X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Fmule-charset.c;h=05d38ce8d59b017d53eaa306a42c683d56689a0a;hb=7b4b1b26bb371112bf3e18732864bc08584127b6;hp=7aabbc02638cd4c972d113546f154cb2e5ebf867;hpb=3890a2e3064a7f562107c58e59d928284ec04741;p=chise%2Fxemacs-chise.git- diff --git a/src/mule-charset.c b/src/mule-charset.c index 7aabbc0..05d38ce 100644 --- a/src/mule-charset.c +++ b/src/mule-charset.c @@ -1,6 +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 MORIOKA Tomohiko This file is part of XEmacs. @@ -24,93 +25,1556 @@ Boston, MA 02111-1307, USA. */ /* Rewritten by Ben Wing . */ #include +#ifdef UTF2000 +#include +#endif #include "lisp.h" -#include "buffer.h" -#include "chartab.h" -#include "elhash.h" -#include "lstream.h" -#include "device.h" -#include "faces.h" -#include "mule-ccl.h" +#include "buffer.h" +#include "chartab.h" +#include "elhash.h" +#include "lstream.h" +#include "device.h" +#include "faces.h" +#include "mule-ccl.h" + +/* The various pre-defined charsets. */ + +Lisp_Object Vcharset_ascii; +Lisp_Object Vcharset_control_1; +Lisp_Object Vcharset_latin_iso8859_1; +Lisp_Object Vcharset_latin_iso8859_2; +Lisp_Object Vcharset_latin_iso8859_3; +Lisp_Object Vcharset_latin_iso8859_4; +Lisp_Object Vcharset_thai_tis620; +Lisp_Object Vcharset_greek_iso8859_7; +Lisp_Object Vcharset_arabic_iso8859_6; +Lisp_Object Vcharset_hebrew_iso8859_8; +Lisp_Object Vcharset_katakana_jisx0201; +Lisp_Object Vcharset_latin_jisx0201; +Lisp_Object Vcharset_cyrillic_iso8859_5; +Lisp_Object Vcharset_latin_iso8859_9; +Lisp_Object Vcharset_japanese_jisx0208_1978; +Lisp_Object Vcharset_chinese_gb2312; +Lisp_Object Vcharset_japanese_jisx0208; +Lisp_Object Vcharset_japanese_jisx0208_1990; +Lisp_Object Vcharset_korean_ksc5601; +Lisp_Object Vcharset_japanese_jisx0212; +Lisp_Object Vcharset_chinese_cns11643_1; +Lisp_Object Vcharset_chinese_cns11643_2; +#ifdef UTF2000 +Lisp_Object Vcharset_ucs; +Lisp_Object Vcharset_ucs_bmp; +Lisp_Object Vcharset_ucs_cns; +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_chinese_big5; +Lisp_Object Vcharset_ideograph_daikanwa; +Lisp_Object Vcharset_mojikyo; +Lisp_Object Vcharset_mojikyo_2022_1; +Lisp_Object Vcharset_mojikyo_pj_1; +Lisp_Object Vcharset_mojikyo_pj_2; +Lisp_Object Vcharset_mojikyo_pj_3; +Lisp_Object Vcharset_mojikyo_pj_4; +Lisp_Object Vcharset_mojikyo_pj_5; +Lisp_Object Vcharset_mojikyo_pj_6; +Lisp_Object Vcharset_mojikyo_pj_7; +Lisp_Object Vcharset_mojikyo_pj_8; +Lisp_Object Vcharset_mojikyo_pj_9; +Lisp_Object Vcharset_mojikyo_pj_10; +Lisp_Object Vcharset_mojikyo_pj_11; +Lisp_Object Vcharset_mojikyo_pj_12; +Lisp_Object Vcharset_mojikyo_pj_13; +Lisp_Object Vcharset_mojikyo_pj_14; +Lisp_Object Vcharset_mojikyo_pj_15; +Lisp_Object Vcharset_mojikyo_pj_16; +Lisp_Object Vcharset_mojikyo_pj_17; +Lisp_Object Vcharset_mojikyo_pj_18; +Lisp_Object Vcharset_mojikyo_pj_19; +Lisp_Object Vcharset_mojikyo_pj_20; +Lisp_Object Vcharset_mojikyo_pj_21; +Lisp_Object Vcharset_ethiopic_ucs; +#endif +Lisp_Object Vcharset_chinese_big5_1; +Lisp_Object Vcharset_chinese_big5_2; + +#ifdef ENABLE_COMPOSITE_CHARS +Lisp_Object Vcharset_composite; + +/* Hash tables for composite chars. One maps string representing + composed chars to their equivalent chars; one goes the + other way. */ +Lisp_Object Vcomposite_char_char2string_hash_table; +Lisp_Object Vcomposite_char_string2char_hash_table; + +static int composite_char_row_next; +static int composite_char_col_next; + +#endif /* ENABLE_COMPOSITE_CHARS */ + +struct charset_lookup *chlook; + +static const struct lrecord_description charset_lookup_description_1[] = { + { XD_LISP_OBJECT_ARRAY, offsetof (struct charset_lookup, charset_by_leading_byte), +#ifdef UTF2000 + 128+4*128 +#else + 128+4*128*2 +#endif + }, { XD_END } +}; + +static const struct struct_description charset_lookup_description = { + sizeof (struct charset_lookup), + charset_lookup_description_1 +}; + +#ifndef UTF2000 +/* Table of number of bytes in the string representation of a character + indexed by the first byte of that representation. + + rep_bytes_by_first_byte(c) is more efficient than the equivalent + canonical computation: + + XCHARSET_REP_BYTES (CHARSET_BY_LEADING_BYTE (c)) */ + +const Bytecount rep_bytes_by_first_byte[0xA0] = +{ /* 0x00 - 0x7f are for straight ASCII */ + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + /* 0x80 - 0x8f are for Dimension-1 official charsets */ +#ifdef CHAR_IS_UCS4 + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, +#else + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, +#endif + /* 0x90 - 0x9d are for Dimension-2 official charsets */ + /* 0x9e is for Dimension-1 private charsets */ + /* 0x9f is for Dimension-2 private charsets */ + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4 +}; +#endif + +#ifdef UTF2000 + +#define BT_UINT8_MIN 0 +#define BT_UINT8_MAX (UCHAR_MAX - 3) +#define BT_UINT8_t (UCHAR_MAX - 2) +#define BT_UINT8_nil (UCHAR_MAX - 1) +#define BT_UINT8_unbound UCHAR_MAX + +INLINE_HEADER int INT_UINT8_P (Lisp_Object obj); +INLINE_HEADER int UINT8_VALUE_P (Lisp_Object obj); +INLINE_HEADER unsigned char UINT8_ENCODE (Lisp_Object obj); +INLINE_HEADER Lisp_Object UINT8_DECODE (unsigned char n); +INLINE_HEADER unsigned short UINT8_TO_UINT16 (unsigned char n); + +INLINE_HEADER int +INT_UINT8_P (Lisp_Object obj) +{ + if (INTP (obj)) + { + int num = XINT (obj); + + return (BT_UINT8_MIN <= num) && (num <= BT_UINT8_MAX); + } + else + return 0; +} + +INLINE_HEADER int +UINT8_VALUE_P (Lisp_Object obj) +{ + return EQ (obj, Qunbound) + || EQ (obj, Qnil) || EQ (obj, Qt) || INT_UINT8_P (obj); +} + +INLINE_HEADER unsigned char +UINT8_ENCODE (Lisp_Object obj) +{ + if (EQ (obj, Qunbound)) + return BT_UINT8_unbound; + else if (EQ (obj, Qnil)) + return BT_UINT8_nil; + else if (EQ (obj, Qt)) + return BT_UINT8_t; + else + return XINT (obj); +} + +INLINE_HEADER Lisp_Object +UINT8_DECODE (unsigned char n) +{ + if (n == BT_UINT8_unbound) + return Qunbound; + else if (n == BT_UINT8_nil) + return Qnil; + else if (n == BT_UINT8_t) + return Qt; + else + return make_int (n); +} + +static Lisp_Object +mark_uint8_byte_table (Lisp_Object obj) +{ + return Qnil; +} + +static void +print_uint8_byte_table (Lisp_Object obj, + Lisp_Object printcharfun, int escapeflag) +{ + Lisp_Uint8_Byte_Table *bte = XUINT8_BYTE_TABLE (obj); + int i; + struct gcpro gcpro1, gcpro2; + GCPRO2 (obj, printcharfun); + + write_c_string ("\n#property[i]; + if ( (i & 15) == 0 ) + write_c_string ("\n ", printcharfun); + write_c_string (" ", printcharfun); + if (n == BT_UINT8_unbound) + write_c_string ("void", printcharfun); + else if (n == BT_UINT8_nil) + write_c_string ("nil", printcharfun); + else if (n == BT_UINT8_t) + write_c_string ("t", printcharfun); + else + { + char buf[4]; + + sprintf (buf, "%hd", n); + write_c_string (buf, printcharfun); + } + } + UNGCPRO; + write_c_string (">", printcharfun); +} + +static int +uint8_byte_table_equal (Lisp_Object obj1, Lisp_Object obj2, int depth) +{ + Lisp_Uint8_Byte_Table *te1 = XUINT8_BYTE_TABLE (obj1); + Lisp_Uint8_Byte_Table *te2 = XUINT8_BYTE_TABLE (obj2); + int i; + + for (i = 0; i < 256; i++) + if (te1->property[i] != te2->property[i]) + return 0; + return 1; +} + +static unsigned long +uint8_byte_table_hash (Lisp_Object obj, int depth) +{ + Lisp_Uint8_Byte_Table *te = XUINT8_BYTE_TABLE (obj); + int i; + hashcode_t hash = 0; + + for (i = 0; i < 256; i++) + hash = HASH2 (hash, te->property[i]); + return hash; +} + +DEFINE_LRECORD_IMPLEMENTATION ("uint8-byte-table", uint8_byte_table, + mark_uint8_byte_table, + print_uint8_byte_table, + 0, uint8_byte_table_equal, + uint8_byte_table_hash, + 0 /* uint8_byte_table_description */, + Lisp_Uint8_Byte_Table); + +static Lisp_Object +make_uint8_byte_table (unsigned char initval) +{ + Lisp_Object obj; + int i; + Lisp_Uint8_Byte_Table *cte; + + cte = alloc_lcrecord_type (Lisp_Uint8_Byte_Table, + &lrecord_uint8_byte_table); + + for (i = 0; i < 256; i++) + cte->property[i] = initval; + + XSETUINT8_BYTE_TABLE (obj, cte); + return obj; +} + +static int +uint8_byte_table_same_value_p (Lisp_Object obj) +{ + Lisp_Uint8_Byte_Table *bte = XUINT8_BYTE_TABLE (obj); + unsigned char v0 = bte->property[0]; + int i; + + for (i = 1; i < 256; i++) + { + if (bte->property[i] != v0) + return 0; + } + return -1; +} + + +#define BT_UINT16_MIN 0 +#define BT_UINT16_MAX (USHRT_MAX - 3) +#define BT_UINT16_t (USHRT_MAX - 2) +#define BT_UINT16_nil (USHRT_MAX - 1) +#define BT_UINT16_unbound USHRT_MAX + +INLINE_HEADER int INT_UINT16_P (Lisp_Object obj); +INLINE_HEADER int UINT16_VALUE_P (Lisp_Object obj); +INLINE_HEADER unsigned short UINT16_ENCODE (Lisp_Object obj); +INLINE_HEADER Lisp_Object UINT16_DECODE (unsigned short us); + +INLINE_HEADER int +INT_UINT16_P (Lisp_Object obj) +{ + if (INTP (obj)) + { + int num = XINT (obj); + + return (BT_UINT16_MIN <= num) && (num <= BT_UINT16_MAX); + } + else + return 0; +} + +INLINE_HEADER int +UINT16_VALUE_P (Lisp_Object obj) +{ + return EQ (obj, Qunbound) + || EQ (obj, Qnil) || EQ (obj, Qt) || INT_UINT16_P (obj); +} + +INLINE_HEADER unsigned short +UINT16_ENCODE (Lisp_Object obj) +{ + if (EQ (obj, Qunbound)) + return BT_UINT16_unbound; + else if (EQ (obj, Qnil)) + return BT_UINT16_nil; + else if (EQ (obj, Qt)) + return BT_UINT16_t; + else + return XINT (obj); +} + +INLINE_HEADER Lisp_Object +UINT16_DECODE (unsigned short n) +{ + if (n == BT_UINT16_unbound) + return Qunbound; + else if (n == BT_UINT16_nil) + return Qnil; + else if (n == BT_UINT16_t) + return Qt; + else + return make_int (n); +} + +INLINE_HEADER unsigned short +UINT8_TO_UINT16 (unsigned char n) +{ + if (n == BT_UINT8_unbound) + return BT_UINT16_unbound; + else if (n == BT_UINT8_nil) + return BT_UINT16_nil; + else if (n == BT_UINT8_t) + return BT_UINT16_t; + else + return n; +} + +static Lisp_Object +mark_uint16_byte_table (Lisp_Object obj) +{ + return Qnil; +} + +static void +print_uint16_byte_table (Lisp_Object obj, + Lisp_Object printcharfun, int escapeflag) +{ + Lisp_Uint16_Byte_Table *bte = XUINT16_BYTE_TABLE (obj); + int i; + struct gcpro gcpro1, gcpro2; + GCPRO2 (obj, printcharfun); + + write_c_string ("\n#property[i]; + if ( (i & 15) == 0 ) + write_c_string ("\n ", printcharfun); + write_c_string (" ", printcharfun); + if (n == BT_UINT16_unbound) + write_c_string ("void", printcharfun); + else if (n == BT_UINT16_nil) + write_c_string ("nil", printcharfun); + else if (n == BT_UINT16_t) + write_c_string ("t", printcharfun); + else + { + char buf[7]; + + sprintf (buf, "%hd", n); + write_c_string (buf, printcharfun); + } + } + UNGCPRO; + write_c_string (">", printcharfun); +} + +static int +uint16_byte_table_equal (Lisp_Object obj1, Lisp_Object obj2, int depth) +{ + Lisp_Uint16_Byte_Table *te1 = XUINT16_BYTE_TABLE (obj1); + Lisp_Uint16_Byte_Table *te2 = XUINT16_BYTE_TABLE (obj2); + int i; + + for (i = 0; i < 256; i++) + if (te1->property[i] != te2->property[i]) + return 0; + return 1; +} + +static unsigned long +uint16_byte_table_hash (Lisp_Object obj, int depth) +{ + Lisp_Uint16_Byte_Table *te = XUINT16_BYTE_TABLE (obj); + int i; + hashcode_t hash = 0; + + for (i = 0; i < 256; i++) + hash = HASH2 (hash, te->property[i]); + return hash; +} + +DEFINE_LRECORD_IMPLEMENTATION ("uint16-byte-table", uint16_byte_table, + mark_uint16_byte_table, + print_uint16_byte_table, + 0, uint16_byte_table_equal, + uint16_byte_table_hash, + 0 /* uint16_byte_table_description */, + Lisp_Uint16_Byte_Table); + +static Lisp_Object +make_uint16_byte_table (unsigned short initval) +{ + Lisp_Object obj; + int i; + Lisp_Uint16_Byte_Table *cte; + + cte = alloc_lcrecord_type (Lisp_Uint16_Byte_Table, + &lrecord_uint16_byte_table); + + for (i = 0; i < 256; i++) + cte->property[i] = initval; + + XSETUINT16_BYTE_TABLE (obj, cte); + return obj; +} + +static Lisp_Object +expand_uint8_byte_table_to_uint16 (Lisp_Object table) +{ + Lisp_Object obj; + int i; + Lisp_Uint8_Byte_Table* bte = XUINT8_BYTE_TABLE(table); + Lisp_Uint16_Byte_Table* cte; + + cte = alloc_lcrecord_type (Lisp_Uint16_Byte_Table, + &lrecord_uint16_byte_table); + for (i = 0; i < 256; i++) + { + cte->property[i] = UINT8_TO_UINT16 (bte->property[i]); + } + XSETUINT16_BYTE_TABLE (obj, cte); + return obj; +} + +static int +uint16_byte_table_same_value_p (Lisp_Object obj) +{ + Lisp_Uint16_Byte_Table *bte = XUINT16_BYTE_TABLE (obj); + unsigned short v0 = bte->property[0]; + int i; + + for (i = 1; i < 256; i++) + { + if (bte->property[i] != v0) + return 0; + } + return -1; +} + + +static Lisp_Object +mark_byte_table (Lisp_Object obj) +{ + Lisp_Byte_Table *cte = XBYTE_TABLE (obj); + int i; + + for (i = 0; i < 256; i++) + { + mark_object (cte->property[i]); + } + return Qnil; +} + +static void +print_byte_table (Lisp_Object obj, Lisp_Object printcharfun, int escapeflag) +{ + Lisp_Byte_Table *bte = XBYTE_TABLE (obj); + int i; + struct gcpro gcpro1, gcpro2; + GCPRO2 (obj, printcharfun); + + write_c_string ("\n#property[i]; + if ( (i & 15) == 0 ) + write_c_string ("\n ", printcharfun); + write_c_string (" ", printcharfun); + if (EQ (elt, Qunbound)) + write_c_string ("void", printcharfun); + else + print_internal (elt, printcharfun, escapeflag); + } + UNGCPRO; + write_c_string (">", printcharfun); +} + +static int +byte_table_equal (Lisp_Object obj1, Lisp_Object obj2, int depth) +{ + Lisp_Byte_Table *cte1 = XBYTE_TABLE (obj1); + Lisp_Byte_Table *cte2 = XBYTE_TABLE (obj2); + int i; + + for (i = 0; i < 256; i++) + if (BYTE_TABLE_P (cte1->property[i])) + { + if (BYTE_TABLE_P (cte2->property[i])) + { + if (!byte_table_equal (cte1->property[i], + cte2->property[i], depth + 1)) + return 0; + } + else + return 0; + } + else + if (!internal_equal (cte1->property[i], cte2->property[i], depth + 1)) + return 0; + return 1; +} + +static unsigned long +byte_table_hash (Lisp_Object obj, int depth) +{ + Lisp_Byte_Table *cte = XBYTE_TABLE (obj); + + return internal_array_hash (cte->property, 256, depth); +} + +static const struct lrecord_description byte_table_description[] = { + { XD_LISP_OBJECT_ARRAY, offsetof(Lisp_Byte_Table, property), 256 }, + { XD_END } +}; + +DEFINE_LRECORD_IMPLEMENTATION ("byte-table", byte_table, + mark_byte_table, + print_byte_table, + 0, byte_table_equal, + byte_table_hash, + byte_table_description, + Lisp_Byte_Table); + +static Lisp_Object +make_byte_table (Lisp_Object initval) +{ + Lisp_Object obj; + int i; + Lisp_Byte_Table *cte; + + cte = alloc_lcrecord_type (Lisp_Byte_Table, &lrecord_byte_table); + + for (i = 0; i < 256; i++) + cte->property[i] = initval; + + XSETBYTE_TABLE (obj, cte); + return obj; +} + +static int +byte_table_same_value_p (Lisp_Object obj) +{ + Lisp_Byte_Table *bte = XBYTE_TABLE (obj); + Lisp_Object v0 = bte->property[0]; + int i; + + for (i = 1; i < 256; i++) + { + if (!internal_equal (bte->property[i], v0, 0)) + return 0; + } + return -1; +} + + +Lisp_Object get_byte_table (Lisp_Object table, unsigned char idx); +Lisp_Object put_byte_table (Lisp_Object table, unsigned char idx, + Lisp_Object value); + +Lisp_Object +get_byte_table (Lisp_Object table, unsigned char idx) +{ + if (UINT8_BYTE_TABLE_P (table)) + return UINT8_DECODE (XUINT8_BYTE_TABLE(table)->property[idx]); + else if (UINT16_BYTE_TABLE_P (table)) + return UINT16_DECODE (XUINT16_BYTE_TABLE(table)->property[idx]); + else if (BYTE_TABLE_P (table)) + return XBYTE_TABLE(table)->property[idx]; + else + return table; +} + +Lisp_Object +put_byte_table (Lisp_Object table, unsigned char idx, Lisp_Object value) +{ + if (UINT8_BYTE_TABLE_P (table)) + { + if (UINT8_VALUE_P (value)) + { + XUINT8_BYTE_TABLE(table)->property[idx] = UINT8_ENCODE (value); + if (!UINT8_BYTE_TABLE_P (value) && + !UINT16_BYTE_TABLE_P (value) && !BYTE_TABLE_P (value) + && uint8_byte_table_same_value_p (table)) + { + return value; + } + } + else if (UINT16_VALUE_P (value)) + { + Lisp_Object new = expand_uint8_byte_table_to_uint16 (table); + + XUINT16_BYTE_TABLE(new)->property[idx] = UINT16_ENCODE (value); + return new; + } + else + { + Lisp_Object new = make_byte_table (Qnil); + int i; + + for (i = 0; i < 256; i++) + { + XBYTE_TABLE(new)->property[i] + = UINT8_DECODE (XUINT8_BYTE_TABLE(table)->property[i]); + } + XBYTE_TABLE(new)->property[idx] = value; + return new; + } + } + else if (UINT16_BYTE_TABLE_P (table)) + { + if (UINT16_VALUE_P (value)) + { + XUINT16_BYTE_TABLE(table)->property[idx] = UINT16_ENCODE (value); + if (!UINT8_BYTE_TABLE_P (value) && + !UINT16_BYTE_TABLE_P (value) && !BYTE_TABLE_P (value) + && uint16_byte_table_same_value_p (table)) + { + return value; + } + } + else + { + Lisp_Object new = make_byte_table (Qnil); + int i; + + for (i = 0; i < 256; i++) + { + XBYTE_TABLE(new)->property[i] + = UINT16_DECODE (XUINT16_BYTE_TABLE(table)->property[i]); + } + XBYTE_TABLE(new)->property[idx] = value; + return new; + } + } + else if (BYTE_TABLE_P (table)) + { + XBYTE_TABLE(table)->property[idx] = value; + if (!UINT8_BYTE_TABLE_P (value) && + !UINT16_BYTE_TABLE_P (value) && !BYTE_TABLE_P (value) + && byte_table_same_value_p (table)) + { + return value; + } + } + else if (!internal_equal (table, value, 0)) + { + if (UINT8_VALUE_P (table) && UINT8_VALUE_P (value)) + { + table = make_uint8_byte_table (UINT8_ENCODE (table)); + XUINT8_BYTE_TABLE(table)->property[idx] = UINT8_ENCODE (value); + } + else if (UINT16_VALUE_P (table) && UINT16_VALUE_P (value)) + { + table = make_uint16_byte_table (UINT16_ENCODE (table)); + XUINT16_BYTE_TABLE(table)->property[idx] = UINT16_ENCODE (value); + } + else + { + table = make_byte_table (table); + XBYTE_TABLE(table)->property[idx] = value; + } + } + return table; +} + +static Lisp_Object +mark_char_id_table (Lisp_Object obj) +{ + Lisp_Char_ID_Table *cte = XCHAR_ID_TABLE (obj); + + return cte->table; +} + +static void +print_char_id_table (Lisp_Object obj, Lisp_Object printcharfun, int escapeflag) +{ + Lisp_Object table = XCHAR_ID_TABLE (obj)->table; + int i; + struct gcpro gcpro1, gcpro2; + GCPRO2 (obj, printcharfun); + + write_c_string ("#", printcharfun); +} + +static int +char_id_table_equal (Lisp_Object obj1, Lisp_Object obj2, int depth) +{ + Lisp_Object table1 = XCHAR_ID_TABLE (obj1)->table; + Lisp_Object table2 = XCHAR_ID_TABLE (obj2)->table; + int i; + + for (i = 0; i < 256; i++) + { + if (!internal_equal (get_byte_table (table1, i), + get_byte_table (table2, i), 0)) + return 0; + } + return -1; +} + +static unsigned long +char_id_table_hash (Lisp_Object obj, int depth) +{ + Lisp_Char_ID_Table *cte = XCHAR_ID_TABLE (obj); + + return char_id_table_hash (cte->table, depth + 1); +} + +static const struct lrecord_description char_id_table_description[] = { + { XD_LISP_OBJECT, offsetof(Lisp_Char_ID_Table, table) }, + { XD_END } +}; + +DEFINE_LRECORD_IMPLEMENTATION ("char-id-table", char_id_table, + mark_char_id_table, + print_char_id_table, + 0, char_id_table_equal, + char_id_table_hash, + char_id_table_description, + Lisp_Char_ID_Table); + +static Lisp_Object +make_char_id_table (Lisp_Object initval) +{ + Lisp_Object obj; + Lisp_Char_ID_Table *cte; + + cte = alloc_lcrecord_type (Lisp_Char_ID_Table, &lrecord_char_id_table); + + cte->table = make_byte_table (initval); + + XSETCHAR_ID_TABLE (obj, cte); + return obj; +} + + +Lisp_Object +get_char_id_table (Emchar ch, Lisp_Object table) +{ + unsigned int code = ch; + + return + get_byte_table + (get_byte_table + (get_byte_table + (get_byte_table + (XCHAR_ID_TABLE (table)->table, + (unsigned char)(code >> 24)), + (unsigned char) (code >> 16)), + (unsigned char) (code >> 8)), + (unsigned char) code); +} + +void put_char_id_table (Emchar ch, Lisp_Object value, Lisp_Object table); +void +put_char_id_table (Emchar ch, Lisp_Object value, Lisp_Object table) +{ + unsigned int code = ch; + Lisp_Object table1, table2, table3, table4; + + table1 = XCHAR_ID_TABLE (table)->table; + table2 = get_byte_table (table1, (unsigned char)(code >> 24)); + table3 = get_byte_table (table2, (unsigned char)(code >> 16)); + table4 = get_byte_table (table3, (unsigned char)(code >> 8)); + + table4 = put_byte_table (table4, (unsigned char)code, value); + table3 = put_byte_table (table3, (unsigned char)(code >> 8), table4); + table2 = put_byte_table (table2, (unsigned char)(code >> 16), table3); + XCHAR_ID_TABLE (table)->table + = put_byte_table (table1, (unsigned char)(code >> 24), table2); +} + + +Lisp_Object Vchar_attribute_hash_table; +Lisp_Object Vcharacter_composition_table; +Lisp_Object Vcharacter_variant_table; + +Lisp_Object Qideograph_daikanwa; +Lisp_Object Q_decomposition; +Lisp_Object Qucs; +Lisp_Object Q_ucs; +Lisp_Object Qcompat; +Lisp_Object Qisolated; +Lisp_Object Qinitial; +Lisp_Object Qmedial; +Lisp_Object Qfinal; +Lisp_Object Qvertical; +Lisp_Object QnoBreak; +Lisp_Object Qfraction; +Lisp_Object Qsuper; +Lisp_Object Qsub; +Lisp_Object Qcircle; +Lisp_Object Qsquare; +Lisp_Object Qwide; +Lisp_Object Qnarrow; +Lisp_Object Qsmall; +Lisp_Object Qfont; + +Emchar to_char_id (Lisp_Object v, char* err_msg, Lisp_Object err_arg); + +Lisp_Object put_char_ccs_code_point (Lisp_Object character, + Lisp_Object ccs, Lisp_Object value); +Lisp_Object remove_char_ccs (Lisp_Object character, Lisp_Object ccs); + +Emchar +to_char_id (Lisp_Object v, char* err_msg, Lisp_Object err_arg) +{ + if (INTP (v)) + return XINT (v); + if (CHARP (v)) + return XCHAR (v); + else if (EQ (v, Qcompat)) + return -1; + else if (EQ (v, Qisolated)) + return -2; + else if (EQ (v, Qinitial)) + return -3; + else if (EQ (v, Qmedial)) + return -4; + else if (EQ (v, Qfinal)) + return -5; + else if (EQ (v, Qvertical)) + return -6; + else if (EQ (v, QnoBreak)) + return -7; + else if (EQ (v, Qfraction)) + return -8; + else if (EQ (v, Qsuper)) + return -9; + else if (EQ (v, Qsub)) + return -10; + else if (EQ (v, Qcircle)) + return -11; + else if (EQ (v, Qsquare)) + return -12; + else if (EQ (v, Qwide)) + return -13; + else if (EQ (v, Qnarrow)) + return -14; + else if (EQ (v, Qsmall)) + return -15; + else if (EQ (v, Qfont)) + return -16; + else + signal_simple_error (err_msg, err_arg); +} + +DEFUN ("get-composite-char", Fget_composite_char, 1, 1, 0, /* +Return character corresponding with list. +*/ + (list)) +{ + Lisp_Object table = Vcharacter_composition_table; + Lisp_Object rest = list; + + while (CONSP (rest)) + { + Lisp_Object v = Fcar (rest); + Lisp_Object ret; + Emchar c = to_char_id (v, "Invalid value for composition", list); + + ret = get_char_id_table (c, table); + + rest = Fcdr (rest); + if (NILP (rest)) + { + if (!CHAR_ID_TABLE_P (ret)) + return ret; + else + return Qt; + } + else if (!CONSP (rest)) + break; + else if (CHAR_ID_TABLE_P (ret)) + table = ret; + else + signal_simple_error ("Invalid table is found with", list); + } + signal_simple_error ("Invalid value for composition", list); +} + +DEFUN ("char-variants", Fchar_variants, 1, 1, 0, /* +Return variants of CHARACTER. +*/ + (character)) +{ + CHECK_CHAR (character); + return Fcopy_list (get_char_id_table (XCHAR (character), + Vcharacter_variant_table)); +} + + +/* We store the char-attributes in hash tables with the names as the + key and the actual char-id-table object as the value. Occasionally + we need to use them in a list format. These routines provide us + with that. */ +struct char_attribute_list_closure +{ + Lisp_Object *char_attribute_list; +}; + +static int +add_char_attribute_to_list_mapper (Lisp_Object key, Lisp_Object value, + void *char_attribute_list_closure) +{ + /* This function can GC */ + struct char_attribute_list_closure *calcl + = (struct char_attribute_list_closure*) char_attribute_list_closure; + Lisp_Object *char_attribute_list = calcl->char_attribute_list; + + *char_attribute_list = Fcons (key, *char_attribute_list); + return 0; +} + +DEFUN ("char-attribute-list", Fchar_attribute_list, 0, 0, 0, /* +Return the list of all existing character attributes except coded-charsets. +*/ + ()) +{ + Lisp_Object char_attribute_list = Qnil; + struct gcpro gcpro1; + struct char_attribute_list_closure char_attribute_list_closure; + + GCPRO1 (char_attribute_list); + char_attribute_list_closure.char_attribute_list = &char_attribute_list; + elisp_maphash (add_char_attribute_to_list_mapper, + Vchar_attribute_hash_table, + &char_attribute_list_closure); + UNGCPRO; + return char_attribute_list; +} + +DEFUN ("find-char-attribute-table", Ffind_char_attribute_table, 1, 1, 0, /* +Return char-id-table corresponding to ATTRIBUTE. +*/ + (attribute)) +{ + return Fgethash (attribute, Vchar_attribute_hash_table, Qnil); +} + + +/* We store the char-id-tables in hash tables with the attributes as + the key and the actual char-id-table object as the value. Each + char-id-table stores values of an attribute corresponding