1 /* Header for multilingual functions.
2 Copyright (C) 1992, 1995 Free Software Foundation, Inc.
3 Copyright (C) 1995 Sun Microsystems, Inc.
5 This file is part of XEmacs.
7 XEmacs is free software; you can redistribute it and/or modify it
8 under the terms of the GNU General Public License as published by the
9 Free Software Foundation; either version 2, or (at your option) any
12 XEmacs is distributed in the hope that it will be useful, but WITHOUT
13 ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
14 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
17 You should have received a copy of the GNU General Public License
18 along with XEmacs; see the file COPYING. If not, write to
19 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
20 Boston, MA 02111-1307, USA. */
22 /* Synched up with: Mule 2.3. Not in FSF. */
24 /* Rewritten by Ben Wing <ben@xemacs.org>. */
26 #ifndef INCLUDED_mule_charset_h_
27 #define INCLUDED_mule_charset_h_
33 A character set (or "charset") is an ordered set of characters.
34 A particular character in a charset is indexed using one or
35 more "position codes", which are non-negative integers.
36 The number of position codes needed to identify a particular
37 character in a charset is called the "dimension" of the
38 charset. In XEmacs/Mule, all charsets have 1 or 2 dimensions,
39 and the size of all charsets (except for a few special cases)
40 is either 94, 96, 94 by 94, or 96 by 96. The range of
41 position codes used to index characters from any of these
42 types of character sets is as follows:
44 Charset type Position code 1 Position code 2
45 ------------------------------------------------------------
48 94x94 33 - 126 33 - 126
49 96x96 32 - 127 32 - 127
51 Note that in the above cases position codes do not start at
52 an expected value such as 0 or 1. The reason for this will
55 For example, Latin-1 is a 96-character charset, and JISX0208
56 (the Japanese national character set) is a 94x94-character
59 [Note that, although the ranges above define the *valid*
60 position codes for a charset, some of the slots in a particular
61 charset may in fact be empty. This is the case for JISX0208,
62 for example, where (e.g.) all the slots whose first
63 position code is in the range 118 - 127 are empty.]
65 There are three charsets that do not follow the above rules.
66 All of them have one dimension, and have ranges of position
69 Charset name Position code 1
70 ------------------------------------
73 Composite 0 - some large number
75 (The upper bound of the position code for composite characters
76 has not yet been determined, but it will probably be at
79 ASCII is the union of two subsidiary character sets:
80 Printing-ASCII (the printing ASCII character set,
81 consisting of position codes 33 - 126, like for a standard
82 94-character charset) and Control-ASCII (the non-printing
83 characters that would appear in a binary file with codes 0
86 Control-1 contains the non-printing characters that would
87 appear in a binary file with codes 128 - 159.
89 Composite contains characters that are generated by
90 overstriking one or more characters from other charsets.
92 Note that some characters in ASCII, and all characters
93 in Control-1, are "control" (non-printing) characters.
94 These have no printed representation but instead control
95 some other function of the printing (e.g. TAB or 8 moves
96 the current character position to the next tab stop).
97 All other characters in all charsets are "graphic"
98 (printing) characters.
100 When a binary file is read in, the bytes in the file are
101 assigned to character sets as follows:
103 Bytes Character set Range
104 --------------------------------------------------
105 0 - 127 ASCII 0 - 127
106 128 - 159 Control-1 0 - 31
107 160 - 255 Latin-1 32 - 127
109 This is a bit ad-hoc but gets the job done.
114 An "encoding" is a way of numerically representing
115 characters from one or more character sets. If an encoding
116 only encompasses one character set, then the position codes
117 for the characters in that character set could be used
118 directly. This is not possible, however, if more than one
119 character set is to be used in the encoding.
121 For example, the conversion detailed above between bytes in
122 a binary file and characters is effectively an encoding
123 that encompasses the three character sets ASCII, Control-1,
124 and Latin-1 in a stream of 8-bit bytes.
126 Thus, an encoding can be viewed as a way of encoding
127 characters from a specified group of character sets using a
128 stream of bytes, each of which contains a fixed number of
129 bits (but not necessarily 8, as in the common usage of
132 Here are descriptions of a couple of common
136 A. Japanese EUC (Extended Unix Code)
138 This encompasses the character sets:
140 - Katakana-JISX0201 (half-width katakana, the right half of JISX0201).
