X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fchar-ucs.h;h=ef96fbf14f2cf6fbe334d10ecb0f218ac0cfd938;hb=0680bf5c284a73aaa6e267c4b0f1363d2cb48739;hp=a6a22808ab61c7a48f439606c63cc3ee81919ff1;hpb=f4eefc19d374808509e1a7c9b6711c2fbe998a7b;p=chise%2Fxemacs-chise.git- diff --git a/src/char-ucs.h b/src/char-ucs.h index a6a2280..ef96fbf 100644 --- a/src/char-ucs.h +++ b/src/char-ucs.h @@ -29,15 +29,23 @@ Boston, MA 02111-1307, USA. */ #define CHAR_ASCII_P(ch) ((ch) <= 0x7F) -typedef struct +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 { - unsigned char *base; - size_t offset; - size_t size; -} Emchar_to_byte_table; + struct lcrecord_header header; + + Lisp_Object property[256]; +}; -unsigned char -get_byte_from_character_table (Emchar ch, Emchar_to_byte_table* table); +Lisp_Object get_char_code_table (Emchar ch, Lisp_Object table); extern Lisp_Object Vcharset_ucs_bmp; @@ -56,13 +64,26 @@ extern Lisp_Object Vcharset_latin_viscii_upper; typedef int Charset_ID; -#define MIN_LEADING_BYTE 0x80 +#define MIN_LEADING_BYTE -0x40 +#define CHARSET_ID_OFFSET 0x00 + +/* represent normal 80-9F */ +#define LEADING_BYTE_CONTROL_1 (CHARSET_ID_OFFSET - 1) + +/* ISO/IEC 10646 BMP */ +#define LEADING_BYTE_UCS_BMP (CHARSET_ID_OFFSET - 2) + +/* VISCII 1.1 */ +#define LEADING_BYTE_LATIN_VISCII (CHARSET_ID_OFFSET - 3) -#define LEADING_BYTE_UCS_BMP 0x80 -#define LEADING_BYTE_CONTROL_1 0x81 /* represent normal 80-9F */ +#define LEADING_BYTE_HIRAGANA_JISX0208 (CHARSET_ID_OFFSET - 4) +#define LEADING_BYTE_KATAKANA_JISX0208 (CHARSET_ID_OFFSET - 5) +#define MIN_LEADING_BYTE_PRIVATE MIN_LEADING_BYTE +#define MAX_LEADING_BYTE_PRIVATE (CHARSET_ID_OFFSET - 6) -#define CHARSET_ID_OFFSET_94 0x55 + +#define CHARSET_ID_OFFSET_94 (CHARSET_ID_OFFSET - '0') #define MIN_CHARSET_ID_PRIVATE_94 (CHARSET_ID_OFFSET_94 + '0') #define MAX_CHARSET_ID_PRIVATE_94 (CHARSET_ID_OFFSET_94 + '?') @@ -77,7 +98,7 @@ typedef int Charset_ID; #define LEADING_BYTE_LATIN_JISX0201 (CHARSET_ID_OFFSET_94 + 'J') -#define CHARSET_ID_OFFSET_96 0x70 +#define CHARSET_ID_OFFSET_96 (CHARSET_ID_OFFSET_94 + 80) #define LEADING_BYTE_LATIN_VISCII_LOWER (CHARSET_ID_OFFSET_96 + '1') #define LEADING_BYTE_LATIN_VISCII_UPPER (CHARSET_ID_OFFSET_96 + '2') @@ -113,11 +134,7 @@ typedef int Charset_ID; #define LEADING_BYTE_THAI_TIS620 (CHARSET_ID_OFFSET_96 + 'T') -#define MIN_LEADING_BYTE_PRIVATE_1 0xD0 -#define MAX_LEADING_BYTE_PRIVATE_1 0xDF - - -#define CHARSET_ID_OFFSET_94x94 0xB0 +#define CHARSET_ID_OFFSET_94x94 (CHARSET_ID_OFFSET_96 + 80) /* Big5 Level 1 */ #define LEADING_BYTE_CHINESE_BIG5_1 ('0' + CHARSET_ID_OFFSET_94x94) @@ -171,10 +188,8 @@ typedef int Charset_ID; /* DPRK Hangul KPS 9566-1997 */ #define LEADING_BYTE_KOREAN_KPS9566 ('N' + CHARSET_ID_OFFSET_94x94) -#define MIN_LEADING_BYTE_OFFICIAL_2 LEADING_BYTE_JAPANESE_JISX0208_1978 -#define MAX_LEADING_BYTE_OFFICIAL_2 LEADING_BYTE_KOREAN_KPS9566 -#define NUM_LEADING_BYTES 256 +#define NUM_LEADING_BYTES (80 * 3 - MIN_LEADING_BYTE) /************************************************************************/ @@ -215,12 +230,20 @@ struct Lisp_Charset /* Which half of font to be used to display this