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 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. */ |
76 | extern 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. */ | |
85 | extern 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 | 97 | /* If C is not ASCII, make it multibyte. It assumes C < 256. */ |
c0dc8f64 SM |
98 | #define MAKE_CHAR_MULTIBYTE(c) \ |
99 | (eassert ((c) >= 0 && (c) < 256), (c) = unibyte_to_multibyte_table[(c)]) | |
3e411074 | 100 | |
885317d8 | 101 | /* This is the maximum byte length of multibyte form. */ |
0168c3d8 KH |
102 | #define MAX_MULTIBYTE_LENGTH 5 |
103 | ||
b583cead KH |
104 | /* Return a Lisp character whose character code is C. It assumes C is |
105 | a valid character code. */ | |
0168c3d8 KH |
106 | #define make_char(c) make_number (c) |
107 | ||
108 | /* Nonzero iff C is an ASCII byte. */ | |
109 | #define ASCII_BYTE_P(c) ((unsigned) (c) < 0x80) | |
110 | ||
111 | /* Nonzero iff X is a character. */ | |
112 | #define CHARACTERP(x) (NATNUMP (x) && XFASTINT (x) <= MAX_CHAR) | |
113 | ||
f65c6d94 | 114 | /* Nonzero iff C is valid as a character code. GENERICP is not used |
885317d8 KH |
115 | now. */ |
116 | #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR) | |
0168c3d8 KH |
117 | |
118 | /* Check if Lisp object X is a character or not. */ | |
63db3c1b MB |
119 | #define CHECK_CHARACTER(x) \ |
120 | CHECK_TYPE (CHARACTERP (x), Qcharacterp, x) | |
0168c3d8 | 121 | |
8f924df7 KH |
122 | #define CHECK_CHARACTER_CAR(x) \ |
123 | do { \ | |
124 | Lisp_Object tmp = XCAR (x); \ | |
125 | CHECK_CHARACTER (tmp); \ | |
126 | XSETCAR ((x), tmp); \ | |
127 | } while (0) | |
128 | ||
129 | #define CHECK_CHARACTER_CDR(x) \ | |
130 | do { \ | |
131 | Lisp_Object tmp = XCDR (x); \ | |
132 | CHECK_CHARACTER (tmp); \ | |
133 | XSETCDR ((x), tmp); \ | |
134 | } while (0) | |
135 | ||
0168c3d8 KH |
136 | /* Nonzero iff C is an ASCII character. */ |
137 | #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80) | |
138 | ||
139 | /* Nonzero iff C is a character of code less than 0x100. */ | |
140 | #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100) | |
141 | ||
142 | /* Nonzero if character C has a printable glyph. */ | |
143 | #define CHAR_PRINTABLE_P(c) \ | |
144 | (((c) >= 32 && ((c) < 127) \ | |
145 | || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c))))) | |
146 | ||
885317d8 | 147 | /* Return byte length of multibyte form for character C. */ |
0168c3d8 KH |
148 | #define CHAR_BYTES(c) \ |
149 | ( (c) <= MAX_1_BYTE_CHAR ? 1 \ | |
150 | : (c) <= MAX_2_BYTE_CHAR ? 2 \ | |
151 | : (c) <= MAX_3_BYTE_CHAR ? 3 \ | |
152 | : (c) <= MAX_4_BYTE_CHAR ? 4 \ | |
153 | : (c) <= MAX_5_BYTE_CHAR ? 5 \ | |
154 | : 2) | |
155 | ||
43c47483 KH |
156 | |
157 | /* Return the leading code of multibyte form of C. */ | |
158 | #define CHAR_LEADING_CODE(c) \ | |
159 | ((c) <= MAX_1_BYTE_CHAR ? c \ | |
160 | : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \ | |
161 | : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \ | |
162 | : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \ | |
163 | : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \ | |
164 | : (0xC0 | (((c) >> 6) & 0x01))) | |
165 | ||
166 | ||
885317d8 KH |
167 | /* Store multibyte form of the character C in P. The caller should |
168 | allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance. | |
169 | Returns the length of the multibyte form. */ | |
0168c3d8 KH |
170 | |
171 | #define CHAR_STRING(c, p) \ | |
172 | ((unsigned) (c) <= MAX_1_BYTE_CHAR \ | |
173 | ? ((p)[0] = (c), \ | |
174 | 1) \ | |
175 | : (unsigned) (c) <= MAX_2_BYTE_CHAR \ | |
176 | ? ((p)[0] = (0xC0 | ((c) >> 6)), \ | |
177 | (p)[1] = (0x80 | ((c) & 0x3F)), \ | |
178 | 2) \ | |
179 | : (unsigned) (c) <= MAX_3_BYTE_CHAR \ | |
180 | ? ((p)[0] = (0xE0 | ((c) >> 12)), \ | |
181 | (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \ | |
182 | (p)[2] = (0x80 | ((c) & 0x3F)), \ | |
183 | 3) \ | |
f958a2fa | 184 | : char_string ((unsigned) c, p)) |
0168c3d8 | 185 | |
eb41da4c KH |
186 | /* Store multibyte form of byte B in P. The caller should allocate at |
187 | least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the | |
188 | length of the multibyte form. */ | |
1106ea2b KH |
189 | |
190 | #define BYTE8_STRING(b, p) \ | |
191 | ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \ | |
7f464917 | 192 | (p)[1] = (0x80 | ((b) & 0x3F)), \ |
1106ea2b KH |
193 | 2) |
194 | ||
0168c3d8 | 195 | |
885317d8 KH |
196 | /* Store multibyte form of the character C in P. The caller should |
197 | allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance. | |
198 | And, advance P to the end of the multibyte form. */ | |
0168c3d8 | 199 | |
eb41da4c KH |
200 | #define CHAR_STRING_ADVANCE(c, p) \ |
201 | do { \ | |
202 | if ((c) <= MAX_1_BYTE_CHAR) \ | |
203 | *(p)++ = (c); \ | |
204 | else if ((c) <= MAX_2_BYTE_CHAR) \ | |
205 | *(p)++ = (0xC0 | ((c) >> 6)), \ | |
206 | *(p)++ = (0x80 | ((c) & 0x3F)); \ | |
207 | else if ((c) <= MAX_3_BYTE_CHAR) \ | |
208 | *(p)++ = (0xE0 | ((c) >> 12)), \ | |
209 | *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \ | |
210 | *(p)++ = (0x80 | ((c) & 0x3F)); \ | |
211 | else \ | |
212 | (p) += char_string ((c), (p)); \ | |
885317d8 | 213 | } while (0) |
0168c3d8 | 214 | |
eb41da4c | 215 | |
0168c3d8 KH |
216 | /* Nonzero iff BYTE starts a non-ASCII character in a multibyte |
217 | form. */ | |
218 | #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0) | |
219 | ||
b5c7dbe6 KH |
220 | /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a |
221 | multibyte form. */ | |
222 | #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80) | |
223 | ||
885317d8 KH |
224 | /* Nonzero iff BYTE starts a character in a multibyte form. |
225 | This is equivalent to: | |
226 | (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */ | |
227 | #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80) | |
228 | ||
0168c3d8 KH |
229 | /* Just kept for backward compatibility. This macro will be removed |
230 | in the future. */ | |
231 | #define BASE_LEADING_CODE_P LEADING_CODE_P | |
232 | ||
233 | /* How many bytes a character that starts with BYTE occupies in a | |
234 | multibyte form. */ | |
235 | #define BYTES_BY_CHAR_HEAD(byte) \ | |
