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) 2001, 2002
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 /* Nonzero iff C is an ASCII character. */
117 #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
119 /* Nonzero iff C is a character of code less than 0x100. */
120 #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
122 /* Nonzero if character C has a printable glyph. */
123 #define CHAR_PRINTABLE_P(c) \
124 (((c) >= 32 && ((c) < 127) \
125 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
127 /* Return byte length of multibyte form for character C. */
128 #define CHAR_BYTES(c) \
129 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
130 : (c) <= MAX_2_BYTE_CHAR ? 2 \
131 : (c) <= MAX_3_BYTE_CHAR ? 3 \
132 : (c) <= MAX_4_BYTE_CHAR ? 4 \
133 : (c) <= MAX_5_BYTE_CHAR ? 5 \
137 /* Return the leading code of multibyte form of C. */
138 #define CHAR_LEADING_CODE(c) \
139 ((c) <= MAX_1_BYTE_CHAR ? c \
140 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
141 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
142 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
143 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
144 : (0xC0 | (((c) >> 6) & 0x01)))
147 /* Store multibyte form of the character C in P. The caller should
148 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
149 Returns the length of the multibyte form. */
151 #define CHAR_STRING(c, p) \
152 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
155 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
156 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
157 (p)[1] = (0x80 | ((c) & 0x3F)), \
159 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
160 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
161 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
162 (p)[2] = (0x80 | ((c) & 0x3F)), \
164 : char_string (c, p))
166 /* Store multibyte form of byte B in P. The caller should allocate at
167 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
168 length of the multibyte form. */
170 #define BYTE8_STRING(b, p) \
171 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
172 (p)[1] = (0x80 | ((c) & 0x3F)), \
176 /* Store multibyte form of the character C in P. The caller should
177 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
178 And, advance P to the end of the multibyte form. */
180 #define CHAR_STRING_ADVANCE(c, p) \
182 if ((c) <= MAX_1_BYTE_CHAR) \
184 else if ((c) <= MAX_2_BYTE_CHAR) \
185 *(p)++ = (0xC0 | ((c) >> 6)), \
186 *(p)++ = (0x80 | ((c) & 0x3F)); \
187 else if ((c) <= MAX_3_BYTE_CHAR) \
188 *(p)++ = (0xE0 | ((c) >> 12)), \
189 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
190 *(p)++ = (0x80 | ((c) & 0x3F)); \
192 (p) += char_string ((c), (p)); \
196 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
198 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
200 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
202 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
204 /* Nonzero iff BYTE starts a character in a multibyte form.
205 This is equivalent to:
206 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
207 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
209 /* Just kept for backward compatibility. This macro will be removed
211 #define BASE_LEADING_CODE_P LEADING_CODE_P
213 /* How many bytes a character that starts with BYTE occupies in a
215 #define BYTES_BY_CHAR_HEAD(byte) \
216 (!((byte) & 0x80) ? 1 \
217 : !((byte) & 0x20) ? 2 \
218 : !((byte) & 0x10) ? 3 \
219 : !((byte) & 0x08) ? 4 \
223 /* Return the length of the multi-byte form at string STR of length
224 LEN while assuming that STR points a valid multi-byte form. As
225 this macro isn't necessary anymore, all callers will be changed to
226 use BYTES_BY_CHAR_HEAD directly in the future. */
228 #define MULTIBYTE_FORM_LENGTH(str, len) \
229 BYTES_BY_CHAR_HEAD (*(str))
231 /* Parse multibyte string STR of length LENGTH and set BYTES to the
232 byte length of a character at STR while assuming that STR points a
233 valid multibyte form. As this macro isn't necessary anymore, all
234 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
237 #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
238 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
240 /* The byte length of multibyte form at unibyte string P ending at
241 PEND. If STR doesn't point a valid multibyte form, return 0. */
