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" | |
37 | #include "coding.h" | |
fc6b09bf | 38 | #include "disptab.h" |
4ed46869 KH |
39 | |
40 | #else /* not emacs */ | |
41 | ||
42 | #include "mulelib.h" | |
43 | ||
44 | #endif /* emacs */ | |
45 | ||
2e344af3 | 46 | Lisp_Object Qcharset, Qascii, Qeight_bit_control, Qeight_bit_graphic; |
0282eb69 | 47 | Lisp_Object Qunknown; |
4ed46869 KH |
48 | |
49 | /* Declaration of special leading-codes. */ | |
4ed46869 KH |
50 | int leading_code_private_11; /* for private DIMENSION1 of 1-column */ |
51 | int leading_code_private_12; /* for private DIMENSION1 of 2-column */ | |
52 | int leading_code_private_21; /* for private DIMENSION2 of 1-column */ | |
53 | int leading_code_private_22; /* for private DIMENSION2 of 2-column */ | |
54 | ||
2e344af3 KH |
55 | /* Declaration of special charsets. The values are set by |
56 | Fsetup_special_charsets. */ | |
4ed46869 KH |
57 | int charset_latin_iso8859_1; /* ISO8859-1 (Latin-1) */ |
58 | int charset_jisx0208_1978; /* JISX0208.1978 (Japanese Kanji old set) */ | |
59 | int charset_jisx0208; /* JISX0208.1983 (Japanese Kanji) */ | |
60 | int charset_katakana_jisx0201; /* JISX0201.Kana (Japanese Katakana) */ | |
61 | int charset_latin_jisx0201; /* JISX0201.Roman (Japanese Roman) */ | |
62 | int charset_big5_1; /* Big5 Level 1 (Chinese Traditional) */ | |
63 | int charset_big5_2; /* Big5 Level 2 (Chinese Traditional) */ | |
64 | ||
65 | Lisp_Object Qcharset_table; | |
66 | ||
67 | /* A char-table containing information of each character set. */ | |
68 | Lisp_Object Vcharset_table; | |
69 | ||
70 | /* A vector of charset symbol indexed by charset-id. This is used | |
71 | only for returning charset symbol from C functions. */ | |
72 | Lisp_Object Vcharset_symbol_table; | |
73 | ||
74 | /* A list of charset symbols ever defined. */ | |
75 | Lisp_Object Vcharset_list; | |
76 | ||
537efd8d KH |
77 | /* Vector of translation table ever defined. |
78 | ID of a translation table is used to index this vector. */ | |
79 | Lisp_Object Vtranslation_table_vector; | |
b0e3cf2b | 80 | |
c1a08b4c KH |
81 | /* A char-table for characters which may invoke auto-filling. */ |
82 | Lisp_Object Vauto_fill_chars; | |
83 | ||
84 | Lisp_Object Qauto_fill_chars; | |
85 | ||
4ed46869 KH |
86 | /* Tables used by macros BYTES_BY_CHAR_HEAD and WIDTH_BY_CHAR_HEAD. */ |
87 | int bytes_by_char_head[256]; | |
88 | int width_by_char_head[256]; | |
89 | ||
90 | /* Mapping table from ISO2022's charset (specified by DIMENSION, | |
91 | CHARS, and FINAL-CHAR) to Emacs' charset. */ | |
92 | int iso_charset_table[2][2][128]; | |
93 | ||
94 | /* Variables used locally in the macro FETCH_MULTIBYTE_CHAR. */ | |
95 | unsigned char *_fetch_multibyte_char_p; | |
96 | int _fetch_multibyte_char_len; | |
97 | ||
35e623fb RS |
98 | /* Offset to add to a non-ASCII value when inserting it. */ |
99 | int nonascii_insert_offset; | |
100 | ||
4cf9710d RS |
101 | /* Translation table for converting non-ASCII unibyte characters |
102 | to multibyte codes, or nil. */ | |
b4e9dd77 | 103 | Lisp_Object Vnonascii_translation_table; |
4cf9710d | 104 | |
8a73a704 KH |
105 | /* List of all possible generic characters. */ |
106 | Lisp_Object Vgeneric_character_list; | |
107 | ||
046b1f03 RS |
108 | #define min(X, Y) ((X) < (Y) ? (X) : (Y)) |
109 | #define max(X, Y) ((X) > (Y) ? (X) : (Y)) | |
110 | \f | |
93bcb785 KH |
111 | void |
112 | invalid_character (c) | |
113 | int c; | |
114 | { | |
ba7434e5 | 115 | error ("Invalid character: 0%o, %d, 0x%x", c, c, c); |
93bcb785 KH |
116 | } |
117 | ||
2e344af3 KH |
118 | /* Parse string STR of length LENGTH and fetch information of a |
119 | character at STR. Set BYTES to the byte length the character | |
120 | occupies, CHARSET, C1, C2 to proper values of the character. */ | |
121 | ||
122 | #define SPLIT_MULTIBYTE_SEQ(str, length, bytes, charset, c1, c2) \ | |
123 | do { \ | |
124 | (c1) = *(str); \ | |
125 | (bytes) = BYTES_BY_CHAR_HEAD (c1); \ | |
126 | if ((bytes) == 1) \ | |
127 | (charset) = ASCII_BYTE_P (c1) ? CHARSET_ASCII : CHARSET_8_BIT_GRAPHIC; \ | |
128 | else if ((bytes) == 2) \ | |
129 | { \ | |
130 | if ((c1) == LEADING_CODE_8_BIT_CONTROL) \ | |
131 | (charset) = CHARSET_8_BIT_CONTROL, (c1) = (str)[1] - 0x20; \ | |
132 | else \ | |
133 | (charset) = (c1), (c1) = (str)[1] & 0x7F; \ | |
134 | } \ | |
135 | else if ((bytes) == 3) \ | |
136 | { \ | |
137 | if ((c1) < LEADING_CODE_PRIVATE_11) \ | |
138 | (charset) = (c1), (c1) = (str)[1] & 0x7F, (c2) = (str)[2] & 0x7F; \ | |
139 | else \ | |
140 | (charset) = (str)[1], (c1) = (str)[2] & 0x7F; \ | |
141 | } \ | |
142 | else \ | |
143 | (charset) = (str)[1], (c1) = (str)[2] & 0x7F, (c2) = (str)[3] & 0x7F; \ | |
ac4137cc KH |
144 | } while (0) |
145 | ||
146 | /* 1 if CHARSET, C1, and C2 compose a valid character, else 0. */ | |
44c6492d | 147 | #define CHAR_COMPONENTS_VALID_P(charset, c1, c2) \ |
2e344af3 | 148 | ((charset) == CHARSET_ASCII \ |
63f4d579 | 149 | ? ((c1) >= 0 && (c1) <= 0x7F) \ |
2e344af3 KH |
150 | : ((charset) == CHARSET_8_BIT_CONTROL \ |
151 | ? ((c1) >= 0x80 && (c1) <= 0x9F) \ | |
152 | : ((charset) == CHARSET_8_BIT_GRAPHIC \ | |
153 | ? ((c1) >= 0x80 && (c1) <= 0xFF) \ | |
154 | : (CHARSET_DIMENSION (charset) == 1 \ | |
155 | ? ((c1) >= 0x20 && (c1) <= 0x7F) \ | |
156 | : ((c1) >= 0x20 && (c1) <= 0x7F \ | |
157 | && (c2) >= 0x20 && (c2) <= 0x7F))))) | |
93bcb785 | 158 | |
99529c2c KH |
159 | /* Store multi-byte form of the character C in STR. The caller should |
160 | allocate at least 4-byte area at STR in advance. Returns the | |
161 | length of the multi-byte form. If C is an invalid character code, | |
162 | signal an error. | |
4ed46869 | 163 | |
99529c2c KH |
164 | Use macro `CHAR_STRING (C, STR)' instead of calling this function |
165 | directly if C can be an ASCII character. */ | |
4ed46869 KH |
166 | |
167 | int | |
99529c2c | 168 | char_to_string (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))) | |
178 | invalid_character (c); | |
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 | ||
210 | /* If C still has any modifier bits, it is an invalid character. */ | |
211 | if (c & CHAR_MODIFIER_MASK) | |
212 | invalid_character (c); | |
2e344af3 KH |
213 | } |
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 | } |
99529c2c | 224 | else if (c < MAX_CHAR) |
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; | |
239 | if (c1 > 0 && c1 < 32 || c2 > 0 && c2 < 32) | |
240 | invalid_character (c); | |
241 | if (c1) | |
4ed46869 | 242 | { |
99529c2c KH |
243 | *p++ = c1 | 0x80; |
244 | if (c2 > 0) | |
245 | *p++ = c2 | 0x80; | |
4ed46869 KH |
246 | } |
247 | } | |
2e344af3 KH |
248 | else |
249 | invalid_character (c); | |
4ed46869 | 250 | |
2e344af3 | 251 | return (p - str); |
4ed46869 KH |
252 | } |
253 | ||
44c6492d KH |
254 | /* Return the non-ASCII character corresponding to multi-byte form at |
255 | STR of length LEN. If ACTUAL_LEN is not NULL, store the byte | |
256 | length of the multibyte form in *ACTUAL_LEN. | |
537efd8d | 257 | |
99529c2c KH |
258 | Use macros STRING_CHAR or STRING_CHAR_AND_LENGTH instead of calling |
259 | this function directly if you want ot handle ASCII characters as | |
260 | well. */ | |
4ed46869 | 261 | |
dfcf069d | 262 | int |
99529c2c | 263 | string_to_char (str, len, actual_len) |
8867de67 | 264 | const unsigned char *str; |
ac4137cc | 265 | int len, *actual_len; |
4ed46869 | 266 | { |
ac4137cc | 267 | int c, bytes, charset, c1, c2; |
4ed46869 | 268 | |
ac4137cc KH |
269 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, charset, c1, c2); |
270 | c = MAKE_CHAR (charset, c1, c2); | |
4ed46869 | 271 | if (actual_len) |
ac4137cc | 272 | *actual_len = bytes; |
4ed46869 KH |
273 | return c; |
274 | } | |
275 | ||
44c6492d KH |
276 | /* Return the length of the multi-byte form at string STR of length LEN. |
277 | Use the macro MULTIBYTE_FORM_LENGTH instead. */ | |
4ed46869 KH |
278 | int |
279 | multibyte_form_length (str, len) | |
8867de67 | 280 | const unsigned char *str; |
4ed46869 KH |
281 | int len; |
282 | { | |
