Mercurial > emacs
annotate src/charset.c @ 17770:0d0e69cee17d
Make functions setup-LANGUAGE-environment
interactive and add new functions describe-LANGUAGE-support for
all LANGUAGEs supported. Remove resisterations of input methods
which use the function encoded-kbd-select-terminal.
Bug fixed in making coding system iso-2022-kr.
author | Kenichi Handa <handa@m17n.org> |
---|---|
date | Mon, 12 May 1997 07:00:13 +0000 |
parents | 9d39361ce928 |
children | 76c6740bd736 |
rev | line source |
---|---|
17052 | 1 /* Multilingual characters handler. |
2 Ver.1.0 | |
3 Copyright (C) 1995 Free Software Foundation, Inc. | |
4 Copyright (C) 1995 Electrotechnical Laboratory, JAPAN. | |
5 | |
17071 | 6 This file is part of GNU Emacs. |
7 | |
8 GNU Emacs is free software; you can redistribute it and/or modify | |
9 it under the terms of the GNU General Public License as published by | |
10 the Free Software Foundation; either version 2, or (at your option) | |
11 any later version. | |
17052 | 12 |
17071 | 13 GNU Emacs is distributed in the hope that it will be useful, |
14 but WITHOUT ANY WARRANTY; without even the implied warranty of | |
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
16 GNU General Public License for more details. | |
17052 | 17 |
17071 | 18 You should have received a copy of the GNU General Public License |
19 along with GNU Emacs; see the file COPYING. If not, write to | |
20 the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
21 Boston, MA 02111-1307, USA. */ | |
17052 | 22 |
23 /* At first, see the document in `charset.h' to understand the code in | |
24 this file. */ | |
25 | |
26 #include <stdio.h> | |
27 | |
28 #ifdef emacs | |
29 | |
30 #include <sys/types.h> | |
31 #include <config.h> | |
32 #include "lisp.h" | |
33 #include "buffer.h" | |
34 #include "charset.h" | |
35 #include "coding.h" | |
36 | |
37 #else /* not emacs */ | |
38 | |
39 #include "mulelib.h" | |
40 | |
41 #endif /* emacs */ | |
42 | |
43 Lisp_Object Qcharset, Qascii, Qcomposition; | |
44 | |
45 /* Declaration of special leading-codes. */ | |
46 int leading_code_composition; /* for composite characters */ | |
47 int leading_code_private_11; /* for private DIMENSION1 of 1-column */ | |
48 int leading_code_private_12; /* for private DIMENSION1 of 2-column */ | |
49 int leading_code_private_21; /* for private DIMENSION2 of 1-column */ | |
50 int leading_code_private_22; /* for private DIMENSION2 of 2-column */ | |
51 | |
52 /* Declaration of special charsets. */ | |
53 int charset_ascii; /* ASCII */ | |
54 int charset_composition; /* for a composite character */ | |
55 int charset_latin_iso8859_1; /* ISO8859-1 (Latin-1) */ | |
56 int charset_jisx0208_1978; /* JISX0208.1978 (Japanese Kanji old set) */ | |
57 int charset_jisx0208; /* JISX0208.1983 (Japanese Kanji) */ | |
58 int charset_katakana_jisx0201; /* JISX0201.Kana (Japanese Katakana) */ | |
59 int charset_latin_jisx0201; /* JISX0201.Roman (Japanese Roman) */ | |
60 int charset_big5_1; /* Big5 Level 1 (Chinese Traditional) */ | |
61 int charset_big5_2; /* Big5 Level 2 (Chinese Traditional) */ | |
62 | |
63 Lisp_Object Qcharset_table; | |
64 | |
65 /* A char-table containing information of each character set. */ | |
66 Lisp_Object Vcharset_table; | |
67 | |
68 /* A vector of charset symbol indexed by charset-id. This is used | |
69 only for returning charset symbol from C functions. */ | |
70 Lisp_Object Vcharset_symbol_table; | |
71 | |
72 /* A list of charset symbols ever defined. */ | |
73 Lisp_Object Vcharset_list; | |
74 | |
75 /* Tables used by macros BYTES_BY_CHAR_HEAD and WIDTH_BY_CHAR_HEAD. */ | |
76 int bytes_by_char_head[256]; | |
77 int width_by_char_head[256]; | |
78 | |
79 /* Mapping table from ISO2022's charset (specified by DIMENSION, | |
80 CHARS, and FINAL-CHAR) to Emacs' charset. */ | |
81 int iso_charset_table[2][2][128]; | |
82 | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
83 /* Table of pointers to the structure `cmpchar_info' indexed by |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
84 CMPCHAR-ID. */ |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
85 struct cmpchar_info **cmpchar_table; |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
86 /* The current size of `cmpchar_table'. */ |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
87 static int cmpchar_table_size; |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
88 /* Number of the current composite characters. */ |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
89 int n_cmpchars; |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
90 |
17052 | 91 /* Variables used locally in the macro FETCH_MULTIBYTE_CHAR. */ |
92 unsigned char *_fetch_multibyte_char_p; | |
93 int _fetch_multibyte_char_len; | |
94 | |
95 /* Set STR a pointer to the multi-byte form of the character C. If C | |
96 is not a composite character, the multi-byte form is set in WORKBUF | |
97 and STR points WORKBUF. The caller should allocate at least 4-byte | |
98 area at WORKBUF in advance. Returns the length of the multi-byte | |
99 form. | |
100 | |
101 Use macro `CHAR_STRING (C, WORKBUF, STR)' instead of calling this | |
102 function directly if C can be an ASCII character. */ | |
103 | |
104 int | |
105 non_ascii_char_to_string (c, workbuf, str) | |
106 int c; | |
107 unsigned char *workbuf, **str; | |
108 { | |
17322
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
109 int charset, c1, c2; |
17052 | 110 |
111 if (COMPOSITE_CHAR_P (c)) | |
112 { | |
113 int cmpchar_id = COMPOSITE_CHAR_ID (c); | |
114 | |
115 if (cmpchar_id < n_cmpchars) | |
116 { | |
117 *str = cmpchar_table[cmpchar_id]->data; | |
118 return cmpchar_table[cmpchar_id]->len; | |
119 } | |
120 else | |
121 { | |
122 *str = workbuf; | |
123 return 0; | |
124 } | |
125 } | |
126 | |
127 SPLIT_NON_ASCII_CHAR (c, charset, c1, c2); | |
128 | |
129 *str = workbuf; | |
130 *workbuf++ = CHARSET_LEADING_CODE_BASE (charset); | |
131 if (*workbuf = CHARSET_LEADING_CODE_EXT (charset)) | |
132 workbuf++; | |
133 *workbuf++ = c1 | 0x80; | |
17322
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
134 if (c2 >= 0) |
17052 | 135 *workbuf++ = c2 | 0x80; |
136 | |
137 return (workbuf - *str); | |
138 } | |
139 | |
140 /* Return a non-ASCII character of which multi-byte form is at STR of | |
141 length LEN. If ACTUAL_LEN is not NULL, the actual length of the | |
142 character is set to the address ACTUAL_LEN. | |
143 | |
144 Use macro `STRING_CHAR (STR, LEN)' instead of calling this function | |
145 directly if STR can hold an ASCII character. */ | |
146 | |
147 string_to_non_ascii_char (str, len, actual_len) | |
148 unsigned char *str; | |
149 int len, *actual_len; | |
150 { | |
151 int charset; | |
152 unsigned char c1, c2; | |
153 register int c; | |
154 | |
155 if (SPLIT_STRING (str, len, charset, c1, c2) == CHARSET_ASCII) | |
156 { | |
157 if (actual_len) | |
158 *actual_len = 1; | |
159 return (int) *str; | |
160 } | |
161 | |
162 c = MAKE_NON_ASCII_CHAR (charset, c1, c2); | |
163 | |
164 if (actual_len) | |
165 *actual_len = (charset == CHARSET_COMPOSITION | |
166 ? cmpchar_table[COMPOSITE_CHAR_ID (c)]->len | |
167 : BYTES_BY_CHAR_HEAD (*str)); | |
168 return c; | |
169 } | |
170 | |
171 /* Return the length of the multi-byte form at string STR of length LEN. */ | |
172 int | |
173 multibyte_form_length (str, len) | |
174 unsigned char *str; | |
175 int len; | |
176 { | |
177 int charset; | |
178 unsigned char c1, c2; | |
179 register int c; | |
180 | |
181 if (SPLIT_STRING (str, len, charset, c1, c2) == CHARSET_ASCII) | |
182 return 1; | |
183 | |
184 return (charset == CHARSET_COMPOSITION | |
185 ? cmpchar_table[(c1 << 7) | c2]->len | |
186 : BYTES_BY_CHAR_HEAD (*str)); | |
187 } | |
188 | |
189 /* Check if string STR of length LEN contains valid multi-byte form of | |
190 a character. If valid, charset and position codes of the character | |
191 is set at *CHARSET, *C1, and *C2, and return 0. If not valid, | |
192 return -1. This should be used only in the macro SPLIT_STRING | |
193 which checks range of STR in advance. */ | |
194 | |
195 split_non_ascii_string (str, len, charset, c1, c2) | |
196 register unsigned char *str, *c1, *c2; | |
197 register int len, *charset; | |
198 { | |
199 register unsigned int cs = *str++; | |
200 | |
201 if (cs == LEADING_CODE_COMPOSITION) | |
202 { | |
203 int cmpchar_id = str_cmpchar_id (str - 1, len); | |
204 | |
205 if (cmpchar_id < 0) | |
206 return -1; | |
207 *charset = cs, *c1 = cmpchar_id >> 7, *c2 = cmpchar_id & 0x7F; | |
208 } | |
209 else if ((cs < LEADING_CODE_PRIVATE_11 || (cs = *str++) >= 0xA0) | |
210 && CHARSET_DEFINED_P (cs)) | |
211 { | |
212 *charset = cs; | |
213 if (*str < 0xA0) | |
214 return -1; | |
215 *c1 = (*str++) & 0x7F; | |
216 if (CHARSET_DIMENSION (cs) == 2) | |
217 { | |
218 if (*str < 0xA0) | |
219 return -1; | |
220 *c2 = (*str++) & 0x7F; | |
221 } | |
222 } | |
223 else | |
224 return -1; | |
225 return 0; | |
226 } | |
227 | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
228 /* Return a character unified with C (or a character made of CHARSET, |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
229 C1, and C2) in unification table TABLE. If no unification is found |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
230 in TABLE, return C. */ |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
