| 1 | /* Basic character set support. |
| 2 | Copyright (C) 1995, 97, 98, 2000, 2001 Electrotechnical Laboratory, JAPAN. |
| 3 | Licensed to the Free Software Foundation. |
| 4 | Copyright (C) 2001 Free Software Foundation, Inc. |
| 5 | Copyright (C) 2001, 2002 |
| 6 | National Institute of Advanced Industrial Science and Technology (AIST) |
| 7 | Registration Number H13PRO009 |
| 8 | |
| 9 | This file is part of GNU Emacs. |
| 10 | |
| 11 | GNU Emacs is free software; you can redistribute it and/or modify |
| 12 | it under the terms of the GNU General Public License as published by |
| 13 | the Free Software Foundation; either version 2, or (at your option) |
| 14 | any later version. |
| 15 | |
| 16 | GNU Emacs is distributed in the hope that it will be useful, |
| 17 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 18 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 19 | GNU General Public License for more details. |
| 20 | |
| 21 | You should have received a copy of the GNU General Public License |
| 22 | along with GNU Emacs; see the file COPYING. If not, write to |
| 23 | the Free Software Foundation, Inc., 59 Temple Place - Suite 330, |
| 24 | Boston, MA 02111-1307, USA. */ |
| 25 | |
| 26 | #include <config.h> |
| 27 | |
| 28 | #include <stdio.h> |
| 29 | #include <unistd.h> |
| 30 | #include <ctype.h> |
| 31 | #include <sys/types.h> |
| 32 | #include "lisp.h" |
| 33 | #include "character.h" |
| 34 | #include "charset.h" |
| 35 | #include "coding.h" |
| 36 | #include "disptab.h" |
| 37 | #include "buffer.h" |
| 38 | |
| 39 | /*** GENERAL NOTES on CODED CHARACTER SETS (CHARSETS) *** |
| 40 | |
| 41 | A coded character set ("charset" hereafter) is a meaningful |
| 42 | collection (i.e. language, culture, functionality, etc.) of |
| 43 | characters. Emacs handles multiple charsets at once. In Emacs Lisp |
| 44 | code, a charset is represented by a symbol. In C code, a charset is |
| 45 | represented by its ID number or by a pointer to a struct charset. |
| 46 | |
| 47 | The actual information about each charset is stored in two places. |
| 48 | Lispy information is stored in the hash table Vcharset_hash_table as |
| 49 | a vector (charset attributes). The other information is stored in |
| 50 | charset_table as a struct charset. |
| 51 | |
| 52 | */ |
| 53 | |
| 54 | /* List of all charsets. This variable is used only from Emacs |
| 55 | Lisp. */ |
| 56 | Lisp_Object Vcharset_list; |
| 57 | |
| 58 | /* Hash table that contains attributes of each charset. Keys are |
| 59 | charset symbols, and values are vectors of charset attributes. */ |
| 60 | Lisp_Object Vcharset_hash_table; |
| 61 | |
| 62 | /* Table of struct charset. */ |
| 63 | struct charset *charset_table; |
| 64 | |
| 65 | static int charset_table_size; |
| 66 | int charset_table_used; |
| 67 | |
| 68 | Lisp_Object Qcharsetp; |
| 69 | |
| 70 | /* Special charset symbols. */ |
| 71 | Lisp_Object Qascii; |
| 72 | Lisp_Object Qeight_bit; |
| 73 | Lisp_Object Qiso_8859_1; |
| 74 | Lisp_Object Qunicode; |
| 75 | |
| 76 | /* The corresponding charsets. */ |
| 77 | int charset_ascii; |
| 78 | int charset_eight_bit; |
| 79 | int charset_iso_8859_1; |
| 80 | int charset_unicode; |
| 81 | |
| 82 | /* The other special charsets. */ |
| 83 | int charset_jisx0201_roman; |
| 84 | int charset_jisx0208_1978; |
| 85 | int charset_jisx0208; |
| 86 | |
| 87 | /* Value of charset attribute `charset-iso-plane'. */ |
| 88 | Lisp_Object Qgl, Qgr; |
| 89 | |
| 90 | /* The primary charset. It is a charset of unibyte characters. */ |
| 91 | int charset_primary; |
| 92 | |
| 93 | /* List of charsets ordered by the priority. */ |
| 94 | Lisp_Object Vcharset_ordered_list; |
| 95 | |
| 96 | /* Incremented everytime we change Vcharset_ordered_list. This is |
| 97 | unsigned short so that it fits in Lisp_Int and never matches |
| 98 | -1. */ |
| 99 | unsigned short charset_ordered_list_tick; |
| 100 | |
| 101 | /* List of iso-2022 charsets. */ |
| 102 | Lisp_Object Viso_2022_charset_list; |
| 103 | |
| 104 | /* List of emacs-mule charsets. */ |
| 105 | Lisp_Object Vemacs_mule_charset_list; |
| 106 | |
| 107 | struct charset *emacs_mule_charset[256]; |
| 108 | |
| 109 | /* Mapping table from ISO2022's charset (specified by DIMENSION, |
| 110 | CHARS, and FINAL-CHAR) to Emacs' charset. */ |
| 111 | int iso_charset_table[ISO_MAX_DIMENSION][ISO_MAX_CHARS][ISO_MAX_FINAL]; |
| 112 | |
| 113 | Lisp_Object Vcharset_map_directory; |
| 114 | |
| 115 | Lisp_Object Vchar_unified_charset_table; |
| 116 | |
| 117 | /* Defined in chartab.c */ |
| 118 | extern void |
| 119 | map_char_table_for_charset P_ ((void (*c_function) (Lisp_Object, Lisp_Object), |
| 120 | Lisp_Object function, Lisp_Object table, |
| 121 | Lisp_Object arg, struct charset *charset, |
| 122 | unsigned from, unsigned to)); |
| 123 | |
| 124 | #define CODE_POINT_TO_INDEX(charset, code) \ |
| 125 | ((charset)->code_linear_p \ |
| 126 | ? (code) - (charset)->min_code \ |
| 127 | : (((charset)->code_space_mask[(code) >> 24] & 0x8) \ |
| 128 | && ((charset)->code_space_mask[((code) >> 16) & 0xFF] & 0x4) \ |
| 129 | && ((charset)->code_space_mask[((code) >> 8) & 0xFF] & 0x2) \ |
| 130 | && ((charset)->code_space_mask[(code) & 0xFF] & 0x1)) \ |
| 131 | ? (((((code) >> 24) - (charset)->code_space[12]) \ |
| 132 | * (charset)->code_space[11]) \ |
| 133 | + (((((code) >> 16) & 0xFF) - (charset)->code_space[8]) \ |
| 134 | * (charset)->code_space[7]) \ |
| 135 | + (((((code) >> 8) & 0xFF) - (charset)->code_space[4]) \ |
| 136 | * (charset)->code_space[3]) \ |
| 137 | + (((code) & 0xFF) - (charset)->code_space[0]) \ |
| 138 | - ((charset)->char_index_offset)) \ |
| 139 | : -1) |
| 140 | |
| 141 | |
| 142 | /* Convert the character index IDX to code-point CODE for CHARSET. |
| 143 | It is assumed that IDX is in a valid range. */ |
| 144 | |
| 145 | #define INDEX_TO_CODE_POINT(charset, idx) \ |
| 146 | ((charset)->code_linear_p \ |
| 147 | ? (idx) + (charset)->min_code \ |
| 148 | : (idx += (charset)->char_index_offset, \ |
| 149 | (((charset)->code_space[0] + (idx) % (charset)->code_space[2]) \ |
| 150 | | (((charset)->code_space[4] \ |
| 151 | + ((idx) / (charset)->code_space[3] % (charset)->code_space[6])) \ |
| 152 | << 8) \ |
| 153 | | (((charset)->code_space[8] \ |
| 154 | + ((idx) / (charset)->code_space[7] % (charset)->code_space[10])) \ |
| 155 | << 16) \ |
| 156 | | (((charset)->code_space[12] + ((idx) / (charset)->code_space[11])) \ |
| 157 | << 24)))) |
| 158 | |
| 159 | |
| 160 | \f |
| 161 | |
| 162 | /* Set to 1 to warn that a charset map is loaded and thus a buffer |
| 163 | text and a string data may be relocated. */ |
| 164 | int charset_map_loaded; |
| 165 | |
| 166 | struct charset_map_entries |
| 167 | { |
| 168 | struct { |
| 169 | unsigned from, to; |
| 170 | int c; |
| 171 | } entry[0x10000]; |
| 172 | struct charset_map_entries *next; |
| 173 | }; |
| 174 | |
| 175 | /* Load the mapping information for CHARSET from ENTRIES. |
| 176 | |
| 177 | If CONTROL_FLAG is 0, setup CHARSET->min_char and CHARSET->max_char. |
| 178 | |
| 179 | If CONTROL_FLAG is 1, setup CHARSET->min_char, CHARSET->max_char, |
| 180 | CHARSET->decoder, and CHARSET->encoder. |
| 181 | |
| 182 | If CONTROL_FLAG is 2, setup CHARSET->deunifier and |
| 183 | Vchar_unify_table. If Vchar_unified_charset_table is non-nil, |
| 184 | setup it too. */ |
| 185 | |
| 186 | static void |
| 187 | load_charset_map (charset, entries, n_entries, control_flag) |
| 188 | struct charset *charset; |
| 189 | struct charset_map_entries *entries; |
| 190 | int n_entries; |
| 191 | int control_flag; |
| 192 | { |
| 193 | Lisp_Object vec, table; |
| 194 | unsigned max_code = CHARSET_MAX_CODE (charset); |
| 195 | int ascii_compatible_p = charset->ascii_compatible_p; |
| 196 | int min_char, max_char, nonascii_min_char; |
| 197 | int i; |
| 198 | unsigned char *fast_map = charset->fast_map; |
| 199 | |
| 200 | if (n_entries <= 0) |
| 201 | return; |
| 202 | |
| 203 | if (control_flag > 0) |
| 204 | { |
| 205 | int n = CODE_POINT_TO_INDEX (charset, max_code) + 1; |
| 206 | |
| 207 | table = Fmake_char_table (Qnil, Qnil); |
| 208 | if (control_flag == 1) |
| 209 | vec = Fmake_vector (make_number (n), make_number (-1)); |
| 210 | else if (! CHAR_TABLE_P (Vchar_unify_table)) |
| 211 | Vchar_unify_table = Fmake_char_table (Qnil, Qnil); |
| 212 | |
| 213 | charset_map_loaded = 1; |
| 214 | } |
| 215 | |
| 216 | min_char = max_char = entries->entry[0].c; |
| 217 | nonascii_min_char = MAX_CHAR; |
| 218 | for (i = 0; i < n_entries; i++) |
| 219 | { |
| 220 | unsigned from, to; |
| 221 | int from_index, to_index; |
| 222 | int from_c, to_c; |
| 223 | int idx = i % 0x10000; |
| 224 | |
| 225 | if (i > 0 && idx == 0) |
| 226 | entries = entries->next; |
| 227 | from = entries->entry[idx].from; |
| 228 | to = entries->entry[idx].to; |
| 229 | from_c = entries->entry[idx].c; |
| 230 | from_index = CODE_POINT_TO_INDEX (charset, from); |
| 231 | if (from == to) |
| 232 | { |
| 233 | to_index = from_index; |
| 234 | to_c = from_c; |
| 235 | } |
| 236 | else |