ac4137cc | 283 | int bytes; |
4ed46869 | 284 | |
ac4137cc | 285 | PARSE_MULTIBYTE_SEQ (str, len, bytes); |
90d7b74e | 286 | return bytes; |
4ed46869 KH |
287 | } |
288 | ||
ac4137cc KH |
289 | /* Check multibyte form at string STR of length LEN and set variables |
290 | pointed by CHARSET, C1, and C2 to charset and position codes of the | |
291 | character at STR, and return 0. If there's no multibyte character, | |
4ed46869 KH |
292 | return -1. This should be used only in the macro SPLIT_STRING |
293 | which checks range of STR in advance. */ | |
294 | ||
dfcf069d | 295 | int |
99529c2c | 296 | split_string (str, len, charset, c1, c2) |
ac4137cc KH |
297 | const unsigned char *str; |
298 | unsigned char *c1, *c2; | |
299 | int len, *charset; | |
4ed46869 | 300 | { |
ac4137cc | 301 | register int bytes, cs, code1, code2 = -1; |
4ed46869 | 302 | |
ac4137cc KH |
303 | SPLIT_MULTIBYTE_SEQ (str, len, bytes, cs, code1, code2); |
304 | if (cs == CHARSET_ASCII) | |
4ed46869 | 305 | return -1; |
ac4137cc KH |
306 | *charset = cs; |
307 | *c1 = code1; | |
308 | *c2 = code2; | |
5865af0d | 309 | return 0; |
ac4137cc KH |
310 | } |
311 | ||
44c6492d KH |
312 | /* Return 1 iff character C has valid printable glyph. |
313 | Use the macro CHAR_PRINTABLE_P instead. */ | |
ac4137cc KH |
314 | int |
315 | char_printable_p (c) | |
316 | int c; | |
317 | { | |
318 | int charset, c1, c2, chars; | |
319 | ||
2e344af3 | 320 | if (ASCII_BYTE_P (c)) |
ac4137cc | 321 | return 1; |
2e344af3 KH |
322 | else if (SINGLE_BYTE_CHAR_P (c)) |
323 | return 0; | |
324 | else if (c >= MAX_CHAR) | |
99529c2c | 325 | return 0; |
ac4137cc | 326 | |
2e344af3 | 327 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
328 | if (! CHARSET_DEFINED_P (charset)) |
329 | return 0; | |
330 | if (CHARSET_CHARS (charset) == 94 | |
331 | ? c1 <= 32 || c1 >= 127 | |
332 | : c1 < 32) | |
333 | return 0; | |
334 | if (CHARSET_DIMENSION (charset) == 2 | |
335 | && (CHARSET_CHARS (charset) == 94 | |
336 | ? c2 <= 32 || c2 >= 127 | |
337 | : c2 < 32)) | |
338 | return 0; | |
339 | return 1; | |
4ed46869 KH |
340 | } |
341 | ||
537efd8d | 342 | /* Translate character C by translation table TABLE. If C |
b4e9dd77 KH |
343 | is negative, translate a character specified by CHARSET, C1, and C2 |
344 | (C1 and C2 are code points of the character). If no translation is | |
345 | found in TABLE, return C. */ | |
dfcf069d | 346 | int |
b4e9dd77 | 347 | translate_char (table, c, charset, c1, c2) |
23d2a7f1 KH |
348 | Lisp_Object table; |
349 | int c, charset, c1, c2; | |
350 | { | |
351 | Lisp_Object ch; | |
352 | int alt_charset, alt_c1, alt_c2, dimension; | |
353 | ||
0ad3f83d | 354 | if (c < 0) c = MAKE_CHAR (charset, (c1 & 0x7F) , (c2 & 0x7F)); |
23d2a7f1 | 355 | if (!CHAR_TABLE_P (table) |
ac4137cc | 356 | || (ch = Faref (table, make_number (c)), !NATNUMP (ch))) |
23d2a7f1 KH |
357 | return c; |
358 | ||
359 | SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); | |
360 | dimension = CHARSET_DIMENSION (alt_charset); | |
361 | if (dimension == 1 && alt_c1 > 0 || dimension == 2 && alt_c2 > 0) | |
362 | /* CH is not a generic character, just return it. */ | |
363 | return XFASTINT (ch); | |
364 | ||
365 | /* Since CH is a generic character, we must return a specific | |
366 | charater which has the same position codes as C from CH. */ | |
367 | if (charset < 0) | |
368 | SPLIT_CHAR (c, charset, c1, c2); | |
369 | if (dimension != CHARSET_DIMENSION (charset)) | |
370 | /* We can't make such a character because of dimension mismatch. */ | |
371 | return c; | |
23d2a7f1 KH |
372 | return MAKE_CHAR (alt_charset, c1, c2); |
373 | } | |
374 | ||
d2665018 | 375 | /* Convert the unibyte character C to multibyte based on |
b4e9dd77 | 376 | Vnonascii_translation_table or nonascii_insert_offset. If they can't |
d2665018 KH |
377 | convert C to a valid multibyte character, convert it based on |
378 | DEFAULT_NONASCII_INSERT_OFFSET which makes C a Latin-1 character. */ | |
35e623fb | 379 | |
dfcf069d | 380 | int |
35e623fb RS |
381 | unibyte_char_to_multibyte (c) |
382 | int c; | |
383 | { | |
543b4f61 | 384 | if (c < 0400 && c >= 0200) |
35e623fb | 385 | { |
d2665018 KH |
386 | int c_save = c; |
387 | ||
b4e9dd77 | 388 | if (! NILP (Vnonascii_translation_table)) |
bbf12bb3 KH |
389 | { |
390 | c = XINT (Faref (Vnonascii_translation_table, make_number (c))); | |
ac4137cc | 391 | if (c >= 0400 && ! char_valid_p (c, 0)) |
bbf12bb3 KH |
392 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
393 | } | |
394 | else if (c >= 0240 && nonascii_insert_offset > 0) | |
395 | { | |
396 | c += nonascii_insert_offset; | |
ac4137cc | 397 | if (c < 0400 || ! char_valid_p (c, 0)) |
bbf12bb3 KH |
398 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
399 | } | |
400 | else if (c >= 0240) | |
d2665018 | 401 | c = c_save + DEFAULT_NONASCII_INSERT_OFFSET; |
35e623fb RS |
402 | } |
403 | return c; | |
404 | } | |
76d7b829 KH |
405 | |
406 | ||
407 | /* Convert the multibyte character C to unibyte 8-bit character based | |
408 | on Vnonascii_translation_table or nonascii_insert_offset. If | |
409 | REV_TBL is non-nil, it should be a reverse table of | |
410 | Vnonascii_translation_table, i.e. what given by: | |
411 | Fchar_table_extra_slot (Vnonascii_translation_table, make_number (0)) */ | |
412 | ||
413 | int | |
414 | multibyte_char_to_unibyte (c, rev_tbl) | |
415 | int c; | |
416 | Lisp_Object rev_tbl; | |
417 | { | |
418 | if (!SINGLE_BYTE_CHAR_P (c)) | |
419 | { | |
420 | int c_save = c; | |
421 | ||
422 | if (! CHAR_TABLE_P (rev_tbl) | |
423 | && CHAR_TABLE_P (Vnonascii_translation_table)) | |
424 | rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table, | |
425 | make_number (0)); | |
426 | if (CHAR_TABLE_P (rev_tbl)) | |
427 | { | |
428 | Lisp_Object temp; | |
429 | temp = Faref (rev_tbl, make_number (c)); | |
430 | if (INTEGERP (temp)) | |
431 | c = XINT (temp); | |
bbf12bb3 KH |
432 | if (c >= 256) |
433 | c = (c_save & 0177) + 0200; | |
434 | } | |
435 | else | |
436 | { | |
437 | if (nonascii_insert_offset > 0) | |
438 | c -= nonascii_insert_offset; | |
439 | if (c < 128 || c >= 256) | |
440 | c = (c_save & 0177) + 0200; | |
76d7b829 | 441 | } |
76d7b829 KH |
442 | } |
443 | ||
444 | return c; | |
445 | } | |
446 | ||
35e623fb | 447 | \f |
4ed46869 KH |
448 | /* Update the table Vcharset_table with the given arguments (see the |
449 | document of `define-charset' for the meaning of each argument). | |
450 | Several other table contents are also updated. The caller should | |
451 | check the validity of CHARSET-ID and the remaining arguments in | |
452 | advance. */ | |
453 | ||
454 | void | |
455 | update_charset_table (charset_id, dimension, chars, width, direction, | |
456 | iso_final_char, iso_graphic_plane, | |
457 | short_name, long_name, description) | |
458 | Lisp_Object charset_id, dimension, chars, width, direction; | |
459 | Lisp_Object iso_final_char, iso_graphic_plane; | |
460 | Lisp_Object short_name, long_name, description; | |
461 | { | |
462 | int charset = XINT (charset_id); | |
463 | int bytes; | |
464 | unsigned char leading_code_base, leading_code_ext; | |
465 | ||
6dc0722d KH |
466 | if (NILP (CHARSET_TABLE_ENTRY (charset))) |
467 | CHARSET_TABLE_ENTRY (charset) | |
468 | = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); | |
4ed46869 | 469 | |
d78bc582 KH |
470 | if (NILP (long_name)) |
471 | long_name = short_name; | |
472 | if (NILP (description)) | |
473 | description = long_name; | |
474 | ||
4ed46869 KH |
475 | /* Get byte length of multibyte form, base leading-code, and |
476 | extended leading-code of the charset. See the comment under the | |
477 | title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
478 | bytes = XINT (dimension); | |
479 | if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
480 | { | |
481 | /* Official charset, it doesn't have an extended leading-code. */ | |
2e344af3 | 482 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC) |
4ed46869 KH |
483 | bytes += 1; /* For a base leading-code. */ |
484 | leading_code_base = charset; | |
