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 of the License, or
13 (at your option) any later version.
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. If not, see <http://www.gnu.org/licenses/>. */
23 #ifndef EMACS_CHARACTER_H
24 #define EMACS_CHARACTER_H
26 /* character code 1st byte byte sequence
27 -------------- -------- -------------
29 80-7FF C2..DF 110xxxxx 10xxxxxx
30 800-FFFF E0..EF 1110xxxx 10xxxxxx 10xxxxxx
31 10000-1FFFFF F0..F7 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
32 200000-3FFF7F F8 11111000 1000xxxx 10xxxxxx 10xxxxxx 10xxxxxx
33 3FFF80-3FFFFF C0..C1 1100000x 10xxxxxx (for eight-bit-char)
36 invalid 1st byte 80..BF 10xxxxxx
37 F9..FF 11111xxx (xxx != 000)
40 /* Maximum character code ((1 << CHARACTERBITS) - 1). */
41 #define MAX_CHAR 0x3FFFFF
43 /* Maximum Unicode character code. */
44 #define MAX_UNICODE_CHAR 0x10FFFF
46 /* Maximum N-byte character codes. */
47 #define MAX_1_BYTE_CHAR 0x7F
48 #define MAX_2_BYTE_CHAR 0x7FF
49 #define MAX_3_BYTE_CHAR 0xFFFF
50 #define MAX_4_BYTE_CHAR 0x1FFFFF
51 #define MAX_5_BYTE_CHAR 0x3FFF7F
53 /* Minimum leading code of multibyte characters. */
54 #define MIN_MULTIBYTE_LEADING_CODE 0xC0
55 /* Maximum leading code of multibyte characters. */
56 #define MAX_MULTIBYTE_LEADING_CODE 0xF8
58 /* Nonzero iff C is a character that corresponds to a raw 8-bit
60 #define CHAR_BYTE8_P(c) ((c) > MAX_5_BYTE_CHAR)
62 /* Return the character code for raw 8-bit byte BYTE. */
63 #define BYTE8_TO_CHAR(byte) ((byte) + 0x3FFF00)
65 /* Return the raw 8-bit byte for character C. */
66 #define CHAR_TO_BYTE8(c) \
69 : multibyte_char_to_unibyte (c, Qnil))
71 /* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
72 that corresponds to a raw 8-bit byte. */
73 #define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
75 /* Mapping table from unibyte chars to multibyte chars. */
76 extern int unibyte_to_multibyte_table
[256];
78 /* Convert the unibyte character C to the corresponding multibyte
79 character. If C can't be converted, return C. */
80 #define unibyte_char_to_multibyte(c) \
81 ((c) < 256 ? unibyte_to_multibyte_table[(c)] : (c))
83 /* Nth element is 1 iff unibyte char N can be mapped to a multibyte
85 extern char unibyte_has_multibyte_table
[256];
87 #define UNIBYTE_CHAR_HAS_MULTIBYTE_P(c) (unibyte_has_multibyte_table[(c)])
89 /* If C is not ASCII, make it unibyte. */
90 #define MAKE_CHAR_UNIBYTE(c) \
92 if (! ASCII_CHAR_P (c)) \
93 c = CHAR_TO_BYTE8 (c); \
97 /* If C is not ASCII, make it multibyte. It assumes C < 256. */
98 #define MAKE_CHAR_MULTIBYTE(c) ((c) = unibyte_to_multibyte_table[(c)])
100 /* This is the maximum byte length of multibyte form. */
101 #define MAX_MULTIBYTE_LENGTH 5
103 /* Return a Lisp character whose character code is C. It assumes C is
104 a valid character code. */
105 #define make_char(c) make_number (c)
107 /* Nonzero iff C is an ASCII byte. */
108 #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
110 /* Nonzero iff X is a character. */
111 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
113 /* Nonzero iff C is valid as a character code. GENERICP is not used
115 #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
117 /* Check if Lisp object X is a character or not. */
118 #define CHECK_CHARACTER(x) \
119 CHECK_TYPE (CHARACTERP (x), Qcharacterp, x)
121 #define CHECK_CHARACTER_CAR(x) \
123 Lisp_Object tmp = XCAR (x); \
124 CHECK_CHARACTER (tmp); \
125 XSETCAR ((x), tmp); \
128 #define CHECK_CHARACTER_CDR(x) \
130 Lisp_Object tmp = XCDR (x); \
131 CHECK_CHARACTER (tmp); \
132 XSETCDR ((x), tmp); \
135 /* Nonzero iff C is an ASCII character. */
136 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
138 /* Nonzero iff C is a character of code less than 0x100. */
139 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
141 /* Nonzero if character C has a printable glyph. */
142 #define CHAR_PRINTABLE_P(c) \
143 (((c) >= 32 && ((c) < 127) \
144 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
146 /* Return byte length of multibyte form for character C. */
147 #define CHAR_BYTES(c) \
148 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
149 : (c) <= MAX_2_BYTE_CHAR ? 2 \
150 : (c) <= MAX_3_BYTE_CHAR ? 3 \
151 : (c) <= MAX_4_BYTE_CHAR ? 4 \
152 : (c) <= MAX_5_BYTE_CHAR ? 5 \
156 /* Return the leading code of multibyte form of C. */
157 #define CHAR_LEADING_CODE(c) \
158 ((c) <= MAX_1_BYTE_CHAR ? c \
159 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
160 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
161 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
162 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
