Merge changes from emacs-23 branch
[emacs.git] / src / charset.h
blob1fc552a5bd38eb2ff66e05977839d6825c590553
1 /* Header for charset handler.
2 Copyright (C) 2001-2011 Free Software Foundation, Inc.
3 Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
4 2005, 2006, 2007, 2008, 2009, 2010, 2011
5 National Institute of Advanced Industrial Science and Technology (AIST)
6 Registration Number H14PRO021
8 Copyright (C) 2003
9 National Institute of Advanced Industrial Science and Technology (AIST)
10 Registration Number H13PRO009
12 This file is part of GNU Emacs.
14 GNU Emacs is free software: you can redistribute it and/or modify
15 it under the terms of the GNU General Public License as published by
16 the Free Software Foundation, either version 3 of the License, or
17 (at your option) any later version.
19 GNU Emacs is distributed in the hope that it will be useful,
20 but WITHOUT ANY WARRANTY; without even the implied warranty of
21 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 GNU General Public License for more details.
24 You should have received a copy of the GNU General Public License
25 along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
27 #ifndef EMACS_CHARSET_H
28 #define EMACS_CHARSET_H
30 /* Index to arguments of Fdefine_charset_internal. */
32 enum define_charset_arg_index
34 charset_arg_name,
35 charset_arg_dimension,
36 charset_arg_code_space,
37 charset_arg_min_code,
38 charset_arg_max_code,
39 charset_arg_iso_final,
40 charset_arg_iso_revision,
41 charset_arg_emacs_mule_id,
42 charset_arg_ascii_compatible_p,
43 charset_arg_supplementary_p,
44 charset_arg_invalid_code,
45 charset_arg_code_offset,
46 charset_arg_map,
47 charset_arg_subset,
48 charset_arg_superset,
49 charset_arg_unify_map,
50 charset_arg_plist,
51 charset_arg_max
55 /* Indices to charset attributes vector. */
57 enum charset_attr_index
59 /* ID number of the charset. */
60 charset_id,
62 /* Name of the charset (symbol). */
63 charset_name,
65 /* Property list of the charset. */
66 charset_plist,
68 /* If the method of the charset is `MAP', the value is a mapping
69 vector or a file name that contains mapping vector. Otherwise,
70 nil. */
71 charset_map,
73 /* If the method of the charset is `MAP', the value is a vector
74 that maps code points of the charset to characters. The vector
75 is indexed by a character index. A character index is
76 calculated from a code point and the code-space table of the
77 charset. */
78 charset_decoder,
80 /* If the method of the charset is `MAP', the value is a
81 char-table that maps characters of the charset to code
82 points. */
83 charset_encoder,
85 /* If the method of the charset is `SUBSET', the value is a vector
86 that has this form:
88 [ CHARSET-ID MIN-CODE MAX-CODE OFFSET ]
90 CHARSET-ID is an ID number of a parent charset. MIN-CODE and
91 MAX-CODE specify the range of characters inherited from the
92 parent. OFFSET is an integer value to add to a code point of
93 the parent charset to get the corresponding code point of this
94 charset. */
95 charset_subset,
97 /* If the method of the charset is `SUPERSET', the value is a list
98 whose elements have this form:
100 (CHARSET-ID . OFFSET)
102 CHARSET-IDs are ID numbers of parent charsets. OFFSET is an
103 integer value to add to a code point of the parent charset to
104 get the corresponding code point of this charset. */
105 charset_superset,
107 /* The value is a mapping vector or a file name that contains the
108 mapping. This defines how characters in the charset should be
109 unified with Unicode. The value of the member
110 `charset_deunifier' is created from this information. */
111 charset_unify_map,
113 /* If characters in the charset must be unified Unicode, the value
114 is a char table that maps a unified Unicode character code to