485 | leading_code_ext = 0; | |
486 | } | |
487 | else | |
488 | { | |
489 | /* Private charset. */ | |
490 | bytes += 2; /* For base and extended leading-codes. */ | |
491 | leading_code_base | |
492 | = (charset < LEADING_CODE_EXT_12 | |
493 | ? LEADING_CODE_PRIVATE_11 | |
494 | : (charset < LEADING_CODE_EXT_21 | |
495 | ? LEADING_CODE_PRIVATE_12 | |
496 | : (charset < LEADING_CODE_EXT_22 | |
497 | ? LEADING_CODE_PRIVATE_21 | |
498 | : LEADING_CODE_PRIVATE_22))); | |
499 | leading_code_ext = charset; | |
500 | } | |
501 | ||
2e344af3 KH |
502 | if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC |
503 | &&BYTES_BY_CHAR_HEAD (leading_code_base) != bytes) | |
6ef23ebb KH |
504 | error ("Invalid dimension for the charset-ID %d", charset); |
505 | ||
4ed46869 KH |
506 | CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; |
507 | CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
508 | CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
509 | CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
510 | CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
511 | CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
512 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
513 | = make_number (leading_code_base); | |
514 | CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
515 | = make_number (leading_code_ext); | |
516 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
517 | CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
518 | = iso_graphic_plane; | |
519 | CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
520 | CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
521 | CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
522 | CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
523 | ||
524 | { | |
525 | /* If we have already defined a charset which has the same | |
526 | DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
527 | DIRECTION, we must update the entry REVERSE-CHARSET of both | |
528 | charsets. If there's no such charset, the value of the entry | |
529 | is set to nil. */ | |
530 | int i; | |
531 | ||
513ee442 | 532 | for (i = 0; i <= MAX_CHARSET; i++) |
4ed46869 KH |
533 | if (!NILP (CHARSET_TABLE_ENTRY (i))) |
534 | { | |
535 | if (CHARSET_DIMENSION (i) == XINT (dimension) | |
536 | && CHARSET_CHARS (i) == XINT (chars) | |
537 | && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
538 | && CHARSET_DIRECTION (i) != XINT (direction)) | |
539 | { | |
540 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
541 | = make_number (i); | |
542 | CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
543 | break; | |
544 | } | |
545 | } | |
513ee442 | 546 | if (i > MAX_CHARSET) |
4ed46869 KH |
547 | /* No such a charset. */ |
548 | CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
549 | = make_number (-1); | |
550 | } | |
551 | ||
552 | if (charset != CHARSET_ASCII | |
553 | && charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
554 | { | |
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. */ | |
2e344af3 KH |
566 | if (iso_final_char >= 0 |
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 | { | |
600 | if (width == 1) | |
601 | from = LEADING_CODE_EXT_11, to = LEADING_CODE_EXT_12; | |
602 | else | |
603 | from = LEADING_CODE_EXT_12, to = LEADING_CODE_EXT_21; | |
604 | } | |
605 | else | |
606 | { | |
607 | if (width == 1) | |
608 | from = LEADING_CODE_EXT_21, to = LEADING_CODE_EXT_22; | |
609 | else | |
b0e3cf2b | 610 | from = LEADING_CODE_EXT_22, to = LEADING_CODE_EXT_MAX + 1; |
4ed46869 KH |
611 | } |
612 | ||
613 | for (charset = from; charset < to; charset++) | |
614 | if (!CHARSET_DEFINED_P (charset)) break; | |
615 | ||
616 | return make_number (charset < to ? charset : 0); | |
617 | } | |
618 | ||
619 | DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
620 | "Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR.\n\ | |
23d2a7f1 | 621 | If CHARSET-ID is nil, it is decided automatically, which means CHARSET is\n\ |
4ed46869 KH |
622 | treated as a private charset.\n\ |
623 | INFO-VECTOR is a vector of the format:\n\ | |
624 | [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE\n\ | |
625 | SHORT-NAME LONG-NAME DESCRIPTION]\n\ | |
626 | The meanings of each elements is as follows:\n\ | |
627 | DIMENSION (integer) is the number of bytes to represent a character: 1 or 2.\n\ | |
628 | CHARS (integer) is the number of characters in a dimension: 94 or 96.\n\ | |
629 | WIDTH (integer) is the number of columns a character in the charset\n\ | |
630 | occupies on the screen: one of 0, 1, and 2.\n\ | |
631 | \n\ | |
632 | DIRECTION (integer) is the rendering direction of characters in the\n\ | |
277576f6 KH |
633 | charset when rendering. If 0, render from left to right, else\n\ |
634 | render from right to left.\n\ | |
4ed46869 KH |
635 | \n\ |
636 | ISO-FINAL-CHAR (character) is the final character of the\n\ | |
637 | corresponding ISO 2022 charset.\n\ | |
2e344af3 | 638 | It may be -1 if the charset is internal use only.\n\ |
4ed46869 KH |
639 | \n\ |
640 | ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked\n\ | |
641 | while encoding to variants of ISO 2022 coding system, one of the\n\ | |
642 | following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).\n\ | |
2e344af3 | 643 | It may be -1 if the charset is internal use only.\n\ |
4ed46869 KH |
644 | \n\ |
645 | SHORT-NAME (string) is the short name to refer to the charset.\n\ | |
646 | \n\ | |
647 | LONG-NAME (string) is the long name to refer to the charset.\n\ | |
648 | \n\ | |
649 | DESCRIPTION (string) is the description string of the charset.") | |
650 | (charset_id, charset_symbol, info_vector) | |
651 | Lisp_Object charset_id, charset_symbol, info_vector; | |
652 | { | |
653 | Lisp_Object *vec; | |
654 | ||
655 | if (!NILP (charset_id)) | |
656 | CHECK_NUMBER (charset_id, 0); | |
657 | CHECK_SYMBOL (charset_symbol, 1); | |
658 | CHECK_VECTOR (info_vector, 2); | |
659 | ||
660 | if (! NILP (charset_id)) | |
661 | { | |
662 | if (! CHARSET_VALID_P (XINT (charset_id))) | |
663 | error ("Invalid CHARSET: %d", XINT (charset_id)); | |
664 | else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
665 | error ("Already defined charset: %d", XINT (charset_id)); | |
666 | } | |
667 | ||
668 | vec = XVECTOR (info_vector)->contents; | |
669 | if (XVECTOR (info_vector)->size != 9 | |
670 | || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
671 | || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
672 | || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
673 | || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
2e344af3 KH |
674 | || !INTEGERP (vec[4]) |
675 | || !(XINT (vec[4]) == -1 || XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~') | |
676 | || !INTEGERP (vec[5]) | |
677 | || !(XINT (vec[5]) == -1 || XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
4ed46869 KH |
678 | || !STRINGP (vec[6]) |
679 | || !STRINGP (vec[7]) | |
680 | || !STRINGP (vec[8])) | |
681 | error ("Invalid info-vector argument for defining charset %s", | |
682 | XSYMBOL (charset_symbol)->name->data); | |
683 | ||
684 | if (NILP (charset_id)) | |
685 | { | |
686 | charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
687 | if (XINT (charset_id) == 0) | |
688 | error ("There's no room for a new private charset %s", | |
689 | XSYMBOL (charset_symbol)->name->data); | |
690 | } | |
691 | ||
692 | update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
693 | vec[4], vec[5], vec[6], vec[7], vec[8]); | |
6dc0722d | 694 | Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
4ed46869 KH |
695 | CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
696 | Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
697 | return Qnil; | |
698 | } | |
699 | ||
8a73a704 KH |
700 | DEFUN ("generic-character-list", Fgeneric_character_list, |
701 | Sgeneric_character_list, 0, 0, 0, | |
702 | "Return a list of all possible generic characters.\n\ | |
703 | It includes a generic character for a charset not yet defined.") | |
704 | () | |
705 | { | |
706 | return Vgeneric_character_list; | |
