Commit | Line | Data |
---|---|---|
3263d5a2 | 1 | /* Basic character set support. |
aaef169d | 2 | Copyright (C) 2001, 2002, 2003, 2004, 2005, |
8cabe764 | 3 | 2006, 2007, 2008 Free Software Foundation, Inc. |
7976eda0 | 4 | Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, |
8cabe764 | 5 | 2005, 2006, 2007, 2008 |
ce03bf76 KH |
6 | National Institute of Advanced Industrial Science and Technology (AIST) |
7 | Registration Number H14PRO021 | |
4ed46869 | 8 | |
327719ee | 9 | Copyright (C) 2003, 2004 |
3263d5a2 KH |
10 | National Institute of Advanced Industrial Science and Technology (AIST) |
11 | Registration Number H13PRO009 | |
4ed46869 | 12 | |
369314dc KH |
13 | This file is part of GNU Emacs. |
14 | ||
9ec0b715 | 15 | GNU Emacs is free software: you can redistribute it and/or modify |
369314dc | 16 | it under the terms of the GNU General Public License as published by |
9ec0b715 GM |
17 | the Free Software Foundation, either version 3 of the License, or |
18 | (at your option) any later version. | |
4ed46869 | 19 | |
369314dc KH |
20 | GNU Emacs is distributed in the hope that it will be useful, |
21 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
22 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
23 | GNU General Public License for more details. | |
4ed46869 | 24 | |
369314dc | 25 | You should have received a copy of the GNU General Public License |
9ec0b715 | 26 | along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */ |
4ed46869 | 27 | |
68c45bf0 | 28 | #include <config.h> |
68c45bf0 | 29 | |
4ed46869 | 30 | #include <stdio.h> |
3263d5a2 KH |
31 | #include <unistd.h> |
32 | #include <ctype.h> | |
4ed46869 | 33 | #include <sys/types.h> |
4ed46869 | 34 | #include "lisp.h" |
3263d5a2 | 35 | #include "character.h" |
4ed46869 KH |
36 | #include "charset.h" |
37 | #include "coding.h" | |
fc6b09bf | 38 | #include "disptab.h" |
3263d5a2 | 39 | #include "buffer.h" |
4ed46869 | 40 | |
04c2f2c5 | 41 | /*** GENERAL NOTES on CODED CHARACTER SETS (CHARSETS) *** |
4ed46869 | 42 | |
3263d5a2 | 43 | A coded character set ("charset" hereafter) is a meaningful |
04c2f2c5 | 44 | collection (i.e. language, culture, functionality, etc.) of |
3263d5a2 | 45 | characters. Emacs handles multiple charsets at once. In Emacs Lisp |
04c2f2c5 DL |
46 | code, a charset is represented by a symbol. In C code, a charset is |
47 | represented by its ID number or by a pointer to a struct charset. | |
4ed46869 | 48 | |
3263d5a2 KH |
49 | The actual information about each charset is stored in two places. |
50 | Lispy information is stored in the hash table Vcharset_hash_table as | |
51 | a vector (charset attributes). The other information is stored in | |
04c2f2c5 | 52 | charset_table as a struct charset. |
4ed46869 | 53 | |
3263d5a2 | 54 | */ |
4ed46869 | 55 | |
3263d5a2 KH |
56 | /* List of all charsets. This variable is used only from Emacs |
57 | Lisp. */ | |
4ed46869 | 58 | Lisp_Object Vcharset_list; |
4ed46869 | 59 | |
3263d5a2 KH |
60 | /* Hash table that contains attributes of each charset. Keys are |
61 | charset symbols, and values are vectors of charset attributes. */ | |
62 | Lisp_Object Vcharset_hash_table; | |
4ed46869 | 63 | |
3263d5a2 KH |
64 | /* Table of struct charset. */ |
65 | struct charset *charset_table; | |
4ed46869 | 66 | |
3263d5a2 | 67 | static int charset_table_size; |
5af5dd92 | 68 | static int charset_table_used; |
4ed46869 | 69 | |
3263d5a2 | 70 | Lisp_Object Qcharsetp; |
4ed46869 | 71 | |
3263d5a2 KH |
72 | /* Special charset symbols. */ |
73 | Lisp_Object Qascii; | |
2fe1edd1 | 74 | Lisp_Object Qeight_bit; |
3263d5a2 KH |
75 | Lisp_Object Qiso_8859_1; |
76 | Lisp_Object Qunicode; | |
6c652beb | 77 | Lisp_Object Qemacs; |
4ed46869 | 78 | |
3263d5a2 KH |
79 | /* The corresponding charsets. */ |
80 | int charset_ascii; | |
2fe1edd1 | 81 | int charset_eight_bit; |
3263d5a2 KH |
82 | int charset_iso_8859_1; |
83 | int charset_unicode; | |
6c652beb | 84 | int charset_emacs; |
b0e3cf2b | 85 | |
7c7dceee KH |
86 | /* The other special charsets. */ |
87 | int charset_jisx0201_roman; | |
88 | int charset_jisx0208_1978; | |
89 | int charset_jisx0208; | |
c1a08b4c | 90 | |
3263d5a2 KH |
91 | /* Value of charset attribute `charset-iso-plane'. */ |
92 | Lisp_Object Qgl, Qgr; | |
c1a08b4c | 93 | |
d1a04588 KH |
94 | /* Charset of unibyte characters. */ |
95 | int charset_unibyte; | |
4ed46869 | 96 | |
3263d5a2 KH |
97 | /* List of charsets ordered by the priority. */ |
98 | Lisp_Object Vcharset_ordered_list; | |
4ed46869 | 99 | |
6a9c90ec KH |
100 | /* Sub-list of Vcharset_ordered_list that contains all non-preferred |
101 | charsets. */ | |
102 | Lisp_Object Vcharset_non_preferred_head; | |
103 | ||
dbbb237d | 104 | /* Incremented everytime we change Vcharset_ordered_list. This is |
64165ae2 | 105 | unsigned short so that it fits in Lisp_Int and never matches |
dbbb237d KH |
106 | -1. */ |
107 | unsigned short charset_ordered_list_tick; | |
4ed46869 | 108 | |
3263d5a2 KH |
109 | /* List of iso-2022 charsets. */ |
110 | Lisp_Object Viso_2022_charset_list; | |
35e623fb | 111 | |
3263d5a2 KH |
112 | /* List of emacs-mule charsets. */ |
113 | Lisp_Object Vemacs_mule_charset_list; | |
114 | ||
115 | struct charset *emacs_mule_charset[256]; | |
4ed46869 KH |
116 | |
117 | /* Mapping table from ISO2022's charset (specified by DIMENSION, | |
118 | CHARS, and FINAL-CHAR) to Emacs' charset. */ | |
3263d5a2 KH |
119 | int iso_charset_table[ISO_MAX_DIMENSION][ISO_MAX_CHARS][ISO_MAX_FINAL]; |
120 | ||
4beef065 | 121 | Lisp_Object Vcharset_map_path; |
3263d5a2 KH |
122 | |
123 | Lisp_Object Vchar_unified_charset_table; | |
124 | ||
6a9c90ec KH |
125 | Lisp_Object Vcurrent_iso639_language; |
126 | ||
64165ae2 DL |
127 | /* Defined in chartab.c */ |
128 | extern void | |
129 | map_char_table_for_charset P_ ((void (*c_function) (Lisp_Object, Lisp_Object), | |
130 | Lisp_Object function, Lisp_Object table, | |
131 | Lisp_Object arg, struct charset *charset, | |
132 | unsigned from, unsigned to)); | |
133 | ||
69f8de5b KH |
134 | #define CODE_POINT_TO_INDEX(charset, code) \ |
135 | ((charset)->code_linear_p \ | |
136 | ? (code) - (charset)->min_code \ | |
137 | : (((charset)->code_space_mask[(code) >> 24] & 0x8) \ | |
138 | && ((charset)->code_space_mask[((code) >> 16) & 0xFF] & 0x4) \ | |
139 | && ((charset)->code_space_mask[((code) >> 8) & 0xFF] & 0x2) \ | |
140 | && ((charset)->code_space_mask[(code) & 0xFF] & 0x1)) \ | |
141 | ? (((((code) >> 24) - (charset)->code_space[12]) \ | |
142 | * (charset)->code_space[11]) \ | |
143 | + (((((code) >> 16) & 0xFF) - (charset)->code_space[8]) \ | |
144 | * (charset)->code_space[7]) \ | |
145 | + (((((code) >> 8) & 0xFF) - (charset)->code_space[4]) \ | |
146 | * (charset)->code_space[3]) \ | |
820ee249 KH |
147 | + (((code) & 0xFF) - (charset)->code_space[0]) \ |
148 | - ((charset)->char_index_offset)) \ | |
3263d5a2 KH |
149 | : -1) |
150 | ||
151 | ||
152 | /* Convert the character index IDX to code-point CODE for CHARSET. | |
153 | It is assumed that IDX is in a valid range. */ | |
154 | ||
820ee249 KH |
155 | #define INDEX_TO_CODE_POINT(charset, idx) \ |
156 | ((charset)->code_linear_p \ | |
157 | ? (idx) + (charset)->min_code \ | |
158 | : (idx += (charset)->char_index_offset, \ | |
159 | (((charset)->code_space[0] + (idx) % (charset)->code_space[2]) \ | |
160 | | (((charset)->code_space[4] \ | |
161 | + ((idx) / (charset)->code_space[3] % (charset)->code_space[6])) \ | |
162 | << 8) \ | |
163 | | (((charset)->code_space[8] \ | |
164 | + ((idx) / (charset)->code_space[7] % (charset)->code_space[10])) \ | |
165 | << 16) \ | |
166 | | (((charset)->code_space[12] + ((idx) / (charset)->code_space[11])) \ | |
167 | << 24)))) | |
4cf9710d | 168 | |
8a73a704 | 169 | |
046b1f03 | 170 | \f |
93bcb785 | 171 | |
e9ce014c KH |
172 | /* Set to 1 to warn that a charset map is loaded and thus a buffer |
173 | text and a string data may be relocated. */ | |
3263d5a2 | 174 | int charset_map_loaded; |
4ed46869 | 175 | |
e9ce014c | 176 | struct charset_map_entries |
4ed46869 | 177 | { |
e9ce014c KH |
178 | struct { |
179 | unsigned from, to; | |
180 | int c; | |
181 | } entry[0x10000]; | |
182 | struct charset_map_entries *next; | |
183 | }; | |
184 | ||
185 | /* Load the mapping information for CHARSET from ENTRIES. | |
4cf9710d | 186 | |
3263d5a2 | 187 | If CONTROL_FLAG is 0, setup CHARSET->min_char and CHARSET->max_char. |
8a73a704 | 188 | |
3263d5a2 KH |
189 | If CONTROL_FLAG is 1, setup CHARSET->min_char, CHARSET->max_char, |
190 | CHARSET->decoder, and CHARSET->encoder. | |
93bcb785 | 191 | |
3263d5a2 KH |
192 | If CONTROL_FLAG is 2, setup CHARSET->deunifier and |
193 | Vchar_unify_table. If Vchar_unified_charset_table is non-nil, | |
194 | setup it too. */ | |
4ed46869 | 195 | |
3263d5a2 | 196 | static void |
e9ce014c | 197 | load_charset_map (charset, entries, n_entries, control_flag) |
3263d5a2 | 198 | struct charset *charset; |
e9ce014c KH |
199 | struct charset_map_entries *entries; |
200 | int n_entries; | |
3263d5a2 | 201 | int control_flag; |
4ed46869 | 202 | { |
3263d5a2 | 203 | Lisp_Object vec, table; |
3263d5a2 KH |
204 | unsigned max_code = CHARSET_MAX_CODE (charset); |
205 | int ascii_compatible_p = charset->ascii_compatible_p; | |
206 | int min_char, max_char, nonascii_min_char; | |
3263d5a2 | 207 | int i; |
3263d5a2 | 208 | unsigned char *fast_map = charset->fast_map; |
99529c2c | 209 | |
e9ce014c KH |
210 | if (n_entries <= 0) |
211 | return; | |
212 | ||
213 | if (control_flag > 0) | |
8ac5a9cc | 214 | { |
3263d5a2 | 215 | int n = CODE_POINT_TO_INDEX (charset, max_code) + 1; |
6662e69b | 216 | |
374c5cfd | 217 | table = Fmake_char_table (Qnil, Qnil); |
3263d5a2 KH |
218 | if (control_flag == 1) |
219 | vec = Fmake_vector (make_number (n), make_number (-1)); | |
220 | else if (! CHAR_TABLE_P (Vchar_unify_table)) | |
374c5cfd | 221 | Vchar_unify_table = Fmake_char_table (Qnil, Qnil); |
6662e69b | 222 | |
3263d5a2 | 223 | charset_map_loaded = 1; |
2e344af3 | 224 | } |
6662e69b | 225 | |
e9ce014c | 226 | min_char = max_char = entries->entry[0].c; |
3263d5a2 | 227 | nonascii_min_char = MAX_CHAR; |
e9ce014c | 228 | for (i = 0; i < n_entries; i++) |
2e344af3 | 229 | { |
e9ce014c | 230 | unsigned from, to; |
3b4f4446 KH |
231 | int from_index, to_index; |
232 | int from_c, to_c; | |
e9ce014c | 233 | int idx = i % 0x10000; |
3263d5a2 | 234 | |
e9ce014c KH |
235 | if (i > 0 && idx == 0) |
236 | entries = entries->next; | |
237 | from = entries->entry[idx].from; | |
238 | to = entries->entry[idx].to; | |
3b4f4446 KH |
239 | from_c = entries->entry[idx].c; |
240 | from_index = CODE_POINT_TO_INDEX (charset, from); | |
241 | if (from == to) | |
6662e69b | 242 | { |
3b4f4446 KH |
243 | to_index = from_index; |
244 | to_c = from_c; | |
6662e69b | 245 | } |
3b4f4446 | 246 | else |
6662e69b | 247 | { |
3b4f4446 KH |
248 | to_index = CODE_POINT_TO_INDEX (charset, to); |
249 | to_c = from_c + (to_index - from_index); | |
6662e69b | 250 | } |
