1 /* Header for multibyte character handler.
2 Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN.
3 Licensed to the Free Software Foundation.
4 Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008
5 National Institute of Advanced Industrial Science and Technology (AIST)
6 Registration Number H13PRO009
8 This file is part of GNU Emacs.
10 GNU Emacs is free software; you can redistribute it and/or modify
11 it under the terms of the GNU General Public License as published by
12 the Free Software Foundation; either version 3, or (at your option)
15 GNU Emacs is distributed in the hope that it will be useful,
16 but WITHOUT ANY WARRANTY; without even the implied warranty of
17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 GNU General Public License for more details.
20 You should have received a copy of the GNU General Public License
21 along with GNU Emacs; see the file COPYING. If not, write to the
22 Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
23 Boston, MA 02110-1301, USA. */
25 #ifndef EMACS_CHARACTER_H
26 #define EMACS_CHARACTER_H
28 /* character code 1st byte byte sequence
29 -------------- -------- -------------
31 80-7FF C2..DF 110xxxxx 10xxxxxx
32 800-FFFF E0..EF 1110xxxx 10xxxxxx 10xxxxxx
33 10000-1FFFFF F0..F7 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
34 200000-3FFF7F F8 11111000 1000xxxx 10xxxxxx 10xxxxxx 10xxxxxx
35 3FFF80-3FFFFF C0..C1 1100000x 10xxxxxx (for eight-bit-char)
38 invalid 1st byte 80..BF 10xxxxxx
39 F9..FF 11111xxx (xxx != 000)
42 /* Maximum character code ((1 << CHARACTERBITS) - 1). */
43 #define MAX_CHAR 0x3FFFFF
45 /* Maximum Unicode character code. */
46 #define MAX_UNICODE_CHAR 0x10FFFF
48 /* Maximum N-byte character codes. */
49 #define MAX_1_BYTE_CHAR 0x7F
50 #define MAX_2_BYTE_CHAR 0x7FF
51 #define MAX_3_BYTE_CHAR 0xFFFF
52 #define MAX_4_BYTE_CHAR 0x1FFFFF
53 #define MAX_5_BYTE_CHAR 0x3FFF7F
55 /* Minimum leading code of multibyte characters. */
56 #define MIN_MULTIBYTE_LEADING_CODE 0xC0
57 /* Maximum leading code of multibyte characters. */
58 #define MAX_MULTIBYTE_LEADING_CODE 0xF8
60 /* Nonzero iff C is a character that corresponds to a raw 8-bit
62 #define CHAR_BYTE8_P(c) ((c) > MAX_5_BYTE_CHAR)
64 /* Return the character code for raw 8-bit byte BYTE. */
65 #define BYTE8_TO_CHAR(byte) ((byte) + 0x3FFF00)
67 /* Return the raw 8-bit byte for character C. */
68 #define CHAR_TO_BYTE8(c) \
71 : multibyte_char_to_unibyte (c, Qnil))
73 /* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
74 that corresponds to a raw 8-bit byte. */
75 #define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
77 /* Mapping table from unibyte chars to multibyte chars. */
78 extern int unibyte_to_multibyte_table
[256];
80 /* Convert the unibyte character C to the corresponding multibyte
81 character. If C can't be converted, return C. */
82 #define unibyte_char_to_multibyte(c) \
83 ((c) < 256 ? unibyte_to_multibyte_table[(c)] : (c))
85 /* Nth element is 1 iff unibyte char N can be mapped to a multibyte
87 extern char unibyte_has_multibyte_table
[256];
89 #define UNIBYTE_CHAR_HAS_MULTIBYTE_P(c) (unibyte_has_multibyte_table[(c)])
91 /* If C is not ASCII, make it unibyte. */
92 #define MAKE_CHAR_UNIBYTE(c) \
94 if (! ASCII_CHAR_P (c)) \
95 c = CHAR_TO_BYTE8 (c); \
99 /* If C is not ASCII, make it multibyte. It assumes C < 256. */
100 #define MAKE_CHAR_MULTIBYTE(c) ((c) = unibyte_to_multibyte_table[(c)])
102 /* This is the maximum byte length of multibyte form. */
103 #define MAX_MULTIBYTE_LENGTH 5
105 /* Return a Lisp character whose character code is C. It assumes C is
106 a valid character code. */
107 #define make_char(c) make_number (c)
