X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fmule-charset.c;h=2031052975a38ca6c72c427b3d6154b1568c3ac4;hb=aa982acf01b4e35675a624d78c9e5ee109c1203e;hp=d6602d0c5ef68a57a9f3a2ea9eb72d6ebf2c0c62;hpb=2b41d33b3497808bf02b2393893cd09a00538853;p=chise%2Fxemacs-chise.git diff --git a/src/mule-charset.c b/src/mule-charset.c index d6602d0..2031052 100644 --- a/src/mule-charset.c +++ b/src/mule-charset.c @@ -76,6 +76,7 @@ Lisp_Object Vcharset_latin_viscii; Lisp_Object Vcharset_latin_tcvn5712; Lisp_Object Vcharset_latin_viscii_lower; Lisp_Object Vcharset_latin_viscii_upper; +Lisp_Object Vcharset_jis_x0208; Lisp_Object Vcharset_chinese_big5; /* Lisp_Object Vcharset_chinese_big5_cdp; */ Lisp_Object Vcharset_ideograph_hanziku_1; @@ -176,20 +177,6 @@ const Bytecount rep_bytes_by_first_byte[0xA0] = #ifdef UTF2000 -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); -} - -#define XCHARSET_BYTE_SIZE(ccs) CHARSET_BYTE_SIZE (XCHARSET (ccs)) - int decoding_table_check_elements (Lisp_Object v, int dim, int ccs_len); int decoding_table_check_elements (Lisp_Object v, int dim, int ccs_len) @@ -218,58 +205,6 @@ decoding_table_check_elements (Lisp_Object v, int dim, int ccs_len) return 0; } -INLINE_HEADER void -decoding_table_remove_char (Lisp_Object v, int dim, int byte_offset, - int code_point); -INLINE_HEADER void -decoding_table_remove_char (Lisp_Object v, int dim, int byte_offset, - int code_point) -{ - int i = -1; - - while (dim > 0) - { - Lisp_Object nv; - - dim--; - i = ((code_point >> (8 * dim)) & 255) - byte_offset; - nv = XVECTOR_DATA(v)[i]; - if (!VECTORP (nv)) - break; - v = nv; - } - if (i >= 0) - XVECTOR_DATA(v)[i] = Qnil; -} - -INLINE_HEADER void -decoding_table_put_char (Lisp_Object v, int dim, int byte_offset, - int code_point, Lisp_Object character); -INLINE_HEADER void -decoding_table_put_char (Lisp_Object v, int dim, int byte_offset, - int code_point, Lisp_Object character) -{ - int i = -1; - Lisp_Object nv; - int ccs_len = XVECTOR_LENGTH (v); - - while (dim > 0) - { - dim--; - i = ((code_point >> (8 * dim)) & 255) - byte_offset; - nv = XVECTOR_DATA(v)[i]; - if (dim > 0) - { - if (!VECTORP (nv)) - nv = (XVECTOR_DATA(v)[i] = make_vector (ccs_len, Qnil)); - v = nv; - } - else - break; - } - XVECTOR_DATA(v)[i] = character; -} - Lisp_Object put_char_ccs_code_point (Lisp_Object character, Lisp_Object ccs, Lisp_Object value) @@ -279,9 +214,6 @@ put_char_ccs_code_point (Lisp_Object character, || (XCHAR (character) != XINT (value))) { Lisp_Object v = XCHARSET_DECODING_TABLE (ccs); - int dim = XCHARSET_DIMENSION (ccs); - int ccs_len = XCHARSET_BYTE_SIZE (ccs); - int byte_offset = XCHARSET_BYTE_OFFSET (ccs); int code_point; if (CONSP (value)) @@ -331,16 +263,10 @@ put_char_ccs_code_point (Lisp_Object character, Lisp_Object cpos = Fget_char_attribute (character, ccs, Qnil); if (INTP (cpos)) { - decoding_table_remove_char (v, dim, byte_offset, XINT (cpos)); + decoding_table_remove_char (ccs, XINT (cpos)); } } - else - { - XCHARSET_DECODING_TABLE (ccs) - = v = make_vector (ccs_len, Qnil); - } - - decoding_table_put_char (v, dim, byte_offset, code_point, character); + decoding_table_put_char (ccs, code_point, character); } return value; } @@ -357,15 +283,12 @@ remove_char_ccs (Lisp_Object character, Lisp_Object ccs) if (!NILP (cpos)) { - decoding_table_remove_char (decoding_table, - XCHARSET_DIMENSION (ccs), - XCHARSET_BYTE_OFFSET (ccs), - XINT (cpos)); + decoding_table_remove_char (ccs, XINT (cpos)); } } if (CHAR_TABLEP (encoding_table)) { - put_char_id_table (XCHAR_TABLE(encoding_table), character, Qnil); + put_char_id_table (XCHAR_TABLE(encoding_table), character, Qunbound); } return Qt; } @@ -427,6 +350,7 @@ Lisp_Object Qascii, Qlatin_viscii_upper, Qvietnamese_viscii_lower, Qvietnamese_viscii_upper, + Qjis_x0208, Qchinese_big5, /* Qchinese_big5_cdp, */ Qideograph_hanziku_1, @@ -895,7 +819,7 @@ make_charset (Charset_ID id, Lisp_Object name, CHARSET_CCL_PROGRAM (cs) = Qnil; CHARSET_REVERSE_DIRECTION_CHARSET (cs) = Qnil; #ifdef UTF2000 - CHARSET_DECODING_TABLE(cs) = Qnil; + CHARSET_DECODING_TABLE(cs) = Qunbound; CHARSET_MIN_CODE (cs) = min_code; CHARSET_MAX_CODE (cs) = max_code; CHARSET_CODE_OFFSET (cs) = code_offset; @@ -984,12 +908,44 @@ get_unallocated_leading_byte (int dimension) #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) Emchar +decode_defined_char (Lisp_Object ccs, int code_point) +{ + int dim = XCHARSET_DIMENSION (ccs); + Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (ccs); + Emchar char_id = -1; + Lisp_Object mother; + + while (dim > 0) + { + dim--; + decoding_table + = get_ccs_octet_table (decoding_table, ccs, + (code_point >> (dim * 8)) & 255); + } + if (CHARP (decoding_table)) + return XCHAR (decoding_table); + if (char_id >= 0) + return char_id; + else if ( CHARSETP (mother = XCHARSET_MOTHER (ccs)) ) + { + if ( XCHARSET_CONVERSION (ccs) == CONVERSION_IDENTICAL ) + { + if ( EQ (mother, Vcharset_ucs) ) + return DECODE_CHAR (mother, code_point); + else + return decode_defined_char (mother, code_point); + } + } + return -1; +} + +Emchar decode_builtin_char (Lisp_Object charset, int code_point) { Lisp_Object mother = XCHARSET_MOTHER (charset); int final; - if ( CHARSETP (mother) ) + if ( CHARSETP (mother) && (XCHARSET_MAX_CODE (charset) > 0) ) { int code = code_point; @@ -1028,35 +984,26 @@ decode_builtin_char (Lisp_Object charset, int code_point) + (row - (18 + 32)) * 94 + cell - 33; } - return DECODE_CHAR (mother, code + XCHARSET_CODE_OFFSET(charset)); + return + decode_builtin_char (mother, code + XCHARSET_CODE_OFFSET(charset)); } - else if (EQ (charset, Vcharset_chinese_big5)) + if (XCHARSET_MAX_CODE (charset)) { - int c1 = code_point >> 8; - int c2 = code_point & 0xFF; - unsigned int I; - - if ( ( (0xA1 <= c1) && (c1 <= 0xFE) ) - && - ( ((0x40 <= c2) && (c2 <= 0x7E)) || - ((0xA1 <= c2) && (c2 <= 0xFE)) ) ) - { - I = (c1 - 0xA1) * BIG5_SAME_ROW - + c2 - (c2 < 0x7F ? 0x40 : 0x62); - - if (c1 < 0xC9) - { - charset = Vcharset_chinese_big5_1; - } - else - { - charset = Vcharset_chinese_big5_2; - I -= (BIG5_SAME_ROW) * (0xC9 - 0xA1); - } - code_point = ((I / 94 + 33) << 8) | (I % 94 + 33); - } + Emchar cid + = (XCHARSET_DIMENSION (charset) == 1 + ? + code_point - XCHARSET_BYTE_OFFSET (charset) + : + ((code_point >> 8) - XCHARSET_BYTE_OFFSET (charset)) + * XCHARSET_CHARS (charset) + + (code_point & 0xFF) - XCHARSET_BYTE_OFFSET (charset)) + + XCHARSET_CODE_OFFSET (charset); + if ((cid < XCHARSET_MIN_CODE (charset)) + || (XCHARSET_MAX_CODE (charset) < cid)) + return -1; + return cid; } - if ((final = XCHARSET_FINAL (charset)) >= '0') + else if ((final = XCHARSET_FINAL (charset)) >= '0') { if (XCHARSET_DIMENSION (charset) == 1) { @@ -1093,28 +1040,12 @@ decode_builtin_char (Lisp_Object charset, int