URL https://opencores.org/ocsvn/openrisc_me/openrisc_me/trunk

Subversion Repositories openrisc_me

[/] [openrisc/] [trunk/] [gnu-src/] [newlib-1.18.0/] [newlib/] [libc/] [stdlib/] [mbtowc_r.c] - Blame information for rev 309

Go to most recent revision | Details | Compare with Previous | View Log


#include <newlib.h>
#include <stdlib.h>
#include <locale.h>
#include "mbctype.h"
#include <wchar.h>
#include <string.h>
#include <errno.h>
#include "local.h"
 
int (*__mbtowc) (struct _reent *, wchar_t *, const char *, size_t,
                 const char *, mbstate_t *)
#ifdef __CYGWIN__
   = __utf8_mbtowc;
#else
   = __ascii_mbtowc;
#endif
 
int
_DEFUN (_mbtowc_r, (r, pwc, s, n, state),
        struct _reent *r   _AND
        wchar_t       *pwc _AND
        const char    *s   _AND
        size_t         n   _AND
        mbstate_t      *state)
{
  return __mbtowc (r, pwc, s, n, __locale_charset (), state);
}
 
int
_DEFUN (__ascii_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    return 0;
 
  if (n == 0)
    return -2;
 
  *pwc = (wchar_t)*t;
 
  if (*t == '\0')
    return 0;
 
  return 1;
}
 
#ifdef _MB_CAPABLE
typedef enum { ESCAPE, DOLLAR, BRACKET, AT, B, J,
               NUL, JIS_CHAR, OTHER, JIS_C_NUM } JIS_CHAR_TYPE;
typedef enum { ASCII, JIS, A_ESC, A_ESC_DL, JIS_1, J_ESC, J_ESC_BR,
               INV, JIS_S_NUM } JIS_STATE;
typedef enum { COPY_A, COPY_J1, COPY_J2, MAKE_A, 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.
 *************************************************************************************/
 
static JIS_STATE JIS_state_table[JIS_S_NUM][JIS_C_NUM] = {
/*              ESCAPE   DOLLAR    BRACKET   AT       B       J        NUL      JIS_CHAR  OTHER */
/* ASCII */   { A_ESC,   ASCII,    ASCII,    ASCII,   ASCII,  ASCII,   ASCII,   ASCII,    ASCII },
/* JIS */     { J_ESC,   JIS_1,    JIS_1,    JIS_1,   JIS_1,  JIS_1,   INV,     JIS_1,    INV },
/* 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_1 */   { INV,     JIS,      JIS,      JIS,     JIS,    JIS,     INV,     JIS,      INV },
/* 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 },
};
 
static JIS_ACTION JIS_action_table[JIS_S_NUM][JIS_C_NUM] = {
/*              ESCAPE   DOLLAR    BRACKET   AT       B        J        NUL      JIS_CHAR  OTHER */
/* ASCII */   { NOOP,    COPY_A,   COPY_A,   COPY_A,  COPY_A,  COPY_A,  EMPTY,   COPY_A,  COPY_A},
/* JIS */     { NOOP,    COPY_J1,  COPY_J1,  COPY_J1, COPY_J1, COPY_J1, ERROR,   COPY_J1, ERROR },
/* A_ESC */   { COPY_A,  NOOP,     COPY_A,   COPY_A,  COPY_A,  COPY_A,  COPY_A,  COPY_A,  COPY_A},
/* A_ESC_DL */{ COPY_A,  COPY_A,   COPY_A,   NOOP,    NOOP,    COPY_A,  COPY_A,  COPY_A,  COPY_A},
/* JIS_1 */   { ERROR,   COPY_J2,  COPY_J2,  COPY_J2, COPY_J2, COPY_J2, ERROR,   COPY_J2, ERROR },
/* J_ESC */   { ERROR,   ERROR,    NOOP,     ERROR,   ERROR,   ERROR,   ERROR,   ERROR,   ERROR },
/* J_ESC_BR */{ ERROR,   ERROR,    ERROR,    ERROR,   MAKE_A,  MAKE_A,  ERROR,   ERROR,   ERROR },
};
 
/* we override the mbstate_t __count field for more complex encodings and use it store a state value */
#define __state __count
 
#ifdef _MB_EXTENDED_CHARSETS_ISO
int
_DEFUN (__iso_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    return 0;
 
  if (n == 0)
    return -2;
 
  if (*t >= 0xa0)
    {
      int iso_idx = __iso_8859_index (charset + 9);
      if (iso_idx >= 0)
        {
          *pwc = __iso_8859_conv[iso_idx][*t - 0xa0];
          if (*pwc == 0) /* Invalid character */
            {
              r->_errno = EILSEQ;
              return -1;
            }
          return 1;
        }
    }
 
