mirror of
git://sourceware.org/git/newlib-cygwin.git
synced 2025-01-16 03:19:54 +08:00
9c64d2a7ba
* libc/include/sys/_types.h (_mbstate_t): Changed to use
unsigned char internally.
* libc/sys/linux/sys/_types.h: Ditto.
* libc/include/sys/reent.h
* libc/stdlib/mblen.c (mblen): Use function-specific state
value from default reentrancy structure.
* libc/stdlib/mblen_r.c (_mblen_r): If return code from
_mbtowc_r is less than 0, reset state __count value and
return -1.
* libc/stdlib/mbrlen.c (mbrlen): If the input state pointer
is NULL, use the function-specific pointer provided in the
default reentrancy structure.
* libc/stdlib/mbrtowc.c: Add reentrant form of function.
If input state pointer is NULL, use function-specific area
provided in reentrancy structure.
* libc/stdlib/mbsrtowcs.c: Ditto.
* libc/stdlib/wcrtomb.c: Ditto.
* libc/stdlib/wcsrtombs.c: Ditto.
* libc/stdlib/mbstowcs.c: Reformat.
* libc/stdlib/wcstombs.c: Ditto.
* libc/stdlib/mbstowcs_r.c (_mbstowcs_r): If an error occurs,
reset the state's __count value and return -1.
* libc/stdlib/mbtowc.c: Ditto.
* libc/stdlib/mbtowc_r.c (_mbtowc_r): Add restartable functionality.
If number of bytes is used up before completing a valid multibyte
character, return -2 and save the state.
* libc/stdlib/wctomb_r.c (_wctomb_r): Define __state as __count
and change some __count references to __state for clarity.
173 lines
4.7 KiB
C
173 lines
4.7 KiB
C
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <wchar.h>
|
|
#include <locale.h>
|
|
#include "mbctype.h"
|
|
|
|
/* for some conversions, we use the __count field as a place to store a state value */
|
|
#define __state __count
|
|
|
|
int
|
|
_DEFUN (_wctomb_r, (r, s, wchar, state),
|
|
struct _reent *r _AND
|
|
char *s _AND
|
|
wchar_t wchar _AND
|
|
mbstate_t *state)
|
|
{
|
|
if (strlen (r->_current_locale) <= 1)
|
|
{ /* fall-through */ }
|
|
else if (!strcmp (r->_current_locale, "C-UTF-8"))
|
|
{
|
|
if (s == NULL)
|
|
return 0; /* UTF-8 encoding is not state-dependent */
|
|
|
|
if (wchar <= 0x7f)
|
|
{
|
|
*s = wchar;
|
|
return 1;
|
|
}
|
|
else if (wchar >= 0x80 && wchar <= 0x7ff)
|
|
{
|
|
*s++ = 0xc0 | ((wchar & 0x7c0) >> 6);
|
|
*s = 0x80 | (wchar & 0x3f);
|
|
return 2;
|
|
}
|
|
else if (wchar >= 0x800 && wchar <= 0xffff)
|
|
{
|
|
/* UTF-16 surrogates -- must not occur in normal UCS-4 data */
|
|
if (wchar >= 0xd800 && wchar <= 0xdfff)
|
|
return -1;
|
|
|
|
*s++ = 0xe0 | ((wchar & 0xf000) >> 12);
|
|
*s++ = 0x80 | ((wchar & 0xfc0) >> 6);
|
|
*s = 0x80 | (wchar & 0x3f);
|
|
return 3;
|
|
}
|
|
else if (wchar >= 0x10000 && wchar <= 0x1fffff)
|
|
{
|
|
*s++ = 0xf0 | ((wchar & 0x1c0000) >> 18);
|
|
*s++ = 0x80 | ((wchar & 0x3f000) >> 12);
|
|
*s++ = 0x80 | ((wchar & 0xfc0) >> 6);
