X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fmule-charset.c;h=751fd32c25a2b6fbf9d10d14da1297e43b806e3d;hb=a699139716d7a947173ebc9a7130cc0eead5da7f;hp=cace922aba3420bf5ec6c3f6149a0ba553db3312;hpb=42925f796a6f8c643f9ed40412d3069e1c09dc2a;p=chise%2Fxemacs-chise.git- diff --git a/src/mule-charset.c b/src/mule-charset.c index cace922..751fd32 100644 --- a/src/mule-charset.c +++ b/src/mule-charset.c @@ -1,7 +1,7 @@ /* Functions to handle multilingual characters. Copyright (C) 1992, 1995 Free Software Foundation, Inc. Copyright (C) 1995 Sun Microsystems, Inc. - Copyright (C) 1999,2000,2001 MORIOKA Tomohiko + Copyright (C) 1999,2000,2001,2002 MORIOKA Tomohiko This file is part of XEmacs. @@ -77,7 +77,7 @@ Lisp_Object Vcharset_latin_tcvn5712; Lisp_Object Vcharset_latin_viscii_lower; Lisp_Object Vcharset_latin_viscii_upper; Lisp_Object Vcharset_chinese_big5; -Lisp_Object Vcharset_chinese_big5_cdp; +/* Lisp_Object Vcharset_chinese_big5_cdp; */ Lisp_Object Vcharset_ideograph_hanziku_1; Lisp_Object Vcharset_ideograph_hanziku_2; Lisp_Object Vcharset_ideograph_hanziku_3; @@ -176,20 +176,6 @@ 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 +204,6 @@ 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 +213,6 @@ 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 +262,10 @@ 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 +282,12 @@ 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; } @@ -428,7 +350,7 @@ Lisp_Object Qascii, Qvietnamese_viscii_lower, Qvietnamese_viscii_upper, Qchinese_big5, - Qchinese_big5_cdp, + /* Qchinese_big5_cdp, */ Qideograph_hanziku_1, Qideograph_hanziku_2, Qideograph_hanziku_3, @@ -895,7 +817,7 @@ 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; @@ -984,11 +906,95 @@ 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); +#ifdef HAVE_DATABASE + if (EQ (decoding_table, Qunloaded) || + EQ (decoding_table, Qunbound) || + NILP (decoding_table) ) + { + char_id = load_char_decoding_entry_maybe (ccs, code_point); + } +#endif + 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 (EQ (charset, Vcharset_chinese_big5)) + if ( CHARSETP (mother) && (XCHARSET_MAX_CODE (charset) > 0) ) + { + int code = code_point; + + if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 ) + { + int row = code_point >> 8; + int cell = code_point & 255; + + if (row < 16 + 32) + return -1; + else if (row < 16 + 32 + 30) + code = (row - (16 + 32)) * 94 + cell - 33; + else if (row < 18 + 32 + 30) + return -1; + else if (row < 18 + 32 + 60) + code = (row - (18 + 32)) * 94 + cell - 33; + } + else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x60 ) + { + int plane = code_point >> 16; + int row = (code_point >> 8) & 255; + int cell = code_point & 255; + + if (row < 16 + 32) + return -1; + else if (row < 16 + 32 + 30) + code + = (plane - 33) * 94 * 60 + + (row - (16 + 32)) * 94 + + cell - 33; + else if (row < 18 + 32 + 30) + return -1; + else if (row < 18 + 32 + 60) + code + = (plane - 33) * 94 * 60 + + (row - (18 + 32)) * 94 + + cell - 33; + } + return + decode_builtin_char (mother, code + XCHARSET_CODE_OFFSET(charset)); + } +#if 0 + else if (EQ (charset, Vcharset_chinese_big5)) { int c1 = code_point >> 8; int c2 = code_point & 0xFF; @@ -1014,6 +1020,7 @@ decode_builtin_char (Lisp_Object charset, int code_point) code_point = ((I / 94 + 33) << 8) | (I % 94 + 33); } } +#endif if ((final = XCHARSET_FINAL (charset)) >= '0') { if (XCHARSET_DIMENSION (charset) == 1) @@ -1092,7 +1099,8 @@ charset_code_point (Lisp_Object charset, Emchar ch) code = charset_code_point (mother, ch); else code = ch; - if ( (min <= code) && (code <= max) ) + if ( ((max == 0) && CHARSETP (mother)) || + ((min <= code) && (code <= max)) ) { int d = code - XCHARSET_CODE_OFFSET (charset); @@ -1451,6 +1459,7 @@ character set. Recognized properties are: this character set. 