  *pwc = (wchar_t) *t;
 
  if (*t == '\0')
    return 0;
 
  return 1;
}
#endif /* _MB_EXTENDED_CHARSETS_ISO */
 
#ifdef _MB_EXTENDED_CHARSETS_WINDOWS
int
_DEFUN (__cp_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    return 0;
 
  if (n == 0)
    return -2;
 
  if (*t >= 0x80)
    {
      int cp_idx = __cp_index (charset + 2);
      if (cp_idx >= 0)
        {
          *pwc = __cp_conv[cp_idx][*t - 0x80];
          if (*pwc == 0) /* Invalid character */
            {
              r->_errno = EILSEQ;
              return -1;
            }
          return 1;
        }
    }
 
  *pwc = (wchar_t)*t;
 
  if (*t == '\0')
    return 0;
 
  return 1;
}
#endif /* _MB_EXTENDED_CHARSETS_WINDOWS */
 
int
_DEFUN (__utf8_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
  int ch;
  int i = 0;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    return 0;
 
  if (n == 0)
    return -2;
 
  if (state->__count == 0)
    ch = t[i++];
  else
    ch = state->__value.__wchb[0];
 
  if (ch == '\0')
    {
      *pwc = 0;
      state->__count = 0;
      return 0; /* s points to the null character */
    }
 
  if (ch <= 0x7f)
    {
      /* single-byte sequence */
      state->__count = 0;
      *pwc = ch;
      return 1;
    }
  if (ch >= 0xc0 && ch <= 0xdf)
    {
      /* two-byte sequence */
      state->__value.__wchb[0] = ch;
      if (state->__count == 0)
        state->__count = 1;
      else if (n < (size_t)-1)
        ++n;
      if (n < 2)
        return -2;
      ch = t[i++];
      if (ch < 0x80 || ch > 0xbf)
        {
          r->_errno = EILSEQ;
          return -1;
        }
      if (state->__value.__wchb[0] < 0xc2)
        {
          /* overlong UTF-8 sequence */
          r->_errno = EILSEQ;
          return -1;
        }
      state->__count = 0;
      *pwc = (wchar_t)((state->__value.__wchb[0] & 0x1f) << 6)
        |    (wchar_t)(ch & 0x3f);
      return i;
    }
  if (ch >= 0xe0 && ch <= 0xef)
    {
      /* three-byte sequence */
      wchar_t tmp;
      state->__value.__wchb[0] = ch;
      if (state->__count == 0)
        state->__count = 1;
      else if (n < (size_t)-1)
        ++n;
      if (n < 2)
        return -2;
      ch = (state->__count == 1) ? t[i++] : state->__value.__wchb[1];
      if (state->__value.__wchb[0] == 0xe0 && ch < 0xa0)
        {
          /* overlong UTF-8 sequence */
          r->_errno = EILSEQ;
          return -1;
        }
      if (ch < 0x80 || ch > 0xbf)
        {
          r->_errno = EILSEQ;
          return -1;
        }
      state->__value.__wchb[1] = ch;
      if (state->__count == 1)
        state->__count = 2;
      else if (n < (size_t)-1)
        ++n;
      if (n < 3)
        return -2;
      ch = t[i++];
      if (ch < 0x80 || ch > 0xbf)
        {
          r->_errno = EILSEQ;
          return -1;
        }
      state->__count = 0;
      tmp = (wchar_t)((state->__value.__wchb[0] & 0x0f) << 12)
        |    (wchar_t)((state->__value.__wchb[1] & 0x3f) << 6)
        |     (wchar_t)(ch & 0x3f);
      *pwc = tmp;
      return i;
    }
  if (ch >= 0xf0 && ch <= 0xf4)
    {
      /* four-byte sequence */
      wint_t tmp;
      state->__value.__wchb[0] = ch;
      if (state->__count == 0)
        state->__count = 1;
      else if (n < (size_t)-1)
        ++n;
      if (n < 2)
        return -2;
      ch = (state->__count == 1) ? t[i++] : state->__value.__wchb[1];
      if ((state->__value.__wchb[0] == 0xf0 && ch < 0x90)
          || (state->__value.__wchb[0] == 0xf4 && ch >= 0x90))
        {
          /* overlong UTF-8 sequence or result is > 0x10ffff */
          r->_errno = EILSEQ;
          return -1;
        }
      if (ch < 0x80 || ch > 0xbf)
        {
          r->_errno = EILSEQ;
          return -1;
        }
      state->__value.__wchb[1] = ch;
      if (state->__count == 1)
        state->__count = 2;
      else if (n < (size_t)-1)
        ++n;
      if (n < 3)
        return -2;
      ch = (state->__count == 2) ? t[i++] : state->__value.__wchb[2];
      if (ch < 0x80 || ch > 0xbf)
        {
          r->_errno = EILSEQ;
          return -1;
        }
      state->__value.__wchb[2] = ch;
      if (state->__count == 2)
        state->__count = 3;
      else if (n < (size_t)-1)
        ++n;
      if (state->__count == 3 && sizeof(wchar_t) == 2)
        {
          /* On systems which have wchar_t being UTF-16 values, the value
             doesn't fit into a single wchar_t in this case.  So what we
             do here is to store the state with a special value of __count
             and return the first half of a surrogate pair.  The first
             three bytes of a UTF-8 sequence are enough to generate the
             first half of a UTF-16 surrogate pair.  As return value we
             choose to return the number of bytes actually read up to
             here.
             The second half of the surrogate pair is returned in case we
             recognize the special __count value of four, and the next
             byte is actually a valid value.  See below. */
          tmp = (wint_t)((state->__value.__wchb[0] & 0x07) << 18)
            |   (wint_t)((state->__value.__wchb[1] & 0x3f) << 12)
            |   (wint_t)((state->__value.__wchb[2] & 0x3f) << 6);
          state->__count = 4;
          *pwc = 0xd800 | ((tmp - 0x10000) >> 10);
          return i;
        }
      if (n < 4)
        return -2;
      ch = t[i++];
      if (ch < 0x80 || ch > 0xbf)
        {
          r->_errno = EILSEQ;
          return -1;
        }
      tmp = (wint_t)((state->__value.__wchb[0] & 0x07) << 18)
        |   (wint_t)((state->__value.__wchb[1] & 0x3f) << 12)
        |   (wint_t)((state->__value.__wchb[2] & 0x3f) << 6)
        |   (wint_t)(ch & 0x3f);
      if (state->__count == 4 && sizeof(wchar_t) == 2)
        /* Create the second half of the surrogate pair for systems with
           wchar_t == UTF-16 . */
        *pwc = 0xdc00 | (tmp & 0x3ff);
      else
        *pwc = tmp;
      state->__count = 0;
      return i;
    }
 