707 | } | |
708 | ||
3fac5a51 KH |
709 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
710 | Sget_unused_iso_final_char, 2, 2, 0, | |
711 | "Return an unsed ISO's final char for a charset of DIMENISION and CHARS.\n\ | |
712 | DIMENSION is the number of bytes to represent a character: 1 or 2.\n\ | |
713 | CHARS is the number of characters in a dimension: 94 or 96.\n\ | |
714 | \n\ | |
715 | This final char is for private use, thus the range is `0' (48) .. `?' (63).\n\ | |
716 | If there's no unused final char for the specified kind of charset,\n\ | |
717 | return nil.") | |
718 | (dimension, chars) | |
719 | Lisp_Object dimension, chars; | |
720 | { | |
721 | int final_char; | |
722 | ||
723 | CHECK_NUMBER (dimension, 0); | |
724 | CHECK_NUMBER (chars, 1); | |
725 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
726 | error ("Invalid charset dimension %d, it should be 1 or 2", | |
727 | XINT (dimension)); | |
728 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
729 | error ("Invalid charset chars %d, it should be 94 or 96", | |
730 | XINT (chars)); | |
731 | for (final_char = '0'; final_char <= '?'; final_char++) | |
732 | { | |
733 | if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0) | |
734 | break; | |
735 | } | |
736 | return (final_char <= '?' ? make_number (final_char) : Qnil); | |
737 | } | |
738 | ||
4ed46869 KH |
739 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
740 | 4, 4, 0, | |
741 | "Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET.\n\ | |
742 | CHARSET should be defined by `defined-charset' in advance.") | |
743 | (dimension, chars, final_char, charset_symbol) | |
744 | Lisp_Object dimension, chars, final_char, charset_symbol; | |
745 | { | |
746 | int charset; | |
747 | ||
748 | CHECK_NUMBER (dimension, 0); | |
749 | CHECK_NUMBER (chars, 1); | |
750 | CHECK_NUMBER (final_char, 2); | |
751 | CHECK_SYMBOL (charset_symbol, 3); | |
752 | ||
753 | if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
754 | error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
755 | if (XINT (chars) != 94 && XINT (chars) != 96) | |
756 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
757 | if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
758 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
759 | if ((charset = get_charset_id (charset_symbol)) < 0) | |
760 | error ("Invalid charset %s", XSYMBOL (charset_symbol)->name->data); | |
761 | ||
762 | ISO_CHARSET_TABLE (dimension, chars, final_char) = charset; | |
763 | return Qnil; | |
764 | } | |
765 | ||
2e344af3 KH |
766 | /* Return information about charsets in the text at PTR of NBYTES |
767 | bytes, which are NCHARS characters. The value is: | |
f6302ac9 | 768 | |
cfe34140 | 769 | 0: Each character is represented by one byte. This is always |
f6302ac9 | 770 | true for unibyte text. |
2e344af3 KH |
771 | 1: No charsets other than ascii eight-bit-control, |
772 | eight-bit-graphic, and latin-1 are found. | |
773 | 2: Otherwise. | |
1d67c29b | 774 | |
2e344af3 KH |
775 | In addition, if CHARSETS is nonzero, for each found charset N, set |
776 | CHARSETS[N] to 1. For that, callers should allocate CHARSETS | |
777 | (MAX_CHARSET + 1 elements) in advance. It may lookup a translation | |
778 | table TABLE if supplied. For invalid charsets, set CHARSETS[1] to | |
779 | 1 (note that there's no charset whose ID is 1). */ | |
4ed46869 KH |
780 | |
781 | int | |
2e344af3 KH |
782 | find_charset_in_text (ptr, nchars, nbytes, charsets, table) |
783 | unsigned char *ptr; | |
784 | int nchars, nbytes, *charsets; | |
23d2a7f1 | 785 | Lisp_Object table; |
4ed46869 | 786 | { |
2e344af3 | 787 | if (nchars == nbytes) |
0282eb69 | 788 | { |
2e344af3 | 789 | if (charsets && nbytes > 0) |
0282eb69 | 790 | { |
2e344af3 KH |
791 | unsigned char *endp = ptr + nbytes; |
792 | int maskbits = 0; | |
793 | ||
794 | while (ptr < endp && maskbits != 7) | |
795 | { | |
796 | maskbits |= (*ptr < 0x80 ? 1 : *ptr < 0xA0 ? 2 : 4); | |
797 | ptr++; | |
798 | } | |
799 | ||
800 | if (maskbits & 1) | |
801 | charsets[CHARSET_ASCII] = 1; | |
802 | if (maskbits & 2) | |
803 | charsets[CHARSET_8_BIT_CONTROL] = 1; | |
804 | if (maskbits & 4) | |
805 | charsets[CHARSET_8_BIT_GRAPHIC] = 1; | |
0282eb69 | 806 | } |
2e344af3 | 807 | return 0; |
0282eb69 | 808 | } |
2e344af3 | 809 | else |
4ed46869 | 810 | { |
2e344af3 | 811 | int return_val = 1; |
99529c2c | 812 | int bytes, charset, c1, c2; |
05505664 | 813 | |
2e344af3 KH |
814 | if (! CHAR_TABLE_P (table)) |
815 | table = Qnil; | |
05505664 | 816 | |
2e344af3 | 817 | while (nchars-- > 0) |
23d2a7f1 | 818 | { |
2e344af3 KH |
819 | SPLIT_MULTIBYTE_SEQ (ptr, len, bytes, charset, c1, c2); |
820 | ptr += bytes; | |
4ed46869 | 821 | |
2e344af3 KH |
822 | if (!CHARSET_DEFINED_P (charset)) |
823 | charset = 1; | |
824 | else if (! NILP (table)) | |
825 | { | |
826 | int c = translate_char (table, -1, charset, c1, c2); | |
827 | if (c >= 0) | |
828 | charset = CHAR_CHARSET (c); | |
829 | } | |
830 | ||
831 | if (return_val == 1 | |
832 | && charset != CHARSET_ASCII | |
833 | && charset != CHARSET_8_BIT_CONTROL | |
834 | && charset != CHARSET_8_BIT_GRAPHIC | |
835 | && charset != charset_latin_iso8859_1) | |
836 | return_val = 2; | |
837 | ||
838 | if (charsets) | |
839 | charsets[charset] = 1; | |
840 | else if (return_val == 2) | |
841 | break; | |
4ed46869 | 842 | } |
2e344af3 | 843 | return return_val; |
4ed46869 | 844 | } |
4ed46869 KH |
845 | } |
846 | ||
847 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 848 | 2, 3, 0, |
4ed46869 | 849 | "Return a list of charsets in the region between BEG and END.\n\ |
23d2a7f1 | 850 | BEG and END are buffer positions.\n\ |
0282eb69 KH |
851 | Optional arg TABLE if non-nil is a translation table to look up.\n\ |
852 | \n\ | |
853 | If the region contains invalid multiybte characters,\n\ | |
38f02ede | 854 | `unknown' is included in the returned list.\n\ |
0282eb69 | 855 | \n\ |
2e344af3 KH |
856 | If the current buffer is unibyte, the returned list may contain\n\ |
857 | only `ascii', `eight-bit-control', and `eight-bit-graphic'.") | |
23d2a7f1 KH |
858 | (beg, end, table) |
859 | Lisp_Object beg, end, table; | |
4ed46869 | 860 | { |
028d516b | 861 | int charsets[MAX_CHARSET + 1]; |
6ae1f27e | 862 | int from, from_byte, to, stop, stop_byte, i; |
4ed46869 KH |
863 | Lisp_Object val; |
864 | ||
865 | validate_region (&beg, &end); | |
866 | from = XFASTINT (beg); | |
867 | stop = to = XFASTINT (end); | |
6ae1f27e | 868 | |
4ed46869 | 869 | if (from < GPT && GPT < to) |
6ae1f27e RS |
870 | { |
871 | stop = GPT; | |
872 | stop_byte = GPT_BYTE; | |
873 | } | |
874 | else | |
875 | stop_byte = CHAR_TO_BYTE (stop); | |
876 | ||
877 | from_byte = CHAR_TO_BYTE (from); | |
878 | ||
028d516b | 879 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
4ed46869 KH |
880 | while (1) |
881 | { | |
2e344af3 KH |
882 | find_charset_in_text (BYTE_POS_ADDR (from_byte), stop - from, |
883 | stop_byte - from_byte, charsets, table); | |
4ed46869 | 884 | if (stop < to) |
6ae1f27e RS |
885 | { |
886 | from = stop, from_byte = stop_byte; | |
887 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
888 | } | |
4ed46869 KH |
889 | else |
890 | break; | |
891 | } | |
6ae1f27e | 892 | |
4ed46869 | 893 | val = Qnil; |
2e344af3 | 894 | if (charsets[1]) |
0282eb69 | 895 | val = Fcons (Qunknown, val); |
2e344af3 KH |
896 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
897 | if (charsets[i]) | |
898 | val = Fcons (CHARSET_SYMBOL (i), val); | |
899 | if (charsets[0]) | |
900 | val = Fcons (Qascii, val); | |
4ed46869 KH |
901 | return val; |
902 | } | |
903 | ||
904 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 KH |
905 | 1, 2, 0, |
906 | "Return a list of charsets in STR.\n\ | |
0282eb69 KH |
907 | Optional arg TABLE if non-nil is a translation table to look up.\n\ |
908 | \n\ | |
909 | If the region contains invalid multiybte characters,\n\ | |
910 | `unknown' is included in the returned list.\n\ | |
911 | \n\ | |
2e344af3 KH |
912 | If STR is unibyte, the returned list may contain\n\ |