3b4f4446 KH |
251 | if (from_index < 0 || to_index < 0) |
252 | continue; | |
3263d5a2 | 253 | |
3263d5a2 | 254 | if (control_flag < 2) |
6662e69b | 255 | { |
3b4f4446 | 256 | int c; |
6662e69b | 257 | |
3b4f4446 KH |
258 | if (to_c > max_char) |
259 | max_char = to_c; | |
260 | else if (from_c < min_char) | |
261 | min_char = from_c; | |
262 | if (ascii_compatible_p) | |
263 | { | |
264 | if (! ASCII_BYTE_P (from_c)) | |
265 | { | |
266 | if (from_c < nonascii_min_char) | |
267 | nonascii_min_char = from_c; | |
268 | } | |
269 | else if (! ASCII_BYTE_P (to_c)) | |
270 | { | |
271 | nonascii_min_char = 0x80; | |
272 | } | |
273 | } | |
177c0ea7 | 274 | |
3b4f4446 KH |
275 | for (c = from_c; c <= to_c; c++) |
276 | CHARSET_FAST_MAP_SET (c, fast_map); | |
277 | ||
e9ce014c | 278 | if (control_flag == 1) |
3263d5a2 | 279 | { |
e9ce014c | 280 | unsigned code = from; |
e9ce014c | 281 | |
e9ce014c KH |
282 | if (CHARSET_COMPACT_CODES_P (charset)) |
283 | while (1) | |
284 | { | |
3b4f4446 | 285 | ASET (vec, from_index, make_number (from_c)); |
8f924df7 KH |
286 | if (NILP (CHAR_TABLE_REF (table, from_c))) |
287 | CHAR_TABLE_SET (table, from_c, make_number (code)); | |
e9ce014c KH |
288 | if (from_index == to_index) |
289 | break; | |
3b4f4446 | 290 | from_index++, from_c++; |
e9ce014c KH |
291 | code = INDEX_TO_CODE_POINT (charset, from_index); |
292 | } | |
293 | else | |
3b4f4446 | 294 | for (; from_index <= to_index; from_index++, from_c++) |
e9ce014c | 295 | { |
3b4f4446 | 296 | ASET (vec, from_index, make_number (from_c)); |
8f924df7 KH |
297 | if (NILP (CHAR_TABLE_REF (table, from_c))) |
298 | CHAR_TABLE_SET (table, from_c, make_number (from_index)); | |
e9ce014c | 299 | } |
3263d5a2 | 300 | } |
3263d5a2 | 301 | } |
2e344af3 KH |
302 | else |
303 | { | |
69f8de5b | 304 | unsigned code = from; |
3b4f4446 | 305 | |
69f8de5b KH |
306 | while (1) |
307 | { | |
308 | int c1 = DECODE_CHAR (charset, code); | |
8f924df7 | 309 | |
3263d5a2 KH |
310 | if (c1 >= 0) |
311 | { | |
3b4f4446 | 312 | CHAR_TABLE_SET (table, from_c, make_number (c1)); |
16fed1fc | 313 | CHAR_TABLE_SET (Vchar_unify_table, c1, make_number (from_c)); |
3263d5a2 KH |
314 | if (CHAR_TABLE_P (Vchar_unified_charset_table)) |
315 | CHAR_TABLE_SET (Vchar_unified_charset_table, c1, | |
316 | CHARSET_NAME (charset)); | |
317 | } | |
69f8de5b KH |
318 | if (from_index == to_index) |
319 | break; | |
3b4f4446 | 320 | from_index++, from_c++; |
69f8de5b | 321 | code = INDEX_TO_CODE_POINT (charset, from_index); |
3263d5a2 | 322 | } |
2e344af3 | 323 | } |
8ac5a9cc | 324 | } |
3263d5a2 KH |
325 | |
326 | if (control_flag < 2) | |
4ed46869 | 327 | { |
3263d5a2 KH |
328 | CHARSET_MIN_CHAR (charset) = (ascii_compatible_p |
329 | ? nonascii_min_char : min_char); | |
330 | CHARSET_MAX_CHAR (charset) = max_char; | |
e9ce014c | 331 | if (control_flag == 1) |
4ed46869 | 332 | { |
3263d5a2 KH |
333 | CHARSET_DECODER (charset) = vec; |
334 | CHARSET_ENCODER (charset) = table; | |
4ed46869 KH |
335 | } |
336 | } | |
2e344af3 | 337 | else |
8f924df7 | 338 | CHARSET_DEUNIFIER (charset) = table; |
4ed46869 KH |
339 | } |
340 | ||
12bcae05 | 341 | |
3263d5a2 KH |
342 | /* Read a hexadecimal number (preceded by "0x") from the file FP while |
343 | paying attention to comment charcter '#'. */ | |
12bcae05 | 344 | |
3263d5a2 KH |
345 | static INLINE unsigned |
346 | read_hex (fp, eof) | |
347 | FILE *fp; | |
348 | int *eof; | |
12bcae05 | 349 | { |
3263d5a2 KH |
350 | int c; |
351 | unsigned n; | |
12bcae05 | 352 | |
3263d5a2 KH |
353 | while ((c = getc (fp)) != EOF) |
354 | { | |
69f8de5b | 355 | if (c == '#') |
3263d5a2 KH |
356 | { |
357 | while ((c = getc (fp)) != EOF && c != '\n'); | |
358 | } | |
359 | else if (c == '0') | |
360 | { | |
361 | if ((c = getc (fp)) == EOF || c == 'x') | |
362 | break; | |
363 | } | |
8f924df7 | 364 | } |
3263d5a2 KH |
365 | if (c == EOF) |
366 | { | |
367 | *eof = 1; | |
368 | return 0; | |
369 | } | |
370 | *eof = 0; | |
371 | n = 0; | |
372 | if (c == 'x') | |
373 | while ((c = getc (fp)) != EOF && isxdigit (c)) | |
374 | n = ((n << 4) | |
375 | | (c <= '9' ? c - '0' : c <= 'F' ? c - 'A' + 10 : c - 'a' + 10)); | |
376 | else | |
377 | while ((c = getc (fp)) != EOF && isdigit (c)) | |
378 | n = (n * 10) + c - '0'; | |
e9ce014c KH |
379 | if (c != EOF) |
380 | ungetc (c, fp); | |
3263d5a2 KH |
381 | return n; |
382 | } | |
12bcae05 | 383 | |
537efd8d | 384 | |
3263d5a2 | 385 | /* Return a mapping vector for CHARSET loaded from MAPFILE. |
e9ce014c KH |
386 | Each line of MAPFILE has this form |
387 | 0xAAAA 0xCCCC | |
388 | where 0xAAAA is a code-point and 0xCCCC is the corresponding | |
389 | character code, or this form | |
390 | 0xAAAA-0xBBBB 0xCCCC | |
391 | where 0xAAAA and 0xBBBB are code-points specifying a range, and | |
392 | 0xCCCC is the first character code of the range. | |
4ed46869 | 393 | |
3263d5a2 KH |
394 | The returned vector has this form: |
395 | [ CODE1 CHAR1 CODE2 CHAR2 .... ] | |
e9ce014c KH |
396 | where CODE1 is a code-point or a cons of code-points specifying a |
397 | range. */ | |
4ed46869 | 398 | |
c449997d | 399 | extern void add_to_log P_ ((char *, Lisp_Object, Lisp_Object)); |
4ed46869 | 400 | |
e9ce014c KH |
401 | static void |
402 | load_charset_map_from_file (charset, mapfile, control_flag) | |
3263d5a2 KH |
403 | struct charset *charset; |
404 | Lisp_Object mapfile; | |
e9ce014c | 405 | int control_flag; |
4ed46869 | 406 | { |
e9ce014c KH |
407 | unsigned min_code = CHARSET_MIN_CODE (charset); |
408 | unsigned max_code = CHARSET_MAX_CODE (charset); | |
3263d5a2 KH |
409 | int fd; |
410 | FILE *fp; | |
3263d5a2 KH |
411 | int eof; |
412 | Lisp_Object suffixes; | |
e9ce014c KH |
413 | struct charset_map_entries *head, *entries; |
414 | int n_entries; | |
4ed46869 | 415 | |
3263d5a2 KH |
416 | suffixes = Fcons (build_string (".map"), |
417 | Fcons (build_string (".TXT"), Qnil)); | |
4ed46869 | 418 | |
4beef065 | 419 | fd = openp (Vcharset_map_path, mapfile, suffixes, NULL, Qnil); |
3263d5a2 KH |
420 | if (fd < 0 |
421 | || ! (fp = fdopen (fd, "r"))) | |
422 | { | |
423 | add_to_log ("Failure in loading charset map: %S", mapfile, Qnil); | |
e9ce014c | 424 | return; |
3263d5a2 | 425 | } |
4ed46869 | 426 | |
e9ce014c KH |
427 | head = entries = ((struct charset_map_entries *) |
428 | alloca (sizeof (struct charset_map_entries))); | |
429 | n_entries = 0; | |
3263d5a2 KH |
430 | eof = 0; |
431 | while (1) | |
432 | { | |
e9ce014c KH |
433 | unsigned from, to; |
434 | int c; | |
435 | int idx; | |
4ed46869 | 436 | |
e9ce014c | 437 | from = read_hex (fp, &eof); |
3263d5a2 KH |
438 | if (eof) |
439 | break; | |
e9ce014c KH |
440 | if (getc (fp) == '-') |
441 | to = read_hex (fp, &eof); | |
442 | else | |
443 | to = from; | |
444 | c = (int) read_hex (fp, &eof); | |
ac4137cc | 445 | |
e9ce014c KH |
446 | if (from < min_code || to > max_code || from > to || c > MAX_CHAR) |
447 | continue; | |
ac4137cc | 448 | |
e9ce014c | 449 | if (n_entries > 0 && (n_entries % 0x10000) == 0) |
3263d5a2 | 450 | { |
e9ce014c KH |
451 | entries->next = ((struct charset_map_entries *) |
452 | alloca (sizeof (struct charset_map_entries))); | |
453 | entries = entries->next; | |
3263d5a2 | 454 | } |
e9ce014c KH |
455 | idx = n_entries % 0x10000; |
456 | entries->entry[idx].from = from; | |
457 | entries->entry[idx].to = to; | |
458 | entries->entry[idx].c = c; | |
459 | n_entries++; | |
3263d5a2 KH |
460 | } |
461 | fclose (fp); | |
462 | close (fd); | |
177c0ea7 | 463 | |
e9ce014c | 464 | load_charset_map (charset, head, n_entries, control_flag); |
4ed46869 KH |
465 | } |
466 | ||
e9ce014c KH |
467 | static void |
468 | load_charset_map_from_vector (charset, vec, control_flag) | |
469 | struct charset *charset; | |
470 | Lisp_Object vec; | |
471 | int control_flag; | |
23d2a7f1 | 472 | { |
e9ce014c KH |
473 | unsigned min_code = CHARSET_MIN_CODE (charset); |
474 | unsigned max_code = CHARSET_MAX_CODE (charset); | |
475 | struct charset_map_entries *head, *entries; | |
476 | int n_entries; | |
477 | int len = ASIZE (vec); | |
478 | int i; | |
23d2a7f1 | 479 | |
e9ce014c | 480 | if (len % 2 == 1) |
3263d5a2 | 481 | { |
e9ce014c KH |
482 | add_to_log ("Failure in loading charset map: %V", vec, Qnil); |
483 | return; | |
3263d5a2 | 484 | } |
35e623fb | 485 | |
e9ce014c KH |
486 | head = entries = ((struct charset_map_entries *) |
487 | alloca (sizeof (struct charset_map_entries))); | |
488 | n_entries = 0; | |
489 | for (i = 0; i < len; i += 2) | |
35e623fb | 490 | { |
e9ce014c KH |
491 | Lisp_Object val, val2; |
492 | unsigned from, to; | |
493 | int c; | |
494 | int idx; | |
d2665018 | 495 | |
e9ce014c KH |
496 | val = AREF (vec, i); |
497 | if (CONSP (val)) | |
bbf12bb3 | 498 | { |
e9ce014c KH |
499 | val2 = XCDR (val); |
500 | val = XCAR (val); | |
501 | CHECK_NATNUM (val); | |
502 | CHECK_NATNUM (val2); | |
503 | from = XFASTINT (val); | |
504 | to = XFASTINT (val2); | |
bbf12bb3 | 505 | } |
e9ce014c | 506 | else |
bbf12bb3 | 507 | { |
e9ce014c KH |
508 | CHECK_NATNUM (val); |
509 | from = to = XFASTINT (val); | |
bbf12bb3 | 510 | } |
e9ce014c KH |
511 | val = AREF (vec, i + 1); |
512 | CHECK_NATNUM (val); | |
513 | c = XFASTINT (val); | |
76d7b829 | 514 | |
e9ce014c KH |
515 | if (from < min_code || to > max_code || from > to || c > MAX_CHAR) |
516 | continue; | |
76d7b829 | 517 | |
dcc694d7 | 518 | if (n_entries > 0 && (n_entries % 0x10000) == 0) |
e9ce014c KH |
519 | { |
520 | entries->next = ((struct charset_map_entries *) | |
521 | alloca (sizeof (struct charset_map_entries))); | |
522 | entries = entries->next; | |
523 | } | |
524 | idx = n_entries % 0x10000; | |
525 | entries->entry[idx].from = from; | |
526 | entries->entry[idx].to = to; | |
527 | entries->entry[idx].c = c; | |
528 | n_entries++; | |
529 | } | |
76d7b829 | 530 | |
e9ce014c | 531 | load_charset_map (charset, head, n_entries, control_flag); |
ac4137cc KH |
532 | } |
533 | ||
3263d5a2 KH |
534 | static void |
535 | load_charset (charset) | |
536 | struct charset *charset; | |
76d7b829 | 537 | { |
3263d5a2 | 538 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP_DEFERRED) |
76d7b829 | 539 | { |
3263d5a2 | 540 | Lisp_Object map; |
76d7b829 | 541 | |
3263d5a2 KH |
542 | map = CHARSET_MAP (charset); |
543 | if (STRINGP (map)) | |
e9ce014c | 544 | load_charset_map_from_file (charset, map, 1); |
bbf12bb3 | 545 | else |
e9ce014c | 546 | load_charset_map_from_vector (charset, map, 1); |
3263d5a2 | 547 | CHARSET_METHOD (charset) = CHARSET_METHOD_MAP; |
76d7b829 | 548 | } |
4ed46869 | 549 | } |
76d7b829 | 550 | |
3263d5a2 KH |
551 | |
552 | DEFUN ("charsetp", Fcharsetp, Scharsetp, 1, 1, 0, | |
553 | doc: /* Return non-nil if and only if OBJECT is a charset.*/) | |
554 | (object) | |
555 | Lisp_Object object; | |
23d2a7f1 | 556 | { |
3263d5a2 | 557 | return (CHARSETP (object) ? Qt : Qnil); |
76d7b829 KH |
558 | } |
559 | ||
4ed46869 KH |
560 | |
561 | void | |
374c5cfd KH |
562 | map_charset_chars (c_function, function, arg, |
