Commit | Line | Data |
---|---|---|
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 | ||
8 | This file is part of GNU Emacs. | |
9 | ||
b9b1cc14 | 10 | GNU Emacs is free software: you can redistribute it and/or modify |
0168c3d8 | 11 | it under the terms of the GNU General Public License as published by |
b9b1cc14 GM |
12 | the Free Software Foundation, either version 3 of the License, or |
13 | (at your option) any later version. | |
0168c3d8 KH |
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 | |
b9b1cc14 | 21 | along 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 | 70 | |
2afc21f5 SM |
71 | /* Return the raw 8-bit byte for character C, |
72 | or -1 if C doesn't correspond to a byte. */ | |
73 | #define CHAR_TO_BYTE_SAFE(c) \ | |
74 | (CHAR_BYTE8_P (c) \ | |
75 | ? (c) - 0x3FFF00 \ | |
455af463 | 76 | : multibyte_char_to_unibyte_safe (c)) |
2afc21f5 | 77 | |
885317d8 KH |
78 | /* Nonzero iff BYTE is the 1st byte of a multibyte form of a character |
79 | that corresponds to a raw 8-bit byte. */ | |
0168c3d8 KH |
80 | #define CHAR_BYTE8_HEAD_P(byte) ((byte) == 0xC0 || (byte) == 0xC1) |
81 | ||
43c47483 KH |
82 | /* Mapping table from unibyte chars to multibyte chars. */ |
83 | extern int unibyte_to_multibyte_table[256]; | |
3e411074 | 84 | |
43c47483 KH |
85 | /* Convert the unibyte character C to the corresponding multibyte |
86 | character. If C can't be converted, return C. */ | |
87 | #define unibyte_char_to_multibyte(c) \ | |
88 | ((c) < 256 ? unibyte_to_multibyte_table[(c)] : (c)) | |
3e411074 | 89 | |
1acbd067 KH |
90 | /* Nth element is 1 iff unibyte char N can be mapped to a multibyte |
91 | char. */ | |
92 | extern char unibyte_has_multibyte_table[256]; | |
93 | ||
94 | #define UNIBYTE_CHAR_HAS_MULTIBYTE_P(c) (unibyte_has_multibyte_table[(c)]) | |
95 | ||
43c47483 KH |
96 | /* If C is not ASCII, make it unibyte. */ |
97 | #define MAKE_CHAR_UNIBYTE(c) \ | |
98 | do { \ | |
99 | if (! ASCII_CHAR_P (c)) \ | |
100 | c = CHAR_TO_BYTE8 (c); \ | |
101 | } while (0) | |
3e411074 | 102 | |
3e411074 | 103 | |
18a2979d | 104 | /* If C is not ASCII, make it multibyte. Assumes C < 256. */ |
c0dc8f64 SM |
105 | #define MAKE_CHAR_MULTIBYTE(c) \ |
106 | (eassert ((c) >= 0 && (c) < 256), (c) = unibyte_to_multibyte_table[(c)]) | |
3e411074 | 107 | |
885317d8 | 108 | /* This is the maximum byte length of multibyte form. */ |
0168c3d8 KH |
109 | #define MAX_MULTIBYTE_LENGTH 5 |
110 | ||
18a2979d | 111 | /* Return a Lisp character whose character code is C. Assumes C is |
b583cead | 112 | a valid character code. */ |
0168c3d8 KH |
113 | #define make_char(c) make_number (c) |
114 | ||
115 | /* Nonzero iff C is an ASCII byte. */ | |
116 | #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80) | |
117 | ||
118 | /* Nonzero iff X is a character. */ | |
119 | #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR) | |
120 | ||
18a2979d | 121 | /* Nonzero iff C is valid as a character code. GENERICP is not used. */ |
885317d8 | 122 | #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR) |
0168c3d8 KH |
123 | |