'dimension Number of octets used to index a character in this charset. Either 1 or 2. Defaults to 1. + If UTF-2000 feature is enabled, 3 or 4 are also available. 'columns Number of columns used to display a character in this charset. Only used in TTY mode. (Under X, the actual width of a character can be derived from the font used to display the @@ -1459,6 +1468,7 @@ character set. Recognized properties are: 'chars Number of characters in each dimension (94 or 96). Defaults to 94. Note that if the dimension is 2, the character set thus described is 94x94 or 96x96. + If UTF-2000 feature is enabled, 128 or 256 are also available. 'final Final byte of ISO 2022 escape sequence. Must be supplied. Each combination of (DIMENSION, CHARS) defines a separate namespace for final bytes. Note that ISO @@ -1483,6 +1493,13 @@ character set. Recognized properties are: is passed the octets of the character, with the high bit cleared and set depending upon whether the value of the 'graphic property is 0 or 1. +'mother [UTF-2000 only] Base coded-charset. +'code-min [UTF-2000 only] Minimum code-point of a base coded-charset. +'code-max [UTF-2000 only] Maximum code-point of a base coded-charset. +'code-offset [UTF-2000 only] Offset for a code-point of a base + coded-charset. +'conversion [UTF-2000 only] Conversion for a code-point of a base + coded-charset (94x60 or 94x94x60). */ (name, doc_string, props)) { @@ -2050,7 +2067,183 @@ Set mapping-table of CHARSET to TABLE. } return table; } + +DEFUN ("save-charset-mapping-table", Fsave_charset_mapping_table, 1, 1, 0, /* +Save mapping-table of CHARSET. +*/ + (charset)) +{ +#ifdef HAVE_DATABASE + struct Lisp_Charset *cs; + int byte_min, byte_max; + Lisp_Object db; + Lisp_Object db_file; + + charset = Fget_charset (charset); + cs = XCHARSET (charset); + + db_file = char_attribute_system_db_file (CHARSET_NAME (cs), + Qsystem_char_id, 1); + db = Fopen_database (db_file, Qnil, Qnil, build_string ("w+"), Qnil); + + byte_min = CHARSET_BYTE_OFFSET (cs); + byte_max = byte_min + CHARSET_BYTE_SIZE (cs); + switch (CHARSET_DIMENSION (cs)) + { + case 1: + { + Lisp_Object table_c = XCHARSET_DECODING_TABLE (charset); + int cell; + + for (cell = byte_min; cell < byte_max; cell++) + { + Lisp_Object c = get_ccs_octet_table (table_c, charset, cell); + + if (CHARP (c)) + Fput_database (Fprin1_to_string (make_int (cell), Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); + } + } + break; + case 2: + { + Lisp_Object table_r = XCHARSET_DECODING_TABLE (charset); + int row; + + for (row = byte_min; row < byte_max; row++) + { + Lisp_Object table_c = get_ccs_octet_table (table_r, charset, row); + int cell; + + for (cell = byte_min; cell < byte_max; cell++) + { + Lisp_Object c = get_ccs_octet_table (table_c, charset, cell); + + if (CHARP (c)) + Fput_database (Fprin1_to_string (make_int ((row << 8) + | cell), + Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); + } + } + } + break; + case 3: + { + Lisp_Object table_p = XCHARSET_DECODING_TABLE (charset); + int plane; + + for (plane = byte_min; plane < byte_max; plane++) + { + Lisp_Object table_r + = get_ccs_octet_table (table_p, charset, plane); + int row; + + for (row = byte_min; row < byte_max; row++) + { + Lisp_Object table_c + = get_ccs_octet_table (table_r, charset, row); + int cell; + + for (cell = byte_min; cell < byte_max; cell++) + { + Lisp_Object c = get_ccs_octet_table (table_c, charset, + cell); + + if (CHARP (c)) + Fput_database (Fprin1_to_string (make_int ((plane << 16) + | (row << 8) + | cell), + Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); + } + } + } + } + break; + default: + { + Lisp_Object table_g = XCHARSET_DECODING_TABLE (charset); + int group; + + for (group = byte_min; group < byte_max; group++) + { + Lisp_Object table_p + = get_ccs_octet_table (table_g, charset, group); + int plane; + + for (plane = byte_min; plane < byte_max; plane++) + { + Lisp_Object table_r + = get_ccs_octet_table (table_p, charset, plane); + int row; + + for (row = byte_min; row < byte_max; row++) + { + Lisp_Object table_c + = get_ccs_octet_table (table_r, charset, row); + int cell; + + for (cell = byte_min; cell < byte_max; cell++) + { + Lisp_Object c + = get_ccs_octet_table (table_c, charset, cell); + + if (CHARP (c)) + Fput_database (Fprin1_to_string + (make_int (( group << 24) + | (plane << 16) + | (row << 8) + | cell), + Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); + } + } + } + } + } + } + return Fclose_database (db); +#else + return Qnil; #endif +} + +#ifdef HAVE_CHISE_CLIENT +Emchar +load_char_decoding_entry_maybe (Lisp_Object ccs, int code_point) +{ + Lisp_Object db; + Lisp_Object db_file + = char_attribute_system_db_file (XCHARSET_NAME(ccs), Qsystem_char_id, + 0); + + db = Fopen_database (db_file, Qnil, Qnil, build_string ("r"), Qnil); + if (!NILP (db)) + { + Lisp_Object ret + = Fget_database (Fprin1_to_string (make_int (code_point), Qnil), + db, Qnil); + if (!NILP (ret)) + { + ret = Fread (ret); + if (CHARP (ret)) + { + decoding_table_put_char (ccs, code_point, ret); + Fclose_database (db); + return XCHAR (ret); + } + } + Fclose_database (db); + } + return -1; +} +#endif /* HAVE_CHISE_CLIENT */ +#endif /* UTF2000 */ /************************************************************************/ @@ -2075,7 +2268,7 @@ 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; } @@ -2389,6 +2582,7 @@ syms_of_mule_charset (void) DEFSUBR (Fdecode_char); DEFSUBR (Fdecode_builtin_char); DEFSUBR (Fencode_char); + DEFSUBR (Fsave_charset_mapping_table); #endif DEFSUBR (Fmake_char); DEFSUBR (Fchar_charset); @@ -2476,7 +2670,7 @@ syms_of_mule_charset (void) defsymbol (&Qideograph_daikanwa_2, "ideograph-daikanwa-2"); defsymbol (&Qideograph_daikanwa, "ideograph-daikanwa"); defsymbol (&Qchinese_big5, "chinese-big5"); - defsymbol (&Qchinese_big5_cdp, "chinese-big5-cdp"); + /* defsymbol (&Qchinese_big5_cdp, "chinese-big5-cdp"); */ defsymbol (&Qideograph_hanziku_1, "ideograph-hanziku-1"); defsymbol (&Qideograph_hanziku_2, "ideograph-hanziku-2"); defsymbol (&Qideograph_hanziku_3, "ideograph-hanziku-3"); @@ -2570,7 +2764,7 @@ 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, @@ -2609,17 +2803,17 @@ 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, @@ -2628,7 +2822,7 @@ 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, @@ -2637,7 +2831,7 @@ 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 @@ -2913,11 +3107,11 @@ complex_vars_of_mule_charset (void) build_string ("Big5"), build_string ("Big5"), build_string ("Big5 Chinese traditional"), - build_string ("big5"), + 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 = make_charset (LEADING_BYTE_CHINESE_BIG5_CDP, Qchinese_big5_cdp, 256, 2, @@ -2928,6 +3122,7 @@ complex_vars_of_mule_charset (void) build_string ("big5\\.cdp-0"), Qnil, MIN_CHAR_BIG5_CDP, MAX_CHAR_BIG5_CDP, MIN_CHAR_BIG5_CDP, 0, Qnil, CONVERSION_IDENTICAL); +#endif #define DEF_HANZIKU(n) \ staticpro (&Vcharset_ideograph_hanziku_##n); \ Vcharset_ideograph_hanziku_##n = \