  r->_errno = EILSEQ;
  return -1;
}
 
/* Cygwin defines its own doublebyte charset conversion functions
   because the underlying OS requires wchar_t == UTF-16. */
#ifndef  __CYGWIN__
int
_DEFUN (__sjis_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
  int ch;
  int i = 0;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    return 0;  /* not state-dependent */
 
  if (n == 0)
    return -2;
 
  ch = t[i++];
  if (state->__count == 0)
    {
      if (_issjis1 (ch))
        {
          state->__value.__wchb[0] = ch;
          state->__count = 1;
          if (n <= 1)
            return -2;
          ch = t[i++];
        }
    }
  if (state->__count == 1)
    {
      if (_issjis2 (ch))
        {
          *pwc = (((wchar_t)state->__value.__wchb[0]) << 8) + (wchar_t)ch;
          state->__count = 0;
          return i;
        }
      else
        {
          r->_errno = EILSEQ;
          return -1;
        }
    }
 
  *pwc = (wchar_t)*t;
 
  if (*t == '\0')
    return 0;
 
  return 1;
}
 
int
_DEFUN (__eucjp_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
  int ch;
  int i = 0;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    return 0;
 
  if (n == 0)
    return -2;
 
  ch = t[i++];
  if (state->__count == 0)
    {
      if (_iseucjp1 (ch))
        {
          state->__value.__wchb[0] = ch;
          state->__count = 1;
          if (n <= 1)
            return -2;
          ch = t[i++];
        }
    }
  if (state->__count == 1)
    {
      if (_iseucjp2 (ch))
        {
          if (state->__value.__wchb[0] == 0x8f)
            {
              state->__value.__wchb[1] = ch;
              state->__count = 2;
              if (n <= i)
                return -2;
              ch = t[i++];
            }
          else
            {
              *pwc = (((wchar_t)state->__value.__wchb[0]) << 8) + (wchar_t)ch;
              state->__count = 0;
              return i;
            }
        }
      else
        {
          r->_errno = EILSEQ;
          return -1;
        }
    }
  if (state->__count == 2)
    {
      if (_iseucjp2 (ch))
        {
          *pwc = (((wchar_t)state->__value.__wchb[1]) << 8)
                 + (wchar_t)(ch & 0x7f);
          state->__count = 0;
          return i;
        }
      else
        {
          r->_errno = EILSEQ;
          return -1;
        }
    }
 
  *pwc = (wchar_t)*t;
 
  if (*t == '\0')
    return 0;
 
  return 1;
}
 
int
_DEFUN (__jis_mbtowc, (r, pwc, s, n, charset, state),
        struct _reent *r       _AND
        wchar_t       *pwc     _AND
        const char    *s       _AND
        size_t         n       _AND
        const char    *charset _AND
        mbstate_t      *state)
{
  wchar_t dummy;
  unsigned char *t = (unsigned char *)s;
  JIS_STATE curr_state;
  JIS_ACTION action;
  JIS_CHAR_TYPE ch;
  unsigned char *ptr;
  unsigned int i;
  int curr_ch;
 
  if (pwc == NULL)
    pwc = &dummy;
 
  if (s == NULL)
    {
      state->__state = ASCII;
      return 1;  /* state-dependent */
    }
 
  if (n == 0)
    return -2;
 
  curr_state = state->__state;
  ptr = t;
 
