Update Files chapter of Emacs manual.
[emacs.git] / src / charset.h
blobbe02bc0feae43433e32be9d8efc79d4193e9529e
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 #include <verify.h>
32 /* Index to arguments of Fdefine_charset_internal. */
34 enum define_charset_arg_index
36 charset_arg_name,
37 charset_arg_dimension,
38 charset_arg_code_space,
39 charset_arg_min_code,
40 charset_arg_max_code,
41 charset_arg_iso_final,
42 charset_arg_iso_revision,
43 charset_arg_emacs_mule_id,
44 charset_arg_ascii_compatible_p,
45 charset_arg_supplementary_p,
46 charset_arg_invalid_code,
47 charset_arg_code_offset,
48 charset_arg_map,
49 charset_arg_subset,
50 charset_arg_superset,
51 charset_arg_unify_map,
52 charset_arg_plist,
53 charset_arg_max
57 /* Indices to charset attributes vector. */
59 enum charset_attr_index
61 /* ID number of the charset. */
62 charset_id,
64 /* Name of the charset (symbol). */
65 charset_name,
67 /* Property list of the charset. */
68 charset_plist,
70 /* If the method of the charset is `MAP', the value is a mapping
71 vector or a file name that contains mapping vector. Otherwise,
72 nil. */
73 charset_map,
75 /* If the method of the charset is `MAP', the value is a vector
76 that maps code points of the charset to characters. The vector
77 is indexed by a character index. A character index is
78 calculated from a code point and the code-space table of the
79 charset. */
80 charset_decoder,
82 /* If the method of the charset is `MAP', the value is a
83 char-table that maps characters of the charset to code
84 points. */
85 charset_encoder,
87 /* If the method of the charset is `SUBSET', the value is a vector
88 that has this form:
90 [ CHARSET-ID MIN-CODE MAX-CODE OFFSET ]
92 CHARSET-ID is an ID number of a parent charset. MIN-CODE and
93 MAX-CODE specify the range of characters inherited from the
94 parent. OFFSET is an integer value to add to a code point of
95 the parent charset to get the corresponding code point of this
96 charset. */
97 charset_subset,
99 /* If the method of the charset is `SUPERSET', the value is a list
100 whose elements have this form:
102 (CHARSET-ID . OFFSET)
104 CHARSET-IDs are ID numbers of parent charsets. OFFSET is an
105 integer value to add to a code point of the parent charset to
106 get the corresponding code point of this charset. */
107 charset_superset,
109 /* The value is a mapping vector or a file name that contains the
110 mapping. This defines how characters in the charset should be
111 unified with Unicode. The value of the member
112 `charset_deunifier' is created from this information. */
113 charset_unify_map,
115 /* If characters in the charset must be unified Unicode, the value
116 is a char table that maps a unified Unicode character code to
117 the non-unified character code in the charset. */
118 charset_deunifier,
120 /* The length of the charset attribute vector. */
121 charset_attr_max
124 /* Methods for converting code points and characters of charsets. */
126 enum charset_method
128 /* For a charset of this method, a character code is calculated
129 from a character index (which is calculated from a code point)
130 simply by adding an offset value. */
131 CHARSET_METHOD_OFFSET,
133 /* For a charset of this method, a decoder vector and an encoder
134 char-table is used for code point <-> character code
135 conversion. */
136 CHARSET_METHOD_MAP,
138 /* A charset of this method is a subset of another charset. */
139 CHARSET_METHOD_SUBSET,
141 /* A charset of this method is a superset of other charsets. */
142 CHARSET_METHOD_SUPERSET
145 struct charset
147 /* Index to charset_table. */
148 int id;
150 /* Index to Vcharset_hash_table. */
151 ptrdiff_t hash_index;
153 /* Dimension of the charset: 1, 2, 3, or 4. */
154 int dimension;
156 /* Byte code range of each dimension. <code_space>[4N] is a mininum
157 byte code of the (N+1)th dimension, <code_space>[4N+1] is a
158 maximum byte code of the (N+1)th dimension, <code_space>[4N+2] is
159 (<code_space>[4N+1] - <code_space>[4N] + 1), <code_space>[4N+3]
160 is the number of characters contained in the first through (N+1)th
161 dimensions, except that there is no <code_space>[15].
162 We get `char-index' of a `code-point' from this
163 information. */
164 int code_space[15];
166 /* If B is a byte of Nth dimension of a code-point, the (N-1)th bit
167 of code_space_mask[B] is set. This array is used to quickly
168 check if a code-point is in a valid range. */
169 unsigned char *code_space_mask;
171 /* 1 if there's no gap in code-points. */
172 int code_linear_p;
174 /* If the charset is treated as 94-chars in ISO-2022, the value is 0.
