1999-10-16 06:09:09 +00:00
|
|
|
/* Multibyte Character Functions.
|
|
|
|
Copyright (C) 1998 Free Software Foundation, Inc.
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
This file is part of GCC.
|
1999-10-16 06:09:09 +00:00
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
GCC is free software; you can redistribute it and/or modify it under
|
|
|
|
the terms of the GNU General Public License as published by the Free
|
|
|
|
Software Foundation; either version 2, or (at your option) any later
|
|
|
|
version.
|
1999-10-16 06:09:09 +00:00
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
GCC is distributed in the hope that it will be useful, but WITHOUT ANY
|
|
|
|
WARRANTY; without even the implied warranty of MERCHANTABILITY or
|
|
|
|
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
|
|
|
|
for more details.
|
1999-10-16 06:09:09 +00:00
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License
|
2002-02-01 18:16:02 +00:00
|
|
|
along with GCC; see the file COPYING. If not, write to the Free
|
|
|
|
Software Foundation, 59 Temple Place - Suite 330, Boston, MA
|
|
|
|
02111-1307, USA. */
|
1999-10-16 06:09:09 +00:00
|
|
|
|
|
|
|
/* Note regarding cross compilation:
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
In general, translation of multibyte characters to wide characters can
|
1999-10-16 06:09:09 +00:00
|
|
|
only work in a native compiler since the translation function (mbtowc)
|
|
|
|
needs to know about both the source and target character encoding. However,
|
|
|
|
this particular implementation for JIS, SJIS and EUCJP source characters
|
|
|
|
will work for any compiler with a newlib target. Other targets may also
|
|
|
|
work provided that their wchar_t implementation is 2 bytes and the encoding
|
|
|
|
leaves the source character values unchanged (except for removing the
|
|
|
|
state shifting markers). */
|
|
|
|
|
|
|
|
#include "config.h"
|
2002-02-01 18:16:02 +00:00
|
|
|
#ifdef MULTIBYTE_CHARS
|
1999-10-16 06:09:09 +00:00
|
|
|
#include "system.h"
|
|
|
|
#include "mbchar.h"
|
|
|
|
#include <locale.h>
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
typedef enum {ESCAPE, DOLLAR, BRACKET, AT, B, J, NUL, JIS_CHAR, OTHER,
|
|
|
|
JIS_C_NUM} JIS_CHAR_TYPE;
|
1999-10-16 06:09:09 +00:00
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
typedef enum {ASCII, A_ESC, A_ESC_DL, JIS, JIS_1, JIS_2, J_ESC, J_ESC_BR,
|
|
|
|
J2_ESC, J2_ESC_BR, INV, JIS_S_NUM} JIS_STATE;
|
|
|
|
|
|
|
|
typedef enum {COPYA, COPYJ, COPYJ2, MAKE_A, MAKE_J, NOOP,
|
|
|
|
EMPTY, ERROR} JIS_ACTION;
|
|
|
|
|
|
|
|
/* State/action tables for processing JIS encoding:
|
|
|
|
|
|
|
|
Where possible, switches to JIS are grouped with proceding JIS characters
|
|
|
|
and switches to ASCII are grouped with preceding JIS characters.
