Mercurial > hg > xemacs-beta
annotate src/charset.h @ 5724:ede80ef92a74
Make soft links in src for module source files, if built in to the executable.
This ensures that those files are built with the same compiler flags as all
other source files.
See these xemacs-beta messages:
<CAHCOHQn+q=Xuwq+y68dvqi7afAP9f-TdB7=8YiZ8VYO816sjHg@mail.gmail.com>
<f5by5ejqiyk.fsf@calexico.inf.ed.ac.uk>
author | Jerry James <james@xemacs.org> |
---|---|
date | Sat, 02 Mar 2013 14:32:37 -0700 |
parents | 308d34e9f07d |
children |
rev | line source |
---|---|
771 | 1 /* Header for charsets. |
2 Copyright (C) 1992, 1995 Free Software Foundation, Inc. | |
3 Copyright (C) 1995 Sun Microsystems, Inc. | |
5200
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
4 Copyright (C) 2001, 2002, 2010 Ben Wing. |
771 | 5 |
6 This file is part of XEmacs. | |
7 | |
5402
308d34e9f07d
Changed bulk of GPLv2 or later files identified by script
Mats Lidell <matsl@xemacs.org>
parents:
5200
diff
changeset
|
8 XEmacs is free software: you can redistribute it and/or modify it |
771 | 9 under the terms of the GNU General Public License as published by the |
5402
308d34e9f07d
Changed bulk of GPLv2 or later files identified by script
Mats Lidell <matsl@xemacs.org>
parents:
5200
diff
changeset
|
10 Free Software Foundation, either version 3 of the License, or (at your |
308d34e9f07d
Changed bulk of GPLv2 or later files identified by script
Mats Lidell <matsl@xemacs.org>
parents:
5200
diff
changeset
|
11 option) any later version. |
771 | 12 |
13 XEmacs is distributed in the hope that it will be useful, but WITHOUT | |
14 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
15 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License | |
16 for more details. | |
17 | |
18 You should have received a copy of the GNU General Public License | |
5402
308d34e9f07d
Changed bulk of GPLv2 or later files identified by script
Mats Lidell <matsl@xemacs.org>
parents:
5200
diff
changeset
|
19 along with XEmacs. If not, see <http://www.gnu.org/licenses/>. */ |
771 | 20 |
21 /* Synched up with: Mule 2.3. Not in FSF. */ | |
22 | |
23 /* Rewritten by Ben Wing <ben@xemacs.org>. */ | |
24 | |
25 #ifndef INCLUDED_charset_h_ | |
26 #define INCLUDED_charset_h_ | |
27 | |
28 | |
29 | |
30 #ifndef MULE | |
31 | |
32 /************************************************************************/ | |
33 /* fake charset defs */ | |
34 /************************************************************************/ | |
35 | |
36 /* used when MULE is not defined, so that Charset-type stuff can still | |
37 be done */ | |
38 | |
39 #define Vcharset_ascii Qnil | |
40 | |
867 | 41 #define ichar_charset(ch) Vcharset_ascii |
42 #define ichar_leading_byte(ch) LEADING_BYTE_ASCII | |
43 #define ichar_len(ch) 1 | |
44 #define ichar_len_fmt(ch, fmt) 1 | |
771 | 45 #define LEADING_BYTE_ASCII 0x80 |
46 #define NUM_LEADING_BYTES 1 | |
47 #define MIN_LEADING_BYTE 0x80 | |
48 #define CHARSETP(cs) 1 | |
826 | 49 #define charset_by_leading_byte(lb) Vcharset_ascii |
771 | 50 #define XCHARSET_LEADING_BYTE(cs) LEADING_BYTE_ASCII |
51 #define XCHARSET_GRAPHIC(cs) -1 | |
52 #define XCHARSET_COLUMNS(cs) 1 | |
53 #define XCHARSET_DIMENSION(cs) 1 | |
867 | 54 #define BREAKUP_ICHAR(ch, charset, byte1, byte2) do { \ |
771 | 55 (charset) = Vcharset_ascii; \ |
56 (byte1) = (ch); \ | |
57 (byte2) = 0; \ | |
58 } while (0) | |
3659 | 59 #define XCHARSET_CCL_PROGRAM(cs) Qnil |
60 #define XCHARSET_NAME(cs) Qascii | |
5019 | 61 #define Fget_charset(cs) (cs) |
62 #define Fcharset_list() list1 (Vcharset_ascii) | |
771 | 63 |
64 #else /* MULE */ | |
65 | |
66 | |