code_point) } } } - else if (XCHARSET_MAX_CODE (charset)) - { - Emchar cid - = (XCHARSET_DIMENSION (charset) == 1 - ? - code_point - XCHARSET_BYTE_OFFSET (charset) - : - ((code_point >> 8) - XCHARSET_BYTE_OFFSET (charset)) - * XCHARSET_CHARS (charset) - + (code_point & 0xFF) - XCHARSET_BYTE_OFFSET (charset)) - + XCHARSET_CODE_OFFSET (charset); - if ((cid < XCHARSET_MIN_CODE (charset)) - || (XCHARSET_MAX_CODE (charset) < cid)) - return -1; - return cid; - } else return -1; } int -charset_code_point (Lisp_Object charset, Emchar ch) +charset_code_point (Lisp_Object charset, Emchar ch, int defined_only) { Lisp_Object encoding_table = XCHARSET_ENCODING_TABLE (charset); Lisp_Object ret; @@ -1131,14 +1062,23 @@ charset_code_point (Lisp_Object charset, Emchar ch) int code; if ( CHARSETP (mother) ) - code = charset_code_point (mother, ch); + code = charset_code_point (mother, ch, defined_only); + else if (defined_only) + return -1; else code = ch; - if ( (min <= code) && (code <= max) ) + if ( ((max == 0) && CHARSETP (mother)) || + ((min <= code) && (code <= max)) ) { int d = code - XCHARSET_CODE_OFFSET (charset); - if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 ) + if ( XCHARSET_CONVERSION (charset) == CONVERSION_IDENTICAL ) + return d; + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94 ) + return d + 33; + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96 ) + return d + 32; + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 ) { int row = d / 94; int cell = d % 94 + 33; @@ -1149,6 +1089,10 @@ charset_code_point (Lisp_Object charset, Emchar ch) row += 18 + 32; return (row << 8) | cell; } + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94 ) + return ((d / 94 + 33) << 8) | (d % 94 + 33); + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96 ) + return ((d / 96 + 32) << 8) | (d % 96 + 32); else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x60 ) { int plane = d / (94 * 60) + 33; @@ -1161,48 +1105,42 @@ charset_code_point (Lisp_Object charset, Emchar ch) row += 18 + 32; return (plane << 16) | (row << 8) | cell; } - else if (XCHARSET_CHARS (charset) == 94) - { - if (XCHARSET_DIMENSION (charset) == 1) - return d + 33; - else if (XCHARSET_DIMENSION (charset) == 2) - return ((d / 94 + 33) << 8) | (d % 94 + 33); - else if (XCHARSET_DIMENSION (charset) == 3) - return - ( (d / (94 * 94) + 33) << 16) - | ((d / 94 % 94 + 33) << 8) - | (d % 94 + 33); - else /* if (XCHARSET_DIMENSION (charset) == 4) */ - return - ( (d / (94 * 94 * 94) + 33) << 24) - | ((d / (94 * 94) % 94 + 33) << 16) - | ((d / 94 % 94 + 33) << 8) - | (d % 94 + 33); - } - else if (XCHARSET_CHARS (charset) == 96) + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x94 ) + return + ( (d / (94 * 94) + 33) << 16) + | ((d / 94 % 94 + 33) << 8) + | (d % 94 + 33); + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96x96 ) + return + ( (d / (96 * 96) + 32) << 16) + | ((d / 96 % 96 + 32) << 8) + | (d % 96 + 32); + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x94x94 ) + return + ( (d / (94 * 94 * 94) + 33) << 24) + | ((d / (94 * 94) % 94 + 33) << 16) + | ((d / 94 % 94 + 33) << 8) + | (d % 94 + 33); + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96x96x96 ) + return + ( (d / (96 * 96 * 96) + 32) << 24) + | ((d / (96 * 96) % 96 + 32) << 16) + | ((d / 96 % 96 + 32) << 8) + | (d % 96 + 32); + else { - if (XCHARSET_DIMENSION (charset) == 1) - return d + 32; - else if (XCHARSET_DIMENSION (charset) == 2) - return ((d / 96 + 32) << 8) | (d % 96 + 32); - else if (XCHARSET_DIMENSION (charset) == 3) - return - ( (d / (96 * 96) + 32) << 16) - | ((d / 96 % 96 + 32) << 8) - | (d % 96 + 32); - else /* if (XCHARSET_DIMENSION (charset) == 4) */ - return - ( (d / (96 * 96 * 96) + 32) << 24) - | ((d / (96 * 96) % 96 + 32) << 16) - | ((d / 96 % 96 + 32) << 8) - | (d % 96 + 32); + printf ("Unknown CCS-conversion %d is specified!", + XCHARSET_CONVERSION (charset)); + exit (-1); } - else - return code - XCHARSET_CODE_OFFSET (charset); } - else if ( (XCHARSET_CODE_OFFSET (charset) == 0) || + else if ( ( XCHARSET_FINAL (charset) >= '0' ) && + ( XCHARSET_MIN_CODE (charset) == 0 ) + /* + (XCHARSET_CODE_OFFSET (charset) == 0) || (XCHARSET_CODE_OFFSET (charset) - == XCHARSET_MIN_CODE (charset)) ) + == XCHARSET_MIN_CODE (charset)) + */ ) { int d; @@ -2126,7 +2064,7 @@ If corresponding character is not found, nil is returned. if (NILP (defined_only)) c = DECODE_CHAR (charset, c); else - c = DECODE_DEFINED_CHAR (charset, c); + c = decode_defined_char (charset, c); return c >= 0 ? make_char (c) : Qnil; } @@ -2266,16 +2204,17 @@ N defaults to 0 if omitted. } #ifdef UTF2000 -DEFUN ("encode-char", Fencode_char, 2, 2, 0, /* +DEFUN ("encode-char", Fencode_char, 2, 3, 0, /* Return code-point of CHARACTER in specified CHARSET. */ - (character, charset)) + (character, charset, defined_only)) { int code_point; CHECK_CHAR_COERCE_INT (character); charset = Fget_charset (charset); - code_point = charset_code_point (charset, XCHAR (character)); + code_point = charset_code_point (charset, XCHAR (character), + !NILP (defined_only)); if (code_point >= 0) return make_int (code_point); else @@ -2512,6 +2451,7 @@ syms_of_mule_charset (void) defsymbol (&Qlatin_viscii_upper, "latin-viscii-upper"); defsymbol (&Qvietnamese_viscii_lower, "vietnamese-viscii-lower"); defsymbol (&Qvietnamese_viscii_upper, "vietnamese-viscii-upper"); + defsymbol (&Qjis_x0208, "=jis-x0208"); defsymbol (&Qideograph_gt, "ideograph-gt"); defsymbol (&Qideograph_gt_pj_1, "ideograph-gt-pj-1"); defsymbol (&Qideograph_gt_pj_2, "ideograph-gt-pj-2"); @@ -2621,7 +2561,7 @@ complex_vars_of_mule_charset (void) build_string ("UCS"), build_string ("ISO/IEC 10646"), build_string (""), - Qnil, 0, 0xFFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0x7FFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs_bmp); Vcharset_ucs_bmp = make_charset (LEADING_BYTE_UCS_BMP, Qucs_bmp, 256, 2, @@ -2660,17 +2600,17 @@ complex_vars_of_mule_charset (void) build_string ("UCS for CNS 11643"), build_string ("ISO/IEC 10646 for CNS 11643"), build_string (""), - Qnil, 0, 0, 0, 0, - Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs_jis); Vcharset_ucs_jis = make_charset (LEADING_BYTE_UCS_JIS, Qucs_jis, 256, 3, 2, 2, 0, CHARSET_LEFT_TO_RIGHT, build_string ("UCS for JIS"), build_string ("UCS for JIS X 0208, 0212 and 0213"), - build_string ("ISO/IEC 10646 for JIS X 0208, 0212 and 0213"), + build_string + ("ISO/IEC 10646 for JIS X 0208, 0212 and 0213"), build_string (""), - Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs_ks); Vcharset_ucs_ks = make_charset (LEADING_BYTE_UCS_KS, Qucs_ks, 256, 3, @@ -2679,7 +2619,7 @@ complex_vars_of_mule_charset (void) build_string ("UCS for CCS defined by KS"), build_string ("ISO/IEC 10646 for Korean