| 237 | { |
| 238 | to_index = CODE_POINT_TO_INDEX (charset, to); |
| 239 | to_c = from_c + (to_index - from_index); |
| 240 | } |
| 241 | if (from_index < 0 || to_index < 0) |
| 242 | continue; |
| 243 | |
| 244 | if (control_flag < 2) |
| 245 | { |
| 246 | int c; |
| 247 | |
| 248 | if (to_c > max_char) |
| 249 | max_char = to_c; |
| 250 | else if (from_c < min_char) |
| 251 | min_char = from_c; |
| 252 | if (ascii_compatible_p) |
| 253 | { |
| 254 | if (! ASCII_BYTE_P (from_c)) |
| 255 | { |
| 256 | if (from_c < nonascii_min_char) |
| 257 | nonascii_min_char = from_c; |
| 258 | } |
| 259 | else if (! ASCII_BYTE_P (to_c)) |
| 260 | { |
| 261 | nonascii_min_char = 0x80; |
| 262 | } |
| 263 | } |
| 264 | |
| 265 | for (c = from_c; c <= to_c; c++) |
| 266 | CHARSET_FAST_MAP_SET (c, fast_map); |
| 267 | |
| 268 | if (control_flag == 1) |
| 269 | { |
| 270 | unsigned code = from; |
| 271 | |
| 272 | if (CHARSET_COMPACT_CODES_P (charset)) |
| 273 | while (1) |
| 274 | { |
| 275 | ASET (vec, from_index, make_number (from_c)); |
| 276 | CHAR_TABLE_SET (table, from_c, make_number (code)); |
| 277 | if (from_index == to_index) |
| 278 | break; |
| 279 | from_index++, from_c++; |
| 280 | code = INDEX_TO_CODE_POINT (charset, from_index); |
| 281 | } |
| 282 | else |
| 283 | for (; from_index <= to_index; from_index++, from_c++) |
| 284 | { |
| 285 | ASET (vec, from_index, make_number (from_c)); |
| 286 | CHAR_TABLE_SET (table, from_c, make_number (from_index)); |
| 287 | } |
| 288 | } |
| 289 | } |
| 290 | else |
| 291 | { |
| 292 | unsigned code = from; |
| 293 | |
| 294 | while (1) |
| 295 | { |
| 296 | int c1 = DECODE_CHAR (charset, code); |
| 297 | |
| 298 | if (c1 >= 0) |
| 299 | { |
| 300 | CHAR_TABLE_SET (table, from_c, make_number (c1)); |
| 301 | CHAR_TABLE_SET (Vchar_unify_table, c1, make_number (from_c)); |
| 302 | if (CHAR_TABLE_P (Vchar_unified_charset_table)) |
| 303 | CHAR_TABLE_SET (Vchar_unified_charset_table, c1, |
| 304 | CHARSET_NAME (charset)); |
| 305 | } |
| 306 | if (from_index == to_index) |
| 307 | break; |
| 308 | from_index++, from_c++; |
| 309 | code = INDEX_TO_CODE_POINT (charset, from_index); |
| 310 | } |
| 311 | } |
| 312 | } |
| 313 | |
| 314 | if (control_flag < 2) |
| 315 | { |
| 316 | CHARSET_MIN_CHAR (charset) = (ascii_compatible_p |
| 317 | ? nonascii_min_char : min_char); |
| 318 | CHARSET_MAX_CHAR (charset) = max_char; |
| 319 | if (control_flag == 1) |
| 320 | { |
| 321 | CHARSET_DECODER (charset) = vec; |
| 322 | CHARSET_ENCODER (charset) = table; |
| 323 | } |
| 324 | } |
| 325 | else |
| 326 | CHARSET_DEUNIFIER (charset) = table; |
| 327 | } |
| 328 | |
| 329 | |
| 330 | /* Read a hexadecimal number (preceded by "0x") from the file FP while |
| 331 | paying attention to comment charcter '#'. */ |
| 332 | |
| 333 | static INLINE unsigned |
| 334 | read_hex (fp, eof) |
| 335 | FILE *fp; |
| 336 | int *eof; |
| 337 | { |
| 338 | int c; |
| 339 | unsigned n; |
| 340 | |
| 341 | while ((c = getc (fp)) != EOF) |
| 342 | { |
| 343 | if (c == '#') |
| 344 | { |
| 345 | while ((c = getc (fp)) != EOF && c != '\n'); |
| 346 | } |
| 347 | else if (c == '0') |
| 348 | { |
| 349 | if ((c = getc (fp)) == EOF || c == 'x') |
| 350 | break; |
| 351 | } |
| 352 | } |
| 353 | if (c == EOF) |
| 354 | { |
| 355 | *eof = 1; |
| 356 | return 0; |
| 357 | } |
| 358 | *eof = 0; |
| 359 | n = 0; |
| 360 | if (c == 'x') |
| 361 | while ((c = getc (fp)) != EOF && isxdigit (c)) |
| 362 | n = ((n << 4) |
| 363 | | (c <= '9' ? c - '0' : c <= 'F' ? c - 'A' + 10 : c - 'a' + 10)); |
| 364 | else |
| 365 | while ((c = getc (fp)) != EOF && isdigit (c)) |
| 366 | n = (n * 10) + c - '0'; |
| 367 | if (c != EOF) |
| 368 | ungetc (c, fp); |
| 369 | return n; |
| 370 | } |
| 371 | |
| 372 | |
| 373 | /* Return a mapping vector for CHARSET loaded from MAPFILE. |
| 374 | Each line of MAPFILE has this form |
| 375 | 0xAAAA 0xCCCC |
| 376 | where 0xAAAA is a code-point and 0xCCCC is the corresponding |
| 377 | character code, or this form |
| 378 | 0xAAAA-0xBBBB 0xCCCC |
| 379 | where 0xAAAA and 0xBBBB are code-points specifying a range, and |
| 380 | 0xCCCC is the first character code of the range. |
| 381 | |
| 382 | The returned vector has this form: |
| 383 | [ CODE1 CHAR1 CODE2 CHAR2 .... ] |
| 384 | where CODE1 is a code-point or a cons of code-points specifying a |
| 385 | range. */ |
| 386 | |
| 387 | extern void add_to_log P_ ((char *, Lisp_Object, Lisp_Object)); |
| 388 | |
| 389 | static void |
| 390 | load_charset_map_from_file (charset, mapfile, control_flag) |
| 391 | struct charset *charset; |
| 392 | Lisp_Object mapfile; |
| 393 | int control_flag; |
| 394 | { |
| 395 | unsigned min_code = CHARSET_MIN_CODE (charset); |
| 396 | unsigned max_code = CHARSET_MAX_CODE (charset); |
| 397 | int fd; |
| 398 | FILE *fp; |
| 399 | int eof; |
| 400 | Lisp_Object suffixes; |
| 401 | struct charset_map_entries *head, *entries; |
| 402 | int n_entries; |
| 403 | |
| 404 | suffixes = Fcons (build_string (".map"), |
| 405 | Fcons (build_string (".TXT"), Qnil)); |
| 406 | |
| 407 | fd = openp (Fcons (Vcharset_map_directory, Qnil), mapfile, suffixes, |
| 408 | NULL, 0); |
| 409 | if (fd < 0 |
| 410 | || ! (fp = fdopen (fd, "r"))) |
| 411 | { |
| 412 | add_to_log ("Failure in loading charset map: %S", mapfile, Qnil); |
| 413 | return; |
| 414 | } |
| 415 | |
| 416 | head = entries = ((struct charset_map_entries *) |
| 417 | alloca (sizeof (struct charset_map_entries))); |
| 418 | n_entries = 0; |
| 419 | eof = 0; |
| 420 | while (1) |
| 421 | { |
| 422 | unsigned from, to; |
| 423 | int c; |
| 424 | int idx; |
| 425 | |
| 426 | from = read_hex (fp, &eof); |
| 427 | if (eof) |
| 428 | break; |
| 429 | if (getc (fp) == '-') |
| 430 | to = read_hex (fp, &eof); |
| 431 | else |
| 432 | to = from; |
| 433 | c = (int) read_hex (fp, &eof); |
| 434 | |
| 435 | if (from < min_code || to > max_code || from > to || c > MAX_CHAR) |
| 436 | continue; |
| 437 | |
| 438 | if (n_entries > 0 && (n_entries % 0x10000) == 0) |
| 439 | { |
| 440 | entries->next = ((struct charset_map_entries *) |
| 441 | alloca (sizeof (struct charset_map_entries))); |
| 442 | entries = entries->next; |
| 443 | } |
| 444 | idx = n_entries % 0x10000; |
| 445 | entries->entry[idx].from = from; |
| 446 | entries->entry[idx].to = to; |
| 447 | entries->entry[idx].c = c; |
| 448 | n_entries++; |
| 449 | } |
| 450 | fclose (fp); |
| 451 | close (fd); |
| 452 | |
| 453 | load_charset_map (charset, head, n_entries, control_flag); |
| 454 | } |
| 455 | |
| 456 | static void |
| 457 | load_charset_map_from_vector (charset, vec, control_flag) |
| 458 | struct charset *charset; |
| 459 | Lisp_Object vec; |
| 460 | int control_flag; |
| 461 | { |
| 462 | unsigned min_code = CHARSET_MIN_CODE (charset); |
| 463 | unsigned max_code = CHARSET_MAX_CODE (charset); |
| 464 | struct charset_map_entries *head, *entries; |
| 465 | int n_entries; |
| 466 | int len = ASIZE (vec); |
| 467 | int i; |
| 468 | |
| 469 | if (len % 2 == 1) |
| 470 | { |
| 471 | add_to_log ("Failure in loading charset map: %V", vec, Qnil); |
| 472 | return; |
| 473 | } |
| 474 | |
| 475 | head = entries = ((struct charset_map_entries *) |
| 476 | alloca (sizeof (struct charset_map_entries))); |
| 477 | n_entries = 0; |
| 478 | for (i = 0; i < len; i += 2) |
| 479 | { |
| 480 | Lisp_Object val, val2; |
| 481 | unsigned from, to; |
| 482 | int c; |
| 483 | int idx; |
| 484 | |
| 485 | val = AREF (vec, i); |
| 486 | if (CONSP (val)) |
| 487 | { |
| 488 | val2 = XCDR (val); |
| 489 | val = XCAR (val); |
| 490 | CHECK_NATNUM (val); |
| 491 | CHECK_NATNUM (val2); |
| 492 | from = XFASTINT (val); |
| 493 | to = XFASTINT (val2); |
| 494 | } |
| 495 | else |
| 496 | { |
| 497 | CHECK_NATNUM (val); |
| 498 | from = to = XFASTINT (val); |
| 499 | } |
| 500 | val = AREF (vec, i + 1); |
| 501 | CHECK_NATNUM (val); |
| 502 | c = XFASTINT (val); |
| 503 | |
| 504 | if (from < min_code || to > max_code || from > to || c > MAX_CHAR) |
| 505 | continue; |
| 506 | |
| 507 | if ((n_entries % 0x10000) == 0) |
| 508 | { |
| 509 | entries->next = ((struct charset_map_entries *) |
| 510 | alloca (sizeof (struct charset_map_entries))); |
| 511 | entries = entries->next; |
| 512 | } |
| 513 | idx = n_entries % 0x10000; |
| 514 | entries->entry[idx].from = from; |
| 515 | entries->entry[idx].to = to; |
| 516 | entries->entry[idx].c = c; |
| 517 | n_entries++; |
| 518 | } |
| 519 | |
| 520 | load_charset_map (charset, head, n_entries, control_flag); |
| 521 | } |
| 522 | |
| 523 | static void |
| 524 | load_charset (charset) |
| 525 | struct charset *charset; |
| 526 | { |
| 527 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP_DEFERRED) |
| 528 | { |
| 529 | Lisp_Object map; |
| 530 | |
| 531 | map = CHARSET_MAP (charset); |
| 532 | if (STRINGP (map)) |
| 533 | load_charset_map_from_file (charset, map, 1); |
| 534 | else |
| 535 | load_charset_map_from_vector (charset, map, 1); |
| 536 | CHARSET_METHOD (charset) = CHARSET_METHOD_MAP; |
| 537 | } |
| 538 | } |
| 539 | |
| 540 | |
| 541 | DEFUN ("charsetp", Fcharsetp, Scharsetp, 1, 1, 0, |
