* keymap.c (Vminibuffer_local_filename_must_match_map):
[bpt/emacs.git] / src / character.h
CommitLineData
0168c3d8
KH
1/* Header for multibyte character handler.
2 Copyright (C) 1995, 1997, 1998 Electrotechnical Laboratory, JAPAN.
8f924df7 3 Licensed to the Free Software Foundation.
ec62e0ac 4 Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008
0168c3d8
KH
5 National Institute of Advanced Industrial Science and Technology (AIST)
6 Registration Number H13PRO009
7
8This file is part of GNU Emacs.
9
b9b1cc14 10GNU Emacs is free software: you can redistribute it and/or modify
0168c3d8 11it under the terms of the GNU General Public License as published by
b9b1cc14
GM
12the Free Software Foundation, either version 3 of the License, or
13(at your option) any later version.
0168c3d8
KH
14
15GNU Emacs is distributed in the hope that it will be useful,
16but WITHOUT ANY WARRANTY; without even the implied warranty of
17MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18GNU General Public License for more details.
19
20You should have received a copy of the GNU General Public License
b9b1cc14 21along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
0168c3d8
KH
22
23#ifndef EMACS_CHARACTER_H
24#define EMACS_CHARACTER_H
25
885317d8
KH
26/* character code 1st byte byte sequence
27 -------------- -------- -------------
28 0-7F 00..7F 0xxxxxxx
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
c43e85a9
KH
33 3FFF80-3FFFFF C0..C1 1100000x 10xxxxxx (for eight-bit-char)
34 400000-... invalid
0168c3d8 35
c43e85a9
KH
36 invalid 1st byte 80..BF 10xxxxxx
37 F9..FF 11111xxx (xxx != 000)
0168c3d8
KH
38*/
39
885317d8 40/* Maximum character code ((1 << CHARACTERBITS) - 1). */
0168c3d8
KH
41#define MAX_CHAR 0x3FFFFF
42
885317d8 43/* Maximum Unicode character code. */
0168c3d8
KH
44#define MAX_UNICODE_CHAR 0x10FFFF
45
885317d8 46/* Maximum N-byte character codes. */
0168c3d8
KH
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
52
3a0a38de
KH
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
57
8bc28f69
KH
58/* Nonzero iff C is a character that corresponds to a raw 8-bit
59 byte. */
60#define CHAR_BYTE8_P(c) ((c) > MAX_5_BYTE_CHAR)
61
885317d8 62/* Return the character code for raw 8-bit byte BYTE. */
0168c3d8 63#define BYTE8_TO_CHAR(byte) ((byte) + 0x3FFF00)
885317d8
KH
64
65/* Return the raw 8-bit byte for character C. */
8bc28f69
KH
66#define CHAR_TO_BYTE8(c) \
67 (CHAR_BYTE8_P (c) \
68 ? (c) - 0x3FFF00 \
69 : multibyte_char_to_unibyte (c, Qnil))
885317d8
KH
70
71/* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
72 that corresponds to a raw 8-bit byte. */
0168c3d8
KH
73#define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
74
43c47483
KH
75/* Mapping table from unibyte chars to multibyte chars. */
76extern int unibyte_to_multibyte_table[256];
3e411074 77
43c47483
KH
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))
3e411074 82
1acbd067
KH
83/* Nth element is 1 iff unibyte char N can be mapped to a multibyte
84 char. */
85extern char unibyte_has_multibyte_table[256];
86
87#define UNIBYTE_CHAR_HAS_MULTIBYTE_P(c) (unibyte_has_multibyte_table[(c)])
88
43c47483
KH
89/* If C is not ASCII, make it unibyte. */
90#define MAKE_CHAR_UNIBYTE(c) \
91 do { \
92 if (! ASCII_CHAR_P (c)) \
