?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Add separator.pbm tool-bar image
[emacs.git] / src / character.h
blob67d4bca4a239ba29e48807ce8f40f2b5182b195b
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) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
5 National Institute of Advanced Industrial Science and Technology (AIST)
6 Registration Number H13PRO009
7
8 This file is part of GNU Emacs.
9
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 3 of the License, or
13 (at your option) any later version.
14
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.
19
20 You should have received a copy of the GNU General Public License
21 along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */
22
23 #ifndef EMACS_CHARACTER_H
24 #define EMACS_CHARACTER_H
25
26 #include <verify.h>
27
28 INLINE_HEADER_BEGIN
29
30 /* character code 1st byte byte sequence
31 -------------- -------- -------------
32 0-7F 00..7F 0xxxxxxx
33 80-7FF C2..DF 110xxxxx 10xxxxxx
34 800-FFFF E0..EF 1110xxxx 10xxxxxx 10xxxxxx
35 10000-1FFFFF F0..F7 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
36 200000-3FFF7F F8 11111000 1000xxxx 10xxxxxx 10xxxxxx 10xxxxxx
37 3FFF80-3FFFFF C0..C1 1100000x 10xxxxxx (for eight-bit-char)
38 400000-... invalid
39
40 invalid 1st byte 80..BF 10xxxxxx
41 F9..FF 11111xxx (xxx != 000)
42 */
43
44 /* Maximum character code ((1 << CHARACTERBITS) - 1). */
45 #define MAX_CHAR 0x3FFFFF
46
47 /* Maximum Unicode character code. */
48 #define MAX_UNICODE_CHAR 0x10FFFF
49
50 /* Maximum N-byte character codes. */
51 #define MAX_1_BYTE_CHAR 0x7F
52 #define MAX_2_BYTE_CHAR 0x7FF
53 #define MAX_3_BYTE_CHAR 0xFFFF
54 #define MAX_4_BYTE_CHAR 0x1FFFFF
55 #define MAX_5_BYTE_CHAR 0x3FFF7F
56
57 /* Minimum leading code of multibyte characters. */
58 #define MIN_MULTIBYTE_LEADING_CODE 0xC0
59 /* Maximum leading code of multibyte characters. */
60 #define MAX_MULTIBYTE_LEADING_CODE 0xF8
61
62 /* Unicode character values. */
63 enum
64 {
65 NO_BREAK_SPACE = 0x00A0,
66 SOFT_HYPHEN = 0x00AD,
67 ZERO_WIDTH_NON_JOINER = 0x200C,
68 ZERO_WIDTH_JOINER = 0x200D,
69 HYPHEN = 0x2010,
70 NON_BREAKING_HYPHEN = 0x2011,
71 LEFT_SINGLE_QUOTATION_MARK = 0x2018,
72 RIGHT_SINGLE_QUOTATION_MARK = 0x2019,
73 PARAGRAPH_SEPARATOR = 0x2029,
74 LEFT_POINTING_ANGLE_BRACKET = 0x2329,
75 RIGHT_POINTING_ANGLE_BRACKET = 0x232A,
76 LEFT_ANGLE_BRACKET = 0x3008,
77 RIGHT_ANGLE_BRACKET = 0x3009,
78 OBJECT_REPLACEMENT_CHARACTER = 0xFFFC,
79 };
80
81 /* UTF-8 encodings. Use \x escapes, so they are portable to pre-C11
82 compilers and can be concatenated with ordinary string literals. */
83 #define uLSQM "\xE2\x80\x98" /* U+2018 LEFT SINGLE QUOTATION MARK */
84 #define uRSQM "\xE2\x80\x99" /* U+2019 RIGHT SINGLE QUOTATION MARK */
85
86 /* Nonzero iff C is a character that corresponds to a raw 8-bit
87 byte. */
88 #define CHAR_BYTE8_P(c) ((c) > MAX_5_BYTE_CHAR)
89
90 /* Return the character code for raw 8-bit byte BYTE. */
