Lisp_Object Qshort_name, Qlong_name;
#ifdef UTF2000
Lisp_Object Qmin_code, Qmax_code, Qcode_offset;
-Lisp_Object Qmother, Qconversion, Q94x60, Q94x94x60;
+Lisp_Object Qmother, Qconversion, Q94x60, Q94x94x60, Qbig5_1, Qbig5_2;
#endif
Lisp_Object Qascii,
#define BIG5_SAME_ROW (0xFF - 0xA1 + 0x7F - 0x40)
Emchar
+decode_defined_char (Lisp_Object ccs, int code_point)
+{
+ int dim = XCHARSET_DIMENSION (ccs);
+ Lisp_Object decoding_table = XCHARSET_DECODING_TABLE (ccs);
+ Emchar char_id = -1;
+ Lisp_Object mother;
+
+ while (dim > 0)
+ {
+ dim--;
+ decoding_table
+ = get_ccs_octet_table (decoding_table, ccs,
+ (code_point >> (dim * 8)) & 255);
+ }
+ if (CHARP (decoding_table))
+ return XCHAR (decoding_table);
+#ifdef HAVE_DATABASE
+ if (EQ (decoding_table, Qunloaded) ||
+ EQ (decoding_table, Qunbound) ||
+ NILP (decoding_table) )
+ {
+ char_id = load_char_decoding_entry_maybe (ccs, code_point);
+ }
+#endif
+ if (char_id >= 0)
+ return char_id;
+ else if ( CHARSETP (mother = XCHARSET_MOTHER (ccs)) )
+ {
+ if ( XCHARSET_CONVERSION (ccs) == CONVERSION_IDENTICAL )
+ {
+ if ( EQ (mother, Vcharset_ucs) )
+ return DECODE_CHAR (mother, code_point);
+ else
+ return decode_defined_char (mother, code_point);
+ }
+ else if ( XCHARSET_CONVERSION (ccs) == CONVERSION_BIG5_1 )
+ {
+ unsigned int I
+ = (((code_point >> 8) & 0x7F) - 33) * 94
+ + (( code_point & 0x7F) - 33);
+ unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1;
+ unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40);
+
+ b2 += b2 < 0x3F ? 0x40 : 0x62;
+ return decode_defined_char (mother, (b1 << 8) | b2);
+ }
+ else if ( XCHARSET_CONVERSION (ccs) == CONVERSION_BIG5_2 )
+ {
+ unsigned int I
+ = (((code_point >> 8) & 0x7F) - 33) * 94
+ + (( code_point & 0x7F) - 33)
+ + BIG5_SAME_ROW * (0xC9 - 0xA1);
+ unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1;
+ unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40);
+
+ b2 += b2 < 0x3F ? 0x40 : 0x62;
+ return decode_defined_char (mother, (b1 << 8) | b2);
+ }
+ }
+ return -1;
+}
+
+Emchar
decode_builtin_char (Lisp_Object charset, int code_point)
{
Lisp_Object mother = XCHARSET_MOTHER (charset);
int final;
- if ( CHARSETP (mother) )
+ if ( CHARSETP (mother) && (XCHARSET_MAX_CODE (charset) > 0) )
{
int code = code_point;
+ (row - (18 + 32)) * 94
+ cell - 33;
}
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_BIG5_1 )
+ {
+ unsigned int I
+ = (((code_point >> 8) & 0x7F) - 33) * 94
+ + (( code_point & 0x7F) - 33);
+ unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1;
+ unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40);
+
+ b2 += b2 < 0x3F ? 0x40 : 0x62;
+ code = (b1 << 8) | b2;
+ }
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_BIG5_2 )
+ {
+ unsigned int I
+ = (((code_point >> 8) & 0x7F) - 33) * 94
+ + (( code_point & 0x7F) - 33)
+ + BIG5_SAME_ROW * (0xC9 - 0xA1);
+ unsigned char b1 = I / (0xFF - 0xA1 + 0x7F - 0x40) + 0xA1;
+ unsigned char b2 = I % (0xFF - 0xA1 + 0x7F - 0x40);
+
+ b2 += b2 < 0x3F ? 0x40 : 0x62;
+ code = (b1 << 8) | b2;
+ }
return
decode_builtin_char (mother, code + XCHARSET_CODE_OFFSET(charset));
}
code = charset_code_point (mother, ch);
else
code = ch;
- if ( (min <= code) && (code <= max) )
+ if ( ((max == 0) && CHARSETP (mother)) ||
+ ((min <= code) && (code <= max)) )
{
int d = code - XCHARSET_CODE_OFFSET (charset);
- if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 )
+ if ( XCHARSET_CONVERSION (charset) == CONVERSION_IDENTICAL )
+ return d;
