X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;ds=sidebyside;f=src%2Fmule-charset.c;h=e32fcf230a6ee16f6216fd47237366021fcc6e45;hb=6e8f204c9e1f490b2752de46c111744d1deb3ee0;hp=1f6c3f1a876ebafe3e988b82e5ecb1631964c497;hpb=66f49b31d02dd227044cba02e2681a193041b3c9;p=chise%2Fxemacs-chise.git.1 diff --git a/src/mule-charset.c b/src/mule-charset.c index 1f6c3f1..e32fcf2 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) 1999,2000,2001,2002,2003 MORIOKA Tomohiko + Copyright (C) 1999,2000,2001,2002,2003,2004 MORIOKA Tomohiko This file is part of XEmacs. @@ -22,10 +22,10 @@ Boston, MA 02111-1307, USA. */ /* Rewritten by Ben Wing . */ -/* Rewritten by MORIOKA Tomohiko for XEmacs UTF-2000. */ +/* Rewritten by MORIOKA Tomohiko for XEmacs CHISE. */ #include -#ifdef CHISE +#ifdef HAVE_LIBCHISE #include #endif #ifdef UTF2000 @@ -67,6 +67,7 @@ Lisp_Object Vcharset_japanese_jisx0212; Lisp_Object Vcharset_chinese_cns11643_1; Lisp_Object Vcharset_chinese_cns11643_2; #ifdef UTF2000 +Lisp_Object Vcharset_system_char_id; Lisp_Object Vcharset_ucs; Lisp_Object Vcharset_ucs_bmp; Lisp_Object Vcharset_ucs_smp; @@ -175,13 +176,101 @@ decoding_table_check_elements (Lisp_Object v, int dim, int ccs_len) return 0; } +void +decoding_table_put_char (Lisp_Object ccs, + int code_point, Lisp_Object character) +{ +#if 1 + 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); + } + else /* if (dim == 4) */ + { + 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); + } +#else + Lisp_Object v = XCHARSET_DECODING_TABLE (ccs); + int dim = XCHARSET_DIMENSION (ccs); + int byte_offset = XCHARSET_BYTE_OFFSET (ccs); + 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)) + { + if (EQ (nv, character)) + return; + else + nv = (XVECTOR_DATA(v)[i] = make_vector (ccs_len, Qnil)); + } + v = nv; + } + else + break; + } + XVECTOR_DATA(v)[i] = character; +#endif +} + Lisp_Object put_char_ccs_code_point (Lisp_Object character, Lisp_Object ccs, Lisp_Object value) { - if (!EQ (XCHARSET_NAME (ccs), Qmap_ucs) - || !INTP (value) - || (XCHAR (character) != XINT (value))) + if ( !( EQ (XCHARSET_NAME (ccs), Qmap_ucs) + && INTP (value) && (XINT (value) < 0xF0000) + && XCHAR (character) == XINT (value) ) + || !INTP (value) ) { Lisp_Object v = XCHARSET_DECODING_TABLE (ccs); int code_point; @@ -298,14 +387,15 @@ Lisp_Object Qascii, Qcyrillic_iso8859_5, Qlatin_iso8859_9, Qmap_jis_x0208_1978, - Qchinese_gb2312, - Qchinese_gb12345, + Qmap_gb2312, + Qmap_gb12345, Qmap_jis_x0208_1983, - Qkorean_ksc5601, + Qmap_ks_x1001, Qmap_jis_x0212, - Qchinese_cns11643_1, - Qchinese_cns11643_2, + Qmap_cns11643_1, + Qmap_cns11643_2, #ifdef UTF2000 + Qsystem_char_id, Qmap_ucs, Qucs, Qucs_bmp, Qucs_smp, @@ -318,7 +408,7 @@ Lisp_Object Qascii, Qvietnamese_viscii_upper, Qmap_jis_x0208, Qmap_jis_x0208_1990, - Qchinese_big5, + Qmap_big5, Qethiopic_ucs, #endif Qchinese_big5_1, @@ -818,15 +908,22 @@ get_unallocated_leading_byte (int dimension) #else if (dimension == 1) { - if (chlook->next_allocated_1_byte_leading_byte > MAX_LEADING_BYTE_PRIVATE_1) + if (chlook->next_allocated_1_byte_leading_byte > + MAX_LEADING_BYTE_PRIVATE_1) lb = 0; else lb = chlook->next_allocated_1_byte_leading_byte++; } else { - if (chlook->next_allocated_2_byte_leading_byte > MAX_LEADING_BYTE_PRIVATE_2) - lb = 0; + /* awfully fragile, but correct */ +#if MAX_LEADING_BYTE_PRIVATE_2 == 255 + if (chlook->next_allocated_2_byte_leading_byte == 0) +#else + if (chlook->next_allocated_2_byte_leading_byte > + MAX_LEADING_BYTE_PRIVATE_2) +#endif + lb = 0; else lb = chlook->next_allocated_2_byte_leading_byte++; } @@ -845,8 +942,76 @@ get_unallocated_leading_byte (int dimension) #define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) +static int +decode_ccs_conversion (int conv_type, int code_point) +{ + if ( conv_type == CONVERSION_IDENTICAL ) + { + return code_point; + } + if ( conv_type == CONVERSION_94x60 ) + { + int row = code_point >> 8; + int cell = code_point & 255; + + if (row < 16 + 32) + return -1; + else if (row < 16 + 32 + 30) + return (row - (16 + 32)) * 94 + cell - 33; + else if (row < 18 + 32 + 30) + return -1; + else if (row < 18 + 32 + 60) + return (row - (18 + 32)) * 94 + cell - 33; + } + else if ( conv_type == 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) + return + (plane - 33) * 94 * 60 + + (row - (16 + 32)) * 94 + + cell - 33; + else if (row < 18 + 32 + 30) + return -1; + else if (row < 18 + 32 + 60) + return + (plane - 33) * 94 * 60 + + (row - (18 + 32)) * 94 + + cell - 33; + } + else if ( conv_type == CONVERSION_BIG5_1 ) + { + unsigned int I + = (((code_point >> 8) & 0x7F) - 33) * 94 + + (( code_point & 0x7F) - 33); + unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; + unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); + + b2 += b2 < 0x3F ? 0x40 : 0x62; + return (b1 << 8) | b2; + } + else if ( conv_type == CONVERSION_BIG5_2 ) + { + unsigned int I + = (((code_point >> 8) & 0x7F) - 33) * 94 + + (( code_point & 0x7F) - 33) + + BIG5_SAME_ROW * (0xC9 - 0xA1); + unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; + unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); + + b2 += b2 < 0x3F ? 0x40 : 0x62; + return (b1 << 8) | b2; + } + return -1; +} + Emchar -decode_defined_char (Lisp_Object ccs, int code_point) +decode_defined_char (Lisp_Object ccs, int code_point, int without_inheritance) { int dim = XCHARSET_DIMENSION (ccs); Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (ccs); @@ -862,45 +1027,28 @@ decode_defined_char (Lisp_Object ccs, int code_point) } if (CHARP (decoding_table)) return XCHAR (decoding_table); -#ifdef HAVE_CHISE_CLIENT +#ifdef HAVE_CHISE if (EQ (decoding_table, Qunloaded)) { char_id = load_char_decoding_entry_maybe (ccs, code_point); } -#endif +#endif /* HAVE_CHISE */ if (char_id >= 0) return char_id; - else if ( CHARSETP (mother = XCHARSET_MOTHER (ccs)) ) + else if ( !without_inheritance + && CHARSETP (mother = XCHARSET_MOTHER (ccs)) ) { - if ( XCHARSET_CONVERSION (ccs) == CONVERSION_IDENTICAL ) + int code + = decode_ccs_conversion (XCHARSET_CONVERSION (ccs), code_point); + + if (code >= 0) { + code += XCHARSET_CODE_OFFSET(ccs); if ( EQ (mother, Vcharset_ucs) ) - return DECODE_CHAR (mother, code_point); + return DECODE_CHAR (mother, code, without_inheritance); else - return decode_defined_char (mother, code_point); - } - else if ( XCHARSET_CONVERSION (ccs) == CONVERSION_BIG5_1 ) - { - unsigned int I - = (((code_point >> 8) & 0x7F) - 33) * 94 - + (( code_point & 0x7F) - 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_defined_char (mother, (b1 << 8) | b2); - } - else if ( XCHARSET_CONVERSION (ccs) == CONVERSION_BIG5_2 ) - { - unsigned int I - = (((code_point >> 8) & 0x7F) - 33) * 94 - + (( code_point & 0x7F) - 33) - + BIG5_SAME_ROW * (0xC9 - 0xA1); - unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; - unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); - - b2 += b2 < 0x3F ? 