236 | (!((byte) & 0x80) ? 1 \ | |
237 | : !((byte) & 0x20) ? 2 \ | |
238 | : !((byte) & 0x10) ? 3 \ | |
239 | : !((byte) & 0x08) ? 4 \ | |
240 | : 5) | |
241 | ||
242 | ||
243 | /* Return the length of the multi-byte form at string STR of length | |
244 | LEN while assuming that STR points a valid multi-byte form. As | |
245 | this macro isn't necessary anymore, all callers will be changed to | |
246 | use BYTES_BY_CHAR_HEAD directly in the future. */ | |
247 | ||
248 | #define MULTIBYTE_FORM_LENGTH(str, len) \ | |
249 | BYTES_BY_CHAR_HEAD (*(str)) | |
250 | ||
251 | /* Parse multibyte string STR of length LENGTH and set BYTES to the | |
252 | byte length of a character at STR while assuming that STR points a | |
253 | valid multibyte form. As this macro isn't necessary anymore, all | |
254 | callers will be changed to use BYTES_BY_CHAR_HEAD directly in the | |
255 | future. */ | |
256 | ||
257 | #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \ | |
258 | (bytes) = BYTES_BY_CHAR_HEAD (*(str)) | |
259 | ||
260 | /* The byte length of multibyte form at unibyte string P ending at | |
261 | PEND. If STR doesn't point a valid multibyte form, return 0. */ | |
262 | ||
263 | #define MULTIBYTE_LENGTH(p, pend) \ | |
264 | (p >= pend ? 0 \ | |
265 | : !((p)[0] & 0x80) ? 1 \ | |
266 | : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \ | |
267 | : ((p)[0] & 0xE0) == 0xC0 ? 2 \ | |
268 | : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \ | |
269 | : ((p)[0] & 0xF0) == 0xE0 ? 3 \ | |
270 | : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \ | |
271 | : ((p)[0] & 0xF8) == 0xF0 ? 4 \ | |
272 | : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \ | |
273 | : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \ | |
274 | : 0) | |
275 | ||
276 | ||
277 | /* Like MULTIBYTE_LENGTH but don't check the ending address. */ | |
278 | ||
279 | #define MULTIBYTE_LENGTH_NO_CHECK(p) \ | |
280 | (!((p)[0] & 0x80) ? 1 \ | |
281 | : ((p)[1] & 0xC0) != 0x80 ? 0 \ | |
282 | : ((p)[0] & 0xE0) == 0xC0 ? 2 \ | |
283 | : ((p)[2] & 0xC0) != 0x80 ? 0 \ | |
284 | : ((p)[0] & 0xF0) == 0xE0 ? 3 \ | |
285 | : ((p)[3] & 0xC0) != 0x80 ? 0 \ | |
286 | : ((p)[0] & 0xF8) == 0xF0 ? 4 \ | |
287 | : ((p)[4] & 0xC0) != 0x80 ? 0 \ | |
288 | : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \ | |
289 | : 0) | |
290 | ||
8f924df7 KH |
291 | /* If P is before LIMIT, advance P to the next character boundary. It |
292 | assumes that P is already at a character boundary of the sane | |
293 | mulitbyte form whose end address is LIMIT. */ | |
294 | ||
295 | #define NEXT_CHAR_BOUNDARY(p, limit) \ | |
296 | do { \ | |
297 | if ((p) < (limit)) \ | |
298 | (p) += BYTES_BY_CHAR_HEAD (*(p)); \ | |
299 | } while (0) | |
300 | ||
301 | ||
302 | /* If P is after LIMIT, advance P to the previous character boundary. | |
303 | It assumes that P is already at a character boundary of the sane | |
304 | mulitbyte form whose beginning address is LIMIT. */ | |
305 | ||
306 | #define PREV_CHAR_BOUNDARY(p, limit) \ | |
307 | do { \ | |
308 | if ((p) > (limit)) \ | |
309 | { \ | |
310 | const unsigned char *p0 = (p); \ | |
311 | do { \ | |