175 If the charset is treated as 96-chars in ISO-2022, the value is 1. */
176 int iso_chars_96;
178 /* ISO final byte of the charset: 48..127. It may be -1 if the
179 charset doesn't conform to ISO-2022. */
180 int iso_final;
182 /* ISO revision number of the charset. */
183 int iso_revision;
185 /* If the charset is identical to what supported by Emacs 21 and the
186 priors, the identification number of the charset used in those
187 version. Otherwise, -1. */
188 int emacs_mule_id;
190 /* Nonzero if the charset is compatible with ASCII. */
191 int ascii_compatible_p;
193 /* Nonzero if the charset is supplementary. */
194 int supplementary_p;
196 /* Nonzero if all the code points are representable by Lisp_Int. */
197 int compact_codes_p;
199 /* The method for encoding/decoding characters of the charset. */
200 enum charset_method method;
202 /* Mininum and Maximum code points of the charset. */
203 unsigned min_code, max_code;
205 /* Offset value used by macros CODE_POINT_TO_INDEX and
206 INDEX_TO_CODE_POINT. . */
207 unsigned char_index_offset;
209 /* Mininum and Maximum character codes of the charset. If the
210 charset is compatible with ASCII, min_char is a minimum non-ASCII
211 character of the charset. If the method of charset is
212 CHARSET_METHOD_OFFSET, even if the charset is unified, min_char
213 and max_char doesn't change. */
214 int min_char, max_char;
216 /* The code returned by ENCODE_CHAR if a character is not encodable
217 by the charset. */
218 unsigned invalid_code;
220 /* If the method of the charset is CHARSET_METHOD_MAP, this is a
221 table of bits used to quickly and roughly guess if a character
222 belongs to the charset.
224 The first 64 elements are 512 bits for characters less than
225 0x10000. Each bit corresponds to 128-character block. The last
226 126 elements are 1008 bits for the greater characters
227 (0x10000..0x3FFFFF). Each bit corresponds to 4096-character
228 block.
230 If a bit is 1, at least one character in the corresponding block is
231 in this charset. */
232 unsigned char fast_map[190];
234 /* Offset value to calculate a character code from code-point, and
235 visa versa. */
236 int code_offset;
238 int unified_p;
241 /* Hash table of charset symbols vs. the correponding attribute
242 vectors. */
243 extern Lisp_Object Vcharset_hash_table;
245 /* Table of struct charset. */
246 extern struct charset *charset_table;
248 #define CHARSET_FROM_ID(id) (charset_table + (id))
250 extern Lisp_Object Vcharset_ordered_list;
251 extern Lisp_Object Vcharset_non_preferred_head;
253 /* Incremented everytime we change the priority of charsets. */
254 extern unsigned short charset_ordered_list_tick;
256 extern Lisp_Object Viso_2022_charset_list;
257 extern Lisp_Object Vemacs_mule_charset_list;
259 extern int emacs_mule_charset[256];
261 /* Macros to access information about charset. */
263 /* Return the attribute vector of charset whose symbol is SYMBOL. */
264 #define CHARSET_SYMBOL_ATTRIBUTES(symbol) \
265 Fgethash ((symbol), Vcharset_hash_table, Qnil)
267 #define CHARSET_ATTR_ID(attrs) AREF ((attrs), charset_id)
268 #define CHARSET_ATTR_NAME(attrs) AREF ((attrs), charset_name)
269 #define CHARSET_ATTR_PLIST(attrs) AREF ((attrs), charset_plist)
270 #define CHARSET_ATTR_MAP(attrs) AREF ((attrs), charset_map)
271 #define CHARSET_ATTR_DECODER(attrs) AREF ((attrs), charset_decoder)
272 #define CHARSET_ATTR_ENCODER(attrs) AREF ((attrs), charset_encoder)
273 #define CHARSET_ATTR_SUBSET(attrs) AREF ((attrs), charset_subset)