  for (i = 0; i < n; ++i)
    {
      curr_ch = t[i];
      switch (curr_ch)
        {
        case 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;
        case EMPTY:
          state->__state = ASCII;
          *pwc = (wchar_t)0;
          return 0;
        case COPY_A:
          state->__state = ASCII;
          *pwc = (wchar_t)*ptr;
          return (i + 1);
        case COPY_J1:
          state->__value.__wchb[0] = t[i];
          break;
        case COPY_J2:
          state->__state = JIS;
          *pwc = (((wchar_t)state->__value.__wchb[0]) << 8) + (wchar_t)(t[i]);
          return (i + 1);
        case MAKE_A:
          ptr = (unsigned char *)(t + i + 1);
          break;
        case ERROR:
        default:
          r->_errno = EILSEQ;
          return -1;
        }
 
    }
 
  state->__state = curr_state;
  return -2;  /* n < bytes needed */
}
#endif /* !__CYGWIN__*/
#endif /* _MB_CAPABLE */

Browse

Tools

Subversion Repositories openrisc_me

[/] [openrisc/] [trunk/] [gnu-src/] [newlib-1.18.0/] [newlib/] [libc/] [stdlib/] [mbtowc_r.c] - Blame information for rev 309

Line No.	Rev	Author	Line
1	207	jeremybenn	`#include <newlib.h>`
2			`#include <stdlib.h>`
3			`#include <locale.h>`
4			`#include "mbctype.h"`
5			`#include <wchar.h>`
6			`#include <string.h>`
7			`#include <errno.h>`
8			`#include "local.h"`
9
10			`int (__mbtowc) (struct _reent , wchar_t , const char , size_t,`
11			`const char , mbstate_t )`
12			`#ifdef __CYGWIN__`
13			`= __utf8_mbtowc;`
14			`#else`
15			`= __ascii_mbtowc;`
16			`#endif`
17
18			`int`
19			`_DEFUN (_mbtowc_r, (r, pwc, s, n, state),`
20			`struct _reent *r _AND`
21			`wchar_t *pwc _AND`
22			`const char *s _AND`
23			`size_t n _AND`
24			`mbstate_t *state)`
25			`{`
26			`return __mbtowc (r, pwc, s, n, __locale_charset (), state);`
27			`}`
28
29			`int`
30			`_DEFUN (__ascii_mbtowc, (r, pwc, s, n, charset, state),`
31			`struct _reent *r _AND`
32			`wchar_t *pwc _AND`
33			`const char *s _AND`
34			`size_t n _AND`
35			`const char *charset _AND`
36			`mbstate_t *state)`
37			`{`
38			`wchar_t dummy;`
39			`unsigned char t = (unsigned char )s;`
40
41			`if (pwc == NULL)`
42			`pwc = &dummy;`
43
44			`if (s == NULL)`
45			`return 0;`
46
47			`if (n == 0)`
48			`return -2;`
49
50			`pwc = (wchar_t)t;`
51
52			`if (*t == '\0')`
53			`return 0;`
54
55			`return 1;`
56			`}`
57
58			`#ifdef _MB_CAPABLE`
59			`typedef enum { ESCAPE, DOLLAR, BRACKET, AT, B, J,`
60			`NUL, JIS_CHAR, OTHER, JIS_C_NUM } JIS_CHAR_TYPE;`
61			`typedef enum { ASCII, JIS, A_ESC, A_ESC_DL, JIS_1, J_ESC, J_ESC_BR,`
62			`INV, JIS_S_NUM } JIS_STATE;`
63			`typedef enum { COPY_A, COPY_J1, COPY_J2, MAKE_A, NOOP, EMPTY, ERROR } JIS_ACTION;`
64
65			`/**************************************************************************************`
66			`* state/action tables for processing JIS encoding`
67			`* Where possible, switches to JIS are grouped with proceding JIS characters and switches`
68			`* to ASCII are grouped with preceding JIS characters. Thus, maximum returned length`
69			`* is 2 (switch to JIS) + 2 (JIS characters) + 2 (switch back to ASCII) = 6.`
70			`*************************************************************************************/`
71
72			`static JIS_STATE JIS_state_table[JIS_S_NUM][JIS_C_NUM] = {`
73			`/* ESCAPE DOLLAR BRACKET AT B J NUL JIS_CHAR OTHER */`
74			`/* ASCII */ { A_ESC, ASCII, ASCII, ASCII, ASCII, ASCII, ASCII, ASCII, ASCII },`
75			`/* JIS */ { J_ESC, JIS_1, JIS_1, JIS_1, JIS_1, JIS_1, INV, JIS_1, INV },`
76			`/* A_ESC */ { ASCII, A_ESC_DL, ASCII, ASCII, ASCII, ASCII, ASCII, ASCII, ASCII },`