124 | /* Check if Lisp object X is a character or not. */ | |
63db3c1b MB |
125 | #define CHECK_CHARACTER(x) \ |
126 | CHECK_TYPE (CHARACTERP (x), Qcharacterp, x) | |
0168c3d8 | 127 | |
8f924df7 KH |
128 | #define CHECK_CHARACTER_CAR(x) \ |
129 | do { \ | |
130 | Lisp_Object tmp = XCAR (x); \ | |
131 | CHECK_CHARACTER (tmp); \ | |
132 | XSETCAR ((x), tmp); \ | |
133 | } while (0) | |
134 | ||
135 | #define CHECK_CHARACTER_CDR(x) \ | |
136 | do { \ | |
137 | Lisp_Object tmp = XCDR (x); \ | |
138 | CHECK_CHARACTER (tmp); \ | |
139 | XSETCDR ((x), tmp); \ | |
140 | } while (0) | |
141 | ||
0168c3d8 KH |
142 | /* Nonzero iff C is an ASCII character. */ |
143 | #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80) | |
144 | ||
145 | /* Nonzero iff C is a character of code less than 0x100. */ | |
146 | #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100) | |
147 | ||
148 | /* Nonzero if character C has a printable glyph. */ | |
149 | #define CHAR_PRINTABLE_P(c) \ | |
150 | (((c) >= 32 && ((c) < 127) \ | |
151 | || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c))))) | |
152 | ||
885317d8 | 153 | /* Return byte length of multibyte form for character C. */ |
0168c3d8 KH |
154 | #define CHAR_BYTES(c) \ |
155 | ( (c) <= MAX_1_BYTE_CHAR ? 1 \ | |
156 | : (c) <= MAX_2_BYTE_CHAR ? 2 \ | |
157 | : (c) <= MAX_3_BYTE_CHAR ? 3 \ | |
158 | : (c) <= MAX_4_BYTE_CHAR ? 4 \ | |
159 | : (c) <= MAX_5_BYTE_CHAR ? 5 \ | |
160 | : 2) | |
161 | ||
43c47483 KH |
162 | |
163 | /* Return the leading code of multibyte form of C. */ | |
164 | #define CHAR_LEADING_CODE(c) \ | |
165 | ((c) <= MAX_1_BYTE_CHAR ? c \ | |
166 | : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \ | |
167 | : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \ | |
168 | : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \ | |
169 | : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \ | |
170 | : (0xC0 | (((c) >> 6) & 0x01))) | |
171 | ||
172 | ||
885317d8 KH |
173 | /* Store multibyte form of the character C in P. The caller should |
174 | allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance. | |
175 | Returns the length of the multibyte form. */ | |
0168c3d8 KH |
176 | |
177 | #define CHAR_STRING(c, p) \ | |
178 | ((unsigned) (c) <= MAX_1_BYTE_CHAR \ | |
179 | ? ((p)[0] = (c), \ | |
180 | 1) \ | |
181 | : (unsigned) (c) <= MAX_2_BYTE_CHAR \ | |
182 | ? ((p)[0] = (0xC0 | ((c) >> 6)), \ | |
183 | (p)[1] = (0x80 | ((c) & 0x3F)), \ | |
184 | 2) \ | |
185 | : (unsigned) (c) <= MAX_3_BYTE_CHAR \ | |
186 | ? ((p)[0] = (0xE0 | ((c) >> 12)), \ | |
187 | (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \ | |
188 | (p)[2] = (0x80 | ((c) & 0x3F)), \ | |
189 | 3) \ | |
f958a2fa | 190 | : char_string ((unsigned) c, p)) |
0168c3d8 | 191 | |
eb41da4c KH |
192 | /* Store multibyte form of byte B in P. The caller should allocate at |
193 | least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the | |
194 | length of the multibyte form. */ | |
1106ea2b KH |
195 | |
196 | #define BYTE8_STRING(b, p) \ | |
197 | ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \ | |
7f464917 | 198 | (p)[1] = (0x80 | ((b) & 0x3F)), \ |
1106ea2b KH |
199 | 2) |
200 | ||
0168c3d8 | 201 | |
18a2979d EZ |
202 | /* Store multibyte form of the character C in P and advance P to the |
203 | end of the multibyte form. The caller should allocate at least | |
204 | MAX_MULTIBYTE_LENGTH bytes area at P in advance. */ | |
0168c3d8 | 205 | |
eb41da4c KH |
206 | #define CHAR_STRING_ADVANCE(c, p) \ |
207 | do { \ | |
208 | if ((c) <= MAX_1_BYTE_CHAR) \ | |
209 | *(p)++ = (c); \ | |
210 | else if ((c) <= MAX_2_BYTE_CHAR) \ | |
211 | *(p)++ = (0xC0 | ((c) >> 6)), \ | |
212 | *(p)++ = (0x80 | ((c) & 0x3F)); \ | |
213 | else if ((c) <= MAX_3_BYTE_CHAR) \ | |
214 | *(p)++ = (0xE0 | ((c) >> 12)), \ | |
215 | *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \ | |
216 | *(p)++ = (0x80 | ((c) & 0x3F)); \ | |
217 | else \ | |
218 | (p) += char_string ((c), (p)); \ | |
885317d8 | 219 | } while (0) |
0168c3d8 | 220 | |
eb41da4c | 221 | |
0168c3d8 KH |
222 | /* Nonzero iff BYTE starts a non-ASCII character in a multibyte |
223 | form. */ | |
224 | #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0) | |
225 | ||
b5c7dbe6 KH |
226 | /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a |
227 | multibyte form. */ | |
228 | #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80) | |
229 | ||
885317d8 KH |
230 | /* Nonzero iff BYTE starts a character in a multibyte form. |
231 | This is equivalent to: | |
232 | (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */ | |
233 | #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80) | |
234 | ||
18a2979d EZ |
235 | /* Kept for backward compatibility. This macro will be removed in the |
236 | future. */ | |
0168c3d8 KH |
237 | #define BASE_LEADING_CODE_P LEADING_CODE_P |
238 | ||
239 | /* How many bytes a character that starts with BYTE occupies in a | |
240 | multibyte form. */ | |
241 | #define BYTES_BY_CHAR_HEAD(byte) \ | |
242 | (!((byte) & 0x80) ? 1 \ | |
243 | : !((byte) & 0x20) ? 2 \ | |
244 | : !((byte) & 0x10) ? 3 \ | |
245 | : !((byte) & 0x08) ? 4 \ | |
246 | : 5) | |
247 | ||
248 | ||
249 | /* Return the length of the multi-byte form at string STR of length | |
250 | LEN while assuming that STR points a valid multi-byte form. As | |
251 | this macro isn't necessary anymore, all callers will be changed to | |
252 | use BYTES_BY_CHAR_HEAD directly in the future. */ | |
253 | ||
254 | #define MULTIBYTE_FORM_LENGTH(str, len) \ | |
255 | BYTES_BY_CHAR_HEAD (*(str)) | |
256 | ||
257 | /* Parse multibyte string STR of length LENGTH and set BYTES to the | |
258 | byte length of a character at STR while assuming that STR points a | |
259 | valid multibyte form. As this macro isn't necessary anymore, all | |
260 | callers will be changed to use BYTES_BY_CHAR_HEAD directly in the | |
261 | future. */ | |
262 | ||
263 | #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \ | |
264 | (bytes) = BYTES_BY_CHAR_HEAD (*(str)) | |
265 | ||
266 | /* The byte length of multibyte form at unibyte string P ending at | |
