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 /* If C is not ASCII, make it unibyte. */
81 #define MAKE_CHAR_UNIBYTE(c) \
83 if (! ASCII_CHAR_P (c)) \
84 c = CHAR_TO_BYTE8 (c); \
88 /* If C is not ASCII, make it multibyte. It assumes C < 256. */
89 #define MAKE_CHAR_MULTIBYTE(c) ((c) = unibyte_to_multibyte_table[(c)])
91 /* This is the maximum byte length of multibyte form. */
92 #define MAX_MULTIBYTE_LENGTH 5
94 /* Return a Lisp character whose character code is C. It assumes C is
95 a valid character code. */
96 #define make_char(c) make_number (c)
98 /* Nonzero iff C is an ASCII byte. */
99 #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
101 /* Nonzero iff X is a character. */
102 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
104 /* Nonzero iff C is valid as a character code. GENERICP is not used
106 #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
108 /* Check if Lisp object X is a character or not. */
109 #define CHECK_CHARACTER(x) \
111 if (! CHARACTERP(x)) x = wrong_type_argument (Qcharacterp, (x)); \
114 #define CHECK_CHARACTER_CAR(x) \
116 Lisp_Object tmp = XCAR (x); \
117 CHECK_CHARACTER (tmp); \
118 XSETCAR ((x), tmp); \
121 #define CHECK_CHARACTER_CDR(x) \
123 Lisp_Object tmp = XCDR (x); \
124 CHECK_CHARACTER (tmp); \
125 XSETCDR ((x), tmp); \
128 /* Nonzero iff C is an ASCII character. */
129 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
131 /* Nonzero iff C is a character of code less than 0x100. */
132 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
134 /* Nonzero if character C has a printable glyph. */
135 #define CHAR_PRINTABLE_P(c) \
136 (((c) >= 32 && ((c) < 127) \
137 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
139 /* Return byte length of multibyte form for character C. */
140 #define CHAR_BYTES(c) \
141 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
142 : (c) <= MAX_2_BYTE_CHAR ? 2 \
143 : (c) <= MAX_3_BYTE_CHAR ? 3 \
144 : (c) <= MAX_4_BYTE_CHAR ? 4 \
145 : (c) <= MAX_5_BYTE_CHAR ? 5 \
149 /* Return the leading code of multibyte form of C. */
150 #define CHAR_LEADING_CODE(c) \
151 ((c) <= MAX_1_BYTE_CHAR ? c \
152 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
153 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
154 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
155 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
156 : (0xC0 | (((c) >> 6) & 0x01)))
159 /* Store multibyte form of the character C in P. The caller should
160 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
161 Returns the length of the multibyte form. */
163 #define CHAR_STRING(c, p) \
164 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
167 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
168 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
169 (p)[1] = (0x80 | ((c) & 0x3F)), \
171 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
172 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
173 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
174 (p)[2] = (0x80 | ((c) & 0x3F)), \
176 : char_string (c, p))
178 /* Store multibyte form of byte B in P. The caller should allocate at
179 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
180 length of the multibyte form. */
182 #define BYTE8_STRING(b, p) \
183 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
184 (p)[1] = (0x80 | ((c) & 0x3F)), \
188 /* Store multibyte form of the character C in P. The caller should
189 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
190 And, advance P to the end of the multibyte form. */
192 #define CHAR_STRING_ADVANCE(c, p) \
194 if ((c) <= MAX_1_BYTE_CHAR) \
196 else if ((c) <= MAX_2_BYTE_CHAR) \
197 *(p)++ = (0xC0 | ((c) >> 6)), \
198 *(p)++ = (0x80 | ((c) & 0x3F)); \
199 else if ((c) <= MAX_3_BYTE_CHAR) \
200 *(p)++ = (0xE0 | ((c) >> 12)), \
201 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
202 *(p)++ = (0x80 | ((c) & 0x3F)); \
204 (p) += char_string ((c), (p)); \
208 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
210 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
212 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
214 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
216 /* Nonzero iff BYTE starts a character in a multibyte form.