563 | charset, from, to) | |
564 | void (*c_function) P_ ((Lisp_Object, Lisp_Object)); | |
565 | Lisp_Object function, arg; | |
566 | struct charset *charset; | |
567 | unsigned from, to; | |
4ed46869 | 568 | { |
3263d5a2 | 569 | Lisp_Object range; |
374c5cfd | 570 | int partial; |
3263d5a2 | 571 | |
8f924df7 | 572 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP_DEFERRED) |
3263d5a2 KH |
573 | load_charset (charset); |
574 | ||
374c5cfd KH |
575 | partial = (from > CHARSET_MIN_CODE (charset) |
576 | || to < CHARSET_MAX_CODE (charset)); | |
577 | ||
578 | if (CHARSET_UNIFIED_P (charset) | |
579 | && CHAR_TABLE_P (CHARSET_DEUNIFIER (charset))) | |
4ed46869 | 580 | { |
374c5cfd KH |
581 | map_char_table_for_charset (c_function, function, |
582 | CHARSET_DEUNIFIER (charset), arg, | |
583 | partial ? charset : NULL, from, to); | |
4ed46869 | 584 | } |
374c5cfd | 585 | |
3263d5a2 | 586 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_OFFSET) |
4ed46869 | 587 | { |
374c5cfd KH |
588 | int from_idx = CODE_POINT_TO_INDEX (charset, from); |
589 | int to_idx = CODE_POINT_TO_INDEX (charset, to); | |
590 | int from_c = from_idx + CHARSET_CODE_OFFSET (charset); | |
591 | int to_c = to_idx + CHARSET_CODE_OFFSET (charset); | |
592 | ||
593 | range = Fcons (make_number (from_c), make_number (to_c)); | |
3263d5a2 | 594 | if (NILP (function)) |
5af5dd92 | 595 | (*c_function) (arg, range); |
3263d5a2 KH |
596 | else |
597 | call2 (function, range, arg); | |
c83ef371 | 598 | } |
3263d5a2 KH |
599 | else if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP) |
600 | { | |
601 | if (! CHAR_TABLE_P (CHARSET_ENCODER (charset))) | |
602 | return; | |
374c5cfd KH |
603 | map_char_table_for_charset (c_function, function, |
604 | CHARSET_ENCODER (charset), arg, | |
605 | partial ? charset : NULL, from, to); | |
3263d5a2 | 606 | } |
374c5cfd | 607 | else if (CHARSET_METHOD (charset) == CHARSET_METHOD_SUBSET) |
4ed46869 | 608 | { |
374c5cfd KH |
609 | Lisp_Object subset_info; |
610 | int offset; | |
611 | ||
612 | subset_info = CHARSET_SUBSET (charset); | |
613 | charset = CHARSET_FROM_ID (XFASTINT (AREF (subset_info, 0))); | |
614 | offset = XINT (AREF (subset_info, 3)); | |
615 | from -= offset; | |
616 | if (from < XFASTINT (AREF (subset_info, 1))) | |
617 | from = XFASTINT (AREF (subset_info, 1)); | |
618 | to -= offset; | |
619 | if (to > XFASTINT (AREF (subset_info, 2))) | |
620 | to = XFASTINT (AREF (subset_info, 2)); | |
621 | map_charset_chars (c_function, function, arg, charset, from, to); | |
4ed46869 | 622 | } |
374c5cfd KH |
623 | else /* i.e. CHARSET_METHOD_SUPERSET */ |
624 | { | |
625 | Lisp_Object parents; | |
4ed46869 | 626 | |
374c5cfd KH |
627 | for (parents = CHARSET_SUPERSET (charset); CONSP (parents); |
628 | parents = XCDR (parents)) | |
bbf12bb3 | 629 | { |
374c5cfd KH |
630 | int offset; |
631 | unsigned this_from, this_to; | |
632 | ||
633 | charset = CHARSET_FROM_ID (XFASTINT (XCAR (XCAR (parents)))); | |
634 | offset = XINT (XCDR (XCAR (parents))); | |
635 | this_from = from - offset; | |
636 | this_to = to - offset; | |
637 | if (this_from < CHARSET_MIN_CODE (charset)) | |
638 | this_from = CHARSET_MIN_CODE (charset); | |
639 | if (this_to > CHARSET_MAX_CODE (charset)) | |
640 | this_to = CHARSET_MAX_CODE (charset); | |
111daccf KH |
641 | map_charset_chars (c_function, function, arg, charset, |
642 | this_from, this_to); | |
bbf12bb3 | 643 | } |
35e623fb | 644 | } |
4ed46869 KH |
645 | } |
646 | ||
374c5cfd | 647 | DEFUN ("map-charset-chars", Fmap_charset_chars, Smap_charset_chars, 2, 5, 0, |
04c2f2c5 | 648 | doc: /* Call FUNCTION for all characters in CHARSET. |
374c5cfd | 649 | FUNCTION is called with an argument RANGE and the optional 3rd |
3263d5a2 | 650 | argument ARG. |
4ed46869 | 651 | |
374c5cfd KH |
652 | RANGE is a cons (FROM . TO), where FROM and TO indicate a range of |
653 | characters contained in CHARSET. | |
4ed46869 | 654 | |
374c5cfd | 655 | The optional 4th and 5th arguments FROM-CODE and TO-CODE specify the |
8f924df7 | 656 | range of code points of target characters. */) |
374c5cfd KH |
657 | (function, charset, arg, from_code, to_code) |
658 | Lisp_Object function, charset, arg, from_code, to_code; | |
4ed46869 | 659 | { |
374c5cfd | 660 | struct charset *cs; |
16fed1fc | 661 | unsigned from, to; |
4ed46869 | 662 | |
374c5cfd KH |
663 | CHECK_CHARSET_GET_CHARSET (charset, cs); |
664 | if (NILP (from_code)) | |
16fed1fc | 665 | from = CHARSET_MIN_CODE (cs); |
970b7474 | 666 | else |
4ed46869 | 667 | { |
970b7474 KH |
668 | CHECK_NATNUM (from_code); |
669 | from = XINT (from_code); | |
670 | if (from < CHARSET_MIN_CODE (cs)) | |
671 | from = CHARSET_MIN_CODE (cs); | |
4ed46869 | 672 | } |
374c5cfd | 673 | if (NILP (to_code)) |
970b7474 | 674 | to = CHARSET_MAX_CODE (cs); |
4ed46869 KH |
675 | else |
676 | { | |
970b7474 KH |
677 | CHECK_NATNUM (to_code); |
678 | to = XINT (to_code); | |
679 | if (to > CHARSET_MAX_CODE (cs)) | |
680 | to = CHARSET_MAX_CODE (cs); | |
4ed46869 | 681 | } |
16fed1fc | 682 | map_charset_chars (NULL, function, arg, cs, from, to); |
3263d5a2 | 683 | return Qnil; |
35e623fb | 684 | } |
4ed46869 | 685 | |
4ed46869 | 686 | |
3263d5a2 KH |
687 | /* Define a charset according to the arguments. The Nth argument is |
688 | the Nth attribute of the charset (the last attribute `charset-id' | |
689 | is not included). See the docstring of `define-charset' for the | |
690 | detail. */ | |
4ed46869 | 691 | |
3263d5a2 KH |
692 | DEFUN ("define-charset-internal", Fdefine_charset_internal, |
693 | Sdefine_charset_internal, charset_arg_max, MANY, 0, | |
04c2f2c5 DL |
694 | doc: /* For internal use only. |
695 | usage: (define-charset-internal ...) */) | |
3263d5a2 KH |
696 | (nargs, args) |
697 | int nargs; | |
698 | Lisp_Object *args; | |
4ed46869 | 699 | { |
3263d5a2 KH |
700 | /* Charset attr vector. */ |
701 | Lisp_Object attrs; | |
702 | Lisp_Object val; | |
703 | unsigned hash_code; | |
704 | struct Lisp_Hash_Table *hash_table = XHASH_TABLE (Vcharset_hash_table); | |
69f8de5b | 705 | int i, j; |
3263d5a2 KH |
706 | struct charset charset; |
707 | int id; | |
708 | int dimension; | |
709 | int new_definition_p; | |
710 | int nchars; | |
711 | ||
712 | if (nargs != charset_arg_max) | |
713 | return Fsignal (Qwrong_number_of_arguments, | |
714 | Fcons (intern ("define-charset-internal"), | |
715 | make_number (nargs))); | |
716 | ||
717 | attrs = Fmake_vector (make_number (charset_attr_max), Qnil); | |
718 | ||
719 | CHECK_SYMBOL (args[charset_arg_name]); | |
720 | ASET (attrs, charset_name, args[charset_arg_name]); | |
721 | ||
722 | val = args[charset_arg_code_space]; | |
723 | for (i = 0, dimension = 0, nchars = 1; i < 4; i++) | |
76d7b829 | 724 | { |
3263d5a2 KH |
725 | int min_byte, max_byte; |
726 | ||
727 | min_byte = XINT (Faref (val, make_number (i * 2))); | |
728 | max_byte = XINT (Faref (val, make_number (i * 2 + 1))); | |
729 | if (min_byte < 0 || min_byte > max_byte || max_byte >= 256) | |
730 | error ("Invalid :code-space value"); | |
731 | charset.code_space[i * 4] = min_byte; | |
732 | charset.code_space[i * 4 + 1] = max_byte; | |
733 | charset.code_space[i * 4 + 2] = max_byte - min_byte + 1; | |
734 | nchars *= charset.code_space[i * 4 + 2]; | |
735 | charset.code_space[i * 4 + 3] = nchars; | |
736 | if (max_byte > 0) | |
737 | dimension = i + 1; | |
738 | } | |
4ed46869 | 739 | |
3263d5a2 KH |
740 | val = args[charset_arg_dimension]; |
741 | if (NILP (val)) | |
742 | charset.dimension = dimension; | |
743 | else | |
4ed46869 | 744 | { |
3263d5a2 KH |
745 | CHECK_NATNUM (val); |
746 | charset.dimension = XINT (val); | |
747 | if (charset.dimension < 1 || charset.dimension > 4) | |
748 | args_out_of_range_3 (val, make_number (1), make_number (4)); | |
4ed46869 KH |
749 | } |
750 | ||
3263d5a2 KH |
751 | charset.code_linear_p |
752 | = (charset.dimension == 1 | |
753 | || (charset.code_space[2] == 256 | |
754 | && (charset.dimension == 2 | |
755 | || (charset.code_space[6] == 256 | |
756 | && (charset.dimension == 3 | |
757 | || charset.code_space[10] == 256))))); | |
758 | ||
69f8de5b | 759 | if (! charset.code_linear_p) |
4ed46869 | 760 | { |
69f8de5b | 761 | charset.code_space_mask = (unsigned char *) xmalloc (256); |
33df3183 | 762 | bzero (charset.code_space_mask, 256); |
69f8de5b KH |
763 | for (i = 0; i < 4; i++) |
764 | for (j = charset.code_space[i * 4]; j <= charset.code_space[i * 4 + 1]; | |
765 | j++) | |
766 | charset.code_space_mask[j] |= (1 << i); | |
4ed46869 KH |
767 | } |
768 | ||
3263d5a2 | 769 | charset.iso_chars_96 = charset.code_space[2] == 96; |
4ed46869 | 770 | |
3263d5a2 KH |
771 | charset.min_code = (charset.code_space[0] |
772 | | (charset.code_space[4] << 8) | |
773 | | (charset.code_space[8] << 16) | |
774 | | (charset.code_space[12] << 24)); | |
775 | charset.max_code = (charset.code_space[1] | |
776 | | (charset.code_space[5] << 8) | |
777 | | (charset.code_space[9] << 16) | |
778 | | (charset.code_space[13] << 24)); | |
820ee249 | 779 | charset.char_index_offset = 0; |
8a73a704 | 780 | |
820ee249 KH |
781 | val = args[charset_arg_min_code]; |
782 | if (! NILP (val)) | |
783 | { | |
784 | unsigned code; | |
fdb82f93 | 785 | |
820ee249 KH |
786 | if (INTEGERP (val)) |
787 | code = XINT (val); | |
788 | else | |
789 | { | |
790 | CHECK_CONS (val); | |
8f924df7 KH |
791 | CHECK_NUMBER_CAR (val); |
792 | CHECK_NUMBER_CDR (val); | |
820ee249 KH |
793 | code = (XINT (XCAR (val)) << 16) | (XINT (XCDR (val))); |
794 | } | |
795 | if (code < charset.min_code | |
796 | || code > charset.max_code) | |
797 | args_out_of_range_3 (make_number (charset.min_code), | |
798 | make_number (charset.max_code), val); | |
799 | charset.char_index_offset = CODE_POINT_TO_INDEX (&charset, code); | |
800 | charset.min_code = code; | |
801 | } | |
3fac5a51 | 802 | |
820ee249 KH |
803 | val = args[charset_arg_max_code]; |
804 | if (! NILP (val)) | |
3fac5a51 | 805 | { |
820ee249 KH |
806 | unsigned code; |
807 | ||
808 | if (INTEGERP (val)) | |
809 | code = XINT (val); | |
810 | else | |
811 | { | |
812 | CHECK_CONS (val); | |
8f924df7 KH |
813 | CHECK_NUMBER_CAR (val); |
814 | CHECK_NUMBER_CDR (val); | |
820ee249 KH |
815 | code = (XINT (XCAR (val)) << 16) | (XINT (XCDR (val))); |
816 | } | |
817 | if (code < charset.min_code | |
818 | || code > charset.max_code) | |
819 | args_out_of_range_3 (make_number (charset.min_code), | |
820 | make_number (charset.max_code), val); | |
821 | charset.max_code = code; | |
3fac5a51 | 822 | } |
3fac5a51 | 823 | |
e9ce014c | 824 | charset.compact_codes_p = charset.max_code < 0x1000000; |
4ed46869 | 825 | |
3263d5a2 KH |
826 | val = args[charset_arg_invalid_code]; |
827 | if (NILP (val)) | |
828 | { | |
829 | if (charset.min_code > 0) | |
830 | charset.invalid_code = 0; | |
bbf12bb3 KH |
831 | else |
832 | { | |
3263d5a2 KH |
833 | XSETINT (val, charset.max_code + 1); |
834 | if (XINT (val) == charset.max_code + 1) | |