|
|
|
|
Thus, maximum returned length is:
|
|
|
|
2 (switch to JIS) + 2 (JIS characters) + 2 (switch back to ASCII) = 6. */
|
1999-10-16 06:09:09 +00:00
|
|
|
|
|
|
|
static JIS_STATE JIS_state_table[JIS_S_NUM][JIS_C_NUM] = {
|
2002-02-01 18:16:02 +00:00
|
|
|
/* ESCAPE DOLLAR BRACKET AT B J NUL JIS_CHAR OTH*/
|
1999-10-16 06:09:09 +00:00
|
|
|
/*ASCII*/ { A_ESC, ASCII, ASCII, ASCII, ASCII, ASCII, ASCII,ASCII,ASCII},
|
|
|
|
/*A_ESC*/ { ASCII, A_ESC_DL,ASCII, ASCII, ASCII, ASCII, ASCII,ASCII,ASCII},
|
|
|
|
/*A_ESC_DL*/{ ASCII, ASCII, ASCII, JIS, JIS, ASCII, ASCII,ASCII,ASCII},
|
|
|
|
/*JIS*/ { J_ESC, JIS_1, JIS_1, JIS_1, JIS_1, JIS_1, INV, JIS_1,INV },
|
|
|
|
/*JIS_1*/ { INV, JIS_2, JIS_2, JIS_2, JIS_2, JIS_2, INV, JIS_2,INV },
|
|
|
|
/*JIS_2*/ { J2_ESC,JIS, JIS, JIS, JIS, JIS, INV, JIS, JIS },
|
|
|
|
/*J_ESC*/ { INV, INV, J_ESC_BR, INV, INV, INV, INV, INV, INV },
|
|
|
|
/*J_ESC_BR*/{ INV, INV, INV, INV, ASCII, ASCII, INV, INV, INV },
|
|
|
|
/*J2_ESC*/ { INV, INV, J2_ESC_BR,INV, INV, INV, INV, INV, INV },
|
|
|
|
/*J2_ESC_BR*/{INV, INV, INV, INV, ASCII, ASCII, INV, INV, INV },
|
|
|
|
};
|
|
|
|
|
|
|
|
static JIS_ACTION JIS_action_table[JIS_S_NUM][JIS_C_NUM] = {
|
2002-02-01 18:16:02 +00:00
|
|
|
/* ESCAPE DOLLAR BRACKET AT B J NUL JIS_CHAR OTH */
|
1999-10-16 06:09:09 +00:00
|
|
|
/*ASCII */ {NOOP, COPYA, COPYA, COPYA, COPYA, COPYA, EMPTY, COPYA, COPYA},
|
|
|
|
/*A_ESC */ {COPYA, NOOP, COPYA, COPYA, COPYA, COPYA, COPYA, COPYA, COPYA},
|
|
|
|
/*A_ESC_DL */{COPYA, COPYA, COPYA, MAKE_J, MAKE_J, COPYA, COPYA, COPYA, COPYA},
|
2002-02-01 18:16:02 +00:00
|
|
|
/*JIS */ {NOOP, NOOP, NOOP, NOOP, NOOP, NOOP, ERROR, NOOP, ERROR},
|
|
|
|
/*JIS_1 */ {ERROR, NOOP, NOOP, NOOP, NOOP, NOOP, ERROR, NOOP, ERROR},
|
1999-10-16 06:09:09 +00:00
|
|
|
/*JIS_2 */ {NOOP, COPYJ2,COPYJ2,COPYJ2, COPYJ2, COPYJ2,ERROR, COPYJ2,COPYJ2},
|
2002-02-01 18:16:02 +00:00
|
|
|
/*J_ESC */ {ERROR, ERROR, NOOP, ERROR, ERROR, ERROR, ERROR, ERROR, ERROR},
|
|
|
|
/*J_ESC_BR */{ERROR, ERROR, ERROR, ERROR, NOOP, NOOP, ERROR, ERROR, ERROR},
|
|
|
|
/*J2_ESC */ {ERROR, ERROR, NOOP, ERROR, ERROR, ERROR, ERROR, ERROR, ERROR},
|
|
|
|
/*J2_ESC_BR*/{ERROR, ERROR, ERROR, ERROR, COPYJ, COPYJ, ERROR, ERROR, ERROR},
|
1999-10-16 06:09:09 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
const char *literal_codeset = NULL;
|
|
|
|
|
|
|
|
/* Store into *PWC (if PWC is not null) the wide character
|
|
|
|
corresponding to the multibyte character at the start of the
|
|
|
|
buffer S of size N. Return the number of bytes in the multibyte
|
|
|
|
character. Return -1 if the bytes do not form a valid character,
|
|
|
|
or 0 if S is null or points to a null byte.