|
|
*s = 0x80 | (wchar & 0x3f);
|
|
return 4;
|
|
}
|
|
else if (wchar >= 0x200000 && wchar <= 0x3ffffff)
|
|
{
|
|
*s++ = 0xf8 | ((wchar & 0x3000000) >> 24);
|
|
*s++ = 0x80 | ((wchar & 0xfc0000) >> 18);
|
|
*s++ = 0x80 | ((wchar & 0x3f000) >> 12);
|
|
*s++ = 0x80 | ((wchar & 0xfc0) >> 6);
|
|
*s = 0x80 | (wchar & 0x3f);
|
|
return 5;
|
|
}
|
|
else if (wchar >= 0x4000000 && wchar <= 0x7fffffff)
|
|
{
|
|
*s++ = 0xfc | ((wchar & 0x40000000) >> 30);
|
|
*s++ = 0x80 | ((wchar & 0x3f000000) >> 24);
|
|
*s++ = 0x80 | ((wchar & 0xfc0000) >> 18);
|
|
*s++ = 0x80 | ((wchar & 0x3f000) >> 12);
|
|
*s++ = 0x80 | ((wchar & 0xfc0) >> 6);
|
|
*s = 0x80 | (wchar & 0x3f);
|
|
return 6;
|
|
}
|
|
else
|
|
return -1;
|
|
}
|
|
else if (!strcmp (r->_current_locale, "C-SJIS"))
|
|
{
|
|
unsigned char char2 = (unsigned char)wchar;
|
|
unsigned char char1 = (unsigned char)(wchar >> 8);
|
|
|
|
if (s == NULL)
|
|
return 0; /* not state-dependent */
|
|
|
|
if (char1 != 0x00)
|
|
{
|
|
/* first byte is non-zero..validate multi-byte char */
|
|
if (_issjis1(char1) && _issjis2(char2))
|
|
{
|
|
*s++ = (char)char1;
|
|
*s = (char)char2;
|
|
return 2;
|
|
}
|
|
else
|
|
return -1;
|
|
}
|
|
}
|
|
else if (!strcmp (r->_current_locale, "C-EUCJP"))
|
|
{
|
|
unsigned char char2 = (unsigned char)wchar;
|
|
unsigned char char1 = (unsigned char)(wchar >> 8);
|
|
|
|
if (s == NULL)
|
|
return 0; /* not state-dependent */
|
|
|
|
if (char1 != 0x00)
|
|
{
|
|
/* first byte is non-zero..validate multi-byte char */
|
|
if (_iseucjp (char1) && _iseucjp (char2))
|
|
{
|
|
*s++ = (char)char1;
|
|
*s = (char)char2;
|
|
return 2;
|
|
}
|
|
else
|
|
return -1;
|
|
}
|
|
}
|
|
else if (!strcmp (r->_current_locale, "C-JIS"))
|
|
{
|
|
int cnt = 0;
|
|
unsigned char char2 = (unsigned char)wchar;
|
|
unsigned char char1 = (unsigned char)(wchar >> 8);
|
|
|
|
if (s == NULL)
|
|
return 1; /* state-dependent */
|
|
|
|
if (char1 != 0x00)
|
|
{
|
|
/* first byte is non-zero..validate multi-byte char */
|
|
if (_isjis (char1) && _isjis (char2))
|
|
{
|
|
if (state->__state == 0)
|
|
{
|
|
/* must switch from ASCII to JIS state */
|
|
state->__state = 1;
|
|
*s++ = ESC_CHAR;
|
|
*s++ = '$';
|
|
*s++ = 'B';
|
|
cnt = 3;
|
|
}
|
|
*s++ = (char)char1;
|
|
*s = (char)char2;
|
|
return cnt + 2;
|
|
}
|
|
else
|
|
return -1;
|
|
}
|
|
else
|
|
{
|
|
if (state->__state != 0)
|
|
{
|
|
/* must switch from JIS to ASCII state */
|
|
state->__state = 0;
|
|
*s++ = ESC_CHAR;
|
|
*s++ = '(';
|
|
*s++ = 'B';
|
|
cnt = 3;
|
|
}
|
|
*s = (char)char2;
|
|
return cnt + 1;
|
|
}
|
|
}
|
|
|
|
if (s == NULL)
|
|
return 0;
|
|
|
|
/* otherwise we are dealing with a single byte character */
|
|
*s = (char) wchar;
|
|
return 1;
|
|
}
|
|
|
|
|