231 unify_char (table, c, charset, c1, c2) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
232 Lisp_Object table; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
233 int c, charset, c1, c2; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
234 { |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
235 Lisp_Object ch; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
236 int alt_charset, alt_c1, alt_c2, dimension; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
237 |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
238 if (c < 0) c = MAKE_CHAR (charset, c1, c2); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
239 if (!CHAR_TABLE_P (table) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
240 || (ch = Faref (table, make_number (c)), !INTEGERP (ch)) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
241 || XINT (ch) < 0) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
242 return c; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
243 |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
244 SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
245 dimension = CHARSET_DIMENSION (alt_charset); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
246 if (dimension == 1 && alt_c1 > 0 || dimension == 2 && alt_c2 > 0) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
247 /* CH is not a generic character, just return it. */ |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
248 return XFASTINT (ch); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
249 |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
250 /* Since CH is a generic character, we must return a specific |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
251 charater which has the same position codes as C from CH. */ |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
252 if (charset < 0) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
253 SPLIT_CHAR (c, charset, c1, c2); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
254 if (dimension != CHARSET_DIMENSION (charset)) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
255 /* We can't make such a character because of dimension mismatch. */ |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
256 return c; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
257 if (!alt_c1) alt_c1 = c1; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
258 if (!alt_c2) alt_c2 = c2; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
259 return MAKE_CHAR (alt_charset, c1, c2); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
260 } |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
261 |
17052 | 262 /* Update the table Vcharset_table with the given arguments (see the |
263 document of `define-charset' for the meaning of each argument). | |
264 Several other table contents are also updated. The caller should | |
265 check the validity of CHARSET-ID and the remaining arguments in | |
266 advance. */ | |
267 | |
268 void | |
269 update_charset_table (charset_id, dimension, chars, width, direction, | |
270 iso_final_char, iso_graphic_plane, | |
271 short_name, long_name, description) | |
272 Lisp_Object charset_id, dimension, chars, width, direction; | |
273 Lisp_Object iso_final_char, iso_graphic_plane; | |
274 Lisp_Object short_name, long_name, description; | |
275 { | |
276 int charset = XINT (charset_id); | |
277 int bytes; | |
278 unsigned char leading_code_base, leading_code_ext; | |
279 | |
17322
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
280 if (NILP (CHARSET_TABLE_ENTRY (charset))) |
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
281 CHARSET_TABLE_ENTRY (charset) |
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
282 = Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil); |
17052 | 283 |
284 /* Get byte length of multibyte form, base leading-code, and | |
285 extended leading-code of the charset. See the comment under the | |
286 title "GENERAL NOTE on CHARACTER SET (CHARSET)" in charset.h. */ | |
287 bytes = XINT (dimension); | |
288 if (charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
289 { | |
290 /* Official charset, it doesn't have an extended leading-code. */ | |
291 if (charset != CHARSET_ASCII) | |
292 bytes += 1; /* For a base leading-code. */ | |
293 leading_code_base = charset; | |
294 leading_code_ext = 0; | |
295 } | |
296 else | |
297 { | |
298 /* Private charset. */ | |
299 bytes += 2; /* For base and extended leading-codes. */ | |
300 leading_code_base | |
301 = (charset < LEADING_CODE_EXT_12 | |
302 ? LEADING_CODE_PRIVATE_11 | |
303 : (charset < LEADING_CODE_EXT_21 | |
304 ? LEADING_CODE_PRIVATE_12 | |
305 : (charset < LEADING_CODE_EXT_22 | |
306 ? LEADING_CODE_PRIVATE_21 | |
307 : LEADING_CODE_PRIVATE_22))); | |
308 leading_code_ext = charset; | |
309 } | |
310 | |
311 CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id; | |
312 CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes); | |
313 CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension; | |
314 CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars; | |
315 CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width; | |
316 CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction; | |
317 CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX) | |
318 = make_number (leading_code_base); | |
319 CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX) | |
320 = make_number (leading_code_ext); | |
321 CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char; | |
322 CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX) | |
323 = iso_graphic_plane; | |
324 CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name; | |
325 CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name; | |
326 CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description; | |
327 CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil; | |
328 | |
329 { | |
330 /* If we have already defined a charset which has the same | |
331 DIMENSION, CHARS and ISO-FINAL-CHAR but the different | |
332 DIRECTION, we must update the entry REVERSE-CHARSET of both | |
333 charsets. If there's no such charset, the value of the entry | |
334 is set to nil. */ | |
335 int i; | |
336 | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
337 for (i = 0; i <= MAX_CHARSET; i++) |
17052 | 338 if (!NILP (CHARSET_TABLE_ENTRY (i))) |
339 { | |
340 if (CHARSET_DIMENSION (i) == XINT (dimension) | |
341 && CHARSET_CHARS (i) == XINT (chars) | |
342 && CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char) | |
343 && CHARSET_DIRECTION (i) != XINT (direction)) | |
344 { | |
345 CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
346 = make_number (i); | |
347 CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id; | |
348 break; | |
349 } | |
350 } | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
351 if (i > MAX_CHARSET) |
17052 | 352 /* No such a charset. */ |
353 CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX) | |
354 = make_number (-1); | |
355 } | |
356 | |
357 if (charset != CHARSET_ASCII | |
358 && charset < MIN_CHARSET_PRIVATE_DIMENSION1) | |
359 { | |
360 /* Update tables bytes_by_char_head and width_by_char_head. */ | |
361 bytes_by_char_head[leading_code_base] = bytes; | |
362 width_by_char_head[leading_code_base] = XINT (width); | |
363 | |
364 /* Update table emacs_code_class. */ | |
365 emacs_code_class[charset] = (bytes == 2 | |
366 ? EMACS_leading_code_2 | |
367 : (bytes == 3 | |
368 ? EMACS_leading_code_3 | |
369 : EMACS_leading_code_4)); | |
370 } | |
371 | |
372 /* Update table iso_charset_table. */ | |
373 if (ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0) | |
374 ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset; | |
375 } | |
376 | |
377 #ifdef emacs | |
378 | |
379 /* Return charset id of CHARSET_SYMBOL, or return -1 if CHARSET_SYMBOL | |
380 is invalid. */ | |
381 int | |
382 get_charset_id (charset_symbol) | |
383 Lisp_Object charset_symbol; | |
384 { | |
385 Lisp_Object val; | |
386 int charset; | |
387 | |
388 return ((SYMBOLP (charset_symbol) | |
389 && (val = Fget (charset_symbol, Qcharset), VECTORP (val)) | |
390 && (charset = XINT (XVECTOR (val)->contents[CHARSET_ID_IDX]), | |
391 CHARSET_VALID_P (charset))) | |
392 ? charset : -1); | |
393 } | |
394 | |
395 /* Return an identification number for a new private charset of | |
396 DIMENSION and WIDTH. If there's no more room for the new charset, | |
397 return 0. */ | |
398 Lisp_Object | |
399 get_new_private_charset_id (dimension, width) | |
400 int dimension, width; | |
401 { | |
402 int charset, from, to; | |
403 | |
404 if (dimension == 1) | |
405 { | |
406 if (width == 1) | |
407 from = LEADING_CODE_EXT_11, to = LEADING_CODE_EXT_12; | |
408 else | |
409 from = LEADING_CODE_EXT_12, to = LEADING_CODE_EXT_21; | |
410 } | |
411 else | |
412 { | |
413 if (width == 1) | |
414 from = LEADING_CODE_EXT_21, to = LEADING_CODE_EXT_22; | |
415 else | |
416 from = LEADING_CODE_EXT_22, to = LEADING_CODE_EXT_MAX - 1; | |
417 } | |
418 | |
419 for (charset = from; charset < to; charset++) | |
420 if (!CHARSET_DEFINED_P (charset)) break; | |
421 | |
422 return make_number (charset < to ? charset : 0); | |
423 } | |
424 | |
425 DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0, | |
426 "Define CHARSET-ID as the identification number of CHARSET with INFO-VECTOR.\n\ | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
