* process.c (socket_options): Make it const.
[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.
76b6f707 4 Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009
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 64
4c0354d7
SM
65#define UNIBYTE_TO_CHAR(byte) \
66 (ASCII_BYTE_P (byte) ? (byte) : BYTE8_TO_CHAR (byte))
67
885317d8 68/* Return the raw 8-bit byte for character C. */
8bc28f69
KH
69#define CHAR_TO_BYTE8(c) \
70 (CHAR_BYTE8_P (c) \
71 ? (c) - 0x3FFF00 \
72 : multibyte_char_to_unibyte (c, Qnil))
885317d8 73
2afc21f5
SM
74/* Return the raw 8-bit byte for character C,
75 or -1 if C doesn't correspond to a byte. */
76#define CHAR_TO_BYTE_SAFE(c) \
77 (CHAR_BYTE8_P (c) \
78 ? (c) - 0x3FFF00 \
455af463 79 : multibyte_char_to_unibyte_safe (c))
2afc21f5 80
885317d8
KH
81/* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
82 that corresponds to a raw 8-bit byte. */
0168c3d8
KH
83#define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
84
43c47483
KH
85/* If C is not ASCII, make it unibyte. */
86#define MAKE_CHAR_UNIBYTE(c) \
87 do { \
88 if (! ASCII_CHAR_P (c)) \
89 c = CHAR_TO_BYTE8 (c); \
90 } while (0)
3e411074 91
3e411074 92
18a2979d 93/* If C is not ASCII, make it multibyte. Assumes C < 256. */
c0dc8f64 94#define MAKE_CHAR_MULTIBYTE(c) \
4c0354d7 95 (eassert ((c) >= 0 && (c) < 256), (c) = UNIBYTE_TO_CHAR (c))
3e411074 96
885317d8 97/* This is the maximum byte length of multibyte form. */
0168c3d8
KH
98#define MAX_MULTIBYTE_LENGTH 5
99
18a2979d 100/* Return a Lisp character whose character code is C. Assumes C is
b583cead 101 a valid character code. */
0168c3d8
KH
102#define make_char(c) make_number (c)
103
104/* Nonzero iff C is an ASCII byte. */
105#define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80)
106
107/* Nonzero iff X is a character. */
108#define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
109
18a2979d 110/* Nonzero iff C is valid as a character code. GENERICP is not used. */
885317d8 111#define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR)
0168c3d8
KH
112
113/* Check if Lisp object X is a character or not. */
63db3c1b
MB
114#define CHECK_CHARACTER(x) \
115 CHECK_TYPE (CHARACTERP (x), Qcharacterp, x)
0168c3d8 116
8f924df7
KH
117#define CHECK_CHARACTER_CAR(x) \
118 do { \
119 Lisp_Object tmp = XCAR (x); \
120 CHECK_CHARACTER (tmp); \
121 XSETCAR ((x), tmp); \
122 } while (0)
123
124#define CHECK_CHARACTER_CDR(x) \
125 do { \
126 Lisp_Object tmp = XCDR (x); \
127 CHECK_CHARACTER (tmp); \
128 XSETCDR ((x), tmp); \
129 } while (0)
130
0168c3d8
KH
131/* Nonzero iff C is an ASCII character. */
132#define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80)
133
134/* Nonzero iff C is a character of code less than 0x100. */
135#define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100)
136
137/* Nonzero if character C has a printable glyph. */
138#define CHAR_PRINTABLE_P(c) \
139 (((c) >= 32 && ((c) < 127) \
140 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c)))))
141
885317d8 142/* Return byte length of multibyte form for character C. */
0168c3d8
KH
143#define CHAR_BYTES(c) \
144 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
145 : (c) <= MAX_2_BYTE_CHAR ? 2 \
146 : (c) <= MAX_3_BYTE_CHAR ? 3 \
147 : (c) <= MAX_4_BYTE_CHAR ? 4 \
148 : (c) <= MAX_5_BYTE_CHAR ? 5 \
149 : 2)
150
43c47483
KH
151
152/* Return the leading code of multibyte form of C. */
153#define CHAR_LEADING_CODE(c) \
154 ((c) <= MAX_1_BYTE_CHAR ? c \
155 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
156 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
157 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
158 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
159 : (0xC0 | (((c) >> 6) & 0x01)))
160
161
885317d8
