1 /* Header for multibyte character handler.
2 Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN.
3 Licensed to the Free Software Foundation.
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 2, 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
22 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 Boston, MA 02111-1307, 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
38 3FFF80-3FFFFF C0..C1 1100000x 10xxxxxx
41 /* Maximum character code ((1 << CHARACTERBITS) - 1). */
42 #define MAX_CHAR 0x3FFFFF
44 /* Maximum Unicode character code. */
45 #define MAX_UNICODE_CHAR 0x10FFFF
47 /* Maximum N-byte character codes. */
48 #define MAX_1_BYTE_CHAR 0x7F
49 #define MAX_2_BYTE_CHAR 0x7FF
50 #define MAX_3_BYTE_CHAR 0xFFFF
51 #define MAX_4_BYTE_CHAR 0x1FFFFF
52 #define MAX_5_BYTE_CHAR 0x3FFF7F
54 /* Leading code range of Latin-1 chars. */
55 #define LEADING_CODE_LATIN_1_MIN 0xC2
56 #define LEADING_CODE_LATIN_1_MAX 0xC3
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 /* If C is not ASCII, make it unibyte. */
84 #define MAKE_CHAR_UNIBYTE(c) \
86 if (! ASCII_CHAR_P (c)) \
87 c = CHAR_TO_BYTE8 (c); \
91 /* If C is not ASCII, make it multibyte. It assumes C < 256. */
92 #define MAKE_CHAR_MULTIBYTE(c) ((c) = unibyte_to_multibyte_table[(c)])
94 /* This is the maximum byte length of multibyte form. */
95 #define MAX_MULTIBYTE_LENGTH 5
97 /* Return a Lisp character whose character code is C. */
98 #define make_char(c) make_number (c)
100 /* Nonzero iff C is an ASCII byte. */
101 #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
103 /* Nonzero iff X is a character. */
104 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
106 /* Nonzero iff C is valid as a character code. GENERICP is not used
108 #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
110 /* Check if Lisp object X is a character or not. */
111 #define CHECK_CHARACTER(x) \
113 if (! CHARACTERP(x)) x = wrong_type_argument (Qcharacterp, (x)); \
116 #define CHECK_CHARACTER_CAR(x) \
118 Lisp_Object tmp = XCAR (x); \
119 CHECK_CHARACTER (tmp); \
120 XSETCAR ((x), tmp); \
123 #define CHECK_CHARACTER_CDR(x) \
125 Lisp_Object tmp = XCDR (x); \
126 CHECK_CHARACTER (tmp); \
127 XSETCDR ((x), tmp); \
130 /* Nonzero iff C is an ASCII character. */
131 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
133 /* Nonzero iff C is a character of code less than 0x100. */
134 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
136 /* Nonzero if character C has a printable glyph. */
137 #define CHAR_PRINTABLE_P(c) \
138 (((c) >= 32 && ((c) < 127) \
139 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
141 /* Return byte length of multibyte form for character C. */
142 #define CHAR_BYTES(c) \
143 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
144 : (c) <= MAX_2_BYTE_CHAR ? 2 \
145 : (c) <= MAX_3_BYTE_CHAR ? 3 \
146 : (c) <= MAX_4_BYTE_CHAR ? 4 \
147 : (c) <= MAX_5_BYTE_CHAR ? 5 \
151 /* Return the leading code of multibyte form of C. */
152 #define CHAR_LEADING_CODE(c) \
153 ((c) <= MAX_1_BYTE_CHAR ? c \
154 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
155 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
156 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
157 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
158 : (0xC0 | (((c) >> 6) & 0x01)))
161 /* Store multibyte form of the character C in P. The caller should
162 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
163 Returns the length of the multibyte form. */
165 #define CHAR_STRING(c, p) \
166 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
169 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
170 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
171 (p)[1] = (0x80 | ((c) & 0x3F)), \
173 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
174 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
175 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
176 (p)[2] = (0x80 | ((c) & 0x3F)), \
178 : char_string (c, p))
180 /* Store multibyte form of byte B in P. The caller should allocate at
181 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
182 length of the multibyte form. */
184 #define BYTE8_STRING(b, p) \
185 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
186 (p)[1] = (0x80 | ((c) & 0x3F)), \
190 /* Store multibyte form of the character C in P. The caller should
191 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
192 And, advance P to the end of the multibyte form. */
194 #define CHAR_STRING_ADVANCE(c, p) \
196 if ((c) <= MAX_1_BYTE_CHAR) \
198 else if ((c) <= MAX_2_BYTE_CHAR) \
199 *(p)++ = (0xC0 | ((c) >> 6)), \
200 *(p)++ = (0x80 | ((c) & 0x3F)); \
201 else if ((c) <= MAX_3_BYTE_CHAR) \
202 *(p)++ = (0xE0 | ((c) >> 12)), \
203 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
204 *(p)++ = (0x80 | ((c) & 0x3F)); \
206 (p) += char_string ((c), (p)); \
210 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
212 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
214 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
216 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
218 /* Nonzero iff BYTE starts a character in a multibyte form.
