#ifdef emacs
#include <config.h>
#endif
#include <stdio.h>
#ifdef emacs
#include <sys/types.h>
#include "lisp.h"
#include "buffer.h"
#include "charset.h"
#include "composite.h"
#include "coding.h"
#include "disptab.h"
#else
#include "mulelib.h"
#endif
Lisp_Object Qcharset, Qascii, Qeight_bit_control, Qeight_bit_graphic;
Lisp_Object Qunknown;
EMACS_INT leading_code_private_11;
EMACS_INT leading_code_private_12;
EMACS_INT leading_code_private_21;
EMACS_INT leading_code_private_22;
int charset_latin_iso8859_1;
int charset_jisx0208_1978;
int charset_jisx0208;
int charset_katakana_jisx0201;
int charset_latin_jisx0201;
int charset_big5_1;
int charset_big5_2;
int charset_mule_unicode_0100_24ff;
int charset_mule_unicode_2500_33ff;
int charset_mule_unicode_e000_ffff;
Lisp_Object Qcharset_table;
Lisp_Object Vcharset_table;
Lisp_Object Vcharset_symbol_table;
Lisp_Object Vcharset_list;
Lisp_Object Vtranslation_table_vector;
Lisp_Object Vauto_fill_chars;
Lisp_Object Qauto_fill_chars;
int bytes_by_char_head[256];
int width_by_char_head[256];
int iso_charset_table[2][2][128];
unsigned char *_fetch_multibyte_char_p;
int _fetch_multibyte_char_len;
EMACS_INT nonascii_insert_offset;
Lisp_Object Vnonascii_translation_table;
Lisp_Object Vgeneric_character_list;
void
invalid_character (c)
int c;
{
error ("Invalid character: %d, #o%o, #x%x", c, c, c);
}
#define SPLIT_MULTIBYTE_SEQ(str, length, bytes, charset, c1, c2) \
do { \
(c1) = *(str); \
(bytes) = BYTES_BY_CHAR_HEAD (c1); \
if ((bytes) == 1) \
(charset) = ASCII_BYTE_P (c1) ? CHARSET_ASCII : CHARSET_8_BIT_GRAPHIC; \
else if ((bytes) == 2) \
{ \
if ((c1) == LEADING_CODE_8_BIT_CONTROL) \
(charset) = CHARSET_8_BIT_CONTROL, (c1) = (str)[1] - 0x20; \
else \
(charset) = (c1), (c1) = (str)[1] & 0x7F; \
} \
else if ((bytes) == 3) \
{ \
if ((c1) < LEADING_CODE_PRIVATE_11) \
(charset) = (c1), (c1) = (str)[1] & 0x7F, (c2) = (str)[2] & 0x7F; \
else \
(charset) = (str)[1], (c1) = (str)[2] & 0x7F; \
} \
else \
(charset) = (str)[1], (c1) = (str)[2] & 0x7F, (c2) = (str)[3] & 0x7F; \
} while (0)
#define CHAR_COMPONENTS_VALID_P(charset, c1, c2) \
((charset) == CHARSET_ASCII \
? ((c1) >= 0 && (c1) <= 0x7F) \
: ((charset) == CHARSET_8_BIT_CONTROL \
? ((c1) >= 0x80 && (c1) <= 0x9F) \
: ((charset) == CHARSET_8_BIT_GRAPHIC \
? ((c1) >= 0x80 && (c1) <= 0xFF) \
: (CHARSET_DIMENSION (charset) == 1 \
? ((c1) >= 0x20 && (c1) <= 0x7F) \
: ((c1) >= 0x20 && (c1) <= 0x7F \
&& (c2) >= 0x20 && (c2) <= 0x7F)))))
int
char_to_string_1 (c, str)
int c;
unsigned char *str;
{
unsigned char *p = str;
if (c & CHAR_MODIFIER_MASK)
{
if (! SINGLE_BYTE_CHAR_P ((c & ~CHAR_MODIFIER_MASK)))
return -1;
if (c & CHAR_META)
{
c = (c & ~CHAR_META) | 0x80;
}
if (c & CHAR_SHIFT)
{
if ((c & 0377) >= 'A' && (c & 0377) <= 'Z')
c &= ~CHAR_SHIFT;
else if ((c & 0377) >= 'a' && (c & 0377) <= 'z')
c = (c & ~CHAR_SHIFT) - ('a' - 'A');
}
if (c & CHAR_CTL)
{
if (c == (CHAR_CTL | ' '))
c = 0;
else if (c == (CHAR_CTL | '?'))