KH
162/* Store multibyte form of the character C in P. The caller should
163 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
164 Returns the length of the multibyte form. */
0168c3d8
KH
165
166#define CHAR_STRING(c, p) \
167 ((unsigned) (c) <= MAX_1_BYTE_CHAR \
168 ? ((p)[0] = (c), \
169 1) \
170 : (unsigned) (c) <= MAX_2_BYTE_CHAR \
171 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
172 (p)[1] = (0x80 | ((c) & 0x3F)), \
173 2) \
174 : (unsigned) (c) <= MAX_3_BYTE_CHAR \
175 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
176 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
177 (p)[2] = (0x80 | ((c) & 0x3F)), \
178 3) \
f958a2fa 179 : char_string ((unsigned) c, p))
0168c3d8 180
eb41da4c
KH
181/* Store multibyte form of byte B in P. The caller should allocate at
182 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
183 length of the multibyte form. */
1106ea2b
KH
184
185#define BYTE8_STRING(b, p) \
186 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
7f464917 187 (p)[1] = (0x80 | ((b) & 0x3F)), \
1106ea2b
KH
188 2)
189
0168c3d8 190
18a2979d
EZ
191/* Store multibyte form of the character C in P and advance P to the
192 end of the multibyte form. The caller should allocate at least
193 MAX_MULTIBYTE_LENGTH bytes area at P in advance. */
0168c3d8 194
eb41da4c
KH
195#define CHAR_STRING_ADVANCE(c, p) \
196 do { \
197 if ((c) <= MAX_1_BYTE_CHAR) \
198 *(p)++ = (c); \
199 else if ((c) <= MAX_2_BYTE_CHAR) \
200 *(p)++ = (0xC0 | ((c) >> 6)), \
201 *(p)++ = (0x80 | ((c) & 0x3F)); \
202 else if ((c) <= MAX_3_BYTE_CHAR) \
203 *(p)++ = (0xE0 | ((c) >> 12)), \
204 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
205 *(p)++ = (0x80 | ((c) & 0x3F)); \
206 else \
207 (p) += char_string ((c), (p)); \
885317d8 208 } while (0)
0168c3d8 209
eb41da4c 210
0168c3d8
KH
211/* Nonzero iff BYTE starts a non-ASCII character in a multibyte
212 form. */
213#define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
214
b5c7dbe6
KH
215/* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
216 multibyte form. */
217#define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
218
885317d8
KH
219/* Nonzero iff BYTE starts a character in a multibyte form.
220 This is equivalent to:
221 (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */
222#define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
223
18a2979d
EZ
224/* Kept for backward compatibility. This macro will be removed in the
225 future. */
0168c3d8
KH
226#define BASE_LEADING_CODE_P LEADING_CODE_P
227
228/* How many bytes a character that starts with BYTE occupies in a
229 multibyte form. */
230#define BYTES_BY_CHAR_HEAD(byte) \
231 (!((byte) & 0x80) ? 1 \
232 : !((byte) & 0x20) ? 2 \
233 : !((byte) & 0x10) ? 3 \
234 : !((byte) & 0x08) ? 4 \
235 : 5)
236
237
238/* Return the length of the multi-byte form at string STR of length
239 LEN while assuming that STR points a valid multi-byte form. As
240 this macro isn't necessary anymore, all callers will be changed to
241 use BYTES_BY_CHAR_HEAD directly in the future. */
242
243#define MULTIBYTE_FORM_LENGTH(str, len) \
244 BYTES_BY_CHAR_HEAD (*(str))
245
246/* Parse multibyte string STR of length LENGTH and set BYTES to the
247 byte length of a character at STR while assuming that STR points a
248 valid multibyte form. As this macro isn't necessary anymore, all
249 callers will be changed to use BYTES_BY_CHAR_HEAD directly in the
250 future. */
251
252#define PARSE_MULTIBYTE_SEQ(str, length, bytes) \
253 (bytes) = BYTES_BY_CHAR_HEAD (*(str))
254
255/* The byte length of multibyte form at unibyte string P ending at
18a2979d 256 PEND. If STR doesn't point to a valid multibyte form, return 0. */
0168c3d8
KH
257
258#define MULTIBYTE_LENGTH(p, pend) \