274 #define CHARSET_ATTR_SUPERSET(attrs) AREF ((attrs), charset_superset)
275 #define CHARSET_ATTR_UNIFY_MAP(attrs) AREF ((attrs), charset_unify_map)
276 #define CHARSET_ATTR_DEUNIFIER(attrs) AREF ((attrs), charset_deunifier)
278 #define CHARSET_SYMBOL_ID(symbol) \
279 CHARSET_ATTR_ID (CHARSET_SYMBOL_ATTRIBUTES (symbol))
281 /* Return an index to Vcharset_hash_table of the charset whose symbol
282 is SYMBOL. */
283 #define CHARSET_SYMBOL_HASH_INDEX(symbol) \
284 hash_lookup (XHASH_TABLE (Vcharset_hash_table), symbol, NULL)
286 /* Return the attribute vector of CHARSET. */
287 #define CHARSET_ATTRIBUTES(charset) \
288 (HASH_VALUE (XHASH_TABLE (Vcharset_hash_table), (charset)->hash_index))
290 #define CHARSET_ID(charset) ((charset)->id)
291 #define CHARSET_HASH_INDEX(charset) ((charset)->hash_index)
292 #define CHARSET_DIMENSION(charset) ((charset)->dimension)
293 #define CHARSET_CODE_SPACE(charset) ((charset)->code_space)
294 #define CHARSET_CODE_LINEAR_P(charset) ((charset)->code_linear_p)
295 #define CHARSET_ISO_CHARS_96(charset) ((charset)->iso_chars_96)
296 #define CHARSET_ISO_FINAL(charset) ((charset)->iso_final)
297 #define CHARSET_ISO_PLANE(charset) ((charset)->iso_plane)
298 #define CHARSET_ISO_REVISION(charset) ((charset)->iso_revision)
299 #define CHARSET_EMACS_MULE_ID(charset) ((charset)->emacs_mule_id)
300 #define CHARSET_ASCII_COMPATIBLE_P(charset) ((charset)->ascii_compatible_p)
301 #define CHARSET_COMPACT_CODES_P(charset) ((charset)->compact_codes_p)
302 #define CHARSET_METHOD(charset) ((charset)->method)
303 #define CHARSET_MIN_CODE(charset) ((charset)->min_code)
304 #define CHARSET_MAX_CODE(charset) ((charset)->max_code)
305 #define CHARSET_INVALID_CODE(charset) ((charset)->invalid_code)
306 #define CHARSET_MIN_CHAR(charset) ((charset)->min_char)
307 #define CHARSET_MAX_CHAR(charset) ((charset)->max_char)
308 #define CHARSET_CODE_OFFSET(charset) ((charset)->code_offset)
309 #define CHARSET_UNIFIED_P(charset) ((charset)->unified_p)
311 #define CHARSET_NAME(charset) \
312 (CHARSET_ATTR_NAME (CHARSET_ATTRIBUTES (charset)))
313 #define CHARSET_MAP(charset) \
314 (CHARSET_ATTR_MAP (CHARSET_ATTRIBUTES (charset)))
315 #define CHARSET_DECODER(charset) \
316 (CHARSET_ATTR_DECODER (CHARSET_ATTRIBUTES (charset)))
317 #define CHARSET_ENCODER(charset) \
318 (CHARSET_ATTR_ENCODER (CHARSET_ATTRIBUTES (charset)))
319 #define CHARSET_SUBSET(charset) \
320 (CHARSET_ATTR_SUBSET (CHARSET_ATTRIBUTES (charset)))
321 #define CHARSET_SUPERSET(charset) \
322 (CHARSET_ATTR_SUPERSET (CHARSET_ATTRIBUTES (charset)))
323 #define CHARSET_UNIFY_MAP(charset) \
324 (CHARSET_ATTR_UNIFY_MAP (CHARSET_ATTRIBUTES (charset)))
325 #define CHARSET_DEUNIFIER(charset) \
326 (CHARSET_ATTR_DEUNIFIER (CHARSET_ATTRIBUTES (charset)))
329 /* Nonzero if OBJ is a valid charset symbol. */
330 #define CHARSETP(obj) (CHARSET_SYMBOL_HASH_INDEX (obj) >= 0)
332 /* Check if X is a valid charset symbol. If not, signal an error. */
333 #define CHECK_CHARSET(x) \
334 do { \
335 if (! SYMBOLP (x) || CHARSET_SYMBOL_HASH_INDEX (x) < 0) \
336 wrong_type_argument (Qcharsetp, (x)); \
337 } while (0)
340 /* Check if X is a valid charset symbol. If valid, set ID to the id
341 number of the charset. Otherwise, signal an error. */
342 #define CHECK_CHARSET_GET_ID(x, id) \
343 do { \
344 ptrdiff_t idx; \
346 if (! SYMBOLP (x) || (idx = CHARSET_SYMBOL_HASH_INDEX (x)) < 0) \
347 wrong_type_argument (Qcharsetp, (x)); \
348 id = XINT (AREF (HASH_VALUE (XHASH_TABLE (Vcharset_hash_table), idx), \