427 If CHARSET-ID is nil, it is decided automatically, which means CHARSET is\n\ |
17052 | 428 treated as a private charset.\n\ |
429 INFO-VECTOR is a vector of the format:\n\ | |
430 [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE\n\ | |
431 SHORT-NAME LONG-NAME DESCRIPTION]\n\ | |
432 The meanings of each elements is as follows:\n\ | |
433 DIMENSION (integer) is the number of bytes to represent a character: 1 or 2.\n\ | |
434 CHARS (integer) is the number of characters in a dimension: 94 or 96.\n\ | |
435 WIDTH (integer) is the number of columns a character in the charset\n\ | |
436 occupies on the screen: one of 0, 1, and 2.\n\ | |
437 \n\ | |
438 DIRECTION (integer) is the rendering direction of characters in the\n\ | |
439 charset when rendering. If 0, render from right to left, else\n\ | |
440 render from left to right.\n\ | |
441 \n\ | |
442 ISO-FINAL-CHAR (character) is the final character of the\n\ | |
443 corresponding ISO 2022 charset.\n\ | |
444 \n\ | |
445 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked\n\ | |
446 while encoding to variants of ISO 2022 coding system, one of the\n\ | |
447 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).\n\ | |
448 \n\ | |
449 SHORT-NAME (string) is the short name to refer to the charset.\n\ | |
450 \n\ | |
451 LONG-NAME (string) is the long name to refer to the charset.\n\ | |
452 \n\ | |
453 DESCRIPTION (string) is the description string of the charset.") | |
454 (charset_id, charset_symbol, info_vector) | |
455 Lisp_Object charset_id, charset_symbol, info_vector; | |
456 { | |
457 Lisp_Object *vec; | |
458 | |
459 if (!NILP (charset_id)) | |
460 CHECK_NUMBER (charset_id, 0); | |
461 CHECK_SYMBOL (charset_symbol, 1); | |
462 CHECK_VECTOR (info_vector, 2); | |
463 | |
464 if (! NILP (charset_id)) | |
465 { | |
466 if (! CHARSET_VALID_P (XINT (charset_id))) | |
467 error ("Invalid CHARSET: %d", XINT (charset_id)); | |
468 else if (CHARSET_DEFINED_P (XINT (charset_id))) | |
469 error ("Already defined charset: %d", XINT (charset_id)); | |
470 } | |
471 | |
472 vec = XVECTOR (info_vector)->contents; | |
473 if (XVECTOR (info_vector)->size != 9 | |
474 || !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2) | |
475 || !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96) | |
476 || !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2) | |
477 || !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1) | |
478 || !INTEGERP (vec[4]) || !(XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~') | |
479 || !INTEGERP (vec[5]) || !(XINT (vec[5]) == 0 || XINT (vec[5]) == 1) | |
480 || !STRINGP (vec[6]) | |
481 || !STRINGP (vec[7]) | |
482 || !STRINGP (vec[8])) | |
483 error ("Invalid info-vector argument for defining charset %s", | |
484 XSYMBOL (charset_symbol)->name->data); | |
485 | |
486 if (NILP (charset_id)) | |
487 { | |
488 charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2])); | |
489 if (XINT (charset_id) == 0) | |
490 error ("There's no room for a new private charset %s", | |
491 XSYMBOL (charset_symbol)->name->data); | |
492 } | |
493 | |
494 update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3], | |
495 vec[4], vec[5], vec[6], vec[7], vec[8]); | |
17322
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
496 Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id))); |
17052 | 497 CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol; |
498 Vcharset_list = Fcons (charset_symbol, Vcharset_list); | |
499 return Qnil; | |
500 } | |
501 | |
502 DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset, | |
503 4, 4, 0, | |
504 "Declare a charset of DIMENSION, CHARS, FINAL-CHAR is the same as CHARSET.\n\ | |
505 CHARSET should be defined by `defined-charset' in advance.") | |
506 (dimension, chars, final_char, charset_symbol) | |
507 Lisp_Object dimension, chars, final_char, charset_symbol; | |
508 { | |
509 int charset; | |
510 | |
511 CHECK_NUMBER (dimension, 0); | |
512 CHECK_NUMBER (chars, 1); | |
513 CHECK_NUMBER (final_char, 2); | |
514 CHECK_SYMBOL (charset_symbol, 3); | |
515 | |
516 if (XINT (dimension) != 1 && XINT (dimension) != 2) | |
517 error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension)); | |
518 if (XINT (chars) != 94 && XINT (chars) != 96) | |
519 error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars)); | |
520 if (XINT (final_char) < '0' || XFASTINT (final_char) > '~') | |
521 error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars)); | |
522 if ((charset = get_charset_id (charset_symbol)) < 0) | |
523 error ("Invalid charset %s", XSYMBOL (charset_symbol)->name->data); | |
524 | |
525 ISO_CHARSET_TABLE (dimension, chars, final_char) = charset; | |
526 return Qnil; | |
527 } | |
528 | |
529 /* Return number of different charsets in STR of length LEN. In | |
530 addition, for each found charset N, CHARSETS[N] is set 1. The | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
531 caller should allocate CHARSETS (MAX_CHARSET + 1 bytes) in advance. |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
532 It may lookup a unification table TABLE if supplied. */ |
17052 | 533 |
534 int | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
535 find_charset_in_str (str, len, charsets, table) |
17052 | 536 unsigned char *str, *charsets; |
537 int len; | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
538 Lisp_Object table; |
17052 | 539 { |
540 int num = 0; | |
541 | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
542 if (! CHAR_TABLE_P (table)) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
543 table = Qnil; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
544 |
17052 | 545 while (len > 0) |
546 { | |
547 int bytes = BYTES_BY_CHAR_HEAD (*str); | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
548 int charset; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
549 |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
550 if (NILP (table)) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
551 charset = CHARSET_AT (str); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
552 else |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
553 { |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
554 int c, charset; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
555 unsigned char c1, c2; |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
556 |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
557 SPLIT_STRING(str, bytes, charset, c1, c2); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
558 if ((c = unify_char (table, -1, charset, c1, c2)) >= 0) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
559 charset = CHAR_CHARSET (c); |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
560 } |
17052 | 561 |
562 if (!charsets[charset]) | |
563 { | |
564 charsets[charset] = 1; | |
565 num += 1; | |
566 } | |
567 str += bytes; | |
568 len -= bytes; | |
569 } | |
570 return num; | |
571 } | |
572 | |
573 DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region, | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
574 2, 3, 0, |
17052 | 575 "Return a list of charsets in the region between BEG and END.\n\ |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
576 BEG and END are buffer positions.\n\ |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
577 Optional arg TABLE if non-nil is a unification table to look up.") |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
578 (beg, end, table) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
579 Lisp_Object beg, end, table; |
17052 | 580 { |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
581 char charsets[MAX_CHARSET + 1]; |
17052 | 582 int from, to, stop, i; |
583 Lisp_Object val; | |
584 | |
585 validate_region (&beg, &end); | |
586 from = XFASTINT (beg); | |
587 stop = to = XFASTINT (end); | |
588 if (from < GPT && GPT < to) | |
589 stop = GPT; | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
590 bzero (charsets, MAX_CHARSET + 1); |
17052 | 591 while (1) |
592 { | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
593 find_charset_in_str (POS_ADDR (from), stop - from, charsets, table); |
17052 | 594 if (stop < to) |
595 from = stop, stop = to; | |
596 else | |
597 break; | |
598 } | |
599 val = Qnil; | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
600 for (i = MAX_CHARSET; i >= 0; i--) |
17052 | 601 if (charsets[i]) |
602 val = Fcons (CHARSET_SYMBOL (i), val); | |
603 return val; | |
604 } | |
605 | |
606 DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string, | |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
607 1, 2, 0, |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
608 "Return a list of charsets in STR.\n\ |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
609 Optional arg TABLE if non-nil is a unification table to look up.") |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
610 (str, table) |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
611 Lisp_Object str, table; |
17052 | 612 { |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
613 char charsets[MAX_CHARSET + 1]; |
17052 | 614 int i; |
615 Lisp_Object val; | |
616 | |
617 CHECK_STRING (str, 0); | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
618 bzero (charsets, MAX_CHARSET + 1); |
17727
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
619 find_charset_in_str (XSTRING (str)->data, XSTRING (str)->size, |
9d39361ce928
(unify_char): New function.