163 : (0xC0 | (((c) >> 6) & 0x01)))
166 /* Store multibyte form of the character C in P. The caller should
167 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
168 Returns the length of the multibyte form. */
170 #define CHAR_STRING(c, p) \
171 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
174 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
175 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
176 (p)[1] = (0x80 | ((c) & 0x3F)), \
178 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
179 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
180 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
181 (p)[2] = (0x80 | ((c) & 0x3F)), \
183 : char_string ((unsigned) c, p))
185 /* Store multibyte form of byte B in P. The caller should allocate at
186 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
187 length of the multibyte form. */
189 #define BYTE8_STRING(b, p) \
190 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
191 (p)[1] = (0x80 | ((b) & 0x3F)), \
195 /* Store multibyte form of the character C in P. The caller should
196 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
197 And, advance P to the end of the multibyte form. */
199 #define CHAR_STRING_ADVANCE(c, p) \
201 if ((c) <= MAX_1_BYTE_CHAR) \
203 else if ((c) <= MAX_2_BYTE_CHAR) \
204 *(p)++ = (0xC0 | ((c) >> 6)), \
205 *(p)++ = (0x80 | ((c) & 0x3F)); \
206 else if ((c) <= MAX_3_BYTE_CHAR) \
207 *(p)++ = (0xE0 | ((c) >> 12)), \
208 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
209 *(p)++ = (0x80 | ((c) & 0x3F)); \
211 (p) += char_string ((c), (p)); \
215 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
217 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
219 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
221 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
223 /* Nonzero iff BYTE starts a character in a multibyte form.
224 This is equivalent to:
225 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
226 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
228 /* Just kept for backward compatibility. This macro will be removed
230 #define BASE_LEADING_CODE_P LEADING_CODE_P
232 /* How many bytes a character that starts with BYTE occupies in a
234 #define BYTES_BY_CHAR_HEAD(byte) \
235 (!((byte) & 0x80) ? 1 \
236 : !((byte) & 0x20) ? 2 \
237 : !((byte) & 0x10) ? 3 \
238 : !((byte) & 0x08) ? 4 \
242 /* Return the length of the multi-byte form at string STR of length
243 LEN while assuming that STR points a valid multi-byte form. As
244 this macro isn't necessary anymore, all callers will be changed to
245 use BYTES_BY_CHAR_HEAD directly in the future. */
247 #define MULTIBYTE_FORM_LENGTH(str, len) \
248 BYTES_BY_CHAR_HEAD (*(str))
250 /* Parse multibyte string STR of length LENGTH and set BYTES to the
251 byte length of a character at STR while assuming that STR points a
252 valid multibyte form. As this macro isn't necessary anymore, all
253 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
256 #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
257 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
259 /* The byte length of multibyte form at unibyte string P ending at
260 PEND. If STR doesn't point a valid multibyte form, return 0. */
262 #define MULTIBYTE_LENGTH(p, pend) \
264 : !((p)[0] & 0x80) ? 1 \
265 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
266 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
267 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
268 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
269 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
270 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
271 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
272 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
276 /* Like MULTIBYTE_LENGTH but don't check the ending address. */
278 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
279 (!((p)[0] & 0x80) ? 1 \
280 : ((p)[1] & 0xC0) != 0x80 ? 0 \
281 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
282 : ((p)[2] & 0xC0) != 0x80 ? 0 \
283 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
284 : ((p)[3] & 0xC0) != 0x80 ? 0 \
285 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
286 : ((p)[4] & 0xC0) != 0x80 ? 0 \
287 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
290 /* If P is before LIMIT, advance P to the next character boundary. It
291 assumes that P is already at a character boundary of the sane
292 mulitbyte form whose end address is LIMIT. */
294 #define NEXT_CHAR_BOUNDARY(p, limit) \
297 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
301 /* If P is after LIMIT, advance P to the previous character boundary.
