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 | |
43c47483 | 104 | /* If C is not ASCII, make it multibyte. It 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 | ||
b583cead KH |
111 | /* Return a Lisp character whose character code is C. It assumes C is |
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 | ||
f65c6d94 | 121 | /* Nonzero iff C is valid as a character code. GENERICP is not used |
885317d8 KH |
122 | now. */ |
123 | #define CHAR_VALID_P(c, genericp) ((unsigned) (c) <= MAX_CHAR) | |
0168c3d8 KH |
124 | |
125 | /* Check if Lisp object X is a character or not. */ | |
63db3c1b MB |
126 | #define CHECK_CHARACTER(x) \ |
127 | CHECK_TYPE (CHARACTERP (x), Qcharacterp, x) | |
0168c3d8 | 128 | |
8f924df7 KH |
129 | #define CHECK_CHARACTER_CAR(x) \ |
130 | do { \ | |
131 | Lisp_Object tmp = XCAR (x); \ | |
132 | CHECK_CHARACTER (tmp); \ | |
133 | XSETCAR ((x), tmp); \ | |
134 | } while (0) | |
135 | ||
136 | #define CHECK_CHARACTER_CDR(x) \ | |
137 | do { \ | |
138 | Lisp_Object tmp = XCDR (x); \ | |
139 | CHECK_CHARACTER (tmp); \ | |
140 | XSETCDR ((x), tmp); \ | |
141 | } while (0) | |
142 | ||
0168c3d8 KH |
143 | /* Nonzero iff C is an ASCII character. */ |
144 | #define ASCII_CHAR_P(c) ((unsigned) (c) < 0x80) | |
145 | ||
146 | /* Nonzero iff C is a character of code less than 0x100. */ | |
147 | #define SINGLE_BYTE_CHAR_P(c) ((unsigned) (c) < 0x100) | |
148 | ||
149 | /* Nonzero if character C has a printable glyph. */ | |
150 | #define CHAR_PRINTABLE_P(c) \ | |
151 | (((c) >= 32 && ((c) < 127) \ | |
152 | || ! NILP (CHAR_TABLE_REF (Vprintable_chars, (c))))) | |
153 | ||
885317d8 | 154 | /* Return byte length of multibyte form for character C. */ |
0168c3d8 KH |
155 | #define CHAR_BYTES(c) \ |
156 | ( (c) <= MAX_1_BYTE_CHAR ? 1 \ | |
157 | : (c) <= MAX_2_BYTE_CHAR ? 2 \ | |
158 | : (c) <= MAX_3_BYTE_CHAR ? 3 \ | |
159 | : (c) <= MAX_4_BYTE_CHAR ? 4 \ | |
160 | : (c) <= MAX_5_BYTE_CHAR ? 5 \ | |
161 | : 2) | |
162 | ||
43c47483 KH |
163 | |
164 | /* Return the leading code of multibyte form of C. */ | |
165 | #define CHAR_LEADING_CODE(c) \ | |
166 | ((c) <= MAX_1_BYTE_CHAR ? c \ | |
167 | : (c) <= MAX_2_BYTE_CHAR ? (0xC0 | ((c) >> 6)) \ | |
168 | : (c) <= MAX_3_BYTE_CHAR ? (0xE0 | ((c) >> 12)) \ | |
169 | : (c) <= MAX_4_BYTE_CHAR ? (0xF0 | ((c) >> 18)) \ | |
170 | : (c) <= MAX_5_BYTE_CHAR ? 0xF8 \ | |
171 | : (0xC0 | (((c) >> 6) & 0x01))) | |
172 | ||
173 | ||
885317d8 KH |
174 | /* Store multibyte form of the character C in P. The caller should |
175 | allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance. | |
176 | Returns the length of the multibyte form. */ | |
0168c3d8 KH |
177 | |
178 | #define CHAR_STRING(c, p) \ | |
179 | ((unsigned) (c) <= MAX_1_BYTE_CHAR \ | |
180 | ? ((p)[0] = (c), \ | |
181 | 1) \ | |
182 | : (unsigned) (c) <= MAX_2_BYTE_CHAR \ | |
183 | ? ((p)[0] = (0xC0 | ((c) >> 6)), \ | |
184 | (p)[1] = (0x80 | ((c) & 0x3F)), \ | |
185 | 2) \ | |
186 | : (unsigned) (c) <= MAX_3_BYTE_CHAR \ | |
