X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fchar-ucs.h;h=7b68791a667fe5097b61989915008b8f1840ff89;hb=15fd1e6e063e07e811ab9d9e4611682baa4df555;hp=7163c9ebd58d609159bf38664ba19c6b923ab0a6;hpb=61011fe65e02e28fcff91958e774ae14503387bb;p=chise%2Fxemacs-chise.git diff --git a/src/char-ucs.h b/src/char-ucs.h index 7163c9e..7b68791 100644 --- a/src/char-ucs.h +++ b/src/char-ucs.h @@ -1,6 +1,5 @@ /* Header for UCS-4 character representation. - Copyright (C) 1999 Electrotechnical Laboratory, JAPAN. - Licensed to the Free Software Foundation. + Copyright (C) 1999,2000,2001,2002 MORIOKA Tomohiko This file is part of XEmacs. @@ -19,86 +18,141 @@ along with XEmacs; see the file COPYING. If not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ -/* Rewritten by MORIOKA Tomohiko . */ +#ifndef INCLUDED_char_ucs_h_ +#define INCLUDED_char_ucs_h_ -#ifndef _XEMACS_CHAR_UCS_H -#define _XEMACS_CHAR_UCS_H +#include "chartab.h" +#include "elhash.h" + +extern Lisp_Object Vchar_attribute_hash_table; #define valid_char_p(ch) 1 #define CHAR_ASCII_P(ch) ((ch) <= 0x7F) -DECLARE_LRECORD (char_byte_table, struct Lisp_Char_Byte_Table); -#define XCHAR_BYTE_TABLE(x) \ - XRECORD (x, char_byte_table, struct Lisp_Char_Byte_Table) -#define XSETCHAR_BYTE_TABLE(x, p) XSETRECORD (x, p, char_byte_table) -#define CHAR_BYTE_TABLE_P(x) RECORDP (x, char_byte_table) -#define GC_CHAR_BYTE_TABLE_P(x) GC_RECORDP (x, char_byte_table) -/* #define CHECK_CHAR_BYTE_TABLE(x) CHECK_RECORD (x, char_byte_table) - char table entries should never escape to Lisp */ - -struct Lisp_Char_Byte_Table -{ - struct lcrecord_header header; +/************************************************************************/ +/* Exported functions */ +/************************************************************************/ - Lisp_Object property[256]; -}; +extern Lisp_Object Vcharset_ucs; +extern Lisp_Object Vcharset_latin_jisx0201; +extern Lisp_Object Vcharset_chinese_big5; +extern Lisp_Object Vcharset_chinese_big5_1; +extern Lisp_Object Vcharset_chinese_big5_2; +extern Lisp_Object Vcharset_japanese_jisx0208; +extern Lisp_Object Vcharset_japanese_jisx0208_1990; +extern Lisp_Object Vcharset_japanese_jisx0212; -Lisp_Object get_char_code_table (Emchar ch, Lisp_Object table); +EXFUN (Fget_charset, 1); +extern Lisp_Object Qucs; -extern Lisp_Object Vcharset_ucs_bmp; -extern Lisp_Object Vcharset_latin_jisx0201; -extern Lisp_Object Vcharset_latin_iso8859_2; -extern Lisp_Object Vcharset_latin_iso8859_3; -extern Lisp_Object Vcharset_latin_iso8859_4; -extern Lisp_Object Vcharset_latin_iso8859_9; -extern Lisp_Object Vcharset_latin_viscii_lower; -extern Lisp_Object Vcharset_latin_viscii_upper; +Lisp_Object put_char_ccs_code_point (Lisp_Object character, + Lisp_Object ccs, Lisp_Object value); +Lisp_Object remove_char_ccs (Lisp_Object character, Lisp_Object ccs); /************************************************************************/ /* Definition of leading bytes */ /************************************************************************/ -typedef int Charset_ID; +typedef short Charset_ID; + +#define MIN_LEADING_BYTE -0x200 -#define MIN_LEADING_BYTE -0x40 -#define CHARSET_ID_OFFSET 0x00 +/* ISO/IEC 10646 */ +#define LEADING_BYTE_UCS (MIN_LEADING_BYTE + 1) /* represent normal 80-9F */ -#define LEADING_BYTE_CONTROL_1 (CHARSET_ID_OFFSET - 1) +#define LEADING_BYTE_CONTROL_1 (MIN_LEADING_BYTE + 2) /* ISO/IEC 10646 BMP */ -#define LEADING_BYTE_UCS_BMP (CHARSET_ID_OFFSET - 2) +#define LEADING_BYTE_UCS_BMP (MIN_LEADING_BYTE + 3) -/* VISCII 1.1 */ -#define LEADING_BYTE_LATIN_VISCII (CHARSET_ID_OFFSET - 3) +/* ISO/IEC 10646 SMP */ +#define LEADING_BYTE_UCS_SMP (MIN_LEADING_BYTE + 4) -/* MULE VISCII-LOWER (CHARSET_ID_OFFSET_96 + '1') */ -#define LEADING_BYTE_LATIN_VISCII_LOWER (CHARSET_ID_OFFSET - 4) +/* ISO/IEC 10646 SIP */ +#define LEADING_BYTE_UCS_SIP (MIN_LEADING_BYTE + 5) -/* MULE VISCII-UPPER (CHARSET_ID_OFFSET_96 + '2') */ -#define LEADING_BYTE_LATIN_VISCII_UPPER (CHARSET_ID_OFFSET - 5) +/* ISO/IEC 10646 for CNS */ +#define LEADING_BYTE_UCS_CNS (MIN_LEADING_BYTE + 6) + +/* ISO/IEC 10646 for JIS */ +#define LEADING_BYTE_UCS_JIS (MIN_LEADING_BYTE + 7) + +/* ISO/IEC 10646 for KS */ +#define LEADING_BYTE_UCS_KS (MIN_LEADING_BYTE + 8) + +/* ISO/IEC 10646 for Big5 */ +#define LEADING_BYTE_UCS_BIG5 (MIN_LEADING_BYTE + 9) + +/* Japanese JIS X0208 Common 2/4 2/{(8),9,10,11} 4/2 (B) */ +#define LEADING_BYTE_JIS_X0208 (MIN_LEADING_BYTE + 10) + +/* Japanese JIS X0208-1990 2/4 2/{(8),9,10,11} 4/2 (B) */ +#define LEADING_BYTE_JAPANESE_JISX0208_1990 (MIN_LEADING_BYTE + 11) + +/* Chinese GB 12345-1990 */ +#define LEADING_BYTE_CHINESE_GB12345 (MIN_LEADING_BYTE + 12) + +#define LEADING_BYTE_CHINESE_BIG5 (MIN_LEADING_BYTE + 20) /* Big5 Level 1 2/4 2/{(8),9,10,11} 4/0 '0' */ -#define LEADING_BYTE_CHINESE_BIG5_1 (CHARSET_ID_OFFSET - 6) +#define LEADING_BYTE_CHINESE_BIG5_1 (MIN_LEADING_BYTE + 21) /* Big5 Level 2 2/4 2/{(8),9,10,11} 4/0 '1' */ -#define LEADING_BYTE_CHINESE_BIG5_2 (CHARSET_ID_OFFSET - 7) +#define LEADING_BYTE_CHINESE_BIG5_2 (MIN_LEADING_BYTE + 22) -#define LEADING_BYTE_HIRAGANA_JISX0208 (CHARSET_ID_OFFSET - 8) -#define LEADING_BYTE_KATAKANA_JISX0208 (CHARSET_ID_OFFSET - 9) - -#define MIN_LEADING_BYTE_PRIVATE MIN_LEADING_BYTE -#define MAX_LEADING_BYTE_PRIVATE (CHARSET_ID_OFFSET - 10) +/* VISCII 1.1 */ +#define LEADING_BYTE_LATIN_VISCII (MIN_LEADING_BYTE + 24) +/* MULE VISCII-LOWER (CHARSET_ID_OFFSET_96 + '1') */ +#define LEADING_BYTE_LATIN_VISCII_LOWER (MIN_LEADING_BYTE + 25) -/* #define CHARSET_ID_OFFSET_94 (CHARSET_ID_OFFSET - '0') */ +/* MULE VISCII-UPPER (CHARSET_ID_OFFSET_96 + '2') */ +#define LEADING_BYTE_LATIN_VISCII_UPPER (MIN_LEADING_BYTE + 26) + +#define LEADING_BYTE_ETHIOPIC_UCS (MIN_LEADING_BYTE + 27) + +#define LEADING_BYTE_DAIKANWA_0 (MIN_LEADING_BYTE + 28) +#define LEADING_BYTE_DAIKANWA_1 (MIN_LEADING_BYTE + 29) +#define LEADING_BYTE_DAIKANWA_2 (MIN_LEADING_BYTE + 30) +#define LEADING_BYTE_DAIKANWA_3 (MIN_LEADING_BYTE + 31) + +#define LEADING_BYTE_GT (MIN_LEADING_BYTE + 40) +#define LEADING_BYTE_GT_PJ_1 (MIN_LEADING_BYTE + 41) +#define LEADING_BYTE_GT_PJ_2 (MIN_LEADING_BYTE + 42) +#define LEADING_BYTE_GT_PJ_3 (MIN_LEADING_BYTE + 43) +#define LEADING_BYTE_GT_PJ_4 (MIN_LEADING_BYTE + 44) +#define LEADING_BYTE_GT_PJ_5 (MIN_LEADING_BYTE + 45) +#define LEADING_BYTE_GT_PJ_6 (MIN_LEADING_BYTE + 46) +#define LEADING_BYTE_GT_PJ_7 (MIN_LEADING_BYTE + 47) +#define LEADING_BYTE_GT_PJ_8 (MIN_LEADING_BYTE + 48) +#define LEADING_BYTE_GT_PJ_9 (MIN_LEADING_BYTE + 49) +#define LEADING_BYTE_GT_PJ_10 (MIN_LEADING_BYTE + 50) +#define LEADING_BYTE_GT_PJ_11 (MIN_LEADING_BYTE + 51) + +#define LEADING_BYTE_CHINA3_JEF (MIN_LEADING_BYTE + 82) +#define LEADING_BYTE_CBETA (MIN_LEADING_BYTE + 83) +/* #define LEADING_BYTE_CHINESE_BIG5_CDP (MIN_LEADING_BYTE + 84) */ +#define LEADING_BYTE_HANZIKU_1 (MIN_LEADING_BYTE + 85) +#define LEADING_BYTE_HANZIKU_2 (MIN_LEADING_BYTE + 86) +#define LEADING_BYTE_HANZIKU_3 (MIN_LEADING_BYTE + 87) +#define LEADING_BYTE_HANZIKU_4 (MIN_LEADING_BYTE + 88) +#define LEADING_BYTE_HANZIKU_5 (MIN_LEADING_BYTE + 89) +#define LEADING_BYTE_HANZIKU_6 (MIN_LEADING_BYTE + 90) +#define LEADING_BYTE_HANZIKU_7 (MIN_LEADING_BYTE + 91) +#define LEADING_BYTE_HANZIKU_8 (MIN_LEADING_BYTE + 92) +#define LEADING_BYTE_HANZIKU_9 (MIN_LEADING_BYTE + 93) +#define LEADING_BYTE_HANZIKU_10 (MIN_LEADING_BYTE + 94) +#define LEADING_BYTE_HANZIKU_11 (MIN_LEADING_BYTE + 95) +#define LEADING_BYTE_HANZIKU_12 (MIN_LEADING_BYTE + 96) + +#define MIN_LEADING_BYTE_PRIVATE (MIN_LEADING_BYTE + 97) +#define MAX_LEADING_BYTE_PRIVATE -1 -/* #define MIN_CHARSET_ID_PRIVATE_94 (CHARSET_ID_OFFSET_94 + '0') */ -/* #define MAX_CHARSET_ID_PRIVATE_94 (CHARSET_ID_OFFSET_94 + '?') */ /* ISO 646 IRV */ #define LEADING_BYTE_ASCII 6 /* (CHARSET_ID_OFFSET_94 + 'B') */ @@ -142,9 +196,11 @@ typedef int Charset_ID; /* TIS620-2533 */ #define LEADING_BYTE_THAI_TIS620 166 /* (CHARSET_ID_OFFSET_96 + 'T') */ +/* Right-hand Part of the VSCII-2 (TCVN 5712:1983) */ +#define LEADING_BYTE_LATIN_TCVN5712 180 /* F = 5/10 0x5A `Z' */ -/* #define CHARSET_ID_OFFSET_94x94 (CHARSET_ID_OFFSET_96 + 80) */ +/* #define CHARSET_ID_OFFSET_94x94 (CHARSET_ID_OFFSET_96 + 80) */ /* #define MIN_LEADING_BYTE_PRIVATE_2 ('0' + CHARSET_ID_OFFSET_94x94) */ /* #define MAX_LEADING_BYTE_PRIVATE_2 ('?' + CHARSET_ID_OFFSET_94x94) */ @@ -205,7 +261,10 @@ struct Lisp_Charset int id; Lisp_Object name; - Lisp_Object doc_string, registry, short_name, long_name; + Lisp_Object doc_string; + Lisp_Object registry; + Lisp_Object short_name; + Lisp_Object long_name; Lisp_Object reverse_direction_charset; @@ -221,48 +280,43 @@ struct Lisp_Charset /* Direction of this character set */ unsigned int direction; - /* Type of this character set (94, 96, 94x94, 96x96) */ - unsigned int type; - - /* Number of bytes used in encoding of this character set (1 or 2) */ - unsigned int dimension; + /* Number of bytes used in encoding of this character set (1 .. 4) */ + unsigned short dimension; - /* Number of chars in each dimension (usually 94 or 96) */ - unsigned int chars; + /* Number of chars in each dimension (94, 96, 128, 256) */ + unsigned short chars; /* Which half of font to be used to display this character set */ unsigned int graphic; - /* Byte->character mapping table */ + /* Code-point->character mapping table */ Lisp_Object decoding_table; + /* Base CCS */ + Lisp_Object mother; + /* Range of character code */ - Emchar ucs_min, ucs_max; + int min_code, max_code; /* Offset for external code */ - Emchar code_offset; + int code_offset; + + /* Type of conversion from mother CCS */ + unsigned char conversion; /* Offset for each byte */ - Emchar byte_offset; + int byte_offset; }; +typedef struct Lisp_Charset Lisp_Charset; -DECLARE_LRECORD (charset, struct Lisp_Charset); -#define XCHARSET(x) XRECORD (x, charset, struct Lisp_Charset) +DECLARE_LRECORD (charset, Lisp_Charset); +#define XCHARSET(x) XRECORD (x, charset, Lisp_Charset) #define XSETCHARSET(x, p) XSETRECORD (x, p, charset) #define CHARSETP(x) RECORDP (x, charset) #define GC_CHARSETP(x) GC_RECORDP (x, charset) #define CHECK_CHARSET(x) CHECK_RECORD (x, charset) #define CONCHECK_CHARSET(x) CONCHECK_RECORD (x, charset) -#define CHARSET_TYPE_94 0 /* This charset includes 94 characters. */ -#define CHARSET_TYPE_94X94 1 /* This charset includes 94x94 characters. */ -#define CHARSET_TYPE_96 2 /* This charset includes 96 characters. */ -#define CHARSET_TYPE_96X96 3 /* This charset includes 96x96 characters. */ -#define CHARSET_TYPE_128 4 /* This charset includes 128 characters. */ -#define CHARSET_TYPE_128X128 5 /* This charset includes 128x128 characters. */ -#define CHARSET_TYPE_256 6 /* This charset includes 256 characters. */ -#define CHARSET_TYPE_256X256 7 /* This charset includes 256x256 characters. */ - #define CHARSET_LEFT_TO_RIGHT 0 #define CHARSET_RIGHT_TO_LEFT 1 @@ -274,7 +328,6 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define CHARSET_LONG_NAME(cs) ((cs)->long_name) #define CHARSET_COLUMNS(cs) ((cs)->columns) #define CHARSET_GRAPHIC(cs) ((cs)->graphic) -#define CHARSET_TYPE(cs) ((cs)->type) #define CHARSET_DIRECTION(cs) ((cs)->direction) #define CHARSET_FINAL(cs) ((cs)->final) #define CHARSET_DOC_STRING(cs) ((cs)->doc_string) @@ -284,18 +337,52 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define CHARSET_CHARS(cs) ((cs)->chars) #define CHARSET_REVERSE_DIRECTION_CHARSET(cs) ((cs)->reverse_direction_charset) #define CHARSET_DECODING_TABLE(cs) ((cs)->decoding_table) -#define CHARSET_UCS_MIN(cs) ((cs)->ucs_min) -#define CHARSET_UCS_MAX(cs) ((cs)->ucs_max) +#define CHARSET_MIN_CODE(cs) ((cs)->min_code) +#define CHARSET_MAX_CODE(cs) ((cs)->max_code) #define CHARSET_CODE_OFFSET(cs) ((cs)->code_offset) #define CHARSET_BYTE_OFFSET(cs) ((cs)->byte_offset) +#define CHARSET_MOTHER(cs) ((cs)->mother) +#define CHARSET_CONVERSION(cs) ((cs)->conversion) + +INLINE_HEADER int CHARSET_BYTE_SIZE (Lisp_Charset* cs); +INLINE_HEADER int +CHARSET_BYTE_SIZE (Lisp_Charset* cs) +{ + /* ad-hoc method for `ascii' */ + if ((CHARSET_CHARS (cs) == 94) && + (CHARSET_BYTE_OFFSET (cs) != 33)) + return 128 - CHARSET_BYTE_OFFSET (cs); + else + return CHARSET_CHARS (cs); +} +INLINE_HEADER