c = 127;
else if ((c & 0137) >= 0101 && (c & 0137) <= 0132)
c &= (037 | (~0177 & ~CHAR_CTL));
else if ((c & 0177) >= 0100 && (c & 0177) <= 0137)
c &= (037 | (~0177 & ~CHAR_CTL));
}
c &= ~CHAR_MODIFIER_MASK;
}
if (SINGLE_BYTE_CHAR_P (c))
{
if (ASCII_BYTE_P (c) || c >= 0xA0)
*p++ = c;
else
{
*p++ = LEADING_CODE_8_BIT_CONTROL;
*p++ = c + 0x20;
}
}
else if (CHAR_VALID_P (c, 0))
{
int charset, c1, c2;
SPLIT_CHAR (c, charset, c1, c2);
if (charset >= LEADING_CODE_EXT_11)
*p++ = (charset < LEADING_CODE_EXT_12
? LEADING_CODE_PRIVATE_11
: (charset < LEADING_CODE_EXT_21
? LEADING_CODE_PRIVATE_12
: (charset < LEADING_CODE_EXT_22
? LEADING_CODE_PRIVATE_21
: LEADING_CODE_PRIVATE_22)));
*p++ = charset;
if ((c1 > 0 && c1 < 32) || (c2 > 0 && c2 < 32))
return -1;
if (c1)
{
*p++ = c1 | 0x80;
if (c2 > 0)
*p++ = c2 | 0x80;
}
}
else
return -1;
return (p - str);
}
int
char_to_string (c, str)
int c;
unsigned char *str;
{
int len;
len = char_to_string_1 (c, str);
if (len == -1)
invalid_character (c);
return len;
}
int
string_to_char (str, len, actual_len)
const unsigned char *str;
int len, *actual_len;
{
int c, bytes, charset, c1, c2;
SPLIT_MULTIBYTE_SEQ (str, len, bytes, charset, c1, c2);
c = MAKE_CHAR (charset, c1, c2);
if (actual_len)
*actual_len = bytes;
return c;
}
int
multibyte_form_length (str, len)
const unsigned char *str;
int len;
{
int bytes;
PARSE_MULTIBYTE_SEQ (str, len, bytes);
return bytes;
}
int
split_string (str, len, charset, c1, c2)
const unsigned char *str;
unsigned char *c1, *c2;
int len, *charset;
{
register int bytes, cs, code1, code2 = -1;
SPLIT_MULTIBYTE_SEQ (str, len, bytes, cs, code1, code2);
if (cs == CHARSET_ASCII)
return -1;
*charset = cs;
*c1 = code1;
*c2 = code2;
return 0;
}
int
char_printable_p (c)
int c;
{
int charset, c1, c2;
if (ASCII_BYTE_P (c))
return 1;
else if (SINGLE_BYTE_CHAR_P (c))
return 0;
else if (c >= MAX_CHAR)
return 0;
SPLIT_CHAR (c, charset, c1, c2);
if (! CHARSET_DEFINED_P (charset))
return 0;
if (CHARSET_CHARS (charset) == 94
? c1 <= 32 || c1 >= 127
: c1 < 32)
return 0;
if (CHARSET_DIMENSION (charset) == 2
&& (CHARSET_CHARS (charset) == 94
? c2 <= 32 || c2 >= 127
: c2 < 32))
return 0;
return 1;
}
int
translate_char (table, c, charset, c1, c2)
Lisp_Object table;
int c, charset, c1, c2;
{
Lisp_Object ch;
int alt_charset, alt_c1, alt_c2, dimension;
if (c < 0) c = MAKE_CHAR (charset, (c1 & 0x7F) , (c2 & 0x7F));
if (!CHAR_TABLE_P (table)
|| (ch = Faref (table, make_number (c)), !NATNUMP (ch)))
return c;
SPLIT_CHAR (XFASTINT (ch), alt_charset, alt_c1, alt_c2);
dimension = CHARSET_DIMENSION (alt_charset);
if ((dimension == 1 && alt_c1 > 0) || (dimension == 2 && alt_c2 > 0))
return XFASTINT (ch);
if (charset < 0)
SPLIT_CHAR (c, charset, c1, c2);
if (dimension != CHARSET_DIMENSION (charset))
return c;
return MAKE_CHAR (alt_charset, c1, c2);
}
int
unibyte_char_to_multibyte (c)
int c;
{
if (c < 0400 && c >= 0200)
{
int c_save = c;
if (! NILP (Vnonascii_translation_table))
{
c = XINT (Faref (Vnonascii_translation_table, make_number (c)));
if (c >= 0400 && ! char_valid_p (c, 0))
c = c_save + DEFAULT_NONASCII_INSERT_OFFSET;
}
else if (c >= 0240 && nonascii_insert_offset > 0)
{
c += nonascii_insert_offset;
if (c < 0400 || ! char_valid_p (c, 0))
c = c_save + DEFAULT_NONASCII_INSERT_OFFSET;
}
else if (c >= 0240)
c = c_save + DEFAULT_NONASCII_INSERT_OFFSET;
}
return c;
}
int
multibyte_char_to_unibyte (c, rev_tbl)
int c;
Lisp_Object rev_tbl;
{
if (!SINGLE_BYTE_CHAR_P (c))
{
int c_save = c;
if (! CHAR_TABLE_P (rev_tbl)
&& CHAR_TABLE_P (Vnonascii_translation_table))
rev_tbl = Fchar_table_extra_slot (Vnonascii_translation_table,
make_number (0));
if (CHAR_TABLE_P (rev_tbl))
{
Lisp_Object temp;
temp = Faref (rev_tbl, make_number (c));
if (INTEGERP (temp))
c = XINT (temp);
if (c >= 256)
c = (c_save & 0177) + 0200;
}
else
{
if (nonascii_insert_offset > 0)
c -= nonascii_insert_offset;
if (c < 128 || c >= 256)
c = (c_save & 0177) + 0200;
}
}
return c;
}
void
update_charset_table (charset_id, dimension, chars, width, direction,
iso_final_char, iso_graphic_plane,
short_name, long_name, description)
Lisp_Object charset_id, dimension, chars, width, direction;
Lisp_Object iso_final_char, iso_graphic_plane;
Lisp_Object short_name, long_name, description;
{
int charset = XINT (charset_id);
int bytes;
unsigned char leading_code_base, leading_code_ext;
if (NILP (CHARSET_TABLE_ENTRY (charset)))
CHARSET_TABLE_ENTRY (charset)
= Fmake_vector (make_number (CHARSET_MAX_IDX), Qnil);