187 | ? ((p)[0] = (0xE0 | ((c) >> 12)), \ | |
188 | (p)[1] = (0x80 | (((c) >> 6) & 0x3F)), \ | |
189 | (p)[2] = (0x80 | ((c) & 0x3F)), \ | |
190 | 3) \ | |
f958a2fa | 191 | : char_string ((unsigned) c, p)) |
0168c3d8 | 192 | |
eb41da4c KH |
193 | /* Store multibyte form of byte B in P. The caller should allocate at |
194 | least MAX_MULTIBYTE_LENGTH bytes area at P in advance. Returns the | |
195 | length of the multibyte form. */ | |
1106ea2b KH |
196 | |
197 | #define BYTE8_STRING(b, p) \ | |
198 | ((p)[0] = (0xC0 | (((b) >> 6) & 0x01)), \ | |
7f464917 | 199 | (p)[1] = (0x80 | ((b) & 0x3F)), \ |
1106ea2b KH |
200 | 2) |
201 | ||
0168c3d8 | 202 | |
885317d8 KH |
203 | /* Store multibyte form of the character C in P. The caller should |
204 | allocate at least MAX_MULTIBYTE_LENGTH bytes area at P in advance. | |
205 | And, advance P to the end of the multibyte form. */ | |
0168c3d8 | 206 | |
eb41da4c KH |
207 | #define CHAR_STRING_ADVANCE(c, p) \ |
208 | do { \ | |
209 | if ((c) <= MAX_1_BYTE_CHAR) \ | |
210 | *(p)++ = (c); \ | |
211 | else if ((c) <= MAX_2_BYTE_CHAR) \ | |
212 | *(p)++ = (0xC0 | ((c) >> 6)), \ | |
213 | *(p)++ = (0x80 | ((c) & 0x3F)); \ | |
214 | else if ((c) <= MAX_3_BYTE_CHAR) \ | |
215 | *(p)++ = (0xE0 | ((c) >> 12)), \ | |
216 | *(p)++ = (0x80 | (((c) >> 6) & 0x3F)), \ | |
217 | *(p)++ = (0x80 | ((c) & 0x3F)); \ | |
218 | else \ | |
219 | (p) += char_string ((c), (p)); \ | |
885317d8 | 220 | } while (0) |
0168c3d8 | 221 | |
eb41da4c | 222 | |
0168c3d8 KH |
223 | /* Nonzero iff BYTE starts a non-ASCII character in a multibyte |
224 | form. */ | |
225 | #define LEADING_CODE_P(byte) (((byte) & 0xC0) == 0xC0) | |
226 | ||
b5c7dbe6 KH |
227 | /* Nonzero iff BYTE is a trailing code of a non-ASCII character in a |
228 | multibyte form. */ | |
229 | #define TRAILING_CODE_P(byte) (((byte) & 0xC0) == 0x80) | |
230 | ||
885317d8 KH |
231 | /* Nonzero iff BYTE starts a character in a multibyte form. |
232 | This is equivalent to: | |
233 | (ASCII_BYTE_P (byte) || LEADING_CODE_P (byte)) */ | |
234 | #define CHAR_HEAD_P(byte) (((byte) & 0xC0) != 0x80) | |
235 | ||
0168c3d8 KH |
236 | /* Just kept for backward compatibility. This macro will be removed |
237 | in the future. */ | |
238 | #define BASE_LEADING_CODE_P LEADING_CODE_P | |
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 | /* Return the length of the multi-byte form at string STR of length | |
251 | LEN while assuming that STR points a valid multi-byte form. As | |
252 | this macro isn't necessary anymore, all callers will be changed to | |
253 | use BYTES_BY_CHAR_HEAD directly in the future. */ | |
254 | ||
255 | #define MULTIBYTE_FORM_LENGTH(str, len) \ | |
256 | BYTES_BY_CHAR_HEAD (*(str)) | |
257 | ||
258 | /* Parse multibyte string STR of length LENGTH and set BYTES to the | |
259 | byte length of a character at STR while assuming that STR points a | |
260 | valid multibyte form. As this macro isn't necessary anymore, all | |