349 charset_id)); \
350 } while (0)
353 /* Check if X is a valid charset symbol. If valid, set ATTR to the
354 attr vector of the charset. Otherwise, signal an error. */
355 #define CHECK_CHARSET_GET_ATTR(x, attr) \
356 do { \
357 if (!SYMBOLP (x) || NILP (attr = CHARSET_SYMBOL_ATTRIBUTES (x))) \
358 wrong_type_argument (Qcharsetp, (x)); \
359 } while (0)
362 #define CHECK_CHARSET_GET_CHARSET(x, charset) \
363 do { \
364 int csid; \
365 CHECK_CHARSET_GET_ID (x, csid); \
366 charset = CHARSET_FROM_ID (csid); \
367 } while (0)
370 /* Lookup Vcharset_ordered_list and return the first charset that
371 contains the character C. */
372 #define CHAR_CHARSET(c) \
373 ((c) < 0x80 ? CHARSET_FROM_ID (charset_ascii) \
374 : char_charset ((c), Qnil, NULL))
376 #if 0
377 /* Char-table of charset-sets. Each element is a bool vector indexed
378 by a charset ID. */
379 extern Lisp_Object Vchar_charset_set;
381 /* Charset-bag of character C. */
382 #define CHAR_CHARSET_SET(c) \
383 CHAR_TABLE_REF (Vchar_charset_set, c)
385 /* Check if two characters C1 and C2 belong to the same charset. */
386 #define SAME_CHARSET_P(c1, c2) \
387 intersection_p (CHAR_CHARSET_SET (c1), CHAR_CHARSET_SET (c2))
389 #endif
392 /* Return a character correponding to the code-point CODE of CHARSET.
393 Try some optimization before calling decode_char. */
395 #define DECODE_CHAR(charset, code) \
396 ((ASCII_BYTE_P (code) && (charset)->ascii_compatible_p) \
397 ? (code) \
398 : ((code) < (charset)->min_code || (code) > (charset)->max_code) \
399 ? -1 \
400 : (charset)->unified_p \
401 ? decode_char ((charset), (code)) \
402 : (charset)->method == CHARSET_METHOD_OFFSET \
403 ? ((charset)->code_linear_p \
404 ? (code) - (charset)->min_code + (charset)->code_offset \
405 : decode_char ((charset), (code))) \
406 : (charset)->method == CHARSET_METHOD_MAP \
407 ? (((charset)->code_linear_p \
408 && VECTORP (CHARSET_DECODER (charset))) \
409 ? XINT (AREF (CHARSET_DECODER (charset), \
410 (code) - (charset)->min_code)) \
411 : decode_char ((charset), (code))) \
412 : decode_char ((charset), (code)))
415 /* If CHARSET is a simple offset base charset, return it's offset,
416 otherwise return -1. */
417 #define CHARSET_OFFSET(charset) \
418 (((charset)->method == CHARSET_METHOD_OFFSET \
419 && (charset)->code_linear_p \
420 && ! (charset)->unified_p) \
421 ? (charset)->code_offset - (charset)->min_code \
422 : -1)
424 extern Lisp_Object charset_work;
426 /* Return a code point of CHAR in CHARSET.
427 Try some optimization before calling encode_char. */
429 #define ENCODE_CHAR(charset, c) \
430 (verify_expr \
431 (sizeof (c) <= sizeof (int), \
432 (ASCII_CHAR_P (c) && (charset)->ascii_compatible_p \
433 ? (c) \
434 : ((charset)->unified_p \
435 || (charset)->method == CHARSET_METHOD_SUBSET \
436 || (charset)->method == CHARSET_METHOD_SUPERSET) \
437 ? encode_char (charset, c) \
438 : (c) < (charset)->min_char || (c) > (charset)->max_char \
439 ? (charset)->invalid_code \
440 : (charset)->method == CHARSET_METHOD_OFFSET \
441 ? ((charset)->code_linear_p \
442 ? (c) - (charset)->code_offset + (charset)->min_code \
443 : encode_char (charset, c)) \
444 : (charset)->method == CHARSET_METHOD_MAP \
445 ? (((charset)->compact_codes_p \
446 && CHAR_TABLE_P (CHARSET_ENCODER (charset))) \
447 ? (charset_work = CHAR_TABLE_REF (CHARSET_ENCODER (charset), c), \
448 (NILP (charset_work) \
449 ? (charset)->invalid_code \
450 : XFASTINT (charset_work))) \
451 : encode_char (charset, c)) \