145 Note that Printing-ASCII and Katakana-JISX0201 are 94-character
146 charsets, while Japanese-JISX0208 is a 94x94-character charset.
148 The encoding is as follows:
150 Character set Representation (PC == position-code)
151 ------------- --------------
153 Japanese-JISX0208 PC1 + 0x80 | PC2 + 0x80
154 Katakana-JISX0201 0x8E | PC1 + 0x80
159 This encompasses the character sets:
161 - Latin-JISX0201 (the left half of JISX0201; this character set is
162 very similar to Printing-ASCII and is a 94-character charset)
167 Unlike Japanese EUC, this is a "modal" encoding, which
168 means that there are multiple states that the encoding can
169 be in, which affect how the bytes are to be interpreted.
170 Special sequences of bytes (called "escape sequences")
171 are used to change states.
173 The encoding is as follows:
175 Character set Representation
176 ------------- --------------
179 Katakana-JISX0201 PC1
180 Japanese-JISX0208 PC1 | PC2
182 Escape sequence ASCII equivalent Meaning
183 --------------- ---------------- -------
184 0x1B 0x28 0x42 ESC ( B invoke Printing-ASCII
185 0x1B 0x28 0x4A ESC ( J invoke Latin-JISX0201
186 0x1B 0x28 0x49 ESC ( I invoke Katakana-JISX0201
187 0x1B 0x24 0x42 ESC $ B invoke Japanese-JISX0208
189 Initially, Printing-ASCII is invoked.
191 3. Internal Mule Encodings
192 ==========================
194 In XEmacs/Mule, each character set is assigned a unique number,
195 called a "leading byte". This is used in the encodings of a
196 character. Leading bytes are in the range 0x80 - 0xFF
197 (except for ASCII, which has a leading byte of 0), although
198 some leading bytes are reserved.
200 Charsets whose leading byte is in the range 0x80 - 0x9F are
201 called "official" and are used for built-in charsets.
202 Other charsets are called "private" and have leading bytes
203 in the range 0xA0 - 0xFF; these are user-defined charsets.
207 Character set Leading byte
208 ------------- ------------
211 Dimension-1 Official 0x81 - 0x8D
214 Dimension-2 Official 0x90 - 0x99
215 (0x9A - 0x9D are free;
216 0x9E and 0x9F are reserved)
217 Dimension-1 Private 0xA0 - 0xEF
218 Dimension-2 Private 0xF0 - 0xFF
220 There are two internal encodings for characters in XEmacs/Mule.
221 One is called "string encoding" and is an 8-bit encoding that
222 is used for representing characters in a buffer or string.
223 It uses 1 to 4 bytes per character. The other is called
224 "character encoding" and is a 19-bit encoding that is used
225 for representing characters individually in a variable.
227 (In the following descriptions, we'll ignore composite
228 characters for the moment. We also give a general (structural)
229 overview first, followed later by the exact details.)
231 A. Internal String Encoding
233 ASCII characters are encoded using their position code directly.
234 Other characters are encoded using their leading byte followed
235 by their position code(s) with the high bit set. Characters
236 in private character sets have their leading byte prefixed with
237 a "leading byte prefix", which is either 0x9E or 0x9F. (No
238 character sets are ever assigned these leading bytes.) Specifically:
240 Character set Encoding (PC == position-code)
241 ------------- -------- (LB == leading-byte)
243 Control-1 LB | PC1 + 0xA0
244 Dimension-1 official LB | PC1 + 0x80
245 Dimension-1 private 0x9E | LB | PC1 + 0x80
246 Dimension-2 official LB | PC1 | PC2 + 0x80
247 Dimension-2 private 0x9F | LB | PC1 + 0x80 | PC2 + 0x80
249 The basic characteristic of this encoding is that the first byte
250 of all characters is in the range 0x00 - 0x9F, and the second and
251 following bytes of all characters is in the range 0xA0 - 0xFF.
252 This means that it is impossible to get out of sync, or more
255 1. Given any byte position, the beginning of the character it is
256 within can be determined in constant time.
257 2. Given any byte position at the beginning of a character, the
258 beginning of the next character can be determined in constant
260 3. Given any byte position at the beginning of a character, the
261 beginning of the previous character can be determined in constant
263 4. Textual searches can simply treat encoded strings as if they
264 were encoded in a one-byte-per-character fashion rather than
265 the actual multi-byte encoding.