18a2979d | 267 | PEND. If STR doesn't point to a valid multibyte form, return 0. */ |
0168c3d8 KH |
268 | |
269 | #define MULTIBYTE_LENGTH(p, pend) \ | |
270 | (p >= pend ? 0 \ | |
271 | : !((p)[0] & 0x80) ? 1 \ | |
272 | : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \ | |
273 | : ((p)[0] & 0xE0) == 0xC0 ? 2 \ | |
274 | : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \ | |
275 | : ((p)[0] & 0xF0) == 0xE0 ? 3 \ | |
276 | : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \ | |
277 | : ((p)[0] & 0xF8) == 0xF0 ? 4 \ | |
278 | : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \ | |
279 | : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \ | |
280 | : 0) | |
281 | ||
282 | ||
18a2979d | 283 | /* Like MULTIBYTE_LENGTH, but don't check the ending address. */ |
0168c3d8 KH |
284 | |
285 | #define MULTIBYTE_LENGTH_NO_CHECK(p) \ | |
286 | (!((p)[0] & 0x80) ? 1 \ | |
287 | : ((p)[1] & 0xC0) != 0x80 ? 0 \ | |
288 | : ((p)[0] & 0xE0) == 0xC0 ? 2 \ | |
289 | : ((p)[2] & 0xC0) != 0x80 ? 0 \ | |
290 | : ((p)[0] & 0xF0) == 0xE0 ? 3 \ | |
291 | : ((p)[3] & 0xC0) != 0x80 ? 0 \ | |
292 | : ((p)[0] & 0xF8) == 0xF0 ? 4 \ | |
293 | : ((p)[4] & 0xC0) != 0x80 ? 0 \ | |
294 | : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \ | |
295 | : 0) | |
296 | ||
18a2979d EZ |
297 | /* If P is before LIMIT, advance P to the next character boundary. |
298 | Assumes that P is already at a character boundary of the same | |
8f924df7 KH |
299 | mulitbyte form whose end address is LIMIT. */ |
300 | ||
301 | #define NEXT_CHAR_BOUNDARY(p, limit) \ | |
302 | do { \ | |
303 | if ((p) < (limit)) \ | |
304 | (p) += BYTES_BY_CHAR_HEAD (*(p)); \ | |
305 | } while (0) | |
306 | ||
307 | ||
308 | /* If P is after LIMIT, advance P to the previous character boundary. | |
18a2979d | 309 | Assumes that P is already at a character boundary of the same |
8f924df7 KH |
310 | mulitbyte form whose beginning address is LIMIT. */ |
311 | ||
312 | #define PREV_CHAR_BOUNDARY(p, limit) \ | |
313 | do { \ | |
314 | if ((p) > (limit)) \ | |
315 | { \ | |
316 | const unsigned char *p0 = (p); \ | |
317 | do { \ | |
318 | p0--; \ | |
319 | } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \ | |
320 | (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \ | |
321 | } \ | |
322 | } while (0) | |
0168c3d8 KH |
323 | |
324 | /* Return the character code of character whose multibyte form is at | |
325 | P. The argument LEN is ignored. It will be removed in the | |
326 | future. */ | |
327 | ||
328 | #define STRING_CHAR(p, len) \ | |
329 | (!((p)[0] & 0x80) \ | |
330 | ? (p)[0] \ | |
331 | : ! ((p)[0] & 0x20) \ | |
332 | ? (((((p)[0] & 0x1F) << 6) \ | |
333 | | ((p)[1] & 0x3F)) \ | |
334 | + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \ | |
335 | : ! ((p)[0] & 0x10) \ | |
336 | ? ((((p)[0] & 0x0F) << 12) \ | |
337 | | (((p)[1] & 0x3F) << 6) \ | |
338 | | ((p)[2] & 0x3F)) \ | |
eb41da4c | 339 | : string_char ((p), NULL, NULL)) |
0168c3d8 KH |
340 | |
341 | ||
18a2979d | 342 | /* Like STRING_CHAR, but set ACTUAL_LEN to the length of multibyte |
0168c3d8 KH |