835 | charset.invalid_code = charset.max_code + 1; | |
836 | else | |
837 | error ("Attribute :invalid-code must be specified"); | |
76d7b829 | 838 | } |
76d7b829 | 839 | } |
3263d5a2 KH |
840 | else |
841 | { | |
842 | CHECK_NATNUM (val); | |
843 | charset.invalid_code = XFASTINT (val); | |
844 | } | |
4ed46869 | 845 | |
3263d5a2 KH |
846 | val = args[charset_arg_iso_final]; |
847 | if (NILP (val)) | |
848 | charset.iso_final = -1; | |
849 | else | |
850 | { | |
851 | CHECK_NUMBER (val); | |
852 | if (XINT (val) < '0' || XINT (val) > 127) | |
853 | error ("Invalid iso-final-char: %d", XINT (val)); | |
854 | charset.iso_final = XINT (val); | |
855 | } | |
4ed46869 | 856 | |
3263d5a2 KH |
857 | val = args[charset_arg_iso_revision]; |
858 | if (NILP (val)) | |
859 | charset.iso_revision = -1; | |
860 | else | |
4ed46869 | 861 | { |
3263d5a2 KH |
862 | CHECK_NUMBER (val); |
863 | if (XINT (val) > 63) | |
864 | args_out_of_range (make_number (63), val); | |
865 | charset.iso_revision = XINT (val); | |
4ed46869 | 866 | } |
4ed46869 | 867 | |
3263d5a2 KH |
868 | val = args[charset_arg_emacs_mule_id]; |
869 | if (NILP (val)) | |
870 | charset.emacs_mule_id = -1; | |
4ed46869 KH |
871 | else |
872 | { | |
3263d5a2 KH |
873 | CHECK_NATNUM (val); |
874 | if ((XINT (val) > 0 && XINT (val) <= 128) || XINT (val) >= 256) | |
875 | error ("Invalid emacs-mule-id: %d", XINT (val)); | |
876 | charset.emacs_mule_id = XINT (val); | |
c83ef371 | 877 | } |
f6302ac9 | 878 | |
3263d5a2 | 879 | charset.ascii_compatible_p = ! NILP (args[charset_arg_ascii_compatible_p]); |
1d67c29b | 880 | |
3263d5a2 | 881 | charset.supplementary_p = ! NILP (args[charset_arg_supplementary_p]); |
4ed46869 | 882 | |
3263d5a2 KH |
883 | charset.unified_p = 0; |
884 | ||
885 | bzero (charset.fast_map, sizeof (charset.fast_map)); | |
886 | ||
887 | if (! NILP (args[charset_arg_code_offset])) | |
888 | { | |
889 | val = args[charset_arg_code_offset]; | |
890 | CHECK_NUMBER (val); | |
891 | ||
892 | charset.method = CHARSET_METHOD_OFFSET; | |
893 | charset.code_offset = XINT (val); | |
894 | ||
895 | i = CODE_POINT_TO_INDEX (&charset, charset.min_code); | |
896 | charset.min_char = i + charset.code_offset; | |
897 | i = CODE_POINT_TO_INDEX (&charset, charset.max_code); | |
898 | charset.max_char = i + charset.code_offset; | |
899 | if (charset.max_char > MAX_CHAR) | |
900 | error ("Unsupported max char: %d", charset.max_char); | |
901 | ||
f148205f KH |
902 | i = (charset.min_char >> 7) << 7; |
903 | for (; i < 0x10000 && i <= charset.max_char; i += 128) | |
3263d5a2 | 904 | CHARSET_FAST_MAP_SET (i, charset.fast_map); |
f148205f | 905 | i = (i >> 12) << 12; |
3263d5a2 KH |
906 | for (; i <= charset.max_char; i += 0x1000) |
907 | CHARSET_FAST_MAP_SET (i, charset.fast_map); | |
908 | } | |
909 | else if (! NILP (args[charset_arg_map])) | |
910 | { | |
911 | val = args[charset_arg_map]; | |
912 | ASET (attrs, charset_map, val); | |
913 | if (STRINGP (val)) | |
e9ce014c KH |
914 | load_charset_map_from_file (&charset, val, 0); |
915 | else | |
916 | load_charset_map_from_vector (&charset, val, 0); | |
3263d5a2 KH |
917 | charset.method = CHARSET_METHOD_MAP_DEFERRED; |
918 | } | |
374c5cfd | 919 | else if (! NILP (args[charset_arg_subset])) |
3263d5a2 | 920 | { |
374c5cfd KH |
921 | Lisp_Object parent; |
922 | Lisp_Object parent_min_code, parent_max_code, parent_code_offset; | |
923 | struct charset *parent_charset; | |
924 | ||
925 | val = args[charset_arg_subset]; | |
926 | parent = Fcar (val); | |
927 | CHECK_CHARSET_GET_CHARSET (parent, parent_charset); | |
928 | parent_min_code = Fnth (make_number (1), val); | |
929 | CHECK_NATNUM (parent_min_code); | |
930 | parent_max_code = Fnth (make_number (2), val); | |
931 | CHECK_NATNUM (parent_max_code); | |
932 | parent_code_offset = Fnth (make_number (3), val); | |
933 | CHECK_NUMBER (parent_code_offset); | |
934 | val = Fmake_vector (make_number (4), Qnil); | |
935 | ASET (val, 0, make_number (parent_charset->id)); | |
936 | ASET (val, 1, parent_min_code); | |
937 | ASET (val, 2, parent_max_code); | |
938 | ASET (val, 3, parent_code_offset); | |
939 | ASET (attrs, charset_subset, val); | |
940 | ||
941 | charset.method = CHARSET_METHOD_SUBSET; | |
942 | /* Here, we just copy the parent's fast_map. It's not accurate, | |
943 | but at least it works for quickly detecting which character | |
944 | DOESN'T belong to this charset. */ | |
945 | for (i = 0; i < 190; i++) | |
946 | charset.fast_map[i] = parent_charset->fast_map[i]; | |
947 | ||
948 | /* We also copy these for parents. */ | |
949 | charset.min_char = parent_charset->min_char; | |
950 | charset.max_char = parent_charset->max_char; | |
951 | } | |
952 | else if (! NILP (args[charset_arg_superset])) | |
0282eb69 | 953 | { |
374c5cfd KH |
954 | val = args[charset_arg_superset]; |
955 | charset.method = CHARSET_METHOD_SUPERSET; | |
3263d5a2 | 956 | val = Fcopy_sequence (val); |
374c5cfd | 957 | ASET (attrs, charset_superset, val); |
3263d5a2 KH |
958 | |
959 | charset.min_char = MAX_CHAR; | |
960 | charset.max_char = 0; | |
961 | for (; ! NILP (val); val = Fcdr (val)) | |
0282eb69 | 962 | { |
3263d5a2 KH |
963 | Lisp_Object elt, car_part, cdr_part; |
964 | int this_id, offset; | |
965 | struct charset *this_charset; | |
2e344af3 | 966 | |
3263d5a2 KH |
967 | elt = Fcar (val); |
968 | if (CONSP (elt)) | |
2e344af3 | 969 | { |
3263d5a2 KH |
970 | car_part = XCAR (elt); |
971 | cdr_part = XCDR (elt); | |
972 | CHECK_CHARSET_GET_ID (car_part, this_id); | |
973 | CHECK_NUMBER (cdr_part); | |
974 | offset = XINT (cdr_part); | |
177c0ea7 | 975 | } |
3263d5a2 | 976 | else |
4ed46869 | 977 | { |
3263d5a2 KH |
978 | CHECK_CHARSET_GET_ID (elt, this_id); |
979 | offset = 0; | |
4ed46869 | 980 | } |
3263d5a2 KH |
981 | XSETCAR (val, Fcons (make_number (this_id), make_number (offset))); |
982 | ||
983 | this_charset = CHARSET_FROM_ID (this_id); | |
984 | if (charset.min_char > this_charset->min_char) | |
985 | charset.min_char = this_charset->min_char; | |
986 | if (charset.max_char < this_charset->max_char) | |
987 | charset.max_char = this_charset->max_char; | |
988 | for (i = 0; i < 190; i++) | |
989 | charset.fast_map[i] |= this_charset->fast_map[i]; | |
0282eb69 | 990 | } |
0282eb69 | 991 | } |
2e344af3 | 992 | else |
3263d5a2 | 993 | error ("None of :code-offset, :map, :parents are specified"); |
05505664 | 994 | |
3263d5a2 KH |
995 | val = args[charset_arg_unify_map]; |
996 | if (! NILP (val) && !STRINGP (val)) | |
997 | CHECK_VECTOR (val); | |
998 | ASET (attrs, charset_unify_map, val); | |
05505664 | 999 | |
3263d5a2 KH |
1000 | CHECK_LIST (args[charset_arg_plist]); |
1001 | ASET (attrs, charset_plist, args[charset_arg_plist]); | |
4ed46869 | 1002 | |
3263d5a2 KH |
1003 | charset.hash_index = hash_lookup (hash_table, args[charset_arg_name], |
1004 | &hash_code); | |
1005 | if (charset.hash_index >= 0) | |
1006 | { | |
1007 | new_definition_p = 0; | |
4f65af01 | 1008 | id = XFASTINT (CHARSET_SYMBOL_ID (args[charset_arg_name])); |
3263d5a2 KH |
1009 | HASH_VALUE (hash_table, charset.hash_index) = attrs; |
1010 | } | |
1a45ff10 | 1011 | else |
3263d5a2 KH |
1012 | { |
1013 | charset.hash_index = hash_put (hash_table, args[charset_arg_name], attrs, | |
1014 | hash_code); | |
1015 | if (charset_table_used == charset_table_size) | |
1016 | { | |
2fe1edd1 KH |
1017 | struct charset *new_table |
1018 | = (struct charset *) xmalloc (sizeof (struct charset) | |
1019 | * (charset_table_size + 16)); | |
1020 | bcopy (charset_table, new_table, | |
1021 | sizeof (struct charset) * charset_table_size); | |
1022 | charset_table_size += 16; | |
1023 | charset_table = new_table; | |
3263d5a2 KH |
1024 | } |
1025 | id = charset_table_used++; | |
3263d5a2 KH |
1026 | new_definition_p = 1; |
1027 | } | |
2e344af3 | 1028 | |
4f65af01 | 1029 | ASET (attrs, charset_id, make_number (id)); |
3263d5a2 KH |
1030 | charset.id = id; |
1031 | charset_table[id] = charset; | |
2e344af3 | 1032 | |
3263d5a2 | 1033 | if (charset.iso_final >= 0) |
4ed46869 | 1034 | { |
3263d5a2 KH |
1035 | ISO_CHARSET_TABLE (charset.dimension, charset.iso_chars_96, |
1036 | charset.iso_final) = id; | |
1037 | if (new_definition_p) | |
1038 | Viso_2022_charset_list = nconc2 (Viso_2022_charset_list, | |
1039 | Fcons (make_number (id), Qnil)); | |
7c7dceee KH |
1040 | if (ISO_CHARSET_TABLE (1, 0, 'J') == id) |
1041 | charset_jisx0201_roman = id; | |
1042 | else if (ISO_CHARSET_TABLE (2, 0, '@') == id) | |
1043 | charset_jisx0208_1978 = id; | |
1044 | else if (ISO_CHARSET_TABLE (2, 0, 'B') == id) | |
1045 | charset_jisx0208 = id; | |
4ed46869 | 1046 | } |
d0cf2d48 | 1047 | |
3263d5a2 | 1048 | if (charset.emacs_mule_id >= 0) |
4ed46869 | 1049 | { |
3263d5a2 | 1050 | emacs_mule_charset[charset.emacs_mule_id] = CHARSET_FROM_ID (id); |
4f65af01 KH |
1051 | if (charset.emacs_mule_id < 0xA0) |
1052 | emacs_mule_bytes[charset.emacs_mule_id] = charset.dimension + 1; | |
3b1ae89b KH |
1053 | else |
1054 | emacs_mule_bytes[charset.emacs_mule_id] = charset.dimension + 2; | |
3263d5a2 KH |
1055 | if (new_definition_p) |
1056 | Vemacs_mule_charset_list = nconc2 (Vemacs_mule_charset_list, | |
1057 | Fcons (make_number (id), Qnil)); | |
4ed46869 KH |
1058 | } |
1059 | ||
3263d5a2 KH |
1060 | if (new_definition_p) |
1061 | { | |
1062 | Vcharset_list = Fcons (args[charset_arg_name], Vcharset_list); | |
8055c66a KH |
1063 | if (charset.supplementary_p) |
1064 | Vcharset_ordered_list = nconc2 (Vcharset_ordered_list, | |
1065 | Fcons (make_number (id), Qnil)); | |
1066 | else | |
880820fe KH |
1067 | { |
1068 | Lisp_Object tail; | |
1069 | ||
1070 | for (tail = Vcharset_ordered_list; CONSP (tail); tail = XCDR (tail)) | |
1071 | { | |
1072 | struct charset *cs = CHARSET_FROM_ID (XINT (XCAR (tail))); | |
1073 | ||
1074 | if (cs->supplementary_p) | |
1075 | break; | |
1076 | } | |
1077 | if (EQ (tail, Vcharset_ordered_list)) | |
1078 | Vcharset_ordered_list = Fcons (make_number (id), | |
1079 | Vcharset_ordered_list); | |
1080 | else if (NILP (tail)) | |
1081 | Vcharset_ordered_list = nconc2 (Vcharset_ordered_list, | |
1082 | Fcons (make_number (id), Qnil)); | |
1083 | else | |
1084 | { | |
1085 | val = Fcons (XCAR (tail), XCDR (tail)); | |
1086 | XSETCDR (tail, val); | |
1087 | XSETCAR (tail, make_number (id)); | |
1088 | } | |
1089 | } | |
dbbb237d | 1090 | charset_ordered_list_tick++; |
3263d5a2 | 1091 | } |
4ed46869 | 1092 | |
3263d5a2 | 1093 | return Qnil; |
4ed46869 KH |
1094 | } |
1095 | ||
2fe1edd1 KH |
1096 | |
1097 | /* Same as Fdefine_charset_internal but arguments are more convenient | |
1098 | to call from C (typically in syms_of_charset). This can define a | |
1099 | charset of `offset' method only. Return the ID of the new | |
1100 | charset. */ | |
1101 | ||
1102 | static int | |
1103 | define_charset_internal (name, dimension, code_space, min_code, max_code, | |
1104 | iso_final, iso_revision, emacs_mule_id, | |
7acf89e6 | 1105 | ascii_compatible, supplementary, |
2fe1edd1 KH |
1106 | code_offset) |