312 | p0--; \ | |
313 | } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \ | |
314 | (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \ | |
315 | } \ | |
316 | } while (0) | |
0168c3d8 KH |
317 | |
318 | /* Return the character code of character whose multibyte form is at | |
319 | P. The argument LEN is ignored. It will be removed in the | |
320 | future. */ | |
321 | ||
322 | #define STRING_CHAR(p, len) \ | |
323 | (!((p)[0] & 0x80) \ | |
324 | ? (p)[0] \ | |
325 | : ! ((p)[0] & 0x20) \ | |
326 | ? (((((p)[0] & 0x1F) << 6) \ | |
327 | | ((p)[1] & 0x3F)) \ | |
328 | + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \ | |
329 | : ! ((p)[0] & 0x10) \ | |
330 | ? ((((p)[0] & 0x0F) << 12) \ | |
331 | | (((p)[1] & 0x3F) << 6) \ | |
332 | | ((p)[2] & 0x3F)) \ | |
eb41da4c | 333 | : string_char ((p), NULL, NULL)) |
0168c3d8 KH |
334 | |
335 | ||
336 | /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte | |
337 | form. The argument LEN is ignored. It will be removed in the | |
338 | future. */ | |
339 | ||
340 | #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \ | |
341 | (!((p)[0] & 0x80) \ | |
342 | ? ((actual_len) = 1, (p)[0]) \ | |
343 | : ! ((p)[0] & 0x20) \ | |
344 | ? ((actual_len) = 2, \ | |
345 | (((((p)[0] & 0x1F) << 6) \ | |
346 | | ((p)[1] & 0x3F)) \ | |
347 | + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \ | |
348 | : ! ((p)[0] & 0x10) \ | |
349 | ? ((actual_len) = 3, \ | |
350 | ((((p)[0] & 0x0F) << 12) \ | |
351 | | (((p)[1] & 0x3F) << 6) \ | |
352 | | ((p)[2] & 0x3F))) \ | |
eb41da4c | 353 | : string_char ((p), NULL, &actual_len)) |
0168c3d8 KH |
354 | |
355 | ||
b583cead | 356 | /* Like STRING_CHAR but advance P to the end of multibyte form. */ |
0168c3d8 KH |
357 | |
358 | #define STRING_CHAR_ADVANCE(p) \ | |
359 | (!((p)[0] & 0x80) \ | |
360 | ? *(p)++ \ | |
361 | : ! ((p)[0] & 0x20) \ | |
362 | ? ((p) += 2, \ | |
363 | ((((p)[-2] & 0x1F) << 6) \ | |
364 | | ((p)[-1] & 0x3F) \ | |
8f924df7 | 365 | | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \ |
0168c3d8 KH |
366 | : ! ((p)[0] & 0x10) \ |
367 | ? ((p) += 3, \ | |
368 | ((((p)[-3] & 0x0F) << 12) \ | |
369 | | (((p)[-2] & 0x3F) << 6) \ | |
370 | | ((p)[-1] & 0x3F))) \ | |
eb41da4c | 371 | : string_char ((p), &(p), NULL)) |
0168c3d8 KH |
372 | |
373 | ||
374 | /* Fetch the "next" character from Lisp string STRING at byte position | |
375 | BYTEIDX, character position CHARIDX. Store it into OUTPUT. | |
376 | ||
377 | All the args must be side-effect-free. | |
378 | BYTEIDX and CHARIDX must be lvalues; | |
379 | we increment them past the character fetched. */ | |
380 | ||
381 | #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 382 | do \ |
0168c3d8 KH |
383 | { \ |
384 | CHARIDX++; \ | |
385 | if (STRING_MULTIBYTE (STRING)) \ | |
386 | { \ | |
f1c99628 | 387 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
0168c3d8 KH |
388 | int len; \ |
389 | \ | |
390 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
391 | BYTEIDX += len; \ | |
392 | } \ | |
393 | else \ | |
f1c99628 SM |
394 | { \ |
395 | OUTPUT = SREF (STRING, BYTEIDX); \ | |
396 | BYTEIDX++; \ | |