if (NILP (long_name))
long_name = short_name;
if (NILP (description))
description = long_name;
bytes = XINT (dimension);
if (charset < MIN_CHARSET_PRIVATE_DIMENSION1)
{
if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC)
bytes += 1;
leading_code_base = charset;
leading_code_ext = 0;
}
else
{
bytes += 2;
leading_code_base
= (charset < LEADING_CODE_EXT_12
? LEADING_CODE_PRIVATE_11
: (charset < LEADING_CODE_EXT_21
? LEADING_CODE_PRIVATE_12
: (charset < LEADING_CODE_EXT_22
? LEADING_CODE_PRIVATE_21
: LEADING_CODE_PRIVATE_22)));
leading_code_ext = charset;
if (BYTES_BY_CHAR_HEAD (leading_code_base) != bytes)
error ("Invalid dimension for the charset-ID %d", charset);
}
CHARSET_TABLE_INFO (charset, CHARSET_ID_IDX) = charset_id;
CHARSET_TABLE_INFO (charset, CHARSET_BYTES_IDX) = make_number (bytes);
CHARSET_TABLE_INFO (charset, CHARSET_DIMENSION_IDX) = dimension;
CHARSET_TABLE_INFO (charset, CHARSET_CHARS_IDX) = chars;
CHARSET_TABLE_INFO (charset, CHARSET_WIDTH_IDX) = width;
CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX) = direction;
CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_BASE_IDX)
= make_number (leading_code_base);
CHARSET_TABLE_INFO (charset, CHARSET_LEADING_CODE_EXT_IDX)
= make_number (leading_code_ext);
CHARSET_TABLE_INFO (charset, CHARSET_ISO_FINAL_CHAR_IDX) = iso_final_char;
CHARSET_TABLE_INFO (charset, CHARSET_ISO_GRAPHIC_PLANE_IDX)
= iso_graphic_plane;
CHARSET_TABLE_INFO (charset, CHARSET_SHORT_NAME_IDX) = short_name;
CHARSET_TABLE_INFO (charset, CHARSET_LONG_NAME_IDX) = long_name;
CHARSET_TABLE_INFO (charset, CHARSET_DESCRIPTION_IDX) = description;
CHARSET_TABLE_INFO (charset, CHARSET_PLIST_IDX) = Qnil;
{
int i;
for (i = 0; i <= MAX_CHARSET; i++)
if (!NILP (CHARSET_TABLE_ENTRY (i)))
{
if (CHARSET_DIMENSION (i) == XINT (dimension)
&& CHARSET_CHARS (i) == XINT (chars)
&& CHARSET_ISO_FINAL_CHAR (i) == XINT (iso_final_char)
&& CHARSET_DIRECTION (i) != XINT (direction))
{
CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX)
= make_number (i);
CHARSET_TABLE_INFO (i, CHARSET_REVERSE_CHARSET_IDX) = charset_id;
break;
}
}
if (i > MAX_CHARSET)
CHARSET_TABLE_INFO (charset, CHARSET_REVERSE_CHARSET_IDX)
= make_number (-1);
}
if (charset != CHARSET_ASCII && charset != CHARSET_8_BIT_GRAPHIC
&& charset < MIN_CHARSET_PRIVATE_DIMENSION1)
{
bytes_by_char_head[leading_code_base] = bytes;
width_by_char_head[leading_code_base] = XINT (width);
emacs_code_class[charset] = (bytes == 2
? EMACS_leading_code_2
: (bytes == 3
? EMACS_leading_code_3
: EMACS_leading_code_4));
}
if (XINT (iso_final_char) >= 0
&& ISO_CHARSET_TABLE (dimension, chars, iso_final_char) < 0)
ISO_CHARSET_TABLE (dimension, chars, iso_final_char) = charset;
}
#ifdef emacs
int
get_charset_id (charset_symbol)
Lisp_Object charset_symbol;
{
Lisp_Object val;
int charset;
if (SYMBOLP (charset_symbol)
&& VECTORP (val = Fget (charset_symbol, Qcharset))
&& CHARSET_VALID_P (charset =
XINT (XVECTOR (val)->contents[CHARSET_ID_IDX])))
return charset;
else
return -1;
}
Lisp_Object
get_new_private_charset_id (dimension, width)
int dimension, width;
{
int charset, from, to;
if (dimension == 1)
{
from = LEADING_CODE_EXT_11;
to = LEADING_CODE_EXT_21;
}
else
{
from = LEADING_CODE_EXT_21;
to = LEADING_CODE_EXT_MAX + 1;
}
for (charset = from; charset < to; charset++)
if (!CHARSET_DEFINED_P (charset)) break;
return make_number (charset < to ? charset : 0);
}
DEFUN ("define-charset", Fdefine_charset, Sdefine_charset, 3, 3, 0,
doc: )
(charset_id, charset_symbol, info_vector)
Lisp_Object charset_id, charset_symbol, info_vector;
{
Lisp_Object *vec;
if (!NILP (charset_id))
CHECK_NUMBER (charset_id);
CHECK_SYMBOL (charset_symbol);
CHECK_VECTOR (info_vector);
if (! NILP (charset_id))
{
if (! CHARSET_VALID_P (XINT (charset_id)))
error ("Invalid CHARSET: %d", XINT (charset_id));
else if (CHARSET_DEFINED_P (XINT (charset_id)))
error ("Already defined charset: %d", XINT (charset_id));
}
vec = XVECTOR (info_vector)->contents;
if (XVECTOR (info_vector)->size != 9
|| !INTEGERP (vec[0]) || !(XINT (vec[0]) == 1 || XINT (vec[0]) == 2)
|| !INTEGERP (vec[1]) || !(XINT (vec[1]) == 94 || XINT (vec[1]) == 96)
|| !INTEGERP (vec[2]) || !(XINT (vec[2]) == 1 || XINT (vec[2]) == 2)
|| !INTEGERP (vec[3]) || !(XINT (vec[3]) == 0 || XINT (vec[3]) == 1)
|| !INTEGERP (vec[4])
|| !(XINT (vec[4]) == -1 || (XINT (vec[4]) >= '0' && XINT (vec[4]) <= '~'))
|| !INTEGERP (vec[5])
|| !(XINT (vec[5]) == -1 || XINT (vec[5]) == 0 || XINT (vec[5]) == 1)