91 #define BYTE8_TO_CHAR(byte) ((byte) + 0x3FFF00)
92
93 #define UNIBYTE_TO_CHAR(byte) \
94 (ASCII_CHAR_P (byte) ? (byte) : BYTE8_TO_CHAR (byte))
95
96 /* Return the raw 8-bit byte for character C. */
97 #define CHAR_TO_BYTE8(c) (CHAR_BYTE8_P (c) ? (c) - 0x3FFF00 : (c & 0xFF))
98
99 /* Return the raw 8-bit byte for character C,
100 or -1 if C doesn't correspond to a byte. */
101 #define CHAR_TO_BYTE_SAFE(c) \
102 (ASCII_CHAR_P (c) ? c : (CHAR_BYTE8_P (c) ? (c) - 0x3FFF00 : -1))
103
104 /* Nonzero iff BYTE is the 1st byte of a multibyte form of a character
105 that corresponds to a raw 8-bit byte. */
106 #define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1)
107
108 /* If C is not ASCII, make it unibyte. */
109 #define MAKE_CHAR_UNIBYTE(c) \
110 do { \
111 if (! ASCII_CHAR_P (c)) \
112 c = CHAR_TO_BYTE8 (c); \
113 } while (false)
114
115
116 /* If C is not ASCII, make it multibyte. Assumes C < 256. */
117 #define MAKE_CHAR_MULTIBYTE(c) \
118 (eassert ((c) >= 0 && (c) < 256), (c) = UNIBYTE_TO_CHAR (c))
119
120 /* This is the maximum byte length of multibyte form. */
121 #define MAX_MULTIBYTE_LENGTH 5
122
123 /* Nonzero iff X is a character. */
124 #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR)
125
126 /* Nonzero iff C is valid as a character code. */
127 #define CHAR_VALID_P(c) UNSIGNED_CMP (c, <=, MAX_CHAR)
128
129 /* Check if Lisp object X is a character or not. */
130 #define CHECK_CHARACTER(x) \
131 CHECK_TYPE (CHARACTERP (x), Qcharacterp, x)
132
133 #define CHECK_CHARACTER_CAR(x) \
134 do { \
135 Lisp_Object tmp = XCAR (x); \
136 CHECK_CHARACTER (tmp); \
137 XSETCAR ((x), tmp); \
138 } while (false)
139
140 #define CHECK_CHARACTER_CDR(x) \
141 do { \
142 Lisp_Object tmp = XCDR (x); \
143 CHECK_CHARACTER (tmp); \
144 XSETCDR ((x), tmp); \
145 } while (false)
146
147 /* Nonzero iff C is a character of code less than 0x100. */
148 #define SINGLE_BYTE_CHAR_P(c) UNSIGNED_CMP (c, <, 0x100)
149
150 /* Nonzero if character C has a printable glyph. */
151 #define CHAR_PRINTABLE_P(c) \
152 (((c) >= 32 && (c) < 127) \
153 || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c))))
154
155 /* Return byte length of multibyte form for character C. */
156 #define CHAR_BYTES(c) \
157 ( (c) <= MAX_1_BYTE_CHAR ? 1 \
158 : (c) <= MAX_2_BYTE_CHAR ? 2 \
159 : (c) <= MAX_3_BYTE_CHAR ? 3 \
160 : (c) <= MAX_4_BYTE_CHAR ? 4 \
161 : (c) <= MAX_5_BYTE_CHAR ? 5 \
162 : 2)
163
164
165 /* Return the leading code of multibyte form of C. */
166 #define CHAR_LEADING_CODE(c) \
167 ((c) <= MAX_1_BYTE_CHAR ? c \
168 : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \
169 : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \
170 : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \
171 : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \
172 : (0xC0 | (((c) >> 6) & 0x01)))
173
174
175 /* Store multibyte form of the character C in P. The caller should
176 allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance.