913 | only `ascii', `eight-bit-control', and `eight-bit-graphic'.") | |
23d2a7f1 KH |
914 | (str, table) |
915 | Lisp_Object str, table; | |
4ed46869 | 916 | { |
a29e3b1b | 917 | int charsets[MAX_CHARSET + 1]; |
4ed46869 KH |
918 | int i; |
919 | Lisp_Object val; | |
920 | ||
921 | CHECK_STRING (str, 0); | |
87b089ad | 922 | |
a29e3b1b | 923 | bzero (charsets, (MAX_CHARSET + 1) * sizeof (int)); |
2e344af3 KH |
924 | find_charset_in_text (XSTRING (str)->data, XSTRING (str)->size, |
925 | STRING_BYTES (XSTRING (str)), charsets, table); | |
926 | ||
4ed46869 | 927 | val = Qnil; |
2e344af3 | 928 | if (charsets[1]) |
0282eb69 | 929 | val = Fcons (Qunknown, val); |
2e344af3 KH |
930 | for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--) |
931 | if (charsets[i]) | |
932 | val = Fcons (CHARSET_SYMBOL (i), val); | |
933 | if (charsets[0]) | |
934 | val = Fcons (Qascii, val); | |
4ed46869 KH |
935 | return val; |
936 | } | |
2e344af3 | 937 | |
4ed46869 KH |
938 | \f |
939 | DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
513ee442 | 940 | "") |
4ed46869 KH |
941 | (charset, code1, code2) |
942 | Lisp_Object charset, code1, code2; | |
943 | { | |
ac4137cc KH |
944 | int charset_id, c1, c2; |
945 | ||
4ed46869 | 946 | CHECK_NUMBER (charset, 0); |
ac4137cc KH |
947 | charset_id = XINT (charset); |
948 | if (!CHARSET_DEFINED_P (charset_id)) | |
949 | error ("Invalid charset ID: %d", XINT (charset)); | |
4ed46869 KH |
950 | |
951 | if (NILP (code1)) | |
ac4137cc | 952 | c1 = 0; |
4ed46869 | 953 | else |
ac4137cc KH |
954 | { |
955 | CHECK_NUMBER (code1, 1); | |
956 | c1 = XINT (code1); | |
957 | } | |
4ed46869 | 958 | if (NILP (code2)) |
ac4137cc | 959 | c2 = 0; |
4ed46869 | 960 | else |
ac4137cc KH |
961 | { |
962 | CHECK_NUMBER (code2, 2); | |
963 | c2 = XINT (code2); | |
964 | } | |
4ed46869 | 965 | |
2e344af3 KH |
966 | if (charset_id == CHARSET_ASCII) |
967 | { | |
968 | if (c1 < 0 || c1 > 0x7F) | |
969 | goto invalid_code_posints; | |
970 | return make_number (c1); | |
971 | } | |
972 | else if (charset_id == CHARSET_8_BIT_CONTROL) | |
973 | { | |
974 | if (c1 < 0x80 || c1 > 0x9F) | |
975 | goto invalid_code_posints; | |
976 | return make_number (c1); | |
977 | } | |
978 | else if (charset_id == CHARSET_8_BIT_GRAPHIC) | |
979 | { | |
980 | if (c1 < 0xA0 || c1 > 0xFF) | |
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 | { | |
1007 | Lisp_Object val; | |
0282eb69 | 1008 | int c, charset, c1, c2; |
4ed46869 KH |
1009 | |
1010 | CHECK_NUMBER (ch, 0); | |
0282eb69 KH |
1011 | c = XFASTINT (ch); |
1012 | if (!CHAR_VALID_P (c, 1)) | |
1013 | return Fcons (Qunknown, Fcons (ch, Qnil)); | |
4ed46869 | 1014 | SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); |
6dc0722d | 1015 | return (c2 >= 0 |
4ed46869 KH |
1016 | ? Fcons (CHARSET_SYMBOL (charset), |
1017 | Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
1018 | : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
1019 | } | |
1020 | ||
1021 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
1022 | "Return charset of CHAR.") | |
1023 | (ch) | |
1024 | Lisp_Object ch; | |
1025 | { | |
1026 | CHECK_NUMBER (ch, 0); | |
1027 | ||
1028 | return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
1029 | } | |
1030 | ||
90d7b74e | 1031 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
ac4137cc | 1032 | "Return charset of a character in the current buffer at position POS.\n\ |
e6e114f2 KH |
1033 | If POS is nil, it defauls to the current point.\n\ |
1034 | If POS is out of range, the value is nil.") | |
90d7b74e KH |
1035 | (pos) |
1036 | Lisp_Object pos; | |
1037 | { | |
2e344af3 KH |
1038 | Lisp_Object ch; |
1039 | int charset; | |
ac4137cc | 1040 | |
2e344af3 KH |
1041 | ch = Fchar_after (pos); |
1042 | if (! INTEGERP (ch)) | |
1043 | return ch; | |
1044 | charset = CHAR_CHARSET (XINT (ch)); | |
90d7b74e KH |
1045 | return CHARSET_SYMBOL (charset); |
1046 | } | |
1047 | ||
4ed46869 | 1048 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
2b71bb78 KH |
1049 | "Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR.\n\ |
1050 | \n\ | |
1051 | ISO 2022's designation sequence (escape sequence) distinguishes charsets\n\ | |
1052 | by their DIMENSION, CHARS, and FINAL-CHAR,\n\ | |
1053 | where as Emacs distinguishes them by charset symbol.\n\ | |
1054 | See the documentation of the function `charset-info' for the meanings of\n\ | |
1055 | DIMENSION, CHARS, and FINAL-CHAR.") | |
4ed46869 KH |
1056 | (dimension, chars, final_char) |
1057 | Lisp_Object dimension, chars, final_char; | |
1058 | { | |
1059 | int charset; | |
1060 | ||
1061 | CHECK_NUMBER (dimension, 0); | |
1062 | CHECK_NUMBER (chars, 1); | |
1063 | CHECK_NUMBER (final_char, 2); | |
1064 | ||
1065 | if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
1066 | return Qnil; | |
1067 | return CHARSET_SYMBOL (charset); | |
1068 | } | |
1069 | ||
9d3d8cba KH |
1070 | /* If GENERICP is nonzero, return nonzero iff C is a valid normal or |
1071 | generic character. If GENERICP is zero, return nonzero iff C is a | |
1072 | valid normal character. Do not call this function directly, | |
1073 | instead use macro CHAR_VALID_P. */ | |
1074 | int | |
1075 | char_valid_p (c, genericp) | |
1076 | int c, genericp; | |
1077 | { | |
1078 | int charset, c1, c2; | |
1079 | ||
1080 | if (c < 0) | |
1081 | return 0; | |
1082 | if (SINGLE_BYTE_CHAR_P (c)) | |
1083 | return 1; | |
2e344af3 | 1084 | SPLIT_CHAR (c, charset, c1, c2); |
ac4137cc KH |
1085 | if (genericp) |
1086 | { | |
1087 | if (c1) | |
1088 | { | |
1089 | if (c2 <= 0) c2 = 0x20; | |
1090 | } | |
1091 | else | |
1092 | { | |
1093 | if (c2 <= 0) c1 = c2 = 0x20; | |
1094 | } | |
1095 | } | |
1096 | return (CHARSET_DEFINED_P (charset) | |
44c6492d | 1097 | && CHAR_COMPONENTS_VALID_P (charset, c1, c2)); |
9d3d8cba KH |
1098 | } |
1099 | ||
1100 | DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0, | |
a9d02884 DL |
1101 | "Return t if OBJECT is a valid normal character.\n\ |
1102 | If optional arg GENERICP is non-nil, also return t if OBJECT is\n\ | |
9d3d8cba KH |
1103 | a valid generic character.") |
1104 | (object, genericp) | |
1105 | Lisp_Object object, genericp; | |
1106 | { | |
1107 | if (! NATNUMP (object)) | |
1108 | return Qnil; | |
1109 | return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil); | |
1110 | } | |
1111 | ||
d2665018 KH |
1112 | DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte, |
1113 | Sunibyte_char_to_multibyte, 1, 1, 0, | |
1114 | "Convert the unibyte character CH to multibyte character.\n\ | |
537efd8d | 1115 | The conversion is done based on `nonascii-translation-table' (which see)\n\ |
340b8d58 | 1116 | or `nonascii-insert-offset' (which see).") |
d2665018 KH |
1117 | (ch) |
1118 | Lisp_Object ch; | |
1119 | { | |
1120 | int c; | |
1121 | ||
1122 | CHECK_NUMBER (ch, 0); | |
1123 | c = XINT (ch); | |
1124 | if (c < 0 || c >= 0400) | |
1125 | error ("Invalid unibyte character: %d", c); | |
1126 | c = unibyte_char_to_multibyte (c); | |
1127 | if (c < 0) | |
1128 | error ("Can't convert to multibyte character: %d", XINT (ch)); | |
1129 | return make_number (c); | |
1130 | } | |
1131 | ||
1bcc1567 RS |
1132 | DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte, |
1133 | Smultibyte_char_to_unibyte, 1, 1, 0, | |
1134 | "Convert the multibyte character CH to unibyte character.\n\ | |
1135 | The conversion is done based on `nonascii-translation-table' (which see)\n\ | |
1136 | or `nonascii-insert-offset' (which see).") | |
1137 | (ch) | |
1138 | Lisp_Object ch; | |
1139 | { | |
1140 | int c; | |
1141 | ||
1142 | CHECK_NUMBER (ch, 0); | |
1143 | c = XINT (ch); | |
ac4137cc | 1144 | if (! CHAR_VALID_P (c, 0)) |
1bcc1567 RS |
1145 | error ("Invalid multibyte character: %d", c); |
1146 | c = multibyte_char_to_unibyte (c, Qnil); | |
1147 | if (c < 0) | |
1148 | error ("Can't convert to unibyte character: %d", XINT (ch)); | |
1149 | return make_number (c); | |
1150 | } | |
1151 | ||
4ed46869 | 1152 | DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, |
f78643ef | 1153 | "Return 1 regardless of the argument CHAR.\n\ |