77			`/* A_ESC_DL */{ ASCII, ASCII, ASCII, JIS, JIS, ASCII, ASCII, ASCII, ASCII },`
78			`/* JIS_1 */ { INV, JIS, JIS, JIS, JIS, JIS, INV, JIS, INV },`
79			`/* J_ESC */ { INV, INV, J_ESC_BR, INV, INV, INV, INV, INV, INV },`
80			`/* J_ESC_BR */{ INV, INV, INV, INV, ASCII, ASCII, INV, INV, INV },`
81			`};`
82
83			`static JIS_ACTION JIS_action_table[JIS_S_NUM][JIS_C_NUM] = {`
84			`/* ESCAPE DOLLAR BRACKET AT B J NUL JIS_CHAR OTHER */`
85			`/* ASCII */ { NOOP, COPY_A, COPY_A, COPY_A, COPY_A, COPY_A, EMPTY, COPY_A, COPY_A},`
86			`/* JIS */ { NOOP, COPY_J1, COPY_J1, COPY_J1, COPY_J1, COPY_J1, ERROR, COPY_J1, ERROR },`
87			`/* A_ESC */ { COPY_A, NOOP, COPY_A, COPY_A, COPY_A, COPY_A, COPY_A, COPY_A, COPY_A},`
88			`/* A_ESC_DL */{ COPY_A, COPY_A, COPY_A, NOOP, NOOP, COPY_A, COPY_A, COPY_A, COPY_A},`
89			`/* JIS_1 */ { ERROR, COPY_J2, COPY_J2, COPY_J2, COPY_J2, COPY_J2, ERROR, COPY_J2, ERROR },`
90			`/* J_ESC */ { ERROR, ERROR, NOOP, ERROR, ERROR, ERROR, ERROR, ERROR, ERROR },`
91			`/* J_ESC_BR */{ ERROR, ERROR, ERROR, ERROR, MAKE_A, MAKE_A, ERROR, ERROR, ERROR },`
92			`};`
93
94			`/* we override the mbstate_t __count field for more complex encodings and use it store a state value */`
95			`#define __state __count`
96
97			`#ifdef _MB_EXTENDED_CHARSETS_ISO`
98			`int`
99			`_DEFUN (__iso_mbtowc, (r, pwc, s, n, charset, state),`
100			`struct _reent *r _AND`
101			`wchar_t *pwc _AND`
102			`const char *s _AND`
103			`size_t n _AND`
104			`const char *charset _AND`
105			`mbstate_t *state)`
106			`{`
107			`wchar_t dummy;`
108			`unsigned char t = (unsigned char )s;`
109
110			`if (pwc == NULL)`
111			`pwc = &dummy;`
112
113			`if (s == NULL)`
114			`return 0;`
115
116			`if (n == 0)`
117			`return -2;`
118
119			`if (*t >= 0xa0)`
120			`{`
121			`int iso_idx = __iso_8859_index (charset + 9);`
122			`if (iso_idx >= 0)`
123			`{`
124			`pwc = __iso_8859_conv[iso_idx][t - 0xa0];`
125			`if (pwc == 0) / Invalid character */`
126			`{`
127			`r->_errno = EILSEQ;`
128			`return -1;`
129			`}`
130			`return 1;`
131			`}`
132			`}`
133
134			`pwc = (wchar_t) t;`
135
136			`if (*t == '\0')`
137			`return 0;`
138
139			`return 1;`
140			`}`
141			`#endif /* _MB_EXTENDED_CHARSETS_ISO */`
142
143			`#ifdef _MB_EXTENDED_CHARSETS_WINDOWS`
144			`int`
145			`_DEFUN (__cp_mbtowc, (r, pwc, s, n, charset, state),`
146			`struct _reent *r _AND`
147			`wchar_t *pwc _AND`
148			`const char *s _AND`
149			`size_t n _AND`
150			`const char *charset _AND`
151			`mbstate_t *state)`
152			`{`
153			`wchar_t dummy;`
154			`unsigned char t = (unsigned char )s;`
155
156			`if (pwc == NULL)`
157			`pwc = &dummy;`
158
159			`if (s == NULL)`
160			`return 0;`
161
162			`if (n == 0)`
163			`return -2;`
164
165			`if (*t >= 0x80)`
166			`{`
167			`int cp_idx = __cp_index (charset + 2);`
168			`if (cp_idx >= 0)`
169			`{`
170			`pwc = __cp_conv[cp_idx][t - 0x80];`
171			`if (pwc == 0) / Invalid character */`
172			`{`
173			`r->_errno = EILSEQ;`
174			`return -1;`
175			`}`
176			`return 1;`
177			`}`
178			`}`
179
180			`pwc = (wchar_t)t;`
181
182			`if (*t == '\0')`
183			`return 0;`
184
185			`return 1;`
186			`}`
187			`#endif /* _MB_EXTENDED_CHARSETS_WINDOWS */`
188
189			`int`
190			`_DEFUN (__utf8_mbtowc, (r, pwc, s, n, charset, state),`
191			`struct _reent *r _AND`
192			`wchar_t *pwc _AND`
193			`const char *s _AND`
194			`size_t n _AND`
195			`const char *charset _AND`
196			`mbstate_t *state)`
197			`{`
198			`wchar_t dummy;`
199			`unsigned char t = (unsigned char )s;`
200			`int ch;`
201			`int i = 0;`