302 It assumes that P is already at a character boundary of the sane
303 mulitbyte form whose beginning address is LIMIT. */
305 #define PREV_CHAR_BOUNDARY(p, limit) \
309 const unsigned char *p0 = (p); \
312 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
313 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
317 /* Return the character code of character whose multibyte form is at
318 P. The argument LEN is ignored. It will be removed in the
321 #define STRING_CHAR(p, len) \
324 : ! ((p)[0] & 0x20) \
325 ? (((((p)[0] & 0x1F) << 6) \
327 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
328 : ! ((p)[0] & 0x10) \
329 ? ((((p)[0] & 0x0F) << 12) \
330 | (((p)[1] & 0x3F) << 6) \
332 : string_char ((p), NULL, NULL))
335 /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte
336 form. The argument LEN is ignored. It will be removed in the
339 #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
341 ? ((actual_len) = 1, (p)[0]) \
342 : ! ((p)[0] & 0x20) \
343 ? ((actual_len) = 2, \
344 (((((p)[0] & 0x1F) << 6) \
346 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
347 : ! ((p)[0] & 0x10) \
348 ? ((actual_len) = 3, \
349 ((((p)[0] & 0x0F) << 12) \
350 | (((p)[1] & 0x3F) << 6) \
351 | ((p)[2] & 0x3F))) \
352 : string_char ((p), NULL, &actual_len))
355 /* Like STRING_CHAR but advance P to the end of multibyte form. */
357 #define STRING_CHAR_ADVANCE(p) \
360 : ! ((p)[0] & 0x20) \
362 ((((p)[-2] & 0x1F) << 6) \
364 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
365 : ! ((p)[0] & 0x10) \
367 ((((p)[-3] & 0x0F) << 12) \
368 | (((p)[-2] & 0x3F) << 6) \
369 | ((p)[-1] & 0x3F))) \
370 : string_char ((p), &(p), NULL))
373 /* Fetch the "next" character from Lisp string STRING at byte position
374 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
376 All the args must be side-effect-free.
377 BYTEIDX and CHARIDX must be lvalues;
378 we increment them past the character fetched. */
380 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
384 if (STRING_MULTIBYTE (STRING)) \
386 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
389 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
394 OUTPUT = SREF (STRING, BYTEIDX); \
400 /* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve
401 if STRING is unibyte. */
403 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
407 if (STRING_MULTIBYTE (STRING)) \
409 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
412 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
417 OUTPUT = SREF (STRING, BYTEIDX); \
419 MAKE_CHAR_MULTIBYTE (OUTPUT); \
425 /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
427 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
430 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
433 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
440 /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
443 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
447 if (!NILP (current_buffer->enable_multibyte_characters)) \
449 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
452 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
457 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
464 /* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */
466 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
469 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
472 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
479 /* Increase the buffer byte position POS_BYTE of the current buffer to
480 the next character boundary. No range checking of POS. */
482 #define INC_POS(pos_byte) \
484 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
485 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
489 /* Decrease the buffer byte position POS_BYTE of the current buffer to
490 the previous character boundary. No range checking of POS. */
492 #define DEC_POS(pos_byte) \
497 if (pos_byte < GPT_BYTE) \
498 p = BEG_ADDR + pos_byte - BEG_BYTE; \
500 p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\
501 while (!CHAR_HEAD_P (*p)) \
508 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
510 #define INC_BOTH(charpos, bytepos) \
514 if (NILP (current_buffer->enable_multibyte_characters)) \
517 INC_POS ((bytepos)); \