|
|
|
|
|
|
|
|
This function behaves like the Standard C function mbtowc, except
|
|
|
|
it treats locale names of the form "C-..." specially. */
|
1999-10-16 06:09:09 +00:00
|
|
|
|
|
|
|
int
|
|
|
|
local_mbtowc (pwc, s, n)
|
2002-02-01 18:16:02 +00:00
|
|
|
wchar_t *pwc;
|
|
|
|
const char *s;
|
|
|
|
size_t n;
|
1999-10-16 06:09:09 +00:00
|
|
|
{
|
|
|
|
static JIS_STATE save_state = ASCII;
|
|
|
|
JIS_STATE curr_state = save_state;
|
2002-02-01 18:16:02 +00:00
|
|
|
const unsigned char *t = (const unsigned char *) s;
|
1999-10-16 06:09:09 +00:00
|
|
|
|
|
|
|
if (s != NULL && n == 0)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
if (literal_codeset == NULL || strlen (literal_codeset) <= 1)
|
2002-02-01 18:16:02 +00:00
|
|
|
/* This must be the "C" locale or unknown locale -- fall thru */
|
|
|
|
;
|
1999-10-16 06:09:09 +00:00
|
|
|
else if (! strcmp (literal_codeset, "C-SJIS"))
|
|
|
|
{
|
|
|
|
int char1;
|
|
|
|
if (s == NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
/* Not state-dependent. */
|
|
|
|
return 0;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
char1 = *t;
|
|
|
|
if (ISSJIS1 (char1))
|
|
|
|
{
|
|
|
|
int char2 = t[1];
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (n <= 1)
|
|
|
|
return -1;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (ISSJIS2 (char2))
|
|
|
|
{
|
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (((wchar_t) *t) << 8) + (wchar_t) (*(t + 1));
|
1999-10-16 06:09:09 +00:00
|
|
|
return 2;
|
|
|
|
}
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
return -1;
|
|
|
|
}
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (wchar_t) *t;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (*t == '\0')
|
|
|
|
return 0;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
else if (! strcmp (literal_codeset, "C-EUCJP"))
|
|
|
|
{
|
|
|
|
int char1;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (s == NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
/* Not state-dependent. */
|
|
|
|
return 0;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
char1 = *t;
|
|
|
|
if (ISEUCJP (char1))
|
|
|
|
{
|
|
|
|
int char2 = t[1];
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (n <= 1)
|
|
|
|
return -1;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (ISEUCJP (char2))
|
|
|
|
{
|
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (((wchar_t) *t) << 8) + (wchar_t) (*(t + 1));
|
1999-10-16 06:09:09 +00:00
|
|
|
return 2;
|
|
|
|
}
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
return -1;
|
|
|
|
}
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (wchar_t) *t;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (*t == '\0')
|
|
|
|
return 0;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
else if (! strcmp (literal_codeset, "C-JIS"))
|
|
|
|
{
|
|
|
|
JIS_ACTION action;
|
|
|
|
JIS_CHAR_TYPE ch;
|
2002-02-01 18:16:02 +00:00
|
|
|
const unsigned char *ptr;
|
|
|
|
size_t i, curr_ch;
|
1999-10-16 06:09:09 +00:00
|
|
|
|
|
|
|
if (s == NULL)
|
|
|
|
{
|
|
|
|
save_state = ASCII;
|
2002-02-01 18:16:02 +00:00
|
|
|
/* State-dependent. */
|
|
|
|
return 1;
|
1999-10-16 06:09:09 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
ptr = t;
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
for (i = 0; i < n; i++)
|
1999-10-16 06:09:09 +00:00
|
|
|
{
|
|
|
|
curr_ch = t[i];
|
|
|
|
switch (curr_ch)
|
|
|
|
{
|
|
|
|
case JIS_ESC_CHAR:
|
|
|
|
ch = ESCAPE;
|
|
|
|
break;
|
|
|
|
case '$':
|
|
|
|
ch = DOLLAR;
|
|
|
|
break;
|
|
|
|
case '@':
|
|
|
|
ch = AT;
|
|
|
|
break;
|
|
|
|
case '(':
|
|
|
|
ch = BRACKET;
|
|
|
|
break;
|
|
|
|
case 'B':
|
|
|
|
ch = B;
|
|
|
|
break;
|
|
|
|
case 'J':
|
|
|
|
ch = J;
|
|
|
|
break;
|
|
|
|
case '\0':
|
|
|
|
ch = NUL;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
if (ISJIS (curr_ch))
|
|
|
|
ch = JIS_CHAR;
|
|
|
|
else
|
|
|
|
ch = OTHER;
|
|
|
|
}
|
|
|
|
|
|
|
|