243 #define MULTIBYTE_LENGTH(p, pend) \
245 : !((p)[0] & 0x80) ? 1 \
246 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
247 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
248 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
249 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
250 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
251 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
252 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
253 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
257 /* Like MULTIBYTE_LENGTH but don't check the ending address. */
259 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
260 (!((p)[0] & 0x80) ? 1 \
261 : ((p)[1] & 0xC0) != 0x80 ? 0 \
262 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
263 : ((p)[2] & 0xC0) != 0x80 ? 0 \
264 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
265 : ((p)[3] & 0xC0) != 0x80 ? 0 \
266 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
267 : ((p)[4] & 0xC0) != 0x80 ? 0 \
268 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
272 /* Return the character code of character whose multibyte form is at
273 P. The argument LEN is ignored. It will be removed in the
276 #define STRING_CHAR(p, len) \
279 : ! ((p)[0] & 0x20) \
280 ? (((((p)[0] & 0x1F) << 6) \
282 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
283 : ! ((p)[0] & 0x10) \
284 ? ((((p)[0] & 0x0F) << 12) \
285 | (((p)[1] & 0x3F) << 6) \
287 : string_char ((p), NULL, NULL))
290 /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte
291 form. The argument LEN is ignored. It will be removed in the
294 #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
296 ? ((actual_len) = 1, (p)[0]) \
297 : ! ((p)[0] & 0x20) \
298 ? ((actual_len) = 2, \
299 (((((p)[0] & 0x1F) << 6) \
301 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
302 : ! ((p)[0] & 0x10) \
303 ? ((actual_len) = 3, \
304 ((((p)[0] & 0x0F) << 12) \
305 | (((p)[1] & 0x3F) << 6) \
306 | ((p)[2] & 0x3F))) \
307 : string_char ((p), NULL, &actual_len))
310 /* Like STRING_CHAR but advacen P to the end of multibyte form. */
312 #define STRING_CHAR_ADVANCE(p) \
315 : ! ((p)[0] & 0x20) \
317 ((((p)[-2] & 0x1F) << 6) \
319 | (((unsigned char) (p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
320 : ! ((p)[0] & 0x10) \
322 ((((p)[-3] & 0x0F) << 12) \
323 | (((p)[-2] & 0x3F) << 6) \
324 | ((p)[-1] & 0x3F))) \
325 : string_char ((p), &(p), NULL))
328 /* Fetch the "next" character from Lisp string STRING at byte position
329 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
331 All the args must be side-effect-free.
332 BYTEIDX and CHARIDX must be lvalues;
333 we increment them past the character fetched. */
335 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
339 if (STRING_MULTIBYTE (STRING)) \
341 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
344 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
348 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
352 /* Like FETCH_STRING_CHAR_ADVANCE */
354 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
358 if (STRING_MULTIBYTE (STRING)) \
360 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
363 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
368 OUTPUT = XSTRING (STRING)->data[BYTEIDX++]; \
369 MAKE_CHAR_MULTIBYTE (OUTPUT); \
375 /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
377 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
380 unsigned char *ptr = &XSTRING (STRING)->data[BYTEIDX]; \
383 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
390 /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
393 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
397 if (!NILP (current_buffer->enable_multibyte_characters)) \
399 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
402 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
407 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
414 /* Like FETCH_CHAR_ADVANCE but assumes STRING is multibyte. */
416 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
419 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
422 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
429 /* Increase the buffer byte position POS_BYTE of the current buffer to
430 the next character boundary. No range checking of POS. */
432 #define INC_POS(pos_byte) \
434 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
435 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