522 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
524 #define DEC_BOTH(charpos, bytepos) \
528 if (NILP (current_buffer->enable_multibyte_characters)) \
531 DEC_POS ((bytepos)); \
536 /* Increase the buffer byte position POS_BYTE of the current buffer to
537 the next character boundary. This macro relies on the fact that
538 *GPT_ADDR and *Z_ADDR are always accessible and the values are
539 '\0'. No range checking of POS_BYTE. */
541 #define BUF_INC_POS(buf, pos_byte) \
543 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
544 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
548 /* Decrease the buffer byte position POS_BYTE of the current buffer to
549 the previous character boundary. No range checking of POS_BYTE. */
551 #define BUF_DEC_POS(buf, pos_byte) \
555 if (pos_byte < BUF_GPT_BYTE (buf)) \
556 p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \
558 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\
559 while (!CHAR_HEAD_P (*p)) \
567 /* If C is a character to be unified with a Unicode character, return
568 the unified Unicode character. */
570 #define MAYBE_UNIFY_CHAR(c) \
571 if (c > MAX_UNICODE_CHAR \
572 && CHAR_TABLE_P (Vchar_unify_table)) \
577 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
582 Funify_charset (val, Qnil, Qnil); \
583 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
585 if ((unified = XINT (val)) >= 0) \
592 /* Return the width of ASCII character C. The width is measured by
593 how many columns occupied on the screen when displayed in the
596 #define ASCII_CHAR_WIDTH(c) \
599 ? XFASTINT (current_buffer->tab_width) \
600 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
603 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
605 /* Return the width of character C. The width is measured by how many
606 columns occupied on the screen when displayed in the current
609 #define CHAR_WIDTH(c) \
611 ? ASCII_CHAR_WIDTH (c) \
612 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
614 extern int char_resolve_modifier_mask
P_ ((int));
615 extern int char_string
P_ ((unsigned, unsigned char *));
616 extern int string_char
P_ ((const unsigned char *,
617 const unsigned char **, int *));
619 extern int translate_char
P_ ((Lisp_Object
, int c
));
620 extern int char_printable_p
P_ ((int c
));
621 extern void parse_str_as_multibyte
P_ ((const unsigned char *, int, int *,
623 extern int parse_str_to_multibyte
P_ ((unsigned char *, int));
624 extern int str_as_multibyte
P_ ((unsigned char *, int, int, int *));
625 extern int str_to_multibyte
P_ ((unsigned char *, int, int));
626 extern int str_as_unibyte
P_ ((unsigned char *, int));
627 extern int strwidth
P_ ((unsigned char *, int));
628 extern int c_string_width
P_ ((const unsigned char *, int, int, int *, int *));
629 extern int lisp_string_width
P_ ((Lisp_Object
, int, int *, int *));
631 extern Lisp_Object Vprintable_chars
;
633 extern Lisp_Object Qcharacterp
, Qauto_fill_chars
;
634 extern Lisp_Object Vtranslation_table_vector
;
635 extern Lisp_Object Vchar_width_table
;
636 extern Lisp_Object Vchar_direction_table
;
637 extern Lisp_Object Vchar_unify_table
;
639 extern Lisp_Object string_escape_byte8
P_ ((Lisp_Object
));
641 /* Return a translation table of id number ID. */
642 #define GET_TRANSLATION_TABLE(id) \
643 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
645 /* A char-table for characters which may invoke auto-filling. */
646 extern Lisp_Object Vauto_fill_chars
;
648 extern Lisp_Object Vchar_script_table
;
649 extern Lisp_Object Vscript_representative_chars
;
651 /* Copy LEN bytes from FROM to TO. This macro should be used only
652 when a caller knows that LEN is short and the obvious copy loop is
653 faster than calling bcopy which has some overhead. Copying a
654 multibyte sequence of a character is the typical case. */
656 #define BCOPY_SHORT(from, to, len) \
659 unsigned char *from_p = from, *to_p = to; \
660 while (i--) *to_p++ = *from_p++; \
663 #define DEFSYM(sym, name) \
664 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
666 #endif /* EMACS_CHARACTER_H */
668 /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac
669 (do not change this comment) */