Kenichi Handa <handa@m17n.org>
parents:
17366
diff
changeset
|
620 charsets, table); |
17052 | 621 val = Qnil; |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
622 for (i = MAX_CHARSET; i >= 0; i--) |
17052 | 623 if (charsets[i]) |
624 val = Fcons (CHARSET_SYMBOL (i), val); | |
625 return val; | |
626 } | |
627 | |
628 DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0, | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
629 "") |
17052 | 630 (charset, code1, code2) |
631 Lisp_Object charset, code1, code2; | |
632 { | |
633 CHECK_NUMBER (charset, 0); | |
634 | |
635 if (NILP (code1)) | |
636 XSETFASTINT (code1, 0); | |
637 else | |
638 CHECK_NUMBER (code1, 1); | |
639 if (NILP (code2)) | |
640 XSETFASTINT (code2, 0); | |
641 else | |
642 CHECK_NUMBER (code2, 2); | |
643 | |
644 if (!CHARSET_DEFINED_P (XINT (charset))) | |
645 error ("Invalid charset: %d", XINT (charset)); | |
646 | |
647 return make_number (MAKE_CHAR (XINT (charset), XINT (code1), XINT (code2))); | |
648 } | |
649 | |
650 DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0, | |
651 "Return list of charset and one or two position-codes of CHAR.") | |
652 (ch) | |
653 Lisp_Object ch; | |
654 { | |
655 Lisp_Object val; | |
17322
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
656 int charset, c1, c2; |
17052 | 657 |
658 CHECK_NUMBER (ch, 0); | |
659 SPLIT_CHAR (XFASTINT (ch), charset, c1, c2); | |
17322
a7d9b8e167ca
(non_ascii_char_to_string, update_charset_table,
Kenichi Handa <handa@m17n.org>
parents:
17186
diff
changeset
|
660 return (c2 >= 0 |
17052 | 661 ? Fcons (CHARSET_SYMBOL (charset), |
662 Fcons (make_number (c1), Fcons (make_number (c2), Qnil))) | |
663 : Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil))); | |
664 } | |
665 | |
666 DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0, | |
667 "Return charset of CHAR.") | |
668 (ch) | |
669 Lisp_Object ch; | |
670 { | |
671 CHECK_NUMBER (ch, 0); | |
672 | |
673 return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch))); | |
674 } | |
675 | |
676 DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0, | |
677 "Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR.") | |
678 (dimension, chars, final_char) | |
679 Lisp_Object dimension, chars, final_char; | |
680 { | |
681 int charset; | |
682 | |
683 CHECK_NUMBER (dimension, 0); | |
684 CHECK_NUMBER (chars, 1); | |
685 CHECK_NUMBER (final_char, 2); | |
686 | |
687 if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0) | |
688 return Qnil; | |
689 return CHARSET_SYMBOL (charset); | |
690 } | |
691 | |
692 DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0, | |
693 "Return byte length of multi-byte form of CHAR.") | |
694 (ch) | |
695 Lisp_Object ch; | |
696 { | |
697 Lisp_Object val; | |
698 int bytes; | |
699 | |
700 CHECK_NUMBER (ch, 0); | |
701 if (COMPOSITE_CHAR_P (XFASTINT (ch))) | |
702 { | |
703 unsigned int id = COMPOSITE_CHAR_ID (XFASTINT (ch)); | |
704 | |
705 bytes = (id < n_cmpchars ? cmpchar_table[id]->len : 1); | |
706 } | |
707 else | |
708 { | |
709 int charset = CHAR_CHARSET (XFASTINT (ch)); | |
710 | |
711 bytes = CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1; | |
712 } | |
713 | |
714 XSETFASTINT (val, bytes); | |
715 return val; | |
716 } | |
717 | |
718 /* Return the width of character of which multi-byte form starts with | |
719 C. The width is measured by how many columns occupied on the | |
720 screen when displayed in the current buffer. */ | |
721 | |
722 #define ONE_BYTE_CHAR_WIDTH(c) \ | |
723 (c < 0x20 \ | |
724 ? (c == '\t' \ | |
17366
f7809ff2c65f
(ONE_BYTE_CHAR_WIDTH, Fconcat_chars): Don't use Lisp_Object as integer.