1107 | Lisp_Object name; | |
1108 | int dimension; | |
1109 | unsigned char *code_space; | |
1110 | unsigned min_code, max_code; | |
1111 | int iso_final, iso_revision, emacs_mule_id; | |
7acf89e6 | 1112 | int ascii_compatible, supplementary; |
2fe1edd1 KH |
1113 | int code_offset; |
1114 | { | |
1115 | Lisp_Object args[charset_arg_max]; | |
1116 | Lisp_Object plist[14]; | |
1117 | Lisp_Object val; | |
1118 | int i; | |
1119 | ||
1120 | args[charset_arg_name] = name; | |
1121 | args[charset_arg_dimension] = make_number (dimension); | |
1122 | val = Fmake_vector (make_number (8), make_number (0)); | |
1123 | for (i = 0; i < 8; i++) | |
1124 | ASET (val, i, make_number (code_space[i])); | |
1125 | args[charset_arg_code_space] = val; | |
1126 | args[charset_arg_min_code] = make_number (min_code); | |
1127 | args[charset_arg_max_code] = make_number (max_code); | |
1128 | args[charset_arg_iso_final] | |
1129 | = (iso_final < 0 ? Qnil : make_number (iso_final)); | |
1130 | args[charset_arg_iso_revision] = make_number (iso_revision); | |
1131 | args[charset_arg_emacs_mule_id] | |
1132 | = (emacs_mule_id < 0 ? Qnil : make_number (emacs_mule_id)); | |
1133 | args[charset_arg_ascii_compatible_p] = ascii_compatible ? Qt : Qnil; | |
7acf89e6 | 1134 | args[charset_arg_supplementary_p] = supplementary ? Qt : Qnil; |
2fe1edd1 KH |
1135 | args[charset_arg_invalid_code] = Qnil; |
1136 | args[charset_arg_code_offset] = make_number (code_offset); | |
1137 | args[charset_arg_map] = Qnil; | |
1138 | args[charset_arg_subset] = Qnil; | |
1139 | args[charset_arg_superset] = Qnil; | |
1140 | args[charset_arg_unify_map] = Qnil; | |
1141 | ||
1142 | plist[0] = intern (":name"); | |
1143 | plist[1] = args[charset_arg_name]; | |
1144 | plist[2] = intern (":dimension"); | |
1145 | plist[3] = args[charset_arg_dimension]; | |
1146 | plist[4] = intern (":code-space"); | |
1147 | plist[5] = args[charset_arg_code_space]; | |
1148 | plist[6] = intern (":iso-final-char"); | |
1149 | plist[7] = args[charset_arg_iso_final]; | |
1150 | plist[8] = intern (":emacs-mule-id"); | |
1151 | plist[9] = args[charset_arg_emacs_mule_id]; | |
1152 | plist[10] = intern (":ascii-compatible-p"); | |
1153 | plist[11] = args[charset_arg_ascii_compatible_p]; | |
1154 | plist[12] = intern (":code-offset"); | |
1155 | plist[13] = args[charset_arg_code_offset]; | |
1156 | ||
1157 | args[charset_arg_plist] = Flist (14, plist); | |
1158 | Fdefine_charset_internal (charset_arg_max, args); | |
1159 | ||
1160 | return XINT (CHARSET_SYMBOL_ID (name)); | |
1161 | } | |
1162 | ||
1163 | ||
3263d5a2 KH |
1164 | DEFUN ("define-charset-alias", Fdefine_charset_alias, |
1165 | Sdefine_charset_alias, 2, 2, 0, | |
1166 | doc: /* Define ALIAS as an alias for charset CHARSET. */) | |
1167 | (alias, charset) | |
1168 | Lisp_Object alias, charset; | |
4ed46869 | 1169 | { |
3263d5a2 KH |
1170 | Lisp_Object attr; |
1171 | ||
1172 | CHECK_CHARSET_GET_ATTR (charset, attr); | |
1173 | Fputhash (alias, attr, Vcharset_hash_table); | |
528623a0 | 1174 | Vcharset_list = Fcons (alias, Vcharset_list); |
3263d5a2 KH |
1175 | return Qnil; |
1176 | } | |
4ed46869 | 1177 | |
4ed46869 | 1178 | |
3263d5a2 | 1179 | DEFUN ("charset-plist", Fcharset_plist, Scharset_plist, 1, 1, 0, |
56a46d1d | 1180 | doc: /* Return the property list of CHARSET. */) |
3263d5a2 KH |
1181 | (charset) |
1182 | Lisp_Object charset; | |
1183 | { | |
1184 | Lisp_Object attrs; | |
1185 | ||
1186 | CHECK_CHARSET_GET_ATTR (charset, attrs); | |
1187 | return CHARSET_ATTR_PLIST (attrs); | |
1188 | } | |
1189 | ||
1190 | ||
1191 | DEFUN ("set-charset-plist", Fset_charset_plist, Sset_charset_plist, 2, 2, 0, | |
1192 | doc: /* Set CHARSET's property list to PLIST. */) | |
1193 | (charset, plist) | |
1194 | Lisp_Object charset, plist; | |
1195 | { | |
1196 | Lisp_Object attrs; | |
1197 | ||
1198 | CHECK_CHARSET_GET_ATTR (charset, attrs); | |
1199 | CHARSET_ATTR_PLIST (attrs) = plist; | |
1200 | return plist; | |
1201 | } | |
1202 | ||
1203 | ||
dbbb237d | 1204 | DEFUN ("unify-charset", Funify_charset, Sunify_charset, 1, 3, 0, |
56a46d1d DL |
1205 | doc: /* Unify characters of CHARSET with Unicode. |
1206 | This means reading the relevant file and installing the table defined | |
dbbb237d KH |
1207 | by CHARSET's `:unify-map' property. |
1208 | ||
64165ae2 DL |
1209 | Optional second arg UNIFY-MAP is a file name string or a vector. It has |
1210 | the same meaning as the `:unify-map' attribute in the function | |
dbbb237d KH |
1211 | `define-charset' (which see). |
1212 | ||
1213 | Optional third argument DEUNIFY, if non-nil, means to de-unify CHARSET. */) | |
1214 | (charset, unify_map, deunify) | |
1215 | Lisp_Object charset, unify_map, deunify; | |
8a73a704 | 1216 | { |
3263d5a2 KH |
1217 | int id; |
1218 | struct charset *cs; | |
8f924df7 | 1219 | |
3263d5a2 KH |
1220 | CHECK_CHARSET_GET_ID (charset, id); |
1221 | cs = CHARSET_FROM_ID (id); | |
1222 | if (CHARSET_METHOD (cs) == CHARSET_METHOD_MAP_DEFERRED) | |
1223 | load_charset (cs); | |
dbbb237d KH |
1224 | if (NILP (deunify) |
1225 | ? CHARSET_UNIFIED_P (cs) && ! NILP (CHARSET_DEUNIFIER (cs)) | |
1226 | : ! CHARSET_UNIFIED_P (cs)) | |
3263d5a2 | 1227 | return Qnil; |
dbbb237d | 1228 | |
3263d5a2 | 1229 | CHARSET_UNIFIED_P (cs) = 0; |
dbbb237d KH |
1230 | if (NILP (deunify)) |
1231 | { | |
1232 | if (CHARSET_METHOD (cs) != CHARSET_METHOD_OFFSET) | |
8f924df7 | 1233 | error ("Can't unify charset: %s", SDATA (SYMBOL_NAME (charset))); |
dbbb237d KH |
1234 | if (NILP (unify_map)) |
1235 | unify_map = CHARSET_UNIFY_MAP (cs); | |
1236 | if (STRINGP (unify_map)) | |
1237 | load_charset_map_from_file (cs, unify_map, 2); | |
1238 | else if (VECTORP (unify_map)) | |
1239 | load_charset_map_from_vector (cs, unify_map, 2); | |
1240 | else if (NILP (unify_map)) | |
1241 | error ("No unify-map for charset"); | |
1242 | else | |
1243 | error ("Bad unify-map arg"); | |
1244 | CHARSET_UNIFIED_P (cs) = 1; | |
1245 | } | |
1246 | else if (CHAR_TABLE_P (Vchar_unify_table)) | |
1247 | { | |
1248 | int min_code = CHARSET_MIN_CODE (cs); | |
1249 | int max_code = CHARSET_MAX_CODE (cs); | |
1250 | int min_char = DECODE_CHAR (cs, min_code); | |
1251 | int max_char = DECODE_CHAR (cs, max_code); | |
8f924df7 | 1252 | |
dbbb237d KH |
1253 | char_table_set_range (Vchar_unify_table, min_char, max_char, Qnil); |
1254 | } | |
8f924df7 | 1255 | |
3263d5a2 | 1256 | return Qnil; |
8a73a704 KH |
1257 | } |
1258 | ||
3fac5a51 KH |
1259 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
1260 | Sget_unused_iso_final_char, 2, 2, 0, | |
3263d5a2 | 1261 | doc: /* |
d0cf2d48 | 1262 | Return an unused ISO final char for a charset of DIMENSION and CHARS. |
fdb82f93 PJ |
1263 | DIMENSION is the number of bytes to represent a character: 1 or 2. |
1264 | CHARS is the number of characters in a dimension: 94 or 96. | |
1265 | ||
1266 | This final char is for private use, thus the range is `0' (48) .. `?' (63). | |
1721b6af | 1267 | If there's no unused final char for the specified kind of charset, |
fdb82f93 PJ |
1268 | return nil. */) |
1269 | (dimension, chars) | |
3fac5a51 KH |
1270 | Lisp_Object dimension, chars; |
1271 | { | |
1272 | int final_char; | |
1273 | ||
b7826503 PJ |
1274 | CHECK_NUMBER (dimension); |
1275 | CHECK_NUMBER (chars); | |
3263d5a2 KH |
1276 | if (XINT (dimension) != 1 && XINT (dimension) != 2 && XINT (dimension) != 3) |
1277 | args_out_of_range_3 (dimension, make_number (1), make_number (3)); | |
3fac5a51 | 1278 | if (XINT (chars) != 94 && XINT (chars) != 96) |
3263d5a2 | 1279 | args_out_of_range_3 (chars, make_number (94), make_number (96)); |
3fac5a51 | 1280 | for (final_char = '0'; final_char <= '?'; final_char++) |
3263d5a2 KH |
1281 | if (ISO_CHARSET_TABLE (XINT (dimension), XINT (chars), final_char) < 0) |
1282 | break; | |
3fac5a51 KH |
1283 | return (final_char <= '?' ? make_number (final_char) : Qnil); |
1284 | } | |
1285 | ||
3263d5a2 KH |
1286 | static void |
1287 | check_iso_charset_parameter (dimension, chars, final_char) | |
1288 | Lisp_Object dimension, chars, final_char; | |
4ed46869 | 1289 | { |
3263d5a2 KH |
1290 | CHECK_NATNUM (dimension); |
1291 | CHECK_NATNUM (chars); | |
1292 | CHECK_NATNUM (final_char); | |
4ed46869 | 1293 | |
3263d5a2 KH |
1294 | if (XINT (dimension) > 3) |
1295 | error ("Invalid DIMENSION %d, it should be 1, 2, or 3", XINT (dimension)); | |
4ed46869 KH |
1296 | if (XINT (chars) != 94 && XINT (chars) != 96) |
1297 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
3263d5a2 | 1298 | if (XINT (final_char) < '0' || XINT (final_char) > '~') |
4ed46869 | 1299 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); |
3263d5a2 KH |
1300 | } |
1301 | ||
1302 | ||
1303 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, | |
1304 | 4, 4, 0, | |
cefd8c4f KH |
1305 | doc: /* Declare an equivalent charset for ISO-2022 decoding. |
1306 | ||
1307 | On decoding by an ISO-2022 base coding system, when a charset | |
1308 | specified by DIMENSION, CHARS, and FINAL-CHAR is designated, behave as | |
1309 | if CHARSET is designated instead. */) | |
3263d5a2 KH |
1310 | (dimension, chars, final_char, charset) |
1311 | Lisp_Object dimension, chars, final_char, charset; | |
1312 | { | |
1313 | int id; | |
82215ce9 | 1314 | int chars_flag; |
4ed46869 | 1315 | |
3263d5a2 KH |
1316 | CHECK_CHARSET_GET_ID (charset, id); |
1317 | check_iso_charset_parameter (dimension, chars, final_char); | |
82215ce9 KH |
1318 | chars_flag = XINT (chars) == 96; |
1319 | ISO_CHARSET_TABLE (XINT (dimension), chars_flag, XINT (final_char)) = id; | |
4ed46869 KH |
1320 | return Qnil; |
1321 | } | |
1322 | ||
3263d5a2 | 1323 | |
2e344af3 KH |
1324 | /* Return information about charsets in the text at PTR of NBYTES |
1325 | bytes, which are NCHARS characters. The value is: | |
f6302ac9 | 1326 | |
cfe34140 | 1327 | 0: Each character is represented by one byte. This is always |
3263d5a2 KH |
1328 | true for a unibyte string. For a multibyte string, true if |
1329 | it contains only ASCII characters. | |
1330 | ||
28c026cd DL |
1331 | 1: No charsets other than ascii, control-1, and latin-1 are |
1332 | found. | |
1d67c29b | 1333 | |
3263d5a2 KH |
1334 | 2: Otherwise. |
1335 | */ | |
4ed46869 KH |
1336 | |
1337 | int | |
3263d5a2 KH |
1338 | string_xstring_p (string) |
1339 | Lisp_Object string; | |
4ed46869 | 1340 | { |
8f924df7 KH |
1341 | const unsigned char *p = SDATA (string); |
1342 | const unsigned char *endp = p + SBYTES (string); | |
3263d5a2 | 1343 | |
8f924df7 | 1344 | if (SCHARS (string) == SBYTES (string)) |
3263d5a2 KH |
1345 | return 0; |
1346 | ||
3263d5a2 | 1347 | while (p < endp) |
0282eb69 | 1348 | { |
3263d5a2 | 1349 | int c = STRING_CHAR_ADVANCE (p); |
2e344af3 | 1350 | |
3cc67a4d | 1351 | if (c >= 0x100) |
3263d5a2 | 1352 | return 2; |
0282eb69 | 1353 | } |
3263d5a2 KH |
1354 | return 1; |
1355 | } | |
05505664 | 1356 | |
05505664 | 1357 | |
3263d5a2 | 1358 | /* Find charsets in the string at PTR of NCHARS and NBYTES. |
4ed46869 | 1359 | |