397 | } \ | |
0168c3d8 | 398 | } \ |
a1c2ac9a | 399 | while (0) |
0168c3d8 | 400 | |
b583cead KH |
401 | /* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve |
402 | if STRING is unibyte. */ | |
43c47483 KH |
403 | |
404 | #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 405 | do \ |
43c47483 KH |
406 | { \ |
407 | CHARIDX++; \ | |
408 | if (STRING_MULTIBYTE (STRING)) \ | |
409 | { \ | |
f1c99628 | 410 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
43c47483 KH |
411 | int len; \ |
412 | \ | |
413 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
414 | BYTEIDX += len; \ | |
415 | } \ | |
416 | else \ | |
417 | { \ | |
f1c99628 SM |
418 | OUTPUT = SREF (STRING, BYTEIDX); \ |
419 | BYTEIDX++; \ | |
43c47483 KH |
420 | MAKE_CHAR_MULTIBYTE (OUTPUT); \ |
421 | } \ | |
422 | } \ | |
a1c2ac9a | 423 | while (0) |
43c47483 | 424 | |
0168c3d8 KH |
425 | |
426 | /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */ | |
427 | ||
428 | #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 429 | do \ |
0168c3d8 | 430 | { \ |
f1c99628 | 431 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
0168c3d8 KH |
432 | int len; \ |
433 | \ | |
434 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
435 | BYTEIDX += len; \ | |
436 | CHARIDX++; \ | |
437 | } \ | |
a1c2ac9a | 438 | while (0) |
0168c3d8 KH |
439 | |
440 | ||
441 | /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current | |
442 | buffer. */ | |
443 | ||
444 | #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 445 | do \ |
0168c3d8 KH |
446 | { \ |
447 | CHARIDX++; \ | |
448 | if (!NILP (current_buffer->enable_multibyte_characters)) \ | |
449 | { \ | |
450 | unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \ | |
451 | int len; \ | |
452 | \ | |
453 | OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
454 | BYTEIDX += len; \ | |
455 | } \ | |
456 | else \ | |
457 | { \ | |
458 | OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \ | |
459 | BYTEIDX++; \ | |
460 | } \ | |
461 | } \ | |
a1c2ac9a | 462 | while (0) |
0168c3d8 KH |
463 | |
464 | ||
b583cead | 465 | /* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */ |
0168c3d8 KH |
466 | |
467 | #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 468 | do \ |
0168c3d8 KH |
469 | { \ |
470 | unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \ | |
471 | int len; \ | |
472 | \ | |
473 | OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
474 | BYTEIDX += len; \ | |
475 | CHARIDX++; \ | |
476 | } \ | |
a1c2ac9a | 477 | while (0) |
0168c3d8 KH |
478 | |
479 | ||
480 | /* Increase the buffer byte position POS_BYTE of the current buffer to | |
481 | the next character boundary. No range checking of POS. */ | |
482 | ||
483 | #define INC_POS(pos_byte) \ | |
484 | do { \ | |
485 | unsigned char *p = BYTE_POS_ADDR (pos_byte); \ | |
486 | pos_byte += BYTES_BY_CHAR_HEAD (*p); \ | |
487 | } while (0) | |
488 | ||
489 | ||
490 | /* Decrease the buffer byte position POS_BYTE of the current buffer to | |
491 | the previous character boundary. No range checking of POS. */ | |
492 | ||
493 | #define DEC_POS(pos_byte) \ | |
494 | do { \ | |