|| !STRINGP (vec[6])
|| !STRINGP (vec[7])
|| !STRINGP (vec[8]))
error ("Invalid info-vector argument for defining charset %s",
SDATA (SYMBOL_NAME (charset_symbol)));
if (NILP (charset_id))
{
charset_id = get_new_private_charset_id (XINT (vec[0]), XINT (vec[2]));
if (XINT (charset_id) == 0)
error ("There's no room for a new private charset %s",
SDATA (SYMBOL_NAME (charset_symbol)));
}
update_charset_table (charset_id, vec[0], vec[1], vec[2], vec[3],
vec[4], vec[5], vec[6], vec[7], vec[8]);
Fput (charset_symbol, Qcharset, CHARSET_TABLE_ENTRY (XINT (charset_id)));
CHARSET_SYMBOL (XINT (charset_id)) = charset_symbol;
Vcharset_list = Fcons (charset_symbol, Vcharset_list);
Fupdate_coding_systems_internal ();
return Qnil;
}
DEFUN ("generic-character-list", Fgeneric_character_list,
Sgeneric_character_list, 0, 0, 0,
doc: )
()
{
return Vgeneric_character_list;
}
DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char,
Sget_unused_iso_final_char, 2, 2, 0,
doc: )
(dimension, chars)
Lisp_Object dimension, chars;
{
int final_char;
CHECK_NUMBER (dimension);
CHECK_NUMBER (chars);
if (XINT (dimension) != 1 && XINT (dimension) != 2)
error ("Invalid charset dimension %d, it should be 1 or 2",
XINT (dimension));
if (XINT (chars) != 94 && XINT (chars) != 96)
error ("Invalid charset chars %d, it should be 94 or 96",
XINT (chars));
for (final_char = '0'; final_char <= '?'; final_char++)
{
if (ISO_CHARSET_TABLE (dimension, chars, make_number (final_char)) < 0)
break;
}
return (final_char <= '?' ? make_number (final_char) : Qnil);
}
DEFUN ("declare-equiv-charset", Fdeclare_equiv_charset, Sdeclare_equiv_charset,
4, 4, 0,
doc: )
(dimension, chars, final_char, charset)
Lisp_Object dimension, chars, final_char, charset;
{
int charset_id;
CHECK_NUMBER (dimension);
CHECK_NUMBER (chars);
CHECK_NUMBER (final_char);
CHECK_SYMBOL (charset);
if (XINT (dimension) != 1 && XINT (dimension) != 2)
error ("Invalid DIMENSION %d, it should be 1 or 2", XINT (dimension));
if (XINT (chars) != 94 && XINT (chars) != 96)
error ("Invalid CHARS %d, it should be 94 or 96", XINT (chars));
if (XINT (final_char) < '0' || XFASTINT (final_char) > '~')
error ("Invalid FINAL-CHAR %c, it should be `0'..`~'", XINT (chars));
if ((charset_id = get_charset_id (charset)) < 0)
error ("Invalid charset %s", SDATA (SYMBOL_NAME (charset)));
ISO_CHARSET_TABLE (dimension, chars, final_char) = charset_id;
return Qnil;
}
int
find_charset_in_text (ptr, nchars, nbytes, charsets, table)
const unsigned char *ptr;
int nchars, nbytes, *charsets;
Lisp_Object table;
{
if (nchars == nbytes)
{
if (charsets && nbytes > 0)
{
const unsigned char *endp = ptr + nbytes;
int maskbits = 0;
while (ptr < endp && maskbits != 7)
{
maskbits |= (*ptr < 0x80 ? 1 : *ptr < 0xA0 ? 2 : 4);
ptr++;
}
if (maskbits & 1)
charsets[CHARSET_ASCII] = 1;
if (maskbits & 2)
charsets[CHARSET_8_BIT_CONTROL] = 1;
if (maskbits & 4)
charsets[CHARSET_8_BIT_GRAPHIC] = 1;
}
return 0;
}
else
{
int return_val = 1;
int bytes, charset, c1, c2;
if (! CHAR_TABLE_P (table))
table = Qnil;
while (nchars-- > 0)
{
SPLIT_MULTIBYTE_SEQ (ptr, len, bytes, charset, c1, c2);
ptr += bytes;
if (!CHARSET_DEFINED_P (charset))
charset = 1;
else if (! NILP (table))
{
int c = translate_char (table, -1, charset, c1, c2);
if (c >= 0)
charset = CHAR_CHARSET (c);
}
if (return_val == 1
&& charset != CHARSET_ASCII
&& charset != CHARSET_8_BIT_CONTROL
&& charset != CHARSET_8_BIT_GRAPHIC
&& charset != charset_latin_iso8859_1)
return_val = 2;
if (charsets)
charsets[charset] = 1;
else if (return_val == 2)
break;
}
return return_val;
}
}
DEFUN ("find-charset-region", Ffind_charset_region, Sfind_charset_region,
2, 3, 0,
doc: )
(beg, end, table)
Lisp_Object beg, end, table;
{
int charsets[MAX_CHARSET + 1];
int from, from_byte, to, stop, stop_byte, i;
Lisp_Object val;
validate_region (&beg, &end);
from = XFASTINT (beg);
stop = to = XFASTINT (end);
if (from < GPT && GPT < to)
{
stop = GPT;
stop_byte = GPT_BYTE;
}
else
stop_byte = CHAR_TO_BYTE (stop);
from_byte = CHAR_TO_BYTE (from);
bzero (charsets, (MAX_CHARSET + 1) * sizeof (int));
while (1)
{
find_charset_in_text (BYTE_POS_ADDR (from_byte), stop - from,
stop_byte - from_byte, charsets, table);
if (stop < to)
{
from = stop, from_byte = stop_byte;
stop = to, stop_byte = CHAR_TO_BYTE (stop);
}
else
break;
}
val = Qnil;
if (charsets[1])
val = Fcons (Qunknown, val);
for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--)
if (charsets[i])
val = Fcons (CHARSET_SYMBOL (i), val);