109 /* Nonzero iff C is an ASCII byte. */
110 #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
112 /* Nonzero iff X is a character. */
113 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
115 /* Nonzero iff C is valid as a character code. GENERICP is not used
117 #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
119 /* Check if Lisp object X is a character or not. */
120 #define CHECK_CHARACTER(x) \
121 CHECK_TYPE (CHARACTERP (x), Qcharacterp, x)
123 #define CHECK_CHARACTER_CAR(x) \
125 Lisp_Object tmp = XCAR (x); \
126 CHECK_CHARACTER (tmp); \
127 XSETCAR ((x), tmp); \
130 #define CHECK_CHARACTER_CDR(x) \
132 Lisp_Object tmp = XCDR (x); \
133 CHECK_CHARACTER (tmp); \
134 XSETCDR ((x), tmp); \
137 /* Nonzero iff C is an ASCII character. */
138 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
140 /* Nonzero iff C is a character of code less than 0x100. */
141 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
143 /* Nonzero if character C has a printable glyph. */
144 #define CHAR_PRINTABLE_P(c) \
145 (((c) >= 32 && ((c) < 127) \
146 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
148 /* Return byte length of multibyte form for character C. */
149 #define CHAR_BYTES(c) \
150 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
151 : (c) <= MAX_2_BYTE_CHAR ? 2 \
152 : (c) <= MAX_3_BYTE_CHAR ? 3 \
153 : (c) <= MAX_4_BYTE_CHAR ? 4 \
154 : (c) <= MAX_5_BYTE_CHAR ? 5 \
158 /* Return the leading code of multibyte form of C. */
159 #define CHAR_LEADING_CODE(c) \
160 ((c) <= MAX_1_BYTE_CHAR ? c \
161 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
162 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
163 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
164 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
165 : (0xC0 | (((c) >> 6) & 0x01)))
168 /* Store multibyte form of the character C in P. The caller should
169 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
170 Returns the length of the multibyte form. */
172 #define CHAR_STRING(c, p) \
173 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
176 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
177 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
178 (p)[1] = (0x80 | ((c) & 0x3F)), \
180 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
181 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
182 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
183 (p)[2] = (0x80 | ((c) & 0x3F)), \
185 : char_string ((unsigned) c, p))
187 /* Store multibyte form of byte B in P. The caller should allocate at
188 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
189 length of the multibyte form. */
191 #define BYTE8_STRING(b, p) \
192 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
193 (p)[1] = (0x80 | ((b) & 0x3F)), \
197 /* Store multibyte form of the character C in P. The caller should
198 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
199 And, advance P to the end of the multibyte form. */
201 #define CHAR_STRING_ADVANCE(c, p) \
203 if ((c) <= MAX_1_BYTE_CHAR) \
205 else if ((c) <= MAX_2_BYTE_CHAR) \
206 *(p)++ = (0xC0 | ((c) >> 6)), \
207 *(p)++ = (0x80 | ((c) & 0x3F)); \
208 else if ((c) <= MAX_3_BYTE_CHAR) \
209 *(p)++ = (0xE0 | ((c) >> 12)), \
210 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
211 *(p)++ = (0x80 | ((c) & 0x3F)); \
213 (p) += char_string ((c), (p)); \
217 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
219 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
221 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
223 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
225 /* Nonzero iff BYTE starts a character in a multibyte form.