67 /************************************************************************/ | |
68 /* Definition of leading bytes */ | |
69 /************************************************************************/ | |
70 | |
71 #define MIN_LEADING_BYTE 0x7F | |
72 | |
73 /** The following are for 1-byte characters in an official charset. **/ | |
74 enum LEADING_BYTE_OFFICIAL_1 | |
75 { | |
76 MIN_LEADING_BYTE_OFFICIAL_1 = 0x80, | |
77 /* LEADING_BYTE_LATIN_ISO8859_1 *MUST* be equal to | |
78 MIN_LEADING_BYTE_OFFICIAL_1. */ | |
79 LEADING_BYTE_LATIN_ISO8859_1 = /* 0x80 Right half of ISO 8859-1 */ | |
80 MIN_LEADING_BYTE_OFFICIAL_1, | |
81 LEADING_BYTE_LATIN_ISO8859_2, /* 0x81 Right half of ISO 8859-2 */ | |
82 LEADING_BYTE_LATIN_ISO8859_3, /* 0x82 Right half of ISO 8859-3 */ | |
83 LEADING_BYTE_LATIN_ISO8859_4, /* 0x83 Right half of ISO 8859-4 */ | |
84 LEADING_BYTE_THAI_TIS620, /* 0x84 TIS620-2533 */ | |
85 LEADING_BYTE_GREEK_ISO8859_7, /* 0x85 Right half of ISO 8859-7 */ | |
86 LEADING_BYTE_ARABIC_ISO8859_6, /* 0x86 Right half of ISO 8859-6 */ | |
87 LEADING_BYTE_HEBREW_ISO8859_8, /* 0x87 Right half of ISO 8859-8 */ | |
88 LEADING_BYTE_KATAKANA_JISX0201, /* 0x88 Right half of JIS X0201-1976 */ | |
89 LEADING_BYTE_LATIN_JISX0201, /* 0x89 Left half of JIS X0201-1976 */ | |
90 LEADING_BYTE_CYRILLIC_ISO8859_5,/* 0x8A Right half of ISO 8859-5 */ | |
91 LEADING_BYTE_LATIN_ISO8859_9, /* 0x8B Right half of ISO 8859-9 */ | |
92 LEADING_BYTE_LATIN_ISO8859_15, /* 0x8C Right half of ISO 8859-15 */ | |
93 #ifdef ENABLE_COMPOSITE_CHARS | |
94 LEADING_BYTE_COMPOSITE, /* 0x8D For a composite character */ | |
95 MAX_LEADING_BYTE_OFFICIAL_1 = | |
96 LEADING_BYTE_COMPOSITE - 1, | |
97 #else | |
98 /* Does not need to be the last entry, but simplifies things */ | |
99 LEADING_BYTE_COMPOSITE_REPLACEMENT, /* 0x8D Replaces ESC 0 - ESC 4 in a | |
100 buffer */ | |
101 MAX_LEADING_BYTE_OFFICIAL_1 = | |
102 LEADING_BYTE_COMPOSITE_REPLACEMENT, | |
103 #endif | |
104 /* 0x8E Unused */ | |
105 }; | |
106 | |
107 /* These next 2 + LEADING_BYTE_COMPOSITE need special treatment in a string | |
108 and/or character */ | |
109 | |
110 #define LEADING_BYTE_ASCII 0x7F /* Not used except in arrays | |
111 indexed by leading byte */ | |
112 #define LEADING_BYTE_CONTROL_1 0x8F /* represent normal 80-9F */ | |
113 | |
114 /** The following are for 2-byte characters in an official charset. **/ | |
115 enum LEADING_BYTE_OFFICIAL_2 | |
116 { | |
117 MIN_LEADING_BYTE_OFFICIAL_2 = 0x90, | |
118 LEADING_BYTE_JAPANESE_JISX0208_1978 = | |
119 MIN_LEADING_BYTE_OFFICIAL_2, /* 0x90 Japanese JIS X0208-1978 */ | |
120 LEADING_BYTE_CHINESE_GB2312, /* 0x91 Chinese Hanzi GB2312-1980 */ | |
121 LEADING_BYTE_JAPANESE_JISX0208, /* 0x92 Japanese JIS X0208-1983 */ | |
122 LEADING_BYTE_KOREAN_KSC5601, /* 0x93 Hangul KS C5601-1987 */ | |
123 LEADING_BYTE_JAPANESE_JISX0212, /* 0x94 Japanese JIS X0212-1990 */ | |
124 LEADING_BYTE_CHINESE_CNS11643_1, /* 0x95 Chinese CNS11643 Set 1 */ | |
125 LEADING_BYTE_CHINESE_CNS11643_2, /* 0x96 Chinese CNS11643 Set 2 */ | |
126 LEADING_BYTE_CHINESE_BIG5_1, /* 0x97 Big5 Level 1 */ | |
127 LEADING_BYTE_CHINESE_BIG5_2, /* 0x98 Big5 Level 2 */ | |
128 MAX_LEADING_BYTE_OFFICIAL_2 = | |
129 LEADING_BYTE_CHINESE_BIG5_2, | |
130 | |
131 /* 0x99 unused */ | |
132 /* 0x9A unused */ | |
133 /* 0x9B unused */ | |
134 /* 0x9C unused */ | |
135 /* 0x9D unused */ | |
136 }; | |
137 | |
138 | |
139 /** The following are for 1- and 2-byte characters in a private charset. **/ | |
140 | |
141 #define PRE_LEADING_BYTE_PRIVATE_1 0x9E /* 1-byte char-set */ | |
142 #define PRE_LEADING_BYTE_PRIVATE_2 0x9F /* 2-byte char-set */ | |
143 | |
144 #define MIN_LEADING_BYTE_PRIVATE_1 0xA0 | |