Karl Heuer <kwzh@gnu.org>
parents:
17322
diff
changeset
|
725 ? XFASTINT (current_buffer->tab_width) \ |
17052 | 726 : (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \ |
727 : (c < 0x7f \ | |
728 ? 1 \ | |
729 : (c == 0x7F \ | |
730 ? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \ | |
731 : ((! NILP (current_buffer->enable_multibyte_characters) \ | |
732 && BASE_LEADING_CODE_P (c)) \ | |
733 ? WIDTH_BY_CHAR_HEAD (c) \ | |
734 : 4)))) \ | |
735 | |
736 | |
737 DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0, | |
738 "Return width of CHAR when displayed in the current buffer.\n\ | |
739 The width is measured by how many columns it occupies on the screen.") | |
740 (ch) | |
741 Lisp_Object ch; | |
742 { | |
743 Lisp_Object val; | |
744 int c; | |
745 | |
746 CHECK_NUMBER (ch, 0); | |
747 | |
748 c = XFASTINT (ch); | |
749 if (SINGLE_BYTE_CHAR_P (c)) | |
750 XSETFASTINT (val, ONE_BYTE_CHAR_WIDTH (c)); | |
751 else if (COMPOSITE_CHAR_P (c)) | |
752 { | |
753 int id = COMPOSITE_CHAR_ID (XFASTINT (ch)); | |
754 XSETFASTINT (val, (id < n_cmpchars ? cmpchar_table[id]->width : 0)); | |
755 } | |
756 else | |
757 { | |
758 int charset = CHAR_CHARSET (c); | |
759 | |
760 XSETFASTINT (val, CHARSET_WIDTH (charset)); | |
761 } | |
762 return val; | |
763 } | |
764 | |
765 /* Return width of string STR of length LEN when displayed in the | |
766 current buffer. The width is measured by how many columns it | |
767 occupies on the screen. */ | |
768 int | |
769 strwidth (str, len) | |
770 unsigned char *str; | |
771 int len; | |
772 { | |
773 unsigned char *endp = str + len; | |
774 int width = 0; | |
775 | |
776 while (str < endp) { | |
777 if (*str == LEADING_CODE_COMPOSITION) | |
778 { | |
779 int id = str_cmpchar_id (str, endp - str); | |
780 | |
781 if (id < 0) | |
782 { | |
783 width += 4; | |
784 str++; | |
785 } | |
786 else | |
787 { | |
788 width += cmpchar_table[id]->width; | |
789 str += cmpchar_table[id]->len; | |
790 } | |
791 } | |
792 else | |
793 { | |
794 width += ONE_BYTE_CHAR_WIDTH (*str); | |
795 str += BYTES_BY_CHAR_HEAD (*str); | |
796 } | |
797 } | |
798 return width; | |
799 } | |
800 | |
801 DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0, | |
802 "Return width of STRING when displayed in the current buffer.\n\ | |
803 Width is measured by how many columns it occupies on the screen.\n\ | |
804 When calculating width of a multi-byte character in STRING,\n\ | |
805 only the base leading-code is considered and the validity of\n\ | |
806 the following bytes are not checked.") | |
807 (str) | |
808 Lisp_Object str; | |
809 { | |
810 Lisp_Object val; | |
811 | |
812 CHECK_STRING (str, 0); | |
813 XSETFASTINT (val, strwidth (XSTRING (str)->data, XSTRING (str)->size)); | |
814 return val; | |
815 } | |
816 | |
817 DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0, | |
818 "Return the direction of CHAR.\n\ | |
819 The returned value is 0 for left-to-right and 1 for right-to-left.") | |
820 (ch) | |
821 Lisp_Object ch; | |
822 { | |
823 int charset; | |
824 | |
825 CHECK_NUMBER (ch, 0); | |
826 charset = CHAR_CHARSET (XFASTINT (ch)); | |
827 if (!CHARSET_DEFINED_P (charset)) | |
828 error ("Invalid character: %d", XINT (ch)); | |
829 return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX); | |
830 } | |
831 | |
832 DEFUN ("chars-in-string", Fchars_in_string, Schars_in_string, 1, 1, 0, | |
833 "Return number of characters in STRING.") | |
834 (str) | |
835 Lisp_Object str; | |
836 { | |
837 Lisp_Object val; | |
838 unsigned char *p, *endp; | |
839 int chars; | |
840 | |
841 CHECK_STRING (str, 0); | |
842 | |
843 p = XSTRING (str)->data; endp = p + XSTRING (str)->size; | |
844 chars = 0; | |
845 while (p < endp) | |
846 { | |
847 if (*p == LEADING_CODE_COMPOSITION) | |
848 { | |
849 p++; | |
850 while (p < endp && ! CHAR_HEAD_P (p)) p++; | |
851 } | |
852 else | |
853 p += BYTES_BY_CHAR_HEAD (*p); | |
854 chars++; | |
855 } | |
856 | |
857 XSETFASTINT (val, chars); | |
858 return val; | |
859 } | |
860 | |
861 DEFUN ("char-boundary-p", Fchar_boundary_p, Schar_boundary_p, 1, 1, 0, | |
862 "Return non-nil value if POS is at character boundary of multibyte form.\n\ | |
863 The return value is:\n\ | |
864 0 if POS is at an ASCII character or at the end of range,\n\ | |
865 1 if POS is at a head of 2-byte length multi-byte form,\n\ | |
866 2 if POS is at a head of 3-byte length multi-byte form,\n\ | |
867 3 if POS is at a head of 4-byte length multi-byte form,\n\ | |
868 4 if POS is at a head of multi-byte form of a composite character.\n\ | |
869 If POS is out of range or not at character boundary, return NIL.") | |
870 (pos) | |
871 Lisp_Object pos; | |
872 { | |
873 Lisp_Object val; | |
874 int n; | |
875 | |
876 CHECK_NUMBER_COERCE_MARKER (pos, 0); | |
877 | |
878 n = XINT (pos); | |
879 if (n < BEGV || n > ZV) | |
880 return Qnil; | |
881 | |
882 if (n == ZV || NILP (current_buffer->enable_multibyte_characters)) | |
883 XSETFASTINT (val, 0); | |
884 else | |
885 { | |
886 unsigned char *p = POS_ADDR (n); | |
887 | |
888 if (SINGLE_BYTE_CHAR_P (*p)) | |
889 XSETFASTINT (val, 0); | |
890 else if (*p == LEADING_CODE_COMPOSITION) | |
891 XSETFASTINT (val, 4); | |
892 else if (BYTES_BY_CHAR_HEAD (*p) > 1) | |
893 XSETFASTINT (val, BYTES_BY_CHAR_HEAD (*p) - 1); | |
894 else | |
895 val = Qnil; | |
896 } | |
897 return val; | |
898 } | |
899 | |
900 DEFUN ("concat-chars", Fconcat_chars, Sconcat_chars, 1, MANY, 0, | |
901 "Concatenate all the argument characters and make the result a string.") | |
17366
f7809ff2c65f
(ONE_BYTE_CHAR_WIDTH, Fconcat_chars): Don't use Lisp_Object as integer.
Karl Heuer <kwzh@gnu.org>
parents:
17322
diff
changeset
|
902 (n, args) |
f7809ff2c65f
(ONE_BYTE_CHAR_WIDTH, Fconcat_chars): Don't use Lisp_Object as integer.
Karl Heuer <kwzh@gnu.org>
parents:
17322
diff
changeset
|
903 int n; |
17052 | 904 Lisp_Object *args; |
905 { | |
17366
f7809ff2c65f
(ONE_BYTE_CHAR_WIDTH, Fconcat_chars): Don't use Lisp_Object as integer.
Karl Heuer <kwzh@gnu.org>
parents:
17322
diff
changeset
|
906 int i; |
17052 | 907 unsigned char *buf |
908 = (unsigned char *) malloc (MAX_LENGTH_OF_MULTI_BYTE_FORM * n); | |
909 unsigned char *p = buf; | |
910 Lisp_Object val; | |
911 | |
912 for (i = 0; i < n; i++) | |
913 { | |
914 int c, len; | |
915 unsigned char *str; | |
916 | |
917 if (!INTEGERP (args[i])) | |
918 { | |
919 free (buf); | |
920 CHECK_NUMBER (args[i], 0); | |
921 } | |
922 c = XINT (args[i]); | |
923 len = CHAR_STRING (c, p, str); | |
924 if (p != str) | |
925 /* C is a composite character. */ | |
926 bcopy (str, p, len); | |
927 p += len; | |
928 } | |
929 | |
930 val = make_string (buf, p - buf); | |
931 free (buf); | |
932 return val; | |
933 } | |
934 | |
935 #endif /* emacs */ | |
936 | |
937 /*** Composite characters staffs ***/ | |
938 | |
939 /* Each composite character is identified by CMPCHAR-ID which is | |
940 assigned when Emacs needs the character code of the composite | |
941 character (e.g. when displaying it on the screen). See the | |
942 document "GENERAL NOTE on COMPOSITE CHARACTER" in `charset.h' how a | |
943 composite character is represented in Emacs. */ | |