439 /* Decrease the buffer byte position POS_BYTE of the current buffer to
440 the previous character boundary. No range checking of POS. */
442 #define DEC_POS(pos_byte) \
447 if (pos_byte < GPT_BYTE) \
448 p = BEG_ADDR + pos_byte - 1; \
450 p = BEG_ADDR + GAP_SIZE + pos_byte - 1; \
451 while (!CHAR_HEAD_P (*p)) \
458 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
460 #define INC_BOTH(charpos, bytepos) \
464 if (NILP (current_buffer->enable_multibyte_characters)) \
467 INC_POS ((bytepos)); \
472 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
474 #define DEC_BOTH(charpos, bytepos) \
478 if (NILP (current_buffer->enable_multibyte_characters)) \
481 DEC_POS ((bytepos)); \
486 /* Increase the buffer byte position POS_BYTE of the current buffer to
487 the next character boundary. This macro relies on the fact that
488 *GPT_ADDR and *Z_ADDR are always accessible and the values are
489 '\0'. No range checking of POS_BYTE. */
491 #define BUF_INC_POS(buf, pos_byte) \
493 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
494 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
498 /* Decrease the buffer byte position POS_BYTE of the current buffer to
499 the previous character boundary. No range checking of POS_BYTE. */
501 #define BUF_DEC_POS(buf, pos_byte) \
505 if (pos_byte < BUF_GPT_BYTE (buf)) \
506 p = BUF_BEG_ADDR (buf) + pos_byte - 1; \
508 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - 1; \
509 while (!CHAR_HEAD_P (*p)) \
517 #define MAYBE_UNIFY_CHAR(c) \
518 if (c > MAX_UNICODE_CHAR \
519 && CHAR_TABLE_P (Vchar_unify_table)) \
524 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
529 Funify_charset (val, Qnil, Qnil); \
530 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
532 if ((unified = XINT (val)) >= 0) \
539 /* Return the width of ASCII character C. The width is measured by
540 how many columns occupied on the screen when displayed in the
543 #define ASCII_CHAR_WIDTH(c) \
546 ? XFASTINT (current_buffer->tab_width) \
547 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
550 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
552 /* Return the width of character C. The width is measured by how many
553 columns occupied on the screen when displayed in the current
556 #define CHAR_WIDTH(c) \
558 ? ASCII_CHAR_WIDTH (c) \
559 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
561 extern int char_resolve_modifier_mask
P_ ((int));
562 extern int char_string
P_ ((int, unsigned char *));
563 extern int string_char
P_ ((const unsigned char *,
564 const unsigned char **, int *));
566 extern int translate_char
P_ ((Lisp_Object
, int c
));
567 extern int char_printable_p
P_ ((int c
));
568 extern void parse_str_as_multibyte
P_ ((unsigned char *, int, int *, int *));
569 extern int parse_str_to_multibyte
P_ ((unsigned char *, int));
570 extern int str_as_multibyte
P_ ((unsigned char *, int, int, int *));
571 extern int str_to_multibyte
P_ ((unsigned char *, int, int));
572 extern int str_as_unibyte
P_ ((unsigned char *, int));
573 extern int strwidth
P_ ((unsigned char *, int));
574 extern int c_string_width
P_ ((unsigned char *, int, int, int *, int *));
575 extern int lisp_string_width
P_ ((Lisp_Object
, int, int *, int *));
577 extern Lisp_Object Vprintable_chars
;
579 extern Lisp_Object Qcharacterp
, Qauto_fill_chars
;
580 extern Lisp_Object Vtranslation_table_vector
;
581 extern Lisp_Object Vchar_width_table
;
582 extern Lisp_Object Vchar_direction_table
;
583 extern Lisp_Object Vchar_unify_table
;
585 extern Lisp_Object string_escape_byte8
P_ ((Lisp_Object
));
587 /* Return a translation table of id number ID. */
588 #define GET_TRANSLATION_TABLE(id) \
589 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
591 /* A char-table for characters which may invoke auto-filling. */
592 extern Lisp_Object Vauto_fill_chars
;
594 extern Lisp_Object Vchar_script_table
;
596 /* Copy LEN bytes from FROM to TO. This macro should be used only
597 when a caller knows that LEN is short and the obvious copy loop is
598 faster than calling bcopy which has some overhead. Copying a
599 multibyte sequence of a character is the typical case. */
601 #define BCOPY_SHORT(from, to, len) \
604 unsigned char *from_p = from, *to_p = to; \
605 while (i--) *to_p++ = *from_p++; \
608 #define DEFSYM(sym, name) \
609 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
611 #endif /* EMACS_CHARACTER_H */