with + characters. Occasionally we need to get attributes of a character + in a association-list format. These routines provide us with + that. */ +struct char_attribute_alist_closure +{ + Emchar char_id; + Lisp_Object *char_attribute_alist; +}; + +static int +add_char_attribute_alist_mapper (Lisp_Object key, Lisp_Object value, + void *char_attribute_alist_closure) +{ + /* This function can GC */ + struct char_attribute_alist_closure *caacl = + (struct char_attribute_alist_closure*) char_attribute_alist_closure; + Lisp_Object ret = get_char_id_table (caacl->char_id, value); + if (!UNBOUNDP (ret)) + { + Lisp_Object *char_attribute_alist = caacl->char_attribute_alist; + *char_attribute_alist + = Fcons (Fcons (key, ret), *char_attribute_alist); + } + return 0; +} + +DEFUN ("char-attribute-alist", Fchar_attribute_alist, 1, 1, 0, /* +Return the alist of attributes of CHARACTER. +*/ + (character)) +{ + Lisp_Object alist = Qnil; + int i; + + CHECK_CHAR (character); + { + struct gcpro gcpro1; + struct char_attribute_alist_closure char_attribute_alist_closure; + + GCPRO1 (alist); + char_attribute_alist_closure.char_id = XCHAR (character); + char_attribute_alist_closure.char_attribute_alist = &alist; + elisp_maphash (add_char_attribute_alist_mapper, + Vchar_attribute_hash_table, + &char_attribute_alist_closure); + UNGCPRO; + } + + for (i = 0; i < countof (chlook->charset_by_leading_byte); i++) + { + Lisp_Object ccs = chlook->charset_by_leading_byte[i]; -/* The various pre-defined charsets. */ + if (!NILP (ccs)) + { + Lisp_Object encoding_table = XCHARSET_ENCODING_TABLE (ccs); + Lisp_Object cpos; + + if ( CHAR_ID_TABLE_P (encoding_table) + && INTP (cpos = get_char_id_table (XCHAR (character), + encoding_table)) ) + { + alist = Fcons (Fcons (ccs, cpos), alist); + } + } + } + return alist; +} -Lisp_Object Vcharset_ascii; -Lisp_Object Vcharset_control_1; -Lisp_Object Vcharset_latin_iso8859_1; -Lisp_Object Vcharset_latin_iso8859_2; -Lisp_Object Vcharset_latin_iso8859_3; -Lisp_Object Vcharset_latin_iso8859_4; -Lisp_Object Vcharset_thai_tis620; -Lisp_Object Vcharset_greek_iso8859_7; -Lisp_Object Vcharset_arabic_iso8859_6; -Lisp_Object Vcharset_hebrew_iso8859_8; -Lisp_Object Vcharset_katakana_jisx0201; -Lisp_Object Vcharset_latin_jisx0201; -Lisp_Object Vcharset_cyrillic_iso8859_5; -Lisp_Object Vcharset_latin_iso8859_9; -Lisp_Object Vcharset_japanese_jisx0208_1978; -Lisp_Object Vcharset_chinese_gb2312; -Lisp_Object Vcharset_japanese_jisx0208; -Lisp_Object Vcharset_korean_ksc5601; -Lisp_Object Vcharset_japanese_jisx0212; -Lisp_Object Vcharset_chinese_cns11643_1; -Lisp_Object Vcharset_chinese_cns11643_2; -Lisp_Object Vcharset_chinese_big5_1; -Lisp_Object Vcharset_chinese_big5_2; +DEFUN ("get-char-attribute", Fget_char_attribute, 2, 3, 0, /* +Return the value of CHARACTER's ATTRIBUTE. +Return DEFAULT-VALUE if the value is not exist. +*/ + (character, attribute, default_value)) +{ + Lisp_Object ccs; -#ifdef ENABLE_COMPOSITE_CHARS -Lisp_Object Vcharset_composite; + CHECK_CHAR (character); + if (!NILP (ccs = Ffind_charset (attribute))) + { + Lisp_Object encoding_table = XCHARSET_ENCODING_TABLE (ccs); -/* Hash tables for composite chars. One maps string representing - composed chars to their equivalent chars; one goes the - other way. */ -Lisp_Object Vcomposite_char_char2string_hash_table; -Lisp_Object Vcomposite_char_string2char_hash_table; + if (CHAR_ID_TABLE_P (encoding_table)) + return get_char_id_table (XCHAR (character), encoding_table); + } + else + { + Lisp_Object table = Fgethash (attribute, + Vchar_attribute_hash_table, + Qunbound); + if (!UNBOUNDP (table)) + { + Lisp_Object ret = get_char_id_table (XCHAR (character), table); + if (!UNBOUNDP (ret)) + return ret; + } + } + return default_value; +} -static int composite_char_row_next; -static int composite_char_col_next; +DEFUN ("put-char-attribute", Fput_char_attribute, 3, 3, 0, /* +Store CHARACTER's ATTRIBUTE with VALUE. +*/ + (character, attribute, value)) +{ + Lisp_Object ccs; -#endif /* ENABLE_COMPOSITE_CHARS */ + CHECK_CHAR (character); + ccs = Ffind_charset (attribute); + if (!NILP (ccs)) + { + return put_char_ccs_code_point (character, ccs, value); + } + else if (EQ (attribute, Q_decomposition)) + { + Lisp_Object seq; -struct charset_lookup *chlook; + if (!CONSP (value)) + signal_simple_error ("Invalid value for ->decomposition", + value); -static const struct lrecord_description charset_lookup_description_1[] = { - { XD_LISP_OBJECT_ARRAY, offsetof (struct charset_lookup, charset_by_leading_byte), 128+4*128*2 }, - { XD_END } -}; + if (CONSP (Fcdr (value))) + { + Lisp_Object rest = value; + Lisp_Object table = Vcharacter_composition_table; + size_t len; + int i = 0; + + GET_EXTERNAL_LIST_LENGTH (rest, len); + seq = make_vector (len, Qnil); + + while (CONSP (rest)) + { + Lisp_Object v = Fcar (rest); + Lisp_Object ntable; + Emchar c + = to_char_id (v, "Invalid value for ->decomposition", value); + + if (c < 0) + XVECTOR_DATA(seq)[i++] = v; + else + XVECTOR_DATA(seq)[i++] = make_char (c); + rest = Fcdr (rest); + if (!CONSP (rest)) + { + put_char_id_table (c, character, table); + break; + } + else + { + ntable = get_char_id_table (c, table); + if (!CHAR_ID_TABLE_P (ntable)) + { + ntable = make_char_id_table (Qnil); + put_char_id_table (c, ntable, table); + } + table = ntable; + } + } + } + else + { + Lisp_Object v = Fcar (value); + + if (INTP (v)) + { + Emchar c = XINT (v); + Lisp_Object ret + = get_char_id_table (c, Vcharacter_variant_table); + + if (NILP (Fmemq (v, ret))) + { + put_char_id_table (c, Fcons (character, ret), + Vcharacter_variant_table); + } + } + seq = make_vector (1, v); + } + value = seq; + } + else if (EQ (attribute, Q_ucs)) + { + Lisp_Object ret; + Emchar c; -static const struct struct_description charset_lookup_description = { - sizeof (struct charset_lookup), - charset_lookup_description_1 -}; + if (!INTP (value)) + signal_simple_error ("Invalid value for ->ucs", value); -/* Table of number of bytes in the string representation of a character - indexed by the first byte of that representation. + c = XINT (value); - rep_bytes_by_first_byte(c) is more efficient than the equivalent - canonical computation: + ret = get_char_id_table (c, Vcharacter_variant_table); + if (NILP (Fmemq (character, ret))) + { + put_char_id_table (c, Fcons (character, ret), + Vcharacter_variant_table); + } + } + { + Lisp_Object table = Fgethash (attribute, + Vchar_attribute_hash_table, + Qnil); + + if (NILP (table)) + { + table = make_char_id_table (Qunbound); + Fputhash (attribute, table, Vchar_attribute_hash_table); + } + put_char_id_table (XCHAR (character), value, table); + return value; + } +} + +DEFUN ("remove-char-attribute", Fremove_char_attribute, 2, 2, 0, /* +Remove CHARACTER's ATTRIBUTE. +*/ + (character, attribute)) +{ + Lisp_Object ccs; - XCHARSET_REP_BYTES (CHARSET_BY_LEADING_BYTE (c)) */ + CHECK_CHAR (character); + ccs = Ffind_charset (attribute); + if (!NILP (ccs)) + { + return remove_char_ccs (character, ccs); + } + else + { + Lisp_Object table = Fgethash (attribute, + Vchar_attribute_hash_table, + Qunbound); + if (!UNBOUNDP (table)) + { + put_char_id_table (XCHAR (character), Qunbound, table); + return Qt; + } + } + return Qnil; +} -const Bytecount rep_bytes_by_first_byte[0xA0] = -{ /* 0x00 - 0x7f are for straight ASCII */ - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - /* 0x80 - 0x8f are for Dimension-1 official charsets */ - 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - /* 0x90 - 0x9d are for Dimension-2 official charsets */ - /* 0x9e is for Dimension-1 private charsets */ - /* 0x9f is for Dimension-2 private charsets */ - 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4 -}; +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) +{ + int i; + + if (XVECTOR_LENGTH (v) > ccs_len) + return -1; + + for (i = 0; i < XVECTOR_LENGTH (v); i++) + { + Lisp_Object c = XVECTOR_DATA(v)[i]; + + if (!NILP (c) && !CHARP (c)) + { + if (VECTORP (c)) + { + int ret = decoding_table_check_elements (c, dim - 1, ccs_len); + if (ret) + return ret; + } + else + return -2; + } + } + 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_older_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) +{ + Lisp_Object encoding_table; + + if (!EQ (XCHARSET_NAME (ccs), Qucs) + || (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)) + { /* obsolete representation: value must be a list of bytes */ + Lisp_Object ret = Fcar (value); + Lisp_Object rest; + + if (!INTP (ret)) + signal_simple_error ("Invalid value for coded-charset", value); + code_point = XINT (ret); + if (XCHARSET_GRAPHIC (ccs) == 1) + code_point &= 0x7F; + rest = Fcdr (value); + while (!NILP (rest)) + { + int j; + + if (!CONSP (rest)) + signal_simple_error ("Invalid value for coded-charset", + value); + ret = Fcar (rest); + if (!INTP (ret)) + signal_simple_error ("Invalid value for coded-charset", + value); + j = XINT (ret); + if (XCHARSET_GRAPHIC (ccs) == 1) + j &= 0x7F; + code_point = (code_point << 8) | j; + rest = Fcdr (rest); + } + value = make_int (code_point); + } + else if (INTP (value)) + { + code_point = XINT (value); + if (XCHARSET_GRAPHIC (ccs) == 1) + { + code_point &= 0x7F7F7F7F; + value = make_int (code_point); + } + } + else + signal_simple_error ("Invalid value for coded-charset", value); + + if (VECTORP (v)) + { + Lisp_Object cpos = Fget_char_attribute (character, ccs, Qnil); + if (!NILP (cpos)) + { + decoding_table_remove_char (v, dim, byte_offset, XINT (cpos)); + } + } + else + { + XCHARSET_DECODING_TABLE (ccs) + = v = make_older_vector (ccs_len, Qnil); + } + + decoding_table_put_char (v, dim, byte_offset, code_point, character); + } + if (NILP (encoding_table = XCHARSET_ENCODING_TABLE (ccs))) + { + XCHARSET_ENCODING_TABLE (ccs) + = encoding_table = make_char_id_table (Qnil); + } + put_char_id_table (XCHAR (character), value, encoding_table); + return Qt; +} + +Lisp_Object +remove_char_ccs (Lisp_Object character, Lisp_Object ccs) +{ + Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (ccs); + Lisp_Object encoding_table = XCHARSET_ENCODING_TABLE (ccs); + + if (VECTORP (decoding_table)) + { + Lisp_Object cpos = Fget_char_attribute (character, ccs, Qnil); + + if (!NILP (cpos)) + { + decoding_table_remove_char (decoding_table, + XCHARSET_DIMENSION (ccs), + XCHARSET_BYTE_OFFSET (ccs), + XINT (cpos)); + } + } + if (CHAR_ID_TABLE_P (encoding_table)) + { + put_char_id_table (XCHAR (character), Qnil, encoding_table); + } + return Qt; +} + +EXFUN (Fmake_char, 3); +EXFUN (Fdecode_char, 2); + +DEFUN ("define-char", Fdefine_char, 1, 1, 0, /* +Store character's ATTRIBUTES. +*/ + (attributes)) +{ + Lisp_Object rest = attributes; + Lisp_Object code = Fcdr (Fassq (Qucs, attributes)); + Lisp_Object character; +#if 0 + Lisp_Object daikanwa = Qnil; +#endif + + if (NILP (code)) + { + while (CONSP (rest)) + { + Lisp_Object cell = Fcar (rest); + Lisp_Object ccs; + + if (!LISTP (cell)) + signal_simple_error ("Invalid argument", attributes); + if (!NILP (ccs = Ffind_charset (Fcar (cell))) + && ((XCHARSET_FINAL (ccs) != 0) || + (XCHARSET_UCS_MAX (ccs) > 0)) ) + { + cell = Fcdr (cell); + if (CONSP (cell)) + character = Fmake_char (ccs, Fcar (cell), Fcar (Fcdr (cell))); + else + character = Fdecode_char (ccs, cell); + goto setup_attributes; + } + rest = Fcdr (rest); + } + if (!NILP (code = Fcdr (Fassq (Q_ucs, attributes)))) + { + if (!INTP (code)) + signal_simple_error ("Invalid argument", attributes); + else + character = make_char (XINT (code) + 0x100000); + goto setup_attributes; + } + return Qnil; + } + else if (!INTP (code)) + signal_simple_error ("Invalid argument", attributes); + else + character = make_char (XINT (code)); + + setup_attributes: + rest = attributes; + while (CONSP (rest)) + { + Lisp_Object cell = Fcar (rest); +#if 0 + Lisp_Object key = Fcar (cell); + Lisp_Object value = Fcdr (cell); +#endif + + if (!LISTP (cell)) + signal_simple_error ("Invalid argument", attributes); + +#if 0 + if (EQ (key, Qmorohashi_daikanwa)) + { + size_t len; + GET_EXTERNAL_LIST_LENGTH (value, len); + + if (len == 1) + { + if (NILP (daikanwa)) + daikanwa = Fcdr (Fassq (Qideograph_daikanwa, rest)); + if (EQ (Fcar (value), daikanwa)) + goto ignored; + } + } + else if (EQ (key, Qideograph_daikanwa)) + daikanwa = value; +#endif + + Fput_char_attribute (character, Fcar (cell), Fcdr (cell)); +#if 0 + ignored: +#endif + rest = Fcdr (rest); + } + return character; +} + +Lisp_Object Vutf_2000_version; +#endif + +#ifndef UTF2000 +int leading_code_private_11; +#endif Lisp_Object Qcharsetp; @@ -138,10 +1602,46 @@ Lisp_Object Qascii, Qjapanese_jisx0208_1978, Qchinese_gb2312, Qjapanese_jisx0208, + Qjapanese_jisx0208_1990, Qkorean_ksc5601, Qjapanese_jisx0212, Qchinese_cns11643_1, Qchinese_cns11643_2, +#ifdef UTF2000 + Qucs_bmp, + Qucs_cns, + Qlatin_viscii, + Qlatin_tcvn5712, + Qlatin_viscii_lower, + Qlatin_viscii_upper, + Qvietnamese_viscii_lower, + Qvietnamese_viscii_upper, + Qchinese_big5, + Qmojikyo, + Qmojikyo_2022_1, + Qmojikyo_pj_1, + Qmojikyo_pj_2, + Qmojikyo_pj_3, + Qmojikyo_pj_4, + Qmojikyo_pj_5, + Qmojikyo_pj_6, + Qmojikyo_pj_7, + Qmojikyo_pj_8, + Qmojikyo_pj_9, + Qmojikyo_pj_10, + Qmojikyo_pj_11, + Qmojikyo_pj_12, + Qmojikyo_pj_13, + Qmojikyo_pj_14, + Qmojikyo_pj_15, + Qmojikyo_pj_16, + Qmojikyo_pj_17, + Qmojikyo_pj_18, + Qmojikyo_pj_19, + Qmojikyo_pj_20, + Qmojikyo_pj_21, + Qethiopic_ucs, +#endif Qchinese_big5_1, Qchinese_big5_2, Qcomposite; @@ -150,9 +1650,6 @@ Lisp_Object Ql2r, Qr2l; Lisp_Object Vcharset_hash_table; -static Bufbyte next_allocated_1_byte_leading_byte; -static Bufbyte next_allocated_2_byte_leading_byte; - /* Composite characters are characters constructed by overstriking two or more regular characters. @@ -196,11 +1693,54 @@ Bytecount non_ascii_set_charptr_emchar (Bufbyte *str, Emchar c) { Bufbyte *p; - Bufbyte lb; +#ifndef UTF2000 + Charset_ID lb; int c1, c2; Lisp_Object charset; +#endif p = str; +#ifdef UTF2000 + if ( c <= 0x7f ) + { + *p++ = c; + } + else if ( c <= 0x7ff ) + { + *p++ = (c >> 6) | 0xc0; + *p++ = (c & 0x3f) | 0x80; + } + else if ( c <= 0xffff ) + { + *p++ = (c >> 12) | 0xe0; + *p++ = ((c >> 6) & 0x3f) | 0x80; + *p++ = (c & 0x3f) | 0x80; + } + else if ( c <= 0x1fffff ) + { + *p++ = (c >> 18) | 0xf0; + *p++ = ((c >> 12) & 0x3f) | 0x80; + *p++ = ((c >> 6) & 0x3f) | 0x80; + *p++ = (c & 0x3f) | 0x80; + } + else if ( c <= 0x3ffffff ) + { + *p++ = (c >> 24) | 0xf8; + *p++ = ((c >> 18) & 0x3f) | 0x80; + *p++ = ((c >> 12) & 0x3f) | 0x80; + *p++ = ((c >> 6) & 0x3f) | 0x80; + *p++ = (c & 0x3f) | 0x80; + } + else + { + *p++ = (c >> 30) | 0xfc; + *p++ = ((c >> 24) & 0x3f) | 0x80; + *p++ = ((c >> 18) & 0x3f) | 0x80; + *p++ = ((c >> 12) & 0x3f) | 0x80; + *p++ = ((c >> 6) & 0x3f) | 0x80; + *p++ = (c & 0x3f) | 0x80; + } +#else BREAKUP_CHAR (c, charset, c1, c2); lb = CHAR_LEADING_BYTE (c); if (LEADING_BYTE_PRIVATE_P (lb)) @@ -211,7 +1751,7 @@ non_ascii_set_charptr_emchar (Bufbyte *str, Emchar c) *p++ = c1 | 0x80; if (c2) *p++ = c2 | 0x80; - +#endif return (p - str); } @@ -222,6 +1762,49 @@ non_ascii_set_charptr_emchar (Bufbyte *str, Emchar c) Emchar non_ascii_charptr_emchar (const Bufbyte *str) { +#ifdef UTF2000 + Bufbyte b; + Emchar ch; + int len; + + b = *str++; + if ( b >= 0xfc ) + { + ch = (b & 0x01); + len = 5; + } + else if ( b >= 0xf8 ) + { + ch = b & 0x03; + len = 4; + } + else if ( b >= 0xf0 ) + { + ch = b & 0x07; + len = 3; + } + else if ( b >= 0xe0 ) + { + ch = b & 0x0f; + len = 2; + } + else if ( b >= 0xc0 ) + { + ch = b & 0x1f; + len = 1; + } + else + { + ch = b; + len = 0; + } + for( ; len > 0; len-- ) + { + b = *str++; + ch = ( ch << 6 ) | ( b & 0x3f ); + } + return ch; +#else Bufbyte i0 = *str, i1, i2 = 0; Lisp_Object charset; @@ -238,11 +1821,13 @@ non_ascii_charptr_emchar (const Bufbyte *str) i2 = *++str & 0x7F; return MAKE_CHAR (charset, i1, i2); +#endif } /* Return whether CH is a valid Emchar, assuming it's non-ASCII. Do not call this directly. Use the macro valid_char_p() instead. */ +#ifndef UTF2000 int non_ascii_valid_char_p (Emchar ch) { @@ -319,32 +1904,25 @@ non_ascii_valid_char_p (Emchar ch) return (XCHARSET_CHARS (charset) == 96); } } +#endif /************************************************************************/ /* Basic string functions */ /************************************************************************/ -/* Copy the character pointed to by PTR into STR, assuming it's - non-ASCII. Do not call this directly. Use the macro - charptr_copy_char() instead. */ +/* Copy the character pointed to by SRC into DST. Do not call this + directly. Use the macro charptr_copy_char() instead. + Return the number of bytes copied. */ Bytecount -non_ascii_charptr_copy_char (const Bufbyte *ptr, Bufbyte *str) +non_ascii_charptr_copy_char (const Bufbyte *src, Bufbyte *dst) { - Bufbyte *strptr = str; - *strptr = *ptr++; - switch (REP_BYTES_BY_FIRST_BYTE (*strptr)) - { - /* Notice fallthrough. */ - case 4: *++strptr = *ptr++; - case 3: *++strptr = *ptr++; - case 2: *++strptr = *ptr; - break; - default: - abort (); - } - return strptr + 1 - str; + unsigned int bytes = REP_BYTES_BY_FIRST_BYTE (*src); + unsigned int i; + for (i = bytes; i; i--, dst++, src++) + *dst = *src; + return bytes; } @@ -361,26 +1939,15 @@ Lstream_get_emchar_1 (Lstream *stream, int ch) { Bufbyte str[MAX_EMCHAR_LEN]; Bufbyte *strptr = str; + unsigned int bytes; - str[0] = (Bufbyte) ch; - switch (REP_BYTES_BY_FIRST_BYTE (ch)) - { - /* Notice fallthrough. */ - case 4: - ch = Lstream_getc (stream); - assert (ch >= 0); - *++strptr = (Bufbyte) ch; - case 3: - ch = Lstream_getc (stream); - assert (ch >= 0); - *++strptr = (Bufbyte) ch; - case 2: - ch = Lstream_getc (stream); - assert (ch >= 0); - *++strptr = (Bufbyte) ch; - break; - default: - abort (); + str[0] = (Bufbyte) ch; + + for (bytes = REP_BYTES_BY_FIRST_BYTE (ch) - 1; bytes; bytes--) + { + int c = Lstream_getc (stream); + bufpos_checking_assert (c >= 0); + *++strptr = (Bufbyte) c; } return charptr_emchar (str); } @@ -416,6 +1983,10 @@ mark_charset (Lisp_Object obj) mark_object (cs->doc_string); mark_object (cs->registry); mark_object (cs->ccl_program); +#ifdef UTF2000 + mark_object (cs->encoding_table); + /* mark_object (cs->decoding_table); */ +#endif return cs->name; } @@ -438,11 +2009,9 @@ print_charset (Lisp_Object obj, Lisp_Object printcharfun, int escapeflag) print_internal (CHARSET_LONG_NAME (cs), printcharfun, 1); write_c_string (" ", printcharfun); print_internal (CHARSET_DOC_STRING (cs), printcharfun, 1); - sprintf (buf, " %s %s cols=%d g%d final='%c' reg=", - CHARSET_TYPE (cs) == CHARSET_TYPE_94 ? "94" : - CHARSET_TYPE (cs) == CHARSET_TYPE_96 ? "96" : - CHARSET_TYPE (cs) == CHARSET_TYPE_94X94 ? "94x94" : - "96x96", + sprintf (buf, " %d^%d %s cols=%d g%d final='%c' reg=", + CHARSET_CHARS (cs), + CHARSET_DIMENSION (cs), CHARSET_DIRECTION (cs) == CHARSET_LEFT_TO_RIGHT ? "l2r" : "r2l", CHARSET_COLUMNS (cs), CHARSET_GRAPHIC (cs), @@ -461,20 +2030,30 @@ static const struct lrecord_description charset_description[] = { { XD_LISP_OBJECT, offsetof (Lisp_Charset, long_name) }, { XD_LISP_OBJECT, offsetof (Lisp_Charset, reverse_direction_charset) }, { XD_LISP_OBJECT, offsetof (Lisp_Charset, ccl_program) }, +#ifdef UTF2000 + { XD_LISP_OBJECT, offsetof (Lisp_Charset, decoding_table) }, + { XD_LISP_OBJECT, offsetof (Lisp_Charset, encoding_table) }, +#endif { XD_END } }; DEFINE_LRECORD_IMPLEMENTATION ("charset", charset, - mark_charset, print_charset, 0, 0, 0, charset_description, + mark_charset, print_charset, 0, 0, 0, + charset_description, Lisp_Charset); -/* Make a new charset. */ +/* Make a new charset. */ +/* #### SJT Should generic properties be allowed? */ static Lisp_Object -make_charset (int id, Lisp_Object name, unsigned char rep_bytes, - unsigned char type, unsigned char columns, unsigned char graphic, - Bufbyte final, unsigned char direction, Lisp_Object short_name, +make_charset (Charset_ID id, Lisp_Object name, + unsigned short chars, unsigned char dimension, + unsigned char columns, unsigned char graphic, + Bufbyte final, unsigned char direction, Lisp_Object short_name, Lisp_Object long_name, Lisp_Object doc, - Lisp_Object reg) + Lisp_Object reg, + Lisp_Object decoding_table, + Emchar ucs_min, Emchar ucs_max, + Emchar code_offset, unsigned char byte_offset) { Lisp_Object obj; Lisp_Charset *cs = alloc_lcrecord_type (Lisp_Charset, &lrecord_charset); @@ -487,9 +2066,9 @@ make_charset (int id, Lisp_Object name, unsigned char rep_bytes, CHARSET_NAME (cs) = name; CHARSET_SHORT_NAME (cs) = short_name; CHARSET_LONG_NAME (cs) = long_name; - CHARSET_REP_BYTES (cs) = rep_bytes; + CHARSET_CHARS (cs) = chars; + CHARSET_DIMENSION (cs) = dimension; CHARSET_DIRECTION (cs) = direction; - CHARSET_TYPE (cs) = type; CHARSET_COLUMNS (cs) = columns; CHARSET_GRAPHIC (cs) = graphic; CHARSET_FINAL (cs) = final; @@ -497,23 +2076,46 @@ make_charset (int id, Lisp_Object name, unsigned char rep_bytes, CHARSET_REGISTRY (cs) = reg; CHARSET_CCL_PROGRAM (cs) = Qnil; CHARSET_REVERSE_DIRECTION_CHARSET (cs) = Qnil; +#ifdef UTF2000 + CHARSET_DECODING_TABLE(cs) = Qnil; + CHARSET_ENCODING_TABLE(cs) = Qnil; + CHARSET_UCS_MIN(cs) = ucs_min; + CHARSET_UCS_MAX(cs) = ucs_max; + CHARSET_CODE_OFFSET(cs) = code_offset; + CHARSET_BYTE_OFFSET(cs) = byte_offset; +#endif - CHARSET_DIMENSION (cs) = (CHARSET_TYPE (cs) == CHARSET_TYPE_94 || - CHARSET_TYPE (cs) == CHARSET_TYPE_96) ? 1 : 2; - CHARSET_CHARS (cs) = (CHARSET_TYPE (cs) == CHARSET_TYPE_94 || - CHARSET_TYPE (cs) == CHARSET_TYPE_94X94) ? 94 : 96; +#ifndef UTF2000 + if (id == LEADING_BYTE_ASCII) + CHARSET_REP_BYTES (cs) = 1; + else if (id < 0xA0) + CHARSET_REP_BYTES (cs) = CHARSET_DIMENSION (cs) + 1; + else + CHARSET_REP_BYTES (cs) = CHARSET_DIMENSION (cs) + 2; +#endif if (final) { /* some charsets do not have final characters. This includes ASCII, Control-1, Composite, and the two faux private charsets. */ - assert (NILP (chlook->charset_by_attributes[type][final][direction])); - chlook->charset_by_attributes[type][final][direction] = obj; + unsigned char iso2022_type + = (dimension == 1 ? 0 : 2) + (chars == 94 ? 0 : 1); +#if UTF2000 + if (code_offset == 0) + { + assert (NILP (chlook->charset_by_attributes[iso2022_type][final])); + chlook->charset_by_attributes[iso2022_type][final] = obj; + } +#else + assert (NILP + (chlook->charset_by_attributes[iso2022_type][final][direction])); + chlook->charset_by_attributes[iso2022_type][final][direction] = obj; +#endif } - assert (NILP (chlook->charset_by_leading_byte[id - 128])); - chlook->charset_by_leading_byte[id - 128] = obj; + assert (NILP (chlook->charset_by_leading_byte[id - MIN_LEADING_BYTE])); + chlook->charset_by_leading_byte[id - MIN_LEADING_BYTE] = obj; /* Some charsets are "faux" and don't have names or really exist at all except in the leading-byte table. */ @@ -525,22 +2127,29 @@ make_charset (int id, Lisp_Object name, unsigned char rep_bytes, static int get_unallocated_leading_byte (int dimension) { - int lb; + Charset_ID lb; +#ifdef UTF2000 + if (chlook->next_allocated_leading_byte > MAX_LEADING_BYTE_PRIVATE) + lb = 0; + else + lb = chlook->next_allocated_leading_byte++; +#else if (dimension == 1) { - if (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 = next_allocated_1_byte_leading_byte++; + lb = chlook->next_allocated_1_byte_leading_byte++; } else { - if (next_allocated_2_byte_leading_byte > MAX_LEADING_BYTE_PRIVATE_2) + if (chlook->next_allocated_2_byte_leading_byte > MAX_LEADING_BYTE_PRIVATE_2) lb = 0; else - lb = next_allocated_2_byte_leading_byte++; + lb = chlook->next_allocated_2_byte_leading_byte++; } +#endif if (!lb) signal_simple_error @@ -550,6 +2159,321 @@ get_unallocated_leading_byte (int dimension) return lb; } +#ifdef UTF2000 +/* Number of Big5 characters which have the same code in 1st byte. */ + +#define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40) + +Emchar +make_builtin_char (Lisp_Object charset, int c1, int c2) +{ + 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)) + { + case 94: + return MIN_CHAR_94 + + (XCHARSET_FINAL (charset) - '0') * 94 + (c1 - 33); + case 96: + return MIN_CHAR_96 + + (XCHARSET_FINAL (charset) - '0') * 96 + (c1 - 32); + default: + abort (); + } + } + else + { + if (EQ (charset, Vcharset_chinese_big5)) + { + int B1 = c1, B2 = c2; + unsigned int I + = (B1 - 0xA1) * BIG5_SAME_ROW + + B2 - (B2 < 0x7F ? 