if (charsets[0])
val = Fcons (Qascii, val);
return val;
}
DEFUN ("find-charset-string", Ffind_charset_string, Sfind_charset_string,
1, 2, 0,
doc: )
(str, table)
Lisp_Object str, table;
{
int charsets[MAX_CHARSET + 1];
int i;
Lisp_Object val;
CHECK_STRING (str);
bzero (charsets, (MAX_CHARSET + 1) * sizeof (int));
find_charset_in_text (SDATA (str), SCHARS (str),
SBYTES (str), charsets, table);
val = Qnil;
if (charsets[1])
val = Fcons (Qunknown, val);
for (i = MAX_CHARSET; i >= MIN_CHARSET_OFFICIAL_DIMENSION1; i--)
if (charsets[i])
val = Fcons (CHARSET_SYMBOL (i), val);
if (charsets[0])
val = Fcons (Qascii, val);
return val;
}
DEFUN ("make-char-internal", Fmake_char_internal, Smake_char_internal, 1, 3, 0,
doc: )
(charset, code1, code2)
Lisp_Object charset, code1, code2;
{
int charset_id, c1, c2;
CHECK_NUMBER (charset);
charset_id = XINT (charset);
if (!CHARSET_DEFINED_P (charset_id))
error ("Invalid charset ID: %d", XINT (charset));
if (NILP (code1))
c1 = 0;
else
{
CHECK_NUMBER (code1);
c1 = XINT (code1);
}
if (NILP (code2))
c2 = 0;
else
{
CHECK_NUMBER (code2);
c2 = XINT (code2);
}
if (charset_id == CHARSET_ASCII)
{
if (c1 < 0 || c1 > 0x7F)
goto invalid_code_posints;
return make_number (c1);
}
else if (charset_id == CHARSET_8_BIT_CONTROL)
{
if (NILP (code1))
c1 = 0x80;
else if (c1 < 0x80 || c1 > 0x9F)
goto invalid_code_posints;
return make_number (c1);
}
else if (charset_id == CHARSET_8_BIT_GRAPHIC)
{
if (NILP (code1))
c1 = 0xA0;
else if (c1 < 0xA0 || c1 > 0xFF)
goto invalid_code_posints;
return make_number (c1);
}
else if (c1 < 0 || c1 > 0xFF || c2 < 0 || c2 > 0xFF)
goto invalid_code_posints;
c1 &= 0x7F;
c2 &= 0x7F;
if (c1 == 0
? c2 != 0
: (c2 == 0
? !CHAR_COMPONENTS_VALID_P (charset_id, c1, 0x20)
: !CHAR_COMPONENTS_VALID_P (charset_id, c1, c2)))
goto invalid_code_posints;
return make_number (MAKE_CHAR (charset_id, c1, c2));
invalid_code_posints:
error ("Invalid code points for charset ID %d: %d %d", charset_id, c1, c2);
}
DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
int c, charset, c1, c2;
CHECK_NUMBER (ch);
c = XFASTINT (ch);
if (!CHAR_VALID_P (c, 1))
return Fcons (Qunknown, Fcons (ch, Qnil));
SPLIT_CHAR (XFASTINT (ch), charset, c1, c2);
return (c2 >= 0
? Fcons (CHARSET_SYMBOL (charset),
Fcons (make_number (c1), Fcons (make_number (c2), Qnil)))
: Fcons (CHARSET_SYMBOL (charset), Fcons (make_number (c1), Qnil)));
}
DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
CHECK_NUMBER (ch);
return CHARSET_SYMBOL (CHAR_CHARSET (XINT (ch)));
}
DEFUN ("charset-after", Fcharset_after, Scharset_after, 0, 1, 0,
doc: )
(pos)
Lisp_Object pos;
{
Lisp_Object ch;
int charset;
ch = Fchar_after (pos);
if (! INTEGERP (ch))
return ch;
charset = CHAR_CHARSET (XINT (ch));
return CHARSET_SYMBOL (charset);
}
DEFUN ("iso-charset", Fiso_charset, Siso_charset, 3, 3, 0,
doc: )
(dimension, chars, final_char)
Lisp_Object dimension, chars, final_char;
{
int charset;
CHECK_NUMBER (dimension);
CHECK_NUMBER (chars);
CHECK_NUMBER (final_char);
if ((charset = ISO_CHARSET_TABLE (dimension, chars, final_char)) < 0)
return Qnil;
return CHARSET_SYMBOL (charset);
}
int
char_valid_p (c, genericp)
int c, genericp;
{
int charset, c1, c2;
if (c < 0 || c >= MAX_CHAR)
return 0;
if (SINGLE_BYTE_CHAR_P (c))
return 1;
SPLIT_CHAR (c, charset, c1, c2);
if (genericp)
{
if (c1)
{
if (c2 <= 0) c2 = 0x20;
}
else
{
if (c2 <= 0) c1 = c2 = 0x20;
}
}
return (CHARSET_DEFINED_P (charset)
&& CHAR_COMPONENTS_VALID_P (charset, c1, c2));
}
DEFUN ("char-valid-p", Fchar_valid_p, Schar_valid_p, 1, 2, 0,
doc: )
(object, genericp)
Lisp_Object object, genericp;
{
if (! NATNUMP (object))
return Qnil;
return (CHAR_VALID_P (XFASTINT (object), !NILP (genericp)) ? Qt : Qnil);
}
DEFUN ("unibyte-char-to-multibyte", Funibyte_char_to_multibyte,
Sunibyte_char_to_multibyte, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
int c;
CHECK_NUMBER (ch);
c = XINT (ch);
if (c < 0 || c >= 0400)
error ("Invalid unibyte character: %d", c);
c = unibyte_char_to_multibyte (c);
if (c < 0)
error ("Can't convert to multibyte character: %d", XINT (ch));
return make_number (c);
}
DEFUN ("multibyte-char-to-unibyte", Fmultibyte_char_to_unibyte,
Smultibyte_char_to_unibyte, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
int c;
CHECK_NUMBER (ch);
c = XINT (ch);
if (! CHAR_VALID_P (c, 0))
error ("Invalid multibyte character: %d", c);
c = multibyte_char_to_unibyte (c, Qnil);
if (c < 0)
error ("Can't convert to unibyte character: %d", XINT (ch));