261 | callers will be changed to use BYTES_BY_CHAR_HEAD directly in the | |
262 | future. */ | |
263 | ||
264 | #define PARSE_MULTIBYTE_SEQ(str, length, bytes) \ | |
265 | (bytes) = BYTES_BY_CHAR_HEAD (*(str)) | |
266 | ||
267 | /* The byte length of multibyte form at unibyte string P ending at | |
268 | PEND. If STR doesn't point a valid multibyte form, return 0. */ | |
269 | ||
270 | #define MULTIBYTE_LENGTH(p, pend) \ | |
271 | (p >= pend ? 0 \ | |
272 | : !((p)[0] & 0x80) ? 1 \ | |
273 | : ((p + 1 >= pend) || (((p)[1] & 0xC0) != 0x80)) ? 0 \ | |
274 | : ((p)[0] & 0xE0) == 0xC0 ? 2 \ | |
275 | : ((p + 2 >= pend) || (((p)[2] & 0xC0) != 0x80)) ? 0 \ | |
276 | : ((p)[0] & 0xF0) == 0xE0 ? 3 \ | |
277 | : ((p + 3 >= pend) || (((p)[3] & 0xC0) != 0x80)) ? 0 \ | |
278 | : ((p)[0] & 0xF8) == 0xF0 ? 4 \ | |
279 | : ((p + 4 >= pend) || (((p)[4] & 0xC0) != 0x80)) ? 0 \ | |
280 | : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \ | |
281 | : 0) | |
282 | ||
283 | ||
284 | /* Like MULTIBYTE_LENGTH but don't check the ending address. */ | |
285 | ||
286 | #define MULTIBYTE_LENGTH_NO_CHECK(p) \ | |
287 | (!((p)[0] & 0x80) ? 1 \ | |
288 | : ((p)[1] & 0xC0) != 0x80 ? 0 \ | |
289 | : ((p)[0] & 0xE0) == 0xC0 ? 2 \ | |
290 | : ((p)[2] & 0xC0) != 0x80 ? 0 \ | |
291 | : ((p)[0] & 0xF0) == 0xE0 ? 3 \ | |
292 | : ((p)[3] & 0xC0) != 0x80 ? 0 \ | |
293 | : ((p)[0] & 0xF8) == 0xF0 ? 4 \ | |
294 | : ((p)[4] & 0xC0) != 0x80 ? 0 \ | |
295 | : (p)[0] == 0xF8 && ((p)[1] & 0xF0) == 0x80 ? 5 \ | |
296 | : 0) | |
297 | ||
8f924df7 KH |
298 | /* If P is before LIMIT, advance P to the next character boundary. It |
299 | assumes that P is already at a character boundary of the sane | |
300 | mulitbyte form whose end address is LIMIT. */ | |
301 | ||
302 | #define NEXT_CHAR_BOUNDARY(p, limit) \ | |
303 | do { \ | |
304 | if ((p) < (limit)) \ | |
305 | (p) += BYTES_BY_CHAR_HEAD (*(p)); \ | |
306 | } while (0) | |
307 | ||
308 | ||
309 | /* If P is after LIMIT, advance P to the previous character boundary. | |
310 | It assumes that P is already at a character boundary of the sane | |
311 | mulitbyte form whose beginning address is LIMIT. */ | |
312 | ||
313 | #define PREV_CHAR_BOUNDARY(p, limit) \ | |
314 | do { \ | |
315 | if ((p) > (limit)) \ | |
316 | { \ | |
317 | const unsigned char *p0 = (p); \ | |
318 | do { \ | |
319 | p0--; \ | |
320 | } while (p0 >= limit && ! CHAR_HEAD_P (*p0)); \ | |
321 | (p) = (BYTES_BY_CHAR_HEAD (*p0) == (p) - p0) ? p0 : (p) - 1; \ | |
322 | } \ | |
323 | } while (0) | |
0168c3d8 KH |
324 | |
325 | /* Return the character code of character whose multibyte form is at | |
326 | P. The argument LEN is ignored. It will be removed in the | |
327 | future. */ | |
328 | ||
329 | #define STRING_CHAR(p, len) \ | |
330 | (!((p)[0] & 0x80) \ | |
331 | ? (p)[0] \ | |
332 | : ! ((p)[0] & 0x20) \ | |
333 | ? (((((p)[0] & 0x1F) << 6) \ | |
334 | | ((p)[1] & 0x3F)) \ | |
335 | + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0)) \ | |
336 | : ! ((p)[0] & 0x10) \ | |