944 | |
945 /* If `static' is defined, it means that it is defined to null string. */ | |
946 #ifndef static | |
947 /* The following function is copied from lread.c. */ | |
948 static int | |
949 hash_string (ptr, len) | |
950 unsigned char *ptr; | |
951 int len; | |
952 { | |
953 register unsigned char *p = ptr; | |
954 register unsigned char *end = p + len; | |
955 register unsigned char c; | |
956 register int hash = 0; | |
957 | |
958 while (p != end) | |
959 { | |
960 c = *p++; | |
961 if (c >= 0140) c -= 40; | |
962 hash = ((hash<<3) + (hash>>28) + c); | |
963 } | |
964 return hash & 07777777777; | |
965 } | |
966 #endif | |
967 | |
968 #define CMPCHAR_HASH_TABLE_SIZE 0xFFF | |
969 | |
970 static int *cmpchar_hash_table[CMPCHAR_HASH_TABLE_SIZE]; | |
971 | |
972 /* Each element of `cmpchar_hash_table' is a pointer to an array of | |
973 integer, where the 1st element is the size of the array, the 2nd | |
974 element is how many elements are actually used in the array, and | |
975 the remaining elements are CMPCHAR-IDs of composite characters of | |
976 the same hash value. */ | |
977 #define CMPCHAR_HASH_SIZE(table) table[0] | |
978 #define CMPCHAR_HASH_USED(table) table[1] | |
979 #define CMPCHAR_HASH_CMPCHAR_ID(table, i) table[i] | |
980 | |
981 /* Return CMPCHAR-ID of the composite character in STR of the length | |
982 LEN. If the composite character has not yet been registered, | |
983 register it in `cmpchar_table' and assign new CMPCHAR-ID. This | |
984 is the sole function for assigning CMPCHAR-ID. */ | |
985 int | |
986 str_cmpchar_id (str, len) | |
987 unsigned char *str; | |
988 int len; | |
989 { | |
990 int hash_idx, *hashp; | |
991 unsigned char *buf; | |
992 int embedded_rule; /* 1 if composition rule is embedded. */ | |
993 int chars; /* number of components. */ | |
994 int i; | |
995 struct cmpchar_info *cmpcharp; | |
996 | |
997 if (len < 5) | |
998 /* Any composite char have at least 3-byte length. */ | |
999 return -1; | |
1000 | |
1001 /* The second byte 0xFF means compostion rule is embedded. */ | |
1002 embedded_rule = (str[1] == 0xFF); | |
1003 | |
1004 /* At first, get the actual length of the composite character. */ | |
1005 { | |
1006 unsigned char *p, *endp = str + 1, *lastp = str + len; | |
1007 int bytes; | |
1008 | |
1009 while (endp < lastp && ! CHAR_HEAD_P (endp)) endp++; | |
1010 chars = 0; | |
1011 p = str + 1 + embedded_rule; | |
1012 while (p < endp) | |
1013 { | |
1014 /* No need of checking if *P is 0xA0 because | |
1015 BYTES_BY_CHAR_HEAD (0x80) surely returns 2. */ | |
1016 p += (bytes = BYTES_BY_CHAR_HEAD (*p - 0x20) + embedded_rule); | |
1017 chars++; | |
1018 } | |
1019 len = (p -= embedded_rule) - str; | |
1020 if (p > endp) | |
1021 len -= - bytes, chars--; | |
1022 | |
1023 if (chars < 2 || chars > MAX_COMPONENT_COUNT) | |
1024 /* Invalid number of components. */ | |
1025 return -1; | |
1026 } | |
1027 hash_idx = hash_string (str, len) % CMPCHAR_HASH_TABLE_SIZE; | |
1028 hashp = cmpchar_hash_table[hash_idx]; | |
1029 | |
1030 /* Then, look into the hash table. */ | |
1031 if (hashp != NULL) | |
1032 /* Find the correct one among composite characters of the same | |
1033 hash value. */ | |
1034 for (i = 2; i < CMPCHAR_HASH_USED (hashp); i++) | |
1035 { | |
1036 cmpcharp = cmpchar_table[CMPCHAR_HASH_CMPCHAR_ID (hashp, i)]; | |
1037 if (len == cmpcharp->len | |
1038 && ! bcmp (str, cmpcharp->data, len)) | |
1039 return CMPCHAR_HASH_CMPCHAR_ID (hashp, i); | |
1040 } | |
1041 | |
1042 /* We have to register the composite character in cmpchar_table. */ | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
1043 if (n_cmpchars > (CHAR_FIELD2_MASK | CHAR_FIELD3_MASK)) |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
1044 /* No, we have no more room for a new composite character. */ |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
1045 return -1; |
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
1046 |
17052 | 1047 /* Make the entry in hash table. */ |
1048 if (hashp == NULL) | |
1049 { | |
1050 /* Make a table for 8 composite characters initially. */ | |
1051 hashp = (cmpchar_hash_table[hash_idx] | |
1052 = (int *) xmalloc (sizeof (int) * (2 + 8))); | |
1053 CMPCHAR_HASH_SIZE (hashp) = 10; | |
1054 CMPCHAR_HASH_USED (hashp) = 2; | |
1055 } | |
1056 else if (CMPCHAR_HASH_USED (hashp) >= CMPCHAR_HASH_SIZE (hashp)) | |
1057 { | |
1058 CMPCHAR_HASH_SIZE (hashp) += 8; | |
1059 hashp = (cmpchar_hash_table[hash_idx] | |
1060 = (int *) xrealloc (hashp, | |
1061 sizeof (int) * CMPCHAR_HASH_SIZE (hashp))); | |
1062 } | |
1063 CMPCHAR_HASH_CMPCHAR_ID (hashp, CMPCHAR_HASH_USED (hashp)) = n_cmpchars; | |
1064 CMPCHAR_HASH_USED (hashp)++; | |
1065 | |
1066 /* Set information of the composite character in cmpchar_table. */ | |
1067 if (cmpchar_table_size == 0) | |
1068 { | |
1069 /* This is the first composite character to be registered. */ | |
1070 cmpchar_table_size = 256; | |
1071 cmpchar_table | |
1072 = (struct cmpchar_info **) xmalloc (sizeof (cmpchar_table[0]) | |
1073 * cmpchar_table_size); | |
1074 } | |
1075 else if (cmpchar_table_size <= n_cmpchars) | |
1076 { | |
1077 cmpchar_table_size += 256; | |
1078 cmpchar_table | |
1079 = (struct cmpchar_info **) xrealloc (cmpchar_table, | |
1080 sizeof (cmpchar_table[0]) | |
1081 * cmpchar_table_size); | |
1082 } | |
1083 | |
1084 cmpcharp = (struct cmpchar_info *) xmalloc (sizeof (struct cmpchar_info)); | |
1085 | |
1086 cmpcharp->len = len; | |
1087 cmpcharp->data = (unsigned char *) xmalloc (len + 1); | |
1088 bcopy (str, cmpcharp->data, len); | |
1089 cmpcharp->data[len] = 0; | |
1090 cmpcharp->glyph_len = chars; | |
1091 cmpcharp->glyph = (GLYPH *) xmalloc (sizeof (GLYPH) * chars); | |
1092 if (embedded_rule) | |
1093 { | |
1094 cmpcharp->cmp_rule = (unsigned char *) xmalloc (chars); | |
1095 cmpcharp->col_offset = (float *) xmalloc (sizeof (float) * chars); | |
1096 } | |
1097 else | |
1098 { | |
1099 cmpcharp->cmp_rule = NULL; | |
1100 cmpcharp->col_offset = NULL; | |
1101 } | |
1102 | |
1103 /* Setup GLYPH data and composition rules (if any) so as not to make | |
1104 them every time on displaying. */ | |
1105 { | |
1106 unsigned char *bufp; | |
1107 int width; | |
1108 float leftmost = 0.0, rightmost = 1.0; | |
1109 | |
1110 if (embedded_rule) | |
1111 /* At first, col_offset[N] is set to relative to col_offset[0]. */ | |
1112 cmpcharp->col_offset[0] = 0; | |
1113 | |
1114 for (i = 0, bufp = cmpcharp->data + 1; i < chars; i++) | |
1115 { | |
1116 if (embedded_rule) | |
1117 cmpcharp->cmp_rule[i] = *bufp++; | |
1118 | |
1119 if (*bufp == 0xA0) /* This is an ASCII character. */ | |
1120 { | |
1121 cmpcharp->glyph[i] = FAST_MAKE_GLYPH ((*++bufp & 0x7F), 0); | |
1122 width = 1; | |
1123 bufp++; | |
1124 } | |
1125 else /* Multibyte character. */ | |
1126 { | |
1127 /* Make `bufp' point normal multi-byte form temporally. */ | |
1128 *bufp -= 0x20; | |
1129 cmpcharp->glyph[i] | |
1130 = FAST_MAKE_GLYPH (string_to_non_ascii_char (bufp, 4, 0), 0); | |
1131 width = WIDTH_BY_CHAR_HEAD (*bufp); | |
1132 *bufp += 0x20; | |
1133 bufp += BYTES_BY_CHAR_HEAD (*bufp - 0x20); | |