343 | form. The argument LEN is ignored. It will be removed in the |
344 | future. */ | |
345 | ||
346 | #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \ | |
347 | (!((p)[0] & 0x80) \ | |
348 | ? ((actual_len) = 1, (p)[0]) \ | |
349 | : ! ((p)[0] & 0x20) \ | |
350 | ? ((actual_len) = 2, \ | |
351 | (((((p)[0] & 0x1F) << 6) \ | |
352 | | ((p)[1] & 0x3F)) \ | |
353 | + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \ | |
354 | : ! ((p)[0] & 0x10) \ | |
355 | ? ((actual_len) = 3, \ | |
356 | ((((p)[0] & 0x0F) << 12) \ | |
357 | | (((p)[1] & 0x3F) << 6) \ | |
358 | | ((p)[2] & 0x3F))) \ | |
eb41da4c | 359 | : string_char ((p), NULL, &actual_len)) |
0168c3d8 KH |
360 | |
361 | ||
18a2979d | 362 | /* Like STRING_CHAR, but advance P to the end of multibyte form. */ |
0168c3d8 KH |
363 | |
364 | #define STRING_CHAR_ADVANCE(p) \ | |
365 | (!((p)[0] & 0x80) \ | |
366 | ? *(p)++ \ | |
367 | : ! ((p)[0] & 0x20) \ | |
368 | ? ((p) += 2, \ | |
369 | ((((p)[-2] & 0x1F) << 6) \ | |
370 | | ((p)[-1] & 0x3F) \ | |
8f924df7 | 371 | | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \ |
0168c3d8 KH |
372 | : ! ((p)[0] & 0x10) \ |
373 | ? ((p) += 3, \ | |
374 | ((((p)[-3] & 0x0F) << 12) \ | |
375 | | (((p)[-2] & 0x3F) << 6) \ | |
376 | | ((p)[-1] & 0x3F))) \ | |
eb41da4c | 377 | : string_char ((p), &(p), NULL)) |
0168c3d8 KH |
378 | |
379 | ||
380 | /* Fetch the "next" character from Lisp string STRING at byte position | |
381 | BYTEIDX, character position CHARIDX. Store it into OUTPUT. | |
382 | ||
383 | All the args must be side-effect-free. | |
384 | BYTEIDX and CHARIDX must be lvalues; | |
385 | we increment them past the character fetched. */ | |
386 | ||
387 | #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 388 | do \ |
0168c3d8 KH |
389 | { \ |
390 | CHARIDX++; \ | |
391 | if (STRING_MULTIBYTE (STRING)) \ | |
392 | { \ | |
f1c99628 | 393 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
0168c3d8 KH |
394 | int len; \ |
395 | \ | |
396 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
397 | BYTEIDX += len; \ | |
398 | } \ | |
399 | else \ | |
f1c99628 SM |
400 | { \ |
401 | OUTPUT = SREF (STRING, BYTEIDX); \ | |
402 | BYTEIDX++; \ | |
403 | } \ | |
0168c3d8 | 404 | } \ |
a1c2ac9a | 405 | while (0) |
0168c3d8 | 406 | |
18a2979d EZ |
407 | /* Like FETCH_STRING_CHAR_ADVANCE, but return a multibyte character |
408 | even if STRING is unibyte. */ | |
43c47483 KH |
409 | |
410 | #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 411 | do \ |
43c47483 KH |
412 | { \ |
413 | CHARIDX++; \ | |
414 | if (STRING_MULTIBYTE (STRING)) \ | |
415 | { \ | |
f1c99628 | 416 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
43c47483 KH |
417 | int len; \ |
418 | \ | |
419 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
420 | BYTEIDX += len; \ | |
421 | } \ | |
422 | else \ | |
423 | { \ | |
f1c99628 SM |
424 | OUTPUT = SREF (STRING, BYTEIDX); \ |
425 | BYTEIDX++; \ | |
43c47483 KH |
426 | MAKE_CHAR_MULTIBYTE (OUTPUT); \ |
427 | } \ | |
428 | } \ | |
a1c2ac9a | 429 | while (0) |
43c47483 | 430 | |
0168c3d8 | 431 | |