| 542 | doc: /* Return non-nil if and only if OBJECT is a charset.*/) |
| 543 | (object) |
| 544 | Lisp_Object object; |
| 545 | { |
| 546 | return (CHARSETP (object) ? Qt : Qnil); |
| 547 | } |
| 548 | |
| 549 | |
| 550 | void |
| 551 | map_charset_chars (c_function, function, arg, |
| 552 | charset, from, to) |
| 553 | void (*c_function) P_ ((Lisp_Object, Lisp_Object)); |
| 554 | Lisp_Object function, arg; |
| 555 | struct charset *charset; |
| 556 | unsigned from, to; |
| 557 | |
| 558 | { |
| 559 | Lisp_Object range; |
| 560 | int partial; |
| 561 | |
| 562 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP_DEFERRED) |
| 563 | load_charset (charset); |
| 564 | |
| 565 | partial = (from > CHARSET_MIN_CODE (charset) |
| 566 | || to < CHARSET_MAX_CODE (charset)); |
| 567 | |
| 568 | if (CHARSET_UNIFIED_P (charset) |
| 569 | && CHAR_TABLE_P (CHARSET_DEUNIFIER (charset))) |
| 570 | { |
| 571 | map_char_table_for_charset (c_function, function, |
| 572 | CHARSET_DEUNIFIER (charset), arg, |
| 573 | partial ? charset : NULL, from, to); |
| 574 | } |
| 575 | |
| 576 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_OFFSET) |
| 577 | { |
| 578 | int from_idx = CODE_POINT_TO_INDEX (charset, from); |
| 579 | int to_idx = CODE_POINT_TO_INDEX (charset, to); |
| 580 | int from_c = from_idx + CHARSET_CODE_OFFSET (charset); |
| 581 | int to_c = to_idx + CHARSET_CODE_OFFSET (charset); |
| 582 | |
| 583 | range = Fcons (make_number (from_c), make_number (to_c)); |
| 584 | if (NILP (function)) |
| 585 | (*c_function) (range, arg); |
| 586 | else |
| 587 | call2 (function, range, arg); |
| 588 | } |
| 589 | else if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP) |
| 590 | { |
| 591 | if (! CHAR_TABLE_P (CHARSET_ENCODER (charset))) |
| 592 | return; |
| 593 | if (CHARSET_ASCII_COMPATIBLE_P (charset) && from <= 127) |
| 594 | { |
| 595 | range = Fcons (make_number (from), make_number (to)); |
| 596 | if (to >= 128) |
| 597 | XSETCAR (range, make_number (127)); |
| 598 | |
| 599 | if (NILP (function)) |
| 600 | (*c_function) (range, arg); |
| 601 | else |
| 602 | call2 (function, range, arg); |
| 603 | } |
| 604 | map_char_table_for_charset (c_function, function, |
| 605 | CHARSET_ENCODER (charset), arg, |
| 606 | partial ? charset : NULL, from, to); |
| 607 | } |
| 608 | else if (CHARSET_METHOD (charset) == CHARSET_METHOD_SUBSET) |
| 609 | { |
| 610 | Lisp_Object subset_info; |
| 611 | int offset; |
| 612 | |
| 613 | subset_info = CHARSET_SUBSET (charset); |
| 614 | charset = CHARSET_FROM_ID (XFASTINT (AREF (subset_info, 0))); |
| 615 | offset = XINT (AREF (subset_info, 3)); |
| 616 | from -= offset; |
| 617 | if (from < XFASTINT (AREF (subset_info, 1))) |
| 618 | from = XFASTINT (AREF (subset_info, 1)); |
| 619 | to -= offset; |
| 620 | if (to > XFASTINT (AREF (subset_info, 2))) |
| 621 | to = XFASTINT (AREF (subset_info, 2)); |
| 622 | map_charset_chars (c_function, function, arg, charset, from, to); |
| 623 | } |
| 624 | else /* i.e. CHARSET_METHOD_SUPERSET */ |
| 625 | { |
| 626 | Lisp_Object parents; |
| 627 | |
| 628 | for (parents = CHARSET_SUPERSET (charset); CONSP (parents); |
| 629 | parents = XCDR (parents)) |
| 630 | { |
| 631 | int offset; |
| 632 | unsigned this_from, this_to; |
| 633 | |
| 634 | charset = CHARSET_FROM_ID (XFASTINT (XCAR (XCAR (parents)))); |
| 635 | offset = XINT (XCDR (XCAR (parents))); |
| 636 | this_from = from - offset; |
| 637 | this_to = to - offset; |
| 638 | if (this_from < CHARSET_MIN_CODE (charset)) |
| 639 | this_from = CHARSET_MIN_CODE (charset); |
| 640 | if (this_to > CHARSET_MAX_CODE (charset)) |
| 641 | this_to = CHARSET_MAX_CODE (charset); |
| 642 | map_charset_chars (c_function, function, arg, charset, from, to); |
| 643 | } |
| 644 | } |
| 645 | } |
| 646 | |
| 647 | |
| 648 | DEFUN ("map-charset-chars", Fmap_charset_chars, Smap_charset_chars, 2, 5, 0, |
| 649 | doc: /* Call FUNCTION for all characters in CHARSET. |
| 650 | FUNCTION is called with an argument RANGE and the optional 3rd |
| 651 | argument ARG. |
| 652 | |
| 653 | RANGE is a cons (FROM . TO), where FROM and TO indicate a range of |
| 654 | characters contained in CHARSET. |
| 655 | |
| 656 | The optional 4th and 5th arguments FROM-CODE and TO-CODE specify the |
| 657 | range of code points of targer characters. */) |
| 658 | (function, charset, arg, from_code, to_code) |
| 659 | Lisp_Object function, charset, arg, from_code, to_code; |
| 660 | { |
| 661 | struct charset *cs; |
| 662 | unsigned from, to; |
| 663 | |
| 664 | CHECK_CHARSET_GET_CHARSET (charset, cs); |
| 665 | if (NILP (from_code)) |
| 666 | from = CHARSET_MIN_CODE (cs); |
| 667 | else |
| 668 | { |
| 669 | CHECK_NATNUM (from_code); |
| 670 | from = XINT (from_code); |
| 671 | if (from < CHARSET_MIN_CODE (cs)) |
| 672 | from = CHARSET_MIN_CODE (cs); |
| 673 | } |
| 674 | if (NILP (to_code)) |
| 675 | to = CHARSET_MAX_CODE (cs); |
| 676 | else |
| 677 | { |
| 678 | CHECK_NATNUM (to_code); |
| 679 | to = XINT (to_code); |
| 680 | if (to > CHARSET_MAX_CODE (cs)) |
| 681 | to = CHARSET_MAX_CODE (cs); |
| 682 | } |
| 683 | map_charset_chars (NULL, function, arg, cs, from, to); |
| 684 | return Qnil; |
| 685 | } |
| 686 | |
| 687 | |
| 688 | /* Define a charset according to the arguments. The Nth argument is |
| 689 | the Nth attribute of the charset (the last attribute `charset-id' |
| 690 | is not included). See the docstring of `define-charset' for the |
| 691 | detail. */ |
| 692 | |
| 693 | DEFUN ("define-charset-internal", Fdefine_charset_internal, |
| 694 | Sdefine_charset_internal, charset_arg_max, MANY, 0, |
| 695 | doc: /* For internal use only. |
| 696 | usage: (define-charset-internal ...) */) |
| 697 | (nargs, args) |
| 698 | int nargs; |
| 699 | Lisp_Object *args; |
| 700 | { |
| 701 | /* Charset attr vector. */ |
| 702 | Lisp_Object attrs; |
| 703 | Lisp_Object val; |
| 704 | unsigned hash_code; |
| 705 | struct Lisp_Hash_Table *hash_table = XHASH_TABLE (Vcharset_hash_table); |
| 706 | int i, j; |
| 707 | struct charset charset; |
| 708 | int id; |
| 709 | int dimension; |
| 710 | int new_definition_p; |
| 711 | int nchars; |
| 712 | |
| 713 | if (nargs != charset_arg_max) |
| 714 | return Fsignal (Qwrong_number_of_arguments, |
| 715 | Fcons (intern ("define-charset-internal"), |
| 716 | make_number (nargs))); |
| 717 | |
| 718 | attrs = Fmake_vector (make_number (charset_attr_max), Qnil); |
| 719 | |
| 720 | CHECK_SYMBOL (args[charset_arg_name]); |
| 721 | ASET (attrs, charset_name, args[charset_arg_name]); |
| 722 | |
| 723 | val = args[charset_arg_code_space]; |
| 724 | for (i = 0, dimension = 0, nchars = 1; i < 4; i++) |
| 725 | { |
| 726 | int min_byte, max_byte; |
| 727 | |
| 728 | min_byte = XINT (Faref (val, make_number (i * 2))); |
| 729 | max_byte = XINT (Faref (val, make_number (i * 2 + 1))); |
| 730 | if (min_byte < 0 || min_byte > max_byte || max_byte >= 256) |
| 731 | error ("Invalid :code-space value"); |
| 732 | charset.code_space[i * 4] = min_byte; |
| 733 | charset.code_space[i * 4 + 1] = max_byte; |
| 734 | charset.code_space[i * 4 + 2] = max_byte - min_byte + 1; |
| 735 | nchars *= charset.code_space[i * 4 + 2]; |
| 736 | charset.code_space[i * 4 + 3] = nchars; |
| 737 | if (max_byte > 0) |
| 738 | dimension = i + 1; |
| 739 | } |
| 740 | |
| 741 | val = args[charset_arg_dimension]; |
| 742 | if (NILP (val)) |
| 743 | charset.dimension = dimension; |
| 744 | else |
| 745 | { |
| 746 | CHECK_NATNUM (val); |
| 747 | charset.dimension = XINT (val); |
| 748 | if (charset.dimension < 1 || charset.dimension > 4) |
| 749 | args_out_of_range_3 (val, make_number (1), make_number (4)); |
| 750 | } |
| 751 | |
| 752 | charset.code_linear_p |
| 753 | = (charset.dimension == 1 |
| 754 | || (charset.code_space[2] == 256 |
| 755 | && (charset.dimension == 2 |
| 756 | || (charset.code_space[6] == 256 |
| 757 | && (charset.dimension == 3 |
| 758 | || charset.code_space[10] == 256))))); |
| 759 | |
| 760 | if (! charset.code_linear_p) |
| 761 | { |
| 762 | charset.code_space_mask = (unsigned char *) xmalloc (256); |
| 763 | bzero (charset.code_space_mask, 256); |
| 764 | for (i = 0; i < 4; i++) |
| 765 | for (j = charset.code_space[i * 4]; j <= charset.code_space[i * 4 + 1]; |
| 766 | j++) |
| 767 | charset.code_space_mask[j] |= (1 << i); |
| 768 | } |
| 769 | |
| 770 | charset.iso_chars_96 = charset.code_space[2] == 96; |
| 771 | |
| 772 | charset.min_code = (charset.code_space[0] |
| 773 | | (charset.code_space[4] << 8) |
| 774 | | (charset.code_space[8] << 16) |
| 775 | | (charset.code_space[12] << 24)); |
| 776 | charset.max_code = (charset.code_space[1] |
| 777 | | (charset.code_space[5] << 8) |
| 778 | | (charset.code_space[9] << 16) |
| 779 | | (charset.code_space[13] << 24)); |
| 780 | charset.char_index_offset = 0; |
| 781 | |
| 782 | val = args[charset_arg_min_code]; |
| 783 | if (! NILP (val)) |
| 784 | { |
| 785 | unsigned code; |
| 786 | |
| 787 | if (INTEGERP (val)) |
| 788 | code = XINT (val); |
| 789 | else |
| 790 | { |
| 791 | CHECK_CONS (val); |
| 792 | CHECK_NUMBER (XCAR (val)); |