3496 | 145 #define MAX_LEADING_BYTE_PRIVATE_1 0xC0 |
146 #define MIN_LEADING_BYTE_PRIVATE_2 0xC1 | |
771 | 147 #define MAX_LEADING_BYTE_PRIVATE_2 0xFF |
148 | |
149 #define NUM_LEADING_BYTES 129 | |
150 | |
151 | |
152 /************************************************************************/ | |
153 /* Operations on leading bytes */ | |
154 /************************************************************************/ | |
155 | |
156 /* Is this leading byte for a private charset? */ | |
157 | |
826 | 158 #define leading_byte_private_p(lb) ((lb) >= MIN_LEADING_BYTE_PRIVATE_1) |
771 | 159 |
160 /* Is this a prefix for a private leading byte? */ | |
161 | |
826 | 162 DECLARE_INLINE_HEADER ( |
163 int | |
867 | 164 leading_byte_prefix_p (Ibyte lb) |
826 | 165 ) |
771 | 166 { |
167 return (lb == PRE_LEADING_BYTE_PRIVATE_1 || | |
168 lb == PRE_LEADING_BYTE_PRIVATE_2); | |
169 } | |
170 | |
171 /* Given a private leading byte, return the leading byte prefix stored | |
172 in a string. */ | |
173 | |
826 | 174 #define private_leading_byte_prefix(lb) \ |
771 | 175 ((unsigned int) (lb) < MIN_LEADING_BYTE_PRIVATE_2 ? \ |
176 PRE_LEADING_BYTE_PRIVATE_1 : \ | |
177 PRE_LEADING_BYTE_PRIVATE_2) | |
178 | |
179 | |
180 /************************************************************************/ | |
181 /* Information about a particular character set */ | |
182 /************************************************************************/ | |
183 | |
184 struct Lisp_Charset | |
185 { | |
5127
a9c41067dd88
more cleanups, terminology clarification, lots of doc work
Ben Wing <ben@xemacs.org>
parents:
5125
diff
changeset
|
186 NORMAL_LISP_OBJECT_HEADER header; |
771 | 187 |
188 int id; | |
189 Lisp_Object name; | |
190 Lisp_Object doc_string; | |
3659 | 191 Lisp_Object registries; |
771 | 192 Lisp_Object short_name; |
193 Lisp_Object long_name; | |
194 | |
195 Lisp_Object reverse_direction_charset; | |
196 | |
197 Lisp_Object ccl_program; | |
198 | |
2367 | 199 /* Unicode translation tables. See unicode.c for the format of these |
200 tables, and discussion of how they are initialized. | |
201 */ | |
771 | 202 void *to_unicode_table; |
203 void *from_unicode_table; | |
204 int from_unicode_levels; | |
205 | |
2367 | 206 /* Final byte of this character set in ISO2022 designating escape |
207 sequence */ | |
867 | 208 Ibyte final; |
771 | 209 |
210 /* Number of bytes (1 - 4) required in the internal representation | |
211 for characters in this character set. This is *not* the | |
212 same as the dimension of the character set). */ | |
213 int rep_bytes; | |
214 | |
215 /* Number of columns a character in this charset takes up, on TTY | |
216 devices. Not used for X devices. */ | |
217 int columns; | |
218 | |
219 /* Direction of this character set */ | |
220 int direction; | |
221 | |
222 /* Type of this character set (94, 96, 94x94, 96x96) */ | |
223 int type; | |
224 | |
225 /* Number of bytes used in encoding of this character set (1 or 2) */ | |
226 int dimension; | |
227 | |
228 /* Number of chars in each dimension (usually 94 or 96) */ | |
229 int chars; | |
230 | |
231 /* Which half of font to be used to display this character set */ | |
232 int graphic; | |
233 | |
3439 | 234 /* If set, this charset should be written out in ISO-2022-based coding |
235 systems using the escape sequence for UTF-8, not using our internal | |
236 representation and the associated real ISO 2022 designation. */ | |
237 unsigned int encode_as_utf_8 :1; | |
238 | |
771 | 239 /* If set, this is a "temporary" charset created when we encounter |
240 an unknown final. This is so that we can successfully compile | |