18a2979d | 432 | /* Like FETCH_STRING_CHAR_ADVANCE, but assumes STRING is multibyte. */ |
0168c3d8 KH |
433 | |
434 | #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 435 | do \ |
0168c3d8 | 436 | { \ |
f1c99628 | 437 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
0168c3d8 KH |
438 | int len; \ |
439 | \ | |
440 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
441 | BYTEIDX += len; \ | |
442 | CHARIDX++; \ | |
443 | } \ | |
a1c2ac9a | 444 | while (0) |
0168c3d8 KH |
445 | |
446 | ||
18a2979d | 447 | /* Like FETCH_STRING_CHAR_ADVANCE, but fetch character from the current |
0168c3d8 KH |
448 | buffer. */ |
449 | ||
450 | #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 451 | do \ |
0168c3d8 KH |
452 | { \ |
453 | CHARIDX++; \ | |
454 | if (!NILP (current_buffer->enable_multibyte_characters)) \ | |
455 | { \ | |
456 | unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \ | |
457 | int len; \ | |
458 | \ | |
459 | OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
460 | BYTEIDX += len; \ | |
461 | } \ | |
462 | else \ | |
463 | { \ | |
464 | OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \ | |
465 | BYTEIDX++; \ | |
466 | } \ | |
467 | } \ | |
a1c2ac9a | 468 | while (0) |
0168c3d8 KH |
469 | |
470 | ||
18a2979d | 471 | /* Like FETCH_CHAR_ADVANCE, but assumes the current buffer is multibyte. */ |
0168c3d8 KH |
472 | |
473 | #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 474 | do \ |
0168c3d8 KH |
475 | { \ |
476 | unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \ | |
477 | int len; \ | |
478 | \ | |
479 | OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
480 | BYTEIDX += len; \ | |
481 | CHARIDX++; \ | |
482 | } \ | |
a1c2ac9a | 483 | while (0) |
0168c3d8 KH |
484 | |
485 | ||
18a2979d | 486 | /* Increment the buffer byte position POS_BYTE of the current buffer to |
0168c3d8 KH |
487 | the next character boundary. No range checking of POS. */ |
488 | ||
489 | #define INC_POS(pos_byte) \ | |
490 | do { \ | |
491 | unsigned char *p = BYTE_POS_ADDR (pos_byte); \ | |
492 | pos_byte += BYTES_BY_CHAR_HEAD (*p); \ | |
493 | } while (0) | |
494 | ||
495 | ||
18a2979d | 496 | /* Decrement the buffer byte position POS_BYTE of the current buffer to |
0168c3d8 KH |
497 | the previous character boundary. No range checking of POS. */ |
498 | ||
499 | #define DEC_POS(pos_byte) \ | |
500 | do { \ | |
501 | unsigned char *p; \ | |
502 | \ | |
503 | pos_byte--; \ | |
504 | if (pos_byte < GPT_BYTE) \ | |
f1c99628 | 505 | p = BEG_ADDR + pos_byte - BEG_BYTE; \ |
0168c3d8 | 506 | else \ |
f1c99628 | 507 | p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\ |
0168c3d8 KH |
508 | while (!CHAR_HEAD_P (*p)) \ |
509 | { \ | |
510 | p--; \ | |
511 | pos_byte--; \ | |
512 | } \ | |
513 | } while (0) | |
514 | ||
515 | /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */ | |
516 | ||
517 | #define INC_BOTH(charpos, bytepos) \ | |
518 | do \ | |
519 | { \ | |
520 | (charpos)++; \ | |
521 | if (NILP (current_buffer->enable_multibyte_characters)) \ | |
522 | (bytepos)++; \ | |
523 | else \ | |