return make_number (c);
}
DEFUN ("char-bytes", Fchar_bytes, Schar_bytes, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
CHECK_NUMBER (ch);
return make_number (1);
}
int
char_bytes (c)
int c;
{
int charset;
if (ASCII_BYTE_P (c) || (c & ~((1 << CHARACTERBITS) -1)))
return 1;
if (SINGLE_BYTE_CHAR_P (c) && c >= 0xA0)
return 1;
charset = CHAR_CHARSET (c);
return (CHARSET_DEFINED_P (charset) ? CHARSET_BYTES (charset) : 1);
}
#define ONE_BYTE_CHAR_WIDTH(c) \
(c < 0x20 \
? (c == '\t' \
? XFASTINT (current_buffer->tab_width) \
: (c == '\n' ? 0 : (NILP (current_buffer->ctl_arrow) ? 4 : 2))) \
: (c < 0x7f \
? 1 \
: (c == 0x7F \
? (NILP (current_buffer->ctl_arrow) ? 4 : 2) \
: ((! NILP (current_buffer->enable_multibyte_characters) \
&& BASE_LEADING_CODE_P (c)) \
? WIDTH_BY_CHAR_HEAD (c) \
: 4))))
DEFUN ("char-width", Fchar_width, Schar_width, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
Lisp_Object val, disp;
int c;
struct Lisp_Char_Table *dp = buffer_display_table ();
CHECK_NUMBER (ch);
c = XINT (ch);
disp = dp ? DISP_CHAR_VECTOR (dp, c) : Qnil;
if (VECTORP (disp))
XSETINT (val, XVECTOR (disp)->size);
else if (SINGLE_BYTE_CHAR_P (c))
XSETINT (val, ONE_BYTE_CHAR_WIDTH (c));
else
{
int charset = CHAR_CHARSET (c);
XSETFASTINT (val, CHARSET_WIDTH (charset));
}
return val;
}
int
strwidth (str, len)
unsigned char *str;
int len;
{
return c_string_width (str, len, -1, NULL, NULL);
}
int
c_string_width (str, len, precision, nchars, nbytes)
const unsigned char *str;
int len, precision, *nchars, *nbytes;
{
int i = 0, i_byte = 0;
int width = 0;
int chars;
struct Lisp_Char_Table *dp = buffer_display_table ();
while (i_byte < len)
{
int bytes, thiswidth;
Lisp_Object val;
if (dp)
{
int c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes);
chars = 1;
val = DISP_CHAR_VECTOR (dp, c);
if (VECTORP (val))
thiswidth = XVECTOR (val)->size;
else
thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]);
}
else
{
chars = 1;
PARSE_MULTIBYTE_SEQ (str + i_byte, len - i_byte, bytes);
thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]);
}
if (precision > 0
&& (width + thiswidth > precision))
{
*nchars = i;
*nbytes = i_byte;
return width;
}
i++;
i_byte += bytes;
width += thiswidth;
}
if (precision > 0)
{
*nchars = i;
*nbytes = i_byte;
}
return width;
}
int
lisp_string_width (string, precision, nchars, nbytes)
Lisp_Object string;
int precision, *nchars, *nbytes;
{
int len = SCHARS (string);
int len_byte = SBYTES (string);
int multibyte = len < len_byte;
const unsigned char *str = SDATA (string);
int i = 0, i_byte = 0;
int width = 0;
struct Lisp_Char_Table *dp = buffer_display_table ();
while (i < len)
{
int chars, bytes, thiswidth;
Lisp_Object val;
int cmp_id;
int ignore, end;
if (find_composition (i, -1, &ignore, &end, &val, string)
&& ((cmp_id = get_composition_id (i, i_byte, end - i, val, string))
>= 0))
{
thiswidth = composition_table[cmp_id]->width;
chars = end - i;
bytes = string_char_to_byte (string, end) - i_byte;
}
else if (dp)
{
int c;
if (multibyte)
c = STRING_CHAR_AND_LENGTH (str + i_byte, len - i_byte, bytes);
else
c = str[i_byte], bytes = 1;
chars = 1;
val = DISP_CHAR_VECTOR (dp, c);
if (VECTORP (val))
thiswidth = XVECTOR (val)->size;
else
thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]);
}
else
{
chars = 1;
if (multibyte)
PARSE_MULTIBYTE_SEQ (str + i_byte, len_byte - i_byte, bytes);
else
bytes = 1;
thiswidth = ONE_BYTE_CHAR_WIDTH (str[i_byte]);
}
if (precision > 0
&& (width + thiswidth > precision))
{
*nchars = i;
*nbytes = i_byte;
return width;
}
i += chars;
i_byte += bytes;
width += thiswidth;
}
if (precision > 0)
{
*nchars = i;
*nbytes = i_byte;
}
return width;
}
DEFUN ("string-width", Fstring_width, Sstring_width, 1, 1, 0,
doc: )
(string)
Lisp_Object string;
{
Lisp_Object val;
CHECK_STRING (string);
XSETFASTINT (val, lisp_string_width (string, -1, NULL, NULL));
return val;
}
DEFUN ("char-direction", Fchar_direction, Schar_direction, 1, 1, 0,
doc: )
(ch)
Lisp_Object ch;
{
int charset;
CHECK_NUMBER (ch);
charset = CHAR_CHARSET (XFASTINT (ch));
if (!CHARSET_DEFINED_P (charset))
invalid_character (XINT (ch));
return CHARSET_TABLE_INFO (charset, CHARSET_DIRECTION_IDX);
}
int
chars_in_text (ptr, nbytes)
const unsigned char *ptr;
int nbytes;
{
if (current_buffer == 0
|| NILP (current_buffer->enable_multibyte_characters))
return nbytes;
return multibyte_chars_in_text (ptr, nbytes);
}
int
multibyte_chars_in_text (ptr, nbytes)
const unsigned char *ptr;