267 None of the standard non-modal encodings meet all of these
268 conditions. For example, EUC satisfies only (2) and (3), while
269 Shift-JIS and Big5 (not yet described) satisfy only (2). (All
270 non-modal encodings must satisfy (2), in order to be unambiguous.)
272 B. Internal Character Encoding
274 One 19-bit word represents a single character. The word is
275 separated into three fields:
277 Bit number: 18 17 16 15 14 13 12 11 10 09 08 07 06 05 04 03 02 01 00
278 <------------> <------------------> <------------------>
281 Note that fields 2 and 3 hold 7 bits each, while field 1 holds 5 bits.
283 Character set Field 1 Field 2 Field 3
284 ------------- ------- ------- -------
289 Dimension-1 official 0 LB - 0x80 PC1
290 range: (01 - 0D) (20 - 7F)
291 Dimension-1 private 0 LB - 0x80 PC1
292 range: (20 - 6F) (20 - 7F)
293 Dimension-2 official LB - 0x8F PC1 PC2
294 range: (01 - 0A) (20 - 7F) (20 - 7F)
295 Dimension-2 private LB - 0xE1 PC1 PC2
296 range: (0F - 1E) (20 - 7F) (20 - 7F)
299 Note that character codes 0 - 255 are the same as the "binary encoding"
304 About Unicode support:
306 Adding Unicode support is very desirable. Unicode will likely be a
307 very common representation in the future, and thus we should
308 represent Unicode characters using three bytes instead of four.
309 This means we need to find leading bytes for Unicode. Given that
310 there are 65,536 characters in Unicode and we can attach 96x96 =
311 9,216 characters per leading byte, we need eight leading bytes for
312 Unicode. We currently have four free (0x9A - 0x9D), and with a
313 little bit of rearranging we can get five: ASCII doesn't really
314 need to take up a leading byte. (We could just as well use 0x7F,
315 with a little change to the functions that assume that 0x80 is the
316 lowest leading byte.) This means we still need to dump three
317 leading bytes and move them into private space. The CNS charsets
318 are good candidates since they are rarely used, and
319 JAPANESE_JISX0208_1978 is becoming less and less used and could
323 /************************************************************************/
324 /* Definition of leading bytes */
325 /************************************************************************/
327 typedef unsigned char Charset_ID;
329 #define MIN_LEADING_BYTE 0x80
330 /* These need special treatment in a string and/or character */
331 #define LEADING_BYTE_ASCII 0x8E /* Omitted in a buffer */
332 #ifdef ENABLE_COMPOSITE_CHARS
334 #define LEADING_BYTE_COMPOSITE 0x80 /* for a composite character */
335 #define LEADING_BYTE_CONTROL_1 0x8F /* represent normal 80-9F */
337 /* Note the gap in each official charset can cause core dump
338 as first and last values are used to determine whether
339 charset is defined or not in non_ascii_valid_char_p */
341 /** The following are for 1-byte characters in an official charset. **/
342 enum LEADING_BYTE_OFFICIAL_1
344 LEADING_BYTE_LATIN_ISO8859_1 = 0x81, /* Right half of ISO 8859-1 */
345 LEADING_BYTE_LATIN_ISO8859_2, /* 0x82 Right half of ISO 8859-2 */
346 LEADING_BYTE_LATIN_ISO8859_3, /* 0x83 Right half of ISO 8859-3 */
347 LEADING_BYTE_LATIN_ISO8859_4, /* 0x84 Right half of ISO 8859-4 */