259 (p >= pend ? 0 \
260 : !((p)[0] & 0x80) ? 1 \
261 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
262 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
263 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
264 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
265 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
266 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
267 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
268 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
269 : 0)
270
271
18a2979d 272/* Like MULTIBYTE_LENGTH, but don't check the ending address. */
0168c3d8
KH
273
274#define MULTIBYTE_LENGTH_NO_CHECK(p) \
275 (!((p)[0] & 0x80) ? 1 \
276 : ((p)[1] & 0xC0) != 0x80 ? 0 \
277 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
278 : ((p)[2] & 0xC0) != 0x80 ? 0 \
279 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
280 : ((p)[3] & 0xC0) != 0x80 ? 0 \
281 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
282 : ((p)[4] & 0xC0) != 0x80 ? 0 \
283 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
284 : 0)
285
18a2979d
EZ
286/* If P is before LIMIT, advance P to the next character boundary.
287 Assumes that P is already at a character boundary of the same
8f924df7
KH
288 mulitbyte form whose end address is LIMIT. */
289
290#define NEXT_CHAR_BOUNDARY(p, limit) \
291 do { \
292 if ((p) < (limit)) \
293 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
294 } while (0)
295
296
297/* If P is after LIMIT, advance P to the previous character boundary.
18a2979d 298 Assumes that P is already at a character boundary of the same
8f924df7
KH
299 mulitbyte form whose beginning address is LIMIT. */
300
301#define PREV_CHAR_BOUNDARY(p, limit) \
302 do { \
303 if ((p) > (limit)) \
304 { \
305 const unsigned char *p0 = (p); \
306 do { \
307 p0--; \
308 } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \
309 (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \
310 } \
311 } while (0)
0168c3d8
KH
312
313/* Return the character code of character whose multibyte form is at
314 P. The argument LEN is ignored. It will be removed in the
315 future. */
316
317#define STRING_CHAR(p, len) \
318 (!((p)[0] & 0x80) \
319 ? (p)[0] \
320 : ! ((p)[0] & 0x20) \
321 ? (((((p)[0] & 0x1F) << 6) \
322 | ((p)[1] & 0x3F)) \
323 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
324 : ! ((p)[0] & 0x10) \
325 ? ((((p)[0] & 0x0F) << 12) \
326 | (((p)[1] & 0x3F) << 6) \
327 | ((p)[2] & 0x3F)) \
eb41da4c 328 : string_char ((p), NULL, NULL))
0168c3d8
KH
329
330
18a2979d 331/* Like STRING_CHAR, but set ACTUAL_LEN to the length of multibyte
0168c3d8
KH
332 form. The argument LEN is ignored. It will be removed in the
333 future. */
334
335#define STRING_CHAR_AND_LENGTH(p, len, actual_len) \
336 (!((p)[0] & 0x80) \
337 ? ((actual_len) = 1, (p)[0]) \
338 : ! ((p)[0] & 0x20) \
339 ? ((actual_len) = 2, \
340 (((((p)[0] & 0x1F) << 6) \
341 | ((p)[1] & 0x3F)) \
342 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \
343 : ! ((p)[0] & 0x10) \
344 ? ((actual_len) = 3, \
345 ((((p)[0] & 0x0F) << 12) \
346 | (((p)[1] & 0x3F) << 6) \
347 | ((p)[2] & 0x3F))) \
eb41da4c 348 : string_char ((p), NULL, &actual_len))
0168c3d8
KH
349
350
18a2979d 351/* Like STRING_CHAR, but advance P to the end of multibyte form. */
0168c3d8
KH
352
353#define STRING_CHAR_ADVANCE(p) \
354 (!((p)[0] & 0x80) \
355 ? *(p)++ \
356 : ! ((p)[0] & 0x20) \
357 ? ((p) += 2, \
358 ((((p)[-2] & 0x1F) << 6) \
359 | ((p)[-1] & 0x3F) \
8f924df7 360 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
0168c3d8
KH
361 : ! ((p)[0] & 0x10) \
362 ? ((p) += 3, \
363 ((((p)[-3] & 0x0F) << 12) \
364 | (((p)[-2] & 0x3F) << 6) \
365 | ((p)[-1] & 0x3F))) \
eb41da4c 366 : string_char ((p), &(p), NULL))