337 | ? ((((p)[0] & 0x0F) << 12) \ | |
338 | | (((p)[1] & 0x3F) << 6) \ | |
339 | | ((p)[2] & 0x3F)) \ | |
eb41da4c | 340 | : string_char ((p), NULL, NULL)) |
0168c3d8 KH |
341 | |
342 | ||
343 | /* Like STRING_CHAR but set ACTUAL_LEN to the length of multibyte | |
344 | form. The argument LEN is ignored. It will be removed in the | |
345 | future. */ | |
346 | ||
347 | #define STRING_CHAR_AND_LENGTH(p, len, actual_len) \ | |
348 | (!((p)[0] & 0x80) \ | |
349 | ? ((actual_len) = 1, (p)[0]) \ | |
350 | : ! ((p)[0] & 0x20) \ | |
351 | ? ((actual_len) = 2, \ | |
352 | (((((p)[0] & 0x1F) << 6) \ | |
353 | | ((p)[1] & 0x3F)) \ | |
354 | + (((unsigned char) (p)[0]) < 0xC2 ? 0x3FFF80 : 0))) \ | |
355 | : ! ((p)[0] & 0x10) \ | |
356 | ? ((actual_len) = 3, \ | |
357 | ((((p)[0] & 0x0F) << 12) \ | |
358 | | (((p)[1] & 0x3F) << 6) \ | |
359 | | ((p)[2] & 0x3F))) \ | |
eb41da4c | 360 | : string_char ((p), NULL, &actual_len)) |
0168c3d8 KH |
361 | |
362 | ||
b583cead | 363 | /* Like STRING_CHAR but advance P to the end of multibyte form. */ |
0168c3d8 KH |
364 | |
365 | #define STRING_CHAR_ADVANCE(p) \ | |
366 | (!((p)[0] & 0x80) \ | |
367 | ? *(p)++ \ | |
368 | : ! ((p)[0] & 0x20) \ | |
369 | ? ((p) += 2, \ | |
370 | ((((p)[-2] & 0x1F) << 6) \ | |
371 | | ((p)[-1] & 0x3F) \ | |
8f924df7 | 372 | | ((unsigned char) ((p)[-2]) < 0xC2 ? 0x3FFF80 : 0))) \ |
0168c3d8 KH |
373 | : ! ((p)[0] & 0x10) \ |
374 | ? ((p) += 3, \ | |
375 | ((((p)[-3] & 0x0F) << 12) \ | |
376 | | (((p)[-2] & 0x3F) << 6) \ | |
377 | | ((p)[-1] & 0x3F))) \ | |
eb41da4c | 378 | : string_char ((p), &(p), NULL)) |
0168c3d8 KH |
379 | |
380 | ||
381 | /* Fetch the "next" character from Lisp string STRING at byte position | |
382 | BYTEIDX, character position CHARIDX. Store it into OUTPUT. | |
383 | ||
384 | All the args must be side-effect-free. | |
385 | BYTEIDX and CHARIDX must be lvalues; | |
386 | we increment them past the character fetched. */ | |
387 | ||
388 | #define FETCH_STRING_CHAR_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 389 | do \ |
0168c3d8 KH |
390 | { \ |
391 | CHARIDX++; \ | |
392 | if (STRING_MULTIBYTE (STRING)) \ | |
393 | { \ | |
f1c99628 | 394 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
0168c3d8 KH |
395 | int len; \ |
396 | \ | |
397 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
398 | BYTEIDX += len; \ | |
399 | } \ | |
400 | else \ | |
f1c99628 SM |
401 | { \ |
402 | OUTPUT = SREF (STRING, BYTEIDX); \ | |
403 | BYTEIDX++; \ | |
404 | } \ | |
0168c3d8 | 405 | } \ |
a1c2ac9a | 406 | while (0) |
0168c3d8 | 407 | |
b583cead KH |
408 | /* Like FETCH_STRING_CHAR_ADVANCE but return a multibyte character eve |
409 | if STRING is unibyte. */ | |
43c47483 KH |
410 | |
411 | #define FETCH_STRING_CHAR_AS_MULTIBYTE_ADVANCE(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 412 | do \ |
43c47483 KH |
413 | { \ |
414 | CHARIDX++; \ | |
415 | if (STRING_MULTIBYTE (STRING)) \ | |
416 | { \ | |
f1c99628 | 417 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