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94 )
+ return d + 33;
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96 )
+ return d + 32;
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x60 )
{
int row = d / 94;
int cell = d % 94 + 33;
row += 18 + 32;
return (row << 8) | cell;
}
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_BIG5_1 )
+ {
+ int B1 = d >> 8, B2 = d & 0xFF;
+ unsigned int I
+ = (B1 - 0xA1) * BIG5_SAME_ROW + B2
+ - (B2 < 0x7F ? 0x40 : 0x62);
+
+ if (B1 < 0xC9)
+ {
+ return ((I / 94 + 33) << 8) | (I % 94 + 33);
+ }
+ }
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_BIG5_2 )
+ {
+ int B1 = d >> 8, B2 = d & 0xFF;
+ unsigned int I
+ = (B1 - 0xA1) * BIG5_SAME_ROW + B2
+ - (B2 < 0x7F ? 0x40 : 0x62);
+
+ if (B1 >= 0xC9)
+ {
+ I -= (BIG5_SAME_ROW) * (0xC9 - 0xA1);
+ return ((I / 94 + 33) << 8) | (I % 94 + 33);
+ }
+ }
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94 )
+ return ((d / 94 + 33) << 8) | (d % 94 + 33);
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96 )
+ return ((d / 96 + 32) << 8) | (d % 96 + 32);
else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x60 )
{
int plane = d / (94 * 60) + 33;
row += 18 + 32;
return (plane << 16) | (row << 8) | cell;
}
- else if (XCHARSET_CHARS (charset) == 94)
- {
- if (XCHARSET_DIMENSION (charset) == 1)
- return d + 33;
- else if (XCHARSET_DIMENSION (charset) == 2)
- return ((d / 94 + 33) << 8) | (d % 94 + 33);
- else if (XCHARSET_DIMENSION (charset) == 3)
- return
- ( (d / (94 * 94) + 33) << 16)
- | ((d / 94 % 94 + 33) << 8)
- | (d % 94 + 33);
- else /* if (XCHARSET_DIMENSION (charset) == 4) */
- return
- ( (d / (94 * 94 * 94) + 33) << 24)
- | ((d / (94 * 94) % 94 + 33) << 16)
- | ((d / 94 % 94 + 33) << 8)
- | (d % 94 + 33);
- }
- else if (XCHARSET_CHARS (charset) == 96)
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x94 )
+ return
+ ( (d / (94 * 94) + 33) << 16)
+ | ((d / 94 % 94 + 33) << 8)
+ | (d % 94 + 33);
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96x96 )
+ return
+ ( (d / (96 * 96) + 32) << 16)
+ | ((d / 96 % 96 + 32) << 8)
+ | (d % 96 + 32);
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_94x94x94x94 )
+ return
+ ( (d / (94 * 94 * 94) + 33) << 24)
+ | ((d / (94 * 94) % 94 + 33) << 16)
+ | ((d / 94 % 94 + 33) << 8)
+ | (d % 94 + 33);
+ else if ( XCHARSET_CONVERSION (charset) == CONVERSION_96x96x96x96 )
+ return
+ ( (d / (96 * 96 * 96) + 32) << 24)
+ | ((d / (96 * 96) % 96 + 32) << 16)
+ | ((d / 96 % 96 + 32) << 8)
+ | (d % 96 + 32);
+ else
{
- if (XCHARSET_DIMENSION (charset) == 1)
- return d + 32;
- else if (XCHARSET_DIMENSION (charset) == 2)
- return ((d / 96 + 32) << 8) | (d % 96 + 32);
- else if (XCHARSET_DIMENSION (charset) == 3)
- return
- ( (d / (96 * 96) + 32) << 16)
- | ((d / 96 % 96 + 32) << 8)
- | (d % 96 + 32);
- else /* if (XCHARSET_DIMENSION (charset) == 4) */
- return
- ( (d / (96 * 96 * 96) + 32) << 24)
- | ((d / (96 * 96) % 96 + 32) << 16)
- | ((d / 96 % 96 + 32) << 8)
- | (d % 96 + 32);
+ printf ("Unknown CCS-conversion %d is specified!",
+ XCHARSET_CONVERSION (charset));
+ exit (-1);
}
- else
- return code - XCHARSET_CODE_OFFSET (charset);
}
else if ( (XCHARSET_CODE_OFFSET (charset) == 0) ||
(XCHARSET_CODE_OFFSET (charset)
'code-offset [UTF-2000 only] Offset for a code-point of a base
coded-charset.
'conversion [UTF-2000 only] Conversion for a code-point of a base
- coded-charset (94x60 or 94x94x60).
+ coded-charset (94x60, 94x94x60, big5-1 or big5-2).
