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
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 /* Nonzero iff C is a character that corresponds to a raw 8-bit
57 #define CHAR_BYTE8_P(c) ((c) > MAX_5_BYTE_CHAR)
59 /* Return the character code for raw 8-bit byte BYTE. */
60 #define BYTE8_TO_CHAR(byte) ((byte) + 0x3FFF00)
62 /* Return the raw 8-bit byte for character C. */
63 #define CHAR_TO_BYTE8(c) \
66 : multibyte_char_to_unibyte (c, Qnil))
68 /* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
69 that corresponds to a raw 8-bit byte. */
70 #define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
72 /* Mapping table from unibyte chars to multibyte chars. */
73 extern int unibyte_to_multibyte_table
[256];
75 /* Convert the unibyte character C to the corresponding multibyte
76 character. If C can't be converted, return C. */
77 #define unibyte_char_to_multibyte(c) \
78 ((c) < 256 ? unibyte_to_multibyte_table[(c)] : (c))
80 /* Nth element is 1 iff unibyte char N can be mapped to a multibyte
82 extern char unibyte_has_multibyte_table
[256];
84 #define UNIBYTE_CHAR_HAS_MULTIBYTE_P(c) (unibyte_has_multibyte_table[(c)])
86 /* If C is not ASCII, make it unibyte. */
87 #define MAKE_CHAR_UNIBYTE(c) \
89 if (! ASCII_CHAR_P (c)) \
90 c = CHAR_TO_BYTE8 (c); \
94 /* If C is not ASCII, make it multibyte. It assumes C < 256. */
95 #define MAKE_CHAR_MULTIBYTE(c) ((c) = unibyte_to_multibyte_table[(c)])
97 /* This is the maximum byte length of multibyte form. */
98 #define MAX_MULTIBYTE_LENGTH 5
100 /* Return a Lisp character whose character code is C. It assumes C is
101 a valid character code. */
102 #define make_char(c) make_number (c)
104 /* Nonzero iff C is an ASCII byte. */
105 #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
107 /* Nonzero iff X is a character. */
108 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
110 /* Nonzero iff C is valid as a character code. GENERICP is not used
112 #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
114 /* Check if Lisp object X is a character or not. */
115 #define CHECK_CHARACTER(x) \
117 if (! CHARACTERP(x)) x = wrong_type_argument (Qcharacterp, (x)); \
120 #define CHECK_CHARACTER_CAR(x) \
122 Lisp_Object tmp = XCAR (x); \
123 CHECK_CHARACTER (tmp); \
124 XSETCAR ((x), tmp); \
127 #define CHECK_CHARACTER_CDR(x) \
129 Lisp_Object tmp = XCDR (x); \
130 CHECK_CHARACTER (tmp); \
131 XSETCDR ((x), tmp); \
134 /* Nonzero iff C is an ASCII character. */
135 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
137 /* Nonzero iff C is a character of code less than 0x100. */
138 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
140 /* Nonzero if character C has a printable glyph. */
141 #define CHAR_PRINTABLE_P(c) \
142 (((c) >= 32 && ((c) < 127) \
143 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
145 /* Return byte length of multibyte form for character C. */
146 #define CHAR_BYTES(c) \
147 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
148 : (c) <= MAX_2_BYTE_CHAR ? 2 \
149 : (c) <= MAX_3_BYTE_CHAR ? 3 \
150 : (c) <= MAX_4_BYTE_CHAR ? 4 \
151 : (c) <= MAX_5_BYTE_CHAR ? 5 \
155 /* Return the leading code of multibyte form of C. */
156 #define CHAR_LEADING_CODE(c) \
157 ((c) <= MAX_1_BYTE_CHAR ? c \
158 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
159 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
160 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
161 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
162 : (0xC0 | (((c) >> 6) & 0x01)))
165 /* Store multibyte form of the character C in P. The caller should
166 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
167 Returns the length of the multibyte form. */
169 #define CHAR_STRING(c, p) \
170 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
173 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
174 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
175 (p)[1] = (0x80 | ((c) & 0x3F)), \
177 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
178 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
179 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
180 (p)[2] = (0x80 | ((c) & 0x3F)), \
182 : char_string (c, p))
184 /* Store multibyte form of byte B in P. The caller should allocate at
185 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
186 length of the multibyte form. */
188 #define BYTE8_STRING(b, p) \
189 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
190 (p)[1] = (0x80 | ((b) & 0x3F)), \
194 /* Store multibyte form of the character C in P. The caller should
195 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
196 And, advance P to the end of the multibyte form. */
198 #define CHAR_STRING_ADVANCE(c, p) \
200 if ((c) <= MAX_1_BYTE_CHAR) \
202 else if ((c) <= MAX_2_BYTE_CHAR) \
203 *(p)++ = (0xC0 | ((c) >> 6)), \
204 *(p)++ = (0x80 | ((c) & 0x3F)); \
205 else if ((c) <= MAX_3_BYTE_CHAR) \
206 *(p)++ = (0xE0 | ((c) >> 12)), \
207 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
208 *(p)++ = (0x80 | ((c) & 0x3F)); \
210 (p) += char_string ((c), (p)); \
214 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
216 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
218 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
220 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
222 /* Nonzero iff BYTE starts a character in a multibyte form.