character set */ unsigned int graphic; - /* byte->character mapping table */ - Emchar* decoding_table; + /* Byte->character mapping table */ + Lisp_Object decoding_table; + + /* Character->byte mapping table */ + Lisp_Object encoding_table; + + /* Range of character code */ + Emchar ucs_min, ucs_max; - /* character->byte mapping table */ - Emchar_to_byte_table* to_byte1_table; - Emchar_to_byte_table* to_byte2_table; + /* Offset for external code */ + Emchar code_offset; + + /* Offset for each byte */ + Emchar byte_offset; }; DECLARE_LRECORD (charset, struct Lisp_Charset); @@ -231,19 +254,21 @@ DECLARE_LRECORD (charset, struct Lisp_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_96 1 /* This charset includes 96 characters. */ -#define CHARSET_TYPE_94X94 2 /* This charset includes 94x94 characters. */ -#define CHARSET_TYPE_96X96 3 /* This charset includes 96x96 characters. */ -#define CHARSET_TYPE_128X128 4 /* This charset includes 128x128 characters. */ -#define CHARSET_TYPE_256X256 5 /* This charset includes 256x256 characters. */ +#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 /* Leading byte and id have been regrouped. -- OG */ #define CHARSET_ID(cs) ((cs)->id) -#define CHARSET_LEADING_BYTE(cs) ((Bufbyte)(CHARSET_ID(cs))) +#define CHARSET_LEADING_BYTE(cs) (CHARSET_ID(cs)) #define CHARSET_NAME(cs) ((cs)->name) #define CHARSET_SHORT_NAME(cs) ((cs)->short_name) #define CHARSET_LONG_NAME(cs) ((cs)->long_name) @@ -259,8 +284,11 @@ 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_TO_BYTE1_TABLE(cs) ((cs)->to_byte1_table) -#define CHARSET_TO_BYTE2_TABLE(cs) ((cs)->to_byte2_table) +#define CHARSET_ENCODING_TABLE(cs) ((cs)->encoding_table) +#define CHARSET_UCS_MIN(cs) ((cs)->ucs_min) +#define CHARSET_UCS_MAX(cs) ((cs)->ucs_max) +#define CHARSET_CODE_OFFSET(cs) ((cs)->code_offset) +#define CHARSET_BYTE_OFFSET(cs) ((cs)->byte_offset) #define XCHARSET_ID(cs) CHARSET_ID (XCHARSET (cs)) @@ -280,8 +308,11 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define XCHARSET_REVERSE_DIRECTION_CHARSET(cs) \ CHARSET_REVERSE_DIRECTION_CHARSET (XCHARSET (cs)) #define XCHARSET_DECODING_TABLE(cs) CHARSET_DECODING_TABLE(XCHARSET(cs)) -#define XCHARSET_TO_BYTE1_TABLE(cs) CHARSET_TO_BYTE1_TABLE(XCHARSET(cs)) -#define XCHARSET_TO_BYTE2_TABLE(cs) CHARSET_TO_BYTE2_TABLE(XCHARSET(cs)) +#define XCHARSET_ENCODING_TABLE(cs) CHARSET_ENCODING_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_CODE_OFFSET(cs) CHARSET_CODE_OFFSET(XCHARSET(cs)) +#define XCHARSET_BYTE_OFFSET(cs) CHARSET_BYTE_OFFSET(XCHARSET(cs)) /* Table of charsets indexed by (leading byte - 128). */ extern Lisp_Object charset_by_leading_byte[NUM_LEADING_BYTES]; @@ -331,9 +362,18 @@ CHARSET_BY_LEADING_BYTE (Charset_ID lb) #define MIN_CHAR_THAI 0x0E00 #define MAX_CHAR_THAI 0x0E5F -#define MIN_CHAR_HALFWIDTH_KATAKANA 0xFF60 +#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 MIN_CHAR_94 0xE90940 #define MAX_CHAR_94 (MIN_CHAR_94 + 94 * 80 - 1) #define MIN_CHAR_96 (MIN_CHAR_94 + 94 * 80) @@ -352,32 +392,47 @@ INLINE Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2); INLINE Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2) { - Emchar* decoding_table; - - if ((decoding_table = XCHARSET_DECODING_TABLE (charset)) != NULL) - return decoding_table[c1 - (XCHARSET_CHARS (charset) == 94 ? 