*/
(name, doc_string, props))
{
conversion = CONVERSION_94x60;
else if (EQ (value, Q94x94x60))
conversion = CONVERSION_94x94x60;
+ else if (EQ (value, Qbig5_1))
+ conversion = CONVERSION_BIG5_1;
+ else if (EQ (value, Qbig5_2))
+ conversion = CONVERSION_BIG5_2;
else
signal_simple_error ("Unrecognized conversion", value);
}
db_file = char_attribute_system_db_file (CHARSET_NAME (cs),
Qsystem_char_id, 1);
- db = Fopen_database (db_file, Qnil, Qnil, Qnil, Qnil);
-
+ db = Fopen_database (db_file, Qnil, Qnil, build_string ("w+"), Qnil);
+
byte_min = CHARSET_BYTE_OFFSET (cs);
byte_max = byte_min + CHARSET_BYTE_SIZE (cs);
switch (CHARSET_DIMENSION (cs))
= char_attribute_system_db_file (XCHARSET_NAME(ccs), Qsystem_char_id,
0);
- db = Fopen_database (db_file, Qnil, Qnil, Qnil, Qnil);
+ db = Fopen_database (db_file, Qnil, Qnil, build_string ("r"), Qnil);
if (!NILP (db))
{
Lisp_Object ret
if (NILP (defined_only))
c = DECODE_CHAR (charset, c);
else
- c = DECODE_DEFINED_CHAR (charset, c);
+ c = decode_defined_char (charset, c);
return c >= 0 ? make_char (c) : Qnil;
}
defsymbol (&Qconversion, "conversion");
defsymbol (&Q94x60, "94x60");
defsymbol (&Q94x94x60, "94x94x60");
+ defsymbol (&Qbig5_1, "big5-1");
+ defsymbol (&Qbig5_2, "big5-2");
#endif
defsymbol (&Ql2r, "l2r");
build_string ("UCS"),
build_string ("ISO/IEC 10646"),
build_string (""),
- Qnil, 0, 0xFFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0x7FFFFFFF, 0, 0, Qnil, CONVERSION_IDENTICAL);
staticpro (&Vcharset_ucs_bmp);
Vcharset_ucs_bmp =
make_charset (LEADING_BYTE_UCS_BMP, Qucs_bmp, 256, 2,
build_string ("UCS for CNS 11643"),
build_string ("ISO/IEC 10646 for CNS 11643"),
build_string (""),
- Qnil, 0, 0, 0, 0,
- Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
staticpro (&Vcharset_ucs_jis);
Vcharset_ucs_jis =
make_charset (LEADING_BYTE_UCS_JIS, Qucs_jis, 256, 3,
2, 2, 0, CHARSET_LEFT_TO_RIGHT,
build_string ("UCS for JIS"),
build_string ("UCS for JIS X 0208, 0212 and 0213"),
- build_string ("ISO/IEC 10646 for JIS X 0208, 0212 and 0213"),
+ build_string
+ ("ISO/IEC 10646 for JIS X 0208, 0212 and 0213"),
build_string (""),
- Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
staticpro (&Vcharset_ucs_ks);
Vcharset_ucs_ks =
make_charset (LEADING_BYTE_UCS_KS, Qucs_ks, 256, 3,
build_string ("UCS for CCS defined by KS"),
build_string ("ISO/IEC 10646 for Korean Standards"),
build_string (""),
- Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
staticpro (&Vcharset_ucs_big5);
Vcharset_ucs_big5 =
make_charset (LEADING_BYTE_UCS_BIG5, Qucs_big5, 256, 3,
build_string ("UCS for Big5"),
build_string ("ISO/IEC 10646 for Big5"),
build_string (""),
- Qnil, 0, 0, 0, 0, Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0, 0, 0, Vcharset_ucs, CONVERSION_IDENTICAL);
#else
# define MIN_CHAR_THAI 0
# define MAX_CHAR_THAI 0
build_string ("TIS620.2529 (Thai)"),
build_string ("tis620"),
Qnil, MIN_CHAR_THAI, MAX_CHAR_THAI,
- MIN_CHAR_THAI, 32, Qnil, CONVERSION_IDENTICAL);
+ MIN_CHAR_THAI, 32, Qnil, CONVERSION_96);
staticpro (&Vcharset_greek_iso8859_7);
Vcharset_greek_iso8859_7 =
make_charset (LEADING_BYTE_GREEK_ISO8859_7, Qgreek_iso8859_7, 96, 1,
Qnil,
MIN_CHAR_JIS_X0208_1990,
MAX_CHAR_JIS_X0208_1990, MIN_CHAR_JIS_X0208_1990, 33,
- Qnil, CONVERSION_IDENTICAL);
+ Qnil, CONVERSION_94x94);
#endif
staticpro (&Vcharset_korean_ksc5601);
Vcharset_korean_ksc5601 =
build_string
("Big5 Level-1 Chinese traditional"),
build_string ("big5"),
- Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0, 0, 33, /* Qnil, CONVERSION_IDENTICAL */
+ Vcharset_chinese_big5, CONVERSION_BIG5_1);
staticpro (&Vcharset_chinese_big5_2);
Vcharset_chinese_big5_2 =
make_charset (LEADING_BYTE_CHINESE_BIG5_2, Qchinese_big5_2, 94, 2,
build_string
("Big5 Level-2 Chinese traditional"),
build_string ("big5"),
- Qnil, 0, 0, 0, 33, Qnil, CONVERSION_IDENTICAL);
+ Qnil, 0, 0, 0, 33, /* Qnil, CONVERSION_IDENTICAL */
+ Vcharset_chinese_big5, CONVERSION_BIG5_2);
#ifdef ENABLE_COMPOSITE_CHARS
/* #### For simplicity, we put composite chars into a 96x96 charset.