115 the non-unified character code in the charset. */
116 charset_deunifier,
118 /* The length of the charset attribute vector. */
119 charset_attr_max
122 /* Methods for converting code points and characters of charsets. */
124 enum charset_method
126 /* For a charset of this method, a character code is calculated
127 from a character index (which is calculated from a code point)
128 simply by adding an offset value. */
129 CHARSET_METHOD_OFFSET,
131 /* For a charset of this method, a decoder vector and an encoder
132 char-table is used for code point <-> character code
133 conversion. */
134 CHARSET_METHOD_MAP,
136 /* A charset of this method is a subset of another charset. */
137 CHARSET_METHOD_SUBSET,
139 /* A charset of this method is a superset of other charsets. */
140 CHARSET_METHOD_SUPERSET
143 struct charset
145 /* Index to charset_table. */
146 int id;
148 /* Index to Vcharset_hash_table. */
149 int hash_index;
151 /* Dimension of the charset: 1, 2, 3, or 4. */
152 int dimension;
154 /* Byte code range of each dimension. <code_space>[4N] is a mininum
155 byte code of the (N+1)th dimension, <code_space>[4N+1] is a
156 maximum byte code of the (N+1)th dimension, <code_space>[4N+2] is
157 (<code_space>[4N+1] - <code_space>[4N] + 1), <code_space>[4N+3]
158 is a number of characters containd in the first to (N+1)th
159 dismesions. We get `char-index' of a `code-point' from this
160 information. */
161 int code_space[16];
163 /* If B is a byte of Nth dimension of a code-point, the (N-1)th bit
164 of code_space_mask[B] is set. This array is used to quickly
165 check if a code-point is in a valid range. */
166 unsigned char *code_space_mask;
168 /* 1 if there's no gap in code-points. */
169 int code_linear_p;
171 /* If the charset is treated as 94-chars in ISO-2022, the value is 0.
172 If the charset is treated as 96-chars in ISO-2022, the value is 1. */
173 int iso_chars_96;
175 /* ISO final byte of the charset: 48..127. It may be -1 if the
176 charset doesn't conform to ISO-2022. */
177 int iso_final;
179 /* ISO revision number of the charset. */
180 int iso_revision;
182 /* If the charset is identical to what supported by Emacs 21 and the
183 priors, the identification number of the charset used in those
184 version. Otherwise, -1. */
185 int emacs_mule_id;
187 /* Nonzero if the charset is compatible with ASCII. */
188 int ascii_compatible_p;
190 /* Nonzero if the charset is supplementary. */
191 int supplementary_p;
193 /* Nonzero if all the code points are representable by Lisp_Int. */
194 int compact_codes_p;
196 /* The method for encoding/decoding characters of the charset. */
197 enum charset_method method;
199 /* Mininum and Maximum code points of the charset. */
200 unsigned min_code, max_code;
202 /* Offset value used by macros CODE_POINT_TO_INDEX and
203 INDEX_TO_CODE_POINT. . */
204 unsigned char_index_offset;
206 /* Mininum and Maximum character codes of the charset. If the
207 charset is compatible with ASCII, min_char is a minimum non-ASCII
208 character of the charset. If the method of charset is
209 CHARSET_METHOD_OFFSET, even if the charset is unified, min_char
210 and max_char doesn't change. */
211 int min_char, max_char;
213 /* The code returned by ENCODE_CHAR if a character is not encodable
214 by the charset. */
215 unsigned invalid_code;
217 /* If the method of the charset is CHARSET_METHOD_MAP, this is a
218 table of bits used to quickly and roughly guess if a character
219 belongs to the charset.
221 The first 64 elements are 512 bits for characters less than
222 0x10000. Each bit corresponds to 128-character block. The last
223 126 elements are 1008 bits for the greater characters
224 (0x10000..0x3FFFFF). Each bit corresponds to 4096-character
225 block.