3cc67a4d KH |
1360 | CHARSETS is a vector. If Nth element is non-nil, it means the |
1361 | charset whose id is N is already found. | |
2e344af3 | 1362 | |
3263d5a2 | 1363 | It may lookup a translation table TABLE if supplied. */ |
2e344af3 | 1364 | |
3263d5a2 | 1365 | static void |
3cc67a4d | 1366 | find_charsets_in_text (ptr, nchars, nbytes, charsets, table, multibyte) |
dbbb237d | 1367 | const unsigned char *ptr; |
42ca828e | 1368 | EMACS_INT nchars, nbytes; |
3263d5a2 | 1369 | Lisp_Object charsets, table; |
3cc67a4d | 1370 | int multibyte; |
3263d5a2 | 1371 | { |
dbbb237d | 1372 | const unsigned char *pend = ptr + nbytes; |
3263d5a2 KH |
1373 | |
1374 | if (nchars == nbytes) | |
3263d5a2 | 1375 | { |
3cc67a4d KH |
1376 | if (multibyte) |
1377 | ASET (charsets, charset_ascii, Qt); | |
1378 | else | |
1379 | while (ptr < pend) | |
1380 | { | |
1381 | int c = *ptr++; | |
1382 | ||
1383 | if (!NILP (table)) | |
1384 | c = translate_char (table, c); | |
1385 | if (ASCII_BYTE_P (c)) | |
1386 | ASET (charsets, charset_ascii, Qt); | |
1387 | else | |
1388 | ASET (charsets, charset_eight_bit, Qt); | |
1389 | } | |
1390 | } | |
1391 | else | |
1392 | { | |
1393 | while (ptr < pend) | |
3263d5a2 | 1394 | { |
3cc67a4d KH |
1395 | int c = STRING_CHAR_ADVANCE (ptr); |
1396 | struct charset *charset; | |
3263d5a2 | 1397 | |
3cc67a4d KH |
1398 | if (!NILP (table)) |
1399 | c = translate_char (table, c); | |
1400 | charset = CHAR_CHARSET (c); | |
1401 | ASET (charsets, CHARSET_ID (charset), Qt); | |
4ed46869 | 1402 | } |
4ed46869 | 1403 | } |
4ed46869 KH |
1404 | } |
1405 | ||
1406 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
23d2a7f1 | 1407 | 2, 3, 0, |
fdb82f93 PJ |
1408 | doc: /* Return a list of charsets in the region between BEG and END. |
1409 | BEG and END are buffer positions. | |
1410 | Optional arg TABLE if non-nil is a translation table to look up. | |
1411 | ||
fdb82f93 PJ |
1412 | If the current buffer is unibyte, the returned list may contain |
1413 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) | |
1414 | (beg, end, table) | |
23d2a7f1 | 1415 | Lisp_Object beg, end, table; |
4ed46869 | 1416 | { |
3263d5a2 | 1417 | Lisp_Object charsets; |
42ca828e DL |
1418 | EMACS_INT from, from_byte, to, stop, stop_byte; |
1419 | int i; | |
4ed46869 | 1420 | Lisp_Object val; |
3cc67a4d | 1421 | int multibyte = ! NILP (current_buffer->enable_multibyte_characters); |
4ed46869 KH |
1422 | |
1423 | validate_region (&beg, &end); | |
1424 | from = XFASTINT (beg); | |
1425 | stop = to = XFASTINT (end); | |
6ae1f27e | 1426 | |
4ed46869 | 1427 | if (from < GPT && GPT < to) |
6ae1f27e RS |
1428 | { |
1429 | stop = GPT; | |
1430 | stop_byte = GPT_BYTE; | |
1431 | } | |
1432 | else | |
1433 | stop_byte = CHAR_TO_BYTE (stop); | |
1434 | ||
1435 | from_byte = CHAR_TO_BYTE (from); | |
1436 | ||
3263d5a2 | 1437 | charsets = Fmake_vector (make_number (charset_table_used), Qnil); |
4ed46869 KH |
1438 | while (1) |
1439 | { | |
3263d5a2 | 1440 | find_charsets_in_text (BYTE_POS_ADDR (from_byte), stop - from, |
3cc67a4d KH |
1441 | stop_byte - from_byte, charsets, table, |
1442 | multibyte); | |
4ed46869 | 1443 | if (stop < to) |
6ae1f27e RS |
1444 | { |
1445 | from = stop, from_byte = stop_byte; | |
1446 | stop = to, stop_byte = CHAR_TO_BYTE (stop); | |
1447 | } | |
4ed46869 KH |
1448 | else |
1449 | break; | |
1450 | } | |
6ae1f27e | 1451 | |
4ed46869 | 1452 | val = Qnil; |
3263d5a2 | 1453 | for (i = charset_table_used - 1; i >= 0; i--) |
3cc67a4d | 1454 | if (!NILP (AREF (charsets, i))) |
3263d5a2 | 1455 | val = Fcons (CHARSET_NAME (charset_table + i), val); |
4ed46869 KH |
1456 | return val; |
1457 | } | |
1458 | ||
1459 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
23d2a7f1 | 1460 | 1, 2, 0, |
fdb82f93 PJ |
1461 | doc: /* Return a list of charsets in STR. |
1462 | Optional arg TABLE if non-nil is a translation table to look up. | |
1463 | ||
fdb82f93 | 1464 | If STR is unibyte, the returned list may contain |
3263d5a2 | 1465 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) |
fdb82f93 | 1466 | (str, table) |
23d2a7f1 | 1467 | Lisp_Object str, table; |
4ed46869 | 1468 | { |
3263d5a2 | 1469 | Lisp_Object charsets; |
4ed46869 KH |
1470 | int i; |
1471 | Lisp_Object val; | |
1472 | ||
b7826503 | 1473 | CHECK_STRING (str); |
87b089ad | 1474 | |
3263d5a2 | 1475 | charsets = Fmake_vector (make_number (charset_table_used), Qnil); |
8f924df7 | 1476 | find_charsets_in_text (SDATA (str), SCHARS (str), SBYTES (str), |
3cc67a4d KH |
1477 | charsets, table, |
1478 | STRING_MULTIBYTE (str)); | |
4ed46869 | 1479 | val = Qnil; |
3263d5a2 | 1480 | for (i = charset_table_used - 1; i >= 0; i--) |
3cc67a4d | 1481 | if (!NILP (AREF (charsets, i))) |
3263d5a2 | 1482 | val = Fcons (CHARSET_NAME (charset_table + i), val); |
4ed46869 KH |
1483 | return val; |
1484 | } | |
2e344af3 | 1485 | |
4ed46869 | 1486 | \f |
3263d5a2 KH |
1487 | |
1488 | /* Return a character correponding to the code-point CODE of | |
1489 | CHARSET. */ | |
1490 | ||
1491 | int | |
1492 | decode_char (charset, code) | |
1493 | struct charset *charset; | |
1494 | unsigned code; | |
4ed46869 | 1495 | { |
3263d5a2 KH |
1496 | int c, char_index; |
1497 | enum charset_method method = CHARSET_METHOD (charset); | |
ac4137cc | 1498 | |
3263d5a2 KH |
1499 | if (code < CHARSET_MIN_CODE (charset) || code > CHARSET_MAX_CODE (charset)) |
1500 | return -1; | |
4ed46869 | 1501 | |
3263d5a2 | 1502 | if (method == CHARSET_METHOD_MAP_DEFERRED) |
ac4137cc | 1503 | { |
3263d5a2 KH |
1504 | load_charset (charset); |
1505 | method = CHARSET_METHOD (charset); | |
ac4137cc | 1506 | } |
4ed46869 | 1507 | |
374c5cfd | 1508 | if (method == CHARSET_METHOD_SUBSET) |
2e344af3 | 1509 | { |
374c5cfd KH |
1510 | Lisp_Object subset_info; |
1511 | ||
1512 | subset_info = CHARSET_SUBSET (charset); | |
1513 | charset = CHARSET_FROM_ID (XFASTINT (AREF (subset_info, 0))); | |
1514 | code -= XINT (AREF (subset_info, 3)); | |
1515 | if (code < XFASTINT (AREF (subset_info, 1)) | |
1516 | || code > XFASTINT (AREF (subset_info, 2))) | |
1517 | c = -1; | |
1518 | else | |
1519 | c = DECODE_CHAR (charset, code); | |
2e344af3 | 1520 | } |
374c5cfd | 1521 | else if (method == CHARSET_METHOD_SUPERSET) |
2e344af3 | 1522 | { |
3263d5a2 | 1523 | Lisp_Object parents; |
4ed46869 | 1524 | |
374c5cfd | 1525 | parents = CHARSET_SUPERSET (charset); |
3263d5a2 KH |
1526 | c = -1; |
1527 | for (; CONSP (parents); parents = XCDR (parents)) | |
1528 | { | |
1529 | int id = XINT (XCAR (XCAR (parents))); | |
1530 | int code_offset = XINT (XCDR (XCAR (parents))); | |
374c5cfd | 1531 | unsigned this_code = code - code_offset; |
4ed46869 | 1532 | |
3263d5a2 KH |
1533 | charset = CHARSET_FROM_ID (id); |
1534 | if ((c = DECODE_CHAR (charset, this_code)) >= 0) | |
1535 | break; | |
1536 | } | |
1537 | } | |
1538 | else | |
ac4137cc | 1539 | { |
3263d5a2 | 1540 | char_index = CODE_POINT_TO_INDEX (charset, code); |
69f8de5b KH |
1541 | if (char_index < 0) |
1542 | return -1; | |
4ed46869 | 1543 | |
3263d5a2 | 1544 | if (method == CHARSET_METHOD_MAP) |
ac4137cc | 1545 | { |
3263d5a2 | 1546 | Lisp_Object decoder; |
4ed46869 | 1547 | |
3263d5a2 KH |
1548 | decoder = CHARSET_DECODER (charset); |
1549 | if (! VECTORP (decoder)) | |
1550 | return -1; | |
1551 | c = XINT (AREF (decoder, char_index)); | |
ac4137cc KH |
1552 | } |
1553 | else | |
1554 | { | |
3263d5a2 | 1555 | c = char_index + CHARSET_CODE_OFFSET (charset); |
ac4137cc KH |
1556 | } |
1557 | } | |
4ed46869 | 1558 | |
3263d5a2 KH |
1559 | if (CHARSET_UNIFIED_P (charset) |
1560 | && c >= 0) | |
c449997d KH |
1561 | { |
1562 | MAYBE_UNIFY_CHAR (c); | |
1563 | } | |
ac4137cc | 1564 | |
3263d5a2 | 1565 | return c; |
90d7b74e KH |
1566 | } |
1567 | ||
374c5cfd KH |
1568 | /* Variable used temporarily by the macro ENCODE_CHAR. */ |
1569 | Lisp_Object charset_work; | |
4ed46869 | 1570 | |
3263d5a2 | 1571 | /* Return a code-point of CHAR in CHARSET. If CHAR doesn't belong to |
28c026cd DL |
1572 | CHARSET, return CHARSET_INVALID_CODE (CHARSET). If STRICT is true, |
1573 | use CHARSET's strict_max_char instead of max_char. */ | |
4ed46869 | 1574 | |
3263d5a2 KH |
1575 | unsigned |
1576 | encode_char (charset, c) | |
1577 | struct charset *charset; | |
9b6a601f | 1578 | int c; |
9d3d8cba | 1579 | { |
3263d5a2 KH |
1580 | unsigned code; |
1581 | enum charset_method method = CHARSET_METHOD (charset); | |
9d3d8cba | 1582 | |
3263d5a2 | 1583 | if (CHARSET_UNIFIED_P (charset)) |
ac4137cc | 1584 | { |
374c5cfd | 1585 | Lisp_Object deunifier, deunified; |
4ed46869 | 1586 | |
3263d5a2 KH |
1587 | deunifier = CHARSET_DEUNIFIER (charset); |
1588 | if (! CHAR_TABLE_P (deunifier)) | |
ac4137cc | 1589 | { |
dbbb237d | 1590 | Funify_charset (CHARSET_NAME (charset), Qnil, Qnil); |
3263d5a2 | 1591 | deunifier = CHARSET_DEUNIFIER (charset); |
ac4137cc | 1592 | } |
374c5cfd KH |
1593 | deunified = CHAR_TABLE_REF (deunifier, c); |
1594 | if (! NILP (deunified)) | |
1595 | c = XINT (deunified); | |
ac4137cc | 1596 | } |
9d3d8cba | 1597 | |
374c5cfd KH |
1598 | if (method == CHARSET_METHOD_SUBSET) |
1599 | { | |
1600 | Lisp_Object subset_info; | |
1601 | struct charset *this_charset; | |
1602 | ||
1603 | subset_info = CHARSET_SUBSET (charset); | |
1604 | this_charset = CHARSET_FROM_ID (XFASTINT (AREF (subset_info, 0))); | |
1605 | code = ENCODE_CHAR (this_charset, c); | |
1606 | if (code == CHARSET_INVALID_CODE (this_charset) | |
1607 | || code < XFASTINT (AREF (subset_info, 1)) | |
1608 | || code > XFASTINT (AREF (subset_info, 2))) | |
1609 | return CHARSET_INVALID_CODE (charset); | |
1610 | code += XINT (AREF (subset_info, 3)); | |
1611 | return code; | |
1612 | } | |
9d3d8cba | 1613 | |
374c5cfd | 1614 | if (method == CHARSET_METHOD_SUPERSET) |
859f2b3c | 1615 | { |
3263d5a2 | 1616 | Lisp_Object parents; |
d2665018 | 1617 | |
374c5cfd | 1618 | parents = CHARSET_SUPERSET (charset); |
3263d5a2 | 1619 | for (; CONSP (parents); parents = XCDR (parents)) |
beeedaad | 1620 | { |
3263d5a2 KH |
1621 | int id = XINT (XCAR (XCAR (parents))); |
1622 | int code_offset = XINT (XCDR (XCAR (parents))); | |
1623 | struct charset *this_charset = CHARSET_FROM_ID (id); | |
d2665018 | 1624 | |
3263d5a2 | 1625 | code = ENCODE_CHAR (this_charset, c); |
dbbb237d KH |
1626 | if (code != CHARSET_INVALID_CODE (this_charset)) |
1627 | return code + code_offset; | |
beeedaad | 1628 | } |
3263d5a2 KH |
1629 | return CHARSET_INVALID_CODE (charset); |
1630 | } | |
1bcc1567 | 1631 | |
15c85a88 KH |
1632 | if (! CHARSET_FAST_MAP_REF ((c), charset->fast_map) |
1633 | || c < CHARSET_MIN_CHAR (charset) || c > CHARSET_MAX_CHAR (charset)) | |
1634 | return CHARSET_INVALID_CODE (charset); | |
1bcc1567 | 1635 | |
3263d5a2 | 1636 | if (method == CHARSET_METHOD_MAP_DEFERRED) |
beeedaad | 1637 | { |
3263d5a2 KH |
1638 | load_charset (charset); |
1639 | method = CHARSET_METHOD (charset); | |
859f2b3c | 1640 | } |
9b6a601f | 1641 | |
3263d5a2 | 1642 | if (method == CHARSET_METHOD_MAP) |