93 c = CHAR_TO_BYTE8 (c); \
94 } while (0)
3e411074 95
3e411074 96
43c47483
KH
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)])
3e411074 99
885317d8 100/* This is the maximum byte length of multibyte form. */
0168c3d8
KH
101#define MAX_MULTIBYTE_LENGTH 5
102
b583cead
KH
103/* Return a Lisp character whose character code is C. It assumes C is
104 a valid character code. */
0168c3d8
KH
105#define make_char(c) make_number (c)
106
107/* Nonzero iff C is an ASCII byte. */
108#define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
109
110/* Nonzero iff X is a character. */
111#define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
112
f65c6d94 113/* Nonzero iff C is valid as a character code. GENERICP is not used
885317d8
KH
114 now. */
115#define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
0168c3d8
KH
116
117/* Check if Lisp object X is a character or not. */
63db3c1b
MB
118#define CHECK_CHARACTER(x) \
119 CHECK_TYPE (CHARACTERP (x), Qcharacterp, x)
0168c3d8 120
8f924df7
KH
121#define CHECK_CHARACTER_CAR(x) \
122 do { \
123 Lisp_Object tmp = XCAR (x); \
124 CHECK_CHARACTER (tmp); \
125 XSETCAR ((x), tmp); \
126 } while (0)
127
128#define CHECK_CHARACTER_CDR(x) \
129 do { \
130 Lisp_Object tmp = XCDR (x); \
131 CHECK_CHARACTER (tmp); \
132 XSETCDR ((x), tmp); \
133 } while (0)
134
0168c3d8
KH
135/* Nonzero iff C is an ASCII character. */
136#define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
137
138/* Nonzero iff C is a character of code less than 0x100. */
139#define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
140
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)))))
145
885317d8 146/* Return byte length of multibyte form for character C. */
0168c3d8
KH
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 \
153 : 2)
154
43c47483
KH
155
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)))
164
165
885317d8
KH
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. */
0168c3d8
KH
169
170#define CHAR_STRING(c, p) \
171 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
172 ? ((p)[0] = (c), \
173 1) \
174 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
175 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
176 (p)[1] = (0x80 | ((c) & 0x3F)), \
177 2) \
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)), \
182 3) \
f958a2fa 183 : char_string ((unsigned) c, p))
0168c3d8 184
eb41da4c
KH
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. */
1106ea2b
KH
188
189#define BYTE8_STRING(b, p) \
190 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
7f464917 191 (p)[1] = (0x80 | ((b) & 0x3F)), \
1106ea2b
KH
192 2)
193
0168c3d8 194
885317d8
KH
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. */
0168c3d8 198
eb41da4c
KH
199#define CHAR_STRING_ADVANCE(c, p) \
200 do { \
201 if ((c) <= MAX_1_BYTE_CHAR) \
202 *(p)++ = (c); \
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)); \
210 else \
211 (p) += char_string ((c), (p)); \
885317d8 212 } while (0)
0168c3d8 213
eb41da4c 214
0168c3d8
KH
215/* Nonzero iff BYTE starts a non-ASCII character in a multibyte
216 form. */
217#define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
218
b5c7dbe6
KH
219/* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
220 multibyte form. */