177 Returns the length of the multibyte form. */
178
179 #define CHAR_STRING(c, p) \
180 (UNSIGNED_CMP (c, <=, MAX_1_BYTE_CHAR) \
181 ? ((p)[0] = (c), \
182 1) \
183 : UNSIGNED_CMP (c, <=, MAX_2_BYTE_CHAR) \
184 ? ((p)[0] = (0xC0 | ((c) >> 6)), \
185 (p)[1] = (0x80 | ((c) & 0x3F)), \
186 2) \
187 : UNSIGNED_CMP (c, <=, MAX_3_BYTE_CHAR) \
188 ? ((p)[0] = (0xE0 | ((c) >> 12)), \
189 (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \
190 (p)[2] = (0x80 | ((c) & 0x3F)), \
191 3) \
192 : verify_expr (sizeof (c) <= sizeof (unsigned), char_string (c, p)))
193
194 /* Store multibyte form of byte B in P. The caller should allocate at
195 least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the
196 length of the multibyte form. */
197
198 #define BYTE8_STRING(b, p) \
199 ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \
200 (p)[1] = (0x80 | ((b) & 0x3F)), \
201 2)
202
203
204 /* Store multibyte form of the character C in P and advance P to the
205 end of the multibyte form. The caller should allocate at least
206 MAX_MULTIBYTE_LENGTH bytes area at P in advance. */
207
208 #define CHAR_STRING_ADVANCE(c, p) \
209 do { \
210 if ((c) <= MAX_1_BYTE_CHAR) \
211 *(p)++ = (c); \
212 else if ((c) <= MAX_2_BYTE_CHAR) \
213 *(p)++ = (0xC0 | ((c) >> 6)), \
214 *(p)++ = (0x80 | ((c) & 0x3F)); \
215 else if ((c) <= MAX_3_BYTE_CHAR) \
216 *(p)++ = (0xE0 | ((c) >> 12)), \
217 *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \
218 *(p)++ = (0x80 | ((c) & 0x3F)); \
219 else \
220 { \
221 verify (sizeof (c) <= sizeof (unsigned)); \
222 (p) += char_string (c, p); \
223 } \
224 } while (false)
225
226
227 /* Nonzero iff BYTE starts a non-ASCII character in a multibyte
228 form. */
229 #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0)
230
231 /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a
232 multibyte form. */
233 #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80)
234
235 /* Nonzero iff BYTE starts a character in a multibyte form.
236 This is equivalent to:
237 (ASCII_CHAR_P (byte) || LEADING_CODE_P (byte)) */
238 #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80)
239
240 /* How many bytes a character that starts with BYTE occupies in a
241 multibyte form. */
242 #define BYTES_BY_CHAR_HEAD(byte) \
243 (!((byte) & 0x80) ? 1 \
244 : !((byte) & 0x20) ? 2 \
245 : !((byte) & 0x10) ? 3 \
246 : !((byte) & 0x08) ? 4 \
247 : 5)
248
249
250 /* The byte length of multibyte form at unibyte string P ending at
251 PEND. If STR doesn't point to a valid multibyte form, return 0. */
252
253 #define MULTIBYTE_LENGTH(p, pend) \
254 (p >= pend ? 0 \
255 : !((p)[0] & 0x80) ? 1 \
256 : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \
257 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
258 : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \
259 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
260 : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \
261 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
262 : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \
263 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
264 : 0)
265
266
267 /* Like MULTIBYTE_LENGTH, but don't check the ending address. */
268
269 #define MULTIBYTE_LENGTH_NO_CHECK(p) \
270 (!((p)[0] & 0x80) ? 1 \
271 : ((p)[1] & 0xC0) != 0x80 ? 0 \
272 : ((p)[0] & 0xE0) == 0xC0 ? 2 \
273 : ((p)[2] & 0xC0) != 0x80 ? 0 \
274 : ((p)[0] & 0xF0) == 0xE0 ? 3 \
275 : ((p)[3] & 0xC0) != 0x80 ? 0 \
276 : ((p)[0] & 0xF8) == 0xF0 ? 4 \
277 : ((p)[4] & 0xC0) != 0x80 ? 0 \
278 : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \
279 : 0)
280
281 /* If P is before LIMIT, advance P to the next character boundary.