524 | INC_POS ((bytepos)); \ | |
525 | } \ | |
526 | while (0) | |
527 | ||
528 | ||
529 | /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */ | |
530 | ||
531 | #define DEC_BOTH(charpos, bytepos) \ | |
532 | do \ | |
533 | { \ | |
534 | (charpos)--; \ | |
535 | if (NILP (current_buffer->enable_multibyte_characters)) \ | |
536 | (bytepos)--; \ | |
537 | else \ | |
538 | DEC_POS ((bytepos)); \ | |
539 | } \ | |
540 | while (0) | |
541 | ||
542 | ||
18a2979d | 543 | /* Increment the buffer byte position POS_BYTE of the current buffer to |
0168c3d8 KH |
544 | the next character boundary. This macro relies on the fact that |
545 | *GPT_ADDR and *Z_ADDR are always accessible and the values are | |
546 | '\0'. No range checking of POS_BYTE. */ | |
547 | ||
548 | #define BUF_INC_POS(buf, pos_byte) \ | |
549 | do { \ | |
550 | unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \ | |
551 | pos_byte += BYTES_BY_CHAR_HEAD (*p); \ | |
552 | } while (0) | |
553 | ||
554 | ||
18a2979d | 555 | /* Decrement the buffer byte position POS_BYTE of the current buffer to |
0168c3d8 KH |
556 | the previous character boundary. No range checking of POS_BYTE. */ |
557 | ||
558 | #define BUF_DEC_POS(buf, pos_byte) \ | |
559 | do { \ | |
560 | unsigned char *p; \ | |
561 | pos_byte--; \ | |
562 | if (pos_byte < BUF_GPT_BYTE (buf)) \ | |
f1c99628 | 563 | p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \ |
0168c3d8 | 564 | else \ |
f1c99628 | 565 | p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\ |
0168c3d8 KH |
566 | while (!CHAR_HEAD_P (*p)) \ |
567 | { \ | |
568 | p--; \ | |
569 | pos_byte--; \ | |
570 | } \ | |
571 | } while (0) | |
572 | ||
573 | ||
b583cead KH |
574 | /* If C is a character to be unified with a Unicode character, return |
575 | the unified Unicode character. */ | |
576 | ||
55870ffc KH |
577 | #define MAYBE_UNIFY_CHAR(c) \ |
578 | do { \ | |
579 | if (c > MAX_UNICODE_CHAR && c <= MAX_5_BYTE_CHAR) \ | |
580 | { \ | |
581 | Lisp_Object val; \ | |
582 | val = CHAR_TABLE_REF (Vchar_unify_table, c); \ | |
583 | if (INTEGERP (val)) \ | |
584 | c = XINT (val); \ | |
585 | else if (! NILP (val)) \ | |
586 | c = maybe_unify_char (c, val); \ | |
587 | } \ | |
588 | } while (0) | |
0168c3d8 | 589 | |
fc9d9d2a | 590 | |
0168c3d8 | 591 | /* Return the width of ASCII character C. The width is measured by |
18a2979d | 592 | how many columns C will occupy on the screen when displayed in the |
0168c3d8 KH |
593 | current buffer. */ |
594 | ||
595 | #define ASCII_CHAR_WIDTH(c) \ | |
596 | (c < 0x20 \ | |
597 | ? (c == '\t' \ | |
598 | ? XFASTINT (current_buffer->tab_width) \ | |
599 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ | |
600 | : (c < 0x7f \ | |
601 | ? 1 \ | |
602 | : ((NILP (current_buffer->ctl_arrow) ? 4 : 2)))) | |
603 | ||
604 | /* Return the width of character C. The width is measured by how many | |
18a2979d | 605 | columns C will occupy on the screen when displayed in the current |
0168c3d8 KH |
606 | buffer. */ |
607 | ||
608 | #define CHAR_WIDTH(c) \ | |
609 | (ASCII_CHAR_P (c) \ | |
610 | ? ASCII_CHAR_WIDTH (c) \ | |