221#define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
222
885317d8
KH
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)
227
0168c3d8
KH
228/* Just kept for backward compatibility. This macro will be removed
229 in the future. */
230#define BASE_LEADING_CODE_P LEADING_CODE_P
231
232/* How many bytes a character that starts with BYTE occupies in a
233 multibyte form. */
234#define BYTES_BY_CHAR_HEAD(byte) \
235 (!((byte) & 0x80) ? 1 \
236 : !((byte) & 0x20) ? 2 \
237 : !((byte) & 0x10) ? 3 \
238 : !((byte) & 0x08) ? 4 \
239 : 5)
240
241
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. */
246
247#define MULTIBYTE_FORM_LENGTH(str, len) \
248 BYTES_BY_CHAR_HEAD (*(str))
249
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
254 future. */
255
256#define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
257 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
258
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. */
261
262#define MULTIBYTE_LENGTH(p, pend) \
263 (p >= pend ? 0 \
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 \
273 : 0)
274
275
276/* Like MULTIBYTE_LENGTH but don't check the ending address. */
277
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 \
288 : 0)
289
8f924df7
KH
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. */
293
294#define NEXT_CHAR_BOUNDARY(p, limit) \
295 do { \
296 if ((p) < (limit)) \
297 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
298 } while (0)
299
300
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. */
304
305#define PREV_CHAR_BOUNDARY(p, limit) \
306 do { \
307 if ((p) > (limit)) \
308 { \
309 const unsigned char *p0 = (p); \
310 do { \
311 p0--; \
312 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
313 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
314 } \
315 } while (0)
0168c3d8
KH
316
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
319 future. */
320
321#define STRING_CHAR(p, len) \
322 (!((p)[0] & 0x80) \
323 ? (p)[0] \
324 : ! ((p)[0] & 0x20) \
325 ? (((((p)[0] & 0x1F) << 6) \
326 | ((p)[1] & 0x3F)) \
327 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
328 : ! ((p)[0] & 0x10) \
329 ? ((((p)[0] & 0x0F) << 12) \
330 | (((p)[1] & 0x3F) << 6) \
331 | ((p)[2] & 0x3F)) \
eb41da4c 332 : string_char ((p), NULL, NULL))
0168c3d8
KH
333
334
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
337 future. */
338
339#define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
340 (!((p)[0] & 0x80) \
341 ? ((actual_len) = 1, (p)[0]) \
342 : ! ((p)[0] & 0x20) \
343 ? ((actual_len) = 2, \
344 (((((p)[0] & 0x1F) << 6) \
345 | ((p)[1] & 0x3F)) \
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))) \
eb41da4c 352 : string_char ((p), NULL, &actual_len))
0168c3d8
KH
353
354
b583cead 355/* Like STRING_CHAR but advance P to the end of multibyte form. */
0168c3d8
KH
356
357#define STRING_CHAR_ADVANCE(p) \
358 (!((p)[0] & 0x80) \
359 ? *(p)++ \
360 : ! ((p)[0] & 0x20) \
361 ? ((p) += 2, \
362 ((((p)[-2] & 0x1F) << 6) \
363 | ((p)[-1] & 0x3F) \
8f924df7 364 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
0168c3d8
KH
365 : ! ((p)[0] & 0x10) \
366 ? ((p) += 3, \
367 ((((p)[-3] & 0x0F) << 12) \
368 | (((p)[-2] & 0x3F) << 6) \
369 | ((p)[-1] & 0x3F))) \
eb41da4c 370 : string_char ((p), &(p), NULL))