0x40 : 0x62; - return decode_defined_char (mother, (b1 << 8) | b2); + return decode_defined_char (mother, code, + without_inheritance); } } return -1; @@ -916,68 +1064,16 @@ decode_builtin_char (Lisp_Object charset, int code_point) { if ( CHARSETP (mother) ) { - 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; - } - else if ( XCHARSET_CONVERSION (charset) == CONVERSION_BIG5_1 ) - { - unsigned int I - = (((code_point >> 8) & 0x7F) - 33) * 94 - + (( code_point & 0x7F) - 33); - unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; - unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); + EMACS_INT code + = decode_ccs_conversion (XCHARSET_CONVERSION (charset), + code_point); - b2 += b2 < 0x3F ? 0x40 : 0x62; - code = (b1 << 8) | b2; - } - else if ( XCHARSET_CONVERSION (charset) == CONVERSION_BIG5_2 ) - { - unsigned int I - = (((code_point >> 8) & 0x7F) - 33) * 94 - + (( code_point & 0x7F) - 33) - + BIG5_SAME_ROW * (0xC9 - 0xA1); - unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1; - unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40); - - b2 += b2 < 0x3F ? 0x40 : 0x62; - code = (b1 << 8) | b2; - } - return - decode_builtin_char (mother, code + XCHARSET_CODE_OFFSET(charset)); + if (code >= 0) + return + decode_builtin_char (mother, + code + XCHARSET_CODE_OFFSET(charset)); + else + return -1; } else { @@ -1159,6 +1255,8 @@ charset_code_point (Lisp_Object charset, Emchar ch, int defined_only) exit (-1); } } + else if (defined_only) + return -1; else if ( ( XCHARSET_FINAL (charset) >= '0' ) && ( XCHARSET_MIN_CODE (charset) == 0 ) /* @@ -1219,6 +1317,55 @@ charset_code_point (Lisp_Object charset, Emchar ch, int defined_only) } int +encode_char_2 (Emchar ch, Lisp_Object* charset) +{ + Lisp_Object charsets = Vdefault_coded_charset_priority_list; + int code_point; + + while (!NILP (charsets)) + { + *charset = Ffind_charset (Fcar (charsets)); + if ( !NILP (*charset) + && (XCHARSET_DIMENSION (*charset) <= 2) ) + { + code_point = charset_code_point (*charset, ch, 0); + if (code_point >= 0) + return code_point; + + if ( !NILP (Vdisplay_coded_charset_priority_use_inheritance) && + NILP (Vdisplay_coded_charset_priority_use_hierarchy_order) ) + { + code_point = encode_char_2_search_children (ch, charset); + if (code_point >= 0) + return code_point; + } + } + charsets = Fcdr (charsets); + } + + if ( !NILP (Vdisplay_coded_charset_priority_use_inheritance) && + !NILP (Vdisplay_coded_charset_priority_use_hierarchy_order) ) + { + charsets = Vdefault_coded_charset_priority_list; + while (!NILP (charsets)) + { + *charset = Ffind_charset (Fcar (charsets)); + if ( !NILP (*charset) + && (XCHARSET_DIMENSION (*charset) <= 2) ) + { + code_point = encode_char_2_search_children (ch, charset); + if (code_point >= 0) + return code_point; + } + charsets = Fcdr (charsets); + } + } + + /* otherwise --- maybe for bootstrap */ + return encode_builtin_char_1 (ch, charset); +} + +int encode_builtin_char_1 (Emchar c, Lisp_Object* charset) { if (c <= MAX_CHAR_BASIC_LATIN) @@ -1340,6 +1487,8 @@ encode_builtin_char_1 (Emchar c, Lisp_Object* charset) } Lisp_Object Vdefault_coded_charset_priority_list; +Lisp_Object Vdisplay_coded_charset_priority_use_inheritance; +Lisp_Object Vdisplay_coded_charset_priority_use_hierarchy_order; #endif @@ -2068,7 +2217,7 @@ Set mapping-table of CHARSET to TABLE. return table; } -#ifdef HAVE_CHISE_CLIENT +#ifdef HAVE_CHISE DEFUN ("save-charset-mapping-table", Fsave_charset_mapping_table, 1, 1, 0, /* Save mapping-table of CHARSET. */ @@ -2076,15 +2225,34 @@ Save mapping-table of CHARSET. { struct Lisp_Charset *cs; int byte_min, byte_max; +#ifdef HAVE_LIBCHISE + CHISE_CCS dt_ccs; +#else /* HAVE_LIBCHISE */ Lisp_Object db; Lisp_Object db_file; +#endif /* not HAVE_LIBCHISE */ charset = Fget_charset (charset); cs = XCHARSET (charset); +#ifdef HAVE_LIBCHISE + if ( open_chise_data_source_maybe () ) + return -1; + + dt_ccs + = chise_ds_get_ccs (default_chise_data_source, + XSTRING_DATA (Fsymbol_name (XCHARSET_NAME(charset)))); + if (dt_ccs == NULL) + { + printf ("Can't open decoding-table %s\n", + XSTRING_DATA (Fsymbol_name (XCHARSET_NAME(charset)))); + return -1; + } +#else /* HAVE_LIBCHISE */ 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); +#endif /* not HAVE_LIBCHISE */ byte_min = CHARSET_BYTE_OFFSET (cs); byte_max = byte_min + CHARSET_BYTE_SIZE (cs); @@ -2100,9 +2268,15 @@ Save mapping-table of CHARSET. 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); + { +#ifdef HAVE_LIBCHISE + chise_ccs_set_decoded_char (dt_ccs, cell, XCHAR (c)); +#else /* HAVE_LIBCHISE */ + Fput_database (Fprin1_to_string (make_int (cell), Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); +#endif /* not HAVE_LIBCHISE */ + } } } break; @@ -2121,11 +2295,19 @@ Save mapping-table of CHARSET. 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); + { +#ifdef HAVE_LIBCHISE + chise_ccs_set_decoded_char + (dt_ccs, + (row << 8) | cell, XCHAR (c)); +#else /* HAVE_LIBCHISE */ + Fput_database (Fprin1_to_string (make_int ((row << 8) + | cell), + Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); +#endif /* not HAVE_LIBCHISE */ + } } } } @@ -2153,12 +2335,23 @@ Save mapping-table of 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); + { +#ifdef HAVE_LIBCHISE + chise_ccs_set_decoded_char + (dt_ccs, + (plane << 16) + | (row << 8) + | cell, XCHAR (c)); +#else /* HAVE_LIBCHISE */ + Fput_database (Fprin1_to_string + (make_int ((plane << 16) + | (row << 8) + | cell), + Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); +#endif /* not HAVE_LIBCHISE */ + } } } } @@ -2193,21 +2386,37 @@ Save mapping-table of CHARSET. = 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); + { +#ifdef HAVE_LIBCHISE + chise_ccs_set_decoded_char + (dt_ccs, + ( group << 24) + | (plane << 16) + | (row << 8) + | cell, XCHAR (c)); +#else /* HAVE_LIBCHISE */ + Fput_database (Fprin1_to_string + (make_int (( group << 24) + | (plane << 16) + | (row << 8) + | cell), + Qnil), + Fprin1_to_string (c, Qnil), + db, Qt); +#endif /* not HAVE_LIBCHISE */ + } } } } } } } +#ifdef HAVE_LIBCHISE + chise_ccs_sync (dt_ccs); + return Qnil; +#else /* HAVE_LIBCHISE */ return Fclose_database (db); +#endif /* not HAVE_LIBCHISE */ } DEFUN ("reset-charset-mapping-table", Freset_charset_mapping_table, 1, 1, 0, /* @@ -2215,13 +2424,35 @@ Reset mapping-table of CCS with database file. */ (ccs)) { +#ifdef HAVE_LIBCHISE + CHISE_CCS chise_ccs; +#else Lisp_Object db_file; +#endif ccs = Fget_charset (ccs); + +#ifdef HAVE_LIBCHISE + if ( open_chise_data_source_maybe () ) + return -1; + + chise_ccs = chise_ds_get_ccs (default_chise_data_source, + XSTRING_DATA (Fsymbol_name + (XCHARSET_NAME(ccs)))); + if (chise_ccs == NULL) + return Qnil; +#else db_file = char_attribute_system_db_file (XCHARSET_NAME(ccs), Qsystem_char_id, 0); +#endif - if (!NILP (Ffile_exists_p (db_file))) + if ( +#ifdef HAVE_LIBCHISE + chise_ccs_setup_db (chise_ccs, 0) == 0 +#else + !NILP (Ffile_exists_p (db_file)) +#endif + ) { XCHARSET_DECODING_TABLE(ccs) = Qunloaded; return Qt; @@ -2232,64 +2463,24 @@ Reset mapping-table of CCS with database file. Emchar load_char_decoding_entry_maybe (Lisp_Object ccs, int code_point) { -#ifdef CHISE - Lisp_Object db_dir = Vexec_directory; - CHISE_DS ds; - CHISE_Decoding_Table *dt_ccs; - int modemask; - int accessmask = 0; - DBTYPE