227 If a bit is 1, at least one character in the corresponding block is
228 in this charset. */
229 unsigned char fast_map[190];
231 /* Offset value to calculate a character code from code-point, and
232 visa versa. */
233 int code_offset;
235 int unified_p;
238 /* Hash table of charset symbols vs. the correponding attribute
239 vectors. */
240 extern Lisp_Object Vcharset_hash_table;
242 /* Table of struct charset. */
243 extern struct charset *charset_table;
245 #define CHARSET_FROM_ID(id) (charset_table + (id))
247 extern Lisp_Object Vcharset_ordered_list;
248 extern Lisp_Object Vcharset_non_preferred_head;
250 /* Incremented everytime we change the priority of charsets. */
251 extern unsigned short charset_ordered_list_tick;
253 extern Lisp_Object Viso_2022_charset_list;
254 extern Lisp_Object Vemacs_mule_charset_list;
256 extern int emacs_mule_charset[256];
258 /* Macros to access information about charset. */
260 /* Return the attribute vector of charset whose symbol is SYMBOL. */
261 #define CHARSET_SYMBOL_ATTRIBUTES(symbol) \
262 Fgethash ((symbol), Vcharset_hash_table, Qnil)
264 #define CHARSET_ATTR_ID(attrs) AREF ((attrs), charset_id)
265 #define CHARSET_ATTR_NAME(attrs) AREF ((attrs), charset_name)
266 #define CHARSET_ATTR_PLIST(attrs) AREF ((attrs), charset_plist)
267 #define CHARSET_ATTR_MAP(attrs) AREF ((attrs), charset_map)
268 #define CHARSET_ATTR_DECODER(attrs) AREF ((attrs), charset_decoder)
269 #define CHARSET_ATTR_ENCODER(attrs) AREF ((attrs), charset_encoder)
270 #define CHARSET_ATTR_SUBSET(attrs) AREF ((attrs), charset_subset)
271 #define CHARSET_ATTR_SUPERSET(attrs) AREF ((attrs), charset_superset)
272 #define CHARSET_ATTR_UNIFY_MAP(attrs) AREF ((attrs), charset_unify_map)
273 #define CHARSET_ATTR_DEUNIFIER(attrs) AREF ((attrs), charset_deunifier)
275 #define CHARSET_SYMBOL_ID(symbol) \
276 CHARSET_ATTR_ID (CHARSET_SYMBOL_ATTRIBUTES (symbol))
278 /* Return an index to Vcharset_hash_table of the charset whose symbol
279 is SYMBOL. */
280 #define CHARSET_SYMBOL_HASH_INDEX(symbol) \
281 hash_lookup (XHASH_TABLE (Vcharset_hash_table), symbol, NULL)
283 /* Return the attribute vector of CHARSET. */
284 #define CHARSET_ATTRIBUTES(charset) \
285 (HASH_VALUE (XHASH_TABLE (Vcharset_hash_table), (charset)->hash_index))
287 #define CHARSET_ID(charset) ((charset)->id)
288 #define CHARSET_HASH_INDEX(charset) ((charset)->hash_index)
289 #define CHARSET_DIMENSION(charset) ((charset)->dimension)
290 #define CHARSET_CODE_SPACE(charset) ((charset)->code_space)
291 #define CHARSET_CODE_LINEAR_P(charset) ((charset)->code_linear_p)
292 #define CHARSET_ISO_CHARS_96(charset) ((charset)->iso_chars_96)
293 #define CHARSET_ISO_FINAL(charset) ((charset)->iso_final)
294 #define CHARSET_ISO_PLANE(charset) ((charset)->iso_plane)
295 #define CHARSET_ISO_REVISION(charset) ((charset)->iso_revision)
296 #define CHARSET_EMACS_MULE_ID(charset) ((charset)->emacs_mule_id)
297 #define CHARSET_ASCII_COMPATIBLE_P(charset) ((charset)->ascii_compatible_p)
298 #define CHARSET_COMPACT_CODES_P(charset) ((charset)->compact_codes_p)
299 #define CHARSET_METHOD(charset) ((charset)->method)
300 #define CHARSET_MIN_CODE(charset) ((charset)->min_code)