| 793 | CHECK_NUMBER (XCDR (val)); |
| 794 | code = (XINT (XCAR (val)) << 16) | (XINT (XCDR (val))); |
| 795 | } |
| 796 | if (code < charset.min_code |
| 797 | || code > charset.max_code) |
| 798 | args_out_of_range_3 (make_number (charset.min_code), |
| 799 | make_number (charset.max_code), val); |
| 800 | charset.char_index_offset = CODE_POINT_TO_INDEX (&charset, code); |
| 801 | charset.min_code = code; |
| 802 | } |
| 803 | |
| 804 | val = args[charset_arg_max_code]; |
| 805 | if (! NILP (val)) |
| 806 | { |
| 807 | unsigned code; |
| 808 | |
| 809 | if (INTEGERP (val)) |
| 810 | code = XINT (val); |
| 811 | else |
| 812 | { |
| 813 | CHECK_CONS (val); |
| 814 | CHECK_NUMBER (XCAR (val)); |
| 815 | CHECK_NUMBER (XCDR (val)); |
| 816 | code = (XINT (XCAR (val)) << 16) | (XINT (XCDR (val))); |
| 817 | } |
| 818 | if (code < charset.min_code |
| 819 | || code > charset.max_code) |
| 820 | args_out_of_range_3 (make_number (charset.min_code), |
| 821 | make_number (charset.max_code), val); |
| 822 | charset.max_code = code; |
| 823 | } |
| 824 | |
| 825 | charset.compact_codes_p = charset.max_code < 0x1000000; |
| 826 | |
| 827 | val = args[charset_arg_invalid_code]; |
| 828 | if (NILP (val)) |
| 829 | { |
| 830 | if (charset.min_code > 0) |
| 831 | charset.invalid_code = 0; |
| 832 | else |
| 833 | { |
| 834 | XSETINT (val, charset.max_code + 1); |
| 835 | if (XINT (val) == charset.max_code + 1) |
| 836 | charset.invalid_code = charset.max_code + 1; |
| 837 | else |
| 838 | error ("Attribute :invalid-code must be specified"); |
| 839 | } |
| 840 | } |
| 841 | else |
| 842 | { |
| 843 | CHECK_NATNUM (val); |
| 844 | charset.invalid_code = XFASTINT (val); |
| 845 | } |
| 846 | |
| 847 | val = args[charset_arg_iso_final]; |
| 848 | if (NILP (val)) |
| 849 | charset.iso_final = -1; |
| 850 | else |
| 851 | { |
| 852 | CHECK_NUMBER (val); |
| 853 | if (XINT (val) < '0' || XINT (val) > 127) |
| 854 | error ("Invalid iso-final-char: %d", XINT (val)); |
| 855 | charset.iso_final = XINT (val); |
| 856 | } |
| 857 | |
| 858 | val = args[charset_arg_iso_revision]; |
| 859 | if (NILP (val)) |
| 860 | charset.iso_revision = -1; |
| 861 | else |
| 862 | { |
| 863 | CHECK_NUMBER (val); |
| 864 | if (XINT (val) > 63) |
| 865 | args_out_of_range (make_number (63), val); |
| 866 | charset.iso_revision = XINT (val); |
| 867 | } |
| 868 | |
| 869 | val = args[charset_arg_emacs_mule_id]; |
| 870 | if (NILP (val)) |
| 871 | charset.emacs_mule_id = -1; |
| 872 | else |
| 873 | { |
| 874 | CHECK_NATNUM (val); |
| 875 | if ((XINT (val) > 0 && XINT (val) <= 128) || XINT (val) >= 256) |
| 876 | error ("Invalid emacs-mule-id: %d", XINT (val)); |
| 877 | charset.emacs_mule_id = XINT (val); |
| 878 | } |
| 879 | |
| 880 | charset.ascii_compatible_p = ! NILP (args[charset_arg_ascii_compatible_p]); |
| 881 | |
| 882 | charset.supplementary_p = ! NILP (args[charset_arg_supplementary_p]); |
| 883 | |
| 884 | charset.unified_p = 0; |
| 885 | |
| 886 | bzero (charset.fast_map, sizeof (charset.fast_map)); |
| 887 | |
| 888 | if (! NILP (args[charset_arg_code_offset])) |
| 889 | { |
| 890 | val = args[charset_arg_code_offset]; |
| 891 | CHECK_NUMBER (val); |
| 892 | |
| 893 | charset.method = CHARSET_METHOD_OFFSET; |
| 894 | charset.code_offset = XINT (val); |
| 895 | |
| 896 | i = CODE_POINT_TO_INDEX (&charset, charset.min_code); |
| 897 | charset.min_char = i + charset.code_offset; |
| 898 | i = CODE_POINT_TO_INDEX (&charset, charset.max_code); |
| 899 | charset.max_char = i + charset.code_offset; |
| 900 | if (charset.max_char > MAX_CHAR) |
| 901 | error ("Unsupported max char: %d", charset.max_char); |
| 902 | |
| 903 | for (i = charset.min_char; i < 0x10000 && i <= charset.max_char; |
| 904 | i += 128) |
| 905 | CHARSET_FAST_MAP_SET (i, charset.fast_map); |
| 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)) |
| 914 | load_charset_map_from_file (&charset, val, 0); |
| 915 | else |
| 916 | load_charset_map_from_vector (&charset, val, 0); |
| 917 | charset.method = CHARSET_METHOD_MAP_DEFERRED; |
| 918 | } |
| 919 | else if (! NILP (args[charset_arg_subset])) |
| 920 | { |
| 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])) |
| 953 | { |
| 954 | val = args[charset_arg_superset]; |
| 955 | charset.method = CHARSET_METHOD_SUPERSET; |
| 956 | val = Fcopy_sequence (val); |
| 957 | ASET (attrs, charset_superset, val); |
| 958 | |
| 959 | charset.min_char = MAX_CHAR; |
| 960 | charset.max_char = 0; |
| 961 | for (; ! NILP (val); val = Fcdr (val)) |
| 962 | { |
| 963 | Lisp_Object elt, car_part, cdr_part; |
| 964 | int this_id, offset; |
| 965 | struct charset *this_charset; |
| 966 | |
| 967 | elt = Fcar (val); |
| 968 | if (CONSP (elt)) |
| 969 | { |
| 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); |
| 975 | } |
| 976 | else |
| 977 | { |
| 978 | CHECK_CHARSET_GET_ID (elt, this_id); |
| 979 | offset = 0; |
| 980 | } |
| 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]; |
| 990 | } |
| 991 | } |
| 992 | else |
| 993 | error ("None of :code-offset, :map, :parents are specified"); |
| 994 | |
| 995 | val = args[charset_arg_unify_map]; |
| 996 | if (! NILP (val) && !STRINGP (val)) |
| 997 | CHECK_VECTOR (val); |
| 998 | ASET (attrs, charset_unify_map, val); |
| 999 | |
| 1000 | CHECK_LIST (args[charset_arg_plist]); |
| 1001 | ASET (attrs, charset_plist, args[charset_arg_plist]); |
| 1002 | |
| 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; |
| 1008 | id = XFASTINT (CHARSET_SYMBOL_ID (args[charset_arg_name])); |
| 1009 | HASH_VALUE (hash_table, charset.hash_index) = attrs; |
| 1010 | } |
| 1011 | else |
| 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 | { |
| 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; |
| 1024 | } |
| 1025 | id = charset_table_used++; |
| 1026 | new_definition_p = 1; |
| 1027 | } |
| 1028 | |
| 1029 | ASET (attrs, charset_id, make_number (id)); |
| 1030 | charset.id = id; |
| 1031 | charset_table[id] = charset; |
| 1032 | |
| 1033 | if (charset.iso_final >= 0) |
| 1034 | { |
| 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)); |
| 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; |
| 1046 | } |
| 1047 | |
| 1048 | if (charset.emacs_mule_id >= 0) |
| 1049 | { |
| 1050 | emacs_mule_charset[charset.emacs_mule_id] = CHARSET_FROM_ID (id); |
| 1051 | if (charset.emacs_mule_id < 0xA0) |
| 1052 | emacs_mule_bytes[charset.emacs_mule_id] = charset.dimension + 1; |
| 1053 | if (new_definition_p) |
| 1054 | Vemacs_mule_charset_list = nconc2 (Vemacs_mule_charset_list, |
| 1055 | Fcons (make_number (id), Qnil)); |
| 1056 | } |
| 1057 | |
| 1058 | if (new_definition_p) |
| 1059 | { |
| 1060 | Vcharset_list = Fcons (args[charset_arg_name], Vcharset_list); |
| 1061 | Vcharset_ordered_list = nconc2 (Vcharset_ordered_list, |
| 1062 | Fcons (make_number (id), Qnil)); |
| 1063 | charset_ordered_list_tick++; |
| 1064 | } |
| 1065 | |
| 1066 | return Qnil; |
| 1067 | } |
| 1068 | |
| 1069 | |
| 1070 | /* Same as Fdefine_charset_internal but arguments are more convenient |
| 1071 | to call from C (typically in syms_of_charset). This can define a |
| 1072 | charset of `offset' method only. Return the ID of the new |
| 1073 | charset. */ |
| 1074 | |
| 1075 | static int |
| 1076 | define_charset_internal (name, dimension, code_space, min_code, max_code, |
| 1077 | iso_final, iso_revision, emacs_mule_id, |
| 1078 | ascii_compatible, supprementary, |
| 1079 | code_offset) |
| 1080 | Lisp_Object name; |
| 1081 | int dimension; |
| 1082 | unsigned char *code_space; |
| 1083 | unsigned min_code, max_code; |
| 1084 | int iso_final, iso_revision, emacs_mule_id; |
| 1085 | int ascii_compatible, supprementary; |
| 1086 | int code_offset; |
| 1087 | { |
| 1088 | Lisp_Object args[charset_arg_max]; |
| 1089 | Lisp_Object plist[14]; |
| 1090 | Lisp_Object val; |
| 1091 | int i; |
| 1092 | |
| 1093 | args[charset_arg_name] = name; |
| 1094 | args[charset_arg_dimension] = make_number (dimension); |
| 1095 | val = Fmake_vector (make_number (8), make_number (0)); |
| 1096 | for (i = 0; i < 8; i++) |
| 1097 | ASET (val, i, make_number (code_space[i])); |
| 1098 | args[charset_arg_code_space] = val; |
| 1099 | args[charset_arg_min_code] = make_number (min_code); |
| 1100 | args[charset_arg_max_code] = make_number (max_code); |
| 1101 | args[charset_arg_iso_final] |
| 1102 | = (iso_final < 0 ? Qnil : make_number (iso_final)); |
| 1103 | args[charset_arg_iso_revision] = make_number (iso_revision); |
| 1104 | args[charset_arg_emacs_mule_id] |
| 1105 | = (emacs_mule_id < 0 ? Qnil : make_number (emacs_mule_id)); |
| 1106 | args[charset_arg_ascii_compatible_p] = ascii_compatible ? Qt : Qnil; |
| 1107 | args[charset_arg_supplementary_p] = supprementary ? Qt : Qnil; |
| 1108 | args[charset_arg_invalid_code] = Qnil; |
| 1109 | args[charset_arg_code_offset] = make_number (code_offset); |
| 1110 | args[charset_arg_map] = Qnil; |
| 1111 | args[charset_arg_subset] = Qnil; |
| 1112 | args[charset_arg_superset] = Qnil; |
| 1113 | args[charset_arg_unify_map] = Qnil; |
| 1114 | |
| 1115 | plist[0] = intern (":name"); |