43c47483 KH |
418 | int len; \ |
419 | \ | |
420 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
421 | BYTEIDX += len; \ | |
422 | } \ | |
423 | else \ | |
424 | { \ | |
f1c99628 SM |
425 | OUTPUT = SREF (STRING, BYTEIDX); \ |
426 | BYTEIDX++; \ | |
43c47483 KH |
427 | MAKE_CHAR_MULTIBYTE (OUTPUT); \ |
428 | } \ | |
429 | } \ | |
a1c2ac9a | 430 | while (0) |
43c47483 | 431 | |
0168c3d8 KH |
432 | |
433 | /* Like FETCH_STRING_CHAR_ADVANCE but assumes STRING is multibyte. */ | |
434 | ||
435 | #define FETCH_STRING_CHAR_ADVANCE_NO_CHECK(OUTPUT, STRING, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 436 | do \ |
0168c3d8 | 437 | { \ |
f1c99628 | 438 | unsigned char *ptr = &SDATA (STRING)[BYTEIDX]; \ |
0168c3d8 KH |
439 | int len; \ |
440 | \ | |
441 | OUTPUT = STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
442 | BYTEIDX += len; \ | |
443 | CHARIDX++; \ | |
444 | } \ | |
a1c2ac9a | 445 | while (0) |
0168c3d8 KH |
446 | |
447 | ||
448 | /* Like FETCH_STRING_CHAR_ADVANCE but fetch character from the current | |
449 | buffer. */ | |
450 | ||
451 | #define FETCH_CHAR_ADVANCE(OUTPUT, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 452 | do \ |
0168c3d8 KH |
453 | { \ |
454 | CHARIDX++; \ | |
455 | if (!NILP (current_buffer->enable_multibyte_characters)) \ | |
456 | { \ | |
457 | unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \ | |
458 | int len; \ | |
459 | \ | |
460 | OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
461 | BYTEIDX += len; \ | |
462 | } \ | |
463 | else \ | |
464 | { \ | |
465 | OUTPUT = *(BYTE_POS_ADDR (BYTEIDX)); \ | |
466 | BYTEIDX++; \ | |
467 | } \ | |
468 | } \ | |
a1c2ac9a | 469 | while (0) |
0168c3d8 KH |
470 | |
471 | ||
b583cead | 472 | /* Like FETCH_CHAR_ADVANCE but assumes the current buffer is multibyte. */ |
0168c3d8 KH |
473 | |
474 | #define FETCH_CHAR_ADVANCE_NO_CHECK(OUTPUT, CHARIDX, BYTEIDX) \ | |
a1c2ac9a | 475 | do \ |
0168c3d8 KH |
476 | { \ |
477 | unsigned char *ptr = BYTE_POS_ADDR (BYTEIDX); \ | |
478 | int len; \ | |
479 | \ | |
480 | OUTPUT= STRING_CHAR_AND_LENGTH (ptr, 0, len); \ | |
481 | BYTEIDX += len; \ | |
482 | CHARIDX++; \ | |
483 | } \ | |
a1c2ac9a | 484 | while (0) |
0168c3d8 KH |
485 | |
486 | ||
487 | /* Increase the buffer byte position POS_BYTE of the current buffer to | |
488 | the next character boundary. No range checking of POS. */ | |
489 | ||
490 | #define INC_POS(pos_byte) \ | |
491 | do { \ | |
492 | unsigned char *p = BYTE_POS_ADDR (pos_byte); \ | |
493 | pos_byte += BYTES_BY_CHAR_HEAD (*p); \ | |
494 | } while (0) | |
495 | ||
496 | ||
497 | /* Decrease the buffer byte position POS_BYTE of the current buffer to | |
498 | the previous character boundary. No range checking of POS. */ | |
499 | ||
500 | #define DEC_POS(pos_byte) \ | |
501 | do { \ | |
502 | unsigned char *p; \ | |
503 | \ | |
504 | pos_byte--; \ | |
505 | if (pos_byte < GPT_BYTE) \ | |
f1c99628 | 506 | p = BEG_ADDR + pos_byte - BEG_BYTE; \ |
0168c3d8 | 507 | else \ |