Lisp_Object CHARSET_ENCODING_TABLE (Lisp_Charset* cs); +INLINE_HEADER Lisp_Object +CHARSET_ENCODING_TABLE (Lisp_Charset* cs) +{ + return Fgethash (CHARSET_NAME(cs), + Vchar_attribute_hash_table, + Qnil); +} + +#define CONVERSION_IDENTICAL 0 +#define CONVERSION_94 1 +#define CONVERSION_96 2 +#define CONVERSION_94x60 3 +#define CONVERSION_94x94 4 +#define CONVERSION_96x96 5 +#define CONVERSION_94x94x60 6 +#define CONVERSION_94x94x94 7 +#define CONVERSION_96x96x96 8 +#define CONVERSION_94x94x94x60 9 +#define CONVERSION_94x94x94x94 10 +#define CONVERSION_96x96x96x96 11 #define XCHARSET_ID(cs) CHARSET_ID (XCHARSET (cs)) #define XCHARSET_NAME(cs) CHARSET_NAME (XCHARSET (cs)) #define XCHARSET_SHORT_NAME(cs) CHARSET_SHORT_NAME (XCHARSET (cs)) #define XCHARSET_LONG_NAME(cs) CHARSET_LONG_NAME (XCHARSET (cs)) #define XCHARSET_GRAPHIC(cs) CHARSET_GRAPHIC (XCHARSET (cs)) -#define XCHARSET_TYPE(cs) CHARSET_TYPE (XCHARSET (cs)) #define XCHARSET_DIRECTION(cs) CHARSET_DIRECTION (XCHARSET (cs)) #define XCHARSET_FINAL(cs) CHARSET_FINAL (XCHARSET (cs)) #define XCHARSET_DOC_STRING(cs) CHARSET_DOC_STRING (XCHARSET (cs)) @@ -304,42 +391,66 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define XCHARSET_CCL_PROGRAM(cs) CHARSET_CCL_PROGRAM (XCHARSET (cs)) #define XCHARSET_DIMENSION(cs) CHARSET_DIMENSION (XCHARSET (cs)) #define XCHARSET_CHARS(cs) CHARSET_CHARS (XCHARSET (cs)) +#define XCHARSET_BYTE_SIZE(cs) CHARSET_BYTE_SIZE (XCHARSET (cs)) #define XCHARSET_REVERSE_DIRECTION_CHARSET(cs) \ CHARSET_REVERSE_DIRECTION_CHARSET (XCHARSET (cs)) #define XCHARSET_DECODING_TABLE(cs) CHARSET_DECODING_TABLE(XCHARSET(cs)) -#define XCHARSET_UCS_MIN(cs) CHARSET_UCS_MIN(XCHARSET(cs)) -#define XCHARSET_UCS_MAX(cs) CHARSET_UCS_MAX(XCHARSET(cs)) +#define XCHARSET_ENCODING_TABLE(cs) CHARSET_ENCODING_TABLE(XCHARSET(cs)) +#define XCHARSET_MIN_CODE(cs) CHARSET_MIN_CODE(XCHARSET(cs)) +#define XCHARSET_MAX_CODE(cs) CHARSET_MAX_CODE(XCHARSET(cs)) #define XCHARSET_CODE_OFFSET(cs) CHARSET_CODE_OFFSET(XCHARSET(cs)) #define XCHARSET_BYTE_OFFSET(cs) CHARSET_BYTE_OFFSET(XCHARSET(cs)) +#define XCHARSET_MOTHER(cs) CHARSET_MOTHER(XCHARSET(cs)) +#define XCHARSET_CONVERSION(cs) CHARSET_CONVERSION(XCHARSET(cs)) + +struct charset_lookup { + /* Table of charsets indexed by (leading byte - 128). */ + Lisp_Object charset_by_leading_byte[NUM_LEADING_BYTES]; + + /* Table of charsets indexed by type/final-byte/direction. */ + Lisp_Object charset_by_attributes[4][128]; -/* Table of charsets indexed by (leading byte - 128). */ -extern Lisp_Object charset_by_leading_byte[NUM_LEADING_BYTES]; + Charset_ID next_allocated_leading_byte; +}; -/* Table of charsets indexed by type/final-byte. */ -extern Lisp_Object charset_by_attributes[4][128]; +extern struct charset_lookup *chlook; #ifdef ERROR_CHECK_TYPECHECK /* int not Bufbyte even though that is the actual type of a leading byte. This way, out-ot-range values will get caught rather than automatically truncated. */ -INLINE Lisp_Object CHARSET_BY_LEADING_BYTE (Charset_ID lb); -INLINE Lisp_Object +INLINE_HEADER Lisp_Object CHARSET_BY_LEADING_BYTE (Charset_ID lb); +INLINE_HEADER Lisp_Object CHARSET_BY_LEADING_BYTE (Charset_ID lb) { assert (lb >= MIN_LEADING_BYTE && lb < (MIN_LEADING_BYTE + NUM_LEADING_BYTES)); - return charset_by_leading_byte[lb - MIN_LEADING_BYTE]; + return chlook->charset_by_leading_byte[lb - MIN_LEADING_BYTE]; } #else #define CHARSET_BY_LEADING_BYTE(lb) \ - (charset_by_leading_byte[(lb) - MIN_LEADING_BYTE]) + (chlook->charset_by_leading_byte[(lb) - MIN_LEADING_BYTE]) #endif -#define CHARSET_BY_ATTRIBUTES(type, final, dir) \ - (charset_by_attributes[type][final]) +INLINE_HEADER Lisp_Object +CHARSET_BY_ATTRIBUTES (int chars, int dimension, int final, int dir); +INLINE_HEADER Lisp_Object +CHARSET_BY_ATTRIBUTES (int chars, int dimension, int final, int dir) +{ + int type; + + if (chars == 94) + type = 0; + else if (chars == 96) + type = 1; + else + return Qnil; + type += dimension == 1 ? 