301 #define CHARSET_MAX_CODE(charset) ((charset)->max_code)
302 #define CHARSET_INVALID_CODE(charset) ((charset)->invalid_code)
303 #define CHARSET_MIN_CHAR(charset) ((charset)->min_char)
304 #define CHARSET_MAX_CHAR(charset) ((charset)->max_char)
305 #define CHARSET_CODE_OFFSET(charset) ((charset)->code_offset)
306 #define CHARSET_UNIFIED_P(charset) ((charset)->unified_p)
308 #define CHARSET_NAME(charset) \
309 (CHARSET_ATTR_NAME (CHARSET_ATTRIBUTES (charset)))
310 #define CHARSET_MAP(charset) \
311 (CHARSET_ATTR_MAP (CHARSET_ATTRIBUTES (charset)))
312 #define CHARSET_DECODER(charset) \
313 (CHARSET_ATTR_DECODER (CHARSET_ATTRIBUTES (charset)))
314 #define CHARSET_ENCODER(charset) \
315 (CHARSET_ATTR_ENCODER (CHARSET_ATTRIBUTES (charset)))
316 #define CHARSET_SUBSET(charset) \
317 (CHARSET_ATTR_SUBSET (CHARSET_ATTRIBUTES (charset)))
318 #define CHARSET_SUPERSET(charset) \
319 (CHARSET_ATTR_SUPERSET (CHARSET_ATTRIBUTES (charset)))
320 #define CHARSET_UNIFY_MAP(charset) \
321 (CHARSET_ATTR_UNIFY_MAP (CHARSET_ATTRIBUTES (charset)))
322 #define CHARSET_DEUNIFIER(charset) \
323 (CHARSET_ATTR_DEUNIFIER (CHARSET_ATTRIBUTES (charset)))
326 /* Nonzero if OBJ is a valid charset symbol. */
327 #define CHARSETP(obj) (CHARSET_SYMBOL_HASH_INDEX (obj) >= 0)
329 /* Check if X is a valid charset symbol. If not, signal an error. */
330 #define CHECK_CHARSET(x) \
331 do { \
332 if (! SYMBOLP (x) || CHARSET_SYMBOL_HASH_INDEX (x) < 0) \
333 wrong_type_argument (Qcharsetp, (x)); \
334 } while (0)
337 /* Check if X is a valid charset symbol. If valid, set ID to the id
338 number of the charset. Otherwise, signal an error. */
339 #define CHECK_CHARSET_GET_ID(x, id) \
340 do { \
341 int idx; \
343 if (! SYMBOLP (x) || (idx = CHARSET_SYMBOL_HASH_INDEX (x)) < 0) \
344 wrong_type_argument (Qcharsetp, (x)); \
345 id = XINT (AREF (HASH_VALUE (XHASH_TABLE (Vcharset_hash_table), idx), \
346 charset_id)); \
347 } while (0)
350 /* Check if X is a valid charset symbol. If valid, set ATTR to the
351 attr vector of the charset. Otherwise, signal an error. */
352 #define CHECK_CHARSET_GET_ATTR(x, attr) \
353 do { \
354 if (!SYMBOLP (x) || NILP (attr = CHARSET_SYMBOL_ATTRIBUTES (x))) \
355 wrong_type_argument (Qcharsetp, (x)); \
356 } while (0)
359 #define CHECK_CHARSET_GET_CHARSET(x, charset) \
360 do { \
361 int id; \
362 CHECK_CHARSET_GET_ID (x, id); \
363 charset = CHARSET_FROM_ID (id); \
364 } while (0)
367 /* Lookup Vcharset_ordered_list and return the first charset that
368 contains the character C. */
369 #define CHAR_CHARSET(c) \
370 ((c) < 0x80 ? CHARSET_FROM_ID (charset_ascii) \
371 : char_charset ((c), Qnil, NULL))
373 #if 0
374 /* Char-table of charset-sets. Each element is a bool vector indexed
375 by a charset ID. */
376 extern Lisp_Object Vchar_charset_set;
378 /* Charset-bag of character C. */
379 #define CHAR_CHARSET_SET(c) \
380 CHAR_TABLE_REF (Vchar_charset_set, c)
382 /* Check if two characters C1 and C2 belong to the same charset. */
383 #define SAME_CHARSET_P(c1, c2) \
384 intersection_p (CHAR_CHARSET_SET (c1), CHAR_CHARSET_SET (c2))
386 #endif
389 /* Return a character correponding to the code-point CODE of CHARSET.