real_subtype; - int status; +#ifdef HAVE_LIBCHISE CHISE_Char_ID char_id; - if (NILP (db_dir)) - db_dir = build_string ("../lib-src"); - db_dir = Fexpand_file_name (build_string ("char-db"), db_dir); - - status = chise_open_data_source (&ds, CHISE_DS_Berkeley_DB, - XSTRING_DATA (db_dir)); - if (status) - { - chise_close_data_source (&ds); - return -1; - } - - modemask = 0755; /* rwxr-xr-x */ - real_subtype = DB_HASH; - accessmask = DB_RDONLY; - - status - = chise_open_decoding_table (&dt_ccs, &ds, - XSTRING_DATA (Fsymbol_name - (XCHARSET_NAME(ccs))), - real_subtype, - accessmask, modemask); - if (status) - { - printf ("Can't open decoding-table %s\n", - XSTRING_DATA (Fsymbol_name (XCHARSET_NAME(ccs)))); - chise_close_decoding_table (dt_ccs); - chise_close_data_source (&ds); - return -1; - } + if ( open_chise_data_source_maybe () ) + return -1; - char_id = chise_dt_get_char (dt_ccs, code_point); - /* - printf ("%s's 0x%X (%d) => 0x%X\n", - XSTRING_DATA (Fsymbol_name (XCHARSET_NAME(ccs))), - code_point, code_point, char_id); - */ + char_id + = chise_ds_decode_char (default_chise_data_source, + XSTRING_DATA(Fsymbol_name (XCHARSET_NAME(ccs))), + code_point); if (char_id >= 0) decoding_table_put_char (ccs, code_point, make_char (char_id)); else decoding_table_put_char (ccs, code_point, Qnil); - chise_close_decoding_table (dt_ccs); - - chise_close_data_source (&ds); - + /* chise_ccst_close (dt_ccs); */ return char_id; -#else +#else /* HAVE_LIBCHISE */ Lisp_Object db; Lisp_Object db_file = char_attribute_system_db_file (XCHARSET_NAME(ccs), Qsystem_char_id, @@ -2315,9 +2506,173 @@ load_char_decoding_entry_maybe (Lisp_Object ccs, int code_point) Fclose_database (db); } return -1; -#endif +#endif /* not HAVE_LIBCHISE */ } -#endif /* HAVE_CHISE_CLIENT */ + +#ifdef HAVE_LIBCHISE +DEFUN ("save-charset-properties", Fsave_charset_properties, 1, 1, 0, /* +Save properties of CHARSET. +*/ + (charset)) +{ + struct Lisp_Charset *cs; + CHISE_Property property; + Lisp_Object ccs; + unsigned char* feature_name; + + ccs = Fget_charset (charset); + cs = XCHARSET (ccs); + + if ( open_chise_data_source_maybe () ) + return -1; + + if ( SYMBOLP (charset) && !EQ (charset, XCHARSET_NAME (ccs)) ) + { + property = chise_ds_get_property (default_chise_data_source, + "true-name"); + feature_name = XSTRING_DATA (Fsymbol_name (charset)); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string (CHARSET_NAME (cs), + Qnil))); + chise_property_sync (property); + } + charset = XCHARSET_NAME (ccs); + feature_name = XSTRING_DATA (Fsymbol_name (charset)); + + property = chise_ds_get_property (default_chise_data_source, + "description"); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string + (CHARSET_DOC_STRING (cs), Qnil))); + chise_property_sync (property); + + property = chise_ds_get_property (default_chise_data_source, "type"); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, "CCS"); + chise_property_sync (property); + + property = chise_ds_get_property (default_chise_data_source, "chars"); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string (make_int + (CHARSET_CHARS (cs)), + Qnil))); + chise_property_sync (property); + + property = chise_ds_get_property (default_chise_data_source, "dimension"); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string (make_int + (CHARSET_DIMENSION (cs)), + Qnil))); + chise_property_sync (property); + + if ( CHARSET_FINAL (cs) != 0 ) + { + property = chise_ds_get_property (default_chise_data_source, + "final-byte"); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string (make_int + (CHARSET_FINAL (cs)), + Qnil))); + chise_property_sync (property); + } + + if ( !NILP (CHARSET_MOTHER (cs)) ) + { + Lisp_Object mother = CHARSET_MOTHER (cs); + + if ( CHARSETP (mother) ) + mother = XCHARSET_NAME (mother); + + property = chise_ds_get_property (default_chise_data_source, + "mother"); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string (mother, Qnil))); + chise_property_sync (property); + } + + if ( CHARSET_MAX_CODE (cs) != 0 ) + { + char str[16]; + + property = chise_ds_get_property (default_chise_data_source, + "mother-code-min"); + if ( CHARSET_MIN_CODE (cs) == 0 ) + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, "0"); + else + { + sprintf (str, "#x%X", CHARSET_MIN_CODE (cs)); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, str); + } + chise_property_sync (property); + + property = chise_ds_get_property (default_chise_data_source, + "mother-code-max"); + sprintf (str, "#x%X", CHARSET_MAX_CODE (cs)); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, str); + chise_property_sync (property); + + property = chise_ds_get_property (default_chise_data_source, + "mother-code-offset"); + if ( CHARSET_CODE_OFFSET (cs) == 0 ) + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, "0"); + else + { + sprintf (str, "#x%X", CHARSET_CODE_OFFSET (cs)); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, str); + } + chise_property_sync (property); + + property = chise_ds_get_property (default_chise_data_source, + "mother-code-conversion"); + if ( CHARSET_CONVERSION (cs) == CONVERSION_IDENTICAL ) + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, "identical"); + else + { + Lisp_Object sym = Qnil; + + if ( CHARSET_CONVERSION (cs) == CONVERSION_94x60 ) + sym = Q94x60; + else if ( CHARSET_CONVERSION (cs) == CONVERSION_94x94x60 ) + sym = Q94x94x60; + else if ( CHARSET_CONVERSION (cs) == CONVERSION_BIG5_1 ) + sym = Qbig5_1; + else if ( CHARSET_CONVERSION (cs) == CONVERSION_BIG5_2 ) + sym = Qbig5_2; + if ( !NILP (sym) ) + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, XSTRING_DATA (Fprin1_to_string (sym, Qnil))); + else + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + property, "unknown"); + } + chise_property_sync (property); + } + return Qnil; +} +#endif /* HAVE_LIBCHISE */ + +#endif /* HAVE_CHISE */ #endif /* UTF2000 */ @@ -2326,12 +2681,13 @@ load_char_decoding_entry_maybe (Lisp_Object ccs, int code_point) /************************************************************************/ #ifdef UTF2000 -DEFUN ("decode-char", Fdecode_char, 2, 3, 0, /* +DEFUN ("decode-char", Fdecode_char, 2, 4, 0, /* Make a character from CHARSET and code-point CODE. If DEFINED_ONLY is non-nil, builtin character is not returned. +If WITHOUT_INHERITANCE is non-nil, inherited character is not returned. If corresponding character is not found, nil is returned. */ - (charset, code, defined_only)) + (charset, code, defined_only, without_inheritance)) { int c; @@ -2341,9 +2697,9 @@ If corresponding character is not found, nil is returned. if (XCHARSET_GRAPHIC (charset) == 1) c &= 0x7F7F7F7F; if (NILP (defined_only)) - c = DECODE_CHAR (charset, c); + c = DECODE_CHAR (charset, c, !NILP (without_inheritance)); else - c = decode_defined_char (charset, c); + c = decode_defined_char (charset, c, !NILP (without_inheritance)); return c >= 0 ? make_char (c) : Qnil; } @@ -2352,13 +2708,14 @@ Make a builtin character from CHARSET and code-point CODE. */ (charset, code)) { - int c; + EMACS_INT c; + Emchar ch; charset = Fget_charset (charset); CHECK_INT (code); if (EQ (charset, Vcharset_latin_viscii)) { - Lisp_Object chr = Fdecode_char (charset, code, Qnil); + Lisp_Object chr = Fdecode_char (charset, code, Qnil, Qnil); Lisp_Object ret; if (!NILP (chr)) @@ -2386,8 +2743,9 @@ Make a builtin character from CHARSET and code-point CODE. if (XCHARSET_GRAPHIC (charset) == 1) c &= 0x7F7F7F7F; #endif - c = decode_builtin_char (charset, c); - return c >= 0 ? make_char (c) : Fdecode_char (charset, code, Qnil); + ch = decode_builtin_char (charset, c); + return + ch >= 0 ? make_char (ch) : Fdecode_char (charset, code, Qnil, Qnil); } #endif @@ -2649,18 +3007,22 @@ syms_of_mule_charset (void) DEFSUBR (Fcharset_id); DEFSUBR (Fset_charset_ccl_program); DEFSUBR (Fset_charset_registry); + #ifdef UTF2000 DEFSUBR (Fcharset_mapping_table); DEFSUBR (Fset_charset_mapping_table); -#ifdef HAVE_CHISE_CLIENT +#ifdef HAVE_CHISE DEFSUBR (Fsave_charset_mapping_table); DEFSUBR (Freset_charset_mapping_table); -#endif - +#ifdef HAVE_LIBCHISE + DEFSUBR (Fsave_charset_properties); +#endif /* HAVE_LIBCHISE */ +#endif /* HAVE_CHISE */ DEFSUBR (Fdecode_char); DEFSUBR (Fdecode_builtin_char); DEFSUBR (Fencode_char); #endif + DEFSUBR (Fmake_char); DEFSUBR (Fchar_charset); DEFSUBR (Fchar_octet); @@ -2711,15 +3073,16 @@ syms_of_mule_charset (void) defsymbol (&Qlatin_jisx0201, "latin-jisx0201"); defsymbol (&Qcyrillic_iso8859_5, "cyrillic-iso8859-5"); defsymbol (&Qlatin_iso8859_9, "latin-iso8859-9"); - defsymbol (&Qmap_jis_x0208_1978, "=jis-x0208-1978"); - defsymbol (&Qchinese_gb2312, "chinese-gb2312"); - defsymbol (&Qchinese_gb12345, "chinese-gb12345"); - defsymbol (&Qmap_jis_x0208_1983, "=jis-x0208-1983"); - defsymbol (&Qkorean_ksc5601, "korean-ksc5601"); + defsymbol (&Qmap_jis_x0208_1978, "=jis-x0208@1978"); + defsymbol (&Qmap_gb2312, "=gb2312"); + defsymbol (&Qmap_gb12345, "=gb12345"); + defsymbol (&Qmap_jis_x0208_1983, "=jis-x0208@1983"); + defsymbol (&Qmap_ks_x1001, "=ks-x1001"); defsymbol (&Qmap_jis_x0212, "=jis-x0212"); - defsymbol (&Qchinese_cns11643_1, "chinese-cns11643-1"); - defsymbol (&Qchinese_cns11643_2, "chinese-cns11643-2"); + defsymbol (&Qmap_cns11643_1, "=cns11643-1"); + defsymbol (&Qmap_cns11643_2, "=cns11643-2"); #ifdef UTF2000 + defsymbol (&Qsystem_char_id, "system-char-id"); defsymbol (&Qmap_ucs, "=ucs"); defsymbol (&Qucs, "ucs"); defsymbol (&Qucs_bmp, "ucs-bmp"); @@ -2732,8 +3095,8 @@ syms_of_mule_charset (void) defsymbol (&Qvietnamese_viscii_lower, "vietnamese-viscii-lower"); defsymbol (&Qvietnamese_viscii_upper, "vietnamese-viscii-upper"); defsymbol (&Qmap_jis_x0208, "=jis-x0208"); - defsymbol (&Qmap_jis_x0208_1990, "=jis-x0208-1990"); - defsymbol (&Qchinese_big5, "chinese-big5"); + defsymbol (&Qmap_jis_x0208_1990, "=jis-x0208@1990"); + defsymbol (&Qmap_big5, "=big5"); defsymbol (&Qethiopic_ucs, "ethiopic-ucs"); #endif defsymbol (&Qchinese_big5_1, "chinese-big5-1"); @@ -2791,6 +3154,16 @@ Leading-code of private TYPE9N charset of column-width 1. &Vdefault_coded_charset_priority_list /* Default order of preferred coded-character-sets. */ ); + Vdisplay_coded_charset_priority_use_inheritance = Qt; + DEFVAR_LISP ("display-coded-charset-priority-use-inheritance", + &Vdisplay_coded_charset_priority_use_inheritance /* +If non-nil, use character inheritance. +*/ ); + Vdisplay_coded_charset_priority_use_hierarchy_order = Qt; + DEFVAR_LISP ("display-coded-charset-priority-use-hierarchy-order", + &Vdisplay_coded_charset_priority_use_hierarchy_order /* +If non-nil, prefer nearest character in hierarchy order. +*/ ); #endif } @@ -2805,6 +3178,15 @@ complex_vars_of_mule_charset (void) ease