202
203			`if (pwc == NULL)`
204			`pwc = &dummy;`
205
206			`if (s == NULL)`
207			`return 0;`
208
209			`if (n == 0)`
210			`return -2;`
211
212			`if (state->__count == 0)`
213			`ch = t[i++];`
214			`else`
215			`ch = state->__value.__wchb[0];`
216
217			`if (ch == '\0')`
218			`{`
219			`*pwc = 0;`
220			`state->__count = 0;`
221			`return 0; /* s points to the null character */`
222			`}`
223
224			`if (ch <= 0x7f)`
225			`{`
226			`/* single-byte sequence */`
227			`state->__count = 0;`
228			`*pwc = ch;`
229			`return 1;`
230			`}`
231			`if (ch >= 0xc0 && ch <= 0xdf)`
232			`{`
233			`/* two-byte sequence */`
234			`state->__value.__wchb[0] = ch;`
235			`if (state->__count == 0)`
236			`state->__count = 1;`
237			`else if (n < (size_t)-1)`
238			`++n;`
239			`if (n < 2)`
240			`return -2;`
241			`ch = t[i++];`
242			`if (ch < 0x80 \|\| ch > 0xbf)`
243			`{`
244			`r->_errno = EILSEQ;`
245			`return -1;`
246			`}`
247			`if (state->__value.__wchb[0] < 0xc2)`
248			`{`
249			`/* overlong UTF-8 sequence */`
250			`r->_errno = EILSEQ;`
251			`return -1;`
252			`}`
253			`state->__count = 0;`
254			`*pwc = (wchar_t)((state->__value.__wchb[0] & 0x1f) << 6)`
255			`\| (wchar_t)(ch & 0x3f);`
256			`return i;`
257			`}`
258			`if (ch >= 0xe0 && ch <= 0xef)`
259			`{`
260			`/* three-byte sequence */`
261			`wchar_t tmp;`
262			`state->__value.__wchb[0] = ch;`
263			`if (state->__count == 0)`
264			`state->__count = 1;`
265			`else if (n < (size_t)-1)`
266			`++n;`
267			`if (n < 2)`
268			`return -2;`
269			`ch = (state->__count == 1) ? t[i++] : state->__value.__wchb[1];`
270			`if (state->__value.__wchb[0] == 0xe0 && ch < 0xa0)`
271			`{`
272			`/* overlong UTF-8 sequence */`
273			`r->_errno = EILSEQ;`
274			`return -1;`
275			`}`
276			`if (ch < 0x80 \|\| ch > 0xbf)`
277			`{`
278			`r->_errno = EILSEQ;`
279			`return -1;`
280			`}`
281			`state->__value.__wchb[1] = ch;`
282			`if (state->__count == 1)`
283			`state->__count = 2;`
284			`else if (n < (size_t)-1)`
285			`++n;`
286			`if (n < 3)`
287			`return -2;`
288			`ch = t[i++];`
289			`if (ch < 0x80 \|\| ch > 0xbf)`
290			`{`
291			`r->_errno = EILSEQ;`
292			`return -1;`
293			`}`
294			`state->__count = 0;`
295			`tmp = (wchar_t)((state->__value.__wchb[0] & 0x0f) << 12)`
296			`\| (wchar_t)((state->__value.__wchb[1] & 0x3f) << 6)`
297			`\| (wchar_t)(ch & 0x3f);`
298			`*pwc = tmp;`
299			`return i;`
300			`}`
301			`if (ch >= 0xf0 && ch <= 0xf4)`
302			`{`
303			`/* four-byte sequence */`
304			`wint_t tmp;`
305			`state->__value.__wchb[0] = ch;`
306			`if (state->__count == 0)`
307			`state->__count = 1;`
308			`else if (n < (size_t)-1)`
309			`++n;`
310			`if (n < 2)`
311			`return -2;`
312			`ch = (state->__count == 1) ? t[i++] : state->__value.__wchb[1];`
313			`if ((state->__value.__wchb[0] == 0xf0 && ch < 0x90)`
314			`\|\| (state->__value.__wchb[0] == 0xf4 && ch >= 0x90))`
315			`{`
316			`/* overlong UTF-8 sequence or result is > 0x10ffff */`
317			`r->_errno = EILSEQ;`
318			`return -1;`
319			`}`
320			`if (ch < 0x80 \|\| ch > 0xbf)`
321			`{`
322			`r->_errno = EILSEQ;`
323			`return -1;`
324			`}`
325			`state->__value.__wchb[1] = ch;`
326			`if (state->__count == 1)`
327			`state->__count = 2;`
328			`else if (n < (size_t)-1)`
329			`++n;`
330			`if (n < 3)`
331			`return -2;`
332			`ch = (state->__count == 2) ? t[i++] : state->__value.__wchb[2];`
333			`if (ch < 0x80 \|\| ch > 0xbf)`
334			`{`
335			`r->_errno = EILSEQ;`
336			`return -1;`
337			`}`
338			`state->__value.__wchb[2] = ch;`
339			`if (state->__count == 2)`
340			`state->__count = 3;`
