X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fchar-ucs.h;h=a47ea808fc503afca1d736808b851c6e393fe0a5;hb=d4fd14df47400ce4a63d921ec1a4c9de649622dc;hp=68552193d39503bbfe4e5c84ab4555dc16b4fc3d;hpb=2da37f54fd796d4607db9f555c255ea750be7712;p=chise%2Fxemacs-chise.git diff --git a/src/char-ucs.h b/src/char-ucs.h index 6855219..a47ea80 100644 --- a/src/char-ucs.h +++ b/src/char-ucs.h @@ -29,46 +29,33 @@ 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; -unsigned char -get_byte_from_character_table (Emchar ch, Emchar_to_byte_table* table); + Lisp_Object property[256]; +}; +Lisp_Object get_char_code_table (Emchar ch, Lisp_Object table); -extern Lisp_Object Vcharset_ucs_bmp; +extern Lisp_Object Vcharset_ucs_bmp; extern Lisp_Object Vcharset_latin_jisx0201; -extern Emchar latin_jisx0201_to_ucs[94]; -extern Emchar_to_byte_table* ucs_to_latin_jisx0201; - extern Lisp_Object Vcharset_latin_iso8859_2; -extern Emchar latin_iso8859_2_to_ucs[96]; -extern Emchar_to_byte_table* ucs_to_latin_iso8859_2; - extern Lisp_Object Vcharset_latin_iso8859_3; -extern Emchar latin_iso8859_3_to_ucs[96]; -extern Emchar_to_byte_table* ucs_to_latin_iso8859_3; - extern Lisp_Object Vcharset_latin_iso8859_4; -extern Emchar latin_iso8859_4_to_ucs[96]; -extern Emchar_to_byte_table* ucs_to_latin_iso8859_4; - extern Lisp_Object Vcharset_latin_iso8859_9; -extern Emchar latin_iso8859_9_to_ucs[96]; -extern Emchar_to_byte_table* ucs_to_latin_iso8859_9; - extern Lisp_Object Vcharset_latin_viscii_lower; -extern Emchar latin_viscii_lower_to_ucs[96]; -extern Emchar_to_byte_table* ucs_to_latin_viscii_lower; - extern Lisp_Object Vcharset_latin_viscii_upper; -extern Emchar latin_viscii_upper_to_ucs[96]; -extern Emchar_to_byte_table* ucs_to_latin_viscii_upper; /************************************************************************/ @@ -81,7 +68,9 @@ typedef int Charset_ID; #define LEADING_BYTE_UCS_BMP 0x80 #define LEADING_BYTE_CONTROL_1 0x81 /* represent normal 80-9F */ - +#define LEADING_BYTE_HIRAGANA_JISX0208 0x82 +#define LEADING_BYTE_KATAKANA_JISX0208 0x83 +#define LEADING_BYTE_LATIN_VISCII 0x84 #define CHARSET_ID_OFFSET_94 0x55 @@ -217,11 +206,6 @@ struct Lisp_Charset /* Final byte of this character set in ISO2022 designating escape sequence */ Bufbyte final; - /* Number of bytes (1 - 4) required in the internal representation - for characters in this character set. This is *not* the - same as the dimension of the character set). */ - unsigned int rep_bytes; - /* Number of columns a character in this charset takes up, on TTY devices. Not used for X devices. */ unsigned int columns; @@ -240,6 +224,21 @@ struct Lisp_Charset /* Which half of font to be used to display this character set */ unsigned int graphic; + + /* 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; + + /* Offset for external code */ + Emchar code_offset; + + /* Offset for each byte */ + Emchar byte_offset; }; DECLARE_LRECORD (charset, struct Lisp_Charset); @@ -250,12 +249,14 @@ 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 @@ -266,7 +267,6 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define CHARSET_NAME(cs) ((cs)->name) #define CHARSET_SHORT_NAME(cs) ((cs)->short_name) #define CHARSET_LONG_NAME(cs) ((cs)->long_name) -#define CHARSET_REP_BYTES(cs) ((cs)->rep_bytes) #define CHARSET_COLUMNS(cs) ((cs)->columns) #define CHARSET_GRAPHIC(cs) ((cs)->graphic) #define CHARSET_TYPE(cs) ((cs)->type) @@ -278,14 +278,18 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define CHARSET_DIMENSION(cs) ((cs)->dimension) #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_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)) #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_REP_BYTES(cs) CHARSET_REP_BYTES (XCHARSET (cs)) -#define XCHARSET_COLUMNS(cs) CHARSET_COLUMNS (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)) @@ -298,6 +302,12 @@ DECLARE_LRECORD (charset, struct Lisp_Charset); #define XCHARSET_CHARS(cs) CHARSET_CHARS (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_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]; @@ -347,9 +357,18 @@ 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) @@ -368,41 +387,47 @@ INLINE Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2); INLINE Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2) { - 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_latin_iso8859_2)) - return latin_iso8859_2_to_ucs[c1 - 32]; - else if (EQ (charset, Vcharset_latin_iso8859_3)) - return latin_iso8859_3_to_ucs[c1 - 32]; - else if (EQ (charset, Vcharset_latin_iso8859_4)) - return latin_iso8859_4_to_ucs[c1 - 32]; - 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_latin_iso8859_9)) - return latin_iso8859_9_to_ucs[c1 - 32]; - 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 32; - else if (EQ (charset, Vcharset_latin_jisx0201)) - return latin_jisx0201_to_ucs[c1 - 33]; - else if (EQ (charset, Vcharset_latin_viscii_lower)) - return latin_viscii_lower_to_ucs[c1 - 32]; - else if (EQ (charset, Vcharset_latin_viscii_upper)) - return latin_viscii_upper_to_ucs[c1 - 32]; + /* 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)) @@ -435,100 +460,78 @@ 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_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 <= MAX_CHAR_BASIC_LATIN) - { - *charset = Vcharset_ascii; - *c1 = c; - *c2 = 0; - } - else if (c < 0xA0) - { - *charset = Vcharset_control_1; - *c1 = c & 0x7f; - *c2 = 0; - } - else if (c <= 0xff) - { - *charset = Vcharset_latin_iso8859_1; - *c1 = c & 0x7f; - *c2 = 0; - } - else if (c <= 0x17f) + if (c < MIN_CHAR_OBS_94x94) { - *charset - = CHARSET_BY_LEADING_BYTE (latin_a_char_to_charset[c - 0x100]); - - if (EQ (*charset, Vcharset_latin_iso8859_2)) - { - *c1 = get_byte_from_character_table (c, ucs_to_latin_iso8859_2); - *c2 = 0; - } - else if (EQ (*charset, Vcharset_latin_iso8859_3)) + Lisp_Object charsets = Vdefault_coded_charset_priority_list; + while (!EQ (charsets, Qnil)) { - *c1 = get_byte_from_character_table (c, ucs_to_latin_iso8859_3); - *c2 = 0; + *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); } - else if (EQ (*charset, Vcharset_latin_iso8859_4)) + /* otherwise --- maybe for bootstrap */ + if (c <= MAX_CHAR_BASIC_LATIN) { - *c1 = get_byte_from_character_table (c, ucs_to_latin_iso8859_4); - *c2 = 0; + *charset = Vcharset_ascii; + *c1 = charset_get_byte1 (*charset, c); + *c2 = charset_get_byte2 (*charset, c); } - else if (EQ (*charset, Vcharset_latin_iso8859_9)) + else if (c < 0xA0) { - *c1 = get_byte_from_character_table (c, ucs_to_latin_iso8859_9); - *c2 = 0; + *charset = Vcharset_control_1; + *c1 = charset_get_byte1 (*charset, c); + *c2 = charset_get_byte2 (*charset, c); } - else + else if (c <= 0xff) { - *c1 = latin_a_char_to_byte1[c - 0x100]; - *c2 = latin_a_char_to_byte2[c - 0x100]; + *charset = Vcharset_latin_iso8859_1; + *c1 = charset_get_byte1 (*charset, c); + *c2 = charset_get_byte2 (*charset, c); } - } - else if (c < MIN_CHAR_GREEK) - { - if ( (*c1 = get_byte_from_character_table (c, ucs_to_latin_iso8859_2)) ) + else if ((MIN_CHAR_GREEK <= c) && (c <= MAX_CHAR_GREEK)) { - *charset = Vcharset_latin_iso8859_2; + *charset = Vcharset_greek_iso8859_7; + *c1 = c - MIN_CHAR_GREEK + 0x20; *c2 = 0; } - else if ( (*c1 = - get_byte_from_character_table (c, ucs_to_latin_iso8859_3)) ) + else if ((MIN_CHAR_CYRILLIC <= c) && (c <= MAX_CHAR_CYRILLIC)) { - *charset = Vcharset_latin_iso8859_3; + *charset = Vcharset_cyrillic_iso8859_5; + *c1 = c - MIN_CHAR_CYRILLIC + 0x20; *c2 = 0; } - else if ( (*c1 = - get_byte_from_character_table (c, ucs_to_latin_iso8859_4)) ) + else if ((MIN_CHAR_HEBREW <= c) && (c <= MAX_CHAR_HEBREW)) { - *charset = Vcharset_latin_iso8859_4; + *charset = Vcharset_hebrew_iso8859_8; + *c1 = c - MIN_CHAR_HEBREW + 0x20; *c2 = 0; } - else if ( (*c1 = - get_byte_from_character_table (c, ucs_to_latin_iso8859_9)) ) + else if ((MIN_CHAR_THAI <= c) && (c <= MAX_CHAR_THAI)) { - *charset = Vcharset_latin_iso8859_9; + *charset = Vcharset_thai_tis620; + *c1 = c - MIN_CHAR_THAI + 0x20; *c2 = 0; } - else if ( (*c1 = - get_byte_from_character_table (c, - ucs_to_latin_viscii_lower)) ) + else if ((MIN_CHAR_HALFWIDTH_KATAKANA <= c) + && (c <= MAX_CHAR_HALFWIDTH_KATAKANA)) { - *charset = Vcharset_latin_viscii_lower; - *c2 = 0; - } - else if ( (*c1 = - get_byte_from_character_table (c, - ucs_to_latin_viscii_upper)) ) - { - *charset = Vcharset_latin_viscii_upper; + *charset = Vcharset_katakana_jisx0201; + *c1 = c - MIN_CHAR_HALFWIDTH_KATAKANA + 33; *c2 = 0; } else @@ -538,79 +541,14 @@ breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2) *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) + else if (c <= MAX_CHAR_OBS_94x94) { - *charset = Vcharset_thai_tis620; - *c1 = c - MIN_CHAR_THAI + 0x20; - *c2 = 0; - } - else if (c < MIN_CHAR_HALFWIDTH_KATAKANA) - { - if ( (*c1 = get_byte_from_character_table (c, ucs_to_latin_jisx0201)) ) - { - *charset = Vcharset_latin_jisx0201; - *c2 = 0; - } - else if ( (*c1 = get_byte_from_character_table - (c, ucs_to_latin_viscii_lower)) ) - { - *charset = Vcharset_latin_viscii_lower; - *c2 = 0; - } - else if ( (*c1 = get_byte_from_character_table - (c, ucs_to_latin_viscii_upper)) ) - { - *charset = Vcharset_latin_viscii_upper; - *c2 = 0; - } - else - { - *charset = Vcharset_ucs_bmp; - *c1 = c >> 8; - *c2 = c & 0xff; - } - } - else if (c <= MAX_CHAR_HALFWIDTH_KATAKANA) - { - *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) { @@ -671,22 +609,13 @@ CHAR_CHARSET (Emchar ch) #define CHAR_LEADING_BYTE(c) (XCHARSET_LEADING_BYTE(CHAR_CHARSET(c))) - -#ifdef ENABLE_COMPOSITE_CHARS -/************************************************************************/ -/* Composite characters */ -/************************************************************************/ - -Emchar lookup_composite_char (Bufbyte *str, int len); -Lisp_Object composite_char_string (Emchar ch); -#endif /* ENABLE_COMPOSITE_CHARS */ +#define CHAR_COLUMNS(c) (CHARSET_COLUMNS(XCHARSET(CHAR_CHARSET(c)))) /************************************************************************/ /* Exported functions */ /************************************************************************/ -EXFUN (Ffind_charset, 1); EXFUN (Fget_charset, 1); extern Lisp_Object Vcharset_chinese_big5_1;