0 : 2; + return chlook->charset_by_attributes[type][final]; +} /************************************************************************/ @@ -348,299 +459,340 @@ CHARSET_BY_LEADING_BYTE (Charset_ID lb) #define MAX_CHAR_BASIC_LATIN 0x007F -#define MIN_CHAR_GREEK 0x0370 -#define MAX_CHAR_GREEK 0x03CF - -#define MIN_CHAR_CYRILLIC 0x0400 -#define MAX_CHAR_CYRILLIC 0x045F - +/* #define MIN_CHAR_HEBREW 0x0590 #define MAX_CHAR_HEBREW 0x05EF +*/ #define MIN_CHAR_THAI 0x0E00 #define MAX_CHAR_THAI 0x0E5F +/* #define MIN_CHAR_HIRAGANA 0x3041 #define MAX_CHAR_HIRAGANA 0x3093 #define MIN_CHAR_KATAKANA 0x30A1 #define MAX_CHAR_KATAKANA 0x30F6 +*/ #define MIN_CHAR_HALFWIDTH_KATAKANA 0xFF61 #define MAX_CHAR_HALFWIDTH_KATAKANA 0xFF9F -#define MIN_CHAR_OBS_94x94 0xE00000 -#define MAX_CHAR_OBS_94x94 (0xE00000 + 94 * 94 * 14 - 1) +#define MAX_CHAR_BMP 0x00FFFF + +#define MIN_CHAR_SMP 0x010000 +#define MAX_CHAR_SMP 0x01FFFF + +#define MIN_CHAR_SIP 0x020000 +#define MAX_CHAR_SIP 0x02FFFF + +#define MIN_CHAR_DAIKANWA 0x00E00000 +#define MAX_CHAR_DAIKANWA (MIN_CHAR_DAIKANWA + 50100) /* 0xE0FFFF */ +#define MIN_CHAR_CBETA 0x00E20000 +#define MAX_CHAR_CBETA 0x00E2FFFF +#define MIN_CHAR_CHINA3_JEF 0x00E80000 +#define MAX_CHAR_CHINA3_JEF 0x00E8FFFF -#define MIN_CHAR_94 0xE90940 +#define MIN_CHAR_94 0x00E90940 #define MAX_CHAR_94 (MIN_CHAR_94 + 94 * 80 - 1) #define MIN_CHAR_96 (MIN_CHAR_94 + 94 * 80) #define MAX_CHAR_96 (MIN_CHAR_96 + 96 * 80 - 1) #define MIN_CHAR_94x94 0xE9F6C0 +#define MIN_CHAR_JIS_X0208_1990 (MIN_CHAR_94x94 + 94 * 94 * 79) #define MAX_CHAR_94x94 (MIN_CHAR_94x94 + 94 * 94 * 80 - 1) +#define MAX_CHAR_JIS_X0208_1990 MAX_CHAR_94x94 #define MIN_CHAR_96x96 0xF4C000 #define MAX_CHAR_96x96 (MIN_CHAR_96x96 + 96 * 96 * 80 - 1) +#define MIN_CHAR_GT 0x61000000 +#define MAX_CHAR_GT (MIN_CHAR_GT + 66773) +#define MIN_CHAR_BIG5_CDP 0x62000000 +#define MAX_CHAR_BIG5_CDP 0x6200FFFF +#define MIN_CHAR_HANZIKU_1 (0x62000000 + 65536 * 1) +#define MAX_CHAR_HANZIKU_1 (0x62000000 + 65536 * 1 + 65535) +#define MIN_CHAR_HANZIKU_2 (0x62000000 + 65536 * 2) +#define MAX_CHAR_HANZIKU_2 (0x62000000 + 65536 * 2 + 65535) +#define MIN_CHAR_HANZIKU_3 (0x62000000 + 65536 * 3) +#define MAX_CHAR_HANZIKU_3 (0x62000000 + 65536 * 3 + 65535) +#define MIN_CHAR_HANZIKU_4 (0x62000000 + 65536 * 4) +#define MAX_CHAR_HANZIKU_4 (0x62000000 + 65536 * 4 + 65535) +#define MIN_CHAR_HANZIKU_5 (0x62000000 + 65536 * 5) +#define MAX_CHAR_HANZIKU_5 (0x62000000 + 65536 * 5 + 65535) +#define MIN_CHAR_HANZIKU_6 (0x62000000 + 65536 * 6) +#define MAX_CHAR_HANZIKU_6 (0x62000000 + 65536 * 6 + 65535) +#define MIN_CHAR_HANZIKU_7 (0x62000000 + 65536 * 7) +#define MAX_CHAR_HANZIKU_7 (0x62000000 + 65536 * 7 + 65535) +#define MIN_CHAR_HANZIKU_8 (0x62000000 + 65536 * 8) +#define MAX_CHAR_HANZIKU_8 (0x62000000 + 65536 * 8 + 65535) +#define MIN_CHAR_HANZIKU_9 (0x62000000 + 65536 * 9) +#define MAX_CHAR_HANZIKU_9 (0x62000000 + 65536 * 9 + 65535) +#define MIN_CHAR_HANZIKU_10 (0x62000000 + 65536 * 10) +#define MAX_CHAR_HANZIKU_10 (0x62000000 + 65536 * 10 + 65535) +#define MIN_CHAR_HANZIKU_11 (0x62000000 + 65536 * 11) +#define MAX_CHAR_HANZIKU_11 (0x62000000 + 65536 * 11 + 65535) +#define MIN_CHAR_HANZIKU_12 (0x62000000 + 65536 * 12) +#define MAX_CHAR_HANZIKU_12 (0x62000000 + 65536 * 12 + 65535) + +Emchar decode_builtin_char (Lisp_Object charset, int code_point); + +INLINE_HEADER Lisp_Object +get_ccs_octet_table (Lisp_Object table, Lisp_Object ccs, int code); +INLINE_HEADER Lisp_Object +get_ccs_octet_table (Lisp_Object table, Lisp_Object ccs, int code) +{ + int byte_offset = XCHARSET_BYTE_OFFSET (ccs); -/* Return a character whose charset is CHARSET and position-codes - are C1 and C2. TYPE9N character ignores C2. */ + if (VECTORP (table)) + { + int idx = code - byte_offset; -INLINE Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2); -INLINE Emchar -MAKE_CHAR (Lisp_Object charset, int c1, int c2) + if (idx < XVECTOR_LENGTH(table)) + return XVECTOR_DATA(table)[idx]; + else + return Qunbound; + } + else + return table; +} + +INLINE_HEADER Lisp_Object +put_ccs_octet_table (Lisp_Object table, Lisp_Object ccs, int code, + Lisp_Object value); +INLINE_HEADER