226 This is equivalent to:
227 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
228 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
230 /* Just kept for backward compatibility. This macro will be removed
232 #define BASE_LEADING_CODE_P LEADING_CODE_P
234 /* How many bytes a character that starts with BYTE occupies in a
236 #define BYTES_BY_CHAR_HEAD(byte) \
237 (!((byte) & 0x80) ? 1 \
238 : !((byte) & 0x20) ? 2 \
239 : !((byte) & 0x10) ? 3 \
240 : !((byte) & 0x08) ? 4 \
244 /* Return the length of the multi-byte form at string STR of length
245 LEN while assuming that STR points a valid multi-byte form. As
246 this macro isn't necessary anymore, all callers will be changed to
247 use BYTES_BY_CHAR_HEAD directly in the future. */
249 #define MULTIBYTE_FORM_LENGTH(str, len) \
250 BYTES_BY_CHAR_HEAD (*(str))
252 /* Parse multibyte string STR of length LENGTH and set BYTES to the
253 byte length of a character at STR while assuming that STR points a
254 valid multibyte form. As this macro isn't necessary anymore, all
255 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
258 #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
259 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
261 /* The byte length of multibyte form at unibyte string P ending at
262 PEND. If STR doesn't point a valid multibyte form, return 0. */
264 #define MULTIBYTE_LENGTH(p, pend) \
266 : !((p)[0] & 0x80) ? 1 \
267 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
268 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
269 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
270 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
271 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
272 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
273 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
274 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
278 /* Like MULTIBYTE_LENGTH but don't check the ending address. */
280 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
281 (!((p)[0] & 0x80) ? 1 \
282 : ((p)[1] & 0xC0) != 0x80 ? 0 \
283 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
284 : ((p)[2] & 0xC0) != 0x80 ? 0 \
285 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
286 : ((p)[3] & 0xC0) != 0x80 ? 0 \
287 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
288 : ((p)[4] & 0xC0) != 0x80 ? 0 \
289 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
292 /* If P is before LIMIT, advance P to the next character boundary. It
293 assumes that P is already at a character boundary of the sane
294 mulitbyte form whose end address is LIMIT. */
296 #define NEXT_CHAR_BOUNDARY(p, limit) \
299 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
303 /* If P is after LIMIT, advance P to the previous character boundary.
304 It assumes that P is already at a character boundary of the sane
305 mulitbyte form whose beginning address is LIMIT. */
307 #define PREV_CHAR_BOUNDARY(p, limit) \
311 const unsigned char *p0 = (p); \
314 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
315 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
319 /* Return the character code of character whose multibyte form is at
320 P. The argument LEN is ignored. It will be removed in the
323 #define STRING_CHAR(p, len) \
326 : ! ((p)[0] & 0x20) \
327 ? (((((p)[0] & 0x1F) << 6) \
329 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
330 : ! ((p)[0] & 0x10) \
331 ? ((((p)[0] & 0x0F) << 12) \
332 | (((p)[1] & 0x3F) << 6) \
334 : string_char ((p), NULL, NULL))
337 /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte
338 form. The argument LEN is ignored. It will be removed in the
341 #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
343 ? ((actual_len) = 1, (p)[0]) \
344 : ! ((p)[0] & 0x20) \
345 ? ((actual_len) = 2, \
346 (((((p)[0] & 0x1F) << 6) \
348 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
349 : ! ((p)[0] & 0x10) \
350 ? ((actual_len) = 3, \
351 ((((p)[0] & 0x0F) << 12) \
352 | (((p)[1] & 0x3F) << 6) \
353 | ((p)[2] & 0x3F))) \
354 : string_char ((p), NULL, &actual_len))
357 /* Like STRING_CHAR but advance P to the end of multibyte form. */
359 #define STRING_CHAR_ADVANCE(p) \
362 : ! ((p)[0] & 0x20) \
364 ((((p)[-2] & 0x1F) << 6) \
366 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
367 : ! ((p)[0] & 0x10) \
369 ((((p)[-3] & 0x0F) << 12) \
370 | (((p)[-2] & 0x3F) << 6) \
371 | ((p)[-1] & 0x3F))) \
372 : string_char ((p), &(p), NULL))