60022cb7 | 1154 | This is now an obsolete function. We keep it just for backward compatibility.") |
4ed46869 KH |
1155 | (ch) |
1156 | Lisp_Object ch; | |
1157 | { | |
1158 | Lisp_Object val; | |
4ed46869 KH |
1159 | |
1160 | CHECK_NUMBER (ch, 0); | |
9b6a601f KH |
1161 | return make_number (1); |
1162 | } | |
1163 | ||
1164 | /* Return how many bytes C will occupy in a multibyte buffer. | |
1165 | Don't call this function directly, instead use macro CHAR_BYTES. */ | |
1166 | int | |
1167 | char_bytes (c) | |
1168 | int c; | |
1169 | { | |
99529c2c | 1170 | int charset; |
9b6a601f | 1171 | |
2e344af3 KH |
1172 | if (ASCII_BYTE_P (c) || (c & ~((1 << CHARACTERBITS) -1))) |
1173 | return 1; | |
1174 | if (SINGLE_BYTE_CHAR_P (c) && c >= 0xA0) | |
8ac5a9cc KH |
1175 | return 1; |
1176 | ||
99529c2c KH |
1177 | charset = CHAR_CHARSET (c); |
1178 | return (CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1); | |
4ed46869 KH |
1179 | } |
1180 | ||
1181 | /* Return the width of character of which multi-byte form starts with | |
1182 | C. The width is measured by how many columns occupied on the | |
1183 | screen when displayed in the current buffer. */ | |
1184 | ||
1185 | #define ONE_BYTE_CHAR_WIDTH(c) \ | |
1186 | (c < 0x20 \ | |
1187 | ? (c == '\t' \ | |
53316e55 | 1188 | ? XFASTINT (current_buffer->tab_width) \ |
4ed46869 KH |
1189 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
1190 | : (c < 0x7f \ | |
1191 | ? 1 \ | |
1192 | : (c == 0x7F \ | |
1193 | ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
1194 | : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
1195 | && BASE_LEADING_CODE_P (c)) \ | |
1196 | ? WIDTH_BY_CHAR_HEAD (c) \ | |
b4e9dd77 | 1197 | : 4)))) |
4ed46869 KH |
1198 | |
1199 | DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
1200 | "Return width of CHAR when displayed in the current buffer.\n\ | |
1201 | The width is measured by how many columns it occupies on the screen.") | |
1202 | (ch) | |
1203 | Lisp_Object ch; | |
1204 | { | |
859f2b3c | 1205 | Lisp_Object val, disp; |
4ed46869 | 1206 | int c; |
51c4025f | 1207 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 KH |
1208 | |
1209 | CHECK_NUMBER (ch, 0); | |
1210 | ||
859f2b3c RS |
1211 | c = XINT (ch); |
1212 | ||
1213 | /* Get the way the display table would display it. */ | |
51c4025f | 1214 | disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil; |
859f2b3c RS |
1215 | |
1216 | if (VECTORP (disp)) | |
1217 | XSETINT (val, XVECTOR (disp)->size); | |
1218 | else if (SINGLE_BYTE_CHAR_P (c)) | |
1219 | XSETINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
4ed46869 KH |
1220 | else |
1221 | { | |
1222 | int charset = CHAR_CHARSET (c); | |
1223 | ||
1224 | XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
1225 | } | |
1226 | return val; | |
1227 | } | |
1228 | ||
1229 | /* Return width of string STR of length LEN when displayed in the | |
1230 | current buffer. The width is measured by how many columns it | |
1231 | occupies on the screen. */ | |
859f2b3c | 1232 | |
4ed46869 KH |
1233 | int |
1234 | strwidth (str, len) | |
1235 | unsigned char *str; | |
1236 | int len; | |
1237 | { | |
1238 | unsigned char *endp = str + len; | |
1239 | int width = 0; | |
c4a4e28f | 1240 | struct Lisp_Char_Table *dp = buffer_display_table (); |
4ed46869 | 1241 | |
859f2b3c RS |
1242 | while (str < endp) |
1243 | { | |
99529c2c KH |
1244 | Lisp_Object disp; |
1245 | int thislen; | |
1246 | int c = STRING_CHAR_AND_LENGTH (str, endp - str, thislen); | |
859f2b3c | 1247 | |
99529c2c KH |
1248 | /* Get the way the display table would display it. */ |
1249 | if (dp) | |
1250 | disp = DISP_CHAR_VECTOR (dp, c); | |
859f2b3c | 1251 | else |
99529c2c KH |
1252 | disp = Qnil; |
1253 | ||
1254 | if (VECTORP (disp)) | |
1255 | width += XVECTOR (disp)->size; | |
1256 | else | |
1257 | width += ONE_BYTE_CHAR_WIDTH (*str); | |
1258 | ||
1259 | str += thislen; | |
859f2b3c | 1260 | } |
4ed46869 KH |
1261 | return width; |
1262 | } | |
1263 | ||
1264 | DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
1265 | "Return width of STRING when displayed in the current buffer.\n\ | |
1266 | Width is measured by how many columns it occupies on the screen.\n\ | |
046b1f03 RS |
1267 | When calculating width of a multibyte character in STRING,\n\ |
1268 | only the base leading-code is considered; the validity of\n\ | |
1269 | the following bytes is not checked.") | |
4ed46869 KH |
1270 | (str) |
1271 | Lisp_Object str; | |
1272 | { | |
1273 | Lisp_Object val; | |
1274 | ||
1275 | CHECK_STRING (str, 0); | |
fc932ac6 RS |
1276 | XSETFASTINT (val, strwidth (XSTRING (str)->data, |
1277 | STRING_BYTES (XSTRING (str)))); | |
4ed46869 KH |
1278 | return val; |
1279 | } | |
1280 | ||
1281 | DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
1282 | "Return the direction of CHAR.\n\ | |
1283 | The returned value is 0 for left-to-right and 1 for right-to-left.") | |
1284 | (ch) | |
1285 | Lisp_Object ch; | |
1286 | { | |
1287 | int charset; | |
1288 | ||
1289 | CHECK_NUMBER (ch, 0); | |
1290 | charset = CHAR_CHARSET (XFASTINT (ch)); | |
1291 | if (!CHARSET_DEFINED_P (charset)) | |
93bcb785 | 1292 | invalid_character (XINT (ch)); |
4ed46869 KH |
1293 | return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); |
1294 | } | |
1295 | ||
af4fecb4 | 1296 | DEFUN ("chars-in-region", Fchars_in_region, Schars_in_region, 2, 2, 0, |
6ae1f27e | 1297 | "Return number of characters between BEG and END.") |
046b1f03 RS |
1298 | (beg, end) |
1299 | Lisp_Object beg, end; | |
1300 | { | |
6ae1f27e | 1301 | int from, to; |
046b1f03 | 1302 | |
17e7ef1b RS |
1303 | CHECK_NUMBER_COERCE_MARKER (beg, 0); |
1304 | CHECK_NUMBER_COERCE_MARKER (end, 1); | |
1305 | ||
046b1f03 | 1306 | from = min (XFASTINT (beg), XFASTINT (end)); |
a8a35e61 | 1307 | to = max (XFASTINT (beg), XFASTINT (end)); |
046b1f03 | 1308 | |
a8c21066 | 1309 | return make_number (to - from); |
6ae1f27e | 1310 | } |
9036eb45 | 1311 | |
87b089ad RS |
1312 | /* Return the number of characters in the NBYTES bytes at PTR. |
1313 | This works by looking at the contents and checking for multibyte sequences. | |
1314 | However, if the current buffer has enable-multibyte-characters = nil, | |
1315 | we treat each byte as a character. */ | |
1316 | ||
6ae1f27e RS |
1317 | int |
1318 | chars_in_text (ptr, nbytes) | |
1319 | unsigned char *ptr; | |
1320 | int nbytes; | |
1321 | { | |
87b089ad RS |
1322 | /* current_buffer is null at early stages of Emacs initialization. */ |
1323 | if (current_buffer == 0 | |
1324 | || NILP (current_buffer->enable_multibyte_characters)) | |
6ae1f27e | 1325 | return nbytes; |
a8a35e61 | 1326 | |
ac4137cc | 1327 | return multibyte_chars_in_text (ptr, nbytes); |
046b1f03 RS |
1328 | } |
1329 | ||
87b089ad RS |
1330 | /* Return the number of characters in the NBYTES bytes at PTR. |
1331 | This works by looking at the contents and checking for multibyte sequences. | |
1332 | It ignores enable-multibyte-characters. */ | |
1333 | ||
1334 | int | |
1335 | multibyte_chars_in_text (ptr, nbytes) | |
1336 | unsigned char *ptr; | |
1337 | int nbytes; | |
1338 | { | |
ac4137cc KH |
1339 | unsigned char *endp; |
1340 | int chars, bytes; | |
87b089ad RS |
1341 | |
1342 | endp = ptr + nbytes; | |
1343 | chars = 0; | |
1344 | ||
1345 | while (ptr < endp) | |
1346 | { | |
2e344af3 KH |
1347 | PARSE_MULTIBYTE_SEQ (ptr, endp - ptr, bytes); |
1348 | ptr += bytes; | |
1349 | chars++; | |
1350 | } | |
1351 | ||
1352 | return chars; | |
1353 | } | |
1354 | ||
1355 | /* Parse unibyte text at STR of LEN bytes as a multibyte text, and | |
1356 | count the numbers of characters and bytes in it. On counting | |
1357 | bytes, pay attention to that 8-bit characters in the range | |
1358 | 0x80..0x9F are represented by 2-byte in a multibyte text. */ | |
1359 | void | |
1360 | parse_str_as_multibyte (str, len, nchars, nbytes) | |
1361 | unsigned char *str; | |
1362 | int len, *nchars, *nbytes; | |
1363 | { | |
1364 | unsigned char *endp = str + len; | |
1365 | int n, chars = 0, bytes = 0; | |
1366 | ||