611 | : XINT (CHAR_TABLE_REF (Vchar_width_table, c))) | |
612 | ||
eb41da4c | 613 | extern int char_resolve_modifier_mask P_ ((int)); |
f958a2fa | 614 | extern int char_string P_ ((unsigned, unsigned char *)); |
eb41da4c KH |
615 | extern int string_char P_ ((const unsigned char *, |
616 | const unsigned char **, int *)); | |
0168c3d8 KH |
617 | |
618 | extern int translate_char P_ ((Lisp_Object, int c)); | |
619 | extern int char_printable_p P_ ((int c)); | |
8f924df7 KH |
620 | extern void parse_str_as_multibyte P_ ((const unsigned char *, int, int *, |
621 | int *)); | |
0168c3d8 KH |
622 | extern int parse_str_to_multibyte P_ ((unsigned char *, int)); |
623 | extern int str_as_multibyte P_ ((unsigned char *, int, int, int *)); | |
624 | extern int str_to_multibyte P_ ((unsigned char *, int, int)); | |
625 | extern int str_as_unibyte P_ ((unsigned char *, int)); | |
0460afe5 KH |
626 | extern EMACS_INT str_to_unibyte P_ ((const unsigned char *, unsigned char *, |
627 | EMACS_INT, int)); | |
0168c3d8 | 628 | extern int strwidth P_ ((unsigned char *, int)); |
8f924df7 | 629 | extern int c_string_width P_ ((const unsigned char *, int, int, int *, int *)); |
0168c3d8 KH |
630 | extern int lisp_string_width P_ ((Lisp_Object, int, int *, int *)); |
631 | ||
632 | extern Lisp_Object Vprintable_chars; | |
633 | ||
634 | extern Lisp_Object Qcharacterp, Qauto_fill_chars; | |
635 | extern Lisp_Object Vtranslation_table_vector; | |
636 | extern Lisp_Object Vchar_width_table; | |
637 | extern Lisp_Object Vchar_direction_table; | |
638 | extern Lisp_Object Vchar_unify_table; | |
a48a6418 | 639 | extern Lisp_Object Vunicode_category_table; |
0168c3d8 | 640 | |
fac2bdc4 DL |
641 | extern Lisp_Object string_escape_byte8 P_ ((Lisp_Object)); |
642 | ||
0168c3d8 KH |
643 | /* Return a translation table of id number ID. */ |
644 | #define GET_TRANSLATION_TABLE(id) \ | |
645 | (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)])) | |
646 | ||
647 | /* A char-table for characters which may invoke auto-filling. */ | |
648 | extern Lisp_Object Vauto_fill_chars; | |
649 | ||
e18ef64a | 650 | extern Lisp_Object Vchar_script_table; |
e0d6e5a5 | 651 | extern Lisp_Object Vscript_representative_chars; |
b5c7dbe6 | 652 | |
0168c3d8 KH |
653 | /* Copy LEN bytes from FROM to TO. This macro should be used only |
654 | when a caller knows that LEN is short and the obvious copy loop is | |
655 | faster than calling bcopy which has some overhead. Copying a | |
656 | multibyte sequence of a character is the typical case. */ | |
657 | ||
658 | #define BCOPY_SHORT(from, to, len) \ | |
659 | do { \ | |
660 | int i = len; \ | |
661 | unsigned char *from_p = from, *to_p = to; \ | |
662 | while (i--) *to_p++ = *from_p++; \ | |
663 | } while (0) | |
664 | ||
665 | #define DEFSYM(sym, name) \ | |
666 | do { (sym) = intern ((name)); staticpro (&(sym)); } while (0) | |
667 | ||
668 | #endif /* EMACS_CHARACTER_H */ | |
fbaf0946 MB |
669 | |
670 | /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac | |
671 | (do not change this comment) */ |