33 : 32)]; - else if (EQ (charset, Vcharset_ascii)) - return c1; - else if (EQ (charset, Vcharset_control_1)) - return c1 | 0x80; - else if (EQ (charset, Vcharset_ucs_bmp)) - return (c1 << 8) | c2; - else if (EQ (charset, Vcharset_latin_iso8859_1)) - return c1 | 0x80; - else if (EQ (charset, Vcharset_cyrillic_iso8859_5)) - return c1 + MIN_CHAR_CYRILLIC - 0x20; - else if (EQ (charset, Vcharset_greek_iso8859_7)) - return c1 + MIN_CHAR_GREEK - 0x20; - else if (EQ (charset, Vcharset_hebrew_iso8859_8)) - return c1 + MIN_CHAR_HEBREW - 0x20; - else if (EQ (charset, Vcharset_thai_tis620)) - return c1 + MIN_CHAR_THAI - 0x20; - else if (EQ (charset, Vcharset_katakana_jisx0201)) + 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)) + { + 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); + } + if (EQ (charset, Vcharset_katakana_jisx0201)) if (c1 < 0x60) - return c1 + MIN_CHAR_HALFWIDTH_KATAKANA - 0x20; + return c1 + MIN_CHAR_HALFWIDTH_KATAKANA - 33; else /* return MIN_CHAR_94 + ('I' - '0') * 94 + (c1 - 33); */ return ' '; + else if (XCHARSET_UCS_MAX (charset)) + { + 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; + } else if (XCHARSET_DIMENSION (charset) == 1) { switch (XCHARSET_CHARS (charset)) @@ -410,23 +465,19 @@ MAKE_CHAR (Lisp_Object charset, int c1, int c2) } } -extern Charset_ID latin_a_char_to_charset[128]; -extern unsigned char latin_a_char_to_byte1[128]; -extern unsigned char latin_a_char_to_byte2[128]; - unsigned char charset_get_byte1 (Lisp_Object charset, Emchar ch); unsigned char charset_get_byte2 (Lisp_Object charset, Emchar ch); -extern Lisp_Object Vdefault_preferred_coded_charset_list; +extern Lisp_Object Vdefault_coded_charset_priority_list; EXFUN (Ffind_charset, 1); INLINE void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2); INLINE void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2) { - if (c < 0x100) + if (c < MIN_CHAR_OBS_94x94) { - Lisp_Object charsets = Vdefault_preferred_coded_charset_list; + Lisp_Object charsets = Vdefault_coded_charset_priority_list; while (!EQ (charsets, Qnil)) { *charset = Ffind_charset (Fcar (charsets)); @@ -438,7 +489,7 @@ breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2) } charsets = Fcdr (charsets); } - /* otherwise */ + /* otherwise --- maybe for bootstrap */ if (c <= MAX_CHAR_BASIC_LATIN) { *charset = Vcharset_ascii; @@ -457,113 +508,52 @@ breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2) *c1 = charset_get_byte1 (*charset, c); *c2 = charset_get_byte2 (*charset, c); } - else + else if ((MIN_CHAR_GREEK <= c) && (c <= MAX_CHAR_GREEK)) { - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; + *charset = Vcharset_greek_iso8859_7; + *c1 = c - MIN_CHAR_GREEK + 0x20; + *c2 = 0; } - } - else if (c <= 0x17f) - { - *charset - = CHARSET_BY_LEADING_BYTE (latin_a_char_to_charset[c - 0x100]); - if (XCHARSET_TO_BYTE1_TABLE (*charset) != NULL) + else if ((MIN_CHAR_CYRILLIC <= c) && (c <= MAX_CHAR_CYRILLIC)) { - *c1 = charset_get_byte1 (*charset, c); - *c2 = charset_get_byte2 (*charset, c); + *charset = Vcharset_cyrillic_iso8859_5; + *c1 = c - MIN_CHAR_CYRILLIC + 0x20; + *c2 = 0; } - else + else if ((MIN_CHAR_HEBREW <= c) && (c <= MAX_CHAR_HEBREW)) { - *c1 = latin_a_char_to_byte1[c - 0x100]; - *c2 = latin_a_char_to_byte2[c - 0x100]; + *charset = Vcharset_hebrew_iso8859_8; + *c1 = c - MIN_CHAR_HEBREW + 0x20; + *c2 = 0; } - } - else if (c < MIN_CHAR_GREEK) - { - Lisp_Object charsets = Vdefault_preferred_coded_charset_list; - while (!EQ (charsets, Qnil)) + else if ((MIN_CHAR_THAI <= c) && (c <= MAX_CHAR_THAI)) { - *charset = Ffind_charset (Fcar (charsets)); - if (!EQ (*charset, Qnil) - && (*c1 = charset_get_byte1 (*charset, c)) ) - { - *c2 = charset_get_byte2 (*charset, c); - return; - } - charsets = Fcdr (charsets); + *charset = Vcharset_thai_tis620; + *c1 = c - MIN_CHAR_THAI + 0x20; + *c2 = 0; } - /* otherwise */ - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; - } - else if (c <= MAX_CHAR_GREEK) - { - *charset = Vcharset_greek_iso8859_7; - *c1 = c - MIN_CHAR_GREEK + 0x20; - *c2 = 0; - } - else if (c < MIN_CHAR_CYRILLIC) - { - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; - } - else if (c <= MAX_CHAR_CYRILLIC) - { - *charset = Vcharset_cyrillic_iso8859_5; - *c1 = c - MIN_CHAR_CYRILLIC + 0x20; - *c2 = 0; - } - else if (c < MIN_CHAR_HEBREW) - { - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; - } - else if (c <= MAX_CHAR_HEBREW) - { - *charset = Vcharset_hebrew_iso8859_8; - *c1 = c - MIN_CHAR_HEBREW + 0x20; - *c2 = 0; - } - else if (c < MIN_CHAR_THAI) - { - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; - } - else if (c <= MAX_CHAR_THAI) - { - *charset = Vcharset_thai_tis620; - *c1 = c - MIN_CHAR_THAI + 0x20; - *c2 = 0; - } - else if (c < MIN_CHAR_HALFWIDTH_KATAKANA) - { - Lisp_Object charsets = Vdefault_preferred_coded_charset_list; - while (!EQ (charsets, Qnil)) + else if ((MIN_CHAR_HALFWIDTH_KATAKANA <= c) + && (c <= MAX_CHAR_HALFWIDTH_KATAKANA)) { - *charset = Ffind_charset (Fcar (charsets)); - if (!EQ (*charset, Qnil) - && (*c1 = charset_get_byte1 (*charset, c)) ) - { - *c2 = charset_get_byte2 (*charset, c); - return; - } - charsets = Fcdr (charsets); + *charset = Vcharset_katakana_jisx0201; + *c1 = c - MIN_CHAR_HALFWIDTH_KATAKANA + 33; + *c2 = 0; + } + else + { + *charset = Vcharset_ucs_bmp; + *c1 = c >> 8; + *c2 = c & 0xff; } - /* otherwise */ - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; } - else if (c <= MAX_CHAR_HALFWIDTH_KATAKANA) + else if (c <= MAX_CHAR_OBS_94x94) { - *charset = Vcharset_katakana_jisx0201; - *c1 = c - MIN_CHAR_HALFWIDTH_KATAKANA + 0x20; - *c2 = 0; + *charset + = CHARSET_BY_ATTRIBUTES (CHARSET_TYPE_94X94, + ((c - MIN_CHAR_OBS_94x94) / (94 * 94)) + '@', + CHARSET_LEFT_TO_RIGHT); + *c1 = (((c - MIN_CHAR_OBS_94x94) / 94) % 94) + 33; + *c2 = ((c - MIN_CHAR_OBS_94x94) % 94) + 33; } else if (c <= MAX_CHAR_94) { @@ -622,8 +612,7 @@ CHAR_CHARSET (Emchar ch) return charset; } -#define CHAR_LEADING_BYTE(c) (XCHARSET_LEADING_BYTE(CHAR_CHARSET(c))) - +#define CHAR_CHARSET_ID(c) (XCHARSET_ID(CHAR_CHARSET(c))) #define CHAR_COLUMNS(c) (CHARSET_COLUMNS(XCHARSET(CHAR_CHARSET(c))))