0x40 : 0x62); + + if (B1 < 0xC9) + { + charset = Vcharset_chinese_big5_1; + } + else + { + charset = Vcharset_chinese_big5_2; + I -= (BIG5_SAME_ROW) * (0xC9 - 0xA1); + } + c1 = I / 94 + 33; + c2 = I % 94 + 33; + } + switch (XCHARSET_CHARS (charset)) + { + case 94: + return MIN_CHAR_94x94 + + (XCHARSET_FINAL (charset) - '0') * 94 * 94 + + (c1 - 33) * 94 + (c2 - 33); + case 96: + return MIN_CHAR_96x96 + + (XCHARSET_FINAL (charset) - '0') * 96 * 96 + + (c1 - 32) * 96 + (c2 - 32); + default: + abort (); + } + } +} + +int +range_charset_code_point (Lisp_Object charset, Emchar ch) +{ + int d; + + if ((XCHARSET_UCS_MIN (charset) <= ch) + && (ch <= XCHARSET_UCS_MAX (charset))) + { + d = ch - XCHARSET_UCS_MIN (charset) + XCHARSET_CODE_OFFSET (charset); + + if (XCHARSET_CHARS (charset) == 256) + return d; + else if (XCHARSET_DIMENSION (charset) == 1) + return d + XCHARSET_BYTE_OFFSET (charset); + else if (XCHARSET_DIMENSION (charset) == 2) + return + ((d / XCHARSET_CHARS (charset) + + XCHARSET_BYTE_OFFSET (charset)) << 8) + | (d % XCHARSET_CHARS (charset) + XCHARSET_BYTE_OFFSET (charset)); + else if (XCHARSET_DIMENSION (charset) == 3) + return + ((d / (XCHARSET_CHARS (charset) * XCHARSET_CHARS (charset)) + + XCHARSET_BYTE_OFFSET (charset)) << 16) + | ((d / XCHARSET_CHARS (charset) + % XCHARSET_CHARS (charset) + + XCHARSET_BYTE_OFFSET (charset)) << 8) + | (d % XCHARSET_CHARS (charset) + XCHARSET_BYTE_OFFSET (charset)); + else /* if (XCHARSET_DIMENSION (charset) == 4) */ + return + ((d / (XCHARSET_CHARS (charset) + * XCHARSET_CHARS (charset) * XCHARSET_CHARS (charset)) + + XCHARSET_BYTE_OFFSET (charset)) << 24) + | ((d / (XCHARSET_CHARS (charset) * XCHARSET_CHARS (charset)) + % XCHARSET_CHARS (charset) + + XCHARSET_BYTE_OFFSET (charset)) << 16) + | ((d / XCHARSET_CHARS (charset) % XCHARSET_CHARS (charset) + + XCHARSET_BYTE_OFFSET (charset)) << 8) + | (d % XCHARSET_CHARS (charset) + XCHARSET_BYTE_OFFSET (charset)); + } + else if (XCHARSET_CODE_OFFSET (charset) == 0) + { + if (XCHARSET_DIMENSION (charset) == 1) + { + if (XCHARSET_CHARS (charset) == 94) + { + if (((d = ch - (MIN_CHAR_94 + + (XCHARSET_FINAL (charset) - '0') * 94)) >= 0) + && (d < 94)) + return d + 33; + } + else if (XCHARSET_CHARS (charset) == 96) + { + if (((d = ch - (MIN_CHAR_96 + + (XCHARSET_FINAL (charset) - '0') * 96)) >= 0) + && (d < 96)) + return d + 32; + } + else + return -1; + } + else if (XCHARSET_DIMENSION (charset) == 2) + { + if (XCHARSET_CHARS (charset) == 94) + { + if (((d = ch - (MIN_CHAR_94x94 + + (XCHARSET_FINAL (charset) - '0') * 94 * 94)) + >= 0) + && (d < 94 * 94)) + return (((d / 94) + 33) << 8) | (d % 94 + 33); + } + else if (XCHARSET_CHARS (charset) == 96) + { + if (((d = ch - (MIN_CHAR_96x96 + + (XCHARSET_FINAL (charset) - '0') * 96 * 96)) + >= 0) + && (d < 96 * 96)) + return (((d / 96) + 32) << 8) | (d % 96 + 32); + } + else + return -1; + } + } + if (EQ (charset, Vcharset_mojikyo_2022_1) + && (MIN_CHAR_MOJIKYO < ch) && (ch < MIN_CHAR_MOJIKYO + 94 * 60 * 94)) + { + int m = ch - MIN_CHAR_MOJIKYO - 1; + int byte1 = m / (94 * 60) + 33; + int byte2 = (m % (94 * 60)) / 94; + int byte3 = m % 94 + 33; + + if (byte2 < 30) + byte2 += 16 + 32; + else + byte2 += 18 + 32; + return (byte1 << 16) | (byte2 << 8) | byte3; + } + return -1; +} + +int +encode_builtin_char_1 (Emchar c, Lisp_Object* charset) +{ + if (c <= MAX_CHAR_BASIC_LATIN) + { + *charset = Vcharset_ascii; + return c; + } + else if (c < 0xA0) + { + *charset = Vcharset_control_1; + return c & 0x7F; + } + else if (c <= 0xff) + { + *charset = Vcharset_latin_iso8859_1; + return c & 0x7F; + } + /* + else if ((MIN_CHAR_GREEK <= c) && (c <= MAX_CHAR_GREEK)) + { + *charset = Vcharset_greek_iso8859_7; + return c - MIN_CHAR_GREEK + 0x20; + } + else if ((MIN_CHAR_CYRILLIC <= c) && (c <= MAX_CHAR_CYRILLIC)) + { + *charset = Vcharset_cyrillic_iso8859_5; + return c - MIN_CHAR_CYRILLIC + 0x20; + } + */ + else if ((MIN_CHAR_HEBREW <= c) && (c <= MAX_CHAR_HEBREW)) + { + *charset = Vcharset_hebrew_iso8859_8; + return c - MIN_CHAR_HEBREW + 0x20; + } + else if ((MIN_CHAR_THAI <= c) && (c <= MAX_CHAR_THAI)) + { + *charset = Vcharset_thai_tis620; + return c - MIN_CHAR_THAI + 0x20; + } + /* + else if ((MIN_CHAR_HALFWIDTH_KATAKANA <= c) + && (c <= MAX_CHAR_HALFWIDTH_KATAKANA)) + { + return list2 (Vcharset_katakana_jisx0201, + make_int (c - MIN_CHAR_HALFWIDTH_KATAKANA + 33)); + } + */ + else if (c <= MAX_CHAR_BMP) + { + *charset = Vcharset_ucs_bmp; + return c; + } + else if (c < MIN_CHAR_DAIKANWA) + { + *charset = Vcharset_ucs; + return c; + } + /* + else if (c <= MAX_CHAR_DAIKANWA) + { + *charset = Vcharset_ideograph_daikanwa; + return c - MIN_CHAR_DAIKANWA; + } + */ + else if (c <= MAX_CHAR_MOJIKYO) + { + *charset = Vcharset_mojikyo; + return c - MIN_CHAR_MOJIKYO; + } + else if (c < MIN_CHAR_94) + { + *charset = Vcharset_ucs; + return c; + } + else if (c <= MAX_CHAR_94) + { + *charset = CHARSET_BY_ATTRIBUTES (94, 1, + ((c - MIN_CHAR_94) / 94) + '0', + CHARSET_LEFT_TO_RIGHT); + if (!NILP (*charset)) + return ((c - MIN_CHAR_94) % 94) + 33; + else + { + *charset = Vcharset_ucs; + return c; + } + } + else if (c <= MAX_CHAR_96) + { + *charset = CHARSET_BY_ATTRIBUTES (96, 1, + ((c - MIN_CHAR_96) / 96) + '0', + CHARSET_LEFT_TO_RIGHT); + if (!NILP (*charset)) + return ((c - MIN_CHAR_96) % 96) + 32; + else + { + *charset = Vcharset_ucs; + return c; + } + } + else if (c <= MAX_CHAR_94x94) + { + *charset + = CHARSET_BY_ATTRIBUTES (94, 2, + ((c - MIN_CHAR_94x94) / (94 * 94)) + '0', + CHARSET_LEFT_TO_RIGHT); + if (!NILP (*charset)) + return (((((c - MIN_CHAR_94x94) / 94) % 94) + 33) << 8) + | (((c - MIN_CHAR_94x94) % 94) + 33); + else + { + *charset = Vcharset_ucs; + return c; + } + } + else if (c <= MAX_CHAR_96x96) + { + *charset + = CHARSET_BY_ATTRIBUTES (96, 2, + ((c - MIN_CHAR_96x96) / (96 * 96)) + '0', + CHARSET_LEFT_TO_RIGHT); + if (!NILP (*charset)) + return ((((c - MIN_CHAR_96x96) / 96) % 96) + 32) << 8 + | (((c - MIN_CHAR_96x96) % 96) + 32); + else + { + *charset = Vcharset_ucs; + return c; + } + } + else + { + *charset = Vcharset_ucs; + return c; + } +} + +Lisp_Object Vdefault_coded_charset_priority_list; +#endif + /************************************************************************/ /* Basic charset Lisp functions */ @@ -609,7 +2533,7 @@ add_charset_to_list_mapper (Lisp_Object key, Lisp_Object value, (struct charset_list_closure*) charset_list_closure; Lisp_Object *charset_list = chcl->charset_list; - *charset_list = Fcons (XCHARSET_NAME (value), *charset_list); + *charset_list = Fcons (key /* XCHARSET_NAME (value) */, *charset_list); return 0; } @@ -632,13 +2556,14 @@ Return a list of the names of all defined charsets. } DEFUN ("charset-name", Fcharset_name, 1, 1, 0, /* -Return the name of the given charset. +Return the name of charset CHARSET. */ (charset)) { return XCHARSET_NAME (Fget_charset (charset)); } +/* #### SJT Should generic properties be allowed? */ DEFUN ("make-charset", Fmake_charset, 3, 3, 0, /* Define a new character set. This function is for use with Mule support. @@ -690,12 +2615,11 @@ character set. Recognized properties are: { int id, dimension = 1, chars = 94, graphic = 0, final = 0, columns = -1; int direction = CHARSET_LEFT_TO_RIGHT; - int type; Lisp_Object registry = Qnil; Lisp_Object charset; - Lisp_Object rest, keyword, value; Lisp_Object ccl_program = Qnil; Lisp_Object short_name = Qnil, long_name = Qnil; + int byte_offset = -1; CHECK_SYMBOL (name); if (!NILP (doc_string)) @@ -705,85 +2629,94 @@ character set. Recognized properties are: if (!NILP (charset)) signal_simple_error ("Cannot redefine existing charset", name); - EXTERNAL_PROPERTY_LIST_LOOP (rest, keyword, value, props) - { - if (EQ (keyword, Qshort_name)) - { - CHECK_STRING (value); - short_name = value; - } - - if (EQ (keyword, Qlong_name)) - { - CHECK_STRING (value); - long_name = value; - } - - else if (EQ (keyword, Qdimension)) - { - CHECK_INT (value); - dimension = XINT (value); - if (dimension < 1 || dimension > 2) - signal_simple_error ("Invalid value for 'dimension", value); - } - - else if (EQ (keyword, Qchars)) - { - CHECK_INT (value); - chars = XINT (value); - if (chars != 94 && chars != 96) - signal_simple_error ("Invalid value for 'chars", value); - } - - else if (EQ (keyword, Qcolumns)) - { - CHECK_INT (value); - columns = XINT (value); - if (columns != 1 && columns != 2) - signal_simple_error ("Invalid value for 'columns", value); - } - - else if (EQ (keyword, Qgraphic)) - { - CHECK_INT (value); - graphic = XINT (value); - if (graphic < 0 || graphic > 1) - signal_simple_error ("Invalid value for 'graphic", value); - } - - else if (EQ (keyword, Qregistry)) - { - CHECK_STRING (value); - registry = value; - } - - else if (EQ (keyword, Qdirection)) - { - if (EQ (value, Ql2r)) - direction = CHARSET_LEFT_TO_RIGHT; - else if (EQ (value, Qr2l)) - direction = CHARSET_RIGHT_TO_LEFT; - else - signal_simple_error ("Invalid value for 'direction", value); - } - - else if (EQ (keyword, Qfinal)) - { - CHECK_CHAR_COERCE_INT (value); - final = XCHAR (value); - if (final < '0' || final > '~') - signal_simple_error ("Invalid value for 'final", value); - } - - else if (EQ (keyword, Qccl_program)) - { - CHECK_VECTOR (value); - ccl_program = value; - } - - else - signal_simple_error ("Unrecognized property", keyword); - } + { + EXTERNAL_PROPERTY_LIST_LOOP_3 (keyword, value, props) + { + if (EQ (keyword, Qshort_name)) + { + CHECK_STRING (value); + short_name = value; + } + + if (EQ (keyword, Qlong_name)) + { + CHECK_STRING (value); + long_name = value; + } + + else if (EQ (keyword, Qdimension)) + { + CHECK_INT (value); + dimension = XINT (value); + if (dimension < 1 || dimension > 2) + signal_simple_error ("Invalid value for 'dimension", value); + } + + else if (EQ (keyword, Qchars)) + { + CHECK_INT (value); + chars = XINT (value); + if (chars != 94 && chars != 96) + signal_simple_error ("Invalid value for 'chars", value); + } + + else if (EQ (keyword, Qcolumns)) + { + CHECK_INT (value); + columns = XINT (value); + if (columns != 1 && columns != 2) + signal_simple_error ("Invalid value for 'columns", value); + } + + else if (EQ (keyword, Qgraphic)) + { + CHECK_INT (value); + graphic = XINT (value); +#ifdef UTF2000 + if (graphic < 0 || graphic > 2) +#else + if (graphic < 0 || graphic > 1) +#endif + signal_simple_error ("Invalid value for 'graphic", value); + } + + else if (EQ (keyword, Qregistry)) + { + CHECK_STRING (value); + registry = value; + } + + else if (EQ (keyword, Qdirection)) + { + if (EQ (value, Ql2r)) + direction = CHARSET_LEFT_TO_RIGHT; + else if (EQ (value, Qr2l)) + direction = CHARSET_RIGHT_TO_LEFT; + else + signal_simple_error ("Invalid value for 'direction", value); + } + + else if (EQ (keyword, Qfinal)) + { + CHECK_CHAR_COERCE_INT (value); + final = XCHAR (value); + if (final < '0' || final > '~') + signal_simple_error ("Invalid value for 'final", value); + } + + else if (EQ (keyword, Qccl_program)) + { + struct ccl_program test_ccl; + + if (setup_ccl_program (&test_ccl, value) < 0) + signal_simple_error ("Invalid value for 'ccl-program", value); + ccl_program = value; + } + + else + signal_simple_error ("Unrecognized property", keyword); + } + } if (!final) error ("'final must be specified"); @@ -792,13 +2725,10 @@ character set. Recognized properties are: ("Final must be in the range 0x30 - 0x5F for dimension == 2", make_char (final)); - if (dimension == 1) - type = (chars == 94) ? CHARSET_TYPE_94 : CHARSET_TYPE_96; - else - type = (chars == 94) ? CHARSET_TYPE_94X94 : CHARSET_TYPE_96X96; - - if (!NILP (CHARSET_BY_ATTRIBUTES (type, final, CHARSET_LEFT_TO_RIGHT)) || - !NILP (CHARSET_BY_ATTRIBUTES (type, final, CHARSET_RIGHT_TO_LEFT))) + if (!NILP (CHARSET_BY_ATTRIBUTES (chars, dimension, final, + CHARSET_LEFT_TO_RIGHT)) || + !NILP (CHARSET_BY_ATTRIBUTES (chars, dimension, final, + CHARSET_RIGHT_TO_LEFT))) error ("Character set already defined for this DIMENSION/CHARS/FINAL combo"); @@ -818,8 +2748,21 @@ character set. Recognized properties are: if (columns == -1) columns = dimension; - charset = make_charset (id, name, dimension + 2, type, columns, graphic, - final, direction, short_name, long_name, doc_string, registry); + + if (byte_offset < 0) + { + if (chars == 94) + byte_offset = 33; + else if (chars == 96) + byte_offset = 32; + else + byte_offset = 0; + } + + charset = make_charset (id, name, chars, dimension, columns, graphic, + final, direction, short_name, long_name, + doc_string, registry, + Qnil, 0, 0, 0, byte_offset); if (!NILP (ccl_program)) XCHARSET_CCL_PROGRAM (charset) = ccl_program; return