390 Try some optimization before calling decode_char. */
392 #define DECODE_CHAR(charset, code) \
393 ((ASCII_BYTE_P (code) && (charset)->ascii_compatible_p) \
394 ? (code) \
395 : ((code) < (charset)->min_code || (code) > (charset)->max_code) \
396 ? -1 \
397 : (charset)->unified_p \
398 ? decode_char ((charset), (code)) \
399 : (charset)->method == CHARSET_METHOD_OFFSET \
400 ? ((charset)->code_linear_p \
401 ? (code) - (charset)->min_code + (charset)->code_offset \
402 : decode_char ((charset), (code))) \
403 : (charset)->method == CHARSET_METHOD_MAP \
404 ? (((charset)->code_linear_p \
405 && VECTORP (CHARSET_DECODER (charset))) \
406 ? XINT (AREF (CHARSET_DECODER (charset), \
407 (code) - (charset)->min_code)) \
408 : decode_char ((charset), (code))) \
409 : decode_char ((charset), (code)))
412 /* If CHARSET is a simple offset base charset, return it's offset,
413 otherwise return -1. */
414 #define CHARSET_OFFSET(charset) \
415 (((charset)->method == CHARSET_METHOD_OFFSET \
416 && (charset)->code_linear_p \
417 && ! (charset)->unified_p) \
418 ? (charset)->code_offset - (charset)->min_code \
419 : -1)
421 extern Lisp_Object charset_work;
423 /* Return a code point of CHAR in CHARSET.
424 Try some optimization before calling encode_char. */
426 #define ENCODE_CHAR(charset, c) \
427 ((ASCII_CHAR_P (c) && (charset)->ascii_compatible_p) \
428 ? (c) \
429 : ((charset)->unified_p \
430 || (charset)->method == CHARSET_METHOD_SUBSET \
431 || (charset)->method == CHARSET_METHOD_SUPERSET) \
432 ? encode_char ((charset), (c)) \
433 : ((c) < (charset)->min_char || (c) > (charset)->max_char) \
434 ? (charset)->invalid_code \
435 : (charset)->method == CHARSET_METHOD_OFFSET \
436 ? ((charset)->code_linear_p \
437 ? (c) - (charset)->code_offset + (charset)->min_code \
438 : encode_char ((charset), (c))) \
439 : (charset)->method == CHARSET_METHOD_MAP \
440 ? (((charset)->compact_codes_p \
441 && CHAR_TABLE_P (CHARSET_ENCODER (charset))) \
442 ? (charset_work = CHAR_TABLE_REF (CHARSET_ENCODER (charset), (c)), \
443 (NILP (charset_work) \
444 ? (charset)->invalid_code \
445 : XFASTINT (charset_work))) \
446 : encode_char ((charset), (c))) \
447 : encode_char ((charset), (c)))
450 /* Set to 1 when a charset map is loaded to warn that a buffer text
451 and a string data may be relocated. */
452 extern int charset_map_loaded;
455 /* Set CHARSET to the charset highest priority of C, CODE to the
456 code-point of C in CHARSET. */
457 #define SPLIT_CHAR(c, charset, code) \
458 ((charset) = char_charset ((c), Qnil, &(code)))
461 #define ISO_MAX_DIMENSION 3
462 #define ISO_MAX_CHARS 2
463 #define ISO_MAX_FINAL 0x80 /* only 0x30..0xFF are used */
465 /* Mapping table from ISO2022's charset (specified by DIMENSION,
466 CHARS, and FINAL_CHAR) to Emacs' charset ID. Should be accessed by
467 macro ISO_CHARSET_TABLE (DIMENSION, CHARS, FINAL_CHAR). */