223 This is equivalent to:
224 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
225 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
227 /* Just kept for backward compatibility. This macro will be removed
229 #define BASE_LEADING_CODE_P LEADING_CODE_P
231 /* How many bytes a character that starts with BYTE occupies in a
233 #define BYTES_BY_CHAR_HEAD(byte) \
234 (!((byte) & 0x80) ? 1 \
235 : !((byte) & 0x20) ? 2 \
236 : !((byte) & 0x10) ? 3 \
237 : !((byte) & 0x08) ? 4 \
241 /* Return the length of the multi-byte form at string STR of length
242 LEN while assuming that STR points a valid multi-byte form. As
243 this macro isn't necessary anymore, all callers will be changed to
244 use BYTES_BY_CHAR_HEAD directly in the future. */
246 #define MULTIBYTE_FORM_LENGTH(str, len) \
247 BYTES_BY_CHAR_HEAD (*(str))
249 /* Parse multibyte string STR of length LENGTH and set BYTES to the
250 byte length of a character at STR while assuming that STR points a
251 valid multibyte form. As this macro isn't necessary anymore, all
252 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
255 #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
256 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
258 /* The byte length of multibyte form at unibyte string P ending at
259 PEND. If STR doesn't point a valid multibyte form, return 0. */
261 #define MULTIBYTE_LENGTH(p, pend) \
263 : !((p)[0] & 0x80) ? 1 \
264 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
265 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
266 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
267 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
268 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
269 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
270 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
271 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
275 /* Like MULTIBYTE_LENGTH but don't check the ending address. */
277 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
278 (!((p)[0] & 0x80) ? 1 \
279 : ((p)[1] & 0xC0) != 0x80 ? 0 \
280 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
281 : ((p)[2] & 0xC0) != 0x80 ? 0 \
282 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
283 : ((p)[3] & 0xC0) != 0x80 ? 0 \
284 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
285 : ((p)[4] & 0xC0) != 0x80 ? 0 \
286 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
289 /* If P is before LIMIT, advance P to the next character boundary. It
290 assumes that P is already at a character boundary of the sane
291 mulitbyte form whose end address is LIMIT. */
293 #define NEXT_CHAR_BOUNDARY(p, limit) \
296 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
300 /* If P is after LIMIT, advance P to the previous character boundary.
301 It assumes that P is already at a character boundary of the sane
302 mulitbyte form whose beginning address is LIMIT. */
304 #define PREV_CHAR_BOUNDARY(p, limit) \
308 const unsigned char *p0 = (p); \
311 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
312 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
316 /* Return the character code of character whose multibyte form is at
317 P. The argument LEN is ignored. It will be removed in the
320 #define STRING_CHAR(p, len) \
323 : ! ((p)[0] & 0x20) \
324 ? (((((p)[0] & 0x1F) << 6) \
326 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
327 : ! ((p)[0] & 0x10) \
328 ? ((((p)[0] & 0x0F) << 12) \
329 | (((p)[1] & 0x3F) << 6) \
331 : string_char ((p), NULL, NULL))
334 /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte
335 form. The argument LEN is ignored. It will be removed in the
338 #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
340 ? ((actual_len) = 1, (p)[0]) \
341 : ! ((p)[0] & 0x20) \
342 ? ((actual_len) = 2, \
343 (((((p)[0] & 0x1F) << 6) \
345 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
346 : ! ((p)[0] & 0x10) \
347 ? ((actual_len) = 3, \
348 ((((p)[0] & 0x0F) << 12) \
349 | (((p)[1] & 0x3F) << 6) \
350 | ((p)[2] & 0x3F))) \
351 : string_char ((p), NULL, &actual_len))
354 /* Like STRING_CHAR but advance P to the end of multibyte form. */
356 #define STRING_CHAR_ADVANCE(p) \
359 : ! ((p)[0] & 0x20) \
361 ((((p)[-2] & 0x1F) << 6) \
363 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
364 : ! ((p)[0] & 0x10) \
366 ((((p)[-3] & 0x0F) << 12) \
367 | (((p)[-2] & 0x3F) << 6) \
368 | ((p)[-1] & 0x3F))) \
369 : string_char ((p), &(p), NULL))