charset; @@ -833,8 +2776,8 @@ NEW-NAME is the name of the new charset. Return the new charset. (charset, new_name)) { Lisp_Object new_charset = Qnil; - int id, dimension, columns, graphic, final; - int direction, type; + int id, chars, dimension, columns, graphic, final; + int direction; Lisp_Object registry, doc_string, short_name, long_name; Lisp_Charset *cs; @@ -849,9 +2792,9 @@ NEW-NAME is the name of the new charset. Return the new charset. cs = XCHARSET (charset); - type = CHARSET_TYPE (cs); - columns = CHARSET_COLUMNS (cs); + chars = CHARSET_CHARS (cs); dimension = CHARSET_DIMENSION (cs); + columns = CHARSET_COLUMNS (cs); id = get_unallocated_leading_byte (dimension); graphic = CHARSET_GRAPHIC (cs); @@ -864,9 +2807,19 @@ NEW-NAME is the name of the new charset. Return the new charset. long_name = CHARSET_LONG_NAME (cs); registry = CHARSET_REGISTRY (cs); - new_charset = make_charset (id, new_name, dimension + 2, type, columns, + new_charset = make_charset (id, new_name, chars, dimension, columns, graphic, final, direction, short_name, long_name, - doc_string, registry); + doc_string, registry, +#ifdef UTF2000 + CHARSET_DECODING_TABLE(cs), + CHARSET_UCS_MIN(cs), + CHARSET_UCS_MAX(cs), + CHARSET_CODE_OFFSET(cs), + CHARSET_BYTE_OFFSET(cs) +#else + Qnil, 0, 0, 0, 0 +#endif +); CHARSET_REVERSE_DIRECTION_CHARSET (cs) = new_charset; XCHARSET_REVERSE_DIRECTION_CHARSET (new_charset) = charset; @@ -874,6 +2827,16 @@ NEW-NAME is the name of the new charset. Return the new charset. return new_charset; } +DEFUN ("define-charset-alias", Fdefine_charset_alias, 2, 2, 0, /* +Define symbol ALIAS as an alias for CHARSET. +*/ + (alias, charset)) +{ + CHECK_SYMBOL (alias); + charset = Fget_charset (charset); + return Fputhash (alias, charset, Vcharset_hash_table); +} + /* #### Reverse direction charsets not yet implemented. */ #if 0 DEFUN ("charset-reverse-direction-charset", Fcharset_reverse_direction_charset, @@ -898,7 +2861,6 @@ will be returned if character sets exist for both directions). (dimension, chars, final, direction)) { int dm, ch, fi, di = -1; - int type; Lisp_Object obj = Qnil; CHECK_INT (dimension); @@ -927,19 +2889,14 @@ will be returned if character sets exist for both directions). signal_simple_error ("Final must be in the range 0x30 - 0x5F for dimension == 2", final); - if (dm == 1) - type = (ch == 94) ? CHARSET_TYPE_94 : CHARSET_TYPE_96; - else - type = (ch == 94) ? CHARSET_TYPE_94X94 : CHARSET_TYPE_96X96; - - if (di == -1) + if (di == -1) { - obj = CHARSET_BY_ATTRIBUTES (type, fi, CHARSET_LEFT_TO_RIGHT); + obj = CHARSET_BY_ATTRIBUTES (ch, dm, fi, CHARSET_LEFT_TO_RIGHT); if (NILP (obj)) - obj = CHARSET_BY_ATTRIBUTES (type, fi, CHARSET_RIGHT_TO_LEFT); + obj = CHARSET_BY_ATTRIBUTES (ch, dm, fi, CHARSET_RIGHT_TO_LEFT); } else - obj = CHARSET_BY_ATTRIBUTES (type, fi, di); + obj = CHARSET_BY_ATTRIBUTES (ch, dm, fi, di); if (CHARSETP (obj)) return XCHARSET_NAME (obj); @@ -979,7 +2936,7 @@ Return dimension of CHARSET. } DEFUN ("charset-property", Fcharset_property, 2, 2, 0, /* -Return property PROP of CHARSET. +Return property PROP of CHARSET, a charset object or symbol naming a charset. Recognized properties are those listed in `make-charset', as well as 'name and 'doc-string. */ @@ -1007,10 +2964,8 @@ Recognized properties are those listed in `make-charset', as well as if (EQ (prop, Qreverse_direction_charset)) { Lisp_Object obj = CHARSET_REVERSE_DIRECTION_CHARSET (cs); - if (NILP (obj)) - return Qnil; - else - return XCHARSET_NAME (obj); + /* #### Is this translation OK? If so, error checking sufficient? */ + return CHARSETP (obj) ? XCHARSET_NAME (obj) : obj; } signal_simple_error ("Unrecognized charset property name", prop); return Qnil; /* not reached */ @@ -1032,8 +2987,11 @@ Set the 'ccl-program property of CHARSET to CCL-PROGRAM. */ (charset, ccl_program)) { + struct ccl_program test_ccl; + charset = Fget_charset (charset); - CHECK_VECTOR (ccl_program); + if (setup_ccl_program (&test_ccl, ccl_program) < 0) + signal_simple_error ("Invalid ccl-program", ccl_program); XCHARSET_CCL_PROGRAM (charset) = ccl_program; return Qnil; } @@ -1052,7 +3010,6 @@ invalidate_charset_font_caches (Lisp_Object charset) } } -/* Japanese folks may want to (set-charset-registry 'ascii "jisx0201") */ DEFUN ("set-charset-registry", Fset_charset_registry, 2, 2, 0, /* Set the 'registry property of CHARSET to REGISTRY. */ @@ -1066,11 +3023,193 @@ Set the 'registry property of CHARSET to REGISTRY. return Qnil; } +#ifdef UTF2000 +DEFUN ("charset-mapping-table", Fcharset_mapping_table, 1, 1, 0, /* +Return mapping-table of CHARSET. +*/ + (charset)) +{ + return XCHARSET_DECODING_TABLE (Fget_charset (charset)); +} + +DEFUN ("set-charset-mapping-table", Fset_charset_mapping_table, 2, 2, 0, /* +Set mapping-table of CHARSET to TABLE. +*/ + (charset, table)) +{ + struct Lisp_Charset *cs; + size_t i; + int byte_offset; + + charset = Fget_charset (charset); + cs = XCHARSET (charset); + + if (NILP (table)) + { + if (VECTORP (CHARSET_DECODING_TABLE(cs))) + make_vector_newer (CHARSET_DECODING_TABLE(cs)); + CHARSET_DECODING_TABLE(cs) = Qnil; + return table; + } + else if (VECTORP (table)) + { + int ccs_len = CHARSET_BYTE_SIZE (cs); + int ret = decoding_table_check_elements (table, + CHARSET_DIMENSION (cs), + ccs_len); + if (ret) + { + if (ret == -1) + signal_simple_error ("Too big table", table); + else if (ret == -2) + signal_simple_error ("Invalid element is found", table); + else + signal_simple_error ("Something wrong", table); + } + CHARSET_DECODING_TABLE(cs) = Qnil; + } + else + signal_error (Qwrong_type_argument, + list2 (build_translated_string ("vector-or-nil-p"), + table)); + + byte_offset = CHARSET_BYTE_OFFSET (cs); + switch (CHARSET_DIMENSION (cs)) + { + case 1: + for (i = 0; i < XVECTOR_LENGTH (table); i++) + { + Lisp_Object c = XVECTOR_DATA(table)[i]; + + if (CHARP (c)) + put_char_ccs_code_point (c, charset, + make_int (i + byte_offset)); + } + break; + case 2: + for (i = 0; i < XVECTOR_LENGTH (table); i++) + { + Lisp_Object v = XVECTOR_DATA(table)[i]; + + if (VECTORP (v)) + { + size_t j; + + for (j = 0; j < XVECTOR_LENGTH (v); j++) + { + Lisp_Object c = XVECTOR_DATA(v)[j]; + + if (CHARP (c)) + put_char_ccs_code_point + (c, charset, + make_int ( ( (i + byte_offset) << 8 ) + | (j + byte_offset) + ) ); + } + } + else if (CHARP (v)) + put_char_ccs_code_point (v, charset, + make_int (i + byte_offset)); + } + break; + } + return table; +} +#endif + /************************************************************************/ /* Lisp primitives for working with characters */ /************************************************************************/ +#ifdef UTF2000 +DEFUN ("decode-char", Fdecode_char, 2, 2, 0, /* +Make a character from CHARSET and code-point CODE. +*/ + (charset, code)) +{ + int c; + + charset = Fget_charset (charset); + CHECK_INT (code); + c = XINT (code); + if (XCHARSET_GRAPHIC (charset) == 1) + c &= 0x7F7F7F7F; + return make_char (DECODE_CHAR (charset, c)); +} + +DEFUN ("decode-builtin-char", Fdecode_builtin_char, 2, 2, 0, /* +Make a builtin character from CHARSET and code-point CODE. +*/ + (charset, code)) +{ + int c; + int final; + + charset = Fget_charset (charset); + CHECK_INT (code); + c = XINT (code); + + if ((final = XCHARSET_FINAL (charset)) >= '0') + { + if (XCHARSET_DIMENSION (charset) == 1) + { + switch (XCHARSET_CHARS (charset)) + { + case 94: + return + make_char (MIN_CHAR_94 + (final - '0') * 94 + + ((c & 0x7F) - 33)); + case 96: + return + make_char (MIN_CHAR_96 + (final - '0') * 96 + + ((c & 0x7F) - 32)); + default: + return Fdecode_char (charset, code); + } + } + else + { + switch (XCHARSET_CHARS (charset)) + { + case 94: + return + make_char (MIN_CHAR_94x94 + + (final - '0') * 94 * 94 + + (((c >> 8) & 0x7F) - 33) * 94 + + ((c & 0x7F) - 33)); + case 96: + return + make_char (MIN_CHAR_96x96 + + (final - '0') * 96 * 96 + + (((c >> 8) & 0x7F) - 32) * 96 + + ((c & 0x7F) - 32)); + default: + return Fdecode_char (charset, code); + } + } + } + else if (XCHARSET_UCS_MAX (charset)) + { + Emchar cid + = (XCHARSET_DIMENSION (charset) == 1 + ? + c - XCHARSET_BYTE_OFFSET (charset) + : + ((c >> 8) - XCHARSET_BYTE_OFFSET (charset)) + * XCHARSET_CHARS (charset) + + (c & 0xFF) - XCHARSET_BYTE_OFFSET (charset)) + - XCHARSET_CODE_OFFSET (charset) + XCHARSET_UCS_MIN (charset); + if ((cid < XCHARSET_UCS_MIN (charset)) + || (XCHARSET_UCS_MAX (charset) < cid)) + return Fdecode_char (charset, code); + return make_char (cid); + } + else + return Fdecode_char (charset, code); +} +#endif + DEFUN ("make-char", Fmake_char, 2, 3, 0, /* Make a character from CHARSET and octets ARG1 and ARG2. ARG2 is required only for characters from two-dimensional charsets. @@ -1088,15 +3227,24 @@ character s with caron. if (EQ (charset, Vcharset_ascii)) lowlim = 0, highlim = 127; else if (EQ (charset, Vcharset_control_1)) lowlim = 0, highlim = 31; +#ifdef UTF2000 + else if (CHARSET_CHARS (cs) == 256) lowlim = 0, highlim = 255; +#endif else if (CHARSET_CHARS (cs) == 94) lowlim = 33, highlim = 126; else /* CHARSET_CHARS (cs) == 96) */ lowlim = 32, highlim = 127; CHECK_INT (arg1); /* It is useful (and safe, according to Olivier Galibert) to strip - the 8th bit off ARG1 and ARG2 becaue it allows programmers to + the 8th bit off ARG1 and ARG2 because it allows programmers to write (make-char 'latin-iso8859-2 CODE) where code is the actual Latin 2 code of the character. */ - a1 = XINT (arg1) & 0x7f; +#ifdef UTF2000 + a1 = XINT (arg1); + if (highlim < 128) + a1 &= 0x7f; +#else + a1 = XINT (arg1); +#endif if (a1 < lowlim || a1 > highlim) args_out_of_range_3 (arg1, make_int (lowlim), make_int (highlim)); @@ -1109,7 +3257,13 @@ character s with caron. } CHECK_INT (arg2); +#ifdef UTF2000 + a2 = XINT (arg2); + if (highlim < 128) + a2 &= 0x7f; +#else a2 = XINT (arg2) & 0x7f; +#endif if (a2 < lowlim || a2 > highlim) args_out_of_range_3 (arg2, make_int (lowlim), make_int (highlim)); @@ -1117,28 +3271,27 @@ character s with caron. } DEFUN ("char-charset", Fchar_charset, 1, 1, 0, /* -Return the character set of char CH. +Return the character set of CHARACTER. */ - (ch)) + (character)) { - CHECK_CHAR_COERCE_INT (ch); + CHECK_CHAR_COERCE_INT (character); - return XCHARSET_NAME (CHARSET_BY_LEADING_BYTE - (CHAR_LEADING_BYTE (XCHAR (ch)))); + return XCHARSET_NAME (CHAR_CHARSET (XCHAR (character))); } DEFUN ("char-octet", Fchar_octet, 1, 2, 0, /* -Return the octet numbered N (should be 0 or 1) of char CH. +Return the octet numbered N (should be 0 or 1) of CHARACTER. N defaults to 0 if omitted. */ - (ch, n)) + (character, n)) { Lisp_Object charset; int octet0, octet1; - CHECK_CHAR_COERCE_INT (ch); + CHECK_CHAR_COERCE_INT (character); - BREAKUP_CHAR (XCHAR (ch), charset, octet0, octet1); + BREAKUP_CHAR (XCHAR (character), charset, octet0, octet1); if (NILP (n) || EQ (n, Qzero)) return make_int (octet0); @@ -1149,7 +3302,7 @@ N defaults to 0 if omitted. } DEFUN ("split-char", Fsplit_char, 1, 1, 0, /* -Return list of charset and one or two position-codes of CHAR. +Return list of charset and one or two position-codes of CHARACTER. */ (character)) { @@ -1157,11 +3310,27 @@ Return list of charset and one or two position-codes of CHAR. struct gcpro gcpro1, gcpro2; Lisp_Object charset = Qnil; Lisp_Object rc = Qnil; +#ifdef UTF2000 + int code_point; + int dimension; +#else int c1, c2; +#endif GCPRO2 (charset, rc); CHECK_CHAR_COERCE_INT (character); +#ifdef UTF2000 + code_point = ENCODE_CHAR (XCHAR (character), charset); + dimension = XCHARSET_DIMENSION (charset); + while (dimension > 0) + { + rc = Fcons (make_int (code_point & 255), rc); + code_point >>= 8; + dimension--; + } + rc = Fcons (XCHARSET_NAME (charset), rc); +#else BREAKUP_CHAR (XCHAR (character), charset, c1, c2); if (XCHARSET_DIMENSION (Fget_charset (charset)) == 2) @@ -1172,6 +3341,7 @@ Return list of charset and one or two position-codes of CHAR. { rc = list2 (XCHARSET_NAME (charset), make_int (c1)); } +#endif UNGCPRO; return rc; @@ -1259,6 +3429,12 @@ Return a string of the characters comprising a composite character. void syms_of_mule_charset (void) { +#ifdef UTF2000 + INIT_LRECORD_IMPLEMENTATION (uint8_byte_table); + INIT_LRECORD_IMPLEMENTATION (uint16_byte_table); + INIT_LRECORD_IMPLEMENTATION (byte_table); + INIT_LRECORD_IMPLEMENTATION (char_id_table); +#endif INIT_LRECORD_IMPLEMENTATION (charset); DEFSUBR (Fcharsetp); @@ -1269,6 +3445,7 @@ syms_of_mule_charset (void) DEFSUBR (Fmake_charset); DEFSUBR (Fmake_reverse_direction_charset); /* DEFSUBR (Freverse_direction_charset); */ + DEFSUBR (Fdefine_charset_alias); DEFSUBR (Fcharset_from_attributes); DEFSUBR (Fcharset_short_name); DEFSUBR (Fcharset_long_name); @@ -1278,7 +3455,24 @@ syms_of_mule_charset (void) DEFSUBR (Fcharset_id); DEFSUBR (Fset_charset_ccl_program); DEFSUBR (Fset_charset_registry); +#ifdef UTF2000 + DEFSUBR (Fchar_attribute_list); + DEFSUBR (Ffind_char_attribute_table); + DEFSUBR (Fchar_attribute_alist); + DEFSUBR (Fget_char_attribute); + DEFSUBR (Fput_char_attribute); + DEFSUBR (Fremove_char_attribute); + DEFSUBR (Fdefine_char); + DEFSUBR (Fchar_variants); + DEFSUBR (Fget_composite_char); + DEFSUBR (Fcharset_mapping_table); + DEFSUBR (Fset_charset_mapping_table); +#endif +#ifdef UTF2000 + DEFSUBR (Fdecode_char); + DEFSUBR (Fdecode_builtin_char); +#endif DEFSUBR (Fmake_char); DEFSUBR (Fchar_charset); DEFSUBR (Fchar_octet); @@ -1320,10 +3514,66 @@ syms_of_mule_charset (void) defsymbol (&Qjapanese_jisx0208_1978, "japanese-jisx0208-1978"); defsymbol (&Qchinese_gb2312, "chinese-gb2312"); defsymbol (&Qjapanese_jisx0208, "japanese-jisx0208"); + defsymbol (&Qjapanese_jisx0208_1990, "japanese-jisx0208-1990"); defsymbol (&Qkorean_ksc5601, "korean-ksc5601"); defsymbol (&Qjapanese_jisx0212, "japanese-jisx0212"); defsymbol (&Qchinese_cns11643_1, "chinese-cns11643-1"); defsymbol (&Qchinese_cns11643_2, "chinese-cns11643-2"); +#ifdef UTF2000 + defsymbol (&Q_ucs, "->ucs"); + defsymbol (&Q_decomposition, "->decomposition"); + defsymbol (&Qcompat, "compat"); + defsymbol (&Qisolated, "isolated"); + defsymbol (&Qinitial, "initial"); + defsymbol (&Qmedial, "medial"); + defsymbol (&Qfinal, "final"); + defsymbol (&Qvertical, "vertical"); + defsymbol (&QnoBreak, "noBreak"); + defsymbol (&Qfraction, "fraction"); + defsymbol (&Qsuper, "super"); + defsymbol (&Qsub, "sub"); + defsymbol (&Qcircle, "circle"); + defsymbol (&Qsquare, "square"); + defsymbol (&Qwide, "wide"); + defsymbol (&Qnarrow, "narrow"); + defsymbol (&Qsmall, "small"); + defsymbol (&Qfont, "font"); + defsymbol (&Qucs, "ucs"); + defsymbol (&Qucs_bmp, "ucs-bmp"); + defsymbol (&Qucs_cns, "ucs-cns"); + defsymbol (&Qlatin_viscii, "latin-viscii"); + defsymbol (&Qlatin_tcvn5712, "latin-tcvn5712"); + defsymbol (&Qlatin_viscii_lower, "latin-viscii-lower"); + defsymbol (&Qlatin_viscii_upper, "latin-viscii-upper"); + defsymbol (&Qvietnamese_viscii_lower, "vietnamese-viscii-lower"); + defsymbol (&Qvietnamese_viscii_upper, "vietnamese-viscii-upper"); + defsymbol (&Qideograph_daikanwa, "ideograph-daikanwa"); + defsymbol (&Qchinese_big5, "chinese-big5"); + defsymbol (&Qmojikyo, "mojikyo"); + defsymbol (&Qmojikyo_2022_1, "mojikyo-2022-1"); + defsymbol (&Qmojikyo_pj_1, "mojikyo-pj-1"); + defsymbol (&Qmojikyo_pj_2, "mojikyo-pj-2"); + defsymbol (&Qmojikyo_pj_3, "mojikyo-pj-3"); + defsymbol (&Qmojikyo_pj_4, "mojikyo-pj-4"); + defsymbol (&Qmojikyo_pj_5, "mojikyo-pj-5"); + defsymbol (&Qmojikyo_pj_6, "mojikyo-pj-6"); + defsymbol (&Qmojikyo_pj_7, "mojikyo-pj-7"); + defsymbol (&Qmojikyo_pj_8, "mojikyo-pj-8"); + defsymbol (&Qmojikyo_pj_9, "mojikyo-pj-9"); + defsymbol (&Qmojikyo_pj_10, "mojikyo-pj-10"); + defsymbol (&Qmojikyo_pj_11, "mojikyo-pj-11"); + defsymbol (&Qmojikyo_pj_12, "mojikyo-pj-12"); + defsymbol (&Qmojikyo_pj_13, "mojikyo-pj-13"); + defsymbol (&Qmojikyo_pj_14, "mojikyo-pj-14"); + defsymbol (&Qmojikyo_pj_15, "mojikyo-pj-15"); + defsymbol (&Qmojikyo_pj_16, "mojikyo-pj-16"); + defsymbol (&Qmojikyo_pj_17, "mojikyo-pj-17"); + defsymbol (&Qmojikyo_pj_18, "mojikyo-pj-18"); + defsymbol (&Qmojikyo_pj_19, "mojikyo-pj-19"); + defsymbol (&Qmojikyo_pj_20, "mojikyo-pj-20"); + defsymbol (&Qmojikyo_pj_21, "mojikyo-pj-21"); + defsymbol (&Qethiopic_ucs, "ethiopic-ucs"); +#endif defsymbol (&Qchinese_big5_1, "chinese-big5-1"); defsymbol (&Qchinese_big5_2, "chinese-big5-2"); @@ -1333,7 +3583,10 @@ syms_of_mule_charset (void) void vars_of_mule_charset (void) { - int i, j, k; + int i, j; +#ifndef UTF2000 + int k; +#endif chlook = xnew (struct charset_lookup); dumpstruct (&chlook, &charset_lookup_description); @@ -1342,14 +3595,52 @@ vars_of_mule_charset (void) for (i = 0; i < countof (chlook->charset_by_leading_byte); i++) chlook->charset_by_leading_byte[i] = Qnil; +#ifdef UTF2000 + /* Table of charsets indexed by type/final-byte. */ + for (i = 0; i < countof (chlook->charset_by_attributes); i++) + for (j = 0; j < countof (chlook->charset_by_attributes[0]); j++) + chlook->charset_by_attributes[i][j] = Qnil; +#else /* Table of charsets indexed by type/final-byte/direction. */ for (i = 0; i < countof (chlook->charset_by_attributes); i++) for (j = 0; j < countof (chlook->charset_by_attributes[0]); j++) for (k = 0; k < countof (chlook->charset_by_attributes[0][0]); k++) chlook->charset_by_attributes[i][j][k] = Qnil; +#endif + +#ifdef UTF2000 + chlook->next_allocated_leading_byte = MIN_LEADING_BYTE_PRIVATE; +#else + chlook->next_allocated_1_byte_leading_byte = MIN_LEADING_BYTE_PRIVATE_1; + chlook->next_allocated_2_byte_leading_byte = MIN_LEADING_BYTE_PRIVATE_2; +#endif + +#ifndef UTF2000 + leading_code_private_11 = PRE_LEADING_BYTE_PRIVATE_1; + DEFVAR_INT ("leading-code-private-11", &leading_code_private_11 /* +Leading-code of private TYPE9N charset of column-width 1. +*/ ); + leading_code_private_11 = PRE_LEADING_BYTE_PRIVATE_1; +#endif + +#ifdef UTF2000 + Vutf_2000_version = build_string("0.17 (Hōryūji)"); + DEFVAR_LISP ("utf-2000-version", &Vutf_2000_version /* +Version number of UTF-2000. +*/ ); + + staticpro (&Vcharacter_composition_table); + Vcharacter_composition_table = make_char_id_table (Qnil); + + staticpro (&Vcharacter_variant_table); + Vcharacter_variant_table = make_char_id_table (Qnil); - next_allocated_1_byte_leading_byte = MIN_LEADING_BYTE_PRIVATE_1; - next_allocated_2_byte_leading_byte = MIN_LEADING_BYTE_PRIVATE_2; + Vdefault_coded_charset_priority_list = Qnil; + DEFVAR_LISP ("default-coded-charset-priority-list", + &Vdefault_coded_charset_priority_list /* +Default order of preferred coded-character-sets. +*/ ); +#endif } void @@ -1362,221 +3653,398 @@ complex_vars_of_mule_charset (void) /* Predefined character sets. We store them into variables for ease of access. */ +#ifdef UTF2000 + staticpro (&Vchar_attribute_hash_table); + Vchar_attribute_hash_table + = make_lisp_hash_table (16, HASH_TABLE_NON_WEAK, HASH_TABLE_EQ); + + staticpro (&Vcharset_ucs); + Vcharset_ucs = + make_charset (LEADING_BYTE_UCS, Qucs, 256, 4, + 1, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("UCS"), + build_string ("UCS"), + build_string ("ISO/IEC 10646"), + build_string (""), + Qnil, 0, 0xFFFFFFF, 0, 0); + staticpro (&Vcharset_ucs_bmp); + Vcharset_ucs_bmp = + make_charset (LEADING_BYTE_UCS_BMP, Qucs_bmp, 256, 2, + 1, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("BMP"), + build_string ("BMP"), + build_string ("ISO/IEC 10646 Group 0 Plane 0 (BMP)"), + build_string ("\\(ISO10646.*-1\\|UNICODE[23]?-0\\)"), + Qnil, 0, 0xFFFF, 0, 0); + staticpro (&Vcharset_ucs_cns); + Vcharset_ucs_cns = + make_charset (LEADING_BYTE_UCS_CNS, Qucs_cns, 256, 4, + 1, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("UCS for CNS"), + build_string ("UCS for CNS 11643"), + build_string ("ISO/IEC 10646 for CNS 11643"), + build_string (""), + Qnil, 0, 0xFFFFFFF, 0, 0); +#else +# define MIN_CHAR_THAI 0 +# define MAX_CHAR_THAI 0 +# define MIN_CHAR_HEBREW 0 +# define MAX_CHAR_HEBREW 0 +# define MIN_CHAR_HALFWIDTH_KATAKANA 0 +# define MAX_CHAR_HALFWIDTH_KATAKANA 0 +#endif staticpro (&Vcharset_ascii); Vcharset_ascii = - make_charset (LEADING_BYTE_ASCII, Qascii, 1, - CHARSET_TYPE_94, 1, 0, 'B', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_ASCII, Qascii, 94, 1, + 1, 0, 'B', CHARSET_LEFT_TO_RIGHT, build_string ("ASCII"), build_string ("ASCII)"), build_string ("ASCII (ISO646 IRV)"), - build_string ("\\(iso8859-[0-9]*\\|-ascii\\)")); + build_string ("\\(iso8859-[0-9]*\\|-ascii\\)"), + Qnil, 0, 0x7F, 0, 0); staticpro (&Vcharset_control_1); Vcharset_control_1 = - make_charset (LEADING_BYTE_CONTROL_1, Qcontrol_1, 2, - CHARSET_TYPE_94, 1, 1, 0, - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CONTROL_1, Qcontrol_1, 94, 1, + 1, 1, 0, CHARSET_LEFT_TO_RIGHT, build_string ("C1"), build_string ("Control characters"), build_string ("Control characters 128-191"), - build_string ("")); + build_string (""), + Qnil, 0x80, 0x9F, 0, 0); staticpro (&Vcharset_latin_iso8859_1); Vcharset_latin_iso8859_1 = - make_charset (LEADING_BYTE_LATIN_ISO8859_1, Qlatin_iso8859_1, 2, - CHARSET_TYPE_96, 1, 1, 'A', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_LATIN_ISO8859_1, Qlatin_iso8859_1, 96, 1, + 1, 1, 'A', CHARSET_LEFT_TO_RIGHT, build_string ("Latin-1"), build_string ("ISO8859-1 (Latin-1)"), build_string ("ISO8859-1 (Latin-1)"), - build_string ("iso8859-1")); + build_string ("iso8859-1"), + Qnil, 0xA0, 0xFF, 0, 32); staticpro (&Vcharset_latin_iso8859_2); Vcharset_latin_iso8859_2 = - make_charset (LEADING_BYTE_LATIN_ISO8859_2, Qlatin_iso8859_2, 2, - CHARSET_TYPE_96, 1, 1, 'B', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_LATIN_ISO8859_2, Qlatin_iso8859_2, 96, 1, + 1, 1, 'B', CHARSET_LEFT_TO_RIGHT, build_string ("Latin-2"), build_string ("ISO8859-2 (Latin-2)"), build_string ("ISO8859-2 (Latin-2)"), - build_string ("iso8859-2")); + build_string ("iso8859-2"), + Qnil, 0, 0, 0, 32); staticpro (&Vcharset_latin_iso8859_3); Vcharset_latin_iso8859_3 = - make_charset (LEADING_BYTE_LATIN_ISO8859_3, Qlatin_iso8859_3, 2, - CHARSET_TYPE_96, 1, 1, 'C', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_LATIN_ISO8859_3, Qlatin_iso8859_3, 96, 1, + 1, 1, 'C', CHARSET_LEFT_TO_RIGHT, build_string ("Latin-3"), build_string ("ISO8859-3 (Latin-3)"), build_string ("ISO8859-3 (Latin-3)"), - build_string ("iso8859-3")); + build_string ("iso8859-3"), + Qnil, 0, 0, 0, 32); staticpro (&Vcharset_latin_iso8859_4); Vcharset_latin_iso8859_4 = - make_charset (LEADING_BYTE_LATIN_ISO8859_4, Qlatin_iso8859_4, 2, - CHARSET_TYPE_96, 1, 1, 'D', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_LATIN_ISO8859_4, Qlatin_iso8859_4, 96, 1, + 1, 1, 'D', CHARSET_LEFT_TO_RIGHT, build_string ("Latin-4"), build_string ("ISO8859-4 (Latin-4)"), build_string ("ISO8859-4 (Latin-4)"), - build_string ("iso8859-4")); + build_string ("iso8859-4"), + Qnil, 0, 0, 0, 32); staticpro (&Vcharset_thai_tis620); Vcharset_thai_tis620 = - make_charset (LEADING_BYTE_THAI_TIS620, Qthai_tis620, 2, - CHARSET_TYPE_96, 1, 1, 'T', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_THAI_TIS620, Qthai_tis620, 96, 1, + 1, 1, 'T', CHARSET_LEFT_TO_RIGHT, build_string ("TIS620"), build_string ("TIS620 (Thai)"), build_string ("TIS620.2529 (Thai)"), - build_string ("tis620")); + build_string ("tis620"), + Qnil, MIN_CHAR_THAI, MAX_CHAR_THAI, 0, 32); staticpro (&Vcharset_greek_iso8859_7); Vcharset_greek_iso8859_7 = - make_charset (LEADING_BYTE_GREEK_ISO8859_7, Qgreek_iso8859_7, 2, - CHARSET_TYPE_96, 1, 1, 'F', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_GREEK_ISO8859_7, Qgreek_iso8859_7, 96, 1, + 1, 1, 'F', CHARSET_LEFT_TO_RIGHT, build_string ("ISO8859-7"), build_string ("ISO8859-7 (Greek)"), build_string ("ISO8859-7 (Greek)"), - build_string ("iso8859-7")); + build_string ("iso8859-7"), + Qnil, + 0 /* MIN_CHAR_GREEK */, + 0 /* MAX_CHAR_GREEK */, 0, 32); staticpro (&Vcharset_arabic_iso8859_6); Vcharset_arabic_iso8859_6 = - make_charset (LEADING_BYTE_ARABIC_ISO8859_6, Qarabic_iso8859_6, 2, - CHARSET_TYPE_96, 1, 1, 'G', - CHARSET_RIGHT_TO_LEFT, + make_charset (LEADING_BYTE_ARABIC_ISO8859_6, Qarabic_iso8859_6, 96, 1, + 1, 1, 'G', CHARSET_RIGHT_TO_LEFT, build_string ("ISO8859-6"), build_string ("ISO8859-6 (Arabic)"), build_string ("ISO8859-6 (Arabic)"), - build_string ("iso8859-6")); + build_string ("iso8859-6"), + Qnil, 0, 0, 0, 32); staticpro (&Vcharset_hebrew_iso8859_8); Vcharset_hebrew_iso8859_8 = - make_charset (LEADING_BYTE_HEBREW_ISO8859_8, Qhebrew_iso8859_8, 2, - CHARSET_TYPE_96, 1, 1, 'H', - CHARSET_RIGHT_TO_LEFT, + make_charset (LEADING_BYTE_HEBREW_ISO8859_8, Qhebrew_iso8859_8, 96, 1, + 1, 1, 'H', CHARSET_RIGHT_TO_LEFT, build_string ("ISO8859-8"), build_string ("ISO8859-8 (Hebrew)"), build_string ("ISO8859-8 (Hebrew)"), - build_string ("iso8859-8")); + build_string ("iso8859-8"), + Qnil, MIN_CHAR_HEBREW, MAX_CHAR_HEBREW, 0, 32); staticpro (&Vcharset_katakana_jisx0201); Vcharset_katakana_jisx0201 = - make_charset (LEADING_BYTE_KATAKANA_JISX0201, Qkatakana_jisx0201, 2, - CHARSET_TYPE_94, 1, 1, 'I', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_KATAKANA_JISX0201, Qkatakana_jisx0201, 94, 1, + 1, 1, 'I', CHARSET_LEFT_TO_RIGHT, build_string ("JISX0201 Kana"), build_string ("JISX0201.1976 (Japanese Kana)"), build_string ("JISX0201.1976 Japanese Kana"), - build_string ("jisx0201.1976")); + build_string ("jisx0201\\.1976"), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_latin_jisx0201); Vcharset_latin_jisx0201 = - make_charset (LEADING_BYTE_LATIN_JISX0201, Qlatin_jisx0201, 2, - CHARSET_TYPE_94, 1, 0, 'J', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_LATIN_JISX0201, Qlatin_jisx0201, 94, 1, + 1, 0, 'J', CHARSET_LEFT_TO_RIGHT, build_string ("JISX0201 Roman"), build_string ("JISX0201.1976 (Japanese Roman)"), build_string ("JISX0201.1976 Japanese Roman"), - build_string ("jisx0201.1976")); + build_string ("jisx0201\\.1976"), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_cyrillic_iso8859_5); Vcharset_cyrillic_iso8859_5 = - make_charset (LEADING_BYTE_CYRILLIC_ISO8859_5, Qcyrillic_iso8859_5, 2, - CHARSET_TYPE_96, 1, 1, 'L', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CYRILLIC_ISO8859_5, Qcyrillic_iso8859_5, 96, 1, + 1, 1, 'L', CHARSET_LEFT_TO_RIGHT, build_string ("ISO8859-5"), build_string ("ISO8859-5 (Cyrillic)"), build_string ("ISO8859-5 (Cyrillic)"), - build_string ("iso8859-5")); + build_string ("iso8859-5"), + Qnil, + 0 /* MIN_CHAR_CYRILLIC */, + 0 /* MAX_CHAR_CYRILLIC */, 0, 32); staticpro (&Vcharset_latin_iso8859_9); Vcharset_latin_iso8859_9 = - make_charset (LEADING_BYTE_LATIN_ISO8859_9, Qlatin_iso8859_9, 2, - CHARSET_TYPE_96, 1, 1, 'M', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_LATIN_ISO8859_9, Qlatin_iso8859_9, 96, 1, + 1, 1, 'M', CHARSET_LEFT_TO_RIGHT, build_string ("Latin-5"), build_string ("ISO8859-9 (Latin-5)"), build_string ("ISO8859-9 (Latin-5)"), - build_string ("iso8859-9")); + build_string ("iso8859-9"), + Qnil, 0, 0, 0, 32); staticpro (&Vcharset_japanese_jisx0208_1978); Vcharset_japanese_jisx0208_1978 = - make_charset (LEADING_BYTE_JAPANESE_JISX0208_1978, Qjapanese_jisx0208_1978, 3, - CHARSET_TYPE_94X94, 2, 0, '@', - CHARSET_LEFT_TO_RIGHT, - build_string ("JISX0208.1978"), - build_string ("JISX0208.1978 (Japanese)"), + make_charset (LEADING_BYTE_JAPANESE_JISX0208_1978, + Qjapanese_jisx0208_1978, 94, 2, + 2, 0, '@', CHARSET_LEFT_TO_RIGHT, + build_string ("JIS X0208:1978"), + build_string ("JIS X0208:1978 (Japanese)"), build_string - ("JISX0208.1978 Japanese Kanji (so called \"old JIS\")"), - build_string ("\\(jisx0208\\|jisc6226\\)\\.1978")); + ("JIS X0208:1978 Japanese Kanji (so called \"old JIS\")"), + build_string ("\\(jisx0208\\|jisc6226\\)\\.1978"), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_chinese_gb2312); Vcharset_chinese_gb2312 = - make_charset (LEADING_BYTE_CHINESE_GB2312, Qchinese_gb2312, 3, - CHARSET_TYPE_94X94, 2, 0, 'A', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CHINESE_GB2312, Qchinese_gb2312, 94, 2, + 2, 0, 'A', CHARSET_LEFT_TO_RIGHT, build_string ("GB2312"), build_string ("GB2312)"), build_string ("GB2312 Chinese simplified"), - build_string ("gb2312")); + build_string ("gb2312"), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_japanese_jisx0208); Vcharset_japanese_jisx0208 = - make_charset (LEADING_BYTE_JAPANESE_JISX0208, Qjapanese_jisx0208, 3, - CHARSET_TYPE_94X94, 2, 0, 'B', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_JAPANESE_JISX0208, Qjapanese_jisx0208, 94, 2, + 2, 0, 'B', CHARSET_LEFT_TO_RIGHT, build_string ("JISX0208"), - build_string ("JISX0208.1983/1990 (Japanese)"), - build_string ("JISX0208.1983/1990 Japanese Kanji"), - build_string ("jisx0208.19\\(83\\|90\\)")); + build_string ("JIS X0208:1983 (Japanese)"), + build_string ("JIS X0208:1983 Japanese Kanji"), + build_string ("jisx0208\\.1983"), + Qnil, 0, 0, 0, 33); +#ifdef UTF2000 + staticpro (&Vcharset_japanese_jisx0208_1990); + Vcharset_japanese_jisx0208_1990 = + make_charset (LEADING_BYTE_JAPANESE_JISX0208_1990, + Qjapanese_jisx0208_1990, 94, 2, + 2, 0, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("JISX0208-1990"), + build_string ("JIS X0208:1990 (Japanese)"), + build_string ("JIS X0208:1990 Japanese Kanji"), + build_string ("jisx0208\\.1990"), + Qnil, + MIN_CHAR_JIS_X0208_1990, + MAX_CHAR_JIS_X0208_1990, 0, 33); +#endif staticpro (&Vcharset_korean_ksc5601); Vcharset_korean_ksc5601 = - make_charset (LEADING_BYTE_KOREAN_KSC5601, Qkorean_ksc5601, 3, - CHARSET_TYPE_94X94, 2, 0, 'C', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_KOREAN_KSC5601, Qkorean_ksc5601, 94, 2, + 2, 0, 'C', CHARSET_LEFT_TO_RIGHT, build_string ("KSC5601"), build_string ("KSC5601 (Korean"), build_string ("KSC5601 Korean Hangul and Hanja"), - build_string ("ksc5601")); + build_string ("ksc5601"), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_japanese_jisx0212); Vcharset_japanese_jisx0212 = - make_charset (LEADING_BYTE_JAPANESE_JISX0212, Qjapanese_jisx0212, 3, - CHARSET_TYPE_94X94, 2, 0, 'D', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_JAPANESE_JISX0212, Qjapanese_jisx0212, 94, 2, + 2, 0, 'D', CHARSET_LEFT_TO_RIGHT, build_string ("JISX0212"), build_string ("JISX0212 (Japanese)"), build_string ("JISX0212 Japanese Supplement"), - build_string ("jisx0212")); + build_string ("jisx0212"), + Qnil, 0, 0, 0, 33); #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, 3, - CHARSET_TYPE_94X94, 2, 0, 'G', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CHINESE_CNS11643_1, Qchinese_cns11643_1, 94, 2, + 2, 0, 'G', CHARSET_LEFT_TO_RIGHT, build_string ("CNS11643-1"), build_string ("CNS11643-1 (Chinese traditional)"), build_string ("CNS 11643 Plane 1 Chinese traditional"), - build_string (CHINESE_CNS_PLANE_RE("1"))); + build_string (CHINESE_CNS_PLANE_RE("1")), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_chinese_cns11643_2); Vcharset_chinese_cns11643_2 = - make_charset (LEADING_BYTE_CHINESE_CNS11643_2, Qchinese_cns11643_2, 3, - CHARSET_TYPE_94X94, 2, 0, 'H', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CHINESE_CNS11643_2, Qchinese_cns11643_2, 94, 2, + 2, 0, 'H', CHARSET_LEFT_TO_RIGHT, build_string ("CNS11643-2"), build_string ("CNS11643-2 (Chinese traditional)"), build_string ("CNS 11643 Plane 2 Chinese traditional"), - build_string (CHINESE_CNS_PLANE_RE("2"))); + build_string (CHINESE_CNS_PLANE_RE("2")), + Qnil, 0, 0, 0, 33); +#ifdef UTF2000 + staticpro (&Vcharset_latin_tcvn5712); + Vcharset_latin_tcvn5712 = + make_charset (LEADING_BYTE_LATIN_TCVN5712, Qlatin_tcvn5712, 96, 1, + 1, 1, 'Z', CHARSET_LEFT_TO_RIGHT, + build_string ("TCVN 5712"), + build_string ("TCVN 5712 (VSCII-2)"), + build_string ("Vietnamese TCVN 5712:1983 (VSCII-2)"), + build_string ("tcvn5712-1"), + Qnil, 0, 0, 0, 32); + staticpro (&Vcharset_latin_viscii_lower); + Vcharset_latin_viscii_lower = + make_charset (LEADING_BYTE_LATIN_VISCII_LOWER, Qlatin_viscii_lower, 96, 1, + 1, 1, '1', CHARSET_LEFT_TO_RIGHT, + build_string ("VISCII lower"), + build_string ("VISCII lower (Vietnamese)"), + build_string ("VISCII lower (Vietnamese)"), + build_string ("MULEVISCII-LOWER"), + Qnil, 0, 0, 0, 32); + staticpro (&Vcharset_latin_viscii_upper); + Vcharset_latin_viscii_upper = + make_charset (LEADING_BYTE_LATIN_VISCII_UPPER, Qlatin_viscii_upper, 96, 1, + 1, 1, '2', CHARSET_LEFT_TO_RIGHT, + build_string ("VISCII upper"), + build_string ("VISCII upper (Vietnamese)"), + build_string ("VISCII upper (Vietnamese)"), + build_string ("MULEVISCII-UPPER"), + Qnil, 0, 0, 0, 32); + staticpro (&Vcharset_latin_viscii); + Vcharset_latin_viscii = + make_charset (LEADING_BYTE_LATIN_VISCII, Qlatin_viscii, 256, 1, + 1, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("VISCII"), + build_string ("VISCII 1.1 (Vietnamese)"), + build_string ("VISCII 1.1 (Vietnamese)"), + build_string ("VISCII1\\.1"), + Qnil, 0, 0, 0, 0); + staticpro (&Vcharset_chinese_big5); + Vcharset_chinese_big5 = + make_charset (LEADING_BYTE_CHINESE_BIG5, Qchinese_big5, 256, 2, + 2, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("Big5"), + build_string ("Big5"), + build_string ("Big5 Chinese traditional"), + build_string ("big5"), + Qnil, 0, 0, 0, 0); + staticpro (&Vcharset_ideograph_daikanwa); + Vcharset_ideograph_daikanwa = + make_charset (LEADING_BYTE_DAIKANWA, Qideograph_daikanwa, 256, 2, + 2, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("Daikanwa"), + build_string ("Morohashi's Daikanwa"), + build_string ("Daikanwa dictionary by MOROHASHI Tetsuji"), + build_string ("Daikanwa"), + Qnil, MIN_CHAR_DAIKANWA, MAX_CHAR_DAIKANWA, 0, 0); + staticpro (&Vcharset_mojikyo); + Vcharset_mojikyo = + make_charset (LEADING_BYTE_MOJIKYO, Qmojikyo, 256, 3, + 2, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("Mojikyo"), + build_string ("Mojikyo"), + build_string ("Konjaku-Mojikyo"), + build_string (""), + Qnil, MIN_CHAR_MOJIKYO, MAX_CHAR_MOJIKYO, 0, 0); + staticpro (&Vcharset_mojikyo_2022_1); + Vcharset_mojikyo_2022_1 = + make_charset (LEADING_BYTE_MOJIKYO_2022_1, Qmojikyo_2022_1, 94, 3, + 2, 2, ':', CHARSET_LEFT_TO_RIGHT, + build_string ("Mojikyo-2022-1"), + build_string ("Mojikyo ISO-2022 Part 1"), + build_string ("Konjaku-Mojikyo for ISO/IEC 2022 Part 1"), + build_string (""), + Qnil, 0, 0, 0, 33); + +#define DEF_MOJIKYO_PJ(n) \ + staticpro (&Vcharset_mojikyo_pj_##n); \ + Vcharset_mojikyo_pj_##n = \ + make_charset (LEADING_BYTE_MOJIKYO_PJ_##n, Qmojikyo_pj_##n, 94, 2, \ + 2, 0, 0, CHARSET_LEFT_TO_RIGHT, \ + build_string ("Mojikyo-PJ-"#n), \ + build_string ("Mojikyo (pseudo JIS encoding) part "#n), \ + build_string \ + ("Konjaku-Mojikyo (pseudo JIS encoding) part "#n), \ + build_string \ + ("\\(MojikyoPJ-"#n "\\|jisx0208\\.Mojikyo-"#n "\\)$"), \ + Qnil, 0, 0, 0, 33); + + DEF_MOJIKYO_PJ (1); + DEF_MOJIKYO_PJ (2); + DEF_MOJIKYO_PJ (3); + DEF_MOJIKYO_PJ (4); + DEF_MOJIKYO_PJ (5); + DEF_MOJIKYO_PJ (6); + DEF_MOJIKYO_PJ (7); + DEF_MOJIKYO_PJ (8); + DEF_MOJIKYO_PJ (9); + DEF_MOJIKYO_PJ (10); + DEF_MOJIKYO_PJ (11); + DEF_MOJIKYO_PJ (12); + DEF_MOJIKYO_PJ (13); + DEF_MOJIKYO_PJ (14); + DEF_MOJIKYO_PJ (15); + DEF_MOJIKYO_PJ (16); + DEF_MOJIKYO_PJ (17); + DEF_MOJIKYO_PJ (18); + DEF_MOJIKYO_PJ (19); + DEF_MOJIKYO_PJ (20); + DEF_MOJIKYO_PJ (21); + + staticpro (&Vcharset_ethiopic_ucs); + Vcharset_ethiopic_ucs = + make_charset (LEADING_BYTE_ETHIOPIC_UCS, Qethiopic_ucs, 256, 2, + 2, 2, 0, CHARSET_LEFT_TO_RIGHT, + build_string ("Ethiopic (UCS)"), + build_string ("Ethiopic (UCS)"), + build_string ("Ethiopic of UCS"), + build_string ("Ethiopic-Unicode"), + Qnil, 0x1200, 0x137F, 0x1200, 0); +#endif staticpro (&Vcharset_chinese_big5_1); Vcharset_chinese_big5_1 = - make_charset (LEADING_BYTE_CHINESE_BIG5_1, Qchinese_big5_1, 3, - CHARSET_TYPE_94X94, 2, 0, '0', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CHINESE_BIG5_1, Qchinese_big5_1, 94, 2, + 2, 0, '0', CHARSET_LEFT_TO_RIGHT, build_string ("Big5"), build_string ("Big5 (Level-1)"), build_string ("Big5 Level-1 Chinese traditional"), - build_string ("big5")); + build_string ("big5"), + Qnil, 0, 0, 0, 33); staticpro (&Vcharset_chinese_big5_2); Vcharset_chinese_big5_2 = - make_charset (LEADING_BYTE_CHINESE_BIG5_2, Qchinese_big5_2, 3, - CHARSET_TYPE_94X94, 2, 0, '1', - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_CHINESE_BIG5_2, Qchinese_big5_2, 94, 2, + 2, 0, '1', CHARSET_LEFT_TO_RIGHT, build_string ("Big5"), build_string ("Big5 (Level-2)"), build_string ("Big5 Level-2 Chinese traditional"), - build_string ("big5")); - + build_string ("big5"), + Qnil, 0, 0, 0, 33); #ifdef ENABLE_COMPOSITE_CHARS /* #### For simplicity, we put composite chars into a 96x96 charset. @@ -1584,9 +4052,8 @@ complex_vars_of_mule_charset (void) room, esp. as we don't yet recycle numbers. */ staticpro (&Vcharset_composite); Vcharset_composite = - make_charset (LEADING_BYTE_COMPOSITE, Qcomposite, 3, - CHARSET_TYPE_96X96, 2, 0, 0, - CHARSET_LEFT_TO_RIGHT, + make_charset (LEADING_BYTE_COMPOSITE, Qcomposite, 96, 2, + 2, 0, 0, CHARSET_LEFT_TO_RIGHT, build_string ("Composite"), build_string ("Composite characters"), build_string ("Composite characters"),