375 /* Fetch the "next" character from Lisp string STRING at byte position
376 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
378 All the args must be side-effect-free.
379 BYTEIDX and CHARIDX must be lvalues;
380 we increment them past the character fetched. */
382 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
386 if (STRING_MULTIBYTE (STRING)) \
388 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
391 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
396 OUTPUT = SREF (STRING, BYTEIDX); \
402 /* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve
403 if STRING is unibyte. */
405 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
409 if (STRING_MULTIBYTE (STRING)) \
411 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
414 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
419 OUTPUT = SREF (STRING, BYTEIDX); \
421 MAKE_CHAR_MULTIBYTE (OUTPUT); \
427 /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
429 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
432 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
435 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
442 /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
445 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
449 if (!NILP (current_buffer->enable_multibyte_characters)) \
451 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
454 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
459 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
466 /* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */
468 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
471 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
474 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
481 /* Increase the buffer byte position POS_BYTE of the current buffer to
482 the next character boundary. No range checking of POS. */
484 #define INC_POS(pos_byte) \
486 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
487 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
491 /* Decrease the buffer byte position POS_BYTE of the current buffer to
492 the previous character boundary. No range checking of POS. */
494 #define DEC_POS(pos_byte) \
499 if (pos_byte < GPT_BYTE) \
500 p = BEG_ADDR + pos_byte - BEG_BYTE; \
502 p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\
503 while (!CHAR_HEAD_P (*p)) \
510 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
512 #define INC_BOTH(charpos, bytepos) \
516 if (NILP (current_buffer->enable_multibyte_characters)) \
519 INC_POS ((bytepos)); \
524 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
526 #define DEC_BOTH(charpos, bytepos) \
530 if (NILP (current_buffer->enable_multibyte_characters)) \
533 DEC_POS ((bytepos)); \
538 /* Increase the buffer byte position POS_BYTE of the current buffer to
539 the next character boundary. This macro relies on the fact that
540 *GPT_ADDR and *Z_ADDR are always accessible and the values are
541 '\0'. No range checking of POS_BYTE. */
543 #define BUF_INC_POS(buf, pos_byte) \
545 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
546 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
550 /* Decrease the buffer byte position POS_BYTE of the current buffer to
551 the previous character boundary. No range checking of POS_BYTE. */
553 #define BUF_DEC_POS(buf, pos_byte) \
557 if (pos_byte < BUF_GPT_BYTE (buf)) \
558 p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \
560 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\
561 while (!CHAR_HEAD_P (*p)) \
569 /* If C is a character to be unified with a Unicode character, return
570 the unified Unicode character. */
572 #define MAYBE_UNIFY_CHAR(c) \
573 if (c > MAX_UNICODE_CHAR \
574 && CHAR_TABLE_P (Vchar_unify_table)) \
579 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
584 Funify_charset (val, Qnil, Qnil); \
585 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
587 if ((unified = XINT (val)) >= 0) \
594 /* Return the width of ASCII character C. The width is measured by
595 how many columns occupied on the screen when displayed in the
598 #define ASCII_CHAR_WIDTH(c) \
601 ? XFASTINT (current_buffer->tab_width) \
602 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
605 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
607 /* Return the width of character C. The width is measured by how many
608 columns occupied on the screen when displayed in the current
611 #define CHAR_WIDTH(c) \
613 ? ASCII_CHAR_WIDTH (c) \
614 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
616 extern int char_resolve_modifier_mask
P_ ((int));
617 extern int char_string
P_ ((unsigned, unsigned char *));
618 extern int string_char
P_ ((const unsigned char *,
619 const unsigned char **, int *));
621 extern int translate_char
P_ ((Lisp_Object
, int c
));
622 extern int char_printable_p
P_ ((int c
));
623 extern void parse_str_as_multibyte
P_ ((const unsigned char *, int, int *,
625 extern int parse_str_to_multibyte
P_ ((unsigned char *, int));
626 extern int str_as_multibyte
P_ ((unsigned char *, int, int, int *));
627 extern int str_to_multibyte
P_ ((unsigned char *, int, int));
628 extern int str_as_unibyte
P_ ((unsigned char *, int));
629 extern int strwidth
P_ ((unsigned char *, int));
630 extern int c_string_width
P_ ((const unsigned char *, int, int, int *, int *));
631 extern int lisp_string_width
P_ ((Lisp_Object
, int, int *, int *));
633 extern Lisp_Object Vprintable_chars
;
635 extern Lisp_Object Qcharacterp
, Qauto_fill_chars
;
636 extern Lisp_Object Vtranslation_table_vector
;
637 extern Lisp_Object Vchar_width_table
;
638 extern Lisp_Object Vchar_direction_table
;
639 extern Lisp_Object Vchar_unify_table
;
641 extern Lisp_Object string_escape_byte8
P_ ((Lisp_Object
));
643 /* Return a translation table of id number ID. */
644 #define GET_TRANSLATION_TABLE(id) \
645 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
647 /* A char-table for characters which may invoke auto-filling. */
648 extern Lisp_Object Vauto_fill_chars
;
650 extern Lisp_Object Vchar_script_table
;
651 extern Lisp_Object Vscript_representative_chars
;
653 /* Copy LEN bytes from FROM to TO. This macro should be used only
654 when a caller knows that LEN is short and the obvious copy loop is
655 faster than calling bcopy which has some overhead. Copying a
656 multibyte sequence of a character is the typical case. */
658 #define BCOPY_SHORT(from, to, len) \
661 unsigned char *from_p = from, *to_p = to; \
662 while (i--) *to_p++ = *from_p++; \
665 #define DEFSYM(sym, name) \
666 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
668 #endif /* EMACS_CHARACTER_H */
670 /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac
671 (do not change this comment) */