| 1116 | plist[1] = args[charset_arg_name]; |
| 1117 | plist[2] = intern (":dimension"); |
| 1118 | plist[3] = args[charset_arg_dimension]; |
| 1119 | plist[4] = intern (":code-space"); |
| 1120 | plist[5] = args[charset_arg_code_space]; |
| 1121 | plist[6] = intern (":iso-final-char"); |
| 1122 | plist[7] = args[charset_arg_iso_final]; |
| 1123 | plist[8] = intern (":emacs-mule-id"); |
| 1124 | plist[9] = args[charset_arg_emacs_mule_id]; |
| 1125 | plist[10] = intern (":ascii-compatible-p"); |
| 1126 | plist[11] = args[charset_arg_ascii_compatible_p]; |
| 1127 | plist[12] = intern (":code-offset"); |
| 1128 | plist[13] = args[charset_arg_code_offset]; |
| 1129 | |
| 1130 | args[charset_arg_plist] = Flist (14, plist); |
| 1131 | Fdefine_charset_internal (charset_arg_max, args); |
| 1132 | |
| 1133 | return XINT (CHARSET_SYMBOL_ID (name)); |
| 1134 | } |
| 1135 | |
| 1136 | |
| 1137 | DEFUN ("define-charset-alias", Fdefine_charset_alias, |
| 1138 | Sdefine_charset_alias, 2, 2, 0, |
| 1139 | doc: /* Define ALIAS as an alias for charset CHARSET. */) |
| 1140 | (alias, charset) |
| 1141 | Lisp_Object alias, charset; |
| 1142 | { |
| 1143 | Lisp_Object attr; |
| 1144 | |
| 1145 | CHECK_CHARSET_GET_ATTR (charset, attr); |
| 1146 | Fputhash (alias, attr, Vcharset_hash_table); |
| 1147 | Vcharset_list = Fcons (alias, Vcharset_list); |
| 1148 | return Qnil; |
| 1149 | } |
| 1150 | |
| 1151 | |
| 1152 | DEFUN ("primary-charset", Fprimary_charset, Sprimary_charset, 0, 0, 0, |
| 1153 | doc: /* Return the primary charset (set by `set-primary-charset'). */) |
| 1154 | () |
| 1155 | { |
| 1156 | return CHARSET_NAME (CHARSET_FROM_ID (charset_primary)); |
| 1157 | } |
| 1158 | |
| 1159 | |
| 1160 | DEFUN ("set-primary-charset", Fset_primary_charset, Sset_primary_charset, |
| 1161 | 1, 1, 0, |
| 1162 | doc: /* Set the primary charset to CHARSET. |
| 1163 | This determines how unibyte/multibyte conversion is done. See also |
| 1164 | function `primary-charset'. */) |
| 1165 | (charset) |
| 1166 | Lisp_Object charset; |
| 1167 | { |
| 1168 | int id; |
| 1169 | |
| 1170 | CHECK_CHARSET_GET_ID (charset, id); |
| 1171 | charset_primary = id; |
| 1172 | return Qnil; |
| 1173 | } |
| 1174 | |
| 1175 | |
| 1176 | DEFUN ("charset-plist", Fcharset_plist, Scharset_plist, 1, 1, 0, |
| 1177 | doc: /* Return the property list of CHARSET. */) |
| 1178 | (charset) |
| 1179 | Lisp_Object charset; |
| 1180 | { |
| 1181 | Lisp_Object attrs; |
| 1182 | |
| 1183 | CHECK_CHARSET_GET_ATTR (charset, attrs); |
| 1184 | return CHARSET_ATTR_PLIST (attrs); |
| 1185 | } |
| 1186 | |
| 1187 | |
| 1188 | DEFUN ("set-charset-plist", Fset_charset_plist, Sset_charset_plist, 2, 2, 0, |
| 1189 | doc: /* Set CHARSET's property list to PLIST. */) |
| 1190 | (charset, plist) |
| 1191 | Lisp_Object charset, plist; |
| 1192 | { |
| 1193 | Lisp_Object attrs; |
| 1194 | |
| 1195 | CHECK_CHARSET_GET_ATTR (charset, attrs); |
| 1196 | CHARSET_ATTR_PLIST (attrs) = plist; |
| 1197 | return plist; |
| 1198 | } |
| 1199 | |
| 1200 | |
| 1201 | DEFUN ("unify-charset", Funify_charset, Sunify_charset, 1, 3, 0, |
| 1202 | doc: /* Unify characters of CHARSET with Unicode. |
| 1203 | This means reading the relevant file and installing the table defined |
| 1204 | by CHARSET's `:unify-map' property. |
| 1205 | |
| 1206 | Optional second arg UNIFY-MAP is a file name string or a vector. It has |
| 1207 | the same meaning as the `:unify-map' attribute in the function |
| 1208 | `define-charset' (which see). |
| 1209 | |
| 1210 | Optional third argument DEUNIFY, if non-nil, means to de-unify CHARSET. */) |
| 1211 | (charset, unify_map, deunify) |
| 1212 | Lisp_Object charset, unify_map, deunify; |
| 1213 | { |
| 1214 | int id; |
| 1215 | struct charset *cs; |
| 1216 | |
| 1217 | CHECK_CHARSET_GET_ID (charset, id); |
| 1218 | cs = CHARSET_FROM_ID (id); |
| 1219 | if (CHARSET_METHOD (cs) == CHARSET_METHOD_MAP_DEFERRED) |
| 1220 | load_charset (cs); |
| 1221 | if (NILP (deunify) |
| 1222 | ? CHARSET_UNIFIED_P (cs) && ! NILP (CHARSET_DEUNIFIER (cs)) |
| 1223 | : ! CHARSET_UNIFIED_P (cs)) |
| 1224 | return Qnil; |
| 1225 | |
| 1226 | CHARSET_UNIFIED_P (cs) = 0; |
| 1227 | if (NILP (deunify)) |
| 1228 | { |
| 1229 | if (CHARSET_METHOD (cs) != CHARSET_METHOD_OFFSET) |
| 1230 | error ("Can't unify charset: %s", XSYMBOL (charset)->name->data); |
| 1231 | if (NILP (unify_map)) |
| 1232 | unify_map = CHARSET_UNIFY_MAP (cs); |
| 1233 | if (STRINGP (unify_map)) |
| 1234 | load_charset_map_from_file (cs, unify_map, 2); |
| 1235 | else if (VECTORP (unify_map)) |
| 1236 | load_charset_map_from_vector (cs, unify_map, 2); |
| 1237 | else if (NILP (unify_map)) |
| 1238 | error ("No unify-map for charset"); |
| 1239 | else |
| 1240 | error ("Bad unify-map arg"); |
| 1241 | CHARSET_UNIFIED_P (cs) = 1; |
| 1242 | } |
| 1243 | else if (CHAR_TABLE_P (Vchar_unify_table)) |
| 1244 | { |
| 1245 | int min_code = CHARSET_MIN_CODE (cs); |
| 1246 | int max_code = CHARSET_MAX_CODE (cs); |
| 1247 | int min_char = DECODE_CHAR (cs, min_code); |
| 1248 | int max_char = DECODE_CHAR (cs, max_code); |
| 1249 | |
| 1250 | char_table_set_range (Vchar_unify_table, min_char, max_char, Qnil); |
| 1251 | } |
| 1252 | |
| 1253 | return Qnil; |
| 1254 | } |
| 1255 | |
| 1256 | DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char, |
| 1257 | Sget_unused_iso_final_char, 2, 2, 0, |
| 1258 | doc: /* |
| 1259 | Return an unsed ISO final char for a charset of DIMENISION and CHARS. |
| 1260 | DIMENSION is the number of bytes to represent a character: 1 or 2. |
| 1261 | CHARS is the number of characters in a dimension: 94 or 96. |
| 1262 | |
| 1263 | This final char is for private use, thus the range is `0' (48) .. `?' (63). |
| 1264 | If there's no unused final char for the specified kind of charset, |
| 1265 | return nil. */) |
| 1266 | (dimension, chars) |
| 1267 | Lisp_Object dimension, chars; |
| 1268 | { |
| 1269 | int final_char; |
| 1270 | |
| 1271 | CHECK_NUMBER (dimension); |
| 1272 | CHECK_NUMBER (chars); |
| 1273 | if (XINT (dimension) != 1 && XINT (dimension) != 2 && XINT (dimension) != 3) |
| 1274 | args_out_of_range_3 (dimension, make_number (1), make_number (3)); |
| 1275 | if (XINT (chars) != 94 && XINT (chars) != 96) |
| 1276 | args_out_of_range_3 (chars, make_number (94), make_number (96)); |
| 1277 | for (final_char = '0'; final_char <= '?'; final_char++) |
| 1278 | if (ISO_CHARSET_TABLE (XINT (dimension), XINT (chars), final_char) < 0) |
| 1279 | break; |
| 1280 | return (final_char <= '?' ? make_number (final_char) : Qnil); |
| 1281 | } |
| 1282 | |
| 1283 | static void |
| 1284 | check_iso_charset_parameter (dimension, chars, final_char) |
| 1285 | Lisp_Object dimension, chars, final_char; |
| 1286 | { |
| 1287 | CHECK_NATNUM (dimension); |
| 1288 | CHECK_NATNUM (chars); |
| 1289 | CHECK_NATNUM (final_char); |
| 1290 | |
| 1291 | if (XINT (dimension) > 3) |
| 1292 | error ("Invalid DIMENSION %d, it should be 1, 2, or 3", XINT (dimension)); |
| 1293 | if (XINT (chars) != 94 && XINT (chars) != 96) |
| 1294 | error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); |
| 1295 | if (XINT (final_char) < '0' || XINT (final_char) > '~') |
| 1296 | error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); |
| 1297 | } |
| 1298 | |
| 1299 | |
| 1300 | DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, |
| 1301 | 4, 4, 0, |
| 1302 | doc: /* |
| 1303 | Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET. |
| 1304 | CHARSET should be defined by `define-charset' in advance. */) |
| 1305 | (dimension, chars, final_char, charset) |
| 1306 | Lisp_Object dimension, chars, final_char, charset; |
| 1307 | { |
| 1308 | int id; |
| 1309 | |
| 1310 | CHECK_CHARSET_GET_ID (charset, id); |
| 1311 | check_iso_charset_parameter (dimension, chars, final_char); |
| 1312 | |
| 1313 | ISO_CHARSET_TABLE (XINT (dimension), XINT (chars), XINT (final_char)) = id; |
| 1314 | return Qnil; |
| 1315 | } |
| 1316 | |
| 1317 | |
| 1318 | /* Return information about charsets in the text at PTR of NBYTES |
| 1319 | bytes, which are NCHARS characters. The value is: |
| 1320 | |
| 1321 | 0: Each character is represented by one byte. This is always |
| 1322 | true for a unibyte string. For a multibyte string, true if |
| 1323 | it contains only ASCII characters. |
| 1324 | |
| 1325 | 1: No charsets other than ascii, control-1, and latin-1 are |
| 1326 | found. |
| 1327 | |
| 1328 | 2: Otherwise. |
| 1329 | */ |
| 1330 | |
| 1331 | int |
| 1332 | string_xstring_p (string) |
| 1333 | Lisp_Object string; |
| 1334 | { |
| 1335 | const unsigned char *p = XSTRING (string)->data; |
| 1336 | const unsigned char *endp = p + STRING_BYTES (XSTRING (string)); |
| 1337 | struct charset *charset; |
| 1338 | |
| 1339 | if (XSTRING (string)->size == STRING_BYTES (XSTRING (string))) |
| 1340 | return 0; |
| 1341 | |
| 1342 | charset = CHARSET_FROM_ID (charset_iso_8859_1); |
| 1343 | while (p < endp) |
| 1344 | { |
| 1345 | int c = STRING_CHAR_ADVANCE (p); |
| 1346 | |
| 1347 | /* Fixme: comparison of unsigned expression < 0 is always false */ |