241 and load such files. We allow a real charset to be created on top | |
242 of this temporary charset. */ | |
243 unsigned int temporary :1; | |
244 }; | |
245 typedef struct Lisp_Charset Lisp_Charset; | |
246 | |
5118
e0db3c197671
merge up to latest default branch, doesn't compile yet
Ben Wing <ben@xemacs.org>
parents:
4096
diff
changeset
|
247 DECLARE_LISP_OBJECT (charset, Lisp_Charset); |
771 | 248 #define XCHARSET(x) XRECORD (x, charset, Lisp_Charset) |
249 #define wrap_charset(p) wrap_record (p, charset) | |
250 #define CHARSETP(x) RECORDP (x, charset) | |
251 #define CHECK_CHARSET(x) CHECK_RECORD (x, charset) | |
252 #define CONCHECK_CHARSET(x) CONCHECK_RECORD (x, charset) | |
253 | |
254 #define CHARSET_TYPE_94 0 /* This charset includes 94 characters. */ | |
255 #define CHARSET_TYPE_96 1 /* This charset includes 96 characters. */ | |
256 #define CHARSET_TYPE_94X94 2 /* This charset includes 94x94 characters. */ | |
257 #define CHARSET_TYPE_96X96 3 /* This charset includes 96x96 characters. */ | |
258 | |
259 #define CHARSET_LEFT_TO_RIGHT 0 | |
260 #define CHARSET_RIGHT_TO_LEFT 1 | |
261 | |
262 /* Leading byte and id have been regrouped. -- OG */ | |
263 #define CHARSET_ID(cs) ((cs)->id) | |
867 | 264 #define CHARSET_LEADING_BYTE(cs) ((Ibyte) CHARSET_ID (cs)) |
771 | 265 #define CHARSET_NAME(cs) ((cs)->name) |
266 #define CHARSET_SHORT_NAME(cs) ((cs)->short_name) | |
267 #define CHARSET_LONG_NAME(cs) ((cs)->long_name) | |
268 #define CHARSET_REP_BYTES(cs) ((cs)->rep_bytes) | |
269 #define CHARSET_COLUMNS(cs) ((cs)->columns) | |
270 #define CHARSET_GRAPHIC(cs) ((cs)->graphic) | |
3439 | 271 #define CHARSET_ENCODE_AS_UTF_8(cs) ((cs)->encode_as_utf_8) |
771 | 272 #define CHARSET_TYPE(cs) ((cs)->type) |
273 #define CHARSET_DIRECTION(cs) ((cs)->direction) | |
274 #define CHARSET_FINAL(cs) ((cs)->final) | |
275 #define CHARSET_DOC_STRING(cs) ((cs)->doc_string) | |
3659 | 276 #define CHARSET_REGISTRIES(cs) ((cs)->registries) |
771 | 277 #define CHARSET_CCL_PROGRAM(cs) ((cs)->ccl_program) |
278 #define CHARSET_DIMENSION(cs) ((cs)->dimension) | |
279 #define CHARSET_CHARS(cs) ((cs)->chars) | |
280 #define CHARSET_REVERSE_DIRECTION_CHARSET(cs) ((cs)->reverse_direction_charset) | |
281 #define CHARSET_TO_UNICODE_TABLE(cs) ((cs)->to_unicode_table) | |
282 #define CHARSET_FROM_UNICODE_TABLE(cs) ((cs)->from_unicode_table) | |
283 #define CHARSET_FROM_UNICODE_LEVELS(cs) ((cs)->from_unicode_levels) | |
284 | |
826 | 285 #define CHARSET_PRIVATE_P(cs) leading_byte_private_p (CHARSET_LEADING_BYTE (cs)) |
771 | 286 |
287 #define XCHARSET_ID(cs) CHARSET_ID (XCHARSET (cs)) | |
288 #define XCHARSET_NAME(cs) CHARSET_NAME (XCHARSET (cs)) | |
289 #define XCHARSET_SHORT_NAME(cs) CHARSET_SHORT_NAME (XCHARSET (cs)) | |
290 #define XCHARSET_LONG_NAME(cs) CHARSET_LONG_NAME (XCHARSET (cs)) | |
291 #define XCHARSET_REP_BYTES(cs) CHARSET_REP_BYTES (XCHARSET (cs)) | |
292 #define XCHARSET_COLUMNS(cs) CHARSET_COLUMNS (XCHARSET (cs)) | |
293 #define XCHARSET_GRAPHIC(cs) CHARSET_GRAPHIC (XCHARSET (cs)) | |
3439 | 294 #define XCHARSET_ENCODE_AS_UTF_8(cs) CHARSET_ENCODE_AS_UTF_8 (XCHARSET (cs)) |
771 | 295 #define XCHARSET_TYPE(cs) CHARSET_TYPE (XCHARSET (cs)) |
296 #define XCHARSET_DIRECTION(cs) CHARSET_DIRECTION (XCHARSET (cs)) | |
297 #define XCHARSET_FINAL(cs) CHARSET_FINAL (XCHARSET (cs)) | |
298 #define XCHARSET_DOC_STRING(cs) CHARSET_DOC_STRING (XCHARSET (cs)) | |
3659 | 299 #define XCHARSET_REGISTRIES(cs) CHARSET_REGISTRIES (XCHARSET (cs)) |
771 | 300 #define XCHARSET_LEADING_BYTE(cs) CHARSET_LEADING_BYTE (XCHARSET (cs)) |
301 #define XCHARSET_CCL_PROGRAM(cs) CHARSET_CCL_PROGRAM (XCHARSET (cs)) | |