0168c3d8
KH
367
368
369/* Fetch the "next" character from Lisp string STRING at byte position
370 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
371
372 All the args must be side-effect-free.
373 BYTEIDX and CHARIDX must be lvalues;
374 we increment them past the character fetched. */
375
376#define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
a1c2ac9a 377 do \
0168c3d8
KH
378 { \
379 CHARIDX++; \
380 if (STRING_MULTIBYTE (STRING)) \
381 { \
f1c99628 382 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
0168c3d8
KH
383 int len; \
384 \
385 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
386 BYTEIDX += len; \
387 } \
388 else \
f1c99628
SM
389 { \
390 OUTPUT = SREF (STRING, BYTEIDX); \
391 BYTEIDX++; \
392 } \
0168c3d8 393 } \
a1c2ac9a 394 while (0)
0168c3d8 395
18a2979d
EZ
396/* Like FETCH_STRING_CHAR_ADVANCE, but return a multibyte character
397 even if STRING is unibyte. */
43c47483
KH
398
399#define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
a1c2ac9a 400 do \
43c47483
KH
401 { \
402 CHARIDX++; \
403 if (STRING_MULTIBYTE (STRING)) \
404 { \
f1c99628 405 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
43c47483
KH
406 int len; \
407 \
408 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
409 BYTEIDX += len; \
410 } \
411 else \
412 { \
f1c99628
SM
413 OUTPUT = SREF (STRING, BYTEIDX); \
414 BYTEIDX++; \
43c47483
KH
415 MAKE_CHAR_MULTIBYTE (OUTPUT); \
416 } \
417 } \
a1c2ac9a 418 while (0)
43c47483 419
0168c3d8 420
18a2979d 421/* Like FETCH_STRING_CHAR_ADVANCE, but assumes STRING is multibyte. */
0168c3d8
KH
422
423#define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
a1c2ac9a 424 do \
0168c3d8 425 { \
f1c99628 426 unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \
0168c3d8
KH
427 int len; \
428 \
429 OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \
430 BYTEIDX += len; \
431 CHARIDX++; \
432 } \
a1c2ac9a 433 while (0)
0168c3d8
KH
434
435
18a2979d 436/* Like FETCH_STRING_CHAR_ADVANCE, but fetch character from the current
0168c3d8
KH
437 buffer. */
438
439#define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
a1c2ac9a 440 do \
0168c3d8
KH
441 { \
442 CHARIDX++; \
443 if (!NILP (current_buffer->enable_multibyte_characters)) \
444 { \
445 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
446 int len; \
447 \
448 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
449 BYTEIDX += len; \
450 } \
451 else \
452 { \
453 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
454 BYTEIDX++; \
455 } \
456 } \
a1c2ac9a 457 while (0)
0168c3d8
KH
458
459
18a2979d 460/* Like FETCH_CHAR_ADVANCE, but assumes the current buffer is multibyte. */
0168c3d8
KH
461
462#define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
a1c2ac9a 463 do \
0168c3d8
KH
464 { \
465 unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \
466 int len; \
467 \
468 OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \
469 BYTEIDX += len; \
470 CHARIDX++; \
471 } \
a1c2ac9a 472 while (0)
0168c3d8
KH
473
474
18a2979d 475/* Increment the buffer byte position POS_BYTE of the current buffer to
0168c3d8
KH
476 the next character boundary. No range checking of POS. */
477
478#define INC_POS(pos_byte) \
479 do { \
480 unsigned char *p = BYTE_POS_ADDR (pos_byte); \
481 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
482 } while (0)