219 This is equivalent to:
220 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
221 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
223 /* Just kept for backward compatibility. This macro will be removed
225 #define BASE_LEADING_CODE_P LEADING_CODE_P
227 /* How many bytes a character that starts with BYTE occupies in a
229 #define BYTES_BY_CHAR_HEAD(byte) \
230 (!((byte) & 0x80) ? 1 \
231 : !((byte) & 0x20) ? 2 \
232 : !((byte) & 0x10) ? 3 \
233 : !((byte) & 0x08) ? 4 \
237 /* Return the length of the multi-byte form at string STR of length
238 LEN while assuming that STR points a valid multi-byte form. As
239 this macro isn't necessary anymore, all callers will be changed to
240 use BYTES_BY_CHAR_HEAD directly in the future. */
242 #define MULTIBYTE_FORM_LENGTH(str, len) \
243 BYTES_BY_CHAR_HEAD (*(str))
245 /* Parse multibyte string STR of length LENGTH and set BYTES to the
246 byte length of a character at STR while assuming that STR points a
247 valid multibyte form. As this macro isn't necessary anymore, all
248 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
251 #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
252 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
254 /* The byte length of multibyte form at unibyte string P ending at
255 PEND. If STR doesn't point a valid multibyte form, return 0. */
257 #define MULTIBYTE_LENGTH(p, pend) \
259 : !((p)[0] & 0x80) ? 1 \
260 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
261 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
262 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
263 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
264 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
265 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
266 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
267 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
271 /* Like MULTIBYTE_LENGTH but don't check the ending address. */
273 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
274 (!((p)[0] & 0x80) ? 1 \
275 : ((p)[1] & 0xC0) != 0x80 ? 0 \
276 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
277 : ((p)[2] & 0xC0) != 0x80 ? 0 \
278 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
279 : ((p)[3] & 0xC0) != 0x80 ? 0 \
280 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
281 : ((p)[4] & 0xC0) != 0x80 ? 0 \
282 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
285 /* If P is before LIMIT, advance P to the next character boundary. It
286 assumes that P is already at a character boundary of the sane
287 mulitbyte form whose end address is LIMIT. */
289 #define NEXT_CHAR_BOUNDARY(p, limit) \
292 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
296 /* If P is after LIMIT, advance P to the previous character boundary.
297 It assumes that P is already at a character boundary of the sane
298 mulitbyte form whose beginning address is LIMIT. */
300 #define PREV_CHAR_BOUNDARY(p, limit) \
304 const unsigned char *p0 = (p); \
307 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
308 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
312 /* Return the character code of character whose multibyte form is at
313 P. The argument LEN is ignored. It will be removed in the
316 #define STRING_CHAR(p, len) \
319 : ! ((p)[0] & 0x20) \
320 ? (((((p)[0] & 0x1F) << 6) \
322 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
323 : ! ((p)[0] & 0x10) \
324 ? ((((p)[0] & 0x0F) << 12) \
325 | (((p)[1] & 0x3F) << 6) \
327 : string_char ((p), NULL, NULL))
330 /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte
331 form. The argument LEN is ignored. It will be removed in the
334 #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
336 ? ((actual_len) = 1, (p)[0]) \
337 : ! ((p)[0] & 0x20) \
338 ? ((actual_len) = 2, \
339 (((((p)[0] & 0x1F) << 6) \
341 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
342 : ! ((p)[0] & 0x10) \
343 ? ((actual_len) = 3, \
344 ((((p)[0] & 0x0F) << 12) \
345 | (((p)[1] & 0x3F) << 6) \
346 | ((p)[2] & 0x3F))) \
347 : string_char ((p), NULL, &actual_len))
350 /* Like STRING_CHAR but advacen P to the end of multibyte form. */
352 #define STRING_CHAR_ADVANCE(p) \
355 : ! ((p)[0] & 0x20) \
357 ((((p)[-2] & 0x1F) << 6) \
359 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
360 : ! ((p)[0] & 0x10) \
362 ((((p)[-3] & 0x0F) << 12) \
363 | (((p)[-2] & 0x3F) << 6) \
364 | ((p)[-1] & 0x3F))) \
365 : string_char ((p), &(p), NULL))
