X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fchartab.c;h=ccfed1106d302191fdd22b2bd620a9004446c7ba;hb=bbf3133177a15f214e3d0060e168ed86dbeed8a2;hp=532c71fac7fdabf01c18f90af4022326244612ab;hpb=6904fbf5fc782f4a2b47a03e4318be17a5faa2b8;p=chise%2Fxemacs-chise.git.1 diff --git a/src/chartab.c b/src/chartab.c index 532c71f..ccfed11 100644 --- a/src/chartab.c +++ b/src/chartab.c @@ -4,7 +4,8 @@ Copyright (C) 1995, 1996 Ben Wing. Copyright (C) 1995, 1997, 1999 Electrotechnical Laboratory, JAPAN. Licensed to the Free Software Foundation. - Copyright (C) 1999,2000,2001,2002,2003,2004,2005 MORIOKA Tomohiko + Copyright (C) 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2008, + 2010 MORIOKA Tomohiko This file is part of XEmacs. @@ -354,7 +355,7 @@ save_uint8_byte_table (Lisp_Uint8_Byte_Table *ct, Lisp_Char_Table* root, { if (ct->property[i] == BT_UINT8_unloaded) { - c1 = c + unit; + c += unit; } else if (ct->property[i] != BT_UINT8_unbound) { @@ -678,7 +679,7 @@ save_uint16_byte_table (Lisp_Uint16_Byte_Table *ct, Lisp_Char_Table* root, { if (ct->property[i] == BT_UINT16_unloaded) { - c1 = c + unit; + c += unit; } else if (ct->property[i] != BT_UINT16_unbound) { @@ -1130,7 +1131,14 @@ make_char_id_table (Lisp_Object initval) Lisp_Object Qcomposition; -Lisp_Object Qmap_decomposition; +Lisp_Object Qrep_decomposition; +Lisp_Object Qto_decomposition_at_superscript; +Lisp_Object Qto_decomposition_at_circled; +Lisp_Object Q_canonical; +Lisp_Object Q_halfwidth_of; +Lisp_Object Q_superscript_of; +Lisp_Object Q_subscript_of; +Lisp_Object Q_circled_of; Lisp_Object Q_decomposition; Lisp_Object Q_identical; Lisp_Object Q_identical_from; @@ -1714,15 +1722,17 @@ once per character). When Mule support exists, the types of ranges that can be assigned values are --- all characters +-- all characters (represented by t) -- an entire charset --- a single row in a two-octet charset +-- a single row in a two-octet charset (represented by a vector of two + elements: a two-octet charset and a row number; the row must be an + integer, not a character) -- a single character When Mule support is not present, the types of ranges that can be assigned values are --- all characters +-- all characters (represented by t) -- a single character To create a char table, use `make-char-table'. @@ -2224,8 +2234,11 @@ Find value for CHARACTER in CHAR-TABLE. } DEFUN ("get-range-char-table", Fget_range_char_table, 2, 3, 0, /* -Find value for a range in CHAR-TABLE. +Find value for RANGE in CHAR-TABLE. If there is more than one value, return MULTI (defaults to nil). + +Valid values for RANGE are single characters, charsets, a row in a +two-octet charset, and all characters. See `put-char-table'. */ (range, char_table, multi)) { @@ -2620,8 +2633,9 @@ one of the following: -- t (all characters are affected) -- A charset (only allowed when Mule support is present) --- A vector of two elements: a two-octet charset and a row number - (only allowed when Mule support is present) +-- A vector of two elements: a two-octet charset and a row number; the row + must be an integer, not a character (only allowed when Mule support is + present) -- A single character VALUE must be a value appropriate for the type of CHAR-TABLE. @@ -3096,8 +3110,8 @@ slow_map_char_table_fun (struct chartab_range *range, } DEFUN ("map-char-table", Fmap_char_table, 2, 3, 0, /* -Map FUNCTION over entries in CHAR-TABLE, calling it with two args, -each key and value in the table. +Map FUNCTION over CHAR-TABLE until it returns non-nil; return that value. +FUNCTION is called with two arguments, each key and entry in the table. RANGE specifies a subrange to map over and is in the same format as the RANGE argument to `put-range-table'. If omitted or t, it defaults to @@ -3385,20 +3399,21 @@ Return DEFAULT-VALUE if the value is not exist. return default_value; } -void put_char_composition (Lisp_Object character, Lisp_Object value); -void +Lisp_Object +put_char_composition (Lisp_Object character, Lisp_Object value); +Lisp_Object put_char_composition (Lisp_Object character, Lisp_Object value) { if (!CONSP (value)) - signal_simple_error ("Invalid value for ->decomposition", + signal_simple_error ("Invalid value for =decomposition", value); - if (CONSP (Fcdr (value))) + if (CONSP (XCDR (value))) { - if (NILP (Fcdr (Fcdr (value)))) + if (NILP (Fcdr (XCDR (value)))) { - Lisp_Object base = Fcar (value); - Lisp_Object modifier = Fcar (Fcdr (value)); + Lisp_Object base = XCAR (value); + Lisp_Object modifier = XCAR (XCDR (value)); if (INTP (base)) { @@ -3408,7 +3423,7 @@ put_char_composition (Lisp_Object character, Lisp_Object value) if (INTP (modifier)) { modifier = make_char (XINT (modifier)); - Fsetcar (Fcdr (value), modifier); + Fsetcar (XCDR (value), modifier); } if (CHARP (base)) { @@ -3424,10 +3439,42 @@ put_char_composition (Lisp_Object character, Lisp_Object value) else Fsetcdr (ret, character); } + else if (EQ (base, Qnarrow)) + return Q_halfwidth_of; + else if (EQ (base, Qsuper)) + return Q_superscript_of; + else if (EQ (base, Qsub)) + return Q_subscript_of; + else if (EQ (base, Qcircle)) + return Q_circled_of; + else if ( EQ (base, Qisolated)|| + EQ (base, Qinitial) || + EQ (base, Qmedial) || + EQ (base, Qfinal) ) + return + Fintern (concat2 (build_string ("<-formed@"), + Fsymbol_name (base)), + Qnil); + else if (SYMBOLP (base)) + return + Fintern (concat2 (build_string ("<-"), + Fsymbol_name (base)), + Qnil); } + else if (EQ (XCAR (value), Qsuper)) + return Qto_decomposition_at_superscript; + else if (EQ (XCAR (value), Qcircle)) + return Qto_decomposition_at_circled; + else + return + Fintern (concat2 (build_string ("=>decomposition@"), + Fsymbol_name (XCAR (value))), + Qnil); } else { + return Q_canonical; +#if 0 Lisp_Object v = Fcar (value); if (INTP (v)) @@ -3448,7 +3495,9 @@ put_char_composition (Lisp_Object character, Lisp_Object value) Fcons (character, ret)); } } +#endif } + return Qrep_decomposition; } static Lisp_Object @@ -3485,12 +3534,14 @@ Store CHARACTER's ATTRIBUTE with VALUE. value = put_char_ccs_code_point (character, ccs, value); attribute = XCHARSET_NAME (ccs); } - else if (EQ (attribute, Qmap_decomposition)) - put_char_composition (character, value); - else if (EQ (attribute, Q_decomposition)) + else if ( EQ (attribute, Qrep_decomposition) || + EQ (attribute, Q_decomposition) ) { - attribute = Qmap_decomposition; - put_char_composition (character, value); + value = Fcopy_sequence (Fchar_refs_simplify_char_specs (value)); + attribute = put_char_composition (character, value); + if ( !EQ (attribute, Qrep_decomposition) && + SYMBOLP (XCAR (value)) ) + value = XCDR (value); } else if (EQ (attribute, Qto_ucs)) { @@ -3511,22 +3562,26 @@ Store CHARACTER's ATTRIBUTE with VALUE. Fput_char_attribute (make_char (c), Q_ucs_unified, Fcons (character, ret)); } - else if ( EQ (attribute, Q_subsumptive) || - EQ (attribute, Q_subsumptive_from) || - EQ (attribute, Q_denotational) || - EQ (attribute, Q_denotational_from) || - EQ (attribute, Q_identical) || - EQ (attribute, Q_identical_from) || - EQ (attribute, Q_component) || - EQ (attribute, Q_component_of) || - !NILP (Fstring_match - (build_string ("^\\(<-\\|->\\)\\(" - "fullwidth\\|halfwidth" - "\\|simplified\\|vulgar\\|wrong" - "\\|same\\|original\\|ancient" - "\\|Oracle-Bones\\)[^*]*$"), + if ( EQ (attribute, Q_subsumptive) || + EQ (attribute, Q_subsumptive_from) || + EQ (attribute, Q_denotational) || + EQ (attribute, Q_denotational_from) || + EQ (attribute, Q_identical) || + EQ (attribute, Q_identical_from) || + EQ (attribute, Q_canonical) || + EQ (attribute, Q_halfwidth_of) || + EQ (attribute, Q_superscript_of) || + EQ (attribute, Q_subscript_of) || + EQ (attribute, Q_circled_of) || + EQ (attribute, Q_component) || + EQ (attribute, Q_component_of) || + ( !EQ (attribute, Q_ucs_unified) + && !NILP (Fstring_match + (build_string ("^\\(<-\\|->\\)[^*]*$"), Fsymbol_name (attribute), - Qnil, Qnil)) ) + Qnil, Qnil)) + ) + ) { Lisp_Object