302 #define XCHARSET_DIMENSION(cs) CHARSET_DIMENSION (XCHARSET (cs)) | |
303 #define XCHARSET_CHARS(cs) CHARSET_CHARS (XCHARSET (cs)) | |
3659 | 304 |
771 | 305 #define XCHARSET_PRIVATE_P(cs) CHARSET_PRIVATE_P (XCHARSET (cs)) |
306 #define XCHARSET_REVERSE_DIRECTION_CHARSET(cs) \ | |
307 CHARSET_REVERSE_DIRECTION_CHARSET (XCHARSET (cs)) | |
308 #define XCHARSET_TO_UNICODE_TABLE(cs) \ | |
309 CHARSET_TO_UNICODE_TABLE (XCHARSET (cs)) | |
310 #define XCHARSET_FROM_UNICODE_TABLE(cs) \ | |
311 CHARSET_FROM_UNICODE_TABLE (XCHARSET (cs)) | |
312 #define XCHARSET_FROM_UNICODE_LEVELS(cs) \ | |
313 CHARSET_FROM_UNICODE_LEVELS (XCHARSET (cs)) | |
314 | |
315 struct charset_lookup | |
316 { | |
317 /* Table of charsets indexed by leading byte. */ | |
318 Lisp_Object charset_by_leading_byte[NUM_LEADING_BYTES]; | |
319 | |
320 /* Table of charsets indexed by type/final-byte/direction. */ | |
321 Lisp_Object charset_by_attributes[4][128][2]; | |
867 | 322 Ibyte next_allocated_1_byte_leading_byte; |
323 Ibyte next_allocated_2_byte_leading_byte; | |
771 | 324 }; |
325 | |
1111 | 326 extern struct charset_lookup *chlook; |
327 | |
826 | 328 DECLARE_INLINE_HEADER ( |
329 Lisp_Object | |
330 charset_by_leading_byte (int lb) | |
331 ) | |
771 | 332 { |
800 | 333 #ifdef ERROR_CHECK_TEXT |
771 | 334 /* When error-checking is on, x86 GCC 2.95.2 -O3 miscompiles the |
335 following unless we introduce `tem'. */ | |
336 int tem = lb; | |
800 | 337 text_checking_assert (tem >= MIN_LEADING_BYTE && tem <= 0xFF); |
771 | 338 #endif |
339 return chlook->charset_by_leading_byte[lb - MIN_LEADING_BYTE]; | |
340 } | |
341 | |
826 | 342 DECLARE_INLINE_HEADER ( |
343 Lisp_Object | |
344 charset_by_attributes (int type, int final, int dir) | |
345 ) | |
771 | 346 { |
347 type_checking_assert (type < countof (chlook->charset_by_attributes) && | |
348 final < countof (chlook->charset_by_attributes[0]) && | |
349 dir < countof (chlook->charset_by_attributes[0][0])); | |
350 return chlook->charset_by_attributes[type][final][dir]; | |
351 } | |
352 | |
353 | |
354 /************************************************************************/ | |
355 /* Dealing with characters */ | |
356 /************************************************************************/ | |
357 | |
358 /* The bit fields of character are divided into 3 parts: | |
3496 | 359 FIELD1(7bits):FIELD2(7bits):FIELD3(7bits) */ |
771 | 360 |
3496 | 361 #define ICHAR_FIELD1_MASK (0x7F << 14) |
867 | 362 #define ICHAR_FIELD2_MASK (0x7F << 7) |
363 #define ICHAR_FIELD3_MASK 0x7F | |
771 | 364 |
365 /* Macros to access each field of a character code of C. */ | |
366 | |
867 | 367 #define ichar_field1(c) (((c) & ICHAR_FIELD1_MASK) >> 14) |
368 #define ichar_field2(c) (((c) & ICHAR_FIELD2_MASK) >> 7) | |
369 #define ichar_field3(c) ((c) & ICHAR_FIELD3_MASK) | |
771 | 370 |
371 /* Field 1, if non-zero, usually holds a leading byte for a | |
372 dimension-2 charset. Field 2, if non-zero, usually holds a leading | |
373 byte for a dimension-1 charset. */ | |
374 | |
375 /* Converting between field values and leading bytes. */ | |
376 | |
377 #define FIELD2_TO_OFFICIAL_LEADING_BYTE (MIN_LEADING_BYTE_OFFICIAL_1 - 1) | |
378 #define FIELD2_TO_PRIVATE_LEADING_BYTE 0x80 | |
379 | |
380 #define FIELD1_TO_OFFICIAL_LEADING_BYTE (MIN_LEADING_BYTE_OFFICIAL_2 - 1) | |
3496 | 381 #define FIELD1_TO_PRIVATE_LEADING_BYTE 0x80 |
771 | 382 |
383 /* Minimum and maximum allowed values for the fields. */ | |
384 | |
867 | 385 #define MIN_ICHAR_FIELD2_OFFICIAL \ |
771 | 386 (MIN_LEADING_BYTE_OFFICIAL_1 - FIELD2_TO_OFFICIAL_LEADING_BYTE) |
867 | 387 #define MAX_ICHAR_FIELD2_OFFICIAL \ |