368 /* Fetch the "next" character from Lisp string STRING at byte position
369 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
371 All the args must be side-effect-free.
372 BYTEIDX and CHARIDX must be lvalues;
373 we increment them past the character fetched. */
375 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
379 if (STRING_MULTIBYTE (STRING)) \
381 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
384 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
388 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
392 /* Like FETCH_STRING_CHAR_ADVANCE */
394 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
398 if (STRING_MULTIBYTE (STRING)) \
400 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
403 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
408 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
409 MAKE_CHAR_MULTIBYTE (OUTPUT); \
415 /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
417 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
420 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
423 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
430 /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
433 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
437 if (!NILP (current_buffer->enable_multibyte_characters)) \
439 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
442 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
447 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
454 /* Like FETCH_CHAR_ADVANCE but assumes STRING is multibyte. */
456 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
459 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
462 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
469 /* Increase the buffer byte position POS_BYTE of the current buffer to
470 the next character boundary. No range checking of POS. */
472 #define INC_POS(pos_byte) \
474 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
475 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
479 /* Decrease the buffer byte position POS_BYTE of the current buffer to
480 the previous character boundary. No range checking of POS. */
482 #define DEC_POS(pos_byte) \
487 if (pos_byte < GPT_BYTE) \
488 p = BEG_ADDR + pos_byte - 1; \
490 p = BEG_ADDR + GAP_SIZE + pos_byte - 1; \
491 while (!CHAR_HEAD_P (*p)) \
498 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
500 #define INC_BOTH(charpos, bytepos) \
504 if (NILP (current_buffer->enable_multibyte_characters)) \
507 INC_POS ((bytepos)); \
512 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
514 #define DEC_BOTH(charpos, bytepos) \
518 if (NILP (current_buffer->enable_multibyte_characters)) \
521 DEC_POS ((bytepos)); \
526 /* Increase the buffer byte position POS_BYTE of the current buffer to
527 the next character boundary. This macro relies on the fact that
528 *GPT_ADDR and *Z_ADDR are always accessible and the values are
529 '\0'. No range checking of POS_BYTE. */
531 #define BUF_INC_POS(buf, pos_byte) \
533 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
534 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
538 /* Decrease the buffer byte position POS_BYTE of the current buffer to
539 the previous character boundary. No range checking of POS_BYTE. */
541 #define BUF_DEC_POS(buf, pos_byte) \
545 if (pos_byte < BUF_GPT_BYTE (buf)) \
546 p = BUF_BEG_ADDR (buf) + pos_byte - 1; \
548 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - 1; \
549 while (!CHAR_HEAD_P (*p)) \
557 #define MAYBE_UNIFY_CHAR(c) \
558 if (c > MAX_UNICODE_CHAR \
559 && CHAR_TABLE_P (Vchar_unify_table)) \
564 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
569 Funify_charset (val, Qnil, Qnil); \
570 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
572 if ((unified = XINT (val)) >= 0) \
579 /* Return the width of ASCII character C. The width is measured by
580 how many columns occupied on the screen when displayed in the
583 #define ASCII_CHAR_WIDTH(c) \
586 ? XFASTINT (current_buffer->tab_width) \
587 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
590 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
592 /* Return the width of character C. The width is measured by how many
593 columns occupied on the screen when displayed in the current
596 #define CHAR_WIDTH(c) \
598 ? ASCII_CHAR_WIDTH (c) \
599 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
601 extern int char_resolve_modifier_mask
P_ ((int));
602 extern int char_string
P_ ((int, unsigned char *));
603 extern int string_char
P_ ((const unsigned char *,
604 const unsigned char **, int *));
606 extern int translate_char
P_ ((Lisp_Object
, int c
));
607 extern int char_printable_p
P_ ((int c
));
608 extern void parse_str_as_multibyte
P_ ((const unsigned char *, int, int *,
610 extern int parse_str_to_multibyte
P_ ((unsigned char *, int));
611 extern int str_as_multibyte
P_ ((unsigned char *, int, int, int *));
612 extern int str_to_multibyte
P_ ((unsigned char *, int, int));
613 extern int str_as_unibyte
P_ ((unsigned char *, int));
614 extern int strwidth
P_ ((unsigned char *, int));
615 extern int c_string_width
P_ ((const unsigned char *, int, int, int *, int *));
616 extern int lisp_string_width
P_ ((Lisp_Object
, int, int *, int *));
618 extern Lisp_Object Vprintable_chars
;
620 extern Lisp_Object Qcharacterp
, Qauto_fill_chars
;
621 extern Lisp_Object Vtranslation_table_vector
;
622 extern Lisp_Object Vchar_width_table
;
623 extern Lisp_Object Vchar_direction_table
;
624 extern Lisp_Object Vchar_unify_table
;
626 extern Lisp_Object string_escape_byte8
P_ ((Lisp_Object
));
628 /* Return a translation table of id number ID. */
629 #define GET_TRANSLATION_TABLE(id) \
630 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
632 /* A char-table for characters which may invoke auto-filling. */
633 extern Lisp_Object Vauto_fill_chars
;
635 extern Lisp_Object Vchar_script_table
;
637 /* Copy LEN bytes from FROM to TO. This macro should be used only
638 when a caller knows that LEN is short and the obvious copy loop is
639 faster than calling bcopy which has some overhead. Copying a
640 multibyte sequence of a character is the typical case. */
642 #define BCOPY_SHORT(from, to, len) \
645 unsigned char *from_p = from, *to_p = to; \
646 while (i--) *to_p++ = *from_p++; \
649 #define DEFSYM(sym, name) \
650 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
652 #endif /* EMACS_CHARACTER_H */