341			`else if (n < (size_t)-1)`
342			`++n;`
343			`if (state->__count == 3 && sizeof(wchar_t) == 2)`
344			`{`
345			`/* On systems which have wchar_t being UTF-16 values, the value`
346			`doesn't fit into a single wchar_t in this case. So what we`
347			`do here is to store the state with a special value of __count`
348			`and return the first half of a surrogate pair. The first`
349			`three bytes of a UTF-8 sequence are enough to generate the`
350			`first half of a UTF-16 surrogate pair. As return value we`
351			`choose to return the number of bytes actually read up to`
352			`here.`
353			`The second half of the surrogate pair is returned in case we`
354			`recognize the special __count value of four, and the next`
355			`byte is actually a valid value. See below. */`
356			`tmp = (wint_t)((state->__value.__wchb[0] & 0x07) << 18)`
357			`\| (wint_t)((state->__value.__wchb[1] & 0x3f) << 12)`
358			`\| (wint_t)((state->__value.__wchb[2] & 0x3f) << 6);`
359			`state->__count = 4;`
360			`*pwc = 0xd800 \| ((tmp - 0x10000) >> 10);`
361			`return i;`
362			`}`
363			`if (n < 4)`
364			`return -2;`
365			`ch = t[i++];`
366			`if (ch < 0x80 \|\| ch > 0xbf)`
367			`{`
368			`r->_errno = EILSEQ;`
369			`return -1;`
370			`}`
371			`tmp = (wint_t)((state->__value.__wchb[0] & 0x07) << 18)`
372			`\| (wint_t)((state->__value.__wchb[1] & 0x3f) << 12)`
373			`\| (wint_t)((state->__value.__wchb[2] & 0x3f) << 6)`
374			`\| (wint_t)(ch & 0x3f);`
375			`if (state->__count == 4 && sizeof(wchar_t) == 2)`
376			`/* Create the second half of the surrogate pair for systems with`
377			`wchar_t == UTF-16 . */`
378			`*pwc = 0xdc00 \| (tmp & 0x3ff);`
379			`else`
380			`*pwc = tmp;`
381			`state->__count = 0;`
382			`return i;`
383			`}`
384
385			`r->_errno = EILSEQ;`
386			`return -1;`
387			`}`
388
389			`/* Cygwin defines its own doublebyte charset conversion functions`
390			`because the underlying OS requires wchar_t == UTF-16. */`
391			`#ifndef __CYGWIN__`
392			`int`
393			`_DEFUN (__sjis_mbtowc, (r, pwc, s, n, charset, state),`
394			`struct _reent *r _AND`
395			`wchar_t *pwc _AND`
396			`const char *s _AND`
397			`size_t n _AND`
398			`const char *charset _AND`
399			`mbstate_t *state)`
400			`{`
401			`wchar_t dummy;`
402			`unsigned char t = (unsigned char )s;`
403			`int ch;`
404			`int i = 0;`
405
406			`if (pwc == NULL)`
407			`pwc = &dummy;`
408
409			`if (s == NULL)`
410			`return 0; /* not state-dependent */`
411
412			`if (n == 0)`
413			`return -2;`
414
415			`ch = t[i++];`
416			`if (state->__count == 0)`
417			`{`
418			`if (_issjis1 (ch))`
419			`{`
420			`state->__value.__wchb[0] = ch;`
421			`state->__count = 1;`
422			`if (n <= 1)`
423			`return -2;`
424			`ch = t[i++];`
425			`}`
426			`}`
427			`if (state->__count == 1)`
428			`{`
429			`if (_issjis2 (ch))`
430			`{`
431			`*pwc = (((wchar_t)state->__value.__wchb[0]) << 8) + (wchar_t)ch;`
432			`state->__count = 0;`
433			`return i;`
434			`}`
435			`else`
436			`{`
437			`r->_errno = EILSEQ;`
438			`return -1;`
439			`}`
440			`}`
441
442			`pwc = (wchar_t)t;`
443
444			`if (*t == '\0')`
445			`return 0;`
446
447			`return 1;`
448			`}`
449
450			`int`
451			`_DEFUN (__eucjp_mbtowc, (r, pwc, s, n, charset, state),`
452			`struct _reent *r _AND`
453			`wchar_t *pwc _AND`
454			`const char *s _AND`
455			`size_t n _AND`
456			`const char *charset _AND`
457			`mbstate_t *state)`
458			`{`
459			`wchar_t dummy;`
460			`unsigned char t = (unsigned char )s;`
461			`int ch;`
462			`int i = 0;`
463
464			`if (pwc == NULL)`
465			`pwc = &dummy;`
466
467			`if (s == NULL)`
468			`return 0;`
469
470			`if (n == 0)`
471			`return -2;`
472
473			`ch = t[i++];`
474			`if (state->__count == 0)`
475			`{`
476			`if (_iseucjp1 (ch))`
477			`{`
478			`state->__value.__wchb[0] = ch;`
479			`state->__count = 1;`
480			`if (n <= 1)`
481			`return -2;`
482			`ch = t[i++];`
483			`}`
484			`}`
485			`if (state->__count == 1)`
486			`{`
487			`if (_iseucjp2 (ch))`
488			`{`
489			`if (state->__value.__wchb[0] == 0x8f)`
490			`{`
491			`state->__value.__wchb[1] = ch;`
492			`state->__count = 2;`
493			`if (n <= i)`
494			`return -2;`
495			`ch = t[i++];`
496			`}`
497			`else`
498			`{`
499			`*pwc = (((wchar_t)state->__value.__wchb[0]) << 8) + (wchar_t)ch;`
500			`state->__count = 0;`
501			`return i;`
502			`}`
503			`}`
504			`else`
505			`{`
506			`r->_errno = EILSEQ;`
507			`return -1;`
508			`}`
509			`}`
510			`if (state->__count == 2)`
511			`{`
512			`if (_iseucjp2 (ch))`
513			`{`
514			`*pwc = (((wchar_t)state->__value.__wchb[1]) << 8)`
515			`+ (wchar_t)(ch & 0x7f);`
516			`state->__count = 0;`
517			`return i;`
518			`}`
519			`else`
520			`{`
521			`r->_errno = EILSEQ;`
522			`return -1;`
523			`}`
524			`}`
525
526			`pwc = (wchar_t)t;`
527
528			`if (*t == '\0')`
529			`return 0;`
530
531			`return 1;`
532			`}`
533
534			`int`
535			`_DEFUN (__jis_mbtowc, (r, pwc, s, n, charset, state),`
536			`struct _reent *r _AND`
537			`wchar_t *pwc _AND`
538			`const char *s _AND`
539			`size_t n _AND`
540			`const char *charset _AND`
541			`mbstate_t *state)`
542			`{`
543			`wchar_t dummy;`
544			`unsigned char t = (unsigned char )s;`
545			`JIS_STATE curr_state;`
546			`JIS_ACTION action;`
547			`JIS_CHAR_TYPE ch;`
548			`unsigned char *ptr;`
549			`unsigned int i;`
550			`int curr_ch;`
551
552			`if (pwc == NULL)`
553			`pwc = &dummy;`
554
555			`if (s == NULL)`
556			`{`
557			`state->__state = ASCII;`
558			`return 1; /* state-dependent */`
559			`}`
560
561			`if (n == 0)`
562			`return -2;`
563
564			`curr_state = state->__state;`
565			`ptr = t;`
566
567			`for (i = 0; i < n; ++i)`
568			`{`
569			`curr_ch = t[i];`
570			`switch (curr_ch)`
571			`{`
572			`case ESC_CHAR:`
573			`ch = ESCAPE;`
574			`break;`
575			`case '$':`
576			`ch = DOLLAR;`
577			`break;`
578			`case '@':`
579			`ch = AT;`
580			`break;`
581			`case '(':`
582			`ch = BRACKET;`
583			`break;`
584			`case 'B':`
585			`ch = B;`
586			`break;`
587			`case 'J':`
588			`ch = J;`
589			`break;`
590			`case '\0':`
591			`ch = NUL;`
592			`break;`
593			`default:`
594			`if (_isjis (curr_ch))`
595			`ch = JIS_CHAR;`
596			`else`
597			`ch = OTHER;`
598			`}`
599
600			`action = JIS_action_table[curr_state][ch];`
601			`curr_state = JIS_state_table[curr_state][ch];`
602
603			`switch (action)`
604			`{`
605			`case NOOP:`
606			`break;`
607			`case EMPTY:`
608			`state->__state = ASCII;`
609			`*pwc = (wchar_t)0;`
610			`return 0;`
611			`case COPY_A:`
612			`state->__state = ASCII;`
613			`pwc = (wchar_t)ptr;`
614			`return (i + 1);`
615			`case COPY_J1:`
616			`state->__value.__wchb[0] = t[i];`
617			`break;`
618			`case COPY_J2:`
619			`state->__state = JIS;`
620			`*pwc = (((wchar_t)state->__value.__wchb[0]) << 8) + (wchar_t)(t[i]);`
621			`return (i + 1);`
622			`case MAKE_A:`
623			`ptr = (unsigned char *)(t + i + 1);`
624			`break;`
625			`case ERROR:`
626			`default:`
627			`r->_errno = EILSEQ;`
628			`return -1;`
629			`}`
630
631			`}`
632
633			`state->__state = curr_state;`
634			`return -2; /* n < bytes needed */`
635			`}`
636			`#endif /* !__CYGWIN__*/`
637			`#endif /* _MB_CAPABLE */`