495 | unsigned char *p; \ | |
496 | \ | |
497 | pos_byte--; \ | |
498 | if (pos_byte < GPT_BYTE) \ | |
f1c99628 | 499 | p = BEG_ADDR + pos_byte - BEG_BYTE; \ |
0168c3d8 | 500 | else \ |
f1c99628 | 501 | p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\ |
0168c3d8 KH |
502 | while (!CHAR_HEAD_P (*p)) \ |
503 | { \ | |
504 | p--; \ | |
505 | pos_byte--; \ | |
506 | } \ | |
507 | } while (0) | |
508 | ||
509 | /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */ | |
510 | ||
511 | #define INC_BOTH(charpos, bytepos) \ | |
512 | do \ | |
513 | { \ | |
514 | (charpos)++; \ | |
515 | if (NILP (current_buffer->enable_multibyte_characters)) \ | |
516 | (bytepos)++; \ | |
517 | else \ | |
518 | INC_POS ((bytepos)); \ | |
519 | } \ | |
520 | while (0) | |
521 | ||
522 | ||
523 | /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */ | |
524 | ||
525 | #define DEC_BOTH(charpos, bytepos) \ | |
526 | do \ | |
527 | { \ | |
528 | (charpos)--; \ | |
529 | if (NILP (current_buffer->enable_multibyte_characters)) \ | |
530 | (bytepos)--; \ | |
531 | else \ | |
532 | DEC_POS ((bytepos)); \ | |
533 | } \ | |
534 | while (0) | |
535 | ||
536 | ||
537 | /* Increase the buffer byte position POS_BYTE of the current buffer to | |
538 | the next character boundary. This macro relies on the fact that | |
539 | *GPT_ADDR and *Z_ADDR are always accessible and the values are | |
540 | '\0'. No range checking of POS_BYTE. */ | |
541 | ||
542 | #define BUF_INC_POS(buf, pos_byte) \ | |
543 | do { \ | |
544 | unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \ | |
545 | pos_byte += BYTES_BY_CHAR_HEAD (*p); \ | |
546 | } while (0) | |
547 | ||
548 | ||
549 | /* Decrease the buffer byte position POS_BYTE of the current buffer to | |
550 | the previous character boundary. No range checking of POS_BYTE. */ | |
551 | ||
552 | #define BUF_DEC_POS(buf, pos_byte) \ | |
553 | do { \ | |
554 | unsigned char *p; \ | |
555 | pos_byte--; \ | |
556 | if (pos_byte < BUF_GPT_BYTE (buf)) \ | |
f1c99628 | 557 | p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \ |
0168c3d8 | 558 | else \ |
f1c99628 | 559 | p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\ |
0168c3d8 KH |
560 | while (!CHAR_HEAD_P (*p)) \ |
561 | { \ | |
562 | p--; \ | |
563 | pos_byte--; \ | |
564 | } \ | |
565 | } while (0) | |
566 | ||
567 | ||
b583cead KH |
568 | /* If C is a character to be unified with a Unicode character, return |
569 | the unified Unicode character. */ | |
570 | ||
fc9d9d2a | 571 | #define MAYBE_UNIFY_CHAR(c) \ |
eb41da4c KH |
572 | if (c > MAX_UNICODE_CHAR \ |
573 | && CHAR_TABLE_P (Vchar_unify_table)) \ | |
fc9d9d2a KH |
574 | { \ |
575 | Lisp_Object val; \ | |
576 | int unified; \ | |
577 | \ | |
578 | val = CHAR_TABLE_REF (Vchar_unify_table, c); \ | |
579 | if (! NILP (val)) \ | |
580 | { \ | |
581 | if (SYMBOLP (val)) \ | |
582 | { \ | |
b5c7dbe6 | 583 | Funify_charset (val, Qnil, Qnil); \ |
fc9d9d2a KH |
584 | val = CHAR_TABLE_REF (Vchar_unify_table, c); \ |
585 | } \ | |
586 | if ((unified = XINT (val)) >= 0) \ | |
587 | c = unified; \ | |
588 | } \ | |
589 | } \ | |
0168c3d8 KH |
590 | else |
591 | ||