452 : encode_char (charset, c))))
455 /* Set to 1 when a charset map is loaded to warn that a buffer text
456 and a string data may be relocated. */
457 extern int charset_map_loaded;
460 /* Set CHARSET to the charset highest priority of C, CODE to the
461 code-point of C in CHARSET. */
462 #define SPLIT_CHAR(c, charset, code) \
463 ((charset) = char_charset ((c), Qnil, &(code)))
466 #define ISO_MAX_DIMENSION 3
467 #define ISO_MAX_CHARS 2
468 #define ISO_MAX_FINAL 0x80 /* only 0x30..0xFF are used */
470 /* Mapping table from ISO2022's charset (specified by DIMENSION,
471 CHARS, and FINAL_CHAR) to Emacs' charset ID. Should be accessed by
472 macro ISO_CHARSET_TABLE (DIMENSION, CHARS, FINAL_CHAR). */
473 extern int iso_charset_table[ISO_MAX_DIMENSION][ISO_MAX_CHARS][ISO_MAX_FINAL];
475 /* A charset of type iso2022 who has DIMENSION, CHARS, and FINAL
476 (final character). */
477 #define ISO_CHARSET_TABLE(dimension, chars_96, final) \
478 iso_charset_table[(dimension) - 1][(chars_96)][(final)]
480 /* Nonzero if the charset who has FAST_MAP may contain C. */
481 #define CHARSET_FAST_MAP_REF(c, fast_map) \
482 ((c) < 0x10000 \
483 ? fast_map[(c) >> 10] & (1 << (((c) >> 7) & 7)) \
484 : fast_map[((c) >> 15) + 62] & (1 << (((c) >> 12) & 7)))
486 #define CHARSET_FAST_MAP_SET(c, fast_map) \
487 do { \
488 if ((c) < 0x10000) \
489 (fast_map)[(c) >> 10] |= 1 << (((c) >> 7) & 7); \
490 else \
491 (fast_map)[((c) >> 15) + 62] |= 1 << (((c) >> 12) & 7); \
492 } while (0)
496 /* 1 if CHARSET may contain the character C. */
497 #define CHAR_CHARSET_P(c, charset) \
498 ((ASCII_CHAR_P (c) && (charset)->ascii_compatible_p) \
499 || ((CHARSET_UNIFIED_P (charset) \
500 || (charset)->method == CHARSET_METHOD_SUBSET \
501 || (charset)->method == CHARSET_METHOD_SUPERSET) \
502 ? encode_char ((charset), (c)) != (charset)->invalid_code \
503 : (CHARSET_FAST_MAP_REF ((c), (charset)->fast_map) \
504 && ((charset)->method == CHARSET_METHOD_OFFSET \
505 ? (c) >= (charset)->min_char && (c) <= (charset)->max_char \
506 : ((charset)->method == CHARSET_METHOD_MAP \
507 && (charset)->compact_codes_p \
508 && CHAR_TABLE_P (CHARSET_ENCODER (charset))) \
509 ? ! NILP (CHAR_TABLE_REF (CHARSET_ENCODER (charset), (c))) \
510 : encode_char ((charset), (c)) != (charset)->invalid_code))))
513 /* Special macros for emacs-mule encoding. */
515 /* Leading-code followed by extended leading-code. DIMENSION/COLUMN */
516 #define EMACS_MULE_LEADING_CODE_PRIVATE_11 0x9A /* 1/1 */
517 #define EMACS_MULE_LEADING_CODE_PRIVATE_12 0x9B /* 1/2 */
518 #define EMACS_MULE_LEADING_CODE_PRIVATE_21 0x9C /* 2/2 */
519 #define EMACS_MULE_LEADING_CODE_PRIVATE_22 0x9D /* 2/2 */
523 extern Lisp_Object Qcharsetp;
525 extern Lisp_Object Qascii;
526 extern int charset_ascii, charset_eight_bit;
527 extern int charset_unicode;
528 extern int charset_jisx0201_roman;
529 extern int charset_jisx0208_1978;
530 extern int charset_jisx0208;
531 extern int charset_ksc5601;
533 extern int charset_unibyte;
535 extern struct charset *char_charset (int, Lisp_Object, unsigned *);
536 extern Lisp_Object charset_attributes (int);
538 extern int maybe_unify_char (int, Lisp_Object);
539 extern int decode_char (struct charset *, unsigned);
540 extern unsigned encode_char (struct charset *, int);
541 extern int string_xstring_p (Lisp_Object);
543 extern void map_charset_chars (void (*) (Lisp_Object, Lisp_Object),
544 Lisp_Object, Lisp_Object,
545 struct charset *, unsigned, unsigned);
547 #endif /* EMACS_CHARSET_H */