348 LEADING_BYTE_THAI_TIS620, /* 0x85 TIS620-2533 */
349 LEADING_BYTE_GREEK_ISO8859_7, /* 0x86 Right half of ISO 8859-7 */
350 LEADING_BYTE_ARABIC_ISO8859_6, /* 0x87 Right half of ISO 8859-6 */
351 LEADING_BYTE_HEBREW_ISO8859_8, /* 0x88 Right half of ISO 8859-8 */
352 LEADING_BYTE_KATAKANA_JISX0201, /* 0x89 Right half of JIS X0201-1976 */
353 LEADING_BYTE_LATIN_JISX0201, /* 0x8A Left half of JIS X0201-1976 */
354 LEADING_BYTE_CYRILLIC_ISO8859_5,/* 0x8B Right half of ISO 8859-5 */
355 LEADING_BYTE_LATIN_ISO8859_9 /* 0x8C Right half of ISO 8859-9 */
359 #define MIN_LEADING_BYTE_OFFICIAL_1 LEADING_BYTE_LATIN_ISO8859_1
360 #define MAX_LEADING_BYTE_OFFICIAL_1 LEADING_BYTE_LATIN_ISO8859_9
362 /** The following are for 2-byte characters in an official charset. **/
363 enum LEADING_BYTE_OFFICIAL_2
365 LEADING_BYTE_JAPANESE_JISX0208_1978 = 0x90, /* Japanese JIS X0208-1978 */
366 LEADING_BYTE_CHINESE_GB2312, /* 0x91 Chinese Hanzi GB2312-1980 */
367 LEADING_BYTE_JAPANESE_JISX0208, /* 0x92 Japanese JIS X0208-1983 */
368 LEADING_BYTE_KOREAN_KSC5601, /* 0x93 Hangul KS C5601-1987 */
369 LEADING_BYTE_JAPANESE_JISX0212, /* 0x94 Japanese JIS X0212-1990 */
370 LEADING_BYTE_CHINESE_CNS11643_1, /* 0x95 Chinese CNS11643 Set 1 */
371 LEADING_BYTE_CHINESE_CNS11643_2, /* 0x96 Chinese CNS11643 Set 2 */
372 LEADING_BYTE_CHINESE_BIG5_1, /* 0x97 Big5 Level 1 */
373 LEADING_BYTE_CHINESE_BIG5_2 /* 0x98 Big5 Level 2 */
380 #define MIN_LEADING_BYTE_OFFICIAL_2 LEADING_BYTE_JAPANESE_JISX0208_1978
381 #define MAX_LEADING_BYTE_OFFICIAL_2 LEADING_BYTE_CHINESE_BIG5_2
383 /** The following are for 1- and 2-byte characters in a private charset. **/
385 #define PRE_LEADING_BYTE_PRIVATE_1 0x9E /* 1-byte char-set */
386 #define PRE_LEADING_BYTE_PRIVATE_2 0x9F /* 2-byte char-set */
388 #define MIN_LEADING_BYTE_PRIVATE_1 0xA0
389 #define MAX_LEADING_BYTE_PRIVATE_1 0xEF
390 #define MIN_LEADING_BYTE_PRIVATE_2 0xF0
391 #define MAX_LEADING_BYTE_PRIVATE_2 0xFF
393 #define NUM_LEADING_BYTES 128
396 /************************************************************************/
397 /* Operations on leading bytes */
398 /************************************************************************/
400 /* Is this leading byte for a private charset? */
402 #define LEADING_BYTE_PRIVATE_P(lb) ((lb) >= MIN_LEADING_BYTE_PRIVATE_1)
404 /* Is this a prefix for a private leading byte? */
406 INLINE_HEADER int LEADING_BYTE_PREFIX_P (Bufbyte lb);
408 LEADING_BYTE_PREFIX_P (Bufbyte lb)
410 return (lb == PRE_LEADING_BYTE_PRIVATE_1 ||
411 lb == PRE_LEADING_BYTE_PRIVATE_2);
414 /* Given a private leading byte, return the leading byte prefix stored
417 #define PRIVATE_LEADING_BYTE_PREFIX(lb) \
418 ((unsigned int) (lb) < MIN_LEADING_BYTE_PRIVATE_2 ? \
419 PRE_LEADING_BYTE_PRIVATE_1 : \
420 PRE_LEADING_BYTE_PRIVATE_2)
423 /************************************************************************/
424 /* Operations on individual bytes */
426 /************************************************************************/
428 /* These are carefully designed to work if BYTE is signed or unsigned. */
429 /* Note that SPC and DEL are considered ASCII, not control. */
431 #define BYTE_ASCII_P(byte) (((byte) & ~0x7f) == 0)