int nbytes;
{
const unsigned char *endp;
int chars, bytes;
endp = ptr + nbytes;
chars = 0;
while (ptr < endp)
{
PARSE_MULTIBYTE_SEQ (ptr, endp - ptr, bytes);
ptr += bytes;
chars++;
}
return chars;
}
void
parse_str_as_multibyte (str, len, nchars, nbytes)
const unsigned char *str;
int len, *nchars, *nbytes;
{
const unsigned char *endp = str + len;
int n, chars = 0, bytes = 0;
while (str < endp)
{
if (UNIBYTE_STR_AS_MULTIBYTE_P (str, endp - str, n))
str += n, bytes += n;
else
str++, bytes += 2;
chars++;
}
*nchars = chars;
*nbytes = bytes;
return;
}
int
str_as_multibyte (str, len, nbytes, nchars)
unsigned char *str;
int len, nbytes, *nchars;
{
unsigned char *p = str, *endp = str + nbytes;
unsigned char *to;
int chars = 0;
int n;
while (p < endp && UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n))
p += n, chars++;
if (nchars)
*nchars = chars;
if (p == endp)
return nbytes;
to = p;
nbytes = endp - p;
endp = str + len;
safe_bcopy (p, endp - nbytes, nbytes);
p = endp - nbytes;
while (p < endp)
{
if (UNIBYTE_STR_AS_MULTIBYTE_P (p, endp - p, n))
{
while (n--)
*to++ = *p++;
}
else
{
*to++ = LEADING_CODE_8_BIT_CONTROL;
*to++ = *p++ + 0x20;
}
chars++;
}
if (nchars)
*nchars = chars;
return (to - str);
}
int
parse_str_to_multibyte (str, len)
unsigned char *str;
int len;
{
unsigned char *endp = str + len;
int bytes;
for (bytes = 0; str < endp; str++)
bytes += (*str < 0x80 || *str >= 0xA0) ? 1 : 2;
return bytes;
}
int
str_to_multibyte (str, len, bytes)
unsigned char *str;
int len, bytes;
{
unsigned char *p = str, *endp = str + bytes;
unsigned char *to;
while (p < endp && (*p < 0x80 || *p >= 0xA0)) p++;
if (p == endp)
return bytes;
to = p;
bytes = endp - p;
endp = str + len;
safe_bcopy (p, endp - bytes, bytes);
p = endp - bytes;
while (p < endp)
{
if (*p < 0x80 || *p >= 0xA0)
*to++ = *p++;
else
*to++ = LEADING_CODE_8_BIT_CONTROL, *to++ = *p++ + 0x20;
}
return (to - str);
}
int
str_as_unibyte (str, bytes)
unsigned char *str;
int bytes;
{
unsigned char *p = str, *endp = str + bytes;
unsigned char *to = str;
while (p < endp && *p != LEADING_CODE_8_BIT_CONTROL) p++;
to = p;
while (p < endp)
{
if (*p == LEADING_CODE_8_BIT_CONTROL)
*to++ = *(p + 1) - 0x20, p += 2;
else
*to++ = *p++;
}
return (to - str);
}
DEFUN ("string", Fstring, Sstring, 0, MANY, 0,
doc: )
(n, args)
int n;
Lisp_Object *args;
{
int i, bufsize;
unsigned char *buf, *p;
int c;
int multibyte = 0;
Lisp_Object ret;
USE_SAFE_ALLOCA;
bufsize = MAX_MULTIBYTE_LENGTH * n;
SAFE_ALLOCA (buf, unsigned char *, bufsize);
p = buf;
for (i = 0; i < n; i++)
{
CHECK_NUMBER (args[i]);
if (!multibyte && !SINGLE_BYTE_CHAR_P (XFASTINT (args[i])))
multibyte = 1;
}
for (i = 0; i < n; i++)
{
c = XINT (args[i]);
if (multibyte)
p += CHAR_STRING (c, p);
else
*p++ = c;
}
ret = make_string_from_bytes (buf, n, p - buf);
SAFE_FREE ();
return ret;
}
#endif
int
charset_id_internal (charset_name)
char *charset_name;
{
Lisp_Object val;
val= Fget (intern (charset_name), Qcharset);
if (!VECTORP (val))
error ("Charset %s is not defined", charset_name);
return (XINT (XVECTOR (val)->contents[0]));
}
DEFUN ("setup-special-charsets", Fsetup_special_charsets,
Ssetup_special_charsets, 0, 0, 0, doc: )
()
{
charset_latin_iso8859_1 = charset_id_internal ("latin-iso8859-1");
charset_jisx0208_1978 = charset_id_internal ("japanese-jisx0208-1978");
charset_jisx0208 = charset_id_internal ("japanese-jisx0208");
charset_katakana_jisx0201 = charset_id_internal ("katakana-jisx0201");
charset_latin_jisx0201 = charset_id_internal ("latin-jisx0201");
charset_big5_1 = charset_id_internal ("chinese-big5-1");
charset_big5_2 = charset_id_internal ("chinese-big5-2");
charset_mule_unicode_0100_24ff
= charset_id_internal ("mule-unicode-0100-24ff");
charset_mule_unicode_2500_33ff
= charset_id_internal ("mule-unicode-2500-33ff");
charset_mule_unicode_e000_ffff
= charset_id_internal ("mule-unicode-e000-ffff");
return Qnil;
}
void
init_charset_once ()
{
int i, j, k;
staticpro (&Vcharset_table);
staticpro (&Vcharset_symbol_table);
staticpro (&Vgeneric_character_list);
Qcharset_table = intern ("charset-table");
staticpro (&Qcharset_table);
Qchar_table_extra_slots = intern ("char-table-extra-slots");
Fput (Qcharset_table, Qchar_table_extra_slots, make_number (0));
Vcharset_table = Fmake_char_table (Qcharset_table, Qnil);
Qunknown = intern ("unknown");
staticpro (&Qunknown);
Vcharset_symbol_table = Fmake_vector (make_number (MAX_CHARSET + 1),
Qunknown);
for (i = 0; i < 2; i++)