| 1348 | if (ENCODE_CHAR (charset, c) < 0) |
| 1349 | return 2; |
| 1350 | } |
| 1351 | return 1; |
| 1352 | } |
| 1353 | |
| 1354 | |
| 1355 | /* Find charsets in the string at PTR of NCHARS and NBYTES. |
| 1356 | |
| 1357 | CHARSETS is a vector. Each element is a cons of CHARSET and |
| 1358 | FOUND-FLAG. CHARSET is a charset id, and FOUND-FLAG is nil or t. |
| 1359 | FOUND-FLAG t (or nil) means that the corresponding charset is |
| 1360 | already found (or not yet found). |
| 1361 | |
| 1362 | It may lookup a translation table TABLE if supplied. */ |
| 1363 | |
| 1364 | static void |
| 1365 | find_charsets_in_text (ptr, nchars, nbytes, charsets, table) |
| 1366 | const unsigned char *ptr; |
| 1367 | int nchars, nbytes; |
| 1368 | Lisp_Object charsets, table; |
| 1369 | { |
| 1370 | const unsigned char *pend = ptr + nbytes; |
| 1371 | int ncharsets = ASIZE (charsets); |
| 1372 | |
| 1373 | if (nchars == nbytes) |
| 1374 | return; |
| 1375 | |
| 1376 | while (ptr < pend) |
| 1377 | { |
| 1378 | int c = STRING_CHAR_ADVANCE (ptr); |
| 1379 | int i; |
| 1380 | int all_found = 1; |
| 1381 | Lisp_Object elt; |
| 1382 | |
| 1383 | if (!NILP (table)) |
| 1384 | c = translate_char (table, c); |
| 1385 | for (i = 0; i < ncharsets; i++) |
| 1386 | { |
| 1387 | elt = AREF (charsets, i); |
| 1388 | if (NILP (XCDR (elt))) |
| 1389 | { |
| 1390 | struct charset *charset = CHARSET_FROM_ID (XINT (XCAR (elt))); |
| 1391 | |
| 1392 | if (ENCODE_CHAR (charset, c) != CHARSET_INVALID_CODE (charset)) |
| 1393 | XCDR (elt) = Qt; |
| 1394 | else |
| 1395 | all_found = 0; |
| 1396 | } |
| 1397 | } |
| 1398 | if (all_found) |
| 1399 | break; |
| 1400 | } |
| 1401 | } |
| 1402 | |
| 1403 | /* Fixme: returns nil for unibyte. */ |
| 1404 | DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, |
| 1405 | 2, 3, 0, |
| 1406 | doc: /* Return a list of charsets in the region between BEG and END. |
| 1407 | BEG and END are buffer positions. |
| 1408 | Optional arg TABLE if non-nil is a translation table to look up. |
| 1409 | |
| 1410 | If the current buffer is unibyte, the returned list may contain |
| 1411 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) |
| 1412 | (beg, end, table) |
| 1413 | Lisp_Object beg, end, table; |
| 1414 | { |
| 1415 | Lisp_Object charsets; |
| 1416 | int from, from_byte, to, stop, stop_byte, i; |
| 1417 | Lisp_Object val; |
| 1418 | |
| 1419 | validate_region (&beg, &end); |
| 1420 | from = XFASTINT (beg); |
| 1421 | stop = to = XFASTINT (end); |
| 1422 | |
| 1423 | if (from < GPT && GPT < to) |
| 1424 | { |
| 1425 | stop = GPT; |
| 1426 | stop_byte = GPT_BYTE; |
| 1427 | } |
| 1428 | else |
| 1429 | stop_byte = CHAR_TO_BYTE (stop); |
| 1430 | |
| 1431 | from_byte = CHAR_TO_BYTE (from); |
| 1432 | |
| 1433 | charsets = Fmake_vector (make_number (charset_table_used), Qnil); |
| 1434 | for (i = 0; i < charset_table_used; i++) |
| 1435 | ASET (charsets, i, Fcons (make_number (i), Qnil)); |
| 1436 | |
| 1437 | while (1) |
| 1438 | { |
| 1439 | find_charsets_in_text (BYTE_POS_ADDR (from_byte), stop - from, |
| 1440 | stop_byte - from_byte, charsets, table); |
| 1441 | if (stop < to) |
| 1442 | { |
| 1443 | from = stop, from_byte = stop_byte; |
| 1444 | stop = to, stop_byte = CHAR_TO_BYTE (stop); |
| 1445 | } |
| 1446 | else |
| 1447 | break; |
| 1448 | } |
| 1449 | |
| 1450 | val = Qnil; |
| 1451 | for (i = charset_table_used - 1; i >= 0; i--) |
| 1452 | if (!NILP (XCDR (AREF (charsets, i)))) |
| 1453 | val = Fcons (CHARSET_NAME (charset_table + i), val); |
| 1454 | return val; |
| 1455 | } |
| 1456 | |
| 1457 | /* Fixme: returns nil for unibyte. */ |
| 1458 | DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, |
| 1459 | 1, 2, 0, |
| 1460 | doc: /* Return a list of charsets in STR. |
| 1461 | Optional arg TABLE if non-nil is a translation table to look up. |
| 1462 | |
| 1463 | If STR is unibyte, the returned list may contain |
| 1464 | only `ascii', `eight-bit-control', and `eight-bit-graphic'. */) |
| 1465 | (str, table) |
| 1466 | Lisp_Object str, table; |
| 1467 | { |
| 1468 | Lisp_Object charsets; |
| 1469 | int i; |
| 1470 | Lisp_Object val; |
| 1471 | |
| 1472 | CHECK_STRING (str); |
| 1473 | |
| 1474 | charsets = Fmake_vector (make_number (charset_table_used), Qnil); |
| 1475 | for (i = 0; i < charset_table_used; i++) |
| 1476 | ASET (charsets, i, Fcons (make_number (i), Qnil)); |
| 1477 | find_charsets_in_text (XSTRING (str)->data, XSTRING (str)->size, |
| 1478 | STRING_BYTES (XSTRING (str)), charsets, table); |
| 1479 | |
| 1480 | val = Qnil; |
| 1481 | for (i = charset_table_used - 1; i >= 0; i--) |
| 1482 | if (!NILP (XCDR (AREF (charsets, i)))) |
| 1483 | val = Fcons (CHARSET_NAME (charset_table + i), val); |
| 1484 | return val; |
| 1485 | } |
| 1486 | |
| 1487 | \f |
| 1488 | |
| 1489 | /* Return a character correponding to the code-point CODE of |
| 1490 | CHARSET. */ |
| 1491 | |
| 1492 | int |
| 1493 | decode_char (charset, code) |
| 1494 | struct charset *charset; |
| 1495 | unsigned code; |
| 1496 | { |
| 1497 | int c, char_index; |
| 1498 | enum charset_method method = CHARSET_METHOD (charset); |
| 1499 | |
| 1500 | if (code < CHARSET_MIN_CODE (charset) || code > CHARSET_MAX_CODE (charset)) |
| 1501 | return -1; |
| 1502 | |
| 1503 | if (method == CHARSET_METHOD_MAP_DEFERRED) |
| 1504 | { |
| 1505 | load_charset (charset); |
| 1506 | method = CHARSET_METHOD (charset); |
| 1507 | } |
| 1508 | |
| 1509 | if (method == CHARSET_METHOD_SUBSET) |
| 1510 | { |
| 1511 | Lisp_Object subset_info; |
| 1512 | |
| 1513 | subset_info = CHARSET_SUBSET (charset); |
| 1514 | charset = CHARSET_FROM_ID (XFASTINT (AREF (subset_info, 0))); |
| 1515 | code -= XINT (AREF (subset_info, 3)); |
| 1516 | if (code < XFASTINT (AREF (subset_info, 1)) |
| 1517 | || code > XFASTINT (AREF (subset_info, 2))) |
| 1518 | c = -1; |
| 1519 | else |
| 1520 | c = DECODE_CHAR (charset, code); |
| 1521 | } |
| 1522 | else if (method == CHARSET_METHOD_SUPERSET) |
| 1523 | { |
| 1524 | Lisp_Object parents; |
| 1525 | |
| 1526 | parents = CHARSET_SUPERSET (charset); |
| 1527 | c = -1; |
| 1528 | for (; CONSP (parents); parents = XCDR (parents)) |
| 1529 | { |
| 1530 | int id = XINT (XCAR (XCAR (parents))); |
| 1531 | int code_offset = XINT (XCDR (XCAR (parents))); |
| 1532 | unsigned this_code = code - code_offset; |
| 1533 | |
| 1534 | charset = CHARSET_FROM_ID (id); |
| 1535 | if ((c = DECODE_CHAR (charset, this_code)) >= 0) |
| 1536 | break; |
| 1537 | } |
| 1538 | } |
| 1539 | else |
| 1540 | { |
| 1541 | char_index = CODE_POINT_TO_INDEX (charset, code); |
| 1542 | if (char_index < 0) |
| 1543 | return -1; |
| 1544 | |
| 1545 | if (method == CHARSET_METHOD_MAP) |
| 1546 | { |
| 1547 | Lisp_Object decoder; |
| 1548 | |
| 1549 | decoder = CHARSET_DECODER (charset); |
| 1550 | if (! VECTORP (decoder)) |
| 1551 | return -1; |
| 1552 | c = XINT (AREF (decoder, char_index)); |
| 1553 | } |
| 1554 | else |
| 1555 | { |
| 1556 | c = char_index + CHARSET_CODE_OFFSET (charset); |
| 1557 | } |
| 1558 | } |
| 1559 | |
| 1560 | if (CHARSET_UNIFIED_P (charset) |
| 1561 | && c >= 0) |
| 1562 | { |
| 1563 | MAYBE_UNIFY_CHAR (c); |
| 1564 | } |
| 1565 | |
| 1566 | return c; |
| 1567 | } |
| 1568 | |
| 1569 | /* Variable used temporarily by the macro ENCODE_CHAR. */ |
| 1570 | Lisp_Object charset_work; |
| 1571 | |
| 1572 | /* Return a code-point of CHAR in CHARSET. If CHAR doesn't belong to |
| 1573 | CHARSET, return CHARSET_INVALID_CODE (CHARSET). If STRICT is true, |
| 1574 | use CHARSET's strict_max_char instead of max_char. */ |
| 1575 | |
| 1576 | unsigned |
| 1577 | encode_char (charset, c) |
| 1578 | struct charset *charset; |
| 1579 | int c; |
| 1580 | { |
| 1581 | unsigned code; |
| 1582 | enum charset_method method = CHARSET_METHOD (charset); |
| 1583 | |
| 1584 | if (CHARSET_UNIFIED_P (charset)) |
| 1585 | { |
| 1586 | Lisp_Object deunifier, deunified; |
| 1587 | |
| 1588 | deunifier = CHARSET_DEUNIFIER (charset); |
| 1589 | if (! CHAR_TABLE_P (deunifier)) |
| 1590 | { |
| 1591 | Funify_charset (CHARSET_NAME (charset), Qnil, Qnil); |
| 1592 | deunifier = CHARSET_DEUNIFIER (charset); |
| 1593 | } |
| 1594 | deunified = CHAR_TABLE_REF (deunifier, c); |
| 1595 | if (! NILP (deunified)) |
| 1596 | c = XINT (deunified); |
| 1597 | } |
| 1598 | |
| 1599 | if (! CHARSET_FAST_MAP_REF ((c), charset->fast_map) |
| 1600 | || c < CHARSET_MIN_CHAR (charset) || c > CHARSET_MAX_CHAR (charset)) |
| 1601 | return CHARSET_INVALID_CODE (charset); |
| 1602 | |
| 1603 | if (method == CHARSET_METHOD_SUBSET) |
| 1604 | { |
| 1605 | Lisp_Object subset_info; |
| 1606 | struct charset *this_charset; |
| 1607 | |
| 1608 | subset_info = CHARSET_SUBSET (charset); |
| 1609 | this_charset = CHARSET_FROM_ID (XFASTINT (AREF (subset_info, 0))); |
| 1610 | code = ENCODE_CHAR (this_charset, c); |
| 1611 | if (code == CHARSET_INVALID_CODE (this_charset) |
| 1612 | || code < XFASTINT (AREF (subset_info, 1)) |
| 1613 | || code > XFASTINT (AREF (subset_info, 2))) |
| 1614 | return CHARSET_INVALID_CODE (charset); |
| 1615 | code += XINT (AREF (subset_info, 3)); |