1134 } | |
1135 | |
1136 if (embedded_rule && i > 0) | |
1137 { | |
1138 /* Reference points (global_ref and new_ref) are | |
1139 encoded as below: | |
1140 | |
1141 0--1--2 -- ascent | |
1142 | | | |
1143 | | | |
1144 | 4 -+--- center | |
1145 -- 3 5 -- baseline | |
1146 | | | |
1147 6--7--8 -- descent | |
1148 | |
1149 Now, we calculate the column offset of the new glyph | |
1150 from the left edge of the first glyph. This can avoid | |
1151 the same calculation everytime displaying this | |
1152 composite character. */ | |
1153 | |
1154 /* Reference points of global glyph and new glyph. */ | |
1155 int global_ref = (cmpcharp->cmp_rule[i] - 0xA0) / 9; | |
1156 int new_ref = (cmpcharp->cmp_rule[i] - 0xA0) % 9; | |
1157 /* Column offset relative to the first glyph. */ | |
1158 float left = (leftmost | |
1159 + (global_ref % 3) * (rightmost - leftmost) / 2.0 | |
1160 - (new_ref % 3) * width / 2.0); | |
1161 | |
1162 cmpcharp->col_offset[i] = left; | |
1163 if (left < leftmost) | |
1164 leftmost = left; | |
1165 if (left + width > rightmost) | |
1166 rightmost = left + width; | |
1167 } | |
1168 else | |
1169 { | |
1170 if (width > rightmost) | |
1171 rightmost = width; | |
1172 } | |
1173 } | |
1174 if (embedded_rule) | |
1175 { | |
1176 /* Now col_offset[N] are relative to the left edge of the | |
1177 first component. Make them relative to the left edge of | |
1178 overall glyph. */ | |
1179 for (i = 0; i < chars; i++) | |
1180 cmpcharp->col_offset[i] -= leftmost; | |
1181 /* Make rightmost holds width of overall glyph. */ | |
1182 rightmost -= leftmost; | |
1183 } | |
1184 | |
1185 cmpcharp->width = rightmost; | |
1186 if (cmpcharp->width < rightmost) | |
1187 /* To get a ceiling integer value. */ | |
1188 cmpcharp->width++; | |
1189 } | |
1190 | |
1191 cmpchar_table[n_cmpchars] = cmpcharp; | |
1192 | |
1193 return n_cmpchars++; | |
1194 } | |
1195 | |
1196 /* Return the Nth element of the composite character C. */ | |
1197 int | |
1198 cmpchar_component (c, n) | |
1199 unsigned int c, n; | |
1200 { | |
1201 int id = COMPOSITE_CHAR_ID (c); | |
1202 | |
1203 if (id >= n_cmpchars /* C is not a valid composite character. */ | |
1204 || n >= cmpchar_table[id]->glyph_len) /* No such component. */ | |
1205 return -1; | |
1206 /* No face data is stored in glyph code. */ | |
1207 return ((int) (cmpchar_table[id]->glyph[n])); | |
1208 } | |
1209 | |
1210 DEFUN ("cmpcharp", Fcmpcharp, Scmpcharp, 1, 1, 0, | |
1211 "T if CHAR is a composite character.") | |
1212 (ch) | |
1213 Lisp_Object ch; | |
1214 { | |
1215 CHECK_NUMBER (ch, 0); | |
1216 return (COMPOSITE_CHAR_P (XINT (ch)) ? Qt : Qnil); | |
1217 } | |
1218 | |
1219 DEFUN ("composite-char-component", Fcmpchar_component, Scmpchar_component, | |
1220 2, 2, 0, | |
1221 "Return the IDXth component character of composite character CHARACTER.") | |
1222 (character, idx) | |
1223 Lisp_Object character, idx; | |
1224 { | |
1225 int c; | |
1226 | |
1227 CHECK_NUMBER (character, 0); | |
1228 CHECK_NUMBER (idx, 1); | |
1229 | |
1230 if ((c = cmpchar_component (XINT (character), XINT (idx))) < 0) | |
1231 args_out_of_range (character, idx); | |
1232 | |
1233 return make_number (c); | |
1234 } | |
1235 | |
1236 DEFUN ("composite-char-composition-rule", Fcmpchar_cmp_rule, Scmpchar_cmp_rule, | |
1237 2, 2, 0, | |
17121
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1238 "Return the Nth composition rule embedded in composite character CHARACTER.\n\ |
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1239 The returned rule is for composing the Nth component\n\ |
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1240 on the (N-1)th component. If N is 0, the returned value is always 255.") |
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1241 (character, n) |
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1242 Lisp_Object character, n; |
17052 | 1243 { |
1244 int id, i; | |
1245 | |
1246 CHECK_NUMBER (character, 0); | |
17121
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1247 CHECK_NUMBER (n, 1); |
17052 | 1248 |
1249 id = COMPOSITE_CHAR_ID (XINT (character)); | |
1250 if (id < 0 || id >= n_cmpchars) | |
1251 error ("Invalid composite character: %d", XINT (character)); | |
17121
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1252 i = XINT (n); |
17052 | 1253 if (i > cmpchar_table[id]->glyph_len) |
17121
f8e2cd94c274
(Fcmpchar_cmp_rule): Escape newlines in docstring.
Kenichi Handa <handa@m17n.org>
parents:
17071
diff
changeset
|
1254 args_out_of_range (character, n); |
17052 | 1255 |
1256 return make_number (cmpchar_table[id]->cmp_rule[i]); | |
1257 } | |
1258 | |
1259 DEFUN ("composite-char-composition-rule-p", Fcmpchar_cmp_rule_p, | |
1260 Scmpchar_cmp_rule_p, 1, 1, 0, | |
1261 "Return non-nil if composite character CHARACTER contains a embedded rule.") | |
1262 (character) | |
1263 Lisp_Object character; | |
1264 { | |
1265 int id; | |
1266 | |
1267 CHECK_NUMBER (character, 0); | |
1268 id = COMPOSITE_CHAR_ID (XINT (character)); | |
1269 if (id < 0 || id >= n_cmpchars) | |
1270 error ("Invalid composite character: %d", XINT (character)); | |
1271 | |
1272 return (cmpchar_table[id]->cmp_rule ? Qt : Qnil); | |
1273 } | |
1274 | |
1275 DEFUN ("composite-char-component-count", Fcmpchar_cmp_count, | |
1276 Scmpchar_cmp_count, 1, 1, 0, | |
1277 "Return number of compoents of composite character CHARACTER.") | |
1278 (character) | |
1279 Lisp_Object character; | |
1280 { | |
1281 int id; | |
1282 | |
1283 CHECK_NUMBER (character, 0); | |
1284 id = COMPOSITE_CHAR_ID (XINT (character)); | |
1285 if (id < 0 || id >= n_cmpchars) | |
1286 error ("Invalid composite character: %d", XINT (character)); | |
1287 | |
1288 return (make_number (cmpchar_table[id]->glyph_len)); | |
1289 } | |
1290 | |
1291 DEFUN ("compose-string", Fcompose_string, Scompose_string, | |
1292 1, 1, 0, | |
1293 "Return one char string composed from all characters in STRING.") | |
1294 (str) | |
1295 Lisp_Object str; | |
1296 { | |
1297 unsigned char buf[MAX_LENGTH_OF_MULTI_BYTE_FORM], *p, *pend, *ptemp; | |
1298 int len, i; | |
1299 | |
1300 CHECK_STRING (str, 0); | |
1301 | |
1302 buf[0] = LEADING_CODE_COMPOSITION; | |
1303 p = XSTRING (str)->data; | |
1304 pend = p + XSTRING (str)->size; | |
1305 i = 1; | |
1306 while (p < pend) | |
1307 { | |
1308 if (*p < 0x20 || *p == 127) /* control code */ | |
1309 error ("Invalid component character: %d", *p); | |
1310 else if (*p < 0x80) /* ASCII */ | |
1311 { | |
1312 if (i + 2 >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1313 error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1314 /* Prepend an ASCII charset indicator 0xA0, set MSB of the | |
1315 code itself. */ | |
1316 buf[i++] = 0xA0; | |
1317 buf[i++] = *p++ + 0x80; | |
1318 } | |
1319 else if (*p == LEADING_CODE_COMPOSITION) /* composite char */ | |
1320 { | |
1321 /* Already composed. Eliminate the heading | |
1322 LEADING_CODE_COMPOSITION, keep the remaining bytes | |
1323 unchanged. */ | |
1324 p++; | |
1325 ptemp = p; | |
1326 while (! CHAR_HEAD_P (p)) p++; | |
1327 if (i + (p - ptemp) >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1328 error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1329 bcopy (ptemp, buf + i, p - ptemp); | |