for (j = 0; j < 2; j++)
for (k = 0; k < 128; k++)
iso_charset_table [i][j][k] = -1;
for (i = 0; i < 256; i++)
bytes_by_char_head[i] = 1;
bytes_by_char_head[LEADING_CODE_PRIVATE_11] = 3;
bytes_by_char_head[LEADING_CODE_PRIVATE_12] = 3;
bytes_by_char_head[LEADING_CODE_PRIVATE_21] = 4;
bytes_by_char_head[LEADING_CODE_PRIVATE_22] = 4;
for (i = 0; i < 128; i++)
width_by_char_head[i] = 1;
for (; i < 256; i++)
width_by_char_head[i] = 4;
width_by_char_head[LEADING_CODE_PRIVATE_11] = 1;
width_by_char_head[LEADING_CODE_PRIVATE_12] = 2;
width_by_char_head[LEADING_CODE_PRIVATE_21] = 1;
width_by_char_head[LEADING_CODE_PRIVATE_22] = 2;
{
Lisp_Object val;
val = Qnil;
for (i = 0x81; i < 0x90; i++)
val = Fcons (make_number ((i - 0x70) << 7), val);
for (; i < 0x9A; i++)
val = Fcons (make_number ((i - 0x8F) << 14), val);
for (i = 0xA0; i < 0xF0; i++)
val = Fcons (make_number ((i - 0x70) << 7), val);
for (; i < 0xFF; i++)
val = Fcons (make_number ((i - 0xE0) << 14), val);
Vgeneric_character_list = Fnreverse (val);
}
nonascii_insert_offset = 0;
Vnonascii_translation_table = Qnil;
}
#ifdef emacs
void
syms_of_charset ()
{
Qcharset = intern ("charset");
staticpro (&Qcharset);
Qascii = intern ("ascii");
staticpro (&Qascii);
Qeight_bit_control = intern ("eight-bit-control");
staticpro (&Qeight_bit_control);
Qeight_bit_graphic = intern ("eight-bit-graphic");
staticpro (&Qeight_bit_graphic);
update_charset_table (make_number (CHARSET_ASCII),
make_number (1), make_number (94),
make_number (1),
make_number (0),
make_number ('B'),
make_number (0),
build_string ("ASCII"),
Qnil,
build_string ("ASCII (ISO646 IRV)"));
CHARSET_SYMBOL (CHARSET_ASCII) = Qascii;
Fput (Qascii, Qcharset, CHARSET_TABLE_ENTRY (CHARSET_ASCII));
update_charset_table (make_number (CHARSET_8_BIT_CONTROL),
make_number (1), make_number (96),
make_number (4),
make_number (0),
make_number (-1),
make_number (-1),
build_string ("8-bit control code (0x80..0x9F)"),
Qnil,
Qnil);
CHARSET_SYMBOL (CHARSET_8_BIT_CONTROL) = Qeight_bit_control;
Fput (Qeight_bit_control, Qcharset,
CHARSET_TABLE_ENTRY (CHARSET_8_BIT_CONTROL));
update_charset_table (make_number (CHARSET_8_BIT_GRAPHIC),
make_number (1), make_number (96),
make_number (4),
make_number (0),
make_number (-1),
make_number (-1),
build_string ("8-bit graphic char (0xA0..0xFF)"),
Qnil,
Qnil);
CHARSET_SYMBOL (CHARSET_8_BIT_GRAPHIC) = Qeight_bit_graphic;
Fput (Qeight_bit_graphic, Qcharset,
CHARSET_TABLE_ENTRY (CHARSET_8_BIT_GRAPHIC));
Qauto_fill_chars = intern ("auto-fill-chars");
staticpro (&Qauto_fill_chars);
Fput (Qauto_fill_chars, Qchar_table_extra_slots, make_number (0));
defsubr (&Sdefine_charset);
defsubr (&Sgeneric_character_list);
defsubr (&Sget_unused_iso_final_char);
defsubr (&Sdeclare_equiv_charset);
defsubr (&Sfind_charset_region);
defsubr (&Sfind_charset_string);
defsubr (&Smake_char_internal);
defsubr (&Ssplit_char);
defsubr (&Schar_charset);
defsubr (&Scharset_after);
defsubr (&Siso_charset);
defsubr (&Schar_valid_p);
defsubr (&Sunibyte_char_to_multibyte);
defsubr (&Smultibyte_char_to_unibyte);
defsubr (&Schar_bytes);
defsubr (&Schar_width);
defsubr (&Sstring_width);
defsubr (&Schar_direction);
defsubr (&Sstring);
defsubr (&Ssetup_special_charsets);
DEFVAR_LISP ("charset-list", &Vcharset_list,
doc: );
Vcharset_list = Fcons (Qascii, Fcons (Qeight_bit_control,
Fcons (Qeight_bit_graphic, Qnil)));
DEFVAR_LISP ("translation-table-vector", &Vtranslation_table_vector,
doc: );
Vtranslation_table_vector = Fmake_vector (make_number (16), Qnil);
DEFVAR_INT ("leading-code-private-11", &leading_code_private_11,
doc: );
leading_code_private_11 = LEADING_CODE_PRIVATE_11;
DEFVAR_INT ("leading-code-private-12", &leading_code_private_12,
doc: );
leading_code_private_12 = LEADING_CODE_PRIVATE_12;
DEFVAR_INT ("leading-code-private-21", &leading_code_private_21,
doc: );
leading_code_private_21 = LEADING_CODE_PRIVATE_21;
DEFVAR_INT ("leading-code-private-22", &leading_code_private_22,
doc: );
leading_code_private_22 = LEADING_CODE_PRIVATE_22;
DEFVAR_INT ("nonascii-insert-offset", &nonascii_insert_offset,
doc: );
nonascii_insert_offset = 0;
DEFVAR_LISP ("nonascii-translation-table", &Vnonascii_translation_table,
doc: );
Vnonascii_translation_table = Qnil;
DEFVAR_LISP ("auto-fill-chars", &Vauto_fill_chars,
doc: );
Vauto_fill_chars = Fmake_char_table (Qauto_fill_chars, Qnil);
CHAR_TABLE_SET (Vauto_fill_chars, make_number (' '), Qt);
CHAR_TABLE_SET (Vauto_fill_chars, make_number ('\n'), Qt);
}
#endif