1367 | while (str < endp) | |
1368 | { | |
1369 | if (UNIBYTE_STR_AS_MULTIBYTE_P (str, endp - str, n)) | |
1370 | str += n, bytes += n; | |
1371 | else | |
1372 | str++, bytes += 2; | |
1373 | chars++; | |
1374 | } | |
1375 | *nchars = chars; | |
1376 | *nbytes = bytes; | |
1377 | return; | |
1378 | } | |
1379 | ||
1380 | /* Arrange unibyte text at STR of NBYTES bytes as a multibyte text. | |
1381 | It actually converts only 8-bit characters in the range 0x80..0x9F | |
1382 | that don't contruct multibyte characters to multibyte forms. If | |
1383 | NCHARS is nonzero, set *NCHARS to the number of characters in the | |
1384 | text. It is assured that we can use LEN bytes at STR as a work | |
1385 | area and that is enough. Return the number of bytes of the | |
1386 | resulting text. */ | |
1387 | ||
1388 | int | |
1389 | str_as_multibyte (str, len, nbytes, nchars) | |
1390 | unsigned char *str; | |
1391 | int len, nbytes, *nchars; | |
1392 | { | |
1393 | unsigned char *p = str, *endp = str + nbytes; | |
1394 | unsigned char *to; | |
1395 | int chars = 0; | |
1396 | int n; | |
1397 | ||
1398 | while (p < endp && UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
1399 | p += n, chars++; | |
1400 | if (nchars) | |
1401 | *nchars = chars; | |
1402 | if (p == endp) | |
1403 | return nbytes; | |
1404 | ||
1405 | to = p; | |
1406 | nbytes = endp - p; | |
1407 | endp = str + len; | |
1408 | safe_bcopy (p, endp - nbytes, nbytes); | |
1409 | p = endp - nbytes; | |
1410 | while (p < endp) | |
1411 | { | |
1412 | if (UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n)) | |
ac4137cc | 1413 | { |
2e344af3 KH |
1414 | while (n--) |
1415 | *to++ = *p++; | |
1416 | } | |
ac4137cc KH |
1417 | else |
1418 | { | |
2e344af3 KH |
1419 | *to++ = LEADING_CODE_8_BIT_CONTROL; |
1420 | *to++ = *p++ + 0x20; | |
ac4137cc | 1421 | } |
87b089ad RS |
1422 | chars++; |
1423 | } | |
2e344af3 KH |
1424 | if (nchars) |
1425 | *nchars = chars; | |
1426 | return (to - str); | |
1427 | } | |
87b089ad | 1428 | |
2e344af3 KH |
1429 | /* Convert unibyte text at STR of NBYTES bytes to a multibyte text |
1430 | that contains the same single-byte characters. It actually | |
1431 | converts all 8-bit characters to multibyte forms. It is assured | |
1432 | that we can use LEN bytes at STR as a work area and that is | |
1433 | enough. */ | |
1434 | ||
1435 | int | |
1436 | str_to_multibyte (str, len, bytes) | |
1437 | unsigned char *str; | |
1438 | int len, bytes; | |
1439 | { | |
1440 | unsigned char *p = str, *endp = str + bytes; | |
1441 | unsigned char *to; | |
1442 | int c; | |
1443 | ||
1444 | while (p < endp && (*p < 0x80 || *p >= 0xA0)) p++; | |
1445 | if (p == endp) | |
1446 | return bytes; | |
1447 | to = p; | |
1448 | bytes = endp - p; | |
1449 | endp = str + len; | |
1450 | safe_bcopy (p, endp - bytes, bytes); | |
1451 | p = endp - bytes; | |
1452 | while (p < endp) | |
1453 | { | |
1454 | if (*p < 0x80 || *p >= 0xA0) | |
1455 | *to++ = *p++; | |
1456 | else | |
1457 | *to++ = LEADING_CODE_8_BIT_CONTROL, *to++ = *p++ + 0x20; | |
1458 | } | |
1459 | return (to - str); | |
87b089ad RS |
1460 | } |
1461 | ||
2e344af3 KH |
1462 | /* Arrange multibyte text at STR of LEN bytes as a unibyte text. It |
1463 | actually converts only 8-bit characters in the range 0x80..0x9F to | |
1464 | unibyte forms. */ | |
1465 | ||
1466 | int | |
1467 | str_as_unibyte (str, bytes) | |
1468 | unsigned char *str; | |
1469 | int bytes; | |
1470 | { | |
1471 | unsigned char *p = str, *endp = str + bytes; | |
1472 | unsigned char *to = str; | |
1473 | ||
1474 | while (p < endp && *p != LEADING_CODE_8_BIT_CONTROL) p++; | |
1475 | to = p; | |
1476 | while (p < endp) | |
1477 | { | |
1478 | if (*p == LEADING_CODE_8_BIT_CONTROL) | |
1479 | *to++ = *(p + 1) - 0x20, p += 2; | |
1480 | else | |
1481 | *to++ = *p++; | |
1482 | } | |
1483 | return (to - str); | |
1484 | } | |
1485 | ||
1486 | \f | |
87b089ad | 1487 | DEFUN ("string", Fstring, Sstring, 1, MANY, 0, |
4ed46869 | 1488 | "Concatenate all the argument characters and make the result a string.") |
53316e55 KH |
1489 | (n, args) |
1490 | int n; | |
4ed46869 KH |
1491 | Lisp_Object *args; |
1492 | { | |
53316e55 | 1493 | int i; |
99529c2c | 1494 | unsigned char *buf = (unsigned char *) alloca (MAX_MULTIBYTE_LENGTH * n); |
4ed46869 | 1495 | unsigned char *p = buf; |
2e344af3 | 1496 | int c; |
4ed46869 KH |
1497 | |
1498 | for (i = 0; i < n; i++) | |
1499 | { | |
d0c037d8 | 1500 | CHECK_NUMBER (args[i], 0); |
4ed46869 | 1501 | c = XINT (args[i]); |
99529c2c | 1502 | p += CHAR_STRING (c, p); |
4ed46869 KH |
1503 | } |
1504 | ||
2e344af3 | 1505 | return make_string_from_bytes (buf, n, p - buf); |
4ed46869 KH |
1506 | } |
1507 | ||
1508 | #endif /* emacs */ | |
1509 | \f | |
dfcf069d | 1510 | int |
4ed46869 KH |
1511 | charset_id_internal (charset_name) |
1512 | char *charset_name; | |
1513 | { | |
76d7b829 | 1514 | Lisp_Object val; |
4ed46869 | 1515 | |
76d7b829 | 1516 | val= Fget (intern (charset_name), Qcharset); |
4ed46869 KH |
1517 | if (!VECTORP (val)) |
1518 | error ("Charset %s is not defined", charset_name); | |
1519 | ||
1520 | return (XINT (XVECTOR (val)->contents[0])); | |
1521 | } | |
1522 | ||
1523 | DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
1524 | Ssetup_special_charsets, 0, 0, 0, "Internal use only.") | |
1525 | () | |
1526 | { | |
1527 | charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1528 | charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1529 | charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1530 | charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1531 | charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1532 | charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1533 | charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1534 | return Qnil; | |
1535 | } | |
1536 | ||
dfcf069d | 1537 | void |
4ed46869 KH |
1538 | init_charset_once () |
1539 | { | |
1540 | int i, j, k; | |
1541 | ||
1542 | staticpro (&Vcharset_table); | |
1543 | staticpro (&Vcharset_symbol_table); | |
8a73a704 | 1544 | staticpro (&Vgeneric_character_list); |
4ed46869 KH |
1545 | |
1546 | /* This has to be done here, before we call Fmake_char_table. */ | |
1547 | Qcharset_table = intern ("charset-table"); | |
1548 | staticpro (&Qcharset_table); | |
1549 | ||
1550 | /* Intern this now in case it isn't already done. | |
1551 | Setting this variable twice is harmless. | |
1552 | But don't staticpro it here--that is done in alloc.c. */ | |
1553 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1554 | ||
1555 | /* Now we are ready to set up this property, so we can | |
1556 | create the charset table. */ | |
1557 | Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1558 | Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1559 | ||
0282eb69 KH |
1560 | Qunknown = intern ("unknown"); |
1561 | staticpro (&Qunknown); | |
1562 | Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), | |
1563 | Qunknown); | |
4ed46869 KH |
1564 | |
1565 | /* Setup tables. */ | |
1566 | for (i = 0; i < 2; i++) | |
1567 | for (j = 0; j < 2; j++) | |
1568 | for (k = 0; k < 128; k++) | |
1569 | iso_charset_table [i][j][k] = -1; | |
1570 | ||
60383934 | 1571 | for (i = 0; i < 256; i++) |
2e344af3 | 1572 | bytes_by_char_head[i] = 1; |
a5236564 | 1573 | for (i = 128; i < MIN_CHARSET_OFFICIAL_DIMENSION2; i++) |
2e344af3 | 1574 | bytes_by_char_head[i] = 2; |
a5236564 | 1575 | for (; i <= MAX_CHARSET_OFFICIAL_DIMENSION2; i++) |
2e344af3 | 1576 | bytes_by_char_head[i] = 3; |
a5236564 KH |
1577 | for (; i < 160; i++) |
1578 | bytes_by_char_head[i] = 2; | |
2e344af3 KH |
1579 | bytes_by_char_head[LEADING_CODE_PRIVATE_11] = 3; |
1580 | bytes_by_char_head[LEADING_CODE_PRIVATE_12] = 3; | |
1581 | bytes_by_char_head[LEADING_CODE_PRIVATE_21] = 4; | |
1582 | bytes_by_char_head[LEADING_CODE_PRIVATE_22] = 4; | |
4ed46869 KH |
1583 | |
1584 | for (i = 0; i < 128; i++) | |
2e344af3 | 1585 | width_by_char_head[i] = 1; |
4ed46869 | 1586 | for (; i < 256; i++) |
2e344af3 KH |
1587 | width_by_char_head[i] = 4; |