217 This is equivalent to:
218 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
219 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
221 /* Just kept for backward compatibility. This macro will be removed
223 #define BASE_LEADING_CODE_P LEADING_CODE_P
225 /* How many bytes a character that starts with BYTE occupies in a
227 #define BYTES_BY_CHAR_HEAD(byte) \
228 (!((byte) & 0x80) ? 1 \
229 : !((byte) & 0x20) ? 2 \
230 : !((byte) & 0x10) ? 3 \
231 : !((byte) & 0x08) ? 4 \
235 /* Return the length of the multi-byte form at string STR of length
236 LEN while assuming that STR points a valid multi-byte form. As
237 this macro isn't necessary anymore, all callers will be changed to
238 use BYTES_BY_CHAR_HEAD directly in the future. */
240 #define MULTIBYTE_FORM_LENGTH(str, len) \
241 BYTES_BY_CHAR_HEAD (*(str))
243 /* Parse multibyte string STR of length LENGTH and set BYTES to the
244 byte length of a character at STR while assuming that STR points a
245 valid multibyte form. As this macro isn't necessary anymore, all
246 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
249 #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
250 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
252 /* The byte length of multibyte form at unibyte string P ending at
253 PEND. If STR doesn't point a valid multibyte form, return 0. */
255 #define MULTIBYTE_LENGTH(p, pend) \
257 : !((p)[0] & 0x80) ? 1 \
258 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
259 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
260 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
261 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
262 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
263 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
264 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
265 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
269 /* Like MULTIBYTE_LENGTH but don't check the ending address. */
271 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
272 (!((p)[0] & 0x80) ? 1 \
273 : ((p)[1] & 0xC0) != 0x80 ? 0 \
274 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
275 : ((p)[2] & 0xC0) != 0x80 ? 0 \
276 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
277 : ((p)[3] & 0xC0) != 0x80 ? 0 \
278 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
279 : ((p)[4] & 0xC0) != 0x80 ? 0 \
280 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
283 /* If P is before LIMIT, advance P to the next character boundary. It
284 assumes that P is already at a character boundary of the sane
285 mulitbyte form whose end address is LIMIT. */
287 #define NEXT_CHAR_BOUNDARY(p, limit) \
290 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
294 /* If P is after LIMIT, advance P to the previous character boundary.
295 It assumes that P is already at a character boundary of the sane
296 mulitbyte form whose beginning address is LIMIT. */
298 #define PREV_CHAR_BOUNDARY(p, limit) \
302 const unsigned char *p0 = (p); \
305 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
306 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
310 /* Return the character code of character whose multibyte form is at
311 P. The argument LEN is ignored. It will be removed in the
314 #define STRING_CHAR(p, len) \
317 : ! ((p)[0] & 0x20) \
318 ? (((((p)[0] & 0x1F) << 6) \
320 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
321 : ! ((p)[0] & 0x10) \
322 ? ((((p)[0] & 0x0F) << 12) \
323 | (((p)[1] & 0x3F) << 6) \
325 : string_char ((p), NULL, NULL))
328 /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte
329 form. The argument LEN is ignored. It will be removed in the
332 #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
334 ? ((actual_len) = 1, (p)[0]) \
335 : ! ((p)[0] & 0x20) \
336 ? ((actual_len) = 2, \
337 (((((p)[0] & 0x1F) << 6) \
339 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
340 : ! ((p)[0] & 0x10) \
341 ? ((actual_len) = 3, \
342 ((((p)[0] & 0x0F) << 12) \
343 | (((p)[1] & 0x3F) << 6) \
344 | ((p)[2] & 0x3F))) \
345 : string_char ((p), NULL, &actual_len))
348 /* Like STRING_CHAR but advance P to the end of multibyte form. */
350 #define STRING_CHAR_ADVANCE(p) \
353 : ! ((p)[0] & 0x20) \
355 ((((p)[-2] & 0x1F) << 6) \
357 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
358 : ! ((p)[0] & 0x10) \
360 ((((p)[-3] & 0x0F) << 12) \
361 | (((p)[-2] & 0x3F) << 6) \
362 | ((p)[-1] & 0x3F))) \
363 : string_char ((p), &(p), NULL))