282 Assumes that P is already at a character boundary of the same
283 multibyte form whose end address is LIMIT. */
284
285 #define NEXT_CHAR_BOUNDARY(p, limit) \
286 do { \
287 if ((p) < (limit)) \
288 (p) += BYTES_BY_CHAR_HEAD (*(p)); \
289 } while (false)
290
291
292 /* If P is after LIMIT, advance P to the previous character boundary.
293 Assumes that P is already at a character boundary of the same
294 multibyte form whose beginning address is LIMIT. */
295
296 #define PREV_CHAR_BOUNDARY(p, limit) \
297 do { \
298 if ((p) > (limit)) \
299 { \
300 const unsigned char *chp = (p); \
301 do { \
302 chp--; \
303 } while (chp >= limit && ! CHAR_HEAD_P (*chp)); \
304 (p) = (BYTES_BY_CHAR_HEAD (*chp) == (p) - chp) ? chp : (p) - 1; \
305 } \
306 } while (false)
307
308 /* Return the character code of character whose multibyte form is at
309 P. Note that this macro unifies CJK characters whose codepoints
310 are in the Private Use Areas (PUAs), so it might return a different
311 codepoint from the one actually stored at P. */
312
313 #define STRING_CHAR(p) \
314 (!((p)[0] & 0x80) \
315 ? (p)[0] \
316 : ! ((p)[0] & 0x20) \
317 ? (((((p)[0] & 0x1F) << 6) \
318 | ((p)[1] & 0x3F)) \
319 + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \
320 : ! ((p)[0] & 0x10) \
321 ? ((((p)[0] & 0x0F) << 12) \
322 | (((p)[1] & 0x3F) << 6) \
323 | ((p)[2] & 0x3F)) \
324 : string_char ((p), NULL, NULL))
325
326
327 /* Like STRING_CHAR, but set ACTUAL_LEN to the length of multibyte
328 form.
329
330 Note: This macro returns the actual length of the character's
331 multibyte sequence as it is stored in a buffer or string. The
332 character it returns might have a different codepoint that has a
333 different multibyte sequence of a different length, due to possible
334 unification of CJK characters inside string_char. Therefore do NOT
335 assume that the length returned by this macro is identical to the
336 length of the multibyte sequence of the character it returns. */
337
338 #define STRING_CHAR_AND_LENGTH(p, actual_len) \
339 (!((p)[0] & 0x80) \
340 ? ((actual_len) = 1, (p)[0]) \
341 : ! ((p)[0] & 0x20) \
342 ? ((actual_len) = 2, \
343 (((((p)[0] & 0x1F) << 6) \
344 | ((p)[1] & 0x3F)) \
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))
352
353
354 /* Like STRING_CHAR, but advance P to the end of multibyte form. */
355
356 #define STRING_CHAR_ADVANCE(p) \
357 (!((p)[0] & 0x80) \
358 ? *(p)++ \
359 : ! ((p)[0] & 0x20) \
360 ? ((p) += 2, \
361 ((((p)[-2] & 0x1F) << 6) \
362 | ((p)[-1] & 0x3F) \
363 | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \
364 : ! ((p)[0] & 0x10) \
365 ? ((p) += 3, \
366 ((((p)[-3] & 0x0F) << 12) \
367 | (((p)[-2] & 0x3F) << 6) \
368 | ((p)[-1] & 0x3F))) \
369 : string_char ((p), &(p), NULL))