771 | 388 (MAX_LEADING_BYTE_OFFICIAL_1 - FIELD2_TO_OFFICIAL_LEADING_BYTE) |
389 | |
867 | 390 #define MIN_ICHAR_FIELD1_OFFICIAL \ |
771 | 391 (MIN_LEADING_BYTE_OFFICIAL_2 - FIELD1_TO_OFFICIAL_LEADING_BYTE) |
867 | 392 #define MAX_ICHAR_FIELD1_OFFICIAL \ |
771 | 393 (MAX_LEADING_BYTE_OFFICIAL_2 - FIELD1_TO_OFFICIAL_LEADING_BYTE) |
394 | |
867 | 395 #define MIN_ICHAR_FIELD2_PRIVATE \ |
771 | 396 (MIN_LEADING_BYTE_PRIVATE_1 - FIELD2_TO_PRIVATE_LEADING_BYTE) |
867 | 397 #define MAX_ICHAR_FIELD2_PRIVATE \ |
771 | 398 (MAX_LEADING_BYTE_PRIVATE_1 - FIELD2_TO_PRIVATE_LEADING_BYTE) |
399 | |
867 | 400 #define MIN_ICHAR_FIELD1_PRIVATE \ |
771 | 401 (MIN_LEADING_BYTE_PRIVATE_2 - FIELD1_TO_PRIVATE_LEADING_BYTE) |
867 | 402 #define MAX_ICHAR_FIELD1_PRIVATE \ |
771 | 403 (MAX_LEADING_BYTE_PRIVATE_2 - FIELD1_TO_PRIVATE_LEADING_BYTE) |
404 | |
405 /* Minimum character code of each <type> character. */ | |
406 | |
867 | 407 #define MIN_CHAR_OFFICIAL_TYPE9N (MIN_ICHAR_FIELD2_OFFICIAL << 7) |
408 #define MIN_CHAR_PRIVATE_TYPE9N (MIN_ICHAR_FIELD2_PRIVATE << 7) | |
409 #define MIN_CHAR_OFFICIAL_TYPE9NX9N (MIN_ICHAR_FIELD1_OFFICIAL << 14) | |
410 #define MIN_CHAR_PRIVATE_TYPE9NX9N (MIN_ICHAR_FIELD1_PRIVATE << 14) | |
3496 | 411 #define MIN_CHAR_COMPOSITION (0x7F << 14) |
771 | 412 |
413 /* Leading byte of a character. | |
414 | |
415 NOTE: This takes advantage of the fact that | |
416 FIELD2_TO_OFFICIAL_LEADING_BYTE and | |
417 FIELD2_TO_PRIVATE_LEADING_BYTE are the same. | |
418 */ | |
419 | |
826 | 420 DECLARE_INLINE_HEADER ( |
867 | 421 Ibyte |
422 ichar_leading_byte (Ichar c) | |
826 | 423 ) |
771 | 424 { |
867 | 425 if (ichar_ascii_p (c)) |
771 | 426 return LEADING_BYTE_ASCII; |
427 else if (c < 0xA0) | |
428 return LEADING_BYTE_CONTROL_1; | |
429 else if (c < MIN_CHAR_OFFICIAL_TYPE9NX9N) | |
867 | 430 return ichar_field2 (c) + FIELD2_TO_OFFICIAL_LEADING_BYTE; |
771 | 431 else if (c < MIN_CHAR_PRIVATE_TYPE9NX9N) |
867 | 432 return ichar_field1 (c) + FIELD1_TO_OFFICIAL_LEADING_BYTE; |
771 | 433 else if (c < MIN_CHAR_COMPOSITION) |
867 | 434 return ichar_field1 (c) + FIELD1_TO_PRIVATE_LEADING_BYTE; |
771 | 435 else |
436 { | |
437 #ifdef ENABLE_COMPOSITE_CHARS | |
438 return LEADING_BYTE_COMPOSITE; | |
439 #else | |
2500 | 440 ABORT(); |
771 | 441 return 0; |
442 #endif /* ENABLE_COMPOSITE_CHARS */ | |
443 } | |
444 } | |
445 | |
826 | 446 DECLARE_INLINE_HEADER ( |
447 Bytecount | |
867 | 448 ichar_len (Ichar c) |
826 | 449 ) |
450 { | |
867 | 451 if (ichar_ascii_p (c)) |
826 | 452 return 1; |
453 else if (c < MIN_CHAR_OFFICIAL_TYPE9NX9N) | |
454 return 2; | |
455 else if (c < MIN_CHAR_PRIVATE_TYPE9NX9N) | |
456 return 3; /* dimension-2 official or dimension-1 private */ | |
457 else if (c < MIN_CHAR_COMPOSITION) | |
458 return 4; | |
459 else | |
460 { | |
461 #ifdef ENABLE_COMPOSITE_CHARS | |
462 #error Not yet implemented | |
463 #else | |
2500 | 464 ABORT(); |
826 | 465 return 0; |
466 #endif /* ENABLE_COMPOSITE_CHARS */ | |
467 } | |
468 } | |
469 | |
470 DECLARE_INLINE_HEADER ( | |
471 Bytecount | |
867 | 472 ichar_len_fmt (Ichar c, Internal_Format fmt) |
826 | 473 ) |
474 { | |
475 switch (fmt) | |
476 { | |
477 case FORMAT_DEFAULT: | |
867 | 478 return ichar_len (c); |
826 | 479 case FORMAT_16_BIT_FIXED: |
480 return 2; | |
481 case FORMAT_32_BIT_FIXED: | |
482 return 4; | |
483 default: | |
484 text_checking_assert (fmt == FORMAT_8_BIT_FIXED); | |
485 return 1; | |
486 } | |
487 } | |
488 | |
867 | 489 #define ichar_charset(c) charset_by_leading_byte (ichar_leading_byte (c)) |
771 | 490 |
491 /* Return a character whose charset is CHARSET and position-codes are C1 | |