483
484
18a2979d 485/* Decrement the buffer byte position POS_BYTE of the current buffer to
0168c3d8
KH
486 the previous character boundary. No range checking of POS. */
487
488#define DEC_POS(pos_byte) \
489 do { \
490 unsigned char *p; \
491 \
492 pos_byte--; \
493 if (pos_byte < GPT_BYTE) \
f1c99628 494 p = BEG_ADDR + pos_byte - BEG_BYTE; \
0168c3d8 495 else \
f1c99628 496 p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\
0168c3d8
KH
497 while (!CHAR_HEAD_P (*p)) \
498 { \
499 p--; \
500 pos_byte--; \
501 } \
502 } while (0)
503
504/* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
505
506#define INC_BOTH(charpos, bytepos) \
507 do \
508 { \
509 (charpos)++; \
510 if (NILP (current_buffer->enable_multibyte_characters)) \
511 (bytepos)++; \
512 else \
513 INC_POS ((bytepos)); \
514 } \
515 while (0)
516
517
518/* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
519
520#define DEC_BOTH(charpos, bytepos) \
521 do \
522 { \
523 (charpos)--; \
524 if (NILP (current_buffer->enable_multibyte_characters)) \
525 (bytepos)--; \
526 else \
527 DEC_POS ((bytepos)); \
528 } \
529 while (0)
530
531
18a2979d 532/* Increment the buffer byte position POS_BYTE of the current buffer to
0168c3d8
KH
533 the next character boundary. This macro relies on the fact that
534 *GPT_ADDR and *Z_ADDR are always accessible and the values are
535 '\0'. No range checking of POS_BYTE. */
536
537#define BUF_INC_POS(buf, pos_byte) \
538 do { \
539 unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \
540 pos_byte += BYTES_BY_CHAR_HEAD (*p); \
541 } while (0)
542
543
18a2979d 544/* Decrement the buffer byte position POS_BYTE of the current buffer to
0168c3d8
KH
545 the previous character boundary. No range checking of POS_BYTE. */
546
547#define BUF_DEC_POS(buf, pos_byte) \
548 do { \
549 unsigned char *p; \
550 pos_byte--; \
551 if (pos_byte < BUF_GPT_BYTE (buf)) \
f1c99628 552 p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \
0168c3d8 553 else \
f1c99628 554 p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\
0168c3d8
KH
555 while (!CHAR_HEAD_P (*p)) \
556 { \
557 p--; \
558 pos_byte--; \
559 } \
560 } while (0)
561
562
b583cead
KH
563/* If C is a character to be unified with a Unicode character, return
564 the unified Unicode character. */
565
55870ffc
KH
566#define MAYBE_UNIFY_CHAR(c) \
567 do { \
568 if (c > MAX_UNICODE_CHAR && c <= MAX_5_BYTE_CHAR) \
569 { \
570 Lisp_Object val; \
571 val = CHAR_TABLE_REF (Vchar_unify_table, c); \
572 if (INTEGERP (val)) \
573 c = XINT (val); \
574 else if (! NILP (val)) \
575 c = maybe_unify_char (c, val); \
576 } \
577 } while (0)
0168c3d8 578
fc9d9d2a 579
0168c3d8 580/* Return the width of ASCII character C. The width is measured by
18a2979d 581 how many columns C will occupy on the screen when displayed in the
0168c3d8
KH
582 current buffer. */
583
584#define ASCII_CHAR_WIDTH(c) \
585 (c < 0x20 \
586 ? (c == '\t' \
587 ? XFASTINT (current_buffer->tab_width) \
588 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
589 : (c < 0x7f \
590 ? 1 \
591 : ((NILP (current_buffer->ctl_arrow) ? 4 : 2))))
592
593/* Return the width of character C. The width is measured by how many
18a2979d 594 columns C will occupy on the screen when displayed in the current
0168c3d8
KH
595 buffer. */
596
597#define CHAR_WIDTH(c) \