rest = value; Lisp_Object ret; @@ -3583,7 +3638,17 @@ Store CHARACTER's ATTRIBUTE with VALUE. if (CONSP (ret)) ret = Fdefine_char (ret); - + else if (INTP (ret)) + { + int code_point = XINT (ret); + Emchar cid = DECODE_CHAR (Vcharset_ucs, code_point, 0); + + if (cid >= 0) + ret = make_char (cid); + else + ret = make_char (code_point); + } + if ( !NILP (ret) && !EQ (ret, character) ) { Lisp_Object ffv; @@ -3602,7 +3667,11 @@ Store CHARACTER's ATTRIBUTE with VALUE. UNGCPRO; } #if 1 - else if (EQ (attribute, Qideographic_structure)) + else if ( EQ (attribute, Qideographic_structure) || + !NILP (Fstring_match + (build_string ("^=>decomposition\\(\\|@[^*]+\\)$"), + Fsymbol_name (attribute), + Qnil, Qnil)) ) value = Fcopy_sequence (Fchar_refs_simplify_char_specs (value)); #endif return put_char_attribute (character, attribute, value); @@ -3852,16 +3921,21 @@ Save values of ATTRIBUTE into database file. if ( !NILP (Ffind_charset (attribute)) ) filter = NULL; - else if ( EQ (attribute, Qideographic_structure) - || EQ (attribute, Q_identical) - || EQ (attribute, Q_identical_from) - || !NILP (Fstring_match - (build_string ("^\\(<-\\|->\\)\\(simplified" - "\\|same\\|vulgar\\|wrong" - "\\|original\\|ancient" - "\\|Oracle-Bones\\)[^*]*$"), - Fsymbol_name (attribute), - Qnil, Qnil)) ) + else if ( EQ (attribute, Qideographic_structure) || + EQ (attribute, Q_identical) || + EQ (attribute, Q_identical_from) || + EQ (attribute, Q_canonical) || + EQ (attribute, Q_halfwidth_of) || + EQ (attribute, Q_superscript_of) || + EQ (attribute, Q_subscript_of) || + EQ (attribute, Q_circled_of) || + !NILP (Fstring_match + (build_string ("^\\(<-\\|->\\)\\(simplified" + "\\|same\\|vulgar\\|wrong" + "\\|original\\|ancient" + "\\|Oracle-Bones\\)[^*]*$"), + Fsymbol_name (attribute), + Qnil, Qnil)) ) filter = &Fchar_refs_simplify_char_specs; else filter = NULL; @@ -4103,12 +4177,18 @@ Load values of ATTRIBUTE into database file. #endif /* HAVE_CHISE */ DEFUN ("map-char-attribute", Fmap_char_attribute, 2, 3, 0, /* -Map FUNCTION over entries in ATTRIBUTE, calling it with two args, -each key and value in the table. +Map FUNCTION over ATTRIBUTE until it returns non-nil; return that value. +FUNCTION is called with two arguments, each key and entry in the table. -RANGE specifies a subrange to map over and is in the same format as -the RANGE argument to `put-range-table'. If omitted or t, it defaults to +RANGE specifies a subrange to map over. If omitted or t, it defaults to the entire table. + +Both RANGE and the keys passed to FUNCTION are in the same format as the +RANGE argument to `put-char-table'. N.B. This function does NOT map over +all characters in RANGE, but over the subranges that have been assigned to. +Thus this function is most suitable for searching a char-table, or for +populating one char-table based on the contents of another. The current +implementation does not coalesce ranges all of whose values are the same. */ (function, attribute, range)) { @@ -4153,13 +4233,27 @@ the entire table. return slarg.retval; } +static Lisp_Object +allocate_character () +{ + int cid = XINT (Vnext_defined_char_id); + + if (cid <= 0xE00000) + { + Vnext_defined_char_id = make_int (cid + 1); + return make_char (cid); + } + else + return Qnil; +} + DEFUN ("define-char", Fdefine_char, 1, 1, 0, /* Store character's ATTRIBUTES. */ (attributes)) { Lisp_Object rest; - Lisp_Object code = Fcdr (Fassq (Qmap_ucs, attributes)); + Lisp_Object code = Fcdr (Fassq (Qrep_ucs, attributes)); Lisp_Object character; if (NILP (code)) @@ -4201,7 +4295,7 @@ Store character's ATTRIBUTES. } rest = Fcdr (rest); } -#if 1 +#if 0 { int cid = XINT (Vnext_defined_char_id); @@ -4213,16 +4307,9 @@ Store character's ATTRIBUTES. } } #else - if ( (!NILP (code = Fcdr (Fassq (Qto_ucs, attributes)))) ) - { - if (!INTP (code)) - signal_simple_error ("Invalid argument", attributes); - else - character = make_char (XINT (code) + 0x100000); - goto