of access. */ #ifdef UTF2000 + staticpro (&Vcharset_system_char_id); + Vcharset_system_char_id = + make_charset (LEADING_BYTE_SYSTEM_CHAR_ID, Qsystem_char_id, 256, 4, + 1, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("SCID"), + build_string ("CHAR-ID"), + build_string ("System char-id"), + build_string (""), + Qnil, 0, 0x7FFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs); Vcharset_ucs = make_charset (LEADING_BYTE_UCS, Qmap_ucs, 256, 4, @@ -2813,7 +3195,7 @@ complex_vars_of_mule_charset (void) build_string ("UCS"), build_string ("ISO/IEC 10646"), build_string (""), - Qnil, 0, 0x7FFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL); + Qnil, 0, 0xEFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs_bmp); Vcharset_ucs_bmp = make_charset (LEADING_BYTE_UCS_BMP, Qucs_bmp, 256, 2, @@ -2822,7 +3204,7 @@ complex_vars_of_mule_charset (void) build_string ("UCS-BMP"), build_string ("ISO/IEC 10646 Group 0 Plane 0 (BMP)"), build_string - ("\\(ISO10646.*-[01]\\|UCS00-0\\|UNICODE[23]?-0\\)"), + ("\\(ISO10646\\(\\.[0-9]+\\)?-[01]\\|UCS00-0\\|UNICODE[23]?-0\\)"), Qnil, 0, 0xFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_ucs_smp); Vcharset_ucs_smp = @@ -3015,7 +3397,7 @@ complex_vars_of_mule_charset (void) CONVERSION_IDENTICAL); staticpro (&Vcharset_chinese_gb2312); Vcharset_chinese_gb2312 = - make_charset (LEADING_BYTE_CHINESE_GB2312, Qchinese_gb2312, 94, 2, + make_charset (LEADING_BYTE_CHINESE_GB2312, Qmap_gb2312, 94, 2, 2, 0, 'A', CHARSET_LEFT_TO_RIGHT, build_string ("GB2312"), build_string ("GB2312)"), @@ -3024,7 +3406,7 @@ complex_vars_of_mule_charset (void) Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_chinese_gb12345); Vcharset_chinese_gb12345 = - make_charset (LEADING_BYTE_CHINESE_GB12345, Qchinese_gb12345, 94, 2, + make_charset (LEADING_BYTE_CHINESE_GB12345, Qmap_gb12345, 94, 2, 2, 0, 0, CHARSET_LEFT_TO_RIGHT, build_string ("G1"), build_string ("GB 12345)"), @@ -3065,7 +3447,7 @@ complex_vars_of_mule_charset (void) #endif staticpro (&Vcharset_korean_ksc5601); Vcharset_korean_ksc5601 = - make_charset (LEADING_BYTE_KOREAN_KSC5601, Qkorean_ksc5601, 94, 2, + make_charset (LEADING_BYTE_KOREAN_KSC5601, Qmap_ks_x1001, 94, 2, 2, 0, 'C', CHARSET_LEFT_TO_RIGHT, build_string ("KSC5601"), build_string ("KSC5601 (Korean"), @@ -3085,7 +3467,7 @@ complex_vars_of_mule_charset (void) #define CHINESE_CNS_PLANE_RE(n) "cns11643[.-]\\(.*[.-]\\)?" n "$" staticpro (&Vcharset_chinese_cns11643_1); Vcharset_chinese_cns11643_1 = - make_charset (LEADING_BYTE_CHINESE_CNS11643_1, Qchinese_cns11643_1, 94, 2, + make_charset (LEADING_BYTE_CHINESE_CNS11643_1, Qmap_cns11643_1, 94, 2, 2, 0, 'G', CHARSET_LEFT_TO_RIGHT, build_string ("CNS11643-1"), build_string ("CNS11643-1 (Chinese traditional)"), @@ -3095,7 +3477,7 @@ complex_vars_of_mule_charset (void) Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_chinese_cns11643_2); Vcharset_chinese_cns11643_2 = - make_charset (LEADING_BYTE_CHINESE_CNS11643_2, Qchinese_cns11643_2, 94, 2, + make_charset (LEADING_BYTE_CHINESE_CNS11643_2, Qmap_cns11643_2, 94, 2, 2, 0, 'H', CHARSET_LEFT_TO_RIGHT, build_string ("CNS11643-2"), build_string ("CNS11643-2 (Chinese traditional)"), @@ -3142,7 +3524,7 @@ complex_vars_of_mule_charset (void) Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL); staticpro (&Vcharset_chinese_big5); Vcharset_chinese_big5 = - make_charset (LEADING_BYTE_CHINESE_BIG5, Qchinese_big5, 256, 2, + make_charset (LEADING_BYTE_CHINESE_BIG5, Qmap_big5, 256, 2, 2, 2, 0, CHARSET_LEFT_TO_RIGHT, build_string ("Big5"), build_string ("Big5"),