action = JIS_action_table[curr_state][ch];
|
|
|
|
curr_state = JIS_state_table[curr_state][ch];
|
|
|
|
|
|
|
|
switch (action)
|
|
|
|
{
|
|
|
|
case NOOP:
|
|
|
|
break;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
case EMPTY:
|
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (wchar_t) 0;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
save_state = curr_state;
|
|
|
|
return i;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
case COPYA:
|
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (wchar_t) *ptr;
|
1999-10-16 06:09:09 +00:00
|
|
|
save_state = curr_state;
|
2002-02-01 18:16:02 +00:00
|
|
|
return i + 1;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
case COPYJ:
|
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (((wchar_t) *ptr) << 8) + (wchar_t) (*(ptr + 1));
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
save_state = curr_state;
|
2002-02-01 18:16:02 +00:00
|
|
|
return i + 1;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
case COPYJ2:
|
|
|
|
if (pwc != NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
*pwc = (((wchar_t) *ptr) << 8) + (wchar_t) (*(ptr + 1));
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
save_state = curr_state;
|
2002-02-01 18:16:02 +00:00
|
|
|
return ptr - t + 2;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
case MAKE_A:
|
|
|
|
case MAKE_J:
|
2002-02-01 18:16:02 +00:00
|
|
|
ptr = (const unsigned char *) (t + i + 1);
|
1999-10-16 06:09:09 +00:00
|
|
|
break;
|
2002-02-01 18:16:02 +00:00
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
case ERROR:
|
|
|
|
default:
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
/* More than n bytes needed. */
|
|
|
|
return -1;
|
1999-10-16 06:09:09 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef CROSS_COMPILE
|
|
|
|
if (s == NULL)
|
2002-02-01 18:16:02 +00:00
|
|
|
/* Not state-dependent. */
|
|
|
|
return 0;
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
if (pwc != NULL)
|
|
|
|
*pwc = *s;
|
|
|
|
return 1;
|
|
|
|
#else
|
2002-02-01 18:16:02 +00:00
|
|
|
|
|
|
|
/* This must be the "C" locale or unknown locale. */
|
1999-10-16 06:09:09 +00:00
|
|
|
return mbtowc (pwc, s, n);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
/* Return the number of bytes in the multibyte character at the start
|
|
|
|
of the buffer S of size N. Return -1 if the bytes do not form a
|
|
|
|
valid character, or 0 if S is null or points to a null byte.
|
|
|
|
|
|
|
|
This function behaves like the Standard C function mblen, except
|
|
|
|
it treats locale names of the form "C-..." specially. */
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
int
|
|
|
|
local_mblen (s, n)
|
2002-02-01 18:16:02 +00:00
|
|
|
const char *s;
|
|
|
|
size_t n;
|
1999-10-16 06:09:09 +00:00
|
|
|
{
|
|
|
|
return local_mbtowc (NULL, s, n);
|
|
|
|
}
|
|
|
|
|
2002-02-01 18:16:02 +00:00
|
|
|
/* Return the maximum mumber of bytes in a multibyte character.
|
|
|
|
|
|
|
|
This function returns the same value as the Standard C macro MB_CUR_MAX,
|
|
|
|
except it treats locale names of the form "C-..." specially. */
|
|
|
|
|
1999-10-16 06:09:09 +00:00
|
|
|
int
|
|
|
|
local_mb_cur_max ()
|
|
|
|
{
|
|
|
|
if (literal_codeset == NULL || strlen (literal_codeset) <= 1)
|
|
|
|
;
|
|
|
|
else if (! strcmp (literal_codeset, "C-SJIS"))
|
|
|
|
return 2;
|
|
|
|
else if (! strcmp (literal_codeset, "C-EUCJP"))
|
|
|
|
return 2;
|
|
|
|
else if (! strcmp (literal_codeset, "C-JIS"))
|
|
|
|
return 8; /* 3 + 2 + 3 */
|
|
|
|
|
|
|
|
#ifdef CROSS_COMPILE
|
|
|
|
return 1;
|
|
|
|
#else
|
|
|
|
if (MB_CUR_MAX > 0)
|
|
|
|
return MB_CUR_MAX;
|
|
|
|
|
|
|
|
return 1; /* default */
|
|
|
|
#endif
|
|
|
|
}
|
2002-02-01 18:16:02 +00:00
|
|
|
#else /* MULTIBYTE_CHARS */
|
|
|
|
extern int dummy; /* silence 'ANSI C forbids an empty source file' warning */
|
1999-10-16 06:09:09 +00:00
|
|
|
#endif /* MULTIBYTE_CHARS */
|