0168c3d8
KH
371
372
373/* Fetch the "next" character from Lisp string STRING at byte position
374 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
375
376 All the args must be side-effect-free.
377 BYTEIDX and CHARIDX must be lvalues;
378 we increment them past the character fetched. */
379
380#define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
a1c2ac9a 381 do \
0168c3d8
KH
382 { \
383 CHARIDX++; \
384 if (STRING_MULTIBYTE (STRING)) \
385 { \
f1c99628 386 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
0168c3d8
KH
387 int len; \
388 \
389 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
390 BYTEIDX += len; \
391 } \
392 else \
f1c99628
SM
393 { \
394 OUTPUT = SREF (STRING, BYTEIDX); \
395 BYTEIDX++; \
396 } \
0168c3d8 397 } \
a1c2ac9a 398 while (0)
0168c3d8 399
b583cead
KH
400/* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve
401 if STRING is unibyte. */
43c47483
KH
402
403#define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
a1c2ac9a 404 do \
43c47483
KH
405 { \
406 CHARIDX++; \
407 if (STRING_MULTIBYTE (STRING)) \
408 { \
f1c99628 409 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
43c47483
KH
410 int len; \
411 \
412 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
413 BYTEIDX += len; \
414 } \
415 else \
416 { \
f1c99628
SM
417 OUTPUT = SREF (STRING, BYTEIDX); \
418 BYTEIDX++; \
43c47483
KH
419 MAKE_CHAR_MULTIBYTE (OUTPUT); \
420 } \
421 } \
a1c2ac9a 422 while (0)
43c47483 423
0168c3d8
KH
424
425/* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */
426
427#define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
a1c2ac9a 428 do \
0168c3d8 429 { \
f1c99628 430 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
0168c3d8
KH
431 int len; \
432 \
433 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
434 BYTEIDX += len; \
435 CHARIDX++; \
436 } \
a1c2ac9a 437 while (0)
0168c3d8
KH
438
439
440/* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current
441 buffer. */
442
443#define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
a1c2ac9a 444 do \
0168c3d8
KH
445 { \
446 CHARIDX++; \
447 if (!NILP (current_buffer->enable_multibyte_characters)) \
448 { \
449 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
450 int len; \
451 \
452 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
453 BYTEIDX += len; \
454 } \
455 else \
456 { \
457 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
458 BYTEIDX++; \
459 } \
460 } \
a1c2ac9a 461 while (0)
0168c3d8
KH
462
463
b583cead 464/* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */
0168c3d8
KH
465
466#define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
a1c2ac9a 467 do \
0168c3d8
KH
468 { \
469 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
470 int len; \
471 \
472 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
473 BYTEIDX += len; \
474 CHARIDX++; \
475 } \
a1c2ac9a 476 while (0)
0168c3d8
KH
477
478
479/* Increase the buffer byte position POS_BYTE of the current buffer to
480 the next character boundary. No range checking of POS. */
481
482#define INC_POS(pos_byte) \
483 do { \
484 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
485 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
486 } while (0)
487
488
489/* Decrease the buffer byte position POS_BYTE of the current buffer to
490 the previous character boundary. No range checking of POS. */
491
492#define DEC_POS(pos_byte) \
493 do { \
494 unsigned char *p; \
495 \
496 pos_byte--; \
497 if (pos_byte < GPT_BYTE) \
f1c99628 498 p = BEG_ADDR + pos_byte - BEG_BYTE; \
0168c3d8 499 else \
f1c99628 500 p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\
0168c3d8
KH
501 while (!CHAR_HEAD_P (*p)) \
502 { \
503 p--; \
504 pos_byte--; \
505 } \
506 } while (0)
507
508/* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
509
510#define INC_BOTH(charpos, bytepos) \
511 do \
512 { \
513 (charpos)++; \
514 if (NILP (current_buffer->enable_multibyte_characters)) \
515 (bytepos)++; \
516 else \
517 INC_POS ((bytepos)); \
518 } \
519 while (0)
520
521
522/* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
523
524#define DEC_BOTH(charpos, bytepos) \
525 do \
526 { \
527 (charpos)--; \
528 if (NILP (current_buffer->enable_multibyte_characters)) \
529 (bytepos)--; \
530 else \
531 DEC_POS ((bytepos)); \
532 } \