| 1616 | return code; |
| 1617 | } |
| 1618 | |
| 1619 | if (method == CHARSET_METHOD_SUPERSET) |
| 1620 | { |
| 1621 | Lisp_Object parents; |
| 1622 | |
| 1623 | parents = CHARSET_SUPERSET (charset); |
| 1624 | for (; CONSP (parents); parents = XCDR (parents)) |
| 1625 | { |
| 1626 | int id = XINT (XCAR (XCAR (parents))); |
| 1627 | int code_offset = XINT (XCDR (XCAR (parents))); |
| 1628 | struct charset *this_charset = CHARSET_FROM_ID (id); |
| 1629 | |
| 1630 | code = ENCODE_CHAR (this_charset, c); |
| 1631 | if (code != CHARSET_INVALID_CODE (this_charset)) |
| 1632 | return code + code_offset; |
| 1633 | } |
| 1634 | return CHARSET_INVALID_CODE (charset); |
| 1635 | } |
| 1636 | |
| 1637 | if (method == CHARSET_METHOD_MAP_DEFERRED) |
| 1638 | { |
| 1639 | load_charset (charset); |
| 1640 | method = CHARSET_METHOD (charset); |
| 1641 | } |
| 1642 | |
| 1643 | if (method == CHARSET_METHOD_MAP) |
| 1644 | { |
| 1645 | Lisp_Object encoder; |
| 1646 | Lisp_Object val; |
| 1647 | |
| 1648 | encoder = CHARSET_ENCODER (charset); |
| 1649 | if (! CHAR_TABLE_P (CHARSET_ENCODER (charset))) |
| 1650 | return CHARSET_INVALID_CODE (charset); |
| 1651 | val = CHAR_TABLE_REF (encoder, c); |
| 1652 | if (NILP (val)) |
| 1653 | return CHARSET_INVALID_CODE (charset); |
| 1654 | code = XINT (val); |
| 1655 | if (! CHARSET_COMPACT_CODES_P (charset)) |
| 1656 | code = INDEX_TO_CODE_POINT (charset, code); |
| 1657 | } |
| 1658 | else /* method == CHARSET_METHOD_OFFSET */ |
| 1659 | { |
| 1660 | code = c - CHARSET_CODE_OFFSET (charset); |
| 1661 | code = INDEX_TO_CODE_POINT (charset, code); |
| 1662 | } |
| 1663 | |
| 1664 | return code; |
| 1665 | } |
| 1666 | |
| 1667 | |
| 1668 | DEFUN ("decode-char", Fdecode_char, Sdecode_char, 2, 3, 0, |
| 1669 | doc: /* Decode the pair of CHARSET and CODE-POINT into a character. |
| 1670 | Return nil if CODE-POINT is not valid in CHARSET. |
| 1671 | |
| 1672 | CODE-POINT may be a cons (HIGHER-16-BIT-VALUE . LOWER-16-BIT-VALUE). |
| 1673 | |
| 1674 | Optional argument RESTRICTION specifies a way to map the pair of CCS |
| 1675 | and CODE-POINT to a chracter. Currently not supported and just ignored. */) |
| 1676 | (charset, code_point, restriction) |
| 1677 | Lisp_Object charset, code_point, restriction; |
| 1678 | { |
| 1679 | int c, id; |
| 1680 | unsigned code; |
| 1681 | struct charset *charsetp; |
| 1682 | |
| 1683 | CHECK_CHARSET_GET_ID (charset, id); |
| 1684 | if (CONSP (code_point)) |
| 1685 | { |
| 1686 | CHECK_NATNUM (XCAR (code_point)); |
| 1687 | CHECK_NATNUM (XCDR (code_point)); |
| 1688 | code = (XINT (XCAR (code_point)) << 16) | (XINT (XCDR (code_point))); |
| 1689 | } |
| 1690 | else |
| 1691 | { |
| 1692 | CHECK_NATNUM (code_point); |
| 1693 | code = XINT (code_point); |
| 1694 | } |
| 1695 | charsetp = CHARSET_FROM_ID (id); |
| 1696 | c = DECODE_CHAR (charsetp, code); |
| 1697 | return (c >= 0 ? make_number (c) : Qnil); |
| 1698 | } |
| 1699 | |
| 1700 | |
| 1701 | DEFUN ("encode-char", Fencode_char, Sencode_char, 2, 3, 0, |
| 1702 | doc: /* Encode the character CH into a code-point of CHARSET. |
| 1703 | Return nil if CHARSET doesn't include CH. |
| 1704 | |
| 1705 | Optional argument RESTRICTION specifies a way to map CHAR to a |
| 1706 | code-point in CCS. Currently not supported and just ignored. */) |
| 1707 | (ch, charset, restriction) |
| 1708 | Lisp_Object ch, charset, restriction; |
| 1709 | { |
| 1710 | int id; |
| 1711 | unsigned code; |
| 1712 | struct charset *charsetp; |
| 1713 | |
| 1714 | CHECK_CHARSET_GET_ID (charset, id); |
| 1715 | CHECK_NATNUM (ch); |
| 1716 | charsetp = CHARSET_FROM_ID (id); |
| 1717 | code = ENCODE_CHAR (charsetp, XINT (ch)); |
| 1718 | if (code == CHARSET_INVALID_CODE (charsetp)) |
| 1719 | return Qnil; |
| 1720 | if (code > 0x7FFFFFF) |
| 1721 | return Fcons (make_number (code >> 16), make_number (code & 0xFFFF)); |
| 1722 | return make_number (code); |
| 1723 | } |
| 1724 | |
| 1725 | |
| 1726 | DEFUN ("make-char", Fmake_char, Smake_char, 1, 5, 0, |
| 1727 | doc: |
| 1728 | /* Return a character of CHARSET whose position codes are CODEn. |
| 1729 | |
| 1730 | CODE1 through CODE4 are optional, but if you don't supply sufficient |
| 1731 | position codes, it is assumed that the minimum code in each dimension |
| 1732 | is specified. */) |
| 1733 | (charset, code1, code2, code3, code4) |
| 1734 | Lisp_Object charset, code1, code2, code3, code4; |
| 1735 | { |
| 1736 | int id, dimension; |
| 1737 | struct charset *charsetp; |
| 1738 | unsigned code; |
| 1739 | int c; |
| 1740 | |
| 1741 | CHECK_CHARSET_GET_ID (charset, id); |
| 1742 | charsetp = CHARSET_FROM_ID (id); |
| 1743 | |
| 1744 | dimension = CHARSET_DIMENSION (charsetp); |
| 1745 | if (NILP (code1)) |
| 1746 | code = (CHARSET_ASCII_COMPATIBLE_P (charsetp) |
| 1747 | ? 0 : CHARSET_MIN_CODE (charsetp)); |
| 1748 | else |
| 1749 | { |
| 1750 | CHECK_NATNUM (code1); |
| 1751 | if (XFASTINT (code1) >= 0x100) |
| 1752 | args_out_of_range (make_number (0xFF), code1); |
| 1753 | code = XFASTINT (code1); |
| 1754 | |
| 1755 | if (dimension > 1) |
| 1756 | { |
| 1757 | code <<= 8; |
| 1758 | if (NILP (code2)) |
| 1759 | code |= charsetp->code_space[(dimension - 2) * 4]; |
| 1760 | else |
| 1761 | { |
| 1762 | CHECK_NATNUM (code2); |
| 1763 | if (XFASTINT (code2) >= 0x100) |
| 1764 | args_out_of_range (make_number (0xFF), code2); |
| 1765 | code |= XFASTINT (code2); |
| 1766 | } |
| 1767 | |
| 1768 | if (dimension > 2) |
| 1769 | { |
| 1770 | code <<= 8; |
| 1771 | if (NILP (code3)) |
| 1772 | code |= charsetp->code_space[(dimension - 3) * 4]; |
| 1773 | else |
| 1774 | { |
| 1775 | CHECK_NATNUM (code3); |
| 1776 | if (XFASTINT (code3) >= 0x100) |
| 1777 | args_out_of_range (make_number (0xFF), code3); |
| 1778 | code |= XFASTINT (code3); |
| 1779 | } |
| 1780 | |
| 1781 | if (dimension > 3) |
| 1782 | { |
| 1783 | code <<= 8; |
| 1784 | if (NILP (code4)) |
| 1785 | code |= charsetp->code_space[0]; |
| 1786 | else |
| 1787 | { |
| 1788 | CHECK_NATNUM (code4); |
| 1789 | if (XFASTINT (code4) >= 0x100) |
| 1790 | args_out_of_range (make_number (0xFF), code4); |
| 1791 | code |= XFASTINT (code4); |
| 1792 | } |
| 1793 | } |
| 1794 | } |
| 1795 | } |
| 1796 | } |
| 1797 | |
| 1798 | if (CHARSET_ISO_FINAL (charsetp) >= 0) |
| 1799 | code &= 0x7F7F7F7F; |
| 1800 | c = DECODE_CHAR (charsetp, code); |
| 1801 | if (c < 0) |
| 1802 | error ("Invalid code(s)"); |
| 1803 | return make_number (c); |
| 1804 | } |
| 1805 | |
| 1806 | |
| 1807 | /* Return the first charset in CHARSET_LIST that contains C. |
| 1808 | CHARSET_LIST is a list of charset IDs. If it is nil, use |
| 1809 | Vcharset_ordered_list. */ |
| 1810 | |
| 1811 | struct charset * |
| 1812 | char_charset (c, charset_list, code_return) |
| 1813 | int c; |
| 1814 | Lisp_Object charset_list; |
| 1815 | unsigned *code_return; |
| 1816 | { |
| 1817 | if (NILP (charset_list)) |
| 1818 | charset_list = Vcharset_ordered_list; |
| 1819 | |
| 1820 | while (CONSP (charset_list)) |
| 1821 | { |
| 1822 | struct charset *charset = CHARSET_FROM_ID (XINT (XCAR (charset_list))); |
| 1823 | unsigned code = ENCODE_CHAR (charset, c); |
| 1824 | |
| 1825 | if (code != CHARSET_INVALID_CODE (charset)) |
| 1826 | { |
| 1827 | if (code_return) |
| 1828 | *code_return = code; |
| 1829 | return charset; |
| 1830 | } |
| 1831 | charset_list = XCDR (charset_list); |
| 1832 | } |
| 1833 | return NULL; |
| 1834 | } |
| 1835 | |
| 1836 | |
| 1837 | /* Fixme: `unknown' can't happen now? */ |
| 1838 | DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, |
| 1839 | doc: /*Return list of charset and one to three position-codes of CHAR. |
| 1840 | If CHAR is invalid as a character code, return a list `(unknown CHAR)'. */) |
| 1841 | (ch) |
| 1842 | Lisp_Object ch; |
| 1843 | { |
| 1844 | struct charset *charset; |
| 1845 | int c, dimension; |
| 1846 | unsigned code; |
| 1847 | Lisp_Object val; |
| 1848 | |
| 1849 | CHECK_CHARACTER (ch); |
| 1850 | c = XFASTINT (ch); |
| 1851 | charset = CHAR_CHARSET (c); |
| 1852 | if (! charset) |
| 1853 | return Fcons (intern ("unknown"), Fcons (ch, Qnil)); |
| 1854 | |
| 1855 | code = ENCODE_CHAR (charset, c); |
| 1856 | if (code == CHARSET_INVALID_CODE (charset)) |
| 1857 | abort (); |
| 1858 | dimension = CHARSET_DIMENSION (charset); |
| 1859 | val = (dimension == 1 ? Fcons (make_number (code), Qnil) |
| 1860 | : dimension == 2 ? Fcons (make_number (code >> 8), |
| 1861 | Fcons (make_number (code & 0xFF), Qnil)) |
| 1862 | : Fcons (make_number (code >> 16), |
| 1863 | Fcons (make_number ((code >> 8) & 0xFF), |
| 1864 | Fcons (make_number (code & 0xFF), Qnil)))); |
| 1865 | return Fcons (CHARSET_NAME (charset), val); |
| 1866 | } |
| 1867 | |
| 1868 | |
| 1869 | DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, |
| 1870 | doc: /* Return the charset of highest priority that contains CHAR. */) |
| 1871 | (ch) |
| 1872 | Lisp_Object ch; |
| 1873 | { |
| 1874 | struct charset *charset; |
| 1875 | |
| 1876 | CHECK_CHARACTER (ch); |
| 1877 | charset = CHAR_CHARSET (XINT (ch)); |
| 1878 | return (CHARSET_NAME (charset)); |