370
371
372 /* Fetch the "next" character from Lisp string STRING at byte position
373 BYTEIDX, character position CHARIDX. Store it into OUTPUT.
374
375 All the args must be side-effect-free.
376 BYTEIDX and CHARIDX must be lvalues;
377 we increment them past the character fetched. */
378
379 #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
380 do \
381 { \
382 CHARIDX++; \
383 if (STRING_MULTIBYTE (STRING)) \
384 { \
385 unsigned char *chp = &SDATA (STRING)[BYTEIDX]; \
386 int chlen; \
387 \
388 OUTPUT = STRING_CHAR_AND_LENGTH (chp, chlen); \
389 BYTEIDX += chlen; \
390 } \
391 else \
392 { \
393 OUTPUT = SREF (STRING, BYTEIDX); \
394 BYTEIDX++; \
395 } \
396 } \
397 while (false)
398
399 /* Like FETCH_STRING_CHAR_ADVANCE, but return a multibyte character
400 even if STRING is unibyte. */
401
402 #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \
403 do \
404 { \
405 CHARIDX++; \
406 if (STRING_MULTIBYTE (STRING)) \
407 { \
408 unsigned char *chp = &SDATA (STRING)[BYTEIDX]; \
409 int chlen; \
410 \
411 OUTPUT = STRING_CHAR_AND_LENGTH (chp, chlen); \
412 BYTEIDX += chlen; \
413 } \
414 else \
415 { \
416 OUTPUT = SREF (STRING, BYTEIDX); \
417 BYTEIDX++; \
418 MAKE_CHAR_MULTIBYTE (OUTPUT); \
419 } \
420 } \
421 while (false)
422
423
424 /* Like FETCH_STRING_CHAR_ADVANCE, but assumes STRING is multibyte. */
425
426 #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \
427 do \
428 { \
429 unsigned char *fetch_ptr = &SDATA (STRING)[BYTEIDX]; \
430 int fetch_len; \
431 \
432 OUTPUT = STRING_CHAR_AND_LENGTH (fetch_ptr, fetch_len); \
433 BYTEIDX += fetch_len; \
434 CHARIDX++; \
435 } \
436 while (false)
437
438
439 /* Like FETCH_STRING_CHAR_ADVANCE, but fetch character from the current
440 buffer. */
441
442 #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \
443 do \
444 { \
445 CHARIDX++; \
446 if (!NILP (BVAR (current_buffer, enable_multibyte_characters))) \
447 { \
448 unsigned char *chp = BYTE_POS_ADDR (BYTEIDX); \
449 int chlen; \
450 \
451 OUTPUT = STRING_CHAR_AND_LENGTH (chp, chlen); \
452 BYTEIDX += chlen; \
453 } \
454 else \
455 { \
456 OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \
457 BYTEIDX++; \
458 } \
459 } \
460 while (false)
461
462
463 /* Like FETCH_CHAR_ADVANCE, but assumes the current buffer is multibyte. */
464
465 #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \
466 do \
467 { \
468 unsigned char *chp = BYTE_POS_ADDR (BYTEIDX); \
469 int chlen; \
470 \
471 OUTPUT = STRING_CHAR_AND_LENGTH (chp, chlen); \
472 BYTEIDX += chlen; \
473 CHARIDX++; \
474 } \
475 while (false)
476
477
478 /* Increment the buffer byte position POS_BYTE of the current buffer to
479 the next character boundary. No range checking of POS. */
480
481 #define INC_POS(pos_byte) \
482 do { \
483 unsigned char *chp = BYTE_POS_ADDR (pos_byte); \
484 pos_byte += BYTES_BY_CHAR_HEAD (*chp); \
485 } while (false)
486
487
488 /* Decrement the buffer byte position POS_BYTE of the current buffer to
489 the previous character boundary. No range checking of POS. */
490
491 #define DEC_POS(pos_byte) \
492 do { \
493 unsigned char *chp; \
494 \
495 pos_byte--; \
496 if (pos_byte < GPT_BYTE) \
497 chp = BEG_ADDR + pos_byte - BEG_BYTE; \
498 else \
499 chp = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE; \
500 while (!CHAR_HEAD_P (*chp)) \
501 { \
502 chp--; \
503 pos_byte--; \
504 } \
505 } while (false)
506