533 while (0)
534
535
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. */
540
541#define BUF_INC_POS(buf, pos_byte) \
542 do { \
543 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
544 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
545 } while (0)
546
547
548/* Decrease the buffer byte position POS_BYTE of the current buffer to
549 the previous character boundary. No range checking of POS_BYTE. */
550
551#define BUF_DEC_POS(buf, pos_byte) \
552 do { \
553 unsigned char *p; \
554 pos_byte--; \
555 if (pos_byte < BUF_GPT_BYTE (buf)) \
f1c99628 556 p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \
0168c3d8 557 else \
f1c99628 558 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\
0168c3d8
KH
559 while (!CHAR_HEAD_P (*p)) \
560 { \
561 p--; \
562 pos_byte--; \
563 } \
564 } while (0)
565
566
b583cead
KH
567/* If C is a character to be unified with a Unicode character, return
568 the unified Unicode character. */
569
fc9d9d2a 570#define MAYBE_UNIFY_CHAR(c) \
eb41da4c
KH
571 if (c > MAX_UNICODE_CHAR \
572 && CHAR_TABLE_P (Vchar_unify_table)) \
fc9d9d2a
KH
573 { \
574 Lisp_Object val; \
575 int unified; \
576 \
577 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
578 if (! NILP (val)) \
579 { \
580 if (SYMBOLP (val)) \
581 { \
b5c7dbe6 582 Funify_charset (val, Qnil, Qnil); \
fc9d9d2a
KH
583 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
584 } \
585 if ((unified = XINT (val)) >= 0) \
586 c = unified; \
587 } \
588 } \
0168c3d8
KH
589 else
590
fc9d9d2a 591
0168c3d8
KH
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
594 current buffer. */
595
596#define ASCII_CHAR_WIDTH(c) \
597 (c < 0x20 \
598 ? (c == '\t' \
599 ? XFASTINT (current_buffer->tab_width) \
600 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
601 : (c < 0x7f \
602 ? 1 \
603 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
604
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
607 buffer. */
608
609#define CHAR_WIDTH(c) \
610 (ASCII_CHAR_P (c) \
611 ? ASCII_CHAR_WIDTH (c) \
612 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
613
eb41da4c 614extern int char_resolve_modifier_mask P_ ((int));
f958a2fa 615extern int char_string P_ ((unsigned, unsigned char *));
eb41da4c
KH
616extern int string_char P_ ((const unsigned char *,
617 const unsigned char **, int *));
0168c3d8
KH
618
619extern int translate_char P_ ((Lisp_Object, int c));
620extern int char_printable_p P_ ((int c));
8f924df7
KH
621extern void parse_str_as_multibyte P_ ((const unsigned char *, int, int *,
622 int *));
0168c3d8
KH
623extern int parse_str_to_multibyte P_ ((unsigned char *, int));
624extern int str_as_multibyte P_ ((unsigned char *, int, int, int *));
625extern int str_to_multibyte P_ ((unsigned char *, int, int));
626extern int str_as_unibyte P_ ((unsigned char *, int));
627extern int strwidth P_ ((unsigned char *, int));
8f924df7 628extern int c_string_width P_ ((const unsigned char *, int, int, int *, int *));
0168c3d8
KH
629extern int lisp_string_width P_ ((Lisp_Object, int, int *, int *));
630
631extern Lisp_Object Vprintable_chars;
632
633extern Lisp_Object Qcharacterp, Qauto_fill_chars;
634extern Lisp_Object Vtranslation_table_vector;
635extern Lisp_Object Vchar_width_table;
636extern Lisp_Object Vchar_direction_table;
637extern Lisp_Object Vchar_unify_table;
638
fac2bdc4
DL
639extern Lisp_Object string_escape_byte8 P_ ((Lisp_Object));
640
0168c3d8
KH
641/* Return a translation table of id number ID. */
642#define GET_TRANSLATION_TABLE(id) \
643 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
644
645/* A char-table for characters which may invoke auto-filling. */
646extern Lisp_Object Vauto_fill_chars;
647
e18ef64a 648extern Lisp_Object Vchar_script_table;
e0d6e5a5 649extern Lisp_Object Vscript_representative_chars;
b5c7dbe6 650
0168c3d8
KH
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. */
655
656#define BCOPY_SHORT(from, to, len) \
657 do { \
658 int i = len; \
659 unsigned char *from_p = from, *to_p = to; \
660 while (i--) *to_p++ = *from_p++; \
661 } while (0)
662
663#define DEFSYM(sym, name) \
664 do { (sym) = intern ((name)); staticpro (&(sym)); } while (0)
665
666#endif /* EMACS_CHARACTER_H */
fbaf0946
MB
667
668/* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac
669 (do not change this comment) */