| 1879 | } |
| 1880 | |
| 1881 | |
| 1882 | DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0, |
| 1883 | doc: /* |
| 1884 | Return charset of a character in the current buffer at position POS. |
| 1885 | If POS is nil, it defauls to the current point. |
| 1886 | If POS is out of range, the value is nil. */) |
| 1887 | (pos) |
| 1888 | Lisp_Object pos; |
| 1889 | { |
| 1890 | Lisp_Object ch; |
| 1891 | struct charset *charset; |
| 1892 | |
| 1893 | ch = Fchar_after (pos); |
| 1894 | if (! INTEGERP (ch)) |
| 1895 | return ch; |
| 1896 | charset = CHAR_CHARSET (XINT (ch)); |
| 1897 | return (CHARSET_NAME (charset)); |
| 1898 | } |
| 1899 | |
| 1900 | |
| 1901 | DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, |
| 1902 | doc: /* |
| 1903 | Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR. |
| 1904 | |
| 1905 | ISO 2022's designation sequence (escape sequence) distinguishes charsets |
| 1906 | by their DIMENSION, CHARS, and FINAL-CHAR, |
| 1907 | where as Emacs distinguishes them by charset symbol. |
| 1908 | See the documentation of the function `charset-info' for the meanings of |
| 1909 | DIMENSION, CHARS, and FINAL-CHAR. */) |
| 1910 | (dimension, chars, final_char) |
| 1911 | Lisp_Object dimension, chars, final_char; |
| 1912 | { |
| 1913 | int id; |
| 1914 | |
| 1915 | check_iso_charset_parameter (dimension, chars, final_char); |
| 1916 | id = ISO_CHARSET_TABLE (XFASTINT (dimension), XFASTINT (chars), |
| 1917 | XFASTINT (final_char)); |
| 1918 | return (id >= 0 ? CHARSET_NAME (CHARSET_FROM_ID (id)) : Qnil); |
| 1919 | } |
| 1920 | |
| 1921 | |
| 1922 | DEFUN ("clear-charset-maps", Fclear_charset_maps, Sclear_charset_maps, |
| 1923 | 0, 0, 0, |
| 1924 | doc: /* |
| 1925 | Clear encoder and decoder of charsets that are loaded from mapfiles. */) |
| 1926 | () |
| 1927 | { |
| 1928 | int i; |
| 1929 | struct charset *charset; |
| 1930 | Lisp_Object attrs; |
| 1931 | |
| 1932 | for (i = 0; i < charset_table_used; i++) |
| 1933 | { |
| 1934 | charset = CHARSET_FROM_ID (i); |
| 1935 | attrs = CHARSET_ATTRIBUTES (charset); |
| 1936 | |
| 1937 | if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP) |
| 1938 | { |
| 1939 | CHARSET_ATTR_DECODER (attrs) = Qnil; |
| 1940 | CHARSET_ATTR_ENCODER (attrs) = Qnil; |
| 1941 | CHARSET_METHOD (charset) = CHARSET_METHOD_MAP_DEFERRED; |
| 1942 | } |
| 1943 | |
| 1944 | if (CHARSET_UNIFIED_P (charset)) |
| 1945 | CHARSET_ATTR_DEUNIFIER (attrs) = Qnil; |
| 1946 | } |
| 1947 | |
| 1948 | if (CHAR_TABLE_P (Vchar_unified_charset_table)) |
| 1949 | { |
| 1950 | Foptimize_char_table (Vchar_unified_charset_table); |
| 1951 | Vchar_unify_table = Vchar_unified_charset_table; |
| 1952 | Vchar_unified_charset_table = Qnil; |
| 1953 | } |
| 1954 | |
| 1955 | return Qnil; |
| 1956 | } |
| 1957 | |
| 1958 | DEFUN ("charset-priority-list", Fcharset_priority_list, |
| 1959 | Scharset_priority_list, 0, 1, 0, |
| 1960 | doc: /* Return the list of charsets ordered by priority. |
| 1961 | HIGHESTP non-nil means just return the highest priority one. */) |
| 1962 | (highestp) |
| 1963 | Lisp_Object highestp; |
| 1964 | { |
| 1965 | Lisp_Object val = Qnil, list = Vcharset_ordered_list; |
| 1966 | |
| 1967 | if (!NILP (highestp)) |
| 1968 | return CHARSET_NAME (CHARSET_FROM_ID (XINT (Fcar (list)))); |
| 1969 | |
| 1970 | while (!NILP (list)) |
| 1971 | { |
| 1972 | val = Fcons (CHARSET_NAME (CHARSET_FROM_ID (XINT (XCAR (list)))), val); |
| 1973 | list = XCDR (list); |
| 1974 | } |
| 1975 | return Fnreverse (val); |
| 1976 | } |
| 1977 | |
| 1978 | DEFUN ("set-charset-priority", Fset_charset_priority, Sset_charset_priority, |
| 1979 | 1, MANY, 0, |
| 1980 | doc: /* Assign higher priority to the charsets given as arguments. |
| 1981 | usage: (set-charset-priority &rest charsets) */) |
| 1982 | (nargs, args) |
| 1983 | int nargs; |
| 1984 | Lisp_Object *args; |
| 1985 | { |
| 1986 | Lisp_Object new_head = Qnil, old_list, arglist[2]; |
| 1987 | int i, id; |
| 1988 | |
| 1989 | old_list = Fcopy_sequence (Vcharset_ordered_list); |
| 1990 | for (i = 0; i < nargs; i++) |
| 1991 | { |
| 1992 | CHECK_CHARSET_GET_ID (args[i], id); |
| 1993 | old_list = Fdelq (make_number (id), old_list); |
| 1994 | new_head = Fcons (make_number (id), new_head); |
| 1995 | } |
| 1996 | arglist[0] = Fnreverse (new_head); |
| 1997 | arglist[1] = old_list; |
| 1998 | Vcharset_ordered_list = Fnconc (2, arglist); |
| 1999 | charset_ordered_list_tick++; |
| 2000 | return Qnil; |
| 2001 | } |
| 2002 | \f |
| 2003 | void |
| 2004 | init_charset () |
| 2005 | { |
| 2006 | |
| 2007 | } |
| 2008 | |
| 2009 | |
| 2010 | void |
| 2011 | init_charset_once () |
| 2012 | { |
| 2013 | int i, j, k; |
| 2014 | |
| 2015 | for (i = 0; i < ISO_MAX_DIMENSION; i++) |
| 2016 | for (j = 0; j < ISO_MAX_CHARS; j++) |
| 2017 | for (k = 0; k < ISO_MAX_FINAL; k++) |
| 2018 | iso_charset_table[i][j][k] = -1; |
| 2019 | |
| 2020 | for (i = 0; i < 255; i++) |
| 2021 | emacs_mule_charset[i] = NULL; |
| 2022 | |
| 2023 | charset_jisx0201_roman = -1; |
| 2024 | charset_jisx0208_1978 = -1; |
| 2025 | charset_jisx0208 = -1; |
| 2026 | |
| 2027 | #if 0 |
| 2028 | Vchar_charset_set = Fmake_char_table (Qnil, Qnil); |
| 2029 | CHAR_TABLE_SET (Vchar_charset_set, make_number (97), Qnil); |
| 2030 | |
| 2031 | DEFSYM (Qcharset_encode_table, "charset-encode-table"); |
| 2032 | |
| 2033 | /* Intern this now in case it isn't already done. |
| 2034 | Setting this variable twice is harmless. |
| 2035 | But don't staticpro it here--that is done in alloc.c. */ |
| 2036 | Qchar_table_extra_slots = intern ("char-table-extra-slots"); |
| 2037 | |
| 2038 | /* Now we are ready to set up this property, so we can create syntax |
| 2039 | tables. */ |
| 2040 | Fput (Qcharset_encode_table, Qchar_table_extra_slots, make_number (0)); |
| 2041 | #endif |
| 2042 | } |
| 2043 | |
| 2044 | #ifdef emacs |
| 2045 | |
| 2046 | void |
| 2047 | syms_of_charset () |
| 2048 | { |
| 2049 | char *p; |
| 2050 | |
| 2051 | DEFSYM (Qcharsetp, "charsetp"); |
| 2052 | |
| 2053 | DEFSYM (Qascii, "ascii"); |
| 2054 | DEFSYM (Qunicode, "unicode"); |
| 2055 | DEFSYM (Qeight_bit, "eight-bit"); |
| 2056 | DEFSYM (Qiso_8859_1, "iso-8859-1"); |
| 2057 | |
| 2058 | DEFSYM (Qgl, "gl"); |
| 2059 | DEFSYM (Qgr, "gr"); |
| 2060 | |
| 2061 | p = (char *) xmalloc (30000); |
| 2062 | |
| 2063 | staticpro (&Vcharset_ordered_list); |
| 2064 | Vcharset_ordered_list = Qnil; |
| 2065 | |
| 2066 | staticpro (&Viso_2022_charset_list); |
| 2067 | Viso_2022_charset_list = Qnil; |
| 2068 | |
| 2069 | staticpro (&Vemacs_mule_charset_list); |
| 2070 | Vemacs_mule_charset_list = Qnil; |
| 2071 | |
| 2072 | staticpro (&Vcharset_hash_table); |
| 2073 | Vcharset_hash_table = Fmakehash (Qeq); |
| 2074 | |
| 2075 | charset_table_size = 128; |
| 2076 | charset_table = ((struct charset *) |
| 2077 | xmalloc (sizeof (struct charset) * charset_table_size)); |
| 2078 | charset_table_used = 0; |
| 2079 | |
| 2080 | staticpro (&Vchar_unified_charset_table); |
| 2081 | Vchar_unified_charset_table = Fmake_char_table (Qnil, make_number (-1)); |
| 2082 | |
| 2083 | defsubr (&Scharsetp); |
| 2084 | defsubr (&Smap_charset_chars); |
| 2085 | defsubr (&Sdefine_charset_internal); |
| 2086 | defsubr (&Sdefine_charset_alias); |
| 2087 | defsubr (&Sprimary_charset); |
| 2088 | defsubr (&Sset_primary_charset); |
| 2089 | defsubr (&Scharset_plist); |
| 2090 | defsubr (&Sset_charset_plist); |
| 2091 | defsubr (&Sunify_charset); |
| 2092 | defsubr (&Sget_unused_iso_final_char); |
| 2093 | defsubr (&Sdeclare_equiv_charset); |
| 2094 | defsubr (&Sfind_charset_region); |
| 2095 | defsubr (&Sfind_charset_string); |
| 2096 | defsubr (&Sdecode_char); |
| 2097 | defsubr (&Sencode_char); |
| 2098 | defsubr (&Ssplit_char); |
| 2099 | defsubr (&Smake_char); |
| 2100 | defsubr (&Schar_charset); |
| 2101 | defsubr (&Scharset_after); |
| 2102 | defsubr (&Siso_charset); |
| 2103 | defsubr (&Sclear_charset_maps); |
| 2104 | defsubr (&Scharset_priority_list); |
| 2105 | defsubr (&Sset_charset_priority); |
| 2106 | |
| 2107 | DEFVAR_LISP ("charset-map-directory", &Vcharset_map_directory, |
| 2108 | doc: /* Directory of charset map files that come with GNU Emacs. |
| 2109 | The default value is sub-directory "charsets" of `data-directory'. */); |
| 2110 | Vcharset_map_directory = Fexpand_file_name (build_string ("charsets"), |
| 2111 | Vdata_directory); |
| 2112 | |
| 2113 | DEFVAR_LISP ("charset-list", &Vcharset_list, |
| 2114 | doc: /* List of all charsets ever defined. */); |
| 2115 | Vcharset_list = Qnil; |
| 2116 | |
| 2117 | charset_ascii |
| 2118 | = define_charset_internal (Qascii, 1, "\x00\x7F\x00\x00\x00\x00", |
| 2119 | 0, 127, 'B', -1, 0, 1, 0, 0); |
| 2120 | charset_unicode |
| 2121 | = define_charset_internal (Qunicode, 3, "\x00\xFF\x00\xFF\x00\x10", |
| 2122 | 0, MAX_UNICODE_CHAR, -1, 0, -1, 1, 0, 0); |
| 2123 | charset_eight_bit |
| 2124 | = define_charset_internal (Qeight_bit, 1, "\x80\xFF\x00\x00\x00\x00", |
| 2125 | 128, 255, -1, 0, -1, 0, 0, |
| 2126 | MAX_5_BYTE_CHAR + 1); |
| 2127 | } |
| 2128 | |
| 2129 | #endif /* emacs */ |