372 /* Fetch the "next" character from Lisp string STRING at byte position
373 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
375 All the args must be side-effect-free.
376 BYTEIDX and CHARIDX must be lvalues;
377 we increment them past the character fetched. */
379 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
383 if (STRING_MULTIBYTE (STRING)) \
385 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
388 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
392 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
396 /* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve
397 if STRING is unibyte. */
399 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
403 if (STRING_MULTIBYTE (STRING)) \
405 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
408 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
413 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
414 MAKE_CHAR_MULTIBYTE (OUTPUT); \
420 /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
422 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
425 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
428 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
435 /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
438 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
442 if (!NILP (current_buffer->enable_multibyte_characters)) \
444 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
447 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
452 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
459 /* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */
461 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
464 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
467 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
474 /* Increase the buffer byte position POS_BYTE of the current buffer to
475 the next character boundary. No range checking of POS. */
477 #define INC_POS(pos_byte) \
479 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
480 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
484 /* Decrease the buffer byte position POS_BYTE of the current buffer to
485 the previous character boundary. No range checking of POS. */
487 #define DEC_POS(pos_byte) \
492 if (pos_byte < GPT_BYTE) \
493 p = BEG_ADDR + pos_byte - 1; \
495 p = BEG_ADDR + GAP_SIZE + pos_byte - 1; \
496 while (!CHAR_HEAD_P (*p)) \
503 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
505 #define INC_BOTH(charpos, bytepos) \
509 if (NILP (current_buffer->enable_multibyte_characters)) \
512 INC_POS ((bytepos)); \
517 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
519 #define DEC_BOTH(charpos, bytepos) \
523 if (NILP (current_buffer->enable_multibyte_characters)) \
526 DEC_POS ((bytepos)); \
531 /* Increase the buffer byte position POS_BYTE of the current buffer to
532 the next character boundary. This macro relies on the fact that
533 *GPT_ADDR and *Z_ADDR are always accessible and the values are
534 '\0'. No range checking of POS_BYTE. */
536 #define BUF_INC_POS(buf, pos_byte) \
538 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
539 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
543 /* Decrease the buffer byte position POS_BYTE of the current buffer to
544 the previous character boundary. No range checking of POS_BYTE. */
546 #define BUF_DEC_POS(buf, pos_byte) \
550 if (pos_byte < BUF_GPT_BYTE (buf)) \
551 p = BUF_BEG_ADDR (buf) + pos_byte - 1; \
553 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - 1; \
554 while (!CHAR_HEAD_P (*p)) \
562 /* If C is a character to be unified with a Unicode character, return
563 the unified Unicode character. */
565 #define MAYBE_UNIFY_CHAR(c) \
566 if (c > MAX_UNICODE_CHAR \
567 && CHAR_TABLE_P (Vchar_unify_table)) \
572 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
577 Funify_charset (val, Qnil, Qnil); \
578 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
580 if ((unified = XINT (val)) >= 0) \
587 /* Return the width of ASCII character C. The width is measured by
588 how many columns occupied on the screen when displayed in the
591 #define ASCII_CHAR_WIDTH(c) \
594 ? XFASTINT (current_buffer->tab_width) \
595 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
598 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
600 /* Return the width of character C. The width is measured by how many
601 columns occupied on the screen when displayed in the current
604 #define CHAR_WIDTH(c) \
606 ? ASCII_CHAR_WIDTH (c) \
607 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
609 extern int char_resolve_modifier_mask
P_ ((int));
610 extern int char_string
P_ ((int, unsigned char *));
611 extern int string_char
P_ ((const unsigned char *,
612 const unsigned char **, int *));
614 extern int translate_char
P_ ((Lisp_Object
, int c
));
615 extern int char_printable_p
P_ ((int c
));
616 extern void parse_str_as_multibyte
P_ ((const unsigned char *, int, int *,
618 extern int parse_str_to_multibyte
P_ ((unsigned char *, int));
619 extern int str_as_multibyte
P_ ((unsigned char *, int, int, int *));
620 extern int str_to_multibyte
P_ ((unsigned char *, int, int));
621 extern int str_as_unibyte
P_ ((unsigned char *, int));
622 extern int strwidth
P_ ((unsigned char *, int));
623 extern int c_string_width
P_ ((const unsigned char *, int, int, int *, int *));
624 extern int lisp_string_width
P_ ((Lisp_Object
, int, int *, int *));
626 extern Lisp_Object Vprintable_chars
;
628 extern Lisp_Object Qcharacterp
, Qauto_fill_chars
;
629 extern Lisp_Object Vtranslation_table_vector
;
630 extern Lisp_Object Vchar_width_table
;
631 extern Lisp_Object Vchar_direction_table
;
632 extern Lisp_Object Vchar_unify_table
;
634 extern Lisp_Object string_escape_byte8
P_ ((Lisp_Object
));
636 /* Return a translation table of id number ID. */
637 #define GET_TRANSLATION_TABLE(id) \
638 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
640 /* A char-table for characters which may invoke auto-filling. */
641 extern Lisp_Object Vauto_fill_chars
;
643 extern Lisp_Object Vchar_script_table
;
645 /* Copy LEN bytes from FROM to TO. This macro should be used only
646 when a caller knows that LEN is short and the obvious copy loop is
647 faster than calling bcopy which has some overhead. Copying a
648 multibyte sequence of a character is the typical case. */
650 #define BCOPY_SHORT(from, to, len) \
653 unsigned char *from_p = from, *to_p = to; \
654 while (i--) *to_p++ = *from_p++; \
657 #define DEFSYM(sym, name) \
658 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
660 #endif /* EMACS_CHARACTER_H */
662 /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac
663 (do not change this comment) */