598 (ASCII_CHAR_P (c) \
599 ? ASCII_CHAR_WIDTH (c) \
600 : XINT (CHAR_TABLE_REF (Vchar_width_table, c)))
601
c19cab20
KH
602/* If C is a variation selector, return the index numnber of the
603 variation selector (1..256). Otherwise, return 0. */
604
605#define CHAR_VARIATION_SELECTOR_P(c) \
606 ((c) < 0xFE00 ? 0 \
607 : (c) <= 0xFE0F ? (c) - 0xFE00 + 1 \
608 : (c) < 0xE0100 ? 0 \
609 : (c) <= 0xE01EF ? (c) - 0xE0100 + 17 \
610 : 0)
611
612/* If C is a high surrogate, return 1. If C is a low surrogate,
613 return 0. Otherwise, return 0. */
614
615#define CHAR_SURROGATE_PAIR_P(c) \
616 ((c) < 0xD800 ? 0 \
617 : (c) <= 0xDBFF ? 1 \
618 : (c) <= 0xDFFF ? 2 \
619 : 0)
620
621
eb41da4c 622extern int char_resolve_modifier_mask P_ ((int));
f958a2fa 623extern int char_string P_ ((unsigned, unsigned char *));
eb41da4c
KH
624extern int string_char P_ ((const unsigned char *,
625 const unsigned char **, int *));
0168c3d8
KH
626
627extern int translate_char P_ ((Lisp_Object, int c));
628extern int char_printable_p P_ ((int c));
8f924df7
KH
629extern void parse_str_as_multibyte P_ ((const unsigned char *, int, int *,
630 int *));
0168c3d8
KH
631extern int parse_str_to_multibyte P_ ((unsigned char *, int));
632extern int str_as_multibyte P_ ((unsigned char *, int, int, int *));
633extern int str_to_multibyte P_ ((unsigned char *, int, int));
634extern int str_as_unibyte P_ ((unsigned char *, int));
0460afe5
KH
635extern EMACS_INT str_to_unibyte P_ ((const unsigned char *, unsigned char *,
636 EMACS_INT, int));
0168c3d8 637extern int strwidth P_ ((unsigned char *, int));
8f924df7 638extern int c_string_width P_ ((const unsigned char *, int, int, int *, int *));
0168c3d8
KH
639extern int lisp_string_width P_ ((Lisp_Object, int, int *, int *));
640
641extern Lisp_Object Vprintable_chars;
642
643extern Lisp_Object Qcharacterp, Qauto_fill_chars;
644extern Lisp_Object Vtranslation_table_vector;
645extern Lisp_Object Vchar_width_table;
646extern Lisp_Object Vchar_direction_table;
647extern Lisp_Object Vchar_unify_table;
a48a6418 648extern Lisp_Object Vunicode_category_table;
0168c3d8 649
fac2bdc4
DL
650extern Lisp_Object string_escape_byte8 P_ ((Lisp_Object));
651
0168c3d8
KH
652/* Return a translation table of id number ID. */
653#define GET_TRANSLATION_TABLE(id) \
654 (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)]))
655
656/* A char-table for characters which may invoke auto-filling. */
657extern Lisp_Object Vauto_fill_chars;
658
e18ef64a 659extern Lisp_Object Vchar_script_table;
e0d6e5a5 660extern Lisp_Object Vscript_representative_chars;
b5c7dbe6 661
0168c3d8
KH
662/* Copy LEN bytes from FROM to TO. This macro should be used only
663 when a caller knows that LEN is short and the obvious copy loop is
664 faster than calling bcopy which has some overhead. Copying a
665 multibyte sequence of a character is the typical case. */
666
667#define BCOPY_SHORT(from, to, len) \
668 do { \
669 int i = len; \
670 unsigned char *from_p = from, *to_p = to; \
671 while (i--) *to_p++ = *from_p++; \
672 } while (0)
673
674#define DEFSYM(sym, name) \
5e2327cf 675 do { (sym) = intern_c_string ((name)); staticpro (&(sym)); } while (0)
0168c3d8
KH
676
677#endif /* EMACS_CHARACTER_H */
fbaf0946
MB
678
679/* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac
680 (do not change this comment) */