3f62427c | 1643 | { |
3263d5a2 | 1644 | Lisp_Object encoder; |
beeedaad | 1645 | Lisp_Object val; |
9b6a601f | 1646 | |
3263d5a2 KH |
1647 | encoder = CHARSET_ENCODER (charset); |
1648 | if (! CHAR_TABLE_P (CHARSET_ENCODER (charset))) | |
1649 | return CHARSET_INVALID_CODE (charset); | |
1650 | val = CHAR_TABLE_REF (encoder, c); | |
374c5cfd KH |
1651 | if (NILP (val)) |
1652 | return CHARSET_INVALID_CODE (charset); | |
e9ce014c KH |
1653 | code = XINT (val); |
1654 | if (! CHARSET_COMPACT_CODES_P (charset)) | |
1655 | code = INDEX_TO_CODE_POINT (charset, code); | |
3263d5a2 | 1656 | } |
820ee249 | 1657 | else /* method == CHARSET_METHOD_OFFSET */ |
beeedaad | 1658 | { |
3263d5a2 KH |
1659 | code = c - CHARSET_CODE_OFFSET (charset); |
1660 | code = INDEX_TO_CODE_POINT (charset, code); | |
3f62427c | 1661 | } |
8ac5a9cc | 1662 | |
3263d5a2 | 1663 | return code; |
4ed46869 KH |
1664 | } |
1665 | ||
4ed46869 | 1666 | |
3263d5a2 KH |
1667 | DEFUN ("decode-char", Fdecode_char, Sdecode_char, 2, 3, 0, |
1668 | doc: /* Decode the pair of CHARSET and CODE-POINT into a character. | |
1669 | Return nil if CODE-POINT is not valid in CHARSET. | |
4ed46869 | 1670 | |
3263d5a2 | 1671 | CODE-POINT may be a cons (HIGHER-16-BIT-VALUE . LOWER-16-BIT-VALUE). |
859f2b3c | 1672 | |
3263d5a2 | 1673 | Optional argument RESTRICTION specifies a way to map the pair of CCS |
d0cf2d48 | 1674 | and CODE-POINT to a character. Currently not supported and just ignored. */) |
3263d5a2 KH |
1675 | (charset, code_point, restriction) |
1676 | Lisp_Object charset, code_point, restriction; | |
4ed46869 | 1677 | { |
3263d5a2 KH |
1678 | int c, id; |
1679 | unsigned code; | |
1680 | struct charset *charsetp; | |
859f2b3c | 1681 | |
3263d5a2 KH |
1682 | CHECK_CHARSET_GET_ID (charset, id); |
1683 | if (CONSP (code_point)) | |
1684 | { | |
8f924df7 KH |
1685 | CHECK_NATNUM_CAR (code_point); |
1686 | CHECK_NATNUM_CDR (code_point); | |
69f8de5b | 1687 | code = (XINT (XCAR (code_point)) << 16) | (XINT (XCDR (code_point))); |
3263d5a2 | 1688 | } |
4ed46869 KH |
1689 | else |
1690 | { | |
3263d5a2 KH |
1691 | CHECK_NATNUM (code_point); |
1692 | code = XINT (code_point); | |
4ed46869 | 1693 | } |
3263d5a2 KH |
1694 | charsetp = CHARSET_FROM_ID (id); |
1695 | c = DECODE_CHAR (charsetp, code); | |
1696 | return (c >= 0 ? make_number (c) : Qnil); | |
4ed46869 KH |
1697 | } |
1698 | ||
859f2b3c | 1699 | |
3263d5a2 KH |
1700 | DEFUN ("encode-char", Fencode_char, Sencode_char, 2, 3, 0, |
1701 | doc: /* Encode the character CH into a code-point of CHARSET. | |
1702 | Return nil if CHARSET doesn't include CH. | |
17e7ef1b | 1703 | |
d0cf2d48 | 1704 | Optional argument RESTRICTION specifies a way to map CH to a |
3263d5a2 KH |
1705 | code-point in CCS. Currently not supported and just ignored. */) |
1706 | (ch, charset, restriction) | |
1707 | Lisp_Object ch, charset, restriction; | |
4ed46869 | 1708 | { |
16fed1fc | 1709 | int id; |
3263d5a2 KH |
1710 | unsigned code; |
1711 | struct charset *charsetp; | |
046b1f03 | 1712 | |
3263d5a2 KH |
1713 | CHECK_CHARSET_GET_ID (charset, id); |
1714 | CHECK_NATNUM (ch); | |
3263d5a2 | 1715 | charsetp = CHARSET_FROM_ID (id); |
16fed1fc | 1716 | code = ENCODE_CHAR (charsetp, XINT (ch)); |
3263d5a2 KH |
1717 | if (code == CHARSET_INVALID_CODE (charsetp)) |
1718 | return Qnil; | |
1719 | if (code > 0x7FFFFFF) | |
1720 | return Fcons (make_number (code >> 16), make_number (code & 0xFFFF)); | |
1721 | return make_number (code); | |
beeedaad KH |
1722 | } |
1723 | ||
beeedaad | 1724 | |
b121a744 KH |
1725 | DEFUN ("make-char", Fmake_char, Smake_char, 1, 5, 0, |
1726 | doc: | |
1727 | /* Return a character of CHARSET whose position codes are CODEn. | |
1728 | ||
1729 | CODE1 through CODE4 are optional, but if you don't supply sufficient | |
1730 | position codes, it is assumed that the minimum code in each dimension | |
04c2f2c5 | 1731 | is specified. */) |
b121a744 KH |
1732 | (charset, code1, code2, code3, code4) |
1733 | Lisp_Object charset, code1, code2, code3, code4; | |
beeedaad | 1734 | { |
3263d5a2 KH |
1735 | int id, dimension; |
1736 | struct charset *charsetp; | |
b121a744 KH |
1737 | unsigned code; |
1738 | int c; | |
87b089ad | 1739 | |
3263d5a2 KH |
1740 | CHECK_CHARSET_GET_ID (charset, id); |
1741 | charsetp = CHARSET_FROM_ID (id); | |
4ed46869 | 1742 | |
b121a744 KH |
1743 | dimension = CHARSET_DIMENSION (charsetp); |
1744 | if (NILP (code1)) | |
d47073ca KH |
1745 | code = (CHARSET_ASCII_COMPATIBLE_P (charsetp) |
1746 | ? 0 : CHARSET_MIN_CODE (charsetp)); | |
3263d5a2 | 1747 | else |
859f2b3c | 1748 | { |
b121a744 KH |
1749 | CHECK_NATNUM (code1); |
1750 | if (XFASTINT (code1) >= 0x100) | |
1751 | args_out_of_range (make_number (0xFF), code1); | |
1752 | code = XFASTINT (code1); | |
859f2b3c | 1753 | |
b0a1e45e | 1754 | if (dimension > 1) |
beeedaad | 1755 | { |
b121a744 | 1756 | code <<= 8; |
b0a1e45e KH |
1757 | if (NILP (code2)) |
1758 | code |= charsetp->code_space[(dimension - 2) * 4]; | |
beeedaad | 1759 | else |
b121a744 | 1760 | { |
b0a1e45e KH |
1761 | CHECK_NATNUM (code2); |
1762 | if (XFASTINT (code2) >= 0x100) | |
1763 | args_out_of_range (make_number (0xFF), code2); | |
1764 | code |= XFASTINT (code2); | |
b121a744 | 1765 | } |
99529c2c | 1766 | |
b0a1e45e | 1767 | if (dimension > 2) |
b121a744 KH |
1768 | { |
1769 | code <<= 8; | |
b0a1e45e KH |
1770 | if (NILP (code3)) |
1771 | code |= charsetp->code_space[(dimension - 3) * 4]; | |
b121a744 KH |
1772 | else |
1773 | { | |
b0a1e45e KH |
1774 | CHECK_NATNUM (code3); |
1775 | if (XFASTINT (code3) >= 0x100) | |
1776 | args_out_of_range (make_number (0xFF), code3); | |
1777 | code |= XFASTINT (code3); | |
1778 | } | |
1779 | ||
1780 | if (dimension > 3) | |
1781 | { | |
1782 | code <<= 8; | |
1783 | if (NILP (code4)) | |
1784 | code |= charsetp->code_space[0]; | |
1785 | else | |
1786 | { | |
1787 | CHECK_NATNUM (code4); | |
1788 | if (XFASTINT (code4) >= 0x100) | |
1789 | args_out_of_range (make_number (0xFF), code4); | |
1790 | code |= XFASTINT (code4); | |
1791 | } | |
b121a744 KH |
1792 | } |
1793 | } | |
beeedaad | 1794 | } |
859f2b3c | 1795 | } |
beeedaad | 1796 | |
b121a744 KH |
1797 | if (CHARSET_ISO_FINAL (charsetp) >= 0) |
1798 | code &= 0x7F7F7F7F; | |
1799 | c = DECODE_CHAR (charsetp, code); | |
1800 | if (c < 0) | |
1801 | error ("Invalid code(s)"); | |
3263d5a2 | 1802 | return make_number (c); |
4ed46869 KH |
1803 | } |
1804 | ||
beeedaad | 1805 | |
3263d5a2 KH |
1806 | /* Return the first charset in CHARSET_LIST that contains C. |
1807 | CHARSET_LIST is a list of charset IDs. If it is nil, use | |
1808 | Vcharset_ordered_list. */ | |
beeedaad | 1809 | |
3263d5a2 KH |
1810 | struct charset * |
1811 | char_charset (c, charset_list, code_return) | |
1812 | int c; | |
1813 | Lisp_Object charset_list; | |
1814 | unsigned *code_return; | |
2e344af3 | 1815 | { |
8a1816bb KH |
1816 | int maybe_null = 0; |
1817 | ||
3263d5a2 KH |
1818 | if (NILP (charset_list)) |
1819 | charset_list = Vcharset_ordered_list; | |
8a1816bb KH |
1820 | else |
1821 | maybe_null = 1; | |
beeedaad | 1822 | |
6c652beb | 1823 | while (CONSP (charset_list)) |
2e344af3 | 1824 | { |
3263d5a2 KH |
1825 | struct charset *charset = CHARSET_FROM_ID (XINT (XCAR (charset_list))); |
1826 | unsigned code = ENCODE_CHAR (charset, c); | |
beeedaad | 1827 | |
3263d5a2 | 1828 | if (code != CHARSET_INVALID_CODE (charset)) |
beeedaad | 1829 | { |
3263d5a2 KH |
1830 | if (code_return) |
1831 | *code_return = code; | |
1832 | return charset; | |
3f62427c | 1833 | } |
3263d5a2 | 1834 | charset_list = XCDR (charset_list); |
6c652beb KH |
1835 | if (c <= MAX_UNICODE_CHAR |
1836 | && EQ (charset_list, Vcharset_non_preferred_head)) | |
1837 | return CHARSET_FROM_ID (charset_unicode); | |
3f62427c | 1838 | } |
8a1816bb KH |
1839 | return (maybe_null ? NULL |
1840 | : c <= MAX_5_BYTE_CHAR ? CHARSET_FROM_ID (charset_emacs) | |
6a9c90ec | 1841 | : CHARSET_FROM_ID (charset_eight_bit)); |
3f62427c KH |
1842 | } |
1843 | ||
2e344af3 | 1844 | |
3263d5a2 | 1845 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, |
3cc67a4d | 1846 | doc: |
d0cf2d48 | 1847 | /*Return list of charset and one to four position-codes of CH. |
3cc67a4d KH |
1848 | The charset is decided by the current priority order of charsets. |
1849 | A position-code is a byte value of each dimension of the code-point of | |
d0cf2d48 | 1850 | CH in the charset. */) |
3263d5a2 KH |
1851 | (ch) |
1852 | Lisp_Object ch; | |
4ed46869 | 1853 | { |
3263d5a2 KH |
1854 | struct charset *charset; |
1855 | int c, dimension; | |
1856 | unsigned code; | |
4ed46869 KH |
1857 | Lisp_Object val; |
1858 | ||
3263d5a2 KH |
1859 | CHECK_CHARACTER (ch); |
1860 | c = XFASTINT (ch); | |
1861 | charset = CHAR_CHARSET (c); | |
1862 | if (! charset) | |
3cc67a4d | 1863 | abort (); |
3263d5a2 KH |
1864 | code = ENCODE_CHAR (charset, c); |
1865 | if (code == CHARSET_INVALID_CODE (charset)) | |
1866 | abort (); | |
1867 | dimension = CHARSET_DIMENSION (charset); | |
3cc67a4d KH |
1868 | for (val = Qnil; dimension > 0; dimension--) |
1869 | { | |
1870 | val = Fcons (make_number (code & 0xFF), val); | |
1871 | code >>= 8; | |
1872 | } | |
3263d5a2 | 1873 | return Fcons (CHARSET_NAME (charset), val); |
4ed46869 KH |
1874 | } |
1875 | ||
740f080d | 1876 | |
3263d5a2 | 1877 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, |
327719ee | 1878 | doc: /* Return the charset of highest priority that contains CH. */) |
fdb82f93 | 1879 | (ch) |
4ed46869 KH |
1880 | Lisp_Object ch; |
1881 | { | |
3263d5a2 | 1882 | struct charset *charset; |
4ed46869 | 1883 | |
3263d5a2 KH |
1884 | CHECK_CHARACTER (ch); |
1885 | charset = CHAR_CHARSET (XINT (ch)); | |
1886 | return (CHARSET_NAME (charset)); | |
4ed46869 KH |
1887 | } |
1888 | ||
17e7ef1b | 1889 | |
3263d5a2 KH |
1890 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
1891 | doc: /* | |
1892 | Return charset of a character in the current buffer at position POS. | |
1893 | If POS is nil, it defauls to the current point. | |
1894 | If POS is out of range, the value is nil. */) | |
1895 | (pos) | |
1896 | Lisp_Object pos; | |
2e344af3 | 1897 | { |
3263d5a2 KH |
1898 | Lisp_Object ch; |
1899 | struct charset *charset; | |
046b1f03 | 1900 | |
3263d5a2 KH |
1901 | ch = Fchar_after (pos); |
1902 | if (! INTEGERP (ch)) | |
1903 | return ch; | |
1904 | charset = CHAR_CHARSET (XINT (ch)); | |
1905 | return (CHARSET_NAME (charset)); | |
6ae1f27e | 1906 | } |
9036eb45 | 1907 | |
87b089ad | 1908 | |
3263d5a2 KH |
1909 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
1910 | doc: /* | |
1911 | Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR. | |
1912 | ||
1913 | ISO 2022's designation sequence (escape sequence) distinguishes charsets | |
1914 | by their DIMENSION, CHARS, and FINAL-CHAR, | |
d0cf2d48 | 1915 | whereas Emacs distinguishes them by charset symbol. |
3263d5a2 KH |
1916 | See the documentation of the function `charset-info' for the meanings of |
1917 | DIMENSION, CHARS, and FINAL-CHAR. */) | |