Standards"), build_string (""), - Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs_big5); Vcharset_ucs_big5 = make_charset (LEADING_BYTE_UCS_BIG5, Qucs_big5, 256, 3, @@ -2688,7 +2628,7 @@ complex_vars_of_mule_charset (void) build_string ("UCS for Big5"), build_string ("ISO/IEC 10646 for Big5"), build_string (""), - Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL); #else # define MIN_CHAR_THAI 0 # define MAX_CHAR_THAI 0 @@ -2759,8 +2699,7 @@ complex_vars_of_mule_charset (void) build_string ("TIS620 (Thai)"), build_string ("TIS620.2529 (Thai)"), build_string ("tis620"), - Qnil, MIN_CHAR_THAI, MAX_CHAR_THAI, - MIN_CHAR_THAI, 32, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_greek_iso8859_7); Vcharset_greek_iso8859_7 = make_charset (LEADING_BYTE_GREEK_ISO8859_7, Qgreek_iso8859_7, 96, 1, @@ -2827,6 +2766,21 @@ complex_vars_of_mule_charset (void) build_string ("ISO8859-9 (Latin-5)"), build_string ("iso8859-9"), Qnil, 0, 0, 0, 32, Qnil, CONVERSION_IDENTICAL); +#ifdef UTF2000 + staticpro (&Vcharset_jis_x0208); + Vcharset_jis_x0208 = + make_charset (LEADING_BYTE_JIS_X0208, + Qjis_x0208, 94, 2, + 2, 0, 'B', CHARSET_LEFT_TO_RIGHT, + build_string ("JIS X0208"), + build_string ("JIS X0208 Common"), + build_string ("JIS X0208 Common part"), + build_string ("jisx0208\\.1990"), + Qnil, + MIN_CHAR_JIS_X0208_1990, + MAX_CHAR_JIS_X0208_1990, MIN_CHAR_JIS_X0208_1990, 33, + Qnil, CONVERSION_94x94); +#endif staticpro (&Vcharset_japanese_jisx0208_1978); Vcharset_japanese_jisx0208_1978 = make_charset (LEADING_BYTE_JAPANESE_JISX0208_1978, @@ -2837,7 +2791,13 @@ complex_vars_of_mule_charset (void) build_string ("JIS X0208:1978 Japanese Kanji (so called \"old JIS\")"), build_string ("\\(jisx0208\\|jisc6226\\)\\.1978"), - Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 33, +#ifdef UTF2000 + Vcharset_jis_x0208, +#else + Qnil, +#endif + CONVERSION_IDENTICAL); staticpro (&Vcharset_chinese_gb2312); Vcharset_chinese_gb2312 = make_charset (LEADING_BYTE_CHINESE_GB2312, Qchinese_gb2312, 94, 2, @@ -2864,7 +2824,13 @@ complex_vars_of_mule_charset (void) build_string ("JIS X0208:1983 (Japanese)"), build_string ("JIS X0208:1983 Japanese Kanji"), build_string ("jisx0208\\.1983"), - Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0, 0, 33, +#ifdef UTF2000 + Vcharset_jis_x0208, +#else + Qnil, +#endif + CONVERSION_IDENTICAL); #ifdef UTF2000 staticpro (&Vcharset_japanese_jisx0208_1990); Vcharset_japanese_jisx0208_1990 = @@ -2876,9 +2842,11 @@ complex_vars_of_mule_charset (void) build_string ("JIS X0208:1990 Japanese Kanji"), build_string ("jisx0208\\.1990"), Qnil, - MIN_CHAR_JIS_X0208_1990, - MAX_CHAR_JIS_X0208_1990, MIN_CHAR_JIS_X0208_1990, 33, - Qnil, CONVERSION_IDENTICAL); + 0x2121 /* MIN_CHAR_JIS_X0208_1990 */, + 0x7426 /* MAX_CHAR_JIS_X0208_1990 */, + 0 /* MIN_CHAR_JIS_X0208_1990 */, 33, + Vcharset_jis_x0208 /* Qnil */, + CONVERSION_IDENTICAL /* CONVERSION_94x94 */); #endif staticpro (&Vcharset_korean_ksc5601); Vcharset_korean_ksc5601 = @@ -2966,9 +2934,8 @@ complex_vars_of_mule_charset (void) build_string ("Big5 Chinese traditional"), build_string ("big5-0"), Qnil, - 0 /* MIN_CHAR_BIG5_CDP */, - 0 /* MAX_CHAR_BIG5_CDP */, 0, 0, - Qnil, CONVERSION_IDENTICAL); + MIN_CHAR_BIG5_CDP, MAX_CHAR_BIG5_CDP, + MIN_CHAR_BIG5_CDP, 0, Qnil, CONVERSION_IDENTICAL); #if 0 staticpro (&Vcharset_chinese_big5_cdp); Vcharset_chinese_big5_cdp =