492 and C2. TYPE9N character ignores C2. (For typical charsets, i.e. not | |
493 ASCII, Control-1 or Composite, C1 and C2 will be in the range of 32 to | |
494 127 or 33 to 126. See `make-char'.) | |
495 | |
496 NOTE: This takes advantage of the fact that | |
497 FIELD2_TO_OFFICIAL_LEADING_BYTE and | |
498 FIELD2_TO_PRIVATE_LEADING_BYTE are the same. | |
499 */ | |
500 | |
826 | 501 DECLARE_INLINE_HEADER ( |
867 | 502 Ichar |
503 make_ichar (Lisp_Object charset, int c1, int c2) | |
826 | 504 ) |
771 | 505 { |
867 | 506 Ichar retval; |
771 | 507 if (EQ (charset, Vcharset_ascii)) |
826 | 508 retval = c1; |
771 | 509 else if (EQ (charset, Vcharset_control_1)) |
826 | 510 retval = c1 | 0x80; |
771 | 511 #ifdef ENABLE_COMPOSITE_CHARS |
512 else if (EQ (charset, Vcharset_composite)) | |
826 | 513 retval = (0x1F << 14) | ((c1) << 7) | (c2); |
771 | 514 #endif |
515 else if (XCHARSET_DIMENSION (charset) == 1) | |
826 | 516 retval = ((XCHARSET_LEADING_BYTE (charset) - |
517 FIELD2_TO_OFFICIAL_LEADING_BYTE) << 7) | (c1); | |
771 | 518 else if (!XCHARSET_PRIVATE_P (charset)) |
826 | 519 retval = ((XCHARSET_LEADING_BYTE (charset) - |
520 FIELD1_TO_OFFICIAL_LEADING_BYTE) << 14) | ((c1) << 7) | (c2); | |
771 | 521 else |
826 | 522 retval = ((XCHARSET_LEADING_BYTE (charset) - |
523 FIELD1_TO_PRIVATE_LEADING_BYTE) << 14) | ((c1) << 7) | (c2); | |
867 | 524 text_checking_assert (valid_ichar_p (retval)); |
826 | 525 return retval; |
771 | 526 } |
527 | |
867 | 528 /* BREAKUP_ICHAR_1_UNSAFE assumes that the charset has already been |
771 | 529 calculated, and just computes c1 and c2. |
530 | |
867 | 531 BREAKUP_ICHAR also computes and stores the charset. */ |
771 | 532 |
867 | 533 #define BREAKUP_ICHAR_1_UNSAFE(c, charset, c1, c2) \ |
771 | 534 XCHARSET_DIMENSION (charset) == 1 \ |
867 | 535 ? ((c1) = ichar_field3 (c), (c2) = 0) \ |
536 : ((c1) = ichar_field2 (c), \ | |
537 (c2) = ichar_field3 (c)) | |
771 | 538 |
826 | 539 DECLARE_INLINE_HEADER ( |
540 void | |
867 | 541 breakup_ichar_1 (Ichar c, Lisp_Object *charset, int *c1, int *c2) |
826 | 542 ) |
771 | 543 { |
867 | 544 text_checking_assert (valid_ichar_p (c)); |
545 *charset = ichar_charset (c); | |
546 BREAKUP_ICHAR_1_UNSAFE (c, *charset, *c1, *c2); | |
771 | 547 } |
548 | |
867 | 549 /* BREAKUP_ICHAR separates an Ichar into its components. The charset of |
771 | 550 character C is set to CHARSET, and the position-codes of C are set to C1 |
551 and C2. C2 of TYPE9N character is 0. */ | |
552 | |
867 | 553 #define BREAKUP_ICHAR(c, charset, c1, c2) \ |
554 breakup_ichar_1 (c, &(charset), &(c1), &(c2)) | |
771 | 555 |
5200
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
556 /* Forward compatibility from ben-unicode-internal: Convert a charset |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
557 codepoint into a character in the internal string representation. |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
558 Return number of bytes written out. FAIL controls failure mode when |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
559 charset conversion to Unicode is not possible (unused as of yet). */ |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
560 DECLARE_INLINE_HEADER ( |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
561 Bytecount |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
562 charset_codepoint_to_itext (Lisp_Object charset, int c1, int c2, Ibyte *ptr, |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
563 enum converr UNUSED (fail)) |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
564 ) |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
565 { |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