f1c99628 | 508 | p = BEG_ADDR + GAP_SIZE + pos_byte - BEG_BYTE;\ |
0168c3d8 KH |
509 | while (!CHAR_HEAD_P (*p)) \ |
510 | { \ | |
511 | p--; \ | |
512 | pos_byte--; \ | |
513 | } \ | |
514 | } while (0) | |
515 | ||
516 | /* Increment both CHARPOS and BYTEPOS, each in the appropriate way. */ | |
517 | ||
518 | #define INC_BOTH(charpos, bytepos) \ | |
519 | do \ | |
520 | { \ | |
521 | (charpos)++; \ | |
522 | if (NILP (current_buffer->enable_multibyte_characters)) \ | |
523 | (bytepos)++; \ | |
524 | else \ | |
525 | INC_POS ((bytepos)); \ | |
526 | } \ | |
527 | while (0) | |
528 | ||
529 | ||
530 | /* Decrement both CHARPOS and BYTEPOS, each in the appropriate way. */ | |
531 | ||
532 | #define DEC_BOTH(charpos, bytepos) \ | |
533 | do \ | |
534 | { \ | |
535 | (charpos)--; \ | |
536 | if (NILP (current_buffer->enable_multibyte_characters)) \ | |
537 | (bytepos)--; \ | |
538 | else \ | |
539 | DEC_POS ((bytepos)); \ | |
540 | } \ | |
541 | while (0) | |
542 | ||
543 | ||
544 | /* Increase the buffer byte position POS_BYTE of the current buffer to | |
545 | the next character boundary. This macro relies on the fact that | |
546 | *GPT_ADDR and *Z_ADDR are always accessible and the values are | |
547 | '\0'. No range checking of POS_BYTE. */ | |
548 | ||
549 | #define BUF_INC_POS(buf, pos_byte) \ | |
550 | do { \ | |
551 | unsigned char *p = BUF_BYTE_ADDRESS (buf, pos_byte); \ | |
552 | pos_byte += BYTES_BY_CHAR_HEAD (*p); \ | |
553 | } while (0) | |
554 | ||
555 | ||
556 | /* Decrease the buffer byte position POS_BYTE of the current buffer to | |
557 | the previous character boundary. No range checking of POS_BYTE. */ | |
558 | ||
559 | #define BUF_DEC_POS(buf, pos_byte) \ | |
560 | do { \ | |
561 | unsigned char *p; \ | |
562 | pos_byte--; \ | |
563 | if (pos_byte < BUF_GPT_BYTE (buf)) \ | |
f1c99628 | 564 | p = BUF_BEG_ADDR (buf) + pos_byte - BEG_BYTE; \ |
0168c3d8 | 565 | else \ |
f1c99628 | 566 | p = BUF_BEG_ADDR (buf) + BUF_GAP_SIZE (buf) + pos_byte - BEG_BYTE;\ |
0168c3d8 KH |
567 | while (!CHAR_HEAD_P (*p)) \ |
568 | { \ | |
569 | p--; \ | |
570 | pos_byte--; \ | |
571 | } \ | |
572 | } while (0) | |
573 | ||
574 | ||
b583cead KH |
575 | /* If C is a character to be unified with a Unicode character, return |
576 | the unified Unicode character. */ | |
577 | ||
fc9d9d2a | 578 | #define MAYBE_UNIFY_CHAR(c) \ |
eb41da4c KH |
579 | if (c > MAX_UNICODE_CHAR \ |
580 | && CHAR_TABLE_P (Vchar_unify_table)) \ | |
fc9d9d2a KH |
581 | { \ |
582 | Lisp_Object val; \ | |
583 | int unified; \ | |
584 | \ | |
585 | val = CHAR_TABLE_REF (Vchar_unify_table, c); \ | |
586 | if (! NILP (val)) \ | |
587 | { \ | |
588 | if (SYMBOLP (val)) \ | |
589 | { \ | |
b5c7dbe6 | 590 | Funify_charset (val, Qnil, Qnil); \ |
fc9d9d2a KH |
591 | val = CHAR_TABLE_REF (Vchar_unify_table, c); \ |
592 | } \ | |
593 | if ((unified = XINT (val)) >= 0) \ | |
594 | c = unified; \ | |
595 | } \ | |
596 | } \ | |
0168c3d8 KH |
597 | else |
598 | ||
fc9d9d2a | 599 | |
0168c3d8 KH |