432 #define BYTE_C0_P(byte) (((byte) & ~0x1f) == 0)
433 #define BYTE_C1_P(byte) (((byte) & ~0x1f) == 0x80)
436 /************************************************************************/
437 /* Operations on individual bytes */
438 /* in a Mule-formatted string */
439 /************************************************************************/
441 /* Does BYTE represent the first byte of a character? */
443 #define BUFBYTE_FIRST_BYTE_P(byte) ((byte) < 0xA0)
445 /* Does BYTE represent the first byte of a multi-byte character? */
447 #define BUFBYTE_LEADING_BYTE_P(byte) BYTE_C1_P (byte)
450 /************************************************************************/
451 /* Information about a particular character set */
452 /************************************************************************/
456 struct lcrecord_header header;
460 Lisp_Object doc_string;
461 Lisp_Object registry;
462 Lisp_Object short_name;
463 Lisp_Object long_name;
465 Lisp_Object reverse_direction_charset;
467 Lisp_Object ccl_program;
469 /* Final byte of this character set in ISO2022 designating escape sequence */
472 /* Number of bytes (1 - 4) required in the internal representation
473 for characters in this character set. This is *not* the
474 same as the dimension of the character set). */
475 unsigned int rep_bytes;
477 /* Number of columns a character in this charset takes up, on TTY
478 devices. Not used for X devices. */
479 unsigned int columns;
481 /* Direction of this character set */
482 unsigned int direction;
484 /* Type of this character set (94, 96, 94x94, 96x96) */
487 /* Number of bytes used in encoding of this character set (1 or 2) */
488 unsigned int dimension;
490 /* Number of chars in each dimension (usually 94 or 96) */
493 /* Which half of font to be used to display this character set */
494 unsigned int graphic;
496 typedef struct Lisp_Charset Lisp_Charset;
498 DECLARE_LRECORD (charset, Lisp_Charset);
499 #define XCHARSET(x) XRECORD (x, charset, Lisp_Charset)
500 #define XSETCHARSET(x, p) XSETRECORD (x, p, charset)
501 #define CHARSETP(x) RECORDP (x, charset)
502 #define CHECK_CHARSET(x) CHECK_RECORD (x, charset)
503 #define CONCHECK_CHARSET(x) CONCHECK_RECORD (x, charset)
505 #define CHARSET_TYPE_94 0 /* This charset includes 94 characters. */
506 #define CHARSET_TYPE_96 1 /* This charset includes 96 characters. */
507 #define CHARSET_TYPE_94X94 2 /* This charset includes 94x94 characters. */
508 #define CHARSET_TYPE_96X96 3 /* This charset includes 96x96 characters. */
510 #define CHARSET_LEFT_TO_RIGHT 0
511 #define CHARSET_RIGHT_TO_LEFT 1
513 /* Leading byte and id have been regrouped. -- OG */
514 #define CHARSET_ID(cs) ((cs)->id)
515 #define CHARSET_LEADING_BYTE(cs) ((Bufbyte) CHARSET_ID(cs))
516 #define CHARSET_NAME(cs) ((cs)->name)
517 #define CHARSET_SHORT_NAME(cs) ((cs)->short_name)
518 #define CHARSET_LONG_NAME(cs) ((cs)->long_name)
519 #define CHARSET_REP_BYTES(cs) ((cs)->rep_bytes)
520 #define CHARSET_COLUMNS(cs) ((cs)->columns)
521 #define CHARSET_GRAPHIC(cs) ((cs)->graphic)
522 #define CHARSET_TYPE(cs) ((cs)->type)
523 #define CHARSET_DIRECTION(cs) ((cs)->direction)
524 #define CHARSET_FINAL(cs) ((cs)->final)