setup_attributes; - } + if ( NILP (character = allocate_character ()) ) #endif - return Qnil; + return Qnil; } else if (!INTP (code)) signal_simple_error ("Invalid argument", attributes); @@ -4283,6 +4370,62 @@ Retrieve the character of the given ATTRIBUTES. /************************************************************************/ +/* Character Feature Property */ +/************************************************************************/ + +#ifdef HAVE_LIBCHISE +DEFUN ("char-feature-property", Fchar_feature_property, 2, 3, 0, /* +Return the value of FEATURE's PROPERTY. +Return DEFAULT-VALUE if the value is not exist. +*/ + (feature, property, default_value)) +{ + unsigned char* feature_name; + unsigned char* property_name; + CHISE_Value value; + int status; + + feature_name = XSTRING_DATA (Fsymbol_name (feature)); + property_name = XSTRING_DATA (Fsymbol_name (property)); + status + = chise_feature_load_property_value (chise_ds_get_feature + (default_chise_data_source, + feature_name), + chise_ds_get_property + (default_chise_data_source, + property_name), + &value); + if (!status) + return read_from_c_string (chise_value_data (&value), + chise_value_size (&value) ); + else + return default_value; +} + +DEFUN ("put-char-feature-property", Fput_char_feature_property, 3, 3, 0, /* +Store FEATURE's PROPERTY with VALUE. +*/ + (feature, property, value)) +{ + unsigned char* feature_name; + unsigned char* property_name; + CHISE_Property prop; + + feature_name = XSTRING_DATA (Fsymbol_name (feature)); + property_name = XSTRING_DATA (Fsymbol_name (property)); + prop = chise_ds_get_property (default_chise_data_source, + property_name); + chise_feature_set_property_value + (chise_ds_get_feature (default_chise_data_source, feature_name), + prop, XSTRING_DATA (Fprin1_to_string + (value, Qnil))); + chise_property_sync (prop); + return Qnil; +} +#endif + + +/************************************************************************/ /* Char table read syntax */ /************************************************************************/ @@ -4620,6 +4763,12 @@ word_boundary_p (Emchar c1, Emchar c2) void syms_of_chartab (void) { +#if defined(UTF2000) || defined(HAVE_CONCORD) + defsymbol (&Q_subsumptive, "->subsumptive"); + defsymbol (&Q_subsumptive_from, "<-subsumptive"); + defsymbol (&Q_denotational, "->denotational"); + defsymbol (&Q_denotational_from, "<-denotational"); +#endif #ifdef UTF2000 INIT_LRECORD_IMPLEMENTATION (uint8_byte_table); INIT_LRECORD_IMPLEMENTATION (uint16_byte_table); @@ -4627,16 +4776,20 @@ syms_of_chartab (void) defsymbol (&Qto_ucs, "=>ucs"); defsymbol (&Q_ucs_unified, "->ucs-unified"); - defsymbol (&Q_subsumptive, "->subsumptive"); - defsymbol (&Q_subsumptive_from, "<-subsumptive"); - defsymbol (&Q_denotational, "->denotational"); - defsymbol (&Q_denotational_from, "<-denotational"); defsymbol (&Q_identical, "->identical"); defsymbol (&Q_identical_from, "<-identical"); defsymbol (&Q_component, "->ideographic-component-forms"); defsymbol (&Q_component_of, "<-ideographic-component-forms"); defsymbol (&Qcomposition, "composition"); - defsymbol (&Qmap_decomposition, "=decomposition"); + defsymbol (&Qrep_decomposition, "=decomposition"); + defsymbol (&Qto_decomposition_at_superscript, + "=>decomposition@superscript"); + defsymbol (&Qto_decomposition_at_circled, "=>decomposition@circled"); + defsymbol (&Q_canonical, "->canonical"); + defsymbol (&Q_halfwidth_of, "<-halfwidth"); + defsymbol (&Q_superscript_of, "<-superscript"); + defsymbol (&Q_subscript_of, "<-subscript"); + defsymbol (&Q_circled_of, "<-circled"); defsymbol (&Q_decomposition, "->decomposition"); defsymbol (&Qcompat, "compat"); defsymbol (&Qisolated, "isolated"); @@ -4683,6 +4836,10 @@ syms_of_chartab (void) DEFSUBR (Fchar_variants); DEFSUBR (Fget_composite_char); +#ifdef HAVE_LIBCHISE + DEFSUBR (Fchar_feature_property); + DEFSUBR (Fput_char_feature_property); +#endif /* HAVE_LIBCHISE */ #endif INIT_LRECORD_IMPLEMENTATION (char_table);