600 | /* Return the width of ASCII character C. The width is measured by |
601 | how many columns occupied on the screen when displayed in the | |
602 | current buffer. */ | |
603 | ||
604 | #define ASCII_CHAR_WIDTH(c) \ | |
605 | (c < 0x20 \ | |
606 | ? (c == '\t' \ | |
607 | ? XFASTINT (current_buffer->tab_width) \ | |
608 | : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ | |
609 | : (c < 0x7f \ | |
610 | ? 1 \ | |
611 | : ((NILP (current_buffer->ctl_arrow) ? 4 : 2)))) | |
612 | ||
613 | /* Return the width of character C. The width is measured by how many | |
614 | columns occupied on the screen when displayed in the current | |
615 | buffer. */ | |
616 | ||
617 | #define CHAR_WIDTH(c) \ | |
618 | (ASCII_CHAR_P (c) \ | |
619 | ? ASCII_CHAR_WIDTH (c) \ | |
620 | : XINT (CHAR_TABLE_REF (Vchar_width_table, c))) | |
621 | ||
eb41da4c | 622 | extern int char_resolve_modifier_mask P_ ((int)); |
f958a2fa | 623 | extern int char_string P_ ((unsigned, unsigned char *)); |
eb41da4c KH |
624 | extern int string_char P_ ((const unsigned char *, |
625 | const unsigned char **, int *)); | |
0168c3d8 KH |
626 | |
627 | extern int translate_char P_ ((Lisp_Object, int c)); | |
628 | extern int char_printable_p P_ ((int c)); | |
8f924df7 KH |
629 | extern void parse_str_as_multibyte P_ ((const unsigned char *, int, int *, |
630 | int *)); | |
0168c3d8 KH |
631 | extern int parse_str_to_multibyte P_ ((unsigned char *, int)); |
632 | extern int str_as_multibyte P_ ((unsigned char *, int, int, int *)); | |
633 | extern int str_to_multibyte P_ ((unsigned char *, int, int)); | |
634 | extern int str_as_unibyte P_ ((unsigned char *, int)); | |
0460afe5 KH |
635 | extern EMACS_INT str_to_unibyte P_ ((const unsigned char *, unsigned char *, |
636 | EMACS_INT, int)); | |
0168c3d8 | 637 | extern int strwidth P_ ((unsigned char *, int)); |
8f924df7 | 638 | extern int c_string_width P_ ((const unsigned char *, int, int, int *, int *)); |
0168c3d8 KH |
639 | extern int lisp_string_width P_ ((Lisp_Object, int, int *, int *)); |
640 | ||
641 | extern Lisp_Object Vprintable_chars; | |
642 | ||
643 | extern Lisp_Object Qcharacterp, Qauto_fill_chars; | |
644 | extern Lisp_Object Vtranslation_table_vector; | |
645 | extern Lisp_Object Vchar_width_table; | |
646 | extern Lisp_Object Vchar_direction_table; | |
647 | extern Lisp_Object Vchar_unify_table; | |
a48a6418 | 648 | extern Lisp_Object Vunicode_category_table; |
0168c3d8 | 649 | |
fac2bdc4 DL |
650 | extern 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. */ | |
657 | extern Lisp_Object Vauto_fill_chars; | |
658 | ||
e18ef64a | 659 | extern Lisp_Object Vchar_script_table; |
e0d6e5a5 | 660 | extern 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) \ | |
675 | do { (sym) = intern ((name)); staticpro (&(sym)); } while (0) | |
676 | ||
677 | #endif /* EMACS_CHARACTER_H */ | |
fbaf0946 MB |
678 | |
679 | /* arch-tag: 4ef86004-2eff-4073-8cea-cfcbcf7188ac | |
680 | (do not change this comment) */ |