525 #define CHARSET_DOC_STRING(cs) ((cs)->doc_string)
526 #define CHARSET_REGISTRY(cs) ((cs)->registry)
527 #define CHARSET_CCL_PROGRAM(cs) ((cs)->ccl_program)
528 #define CHARSET_DIMENSION(cs) ((cs)->dimension)
529 #define CHARSET_CHARS(cs) ((cs)->chars)
530 #define CHARSET_REVERSE_DIRECTION_CHARSET(cs) ((cs)->reverse_direction_charset)
533 #define CHARSET_PRIVATE_P(cs) LEADING_BYTE_PRIVATE_P (CHARSET_LEADING_BYTE (cs))
535 #define XCHARSET_ID(cs) CHARSET_ID (XCHARSET (cs))
536 #define XCHARSET_NAME(cs) CHARSET_NAME (XCHARSET (cs))
537 #define XCHARSET_SHORT_NAME(cs) CHARSET_SHORT_NAME (XCHARSET (cs))
538 #define XCHARSET_LONG_NAME(cs) CHARSET_LONG_NAME (XCHARSET (cs))
539 #define XCHARSET_REP_BYTES(cs) CHARSET_REP_BYTES (XCHARSET (cs))
540 #define XCHARSET_COLUMNS(cs) CHARSET_COLUMNS (XCHARSET (cs))
541 #define XCHARSET_GRAPHIC(cs) CHARSET_GRAPHIC (XCHARSET (cs))
542 #define XCHARSET_TYPE(cs) CHARSET_TYPE (XCHARSET (cs))
543 #define XCHARSET_DIRECTION(cs) CHARSET_DIRECTION (XCHARSET (cs))
544 #define XCHARSET_FINAL(cs) CHARSET_FINAL (XCHARSET (cs))
545 #define XCHARSET_DOC_STRING(cs) CHARSET_DOC_STRING (XCHARSET (cs))
546 #define XCHARSET_REGISTRY(cs) CHARSET_REGISTRY (XCHARSET (cs))
547 #define XCHARSET_LEADING_BYTE(cs) CHARSET_LEADING_BYTE (XCHARSET (cs))
548 #define XCHARSET_CCL_PROGRAM(cs) CHARSET_CCL_PROGRAM (XCHARSET (cs))
549 #define XCHARSET_DIMENSION(cs) CHARSET_DIMENSION (XCHARSET (cs))
550 #define XCHARSET_CHARS(cs) CHARSET_CHARS (XCHARSET (cs))
551 #define XCHARSET_PRIVATE_P(cs) CHARSET_PRIVATE_P (XCHARSET (cs))
552 #define XCHARSET_REVERSE_DIRECTION_CHARSET(cs) \
553 CHARSET_REVERSE_DIRECTION_CHARSET (XCHARSET (cs))
555 struct charset_lookup {
556 /* Table of charsets indexed by (leading byte - MIN_LEADING_BYTE). */
557 Lisp_Object charset_by_leading_byte[NUM_LEADING_BYTES];
559 /* Table of charsets indexed by type/final-byte/direction. */
560 Lisp_Object charset_by_attributes[4][128][2];
562 Charset_ID next_allocated_1_byte_leading_byte;
563 Charset_ID next_allocated_2_byte_leading_byte;
566 INLINE_HEADER Lisp_Object CHARSET_BY_LEADING_BYTE (Bufbyte lb);
567 INLINE_HEADER Lisp_Object
568 CHARSET_BY_LEADING_BYTE (Bufbyte lb)
570 extern struct charset_lookup *chlook;
572 #ifdef ERROR_CHECK_TYPECHECK
573 /* When error-checking is on, x86 GCC 2.95.2 -O3 miscompiles the
574 following unless we introduce `tem'. */
576 type_checking_assert (tem >= MIN_LEADING_BYTE &&
577 tem < (MIN_LEADING_BYTE + NUM_LEADING_BYTES));
579 return chlook->charset_by_leading_byte[lb - MIN_LEADING_BYTE];
582 INLINE_HEADER Lisp_Object
583 CHARSET_BY_ATTRIBUTES (unsigned int type, unsigned char final, int dir);
584 INLINE_HEADER Lisp_Object
585 CHARSET_BY_ATTRIBUTES (unsigned int type, unsigned char final, int dir)
587 extern struct charset_lookup *chlook;
589 type_checking_assert (type < countof (chlook->charset_by_attributes) &&
590 final < countof (chlook->charset_by_attributes[0]) &&
591 dir < countof (chlook->charset_by_attributes[0][0]));
592 return chlook->charset_by_attributes[type][final][dir];
595 /* Table of number of bytes in the string representation of a character
596 indexed by the first byte of that representation.