566 Ichar ch; |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
567 |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
568 if (EQ (charset, Vcharset_ascii)) |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
569 { |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
570 ptr[0] = (Ibyte) c2; |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
571 return 1; |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
572 } |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
573 |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
574 ch = make_ichar (charset, c1, c2); |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
575 |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
576 /* We can't rely on the converted character being non-ASCII. For |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
577 example, JISX0208 codepoint (33, 64) == Unicode 0x5C (ASCII |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
578 backslash). */ |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
579 return set_itext_ichar (ptr, ch); |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
580 } |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
581 |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
582 /* Forward compatibility from ben-unicode-internal */ |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
583 |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
584 DECLARE_INLINE_HEADER ( |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
585 void |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
586 buffer_itext_to_charset_codepoint (const Ibyte *ptr, |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
587 struct buffer *UNUSED (buf), |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
588 Lisp_Object *charset, int *c1, int *c2, |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
589 enum converr UNUSED (fail)) |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
590 ) |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
591 { |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
592 Ichar ch = itext_ichar (ptr); |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
593 breakup_ichar_1 (ch, charset, c1, c2); |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
594 } |
70ed8a0d8da8
port Mule-ization of mule-wnnfns.c from ben-unicode-internal
Ben Wing <ben@xemacs.org>
parents:
5127
diff
changeset
|
595 |
788 | 596 void get_charset_limits (Lisp_Object charset, int *low, int *high); |
867 | 597 int ichar_to_unicode (Ichar chr); |
788 | 598 |
3439 | 599 EXFUN (Fcharset_name, 1); |
600 | |
771 | 601 #endif /* MULE */ |
602 | |
3439 | 603 /* ISO 10646 UTF-16, UCS-4, UTF-8, UTF-7, etc. */ |
604 | |
605 enum unicode_type | |
606 { | |
607 UNICODE_UTF_16, | |
608 UNICODE_UTF_8, | |
609 UNICODE_UTF_7, | |
4096 | 610 UNICODE_UCS_4, |
611 UNICODE_UTF_32 | |
3439 | 612 }; |
613 | |
614 void encode_unicode_char (Lisp_Object USED_IF_MULE (charset), int h, | |
615 int USED_IF_MULE (l), unsigned_char_dynarr *dst, | |
4096 | 616 enum unicode_type type, unsigned int little_endian, |
617 int write_error_characters_as_such); | |
618 | |
619 #define UNICODE_ERROR_OCTET_RANGE_START 0x200000 | |
620 | |
621 #define valid_utf_16_first_surrogate(ch) (((ch) & 0xFC00) == 0xD800) | |
622 #define valid_utf_16_last_surrogate(ch) (((ch) & 0xFC00) == 0xDC00) | |
623 #define valid_utf_16_surrogate(ch) (((ch) & 0xF800) == 0xD800) | |
3439 | 624 |
3676 | 625 void set_charset_registries(Lisp_Object charset, Lisp_Object registries); |
626 | |
3439 | 627 EXFUN (Funicode_to_char, 2); |
628 EXFUN (Fchar_to_unicode, 1); | |
629 | |
771 | 630 #endif /* INCLUDED_charset_h_ */ |