507 /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */
508
509 #define INC_BOTH(charpos, bytepos) \
510 do \
511 { \
512 (charpos)++; \
513 if (NILP (BVAR (current_buffer, enable_multibyte_characters))) \
514 (bytepos)++; \
515 else \
516 INC_POS ((bytepos)); \
517 } \
518 while (false)
519
520
521 /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */
522
523 #define DEC_BOTH(charpos, bytepos) \
524 do \
525 { \
526 (charpos)--; \
527 if (NILP (BVAR (current_buffer, enable_multibyte_characters))) \
528 (bytepos)--; \
529 else \
530 DEC_POS ((bytepos)); \
531 } \
532 while (false)
533
534
535 /* Increment the buffer byte position POS_BYTE of the current buffer to
536 the next character boundary. This macro relies on the fact that
537 *GPT_ADDR and *Z_ADDR are always accessible and the values are
538 '\0'. No range checking of POS_BYTE. */
539
540 #define BUF_INC_POS(buf, pos_byte) \
541 do { \
542 unsigned char *chp = BUF_BYTE_ADDRESS (buf, pos_byte); \
543 pos_byte += BYTES_BY_CHAR_HEAD (*chp); \
544 } while (false)
545
546
547 /* Decrement the buffer byte position POS_BYTE of the current buffer to
548 the previous character boundary. No range checking of POS_BYTE. */
549
550 #define BUF_DEC_POS(buf, pos_byte) \
551 do { \
552 unsigned char *chp; \
553 pos_byte--; \
554 if (pos_byte < BUF_GPT_BYTE (buf)) \
555 chp = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \
556 else \
557 chp = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\
558 while (!CHAR_HEAD_P (*chp)) \
559 { \
560 chp--; \
561 pos_byte--; \
562 } \
563 } while (false)
564
565
566 /* Return a non-outlandish value for the tab width. */
567
568 #define SANE_TAB_WIDTH(buf) \
569 sanitize_tab_width (XFASTINT (BVAR (buf, tab_width)))
570 INLINE int
571 sanitize_tab_width (EMACS_INT width)
572 {
573 return 0 < width && width <= 1000 ? width : 8;
574 }
575
576 /* Return the width of ASCII character C. The width is measured by
577 how many columns C will occupy on the screen when displayed in the
578 current buffer. */
579
580 #define ASCII_CHAR_WIDTH(c) \
581 (c < 0x20 \
582 ? (c == '\t' \
583 ? SANE_TAB_WIDTH (current_buffer) \
584 : (c == '\n' ? 0 : (NILP (BVAR (current_buffer, ctl_arrow)) ? 4 : 2))) \
585 : (c < 0x7f \
586 ? 1 \
587 : ((NILP (BVAR (current_buffer, ctl_arrow)) ? 4 : 2))))
588
589 /* Return a non-outlandish value for a character width. */
590
591 INLINE int
592 sanitize_char_width (EMACS_INT width)
593 {
594 return 0 <= width && width <= 1000 ? width : 1000;
595 }
596
597 /* Return the width of character C. The width is measured by how many
598 columns C will occupy on the screen when displayed in the current
599 buffer. */
600
601 #define CHAR_WIDTH(c) \
602 (ASCII_CHAR_P (c) \
603 ? ASCII_CHAR_WIDTH (c) \
604 : sanitize_char_width (XINT (CHAR_TABLE_REF (Vchar_width_table, c))))
605
606 /* If C is a variation selector, return the index of the
607 variation selector (1..256). Otherwise, return 0. */
608
609 #define CHAR_VARIATION_SELECTOR_P(c) \
610 ((c) < 0xFE00 ? 0 \
611 : (c) <= 0xFE0F ? (c) - 0xFE00 + 1 \
612 : (c) < 0xE0100 ? 0 \
613 : (c) <= 0xE01EF ? (c) - 0xE0100 + 17 \
614 : 0)
615
616 /* If C is a high surrogate, return 1. If C is a low surrogate,
617 return 2. Otherwise, return 0. */
618
619 #define CHAR_SURROGATE_PAIR_P(c) \
620 ((c) < 0xD800 ? 0 \
621 : (c) <= 0xDBFF ? 1 \
622 : (c) <= 0xDFFF ? 2 \
623 : 0)
624
625 /* Data type for Unicode general category.