Lisp_Object +put_ccs_octet_table (Lisp_Object table, Lisp_Object ccs, int code, + Lisp_Object value) { - Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (charset); - int idx; - Lisp_Object ch; - - if (!EQ (decoding_table, Qnil) - && (0 <= (idx = c1 - XCHARSET_BYTE_OFFSET (charset))) - && (idx < XVECTOR_LENGTH (decoding_table)) - && !EQ (ch = XVECTOR_DATA(decoding_table)[idx], Qnil)) + int byte_offset = XCHARSET_BYTE_OFFSET (ccs); + int ccs_len = XCHARSET_BYTE_SIZE (ccs); + + if (VECTORP (table)) { - if (VECTORP (ch)) - { - if ((0 <= (idx = c2 - XCHARSET_BYTE_OFFSET (charset))) - && (idx < XVECTOR_LENGTH (ch)) - && !EQ (ch = XVECTOR_DATA(ch)[idx], Qnil)) - return XCHAR (ch); - } - else - return XCHAR (ch); + XVECTOR_DATA(table)[code - byte_offset] = value; + return table; + } + else if (EQ (table, value)) + return table; + else + { + table = make_vector (ccs_len, table); + XVECTOR_DATA(table)[code - byte_offset] = value; + return table; + } +} + +INLINE_HEADER void +decoding_table_put_char (Lisp_Object ccs, + int code_point, Lisp_Object character); +INLINE_HEADER void +decoding_table_put_char (Lisp_Object ccs, + int code_point, Lisp_Object character) +{ + Lisp_Object table1 = XCHARSET_DECODING_TABLE (ccs); + int dim = XCHARSET_DIMENSION (ccs); + + if (dim == 1) + XCHARSET_DECODING_TABLE (ccs) + = put_ccs_octet_table (table1, ccs, code_point, character); + else if (dim == 2) + { + Lisp_Object table2 + = get_ccs_octet_table (table1, ccs, (unsigned char)(code_point >> 8)); + + table2 = put_ccs_octet_table (table2, ccs, + (unsigned char)code_point, character); + XCHARSET_DECODING_TABLE (ccs) + = put_ccs_octet_table (table1, ccs, + (unsigned char)(code_point >> 8), table2); + } + else if (dim == 3) + { + Lisp_Object table2 + = get_ccs_octet_table (table1, ccs, (unsigned char)(code_point >> 16)); + Lisp_Object table3 + = get_ccs_octet_table (table2, ccs, (unsigned char)(code_point >> 8)); + + table3 = put_ccs_octet_table (table3, ccs, + (unsigned char)code_point, character); + table2 = put_ccs_octet_table (table2, ccs, + (unsigned char)(code_point >> 8), table3); + XCHARSET_DECODING_TABLE (ccs) + = put_ccs_octet_table (table1, ccs, + (unsigned char)(code_point >> 16), table2); } - if (EQ (charset, Vcharset_katakana_jisx0201)) - if (c1 < 0x60) - return c1 + MIN_CHAR_HALFWIDTH_KATAKANA - 33; - else - /* return MIN_CHAR_94 + ('I' - '0') * 94 + (c1 - 33); */ - return ' '; - else if (XCHARSET_UCS_MAX (charset)) + else /* if (dim == 4) */ { - Emchar code - = (XCHARSET_DIMENSION (charset) == 1 - ? - c1 - XCHARSET_BYTE_OFFSET (charset) - : - (c1 - XCHARSET_BYTE_OFFSET (charset)) * XCHARSET_CHARS (charset) - + c2 - XCHARSET_BYTE_OFFSET (charset)) - - XCHARSET_CODE_OFFSET (charset) + XCHARSET_UCS_MIN (charset); - if ((code < XCHARSET_UCS_MIN (charset)) - || (XCHARSET_UCS_MAX (charset) < code)) - signal_simple_error ("Arguments makes invalid character", - make_char (code)); - return code; + Lisp_Object table2 + = get_ccs_octet_table (table1, ccs, (unsigned char)(code_point >> 24)); + Lisp_Object table3 + = get_ccs_octet_table (table2, ccs, (unsigned char)(code_point >> 16)); + Lisp_Object table4 + = get_ccs_octet_table (table3, ccs, (unsigned char)(code_point >> 8)); + + table4 = put_ccs_octet_table (table4, ccs, + (unsigned char)code_point, character); + table3 = put_ccs_octet_table (table3, ccs, + (unsigned char)(code_point >> 8), table4); + table2 = put_ccs_octet_table (table2, ccs, + (unsigned char)(code_point >> 16), table3); + XCHARSET_DECODING_TABLE (ccs) + = put_ccs_octet_table (table1, ccs, + (unsigned char)(code_point >> 24), table2); } +} + +INLINE_HEADER void +decoding_table_remove_char (Lisp_Object ccs, int code_point); +INLINE_HEADER void +decoding_table_remove_char (Lisp_Object ccs, int code_point) +{ + decoding_table_put_char (ccs, code_point, Qunbound); +} + +Emchar decode_defined_char (Lisp_Object charset, int code_point); + +INLINE_HEADER Emchar DECODE_CHAR (Lisp_Object charset, int code_point); +INLINE_HEADER Emchar +DECODE_CHAR (Lisp_Object charset, int code_point) +{ + Emchar