598 This value can be derived in other ways -- e.g. something like
599 XCHARSET_REP_BYTES (CHARSET_BY_LEADING_BYTE (first_byte))
600 but it's faster this way. */
601 extern const Bytecount rep_bytes_by_first_byte[0xA0];
603 /* Number of bytes in the string representation of a character. */
604 INLINE_HEADER int REP_BYTES_BY_FIRST_BYTE (Bufbyte fb);
606 REP_BYTES_BY_FIRST_BYTE (Bufbyte fb)
608 type_checking_assert (fb < 0xA0);
609 return rep_bytes_by_first_byte[fb];
613 /************************************************************************/
614 /* Dealing with characters */
615 /************************************************************************/
617 #define CHAR_ASCII_P(ch) ((ch) <= 0x7F)
619 /* The bit fields of character are divided into 3 parts:
620 FIELD1(5bits):FIELD2(7bits):FIELD3(7bits) */
622 #define CHAR_FIELD1_MASK (0x1F << 14)
623 #define CHAR_FIELD2_MASK (0x7F << 7)
624 #define CHAR_FIELD3_MASK 0x7F
626 /* Macros to access each field of a character code of C. */
628 #define CHAR_FIELD1(c) (((c) & CHAR_FIELD1_MASK) >> 14)
629 #define CHAR_FIELD2(c) (((c) & CHAR_FIELD2_MASK) >> 7)
630 #define CHAR_FIELD3(c) ((c) & CHAR_FIELD3_MASK)
632 /* Field 1, if non-zero, usually holds a leading byte for a
633 dimension-2 charset. Field 2, if non-zero, usually holds a leading
634 byte for a dimension-1 charset. */
636 /* Converting between field values and leading bytes. */
638 #define FIELD2_TO_OFFICIAL_LEADING_BYTE 0x80
639 #define FIELD2_TO_PRIVATE_LEADING_BYTE 0x80
641 #define FIELD1_TO_OFFICIAL_LEADING_BYTE 0x8F
642 #define FIELD1_TO_PRIVATE_LEADING_BYTE 0xE1
644 /* Minimum and maximum allowed values for the fields. */
646 #define MIN_CHAR_FIELD2_OFFICIAL \
647 (MIN_LEADING_BYTE_OFFICIAL_1 - FIELD2_TO_OFFICIAL_LEADING_BYTE)
648 #define MAX_CHAR_FIELD2_OFFICIAL \
649 (MAX_LEADING_BYTE_OFFICIAL_1 - FIELD2_TO_OFFICIAL_LEADING_BYTE)
651 #define MIN_CHAR_FIELD1_OFFICIAL \
652 (MIN_LEADING_BYTE_OFFICIAL_2 - FIELD1_TO_OFFICIAL_LEADING_BYTE)
653 #define MAX_CHAR_FIELD1_OFFICIAL \
654 (MAX_LEADING_BYTE_OFFICIAL_2 - FIELD1_TO_OFFICIAL_LEADING_BYTE)
656 #define MIN_CHAR_FIELD2_PRIVATE \
657 (MIN_LEADING_BYTE_PRIVATE_1 - FIELD2_TO_PRIVATE_LEADING_BYTE)
658 #define MAX_CHAR_FIELD2_PRIVATE \
659 (MAX_LEADING_BYTE_PRIVATE_1 - FIELD2_TO_PRIVATE_LEADING_BYTE)
661 #define MIN_CHAR_FIELD1_PRIVATE \
662 (MIN_LEADING_BYTE_PRIVATE_2 - FIELD1_TO_PRIVATE_LEADING_BYTE)
663 #define MAX_CHAR_FIELD1_PRIVATE \
664 (MAX_LEADING_BYTE_PRIVATE_2 - FIELD1_TO_PRIVATE_LEADING_BYTE)
666 /* Minimum character code of each <type> character. */
668 #define MIN_CHAR_OFFICIAL_TYPE9N (MIN_CHAR_FIELD2_OFFICIAL << 7)
669 #define MIN_CHAR_PRIVATE_TYPE9N (MIN_CHAR_FIELD2_PRIVATE << 7)
670 #define MIN_CHAR_OFFICIAL_TYPE9NX9N (MIN_CHAR_FIELD1_OFFICIAL << 14)
671 #define MIN_CHAR_PRIVATE_TYPE9NX9N (MIN_CHAR_FIELD1_PRIVATE << 14)
672 #define MIN_CHAR_COMPOSITION (0x1F << 14)
674 /* Leading byte of a character.
