | update_charset_table (make_number (CHARSET_ASCII), |
1734 | make_number (1), make_number (94), | |
1735 | make_number (1), | |
1736 | make_number (0), | |
1737 | make_number ('B'), | |
1738 | make_number (0), | |
1739 | build_string ("ASCII"), | |
d78bc582 | 1740 | Qnil, /* same as above */ |
4ed46869 KH |
1741 | build_string ("ASCII (ISO646 IRV)")); |
1742 | CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1743 | Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1744 | ||
2e344af3 KH |
1745 | update_charset_table (make_number (CHARSET_8_BIT_CONTROL), |
1746 | make_number (1), make_number (96), | |
797a084a | 1747 | make_number (4), |
2e344af3 KH |
1748 | make_number (0), |
1749 | make_number (-1), | |
1750 | make_number (-1), | |
1751 | build_string ("8-bit control code (0x80..0x9F)"), | |
d78bc582 KH |
1752 | Qnil, /* same as above */ |
1753 | Qnil); /* same as above */ | |
2e344af3 KH |
1754 | CHARSET_SYMBOL (CHARSET_8_BIT_CONTROL) = Qeight_bit_control; |
1755 | Fput (Qeight_bit_control, Qcharset, | |
1756 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_CONTROL)); | |
1757 | ||
1758 | update_charset_table (make_number (CHARSET_8_BIT_GRAPHIC), | |
1759 | make_number (1), make_number (96), | |
797a084a | 1760 | make_number (4), |
2e344af3 KH |
1761 | make_number (0), |
1762 | make_number (-1), | |
1763 | make_number (-1), | |
2e344af3 | 1764 | build_string ("8-bit graphic char (0xA0..0xFF)"), |
d78bc582 KH |
1765 | Qnil, /* same as above */ |
1766 | Qnil); /* same as above */ | |
2e344af3 KH |
1767 | CHARSET_SYMBOL (CHARSET_8_BIT_GRAPHIC) = Qeight_bit_graphic; |
1768 | Fput (Qeight_bit_graphic, Qcharset, | |
1769 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_GRAPHIC)); | |
1770 | ||
c1a08b4c KH |
1771 | Qauto_fill_chars = intern ("auto-fill-chars"); |
1772 | staticpro (&Qauto_fill_chars); | |
1773 | Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0)); | |
1774 | ||
4ed46869 | 1775 | defsubr (&Sdefine_charset); |
8a73a704 | 1776 | defsubr (&Sgeneric_character_list); |
3fac5a51 | 1777 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
1778 | defsubr (&Sdeclare_equiv_charset); |
1779 | defsubr (&Sfind_charset_region); | |
1780 | defsubr (&Sfind_charset_string); | |
1781 | defsubr (&Smake_char_internal); | |
1782 | defsubr (&Ssplit_char); | |
1783 | defsubr (&Schar_charset); | |
90d7b74e | 1784 | defsubr (&Scharset_after); |
4ed46869 | 1785 | defsubr (&Siso_charset); |
9d3d8cba | 1786 | defsubr (&Schar_valid_p); |
d2665018 | 1787 | defsubr (&Sunibyte_char_to_multibyte); |
1bcc1567 | 1788 | defsubr (&Smultibyte_char_to_unibyte); |
4ed46869 KH |
1789 | defsubr (&Schar_bytes); |
1790 | defsubr (&Schar_width); | |
1791 | defsubr (&Sstring_width); | |
1792 | defsubr (&Schar_direction); | |
af4fecb4 | 1793 | defsubr (&Schars_in_region); |
87b089ad | 1794 | defsubr (&Sstring); |
4ed46869 KH |
1795 | defsubr (&Ssetup_special_charsets); |
1796 | ||
1797 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
1798 | "List of charsets ever defined."); | |
2e344af3 KH |
1799 | Vcharset_list = Fcons (Qascii, Fcons (Qeight_bit_control, |
1800 | Fcons (Qeight_bit_graphic, Qnil))); | |
4ed46869 | 1801 | |
537efd8d | 1802 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, |
b4e9dd77 KH |
1803 | "Vector of cons cell of a symbol and translation table ever defined.\n\ |
1804 | An ID of a translation table is an index of this vector."); | |
537efd8d | 1805 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
b0e3cf2b | 1806 | |
4ed46869 KH |
1807 | DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, |
1808 | "Leading-code of private TYPE9N charset of column-width 1."); | |
1809 | leading_code_private_11 = LEADING_CODE_PRIVATE_11; | |
1810 | ||
1811 | DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
1812 | "Leading-code of private TYPE9N charset of column-width 2."); | |
1813 | leading_code_private_12 = LEADING_CODE_PRIVATE_12; | |
1814 | ||
1815 | DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
1816 | "Leading-code of private TYPE9Nx9N charset of column-width 1."); | |
1817 | leading_code_private_21 = LEADING_CODE_PRIVATE_21; | |
1818 | ||
1819 | DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
1820 | "Leading-code of private TYPE9Nx9N charset of column-width 2."); | |
1821 | leading_code_private_22 = LEADING_CODE_PRIVATE_22; | |
35e623fb RS |
1822 | |
1823 | DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset, | |
d2665018 | 1824 | "Offset for converting non-ASCII unibyte codes 0240...0377 to multibyte.\n\ |
4cf9710d RS |
1825 | This is used for converting unibyte text to multibyte,\n\ |
1826 | and for inserting character codes specified by number.\n\n\ | |
3e8ceaac RS |
1827 | This serves to convert a Latin-1 or similar 8-bit character code\n\ |
1828 | to the corresponding Emacs multibyte character code.\n\ | |
1829 | Typically the value should be (- (make-char CHARSET 0) 128),\n\ | |
1830 | for your choice of character set.\n\ | |
537efd8d | 1831 | If `nonascii-translation-table' is non-nil, it overrides this variable."); |
35e623fb | 1832 | nonascii_insert_offset = 0; |
b0e3cf2b | 1833 | |
b4e9dd77 | 1834 | DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table, |
537efd8d | 1835 | "Translation table to convert non-ASCII unibyte codes to multibyte.\n\ |
4cf9710d RS |
1836 | This is used for converting unibyte text to multibyte,\n\ |
1837 | and for inserting character codes specified by number.\n\n\ | |
1838 | Conversion is performed only when multibyte characters are enabled,\n\ | |
1839 | and it serves to convert a Latin-1 or similar 8-bit character code\n\ | |
1840 | to the corresponding Emacs character code.\n\n\ | |
da4d65af | 1841 | If this is nil, `nonascii-insert-offset' is used instead.\n\ |
b4e9dd77 KH |
1842 | See also the docstring of `make-translation-table'."); |
1843 | Vnonascii_translation_table = Qnil; | |
4cf9710d | 1844 | |
c1a08b4c KH |
1845 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, |
1846 | "A char-table for characters which invoke auto-filling.\n\ | |
8dd61baf | 1847 | Such characters have value t in this table."); |
c1a08b4c | 1848 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); |
60022cb7 AS |
1849 | CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt); |
1850 | CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt); | |
4ed46869 KH |
1851 | } |
1852 | ||
1853 | #endif /* emacs */ |