fc9d9d2a | 592 | |
0168c3d8 KH |
593 | /* Return the width of ASCII character C. The width is measured by |
594 | how many columns occupied on the screen when displayed in the | |
595 | current buffer. */ | |
596 | ||
597 | #define ASCII_CHAR_WIDTH(c) \ | |
598 | (c < 0x20 \ | |
599 | ? (c == '\t' \ | |
600 | ? XFASTINT (current_buffer->tab_width) \ | |
601 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ | |
602 | : (c < 0x7f \ | |
603 | ? 1 \ | |
604 | : ((NILP (current_buffer->ctl_arrow) ? 4 : 2)))) | |
605 | ||
606 | /* Return the width of character C. The width is measured by how many | |
607 | columns occupied on the screen when displayed in the current | |
608 | buffer. */ | |
609 | ||
610 | #define CHAR_WIDTH(c) \ | |
611 | (ASCII_CHAR_P (c) \ | |
612 | ? ASCII_CHAR_WIDTH (c) \ | |
613 | : XINT (CHAR_TABLE_REF (Vchar_width_table, c))) | |
614 | ||
eb41da4c | 615 | extern int char_resolve_modifier_mask P_ ((int)); |
f958a2fa | 616 | extern int char_string P_ ((unsigned, unsigned char *)); |
eb41da4c KH |
617 | extern int string_char P_ ((const unsigned char *, |
618 | const unsigned char **, int *)); | |
0168c3d8 KH |
619 | |
620 | extern int translate_char P_ ((Lisp_Object, int c)); | |
621 | extern int char_printable_p P_ ((int c)); | |
8f924df7 KH |
622 | extern void parse_str_as_multibyte P_ ((const unsigned char *, int, int *, |
623 | int *)); | |
0168c3d8 KH |
624 | extern int parse_str_to_multibyte P_ ((unsigned char *, int)); |
625 | extern int str_as_multibyte P_ ((unsigned char *, int, int, int *)); | |
626 | extern int str_to_multibyte P_ ((unsigned char *, int, int)); | |
627 | extern int str_as_unibyte P_ ((unsigned char *, int)); | |
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; | |
639 | ||
fac2bdc4 DL |
640 | extern Lisp_Object string_escape_byte8 P_ ((Lisp_Object)); |
641 | ||
0168c3d8 KH |
642 | /* Return a translation table of id number ID. */ |
643 | #define GET_TRANSLATION_TABLE(id) \ | |
644 | (XCDR(XVECTOR(Vtranslation_table_vector)->contents[(id)])) | |
645 | ||
646 | /* A char-table for characters which may invoke auto-filling. */ | |
647 | extern Lisp_Object Vauto_fill_chars; | |
648 | ||
e18ef64a | 649 | extern Lisp_Object Vchar_script_table; |
e0d6e5a5 | 650 | extern Lisp_Object Vscript_representative_chars; |
b5c7dbe6 | 651 | |
0168c3d8 KH |
652 | /* Copy LEN bytes from FROM to TO. This macro should be used only |
653 | when a caller knows that LEN is short and the obvious copy loop is | |
654 | faster than calling bcopy which has some overhead. Copying a | |
655 | multibyte sequence of a character is the typical case. */ | |
656 | ||
657 | #define BCOPY_SHORT(from, to, len) \ | |
658 | do { \ | |
659 | int i = len; \ | |
660 | unsigned char *from_p = from, *to_p = to; \ | |
661 | while (i--) *to_p++ = *from_p++; \ | |
662 | } while (0) | |
663 | ||
664 | #define DEFSYM(sym, name) \ | |
665 | do { (sym) = intern ((name)); staticpro (&(sym)); } while (0) | |
666 | ||
667 | #endif /* EMACS_CHARACTER_H */ | |
fbaf0946 MB |
668 | |
669 | /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac | |
670 | (do not change this comment) */ |