676 NOTE: This takes advantage of the fact that
677 FIELD2_TO_OFFICIAL_LEADING_BYTE and
678 FIELD2_TO_PRIVATE_LEADING_BYTE are the same.
681 INLINE_HEADER Bufbyte CHAR_LEADING_BYTE (Emchar c);
682 INLINE_HEADER Bufbyte
683 CHAR_LEADING_BYTE (Emchar c)
685 if (CHAR_ASCII_P (c))
686 return LEADING_BYTE_ASCII;
688 return LEADING_BYTE_CONTROL_1;
689 else if (c < MIN_CHAR_OFFICIAL_TYPE9NX9N)
690 return CHAR_FIELD2 (c) + FIELD2_TO_OFFICIAL_LEADING_BYTE;
691 else if (c < MIN_CHAR_PRIVATE_TYPE9NX9N)
692 return CHAR_FIELD1 (c) + FIELD1_TO_OFFICIAL_LEADING_BYTE;
693 else if (c < MIN_CHAR_COMPOSITION)
694 return CHAR_FIELD1 (c) + FIELD1_TO_PRIVATE_LEADING_BYTE;
697 #ifdef ENABLE_COMPOSITE_CHARS
698 return LEADING_BYTE_COMPOSITE;
702 #endif /* ENABLE_COMPOSITE_CHARS */
706 #define CHAR_CHARSET(c) CHARSET_BY_LEADING_BYTE (CHAR_LEADING_BYTE (c))
708 /* Return a character whose charset is CHARSET and position-codes
709 are C1 and C2. TYPE9N character ignores C2.
711 NOTE: This takes advantage of the fact that
712 FIELD2_TO_OFFICIAL_LEADING_BYTE and
713 FIELD2_TO_PRIVATE_LEADING_BYTE are the same.
716 INLINE_HEADER Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2);
718 MAKE_CHAR (Lisp_Object charset, int c1, int c2)
720 if (EQ (charset, Vcharset_ascii))
722 else if (EQ (charset, Vcharset_control_1))
724 #ifdef ENABLE_COMPOSITE_CHARS
725 else if (EQ (charset, Vcharset_composite))
726 return (0x1F << 14) | ((c1) << 7) | (c2);
728 else if (XCHARSET_DIMENSION (charset) == 1)
729 return ((XCHARSET_LEADING_BYTE (charset) -
730 FIELD2_TO_OFFICIAL_LEADING_BYTE) << 7) | (c1);
731 else if (!XCHARSET_PRIVATE_P (charset))
732 return ((XCHARSET_LEADING_BYTE (charset) -
733 FIELD1_TO_OFFICIAL_LEADING_BYTE) << 14) | ((c1) << 7) | (c2);
735 return ((XCHARSET_LEADING_BYTE (charset) -
736 FIELD1_TO_PRIVATE_LEADING_BYTE) << 14) | ((c1) << 7) | (c2);
739 /* The charset of character C is set to CHARSET, and the
740 position-codes of C are set to C1 and C2. C2 of TYPE9N character
743 /* BREAKUP_CHAR_1_UNSAFE assumes that the charset has already been
744 calculated, and just computes c1 and c2.
746 BREAKUP_CHAR also computes and stores the charset. */
748 #define BREAKUP_CHAR_1_UNSAFE(c, charset, c1, c2) \
749 XCHARSET_DIMENSION (charset) == 1 \
750 ? ((c1) = CHAR_FIELD3 (c), (c2) = 0) \
751 : ((c1) = CHAR_FIELD2 (c), \
752 (c2) = CHAR_FIELD3 (c))
754 INLINE_HEADER void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2);
756 breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2)
758 *charset = CHAR_CHARSET (c);
759 BREAKUP_CHAR_1_UNSAFE (c, *charset, *c1, *c2);
762 #define BREAKUP_CHAR(c, charset, c1, c2) \
763 breakup_char_1 (c, &(charset), &(c1), &(c2))
767 #ifdef ENABLE_COMPOSITE_CHARS
768 /************************************************************************/
769 /* Composite characters */
770 /************************************************************************/
772 Emchar lookup_composite_char (Bufbyte *str, int len);
773 Lisp_Object composite_char_string (Emchar ch);
774 #endif /* ENABLE_COMPOSITE_CHARS */
777 /************************************************************************/
778 /* Exported functions */
779 /************************************************************************/
781 EXFUN (Ffind_charset, 1);
782 EXFUN (Fget_charset, 1);
784 extern Lisp_Object Vcharset_chinese_big5_1;
785 extern Lisp_Object Vcharset_chinese_big5_2;
786 extern Lisp_Object Vcharset_japanese_jisx0208;
788 Emchar Lstream_get_emchar_1 (Lstream *stream, int first_char);
789 int Lstream_fput_emchar (Lstream *stream, Emchar ch);
790 void Lstream_funget_emchar (Lstream *stream, Emchar ch);
792 int copy_internal_to_external (const Bufbyte *internal, Bytecount len,
793 unsigned char *external);
794 Bytecount copy_external_to_internal (const unsigned char *external,
795 int len, Bufbyte *internal);
797 #endif /* INCLUDED_mule_charset_h_ */