1588 | width_by_char_head[LEADING_CODE_PRIVATE_11] = 1; | |
1589 | width_by_char_head[LEADING_CODE_PRIVATE_12] = 2; | |
1590 | width_by_char_head[LEADING_CODE_PRIVATE_21] = 1; | |
1591 | width_by_char_head[LEADING_CODE_PRIVATE_22] = 2; | |
8a73a704 KH |
1592 | |
1593 | { | |
76d7b829 | 1594 | Lisp_Object val; |
8a73a704 | 1595 | |
76d7b829 | 1596 | val = Qnil; |
8a73a704 KH |
1597 | for (i = 0x81; i < 0x90; i++) |
1598 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1599 | for (; i < 0x9A; i++) | |
1600 | val = Fcons (make_number ((i - 0x8F) << 14), val); | |
1601 | for (i = 0xA0; i < 0xF0; i++) | |
1602 | val = Fcons (make_number ((i - 0x70) << 7), val); | |
1603 | for (; i < 0xFF; i++) | |
1604 | val = Fcons (make_number ((i - 0xE0) << 14), val); | |
8a73a704 KH |
1605 | Vgeneric_character_list = Fnreverse (val); |
1606 | } | |
bbf12bb3 KH |
1607 | |
1608 | nonascii_insert_offset = 0; | |
1609 | Vnonascii_translation_table = Qnil; | |
4ed46869 KH |
1610 | } |
1611 | ||
1612 | #ifdef emacs | |
1613 | ||
dfcf069d | 1614 | void |
4ed46869 KH |
1615 | syms_of_charset () |
1616 | { | |
2e344af3 KH |
1617 | Qcharset = intern ("charset"); |
1618 | staticpro (&Qcharset); | |
1619 | ||
4ed46869 KH |
1620 | Qascii = intern ("ascii"); |
1621 | staticpro (&Qascii); | |
1622 | ||
2e344af3 KH |
1623 | Qeight_bit_control = intern ("eight-bit-control"); |
1624 | staticpro (&Qeight_bit_control); | |
1625 | ||
1626 | Qeight_bit_graphic = intern ("eight-bit-graphic"); | |
1627 | staticpro (&Qeight_bit_graphic); | |
4ed46869 | 1628 | |
2e344af3 KH |
1629 | /* Define special charsets ascii, eight-bit-control, and |
1630 | eight-bit-graphic. */ | |
4ed46869 KH |
1631 | update_charset_table (make_number (CHARSET_ASCII), |
1632 | make_number (1), make_number (94), | |
1633 | make_number (1), | |
1634 | make_number (0), | |
1635 | make_number ('B'), | |
1636 | make_number (0), | |
1637 | build_string ("ASCII"), | |
d78bc582 | 1638 | Qnil, /* same as above */ |
4ed46869 KH |
1639 | build_string ("ASCII (ISO646 IRV)")); |
1640 | CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1641 | Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1642 | ||
2e344af3 KH |
1643 | update_charset_table (make_number (CHARSET_8_BIT_CONTROL), |
1644 | make_number (1), make_number (96), | |
cd550f91 | 1645 | make_number (1), |
2e344af3 KH |
1646 | make_number (0), |
1647 | make_number (-1), | |
1648 | make_number (-1), | |
1649 | build_string ("8-bit control code (0x80..0x9F)"), | |
d78bc582 KH |
1650 | Qnil, /* same as above */ |
1651 | Qnil); /* same as above */ | |
2e344af3 KH |
1652 | CHARSET_SYMBOL (CHARSET_8_BIT_CONTROL) = Qeight_bit_control; |
1653 | Fput (Qeight_bit_control, Qcharset, | |
1654 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_CONTROL)); | |
1655 | ||
1656 | update_charset_table (make_number (CHARSET_8_BIT_GRAPHIC), | |
1657 | make_number (1), make_number (96), | |
cd550f91 | 1658 | make_number (1), |
2e344af3 KH |
1659 | make_number (0), |
1660 | make_number (-1), | |
1661 | make_number (-1), | |
2e344af3 | 1662 | build_string ("8-bit graphic char (0xA0..0xFF)"), |
d78bc582 KH |
1663 | Qnil, /* same as above */ |
1664 | Qnil); /* same as above */ | |
2e344af3 KH |
1665 | CHARSET_SYMBOL (CHARSET_8_BIT_GRAPHIC) = Qeight_bit_graphic; |
1666 | Fput (Qeight_bit_graphic, Qcharset, | |
1667 | CHARSET_TABLE_ENTRY (CHARSET_8_BIT_GRAPHIC)); | |
1668 | ||
c1a08b4c KH |
1669 | Qauto_fill_chars = intern ("auto-fill-chars"); |
1670 | staticpro (&Qauto_fill_chars); | |
1671 | Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0)); | |
1672 | ||
4ed46869 | 1673 | defsubr (&Sdefine_charset); |
8a73a704 | 1674 | defsubr (&Sgeneric_character_list); |
3fac5a51 | 1675 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
1676 | defsubr (&Sdeclare_equiv_charset); |
1677 | defsubr (&Sfind_charset_region); | |
1678 | defsubr (&Sfind_charset_string); | |
1679 | defsubr (&Smake_char_internal); | |
1680 | defsubr (&Ssplit_char); | |
1681 | defsubr (&Schar_charset); | |
90d7b74e | 1682 | defsubr (&Scharset_after); |
4ed46869 | 1683 | defsubr (&Siso_charset); |
9d3d8cba | 1684 | defsubr (&Schar_valid_p); |
d2665018 | 1685 | defsubr (&Sunibyte_char_to_multibyte); |
1bcc1567 | 1686 | defsubr (&Smultibyte_char_to_unibyte); |
4ed46869 KH |
1687 | defsubr (&Schar_bytes); |
1688 | defsubr (&Schar_width); | |
1689 | defsubr (&Sstring_width); | |
1690 | defsubr (&Schar_direction); | |
af4fecb4 | 1691 | defsubr (&Schars_in_region); |
87b089ad | 1692 | defsubr (&Sstring); |
4ed46869 KH |
1693 | defsubr (&Ssetup_special_charsets); |
1694 | ||
1695 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
1696 | "List of charsets ever defined."); | |
2e344af3 KH |
1697 | Vcharset_list = Fcons (Qascii, Fcons (Qeight_bit_control, |
1698 | Fcons (Qeight_bit_graphic, Qnil))); | |
4ed46869 | 1699 | |
537efd8d | 1700 | DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector, |
b4e9dd77 KH |
1701 | "Vector of cons cell of a symbol and translation table ever defined.\n\ |
1702 | An ID of a translation table is an index of this vector."); | |
537efd8d | 1703 | Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil); |
b0e3cf2b | 1704 | |
4ed46869 KH |
1705 | DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, |
1706 | "Leading-code of private TYPE9N charset of column-width 1."); | |
1707 | leading_code_private_11 = LEADING_CODE_PRIVATE_11; | |
1708 | ||
1709 | DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
1710 | "Leading-code of private TYPE9N charset of column-width 2."); | |
1711 | leading_code_private_12 = LEADING_CODE_PRIVATE_12; | |
1712 | ||
1713 | DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
1714 | "Leading-code of private TYPE9Nx9N charset of column-width 1."); | |
1715 | leading_code_private_21 = LEADING_CODE_PRIVATE_21; | |
1716 | ||
1717 | DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
1718 | "Leading-code of private TYPE9Nx9N charset of column-width 2."); | |
1719 | leading_code_private_22 = LEADING_CODE_PRIVATE_22; | |
35e623fb RS |
1720 | |
1721 | DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset, | |
d2665018 | 1722 | "Offset for converting non-ASCII unibyte codes 0240...0377 to multibyte.\n\ |
4cf9710d RS |
1723 | This is used for converting unibyte text to multibyte,\n\ |
1724 | and for inserting character codes specified by number.\n\n\ | |
3e8ceaac RS |
1725 | This serves to convert a Latin-1 or similar 8-bit character code\n\ |
1726 | to the corresponding Emacs multibyte character code.\n\ | |
1727 | Typically the value should be (- (make-char CHARSET 0) 128),\n\ | |
1728 | for your choice of character set.\n\ | |
537efd8d | 1729 | If `nonascii-translation-table' is non-nil, it overrides this variable."); |
35e623fb | 1730 | nonascii_insert_offset = 0; |
b0e3cf2b | 1731 | |
b4e9dd77 | 1732 | DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table, |
537efd8d | 1733 | "Translation table to convert non-ASCII unibyte codes to multibyte.\n\ |
4cf9710d RS |
1734 | This is used for converting unibyte text to multibyte,\n\ |
1735 | and for inserting character codes specified by number.\n\n\ | |
1736 | Conversion is performed only when multibyte characters are enabled,\n\ | |
1737 | and it serves to convert a Latin-1 or similar 8-bit character code\n\ | |
1738 | to the corresponding Emacs character code.\n\n\ | |
da4d65af | 1739 | If this is nil, `nonascii-insert-offset' is used instead.\n\ |
b4e9dd77 KH |
1740 | See also the docstring of `make-translation-table'."); |
1741 | Vnonascii_translation_table = Qnil; | |
4cf9710d | 1742 | |
c1a08b4c KH |
1743 | DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars, |
1744 | "A char-table for characters which invoke auto-filling.\n\ | |
8dd61baf | 1745 | Such characters have value t in this table."); |
c1a08b4c | 1746 | Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil); |
60022cb7 AS |
1747 | CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt); |
1748 | CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt); | |
4ed46869 KH |
1749 | } |
1750 | ||
1751 | #endif /* emacs */ |