626
627 The order of members must be in sync with the 8th element of the
628 member of unidata-prop-alist (in admin/unidata/unidata-gen.el) for
629 Unicode character property `general-category'. */
630
631 typedef enum {
632 UNICODE_CATEGORY_UNKNOWN = 0,
633 UNICODE_CATEGORY_Lu,
634 UNICODE_CATEGORY_Ll,
635 UNICODE_CATEGORY_Lt,
636 UNICODE_CATEGORY_Lm,
637 UNICODE_CATEGORY_Lo,
638 UNICODE_CATEGORY_Mn,
639 UNICODE_CATEGORY_Mc,
640 UNICODE_CATEGORY_Me,
641 UNICODE_CATEGORY_Nd,
642 UNICODE_CATEGORY_Nl,
643 UNICODE_CATEGORY_No,
644 UNICODE_CATEGORY_Pc,
645 UNICODE_CATEGORY_Pd,
646 UNICODE_CATEGORY_Ps,
647 UNICODE_CATEGORY_Pe,
648 UNICODE_CATEGORY_Pi,
649 UNICODE_CATEGORY_Pf,
650 UNICODE_CATEGORY_Po,
651 UNICODE_CATEGORY_Sm,
652 UNICODE_CATEGORY_Sc,
653 UNICODE_CATEGORY_Sk,
654 UNICODE_CATEGORY_So,
655 UNICODE_CATEGORY_Zs,
656 UNICODE_CATEGORY_Zl,
657 UNICODE_CATEGORY_Zp,
658 UNICODE_CATEGORY_Cc,
659 UNICODE_CATEGORY_Cf,
660 UNICODE_CATEGORY_Cs,
661 UNICODE_CATEGORY_Co,
662 UNICODE_CATEGORY_Cn
663 } unicode_category_t;
664
665 extern EMACS_INT char_resolve_modifier_mask (EMACS_INT) ATTRIBUTE_CONST;
666 extern int char_string (unsigned, unsigned char *);
667 extern int string_char (const unsigned char *,
668 const unsigned char **, int *);
669
670 extern int translate_char (Lisp_Object, int c);
671 extern ptrdiff_t count_size_as_multibyte (const unsigned char *, ptrdiff_t);
672 extern ptrdiff_t str_as_multibyte (unsigned char *, ptrdiff_t, ptrdiff_t,
673 ptrdiff_t *);
674 extern ptrdiff_t str_to_multibyte (unsigned char *, ptrdiff_t, ptrdiff_t);
675 extern ptrdiff_t str_as_unibyte (unsigned char *, ptrdiff_t);
676 extern ptrdiff_t str_to_unibyte (const unsigned char *, unsigned char *,
677 ptrdiff_t);
678 extern ptrdiff_t strwidth (const char *, ptrdiff_t);
679 extern ptrdiff_t c_string_width (const unsigned char *, ptrdiff_t, int,
680 ptrdiff_t *, ptrdiff_t *);
681 extern ptrdiff_t lisp_string_width (Lisp_Object, ptrdiff_t,
682 ptrdiff_t *, ptrdiff_t *);
683
684 extern Lisp_Object Vchar_unify_table;
685 extern Lisp_Object string_escape_byte8 (Lisp_Object);
686
687 extern bool alphabeticp (int);
688 extern bool decimalnump (int);
689 extern bool graphicp (int);
690 extern bool printablep (int);
691
692 /* Return a translation table of id number ID. */
693 #define GET_TRANSLATION_TABLE(id) \
694 (XCDR (XVECTOR (Vtranslation_table_vector)->contents[(id)]))
695
696 /* Look up the element in char table OBJ at index CH, and return it as
697 an integer. If the element is not a character, return CH itself. */
698
699 INLINE int
700 char_table_translate (Lisp_Object obj, int ch)
701 {
702 /* This internal function is expected to be called with valid arguments,
703 so there is a eassert instead of CHECK_xxx for the sake of speed. */
704 eassert (CHAR_VALID_P (ch));
705 eassert (CHAR_TABLE_P (obj));
706 obj = CHAR_TABLE_REF (obj, ch);
707 return CHARACTERP (obj) ? XINT (obj) : ch;
708 }
709
710 INLINE_HEADER_END
711
712 #endif /* EMACS_CHARACTER_H */
Emacs repository mirror