366 /* Fetch the "next" character from Lisp string STRING at byte position
367 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
369 All the args must be side-effect-free.
370 BYTEIDX and CHARIDX must be lvalues;
371 we increment them past the character fetched. */
373 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
377 if (STRING_MULTIBYTE (STRING)) \
379 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
382 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
386 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
390 /* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve
391 if STRING is unibyte. */
393 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
397 if (STRING_MULTIBYTE (STRING)) \
399 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
402 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
407 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
408 MAKE_CHAR_MULTIBYTE (OUTPUT); \
414 /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
416 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
419 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
422 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
429 /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
432 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
436 if (!NILP (current_buffer->enable_multibyte_characters)) \
438 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
441 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
446 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
453 /* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */
455 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
458 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
461 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
468 /* Increase the buffer byte position POS_BYTE of the current buffer to
469 the next character boundary. No range checking of POS. */
471 #define INC_POS(pos_byte) \
473 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
474 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
478 /* Decrease the buffer byte position POS_BYTE of the current buffer to
479 the previous character boundary. No range checking of POS. */
481 #define DEC_POS(pos_byte) \
486 if (pos_byte < GPT_BYTE) \
487 p = BEG_ADDR + pos_byte - 1; \
489 p = BEG_ADDR + GAP_SIZE + pos_byte - 1; \
490 while (!CHAR_HEAD_P (*p)) \
497 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
499 #define INC_BOTH(charpos, bytepos) \
503 if (NILP (current_buffer->enable_multibyte_characters)) \
506 INC_POS ((bytepos)); \
511 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
513 #define DEC_BOTH(charpos, bytepos) \
517 if (NILP (current_buffer->enable_multibyte_characters)) \
520 DEC_POS ((bytepos)); \
525 /* Increase the buffer byte position POS_BYTE of the current buffer to
526 the next character boundary. This macro relies on the fact that
527 *GPT_ADDR and *Z_ADDR are always accessible and the values are
528 '\0'. No range checking of POS_BYTE. */
530 #define BUF_INC_POS(buf, pos_byte) \
532 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
533 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
537 /* Decrease the buffer byte position POS_BYTE of the current buffer to
538 the previous character boundary. No range checking of POS_BYTE. */
540 #define BUF_DEC_POS(buf, pos_byte) \
544 if (pos_byte < BUF_GPT_BYTE (buf)) \
545 p = BUF_BEG_ADDR (buf) + pos_byte - 1; \
547 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - 1; \
548 while (!CHAR_HEAD_P (*p)) \
556 /* If C is a character to be unified with a Unicode character, return
557 the unified Unicode character. */
559 #define MAYBE_UNIFY_CHAR(c) \
560 if (c > MAX_UNICODE_CHAR \
561 && CHAR_TABLE_P (Vchar_unify_table)) \
566 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
571 Funify_charset (val, Qnil, Qnil); \
572 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
574 if ((unified = XINT (val)) >= 0) \
581 /* Return the width of ASCII character C. The width is measured by
582 how many columns occupied on the screen when displayed in the
585 #define ASCII_CHAR_WIDTH(c) \
588 ? XFASTINT (current_buffer->tab_width) \
589 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
592 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
594 /* Return the width of character C. The width is measured by how many
595 columns occupied on the screen when displayed in the current
598 #define CHAR_WIDTH(c) \
600 ? ASCII_CHAR_WIDTH (c) \
601 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
603 extern int char_resolve_modifier_mask
P_ ((int));
604 extern int char_string
P_ ((int, unsigned char *));
605 extern int string_char
P_ ((const unsigned char *,
606 const unsigned char **, int *));
608 extern int translate_char
P_ ((Lisp_Object
, int c
));
609 extern int char_printable_p
P_ ((int c
));
610 extern void parse_str_as_multibyte
P_ ((const unsigned char *, int, int *,
612 extern int parse_str_to_multibyte
P_ ((unsigned char *, int));
613 extern int str_as_multibyte
P_ ((unsigned char *, int, int, int *));
614 extern int str_to_multibyte
P_ ((unsigned char *, int, int));
615 extern int str_as_unibyte
P_ ((unsigned char *, int));
616 extern int strwidth
P_ ((unsigned char *, int));
617 extern int c_string_width
P_ ((const unsigned char *, int, int, int *, int *));
618 extern int lisp_string_width
P_ ((Lisp_Object
, int, int *, int *));
620 extern Lisp_Object Vprintable_chars
;
622 extern Lisp_Object Qcharacterp
, Qauto_fill_chars
;
623 extern Lisp_Object Vtranslation_table_vector
;
624 extern Lisp_Object Vchar_width_table
;
625 extern Lisp_Object Vchar_direction_table
;
626 extern Lisp_Object Vchar_unify_table
;
628 extern Lisp_Object string_escape_byte8
P_ ((Lisp_Object
));
630 /* Return a translation table of id number ID. */
631 #define GET_TRANSLATION_TABLE(id) \
632 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
634 /* A char-table for characters which may invoke auto-filling. */
635 extern Lisp_Object Vauto_fill_chars
;
637 extern Lisp_Object Vchar_script_table
;
639 /* Copy LEN bytes from FROM to TO. This macro should be used only
640 when a caller knows that LEN is short and the obvious copy loop is
641 faster than calling bcopy which has some overhead. Copying a
642 multibyte sequence of a character is the typical case. */
644 #define BCOPY_SHORT(from, to, len) \
647 unsigned char *from_p = from, *to_p = to; \
648 while (i--) *to_p++ = *from_p++; \
651 #define DEFSYM(sym, name) \
652 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
654 #endif /* EMACS_CHARACTER_H */
656 /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac
657 (do not change this comment) */