char_id = decode_defined_char (charset, code_point); + + if (char_id >= 0) + return char_id; else if (XCHARSET_DIMENSION (charset) == 1) + return decode_builtin_char (charset, code_point); + else if (EQ (charset, Vcharset_chinese_big5_1)) { - switch (XCHARSET_CHARS (charset)) - { - case 94: - return MIN_CHAR_94 - + (XCHARSET_FINAL (charset) - '0') * 94 + (c1 - 33); - case 96: - return MIN_CHAR_96 - + (XCHARSET_FINAL (charset) - '0') * 96 + (c1 - 32); - default: - abort (); - } + unsigned int I + = ((code_point >> 8) - 33) * (0xFF - 0xA1) + + ((code_point & 0xFF) - 33); + unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; + unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); + + b2 += b2 < 0x3F ? 0x40 : 0x62; + return DECODE_CHAR (Vcharset_chinese_big5, (b1 << 8) | b2); } - else + else if (EQ (charset, Vcharset_chinese_big5_2)) { - switch (XCHARSET_CHARS (charset)) - { - case 94: - return MIN_CHAR_94x94 - + (XCHARSET_FINAL (charset) - '0') * 94 * 94 - + (c1 - 33) * 94 + (c2 - 33); - case 96: - return MIN_CHAR_96x96 - + (XCHARSET_FINAL (charset) - '0') * 96 * 96 - + (c1 - 32) * 96 + (c2 - 32); - default: - abort (); - } + unsigned int I + = ((code_point >> 8) - 33) * (0xFF - 0xA1) + + ((code_point & 0xFF) - 33); + unsigned char b1, b2; + + I += (0xFF - 0xA1 + 0x7F - 0x40) * (0xC9 - 0xA1); + b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; + b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); + b2 += b2 < 0x3F ? 0x40 : 0x62; + return DECODE_CHAR (Vcharset_chinese_big5, (b1 << 8) | b2); } + return decode_builtin_char (charset, code_point); +} + +/* Return a character whose charset is CHARSET and position-codes + are C1 and C2. TYPE9N character ignores C2. */ + +INLINE_HEADER Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2); +INLINE_HEADER Emchar +MAKE_CHAR (Lisp_Object charset, int c1, int c2) +{ + if (XCHARSET_DIMENSION (charset) == 1) + return DECODE_CHAR (charset, c1); + else + return DECODE_CHAR (charset, (c1 << 8) | c2); } extern Lisp_Object Vcharacter_attribute_table; -Lisp_Object range_charset_code_point (Lisp_Object charset, Emchar ch); -Lisp_Object charset_code_point (Lisp_Object charset, Emchar ch); +int encode_builtin_char_1 (Emchar c, Lisp_Object* charset); +int charset_code_point (Lisp_Object charset, Emchar ch, int defined_only); +int range_charset_code_point (Lisp_Object charset, Emchar ch); extern Lisp_Object Vdefault_coded_charset_priority_list; EXFUN (Ffind_charset, 1); -INLINE Lisp_Object SPLIT_CHAR (Emchar c); -INLINE Lisp_Object -SPLIT_CHAR (Emchar c) +INLINE_HEADER int encode_char_1 (Emchar ch, Lisp_Object* charset); +INLINE_HEADER int +encode_char_1 (Emchar ch, Lisp_Object* charset) { - Lisp_Object cdef = get_char_code_table (c, Vcharacter_attribute_table); + Lisp_Object charsets = Vdefault_coded_charset_priority_list; - if (!EQ (cdef, Qnil)) + while (!NILP (charsets)) { - Lisp_Object charsets = Vdefault_coded_charset_priority_list; - Lisp_Object field; - - while (!EQ (charsets, Qnil)) + *charset = Ffind_charset (Fcar (charsets)); + if (!NILP (*charset)) { - Lisp_Object charset = Ffind_charset (Fcar (charsets)); - - if (!EQ (charset, Qnil)) - { - if (!NILP (field = Fassq (charset, cdef))) - return field; - else if (!NILP (field = range_charset_code_point (charset, c))) - return Fcons (charset, field); - } - charsets = Fcdr (charsets); + int code_point = charset_code_point (*charset, ch, 0); + + if (code_point >= 0) + return code_point; } + charsets = Fcdr (charsets); } /* otherwise --- maybe for bootstrap */ - if (c < MIN_CHAR_OBS_94x94) - { - if (c <= MAX_CHAR_BASIC_LATIN) - { - return list2 (Vcharset_ascii, make_int (c)); - } - else if (c < 0xA0) - { - return list2 (Vcharset_control_1, make_int (c & 0x7F)); - } - else if (c <= 0xff) - { - return list2 (Vcharset_latin_iso8859_1, make_int (c & 0x7F)); - } - else if ((MIN_CHAR_GREEK <= c) && (c <= MAX_CHAR_GREEK)) - { - return list2 (Vcharset_greek_iso8859_7, - make_int (c - MIN_CHAR_GREEK + 0x20)); - } - else if ((MIN_CHAR_CYRILLIC <= c) && (c <= MAX_CHAR_CYRILLIC)) - { - return list2 (Vcharset_cyrillic_iso8859_5, - make_int (c - MIN_CHAR_CYRILLIC + 0x20)); - } - else if ((MIN_CHAR_HEBREW <= c) && (c <= MAX_CHAR_HEBREW)) - { - return list2 (Vcharset_hebrew_iso8859_8, - make_int (c - MIN_CHAR_HEBREW + 0x20)); - } - else if ((MIN_CHAR_THAI <= c) && (c <= MAX_CHAR_THAI)) - { - return list2 (Vcharset_thai_tis620, - make_int (c - MIN_CHAR_THAI + 0x20)); - } - else if ((MIN_CHAR_HALFWIDTH_KATAKANA <= c) - && (c <= MAX_CHAR_HALFWIDTH_KATAKANA)) - { - return list2 (Vcharset_katakana_jisx0201, - make_int (c - MIN_CHAR_HALFWIDTH_KATAKANA + 33)); - } - else - { - return list3 (Vcharset_ucs_bmp, - make_int (c >> 8), make_int (c & 0xff)); - } - } - else if (c <= MAX_CHAR_OBS_94x94) - { - return list3 (CHARSET_BY_ATTRIBUTES - (CHARSET_TYPE_94X94, - ((c - MIN_CHAR_OBS_94x94) / (94 * 94)) + '@', - CHARSET_LEFT_TO_RIGHT), - make_int ((((c - MIN_CHAR_OBS_94x94) / 94) % 94) + 33), - make_int (((c - MIN_CHAR_OBS_94x94) % 94) + 33)); - } - else if (c <= MAX_CHAR_94) - { - return list2 (CHARSET_BY_ATTRIBUTES (CHARSET_TYPE_94, - ((c - MIN_CHAR_94) / 94) + '0', - CHARSET_LEFT_TO_RIGHT), - make_int (((c - MIN_CHAR_94) % 94) + 33)); - } - else if (c <= MAX_CHAR_96) - { - return list2 (CHARSET_BY_ATTRIBUTES (CHARSET_TYPE_96, - ((c - MIN_CHAR_96) / 96) + '0', - CHARSET_LEFT_TO_RIGHT), - make_int (((c - MIN_CHAR_96) % 96) + 32)); - } - else if (c <= MAX_CHAR_94x94) - { - return list3 (CHARSET_BY_ATTRIBUTES - (CHARSET_TYPE_94X94, - ((c - MIN_CHAR_94x94) / (94 * 94)) + '0', - CHARSET_LEFT_TO_RIGHT), - make_int ((((c - MIN_CHAR_94x94) / 94) % 94) + 33), - make_int (((c - MIN_CHAR_94x94) % 94) + 33)); - } - else if (c <= MAX_CHAR_96x96) - { - return list3 (CHARSET_BY_ATTRIBUTES - (CHARSET_TYPE_96X96, - ((c - MIN_CHAR_96x96) / (96 * 96)) + '0', - CHARSET_LEFT_TO_RIGHT), - make_int ((((c - MIN_CHAR_96x96) / 96) % 96) + 32), - make_int (((c - MIN_CHAR_96x96) % 96) + 32)); - } - else - { - return Qnil; - } + return encode_builtin_char_1 (ch, charset); } -INLINE void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2); -INLINE void +#define ENCODE_CHAR(ch, charset) encode_char_1 (ch, &(charset)) + +INLINE_HEADER void +breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2); +INLINE_HEADER void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2) { - Lisp_Object ret = SPLIT_CHAR (c); + int code_point = encode_char_1 (c, charset); - *charset = Fcar (ret); - ret = Fcdr (ret); - if (INTP (Fcar (ret))) + if (code_point >= 0) { - *c1 = XINT (Fcar (ret)); - ret = Fcdr (ret); - if (INTP (Fcar (ret))) - *c2 = XINT (Fcar (ret)); - else - *c2 = 0; - } - else - { - *c1 = *c2 = 0; + int dim = XCHARSET_DIMENSION (*charset); + + if (dim == 1) + { + *c1 = code_point; + *c2 = 0; + } + else /* if (dim == 2) */ + { + *c1 = code_point >> 8; + *c2 = code_point & 255; + } } + else{ + *c1 = *c2 = 0; + } } #define BREAKUP_CHAR(ch, charset, b1, b2) \ breakup_char_1 (ch, &(charset), &(b1), &(b2)) -INLINE Lisp_Object CHAR_CHARSET (Emchar ch); -INLINE Lisp_Object +INLINE_HEADER Lisp_Object CHAR_CHARSET (Emchar ch); +INLINE_HEADER Lisp_Object CHAR_CHARSET (Emchar ch) { Lisp_Object charset; - int b1, b2; - - BREAKUP_CHAR(ch, charset, b1, b2); + + ENCODE_CHAR (ch, charset); return charset; } #define CHAR_CHARSET_ID(c) (XCHARSET_ID(CHAR_CHARSET(c))) #define CHAR_COLUMNS(c) (CHARSET_COLUMNS(XCHARSET(CHAR_CHARSET(c)))) - -/************************************************************************/ -/* Exported functions */ -/************************************************************************/ -EXFUN (Fget_charset, 1); +typedef struct Charc +{ + Lisp_Object charset; + int code_point; +} Charc; -extern Lisp_Object Vcharset_chinese_big5_1; -extern Lisp_Object Vcharset_chinese_big5_2; -extern Lisp_Object Vcharset_japanese_jisx0208; -extern Lisp_Object Vcharset_japanese_jisx0212; +INLINE_HEADER Charc CHAR_TO_CHARC (Emchar ch); +INLINE_HEADER Charc +CHAR_TO_CHARC (Emchar ch) +{ + Charc cc; + + cc.code_point = encode_char_1 (ch, &cc.charset); + return cc; +} -#endif /* _XEMACS_CHAR_UCS_H */ +#endif /* INCLUDED_char_ucs_h_ */