1330 i += p - ptemp; | |
1331 } | |
1332 else /* multibyte char */ | |
1333 { | |
1334 /* Add 0x20 to the base leading-code, keep the remaining | |
1335 bytes unchanged. */ | |
1336 len = BYTES_BY_CHAR_HEAD (*p); | |
1337 if (i + len >= MAX_LENGTH_OF_MULTI_BYTE_FORM) | |
1338 error ("Too long string to be composed: %s", XSTRING (str)->data); | |
1339 bcopy (p, buf + i, len); | |
1340 buf[i] += 0x20; | |
1341 p += len, i += len; | |
1342 } | |
1343 } | |
1344 | |
1345 if (i < 5) | |
1346 /* STR contains only one character, which can't be composed. */ | |
1347 error ("Too short string to be composed: %s", XSTRING (str)->data); | |
1348 | |
1349 return make_string (buf, i); | |
1350 } | |
1351 | |
1352 | |
1353 charset_id_internal (charset_name) | |
1354 char *charset_name; | |
1355 { | |
1356 Lisp_Object val = Fget (intern (charset_name), Qcharset); | |
1357 | |
1358 if (!VECTORP (val)) | |
1359 error ("Charset %s is not defined", charset_name); | |
1360 | |
1361 return (XINT (XVECTOR (val)->contents[0])); | |
1362 } | |
1363 | |
1364 DEFUN ("setup-special-charsets", Fsetup_special_charsets, | |
1365 Ssetup_special_charsets, 0, 0, 0, "Internal use only.") | |
1366 () | |
1367 { | |
1368 charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1"); | |
1369 charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978"); | |
1370 charset_jisx0208 = charset_id_internal ("japanese-jisx0208"); | |
1371 charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201"); | |
1372 charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201"); | |
1373 charset_big5_1 = charset_id_internal ("chinese-big5-1"); | |
1374 charset_big5_2 = charset_id_internal ("chinese-big5-2"); | |
1375 return Qnil; | |
1376 } | |
1377 | |
1378 init_charset_once () | |
1379 { | |
1380 int i, j, k; | |
1381 | |
1382 staticpro (&Vcharset_table); | |
1383 staticpro (&Vcharset_symbol_table); | |
1384 | |
1385 /* This has to be done here, before we call Fmake_char_table. */ | |
1386 Qcharset_table = intern ("charset-table"); | |
1387 staticpro (&Qcharset_table); | |
1388 | |
1389 /* Intern this now in case it isn't already done. | |
1390 Setting this variable twice is harmless. | |
1391 But don't staticpro it here--that is done in alloc.c. */ | |
1392 Qchar_table_extra_slots = intern ("char-table-extra-slots"); | |
1393 | |
1394 /* Now we are ready to set up this property, so we can | |
1395 create the charset table. */ | |
1396 Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0)); | |
1397 Vcharset_table = Fmake_char_table (Qcharset_table, Qnil); | |
1398 | |
17186
65bc06706714
Adjusted for the change of MAX_CHARSET.
Kenichi Handa <handa@m17n.org>
parents:
17121
diff
changeset
|
1399 Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1), Qnil); |
17052 | 1400 |
1401 /* Setup tables. */ | |
1402 for (i = 0; i < 2; i++) | |
1403 for (j = 0; j < 2; j++) | |
1404 for (k = 0; k < 128; k++) | |
1405 iso_charset_table [i][j][k] = -1; | |
1406 | |
1407 bzero (cmpchar_hash_table, sizeof cmpchar_hash_table); | |
1408 cmpchar_table_size = n_cmpchars = 0; | |
1409 | |
1410 for (i = 0; i < 256; i++) | |
1411 BYTES_BY_CHAR_HEAD (i) = 1; | |
1412 BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_11) = 3; | |
1413 BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_12) = 3; | |
1414 BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_21) = 4; | |
1415 BYTES_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_22) = 4; | |
1416 /* The following doesn't reflect the actual bytes, but just to tell | |
1417 that it is a start of a multibyte character. */ | |
1418 BYTES_BY_CHAR_HEAD (LEADING_CODE_COMPOSITION) = 2; | |
1419 | |
1420 for (i = 0; i < 128; i++) | |
1421 WIDTH_BY_CHAR_HEAD (i) = 1; | |
1422 for (; i < 256; i++) | |
1423 WIDTH_BY_CHAR_HEAD (i) = 4; | |
1424 WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_11) = 1; | |
1425 WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_12) = 2; | |
1426 WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_21) = 1; | |
1427 WIDTH_BY_CHAR_HEAD (LEADING_CODE_PRIVATE_22) = 2; | |
1428 } | |
1429 | |
1430 #ifdef emacs | |
1431 | |
1432 syms_of_charset () | |
1433 { | |
1434 Qascii = intern ("ascii"); | |
1435 staticpro (&Qascii); | |
1436 | |
1437 Qcharset = intern ("charset"); | |
1438 staticpro (&Qcharset); | |
1439 | |
1440 /* Define ASCII charset now. */ | |
1441 update_charset_table (make_number (CHARSET_ASCII), | |
1442 make_number (1), make_number (94), | |
1443 make_number (1), | |
1444 make_number (0), | |
1445 make_number ('B'), | |
1446 make_number (0), | |
1447 build_string ("ASCII"), | |
1448 build_string ("ASCII"), | |
1449 build_string ("ASCII (ISO646 IRV)")); | |
1450 CHARSET_SYMBOL (CHARSET_ASCII) = Qascii; | |
1451 Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII)); | |
1452 | |
1453 Qcomposition = intern ("composition"); | |
1454 staticpro (&Qcomposition); | |
1455 CHARSET_SYMBOL (CHARSET_COMPOSITION) = Qcomposition; | |
1456 | |
1457 defsubr (&Sdefine_charset); | |
1458 defsubr (&Sdeclare_equiv_charset); | |
1459 defsubr (&Sfind_charset_region); | |
1460 defsubr (&Sfind_charset_string); | |
1461 defsubr (&Smake_char_internal); | |
1462 defsubr (&Ssplit_char); | |
1463 defsubr (&Schar_charset); | |
1464 defsubr (&Siso_charset); | |
1465 defsubr (&Schar_bytes); | |
1466 defsubr (&Schar_width); | |
1467 defsubr (&Sstring_width); | |
1468 defsubr (&Schar_direction); | |
1469 defsubr (&Schars_in_string); | |
1470 defsubr (&Schar_boundary_p); | |
1471 defsubr (&Sconcat_chars); | |
1472 defsubr (&Scmpcharp); | |
1473 defsubr (&Scmpchar_component); | |
1474 defsubr (&Scmpchar_cmp_rule); | |
1475 defsubr (&Scmpchar_cmp_rule_p); | |
1476 defsubr (&Scmpchar_cmp_count); | |
1477 defsubr (&Scompose_string); | |
1478 defsubr (&Ssetup_special_charsets); | |
1479 | |
1480 DEFVAR_LISP ("charset-list", &Vcharset_list, | |
1481 "List of charsets ever defined."); | |
1482 Vcharset_list = Fcons (Qascii, Qnil); | |
1483 | |
1484 DEFVAR_INT ("leading-code-composition", &leading_code_composition, | |
1485 "Leading-code of composite characters."); | |
1486 leading_code_composition = LEADING_CODE_COMPOSITION; | |
1487 | |
1488 DEFVAR_INT ("leading-code-private-11", &leading_code_private_11, | |
1489 "Leading-code of private TYPE9N charset of column-width 1."); | |
1490 leading_code_private_11 = LEADING_CODE_PRIVATE_11; | |
1491 | |
1492 DEFVAR_INT ("leading-code-private-12", &leading_code_private_12, | |
1493 "Leading-code of private TYPE9N charset of column-width 2."); | |
1494 leading_code_private_12 = LEADING_CODE_PRIVATE_12; | |
1495 | |
1496 DEFVAR_INT ("leading-code-private-21", &leading_code_private_21, | |
1497 "Leading-code of private TYPE9Nx9N charset of column-width 1."); | |
1498 leading_code_private_21 = LEADING_CODE_PRIVATE_21; | |
1499 | |
1500 DEFVAR_INT ("leading-code-private-22", &leading_code_private_22, | |
1501 "Leading-code of private TYPE9Nx9N charset of column-width 2."); | |
1502 leading_code_private_22 = LEADING_CODE_PRIVATE_22; | |
1503 } | |
1504 | |
1505 #endif /* emacs */ |