468 extern int iso_charset_table[ISO_MAX_DIMENSION][ISO_MAX_CHARS][ISO_MAX_FINAL];
470 /* A charset of type iso2022 who has DIMENSION, CHARS, and FINAL
471 (final character). */
472 #define ISO_CHARSET_TABLE(dimension, chars_96, final) \
473 iso_charset_table[(dimension) - 1][(chars_96)][(final)]
475 /* Nonzero if the charset who has FAST_MAP may contain C. */
476 #define CHARSET_FAST_MAP_REF(c, fast_map) \
477 ((c) < 0x10000 \
478 ? fast_map[(c) >> 10] & (1 << (((c) >> 7) & 7)) \
479 : fast_map[((c) >> 15) + 62] & (1 << (((c) >> 12) & 7)))
481 #define CHARSET_FAST_MAP_SET(c, fast_map) \
482 do { \
483 if ((c) < 0x10000) \
484 (fast_map)[(c) >> 10] |= 1 << (((c) >> 7) & 7); \
485 else \
486 (fast_map)[((c) >> 15) + 62] |= 1 << (((c) >> 12) & 7); \
487 } while (0)
491 /* 1 if CHARSET may contain the character C. */
492 #define CHAR_CHARSET_P(c, charset) \
493 ((ASCII_CHAR_P (c) && (charset)->ascii_compatible_p) \
494 || ((CHARSET_UNIFIED_P (charset) \
495 || (charset)->method == CHARSET_METHOD_SUBSET \
496 || (charset)->method == CHARSET_METHOD_SUPERSET) \
497 ? encode_char ((charset), (c)) != (charset)->invalid_code \
498 : (CHARSET_FAST_MAP_REF ((c), (charset)->fast_map) \
499 && ((charset)->method == CHARSET_METHOD_OFFSET \
500 ? (c) >= (charset)->min_char && (c) <= (charset)->max_char \
501 : ((charset)->method == CHARSET_METHOD_MAP \
502 && (charset)->compact_codes_p \
503 && CHAR_TABLE_P (CHARSET_ENCODER (charset))) \
504 ? ! NILP (CHAR_TABLE_REF (CHARSET_ENCODER (charset), (c))) \
505 : encode_char ((charset), (c)) != (charset)->invalid_code))))
508 /* Special macros for emacs-mule encoding. */
510 /* Leading-code followed by extended leading-code. DIMENSION/COLUMN */
511 #define EMACS_MULE_LEADING_CODE_PRIVATE_11 0x9A /* 1/1 */
512 #define EMACS_MULE_LEADING_CODE_PRIVATE_12 0x9B /* 1/2 */
513 #define EMACS_MULE_LEADING_CODE_PRIVATE_21 0x9C /* 2/2 */
514 #define EMACS_MULE_LEADING_CODE_PRIVATE_22 0x9D /* 2/2 */
518 extern Lisp_Object Qcharsetp;
520 extern Lisp_Object Qascii, Qunicode;
521 extern int charset_ascii, charset_eight_bit;
522 extern int charset_iso_8859_1;
523 extern int charset_unicode;
524 extern int charset_jisx0201_roman;
525 extern int charset_jisx0208_1978;
526 extern int charset_jisx0208;
527 extern int charset_ksc5601;
529 extern int charset_unibyte;
531 extern struct charset *char_charset (int, Lisp_Object, unsigned *);
532 extern Lisp_Object charset_attributes (int);
534 extern int maybe_unify_char (int, Lisp_Object);
535 extern int decode_char (struct charset *, unsigned);
536 extern unsigned encode_char (struct charset *, int);
537 extern int string_xstring_p (Lisp_Object);
539 extern void map_charset_chars (void (*) (Lisp_Object, Lisp_Object),
540 Lisp_Object, Lisp_Object,
541 struct charset *, unsigned, unsigned);
543 #endif /* EMACS_CHARSET_H */