1918 | (dimension, chars, final_char) | |
1919 | Lisp_Object dimension, chars, final_char; | |
6ae1f27e | 1920 | { |
3263d5a2 | 1921 | int id; |
82215ce9 | 1922 | int chars_flag; |
a8a35e61 | 1923 | |
3263d5a2 | 1924 | check_iso_charset_parameter (dimension, chars, final_char); |
82215ce9 KH |
1925 | chars_flag = XFASTINT (chars) == 96; |
1926 | id = ISO_CHARSET_TABLE (XFASTINT (dimension), chars_flag, | |
3263d5a2 KH |
1927 | XFASTINT (final_char)); |
1928 | return (id >= 0 ? CHARSET_NAME (CHARSET_FROM_ID (id)) : Qnil); | |
046b1f03 RS |
1929 | } |
1930 | ||
87b089ad | 1931 | |
3263d5a2 KH |
1932 | DEFUN ("clear-charset-maps", Fclear_charset_maps, Sclear_charset_maps, |
1933 | 0, 0, 0, | |
1934 | doc: /* | |
1935 | Clear encoder and decoder of charsets that are loaded from mapfiles. */) | |
1936 | () | |
87b089ad | 1937 | { |
53316e55 | 1938 | int i; |
3263d5a2 KH |
1939 | struct charset *charset; |
1940 | Lisp_Object attrs; | |
87b089ad | 1941 | |
3263d5a2 | 1942 | for (i = 0; i < charset_table_used; i++) |
87b089ad | 1943 | { |
3263d5a2 KH |
1944 | charset = CHARSET_FROM_ID (i); |
1945 | attrs = CHARSET_ATTRIBUTES (charset); | |
2e344af3 | 1946 | |
3263d5a2 KH |
1947 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP) |
1948 | { | |
1949 | CHARSET_ATTR_DECODER (attrs) = Qnil; | |
1950 | CHARSET_ATTR_ENCODER (attrs) = Qnil; | |
1951 | CHARSET_METHOD (charset) = CHARSET_METHOD_MAP_DEFERRED; | |
1952 | } | |
2e344af3 | 1953 | |
3263d5a2 KH |
1954 | if (CHARSET_UNIFIED_P (charset)) |
1955 | CHARSET_ATTR_DEUNIFIER (attrs) = Qnil; | |
2e344af3 | 1956 | } |
2e344af3 | 1957 | |
3263d5a2 | 1958 | if (CHAR_TABLE_P (Vchar_unified_charset_table)) |
2e344af3 | 1959 | { |
d0827857 | 1960 | Foptimize_char_table (Vchar_unified_charset_table, Qnil); |
3263d5a2 KH |
1961 | Vchar_unify_table = Vchar_unified_charset_table; |
1962 | Vchar_unified_charset_table = Qnil; | |
87b089ad | 1963 | } |
740f080d | 1964 | |
3263d5a2 | 1965 | return Qnil; |
740f080d KH |
1966 | } |
1967 | ||
8ddf5e57 DL |
1968 | DEFUN ("charset-priority-list", Fcharset_priority_list, |
1969 | Scharset_priority_list, 0, 1, 0, | |
1970 | doc: /* Return the list of charsets ordered by priority. | |
1971 | HIGHESTP non-nil means just return the highest priority one. */) | |
1972 | (highestp) | |
1973 | Lisp_Object highestp; | |
2e344af3 | 1974 | { |
8ddf5e57 | 1975 | Lisp_Object val = Qnil, list = Vcharset_ordered_list; |
2e344af3 | 1976 | |
8ddf5e57 | 1977 | if (!NILP (highestp)) |
16fed1fc | 1978 | return CHARSET_NAME (CHARSET_FROM_ID (XINT (Fcar (list)))); |
2e344af3 | 1979 | |
8ddf5e57 | 1980 | while (!NILP (list)) |
2e344af3 | 1981 | { |
16fed1fc | 1982 | val = Fcons (CHARSET_NAME (CHARSET_FROM_ID (XINT (XCAR (list)))), val); |
8ddf5e57 | 1983 | list = XCDR (list); |
2e344af3 | 1984 | } |
8ddf5e57 | 1985 | return Fnreverse (val); |
2e344af3 KH |
1986 | } |
1987 | ||
8ddf5e57 DL |
1988 | DEFUN ("set-charset-priority", Fset_charset_priority, Sset_charset_priority, |
1989 | 1, MANY, 0, | |
1990 | doc: /* Assign higher priority to the charsets given as arguments. | |
1991 | usage: (set-charset-priority &rest charsets) */) | |
1992 | (nargs, args) | |
1993 | int nargs; | |
4ed46869 KH |
1994 | Lisp_Object *args; |
1995 | { | |
af7c60ca | 1996 | Lisp_Object new_head, old_list, arglist[2]; |
321c819c | 1997 | Lisp_Object list_2022, list_emacs_mule; |
16fed1fc | 1998 | int i, id; |
4ed46869 | 1999 | |
8ddf5e57 | 2000 | old_list = Fcopy_sequence (Vcharset_ordered_list); |
af7c60ca | 2001 | new_head = Qnil; |
8ddf5e57 | 2002 | for (i = 0; i < nargs; i++) |
4ed46869 | 2003 | { |
8ddf5e57 | 2004 | CHECK_CHARSET_GET_ID (args[i], id); |
af7c60ca KH |
2005 | if (! NILP (Fmemq (make_number (id), old_list))) |
2006 | { | |
2007 | old_list = Fdelq (make_number (id), old_list); | |
2008 | new_head = Fcons (make_number (id), new_head); | |
2009 | } | |
5729c92f | 2010 | } |
8ddf5e57 | 2011 | arglist[0] = Fnreverse (new_head); |
6a9c90ec | 2012 | arglist[1] = Vcharset_non_preferred_head = old_list; |
8ddf5e57 | 2013 | Vcharset_ordered_list = Fnconc (2, arglist); |
dbbb237d | 2014 | charset_ordered_list_tick++; |
5729c92f | 2015 | |
321c819c | 2016 | for (old_list = Vcharset_ordered_list, list_2022 = list_emacs_mule = Qnil; |
d017b41e | 2017 | CONSP (old_list); old_list = XCDR (old_list)) |
5729c92f | 2018 | { |
e77415b0 | 2019 | if (! NILP (Fmemq (XCAR (old_list), Viso_2022_charset_list))) |
321c819c KH |
2020 | list_2022 = Fcons (XCAR (old_list), list_2022); |
2021 | if (! NILP (Fmemq (XCAR (old_list), Vemacs_mule_charset_list))) | |
2022 | list_emacs_mule = Fcons (XCAR (old_list), list_emacs_mule); | |
4ed46869 | 2023 | } |
321c819c KH |
2024 | Viso_2022_charset_list = Fnreverse (list_2022); |
2025 | Vemacs_mule_charset_list = Fnreverse (list_emacs_mule); | |
4ed46869 | 2026 | |
8ddf5e57 | 2027 | return Qnil; |
4ed46869 KH |
2028 | } |
2029 | ||
d5b33309 KH |
2030 | DEFUN ("charset-id-internal", Fcharset_id_internal, Scharset_id_internal, |
2031 | 0, 1, 0, | |
2032 | doc: /* Internal use only. | |
2033 | Return charset identification number of CHARSET. */) | |
2034 | (charset) | |
2035 | Lisp_Object charset; | |
4ed46869 | 2036 | { |
d5b33309 | 2037 | int id; |
4ed46869 | 2038 | |
d5b33309 KH |
2039 | CHECK_CHARSET_GET_ID (charset, id); |
2040 | return make_number (id); | |
4ed46869 KH |
2041 | } |
2042 | ||
4ed46869 | 2043 | \f |
3263d5a2 KH |
2044 | void |
2045 | init_charset () | |
4ed46869 | 2046 | { |
4beef065 KH |
2047 | Vcharset_map_path |
2048 | = Fcons (Fexpand_file_name (build_string ("charsets"), Vdata_directory), | |
2049 | Qnil); | |
4ed46869 KH |
2050 | } |
2051 | ||
4ed46869 | 2052 | |
dfcf069d | 2053 | void |
4ed46869 KH |
2054 | init_charset_once () |
2055 | { | |
2056 | int i, j, k; | |
2057 | ||
3263d5a2 KH |
2058 | for (i = 0; i < ISO_MAX_DIMENSION; i++) |
2059 | for (j = 0; j < ISO_MAX_CHARS; j++) | |
2060 | for (k = 0; k < ISO_MAX_FINAL; k++) | |
2061 | iso_charset_table[i][j][k] = -1; | |
4ed46869 | 2062 | |
60383934 | 2063 | for (i = 0; i < 256; i++) |
3263d5a2 | 2064 | emacs_mule_charset[i] = NULL; |
4ed46869 | 2065 | |
7c7dceee KH |
2066 | charset_jisx0201_roman = -1; |
2067 | charset_jisx0208_1978 = -1; | |
2068 | charset_jisx0208 = -1; | |
4ed46869 KH |
2069 | |
2070 | for (i = 0; i < 128; i++) | |
14e3d523 | 2071 | unibyte_to_multibyte_table[i] = i; |
4ed46869 | 2072 | for (; i < 256; i++) |
170e4589 | 2073 | unibyte_to_multibyte_table[i] = BYTE8_TO_CHAR (i); |
4ed46869 KH |
2074 | } |
2075 | ||
2076 | #ifdef emacs | |
2077 | ||
dfcf069d | 2078 | void |
4ed46869 KH |
2079 | syms_of_charset () |
2080 | { | |
3263d5a2 KH |
2081 | DEFSYM (Qcharsetp, "charsetp"); |
2082 | ||
2083 | DEFSYM (Qascii, "ascii"); | |
2084 | DEFSYM (Qunicode, "unicode"); | |
6c652beb | 2085 | DEFSYM (Qemacs, "emacs"); |
2fe1edd1 | 2086 | DEFSYM (Qeight_bit, "eight-bit"); |
3263d5a2 KH |
2087 | DEFSYM (Qiso_8859_1, "iso-8859-1"); |
2088 | ||
2089 | DEFSYM (Qgl, "gl"); | |
2090 | DEFSYM (Qgr, "gr"); | |
2091 | ||
3263d5a2 KH |
2092 | staticpro (&Vcharset_ordered_list); |
2093 | Vcharset_ordered_list = Qnil; | |
2094 | ||
2095 | staticpro (&Viso_2022_charset_list); | |
2096 | Viso_2022_charset_list = Qnil; | |
2097 | ||
2098 | staticpro (&Vemacs_mule_charset_list); | |
2099 | Vemacs_mule_charset_list = Qnil; | |
2100 | ||
3943ed76 KH |
2101 | /* Don't staticpro them here. It's done in syms_of_fns. */ |
2102 | QCtest = intern (":test"); | |
2103 | Qeq = intern ("eq"); | |
2104 | ||
3263d5a2 | 2105 | staticpro (&Vcharset_hash_table); |
8f924df7 KH |
2106 | { |
2107 | Lisp_Object args[2]; | |
2108 | args[0] = QCtest; | |
2109 | args[1] = Qeq; | |
2110 | Vcharset_hash_table = Fmake_hash_table (2, args); | |
2111 | } | |
3263d5a2 KH |
2112 | |
2113 | charset_table_size = 128; | |
2114 | charset_table = ((struct charset *) | |
2115 | xmalloc (sizeof (struct charset) * charset_table_size)); | |
2116 | charset_table_used = 0; | |
2117 | ||
2118 | staticpro (&Vchar_unified_charset_table); | |
2119 | Vchar_unified_charset_table = Fmake_char_table (Qnil, make_number (-1)); | |
2120 | ||
2121 | defsubr (&Scharsetp); | |
2122 | defsubr (&Smap_charset_chars); | |
2123 | defsubr (&Sdefine_charset_internal); | |
2124 | defsubr (&Sdefine_charset_alias); | |
3263d5a2 KH |
2125 | defsubr (&Scharset_plist); |
2126 | defsubr (&Sset_charset_plist); | |
2127 | defsubr (&Sunify_charset); | |
3fac5a51 | 2128 | defsubr (&Sget_unused_iso_final_char); |
4ed46869 KH |
2129 | defsubr (&Sdeclare_equiv_charset); |
2130 | defsubr (&Sfind_charset_region); | |
2131 | defsubr (&Sfind_charset_string); | |
3263d5a2 KH |
2132 | defsubr (&Sdecode_char); |
2133 | defsubr (&Sencode_char); | |
4ed46869 | 2134 | defsubr (&Ssplit_char); |
3263d5a2 | 2135 | defsubr (&Smake_char); |
4ed46869 | 2136 | defsubr (&Schar_charset); |
90d7b74e | 2137 | defsubr (&Scharset_after); |
4ed46869 | 2138 | defsubr (&Siso_charset); |
3263d5a2 | 2139 | defsubr (&Sclear_charset_maps); |
8ddf5e57 DL |
2140 | defsubr (&Scharset_priority_list); |
2141 | defsubr (&Sset_charset_priority); | |
d5b33309 | 2142 | defsubr (&Scharset_id_internal); |
3263d5a2 | 2143 | |
4beef065 KH |
2144 | DEFVAR_LISP ("charset-map-path", &Vcharset_map_path, |
2145 | doc: /* *Lisp of directories to search for charset map files. */); | |
2146 | Vcharset_map_path = Qnil; | |
4ed46869 KH |
2147 | |
2148 | DEFVAR_LISP ("charset-list", &Vcharset_list, | |
528623a0 | 2149 | doc: /* List of all charsets ever defined. */); |
3263d5a2 KH |
2150 | Vcharset_list = Qnil; |
2151 | ||
6a9c90ec KH |
2152 | DEFVAR_LISP ("current-iso639-language", &Vcurrent_iso639_language, |
2153 | doc: /* ISO639 language mnemonic symbol for the current language environment. | |
2154 | If the current language environment is for multiple languages (e.g. "Latin-1"), | |
2155 | the value may be a list of mnemonics. */); | |
2156 | Vcurrent_iso639_language = Qnil; | |
2157 | ||
2fe1edd1 KH |
2158 | charset_ascii |
2159 | = define_charset_internal (Qascii, 1, "\x00\x7F\x00\x00\x00\x00", | |
2160 | 0, 127, 'B', -1, 0, 1, 0, 0); | |
14e3d523 KH |
2161 | charset_iso_8859_1 |
2162 | = define_charset_internal (Qiso_8859_1, 1, "\x00\xFF\x00\x00\x00\x00", | |
2163 | 0, 255, -1, -1, -1, 1, 0, 0); | |
2fe1edd1 | 2164 | charset_unicode |
73fbf2d9 | 2165 | = define_charset_internal (Qunicode, 3, "\x00\xFF\x00\xFF\x00\x10", |
2fe1edd1 | 2166 | 0, MAX_UNICODE_CHAR, -1, 0, -1, 1, 0, 0); |
6c652beb KH |
2167 | charset_emacs |
2168 | = define_charset_internal (Qemacs, 3, "\x00\xFF\x00\xFF\x00\x3F", | |
2169 | 0, MAX_5_BYTE_CHAR, -1, 0, -1, 1, 1, 0); | |
2fe1edd1 KH |
2170 | charset_eight_bit |
2171 | = define_charset_internal (Qeight_bit, 1, "\x80\xFF\x00\x00\x00\x00", | |
880820fe | 2172 | 128, 255, -1, 0, -1, 0, 1, |
2fe1edd1 | 2173 | MAX_5_BYTE_CHAR + 1); |
4ed46869 KH |
2174 | } |
2175 | ||
2176 | #endif /* emacs */ | |
cefd8c4f KH |
2177 | |
2178 | /* arch-tag: 66a89b8d-4c28-47d3-9ca1-56f78440d69f | |
2179 | (do not change this comment) */ |